Characterization of brain type I receptors for adrenocorticosteroid hormones

MISSING IMAGE

Material Information

Title:
Characterization of brain type I receptors for adrenocorticosteroid hormones
Physical Description:
xiii, 185 leaves : ill. ; 29 cm.
Language:
English
Creator:
Emadian, Seyed Mohammad, 1958-
Publication Date:

Subjects

Subjects / Keywords:
Corticotropin -- physiology   ( mesh )
Receptors, Cell Surface -- physiology   ( mesh )
Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- Neuroscience -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1987.
Bibliography:
Bibliography: leaves 145-183.
Statement of Responsibility:
by Seyed Mohammad Emadian.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000903090
oclc - 17889880
notis - AEL1997
sobekcm - AA00006108_00001
System ID:
AA00006108:00001

Full Text













CHARACTERIZATION OF BRAIN TYPE I RECEPTORS
FOR ADRENOCORTICOSTEROID HORMONES












By

SEYED MOHAMMAD EMADIAN


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


1987




















Dedicated to my parents and my wife without whose continuous

support this work would not have been possible















ACKNOWLEDGMENTS

The experiments in this dissertation were conducted under the

guidance of my mentor and chairman of my supervisory committee, Dr.

William G. Luttge. I am indebted to Dr. Luttge for being a superb

mentor, teacher, colleague, friend, and for his continuous support

throughout the course of my dissertation work. I further would like to

extend my gratitude to my other supervisory committee members, Drs. A.J.

Dunn, M.J. Fregly, M.B. Heaton and K.T. Shiverick, for their guidance

and critical evaluation of my dissertation. I am also grateful to Dr.

L.C. Garg for his participation in the discussion of this work and his

helpful and valuable contributions during our committee meetings. I

would like to thank Dr. C.L. Densmore and Ms. Y.-C. Chou for their

helpful comments and evaluation of the results of this work during our

laboratory meetings and Ms. Mary E. Rupp and Mrs. Cameron F. Bloom for

technical assistance. Last, but not least, I would like to acknowledge

the financial assistance provided to me by the Department of

Neuroscience and Center for Neurobiological Sciences during the course

of my graduate studies.

















TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS . . .. iii

LIST OF TABLES . . ... ....... .vi

LIST OF FIGURES . . .. vii

ABBREVIATIONS . . .. ix

ABSTRACT . . ... xii

CHAPTERS

I REVIEW OF THE LITERATURE . . 1

Control of ALDO Secretion . . 4
Peripheral Site of Action and Effects of
Mineralocorticoids . . 5
Central Site of Action and Effects of
Mineralocorticoids . . 11
Molecular Mechanisms of Mineralocorticoid
Action . . .. 18
Mechanisms of Receptor Regulation . .. 26
In Vitro Analysis of Type I Receptors
by Radioreceptor Assay . ... .36

II EXPERIMENTAL PROCEDURES . ... .43

Isotopes, Steroids and Other Chemicals . .. .43
Experimental Animals. . . ... 44
Buffers . . ... . .44
Cytosol Preparation . .. .45
Aging . . . 45
Hydrophobic Interaction Chromatography . .. 45
Steroid Binding Determination . ... 46
Cytosolic Protein Concentration . ... 47

III EFFECTS OF POLYHYDRIC AND MONOHYDRIC COMPOUNDS ON THE
STABILITY OF TYPE I RECEPTORS FOR ADRENOCORTICOSTEROID
HORMONES IN BRAIN CYTOSOL . ... .48

Summary . . .. . 48
Introduction . . .... .49
Materials and Methods . . .50
Results . . .. . 50
Discussion . . ... .. .. 58











Proposed Mechanism of Action of Poly-
and Monohydric Compounds . ... 58
Possible Mode of Action of Poly- and
Monohydric Compounds on Unoccupied
Type I Receptors in Crude Cytosol .. .64

IV A NOVEL EFFECT OF MOLYBDATE ON THE BINDING OF
[ H]ALDOSTERONE TO GEL FILTERED TYPE I
RECEPTORS IN BRAIN CYTOSOL . ... 70

Summary . . .. . 70
Introduction .... .. . .71
Materials and Methods . . .. 72
Results . . ... . 73
Discussion . .. 88

V DIFFERENTIAL INACTIVATION OF TYPE II RECEPTORS FOR
ADRENAL STEROIDS IN WHOLE BRAIN CYTOSOL:
RECOVERY OF INTACT TYPE I RECEPTORS . .. 93

Summary . . .. 93
Introduction . . 94
Materials and Methods . .. 96
Results and Discussion . . 97
Effects of DTNB on Unoccupied Type I Receptors 97
Effects of DTNB on Occupied Receptors .. 101
Effects of DCC Pretreatment of Cytosol on
Ligand Binding Capacity of Unoccupied
Type I and Type II Receptors ... .106

VI IN VITRO TRANSFORMATION OF ALDOSTERONE-TYPE I
RECEPTOR COMPLEXES TO A DNA-BINDING STATE ... 122

Summary . . .. ... .. 122
Introduction . . ... 123
Materials and Methods . ... 129
Results . . ... ... 130
Discussion . . ... .. 139

VII CONCLUDING REMARKS . .. .142

REFERENCES . . .. ... 145

BIOGRAPHICAL SKETCH . . ... .184

















LIST OF TABLES


Page


Table 2-1:

Table 4-1:



Table 5-1:


Table 5-2:



Table 6-1:


Buffers . . .. .. 44

Binding Parameters for Type I Receptors Obtained
from Crude and Gel Filtered Cytosol in the
Presence and Absence of 2 mM Molybdate 82

Effects of 1 mM DTNB on Whole Brain Cytosolic
Unoccupied Type I Receptors . .. 100

Effects of Two Consecutive DCC Pretreatments
of Cytosol on the Residual DEX Specific
Binding .. . 111

Comparison of the Binding of ALDO-Type I and
DEX-Type II Receptor Complexes to
DNA-cellulose Before and After Salt- and/or
Temperature-Induced Activation .. 132


Table 6-2: Effects of 5 mM Molybdate on the Binding of
ALDO-Type I Receptor Complexes to DNA-
Cellulose Before and After Salt- and
Temperature-Induced Activation .. 136

















LIST OF FIGURES


Page


Figure 3-1:




Figure 3-2:




Figure 3-3:




Figure 3-4:




Figure 3-5:





Figure 3-6:




Figure 4-1:





Figure 4-2:





Figure 4-3:


Effects of Polyhydric Compounds on the Stability
of Unoccupied Type I Receptors in Whole Brain
Cytosol . . .

Effects of 10% Glycerol on the Stability of
Unoccupied Type II Receptors in Whole brain
Cytosol . . .

Effects of Monohydric Compounds on the Stability
of Unoccupied Type I Receptors in Whole Brain
Cytosol . . .

Dose-Response Analysis of the Effects of Ethanol
on the Binding of ALDO to Type I Receptors in
Whole Brain Cytosol . .

Scatchard Analysis of the Effects of 10% Ethanol
on the Equilibrium Binding Parameters of ALDO
Binding to Type I Receptors in Whole Brain
Cytosol . . .

Comparison of the Surface Hydrophobic Properties
of ALDO-Type I and DEX-Type II Receptor
Complexes on Pentyl-Agarose Columns .

Effects of Gel Filtration, Molybdate and/or
Dithiothreitol on the Binding Capacity and
Stability of Unoccupied Type I Receptors From
Whole Brain Cytosol . .


. 52




S. 55




. 57




S. 60





S. 62


. 75


Dose-Response Analysis of the Effects of Gel
Filtration and Molybdate on the Binding
Capacity and Stability of Unoccupied Type I
Receptors frdm Whole Brain Cytosol ... 78

Effects of Molybdate and/or Gel Filtration
on the Equilibrium Binding Parameters of ALDO
Binding to Unoccupied Type I Receptors from
Whole Brain Cytosol . .. 81


vii











Figure 4-4:






Figure 4-5:



Figure 5-1:


Figure 5-2:



Figure 5-3:





Figure 5-4:


Stability of Unoccupied Type I Receptors from
Whole Brain Cytosol Following Gel Filtration
in the Absence of Molybdate: Effects of
Molybdate Addition Following Gel Filtration
and Aging . . .. ... 85

Effects of Dilution and/or Molybdate on the
Binding Capacity of Unoccupied Type I
Receptors from Whole Brain Cytosol ... 87

Effects of 1 mM DTNB on Receptors Bound to
ALDO in Whole Brain Cytosol . .103

Effects of DCC Pretreatment and/or Aging of
Unlabeled Whole Brain Cytosol on the Binding
of ALDO to Type I and DEX to Type II Receptors .105

Effects of Pretreatment of Whole Brain Cytosol
with Various Concentrations of DCC and
Different Duration of DCC Pretreatment on
the Binding of DEX to Type II Receptors ... .109

Effects of DCC Pretreatment of Whole Brain
Cytosol on the Binding of ALDO and DEX to
Type I and Type II Receptors in the Presence
or Absence of a Steroid Competitor .. 114


Figure 5-5: Effects of 300 mM KC1 in the Presence and
Absence of 20 mM Molybdate on the Binding
of ALDO to Type I and DEX to Type II
Receptors . . .


Figure 5-6:



Figure 6-1:


Effects of 300 mM KC1 on the Specific Binding
of ALDO and DEX Prior and Subsequent to DCC
Pretreatment . . ... 119

Comparison of the Surface Hydrophobic Properties
of Unactivated and Activated ALDO-Type I
Receptor Complexes on Pentyl-Agarose
Columns . . .. ... 138


viii
















ABBREVIATIONS


AII: angiotensin II

Ach: acetylcholine

ACTH: adrenocorticotropic hormone

ADP: adenosine diphosphate

AIPs: aldosterone-induced proteins

ALDO: aldosterone

ANF: atrial natriuretic factor

ASF: aldosterone-stimulating factor

ATPase: adenosine triphosphatase

BMAX: maximal binding

BNS: nonspecific binding

B : total binding

CBG: corticosteriod-binding globulin

CPM: counts per minute

CNS: central nervous system

CORT: corticosterone

CSF: cerebrospinal fluid

DCC: dextran-coated charcoal

DEAE-Cellulsoe: diethylaminoethyl-Cellulose

DEX: dexamethasone

DNA: deoxyribonucleic acid

DNA-C: deoxyribonucleic acid-cellulose

DOC: deoxycorticosterone











DOCA: deoxycorticosterone acelate

DPM: disintegration per minute

DTNB: 5,5'-dithiobis (2-nitrobenzoic acid)

DTT: dithiothreitol

E: erythritol

EEDQ: N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinolone

EG: ethylene glycol

Et: ethanol

G: glycerol

HEPES: 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid

hn-RNA: heterogeneous nuclear-ribonucleic acid

Kd: equilibrium dissociation constant

MMTS: methyl methanethiosulfonate

MMTV: mouse mammary tumor virus

mRNA: messenger ribonucleic acid

Mt: methanol

NAD+: nicotinamide-adenine dinucleotide

NADPH: nicotinamide-adenine dinucleotide phosphate

NEM: N-ethylmaleimide

NM: nuclear matrix

P: propanol

PMSF: phenylmethyl sulfonylfluoride

R: ribitol

RPM: revolutions per minute

S: sorbitol

SCO: subcommissural organ

SDS-PAGE: sodium dodocylsulfate-polyacrylamide gel electrophoresis











TA: triamcinolone acetonide

T : melting temperature
X: xyto
X: xylitol

















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


CHARACTERIZATION OF BRAIN TYPE I RECEPTORS
FOR ADRENOCORTICOSTEROID HORMONES

By

Seyed Mohammad Emadian

May 1987

Chairman: William G. Luttge, Ph.D.
Major Department: Neuroscience

Preponderance of evidence suggests that the potent effects of

adrenocorticosteroid hormones on a number of physiological and

behavioral functions are initiated through stereospecific binding of

these steroids with soluble intracellular receptors, followed by the

interaction of the steroid-receptor complexes with the genome. Despite

the wealth of our knowledge regarding the properties of steroid hormone

receptors in general, practical difficulties such as lack of specific

available ligands, receptor instability under cell-free conditions, low

receptor concentration in whole tissue preparations, etc., have greatly

limited our knowledge of properties and regulation of Type I receptors

for adrenocorticosteroid hormones.

The work presented in this dissertation has overcome some of these

difficulties and describes a number of distinct in vitro properties

of these receptors: 1) Unlike Type II (glucocorticoid) receptors, the

extraction of unoccupied Type I receptors into an aqueous environment

was found to result in a reduction in the binding capacity of these










receptors--possibly by inducing a conformation of the receptors with an

infolded hydrophobic steroid-binding pocket. A reduction in hydrophobic

interactions by including monohydric compounds in cytosol was shown to

result in an increase in the binding capacity of these receptors for the

ligand. 2) Although the stabilizing effects of molybdate on steroid

receptors is well established, this oxyanion was found to produce a

paradoxical effect on the binding capacity of unoccupied Type I

receptors in crude and gel filtered cytosol preparations: a reduction

in receptor binding capacity in crude, and an increase in receptor

binding capacity in gel filtered cytosol was observed in the presence of

molybdate. 3) Whereas unoccupied Type II receptors could be inactivated

by removal of endogenous sulfhydryl reducing agent from cytosol using

dextran-coated charcoal (DCC), Type I receptors appeared insensitive to

the effects of dithiothreitol (a potent sulfhydryl reducing reagent) and

DCC pretreatment of cytosol (a mild oxidizing condition). The

differences between these two receptors were used to inactivate

selectively Type II receptors. 4) Lastly, conditions that induced

transformation (activation) of glucocorticoid-Type II receptors to a

DNA-binding state (i.e., 300 mM KC1 and incubation at 22C) were found

to produce minimal activation of Type I receptors. Possible functional

and physiological significance of these findings is discussed.


xiii
















CHAPTER I
REVIEW OF THE LITERATURE

The actions of adrenocorticosteroid hormones on a wide variety of

physiological and behavioral functions in mammals are well characterized

(for review see Fregly and Luttge, 1982; Luttge, 1983; Rees and Gray,

1983; De Kloet 1984; Munck et al., 1984; Meyer, 1985; Funder, 1986;

De Kloet et al., 1986; McEwen et al., 1986b; Munck and Guyre, 1986). As

a result of extensive research, significant developments have occurred

in the past three decades with regard to the mechanism of action of

these hormones (for review see Reich and Scott, 1979; Fanestil and

Kipnowski, 1982; Lan et al., 1984; Rossier, et al., 1985; Marver, 1986;

Scheidereit et al., 1986; Slater et al., 1986). Although limited

experimental evidence suggests that the actions of adrenocorticosteroid

hormones may result from direct interactions of these steroids with

components of membrane (nongenomic response) (Chaplin et al., 1981;

Towle and Sze, 1983), the bulk of evidence suggests that steroid

hormones, in general, exert their effect via a stereospecific binding

with soluble intracellular protein receptors. The resulting

steroid-receptor complexes undergo certain physicochemical changes that

increase the overall affinity of the complexes for nuclei, chromatin and

DNA and other polyanions (for detail see Chapter VI). This

"transformation" process (also known as "activation") allows for a

direct interaction of the transformed complexes with nuclear acceptor

site(s) which in turn modulate the transcription of certain mRNAs and

hence cellular protein synthesis. It is important to note that steroid











hormone receptors are the only regulators of gene expression in the

eukaryotic cells thus far described. Indeed, recent studies have

characterized specific palindromic sequences within the DNA

molecule--hormone response elements--to which the transformed steroid

receptors bind in order to modulate the transcription of certain mRNAs

(e.g., Arnemann et al., 1987; Chambone et al., 1987; Drouin et al.,

1987; Karin et al., 1987; Miesfeld et al., 1987).

Long before isolation and purification of aldosterone (ALDO)

(Grundy et al., 1952; Simpson and Tait, 1952), the most potent naturally

occurring mineralocorticoid hormone, it was known that blood borne

"factors" controlled electrolyte balance within the body (Addison,

1855). Lucas (1926) provided the first evidence for the adrenal

cortical regulation of mineral metabolism by adrenalectomizing dogs.

However, despite these and many other preliminary observations, the

discovery of ALDO production by zona glomerulosa of the adrenal glands

came long after the discovery of the main corticosteroid hormones

produced by the zonae fasciculata and reticularis. The earlier

investigators attributed hyponatremia, hypochloremia, acidosis,

hyperkalemia, and other symptoms associated with Addison's disease to

the lack of deoxycorticosterone (DOC) secretion (Thorn et al., 1942).

