Citation
Design and evaluation of soft anticholinergics based on methscopolamine

Material Information

Title:
Design and evaluation of soft anticholinergics based on methscopolamine
Creator:
Kumar, Gondi N., 1960-
Publication Date:
Language:
English
Physical Description:
xiii, 123 leaves : illustrations ; 29 cm

Subjects

Subjects / Keywords:
Cholinergic Antagonists -- chemical synthesis ( mesh )
Cholinergic Antagonists -- pharmacology ( mesh )
Scopolamine Derivatives -- chemical synthesis ( mesh )
Scopolamine Derivatives -- pharmacology ( mesh )
Mydriatics -- pharmacology ( mesh )
Mydriatics -- chemical synthesis ( mesh )
Drug Design ( mesh )
Research ( mesh )
Department of Medicinal Chemistry thesis Ph. D ( mesh )
Medicinal Chemistry thesis, Ph. D
Dissertations, Academic -- College of Pharmacy -- Department of Medicinal Chemistry -- UF ( mesh )
Dissertations, Academic -- Medicinal Chemistry -- UF
Parasympatholytics -- chemical synthesis.
Scopolamine Derivatives.
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )
Academic theses ( lcgft )
Academic theses ( fast )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 116-122).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Gondi N. Kumar.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
027795778 ( ALEPH )
27961528 ( OCLC )

Downloads

This item has the following downloads:


Full Text
DESIGN AND EVALUATION OF SOFT ANTICHOLINERGICS
BASED ON METHSCOPOLAMINE
By
GONDI N. KUMAR
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
1992


To my parents, Kondaiah and Sakuntala


ACKNOWLEDGEMENTS
I thank my principal advisor Dr. R.H. Hammer for his excellent guidance
and care throughout this project. I would like to thank my coadvisor Dr. N.S.
Bodor for his example, thoughtful input and constant encouragement. I would
like to thank the member of my supervisory committee, Drs. J. Simpkins, M.O.
James and E. Enholm, for their interest in the project. I would like to thank the
Department of Medicinal Chemistry for the continuous financial support in the
form of an assistantship.
I would like to acknowledge the help I received from Drs. Z. Gunes, L.
Prokai and J. Sastry. The help I received from Dr. W. Wu and L. Prokai needs
special mention. I would to thank the other members at the Center: Laurie
Johnston, Joan Martignago, Julie Berger, Kathy Eberst and Robert Wong.
I would like to thank all friends and colleagues at the Center who made
this tenure pleasant and enjoyable. I cherish the support and encouragement
received from my siblings, Sujatha, Sunitha and Suresh, and my grandparents.
Most importantly, I would like to thank my wife Uma, for her
understanding and love, and for coping with me while contending with her own
dissertation.


TABLE OF CONTENTS
Page
Acknowledgements iii
List of Table vi
List of Figures vii
Key to Symbols x
Abstract xii
Chapters
1 Introduction 1
Drug Design 1
Soft Drugs 3
Autonomic Nervous System 8
Structure and Drug Deliver to the Eye 11
Skin Structure and Drug Delivery 17
In vitro Penetration Studies 20
Sweat Glands Structure and Pharmacology 24
2 Present Study and its Significance 27
Scopolamine and Methscopolamine 27
Design and Rationale 31
Objectives 34
3 Materials and Methods 36
Materials 36
Methods 37
4 Results and Discussion 53
Syntheses 53
Anticholinergic Activity 63
In Vitro Stability 83
In Vitro Penetration Studies 98
IV


5
Conclusions
112
References 116
Biographical sketch 123
v


LIST OF TABLES
Table Page
1 In vitro anticholinergic activity guinea pig ileum assay 66
2 In vitro stability of soft drugs 3 (a-f) in biological media 91
3 In vitro stability of soft drugs 6 (a-e) in biological media .... 92
4 The n-Octanol/water partition coefficients and the hairless
mice skin permeability coefficients of 1 and soft drugs 102
VI


LIST OF FIGURES
Figure Page
1 A schematic representation of the metabolic fate of a
conventional drug (D) in vivo 6
2 A schematic representation of the metabolic fate of a
soft drug (SD) in vivo 7
3 A schematic cross section of the eye 12
4 A schematic cross section of the skin showing the
structural elements and three potential routes of
penetration of a diffusant 18
5 A schematic representation of the amount of material
penetrating the skin as a function of time in an in vitro
diffusion experiment 22
6 Structures of methscopolamine, hypothetical metabolites
and the soft drugs of this study 33
7 Synthesis of phenylmalonic acid analogs of
methscopolamine (3 a-f) 39
8 Synthesis of phenylsuccinic analogs of
methscopolamine (6 a-e) 41
9 Representative dose response curves from guinea pig
ileum assay 64
10 A representative Schild plot 65
11 Mydriatic dose response curves in treated rabbit eye 70
12 Mydriatic dose response curves in rabbit treated eye
after unilateral instillation (1, 3 a-e) 71
vii


13 Mydriatic dose response curves in rabbit treated eye
after unilateral instillation (1, 6 a-c) 72
14 AUC24hrs for the treated eye after unilateral instillation
into the rabbit eye 73
15 Duration of mydriatic activity after unilateral instillation
into the rabbit eye 74
16 Time vs mydriatic response curves in the treated eye
after unilateral administration of equieffective doses 75
17 Time vs mydriatic response curves in the untreated eye
after unilateral administration of equieffective doses 77
18 AUC6hrs for the untreated eye after unilateral instillation
of equieffective doses 78
19 Mydriatic activity after intravenous administration of
equipotent doses to rabbits 81
20 Muscarinolytic activity against acetylcholine induced
bradycardia in rats after intravenous administration of
equipotent doses 84
21 pH profile of compound 3a in buffers at 63C 86
22 Pathways of degradation of 3a in buffers 87
23 Arrhenius plot for compound 3a in pH 4 buffer 88
24 Hydrolytic rates of 3a in rabbit tissues 93
25 Hydrolytic rates of 3a in pure enzyme preparations 94
26 Mydriatic activity in treated rabbit eye after unilateral
administration of equieffective doses of 3a and 3g 97
27 A representative trace of the in vitro penetration of 3c
across hairless mice skin 101
28 Relationship between log Partition Coefficient and
log Permeability Coefficient 103
VIII


29 Network of alternate pathways for percutaneous
absorption 106
30 The time course of penetration of 1, 3a and 6a across
rabbit cornea in vitro 108
31 In vitro permeability coefficients across rabbit cornea 109
IX


KEY TO SYMBOLS
A0
AT
AUC
C
CDCIg
cm
DMSO-d6
gm
hr
HPLC
IR
iv
k
kg
log P
log Kp
lbs
M +
angstrom
absolute temperature
area under the curve
degrees centigrade
deuterated chloroform
centimeter
deuterated dimethyl sulfoxide
gram
hour
high pressure liquid chromatography
infrared
intravenous
degradation rate constant
kilogram
log partition coefficient
log permeability coefficient
pounds
molecular ion
x


mAChR
mg
min
ml
M
mM
mp
n
NMR
P
QSAR
r2
SEM
*1/2
TLC
vs
v/v
w/v
6
V
v g
tA
muscarinic acetylcholine receptor
milligram
minutes
milliliter
moles
millimoles
melting point
number of samples or readings
nuclear magnetic resonance
probability
quantitative structure activity relationship
regression coefficient
standard error of mean
half life
thin layer chromatography
versus
volume/volume
weight/volume
parts per million
wave length (micrometer)
microgram
microliter
XI


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
DESIGN AND EVALUATION OF SOFT ANTICHOLINERGICS
BASED ON METHSCOPOLAMINE
By
Gondi N. Kumar
August, 1992
Chairperson: Richard H. Hammer
Cochairperson: Nicholas S. Bodor
Major Department: Medicinal Chemistry
The present study involves the development of a new type of
anticholinergics, called soft anticholinergics, which are based on
methscopolamine. They are designed to act at the site of application but not to
act systemically due to their rapid metabolic inactivation in the systemic
circulation. Two hypothetical metabolites of methscopolamine were chosen as
the lead compounds and they were inactivated by esterification with cyclic and
alicyclic alcohols to yield two series of compounds.
The compounds synthesized were found to be potent anticholinergics in
both in vitro and in vivo assays. The activity was found to decrease with
increasing side chain length. The soft drugs with short side chain (2 or 3
carbons) were found to be very short acting mydriatics in the rabbit eyes after
xii


unilateral topical administration of equieffective doses compared to a very long
duration of activity with methscopolamine. The untreated eye was found to
dilate in methscopolamine treated animals but not in soft drug treated animals,
which indicates a lack of systemic activity of the topically administered soft
drugs. One soft drug (phenylmalonic ethyl analog) exhibited a very short
duration of systemic anticholinergic activity compared to methscopolamine on
intravenous administration in rabbits and rats.
The in vitro hydrolytic rates of soft drugs in biological media were
significantly higher than those of methscopolamine. The soft drugs of
phenylmalonic series degraded, in vitro, to yield an active metabolite which is
the decarboxylated product of the lead hypothetical metabolite. The soft drugs
of the phenylsuccinic series degraded to yield the lead carboxylate metabolite
which is essentially inactive.
The transdermal permeabilities of the soft drugs were found to be directly
dependent on the octanol/water partition coefficients. The transcorneal
permeability of one of the soft drugs tested was found to be higher than
methscopolamine.
The short duration of mydriatic action coupled with a lack of systemic
activity on topical administration would probably increase the therapeutic index
and make these compounds promising candidates for further studies.
xiii


CHAPTER 1
INTRODUCTION
Drug Design
The biological and pharmacological properties of a compound are
determined by its chemical structure. Structural variations should be able to
change the biological properties as well. Numerous skillful variations have been
made by medicinal chemists to yield therapeutically useful drugs. The most
rational route for the design of better drugs would be direct design based on
the biomolecular processes underlying diseased states. Since the biomolecular
processes underlying many diseased states is still not well understood, such an
approach is not always feasible. In fact, many of the presently important types
of drugs have been discovered accidentally (Austel, 1989). Minimization of
chance discovery can be achieved by systematic empirical procedures which
incorporate methods that allow detection and objective expression of
relationships between chemical structures and their biological properties.
Lead Compound
The general scheme for drug discovery starts with identification of a lead
compound or structure. A lead compound is one which has some if not all the
1


2
desirable properties expected of a better drug. The lead compound is then
subjected to structural variations to optimize its biological properties in the
desired direction. A new drug must be able to improve current therapy. The
new drug, therefore, has to fulfill many more and stricter requirements than the
lead compound. Two aspects of this strategy is lead finding and lead
optimization (Testa, 1984).
Lead Finding
Lead finding includes the identification of novel compounds displaying an
interesting biological activity and compounds with a newly discovered biological
activity.
Mass (random) screening, rational approach (based on correct or
incorrect hypothesis), natural products, drug metabolites, unexpected side
effects and the most important of all, serendipity are the methods used to
accomplish lead finding.
Lead Optimization
Lead optimization is the starting point for systematic modification aiming
at increasing the desired activities and decreasing the undesired activities.
Quantitative methods based on biotransformation considerations and
quantitative structure activity relationships (QSAR) are the methods used for
lead optimization. QSAR method advanced by Hansch (1981), which
mathematically correlates physicochemical properties like partition coefficient,
lipophilicity etc. to biological activity, is used extensively in lead optimization.


3
The other approach used for lead optimization takes the metabolic processes of
the body into consideration. Ariens and Simonis (1982) introduced the design
of poorly metabolizable drugs called "hard drugs." Hard drugs are
"metabolically stabilized" and display some advantages, for example: longer
half-lives, decreased intra- and interpatient variability, decreased species
difference and decreased number and significance of active metabolites.
In contrast, the other group "prodrugs" and "soft drugs" are usually
rapidly metabolized due to the incorporation of labile groups into the molecules.
Prodrugs are compounds with little or no activity, but which undergo metabolic
transformation in vivo to yield the active drug (Notari, 1981). Prodrugs
essentially alter the pharmacokinetic rather than the pharmacodynamic profile of
a drug.
Soft Drugs
The therapeutic index (Tl) is the most important characteristic of a drug.
The toxicity of a drug is the result of summation of many effects of not only of
the drug itself but also its metabolites, reactive intermediates, and various
compounds resulting from direct or indirect interaction with cell components. It
is thus apparent that there is a need to include metabolism considerations in
the general drug design process.
Bodor (1981) has suggested an approach for the design of safe and
better drugs which aims to control and direct the metabolism by drug design.


