Citation
Assay development, tissue distribution and pharmacodynamics of a novel estrogen-chemical delivery system for the brain

Material Information

Title:
Assay development, tissue distribution and pharmacodynamics of a novel estrogen-chemical delivery system for the brain
Creator:
Rahimy, Mohamad H., 1954-
Publication Date:
Language:
English
Physical Description:
xii, 202 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Anterior pituitary ( jstor )
Blood brain barrier ( jstor )
Dosage ( jstor )
Estrogens ( jstor )
Hormones ( jstor )
Hydrolysis ( jstor )
Metabolites ( jstor )
Plasmas ( jstor )
Rats ( jstor )
Steroids ( jstor )
Brain ( mesh )
Department of Pharmacodynamics thesis Ph.D ( mesh )
Dissertations, Academic ( mesh )
Dissertations, Academic -- College of Pharmacy -- Department of Pharmacodynamics -- UF ( mesh )
Drug Administration Routes ( mesh )
Drug Delivery Systems ( mesh )
Estrogens -- administration & dosage ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 175-201).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Mohamad H. Rahimy.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Mohamad Hossein Rahimy. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
25072883 ( OCLC )
023234089 ( ALEPH )

Downloads

This item has the following downloads:


Full Text










ASSAY DEVELOPMENT, TISSUE DISTRIBUTION
AND PHARMACODYNAMICS OF A NOVEL ESTROGEN-CHEMICAL
DELIVERY SYSTEM FOR THE BRAIN











BY


MOHAMAD H. RAHIMY


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY



UNIVERSITY OF FLORIDA


1990





To my parents for their encouragement and unwavering
support toward my education, and my wife, Mahbobeh, and my son, Ehsan,

whose patience and sacrifices helped to make this work possible.


I


I
I













ACKNOWLEDGEMENTS


This work would never have come to fruition without the

encouragement, assistance and advice of many individuals whom I am very grateful to. First, I wish to express my sincere appreciation and gratitude to my mentor, Dr. James W. Simpkins, for his expert guidance, encouragement, and support. Throughout my graduate study at the University of Florida, I had ample opportunity to learn by experience under the skillful guidance of Dr. Simpkins. I also wish to express great thanks to the other members of my committee, Dr. Nicholas Bodor, Dr. William Millard, Dr. Edwin Meyer, and Dr. Ralph Dawson, who have imparted valuable advice as well as their critical evaluation of my work. I would also like to extend thanks to Dr. Michael Meldrum, Dr. Michael Katovich, Dr. Wesley Anderson, and Dr. Anna Ratka for their advice and assistance.

I would like to thank the many others who contributed their time and efforts, especially Victoria Red Patterson, Becky Hamilton, Lee Glancey, Terry Romano, Debby Andreadis, Billie Jean Goins, Roxane Federline, and Denise Blake, who assisted me in various aspects of this work. The assistance of Anup Zutshi regarding the kinetic analysis is greatly appreciated. My personal thanks go to Dr. Lee Ann Burgland and Dave Wallace whose cooperations during these many years made graduate school more bearable. I extend thanks to new graduate students Singh Meharvan, Melanie King, Melanie Pecins, and Jean Bishop-Sparks who have already taken over the reins in the lab for accepting the challenge.


iii









Finally, very special thanks go to my parents who have always inspired me to pursue an academic career, and to my supportive and considerate family (my wife & my son) who always managed to create an environment in which I could devote the many years required to accomplish this work.


iv















TABLE OF CONTENTS

ACKN OW LEDGEM EN TS.........................................................................................iii

LIST OF TABLES..........................................................................................................viii

LIST OF FIGURES..................................................................................................... ix

A BSTRACT................................................................................................................... xi

CHAPTERS

1. IN TRODUCTION ................................................................................................1

2. REVIEW OF TH E LITERATURE.........................................................................7
Estrogen Horm ones........................................................................................... 7
H istorical Observations ............................................................................. 7
Endocrinological and Biochemical Considerations ............................8
Biosynthesis ........................................................................................... 8
Secretion and transport....................................................................... 10
M etabolism and excretion........................................................................11
M echanism of action.................................................................................12
Estrogen Receptors....................................................................................... 15
Intracellular/cytosolic receptors ......................................................... 15
M em brane receptors............................................................................. 16
Estrogen-receptor binding kinetics.................................................... 17
Role of Estrogen in the M enstrual Cycle ................................................ 19
Feedback regulation.............................................................................. 20
Role of Estrogen in the Rat Estrous Cycle............................................. 22
Feedback regulation............................................................................. 23
Blood-Brain Barrier............................................................................................25
H istorical Overview .................................................................................. 25
Potential Asset to Utilize in the Design of Brain-Specific Drug
Delivery.....................................................................................................28
Therapies Aimed at Targeting/Enhancing Brain Estradiol Levels.........32
Fertility Regulation.................................................................................... 32
M enopausal Syndrom e .............................................................................. 33
Prostatic Cancer ........................................................................................... 34
Body W eight Regulation.......................................................................... 35
Libido/Sexual Dysfunction....................................................................... 36
Disorders of Depression............................................................................. 37
Cognitive Impairment of Menopausal Alzheimer's Type.................41


V










3. GEN ERAL M ATERIALS AN D M ETHODS.................................................. 44
Drugs and Solutions......................................................................................... 44
Estradiol and Standard Solution.............................................................. 44
Estradiol-Chem ical Delivery System ....................................................... 44
Estradiol Pellet.............................................................................................. 45
M orphine Pellets......................................................................................... 45
Anim als ...................................................................................................................46
Drug Treatm ent.................................................................................................. 47
Steroid Treatment.........................................................................................47
M orphine and N aloxone Treatm ent...................................................... 48
Plasm a Horm ones Radioim m unoassays .................................................... 48
Protein Horm one Assays ........................................................................... 48
Steroid Horm one Assays........................................................................... 49
Statistical Analysis............................................................................................ 50

4. DEVELOPMENT OF AN ANALYTICAL METHOD FOR THE
QUANTITATION OF E2-CDS METABOLITES IN A WIDE
VARIETY OF TISSUES IN THE RAT................................................................51
Introduction....................................................................................................... 51
M aterials and M ethods..................................................................................... 52
In Vitro M ethodology ................................................................................. 53
Specificity of the estradiol antibody for E2 ..................................... 53
Selective solvent extraction of steroids from tissues...........53
Hydrolysis of E2-Q+ in various tissue extracts ................................ 55
Solid-phase extraction and separation of E2 by C18 columns.....55 Radioimm unoassay of E2 ................................................................... 56
Calculations........................................................................................... 56
In Vivo Studies ............................................................................................ 57
Results......................................................................................................................58
In Vitro M ethodology ................................................................................. 58
Cross-reactivity of the E2 antibody with E2-Q+ ............................... 58
Recovery of E2 ........................................................................................ 58
Precision of the E2 extraction-assay m ethod .................................... 59
Recovery of E2-Q+ ................................................................................. 60
Precision of the E2-Q+ extraction-assay m ethod............................. 60
Distribution of E2 and E2-Q+ in vivo............................................... 61
Discussion................................................................................................................62

5. DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
M ETABOLITES IN M ALE RATS.......................................................................75
Introduction....................................................................................................... 75
M aterials and M ethods..................................................................................... 77
Results....................................................................................................................78
Discussion................................................................................................................80




Vi










6. DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
M ETABOLITES IN FEM ALE RATS...................................................................90
Introduction....................................................................................................... 90
M aterials and M ethods..................................................................................... 91
Experim ent 1.............................................................................................. 91
Experim ent 2.............................................................................................. 92
Results......................................................................................................................94
Experim ent 1.............................................................................................. 94
Experim ent 2.............................................................................................. 96
Discussion................................................................................................................96

7. EVALUATION OF THE PHARMACODYNAMIC EFFECTS OF E2CDS IN OVARIECTOM IZED FEM ALE RATS.................................................115
Introduction............................................................................................................115
M aterials and M ethods.........................................................................................116
Results......................................................................................................................117
Discussion................................................................................................................119

8. EFFECTS OF THE E2-CDS OR CASTRATION ON ANDROGEN
AND ANDROGEN-DEPENDENT TISSUES IN MALE RATS.........128 Introduction............................................................................................................128
M aterials and M ethods.........................................................................................131
Results......................................................................................................................133
Effect of CAST or E2-CDS on Plasm a T Levels...........................................133
Effect of CAST or E2-CDS on Tissue W eights............................................134
Effect of CAST or E2-CDS on Plasm a H orm ones ......................................135
Discussion................................................................................................................135

9. EFFECTS OF THE E2-CDS OR E2 PELLET ON TAIL-SKIN
TEMPERATURE RESPONSES IN OVARIECTOMIZED FEMALE
RATS ........................................................................................................................148
Introduction............................................................................................................148
M aterials and M ethods......................................................................................... 151
Results......................................................................................................................153
Discussion................................................................................................................155

10. GEN ERAL DISCUSSION .....................................................................................163

REFEREN CES...............................................................................................................175

BIOGRAPHICAL SKETCH ......................................................................................... 202


Vii














LIST OF TABLES


Table Page

1. Recovery and Precision Determinations for Biological Samples
Spiked with E2 -------............................................72

2. Recovery and Precision Determinations for Biological Samples
Spiked w ith E2-Q + ...................................................................................... 73

3. Percent Hydrolysis of E2-Q+ in Supernatants of a Variety of
T issu es........................................................................................................... 74

4. Effects of Dose on the Extent of Oxidation and Hydrolysis of E2CDS in a Variety of Tissues in vivo ............................................................111

5. Effects of the E2-CDS on the Clearance of E2-Q+ from a Variety of
T issu es................................................................................................................112

6. Effects of the E2-CDS on the Clearance of E2 from a Variety of
T issu e s................................................................................................................113

7. Effects of an Equimolar Dose of E2 on the Tissue Concentrations
o f E 2 ....................................................................................................................114

8. Dose and Time-Dependent Effects of the E2-CDS on Peripheral
Tissue Weights in Ovariectomized Rats ....................................................127

9. Effects of the E2-CDS or CAST on Plasma Hormone
Concentrations at 7 Days after the Last Treatment in Male Rats...........146

10. Effects of the E2-CDS or CAST on Plasma Hormone
Concentrations at 14 Days after the Last Treatment in Male Rats.........147

11. Effects of the E2-CDS or E2 Pellet on Basal Temperature, Maximal
Change in TST, and Area Under the 90 Min TST Curve in
Ovariectomized, Morphine-Dependent Rats ............................................161

12. Effects of the E2-CDS or E2 Pellet on Plasma Hormone
Concentrations in Ovariectomized, Morphine-Dependent Rats..........162


viii














LIST OF FIGURES


Figure Page

1. Schematic representation of in vitro synthesis and in vivo
transformation of the estradiol-chemical delivery system (E2C D S) .............................................................................................................. 67

2. Inhibition of 1251-E2 binding to an E2 antibody caused by E2 or E2Q + ........................................................................................................................68

3. Recovery of known concentrations of E2-Q+ and E2 added to
brain tissue homogenates prior to extraction.......................................69

4. Inhibition of 1251-E2 binding to an E2 antibody caused by
increasing amounts of brain tissue from rats treated with the
estradiol-chemical delivery system........................................................ 70

5. Effects of a single iv dose of the estradiol-chemical delivery
system on serum and brain concentrations of E2-Q+ and E2 or the
H PC D vehicle.............................................................................................. 71

6. Effects of a single iv dose of the E2-CDS on brain and plasma
concentrations of E2-Q+ and E2 in intact male rats.............................84

7. Effects of a single iv dose of the E2-CDS on liver and fat
concentrations of E2-Q+ and E2 in intact male rats.............................86

8. Effects of a single iv dose of the E2-CDS on kidney, heart, lung,
and anterior pituitary concentrations of E2-Q+ and E2 in intact
m ale rats..................................................................................................... . 87

9. Brain and anterior pituitary contents of the E2-Q+ and E2
following a single iv dose of the E2-CDS. .............................................88

10. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in whole brain of ovariectomized rats............................100

11. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in hypothalamus of ovariectomized rats.......................102


ix









12. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in plasma of ovariectomized rats.....................................104

13. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in liver of ovariectomized rats.........................................106

14. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in fat of ovariectomized rats.............................................108

15. Effects of a single iv dose of the E2-CDS on plasma E2-Q+ or
plasma E2 concentrations in ovariectomized rats....................................110

16. Dose and time-dependent effects of the E2-CDS on plasma LH
responses in ovariectom ized rats.................................................................124

17. Dose and time-dependent effects of the E2-CDS on plasma FSH
responses in ovariectom ized rats.................................................................125

18. Dose and time-dependent effects of the E2-CDS on plasma PRL
and GH responses in ovariectomized rats..................................................126

19. Effects of the E2-CDS or CAST on plasma testosterone
concentrations either 7 days or 14 days after the final treatment..........141

20. Effects of the E2-CDS or CAST on ventral prostate weight either 7
days or 14 days after the final treatment.....................................................142

21. Effects of the E2-CDS or CAST on seminal vesicle weight either 7
days or 14 days after the final treatment. ....................................................143

22. Effects of the E2-CDS or CAST on testis weight either 7 days or 14
days after the final treatm ent........................................................................144

23. Effects of the E2-CDS or CAST on anterior pituitary weight either
7 days or 14 days after the final treatment..................................................145

24. Effects of the E2-CDS or E2 pellet on the mean TST responses
induced by naloxone administration to morphine-dependent,
ovariectom ized rats.........................................................................................159

25. Effects of the E2-CDS or E2 pellet on the mean RT responses
induced by naloxone administration to morphine-dependent,
ovariectom ized rats.........................................................................................160


X













Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


ASSAY DEVELOPMENT, TISSUE DISTRIBUTION
AND PHARMACODYNAMICS OF A NOVEL ESTROGEN-CHEMICAL DELIVERY SYSTEM FOR THE BRAIN


By

Mohamad H. Rahimy

August, 1990




Chairman: Dr. James W. Simpkins Major Department: Pharmacodynamics


Enhanced delivery and sustained release of estradiol (E2) in the brain is desirable for effective treatments of the menopausal hot flush, prostatic adenocarcinoma, and fertility regulation. Our studies thus evaluated an E2chemical delivery system (E2-CDS) for the brain, which is based upon the interconvertible dihydropyridine <* pyridinium ion redox reaction. The E2CDS requires multiple, facile chemical conversions, including the oxidation of E2-CDS to the corresponding quaternary ion (E2-Q+), which provides the basis of locking the molecule within the brain, and the subsequent slow hydrolysis of E2-Q+ by esterases to E2 in that tissue.

Initially, studies were undertaken to develop a reliable, specific, and sensitive method to simultaneously measure E2-Q+ and E2 (two metabolites


xi








of E2-CDS) in various biological tissues. This method utilized the following steps: (1) selective solvent extraction of E2-Q+ and E2 from the tissues; (2) basecatalyzed hydrolysis of E2-Q+ to E2 in NaOH; (3) solid-phase purification of E2 with C18 reversed-phase extraction columns; and (4) radioimmunoassay of E2.

Subsequently, the in vivo tissue distributions of E2-Q+ and E2 were determined in both male and female rats. The results revealed that the disappearance of E2-Q+ as well as E2 was slow in brain tissue with a t1/2 = 8-9 days. By contrast, both of these metabolites exhibited relatively rapid clearance from the plasma, liver, lung, kidney, heart, fat, and uterus.

After documenting the kinetic behaviors of E2-CDS, time-course studies were then conducted to assess the dynamic effects of E2-CDS on responses which are known to be affected by E2. The E2-CDS consistently exhibited prolonged and sustained suppression of pituitary gonadotropins secretion, i.e. LH and FSH in a dose- and time-dependent manner.

Finally, the therapeutic potentials of E2-CDS were investigated in male and female rats. Studies in the male rat demonstrated that E2-CDS is as effective as castration in both suppressing the plasma testosterone levels and reducing the weights of androgen-responsive tissues. Further studies in the female rat, examining the effects of E2-CDS on tail-skin temperature (TST) responses, revealed that E2-CDS can significantly attenuate the rise in TST.

Collectively, the results of these studies are consistent with the
proposed mechanism of this drug delivery system, that is, the preferential retention of E2-Q+ by the brain, and the subsequent slow release of E2 locally in that tissue. Furthermore, the profound pharmacodynamic effects of this delivery system support the view that E2-CDS may be potentially useful for fertility regulation, the effective treatments of prostatic cancer, and certain brain-mediated estrogen withdrawal symptoms, i.e. menopausal hot flushes.


xii













CHAPTER 1
INTRODUCTION


Estrogens exhibit a myriad of important regulatory roles in the growth, development, and maintenance of the structures and functions which are necessary for the continuation of the species. Their therapeutic applications for certain clinical problems have been appreciated since the turn of the century, when ovarian grafts were shown to prevent uterine atrophy and loss of sexual function in castrated animals (Knauer, 1900). Estrogen hormones have broad therapeutic applications and in most cases the steroids are used primarily for their central actions (Meites & Nicoll, 1965). Among these are the reproductive-related applications, including fertility regulation, sexual dysfunction, and the replacement therapy in postmenopausal patients; and the non-reproductive applications, including treatment of postmenopausal depression and cancer therapy. Nevertheless, the full spectrum of potential clinical benefits and applications of estrogen therapy has yet to be uncovered.

Physiologically, estrogen hormones exert two modes of action on the central nervous system (CNS), particularly on the brain. First, during the critical period of fetal/neonatal life, estrogens affect permanently some features of the brain structure and function which result in neuronal differentiation (Allen et al., 1989; Goy & McEwen, 1980). Second, during the course of adult life, these hormones exert their effects in a modulatory and reversible mode that influence a myriad of adult brain functions (McEwen, 1988; McEwen & Parsons, 1982).


1





2


Therapeutically, this latter central action of estrogens is of significant interest due to the existence of several clinical conditions which are influenced only by the presence of estrogens in the brain. For instance, after menopause the decline in ovarian function leads to a number of central nervous system (CNS)-mediated estrogen-withdrawal symptoms (Notelovitz, 1986). The symptoms are clearly caused by brain deprivation of estrogen (Judd, 1983; Lauritzen, 1973, 1982) since they can be alleviated by the replacement of estrogen (Campbell & Whitehead, 1977; Upton, 1984). Furthermore, evidence suggests that the brain is the primary locus where estradiol (E2) exerts its effect to inhibit the secretion of luteinizing hormonereleasing hormone (LHRH) from the hypothalamus (Goodman & Knobil, 1981; Kalra & Kalra, 1980, 1983, 1989; Plant, 1986) and hence of LH from the anterior pituitary and eventually of gonadal steroid hormones. As such, the E2 hormone has been and continues to be used therapeutically for (1) fertility regulation (Briggs, 1976; Davidson, 1969) and (2) treatment of androgendependent prostatic adenocarcinoma (van Steenbrugge et al., 1988) by virtue of suppressing plasma androgen levels (Carlstrom et al., 1989). Additionally, estrogens are believed to act centrally to stimulate male and female sexual behaviors (Beyer et al., 1976; Christensen & Clemens, 1974; MacLusky et al., 1984), to regulate body weight (Palmer & Gray, 1986; Pliner & Fleming, 1983), and may have influences on mood (Klaiber et al., 1976, 1979; Lauritzen & van Keep, 1978; Schneider et al., 1977), and on cognitive functioning (Fillit et al., 1986; Hackman & Galbraith, 1976).

Potential adverse effects and toxicity have, however, been associated with the currently used estrogens. Estrogen hormones are intrinsically lipophilic (Abraham, 1974). The high lipophilicity of these steroids ensures their rapid penetration of biological membranes, including the blood-brain





3


barrier (BBB), thus enabling access to all cells and organs. Indeed, when these hormones are used therapeutically to specifically target the CNS, the steroids equilibrate among all body tissues due to their high lipophilicity (Pardridge & Meitus, 1979). Moreover, when inside the CNS, there is no mechanism to prevent their redistribution back to the periphery as blood levels of the steroids decline (Davson, 1976; Schanker, 1965). So, even if estrogens can easily gain access to the CNS, they are poorly retained by the brain. As a result, only a fraction of the administered estrogen dose accumulates at or near the site of action in the brain. This property of the estrogens necessitates either frequent dosing or the administration of a depot form of the estrogen in order to maintain therapeutically effective concentrations in the brain (Schanker, 1965; Spona & Schneider, 1977). Both of these treatment strategies lead to sustained increases in peripheral estrogen levels. Since estrogen receptors are present in many peripheral tissues (Walters, 1985), where they mediate a myriad of physiological and pharmacological effects (Murad & Haynes, 1985), it further creates the potential of untoward peripheral toxicities. In fact, constant increases in peripheral tissue exposure to estrogens have been shown in numerous retrospective studies to precipitate various peripheral toxicities, including increased risk of breast and endometrial cancer (Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger et al., 1988; Persson, 1985; Thomas, 1988), cardiovascular complications (Barrett-Conner et al., 1989; Inman & Vessey, 1968; Kaplan, 1978; Thomas, 1988), and marked interference with hepatic metabolism (Burkman, 1988).

In addition to the peripheral toxicities mentioned above, constant exposure to high levels of E2 valerate has been shown to induce neuronal degeneration in the hypothalamic arcuate nucleus of both male and female rats (Brawer et al., 1980, 1983). Furthermore, other experimental conditions





4


which result in constant exposure of the hypothalamus to endogenous estrogens, i.e. constant exposure to illumination can also induce similar arcuate nucleus neuropathological lesion in the rat (Brawer et al., 1983). Conversely, experimental manipulations that essentially eliminate circulating E2 levels (e.g. ovariectomy) greatly reduce the magnitude of the arcuate nucleus neuropathological responses to constant illumination or senescence in the female rat. Although there is no direct evidence as yet that E2 is the primary neurotoxic agent responsible for the arcuate lesion, it is noteworthy that this region of the hypothalamus is particularly rich in estrogen receptors as well as E2-concentrating neurons (Pfaff & Keiner, 1973). Thus, it may be that this region of the hypothalamus is exquisitely susceptible to E2 in any concentration for prolonged period of time in the adult female rat.

Given the aforementioned evidence for: (i) the central actions and the therapeutic implications of estrogen hormones, and (ii) the major limiting factors associated with the use of currently available estrogen medications, a brain-estrogen delivery system with sustained release of estrogen in that tissue is dearly warranted.

Over the past two decades, the attention and efforts of pharmaceutical research were generally focused on the strategy of improving the efficacy as well as the specificity of pharmaceutical products in order to minimize or even abolish their adverse effects. To fulfill these objectives, novel drug delivery systems have been designed and formulated to achieve ratecontrolled and targeted-organ delivery (Bodor, 1987; Bodor et al., 1981, 1987; Bodor & Farag, 1983; Bodor & Simpkins, 1983). This strategy would not only ensure a therapeutic agent preferentially gets to its intended site of action, but it does so at the desired rate in order to satisfy the therapeutic criteria.





5


A remarkable example is the design of an estradiol-chemical delivery system (E2-CDS) for the enhanced and sustained release of E2 in the brain (Bodor et al., 1987). The E2-CDS exploits the unique architecture of the BBB, which normally excludes a variety of pharmacological agents from the CNS due to their physicochemical properties (Bodor & Brewster, 1983). The E2CDS is a redox-based chemical-delivery system and the mechanism of its drug delivery is based upon an interconvertible dihydropyridine <- pyridinium ion redox carrier (Bodor et al., 1987). After systemic administration of the E2CDS, it distributes throughout the body, then, the carrier moiety is quickly oxidized to the corresponding quaternary pyridinium ion (E2-Q+) in the brain as well as in peripheral tissues. The charged pyridinium-drug complex is thus locked into the CNS while the same moiety rapidly clears from the periphery because of a 40,000-fold increase in its hydrophilicity. Sustained release of the active, parent drug from the charged pyridinium-drug complex occurs in the brain as a result of enzymatic hydrolysis of the ester linkage. The enzymes involved in cleavage of the ester bond are believed to be nonspecific esterases.

Collectively, the ability to preferentially deliver E2 to the brain, thus sparing non-target peripheral site tissues, should improve the therapeutic index of E2 by (i) increasing the concentrations and/or residence time of E2 at its receptor site in the brain and (ii), equally important, decreasing the concentrations and/or residence time of E2 at the potential peripheral sites of toxicities, thereby decreasing untoward peripheral side effects.

To document the predictive biotransformation behaviors of the E2-CDS (Bodor et al., 1987), and to further substantiate its effectiveness over the currently used estrogens, extensive and long-term pharmacokinetic and pharmacodynamic studies were conducted. These studies included the





6


following: (1) development of a reliable, sensitive, and specific method for the simultaneous quantitation of E2-Q+ and E2, two metabolites of the E2-CDS, in a wide variety of rat tissues; (2) determination of the tissue distributions of these metabolites in the whole brain, hypothalamus, anterior pituitary, lung, liver, heart, kidney, uterus, adipose tissues, and plasma of the rat; (3) evaluation of the pharmacodynamic consequences of E2-CDS following its administration into ovariectomized female rats; and finally (4) assessment of the therapeutic potentials of E2-CDS in animal models for (i) the menopausal hot flush and (ii) androgen-dependent prostatic hyperplasia.














CHAPTER 2
REVIEW OF THE LITERATURE

This chapter will first present a historical review with respect to the
endocrinology/ neuroendocrinology of estrogen hormones. This will include some evidence pertinent to their physiological/pharmacological actions in the central nervous system (CNS). Furthermore, since the unique architecture of the brain, the blood-brain barrier (BBB), is of central asset in the design and synthesis of the estrogen-delivery system under investigation, a historical account of the BBB will be discussed. Attempts will be made to identify problems associated with the brain delivery of existing drugs. Finally, example of certain clinical conditions which require the presence of estrogen in the brain as a therapeutic agent will be discussed as well. The purpose of this diverse literature review is to identify and describe the concepts and rationale which were the basis in the design and synthesis of the E2-CDS, which will be evaluated in detail in later chapters.


Estrogen Hormones


Historical Observations


Ovarian endocrine activity was first demonstrated experimentally by Knauer in 1896 (quoted by Tepperman, 1981). Independently, Sobotta (1896) described the origin of corpus luteum at the same time. Shortly thereafter, Beard (1897) postulated that the corpus luteum might serve a necessary function during pregnancy. The observation by Knauer (1900), who


7





8


demonstrated that ovarian transplants prevented uterine atrophy and loss of sexual function in castrated animals, established the hormonal nature of ovarian control of the female reproductive system. Further supporting evidence was provided by Fraenkel (1903), who showed that destruction of the corpora lutea in pregnant rabbits causes abortion. In 1923, Allen and Doisy developed a simple, quantitative bioassay for ovarian extracts based upon changes produced in the vaginal smear of the rat. Two years later, Loewe (1925) reported on a female sex hormone in the blood of various species. Shortly thereafter, Loewe and Lange (1926) discovered a female sex hormone in the urine of menstruating women with the observation that the concentration of the hormone in the urine varied with the phases of the menstrual cycle. These observations set the stage for chemists, who soon isolated independently an active estrogen substance from urine in crystalline form (Butenandt, 1929; Doisy et al., 1929, 1930). In 1935, Doisy et al. (quoted by Tepperman, 1981) characterized the chemical structure of estradiol-17s. However, it was the contributions of Corner and Allen (1929) that firmly established the endocrine function of the corpus luteum. They clearly demonstrated that the abortion following extirpation of the corpora lutea in pregnant rabbits can be prevented by the injection of luteal extracts.


Endocrinological and Biochemical Considerations


Biosynthesis

Estrogens are primarily produced by the follicles and corpus luteum of the ovary and by the placenta during the second and third trimesters of pregnancy. Ovaries secrete estradiol (E2) and estrone, whereas the placenta produces these and estriol (Ross, 1985; Schwartz, 1981). All these hormones





9


exhibit estrogenic activity; however, 17 P-E2 is the major and most potent estrogen produced by the ovaries of most species including the human and the rat. Ovaries are capable of synthesizing cholestrol de novo from acetate and subsequently converting it to other steroids, including estrogens, progestins, and androgens (Miller, 1988; Schwartz, 1981) Data obtained from various enzyme kinetics and steroidal precursor-product relationships indicate the involvement of very large number of distinct enzymes, most are members of the cytochrome P 450 oxidases, in the conversion of cholestrol to active steroid hormones (Miller, 1988; Ross, 1985; Schwartz, 1981). Estradiol is formed from either androstendione or testosterone via an aromatization reaction. This reaction is of central importance in estrogen formation, and it is not limited to the gonads or placenta rather a wide variety of peripheral tissues as well as the CNS can aromatize the A ring of androstendione and testosterone to form estrogens (Canick et al., 1986; Michael et al., 1986).

The formation of estrogens is regulated by the concerted actions of two pituitary gonadotropic hormones: follicle-stimulating hormone (FSH) and luteinizing hormone (LH) (Richards & Hedin, 1988). FSH influences the growth and maturation of ovarian follicles, whereas LH stimulates the synthesis and secretion of estrogens (Richards & Hedin, 1988). The synthesis and release of LH and FSH are, in turn, regulated by the hypothalamic gonadotropin-releasing hormone (GnRH/LHRH) (Dalkin et al., 1989; Davidson, 1969). Furthermore, feedback effects of E2 and other gonadal factors on the anterior pituitary and primarily on the hypothalamus influence the synthesis and secretion of LHRH (Kalra & Kalra, 1983, 1989; Plant, 1986; Rosie et al., 1990).





10


Secretion and transport
The E2-producing cells in the ovary and corpus luteum do not
characteristically prepackage large amounts of steroid hormone for release. Rather, these endocrine tissues store the hormone precursor, cholesterolester, in the form of lipid droplets inside the hormone-producing cells (Rossmanith et al., 1990). This indicates that, in these secretory cells, the signal for E2 release is perhaps tightly coupled to that of estrogen-hormone synthesis. Thus, the newly synthesized hormone will be released into the circulation for transport to target tissues. The signals for release are primarily those of the anterior pituitary tropic factors (LH & FSH). It is believed that E2 is released in pulsatile fashion and this perhaps is the result of the episodic modulating influence of LH (Rossmanith et al., 1990).

Estradiol hormone, like other steroid hormones, when secreted into the blood, it is primarily transported by carrier proteins. E2 may be transported by: (a) plasma albumin (60%) with low affinity binding; (b) sex hormone-binding globulin (38%) with high affinity binding; and (c) in free (dialyzable) form (2%) (Moutsatsou & Oakey, 1988; Pardridge, 1988a). The carrier-bound E2 is biologically inactive and sequestered in plasma while only the free fraction is regarded as biologically active hormone. Recent studies have, however, suggested that the carrier-bound pool of E2 may also be available for uptake by the target tissues (Pardridge, 1988a). It was suggested that there are two possible mechanisms for the delivery of carrier-bound E2 to target tissues (Pardridge, 1988a). One mechanism involves interactions between the carrier protein surface and the surface of the organ microcirculation that results in a conformational change about the carrierbinding site and thus enhanced dissociation of E2 hormone. The second





11


mechanism involves a receptor-mediated transcytosis of carrier-bound hormone complex in the limiting membrane of the organ microcirculation. However, other investigators have argued against this hypothesis on the grounds that the concept does not readily reconcile with physiological findings (Mendel et al., 1988). That is, the rate of protein-bound hormone dissociation (Kd) is a potentially important factor in the model of "free hormone hypothesis." Thus, a more comprehensive model (equation) which formally takes into account the rate-limiting effects of protein-bound hormone dissociation is more relevant to the experimental observations. The current Pardridge's model (the protein-bound hormone hypothesis) is, however, deficient in this respect.

Metabolism and excretion

The plasma concentration of E2 at any time represents the net
difference between the rate of E2 secretion and the rate of metabolism in the liver and excretion by the kidneys. There is no apparent limit to the capacity of these organs to metabolize and excrete the E2 hormone. The liver is the primary organ for metabolizing E2 hormone (Bolt, 1979). The rate of turnover of E2 hormone is rather rapid. It has a half-life of about 90 min (von Schoultz et al., 1989). The estrogen is oxidized by the action of a stereospecific dehydrogenase enzyme, using pyridine nucleotides as cofactors, to less active products such as estrone and estriol. The oxidized metabolites are then conjugated as sulfates or glucuronides. The conjugation process renders these metabolites highly water soluble which then are quickly excreted in the urine or bile (Bolt, 1979). These conjugates are biologically inactive; however, the biliary metabolites may undergo further metabolism by action of the intestinal flora. The products are then reabsorbed into the portal circulation





12


and transported to the liver, a process called enterohepatic circulation (Bolt, 1979). A metabolite of E2, which comprises at least 20% of the total amount secreted in humans, is the 2-hydroxyl derivative. These metabolites, referred to as catechol estrogens, are shown to have biological activity. The biological activity of the catechol estrogen appears to involve an interaction with catecholamine synthesis, receptors or effectors (Weisz & Crowley, 1986). The conversion to catechol estrogen can occur in a number of tissues, including the CNS (Weisz & Crowley, 1986).

Mechanism of action

Based on the gross anatomical, histological, and biochemical evidence, E2 is shown to have growth-promoting activities on cells of the target organs such as the anterior pituitary, uterus, vagina, Graafian follicles of the ovary and the mammary gland by increasing protein synthesis and mitotic activity. As early as 1953, Szego and Roberts, seeking an understanding of the mechanism of E2 action, demonstrated accumulation of ribonucleic acid (RNA) and protein in estrogen-stimulated uterine tissues. Mueller et al. (1958) showed that most of the E2 effects on RNA and protein synthesis can be blocked by a translation inhibitor (puromycin) and a transcription inhibitor (actinomycin D). These observations led to the proposal that estrogen hormones and steroids in general stimulate or activate the production of nucleic acid templates (mRNAs) and, hence, gene expression (Mueller et al., 1958). Soon after the technological advances of the 1960's and, thus, the availability of tritium-labeled estradiol (3H-E2), Jensen and Jacobsen (1962) discovered that the estrogen target tissues (uterus and vagina) selectively concentrated the labeled E2. These investigators were also the first to demonstrate the binding of E2 to a specific cytosolic receptor protein (Jensen &





13


Jacobsen, 1962). Further studies demonstrating the nuclear localization of E2 receptors (Jensen & DeSombre, 1972) or increases in thymidine incorporation, mRNA polymerase, mRNA synthesis, and protein synthesis (Hamilton, 1968) supported the concept that the primary site of estrogen action in target tissues is within the nuclear genome. Furthermore, voluminous body of available evidence suggests that steroid hormones in general interact with intracellular stereospecific receptors and upon binding the whole receptor-hormone complex translocates into the nucleus (Walters, 1985). The hormone-receptor complex then alters nuclear gene transcription, leading to the production of all classes of RNA before regulating cytoplasmic protein synthesis (Walters, 1985). These actions generally occur with a delay of several hours or days between the arrival of steroids at the target tissue and the first detectable changes in cellular function.

Several recent contributions to the concept of gene expression by
steroid hormones have included the following: (1) demonstration of specific DNA sequences that serves as the actual nuclear acceptor site for the steroid receptor (Spelsberg et al., 1984); (2) evidence for the binding of the occupied and/or transformed steroid receptors to DNA components (Gehring & Tomkins, 1974); (3) assessment of gene transcription in purified nuclei, including the demonstration that the estrogen receptor-induced increase in ovalbumin mRNA transcription is not only dose-dependent but also tissuespecific (Taylor & Smith, 1982); and (4) isolation of hormone-induced mRNA sequences and subsequent cloning of their complimentary DNAs (cDNAs) (O'Malley et al., 1979).

In addition to the direct genomic actions (delayed effects) of estrogens described above, some target tissues exhibit very rapid responses to estrogen exposure that are difficult to reconcile with the concept of direct gene





14


regulation. For instance, the CNS neurones produce diversity of both rapid and delayed neuroendocrine and behavioral effects in response to estrogen exposure (Majewska, 1987). Various studies have suggested that the neuronal plasma membrane may serve as a direct target for the rapid action of estrogens, which may lead to modification of neurotransmitter release or their receptor/effector systems. First, neurophysiological studies on CNS neurons have demonstrated rapid modulation of neuronal excitability, including a brief hyperpolarization and increase in potassium conductance of the postsynaptic membranes of medial amygdala neurones (Nabekura et al., 1986), increase the firing rate of medial preoptic and septal neurones (Kelly et al., 1978), and increase the cerebellar neuronal responsiveness to iontophoretically applied glutamate (Smith et al., 1987) after application of physiological levels of 17 P-E2. Second, biochemical studies have also provided evidence for the direct actions of estrogens on neuronal membranes. This included an increase in amphetamine-stimulated striatal dopamine release in vitro superfusion system with 17 P-E2 or diethylstilbesterol, but not with 17 a-E2 (Becker, 1990) as well as in vivo microdialysis in freely moving rats (Becker & Beer, 1986). Also, 17 P-E2 is shown to enhance the responses of adenylate cyclase to biogenic amines in striatal neurons culture (Maus et al., 1989). Finally, morphological experiments have also demonstrated rapid plasma membrane ultrastructural modifications in response to sex steroids application (Garcia-Segura et al., 1987, 1989). Employing freeze-fracture techniques, within 1 min, physiological concentrations of 17 P-E2 increased the density of exoendocytotic pits in cerebrocortical and hypothalamic neuronal membranes in culture, which was blocked by estrogen antagonist tamoxifen (Garcia-Segura et al., 1987, 1989). These rapid effects (from seconds to minutes of latency)





15


with estrogen hormone on neuronal membrane excitability (Garcia-Segura et al., 1987, 1989; Kelly et al., 1978; Nabekura et al., 1986; Smith et al., 1987) occur much too rapidly to be accounted for by new mRNA synthesis and translation into proteins. This has led some investigators to suggest a possible action of gonadal steroids directly on neuronal membrane function/components. Moreover, Pietras and Szego (1979) have reported an increase in cyclic AMP concentrations in uterus within 15 minutes after E2 treatment. Although actinomycin D (an RNA synthesis inhibitor) can effectively prevent the full expression of long-term E2 effects on target tissues, the rapid or short-term effects of E2 seem to be independent of RNA/protein synthesis. These effects are most likely mediated by membrane-associated E2 receptors (Pietras & Szego, 1979; Towle & Sze, 1983).


Estrogen Receptors


Intracellular/cytosolic receptors

The pioneering studies of Glascock and Hoekstra (1959) and of Jensen and Jacobson (1962) utilizing radiolabeled estrogen demonstrated selective localization and retention of the label in tissues known to be targets for estrogen action. Jensen and Jacobsen (1962) in their studies also demonstrated E2 binding to a specific cytosolic receptor protein. The application of sucrosedensity gradient centrifugation then led to further characterization of the cytosolic estrogen receptor (Toft & Gorski, 1966). Subsequent studies provided further evidence to satisfy the criteria for an E2 protein receptor. These criteria included the stereospecificity for E2 binding (Noteboom & Gorski, 1965), saturable or limited number of binding sites (Gorski et al., 1968; Noteboom & Gorski, 1965), size determination by gel filtration





16


chromatography (Gorski et al., 1968; O'Malley et al., 1969), sucrose gradient analysis (Jensen & DeSombre, 1973; Toft & Gorski, 1966), and sensitivity to heat and proteases but not to nucleases (Gorski et al., 1968; Noteboom & Gorski, 1965; O'Malley et al., 1969).

The question of the subcellular localization of the E2 receptors was then resolved by a number of experimental criteria. These included subcellular fractionation by differential centrifugation after 3H-E2 exposure in vivo or in vitro (Jensen & Jacobson, 1962; King et al., 1965, Mowles et al., 1971) which provided early evidence for nuclear localization of E2 receptors. Conversely, E2 receptors remained in the soluble, high-speed cytosol fraction in tissues which were not previously exposed to the E2 hormone (Jensen et al., 1968, 1973). Taken together, available data indicate that the unoccupied receptors migrate between both the cytoplasmic and nuclear compartments but are believed to be primarily concentrated in the nuclei in a reversible equilibrium binding state with the nuclear components. Binding of E2 to the cytosolic receptor results in a transient biologically active occupied transformed receptor. Then the transformed receptor translocates into the nucleus with an enhanced affinity for the nuclear acceptor sites, favoring receptor binding to the acceptor site on the DNA component. This triggers gene expression (changes in mRNA synthesis and modification) and the expression of proteins in the target cell. These effects are generally observed with a delay of several hours or even days between the exposure of the target tissue to E2 and the first detectable changes in cellular function. Membrane receptors

Numerous documented reports have emerged to indicate that the rapid actions (from seconds to minutes of latency) of gonadal steroids on





17


some target tissues may be caused by direct interaction with plasma membrane receptor/effector components (Becker, 1990; Garcia-Segura et al., 1987, 1989; Kelly et al., 1978; Majewska, 1987; Nabekura et al., 1986; Smith et al., 1987; Towle & Sze, 1983). In fact, biochemical studies have demonstrated specific binding sites for sex steroids in synaptosomal plasma membranes prepared from the rat brain (Towle & Sze, 1983). Furthermore, the presence of steroid binding sites have also been demonstrated on plasma membranes of other target tissues as well, including liver (Suyemitsu & Terrayama, 1975), pituitary (Koch et al., 1977), and uterus (Pietras & Szego, 1979). In all these instances, the exact physiological/pharmacological function of the membrane binding sites for steroids has yet to be determined. However, these binding observations are compatible with the rapid non-genomic effects of estrogen, which are not easily accommodated within the genomic model (McEwen et al., 1982, 1984; Majewska, 1987). Collectively, the presence, if real, of these speculative membrane receptors can account for the rapid neurotropic effects of E2. Furthermore, these receptors may be involved in the modification of CNS neurotransmission.

Estrogen-receptor binding kinetics

Estrogen target tissues, i.e., brain, anterior pituitary, uterus, etc., apparently contain a single, specific estrogen binding component (type I estrogen receptor). The unoccupied receptors migrate between both the cytoplasmic and nuclear components (Walters, 1985). The cytosolic receptors bind E2 with high affinity such that the steroid-receptor complex remains intact for translocation into the nucleus (Towle & Sze, 1983). In the rat uterus, estrogen-filled binding sites do not undergo detectable degradation over a 24-hr period at temperature up to 30*C (Walters, 1985). Following a





18


single injection of E2, the translocation of cytosolic estrogen receptor (ERc) to the nucleus has been reported to be nearly complete within 1 hr in the rat uterus (Jakesz et al., 1983). When ERc was assessed by exchange assay 6 hrs after hormone administration, ERc levels continued to remain very low. However, an increase in nuclear estrogen receptor (ERn) was concomitantly observed following E2 injection, reaching maximal levels after 1 hr (Jakesz et al., 1983). Following an apparently near quantitative translocation of ERc to the nucleus, ERn concentrations declined to 30% after 6 hrs. However, with repeated injections of estrogen which maintained continuous receptor saturating concentrations of [3H]E2 over a 6-hr period, conservation of total cellular receptors in both cytoplasmic and nuclear fractions were observed (Jakesz et al., 1983). It is important, however, to note that under continuous steroid exposure qualitative changes in receptor properties (down regulation) occur over time in both cytosol and nuclear compartments. It is thought that the ERc present at 6 hr after estrogen administration originate from a replenished pool of receptors. This replenished pool has been reported to be partially dependent on protein synthesis. However, inhibition of protein synthesis by cycloheximide did not inhibit replenishment after estrogen exposure. Thus, estrogen target tissues, particularly uterus may represent a system in which estrogen receptors replenishment appears to be due entirely to receptor recycling (Jakesz et al., 1983).

Determination of the kinetic binding parameters indicated a high
affinity estrogen-binding site (Kd = 10-10 M) for brain ERc and a Bmax of 3 fmol/mg protein (Walters, 1985). These binding parameters are similar to estrogen binding kinetics of other tissues. Rate of association with the receptor has been reported to be 4.4 x 105 M-1 S-1 while rate of dissociation was

2.4 x 105 M-1 S-1 with t1/2 = 80 hrs at 0 to 4*C. Half-life for the clearance of





19


nuclear estrogen-receptor complexs were estimated to be 2 hrs for 17 P-E2 (Walters, 1985).

Finally, the maximum biological responses seemed to be determined not by the level of hormone binding to cytosol receptors, but by retention of a small proportion (10 to 15%) of the receptors on a limited number of saturable estrogen receptor-binding sites (acceptor sites) in the nuclear DNA (Walters, 1985). This is supported by the observation of a good correlation between the duration of nuclear occupancy and uterine growth stimulation in a series of short and long acting estrogen and their derivatives.

Role of Estrogen in the Menstrual Cycle

During the normal reproductive life (30 to 40 years), a female

menstruates 300 to 500 times (Tepperman, 1981). The menstrual cycle is characterized by monthly rhythmic changes in the secretion pattern of the reproductive hormones and the corresponding changes in the sexual organs as well. The duration of the cycle averages 28 days. The significance of the menstrual cycle are (1) maturation of an ovum and ovulation; (2) preparation of the uterine endometrium for implantation of an embryo; and (3) expression of the secondary sex characteristics associated with the procreative act. The coordination among these events is achieved by precisely timed fluctuation in the production and secretion rates of a number of hormones associated with the hypothalamic-pituitary-ovarian axis. In humans and primates the ovarian E2 is perhaps the major driving force for the initiation/maintenance of the cycle. This implies that the primary stimulus in triggering the initiation of the preovulatory gonadotropin surge, the central event in the cycle, is caused by the rise in E2 levels during the late





20


follicular phase. The dynamic changes of the ovaries have a periodicity of once every 28 days; however, these events are, in turn, regulated by the cyclic changes of pituitary gonadotropins (Ross, 1985; Schwartz, 1981; Yen, 1978). The pattern of gonadotropin secretion is, in turn, maintained by the pulsatile mode of the hypothalamic LHRH secretion (Levine & Ramirez, 1982; Marshal & Kelch, 1986). Finally, the whole cascade of events is regulated by negative and positive feedback effects of ovarian hormones (McCann, 1982; Pohl & Knobil, 1982). In a sense, the dynamic interplay between the brain and pituitary and the dynamics of the ovarian feedback mechanisms govern the reproductive cycle. That is, the momentum gained from one phase of the cycle powers the next phase, and continues on into the next cycle. Feedback regulation

The hypothalamic-pituitary-ovarian-endometrial axis in females changes markedly with the menstrual cycle. In humans and nonhuman primates, the cyclic pattern of gonadotropin secretion (the LH surge particularly) seems to depend more on ovarian estrogen than neural signals. This is evidenced by the observation that menstrual cycles in primates were maintained with a constant dose of LHRH per pulse with no variation in amplitude or frequency of LHRH administration (Knobil, 1980). These observations suggested that the hypothalamus may only play a permissive role in the control of the menstrual cycle (Knobil, 1980). However, more recent observations in human and nonhuman primates indicate that normal menstrual cyclicity does depend on LHRH pulse parameter that changes in frequency and amplitude during the menstrual cycle (Dalkin et al. 1989; Ferin et al., 1984; Marshal & Kelch, 1986). Studies in animals as well as in humans have shown that LHRH and gonadotropins are secreted in episodic fashion





21


(Levine & Ramirez, 1982), and the dynamics of the LHRH pulsatile mode is essential for the differential synthesis and release of gonadotropins (Marshal & Kelch, 1986). Furthermore, ovarian hormones concurrently modulate LHRH pulse parameters and, therefore, are important regulators of gonadotropin secretion. For instance, the preovulatory rise in E2 increases the LH pulse frequency, and since there is a good concordance between LH pulses in plasma and LHRH pulses in hypothalamic portal blood, this indicates that a transient elevation in E2 levels increases the frequency of LHRH secretion (Marshal & Kelch, 1986). The feedback effects of E2 not only modulate the release of LHRH from the hypothalamus but also the responsiveness of the pituitary to LHRH signals (Marshal & Kelch, 1986). Therefore, the effects of E2 on the gonadotropins LH and FSH are coordinated at the levels of both the hypothalamus and pituitary.
Estradiol displays both inhibitory and stimulatory effects on the

secretion of these hormones. Inhibitory, or negative feedback effects, are seen during periods of basal LH secretion throughout the menstrual cycle. During the follicular phase, the follicle secretes low levels of E2, and during the luteal phase, the corpus luteum secretes large quantities of both estrogen and progesterone. The combined effects of these steroids inhibit the secretion of LHRH and consequently reduce the release of pituitary gonadotropins. However, the positive feedback effects, or stimulation, are observed after a transient and progressive increase in the titers of E2. That is, an increase in serum E2 levels of 150 pg/ml or greater for 24 to 36 hours during the late follicular phase of the menstrual cycle (Ross, 1981). This occurs in response to the increase in LH pulse frequency which subsequently results in the preovulatory surge of LH. This condition subsequently causes ovulation of the ovum from the Graafian follicle and the formation of a new corpus





22


luteum. In the absence of gonadal steroids such as following ovariectomy or menopause, the negative feedback effects of estrogens on LHRH and gonadotropins are removed and thus, serum LH and FSH levels increase. From the clinical point of view, constant exposure to estrogens (or maintenance of their elevated levels) prevents the preovulatory surge of LH by exerting a negative feedback mechanism on the hypothalamic-pituitary unit. This strategy represents the primary mechanism by which the E2 component of contraceptives prevents ovulation. Collectively, the effects of E2 on the hypothalamic-pituitary unit depend on the exact duration or magnitude (or both) of exposure to the hormone.


Role of Estrogen in the Rat Estrous Cycle

Rats have 4 to 5 days of estrous cycle (Long & Evans, 1922). The estrous cycle, like the menstrual cycle in human and nonhuman primates, represents an extraordinary sequence of events in hormonal and behavioral changes, and it is verified by cyclic changes in vaginal cell morphology (Long & Evans, 1922). The normal cycle consists of one day of estrus, followed by two days of diestrus (I and II), and one day of proestrus. The dynamic relation between the hypothalamic releasing factor, pituitary gonadotropins, and ovarian steroids during the estrous cycle have been reviewed in considerable detail (Kalra & Kalra, 1983). Estrogen hormones are responsible for the maintenance of the estrous cycle, as in the menstrual cycle. On the evening of estrus, the serum E2 concentrations reach their lowest levels (15 to 20 pg/ml) as the corpus luteum involutes. However, during the days of diestrus as follicular growth and maturation progress under the influence of FSH stimulation, ovaries produce increasing concentrations of estrogens. E2 levels





23


peak to 40 to 80 pg/ml on the evening of the proestrus day of the cycle (Butcher et al., 1974; Kalra & Kalra, 1977). E2 levels then decline rapidly to basal levels on estrus. The production of E2 by the follicles of the ovaries is controlled via feedback mechanism (Richards & Hedin, 1988). Feedback regulation

Ovarian cycle of the rat, like the human, is regulated by the pituitary

gonadotropins LH and FSH. However, in the rat, the gonadotropins appear to be primarily under the stimulatory influence of the hypothalamic LHRH neuronal activity (Kalra & Kalra, 1980, 1983; McCann, 1982). This is evidenced by the fact that, in ovariectomized steroid-treated rat, the positive feedback effects of estrogen are expressed as a daily signal for mid-afternoon LH hypersecretion which ensues for several days if concentrations of E2 are maintained at or greater than E2 levels seen on the proestrus day (Legan et al., 1975). As in the human and primate, the feedback effects of E2 on the gonadotropin secretion are coordinated at the levels of both the hypothalamus and the anterior pituitary gland (Kalra & Kalra, 1983, 1989; plant, 1986).

In the rat estrous cycle, E2 also exhibits both inhibitory and stimulatory effects on the hypothalamic-pituitary unit. The inhibitory, negative feedback, effects are observed during periods of basal LH and FSH secretion throughout the estrous cycle or after chronic exposure of ovariectomized rats to E2. The inhibitory effects of E2 are exerted within the hypothalamus (the medial basal and medial preoptic areas) and/or the anterior pituitary. Implantation of E2 in the medial basal hypothalamus (MBH) (Smith & Davidson, 1974) or chronic elevation in serum E2 levels (Henderson et al., 1977) were shown to suppress LH levels and increase the pituitary responsiveness to LHRH,





24


suggesting that the primary site of E2 action is perhaps within the MBH. However, the stimulatory effects, or positive feedback, are observed after a transient increase in follicular estrogen secretion from diestrus II through proestrus or after more than 48 hours of sustained E2 exposure to ovariectomized rats (Kalra & Kalra, 1983; 1989; Legan et al., 1975). The positive feedback action of E2 in the estrous cycle is believed to be exerted primarily on the medial preoptic area of the ventral diencephalon since lesioning of this area (Wiegand et al., 1980), or interuption of its connections to the MBH (Halasz & Gorski, 1967), or implantation of antiestrogen clomiphene into this region (Docke et al., 1989) completely prevent the LH surge and ovulation in the rat. The preovulatory rise of E2 increases the hypothalamic LHRH mRNA (Rosie et al., 1990), the LIRH pulse frequency and pulse amplitude at mid-cycle (Dalkin et al., 1989; Marshal & Kelch, 1986). However, at the pituitary level E2 increases the responsiveness of the gonadotrophs to LHRH on the afternoon of proestrus. This allows a priming effect and thus augments the response of the pituitary to subsequent LHRH messages. The increase in pituitary responsiveness must precede the hypothalamic LHRH message for the LH surge to occur.





25


Blood-Brain Barrier


Historical Overview

The blood-brain barrier is vital to brain function. Neurons are

extremely sensitive to the ratio of the concentrations of ions across their plasma membranes. Furthermore, the concentrations of excitatory and inhibitory substances in the extracellular neuronal environment must be tightly regulated for optimal brain functioning. Perhaps because of the functional intricacy of neurons, the mammalian CNS acquired this ultrastable cellular environment to perform effectively. This stable extracellular environment is achieved by several morphological and enzymatic components collectively referred to as the blood-brain barrier (BBB). The BBB is not an absolute barrier, but rather a selective and protective barrier (Neuwelt, 1989; Rapoport, 1976; Suckling et al., 1986). Many pharmacological agents are excluded from entering the brain because of the existence of BBB (Bodor & Brewster, 1983; Schanker, 1965).

The concept of barrier was originally postulated by Ehrlich at the end of the 19th Century (Levin, 1977; Pardridge et al., 1975; van Deurs, 1980). When vital dyes, such as trypan blue, were injected into the bloodstream, the dyes penetrated almost all organs of the body but did not enter the brain tissue or the cerebral ventricles (Levin, 1977; Pardridge et al., 1975; van Deurs, 1980). However, when dyes were injected into the cerebral ventricles, brain tissue was stained and the dye did not readily enter the bloodstream. Thus, the lack of staining by the dyes was not an intrinsic property of brain tissue, rather the existence of a dynamic barrier interface between the blood and the brain.





26


Subsequent studies which utilized different compounds, including drugs and radioactive tracers further substantiated the concept of BBB. Furthermore, it was later discovered that many small molecules were similarly excluded from transport into the brain. This, then, led to the suggestion that BBB is absolute; a concept which was soon thereafter refuted when the nutrient requirements of the brain were elucidated (Davson, 1976).

Ultrastructural studies have shown that there are several differences between the systemic capillaries and the cerebral capillaries which explain their permeability differences. Morphologically, the ultrastructural feature which most distinguishes CNS microvessels is the endothelial cell lining of the these vessels (Brightman, 1977). The morphological features of the brain capillaries were elucidated using horseradish peroxidase (HRP). This relatively small enzyme has a high affinity for radiopaque substances, such as osmium tetroxide which can be visualized, and it does not cross cerebral capillaries (Brightman, 1977). When introduced directly into the brain, it readily diffused throughout the extracellular space but did not pass between the endothelial cells of cerebral capillaries (Reese & Karnovsky, 1967). So the anatomical basis for the BBB was identified as the endothelial lining of the cerebral capillaries. Unlike systemic endothelial capillaries, the cerebral counterparts are joined by tight junctions (Brightman & Reese, 1969). These tight junctions form a zona occludens and provide, for molecules like HRP, an absolute barrier. Morphologically, these junctions consist of aligned intramembranous ridges and grooves which are in close apposition (Oldendorf, 1977; Shivers, 1979). Two additional morphological features of the cerebral capillaries contribute to the BBB as well: (i) cerebral endothelial cells have a paucity of vesicles and of vesicular transport feature and (ii) the perivascular





27


astrocytic endfeet seem to be involved in the regulation of nutrients flux and uptake of substances from the circulation (Broadwell & Salcmen, 1981).

In addition to the aforementioned structural features contributing to
the BBB, the presence of various enzymes in and around the endothelial cells of cerebral capillaries may play a vital role in limiting and perhaps protecting the brain from a variety of blood-borne substances (Hardebo & Owman, 1980). Thus, the presence of catechol-o-methyltransferase, monoamine oxidase, aromatic amino acid decarboxylase and gamma-aminobutyric acid transaminase in the vicinity of cerebral vasculature may restrict the entry of various blood-borne chemicals, i.e., neurotransmitters/neuromodulators or therapeutic agents into the CNS. Such a protective mechanism against circulating neuroactive substances is essential since optimal CNS function requires a delicate balance between neurotransmitter release, metabolism and uptake in the vicinity of neurons. Finally, the enzymatic component of the BBB may also play a role in excluding some of the lipophilic compounds from the CNS which otherwise might passively diffuse through the barrier.

Regarding the nutrient requirements of the brain, there are numerous specialized carrier transport systems which are localized within the BBB for uptake of nutrients from the circulation (Fenstermacher, 1985; Pardridge, 1981, 1983, 1986, 1987). These include, specific carrier-mediated transport systems for numerous classes of nutrients (Pardridge, 1981, 1983), receptormediated transport mechanisms for plasma proteins and peptides (Pardridge, 1986), and plasma protein-mediated transport of protein-bound substances (Pardridge, 1988a, 1988b). These transport systems are localized on the luminal (or blood side) as well as on the antiluminal border (or brain side) of the BBB (Pardridge, 1988a, 1988b), and are characterized by saturability and specificity. These transport mechanisms of the BBB, therefore, provide





28


means for bidirectional movement of selective molecules. However, with few exceptions, these carrier systems are not involved in transport of chemotherapeutic agents across the BBB (Greig et al., 1987).


Potential Asset to Utilize in the Design of Brain-Specific Drug Delivery

The unique architecture of the BBB allows only the transport of
compounds either by specific transport systems or by simple diffusion directly through cell membranes if they are to gain access to the brain parenchyma and extracellular spaces. Therapeutic agents are no exception. They can access the brain through either of these routes. Furthermore, the bulk transport of materials is limited due to the sealing of endothelial gap junctions and the lack of vesicular transport system in the cerebral capillaries. As a result, most drugs that enter the CNS must do so by passive diffusion through the phospholipid cellular matrix of capillary endothelial cells. The lipophilicity of drugs, defined by their octanol-water partition coefficient, correlates with their ability to penetrate the BBB for several classes of drugs, including narcotics (Oldendorf et al.,1972), barbiturates (Levin, 1980) and P-receptor antagonists (Cruickshank et al., 1979). Furthermore, drugs which cannot penetrate the BBB can gain access to limited areas of the brain around the circumventricular organs. Collectively, the CNS has evolved mechanisms to protect itself by excluding hydrophilic (polar substances) and other compounds which may be harmful to its optimal functioning. Unfortunately, this barrier mechanism impedes the delivery and transport of many potentially useful therapeutic agents to the CNS, thus severely complicating the effective treatment of brain diseases.





29


A strategy that could achieve an improved delivery of drugs to the brain with sustained release in that tissue would be of great advantage. A general approach to increase brain concentration of drugs and thus, their effectiveness has been the design of lipid soluble prodrug from water soluble drugs (Bodor, 1981, 1985; Bodor & Kaminski, 1987; Sinkula & Yalkowsky, 1975; Stella, 1975). Prodrugs are pharmacological agents which have been transiently modified to improve their lipophilicity as well as to hinder their rapid metabolic inactivation via enzymes. Ideally, the prodrug is biologically inactive but reverts to the active, parent drug in vivo at, or around, the site of action. This transformation can be mediated by an enzyme or may occur chemically as a result of designed instability in the structure of the prodrug. The purpose of prodrug modification is to increase the concentration of the active drug at or near its site of action, thereby increasing its potency/efficacy. By temporary masking the polar groups of a drug, the lipophilicity of the drug is increased and thus, its ability to enter the brain parenchyma is enhanced. Once in the brain, hydrolysis of the masking groups will release the active drug.

Nevertheless, potential problems are associated with the prodrug
approach (Gorrod, 1980). For instance, by increasing the lipophilicity of a drug via the prodrug approach, it may not only improve its diffusion through the BBB to gain access to the CNS, but also ensures the uptake of the compound into all other tissues and thus, exposure to a greater drug burden. This is a major limiting factor in the use of prodrugs, especially those with cytotoxicity, i.e. antineoplastic agents, or those with broad spectrum of peripheral site of actions such as steroids. Furthermore, even if enhanced CNS delivery/uptake is achieved via the prodrug approach, the efflux of the drug





30


is concurrently enhanced from the CNS. This results in poor retention and minimal or no improvement in the biological half-life of the drug.

To overcome the problem of potential general toxicity associated with enhanced lipophilicity of prodrugs, novel redox-based chemical delivery system (CDS) for drugs has been designed which exploits the unique architecture of the CNS BBB (Bodor, 1987; Bodor & Farag, 1983, 1984; Bodor & Simpkins, 1983; Bodor et al., 1981, 1987, 1988). By definition, a CDS is a biologically inert molecule which requires several chemical conversions leading to the active, parent drug at or near the site of action (Bodor, 1987; Bodor & Brewster, 1983). The multiple, facile chemical conversions may lead to (a) selectivity in drug delivery; (b) improve the drug half-life; and (c) decrease the toxicity of the drug. The redox-based CDS utilizes a carrier molecule that can exist as a lipid soluble (in the reduced state) or water soluble (in the oxidized state). The mechanism of its drug delivery is based upon an interconvertible dihydropyridine 4> pyridinium ion carrier (Bodor, 1987). In this brain-specific CDS, the lipoidal dihydropyridine moiety is attached to the drug, thus increasing its lipid solubility and thereby enhancing its permeability through the BBB. The reduced dihydropyridine can be oxidized, after its administration, to the pyridinium ion in the brain as well as in the periphery including systemic circulation. The charged pyridiniumdrug complex is thus locked into the brain while the same moiety rapidly clears from the periphery by renal or biliary processes due to its increased hydrophilicity. Sustained release of the active, parent drug from the charged pyridinium-drug complex occurs in the brain as a result of the enzymatic hydrolysis of the ester (or amide, etc.) linkage between the drug and the pyridinium moiety.





31


Collectively, the ability to preferentially deliver and sustain the release of a drug in the brain, thus sparing non-target site tissues, should improve the therapeutic index of the drug by (i) increasing the concentrations and/or residence time of the drug at its receptor site in the brain and (ii), equally important, decreasing the concentrations and/or residence time of the drug at the potential peripheral sites of toxicities, thereby decreasing its untoward side effects. Furthermore, this approach may be potentially advantageous in the treatment of brain diseases by virtue of the need for lower or less frequent doses of the drug.

This redox-based CDS has been applied successfully to brain-specific

delivery of a wide variety of therapeutic agents, including phenylethylamine (Bodor et al., 1981; Bodor & Farag, 1983), dopamine (Bodor & Simpkins, 1983; Simpkins & Bodor, 1985), gamma-aminobutyric acid (Anderson et al., 1987b), P-adrenergic blocking agents (Bodor et al., 1988), antitumor drugs (Bodor & Brewster, 1983), antiviral agents and antibiotics (Bodor & Brewster, 1983), testosterone (Bodor & Farag, 1984), estradiol (Bodor et al., 1987), and norethindrone (Brewster et al., 1987).

The application of this redox-based CDS to estrogens, particularly E2
(Bodor et al., 1987), has important clinical and research implications since the hormone plays major role in the reproductive and nonreproductive functions by influencing the brain. Estrogens are intrinsically lipophilic and readily enter the CNS; however, when inside the CNS, there is no mechanism to prevent their redistribution back to the periphery and thus exhibit poor retention. Furthermore, because of their inherent lipophilicity, estrogens equilibrate among all body tissues. This property of the steroid necessitates either frequent dosing or the administration of a depot form of the estrogen in order to maintain therapeutically effective concentrations in





32


the brain. Both of these treatment strategies lead to sustained increases in peripheral estrogen levels. However, when estrogen is attached to the dihydropyridine carrier, the E2-CDS enhances brain-specific delivery of estrogen by (i) locking the estrogen into the brain following the oxidation of E2-CDS to the charged pyridinium moiety (E2-Q+); and (ii) enhancing the rate of elimination of the lipoidal estrogen, in the inactive E2-Q+ form, from the periphery following its oxidation to the charged, more hydrophilic compound.


Therapies Aimed at Targeting/Enhancing Brain Estradiol Levels


Fertility Regulation

Fertility control may be achieved by a wide variety of mechanical,
surgical, and chemical methods. The chemical (steroidal) methods of fertility control was first introduced by Pincus and Chang (quoted by Tepperman, 1981) and since then it has had important repercussions on population growth. Commonly used steroid contraceptives consist of synthetic estrogens in combination with progestins. When given at pharmacological doses and/or constant exposure, E2 inhibits (via negative-feedback mechanism) the secretion of gonadotropin hormone-releasing hormone (GnRH) from the hypothalamus (Kalra & Kalra, 1983, 1989; Plant, 1986) and hence, of gonadotropins (LH and FSH) from the anterior pituitary (Kalra & Kalra, 1989). The inhibition of the hypothalamic-pituitary-ovarian axis prevents follicular development and therefore ovulation (Briggs, 1976). However, the use of oral contraceptives has been associated with many adverse metabolic changes, including increased risk of coronary atherosclerosis, myocardial infarction in





33


smokers, liver tumors, hypertension, and changes in glucose metabolism that appear to be estrogen related (Drill & Calhoun, 1972; Firsch & Frank, 1977; Fotherby, 1985; Inman et al., 1970; Kaplan, 1978; Mays et al., 1976; Thomas, 1988). To reduce the magnitude and the spectrum of these dose-related adverse effects of estrogen thus would be to reduce the dose (Bottiger et al., 1980) and/or to use a more natural hormone (Ottosson, 1984). Given the combined decrease in the contraceptive components, there is the possibility that suppression of the hypothalamic-pituitary-ovarian function may not be as effective as with higher dose formulations. Therefore, preferential brain delivery of E2 with the CDS may provide an effective, long-acting contraceptive by virtue of sustained release of E2 in that tissue. Furthermore, the adverse peripheral effects associated with the currently used contraceptive steroids may be avoided by lowering the dose or the frequency of ingestion.


Menopausal Syndrome

The cessation of menses, menopause, near the age of 50 is the result of the decreasing production of ovarian estrogens/progestins (Notelovitz, 1986). This loss of ovarian hormones in 75% to 85% of women leads to a number of brain-mediated steroid-withdrawal symptoms (Casper & Yen, 1985; Lauritzen, 1973; Yen, 1977), the most frequent being hot flushes (Clayden et al., 1974; Meldrum et al., 1979). These (patho)physiological alterations appear to be the result of autonomic discharge which causes peripheral vasodilation and heat loss (Nesheim & Saetre, 1982). Replacement therapy with estrogens and/or progestins has been shown to be effective in most menopausal patients in alleviating the symptoms of the disease (Campbell & Whitehead, 1977; Huppert, 1987; Upton, 1984). However, numerous retrospective studies





34


indicated an increased risk of peripheral toxicities, including the risk of breast and endometrial cancer (Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger et al., 1988; Persson, 1985; Trapido et al., 1984), cardiovascular morbidity (Barrett-Conner et al., 1989; Kaplan, 1978; Thomas, 1988), and interference with hepatic metabolism (Burkman, 1988). These adverse effects of estrogens are dose dependent. Currently used estrogens are administered either in frequent doses, or as a depot form, in order to maintain therapeutically effective levels in the brain. Both of these treatment strategies lead to sustained increases in peripheral estrogen levels and thus peripheral toxicities. Since currently employed estrogen therapy is contraindicated in many postmenopausal women, and in as much as some women do not respond to the existing steroid medications, the brain-enhanced E2-CDS with sustained release of E2 may be more effective in the treatment of menopausal symptoms by providing sufficient E2 to the brain while avoiding peripheral toxicities.


Prostatic Cancer

The primary objective of hormone therapy in prostate cancer patients is to induce an effective androgen suppression, thus abolishing the growth promoting effects of androgens on the diseased prostate ( Brendler, 1988; Cabot, 1896; Huggins & Hodges, 1941; Isaacs et al., 1983; Moore, 1944; White, 1895). Currently a variety of surgical and therapeutic means for inhibiting androgen production (in the testis) or blocking androgen action (in the prostate) are being used. These include castration, high-dose estrogen therapy, GnRH analogues, and antiandrogens (Santen & English, 1989). Castration or high-dose estrogen therapy remain, however, as the treatment





35


of choice for the endocrine-dependent management of prostatic cancer (van Steenbrugge et al., 1988). Both treatments are reported to be equally effective in (i) suppressing the circulating testosterone levels (Carlstrom et al., 1989); and (ii) controlling the symptoms of advanced prostatic cancer in 70-80% of patients (Klein, 1979). In contrast to castration, high-dose estrogen therapy inhibits (via negative feedback mechanism) the hypothalamic-pituitarygonadal function leading to chemical castration. However, high-dose estrogen therapy has been shown to cause severe cardiovascular complications in patients (Henriksson & Edhag, 1986) due to the alterations in liver metabolism (von Schoultz et al., 1989). The E2-CDS may be potentially useful in the treatment of androgen-dependent prostatic cancer by the virtue of its sustained suppression of the hypothalamic-pituitary-testis function leading to chemical castration and thus regression of the tumorous tissue.


Body Weight Regulation

Food intake and body weight may vary during the estrous cycle of the rat (Tarttelin & Gorski, 1971) and the menstrual cycle of primates (Czaja, 1978) including women (Pliner & Fleming, 1983). A consistent observation is that food intake and body weight decrease during the follicular phase of the ovarian cycle when circulating E2 levels increase. Conversely, food intake and body weight increase during the luteal phase of the ovarian cycle when E2 levels decline and progesterone levels elevate.

Although relatively few studies have evaluated the potential
modulatory effects of E2 on body weight regulation in human subjects, the available data support a suppressive role of E2 in appetite and body weight. Morton et al. (1953) in a study involving menstruating women with





36


premenstrual syndrome (PMS) reported 37% had a craving for sweets during the luteal phase of their menstrual cycles. Increased appetite was reported to be a frequent PMS symptom in 45 women examined by Fortin et al. (as reported by Smith & Sauder, 1969). Pliner and Fleming (1983) studied 34 women and observed that the decreased food intake during the follicular phase was associated with significant weight loss and that the increased food intake during the luteal phase was concurrent with weight gain. Collectively, endogenous estrogen in women has consistently been shown, albeit subtle, to have a suppressive effect on food intake and body weight.

The potential feasibility of E2-CDS to reduce body weight has been extensively evaluated in the rat (Estes et al., 1988; Simpkins et al., 1988, 1989a,b). Interestingly, a separation between the effects of E2-CDS on body weight and food intake has been demonstrated. That is, despite weight reduction, no consistent reduction in food intake has been observed, indicating that mechanisms other than reduced appetite are responsible for the weight loss. Collectively, available data indicate that the E2-CDS chronically suppresses body weight following a single administration in the rat. Furthermore, the compensatory hyperphasia in response to the weight loss is prevented by the E2-CDS.


Libido/Sexual Dysfunction

Certain regions of the hypothalamus have been identified to be involved in the central integration of sexual behaviors (Christensen & Clemens, 1974). The first direct evidence that sex steroids influence the central structures associated with mediating and integrating sexual behaviors was presented when Harris implanted estrogen directly into the cat's





37


hypothalamus. Furthermore, hypothalamic lesions prevent sexual behavior in animals, even in the presence of adequate estrogen (Beyer et al., 1976).

Male sexual behaviors are composed of two distinct components: (1)

proception, or the awareness and pursuit of a receptive female and mounting to achieve intromission, and (2) comsumation, or penile erection, intromission and eventual ejaculation (Davidson, 1972). It is believed that the expression of the proceptive components are dependent upon the aromatization of testosterone to E2 in the brain, particularly in the regions of the preoptic area of the hypothalamus and the amygdala (Beyer et al., 1976; MacLusky et al., 1984). When E2 was implanted into the preoptic area of the hypothalamus, it effectively restored mounting and intromissions in the castrated rat (Beyer et al., 1976). Other studies have shown that the full restoration of masculine sexual behaviors in castrated rats require E2 action in the brain and dihydrotestosterone in the peripheral tissues (Lisk & Greenwald, 1983). Given the aforementioned evidence regarding brain mediation of sexual behaviors by E2, the E2-CDS may be potentially useful in the treatment of sexual dysfunctions or psychological impotence that are not caused by deficits in peripheral androgen-responsive tissue.


Disorders of Depression


It is generally believed that the underlying mechanism(s) of major

mood disorders (mania and depression) may include abnormal functions of monoamine transmission (Bunney & Garland, 1982; Leonard & Kuschinsky, 1982). Pharmacological evidence suggests that mania is the result of hyperactivity while depression is due to hypoactivity of monoamines. Depression (unipolar and bipolar) respond well to antidepressant drugs, such





38


as tricyclic antidepressants or monoamine oxidase inhibitors, which inhibit the monoamines re-uptake or their respective metabolism (Bunney & Garland, 1982; Leonard & Kuschinsky, 1982).

The biological basis for the involvement of estrogens in depression
comes from two gynecological problems: premenstrual and postmenopausal syndromes. The premenstrual syndrome (PMS) refers to the various mood changes in relation to the menstrual cycle that are experienced by a large population of fertile age women. When the relationship between symptom development and normal variations during the menstrual cycle was examined (Backstrom et al., 1985), a consistent observation was that there were very few negative symptoms, rather an increased sense of well being during the preovulatory E2 peak of the menstrual cycle. However, the maximum degree of symptoms or mood changes occurred during the luteal phase of the menstrual cycle when progesterone levels are increased. Clinical studies as early as in 1932 (Bowman & Bender) suggested a possible therapeutic role of estrogen in the treatment of depression. More recently, Klaiber et al. (1979) reported on a double-blind study performed to assess the therapeutic efficacy of estrogen in the treatment of severely depressed women. The estrogen treatment significantly decreased the degree of symptoms compared with placebo (Klaiber et al., 1979).

Since the 1930's, numerous other clinical studies have provided

evidence regarding the influence of estrogens on the well-being and mental performance in postmenopausal women. Hawkinson (1938) reported a significant improvement in menopausal symptoms including depression and anxiety. When subjective indices of moods in ovariectomized women were evaluated in response to estrogen replacement therapy by Rauramo (1975), estrogen treatment resulted in an elevation in moods to that which was





39


described as being close to normal conditions. A recent study conducted by Gerdis et al. (1982), using a variety of psychometric measures to estimate depression, reported that three weeks estrogen treatment (Premarin) in postmenopausal women significantly improved the symptoms of depression. Furthermore, DeLignieres and Vincens (1982) reported improvement of symptoms of depression, aggression, and anxiety in postmenopausal women that were treated percutaneously with E2 for three months. In another study (Klaiber et al., 1979), postmenopausal women with primary, recurrent unipolar depression and a history of unsuccessful therapy of their depression were treated with estrogen. The evaluation of their progress by the Hamilton Rating Scale for Depression indicated a dramatic improvement in mean Hamilton scores in some patients (Klaiber et al., 1979). These studies suggested that estrogens may have antidepressant activity in postmenopausal patients.
Another gynecological problem associated with gonadal steroid
withdrawal is the postpartum psychosis or depression. During the third trimester of pregnancy, E2 levels increase to about 10 to 40 ng/ml (~ 1000-fold increase over the follicular phase of the ovarian cycle) while progesterone levels increase to about 100 to 400 ng/ml (- 100- to 400-fold increase over the follicular phase) (Ross, 1985; Schwartz, 1981). After parturition, gonadal steroid levels fall precipitously and thus, it leads to CNS-steroid withdrawal. Although the etiology of postpartum depression is yet unknown, it is speculated that the decline in gonadal steroid levels after parturition may be the primary factor leading to postpartum depression. In severe cases, it has been reported that administration of estrogen to patients alliveated postpartum depression by suppressing lactation (Yalom et al., 1968). Perhaps, since estrogen hormones influence a variety of CNS functions, by modifying





40


central neurotransmitters/neuromodulators levels involved in mood, or enzymes necessary for their synthesis, or their receptor/effector systems, estrogen therapy may very well benefit the postpartum depressed patients (Luine et al., 1975; Maggi & Perez, 1985). Thus, there is reason to suggest the notion that estrogen therapy with the E2-CDS may be useful for the treatment of postpartum depression.

Various biochemical studies in the rat, examining the effect of estrogens on biogenic amines and their enzymes, strongly support the behavioral and symptomatic improvements which are observed in postmenopausal patients. Estrogens are reported to inhibit the re-uptake of norepinephrine in the rat brain (Luine et al., 1975) and decrease the activity of monoamine oxidase (Holzbauer & Youdim, 1973; Luine et al., 1975). Additionally, the activity of monoamine oxidase (MAO) in plasma of postmenopausal patients were also reduced in response to estrogen treatment (Klaiber et al., 1976). In fact, MAO activity is lower in women during the preovulatory phase of the menstrual cycle when E2 levels are highest. In contrast, MAO activity increases during the luteal phase when progesterone levels are highest (Klaiber et al., 1976).

Collectively, the evidence mentioned above support the notion that estrogen therapy may have a rational scientific basis for treatment of depression associated with the decline in endogenous estrogen production. Estrogen treatment of these conditions may influence brain function via effects on a number of neurotransmitter systems involved in mood and other emotional behaviors.





41


Cognitive Impairment of Menopausal Alzheimer's Type

Since the 1950's, several lines of evidence have accrued to suggest that estrogen hormones may influence certain cognitive functions of the female. First, the decreasing production of endogenous estrogens/progestins after menopause or ovariectomy has been shown to cause changes in cognitive functions, especially in memory, that are prominent among the somatic and behavioral symptoms of menopause (Kopera, 1973; Malleson, 1953; Lauritzen & van Keep, 1978). Second, estrogen treatment of menopausal women with senile dementia of Alzheimer's type is shown to improve both symptomatic treatment and prevents or slows the progression of dementia in these patients (Campbell & Whitehead, 1977; Fedor-Freybergh, 1977; Fillit et al., 1986; Furuhjelm & Fedor-Freybergh, 1976; Hackman & Galbraith, 1976; Honjo et al., 1989; Michael et al., 1970; Sherwin, 1988; Vanhulle & Demol, 1976). Third, numerous biochemical studies have demonstrated that estrogen hormones modulate/enhance cholinergic transmission or activity in regions of the brain that are important for cognitive functions (Eleftheriou & Dobson, 1972; Iramain et al., 1980; Luine et al., 1975, 1980, 1983, 1985; O'Malley et al., 1987). Fourth, estrogen receptors have been identified in nuclei of the basal forebrain structures in the rat (Luine et al., 1975; Morrel et al., 1975; Pfaff & Keiner, 1973), the major loci of cell bodies of cholinergic neurons which innervate the cerebral cortex, limbic system, hippocampus and hypothalamus. These brain regions are believed to be involved in the pathology of Alzheimer's Disease (Coyle et al., 1983). Finally, perhaps the most interesting observation, is that the female:male sex ratio in the prevalence of Alzheimer's Disease is 2:1 (Sulkava et al., 1985).





42


Taken together the findings reported in the literature suggest a strong link among estrogen, the cholinergic system, and cognitive functioning in women. Using different estrogen preparations and various Memory Tests/instruments to assess cognitive functioning, results indicated that estrogen therapy resulted in improvement in attention span, orientation, memory, mood, and social interaction in a majority of postmenopausal women with cognitive impairments (Fillet et al., 1986; Furuhjelm & FedorFreyberg, 1976; Hackman & Galbraith, 1976; Honjo et al., 1989; Vanhulle & Demol, 1976). Furthermore, the improvement in cognitive functioning, in those patients who benefited, was correlated with an increase in circulating concentrations of estrogens (Sherwin, 1988). Additionally, biochemical data obtained from the rat brain further support the involvement of estrogens in cognition. Studies in the rat brain have shown that cholinergic neurons respond to administration of estrogen by (1) increasing the activity of choline acetyltransferase (ChAT, the enzyme that synthesizes acetylcholine) (Eleftheriou & Dobson, 1972; Iramain et al., 1980; Kaufman et al., 1988; Luine et al., 1975, 1980, 1983, 1985; ) in the basal forebrain, cortex, hippocampus and hypothalamus; (2) increasing high affinity choline uptake as well as acetylcholine synthesis (O'Malley et al., 1987) in cerebral cortex; and furthermore (3) the responses of cholinergic system to estrogen administration (i.e., enzyme activity) were observed in female but not male rats and positively related to the dose of E2 and blocked by an estrogen antagonist (Luine et al., 1980, 1983). Finally, E2 application has been shown to enhance the excitatory actions of Glu, and Glu is known to be essential in long-term potentiation and memory formation (Smith et al., 1987).

Collectively, available data support the role of estrogens in cognitive functioning. However, the hypothesis does not exclude the possibility of the





43


involvement of other neurotransmitter systems and estrogens in memory. Thus, if estrogens serve to maintain or enhance the activity of cholinergic neurons or serve as a trophic substance which directly or indirectly acts on cholinergic neurons, the idea of enhanced brain exposure to E2 may be an improvement in cognition. Further, since any therapy which is aimed at treating Alzheimer's type dementia must be chronic in its application to the patient, the sustained release of E2 from the E2-CDS is an additional useful benefit. As such, a careful evaluation of the E2-CDS for the mechanism by which it improves cholinergic function and for its potential application to Alzheimer's Disease patients is clearly warranted.













CHAPTER 3
GENERAL MATERIALS AND METHODS


Drugs and Solutions


Estradiol and Standard Solution

Estradiol-17P (E2) was purchased from Steraloids, Inc. (Wilton, NH). Standard solutions of E2 were prepared in ethanol for the in vitro studies involving methodology development. Solutions of E2 were stored at -20*C (stock solution) or 4*C (working standard).

Estradiol-175 incorporated in 2-hydroxypropyl-@-cyclodextrin (HPCD) was provided by the Pharmatec, Inc. (Alachua, FL). Aqueous solution of E2 was prepared on the day of experiment in 20% HPCD (wt:vol) for in vivo injection.


Estradiol-Chemical Delivery System

Estradiol-chemical delivery system, E2-CDS (3-hydroxy-170-{[(1-methyl1,4-dihydropyridine-3-yl)carbonyl] oxy}-estra-1,3,5-(10)-triene) and E2-Q+ (1methyl-3-{[(3-hydroxyestra-1,3,5-(10)-triene-17p-yl)oxy]carbonyl} pyridinium iodide) were synthesized as previously reported (Bodor et al., 1987). Briefly, the 3, 17p-dinicotinate ester of E2 was made by refluxing 1713-E2 with nicotinoyl chloride or nicotinic anhydride in pyridine. This derivative was selectively hydrolyzed to the 17-monoester of E2 with potassium bicarbonate in 95% methanol. The monoester of E2 was then quaternized with methyl


44





45


iodide. The delivery system, E2-CDS, was then prepared by reduction of the obtained E2-Q+ with Na2S2O4. The structure of each intermediate and the final product (E2-CDS) was confirmed by the nuclear magnetic resonance and elemental analysis: mp 115-130*C. The yields at each synthetic step were 6494%. Solutions of the E2-CDS in water containing 20% 2-hydroxypropyl-Pcyclodextrin (wt:vol) were prepared for injection. For E2-Q+, the working standard solutions were prepared in water/acetone (80:20;vol:vol) to be used in the in vitro methodology development.


Estradiol Pellet

Pellets, weighing 100 mg each, were prepared from crystalline E2 and

cholestrol (CHOL) powder. Both E2 and CHOL were thoroughly mixed in ratio of 0.5% E2 and 95.5% CHOL and melted in an oil-bath (200*C). Using a heated pasteur pipette, aliquots of the homogeneous mixture were transferred into small molds made of aluminum foil. After cooling, the solidified pellets were unwrapped from the foil and each pellet weight was adjusted to exactly 100 mg. These pellets were used in the experiments described in Chapter 9.


Morphine Pellets

Morphine pellets were compounded in our laboratory, as previously reported (Simpkins et al., 1983), by the method originally described by Gibson and Tingstad (1970). Each pellet contained 75 mg morphine free base (Merk, St. Louis, MO), 37.5 mg microcrystalline cellulose (Avisil, FMC Corporation, Philadelphia, Pa), 0.56 mg Cab-o-sil (Cabot Corporation, Boston, MA) and 1.13 mg magnesium sterate (Fisher Chemical Co., Fair Lawn, NJ).





46


Animals

The laboratory rat was chosen as the experimental animal for all

experiments herewith. Adult male and female Charles River (CD) rats (aged 3-5 months) were purchased from Charles River Breeding Laboratories (Wilmington, MA). These rats weighed 200-250 g upon arrival and were allowed several days to adjust to the animal quarters before conducting an experiment. Animals were housed in a temperature- (24 1C) and light(lights on 0500 to 1900 hr daily) controlled room and provided with Purina rat chow and tap water ad libitum. After a 7-day acclimation period, animals were randomly divided into various experimental groups of 7-8 rats per group. This number of rats per group is standard for the field and is based on our estimates of experimental error in response to the drugs that were evaluated in these studies.

In experiments which required surgical procedures, animals were
anesthetized with Metofane (Methoxy Flurane, Pitman-Moore Inc., Crossing, NJ). The surgical procedures consisted of subcutaneous (sc) implantation of drugs or steroids, gonadectomy, and atrial cannulation. Female rats were bilaterally ovariectomized (OVX) by a small incision made through the dorsal peritoneal cavity. Male rats were castrated (CAST) by exteriorizing the testicles through a midline ventral incision. For atrial cannulation, in order to facilitate frequent blood sampling from unrestrained animals, a small incision in the neck-chest area was made to expose the external jugular vein. A Silastic catheter (i.d. 0.5 mm, o.d. 1 mm) was then positioned into the right atrium via the external jugular vein. This surgical procedure was done under sodium pentobarbital anesthesia, according to the guidelines described





47


by Steffens (1969). All animals were then monitored for post surgical recovery before conducting an experiment on them.

Two methods were employed for collecting blood samples. In most experiments, animals were killed by decapitation and the trunk blood was collected in heparinized tubes. In studies which required frequent blood sampling, animals equipped with atrial cannula were transferred to special sampling chambers and serial blood samples (1 ml) were removed through the cannula. All blood samples were collected in a room separate from the animal quarters. The blood samples were centrifuged and the plasma separated and stored at -20'C until hormone analysis by the radioimmunoassay (RIA).


Drug Treatment


Steroid Treatment

The E2-CDS treatments described in this dissertation were only given intravenously (iv). An aqueous solution of E2-CDS was prepared in 20% HPCD on the day of injection and administered iv (via tail vein), a procedure which required a brief restraint of the rat without anesthesia.

The 17 P-E2 treatments consisted of either iv administration (Chapters

6 & 7) or sc implantation of E2 pellet (Chapter 9). Implantation of the E2 pellet was performed in animals under light metofane anesthesia. It should be pointed out that these pellets did require pre-conditioning (i.e. soaking in PBS) before implantation. After implantation, each E2 pellet produces a transient high concentration of E2 in plasma that is followed by a sustained blood E2 levels for about 2 weeks.





48


Morphine and Naloxone Treatment

In experiments which examined the effects of E2-CDS on the tail-skin temperature (Chapter 9), animals were addicted to morphine. Morphine dependency was produced by sc implantation behind the neck region of one pellet containing 75 mg morphine free base. Two days after the first morphine pellet, two additional morphine pellets were sc implanted. This regimen of morphine treatment has been utilized in our laboratory to consistently produce typical symptoms of morphine dependency and tolerance (Simpkins et al., 1983, 1984) as measured by several tests of analgesia and withdrawal (Gibson &Tingstad, 1970; Simpkins et al, 1983, 1984). These morphine pellets produce serum morphine concentrations of 300 ng/ml by one hr after implantation and remain elevated at this level through 48 hrs (Derendorf & Kaltenbach, 1986). Thus, the sustained release of morphine achieved by the pellet is presumed to produce persistent stimulation of opiate receptors utilizing our treatment regimen.

Naloxone HCI from Dupont Pharmaceuticals (Garden City, NJ) was dissolved in saline and administered (0.5 mg/kg b.w.) subcutaneously.


Plasma Hormone Radioimmunoassays


Protein Hormone Assays

Plasma luteinizing hormone (LH), follicle-stimulating hormone (FSH), growth hormone (GH), and prolactin (PRL) concentrations were measured in duplicate by radioimmunoassay (RIA) using NIADDK kits provided by the National Hormone and Pituitary Program (Baltimore, MD). Plasma LH, FSH,





49


GH, and PRL values are expressed as ng/ml of either the LH-RP-2, FSH-RP-2, GH-RP-2 or the PRL-RP-3 reference standards, respectively. Values for unknown plasma samples were derived from the 10 to 90% (linear inhibition portion) of the respective standard curves. Radioiodinations of the labeled hormones were performed in our laboratory using standard procedures for a chloramine-T iodination with gel filtration chromatography to separate free iodine from hormone-bound iodine.

The ranges of hormone assay detectability were (1) 0.25 to 20 ng/ml for LH in 50 ul; (2) 2.5 to 200 ng/ml for FSH in 100 ul; (3) 2.5 to 320 ng/ml for GH in 25 ul; and (4) 0.25 to 50 ng/ml for PRL in 50 ul of plasma sample. Plasma samples containing undetectable LH, FSH, GH, or PRL were assigned the respective assay sensitivity of 0.25, 2.5, 2.5 and 0.25 ng/ml for these hormones. All samples for each hormone in an experiment were assayed in a single run.


Steroid Hormone Assays

Coat-A-Count Estradiol kits--employing solid-phase [1251]radioimmunoassay--designed for the quantitative measurement of E2 in serum were purchased from Diagnostic Products Corporation (Los Angeles, CA). Each kit is equipped with human serum-based standards having E2 values ranging from 20 to 3600 pg/ml (0.07 to 13.2 nmol/l) (technical information from Diagnostic Products). The cross-reactivity of the E2 antibody has been reported to be <0.3% for E2-Q+ at the concentration of 15 ng/ml and higher (Rahimy et al., 1989a). The cross-reactivity for estriol and estrone has been reported to be 0.32% and 1.1%, respectively (technical information from Diagnostic Products).





50


Coat-A-Count Testosterone kits--employing solid-phase [1251]radioimmunoassay--purchased from Diagnostic Products were used for the plasma testosterone (T) assay. The RIA sensitivity for the T assay was 0.2 ng/ml. The cross-reactivity of the T antibody to the DHT and E2 has been reported to be 3.3% and 0.02%, respectively (technical information from Diagnostic Products).


Statistical Analysis

For the experimental data that were normally distributed, the
significance of differences among groups were determined by one- or two-way analysis of variance (ANOVA) (Zar, 1974). Where necessary, data were transformed (ln) prior to ANOVA. Subsequent pairwise comparisons were made by Dunnett's or Scheffe's multiple range tests using Statview 512+ program from BrainPower, Inc. (Calabasas, CA) for the MacIntosh computer. Where appropriate, data were subjected to area under the curve (AUC) analysis using the trapezoid method (Tallarida & Murray, 1981) and group means for AUC were subjected to ANOVA and Scheffe's tests. The level of probability for all tests were p < 0.05. The specific statistical design employed in each experiment is indicated in the figure legends.













CHAPTER 4
DEVELOPMENT OF AN ANALYTICAL METHOD FOR THE
QUANTITATION OF E2-CDS METABOLITES IN A WIDE VARIETY OF TISSUES IN THE RAT


Introduction

The estradiol-chemical delivery system (E2-CDS) offers a novel
approach to non-invasively enhance brain delivery and sustained release of E2 (Bodor et al., 1987). The E2-CDS is a redox-based chemical-delivery system and exploits the unique architecture of the BBB, which normally excludes a variety of pharmacological agents from the brain (Bodor & Brewster, 1983). The mechanism of E2-CDS drug delivery is based on an interconvertible dihydropyridine <- pyridinium salt carrier. Figure 1 schematically shows the structures and the mechanisms leading to brain-enhanced and sustained release of E2. Estradiol, when it is chemically attached to the lipoidal carrier, dihydropyridine, its lipophilicity is further increased and thus, the ability to enter the brain is enhanced. After systemic administration of the E2-CDS, the carrier system is then quickly oxidized to the corresponding quaternary pyridinium salt (E2-Q+). This charged moiety of the carrier system reduces its rate of exit from the brain, thereby locking a depot of E2-Q+ into the CNS. Subsequent hydrolysis of the E2-Q+ with nonspecific esterases provides sustained release of the active species (E2) in the brain. Since the E2-Q+ is hydrophilic (40,000-fold greater than E2-CDS), its elimination rate from the periphery is predictably much faster than from the brain.


51





52


Application of the E2-CDS to biological systems creates a problem concerning the separation and quantitation of tissue levels of the active species, E2, in the presence of an excess concentration of the inactive conjugated form of the drug, E2-Q+. Furthermore, pharmacokinetic and pharmacodynamic studies of the E2-CDS require a reliable and sensitive method for the quantitative analysis of E2 as well as E2-Q+ in biological tissues and fluids. Conventional methods of assaying steroids, especially conjugated forms of E2, are inefficient and extremely time-consuming. (Bonney et al., 1984; Cortes-Gollegos & Gallegos, 1975; Hoffmann, 1983; Paradisi et al., 1980). RIA is the only method which has been used on a large scale for the analysis of steroid hormones in plasma or serum. Previously described methods for measuring steroidal hormones and their conjugates in biological tissues have been severely hindered by lengthy extraction and repeated purification procedures which are required prior to the use of RIA procedure.

Therefore, the objectives of these experiments were to develop a
sensitive, specific method that permits a rapid and reliable quantitation of both E2 and E2-Q+ in various biological tissues and fluids. This method development was necessitated by the design of the E2-CDS which exhibits the predictive, multiple, facile enzymatic conversion to the charged quaternary ion (E2-Q+) and its subsequent hydrolysis to slowly liberate E2.


Materials and Methods

Initially, several in vitro experiments were conducted to optimize the selective extraction, purification and quantitation of both E2-Q+ and E2 from a wide variety of biological tissues and fluids. Subsequently, an in vivo study was undertaken to assess the reliability and applicability of the in vitro





53


methodology by determining simultaneously the levels of both E2-Q+ and E2 in several tissues following administration of E2-CDS in the rat.


In Vitro Methodology


Specificity of the estradiol antibody for E7

The presence of the estradiol conjugate (E2-Q+) in biological tissues

could lead to overestimation of tissue E2 concentrations if E2-Q+ cross-reacts with the E2 antibody. Therefore, possible cross-reaction with the E2-Q+ was examined by adding E2-Q+ standards to the RIA of E2 at concentrations of 2, 5, 15, 50, 180, and 360 ng/ml. These doses of E2-Q+ were 100-fold greater than the corresponding E2 standards used to generate standard curves in the RIA of E2. Selective solvent extraction of steroids from tissues

Tissues including brain, anterior pituitary, kidney, lung, liver, heart, and adipose tissue were dissected from adult male Charles River (CD) rats (Charles River Breeding Laboratories, Wilmington, MA) immediately following decapitation. All tissues were rinsed in ice-cold saline and stripped of surrounding connective tissue and fat, blotted dry on paper then weighed to the nearest 0.1 mg. Tissue samples of known wet weight were then homogenized (using a Brinkman Polytron Homogenizer; Model PT 10/35) at moderate speed (setting at 6 for two 15-second periods) in an appropriate solvent system (depending on the type of tissue and compound of interest to be extracted). The appropriateness of the solvent system was determined by screening various organic solvents for effective extraction and acceptable recoveries of the steroids, with high reproducibility in the assay procedure. Tissue homogenate pools, at a concentration of 100 mg tissue/ml solvent, were prepared as follows: for E2 extraction from brain, anterior pituitary,





54


liver, and kidney, 100% methanol was used; for E2 extraction from lung, heart, and fat, 100% acetone was used. The reason for using different solvents was that methanol extraction yielded high and consistent (low CV, high CC) recovery of E2 from brain, anterior pituitary, liver, and kidney but resulted in low or inconsistent recovery of E2 from lung, heart, and adipose tissue. However, using 100% acetone, an acceptable and consistent recovery of E2 was observed for lung, heart, and adipose tissue. For E2-Q+ extraction, all tissue homogenate pools were prepared in water/acetone (50:50; v:v).