Despite the potent mineralocorticoid activity of DOC, this theory was

discredited when extracts from adrenal glands were found to contain only

minute amounts of DOC. The discovery of ALDO came about some years

later as a result of improved quantitative bioassay and chromatographic

techniques (Grundy et al., 1952; Simpson and Tait, 1952; Simpson et al.,

1954; Simpson and Tait, 1955).









3

Because of the potent action of ALDO on mineral metabolism (for

detail see below), the high affinity binders for this steroid extracted

from the kidney were dubbed as "mineralocorticoid receptors." However,

since the endogenous glucocorticoid, corticosterone (CORT), was found to

have comparable affinty for these receptors (see below), a less

ambiguous term, Type I receptors, will be used in this dissertation to

refer to these receptors. This terminology will also be used to

distinguish these receptors from the lower affinity, higher capacity

ALDO binders, the Type II (glucocorticoid) receptors. Although the

principal target tissues for the action of ALDO in mammals have been

considered to be the kidney, intestines and salivary glands, because

electrolyte exchange occurs throughout the organism, other tissues may

also be considered as targets for ALDO. This claim is further supported

by the fact that Type I receptors have been identified in the pituitary

gland (Moguilewsky and Raynaud, 1980; De Nicola et al., 1981), parotid

gland (Funder et al., 1972), mammary gland (Quirk et al., 1983), lung

(Krozowski and Funder, 1981), bovine ciliary body (Starka et al., 1977),

rabbit lens epithelium (Hampl et al., 1981), human ocular lens (Hampl,

et al., 1984; Starka et al., 1986), whole brain cytosol (Anderson and

Fanestil, 1976; Emadian et al., 1986), various brain regions

(Moguilewsky and Raynaud, 1980; De Nicola et al., 1981; Veldhuis et al.,

1982b; Coirini et al., 1983), spinal cord (Orti et al., 1986), human

mononuclear leukocytes (Armanini et al., 1986), rat epididymis (Hinton

and Keefer, 1985; Pearce et al., 1986a), "arterial walls" (Kornel, 1981;

Kornel et al., 1982a & b) and rat cochlear and vestibular organs (Rarey

and Luttge, 1987).








4

Control of ALDO Secretion

In mammals, several known physiological modulators of ALDO

secretion from the adrenal cortex include the renin-angiotensin II

system, potassium ion and adrenocorticotropic hormone (ACTH). An

extensive body of research has implicated the renin-angiotensin II

system as the principal regulator of ALDO secretion (e.g., Kaplan, 1965;

Davis, 1975; Fraser et al., 1979). Although ACTH stimulation of ALDO

steroidogenesis via the adenylate cyclase cascade mechanism is well

characterized, ACTH is apparently not involved in the maintenance of

ALDO secretion under basal conditions. Note, for example, that

suppression of ACTH secretion with dexamethasone (DEX), a synthetic

glucocorticoid, and other manipulations (e.g., hypophysectomy) does not

alter the circadian rhythm of ALDO secretion (Fraser et al., 1979).

Furthermore, infusion of ACTH leads to sustained increases in plasma

ALDO concentration, suggesting that ACTH may mediate only acute

increases in ALDO secretion physiologically (Kem et al., 1975).

A number of other factors including aldosterone-stimulating factor

(ASF, secreted by the anterior pituitary) (Carey and Sen, 1986); atrial

natriuretic factor (ANF, a family of peptides isolated from the atria of

rat, human and other species) (Chartier et al., 1984; Takagi et al.,

1986; Aguilera, 1987; Naruse et al., 1987; Sagnella et al., 1987);

acetylcholine, ACh (Kojima et al., 1986); dopamine (Carey and Drake,

1986; for review see Campbell et al., 1981); vasopressin (Woodcock et

al., 1986) and serotonin (Matsuoka et al., 1985) have been recently

implicated in the regulation of ALDO secretion. Large reductions in the

concentration of sodium in serum have also been shown to stimulate ALDO









5

secretion; however, the physiological role of this hyponatremia-induced

ALDO secretion awaits further exploration (Davis, 1975).

Peripheral Site of Action and Effects of Mineralocorticoids

As mentioned above, the main peripheral action of ALDO is to

maintain electrolyte balance in the organism. A secondary action is to

help maintain proper blood volume and pressure. These actions of ALDO

are achieved primarily through reabsorption of sodium and hence water

(due to osmotic activity of sodium) and excretion of potassium and

hydrogen ions by the kidney. In other words, an increase in the

circulating levels of ALDO results in a decrease in the ratio of sodium

to potassium in urine, hence an increase in plasma and extracellular

fluid volume. An ALDO-mediated reduction in the Na/K ratio is also

reported in saliva, sweat and feces.

The antinatriuretic and kaliuretic effects of ALDO on kidney are

well established. Due to a remarkable anatomical complexity and

specialization in this organ (for review see Jacobson, 1981), the exact

sites at which these effects are mediated remained controversial for

some time (Knox et al., 1980). Because ALDO, deoxycorticosterone

acetate (DOCA, an acetate derivative of DOC) and other steroids with

mineralocorticoid activity were found to influence electrolyte balance

without changing renal plasma flow and glomerular filtration rate

(Barger et al., 1958; Ganong and Mulrow, 1958), the distal tubule was

postulated to be the main locus of mineralocorticoid action (Hierholzer

et al., 1965; Wright and Giebisch, 1978). Studies with isolated,

perfused tubules from rabbit kidneys, however, provided evidence

suggesting that mineralocorticoids (administered acutely or chronically)

increase sodium reabsorption and potassium secretion in the cortical








6

collecting tubules and not the distal tubules (O'Neil and Helman, 1977;

Schwartz and Burg, 1978; Stokes et al., 1981). Furthermore,

mineralocorticoids were also found to increase discretely the

transtubular electrical potential (lumen negative) across the cortical

collecting tubules (Gross et al., 1975; Gross and Kokko, 1977), a

finding not confirmed in the study by Schwartz and Burg (1978).

Determination of pH profile along the rat papillary collecting duct by

micropuncture technique suggested the action of ALDO on distal papillary

collecting duct (Higashihara et al., 1984). In addition to the

possibility that ALDO may influence multiple segments of the nephron,

the uncertainty as to the site of action of mineralocorticoids in kidney

and other target tissues may also be due to species, tissue and/or

methodological differences between the aforementioned studies. For

example, in contrast to the results obtained from the rabbit nephron,

ALDO was found to normalize sodium transport in the rat distal tubule

(Hierholzer and Wiederholt, 1976; Horisberger and Diezi, 1983) without

changing the transtubular potential (Hierholzer and Wiederholt, 1976).

Moreover, ALDO action at receptor sites other that Type I receptors

cannot be ruled out. Note that the administration of DEX (a specific

synthetic glucocorticoid) to adrenalectomized rats has been shown

recently to increase Na /H+ exchange activity in isolated renal

brush border membrane vesicles (Kinsella et al., 1985). Other

investigators, however, reported that glucocorticoids have kaliuretic

and not antinatriuretic effects (e.g., Bia et al., 1982; Campen et al.,

1983; for review see Field and Giebisch; 1985). The use of in situ

microperfusion of single superficial distal tubles in adrenalectomized

rats revealed no direct effect of DEX on potassium secretion by the










microperfused tubule; however, an overall increase in urinary flow and

sodium excretion was a common finding in these studies (Field et al.,

1984). In contrast, infusion of ALDO under identical conditions led to

a 90% stimulation of potassium secretion by the microperfused tubule and

a marked reduction in the kidney sodium excretion and urinary flow rate.

No apparent effect on kidney potassium excretion was observed under this

latter condition (Field et al., 1984). It is postulated that mineral

exchange represents the net result of direct and synergistic actions of

the steroids on renal tubules plus secondary effects mediated by changes

in urinary flow rate (Field et al., 1984). Parenthetically, recent

studies on the effects of corticosteroids on Na+ transport in distal

colon suggest a different cellular mechanism for ALDO and DEX (Na+

transport in response to ALDO, but not DEX, was found to be amiloride

sensitive) (Jorkasky et al., 1985). Furthermore, since basal plasma

levels of ALDO (5 ng/dl) in the presence of potassium loading

significantly increased transepithelial potential difference and the

area of the basolateral membrane in the late distal convolution (a

characteristic feature of "potassium adaptation" when

pharmacological doses of ALDO are used in potassium-loaded

adrenalectomized animals), Hirsch et al. (1984) postulated a

differential mechanism of ALDO action on different segments of nephron.

Note that neither the acute nor chronic administration of ALDO in the

absence of potassium loading has any effect on the late distal

convolution, whereas this latter treatment increased surface density of

the basolateral membrane in initial collecting tubule cells (Hirsch et

al., 1984).








8

Autoradiographic and binding assay studies focusing on the regional

distribution of Type I receptors in various segments of renal nephron

have yielded fruitful results regarding the potential site(s) of

mineralocorticoid action along the nephron. Radioreceptor assays of

cytosolic fractions obtained from rat kidney cortical tubules enriched

in proximal or distal segments revealed a higher concentration of

[3H]ALDO binding in the latter (in the presence or absence of a

glucocorticoid competitor), although the precise location of Type I

receptors along the nephron was not identified (Scholer et al., 1979).

In the rabbit nephron, high specific [3H]ALDO binding ( >10-16

mol/cm tubule length) was found in the branched, cortical and outer

medullary collecting tubules, whereas specific binding was negligible in

the proximal tubules, pars recta, medullary thick ascending limb,

cortical thick ascending limb and distal convoluted tubules ( <10-17

mol/cm tubule length) (Doucet and Katz, 1981a, Farman et al., 1982a).

In contrast, using CORT as a ligand, receptor binding was demonstrated

throughout the rat nephron, albeit the greatest concentration was found

in the cortical collecting tubules (Lee et al., 1983). Autoradiographic

studies are consistent with radioreceptor assays, in that, specific ALDO

binding was observed along the distal and cortical collecting tubules

(Vandewalle et al., 1981; Farman et al., 1982b; Farman and Bonvalet,

1983), but not in the glomeruli (Farman et al., 1982c).

Recent research on the endocrine basis of experimental hypertension

suggest a role for a renally-independent (Berecek and Bohr, 1978;

Onoyama et al., 1979) intraarterial mechanism of mineralocorticoid

action (Kornel, 1981; Kornel et al., 1982a & b; for review see Kotchen

and Guthrie, 1980). Chronic administration of DOCA was found to








9

increase the turnover of arterial tissue potassium and chloride ions and

to increase cell membrane leakinesss" to Na+ (Friedman et al., 1975;

Jones and Hart, 1975; Friedman and Friedman, 1976). In agreement with

these results, ALDO was found to increase significantly sodium flux

along the porcine arterial wall (a tissue with morphological resemblance

to human arterial wall) (Llaurado et al., 1983). An increase in

peripheral resistance and vascular reactivity was also reported in the

DOCA-treated animals (Hansen and Bohr, 1975; Berecek and Bohr, 1977 &

1978). Treatment of normal subjects with the synthetic

mineralocorticoid, 9-a-fluorocortisol, for several weeks resulted in a

rise in blood pressure in those subjects, an effect which could be

reversed by administration of spirolactone (a mineralocorticoid

antagonist) (Nicholls et al., 1979).

Clinically, steroid-induced hypertension is usually associated with

three types of adrenal disorders. 1) Excess secretion of aldosterone

from an autonomous adrenocortical adenoma in primary hyperaldosteronism

(first described by Conn, 1955). This hypertension can be remedied

usually through surgical procedures and/or by administration of

antimineralocorticoids (e.g., spironolactones) (Lim et al., 1986). 2)

Cushing's syndrome, characterized by an excessive secretion of cortisol

(a glucocorticoid) and DOC, is accompanied by hypertension in about

80-90% of the patients (Gomez-Sanchez, 1982). 3) In patients with

11--dehydrogenase and 5a-reductase ("apparent mineralocorticoid

excess," a juvenile disease characterized by low or undetectable levels

of renin and low plasma ALDO, accompanied by hypokalemia, Monder et al.,

1986) and 17a-hydroxylase deficiency (where there is an increase in

ACTH-induced ALDO production) there is a high incidence of hypertension.








10

The exact mechanism by which mineralocorticoids induce hypertension

awaits further elucidation. However, since mineralocorticoids can

influence electrolyte distribution within the arterial walls by

increasing sodium transport and since "specific" mineralocorticoid

receptors along the arterial, but not the venous, walls are known to

exist (Kornel, 1981; Kornel et al., 1982a & b; Moura and Worcel, 1984),

there is strong support for an intraarterial, mineralocorticoid-induced,

receptor-mediated basis for this form of experimental hypertension.

Studies investigating the effects of ALDO, CORT and DOCA on hypertension

in rats suggests that a common mediator of steroid-induced hypertension

may be an increase in extracellular fluid and plasma volume (Haack et

al., 1977; Ishii et al., 1985). Although aldosterone has been

implicated in the pathogenesis of essential hypertension in experimental

animal models (e.g., Nowaczynski et al., 1983), the role of

mineralocorticoids in essential hypertension is still a matter of

controversy (for review see Fraser et al., 1981; Gomez-Sanchez, 1982).

Recently, a central nervous system mediated mechanism of ALDO

induced-hypertension has been suggested (Gomez-Sanchez, 1986) (for

detail see below).

Some other peripheral effects of mineralocorticoid hormones include

a redistribution of potasium within the body in addition to the

kaliuretic effect discussed above (Young, 1979; Young and Jackson,

1982); a reduction in the concentration of zinc in plasma (Latman et

al., 1984); an increase in renal cortical and medullary citrate synthase

activity (Kinne and Kirsten, 1968) and synthesis (Law and Edelman,

1978b), a phenomenon independent of sodium transport (at least in

cultured cells) (Johnson and Green, 1981); an induction of renal









11

flavokinase enzymic activity (Trachewsky et al., 1985); an increase in

thymidine kinase synthesis in immature rat kidney (Bukhari et al.,

1985); an increase in guanylate cyclase activity in liver, kidney,

heart, lung, spleen and ilium (Vesely, 1980); an increase in isocitrate

dehydrogenase and glutamate-oxaloacetate transaminase activity in rat

kidney and toad urinary bladder (Kinne and Kirsten, 1968; Kirsten et

al., 1968; Kirsten et al., 1970; Kirsten and Kirsten, 1972); an

activation of intestinal mucosal carbonic anhydrase in adrenalectomized

rats (Scott and Sapirstein, 1975; Sapirstein and Scott, 1975; Voute and

Meier, 1978; Suzuki, 1981; Suzuki et al., 1983; Suzuki and Ozaki, 1984);

an increase in Na-K-ATPase activity along the nephron (Katz et al.,

1979; Garg et al., 1981; Le Hir et al., 1982; O'Neil and Dubinsky, 1984;

for detail see below); an increase in intestinal Mg -HCO3-

ATPase activity (Suzuki, 1981; Suzuki et al., 1983); a diminution of

bile flow and liver uptake and a decrease in biliary excretion of sodium

and water (Afifi et al., 1979; Afifi, 1981); an increase in red blood

cell sodium uptake (Rettori et al., 1969; Afifi et al., 1979; Stern et

al., 1983) and a non-genomically-induced increase in membrane

phospholipid polyunsaturated fatty acid content (Goodman, 1981).

Central Site of Action and Effects of Mineralocorticoids

Although specific macromolecular binding to ALDO in whole brain

cytosol was first described by Anderson and Fanestil (1976), prior to

and since this discovery investigators have implicated the involvement

of mineralocorticoids in the central nervous system (CNS) electrolyte

balance (Woodbury, 1958); neuronal electrical activities (Kraulis et

al., 1975); feedback mechanisms involving ALDO secretion (Dundore et

al., 1984; Birmingham et al., 1974; Grizzle and Dunlap, 1984);








12

inhibition of CORT-induced serotonin turnover in dorsal hippocampal and

raphe areas (by inhibiting [3H]CORT nuclear uptake) (De Kloet et

al., 1983); inhibition of brain growth (Devenport, 1979); influences on

glial-cell function (Beaumont, 1985) and inhibition of CORT-induced

exploratory behavior (Veldhuis et al., 1982a). In addition,

mineralocorticoids have been shown to augment passive avoidance

responses in adrenalectomized and adrenally intact "pre-stressed" rats,

an effect not observed with CORT administration (Weiss and Gray, 1973).