4
This is the concept of "soft drugs," which are defined as "biologically active,
therapeutically useful chemical compounds (drugs) characterized by a
predictable and controllable in vivo destruction (metabolism) to non-toxic
moieties, after they achieve their therapeutic role" (Bodor, 1984, p. 262).
The main objective in the soft drug design is deliberate simplification of
the metabolism. Ideally, a soft drug is metabolized in a single step to an
inactive metabolite. This metabolite may be further metabolized or conjugated
before elimination. This process avoids formation of active metabolites or
reactive intermediates. Since the main objective is an optimized therapeutic
index, the new soft drug does not necessarily have to be the most potent agent
of the series.
There are at least five ways to attempt rational soft drug design (Bodor,
1984).
1. Soft analogs: Bioisosteric/isoelectronic replacement in the known bioactive
compounds to yield structural analogs which are better metabolized.
2. Activated soft compounds: A non-toxic inactive compound is activated by
the incorporation of a group which will provide the desired pharmacological
effect.
3. Active metabolites: A derivative of an active metabolite which is in highest
oxidation state and which then will undergo one-step deactivation.
4. Natural soft drugs: Pro-soft drug approach to deliver endogenous
substances such as steroids and autocoids.


5
5. Inactive metabolite approach: The inactive metabolite approach starts with
the identification of an inactive metabolite of a drug. The inactive metabolite is
then chemically modified to produce a compound with structural resemblance
(isosteric and/or isoelectronic) to the original active drug. This newly
synthesized compound should display sufficient pharmacological activity and
degrade in vivo to yield the original inactive metabolite without the generation of
any toxic intermediates.
This design has been applied successfully to such diverse agents like
DDT (Bodor, 1984), 6-blockers (Bodor et al. 1988), steroids (Druzgala et al.
1991) and anticholinergics (Bodor et al. 1980; Hammer et al. 1988). The
metabolic patterns of a conventional drug (D) and a soft drug (SD) are depicted
in Figures 1 and 2, respectively. A conventional drug (D) has the potential to
generate metabolites (active and/or inactive) and reactive intermediates. In
comparison a soft drug (SD) is metabolized in a single step to the inactive
metabolite, which is then either eliminated directly or metabolized further
(conjugation) before elimination. Thus, the soft drug approach enables the
separation of therapeutic properties from deleterious side effects.
Short Duration of Action
A short duration of action is important if a drug results in undesirable
side effects. If the plasma concentration is high over a long period of time, a
transaxonal transport of drugs may occur across the blood brain barrier
producing central side effects also (Mager et al. 1972). A short duration of


6
D
Figure 1: A schematic representation of the metabolic fate of a conventional drug
(D) in vivo.


7
Direct
Delivery
Process
>
Soft
Drug
SD
Active
(Inactive)
M ,,.M
1
Inactive
n
Elimination
Figure 2: A schematic representation of the metabolic fate of a soft drug (SD) in vivo.


8
action is necessary for diagnostic agents as well. It can be achieved by the
introduction of metabolically labile groups into the molecule (COOR,
OCO(CH2)nR, CH2S03Na etc) (Mager, 1984). If metabolically unstable groups
are placed in an accessible position in the molecule, the resulting entity may be
more rapidly metabolized, conjugated, and rapidly eliminated after fulfilling its
intended pharmacological activity.
Autonomic Nervous System
Acetylcholine acts as the neurotransmitter in all postganglionic
parasympathetic nerves as well as in all preganglionic fibers of peripheral
autonomic ganglia, motor nerves to skeletal muscles, and certain synapses
within the central nervous system. Binding of acetylcholine to the cholinergic
receptors initiates a series of biochemical and/or electrophysiological events
that eventually lead to the final cellular response.
The central and peripheral actions of acetylcholine are exerted at two
main types of receptors. In 1914, Sir Henry Dale classified cholinergic
receptors into two types: nicotinic and muscarinic, to describe the different
physiological effects of acetylcholine that were mimicked by the alkaloids
nicotine and muscarine. Nicotinic and muscarinic receptors differ not only in
their affinities for agonists and antagonists, but also in many other fundamental
respects such as location, function, molecular architecture and receptor-effector
coupling mechanisms (Wess et al. 1990). The muscarinic acetylcholine


9
receptors (mAChR) is an integral membrane protein which is accessible to
muscarinic ligands from the extracellular space. Studies with various probes
has showed the presence in the ligand-binding site of at least one disulfide
bridge, several thiol groups, some strong nucleophilic groups and one or more
tyrosine residues (Sokolovsky, 1984). Studies on antagonist binding showed a
strong correlation between the affinity of the ligands for the mAChR and their
hydrophobicity, indicating that binding of the bulky antagonists involves
hydrophobic interaction with the receptor (Jarv & Bartfai, 1982).
Several pharmacologically distinct subtypes of mAChRs have been
identified, which may represent potential targets for new, therapeutically useful
drugs. Based on their affinity for pirenzepine, the mAChRs have been classified
into two groups, M, and M2. The M, receptors (high affinity to pirenzepine) are
mainly present in forebrain (thought to be involved in higher brain functions
such as learning and memory as well as locomotor and behavioral effects) and
in peripheral autonomic ganglia such as intramural ganglia of stomach wall
(Wess et al. 1990). The M2 receptors (low affinity to pirenzepine) are located in
brain (implicated in the regulation of vegetative, sensory and motor functions)
and in peripheral effector organs such as heart, glands or smooth muscle
(Mitchelson, 1984).
Muscarinic Antagonists
Muscarinic antagonists competitively block the physiological actions of
acetylcholine mediated by muscarinic acetylcholine receptors. The use of drugs


10
that antagonize the action of acetylcholine long preceded the discovery of
acetylcholine itself. The prototypical antimuscarinic agents are the plant
alkaloids atropine and scopolamine found in several solanaceous species which
have been known to humans for millennia.
The structural elements of cholinergic antagonists (Wess, 1990) are
1. a cationic "head group" which is either a tertiary base which is protonated at
physiological pH or a quaternary ammonium moiety;
2. some "heavy blocking moieties," eg: alicyclic or aromatic rings, for
hydrophobic interaction with the receptor;
3. an interconnecting structural element (ester or amide) of definite length;
4. an "anchoring group," e.g. hydroxyl group(s) are often present at key
positions (Barlow and Ramtoola, 1980).
The major pharmacological effects produced by anticholinergics are:
reduction of lacrimal, salivary, bronchial and gastro-intestinal secretion;
suppression of sweating; increase in heart rate; depression of smooth muscle
activity in the bronchial, gastro-intestinal and genitourinary tract; mydriasis and
cycloplegia.
The variety of pharmacological responses of atropine/ scopolamine has
led to intense efforts to develop antimuscarinic agents which display selectivity
for certain organs or organ systems. But the success has been limited. The
clinical agents available today display essentially the same pharmacological
profile as atropine/scopolamine.


11
Stereochemistry of Anticholinergics
The most interesting aspect of biologically active compounds such as
hormones is their selective interaction with specific receptors. This interaction is
largely determined by their chemical complimentarity. Another important aspect
of this interaction is the spatial arrangement of atoms in the interacting ligand.
This is well exemplified in nature by the predominance of L-amino acids and D-
sugars. This stereospecific interaction is also true with xenobiotics. More than
75 years ago Cushny observed differences in activity between (-) hyoscyamine
and the corresponding racemate, atropine (Triggle & Triggle, 1976). High
eudismic ratios (eutomer/distomer; ratio of activity between the more active and
less active isomers) have been reported for hyoscyamine (Barlow et al. 1973)
and hyoscine (Triggle & Triggle, 1976). It has been shown that for the
appearance of enantioselectivity that an asymmetric center should be present in
a part of the molecule that actually contributes for binding (Ariens, 1966).
Structure and Drug Delivery to the Eve
A horizontal cross-section of an eye is depicted in Figure 3. The eye is
covered by three layers. The outer layer, sclera, is protective in function. It is
white in color and opaque with a transparent anterior portion (the cornea). The
middle layer is mainly vascular and is made up of the choroid, the ciliary body
and the iris. The innermost layer is the retina, a predominantly nervous tissue.
Within the three coats the eye is divided into two compartments by the lens.


Figure 3: A schematic cross section of the eye (Berman, 1991).


13
The frontal compartment contains aqueous humor, and is itself divided into
anterior and posterior chambers by the iris. The compartment behind the lens
contains the vitreous humor.
The important sites for drug action in the eye are the iris and ciliary
muscle, the blood vessels, the extraocular muscles and the lacrimal gland.
Autonomic Systems in Eve
The iris-ciliary body is composed of smooth muscles which are
innervated by the autonomic nervous system. The iris is a forward extension of
the choroid and arises from the anterior face of the ciliary body. The pupil
forms a central aperture in the iris. Two different types of muscle layers are
contained in the iris. These are dilator and sphincter muscles. The dilator
muscles receive sympathetic innervation. An increase in activity of these
muscles produces mydriasis. The sphincter muscles receive parasympathetic
innervation. An increase in activity of these muscles produces miosis. The
ciliary muscle is composed of smooth muscle units which receive
parasympathetic innervation. Stimulation of these muscles causes contraction,
which in turn results in a lessening of the tension on the suspensory ligaments
which causes lens to change shape, so that it becomes more convex and
allows near objects to be brought into focus at the retina.
Clinical uses of mydriatics include their diagnostic use for accurate
examination of retina and optic disc and in the treatment of acute iritis,
iridocyclitis and keratitis. Complete cycloplegia may be useful in certain clinical


14
conditions such as iridocyclitis and for precise measurement of refractive errors
(Weiner, 1980).
Clinically used anticholinergic agents for ocular use are atropine,
scopolamine, homatropine, cyclopentolate and tropicamide (Bartlett et al. 1989).
Atropine and scopolamine have a very long duration of action (exceeding 3
days). Homatropine and cyclopentolate have durations of about 1-3 days.
Tropicamide has the shortest duration of action lasting about 6 hours.
Tropicamide has been reported to be an unreliable cycloplegic (Gettes, 1961).
Systemic side effects of ocular drugs. Tropicamide syncope has been
reported as a complication of the drug (Schmidt, 1970). Scopolamine has been
reported to produce CNS toxicity (hallucinations and ataxia) (Freund & Merin,
1970). Homatropine has been reported to produce ataxia, hallucinations and
speech difficulty in children (Fraunfelder, 1989). Atropine has been reported to
produce at least 6 deaths involving ocular administration (Gray, 1979).
Cyclopentolate produced CNS side effects including grand mal seizures
(Kennerdell & Wucher, 1972).
Ocular Drug Delivery
Local administration (topical) has been the conventional form of drug
delivery to the eye. The topically applied ophthalmic drugs are in contact with
three absorptive layers: the cornea, the conjunctiva and the nasal mucosa
which results in ocular as well as systemic absorption. It has been shown that


15
about 10% or less of a topical dose is usually absorbed into the eye, the
remaining being systemically available, thereby eliciting unwanted side effects.
Some of the examples with narrow therapeutic indices include (3-blockers (Berry
& van Buskirk, 1984) and anticholinergics (Gray, 1979).
Some approaches that have been tried to minimize the systemic
absorption of ocularly applied drugs are
1. manipulation of physical properties of the vehicle (viscosity etc) to increase
the residence time of the drug at the site of administration (Chang et al. 1988);
2. prodrugs to enhance ocular absorption with the consequent reduction in the
dose (Chang et al. 1987);
3. soft drugs, which act locally at the site of application and are metabolized
rapidly to non-toxic moieties in systemic circulation, e.g. soft steroids (Druzgala
et al. 1991);
4. design of chemical delivery systems based on the unique enzymatic systems
of the eye for selective activation in the target tissue (iris-ciliary body), e.g.
alprenolol ketoxime (Bodor & Prokai, 1990); and
5. choosing an oculoselective drug. A new noncardioselective but topically
oculoselective 8-blocker has been reported to be advantageous over the
traditional 8-blockers (Bauer et al. 1991).
Another strategy is to select a less potent analog and then offset the loss
in potency by enhancing its ocular absorption using prodrugs. Some success
has been reported with 6-blockers using this approach (Nathanson, 1985).