E7 recovery estimation. Duplicate aliquots of homogenate, each having a concentration of 100 mg tissue/ml solvent, were spiked with 90, 180, or 360 pg E2 and vortexed for 1 min. The added steroid was allowed to equilibrate with the homogenate for 30 min at room temperature. The spiked homogenates were then centrifuged for 10 min at 1,500 x g. The supernatant was decanted into a clean test tube and the pellet discarded. For plasma, duplicate 1 ml aliquots of plasma (from male rats) were spiked with the aforementioned doses of E2 standards, samples were allowed to equilibrate for 30 min, but no subsequent centrifugation was done. Following extraction, E2 recovery was determined as described below.

E?-Q+ recovery estimation. The procedure used for E2-Q+ adhered closely to that described for E2 recovery. Duplicate aliquots of homogenate were spiked with 150, 300, and 600 pg E2-Q+ and processed similarly to E2spiked homogenates. Serving as a control for the percent of hydrolysis of E2Q+, separate duplicate aliquots of homogenates were processed without the addition of E2-Q+ standards until after the supernatant were isolated. These supernatants were spiked with 300 pg E2-Q+.





55


Blanks. Tissue homogenates or plasma pools were extracted to

determine residual E2 concentrations and thereby served as the estimate of hormone background.

Hydrolysis of E7-0+ in various tissue extracts
In preparation for base-catalyzed hydrolysis of E2-Q+, the final volume of all E2-Q+ extracts, including the control samples, was brought to 900 ul by the addition of 50% water:50% acetone. To each tube containing E2-Q+ solution, 100 ul of 1ON NaOH were added to make the reaction medium 1N (pH >13). All tubes were vortexed and allowed to reach steady-state equilibrium for 20 min at room temperature. Under the conditions used, a time-course evaluation of the rate of E2-Q+ hydrolysis indicated that the steady-state equilibrium was achieved in less than 15 min and thus, longer incubation times did not work better. Also, it was found that the hydrolysis of E2-Q+ under basic conditions was maximized in an aqueous/organic solvent (50% water:50% acetone). After the hydrolysis, the pH of the reaction medium was adjusted to a pH in the range of 6 to 8 with NaH2PO4 and HCl. This is an optimal pH range for the maintenance of the integrity of the C18 columns. Samples with pH outside this range damaged the column sorbents, and the presence of column material in the assay sample interfered with the RIA for E2.

Solid-phase extraction and separation of E, by C18 columns

After C18 columns were conditioned with 2 column volumes (6 ml) of HPLC grade methanol followed by a column volume wash (3 ml) with distilled water, the samples to be extracted (E2 and E2-Q+ hydrolyzed extracts) were applied to the columns. Approximately 1 to 2 min were allowed for the column adsorption to be completed, then the columns were washed with 2





56


ml water:acetone (80:20; v:v). The columns were then allowed to air dry for 3 min before samples were eluted. With aqueous samples, the small amount of residual water in the column was removed with either 50-100 ul of HPLC grade hexane or by air-drying for >3 min. For sample elution, 2 aliquots of 500 ul methanol were applied sequentially to each column and the steroid was eluted under vacuum pressure. The eluates were collected separately into glass test tubes in the vacuum manifold and then dried under a stream of nitrogen gas. Methanol was used to elute E2 from the columns since we observed a column extraction efficiency of 92% with methanol, whereas two other solvents tested were less efficient in eluting E2 (88% with acetone and 49% with acetonitrile).

Radioimmunoassay of E7

The dried residues of the E2 samples were reconstituted in 300 ul of the assay buffer (kit Zero Calibrator; Lot # 10E2Z003; 100 ml) then, after vortexing for 1 min, samples were equilibrated for 30 to 60 min at room temperature. Duplicate 100 ul aliquots of each reconstituted E2 samples were assayed by RIA.

Calculations

If the mean blank (tissue homogenates that did not contain the E2 spike) values for an assay were greater than the limits of sensitivity of the standard curve, (i.e. if detectable E2 was present in the tissue), these values were subtracted from all spiked samples. Also, values calculated from the RIA run were adjusted for the volume of the aliquot taken for the RIA, experimental losses during solvent extraction and chromatography (determined by the addition of internal standard), and the weight of the tissue sample used (for the in vivo experiment).





57


In Vivo Studies

To evaluate the applicability and reliability of this procedure to in vivo condition, adult male Charles River (CD) rats received a single intravenous injection (tail vein) of the E2-CDS at a dose of 1.0 mg/kg body weight or the drug's vehicle, 2-hydroxypropyl-p-cyclodextrin (HPCD). Animals (6 per group) were then killed by decapitation 1, 7, and 14 days after the drug injection and the trunk blood was collected in heparinized tubes. The blood was centrifuged and the plasma separated and stored at -20'C until hormone analyses. Brains were removed, rinsed with ice-cold saline solution and stripped of their pia matter and immediately stored at -80*C until hormone analysis. Plasma and brain samples were each processed and assayed by the method described under the In Vitro Methodology section.

The 1.0 mg/kg dose of E2-CDS was chosen for investigation in this study for the following reasons: 1) we anticipated that the tissue concentrations of the E2-CDS metabolites, E2-Q+ and E2, would be in a quantitatable range when using this dose and the application of RIA procedure, over the time-course chosen for this study and 2) our previous pharmacological observations indicated that this dose of the E2-CDS is capable of causing chronic suppression of gonadotropins in castrated rats.





58


Results


In Vitro Methodology


Cross-reactivity of the E7 antibody with E?-O+
The inhibition of binding of 1251-E2 to the E2 antibody caused by E2 and E2-Q+ is shown in Figure 2. While E2 effectively competed for binding with the labeled hormone (IC50 = 368 pg/ml; 1.35 pM), E2-Q+ was ineffective in displacing the 1251-E2 in the RIA (IC50 = 129,000 pg/ml; 326.50 pM). At concentrations of E2-Q+ of 15 ng/ml and higher, the cross-reactivity of the E2 antibody for E2-Q+ was <0.3%. Cross-reactivity of the E2 antibody for estriol and estrone has been determined to be 0.32% and 1.1%, respectively (technical information from Diagnostic Products). Recovery of E,

Recovery of E2 was assessed by determining the % of each of three doses of E2 recovered from the brain, liver, kidney, lung, heart, and fat homogenates (Table 1). Due to limitations in the amount of tissue available, only one dose (180 pg/ml) was tested for homogenates of anterior pituitary glands. Recovery of E2 from each tissue evaluated was found to be dependent upon the organic solvent used in the extraction step. For brain, anterior pituitary, liver and kidney, E2 extraction with 100% methanol was found to be superior to two other solvents (acetone and acetonitrile) resulting in E2 recoveries (average of the three doses tested) of 77% for brain, 78% for anterior pituitary, 72% for liver and 71% for kidney (Table 1). Methanol was determined to be a poor solvent for extracting E2 from lung, heart and fat tissue, but 100% acetone achieved acceptable recovery of E2 in these three





59


tissues. The mean recovery of E2 was 57% for lung, 62% for heart and 64% for fat tissue. The E2 recovery from plasma was 81%. Precision of the E7 extraction-assay method

The precision of the method of estimating E2 concentrations in a
variety of tissues was determined in three ways. First, we determined the coefficient of variation (CV) for quadruplicate samples of each tissue at 3 different E2 dose levels. Second, we determined the correlation coefficient

(CC) of the E2 dose-RIA response for each tissue. And third, we determined the tissue weight-RIA response for brain samples taken at various times after rats were treated with the E2-CDS.

The CV for quadruplicate determinations of E2-spiked tissue
homogenates or plasma pools were in the range of 0.8 to 7.9% with the majority of E2 doses in each tissue showing a CV of less than 3.0% (Table 1).

The CC for the E2 dose-RIA response was 0.97 or greater for all tissues, indicating that over the E2 dose-range tested, the procedure accurately estimated the E2 concentration of the tissue (Table 1; Figure 3, lower panel).

To determine the accuracy of the method in estimating E2
concentrations in tissues of different weights and hence, to evaluate for tissue components which may interfere with various steps in the E2 assay procedure, we evaluated tissue wet weights over the range of 2 to 100 mg (Figure 4, upper panel). Brain tissue from rats treated 1 or 7 days before with the E2-CDS was extracted and assayed for E2. The inhibition of binding of 1251E2 was correlated with the weight of tissue used for both samples taken at day

1 (CC = 0.99) and 7 days (CC = 0.99) after E2-CDS administration. The parallelism of the observed inhibition curves indicated that the E2 concentration measured is independent of the weight of tissues used in the





60


determination. Also, the rightward shift in the tissue weight-RIA response curve at day 7 indicates a decrease in brain E2 concentrations with increasing time after drug administration.

Recovery of E7-O+
The recovery of E2-Q+ is dependent upon the efficiency of three processes: (a) the extraction efficiency of E2-Q+ from tissues with water:acetone; (b) the % hydrolysis achieved under basic conditions; and (c) the % recovery of E2 following its formation from the hydrolysis of E2-Q+. The recovery of E2 was determined as described above and the parameters described below after the adjustment for E2 recovery. The recovery of E2-Q+ was determined as the percent of the spiked concentration of E2-Q+ which was assayed. The % hydrolysis of E2-Q+ was determined experimentally for each tissue and the extraction efficiency in water:acetone was calculated as the recovery times the reciprocal of the % hydrolysis.
The extraction efficiency of E2-Q+ with water:acetone ranged from

65.4% for heart to 80.7% for liver tissue (Table 2). For each tissue evaluated, the % hydrolysis of E2-Q+ was the primary factor limiting the recovery of this species. The % hydrolysis ranged from 69% in liver to 37% in adipose tissue (Table 3). For all tissues except fat and plasma, the % hydrolysis was greater than 50% (Table 3).

Precision of the E7-Q+ extraction-assay method

The precision of the E2-Q+ method was determined using the same parameters used for demonstrating precision of the E2 method.

The CV for the quadruplicate determinations of E2-Q+-spiked tissue
homogenates or plasma pools ranged from 0.9 to 7.3% with a majority of the E2-Q+ doses for each tissue showing a CV of less than 5.0% (Table 2).





61


The CC for the E2-Q+ dose-RIA response was 0.98 or higher for each tissue evaluated (Table 2; Figure 3, upper panel). Thus over the E2-Q+ dose range tested, the method accurately estimated E2-Q+ concentrations in each of the tissues.

Increasing brain tissue wet weight from 0.25 to 50 mg caused a highly correlated decrease in binding of 1251-E2 in the RIA used at day 1 (CC = 0.99), day 7 (CC = 0.99) and day 14 (CC = 0.99) after the treatment of rats with E2-CDS (Figure 4, lower panel). The inhibition curves caused by increasing brain tissue wet weight from animals at 1, 7 and 14 days posttreatment were parallel and the rightward shift was indicative of the time-dependent reduction in brain E2-Q+ concentrations (Figure 4, lower panel). Distribution of E, and E2-0+ in vivo

Figure 5 shows brain and serum levels of E2-Q+ (upper panel) and E2 (lower panel) at various times following administration of the E2-CDS (1 mg/kg) or the HPCD vehicle (day 0 values). Brain E2 concentrations were increased to 11.1 1.4 ng/g tissue (mean SEM) on day 1 and remained greater than 3.5 1.1 ng/g at day 14 after drug treatment. A small amount of E2 was detected in brains of HIPCD-treated (control) male rats (0.2 0.07 ng/g), a likely result of the aromatization of testosterone in the brain (Michael et al., 1986). Brain concentrations of E2 exceeded serum levels of the hormone by 39-, 41- and 82-fold at 1, 7 and 14 days, respectively, after E2-CDS treatment (Figure 5).

E2-Q+ levels were increased to 200.9 8.8 ng/g brain tissue at 1 day after administration of the E2-CDS, and E2-Q+ levels remained elevated to 67.0 17.2 ng/g at 14 days postinjection (Figure 5). The brain to serum ratio for E2Q+ was 33, 70 and 294 at 1, 7 and 14 days, respectively.





62


In brain tissue, E2-Q+ levels exceeded E2 levels by 18-, 22- and 19-fold

and in plasma E2-Q+ levels were 6, 13 and 22-fold higher than E2 at 1, 7 and 14 days, respectively.


Discussion

This novel, but predictable, metabolism of the E2-CDS presents several problems for the quantitation of E2-Q+ and E2, two metabolites of the E2-CDS. First, since estradiol is active at tissue concentrations of low pg/g, the assay method for the E2-CDS metabolite, E2, must be extremely sensitive. Second, since E2-Q+, the moiety "locked" in the brain, is expected to be present in concentrations much higher than E2, the assay method must be capable of distinguishing low levels of E2 in the presence of high concentrations of E2Q+. Third, the accuracy of the E2 determination is dependent upon the stability of E2-Q+ against hydrolysis (enzymatic or spontaneous) throughout the procedures. However, under the conditions utilized, E2-Q+ was quite stable. When 300 pg E2-Q+ were added to tissue homogenates and evaluated for spontaneous hydrolysis throughout the procedures used, the E2 recovery was below the sensitivity of the assay. Finally, as for any assay method, necessary features are (a) a high recovery of the species of interest; (b) accuracy of the determinations; and (c) reliability of the method through a wide range of hormone concentrations and tissue weights. We have provided evidence for each of the features of the present method for simultaneous determinations of E2-Q+ and E2.

The RIA procedure provides the needed sensitive endpoint for the
determination of E2 and E2-Q+ levels. This RIA for E2 is sensitive from 0.8 to

1.2 pg E2/assay tube and exhibits a highly correlated inhibition of 1251-E2





63


binding over the range of 20 to 3600 pg/ml. This level of sensitivity is substantially greater than the recently published HPLC methods which report levels of sensitivity of 50 and 10 ng/ml of plasma for E2 and E2-Q+, respectively (Mullersman et al., 1988). Additionally, the antibody used was very specific for estradiol, showing a cross-reactivity of <0.3% for E2-Q+ and has also been described to cross-react with estriol and estrone at the level of

0.3 and 1.1%, respectively. In brief, the RIA described here is sensitive and specific for E2.

The recovery of E2 is dependent upon the extraction efficiency of the organic solvent used and the elution efficiency of E2 loaded on the C18 columns. Since column elution of E2 with 100% methanol was essentially quantitative, the recovery of E2 was equivalent to the extraction efficiency. Methanol extraction yielded high and consistent (low CV, high CC) recovery of E2 from brain, anterior pituitary, liver and kidney but resulted in low or inconsistent recovery of E2 from lung, heart and adipose tissue. However, using 100% acetone, an acceptable and consistent recovery of E2 was observed for lung, heart and adipose tissue. While the reason for this tissue-specific solvent extraction is not clear, the results indicate that the judicious choice of solvent allows for the reliable estimate of E2 concentrations in a variety of tissues. Indeed, E2 was precisely measured in a variety of tissues over a 4-fold change in the concentration of E2 or over a 50-fold change in the amount of tissue used in the determination (Figures 3 & 4).

The recovery of E2-Q+ was limited primarily by the percent hydrolysis of the E2-Q+, since extraction efficiency in water:acetone (50:50; v:v) ranged from 65 to 81% for an individual tissue. We observed that our base-catalyzed hydrolysis of E2-Q+ yielded values of 54% to 69% for all tissues except fat (37%) and serum (30%). While base hydrolysis is not complete, the % hydrolysis





64


was consistent for each tissue and the variation in % hydrolysis was low (CV = 0.8% for plasma to 5.0% for brain tissue). Furthermore, other methods of hydrolyzing E2 conjugates, such as enzyme- or acid-catalyzed hydrolysis, are much less efficient (11% and 0.8% net hydrolysis, respectively) and require 18 to 24 hours to conduct (Bain et al., 1984; Czekala et al., 1981; Saumande & Batra, 1984; Segal et al., 1960) versus 20 min for base-catalyzed hydrolysis under the present conditions.

Analysis of E2-Q+ in adipose tissue was complicated by two factors. First, E2-Q+ is hydrophilic (due to its charge and polarity) and requires an aqueous solvent for its effective extraction. This requirement creates problems in separating the supernatant from the pellet because of the formation of a superficial layer of lipid above the supernatant phase. Second, extensive loss of E2 from the supernatant occurs after E2-Q+ was hydrolyzed. This is due to the presence of fatty droplets in the reaction medium into which E2 partitioned from the aqueous supernatant; and therefore, it was not recovered efficiently when the supernatant was transferred onto the C18 column. These two conditions reduced the recovery of E2-Q+ from adipose tissue.

Analysis of E2-Q+ in heart and kidney tissues posed a different problem. After hydrolysis of E2-Q+ extracts in 1N NaOH solution, the hydrolyzed supernatants from heart and kidney required less HC and NaH2PO4 than other tissues to adjust the pH to the range of 6 to 8; the optimal pH range for C18 column function. Without determining the exact amount of acid and buffer needed to achieve the optimal pH range for each tissue, erroneously high E2-Q+ levels were calculated due to contamination of the assay tube by the column sorbent.





65


E2-Q+ recovery from plasma samples was low likely because of protein precipitation caused by hydrolysis and subsequent neutralization. Centrifugation was needed to separate the supernatants from precipitates before their application onto the C18 columns. E2 released during the hydrolysis step likely interacted extensively with albumin and sex steroidbinding globulin and was unavailable for column extraction.

The separation of E2 and E2-Q+ was achieved by 3 different techniques in the process of extraction and assay of these two products of the E2-CDS. First, samples were divided and differentially extracted for E2 (methanol or acetone) and for E2-Q+ (50% water:50% acetone). Although this procedure, which depends upon the solubilization of the lipophilic E2 in methanol or acetone and the more hydrophilic E2-Q+ in water/acetone effectively extracted the intended steroid, separation of the two species was not complete. However, when extracts were loaded onto the C18 column and eluted with 100% methanol, only E2 was preferentially extracted and eluted by more than 92%. Thus, the column chromatography effectively separated the two species. Finally, the low cross-reactivity of the E2 antibody for E2-Q+ (<0.3%) ensured that in samples extracted and chromatographed for E2, virtually no E2-Q+ was measured. Moreover, analysis of tissue samples treated with the E2-CDS required various dilutions for each time point which ensured the expected E2 values to fit an appropriate part of the standard curve (ED20 at 800 pg/ml to ED80 at 25 pg/ml). Indeed, dilutions which were performed prior to loading the extracts onto the C18 column minimized nonspecific interference by E2-Q+ and lipids. Lipids decrease steroid radioimmunoassay accuracy and reproducibility (Rash et al., 1980).





66


Application of these methods to brain and plasma samples obtained at various times after treatment with the E2-CDS revealed that both E2 and E2Q+ can be quantitated throughout the 14-day time-course of the study. As predicted, based upon previous reports on brain levels of E2-Q+ (Bodor et al., 1987; Mullersman et al., 1988), this "locked-in" form of the E2-CDS reached a 33-fold higher concentration in the brain than plasma by 1 day after treatment with E2-CDS and these brain-blood ratios increased to 294-fold by 14 days. Brain E2 concentrations were similarly and dramatically elevated relative to plasma. These observations are consistent with the proposed brain-enhanced delivery of E2 with the redox-based E2-CDS and indicate that the observed distribution pattern of E2-Q+ and E2 may explain the long-term pharmacological effects of the E2-CDS (Anderson et al., 1988ab; Estes et al., 1987a,b; Simpkins et al., 1986).

In summary, the described technique for the simultaneous
measurement of E2-Q+ and E2 is sensitive, reliable, specific and applicable to a wide variety of tissues in the body. The additional feature of rapidity of the method allows for the determination of about 100 samples in one day. Collectively, these characteristics indicate that the described techniques could be applied to the quantiation of other conjugates of steroid hormones.





67


H




110/


ESTRADIOL


C

CHEMICAL
TRANSFORMATIONS H11/


E2 -CDS


S



C H+
I
HO ~ ~ ~ i oo 1 -4


IN VI
OXIDA


VO
TION


ELIMINATION FROM GENERAL CIRCULATORY
SYSTEM


E2Q +


Figure 1. Schematic representation of in vitro synthesis and in vivo
transformation of the estradiol-chemical delivery system (E2-CDS).
E2-Q+ is the charged quaternary form of the E2-CDS which is
"locked" into the brain and quickly eliminated from the
peripheral tissues. Subsequent hydrolysis of E2-Q+ with nonspecific esterases results in sustained and slow release of estradiol
in the brain. The trigonelline, carrier moiety, formed upon
hydrolysis of E2-Q+ is non-toxic and is cleared from the brain rather quickly. Although the in vivo rate constants for these
reactions are unknown among different tissues, the oxidation of
E2-CDS to E2-Q+ is quite rapid in all tissues analyzed (t1/2 = 29
min). However, the rate constant for hydrolytic enzymes may
differ among various tissues.


VO


IN VI
HYDRO


LYSI








68


-


80


S


0


102


10


105


101


PG/ML











Figure 2. Inhibition of 1251-E2 binding to an E2 antibody caused by E2 (left,

open circle) or E2-Q+ (right, closed circle). Possible cross-reaction

with the E2-Q+ was examined by adding E2-Q+ standards to the

RIA of E2 at concentrations of 2, 5, 15, 50, 180, and 360 ng/ml.

These doses of E2-Q+ were 100-fold greater than the corresponding

E2 standards used to generate standard curves in the RIA of E2.

Data are expressed on logit-log graph (% bound on ordinate is

based on logit = log (percent bound/100 percent bound)). Crossreactivity data indicated that E2 antibody used was specific for E2

and cross-reacts with E2-Q+ to <0.3% at concentration of 15 ng/ml

and greater.


97 95


a
z
0

z
W C.)
a.


70 60 50 40 30 20


10


106


1


104






69


300
I E2-Q+

y =10.23 + 0.40 x 200 r =0.994
to


0
100





0 100 200 300 400 500 600 700

300- E2


y =8.05 + 0.70 x C> r =0.994
C 200ba

0
U
100





0 100 200 300 400
Added Conc. (pg/100 mg tissue)


Figure 3. Recovery of known concentrations of E2-Q+ (upper panel) and E2
(lower panel) added to brain tissue homogenates prior to
extraction. Duplicate aliquots of homogenate were spiked with
150, 300, or 600 pg E2-Q+ (upper panel) or 90, 180, or 360 pg E2
(lower panel). After equilibration for 30 min followed by solvent
extraction, the spiked homogenates were centrifuged and the
supernatant was analyzed for E2-Q+ or E2 by the RIA for E2. The
results indicated that the assay method used accurately determines
E2-Q+ and E2 over a wide range of tissue concentrations.







70


6 Day 1 0 Day 7


0


100


6Day I 0 Day 7 O Day 14











D 1


10


1000


100


TISSUE (MG)/TUBE


Inhibition of 1251-E2 binding to an E2 antibody caused by increasing amounts of brain tissue from rats treated with the estradiolchemical delivery system (1.0 mg/kg). The upper panel depicts brain tissue wet weights extracted over the range of 2 to 100 mg for E2 at 1 or 7 days after treatment with E2-CDS. The lower panel depicts brain tissue extracted over the range of 0.25 to 50 mg for E2Q+ at 1, 7 or 14 days after treatment with E2-CDS. Data are expressed on logit-log graph (% bound on ordinate is based on logit = log (percent bound/100 percent bound)). The results indicated that the assay method used accurately measures E2-Q+ and E2 concentrations over a wide range of tissue weights.


70 60 50

40 30


20 -


a
z
0 I->

z
0
cc


0
z
0 0.
z
W
0
w a.


10


10


80 70 60 50

40 30


20 1


10


0.1


Figure 4.


i I


1





71


30000 25000

20000 15000 10000 5000


1400012000100008000 6000

4000 2000


0 10


0

0

0




0

0

0
0


o Brain
* Serum


14


1


7


Days post Injection


Figure 5. Effects of a single iv dose of the estradiol-chemical delivery system
on serum and brain levels of E2-Q+ (upper panel) and E2 (lower
panel) or the HPCD vehicle (day 0). Each point represents the
group mean SEM. N = 6 animals per group.


E2-Q+

E Brain Serum










7 14


0


to


(N





72


Table 1: Recovery and Precision Determination for Biological Samples Spiked with E2



E2 Added E2 Assayedb Recovery CVC CCd
Tissuea (pg) (pg) (%) (%) (r)


Brain Anter. Pituit. Plasma Kidney Lung Heart Liver Fat


a
b
c
d
e


90 180 360

180

90 180 360

90 180 360

90 180 360

90 180 360

90 180 360

90 180 360


73.9
130.4 268.1

141.0

63.4 151.8 321.3

71.5 121.8 238.8

52.0 100.0 206.1

52.6 112.6 230.8

69.4 126.5
245.4

66.4 97.4 228.0


82.1
72.4 74.5

78.3

70.5
84.3 89.2

79.4 67.6 66.3

57.8
56.4 57.3

58.5 62.6
64.3

77.1 70.3 68.2

73.8
54.1 63.3


7.9 3.3
3.4

1.8

2.5 3.0 2.9

2.2 1.2 1.8

6.0 3.7
2.4

2.7 2.8
4.4

1.7 1.9 2.5

3.0 0.8 1.1


0.994 NDe 0.997 0.996 0.993 0.997 0.997 0.975


100 mg of tissue or 1 ml of plasma was used. Mean of n = 4 for each dose of E2 in each tissue. CV = coefficient of variation. CC = correlation coefficient. ND = not determined.





73


Table 2: Recovery and Precision Determinations for Biological Samples Spiked with E2-Q+



E2-Q+ Added E2-Q+ Assayedb EEc Recovery CVd CCe Tissuea (pg) (pg) (%) (%) (%) (r)


Brain


Plasma Kidney Lung Heart Liver


150 300 600

150 300 600

150 300 600

150 300 600

150 300 600

150 300 600

150 300 600


Fat


73.7 129.6
249.9

47.5 87.8 208.3

68.4 137.9 238.3

64.0 131.1 236.7

52.6
104.0 218.5

80.4 164.0 319.2

39.8 78.9 163.3


74.5 65.1 63.1

ND ND ND

73.4 74.0 64.4

76.3 78.2 71.6

64.8 64.1 67.4

80.4 82.0 79.8

70.9 70.2 72.7


49.2 43.2 41.7

31.7 30.0
34.7

45.6 46.0 40.0

42.6 43.7 40.0

35.0
34.7 36.4

53.6
54.7 53.2

26.6 26.3 27.2


1.3
4.6 4.6

6.1 2.8 0.9

1.0 3.0
2.2

1.0 2.6 1.1

2.6
4.6 5.9

2.2 6.7 7.3

2.1 2.6 3.0


0.994 0.989 0.990 0.995 0.989 0.988 0.988


100 mg of tissue or 1 ml of plasma was used. Mean of n = 4 for each dose of E2-Q+ in each tissue. EE = extraction efficiency. CV = coefficient of variation. CC = correlation coefficient. ND = not determined.


a
b
c
d
e
f





74


Table 3: Percent Hydrolysis of E2-Q+ in Supernatants of a Variety of Tissues


E2-Q+ Added E2-Q+ Assayedb Hydrolysis CVC Tissuea (pg) (pg) (%) (%)

Brain 300 198.3 66.1 5.0

Plasma 300 87.8 30.0 2.8

Kidney 300 186.6 62.2 2.9

Lung 300 167.6 55.8 1.8

Heart 300 162.5 54.2 1.7

Liver 300 207.3 69.1 3.5

Fat 300 112.3 37.4 4.1


a
b
c


100 mg of tissue or 1 ml of plasma was used. Mean of n = 4 for each dose of E2-Q+ in each tissue. CV = coefficient of variation.













CHAPTER 5
DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS METABOLITES IN MALE RATS


Introduction

Estrogens are intrinsically lipophilic (Abraham, 1974) and readily cross the blood-brain barrier (BBB) to gain access to the central nervous system (CNS). However, when inside the CNS, there is no mechanism to prevent their redistribution back to the periphery as blood levels of the steroid decline (Davson, 1976). Indeed, when these hormones are used therapeutically to specifically target the CNS, the steroids tend to equilibrate among all body tissues due to their high lipophilicity (Pardridge & Meitus, 1979). As a result, only a fraction of the administered dose accumulates at or near the site of action in the brain. This property of the estrogens necessitates, either frequent dosing, or the administration of a depot form of the estrogen to achieve and maintain therapeutically effective concentrations in the brain (Spona & Schneider, 1977). Both of these treatment strategies lead to sustained increases in peripheral estrogen levels.

Furthermore, estrogen receptors are present in many peripheral tissues (Walters, 1985), where they mediate a myriad of physiological and pharmacological effects (Murad & Haynes, 1985; Walters, 1985). This further creates the potential of untoward peripheral side effects (Thomas, 1988). In fact, constant increases in peripheral tissue exposure to estrogens have been shown in numerous studies to precipitate various peripheral toxicities, including risk of breast and endometrial cancer (Hurst & Rock, 1989; Persson,


75





76


1985; Thomas, 1988), cardiovascular complications (Barrett-Conner et al., 1989; Drill & Calhoun, 1972; Inman & Vessey, 1968; Kaplan, 1978), and alterations in hepatic metabolism (Burkman, 1988).
Since the brain is the primary site where E2 exerts its beneficial effects on the estrogen withdrawal symptoms at the menopause (Casper & Yen, 1985; Lauritzen, 1973; Yen, 1977), to inhibit gonadotropin secretion for fertility regulation (Goodman & Knobil, 1981; Kalra & Kalra, 1989; Plant, 1986), to reduce growth of peripheral steroid-dependent tissue tumors such as the prostate (Rao et al., 1988), and to stimulate male and female sexual behaviors (Beyer et al., 1976; Christensen & Clemens, 1974), a brain-enhanced delivery with sustained release of E2 in that tissue is warranted. The ability to deliver E2 preferentially to the brain, thus sparing non-target site tissues, should improve the therapeutic index of E2 by (i) increasing the concentrations and/or residence time of E2 at its receptor site in the brain and (ii), equally important, decreasing the concentrations and/or residence time of E2 at the potential peripheral sites of toxicities, thereby decreasing untoward peripheral side effects.

Having established a reliable, specific method for the simultaneous quantitation of E2-CDS metabolites in various tissues (Chapter 4; Rahimy et al., 1989a), and thus to demonstrate the effectiveness of the E2-CDS, extensive time-course studies were undertaken to evaluate the tissue distribution of E2Q+ and E2 in both male and female rats. The objective of this study was to evaluate a general tissue distribution of E2-Q+ (the intermediate, oxidized metabolite of the E2-CDS) and E2 (the active, parent steroid released upon hydrolysis of the E2-Q+) in brain, anterior pituitary, lung, liver, kidney, heart, fat, and plasma following a single iv dose of 1 mg/kg of E2-CDS in the male rat.





77


Materials and Methods

Adult, intact male Charles River (CD) rats (225-250 g) received a single iv injection (tail vein) of the E2-CDS at a dose of 1.0 mg/kg body weight or the drug's vehicle, 2-hydroxypropyl-o-cyclodextrin (HPCD). Rats (6-7 per group) were killed by decapitation 1, 7 or 14 days after the drug administration and the trunk blood was collected in heparinized tubes. The blood was centrifuged and the plasma separated and stored at -20'C until hormone analysis. Tissues (brain, anterior pituitary, lung, liver, kidney, heart, and fat) were dissected immediately following decapitation and stored at -80'C until hormone analysis.

Tissue samples of known wet weight at a concentration of 1 mg/20 pl solvent were processed and assayed by the method described in Chapter 4 (Rahimy et al., 1989a). Tissue homogenates and plasma from HPCD-treated rats were also extracted to determine the residual E2 concentrations and thereby served as the estimate of hormone background.

Coat-A-Count Estradiol kits--a solid-phase [125I]-radioimmunoassay-designed for the quantitative measurement of E2 in serum were used for the assay of E2 in all tissue and plasma samples. Cross-reactivity of the E2 antibody was determined to be <0.3% for E2-Q+ at a concentration of 15 ng/ml and higher (Chapter 4). All the purified dried E2 unknowns were reconstituted in 300 gl of the assay buffer (kit Zero Calibrator) and assayed in duplicate by the RIA. The intra-assay coefficient of variation for E2 was 1.56% and all samples were determined in two assay runs.

Calculated values obtained from the RIA run were adjusted for the
volume of the aliquot taken for the RIA, experimental losses during solvent





78


extraction and chromatographic separation (using internal standard), and the weight of the tissue sample used.


Results

The results of this experiment are presented in Figures 6-9. Figure 6
shows brain (upper panels) and plasma (lower panels) concentrations of E2-Q+ and E2 at various times following administration of the E2-CDS. Brain E2-Q+ concentrations increased to 318 14 ng/g tissue (mean SEM) on day 1, followed by a linear decline to 39 2 ng/g on day 14. This result indicated a reduction in E2-Q+ concentration of 46% by 7 days and 88% by 14 days after administration of the E2-CDS. In contrast, plasma concentrations of E2-Q+ increased to 6.1 0.3 ng/ml on day 1, then rapidly decreased by 79% at day 7 and remained at very low levels (0.23 0.03 ng/ml) at day 14 after the E2-CDS treatment.
Brain concentrations of E2 increased to 8 0.5 ng/g (day 1) then

decreased steadily to 2 ng/g (day 14), indicating a sustained-release behavior from brain E2-Q+. In contrast, plasma E2 concentrations increased to only 0.28

0.1 ng/ml (day 1) and steadily declined thereafter.

Figure 7 shows the time-concentration profiles of E2-Q+ (upper panel) and E2 (lower panel) in liver and fat tissues at various times following administration of the E2-CDS. Both E2-Q+ and E2 were detected in these tissues throughout the time-course studied. As expected, these tissues showed rapid clearance of E2-Q+ as well as E2. The E2-Q+ concentrations decreased from 77 10 ng/g and 71.7 19.5 ng/g (day 1) to 5.4 1.3 ng/g and
1.9 0.5 ng/g (day 14) in liver and fat, respectively. This indicated a reduction





79


in concentrations of greater than 83% and 87% from 1 to 7 days and 93% to 98% by 14 days after drug administration in liver and fat, respectively.
Similarly, E2 concentrations in these tissues fell by more than 84% and 80% from day 1 to day 7, and by 14 days after drug administration, the E2 concentrations decreased by 95% and 90%, respectively.

Figure 8 shows the time-concentration profiles in kidney, heart, lung, and anterior pituitary concentrations of E2-Q+ (upper panel) and E2 (lower panel) at various times following administration of the E2-CDS. The E2-Q+ concentrations in these tissues initially increased to 1906 131, 1047 106, 748
28, and 407.6 50.6 ng/g in heart, lung, kidney, and anterior pituitary,

respectively. These E2-Q+ levels decreased rapidly by more than 76%, 79%, 74%, and 80% by day 7 in these 4 tissues, respectively. By 14 days after drug administration, E2-Q+ concentrations decreased by greater than 98% in heart and lung, 96% in kidney, and 93% in anterior pituitary. Despite high initial concentrations of E2-Q+ in these peripheral tissues, brain levels of E2-Q+ exceeded E2-Q+ levels of these tissues by 1.5- to 3-fold at 14 days after administration of the E2-CDS.

Estradiol concentrations in heart, lung, kidney, and anterior pituitary (Figure 8; lower panel) were similarly elevated on day 1 but decreased rapidly by 67% in heart, 83% in lung, 81% in kidney, and 86% in anterior pituitary by day 7. From day 1 to day 14, the E2 levels in these tissues decreased by more than 95% of the initial concentrations.

Figure 9 depicts brain (upper panels) and anterior pituitary (lower
panels) contents of E2-Q+ and E2 at various times after administration of the E2-CDS. Following a single injection of the E2-CDS, the brain E2-Q+ content was 635 28, 340 23, and 77 3.9 ng /brain at 1, 7 and 14 days, respectively.





80


By contrast, the anterior pituitary content of E2-Q+ was only 1/260 to 1/170 of that observed in the brain.
Similarly, the brain E2 content was 15.8 0.9, 10.4 0.8, and 3.2 0.1

ng/brain at 1, 7 and 14 days following administration, respectively, while the anterior pituitary E2 content was 0.47 0.04, 0.06 0.004, and 0.035 0.006 ng/anterior pituitary at these sampling times. As such, the anterior pituitary E2 content was only 1/90 to 1/34 of that observed in the brain throughout the time-course studied. Thus, the absolute amounts (contents) of E2 and E2-Q+ were many fold higher in the brain even though the anterior pituitary concentrations of E2 and E2-Q+ were initially higher than in the brain.


Discussion

The results of this single-dose distribution study demonstrated that both E2-Q+ and E2, two metabolites of the E2-CDS, were present in all tissues analyzed up to 14 days (the last sampling time) after treatment of male rats with the E2-CDS. Moreover, over the time-course studied, the distribution profiles indicated that: a) regardless of the tissues evaluated, E2-Q+ levels were many fold higher than E2 levels at each time point in a particular tissue, indicating a slow rate of hydrolysis of E2-Q+ to E2; b) the increased brain/plasma ratios of E2-Q+ as well as E2, confirmed that "locking" of the charged moiety, E2-Q+, into the brain had occurred; and c) E2-Q+ is retained in the CNS tissue but is rapidly cleared from the peripheral tissues, an observation which is predicted by the inherent physicochemical properties of the delivery system.

Brain-distribution profile revealed that E2-Q+, the quaternary form of the delivery system, persists in the brain with a half-life of about 8 days, but it





81


is rapidly cleared from the periphery. This is in accordance with that reported in other studies (Mullersman et al., 1988). The half-life of the lipophilic E2CDS in brain tissue is only 29.2 min (Bodor et al., 1987), indicating rapid oxidation of the delivery system to E2-Q+. From this store of E2-Q+, E2 can be slowly released chronically in the brain through nonspecific hydrolysis.
As predicted from the physicochemical properties of E2-CDS as well as previous reports on brain levels of E2-Q+ (Boder et al., 1987; Mullersman et al., 1988; Simpkins et al., 1986), this "locked-in" form of the E2-CDS reached a 52-fold higher concentration in the brain than plasma by day 1 after treatment with the E2-CDS, and these brain-blood ratios increased to 132-fold at day 7 and to about 170-fold by 14 days. Furthermore, from day 1 through 14, the content of E2-Q+ in the brain was 6- to 23-times the content of E2-Q+ in the blood. Thus, a portion of the E2-Q+ found in plasma may arise from brain stores of the compounds. E2-Q+ can be cleared from the brain by bulk flow of cerebrospinal fluid (Boder & Brewster, 1983; Schanker, 1965).
Brain E2 concentrations were similarly elevated relative to plasma.

Estradiol achieved a 28-fold higher concentration in the brain than plasma by day 1 and this ratio increased to more than 50-fold at day 7 and remained at 37-fold by 14 days. Additionally, throughout the time-course studied, the brain E2 content was 3- to 6-times the content of E2 in the blood. These observations indicated that brain E2 is continuously produced, and as such the steady-state brain E2 concentration is dependent on its rate of production from the E2-Q+ and its rate of elimination from the brain by local metabolism and/or redistribution down a concentration gradient to the plasma. Brain stores of E2 could contribute to plasma levels through its partitioning to the periphery down a large concentration gradient.





82


Levels of E2-Q+ in the anterior pituitary were surprisingly high on day

1 then dropped rapidly to below that of brain levels by day 7 and steadily decreased thereafter throughout the observation period. This initial rise in E2-Q+ levels may be attributed to increased anterior pituitary uptake of the E2CDS followed by its rapid metabolism and clearance. Furthermore, the relative elevation of E2-Q+ as well as E2, from day 7 to day 14, in anterior pituitary may be caused by the anatomical relationship between the hypothalamus and anterior pituitary gland. Estradiol released upon the hydrolysis of E2-Q+, or the E2-Q+ itself, which is locked into brain, could be delivered directly to the anterior pituitary by the capillary plexus of the hypophyseal portal system. These capillaries in the median eminence lack features of other brain capillaries and hence are not part of the blood-brain barrier (Traystman, 1983). Thus, the median eminence would not be expected to prevent the efflux of E2-Q+ from the brain, and thus transfer of E2-Q+ to the anterior pituitary can be expected.

High levels of E2-Q+ seen in the kidney are likely because this organ is a major site for the elimination of all metabolites of the E2-CDS. However, the reasons for initial high levels of E2-Q+ in the lung and heart tissues of the male rat are not clear. We speculate that since these organs receive high blood flow, a substantial amount of the E2-CDS is delivered to and taken up by these tissues initially. Additionally, despite higher E2-Q+ levels in heart relative to those of lung, kidney, and anterior pituitary, the heart E2 concentrations were lower than these 3 tissues. Perhaps, a slow rate of hydrolysis of E2-Q+, or a slower E2 metabolism, in heart tissue could contribute to the higher levels of E2-Q+ and lower levels of E2 in this tissue.

This single-dose pharmacokinetic study supports the previous
observations of prolonged pharmacodynamic effects of the E2-CDS following





83


administration of a single dose. LH secretion in castrated male rats was suppressed for greater than 21 days (Estes et al., 1987b; Simpkins et al., 1986), sexual copulatory behavior was stimulated for 28 days (Anderson et al., 1987a), and body weight was suppressed for 36 days (Estes et al., 1988; Simpkins et al., 1988) after doses of E2-CDS of 1 to 3 mg/kg. These prolonged effects of the E2-CDS are consistent with the observation here of the accumulation of E2-Q+, the oxidized form of the delivery system, in the rat brain and its long half-life (t 1/2 = 8 days) in this tissue. From this store of E2Q+, E2 is released through slow hydrolysis and exhibits a half-life similar to that of E2-Q+.

Finally, since we evaluated the tissue distribution of E2-CDS
metabolites in intact male rats, determination of E2-CDS metabolites in the testes, prostate, and seminal vesicle tissues would have certainly added more valuable informations to the results presented in this chapter. Unfortunately, at the time of experimental investigation, these tissues were not collected for analysis of E2-CDS metabolites. Certainly, in future studies involving E2-CDS, these tissues need to be evaluated for E2-Q+ and E2 distribution and clearance. Furthermore, the question of blood-testes barrier must be addressed. This would be of great interest to find out whether E2-Q+ is being "locked" in this tissue similar to CNS tissue or behaves like the rest of peripheral tissues. However, effects of the E2-CDS on weights of these androgen-responsive tissues were examined and the are presented in Chapter 8.

In conclusion, these observations are consistent with the proposed
mechanism of the redox-based E2-CDS and the contribution of the BBB to the chronic retention of the charged, hydrophilic E2-Q+ in the rat brain. This observed tissue distribution pattern of E2-Q+ and E2 may explain the longterm pharmacological effects of the E2-CDS in the male rat.





84


Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on brain (upper panels) and plasma (lower panels) concentrations of E2-Q+ and E2 in intact male rats. Intact male rats were injected with a single iv dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14 days after treatment. Whole brain tissue and plasma were processed and assayed for E2-Q+ and E2 by the method described in Chapter 4. Each point represents the group mean SEM (n = 6-7 rats for each time point).


Figure 6.





85


E2-0+


0
u


10 8

6

4-


0


7


14
14


E2-0+


E2


-V -


01


0.4-


0.3


e 0.20
u


0.0


7


14


7


E2


- - - -


0 1


7


TIME (days)


400 o 300 0 200U

0 100


0


01


14


u
0

U


4


2-


0


-V


0 1


14


-


TIME (days)





86


100
E2-0+

80 -0 Liver
Fat

5 60

0
U
S 40


20


0
0 1 7 14

5
E2

4 -- Liver
Fat

3

0
U
2


1



0 1 7 14
Days Post Injection


Figure 7. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on liver and
fat concentrations of E2-Q+ (upper panels) and E2 (lower panel) in
intact male rats. Intact male rats were injected with a single iv
dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14
days after treatment. Each point represents the group mean SEM
(n = 6-7 rats for each time point).





87


2500
E2-0+

2000- 1 Kidney
0 -Heart bo
bo Lung
5 1500- Ant Pit


U
1000


500


0
0 1 7 14

60
E2
50 Kidney
Heart
40 I Lung
r. Ant Pit

0 30
U
20


10


0'
0 1 7 14
Days Post Injection


Figure 8. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on kidney,
heart, lung, and anterior pituitary concentrations of E2-Q+ (upper
panel) and E2 (lower panel) in intact male rats. Intact male rats were injected with a single iv dose of 1.0 mg E2-CDS/kg bw and killed 1, 7, or 14 days after treatment. Each point represents the
group mean SEM (n = 6-7 rats for each time point).





88


Brain (upper panels) and anterior pituitary (lower panels) contents of the E2-Q+ and E2 following a single iv dose of the E2-CDS (1.0 mg/kg). Intact male rats were injected with a single iv dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14 days after treatment Whole brain tissue and the anterior pituitary were processed and assayed for E2-Q+ and E2 by the method described in Chapter 4. Each point represents the group mean SEM (n = 6-7 rats for each time point). These calculations assumed a brain wet weight of 2 grams (based on personal experience and knowledge).


Figure 9.




Full Text
3
barrier (BBB), thus enabling access to all cells and organs. Indeed, when these
hormones are used therapeutically to specifically target the CNS, the steroids
equilibrate among all body tissues due to their high lipophilicity (Pardridge &
Meitus, 1979). Moreover, when inside the CNS, there is no mechanism to
prevent their redistribution back to the periphery as blood levels of the
steroids decline (Davson, 1976; Schanker, 1965). So, even if estrogens can
easily gain access to the CNS, they are poorly retained by the brain. As a
result, only a fraction of the administered estrogen dose accumulates at or
near the site of action in the brain. This property of the estrogens necessitates
either frequent dosing or the administration of a depot form of the estrogen
in order to maintain therapeutically effective concentrations in the brain
(Schanker, 1965; Spona & Schneider, 1977). Both of these treatment strategies
lead to sustained increases in peripheral estrogen levels. Since estrogen
receptors are present in many peripheral tissues (Walters, 1985), where they
mediate a myriad of physiological and pharmacological effects (Murad &
Haynes, 1985), it further creates the potential of untoward peripheral
toxicities. In fact, constant increases in peripheral tissue exposure to estrogens
have been shown in numerous retrospective studies to precipitate various
peripheral toxicities, including increased risk of breast and endometrial cancer
(Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger et al., 1988; Persson, 1985;
Thomas, 1988), cardiovascular complications (Barrett-Conner et al., 1989;
Inman & Vessey, 1968; Kaplan, 1978; Thomas, 1988), and marked interference
with hepatic metabolism (Burkman, 1988).
In addition to the peripheral toxicities mentioned above, constant
exposure to high levels of E2 valerate has been shown to induce neuronal
degeneration in the hypothalamic arcuate nucleus of both male and female
rats (Brawer et al., 1980, 1983). Furthermore, other experimental conditions


174
administration in experimental animals in our studies, this route is not a
preferred avenue of drug administration in humans. However, it should be
mentioned that recently other routes of E2-CDS administration have been
examined, and it appears that oral mucosal or buccal route may be as effective
as iv route for E2-CDS administration.
In conclusion, the work presented in this dissertation demonstrate that
the redox-based dihydropyridine chemical delivery system is capable of
preferential delivery of E2 to the brain by "locking" it in the form of E2-Q+
which then allows to be released in slow and sustained manner. The results
further support the idea that the E2-CDS may be potentially useful in fertility
regulation and effective treatment of androgen-dependent prostatic diseases
by virtue of selective and sustained suppression of gonadotropin secretion,
and in treatment of brain estradiol deficiencies, i.e. postmenopausal
syndrome. In comparison to the currently used estrogenic medications, the
E2-CDS should achieve the sustained stimulation of brain E2 receptors at
lower doses and with less frequent dosing.


192
Neuwelt E.A. Implications of the blood-brain barrier and its manipulation.
Plenum Medical Book Co., New York (1989).
Neuwelt E.A. and P.A. Barrett. Blood-brain barrier disruption in the
treatment of brain tumors, animal studies. In: Implications of the
blood-brain barrier and its manipulation, Vol. 2, E.A. Neuwelt (ed) 107-
193, Plenum Medical Book Co. New York (1989).
Noteboom W.D. and J. Gorski. Stereospecific binding of estrogens in the rat
uterus. Arch. Biochem. Biophys. 111:559-568 (1965).
Notelovitz M. Climacteric medicine: Cornerstone for mid-life health and
wellness. Public Health Rep. [suppl.]:114-124 (1986).
Oldendorf W.H. The blood-brain barrier. Exp. Eye Res. [Suppl.] 25:177-190
(1977).
Oldendorf W.H., L. Braun, S. Hyman, and S.Z. Oldendorf. Blood-brain
barrier: Penetration of morphine, codeine, heroin, and methadone
after carotid injection. Science 178:984-986 (1972).
O'Malley B.W., W.L. McGuire, P.O. Kohler, and S.G. Korenman. Studies on
the mechanism of steroid hormone regulation of synthesis of specific
proteins. Recent Prog. Horm. Res. 25:105-111 (1969).
O'Malley B.W., D.R. Roop, E.C. Lai, J.L. Nordstrom, J.F. Catterall, G.E.
Swaneck, D.A. Colbert, M.J. Tsai, A. Dugaiczyk, and S.L. Woo. The
ovalbumin gene: Organization, structure, transcription, and
regulation. Recent Prog. Horm. Res. 35:1-46 (1979).
O'Malley C.A., R.D. Hautamaki, M. Kelley, and E.M. Meyer. Effects of
ovariectomy and estradiol benzoate on high affinity choline uptake,
ACh synthesis, and release from rat cerebral cortical synaptosomes.
Brain Res. 403:389-392(1987).
Onoda M. and P.F. Hall. Inhibition of testicular microsomal cyctochrome P-
450 (17 a-hydroxylase/C-17,20-lyase) by estrogens. Endocrinology
109:763-767 (1981).
Ottosson U.B. Oral progesterone and estrogen/progesterone therapy. Effects
of natural and synthetic hormones on subfractions of HDL cholestrol
and liver proteins. Acta. Obstet. Gynecol. Scand. [Suppl.] 12:5-37 (1984).
Ottosson U.B., K. Carlstrom, B.G. Johansson, and B. von Schoultz. Estrogen
induction of liver proteins and high-density lipoprotein cholestrol:


CHAPTER 1
INTRODUCTION
Estrogens exhibit a myriad of important regulatory roles in the growth,
development, and maintenance of the structures and functions which are
necessary for the continuation of the species. Their therapeutic applications
for certain clinical problems have been appreciated since the turn of the
century, when ovarian grafts were shown to prevent uterine atrophy and loss
of sexual function in castrated animals (Knauer, 1900). Estrogen hormones
have broad therapeutic applications and in most cases the steroids are used
primarily for their central actions (Meites & Nicoll, 1965). Among these are
the reproductive-related applications, including fertility regulation, sexual
dysfunction, and the replacement therapy in postmenopausal patients; and
the non-reproductive applications, including treatment of postmenopausal
depression and cancer therapy. Nevertheless, the full spectrum of potential
clinical benefits and applications of estrogen therapy has yet to be uncovered.
Physiologically, estrogen hormones exert two modes of action on the
central nervous system (CNS), particularly on the brain. First, during the
critical period of fetal /neonatal life, estrogens affect permanently some
features of the brain structure and function which result in neuronal
differentiation (Allen et al., 1989; Goy & McEwen, 1980). Second, during the
course of adult life, these hormones exert their effects in a modulatory and
reversible mode that influence a myriad of adult brain functions (McEwen,
1988; McEwen & Parsons, 1982).
l


Plasma FSH (ng/ml)
125
-7 0 7 14 21 28
Days Post Injection
Figure 17. Dose and time-dependent effects of the E2-CDS on plasma FSH
responses in ovariectomized rats. Animals received a single iv
injection of the E2-CDS on day 0 at doses of 0.01, 0.1, and 1.0
mg/kg. Also, the responses to an E2 dose of 0.7 mg/kg bw,
equimolar to the 1.0 mg/kg dose of E2-CDS, is shown for day 1 and
7. Represented are means SEM for n = 7 rats per group per
sampling time. The symbols indicate statistical differences as
follows: *) different from vehicle group (day 0); a) different from
0.01 mg/kg; and b) different from both 0.01 and 0.1 mg/kg. The
significance of interaction between factors (time and dose) was
determined by two-way analysis of variance (ANOVA). The
significance of differences among mean values at each dose level
was determined over time by one-way ANOVA and Dunnett's
test while the significance of differences among mean values of
three dose levels (at each time point) was determined by one-way
ANOVA and Scheffe F-test. The level of probability for all tests
was p<0.05.


140
level in the prostatic tissue (Daehlin et al., 1987). However, in spite of the
possible direct effects of estrogen at the cellular level in addition to the
indirect effects (CAST-like suppression of plasma T levels), the combined
treatment of prostate cancer patients with CAST and estrogens did not
improve survival when compared to either treatment alone (Blackard et al.,
1973, 1975). Further, estrogens have also been shown to decrease 17a-
dehydroxylase, 17 P-dehydrogenase and/or 17-20 desmolase (Kalla et al., 1980;
Onoda & Hall, 1981) activity in the Leydig cells of the testis. Inhibition of
these enzymes can increase the production of pregnenolone and progesterone
and the concomitant decrease in T synthesis in Leydig cells.
In conclusion, because of the potential application of brain-enhanced
estrogen delivery with sustained release in the brain as an alternative, we
conducted this study. Collectively, the results support the concept that the E2-
CDS may be useful in the treatment of androgen-dependent prostatic
hyperplasia. In comparison to the currently used estrogenic products, the E2-
CDS should achieve the sustained stimulation of brain E2 receptors at lower
doses or with less frequent dosing.


152
Morphine dependency was produced after initiation of estrogen
treatment as described previously (Katovich & O'Meara 1986; Simpkins &
Katovich, 1984). Briefly, one morphine pellet, containing 75 mg morphine
free base, was sc implanted at 17 days (long-term treatment, groups 1-3), or at 3
days (short-term treatment, group 4) after initiation of estrogen treatment.
Two days after the first morphine pellet, two additional morphine pellets
were sc implemented. This regimen of morphine treatment has been utilized
in our laboratory (Simpkins et al., 1983; Simpkins & Katovich, 1984) to
consistently produce typical symptoms of morphine dependency, tolerance,
and withdrawal (Wei et al., 1973). Four days after the initiation of morphine
treatment, on the morning of the twenty first day (long-term treatment) or
the seventh day (short-term treatment) after initiation of estrogen treatment,
animals were lightly restrained in wire mesh tunnel cages with a wooden
floor. TST was measured with a copper-constantan thermocouple that was
taped to the dorsal region of the tail at approximately 2 cm from its base so
that the thermocouple contacted the skin near the base of the tail. Rectal
temperature (RT) was measured with a copper-constantan thermocouple
inserted 6 cm beyond the anus and taped to the base of the tail. TST and RT
were recorded at 2-min intervals. Rats were allowed 1 hr to acclimate to the
restraining cages while control measurements were recorded. At the end of
the control period, rats were administered naloxone HC1 (0.5 mg/kg b.w., sc).
TST and RT were recorded for an additional 90 min in a room maintained at
24 1C. At the conclusion of the temperature study, all animals were killed
by decapitation and the trunk blood was collected in heparinized tubes. The
blood was centrifuged and the plasma separated and stored at -20C until
hormone analysis.


CHAPTER 10
GENERAL DISCUSSION
The studies in this dissertation evaluated the pharmacokinetics and
pharmacodynamic consequences of the redox-based estradiol-chemical
delivery system (E2-CDS) for the brain in the rat. The synopses of the major
findings drawn from these studies are (1) development of a technique for the
simultaneous quantitation of E2-Q+ and E2 that is reliable, sensitive and
applicable to a wide variety of tissues (Chapter 4; Rahimy et al., 1989a); (2)
documentation of the preferential deposition and retention of the E2-CDS
metabolites, E2-Q+ and E2, in the CNS tissue with a ti /2 = 8-9 days in male
(Chapter 5; Rahimy et al., 1988,1990a) and in OVX female rats (Chapter 6;
Rahimy et al., 1990b); (3) demonstration of the relatively rapid disappearance
of these metabolites from various peripheral tissues (Chapters 5 & 6; Rahimy
et al., 1988, 1990a,b); (4) determination of the prolonged suppression of
gonadotropin secretion LH and FSH in OVX female rats in a dose- and time-
dependent manner (Chapter 7; Rahimy et al., 1989b, 1990c); (5) demonstration
of the sustained suppression of T secretion and weight of androgen-
responsive tissues equivalent in magnitude to that of castration level
(Chapter 8); and (6) demonstration of significant attenuation of the naloxone-
induced surge in TST of morphine-dependent rats in the face of very low
plasma E2 levels (Chapter 9).
The formidable task of delivering the drugs of choice to the CNS has
long been recognized particularly by neuropharmacologists. This is because of
the unique feature of the brain, the BBB, that allows only lipophilic agents to
163


Plasma GH (ng/ml) Plasma PRL (ng/ml)
126
Figure 18. Dose and time-dependent effects of the E2-CDS on plasma PRL
(upper panel) and GH responses (lower panel) in ovariectomized
rats. Animals received a single iv injection of the E2-CDS on day 0
at doses of 0.01,0.1 and 1.0 mg/kg or an E2 dose of 0.7 mg/kg,
equimolar to the 1.0 mg/kg dose of E2-CDS, is shown for day 1 and
7 only. The symbols indicate statistical differences as follows: *)
different from vehicle group (day 0); a) different from 0.01 mg/kg;
and b) different from both 0.01 and 0.1 mg/kg.


201
Yalom I.D., D.T. Lunde, and R.H. Moos. Postpartum blues syndrome; A
description and related variables. Arch. Gen. Psychiatry 18:16-17 (1968).
Yen S.S.C. The biology of menopause. T. Reprod. Med. 18:287-289 (1977).
Yen S.S.C. The human menstrual cycle (integrative function of the
hypothalamic-pituitary-ovarian-endometrial axis). In: Reproductive
Endocrinology. S.S.C. Yen and R.B. Jaffe (eds), W. B. Saunders Co.,
Philadelphia (1978).
Zar J.H. Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ (1974).


37
hypothalamus. Furthermore, hypothalamic lesions prevent sexual behavior
in animals, even in the presence of adequate estrogen (Beyer et al., 1976).
Male sexual behaviors are composed of two distinct components: (1)
proception, or the awareness and pursuit of a receptive female and mounting
to achieve intromission, and (2) comsumation, or penile erection,
intromission and eventual ejaculation (Davidson, 1972). It is believed that
the expression of the proceptive components are dependent upon the
aromatization of testosterone to E2 in the brain, particularly in the regions of
the preoptic area of the hypothalamus and the amygdala (Beyer et al., 1976;
MacLusky et al., 1984). When E2 was implanted into the preoptic area of the
hypothalamus, it effectively restored mounting and intromissions in the
castrated rat (Beyer et al., 1976). Other studies have shown that the full
restoration of masculine sexual behaviors in castrated rats require E2 action in
the brain and dihydrotestosterone in the peripheral tissues (Lisk &
Greenwald, 1983). Given the aforementioned evidence regarding brain
mediation of sexual behaviors by E2, the E2-CDS may be potentially useful in
the treatment of sexual dysfunctions or psychological impotence that are not
caused by deficits in peripheral androgen-responsive tissue.
Disorders of Depression
It is generally believed that the underlying mechanism(s) of major
mood disorders (mania and depression) may include abnormal functions of
monoamine transmission (Bunney & Garland, 1982; Leonard & Kuschinsky,
1982). Pharmacological evidence suggests that mania is the result of
hyperactivity while depression is due to hypoactivity of monoamines.
Depression (unipolar and bipolar) respond well to antidepressant drugs, such


22
luteum. In the absence of gonadal steroids such as following ovariectomy or
menopause, the negative feedback effects of estrogens on LHRH and
gonadotropins are removed and thus, serum LH and FSH levels increase.
From the clinical point of view, constant exposure to estrogens (or
maintenance of their elevated levels) prevents the preovulatory surge of LH
by exerting a negative feedback mechanism on the hypothalamic-pituitary
unit. This strategy represents the primary mechanism by which the E2
component of contraceptives prevents ovulation. Collectively, the effects of
E2 on the hypothalamic-pituitary unit depend on the exact duration or
magnitude (or both) of exposure to the hormone.
Role of Estrogen in the Rat Estrous Cycle
Rats have 4 to 5 days of estrous cycle (Long & Evans, 1922). The estrous
cycle, like the menstrual cycle in human and nonhuman primates, represents
an extraordinary sequence of events in hormonal and behavioral changes,
and it is verified by cyclic changes in vaginal cell morphology (Long & Evans,
1922). The normal cycle consists of one day of estrus, followed by two days of
diestrus (I and II), and one day of proestrus. The dynamic relation between
the hypothalamic releasing factor, pituitary gonadotropins, and ovarian
steroids during the estrous cycle have been reviewed in considerable detail
(Kalra & Kalra, 1983). Estrogen hormones are responsible for the
maintenance of the estrous cycle, as in the menstrual cycle. On the evening
of estrus, the serum E2 concentrations reach their lowest levels (15 to 20
pg/ml) as the corpus luteum involutes. However, during the days of diestrus
as follicular growth and maturation progress under the influence of FSH
stimulation, ovaries produce increasing concentrations of estrogens. E2 levels


36
premenstrual syndrome (PMS) reported 37% had a craving for sweets during
the luteal phase of their menstrual cycles. Increased appetite was reported to
be a frequent PMS symptom in 45 women examined by Fortin et al. (as
reported by Smith & Sauder, 1969). Pliner and Fleming (1983) studied 34
women and observed that the decreased food intake during the follicular
phase was associated with significant weight loss and that the increased food
intake during the luteal phase was concurrent with weight gain. Collectively,
endogenous estrogen in women has consistently been shown, albeit subtle, to
have a suppressive effect on food intake and body weight.
The potential feasibility of E2-CDS to reduce body weight has been
extensively evaluated in the rat (Estes et al., 1988; Simpkins et al., 1988,
1989a,b). Interestingly, a separation between the effects of E2-CDS on body
weight and food intake has been demonstrated. That is, despite weight
reduction, no consistent reduction in food intake has been observed,
indicating that mechanisms other than reduced appetite are responsible for
the weight loss. Collectively, available data indicate that the E2-CDS
chronically suppresses body weight following a single administration in the
rat. Furthermore, the compensatory hyperphasia in response to the weight
loss is prevented by the E2-CDS.
Libido/Sexual Dysfunction
Certain regions of the hypothalamus have been identified to be
involved in the central integration of sexual behaviors (Christensen &
Clemens, 1974). The first direct evidence that sex steroids influence the
central structures associated with mediating and integrating sexual behaviors
was presented when Harris implanted estrogen directly into the cat's


84
Figure 6. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on brain
(upper panels) and plasma (lower panels) concentrations of E2-Q+
and E2 in intact male rats. Intact male rats were injected with a
single iv dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1,
7, or 14 days after treatment. Whole brain tissue and plasma were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Each point represents the group mean SEM (n = 6-7
rats for each time point).


Table 9: Effects of the E2-CDS (0.5 mg/kg bw) or CAST on Plasma Hormone Concentrations
at 7 Days after the Last Treatment in Male Rats.
Treatment
Group
Inject.
No.
E2
(pg/ml)
E?-Q+
(pg/ml)
LH
(ng/ml)
FSH
(ng/ml)
PRL
(ng/ml)
Intact
-
12.3 2.70
ND
0.50 0.07
10.341.88
28.1514.91
CAST
-
13.06 2.85
ND
5.77 0.42a
27.5711.00a
5.8911.06
E2-CDS
1
40.45 3.39
578.3 46.9
0.30 0.03b
6.6510.56b
35.9715.67
e2-cds
2
52.67 7.69a,b
783.5 90.9
0.28 0.02b
6.5310.56b
43.2116.69
e2-cds
3
70.63 18.29a,b
654.8 54.4
0.39 0.07b
7.3310.98b
80.64 112.88a,b
ND Not determined
a Different from Intact
b Different from CAST
146


LIST OF TABLES
Table Page
1. Recovery and Precision Determinations for Biological Samples
Spiked with E2 72
2. Recovery and Precision Determinations for Biological Samples
Spiked with E2-Q+ 73
3. Percent Hydrolysis of E2-Q+ in Supernatants of a Variety of
Tissues 74
4. Effects of Dose on the Extent of Oxidation and Hydrolysis of E2-
CDS in a Variety of Tissues in vivo Ill
5. Effects of the E2-CDS on the Clearance of E2-Q+ from a Variety of
Tissues 112
6. Effects of the E2-CDS on the Clearance of E2 from a Variety of
Tissues 113
7. Effects of an Equimolar Dose of E2 on the Tissue Concentrations
of E2 114
8. Dose and Time-Dependent Effects of the E2-CDS on Peripheral
Tissue Weights in Ovariectomized Rats 127
9. Effects of the E2-CDS or CAST on Plasma Hormone
Concentrations at 7 Days after the Last Treatment in Male Rats 146
10. Effects of the E2-CDS or CAST on Plasma Hormone
Concentrations at 14 Days after the Last Treatment in Male Rats 147
11. Effects of the E2-CDS or E2 Pellet on Basal Temperature, Maximal
Change in TST, and Area Under the 90 Min TST Curve in
Ovariectomized, Morphine-Dependent Rats 161
12. Effects of the E2-CDS or E2 Pellet on Plasma Hormone
Concentrations in Ovariectomized, Morphine-Dependent Rats 162
viii


190
McCann S.M. Physiology and pharmacology of LHRH and somatostatin.
Ann. Rev. Pharmacol. Toxicol. 22:491-515 (1982).
McEwen B.S. Steroid hormones and the brain: Linking "nature" and
"nurture". Neurochem. Res. 13:663-669 (1988).
McEwen B., P. Biegon, P. Davies, L.C. Krey, V.N. Luine, M.Y. McGinnus, C.M.
Paden, B. Parson, and T.C. Rainbow. Steroid hormones: Hormonal
signals which alter brain cell properties and functions. Recent Prog.
Horm. Res. 38:41-92(1982).
McEwen B.S., A. Biegon, C.T. Fischetee, V.N. Luine, B. Parsons, and T.C.
Rainbow. Towards a neurochemical basis of steroid hormone action.
In: Frontiers in Neuroendocrinology, L. Martini and W. Ganong (eds)
1153-1176 (1984).
McEwen B.S. and B. Parsons. Gonadal steroid action on the brain:
neurochemistry and neuropharmacology. Ann. Rev. Pharmacol.
Toxicol. 22:555-598 (1982).
Meites J. and C.S. Nicoll. Hormonal steroids, biochemistry, pharmacology
and therapeutics: proceedings of the first international congress on
hormonal steroids. Vol. 2, Academic Press, New York (1965).
Meldrum D.R., I.M. Shamoni, A.M. Frumar, I.V. Tataryn, R.J. Chang, and
H.L. Judd. Elevation in skin temperature of the finger as an objective
index of postmenopausal hot flushes: Standardization of the
techniques. Am T Obstet. Gynecol. 135:713-717 (1979).
Mendel C.M., R.A. Weisiger, and R.R. Cavalieri. Uptake of 3,5,3'-
triiodothyronine by the perfused rat liver: Return to the free hormone
hypothesis. Endocrinology 123:1817-1824 (1988).
Michael C.H., H.I. Kantor, and H. Shore. Further psychometric evaluation of
older womenthe effect of estrogen administration. T. Gerontol.
25:337-341 (1970).
Michael R.P., R.W. Bonsall, and H.D. Rees. The nuclear accumulation of pH]
testosterone and pH] estradiol in the brain of the female primate:
Evidence for the aromatization hypothesis. Endocrinology 118:1935-
1944 (1986).
Millard W.J., T.M. Romano, N. Bodor, and J.W. Simpkins. GH secretory
dynamics in animals administered estradiol utilizing a chemical
delivery system. Pharm. Res. 7 (11) (in press, 1990).


159
-2 i 1 1 1 1 1 1 1 1 1 1 1 1 1
-40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100
Time (min)
Figure 24. Effects of the E2-CDS or E2 pellet on the mean TST responses
induced by naloxone administration (0.5 mg/kg; sc) to morphine-
dependent ovariectomized rats. Animals were treated weekly
with either the vehicle (HPCD, x3, 3 injections total over 3 weeks),
E2-CDS (1.0 mg/kg, x3, 3 injections total over 3 weeks), E2 pellet
(0.5 mg, x3, 3 implants total over 3 weeks), or E2-CDS (1.0 mg/kg,
xl, 1 injection for 1 week). Morphine dependency was produced
after initiation of estrogen treatment. Four days after the
initiation of morphine treatment, on the morning of the 21st day
(long-term) or the 7th day (short-term) after the initiation of
estrogen treatment, animals were lightly restrained in wire mesh
cages. TST was then recorded at 2-min intervals. Represented are
the means SEM for n=7-8 rats per group.


ng/g ng/g ng/g
103
HYPOTHAL. E2-Q+
HYPOTHAL. E2


53
methodology by determining simultaneously the levels of both E2-Q+ and E2
in several tissues following administration of E2-CDS in the rat.
In Vitro Methodology
Specificity of the estradiol antibody for E?
The presence of the estradiol conjugate (E2-Q+) in biological tissues
could lead to overestimation of tissue E2 concentrations if E2-Q+ cross-reacts
with the E2 antibody. Therefore, possible cross-reaction with the E2-Q+ was
examined by adding E2-Q+ standards to the RIA of E2 at concentrations of 2, 5,
15, 50,180, and 360 ng/ml. These doses of E2-Q+ were 100-fold greater than the
corresponding E2 standards used to generate standard curves in the RIA of E2.
Selective solvent extraction of steroids from tissues
Tissues including brain, anterior pituitary, kidney, lung, liver, heart,
and adipose tissue were dissected from adult male Charles River (CD) rats
(Charles River Breeding Laboratories, Wilmington, MA) immediately
following decapitation. All tissues were rinsed in ice-cold saline and stripped
of surrounding connective tissue and fat, blotted dry on paper then weighed
to the nearest 0.1 mg. Tissue samples of known wet weight were then
homogenized (using a Brinkman Polytron Homogenizer; Model PT 10/35) at
moderate speed (setting at 6 for two 15-second periods) in an appropriate
solvent system (depending on the type of tissue and compound of interest to
be extracted). The appropriateness of the solvent system was determined by
screening various organic solvents for effective extraction and acceptable
recoveries of the steroids, with high reproducibility in the assay procedure.
Tissue homogenate pools, at a concentration of 100 mg tissue/ml solvent,
were prepared as follows: for E2 extraction from brain, anterior pituitary,


189
Luine V.N. Estradiol increases choline acetyltransferase activity in specific
basal forebrain nuclei and projection areas of female rats. Exp. Neurol.
89:484-490 (1985).
Luine V.N., R.I. Khylchevskaya, and B. McEwen. Effect of gonadal steroids on
activities of monoamine oxidase and choline acetylase in rat brain.
Brain Res. 86:293-306 (1975).
Luine V.N. and B.S. McEwen. Sex differences in cholinergic enzymes of
diagonal band nuclei in the rat preoptic area. Neuroendocrinologv
36:475-482 (1983).
Luine V.N., D. Park, T. Joh, D. Reis, and B. McEwen. Immuno-chemical
demonstration of increased choline acetyltransferase concentration in
rat preoptic area after estradiol administration. Brain Res. 191:273-277
(1980).
MacLusky N.J., A. Philip, C. Hurlburt, and F. Naftolin. Estrogen metabolism
in neuroendocrine structures. In: Metabolism of Hormonal Steroids
in the Neuroendocrine Structures, F. Celotti, F. Naftolin, and L. Martini
(eds) 103-116, Raven Press, New York (1984).
Maggi A. and J. Perez. Role of female gonadal hormones in the CNS: Clinical
and experimental affects. Life Sci. 37:893-906 (1985).
Majewska M.D. Steroids and brain activity. Essential dialogue between body
and mind. Biochem. Pharmacol. 36:3781-3788 (1987).
Malleson J. An endocrine factor in certain affective disorders. Lancet ii:158-
164 (1953).
Marshall J.C. and R.P. Kelch. Gonadotropin-releasing hormone: Role of
pulsatile secretion in the regulation of reproduction. New Eng L Med.
315:1459-1468 (1986).
Maus M., J. Cordier, J. Glowinski, and J. Premont. 17 p-oestradiol
pretreatment of mouse striatal neurons in culture enhances the
responses of adenylate cyclase sensitive to biogenic amines. Eur. T.
Neurosci. 1:154-161 (1989).
Mayer S.E., R.P. Maickel, and B.B. Brodie. Disappearance of various drugs
from the cerebrospinal fluid. T. Pharmac. Exp. Ther. 128:41-43 (1960).
Mays E.T., W.M. Christopherson, M.M. Mahr, and H.C. Williams. Hepatic
changes in young women ingesting contraceptive steroids. TAMA
235:730-732 (1976).


161
Table 11: Effects of the E2-CDS or E2 pellet on Basal Temperature, Maximal
Change in TST, and Area Under the 90 Min TST Curve in Ovariectomized,
Morphine-Dependent Rats
Treatment
Group
Basal RT
Basal TST
Max A TST
AUC
HPCD
(3)*
39.8 0.2
(C)
26.4 0.1
6.4 0.2
(C-Min)
335.5 24.9
E2 Pellet
(3)
38.7 0.2
26.2 0.1
6.4 0.1
361.8 19.4
E2-CDS
(3)
39.1 0.3
26.9 0.4
3.4 0.6a,b
197.8 35.3a,b
e2-cds
(1)
38.6 0.3
26.5 0.1
4.9 0.5
291.5 32.8
* Number of injection(s)/pellet(s)
a Different from HPCD group
b Different from E2-pellet group


20
follicular phase. The dynamic changes of the ovaries have a periodicity of
once every 28 days; however, these events are, in turn, regulated by the cyclic
changes of pituitary gonadotropins (Ross, 1985; Schwartz, 1981; Yen, 1978).
The pattern of gonadotropin secretion is, in turn, maintained by the pulsatile
mode of the hypothalamic LHRH secretion (Levine & Ramirez, 1982; Marshal
& Kelch, 1986). Finally, the whole cascade of events is regulated by negative
and positive feedback effects of ovarian hormones (McCann, 1982; Pohl &
Knobil, 1982). In a sense, the dynamic interplay between the brain and
pituitary and the dynamics of the ovarian feedback mechanisms govern the
reproductive cycle. That is, the momentum gained from one phase of the
cycle powers the next phase, and continues on into the next cycle.
Feedback regulation
The hypothalamic-pituitary-ovarian-endometrial axis in females
changes markedly with the menstrual cycle. In humans and nonhuman
primates, the cyclic pattern of gonadotropin secretion (the LH surge
particularly) seems to depend more on ovarian estrogen than neural signals.
This is evidenced by the observation that menstrual cycles in primates were
maintained with a constant dose of LHRH per pulse with no variation in
amplitude or frequency of LHRH administration (Knobil, 1980). These
observations suggested that the hypothalamus may only play a permissive
role in the control of the menstrual cycle (Knobil, 1980). However, more
recent observations in human and nonhuman primates indicate that normal
menstrual cyclicity does depend on LHRH pulse parameter that changes in
frequency and amplitude during the menstrual cycle (Dalkin et al. 1989; Ferin
et al., 1984; Marshal & Kelch, 1986). Studies in animals as well as in humans
have shown that LHRH and gonadotropins are secreted in episodic fashion


73
Table 2: Recovery and Precision Determinations for Biological Samples
Spiked with E2-Q+
Tissue*
E2-Q+ Added
E2-Q4- Assayedb
(Pg)
EEc
(%)
Recovery
(%)
CVd
(%)
CCe
(r)
Brain
150
73.7
74.5
49.2
1.3
300
129.6
65.1
43.2
4.6
600
249.9
63.1
41.7
4.6
0.994
Plasma
150
47.5
ND
31.7
6.1
300
87.8
ND
30.0
2.8
600
208.3
ND
34.7
0.9
0.989
Kidney
150
68.4
73.4
45.6
1.0
300
137.9
74.0
46.0
3.0
600
238.3
64.4
40.0
2.2
0.990
Lung
150
64.0
76.3
42.6
1.0
300
131.1
78.2
43.7
2.6
600
236.7
71.6
40.0
1.1
0.995
Heart
150
52.6
64.8
35.0
2.6
300
104.0
64.1
34.7
4.6
600
218.5
67.4
36.4
5.9
0.989
Liver
150
80.4
80.4
53.6
2.2
300
164.0
82.0
54.7
6.7
600
319.2
79.8
53.2
7.3
0.988
Fat
150
39.8
70.9
26.6
2.1
300
78.9
70.2
26.3
2.6
600
163.3
72.7
27.2
3.0
0.988
a 100 mg of tissue or 1 ml of plasma was used,
b Mean of n = 4 for each dose of E2-Q+ in each tissue,
c EE = extraction efficiency,
d CV = coefficient of variation,
e CC = correlation coefficient,
f ND = not determined.


Ill
Table 4: Effects of Dose on the Extent of Oxidation and Hydrolysis of E2-CDS
in a Variety of Tissues in vivo
Tissue
Oxidation
Hvdrolvsisb
(-fold increase)3
0.01
E2-CDS Dose (mg/kg)
0.1
1.0
Brain
176
24
23
13
Hypothalamus
112
22
23
19
Ante. Pituitary
NDc
40
31
ND
Plasma
147
14
4
2
Kidney
103
29
22
27
Lung
116
47
32
24
Heart
125
25
21
29
Liver
73
13
15
6
Fat
77
35
22
7
Uterus
21
31
32
26
a Fold increase in E2-Q+ concentrations over a 100-fold increase in the E2-
CDS dose at day 1 after injection (the first sampling time),
b Average % hydrolysis (fraction of E2-Q+ hydrolyzed + total E2-Q+ x 100)
on the first sampling time (day 1).
c Not determined due to incorrect extraction of the tissue at day 1.


118
equimolar E2 dose reduced plasma FSH by 27% at day 1 and 19% at day 7
(Figure 17).
Plasma concentrations of LH and FSH in animals treated with lower
doses of the E2-CDS (0.01 and 0.1 mg/kg) began to gradually increase after 7
days of drug administration (Figures 16 &17).
Plasma PRL concentrations in animals treated with 1.0 mg E2-CDS dose
were increased by 4-, 8-, 13-, and 8-fold at 1, 7,14 and 21 days, respectively or
were at preinjection levels by 28 days after drug administration (Figure 18,
upper panel). Lower doses of the E2-CDS did not effect PRL concentrations.
By contrast, the 0.7 mg/kg dose of E2 increased plasma PRL concentrations by
3-fold on day 1 and PRL returned to preinjection levels by day 7 after drug
administration (Figure 18, upper panel).
Plasma GH concentrations were not altered in response to E2 or E2-CDS
at any dose or time point evaluated (Figure 18, lower panel).
Anterior pituitary weights increased in a dose- and time-dependent
manner in response to E2-CDS administration (Table 8). With the lower
doses of the E2-CDS (0.01 and 0.1 mg/kg), pituitary weights were slightly
increased (22 to 32%) over control group weights by 14 days postinjection, but
they returned to control levels by day 21. However, the highest dose of the E2-
CDS increased pituitary weights significantly from day 7 to day 28 relative to
weights at time 0 and following treatment with lower doses of E2-CDS. The
maximum pituitary gland stimulation occurred at 14 days postinjection and
then pituitary weights began to decrease but remained elevated at 28 days after
the drug administration (Table 8).
Similarly, uterine weights showed a dose- and time-dependent increase
in response to E2-CDS administration (Table 8). Uterine weights were
increased by 20, 54, or 82% on day 1 following treatment with the E2-CDS at


ASSAY DEVELOPMENT, TISSUE DISTRIBUTION
AND PHARMACODYNAMICS OF A NOVEL ESTROGEN-CHEMICAL
DELIVERY SYSTEM FOR THE BRAIN
BY
MOHAMAD H. RAHIMY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1990


60
determination. Also, the rightward shift in the tissue weight-RIA response
curve at day 7 indicates a decrease in brain E2 concentrations with increasing
time after drug administration.
Recovery of E9-O+
The recovery of E2-Q+ is dependent upon the efficiency of three
processes: (a) the extraction efficiency of E2-Q+ from tissues with
watenacetone; (b) the % hydrolysis achieved under basic conditions; and (c)
the % recovery of E2 following its formation from the hydrolysis of E2-Q+.
The recovery of E2 was determined as described above and the parameters
described below after the adjustment for E2 recovery. The recovery of E2-Q+
was determined as the percent of the spiked concentration of E2-Q+ which was
assayed. The % hydrolysis of E2-Q+ was determined experimentally for each
tissue and the extraction efficiency in watenacetone was calculated as the
recovery times the reciprocal of the % hydrolysis.
The extraction efficiency of E2-Q+ with watenacetone ranged from
65.4% for heart to 80.7% for liver tissue (Table 2). For each tissue evaluated,
the % hydrolysis of E2-Q+ was the primary factor limiting the recovery of this
species. The % hydrolysis ranged from 69% in liver to 37% in adipose tissue
(Table 3). For all tissues except fat and plasma, the % hydrolysis was greater
than 50% (Table 3).
Precision of the E7-O+ extraction-assay method
The precision of the E2-Q+ method was determined using the same
parameters used for demonstrating precision of the E2 method.
The CV for the quadruplicate determinations of E2-Q+-spiked tissue
homogenates or plasma pools ranged from 0.9 to 7.3% with a majority of the
E2-Q+ doses for each tissue showing a CV of less than 5.0% (Table 2).


Table 12. Effects of the E2-CDS or E2 Pellet on Plasma Hormone Concentrations in
OVX, Morphine-Dependent Rats.
Treatment
Group
2
E?-Q+
m
FSH
PRL
(pg/ml)
(pg/ml)
(ng/ml)
(ng/ml)
(ng/ml)
HPCD (3)*
3.67 0.95
ND
10.11 2.77
40.53 3.24
15.49 9.93
E2 Pellet (3)
89.80 9.97a,b
ND
0.90 0.19a
13.26 1.23a,c
960.50 100.4a,c
E2-CDS (3)
44.10 6.37a,c
866.66 53.42c
0.43 0.08a
12.16 1.69a,c
677.64 86.72a,c
E2-CDS (1)
13.45 3.53
224.12 17.88
0.57 0.09a
20.43 1.06a
279.50 96.85
ND Not determined
* Number of injection(s)/pellet(s)
a Different from HPCD group
b Different from E2-CDSO) or E2-CDSO) group
c Different from E2-CDSO) group
162


34
indicated an increased risk of peripheral toxicides, including the risk of breast
and endometrial cancer (Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger
et al., 1988; Persson, 1985; Trapido et al., 1984), cardiovascular morbidity
(Barrett-Conner et al., 1989; Kaplan, 1978; Thomas, 1988), and interference
with hepatic metabolism (Burkman, 1988). These adverse effects of estrogens
are dose dependent. Currently used estrogens are administered either in
frequent doses, or as a depot form, in order to maintain therapeutically
effective levels in the brain. Both of these treatment strategies lead to
sustained increases in peripheral estrogen levels and thus peripheral
toxicities. Since currently employed estrogen therapy is contraindicated in
many postmenopausal women, and in as much as some women do not
respond to the existing steroid medications, the brain-enhanced E2-CDS with
sustained release of E2 may be more effective in the treatment of menopausal
symptoms by providing sufficient E2 to the brain while avoiding peripheral
toxicities.
Prostatic Cancer
The primary objective of hormone therapy in prostate cancer patients
is to induce an effective androgen suppression, thus abolishing the growth
promoting effects of androgens on the diseased prostate ( Brendler, 1988;
Cabot, 1896; Huggins & Hodges, 1941; Isaacs et al., 1983; Moore, 1944; White,
1895). Currently a variety of surgical and therapeutic means for inhibiting
androgen production (in the testis) or blocking androgen action (in the
prostate) are being used. These include castration, high-dose estrogen
therapy, GnRH analogues, and antiandrogens (Santen & English, 1989).
Castration or high-dose estrogen therapy remain, however, as the treatment


35
of choice for the endocrine-dependent management of prostatic cancer (van
Steenbrugge et al., 1988). Both treatments are reported to be equally effective
in (i) suppressing the circulating testosterone levels (Carlstrom et al., 1989);
and (ii) controlling the symptoms of advanced prostatic cancer in 70-80% of
patients (Klein, 1979). In contrast to castration, high-dose estrogen therapy
inhibits (via negative feedback mechanism) the hypothalamic-pituitary-
gonadal function leading to chemical castration. However, high-dose
estrogen therapy has been shown to cause severe cardiovascular
complications in patients (Henriksson & Edhag, 1986) due to the alterations in
liver metabolism (von Schoultz et al., 1989). The E2-CDS may be potentially
useful in the treatment of androgen-dependent prostatic cancer by the virtue
of its sustained suppression of the hypothalamic-pituitary-testis function
leading to chemical castration and thus regression of the tumorous tissue.
Body Weight Regulation
Food intake and body weight may vary during the estrous cycle of the
rat (Tarttelin & Gorski, 1971) and the menstrual cycle of primates (Czaja, 1978)
including women (Pliner & Fleming, 1983). A consistent observation is that
food intake and body weight decrease during the follicular phase of the
ovarian cycle when circulating E2 levels increase. Conversely, food intake
and body weight increase during the luteal phase of the ovarian cycle when E2
levels decline and progesterone levels elevate.
Although relatively few studies have evaluated the potential
modulatory effects of E2 on body weight regulation in human subjects, the
available data support a suppressive role of E2 in appetite and body weight.
Morton et al. (1953) in a study involving menstruating women with


49
GH, and PRL values are expressed as ng/ml of either the LH-RP-2, FSH-RP-2,
GH-RP-2 or the PRL-RP-3 reference standards, respectively. Values for
unknown plasma samples were derived from the 10 to 90% (linear inhibition
portion) of the respective standard curves. Radioiodinations of the labeled
hormones were performed in our laboratory using standard procedures for a
chloramine-T iodination with gel filtration chromatography to separate free
iodine from hormone-bound iodine.
The ranges of hormone assay detectability were (1) 0.25 to 20 ng/ml for
LH in 50 ul; (2) 2.5 to 200 ng/ml for FSH in 100 ul; (3) 2.5 to 320 ng/ml for GH
in 25 ul; and (4) 0.25 to 50 ng/ml for PRL in 50 ul of plasma sample. Plasma
samples containing undetectable LH, FSH, GH, or PRL were assigned the
respective assay sensitivity of 0.25, 2.5, 2.5 and 0.25 ng/ml for these hormones.
All samples for each hormone in an experiment were assayed in a single run.
Steroid Hormone Assays
Coat-A-Count Estradiol kitsemploying solid-phase [125I]-
radioimmunoassaydesigned for the quantitative measurement of E2 in
serum were purchased from Diagnostic Products Corporation (Los Angeles,
CA). Each kit is equipped with human serum-based standards having E2
values ranging from 20 to 3600 pg/ml (0.07 to 13.2 nmol/1) (technical
information from Diagnostic Products). The cross-reactivity of the E2
antibody has been reported to be <0.3% for E2-Q+ at the concentration of 15
ng/ml and higher (Rahimy et al., 1989a). The cross-reactivity for estriol and
estrone has been reported to be 0.32% and 1.1%, respectively (technical
information from Diagnostic Products).


137
inhibiting further LHRH and LH release. Recent studies suggest that E2, the
aromatized metabolite of T, is responsible in mediating the feedback effect of
T on the hypothalamic neuronal system (Christensen & Clemens, 1974).
Several lines of evidence have accrued to indicate that E2 formation in the
CNS is important for the effect of T. First, aromatization inhibitors decreased
the effectiveness of T in inducing sexual behavior in CAST rats (Beyer et al.,
1976). Second, intrahypothalamic implant of E2 effectively restored the
masculine sexual behavior in long-term CAST male rats (Christensen &
Clemens, 1974). Finally, estrogen receptors and binding sites have been
identified in the brain of male rats (Krey et al., 1980). Thus, prolonged
exposure to estrogen hormone, whether by E2-CDS or frequent
administration of currently available estrogens, produces sustained
suppression of hypothalamic LHRH secretion and hence, LH inhibition.
Eventually, this leads to T deprivation or chemical castration.
Our results show that, like CAST, the E2-CDS produced a significant T
deprivation which remained suppressed for at least 14 days with a single
injection, or for 28 days with 3 injections given once every 7 days for three
consecutive weeks. Similarly, a single injection of E2-CDS was sufficient to
significantly regress both the ventral prostate and the seminal vesicles
weights during the first week of treatment, while the 3 injections paradigm
produced significant regression (equivalent to CAST) of these two tissues for
at least 4 weeks. The profound suppression at 7 days of circulating T
concentrations and/or of tissue weights (prostate and seminal vesicles) caused
by the single or double injection of E2-CDS, gradually began to recover at 14
days after the final treatment. However, these values remained significantly
suppressed compared with the control values at 14 days after the last
injection. Furthermore, unlike CAST, E2-CDS treatment resulted in pituitary


169
the female heart tissue, contains E2 inducible enzymes, i.e. cytochrome P450
monooxigenases that metabolize E2, this may be the reason why this tissue
exhibits faster clearance of E2-CDS metabolites.
An important issue that requires further explanation is the initial
distribution phase of the E2-CDS throughout the body before its oxidation to
E2-Q+. Since the iv route was chosen here for drug administration, this route
completely eliminates the process of absorption. Thus, the extent of
distribution or accessibility of E2-CDS to its ultimate site of action is
determined by its ability to cross the capillary endothelial cells, then by the
rate of blood flow through organs and tissues and, finally, since the parent
drug likely acts intracellularly, by its rate of diffusion across the cellular
plasma membranes. Since the E2-CDS is very lipophilic in nature, it is
capable of readily traversing the capillary endothelia and the plasma
membranes, thus, reaching inside the cells if it survives metabolism and
elimination during the initial period of distribution. Therefore, of major
concerns are the general factors which may influence the amount of E2-CDS
eventually reaching and residing in different tissues. Since we do not know
the extent of E2-CDS binding to plasma proteins and other tissue components,
for the present, we assume that these important issues are not critical. Thus,
this general treatment of E2-CDS kinetic behavior permits us to draw the
following conclusions regarding the initial tissue distribution of E2-CDS.
First, the quantity of E2-CDS reaching an organ or tissue represents a small
fraction of the total amount of drug administered. Second, the greater the
rate of blood flow to an organ or tissue the higher the quantity of E2-CDS
reaching that organ or tissue. Third, the greater the extraction efficiency from
the circulation of E2-CDS by an organ or tissue the higher the quantity of E2-
CDS that is taken up by that organ or tissue. It should be emphasized that


CHAPTER 3
GENERAL MATERIALS AND METHODS
Drugs and Solutions
Estradiol and Standard Solution
Estradiol-17(3 (E2) was purchased from Steraloids, Inc. (Wilton, NH).
Standard solutions of E2 were prepared in ethanol for the in vitro studies
involving methodology development. Solutions of E2 were stored at -20C
(stock solution) or 4C (working standard).
Estradiol-17(3 incorporated in 2-hydroxypropyl-(3-cyclodextrin (HPCD)
was provided by the Pharmatec, Inc. (Alachua, FL). Aqueous solution of E2
was prepared on the day of experiment in 20% HPCD (wt:vol) for in vivo
injection.
Estradiol-Chemical Delivery System
Estradiol-chemical delivery system, E2-CDS (3-hydroxy-17P-{[(l-methyl-
l,4-dihydropyridine-3-yl)carbonyl] oxy}-estra-l,3,5-(10)-triene) and E2-Q+ (1-
methyl-3-{[(3-hydroxyestra-l,3,5-(10)-triene-17p-yl)oxy]carbonyl} pyridinium
iodide) were synthesized as previously reported (Bodor et al., 1987). Briefly,
the 3,17p-dinicotinate ester of E2 was made by refluxing 17P-E2 with
nicotinoyl chloride or nicotinic anhydride in pyridine. This derivative was
selectively hydrolyzed to the 17-monoester of E2 with potassium bicarbonate
in 95% methanol. The monoester of E2 was then quaternized with methyl
44


TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
LIST OF TABLES viii
LIST OF FIGURES ix
ABSTRACT xi
CHAPTERS
1. INTRODUCTION 1
2. REVIEW OF THE LITERATURE 7
Estrogen Hormones 7
Historical Observations 7
Endocrinological and Biochemical Considerations 8
Biosynthesis 8
Secretion and transport 10
Metabolism and excretion 11
Mechanism of action 12
Estrogen Receptors 15
Intracellular/cytosolic receptors 15
Membrane receptors 16
Estrogen-receptor binding kinetics 17
Role of Estrogen in the Menstrual Cycle 19
Feedback regulation 20
Role of Estrogen in the Rat Estrous Cycle 22
Feedback regulation 23
Blood-Brain Barrier 25
Historical Overview 25
Potential Asset to Utilize in the Design of Brain-Specific Drug
Delivery 28
Therapies Aimed at Targeting/Enhancing Brain Estradiol Levels 32
Fertility Regulation 32
Menopausal Syndrome 33
Prostatic Cancer 34
Body Weight Regulation 35
Libido / Sexual Dysfunction 36
Disorders of Depression 37
Cognitive Impairment of Menopausal Alzheimer's Type 41
v


132
To evaluate the effects of E2-CDS or CAST on androgen and androgen-
dependent sex accessory organs, a total of 5 different treatment conditions
were evaluated: 1) intact control group; animals received no treatment; 2)
CAST group; animals received no further treatment; 3) intact + E2-CDS (xl)
group; animals received a single iv injection (tail vein) of E2-CDS (0.5 mg/kg
bw); 4) intact + E2-CDS (x2) group; animals received E2-CDS (0.5 mg/kg bw, iv)
once a week for 2 consecutive weeks; 5) intact + E2-CDS (3x) group; animals
received E2-CDS (0.5 mg /kg bw, iv) once a week for 3 consecutive weeks.
Animals were then sampled 7 or 14 days after the last treatment (treatment
conditions 2 to 5).
Rats (7-8 per group) were sacrificed by decapitation either 7 or 14 days
after the final administration of E2-CDS or post CAST (treatment conditions
2-5). Immediately following decapitation, the trunk blood was collected in
heparinized glass tubes on ice. The blood was centrifuged, and the plasma
was separated and stored at -20C for subsequent plasma hormone analysis.
Also, the following tissues: prostate (the ventral lobe), seminal vesicles (both
horns), testis, and anterior pituitary were rapidly dissected. The dissection of
accessory sex organs was carried out according to previously reported
guidelines (Lee, 1987). The ventral prostate was carefully grasped with forceps
and dissected to the level of the ducts entering ventrally into the urethra.
The seminal vesicles were retracted and expelled of seminal fluid. All tissues
were blotted dry on paper then weighted to the nearest 0.1 mg.
Plasma LH, FSH, and PRL concentrations were measured in duplicate
by RIA using NIDDK kits. Plasma LH, FSH, and PRL values are expressed as
ng/ml of the LH-RP-2, FSH-RP-2 or the PRL-RP-3 reference standards,
respectively, The intra-assay coefficients of variation were 3.37, 2.27, and
4.07% for LH, FSH, and PRL assays, respectively. Plasma samples containing


57
In Vivo Studies
To evaluate the applicability and reliability of this procedure to in vivo
condition, adult male Charles River (CD) rats received a single intravenous
injection (tail vein) of the E2-CDS at a dose of 1.0 mg/kg body weight or the
drug's vehicle, 2-hydroxypropyl-(}-cyclodextrin (HPCD). Animals (6 per
group) were then killed by decapitation 1, 7, and 14 days after the drug
injection and the trunk blood was collected in heparinized tubes. The blood
was centrifuged and the plasma separated and stored at -20C until hormone
analyses. Brains were removed, rinsed with ice-cold saline solution and
stripped of their pia matter and immediately stored at -80C until hormone
analysis. Plasma and brain samples were each processed and assayed by the
method described under the In Vitro Methodology section.
The 1.0 mg/kg dose of E2-CDS was chosen for investigation in this
study for the following reasons: 1) we anticipated that the tissue
concentrations of the E2-CDS metabolites, E2-Q+ and E2, would be in a
quantitatable range when using this dose and the application of RIA
procedure, over the time-course chosen for this study and 2) our previous
pharmacological observations indicated that this dose of the E2-CDS is capable
of causing chronic suppression of gonadotropins in castrated rats.