Although mineralocorticoid-induced hypertension by actions at renal

levels is well documented (see above), early reports implicated a role

for mineralocorticoid action on brain regulation of experimental

hypertension in rats (e.g., Finch et al., 1972; Zamir et al., 1979).

Recently, it was shown that minute doses of ALDO administered

intracerebroventricularly produce a significant rise in blood pressure

(Gomez-Sanchez, 1986). Infusion of ALDO in the presence of an equimolar

concentration of prorenone (a spironolactone with antimineralocorticoid

activity) antagonized the pressor response to ALDO. Studies of this

nature provide strong circumstantial evidence for hypertensinogenic

effects of mineralocorticoids directly within the CNS; however, the

exact site(s) of action of mineralocorticoids on the CNS regulation of

hypertension remains to be elucidated.

Studies investigating the endocrine basis of salt appetite

implicate a central role for mineralocorticoid action (Fregly and

Waters, 1966; Wolf and Handal, 1966; Weisinger and Woods, 1971; McEwen

et al., 1986b). The classical work by Richter (1936) showed that

adrenalectomy was accompanied by sodium appetite in rats and that

treatment of adrenalectomized rats with DOCA reduced sodium intake to










control levels (Richter, 1956). Similarly, a recent report by McEwen et

al. (1986b) showed that the administration of ALDO supressed salt intake

in adrenalectomized rats as measured in a two-bottle preference test

paradigm (findings that replicate the earlier report by Fregly and

Waters, 1966). The suppression of salt appetite in response to ALDO

could be inhibited by co-administration of a molar excess of RU 28318 (a

specific synthetic antimineralocorticoid). It is noteworthy that this

effect of ALDO could not be mimicked by CORT, a proposed physiological

ligand for brain Type I receptors (for review see Funder, 1986) (for

more detail see below). In fact, only concentrations of CORT 500-fold

in excess of ALDO were able to inhibit the action of ALDO (McEwen et al.

1986b).

In sheep deprived of sodium for 22 h, infusion of 500 mM Na+

via a lateral ventricular approach led to a marked reduction in salt

intake compared to animals that received 150 mM Na+ (Weisinger et

al., 1979). It is postulated that the modulation of Na+

concentration in cerebrospinal fluid (CSF) may translate into changes in

the concentration of Na+ in hypothalamic neurons subserving salt

appetite (Weisinger et al., 1979). Similarly, an earlier investigation

by Richter (1956) showed that low concentrations of sodium in plasma

stimulate salt appetite. The author (Ritcher, 1956) implicated (but did

not show) possible involment of hypothalamic neurons. Other CNS

structures have also been implicated in the regulation of salt intake;

for example, electrolytic lesions centered in the thalamic gustatory

subnucleus of the ventral posterior complex were shown to impair sodium

intake in adrenalectomized rats (Wolf and Dicara, 1974). It is

noteworthy that the impairment in salt appetite following such lesions









14

may be unrelated to the gustatory system since neocortical ablations

which resulted in extensive retrograde cell degeneration in this

diencephalic structure had no effect on salt appetite (Wolf et al.,

1970). Becuase other en passage fibers are damaged by the thalamic

lesion (e.g., ascending reticular pathways to the intralaminar nuclei,

ascending noradrenergic pathways to the hypothalamus and forebrain,

etc.), the involvement of other structures could not be ruled out.

Although a specific centers) in the CNS through which

mineralocorticoids may mediate salt appetite and/or other functional

responses is (are) yet to be described, it is suggested that the actual

mechanism of salt appetite induction may be complex and involve a

synergistic association of a number of factors (Fregly and Rowland,

1985).

Since studies of the regional distribution of specific ALDO

receptors within the CNS have revealed distinct localization in certain

limbic structures such as the hippocampus, hypothalamus, indusium

griseum and amygdala, these brain regions have been implicated as

possible candidates for the central site of action of ALDO (Ermisch and

Ruhle, 1978; Moguilewsky and Raynaud, 1980; De Nicola et al., 1981;

Veldhuis et al., 1982b; Coirini et al., 1983; Birmingham et al., 1984;

Reul and De Kloet, 1985 & 1986). However, co-administration of "stress

levels" of CORT with ALDO was found to suppress preferentially ALDO

uptake by hippocampal formation, amygdala and septum, having little

effect on ALDO suppression of salt appetite or ALDO uptake in

circumventricular organs (McEwen et al., 1986b). Furthermore, bilateral

hippocampectomy, which depleted specific [3H]ALDO binding in the

residual structure by 80%, had no effect on the development of salt










appetite following adrenalectomy in these hippocampectomized rats (Kim,

1960; Magarinos et al., 1986). The increase in salt intake in

adrenalectomized-hippocampectomized animals was found to be suppressible

by administration of ALDO (Magarinos et al., 1986). These studies thus

suggest, at least in rats, that the hippocampus may not be the target

organ for the action of ALDO on salt appetite as elicited by

adrenalectomy.

McEwen et al. (1986b) have concluded that one or more of the

circumventricular organs, the dentate gyrus, Ammon's horn and/or

subiculum may be the most plausible candidates for the action of ALDO on

salt appetite since these structures show significant [ H]ALDO

uptake even in the presence of high circulating levels of CORT. This

notion is consistent with the theory that angiotensin II and ALDO act

synergistically to regulate salt intake (Epstein, 1982; Fluharty and

Epstein, 1983; Stricker, 1983; Sakai et al., 1986), and that some of

these circumventricular organs appear to contain angiotensin II

receptors that can be modulated by exposure to ALDO (Wilson et al.,

1986). Recently, it has been proposed that ALDO and angiotensin II

(All) may mediate thirst and salt appetite by influencing the activities

of septal neurons (Stricker, 1984). Lesions of the septal area in

hypovolemic rats were found to produce sensitization to the sodium

appetite-eliciting effects of All; however, this was only seen in the

presence of ALDO (Stricker, 1984). Although these studies reflect a

synergism between the two hormonal systems in inducing electrolyte and

water balance, it must be kept in mind that, in reality, water and

electrolyte homeostasis may involve complicated interactions of a number

of other systems (for review see Fregly and Rowland, 1985).








16

Furthermore, limited experimental evidence comparing adrenalectomized

mice, hamsters and rats suggests clear differences in the control of

salt appetite in these different species.

A developing area of research regarding mineralocorticoid action on

the CNS is concerned with the role of a poorly understood CNS structure,

the subcommisural organ (SCO). The intraperitoneal or subcutaneous

administration of extracts from SCO was found to induce antinatriuresis

in dogs (Gilbert, 1963). Palkovits and Foldvari (1963) reported an

increased nuclear volume in the cells of the adrenal glomerulosa

indicating increased cellular activity following intraperitoneal

injection of SCO extracts. In contrast, lesions of the SCO decreased

nuclear volume as well as the width of the zona glomerulosa in rats

(Palkovits et al., 1965). An in vitro examination of the zona

glomerulosa cells from rats with SCO lesion revealed a marked reduction

in the rate of ALDO secretion (Palkovits et al., 1965). Furthermore,

mineralocorticoids reduced, and adrenalectomy or chronic dehydration

increased cellular activity in the SCO (Palkovits, 1968). Together,

these observations have led investigators to associate SCO with salt and

water homeostasis (for review see Ziegels, 1976).

Recently, Dundore et al. (1984) found that when ALDO was

administered in the "general vicinity" of the SCO (note that the

injections were performed via a cannula placed in the pineal recess

above the rostral two-thirds of the SCO), it produced a natriuresis (an

action opposite to those of blood-borne ALDO discussed above) and a

"site-specific" (related to cannula placement) reduction in the

cross-sectional area of the adrenal medulla without affecting cell

density. It is proposed, but not conclusively shown, that ALDO acts








17

upon SCO cells (and possibly other structures and cells in the vicinity

of SCO due to diffusion) to inhibit the trophic effects of SCO on

adrenal glomerulosa cells. This, in turn, would result in an increased

urinary Na/K ratio (Dundore et al., 1984). Although an attractive

hypothesis, this study fails to show a change in plasma concentration of

adrenocorticosteroids following intracerebral administration of ALDO.

Furthermore, intracerebral administration of ALDO was found to have no

significant effect on the cortical (steroid producing) cells of the

adrenal gland when examined histologically (Dundore et al., 1984). In

light of the above discrepancies and the fact that Crow (1967), who made

discrete lesions encompassing all or parts of the SCO observed no

changes in urine volume and concentration, it is apparent that the exact

role of the SCO in electrolyte metabolism awaits further experiments.

It is conceivable that research investigating 1) the direct effects of

SCO lesions on plasma levels of corticosteroids, 2) the localization

within the SCO of specific mineralocorticoid receptors by

autoradiographic and radioreceptor assay techniques and 3) the isolation

and purification of SCO factors(s" responsible for the aforementioned

effects is necessary to establish the physiological role of the SCO, and

its secretary factors in salt and water homeostasis.

Some other centrally mediated actions of mineralocorticoid hormones

include the regulation of food intake and maintenance of body weight,

(Devenport and Devenport, 1983; Devenport et al., 1983; Kenyon et al.,

1984); modulation of hypothalamic, thalamic and septal angiotensin II

receptor binding capacity (Wilson et al., 1986) and increase in cerebral

guanylate cyclase activity (Vesely, 1980).









18

Molecular Mechanisms of Mineralocorticoid Action

Recent in vitro radioreceptor assay and ligand competition

techniques suggest that in cytosol lacking extravascular CBG, Type I

receptors have an equal affinity for ALDO and CORT (for review see

Funder, 1986). It is further suggested that some actions of CORT on the

CNS may be mediated through Type I receptors (e.g., De Kloet et al.,

1986; Magarinos et al., 1986). However, because in this dissertation

the emphasis is placed on the role of Type I receptors in electrolyte

homeostasis, this section will focus primarily on our current

understanding of the molecular mechanisms of action of ALDO and other

physiological mineralocorticoids.

Research on the mechanism of mineralocorticoid action has drawn

tremendous benefit from studies carried out using isolated nephrons, the

A6 cell line derived from the Xenopus kidney epithelial cells and

the sodium transporting amphibian epithelia (toad urinary bladder and

frog skin). These biological preparations can be maintained viable for

several hours at room temperature in incubation media of simple

composition. The simple morphological properties of these models render

them ideal in the study of the mechanisms) of mineralocorticoid action;

however, it is noteworthy that although aldosterone is the naturally

occurring mineralocorticoid produced by the interrenal gland in anurans

(Carstensen et al., 1961; Crabbe, 1961), the "mineralocorticoid

response," at least in toad bladder appears to be mediated through

interactions with Type I as well as other glucocorticoid specific

receptors (Geering et al., 1985). Even though the authors dubbed these

other receptors as "type 2" glucocorticoid receptors, the possibility








19

they may be Type IB (for more detail see below) and thus distinct from

classical glucocorticoid (Type II) receptors should also be considered.

In the early 1960's, Williamson (1963) and Edelman and co-workers

(1963) independently proposed the classic two-step genomic mode of ALDO

action. The first step was the binding to specific soluble

intracellular receptors, following which, as a second step, the

steroid-receptor complexes would translocate into the nucleus where they

would interact with the cellular genomic machinery (for a discussion of

the nature of the nuclear acceptor site see Chapter VI). It is this

latter step that was proposed to result in the modulation of synthesis

of a number of structural and regulatory proteins. Several lines of

evidence appeared later in favor of this proposal. For instance, Sharp

and colleagues (1966b) were the first to report saturable ALDO binding

sites (receptors) in the toad bladder. Marver et al. (1974) later

reported that the mineralocorticoid antagonist, spirolactone (SC-26304),

when completed with these receptors, had little or no affinity for

nuclear acceptor sites in vivo and in vitro. More recently, the

response to ALDO in the rat kidney was shown to depend clearly on the

ontogeny of Type I receptors in this target organ (Stephenson et al.,

1984a). Finally, it was possible to correlate the magnitude of

epithelial sodium transport with the binding of ALDO-receptor complexes

to chromatin and the increases in the synthesis of a specific class of

nonmethylated mRNAs and of proteins (Rossier et al., 1974). In this

regard, actinomycin-D (a DNA intercalator which binds specifically to

G--C base pairs and protrudes into the major grooves) was shown to block

ALDO-stimulated sodium transport by inhibiting new mRNA and protein

synthesis (Lahav et al., 1973; Rossier et al., 1974; Horisberger and











Diezi, 1984). Subsequent to these observations a number of studies

replicated the results described (for review see Edelman and Marver,

1980; Fanestil and Park, 1981; Rossier et al., 1985; Garty, 1986;

Marver, 1986) thus further confirming the original model proposed by

Edelman et al. (1963) and Williamson (1963).

It is now well established that mineralocorticoids, in general, are

capable of inducing the synthesis of a group of proteins in target

tissues generally referred to as aldosterone-induced proteins (AIPs).

In the rat kidney and toad urinary bladder, proteins with a variety of

molecular weights have been reported to exhibit increased incorporation

of radiolabeled precursor amino acids as a result of exposure to

aldosterone (Law and Edelman, 1978a & b; Geheb et al., 1981; Reich et

al., 1981; Yang et al; 1981; Blazer-Yost et al., 1982; Geering, et al.

1982; Geheb et al., 1983; Geheb et al., 1984). However, the mechanism

by which AIPs participate in the mineralocorticoid-induced physiological

and functional consequences such as sodium transport still remains a

matter of debate. In fact, only recently was it shown that some AIPs

can be immunoprecipitated by polyclonal antibodies specific for the c-

and B-subunits of Na-K-ATPase (Geering et al., 1982).

Current thinking on the role of AIPs, as related to the action of

mineralocorticoids, points to three distinct, but not necessarily

independent, theories: 1) the sodium pump theory, which argues for

increased sodium pump activity as a result of either activation of

pre-existing pumps or an increase in the synthesis of new pumps; 2) the

sodium permease theory, which holds that AIPs are involved in the

facilitation of passive sodium entry across the plasma membrane and 3)

the energy theory, which associates AIPs with the augmentation of










cellular energy, presumably by enhanced mitochondrial oxidative

phosphorylation. Although each of these theories has some experimental

support, thus far none has gained universal acceptance. It is important

to emphasize that the involvement of AIPs may not be limited to only one

of these mechanisms exclusively; rather the effects of

mineralocorticoids may reflect the contribution of AIPs to some

combination of these three general mechanisms. In this regard, Reich

and Scott (1979) proposed a unifying hypothesis that includes all three

of the above theories (for recent reviews see Rossier et al., 1985;

Garty, 1986; Marver, 1986).

Considerable controversy still surrounds the pump theory. In the

past decade, a large number of laboratories has investigated the

mineralocorticoid induction of Na-K-ATPase activity (e.g., Aperia et

al., 1981; Doucet and Katz, 1981b; Garg et al., 1981; Petty et al.,

1981; Geering et al., 1982; Osore and Gilbert, 1982; Cortas et al.,

1983; El Mernissi and Doucet, 1983; Stern et al., 1983; Aperia and

Larsson, 1984; El Mernissi and Doucet, 1984; O'Neil and Dubinsky, 1984;

Park and Edelman, 1984a & b; Geering et al., 1985; O'Neil and Hayhurst,

1985; Girardet et al., 1986; Johnson et al., 1986; Palmer and Speez,

1986). As a result of this intensive investigation, it is now well

established that following adrenalectomy there is a significant

reduction in Na-K-ATPase activity in nephrons; however, some

investigators find complete restoration of the enzymatic activity

following administration of ALDO (Petty et al., 1981, El Mernissi and

Doucet, 1983; El Mernissi and Doucet, 1984), whereas others find little

or no effect of ALDO on the restoration of enzyme activities in the

target tissues investigated (Doucet and Katz, 1981b).