16
Ocular Drug Metabolism
Aryl hydroxylase, UDP-glucuronosyltransferase activity (Das & Shichi,
1981) and cytochrome P-450 activity (Shichi & Neber, 1980) have been
demonstrated in the eye. Enzymes for mercapturate synthesis have been
shown to occur in the ciliary body (Das & Shichi, 1981). Abundant presence of
nonspecific carboxylesterases in the iris-ciliary body has been reported (Lee,
1983).
Animal Models
Practical, ethical and economic considerations prevent testing of the
potential pharmacological agents in humans directly. Hence experiments on
living animals are important to study the compounds before they can be
clinically tested. The most frequently used animal model for ocular studies is
the rabbit. A few differences exist between the ocular physiology of rabbits and
humans (Maurice & Mishima, 1984).
1. Pattern of blinking: Rabbits blink 4-5 times per hour compared to much
higher number in humans.
2. Lower corneal permeability in humans than rabbits.
3. Humans have a larger lacrimal gland and different lacrimal drainage system
than rabbits.
Similarities include the similar structure of cornea and similar method of
changing the focal power of the eye. J.H. Prince (1964, p. x) states that "while


17
differences exist, numerous enough similarities justify its continued use of rabbit
in ocular work with due caution."
Corneal Permeability
Transcorneal delivery of pharmacological agents in the treatment of
ocular diseases is of immense importance. The cornea acts as a barrier to
xenobiotics entering the eye. The cornea is a trilaminate structure with
hydrophilic stroma sandwiched between two layers of hydrophobic epithelium.
Thus for hydrophilic compounds, the epithelium acts as a barrier whereas the
stroma acts as a barrier for hydrophobic compounds. Since the permeability of
a certain drug plays as significant a role as its potency, compounds like
epinephrine have been manipulated to increase their permeability by suitable
structural modifications (e.g. dipivalyl ester dipivefrin) (Mandell et al. 1978).
Skin Structure and Drug Delivery
Skin is composed of two main tissue layers; the dermis and epidermis,
supported by a cushion of subcutaneous fat (Figure 4). Most simply, the skin
may be envisaged as providing a trilaminate membrane barrier to penetration;
(a) a thin, dead, highly resistant and lipophilic outer layer, known as the stratum
corneum, (b) the remainder of the living epidermis, which has the properties of
an aqueous gel, and (c) the dermis, which supports the vasculature and thus
provides a sink. For charged, hydrophilic molecules, the stratum corneum
provides the greatest barrier to skin penetration. This layer may be likened to a


18
ROUTES OF PENETRATION
Figure 4: A schematic cross section of the skin showing the structural elements
and three potential routes of penetration of a diffusant (Barry, 1983).


19
brick wall; the intercellular spaces between the keratinocytes (bricks) are filled
with a heterogeneous, apolar, lipid mixture (mortar). For small organic
molecules, the intercellular route rather than the alternative intracellular (or
shunt) route has been demonstrated to be the more dominant pathway of
penetration. Conversely, for lipophilic compounds, partitioning from the stratum
corneum into the aqueous, viable tissue, is the rate-limiting step.
Dermal and Transdermal Drug Delivery
The traditional forms used for cutaneous application are ointments,
creams, lotions and solution. Transdermal patches of scopolamine and
nitroglycerin (Dasta & Garaets, 1982) and topical solution of minoxidil (Maibach
& Wester, 1984) are some excellent examples of administering drugs by
transdermal and dermal routes, respectively. Bodor and Chikhale (1991) have
experimentally achieved dermal delivery of an acyclovir chemical delivery
system.
Cutaneous Metabolism
Recent studies have shown that certain compounds are indeed
metabolized in skin during the percutaneous absorption process (Kao et al.
1984). It seems likely that, although some compounds are metabolized
extensively during skin absorption, for many, metabolism will be small or
undetectable. However, a small amount of metabolism may be extremely
important, such as in the activation of potential carcinogens or other potent
compounds. Also, a knowledge of skin metabolism is essential for accurate


20
determination of the pharmacokinetics involved in the skin absorption process.
Esterase activity has been demonstrated in skin of humans and animals
(Tauber & Rost, 1986) along with other enzymes responsible for Phase I and
conjugation reactions (Tauber, 1982). The metabolic ability of skin has been
used to advantage in the design and activation of prodrugs (Higuchi, 1977).
Hadgraft (1980) and Bickers (1980) have reviewed this topic.
In Vitro Penetration Studies
In vitro absorption studies, when properly conducted and interpreted,
have proved very useful for the estimation of in vivo human percutaneous and
transcorneal absorption. Accurate determinations of absorption rates can be
made with in vitro diffusion cells, since frequent sampling can be done directly
beneath/across the membranes. For highly toxic compounds, in vitro studies
may be the only ethical way that human absorption data can be obtained.
For many compounds, the primary barriers to absorption are the non
living surface layer (stratum corneum) in the case of skin and the hydrophobic
epithelium of cornea (Ashton et al. 1991) in the case of eye. The rationale for
measuring absorption by in vitro techniques is that the absorption rates of
hydrophilic molecules are determined by passive diffusion through the stratum
corneum. In the case of eye, the hydrophilic molecules are known to diffuse
through the aqueous channels that exist in the corneal epithelium (Grass &
Robinson, 1988a).


21
Animal Models
Excised skin from a variety of animals, including rats and mice (normal
and hairless), rabbits, guinea pigs, hamsters, pigs, hairless dogs and primates
have been employed in a variety of types of diffusion cells in efforts to predict
percutaneous absorption in man (Barry, 1983). A popular model has been the
hairless mouse. Permeability of hairless mice skin has been shown to be
similar for some compounds (Stoughton, 1975). The in vitro absorption studies
across the rabbit cornea is the only reported technique so far reported.
Solutions to Ficks Laws of Diffusion
In an in vitro diffusion experiment, the amount of substance building up in
the receptor phase is measured as a function time over an extended period.
Often, data are collected for a 48-hour period in the case of skin (6 hours in the
case of cornea) and the type of trace obtained is shown schematically in
Figure 5 (Hadgraft, 1989). During the middle section of the experiment the
amount accumulating approximates to a linear function of time; pseudo-steady
state has been established and a linear concentration gradient exists across the
skin. This portion of the overall curve is described by Ficks first law of
diffusion:
J = DAK (C0 Cr) / h
where J is the flux of substance across the skin, D is the effective diffusion
coefficient, K is the skin-vehicle partition coefficient, A is the cross-sectional area
of the skin of thickness h, and (C0 Cr) is the concentration difference between


22
Figure 5: A schematic representation of the amount of material penetrating the
skin as a function of time in an in vitro diffusion experiment (Scott et al. 1989).


23
the donor and receptor phases, respectively. Often, the concentration in the
receptor phase can be considered insignificantly small compared with the donor
concentration (Cd) and the above equation is often simplified to
J = Kp A Cd
where Kp is the permeability coefficient and
Kp = D K / h
It is very difficult to assign precise values to K, D and h, and it is often more
appropriate to quote permeability coefficients generated form flux experiments
rather than to attempt to deconvolute them into their component parameters.
The linear portion of the graph can be extrapolated back to the time axis. This
yields the lag time which, from a solution to Ficks second law of diffusion, can
be shown to be numerically equal to h2/6D.
Partition Coefficient and Permeability
Membranes like skin and cornea have both hydrophilic and hydrophobic
regions. The currently accepted theory proposes that lipid-soluble substances
pass through membranes because of their lipid content, whereas water-soluble
substances pass through cell membranes because of their hydrated
proteinaceous content. Apart from this, the chemical composition and the
physical characteristics of compounds play an important role in the penetration
process.
The partition coefficient is a measure of the ability of a chemical to
separate between two immiscible phases. This characteristic simulates the


24
partition of a compound in a vehicle and a membrane. The partition coefficient
between water and heptane (Bartek & LaBudde, 1975) or octanol (Roberts &
Anderson, 1977) of some compounds has been used to correlate their
percutaneous absorption. A good correlation has been reported to exist
between in vitro determined permeability and partition coefficients. In the case
of ocularly administered drugs precorneal factors like nasolacrimal drainage and
noncorneal absorption play a significant role in the overall bioavailability of the
drug at the site of action, i.e., the inner compartments of the eye. Hence an
optimal transcorneal absorption is needed to compete with the noncorneal
factors. Partition coefficient has been used to assess the penetration potential
of drugs across the cornea. A parabolic relationship (Schoenwald & Ward,
1979) has been reported for steroids between their log partition coefficient and
log permeability coefficient. Even though relative potency is a significant factor,
a rapid penetration rate can contribute significantly to effectiveness. It is
suggested that Hansch type correlations, linear or parabolic with modifications,
can be used to predict drug transfer across various ocular membranes such as
the cornea, blood-aqueous barrier and blood-vitreous barrier (Lien et al. 1982).
Sweat Glands Structure and Pharmacology
Schiefferdecker introduced the terms "eccrine" and "apocrine" to describe
the two types of simple tubular exocrine glands of the skin (Greaves & Shuster,


25
1989). Eccrine glands discharge fluid without loss of cytoplasmic material in
contrast to the apocrine glands. Apocrine glands are associated with a hair
follicle/ sebaceous gland unit whereas eccrine glands have an independent
duct opening onto the surface. Eccrine glands are distributed over the entire
body surface while apocrine glands are concentrated in axillary and pubic
areas. Eccrine glands are richly innervated by unmyelinated sympathetic nerve
fibers which are cholinergic. Apocrine glands are innervated by adrenergic
nerves which respond to catecholamines. Very few animal species posses a
mechanism of heat dissipation quite as effective as that of man, who can
secrete as much as 1 liter/hour from eccrine sweat glands. The response of
human eccrine glands to acetylcholine and allied substances and the blocking
action of atropine has been demonstrated (Collins et al. 1959).
Antiperspirants
Hyperhidrosis is a cosmetic problem of great social importance.
Aldehydes (formaldehyde and glutaraldehyde) have been shown to have
antisudorific effect (Sato & Dobson, 1969). While glutaraldehyde causes
discoloration of skin, formaldehyde causes sensitization. Zinc and aluminum
salts are used extensively as antiperspirants. The mechanism of action of these
salts is still unclear. An aluminum-containing plug of solid material has been
discovered in the duct of eccrine sweat glands following the application of
aluminum chlorhydrate (Hem & White, 1989). These salts may produce
peridural inflammatory changes which result in eccrine duct occlusion and a


26
degenerative change may occur in the apocrine tubular cell following their
repeated use (Kuno, 1956).
Anticholinergics as Antiperspirants
One of the antisecretory effects exerted by anticholinergics is their
inhibition of eccrine sweating. This aspect is usually the side effect of
systemically administered anticholinergics. A number of studies carried out with
anticholinergics show their effectiveness in inhibiting sweating. A detailed study
carried out (MacMillan et al. 1964) with 95 anticholinergics showed that some of
the most active antiperspirants are quaternary compounds. Scopolamine
methyl bromide was found to be more effective on human subjects than
scopolamine hydrobromide. In another study, topical application of a 1%
solution of scopolamine inhibited sweating for six days, while an 18% solution
has an effect lasting over 30 days (Stoughton, 1964).


CHAPTER 2
PRESENT STUDY AND ITS SIGNIFICANCE
Scopolamine and Methscopolamine
Scopolamine (Hyoscine) is an alkaloid found chiefly in shrubs,
Hyoscyamus niger and Scopolia carnolica. It is an antimuscarinic agent due to
its competitive antagonism of acetylcholine at muscarinic receptors. It is used
therapeutically in both tertiary form as hydrobromide salt and quaternary forms
as methobromide (1) (Figure 6), methonitrate and butylbromide (Extra
Pharmacopoeia, 1989).
Scopolamine has both central and peripheral anticholinergic actions. Its
peripheral effects include increase in heart rate at higher doses, decreased
production of saliva, sweat, bronchial, nasal and intestinal secretions, decreased
intestinal motility and inhibition of micturition. Its ocular effects include dilation
of pupil, paralysis of accommodation and photophobia. Scopolamine has found
wide application in therapy. In its tertiary form it is used as an adjunct to
anaesthetic medication, for prevention of motion sickness (usually as a
transdermal patch) and in its quaternary form as an antispasmodic for gastro
intestinal disorders and as a mydriatic/cycloplegic. Scopolamine ointment was
used experimentally to prevent gustatory sweating with success (Bailey &
27


28
Pearce, 1985). Topical application of 1% solution of scopolamine was found to
prevent thermal and pilocarpine induced sweating (Shelley & Horvath, 1951;
Brun & Hunziker, 1955). MacMillan et al (1964) have tested a series of O-
esters of scopolamine for antiperspirant activity and have reported high efficacy
in preventing thermally induced sweating.
Metabolism of Scopolamine
After transdermal administration of scopolamine to humans, 79%
appeared in urine as conjugates of glucuronic and/or sulfuric acid and 21% as
unchanged drug (Scheurlen et al. 1984). In the mouse the following
metabolites were found after administration of scopolamine hydrobromide:
scopolamine 9-glucuronide, 6-hydroxyhyoscyamine, scopine, aposcopolamine,
nor-scopolamine, nor-scopolamine 9-glucuronide and unchanged scopolamine
(Werner & Schmidt, 1968). In rats the following metabolites were found after
administration of methscopolamine bromide : methaposcopolamine, quaternary
scopine, p-hydroxy methscopolamine, p-methoxy methscopolamine,
methscopolamine glucuronide and two other unidentified metabolites (Sano &
Hakusui, 1974). In an in vitro stability study of scopolamine in different animal
tissues, human serum did not hydrolyze scopolamine whereas rabbit serum
exhibited the highest capacity of hydrolysis. The enzyme which hydrolyses
scopolamine has been shown to be different from cholinesterase (Otorii, 1969).
Incubation of scopolamine with guinea-pig liver microsomal preparations
resulted in the formation of the corresponding nor-alkaloids and N-oxide


29
(Phillipson et al. 1976). In case of atropine, an analogous tropane alkaloid
which differs from scopolamine by the absence of epoxy moiety, a toxic
aldehyde metabolite, atropanal, was detected in rabbits (Matsuda, 1966).
Side Effects of Scopolamine
Systemic administration of scopolamine in therapeutic doses normally
causes dry mouth, drowsiness, euphoria, amnesia and fatigue. Occasionally it
also causes excitement, restlessness, hallucinations or delirium (Extra
Pharmacopoeia, 1987). Transdermal administration of scopolamine as anti
motion-sickness agent produced dilation of pupil (Johnson & Moore, 1983).
Administration of scopolamine eye drops resulted in visual hallucinations,
strange behavior and restlessness (Hamborg et al. 1984; Birkhimer et al. 1984).
The side effects after ocular administration are due to the drainage of the
applied drug into the nasolacrimal duct and subsequent systemic distribution.
In view of the above cited systemic side effects after local administration
of scopolamine, it is advisable to design soft drugs based on scopolamine for
achieving a local action without systemic side effects.
Soft Anticholinergics
Bodor et al. (1980) have reported soft ester open chain analogs of
anticholinergics synthesized from cyclopentyl-phenylacetic acid, phenylacetic
acid and branched aliphatic carboxylic acids. In this class of soft
anticholinergics the ester oxygen and quaternary head are separated by only
one carbon whereas the classical anticholinergics contain a 2-3 carbon bridge.