56
ml watenacetone (80:20; v:v). The columns were then allowed to air dry for 3
min before samples were eluted. With aqueous samples, the small amount
of residual water in the column was removed with either 50-100 ul of HPLC
grade hexane or by air-drying for >3 min. For sample elution, 2 aliquots of 500
ul methanol were applied sequentially to each column and the steroid was
eluted under vacuum pressure. The eluates were collected separately into
glass test tubes in the vacuum manifold and then dried under a stream of
nitrogen gas. Methanol was used to elute E2 from the columns since we
observed a column extraction efficiency of 92% with methanol, whereas two
other solvents tested were less efficient in eluting E2 (88% with acetone and
49% with acetonitrile).
Radioimmunoassay of E?
The dried residues of the E2 samples were reconstituted in 300 ul of the
assay buffer (kit Zero Calibrator; Lot # 10E2Z003; 100 ml) then, after vortexing
for 1 min, samples were equilibrated for 30 to 60 min at room temperature.
Duplicate 100 ul aliquots of each reconstituted E2 samples were assayed by
RIA.
Calculations
If the mean blank (tissue homogenates that did not contain the E2
spike) values for an assay were greater than the limits of sensitivity of the
standard curve, (i.e. if detectable E2 was present in the tissue), these values
were subtracted from all spiked samples. Also, values calculated from the
RIA run were adjusted for the volume of the aliquot taken for the RIA,
experimental losses during solvent extraction and chromatography
(determined by the addition of internal standard), and the weight of the tissue
sample used (for the in vivo experiment).


98
of both E2-Q+ and E2 in the anterior pituitary may be caused by the anatomical
relationship between the hypothalamus and the pituitary gland. Estradiol
released upon the hydrolysis of E2-Q+, or the E2-Q+ itself, which is locked into
brain, could be delivered directly to the pituitary by the capillary plexus of the
hypophyseal portal system. These capillaries in the median eminence lack
features of other brain capillaries and hence are not part of the BBB
(Traystman, 1983). Thus, the median eminence would not be expected to
prevent the efflux of E2-Q+ from brain, and transfer of E2-Q+ to anterior
pituitary can be expected.
Plasma also showed, after day 1, a residual but detectable E2
concentration throughout the time course of the 1.0-mg E2-CDS dose and
through the 21-day time course of the 0.1-mg dose. This prolonged and
residual E2 in plasma is likely to be the result of a continuous redistribution
of E2 liberated from E2-Q+ in the brain or other tissues down its concentration
gradient into the general circulation.
The oxidation of E2-CDS to E2-Q+ in uterine tissue did not appear to be
dose dependent, since a 100-fold increase in dose resulted in a 21-fold increase
in E2-Q+ concentration. This observation is perhaps, in part, an artifact of the
observed uterine hypertrophy in animals treated with the E2-CDS (Anderson
et al., 1988a). At day 1 after E2-CDS treatment, a 100-fold increase in E2-CDS
dose resulted in 54% increase in uterine weight (Chapter 7, Table 8), this
might reduce the values for E2-Q+ when normalized for tissue weight.
However, when the extent of oxidation was estimated per tissue, we observed
only a 32-fold increase over the 100-fold dose range examined. Additionally,
since blood flow to the uterus of OVX rats is low, the expected rapid oxidation
of E2-CDS to E2-Q+ would occur less likely in that tissue. Finally, we cannot
rule out the possibility that in the uterus of OVX rats, the enzymatic oxidation


100
Figure 10. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in whole brain of ovariectomized rats.
Animals received a single iv (tail vein) injection of the E2-CDS on
day 0 at a dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle
panels), or 0.01 mg/kg (lower panels). Animals were killed 1, 7,14,
21, or 28 days after the drug administration. Tissue samples of
known wet weight at a concentration of 1 mg/20 pi solvent were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Also, tissue homogenates from HPCD-treated rats were
analyzed for E2 hormone background. Represented are means
SEM for n = 7 rats per group per sampling time.


165
delivery is the major limiting factor in the prodrug design, specially those
with cytotoxicity or those with broad spectrum of peripheral sites of action
such as steroids.
A more recent, pharmacologic-based strategy which has shown great
potential for drug delivery is the liposome preparation (Weiner et al, 1989).
An assortment of drugs, including peptide and protein compounds, may be
incorporated in the liposomes, which can then be administered by different
routes. The physicochemical properties of the liposomes allows the
encapsulation of a drug molecule either in the aqueous space or intercalation
into the lipid bilayer matrix, depending on the properties of drugs.
Liposomes are, however, taken up by cells lining the reticuloendothelial
system, and do not appear to be useful for drug delivery through the BBB.
The paradoxical reason for this may be the larger size of these vesicles which
prevent them from crossing the BBB via lipid-mediated transport
mechanisms even though liposomes are highly lipid soluble (Pardridge,
1988a).
The most highly developed and promising strategy for improving drug
delivery through the BBB is the coupling of water- or lipid-soluble drugs to
the redox-based dihydropyridine nucleus (Bodor, 1987; Bodor & Brewster,
1983; Bodor et al., 1981). This strategy for the CNS drug delivery offers several
advantages. First, the application of the carrier, dihydropyridine, to a drug
increases the lipid solubility of the drug, because of highly lipid-soluble
nature of the carrier. Second, the carrier exhibits an intermediate enzymatic
oxidation to a quaternary pyridinium ion, which encourages preferential
brain deposition and retention by "locking" the charged, oxidative metabolite
behind the BBB; enzymatic hydrolysis of the charged-pyridinium drug
complex in a subsequent step provides sustained release of the parent drug in


106
Figure 13. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in liver of ovariectomized rats. Animals
received a single iv (tail vein) injection of the E2-CDS on day 0 at a
dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle panels), or
0.01 mg/kg (lower panels). Animals were killed 1, 7,14, 21, or 28
days after the drug administration. Tissue samples of known wet
weight at a concentration of 1 mg/20 pi solvent were processed
and assayed for E2-Q+ and E2 by the method described in Chapter 4.
Also, tissue homogenates from HPCD-treated rats were analyzed
for E2 hormone background. Represented are means SEM for n
= 7 rats per group per sampling time. Indicates below the
sensitivity limit of the assay.


27
astrocytic endfeet seem to be involved in the regulation of nutrients flux and
uptake of substances from the circulation (Broadwell & Salcmen, 1981).
In addition to the aforementioned structural features contributing to
the BBB, the presence of various enzymes in and around the endothelial cells
of cerebral capillaries may play a vital role in limiting and perhaps protecting
the brain from a variety of blood-borne substances (Hardebo & Owman, 1980).
Thus, the presence of catechol-o-methyltransferase, monoamine oxidase,
aromatic amino acid decarboxylase and gamma-aminobutyric acid
transaminase in the vicinity of cerebral vasculature may restrict the entry of
various blood-borne chemicals, i.e., neurotransmitters/neuromodulators or
therapeutic agents into the CNS. Such a protective mechanism against
circulating neuroactive substances is essential since optimal CNS function
requires a delicate balance between neurotransmitter release, metabolism and
uptake in the vicinity of neurons. Finally, the enzymatic component of the
BBB may also play a role in excluding some of the lipophilic compounds
from the CNS which otherwise might passively diffuse through the barrier.
Regarding the nutrient requirements of the brain, there are numerous
specialized carrier transport systems which are localized within the BBB for
uptake of nutrients from the circulation (Fenstermacher, 1985; Pardridge,
1981,1983,1986,1987). These include, specific carrier-mediated transport
systems for numerous classes of nutrients (Pardridge, 1981, 1983), receptor-
mediated transport mechanisms for plasma proteins and peptides (Pardridge,
1986), and plasma protein-mediated transport of protein-bound substances
(Pardridge, 1988a, 1988b). These transport systems are localized on the
luminal (or blood side) as well as on the antiluminal border (or brain side) of
the BBB (Pardridge, 1988a, 1988b), and are characterized by saturability and
specificity. These transport mechanisms of the BBB, therefore, provide


177
Bodor.N. Novel approaches to prodrug design. Drug of the Future 6:165-182
(1981).
Bodor N. Prodrugs versus soft drugs. In: Design of Prodrugs. H. Bundgaard
(ed) 333-354, Elsevier Science Publishers BV, Amsterdam (1985).
Bodor N. Redox drug delivery for targeting drug to the brain. Ann. NY Acad.
Sd. 507:289-306 (1987).
Bodor N. and M.E. Brewster. Problems of delivery of drugs to the brain.
Pharmacol. Ther. 19:337-386 (1983).
Bodor N., A. El-Koussi, M. Kano, T. Nakamura, and M. Khalifa. Improved
delivery through biological membranes 26. Design, synthesis and
pharmacological activity of a novel chemical delivery system for (3-
adrenergic blocking agents. T. Med. Chem. 31:100-106(1988).
Bodor N. and H.H. Farag. Improved delivery through biological membranes
XI. A redox chemical drug-delivery system and its use for brain-specific
delivery of phenethylamine. T. Med. Chem. 26:313-318 (1983).
Bodor N. and H.H. Farag. Improved delivery through biological membranes
XIV. Brain-specific, sustained delivery of testosterone using a redox
chemical delivery system. T Pharm. Sci. 73:385-389 (1984).
Bodor N., H.H. Farag, and M.E. Brewster. Site specific, sustained release of
drugs to the brain. Science 214:1370-1372 (1981).
Bodor N. and J. Kaminski. Prodrug and site-specific chemical delivery
systems. Ann. Rpts. Med. Chem. 22:303-313 (1987).
Bodor N., J. McCornack, and M.E. Brewster. Improved delivery through
bioloical membranes. XXII. Synthesis and distribution of brain-
selective estrogen delivery systems. Int. T. Pharm. 35:47-59 (1987).
Bodor N. and J.W. Simpkins. Redox delivery system for brain-specific,
sustained release of dopamine. Science 221:65-67 (1983).
Bolt H.M. Metabolism of estrogensnatural and synthetic. Pharmacol. Ther.
4:155-181 (1979).
Bonney R.C., C. A. Sparks, R.W. Cheng, M.J. Reed, and V.H. James. The
measurement of steroid hormones in endometrial tissue: A
comparison between two methods of extraction. T. Steroid Biochem.
21:479-481 (1984).


133
undetectable LH or FSH were assigned the respective assay sensitivity (0.25 for
LH and 2.5 ng/ml for FSH). All the samples for each hormone were assayed
in a single run.
Plasma E2 and T concentrations were measured in duplicate by using
Coat-A-Count Estradiol kits and Coat-A-Count Testosterone kits, respectively.
The RIA sensitivity of the T assay was 0.2 ng/ml. The cross-reactivity of the T
antibody to the DHT and E2 has been reported to be 3.3 and 0.02%, respectively
(technical information from Diagnostic Products).
Results
Effect of CAST or E?-CDS on Plasma T Levels
Figure 19 shows the comparative plasma T concentrations of untreated
control, post CAST and animals treated with E2-CDS. CAST resulted in
greater than 99% suppression of plasma T levels which remained suppressed
at about the detection limit of the T assay for the time-course of study.
Likewise, plasma T levels in E2-CDS-treated animals were significantly
suppressed by more than 96, 92, or 95% with 1, 2, or 3 injections, respectively,
at 7 days after the last treatment. However, by 14 days after the last treatment
with E2-CDS, a 76, 82, or 91% reduction in plasma T was observed with 1, 2, or
3 injections, respectively. Even though plasma T levels began to gradually
increase in a manner related to the number of E2-CDS injections, T levels
remained significantly suppressed compared with the control values.


195
Rahimy M.H., J.W. Simpkins, and N. Bodor. Dose and time-course
evaluation of a redox-based estradiol-chemical delivery system for the
brain. I. Tissue distribution. Pharm. Res. 7:1061-1067 (1990b).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Dose and time-course
evaluation of a redox-based estradiol-chemical delivery system for the
brain. II. Pharmacodynamic responses. Pharm. Res. 7 (11) (in press,
1990c).
Rao B.R., A.A. Geldof, C.L. van der Wilt, and H.J. de Voogt. Efficacy and
advantages in the use of low doses of androgen and estrogen
combination in the treatment of prostate cancer. The Prostate 13:69-78
(1988).
Rapoport S.I. The Blood-brain barrier in physiology and medicine. Raven
Press, New York (1976).
Rapoport S.I., M. Ohata, E.D. London. Cerebral blood flow and glucose
utilization following opening of the blood-brain barrier and during
maturation of the rat brain. Fed.Proc. Fed. Am. Soc. Expl. Biol. 40:2322-
2325 (1981).
Rash J.M., I. Jerkunica, and D.S. Sgoutas. Lipid interference in steroid
radioimmunoassay. Clin. Chem. 26:84-88 (1980).
Rauramo L., K. Lagerapitz, P. Engblom, and R. Punnonen. The effect of
castration and peripheral estrogen therapy on some psychological
function. Front. Horm. Res. 8:133-151 (1975).
Reese T.S. and M.J. Karnovsky. Fine structural localization of a blood-brain
barrier to exogenous peroxidase. Cell Biol. 34:207-217 (1967).
Richards J.S. and L. Hedin. Molecular aspects of hormone action on ovarian
follicular development, ovulation, and luteinization. Ann. Rev.
Physiol. 50:441-463 (1988).
Roselli C.E. and J.A. Resko. Regulation of hypothalamic luteinizing
hormone-releasing hormone levels by testosterone and estradiol in
male rhesus monkeys. Brain Res. 509:343-346 (1990).
Rosie R., E. Thomson, and G. Fink. Oestrogen positive feedback stimulates
the synthesis of LHRH mRNA in neurons of the rostral diencephalon
of the rat. T. Endocrinol. 124:285-289 (1990).


158
frequency of the postmenopausal symptoms vary among women, and
symptoms do not occur at all in about 25% of postmenopausal women
(Casper & Yen, 1985). Also, it is important to note that not all
postmenopausal women on hormone replacement therapy respond
positively to the steroid medications (Paterson, 1982), even though it is
generally believed that the flushing response is due to steroid-hormone
withdrawal phenomenon (Casper & Yen, 1985; Lauritzen, 1973; Simpkins &
Katovich, 1984; Yen, 1977).
It is also important to point out that the length of exposure to E2 via E2-
CDS may be as important as the dose administered to the morphine-
dependent rats. Perhaps the variability or heterogeneity in response is due to
the fact that all animals may eventually respond to E2-CDS treatment, but it
may take variable durations of sustained E2 exposure via E2-CDS treatment in
order to observe an attenuation in TST response.
In conclusion, these results support the view that the E2-CDS may be
potentially useful in the treatment of brain E2 deficiencies (i.e., vasomotor hot
flushes). Further experimental as well as clinical investigations pertaining to
the therapeutic efficacy of the E2-CDS in this regard is warranted. In
comparison to the currently used estrogenic products, the E2-CDS should
achieve the sustained stimulation of brain E2 receptors at lower doses or with
less frequent dosing.


Table 10: Effects of the E2-CDS (0.5 mg/kg bw) or CAST on Plasma Hormone Concentrations at 14
Days after the Last Treatment in Male Rats.
Treatment
Group
Inject.
No.
12
(pg/ml)
E?-Q+
(pg/ml)
LH
(ng/ml)
FSH
(ng/ml)
PRL
(ng/ml)
Intact
-
12.3 2.70
ND
0.50 0.07
10.34 1.88
28.15 4.91
CAST
-
7.30 1.87
ND
8.15 0.55a
26.57 1.34a
4.84 0.82
E2-CDS
1
20.07 4.11
190.2 22.9
0.27 0.01b
10.53 0.91b
38.34 9.96
e2-cds
2
13.73 4.31
120.9 18.8
0.27 0.02b
12.00 1.86b
15.93 3.10
e2-cds
3
30.01 4.17
131.5 20.4
0.31 0.03b
11.83 1.30b
12.47 3.19
ND Not determined
a Different from Intact
b Different from CAST
147


40
central neurotransmitters/neuromodulators levels involved in mood, or
enzymes necessary for their synthesis, or their receptor/effector systems,
estrogen therapy may very well benefit the postpartum depressed patients
(Luine et al., 1975; Maggi & Perez, 1985). Thus, there is reason to suggest the
notion that estrogen therapy with the E2-CDS may be useful for the treatment
of postpartum depression.
Various biochemical studies in the rat, examining the effect of
estrogens on biogenic amines and their enzymes, strongly support the
behavioral and symptomatic improvements which are observed in
postmenopausal patients. Estrogens are reported to inhibit the re-uptake of
norepinephrine in the rat brain (Luine et al., 1975) and decrease the activity of
monoamine oxidase (Holzbauer & Youdim, 1973; Luine et al., 1975).
Additionally, the activity of monoamine oxidase (MAO) in plasma of
postmenopausal patients were also reduced in response to estrogen treatment
(Klaiber et al., 1976). In fact, MAO activity is lower in women during the
preovulatory phase of the menstrual cycle when E2 levels are highest. In
contrast, MAO activity increases during the luteal phase when progesterone
levels are highest (Klaiber et al., 1976).
Collectively, the evidence mentioned above support the notion that
estrogen therapy may have a rational scientific basis for treatment of
depression associated with the decline in endogenous estrogen production.
Estrogen treatment of these conditions may influence brain function via
effects on a number of neurotransmitter systems involved in mood and other
emotional behaviors.


186
Jensen E.V. and E.R. DeSombre. Estrogen receptor interaction. Science
182:126-134 (1973).
Jensen E.V. and H.I. Jacobsen. Basic guides to the mechanism of estrogen
action. Recent Prog. Horm. Res. 18:387-414 (1962).
Jonsson G., A.M. Olsson, W. Luttrop, Z. Cekan, K. Purvis, and E. Diczfalusy.
Treatment of prostatic carcinoma with various types of estrogen
derivatives. Vitam. Horm. 33:351-376. (1975).
Judd H.L. Pathophysiology of menopausal hot flushes. In:
Neuroendocrinolgy of Aging, J. Meites (ed) 173-202, Plenum Press, New
York (1983).
Kalla N.R., B.C. Nisula, R. Menard, and D. Lynn Loriaux. The effect of
estradiol on testicular testosterone biosynthesis. Endocrinology 106:35-
39 (1980).
Kalra P.S. and S.P. Kalra. Temporal changes in the hypothalamic and serum
luteinizing hormone-releasing hormone (LHRH) levels and
circulating ovarian steroids during the rat estrous cycle. Acta
Endocrinol. 85:449-455 (1977).
Kalra S.P. and P.S. Kalra. Modulation of hypothalamic luteinizing hormone
releasing hormone levels by intracranial and subcutaneous implants of
gonadal steroids in castrated rats: Effects of androgen and estrogen
antagonists. Endocrinology 106:390-397 (1980).
Kalra S.P. and P.S. Kalra. Neural regulation of luteinizing hormone secretion
in the rat. Endocr. Rev. 4:311-351 (1983).
Kalra S.P. and P.S. Kalra. Do testosterone and estradiol-17p enforce inhibition
or stimulation of luteinizing hormone-releasing hormone secretion?
Biol. Reprod. 41:559-570 (1989).
Kaplan N.M. Cardiovascular complications of oral contraceptives. Ann. Rev.
Med. 29:31-40 (1978).
Katovich M.J. and J. O'Meara. Effect of chronic estrogen on the skin
temperature response to naloxone in morphine-dependent rats. Can. T.
Physiol. Pharmacol. 65:563-567 (1986).
Kaufman H., C. Vadasz, and A. Lajtha. Effects of estradiol and
dexamethasone on choline acetyltransferase activity in various rat
brain regions. Brain Res. 453:389-392 (1988).


CHAPTER 7
EVALUATION OF THE PHARMACODYNAMIC EFFECTS OF E2-CDS IN
OVARIECTOMIZED FEMALE RATS
Introduction
Estrogen hormones have been shown to influence a myriad of CNS
processes including reproductive parameters, i.e., neuroendocrine
modulation of reproductive cycle (Kalra & Kalra, 1989; Maggi & Perez, 1985;
Plant, 1986) and stimulation of sexual behaviors (Beyer et al., 1976;
Christensen & Clemens, 1974) as well as non-reproductive parameters such as
regulation of neurotransmissions involved in sensorimotor functions,
mood, and learning tasks (Fillit et al., 1986; McEwen et al., 1984; Smith, 1989).
These diverse actions of estrogens on the CNS functions are of significant
therapeutic interest after menopause or ovariectomy when endogenous
estrogens decline. In these and certain other cases, since estrogen medications
are primarily used for their central actions, the preferential brain estrogen
delivery is not only beneficial but may produce safer and more potent natural
therapeutic agent.
The evaluation of the tissue distribution patterns of both E2-Q+ and E2
in intact male (Chapter 5; Rahimy et al., 1990a) as well as in OVX female rats
(Chapter 6; Rahimy et al., 1990b) substantiated the major aspect of the
proposed mechanism of this redox-based estrogen delivery system for the
brain. That is, the E2-CDS consistently demonstrated its predictive
pharmacokinetic behaviors including the preferential retention of E2-Q+ and
thus, E2 in the CNS tissue with ti/2 = 8-9 days, while simultaneously
115


PERCENT BOUND
68
PG/ML
Figure 2. Inhibition of 125I-E2 binding to an E2 antibody caused by E2 (left,
open circle) or E2-Q+ (right, closed circle). Possible cross-reaction
with the E2-Q+ was examined by adding E2-Q+ standards to the
RIA of E2 at concentrations of 2, 5, 15, 50,180, and 360 ng/ml.
These doses of E2-Q+ were 100-fold greater than the corresponding
E2 standards used to generate standard curves in the RIA of E2.
Data are expressed on logit-log graph (% bound on ordinate is
based on logit = log (percent bound/100 percent bound)). Cross
reactivity data indicated that E2 antibody used was specific for E2
and cross-reacts with E2-Q+ to <0.3% at concentration of 15 ng/ml
and greater.




130
high-dose estrogen therapy, the two conventional treatment paradigms, still
remain as the treatment of choice for the endocrine-dependent management
of prostatic cancer (van Steenbrugge et al., 1988). Both treatments are reported
to be equally effective (i) in suppressing the circulating T levels (Carlstrom et
al., 1989) and perhaps regression of the prostatic hyperplasia; and (ii) in
controlling the symptoms of advanced prostatic cancer in 70-80% of patients
with an average improved rate of survival of 5 years (Klein, 1979).
However, none of the two treatments is optimal or without major
complication. For instance, high-dose estrogen therapy has been shown to
cause severe cardiovascular complications (Henriksson & Edhag, 1986) and
alterations in liver metabolism (von Schoultz et al., 1989). Estrogens,
specially synthetic compounds, exert profound effects on liver-derived
plasma proteins, coagulation factors, lipoproteins, and triglycerides when
administered orally. However, most of the cardiovascular complications are
the result of arterial ischemic events, and the majority of such events are the
results of acute coronary arterial disease (Henrikson & Edhag, 1986). Recent
studies have suggested that these dose-dependent liver-associated side effects
and thus, cardiovascular complications may be reduced or even abolished
when lowering the dose, provided that adequate T suppression is achieved
(Jonsson et al., 1975; von Schoultz et al., 1989). Furthermore, numerous other
reports also indicated that the marked interference with hepatic metabolism
is associated with the kind of estrogen used (Ottosson, 1984; Ottosson et al.,
1986). For example, the synthetic alkylated estrogen, ethinyl E2, is a potent
estrogen used in the treatment of prostatic cancer; however, it exerts
profound effects on liver metabolism and thus, the use of a natural estrogen
may be advantageous (Ottosson et al., 1986).


179
Butenandt A. Uber "PROGYNON" ein crystallisiertes, weibliches
sexualhormone. Naturwissenschaften 17:879 (1929).
Cabot A.T. The question of castration for enlarged prostate. Ann. Surg.
24:265-309 (1896).
Campbell S. and M. Whitehead. Oestrogen therapy and the menopausal
syndrome. Clin. Obstet. Gynaecol. 4:31-47 (1977)
Canick J.A., D.E. Vaccaro, E.M. Livingston, S.E. Leeman, K.J. Ryan, and T.O.
Fox. Localization of aromatase and 5 a-reductase to neuronal and non
neuronal cells in the fetal rat hypothalamus. Brain Res. 372:277-282
(1986).
Carlstrom K., L. Collste, A. Erikson, P. Henriksson, A. Pousette, R. Stege, and
B. von Schoultz. A comparison of androgen status in patients with
prostatic cancer treated with oral and/or parenteral estrogens or by
orchidectomy. The Prostate 14:177-182 (1989).
Casper, R.E. and S.S.C. Yen. Neuroendocrinology of menopausal flushes: An
hypothesis of flush mechanism. Clin. Endocrinol. 22:257-267 (1985).
Casper R.F., S.S.C. Yen, and M.M. Wilkes. Menopausal flushes: A
neuroendocrine link with pulstil luteinizing hormone secretion.
Science 205:823-925 (1979).
Chen C.L. and J. Meites. Effects of estrogens and progesterone on serum and
pituitary prolactin levels in ovariectomized rats. Endocrinology
86:503-508 (1970).
Christensen L.W. and L.G. Clemens. Intrahypothalamic implants of
testosterone or estradiol and resumption of masculine sexual behavior
in long-term castrated male rats. Endocrinology 95:984-990 (1974).
Clayden J.R., J.W. Bell,and P. Pollard. Menopausal flushing: Double-blind
trial of a non-hormonal medication. Br. Med. T. 1:409-412 (1974).
Cooper K.J., C.P. Fawcett, and S.M. McCann. Inhibitory and facilitatory effects
of estradiol 17|3 on pituitary responsiveness to a luteinizing hormone-
follicle stimulating hormone releasing factor (LH-RF/FSH-RF)
preparation in the ovariectomized rat. Pro. Soc. Expt. Biol. Med.
145:1422-1426 (1974).
Corner G.W. and W.M. Allen. Physiology of the corpus luteum. II.
Production of a special uterine reaction (progestational proliferation) by
extracts of the corpus luteum. Am. T. Physiol. 88:326-346 (1929).


Treatment Treatment
142
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
7 Days After Final Treatment
0 50 100 150
14 Days After Final Treatment
Control
0
Castrated
0
E2-CDS
(xl)
E2-CDS
(x2)
E2-CDS
(x3)
0 50 100 150
Ventral Prostate mg /100 g bw
i
200
200
Figure 20. Effects of the E2-CDS (0.5 mg/kg) or CAST on ventral prostate
weight either 7 days (upper panel) or 14 days (lower panel) after
the final treatment. Animals received weekly iv (tail vein)
injection of either the drug's vehicle (HPCD), or E2-CDS at a dose
of 0.5 mg/kg (1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3
injections; x3), or were castrated (CAST). Represented are means
SEM for n = 7-8 animals per group per sampling time. The symbol
(a) denotes differences from control (HPCD)-treated animals and
the symbol (b) indicates differences from CAST within the panel
as analyzed by ANOVA and Scheffe statistics.


41
Cognitive Impairment of Menopausal Alzheimer's Type
Since the 1950's, several lines of evidence have accrued to suggest that
estrogen hormones may influence certain cognitive functions of the female.
First, the decreasing production of endogenous estrogens/progestins after
menopause or ovariectomy has been shown to cause changes in cognitive
functions, especially in memory, that are prominent among the somatic and
behavioral symptoms of menopause (Kopera, 1973; Malleson, 1953; Lauritzen
& van Keep, 1978). Second, estrogen treatment of menopausal women with
senile dementia of Alzheimer's type is shown to improve both symptomatic
treatment and prevents or slows the progression of dementia in these
patients (Campbell & Whitehead, 1977; Fedor-Freybergh, 1977; Fillit et al.,
1986; Furuhjelm & Fedor-Freybergh, 1976; Hackman & Galbraith, 1976; Honjo
et al., 1989; Michael et al., 1970; Sherwin, 1988; Vanhulle & Demol, 1976).
Third, numerous biochemical studies have demonstrated that estrogen
hormones modulate/enhance cholinergic transmission or activity in regions
of the brain that are important for cognitive functions (Eleftheriou & Dobson,
1972; Iramain et al., 1980; Luine et al., 1975,1980,1983,1985; O'Malley et al.,
1987). Fourth, estrogen receptors have been identified in nuclei of the basal
forebrain structures in the rat (Luine et al., 1975; Morrel et al., 1975; Pfaff &
Keiner, 1973), the major loci of cell bodies of cholinergic neurons which
innervate the cerebral cortex, limbic system, hippocampus and
hypothalamus. These brain regions are believed to be involved in the
pathology of Alzheimer's Disease (Coyle et al., 1983). Finally, perhaps the
most interesting observation, is that the female:male sex ratio in the
prevalence of Alzheimers Disease is 2:1 (Sulkava et al., 1985).


180
Cortes-Gollegos V. and A.J. Gallegos. Estrogen peripheral levels vs. estrogen
tissue concentration in the human female reproductive tract. T. Steroid
Biochem. 6:15-20(1975).
Coyle J.T., D.L. Price, and M.R. DeLong. Alzheimer's disease: A disorder of
cortical cholinergic innervation. Science 219:1184-1190 (1983).
Cruickshank J.M., G. Neil-Dwyer, M.M. Cameron, and J. McAinsh. (3-
Adrenoreceptor-blocking agents and the blood-brain barrier. Clin. Sci.
59:453s-455s (1979).
Czaja J.A. Ovarian influence on primate food intake: Assessment of
progesterone actions. Physiol. Behav. 21:923-928 (1978).
Czekala N.M., J.K. Hodges, and B.L. Lasley. Pregnancy monitoring in diverse
primate species by estrogen and bioactive LH determinations in small
volumes of urine. T. Med. Primatol. 10:1-15 (1981).
Daehlin L., A. Bergh, and J.E. Damber. Direct effects of oestradiol on growth
and morphology of the Dunning R3327H prostatic carcinoma. Urol.
Res. 15:169-172 (1987).
Daehlin L. and J.E. Damber. Blood flow in the Dunning R3327H rat prostatic
adenocarcinoma: Effect of oestradiol and testosterone. Urol. Res.
14:113-117(1986).
Dalkin A.C., D.J. Haisenleder, G.A. Ortolano, T.R. Ellis, and J.C. Marshall. The
frequency of gonadotropin-releasing-hormone stimulation
differentially regulates gonadotropin subunit messenger ribonucleic
acid expression. Endocrinology 125:917-924 (1989).
Davidson J.M. Feedback control of gonadotropin secretion. In: Frontiers in
Neuroendocrinologv. W.F. Ganong and L. Martini (eds) 343-388, Oxford
University Press, New York (1969).
Davidson J.M. Hormones and reproductive behavior. In: Hormones and
Behavior. S. Levine (ed.) 63-103, Academic Press, New York (1972).
Davson H. The blood brain barrier. T Physiol. Lond. 255:1-29 (1976).
DeLignieres B. and M. Vincens. Differential effects of exogenous oestradiol
and progesterone on mood in postmenopausal women: Individual
dose effect relationships. Maturitas 4:67-72 (1982).


176
Bain J.D., L.H. Kasman, A.B. Bercovitz, and B.L. Lasley. A comparison of
three methods of hydrolysis for estrogen conjugates. Steroids 43:603-
619 (1984).
Barraclough C.S. and P.M. Wise. The role of catecholamines in the regulation
of pituitary luteinizing hormone and follicle-stimulating hormone
secretion. Endocr. Rev. 3:91-119 (1982).
Barrett-Connor E., D.L. Wingard, and M.H. Criqui. Postmenopausal estrogen
use and heart disease risk factors in the 1980's. TAMA 261:2095-2100
(1989).
Beard J. The span of gestation and the cause of birth. Gustav Fischer Verlag,
Jena (1897).
Becker J.B. Direct effect of 170-estradiol on striatum: Sex differences in
dopamine release. Synapse 5:157164 (1990).
Becker J.B. and M.E. Beer. The influence of estrogen on nigrostriatal
dopamine activity: Behavioral and neurochemical evidence for both
pre- and postsynaptic components. Behav. Brain Res. 1:27-33 (1986).
Bergkvist L., I. Persson, H-O Adami, and C. Schairer. Risk factors for breast
and endometrial cancer in a cohort of women treated with menopausal
oestrogens. Int. T. Epidemiol. 17:732-737 (1988).
Berkowitz G.S., J.L. Kelsey, T.R. Holford, V.A. Livolsi, M.J. Merino, G.J. Beck,
S. Ort, T.Z. O'Connor, and C. White. Estrogen replacement therapy
and fibrocystic breast disease in postmenopausal women. Am. T.
Epidemiol. 121:238-245(1985).
Berry S.J., D.S. Coffey, P.C. Walsh, and L.L. Ewing. The development of
human benign prostatic hyperplasia with age. T. Urol. 132:474-479
(1984).
Beyer C, G. Morali, F. Naftolin, K. Larsson, and G. Perez-Palacios. Effects of
some antiestrogens and aromatase inhibitors on androgen induced
sexual behavior in castrated male rats. Horm. Behav. 7:353-362 (1976).
Blackard C.E. The Veterans' Administration Cooperative Urological
Research group studies of carcinoma of the prostate: A review. Cancer
Chemother. Rep. 59:225-227 (1975).
Blackard C.E., D.P. Byar, and W.P. Jordan. Orchiectomy for advanced prostatic
carcinoma: a reevaluation. Urology 1:553-560 (1973).


Treatment Treatment
141
Control 0
7 Davs After Final Treatment
Castrated 0
1 a
E2-CDS (xl)
a
E2-CDS (x2)
a
E2-CDS (x3)
a
14 Days After Final Treatment
Control
Castrated
E2-CDS
E2-CDS
E2-CDS
Plasma Testosterone (ng/ml)
Figure 19. Effects of the E2-CDS (0.5 mg/kg) or CAST on plasma testosterone
levels either 7 days (upper panel) or 14 days (lower panel) after the
final treatment. Animals received weekly iv (tail vein) injection
of either the drug's vehicle (HPCD), or E2-CDS (1 injection; xl), E2-
CDS (2 injections; x2), E2-CDS (3 injections; x3), or were castrated
(CAST). Represented are means SEM for n = 7-8 animals per
group per sampling time. The symbol (a) denotes differences from
control (HPCD)-treated animals as analyzed by ANOVA and
Scheffe statistics.


166
the brain. Finally, the carrier system simultaneously enhances the rate of
elimination of the drug, specially if lipoidal in nature, in an inactive form,
from peripheral tissues following the oxidation to a hydrophilic quaternary
form.
The present work evaluated this redox-based chemical-delivery
approach to brain estrogen delivery. To determine reliably the tissue
distribution and thus to document the effectiveness of E2-CDS, a specific and
sensitive method was essential (Chapter 4; Rahimy et al., 1989a). Such criteria
are extremely important, particularly for the E2-CDS, since the E2-CDS
metabolite, E2, is not only present in low concentration but also active at low
pg/g tissue. Furthermore, the intermediate metabolite, E2-Q+, is present in
concentrations much higher than E2. Thus, to accurately quantitate E2 levels
in tissue samples, the assay method must be capable of distinguishing low
levels of E2 in the presence of high concentrations of E2-Q+. These problems
were resolved by employing and optimizing the reproducibility of an RIA
procedure for E2 determination in all tissues and fluid (Chapter 4). The RIA
provides the needed sensitive end point (0.8 to 1.2 pg/assay tube) as well as
the required specificity (cross-reactivity of <0.3% for E2-Q+ at concentration of
15 ng/ml and higher).
Extensive and detailed evaluation of the tissue distribution of E2-CDS
in both intact male (Chapter 5; Rahimy et al., 1988,1990a) and OVX female
rats (Chapter 6; Rahimy et al., 1990b) support the concept of brain-enhanced
delivery and sustained release of E2 using the redox-based carrier system
(Bodor et al., 1987). Furthermore, the distribution patterns of both E2-Q+ and
E2 in male (Chapter 5; Rahimy et al., 1990a) and in female rats (Chapter 6;
Rahimy et al., 1990b) confirmed the major aspect of the proposed mechanism
of the E2-CDS drug delivery. Interestingly, the extent of deposition and the


12
and transported to the liver, a process called enterohepatic circulation (Bolt,
1979). A metabolite of E2, which comprises at least 20% of the total amount
secreted in humans, is the 2-hydroxyl derivative. These metabolites, referred
to as catechol estrogens, are shown to have biological activity. The biological
activity of the catechol estrogen appears to involve an interaction with
catecholamine synthesis, receptors or effectors (Weisz & Crowley, 1986). The
conversion to catechol estrogen can occur in a number of tissues, including
the CNS (Weisz & Crowley, 1986).
Mechanism of action
Based on the gross anatomical, histological, and biochemical evidence,
E2 is shown to have growth-promoting activities on cells of the target organs
such as the anterior pituitary, uterus, vagina, Graafian follicles of the ovary
and the mammary gland by increasing protein synthesis and mitotic activity.
As early as 1953, Szego and Roberts, seeking an understanding of the
mechanism of E2 action, demonstrated accumulation of ribonucleic acid
(RNA) and protein in estrogen-stimulated uterine tissues. Mueller et al.
(1958) showed that most of the E2 effects on RNA and protein synthesis can be
blocked by a translation inhibitor (puromycin) and a transcription inhibitor
(actinomycin D). These observations led to the proposal that estrogen
hormones and steroids in general stimulate or activate the production of
nucleic acid templates (mRNAs) and, hence, gene expression (Mueller et al.,
1958). Soon after the technological advances of the 1960's and, thus, the
availability of tritium-labeled estradiol (3H-E2), Jensen and Jacobsen (1962)
discovered that the estrogen target tissues (uterus and vagina) selectively
concentrated the labeled E2. These investigators were also the first to
demonstrate the binding of E2 to a specific cytosolic receptor protein (Jensen &


164
gain access to the CNS. As a result, many potentially useful therapeutic
agents, i.e. water soluble drugs are excluded from entering the brain. To
overcome this barrier and thus, to enhance CNS drug concentration, various
strategies for drug flux through the BBB have been developed. These include:
(a) physical approaches, i.e. intraventricular/intrathecal infusion and
implantable pump (Hammond, 1988); and (b) chemical approaches, i.e.
prodrugs (Bodor, 1981, 1985,1987; Sinkula & Yalkowsky, 1975; Stella, 1975),
liposomes (Weiner et al., 1989), and the use of membrane transport systems to
deliver nontransportable peptides through the BBB (Pardridge, 1986).
The intraventricular administration of drugs, in addition to being an
invasive technique, is inefficient in cases where the drugs of choice are polar
and highly water soluble. For instance, the intraventricular administration of
polar drugs results in uneven or incomplete distribution in the brain since
these agents are solubilized primarily in the aqueous compartment (CSF).
Furthermore, this approach tends to bath the surface of the brain, perhaps
because the efflux of the drug out of the ventricles and into the superior
sagittal sinus is much faster than diffusion into the brain parenchyma
(Pardridge, 1988a).
A general, more practical approach to increase brain concentrations of
water soluble drugs and thus their therapeutic efficacy has been the design of
prodrug formulation (Bodor, 1981, 1985; Sinkula & Yalkowsky, 1975; Stella,
1975). The purpose of prodrug modification is to increase the concentration
of the active drug at or near its site of action, thereby increasing its efficacy.
However, by increasing the lipophilicity of a drug nonspecifically via the
prodrug approach, it may not only enhance its diffusion through the BBB ,
but also enables the uptake of the compound into all other tissues and thus,
exposure to a greater drug burden. This method of nonselectivity of drug


129
several lines of evidence that strongly support the possibility that prostatic
growth/hyperplasia is dependent upon the endocrine activity of the testis: (1)
prostatic maturation as well as prostatic hyperplasia does not occur in men
who are castrated prior to puberty (Moore, 1944); (2) surgical castration (CAST)
produces regression as well as beneficial effects toward the prostate
hyperplasia of men (Cabot, 1896; Huggins & Hodges, 1941; White, 1895); (3)
numerous studies have reported regression of prostatic hyperplasia with
antiandrogens or estrogens treatment (Brendler, 1988; Foote & Crawford,
1988); and (4) recent biochemical studies have reported an association between
prostatic hyperplasia with an abnormal accumulation of dihydro testosterone
(DHT), a potent testosterone (T) metabolite, in the human prostate (Isaacs et
al., 1983). Whether the increase in the concentration of prostatic DHT is
directly responsible for or is the result of the disease has not been completely
resolved. However, further support for the DHT involvement comes from
males with hereditary deficiency of the 5 oc-reductase enzyme, an enzyme that
converts T to DHT in androgen-target tissues (Imperato-McGinley et al., 1980,
1984). The lack of this enzyme results in very low levels of DHT as well as no
palpable prostatic tissue, indicating that the prostate had not developed
during embryogenesis.
Regarding the therapeutic aspects of the prostatic adenocarcinoma
and/or hyperplasia, currently a variety of surgical and therapeutic means of
inhibiting androgen production or blocking androgen action are being used
(Labrie et al., 1983; Santen & English, 1989). These include surgical CAST,
high-dose estrogen therapy, GnRH analogues, antiandrogens, and
combination of low dose estrogens plus low dose antiandrogens. Each of
these treatment regimens provides temporary benefit to patients, however,
relapse usually occurs within a period of 1 to 2 years. Nevertheless, CAST or


Treatment Treatment
144
7 Days After Final Treatment
Control
0
Castrated
0
E2-CDS
(xl)
E2-CDS
(x2)
E2-CDS
(x3)
Control
0
t 1
800 1000
14 Days After Final Treatment
400
Testes mg/100 g bw
1000
Figure 22. Effects of the E2-CDS (0.5 mg/kg) or CAST on testis weight either 7
days (upper panel) or 14 days (lower panel) after the final
treatment. Animals received weekly iv (tail vein) injection of
either the drug's vehicle (HPCD), or E2-CDS at a dose of 0.5 mg/kg
(1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3 injections; x3),
or were castrated (CAST). Represented are means SEM for n = 7-
8 animals per group per sampling time.


CHAPTER 6
DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN FEMALE RATS
Introduction
Brain-enhanced delivery to and the sustained release of E2 in the brain
may be potentially useful for the effective treatments of vasomotor hot
flushes, prostatic adenocarcinoma, and fertility regulation. The natural
estrogen, 17 (3-E2, administered as either a valerate, benzoate, or dienanthate
is effective in alleviating hot flushes (Campbell & Whitehead, 1977; Doring,
1976; Dusterberg & Nishino, 1982; Klopper, 1976; Lauritzen, 1973). Likewise,
the synthetic alkylating estrogen, ethinyl E2, is a potent and an effective
contraceptive agent (Burkman, 1988). However, these estrogen compounds
act upon all steroid-responsive tissues which limits their therapeutic efficacy.
Moreover, since these steroids equilibrate among all body tissues, only a
fraction of the administered dose accumulates at or near the site of action in
the CNS. These properties of the estrogen then force either frequent dosing,
or the administration of a depot form of the steroid, to maintain
therapeutically effective levels in the brain. These treatment strategies lead to
sustained increases in peripheral estrogen levels, and thus augment the risk
of peripheral toxicities. The redox-based E2-CDS may be potentially useful in
the effective treatments of brain diseases by providing sustained and
sufficient E2 to the brain while avoiding peripheral toxicities.
The present studies were undertaken to determine if the E2-CDS
behaves as predicted on the basis of the physicochemical properties designed
90


8
demonstrated that ovarian transplants prevented uterine atrophy and loss of
sexual function in castrated animals, established the hormonal nature of
ovarian control of the female reproductive system. Further supporting
evidence was provided by Fraenkel (1903), who showed that destruction of
the corpora ltea in pregnant rabbits causes abortion. In 1923, Allen and
Doisy developed a simple, quantitative bioassay for ovarian extracts based
upon changes produced in the vaginal smear of the rat. Two years later,
Loewe (1925) reported on a female sex hormone in the blood of various
species. Shortly thereafter, Loewe and Lange (1926) discovered a female sex
hormone in the urine of menstruating women with the observation that the
concentration of the hormone in the urine varied with the phases of the
menstrual cycle. These observations set the stage for chemists, who soon
isolated independently an active estrogen substance from urine in crystalline
form (Butenandt, 1929; Doisy et al., 1929,1930). In 1935, Doisy et al. (quoted by
Tepperman, 1981) characterized the chemical structure of estradiol-17|3.
However, it was the contributions of Corner and Allen (1929) that firmly
established the endocrine function of the corpus luteum. They clearly
demonstrated that the abortion following extirpation of the corpora ltea in
pregnant rabbits can be prevented by the injection of luteal extracts.
Endocrinological and Biochemical Considerations
Biosynthesis
Estrogens are primarily produced by the follicles and corpus luteum of
the ovary and by the placenta during the second and third trimesters of
pregnancy. Ovaries secrete estradiol (E2) and estrone, whereas the placenta
produces these and estriol (Ross, 1985; Schwartz, 1981). All these hormones


43
involvement of other neurotransmitter systems and estrogens in memory.
Thus, if estrogens serve to maintain or enhance the activity of cholinergic
neurons or serve as a trophic substance which directly or indirectly acts on
cholinergic neurons, the idea of enhanced brain exposure to E2 may be an
improvement in cognition. Further, since any therapy which is aimed at
treating Alzheimer's type dementia must be chronic in its application to the
patient, the sustained release of E2 from the E2-CDS is an additional useful
benefit. As such, a careful evaluation of the E2-CDS for the mechanism by
which it improves cholinergic function and for its potential application to
Alzheimer's Disease patients is clearly warranted.


Tissue Cone, (ng/g) Tissue Cone, (ng/g)
87
Figure 8. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on kidney,
heart, lung, and anterior pituitary concentrations of E2-Q+ (upper
panel) and E2 (lower panel) in intact male rats. Intact male rats
were injected with a single iv dose of 1.0 mg E2-CDS/kg bw and
killed 1, 7, or 14 days after treatment. Each point represents the
group mean SEM (n = 6-7 rats for each time point).


5
A remarkable example is the design of an estradiol-chemical delivery
system (E2-CDS) for the enhanced and sustained release of E2 in the brain
(Bodor et al., 1987). The E2-CDS exploits the unique architecture of the BBB,
which normally excludes a variety of pharmacological agents from the CNS
due to their physicochemical properties (Bodor & Brewster, 1983). The E2-
CDS is a redox-based chemical-delivery system and the mechanism of its drug
delivery is based upon an interconvertible dihydropyridine <=> pyridinium
ion redox carrier (Bodor et al., 1987). After systemic administration of the E2-
CDS, it distributes throughout the body, then, the carrier moiety is quickly
oxidized to the corresponding quaternary pyridinium ion (E2-Q+) in the brain
as well as in peripheral tissues. The charged pyridinium-drug complex is
thus locked into the CNS while the same moiety rapidly clears from the
periphery because of a 40,000-fold increase in its hydrophilicity. Sustained
release of the active, parent drug from the charged pyridinium-drug complex
occurs in the brain as a result of enzymatic hydrolysis of the ester linkage.
The enzymes involved in cleavage of the ester bond are believed to be non
specific esterases.
Collectively, the ability to preferentially deliver E2 to the brain, thus
sparing non-target peripheral site tissues, should improve the therapeutic
index of E2 by (i) increasing the concentrations and/or residence time of E2 at
its receptor site in the brain and (ii), equally important, decreasing the
concentrations and/or residence time of E2 at the potential peripheral sites of
toxicities, thereby decreasing untoward peripheral side effects.
To document the predictive biotransformation behaviors of the E2-CDS
(Bodor et al., 1987), and to further substantiate its effectiveness over the
currently used estrogens, extensive and long-term pharmacokinetic and
pharmacodynamic studies were conducted. These studies included the


108
Figure 14. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in fat of ovariectomized rats. Animals received
a single iv (tail vein) injection of the E2-CDS on day 0 at a dose of
1.0 mg/kg (upper panels), 0.1 mg/kg (middle panels), or 0.01
mg/kg (lower panels). Animals were killed 1, 7,14, 21, or 28 days
after the drug administration. Tissue samples of known wet
weight at a concentration of 1 mg/20 |il solvent were processed
and assayed for E2-Q+ and E2 by the method described in Chapter 4.
Also, tissue homogenates from HPCD-treated rats were analyzed
for E2 hormone background. Represented are means SEM for n
= 7 rats per group per sampling time. Indicates below the
sensitivity limit of the assay.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E8SDNM882_VNZZ25 INGEST_TIME 2015-04-06T19:00:13Z PACKAGE AA00029743_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


Plasma Cone, (ng/ml) Plasma Cone, (ng/ml)
110
Figure 15. Effects of a single iv dose of the E2-CDS (1.0 mg/kg bw) on plasma
E2-Q+ (upper panel) or plasma E2 (lower panel) concentrations in
ovariectomized rats. Animals received a single iv (tail vein)
injection of the E2-CDS on day 0 at a dose of 1.0 mg/kg.
Immediately after drug treatment, animals were transferred to
sampling chambers and blood samples were removed at 0.5,1, 2, 4,
8,12, 24, 48, 96, and 168 hrs postinjection. Plasma samples were
assayed for E2-Q+ and E2 as described in Chapter 4. Represented
are means SEM for n = 5 rats per group per sampling time.


187
Kelly M.J., R.L. Moss, and C.A. Dudley. The effect of ovariectomy on
responsiveness of preoptic septal neurons to microelectrotophoresed
estrogen. Neuroendocrinology 25:204-211 (1978).
King R.J.B., J. Gordon, and D.R. Inmam. The intracellular localization of
oestrogen in the rat tissues. T. Endocrinol. 332:9-16 (1965).
Klaiber E.L., D.M. Broverman, W. Vogel, and Y. Kabayashi. The use of steroid
hormones in depression. In: Psychotropic Action of Hormones, T.M.
Itil, G. Laudahn, and W. Herrmann (eds), 135-157, Spectrum Publ Inc.,
New York (1976).
Klaiber E.L., D.M. Broverman, W. Vogel, and Y. Kabayashi. Estrogen therapy
for sever persistant depressions in women. Arch. Gen. Psych. 36:550-
554 (1979).
Klein L.H. Prostatic carcinoma. New Eng. T. Med. 300:824-833 (1979).
Klopper A. Endocrine and metabolic diseases: Treatment of infertility and
menopausal symptoms. Br. Med. T. 2:414-416 (1976).
Knauer E. Die ovarian-transplantation. Arch. Gvnaekol. 60:322-376 (1900).
Knobil E. The neuroendocrine control of the menstrual cycle. Recent Prog.
Horm. Res. 36:53-74 (1980).
Koch B., B. Lutz-Bucher, B. Briaud, and C. Mialhe. Glucocorticoid binding to
plasma membrane of the adenohypophysis. T. Endocrinol. 73:399-400
(1977).
Kopera H. Estrogens and psychic functions. Front. Horm. Res. 2:118-133
(1973).
Krey L.C., F. Kamel, B.S. McEwen. Parameters of neuroendocrine
aromatizaron and estrogen receptor occupation in the male rat. Brain
Res. 193:277-283(1980).
Labrie F., A. Dupont, A. Belanger, F.A. Lefebyre, L. Cusan, G. Monfette, J.G.
Leberge, J.P. Emond, J.P. Raymond, J.M. Husson, and A.T.A. Fazekas.
New hormone treatment in cancer of the prostate: Combined
administrations of an LHRH agonist and an antiandrogen. T. Steroid
Biochem. 19:999-1007(1983).
Lauritzen C. The management of the pre-menopausal and the post
menopausal patient. In: Aging and Estrogens, P.A. van Keep and C.
Lauritzen (eds), Karger, Basel (1973).


Treatment Treatment
145
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
7 Days After Final Treatment
0 1 2 3 4 5 6
a,b
i
7
14 Days After Final Treatment
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
H a,b
T"
2
-r
3
5
-i
6
Anterior Pituitary mg/100 g bw
Figure 23. Effects of the E2-CDS (0.5 mg/kg) or CAST on anterior pituitary
weight either 7 days (upper panel) or 14 days (lower panel) after
the final treatment. Animals received weekly iv (tail vein)
injection of either the drug's vehicle (HPCD), or E2-CDS at a dose
of 0.5 mg/kg (1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3
injections; x3), or were castrated (CAST). Represented are means
SEM for n = 7-8 animals per group per sampling time. The symbol
(a) denotes differences from control (HPCD)-treated animals and
the symbol (b) indicates differences from CAST within the panel
as analyzed by ANOVA and Scheffe statistics.


32
the brain. Both of these treatment strategies lead to sustained increases in
peripheral estrogen levels. However, when estrogen is attached to the
dihydropyridine carrier, the E2-CDS enhances brain-specific delivery of
estrogen by (i) locking the estrogen into the brain following the oxidation of
E2-CDS to the charged pyridinium moiety (E2-Q+); and (ii) enhancing the rate
of elimination of the lipoidal estrogen, in the inactive E2-Q+ form, from the
periphery following its oxidation to the charged, more hydrophilic
compound.
Therapies Aimed at Targeting/Enhancing Brain Estradiol Levels
Fertility Regulation
Fertility control may be achieved by a wide variety of mechanical,
surgical, and chemical methods. The chemical (steroidal) methods of fertility
control was first introduced by Pincus and Chang (quoted by Tepperman,
1981) and since then it has had important repercussions on population
growth. Commonly used steroid contraceptives consist of synthetic estrogens
in combination with progestins. When given at pharmacological doses
and/or constant exposure, E2 inhibits (via negative-feedback mechanism) the
secretion of gonadotropin hormone-releasing hormone (GnRH) from the
hypothalamus (Kalra & Kalra, 1983,1989; Plant, 1986) and hence, of
gonadotropins (LH and FSH) from the anterior pituitary (Kalra & Kalra, 1989).
The inhibition of the hypothalamic-pituitary-ovarian axis prevents follicular
development and therefore ovulation (Briggs, 1976). However, the use of
oral contraceptives has been associated with many adverse metabolic changes,
including increased risk of coronary atherosclerosis, myocardial infarction in


112
Table 5: Effects of the E2-CDS on the Clearance of E2-Q+ from a Variety of
Tissues
Tissue
Dose
Davs after treatment
(mg/kg)
la
7b
14b
21b
28b
Brain
1.0
329.97
42
82
92
97
Hypothalamus
1.0
310.58
16
80
88
96
Anterior
Pituitary
1.0
NDc
ND
ND
ND
ND
Plasma
1.0
21.11
96
99
>99
UDd
Kidney
1.0
487.56
80
96
98
>99
Lung
1.0
716.94
83
98
99
>99
Heart
1.0
910.39
82
98
>99
>99
Liver
1.0
374.57
94
98
>99
UD
Fat
1.0
176.11
97
99
>99
UD
Uterus
1.0
92.58
89
93
97
>98
a Initial concentrations of E2-Q+ (ng/g wet tissue or ng/ml) 1 day after
administration of the E2-CDS.
b % reduction in E2-Q+ concentrations at various times after E2-CDS
treatment relative to the initial corresponding values,
c Not determined due to incorrect extraction of the tissue at day 1.
d Undetectable (below the sensitivity of RIA for E2).


170
these factors may be important only if the E2-CDS does not achieve an
equilibrium or steady estate within 30 min after its administration.
Otherwise, since the E2-CDS does not bind to estrogen receptors itself, and
since it is very lipophilic, it would be redistributed among all tissues and thus,
the contribution of blood flow would be of minimal importance. However,
after the oxidation of E2-CDS to E2-Q+, it is then the unique property of an
organ or tissue, i.e. BBB, that serves the bases for retaining the oxidized
metabolite within a tissue.
Several other important issues regarding the regional distribution
and/or localization of the E2-CDS metabolites, E2-Q+ and E2, remain to be
resolved. Based upon our observations and comparisons of the whole brain
and the hypothalamus (Chapters 5 & 6) and the comparisons made by Sarkar
et al. (1989) of the hypothalamus, preoptic area and the cerebral cortex, it
seems that the E2-CDS is distributed evenly throughout the brain.
Furthermore, no obvious differences were observed with respect to its
oxidation, hydrolysis, and clearance from these brain regions evaluated to
date. This suggests that the oxidation and the hydrolysis reactions leading to
E2 formation are not unique to a particular set of neurons or even other cells
in the body.
Regarding the enzymatic conversion and cellular localization of the E2-
CDS metabolites, currently there are no documented reports to demonstrate
that the E2-CDS metabolism is an intracellular phenomenon. However, it is
believed that the multiple enzymatic reactions are primarily an intracellular
events. Using rat tissue homogenates as the test matrix, the in vitro
observation showed a faster oxidation rate of E2-CDS to E2-Q+ in tissue
homogenates than in plasma (Bodor et al., 1987). That suggests the
involvement of membrane-bound enzyme, most likely the ubiquitous NAD


67
Figure 1. Schematic representation of in vitro synthesis and in vivo
transformation of the estradiol-chemical delivery system (E2-CDS).
E2-Q+ is the charged quaternary form of the E2-CDS which is
"locked" into the brain and quickly eliminated from the
peripheral tissues. Subsequent hydrolysis of E2-Q+ with non
specific esterases results in sustained and slow release of estradiol
in the brain. The trigonelline, carrier moiety, formed upon
hydrolysis of E2-Q+ is non-toxic and is cleared from the brain
rather quickly. Although the in vivo rate constants for these
reactions are unknown among different tissues, the oxidation of
E2-CDS to E2-Q+ is quite rapid in all tissues analyzed (ti/2 = 29
min). However, the rate constant for hydrolytic enzymes may
differ among various tissues.


91
into its structure. We determined (1) the effects of E2-CDS dose on tissue
concentrations of E2-Q+ and E2, (2) the effects of E2-CDS dose on the rate of
oxidation of E2-CDS to E2-Q+ and hydrolysis of E2-Q+ to E2, and (3) the effects
of E2-CDS dose on the clearance of E2-Q+ and E2 in a variety of tissues in the
female rat. More specifically, our objective was to analyze quantitatively both
E2-Q+ and E2 (two metabolites of the E2-CDS) in brain, hypothalamus, anterior
pituitary, kidney, lung, heart, liver, fat, uterus, and plasma following a single
iv injection of one of several different doses of the E2-CDS over a 28-day time
course in ovariectomized (OVX) rats.
The rationale for using OVX female rat model in this study was
twofold: 1) this animal model exhibits very low endogenous estrogen levels
since both ovaries have been removed, thus allowing reliable determination
of tissue metabolites of the E2-CDS and 2) the results from this animal model
would allow us to make comparison with the results obtained from the male
rat model (Chapter 5) with regard to E2-CDS distribution.
Materials and Methods
Experiment 1
To evaluate the dose- and time-dependent effects of E2-CDS on the
tissue distribution of E2-Q+ and E2 in female rats, all animals were bilaterally
ovariectomized (OVX) under metofane anesthesia. All experiments were
initiated exactly 2 weeks after ovariectomy. On day 15 after ovariectomy, rats
(7 per group per each time point) were administered a single iv injection (tail
vein) of the E2-CDS at doses of 0 (HPCD), 0.01, 0.1, or 1.0 mg/kg body weight or
E2 at a dose of 0.7 mg/kg (equimolar to the 1.0-mg/kg dose of E2-CDS).


58
Results
In Vitro Methodology
Cross-reactivity of the E? antibody with E?-Q+
The inhibition of binding of 1251-E2 to the E2 antibody caused by E2 and
E2-Q+ is shown in Figure 2. While E2 effectively competed for binding with
the labeled hormone (IC50 = 368 pg/ml; 1.35 pM), E2-Q+ was ineffective in
displacing the 125I-E2 in the RIA (IC50 = 129,000 pg/ml; 326.50 pM). At
concentrations of E2-Q+ of 15 ng/ ml and higher, the cross-reactivity of the E2
antibody for E2-Q+ was <0.3%. Cross-reactivity of the E2 antibody for estriol
and estrone has been determined to be 0.32% and 1.1%, respectively (technical
information from Diagnostic Products).
Recovery of E?
Recovery of E2 was assessed by determining the % of each of three doses
of E2 recovered from the brain, liver, kidney, lung, heart, and fat
homogenates (Table 1). Due to limitations in the amount of tissue available,
only one dose (180 pg/ml) was tested for homogenates of anterior pituitary
glands. Recovery of E2 from each tissue evaluated was found to be dependent
upon the organic solvent used in the extraction step. For brain, anterior
pituitary, liver and kidney, E2 extraction with 100% methanol was found to be
superior to two other solvents (acetone and acetonitrile) resulting in E2
recoveries (average of the three doses tested) of 77% for brain, 78% for
anterior pituitary, 72% for liver and 71% for kidney (Table 1). Methanol was
determined to be a poor solvent for extracting E2 from lung, heart and fat
tissue, but 100% acetone achieved acceptable recovery of E2 in these three


171
<=> NADH transhydrogenase, in mediating the oxidation step (Hoek &
Rydstrom, 1988). This enzyme/coenzyme system is primarily located in the
inner mitochondrial membrane. In fact, the E2-CDS was designed to
biomimic this coenzyme moiety for its oxidative metabolism. Furthermore,
since the oxidoreductive enzymes are widespread, it is expected that all tissues
are capable of converting the E2-CDS to the corresponding quaternary
pyridinium salt (E2-Q+). Likewise, the hydrolysis of E2-Q+ to E2 is mediated by
widespread non-specific esterases present in tissues (Bodor et al., 1987;
Brewster et al., 1987). We examined the disappearance of E2-Q+ to E2 in tissue
homogenates and blood, approximately 1.2% of an E2-Q+ dose was converted
to E2 in brain homogenate, 3.1% in whole rat blood, and 2.3% in liver after
120 min incubation of E2-Q+. The slow rate of hydrolysis of E2-Q+ to E2 in an
in vitro condition is consistent with the in vivo sustained release of E2 in
brain tissue over 28 days following administration of E2-CDS in OVX female
rats (Chapter 6; Rahimy et al., 1990b). Certainly, more detailed studies
regarding the cellular localization of these enzymatic events and their
products will lead to further insight into the understanding of mechanism of
action of this delivery system.
The pharmacodynamic data demonstrates that the E2-CDS causes dose-
dependent and chronic suppression of LH and FSH secretion in OVX rats,
despite the rapid peripheral clearance of the E2-CDS metabolites (Chapter 7;
Rahimy et al., 1990c). The time-course and the magnitude of gonadotropin
suppression observed are quite comparable to the previously reported chronic
effects of the E2-CDS following a single iv administration to several animal
models (Anderson et al., 1987a,b, 1988a,b, 1989; Estes et al., 1987a b, 1988; Sarkar
et al., 1989; Simpkins et al., 1986,1988,1989a,b). These sustained and
prolonged pharmacological effects further support the idea that the E2-Q+ is


93
were removed at 0.5,1, 2, 4, 8,12, 24, 48, 96, and 168 hrs post-injection. At each
sampling time, the blood was centrifuged and the plasma collected for E2-Q+
and E2 analysis. Red cells were then resuspended in 0.5 ml heparinized saline
(40 units/ml) and returned to each respective animal before the next blood
sample. The volume of resuspended red cell solution was approximately 1
ml. It should be emphasized that although attempts were made to minimize
the occurrence of potential problems associated with the design of this
experiment, certain problems were unavoidable and thus should be
mentioned. During the initial 12-hrs of the time course of this experiment,
approximately 3 ml of plasma (i.e. 1/2 the total blood volume withdrawn)
were collected from each animal. This volume of plasma was needed to
analyze both metabolites of the E2-CDS in plasma. However, this volume
represents approximately 25% of total blood volume in an adult rat. And
since red cells were resuspended in equal volume of heparinized saline and
returned to each animal before the next blood sample, this procedure might
decrease the plasma concentrations of the E2-CDS metabolites.
Coat-A-Count Estradiol kits, a solid-phase [125I]-radioimmunoassay,
designed for the quantitative measurement of E2 in serum were used for the
assay of E2 in all tissue samples. All the purified dried E2 unknowns were
reconstituted in 300 |il of the assay buffer (kit Zero Calibrator) and assayed in
duplicate by the RIA. The intraassay and interassay coefficients of variation
for E2 were 1.56 and 6.1%, respectively. All samples were determined in 14
RIA runs.


194
Pfaff D. Nature of sex hormone effects on rat sex behavior: Specificity of
effects and individual patterns of response. T. Comp. Physiol. Psychol.
73:349-358 (1970).
Pfaff D. and M. Keiner. Atlas of estradiol-concentrating cells in the central
nervous system of the female rat. T. Comp. Neurol. 151:121-158 (1973).
Pietras R.J. and C.M. Szego. Estrogen receptors in uterine plasma membrane.
T. Steroid Biochem. 11:1471-1488 (1979).
Plant T.M. Gonadal regulation of hypothalamic gonadotropin-releasing
hormone release in primates. End. Rev. 7:75-88 (1986).
Pliner P. and A.S. Fleming. Food intake, body weight and sweetness
preferences over the menstrual cycle in humans. Physiol. Behav.
30:663-666 (1983).
Pohl C.R. and E. Knobil. The role of the central nervous system in the control
of ovarian function in higher primates. Ann. Rev. Physiol. 44:583-
(1982).
Pollard M., P.H. Luckert, and D. Snyder. Prevention and treatment of
experimental prostate cancer in Lobund-Wistar rats. I. Effects of
estradiol, dihydrotestosterone, and castration. The Prostate 15:95-103
(1989).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Distribution of a brain-
enhanced chemical delivery system for estradiol. Pharm. Res. [Suppl.]
5:S-205 (1988).
Rahimy M.H., J.W. Simpkins, and N. Bodor. A rapid, sensitive method for
the simultaneous quantitation of estradiol and estradiol conjugates in a
variety of tissues: Assay development and evaluation of the
distribution of a brain-enhanced estradiol-chemical delivery system. L
Steroid Biochem. 33:179-187 (1989a).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Pharmacodynamics of a novel
estradiol-chemical delivery system for the brain. Pharm. Res. [Suppl.]
6:S-211 (1989b).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Tissue distribution of a brain-
enhanced chemical delivery system for estradiol. Drug Des. Deliv.
6:29-40 (1990a).


38
as tricyclic antidepressants or monoamine oxidase inhibitors, which inhibit
the monoamines re-uptake or their respective metabolism (Bunney &
Garland, 1982; Leonard & Kuschinsky, 1982).
The biological basis for the involvement of estrogens in depression
comes from two gynecological problems: premenstrual and postmenopausal
syndromes. The premenstrual syndrome (PMS) refers to the various mood
changes in relation to the menstrual cycle that are experienced by a large
population of fertile age women. When the relationship between symptom
development and normal variations during the menstrual cycle was
examined (Backstrom et al., 1985), a consistent observation was that there
were very few negative symptoms, rather an increased sense of well being
during the preovulatory E2 peak of the menstrual cycle. However, the
maximum degree of symptoms or mood changes occurred during the luteal
phase of the menstrual cycle when progesterone levels are increased. Clinical
studies as early as in 1932 (Bowman & Bender) suggested a possible
therapeutic role of estrogen in the treatment of depression. More recently,
Klaiber et al. (1979) reported on a double-blind study performed to assess the
therapeutic efficacy of estrogen in the treatment of severely depressed
women. The estrogen treatment significantly decreased the degree of
symptoms compared with placebo (Klaiber et al., 1979).
Since the 1930's, numerous other clinical studies have provided
evidence regarding the influence of estrogens on the well-being and mental
performance in postmenopausal women. Hawkinson (1938) reported a
significant improvement in menopausal symptoms including depression and
anxiety. When subjective indices of moods in ovariectomized women were
evaluated in response to estrogen replacement therapy by Rauramo (1975),
estrogen treatment resulted in an elevation in moods to that which was


59
tissues. The mean recovery of E2 was 57% for lung, 62% for heart and 64% for
fat tissue. The E2 recovery from plasma was 81%.
Precision of the E? extraction-assay method
The precision of the method of estimating E2 concentrations in a
variety of tissues was determined in three ways. First, we determined the
coefficient of variation (CV) for quadruplicate samples of each tissue at 3
different E2 dose levels. Second, we determined the correlation coefficient
(CC) of the E2 dose-RIA response for each tissue. And third, we determined
the tissue weight-RIA response for brain samples taken at various times after
rats were treated with the E2-CDS.
The CV for quadruplicate determinations of E2-spiked tissue
homogenates or plasma pools were in the range of 0.8 to 7.9% with the
majority of E2 doses in each tissue showing a CV of less than 3.0% (Table 1).
The CC for the E2 dose-RIA response was 0.97 or greater for all tissues,
indicating that over the E2 dose-range tested, the procedure accurately
estimated the E2 concentration of the tissue (Table 1; Figure 3, lower panel).
To determine the accuracy of the method in estimating E2
concentrations in tissues of different weights and hence, to evaluate for tissue
components which may interfere with various steps in the E2 assay
procedure, we evaluated tissue wet weights over the range of 2 to 100 mg
(Figure 4, upper panel). Brain tissue from rats treated 1 or 7 days before with
the E2-CDS was extracted and assayed for E2. The inhibition of binding of 125I-
E2 was correlated with the weight of tissue used for both samples taken at day
1 (CC = 0.99) and 7 days (CC = 0.99) after E2-CDS administration. The
parallelism of the observed inhibition curves indicated that the E2
concentration measured is independent of the weight of tissues used in the


77
Materials and Methods
Adult, intact male Charles River (CD) rats (225-250 g) received a single
iv injection (tail vein) of the E2-CDS at a dose of 1.0 mg/kg body weight or the
drug's vehicle, 2-hydroxypropyl-p-cyclodextrin (HPCD). Rats (6-7 per group)
were killed by decapitation 1, 7 or 14 days after the drug administration and
the trunk blood was collected in heparinized tubes. The blood was
centrifuged and the plasma separated and stored at -20C until hormone
analysis. Tissues (brain, anterior pituitary, lung, liver, kidney, heart, and fat)
were dissected immediately following decapitation and stored at -80C until
hormone analysis.
Tissue samples of known wet weight at a concentration of 1 mg/20 pi
solvent were processed and assayed by the method described in Chapter 4
(Rahimy et al., 1989a). Tissue homogenates and plasma from HPCD-treated
rats were also extracted to determine the residual E2 concentrations and
thereby served as the estimate of hormone background.
Coat-A-Count Estradiol kits--a solid-phase [125I]-radioimmunoassay--
designed for the quantitative measurement of E2 in serum were used for the
assay of E2 in all tissue and plasma samples. Cross-reactivity of the E2
antibody was determined to be <0.3% for E2-Q+ at a concentration of 15 ng/ml
and higher (Chapter 4). All the purified dried E2 unknowns were
reconstituted in 300 pi of the assay buffer (kit Zero Calibrator) and assayed in
duplicate by the RIA. The intra-assay coefficient of variation for E2 was 1.56%
and all samples were determined in two assay runs.
Calculated values obtained from the RIA run were adjusted for the
volume of the aliquot taken for the RIA, experimental losses during solvent


172
"locked" behind the BBB and there it serves as a brain depot for E2 (Bodor et
al., 1987).
The present work also evaluated the potential for clinical application of
E2-CDS in several estrogenically responsive animal models including
androgen-dependent prostatic tissue growth/hyperplasia (Chapter 8) and
morphine-dependent, naloxone-withdrawal rat model for menopausal hot
flushes (Chapter 9). The encouraging results obtained from these
experimental studies indicate that the E2-CDS may have potential application
for the effective treatments of androgen-dependent prostatic hyperplasia as
well as vasomotor hot flushes of postmenopausal women. Although we
used normal male rats (Chapter 8) as the model for prostatic adenocarcinoma
to examine the effectiveness of E2-CDS in reducing testosterone levels with
subsequent regression of the prostatic tissue, there are no conclusive data as to
whether this model reflects a complete picture analogous to that of the aging
human prostate. However, since the primary objective of endocrine therapy
in prostate malignancy is the induction of an effective androgen deprivation,
thus abolishing the growth promoting effects of androgens on the diseased
prostate, the E2-CDS is quite capable of inducing an effective chemical
castration-like effect (Chapter 8). Indeed, the E2-CDS at a single dose of 0.5
mg /kg was as effective as CAST not only in suppressing T levels but also in
regression of the androgen-sensitive prostatic tissue. Further experimental
(animal models for prostatic adenocarcinoma) as well as clinical investigation
pertaining to evaluation of therapeutic efficacy of this novel E2-CDS in this
regard as well as the menopausal hot flushes are warranted.
Finally, some of the potential complications or disadvantages of this
drug delivery system need to be mentioned. Although brain-enhanced
delivery with sustained release of E2 in that tissue via a redox-based system


CHAPTER 2
REVIEW OF THE LITERATURE
This chapter will first present a historical review with respect to the
endocrinology/neuroendocrinology of estrogen hormones. This will include
some evidence pertinent to their physiological/pharmacological actions in
the central nervous system (CNS). Furthermore, since the unique
architecture of the brain, the blood-brain barrier (BBB), is of central asset in
the design and synthesis of the estrogen-delivery system under investigation,
a historical account of the BBB will be discussed. Attempts will be made to
identify problems associated with the brain delivery of existing drugs. Finally,
example of certain clinical conditions which require the presence of estrogen
in the brain as a therapeutic agent will be discussed as well. The purpose of
this diverse literature review is to identify and describe the concepts and
rationale which were the basis in the design and synthesis of the E2-CDS,
which will be evaluated in detail in later chapters.
Estrogen Hormones
Historical Observations
Ovarian endocrine activity was first demonstrated experimentally by
Knauer in 1896 (quoted by Tepperman, 1981). Independently, Sobotta (1896)
described the origin of corpus luteum at the same time. Shortly thereafter,
Beard (1897) postulated that the corpus luteum might serve a necessary
function during pregnancy. The observation by Knauer (1900), who
7


50
Coat-A-Count Testosterone kitsemploying solid-phase [12511-
radioimmunoassaypurchased from Diagnostic Products were used for the
plasma testosterone (T) assay. The RIA sensitivity for the T assay was 0.2
ng/ml. The cross-reactivity of the T antibody to the DHT and E2 has been
reported to be 3.3% and 0.02%, respectively (technical information from
Diagnostic Products).
Statistical Analysis
For the experimental data that were normally distributed, the
significance of differences among groups were determined by one- or two-way
analysis of variance (ANOVA) (Zar, 1974). Where necessary, data were
transformed (In) prior to ANOVA. Subsequent pairwise comparisons were
made by Dunnett's or Scheffe's multiple range tests using Statview 512+
program from Brainpower, Inc. (Calabasas, CA) for the Macintosh computer.
Where appropriate, data were subjected to area under the curve (AUC)
analysis using the trapezoid method (Tallarida & Murray, 1981) and group
means for AUC were subjected to ANOVA and Scheffe's tests. The level of
probability for all tests were p < 0.05. The specific statistical design employed
in each experiment is indicated in the figure legends.


122
its rate of elimination from the brain by either local metabolism or its
redistribution down a concentration gradient into the general circulation.
We observed a significant elevation in plasma PRL in response to the
highest dose of E2-CDS (1.0 mg/kg), whereas lower doses had no significant
effect on plasma PRL concentrations. It appears then that elevations in
plasma PRL correlate with the administration of E2-CDS at doses which result
in the transient elevation of plasma E2 levels, but not at doses at which
plasma E2 remains low (Chapter 6). This apparent stimulation of PRL
production by the E2-CDS would appear to be due to the well described actions
of E2 on the anterior pituitary lactotrophes (Chen & Meites, 1970). However,
the lack of a positive temporal correlation between plasma PRL (present
study) and plasma E2 levels (Chapter 6) suggests the possibility that E2 released
in the brain might be responsible for a direct stimulation of the anterior
pituitary. This can be explained by the anatomical relationship between the
hypothalamus and the anterior pituitary gland. Estradiol released from the
E2-Q+/ or the E2-Q+ itself, which is "locked" into the brain, could be delivered
directly to the anterior pituitary by the capillary plexus of the hypophyseal
portal system (Traystman, 1983). These capillaries in the median eminence
lack features of other brain capillaries and hence are not part of the BBB
(Traystman, 1983).
The E2-CDS had no significant effects on the mean plasma GH
concentrations over the 28 days time-course at any of the 3 doses examined.
However, a careful analysis of the effects of E2-CDS on pulsatile GH secretion
(Millard et al., 1990) revealed that while mean GH levels are not changed,
baseline GH values were elevated and GH pulse amplitudes were moderately
reduced at 7 days after E2-CDS administration.


Tissue Cone, (ng/g) Tissue Cone, (ng/g)
86
Figure 7. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on liver and
fat concentrations of E2-Q+ (upper panels) and E2 (lower panel) in
intact male rats. Intact male rats were injected with a single iv
dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14
days after treatment. Each point represents the group mean SEM
(n = 6-7 rats for each time point).


21
(Levine & Ramirez, 1982), and the dynamics of the LHRH pulsatile mode is
essential for the differential synthesis and release of gonadotropins (Marshal
& Kelch, 1986). Furthermore, ovarian hormones concurrently modulate
LHRH pulse parameters and, therefore, are important regulators of
gonadotropin secretion. For instance, the preovulatory rise in E2 increases
the LH pulse frequency, and since there is a good concordance between LH
pulses in plasma and LHRH pulses in hypothalamic portal blood, this
indicates that a transient elevation in E2 levels increases the frequency of
LHRH secretion (Marshal & Kelch, 1986). The feedback effects of E2 not only
modulate the release of LHRH from the hypothalamus but also the
responsiveness of the pituitary to LHRH signals (Marshal & Kelch, 1986).
Therefore, the effects of E2 on the gonadotropins LH and FSH are coordinated
at the levels of both the hypothalamus and pituitary.
Estradiol displays both inhibitory and stimulatory effects on the
secretion of these hormones. Inhibitory, or negative feedback effects, are seen
during periods of basal LH secretion throughout the menstrual cycle. During
the follicular phase, the follicle secretes low levels of E2, and during the luteal
phase, the corpus luteum secretes large quantities of both estrogen and
progesterone. The combined effects of these steroids inhibit the secretion of
LHRH and consequently reduce the release of pituitary gonadotropins.
However, the positive feedback effects, or stimulation, are observed after a
transient and progressive increase in the titers of E2. That is, an increase in
serum E2 levels of 150 pg/ml or greater for 24 to 36 hours during the late
follicular phase of the menstrual cycle (Ross, 1981). This occurs in response to
the increase in LH pulse frequency which subsequently results in the
preovulatory surge of LH. This condition subsequently causes ovulation of
the ovum from the Graafian follicle and the formation of a new corpus


23
peak to 40 to 80 pg/ml on the evening of the proestrus day of the cycle
(Butcher et al., 1974; Kalra & Kalra, 1977). E2 levels then decline rapidly to
basal levels on estrus. The production of E2 by the follicles of the ovaries is
controlled via feedback mechanism (Richards & Hedin, 1988).
Feedback regulation
Ovarian cycle of the rat, like the human, is regulated by the pituitary
gonadotropins LH and FSH. However, in the rat, the gonadotropins appear to
be primarily under the stimulatory influence of the hypothalamic LHRH
neuronal activity (Kalra & Kalra, 1980, 1983; McCann, 1982). This is evidenced
by the fact that, in ovariectomized steroid-treated rat, the positive feedback
effects of estrogen are expressed as a daily signal for mid-afternoon LH
hypersecretion which ensues for several days if concentrations of E2 are
maintained at or greater than E2 levels seen on the proestrus day (Legan et al.,
1975). As in the human and primate, the feedback effects of E2 on the
gonadotropin secretion are coordinated at the levels of both the
hypothalamus and the anterior pituitary gland (Kalra & Kalra, 1983, 1989;
plant, 1986).
In the rat estrous cycle, E2 also exhibits both inhibitory and stimulatory
effects on the hypothalamic-pituitary unit. The inhibitory, negative feedback,
effects are observed during periods of basal LH and FSH secretion throughout
the estrous cycle or after chronic exposure of ovariectomized rats to E2. The
inhibitory effects of E2 are exerted within the hypothalamus (the medial basal
and medial preoptic areas) and/or the anterior pituitary. Implantation of E2
in the medial basal hypothalamus (MBH) (Smith & Davidson, 1974) or
chronic elevation in serum E2 levels (Henderson et al., 1977) were shown to
suppress LH levels and increase the pituitary responsiveness to LHRH,


198
neurodegenerative diseases. In: Novel Approaches to the Treatment
of Alzheimer Disease. E.M. Meyer, J.W. Simpkins and J. Yamamoto
(eds) 197-212, Plenum Press, New York (1989b).
Sinkula A.A. and S.H. Yalkowsky. Ratioal for design of biologically reversible
drug derivatives: prodrugs. T. Pharm. Sci. 64:181-193 (1975).
Smith E.R. and J.M. Davidson. Location of feedback receptors; effects of
intracranially implanted steroids on plasma LH and PRL response.
Endocrinology 95:1566-1573 (1974).
Smith S.L. and C. Sauder. Food cravings, depression and premenstrual
problems. Psvchosom. Med. 31:281-287 (1969).
Smith S.S., B.D. Waterhouse, and D.J. Woodward. Sex steroids effects on
extrahypothalamic CNS. I. Estrogen augments neuronal
responsiveness to iontophoretically applied glutamate in the
cerebellum. Brain Res. 422:40-51 (1987).
Spelsberg T.C., B.J. Gosse, B.A. Littlefield, H. Toyoda, and R. Seelke.
Reconstitution of native-like nuclear acceptor sites of the avian
oviduct progesterone receptor: Evidence for involvement of specific
chromatin proteins and specific DNA sequences. Biochemistry
23:5103-5113 (1984).
Spona J. and W. Schneider. Bioavailability of natural estrogens in young
females with secondary amenorrhea. Acta Obstet. Gynecol. Scand.
[Suppl.] 65:33-38 (1977).
Steffens A.B. A method for frequent sampling of blood and continuous
infusion of fluids in the rat without disturbing the animal. Physiol.
Behav. 4:833-836 (1969).
Stella V. Pro-drugs: An overview and definition. In: Prodrugs as Novel
Drug Delivery Systems. T. Higuchi, V. Stella (eds) 1-115,American
Chemical Society, Washington, DC (1975).
Suckling A.J., M.G. Rumsby, M.W.B. Bradbury. The Blood-Brain Barrier in
Health and Disease. VCH Publishers, Chichester, England (1986).
Sulkava R., J. Wikstrom, A. Aromaa, R. Raitasalo, V. Lehtinen, K. Lahtela,
and J. Palo. Prevalence of sever dementia in Finland. Neurology
35:1025-1029 (1985).


92
Rats (7 per group) were killed by decapitation 1, 7,14, 21, or 28 days after
the drug administration and the trunk blood was collected in heparinized
tubes. The blood was centrifuged and the plasma separated and stored at -
20C until hormone analysis. Tissues (whole brain, hypothalamus, anterior
pituitary, kidney, lung, heart, liver, fat, and uterus) were dissected
immediately following decapitation, rinsed in ice-cold saline, stripped of
surrounding connective tissue where necessary, blotted dry on paper, and
then stored at -80C until hormone analysis.
Tissue samples of known wet weight at a concentration of 1 mg/20 pi
solvent were processed and assayed for E2-Q+ and E2 by the method described
in Chapter 4 (Rahimy et al., 1989a). Also, tissue homogenates or plasma from
HPCD-treated rats were analyzed to determine residual E2 concentrations and
thereby served as our estimate of hormone background.
Experiment 2
To accurately assess the kinetics of E2-CDS within the general
circulation, an acute 7-day time-course study with frequent blood sampling
was undertaken. To facilitate frequent blood sampling from unrestrained
animals, a group of OVX rats were equipped with Silastic catheter (i.d. 0.5
mm, o.d. 1 mm). The catheter was positioned in the right atrium via the
external jugular vein under pentobarbital anesthesia, according to the
procedure described by Steffens (1969).
After recovery from the surgical procedure (usually one week, 3 weeks
after OVX), rats were administered iv (via tail vein) with 1.0 mg/kg dose of
the E2-CDS. Immediately after drug treatment, animals were transferred to
special sampling chambers for serial blood sampling. Blood samples (1 ml)


Results
Experiment 1
To estimate the extent of in vivo oxidation of E2-CDS to E2-Q+, we
determined for each tissue the magnitude of increase in E2-Q+ concentrations
over the 100-fold increase in E2-CDS dose, at day 1 after injection (the first
sampling time). The enzymatic oxidation of E2-CDS to E2-Q+ showed a clear
dose dependency in brain, hypothalamus, plasma, kidney, lung, heart, liver,
and fat tissues (Table 4). This dose-related oxidation ranged from 73-fold in
liver to 176-fold in whole brain tissue over the 100-fold increase in E2-CDS
dose administered. The uterus was an exception and showed only a 21-fold
increase in E2-Q+ concentrations over the 100-fold increase in E2-CDS dose
(Table 4).
The in vivo rate of hydrolysis of E2-Q+ to E2 was estimated for each
tissue at each dose of E2-CDS administered by determining the ratio of E2 to
E2-Q+ on the first sampling day (day 1, Table 4). This ratio remained constant
over a 100-fold dose range for the hypothalamus, kidney, heart, and uterus; it
decreased moderately (less than 50%) over the 100-fold dose range for the
brain, lung, and liver and decreased precipitously for plasma and fat tissue
(Table 4).
At 1 day after administration of E2-CDS, all tissues showed a dose-
dependent increase in concentrations of E2-Q+ and E2. Furthermore, the
concentration-time profiles revealed a gradual decline in concentrations of
E2-Q+ and E2 in whole brain (Figure 10) as well as in hypothalamus (Figure
11), with ti/2 = 8-9 days. In contrast, both E2-Q+ and E2 were rapidly cleared


28
means for bidirectional movement of selective molecules. However, with
few exceptions, these carrier systems are not involved in transport of
chemotherapeutic agents across the BBB (Greig et al., 1987).
Potential Asset to Utilize in the Design of Brain-Specific Drug Delivery
The unique architecture of the BBB allows only the transport of
compounds either by specific transport systems or by simple diffusion directly
through cell membranes if they are to gain access to the brain parenchyma
and extracellular spaces. Therapeutic agents are no exception. They can access
the brain through either of these routes. Furthermore, the bulk transport of
materials is limited due to the sealing of endothelial gap junctions and the
lack of vesicular transport system in the cerebral capillaries. As a result, most
drugs that enter the CNS must do so by passive diffusion through the
phospholipid cellular matrix of capillary endothelial cells. The lipophilicity
of drugs, defined by their octanol-water partition coefficient, correlates with
their ability to penetrate the BBB for several classes of drugs, including
narcotics (Oldendorf et al.,1972), barbiturates (Levin, 1980) and P-receptor
antagonists (Cruickshank et al., 1979). Furthermore, drugs which cannot
penetrate the BBB can gain access to limited areas of the brain around the
circumventricular organs. Collectively, the CNS has evolved mechanisms to
protect itself by excluding hydrophilic (polar substances) and other
compounds which may be harmful to its optimal functioning.
Unfortunately, this barrier mechanism impedes the delivery and transport of
many potentially useful therapeutic agents to the CNS, thus severely
complicating the effective treatment of brain diseases.


62
In brain tissue, E2-Q+ levels exceeded E2 levels by 18-, 22- and 19-fold
and in plasma E2-Q+ levels were 6,13 and 22-fold higher than E2 at 1, 7 and 14
days, respectively.
Discussion
This novel, but predictable, metabolism of the E2-CDS presents several
problems for the quantitation of E2-Q+ and E2, two metabolites of the E2-CDS.
First, since estradiol is active at tissue concentrations of low pg/g, the assay
method for the E2-CDS metabolite, E2, must be extremely sensitive. Second,
since E2-Q+, the moiety "locked" in the brain, is expected to be present in
concentrations much higher than E2, the assay method must be capable of
distinguishing low levels of E2 in the presence of high concentrations of E2-
Q+. Third, the accuracy of the E2 determination is dependent upon the
stability of E2-Q+ against hydrolysis (enzymatic or spontaneous) throughout
the procedures. However, under the conditions utilized, E2-Q+ was quite
stable. When 300 pg E2-Q+ were added to tissue homogenates and evaluated
for spontaneous hydrolysis throughout the procedures used, the E2 recovery
was below the sensitivity of the assay. Finally, as for any assay method,
necessary features are (a) a high recovery of the species of interest; (b) accuracy
of the determinations; and (c) reliability of the method through a wide range
of hormone concentrations and tissue weights. We have provided evidence
for each of the features of the present method for simultaneous
determinations of E2-Q+ and E2.
The RIA procedure provides the needed sensitive endpoint for the
determination of E2 and E2-Q+ levels. This RIA for E2 is sensitive from 0.8 to
1.2 pg E2 / assay tube and exhibits a highly correlated inhibition of 125I-E2


113
Table 6: Effects of the E2-CDS on the Clearance of E2 from a Variety of Tissues
Tissue
Dose
Davs after treatment
(mg/kg)
la
7b
14b
21b
28b
Brain
1.0
26.63
40
71
91
96
Hypothalamus
1.0
25.32
24
68
77
91
Anterior
Pituitary
1.0
NDc
ND
ND
ND
ND
Plasma
1.0
0.32
78
87
93
95
Kidney
1.0
135.17
80
98
>99
UDd
Lung
1.0
167.67
95
99
>99
>99
Heart
1.0
156.03
82
99
>99
>99
Liver
1.0
8.43
90
98
>99
UD
Fat
1.0
2.32
84
97
>99
>99
Uterus
1.0
23.07
93
94
97
>99
a Initial concentrations of E2 (ng/g wet tissue or ng/ml) 1 day after
administration of the E2-CDS.
b % reduction in E2 concentrations at various times after E2-CDS
treatment relative to the initial corresponding values,
c Not determined due to incorrect extraction of the tissue at day 1.
d Undetectable (below the sensitivity of RIA for E2).


14
regulation. For instance, the CNS neurones produce diversity of both rapid
and delayed neuroendocrine and behavioral effects in response to estrogen
exposure (Majewska, 1987). Various studies have suggested that the neuronal
plasma membrane may serve as a direct target for the rapid action of
estrogens, which may lead to modification of neurotransmitter release or
their receptor/effector systems. First, neurophysiological studies on CNS
neurons have demonstrated rapid modulation of neuronal excitability,
including a brief hyperpolarization and increase in potassium conductance of
the postsynaptic membranes of medial amygdala neurones (Nabekura et al.,
1986), increase the firing rate of medial preoptic and septal neurones (Kelly et
al., 1978), and increase the cerebellar neuronal responsiveness to
iontophoretically applied glutamate (Smith et al., 1987) after application of
physiological levels of 17 (3-E2- Second, biochemical studies have also
provided evidence for the direct actions of estrogens on neuronal
membranes. This included an increase in amphetamine-stimulated striatal
dopamine release in vitro superfusion system with 17 (3-E2 or
diethylstilbesterol, but not with 17 0C-E2 (Becker, 1990) as well as in vivo
microdialysis in freely moving rats (Becker & Beer, 1986). Also, 17 P-E2 is
shown to enhance the responses of adenylate cyclase to biogenic amines in
striatal neurons culture (Maus et al., 1989). Finally, morphological
experiments have also demonstrated rapid plasma membrane ultrastructural
modifications in response to sex steroids application (Garda-Segura et al.,
1987, 1989). Employing freeze-fracture techniques, within 1 min,
physiological concentrations of 17 P-E2 increased the density of exo-
endocytotic pits in cerebrocortical and hypothalamic neuronal membranes in
culture, which was blocked by estrogen antagonist tamoxifen (Garcia-Segura
et al., 1987,1989). These rapid effects (from seconds to minutes of latency)


30
is concurrently enhanced from the CNS. This results in poor retention and
minimal or no improvement in the biological half-life of the drug.
To overcome the problem of potential general toxicity associated with
enhanced lipophilicity of prodrugs, novel redox-based chemical delivery
system (CDS) for drugs has been designed which exploits the unique
architecture of the CNS BBB (Bodor, 1987; Bodor & Farag, 1983, 1984; Bodor &
Simpkins, 1983; Bodor et al., 1981, 1987,1988). By definition, a CDS is a
biologically inert molecule which requires several chemical conversions
leading to the active, parent drug at or near the site of action (Bodor, 1987;
Bodor & Brewster, 1983). The multiple, facile chemical conversions may lead
to (a) selectivity in drug delivery; (b) improve the drug half-life; and (c)
decrease the toxicity of the drug. The redox-based CDS utilizes a carrier
molecule that can exist as a lipid soluble (in the reduced state) or water
soluble (in the oxidized state). The mechanism of its drug delivery is based
upon an interconvertible dihydropyridine <=> pyridinium ion carrier (Bodor,
1987). In this brain-specific CDS, the lipoidal dihydropyridine moiety is
attached to the drug, thus increasing its lipid solubility and thereby enhancing
its permeability through the BBB. The reduced dihydropyridine can be
oxidized, after its administration, to the pyridinium ion in the brain as well as
in the periphery including systemic circulation. The charged pyridinium-
drug complex is thus locked into the brain while the same moiety rapidly
clears from the periphery by renal or biliary processes due to its increased
hydrophilicity. Sustained release of the active, parent drug from the charged
pyridinium-drug complex occurs in the brain as a result of the enzymatic
hydrolysis of the ester (or amide, etc.) linkage between the drug and the
pyridinium moiety.


74
Table 3: Percent Hydrolysis of E2-Q+ in Supernatants of a Variety of Tissues
Tissue
E2-Q+ Added
(Pg)
E2-Q+ Assayed^
(Pg)
Hydrolysis
(%)
CVc
(%)
Brain
300
198.3
66.1
5.0
Plasma
300
87.8
30.0
2.8
Kidney
300
186.6
62.2
2.9
Lung
300
167.6
55.8
1.8
Heart
300
162.5
54.2
1.7
Liver
300
207.3
69.1
3.5
Fat
300
112.3
37.4
4.1
a 100 mg of tissue or 1 ml of plasma was used,
b Mean of n = 4 for each dose of E2-Q+ in each tissue,
c CV = coefficient of variation.


151
withdrawal rat model developed in our laboratory (Simpkins & Katovich,
1984). This animal model was originally developed to study the
neuroendocrine mechanism(s) leading to the flush response as well as
evaluating new and existing drugs for their efficacy in treating the
menopausal symptoms. We have previously demonstrated that naloxone-
induced withdrawal in the morphine-dependent rat model results in a surge
in TST that is preceded by tachycardia, accompanied by hypersecretion of LH,
and is followed by a significant fall in core body temperature (Simpkins &
Katovich, 1984). Each of these responses is similar in both magnitude and
duration and temporally associated to those observed in menopausal hot
flushes (Simpkins et al., 1983).
Materials and Methods
To evaluate the effects of E2-CDS on TST responses, adult female rats
were bilaterally ovariectomized (OVX) under metofane anesthesia and
experiments were initiated 2 weeks after ovariectomy. On day 15 after
ovariectomy, rats were randomly divided into 4 groups (7-8 rats per group).
These experimental groups were designated and treated as follows: 1) OVX +
HPCD control, animals in this group received weekly iv injection of HPCD
(vehicle) for 3 weeks; 2) OVX + E2 pellet, these animals were subcutaneously
(sc) implanted (under light metofane anesthesia) with an E2 pellet weekly for
3 weeks (removing the old pellet before implanting a new one); 3) OVX + E2-
CDS multiple injections, these animals received weekly iv injection of E2-
CDS at a dose of 1.0 mg/kg b.w. for 3 weeks; and 4) OVX + E2-CDS single
injection, these animals received a single iv injection of E2-CDS at 1 week
before testing.