In addition to this discrepancy in results, the studies cited above

also fail to prove that sodium pump activity plays a significant role in

the action of mineralocorticoids on transepithelial or nephron sodium

transport. Indeed, it is suggested that the observed increase in

Na-K-ATPase activity, at least in some target tissues, may be due to

interaction of mineralocorticoids with Type II receptors, since

long-term pharmacological doses of steroids are often required to

restore Na-K-ATPase activity to control levels. Recently, a number of

investigators (Rayson and Lowther, 1984; El Mernissi and Doucet, 1984;

Garg et al., 1985) reported that they were able to restore the

adrenalectomy-induced reduction in Na-K-ATPase activity in various

segments of rabbit nephron through DEX administration. Even though

similar results were reported by earlier investigators (Fisher et al.,

1975), the small doses of DEX used in the former reports (Rayson and

Lowther, 1984; El Menissi and Doucet, 1984; Garg et al., 1985), provide

stronger evidence that the stimulation of the Na-K-ATPase was probably

through Type II receptors. Note that although the radioreceptor assay

technique suggests that DEX binding to Type I receptors may be

negligible, the pharmacological doses of DEX administered in some of

these studies may allow sufficient interaction of this glucocorticoid

with these receptors to mediate mineralocorticoid action.

In contrast to this suggestion, El Mernissi and Doucet (1984) have

shown that ALDO administration increases the synthesis of new

Na-K-ATPase in the collecting tubules, whereas DEX stimulates the

activity of the pre-exisiting pumps in the thick ascending limb and

distal convoluted tubules. Thus, these results suggest that ALDO and

DEX may act through differential mechanisms at different portions of the










nephron. A differential action of ALDO in the rat proximal and distal

colon has also been described: ALDO response in distal, but not

proximal, colon could be inhibited by amiloride (an acylguanidine

diuretic that inhibits influx of Na through the apical membrane)

(Hirsch et al., 1985). In contrast, in vitro studies evaluating

short-circuit current in colon from adrenalectomized rats suggest that

although ALDO may be the physiological regulator of Na+ transport,

glucocorticoids may also mediate amiloride-sensitive sodium transport

(Will et al., 1985). Geering et al. (1985) maintain that, at least in

toad urinary bladder, ALDO binding to both Type I and Type II receptors

is required for full mineralocorticoid response.

It is also conceivable that DEX actions on Na-K-ATPase activity may

be mediated through a population of receptors distinct from Type I and

Type II receptors called Type IB. The presence of these receptors in a

number of tissues including the rat liver (Litwack and Rosenfield,

1975), colon (Bastl et al., 1984) and kidney (Litwack and Rosenfield,

1975; Markovic et al., 1980; Mayer et al., 1983) has been confirmed (for

review see Mayer and Litwack, 1983). These receptors have also been

isolated from mouse lung cytosol (Goldman and Katsumata, 1986) and were

implicated to mediate the known anti-inflamatory and teratogenic actions

of DEX and phenytoin (Katsumata et al., 1985). Some investigators,

however, have questioned the polymorphic nature of these receptors and

have argued that these atypical receptors may represent a proteolytic

fragments) generated from Type II receptors under in vitro

conditions (e.g., Sherman et al., 1983; Reichman et al., 1984). More

recently, Eisen et al., (1986) showed that kidney cytosol prepared in

the absence, but not the presence, of molybdate (for detail of possible











mechanism of action of this oxyanion see Chapter IV) showed specific

glucocorticoid binders characteristics of Type IB receptors (i.e., the

receptors did not react with a monoclonal antibody raised against Type

II receptors and they were not retained by DEAE-cellulose columns). In

contrast, Goldman and Katsumata (1986) have reported that Type IB

receptors could be detected in lung cytosol only when cytosol was

prepared in hypotonic buffers containing molybdate. Although the

questions as to whether Type IB receptors exist in vivo and whether

they have physiological significance remain to be established, it is

entirely possible that these receptors may be generated from Type II

receptors by specific proteases in target tissues that appear to contain

these receptors.

Experiments with adrenalectomized animals treated with amiloride

prior to injection of ALDO suggest that the increase in Na-K-ATPase

activity may be a secondary adaptation to sodium transport induced by

this steroid (Doucet and Katz, 1981b; Petty et al., 1981). Furthermore,

at least in toad urinary bladder, it appears that the stimulatory action

of ALDO on sodium transport is not mediated by the synthesis of new

(Geering et al., 1982, Park and Edelman, 1984b) or activation of the

pre-existing sodium pumps (Park and Edelman, 1984b).

The idea that AIPs may be a component of the plasma membrane (on

the luminal, urine, surface) of the cell, where they act as a sodium

permease to lower the resistance to Na entry into the cell, was

proposed originally by Crabbe and de Weer (1965) and Sharp et al.

(1966a) and subsequently gained experimental support through work in the

toad urinary bladder (Civan and Hoffman, 1971; Spooner and Edelman,

1975; Palmer et al., 1982; Park and Edelman, 1984a). All of these








25

studies, however, fail to demonstrate whether AIPs act as permeases

themselves, or whether they modulate the conformation or arrangement of

pre-existing permease molecules in the membrane such that they are more

accessible for Na+ entry. A recent study by Kipnowski et al. (1983)

found that ALDO fails to increase sodium transport in tissues where

transport was inhibited by N-ethoxycarbonyl-2-ethoxy-l,2-dihydro-

quinoline (EEDQ) prior to exposure to this steroid. In contrast to

results with cycloheximide (an inhibitor of protein synthesis), EEDQ did

alter ALDO stimulation of the osmotic water flow in response to

antidiuretic hormone. These data do not therefore support the notion

that AIPs are new sodium channels, but instead they are consistent with

the hormonal activation of pre-existing, nonfunctional channels.

The first evidence that mineralocorticoids may induce the synthesis

of a mitochondrial enzyme participating in tricarboxylic acid cycle was

obtained by Kirsten et al. (1968) who found that ALDO increased the

citrate synthase, isocitrate dehydrogenase, glutamate dehydrogenase,

glutamate-oxaloacetyl transaminase and malate dehydrogenase activity in

the toad bladder. The increase in the synthesis of mitochondrial

enzymes was postulated to reflect hormone-controlled mudulation of the

available ratio of ATP/ADP which, in turn, could alter both Na+

conductance and active transport. ALDO has also been shown to increase

the ratio of NADH/NAD in adrenalectomized rat kidneys (Kirsten and

Kirsten, 1972). Similarly, Law and Edelman (1978b) found an increase in

renal cortical and medullary citrate synthase activity three hours after

administration of ALDO to adrenalectomized rats. In this latter study

ALDO was found to enhance incorporation of radiolabeled methionine into

renal, but not hepatic, citrate synthase. The relevance of these








26

observations with reference to the energy theory of mineralocorticoid

action lies in the fact that actinomycin D and specific spironolactones

inhibit the ALDO-induced increase in mitochondrial enzymatic activities,

and that DEX has no effect on the hormonal induction (Law and Edelman,

1978b). Furthermore, this action of ALDO appears to be target specific:

ALDO had no detectable effect on the citrate synthase activity in the

renal papilla or the liver under the same conditions (Law and Edelman,

1978b).

In summary, although much has been done to expand our knowledge of

the molecular mechanisms of mineralocorticoid action, the exact

mechanism by which these hormones exert their effects still awaits

further exploration. It is, nevertheless, clear that the functional

consequences of mineralocorticoid action involves, at least in part, the

genetic machinery in target cells and an ultimate modulation of the

synthesis and/or activation of the pre-existing structural and

regulatory proteins.

Mechanisms of Receptor Regulation

There is considerable evidence to suggest that the in vivo

binding capacity of adrenocorticosteroid hormone receptors is subject to

change in response to a variety of intra- and extracellular signals

(McEwen, 1979; Muldoon, 1980; Svec, 1985b; De Kloet et al., 1986; McEwen

et al., 1986a). For example, hyperkalemic adrenalectomized rats were

shown to have an elevated renal cytosolic Type I receptor binding

capacity (Rafestin-Oblin et al., 1984). Similarly, adrenalectomy

(adrenocorticosteroid deprivation) alone was found to result in a

time-dependent increase in the maximal binding capacity of both Type I

and Type II receptors in all tissues investigated (Beato et al., 1974;








27

McEwen et al., 1974; Gregory et al., 1976; Ichii, 1981; Claire et al.,

1981; Muramatsu et al., 1983, Rafestin-Oblin et al., 1984; Turner, 1986;

Sarrieau, et al., 1986; Luttge and Rupp, unpublished). It is noteworthy

that adrenalectomy was found to induce a differential tissue-specific

up-regulation in Type I and Type II receptor binding capacity in mice.

In the three tissues investigated (i.e.; brain, liver and kidney),

adrenalectomy was found to induce the greatest up-regulation in the

brain cytosolic Type I receptors -- a 10- to 14-fold rise compared to

the adrenally intact control group as measured by saturation and

equilibrium binding analyses (Luttge and Rupp, unpublished).

Adrenalectomy also induced a similar tissue-specific up-regulation in

Type II receptors: The largest increase was found in the kidney

cytosol, followed by amygdala-entorhinal cortex, hippocampus, liver,

cerebral cortex, hypothalamus, pituitary and heart (Turner, 1986).

In Scatchard plots (Scatchard, 1949), an increase in the BMAX

as a mode of "up-regulation" of Type I and Type II receptors induced by

adrenalectomy is a common finding (Beato et al., 1974; McEwen et al.,

1974; Gregory et al., 1976; Claire et al., 1981; Ichii, 1981; Tornello

et al., 1982; Muramatsu et al., 1983; Rafestin-Oblin et al., 1984;

Sapolsky et al., 1984; Luttge and Rupp, unpublished). This increase in

the BMAX has been postulated to reflect an unmasking of receptors

previously occupied by endogenous ligands (Feldman, 1974; McEwen et al.,

1974; Giannopoulos, 1975), "activation" of an already available pool of

receptors to a ligand-binding state, an increase in the rate of receptor

synthesis (McEwen et al., 1974; Claire et al., 1981) and/or a reduction

in the rate of receptor degradation (Claire et al., 1981).










Treatment of adrenalectomized animals with CORT, ALDO and other

steroid agonists that show appreciable affinity for Type I and Type II

receptor binding sites results in a reduction in the ligand binding

capacity of these receptors (Grekin and Sider, 1981; Ichii, 1981;

Tornello et al., 1982; Muramatsu et al., 1983; Rafestin-Oblin et al.,

1984; Sarrieau et al., 1986; Luttge and Rupp, unpublished). Repeated

stress (a condition that induces the production of high circulating

levels of CORT) was also found to result in a significant reduction in

Type II receptor binding capacity in the rat hippocampus (Valeri et al.,

1978; Sapolsky et al., 1983; Sapolsky et al., 1984; Sapolsky, 1985;

Sapolsky et al., 1985b) and amygdala (Sapolsky et al., 1983; Sapolsky et

al., 1984). In addition to regulation of their own receptors, chronic

administration of stress level concentrations of corticosteroids was

found to have neurotoxic effects on hippocampal neurons (Sapolsky, 1985;

Sapolsky et al., 1985b). Similarly, in cultured cell lines (HeLa S3,

AtT-20 and GH1 cells), glucocorticoids were shown to "down-regulate"

their own receptors (Cidlowski and Cidlowski, 1981; Svec and Rudis,

1981; Raaka and Samuels, 1983; Svec, 1985a). This down-regulatory

effect of ligand on its own receptor has been reported for a number of

other receptors; e.g., progesterone receptors (Milgrom et al., 1973b;

Walters and Clark, 1979), thyroid hormone receptors (Raaka and Samuels,

1981), insulin receptors (Gavin et al., 1974; Huang and Cuatrecasas,

1975; Olefsky, 1976) and other peptide receptors (Catt et al., 1979;

Roth and Taylor, 1982).

In GH1 cells, incubation with triamcinolone acetonide (TA) was

found by McIntyre and Samuels (1985) to reduce the half-life of Type II

receptors by more than 50% (from 19.0 to 9.0 h) as determined through









29

the use of dense amino acid labeling of the receptors. Consistent with

this finding, in the presence of TA, the levels of Type II receptors in

these cells were half that measured in the control cells (i.e., about

130 vs. 260 fmole/100 ug DNA, respectively). The addition of TA to the

culture media had little effect on the rate of de novo receptor

synthesis. It was shown that TA promoted the activation of Type II

receptors (transformation to a state with increased affinity for DNA,

see Chapter VI), and that the activated form was degraded at a faster

rate (than the unactivated form) thus accounting for the reduction in

the receptor half-life (McIntyre and Samuels, 1985). This apparent

faster rate of degradation may represent the inability of the activated

receptors to rebind ligand, since recent work from our laboratory has

shown that following heat-induced transformation to a DNA-binding form

the activated receptors do not rebind ligand after dissociation (Chou

and Luttge, 1987). Recently, treatment of rat hepatoma culture cells

with DEX was shown to result in an initial increase (after 6 h),

followed by a reversible (after 72 h) 50% to 95% decrease in Type II

receptor mRNA (after 24 h) in these cells (Okret et al., 1986). In this

latter study, using an immunoprecipitation assay, Type II receptors were

shown to interact specifically with a Type II receptor cDNA clone, hence

providing a more direct evidence for the ligand-induced receptor

autoregulation (Okret et al., 1986).

The up- and down-regulation of the binding capacity of receptors

for adrenal steroids in the central nervous system raises the

possibility that receptor autoregulation, and hence modulation of brain

sensitivity to steroids, may ultimately be involved in such functional

consequences as the feed-back control mechanisms of adrenal steroid










secretion (Cake and Litwack, 1975; Sapolsky et al., 1985a; for review

see Sapolsky et al., 1986).

A second mode of receptor regulation, at least in

adrenocorticosteroid Type II and progesterone receptor systems, may

involve a rapid turning "on" and "off" of the unoccupied receptors in

response to different stimuli. This rapid and reversible loss of

steroid binding capacity has been shown to involve a phosphorylation-

dephosphorylation cycle of the receptors in vitro and possibly

in vivo (for review see Housley et al., 1984). In an earlier

report, Munck and co-workers (1972) proposed that Type II receptors in

thymocytes exist in two states; the steroid-binding form and those that

did not bind steroids. It was proposed that ATP was required to convert

the non-binding receptors to a state that bound steroid (Munck et al.,

1972). Consistent with this finding, receptor stability was reported to

be enhanced by including ATP and other nucleotides (a requirement for

the action of some phosphorylase enzymes) in cytosol preparations

(Nielsen et al., 1977b; Sando et al., 1979a & b). Our laboratory has

recently shown that exogenous ATP can also prevent the destabilizing

actions of as yet unknown microsomal components on Type II receptors

(Densmore and Luttge, 1985). The use of phosphatase inhibitors (e.g.,

fluoride and molybdate) in cytosol preparations has been shown to

improve considerably the stability of Type II receptors (Nielsen et al.,

1977b). Conversely, the addition of highly purified calf intestine

alkaline phosphatase to cytosol preparations markedly destabilized these

receptors (Nielsen et al., 1977a). A striking, positive correlation

between ATP concentration and Type II receptor binding capacity in a

human cell line (i.e., IM-9 cells) was also observed (Wheeler et al.,








31

1981). Lastly, affinity labeling of Type II receptors (from L-cells

grown in the presence of [32P]orthophosphate) with dexamethasone

21-mesylate has provided direct evidence suggesting that these receptors

may be phosphoproteins (Housley and Pratt, 1983). It is important to

note that although these data clearly favor the role of

phosphorylation-dephosphorylation as a possible mechanism of Type II

receptor regulation, there is as yet no experimental evidence to suggest

a similar mechanism for up- and down-regulation of Type I receptors.

A third mechanism for the potential regulation of Type II and

possibly Type I receptors (which may also be related to the

phosphorylation-dephosphorylation cycle (Housley et al., 1982, for

review see Housley et al., 1984)), is the reduction and oxidation of

sulfhydryl groups) in the receptor macromolecule. It is well known

that thiol-disulfide exchanges can markedly affect the activity of a

number of protein macromolecules including androgen, estrogen,

progesterone, ecdysone and insulin receptors, lysozyme, immunoglobulins

and many digestive enzymes. In 1975, Rees and Bell claimed that they

were able to recover partially the loss in binding capacity of rat

thymic cytosol receptors by including sulfhydryl reducing reagents,

e.g.; dithiothreitol (DTT) and 2-mercaptoethanol, in their buffers. Two

years later, Granberg and Ballard (1977) reported that the addition of

2 mM DTT to rat lung cytosol increased the apparent binding of Type II

receptors 10-fold. This increase was proposed to be due to

transformation of an already-available pool of receptors to a

ligand-binding state. Similar effects were also seen in brain, uterus,

thymus and a number of other tissues whereas the presence or absence of

DTT was found to have little effect on Type II receptor binding capacity








32

in liver, kidney and heart cytosol. It was postulated that these latter

tissues contained endogenous, heat-stable, sulfhydryl reducing agents

which stabilize Type II receptor binding capacity. Interestingly, an

analysis of Type II receptors in fetal (Ballard et al., 1974) and adult

(Granberg and Ballard, 1977) rat lung revealed that exogenous DTT was

required only in the cytosol prepared from adults. Together these

observations suggest that variations in the production of sulfhydryl

reducing (and/or oxidizing) reagents may be responsible for differential

stabilization of Type II receptors in various tissues and in the same

tissue during different stages of development.