30
The quaternary head is destroyed on hydrolysis with consequent loss of
potency with the formation of formaldehyde and other inactive moieties. The
soft anticholinergics so designed were found to be potent and hydrolytically
unstable. This concept was also applied for the design of soft anticholinergics
based on propantheline (Brouillette, 1987). The ethylene bridge of
propantheline was shortened by one carbon to produce a new series of
compounds with retention of anticholinergic activity and decreased hydrolytic
stability.
Hammer et al. (1988) have applied the inactive metabolite approach of
designing soft anticholinergics of atropine. They have chosen a hypothetical
metabolite of atropine, an oxidation product of the primary hydroxyl group as
the lead compound. This lead compound was reactivated by esterification with
aliphatic and cycloaliphatic alcohols of varying chain length. In both in vitro
(guinea pig ileum assay) and in vivo (mydriatic activity in rabbit eye) tests, these
compounds were shown to be active and comparable to atropine. Compared
to atropine, these compounds are reported to be less stable in biological media
yielding the inactive metabolite. The soft anticholinergics have been shown to
have short duration of mydriatic activity upon topical administration into rabbit
eye (Hammer et al. 1991) and an ultra-short duration of muscarinolytic activity
on systemic administration in rats (Bodor et al. 1990) compared to atropine.
Soft anticholinergics so far synthesized essentially have two
characteristics:


31
1. retention of anticholinergic activity and
2. increased rates of hydrolysis in biological media.
Tropicamide and tropane alkaloids, viz: atropine and scopolamine are
used in ophthalmic procedures as mydriatic/cycloplegic agents for the
examination of the interior of the eye. With tropane alkaloids, the effect is
usually accompanied by systemic side effects due to drainage of the ocularly
administered drug into the systemic circulation through the nasolacrimal canal.
The mydriatic effect of tropane alkaloids usually lasts for more than two days.
The duration of action with tropicamide is about six hours (Weiner, 1980).
Anticholinergics have been tested as antiperspirants experimentally and
have been found to be active. But prolonged use of traditional anticholinergics
as antiperspirants may not be feasible due to the systemic absorption and
consequent accumulation in the systemic circulation resulting in unwanted side
effects.
Design And Rationale
In view of the above cited undesirable systemic side effects due to the
locally administered drug and extremely long duration of mydriatic action, the
design of soft drugs based on scopolamine is desirable. It is also desirable to
develop a topical antisecretory agent with a selective local action on the eccrine
sweat glands with loss of potency on entering the systemic circulation. The
inactive metabolite approach for the design of soft drugs advanced by Bodor


32
(1984) and successfully adopted for the design of soft drugs of atropine by
Hammer et al. (1988) is applicable to scopolamine/ methscopolamine as well.
A hypothetical carboxylate metabolite of methscopolamine (2) is chosen as the
lead metabolite for the design of the soft drugs. Even though this metabolite
has not been detected in either animals or humans, it is a logical choice as a
lead compound since it is the oxidation product of primary alcoholic group of
methscopolamine. Another hypothetical metabolite (2a) chosen as the lead
compound is designed with a methylene group in the side chain. This
extension of side chain will expose the ester moiety even further and possibly
lessen the steric hinderance making it more amenable for enzymatic hydrolysis.
The quaternary form of scopolamine is chosen as the lead compound instead
of the tertiary form to minimize the central side effects on systemic absorption
(Malatray & Simon, 1972). The hypothetical metabolites are expected to be
highly polar (pKa: ca. phenylmalonate = 2.65 and phenylsuccinate = 4.0) and
ionized at physiological pH, and thus be subject to facile elimination from the
systemic circulation either directly or after conjugation. The hypothetical
metabolites are activated by esterification with suitable aliphatic and
cycloaliphatic alcohols to yield soft drugs (3 or 6) (Figure 6) which will degrade
in vivo, in a single step, to the more polar metabolite (2 or 2a). Strong
nucleophilic groups are shown to be present at the muscarinic receptor site
(Sokolovsky, 1984). There is a possibility that the carboxylate metabolite (2 or
2a) that results on the hydrolysis of the soft drugs will have an unfavorable


33
1
2 (Hypothetical metabolite)
3 (Soft drugs)
2a (Hypothetical metabolite)
Figure 6: Structures of methscopolamine, hypothetical metabolites and the soft
drugs of this study.


34
interaction with the receptor site, making it less active than the original soft
drug. The hydrolysis is expected to be facile in the possible target tissues for
the soft drugs due to the abundant presence of non-specific esterases in ocular
tissues (Lee, 1983) and skin (Tauber, 1982). Thus a shorter local action with
potentially reduced systemic side effects can be visualized with these soft
drugs.
Objectives
The aims of this study are the development and evaluation of soft
anticholinergics based on methscopolamine. These putative soft
anticholinergics are expected to be active locally at the site of application but
are hydrolyzed in a facile manner in systemic circulation to an inactive polar
metabolite. The net result is expected to be an increase in the therapeutic
index with localization of response and avoidance of systemic side effects. A
decrease in the duration of mydriatic activity is expected due to the
incorporation of a labile ester moiety into the molecule. The experimental
protocol is as follows:
1. Syntheses of phenylmalonic and phenylsuccinic acid analogs, and their
respective metabolites.
2. Development of a suitable analytical system to evaluate the stability and
the metabolic pathways of the soft drugs synthesized.


35
3. Evaluation of physical stability and degradation pathways of a selected
soft drug in buffers.
4. Evaluation of in vitro stability in various biological media.
5. Evaluation of in vitro pharmacological activity by guinea-pig ileum assay.
6. Evaluation of in vivo activity in suitable animal models.
7. Studies on penetration characteristics of soft drugs through rabbit cornea
and hairless-mice skin, and correlating the permeabilities with the
partition coefficients.


CHAPTER 3
MATERIALS AND METHODS
Materials
All chemicals used were reagent grade. Scopine hydrochloride,
acetylcholinesterase, pseudocholinesterase, porcine liver esterase, carbachot
and hexamethonium bromide were obtained from Sigma Chemical Company.
Other chemicals were obtained from Aldrich Chemical Company and solvents
from Fisher Scientific. All melting points were recorded using Fisher-Johns
melting point apparatus and are uncorrected. NMR data were recorded with
Varan T-90 NMR spectrometer and are reported in parts per million (5) relative
to tetramethylsilane. All quaternary compounds were dissolved in DMSO-d6 and
other compounds were dissolved in CDCI3. Infrared spectra were recorded with
Perkin Elmer 1420 Ratio Recording Infrared Spectrophotometer. The elemental
analyses were carried out at Atlantic Microiab. Inc., Atlanta, Ga. FAB mass
spectrometry (Kratos MFC 500) of the quaternary compounds and metabolites
for characterization and identification in in vitro stability studies was performed.
Thin layer chromatography was carried out using EM Science DC-plastic foil
plates coated to a thickness of 0.2 mm with silica gel 60 containing Florescent
(254) indicator. The mobile phase consisted of toluene:methanol or
36


37
hexanes:acetone in various proportions. Column chromatography was
performed with silica gel (70-230 mesh) with appropriate mobile phases. All the
animal studies were conducted in accordance with the guidelines set forth in the
Declaration of Helsinki and The Guiding Principles in the Care and Use of
Animals (DHEW Publication, NIH 80-23). The following strains of animals were
used in the studies:
1. Male New Zealand albino rabbits weighing 3 kg (Kel Farm, Alachua, Florida),
2. Male Sprague Dawley rats weighing 300 gms,
3. Male Hartley guinea-pigs weighing 400 gms, and
4. 4-5 weeks old female hairless mice (NIHS-bg-nu-xid).
Rats, guinea pigs and mice were obtained from Harlan Sprague Dawley
Inc., Indianapolis.
Methods
Synthesis of Phenvlmalonic Analogs (3 a-f) (Figure 7)
Synthesis of monoalkvl 2-phenvlpropanedioic acids 4(a-f)
Phenylmalonic acid (9.00 gms, 0.05 moles) of phenylmalonic acid
dissolved in 100 ml of diethyl ether was treated with 5.95 gms (0.05 moles) of
thionyl chloride and 2 drops of dimethylformamide. The mixture was refluxed at
40-50C for 2 hours. The solvent was removed under vacuum. The acid
chloride, dissolved in 50 ml of dry benzene, was treated with 0.055 moles of the
relevant alcohol and the mixture was stirred overnight at room temperature.


38
The mixture was then washed with three 50 ml portions of water to remove any
traces of unreacted phenylmalonic acid. The monoester was then extracted
into 100 ml saturated solution of sodium bicarbonate. The bicarbonate solution
was washed with 50 ml of diethyl ether and neutralized with 10% of hydrochloric
acid. The oily monoester was extracted into 50 ml of diethyl ether, dried over
anhydrous sodium sulfate and crystallized from diethyl ether/petroleum ether.
The purity of the phenylmalonic acid monoesters was confirmed by thin layer
chromatography and their structures were confirmed by NMR spectroscopy.
Synthesis of (2R.2S1 1-alkvl. 3-r3a-f8-methvl-6B.7B-epoxv-8-azabicvclo [3.2.1]
oct-3-vll 2-phenvlpropanedioic acid 5fa-fl
Scopine hydrochloride was dissolved in methanol and neutralized with an
equimolar quantity of methanolic potassium hydroxide. The precipitated
potassium chloride was filtered out, the filtrate was evaporated and the oily
residue was redissolved in chloroform. The filtrate was evaporated and dried
under vacuum. Scopine base was recrystallized from ether/pet.ether to yield
needle shaped crystals.
Phenylmalonic acid monoester (0.01 moles) dissolved in 25 ml of diethyl
ether was treated with 11 millimoles of thionyl chloride and two drops of
dimethylformamide. The mixture was refluxed for 2 hrs at 40-50C. Solvent and
excess thionyl chloride were evaporated under vacuum. The oily acid chloride
was dissolved in dry benzene and to it was added, dropwise, a solution of 0.01
moles of scopine in benzene. After overnight stirring at room temperature, the
reaction mixture was washed with three quantities of 50 ml each of saturated


39
a : ethyl
b : n-propyl
c : n-butyl
d : n-pentyl
e : c-hexyl
f : ethyl-c-hexyl
Figure 7: Synthesis of phenylmalonic acid analogs of methscopolamine (3 a-f).