CHAPTER 9
EFFECTS OF THE E2-CDS OR E2 PELLET ON TAIL-SKIN TEMPERATURE
RESPONSES IN OVARIECTOMIZED FEMALE RATS
Introduction
Over the past 50 years, numerous clinical studies in either natural or
ovariectomy-induced menopausal women have provided evidence that
gonadal steroid withdrawal causes alterations in the central thermoregulatory
system which then leads to vasomotor hot flushes (Casper & Yen, 1985;
Lauritzen, 1973; Yen, 1977). After menopause or ovariectomy, the decreasing
production of ovarian estrogens/progestins leads abruptly to a number of
central nervous system (CNS)-mediated steroid-withdrawal symptoms
(Casper & Yen, 1985; Lauritzen, 1973; Yen, 1977). These symptoms most often
express themselves as hot flushes, perspiration, depression, anxiety, changes
in memory, headaches, insomnia and irritability (Clayden et al., 1974;
Meldrum et al., 1979; Paterson, 1982). However, their intensity and frequency
vary among women and symptoms do not occur at all in about 15 to 25% of
menopausal women (Casper & Yen, 1985). The most frequent symptom of
the menopause is the hot flush, an episodic disturbance of thermoregulation
characterized by a sensation of heat followed by a sudden spreading of flush
and perspiration (Casper & Yen, 1985; Clayden et al., 1974; Lauritzen, 1973;
Meldrum et al., 1979; Paterson, 1982; Yen, 1977). These physiological
alterations appear to be the result of autonomic discharge which causes
peripheral vasodilation and heat loss, with a considerable drop in core body
temperature (Nesheim & Saetre, 1982). Furthermore, each flush is preceded
148


66
Application of these methods to brain and plasma samples obtained at
various times after treatment with the E2-CDS revealed that both E2 and E2-
Q+ can be quantitated throughout the 14-day time-course of the study. As
predicted, based upon previous reports on brain levels of E2-Q+ (Bodor et al.,
1987; Mullersman et al., 1988), this "locked-in" form of the E2-CDS reached a
33-fold higher concentration in the brain than plasma by 1 day after treatment
with E2-CDS and these brain-blood ratios increased to 294-fold by 14 days.
Brain E2 concentrations were similarly and dramatically elevated relative to
plasma. These observations are consistent with the proposed brain-enhanced
delivery of E2 with the redox-based E2-CDS and indicate that the observed
distribution pattern of E2-Q+ and E2 may explain the long-term
pharmacological effects of the E2-CDS (Anderson et al., 1988a,b; Estes et al.,
1987a,b; Simpkins et al., 1986).
In summary, the described technique for the simultaneous
measurement of E2-Q+ and E2 is sensitive, reliable, specific and applicable to a
wide variety of tissues in the body. The additional feature of rapidity of the
method allows for the determination of about 100 samples in one day.
Collectively, these characteristics indicate that the described techniques could
be applied to the quantiation of other conjugates of steroid hormones.


I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
'2si/yu,
es W. Simpkins, /Chairman
ofessor of Pharmacodynamics
I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
I-\X>
Nicholas Bodor
Graduate Research Professor of
Medicinal Chemistry
I certify that I have read this study and that in my opinion it conforr
to acceptable standards of scholarly presentation/nd is fully^Sdequate, in,
scope and quality, as a dissertation for the degree of Dpctpi/of Phos
lliam J.
Associate Professor of
Pharmacodynamics


197
Segal L., B. Segal, and W.R. Nes. The acid-catalyzed solvolysis of
dehydroepiandrosterone sulfate and its significance in the examination
of urinary 17-ketosteroids. T. Biol. Chem. 235:3108-3111 (1960).
Sherwin B.B. Estrogen and/or androgen replacement therapy and cognitive
functioning in surgically menopausal women.
Psychoneuroendocrinologv 13:345-357 (1988).
Shivers R.R. The blood-brain barrier of a reptile. Anoli's carolinensis. A
freeze-fracture study. Brain Res. 169:221-230 (1979).
Silverberg E. and J. Lubera. Cancer statistics. CA 36:9-25 (1986).
Silverberg S.O. and E.L. Makawski. Endometrial carcinoma in young women
taking oral contraceptive agents. Onstet. Gynecol. 46:503-506 (1965).
Simpkins J.W., W.R. Anderson, R. Dawson, Jr., N. Bodor. Effects of a brain-
enhanced chemical delivery system for estradiol on body weight and
food intake in intact and ovariectomized rats. Pharm. Res. 6:592-600
(1989a).
Simpkins J.W., W.R. Anderson, R. Dawson, Jr., A. Seth, M.E. Brewster, K.S.
Estes, and N. Bodor. Chronic weight loss in lean and obese rats with a
brain-enhanced chemical delivery system for estradiol. Physiol. Behav.
44:573-580 (1988).
Simpkins J.W., N. Bodor, and A. Enz. Direct evidence for brain-specific
release of dopamine from a redox delivery system. T. Pharm. Sci.
74:1033-1036 (1985).
Simpkins J.W., M.J. Katovich, and I.C. Song. Similarities between morphine
withdrawal in the rat and the menopausal hot flush. Life Sci. 32:1957-
1966 (1983).
Simpkins J.W. and M.J. Katovich. An animal model for pharmacologic
evaluation of the menopausal hot flush. In: The climacteric in
perspective. M. Notelovitz and P. van Keep (eds) 213-251, MTP Press,
Lancaster, UK (1984).
Simpkins J.W., J. McCornack, K.S. Estes, M.E. Brewster, E. Shek, and N. Bodor.
Sustained brain-specific delivery of estradiol causes long-term
suppression of luteinizing hormone secretion. T. Med. Chem. 29:1809-
1812 (1986).
Simpkins J.W., M.H. Rahimy, and N. Bodor. A brain-enhanced chemical
delivery system for gonadal steroids: Implications for


88
Figure 9. Brain (upper panels) and anterior pituitary (lower panels) contents
of the E2-Q+ and E2 following a single iv dose of the E2-CDS (1.0
mg/kg). Intact male rats were injected with a single iv dose of 1.0
mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14 days after
treatment. Whole brain tissue and the anterior pituitary were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Each point represents the group mean SEM (n = 6-7
rats for each time point). These calculations assumed a brain wet
weight of 2 grams (based on personal experience and knowledge).


39
described as being close to normal conditions. A recent study conducted by
Gerdis et al. (1982), using a variety of psychometric measures to estimate
depression, reported that three weeks estrogen treatment (Premarin) in
postmenopausal women significantly improved the symptoms of depression.
Furthermore, DeLignieres and Vincens (1982) reported improvement of
symptoms of depression, aggression, and anxiety in postmenopausal women
that were treated percutaneously with E2 for three months. In another study
(Klaiber et al., 1979), postmenopausal women with primary, recurrent
unipolar depression and a history of unsuccessful therapy of their depression
were treated with estrogen. The evaluation of their progress by the Hamilton
Rating Scale for Depression indicated a dramatic improvement in mean
Hamilton scores in some patients (Klaiber et al., 1979). These studies
suggested that estrogens may have antidepressant activity in postmenopausal
patients.
Another gynecological problem associated with gonadal steroid
withdrawal is the postpartum psychosis or depression. During the third
trimester of pregnancy, E2 levels increase to about 10 to 40 ng/ml (~ 1000-fold
increase over the follicular phase of the ovarian cycle) while progesterone
levels increase to about 100 to 400 ng/ml (~ 100- to 400-fold increase over the
follicular phase) (Ross, 1985; Schwartz, 1981). After parturition, gonadal
steroid levels fall precipitously and thus, it leads to CNS-steroid withdrawal.
Although the etiology of postpartum depression is yet unknown, it is
speculated that the decline in gonadal steroid levels after parturition may be
the primary factor leading to postpartum depression. In severe cases, it has
been reported that administration of estrogen to patients alliveated
postpartum depression by suppressing lactation (Yalom et al., 1968). Perhaps,
since estrogen hormones influence a variety of CNS functions, by modifying


CHAPTER 4
DEVELOPMENT OF AN ANALYTICAL METHOD FOR THE
QUANTITATION OF E2-CDS METABOLITES IN A WIDE VARIETY OF
TISSUES IN THE RAT
Introduction
The estradiol-chemical delivery system (E2-CDS) offers a novel
approach to non-invasively enhance brain delivery and sustained release of
E2 (Bodor et al., 1987). The E2-CDS is a redox-based chemical-delivery system
and exploits the unique architecture of the BBB, which normally excludes a
variety of pharmacological agents from the brain (Bodor & Brewster, 1983).
The mechanism of E2-CDS drug delivery is based on an interconvertible
dihydropyridine <=> pyridinium salt carrier. Figure 1 schematically shows the
structures and the mechanisms leading to brain-enhanced and sustained
release of E2. Estradiol, when it is chemically attached to the lipoidal carrier,
dihydropyridine, its lipophilicity is further increased and thus, the ability to
enter the brain is enhanced. After systemic administration of the E2-CDS, the
carrier system is then quickly oxidized to the corresponding quaternary
pyridinium salt (E2-Q+). This charged moiety of the carrier system reduces its
rate of exit from the brain, thereby locking a depot of E2-Q+ into the CNS.
Subsequent hydrolysis of the E2-Q+ with nonspecific esterases provides
sustained release of the active species (E2) in the brain. Since the E2-Q+ is
hydrophilic (40,000-fold greater than E2-CDS), its elimination rate from the
periphery is predictably much faster than from the brain.
51


200
Van Deurs B. Structural aspects of brain barriers, with special reference to the
permeability of the cerebral endothelium and choroidal epithelium.
Int. Rev. Cvtol. 65:117-191 (1980).
van Steenbrugge G.J., M. Groen, A. van Kreuningen, F.H. de Jong, M.P.W.
Gallee, and F.H. Schroder. Transplantable human prostatic carcinoma
(PC-82) in athymic nude mice. III. Effects of estrogens on the growth of
the tumor tissue. The Prostate 12:157-171 (1988).
Vanhulle G. and R. Demol. A double-blind study into the influence of estriol
on a number of psychological tests in post-menopausal women. In:
Consensus on Menopausal Research, P.A. van Keep, R.B. Greenblatt,
M. Albeaux-Fernet (eds) 94-99, MTP Press, London (1976).
von Schoultz B., K. Carlstrom, L. Collste, A. Eriksson, P. Henriksson, A.
Pousette, and R. Stege. Estrogen therapy and liver functionmetabolic
effects of oral and parenteral administration. The Prostate 14:389-395
(1989).
Walters M.R. Steroid hormone receptors and the nucleus. End. Rev. 6:512-
543 (1985).
Wei E., Loh H.H. and E.L. Way. Quantitative aspects of precipitated
abstinence in morephine-dependent rats. T. Pharmacol. Exp. Ther.
184:398-403 (1973).
Weiner N., F. Martin, and M. Riaz. Liposomes as a drug delivery system.
Drug Dev. Ind. Pharmacy 15:1523-1554 (1989).
Weiss N.S. Epidemiology of carcinoma of the endometrium. In: Reviews in
Cancer Epidemiology. Vol. 2, A.M. Lilienfeld (ed) 46-60, Elsevier
Science Publishing Company, New York (1983).
Weisz J. and W.R. Crowley. Catechol estrogen formation by the CNS:
Regional distribution of estrogen-2/4-hydroxylase activity in the rat
brain. Neuroendocrinology 43:543-549 (1986).
White J.W. The results of double castration in hypertrophy of the prostate.
Ann. Surg. 22:1-80 (1895).
Wiegand S.J., E. Terasawa, W.E. Bridson, R.W. Goy. Effects of discrete lesions
of preoptic and suprachiasmatic structures in the female rat.
Alterations in the feedback regulation of gonadotrophin secretion.
Neuroendocrinologv 31:147-157 (1980).


6. DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN FEMALE RATS 90
Introduction 90
Materials and Methods 91
Experiment 1 91
Experiment 2 92
Results 94
Experiment 1 94
Experiment 2 96
Discussion 96
7. EVALUATION OF THE PHARMACODYNAMIC EFFECTS OF E2-
CDS IN OVARIECTOMIZED FEMALE RATS 115
Introduction 115
Materials and Methods 116
Results 117
Discussion 119
8. EFFECTS OF THE E2-CDS OR CASTRATION ON ANDROGEN
AND ANDROGEN-DEPENDENT TISSUES IN MALE RATS 128
Introduction 128
Materials and Methods 131
Results 133
Effect of CAST or E2-CDS on Plasma T Levels 133
Effect of CAST or E2-CDS on Tissue Weights 134
Effect of CAST or E2-CDS on Plasma Hormones 135
Discussion 135
9. EFFECTS OF THE E2-CDS OR E2 PELLET ON TAIL-SKIN
TEMPERATURE RESPONSES IN OVARIECTOMIZED FEMALE
RATS 148
Introduction 148
Materials and Methods 151
Results 153
Discussion 155
10. GENERAL DISCUSSION 163
REFERENCES 175
BIOGRAPHICAL SKETCH 202
vii


46
Animals
The laboratory rat was chosen as the experimental animal for all
experiments herewith. Adult male and female Charles River (CD) rats (aged
3-5 months) were purchased from Charles River Breeding Laboratories
(Wilmington, MA). These rats weighed 200-250 g upon arrival and were
allowed several days to adjust to the animal quarters before conducting an
experiment. Animals were housed in a temperature- (24 1C) and light-
(lights on 0500 to 1900 hr daily) controlled room and provided with Purina rat
chow and tap water ad libitum. After a 7-day acclimation period, animals
were randomly divided into various experimental groups of 7-8 rats per
group. This number of rats per group is standard for the field and is based on
our estimates of experimental error in response to the drugs that were
evaluated in these studies.
In experiments which required surgical procedures, animals were
anesthetized with Metofane (Methoxy Flurane, Pitman-Moore Inc., Crossing,
NJ). The surgical procedures consisted of subcutaneous (sc) implantation of
drugs or steroids, gonadectomy, and atrial cannulation. Female rats were
bilaterally ovariectomized (OVX) by a small incision made through the dorsal
peritoneal cavity. Male rats were castrated (CAST) by exteriorizing the
testicles through a midline ventral incision. For atrial cannulation, in order
to facilitate frequent blood sampling from unrestrained animals, a small
incision in the neck-chest area was made to expose the external jugular vein.
A Silastic catheter (i.d. 0.5 mm, o.d. 1 mm) was then positioned into the right
atrium via the external jugular vein. This surgical procedure was done
under sodium pentobarbital anesthesia, according to the guidelines described


REFERENCES
Abraham G.E. Radioimmunoassay of steroids in biological materials. Acta
Endocrinol. (Suppl.) 183:1-42 (1974).
Allen E. and E.A. Doisy. An ovarian hormone: A preliminary report on its
localization, extraction, and partial purification, and action in test
animals. TAMA 81:819-821 (1923).
Allen L.S., M. Hines, J.E. Shryne, and R.A. Gorski. Two sexual dimorphic cell
groups in the human brain. T. Neurosci. 9:497-506 (1989).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Evidence for
the reestablishment of copulatory behavior in castrated male rats with
a brain-enhanced estradiol-chemical delivery system. Pharmacol.
Biochem. Behav. 27:265-271 (1987a).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Evidence for
suppression of serum LH without elevation in serum estradiol or
prolactin with a brain-enhanced redox delivery system for estradiol.
Life Sri. 42:1493-1502 (1988a).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Effects of a
brain-enhanced chemical delivery system for estradiol on body weight
and serum hormones in middle-aged rats. Endocr. Res. 14:131-148
(1988b).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Prolonged
suppression of androgens and androgen-dependent tissues with a
brain-enhanced delivery system for estradiol. Endoc. Soc. Seattle,
Washinton (1989).
Anderson W.R., J.W. Simpkins, P.A. Woodard, W.C. Stern, and N. Bodor.
Anxiolytic activity of a brain delivery system for GABA.
Psychopharmacology 92:157-163 (1987b).
Backstrom T., M. Bixo, and S. Hammarback. Ovarian steroid hormones. Acta
Obstet. Gynecol. Scand. [Suppl.] 130:19-24 (1985).
175


TIME (days) TIME (days)
Plasma Cone, (ng/ml) Brain Cone, (ng/g)
oo
on


45
iodide. The delivery system, E2-CDS, was then prepared by reduction of the
obtained E2-Q+ with Na2S2C>4. The structure of each intermediate and the
final product (E2-CDS) was confirmed by the nuclear magnetic resonance and
elemental analysis: mp 115-130C. The yields at each synthetic step were 64-
94%. Solutions of the E2-CDS in water containing 20% 2-hydroxy propyl-13-
cyclodextrin (wt:vol) were prepared for injection. For E2-Q+, the working
standard solutions were prepared in water/acetone (80:20; vokvol) to be used
in the in vitro methodology development.
Estradiol Pellet
Pellets, weighing 100 mg each, were prepared from crystalline E2 and
cholestrol (CHOL) powder. Both E2 and CHOL were thoroughly mixed in ratio
of 0.5% E2 and 95.5% CHOL and melted in an oil-bath (200C). Using a heated
pasteur pipette, aliquots of the homogeneous mixture were transferred into
small molds made of aluminum foil. After cooling, the solidified pellets
were unwrapped from the foil and each pellet weight was adjusted to exactly
100 mg. These pellets were used in the experiments described in Chapter 9.
Morphine Pellets
Morphine pellets were compounded in our laboratory, as previously
reported (Simpkins et al., 1983), by the method originally described by Gibson
and Tingstad (1970). Each pellet contained 75 mg morphine free base (Merk,
St. Louis, MO), 37.5 mg microcrystalline cellulose (Avisil, FMC Corporation,
Philadelphia, Pa), 0.56 mg Cab-o-sil (Cabot Corporation, Boston, MA) and 1.13
mg magnesium sterate (Fisher Chemical Co., Fair Lawn, NJ).


13
Jacobsen, 1962). Further studies demonstrating the nuclear localization of E2
receptors (Jensen & DeSombre, 1972) or increases in thymidine incorporation,
mRNA polymerase, mRNA synthesis, and protein synthesis (Hamilton, 1968)
supported the concept that the primary site of estrogen action in target tissues
is within the nuclear genome. Furthermore, voluminous body of available
evidence suggests that steroid hormones in general interact with intracellular
stereospecific receptors and upon binding the whole receptor-hormone
complex translocates into the nucleus (Walters, 1985). The hormone-receptor
complex then alters nuclear gene transcription, leading to the production of
all classes of RNA before regulating cytoplasmic protein synthesis (Walters,
1985). These actions generally occur with a delay of several hours or days
between the arrival of steroids at the target tissue and the first detectable
changes in cellular function.
Several recent contributions to the concept of gene expression by
steroid hormones have included the following: (1) demonstration of specific
DNA sequences that serves as the actual nuclear acceptor site for the steroid
receptor (Spelsberg et al., 1984); (2) evidence for the binding of the occupied
and/or transformed steroid receptors to DNA components (Gehring &
Tomkins, 1974); (3) assessment of gene transcription in purified nuclei,
including the demonstration that the estrogen receptor-induced increase in
ovalbumin mRNA transcription is not only dose-dependent but also tissue-
specific (Taylor & Smith, 1982); and (4) isolation of hormone-induced mRNA
sequences and subsequent cloning of their complimentary DNAs (cDNAs)
(O'Malley et al., 1979).
In addition to the direct genomic actions (delayed effects) of estrogens
described above, some target tissues exhibit very rapid responses to estrogen
exposure that are difficult to reconcile with the concept of direct gene


4
which result in constant exposure of the hypothalamus to endogenous
estrogens, i.e. constant exposure to illumination can also induce similar
arcuate nucleus neuropathological lesion in the rat (Brawer et al., 1983).
Conversely, experimental manipulations that essentially eliminate
circulating E2 levels (e.g. ovariectomy) greatly reduce the magnitude of the
arcuate nucleus neuropathological responses to constant illumination or
senescence in the female rat. Although there is no direct evidence as yet that
E2 is the primary neurotoxic agent responsible for the arcuate lesion, it is
noteworthy that this region of the hypothalamus is particularly rich in
estrogen receptors as well as E2-concentrating neurons (Pfaff & Keiner, 1973).
Thus, it may be that this region of the hypothalamus is exquisitely susceptible
to E2 in any concentration for prolonged period of time in the adult female
rat.
Given the aforementioned evidence for: (i) the central actions and the
therapeutic implications of estrogen hormones, and (ii) the major limiting
factors associated with the use of currently available estrogen medications, a
brain-estrogen delivery system with sustained release of estrogen in that
tissue is clearly warranted.
Over the past two decades, the attention and efforts of pharmaceutical
research were generally focused on the strategy of improving the efficacy as
well as the specificity of pharmaceutical products in order to minimize or
even abolish their adverse effects. To fulfill these objectives, novel drug
delivery systems have been designed and formulated to achieve rate-
controlled and targeted-organ delivery (Bodor, 1987; Bodor et al., 1981, 1987;
Bodor & Farag, 1983; Bodor & Simpkins, 1983). This strategy would not only
ensure a therapeutic agent preferentially gets to its intended site of action, but
it does so at the desired rate in order to satisfy the therapeutic criteria.


To my parents for their encouragement and unwavering
support toward my education, and my wife, Mahbobeh, and my son, Ehsan,
whose patience and sacrifices helped to make this work possible.


193
Comparison between estradiol valerate and ethinyl estradiol. Gynecol.
Obstet. Invest. 22:198-205 (1986).
Palmer K. and J.M. Gray. Central vs peripheral effects of estrogen on food
intake and lipoprotein lipase activity in ovariectomized rats. Physiol.
Behav. 37:187-189 (1986).
Paradisi R., S. Lodi, G. Bolelli, and S. Venturoli. Radioimmunoassay of three
oestrogens and three androgens in the same plasma sample after
extraction and chromatographic separation. Acta Endocrinol. 94:229-
234 (1980).
Pardridge W.M. Transport of nutrients and hormones through the blood-
brain barrier. Diabetologia 20:246-254 (1981).
Pardridge W.M. Neuropeptides and the blood-brain barrier. Ann. Rev.
Physiol. 45:73-82 (1983).
Pardridge W.M. Receptor-mediated peptide transport through the blood-
brain barrier. End. Rev. 7:314-330 (1986).
Pardridge W.M. Plasma protein-mediated transport of steroid and thyroid
hormones. Am. I. Physiol. 252:E157-164 (1987).
Pardridge W.M. Selective delivery of sex steroid hormones to tissues in vivo
by albumin and by sex hormone-binding golbulin. In: Steroid Protein
Interactions : Basic and Clinical Aspects. R. Fraira (ed.) 173-192, New
York: NY, Acad. Sci. (1988a).
Pardridge W.M. Recent advances in blood-brain barrier transport. Ann. Rev.
Pharmacol. Toxicol. 28:25-39 (1988b).
Pardridge W.M., J.D. Connor, I.L. Crawford. Permeability changes in the
blood-brain barrier: causes and consequences. CRC Crit. Rev. Toxic.
3:159-199 (1975).
Pardridge W..M. and L.J. Meitus. Transport of steroid hormones through the
rat blood-brain barrier. T. Clin. Invest. 64:145-154 (1979).
Paterson M.E.L. A randomized double-blind cross-over trial into the effect of
norethisterone on climateric symptoms and biochemical profiles. Br. T.
Obstet. Gynecol. 89:464-472 (1982).
Persson I. The risk of endometrial and breast cancer after estrogen treatment.
Acta Obstet. Gynecol. Scand. [Suppl.] 130:59-66 (1985).


139
suppression of circulating T levels were observed even in the face of low
plasma E2 levels, indicating the CNS involvement in mediating T
suppression. Furthermore, when the dynamics of E2-CDS effects on
circulating T levels were compared with that of an equimolar dose of E2
valerate, the E2-CDS significantly suppressed T levels while E2 valerate was
ineffective (Anderson et al., 1989). Thus, as it has been consistently
demonstrated (Anderson et al., 1987a,b, 1988a,b, 1989; Estes et al., 1987a,b, 1988;
Simpkins et al., 1986, 1988, 1989a,b), the prolonged pharmacodynamic effects
of E2-CDS following a single injection are most likely due to "locking" of the
E2-Q+ behind the BBB and there it serves as a brain depot for E2.
Although, in the present study, we used a rat model to investigate the
efficacy of E2-CDS in reducing androgen levels with subsequent regression of
the ventral prostate tissue, there are no conclusive data as to whether the rat
prostate reflects a complete picture analogous to that of the aging human
prostate. However, this model system has been employed for some years to
gain further insight into the processes involved in the initiation and
progression of prostatic hyperplasia (Pollard et al., 1989).
High-dose estrogens have also been reported to be effective in animal
models of prostatic cancer (Daehlin & Damber, 1986). Recently, it was
reported that E2 implants (producing plasma E2 levels as high as 4,442 962
pmole/liter) to tumor-(prostatic carcinoma, PC-82) bearing mice resulted in
tumor growth arrest with a subsequent decline of the tumor volume, which
equals the effect of CAST (van Steenbrugge et al., 1988). Additionally, it was
suggested that the effects of E2 on the PC-82 tumor model were mainly
indirect by suppressory effect on T secretion in the host animal, rather than a
direct effect on the tumor tissue (van Steenbrugge et al., 1988). Conversely a
number of studies demonstrated a direct action of estrogens at the cellular


2
Therapeutically, this latter central action of estrogens is of significant
interest due to the existence of several clinical conditions which are
influenced only by the presence of estrogens in the brain. For instance, after
menopause the decline in ovarian function leads to a number of central
nervous system (CNS)-mediated estrogen-withdrawal symptoms (Notelovitz,
1986). The symptoms are clearly caused by brain deprivation of estrogen
(Judd, 1983; Lauritzen, 1973,1982) since they can be alleviated by the
replacement of estrogen (Campbell & Whitehead, 1977; Upton, 1984).
Furthermore, evidence suggests that the brain is the primary locus where
estradiol (E2) exerts its effect to inhibit the secretion of luteinizing hormone
releasing hormone (LHRH) from the hypothalamus (Goodman & Knobil,
1981; Kalra & Kalra, 1980,1983,1989; Plant, 1986) and hence of LH from the
anterior pituitary and eventually of gonadal steroid hormones. As such, the
E2 hormone has been and continues to be used therapeutically for (1) fertility
regulation (Briggs, 1976; Davidson, 1969) and (2) treatment of androgen-
dependent prostatic adenocarcinoma (van Steenbrugge et al., 1988) by virtue
of suppressing plasma androgen levels (Carlstrom et al., 1989). Additionally,
estrogens are believed to act centrally to stimulate male and female sexual
behaviors (Beyer et al., 1976; Christensen & Clemens, 1974; MacLusky et al.,
1984), to regulate body weight (Palmer & Gray, 1986; Pliner & Fleming, 1983),
and may have influences on mood (Klaiber et al., 1976, 1979; Lauritzen & van
Keep, 1978; Schneider et al., 1977), and on cognitive functioning (Fillit et al.,
1986; Hackman & Galbraith, 1976).
Potential adverse effects and toxicity have, however, been associated
with the currently used estrogens. Estrogen hormones are intrinsically
lipophilic (Abraham, 1974). The high lipophilicity of these steroids ensures
their rapid penetration of biological membranes, including the blood-brain


167
chronic retention of these metabolites by the CNS are quite comparable in
male and OVX female rats. The estimated half-lives of E2-Q+ and E2 (ti /2 = 8-
9) in brain tissue in these studies are in agreement with other reports which
utilized different analytical techniques and E2-CDS doses (Mullersman et al.,
1988). Taken together these findings indicate the following: (1) as predicted
based on the physicochemical properties of the E2-CDS (Bodor et al., 1987), the
"locking of E2-Q+ in the CNS tissue had occurred and the unique features of
the BBB are the contributing factors to the chronic retention of the charged,
hydrophilic E2-Q+; (2) while the enzymatic oxidation of E2-CDS to E2-Q+ and
the hydrolysis of E2-Q+ to E2 exhibit dose-dependency, the disappearance of
these metabolites from the CNS tissue appear to be independent of dose. This
is supported by the fact that consistent results have been obtained in several
studies using doses of E2-CDS ranging from 0.01 mg/kg (Chapter 6 & 7;
Rahimy et al., 1990a, b) to 15 mg/kg dose (Mullersman et al., 1988).
Determination of distribution of E2-Q+ and E2 in peripheral tissues of
male (Chapter 5) and female rats (Chapter 6) revealed some similarities and
differences. The anterior pituitary of both male and female rats showed
slower elimination of E2-Q+ and E2 compared with other peripheral tissues.
In fact, the elimination of these metabolites from this tissue appeared more
like the CNS tissue. The concentrations of E2-Q+ and E2 in anterior pituitary
were below that of brain levels of these compounds initially and steadily
decreased throughout the observation period (Chapters 5 & 6). This relative
persistency of E2-Q+ and specially of E2 in this particular tissue are most likely
because of anatomical relationship between the hypothalamus and the
anterior pituitary gland. Estradiol released from the E2-Q+, or the E2-Q+ itself,
which is "locked" into the brain, could be delivered directly to the anterior
pituitary by the capillary plexus of the hypophyseal portal system (Traystman,


9
exhibit estrogenic activity; however, 17 P-E2 is the major and most potent
estrogen produced by the ovaries of most species including the human and
the rat. Ovaries are capable of synthesizing cholestrol de novo from acetate
and subsequently converting it to other steroids, including estrogens,
progestins, and androgens (Miller, 1988; Schwartz, 1981) Data obtained from
various enzyme kinetics and steroidal precursor-product relationships
indicate the involvement of very large number of distinct enzymes, most are
members of the cytochrome P 450 oxidases, in the conversion of cholestrol to
active steroid hormones (Miller, 1988; Ross, 1985; Schwartz, 1981). Estradiol is
formed from either androstendione or testosterone via an aromatization
reaction. This reaction is of central importance in estrogen formation, and it
is not limited to the gonads or placenta rather a wide variety of peripheral
tissues as well as the CNS can aromatize the A ring of androstendione and
testosterone to form estrogens (Canick et al., 1986; Michael et al., 1986).
The formation of estrogens is regulated by the concerted actions of two
pituitary gonadotropic hormones: follicle-stimulating hormone (FSH) and
luteinizing hormone (LH) (Richards & Hedin, 1988). FSH influences the
growth and maturation of ovarian follicles, whereas LH stimulates the
synthesis and secretion of estrogens (Richards & Hedin, 1988). The synthesis
and release of LH and FSH are, in turn, regulated by the hypothalamic
gonadotropin-releasing hormone (GnRH/LHRH) (Dalkin et al., 1989;
Davidson, 1969). Furthermore, feedback effects of E2 and other gonadal factors
on the anterior pituitary and primarily on the hypothalamus influence the
synthesis and secretion of LHRH (Kalra & Kalra, 1983, 1989; Plant, 1986; Rosie
et al., 1990).


ng/g ng/g
107
LIVER E2-Q+
LIVER E2


61
The CC for the E2-Q+ dose-RIA response was 0.98 or higher for each
tissue evaluated (Table 2; Figure 3, upper panel). Thus over the E2-Q+ dose
range tested, the method accurately estimated E2-Q+ concentrations in each of
the tissues.
Increasing brain tissue wet weight from 0.25 to 50 mg caused a highly
correlated decrease in binding of 125I-E2 in the RIA used at day 1 (CC = 0.99),
day 7 (CC = 0.99) and day 14 (CC = 0.99) after the treatment of rats with E2-CDS
(Figure 4, lower panel). The inhibition curves caused by increasing brain
tissue wet weight from animals at 1, 7 and 14 days posttreatment were parallel
and the rightward shift was indicative of the time-dependent reduction in
brain E2-Q+ concentrations (Figure 4, lower panel).
Distribution of E? and E7-O+ in vivo
Figure 5 shows brain and serum levels of E2-Q+ (upper panel) and E2
(lower panel) at various times following administration of the E2-CDS (1
mg/kg) or the HPCD vehicle (day 0 values). Brain E2 concentrations were
increased to 11.1 1.4 ng/g tissue (mean SEM) on day 1 and remained
greater than 3.5 1.1 ng/g at day 14 after drug treatment. A small amount of
E2 was detected in brains of HPCD-treated (control) male rats (0.2 0.07 ng/g),
a likely result of the aromatization of testosterone in the brain (Michael et al.,
1986). Brain concentrations of E2 exceeded serum levels of the hormone by
39-, 41- and 82-fold at 1, 7 and 14 days, respectively, after E2-CDS treatment
(Figure 5).
E2-Q+ levels were increased to 200.9 8.8 ng/g brain tissue at 1 day after
administration of the E2-CDS, and E2-Q+ levels remained elevated to 67.0
17.2 ng/g at 14 days postinjection (Figure 5). The brain to serum ratio for E2-
Q+ was 33,70 and 294 at 1,7 and 14 days, respectively.


76
1985; Thomas, 1988), cardiovascular complications (Barrett-Conner et al., 1989;
Drill & Calhoun, 1972; Inman & Vessey, 1968; Kaplan, 1978), and alterations
in hepatic metabolism (Burkman, 1988).
Since the brain is the primary site where E2 exerts its beneficial effects
on the estrogen withdrawal symptoms at the menopause (Casper & Yen, 1985;
Lauritzen, 1973; Yen, 1977), to inhibit gonadotropin secretion for fertility
regulation (Goodman & Knobil, 1981; Kalra & Kalra, 1989; Plant, 1986), to
reduce growth of peripheral steroid-dependent tissue tumors such as the
prostate (Rao et al., 1988), and to stimulate male and female sexual behaviors
(Beyer et al., 1976; Christensen & Clemens, 1974), a brain-enhanced delivery
with sustained release of E2 in that tissue is warranted. The ability to deliver
E2 preferentially to the brain, thus sparing non-target site tissues, should
improve the therapeutic index of E2 by (i) increasing the concentrations
and/or residence time of E2 at its receptor site in the brain and (ii), equally
important, decreasing the concentrations and/or residence time of E2 at the
potential peripheral sites of toxicities, thereby decreasing untoward peripheral
side effects.
Having established a reliable, specific method for the simultaneous
quantitation of E2-CDS metabolites in various tissues (Chapter 4; Rahimy et
al., 1989a), and thus to demonstrate the effectiveness of the E2-CDS, extensive
time-course studies were undertaken to evaluate the tissue distribution of E2-
Q+ and E2 in both male and female rats. The objective of this study was to
evaluate a general tissue distribution of E2-Q+ (the intermediate, oxidized
metabolite of the E2-CDS) and E2 (the active, parent steroid released upon
hydrolysis of the E2-Q+) in brain, anterior pituitary, lung, liver, kidney, heart,
fat, and plasma following a single iv dose of 1 mg /kg of E2-CDS in the male
rat.


153
Our previous studies (Simpkins et al., 1983; Simpkins & Katovich,
1984) demonstrated that implants of placebo followed by naloxone
administration as well as implants of morphine which followed by saline
injection did not induce TST surge in these treatment groups. Therefore,
these control groups were not repeated again in the present study.
Plasma LH, FSH, and PRL concentrations were measured in duplicate
by RIA using NIDDK kits. Plasma LH, FSH, and PRL values are expressed as
ng/ml of either the LH-RP-2, FSH-RP-2 or the PRL-RP-3 reference standards,
respectively. The intra-assay coefficients of variation were 4.19%, 3.32%, and
4.4% for LH, FSH, and PRL assays, respectively. Plasma E2 concentrations
were measured in duplicate by the RIA employing Coat-A-Count Estradiol
kits. All the samples for each hormone were assayed in a single run.
Results
There was no significant effect of estradiol on the basal TST (Figure 24;
Table 11) or the basal RT (Figure 25; Table 11) in the morphine-dependent
animals. However, administration of naloxone (0.5 mg/kg) to the morphine-
dependent rats resulted in a rapid increase in TST and a subsequent decline in
RT (Figures 24 & 25; Table 11). The mean maximal elevation in TST were 6.4
0.2, 6.4 0.1,3.4 0.6, and 4.9 0.5C in the HPCD, E2 pellet, multiple E2-CDS,
and single E2-CDS groups, respectively (Table 11). Multiple injections of the
E2-CDS resulted in significant attenuation (more than 47%) of the naloxone-
induced maximal rise in TST (Figure 24; Table 11). A single injection of the
E2-CDS for 7 days also attenuated the maximal rise in TST by 25%, but this
treatment effect was not statistically significant. By contrast, treatment with
17 P-E2 (E2 pellet regimen) had no effect on the surge of TST (Figure 24; Table


104
Figure 12. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in plasma of ovariectomized rats. Animals
received a single iv (tail vein) injection of the E2-CDS on day 0 at a
dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle panels), or
0.01 mg/kg (lower panels). Animals were killed 1, 7,14, 21, or 28
days after the drug administration. Plasma samples of known
aliquot were processed and assayed for E2-Q+ and E2 by the method
described in Chapter 4. Also, plasma samples from HPCD-treated
rats were analyzed for E2 hormone background. Represented are
means SEM for n = 7 rats per group per sampling time. *
Indicates below the sensitivity limit of the assay.


55
Blanks. Tissue homogenates or plasma pools were extracted to
determine residual E2 concentrations and thereby served as the estimate of
hormone background.
Hydrolysis of E?-Q+ in various tissue extracts
In preparation for base-catalyzed hydrolysis of E2-Q+, the final volume
of all E2-Q+ extracts, including the control samples, was brought to 900 ul by
the addition of 50% water:50% acetone. To each tube containing E2-Q+
solution, 100 ul of 10N NaOH were added to make the reaction medium IN
(pH >13). All tubes were vortexed and allowed to reach steady-state
equilibrium for 20 min at room temperature. Under the conditions used, a
time-course evaluation of the rate of E2-Q+ hydrolysis indicated that the
steady-state equilibrium was achieved in less than 15 min and thus, longer
incubation times did not work better. Also, it was found that the hydrolysis
of E2-Q+ under basic conditions was maximized in an aqueous/organic
solvent (50% water:50% acetone). After the hydrolysis, the pH of the reaction
medium was adjusted to a pH in the range of 6 to 8 with NaH2PC>4 and HC1.
This is an optimal pH range for the maintenance of the integrity of the C]8
columns. Samples with pH outside this range damaged the column sorbents,
and the presence of column material in the assay sample interfered with the
RIA for E2.
Solid-phase extraction and separation of E? by Ca columns
After Ci8 columns were conditioned with 2 column volumes (6 ml) of
HPLC grade methanol followed by a column volume wash (3 ml) with
distilled water, the samples to be extracted (E2 and E2-Q+ hydrolyzed extracts)
were applied to the columns. Approximately 1 to 2 min were allowed for the
column adsorption to be completed, then the columns were washed with 2


16
chromatography (Gorski et al., 1968; O'Malley et al., 1969), sucrose gradient
analysis (Jensen & DeSombre, 1973; Toft & Gorski, 1966), and sensitivity to
heat and proteases but not to nucleases (Gorski et al., 1968; Noteboom &
Gorski, 1965; O'Malley et al., 1969).
The question of the subcellular localization of the E2 receptors was then
resolved by a number of experimental criteria. These included subcellular
fractionation by differential centrifugation after 3H-E2 exposure in vivo or in
vitro (Jensen & Jacobson, 1962; King et al., 1965, Mowles et al., 1971) which
provided early evidence for nuclear localization of E2 receptors. Conversely,
E2 receptors remained in the soluble, high-speed cytosol fraction in tissues
which were not previously exposed to the E2 hormone (Jensen et al., 1968,
1973). Taken together, available data indicate that the unoccupied receptors
migrate between both the cytoplasmic and nuclear compartments but are
believed to be primarily concentrated in the nuclei in a reversible equilibrium
binding state with the nuclear components. Binding of E2 to the cytosolic
receptor results in a transient biologically active occupied transformed
receptor. Then the transformed receptor translocates into the nucleus with
an enhanced affinity for the nuclear acceptor sites, favoring receptor binding
to the acceptor site on the DNA component. This triggers gene expression
(changes in mRNA synthesis and modification) and the expression of
proteins in the target cell. These effects are generally observed with a delay of
several hours or even days between the exposure of the target tissue to E2 and
the first detectable changes in cellular function.
Membrane receptors
Numerous documented reports have emerged to indicate that the
rapid actions (from seconds to minutes of latency) of gonadal steroids on


I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
Edwin Meyer
Associate Professor of
Pharmacology and Therapeutics
I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
Kalpn Dawson ^
Associate Professor of
Pharmacodynamics
This dissertation was submitted to the Graduate Faculty of the College
of Pharmacy and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
August, 1990
Dean, College of Pharmacy
Dean, Graduate School


134
Effect of CAST or E9-CDS on Tissue Weights
The effects of CAST and E2-CDS treatment on tissue weights are shown
in Figures 20-23. CAST reduced the ventral prostate weights by more than
67% or 66% at 7 or 14 days after the orchidectomy, respectively (Figure 20).
Treatment with E2-CDS with all paradigms significantly reduced the ventral
prostate weight to CAST level at 7 days after the last injection (Figure 20). The
E2-CDS-induced regression in ventral prostate weight remained significantly
low at 14 days after the last treatment. Interestingly, the 3-injection paradigm
of E2-CDS was equivalent to CAST in reducing the ventral prostate weight by
62% at 14 days after the last injection.
Similarly, CAST significantly reduced the seminal vesicles weight by
52% or 63% at 7 or 14 days post CAST, respectively (Figure 21). Treatment of
intact animals with E2-CDS significantly reduced seminal vesicles weight by
44% to 62% and 34% to 55% at 7 and 14 days after the final treatment,
respectively (Figure 21). The 3-injection paradigm of E2-CDS was sufficient to
reduce and chronically maintain the seminal vesicles weight at CAST level
for the time-course studied.
Treatment of intact male rats with E2-CDS had no significant effect on
testes wet weight or wet weight/100 g bw (Figure 22). The 3-injection regimen
of E2-CDS caused about 19% reduction in testes weight at 7 or 14 days after the
final injection (Figure 22).
Anterior pituitary weights were not changed by CAST (Figure 23). In
contrast, pituitary weights increased significantly in a manner related to the
number of injections in response to E2-CDS administration (Figure 23).


Assayed Cone, (pg/100 mg) Assayed Cone, (pg/100 mg)
69
Figure 3. Recovery of known concentrations of E2-Q+ (upper panel) and E2
(lower panel) added to brain tissue homogenates prior to
extraction. Duplicate aliquots of homogenate were spiked with
150,300, or 600 pg E2-Q+ (upper panel) or 90,180, or 360 pg E2
(lower panel). After equilibration for 30 min followed by solvent
extraction, the spiked homogenates were centrifuged and the
supernatant was analyzed for E2-Q+ or E2 by the RIA for E2. The
results indicated that the assay method used accurately determines
E2-Q+ and E2 over a wide range of tissue concentrations.


6
following: (1) development of a reliable, sensitive, and specific method for
the simultaneous quantitation of E2-Q+ and E2, two metabolites of the E2-CDS,
in a wide variety of rat tissues; (2) determination of the tissue distributions of
these metabolites in the whole brain, hypothalamus, anterior pituitary, lung,
liver, heart, kidney, uterus, adipose tissues, and plasma of the rat; (3)
evaluation of the pharmacodynamic consequences of E2-CDS following its
administration into ovariectomized female rats; and finally (4) assessment of
the therapeutic potentials of E2-CDS in animal models for (i) the menopausal
hot flush and (ii) androgen-dependent prostatic hyperplasia.


65
E2-Q+ recovery from plasma samples was low likely because of protein
precipitation caused by hydrolysis and subsequent neutralization.
Centrifugation was needed to separate the supernatants from precipitates
before their application onto the Cis columns. E2 released during the
hydrolysis step likely interacted extensively with albumin and sex steroid
binding globulin and was unavailable for column extraction.
The separation of E2 and E2-Q+ was achieved by 3 different techniques
in the process of extraction and assay of these two products of the E2-CDS.
First, samples were divided and differentially extracted for E2 (methanol or
acetone) and for E2-Q+ (50% water:50% acetone). Although this procedure,
which depends upon the solubilization of the lipophilic E2 in methanol or
acetone and the more hydrophilic E2-Q+ in water/acetone effectively extracted
the intended steroid, separation of the two species was not complete.
However, when extracts were loaded onto the Cis column and eluted with
100% methanol, only E2 was preferentially extracted and eluted by more than
92%. Thus, the column chromatography effectively separated the two species.
Finally, the low cross-reactivity of the E2 antibody for E2-Q+ (<0.3%) ensured
that in samples extracted and chromatographed for E2, virtually no E2-Q+ was
measured. Moreover, analysis of tissue samples treated with the E2-CDS
required various dilutions for each time point which ensured the expected E2
values to fit an appropriate part of the standard curve (ED20 at 800 pg/ml to
E80 at 25 pg/ml). Indeed, dilutions which were performed prior to loading
the extracts onto the Cis column minimized nonspecific interference by E2-Q+
and lipids. Lipids decrease steroid radioimmunoassay accuracy and
reproducibility (Rash et al., 1980).


24
suggesting that the primary site of E2 action is perhaps within the MBH.
However, the stimulatory effects, or positive feedback, are observed after a
transient increase in follicular estrogen secretion from diestrus II through
proestrus or after more than 48 hours of sustained E2 exposure to
ovariectomized rats (Kalra & Kalra, 1983; 1989; Legan et al., 1975). The
positive feedback action of E2 in the estrous cycle is believed to be exerted
primarily on the medial preoptic area of the ventral diencephalon since
lesioning of this area (Wiegand et al., 1980), or interuption of its connections
to the MBH (Halasz & Gorski, 1967), or implantation of antiestrogen
clomiphene into this region (Docke et al., 1989) completely prevent the LH
surge and ovulation in the rat. The preovulatory rise of E2 increases the
hypothalamic LHRH mRNA (Rosie et al., 1990), the LHRH pulse frequency
and pulse amplitude at mid-cycle (Dalkin et al., 1989; Marshal & Kelch, 1986).
However, at the pituitary level E2 increases the responsiveness of the
gonadotrophs to LHRH on the afternoon of proestrus. This allows a priming
effect and thus augments the response of the pituitary to subsequent LHRH
messages. The increase in pituitary responsiveness must precede the
hypothalamic LHRH message for the LH surge to occur.


150
circuit, in particular the opioid neurons/receptors component, is uncoupled
from the LH release mechanism (Casper & Yen, 1985; Simpkins & Katovich,
1984). Hence, replacement of gonadal steroids in postmenopausal women
would exert a stabilizing influence on these neuronal network or the
mechanisms responsible for the flush response. Nevertheless, the
mechanism(s) through which gonadal steroids exert their stabilizing effects
remains unknown.
Although replacement therapy with estrogens and/or progestins has
been shown to be effective in most patients in alleviating the symptoms of
the menopause (Campbell & Whitehead,1977; Casper & Yen, 1985; Lauritzen,
1973; Upton, 1984; Yen, 1977), numerous retrospective studies indicated an
increased risk of peripheral toxicities, including the risk of breast and
endometrial cancer (Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger et al.,
1988; Persson, 1985), cardiovascular complications (Barrett-Connor et al., 1989;
Kaplan, 1978; Thomas, 1988), and alteration in hepatic metabolism (Burkman,
1988). The potential problem associated with currently used steroids
medication is that these hormones equilibrate among all body tissues due to
their high lipophilicity. As a result, only a fraction of the dose accumulates at
or near the site of action in the brain. Furthermore, steroid receptors are
present in many peripheral tissues creating the potential of untoward
peripheral site effects (Walters, 1985).
Given the aforementioned limiting factors in estrogen replacement
therapy, the preferential brain delivery of E2 with the E2-CDS may offer an
effective treatment strategy for postmenopausal symptoms by providing
sufficient E2 to the brain while avoiding peripheral toxicities.
This study was undertaken to assess the effects of E2-CDS on the rise in
tail-skin temperature (TST) in the morphine-dependent, naloxone-


ASSAY DEVELOPMENT, TISSUE DISTRIBUTION
AND PHARMACODYNAMICS OF A NOVEL ESTROGEN-CHEMICAL
DELIVERY SYSTEM FOR THE BRAIN
BY
MOHAMAD H. RAHIMY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1990

To my parents for their encouragement and unwavering
support toward my education, and my wife, Mahbobeh, and my son, Ehsan,
whose patience and sacrifices helped to make this work possible.

ACKNOWLEDGEMENTS
This work would never have come to fruition without the
encouragement, assistance and advice of many individuals whom I am very
grateful to. First, I wish to express my sincere appreciation and gratitude to
my mentor, Dr. James W. Simpkins, for his expert guidance, encouragement,
and support. Throughout my graduate study at the University of Florida, I
had ample opportunity to learn by experience under the skillful guidance of
Dr. Simpkins. I also wish to express great thanks to the other members of my
committee, Dr. Nicholas Bodor, Dr. William Millard, Dr. Edwin Meyer, and
Dr. Ralph Dawson, who have imparted valuable advice as well as their
critical evaluation of my work. I would also like to extend thanks to Dr.
Michael Meldrum, Dr. Michael Katovich, Dr. Wesley Anderson, and Dr.
Anna Ratka for their advice and assistance.
I would like to thank the many others who contributed their time and
efforts, especially Victoria Red Patterson, Becky Hamilton, Lee Glancey, Terry
Romano, Debby Andreadis, Billie Jean Goins, Roxane Federline, and Denise
Blake, who assisted me in various aspects of this work. The assistance of
Anup Zutshi regarding the kinetic analysis is greatly appreciated. My
personal thanks go to Dr. Lee Ann Burgland and Dave Wallace whose
cooperations during these many years made graduate school more bearable. I
extend thanks to new graduate students Singh Meharvan, Melanie King,
Melanie Pecins, and Jean Bishop-Sparks who have already taken over the
reins in the lab for accepting the challenge.

Finally, very special thanks go to my parents who have always inspired
me to pursue an academic career, and to my supportive and considerate
family (my wife & my son) who always managed to create an environment in
which I could devote the many years required to accomplish this work.
iv

TABLE OF CONTENTS
ACKNOWLEDGEMENTS iii
LIST OF TABLES viii
LIST OF FIGURES ix
ABSTRACT xi
CHAPTERS
1. INTRODUCTION 1
2. REVIEW OF THE LITERATURE 7
Estrogen Hormones 7
Historical Observations 7
Endocrinological and Biochemical Considerations 8
Biosynthesis 8
Secretion and transport 10
Metabolism and excretion 11
Mechanism of action 12
Estrogen Receptors 15
Intracellular/cytosolic receptors 15
Membrane receptors 16
Estrogen-receptor binding kinetics 17
Role of Estrogen in the Menstrual Cycle 19
Feedback regulation 20
Role of Estrogen in the Rat Estrous Cycle 22
Feedback regulation 23
Blood-Brain Barrier 25
Historical Overview 25
Potential Asset to Utilize in the Design of Brain-Specific Drug
Delivery 28
Therapies Aimed at Targeting/Enhancing Brain Estradiol Levels 32
Fertility Regulation 32
Menopausal Syndrome 33
Prostatic Cancer 34
Body Weight Regulation 35
Libido / Sexual Dysfunction 36
Disorders of Depression 37
Cognitive Impairment of Menopausal Alzheimer's Type 41
v

3. GENERAL MATERIALS AND METHODS 44
Drugs and Solutions 44
Estradiol and Standard Solution 44
Estradiol-Chemical Delivery System 44
Estradiol Pellet 45
Morphine Pellets 45
Animals 46
Drug Treatment 47
Steroid Treatment 47
Morphine and Naloxone Treatment 48
Plasma Hormones Radioimmunoassays 48
Protein Hormone Assays 48
Steroid Hormone Assays 49
Statistical Analysis 50
4. DEVELOPMENT OF AN ANALYTICAL METHOD FOR THE
QUANTITATION OF E2-CDS METABOLITES IN A WIDE
VARIETY OF TISSUES IN THE RAT 51
Introduction 51
Materials and Methods 52
In Vitro Methodology 53
Specificity of the estradiol antibody for E2 53
Selective solvent extraction of steroids from tissues 53
Hydrolysis of E2-Q+ in various tissue extracts 55
Solid-phase extraction and separation of E2 by Cis columns 55
Radioimmunoassay of E2 56
Calculations 56
In Vivo Studies 57
Results 58
In Vitro Methodology 58
Cross-reactivity of the E2 antibody with E2-Q+ 58
Recovery of E2 58
Precision of the E2 extraction-assay method 59
Recovery of E2-Q+ 60
Precision of the E2-Q+ extraction-assay method 60
Distribution of E2 and E2-Q+ in vivo 61
Discussion 62
5. DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN MALE RATS 75
Introduction 75
Materials and Methods 77
Results 78
Discussion 80
VI

6. DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN FEMALE RATS 90
Introduction 90
Materials and Methods 91
Experiment 1 91
Experiment 2 92
Results 94
Experiment 1 94
Experiment 2 96
Discussion 96
7. EVALUATION OF THE PHARMACODYNAMIC EFFECTS OF E2-
CDS IN OVARIECTOMIZED FEMALE RATS 115
Introduction 115
Materials and Methods 116
Results 117
Discussion 119
8. EFFECTS OF THE E2-CDS OR CASTRATION ON ANDROGEN
AND ANDROGEN-DEPENDENT TISSUES IN MALE RATS 128
Introduction 128
Materials and Methods 131
Results 133
Effect of CAST or E2-CDS on Plasma T Levels 133
Effect of CAST or E2-CDS on Tissue Weights 134
Effect of CAST or E2-CDS on Plasma Hormones 135
Discussion 135
9. EFFECTS OF THE E2-CDS OR E2 PELLET ON TAIL-SKIN
TEMPERATURE RESPONSES IN OVARIECTOMIZED FEMALE
RATS 148
Introduction 148
Materials and Methods 151
Results 153
Discussion 155
10. GENERAL DISCUSSION 163
REFERENCES 175
BIOGRAPHICAL SKETCH 202
vii

LIST OF TABLES
Table Page
1. Recovery and Precision Determinations for Biological Samples
Spiked with E2 72
2. Recovery and Precision Determinations for Biological Samples
Spiked with E2-Q+ 73
3. Percent Hydrolysis of E2-Q+ in Supernatants of a Variety of
Tissues 74
4. Effects of Dose on the Extent of Oxidation and Hydrolysis of E2-
CDS in a Variety of Tissues in vivo Ill
5. Effects of the E2-CDS on the Clearance of E2-Q+ from a Variety of
Tissues 112
6. Effects of the E2-CDS on the Clearance of E2 from a Variety of
Tissues 113
7. Effects of an Equimolar Dose of E2 on the Tissue Concentrations
of E2 114
8. Dose and Time-Dependent Effects of the E2-CDS on Peripheral
Tissue Weights in Ovariectomized Rats 127
9. Effects of the E2-CDS or CAST on Plasma Hormone
Concentrations at 7 Days after the Last Treatment in Male Rats 146
10. Effects of the E2-CDS or CAST on Plasma Hormone
Concentrations at 14 Days after the Last Treatment in Male Rats 147
11. Effects of the E2-CDS or E2 Pellet on Basal Temperature, Maximal
Change in TST, and Area Under the 90 Min TST Curve in
Ovariectomized, Morphine-Dependent Rats 161
12. Effects of the E2-CDS or E2 Pellet on Plasma Hormone
Concentrations in Ovariectomized, Morphine-Dependent Rats 162
viii

LIST OF FIGURES
Figure Page
1. Schematic representation of in vitro synthesis and in vivo
transformation of the estradiol-chemical delivery system (E2-
CDS) 67
2. Inhibition of 125I-E2 binding to an E2 antibody caused by E2 or E2-
Q+ 68
3. Recovery of known concentrations of E2-Q+ and E2 added to
brain tissue homogenates prior to extraction 69
4. Inhibition of 1251-E2 binding to an E2 antibody caused by
increasing amounts of brain tissue from rats treated with the
estradiol-chemical delivery system 70
5. Effects of a single iv dose of the estradiol-chemical delivery
system on serum and brain concentrations of E2-Q+ and E2 or the
HPCD vehicle 71
6. Effects of a single iv dose of the E2-CDS on brain and plasma
concentrations of E2-Q+ and E2 in intact male rats 84
7. Effects of a single iv dose of the E2-CDS on liver and fat
concentrations of E2-Q+ and E2 in intact male rats 86
8. Effects of a single iv dose of the E2-CDS on kidney, heart, lung,
and anterior pituitary concentrations of E2-Q+ and E2 in intact
male rats 87
9. Brain and anterior pituitary contents of the E2-Q+ and E2
following a single iv dose of the E2-CDS 88
10. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in whole brain of ovariectomized rats 100
11. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in hypothalamus of ovariectomized rats 102
ix

12. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in plasma of ovariectomized rats 104
13. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in liver of ovariectomized rats 106
14. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in fat of ovariectomized rats 108
15. Effects of a single iv dose of the E2-CDS on plasma E2-Q+ or
plasma E2 concentrations in ovariectomized rats 110
16. Dose and time-dependent effects of the E2-CDS on plasma LH
responses in ovariectomized rats 124
17. Dose and time-dependent effects of the E2-CDS on plasma FSH
responses in ovariectomized rats 125
18. Dose and time-dependent effects of the E2-CDS on plasma PRL
and GH responses in ovariectomized rats 126
19. Effects of the E2-CDS or CAST on plasma testosterone
concentrations either 7 days or 14 days after the final treatment 141
20. Effects of the E2-CDS or CAST on ventral prostate weight either 7
days or 14 days after the final treatment 142
21. Effects of the E2-CDS or CAST on seminal vesicle weight either 7
days or 14 days after the final treatment 143
22. Effects of the E2-CDS or CAST on testis weight either 7 days or 14
days after the final treatment 144
23. Effects of the E2-CDS or CAST on anterior pituitary weight either
7 days or 14 days after the final treatment 145
24. Effects of the E2-CDS or E2 pellet on the mean TST responses
induced by naloxone administration to morphine-dependent,
ovariectomized rats 159
25. Effects of the E2-CDS or E2 pellet on the mean RT responses
induced by naloxone administration to morphine-dependent,
ovariectomized rats 160
X

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ASSAY DEVELOPMENT, TISSUE DISTRIBUTION
AND PHARMACODYNAMICS OF A NOVEL ESTROGEN-CHEMICAL
DELIVERY SYSTEM FOR THE BRAIN
By
Mohamad H. Rahimy
August, 1990
Chairman: Dr. James W. Simpkins
Major Department: Pharmacodynamics
Enhanced delivery and sustained release of estradiol (E2) in the brain is
desirable for effective treatments of the menopausal hot flush, prostatic
adenocarcinoma, and fertility regulation. Our studies thus evaluated an E2-
chemical delivery system (E2-CDS) for the brain, which is based upon the
interconvertible dihydropyridine <=> pyridinium ion redox reaction. The E2-
CDS requires multiple, facile chemical conversions, including the oxidation
of E2-CDS to the corresponding quaternary ion (E2-Q+), which provides the
basis of locking the molecule within the brain, and the subsequent slow
hydrolysis of E2-Q+ by esterases to E2 in that tissue.
Initially, studies were undertaken to develop a reliable, specific, and
sensitive method to simultaneously measure E2-Q+ and E2 (two metabolites
xi

of E2-CDS) in various biological tissues. This method utilized the following
steps: (1) selective solvent extraction of E2-Q+ and E2 from the tissues; (2) base-
catalyzed hydrolysis of E2-Q+ to E2 in NaOH; (3) solid-phase purification of E2
with Ci8 reversed-phase extraction columns; and (4) radioimmunoassay of E2.
Subsequently, the in vivo tissue distributions of E2-Q+ and E2 were
determined in both male and female rats. The results revealed that the
disappearance of E2-Q+ as well as E2 was slow in brain tissue with a tj /2 = 8-9
days. By contrast, both of these metabolites exhibited relatively rapid
clearance from the plasma, liver, lung, kidney, heart, fat, and uterus.
After documenting the kinetic behaviors of E2-CDS, time-course
studies were then conducted to assess the dynamic effects of E2-CDS on
responses which are known to be affected by E2. The E2-CDS consistently
exhibited prolonged and sustained suppression of pituitary gonadotropins
secretion, i.e. LH and FSH in a dose- and time-dependent manner.
Finally, the therapeutic potentials of E2-CDS were investigated in male
and female rats. Studies in the male rat demonstrated that E2-CDS is as
effective as castration in both suppressing the plasma testosterone levels and
reducing the weights of androgen-responsive tissues. Further studies in the
female rat, examining the effects of E2-CDS on tail-skin temperature (TST)
responses, revealed that E2-CDS can significantly attenuate the rise in TST.
Collectively, the results of these studies are consistent with the
proposed mechanism of this drug delivery system, that is, the preferential
retention of E2-Q+ by the brain, and the subsequent slow release of E2 locally in
that tissue. Furthermore, the profound pharmacodynamic effects of this
delivery system support the view that E2-CDS may be potentially useful for
fertility regulation, the effective treatments of prostatic cancer, and certain
brain-mediated estrogen withdrawal symptoms, i.e. menopausal hot flushes.
xii

CHAPTER 1
INTRODUCTION
Estrogens exhibit a myriad of important regulatory roles in the growth,
development, and maintenance of the structures and functions which are
necessary for the continuation of the species. Their therapeutic applications
for certain clinical problems have been appreciated since the turn of the
century, when ovarian grafts were shown to prevent uterine atrophy and loss
of sexual function in castrated animals (Knauer, 1900). Estrogen hormones
have broad therapeutic applications and in most cases the steroids are used
primarily for their central actions (Meites & Nicoll, 1965). Among these are
the reproductive-related applications, including fertility regulation, sexual
dysfunction, and the replacement therapy in postmenopausal patients; and
the non-reproductive applications, including treatment of postmenopausal
depression and cancer therapy. Nevertheless, the full spectrum of potential
clinical benefits and applications of estrogen therapy has yet to be uncovered.
Physiologically, estrogen hormones exert two modes of action on the
central nervous system (CNS), particularly on the brain. First, during the
critical period of fetal /neonatal life, estrogens affect permanently some
features of the brain structure and function which result in neuronal
differentiation (Allen et al., 1989; Goy & McEwen, 1980). Second, during the
course of adult life, these hormones exert their effects in a modulatory and
reversible mode that influence a myriad of adult brain functions (McEwen,
1988; McEwen & Parsons, 1982).
l

2
Therapeutically, this latter central action of estrogens is of significant
interest due to the existence of several clinical conditions which are
influenced only by the presence of estrogens in the brain. For instance, after
menopause the decline in ovarian function leads to a number of central
nervous system (CNS)-mediated estrogen-withdrawal symptoms (Notelovitz,
1986). The symptoms are clearly caused by brain deprivation of estrogen
(Judd, 1983; Lauritzen, 1973,1982) since they can be alleviated by the
replacement of estrogen (Campbell & Whitehead, 1977; Upton, 1984).
Furthermore, evidence suggests that the brain is the primary locus where
estradiol (E2) exerts its effect to inhibit the secretion of luteinizing hormone
releasing hormone (LHRH) from the hypothalamus (Goodman & Knobil,
1981; Kalra & Kalra, 1980,1983,1989; Plant, 1986) and hence of LH from the
anterior pituitary and eventually of gonadal steroid hormones. As such, the
E2 hormone has been and continues to be used therapeutically for (1) fertility
regulation (Briggs, 1976; Davidson, 1969) and (2) treatment of androgen-
dependent prostatic adenocarcinoma (van Steenbrugge et al., 1988) by virtue
of suppressing plasma androgen levels (Carlstrom et al., 1989). Additionally,
estrogens are believed to act centrally to stimulate male and female sexual
behaviors (Beyer et al., 1976; Christensen & Clemens, 1974; MacLusky et al.,
1984), to regulate body weight (Palmer & Gray, 1986; Pliner & Fleming, 1983),
and may have influences on mood (Klaiber et al., 1976, 1979; Lauritzen & van
Keep, 1978; Schneider et al., 1977), and on cognitive functioning (Fillit et al.,
1986; Hackman & Galbraith, 1976).
Potential adverse effects and toxicity have, however, been associated
with the currently used estrogens. Estrogen hormones are intrinsically
lipophilic (Abraham, 1974). The high lipophilicity of these steroids ensures
their rapid penetration of biological membranes, including the blood-brain

3
barrier (BBB), thus enabling access to all cells and organs. Indeed, when these
hormones are used therapeutically to specifically target the CNS, the steroids
equilibrate among all body tissues due to their high lipophilicity (Pardridge &
Meitus, 1979). Moreover, when inside the CNS, there is no mechanism to
prevent their redistribution back to the periphery as blood levels of the
steroids decline (Davson, 1976; Schanker, 1965). So, even if estrogens can
easily gain access to the CNS, they are poorly retained by the brain. As a
result, only a fraction of the administered estrogen dose accumulates at or
near the site of action in the brain. This property of the estrogens necessitates
either frequent dosing or the administration of a depot form of the estrogen
in order to maintain therapeutically effective concentrations in the brain
(Schanker, 1965; Spona & Schneider, 1977). Both of these treatment strategies
lead to sustained increases in peripheral estrogen levels. Since estrogen
receptors are present in many peripheral tissues (Walters, 1985), where they
mediate a myriad of physiological and pharmacological effects (Murad &
Haynes, 1985), it further creates the potential of untoward peripheral
toxicities. In fact, constant increases in peripheral tissue exposure to estrogens
have been shown in numerous retrospective studies to precipitate various
peripheral toxicities, including increased risk of breast and endometrial cancer
(Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger et al., 1988; Persson, 1985;
Thomas, 1988), cardiovascular complications (Barrett-Conner et al., 1989;
Inman & Vessey, 1968; Kaplan, 1978; Thomas, 1988), and marked interference
with hepatic metabolism (Burkman, 1988).
In addition to the peripheral toxicities mentioned above, constant
exposure to high levels of E2 valerate has been shown to induce neuronal
degeneration in the hypothalamic arcuate nucleus of both male and female
rats (Brawer et al., 1980, 1983). Furthermore, other experimental conditions

4
which result in constant exposure of the hypothalamus to endogenous
estrogens, i.e. constant exposure to illumination can also induce similar
arcuate nucleus neuropathological lesion in the rat (Brawer et al., 1983).
Conversely, experimental manipulations that essentially eliminate
circulating E2 levels (e.g. ovariectomy) greatly reduce the magnitude of the
arcuate nucleus neuropathological responses to constant illumination or
senescence in the female rat. Although there is no direct evidence as yet that
E2 is the primary neurotoxic agent responsible for the arcuate lesion, it is
noteworthy that this region of the hypothalamus is particularly rich in
estrogen receptors as well as E2-concentrating neurons (Pfaff & Keiner, 1973).
Thus, it may be that this region of the hypothalamus is exquisitely susceptible
to E2 in any concentration for prolonged period of time in the adult female
rat.
Given the aforementioned evidence for: (i) the central actions and the
therapeutic implications of estrogen hormones, and (ii) the major limiting
factors associated with the use of currently available estrogen medications, a
brain-estrogen delivery system with sustained release of estrogen in that
tissue is clearly warranted.
Over the past two decades, the attention and efforts of pharmaceutical
research were generally focused on the strategy of improving the efficacy as
well as the specificity of pharmaceutical products in order to minimize or
even abolish their adverse effects. To fulfill these objectives, novel drug
delivery systems have been designed and formulated to achieve rate-
controlled and targeted-organ delivery (Bodor, 1987; Bodor et al., 1981, 1987;
Bodor & Farag, 1983; Bodor & Simpkins, 1983). This strategy would not only
ensure a therapeutic agent preferentially gets to its intended site of action, but
it does so at the desired rate in order to satisfy the therapeutic criteria.

5
A remarkable example is the design of an estradiol-chemical delivery
system (E2-CDS) for the enhanced and sustained release of E2 in the brain
(Bodor et al., 1987). The E2-CDS exploits the unique architecture of the BBB,
which normally excludes a variety of pharmacological agents from the CNS
due to their physicochemical properties (Bodor & Brewster, 1983). The E2-
CDS is a redox-based chemical-delivery system and the mechanism of its drug
delivery is based upon an interconvertible dihydropyridine <=> pyridinium
ion redox carrier (Bodor et al., 1987). After systemic administration of the E2-
CDS, it distributes throughout the body, then, the carrier moiety is quickly
oxidized to the corresponding quaternary pyridinium ion (E2-Q+) in the brain
as well as in peripheral tissues. The charged pyridinium-drug complex is
thus locked into the CNS while the same moiety rapidly clears from the
periphery because of a 40,000-fold increase in its hydrophilicity. Sustained
release of the active, parent drug from the charged pyridinium-drug complex
occurs in the brain as a result of enzymatic hydrolysis of the ester linkage.
The enzymes involved in cleavage of the ester bond are believed to be non
specific esterases.
Collectively, the ability to preferentially deliver E2 to the brain, thus
sparing non-target peripheral site tissues, should improve the therapeutic
index of E2 by (i) increasing the concentrations and/or residence time of E2 at
its receptor site in the brain and (ii), equally important, decreasing the
concentrations and/or residence time of E2 at the potential peripheral sites of
toxicities, thereby decreasing untoward peripheral side effects.
To document the predictive biotransformation behaviors of the E2-CDS
(Bodor et al., 1987), and to further substantiate its effectiveness over the
currently used estrogens, extensive and long-term pharmacokinetic and
pharmacodynamic studies were conducted. These studies included the

6
following: (1) development of a reliable, sensitive, and specific method for
the simultaneous quantitation of E2-Q+ and E2, two metabolites of the E2-CDS,
in a wide variety of rat tissues; (2) determination of the tissue distributions of
these metabolites in the whole brain, hypothalamus, anterior pituitary, lung,
liver, heart, kidney, uterus, adipose tissues, and plasma of the rat; (3)
evaluation of the pharmacodynamic consequences of E2-CDS following its
administration into ovariectomized female rats; and finally (4) assessment of
the therapeutic potentials of E2-CDS in animal models for (i) the menopausal
hot flush and (ii) androgen-dependent prostatic hyperplasia.

CHAPTER 2
REVIEW OF THE LITERATURE
This chapter will first present a historical review with respect to the
endocrinology/neuroendocrinology of estrogen hormones. This will include
some evidence pertinent to their physiological/pharmacological actions in
the central nervous system (CNS). Furthermore, since the unique
architecture of the brain, the blood-brain barrier (BBB), is of central asset in
the design and synthesis of the estrogen-delivery system under investigation,
a historical account of the BBB will be discussed. Attempts will be made to
identify problems associated with the brain delivery of existing drugs. Finally,
example of certain clinical conditions which require the presence of estrogen
in the brain as a therapeutic agent will be discussed as well. The purpose of
this diverse literature review is to identify and describe the concepts and
rationale which were the basis in the design and synthesis of the E2-CDS,
which will be evaluated in detail in later chapters.
Estrogen Hormones
Historical Observations
Ovarian endocrine activity was first demonstrated experimentally by
Knauer in 1896 (quoted by Tepperman, 1981). Independently, Sobotta (1896)
described the origin of corpus luteum at the same time. Shortly thereafter,
Beard (1897) postulated that the corpus luteum might serve a necessary
function during pregnancy. The observation by Knauer (1900), who
7

8
demonstrated that ovarian transplants prevented uterine atrophy and loss of
sexual function in castrated animals, established the hormonal nature of
ovarian control of the female reproductive system. Further supporting
evidence was provided by Fraenkel (1903), who showed that destruction of
the corpora ltea in pregnant rabbits causes abortion. In 1923, Allen and
Doisy developed a simple, quantitative bioassay for ovarian extracts based
upon changes produced in the vaginal smear of the rat. Two years later,
Loewe (1925) reported on a female sex hormone in the blood of various
species. Shortly thereafter, Loewe and Lange (1926) discovered a female sex
hormone in the urine of menstruating women with the observation that the
concentration of the hormone in the urine varied with the phases of the
menstrual cycle. These observations set the stage for chemists, who soon
isolated independently an active estrogen substance from urine in crystalline
form (Butenandt, 1929; Doisy et al., 1929,1930). In 1935, Doisy et al. (quoted by
Tepperman, 1981) characterized the chemical structure of estradiol-17|3.
However, it was the contributions of Corner and Allen (1929) that firmly
established the endocrine function of the corpus luteum. They clearly
demonstrated that the abortion following extirpation of the corpora ltea in
pregnant rabbits can be prevented by the injection of luteal extracts.
Endocrinological and Biochemical Considerations
Biosynthesis
Estrogens are primarily produced by the follicles and corpus luteum of
the ovary and by the placenta during the second and third trimesters of
pregnancy. Ovaries secrete estradiol (E2) and estrone, whereas the placenta
produces these and estriol (Ross, 1985; Schwartz, 1981). All these hormones

9
exhibit estrogenic activity; however, 17 P-E2 is the major and most potent
estrogen produced by the ovaries of most species including the human and
the rat. Ovaries are capable of synthesizing cholestrol de novo from acetate
and subsequently converting it to other steroids, including estrogens,
progestins, and androgens (Miller, 1988; Schwartz, 1981) Data obtained from
various enzyme kinetics and steroidal precursor-product relationships
indicate the involvement of very large number of distinct enzymes, most are
members of the cytochrome P 450 oxidases, in the conversion of cholestrol to
active steroid hormones (Miller, 1988; Ross, 1985; Schwartz, 1981). Estradiol is
formed from either androstendione or testosterone via an aromatization
reaction. This reaction is of central importance in estrogen formation, and it
is not limited to the gonads or placenta rather a wide variety of peripheral
tissues as well as the CNS can aromatize the A ring of androstendione and
testosterone to form estrogens (Canick et al., 1986; Michael et al., 1986).
The formation of estrogens is regulated by the concerted actions of two
pituitary gonadotropic hormones: follicle-stimulating hormone (FSH) and
luteinizing hormone (LH) (Richards & Hedin, 1988). FSH influences the
growth and maturation of ovarian follicles, whereas LH stimulates the
synthesis and secretion of estrogens (Richards & Hedin, 1988). The synthesis
and release of LH and FSH are, in turn, regulated by the hypothalamic
gonadotropin-releasing hormone (GnRH/LHRH) (Dalkin et al., 1989;
Davidson, 1969). Furthermore, feedback effects of E2 and other gonadal factors
on the anterior pituitary and primarily on the hypothalamus influence the
synthesis and secretion of LHRH (Kalra & Kalra, 1983, 1989; Plant, 1986; Rosie
et al., 1990).

10
Secretion and transport
The E2-producing cells in the ovary and corpus luteum do not
characteristically prepackage large amounts of steroid hormone for release.
Rather, these endocrine tissues store the hormone precursor, cholesterol-
ester, in the form of lipid droplets inside the hormone-producing cells
(Rossmanith et al., 1990). This indicates that, in these secretory cells, the
signal for E2 release is perhaps tightly coupled to that of estrogen-hormone
synthesis. Thus, the newly synthesized hormone will be released into the
circulation for transport to target tissues. The signals for release are primarily
those of the anterior pituitary tropic factors (LH & FSH). It is believed that E2
is released in pulsatile fashion and this perhaps is the result of the episodic
modulating influence of LH (Rossmanith et al., 1990).
Estradiol hormone, like other steroid hormones, when secreted into
the blood, it is primarily transported by carrier proteins. E2 may be
transported by: (a) plasma albumin (60%) with low affinity binding; (b) sex
hormone-binding globulin (38%) with high affinity binding; and (c) in free
(dialyzable) form (2%) (Moutsatsou & Oakey, 1988; Pardridge, 1988a). The
carrier-bound E2 is biologically inactive and sequestered in plasma while only
the free fraction is regarded as biologically active hormone. Recent studies
have, however, suggested that the carrier-bound pool of E2 may also be
available for uptake by the target tissues (Pardridge, 1988a). It was suggested
that there are two possible mechanisms for the delivery of carrier-bound E2 to
target tissues (Pardridge, 1988a). One mechanism involves interactions
between the carrier protein surface and the surface of the organ
microcirculation that results in a conformational change about the carrier
binding site and thus enhanced dissociation of E2 hormone. The second

11
mechanism involves a receptor-mediated transcytosis of carrier-bound
hormone complex in the limiting membrane of the organ microcirculation.
However, other investigators have argued against this hypothesis on the
grounds that the concept does not readily reconcile with physiological
findings (Mendel et al., 1988). That is, the rate of protein-bound hormone
dissociation (K hormone hypothesis. Thus, a more comprehensive model (equation) which
formally takes into account the rate-limiting effects of protein-bound
hormone dissociation is more relevant to the experimental observations.
The current Pardridge's model (the protein-bound hormone hypothesis) is,
however, deficient in this respect.
Metabolism and excretion
The plasma concentration of E2 at any time represents the net
difference between the rate of E2 secretion and the rate of metabolism in the
liver and excretion by the kidneys. There is no apparent limit to the capacity
of these organs to metabolize and excrete the E2 hormone. The liver is the
primary organ for metabolizing E2 hormone (Bolt, 1979). The rate of
turnover of E2 hormone is rather rapid. It has a half-life of about 90 min (von
Schoultz et al., 1989). The estrogen is oxidized by the action of a stereospecific
dehydrogenase enzyme, using pyridine nucleotides as cofactors, to less active
products such as estrone and estriol. The oxidized metabolites are then
conjugated as sulfates or glucuronides. The conjugation process renders these
metabolites highly water soluble which then are quickly excreted in the urine
or bile (Bolt, 1979). These conjugates are biologically inactive; however, the
biliary metabolites may undergo further metabolism by action of the
intestinal flora. The products are then reabsorbed into the portal circulation

12
and transported to the liver, a process called enterohepatic circulation (Bolt,
1979). A metabolite of E2, which comprises at least 20% of the total amount
secreted in humans, is the 2-hydroxyl derivative. These metabolites, referred
to as catechol estrogens, are shown to have biological activity. The biological
activity of the catechol estrogen appears to involve an interaction with
catecholamine synthesis, receptors or effectors (Weisz & Crowley, 1986). The
conversion to catechol estrogen can occur in a number of tissues, including
the CNS (Weisz & Crowley, 1986).
Mechanism of action
Based on the gross anatomical, histological, and biochemical evidence,
E2 is shown to have growth-promoting activities on cells of the target organs
such as the anterior pituitary, uterus, vagina, Graafian follicles of the ovary
and the mammary gland by increasing protein synthesis and mitotic activity.
As early as 1953, Szego and Roberts, seeking an understanding of the
mechanism of E2 action, demonstrated accumulation of ribonucleic acid
(RNA) and protein in estrogen-stimulated uterine tissues. Mueller et al.
(1958) showed that most of the E2 effects on RNA and protein synthesis can be
blocked by a translation inhibitor (puromycin) and a transcription inhibitor
(actinomycin D). These observations led to the proposal that estrogen
hormones and steroids in general stimulate or activate the production of
nucleic acid templates (mRNAs) and, hence, gene expression (Mueller et al.,
1958). Soon after the technological advances of the 1960's and, thus, the
availability of tritium-labeled estradiol (3H-E2), Jensen and Jacobsen (1962)
discovered that the estrogen target tissues (uterus and vagina) selectively
concentrated the labeled E2. These investigators were also the first to
demonstrate the binding of E2 to a specific cytosolic receptor protein (Jensen &

13
Jacobsen, 1962). Further studies demonstrating the nuclear localization of E2
receptors (Jensen & DeSombre, 1972) or increases in thymidine incorporation,
mRNA polymerase, mRNA synthesis, and protein synthesis (Hamilton, 1968)
supported the concept that the primary site of estrogen action in target tissues
is within the nuclear genome. Furthermore, voluminous body of available
evidence suggests that steroid hormones in general interact with intracellular
stereospecific receptors and upon binding the whole receptor-hormone
complex translocates into the nucleus (Walters, 1985). The hormone-receptor
complex then alters nuclear gene transcription, leading to the production of
all classes of RNA before regulating cytoplasmic protein synthesis (Walters,
1985). These actions generally occur with a delay of several hours or days
between the arrival of steroids at the target tissue and the first detectable
changes in cellular function.
Several recent contributions to the concept of gene expression by
steroid hormones have included the following: (1) demonstration of specific
DNA sequences that serves as the actual nuclear acceptor site for the steroid
receptor (Spelsberg et al., 1984); (2) evidence for the binding of the occupied
and/or transformed steroid receptors to DNA components (Gehring &
Tomkins, 1974); (3) assessment of gene transcription in purified nuclei,
including the demonstration that the estrogen receptor-induced increase in
ovalbumin mRNA transcription is not only dose-dependent but also tissue-
specific (Taylor & Smith, 1982); and (4) isolation of hormone-induced mRNA
sequences and subsequent cloning of their complimentary DNAs (cDNAs)
(O'Malley et al., 1979).
In addition to the direct genomic actions (delayed effects) of estrogens
described above, some target tissues exhibit very rapid responses to estrogen
exposure that are difficult to reconcile with the concept of direct gene

14
regulation. For instance, the CNS neurones produce diversity of both rapid
and delayed neuroendocrine and behavioral effects in response to estrogen
exposure (Majewska, 1987). Various studies have suggested that the neuronal
plasma membrane may serve as a direct target for the rapid action of
estrogens, which may lead to modification of neurotransmitter release or
their receptor/effector systems. First, neurophysiological studies on CNS
neurons have demonstrated rapid modulation of neuronal excitability,
including a brief hyperpolarization and increase in potassium conductance of
the postsynaptic membranes of medial amygdala neurones (Nabekura et al.,
1986), increase the firing rate of medial preoptic and septal neurones (Kelly et
al., 1978), and increase the cerebellar neuronal responsiveness to
iontophoretically applied glutamate (Smith et al., 1987) after application of
physiological levels of 17 (3-E2- Second, biochemical studies have also
provided evidence for the direct actions of estrogens on neuronal
membranes. This included an increase in amphetamine-stimulated striatal
dopamine release in vitro superfusion system with 17 (3-E2 or
diethylstilbesterol, but not with 17 0C-E2 (Becker, 1990) as well as in vivo
microdialysis in freely moving rats (Becker & Beer, 1986). Also, 17 P-E2 is
shown to enhance the responses of adenylate cyclase to biogenic amines in
striatal neurons culture (Maus et al., 1989). Finally, morphological
experiments have also demonstrated rapid plasma membrane ultrastructural
modifications in response to sex steroids application (Garda-Segura et al.,
1987, 1989). Employing freeze-fracture techniques, within 1 min,
physiological concentrations of 17 P-E2 increased the density of exo-
endocytotic pits in cerebrocortical and hypothalamic neuronal membranes in
culture, which was blocked by estrogen antagonist tamoxifen (Garcia-Segura
et al., 1987,1989). These rapid effects (from seconds to minutes of latency)

15
with estrogen hormone on neuronal membrane excitability (Garda-Segura et
al., 1987,1989; Kelly et al., 1978; Nabekura et al., 1986; Smith et al., 1987) occur
much too rapidly to be accounted for by new mRNA synthesis and translation
into proteins. This has led some investigators to suggest a possible action of
gonadal steroids directly on neuronal membrane function/components.
Moreover, Pietras and Szego (1979) have reported an increase in cyclic AMP
concentrations in uterus within 15 minutes after E2 treatment. Although
actinomydn D (an RNA synthesis inhibitor) can effectively prevent the full
expression of long-term E2 effects on target tissues, the rapid or short-term
effects of E2 seem to be independent of RNA/protein synthesis. These effects
are most likely mediated by membrane-associated E2 receptors (Pietras &
Szego, 1979; Towle & Sze, 1983).
Estrogen Receptors
Intracellular/cytosolic receptors
The pioneering studies of Glascock and Hoekstra (1959) and of Jensen
and Jacobson (1962) utilizing radiolabeled estrogen demonstrated selective
localization and retention of the label in tissues known to be targets for
estrogen action. Jensen and Jacobsen (1962) in their studies also demonstrated
E2 binding to a spedfic cytosolic receptor protein. The application of sucrose-
density gradient centrifugation then led to further characterization of the
cytosolic estrogen receptor (Toft & Gorski, 1966). Subsequent studies provided
further evidence to satisfy the criteria for an E2 protein receptor. These
criteria included the stereospecifidty for E2 binding (Noteboom & Gorski,
1965), saturable or limited number of binding sites (Gorski et al., 1968;
Noteboom & Gorski, 1965), size determination by gel filtration

16
chromatography (Gorski et al., 1968; O'Malley et al., 1969), sucrose gradient
analysis (Jensen & DeSombre, 1973; Toft & Gorski, 1966), and sensitivity to
heat and proteases but not to nucleases (Gorski et al., 1968; Noteboom &
Gorski, 1965; O'Malley et al., 1969).
The question of the subcellular localization of the E2 receptors was then
resolved by a number of experimental criteria. These included subcellular
fractionation by differential centrifugation after 3H-E2 exposure in vivo or in
vitro (Jensen & Jacobson, 1962; King et al., 1965, Mowles et al., 1971) which
provided early evidence for nuclear localization of E2 receptors. Conversely,
E2 receptors remained in the soluble, high-speed cytosol fraction in tissues
which were not previously exposed to the E2 hormone (Jensen et al., 1968,
1973). Taken together, available data indicate that the unoccupied receptors
migrate between both the cytoplasmic and nuclear compartments but are
believed to be primarily concentrated in the nuclei in a reversible equilibrium
binding state with the nuclear components. Binding of E2 to the cytosolic
receptor results in a transient biologically active occupied transformed
receptor. Then the transformed receptor translocates into the nucleus with
an enhanced affinity for the nuclear acceptor sites, favoring receptor binding
to the acceptor site on the DNA component. This triggers gene expression
(changes in mRNA synthesis and modification) and the expression of
proteins in the target cell. These effects are generally observed with a delay of
several hours or even days between the exposure of the target tissue to E2 and
the first detectable changes in cellular function.
Membrane receptors
Numerous documented reports have emerged to indicate that the
rapid actions (from seconds to minutes of latency) of gonadal steroids on

17
some target tissues may be caused by direct interaction with plasma
membrane receptor/effector components (Becker, 1990; Garcia-Segura et al.,
1987,1989; Kelly et al., 1978; Majewska, 1987; Nabekura et al., 1986; Smith et
al., 1987; Towle & Sze, 1983). In fact, biochemical studies have demonstrated
specific binding sites for sex steroids in synaptosomal plasma membranes
prepared from the rat brain (Towle & Sze, 1983). Furthermore, the presence
of steroid binding sites have also been demonstrated on plasma membranes
of other target tissues as well, including liver (Suyemitsu & Terrayama, 1975),
pituitary (Koch et al., 1977), and uterus (Pietras & Szego, 1979). In all these
instances, the exact physiological/pharmacological function of the membrane
binding sites for steroids has yet to be determined. However, these binding
observations are compatible with the rapid non-genomic effects of estrogen,
which are not easily accommodated within the genomic model (McEwen et
al., 1982,1984; Majewska, 1987). Collectively, the presence, if real, of these
speculative membrane receptors can account for the rapid neurotropic effects
of E2- Furthermore, these receptors may be involved in the modification of
CNS neurotransmission.
Estrogen-receptor binding kinetics
Estrogen target tissues, i.e., brain, anterior pituitary, uterus, etc.,
apparently contain a single, specific estrogen binding component (type I
estrogen receptor). The unoccupied receptors migrate between both the
cytoplasmic and nuclear components (Walters, 1985). The cytosolic receptors
bind E2 with high affinity such that the steroid-receptor complex remains
intact for translocation into the nucleus (Towle & Sze, 1983). In the rat
uterus, estrogen-filled binding sites do not undergo detectable degradation
over a 24-hr period at temperature up to 30C (Walters, 1985). Following a

18
single injection of E2, the translocation of cytosolic estrogen receptor (ERc) to
the nucleus has been reported to be nearly complete within 1 hr in the rat
uterus (Jakesz et al., 1983). When ERc was assessed by exchange assay 6 hrs
after hormone administration, ERc levels continued to remain very low.
However, an increase in nuclear estrogen receptor (ERn) was concomitantly
observed following E2 injection, reaching maximal levels after 1 hr (Jakesz et
al., 1983). Following an apparently near quantitative translocation of ERc to
the nucleus, ERn concentrations declined to ~ 30% after 6 hrs. However, with
repeated injections of estrogen which maintained continuous receptor
saturating concentrations of [3H]E2 over a 6-hr period, conservation of total
cellular receptors in both cytoplasmic and nuclear fractions were observed
(Jakesz et al., 1983). It is important, however, to note that under continuous
steroid exposure qualitative changes in receptor properties (down regulation)
occur over time in both cytosol and nuclear compartments. It is thought that
the ERc present at 6 hr after estrogen administration originate from a
replenished pool of receptors. This replenished pool has been reported to be
partially dependent on protein synthesis. However, inhibition of protein
synthesis by cycloheximide did not inhibit replenishment after estrogen
exposure. Thus, estrogen target tissues, particularly uterus may represent a
system in which estrogen receptors replenishment appears to be due entirely
to receptor recycling (Jakesz et al., 1983).
Determination of the kinetic binding parameters indicated a high
affinity estrogen-binding site (Kd = ~ 10-10 M) for brain ERc and a Bmax of ~ 3
fmol/mg protein (Walters, 1985). These binding parameters are similar to
estrogen binding kinetics of other tissues. Rate of association with the
receptor has been reported to be 4.4 x 105 M-l S-l while rate of dissociation was
2.4 x 105 M-l S-l with ti/2 = 80 hrs at 0 to 4C. Half-life for the clearance of

19
nuclear estrogen-receptor complexs were estimated to be 2 hrs for 17 (3-E2
(Walters, 1985).
Finally, the maximum biological responses seemed to be determined
not by the level of hormone binding to cytosol receptors, but by retention of a
small proportion (10 to 15%) of the receptors on a limited number of saturable
estrogen receptor-binding sites (acceptor sites) in the nuclear DNA (Walters,
1985). This is supported by the observation of a good correlation between the
duration of nuclear occupancy and uterine growth stimulation in a series of
short and long acting estrogen and their derivatives.
Role of Estrogen in the Menstrual Cycle
During the normal reproductive life (30 to 40 years), a female
menstruates 300 to 500 times (Tepperman, 1981). The menstrual cycle is
characterized by monthly rhythmic changes in the secretion pattern of the
reproductive hormones and the corresponding changes in the sexual organs
as well. The duration of the cycle averages 28 days. The significance of the
menstrual cycle are (1) maturation of an ovum and ovulation; (2) preparation
of the uterine endometrium for implantation of an embryo; and (3)
expression of the secondary sex characteristics associated with the procreative
act. The coordination among these events is achieved by precisely timed
fluctuation in the production and secretion rates of a number of hormones
associated with the hypothalamic-pituitary-ovarian axis. In humans and
primates the ovarian E2 is perhaps the major driving force for the
initiation/maintenance of the cycle. This implies that the primary stimulus
in triggering the initiation of the preovulatory gonadotropin surge, the
central event in the cycle, is caused by the rise in E2 levels during the late

20
follicular phase. The dynamic changes of the ovaries have a periodicity of
once every 28 days; however, these events are, in turn, regulated by the cyclic
changes of pituitary gonadotropins (Ross, 1985; Schwartz, 1981; Yen, 1978).
The pattern of gonadotropin secretion is, in turn, maintained by the pulsatile
mode of the hypothalamic LHRH secretion (Levine & Ramirez, 1982; Marshal
& Kelch, 1986). Finally, the whole cascade of events is regulated by negative
and positive feedback effects of ovarian hormones (McCann, 1982; Pohl &
Knobil, 1982). In a sense, the dynamic interplay between the brain and
pituitary and the dynamics of the ovarian feedback mechanisms govern the
reproductive cycle. That is, the momentum gained from one phase of the
cycle powers the next phase, and continues on into the next cycle.
Feedback regulation
The hypothalamic-pituitary-ovarian-endometrial axis in females
changes markedly with the menstrual cycle. In humans and nonhuman
primates, the cyclic pattern of gonadotropin secretion (the LH surge
particularly) seems to depend more on ovarian estrogen than neural signals.
This is evidenced by the observation that menstrual cycles in primates were
maintained with a constant dose of LHRH per pulse with no variation in
amplitude or frequency of LHRH administration (Knobil, 1980). These
observations suggested that the hypothalamus may only play a permissive
role in the control of the menstrual cycle (Knobil, 1980). However, more
recent observations in human and nonhuman primates indicate that normal
menstrual cyclicity does depend on LHRH pulse parameter that changes in
frequency and amplitude during the menstrual cycle (Dalkin et al. 1989; Ferin
et al., 1984; Marshal & Kelch, 1986). Studies in animals as well as in humans
have shown that LHRH and gonadotropins are secreted in episodic fashion

21
(Levine & Ramirez, 1982), and the dynamics of the LHRH pulsatile mode is
essential for the differential synthesis and release of gonadotropins (Marshal
& Kelch, 1986). Furthermore, ovarian hormones concurrently modulate
LHRH pulse parameters and, therefore, are important regulators of
gonadotropin secretion. For instance, the preovulatory rise in E2 increases
the LH pulse frequency, and since there is a good concordance between LH
pulses in plasma and LHRH pulses in hypothalamic portal blood, this
indicates that a transient elevation in E2 levels increases the frequency of
LHRH secretion (Marshal & Kelch, 1986). The feedback effects of E2 not only
modulate the release of LHRH from the hypothalamus but also the
responsiveness of the pituitary to LHRH signals (Marshal & Kelch, 1986).
Therefore, the effects of E2 on the gonadotropins LH and FSH are coordinated
at the levels of both the hypothalamus and pituitary.
Estradiol displays both inhibitory and stimulatory effects on the
secretion of these hormones. Inhibitory, or negative feedback effects, are seen
during periods of basal LH secretion throughout the menstrual cycle. During
the follicular phase, the follicle secretes low levels of E2, and during the luteal
phase, the corpus luteum secretes large quantities of both estrogen and
progesterone. The combined effects of these steroids inhibit the secretion of
LHRH and consequently reduce the release of pituitary gonadotropins.
However, the positive feedback effects, or stimulation, are observed after a
transient and progressive increase in the titers of E2. That is, an increase in
serum E2 levels of 150 pg/ml or greater for 24 to 36 hours during the late
follicular phase of the menstrual cycle (Ross, 1981). This occurs in response to
the increase in LH pulse frequency which subsequently results in the
preovulatory surge of LH. This condition subsequently causes ovulation of
the ovum from the Graafian follicle and the formation of a new corpus

22
luteum. In the absence of gonadal steroids such as following ovariectomy or
menopause, the negative feedback effects of estrogens on LHRH and
gonadotropins are removed and thus, serum LH and FSH levels increase.
From the clinical point of view, constant exposure to estrogens (or
maintenance of their elevated levels) prevents the preovulatory surge of LH
by exerting a negative feedback mechanism on the hypothalamic-pituitary
unit. This strategy represents the primary mechanism by which the E2
component of contraceptives prevents ovulation. Collectively, the effects of
E2 on the hypothalamic-pituitary unit depend on the exact duration or
magnitude (or both) of exposure to the hormone.
Role of Estrogen in the Rat Estrous Cycle
Rats have 4 to 5 days of estrous cycle (Long & Evans, 1922). The estrous
cycle, like the menstrual cycle in human and nonhuman primates, represents
an extraordinary sequence of events in hormonal and behavioral changes,
and it is verified by cyclic changes in vaginal cell morphology (Long & Evans,
1922). The normal cycle consists of one day of estrus, followed by two days of
diestrus (I and II), and one day of proestrus. The dynamic relation between
the hypothalamic releasing factor, pituitary gonadotropins, and ovarian
steroids during the estrous cycle have been reviewed in considerable detail
(Kalra & Kalra, 1983). Estrogen hormones are responsible for the
maintenance of the estrous cycle, as in the menstrual cycle. On the evening
of estrus, the serum E2 concentrations reach their lowest levels (15 to 20
pg/ml) as the corpus luteum involutes. However, during the days of diestrus
as follicular growth and maturation progress under the influence of FSH
stimulation, ovaries produce increasing concentrations of estrogens. E2 levels

23
peak to 40 to 80 pg/ml on the evening of the proestrus day of the cycle
(Butcher et al., 1974; Kalra & Kalra, 1977). E2 levels then decline rapidly to
basal levels on estrus. The production of E2 by the follicles of the ovaries is
controlled via feedback mechanism (Richards & Hedin, 1988).
Feedback regulation
Ovarian cycle of the rat, like the human, is regulated by the pituitary
gonadotropins LH and FSH. However, in the rat, the gonadotropins appear to
be primarily under the stimulatory influence of the hypothalamic LHRH
neuronal activity (Kalra & Kalra, 1980, 1983; McCann, 1982). This is evidenced
by the fact that, in ovariectomized steroid-treated rat, the positive feedback
effects of estrogen are expressed as a daily signal for mid-afternoon LH
hypersecretion which ensues for several days if concentrations of E2 are
maintained at or greater than E2 levels seen on the proestrus day (Legan et al.,
1975). As in the human and primate, the feedback effects of E2 on the
gonadotropin secretion are coordinated at the levels of both the
hypothalamus and the anterior pituitary gland (Kalra & Kalra, 1983, 1989;
plant, 1986).
In the rat estrous cycle, E2 also exhibits both inhibitory and stimulatory
effects on the hypothalamic-pituitary unit. The inhibitory, negative feedback,
effects are observed during periods of basal LH and FSH secretion throughout
the estrous cycle or after chronic exposure of ovariectomized rats to E2. The
inhibitory effects of E2 are exerted within the hypothalamus (the medial basal
and medial preoptic areas) and/or the anterior pituitary. Implantation of E2
in the medial basal hypothalamus (MBH) (Smith & Davidson, 1974) or
chronic elevation in serum E2 levels (Henderson et al., 1977) were shown to
suppress LH levels and increase the pituitary responsiveness to LHRH,

24
suggesting that the primary site of E2 action is perhaps within the MBH.
However, the stimulatory effects, or positive feedback, are observed after a
transient increase in follicular estrogen secretion from diestrus II through
proestrus or after more than 48 hours of sustained E2 exposure to
ovariectomized rats (Kalra & Kalra, 1983; 1989; Legan et al., 1975). The
positive feedback action of E2 in the estrous cycle is believed to be exerted
primarily on the medial preoptic area of the ventral diencephalon since
lesioning of this area (Wiegand et al., 1980), or interuption of its connections
to the MBH (Halasz & Gorski, 1967), or implantation of antiestrogen
clomiphene into this region (Docke et al., 1989) completely prevent the LH
surge and ovulation in the rat. The preovulatory rise of E2 increases the
hypothalamic LHRH mRNA (Rosie et al., 1990), the LHRH pulse frequency
and pulse amplitude at mid-cycle (Dalkin et al., 1989; Marshal & Kelch, 1986).
However, at the pituitary level E2 increases the responsiveness of the
gonadotrophs to LHRH on the afternoon of proestrus. This allows a priming
effect and thus augments the response of the pituitary to subsequent LHRH
messages. The increase in pituitary responsiveness must precede the
hypothalamic LHRH message for the LH surge to occur.