An endogenous protein molecule, capable of modulating steroid

binding capacity of the rat liver cytosolic unoccupied Type II

receptors, has recently been characterized by Pratt and co-workers

(Grippo et al., 1983 & 1985). This endogenous receptor-reducing agent

appears to be an NADPH-dependent thioredoxin. Extraction of this

stabilizing factor from liver cytosol by charcoal resulted in a

significant loss of unoccupied, but not occupied Type II receptor

binding capacity. Addition of DTT to the charcoal-extracted cytosol

resulted in a complete recovery of Type II receptor binding capacity

(Grippo et al., 1983). Similar results have been obtained in our

laboratory. We found that charcoal extraction of steroid-free brain

cytosol prepared in HEPES [4(2-hydroxyethyl)-l-piperazineethanesulfonic

acid] buffer containing 10% glycerol and 2 mM DTT resulted in an

increase in the thermal instability and a decrease in the total binding

capacity of Type II receptors (Emadian et al., 1986). These effects

were reversible after the readdition of the exogenous DTT removed by the

charcoal treatment. In further investigating the effect of the presence








33

or absence of DTT, we found that the omission or removal of this

sulfhydryl reducing reagent produced a significant increase in the

apparent Kd and a slight reduction in the BMAX of [3H]DEX

binding to Type II receptors (Densmore et al., 1984b).

The effect of DTT on Type II receptor binding capacity appears to

be a temperature-independent phenomenon (Rees and Bell, 1975; Granberg

and Ballard, 1977; Grippo et al., 1983; Densmore et al., 1984; Emadian

et al., 1986), whereas the NADPH-dependent thioredoxin appears to act in

a temperature-dependent fashion (Grippo et al., 1983). These findings

suggest that the exogenous sulfhydryl reducing reagents may act

primarily on the receptor [or other components) of the system] to

maintain it in a reduced state -- the conformation necessary for steroid

binding. In contrast, the reducing effects of thioredoxin must be

mediated primarily through an enzymatic process (Grippo et al., 1983 &

1985). Thioredoxin reductase was found to be responsible for the

transfer of reducing equivalents from NADPH to thioredoxin. The reduced

form of the thioredoxin, in turn, is thought to maintain the receptors

in the steroid-binding state (Grippo et al., 1985). The transfer of the

reducing equivalent may be shown schematically as follows:

thioredoxin reductase
oxidized thioredoxin reduced thioredoxin
I NADPH


oxidized thioredoxin
+ oxidized receptor
reduced receptor (incapable of binding ligand)
(ligand binding conformation)

The observation that occupied receptors from charcoal extracted

cytosol (i.e., in the absence of thioredoxin and/or DTT) appear to be

unaffected by charcoal treatment has raised the possibility that the








34

sulfhydryl groups) on the receptor may be located at or near the

steroid binding domain (Grippo et al., 1983). Alternatively, the

binding of ligand to Type II receptor could induce a conformational

change in the receptor molecule which renders more distant sulfhydryl

groups) inaccessible for oxidation. With regards to this latter

hypothesis, a recent preliminary study from our laboratory suggests that

DTT may have no detectable effects on the sedimentation coefficient of

unoccupied Type II receptors (Densmore et al., 1984). It is important

to remember that the method of sucrose density gradient

ultracentrifugation may not be sensitive enough to reveal small

conformational changes in receptor.

Another possible regulatory role of the thiol-disulfide exchange

mechanism involves the interaction of steroid-receptor complexes with

the cellular genomic machinery. Preatreatment of DEX-receptor complexes

with sulfhydryl oxidizing reagents [e.g.; methyl methanethiosulfonate,

MMTS and 5,5'-dithiobis (2-nitobenzoic acid), DTNB] inhibits subsequent

binding of the activated receptor complexes to DNA-cellulose (an

in vitro index for receptor transformation) (Bodwell et al., 1984).

Interestingly, the inhibitory effect of these reagents can be reversed

by the addition of DTT. These results were interpreted to suggest that

sulfhydryl groupss, located at or near DNA-binding domain of the

activated steroid-receptor complex, may regulate the interaction of the

complex with the genome (Bodwell et al., 1984). It is noteworthy that

although MMTS and DTNB were shown to diminish DNA-cellulose binding of

the complexes, the authors (Bodwell et al., 1984) failed to refute

conclusively the possibility that these reagents inhibited DNA-binding

by acting as cross-linking reagents (i.e., cross-linking of the











activated receptors with other cytosolic components such as heat-shock

proteins, mRNAs, etc.). In fact, current data clearly favors the notion

that N-ethylmaleimide (NEM) and iodoacetamide (other sulfhydryl

modifying reagents) inhibit transformation of rat liver Type II

receptors (Kalimi and Love, 1980) probably through inhibition of subunit

dissociation (for detail see Chapter VI). Furthermore, it is possible

that the bulky thiomethyl and thionitrobenzoate complexes (formed by the

interaction of MMTS and DTNB with the receptors, respectively rather

than the oxidation itself, inhibit the binding of the complexes to

DNA-cellulose. More recently, Type II receptors were shown to bind

nuclear matrix (a proteinaceous structural framework implicated in many

nuclear functions, for detail see Chapter VI) through intermolecular

disulfide bond formation (Kaufmann et al., 1986). Sodium

dodocylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of the rat

hepatic Type II receptors extracted from nuclei isolated from animals

injected with [ H]TA in the absence of sulfhydryl-blocking (e.g.,

iodoacetamide) and cross-linking reagents yielded steroid-receptor

complexes that were disulfide-cross-linked to the nuclear matrix.

Conversly, isolation of the nuclei in presence of iodoacetamide,

revealed no receptor-nuclear matrix complex formation (Kaufmann et al.,

1986). These results provide a more direct evidence for the involvement

of sulfydryl groups in the binding of transformed steroid-receptor

complexes to nuclear components (for further discussion see Chapter VI).

Recently, structural analysis of the amino acid sequences of a number of

steroid receptors revealed the presence of a cysteine rich region in the

DNA-binding domain these receptors (e.g., Danielsen et al., 1987).











In contrast to our understanding of the factors regulating Type II

receptor binding activity, our knowledge concerning possible up- and

down-regulatory factors affecting Type I receptors remains scanty. As

discussed above, we have demonstrated several in vitro chemical

differences between the brain cytosolic Type I and Type II receptors

which may play an important role in the in vivo regulation of these

receptors. For example, although charcoal-extraction of steroid-free

brain cytosol in HEPES buffer lacking DTT causes a dramatic reduction in

Type II binding capacity (possibly by removal of endogenous

thioredoxin), this treatment has no effect on the binding capacity of

Type I receptors (Emadian et al., 1986; also see Chapter V). This

finding suggests that the labile thiol-disulfide exchange mechanism

described for Type II receptors may not be involved in the regulation of

Type I receptor binding capacity. In an effort to shed more light on

the role of possible factors and mechanisms involved in the in vitro

and/or in vivo regulation of Type I receptors, an investigation of

possible involvement of sulfhydryl groups in the ligand- and DNA-binding

capacity of these receptors was conducted in this dissertation (see

Chapters V and VI).

In Vitro Analysis of Type I Receptors by Radioreceptor Assay

A clear understanding of the physiological and behavioral

consequences of steroid hormone action requires an unambiguous

characterization of the receptors through which such actions presumably

are mediated. Our knowledge of the properties of Type I receptors as

well as the factors involved in the regulation of these receptors may

indeed lead to further discoveries regarding functional consequences of

these steroids. For example, the recognition of a specific








37

mineralocorticoid receptor led to the discovery of spironolactones

(mineralocorticoid antagonists), which have proven invaluable tools

clinically (as effective antihypertensive agents) and in basic sciences

(see above). Although numerous studies investigating the properties of

steroid hormone receptors have appeared in the literature within the

past two decades, the volume of research on the properties of Type I

receptors has been comparatively minimal (Liao et al., 1983). This is

partially due to the fact that in crude preparations ALDO and other

readily available synthetic mineralocorticoids bind to Type II receptors

with appreciable affinity (Rousseau et al., 1972; Funder et al., 1973a;

Emadian et al., 1986; Emadian and Luttge, unpublished; also see Chapter

V). This indiscriminatory binding of ALDO to Type II receptors can

produce confounding results thus hampering an independent analysis of

Type I receptors. Although the use of mathematical models has proven

effective in determination of binding parameters for Type I receptors

(e.g., Claire et al., 1978; Claire et al., 1985; Emadian et al., 1986),

failure to eliminate Type II-binding interference has led a number of

investigators reach erroneous conclusions in their analysis of Type I

receptor properties (Grekin and Sider, 1980; Marver et al., 1972).

Recently, however, the introduction of very specific synthetic Type II

ligands (e.g., RU 26988, RU 28362 and RU 38486) has made the elimination

of Type II-binding contribution in crude cytosolic preparations

possible. When used in conjunction with radiolabeled ALDO, a molar

excess of the RU compound can saturate all Type II binding sites with

little or no cross-reactivity with Type I receptors, hence leaving these

latter sites intact for ALDO binding. This method has been utilized

successfully by this and a number of other laboratories (Moguilewsky and










Raynaud, 1980; Veldhuis et al., 1982b; Gomez-Sanchez and Gomez-Sanchez,

1983; Wrange and Yu, 1983; Emadian, et al., 1986).

In addition to these methods, Veldhuis et al. (1982b) reported that

they were able to eliminate non-linearities observed in the Scatchard

plots for ALDO binding by co-incubating hippocampal cytosol with

unlabeled CORT. Although the Scatchard plot of ALDO binding in the

presence of a 100-fold concentration of RU 26988 appears comparable to

that obtained in the presence of a 0.6-fold concentration of CORT

(Veldhuis et al., 1982b), in light of the fact that CORT has a higher

affinity for hippocampal Type I receptors (see below), a lower value of

equilibrium maximal binding in the presence of CORT is consistent with

the interaction of CORT with Type I receptors (for detail see below).

To eliminate the confounding effects of ALDO interaction with Type II

receptors, unless otherwise stated, the characterization of Type I

receptors in this dissertation was accomplished in the presence of a

500-fold molar excess of RU 26988, a concentration determined to

eliminate all Type II binding interference (Emadian et al., 1986) (for

detail see Chapter II).

Another pertinent issue in the in vitro analysis of steroid

receptors in crude cytosolic preparations, in general, is possible

contamination and thus binding contribution by low affinity and high

capacity "non-specific binders" (of cytosolic, membrane or of plasma

origin). It is well established that ALDO binds with appreciable

affinity, especially at lower temperatures, to corticosteroid binding

globulin (CBG) (Sandberg et al., 1960; Daughaday et al., 1960; Meyer et

al., 1961; Davidson et al., 1962; Zager et al., 1976; Zipser et al.,

1980), and with a lower affinity to a,-acid glycoprotein and albumin








39

(Chen et al., 1961; Daughaday et al., 1960; Ganguly and Westphal, 1968).

Moreover, using competitive adsorption and gel filtration techniques,

Richardson and co-workers (1977) identified two other plasma

ALDO-binding proteins distinct from CBG, albumin and a -acid

glycoprotein in human plasma. This has been confirmed by Katayama and

Yamaji (1982) who reported that this "ALDO binder" had an apparent

association constant for ALDO one order of magnitude higher than that

for CBG. Although there is no evidence that these plasma binders are

present in rodents, because 1) ligand binding assays are usually

performed at lower temperatures and 2) corticosteroid-binding globulin

(CBG) shows appreciable affinity for ALDO at lower temperatures, an

extensive in situ transcardiac perfusion of brains prior to cytosol

preparation will be performed. This will assure the elimination of

possible contribution of these binders in our in vitro

characterization of Type I receptors. Note that this perfusion was

deemed sufficient in eliminating these binders, since co-incubation of

whole brain cytosol with [3H]ALDO and cortiuic acid (a synthetic

17a-acid derivitive of cortisol that specifically binds CBG, see Alexis

et al., 1983; Little, 1983; Sheppard and Funder, 1986) produced specific

Type I receptor binding similar in magnitude to that found in the

presence of [3H]ADLO alone.

In addition to the above plasma binders, the presence of the

extravascular corticosterone binder (with identical physicochemical and

immunoreactive properties to the plasma CBG) has been identified in the

uterus (Milgrom and Baulieu, 1970; Guerigian et al., 1974; Al-Khouri and

Greenstein, 1980), liver (Koblinsky et al., 1972; Litwack et al., 1973;

Amaral et al., 1974; Suyemitsu and Terayama, 1975; Weiser et al., 1979),








40

kidney (Funder et al., 1973b; Feldman et al., 1973; Strum et al., 1975;

Weiser et al., 1979), pituitary and brain (Koch et al., 1976; De Kloet

and McEwen, 1976a & b; Al-Khouri and Greenstein, 1980; Gray and Luttge,

1982), and a number of other tissues (Mayer et al., 1975; Giannopoulos,

1976; Werthamer et al., 1973) (for review see Siiteri et al., 1982).

The order of steroid binding specificity for these CBG-like binders

determined by competition assays appears to be CORT > DOC >> ALDO >>

DEX. Although it has been shown conclusively that the synthesis of

CBG-like glucocorticoid binders (by convention, referred to as Type III

receptors) occurs in the liver (Perrot-Applanat and Milgrom, 1979;

Weiser et al., 1979), the possible functional roles and the origin of

these binders in other tissues have been the subject of considerable

controversy. Since a comprehensive review of the primary references

surrounding this issue is outside the scope of this dissertation, only

the recent findings relevant to the in vitro characterization of

Type I receptors will be reviewed.

More than a decade ago, it was established that in kidney there

were two distinct classes of corticosteroid binding proteins

representing the Type I and Type II receptors (Funder et al., 1973b).

It was further claimed that the hierarchy of steroid binding specificity

for Type I receptors were ALDO > DOC > CORT (Funder et al., 1973b).

Later, the presence of specific Type I receptors in a number of other

tissues including the CNS were reported (Anderson and Fanestil, 1976;

Moguilewsky and Raynaud, 1980; Lan et al., 1981). However, it appeared

that ALDO, DOC and CORT had equivalent affinity for these sites. The

results from these in vitro competition assays were taken as

evidence for possible in vivo occupancy of Type I receptors by











glucocorticoids in the rat kidney and brain (Lan et al., 1982; Veldhuis

et al., 1982b). In a separate study, Beaumont and Fanestil (1983)

reported that the steroid binding specificity Type I receptors in rat

brain could be altered by different experimental manipulations. These

investigators found that after elimination of Type II binding

interference (in the presence of a molar excess of RU 26988) and plasma

CBG (and possibly Type III receptors), the order of steroid competition

for Type I sites (occupied with [ H]ALDO) was DOC > fludrocortisone >

CORT > ALDO > progesterone > DEX. However, the addition of small

quantities of dialyzed serum to the above preparations yielded a

competition hierarchy similar to that described for kidney, i.e., ALDO >

DOC > CORT. Several independent confirmations of the above observations

were reported for the rat kidney (Krozowski and Funder, 1983; Stephenson

et al., 1984b) and hippocampus (Krozowski and Funder, 1983; De Kloet et

al., 1984). In conclusion, it is suggested that CBG and CBG-like

binders, in addition to affecting the metabolic clearance rate of the

steriods to which they bind with high affinity (for review see Siiteri

et al., 1982), may act in vivo as sequestration sites or "sinks" for

corticosterone thus rendering the Type I sites available for ALDO

binding (Funder et al., 1973b; Krozowski and Funder, 1983; Stephenson et

al., 1984; Funder, 1986). It is further suggested that the hippocampal

Type I receptors are physiological CORT binders and may be involved in a

tonic (permissive) influence on brain function with septo-hippocampal

complex as a primary target (De Kloet et al., 1984; Reul and De Kloet,

1985; Funder, 1986; Magarinos et al., 1986; Reul and De Kloet, 1986).