40
sodium bicarbonate solution and separated. The scopine ester was then
extracted into 50 ml of 10% hydrochloric acid. The acidic solution of scopine
ester was washed with 50 ml of diethyl ether and neutralized with sodium
bicarbonate. The oily scopine ester was extracted into 50 ml of chloroform,
which was dried over anhydrous sodium sulfate and evaporated to yield a
yellowish brown viscous liquid. The structure of the compound was confirmed
with NMR (Solvent: CDCI3) and purity with thin layer chromatography.
Quaternization
Quaternization of scopine ester (5 a-f) was done by reacting 0.005 moles
of scopine ester and 0.006 moles of dimethyl sulfate in diethyl ether. The
precipitated quaternary compound (3 a-f) was filtered and recrystallized from
methanol/ether. The compounds were identified by NMR/Mass spectra and
the purity was confirmed by elemental analysis.
Synthesis of metabolite (3g)
Essentially similar procedure adopted for 5 was followed for the synthesis
of metabolite 3g. Phenylacetic acid was coupled with scopine followed by
quaternization with dimethyl sulfate.
Synthesis of Phenvlsuccinic analogs (6a-e) (Figure 8)
Synthesis of bromoacetate esters 7(b-e1
Ethyl bromoacetate was purchased from Aldrich Chemical Company.
Bromoacetic acid (8.34 gms, 0.06 moles) was dissolved in 75 ml of anhydrous


41
O
OH
ROH
Benzene

O
R
6a
ethyl
6b
n-propyl
6c
n-butyl
6d
c-hexyl
6e
benzyl
2a
H
2a
Pd/C, Hydrogen
COOH
2 eq LDA

Phenylacetic
acid
1. Thionyl
chloride
2. Scopine
3. Dimethyl
ir sulfate
6 (a-e)
Figure 8: Synthesis of phenylsuccinic analogs of methscopolamine (6 a-e).


42
benzene and to it was added 0.05 moles of the relevant alcohol. The mixture
was refluxed using a Dean-Starke tube until no more water generated from the
reaction. The mixture was then washed with 100 ml of saturated sodium
bicarbonate solution to remove excess acid, treated with anhydrous sodium
sulfate and evaporated to yield a yellowish viscous liquid. The esters were then
purified by vacuum distillation at 10 mm of Hg.
Synthesis of monoalkvl 2-phenvlbutanedioic acids 8(a-e1
Phenylacetic acid (1.63 gms, 0.012 moles) was dissolved in 50 ml of
anhydrous tetrahydrofuran and cooled to -70C in acetone-dry ice bath under a
brisk stream of dry nitrogen. Lithium diisopropyl amide monotetrahydrofuran
(17 ml, 1.5 M solution in hexanes- Aldrich Chemical Company, 0.025 moles)
was injected through stopple slowly. The mixture was stirred for 1 hour to yield
a thick white suspension of the dianion. Then 0.012 moles of the bromoacetate
ester dissolved in 5 ml of tetrahydrofuran was injected at once through stopple
and the mixture was stirred for 1 hour to yield a reddish brown liquid which was
allowed to warm to room temperature. The reaction was then stopped by the
addition of 1 ml of water. The mixture was evaporated and the residue was
suspended in 50 ml of ethyl ether. The ether layer was then washed with 100
ml of dilute hydrochloric acid and the reaction products were extracted into 100
mi of saturated sodium bicarbonate solution. The bicarbonate layer was
separated, washed with 50 ml of diethyl ether, neutralized with dilute
hydrochloric acid and the precipitated products were extracted into 50 ml of


43
diethyl ether. The ether layer was treated with anhydrous sodium sulfate and
evaporated to yield an amorphous solid. If the TLC (hexanes:acetone = 2:1)
showed the presence of phenylacetic acid, the product was purified by silica gel
chromatography using hexanes-acetone. The product was crystallized in
ether/petroleum ether.
Synthesis of (2R.2S1 4-alkvl 1-[8-methvl-6B.7B-epoxv-8-azabicvclo [3.2.1] oct-3-
vl] 2-phenvlbutanedioic acid 9fa-e1
4-alkyl 2-phenylbutanedioic acid (0.05 moles) was dissolved in 25 ml of
diethyl ether. To it was added 1 ml of thionyl chloride and the mixture was
refluxed at 50C for 2 hours. The solvent was removed under vacuum. Three
portions of 10 ml each of dry benzene were added and removed under reduced
pressure. The reddish brown oily acid chloride was redissolved in 20 ml of dry
benzene and to it was added dropwise a solution of 0.8 gms of scopine (0.005
moles) in benzene. The mixture was stirred overnight at room temperature.
Then it was washed with saturated sodium bicarbonate and the scopine ester
was extracted into 25 ml of dilute hydrochloric acid. The acid layer was washed
with ether, neutralized with sodium bicarbonate and the scopine ester was
extracted into 25 ml of chloroform. The solvent was removed under vacuum to
yield a viscous brownish liquid. TLC was performed with toluene:acetone (3:1).
Quaternization
Phenylsuccinic diester (0.002 moles) was dissolved in 20 ml of
chloroform and to it was added 0.5 gms of dimethyl sulfate. The mixture was
stirred at room temperature overnight. The white precipitate (6 a-e) was


44
collected on a filter, washed with diethyl ether and crystallized from
ethanol/diethyl ether.
Carboxvlate metabolite (2a)
Debenzylation of 6e was done by subjecting a mixture of 0.85 gms of 6e
and 1 gm of 5% palladium on carbon in 20 ml of acetonitrile to hydrogenolysis
at 30 Ibs/sq.in. in Parr apparatus for 1 hour. The mixture was filtered and the
filtrate was evaporated to yield a white powder, which was crystallized from
methanol/ether.
In Vitro Stability
Analytical method
A high pressure liquid chromatographic (HPLC) method was developed
to assay the soft drugs and metabolites in buffer media and biological fluids.
The system consisted of SP 8810 precision isocratic pump, SP 4290
autosampler with 20 fjI loop, reverse phase Waters Nova-Pak CN 5mm X 10cm
radial-pak column with a guard column, SP 8450 UV/visible detector and
SP4290 integrator. The mobile phase consisted of 2.5 mM potassium
dihydrogen phosphate buffer containing 5 mM of 1-octane sulfonic acid sodium
and acetonitrile in different proportions. The flow rate was 1 ml/min. The
detection was made at 254 nm. The area under the peak was used as a
measure of the concentration. The concentration vs. area under the peak plot
showed linearity (r = 0.999) for the range of 0.2-1 ¡jg of the injected


45
compounds with a detection limit of 0.2 ¡jg (0.4 nm) of injected sample. The
retention times (minutes) for soft drugs and metabolites with 40% v/v
acetonitrile are as follows: 1 5.6, 3a 7.5, 3b 8.6, 3c 9.7, 3d 11.2, 3e -
10.2, 3f 15.2, 3g 6.5, 6a 8.2, 6b 8.5, 6c 8.7, 6d 9.4, 6e 9.2 and 2a -
6.8.
In Vitro Stability in Buffers
USP standard buffer solutions (0.2 M) (USP XXI, 1985) in the pH range
of 1.5 to 9.5 were used in the study. Solutions of 3a (0.05 M) in buffers were
made. The solutions were kept in glass vials with screw caps with septa to
facilitate withdrawal of sample without opening the cap. The solutions were
stored at 25C, 37C, 46C, 56 and 63C. A 100 ¡j\ sample was withdrawn at
various time points and was analyzed by HPLC for 3a and degradation
products. Atropine sulfate monohydrate, 0.5% w/v in pH 3.8 USP buffer was
used as internal standard. Atropine solution was stored at 5C and showed no
degradation during the period of study.
In vitro metabolism in biological media
The samples were analyzed by HPLC as described above. The samples
were either analyzed immediately or stored frozen until analysis.
Rat, rabbit and human plasma. The plasma was obtained by
centrifugation of freshly obtained heparinized blood. The stability studies were
carried out by adding aliquot of the drug stock solution to plasma to obtain a


46
final concentration of 0.25 mM. The plasma was kept at 37C while shaking.
Samples of 0.1 ml were withdrawn at appropriate time intervals and mixed with
0.4 ml of ice cold acetonitrile to stop enzymatic reaction. The sample was then
centrifuged and the supernatant was analyzed by HPLC.
Rabbit liver homogenate. A suspension of freshly obtained rabbit liver
was made in isotonic pH 7.4 phosphate buffer. The protein content was
adjusted to 50 mg per ml. The liver suspension was equilibrated at 37C. An
aliquot of soft drug stock solution was added to obtain a final concentration of
0.25 mM. Sampling, deproteination and analysis were done as described
above.
Pure enzymes. Drug solution (0.25 mM) in phosphate buffer of
appropriate pH was prepared and equilibrated at 37C. An aliquot of pure
enzyme or reconstituted enzyme was added. Sampling, deproteination and
analysis were done as described above.
Ocular tissues. Rabbits were sacrificed by rapid i.v. injection of
pentobarbital (100 mg/kg) into the marginal ear vein. Eyes were enucleated
and rinsed in cold phosphate buffer (pH 7.4). The aqueous humor was
obtained by making a single puncture at limbus. Cornea was excised and iris-
ciliary body was isolated. Whole aqueous humor was used in the study without
further dilution. The homogenates of cornea and iris-ciliary body were made in
phosphate buffer (pH 7.4) using a Tekmer SDT tissuemizer for 2 minutes and
then centrifuged at 6000 rpm for 10 mins. The supernatant was used for the


47
study. Protein content was adjusted to 1.25 mg/ml. An aliquot of stock
solution of 3a was added to the homogenates to yield 0.1 mM solution. The
sampling and analysis was performed as described above.
Determination of protein content. Protein content was determined by
Peterson modification of the micro-Lowry method (Protein Assay Kit, Procedure
No. P 5656, Sigma Chemical Company) using bovine serum albumin as the
standard.
In Vitro Activity
Determination of anticholinergic activity
A guinea pig weighing about 400 gms was sacrificed by decapitation and
the terminal ileum was isolated. The ileum was cut into one inch pieces and
each was suspended in a 30 ml jacketed organ bath containing Tyrodes
solution containing 0.1 mM hexamethonium bromide and constantly aerated
with 5:95 mixture of carbon dioxide and oxygen. One end of the ileum strip
was attached to a fixed support at the bottom of the organ bath and the other
end to a isometric force transducer (Model TRN001, Kent Scientific Corp.,
Connecticut) operated at 10 gms range. The ileum was kept at 0.5 gms
tension. Carbachol was used as an agonist and soft drugs and scopolamine
methyl bromide were used as antagonists. The ileum responded by contracting
upon the addition of carbachol. The contractions were recorded on a
physiograph (Desktop Model DMP 4B, Narco Biosystems Inc. Hounston).


48
Cumulative dose response curves of longitudinal contractions to the
addition of carbachol in the absence and presence of antagonists were
obtained following the method of Van Rossum et al. (1963). The pA2 value, an
empirical parameter which defines the negative logarithm of the molar
concentration of the antagonist which produces a two-fold shift to the right of a
concentration-response curve, was used as a measure of comparison between
the affinities of scopolamine and the soft drugs to the muscarinic receptor. In a
plot, the -Log (antagonist) vs the log (Dose ratio 1), called the Schild plot, the
X-axis intercept is the pA2 value.
In Vivo Activity
Mydriatic activity
Twelve male New Zealand albino rabbits weighing 2.5-3 kg were used in
the study. The study was performed in a light and temperature controlled room
with minimum noise disturbance. The rabbits were divided into two groups of
six each. Drug solution in normal saline (50 pi) was administered into one eye
of the one group of the animals. To the other group 50 pL of normal saline was
administered into one eye. The pupil diameters of both the treated eye and the
untreated eye were measured with locally fabricated Haabs scale (Alexandridis,
1985) before the administration of the drug and at appropriate time intervals
until the diameter reaches the time zero value. The difference between the
diameter at a particular time interval and the diameter at time zero is reported


49
as the mydriatic response. The net mydriatic response was obtained by
deducting the normal saline treated value from the corresponding drug treated
value at a particular time interval for the same animal. The experiment was
repeated after a period of 48 hours by switching the groups. The mydriatic
activity was compared using two parameters viz: recovery time (i.e., time taken
for the pupil to return to 0.75 mm above the time zero value) and 24 hour area
under time vs mydriatic response curve (AUC24hrs).
The mean and the standard error of the mean (SEM) were calculated for
the twelve animals tested. Student t-test was performed to determine the
statistical significance at 4 hours for the treated eye and 30 minutes for the
untreated eye. The AUC was calculated by approximate trapezoids and
triangles method for different time intervals.
Mydriatic effect in rabbits on intravenous administration
Groups of four male New Zealand albino rabbits weighing 3-3.5 kg were
used in the study. Pupil diameters of both the eyes were noted. 200 /l/I of
equipotent doses methscopolamine or soft drugs (3a or 6a) dissolved in normal
saline was then injected into the marginal ear vein. Control group received 200
fj\ of saline. Pupil diameter was recorded at various time intervals until it
reached the basal reading.
Effect on the resting heart rate of rats
Groups of four male Sprague-Dawley rats weighing 250-300 gms were
used. The animal was anesthetized with 50 mg/kg of pentobarbital given


50
intraperitoneally. The heart rate was recorded using a physiograph (Projector
Model Type PMP-4B, Narco Biosystems, Inc). After a 10 minute equilibration
period, a dose of drug dissolved in saline was injected as a bolus into the left
jugular vein. The control group received saline. The heart rate was recorded at
periodic intervals.
Effect on acetylcholine induced bradycardia
Groups of four male Sprague-Dawley rats weighing 250-300 gms were
used. The animal was anesthetized with 50 mg/kg of pentobarbital given
intraperitoneally. The heart rate was recorded using a physiograph (Projector
Model Type PMP-4B, Narco Biosystems, Inc.). After a 10 minute equilibration
period, an equipotent dose of methscopolamine or soft drug (3a) or metabolite
(3g) dissolved in saline was injected into the left jugular vein. Control group
received saline. At appropriate time intervals 50 fj\ of 0.1% w/v solution of
acetylcholine in saline was injected into right jugular vein and the heart rate was
recorded.
In Vitro Penetration Studies
Octanol A/vater partition coefficient
The solution of drug in water (5 ml, 1 mM) was vigorously shaken with 5
ml of n-octanol at 25C in a sealed container for 24 hours (Grass and Robinson,
1984). The content of the drug in water and octanol was quantitated by HPLC.