25
Blood-Brain Barrier
Historical Overview
The blood-brain barrier is vital to brain function. Neurons are
extremely sensitive to the ratio of the concentrations of ions across their
plasma membranes. Furthermore, the concentrations of excitatory and
inhibitory substances in the extracellular neuronal environment must be
tightly regulated for optimal brain functioning. Perhaps because of the
functional intricacy of neurons, the mammalian CNS acquired this
ultrastable cellular environment to perform effectively. This stable
extracellular environment is achieved by several morphological and
enzymatic components collectively referred to as the blood-brain barrier
(BBB). The BBB is not an absolute barrier, but rather a selective and
protective barrier (Neuwelt, 1989; Rapoport, 1976; Suckling et al., 1986). Many
pharmacological agents are excluded from entering the brain because of the
existence of BBB (Bodor & Brewster, 1983; Schanker, 1965).
The concept of barrier was originally postulated by Ehrlich at the end of
the 19th Century (Levin, 1977; Pardridge et al., 1975; van Deurs, 1980). When
vital dyes, such as trypan blue, were injected into the bloodstream, the dyes
penetrated almost all organs of the body but did not enter the brain tissue or
the cerebral ventricles (Levin, 1977; Pardridge et al., 1975; van Deurs, 1980).
However, when dyes were injected into the cerebral ventricles, brain tissue
was stained and the dye did not readily enter the bloodstream. Thus, the lack
of staining by the dyes was not an intrinsic property of brain tissue, rather the
existence of a dynamic barrier interface between the blood and the brain.

26
Subsequent studies which utilized different compounds, including drugs and
radioactive tracers further substantiated the concept of BBB. Furthermore, it
was later discovered that many small molecules were similarly excluded from
transport into the brain. This, then, led to the suggestion that BBB is absolute;
a concept which was soon thereafter refuted when the nutrient requirements
of the brain were elucidated (Davson, 1976).
Ultrastructural studies have shown that there are several differences
between the systemic capillaries and the cerebral capillaries which explain
their permeability differences. Morphologically, the ultrastructural feature
which most distinguishes CNS microvessels is the endothelial cell lining of
the these vessels (Brightman, 1977). The morphological features of the brain
capillaries were elucidated using horseradish peroxidase (HRP). This
relatively small enzyme has a high affinity for radiopaque substances, such as
osmium tetroxide which can be visualized, and it does not cross cerebral
capillaries (Brightman, 1977). When introduced directly into the brain, it
readily diffused throughout the extracellular space but did not pass between
the endothelial cells of cerebral capillaries (Reese & Karnovsky, 1967). So the
anatomical basis for the BBB was identified as the endothelial lining of the
cerebral capillaries. Unlike systemic endothelial capillaries, the cerebral
counterparts are joined by tight junctions (Brightman & Reese, 1969). These
tight junctions form a zona occludens and provide, for molecules like HRP,
an absolute barrier. Morphologically, these junctions consist of aligned intra-
membranous ridges and grooves which are in close apposition (Oldendorf,
1977; Shivers, 1979). Two additional morphological features of the cerebral
capillaries contribute to the BBB as well: (i) cerebral endothelial cells have a
paucity of vesicles and of vesicular transport feature and (ii) the perivascular

27
astrocytic endfeet seem to be involved in the regulation of nutrients flux and
uptake of substances from the circulation (Broadwell & Salcmen, 1981).
In addition to the aforementioned structural features contributing to
the BBB, the presence of various enzymes in and around the endothelial cells
of cerebral capillaries may play a vital role in limiting and perhaps protecting
the brain from a variety of blood-borne substances (Hardebo & Owman, 1980).
Thus, the presence of catechol-o-methyltransferase, monoamine oxidase,
aromatic amino acid decarboxylase and gamma-aminobutyric acid
transaminase in the vicinity of cerebral vasculature may restrict the entry of
various blood-borne chemicals, i.e., neurotransmitters/neuromodulators or
therapeutic agents into the CNS. Such a protective mechanism against
circulating neuroactive substances is essential since optimal CNS function
requires a delicate balance between neurotransmitter release, metabolism and
uptake in the vicinity of neurons. Finally, the enzymatic component of the
BBB may also play a role in excluding some of the lipophilic compounds
from the CNS which otherwise might passively diffuse through the barrier.
Regarding the nutrient requirements of the brain, there are numerous
specialized carrier transport systems which are localized within the BBB for
uptake of nutrients from the circulation (Fenstermacher, 1985; Pardridge,
1981,1983,1986,1987). These include, specific carrier-mediated transport
systems for numerous classes of nutrients (Pardridge, 1981, 1983), receptor-
mediated transport mechanisms for plasma proteins and peptides (Pardridge,
1986), and plasma protein-mediated transport of protein-bound substances
(Pardridge, 1988a, 1988b). These transport systems are localized on the
luminal (or blood side) as well as on the antiluminal border (or brain side) of
the BBB (Pardridge, 1988a, 1988b), and are characterized by saturability and
specificity. These transport mechanisms of the BBB, therefore, provide

28
means for bidirectional movement of selective molecules. However, with
few exceptions, these carrier systems are not involved in transport of
chemotherapeutic agents across the BBB (Greig et al., 1987).
Potential Asset to Utilize in the Design of Brain-Specific Drug Delivery
The unique architecture of the BBB allows only the transport of
compounds either by specific transport systems or by simple diffusion directly
through cell membranes if they are to gain access to the brain parenchyma
and extracellular spaces. Therapeutic agents are no exception. They can access
the brain through either of these routes. Furthermore, the bulk transport of
materials is limited due to the sealing of endothelial gap junctions and the
lack of vesicular transport system in the cerebral capillaries. As a result, most
drugs that enter the CNS must do so by passive diffusion through the
phospholipid cellular matrix of capillary endothelial cells. The lipophilicity
of drugs, defined by their octanol-water partition coefficient, correlates with
their ability to penetrate the BBB for several classes of drugs, including
narcotics (Oldendorf et al.,1972), barbiturates (Levin, 1980) and P-receptor
antagonists (Cruickshank et al., 1979). Furthermore, drugs which cannot
penetrate the BBB can gain access to limited areas of the brain around the
circumventricular organs. Collectively, the CNS has evolved mechanisms to
protect itself by excluding hydrophilic (polar substances) and other
compounds which may be harmful to its optimal functioning.
Unfortunately, this barrier mechanism impedes the delivery and transport of
many potentially useful therapeutic agents to the CNS, thus severely
complicating the effective treatment of brain diseases.

29
A strategy that could achieve an improved delivery of drugs to the
brain with sustained release in that tissue would be of great advantage. A
general approach to increase brain concentration of drugs and thus, their
effectiveness has been the design of lipid soluble prodrug from water soluble
drugs (Bodor, 1981, 1985; Bodor & Kaminski, 1987; Sinkula & Yalkowsky, 1975;
Stella, 1975). Prodrugs are pharmacological agents which have been
transiently modified to improve their lipophilicity as well as to hinder their
rapid metabolic inactivation via enzymes. Ideally, the prodrug is biologically
inactive but reverts to the active, parent drug in vivo at, or around, the site of
action. This transformation can be mediated by an enzyme or may occur
chemically as a result of designed instability in the structure of the prodrug.
The purpose of prodrug modification is to increase the concentration of the
active drug at or near its site of action, thereby increasing its potency/efficacy.
By temporary masking the polar groups of a drug, the lipophilicity of the drug
is increased and thus, its ability to enter the brain parenchyma is enhanced.
Once in the brain, hydrolysis of the masking groups will release the active
drug.
Nevertheless, potential problems are associated with the prodrug
approach (Gorrod, 1980). For instance, by increasing the lipophilicity of a drug
via the prodrug approach, it may not only improve its diffusion through the
BBB to gain access to the CNS, but also ensures the uptake of the compound
into all other tissues and thus, exposure to a greater drug burden. This is a
major limiting factor in the use of prodrugs, especially those with cytotoxicity,
i.e. antineoplastic agents, or those with broad spectrum of peripheral site of
actions such as steroids. Furthermore, even if enhanced CNS
delivery/uptake is achieved via the prodrug approach, the efflux of the drug

30
is concurrently enhanced from the CNS. This results in poor retention and
minimal or no improvement in the biological half-life of the drug.
To overcome the problem of potential general toxicity associated with
enhanced lipophilicity of prodrugs, novel redox-based chemical delivery
system (CDS) for drugs has been designed which exploits the unique
architecture of the CNS BBB (Bodor, 1987; Bodor & Farag, 1983, 1984; Bodor &
Simpkins, 1983; Bodor et al., 1981, 1987,1988). By definition, a CDS is a
biologically inert molecule which requires several chemical conversions
leading to the active, parent drug at or near the site of action (Bodor, 1987;
Bodor & Brewster, 1983). The multiple, facile chemical conversions may lead
to (a) selectivity in drug delivery; (b) improve the drug half-life; and (c)
decrease the toxicity of the drug. The redox-based CDS utilizes a carrier
molecule that can exist as a lipid soluble (in the reduced state) or water
soluble (in the oxidized state). The mechanism of its drug delivery is based
upon an interconvertible dihydropyridine <=> pyridinium ion carrier (Bodor,
1987). In this brain-specific CDS, the lipoidal dihydropyridine moiety is
attached to the drug, thus increasing its lipid solubility and thereby enhancing
its permeability through the BBB. The reduced dihydropyridine can be
oxidized, after its administration, to the pyridinium ion in the brain as well as
in the periphery including systemic circulation. The charged pyridinium-
drug complex is thus locked into the brain while the same moiety rapidly
clears from the periphery by renal or biliary processes due to its increased
hydrophilicity. Sustained release of the active, parent drug from the charged
pyridinium-drug complex occurs in the brain as a result of the enzymatic
hydrolysis of the ester (or amide, etc.) linkage between the drug and the
pyridinium moiety.

31
Collectively, the ability to preferentially deliver and sustain the release
of a drug in the brain, thus sparing non-target site tissues, should improve
the therapeutic index of the drug by (i) increasing the concentrations and/or
residence time of the drug at its receptor site in the brain and (ii), equally
important, decreasing the concentrations and/or residence time of the drug at
the potential peripheral sites of toxicities, thereby decreasing its untoward side
effects. Furthermore, this approach may be potentially advantageous in the
treatment of brain diseases by virtue of the need for lower or less frequent
doses of the drug.
This redox-based CDS has been applied successfully to brain-specific
delivery of a wide variety of therapeutic agents, including phenylethylamine
(Bodor et al., 1981; Bodor & Farag, 1983), dopamine (Bodor & Simpkins, 1983;
Simpkins & Bodor, 1985), gamma-aminobutyric acid (Anderson et al., 1987b),
(3-adrenergic blocking agents (Bodor et al., 1988), antitumor drugs (Bodor &
Brewster, 1983), antiviral agents and antibiotics (Bodor & Brewster, 1983),
testosterone (Bodor & Farag, 1984), estradiol (Bodor et al., 1987), and
norethindrone (Brewster et al., 1987).
The application of this redox-based CDS to estrogens, particularly E2
(Bodor et al., 1987), has important clinical and research implications since the
hormone plays major role in the reproductive and nonreproductive
functions by influencing the brain. Estrogens are intrinsically lipophilic and
readily enter the CNS; however, when inside the CNS, there is no
mechanism to prevent their redistribution back to the periphery and thus
exhibit poor retention. Furthermore, because of their inherent lipophilicity,
estrogens equilibrate among all body tissues. This property of the steroid
necessitates either frequent dosing or the administration of a depot form of
the estrogen in order to maintain therapeutically effective concentrations in

32
the brain. Both of these treatment strategies lead to sustained increases in
peripheral estrogen levels. However, when estrogen is attached to the
dihydropyridine carrier, the E2-CDS enhances brain-specific delivery of
estrogen by (i) locking the estrogen into the brain following the oxidation of
E2-CDS to the charged pyridinium moiety (E2-Q+); and (ii) enhancing the rate
of elimination of the lipoidal estrogen, in the inactive E2-Q+ form, from the
periphery following its oxidation to the charged, more hydrophilic
compound.
Therapies Aimed at Targeting/Enhancing Brain Estradiol Levels
Fertility Regulation
Fertility control may be achieved by a wide variety of mechanical,
surgical, and chemical methods. The chemical (steroidal) methods of fertility
control was first introduced by Pincus and Chang (quoted by Tepperman,
1981) and since then it has had important repercussions on population
growth. Commonly used steroid contraceptives consist of synthetic estrogens
in combination with progestins. When given at pharmacological doses
and/or constant exposure, E2 inhibits (via negative-feedback mechanism) the
secretion of gonadotropin hormone-releasing hormone (GnRH) from the
hypothalamus (Kalra & Kalra, 1983,1989; Plant, 1986) and hence, of
gonadotropins (LH and FSH) from the anterior pituitary (Kalra & Kalra, 1989).
The inhibition of the hypothalamic-pituitary-ovarian axis prevents follicular
development and therefore ovulation (Briggs, 1976). However, the use of
oral contraceptives has been associated with many adverse metabolic changes,
including increased risk of coronary atherosclerosis, myocardial infarction in

33
smokers, liver tumors, hypertension, and changes in glucose metabolism that
appear to be estrogen related (Drill & Calhoun, 1972; Firsch & Frank, 1977;
Fotherby, 1985; Inman et al., 1970; Kaplan, 1978; Mays et al., 1976; Thomas,
1988). To reduce the magnitude and the spectrum of these dose-related
adverse effects of estrogen thus would be to reduce the dose (Bottiger et al.,
1980) and/or to use a more natural hormone (Ottosson, 1984). Given the
combined decrease in the contraceptive components, there is the possibility
that suppression of the hypothalamic-pituitary-ovarian function may not be
as effective as with higher dose formulations. Therefore, preferential brain
delivery of E2 with the CDS may provide an effective, long-acting
contraceptive by virtue of sustained release of E2 in that tissue. Furthermore,
the adverse peripheral effects associated with the currently used contraceptive
steroids may be avoided by lowering the dose or the frequency of ingestion.
Menopausal Syndrome
The cessation of menses, menopause, near the age of 50 is the result of
the decreasing production of ovarian estrogens/progestins (Notelovitz, 1986).
This loss of ovarian hormones in 75% to 85% of women leads to a number of
brain-mediated steroid-withdrawal symptoms (Casper & Yen, 1985; Lauritzen,
1973; Yen, 1977), the most frequent being hot flushes (Clayden et al., 1974;
Meldrum et al., 1979). These (patho)physiological alterations appear to be the
result of autonomic discharge which causes peripheral vasodilation and heat
loss (Nesheim & Saetre, 1982). Replacement therapy with estrogens and/or
progestins has been shown to be effective in most menopausal patients in
alleviating the symptoms of the disease (Campbell & Whitehead, 1977;
Huppert, 1987; Upton, 1984). However, numerous retrospective studies

34
indicated an increased risk of peripheral toxicides, including the risk of breast
and endometrial cancer (Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger
et al., 1988; Persson, 1985; Trapido et al., 1984), cardiovascular morbidity
(Barrett-Conner et al., 1989; Kaplan, 1978; Thomas, 1988), and interference
with hepatic metabolism (Burkman, 1988). These adverse effects of estrogens
are dose dependent. Currently used estrogens are administered either in
frequent doses, or as a depot form, in order to maintain therapeutically
effective levels in the brain. Both of these treatment strategies lead to
sustained increases in peripheral estrogen levels and thus peripheral
toxicities. Since currently employed estrogen therapy is contraindicated in
many postmenopausal women, and in as much as some women do not
respond to the existing steroid medications, the brain-enhanced E2-CDS with
sustained release of E2 may be more effective in the treatment of menopausal
symptoms by providing sufficient E2 to the brain while avoiding peripheral
toxicities.
Prostatic Cancer
The primary objective of hormone therapy in prostate cancer patients
is to induce an effective androgen suppression, thus abolishing the growth
promoting effects of androgens on the diseased prostate ( Brendler, 1988;
Cabot, 1896; Huggins & Hodges, 1941; Isaacs et al., 1983; Moore, 1944; White,
1895). Currently a variety of surgical and therapeutic means for inhibiting
androgen production (in the testis) or blocking androgen action (in the
prostate) are being used. These include castration, high-dose estrogen
therapy, GnRH analogues, and antiandrogens (Santen & English, 1989).
Castration or high-dose estrogen therapy remain, however, as the treatment

35
of choice for the endocrine-dependent management of prostatic cancer (van
Steenbrugge et al., 1988). Both treatments are reported to be equally effective
in (i) suppressing the circulating testosterone levels (Carlstrom et al., 1989);
and (ii) controlling the symptoms of advanced prostatic cancer in 70-80% of
patients (Klein, 1979). In contrast to castration, high-dose estrogen therapy
inhibits (via negative feedback mechanism) the hypothalamic-pituitary-
gonadal function leading to chemical castration. However, high-dose
estrogen therapy has been shown to cause severe cardiovascular
complications in patients (Henriksson & Edhag, 1986) due to the alterations in
liver metabolism (von Schoultz et al., 1989). The E2-CDS may be potentially
useful in the treatment of androgen-dependent prostatic cancer by the virtue
of its sustained suppression of the hypothalamic-pituitary-testis function
leading to chemical castration and thus regression of the tumorous tissue.
Body Weight Regulation
Food intake and body weight may vary during the estrous cycle of the
rat (Tarttelin & Gorski, 1971) and the menstrual cycle of primates (Czaja, 1978)
including women (Pliner & Fleming, 1983). A consistent observation is that
food intake and body weight decrease during the follicular phase of the
ovarian cycle when circulating E2 levels increase. Conversely, food intake
and body weight increase during the luteal phase of the ovarian cycle when E2
levels decline and progesterone levels elevate.
Although relatively few studies have evaluated the potential
modulatory effects of E2 on body weight regulation in human subjects, the
available data support a suppressive role of E2 in appetite and body weight.
Morton et al. (1953) in a study involving menstruating women with

36
premenstrual syndrome (PMS) reported 37% had a craving for sweets during
the luteal phase of their menstrual cycles. Increased appetite was reported to
be a frequent PMS symptom in 45 women examined by Fortin et al. (as
reported by Smith & Sauder, 1969). Pliner and Fleming (1983) studied 34
women and observed that the decreased food intake during the follicular
phase was associated with significant weight loss and that the increased food
intake during the luteal phase was concurrent with weight gain. Collectively,
endogenous estrogen in women has consistently been shown, albeit subtle, to
have a suppressive effect on food intake and body weight.
The potential feasibility of E2-CDS to reduce body weight has been
extensively evaluated in the rat (Estes et al., 1988; Simpkins et al., 1988,
1989a,b). Interestingly, a separation between the effects of E2-CDS on body
weight and food intake has been demonstrated. That is, despite weight
reduction, no consistent reduction in food intake has been observed,
indicating that mechanisms other than reduced appetite are responsible for
the weight loss. Collectively, available data indicate that the E2-CDS
chronically suppresses body weight following a single administration in the
rat. Furthermore, the compensatory hyperphasia in response to the weight
loss is prevented by the E2-CDS.
Libido/Sexual Dysfunction
Certain regions of the hypothalamus have been identified to be
involved in the central integration of sexual behaviors (Christensen &
Clemens, 1974). The first direct evidence that sex steroids influence the
central structures associated with mediating and integrating sexual behaviors
was presented when Harris implanted estrogen directly into the cat's

37
hypothalamus. Furthermore, hypothalamic lesions prevent sexual behavior
in animals, even in the presence of adequate estrogen (Beyer et al., 1976).
Male sexual behaviors are composed of two distinct components: (1)
proception, or the awareness and pursuit of a receptive female and mounting
to achieve intromission, and (2) comsumation, or penile erection,
intromission and eventual ejaculation (Davidson, 1972). It is believed that
the expression of the proceptive components are dependent upon the
aromatization of testosterone to E2 in the brain, particularly in the regions of
the preoptic area of the hypothalamus and the amygdala (Beyer et al., 1976;
MacLusky et al., 1984). When E2 was implanted into the preoptic area of the
hypothalamus, it effectively restored mounting and intromissions in the
castrated rat (Beyer et al., 1976). Other studies have shown that the full
restoration of masculine sexual behaviors in castrated rats require E2 action in
the brain and dihydrotestosterone in the peripheral tissues (Lisk &
Greenwald, 1983). Given the aforementioned evidence regarding brain
mediation of sexual behaviors by E2, the E2-CDS may be potentially useful in
the treatment of sexual dysfunctions or psychological impotence that are not
caused by deficits in peripheral androgen-responsive tissue.
Disorders of Depression
It is generally believed that the underlying mechanism(s) of major
mood disorders (mania and depression) may include abnormal functions of
monoamine transmission (Bunney & Garland, 1982; Leonard & Kuschinsky,
1982). Pharmacological evidence suggests that mania is the result of
hyperactivity while depression is due to hypoactivity of monoamines.
Depression (unipolar and bipolar) respond well to antidepressant drugs, such

38
as tricyclic antidepressants or monoamine oxidase inhibitors, which inhibit
the monoamines re-uptake or their respective metabolism (Bunney &
Garland, 1982; Leonard & Kuschinsky, 1982).
The biological basis for the involvement of estrogens in depression
comes from two gynecological problems: premenstrual and postmenopausal
syndromes. The premenstrual syndrome (PMS) refers to the various mood
changes in relation to the menstrual cycle that are experienced by a large
population of fertile age women. When the relationship between symptom
development and normal variations during the menstrual cycle was
examined (Backstrom et al., 1985), a consistent observation was that there
were very few negative symptoms, rather an increased sense of well being
during the preovulatory E2 peak of the menstrual cycle. However, the
maximum degree of symptoms or mood changes occurred during the luteal
phase of the menstrual cycle when progesterone levels are increased. Clinical
studies as early as in 1932 (Bowman & Bender) suggested a possible
therapeutic role of estrogen in the treatment of depression. More recently,
Klaiber et al. (1979) reported on a double-blind study performed to assess the
therapeutic efficacy of estrogen in the treatment of severely depressed
women. The estrogen treatment significantly decreased the degree of
symptoms compared with placebo (Klaiber et al., 1979).
Since the 1930's, numerous other clinical studies have provided
evidence regarding the influence of estrogens on the well-being and mental
performance in postmenopausal women. Hawkinson (1938) reported a
significant improvement in menopausal symptoms including depression and
anxiety. When subjective indices of moods in ovariectomized women were
evaluated in response to estrogen replacement therapy by Rauramo (1975),
estrogen treatment resulted in an elevation in moods to that which was

39
described as being close to normal conditions. A recent study conducted by
Gerdis et al. (1982), using a variety of psychometric measures to estimate
depression, reported that three weeks estrogen treatment (Premarin) in
postmenopausal women significantly improved the symptoms of depression.
Furthermore, DeLignieres and Vincens (1982) reported improvement of
symptoms of depression, aggression, and anxiety in postmenopausal women
that were treated percutaneously with E2 for three months. In another study
(Klaiber et al., 1979), postmenopausal women with primary, recurrent
unipolar depression and a history of unsuccessful therapy of their depression
were treated with estrogen. The evaluation of their progress by the Hamilton
Rating Scale for Depression indicated a dramatic improvement in mean
Hamilton scores in some patients (Klaiber et al., 1979). These studies
suggested that estrogens may have antidepressant activity in postmenopausal
patients.
Another gynecological problem associated with gonadal steroid
withdrawal is the postpartum psychosis or depression. During the third
trimester of pregnancy, E2 levels increase to about 10 to 40 ng/ml (~ 1000-fold
increase over the follicular phase of the ovarian cycle) while progesterone
levels increase to about 100 to 400 ng/ml (~ 100- to 400-fold increase over the
follicular phase) (Ross, 1985; Schwartz, 1981). After parturition, gonadal
steroid levels fall precipitously and thus, it leads to CNS-steroid withdrawal.
Although the etiology of postpartum depression is yet unknown, it is
speculated that the decline in gonadal steroid levels after parturition may be
the primary factor leading to postpartum depression. In severe cases, it has
been reported that administration of estrogen to patients alliveated
postpartum depression by suppressing lactation (Yalom et al., 1968). Perhaps,
since estrogen hormones influence a variety of CNS functions, by modifying

40
central neurotransmitters/neuromodulators levels involved in mood, or
enzymes necessary for their synthesis, or their receptor/effector systems,
estrogen therapy may very well benefit the postpartum depressed patients
(Luine et al., 1975; Maggi & Perez, 1985). Thus, there is reason to suggest the
notion that estrogen therapy with the E2-CDS may be useful for the treatment
of postpartum depression.
Various biochemical studies in the rat, examining the effect of
estrogens on biogenic amines and their enzymes, strongly support the
behavioral and symptomatic improvements which are observed in
postmenopausal patients. Estrogens are reported to inhibit the re-uptake of
norepinephrine in the rat brain (Luine et al., 1975) and decrease the activity of
monoamine oxidase (Holzbauer & Youdim, 1973; Luine et al., 1975).
Additionally, the activity of monoamine oxidase (MAO) in plasma of
postmenopausal patients were also reduced in response to estrogen treatment
(Klaiber et al., 1976). In fact, MAO activity is lower in women during the
preovulatory phase of the menstrual cycle when E2 levels are highest. In
contrast, MAO activity increases during the luteal phase when progesterone
levels are highest (Klaiber et al., 1976).
Collectively, the evidence mentioned above support the notion that
estrogen therapy may have a rational scientific basis for treatment of
depression associated with the decline in endogenous estrogen production.
Estrogen treatment of these conditions may influence brain function via
effects on a number of neurotransmitter systems involved in mood and other
emotional behaviors.

41
Cognitive Impairment of Menopausal Alzheimer's Type
Since the 1950's, several lines of evidence have accrued to suggest that
estrogen hormones may influence certain cognitive functions of the female.
First, the decreasing production of endogenous estrogens/progestins after
menopause or ovariectomy has been shown to cause changes in cognitive
functions, especially in memory, that are prominent among the somatic and
behavioral symptoms of menopause (Kopera, 1973; Malleson, 1953; Lauritzen
& van Keep, 1978). Second, estrogen treatment of menopausal women with
senile dementia of Alzheimer's type is shown to improve both symptomatic
treatment and prevents or slows the progression of dementia in these
patients (Campbell & Whitehead, 1977; Fedor-Freybergh, 1977; Fillit et al.,
1986; Furuhjelm & Fedor-Freybergh, 1976; Hackman & Galbraith, 1976; Honjo
et al., 1989; Michael et al., 1970; Sherwin, 1988; Vanhulle & Demol, 1976).
Third, numerous biochemical studies have demonstrated that estrogen
hormones modulate/enhance cholinergic transmission or activity in regions
of the brain that are important for cognitive functions (Eleftheriou & Dobson,
1972; Iramain et al., 1980; Luine et al., 1975,1980,1983,1985; O'Malley et al.,
1987). Fourth, estrogen receptors have been identified in nuclei of the basal
forebrain structures in the rat (Luine et al., 1975; Morrel et al., 1975; Pfaff &
Keiner, 1973), the major loci of cell bodies of cholinergic neurons which
innervate the cerebral cortex, limbic system, hippocampus and
hypothalamus. These brain regions are believed to be involved in the
pathology of Alzheimer's Disease (Coyle et al., 1983). Finally, perhaps the
most interesting observation, is that the female:male sex ratio in the
prevalence of Alzheimers Disease is 2:1 (Sulkava et al., 1985).

42
Taken together the findings reported in the literature suggest a strong
link among estrogen, the cholinergic system, and cognitive functioning in
women. Using different estrogen preparations and various Memory
Tests/instruments to assess cognitive functioning, results indicated that
estrogen therapy resulted in improvement in attention span, orientation,
memory, mood, and social interaction in a majority of postmenopausal
women with cognitive impairments (Fillet et al., 1986; Furuhjelm & Fedor-
Freyberg, 1976; Hackman & Galbraith, 1976; Honjo et al., 1989; Vanhulle &
Demol, 1976). Furthermore, the improvement in cognitive functioning, in
those patients who benefited, was correlated with an increase in circulating
concentrations of estrogens (Sherwin, 1988). Additionally, biochemical data
obtained from the rat brain further support the involvement of estrogens in
cognition. Studies in the rat brain have shown that cholinergic neurons
respond to administration of estrogen by (1) increasing the activity of choline
acetyltransferase (ChAT, the enzyme that synthesizes acetylcholine)
(Eleftheriou & Dobson, 1972; Iramain et al., 1980; Kaufman et al., 1988; Luine
et al., 1975,1980,1983,1985;) in the basal forebrain, cortex, hippocampus and
hypothalamus; (2) increasing high affinity choline uptake as well as
acetylcholine synthesis (O'Malley et al., 1987) in cerebral cortex; and
furthermore (3) the responses of cholinergic system to estrogen
administration (i.e., enzyme activity) were observed in female but not male
rats and positively related to the dose of E2 and blocked by an estrogen
antagonist (Luine et al., 1980,1983). Finally, E2 application has been shown to
enhance the excitatory actions of Glu, and Glu is known to be essential in
long-term potentiation and memory formation (Smith et al., 1987).
Collectively, available data support the role of estrogens in cognitive
functioning. However, the hypothesis does not exclude the possibility of the

43
involvement of other neurotransmitter systems and estrogens in memory.
Thus, if estrogens serve to maintain or enhance the activity of cholinergic
neurons or serve as a trophic substance which directly or indirectly acts on
cholinergic neurons, the idea of enhanced brain exposure to E2 may be an
improvement in cognition. Further, since any therapy which is aimed at
treating Alzheimer's type dementia must be chronic in its application to the
patient, the sustained release of E2 from the E2-CDS is an additional useful
benefit. As such, a careful evaluation of the E2-CDS for the mechanism by
which it improves cholinergic function and for its potential application to
Alzheimer's Disease patients is clearly warranted.

CHAPTER 3
GENERAL MATERIALS AND METHODS
Drugs and Solutions
Estradiol and Standard Solution
Estradiol-17(3 (E2) was purchased from Steraloids, Inc. (Wilton, NH).
Standard solutions of E2 were prepared in ethanol for the in vitro studies
involving methodology development. Solutions of E2 were stored at -20C
(stock solution) or 4C (working standard).
Estradiol-17(3 incorporated in 2-hydroxypropyl-(3-cyclodextrin (HPCD)
was provided by the Pharmatec, Inc. (Alachua, FL). Aqueous solution of E2
was prepared on the day of experiment in 20% HPCD (wt:vol) for in vivo
injection.
Estradiol-Chemical Delivery System
Estradiol-chemical delivery system, E2-CDS (3-hydroxy-17P-{[(l-methyl-
l,4-dihydropyridine-3-yl)carbonyl] oxy}-estra-l,3,5-(10)-triene) and E2-Q+ (1-
methyl-3-{[(3-hydroxyestra-l,3,5-(10)-triene-17p-yl)oxy]carbonyl} pyridinium
iodide) were synthesized as previously reported (Bodor et al., 1987). Briefly,
the 3,17p-dinicotinate ester of E2 was made by refluxing 17P-E2 with
nicotinoyl chloride or nicotinic anhydride in pyridine. This derivative was
selectively hydrolyzed to the 17-monoester of E2 with potassium bicarbonate
in 95% methanol. The monoester of E2 was then quaternized with methyl
44

45
iodide. The delivery system, E2-CDS, was then prepared by reduction of the
obtained E2-Q+ with Na2S2C>4. The structure of each intermediate and the
final product (E2-CDS) was confirmed by the nuclear magnetic resonance and
elemental analysis: mp 115-130C. The yields at each synthetic step were 64-
94%. Solutions of the E2-CDS in water containing 20% 2-hydroxy propyl-13-
cyclodextrin (wt:vol) were prepared for injection. For E2-Q+, the working
standard solutions were prepared in water/acetone (80:20; vokvol) to be used
in the in vitro methodology development.
Estradiol Pellet
Pellets, weighing 100 mg each, were prepared from crystalline E2 and
cholestrol (CHOL) powder. Both E2 and CHOL were thoroughly mixed in ratio
of 0.5% E2 and 95.5% CHOL and melted in an oil-bath (200C). Using a heated
pasteur pipette, aliquots of the homogeneous mixture were transferred into
small molds made of aluminum foil. After cooling, the solidified pellets
were unwrapped from the foil and each pellet weight was adjusted to exactly
100 mg. These pellets were used in the experiments described in Chapter 9.
Morphine Pellets
Morphine pellets were compounded in our laboratory, as previously
reported (Simpkins et al., 1983), by the method originally described by Gibson
and Tingstad (1970). Each pellet contained 75 mg morphine free base (Merk,
St. Louis, MO), 37.5 mg microcrystalline cellulose (Avisil, FMC Corporation,
Philadelphia, Pa), 0.56 mg Cab-o-sil (Cabot Corporation, Boston, MA) and 1.13
mg magnesium sterate (Fisher Chemical Co., Fair Lawn, NJ).

46
Animals
The laboratory rat was chosen as the experimental animal for all
experiments herewith. Adult male and female Charles River (CD) rats (aged
3-5 months) were purchased from Charles River Breeding Laboratories
(Wilmington, MA). These rats weighed 200-250 g upon arrival and were
allowed several days to adjust to the animal quarters before conducting an
experiment. Animals were housed in a temperature- (24 1C) and light-
(lights on 0500 to 1900 hr daily) controlled room and provided with Purina rat
chow and tap water ad libitum. After a 7-day acclimation period, animals
were randomly divided into various experimental groups of 7-8 rats per
group. This number of rats per group is standard for the field and is based on
our estimates of experimental error in response to the drugs that were
evaluated in these studies.
In experiments which required surgical procedures, animals were
anesthetized with Metofane (Methoxy Flurane, Pitman-Moore Inc., Crossing,
NJ). The surgical procedures consisted of subcutaneous (sc) implantation of
drugs or steroids, gonadectomy, and atrial cannulation. Female rats were
bilaterally ovariectomized (OVX) by a small incision made through the dorsal
peritoneal cavity. Male rats were castrated (CAST) by exteriorizing the
testicles through a midline ventral incision. For atrial cannulation, in order
to facilitate frequent blood sampling from unrestrained animals, a small
incision in the neck-chest area was made to expose the external jugular vein.
A Silastic catheter (i.d. 0.5 mm, o.d. 1 mm) was then positioned into the right
atrium via the external jugular vein. This surgical procedure was done
under sodium pentobarbital anesthesia, according to the guidelines described

47
by Steffens (1969). All animals were then monitored for post surgical
recovery before conducting an experiment on them.
Two methods were employed for collecting blood samples. In most
experiments, animals were killed by decapitation and the trunk blood was
collected in heparinized tubes. In studies which required frequent blood
sampling, animals equipped with atrial cannula were transferred to special
sampling chambers and serial blood samples (1 ml) were removed through
the cannula. All blood samples were collected in a room separate from the
animal quarters. The blood samples were centrifuged and the plasma
separated and stored at -20C until hormone analysis by the
radioimmunoassay (RIA).
Drug Treatment
Steroid Treatment
The E2-CDS treatments described in this dissertation were only given
intravenously (iv). An aqueous solution of E2-CDS was prepared in 20%
HPCD on the day of injection and administered iv (via tail vein), a procedure
which required a brief restraint of the rat without anesthesia.
The 17 P-E2 treatments consisted of either iv administration (Chapters
6 & 7) or sc implantation of E2 pellet (Chapter 9). Implantation of the E2 pellet
was performed in animals under light metofane anesthesia. It should be
pointed out that these pellets did require pre-conditioning (i.e. soaking in
PBS) before implantation. After implantation, each E2 pellet produces a
transient high concentration of E2 in plasma that is followed by a sustained
blood E2 levels for about 2 weeks.

48
Morphine and Naloxone Treatment
In experiments which examined the effects of E2-CDS on the tail-skin
temperature (Chapter 9), animals were addicted to morphine. Morphine
dependency was produced by sc implantation behind the neck region of one
pellet containing 75 mg morphine free base. Two days after the first
morphine pellet, two additional morphine pellets were sc implanted. This
regimen of morphine treatment has been utilized in our laboratory to
consistently produce typical symptoms of morphine dependency and
tolerance (Simpkins et al., 1983, 1984) as measured by several tests of analgesia
and withdrawal (Gibson &Tingstad, 1970; Simpkins et al, 1983, 1984). These
morphine pellets produce serum morphine concentrations of 300 ng/ml by
one hr after implantation and remain elevated at this level through 48 hrs
(Derendorf & Kaltenbach, 1986). Thus, the sustained release of morphine
achieved by the pellet is presumed to produce persistent stimulation of opiate
receptors utilizing our treatment regimen.
Naloxone HC1 from Dupont Pharmaceuticals (Garden City, NJ) was
dissolved in saline and administered (0.5 mg/kg b.w.) subcutaneously.
Plasma Hormone Radioimmunoassays
Protein Hormone Assays
Plasma luteinizing hormone (LH), follicle-stimulating hormone (FSH),
growth hormone (GH), and prolactin (PRL) concentrations were measured in
duplicate by radioimmunoassay (RIA) using NIADDK kits provided by the
National Hormone and Pituitary Program (Baltimore, MD). Plasma LH, FSH,

49
GH, and PRL values are expressed as ng/ml of either the LH-RP-2, FSH-RP-2,
GH-RP-2 or the PRL-RP-3 reference standards, respectively. Values for
unknown plasma samples were derived from the 10 to 90% (linear inhibition
portion) of the respective standard curves. Radioiodinations of the labeled
hormones were performed in our laboratory using standard procedures for a
chloramine-T iodination with gel filtration chromatography to separate free
iodine from hormone-bound iodine.
The ranges of hormone assay detectability were (1) 0.25 to 20 ng/ml for
LH in 50 ul; (2) 2.5 to 200 ng/ml for FSH in 100 ul; (3) 2.5 to 320 ng/ml for GH
in 25 ul; and (4) 0.25 to 50 ng/ml for PRL in 50 ul of plasma sample. Plasma
samples containing undetectable LH, FSH, GH, or PRL were assigned the
respective assay sensitivity of 0.25, 2.5, 2.5 and 0.25 ng/ml for these hormones.
All samples for each hormone in an experiment were assayed in a single run.
Steroid Hormone Assays
Coat-A-Count Estradiol kitsemploying solid-phase [125I]-
radioimmunoassaydesigned for the quantitative measurement of E2 in
serum were purchased from Diagnostic Products Corporation (Los Angeles,
CA). Each kit is equipped with human serum-based standards having E2
values ranging from 20 to 3600 pg/ml (0.07 to 13.2 nmol/1) (technical
information from Diagnostic Products). The cross-reactivity of the E2
antibody has been reported to be <0.3% for E2-Q+ at the concentration of 15
ng/ml and higher (Rahimy et al., 1989a). The cross-reactivity for estriol and
estrone has been reported to be 0.32% and 1.1%, respectively (technical
information from Diagnostic Products).

50
Coat-A-Count Testosterone kitsemploying solid-phase [12511-
radioimmunoassaypurchased from Diagnostic Products were used for the
plasma testosterone (T) assay. The RIA sensitivity for the T assay was 0.2
ng/ml. The cross-reactivity of the T antibody to the DHT and E2 has been
reported to be 3.3% and 0.02%, respectively (technical information from
Diagnostic Products).
Statistical Analysis
For the experimental data that were normally distributed, the
significance of differences among groups were determined by one- or two-way
analysis of variance (ANOVA) (Zar, 1974). Where necessary, data were
transformed (In) prior to ANOVA. Subsequent pairwise comparisons were
made by Dunnett's or Scheffe's multiple range tests using Statview 512+
program from Brainpower, Inc. (Calabasas, CA) for the Macintosh computer.
Where appropriate, data were subjected to area under the curve (AUC)
analysis using the trapezoid method (Tallarida & Murray, 1981) and group
means for AUC were subjected to ANOVA and Scheffe's tests. The level of
probability for all tests were p < 0.05. The specific statistical design employed
in each experiment is indicated in the figure legends.

CHAPTER 4
DEVELOPMENT OF AN ANALYTICAL METHOD FOR THE
QUANTITATION OF E2-CDS METABOLITES IN A WIDE VARIETY OF
TISSUES IN THE RAT
Introduction
The estradiol-chemical delivery system (E2-CDS) offers a novel
approach to non-invasively enhance brain delivery and sustained release of
E2 (Bodor et al., 1987). The E2-CDS is a redox-based chemical-delivery system
and exploits the unique architecture of the BBB, which normally excludes a
variety of pharmacological agents from the brain (Bodor & Brewster, 1983).
The mechanism of E2-CDS drug delivery is based on an interconvertible
dihydropyridine <=> pyridinium salt carrier. Figure 1 schematically shows the
structures and the mechanisms leading to brain-enhanced and sustained
release of E2. Estradiol, when it is chemically attached to the lipoidal carrier,
dihydropyridine, its lipophilicity is further increased and thus, the ability to
enter the brain is enhanced. After systemic administration of the E2-CDS, the
carrier system is then quickly oxidized to the corresponding quaternary
pyridinium salt (E2-Q+). This charged moiety of the carrier system reduces its
rate of exit from the brain, thereby locking a depot of E2-Q+ into the CNS.
Subsequent hydrolysis of the E2-Q+ with nonspecific esterases provides
sustained release of the active species (E2) in the brain. Since the E2-Q+ is
hydrophilic (40,000-fold greater than E2-CDS), its elimination rate from the
periphery is predictably much faster than from the brain.
51

52
Application of the E2-CDS to biological systems creates a problem
concerning the separation and quantitation of tissue levels of the active
species, E2, in the presence of an excess concentration of the inactive
conjugated form of the drug, E2-Q+. Furthermore, pharmacokinetic and
pharmacodynamic studies of the E2-CDS require a reliable and sensitive
method for the quantitative analysis of E2 as well as E2-Q+ in biological tissues
and fluids. Conventional methods of assaying steroids, especially conjugated
forms of E2, are inefficient and extremely time-consuming. (Bonney et al.,
1984; Cortes-Gollegos & Gallegos, 1975; Hoffmann, 1983; Paradisi et al., 1980).
RIA is the only method which has been used on a large scale for the analysis
of steroid hormones in plasma or serum. Previously described methods for
measuring steroidal hormones and their conjugates in biological tissues have
been severely hindered by lengthy extraction and repeated purification
procedures which are required prior to the use of RIA procedure.
Therefore, the objectives of these experiments were to develop a
sensitive, specific method that permits a rapid and reliable quantitation of
both E2 and E2-Q+ in various biological tissues and fluids. This method
development was necessitated by the design of the E2-CDS which exhibits the
predictive, multiple, facile enzymatic conversion to the charged quaternary
ion (E2-Q+) and its subsequent hydrolysis to slowly liberate E2.
Materials and Methods
Initially, several in vitro experiments were conducted to optimize the
selective extraction, purification and quantitation of both E2-Q+ and E2 from a
wide variety of biological tissues and fluids. Subsequently, an in vivo study
was undertaken to assess the reliability and applicability of the in vitro

53
methodology by determining simultaneously the levels of both E2-Q+ and E2
in several tissues following administration of E2-CDS in the rat.
In Vitro Methodology
Specificity of the estradiol antibody for E?
The presence of the estradiol conjugate (E2-Q+) in biological tissues
could lead to overestimation of tissue E2 concentrations if E2-Q+ cross-reacts
with the E2 antibody. Therefore, possible cross-reaction with the E2-Q+ was
examined by adding E2-Q+ standards to the RIA of E2 at concentrations of 2, 5,
15, 50,180, and 360 ng/ml. These doses of E2-Q+ were 100-fold greater than the
corresponding E2 standards used to generate standard curves in the RIA of E2.
Selective solvent extraction of steroids from tissues
Tissues including brain, anterior pituitary, kidney, lung, liver, heart,
and adipose tissue were dissected from adult male Charles River (CD) rats
(Charles River Breeding Laboratories, Wilmington, MA) immediately
following decapitation. All tissues were rinsed in ice-cold saline and stripped
of surrounding connective tissue and fat, blotted dry on paper then weighed
to the nearest 0.1 mg. Tissue samples of known wet weight were then
homogenized (using a Brinkman Polytron Homogenizer; Model PT 10/35) at
moderate speed (setting at 6 for two 15-second periods) in an appropriate
solvent system (depending on the type of tissue and compound of interest to
be extracted). The appropriateness of the solvent system was determined by
screening various organic solvents for effective extraction and acceptable
recoveries of the steroids, with high reproducibility in the assay procedure.
Tissue homogenate pools, at a concentration of 100 mg tissue/ml solvent,
were prepared as follows: for E2 extraction from brain, anterior pituitary,

54
liver, and kidney, 100% methanol was used; for E2 extraction from lung,
heart, and fat, 100% acetone was used. The reason for using different solvents
was that methanol extraction yielded high and consistent (low CV, high CC)
recovery of E2 from brain, anterior pituitary, liver, and kidney but resulted in
low or inconsistent recovery of E2 from lung, heart, and adipose tissue.
However, using 100% acetone, an acceptable and consistent recovery of E2 was
observed for lung, heart, and adipose tissue. For E2-Q+ extraction, all tissue
homogenate pools were prepared in water/acetone (50:50; v:v).
E? recovery estimation. Duplicate aliquots of homogenate, each having
a concentration of 100 mg tissue/ml solvent, were spiked with 90, 180, or 360
pg E2 and vortexed for 1 min. The added steroid was allowed to equilibrate
with the homogenate for 30 min at room temperature. The spiked
homogenates were then centrifuged for 10 min at 1,500 x g. The supernatant
was decanted into a clean test tube and the pellet discarded. For plasma,
duplicate 1 ml aliquots of plasma (from male rats) were spiked with the
aforementioned doses of E2 standards, samples were allowed to equilibrate for
30 min, but no subsequent centrifugation was done. Following extraction, E2
recovery was determined as described below.
E9-O+ recovery estimation. The procedure used for E2-Q+ adhered
closely to that described for E2 recovery. Duplicate aliquots of homogenate
were spiked with 150,300, and 600 pg E2-Q+ and processed similarly to E2-
spiked homogenates. Serving as a control for the percent of hydrolysis of E2-
Q+, separate duplicate aliquots of homogenates were processed without the
addition of E2-Q+ standards until after the supernatant were isolated. These
supernatants were spiked with 300 pg E2-Q+.

55
Blanks. Tissue homogenates or plasma pools were extracted to
determine residual E2 concentrations and thereby served as the estimate of
hormone background.
Hydrolysis of E?-Q+ in various tissue extracts
In preparation for base-catalyzed hydrolysis of E2-Q+, the final volume
of all E2-Q+ extracts, including the control samples, was brought to 900 ul by
the addition of 50% water:50% acetone. To each tube containing E2-Q+
solution, 100 ul of 10N NaOH were added to make the reaction medium IN
(pH >13). All tubes were vortexed and allowed to reach steady-state
equilibrium for 20 min at room temperature. Under the conditions used, a
time-course evaluation of the rate of E2-Q+ hydrolysis indicated that the
steady-state equilibrium was achieved in less than 15 min and thus, longer
incubation times did not work better. Also, it was found that the hydrolysis
of E2-Q+ under basic conditions was maximized in an aqueous/organic
solvent (50% water:50% acetone). After the hydrolysis, the pH of the reaction
medium was adjusted to a pH in the range of 6 to 8 with NaH2PC>4 and HC1.
This is an optimal pH range for the maintenance of the integrity of the C]8
columns. Samples with pH outside this range damaged the column sorbents,
and the presence of column material in the assay sample interfered with the
RIA for E2.
Solid-phase extraction and separation of E? by Ca columns
After Ci8 columns were conditioned with 2 column volumes (6 ml) of
HPLC grade methanol followed by a column volume wash (3 ml) with
distilled water, the samples to be extracted (E2 and E2-Q+ hydrolyzed extracts)
were applied to the columns. Approximately 1 to 2 min were allowed for the
column adsorption to be completed, then the columns were washed with 2

56
ml watenacetone (80:20; v:v). The columns were then allowed to air dry for 3
min before samples were eluted. With aqueous samples, the small amount
of residual water in the column was removed with either 50-100 ul of HPLC
grade hexane or by air-drying for >3 min. For sample elution, 2 aliquots of 500
ul methanol were applied sequentially to each column and the steroid was
eluted under vacuum pressure. The eluates were collected separately into
glass test tubes in the vacuum manifold and then dried under a stream of
nitrogen gas. Methanol was used to elute E2 from the columns since we
observed a column extraction efficiency of 92% with methanol, whereas two
other solvents tested were less efficient in eluting E2 (88% with acetone and
49% with acetonitrile).
Radioimmunoassay of E?
The dried residues of the E2 samples were reconstituted in 300 ul of the
assay buffer (kit Zero Calibrator; Lot # 10E2Z003; 100 ml) then, after vortexing
for 1 min, samples were equilibrated for 30 to 60 min at room temperature.
Duplicate 100 ul aliquots of each reconstituted E2 samples were assayed by
RIA.
Calculations
If the mean blank (tissue homogenates that did not contain the E2
spike) values for an assay were greater than the limits of sensitivity of the
standard curve, (i.e. if detectable E2 was present in the tissue), these values
were subtracted from all spiked samples. Also, values calculated from the
RIA run were adjusted for the volume of the aliquot taken for the RIA,
experimental losses during solvent extraction and chromatography
(determined by the addition of internal standard), and the weight of the tissue
sample used (for the in vivo experiment).

57
In Vivo Studies
To evaluate the applicability and reliability of this procedure to in vivo
condition, adult male Charles River (CD) rats received a single intravenous
injection (tail vein) of the E2-CDS at a dose of 1.0 mg/kg body weight or the
drug's vehicle, 2-hydroxypropyl-(}-cyclodextrin (HPCD). Animals (6 per
group) were then killed by decapitation 1, 7, and 14 days after the drug
injection and the trunk blood was collected in heparinized tubes. The blood
was centrifuged and the plasma separated and stored at -20C until hormone
analyses. Brains were removed, rinsed with ice-cold saline solution and
stripped of their pia matter and immediately stored at -80C until hormone
analysis. Plasma and brain samples were each processed and assayed by the
method described under the In Vitro Methodology section.
The 1.0 mg/kg dose of E2-CDS was chosen for investigation in this
study for the following reasons: 1) we anticipated that the tissue
concentrations of the E2-CDS metabolites, E2-Q+ and E2, would be in a
quantitatable range when using this dose and the application of RIA
procedure, over the time-course chosen for this study and 2) our previous
pharmacological observations indicated that this dose of the E2-CDS is capable
of causing chronic suppression of gonadotropins in castrated rats.

58
Results
In Vitro Methodology
Cross-reactivity of the E? antibody with E?-Q+
The inhibition of binding of 1251-E2 to the E2 antibody caused by E2 and
E2-Q+ is shown in Figure 2. While E2 effectively competed for binding with
the labeled hormone (IC50 = 368 pg/ml; 1.35 pM), E2-Q+ was ineffective in
displacing the 125I-E2 in the RIA (IC50 = 129,000 pg/ml; 326.50 pM). At
concentrations of E2-Q+ of 15 ng/ ml and higher, the cross-reactivity of the E2
antibody for E2-Q+ was <0.3%. Cross-reactivity of the E2 antibody for estriol
and estrone has been determined to be 0.32% and 1.1%, respectively (technical
information from Diagnostic Products).
Recovery of E?
Recovery of E2 was assessed by determining the % of each of three doses
of E2 recovered from the brain, liver, kidney, lung, heart, and fat
homogenates (Table 1). Due to limitations in the amount of tissue available,
only one dose (180 pg/ml) was tested for homogenates of anterior pituitary
glands. Recovery of E2 from each tissue evaluated was found to be dependent
upon the organic solvent used in the extraction step. For brain, anterior
pituitary, liver and kidney, E2 extraction with 100% methanol was found to be
superior to two other solvents (acetone and acetonitrile) resulting in E2
recoveries (average of the three doses tested) of 77% for brain, 78% for
anterior pituitary, 72% for liver and 71% for kidney (Table 1). Methanol was
determined to be a poor solvent for extracting E2 from lung, heart and fat
tissue, but 100% acetone achieved acceptable recovery of E2 in these three

59
tissues. The mean recovery of E2 was 57% for lung, 62% for heart and 64% for
fat tissue. The E2 recovery from plasma was 81%.
Precision of the E? extraction-assay method
The precision of the method of estimating E2 concentrations in a
variety of tissues was determined in three ways. First, we determined the
coefficient of variation (CV) for quadruplicate samples of each tissue at 3
different E2 dose levels. Second, we determined the correlation coefficient
(CC) of the E2 dose-RIA response for each tissue. And third, we determined
the tissue weight-RIA response for brain samples taken at various times after
rats were treated with the E2-CDS.
The CV for quadruplicate determinations of E2-spiked tissue
homogenates or plasma pools were in the range of 0.8 to 7.9% with the
majority of E2 doses in each tissue showing a CV of less than 3.0% (Table 1).
The CC for the E2 dose-RIA response was 0.97 or greater for all tissues,
indicating that over the E2 dose-range tested, the procedure accurately
estimated the E2 concentration of the tissue (Table 1; Figure 3, lower panel).
To determine the accuracy of the method in estimating E2
concentrations in tissues of different weights and hence, to evaluate for tissue
components which may interfere with various steps in the E2 assay
procedure, we evaluated tissue wet weights over the range of 2 to 100 mg
(Figure 4, upper panel). Brain tissue from rats treated 1 or 7 days before with
the E2-CDS was extracted and assayed for E2. The inhibition of binding of 125I-
E2 was correlated with the weight of tissue used for both samples taken at day
1 (CC = 0.99) and 7 days (CC = 0.99) after E2-CDS administration. The
parallelism of the observed inhibition curves indicated that the E2
concentration measured is independent of the weight of tissues used in the

60
determination. Also, the rightward shift in the tissue weight-RIA response
curve at day 7 indicates a decrease in brain E2 concentrations with increasing
time after drug administration.
Recovery of E9-O+
The recovery of E2-Q+ is dependent upon the efficiency of three
processes: (a) the extraction efficiency of E2-Q+ from tissues with
watenacetone; (b) the % hydrolysis achieved under basic conditions; and (c)
the % recovery of E2 following its formation from the hydrolysis of E2-Q+.
The recovery of E2 was determined as described above and the parameters
described below after the adjustment for E2 recovery. The recovery of E2-Q+
was determined as the percent of the spiked concentration of E2-Q+ which was
assayed. The % hydrolysis of E2-Q+ was determined experimentally for each
tissue and the extraction efficiency in watenacetone was calculated as the
recovery times the reciprocal of the % hydrolysis.
The extraction efficiency of E2-Q+ with watenacetone ranged from
65.4% for heart to 80.7% for liver tissue (Table 2). For each tissue evaluated,
the % hydrolysis of E2-Q+ was the primary factor limiting the recovery of this
species. The % hydrolysis ranged from 69% in liver to 37% in adipose tissue
(Table 3). For all tissues except fat and plasma, the % hydrolysis was greater
than 50% (Table 3).
Precision of the E7-O+ extraction-assay method
The precision of the E2-Q+ method was determined using the same
parameters used for demonstrating precision of the E2 method.
The CV for the quadruplicate determinations of E2-Q+-spiked tissue
homogenates or plasma pools ranged from 0.9 to 7.3% with a majority of the
E2-Q+ doses for each tissue showing a CV of less than 5.0% (Table 2).

61
The CC for the E2-Q+ dose-RIA response was 0.98 or higher for each
tissue evaluated (Table 2; Figure 3, upper panel). Thus over the E2-Q+ dose
range tested, the method accurately estimated E2-Q+ concentrations in each of
the tissues.
Increasing brain tissue wet weight from 0.25 to 50 mg caused a highly
correlated decrease in binding of 125I-E2 in the RIA used at day 1 (CC = 0.99),
day 7 (CC = 0.99) and day 14 (CC = 0.99) after the treatment of rats with E2-CDS
(Figure 4, lower panel). The inhibition curves caused by increasing brain
tissue wet weight from animals at 1, 7 and 14 days posttreatment were parallel
and the rightward shift was indicative of the time-dependent reduction in
brain E2-Q+ concentrations (Figure 4, lower panel).
Distribution of E? and E7-O+ in vivo
Figure 5 shows brain and serum levels of E2-Q+ (upper panel) and E2
(lower panel) at various times following administration of the E2-CDS (1
mg/kg) or the HPCD vehicle (day 0 values). Brain E2 concentrations were
increased to 11.1 1.4 ng/g tissue (mean SEM) on day 1 and remained
greater than 3.5 1.1 ng/g at day 14 after drug treatment. A small amount of
E2 was detected in brains of HPCD-treated (control) male rats (0.2 0.07 ng/g),
a likely result of the aromatization of testosterone in the brain (Michael et al.,
1986). Brain concentrations of E2 exceeded serum levels of the hormone by
39-, 41- and 82-fold at 1, 7 and 14 days, respectively, after E2-CDS treatment
(Figure 5).
E2-Q+ levels were increased to 200.9 8.8 ng/g brain tissue at 1 day after
administration of the E2-CDS, and E2-Q+ levels remained elevated to 67.0
17.2 ng/g at 14 days postinjection (Figure 5). The brain to serum ratio for E2-
Q+ was 33,70 and 294 at 1,7 and 14 days, respectively.

62
In brain tissue, E2-Q+ levels exceeded E2 levels by 18-, 22- and 19-fold
and in plasma E2-Q+ levels were 6,13 and 22-fold higher than E2 at 1, 7 and 14
days, respectively.
Discussion
This novel, but predictable, metabolism of the E2-CDS presents several
problems for the quantitation of E2-Q+ and E2, two metabolites of the E2-CDS.
First, since estradiol is active at tissue concentrations of low pg/g, the assay
method for the E2-CDS metabolite, E2, must be extremely sensitive. Second,
since E2-Q+, the moiety "locked" in the brain, is expected to be present in
concentrations much higher than E2, the assay method must be capable of
distinguishing low levels of E2 in the presence of high concentrations of E2-
Q+. Third, the accuracy of the E2 determination is dependent upon the
stability of E2-Q+ against hydrolysis (enzymatic or spontaneous) throughout
the procedures. However, under the conditions utilized, E2-Q+ was quite
stable. When 300 pg E2-Q+ were added to tissue homogenates and evaluated
for spontaneous hydrolysis throughout the procedures used, the E2 recovery
was below the sensitivity of the assay. Finally, as for any assay method,
necessary features are (a) a high recovery of the species of interest; (b) accuracy
of the determinations; and (c) reliability of the method through a wide range
of hormone concentrations and tissue weights. We have provided evidence
for each of the features of the present method for simultaneous
determinations of E2-Q+ and E2.
The RIA procedure provides the needed sensitive endpoint for the
determination of E2 and E2-Q+ levels. This RIA for E2 is sensitive from 0.8 to
1.2 pg E2 / assay tube and exhibits a highly correlated inhibition of 125I-E2

63
binding over the range of 20 to 3600 pg/ml. This level of sensitivity is
substantially greater than the recently published HPLC methods which report
levels of sensitivity of 50 and 10 ng/ml of plasma for E2 and E2-Q+,
respectively (Mullersman et al., 1988). Additionally, the antibody used was
very specific for estradiol, showing a cross-reactivity of <0.3% for E2-Q+ and
has also been described to cross-react with estriol and estrone at the level of
0.3 and 1.1%, respectively. In brief, the RIA described here is sensitive and
specific for E2.
The recovery of E2 is dependent upon the extraction efficiency of the
organic solvent used and the elution efficiency of E2 loaded on the Cis
columns. Since column elution of E2 with 100% methanol was essentially
quantitative, the recovery of E2 was equivalent to the extraction efficiency.
Methanol extraction yielded high and consistent (low CV, high CC) recovery
of E2 from brain, anterior pituitary, liver and kidney but resulted in low or
inconsistent recovery of E2 from lung, heart and adipose tissue. However,
using 100% acetone, an acceptable and consistent recovery of E2 was observed
for lung, heart and adipose tissue. While the reason for this tissue-specific
solvent extraction is not clear, the results indicate that the judicious choice of
solvent allows for the reliable estimate of E2 concentrations in a variety of
tissues. Indeed, E2 was precisely measured in a variety of tissues over a 4-fold
change in the concentration of E2 or over a 50-fold change in the amount of
tissue used in the determination (Figures 3 & 4).
The recovery of E2-Q+ was limited primarily by the percent hydrolysis
of the E2-Q+, since extraction efficiency in water:acetone (50:50; v:v) ranged
from 65 to 81 % for an individual tissue. We observed that our base-catalyzed
hydrolysis of E2-Q+ yielded values of 54% to 69% for all tissues except fat (37%)
and serum (30%). While base hydrolysis is not complete, the % hydrolysis

64
was consistent for each tissue and the variation in % hydrolysis was low (CV
= 0.8% for plasma to 5.0% for brain tissue). Furthermore, other methods of
hydrolyzing E2 conjugates, such as enzyme- or acid-catalyzed hydrolysis, are
much less efficient (11% and 0.8% net hydrolysis, respectively) and require 18
to 24 hours to conduct (Bain et al., 1984; Czekala et al., 1981; Saumande &
Batra, 1984; Segal et al., 1960) versus 20 min for base-catalyzed hydrolysis
under the present conditions.
Analysis of E2-Q+ in adipose tissue was complicated by two factors.
First, E2-Q+ is hydrophilic (due to its charge and polarity) and requires an
aqueous solvent for its effective extraction. This requirement creates
problems in separating the supernatant from the pellet because of the
formation of a superficial layer of lipid above the supernatant phase. Second,
extensive loss of E2 from the supernatant occurs after E2-Q+ was hydrolyzed.
This is due to the presence of fatty droplets in the reaction medium into
which E2 partitioned from the aqueous supernatant; and therefore, it was not
recovered efficiently when the supernatant was transferred onto the Qs
column. These two conditions reduced the recovery of E2-Q+ from adipose
tissue.
Analysis of E2-Q+ in heart and kidney tissues posed a different problem.
After hydrolysis of E2-Q+ extracts in IN NaOH solution, the hydrolyzed
supernatants from heart and kidney required less HC1 and NaH2P04 than
other tissues to adjust the pH to the range of 6 to 8; the optimal pH range for
Ci8 column function. Without determining the exact amount of acid and
buffer needed to achieve the optimal pH range for each tissue, erroneously
high E2-Q+ levels were calculated due to contamination of the assay tube by
the column sorbent.

65
E2-Q+ recovery from plasma samples was low likely because of protein
precipitation caused by hydrolysis and subsequent neutralization.
Centrifugation was needed to separate the supernatants from precipitates
before their application onto the Cis columns. E2 released during the
hydrolysis step likely interacted extensively with albumin and sex steroid
binding globulin and was unavailable for column extraction.
The separation of E2 and E2-Q+ was achieved by 3 different techniques
in the process of extraction and assay of these two products of the E2-CDS.
First, samples were divided and differentially extracted for E2 (methanol or
acetone) and for E2-Q+ (50% water:50% acetone). Although this procedure,
which depends upon the solubilization of the lipophilic E2 in methanol or
acetone and the more hydrophilic E2-Q+ in water/acetone effectively extracted
the intended steroid, separation of the two species was not complete.
However, when extracts were loaded onto the Cis column and eluted with
100% methanol, only E2 was preferentially extracted and eluted by more than
92%. Thus, the column chromatography effectively separated the two species.
Finally, the low cross-reactivity of the E2 antibody for E2-Q+ (<0.3%) ensured
that in samples extracted and chromatographed for E2, virtually no E2-Q+ was
measured. Moreover, analysis of tissue samples treated with the E2-CDS
required various dilutions for each time point which ensured the expected E2
values to fit an appropriate part of the standard curve (ED20 at 800 pg/ml to
E80 at 25 pg/ml). Indeed, dilutions which were performed prior to loading
the extracts onto the Cis column minimized nonspecific interference by E2-Q+
and lipids. Lipids decrease steroid radioimmunoassay accuracy and
reproducibility (Rash et al., 1980).

66
Application of these methods to brain and plasma samples obtained at
various times after treatment with the E2-CDS revealed that both E2 and E2-
Q+ can be quantitated throughout the 14-day time-course of the study. As
predicted, based upon previous reports on brain levels of E2-Q+ (Bodor et al.,
1987; Mullersman et al., 1988), this "locked-in" form of the E2-CDS reached a
33-fold higher concentration in the brain than plasma by 1 day after treatment
with E2-CDS and these brain-blood ratios increased to 294-fold by 14 days.
Brain E2 concentrations were similarly and dramatically elevated relative to
plasma. These observations are consistent with the proposed brain-enhanced
delivery of E2 with the redox-based E2-CDS and indicate that the observed
distribution pattern of E2-Q+ and E2 may explain the long-term
pharmacological effects of the E2-CDS (Anderson et al., 1988a,b; Estes et al.,
1987a,b; Simpkins et al., 1986).
In summary, the described technique for the simultaneous
measurement of E2-Q+ and E2 is sensitive, reliable, specific and applicable to a
wide variety of tissues in the body. The additional feature of rapidity of the
method allows for the determination of about 100 samples in one day.
Collectively, these characteristics indicate that the described techniques could
be applied to the quantiation of other conjugates of steroid hormones.

67
Figure 1. Schematic representation of in vitro synthesis and in vivo
transformation of the estradiol-chemical delivery system (E2-CDS).
E2-Q+ is the charged quaternary form of the E2-CDS which is
"locked" into the brain and quickly eliminated from the
peripheral tissues. Subsequent hydrolysis of E2-Q+ with non
specific esterases results in sustained and slow release of estradiol
in the brain. The trigonelline, carrier moiety, formed upon
hydrolysis of E2-Q+ is non-toxic and is cleared from the brain
rather quickly. Although the in vivo rate constants for these
reactions are unknown among different tissues, the oxidation of
E2-CDS to E2-Q+ is quite rapid in all tissues analyzed (ti/2 = 29
min). However, the rate constant for hydrolytic enzymes may
differ among various tissues.

PERCENT BOUND
68
PG/ML
Figure 2. Inhibition of 125I-E2 binding to an E2 antibody caused by E2 (left,
open circle) or E2-Q+ (right, closed circle). Possible cross-reaction
with the E2-Q+ was examined by adding E2-Q+ standards to the
RIA of E2 at concentrations of 2, 5, 15, 50,180, and 360 ng/ml.
These doses of E2-Q+ were 100-fold greater than the corresponding
E2 standards used to generate standard curves in the RIA of E2.
Data are expressed on logit-log graph (% bound on ordinate is
based on logit = log (percent bound/100 percent bound)). Cross
reactivity data indicated that E2 antibody used was specific for E2
and cross-reacts with E2-Q+ to <0.3% at concentration of 15 ng/ml
and greater.

Assayed Cone, (pg/100 mg) Assayed Cone, (pg/100 mg)
69
Figure 3. Recovery of known concentrations of E2-Q+ (upper panel) and E2
(lower panel) added to brain tissue homogenates prior to
extraction. Duplicate aliquots of homogenate were spiked with
150,300, or 600 pg E2-Q+ (upper panel) or 90,180, or 360 pg E2
(lower panel). After equilibration for 30 min followed by solvent
extraction, the spiked homogenates were centrifuged and the
supernatant was analyzed for E2-Q+ or E2 by the RIA for E2. The
results indicated that the assay method used accurately determines
E2-Q+ and E2 over a wide range of tissue concentrations.

PERCENT BOUND PERCENT BOUND
70
Figure 4. Inhibition of 125I-E2 binding to an E2 antibody caused by increasing
amounts of brain tissue from rats treated with the estradiol-
chemical delivery system (1.0 mg/kg). The upper panel depicts
brain tissue wet weights extracted over the range of 2 to 100 mg for
E2 at 1 or 7 days after treatment with E2-CDS. The lower panel
depicts brain tissue extracted over the range of 0.25 to 50 mg for E2-
Q+ at 1,7 or 14 days after treatment with E2-CDS. Data are
expressed on logit-log graph (% bound on ordinate is based on
logit = log (percent bound/100 percent bound)). The results
indicated that the assay method used accurately measures E2-Q+
and E2 concentrations over a wide range of tissue weights.

E2 (pg/g or pg/ml) E2-Q+ (pg/g or pg/ml)
71
Days post Injection
Figure 5. Effects of a single iv dose of the estradiol-chemical delivery system
on serum and brain levels of E2-Q+ (upper panel) and E2 (lower
panel) or the HPCD vehicle (day 0). Each point represents the
group mean SEM. N = 6 animals per group.

72
Table 1: Recovery and Precision Determination for Biological Samples Spiked
with E2
Tissue*
E2 Added
(Pg)
E2 Assayed^
(Pg)
Recovery
(%)
CVc
(%)
CCd
(r)
Brain
90
73.9
82.1
7.9
180
130.4
72.4
3.3
360
268.1
74.5
3.4
0.994
Anter. Pituit. 180
141.0
78.3
1.8
NDe
Plasma
90
63.4
70.5
2.5
180
151.8
84.3
3.0
360
321.3
89.2
2.9
0.997
Kidney
90
71.5
79.4
2.2
180
121.8
67.6
1.2
360
238.8
66.3
1.8
0.996
Lung
90
52.0
57.8
6.0
180
100.0
56.4
3.7
360
206.1
57.3
2.4
0.993
Heart
90
52.6
58.5
2.7
180
112.6
62.6
2.8
360
230.8
64.3
4.4
0.997
Liver
90
69.4
77.1
1.7
180
126.5
70.3
1.9
360
245.4
68.2
2.5
0.997
Fat
90
66.4
73.8
3.0
180
97.4
54.1
0.8
360
228.0
63.3
1.1
0.975
a 100 mg of tissue or 1 ml of plasma was used,
b Mean of n = 4 for each dose of E2 in each tissue,
c CV = coefficient of variation,
d CC = correlation coefficient,
e ND = not determined.

73
Table 2: Recovery and Precision Determinations for Biological Samples
Spiked with E2-Q+
Tissue*
E2-Q+ Added
E2-Q4- Assayedb
(Pg)
EEc
(%)
Recovery
(%)
CVd
(%)
CCe
(r)
Brain
150
73.7
74.5
49.2
1.3
300
129.6
65.1
43.2
4.6
600
249.9
63.1
41.7
4.6
0.994
Plasma
150
47.5
ND
31.7
6.1
300
87.8
ND
30.0
2.8
600
208.3
ND
34.7
0.9
0.989
Kidney
150
68.4
73.4
45.6
1.0
300
137.9
74.0
46.0
3.0
600
238.3
64.4
40.0
2.2
0.990
Lung
150
64.0
76.3
42.6
1.0
300
131.1
78.2
43.7
2.6
600
236.7
71.6
40.0
1.1
0.995
Heart
150
52.6
64.8
35.0
2.6
300
104.0
64.1
34.7
4.6
600
218.5
67.4
36.4
5.9
0.989
Liver
150
80.4
80.4
53.6
2.2
300
164.0
82.0
54.7
6.7
600
319.2
79.8
53.2
7.3
0.988
Fat
150
39.8
70.9
26.6
2.1
300
78.9
70.2
26.3
2.6
600
163.3
72.7
27.2
3.0
0.988
a 100 mg of tissue or 1 ml of plasma was used,
b Mean of n = 4 for each dose of E2-Q+ in each tissue,
c EE = extraction efficiency,
d CV = coefficient of variation,
e CC = correlation coefficient,
f ND = not determined.

74
Table 3: Percent Hydrolysis of E2-Q+ in Supernatants of a Variety of Tissues
Tissue
E2-Q+ Added
(Pg)
E2-Q+ Assayed^
(Pg)
Hydrolysis
(%)
CVc
(%)
Brain
300
198.3
66.1
5.0
Plasma
300
87.8
30.0
2.8
Kidney
300
186.6
62.2
2.9
Lung
300
167.6
55.8
1.8
Heart
300
162.5
54.2
1.7
Liver
300
207.3
69.1
3.5
Fat
300
112.3
37.4
4.1
a 100 mg of tissue or 1 ml of plasma was used,
b Mean of n = 4 for each dose of E2-Q+ in each tissue,
c CV = coefficient of variation.

CHAPTER 5
DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN MALE RATS
Introduction
Estrogens are intrinsically lipophilic (Abraham, 1974) and readily cross
the blood-brain barrier (BBB) to gain access to the central nervous system
(CNS). However, when inside the CNS, there is no mechanism to prevent
their redistribution back to the periphery as blood levels of the steroid decline
(Davson, 1976). Indeed, when these hormones are used therapeutically to
specifically target the CNS, the steroids tend to equilibrate among all body
tissues due to their high lipophilicity (Pardridge & Meitus, 1979). As a result,
only a fraction of the administered dose accumulates at or near the site of
action in the brain. This property of the estrogens necessitates, either frequent
dosing, or the administration of a depot form of the estrogen to achieve and
maintain therapeutically effective concentrations in the brain (Spona &
Schneider, 1977). Both of these treatment strategies lead to sustained
increases in peripheral estrogen levels.
Furthermore, estrogen receptors are present in many peripheral tissues
(Walters, 1985), where they mediate a myriad of physiological and
pharmacological effects (Murad & Haynes, 1985; Walters, 1985). This further
creates the potential of untoward peripheral side effects (Thomas, 1988). In
fact, constant increases in peripheral tissue exposure to estrogens have been
shown in numerous studies to precipitate various peripheral toxicities,
including risk of breast and endometrial cancer (Hurst & Rock, 1989; Persson,
75

76
1985; Thomas, 1988), cardiovascular complications (Barrett-Conner et al., 1989;
Drill & Calhoun, 1972; Inman & Vessey, 1968; Kaplan, 1978), and alterations
in hepatic metabolism (Burkman, 1988).
Since the brain is the primary site where E2 exerts its beneficial effects
on the estrogen withdrawal symptoms at the menopause (Casper & Yen, 1985;
Lauritzen, 1973; Yen, 1977), to inhibit gonadotropin secretion for fertility
regulation (Goodman & Knobil, 1981; Kalra & Kalra, 1989; Plant, 1986), to
reduce growth of peripheral steroid-dependent tissue tumors such as the
prostate (Rao et al., 1988), and to stimulate male and female sexual behaviors
(Beyer et al., 1976; Christensen & Clemens, 1974), a brain-enhanced delivery
with sustained release of E2 in that tissue is warranted. The ability to deliver
E2 preferentially to the brain, thus sparing non-target site tissues, should
improve the therapeutic index of E2 by (i) increasing the concentrations
and/or residence time of E2 at its receptor site in the brain and (ii), equally
important, decreasing the concentrations and/or residence time of E2 at the
potential peripheral sites of toxicities, thereby decreasing untoward peripheral
side effects.
Having established a reliable, specific method for the simultaneous
quantitation of E2-CDS metabolites in various tissues (Chapter 4; Rahimy et
al., 1989a), and thus to demonstrate the effectiveness of the E2-CDS, extensive
time-course studies were undertaken to evaluate the tissue distribution of E2-
Q+ and E2 in both male and female rats. The objective of this study was to
evaluate a general tissue distribution of E2-Q+ (the intermediate, oxidized
metabolite of the E2-CDS) and E2 (the active, parent steroid released upon
hydrolysis of the E2-Q+) in brain, anterior pituitary, lung, liver, kidney, heart,
fat, and plasma following a single iv dose of 1 mg /kg of E2-CDS in the male
rat.

77
Materials and Methods
Adult, intact male Charles River (CD) rats (225-250 g) received a single
iv injection (tail vein) of the E2-CDS at a dose of 1.0 mg/kg body weight or the
drug's vehicle, 2-hydroxypropyl-p-cyclodextrin (HPCD). Rats (6-7 per group)
were killed by decapitation 1, 7 or 14 days after the drug administration and
the trunk blood was collected in heparinized tubes. The blood was
centrifuged and the plasma separated and stored at -20C until hormone
analysis. Tissues (brain, anterior pituitary, lung, liver, kidney, heart, and fat)
were dissected immediately following decapitation and stored at -80C until
hormone analysis.
Tissue samples of known wet weight at a concentration of 1 mg/20 pi
solvent were processed and assayed by the method described in Chapter 4
(Rahimy et al., 1989a). Tissue homogenates and plasma from HPCD-treated
rats were also extracted to determine the residual E2 concentrations and
thereby served as the estimate of hormone background.
Coat-A-Count Estradiol kits--a solid-phase [125I]-radioimmunoassay--
designed for the quantitative measurement of E2 in serum were used for the
assay of E2 in all tissue and plasma samples. Cross-reactivity of the E2
antibody was determined to be <0.3% for E2-Q+ at a concentration of 15 ng/ml
and higher (Chapter 4). All the purified dried E2 unknowns were
reconstituted in 300 pi of the assay buffer (kit Zero Calibrator) and assayed in
duplicate by the RIA. The intra-assay coefficient of variation for E2 was 1.56%
and all samples were determined in two assay runs.
Calculated values obtained from the RIA run were adjusted for the
volume of the aliquot taken for the RIA, experimental losses during solvent

78
extraction and chromatographic separation (using internal standard), and the
weight of the tissue sample used.
Results
The results of this experiment are presented in Figures 6-9. Figure 6
shows brain (upper panels) and plasma (lower panels) concentrations of E2-Q+
and E2 at various times following administration of the E2-CDS. Brain E2-Q+
concentrations increased to 318 14 ng/g tissue (mean SEM) on day 1,
followed by a linear decline to 39 2 ng/g on day 14. This result indicated a
reduction in E2-Q+ concentration of 46% by 7 days and 88% by 14 days after
administration of the E2-CDS. In contrast, plasma concentrations of E2-Q+
increased to 6.1 0.3 ng/ml on day 1, then rapidly decreased by 79% at day 7
and remained at very low levels (0.23 0.03 ng/ml) at day 14 after the E2-CDS
treatment.
Brain concentrations of E2 increased to 8 0.5 ng/g (day 1) then
decreased steadily to 2 ng/g (day 14), indicating a sustained-release behavior
from brain E2-Q+. In contrast, plasma E2 concentrations increased to only 0.28
0.1 ng/ml (day 1) and steadily declined thereafter.
Figure 7 shows the time-concentration profiles of E2-Q+ (upper panel)
and E2 (lower panel) in liver and fat tissues at various times following
administration of the E2-CDS. Both E2-Q+ and E2 were detected in these
tissues throughout the time-course studied. As expected, these tissues
showed rapid clearance of E2-Q+ as well as E2. The E2-Q+ concentrations
decreased from 77 10 ng/g and 71.7 19.5 ng/g (day 1) to 5.4 1.3 ng/g and
1.9 0.5 ng/g (day 14) in liver and fat, respectively. This indicated a reduction

79
in concentrations of greater than 83% and 87% from 1 to 7 days and 93% to
98% by 14 days after drug administration in liver and fat, respectively.
Similarly, E2 concentrations in these tissues fell by more than 84% and
80% from day 1 to day 7, and by 14 days after drug administration, the E2
concentrations decreased by 95% and 90%, respectively.
Figure 8 shows the time-concentration profiles in kidney, heart, lung,
and anterior pituitary concentrations of E2-Q+ (upper panel) and E2 (lower
panel) at various times following administration of the E2-CDS. The E2-Q+
concentrations in these tissues initially increased to 1906 131,1047 106, 748
28, and 407.6 50.6 ng/g in heart, lung, kidney, and anterior pituitary,
respectively. These E2-Q+ levels decreased rapidly by more than 76%, 79%,
74%, and 80% by day 7 in these 4 tissues, respectively. By 14 days after drug
administration, E2-Q+ concentrations decreased by greater than 98% in heart
and lung, 96% in kidney, and 93% in anterior pituitary. Despite high initial
concentrations of E2-Q+ in these peripheral tissues, brain levels of E2-Q+
exceeded E2-Q+ levels of these tissues by 1.5- to 3-fold at 14 days after
administration of the E2-CDS.
Estradiol concentrations in heart, lung, kidney, and anterior pituitary
(Figure 8; lower panel) were similarly elevated on day 1 but decreased rapidly
by 67% in heart, 83% in lung, 81% in kidney, and 86% in anterior pituitary by
day 7. From day 1 to day 14, the E2 levels in these tissues decreased by more
than 95% of the initial concentrations.
Figure 9 depicts brain (upper panels) and anterior pituitary (lower
panels) contents of E2-Q+ and E2 at various times after administration of the
E2-CDS. Following a single injection of the E2-CDS, the brain E2-Q+ content
was 635 28,340 23, and 77 3.9 ng/brain at 1,7 and 14 days, respectively.

80
By contrast, the anterior pituitary content of E2-Q+ was only 1/260 to 1/170 of
that observed in the brain.
Similarly, the brain E2 content was 15.8 0.9,10.4 0.8, and 3.2 0.1
ng/brain at 1, 7 and 14 days following administration, respectively, while the
anterior pituitary E2 content was 0.47 0.04, 0.06 0.004, and 0.035 0.006
ng/anterior pituitary at these sampling times. As such, the anterior pituitary
E2 content was only 1/90 to 1/34 of that observed in the brain throughout the
time-course studied. Thus, the absolute amounts (contents) of E2 and E2-Q+
were many fold higher in the brain even though the anterior pituitary
concentrations of E2 and E2-Q+ were initially higher than in the brain.
Discussion
The results of this single-dose distribution study demonstrated that
both E2-Q+ and E2, two metabolites of the E2-CDS, were present in all tissues
analyzed up to 14 days (the last sampling time) after treatment of male rats
with the E2-CDS. Moreover, over the time-course studied, the distribution
profiles indicated that: a) regardless of the tissues evaluated, E2-Q+ levels
were many fold higher than E2 levels at each time point in a particular tissue,
indicating a slow rate of hydrolysis of E2-Q+ to E2; b) the increased
brain/plasma ratios of E2-Q+ as well as E2, confirmed that "locking" of the
charged moiety, E2-Q+, into the brain had occurred; and c) E2-Q+ is retained in
the CNS tissue but is rapidly cleared from the peripheral tissues, an
observation which is predicted by the inherent physicochemical properties of
the delivery system.
Brain-distribution profile revealed that E2-Q+, the quaternary form of
the delivery system, persists in the brain with a half-life of about 8 days, but it

81
is rapidly cleared from the periphery. This is in accordance with that reported
in other studies (Mullersman et al., 1988). The half-life of the lipophilic E2-
CDS in brain tissue is only 29.2 min (Bodor et al., 1987), indicating rapid
oxidation of the delivery system to E2-Q+. From this store of E2-Q+, E2 can be
slowly released chronically in the brain through nonspecific hydrolysis.
As predicted from the physicochemical properties of E2-CDS as well as
previous reports on brain levels of E2-Q+ (Boder et al., 1987; Mullersman et
al., 1988; Simpkins et al., 1986), this "locked-in" form of the E2-CDS reached a
52-fold higher concentration in the brain than plasma by day 1 after treatment
with the E2-CDS, and these brain-blood ratios increased to 132-fold at day 7
and to about 170-fold by 14 days. Furthermore, from day 1 through 14, the
content of E2-Q+ in the brain was 6- to 23-times the content of E2-Q+ in the
blood. Thus, a portion of the E2-Q+ found in plasma may arise from brain
stores of the compounds. E2-Q+ can be cleared from the brain by bulk flow of
cerebrospinal fluid (Boder & Brewster, 1983; Schanker, 1965).
Brain E2 concentrations were similarly elevated relative to plasma.
Estradiol achieved a 28-fold higher concentration in the brain than plasma by
day 1 and this ratio increased to more than 50-fold at day 7 and remained at
37-fold by 14 days. Additionally, throughout the time-course studied, the
brain E2 content was 3- to 6-times the content of E2 in the blood. These
observations indicated that brain E2 is continuously produced, and as such the
steady-state brain E2 concentration is dependent on its rate of production from
the E2-Q+ and its rate of elimination from the brain by local metabolism
and/or redistribution down a concentration gradient to the plasma. Brain
stores of E2 could contribute to plasma levels through its partitioning to the
periphery down a large concentration gradient.

82
Levels of E2-Q+ in the anterior pituitary were surprisingly high on day
1 then dropped rapidly to below that of brain levels by day 7 and steadily
decreased thereafter throughout the observation period. This initial rise in
E2-Q+ levels may be attributed to increased anterior pituitary uptake of the E2-
CDS followed by its rapid metabolism and clearance. Furthermore, the
relative elevation of E2-Q+ as well as E2, from day 7 to day 14, in anterior
pituitary may be caused by the anatomical relationship between the
hypothalamus and anterior pituitary gland. Estradiol released upon the
hydrolysis of E2-Q+, or the E2-Q+ itself, which is locked into brain, could be
delivered directly to the anterior pituitary by the capillary plexus of the
hypophyseal portal system. These capillaries in the median eminence lack
features of other brain capillaries and hence are not part of the blood-brain
barrier (Traystman, 1983). Thus, the median eminence would not be expected
to prevent the efflux of E2-Q+ from the brain, and thus transfer of E2-Q+ to the
anterior pituitary can be expected.
High levels of E2-Q+ seen in the kidney are likely because this organ is a
major site for the elimination of all metabolites of the E2-CDS. However, the
reasons for initial high levels of E2-Q+ in the lung and heart tissues of the
male rat are not clear. We speculate that since these organs receive high
blood flow, a substantial amount of the E2-CDS is delivered to and taken up
by these tissues initially. Additionally, despite higher E2-Q+ levels in heart
relative to those of lung, kidney, and anterior pituitary, the heart E2
concentrations were lower than these 3 tissues. Perhaps, a slow rate of
hydrolysis of E2-Q+, or a slower E2 metabolism, in heart tissue could
contribute to the higher levels of E2-Q+ and lower levels of E2 in this tissue.
This single-dose pharmacokinetic study supports the previous
observations of prolonged pharmacodynamic effects of the E2-CDS following

83
administration of a single dose. LH secretion in castrated male rats was
suppressed for greater than 21 days (Estes et al., 1987b; Simpkins et al., 1986),
sexual copulatory behavior was stimulated for 28 days (Anderson et al.,
1987a), and body weight was suppressed for 36 days (Estes et al., 1988;
Simpkins et al., 1988) after doses of E2-CDS of 1 to 3 mg/kg. These prolonged
effects of the E2-CDS are consistent with the observation here of the
accumulation of E2-Q+, the oxidized form of the delivery system, in the rat
brain and its long half-life (t 1/2 = 8 days) in this tissue. From this store of E2-
Q+, E2 is released through slow hydrolysis and exhibits a half-life similar to
that of E2-Q+.
Finally, since we evaluated the tissue distribution of E2-CDS
metabolites in intact male rats, determination of E2-CDS metabolites in the
testes, prostate, and seminal vesicle tissues would have certainly added more
valuable informations to the results presented in this chapter. Unfortunately,
at the time of experimental investigation, these tissues were not collected for
analysis of E2-CDS metabolites. Certainly, in future studies involving E2-CDS,
these tissues need to be evaluated for E2-Q+ and E2 distribution and clearance.
Furthermore, the question of blood-testes barrier must be addressed. This
would be of great interest to find out whether E2-Q+ is being "locked" in this
tissue similar to CNS tissue or behaves like the rest of peripheral tissues.
However, effects of the E2-CDS on weights of these androgen-responsive
tissues were examined and the are presented in Chapter 8.
In conclusion, these observations are consistent with the proposed
mechanism of the redox-based E2-CDS and the contribution of the BBB to the
chronic retention of the charged, hydrophilic E2-Q+ in the rat brain. This
observed tissue distribution pattern of E2-Q+ and E2 may explain the long
term pharmacological effects of the E2-CDS in the male rat.

84
Figure 6. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on brain
(upper panels) and plasma (lower panels) concentrations of E2-Q+
and E2 in intact male rats. Intact male rats were injected with a
single iv dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1,
7, or 14 days after treatment. Whole brain tissue and plasma were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Each point represents the group mean SEM (n = 6-7
rats for each time point).

TIME (days) TIME (days)
Plasma Cone, (ng/ml) Brain Cone, (ng/g)
oo
on

Tissue Cone, (ng/g) Tissue Cone, (ng/g)
86
Figure 7. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on liver and
fat concentrations of E2-Q+ (upper panels) and E2 (lower panel) in
intact male rats. Intact male rats were injected with a single iv
dose of 1.0 mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14
days after treatment. Each point represents the group mean SEM
(n = 6-7 rats for each time point).

Tissue Cone, (ng/g) Tissue Cone, (ng/g)
87
Figure 8. Effects of a single iv dose of the E2-CDS (1.0 mg/kg) on kidney,
heart, lung, and anterior pituitary concentrations of E2-Q+ (upper
panel) and E2 (lower panel) in intact male rats. Intact male rats
were injected with a single iv dose of 1.0 mg E2-CDS/kg bw and
killed 1, 7, or 14 days after treatment. Each point represents the
group mean SEM (n = 6-7 rats for each time point).

88
Figure 9. Brain (upper panels) and anterior pituitary (lower panels) contents
of the E2-Q+ and E2 following a single iv dose of the E2-CDS (1.0
mg/kg). Intact male rats were injected with a single iv dose of 1.0
mg E2-CDS/kg bw and killed by decapitation 1, 7, or 14 days after
treatment. Whole brain tissue and the anterior pituitary were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Each point represents the group mean SEM (n = 6-7
rats for each time point). These calculations assumed a brain wet
weight of 2 grams (based on personal experience and knowledge).

Ant Pit Content (ng)
o
Ant Pit Content (ng)
Brain Content (ng)
N) *. Os 00
o o o o
o o o o
I I 1 I I 1 I
tfl
N>
I
o
+
Brain Content (ng)
i ro
Ol O U1 o
I I I I I I I
IfS
00
KO

CHAPTER 6
DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN FEMALE RATS
Introduction
Brain-enhanced delivery to and the sustained release of E2 in the brain
may be potentially useful for the effective treatments of vasomotor hot
flushes, prostatic adenocarcinoma, and fertility regulation. The natural
estrogen, 17 (3-E2, administered as either a valerate, benzoate, or dienanthate
is effective in alleviating hot flushes (Campbell & Whitehead, 1977; Doring,
1976; Dusterberg & Nishino, 1982; Klopper, 1976; Lauritzen, 1973). Likewise,
the synthetic alkylating estrogen, ethinyl E2, is a potent and an effective
contraceptive agent (Burkman, 1988). However, these estrogen compounds
act upon all steroid-responsive tissues which limits their therapeutic efficacy.
Moreover, since these steroids equilibrate among all body tissues, only a
fraction of the administered dose accumulates at or near the site of action in
the CNS. These properties of the estrogen then force either frequent dosing,
or the administration of a depot form of the steroid, to maintain
therapeutically effective levels in the brain. These treatment strategies lead to
sustained increases in peripheral estrogen levels, and thus augment the risk
of peripheral toxicities. The redox-based E2-CDS may be potentially useful in
the effective treatments of brain diseases by providing sustained and
sufficient E2 to the brain while avoiding peripheral toxicities.
The present studies were undertaken to determine if the E2-CDS
behaves as predicted on the basis of the physicochemical properties designed
90

91
into its structure. We determined (1) the effects of E2-CDS dose on tissue
concentrations of E2-Q+ and E2, (2) the effects of E2-CDS dose on the rate of
oxidation of E2-CDS to E2-Q+ and hydrolysis of E2-Q+ to E2, and (3) the effects
of E2-CDS dose on the clearance of E2-Q+ and E2 in a variety of tissues in the
female rat. More specifically, our objective was to analyze quantitatively both
E2-Q+ and E2 (two metabolites of the E2-CDS) in brain, hypothalamus, anterior
pituitary, kidney, lung, heart, liver, fat, uterus, and plasma following a single
iv injection of one of several different doses of the E2-CDS over a 28-day time
course in ovariectomized (OVX) rats.
The rationale for using OVX female rat model in this study was
twofold: 1) this animal model exhibits very low endogenous estrogen levels
since both ovaries have been removed, thus allowing reliable determination
of tissue metabolites of the E2-CDS and 2) the results from this animal model
would allow us to make comparison with the results obtained from the male
rat model (Chapter 5) with regard to E2-CDS distribution.
Materials and Methods
Experiment 1
To evaluate the dose- and time-dependent effects of E2-CDS on the
tissue distribution of E2-Q+ and E2 in female rats, all animals were bilaterally
ovariectomized (OVX) under metofane anesthesia. All experiments were
initiated exactly 2 weeks after ovariectomy. On day 15 after ovariectomy, rats
(7 per group per each time point) were administered a single iv injection (tail
vein) of the E2-CDS at doses of 0 (HPCD), 0.01, 0.1, or 1.0 mg/kg body weight or
E2 at a dose of 0.7 mg/kg (equimolar to the 1.0-mg/kg dose of E2-CDS).

92
Rats (7 per group) were killed by decapitation 1, 7,14, 21, or 28 days after
the drug administration and the trunk blood was collected in heparinized
tubes. The blood was centrifuged and the plasma separated and stored at -
20C until hormone analysis. Tissues (whole brain, hypothalamus, anterior
pituitary, kidney, lung, heart, liver, fat, and uterus) were dissected
immediately following decapitation, rinsed in ice-cold saline, stripped of
surrounding connective tissue where necessary, blotted dry on paper, and
then stored at -80C until hormone analysis.
Tissue samples of known wet weight at a concentration of 1 mg/20 pi
solvent were processed and assayed for E2-Q+ and E2 by the method described
in Chapter 4 (Rahimy et al., 1989a). Also, tissue homogenates or plasma from
HPCD-treated rats were analyzed to determine residual E2 concentrations and
thereby served as our estimate of hormone background.
Experiment 2
To accurately assess the kinetics of E2-CDS within the general
circulation, an acute 7-day time-course study with frequent blood sampling
was undertaken. To facilitate frequent blood sampling from unrestrained
animals, a group of OVX rats were equipped with Silastic catheter (i.d. 0.5
mm, o.d. 1 mm). The catheter was positioned in the right atrium via the
external jugular vein under pentobarbital anesthesia, according to the
procedure described by Steffens (1969).
After recovery from the surgical procedure (usually one week, 3 weeks
after OVX), rats were administered iv (via tail vein) with 1.0 mg/kg dose of
the E2-CDS. Immediately after drug treatment, animals were transferred to
special sampling chambers for serial blood sampling. Blood samples (1 ml)

93
were removed at 0.5,1, 2, 4, 8,12, 24, 48, 96, and 168 hrs post-injection. At each
sampling time, the blood was centrifuged and the plasma collected for E2-Q+
and E2 analysis. Red cells were then resuspended in 0.5 ml heparinized saline
(40 units/ml) and returned to each respective animal before the next blood
sample. The volume of resuspended red cell solution was approximately 1
ml. It should be emphasized that although attempts were made to minimize
the occurrence of potential problems associated with the design of this
experiment, certain problems were unavoidable and thus should be
mentioned. During the initial 12-hrs of the time course of this experiment,
approximately 3 ml of plasma (i.e. 1/2 the total blood volume withdrawn)
were collected from each animal. This volume of plasma was needed to
analyze both metabolites of the E2-CDS in plasma. However, this volume
represents approximately 25% of total blood volume in an adult rat. And
since red cells were resuspended in equal volume of heparinized saline and
returned to each animal before the next blood sample, this procedure might
decrease the plasma concentrations of the E2-CDS metabolites.
Coat-A-Count Estradiol kits, a solid-phase [125I]-radioimmunoassay,
designed for the quantitative measurement of E2 in serum were used for the
assay of E2 in all tissue samples. All the purified dried E2 unknowns were
reconstituted in 300 |il of the assay buffer (kit Zero Calibrator) and assayed in
duplicate by the RIA. The intraassay and interassay coefficients of variation
for E2 were 1.56 and 6.1%, respectively. All samples were determined in 14
RIA runs.