Although the sequestration theory is an attractive hypothesis, further

experiments investigating the tissue availability of CORT (that is not











bound to CBG, other plasma binders and Type III receptors) and ALDO

under physiological conditions is required to establish this theory. In

this regard, preliminary studies suggest that the serum levels of ALDO

and CORT under "normal" conditions leads to nuclear uptake of both

hormones in the hippocampus as measured by radioimmunoassay (Yongue and

Roy, 1984). Recently, McEwen et al. (1986b) showed that, in

adrenalectomized rats, only "stress level" concentrations (a 500-fold

molar excess) of CORT was able to block the nuclear uptake of

[3H]ALDO in limbic, but not circumventricular, areas. Further

functional relevence of these findings remain to be established.

A further caution regarding the results obtained from in vitro

competition assays also deserves mention: in the above experiments

cytosolic preparations were performed in the presence of molybdate and

DTT. Our own work has shown that although molybdate stabilizes both

Type I and Type II receptors, it causes a significant reduction in the

binding capacity of Type I receptors (Emadian et al., 1986; also see

Chapter IV). Furthermore, sulfhydryl reducing reagents such as DTT,

while having no effect on the binding of ALDO to Type I receptors,

result in an apparent increase in the equilibrium dissociation constant

of CORT from CBG (Westphal, 1983) and very likely from Type III

receptors. Therefore, under conditions that provide optimal stability

for all of these proteins, one may find different steroid-binding

hierarchy from those discussed above.
















CHAPTER II
EXPERIMENTAL PROCEDURES

In order to avoid repetition, in this chapter a general scheme of

the materials and experimental procedures shared by the studies

described in Chapters III through VI will be presented. Materials and

methods that are unique to a given chapter will be described in that

particular chapter.

Isotopes, Steroids and Other Chemicals

The radiolabeled steroids; i.e., [6,7-3H]Triamcinolone acetonide,

[ H]TA, 9t-fluoro-ll, 16a,17,21-tetra-ol-pregna-1,4-diene-3,20-dione

(specific activity, SA 43.7 Ci/mmol); [6,7-3H]dexamethasone, [ H]DEX,

9c-fluoro-16a-methylprednisolone (SA 37.3-44.1 Ci/mmol) and

[1,2,6,7-3Hjaldosterone, [3H]ALDO, 4-pregnen-ll,21-diol-3,18,

20-trione (SA = 71.5-82.0 Ci/mmol) were purchased from New England

Nuclear. These radiolabeled steroids were checked for purity by

thin-layer silica gel chromatography using a chloroform-methanol (9:1,

volume:volume) solvent system. [1H]TA, [IH]DEX, [ H]ALDO,

Sephadex G-25 (bead size 20-80 u) and sodium molybdate were obtained

from Sigma. HEPES buffer [4(2-hydroxyethyl)-l-peperazineethanesulfonic

acid] and dithiothreitol (DTT) were obtained from Research Organic. RU

26988 [118,17B-dihydroxy-17a-(l-propionyl)-androsta-1,4,6-triene-3-one)]

and RU 26752 [3' (3-oxo-7a-propyl-178-hydroxy-4-androstene) propionic

acid lactone] were kindly supplied by Roussel-Uclaf (France). Prorenone








44

[3(178-hydroxy-68,,78-methylene-3-oxo-4-androsten-17a-yl) propionic acid

y-lactone] was a gift from Searle Co. All other chemicals used in this

dissertation were reagent grade quality.

Experimental Animals

Adult female CD-1 mice (Charles River Laboratories) were subjected

to a combined bilateral adrenalectomy and ovariectomy under barbiturate

anesthesia 3-6 days prior to use [a period shown to result in maximal

up-regulation of Type I receptors (Luttge and Rupp, unpublished

results)] in order to remove known sources of endogenous steroids. They

were then maintained on Purina Rodent Chow and 0.9% NaCl. On the day of

experiment, mice were anesthetized with ether and perfused slowly

through the heart with 15-20 ml ice-cold HEPES-buffered saline

(isotonic, pH 7.60).

Table 2-1

Buffers

20 mM 2 mM
BUFFER HEPES DTT Na MoO


A +

D + +

M + -2 mM

M20 + -20 mM

DM + + 2 mM

DM20 + + 20 mM



The cytosolic concentration of the components) in Buffers D, M,

M20, DM, DM20 as well as other cytosolic components (e.g.,

300 mM KC1; 10% mono- and polyhydric compounds) were obtained by a 9:1











dilution of cytosol with stock buffers containing a 10-fold

concentration of the desired componentss. Under such circumstances,

the control group was diluted accordingly with an equivalent volume of

Buffer A. Buffers were adjusted to pH 7.60 at 0-20C.

Cytosol Preparation

Brains were homogenized (2 x 10 strokes) at 1000 RPM in two volumes

of ice-cold Buffer A or Buffer A containing one or more of the desired

components (see specific experiments). A glass homogenizer with a

teflon pestle milled to a clearance of 0.125 mm on the radius was used

to minimize the rupture of brain cell nuclei. The crude homogenate was

centrifuged at 100K x g for 20 min and the resulting supernatant

recentrifuged at 100K x g for an additional 60 min to yield cytosol.

During this and all other procedures (except "aging" and column

chromatography) attempts were made to maintain the cytosol at 00C.

Aging

In order to examine thermal stability of unoccupied Type I and Type

II receptors, cytosol was incubated for various periods of time at 0

and/or 220C. Following such treatment, "aging" was quenched by

incubating the cytosol with steroid(s) for 24-48 h at 00C for specific

binding determination (see below).

Hydrophobic Interaction Chromatography.

Pently-Agarose columns (0.7 x 21.5 cm) were used to compare the

surface hydrophobic properties of occupied Type I versus Type II

(Chapter III) as well as activated versus unactivated Type I receptors

(Chapter VI). In such experiments, cytosol in Buffer DM20 was

incubated with radiolabeled ligand for Type I or Type II receptors for

48 h at O0C (see "Steroid Binding Determination"). Following incubation








46

with steroid(s), salt- and heat-induced activation (see Chapter VI),

bound-free separation on Sephadex G-25 (0.6 x 14.0 cm) columns were

performed as described previously (Luttge et al., 1984c). Subsequently,

0.8 ml [ H]steroid-bound macromolecular fraction eluted from the

G-25 column was loaded on pently-Agarose columns pre-equilibrated in

buffer D containing 50 mM molybdate and 600 mM KCI. Using this latter

buffer, 20-27 consecutive 0.8 ml fractions were collected directly into

scintillation vials for measurements of radioactivity (see below). In

some experiments, a second bound-free separation on G-25 columns were

performed on the "peak" fractions eluted from the pentyl agarose columns

to assure that the radioactivity eluted was indeed macromolecular-bound

in nature. All chromatographic procedures were performed at 4-60C.

Steroid Binding Determination

After appropriate treatment of cytosol, 0.24-4.0 ml aliquots were

added to pre-cooled glass tubes containing the desired steroid(s) (dried

under N2). The tubes were then allowed to incubate for 24-48 h in

ice. In single saturation dose studies, B determinations for Type

I receptors were performed in the presence of 10-20 nM [3H]ALDO and

500-fold molar excess of RU 26988. For Type II receptors, 20 nM

[3H]DEX or [3H]TA was used as the ligand. In some experiments,

[3H]DEX incubation was in the presence of a 500-fold molar excess of

RU 26752, prorenone or RU 26988 (see Chapter V). Parallel incubation

tubes containing an additional 200-fold molar excess of [1H]ALDO,

[ H]DEX or [ H]TA were used for the determination of BNS for

Type I and Type II receptors, respectively. Equilibrium binding

parameters; i.e., Kd and BMAX were determined using Scatchard's

method (1949) as described previously (Emadian et al., 1986).








47

In all experiments, macromolecular-bound from free steroid was

separated at 4C using G-25 (0.6 x 14 cm) columns (for detail see Luttge

et al., 1984c). Bound fraction (0.8 ml) obtained from each column was

diluted with 5 ml Scinti-Verse fluor (Fisher Scientific) in a

scitillation vial, vortexed vigorously and the radioactivity determined

in a Packard Model 2425 liquid Scintillation spectrometer. All counts

per minute (CPM) data were converted to disintegration per minute (DPM)

by an external standard channels ratio measurement. The counting

efficiency for tritium averaged about 35%.

Cytosolic Protein Concentration

The concentration of protein in cytosol preparations was determined

using the method described by Lowry et al. (1951). In crude

preparations, protein concentration ranged from 6.2 to 7.3 mg/ml

cytosol, whereas the values in gel filtered and 2-fold diluted

preparations (see Chapter IV) averaged about 3.3 and 3.6 mg/ml cytosol,

respectively.
















CHAPTER III
EFFECTS OF POLYHYDRIC AND MONOHYDRIC COMPOUNDS ON THE STABILITY OF
TYPE I RECEPTORS FOR ADRENOCORTICOSTEROID HORMONES IN BRAIN CYTOSOL


Summary

We have shown previously that unoccupied Type I receptors for

adrenal steroids in brain cytosol lose their ability to bind

[ 3HALDO in a time- and temperature-dependent manner (Emadian

et al., 1986). Based on reports that sugars and polyvalent alcohols are

capable of stabilizing a variety of globular proteins, we made an

attempt in this study to stabilize Type I receptors by including

polyhydric compounds in our cytosol preparations. However, unlike our

expectations, the addition of 10% (weight:volume) ethylene glycol,

glycerol, erythritol, xylitol, ribitol or sorbitol to cytosol at 0C

failed to stabilize the receptors and in fact produced a slight

reduction in the binding of [ H]ALDO. The magnitude of this

reduction was greater when cytosol was "aged" for 2 h at 220C prior to

incubation with [ 3H]ALDO. In contrast to these results with

polyhydric compounds, the addition of 10% (weight:volume) ethanol to

cytosol maintained at 0C markedly increased the binding of [3H]ALDO

to Type I receptors. Under identical conditions, methanol slightly

increased, and propanol had no significant effect on the binding

capacity of these receptors. When cytosol was aged at 220C, however,

all of these monohydric compounds led to a marked loss in Type I

receptor binding capacity. An investigation of various doses of ethanol

at 0C on the subsequent binding of [3H]ALDO to Type I receptors










yielded an inverse U-shaped curve with 10% ethanol producing the highest

level of specific binding, as reflected by an increase in maximal

binding in Scatchard plots, and 40% ethanol producing a complete loss in

Type I receptor binding capacity. In view of the above findings and

proposed mechanisms of action of mono- and polyhydric compounds, it is

suggested that Type I receptors in an aqueous cytosolic environment

undergo spontaneous conformational changes which promote the infolding

of hydrophobic steroid binding site(s) on the receptor surface, thus

leading to the observed reduction in Type I receptor binding capacity.

Introduction

Although receptors for steroid hormones appear to be relatively

stable under in vivo conditions (Raaka and Samuels, 1983), these

macromolecules are rendered labile under cell-free conditions. This

increased ability is particularly apparent when working with unoccupied

receptors (Puca et al., 1971; Pratt and Ishii, 1972; Bell and Munck,

1973; Rafestin-Oblin et al., 1977; Emadian et al., 1986), and it can be

a serious problem when experimental conditions necessitate that the

unoccupied receptors must remain stable for extended periods of time

(e.g., during the determination of their hydrodynamic and equilibrium

binding parameters). In previous work with unoccupied and occupied

Type I and Type II receptors for adrenal steroids in brain cytosol we

have systematically examined the stabilizing effectiveness of a number

of factors including temperature, ionic strength, chelators, molybdate,

dithiothreitol, dextran-coated charcoal adsorption, etc. (Gray, 1982;

Densmore et al., 1984a; Densmore et al., 1984b; Densmore and Luttge,

1985; Densmore et al., 1986a; Densmore et al., 1986b; Emadian and

Luttge, 1986; Emadian et al., 1986); however, even with our best buffer








50

formulations, we still observe a time- and temperature-dependent loss of

unoccupied Type I receptor binding capacity (Emadian et al., 1986).

Work in other laboratories has shown that polyhydric compounds (e.g.,

sugars and polyvalent alcohols) can stabilize a wide range of globular

proteins, including a number of enzymes and receptors, against

time-dependent thermal degradation (Feil et al., 1972; Rafestin-Oblin

et al., 1977; Back et al., 1979; Ogle, 1980; Gekko and Timasheff, 1981a

& b; Ogle, 1983). In view of these successes, the experiments presented

here examine the effects of a number of mono- and polyhydric compounds

on the stability and binding capacity of unoccupied Type I receptors for

S3H]ALDO in brain cytosol. The results from these studies suggest

that spontaneous conformational changes that occur as a consequence of

interaction between these receptors and their aqueous cytosolic

environment may ultimately lead to the infolding of hydrophobic steroid

binding site(s) on the surface of Type I receptors.

Materials and Methods

Chemicals. Reagent grade poly- and monohydric compounds were

purchased from Sigma.

Buffers. The final concentration of mono- and polyhydric compounds

in cytosol was obtained by adding concentrated stock buffers containing

each compound to the cytosol to yield the desired concentration of that

compound. The pH is all buffers was adjusted to 7.60 at 0-20C. All

other materials and methods were as described in Chapter II.

Results

Fig. 3-1 shows that a 10% (weight:volume) final concentration of

di- (ethylene glycol, EG), tri- (glycerol, G), tetra- (erythritol, E),

penta- xylitoll, X and ribitol, R) or hexavalent (sorbitol, S) alcohols










Figure 3-1. Effects of Polyhydric Compounds on the Stability of
Unoccupied Type I Receptors in Whole Brain Cytosol. Cytosol

prepared in Buffer A was either diluted with additional Buffer A (No

Poly-OH) or adjusted to 10% (weight:volume) ethylene glycol (EG),

glycerol (G), erythritol (E), Xylitol (X), ribitol (R) and sorbitol (S)

using a concentrated HEPES-buffered solution of each compound.

Subsequently, the cytosol was divided into two parts: one part was

incubated immediately with 10 nM [3H]ALDO plus 5 uM RU 26988 and

either with (BNS) or without (B ) 2 uM [IH]ALDO for 24 h OC

(0-h open bars). The second part was aged either from 2 to 48 h at 0C

(2-, 24- and 48-h open bars) or for 2 h at 220C (closed bars) prior to

incubation with the steroids. At the end of the incubation period,

macromolecular-bound from free steroids were separated on Sephadex G-25

columns and specific binding was determined as described in Chapter II.

The number of independent replicate experiments is indicated near the

base of each bar. All data presented are based on the percent of the

non-aged, "No Poly-OH" control (*) group + SEM. 100% = 28 fmole/mg

protein.







































No Poly-OH EG


E X. R S


100



C 80
o
0

4 60

c
2 40.
6)











to unlabeled brain cytosol produced a small, but significant (p < 0.05),

reduction in the binding of 3 H]ALDO to Type I receptors for adrenal

steroids. Aging unlabeled cytosol at 0C in the presence of any of

these polyhydric compounds did not attenuate, or exacerbate, the loss in

Type I receptor binding capacity seen in the control cytosol (No

Poly-OH). However, when cytosol was aged for 2 h at 220C, the reduction

in the subsequent binding of [ H]ALDO to Type I receptors in the

presence of polyhydric compounds was nearly twice as great as in their

absence. Since a 10% concentration of glycerol has also been routinely

included by many investigators in buffers used for Type II receptors, we

examined the effects this compound on [3H]DEX binding. As shown in

Fig. 3-2, the addition of 10% glycerol to brain cytosol prior to

incubation with [3H]DEX fails to protect the time- and temperature-

dependent loss in Type II receptor binding capacity. It is noteworthy

that, in contrast to the results obtained with Type I receptors (Fig.

3-1), glycerol does not exacerbate the loss in binding capacity when

unoccupied Type II receptors are aged for 2 h at 220C.

Because polyhydric compounds failed to stabilize Type I (or

Type II) receptor binding capacity, we next investigated the effects of

monohydric alcohols on the binding of [ H]ALDO to Type I receptors

(Fig. 3-3). In contrast to the data with polyhydric compounds, when a

10% (weight:volume) final concentration of methanol, and especially

ethanol, but not propanol, was added to cytosol immediately before the

steroids, we observed a significant increase in [ H]ALDO binding to

Type I receptors. If cytosol was aged for 2 h at 0C prior to the

addition of steroids, ethanol still produced an increase in the

subsequent binding of [ H]ALDO to Type I receptors, whereas methanol










Figure 3-2. Effects of 10% Glycerol on the Stability of Unoccupied
Type II Receptors in Whole Brain Cytosol. Cytosol prepared in

Buffer D was either diluted with additional Buffer D (No Poly-OH) or

adjusted to 10% (weight:volume) glycerol using a concentrated

HEPES-buffered solution of glycerol containing 2 mM DTT (G).