51
Transdermal penetration
Female hairless-mice were used for the study. Two-compartment,
vertical diffusion cells (Kresco Engineering, Palo Alto) with a surface area of 7.1
cm2 were used in the study. The animals were sacrificed by cervical
dislocation. Whole skin was dissected carefully from the abdomen and back.
The underlying fat tissue was gently peeled off with a forceps. The skin was
immediately mounted in the cell with dermal side facing the receptor chamber.
The receptor chamber was then filled with 40 ml of isotonic phosphate buffered
saline of pH 7.1 containing 0.1% v/v of formaldehyde. Drug solution (1 ml, 0.1
M) in isotonic phosphate buffered saline was placed on the top of the skin. The
entire system was stirred continuously at 200 rpm and incubated at 32C. At
appropriate time intervals, 500 /jI sample was withdrawn and the receptor was
replenished with an equal quantity of buffer. The sample withdrawn was diluted
with 500 /il of acetonitrile and vortexed. The mixture was analyzed by HPLC for
drug and metabolites.
Transcorneal penetration
Male New Zealand albino rabbits weighing 2.5-3.5 kg were used in the
study. Transcorneal permeation system of Crown Glass Company (Somerville,
NJ) was used in this study. This design is a two equal sized side-by-side cell
assembly which keeps the mounted cornea wrinkle free while maintaining
corneal curvature. The animals were sacrificed by a rapid injection of
pentobarbital (100 mg/kg) into marginal ear vein. The corneas were isolated by


52
making a small incision in the sclera approximately 2 mm from limbus and then
cutting circumferentially. The isolated cornea were washed with bicarbonated
Ringer and immediately mounted on the receptor cell. The donor cell was then
placed properly in position and the two cells were clamped securely. First the
receptor cell was filled with 4 ml of bicarbonated Ringer followed by the donor
cell with 4 ml of 0.025 M solution of drug in normal saline. The system was
kept at 37C by connecting to a circulating water bath and continuously stirred.
Oxygen:Carbon dioxide (95:5) was continuously bubbled into both receptor and
donor chambers. Samples (200 /jI) were withdrawn at appropriate time intervals
form receptor cell and the chamber was replenished by addition of 200 fjI of
bicarbonated Ringer. The samples were analyzed by HPLC without further
treatment.
Statistical analysis
All the values reported are mean SEM (Standard error of mean =
Standard deviation/v(n-1)). Student "t" test was performed to assess the
significance of the results.


CHAPTER 4
RESULTS AND DISCUSSION
Syntheses
The physical and spectral characteristics of the compounds synthesized
are enumerated below:
Monoalkvl 2-phenvlpropanedioic acids (4 a-fi: The compounds were
identified by NMR spectroscopy and the purity was tested with TLC. HOOC-
CH(Phenyl)-COOR 4a ethyl: M.P. 80C, Yield 69%, 4b n-Propyl: 65C, 71%, 4c
n-Butyl: 66C, 72%, 4d n-Pentyl: 75C, 66%, 4e c-Hexyl: 80C, 53% and 4f ethyl-
c-hexyl: 73C, 60%. The melting points of 4a and 4e coincided with the values
reported In the literature (Hammer et al. 1988).
(2R.2S1 ( + 11-alkvl. 3-[3a-i8-methyl-6l3.7B-epoxv-8-azabicvclo [3.2.1) oct-3-vn] 2-
phenvlpropanedioic acids 5 a-fi
5a. (2R,2S) ()1 -ethyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylpropanedioic acid: Yield, 81%, brownish viscous liquid, 1H
NMR (CDCIg): 6 1.2 (t, 3H, CH3), 2.0 (m, 4H, bicyclic), 2.5 (s, 3H, N-CH3), 3.2
(m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s,
1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic).
53


54
5b. (2R,2S) ()1-n-propyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 76.5%, viscous liquid, 1H NMR
(CDCIg): CHg), 3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, O-
CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H,
aromatic).
5c. (2R,2S) ()1-n-butyl, 3-[3o:-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 81%, viscous liquid, 1H NMR
(CDCIg): S 0.9 (t, 3H, CH3), 1.4-2.2 (m, 8H, bicyclic and CH2-CH2), 2.5 (s, 3H, n-
chg), 3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2),
5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic).
5d. (2R,2S) ()1-n-pentyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 79%, viscous liquid, 1H NMR
(CDCI3): s 0.9 (t, 3H, CHg), 1.2-2.2 (m, 10H, bicyclic and CH2-CH2-CH2), 2.5 (s,
3H, N-CH3), 3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H,
0-CH2), 5.0 (s, 1H,Ar-CH), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H,
aromatic).
5§. (2R,2S) ()1-c-hexyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 65%, brownish viscous liquid, 1H
NMR (CDCy: S 1.1-2.2 (m, 14H, bicyclic and cyclohexyl), 2.5 (s, 3H, N-CHg),
3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 1H, CO-O-


55
CH,cyclohexyl), 5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m,
5H, aromatic).
5f. (2R,2S) ()1-ethyl-c-hexyl, 3-[3a-(8-methyl-6(3,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 75%, viscous liquid, 1H NMR
(CDCI3): S 1.1-2.2 (m, 17H, bicyclic and cyclohexyl ethyl), 2.5 (s, 3H, N-CH3),
3.2 (m, 2H, CH-N-CH), 3.3 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7 (m, 2H,
CH-O-CH epoxide), 4.1 (q, 1H, CO-0-CH2,cyclohexylethyl), 4.9 (s, 1H, Ar-CH),
5.1 (m, 1H, CH-O-CO, scopyl) and 7.3 (m, 5H, aromatic).
(2R.2S1 (+11-alkvl. 3-[3a-(8.8-dimethvl-6B.7B-epoxv-8-azabicvclo [3.2.1] oct-3-
vll] 2-phenvlpropanedioic acid methvl sulfates (3 a-fl
39. (2R,2S) ()1-ethyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 82%. m.p.
104-106C. 1H NMR (DMSO-d6): 5 1.2 (t, 3H, CH3), 2.0 (m, 4H, bicyclic), 3.2
(m, 2H, CH-N-CH), 3.3 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7 (m, 2H, CH-
O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-O-
CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 360. Infrared spectrum (KBr pellet):
5.9 /m C=0 stretching, 7.9 /L/m C-N stretching, 8 /L/m epoxy ring breathing,
and 12.2 /L/m "12 /j band of epoxide. The ultraviolet spectrum (in acetonitrile)
showed three maxima at 264 nm (e 45.4), 258 nm (e 85.8) and 252 nm (e
45.4). Elemental analysis: calculated/found; C, 51.79/52.01; H, 6.01/6.09; and
N, 2.86/2.92.
3b. (2R,2S) ()1-n-propyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 81.5%.


56
m.p.116-118C. 1H NMR (DMSO-d6): 5 0.9 (t, 3H, CH3), 1.4-2.2 (m, 6H, bicyclic
and CH2), 3.2 (m, 2H, CH-N-CH), 3.3 3.5 (3s, 9H, 2 X N-CH3 and CH3S04),
3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m,
1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 374. Calculated /found;
C, 53.44/53.64; H, 6.68/6.57; and N, 2.83/2.92.
3c. (2R,2S) ()1-n-butyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate: Yield 84%. m.p.100-
102C. 1H NMR (DMSO-d6): 5 0.9 (t, 3H, CH3), 1.4-2.2 (m, 8H, bicyclic and
CH2-CH2), 3.2 (m, 2H, CH-N-CH), 3.3 3.5 (3s, 9H, 2 X N-CH3 and CH3S04),
3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m,
1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 388. Calculated/found;
C, 55.31/55.21; H, 6.61/6.69; and 2.81/2.76.
3d. (2R,2S) ()1-n-pentyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 79%. m.p.106-
107C. 1H NMR (DMSO-d6): 5 0.9 (t, 3H, CH3), 1.2-2.2 (m, 10H, bicyclic and
CH2-CH2-CH2), 3.2 (m, 2H, CH-N-CH), 3.3 -3.5 (3s, 9H, 2 X N-CH3 and
CH3S04), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H,Ar-CH),
5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M + 402.
Calculated/found; C, 56.14/55.99; H, 6.82/6.89; and N, 2.73/2.64.
3e. (2R,2S) ()1-c-hexyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 75%.
m.p.183C. 1H NMR (DMSO-d6): 5 1.1-2.2 (m, 14H, bicyclic and cyclohexyl),


57
3.2 (m, 2H, CH-N-CH), 3.3 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7 (m, 2H,
CH-O-CH epoxide), 4.2 (q, 1H, CO-O-CH,cyclohexyl), 5.0 (s, 1H, Ar-CH-), 5.1
(m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 414. Calculated/
found; C, 56.18/55.96; H, 6.74/6.67; and N, 2.62/2.60.
3f. (2R,2S) ()1-ethyl-c-hexyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-
azabicyclo [3,2,1] oct-3-yl)] 2-phenylpropanedioic methyl sulfate : Yield 85%.
m.p.145C. 1H NMR (DMSO-d6): s 1.1-2.2 (m, 17H, bicyclic and cyclohexyl
ethyl), 3.2 (m, 2H, CH-N-CH), 3.3 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7
(m, 2H, CH-O-CH epoxide), 4.1 (q, 1H, CO-0-CH2,cyclohexylethyl), 4.9 (s, 1H,
Ar-CH), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 442.
Elemental analysis: calculated/found; C, 58.58/58.49; H, 7.05/7.11; and N,
2.53/2.54.
3g. [3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo [3,2,1] oct-3-yl)]
benzeneacetic acid dimethyl sulfate: White crystalline powder, yield : 82%, M.P.
153C, m+: 288. NMR data: (DMSO-d6) 1.2-2.2 (m, 4H) bicyclic, 2.4-2.7 (3s,
9H) N + (CH3)2 and CH3S04, 2.7-3.7 (m, 8H) N + (CH)2, 0(CH)2 (epoxy), CH2-COO
and Ar-CH2, 4.9 (t, 1H) bicyclic CH-OCO and 7.3 (m, 5H) aromatic. Elemental
analysis: calculated/found; C, 54.14/53.95; H, 6.27/6.31; and N, 3.51/3.46.
The syntheses of the phenylmalonic acid analogs (3 a-f) was
accomplished by sequential esterification of phenylmalonic acid by the relevant
alcohol followed by scopine. The diester obtained was then quaternized.
Dimethylformamide was used as a catalyst in the preparation of acid chlorides.


58
Dimethylformamide is known to form a highly activated complex on reacting
with thionyl chloride which collapses on reaction with an acid to yield acid
chloride (Zaugg et al. 1960). In the first esterification step, the yields were not
usually very high due to the formation of diester of phenylmalonic acid instead
of the monoester alone. In the second esterification step and quaternization
step the yields were high. The quaternary compounds are extremely
hygroscopic and were stored in vacuo.
Several attempts were made to synthesize the hypothetical carboxylate
metabolite (2).
1. esterification of phenylmalonic acid with scopine;
2. esterification of scopine with monobenzyl phenylmalonate and debenzylation
by hydrogenation (in methylene chloride or diethyl carbonate);
3. debenzylation of quaternized scopine ester of monobenzyl phenylmalonate
in dilute acetic acid or acetone;
4. esterification of mono-TBDMS phenylmalonate with scopine and
deprotection with FN(t-Bu)4; and
5. enzymatic hydrolysis of 3a with porcine liver esterase in buffers.
The end product was 3g in all the above reactions instead of 2.
Compound 2 seems to be unstable in reaction conditions. This is interesting
since there are no reports in the literature about decarboxylation of carbenicillin,
the only other derivative of phenylmalonic acid in therapy.