Results
Experiment 1
To estimate the extent of in vivo oxidation of E2-CDS to E2-Q+, we
determined for each tissue the magnitude of increase in E2-Q+ concentrations
over the 100-fold increase in E2-CDS dose, at day 1 after injection (the first
sampling time). The enzymatic oxidation of E2-CDS to E2-Q+ showed a clear
dose dependency in brain, hypothalamus, plasma, kidney, lung, heart, liver,
and fat tissues (Table 4). This dose-related oxidation ranged from 73-fold in
liver to 176-fold in whole brain tissue over the 100-fold increase in E2-CDS
dose administered. The uterus was an exception and showed only a 21-fold
increase in E2-Q+ concentrations over the 100-fold increase in E2-CDS dose
(Table 4).
The in vivo rate of hydrolysis of E2-Q+ to E2 was estimated for each
tissue at each dose of E2-CDS administered by determining the ratio of E2 to
E2-Q+ on the first sampling day (day 1, Table 4). This ratio remained constant
over a 100-fold dose range for the hypothalamus, kidney, heart, and uterus; it
decreased moderately (less than 50%) over the 100-fold dose range for the
brain, lung, and liver and decreased precipitously for plasma and fat tissue
(Table 4).
At 1 day after administration of E2-CDS, all tissues showed a dose-
dependent increase in concentrations of E2-Q+ and E2. Furthermore, the
concentration-time profiles revealed a gradual decline in concentrations of
E2-Q+ and E2 in whole brain (Figure 10) as well as in hypothalamus (Figure
11), with ti/2 = 8-9 days. In contrast, both E2-Q+ and E2 were rapidly cleared

95
from plasma (Figure 12), liver (Figure 13), fat (Figure 14), and anterior
pituitary, kidney, lung, heart, and uterus (Tables 5 and 6). By 28 days (the last
sampling time) after a single injection of 1.0 mg E2-CDS/kg, the E2-Q+
concentrations remained elevated at 9.8 0.7 ng/g wet tissue (mean SEM) in
brain (Figure 10, left column, upper panel) and 10.6 0.2 ng/g in
hypothalamus (Figure 11, left column, upper panel). In contrast, peripheral
tissues concentrations of E2-Q+ were reduced to 2.9 0.1 ng/g in anterior
pituitary, 5.2 2.2 ng/g in kidney, 2.9 0.5 ng/g in lung, 1.7 0.3 ng/g in heart,
2.5 0.7 ng/g in uterus (Table 5, % reduction); and E2-Q+ values were
undetectable in plasma, liver, and fat (Figures 12-14, left columns, upper
panels) 28 days after administration of 1.0 mg E2-CDS/kg dose.
Similarly, E2 concentrations were maintained relatively high in whole
brain (Figure 10, right column of panels) and in hypothalamus (Figure 11,
right column of panels); however, E2 concentrations in peripheral tissues
(except for anterior pituitary and plasma) fell by more than 80% by day 7, and
by 97% by day 21, and were undetectable by day 28 (Table 6).
In contrast with E2 concentrations achieved following E2-CDS
administration, E2 levels in tissues following equimolar E2 administration
were remarkably low (Table 7) and in all tissues examined, the clearance of E2
was rapid (Table 7). A comparison of the tissue levels of E2 achieved at day 1
revealed that following E2-CDS administration, brain (Figure 10) and
hypothalamus (Figure 11) levels of E2 were 88- and 22-fold greater,
respectively, than levels observed in these tissues following 17 P-E2
administration (Table 7). By 7 days after treatment, the E2-CDS produced
brain and hypothalamic E2 concentrations that were 182- and 55-fold greater,
respectively, than those achieved by an equimolar 17 P-E2 dose.

96
Experiment 2
Figure 15 depicts the concentration-time profiles of the E2-CDS
metabolites (E2-Q+, upper panel; E2, lower panel) in plasma of OVX rats. By 30
min after administration of the E2-CDS, plasma E2-Q+ increased to (61.9 3.8
ng/ml). The E2*Q+ levels decreased by 50% at 8 hrs and by greater than 88% at
24 hrs after E2-CDS treatment. Kinetic analysis revealed that the plasma E2-Q+
concentration-time profile fits a sum of two exponentials. The half-lives of
these two phases were ti/2 = 8.16 and ti/2 = 70.38 hrs respectively in plasma.
The area under curve (AUQ) was 835.25 ng/ml hr for the time-course
studied.
Similarly, plasma E2 concentrations were increased to 1.9 0.08 ng/ml
after 30 min of E2-CDS administration. The plasma E2 levels decreased by
50% at 3 hrs and by greater than 91% at 24 hrs after E2-CDS treatment. The
plasma E2 concentration-time profile was best fitted in a sum of three
exponentials. The half-lives of the three phases were ti/2 = 0.14, ti/2 = 2.35,
ti/2 = 38.99 hrs, respectively. The area under curve (AUQ) was 23.85 ng/ml
hr for the time-course studied. The third half-life indicated the presence of
deep compartment (most likely the brain) which slowly releases E2 in plasma.
Discussion
This detailed dose-distribution and time-course study demonstrates
that (i) the enzymatic oxidation of E2-CDS to E2-Q+ is dose dependent, and
with the possible exception of the uterus, the oxidation is not saturable over
the 100-fold dose range tested; (ii) the hydrolysis of E2-Q+ to E2 was dependent
upon the tissues analyzed and appeared to be saturable only in plasma and fat

97
and, to a lesser extent, in brain, lung, and liver; and (iii) the disappearance of
both E2-Q+ and E2 was slow in brain tissue and rapid in all peripheral tissues
tested. Collectively, these data are consistent with the expected behavior of
the E2-CDS (Bodor et al., 1987).
Dose and time-course profiles revealed that E2-Q+ persists in brain
tissue as well as in hypothalamus, with a ti/2 = 8-9 days, but it is rapidly
cleared from the periphery. The previous single-dose distribution study in
intact male rats (Chapter 5) demonstrated a similar half-life for E2-Q+ in brain
tissue. These estimates of the half-life of E2-Q+ in the brain are in accordance
with reports which utilized different analytical techniques and E2-CDS doses
(Mullersman et al., 1988). The slow clearance of the E2-CDS metabolite, E2-Q+,
from the CNS tissue appears to be independent of dose since similar values
have been obtained in studies using doses of E2-CDS ranging from 0.01 mg/kg
(present report) to 15 mg/kg dose (Mullersman et al., 1988). Further, the long
half-life of E2-Q+ in brain tissue does not appear to be an artifact of its
sustained production from E2-CDS since the half-life of the delivery system
itself in brain tissue is only 29.2 min, indicating rapid oxidation to E2-Q+
(Bodor et al., 1987). Thus, as predicted based on the physicochemical
properties of the E2-CDS (Bodor et al., 1987), the unique features of the BBB
appear to contribute to the chronic retention by the brain of the charged,
hydrophilic E2-Q+.
The anterior pituitary exhibited slower elimination of the metabolites
of E2-CDS (E2-Q+ & E2) than other peripheral tissues. By 7 days following
administration of a 1.0-mg E2-CDS dose, the hypothalamic-anterior pituitary
E2-Q+ ratio was 1.3 and then increased to about 4-fold by 28 days. Similarly,
the hypothalamic-anterior pituitary E2 ratio was 1.2 on day 7 and this ratio
was maintained throughout the 28-day time course. This relative persistency

98
of both E2-Q+ and E2 in the anterior pituitary may be caused by the anatomical
relationship between the hypothalamus and the pituitary gland. Estradiol
released upon the hydrolysis of E2-Q+, or the E2-Q+ itself, which is locked into
brain, could be delivered directly to the pituitary by the capillary plexus of the
hypophyseal portal system. These capillaries in the median eminence lack
features of other brain capillaries and hence are not part of the BBB
(Traystman, 1983). Thus, the median eminence would not be expected to
prevent the efflux of E2-Q+ from brain, and transfer of E2-Q+ to anterior
pituitary can be expected.
Plasma also showed, after day 1, a residual but detectable E2
concentration throughout the time course of the 1.0-mg E2-CDS dose and
through the 21-day time course of the 0.1-mg dose. This prolonged and
residual E2 in plasma is likely to be the result of a continuous redistribution
of E2 liberated from E2-Q+ in the brain or other tissues down its concentration
gradient into the general circulation.
The oxidation of E2-CDS to E2-Q+ in uterine tissue did not appear to be
dose dependent, since a 100-fold increase in dose resulted in a 21-fold increase
in E2-Q+ concentration. This observation is perhaps, in part, an artifact of the
observed uterine hypertrophy in animals treated with the E2-CDS (Anderson
et al., 1988a). At day 1 after E2-CDS treatment, a 100-fold increase in E2-CDS
dose resulted in 54% increase in uterine weight (Chapter 7, Table 8), this
might reduce the values for E2-Q+ when normalized for tissue weight.
However, when the extent of oxidation was estimated per tissue, we observed
only a 32-fold increase over the 100-fold dose range examined. Additionally,
since blood flow to the uterus of OVX rats is low, the expected rapid oxidation
of E2-CDS to E2-Q+ would occur less likely in that tissue. Finally, we cannot
rule out the possibility that in the uterus of OVX rats, the enzymatic oxidation

99
of E2-CDS to E2-Q+ is a saturable process and is thereby independent of the
dose of E2-CDS administered.
To demonstrate further the preferential deposition and retention of
estrogen in the CNS with the E2-CDS, one dose of 17 p-E2 (equimolar to the
1.0-mg E2-CDS dose) was also studied. As shown in Table 7, E2 concentrations
in the CNS tissues of rats treated with 17 |3-E2 were slightly increased on day 1,
and were just above the detection limits of the assay at 7 days. In contrast, the
1.0 mg E2-CDS dose resulted in brain E2 concentrations that were 81- and 182-
fold greater than those achieved following 17 (3-E2 injection at 1 and 7 days,
respectively. These data demonstrated that the E2-CDS is much more
effective than 17 f)-E2 itself in delivering and retaining the estrogen in the
brain.
Collectively, these observations are consistent with the proposal that E2
can be preferentially delivered to the brain using a redox-based chemical
delivery system, an inert molecule which requires several steps in its
conversion to the parent drug (Bodor et al., 1981; Bodor & Brewster, 1983).
The multiple, facile enzymatic conversions including oxidation and
hydrolytic cleavage may not only lead to preferential E2 delivery and
sustained release/effects, but may also act to decrease the toxicity of the drug.
A preferential and sustained CNS estrogen delivery can be potentially useful
since estrogens are known to influence a variety of CNS functions (McEwen,
1988; Maggi & Perez, 1985).

100
Figure 10. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in whole brain of ovariectomized rats.
Animals received a single iv (tail vein) injection of the E2-CDS on
day 0 at a dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle
panels), or 0.01 mg/kg (lower panels). Animals were killed 1, 7,14,
21, or 28 days after the drug administration. Tissue samples of
known wet weight at a concentration of 1 mg/20 pi solvent were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Also, tissue homogenates from HPCD-treated rats were
analyzed for E2 hormone background. Represented are means
SEM for n = 7 rats per group per sampling time.

ng/g ng/g ng/g
101
BRAIN E2-Q+
BRAIN E2

102
Figure 11. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in hypothalamus of ovariectomized rats.
Animals received a single iv (tail vein) injection of the E2-CDS on
day 0 at a dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle
panels), or 0.01 mg/kg (lower panels). Animals were killed 1, 7,14,
21, or 28 days after the drug administration. Tissue samples of
known wet weight at a concentration of 1 mg/20 pi solvent were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Also, tissue homogenates from HPCD-treated rats were
analyzed for E2 hormone background. Represented are means
SEM for n = 7 rats per group per sampling time.

ng/g ng/g ng/g
103
HYPOTHAL. E2-Q+
HYPOTHAL. E2

104
Figure 12. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in plasma of ovariectomized rats. Animals
received a single iv (tail vein) injection of the E2-CDS on day 0 at a
dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle panels), or
0.01 mg/kg (lower panels). Animals were killed 1, 7,14, 21, or 28
days after the drug administration. Plasma samples of known
aliquot were processed and assayed for E2-Q+ and E2 by the method
described in Chapter 4. Also, plasma samples from HPCD-treated
rats were analyzed for E2 hormone background. Represented are
means SEM for n = 7 rats per group per sampling time. *
Indicates below the sensitivity limit of the assay.

ng/ml ng/ml ng/ml
105
PLASMA E2-Q+ PLASMA E2

106
Figure 13. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in liver of ovariectomized rats. Animals
received a single iv (tail vein) injection of the E2-CDS on day 0 at a
dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle panels), or
0.01 mg/kg (lower panels). Animals were killed 1, 7,14, 21, or 28
days after the drug administration. Tissue samples of known wet
weight at a concentration of 1 mg/20 pi solvent were processed
and assayed for E2-Q+ and E2 by the method described in Chapter 4.
Also, tissue homogenates from HPCD-treated rats were analyzed
for E2 hormone background. Represented are means SEM for n
= 7 rats per group per sampling time. Indicates below the
sensitivity limit of the assay.

ng/g ng/g
107
LIVER E2-Q+
LIVER E2

108
Figure 14. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in fat of ovariectomized rats. Animals received
a single iv (tail vein) injection of the E2-CDS on day 0 at a dose of
1.0 mg/kg (upper panels), 0.1 mg/kg (middle panels), or 0.01
mg/kg (lower panels). Animals were killed 1, 7,14, 21, or 28 days
after the drug administration. Tissue samples of known wet
weight at a concentration of 1 mg/20 |il solvent were processed
and assayed for E2-Q+ and E2 by the method described in Chapter 4.
Also, tissue homogenates from HPCD-treated rats were analyzed
for E2 hormone background. Represented are means SEM for n
= 7 rats per group per sampling time. Indicates below the
sensitivity limit of the assay.

ng/g ng/g ng/g
109
FAT E2-Q+
FAT E2

Plasma Cone, (ng/ml) Plasma Cone, (ng/ml)
110
Figure 15. Effects of a single iv dose of the E2-CDS (1.0 mg/kg bw) on plasma
E2-Q+ (upper panel) or plasma E2 (lower panel) concentrations in
ovariectomized rats. Animals received a single iv (tail vein)
injection of the E2-CDS on day 0 at a dose of 1.0 mg/kg.
Immediately after drug treatment, animals were transferred to
sampling chambers and blood samples were removed at 0.5,1, 2, 4,
8,12, 24, 48, 96, and 168 hrs postinjection. Plasma samples were
assayed for E2-Q+ and E2 as described in Chapter 4. Represented
are means SEM for n = 5 rats per group per sampling time.

Ill
Table 4: Effects of Dose on the Extent of Oxidation and Hydrolysis of E2-CDS
in a Variety of Tissues in vivo
Tissue
Oxidation
Hvdrolvsisb
(-fold increase)3
0.01
E2-CDS Dose (mg/kg)
0.1
1.0
Brain
176
24
23
13
Hypothalamus
112
22
23
19
Ante. Pituitary
NDc
40
31
ND
Plasma
147
14
4
2
Kidney
103
29
22
27
Lung
116
47
32
24
Heart
125
25
21
29
Liver
73
13
15
6
Fat
77
35
22
7
Uterus
21
31
32
26
a Fold increase in E2-Q+ concentrations over a 100-fold increase in the E2-
CDS dose at day 1 after injection (the first sampling time),
b Average % hydrolysis (fraction of E2-Q+ hydrolyzed + total E2-Q+ x 100)
on the first sampling time (day 1).
c Not determined due to incorrect extraction of the tissue at day 1.

112
Table 5: Effects of the E2-CDS on the Clearance of E2-Q+ from a Variety of
Tissues
Tissue
Dose
Davs after treatment
(mg/kg)
la
7b
14b
21b
28b
Brain
1.0
329.97
42
82
92
97
Hypothalamus
1.0
310.58
16
80
88
96
Anterior
Pituitary
1.0
NDc
ND
ND
ND
ND
Plasma
1.0
21.11
96
99
>99
UDd
Kidney
1.0
487.56
80
96
98
>99
Lung
1.0
716.94
83
98
99
>99
Heart
1.0
910.39
82
98
>99
>99
Liver
1.0
374.57
94
98
>99
UD
Fat
1.0
176.11
97
99
>99
UD
Uterus
1.0
92.58
89
93
97
>98
a Initial concentrations of E2-Q+ (ng/g wet tissue or ng/ml) 1 day after
administration of the E2-CDS.
b % reduction in E2-Q+ concentrations at various times after E2-CDS
treatment relative to the initial corresponding values,
c Not determined due to incorrect extraction of the tissue at day 1.
d Undetectable (below the sensitivity of RIA for E2).

113
Table 6: Effects of the E2-CDS on the Clearance of E2 from a Variety of Tissues
Tissue
Dose
Davs after treatment
(mg/kg)
la
7b
14b
21b
28b
Brain
1.0
26.63
40
71
91
96
Hypothalamus
1.0
25.32
24
68
77
91
Anterior
Pituitary
1.0
NDc
ND
ND
ND
ND
Plasma
1.0
0.32
78
87
93
95
Kidney
1.0
135.17
80
98
>99
UDd
Lung
1.0
167.67
95
99
>99
>99
Heart
1.0
156.03
82
99
>99
>99
Liver
1.0
8.43
90
98
>99
UD
Fat
1.0
2.32
84
97
>99
>99
Uterus
1.0
23.07
93
94
97
>99
a Initial concentrations of E2 (ng/g wet tissue or ng/ml) 1 day after
administration of the E2-CDS.
b % reduction in E2 concentrations at various times after E2-CDS
treatment relative to the initial corresponding values,
c Not determined due to incorrect extraction of the tissue at day 1.
d Undetectable (below the sensitivity of RIA for E2).

114
Table 7: Effects of an Equimolar Dose of E2 (0.7 mg/kg) on the Tissue
Concentrations of E2a
Tissue
Vehicle (HPCD)b
Days after treatment
lc
7c
Brain
0.05 0.03
0.33 0.04
0.09 0.07
Hypothalamus
0.03 0.01
1.16 0.14
0.35 0.08
Ante. Pituitary
0.02 0.01
0.99 0.07
0.28 0.02
Plasma
UDd
0.03 0.01
0.02 0.01
Kidney
UD
0.55 0.13
0.04 0.01
Lung
UD
0.26 0.30
0.01 0.01
Heart
UD
0.14 0.04
0.01 0.01
Liver
0.02 0.01
0.80 0.20
0.11 0.04
Fat
0.05 0.03
1.12 0.29
0.11 0.05
Uterus
UD
2.08 0.65
0.17 0.14
a This dose of E2 (0.7 mg/kg) is equimolar to 1.0 mg E2-CDS/kg dose,
b Residual E2 concentrations (ng/g wet tissue or ng/ml; mean SEM).
c E2 concentrations (ng/g wet tissue or ng/ml; mean SEM) at various
times following administration of E2.
d Undetectable (below the sensitivity of RIA for E2).
Note: For the determination of E2 concentration in this table, we used 200 to
300 mg tissue to obtain a detectable E2 baseline in the RIA since tissue
E2 concentration was very low following 17 P-E2 injection. And since
our assay methodology was optimized for up to 100 mg tissue, using
excess amount of tissue in the assay might result in unreliable E2
values following 17 P-E2 administration.

CHAPTER 7
EVALUATION OF THE PHARMACODYNAMIC EFFECTS OF E2-CDS IN
OVARIECTOMIZED FEMALE RATS
Introduction
Estrogen hormones have been shown to influence a myriad of CNS
processes including reproductive parameters, i.e., neuroendocrine
modulation of reproductive cycle (Kalra & Kalra, 1989; Maggi & Perez, 1985;
Plant, 1986) and stimulation of sexual behaviors (Beyer et al., 1976;
Christensen & Clemens, 1974) as well as non-reproductive parameters such as
regulation of neurotransmissions involved in sensorimotor functions,
mood, and learning tasks (Fillit et al., 1986; McEwen et al., 1984; Smith, 1989).
These diverse actions of estrogens on the CNS functions are of significant
therapeutic interest after menopause or ovariectomy when endogenous
estrogens decline. In these and certain other cases, since estrogen medications
are primarily used for their central actions, the preferential brain estrogen
delivery is not only beneficial but may produce safer and more potent natural
therapeutic agent.
The evaluation of the tissue distribution patterns of both E2-Q+ and E2
in intact male (Chapter 5; Rahimy et al., 1990a) as well as in OVX female rats
(Chapter 6; Rahimy et al., 1990b) substantiated the major aspect of the
proposed mechanism of this redox-based estrogen delivery system for the
brain. That is, the E2-CDS consistently demonstrated its predictive
pharmacokinetic behaviors including the preferential retention of E2-Q+ and
thus, E2 in the CNS tissue with ti/2 = 8-9 days, while simultaneously
115

116
accelerated the elimination of these metabolites from the peripheral tissues.
On the basis of these pharmacokinetic findings, the E2-CDS is expected to
exhibit pharmacodynamic responses with long duration of effects following a
single administration of the delivery system.
The present study was undertaken to determine whether the long half-
lives and the magnitude of E2-CDS metabolites in brain tissue (Chapters 5 &
6) correlate with the duration of pharmacodynamic effects. More specifically,
the objectives were (1) to assess the dose- and time-dependent effects of the E2-
CDS on brain-mediated responses, i.e. anterior pituitary hormones secretion
in OVX rats; (2) to compare E2-CDS with an equimolar dose of 17 P-E2; and (3)
to correlate the half-lives of the E2-CDS metabolites with the duration of
pharmacodynamic effects mediated by E2.
Materials and Methods
All the samples analyzed in this study were obtained from the animals
used and described in the preceding study (Chapter 6); this chapter presents
further data on evaluation of the pharmacodynamic responses of E2-CDS.
Briefly, rats were ovariectomized (OVX) and two weeks later were
administered a single iv injection of the E2-CDS at doses of 0 (HPCD), 0.01, 0.1,
or 1.0 mg/kg body weight or E2 at a dose of 0.7 mg/kg (equimolar to the 1.0 mg
E2-CDS dose). Animals (7 per group) were then killed by decapitation 1, 7,14,
21, or 28 days after the drug administration and plasma, anterior pituitary,
and uterine tissues were collected for subsequent analysis.
Plasma luteinizing hormone (LH), follicle-stimulating hormone (FSH),
growth hormone (GH), and prolactin (PRL) concentrations were measured in
duplicate by the RIA using NIDDK kits. Plasma LH, FSH and GH values are

117
expressed as ng/ml of either the LH-RP-2, FSH-RP-2, or GH-RP-2 reference
standard, respectively and PRL values are expressed as ng/ml of the PRL-RP-3
standard. The intra-assay coefficients of variation were 4.67%, 5.02%, 4.05%,
and 4.96% for LH, FSH, GH, and PRL assays, respectively. All the samples for
each hormone were assayed in a single run.
Results
The E2-CDS caused a dose- and time-dependent suppression of plasma
LH throughout the time-course studied (Figure 16). The maximum LH
reduction occurred at 7 days postinjection. At this time, LH was suppressed by
21, 46 and 86% relative to HPCD control at doses of 0.01, 0.1 and 1.0 mg E2-
CDS/kg, respectively (Figure 16). The plasma LH concentrations in animals
treated with 1.0 mg E2-CDS were significantly reduced by 56, 86, 72, and 56% at
1, 7,14, or 21 days, respectively and remained suppressed by greater than 35%
at 28 days after drug administration. By contrast, equimolar E2 dose (0.7
mg/kg) caused a transient reduction in LH concentrations of 27% on day 1
and 24% on day 7 which were not significantly different from time 0 values
(Figure 16).
Similarly, the E2-CDS caused a dose- and time-dependent suppression
of plasma FSH throughout the time-course studied (Figure 17). The
maximum FSH reduction occurred at 7 days postinjection. FSH was
suppressed by 14, 28, and 58% relative to control at doses of 0.01, 0.1 and 1.0
mg E2-CDS/kg, respectively, on day 7 (Figure 17). The plasma FSH
concentrations in animals treated with 1.0 mg E2-CDS were significantly
reduced by 37,58 and 20% at 1,7 or 14 days, respectively and by 7% (day 21) or
were at preinjection levels by 28 days after drug administration. By contrast,

118
equimolar E2 dose reduced plasma FSH by 27% at day 1 and 19% at day 7
(Figure 17).
Plasma concentrations of LH and FSH in animals treated with lower
doses of the E2-CDS (0.01 and 0.1 mg/kg) began to gradually increase after 7
days of drug administration (Figures 16 &17).
Plasma PRL concentrations in animals treated with 1.0 mg E2-CDS dose
were increased by 4-, 8-, 13-, and 8-fold at 1, 7,14 and 21 days, respectively or
were at preinjection levels by 28 days after drug administration (Figure 18,
upper panel). Lower doses of the E2-CDS did not effect PRL concentrations.
By contrast, the 0.7 mg/kg dose of E2 increased plasma PRL concentrations by
3-fold on day 1 and PRL returned to preinjection levels by day 7 after drug
administration (Figure 18, upper panel).
Plasma GH concentrations were not altered in response to E2 or E2-CDS
at any dose or time point evaluated (Figure 18, lower panel).
Anterior pituitary weights increased in a dose- and time-dependent
manner in response to E2-CDS administration (Table 8). With the lower
doses of the E2-CDS (0.01 and 0.1 mg/kg), pituitary weights were slightly
increased (22 to 32%) over control group weights by 14 days postinjection, but
they returned to control levels by day 21. However, the highest dose of the E2-
CDS increased pituitary weights significantly from day 7 to day 28 relative to
weights at time 0 and following treatment with lower doses of E2-CDS. The
maximum pituitary gland stimulation occurred at 14 days postinjection and
then pituitary weights began to decrease but remained elevated at 28 days after
the drug administration (Table 8).
Similarly, uterine weights showed a dose- and time-dependent increase
in response to E2-CDS administration (Table 8). Uterine weights were
increased by 20, 54, or 82% on day 1 following treatment with the E2-CDS at

119
0.01, 0.1, and 1.0 mg/kg doses, respectively. With the highest dose of the E2-
CDS (1.0 mg/kg), uterine weights were significantly increased by about 3-fold
on day 7 which then weights began to decrease but remained elevated at 28
days after the drug administration (Table 8). It should be noted that even at
the highest dose (1.0 mg E2-CDS/kg), uterine weights were less than those
typically observed in gonad-intact rats (500-625 mg).
An equivalent increase in anterior pituitary as well as uterine weights
was observed with E2 (0.7 mg/kg) compared to 1.0 mg E2-CDS dose on day 1.
However, by day 7 the effects of equimolar E2 were equivalent to the lowest
dose of E2-CDS (0.01 mg/kg).
Discussion
The results of this study demonstrated that the E2-CDS causes a dose-
and time-dependent suppression of gonadotropin (LH & FSH) secretion in
OVX rats with maximum reductions in plasma LH and FSH concentrations
occurring 7 days after E2-CDS administration. The time course of
gonadotropin suppression in OVX rats is comparable to that previously
observed for E2-CDS effects on other parameters and in other animal models.
We have reported long-term suppression of LH in castrated male rats (Estes et
al., 1987b; Simpkins et al., 1986), in OVX female rats (Anderson et al., 1988a),
and stimulation of masculine sexual behavior in castrated male rats for 28
days (Anderson et al., 1987a), and body weight alterations for 36 days
(Simpkins et al., 1988) following a single iv administration of the E2-CDS.
Sarkar et al. (1989) have reported on the suspension of estrous cycles in
female rats for 30 days following E2-CDS treatment. These prolonged
pharmacological effects further support the idea that the intermediate

120
metabolite of the E2-CDS, E2-Q+, is "locked" behind the BBB and there it
serves as a brain depot for E2 (Bodor, et al., 1987). From this store of E2-Q+, E2
is then slowly released through non-specific hydrolysis of the carrier,
resulting in sustained brain exposure to E2.
Since 17-substituted estrogens, such as the E2-CDS and E2-Q+, do not
effectively bind to E2 receptors (Dusterberg & Nishino, 1982; Janoko et al.,
1984), they are not likely to exhibit estrogenic activity. Thus, it is reasonable to
believe that neither the E2-CDS nor the E2-Q+ formed in the brain account for
the prolonged pharmacological effects of this delivery system. Rather locally
released E2 in the brain, particularly the hypothalamus, would appear to
account for the sustained suppression of the gonadotropin secretion.
Our previous evaluation of tissue distribution of the E2-CDS in male
rats (Chapter 5; Rahimy et al., 1990a) and the more detailed dose-response and
time-course evaluation of the E2-CDS distribution in OVX rats (Chapter 6;
Rahimy et al., 1990b) revealed that (i) E2-Q+, the quaternary form of E2-CDS, as
well as E2 persists in the brain with ti/2 = 8-9 days and (ii) the same
metabolites are rapidly eliminated from the peripheral tissues. These
findings together with the absence of a physiologically significant elevation of
plasma E2 concentrations from 7-28 days after the E2-CDS administration
(Chapter 6), provide strong evidence for the local action of E2 in the brain,
presumably on hypothalamic luteinizing hormone-releasing hormone
(LHRH) containing neurons (Sarkar et al., 1989).
The synthesis and secretion of gonadotropins from the anterior
pituitary are differentially regulated by several neuronal (Barraclough &
Wise, 1982; Dalkin et al., 1989; Marshal & Kelch, 1986; Plant, 1986) and
hormonal (Kalra & Kalra, 1980, 1983) factors including the hypothalamic
decapeptide, LHRH, and the action of E2 both in a positive and negative

121
feedback mode at the hypothalamus as well as the anterior pituitary. The
evaluation of the effects of E2-CDS on LHRH neuronal activity (i.e. LHRH
release) showed that portal blood concentrations of LHRH were significantly
reduced for more than 16 days following the treatment (Sarkar et al., 1989).
The reduced LHRH secretion, was in contrast to the increased hypothalamic
LHRH concentrations, suggesting that the inhibition of release resulted in a
tissue buildup of the decapeptide. Furthermore, chronic exposure to E2 has
no significant effects on anterior pituitary responsiveness to LHRH (Cooper et
al., 1974), indicating that the prolonged inhibitory effects of E2-CDS on LH and
FSH are due primarily to sustained suppression of LHRH secretion from the
hypothalamus.
When the dynamics of the E2-CDS effects were compared with that of
an equimolar dose of E2, the E2-CDS showed 100-fold greater efficacy in the
magnitude of inhibition of plasma LH and FSH. In other words, the
magnitude of E2 effects was equivalent to that of the E2-CDS but with 100-fold
lower dose (Figure 15). This marked increased in effectiveness and the
prolonged duration of the E2-CDS effects on LH and FSH secretion are most
likely due to "lock-in" of the E2-Q+ behind the BBB with subsequent slow
release of E2 in the brain.
When the kinetic behaviors of E2-CDS and E2 were compared on molar
basis, the E2-CDS (1.0 mg/kg) produced E2 concentrations in brain tissue
which were 81- and 182-fold greater than after an equimolar E2 (0.7 mg/kg)
treatment at 1 and 7 days postinjection, respectively (Chapter 6). Therefore, it
seems more reasonable to suggest that following the E2-CDS administration,
the brain E2 is continuously produced and as such the steady-state brain
concentrations of E2 is dependent on its rate of production from the E2-Q+ and

122
its rate of elimination from the brain by either local metabolism or its
redistribution down a concentration gradient into the general circulation.
We observed a significant elevation in plasma PRL in response to the
highest dose of E2-CDS (1.0 mg/kg), whereas lower doses had no significant
effect on plasma PRL concentrations. It appears then that elevations in
plasma PRL correlate with the administration of E2-CDS at doses which result
in the transient elevation of plasma E2 levels, but not at doses at which
plasma E2 remains low (Chapter 6). This apparent stimulation of PRL
production by the E2-CDS would appear to be due to the well described actions
of E2 on the anterior pituitary lactotrophes (Chen & Meites, 1970). However,
the lack of a positive temporal correlation between plasma PRL (present
study) and plasma E2 levels (Chapter 6) suggests the possibility that E2 released
in the brain might be responsible for a direct stimulation of the anterior
pituitary. This can be explained by the anatomical relationship between the
hypothalamus and the anterior pituitary gland. Estradiol released from the
E2-Q+/ or the E2-Q+ itself, which is "locked" into the brain, could be delivered
directly to the anterior pituitary by the capillary plexus of the hypophyseal
portal system (Traystman, 1983). These capillaries in the median eminence
lack features of other brain capillaries and hence are not part of the BBB
(Traystman, 1983).
The E2-CDS had no significant effects on the mean plasma GH
concentrations over the 28 days time-course at any of the 3 doses examined.
However, a careful analysis of the effects of E2-CDS on pulsatile GH secretion
(Millard et al., 1990) revealed that while mean GH levels are not changed,
baseline GH values were elevated and GH pulse amplitudes were moderately
reduced at 7 days after E2-CDS administration.

123
The marked increases in anterior pituitary weights of OVX rats treated
with the E2-CDS appears to be due to the direct effects of E2 on the pituitary
gland. These effects of E2 appear to be exerted on the lactotroph population of
the anterior pituitary (Chen & Meites, 1970; Gorski, 1981). E2 is well known to
stimulate PRL secretion and to induce hyperplasia of lactotrophs (Chen &
Meites, 1970; Gorski, 1981). As indicated above, brain E2 likely reaches the
anterior pituitary gland, through the redistribution of the steroid down the
marked concentration gradient from the brain to the pituitary gland
(Traystman, 1983). It should be noted, however, that the effects of E2-CDS on
pituitary weight are dependent upon the OVX condition of the rats. In gonad-
intact rats, E2-CDS does not alter anterior pituitary weight.
The uterotrophic effects of E2-CDS were also dose- and time-dependent.
This effect of E2-CDS likely relates to the extreme sensitivity of OVX rats to
circulating estrogens (Mayer et al., 1960). Thus, even modest elevations in
plasma E2 following administration of E2 or E2-CDS (Chapter 6), result in
stimulation of uterine tissue in OVX rats. However, both the uterus and the
anterior pituitary gland of gonad-intact rats are unresponsive to the estrogen
delivery system (Anderson et al., 1988a). Finally, it should be noted that the
uterine weights observed following E2-CDS were considerably lower than the
500 to 625 mg weights normally seen in gonad-intact rats (Anderson et al.,
1988a).
In conclusion, the prolonged effects of the E2-CDS on gonadotropins
suppression were dose- and time-dependent, and the duration of these
responses are consistent with the long half-lives of the E2-CDS metabolites in
the brain. These results further support the view that the E2-CDS may be
potentially useful in fertility regulation and treatment of brain E2 deficiencies
(i.e. vasomotor hot flushes).

Plasma LH (ng/ml)
124
Figure 16. Dose and time-dependent effects of the E2-CDS on plasma LH
responses in ovariectomized rats. Animals received a single iv
injection of the E2-CDS on day 0 at doses of 0.01, 0.1 and 1.0 mg/kg
bw. Also, the responses to an E2 dose of 0.7 mg/kg, equimolar to
the 1.0 mg /kg dose of E2-CDS, is shown for day 1 and 7.
Represented are means SEM for n = 7 rats per group per
sampling time. The symbols indicate statistical differences as
follows: *) different from vehicle group (day 0); a) different from
0.01 mg/kg; and b) different from both 0.01 and 0.1 mg/kg. The
significance of interaction between factors (time and dose) was
determined by two-way analysis of variance (ANOVA). The
significance of differences among mean values at each dose level
was determined over time by one-way ANOVA and Dunnett's
test while the significance of differences among mean values of
three dose levels (at each time point) was determined by one-way
ANOVA and Scheffe F-test. The level of probability for all tests
was p<0.05.

Plasma FSH (ng/ml)
125
-7 0 7 14 21 28
Days Post Injection
Figure 17. Dose and time-dependent effects of the E2-CDS on plasma FSH
responses in ovariectomized rats. Animals received a single iv
injection of the E2-CDS on day 0 at doses of 0.01, 0.1, and 1.0
mg/kg. Also, the responses to an E2 dose of 0.7 mg/kg bw,
equimolar to the 1.0 mg/kg dose of E2-CDS, is shown for day 1 and
7. Represented are means SEM for n = 7 rats per group per
sampling time. The symbols indicate statistical differences as
follows: *) different from vehicle group (day 0); a) different from
0.01 mg/kg; and b) different from both 0.01 and 0.1 mg/kg. The
significance of interaction between factors (time and dose) was
determined by two-way analysis of variance (ANOVA). The
significance of differences among mean values at each dose level
was determined over time by one-way ANOVA and Dunnett's
test while the significance of differences among mean values of
three dose levels (at each time point) was determined by one-way
ANOVA and Scheffe F-test. The level of probability for all tests
was p<0.05.

Plasma GH (ng/ml) Plasma PRL (ng/ml)
126
Figure 18. Dose and time-dependent effects of the E2-CDS on plasma PRL
(upper panel) and GH responses (lower panel) in ovariectomized
rats. Animals received a single iv injection of the E2-CDS on day 0
at doses of 0.01,0.1 and 1.0 mg/kg or an E2 dose of 0.7 mg/kg,
equimolar to the 1.0 mg/kg dose of E2-CDS, is shown for day 1 and
7 only. The symbols indicate statistical differences as follows: *)
different from vehicle group (day 0); a) different from 0.01 mg/kg;
and b) different from both 0.01 and 0.1 mg/kg.

127
Table 8: Dose and Time-Dependent Effects of the E2-CDS on Peripheral Tissue
Weights in Ovariectomized Rats
Tissue
Drue
Dose
Davs after treatment
Anter.
HPCD
mg/kg
0
11.8
1
7
14
21
28
Pituit.
0.5
(mg)
e2
0.7
13.5
12.8
ND
ND
ND
0.7
0.7
E2-CDS
0.01
12.0
13.1
14.4
11.9
13.0
0.8
0.5
1.5
0.8
0.6
E2-CDS
0.1
12.8
15.6*
14.9*
13.1
12.6
0.9
0.9
1.5
0.5
0.6
E2-CDS
1.0
13.3
19.0*,b
20.7* ,b
16.2*,b
15.0*,c
0.9
1.0
1.3
1.0
0.6
Uterus
HPCD
154.2
(mg)
e2
0.7
8.6
261.4*
182.9
ND
ND
ND
13.0
24.0
E2-CDS
0.01
183.8
179.1
196.4
121.5
107.8
12.3
29.0
42.0
7.0
8.0
E2-CDS
0.1
237.7*
234.4*
210.1*
161.1
128.4
12.0
27.0
33.0
8.0
5.0
E2-CDS
1.0
281 .4*, a
427.0*, b
372.9*, b
267.4*,b 235.4*h
20.0
31.0
28.0
31.0
16.0
Values are the mean tissue weights SEM. ND = Not determined.
* Different from time 0 values,
a Different from 0.01 mg/kg dose,
b Different from 0.01 and 0.1 mg/kg doses,
c Different from 0.1 mg/kg dose.

CHAPTER 8
EFFECTS OF THE E2-CDS OR CASTRATION ON ANDROGEN AND
ANDROGEN-DEPENDENT TISSUES IN MALE RATS
Introduction
Prostate carcinoma is the second most common cancer in males, and its
incidence increases rapidly over the age of 50 (Silverberg & Lubera, 1986).
Likewise, the development of benign prostatic hyperplasia also increases in
incidence with age; however, there is no direct evidence that benign prostatic
hyperplasia is a necessary prerequisite in the development of prostatic cancer
(Berry et al., 1984; Rotkin, 1983). Despite the considerable clinical and
scientific advances in the diagnosis, prognostication, and treatment, these
prostatic diseases remain the second leading cause of morbidity and/or
mortality in senescent males (Santen & English, 1989). Although
epidemiological findings and sporadic reports of a familial occurrence of
prostatic cancer suggest a genetic predisposition (Silverberg & Lubera, 1986),
there is currently little evidence that this is a significant factor in most
patients. A variety of other theories including neoplastic and endocrine
metabolic factors have been proposed to explain the etiology and occurrence
of prostatic adenocarcinoma and/or benign prostatic hyperplasia in men.
Certainly, it is clear that the two major factors necessary for the genesis in
men of these prostatic diseases are the presence of the testis and aging
(Huggins & Hodges, 1941). Studies that attempt to correlate the incidence of
the disease with age-related changes in testosterone levels or other endocrine
factors have shown no consistent relationship as yet. However, there are
128

129
several lines of evidence that strongly support the possibility that prostatic
growth/hyperplasia is dependent upon the endocrine activity of the testis: (1)
prostatic maturation as well as prostatic hyperplasia does not occur in men
who are castrated prior to puberty (Moore, 1944); (2) surgical castration (CAST)
produces regression as well as beneficial effects toward the prostate
hyperplasia of men (Cabot, 1896; Huggins & Hodges, 1941; White, 1895); (3)
numerous studies have reported regression of prostatic hyperplasia with
antiandrogens or estrogens treatment (Brendler, 1988; Foote & Crawford,
1988); and (4) recent biochemical studies have reported an association between
prostatic hyperplasia with an abnormal accumulation of dihydro testosterone
(DHT), a potent testosterone (T) metabolite, in the human prostate (Isaacs et
al., 1983). Whether the increase in the concentration of prostatic DHT is
directly responsible for or is the result of the disease has not been completely
resolved. However, further support for the DHT involvement comes from
males with hereditary deficiency of the 5 oc-reductase enzyme, an enzyme that
converts T to DHT in androgen-target tissues (Imperato-McGinley et al., 1980,
1984). The lack of this enzyme results in very low levels of DHT as well as no
palpable prostatic tissue, indicating that the prostate had not developed
during embryogenesis.
Regarding the therapeutic aspects of the prostatic adenocarcinoma
and/or hyperplasia, currently a variety of surgical and therapeutic means of
inhibiting androgen production or blocking androgen action are being used
(Labrie et al., 1983; Santen & English, 1989). These include surgical CAST,
high-dose estrogen therapy, GnRH analogues, antiandrogens, and
combination of low dose estrogens plus low dose antiandrogens. Each of
these treatment regimens provides temporary benefit to patients, however,
relapse usually occurs within a period of 1 to 2 years. Nevertheless, CAST or

130
high-dose estrogen therapy, the two conventional treatment paradigms, still
remain as the treatment of choice for the endocrine-dependent management
of prostatic cancer (van Steenbrugge et al., 1988). Both treatments are reported
to be equally effective (i) in suppressing the circulating T levels (Carlstrom et
al., 1989) and perhaps regression of the prostatic hyperplasia; and (ii) in
controlling the symptoms of advanced prostatic cancer in 70-80% of patients
with an average improved rate of survival of 5 years (Klein, 1979).
However, none of the two treatments is optimal or without major
complication. For instance, high-dose estrogen therapy has been shown to
cause severe cardiovascular complications (Henriksson & Edhag, 1986) and
alterations in liver metabolism (von Schoultz et al., 1989). Estrogens,
specially synthetic compounds, exert profound effects on liver-derived
plasma proteins, coagulation factors, lipoproteins, and triglycerides when
administered orally. However, most of the cardiovascular complications are
the result of arterial ischemic events, and the majority of such events are the
results of acute coronary arterial disease (Henrikson & Edhag, 1986). Recent
studies have suggested that these dose-dependent liver-associated side effects
and thus, cardiovascular complications may be reduced or even abolished
when lowering the dose, provided that adequate T suppression is achieved
(Jonsson et al., 1975; von Schoultz et al., 1989). Furthermore, numerous other
reports also indicated that the marked interference with hepatic metabolism
is associated with the kind of estrogen used (Ottosson, 1984; Ottosson et al.,
1986). For example, the synthetic alkylated estrogen, ethinyl E2, is a potent
estrogen used in the treatment of prostatic cancer; however, it exerts
profound effects on liver metabolism and thus, the use of a natural estrogen
may be advantageous (Ottosson et al., 1986).

131
A potential problem associated with currently used estrogens is that
these steroid hormones equilibrate among all body tissues due to their high
lipophilicity. As a result, only a fraction of the administered estrogen dose
accumulates at or near the intended site of action. Indeed, when these
hormones are used therapeutically to specifically target the brain, the steroids
must be given either frequently or in high doses in order to maintain
therapeutically effective concentrations in the brain. Both of these treatment
strategies lead to sustained increases in peripheral estrogen levels and, in
particular, the liver is exposed to a greater drug burden. This is a major
limiting factor in the use of these estrogenic products.
Based on the previous pharmacokinetic (Chapters 5 & 6) and
pharmacodynamic (Chapter 7) observations and hence, the therapeutic
potential of the E2-CDS, we evaluated the effects of E2-CDS (0.5 mg/kg b.w.) on
androgen and androgen-responsive tissues in the present study. This dose of
the E2-CDS was chosen because a preliminary dose-response and time-course
study indicated that 0.5 mg/kg is more effective than 1.0 mg/kg dose in
suppressing androgen and androgen-responsive tissues in intact male rats.
Materials and Methods
Adult male Charles River (CD) rats (aged 3-4 months) used in this
study were randomly divided into 9 experimental groups (7-8 rats per group).
Two groups of rats were castrated while 7 other groups remained intact.
CAST was performed by an abdominal incision under metofane anesthesia.
Since surgical stress has previously been found not to affect the experimental
parameters tested here, intact male rats used here were not subjected to sham
operation.

132
To evaluate the effects of E2-CDS or CAST on androgen and androgen-
dependent sex accessory organs, a total of 5 different treatment conditions
were evaluated: 1) intact control group; animals received no treatment; 2)
CAST group; animals received no further treatment; 3) intact + E2-CDS (xl)
group; animals received a single iv injection (tail vein) of E2-CDS (0.5 mg/kg
bw); 4) intact + E2-CDS (x2) group; animals received E2-CDS (0.5 mg/kg bw, iv)
once a week for 2 consecutive weeks; 5) intact + E2-CDS (3x) group; animals
received E2-CDS (0.5 mg /kg bw, iv) once a week for 3 consecutive weeks.
Animals were then sampled 7 or 14 days after the last treatment (treatment
conditions 2 to 5).
Rats (7-8 per group) were sacrificed by decapitation either 7 or 14 days
after the final administration of E2-CDS or post CAST (treatment conditions
2-5). Immediately following decapitation, the trunk blood was collected in
heparinized glass tubes on ice. The blood was centrifuged, and the plasma
was separated and stored at -20C for subsequent plasma hormone analysis.
Also, the following tissues: prostate (the ventral lobe), seminal vesicles (both
horns), testis, and anterior pituitary were rapidly dissected. The dissection of
accessory sex organs was carried out according to previously reported
guidelines (Lee, 1987). The ventral prostate was carefully grasped with forceps
and dissected to the level of the ducts entering ventrally into the urethra.
The seminal vesicles were retracted and expelled of seminal fluid. All tissues
were blotted dry on paper then weighted to the nearest 0.1 mg.
Plasma LH, FSH, and PRL concentrations were measured in duplicate
by RIA using NIDDK kits. Plasma LH, FSH, and PRL values are expressed as
ng/ml of the LH-RP-2, FSH-RP-2 or the PRL-RP-3 reference standards,
respectively, The intra-assay coefficients of variation were 3.37, 2.27, and
4.07% for LH, FSH, and PRL assays, respectively. Plasma samples containing

133
undetectable LH or FSH were assigned the respective assay sensitivity (0.25 for
LH and 2.5 ng/ml for FSH). All the samples for each hormone were assayed
in a single run.
Plasma E2 and T concentrations were measured in duplicate by using
Coat-A-Count Estradiol kits and Coat-A-Count Testosterone kits, respectively.
The RIA sensitivity of the T assay was 0.2 ng/ml. The cross-reactivity of the T
antibody to the DHT and E2 has been reported to be 3.3 and 0.02%, respectively
(technical information from Diagnostic Products).
Results
Effect of CAST or E?-CDS on Plasma T Levels
Figure 19 shows the comparative plasma T concentrations of untreated
control, post CAST and animals treated with E2-CDS. CAST resulted in
greater than 99% suppression of plasma T levels which remained suppressed
at about the detection limit of the T assay for the time-course of study.
Likewise, plasma T levels in E2-CDS-treated animals were significantly
suppressed by more than 96, 92, or 95% with 1, 2, or 3 injections, respectively,
at 7 days after the last treatment. However, by 14 days after the last treatment
with E2-CDS, a 76, 82, or 91% reduction in plasma T was observed with 1, 2, or
3 injections, respectively. Even though plasma T levels began to gradually
increase in a manner related to the number of E2-CDS injections, T levels
remained significantly suppressed compared with the control values.

134
Effect of CAST or E9-CDS on Tissue Weights
The effects of CAST and E2-CDS treatment on tissue weights are shown
in Figures 20-23. CAST reduced the ventral prostate weights by more than
67% or 66% at 7 or 14 days after the orchidectomy, respectively (Figure 20).
Treatment with E2-CDS with all paradigms significantly reduced the ventral
prostate weight to CAST level at 7 days after the last injection (Figure 20). The
E2-CDS-induced regression in ventral prostate weight remained significantly
low at 14 days after the last treatment. Interestingly, the 3-injection paradigm
of E2-CDS was equivalent to CAST in reducing the ventral prostate weight by
62% at 14 days after the last injection.
Similarly, CAST significantly reduced the seminal vesicles weight by
52% or 63% at 7 or 14 days post CAST, respectively (Figure 21). Treatment of
intact animals with E2-CDS significantly reduced seminal vesicles weight by
44% to 62% and 34% to 55% at 7 and 14 days after the final treatment,
respectively (Figure 21). The 3-injection paradigm of E2-CDS was sufficient to
reduce and chronically maintain the seminal vesicles weight at CAST level
for the time-course studied.
Treatment of intact male rats with E2-CDS had no significant effect on
testes wet weight or wet weight/100 g bw (Figure 22). The 3-injection regimen
of E2-CDS caused about 19% reduction in testes weight at 7 or 14 days after the
final injection (Figure 22).
Anterior pituitary weights were not changed by CAST (Figure 23). In
contrast, pituitary weights increased significantly in a manner related to the
number of injections in response to E2-CDS administration (Figure 23).

135
Effect of CAST or E?-CDS on Plasma Hormones
CAST significantly increased the plasma gonadotropins (LH and FSH)
concentrations (Tables 9 & 10). By contrast, the plasma gonadotropin
concentrations in animals treated with E2-CDS were not significantly altered
from control (Tables 9 & 10). Most likely, LH values in animals treated with
E2-CDS may have been significantly suppressed compared with control
values, because more than 50% of LH values were below the detection limit
of assay. However, since the sensitivity of LH assay was assigned as 0.25
ng/ml, statistically significant LH suppression was not achieved. A
significant elevation in plasma PRL concentrations was observed only in
animals treated with 3 injections of E2-CDS (Table 9). Likewise, plasma E2
concentrations were significantly elevated in animals treated with 2 and 3
injections of E2-CDS at 7 days after the last injection (Table 9).
Discussion
The primary objective of endocrine therapy in prostate malignancy is
the induction of an effective androgen deprivation, thus abolishing the
growth promoting effects of androgens on the prostate tissue. These
therapeutic objectives can be achieved by several mechanisms at different
levels of the hypothalamo-pituitary-gonadal axis: 1) suppression of
hypothalamic LHRH and hence, of pituitary LH release, thereby inhibiting T
production by the testis; 2) surgical CAST which eliminates more than 90% of
circulating T; 3) inhibition of androgen synthesis in the testis; and 4) blocking
androgen action at the receptor site in the prostate. Thus, the choice of

136
treatment strategy largely relates to minimizing toxicity while optimizing the
response rate as well as the duration of benefit.
In the present study, we described a newer endocrine approach in the
treatment of androgen-dependent prostatic diseases. The evaluation of E2-
CDS effects on the normal androgen-responsive sex accessory organs with
comparison to that of CAST demonstrated that: 1) the E2-CDS at a single dose
of 0.5 mg/kg was as effective as CAST in suppressing significantly the plasma
T levels for 14 days after treatment; 2) the E2-CDS treatment of intact male rats
resulted in significant regression of the androgen-sensitive ventral prostate as
well as seminal vesicle weight equivalent in magnitude to that of CAST
alone; 3) interestingly, both the profound suppression of T levels and the
prolonged duration of tissue regression, at 14 days after the final treatment,
were observed even in the face of low plasma E2 levels. Furthermore, these
data suggest that the primary site of action where E2 exerts its effects leading to
T suppression is in the central nervous system. Previously, we have reported
long-term suppression of LH in CAST male rats (Simpkins et al., 1986),
stimulation of masculine sexual behavior in CAST male rats for 28 days
(Anderson et al., 1987a), and long-term gonadotropin suppression in
ovariectomized rats (Chapter 7) following a single iv administration of the E2-
CDS. These prolonged pharmacological effects of the E2-CDS are consistent
with the observations of the accumulation of E2-Q+, the oxidized form of the
delivery system, in the brain with an apparent t^ ¡2 = 8-9 days in that tissue
(Chapters 5 & 6). From this store of E2-Q+, E2 is slowly released through non
specific hydrolysis of the carrier, resulting is sustained brain exposure to E2.
The production of T by the Leydig cells of the testis are controlled via a
negative feedback mechanism (Swerdloff, 1986). Increased levels of T exert a
negative feedback on both the hypothalamus and the anterior pituitary, thus

137
inhibiting further LHRH and LH release. Recent studies suggest that E2, the
aromatized metabolite of T, is responsible in mediating the feedback effect of
T on the hypothalamic neuronal system (Christensen & Clemens, 1974).
Several lines of evidence have accrued to indicate that E2 formation in the
CNS is important for the effect of T. First, aromatization inhibitors decreased
the effectiveness of T in inducing sexual behavior in CAST rats (Beyer et al.,
1976). Second, intrahypothalamic implant of E2 effectively restored the
masculine sexual behavior in long-term CAST male rats (Christensen &
Clemens, 1974). Finally, estrogen receptors and binding sites have been
identified in the brain of male rats (Krey et al., 1980). Thus, prolonged
exposure to estrogen hormone, whether by E2-CDS or frequent
administration of currently available estrogens, produces sustained
suppression of hypothalamic LHRH secretion and hence, LH inhibition.
Eventually, this leads to T deprivation or chemical castration.
Our results show that, like CAST, the E2-CDS produced a significant T
deprivation which remained suppressed for at least 14 days with a single
injection, or for 28 days with 3 injections given once every 7 days for three
consecutive weeks. Similarly, a single injection of E2-CDS was sufficient to
significantly regress both the ventral prostate and the seminal vesicles
weights during the first week of treatment, while the 3 injections paradigm
produced significant regression (equivalent to CAST) of these two tissues for
at least 4 weeks. The profound suppression at 7 days of circulating T
concentrations and/or of tissue weights (prostate and seminal vesicles) caused
by the single or double injection of E2-CDS, gradually began to recover at 14
days after the final treatment. However, these values remained significantly
suppressed compared with the control values at 14 days after the last
injection. Furthermore, unlike CAST, E2-CDS treatment resulted in pituitary

138
hyperplasia. This time- and injection-related increase in pituitary weight
appears to be due to the direct effects of E2 on the pituitary gland. E2 is well
known to stimulate PRL secretion and to induce hyperplasia of lactotrophs
(Chen & Meites, 1970). The source of E2 responsible for the stimulation of
anterior pituitary is most likely the brain E2 and not the residual peripheral
E2. This can be explained by the anatomical relationship between the
hypothalamus and the pituitary gland (Traystman, 1983).
We did not observe in intact male rats a statistically significant
suppression in plasma gonadotropins (LH & FSH) in response to E2-CDS
treatment. However, plasma LH showed a progressive decline of 20 to 40%
compared with the basal LH values. In fact, more than 50% of experimental
animals treated with the E2-CDS exhibited plasma LH values which were
below the sensitivity of LH assay. Since plasma samples containing
undetectable LH were assigned the assay sensitivity (0.25 ng LH/ml),
statistically significant LH suppression was not obtained even if LH may have
been significantly suppressed in animals that were treated with the E2-CDS.
Furthermore, since basal LH values of intact male rats were at the limit of the
LH assay sensitivity, it was not possible to make a reliable correlation between
the degree of LH suppression and tissue weight regression in rats treated with
the E2-CDS. Our previous observations in CAST rats showed that LH
concentrations were suppressed by 82-90% for 4-12 days after a single injection
of E2-CDS (Simpkins et al., 1986). It should be noted, however, that CAST rats
are much more sensitive than intact rats to the LH-suppressing effects of E2-
CDS.
Despite the apparent lack of significant LH suppression, plasma T
levels were significantly suppressed by 96% or 76% at 7 or 14 days,
respectively, after a single injection of E2-CDS. This profound and sustained

139
suppression of circulating T levels were observed even in the face of low
plasma E2 levels, indicating the CNS involvement in mediating T
suppression. Furthermore, when the dynamics of E2-CDS effects on
circulating T levels were compared with that of an equimolar dose of E2
valerate, the E2-CDS significantly suppressed T levels while E2 valerate was
ineffective (Anderson et al., 1989). Thus, as it has been consistently
demonstrated (Anderson et al., 1987a,b, 1988a,b, 1989; Estes et al., 1987a,b, 1988;
Simpkins et al., 1986, 1988, 1989a,b), the prolonged pharmacodynamic effects
of E2-CDS following a single injection are most likely due to "locking" of the
E2-Q+ behind the BBB and there it serves as a brain depot for E2.
Although, in the present study, we used a rat model to investigate the
efficacy of E2-CDS in reducing androgen levels with subsequent regression of
the ventral prostate tissue, there are no conclusive data as to whether the rat
prostate reflects a complete picture analogous to that of the aging human
prostate. However, this model system has been employed for some years to
gain further insight into the processes involved in the initiation and
progression of prostatic hyperplasia (Pollard et al., 1989).
High-dose estrogens have also been reported to be effective in animal
models of prostatic cancer (Daehlin & Damber, 1986). Recently, it was
reported that E2 implants (producing plasma E2 levels as high as 4,442 962
pmole/liter) to tumor-(prostatic carcinoma, PC-82) bearing mice resulted in
tumor growth arrest with a subsequent decline of the tumor volume, which
equals the effect of CAST (van Steenbrugge et al., 1988). Additionally, it was
suggested that the effects of E2 on the PC-82 tumor model were mainly
indirect by suppressory effect on T secretion in the host animal, rather than a
direct effect on the tumor tissue (van Steenbrugge et al., 1988). Conversely a
number of studies demonstrated a direct action of estrogens at the cellular

140
level in the prostatic tissue (Daehlin et al., 1987). However, in spite of the
possible direct effects of estrogen at the cellular level in addition to the
indirect effects (CAST-like suppression of plasma T levels), the combined
treatment of prostate cancer patients with CAST and estrogens did not
improve survival when compared to either treatment alone (Blackard et al.,
1973, 1975). Further, estrogens have also been shown to decrease 17a-
dehydroxylase, 17 P-dehydrogenase and/or 17-20 desmolase (Kalla et al., 1980;
Onoda & Hall, 1981) activity in the Leydig cells of the testis. Inhibition of
these enzymes can increase the production of pregnenolone and progesterone
and the concomitant decrease in T synthesis in Leydig cells.
In conclusion, because of the potential application of brain-enhanced
estrogen delivery with sustained release in the brain as an alternative, we
conducted this study. Collectively, the results support the concept that the E2-
CDS may be useful in the treatment of androgen-dependent prostatic
hyperplasia. In comparison to the currently used estrogenic products, the E2-
CDS should achieve the sustained stimulation of brain E2 receptors at lower
doses or with less frequent dosing.

Treatment Treatment
141
Control 0
7 Davs After Final Treatment
Castrated 0
1 a
E2-CDS (xl)
a
E2-CDS (x2)
a
E2-CDS (x3)
a
14 Days After Final Treatment
Control
Castrated
E2-CDS
E2-CDS
E2-CDS
Plasma Testosterone (ng/ml)
Figure 19. Effects of the E2-CDS (0.5 mg/kg) or CAST on plasma testosterone
levels either 7 days (upper panel) or 14 days (lower panel) after the
final treatment. Animals received weekly iv (tail vein) injection
of either the drug's vehicle (HPCD), or E2-CDS (1 injection; xl), E2-
CDS (2 injections; x2), E2-CDS (3 injections; x3), or were castrated
(CAST). Represented are means SEM for n = 7-8 animals per
group per sampling time. The symbol (a) denotes differences from
control (HPCD)-treated animals as analyzed by ANOVA and
Scheffe statistics.

Treatment Treatment
142
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
7 Days After Final Treatment
0 50 100 150
14 Days After Final Treatment
Control
0
Castrated
0
E2-CDS
(xl)
E2-CDS
(x2)
E2-CDS
(x3)
0 50 100 150
Ventral Prostate mg /100 g bw
i
200
200
Figure 20. Effects of the E2-CDS (0.5 mg/kg) or CAST on ventral prostate
weight either 7 days (upper panel) or 14 days (lower panel) after
the final treatment. Animals received weekly iv (tail vein)
injection of either the drug's vehicle (HPCD), or E2-CDS at a dose
of 0.5 mg/kg (1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3
injections; x3), or were castrated (CAST). Represented are means
SEM for n = 7-8 animals per group per sampling time. The symbol
(a) denotes differences from control (HPCD)-treated animals and
the symbol (b) indicates differences from CAST within the panel
as analyzed by ANOVA and Scheffe statistics.

Treatment Treatment
143
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
0 20 40 60 80 100 120 140
14 Days After Final Treatment
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
a,b
t'i
0 20 40 60 80 100 120 140
Seminal Vesicle mg/100 g bw
Figure 21. Effects of the E2-CDS (0.5 mg/kg) or CAST on seminal vesicle
weight either 7 days (upper panel) or 14 days (lower panel) after
the final treatment. Animals received weekly iv (tail vein)
injection of either the drug's vehicle (HPCD), or E2-CDS at a dose
of 0.5 mg/kg (1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3
injections; x3), or were castrated (CAST). The symbol (a) denotes
differences from control (HPCD)-treated animals and the symbol
(b) indicates differences from CAST within the panel as analyzed
by ANOVA and Scheffe statistics.

Treatment Treatment
144
7 Days After Final Treatment
Control
0
Castrated
0
E2-CDS
(xl)
E2-CDS
(x2)
E2-CDS
(x3)
Control
0
t 1
800 1000
14 Days After Final Treatment
400
Testes mg/100 g bw
1000
Figure 22. Effects of the E2-CDS (0.5 mg/kg) or CAST on testis weight either 7
days (upper panel) or 14 days (lower panel) after the final
treatment. Animals received weekly iv (tail vein) injection of
either the drug's vehicle (HPCD), or E2-CDS at a dose of 0.5 mg/kg
(1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3 injections; x3),
or were castrated (CAST). Represented are means SEM for n = 7-
8 animals per group per sampling time.

Treatment Treatment
145
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
7 Days After Final Treatment
0 1 2 3 4 5 6
a,b
i
7
14 Days After Final Treatment
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
H a,b
T"
2
-r
3
5
-i
6
Anterior Pituitary mg/100 g bw
Figure 23. Effects of the E2-CDS (0.5 mg/kg) or CAST on anterior pituitary
weight either 7 days (upper panel) or 14 days (lower panel) after
the final treatment. Animals received weekly iv (tail vein)
injection of either the drug's vehicle (HPCD), or E2-CDS at a dose
of 0.5 mg/kg (1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3
injections; x3), or were castrated (CAST). Represented are means
SEM for n = 7-8 animals per group per sampling time. The symbol
(a) denotes differences from control (HPCD)-treated animals and
the symbol (b) indicates differences from CAST within the panel
as analyzed by ANOVA and Scheffe statistics.

Table 9: Effects of the E2-CDS (0.5 mg/kg bw) or CAST on Plasma Hormone Concentrations
at 7 Days after the Last Treatment in Male Rats.
Treatment
Group
Inject.
No.
E2
(pg/ml)
E?-Q+
(pg/ml)
LH
(ng/ml)
FSH
(ng/ml)
PRL
(ng/ml)
Intact
-
12.3 2.70
ND
0.50 0.07
10.341.88
28.1514.91
CAST
-
13.06 2.85
ND
5.77 0.42a
27.5711.00a
5.8911.06
E2-CDS
1
40.45 3.39
578.3 46.9
0.30 0.03b
6.6510.56b
35.9715.67
e2-cds
2
52.67 7.69a,b
783.5 90.9
0.28 0.02b
6.5310.56b
43.2116.69
e2-cds
3
70.63 18.29a,b
654.8 54.4
0.39 0.07b
7.3310.98b
80.64 112.88a,b
ND Not determined
a Different from Intact
b Different from CAST
146

Table 10: Effects of the E2-CDS (0.5 mg/kg bw) or CAST on Plasma Hormone Concentrations at 14
Days after the Last Treatment in Male Rats.
Treatment
Group
Inject.
No.
12
(pg/ml)
E?-Q+
(pg/ml)
LH
(ng/ml)
FSH
(ng/ml)
PRL
(ng/ml)
Intact
-
12.3 2.70
ND
0.50 0.07
10.34 1.88
28.15 4.91
CAST
-
7.30 1.87
ND
8.15 0.55a
26.57 1.34a
4.84 0.82
E2-CDS
1
20.07 4.11
190.2 22.9
0.27 0.01b
10.53 0.91b
38.34 9.96
e2-cds
2
13.73 4.31
120.9 18.8
0.27 0.02b
12.00 1.86b
15.93 3.10
e2-cds
3
30.01 4.17
131.5 20.4
0.31 0.03b
11.83 1.30b
12.47 3.19
ND Not determined
a Different from Intact
b Different from CAST
147

CHAPTER 9
EFFECTS OF THE E2-CDS OR E2 PELLET ON TAIL-SKIN TEMPERATURE
RESPONSES IN OVARIECTOMIZED FEMALE RATS
Introduction
Over the past 50 years, numerous clinical studies in either natural or
ovariectomy-induced menopausal women have provided evidence that
gonadal steroid withdrawal causes alterations in the central thermoregulatory
system which then leads to vasomotor hot flushes (Casper & Yen, 1985;
Lauritzen, 1973; Yen, 1977). After menopause or ovariectomy, the decreasing
production of ovarian estrogens/progestins leads abruptly to a number of
central nervous system (CNS)-mediated steroid-withdrawal symptoms
(Casper & Yen, 1985; Lauritzen, 1973; Yen, 1977). These symptoms most often
express themselves as hot flushes, perspiration, depression, anxiety, changes
in memory, headaches, insomnia and irritability (Clayden et al., 1974;
Meldrum et al., 1979; Paterson, 1982). However, their intensity and frequency
vary among women and symptoms do not occur at all in about 15 to 25% of
menopausal women (Casper & Yen, 1985). The most frequent symptom of
the menopause is the hot flush, an episodic disturbance of thermoregulation
characterized by a sensation of heat followed by a sudden spreading of flush
and perspiration (Casper & Yen, 1985; Clayden et al., 1974; Lauritzen, 1973;
Meldrum et al., 1979; Paterson, 1982; Yen, 1977). These physiological
alterations appear to be the result of autonomic discharge which causes
peripheral vasodilation and heat loss, with a considerable drop in core body
temperature (Nesheim & Saetre, 1982). Furthermore, each flush is preceded
148

149
by an acceleration in heart rate (Molnar, 1975) and a surge in secretion of
luteinizing hormone (LH) accompanies the flush response (Casper et al., 1979;
Meldrum, 1979). Additionally, flushes can be provoked by warm ambient
temperature, hot drinks, alcoholic beverages, mental stress, and hypoglycemia
(Simpkins & Katovich, 1984).
Although the mechanism(s) involved in the menopausal syndrome is
as yet unknown, numerous clinical and experimental studies have
implicated the hypothalamic noradrenergic system (Casper & Yen, 1985;
Simpkins & Katovich, 1984), luteinizing hormone-releasing hormone
(LHRH) neuronal system (Gambone et al., 1984), and endogenous opioid
peptide system (Casper & Yen, 1985; Simpkins & Katovich, 1984) in its genesis.
The evidence that gonadal steroid hormones modulate or influence the
activity of each of these central hypothalamic neuronal systems is compelling
(McEwen et al., 1984; McEwen & Parsons, 1982; Roselli & Resko, 1990).
Biochemical, autoradiographic and immunohistochemical experiments have
demonstrated that the most dense collections of cells containing gonadal
steroid hormones and their receptors in the rat brain are found in the medial
preoptic area, hypothalamus, and in the limbic system structures (Luine et al.,
1975; Morrel et al., 1975; Pfaff & Keiner, 1973). These brain regions are key
elements in the neural circuits that regulate the neuroendocrine events of
reproduction and behavior (Barraclough & Wise, 1982; Goodman & Knobil;
Plant, 1986; Christensen & Clemens, 1974). A plausible hypothesis currently
held for the mechanism of the hot flush states that normally the opioidergic-
noradrenergic-LHRH neuronal systems are serially involved in coupling or
mediating the regulatory (feedback) influence of gonadal steroids on LH
secretion and perhaps thermoregulation (Casper & Yen, 1985). However, in
postmenopausal, as well as in long-term ovariectomized women, this neural

150
circuit, in particular the opioid neurons/receptors component, is uncoupled
from the LH release mechanism (Casper & Yen, 1985; Simpkins & Katovich,
1984). Hence, replacement of gonadal steroids in postmenopausal women
would exert a stabilizing influence on these neuronal network or the
mechanisms responsible for the flush response. Nevertheless, the
mechanism(s) through which gonadal steroids exert their stabilizing effects
remains unknown.
Although replacement therapy with estrogens and/or progestins has
been shown to be effective in most patients in alleviating the symptoms of
the menopause (Campbell & Whitehead,1977; Casper & Yen, 1985; Lauritzen,
1973; Upton, 1984; Yen, 1977), numerous retrospective studies indicated an
increased risk of peripheral toxicities, including the risk of breast and
endometrial cancer (Bergkvist et al., 1988; Berkowitz et al., 1985; Ettinger et al.,
1988; Persson, 1985), cardiovascular complications (Barrett-Connor et al., 1989;
Kaplan, 1978; Thomas, 1988), and alteration in hepatic metabolism (Burkman,
1988). The potential problem associated with currently used steroids
medication is that these hormones equilibrate among all body tissues due to
their high lipophilicity. As a result, only a fraction of the dose accumulates at
or near the site of action in the brain. Furthermore, steroid receptors are
present in many peripheral tissues creating the potential of untoward
peripheral site effects (Walters, 1985).
Given the aforementioned limiting factors in estrogen replacement
therapy, the preferential brain delivery of E2 with the E2-CDS may offer an
effective treatment strategy for postmenopausal symptoms by providing
sufficient E2 to the brain while avoiding peripheral toxicities.
This study was undertaken to assess the effects of E2-CDS on the rise in
tail-skin temperature (TST) in the morphine-dependent, naloxone-

151
withdrawal rat model developed in our laboratory (Simpkins & Katovich,
1984). This animal model was originally developed to study the
neuroendocrine mechanism(s) leading to the flush response as well as
evaluating new and existing drugs for their efficacy in treating the
menopausal symptoms. We have previously demonstrated that naloxone-
induced withdrawal in the morphine-dependent rat model results in a surge
in TST that is preceded by tachycardia, accompanied by hypersecretion of LH,
and is followed by a significant fall in core body temperature (Simpkins &
Katovich, 1984). Each of these responses is similar in both magnitude and
duration and temporally associated to those observed in menopausal hot
flushes (Simpkins et al., 1983).
Materials and Methods
To evaluate the effects of E2-CDS on TST responses, adult female rats
were bilaterally ovariectomized (OVX) under metofane anesthesia and
experiments were initiated 2 weeks after ovariectomy. On day 15 after
ovariectomy, rats were randomly divided into 4 groups (7-8 rats per group).
These experimental groups were designated and treated as follows: 1) OVX +
HPCD control, animals in this group received weekly iv injection of HPCD
(vehicle) for 3 weeks; 2) OVX + E2 pellet, these animals were subcutaneously
(sc) implanted (under light metofane anesthesia) with an E2 pellet weekly for
3 weeks (removing the old pellet before implanting a new one); 3) OVX + E2-
CDS multiple injections, these animals received weekly iv injection of E2-
CDS at a dose of 1.0 mg/kg b.w. for 3 weeks; and 4) OVX + E2-CDS single
injection, these animals received a single iv injection of E2-CDS at 1 week
before testing.

152
Morphine dependency was produced after initiation of estrogen
treatment as described previously (Katovich & O'Meara 1986; Simpkins &
Katovich, 1984). Briefly, one morphine pellet, containing 75 mg morphine
free base, was sc implanted at 17 days (long-term treatment, groups 1-3), or at 3
days (short-term treatment, group 4) after initiation of estrogen treatment.
Two days after the first morphine pellet, two additional morphine pellets
were sc implemented. This regimen of morphine treatment has been utilized
in our laboratory (Simpkins et al., 1983; Simpkins & Katovich, 1984) to
consistently produce typical symptoms of morphine dependency, tolerance,
and withdrawal (Wei et al., 1973). Four days after the initiation of morphine
treatment, on the morning of the twenty first day (long-term treatment) or
the seventh day (short-term treatment) after initiation of estrogen treatment,
animals were lightly restrained in wire mesh tunnel cages with a wooden
floor. TST was measured with a copper-constantan thermocouple that was
taped to the dorsal region of the tail at approximately 2 cm from its base so
that the thermocouple contacted the skin near the base of the tail. Rectal
temperature (RT) was measured with a copper-constantan thermocouple
inserted 6 cm beyond the anus and taped to the base of the tail. TST and RT
were recorded at 2-min intervals. Rats were allowed 1 hr to acclimate to the
restraining cages while control measurements were recorded. At the end of
the control period, rats were administered naloxone HC1 (0.5 mg/kg b.w., sc).
TST and RT were recorded for an additional 90 min in a room maintained at
24 1C. At the conclusion of the temperature study, all animals were killed
by decapitation and the trunk blood was collected in heparinized tubes. The
blood was centrifuged and the plasma separated and stored at -20C until
hormone analysis.

153
Our previous studies (Simpkins et al., 1983; Simpkins & Katovich,
1984) demonstrated that implants of placebo followed by naloxone
administration as well as implants of morphine which followed by saline
injection did not induce TST surge in these treatment groups. Therefore,
these control groups were not repeated again in the present study.
Plasma LH, FSH, and PRL concentrations were measured in duplicate
by RIA using NIDDK kits. Plasma LH, FSH, and PRL values are expressed as
ng/ml of either the LH-RP-2, FSH-RP-2 or the PRL-RP-3 reference standards,
respectively. The intra-assay coefficients of variation were 4.19%, 3.32%, and
4.4% for LH, FSH, and PRL assays, respectively. Plasma E2 concentrations
were measured in duplicate by the RIA employing Coat-A-Count Estradiol
kits. All the samples for each hormone were assayed in a single run.
Results
There was no significant effect of estradiol on the basal TST (Figure 24;
Table 11) or the basal RT (Figure 25; Table 11) in the morphine-dependent
animals. However, administration of naloxone (0.5 mg/kg) to the morphine-
dependent rats resulted in a rapid increase in TST and a subsequent decline in
RT (Figures 24 & 25; Table 11). The mean maximal elevation in TST were 6.4
0.2, 6.4 0.1,3.4 0.6, and 4.9 0.5C in the HPCD, E2 pellet, multiple E2-CDS,
and single E2-CDS groups, respectively (Table 11). Multiple injections of the
E2-CDS resulted in significant attenuation (more than 47%) of the naloxone-
induced maximal rise in TST (Figure 24; Table 11). A single injection of the
E2-CDS for 7 days also attenuated the maximal rise in TST by 25%, but this
treatment effect was not statistically significant. By contrast, treatment with
17 P-E2 (E2 pellet regimen) had no effect on the surge of TST (Figure 24; Table

154
11). The naloxone-induced surge in TST returned to normal range at the end
of 70 to 90 min of study.
The area under the TST curve (AUC) was significantly reduced by 42%
or 46% with multiple injections of E2-CDS compared with the HPCD control
or the E2 pellet, respectively. Likewise, treatment with a single injection of
E2-CDS reduced the AUC by 13% or 20% compared to the HPCD group or the
E2 pellet, respectively, however these effects were not statistically significant.
The mean maximal decline in RT was 2.3 0.3, 3.1 0.4, 3.3 0.3, and
2.5 0.4C in the HPCD, E2 pellet, multiple E2-CDS, or single E2-CDS groups,
respectively (Figure 25). Estrogen treatment with E2-CDS or E2 pellet had no
significant effect on Rt. The naloxone-induced decline in RT returned to
normal at 3 to 4 hrs after naloxone administration (data not shown).
Plasma E2 concentrations were significantly elevated with the E2*pellet
as well as with the multiple E2-CDS treatment (Table 12). However, the E2-
pellet treatment produced plasma E2 concentrations which were significantly
higher (2- to 6- fold) than those produced by single or multiple E2-CDS
treatment (Table 12).
Plasma gonadotropin concentrations (LH and FSH) were significantly
suppressed relative to HPCD control with the E2-pellet as well as the single
and multiple E2-CDS treatment (Table 12).
Plasma PRL concentrations in animals treated with either the E2 pellet
or the multiple E2-CDS were significantly elevated relative to PRL levels of
the HPCD group or the single-injected E2-CDS group. Furthermore, the
magnitude of PRL stimulation with the E2-pellet treatment was 1.5- and 3.5-
fold greater than that of multiple- or single-injected E2-CDS, respectively
(Table 12).

155
Discussion
In this study, we used the morphine-dependent, naloxone-withdrawal
rat model to evaluate the effectiveness of E2-CDS for the treatment of hot
flushes. To our knowledge, this is the only animal model (Simpkins &
Katovich, 1984) available to evaluate effectively alternative therapies for this
common disease.
The results of this study for the first time demonstrate that: (1) the E2-
CDS can attenuate significantly the naloxone-induced surge in TST of the
morphine-dependent rats; (2) 17 P-E2 pellet treatment has no effect on
naloxone-induced rise in TST; and (3) the significant effects of E2-CDS are
achieved at low plasma E2 concentrations. Furthermore, these data support
the hypothesis that the primary site of action where E2 exerts its stabilizing
effects on the mechanism(s) which control thermoregulation is in the CNS
and not in the periphery (Casper & Yen, 1985; Lauritzen, 1973; Yen, 1977).
This is clearly shown by the fact that treatment with 17 J3-E2 pellet which
produced superphysiological plasma E2 (2-fold greater than those produced by
the multiple injections of E2-CDS) did not exert any stabilizing effect on the
rise of TST. These pellets were shown in our laboratory to release high
concentrations of E2 (280-180 pg/ml) in plasma on day 1 and 2 after
implantation. However, between days 5 and 14 of implantation stable levels
of E2 (100-80 pg/ml) were observed. Contrary to our findings in the present
study, 17 P-E2 has recently been reported to produce significant attenuation in
the magnitude of the flush response using the same animal model (Katovich
& O'Meara, 1986). The reason for this discrepancy seems to be the result of
very high concentrations of plasma E2 in the previous study. Utilizing E2

156
pellet (Katovich & O'Meara, 1986) which produced very high plasma E2
concentrations (7-fold greater than E2 concentrations produced by E2 pellet in
the present study), would certainly lead to higher brain E2 levels. Therefore,
maintaining therapeutically effective E2 levels in the brain will eventually
produce brain-mediated E2 effect. Nevertheless, estrogens or estrogen-esters
are used therapeutically in postmenopausal patients (Campbell & Whitehead,
1977; Casper & Yen, 1985; Lauritzen, 1973; Upton, 1984; Yen, 1977), and these
agents are effective in alleviating menopausal symptoms (Campbell &
Whitehead, 1977; Paterson, 1982; Upton, 1984). Unfortunately, these
estrogenic products are administered either in frequent doses, or as a large
depot form, to achieve and maintain therapeutically effective levels in the
brain. Both of these treatment strategies lead to sustained increases in
peripheral estrogen levels which have been shown in numerous studies to
increase peripheral toxicides (Barrett-Conner et al., 1989; Burkman, 1988;
Campbell & Whitehead,1977; Casper & Yen, 1985; Kaplan, 1978; Lauritzen,
1973; Thomas, 1988; Trapido et al., 1984; Upton, 1984).
Our previous observations, including the E2-CDS kinetics in intact
male rats (Chapter 5) and in OVX rats (Chapter 6) as well as the
pharmacodynamic effects in OVX rats (Chapter 7), long-term suppression of
gonadotropin secretion in CAST male rats (Simpkins et al., 1986), long-term
suppression of androgen secretion and weights of androgen-responsive
tissues (Chapter 8), stimulation of masculine sexual behaviors in CAST male
rats for 28 days (Anderson et al., 1987a) following a single iv administration of
the E2-CDS and our present observation, that E2-CDS but not 17 (3-E2
significantly attenuates the naloxone-induced rise in TST, provide strong
support for the proposed mechanism of the E2-CDS. That is, following E2-
CDS administration, the delivery system undergoes rapid oxidation

157
providing the basis for "locking" the intermediate metabolite, E2-Q+, behind
the BBB, there it serves as a brain depot for E2. From this store of E2-Q+, E2 is
slowly released through non-specific hydrolysis of the carrier, resulting in
sustained brain exposure to E2.
Estrogen treatment of OVX morphine-dependent rats with both the E2-
CDS and E2 pellet resulted in significant suppression of gonadotropins (LH
and FSH) secretion. These effects of E2-CDS on gonadotropins are consistent
with the previously reported pharmacodynamic behaviors of the delivery
system (Anderson et al., 1987a,b, 1988a,b, 1989; Sarkar et al., 1989; Simpkins et
al., 1986,1988,1989a,b). The fact that the magnitude of LH suppression was
greater with E2-CDS treatment in the face of lower plasma E2 concentrations,
indicate that the prolonged and sustained inhibitory effects of E2-CDS are due
primarily to sustained suppression of LHRH secretion from the
hypothalamus (Sarkar et al., 1989). Sarkar et al. (1989) have reported
reduction in LHRH release into the hypophyseal portal system and no change
in pituitary responsiveness to LHRH following E2-CDS treatment.
An important question is whether morphine withdrawal in the rat is
analogous to the postmenopausal flushes. The remarkable similarities
between the symptoms of opiate withdrawal and the postmenopausal
syndrome (Simpkins et al., 1983) suggest a common underlying neuronal
mechanism(s) mediating these changes in both the addicted rats and the
postmenopausal women (Casper & Yen, 1985; Simpkins & Katovich, 1984).
Our observations that E2-CDS treatment in some animals (50% or less) did
not completely stabilize these underlying mechanism(s) in order to prevent
the flush response may argue against this animal model. This heterogeneity
in response to estrogen replacement therapy, however, is observed in
postmenopausal women as well. That is, first of all, the intensity and

158
frequency of the postmenopausal symptoms vary among women, and
symptoms do not occur at all in about 25% of postmenopausal women
(Casper & Yen, 1985). Also, it is important to note that not all
postmenopausal women on hormone replacement therapy respond
positively to the steroid medications (Paterson, 1982), even though it is
generally believed that the flushing response is due to steroid-hormone
withdrawal phenomenon (Casper & Yen, 1985; Lauritzen, 1973; Simpkins &
Katovich, 1984; Yen, 1977).
It is also important to point out that the length of exposure to E2 via E2-
CDS may be as important as the dose administered to the morphine-
dependent rats. Perhaps the variability or heterogeneity in response is due to
the fact that all animals may eventually respond to E2-CDS treatment, but it
may take variable durations of sustained E2 exposure via E2-CDS treatment in
order to observe an attenuation in TST response.
In conclusion, these results support the view that the E2-CDS may be
potentially useful in the treatment of brain E2 deficiencies (i.e., vasomotor hot
flushes). Further experimental as well as clinical investigations pertaining to
the therapeutic efficacy of the E2-CDS in this regard is warranted. In
comparison to the currently used estrogenic products, the E2-CDS should
achieve the sustained stimulation of brain E2 receptors at lower doses or with
less frequent dosing.

159
-2 i 1 1 1 1 1 1 1 1 1 1 1 1 1
-40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100
Time (min)
Figure 24. Effects of the E2-CDS or E2 pellet on the mean TST responses
induced by naloxone administration (0.5 mg/kg; sc) to morphine-
dependent ovariectomized rats. Animals were treated weekly
with either the vehicle (HPCD, x3, 3 injections total over 3 weeks),
E2-CDS (1.0 mg/kg, x3, 3 injections total over 3 weeks), E2 pellet
(0.5 mg, x3, 3 implants total over 3 weeks), or E2-CDS (1.0 mg/kg,
xl, 1 injection for 1 week). Morphine dependency was produced
after initiation of estrogen treatment. Four days after the
initiation of morphine treatment, on the morning of the 21st day
(long-term) or the 7th day (short-term) after the initiation of
estrogen treatment, animals were lightly restrained in wire mesh
cages. TST was then recorded at 2-min intervals. Represented are
the means SEM for n=7-8 rats per group.

160
Time (min)
Figure 25. Effects of the E2-CDS or E2 pellet on the mean RT responses
induced by naloxone administration (0.5 mg/kg; sc) to morphine-
dependent ovariectomized rats. Animals were treated weekly
with either the vehicle (HPCD, x3, 3 injections total over 3 weeks),
E2-CDS (1.0 mg/kg, x3, 3 injections total over 3 weeks), E2 pellet
(0.5 mg, x3, 3 implants total over 3 weeks), or E2-CDS (1.0 mg/kg,
xl, 1 injection for 1 week). Morphine dependency was produced
after initiation of estrogen treatment. Four days after the
initiation of morphine treatment, on the morning of the 21st day
(long-term) or the 7th day (short-term) after the initiation of
estrogen treatment, animals were lightly restrained in wire mesh
cages. RT was then recorded at 2-min intervals. Represented are
the means SEM for n=7-8 rats per group.

161
Table 11: Effects of the E2-CDS or E2 pellet on Basal Temperature, Maximal
Change in TST, and Area Under the 90 Min TST Curve in Ovariectomized,
Morphine-Dependent Rats
Treatment
Group
Basal RT
Basal TST
Max A TST
AUC
HPCD
(3)*
39.8 0.2
(C)
26.4 0.1
6.4 0.2
(C-Min)
335.5 24.9
E2 Pellet
(3)
38.7 0.2
26.2 0.1
6.4 0.1
361.8 19.4
E2-CDS
(3)
39.1 0.3
26.9 0.4
3.4 0.6a,b
197.8 35.3a,b
e2-cds
(1)
38.6 0.3
26.5 0.1
4.9 0.5
291.5 32.8
* Number of injection(s)/pellet(s)
a Different from HPCD group
b Different from E2-pellet group

Table 12. Effects of the E2-CDS or E2 Pellet on Plasma Hormone Concentrations in
OVX, Morphine-Dependent Rats.
Treatment
Group
2
E?-Q+
m
FSH
PRL
(pg/ml)
(pg/ml)
(ng/ml)
(ng/ml)
(ng/ml)
HPCD (3)*
3.67 0.95
ND
10.11 2.77
40.53 3.24
15.49 9.93
E2 Pellet (3)
89.80 9.97a,b
ND
0.90 0.19a
13.26 1.23a,c
960.50 100.4a,c
E2-CDS (3)
44.10 6.37a,c
866.66 53.42c
0.43 0.08a
12.16 1.69a,c
677.64 86.72a,c
E2-CDS (1)
13.45 3.53
224.12 17.88
0.57 0.09a
20.43 1.06a
279.50 96.85
ND Not determined
* Number of injection(s)/pellet(s)
a Different from HPCD group
b Different from E2-CDSO) or E2-CDSO) group
c Different from E2-CDSO) group
162

CHAPTER 10
GENERAL DISCUSSION
The studies in this dissertation evaluated the pharmacokinetics and
pharmacodynamic consequences of the redox-based estradiol-chemical
delivery system (E2-CDS) for the brain in the rat. The synopses of the major
findings drawn from these studies are (1) development of a technique for the
simultaneous quantitation of E2-Q+ and E2 that is reliable, sensitive and
applicable to a wide variety of tissues (Chapter 4; Rahimy et al., 1989a); (2)
documentation of the preferential deposition and retention of the E2-CDS
metabolites, E2-Q+ and E2, in the CNS tissue with a ti /2 = 8-9 days in male
(Chapter 5; Rahimy et al., 1988,1990a) and in OVX female rats (Chapter 6;
Rahimy et al., 1990b); (3) demonstration of the relatively rapid disappearance
of these metabolites from various peripheral tissues (Chapters 5 & 6; Rahimy
et al., 1988, 1990a,b); (4) determination of the prolonged suppression of
gonadotropin secretion LH and FSH in OVX female rats in a dose- and time-
dependent manner (Chapter 7; Rahimy et al., 1989b, 1990c); (5) demonstration
of the sustained suppression of T secretion and weight of androgen-
responsive tissues equivalent in magnitude to that of castration level
(Chapter 8); and (6) demonstration of significant attenuation of the naloxone-
induced surge in TST of morphine-dependent rats in the face of very low
plasma E2 levels (Chapter 9).
The formidable task of delivering the drugs of choice to the CNS has
long been recognized particularly by neuropharmacologists. This is because of
the unique feature of the brain, the BBB, that allows only lipophilic agents to
163

164
gain access to the CNS. As a result, many potentially useful therapeutic
agents, i.e. water soluble drugs are excluded from entering the brain. To
overcome this barrier and thus, to enhance CNS drug concentration, various
strategies for drug flux through the BBB have been developed. These include:
(a) physical approaches, i.e. intraventricular/intrathecal infusion and
implantable pump (Hammond, 1988); and (b) chemical approaches, i.e.
prodrugs (Bodor, 1981, 1985,1987; Sinkula & Yalkowsky, 1975; Stella, 1975),
liposomes (Weiner et al., 1989), and the use of membrane transport systems to
deliver nontransportable peptides through the BBB (Pardridge, 1986).
The intraventricular administration of drugs, in addition to being an
invasive technique, is inefficient in cases where the drugs of choice are polar
and highly water soluble. For instance, the intraventricular administration of
polar drugs results in uneven or incomplete distribution in the brain since
these agents are solubilized primarily in the aqueous compartment (CSF).
Furthermore, this approach tends to bath the surface of the brain, perhaps
because the efflux of the drug out of the ventricles and into the superior
sagittal sinus is much faster than diffusion into the brain parenchyma
(Pardridge, 1988a).
A general, more practical approach to increase brain concentrations of
water soluble drugs and thus their therapeutic efficacy has been the design of
prodrug formulation (Bodor, 1981, 1985; Sinkula & Yalkowsky, 1975; Stella,
1975). The purpose of prodrug modification is to increase the concentration
of the active drug at or near its site of action, thereby increasing its efficacy.
However, by increasing the lipophilicity of a drug nonspecifically via the
prodrug approach, it may not only enhance its diffusion through the BBB ,
but also enables the uptake of the compound into all other tissues and thus,
exposure to a greater drug burden. This method of nonselectivity of drug

165
delivery is the major limiting factor in the prodrug design, specially those
with cytotoxicity or those with broad spectrum of peripheral sites of action
such as steroids.
A more recent, pharmacologic-based strategy which has shown great
potential for drug delivery is the liposome preparation (Weiner et al, 1989).
An assortment of drugs, including peptide and protein compounds, may be
incorporated in the liposomes, which can then be administered by different
routes. The physicochemical properties of the liposomes allows the
encapsulation of a drug molecule either in the aqueous space or intercalation
into the lipid bilayer matrix, depending on the properties of drugs.
Liposomes are, however, taken up by cells lining the reticuloendothelial
system, and do not appear to be useful for drug delivery through the BBB.
The paradoxical reason for this may be the larger size of these vesicles which
prevent them from crossing the BBB via lipid-mediated transport
mechanisms even though liposomes are highly lipid soluble (Pardridge,
1988a).
The most highly developed and promising strategy for improving drug
delivery through the BBB is the coupling of water- or lipid-soluble drugs to
the redox-based dihydropyridine nucleus (Bodor, 1987; Bodor & Brewster,
1983; Bodor et al., 1981). This strategy for the CNS drug delivery offers several
advantages. First, the application of the carrier, dihydropyridine, to a drug
increases the lipid solubility of the drug, because of highly lipid-soluble
nature of the carrier. Second, the carrier exhibits an intermediate enzymatic
oxidation to a quaternary pyridinium ion, which encourages preferential
brain deposition and retention by "locking" the charged, oxidative metabolite
behind the BBB; enzymatic hydrolysis of the charged-pyridinium drug
complex in a subsequent step provides sustained release of the parent drug in

166
the brain. Finally, the carrier system simultaneously enhances the rate of
elimination of the drug, specially if lipoidal in nature, in an inactive form,
from peripheral tissues following the oxidation to a hydrophilic quaternary
form.
The present work evaluated this redox-based chemical-delivery
approach to brain estrogen delivery. To determine reliably the tissue
distribution and thus to document the effectiveness of E2-CDS, a specific and
sensitive method was essential (Chapter 4; Rahimy et al., 1989a). Such criteria
are extremely important, particularly for the E2-CDS, since the E2-CDS
metabolite, E2, is not only present in low concentration but also active at low
pg/g tissue. Furthermore, the intermediate metabolite, E2-Q+, is present in
concentrations much higher than E2. Thus, to accurately quantitate E2 levels
in tissue samples, the assay method must be capable of distinguishing low
levels of E2 in the presence of high concentrations of E2-Q+. These problems
were resolved by employing and optimizing the reproducibility of an RIA
procedure for E2 determination in all tissues and fluid (Chapter 4). The RIA
provides the needed sensitive end point (0.8 to 1.2 pg/assay tube) as well as
the required specificity (cross-reactivity of <0.3% for E2-Q+ at concentration of
15 ng/ml and higher).
Extensive and detailed evaluation of the tissue distribution of E2-CDS
in both intact male (Chapter 5; Rahimy et al., 1988,1990a) and OVX female
rats (Chapter 6; Rahimy et al., 1990b) support the concept of brain-enhanced
delivery and sustained release of E2 using the redox-based carrier system
(Bodor et al., 1987). Furthermore, the distribution patterns of both E2-Q+ and
E2 in male (Chapter 5; Rahimy et al., 1990a) and in female rats (Chapter 6;
Rahimy et al., 1990b) confirmed the major aspect of the proposed mechanism
of the E2-CDS drug delivery. Interestingly, the extent of deposition and the

167
chronic retention of these metabolites by the CNS are quite comparable in
male and OVX female rats. The estimated half-lives of E2-Q+ and E2 (ti /2 = 8-
9) in brain tissue in these studies are in agreement with other reports which
utilized different analytical techniques and E2-CDS doses (Mullersman et al.,
1988). Taken together these findings indicate the following: (1) as predicted
based on the physicochemical properties of the E2-CDS (Bodor et al., 1987), the
"locking of E2-Q+ in the CNS tissue had occurred and the unique features of
the BBB are the contributing factors to the chronic retention of the charged,
hydrophilic E2-Q+; (2) while the enzymatic oxidation of E2-CDS to E2-Q+ and
the hydrolysis of E2-Q+ to E2 exhibit dose-dependency, the disappearance of
these metabolites from the CNS tissue appear to be independent of dose. This
is supported by the fact that consistent results have been obtained in several
studies using doses of E2-CDS ranging from 0.01 mg/kg (Chapter 6 & 7;
Rahimy et al., 1990a, b) to 15 mg/kg dose (Mullersman et al., 1988).
Determination of distribution of E2-Q+ and E2 in peripheral tissues of
male (Chapter 5) and female rats (Chapter 6) revealed some similarities and
differences. The anterior pituitary of both male and female rats showed
slower elimination of E2-Q+ and E2 compared with other peripheral tissues.
In fact, the elimination of these metabolites from this tissue appeared more
like the CNS tissue. The concentrations of E2-Q+ and E2 in anterior pituitary
were below that of brain levels of these compounds initially and steadily
decreased throughout the observation period (Chapters 5 & 6). This relative
persistency of E2-Q+ and specially of E2 in this particular tissue are most likely
because of anatomical relationship between the hypothalamus and the
anterior pituitary gland. Estradiol released from the E2-Q+, or the E2-Q+ itself,
which is "locked" into the brain, could be delivered directly to the anterior
pituitary by the capillary plexus of the hypophyseal portal system (Traystman,

168
1983). These capillaries in the median eminence lack features of other brain
capillaries and hence are not part of the BBB (Traystman, 1983).
Likewise, plasma showed, following administration of E2-CDS to male
(Chapter 5) or OVX female rats (Chapter 6), a residual but detectable levels of
E2 for approximately two weeks after a single iv dose of E2-CDS treatment.
Since E2 is active at low pg concentrations, these plasma levels of E2 may
contribute to some of the long-term effects of E2-CDS, i.e. uterine tissue
stimulation or pituitary responses (Chapter 7). The source of this sustained
residual E2 in plasma is likely to be the result of a continuous redistribution
of E2 liberated from E2-Q+ in the brain down its concentration gradient into
the general circulation.
In contrast to CNS tissue, the distribution of E2-CDS metabolites in
male (Chapter 5) and OVX female rats (Chapter 6) showed sex differences with
respect to the elimination patterns of E2-Q+ and E2 from certain peripheral
tissues. The more obvious one was the heart tissue. Levels of E2-Q+ in heart
tissue of the male rat, following administration of E2-CDS, were
approximately 2-fold greater than levels of this metabolite in female
counterparts (Chapters 5 & 6). Furthermore, the rate of elimination of this
compound from heart tissue of male rat was slower than that of heart tissue
of OVX female rats. Lung tissue of male rat also showed higher levels of both
E2-Q+ and E2 than those of female rats following administration of E2-CDS.
The observed sex differences with respect to E2-CDS metabolism and
elimination patterns of these metabolites from these tissues may be related to
differences in kinetics of enzymes involved. For instance, if there are tissue
specific rates of E2 metabolism that can alter the equilibrium between E2-Q+ <=>
E2 <=> E2 metabolites, that may lead to the observed differences in the
disappearance of E2-Q+ from these tissues. Furthermore, if a certain tissue, i.e.