Subsequently, the cytosol was divided into two parts: one part was

incubated immediately with 20 nM [3H]DEX and either with (BNS)

or without (BT) 4 uM [1H]DEX for 24 h at 0C (0-h open bars).

The second part was aged either from 2 to 48 h at 0OC (2-, 24- and 48-h

open bars) or for 2 h at 220C (closed bars) prior to incubation with the

ligand. At the end of the incubation period, macromolecular-bound from

free steroid was separated on Sephadex G-25 columns and specific binding

was determined as described in Chapter II. The results are the mean

from 2 independent replicate experiments. All data presented are based

on the percent of the non-aged, "No Poly-OH" control (*) group.

100% = 300 fmole/mg protein.











100 -
*
-80-
C
0

t 60-
C
a)
S40-
20..
LO40

20- oCj 1 0 oN cM1

I


No Poly- OH










Figure 3-3. Effects of Monohydric Compounds on the Stability of
Unoccupied Type I Receptors in Whole Brain Cytosol. Cytosol

prepared in Buffer A was either diluted with additional Buffer A (No

Alcohol) or adjusted to 10% (weight:volume) methanol (Mt), ethanol (Et)

and propanol (P) using a concentrated HEPES-buffered solution of each

compound. Subsequently, the cytosol was divided into two parts: one

part incubated immediately with 10 nM [3H]ALDO plus 5 uM RU 26988

and either with (BNS) or without (BT) 2 uM [ H]ALDO for 24 h

at OOC (0-h open bars). The second part was aged for 2 h at 0 or 22C

(2-h open and closed bars, respectively) prior to incubation with the

steroids. At the end of the incubation period, macromolecular-bound

from free steroids were separated on Sephadex G-25 columns and specific

binding was determined as described in Chapter II. All data presented

are based on the percent of the non-aged, "No Alcohol" control (*) group

expressed as the mean + SEM from 6 independent replicate

experiments. 100% = 18 fmole/mg protein.























0
*
6 0-
'-


aL 40- o


20-




No Alcohol


Mt Et








58

had no effect and propanol produced a clear reduction in receptor

binding capacity. When cytosol preparations were aged for 2 h at 220C,

all three alcohols greatly increased the loss in Type I receptor binding

capacity.

An investigation of the effects of various concentrations of

ethanol on the binding of [ H]ALDO to cytosolic Type I receptors

revealed an inverse U-shaped curve with 10% ethanol producing the

greatest binding (120% of control) and 40% ethanol leading to a complete

inactivation of Type I receptors (1% of control) (Fig. 3-4).

Furthermore, a Scatchard (1949) analysis of the equilibrium binding data

revealed that the 10% concentration produced a 46% increase in the

K and a 23% increase in the B of [3H]ALDO binding to
max
Type I receptors in brain cytosol (Fig. 3-5).

Discussion

Proposed Mechanism of Action of Poly- and Monohydric Compounds.

For proteins in an aqueous environment, the non-covalent weak

intramolecular interactions (i.e., electrostatic forces, hydrogen

bonding and Van der Waals forces), as well as the interactions between

protein molecules and their surrounding solvent (i.e., hydrophobic

interactions), act in concert to stabilize the functionally important

tertiary structure of proteins. In some proteins, additional stability

may result from cross-linking, metal complex formation and binding to

other ions and cofactors; however, it is often held that the hydrophobic

interactions, a consequence of net unfavorable interactions between

solvent (i.e., water molecules) and non-polar residues in the protein,

play the most important role in stabilizing proteins (e.g., Creighton,

1984; Fersht, 1985). Therefore, when proteins are extracted from the










Figure 3-4. Dose-Response Analysis of the Effects of Ethanol on the
Binding of ALDO to Type I Receptors in Whole Brain Cytosol. Cytosol

prepared in Buffer A was either diluted with additional Buffer A

(control, *) or adjusted to 1 to 40% (weight:volume) ethanol using a

concentrated HEPES-buffered solution of ethanol. Subsequently, all

groups were incubated with 10 nM [3H]ALDO plus 5 uM RU 26988 and

either with (BNS) or without (BT) 2 uM [IH]ALDO for 24 h at

00C. At the end of the incubation period, macromolecular-bound from

free steroids were separated on Sephadex G-25 columns and specific

binding was determined as described in Chapter II. All data presented

are based on the percent of 0% ethanol control (*) group expressed as

the mean + SEM from 3-12 independent replicate experiments.

100% = 23 fmole/mg protein.














120


100


80


60-


40-


20-


Ethanol Concentration (% w:v)










Figure 3-5. Scatchard Analysis of the Effects of 10% Ethanol on the
Equilibrium Binding Parameters of ALDO Binding to Type I Receptors
in Whole Brain Cytosol. Cytosol prepared in Buffer A was either

diluted with additional Buffer A (open circles) or adjusted to 10%

(weight:volume) ethanol using a concentrated HEPES-buffered solution of

ethanol (closed circles). Subsequently, 0.25 ml aliquots from each group

was incubated with 0.5 to 40 nM [3H]ALDO plus 500-fold molar excess

of RU 26988 and either with (BNS) or without (B ) 200-fold molar

excess of [ H]ALDO for 24 h at 0C. At the end of the incubation

period, macromolecular-bound from free steroids were separated on

Sephadex G-25 columns and specific binding was determined as described

in Chapter II. The Scatchard plot shown is representative from 4

independent replicate experiments. Binding parameters were calculated

using least square linear regression. Open circles: B =
max
21 fmole/mg protein, Kd = 3.0 x 10-10 M. Closed circles:

max = 26 fmole/mg protein, Kd = 4.4 x 10-10 M.

















0.20
LL.

a-

0.10






0.05 0.10 0.15 0.20
BSP (nM)











intracellular environment into a medium that is more aqueous than

cytoplasm, it is not uncommon for them to undergo spontaneous

conformational changes leading to their functional inactivation. To

protect proteins against such in vitro denaturation, various

strategies have been utilized to reduce unfavorable solvent-solute

interactions.

In a series of studies by Gerlsma (1968 & 1970) and Gerlsma and

Stuur (1972 & 1974), polyhydric compounds were shown to increase the

melting temperature (T ) of chymotrypsinogen A, ribonuclease and

lysozyme and hence stabilize the native conformation of these proteins

at a higher temperature. In contrast, monohydric alcohols were found to

destabilize these proteins (i.e., lower the T ). On a molecule per

molecule basis, the increase in the T produced by polyhydric
m
compounds was shown to be related directly to the number of hydroxyl

groups per molecule of the polyhydric compound, whereas the extent of

destabilization produced by monohydric compounds were shown to depend on

the length of their alkyl chain. On the basis of these results, Gerlsma

and Stuur (1972 & 1974) proposed that polyhydric compounds stabilize

proteins by decreasing the hydrogen bond rupturing capacity of the

solvent, whereas monohydric compounds augment such solvent-induced

hydrogen bond rupturing capacity. This proposed mechanism is somewhat

different from that set forth earlier by McDuffie et al. (1962), who

maintained that polyhydric compounds stabilize proteins by inducing the

tetrahedral conformation of water molecules at higher temperatures

which, in turn, increase the hydrophobic interactions. In support of

this hypothesis, Oakenfull & Fenwick (1979) showed that hydrophobic










interactions in aqueous-organic environments are maximized when the

three-dimensional hydrogen-bonded structure of water is most developed.

In a recent quantitative analysis by scanning calorimetric

technique, Back et al. (1979) have shown that sucrose and glycerol

strengthen pairwise hydrophobic interactions in aqueous solutions of

ovalbumin, lysozyme, a-chymotrypsinogen and conalbumin; however, both of

these polyhydric compounds were also shown to reduce the driving force

for transfer of a hydrophobic group from an aqueous to a nonpolar

environment. Other workers have argued that glycerol, the most common

polyhydric compound used to stabilize the activity and native structure

of enzymes and proteins (Feil et al., 1972; Rafestin-Oblin et al., 1977;

Ogle, 1980; Dias et al., 1981; Korge and Timpmann, 1983; Livesey et al.,

1983; Ogle, 1983; Ramaley and Vasantha, 1983; Fojo et al., 1985),

increases the hydrophobic interactions in chymotrypsinogen A,

a-chymotrypsin, y-lactoglobulin and ribonuclease A by being

preferentially excluded from the immediate domain of these proteins

(Gekko and Timasheff, 1981a & b; Timasheff et al., 1976). Collectively,

these results suggest that polyhydric compounds tend to strengthen the

pairwise hydrophobic interactions in proteins by promoting the

three-dimensional hydrogen-bonded structure of water, hence stabilizing

the native, more folded conformation of proteins.

Possible Mode of Action of Poly- and Monohydric Compounds on
Unoccupied Type I receptors in Crude Cytosol.

Recently we reported that, in contrast to unoccupied Type II

adrenal steroid receptors, unoccupied Type I adrenal steroid receptors

lose their binding capacity for [3H]ALDO in crude brain cytosol in a

time- and temperature-dependent manner even in the presence of DTT,

molybdate and/or glycerol (Emadian et al., 1986). However, as discussed











above, since protein molecules are thought to exist in a dynamic state

in aqueous environments (Creighton, 1984), we now think that our earlier

findings with Type I receptors could be a reflection of the transient

fluctuations in the three-dimensional structure of the unoccupied

receptors in cytosol resulting in an exclusion of [3H]ALDO from its

specific hydrophobic binding site.

To arrive at this hypothesis, we first investigated the effects

polyhydric compounds with increasing ability to promote the tetrahedral

conformation of water molecules (i.e., di-, tri-, tetra-, penta-, and

hexavalent alcohols), hence an increased ability to enhance

intramolecular hydrophobic interactions, on the 3H]ALDO binding

capacity of unoccupied Type I receptors. As shown in Figure 3-1, all of

the polyhydric compounds tested failed to stabilize unoccupied Type I

receptors, in fact, all of them produced a slight reduction in the

specific binding of [3H]ALDO, especially when cytosol was aged for 2

h at 220C prior to binding determinations. These results suggest that

increases in the intramolecular hydrophobic interactions within the

unoccupied receptor, following the addition of polyhydric compounds to

aqueous cytosol preparations, may result in the preferential exclusion

of the steroid from its hydrophobic binding sites. If so, one would

expect that a reduction in intramolecular hydrophobic interactions

should increase the binding of [ H]ALDO to Type I receptors. It is

noteworthy that, compared to Type I receptors, inclusion of 10% glycerol

in our buffers did not affect the binding of ligand to Type II receptors

(Figure 3-2). The differential effects of glycerol on Type I and Type

II receptors strongly suggested that there might be differences in the

overall surface hydrophobicity of these receptors. In fact, a








66

comparison of the surface hydrophobic properties of these receptors by

glass fiber filter assay (Luttge et al., 1984) and hydrophobic

interaction chromatography (Figure 3-6) revealed a greater overall

surface hydrophobicity for Type I receptors.

To test the hypothesis that a reduction in hydrophobic interactions

may increase the binding of ligand to Type I receptors, we next

investigated the effects of adding monohydric alcohols to brain cytosol.

As predicted, when compared to the non-aged, "no-alcohol" group, a 10%

final concentration of ethanol in cytosol increased the binding of

[3H]ALDO to Type I receptors by about 20% (Figs. 3-3 and 3-4). This

effect was reflected in Scatchard plots by an increase in the B
max
(Fig. 3-5). In spite of this increase, the ethanol-induced reductions

in intramolecular hydrophobic interactions also appear to reduce the

stability of unoccupied Type I receptors when they are aged at 0C, and

especially at 220C. This increase in the liability of the unoccupied

receptor could easily account for the Kd shift seen in the Scatchard

plot (Fig. 3-5) (see Beck and Goren, 1983 for a theoretical explanation

of this phenomenon) and may be similar to the previously reported

temperature-dependent effects of ethanol on the stability of

ribonuclease and glutamate dehydrogenase (Shukuya and Schwert, 1960;

Brandts and Hunt, 1967).

In summary, the results described in this chapter provide indirect

support for the hypothesis that [3H]ALDO binding to Type I receptors

is dependent upon the exposure and stabilization of hydrophobic steroid

binding sites (and possibly other hydrophobic residues) on the surface

of the receptors. Since the comparatively stable in vivo

conformation of Type I receptors presumably allows for optimal ligand










Figure 3-6. Comparison of the Surface Hydrophobic Properties of
ALDO-Type I and DEX-Type II Receptor Complexes on Pentyl-Agarose
Columns. Cytosol prepared in buffer DM was incubated with either

10 nM [3H]ALDO plus 5 uM RU 26988 (to label Type I receptors, open

circles) or 20 nM [3H]DEX (to label Type II receptors, closed

circles) at 0C for 48 h for determination of B T. The

determinations of BNS for Type I and Type II receptors were

performed in the presence of 2 uM [1H]ALDO and 4 uM [1HIDEX,

respectively. Following separation of macromolecular-bound from free

steroids on Sephadex G-25 columns, hydrophobic interaction column

chromatography was performed as described in Chapter II. The data

presented are expressed as a percent of total BSp eluted from

pentyl-Agarose columns and are representative from 2 independent

replicate experiments.















30

c

m
ro



^20

o

4-
0

I-


5 10 15 20


Fraction Number








69

binding, it seems likely that these receptors undergo a spontaneous

solvent-induced folding upon extraction into the more aqueous cytosolic

environment which results in a reduction in binding capacity and

stability. Although, the mechanisms) and/or intracellular factors

producing the marked stability of Type I and other steroid receptors

in situ awaits further experiments, the present findings may provide

some clues as to the nature of their properties.
















CHAPTER IV
A NOVEL EFFECT OF MOLYBDATE ON THE BINDING OF [ H]ALDOSTERONE
TO GEL FILTERED TYPE I RECEPTORS IN BRAIN CYTOSOL


Summary

Recently, we reported that the addition of sodium molybdate to

crude cytosol at 0C results in a dose-dependent reduction in the

binding of [3H]ALDO to Type I receptors for adrenal steroids

(Emadian et al., 1986). In the experiments outlined here, we show that

Sephadex G-25 gel filtration of whole brain cytosol from

adrenalectomized and ovariectomized CD-1 female mice in the presence of

2 mM molybdate results in a 30-50% increase in the specific binding of

10 nM [3H]ALDO to Type I receptors above the level of [3H]ALDO

binding seen in crude cytosol prepared in 20 mM HEPES buffer (control

group). All specific binding determinations were performed in the

presence of 5 uM RU 26988 -- a specific synthetic glucocorticoid known

to block I3H]ALDO binding to Type II (glucocorticoid) receptors. In

Scatchard plots, this molybdate- and gel filtration-induced increase in

specific [3H]ALDO binding was reflected as a 37% increase in maximal

binding (Bmax) with no change in the equilibrium dissociation

constant (Kd). In contrast, when gel filtration was performed in

the absence of molybdate, there was a marked reduction in the subsequent

specific binding of [3HIALDO to Type I receptors. In Scatchard

plots, this latter effect was reflected as a 62% reduction in the

B max and a 2-fold increase in the Kd when compared to the
control group. The addition of 2 mM molybdate immediately following gel
control group. The addition of 2 mM molybdate immediately following gel








71

filtration yielded specific [3H]ALDO binding comparable to the

control group. Although the addition of 2 mM molybdate prevents the

loss in the binding capacity of unoccupied Type I receptors in crude

cytosol incubated at 220C, Type I receptors in gel filtered cytosol were

very unstable at 220C, and appeared insensitive to the protective

effects of 2, but not 10-100 mM, molybdate. Two mM dithiothreitol (DTT,

a potent sulfhydryl reducing reagent) was unable to prevent the gel

filtration-induced loss in [3H]ALDO binding. Moreover, in gel

filtration experiments, the data obtained in the presence of 2 mM DTT

plus 2 mM molybdate did not differ from those obtained in the presence

of 2 mM molybdate alone. The effect of molybdate on Type I receptors

caused by its presence during gel filtration of cytosol is clearly

different from that seen with these receptors in crude cytosol

preparations, as well as those reported in the literature for other

steroid receptors. Possible mechanisms of action of molybdate on

unoccupied Type I receptors in crude and gel filtered cytosol are

discussed.