59
Alkvl bromoacetates (7 a-e): The structures were confirmed by NMR
spectroscopy. Br-CH2-COO-C2H5 was obtained from Aldrich Chemical
Company. Br-CH2-COO-n-propyl: Yield=84%, B.P. 69C at 10 mm of Hg. Br-
CH2-COO-n-butyl: 90%, 56C. Br-CH2-COO-cyclohexyl: 89%, 71 C. Br-CH2-
COO-benzyl: 84%, 129C.
4-alkvl 2-phenvlbutanedioic acids (8 a-e): The structures were confirmed
by NMR spectroscopy and the purity was tested with TLC. HOOC-CH(Phenyl)-
CH2-COO-R: ethyl: M.P. 97C, Yield 65%. n-propyl: 58C, 66%, n-butyl: 60C,
68%, c-hexyl: 63C, 60% and benzyl: 100C, 62%. The melting point of 8a
coincided with the value reported in literature (Gunes et al 1992), but differs
from the value reported in another report, 87-89C (Kofron and Wideman,
1971).
2R.2S1 f+H-alkvl 1-f3a-f8-methvl-6B.78-eDOXv-8-azabicvclo [3.2.1] oct-3-vh] 2-
phenvlbutanedioic acids (9a-e1:
9a. (2R,2S) (+)4-ethyl 1-[3a-(8-methyl-68,76-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: Viscous dark brownish liquid, Yield, 90%,
NMR data: (CDCI3) 1.2 (t, 3H, CH3), 1.3-2.2 (m, 4H, bicyclic), 2.4 (s, 3H, N-
CHg), 2.8-3.5 (m, 6H, N(CH)2, 0(CH)2 (epoxy), CH2-COO), 4.1 (m, 3H, COO-
CH2), Ar-CH-COO, 5 (t, 1H, bicyclic CH-OCO), 7.3 (m, 5H, aromatic).
9b. (2R,2S) ()4-n-propyl 1-[3a-(8-methyl-68,76-epoxy-8-azabicyclo
[3,2,1] oct-3-yl)] 2-phenylbutanedioic acid: viscous liquid, 85%, NMR data:
(CDCIg) 0.9 (t, 3H, CH3), 1.5 (m, 2H, -CH2-), 2.2 (m, 4H, bicyclic), 2.5 (s, 3H, N-


60
CH3), 2.8-3.5 (m, 6H, CH2-COO, N(CH)2, 0(CH)2 epoxy), 4.1 (m, 3H, COO-CH2)
and Ph-CH, 4.9 (t, 1H, bicyclic CH) and 7.3 (s, 5H, aromatic).
9c. (2R,2S) (+)4-n-butyl 1-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: viscous liquid, 81%, NMR data: (CDCI3) 0.9
(t, 3H, CH3), 1.5 (m, 4H, CH2-CH2-), 2.2 (m, 4H, bicyclic), 2.5 (1s, 3H, N-CH3),
2.8-3.5 (m, 6H, CH2-COO, N(CH)2, 0(CH)2 epoxy), 4.1 (m, 3H, COO-CH2) and
Ph-CH, 4.9 (t, 1H, bicyclic CH) and 7.3 (s, 5H, aromatic).
9d. (2R,2S) ()4-c-hexyl 1-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: viscous liquid, 75%, NMR data: (CDCI3)
1.2-1.7 (m, 10H, Cyclohexyl), 2.2 (m, 4H, bicyclic), 2.5 (s, 3H, N-CH3), 2.8-3.5
(m, 6H, CH2-COO, 0(CH)2 epoxy), 4.1 (q, 1H, Ph-CH), 4.8 (m, 1H, COO-CH,
cyclohexyl), 4.9 (t, 1H, bicyclic CH) and 7.3 (s, 5H, aromatic).
2e. (2R,2S) ()4-benzyl 1-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: dark brownish viscous liquid, 88%, NMR
data: (CDCI3) 1.3-2.2 (m, 4H, bicyclic), 2.4 (s, 3H, N-CH3), 2.7-3.7 (m, 6H,
N + (CH)2, 0(CH)2 epoxy, CH2-COO), 4.1 (m, 1H, Ar-CH-COO), 4.9 (t, 1H,
bicyclic CH-OCO), 5.0 (s, 2H, Ar-CH2) and 7.3 (m, 5H, aromatic).
(2R.2S1 f-t-14-alkyl 1-[3a-(8.8-dimethyl-6B.7B-epoxv-8-azabicyclo [3.2.1] oct-3-yl)]
2-phenvlbutanedioic acid methyl sulfates (6 a-e):
6a. (2R,2S) ()4-ethyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1] oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: White crystals, Yield =
93%, M.P. 118-120C, m + 374. NMR data: (DMSO-d6) 1.2 (t, 3H) CH3, 1.3-2.2
(m, 4H) bicyclic, 2.4-2.7 (3s, 9H) N + (CH3)2, CH^O^, 2.7-3.7 (m, 6H) N + (CH)2,


61
0(CH)2 (epoxy), CH2-COO, 4.1 (m, 3H) COO-CH2, Ar-CH-COO, 5 (t, 1H)
bicyclic CH-OCO, 7.3 (m, 5H) aromatic. Elemental analysis: calculated/found;
C, 54.44/54.63; H, 6.39/6.49; and 2.89/2.96.
6£. (2R,2S) ()4-n-propyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: Extremely hygroscopic
white powder, 94%, M.P. 114C, m + 388, NMR data: (DMSO-d6) 0.9 (t, 3H)
CH3, 1.5 (m, 2H) -CH2-, 2.2 (m, 4H) bicyclic, 2.5-2.7 (3s, 9H) N(CH3)2 and
CHgSO/, 2.8-3.5 (m, 6H) CH2-COO, 4.1 (m, 3H) COO-CH2 and Ph-CH, 4.9 (t,
1H) bicyclic CH and 7.3 (s, 5H) aromatic. Calculated/found; C, 57.38/57.37; H,
6.70/6.67; and N, 2.70/2.78.
§c. (2R,2S) ()4-n-butyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: Extremely
hygroscopic white powder, 87%, M.P. 120C, m+: 402, NMR data: (DMSO-d6)
0.9 (t, 3H) CH3, 1.5 (m, 4H) CH2-CH2-, 2.2 (m, 4H) bicyclic, 2.5-2.7 (3s, 9H)
N(CH3)2 and CH3S04, 2.8-3.5 (m, 6H) CH2-COO, 4.1 (m, 3H) COO-CH2 and Ph-
CH, 4.9 (t, 1H) bicyclic CH and 7.3 (s, 5H) aromatic. Calculated/found; C,
54.24/54.41; H, 6.82/6.77; and N, 2.64/2.65.
6d. (2R,2S) ()4-c-hexyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: Extremely
hygroscopic white powder, 87%, M.P. 118C, m+: 428, NMR data: (DMSO-d6)
1.2-1.7 (m, 10H) Cyclohexyl, 2.2 (m, 4H) bicyclic, 2.5-2.7 (3s, 9H) N(CH3)2 and
CH3S04', 2.8-3.5 (m, 6H) CH2-COO, 4.1 (q, 1H) Ph-CH, 4.8 (m, 1H) COO-CH


62
cyclohexyl, 4.9 (t, 1H) bicyclic CH and 7.3 (s, 5H) aromatic. Calculated/found;
C, 57.90/57.71; H, 6.86/6.88; and N, 2.60/2.59.
6§. (2R,2S) ()4-benzyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1] oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: White crystals, 96%,
M.P. 98C, m+: 436. NMR data: (DMSO-d6) 1.3-2.2 (m, 4H) bicyclic, 2.4-2.7
(3s, 9H) N+(CH3)2 and CH3S04', 2.7-3.7 (m, 6H) N + (CH)2, 0(CH)2 (epoxy), CH2-
COO, 4.1 (m, 1H) Ar-CH-COO, 4.9 (t, 1H) bicyclic CH-OCO, 5.0 (s, 2H) Ar-CH2
and 7.3 (m, 5H) aromatic. Calculated/found; C, 57.44/57.27; H, 6.21/6.17; and
N, 2.48/2.45.
2a. (2R,2S) () 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo [3,2,1] oct-
3-yl)] 2-phenylbutanedioic acid methyl sulfate: White crystals, 80%, M.P. 80C,
m + : 346. NMR data: (DMSO-dg) 1.3-2.2 (m, 4H) bicyclic, 2.4-2.7 (3s, 9H)
N + (CH3)2 and CH3S04, 2.7-3.7 (m, 6H) N + (CH)2, 0(CH)2 (epoxy), CH2-COO,
4.1 (m, 1H) Ar-CH-COO, 4.9 (t, 1H) bicyclic CH-OCO, 7.3 (m, 5H) aromatic and
10.5 (s, 1H) COOH. Calculated/found; C, 50.52/50.61; H, 6.11/6.15; and N,
2.95/3.01.
The esterification of bromoacetic acid with the relevant alcohols by
azeotropic distillation in benzene gave high yields of bromoacetate esters. The
next alkylation step of phenylacetic acid is extremely moisture sensitive.
Precautions were taken to exclude moisture from the reaction setup. Dry
nitrogen was bubbled continuously to maintain a positive pressure inside the


flask. The yields of the alkylated products were in the range of 60 70% and
are in agreement yields reported in the literature (Kofron and Wideman, 1972).
Coupling of scopine and subsequent quaternization with dimethyl sulfate
proceeded smoothly with cumulative yields ranging from 65 90%. Catalytic
hydrogenation of 6e gave the carboxylate metabolite 2a.
Anticholinergic Activity
In Vitro Activity
All the values are means of four determinations. The dose response
curves in the absence and presence of increasing concentrations of antagonist
were generated by van Rossums method (1963). This is the most widely used
method for determination of the affinity of agonists and antagonists to receptors
in isolated organs. In this experiment, hexamethonium was added to the
medium to prevent any interaction of carbachol with the nicotinic receptors. So
the response obtained was exclusively from the action of the compounds on the
muscarinic receptors.
Methscopolamine and all the soft drugs showed reversible competitive
antagonism with parallel dextral shifts of dose response curve with increasing
concentrations of antagonist without depression of the maximal response
(Figure 9). Washing off the antagonist completely restored the original
sensitivity to carbachol clearly indicating the competitive nature of the
antagonism. The pA2 values were calculated according to Schilds method


64
% of maximal reaponae
1.000E-07 1.000E-06 I.OOOE-OS
Agonist (moles)
0 f- 0.12 0.4* -B- 0.S 0.00 -o- 2.4 -A- 4.9
Antagonist concentration ImM)
Figure 9: Representative dose response curves from guinea pig ileum assay.
Agonist: Carbachol. Antagonist: 3c. Each curve represents the response
obtained to increasing concentrations of agonist in the presence of a certain
concentration of antagonist.


Log (Dose Ratio
65
5.00 5.50 6.00 6.50 7.00 7.50 8.00
-Log (Antagonist)
Figure 10: A representative Schild plot. Each value of dose ratio (DR) is the
mean of four determinations at a particular concentration of the antagonist (3c).
X axis intercept is pA2.


66
Compound
pA2
r2
Slope
1
9.50
0.992
-1.067
3a
7.40
0.915
-0.920
3b
7.85
0.935
-1.010
3c
7.40
0.974
-1.003
3d
7.00
0.982
-0.876
3e
5.85
0.967
-1.039
3f
5.90
0.996
-1.503
6a
7.20
0.985
-0.967
6b
6.80
0.950
-0.895
6c
6.03
0.972
-1.080
6d
5.50
0.971
-0.934
6e
5.40
0.994
-0.928
2a
5.12
0.966
-1.102
3g
7.60
0.994
-1.000
Tropicamide
6.02
0.992
-1.028
Table 1: In vitro anticholinergic activity guinea pig ileum assay. All values are
means of four determinations.