169
the female heart tissue, contains E2 inducible enzymes, i.e. cytochrome P450
monooxigenases that metabolize E2, this may be the reason why this tissue
exhibits faster clearance of E2-CDS metabolites.
An important issue that requires further explanation is the initial
distribution phase of the E2-CDS throughout the body before its oxidation to
E2-Q+. Since the iv route was chosen here for drug administration, this route
completely eliminates the process of absorption. Thus, the extent of
distribution or accessibility of E2-CDS to its ultimate site of action is
determined by its ability to cross the capillary endothelial cells, then by the
rate of blood flow through organs and tissues and, finally, since the parent
drug likely acts intracellularly, by its rate of diffusion across the cellular
plasma membranes. Since the E2-CDS is very lipophilic in nature, it is
capable of readily traversing the capillary endothelia and the plasma
membranes, thus, reaching inside the cells if it survives metabolism and
elimination during the initial period of distribution. Therefore, of major
concerns are the general factors which may influence the amount of E2-CDS
eventually reaching and residing in different tissues. Since we do not know
the extent of E2-CDS binding to plasma proteins and other tissue components,
for the present, we assume that these important issues are not critical. Thus,
this general treatment of E2-CDS kinetic behavior permits us to draw the
following conclusions regarding the initial tissue distribution of E2-CDS.
First, the quantity of E2-CDS reaching an organ or tissue represents a small
fraction of the total amount of drug administered. Second, the greater the
rate of blood flow to an organ or tissue the higher the quantity of E2-CDS
reaching that organ or tissue. Third, the greater the extraction efficiency from
the circulation of E2-CDS by an organ or tissue the higher the quantity of E2-
CDS that is taken up by that organ or tissue. It should be emphasized that

170
these factors may be important only if the E2-CDS does not achieve an
equilibrium or steady estate within 30 min after its administration.
Otherwise, since the E2-CDS does not bind to estrogen receptors itself, and
since it is very lipophilic, it would be redistributed among all tissues and thus,
the contribution of blood flow would be of minimal importance. However,
after the oxidation of E2-CDS to E2-Q+, it is then the unique property of an
organ or tissue, i.e. BBB, that serves the bases for retaining the oxidized
metabolite within a tissue.
Several other important issues regarding the regional distribution
and/or localization of the E2-CDS metabolites, E2-Q+ and E2, remain to be
resolved. Based upon our observations and comparisons of the whole brain
and the hypothalamus (Chapters 5 & 6) and the comparisons made by Sarkar
et al. (1989) of the hypothalamus, preoptic area and the cerebral cortex, it
seems that the E2-CDS is distributed evenly throughout the brain.
Furthermore, no obvious differences were observed with respect to its
oxidation, hydrolysis, and clearance from these brain regions evaluated to
date. This suggests that the oxidation and the hydrolysis reactions leading to
E2 formation are not unique to a particular set of neurons or even other cells
in the body.
Regarding the enzymatic conversion and cellular localization of the E2-
CDS metabolites, currently there are no documented reports to demonstrate
that the E2-CDS metabolism is an intracellular phenomenon. However, it is
believed that the multiple enzymatic reactions are primarily an intracellular
events. Using rat tissue homogenates as the test matrix, the in vitro
observation showed a faster oxidation rate of E2-CDS to E2-Q+ in tissue
homogenates than in plasma (Bodor et al., 1987). That suggests the
involvement of membrane-bound enzyme, most likely the ubiquitous NAD

171
<=> NADH transhydrogenase, in mediating the oxidation step (Hoek &
Rydstrom, 1988). This enzyme/coenzyme system is primarily located in the
inner mitochondrial membrane. In fact, the E2-CDS was designed to
biomimic this coenzyme moiety for its oxidative metabolism. Furthermore,
since the oxidoreductive enzymes are widespread, it is expected that all tissues
are capable of converting the E2-CDS to the corresponding quaternary
pyridinium salt (E2-Q+). Likewise, the hydrolysis of E2-Q+ to E2 is mediated by
widespread non-specific esterases present in tissues (Bodor et al., 1987;
Brewster et al., 1987). We examined the disappearance of E2-Q+ to E2 in tissue
homogenates and blood, approximately 1.2% of an E2-Q+ dose was converted
to E2 in brain homogenate, 3.1% in whole rat blood, and 2.3% in liver after
120 min incubation of E2-Q+. The slow rate of hydrolysis of E2-Q+ to E2 in an
in vitro condition is consistent with the in vivo sustained release of E2 in
brain tissue over 28 days following administration of E2-CDS in OVX female
rats (Chapter 6; Rahimy et al., 1990b). Certainly, more detailed studies
regarding the cellular localization of these enzymatic events and their
products will lead to further insight into the understanding of mechanism of
action of this delivery system.
The pharmacodynamic data demonstrates that the E2-CDS causes dose-
dependent and chronic suppression of LH and FSH secretion in OVX rats,
despite the rapid peripheral clearance of the E2-CDS metabolites (Chapter 7;
Rahimy et al., 1990c). The time-course and the magnitude of gonadotropin
suppression observed are quite comparable to the previously reported chronic
effects of the E2-CDS following a single iv administration to several animal
models (Anderson et al., 1987a,b, 1988a,b, 1989; Estes et al., 1987a b, 1988; Sarkar
et al., 1989; Simpkins et al., 1986,1988,1989a,b). These sustained and
prolonged pharmacological effects further support the idea that the E2-Q+ is

172
"locked" behind the BBB and there it serves as a brain depot for E2 (Bodor et
al., 1987).
The present work also evaluated the potential for clinical application of
E2-CDS in several estrogenically responsive animal models including
androgen-dependent prostatic tissue growth/hyperplasia (Chapter 8) and
morphine-dependent, naloxone-withdrawal rat model for menopausal hot
flushes (Chapter 9). The encouraging results obtained from these
experimental studies indicate that the E2-CDS may have potential application
for the effective treatments of androgen-dependent prostatic hyperplasia as
well as vasomotor hot flushes of postmenopausal women. Although we
used normal male rats (Chapter 8) as the model for prostatic adenocarcinoma
to examine the effectiveness of E2-CDS in reducing testosterone levels with
subsequent regression of the prostatic tissue, there are no conclusive data as to
whether this model reflects a complete picture analogous to that of the aging
human prostate. However, since the primary objective of endocrine therapy
in prostate malignancy is the induction of an effective androgen deprivation,
thus abolishing the growth promoting effects of androgens on the diseased
prostate, the E2-CDS is quite capable of inducing an effective chemical
castration-like effect (Chapter 8). Indeed, the E2-CDS at a single dose of 0.5
mg /kg was as effective as CAST not only in suppressing T levels but also in
regression of the androgen-sensitive prostatic tissue. Further experimental
(animal models for prostatic adenocarcinoma) as well as clinical investigation
pertaining to evaluation of therapeutic efficacy of this novel E2-CDS in this
regard as well as the menopausal hot flushes are warranted.
Finally, some of the potential complications or disadvantages of this
drug delivery system need to be mentioned. Although brain-enhanced
delivery with sustained release of E2 in that tissue via a redox-based system

173
has potential application for various clinical conditions, it may have some
drawbacks due to its mechanism of drug delivery that is, locking the drug into
the CNS tissue for prolonged period of time. Potential problems may include
the following: 1) The most frequent, unpleasant side effects of estrogen
hormones, particularly E2, are nausea and vomiting. These adverse effects
are due most likely to their central effects. Having "locked" irreversibly a
depot E2 inside the brain, in the form of E2-Q+, with sustained release of the
hormone over prolonged period of time may cause complications in this
regard. 2) In clinical practice, it is the standard procedure to monitor drug
concentration in patients by taking blood samples and analyzing plasma or
serum to obtain necessary informations about the drug distribution,
metabolism, and perhaps drug concentration at the site of action. However,
with the application of E2-CDS, plasma or serum concentration of the E2-CDS
metabolites do not reflect the brain levels of these compounds, since these
metabolites are preferentially retained by the brain. Thus, this feature of the
E2-CDS may be a disadvantage with regard to drug monitoring. 3) Certain
regions of the brain, i.e. arcuate nucleus of the hypothalamus, are exquisitely
susceptible to prolonged exposure and/or high concentration of estrogen
hormones. Since E2-CDS metabolites persist evenly throughout the brain for
prolonged period of time after drug administration, potential for
neuropathological lesions by this delivery may exist. 4) Perhaps the most
important issue regarding this drug delivery would be the question of
estrogen receptors down regulation or desensitization with prolonged
duration of exposure to the hormone. So, if estrogen receptors down
regulate, like other receptor types, the prolonged residence and the sustained
release of E2 in the brain may not be useful for the duration of residence in
that tissue. And finally, although we chose the iv route for drug

174
administration in experimental animals in our studies, this route is not a
preferred avenue of drug administration in humans. However, it should be
mentioned that recently other routes of E2-CDS administration have been
examined, and it appears that oral mucosal or buccal route may be as effective
as iv route for E2-CDS administration.
In conclusion, the work presented in this dissertation demonstrate that
the redox-based dihydropyridine chemical delivery system is capable of
preferential delivery of E2 to the brain by "locking" it in the form of E2-Q+
which then allows to be released in slow and sustained manner. The results
further support the idea that the E2-CDS may be potentially useful in fertility
regulation and effective treatment of androgen-dependent prostatic diseases
by virtue of selective and sustained suppression of gonadotropin secretion,
and in treatment of brain estradiol deficiencies, i.e. postmenopausal
syndrome. In comparison to the currently used estrogenic medications, the
E2-CDS should achieve the sustained stimulation of brain E2 receptors at
lower doses and with less frequent dosing.

REFERENCES
Abraham G.E. Radioimmunoassay of steroids in biological materials. Acta
Endocrinol. (Suppl.) 183:1-42 (1974).
Allen E. and E.A. Doisy. An ovarian hormone: A preliminary report on its
localization, extraction, and partial purification, and action in test
animals. TAMA 81:819-821 (1923).
Allen L.S., M. Hines, J.E. Shryne, and R.A. Gorski. Two sexual dimorphic cell
groups in the human brain. T. Neurosci. 9:497-506 (1989).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Evidence for
the reestablishment of copulatory behavior in castrated male rats with
a brain-enhanced estradiol-chemical delivery system. Pharmacol.
Biochem. Behav. 27:265-271 (1987a).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Evidence for
suppression of serum LH without elevation in serum estradiol or
prolactin with a brain-enhanced redox delivery system for estradiol.
Life Sri. 42:1493-1502 (1988a).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Effects of a
brain-enhanced chemical delivery system for estradiol on body weight
and serum hormones in middle-aged rats. Endocr. Res. 14:131-148
(1988b).
Anderson W.R., J.W. Simpkins, M.E. Brewster, and N. Bodor. Prolonged
suppression of androgens and androgen-dependent tissues with a
brain-enhanced delivery system for estradiol. Endoc. Soc. Seattle,
Washinton (1989).
Anderson W.R., J.W. Simpkins, P.A. Woodard, W.C. Stern, and N. Bodor.
Anxiolytic activity of a brain delivery system for GABA.
Psychopharmacology 92:157-163 (1987b).
Backstrom T., M. Bixo, and S. Hammarback. Ovarian steroid hormones. Acta
Obstet. Gynecol. Scand. [Suppl.] 130:19-24 (1985).
175

176
Bain J.D., L.H. Kasman, A.B. Bercovitz, and B.L. Lasley. A comparison of
three methods of hydrolysis for estrogen conjugates. Steroids 43:603-
619 (1984).
Barraclough C.S. and P.M. Wise. The role of catecholamines in the regulation
of pituitary luteinizing hormone and follicle-stimulating hormone
secretion. Endocr. Rev. 3:91-119 (1982).
Barrett-Connor E., D.L. Wingard, and M.H. Criqui. Postmenopausal estrogen
use and heart disease risk factors in the 1980's. TAMA 261:2095-2100
(1989).
Beard J. The span of gestation and the cause of birth. Gustav Fischer Verlag,
Jena (1897).
Becker J.B. Direct effect of 170-estradiol on striatum: Sex differences in
dopamine release. Synapse 5:157164 (1990).
Becker J.B. and M.E. Beer. The influence of estrogen on nigrostriatal
dopamine activity: Behavioral and neurochemical evidence for both
pre- and postsynaptic components. Behav. Brain Res. 1:27-33 (1986).
Bergkvist L., I. Persson, H-O Adami, and C. Schairer. Risk factors for breast
and endometrial cancer in a cohort of women treated with menopausal
oestrogens. Int. T. Epidemiol. 17:732-737 (1988).
Berkowitz G.S., J.L. Kelsey, T.R. Holford, V.A. Livolsi, M.J. Merino, G.J. Beck,
S. Ort, T.Z. O'Connor, and C. White. Estrogen replacement therapy
and fibrocystic breast disease in postmenopausal women. Am. T.
Epidemiol. 121:238-245(1985).
Berry S.J., D.S. Coffey, P.C. Walsh, and L.L. Ewing. The development of
human benign prostatic hyperplasia with age. T. Urol. 132:474-479
(1984).
Beyer C, G. Morali, F. Naftolin, K. Larsson, and G. Perez-Palacios. Effects of
some antiestrogens and aromatase inhibitors on androgen induced
sexual behavior in castrated male rats. Horm. Behav. 7:353-362 (1976).
Blackard C.E. The Veterans' Administration Cooperative Urological
Research group studies of carcinoma of the prostate: A review. Cancer
Chemother. Rep. 59:225-227 (1975).
Blackard C.E., D.P. Byar, and W.P. Jordan. Orchiectomy for advanced prostatic
carcinoma: a reevaluation. Urology 1:553-560 (1973).

177
Bodor.N. Novel approaches to prodrug design. Drug of the Future 6:165-182
(1981).
Bodor N. Prodrugs versus soft drugs. In: Design of Prodrugs. H. Bundgaard
(ed) 333-354, Elsevier Science Publishers BV, Amsterdam (1985).
Bodor N. Redox drug delivery for targeting drug to the brain. Ann. NY Acad.
Sd. 507:289-306 (1987).
Bodor N. and M.E. Brewster. Problems of delivery of drugs to the brain.
Pharmacol. Ther. 19:337-386 (1983).
Bodor N., A. El-Koussi, M. Kano, T. Nakamura, and M. Khalifa. Improved
delivery through biological membranes 26. Design, synthesis and
pharmacological activity of a novel chemical delivery system for (3-
adrenergic blocking agents. T. Med. Chem. 31:100-106(1988).
Bodor N. and H.H. Farag. Improved delivery through biological membranes
XI. A redox chemical drug-delivery system and its use for brain-specific
delivery of phenethylamine. T. Med. Chem. 26:313-318 (1983).
Bodor N. and H.H. Farag. Improved delivery through biological membranes
XIV. Brain-specific, sustained delivery of testosterone using a redox
chemical delivery system. T Pharm. Sci. 73:385-389 (1984).
Bodor N., H.H. Farag, and M.E. Brewster. Site specific, sustained release of
drugs to the brain. Science 214:1370-1372 (1981).
Bodor N. and J. Kaminski. Prodrug and site-specific chemical delivery
systems. Ann. Rpts. Med. Chem. 22:303-313 (1987).
Bodor N., J. McCornack, and M.E. Brewster. Improved delivery through
bioloical membranes. XXII. Synthesis and distribution of brain-
selective estrogen delivery systems. Int. T. Pharm. 35:47-59 (1987).
Bodor N. and J.W. Simpkins. Redox delivery system for brain-specific,
sustained release of dopamine. Science 221:65-67 (1983).
Bolt H.M. Metabolism of estrogensnatural and synthetic. Pharmacol. Ther.
4:155-181 (1979).
Bonney R.C., C. A. Sparks, R.W. Cheng, M.J. Reed, and V.H. James. The
measurement of steroid hormones in endometrial tissue: A
comparison between two methods of extraction. T. Steroid Biochem.
21:479-481 (1984).

178
Bottiger L.E., G. Boman, and G. Eklund. Oral contraceptives and
thermoboembolic disease: Effects of lowering oestrogen content.
Lancet 1:1097-1101 (1980).
Bowman K.M. and L. Bender. The treatment of involution melancholia with
ovarian hormone. Am. T. Psvchiat. 11:867-893 (1932).
Brawer J.R., H. Schipper, and F. Naftolin. Ovary-dependent degeneration in
the hypothalamic arcuate nucleus. Endocrinology 107:274-279 (1980).
Brawer J.R., H. Schipper, and B. Robaire. Effects of long term androgen and
estradiol exposure on the hypothalamus. Endocrinology 112:194-199
(1983).
Brendler C.B. The current role of hormone therapy in the clinical treatment
of prostatic cancer. Seminars in Urol. 6:269-278 (1988).
Brewster M.E., K.S. Estes, and N. Bodor. Improved delivery through
biological membranes 32. Synthsis and biological activity of brain-
targeted delivery system for various estradiol derivatives. T. Med.
Chem. 31:244-249 (1987).
Briggs M. Biochemical effects of oral contraceptives. Adv. Steroid Biochem.
Pharmacol. 5:66-160(1976).
Brightman M.W. Morphology of blood-brain interfaces. Expt. Eye Res.
{suppl.] 25:1-25 (1977).
Brightman M.W. and T.S. Reese. Junctions between intimately opposed cell
membranes in the vertebrate brain. T. Cell Biol. 40:648-677 (1969).
Broadwell R.D. and M. Salcman. Expanding the definition of the blood-brain
barrier to protein. Proc. Natl. Acad. Sri. USA 78:7820-7824 (1981).
Bunney W.E., Jr., and B.L. Garland. A second generation catecholamine
hypothesis. Pharmacopsychiat. 15:111-115 (1982).
Burkman R.T. Lipid and lipoprotein changes in relation to oral contraception
and hormone replacement therapy. Fertility and Sterility [Suppl.]
49:39S-50S (1988).
Butcher R.L., W.E. Collins, and N.W. Fugo. Plasma concentration of LH,
FSH, prolactin, progesterone and estradiol-17(J throughout the 4-day
estrous cycle of the rat. Endocrinology 94:1704-1708 (1974).

179
Butenandt A. Uber "PROGYNON" ein crystallisiertes, weibliches
sexualhormone. Naturwissenschaften 17:879 (1929).
Cabot A.T. The question of castration for enlarged prostate. Ann. Surg.
24:265-309 (1896).
Campbell S. and M. Whitehead. Oestrogen therapy and the menopausal
syndrome. Clin. Obstet. Gynaecol. 4:31-47 (1977)
Canick J.A., D.E. Vaccaro, E.M. Livingston, S.E. Leeman, K.J. Ryan, and T.O.
Fox. Localization of aromatase and 5 a-reductase to neuronal and non
neuronal cells in the fetal rat hypothalamus. Brain Res. 372:277-282
(1986).
Carlstrom K., L. Collste, A. Erikson, P. Henriksson, A. Pousette, R. Stege, and
B. von Schoultz. A comparison of androgen status in patients with
prostatic cancer treated with oral and/or parenteral estrogens or by
orchidectomy. The Prostate 14:177-182 (1989).
Casper, R.E. and S.S.C. Yen. Neuroendocrinology of menopausal flushes: An
hypothesis of flush mechanism. Clin. Endocrinol. 22:257-267 (1985).
Casper R.F., S.S.C. Yen, and M.M. Wilkes. Menopausal flushes: A
neuroendocrine link with pulstil luteinizing hormone secretion.
Science 205:823-925 (1979).
Chen C.L. and J. Meites. Effects of estrogens and progesterone on serum and
pituitary prolactin levels in ovariectomized rats. Endocrinology
86:503-508 (1970).
Christensen L.W. and L.G. Clemens. Intrahypothalamic implants of
testosterone or estradiol and resumption of masculine sexual behavior
in long-term castrated male rats. Endocrinology 95:984-990 (1974).
Clayden J.R., J.W. Bell,and P. Pollard. Menopausal flushing: Double-blind
trial of a non-hormonal medication. Br. Med. T. 1:409-412 (1974).
Cooper K.J., C.P. Fawcett, and S.M. McCann. Inhibitory and facilitatory effects
of estradiol 17|3 on pituitary responsiveness to a luteinizing hormone-
follicle stimulating hormone releasing factor (LH-RF/FSH-RF)
preparation in the ovariectomized rat. Pro. Soc. Expt. Biol. Med.
145:1422-1426 (1974).
Corner G.W. and W.M. Allen. Physiology of the corpus luteum. II.
Production of a special uterine reaction (progestational proliferation) by
extracts of the corpus luteum. Am. T. Physiol. 88:326-346 (1929).

180
Cortes-Gollegos V. and A.J. Gallegos. Estrogen peripheral levels vs. estrogen
tissue concentration in the human female reproductive tract. T. Steroid
Biochem. 6:15-20(1975).
Coyle J.T., D.L. Price, and M.R. DeLong. Alzheimer's disease: A disorder of
cortical cholinergic innervation. Science 219:1184-1190 (1983).
Cruickshank J.M., G. Neil-Dwyer, M.M. Cameron, and J. McAinsh. (3-
Adrenoreceptor-blocking agents and the blood-brain barrier. Clin. Sci.
59:453s-455s (1979).
Czaja J.A. Ovarian influence on primate food intake: Assessment of
progesterone actions. Physiol. Behav. 21:923-928 (1978).
Czekala N.M., J.K. Hodges, and B.L. Lasley. Pregnancy monitoring in diverse
primate species by estrogen and bioactive LH determinations in small
volumes of urine. T. Med. Primatol. 10:1-15 (1981).
Daehlin L., A. Bergh, and J.E. Damber. Direct effects of oestradiol on growth
and morphology of the Dunning R3327H prostatic carcinoma. Urol.
Res. 15:169-172 (1987).
Daehlin L. and J.E. Damber. Blood flow in the Dunning R3327H rat prostatic
adenocarcinoma: Effect of oestradiol and testosterone. Urol. Res.
14:113-117(1986).
Dalkin A.C., D.J. Haisenleder, G.A. Ortolano, T.R. Ellis, and J.C. Marshall. The
frequency of gonadotropin-releasing-hormone stimulation
differentially regulates gonadotropin subunit messenger ribonucleic
acid expression. Endocrinology 125:917-924 (1989).
Davidson J.M. Feedback control of gonadotropin secretion. In: Frontiers in
Neuroendocrinologv. W.F. Ganong and L. Martini (eds) 343-388, Oxford
University Press, New York (1969).
Davidson J.M. Hormones and reproductive behavior. In: Hormones and
Behavior. S. Levine (ed.) 63-103, Academic Press, New York (1972).
Davson H. The blood brain barrier. T Physiol. Lond. 255:1-29 (1976).
DeLignieres B. and M. Vincens. Differential effects of exogenous oestradiol
and progesterone on mood in postmenopausal women: Individual
dose effect relationships. Maturitas 4:67-72 (1982).

181
Derendorf H. and M. Kaltenbach. Coulometric high-performance liquid
chromatographic analysis of morphine in biological fluids. T. Pharm.
Sci. 75:1198-1200 (1986).
Docke F., P. Ledwon, B. Sturzebecher, W. Rohde, and G. Dorner. Oestrogen
priming for the positive oestrogen feedback: Site of action. Exp. Clin.
Endocrinol. 94:55-60 (1989).
Doisy E.A., C.D. Veler, and S.A. Thayer. Folliculin from the urine of pregnant
women. Am. T. Physiol. 90:329-330 (1929).
Doisy E.A., C.D. Veler, and S.A. Thayer. The preparation of the crystalline
ovarian hormone from the urine of pregnant women. T. Biol. Chem.
86:499-509 (1930).
Doring V.G.K. Klimakterische beschwerden. Fortschr. Med. 94:1058-1067
(1976).
Drill V. and C.W. Calhoun. Oral contraceptives and thromboembolic disease.
T. Am. Med. Assoc. 219: 583-596 (1972).
Dusterberg B. and Y. Nishino. Pharmacokinetic and pharmacological features
of oestradiol valerate. Maturitas 4:315-324 (1982).
Eleftheriou B.C. and C.L. Dobson. Effects of neonatal and adult treatments
with gonadal hormones on choline acetylase activity in brain regions
of male mice after fighting. Psychopharmacology 27:45-52 (1972).
Estes K.S., M.E. Brewster, and N. Bodor. Use of a chemical redox system for
brain enhanced delivery of estradiol decreases prostate weight. Ann.
NY Acad. Sci. 507:334-336 (1987a).
Estes K.S., M.E. Brewster, and N. Bodor. A redox system for brain targeted
estrogen delivery causes chronic body weight decrease in rats. Life Sci.
42:1077-1084 (1988).
Estes K.S., M.E. Brewster, J.W. Simpkins, and N. Bodor. A novel redox
system for CNS-directed delivery of estradiol causes sustained LH
suppression in castrate rats. Life Sci. 40:1327-1334 (1987b).
Ettinger B., I.M. Golditch, and G. Friedman, Gynecologic consequences of
long-term, unopposed estrogen replacement therapy. Maturitas
10:271-282 (1988).

182
Fedor-Freyberg P. The influence of oestrogen on the well-being and mental
performance in climacteric and postmenopausal women. Acta Qbstet.
Gynaecol. Scand. 64:5-69 (1977).
Fenstermacher J.D. Current models of blood-brain transfer. Trends Neuro.
Sd. 8:449-452(1985).
Ferin M., D. van Vught, and S. Wardlow. The hypothalamic control of the
menstrual cycle and the role of endogenous opioid peptides. Recent
Prog. Horm. Res. 40:441-486 (1984).
Fillit H., H. Weinreb, I. Cholst, V. Luine, B. McEwen, R. Amador, and J.
Zabriskie. Observations in a preliminary open trial of estradiol therapy
for senile dementia-Alzheimer's type. Psychoneuroendocrinologv
11:337-345 (1986).
Firsch I.R. and J. Frank. Oral contraceptives and blood pressure. TAMA
237:2499-2503 (1977).
Foote J.E. and E.D. Crawford. Total androgen suppression: Are there any
advantages? The use of combined therapy in the treatment of
advanced prostate adenocarcinoma. Sem. Urol. 6:291-302 (1988).
Fotherby K. Oral contraceptives, lipids and cardiovascular disease.
Contraception 31:367-394(1985).
Fraenkel L. Die funktion des corpus luteam. Arch. Gvnaekol. 68:483-545
(1903).
Furuhjelm M. and P. Fedor-Freyberg. The influence of estrogens on the
psyche in climacteric and postmenopausal women. In: Consensus on
Menopausal Research. P.A. van Keep, R.B. Greenblatt, M. Albeaux-
Femet (eds) 84-93, MTP Press, London (1976).
Gambone J., D.R. Meldrum, L. Laufer, R.J. Chang, J.K.H. Lu, and H.L. Judd.
Further delineation of hypothalamic dysfunction responsible for
menopausal hot flashes. T. Clin. Endocrinol. Metab. 59:1097-1102
(1984).
Garda-Segura L.M., G. Olmos, R.J. Robbins, P. Hernandez, J.H. Meyer, and F.
Naftolin. Estradiol induces rapid remodelling of plasma membranes
in developing rat cerebrocortical neurons in cuture. Brain Res.
498:339-343 (1989).

183
Garca-Segura L.M., G. Olmos, P. Tranque, and F. Naftolin. Rapid effects of
gonadal steroids upon hypothalamic neuronal membrane
ultrastructure. T. Steroid Biochem. 27:615-623 (1987).
Gehring U. and G.M. Tomkins. New mechanism for steroid
unresponsivenessloss of nuclear binding activity of a steroid-
hormone receptor. Cell 3:301-306 (1974).
Gerdis L.C., E.W. Sonnendecker, and E.S. Polakow. Psychological changes
effected by estrogen-progestogen and clonidine treatment in climacteric
women. Am. T. Obstet. Gynecol. 142:98-103 (1982).
Gibson R.D. and J.F. Tingstad. Formulation of a morphine implantation
pellet suitable for tolerance-physical dependence studies in mice. J.
Pharm. Sri. 59:426-427 (1970).
Glascock R.F. and W.G. Hoekstra Selective acccumulation of tritium-labeled
hexoestrol by the reproductive organs of immature female goats and
sheep. Biochem. T. 72:673-682 (1959).
Goodman R.L. and E. Knobil. The site of action of ovarian steroids in the
regulation of LH secretion. Neuroendocrinologv 32:57-63 (1981).
Gorrod J. Potential hazards of the prodrug approach. Chem. Ind. 11:458-462
(1980).
Gorski J. Prolactin biosynthesis and regulation by estrogens. In: Prolactin.
R.B. Jaffe (ed.) 57-83, Elsevier, New York (1981).
Gorski J., D. Toft, G. Shyamala, D. Smith, and A. Notides. Hormone
receptors: Studies on the interaction of estrogen with the uterus.
Recent Prog. Horm. Res. 24:45-80 (1968).
Goy R.W. and B.S. McEwen. Sexual Differentiation of the Brain. MIT Press,
Cambridge, MA (1980).
Greene G.L., L.E. Closs, H. Fleming, E.R. DeSomber, and E.V. Jensen.
Antibodies to estrogen receptor: Immunochemical similarity of
estrophilin from various mammalian species. Proc. Natl. Acad. SO
USA 74:3681-3685(1977).
Greig N.H., S. Momma, D.J. Sweeney, Q.R. Smith, and S.I. Rapoport.
Facilitated transport of melphalan at the rat blood-brain barrier by large
neutral amino acid carrier system. Cancer Res. 47:1571-1576 (1987).

184
Hackman B.W. and D. Galbraith. Replacement therapy with piperazine
oestrone sulfate ("Harmogen") and its effect on memory. Current Med.
Res. Opin. 4:303-306 (1976).
Halasz B. and R.A. Gorski. Gonadotrophic secretion in female rats after
partial or total interuption of neural afferents to the medial basal
hypothalamus. Endocrinology 80:608-622 (1967).
Hamilton T.H. Control by estrogen of genetic transcription and translation.
Science 161:649-661 (1968).
Hammond D.L. Intrathecal administration: Methodological considerations.
Prog. Brain Res. 77:313-320 (1988).
Hardebo J.E. and C. Owman. Barrier mechanisms for neurotransmitter
monoamines and their precursors at the blood-brain interface. Ann.
Neurol. 8:1-11 (1980).
Hawkinson L.F. The menopausal syndrome. One thousand consecutive
patients treated with estrogen. TAMA 111:390-393 (1938).
Henderson S.R., C. Baker, and G. Fink. Oestradiol-17|3 and pituitary
responsiveness to luteinizing hormone releasing factor in the rat: A
study using rectangular pulses of oestradiol-17P monitored by non
chromatographic radioimmunoassay. T. Endocrinol. 73:441-453 (1977).
Henriksson P. and O. Edhag. Orchidectomy versus oestrogen for prostatic
cancer: Cardiovascular effects. Br. Med T. 293:413-415 (1986).
Hoek J.B., and J. Rydstrom. Physiological roles of nicotinamide nucleotide
transhydrogenase. Biochem. T. 254:1-10 (1988).
Hoffmann B. Use of radioimmunoassay procedures for the determination of
sex hormones in animal tissues. T. Steroid Biochem. 19:947-951 (1983).
Holzbauer M. and M.B.H. Youdim. The oestrous cycle and monoamine
oxidase activity. Br. T. Pharmacol. 48:600-605 (1973).
Honjo H., Y. Ogino, K. Naitoh, M. Urabe, J. Kitawaki, J. Yasuda, T. Yamamoto,
S. Ishihara, H. Okada, T. Yonezawa, K. Hayashi, and T. Nambara. In
vivo effects by estrone sulfate on the central nervous systemsenile
dementia (Alzheimers type). T. Steroid Biochem. 34:521-525 (1989).
Huggins C. and C.V. Hodges. Studies on prostatic cancer. I. The effect of
castration, of estrogen and of androgen injection on serum

185
phosphatases in metastatic carcinoma of the prostate. Cancer Res.
1:293-297 (1941).
Huppert L.C. Hormonal replacement therapy: Benefits, risks, doses. Med.
Clin. North Am. 71:23-39 (1987).
Hurst B.S. and J.A. Rock. Endometriosis: Pathophysiology, diagnosis and
treatment. Obstet. Gynecol. Surv. 44:297-304 (1989).
Imperato-McGinley J. 5 a-Reductase deficien in man. In: Progress in Cancer
Research and Therapy. Vol. 31, F. Bresciani, R.J.B. King, M.E. Lippman,
and M. Namer (eds) 491-496, Raven Press, New York (1984).
Imperato-McGinley J., R.E. Peterson, M. Leshin, J.E. Griffin, G. Cooper, S.
Draghi, M. Berenyi, and J.D. Wilson. Steriod 5 a-reductase deficiency
in a 65 year old male pseudohermaphrodite: The natural history,
ultrastructure of the testis and evidence for inherited enzyme
heterogeneity. T. Clin. Endocrinol. Metab. 50:15-22 (1980).
Inman W.H.W. and M.P. Vessey. Investigation of deaths from pulmonary,
coronary, and cerebral thrombosis and embolism in women of child
bearing age. Br. T. Med. 2:193-199 (1968).
Inman W.H.W., M.P. Vessey, B. Westerholme, and A. Engelund.
Thromboembolic disease and the steroidal content of oral
contraceptives: A report to the committee on safety of drugs. Br. Med.
L 2:203-209 (1970).
Iramain C.A., J.O. Owasoyo, and G.N. Egbunike. Influence of estradiol on
acetylcholinestrase activity in several female rat brain areas and
adenohypophysis. Neurosci. Lett. 16:81-84 (1980).
Isaacs J.T., C.B. Brendler, and P.C. Walsh. Changes in the metabolism of
dihydrotestosterone in the hyperplastic human prostate. T. Clin.
Endocrinol. Metab. 56:139-146 (1983).
Jakesz R., A. Kasid, and M.E. Lippman. Continuous estrogen exposure in the
rat does not induce loss of uterine estrogen receptor. J. Biol. Chem.
258:11798-11806 (1983).
Janocko L., J.M. Larner, and R.B. Hochberg. The interaction of C-17 esters of
estradiol with estrogen receptor. Endocrinology 114:1180-1186 (1984).
Jensen E.V. and E.R. DeSombre. Mechanism of action of the female sex
steroids. Ann. Rev. Biochem. 41:203-230 (1972).

186
Jensen E.V. and E.R. DeSombre. Estrogen receptor interaction. Science
182:126-134 (1973).
Jensen E.V. and H.I. Jacobsen. Basic guides to the mechanism of estrogen
action. Recent Prog. Horm. Res. 18:387-414 (1962).
Jonsson G., A.M. Olsson, W. Luttrop, Z. Cekan, K. Purvis, and E. Diczfalusy.
Treatment of prostatic carcinoma with various types of estrogen
derivatives. Vitam. Horm. 33:351-376. (1975).
Judd H.L. Pathophysiology of menopausal hot flushes. In:
Neuroendocrinolgy of Aging, J. Meites (ed) 173-202, Plenum Press, New
York (1983).
Kalla N.R., B.C. Nisula, R. Menard, and D. Lynn Loriaux. The effect of
estradiol on testicular testosterone biosynthesis. Endocrinology 106:35-
39 (1980).
Kalra P.S. and S.P. Kalra. Temporal changes in the hypothalamic and serum
luteinizing hormone-releasing hormone (LHRH) levels and
circulating ovarian steroids during the rat estrous cycle. Acta
Endocrinol. 85:449-455 (1977).
Kalra S.P. and P.S. Kalra. Modulation of hypothalamic luteinizing hormone
releasing hormone levels by intracranial and subcutaneous implants of
gonadal steroids in castrated rats: Effects of androgen and estrogen
antagonists. Endocrinology 106:390-397 (1980).
Kalra S.P. and P.S. Kalra. Neural regulation of luteinizing hormone secretion
in the rat. Endocr. Rev. 4:311-351 (1983).
Kalra S.P. and P.S. Kalra. Do testosterone and estradiol-17p enforce inhibition
or stimulation of luteinizing hormone-releasing hormone secretion?
Biol. Reprod. 41:559-570 (1989).
Kaplan N.M. Cardiovascular complications of oral contraceptives. Ann. Rev.
Med. 29:31-40 (1978).
Katovich M.J. and J. O'Meara. Effect of chronic estrogen on the skin
temperature response to naloxone in morphine-dependent rats. Can. T.
Physiol. Pharmacol. 65:563-567 (1986).
Kaufman H., C. Vadasz, and A. Lajtha. Effects of estradiol and
dexamethasone on choline acetyltransferase activity in various rat
brain regions. Brain Res. 453:389-392 (1988).

187
Kelly M.J., R.L. Moss, and C.A. Dudley. The effect of ovariectomy on
responsiveness of preoptic septal neurons to microelectrotophoresed
estrogen. Neuroendocrinology 25:204-211 (1978).
King R.J.B., J. Gordon, and D.R. Inmam. The intracellular localization of
oestrogen in the rat tissues. T. Endocrinol. 332:9-16 (1965).
Klaiber E.L., D.M. Broverman, W. Vogel, and Y. Kabayashi. The use of steroid
hormones in depression. In: Psychotropic Action of Hormones, T.M.
Itil, G. Laudahn, and W. Herrmann (eds), 135-157, Spectrum Publ Inc.,
New York (1976).
Klaiber E.L., D.M. Broverman, W. Vogel, and Y. Kabayashi. Estrogen therapy
for sever persistant depressions in women. Arch. Gen. Psych. 36:550-
554 (1979).
Klein L.H. Prostatic carcinoma. New Eng. T. Med. 300:824-833 (1979).
Klopper A. Endocrine and metabolic diseases: Treatment of infertility and
menopausal symptoms. Br. Med. T. 2:414-416 (1976).
Knauer E. Die ovarian-transplantation. Arch. Gvnaekol. 60:322-376 (1900).
Knobil E. The neuroendocrine control of the menstrual cycle. Recent Prog.
Horm. Res. 36:53-74 (1980).
Koch B., B. Lutz-Bucher, B. Briaud, and C. Mialhe. Glucocorticoid binding to
plasma membrane of the adenohypophysis. T. Endocrinol. 73:399-400
(1977).
Kopera H. Estrogens and psychic functions. Front. Horm. Res. 2:118-133
(1973).
Krey L.C., F. Kamel, B.S. McEwen. Parameters of neuroendocrine
aromatizaron and estrogen receptor occupation in the male rat. Brain
Res. 193:277-283(1980).
Labrie F., A. Dupont, A. Belanger, F.A. Lefebyre, L. Cusan, G. Monfette, J.G.
Leberge, J.P. Emond, J.P. Raymond, J.M. Husson, and A.T.A. Fazekas.
New hormone treatment in cancer of the prostate: Combined
administrations of an LHRH agonist and an antiandrogen. T. Steroid
Biochem. 19:999-1007(1983).
Lauritzen C. The management of the pre-menopausal and the post
menopausal patient. In: Aging and Estrogens, P.A. van Keep and C.
Lauritzen (eds), Karger, Basel (1973).

188
Lauritzen C. Das Klimakterium der Frau. P.I.L., Paris.and Schering, Berlin
(1982).
Lauritzen C. and P.A. van Keep. Potential beneficial effects of estrogen
substitution in the post-menopause. Front. Horm. Res. 5:1-25 (1978).
Lee C. Gross dissection of three lobes of the rat prostate. In: Current Concepts
and Approaches to the Study of Prostate Cancer, D.S. Coffey, N.
Bruchovsky, W.A. Gardner, Jr., M.I. Resnick, and J.P. Karr (eds) 577-582,
Alan R. Liss, New York (1987).
Legan S.J., G.A. Coon, and F.J. Karsch. Role of estrogen as initiator of daily LH
surges in the ovariectomized rat. Endocrinology 96:50-56 (1975).
Leonard B.E. and D. Kuschinsky. Current status of the biogenic amine theory
of depression. Neurochem. Inti. 4:339-350 (1982).
Levin E. Are the terms blood-brain barrier and brain capillary permeability
synonymous. Exp. Eve Res. [Suppl.] 25:191-199 (1977).
Levin V.A. Relationship of octanol/water partition coefficients and
molecular weight to rat brain capillary permeabiltity. T. Med. Chem.
23:682-684 (1980).
Levine J.E. and V.D. Ramirez. In vivo release of luteinizing hormone
releasing hormone estimated with push-pull cannula from the medial
basal hypothalami of ovariectomized, steroid-primed rats.
Endrocrinologv 107:1782-89 (1982).
Lisk R.D. and D.P. Greenward. Central plus peripheral stimulation by
androgen is necessary for complete restoration of copulatory behavior
in the male hamster. Neuroendocrinology 36:211-217 (1983).
Loewe S. Nachweis brunsterzeugender stoffe im weiblichen blute. Klin.
Wochenschr. 4:1407-1408 (1925).
Loewe S. and F. Lang. Der gehalt des frauenharns an brunsterzeugender
stoffe in abhangigkeit von ovariellen zyklus. Klin. Wochenschr.
5:1038-1039 (1926).
Long J.A. and H.M. Evans. The oestrus cycle in the rat and its associated
phenomena. Mem. Univ. Calif. 6:1-148 (1922).

189
Luine V.N. Estradiol increases choline acetyltransferase activity in specific
basal forebrain nuclei and projection areas of female rats. Exp. Neurol.
89:484-490 (1985).
Luine V.N., R.I. Khylchevskaya, and B. McEwen. Effect of gonadal steroids on
activities of monoamine oxidase and choline acetylase in rat brain.
Brain Res. 86:293-306 (1975).
Luine V.N. and B.S. McEwen. Sex differences in cholinergic enzymes of
diagonal band nuclei in the rat preoptic area. Neuroendocrinologv
36:475-482 (1983).
Luine V.N., D. Park, T. Joh, D. Reis, and B. McEwen. Immuno-chemical
demonstration of increased choline acetyltransferase concentration in
rat preoptic area after estradiol administration. Brain Res. 191:273-277
(1980).
MacLusky N.J., A. Philip, C. Hurlburt, and F. Naftolin. Estrogen metabolism
in neuroendocrine structures. In: Metabolism of Hormonal Steroids
in the Neuroendocrine Structures, F. Celotti, F. Naftolin, and L. Martini
(eds) 103-116, Raven Press, New York (1984).
Maggi A. and J. Perez. Role of female gonadal hormones in the CNS: Clinical
and experimental affects. Life Sci. 37:893-906 (1985).
Majewska M.D. Steroids and brain activity. Essential dialogue between body
and mind. Biochem. Pharmacol. 36:3781-3788 (1987).
Malleson J. An endocrine factor in certain affective disorders. Lancet ii:158-
164 (1953).
Marshall J.C. and R.P. Kelch. Gonadotropin-releasing hormone: Role of
pulsatile secretion in the regulation of reproduction. New Eng L Med.
315:1459-1468 (1986).
Maus M., J. Cordier, J. Glowinski, and J. Premont. 17 p-oestradiol
pretreatment of mouse striatal neurons in culture enhances the
responses of adenylate cyclase sensitive to biogenic amines. Eur. T.
Neurosci. 1:154-161 (1989).
Mayer S.E., R.P. Maickel, and B.B. Brodie. Disappearance of various drugs
from the cerebrospinal fluid. T. Pharmac. Exp. Ther. 128:41-43 (1960).
Mays E.T., W.M. Christopherson, M.M. Mahr, and H.C. Williams. Hepatic
changes in young women ingesting contraceptive steroids. TAMA
235:730-732 (1976).

190
McCann S.M. Physiology and pharmacology of LHRH and somatostatin.
Ann. Rev. Pharmacol. Toxicol. 22:491-515 (1982).
McEwen B.S. Steroid hormones and the brain: Linking "nature" and
"nurture". Neurochem. Res. 13:663-669 (1988).
McEwen B., P. Biegon, P. Davies, L.C. Krey, V.N. Luine, M.Y. McGinnus, C.M.
Paden, B. Parson, and T.C. Rainbow. Steroid hormones: Hormonal
signals which alter brain cell properties and functions. Recent Prog.
Horm. Res. 38:41-92(1982).
McEwen B.S., A. Biegon, C.T. Fischetee, V.N. Luine, B. Parsons, and T.C.
Rainbow. Towards a neurochemical basis of steroid hormone action.
In: Frontiers in Neuroendocrinology, L. Martini and W. Ganong (eds)
1153-1176 (1984).
McEwen B.S. and B. Parsons. Gonadal steroid action on the brain:
neurochemistry and neuropharmacology. Ann. Rev. Pharmacol.
Toxicol. 22:555-598 (1982).
Meites J. and C.S. Nicoll. Hormonal steroids, biochemistry, pharmacology
and therapeutics: proceedings of the first international congress on
hormonal steroids. Vol. 2, Academic Press, New York (1965).
Meldrum D.R., I.M. Shamoni, A.M. Frumar, I.V. Tataryn, R.J. Chang, and
H.L. Judd. Elevation in skin temperature of the finger as an objective
index of postmenopausal hot flushes: Standardization of the
techniques. Am T Obstet. Gynecol. 135:713-717 (1979).
Mendel C.M., R.A. Weisiger, and R.R. Cavalieri. Uptake of 3,5,3'-
triiodothyronine by the perfused rat liver: Return to the free hormone
hypothesis. Endocrinology 123:1817-1824 (1988).
Michael C.H., H.I. Kantor, and H. Shore. Further psychometric evaluation of
older womenthe effect of estrogen administration. T. Gerontol.
25:337-341 (1970).
Michael R.P., R.W. Bonsall, and H.D. Rees. The nuclear accumulation of pH]
testosterone and pH] estradiol in the brain of the female primate:
Evidence for the aromatization hypothesis. Endocrinology 118:1935-
1944 (1986).
Millard W.J., T.M. Romano, N. Bodor, and J.W. Simpkins. GH secretory
dynamics in animals administered estradiol utilizing a chemical
delivery system. Pharm. Res. 7 (11) (in press, 1990).

191
Miller W.L. Molecular biology of steroid hormone synthesis. Endocr. Rev.
9:295-318 (1988).
Molnar G.W. Body temperature during menopausal hot flushes. T. Appl.
Physiol. 38:499-503 (1975).
Moore R.A. Benign prostatic hypertrophy and carcinoma of the prostate.
Occuarnce and experimental production in animals. Surgery 16:152-
167 (1944).
Morrel J.I., D.B. Kelly, and D.W. Pfaff. Sex steroid binding in the brains of
vertebrates. In: Brain-Endocrine Interaction II, K.M. Knigge, D.E. Scott,
H. Kobayashi, and S. Ishii (eds) 230-256, Karger, Basel (1975).
Morton J.H., H. additon, R.G. Addison, L. Hunt, and J.J. Sullivan. A clinical
study of premenstrual tension. Am. T. Obstet. Gynecol. 65:1182-1191
(1953).
Moutsatsou V. and R.E. Oakey. Oestradiol binding to plama proteins. J.
Steroid Biochem. 29:319-323 (1988).
Mowles T.F., B. Ashkanazy, E. Mix, and H. Sheppard. Hypothalamic and
hypophyseal estradiol-binding complexes. Endocrinology 89:484-493
(1971).
Mueller G.C., A.M. Herranen, and K.F. Jervell. Studies on the mechanism of
action of estrogen. Recent Prog. Horm. Res. 14:95-139 (1958).
Mullersman G., H. Derendorf, M.E. Brewster, K.S. Estes, and N. Bodor. High-
performance liquid chromatographic assay of a central nervous system
(CNS)-directed estradiol chemical delivery system and its application
after intravenous administration to rats. Pharm. Res. 5:172-177 (1988).
Murad F. and R.C. Haynes, Jr. Estrogens and progestins. In: The
Pharmacological Basis of Therapeutics, A.G. Gilman, L.S. Goodman,
T.W. Rail, and F. Murad (eds) 1412-1439, Macmillan, New York (1985).
Nabekura J., Y. Oomura, T. Minami, Y. Mizuno, and A. Fukuda. Mechanism
of the rapid effect of 17 P-estradiol on medial amygdala neurons.
Science 223:226-228 (1986).
Nesheim B.I. and T. Saetre. Changes in skin blood flow and body
temperature during climacteric hot flushes. Maturitas 4:49-55 (1982).

192
Neuwelt E.A. Implications of the blood-brain barrier and its manipulation.
Plenum Medical Book Co., New York (1989).
Neuwelt E.A. and P.A. Barrett. Blood-brain barrier disruption in the
treatment of brain tumors, animal studies. In: Implications of the
blood-brain barrier and its manipulation, Vol. 2, E.A. Neuwelt (ed) 107-
193, Plenum Medical Book Co. New York (1989).
Noteboom W.D. and J. Gorski. Stereospecific binding of estrogens in the rat
uterus. Arch. Biochem. Biophys. 111:559-568 (1965).
Notelovitz M. Climacteric medicine: Cornerstone for mid-life health and
wellness. Public Health Rep. [suppl.]:114-124 (1986).
Oldendorf W.H. The blood-brain barrier. Exp. Eye Res. [Suppl.] 25:177-190
(1977).
Oldendorf W.H., L. Braun, S. Hyman, and S.Z. Oldendorf. Blood-brain
barrier: Penetration of morphine, codeine, heroin, and methadone
after carotid injection. Science 178:984-986 (1972).
O'Malley B.W., W.L. McGuire, P.O. Kohler, and S.G. Korenman. Studies on
the mechanism of steroid hormone regulation of synthesis of specific
proteins. Recent Prog. Horm. Res. 25:105-111 (1969).
O'Malley B.W., D.R. Roop, E.C. Lai, J.L. Nordstrom, J.F. Catterall, G.E.
Swaneck, D.A. Colbert, M.J. Tsai, A. Dugaiczyk, and S.L. Woo. The
ovalbumin gene: Organization, structure, transcription, and
regulation. Recent Prog. Horm. Res. 35:1-46 (1979).
O'Malley C.A., R.D. Hautamaki, M. Kelley, and E.M. Meyer. Effects of
ovariectomy and estradiol benzoate on high affinity choline uptake,
ACh synthesis, and release from rat cerebral cortical synaptosomes.
Brain Res. 403:389-392(1987).
Onoda M. and P.F. Hall. Inhibition of testicular microsomal cyctochrome P-
450 (17 a-hydroxylase/C-17,20-lyase) by estrogens. Endocrinology
109:763-767 (1981).
Ottosson U.B. Oral progesterone and estrogen/progesterone therapy. Effects
of natural and synthetic hormones on subfractions of HDL cholestrol
and liver proteins. Acta. Obstet. Gynecol. Scand. [Suppl.] 12:5-37 (1984).
Ottosson U.B., K. Carlstrom, B.G. Johansson, and B. von Schoultz. Estrogen
induction of liver proteins and high-density lipoprotein cholestrol:

193
Comparison between estradiol valerate and ethinyl estradiol. Gynecol.
Obstet. Invest. 22:198-205 (1986).
Palmer K. and J.M. Gray. Central vs peripheral effects of estrogen on food
intake and lipoprotein lipase activity in ovariectomized rats. Physiol.
Behav. 37:187-189 (1986).
Paradisi R., S. Lodi, G. Bolelli, and S. Venturoli. Radioimmunoassay of three
oestrogens and three androgens in the same plasma sample after
extraction and chromatographic separation. Acta Endocrinol. 94:229-
234 (1980).
Pardridge W.M. Transport of nutrients and hormones through the blood-
brain barrier. Diabetologia 20:246-254 (1981).
Pardridge W.M. Neuropeptides and the blood-brain barrier. Ann. Rev.
Physiol. 45:73-82 (1983).
Pardridge W.M. Receptor-mediated peptide transport through the blood-
brain barrier. End. Rev. 7:314-330 (1986).
Pardridge W.M. Plasma protein-mediated transport of steroid and thyroid
hormones. Am. I. Physiol. 252:E157-164 (1987).
Pardridge W.M. Selective delivery of sex steroid hormones to tissues in vivo
by albumin and by sex hormone-binding golbulin. In: Steroid Protein
Interactions : Basic and Clinical Aspects. R. Fraira (ed.) 173-192, New
York: NY, Acad. Sci. (1988a).
Pardridge W.M. Recent advances in blood-brain barrier transport. Ann. Rev.
Pharmacol. Toxicol. 28:25-39 (1988b).
Pardridge W.M., J.D. Connor, I.L. Crawford. Permeability changes in the
blood-brain barrier: causes and consequences. CRC Crit. Rev. Toxic.
3:159-199 (1975).
Pardridge W..M. and L.J. Meitus. Transport of steroid hormones through the
rat blood-brain barrier. T. Clin. Invest. 64:145-154 (1979).
Paterson M.E.L. A randomized double-blind cross-over trial into the effect of
norethisterone on climateric symptoms and biochemical profiles. Br. T.
Obstet. Gynecol. 89:464-472 (1982).
Persson I. The risk of endometrial and breast cancer after estrogen treatment.
Acta Obstet. Gynecol. Scand. [Suppl.] 130:59-66 (1985).

194
Pfaff D. Nature of sex hormone effects on rat sex behavior: Specificity of
effects and individual patterns of response. T. Comp. Physiol. Psychol.
73:349-358 (1970).
Pfaff D. and M. Keiner. Atlas of estradiol-concentrating cells in the central
nervous system of the female rat. T. Comp. Neurol. 151:121-158 (1973).
Pietras R.J. and C.M. Szego. Estrogen receptors in uterine plasma membrane.
T. Steroid Biochem. 11:1471-1488 (1979).
Plant T.M. Gonadal regulation of hypothalamic gonadotropin-releasing
hormone release in primates. End. Rev. 7:75-88 (1986).
Pliner P. and A.S. Fleming. Food intake, body weight and sweetness
preferences over the menstrual cycle in humans. Physiol. Behav.
30:663-666 (1983).
Pohl C.R. and E. Knobil. The role of the central nervous system in the control
of ovarian function in higher primates. Ann. Rev. Physiol. 44:583-
(1982).
Pollard M., P.H. Luckert, and D. Snyder. Prevention and treatment of
experimental prostate cancer in Lobund-Wistar rats. I. Effects of
estradiol, dihydrotestosterone, and castration. The Prostate 15:95-103
(1989).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Distribution of a brain-
enhanced chemical delivery system for estradiol. Pharm. Res. [Suppl.]
5:S-205 (1988).
Rahimy M.H., J.W. Simpkins, and N. Bodor. A rapid, sensitive method for
the simultaneous quantitation of estradiol and estradiol conjugates in a
variety of tissues: Assay development and evaluation of the
distribution of a brain-enhanced estradiol-chemical delivery system. L
Steroid Biochem. 33:179-187 (1989a).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Pharmacodynamics of a novel
estradiol-chemical delivery system for the brain. Pharm. Res. [Suppl.]
6:S-211 (1989b).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Tissue distribution of a brain-
enhanced chemical delivery system for estradiol. Drug Des. Deliv.
6:29-40 (1990a).

195
Rahimy M.H., J.W. Simpkins, and N. Bodor. Dose and time-course
evaluation of a redox-based estradiol-chemical delivery system for the
brain. I. Tissue distribution. Pharm. Res. 7:1061-1067 (1990b).
Rahimy M.H., J.W. Simpkins, and N. Bodor. Dose and time-course
evaluation of a redox-based estradiol-chemical delivery system for the
brain. II. Pharmacodynamic responses. Pharm. Res. 7 (11) (in press,
1990c).
Rao B.R., A.A. Geldof, C.L. van der Wilt, and H.J. de Voogt. Efficacy and
advantages in the use of low doses of androgen and estrogen
combination in the treatment of prostate cancer. The Prostate 13:69-78
(1988).
Rapoport S.I. The Blood-brain barrier in physiology and medicine. Raven
Press, New York (1976).
Rapoport S.I., M. Ohata, E.D. London. Cerebral blood flow and glucose
utilization following opening of the blood-brain barrier and during
maturation of the rat brain. Fed.Proc. Fed. Am. Soc. Expl. Biol. 40:2322-
2325 (1981).
Rash J.M., I. Jerkunica, and D.S. Sgoutas. Lipid interference in steroid
radioimmunoassay. Clin. Chem. 26:84-88 (1980).
Rauramo L., K. Lagerapitz, P. Engblom, and R. Punnonen. The effect of
castration and peripheral estrogen therapy on some psychological
function. Front. Horm. Res. 8:133-151 (1975).
Reese T.S. and M.J. Karnovsky. Fine structural localization of a blood-brain
barrier to exogenous peroxidase. Cell Biol. 34:207-217 (1967).
Richards J.S. and L. Hedin. Molecular aspects of hormone action on ovarian
follicular development, ovulation, and luteinization. Ann. Rev.
Physiol. 50:441-463 (1988).
Roselli C.E. and J.A. Resko. Regulation of hypothalamic luteinizing
hormone-releasing hormone levels by testosterone and estradiol in
male rhesus monkeys. Brain Res. 509:343-346 (1990).
Rosie R., E. Thomson, and G. Fink. Oestrogen positive feedback stimulates
the synthesis of LHRH mRNA in neurons of the rostral diencephalon
of the rat. T. Endocrinol. 124:285-289 (1990).

196
Ross G.T. Disorders of the ovary and female reproductive tract. In: Textbook
of Endocrinology, J.D. Wilson and D.W. Foster (eds), W. B. Saunders
Co., Philadelphia (1981).
Ross G.T. The ovaries. In: Textbook of Medicine. J.N. Wyngaarden, L.H.
Smith (eds) 1379-1393, Saunders, Philadelphia (1985).
Rossmanith W.G., G.A. Laughlin, J.F. Mortola, and S.S.C. Yen. Secretory
dynamics of oestradiol (E2) and progesterone (P) during periods of
relative pituitary LH quiescence in the midluteal phase of the
menstrual cycle. Clin. Endocrinol. 32:13-23 (1990).
Rotkin I.D. Distribution, and risk of benign prostatic hypertrophy. In:
Benign Prostatic Hypertrophy, F. Hinman, Jr. (ed.) 10-21, Springer-
Verlag, New York (1983).
Santen R.J. and H.F. English. Biological principles underlying treatment
strategies for prostate carcinoma. In: Proceedings of IVth International
Congress of Andrology. M. Serio (ed.) 449-457, Raven Press, Florence,
Italy (1989).
Sarkar D.K., S.J. Friedman, S.S.C. Yen, and S.A. Frautschy. Chronic inhibition
of hypothalamic-pituitary-ovarian axis and body weight gain by brain-
directed delivery of estradiol-17(3 in female rats. Neuroendocrinologv
50:204-210 (1989).
Saumande J. and S.K. Batra. A double antibody radioimmunoassay for free
and conjugated estradiol-17(3 in cow's milk. Steroids 44:137-152 (1984).
Schanker L.S. Passage of drugs into and out of the central nervous system.
Antimicrob. Agents Chemother ,1044-1050 (1965).
Schiess M.C., C.A. Dudley, and R.L. Moss. Estrogen priming affects the
sensitivity of midbrain central gray neurons to microiontophoretically
applied LHRH but not beta-endorphin. Neuroendocrinology 46:24-31
(1987).
Schneider M.A., P.L. Brotherton, and J. Hailes. The effect of exogenous
oestrogens on depression in menopausal women. Med. T. Austr. 2:162-
253 (1977).
Schwartz N.B. (ed.) Dynamics of ovarian function. Raven Press, New York
(1981).

197
Segal L., B. Segal, and W.R. Nes. The acid-catalyzed solvolysis of
dehydroepiandrosterone sulfate and its significance in the examination
of urinary 17-ketosteroids. T. Biol. Chem. 235:3108-3111 (1960).
Sherwin B.B. Estrogen and/or androgen replacement therapy and cognitive
functioning in surgically menopausal women.
Psychoneuroendocrinologv 13:345-357 (1988).
Shivers R.R. The blood-brain barrier of a reptile. Anoli's carolinensis. A
freeze-fracture study. Brain Res. 169:221-230 (1979).
Silverberg E. and J. Lubera. Cancer statistics. CA 36:9-25 (1986).
Silverberg S.O. and E.L. Makawski. Endometrial carcinoma in young women
taking oral contraceptive agents. Onstet. Gynecol. 46:503-506 (1965).
Simpkins J.W., W.R. Anderson, R. Dawson, Jr., N. Bodor. Effects of a brain-
enhanced chemical delivery system for estradiol on body weight and
food intake in intact and ovariectomized rats. Pharm. Res. 6:592-600
(1989a).
Simpkins J.W., W.R. Anderson, R. Dawson, Jr., A. Seth, M.E. Brewster, K.S.
Estes, and N. Bodor. Chronic weight loss in lean and obese rats with a
brain-enhanced chemical delivery system for estradiol. Physiol. Behav.
44:573-580 (1988).
Simpkins J.W., N. Bodor, and A. Enz. Direct evidence for brain-specific
release of dopamine from a redox delivery system. T. Pharm. Sci.
74:1033-1036 (1985).
Simpkins J.W., M.J. Katovich, and I.C. Song. Similarities between morphine
withdrawal in the rat and the menopausal hot flush. Life Sci. 32:1957-
1966 (1983).
Simpkins J.W. and M.J. Katovich. An animal model for pharmacologic
evaluation of the menopausal hot flush. In: The climacteric in
perspective. M. Notelovitz and P. van Keep (eds) 213-251, MTP Press,
Lancaster, UK (1984).
Simpkins J.W., J. McCornack, K.S. Estes, M.E. Brewster, E. Shek, and N. Bodor.
Sustained brain-specific delivery of estradiol causes long-term
suppression of luteinizing hormone secretion. T. Med. Chem. 29:1809-
1812 (1986).
Simpkins J.W., M.H. Rahimy, and N. Bodor. A brain-enhanced chemical
delivery system for gonadal steroids: Implications for

198
neurodegenerative diseases. In: Novel Approaches to the Treatment
of Alzheimer Disease. E.M. Meyer, J.W. Simpkins and J. Yamamoto
(eds) 197-212, Plenum Press, New York (1989b).
Sinkula A.A. and S.H. Yalkowsky. Ratioal for design of biologically reversible
drug derivatives: prodrugs. T. Pharm. Sci. 64:181-193 (1975).
Smith E.R. and J.M. Davidson. Location of feedback receptors; effects of
intracranially implanted steroids on plasma LH and PRL response.
Endocrinology 95:1566-1573 (1974).
Smith S.L. and C. Sauder. Food cravings, depression and premenstrual
problems. Psvchosom. Med. 31:281-287 (1969).
Smith S.S., B.D. Waterhouse, and D.J. Woodward. Sex steroids effects on
extrahypothalamic CNS. I. Estrogen augments neuronal
responsiveness to iontophoretically applied glutamate in the
cerebellum. Brain Res. 422:40-51 (1987).
Spelsberg T.C., B.J. Gosse, B.A. Littlefield, H. Toyoda, and R. Seelke.
Reconstitution of native-like nuclear acceptor sites of the avian
oviduct progesterone receptor: Evidence for involvement of specific
chromatin proteins and specific DNA sequences. Biochemistry
23:5103-5113 (1984).
Spona J. and W. Schneider. Bioavailability of natural estrogens in young
females with secondary amenorrhea. Acta Obstet. Gynecol. Scand.
[Suppl.] 65:33-38 (1977).
Steffens A.B. A method for frequent sampling of blood and continuous
infusion of fluids in the rat without disturbing the animal. Physiol.
Behav. 4:833-836 (1969).
Stella V. Pro-drugs: An overview and definition. In: Prodrugs as Novel
Drug Delivery Systems. T. Higuchi, V. Stella (eds) 1-115,American
Chemical Society, Washington, DC (1975).
Suckling A.J., M.G. Rumsby, M.W.B. Bradbury. The Blood-Brain Barrier in
Health and Disease. VCH Publishers, Chichester, England (1986).
Sulkava R., J. Wikstrom, A. Aromaa, R. Raitasalo, V. Lehtinen, K. Lahtela,
and J. Palo. Prevalence of sever dementia in Finland. Neurology
35:1025-1029 (1985).

199
Suyemitus T. and H. Terrayama. Specific binding sites for natural
glucocorticoides in plasma membranes of rat liver. Endocrinology
96:1499-1508 (1975).
Swerdloff R.S. Physiology of male reproduction: Hypothalamic-pituitary
function. In: Campbell's Urology, P.C. Walsh, R.F. Gittes, and A.D.
Perlmutter (eds) 186-200, Saunders, Philadelphia (1986).
Szego C.M. and S. Roberts. Steroid action and interaction in uterine
metabolism. Recent Prog. Horm. Res. 8:419-470 (1953).
Tallarida R.J. and R.B. Murry. Manual of Pharmacological Calculation with
Computer Programs. Springer, New York (1981).
Tarttelin M.F. and R.A. Gorski. Variations in food and water intake in
normal and acyclic female rats. Physiol. Behav. 7:847-852 (1971).
Taylor R.N. and R.G. Smith. Effects of highly purified estrogen receptors on
gene transcription in isolated nuclei. Biochemistry 21:1781-1787 (1982).
Tepperman J. Metabolic and Endocrine Physiology. Year Book Medical
Publishers, Inc. Chicago (1981).
Thomas D.B. Steroid hormones and medications that alter cancer risks.
Cancer 62:1755-1767 (1988).
Toft D. and J. Gorski. A receptor molecule for estrogens: Isolation from the
rat uterus and preliminary characterization. Proc. Natl. Acad. Sci. USA
55:1574-1581 (1966).
Towle A.C. and P.Y. Sze. Steroid binding to synaptic plasma membrane:
Differential binding of glucocorticoids and gonadal steroids. T. Steroid
Bioch. 18:135-143 (1983).
Trapido E.J., L.A. Brinton, C. Schairer, and R. Hoover. Estrogen replacement
therapy and benign breast disease. T. Natl. Cancer Inst. 73:1101-1105
(1984).
Traystman R.J. Microcirculation of the brain. In: The Physiology and
Pharmacology of the Microcirculation. Vol. 1, N.A. Mortillaro (ed.) 237-
298, Academic Press, New York (1983).
Upton V. Therapeutic considerations in the management of the climactric. L
Reprod. Med. 29:71-79 (1984).

200
Van Deurs B. Structural aspects of brain barriers, with special reference to the
permeability of the cerebral endothelium and choroidal epithelium.
Int. Rev. Cvtol. 65:117-191 (1980).
van Steenbrugge G.J., M. Groen, A. van Kreuningen, F.H. de Jong, M.P.W.
Gallee, and F.H. Schroder. Transplantable human prostatic carcinoma
(PC-82) in athymic nude mice. III. Effects of estrogens on the growth of
the tumor tissue. The Prostate 12:157-171 (1988).
Vanhulle G. and R. Demol. A double-blind study into the influence of estriol
on a number of psychological tests in post-menopausal women. In:
Consensus on Menopausal Research, P.A. van Keep, R.B. Greenblatt,
M. Albeaux-Fernet (eds) 94-99, MTP Press, London (1976).
von Schoultz B., K. Carlstrom, L. Collste, A. Eriksson, P. Henriksson, A.
Pousette, and R. Stege. Estrogen therapy and liver functionmetabolic
effects of oral and parenteral administration. The Prostate 14:389-395
(1989).
Walters M.R. Steroid hormone receptors and the nucleus. End. Rev. 6:512-
543 (1985).
Wei E., Loh H.H. and E.L. Way. Quantitative aspects of precipitated
abstinence in morephine-dependent rats. T. Pharmacol. Exp. Ther.
184:398-403 (1973).
Weiner N., F. Martin, and M. Riaz. Liposomes as a drug delivery system.
Drug Dev. Ind. Pharmacy 15:1523-1554 (1989).
Weiss N.S. Epidemiology of carcinoma of the endometrium. In: Reviews in
Cancer Epidemiology. Vol. 2, A.M. Lilienfeld (ed) 46-60, Elsevier
Science Publishing Company, New York (1983).
Weisz J. and W.R. Crowley. Catechol estrogen formation by the CNS:
Regional distribution of estrogen-2/4-hydroxylase activity in the rat
brain. Neuroendocrinology 43:543-549 (1986).
White J.W. The results of double castration in hypertrophy of the prostate.
Ann. Surg. 22:1-80 (1895).
Wiegand S.J., E. Terasawa, W.E. Bridson, R.W. Goy. Effects of discrete lesions
of preoptic and suprachiasmatic structures in the female rat.
Alterations in the feedback regulation of gonadotrophin secretion.
Neuroendocrinologv 31:147-157 (1980).

201
Yalom I.D., D.T. Lunde, and R.H. Moos. Postpartum blues syndrome; A
description and related variables. Arch. Gen. Psychiatry 18:16-17 (1968).
Yen S.S.C. The biology of menopause. T. Reprod. Med. 18:287-289 (1977).
Yen S.S.C. The human menstrual cycle (integrative function of the
hypothalamic-pituitary-ovarian-endometrial axis). In: Reproductive
Endocrinology. S.S.C. Yen and R.B. Jaffe (eds), W. B. Saunders Co.,
Philadelphia (1978).
Zar J.H. Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ (1974).

BIOGRAPHICAL SKETCH
Mohamad Hossein Rahimy was born on September 5, 1954, in Esfahan,
Iran. It seems pretty old to be a graduate student at that age. But certainly it is
never too late to satisfy the ego, specially with regard to knowledge. After he
graduated from high school in 1974, he was compelled to serve a two-year
term in the late Shah's military. Shortly after the compulsory service, he
made a trip to the Florida Beach, St. Petersburg, that left a lasting impression.
He entered St. Petersburg College in 1979. Two years later, he transfered to the
University of Florida in Gainesville. He earned his B.S. in microbiology and
cell science in December 1982. After graduation, he returned to his homeland
to start his career plans, not knowing that things were changed allot. In a year
or so, he decided to pursue an academic career. Thus, after his return to
Gainesville in 1985, he entered the graduate program of the College of
Pharmacy. In June 1985, he joined the laboratory of Dr. James W. Simpkins at
the Department of Pharmacodynamics, and there he remained until the
present time. Should he ever graduate, he plans to continue his research
career in either pharmaceutical industry or academia.
202

I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
'2si/yu,
es W. Simpkins, /Chairman
ofessor of Pharmacodynamics
I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
I-\X>
Nicholas Bodor
Graduate Research Professor of
Medicinal Chemistry
I certify that I have read this study and that in my opinion it conforr
to acceptable standards of scholarly presentation/nd is fully^Sdequate, in,
scope and quality, as a dissertation for the degree of Dpctpi/of Phos
lliam J.
Associate Professor of
Pharmacodynamics

I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
Edwin Meyer
Associate Professor of
Pharmacology and Therapeutics
I certify that I have read this study and that in my opinion it conforms
to acceptable standards of scholarly presentation and is fully adequate, in
scope and quality, as a dissertation for the degree of Doctor of Philosophy.
Kalpn Dawson ^
Associate Professor of
Pharmacodynamics
This dissertation was submitted to the Graduate Faculty of the College
of Pharmacy and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
August, 1990
Dean, College of Pharmacy
Dean, Graduate School




19
nuclear estrogen-receptor complexs were estimated to be 2 hrs for 17 (3-E2
(Walters, 1985).
Finally, the maximum biological responses seemed to be determined
not by the level of hormone binding to cytosol receptors, but by retention of a
small proportion (10 to 15%) of the receptors on a limited number of saturable
estrogen receptor-binding sites (acceptor sites) in the nuclear DNA (Walters,
1985). This is supported by the observation of a good correlation between the
duration of nuclear occupancy and uterine growth stimulation in a series of
short and long acting estrogen and their derivatives.
Role of Estrogen in the Menstrual Cycle
During the normal reproductive life (30 to 40 years), a female
menstruates 300 to 500 times (Tepperman, 1981). The menstrual cycle is
characterized by monthly rhythmic changes in the secretion pattern of the
reproductive hormones and the corresponding changes in the sexual organs
as well. The duration of the cycle averages 28 days. The significance of the
menstrual cycle are (1) maturation of an ovum and ovulation; (2) preparation
of the uterine endometrium for implantation of an embryo; and (3)
expression of the secondary sex characteristics associated with the procreative
act. The coordination among these events is achieved by precisely timed
fluctuation in the production and secretion rates of a number of hormones
associated with the hypothalamic-pituitary-ovarian axis. In humans and
primates the ovarian E2 is perhaps the major driving force for the
initiation/maintenance of the cycle. This implies that the primary stimulus
in triggering the initiation of the preovulatory gonadotropin surge, the
central event in the cycle, is caused by the rise in E2 levels during the late


183
Garca-Segura L.M., G. Olmos, P. Tranque, and F. Naftolin. Rapid effects of
gonadal steroids upon hypothalamic neuronal membrane
ultrastructure. T. Steroid Biochem. 27:615-623 (1987).
Gehring U. and G.M. Tomkins. New mechanism for steroid
unresponsivenessloss of nuclear binding activity of a steroid-
hormone receptor. Cell 3:301-306 (1974).
Gerdis L.C., E.W. Sonnendecker, and E.S. Polakow. Psychological changes
effected by estrogen-progestogen and clonidine treatment in climacteric
women. Am. T. Obstet. Gynecol. 142:98-103 (1982).
Gibson R.D. and J.F. Tingstad. Formulation of a morphine implantation
pellet suitable for tolerance-physical dependence studies in mice. J.
Pharm. Sri. 59:426-427 (1970).
Glascock R.F. and W.G. Hoekstra Selective acccumulation of tritium-labeled
hexoestrol by the reproductive organs of immature female goats and
sheep. Biochem. T. 72:673-682 (1959).
Goodman R.L. and E. Knobil. The site of action of ovarian steroids in the
regulation of LH secretion. Neuroendocrinologv 32:57-63 (1981).
Gorrod J. Potential hazards of the prodrug approach. Chem. Ind. 11:458-462
(1980).
Gorski J. Prolactin biosynthesis and regulation by estrogens. In: Prolactin.
R.B. Jaffe (ed.) 57-83, Elsevier, New York (1981).
Gorski J., D. Toft, G. Shyamala, D. Smith, and A. Notides. Hormone
receptors: Studies on the interaction of estrogen with the uterus.
Recent Prog. Horm. Res. 24:45-80 (1968).
Goy R.W. and B.S. McEwen. Sexual Differentiation of the Brain. MIT Press,
Cambridge, MA (1980).
Greene G.L., L.E. Closs, H. Fleming, E.R. DeSomber, and E.V. Jensen.
Antibodies to estrogen receptor: Immunochemical similarity of
estrophilin from various mammalian species. Proc. Natl. Acad. SO
USA 74:3681-3685(1977).
Greig N.H., S. Momma, D.J. Sweeney, Q.R. Smith, and S.I. Rapoport.
Facilitated transport of melphalan at the rat blood-brain barrier by large
neutral amino acid carrier system. Cancer Res. 47:1571-1576 (1987).


168
1983). These capillaries in the median eminence lack features of other brain
capillaries and hence are not part of the BBB (Traystman, 1983).
Likewise, plasma showed, following administration of E2-CDS to male
(Chapter 5) or OVX female rats (Chapter 6), a residual but detectable levels of
E2 for approximately two weeks after a single iv dose of E2-CDS treatment.
Since E2 is active at low pg concentrations, these plasma levels of E2 may
contribute to some of the long-term effects of E2-CDS, i.e. uterine tissue
stimulation or pituitary responses (Chapter 7). The source of this sustained
residual E2 in plasma is likely to be the result of a continuous redistribution
of E2 liberated from E2-Q+ in the brain down its concentration gradient into
the general circulation.
In contrast to CNS tissue, the distribution of E2-CDS metabolites in
male (Chapter 5) and OVX female rats (Chapter 6) showed sex differences with
respect to the elimination patterns of E2-Q+ and E2 from certain peripheral
tissues. The more obvious one was the heart tissue. Levels of E2-Q+ in heart
tissue of the male rat, following administration of E2-CDS, were
approximately 2-fold greater than levels of this metabolite in female
counterparts (Chapters 5 & 6). Furthermore, the rate of elimination of this
compound from heart tissue of male rat was slower than that of heart tissue
of OVX female rats. Lung tissue of male rat also showed higher levels of both
E2-Q+ and E2 than those of female rats following administration of E2-CDS.
The observed sex differences with respect to E2-CDS metabolism and
elimination patterns of these metabolites from these tissues may be related to
differences in kinetics of enzymes involved. For instance, if there are tissue
specific rates of E2 metabolism that can alter the equilibrium between E2-Q+ <=>
E2 <=> E2 metabolites, that may lead to the observed differences in the
disappearance of E2-Q+ from these tissues. Furthermore, if a certain tissue, i.e.


33
smokers, liver tumors, hypertension, and changes in glucose metabolism that
appear to be estrogen related (Drill & Calhoun, 1972; Firsch & Frank, 1977;
Fotherby, 1985; Inman et al., 1970; Kaplan, 1978; Mays et al., 1976; Thomas,
1988). To reduce the magnitude and the spectrum of these dose-related
adverse effects of estrogen thus would be to reduce the dose (Bottiger et al.,
1980) and/or to use a more natural hormone (Ottosson, 1984). Given the
combined decrease in the contraceptive components, there is the possibility
that suppression of the hypothalamic-pituitary-ovarian function may not be
as effective as with higher dose formulations. Therefore, preferential brain
delivery of E2 with the CDS may provide an effective, long-acting
contraceptive by virtue of sustained release of E2 in that tissue. Furthermore,
the adverse peripheral effects associated with the currently used contraceptive
steroids may be avoided by lowering the dose or the frequency of ingestion.
Menopausal Syndrome
The cessation of menses, menopause, near the age of 50 is the result of
the decreasing production of ovarian estrogens/progestins (Notelovitz, 1986).
This loss of ovarian hormones in 75% to 85% of women leads to a number of
brain-mediated steroid-withdrawal symptoms (Casper & Yen, 1985; Lauritzen,
1973; Yen, 1977), the most frequent being hot flushes (Clayden et al., 1974;
Meldrum et al., 1979). These (patho)physiological alterations appear to be the
result of autonomic discharge which causes peripheral vasodilation and heat
loss (Nesheim & Saetre, 1982). Replacement therapy with estrogens and/or
progestins has been shown to be effective in most menopausal patients in
alleviating the symptoms of the disease (Campbell & Whitehead, 1977;
Huppert, 1987; Upton, 1984). However, numerous retrospective studies


LIST OF FIGURES
Figure Page
1. Schematic representation of in vitro synthesis and in vivo
transformation of the estradiol-chemical delivery system (E2-
CDS) 67
2. Inhibition of 125I-E2 binding to an E2 antibody caused by E2 or E2-
Q+ 68
3. Recovery of known concentrations of E2-Q+ and E2 added to
brain tissue homogenates prior to extraction 69
4. Inhibition of 1251-E2 binding to an E2 antibody caused by
increasing amounts of brain tissue from rats treated with the
estradiol-chemical delivery system 70
5. Effects of a single iv dose of the estradiol-chemical delivery
system on serum and brain concentrations of E2-Q+ and E2 or the
HPCD vehicle 71
6. Effects of a single iv dose of the E2-CDS on brain and plasma
concentrations of E2-Q+ and E2 in intact male rats 84
7. Effects of a single iv dose of the E2-CDS on liver and fat
concentrations of E2-Q+ and E2 in intact male rats 86
8. Effects of a single iv dose of the E2-CDS on kidney, heart, lung,
and anterior pituitary concentrations of E2-Q+ and E2 in intact
male rats 87
9. Brain and anterior pituitary contents of the E2-Q+ and E2
following a single iv dose of the E2-CDS 88
10. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in whole brain of ovariectomized rats 100
11. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in hypothalamus of ovariectomized rats 102
ix


188
Lauritzen C. Das Klimakterium der Frau. P.I.L., Paris.and Schering, Berlin
(1982).
Lauritzen C. and P.A. van Keep. Potential beneficial effects of estrogen
substitution in the post-menopause. Front. Horm. Res. 5:1-25 (1978).
Lee C. Gross dissection of three lobes of the rat prostate. In: Current Concepts
and Approaches to the Study of Prostate Cancer, D.S. Coffey, N.
Bruchovsky, W.A. Gardner, Jr., M.I. Resnick, and J.P. Karr (eds) 577-582,
Alan R. Liss, New York (1987).
Legan S.J., G.A. Coon, and F.J. Karsch. Role of estrogen as initiator of daily LH
surges in the ovariectomized rat. Endocrinology 96:50-56 (1975).
Leonard B.E. and D. Kuschinsky. Current status of the biogenic amine theory
of depression. Neurochem. Inti. 4:339-350 (1982).
Levin E. Are the terms blood-brain barrier and brain capillary permeability
synonymous. Exp. Eve Res. [Suppl.] 25:191-199 (1977).
Levin V.A. Relationship of octanol/water partition coefficients and
molecular weight to rat brain capillary permeabiltity. T. Med. Chem.
23:682-684 (1980).
Levine J.E. and V.D. Ramirez. In vivo release of luteinizing hormone
releasing hormone estimated with push-pull cannula from the medial
basal hypothalami of ovariectomized, steroid-primed rats.
Endrocrinologv 107:1782-89 (1982).
Lisk R.D. and D.P. Greenward. Central plus peripheral stimulation by
androgen is necessary for complete restoration of copulatory behavior
in the male hamster. Neuroendocrinology 36:211-217 (1983).
Loewe S. Nachweis brunsterzeugender stoffe im weiblichen blute. Klin.
Wochenschr. 4:1407-1408 (1925).
Loewe S. and F. Lang. Der gehalt des frauenharns an brunsterzeugender
stoffe in abhangigkeit von ovariellen zyklus. Klin. Wochenschr.
5:1038-1039 (1926).
Long J.A. and H.M. Evans. The oestrus cycle in the rat and its associated
phenomena. Mem. Univ. Calif. 6:1-148 (1922).


52
Application of the E2-CDS to biological systems creates a problem
concerning the separation and quantitation of tissue levels of the active
species, E2, in the presence of an excess concentration of the inactive
conjugated form of the drug, E2-Q+. Furthermore, pharmacokinetic and
pharmacodynamic studies of the E2-CDS require a reliable and sensitive
method for the quantitative analysis of E2 as well as E2-Q+ in biological tissues
and fluids. Conventional methods of assaying steroids, especially conjugated
forms of E2, are inefficient and extremely time-consuming. (Bonney et al.,
1984; Cortes-Gollegos & Gallegos, 1975; Hoffmann, 1983; Paradisi et al., 1980).
RIA is the only method which has been used on a large scale for the analysis
of steroid hormones in plasma or serum. Previously described methods for
measuring steroidal hormones and their conjugates in biological tissues have
been severely hindered by lengthy extraction and repeated purification
procedures which are required prior to the use of RIA procedure.
Therefore, the objectives of these experiments were to develop a
sensitive, specific method that permits a rapid and reliable quantitation of
both E2 and E2-Q+ in various biological tissues and fluids. This method
development was necessitated by the design of the E2-CDS which exhibits the
predictive, multiple, facile enzymatic conversion to the charged quaternary
ion (E2-Q+) and its subsequent hydrolysis to slowly liberate E2.
Materials and Methods
Initially, several in vitro experiments were conducted to optimize the
selective extraction, purification and quantitation of both E2-Q+ and E2 from a
wide variety of biological tissues and fluids. Subsequently, an in vivo study
was undertaken to assess the reliability and applicability of the in vitro


97
and, to a lesser extent, in brain, lung, and liver; and (iii) the disappearance of
both E2-Q+ and E2 was slow in brain tissue and rapid in all peripheral tissues
tested. Collectively, these data are consistent with the expected behavior of
the E2-CDS (Bodor et al., 1987).
Dose and time-course profiles revealed that E2-Q+ persists in brain
tissue as well as in hypothalamus, with a ti/2 = 8-9 days, but it is rapidly
cleared from the periphery. The previous single-dose distribution study in
intact male rats (Chapter 5) demonstrated a similar half-life for E2-Q+ in brain
tissue. These estimates of the half-life of E2-Q+ in the brain are in accordance
with reports which utilized different analytical techniques and E2-CDS doses
(Mullersman et al., 1988). The slow clearance of the E2-CDS metabolite, E2-Q+,
from the CNS tissue appears to be independent of dose since similar values
have been obtained in studies using doses of E2-CDS ranging from 0.01 mg/kg
(present report) to 15 mg/kg dose (Mullersman et al., 1988). Further, the long
half-life of E2-Q+ in brain tissue does not appear to be an artifact of its
sustained production from E2-CDS since the half-life of the delivery system
itself in brain tissue is only 29.2 min, indicating rapid oxidation to E2-Q+
(Bodor et al., 1987). Thus, as predicted based on the physicochemical
properties of the E2-CDS (Bodor et al., 1987), the unique features of the BBB
appear to contribute to the chronic retention by the brain of the charged,
hydrophilic E2-Q+.
The anterior pituitary exhibited slower elimination of the metabolites
of E2-CDS (E2-Q+ & E2) than other peripheral tissues. By 7 days following
administration of a 1.0-mg E2-CDS dose, the hypothalamic-anterior pituitary
E2-Q+ ratio was 1.3 and then increased to about 4-fold by 28 days. Similarly,
the hypothalamic-anterior pituitary E2 ratio was 1.2 on day 7 and this ratio
was maintained throughout the 28-day time course. This relative persistency


ng/ml ng/ml ng/ml
105
PLASMA E2-Q+ PLASMA E2


82
Levels of E2-Q+ in the anterior pituitary were surprisingly high on day
1 then dropped rapidly to below that of brain levels by day 7 and steadily
decreased thereafter throughout the observation period. This initial rise in
E2-Q+ levels may be attributed to increased anterior pituitary uptake of the E2-
CDS followed by its rapid metabolism and clearance. Furthermore, the
relative elevation of E2-Q+ as well as E2, from day 7 to day 14, in anterior
pituitary may be caused by the anatomical relationship between the
hypothalamus and anterior pituitary gland. Estradiol released upon the
hydrolysis of E2-Q+, or the E2-Q+ itself, which is locked into brain, could be
delivered directly to the anterior pituitary by the capillary plexus of the
hypophyseal portal system. These capillaries in the median eminence lack
features of other brain capillaries and hence are not part of the blood-brain
barrier (Traystman, 1983). Thus, the median eminence would not be expected
to prevent the efflux of E2-Q+ from the brain, and thus transfer of E2-Q+ to the
anterior pituitary can be expected.
High levels of E2-Q+ seen in the kidney are likely because this organ is a
major site for the elimination of all metabolites of the E2-CDS. However, the
reasons for initial high levels of E2-Q+ in the lung and heart tissues of the
male rat are not clear. We speculate that since these organs receive high
blood flow, a substantial amount of the E2-CDS is delivered to and taken up
by these tissues initially. Additionally, despite higher E2-Q+ levels in heart
relative to those of lung, kidney, and anterior pituitary, the heart E2
concentrations were lower than these 3 tissues. Perhaps, a slow rate of
hydrolysis of E2-Q+, or a slower E2 metabolism, in heart tissue could
contribute to the higher levels of E2-Q+ and lower levels of E2 in this tissue.
This single-dose pharmacokinetic study supports the previous
observations of prolonged pharmacodynamic effects of the E2-CDS following


95
from plasma (Figure 12), liver (Figure 13), fat (Figure 14), and anterior
pituitary, kidney, lung, heart, and uterus (Tables 5 and 6). By 28 days (the last
sampling time) after a single injection of 1.0 mg E2-CDS/kg, the E2-Q+
concentrations remained elevated at 9.8 0.7 ng/g wet tissue (mean SEM) in
brain (Figure 10, left column, upper panel) and 10.6 0.2 ng/g in
hypothalamus (Figure 11, left column, upper panel). In contrast, peripheral
tissues concentrations of E2-Q+ were reduced to 2.9 0.1 ng/g in anterior
pituitary, 5.2 2.2 ng/g in kidney, 2.9 0.5 ng/g in lung, 1.7 0.3 ng/g in heart,
2.5 0.7 ng/g in uterus (Table 5, % reduction); and E2-Q+ values were
undetectable in plasma, liver, and fat (Figures 12-14, left columns, upper
panels) 28 days after administration of 1.0 mg E2-CDS/kg dose.
Similarly, E2 concentrations were maintained relatively high in whole
brain (Figure 10, right column of panels) and in hypothalamus (Figure 11,
right column of panels); however, E2 concentrations in peripheral tissues
(except for anterior pituitary and plasma) fell by more than 80% by day 7, and
by 97% by day 21, and were undetectable by day 28 (Table 6).
In contrast with E2 concentrations achieved following E2-CDS
administration, E2 levels in tissues following equimolar E2 administration
were remarkably low (Table 7) and in all tissues examined, the clearance of E2
was rapid (Table 7). A comparison of the tissue levels of E2 achieved at day 1
revealed that following E2-CDS administration, brain (Figure 10) and
hypothalamus (Figure 11) levels of E2 were 88- and 22-fold greater,
respectively, than levels observed in these tissues following 17 P-E2
administration (Table 7). By 7 days after treatment, the E2-CDS produced
brain and hypothalamic E2 concentrations that were 182- and 55-fold greater,
respectively, than those achieved by an equimolar 17 P-E2 dose.


116
accelerated the elimination of these metabolites from the peripheral tissues.
On the basis of these pharmacokinetic findings, the E2-CDS is expected to
exhibit pharmacodynamic responses with long duration of effects following a
single administration of the delivery system.
The present study was undertaken to determine whether the long half-
lives and the magnitude of E2-CDS metabolites in brain tissue (Chapters 5 &
6) correlate with the duration of pharmacodynamic effects. More specifically,
the objectives were (1) to assess the dose- and time-dependent effects of the E2-
CDS on brain-mediated responses, i.e. anterior pituitary hormones secretion
in OVX rats; (2) to compare E2-CDS with an equimolar dose of 17 P-E2; and (3)
to correlate the half-lives of the E2-CDS metabolites with the duration of
pharmacodynamic effects mediated by E2.
Materials and Methods
All the samples analyzed in this study were obtained from the animals
used and described in the preceding study (Chapter 6); this chapter presents
further data on evaluation of the pharmacodynamic responses of E2-CDS.
Briefly, rats were ovariectomized (OVX) and two weeks later were
administered a single iv injection of the E2-CDS at doses of 0 (HPCD), 0.01, 0.1,
or 1.0 mg/kg body weight or E2 at a dose of 0.7 mg/kg (equimolar to the 1.0 mg
E2-CDS dose). Animals (7 per group) were then killed by decapitation 1, 7,14,
21, or 28 days after the drug administration and plasma, anterior pituitary,
and uterine tissues were collected for subsequent analysis.
Plasma luteinizing hormone (LH), follicle-stimulating hormone (FSH),
growth hormone (GH), and prolactin (PRL) concentrations were measured in
duplicate by the RIA using NIDDK kits. Plasma LH, FSH and GH values are


181
Derendorf H. and M. Kaltenbach. Coulometric high-performance liquid
chromatographic analysis of morphine in biological fluids. T. Pharm.
Sci. 75:1198-1200 (1986).
Docke F., P. Ledwon, B. Sturzebecher, W. Rohde, and G. Dorner. Oestrogen
priming for the positive oestrogen feedback: Site of action. Exp. Clin.
Endocrinol. 94:55-60 (1989).
Doisy E.A., C.D. Veler, and S.A. Thayer. Folliculin from the urine of pregnant
women. Am. T. Physiol. 90:329-330 (1929).
Doisy E.A., C.D. Veler, and S.A. Thayer. The preparation of the crystalline
ovarian hormone from the urine of pregnant women. T. Biol. Chem.
86:499-509 (1930).
Doring V.G.K. Klimakterische beschwerden. Fortschr. Med. 94:1058-1067
(1976).
Drill V. and C.W. Calhoun. Oral contraceptives and thromboembolic disease.
T. Am. Med. Assoc. 219: 583-596 (1972).
Dusterberg B. and Y. Nishino. Pharmacokinetic and pharmacological features
of oestradiol valerate. Maturitas 4:315-324 (1982).
Eleftheriou B.C. and C.L. Dobson. Effects of neonatal and adult treatments
with gonadal hormones on choline acetylase activity in brain regions
of male mice after fighting. Psychopharmacology 27:45-52 (1972).
Estes K.S., M.E. Brewster, and N. Bodor. Use of a chemical redox system for
brain enhanced delivery of estradiol decreases prostate weight. Ann.
NY Acad. Sci. 507:334-336 (1987a).
Estes K.S., M.E. Brewster, and N. Bodor. A redox system for brain targeted
estrogen delivery causes chronic body weight decrease in rats. Life Sci.
42:1077-1084 (1988).
Estes K.S., M.E. Brewster, J.W. Simpkins, and N. Bodor. A novel redox
system for CNS-directed delivery of estradiol causes sustained LH
suppression in castrate rats. Life Sci. 40:1327-1334 (1987b).
Ettinger B., I.M. Golditch, and G. Friedman, Gynecologic consequences of
long-term, unopposed estrogen replacement therapy. Maturitas
10:271-282 (1988).


47
by Steffens (1969). All animals were then monitored for post surgical
recovery before conducting an experiment on them.
Two methods were employed for collecting blood samples. In most
experiments, animals were killed by decapitation and the trunk blood was
collected in heparinized tubes. In studies which required frequent blood
sampling, animals equipped with atrial cannula were transferred to special
sampling chambers and serial blood samples (1 ml) were removed through
the cannula. All blood samples were collected in a room separate from the
animal quarters. The blood samples were centrifuged and the plasma
separated and stored at -20C until hormone analysis by the
radioimmunoassay (RIA).
Drug Treatment
Steroid Treatment
The E2-CDS treatments described in this dissertation were only given
intravenously (iv). An aqueous solution of E2-CDS was prepared in 20%
HPCD on the day of injection and administered iv (via tail vein), a procedure
which required a brief restraint of the rat without anesthesia.
The 17 P-E2 treatments consisted of either iv administration (Chapters
6 & 7) or sc implantation of E2 pellet (Chapter 9). Implantation of the E2 pellet
was performed in animals under light metofane anesthesia. It should be
pointed out that these pellets did require pre-conditioning (i.e. soaking in
PBS) before implantation. After implantation, each E2 pellet produces a
transient high concentration of E2 in plasma that is followed by a sustained
blood E2 levels for about 2 weeks.