Introduction

In a series of reports from Pratt's laboratory it was shown that

the addition of sodium molybdate to crude cytosol preparations prevents

the heat-induced loss in the binding capacity of unoccupied

glucocorticoid (Type II) receptors (Nielsen et al., 1977b & c).

Subsequently, Toft and Nishigori (1979) showed that this transition

element oxyanion reversibly blocks the temperature-dependent

transformation (activation) of progesterone receptors to a DNA-binding

state. Later, numerous reports have shown similar stabilizing effects

of molybdate on other steroid receptors (Nishigori and Toft, 1980;








72

Shyamala and Leonard, 1980; Miller et al., 1981) and proteins (Defay et

al., 1984; Nakada et al., 1985; Denison et al., 1986). In contrast to

these findings, recently we have reported that although molybdate

prevents the heat-induced inactivation of Type I receptors, at 0C, this

compound leads to a dose-dependent reduction in the subsequent binding

of [ 3H]ALDO in crude whole brain cytosol preparations (Emadian et

al., 1986). In the experiments below, we will describe yet another

novel effect of molybdate on Type I receptors in Sephadex G-25 gel

filtered whole brain cytosol.

Materials and Methods

Chemicals. All chemicals used in the experiments in this chapter

were reagent grade quality.

Buffers. The cytosolic concentration of molybdate and DTT were

obtained by a 9:1 dilution of cytosol with stock buffers containing a

10-fold concentration of the desired componentss. Under such

circumstances, the control group was diluted accordingly with an

equivalent volume of Buffer A. The final pH for all buffers was

adjusted to 7.60 at 0-20C.

Sephadex G-25 Gel Filtration. Gel filtered receptor preparations

were obtained by loading 0.5 ml aliquots of unlabeled crude cytosol on

Sephadex G-25 columns (0.6 x 14.0 cm) pre-equilibrated in appropriate

buffer(s) at 4-5C. The sample on each column was allowed to penetrate

the gel and the column was then washed with 1.0 ml buffer.

Subsequently, the macromolecular fraction was collected using an

additional 1.0 ml homologous buffer.









73

Results

Gel filtration of cytosol prepared in Buffer A through Sephadex

G-25 columns equilibratedd and eluted with homologous buffer) led to a

70% reduction in the subsequent specific binding of [ H]ALDO to Type

I receptors when compared to the control group (i.e., non-aged crude

cytosol prepared in Buffer A) (Fig. 4-1, A). There was no additional

loss in the binding capacity of Type I receptors when unlabeled crude or

gel filtered cytosol was aged for 2 h at 0C prior to incubation with

steroids; however, when the temperature during aging was raised to 220C,

we observed a 50% reduction in the subsequent Type I receptor binding to

[3H]ALDO (Fig. 4-1, A; 2 h aging). With the exception of a slight

increase in [3H]ALDO binding at 0C, similar results were obtained

when the above procedures were repeated in the presence of 2 mM DTT

(Fig. 4-1, D). This latter increase in BSp values was due to lower

BNS values in crude and especially gel filtered cytosol preparations

containing 2 mM DTT. It should also be noted that unoccupied Type I

receptors in gel filtered cytosol containing DTT appeared to be more

susceptible to the 220C aging-induced loss in ligand binding.

In agreement with our earlier report (Emadian et al., 1986), at

0C, the addition of 2 mM molybdate to unlabeled crude cytosol produced

a slight reduction in the binding of [3H]ALDO to Type I receptors

(Fig. 4-1, M). However, when the molybdate-containing cytosol was

subjected to Sephadex G-25 gel filtration, we observed a 50% increase in

the subsequent binding of [ H]ALDO to Type I receptors. Whereas

aging unlabeled crude or gel filtered cytosol in Buffer M at 0C for 2 h

had no significant effect on the measurable Type I receptor binding,

aging the unlabeled gel filtered, but not crude, cytosol at 220C for 2 h










Figure 4-1. Effects of Gel Filtration, Molybdate and/or
Dithiothreitol on the Binding Capacity and Stability of Unoccupied
Type I Receptors from Whole Brain Cytosol. Nine parts cytosol

prepared in Buffer A was diluted with either one part Buffer A (A) or

adjusted to 2 mM dithiothreitol (D), 2 mM molybdate (M) or 2 mM

dithiothreitol plus 2 mM molybdate (DM) using one part Buffer A

containing a 10-fold concentration of the respective componentss. Each

group either underwent the gel filtration procedure (for detail see

"Materials and Methods" section) at 4-50C (triangles) or was left

undisturbed at 4-50C (circles) during this process. Subsequently, a

portion of the crude or gel filtered cytosol from each group was either

incubated with 10 nM [3H]ALDO and 5 uM RU 26988 and minus (B )

or plus (BNS) 2 uM [IH]ALDO at 0C for 24 h for the

determination of specific binding or was aged at 0 (solid symbols) or

22C (open symbols) for 2 h prior to incubation with the steroids. Gel

filtration reduced the cytosolic protein concentration from 7.3 to

3.3 mg/ml. The data presented are from at least 5 independent replicate

experiments (except for the DM group) and are expressed as a percent of

control* (i.e., specific binding measured in non-aged crude cytosol

prepared in Buffer A) + standard error of the mean (SEM).

100% = 21 fmole/mg protein.










H D M DM


160

140-

120-

o *A
S100-

80-
o

i 60-

40-

20-



0 2 0 2 0 2 0 2
Duration of Aging (h)








76

led to a marked loss in the binding capacity of these receptors

(Fig. 4-1, M; 2 h aging). The data obtained in the presence of Buffer

DM did not differ appreciably from those described in the presence of

Buffer M with the exception, once again, of a slight exacerbation of the

heat-induced loss in the binding of [3H]ALDO to Type I receptors in

gel filtered cytosol following 2 h aging at 220C (Fig. 4-1, DM; 2 h

aging).

To determine the concentration of molybdate needed to produce the

highest increase in the binding of [ H]ALDO to Type I receptors

during gel filtration, we conducted a dose-response analysis of the

effects of this compound on Type I receptors during the gel filtration

procedure. Consistent with our previous observations (Emadian et al.,

1986), the addition of molybdate to HEPES buffer led to a dose-dependent

loss in [3HIALDO specific binding to Type I receptors in crude

cytosol at 0OC (Fig. 4-2). Aging the crude cytosol for 2 h at 0C

reduced the loss in [3H]ALDO binding seen with the two highest doses

of molybdate. In non-aged gel filtered cytosol, the maximal increase in

[3H]ALDO binding to Type I receptors occurred at 2 mM molybdate,

with 10-100 mM concentrations producing binding levels comparable to

those seen in the absence of molybdate (i.e., control group). It is

noteworthy that except for when molybdate was absent during gel

filtration, at all concentrations of this compound tested, the

measurable Type I receptor binding was higher in the non-aged gel

filtered cytosol than non-aged crude cytosol containing the same dose of

molybdate. Furthermore, at all concentrations of molybdate tested,

aging unlabeled gel filtered cytosol for 2 h at 0C, produced Type I

receptor binding comparable to the non-aged gel filtered preparations










Figure 4-2. Dose-Response Analysis of the Effects of Gel Filtration
and Molybdate on the Binding Capacity and Stability of Unoccupied
Type I Receptors from Whole Brain Cytosol. Nine parts cytosol

prepared in Buffer A was diluted with one part Buffer A containing 0,

20, 100, 200 or 1000 mM molybdate to bring the cytosol to a final

concentration of 0, 1, 2, 10, 20 or 100 mM molybdate, respectively.

Each group then either underwent the gel filtration procedure at 4-50C

(open symbols) or was left undisturbed at 4-5C during this process

(solid symbols). Subsequently, a portion of cytosol from each group was

either incubated with 10 nM [3H]ALDO and 5 uM RU 26988 minus

(BT) or plus (BNS) 2 uM [IH]ALDO at 0C for 24 h for the

determination of specific binding (open and solid circles) or was aged

at 0 (open and solid squares) or 22C (open and solid triangles) for 2 h

prior to incubation with the steroids. Gel filtration reduced the

cytosolic protein concentration from 7.3 to 3.3 mg/ml. The data are

from at least 4 independent replicate experiments and are expressed as a

percent of control* (i.e., specific binding measured in non-aged crude

cytosol prepared in Buffer A) + standard error of the mean (SEM).

100% = 18 fmole/mg protein.













160-

140-

120-

0 100-

80-

. 60-

40-

20-


0 2 10 20 00
[Na2 MoO 4] mM








79

containing the same doses of molybdate. Consistent with the results in

Fig. 4-1, Type I receptors in unlabeled gel filtered cytosol containing

0 (i.e., Buffer A) or 2 mM molybdate (i.e., Buffer M) were found to be

extremely labile when aged for 2 h at 220C; however, increasing the

concentration of molybdate to 10-100 mM prevents this loss in binding

capacity. Moreover, as reported earlier (Emadian et al., 1986), when

crude, non-gel filtered cytosol preparations are aged for 2 h at 220C,

the highest level of [3H]ALDO binding was observed in the presence

2 mM molybdate. Further increases in the concentration of molybdate

resulted in lower levels of [3H]ALDO binding, comparable to those

measured in non-aged crude cytosol containing the same concentration of

molybdate.

To gain further insight about the nature of the effects of gel

filtration on the binding of Type I receptors in the presence and

absence of molybdate, Scatchard (1949) analyses of the equilibrium

binding data were performed (Fig. 4-3, Table 4-1). Gel filtration of

cytosol in Buffer M produced no effect on the Kd, but increased the

B of Type I receptors for [3H]ALDO by 36% above the level
max
seen in crude cytosol prepared in Buffer A (control). Gel filtration of

cytosol in Buffer A produced a 62% reduction in the B and a
max
2-fold increase in the Kd of [ 3H]ALDO binding to Type I

receptors. The presence of 2 mM molybdate in crude cytosol produced a

slight increase in the Kd, but had no apparent effect on the

B of Type I receptors when compared to the control group.

To see whether the gel filtration-induced loss in Type I receptor

binding capacity observed in Buffer A was reversible, we investigated

the effects of 2 mM molybdate added to cytosol at different time










Figure 4-3. Effects of Molybdate and/or Gel Filtration on the
Equilibrium Binding Parameters of ALDO Binding to Unoccupied Type I
Receptors from Whole Brain Cytosol. Nine parts cytosol prepared in

Buffer A was diluted with either one part Buffer A (solid circles and

triangles) or one part Buffer A containing 20 mM molybdate to bring the

cytosol to a final concentration of 2 mM molybdate (open circles and

triangles). Subsequently, each group either underwent the gel

filtration procedure (triangles) at 4-50C or was left undisturbed at

4-50C during this process (circles). All groups were then incubated at

0C with 0.1-40 nM [3H]ALDO plus a 500-fold excess of RU 26988 and

either with (BNS) or without (B ) a 200-fold excess of

[1H]ALDO at 0C for 24 h for the determination of specific binding.

The concentration of free [ H]ALDO was determined in every tube by

subtracting BT from the total radioactivity present. The values for

B max and Kd were obtained from the x-intercept and negative
max d
inverse of the slope for each line, respectively, by least square linear

regression method. Gel filtration reduced the cytosolic protein

concentration from 7.3 to 3.3 mg/ml. The plots are representative of 2

independent replicate experiments.







81





60-


.C 50
0
a 40-





EII I I
L- 20 0L 6
a-





5 10 15 20 25 30

B (fmole / mg Protein)
SP


































-4
o o 14


0 00
41












w "
4w Cn%4

4.4


-44



















a0 ra




o 4

u a












c a 0
14,1
00 0 0
Go 0








p00
0I + ) I + 2










. 0 1
Ma














93'
A' a
0 0 a.1














13 X o o
^i ri c^-i r
ns
p .a











intervals following gel filtration. As shown in Fig. 4-4, adding 2 mM

molybdate to cytosol immediately following gel filtration in the

presence of Buffer A led to a significant recovery of Type I receptor

binding capacity; however, the magnitude of this recovery was only

sufficient to increase [ H]ALDO binding to the levels seen in the

control group and never to the levels observed in cytosol gel filtered

in the presence of 2 mM molybdate. The magnitude of this recovery was

also time-dependent, since only molybdate added 1 to 5 h following gel

filtration led to a significant recovery of the binding capacity of

Type I receptors. The addition of molybdate to unlabeled gel filtered

cytosol 24 h following gel filtration did not produce any increase in

the binding capacity of Type I receptors. The effects of 2 mM molybdate

on Type I receptor binding capacity in crude and gel filtered cytosol

were comparable to those described in Figs. 4-1 and 4-2.

A number of earlier reports have implicated the role of endogenous

small molecular weight "factors" in cytosol as possible modulators of

hormone binding to glucocorticoid receptors and subsequent activation

(transformation) and DNA binding of the steroid-receptor complexes

(Bailly et al., 1977; Ishohashi et al., 1980; Leach et al., 1982; Dahmer

et al., 1984). Since during Sephadex G-25 gel filtration many small

molecular weight components are removed from cytosol, we sought to

investigate the effects of dilution of cytosol (which in effect dilutes

the concentration of endogenous factors) possibly involved in the

inhibition of [3H]ALDO binding to Type I receptors) on the binding

capacity of these receptors. As shown in Fig. 5-5, a 2-fold dilution of

crude cytosol prepared in Buffer A or incubated with 2 mM molybdate for

2 h prior to dilution, produced Type I receptor binding comparable to












Figure 4-4. Stability of Unoccupied Type I Receptors from Whole
Brain Cytosol Following Gel Filtration in the Absence of Molybdate:
Effects of Molybdate Addition Following Gel Filtration and Aging.

Nine parts cytosol prepared in Buffer A was diluted with either one part

Buffer A (-M) or Buffer A containing 20 mM molybdate to bring the

cytosol to a final concentration of 2 mM molybdate (+M, solid bars).

Each group then either underwent the gel filtration procedure at 4-50C

(+ G-25) or was left undisturbed at 4-5C during this process (- G-25).

Subsequently, the group that was gel filtered in the absence of

molybdate was divided into two parts. One part was brought to 2 mM

molybdate immediately or 1, 2, 5 and 24 h following gel filtration

(shaded bars). The second part was diluted accordingly with Buffer A

immediately or 1, 2, 5 and 24 h (hash marked bars) following gel

filtration. All other groups were diluted accordingly with appropriate

buffers. Following these steps, all groups incubated with 10 nM

[3 HALDO and 5 uM RU 26988 minus (BT ) or plus (BNS) 2 uM

I H]ALDO at 0C for 24 h for the determination of specific binding.

Gel filtration reduced the cytosolic protein concentration from 7.3 to

3.3 mg/ml. The data are from at least 3 independent experiments and are

expressed as a percent of control* (i.e., specific binding measured in

non-aged crude cytosol prepared in Buffer A) + standard error of the

mean (SEM). 100% = 26 fmole/mg protein.














140


120

o
| 100.
o
0

80.
0
O-


60-
-


40-


20-



M
G-25
M


+ -- -
-+ +


~+ +










Figure 4-5. Effects of Dilution and/or Molybdate on the Binding
Capacity of Unoccupied Type I Receptors from Whole Brain Cytosol.

Nine parts cytosol prepared in Buffer A was incubated at 0C for 2 h

with either one part Buffer A (shaded bars) or Buffer A containing 20 mM

molybdate to bring the cytosol to a final concentration of 2 mM

molybdate (solid bars). One group either underwent the gel filtration

procedure at 4-5C in columns pre-equilibrated with (G-25, M) or without

(G-25) 2 mM molybdate. A second group was diluted by 2-fold (Dil.)

using either Buffer A or Buffer M, where appropriate. A third batch of

undiluted crude cytosol in Buffer A (the control*) or Buffer M (M) along

with the "Dil." groups remained at 4-5C during the gel filtration

process. Subsequently, all groups were incubated with 10 nM

[3H]ALDO and 5 uM RU 26988 minus (B ) or plus (BNS) 2 uM

[1H]ALDO at O0C for 24 h for the determination of specific binding.

Dilution and gel filtration reduced the cytosolic protein concentration

from 7.3 to 3.6 and 3.3 mg/ml, respectively. The data are from 4

independent replicate experiments and are expressed as a percent of

control* + standard error of the mean (SEM). 100% = 25 fmole/mg

protein.












120


100


80


60-


40-


20-


M Dil. Dil. G-25 G-25 G-25 G-25
M M