CHAPTER 5
DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN MALE RATS
Introduction
Estrogens are intrinsically lipophilic (Abraham, 1974) and readily cross
the blood-brain barrier (BBB) to gain access to the central nervous system
(CNS). However, when inside the CNS, there is no mechanism to prevent
their redistribution back to the periphery as blood levels of the steroid decline
(Davson, 1976). Indeed, when these hormones are used therapeutically to
specifically target the CNS, the steroids tend to equilibrate among all body
tissues due to their high lipophilicity (Pardridge & Meitus, 1979). As a result,
only a fraction of the administered dose accumulates at or near the site of
action in the brain. This property of the estrogens necessitates, either frequent
dosing, or the administration of a depot form of the estrogen to achieve and
maintain therapeutically effective concentrations in the brain (Spona &
Schneider, 1977). Both of these treatment strategies lead to sustained
increases in peripheral estrogen levels.
Furthermore, estrogen receptors are present in many peripheral tissues
(Walters, 1985), where they mediate a myriad of physiological and
pharmacological effects (Murad & Haynes, 1985; Walters, 1985). This further
creates the potential of untoward peripheral side effects (Thomas, 1988). In
fact, constant increases in peripheral tissue exposure to estrogens have been
shown in numerous studies to precipitate various peripheral toxicities,
including risk of breast and endometrial cancer (Hurst & Rock, 1989; Persson,
75


199
Suyemitus T. and H. Terrayama. Specific binding sites for natural
glucocorticoides in plasma membranes of rat liver. Endocrinology
96:1499-1508 (1975).
Swerdloff R.S. Physiology of male reproduction: Hypothalamic-pituitary
function. In: Campbell's Urology, P.C. Walsh, R.F. Gittes, and A.D.
Perlmutter (eds) 186-200, Saunders, Philadelphia (1986).
Szego C.M. and S. Roberts. Steroid action and interaction in uterine
metabolism. Recent Prog. Horm. Res. 8:419-470 (1953).
Tallarida R.J. and R.B. Murry. Manual of Pharmacological Calculation with
Computer Programs. Springer, New York (1981).
Tarttelin M.F. and R.A. Gorski. Variations in food and water intake in
normal and acyclic female rats. Physiol. Behav. 7:847-852 (1971).
Taylor R.N. and R.G. Smith. Effects of highly purified estrogen receptors on
gene transcription in isolated nuclei. Biochemistry 21:1781-1787 (1982).
Tepperman J. Metabolic and Endocrine Physiology. Year Book Medical
Publishers, Inc. Chicago (1981).
Thomas D.B. Steroid hormones and medications that alter cancer risks.
Cancer 62:1755-1767 (1988).
Toft D. and J. Gorski. A receptor molecule for estrogens: Isolation from the
rat uterus and preliminary characterization. Proc. Natl. Acad. Sci. USA
55:1574-1581 (1966).
Towle A.C. and P.Y. Sze. Steroid binding to synaptic plasma membrane:
Differential binding of glucocorticoids and gonadal steroids. T. Steroid
Bioch. 18:135-143 (1983).
Trapido E.J., L.A. Brinton, C. Schairer, and R. Hoover. Estrogen replacement
therapy and benign breast disease. T. Natl. Cancer Inst. 73:1101-1105
(1984).
Traystman R.J. Microcirculation of the brain. In: The Physiology and
Pharmacology of the Microcirculation. Vol. 1, N.A. Mortillaro (ed.) 237-
298, Academic Press, New York (1983).
Upton V. Therapeutic considerations in the management of the climactric. L
Reprod. Med. 29:71-79 (1984).


173
has potential application for various clinical conditions, it may have some
drawbacks due to its mechanism of drug delivery that is, locking the drug into
the CNS tissue for prolonged period of time. Potential problems may include
the following: 1) The most frequent, unpleasant side effects of estrogen
hormones, particularly E2, are nausea and vomiting. These adverse effects
are due most likely to their central effects. Having "locked" irreversibly a
depot E2 inside the brain, in the form of E2-Q+, with sustained release of the
hormone over prolonged period of time may cause complications in this
regard. 2) In clinical practice, it is the standard procedure to monitor drug
concentration in patients by taking blood samples and analyzing plasma or
serum to obtain necessary informations about the drug distribution,
metabolism, and perhaps drug concentration at the site of action. However,
with the application of E2-CDS, plasma or serum concentration of the E2-CDS
metabolites do not reflect the brain levels of these compounds, since these
metabolites are preferentially retained by the brain. Thus, this feature of the
E2-CDS may be a disadvantage with regard to drug monitoring. 3) Certain
regions of the brain, i.e. arcuate nucleus of the hypothalamus, are exquisitely
susceptible to prolonged exposure and/or high concentration of estrogen
hormones. Since E2-CDS metabolites persist evenly throughout the brain for
prolonged period of time after drug administration, potential for
neuropathological lesions by this delivery may exist. 4) Perhaps the most
important issue regarding this drug delivery would be the question of
estrogen receptors down regulation or desensitization with prolonged
duration of exposure to the hormone. So, if estrogen receptors down
regulate, like other receptor types, the prolonged residence and the sustained
release of E2 in the brain may not be useful for the duration of residence in
that tissue. And finally, although we chose the iv route for drug


BIOGRAPHICAL SKETCH
Mohamad Hossein Rahimy was born on September 5, 1954, in Esfahan,
Iran. It seems pretty old to be a graduate student at that age. But certainly it is
never too late to satisfy the ego, specially with regard to knowledge. After he
graduated from high school in 1974, he was compelled to serve a two-year
term in the late Shah's military. Shortly after the compulsory service, he
made a trip to the Florida Beach, St. Petersburg, that left a lasting impression.
He entered St. Petersburg College in 1979. Two years later, he transfered to the
University of Florida in Gainesville. He earned his B.S. in microbiology and
cell science in December 1982. After graduation, he returned to his homeland
to start his career plans, not knowing that things were changed allot. In a year
or so, he decided to pursue an academic career. Thus, after his return to
Gainesville in 1985, he entered the graduate program of the College of
Pharmacy. In June 1985, he joined the laboratory of Dr. James W. Simpkins at
the Department of Pharmacodynamics, and there he remained until the
present time. Should he ever graduate, he plans to continue his research
career in either pharmaceutical industry or academia.
202


ng/g ng/g ng/g
101
BRAIN E2-Q+
BRAIN E2


54
liver, and kidney, 100% methanol was used; for E2 extraction from lung,
heart, and fat, 100% acetone was used. The reason for using different solvents
was that methanol extraction yielded high and consistent (low CV, high CC)
recovery of E2 from brain, anterior pituitary, liver, and kidney but resulted in
low or inconsistent recovery of E2 from lung, heart, and adipose tissue.
However, using 100% acetone, an acceptable and consistent recovery of E2 was
observed for lung, heart, and adipose tissue. For E2-Q+ extraction, all tissue
homogenate pools were prepared in water/acetone (50:50; v:v).
E? recovery estimation. Duplicate aliquots of homogenate, each having
a concentration of 100 mg tissue/ml solvent, were spiked with 90, 180, or 360
pg E2 and vortexed for 1 min. The added steroid was allowed to equilibrate
with the homogenate for 30 min at room temperature. The spiked
homogenates were then centrifuged for 10 min at 1,500 x g. The supernatant
was decanted into a clean test tube and the pellet discarded. For plasma,
duplicate 1 ml aliquots of plasma (from male rats) were spiked with the
aforementioned doses of E2 standards, samples were allowed to equilibrate for
30 min, but no subsequent centrifugation was done. Following extraction, E2
recovery was determined as described below.
E9-O+ recovery estimation. The procedure used for E2-Q+ adhered
closely to that described for E2 recovery. Duplicate aliquots of homogenate
were spiked with 150,300, and 600 pg E2-Q+ and processed similarly to E2-
spiked homogenates. Serving as a control for the percent of hydrolysis of E2-
Q+, separate duplicate aliquots of homogenates were processed without the
addition of E2-Q+ standards until after the supernatant were isolated. These
supernatants were spiked with 300 pg E2-Q+.


102
Figure 11. Dose and time-dependent effects of the E2-CDS on E2-Q+
concentrations (left column of panels) and E2 concentrations (right
column of panels) in hypothalamus of ovariectomized rats.
Animals received a single iv (tail vein) injection of the E2-CDS on
day 0 at a dose of 1.0 mg/kg (upper panels), 0.1 mg/kg (middle
panels), or 0.01 mg/kg (lower panels). Animals were killed 1, 7,14,
21, or 28 days after the drug administration. Tissue samples of
known wet weight at a concentration of 1 mg/20 pi solvent were
processed and assayed for E2-Q+ and E2 by the method described in
Chapter 4. Also, tissue homogenates from HPCD-treated rats were
analyzed for E2 hormone background. Represented are means
SEM for n = 7 rats per group per sampling time.


Ant Pit Content (ng)
o
Ant Pit Content (ng)
Brain Content (ng)
N) *. Os 00
o o o o
o o o o
I I 1 I I 1 I
tfl
N>
I
o
+
Brain Content (ng)
i ro
Ol O U1 o
I I I I I I I
IfS
00
KO


E2 (pg/g or pg/ml) E2-Q+ (pg/g or pg/ml)
71
Days post Injection
Figure 5. Effects of a single iv dose of the estradiol-chemical delivery system
on serum and brain levels of E2-Q+ (upper panel) and E2 (lower
panel) or the HPCD vehicle (day 0). Each point represents the
group mean SEM. N = 6 animals per group.


ACKNOWLEDGEMENTS
This work would never have come to fruition without the
encouragement, assistance and advice of many individuals whom I am very
grateful to. First, I wish to express my sincere appreciation and gratitude to
my mentor, Dr. James W. Simpkins, for his expert guidance, encouragement,
and support. Throughout my graduate study at the University of Florida, I
had ample opportunity to learn by experience under the skillful guidance of
Dr. Simpkins. I also wish to express great thanks to the other members of my
committee, Dr. Nicholas Bodor, Dr. William Millard, Dr. Edwin Meyer, and
Dr. Ralph Dawson, who have imparted valuable advice as well as their
critical evaluation of my work. I would also like to extend thanks to Dr.
Michael Meldrum, Dr. Michael Katovich, Dr. Wesley Anderson, and Dr.
Anna Ratka for their advice and assistance.
I would like to thank the many others who contributed their time and
efforts, especially Victoria Red Patterson, Becky Hamilton, Lee Glancey, Terry
Romano, Debby Andreadis, Billie Jean Goins, Roxane Federline, and Denise
Blake, who assisted me in various aspects of this work. The assistance of
Anup Zutshi regarding the kinetic analysis is greatly appreciated. My
personal thanks go to Dr. Lee Ann Burgland and Dave Wallace whose
cooperations during these many years made graduate school more bearable. I
extend thanks to new graduate students Singh Meharvan, Melanie King,
Melanie Pecins, and Jean Bishop-Sparks who have already taken over the
reins in the lab for accepting the challenge.


135
Effect of CAST or E?-CDS on Plasma Hormones
CAST significantly increased the plasma gonadotropins (LH and FSH)
concentrations (Tables 9 & 10). By contrast, the plasma gonadotropin
concentrations in animals treated with E2-CDS were not significantly altered
from control (Tables 9 & 10). Most likely, LH values in animals treated with
E2-CDS may have been significantly suppressed compared with control
values, because more than 50% of LH values were below the detection limit
of assay. However, since the sensitivity of LH assay was assigned as 0.25
ng/ml, statistically significant LH suppression was not achieved. A
significant elevation in plasma PRL concentrations was observed only in
animals treated with 3 injections of E2-CDS (Table 9). Likewise, plasma E2
concentrations were significantly elevated in animals treated with 2 and 3
injections of E2-CDS at 7 days after the last injection (Table 9).
Discussion
The primary objective of endocrine therapy in prostate malignancy is
the induction of an effective androgen deprivation, thus abolishing the
growth promoting effects of androgens on the prostate tissue. These
therapeutic objectives can be achieved by several mechanisms at different
levels of the hypothalamo-pituitary-gonadal axis: 1) suppression of
hypothalamic LHRH and hence, of pituitary LH release, thereby inhibiting T
production by the testis; 2) surgical CAST which eliminates more than 90% of
circulating T; 3) inhibition of androgen synthesis in the testis; and 4) blocking
androgen action at the receptor site in the prostate. Thus, the choice of


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
ASSAY DEVELOPMENT, TISSUE DISTRIBUTION
AND PHARMACODYNAMICS OF A NOVEL ESTROGEN-CHEMICAL
DELIVERY SYSTEM FOR THE BRAIN
By
Mohamad H. Rahimy
August, 1990
Chairman: Dr. James W. Simpkins
Major Department: Pharmacodynamics
Enhanced delivery and sustained release of estradiol (E2) in the brain is
desirable for effective treatments of the menopausal hot flush, prostatic
adenocarcinoma, and fertility regulation. Our studies thus evaluated an E2-
chemical delivery system (E2-CDS) for the brain, which is based upon the
interconvertible dihydropyridine <=> pyridinium ion redox reaction. The E2-
CDS requires multiple, facile chemical conversions, including the oxidation
of E2-CDS to the corresponding quaternary ion (E2-Q+), which provides the
basis of locking the molecule within the brain, and the subsequent slow
hydrolysis of E2-Q+ by esterases to E2 in that tissue.
Initially, studies were undertaken to develop a reliable, specific, and
sensitive method to simultaneously measure E2-Q+ and E2 (two metabolites
xi


42
Taken together the findings reported in the literature suggest a strong
link among estrogen, the cholinergic system, and cognitive functioning in
women. Using different estrogen preparations and various Memory
Tests/instruments to assess cognitive functioning, results indicated that
estrogen therapy resulted in improvement in attention span, orientation,
memory, mood, and social interaction in a majority of postmenopausal
women with cognitive impairments (Fillet et al., 1986; Furuhjelm & Fedor-
Freyberg, 1976; Hackman & Galbraith, 1976; Honjo et al., 1989; Vanhulle &
Demol, 1976). Furthermore, the improvement in cognitive functioning, in
those patients who benefited, was correlated with an increase in circulating
concentrations of estrogens (Sherwin, 1988). Additionally, biochemical data
obtained from the rat brain further support the involvement of estrogens in
cognition. Studies in the rat brain have shown that cholinergic neurons
respond to administration of estrogen by (1) increasing the activity of choline
acetyltransferase (ChAT, the enzyme that synthesizes acetylcholine)
(Eleftheriou & Dobson, 1972; Iramain et al., 1980; Kaufman et al., 1988; Luine
et al., 1975,1980,1983,1985;) in the basal forebrain, cortex, hippocampus and
hypothalamus; (2) increasing high affinity choline uptake as well as
acetylcholine synthesis (O'Malley et al., 1987) in cerebral cortex; and
furthermore (3) the responses of cholinergic system to estrogen
administration (i.e., enzyme activity) were observed in female but not male
rats and positively related to the dose of E2 and blocked by an estrogen
antagonist (Luine et al., 1980,1983). Finally, E2 application has been shown to
enhance the excitatory actions of Glu, and Glu is known to be essential in
long-term potentiation and memory formation (Smith et al., 1987).
Collectively, available data support the role of estrogens in cognitive
functioning. However, the hypothesis does not exclude the possibility of the


155
Discussion
In this study, we used the morphine-dependent, naloxone-withdrawal
rat model to evaluate the effectiveness of E2-CDS for the treatment of hot
flushes. To our knowledge, this is the only animal model (Simpkins &
Katovich, 1984) available to evaluate effectively alternative therapies for this
common disease.
The results of this study for the first time demonstrate that: (1) the E2-
CDS can attenuate significantly the naloxone-induced surge in TST of the
morphine-dependent rats; (2) 17 P-E2 pellet treatment has no effect on
naloxone-induced rise in TST; and (3) the significant effects of E2-CDS are
achieved at low plasma E2 concentrations. Furthermore, these data support
the hypothesis that the primary site of action where E2 exerts its stabilizing
effects on the mechanism(s) which control thermoregulation is in the CNS
and not in the periphery (Casper & Yen, 1985; Lauritzen, 1973; Yen, 1977).
This is clearly shown by the fact that treatment with 17 J3-E2 pellet which
produced superphysiological plasma E2 (2-fold greater than those produced by
the multiple injections of E2-CDS) did not exert any stabilizing effect on the
rise of TST. These pellets were shown in our laboratory to release high
concentrations of E2 (280-180 pg/ml) in plasma on day 1 and 2 after
implantation. However, between days 5 and 14 of implantation stable levels
of E2 (100-80 pg/ml) were observed. Contrary to our findings in the present
study, 17 P-E2 has recently been reported to produce significant attenuation in
the magnitude of the flush response using the same animal model (Katovich
& O'Meara, 1986). The reason for this discrepancy seems to be the result of
very high concentrations of plasma E2 in the previous study. Utilizing E2


25
Blood-Brain Barrier
Historical Overview
The blood-brain barrier is vital to brain function. Neurons are
extremely sensitive to the ratio of the concentrations of ions across their
plasma membranes. Furthermore, the concentrations of excitatory and
inhibitory substances in the extracellular neuronal environment must be
tightly regulated for optimal brain functioning. Perhaps because of the
functional intricacy of neurons, the mammalian CNS acquired this
ultrastable cellular environment to perform effectively. This stable
extracellular environment is achieved by several morphological and
enzymatic components collectively referred to as the blood-brain barrier
(BBB). The BBB is not an absolute barrier, but rather a selective and
protective barrier (Neuwelt, 1989; Rapoport, 1976; Suckling et al., 1986). Many
pharmacological agents are excluded from entering the brain because of the
existence of BBB (Bodor & Brewster, 1983; Schanker, 1965).
The concept of barrier was originally postulated by Ehrlich at the end of
the 19th Century (Levin, 1977; Pardridge et al., 1975; van Deurs, 1980). When
vital dyes, such as trypan blue, were injected into the bloodstream, the dyes
penetrated almost all organs of the body but did not enter the brain tissue or
the cerebral ventricles (Levin, 1977; Pardridge et al., 1975; van Deurs, 1980).
However, when dyes were injected into the cerebral ventricles, brain tissue
was stained and the dye did not readily enter the bloodstream. Thus, the lack
of staining by the dyes was not an intrinsic property of brain tissue, rather the
existence of a dynamic barrier interface between the blood and the brain.


78
extraction and chromatographic separation (using internal standard), and the
weight of the tissue sample used.
Results
The results of this experiment are presented in Figures 6-9. Figure 6
shows brain (upper panels) and plasma (lower panels) concentrations of E2-Q+
and E2 at various times following administration of the E2-CDS. Brain E2-Q+
concentrations increased to 318 14 ng/g tissue (mean SEM) on day 1,
followed by a linear decline to 39 2 ng/g on day 14. This result indicated a
reduction in E2-Q+ concentration of 46% by 7 days and 88% by 14 days after
administration of the E2-CDS. In contrast, plasma concentrations of E2-Q+
increased to 6.1 0.3 ng/ml on day 1, then rapidly decreased by 79% at day 7
and remained at very low levels (0.23 0.03 ng/ml) at day 14 after the E2-CDS
treatment.
Brain concentrations of E2 increased to 8 0.5 ng/g (day 1) then
decreased steadily to 2 ng/g (day 14), indicating a sustained-release behavior
from brain E2-Q+. In contrast, plasma E2 concentrations increased to only 0.28
0.1 ng/ml (day 1) and steadily declined thereafter.
Figure 7 shows the time-concentration profiles of E2-Q+ (upper panel)
and E2 (lower panel) in liver and fat tissues at various times following
administration of the E2-CDS. Both E2-Q+ and E2 were detected in these
tissues throughout the time-course studied. As expected, these tissues
showed rapid clearance of E2-Q+ as well as E2. The E2-Q+ concentrations
decreased from 77 10 ng/g and 71.7 19.5 ng/g (day 1) to 5.4 1.3 ng/g and
1.9 0.5 ng/g (day 14) in liver and fat, respectively. This indicated a reduction


80
By contrast, the anterior pituitary content of E2-Q+ was only 1/260 to 1/170 of
that observed in the brain.
Similarly, the brain E2 content was 15.8 0.9,10.4 0.8, and 3.2 0.1
ng/brain at 1, 7 and 14 days following administration, respectively, while the
anterior pituitary E2 content was 0.47 0.04, 0.06 0.004, and 0.035 0.006
ng/anterior pituitary at these sampling times. As such, the anterior pituitary
E2 content was only 1/90 to 1/34 of that observed in the brain throughout the
time-course studied. Thus, the absolute amounts (contents) of E2 and E2-Q+
were many fold higher in the brain even though the anterior pituitary
concentrations of E2 and E2-Q+ were initially higher than in the brain.
Discussion
The results of this single-dose distribution study demonstrated that
both E2-Q+ and E2, two metabolites of the E2-CDS, were present in all tissues
analyzed up to 14 days (the last sampling time) after treatment of male rats
with the E2-CDS. Moreover, over the time-course studied, the distribution
profiles indicated that: a) regardless of the tissues evaluated, E2-Q+ levels
were many fold higher than E2 levels at each time point in a particular tissue,
indicating a slow rate of hydrolysis of E2-Q+ to E2; b) the increased
brain/plasma ratios of E2-Q+ as well as E2, confirmed that "locking" of the
charged moiety, E2-Q+, into the brain had occurred; and c) E2-Q+ is retained in
the CNS tissue but is rapidly cleared from the peripheral tissues, an
observation which is predicted by the inherent physicochemical properties of
the delivery system.
Brain-distribution profile revealed that E2-Q+, the quaternary form of
the delivery system, persists in the brain with a half-life of about 8 days, but it


127
Table 8: Dose and Time-Dependent Effects of the E2-CDS on Peripheral Tissue
Weights in Ovariectomized Rats
Tissue
Drue
Dose
Davs after treatment
Anter.
HPCD
mg/kg
0
11.8
1
7
14
21
28
Pituit.
0.5
(mg)
e2
0.7
13.5
12.8
ND
ND
ND
0.7
0.7
E2-CDS
0.01
12.0
13.1
14.4
11.9
13.0
0.8
0.5
1.5
0.8
0.6
E2-CDS
0.1
12.8
15.6*
14.9*
13.1
12.6
0.9
0.9
1.5
0.5
0.6
E2-CDS
1.0
13.3
19.0*,b
20.7* ,b
16.2*,b
15.0*,c
0.9
1.0
1.3
1.0
0.6
Uterus
HPCD
154.2
(mg)
e2
0.7
8.6
261.4*
182.9
ND
ND
ND
13.0
24.0
E2-CDS
0.01
183.8
179.1
196.4
121.5
107.8
12.3
29.0
42.0
7.0
8.0
E2-CDS
0.1
237.7*
234.4*
210.1*
161.1
128.4
12.0
27.0
33.0
8.0
5.0
E2-CDS
1.0
281 .4*, a
427.0*, b
372.9*, b
267.4*,b 235.4*h
20.0
31.0
28.0
31.0
16.0
Values are the mean tissue weights SEM. ND = Not determined.
* Different from time 0 values,
a Different from 0.01 mg/kg dose,
b Different from 0.01 and 0.1 mg/kg doses,
c Different from 0.1 mg/kg dose.


156
pellet (Katovich & O'Meara, 1986) which produced very high plasma E2
concentrations (7-fold greater than E2 concentrations produced by E2 pellet in
the present study), would certainly lead to higher brain E2 levels. Therefore,
maintaining therapeutically effective E2 levels in the brain will eventually
produce brain-mediated E2 effect. Nevertheless, estrogens or estrogen-esters
are used therapeutically in postmenopausal patients (Campbell & Whitehead,
1977; Casper & Yen, 1985; Lauritzen, 1973; Upton, 1984; Yen, 1977), and these
agents are effective in alleviating menopausal symptoms (Campbell &
Whitehead, 1977; Paterson, 1982; Upton, 1984). Unfortunately, these
estrogenic products are administered either in frequent doses, or as a large
depot form, to achieve and maintain therapeutically effective levels in the
brain. Both of these treatment strategies lead to sustained increases in
peripheral estrogen levels which have been shown in numerous studies to
increase peripheral toxicides (Barrett-Conner et al., 1989; Burkman, 1988;
Campbell & Whitehead,1977; Casper & Yen, 1985; Kaplan, 1978; Lauritzen,
1973; Thomas, 1988; Trapido et al., 1984; Upton, 1984).
Our previous observations, including the E2-CDS kinetics in intact
male rats (Chapter 5) and in OVX rats (Chapter 6) as well as the
pharmacodynamic effects in OVX rats (Chapter 7), long-term suppression of
gonadotropin secretion in CAST male rats (Simpkins et al., 1986), long-term
suppression of androgen secretion and weights of androgen-responsive
tissues (Chapter 8), stimulation of masculine sexual behaviors in CAST male
rats for 28 days (Anderson et al., 1987a) following a single iv administration of
the E2-CDS and our present observation, that E2-CDS but not 17 (3-E2
significantly attenuates the naloxone-induced rise in TST, provide strong
support for the proposed mechanism of the E2-CDS. That is, following E2-
CDS administration, the delivery system undergoes rapid oxidation


114
Table 7: Effects of an Equimolar Dose of E2 (0.7 mg/kg) on the Tissue
Concentrations of E2a
Tissue
Vehicle (HPCD)b
Days after treatment
lc
7c
Brain
0.05 0.03
0.33 0.04
0.09 0.07
Hypothalamus
0.03 0.01
1.16 0.14
0.35 0.08
Ante. Pituitary
0.02 0.01
0.99 0.07
0.28 0.02
Plasma
UDd
0.03 0.01
0.02 0.01
Kidney
UD
0.55 0.13
0.04 0.01
Lung
UD
0.26 0.30
0.01 0.01
Heart
UD
0.14 0.04
0.01 0.01
Liver
0.02 0.01
0.80 0.20
0.11 0.04
Fat
0.05 0.03
1.12 0.29
0.11 0.05
Uterus
UD
2.08 0.65
0.17 0.14
a This dose of E2 (0.7 mg/kg) is equimolar to 1.0 mg E2-CDS/kg dose,
b Residual E2 concentrations (ng/g wet tissue or ng/ml; mean SEM).
c E2 concentrations (ng/g wet tissue or ng/ml; mean SEM) at various
times following administration of E2.
d Undetectable (below the sensitivity of RIA for E2).
Note: For the determination of E2 concentration in this table, we used 200 to
300 mg tissue to obtain a detectable E2 baseline in the RIA since tissue
E2 concentration was very low following 17 P-E2 injection. And since
our assay methodology was optimized for up to 100 mg tissue, using
excess amount of tissue in the assay might result in unreliable E2
values following 17 P-E2 administration.


10
Secretion and transport
The E2-producing cells in the ovary and corpus luteum do not
characteristically prepackage large amounts of steroid hormone for release.
Rather, these endocrine tissues store the hormone precursor, cholesterol-
ester, in the form of lipid droplets inside the hormone-producing cells
(Rossmanith et al., 1990). This indicates that, in these secretory cells, the
signal for E2 release is perhaps tightly coupled to that of estrogen-hormone
synthesis. Thus, the newly synthesized hormone will be released into the
circulation for transport to target tissues. The signals for release are primarily
those of the anterior pituitary tropic factors (LH & FSH). It is believed that E2
is released in pulsatile fashion and this perhaps is the result of the episodic
modulating influence of LH (Rossmanith et al., 1990).
Estradiol hormone, like other steroid hormones, when secreted into
the blood, it is primarily transported by carrier proteins. E2 may be
transported by: (a) plasma albumin (60%) with low affinity binding; (b) sex
hormone-binding globulin (38%) with high affinity binding; and (c) in free
(dialyzable) form (2%) (Moutsatsou & Oakey, 1988; Pardridge, 1988a). The
carrier-bound E2 is biologically inactive and sequestered in plasma while only
the free fraction is regarded as biologically active hormone. Recent studies
have, however, suggested that the carrier-bound pool of E2 may also be
available for uptake by the target tissues (Pardridge, 1988a). It was suggested
that there are two possible mechanisms for the delivery of carrier-bound E2 to
target tissues (Pardridge, 1988a). One mechanism involves interactions
between the carrier protein surface and the surface of the organ
microcirculation that results in a conformational change about the carrier
binding site and thus enhanced dissociation of E2 hormone. The second


ng/g ng/g ng/g
109
FAT E2-Q+
FAT E2


83
administration of a single dose. LH secretion in castrated male rats was
suppressed for greater than 21 days (Estes et al., 1987b; Simpkins et al., 1986),
sexual copulatory behavior was stimulated for 28 days (Anderson et al.,
1987a), and body weight was suppressed for 36 days (Estes et al., 1988;
Simpkins et al., 1988) after doses of E2-CDS of 1 to 3 mg/kg. These prolonged
effects of the E2-CDS are consistent with the observation here of the
accumulation of E2-Q+, the oxidized form of the delivery system, in the rat
brain and its long half-life (t 1/2 = 8 days) in this tissue. From this store of E2-
Q+, E2 is released through slow hydrolysis and exhibits a half-life similar to
that of E2-Q+.
Finally, since we evaluated the tissue distribution of E2-CDS
metabolites in intact male rats, determination of E2-CDS metabolites in the
testes, prostate, and seminal vesicle tissues would have certainly added more
valuable informations to the results presented in this chapter. Unfortunately,
at the time of experimental investigation, these tissues were not collected for
analysis of E2-CDS metabolites. Certainly, in future studies involving E2-CDS,
these tissues need to be evaluated for E2-Q+ and E2 distribution and clearance.
Furthermore, the question of blood-testes barrier must be addressed. This
would be of great interest to find out whether E2-Q+ is being "locked" in this
tissue similar to CNS tissue or behaves like the rest of peripheral tissues.
However, effects of the E2-CDS on weights of these androgen-responsive
tissues were examined and the are presented in Chapter 8.
In conclusion, these observations are consistent with the proposed
mechanism of the redox-based E2-CDS and the contribution of the BBB to the
chronic retention of the charged, hydrophilic E2-Q+ in the rat brain. This
observed tissue distribution pattern of E2-Q+ and E2 may explain the long
term pharmacological effects of the E2-CDS in the male rat.


160
Time (min)
Figure 25. Effects of the E2-CDS or E2 pellet on the mean RT responses
induced by naloxone administration (0.5 mg/kg; sc) to morphine-
dependent ovariectomized rats. Animals were treated weekly
with either the vehicle (HPCD, x3, 3 injections total over 3 weeks),
E2-CDS (1.0 mg/kg, x3, 3 injections total over 3 weeks), E2 pellet
(0.5 mg, x3, 3 implants total over 3 weeks), or E2-CDS (1.0 mg/kg,
xl, 1 injection for 1 week). Morphine dependency was produced
after initiation of estrogen treatment. Four days after the
initiation of morphine treatment, on the morning of the 21st day
(long-term) or the 7th day (short-term) after the initiation of
estrogen treatment, animals were lightly restrained in wire mesh
cages. RT was then recorded at 2-min intervals. Represented are
the means SEM for n=7-8 rats per group.


PERCENT BOUND PERCENT BOUND
70
Figure 4. Inhibition of 125I-E2 binding to an E2 antibody caused by increasing
amounts of brain tissue from rats treated with the estradiol-
chemical delivery system (1.0 mg/kg). The upper panel depicts
brain tissue wet weights extracted over the range of 2 to 100 mg for
E2 at 1 or 7 days after treatment with E2-CDS. The lower panel
depicts brain tissue extracted over the range of 0.25 to 50 mg for E2-
Q+ at 1,7 or 14 days after treatment with E2-CDS. Data are
expressed on logit-log graph (% bound on ordinate is based on
logit = log (percent bound/100 percent bound)). The results
indicated that the assay method used accurately measures E2-Q+
and E2 concentrations over a wide range of tissue weights.


63
binding over the range of 20 to 3600 pg/ml. This level of sensitivity is
substantially greater than the recently published HPLC methods which report
levels of sensitivity of 50 and 10 ng/ml of plasma for E2 and E2-Q+,
respectively (Mullersman et al., 1988). Additionally, the antibody used was
very specific for estradiol, showing a cross-reactivity of <0.3% for E2-Q+ and
has also been described to cross-react with estriol and estrone at the level of
0.3 and 1.1%, respectively. In brief, the RIA described here is sensitive and
specific for E2.
The recovery of E2 is dependent upon the extraction efficiency of the
organic solvent used and the elution efficiency of E2 loaded on the Cis
columns. Since column elution of E2 with 100% methanol was essentially
quantitative, the recovery of E2 was equivalent to the extraction efficiency.
Methanol extraction yielded high and consistent (low CV, high CC) recovery
of E2 from brain, anterior pituitary, liver and kidney but resulted in low or
inconsistent recovery of E2 from lung, heart and adipose tissue. However,
using 100% acetone, an acceptable and consistent recovery of E2 was observed
for lung, heart and adipose tissue. While the reason for this tissue-specific
solvent extraction is not clear, the results indicate that the judicious choice of
solvent allows for the reliable estimate of E2 concentrations in a variety of
tissues. Indeed, E2 was precisely measured in a variety of tissues over a 4-fold
change in the concentration of E2 or over a 50-fold change in the amount of
tissue used in the determination (Figures 3 & 4).
The recovery of E2-Q+ was limited primarily by the percent hydrolysis
of the E2-Q+, since extraction efficiency in water:acetone (50:50; v:v) ranged
from 65 to 81 % for an individual tissue. We observed that our base-catalyzed
hydrolysis of E2-Q+ yielded values of 54% to 69% for all tissues except fat (37%)
and serum (30%). While base hydrolysis is not complete, the % hydrolysis


26
Subsequent studies which utilized different compounds, including drugs and
radioactive tracers further substantiated the concept of BBB. Furthermore, it
was later discovered that many small molecules were similarly excluded from
transport into the brain. This, then, led to the suggestion that BBB is absolute;
a concept which was soon thereafter refuted when the nutrient requirements
of the brain were elucidated (Davson, 1976).
Ultrastructural studies have shown that there are several differences
between the systemic capillaries and the cerebral capillaries which explain
their permeability differences. Morphologically, the ultrastructural feature
which most distinguishes CNS microvessels is the endothelial cell lining of
the these vessels (Brightman, 1977). The morphological features of the brain
capillaries were elucidated using horseradish peroxidase (HRP). This
relatively small enzyme has a high affinity for radiopaque substances, such as
osmium tetroxide which can be visualized, and it does not cross cerebral
capillaries (Brightman, 1977). When introduced directly into the brain, it
readily diffused throughout the extracellular space but did not pass between
the endothelial cells of cerebral capillaries (Reese & Karnovsky, 1967). So the
anatomical basis for the BBB was identified as the endothelial lining of the
cerebral capillaries. Unlike systemic endothelial capillaries, the cerebral
counterparts are joined by tight junctions (Brightman & Reese, 1969). These
tight junctions form a zona occludens and provide, for molecules like HRP,
an absolute barrier. Morphologically, these junctions consist of aligned intra-
membranous ridges and grooves which are in close apposition (Oldendorf,
1977; Shivers, 1979). Two additional morphological features of the cerebral
capillaries contribute to the BBB as well: (i) cerebral endothelial cells have a
paucity of vesicles and of vesicular transport feature and (ii) the perivascular


131
A potential problem associated with currently used estrogens is that
these steroid hormones equilibrate among all body tissues due to their high
lipophilicity. As a result, only a fraction of the administered estrogen dose
accumulates at or near the intended site of action. Indeed, when these
hormones are used therapeutically to specifically target the brain, the steroids
must be given either frequently or in high doses in order to maintain
therapeutically effective concentrations in the brain. Both of these treatment
strategies lead to sustained increases in peripheral estrogen levels and, in
particular, the liver is exposed to a greater drug burden. This is a major
limiting factor in the use of these estrogenic products.
Based on the previous pharmacokinetic (Chapters 5 & 6) and
pharmacodynamic (Chapter 7) observations and hence, the therapeutic
potential of the E2-CDS, we evaluated the effects of E2-CDS (0.5 mg/kg b.w.) on
androgen and androgen-responsive tissues in the present study. This dose of
the E2-CDS was chosen because a preliminary dose-response and time-course
study indicated that 0.5 mg/kg is more effective than 1.0 mg/kg dose in
suppressing androgen and androgen-responsive tissues in intact male rats.
Materials and Methods
Adult male Charles River (CD) rats (aged 3-4 months) used in this
study were randomly divided into 9 experimental groups (7-8 rats per group).
Two groups of rats were castrated while 7 other groups remained intact.
CAST was performed by an abdominal incision under metofane anesthesia.
Since surgical stress has previously been found not to affect the experimental
parameters tested here, intact male rats used here were not subjected to sham
operation.


of E2-CDS) in various biological tissues. This method utilized the following
steps: (1) selective solvent extraction of E2-Q+ and E2 from the tissues; (2) base-
catalyzed hydrolysis of E2-Q+ to E2 in NaOH; (3) solid-phase purification of E2
with Ci8 reversed-phase extraction columns; and (4) radioimmunoassay of E2.
Subsequently, the in vivo tissue distributions of E2-Q+ and E2 were
determined in both male and female rats. The results revealed that the
disappearance of E2-Q+ as well as E2 was slow in brain tissue with a tj /2 = 8-9
days. By contrast, both of these metabolites exhibited relatively rapid
clearance from the plasma, liver, lung, kidney, heart, fat, and uterus.
After documenting the kinetic behaviors of E2-CDS, time-course
studies were then conducted to assess the dynamic effects of E2-CDS on
responses which are known to be affected by E2. The E2-CDS consistently
exhibited prolonged and sustained suppression of pituitary gonadotropins
secretion, i.e. LH and FSH in a dose- and time-dependent manner.
Finally, the therapeutic potentials of E2-CDS were investigated in male
and female rats. Studies in the male rat demonstrated that E2-CDS is as
effective as castration in both suppressing the plasma testosterone levels and
reducing the weights of androgen-responsive tissues. Further studies in the
female rat, examining the effects of E2-CDS on tail-skin temperature (TST)
responses, revealed that E2-CDS can significantly attenuate the rise in TST.
Collectively, the results of these studies are consistent with the
proposed mechanism of this drug delivery system, that is, the preferential
retention of E2-Q+ by the brain, and the subsequent slow release of E2 locally in
that tissue. Furthermore, the profound pharmacodynamic effects of this
delivery system support the view that E2-CDS may be potentially useful for
fertility regulation, the effective treatments of prostatic cancer, and certain
brain-mediated estrogen withdrawal symptoms, i.e. menopausal hot flushes.
xii


157
providing the basis for "locking" the intermediate metabolite, E2-Q+, behind
the BBB, there it serves as a brain depot for E2. From this store of E2-Q+, E2 is
slowly released through non-specific hydrolysis of the carrier, resulting in
sustained brain exposure to E2.
Estrogen treatment of OVX morphine-dependent rats with both the E2-
CDS and E2 pellet resulted in significant suppression of gonadotropins (LH
and FSH) secretion. These effects of E2-CDS on gonadotropins are consistent
with the previously reported pharmacodynamic behaviors of the delivery
system (Anderson et al., 1987a,b, 1988a,b, 1989; Sarkar et al., 1989; Simpkins et
al., 1986,1988,1989a,b). The fact that the magnitude of LH suppression was
greater with E2-CDS treatment in the face of lower plasma E2 concentrations,
indicate that the prolonged and sustained inhibitory effects of E2-CDS are due
primarily to sustained suppression of LHRH secretion from the
hypothalamus (Sarkar et al., 1989). Sarkar et al. (1989) have reported
reduction in LHRH release into the hypophyseal portal system and no change
in pituitary responsiveness to LHRH following E2-CDS treatment.
An important question is whether morphine withdrawal in the rat is
analogous to the postmenopausal flushes. The remarkable similarities
between the symptoms of opiate withdrawal and the postmenopausal
syndrome (Simpkins et al., 1983) suggest a common underlying neuronal
mechanism(s) mediating these changes in both the addicted rats and the
postmenopausal women (Casper & Yen, 1985; Simpkins & Katovich, 1984).
Our observations that E2-CDS treatment in some animals (50% or less) did
not completely stabilize these underlying mechanism(s) in order to prevent
the flush response may argue against this animal model. This heterogeneity
in response to estrogen replacement therapy, however, is observed in
postmenopausal women as well. That is, first of all, the intensity and


123
The marked increases in anterior pituitary weights of OVX rats treated
with the E2-CDS appears to be due to the direct effects of E2 on the pituitary
gland. These effects of E2 appear to be exerted on the lactotroph population of
the anterior pituitary (Chen & Meites, 1970; Gorski, 1981). E2 is well known to
stimulate PRL secretion and to induce hyperplasia of lactotrophs (Chen &
Meites, 1970; Gorski, 1981). As indicated above, brain E2 likely reaches the
anterior pituitary gland, through the redistribution of the steroid down the
marked concentration gradient from the brain to the pituitary gland
(Traystman, 1983). It should be noted, however, that the effects of E2-CDS on
pituitary weight are dependent upon the OVX condition of the rats. In gonad-
intact rats, E2-CDS does not alter anterior pituitary weight.
The uterotrophic effects of E2-CDS were also dose- and time-dependent.
This effect of E2-CDS likely relates to the extreme sensitivity of OVX rats to
circulating estrogens (Mayer et al., 1960). Thus, even modest elevations in
plasma E2 following administration of E2 or E2-CDS (Chapter 6), result in
stimulation of uterine tissue in OVX rats. However, both the uterus and the
anterior pituitary gland of gonad-intact rats are unresponsive to the estrogen
delivery system (Anderson et al., 1988a). Finally, it should be noted that the
uterine weights observed following E2-CDS were considerably lower than the
500 to 625 mg weights normally seen in gonad-intact rats (Anderson et al.,
1988a).
In conclusion, the prolonged effects of the E2-CDS on gonadotropins
suppression were dose- and time-dependent, and the duration of these
responses are consistent with the long half-lives of the E2-CDS metabolites in
the brain. These results further support the view that the E2-CDS may be
potentially useful in fertility regulation and treatment of brain E2 deficiencies
(i.e. vasomotor hot flushes).


Finally, very special thanks go to my parents who have always inspired
me to pursue an academic career, and to my supportive and considerate
family (my wife & my son) who always managed to create an environment in
which I could devote the many years required to accomplish this work.
iv


11
mechanism involves a receptor-mediated transcytosis of carrier-bound
hormone complex in the limiting membrane of the organ microcirculation.
However, other investigators have argued against this hypothesis on the
grounds that the concept does not readily reconcile with physiological
findings (Mendel et al., 1988). That is, the rate of protein-bound hormone
dissociation (K hormone hypothesis. Thus, a more comprehensive model (equation) which
formally takes into account the rate-limiting effects of protein-bound
hormone dissociation is more relevant to the experimental observations.
The current Pardridge's model (the protein-bound hormone hypothesis) is,
however, deficient in this respect.
Metabolism and excretion
The plasma concentration of E2 at any time represents the net
difference between the rate of E2 secretion and the rate of metabolism in the
liver and excretion by the kidneys. There is no apparent limit to the capacity
of these organs to metabolize and excrete the E2 hormone. The liver is the
primary organ for metabolizing E2 hormone (Bolt, 1979). The rate of
turnover of E2 hormone is rather rapid. It has a half-life of about 90 min (von
Schoultz et al., 1989). The estrogen is oxidized by the action of a stereospecific
dehydrogenase enzyme, using pyridine nucleotides as cofactors, to less active
products such as estrone and estriol. The oxidized metabolites are then
conjugated as sulfates or glucuronides. The conjugation process renders these
metabolites highly water soluble which then are quickly excreted in the urine
or bile (Bolt, 1979). These conjugates are biologically inactive; however, the
biliary metabolites may undergo further metabolism by action of the
intestinal flora. The products are then reabsorbed into the portal circulation


CHAPTER 8
EFFECTS OF THE E2-CDS OR CASTRATION ON ANDROGEN AND
ANDROGEN-DEPENDENT TISSUES IN MALE RATS
Introduction
Prostate carcinoma is the second most common cancer in males, and its
incidence increases rapidly over the age of 50 (Silverberg & Lubera, 1986).
Likewise, the development of benign prostatic hyperplasia also increases in
incidence with age; however, there is no direct evidence that benign prostatic
hyperplasia is a necessary prerequisite in the development of prostatic cancer
(Berry et al., 1984; Rotkin, 1983). Despite the considerable clinical and
scientific advances in the diagnosis, prognostication, and treatment, these
prostatic diseases remain the second leading cause of morbidity and/or
mortality in senescent males (Santen & English, 1989). Although
epidemiological findings and sporadic reports of a familial occurrence of
prostatic cancer suggest a genetic predisposition (Silverberg & Lubera, 1986),
there is currently little evidence that this is a significant factor in most
patients. A variety of other theories including neoplastic and endocrine
metabolic factors have been proposed to explain the etiology and occurrence
of prostatic adenocarcinoma and/or benign prostatic hyperplasia in men.
Certainly, it is clear that the two major factors necessary for the genesis in
men of these prostatic diseases are the presence of the testis and aging
(Huggins & Hodges, 1941). Studies that attempt to correlate the incidence of
the disease with age-related changes in testosterone levels or other endocrine
factors have shown no consistent relationship as yet. However, there are
128


154
11). The naloxone-induced surge in TST returned to normal range at the end
of 70 to 90 min of study.
The area under the TST curve (AUC) was significantly reduced by 42%
or 46% with multiple injections of E2-CDS compared with the HPCD control
or the E2 pellet, respectively. Likewise, treatment with a single injection of
E2-CDS reduced the AUC by 13% or 20% compared to the HPCD group or the
E2 pellet, respectively, however these effects were not statistically significant.
The mean maximal decline in RT was 2.3 0.3, 3.1 0.4, 3.3 0.3, and
2.5 0.4C in the HPCD, E2 pellet, multiple E2-CDS, or single E2-CDS groups,
respectively (Figure 25). Estrogen treatment with E2-CDS or E2 pellet had no
significant effect on Rt. The naloxone-induced decline in RT returned to
normal at 3 to 4 hrs after naloxone administration (data not shown).
Plasma E2 concentrations were significantly elevated with the E2*pellet
as well as with the multiple E2-CDS treatment (Table 12). However, the E2-
pellet treatment produced plasma E2 concentrations which were significantly
higher (2- to 6- fold) than those produced by single or multiple E2-CDS
treatment (Table 12).
Plasma gonadotropin concentrations (LH and FSH) were significantly
suppressed relative to HPCD control with the E2-pellet as well as the single
and multiple E2-CDS treatment (Table 12).
Plasma PRL concentrations in animals treated with either the E2 pellet
or the multiple E2-CDS were significantly elevated relative to PRL levels of
the HPCD group or the single-injected E2-CDS group. Furthermore, the
magnitude of PRL stimulation with the E2-pellet treatment was 1.5- and 3.5-
fold greater than that of multiple- or single-injected E2-CDS, respectively
(Table 12).


15
with estrogen hormone on neuronal membrane excitability (Garda-Segura et
al., 1987,1989; Kelly et al., 1978; Nabekura et al., 1986; Smith et al., 1987) occur
much too rapidly to be accounted for by new mRNA synthesis and translation
into proteins. This has led some investigators to suggest a possible action of
gonadal steroids directly on neuronal membrane function/components.
Moreover, Pietras and Szego (1979) have reported an increase in cyclic AMP
concentrations in uterus within 15 minutes after E2 treatment. Although
actinomydn D (an RNA synthesis inhibitor) can effectively prevent the full
expression of long-term E2 effects on target tissues, the rapid or short-term
effects of E2 seem to be independent of RNA/protein synthesis. These effects
are most likely mediated by membrane-associated E2 receptors (Pietras &
Szego, 1979; Towle & Sze, 1983).
Estrogen Receptors
Intracellular/cytosolic receptors
The pioneering studies of Glascock and Hoekstra (1959) and of Jensen
and Jacobson (1962) utilizing radiolabeled estrogen demonstrated selective
localization and retention of the label in tissues known to be targets for
estrogen action. Jensen and Jacobsen (1962) in their studies also demonstrated
E2 binding to a spedfic cytosolic receptor protein. The application of sucrose-
density gradient centrifugation then led to further characterization of the
cytosolic estrogen receptor (Toft & Gorski, 1966). Subsequent studies provided
further evidence to satisfy the criteria for an E2 protein receptor. These
criteria included the stereospecifidty for E2 binding (Noteboom & Gorski,
1965), saturable or limited number of binding sites (Gorski et al., 1968;
Noteboom & Gorski, 1965), size determination by gel filtration


48
Morphine and Naloxone Treatment
In experiments which examined the effects of E2-CDS on the tail-skin
temperature (Chapter 9), animals were addicted to morphine. Morphine
dependency was produced by sc implantation behind the neck region of one
pellet containing 75 mg morphine free base. Two days after the first
morphine pellet, two additional morphine pellets were sc implanted. This
regimen of morphine treatment has been utilized in our laboratory to
consistently produce typical symptoms of morphine dependency and
tolerance (Simpkins et al., 1983, 1984) as measured by several tests of analgesia
and withdrawal (Gibson &Tingstad, 1970; Simpkins et al, 1983, 1984). These
morphine pellets produce serum morphine concentrations of 300 ng/ml by
one hr after implantation and remain elevated at this level through 48 hrs
(Derendorf & Kaltenbach, 1986). Thus, the sustained release of morphine
achieved by the pellet is presumed to produce persistent stimulation of opiate
receptors utilizing our treatment regimen.
Naloxone HC1 from Dupont Pharmaceuticals (Garden City, NJ) was
dissolved in saline and administered (0.5 mg/kg b.w.) subcutaneously.
Plasma Hormone Radioimmunoassays
Protein Hormone Assays
Plasma luteinizing hormone (LH), follicle-stimulating hormone (FSH),
growth hormone (GH), and prolactin (PRL) concentrations were measured in
duplicate by radioimmunoassay (RIA) using NIADDK kits provided by the
National Hormone and Pituitary Program (Baltimore, MD). Plasma LH, FSH,


79
in concentrations of greater than 83% and 87% from 1 to 7 days and 93% to
98% by 14 days after drug administration in liver and fat, respectively.
Similarly, E2 concentrations in these tissues fell by more than 84% and
80% from day 1 to day 7, and by 14 days after drug administration, the E2
concentrations decreased by 95% and 90%, respectively.
Figure 8 shows the time-concentration profiles in kidney, heart, lung,
and anterior pituitary concentrations of E2-Q+ (upper panel) and E2 (lower
panel) at various times following administration of the E2-CDS. The E2-Q+
concentrations in these tissues initially increased to 1906 131,1047 106, 748
28, and 407.6 50.6 ng/g in heart, lung, kidney, and anterior pituitary,
respectively. These E2-Q+ levels decreased rapidly by more than 76%, 79%,
74%, and 80% by day 7 in these 4 tissues, respectively. By 14 days after drug
administration, E2-Q+ concentrations decreased by greater than 98% in heart
and lung, 96% in kidney, and 93% in anterior pituitary. Despite high initial
concentrations of E2-Q+ in these peripheral tissues, brain levels of E2-Q+
exceeded E2-Q+ levels of these tissues by 1.5- to 3-fold at 14 days after
administration of the E2-CDS.
Estradiol concentrations in heart, lung, kidney, and anterior pituitary
(Figure 8; lower panel) were similarly elevated on day 1 but decreased rapidly
by 67% in heart, 83% in lung, 81% in kidney, and 86% in anterior pituitary by
day 7. From day 1 to day 14, the E2 levels in these tissues decreased by more
than 95% of the initial concentrations.
Figure 9 depicts brain (upper panels) and anterior pituitary (lower
panels) contents of E2-Q+ and E2 at various times after administration of the
E2-CDS. Following a single injection of the E2-CDS, the brain E2-Q+ content
was 635 28,340 23, and 77 3.9 ng/brain at 1,7 and 14 days, respectively.


117
expressed as ng/ml of either the LH-RP-2, FSH-RP-2, or GH-RP-2 reference
standard, respectively and PRL values are expressed as ng/ml of the PRL-RP-3
standard. The intra-assay coefficients of variation were 4.67%, 5.02%, 4.05%,
and 4.96% for LH, FSH, GH, and PRL assays, respectively. All the samples for
each hormone were assayed in a single run.
Results
The E2-CDS caused a dose- and time-dependent suppression of plasma
LH throughout the time-course studied (Figure 16). The maximum LH
reduction occurred at 7 days postinjection. At this time, LH was suppressed by
21, 46 and 86% relative to HPCD control at doses of 0.01, 0.1 and 1.0 mg E2-
CDS/kg, respectively (Figure 16). The plasma LH concentrations in animals
treated with 1.0 mg E2-CDS were significantly reduced by 56, 86, 72, and 56% at
1, 7,14, or 21 days, respectively and remained suppressed by greater than 35%
at 28 days after drug administration. By contrast, equimolar E2 dose (0.7
mg/kg) caused a transient reduction in LH concentrations of 27% on day 1
and 24% on day 7 which were not significantly different from time 0 values
(Figure 16).
Similarly, the E2-CDS caused a dose- and time-dependent suppression
of plasma FSH throughout the time-course studied (Figure 17). The
maximum FSH reduction occurred at 7 days postinjection. FSH was
suppressed by 14, 28, and 58% relative to control at doses of 0.01, 0.1 and 1.0
mg E2-CDS/kg, respectively, on day 7 (Figure 17). The plasma FSH
concentrations in animals treated with 1.0 mg E2-CDS were significantly
reduced by 37,58 and 20% at 1,7 or 14 days, respectively and by 7% (day 21) or
were at preinjection levels by 28 days after drug administration. By contrast,


29
A strategy that could achieve an improved delivery of drugs to the
brain with sustained release in that tissue would be of great advantage. A
general approach to increase brain concentration of drugs and thus, their
effectiveness has been the design of lipid soluble prodrug from water soluble
drugs (Bodor, 1981, 1985; Bodor & Kaminski, 1987; Sinkula & Yalkowsky, 1975;
Stella, 1975). Prodrugs are pharmacological agents which have been
transiently modified to improve their lipophilicity as well as to hinder their
rapid metabolic inactivation via enzymes. Ideally, the prodrug is biologically
inactive but reverts to the active, parent drug in vivo at, or around, the site of
action. This transformation can be mediated by an enzyme or may occur
chemically as a result of designed instability in the structure of the prodrug.
The purpose of prodrug modification is to increase the concentration of the
active drug at or near its site of action, thereby increasing its potency/efficacy.
By temporary masking the polar groups of a drug, the lipophilicity of the drug
is increased and thus, its ability to enter the brain parenchyma is enhanced.
Once in the brain, hydrolysis of the masking groups will release the active
drug.
Nevertheless, potential problems are associated with the prodrug
approach (Gorrod, 1980). For instance, by increasing the lipophilicity of a drug
via the prodrug approach, it may not only improve its diffusion through the
BBB to gain access to the CNS, but also ensures the uptake of the compound
into all other tissues and thus, exposure to a greater drug burden. This is a
major limiting factor in the use of prodrugs, especially those with cytotoxicity,
i.e. antineoplastic agents, or those with broad spectrum of peripheral site of
actions such as steroids. Furthermore, even if enhanced CNS
delivery/uptake is achieved via the prodrug approach, the efflux of the drug


96
Experiment 2
Figure 15 depicts the concentration-time profiles of the E2-CDS
metabolites (E2-Q+, upper panel; E2, lower panel) in plasma of OVX rats. By 30
min after administration of the E2-CDS, plasma E2-Q+ increased to (61.9 3.8
ng/ml). The E2*Q+ levels decreased by 50% at 8 hrs and by greater than 88% at
24 hrs after E2-CDS treatment. Kinetic analysis revealed that the plasma E2-Q+
concentration-time profile fits a sum of two exponentials. The half-lives of
these two phases were ti/2 = 8.16 and ti/2 = 70.38 hrs respectively in plasma.
The area under curve (AUQ) was 835.25 ng/ml hr for the time-course
studied.
Similarly, plasma E2 concentrations were increased to 1.9 0.08 ng/ml
after 30 min of E2-CDS administration. The plasma E2 levels decreased by
50% at 3 hrs and by greater than 91% at 24 hrs after E2-CDS treatment. The
plasma E2 concentration-time profile was best fitted in a sum of three
exponentials. The half-lives of the three phases were ti/2 = 0.14, ti/2 = 2.35,
ti/2 = 38.99 hrs, respectively. The area under curve (AUQ) was 23.85 ng/ml
hr for the time-course studied. The third half-life indicated the presence of
deep compartment (most likely the brain) which slowly releases E2 in plasma.
Discussion
This detailed dose-distribution and time-course study demonstrates
that (i) the enzymatic oxidation of E2-CDS to E2-Q+ is dose dependent, and
with the possible exception of the uterus, the oxidation is not saturable over
the 100-fold dose range tested; (ii) the hydrolysis of E2-Q+ to E2 was dependent
upon the tissues analyzed and appeared to be saturable only in plasma and fat


178
Bottiger L.E., G. Boman, and G. Eklund. Oral contraceptives and
thermoboembolic disease: Effects of lowering oestrogen content.
Lancet 1:1097-1101 (1980).
Bowman K.M. and L. Bender. The treatment of involution melancholia with
ovarian hormone. Am. T. Psvchiat. 11:867-893 (1932).
Brawer J.R., H. Schipper, and F. Naftolin. Ovary-dependent degeneration in
the hypothalamic arcuate nucleus. Endocrinology 107:274-279 (1980).
Brawer J.R., H. Schipper, and B. Robaire. Effects of long term androgen and
estradiol exposure on the hypothalamus. Endocrinology 112:194-199
(1983).
Brendler C.B. The current role of hormone therapy in the clinical treatment
of prostatic cancer. Seminars in Urol. 6:269-278 (1988).
Brewster M.E., K.S. Estes, and N. Bodor. Improved delivery through
biological membranes 32. Synthsis and biological activity of brain-
targeted delivery system for various estradiol derivatives. T. Med.
Chem. 31:244-249 (1987).
Briggs M. Biochemical effects of oral contraceptives. Adv. Steroid Biochem.
Pharmacol. 5:66-160(1976).
Brightman M.W. Morphology of blood-brain interfaces. Expt. Eye Res.
{suppl.] 25:1-25 (1977).
Brightman M.W. and T.S. Reese. Junctions between intimately opposed cell
membranes in the vertebrate brain. T. Cell Biol. 40:648-677 (1969).
Broadwell R.D. and M. Salcman. Expanding the definition of the blood-brain
barrier to protein. Proc. Natl. Acad. Sri. USA 78:7820-7824 (1981).
Bunney W.E., Jr., and B.L. Garland. A second generation catecholamine
hypothesis. Pharmacopsychiat. 15:111-115 (1982).
Burkman R.T. Lipid and lipoprotein changes in relation to oral contraception
and hormone replacement therapy. Fertility and Sterility [Suppl.]
49:39S-50S (1988).
Butcher R.L., W.E. Collins, and N.W. Fugo. Plasma concentration of LH,
FSH, prolactin, progesterone and estradiol-17(J throughout the 4-day
estrous cycle of the rat. Endocrinology 94:1704-1708 (1974).


120
metabolite of the E2-CDS, E2-Q+, is "locked" behind the BBB and there it
serves as a brain depot for E2 (Bodor, et al., 1987). From this store of E2-Q+, E2
is then slowly released through non-specific hydrolysis of the carrier,
resulting in sustained brain exposure to E2.
Since 17-substituted estrogens, such as the E2-CDS and E2-Q+, do not
effectively bind to E2 receptors (Dusterberg & Nishino, 1982; Janoko et al.,
1984), they are not likely to exhibit estrogenic activity. Thus, it is reasonable to
believe that neither the E2-CDS nor the E2-Q+ formed in the brain account for
the prolonged pharmacological effects of this delivery system. Rather locally
released E2 in the brain, particularly the hypothalamus, would appear to
account for the sustained suppression of the gonadotropin secretion.
Our previous evaluation of tissue distribution of the E2-CDS in male
rats (Chapter 5; Rahimy et al., 1990a) and the more detailed dose-response and
time-course evaluation of the E2-CDS distribution in OVX rats (Chapter 6;
Rahimy et al., 1990b) revealed that (i) E2-Q+, the quaternary form of E2-CDS, as
well as E2 persists in the brain with ti/2 = 8-9 days and (ii) the same
metabolites are rapidly eliminated from the peripheral tissues. These
findings together with the absence of a physiologically significant elevation of
plasma E2 concentrations from 7-28 days after the E2-CDS administration
(Chapter 6), provide strong evidence for the local action of E2 in the brain,
presumably on hypothalamic luteinizing hormone-releasing hormone
(LHRH) containing neurons (Sarkar et al., 1989).
The synthesis and secretion of gonadotropins from the anterior
pituitary are differentially regulated by several neuronal (Barraclough &
Wise, 1982; Dalkin et al., 1989; Marshal & Kelch, 1986; Plant, 1986) and
hormonal (Kalra & Kalra, 1980, 1983) factors including the hypothalamic
decapeptide, LHRH, and the action of E2 both in a positive and negative


3. GENERAL MATERIALS AND METHODS 44
Drugs and Solutions 44
Estradiol and Standard Solution 44
Estradiol-Chemical Delivery System 44
Estradiol Pellet 45
Morphine Pellets 45
Animals 46
Drug Treatment 47
Steroid Treatment 47
Morphine and Naloxone Treatment 48
Plasma Hormones Radioimmunoassays 48
Protein Hormone Assays 48
Steroid Hormone Assays 49
Statistical Analysis 50
4. DEVELOPMENT OF AN ANALYTICAL METHOD FOR THE
QUANTITATION OF E2-CDS METABOLITES IN A WIDE
VARIETY OF TISSUES IN THE RAT 51
Introduction 51
Materials and Methods 52
In Vitro Methodology 53
Specificity of the estradiol antibody for E2 53
Selective solvent extraction of steroids from tissues 53
Hydrolysis of E2-Q+ in various tissue extracts 55
Solid-phase extraction and separation of E2 by Cis columns 55
Radioimmunoassay of E2 56
Calculations 56
In Vivo Studies 57
Results 58
In Vitro Methodology 58
Cross-reactivity of the E2 antibody with E2-Q+ 58
Recovery of E2 58
Precision of the E2 extraction-assay method 59
Recovery of E2-Q+ 60
Precision of the E2-Q+ extraction-assay method 60
Distribution of E2 and E2-Q+ in vivo 61
Discussion 62
5. DETERMINATION OF THE TISSUE DISTRIBUTION OF E2-CDS
METABOLITES IN MALE RATS 75
Introduction 75
Materials and Methods 77
Results 78
Discussion 80
VI


136
treatment strategy largely relates to minimizing toxicity while optimizing the
response rate as well as the duration of benefit.
In the present study, we described a newer endocrine approach in the
treatment of androgen-dependent prostatic diseases. The evaluation of E2-
CDS effects on the normal androgen-responsive sex accessory organs with
comparison to that of CAST demonstrated that: 1) the E2-CDS at a single dose
of 0.5 mg/kg was as effective as CAST in suppressing significantly the plasma
T levels for 14 days after treatment; 2) the E2-CDS treatment of intact male rats
resulted in significant regression of the androgen-sensitive ventral prostate as
well as seminal vesicle weight equivalent in magnitude to that of CAST
alone; 3) interestingly, both the profound suppression of T levels and the
prolonged duration of tissue regression, at 14 days after the final treatment,
were observed even in the face of low plasma E2 levels. Furthermore, these
data suggest that the primary site of action where E2 exerts its effects leading to
T suppression is in the central nervous system. Previously, we have reported
long-term suppression of LH in CAST male rats (Simpkins et al., 1986),
stimulation of masculine sexual behavior in CAST male rats for 28 days
(Anderson et al., 1987a), and long-term gonadotropin suppression in
ovariectomized rats (Chapter 7) following a single iv administration of the E2-
CDS. These prolonged pharmacological effects of the E2-CDS are consistent
with the observations of the accumulation of E2-Q+, the oxidized form of the
delivery system, in the brain with an apparent t^ ¡2 = 8-9 days in that tissue
(Chapters 5 & 6). From this store of E2-Q+, E2 is slowly released through non
specific hydrolysis of the carrier, resulting is sustained brain exposure to E2.
The production of T by the Leydig cells of the testis are controlled via a
negative feedback mechanism (Swerdloff, 1986). Increased levels of T exert a
negative feedback on both the hypothalamus and the anterior pituitary, thus


182
Fedor-Freyberg P. The influence of oestrogen on the well-being and mental
performance in climacteric and postmenopausal women. Acta Qbstet.
Gynaecol. Scand. 64:5-69 (1977).
Fenstermacher J.D. Current models of blood-brain transfer. Trends Neuro.
Sd. 8:449-452(1985).
Ferin M., D. van Vught, and S. Wardlow. The hypothalamic control of the
menstrual cycle and the role of endogenous opioid peptides. Recent
Prog. Horm. Res. 40:441-486 (1984).
Fillit H., H. Weinreb, I. Cholst, V. Luine, B. McEwen, R. Amador, and J.
Zabriskie. Observations in a preliminary open trial of estradiol therapy
for senile dementia-Alzheimer's type. Psychoneuroendocrinologv
11:337-345 (1986).
Firsch I.R. and J. Frank. Oral contraceptives and blood pressure. TAMA
237:2499-2503 (1977).
Foote J.E. and E.D. Crawford. Total androgen suppression: Are there any
advantages? The use of combined therapy in the treatment of
advanced prostate adenocarcinoma. Sem. Urol. 6:291-302 (1988).
Fotherby K. Oral contraceptives, lipids and cardiovascular disease.
Contraception 31:367-394(1985).
Fraenkel L. Die funktion des corpus luteam. Arch. Gvnaekol. 68:483-545
(1903).
Furuhjelm M. and P. Fedor-Freyberg. The influence of estrogens on the
psyche in climacteric and postmenopausal women. In: Consensus on
Menopausal Research. P.A. van Keep, R.B. Greenblatt, M. Albeaux-
Femet (eds) 84-93, MTP Press, London (1976).
Gambone J., D.R. Meldrum, L. Laufer, R.J. Chang, J.K.H. Lu, and H.L. Judd.
Further delineation of hypothalamic dysfunction responsible for
menopausal hot flashes. T. Clin. Endocrinol. Metab. 59:1097-1102
(1984).
Garda-Segura L.M., G. Olmos, R.J. Robbins, P. Hernandez, J.H. Meyer, and F.
Naftolin. Estradiol induces rapid remodelling of plasma membranes
in developing rat cerebrocortical neurons in cuture. Brain Res.
498:339-343 (1989).


184
Hackman B.W. and D. Galbraith. Replacement therapy with piperazine
oestrone sulfate ("Harmogen") and its effect on memory. Current Med.
Res. Opin. 4:303-306 (1976).
Halasz B. and R.A. Gorski. Gonadotrophic secretion in female rats after
partial or total interuption of neural afferents to the medial basal
hypothalamus. Endocrinology 80:608-622 (1967).
Hamilton T.H. Control by estrogen of genetic transcription and translation.
Science 161:649-661 (1968).
Hammond D.L. Intrathecal administration: Methodological considerations.
Prog. Brain Res. 77:313-320 (1988).
Hardebo J.E. and C. Owman. Barrier mechanisms for neurotransmitter
monoamines and their precursors at the blood-brain interface. Ann.
Neurol. 8:1-11 (1980).
Hawkinson L.F. The menopausal syndrome. One thousand consecutive
patients treated with estrogen. TAMA 111:390-393 (1938).
Henderson S.R., C. Baker, and G. Fink. Oestradiol-17|3 and pituitary
responsiveness to luteinizing hormone releasing factor in the rat: A
study using rectangular pulses of oestradiol-17P monitored by non
chromatographic radioimmunoassay. T. Endocrinol. 73:441-453 (1977).
Henriksson P. and O. Edhag. Orchidectomy versus oestrogen for prostatic
cancer: Cardiovascular effects. Br. Med T. 293:413-415 (1986).
Hoek J.B., and J. Rydstrom. Physiological roles of nicotinamide nucleotide
transhydrogenase. Biochem. T. 254:1-10 (1988).
Hoffmann B. Use of radioimmunoassay procedures for the determination of
sex hormones in animal tissues. T. Steroid Biochem. 19:947-951 (1983).
Holzbauer M. and M.B.H. Youdim. The oestrous cycle and monoamine
oxidase activity. Br. T. Pharmacol. 48:600-605 (1973).
Honjo H., Y. Ogino, K. Naitoh, M. Urabe, J. Kitawaki, J. Yasuda, T. Yamamoto,
S. Ishihara, H. Okada, T. Yonezawa, K. Hayashi, and T. Nambara. In
vivo effects by estrone sulfate on the central nervous systemsenile
dementia (Alzheimers type). T. Steroid Biochem. 34:521-525 (1989).
Huggins C. and C.V. Hodges. Studies on prostatic cancer. I. The effect of
castration, of estrogen and of androgen injection on serum


18
single injection of E2, the translocation of cytosolic estrogen receptor (ERc) to
the nucleus has been reported to be nearly complete within 1 hr in the rat
uterus (Jakesz et al., 1983). When ERc was assessed by exchange assay 6 hrs
after hormone administration, ERc levels continued to remain very low.
However, an increase in nuclear estrogen receptor (ERn) was concomitantly
observed following E2 injection, reaching maximal levels after 1 hr (Jakesz et
al., 1983). Following an apparently near quantitative translocation of ERc to
the nucleus, ERn concentrations declined to ~ 30% after 6 hrs. However, with
repeated injections of estrogen which maintained continuous receptor
saturating concentrations of [3H]E2 over a 6-hr period, conservation of total
cellular receptors in both cytoplasmic and nuclear fractions were observed
(Jakesz et al., 1983). It is important, however, to note that under continuous
steroid exposure qualitative changes in receptor properties (down regulation)
occur over time in both cytosol and nuclear compartments. It is thought that
the ERc present at 6 hr after estrogen administration originate from a
replenished pool of receptors. This replenished pool has been reported to be
partially dependent on protein synthesis. However, inhibition of protein
synthesis by cycloheximide did not inhibit replenishment after estrogen
exposure. Thus, estrogen target tissues, particularly uterus may represent a
system in which estrogen receptors replenishment appears to be due entirely
to receptor recycling (Jakesz et al., 1983).
Determination of the kinetic binding parameters indicated a high
affinity estrogen-binding site (Kd = ~ 10-10 M) for brain ERc and a Bmax of ~ 3
fmol/mg protein (Walters, 1985). These binding parameters are similar to
estrogen binding kinetics of other tissues. Rate of association with the
receptor has been reported to be 4.4 x 105 M-l S-l while rate of dissociation was
2.4 x 105 M-l S-l with ti/2 = 80 hrs at 0 to 4C. Half-life for the clearance of


17
some target tissues may be caused by direct interaction with plasma
membrane receptor/effector components (Becker, 1990; Garcia-Segura et al.,
1987,1989; Kelly et al., 1978; Majewska, 1987; Nabekura et al., 1986; Smith et
al., 1987; Towle & Sze, 1983). In fact, biochemical studies have demonstrated
specific binding sites for sex steroids in synaptosomal plasma membranes
prepared from the rat brain (Towle & Sze, 1983). Furthermore, the presence
of steroid binding sites have also been demonstrated on plasma membranes
of other target tissues as well, including liver (Suyemitsu & Terrayama, 1975),
pituitary (Koch et al., 1977), and uterus (Pietras & Szego, 1979). In all these
instances, the exact physiological/pharmacological function of the membrane
binding sites for steroids has yet to be determined. However, these binding
observations are compatible with the rapid non-genomic effects of estrogen,
which are not easily accommodated within the genomic model (McEwen et
al., 1982,1984; Majewska, 1987). Collectively, the presence, if real, of these
speculative membrane receptors can account for the rapid neurotropic effects
of E2- Furthermore, these receptors may be involved in the modification of
CNS neurotransmission.
Estrogen-receptor binding kinetics
Estrogen target tissues, i.e., brain, anterior pituitary, uterus, etc.,
apparently contain a single, specific estrogen binding component (type I
estrogen receptor). The unoccupied receptors migrate between both the
cytoplasmic and nuclear components (Walters, 1985). The cytosolic receptors
bind E2 with high affinity such that the steroid-receptor complex remains
intact for translocation into the nucleus (Towle & Sze, 1983). In the rat
uterus, estrogen-filled binding sites do not undergo detectable degradation
over a 24-hr period at temperature up to 30C (Walters, 1985). Following a


81
is rapidly cleared from the periphery. This is in accordance with that reported
in other studies (Mullersman et al., 1988). The half-life of the lipophilic E2-
CDS in brain tissue is only 29.2 min (Bodor et al., 1987), indicating rapid
oxidation of the delivery system to E2-Q+. From this store of E2-Q+, E2 can be
slowly released chronically in the brain through nonspecific hydrolysis.
As predicted from the physicochemical properties of E2-CDS as well as
previous reports on brain levels of E2-Q+ (Boder et al., 1987; Mullersman et
al., 1988; Simpkins et al., 1986), this "locked-in" form of the E2-CDS reached a
52-fold higher concentration in the brain than plasma by day 1 after treatment
with the E2-CDS, and these brain-blood ratios increased to 132-fold at day 7
and to about 170-fold by 14 days. Furthermore, from day 1 through 14, the
content of E2-Q+ in the brain was 6- to 23-times the content of E2-Q+ in the
blood. Thus, a portion of the E2-Q+ found in plasma may arise from brain
stores of the compounds. E2-Q+ can be cleared from the brain by bulk flow of
cerebrospinal fluid (Boder & Brewster, 1983; Schanker, 1965).
Brain E2 concentrations were similarly elevated relative to plasma.
Estradiol achieved a 28-fold higher concentration in the brain than plasma by
day 1 and this ratio increased to more than 50-fold at day 7 and remained at
37-fold by 14 days. Additionally, throughout the time-course studied, the
brain E2 content was 3- to 6-times the content of E2 in the blood. These
observations indicated that brain E2 is continuously produced, and as such the
steady-state brain E2 concentration is dependent on its rate of production from
the E2-Q+ and its rate of elimination from the brain by local metabolism
and/or redistribution down a concentration gradient to the plasma. Brain
stores of E2 could contribute to plasma levels through its partitioning to the
periphery down a large concentration gradient.


185
phosphatases in metastatic carcinoma of the prostate. Cancer Res.
1:293-297 (1941).
Huppert L.C. Hormonal replacement therapy: Benefits, risks, doses. Med.
Clin. North Am. 71:23-39 (1987).
Hurst B.S. and J.A. Rock. Endometriosis: Pathophysiology, diagnosis and
treatment. Obstet. Gynecol. Surv. 44:297-304 (1989).
Imperato-McGinley J. 5 a-Reductase deficien in man. In: Progress in Cancer
Research and Therapy. Vol. 31, F. Bresciani, R.J.B. King, M.E. Lippman,
and M. Namer (eds) 491-496, Raven Press, New York (1984).
Imperato-McGinley J., R.E. Peterson, M. Leshin, J.E. Griffin, G. Cooper, S.
Draghi, M. Berenyi, and J.D. Wilson. Steriod 5 a-reductase deficiency
in a 65 year old male pseudohermaphrodite: The natural history,
ultrastructure of the testis and evidence for inherited enzyme
heterogeneity. T. Clin. Endocrinol. Metab. 50:15-22 (1980).
Inman W.H.W. and M.P. Vessey. Investigation of deaths from pulmonary,
coronary, and cerebral thrombosis and embolism in women of child
bearing age. Br. T. Med. 2:193-199 (1968).
Inman W.H.W., M.P. Vessey, B. Westerholme, and A. Engelund.
Thromboembolic disease and the steroidal content of oral
contraceptives: A report to the committee on safety of drugs. Br. Med.
L 2:203-209 (1970).
Iramain C.A., J.O. Owasoyo, and G.N. Egbunike. Influence of estradiol on
acetylcholinestrase activity in several female rat brain areas and
adenohypophysis. Neurosci. Lett. 16:81-84 (1980).
Isaacs J.T., C.B. Brendler, and P.C. Walsh. Changes in the metabolism of
dihydrotestosterone in the hyperplastic human prostate. T. Clin.
Endocrinol. Metab. 56:139-146 (1983).
Jakesz R., A. Kasid, and M.E. Lippman. Continuous estrogen exposure in the
rat does not induce loss of uterine estrogen receptor. J. Biol. Chem.
258:11798-11806 (1983).
Janocko L., J.M. Larner, and R.B. Hochberg. The interaction of C-17 esters of
estradiol with estrogen receptor. Endocrinology 114:1180-1186 (1984).
Jensen E.V. and E.R. DeSombre. Mechanism of action of the female sex
steroids. Ann. Rev. Biochem. 41:203-230 (1972).


191
Miller W.L. Molecular biology of steroid hormone synthesis. Endocr. Rev.
9:295-318 (1988).
Molnar G.W. Body temperature during menopausal hot flushes. T. Appl.
Physiol. 38:499-503 (1975).
Moore R.A. Benign prostatic hypertrophy and carcinoma of the prostate.
Occuarnce and experimental production in animals. Surgery 16:152-
167 (1944).
Morrel J.I., D.B. Kelly, and D.W. Pfaff. Sex steroid binding in the brains of
vertebrates. In: Brain-Endocrine Interaction II, K.M. Knigge, D.E. Scott,
H. Kobayashi, and S. Ishii (eds) 230-256, Karger, Basel (1975).
Morton J.H., H. additon, R.G. Addison, L. Hunt, and J.J. Sullivan. A clinical
study of premenstrual tension. Am. T. Obstet. Gynecol. 65:1182-1191
(1953).
Moutsatsou V. and R.E. Oakey. Oestradiol binding to plama proteins. J.
Steroid Biochem. 29:319-323 (1988).
Mowles T.F., B. Ashkanazy, E. Mix, and H. Sheppard. Hypothalamic and
hypophyseal estradiol-binding complexes. Endocrinology 89:484-493
(1971).
Mueller G.C., A.M. Herranen, and K.F. Jervell. Studies on the mechanism of
action of estrogen. Recent Prog. Horm. Res. 14:95-139 (1958).
Mullersman G., H. Derendorf, M.E. Brewster, K.S. Estes, and N. Bodor. High-
performance liquid chromatographic assay of a central nervous system
(CNS)-directed estradiol chemical delivery system and its application
after intravenous administration to rats. Pharm. Res. 5:172-177 (1988).
Murad F. and R.C. Haynes, Jr. Estrogens and progestins. In: The
Pharmacological Basis of Therapeutics, A.G. Gilman, L.S. Goodman,
T.W. Rail, and F. Murad (eds) 1412-1439, Macmillan, New York (1985).
Nabekura J., Y. Oomura, T. Minami, Y. Mizuno, and A. Fukuda. Mechanism
of the rapid effect of 17 P-estradiol on medial amygdala neurons.
Science 223:226-228 (1986).
Nesheim B.I. and T. Saetre. Changes in skin blood flow and body
temperature during climacteric hot flushes. Maturitas 4:49-55 (1982).


Plasma LH (ng/ml)
124
Figure 16. Dose and time-dependent effects of the E2-CDS on plasma LH
responses in ovariectomized rats. Animals received a single iv
injection of the E2-CDS on day 0 at doses of 0.01, 0.1 and 1.0 mg/kg
bw. Also, the responses to an E2 dose of 0.7 mg/kg, equimolar to
the 1.0 mg /kg dose of E2-CDS, is shown for day 1 and 7.
Represented are means SEM for n = 7 rats per group per
sampling time. The symbols indicate statistical differences as
follows: *) different from vehicle group (day 0); a) different from
0.01 mg/kg; and b) different from both 0.01 and 0.1 mg/kg. The
significance of interaction between factors (time and dose) was
determined by two-way analysis of variance (ANOVA). The
significance of differences among mean values at each dose level
was determined over time by one-way ANOVA and Dunnett's
test while the significance of differences among mean values of
three dose levels (at each time point) was determined by one-way
ANOVA and Scheffe F-test. The level of probability for all tests
was p<0.05.


196
Ross G.T. Disorders of the ovary and female reproductive tract. In: Textbook
of Endocrinology, J.D. Wilson and D.W. Foster (eds), W. B. Saunders
Co., Philadelphia (1981).
Ross G.T. The ovaries. In: Textbook of Medicine. J.N. Wyngaarden, L.H.
Smith (eds) 1379-1393, Saunders, Philadelphia (1985).
Rossmanith W.G., G.A. Laughlin, J.F. Mortola, and S.S.C. Yen. Secretory
dynamics of oestradiol (E2) and progesterone (P) during periods of
relative pituitary LH quiescence in the midluteal phase of the
menstrual cycle. Clin. Endocrinol. 32:13-23 (1990).
Rotkin I.D. Distribution, and risk of benign prostatic hypertrophy. In:
Benign Prostatic Hypertrophy, F. Hinman, Jr. (ed.) 10-21, Springer-
Verlag, New York (1983).
Santen R.J. and H.F. English. Biological principles underlying treatment
strategies for prostate carcinoma. In: Proceedings of IVth International
Congress of Andrology. M. Serio (ed.) 449-457, Raven Press, Florence,
Italy (1989).
Sarkar D.K., S.J. Friedman, S.S.C. Yen, and S.A. Frautschy. Chronic inhibition
of hypothalamic-pituitary-ovarian axis and body weight gain by brain-
directed delivery of estradiol-17(3 in female rats. Neuroendocrinologv
50:204-210 (1989).
Saumande J. and S.K. Batra. A double antibody radioimmunoassay for free
and conjugated estradiol-17(3 in cow's milk. Steroids 44:137-152 (1984).
Schanker L.S. Passage of drugs into and out of the central nervous system.
Antimicrob. Agents Chemother ,1044-1050 (1965).
Schiess M.C., C.A. Dudley, and R.L. Moss. Estrogen priming affects the
sensitivity of midbrain central gray neurons to microiontophoretically
applied LHRH but not beta-endorphin. Neuroendocrinology 46:24-31
(1987).
Schneider M.A., P.L. Brotherton, and J. Hailes. The effect of exogenous
oestrogens on depression in menopausal women. Med. T. Austr. 2:162-
253 (1977).
Schwartz N.B. (ed.) Dynamics of ovarian function. Raven Press, New York
(1981).


12. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in plasma of ovariectomized rats 104
13. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in liver of ovariectomized rats 106
14. Dose and time-dependent effects of the E2-CDS on E2-Q+ and E2
concentrations in fat of ovariectomized rats 108
15. Effects of a single iv dose of the E2-CDS on plasma E2-Q+ or
plasma E2 concentrations in ovariectomized rats 110
16. Dose and time-dependent effects of the E2-CDS on plasma LH
responses in ovariectomized rats 124
17. Dose and time-dependent effects of the E2-CDS on plasma FSH
responses in ovariectomized rats 125
18. Dose and time-dependent effects of the E2-CDS on plasma PRL
and GH responses in ovariectomized rats 126
19. Effects of the E2-CDS or CAST on plasma testosterone
concentrations either 7 days or 14 days after the final treatment 141
20. Effects of the E2-CDS or CAST on ventral prostate weight either 7
days or 14 days after the final treatment 142
21. Effects of the E2-CDS or CAST on seminal vesicle weight either 7
days or 14 days after the final treatment 143
22. Effects of the E2-CDS or CAST on testis weight either 7 days or 14
days after the final treatment 144
23. Effects of the E2-CDS or CAST on anterior pituitary weight either
7 days or 14 days after the final treatment 145
24. Effects of the E2-CDS or E2 pellet on the mean TST responses
induced by naloxone administration to morphine-dependent,
ovariectomized rats 159
25. Effects of the E2-CDS or E2 pellet on the mean RT responses
induced by naloxone administration to morphine-dependent,
ovariectomized rats 160
X


Treatment Treatment
143
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
0 20 40 60 80 100 120 140
14 Days After Final Treatment
Control 0
Castrated 0
E2-CDS (xl)
E2-CDS (x2)
E2-CDS (x3)
a,b
t'i
0 20 40 60 80 100 120 140
Seminal Vesicle mg/100 g bw
Figure 21. Effects of the E2-CDS (0.5 mg/kg) or CAST on seminal vesicle
weight either 7 days (upper panel) or 14 days (lower panel) after
the final treatment. Animals received weekly iv (tail vein)
injection of either the drug's vehicle (HPCD), or E2-CDS at a dose
of 0.5 mg/kg (1 injection; xl), E2-CDS (2 injections; x2), E2-CDS (3
injections; x3), or were castrated (CAST). The symbol (a) denotes
differences from control (HPCD)-treated animals and the symbol
(b) indicates differences from CAST within the panel as analyzed
by ANOVA and Scheffe statistics.


119
0.01, 0.1, and 1.0 mg/kg doses, respectively. With the highest dose of the E2-
CDS (1.0 mg/kg), uterine weights were significantly increased by about 3-fold
on day 7 which then weights began to decrease but remained elevated at 28
days after the drug administration (Table 8). It should be noted that even at
the highest dose (1.0 mg E2-CDS/kg), uterine weights were less than those
typically observed in gonad-intact rats (500-625 mg).
An equivalent increase in anterior pituitary as well as uterine weights
was observed with E2 (0.7 mg/kg) compared to 1.0 mg E2-CDS dose on day 1.
However, by day 7 the effects of equimolar E2 were equivalent to the lowest
dose of E2-CDS (0.01 mg/kg).
Discussion
The results of this study demonstrated that the E2-CDS causes a dose-
and time-dependent suppression of gonadotropin (LH & FSH) secretion in
OVX rats with maximum reductions in plasma LH and FSH concentrations
occurring 7 days after E2-CDS administration. The time course of
gonadotropin suppression in OVX rats is comparable to that previously
observed for E2-CDS effects on other parameters and in other animal models.
We have reported long-term suppression of LH in castrated male rats (Estes et
al., 1987b; Simpkins et al., 1986), in OVX female rats (Anderson et al., 1988a),
and stimulation of masculine sexual behavior in castrated male rats for 28
days (Anderson et al., 1987a), and body weight alterations for 36 days
(Simpkins et al., 1988) following a single iv administration of the E2-CDS.
Sarkar et al. (1989) have reported on the suspension of estrous cycles in
female rats for 30 days following E2-CDS treatment. These prolonged
pharmacological effects further support the idea that the intermediate


121
feedback mode at the hypothalamus as well as the anterior pituitary. The
evaluation of the effects of E2-CDS on LHRH neuronal activity (i.e. LHRH
release) showed that portal blood concentrations of LHRH were significantly
reduced for more than 16 days following the treatment (Sarkar et al., 1989).
The reduced LHRH secretion, was in contrast to the increased hypothalamic
LHRH concentrations, suggesting that the inhibition of release resulted in a
tissue buildup of the decapeptide. Furthermore, chronic exposure to E2 has
no significant effects on anterior pituitary responsiveness to LHRH (Cooper et
al., 1974), indicating that the prolonged inhibitory effects of E2-CDS on LH and
FSH are due primarily to sustained suppression of LHRH secretion from the
hypothalamus.
When the dynamics of the E2-CDS effects were compared with that of
an equimolar dose of E2, the E2-CDS showed 100-fold greater efficacy in the
magnitude of inhibition of plasma LH and FSH. In other words, the
magnitude of E2 effects was equivalent to that of the E2-CDS but with 100-fold
lower dose (Figure 15). This marked increased in effectiveness and the
prolonged duration of the E2-CDS effects on LH and FSH secretion are most
likely due to "lock-in" of the E2-Q+ behind the BBB with subsequent slow
release of E2 in the brain.
When the kinetic behaviors of E2-CDS and E2 were compared on molar
basis, the E2-CDS (1.0 mg/kg) produced E2 concentrations in brain tissue
which were 81- and 182-fold greater than after an equimolar E2 (0.7 mg/kg)
treatment at 1 and 7 days postinjection, respectively (Chapter 6). Therefore, it
seems more reasonable to suggest that following the E2-CDS administration,
the brain E2 is continuously produced and as such the steady-state brain
concentrations of E2 is dependent on its rate of production from the E2-Q+ and


64
was consistent for each tissue and the variation in % hydrolysis was low (CV
= 0.8% for plasma to 5.0% for brain tissue). Furthermore, other methods of
hydrolyzing E2 conjugates, such as enzyme- or acid-catalyzed hydrolysis, are
much less efficient (11% and 0.8% net hydrolysis, respectively) and require 18
to 24 hours to conduct (Bain et al., 1984; Czekala et al., 1981; Saumande &
Batra, 1984; Segal et al., 1960) versus 20 min for base-catalyzed hydrolysis
under the present conditions.
Analysis of E2-Q+ in adipose tissue was complicated by two factors.
First, E2-Q+ is hydrophilic (due to its charge and polarity) and requires an
aqueous solvent for its effective extraction. This requirement creates
problems in separating the supernatant from the pellet because of the
formation of a superficial layer of lipid above the supernatant phase. Second,
extensive loss of E2 from the supernatant occurs after E2-Q+ was hydrolyzed.
This is due to the presence of fatty droplets in the reaction medium into
which E2 partitioned from the aqueous supernatant; and therefore, it was not
recovered efficiently when the supernatant was transferred onto the Qs
column. These two conditions reduced the recovery of E2-Q+ from adipose
tissue.
Analysis of E2-Q+ in heart and kidney tissues posed a different problem.
After hydrolysis of E2-Q+ extracts in IN NaOH solution, the hydrolyzed
supernatants from heart and kidney required less HC1 and NaH2P04 than
other tissues to adjust the pH to the range of 6 to 8; the optimal pH range for
Ci8 column function. Without determining the exact amount of acid and
buffer needed to achieve the optimal pH range for each tissue, erroneously
high E2-Q+ levels were calculated due to contamination of the assay tube by
the column sorbent.


31
Collectively, the ability to preferentially deliver and sustain the release
of a drug in the brain, thus sparing non-target site tissues, should improve
the therapeutic index of the drug by (i) increasing the concentrations and/or
residence time of the drug at its receptor site in the brain and (ii), equally
important, decreasing the concentrations and/or residence time of the drug at
the potential peripheral sites of toxicities, thereby decreasing its untoward side
effects. Furthermore, this approach may be potentially advantageous in the
treatment of brain diseases by virtue of the need for lower or less frequent
doses of the drug.
This redox-based CDS has been applied successfully to brain-specific
delivery of a wide variety of therapeutic agents, including phenylethylamine
(Bodor et al., 1981; Bodor & Farag, 1983), dopamine (Bodor & Simpkins, 1983;
Simpkins & Bodor, 1985), gamma-aminobutyric acid (Anderson et al., 1987b),
(3-adrenergic blocking agents (Bodor et al., 1988), antitumor drugs (Bodor &
Brewster, 1983), antiviral agents and antibiotics (Bodor & Brewster, 1983),
testosterone (Bodor & Farag, 1984), estradiol (Bodor et al., 1987), and
norethindrone (Brewster et al., 1987).
The application of this redox-based CDS to estrogens, particularly E2
(Bodor et al., 1987), has important clinical and research implications since the
hormone plays major role in the reproductive and nonreproductive
functions by influencing the brain. Estrogens are intrinsically lipophilic and
readily enter the CNS; however, when inside the CNS, there is no
mechanism to prevent their redistribution back to the periphery and thus
exhibit poor retention. Furthermore, because of their inherent lipophilicity,
estrogens equilibrate among all body tissues. This property of the steroid
necessitates either frequent dosing or the administration of a depot form of
the estrogen in order to maintain therapeutically effective concentrations in


138
hyperplasia. This time- and injection-related increase in pituitary weight
appears to be due to the direct effects of E2 on the pituitary gland. E2 is well
known to stimulate PRL secretion and to induce hyperplasia of lactotrophs
(Chen & Meites, 1970). The source of E2 responsible for the stimulation of
anterior pituitary is most likely the brain E2 and not the residual peripheral
E2. This can be explained by the anatomical relationship between the
hypothalamus and the pituitary gland (Traystman, 1983).
We did not observe in intact male rats a statistically significant
suppression in plasma gonadotropins (LH & FSH) in response to E2-CDS
treatment. However, plasma LH showed a progressive decline of 20 to 40%
compared with the basal LH values. In fact, more than 50% of experimental
animals treated with the E2-CDS exhibited plasma LH values which were
below the sensitivity of LH assay. Since plasma samples containing
undetectable LH were assigned the assay sensitivity (0.25 ng LH/ml),
statistically significant LH suppression was not obtained even if LH may have
been significantly suppressed in animals that were treated with the E2-CDS.
Furthermore, since basal LH values of intact male rats were at the limit of the
LH assay sensitivity, it was not possible to make a reliable correlation between
the degree of LH suppression and tissue weight regression in rats treated with
the E2-CDS. Our previous observations in CAST rats showed that LH
concentrations were suppressed by 82-90% for 4-12 days after a single injection
of E2-CDS (Simpkins et al., 1986). It should be noted, however, that CAST rats
are much more sensitive than intact rats to the LH-suppressing effects of E2-
CDS.
Despite the apparent lack of significant LH suppression, plasma T
levels were significantly suppressed by 96% or 76% at 7 or 14 days,
respectively, after a single injection of E2-CDS. This profound and sustained


72
Table 1: Recovery and Precision Determination for Biological Samples Spiked
with E2
Tissue*
E2 Added
(Pg)
E2 Assayed^
(Pg)
Recovery
(%)
CVc
(%)
CCd
(r)
Brain
90
73.9
82.1
7.9
180
130.4
72.4
3.3
360
268.1
74.5
3.4
0.994
Anter. Pituit. 180
141.0
78.3
1.8
NDe
Plasma
90
63.4
70.5
2.5
180
151.8
84.3
3.0
360
321.3
89.2
2.9
0.997
Kidney
90
71.5
79.4
2.2
180
121.8
67.6
1.2
360
238.8
66.3
1.8
0.996
Lung
90
52.0
57.8
6.0
180
100.0
56.4
3.7
360
206.1
57.3
2.4
0.993
Heart
90
52.6
58.5
2.7
180
112.6
62.6
2.8
360
230.8
64.3
4.4
0.997
Liver
90
69.4
77.1
1.7
180
126.5
70.3
1.9
360
245.4
68.2
2.5
0.997
Fat
90
66.4
73.8
3.0
180
97.4
54.1
0.8
360
228.0
63.3
1.1
0.975
a 100 mg of tissue or 1 ml of plasma was used,
b Mean of n = 4 for each dose of E2 in each tissue,
c CV = coefficient of variation,
d CC = correlation coefficient,
e ND = not determined.


99
of E2-CDS to E2-Q+ is a saturable process and is thereby independent of the
dose of E2-CDS administered.
To demonstrate further the preferential deposition and retention of
estrogen in the CNS with the E2-CDS, one dose of 17 p-E2 (equimolar to the
1.0-mg E2-CDS dose) was also studied. As shown in Table 7, E2 concentrations
in the CNS tissues of rats treated with 17 |3-E2 were slightly increased on day 1,
and were just above the detection limits of the assay at 7 days. In contrast, the
1.0 mg E2-CDS dose resulted in brain E2 concentrations that were 81- and 182-
fold greater than those achieved following 17 (3-E2 injection at 1 and 7 days,
respectively. These data demonstrated that the E2-CDS is much more
effective than 17 f)-E2 itself in delivering and retaining the estrogen in the
brain.
Collectively, these observations are consistent with the proposal that E2
can be preferentially delivered to the brain using a redox-based chemical
delivery system, an inert molecule which requires several steps in its
conversion to the parent drug (Bodor et al., 1981; Bodor & Brewster, 1983).
The multiple, facile enzymatic conversions including oxidation and
hydrolytic cleavage may not only lead to preferential E2 delivery and
sustained release/effects, but may also act to decrease the toxicity of the drug.
A preferential and sustained CNS estrogen delivery can be potentially useful
since estrogens are known to influence a variety of CNS functions (McEwen,
1988; Maggi & Perez, 1985).


149
by an acceleration in heart rate (Molnar, 1975) and a surge in secretion of
luteinizing hormone (LH) accompanies the flush response (Casper et al., 1979;
Meldrum, 1979). Additionally, flushes can be provoked by warm ambient
temperature, hot drinks, alcoholic beverages, mental stress, and hypoglycemia
(Simpkins & Katovich, 1984).
Although the mechanism(s) involved in the menopausal syndrome is
as yet unknown, numerous clinical and experimental studies have
implicated the hypothalamic noradrenergic system (Casper & Yen, 1985;
Simpkins & Katovich, 1984), luteinizing hormone-releasing hormone
(LHRH) neuronal system (Gambone et al., 1984), and endogenous opioid
peptide system (Casper & Yen, 1985; Simpkins & Katovich, 1984) in its genesis.
The evidence that gonadal steroid hormones modulate or influence the
activity of each of these central hypothalamic neuronal systems is compelling
(McEwen et al., 1984; McEwen & Parsons, 1982; Roselli & Resko, 1990).
Biochemical, autoradiographic and immunohistochemical experiments have
demonstrated that the most dense collections of cells containing gonadal
steroid hormones and their receptors in the rat brain are found in the medial
preoptic area, hypothalamus, and in the limbic system structures (Luine et al.,
1975; Morrel et al., 1975; Pfaff & Keiner, 1973). These brain regions are key
elements in the neural circuits that regulate the neuroendocrine events of
reproduction and behavior (Barraclough & Wise, 1982; Goodman & Knobil;
Plant, 1986; Christensen & Clemens, 1974). A plausible hypothesis currently
held for the mechanism of the hot flush states that normally the opioidergic-
noradrenergic-LHRH neuronal systems are serially involved in coupling or
mediating the regulatory (feedback) influence of gonadal steroids on LH
secretion and perhaps thermoregulation (Casper & Yen, 1985). However, in
postmenopausal, as well as in long-term ovariectomized women, this neural