67
(Figure 10) (Table 1). The pA2 values unambiguously quantify the potency of
an antagonist for a receptor. The slopes of Schild plots were around unity for
all compounds tested except for compound 3f. If the linear regression
coefficients obtained in the Schild plot are close to unity, the agents are
considered to be pure antagonists (Kenakin, 1982). The deviation of slope from
unity for compound 3f could be due to insufficient equilibration time of the
antagonist for the receptors or the activation of other receptors at high
concentration of the compound which might produce a physiological
antagonism of the agonist responses (Kenakin, 1982). An equilibration time of
ten minutes was allowed for the antagonists before the addition of carbachol in
all determinations. Probably this time was not sufficient for compound 3f to
equilibrate.
The pA2 value of methscopolamine determined in this study is identical to
the value reported in the literature (Triggle & Triggle, 1976). The pA2 values
were found to decrease with increasing chain length. The importance of
hydrophobic forces in cholinergic drug receptors have long been recognized
(van Rossum and Ariens, 1957). In a study carried out with straight chain
esters of phenylacetic acid, the following conclusions have been drawn
(Banerjee & Lien 1990).
CH(R)-COO-CH2-CH2-N(C2H5)2.HCl


68
1. substitution of benzylic carbon with 1 or 2 carbon alkyl groups increased
anticholinergic activity, while further increase in chain length decreased activity,
probably due to steric interactions,
2. unsubstituted phenylacetic acid derivative had very low activity,
3. substitution of benzylic carbon with hydroxymethyl group markedly
enhanced the activity, suggesting the possibility of a weak hydrogen bonding
site at the receptor site. In the present study, -CH2OH group of
methscopolamine is substituted with either -COOR or CH2COOR group. The
most potent compounds of both the series are about 2 log units less potent
than methscopolamine. By design, this compromise of potency is not a matter
of concern as long as the compounds are deactivated, in vivo, to less potent
and less toxic moieties. The pA2 values of both the phenylmalonic and
phenylsuccinic series have been found to decrease with increasing chain length
(Table 1). Contrary to the observation that the unsubstituted phenylacetic acid
esters had very low activity (Benerjee & Lien, 1990), 3g, a phenylacetic acid
derivative of scopine is active. The pA2 values of the phenylmalonic analogs are
about one half to one log unit higher than the corresponding phenylsuccinic
analogs. This could be due to the increased steric hinderance due to the
introduction of a methylene group. 2a, the hypothetical lead metabolite of
phenylsuccinic analogs, is about 100 times less potent than the most active
compound (6a) of the series suggesting that the soft drugs (6 a-e) on
hydrolysis, in vivo, would yield a metabolite which is practically inactive.


69
In Vivo Activity
In vivo activity was studied by three methods viz: mydriatic activity on
unilateral topical instillation into rabbit eye and on systemic administration into
rabbits, and muscarinolytic activity against acetylcholine induced bradycardia on
systemic administration into rats.
Mydriatic activity
The mydriatic dose response curves were obtained by administering
increasing concentrations of the drugs until the maximum dilation was achieved
(Figures 11, 12 & 13). Due to the mechanical restriction, the pupil will not dilate
infinitely. Hence the lowest dose which produces the maximum achievable
dilation was used as the dose for comparison. Methscopolamine (0.05% w/v),
tropicamide (0.33% w/v), 3a (0.75% w/v), 3b (1% w/v), 3c (1.25% w/v), 3d (1%
w/v), 6a (5% w/v), 6b (5% w/v) and 6c (1.5% w/v) produced equieffectiveness.
3e exhibited only 89% of the maximum effect even at higher concentrations. 3f
(0.5% w/v) and 6d (2% w/v) produced intense irritation with erythema, tears
and secretions, and were not studied further. Moderate irritation was noticed
with tropicamide, 3d, 3e and 6c, whereas none to mild irritation was noticed
with methscopolamine, 3a, 3b, 3c, 6a and 6b. The maximal dilation was
achieved about 30 minutes after instillation without any significant difference
between 1, tropicamide and soft drugs.
At equieffective doses, the percentages of AUC24hr for the drugs to that
of methscopolamine (Figure 14) are: 3a 23.2%, 3b 45.6%, 3c 60.6%, 3d -


70
Concentration (w/v)
Concentration (% w/v)
Concentration (% w/v) Concentration w/v)
Figure 11: Mydriatic dose response curves in treated rabbit eye. Each value is the mean of
twelve readings and error bar represents SEM. The arrows denote the equieffective doses.


71
120
% of maximum dilation
100 -
80
eo -
40
20 -
i i i i i iiii i i i i i i i i i i i i i r
0.0001 0.0010 0.0100
Log molar concentration
3a 3b
3c
Trop
3d
3a
till
0.1000
Figure 12: Mydriatic dose response curves in rabbit treated eye after unilateral
instillation. Each value is the mean of twelve readings.


72
% of maximum dilation
Log molar concentration
+ 1 ^Trop 6b -^6c
Figure 13: Mydriatic dose response curves in rabbit treated eye after unilateral
instillation. Each value is the mean of twelve readings.


73
AUC(24hrs)
Control
Figure 14: AUC24hrs for the treated after unilateral instillation into the rabbit eye.
Each value is the mean of twelve readings. Error bar represents SEM.


74
Hours
Figure 15: Duration of mydriatic activity after unilateral instillation into the rabbit
eye. Each value is the mean of twelve readings. Error bar represents SEM.


75
Pupil dilation (mm)
^ 3 -H 3b 3c & Sc Me Br ^ Trop 34 "Ap 3c
1 Control
Figure 16: Time vs mydriatic response curves in the treated eye after unilateral
administration of equieffective doses. Each reading is the mean of twelve
readings.


76
187%, 3e 49.1%, 6a 22.4%, 6b 32.7%, 6c 60% and tropicamide 58.3%.
At equieffective doses the recovery periods for the drugs (Figure 15) are as
follows: methscopolamine 20 hrs, tropicamide 5.7 hrs, 3a 3.7 hrs, 3b 6.8
hrs, 3c 9.8 hrs, 3d 36 hrs, 3e 5.9 hrs, 6a 3.0 hrs, 6b 4.5 hrs and 6c 6.9
hrs. The time vs mydriatic response curves for the drugs at equieffective doses
are summarized in Figure 16. The control eye into which only vehicle was
administered did not show any significant dilation. At 4 hours, 3a and 6a, and
methscopolamine (0.05%) exhibited significant difference in dilation (P < 0.001).
At 4 hours, 3a and 6a, and tropicamide (0.33%) also exhibited significant
difference in dilation (P < 0.005).
The time course of mydriatic activity in the untreated eyes (at
equieffective doses) is depicted in Figure 17. At equieffective doses, the
percentages of AUC6hr to that of methscopolamine (Figure 18) are:
tropicamide 17.6%, 3a 2.3%, 3b 15.7%, 3c 10.9%, 3d 17.5%, 3e 9.8%,
6a 10.4%, 6b 8.1% and 6c 13.3%. At 30 minutes, the difference in dilation
between 3a and 6a, and methscopolamine was found to be significant (P <
0.001). At 30 minutes, the difference in dilation between 3a and tropicamide
was also found significant (P < 0.05).
At equieffective doses all the soft drugs tested, except 3d, showed
significantly shorter duration of mydriatic action in the treated eye than
methscopolamine. Compounds 3a and 6a exhibited significantly shorter
duration than tropicamide. The short duration of mydriatic action could possibly


77
Pupil dilation (mm)
Time (hrs)
-0- 3a 3b 3c "0" 1 Trop 3d 3 Hi* Control
Figure 17: Time vs mydriatic response curves in the untreated eye after
unilateral administration of equieffective doses. Each reading is the mean of
twelve readings.


78
1 Trop 3a 3b So 3d 3o Control 3a 6b 6o
Figure 18: AUC6hrs for the untreated eye after unilateral instillation of equieffective
doses. Each value is the mean of twelve readings. Error bar represents SEM.


79
be due to rapid hydrolysis of soft drugs in the eye. A parabolic relationship has
been reported between chain length and in vitro hydrolytic rates of ester
prodrugs by ocular esterases (Chang and Lee, 1983); compounds with 4-5
carbon chain exhibiting fastest hydrolytic rates. In this study, compounds
synthesized with a 2 carbon alcohol (ethyl, 3a and 6a) and 3 carbon alcohol
(propyl, 3b and 6b) were the shortest acting. In fact, 3d with an n-pentyl ester
was even longer acting than 1. This may be due to binding to ocular pigments
through hydrophobic interactions (Koneru et al. 1986).
The extent of mydriasis is an instantaneous response of the quantity of
the drug residing in the biophasic tissue (iris musculature) and hence the time
course of the pupil response directly reflects the change of drug concentration
in the iris (Mishima, 1981). More than 90% of topically administered drugs have
been reported to be drained into systemic circulation through nasolacrimal duct
without entering the interior of the eye. This results in a high incidence of
systemic side effects after ocular administration of drugs. In this study, the
drugs were administered into only one eye of the rabbit. The other eye served
as an indicator of systemic absorption of the drug and its subsequent side
effects. The untreated eye was observed to dilate in methscopolamine treated
animals but not with soft drug treated animals. This could be due to the
persistence of methscopolamine in the systemic circulation in comparison to the
rapid systemic inactivation of soft drugs.


80
Clinically mydriatics are used more for diagnostic rather than for
therapeutic purposes. Hence persistence of the mydriatic action after a certain
required period of time is both unnecessary and inconvenient. One of the aims
of the present study was to reduce the duration of mydriatic action by
incorporation of a metabolically labile moiety (ester) into the molecule. The
esterases present in the interior of the eye, especially in the iris-ciliary body, will
aid in the metabolism of the drug to the more polar metabolite. The net result
will be a reduction in the duration of activity as shown by the mydriatic activity
studies with soft drugs containing a 2 or 3 carbon ester chain (3a, 3b, 6a and
6b). Further increase in ester chain (3d) gradually increased duration of
mydriatic action. This could be due to two reasons: inefficient hydrolysis by
carboxylesterases and non-specific binding of the drug to ocular pigments by
hydrophobic interactions. The soft drugs with shorter chain length are found to
have no or very low irritation potential on topical administration into the eye.
The irritation potential is found to increase with increasing chain length as was
seen with compounds 3f and 6d. There are reports in the literature which
suggest the relationship between hydrophobic side chains and irritation
potential of pharmaceutical compounds (Fraunfelder, 1989).
Mydriatic activity of intravenous administration to rabbits
Equipotent doses of methscopolamine and soft drugs 3a and 6a were
administered into rabbit ear vein and the mydriatic activity was followed (Figure
19). At all the doses tested, 1 exhibited an extremely long duration of mydriatic


81
Pupil dilation (mm)
3a 6a
Figure 19: Mydriatic activity after intravenous administration of equipotent doses
to rabbits. Each value is the mean of eight readings (from four animals).


82
action indicating the persistence of the compound in systemic circulation for
prolonged periods of time. Comparatively, 3a and 6a exhibited significantly
shorter duration of mydriatic action indicating facile metabolism and elimination
from the circulation. At a dose of 100 /jg of 1 and equipotent doses of 3a and
6a, mydriatic activity lasted for about 30 hours with 1 compared to 4.5 hours
with 3a and 8 hours with 6a.
The long duration of mydriatic action after systemic administration of 1
into rabbits is consistent with the result reported for humans. Prolonged
suppression of resting pulse rate, decreased secretion of saliva and long lasting
mydriatic activity have been reported in humans after intramuscular
administration of 1 (Brand, 1969).
Muscarinolytic activity on intravenous administration into rats
Effect on the resting heart rate. All the compounds tested (1, 3a and 6a)
showed no significant change in the resting heart rate during the one hour
period after the administration.
Effect on acetylcholine induced bradycardia. The muscarinolytic activity
of 1, 3a and 3g were studied by intravenous administration of equipotent doses
into anaesthetized rats followed by bolus intravenous injections of acetylcholine
at different time intervals. The blockade of cholinergic activity was studied by
recording the heart rate (Figure 20). Methscopolamine completely inhibited
activity of acetylcholine even one hour after the administration compared to 10
minutes with 3a and 40 minutes with the metabolite 3g. Saline control did not


83
exhibit any muscarinoiytic activity. From the in vitro metabolism results
described in later sections it can be concluded that rat metabolizes 1 less
efficiently than the soft drug 3a or the metabolite 3g. This study shows that
compared to methscopolamine, the soft drug 3a undergoes more facile
metabolism probably producing an inactive metabolite. The activity profile of
compound 3g is discussed in a later section.
In Vitro Stability
Analytical procedure
An accurate and reproducible qualitative and quantitative method of
analysis is needed for the estimation of compounds and their degradation and
metabolic products. The High Pressure Liquid Chromatography (HPLC) has
been the most widely used method for such purposes. The main advantages
of HPLC are its reproducibility, accuracy of quantitation and economy of
operation. Various types of columns and detectors have been developed for
HPLC. In the present study, a reverse phase CN column was used. The
detection was made with the UV absorbance detector set at 254 nm. The
mobile phase consisted of varying proportions of acetonitrile and phosphate
buffer. Octane sulfonic acid was used as an ion-pairing agent. The inclusion of
an ion-pairing agent has greatly improved the resolution of the peaks. The
concentration vs area under the peak plot showed linearity (r = 0.999) for the
range of 0.2 to 2 ¡jg of injected sample with a detection limit of 0.2 [jg


84
O
-10
-20
-30
-40
-60
-60
-70
0 10 20 30 40 50 60
Time (mins)
Control H 3a 3g 1
% decrease in heart rate
70
Figure 20: Muscarinolytic activity against acetylcholine induced bradycardia in rats
after intravenous administration of equipotent doses. Each value is the mean of
four determinations.


85
(approximately 0.4 nanomoles) of injected sample. Due to low absorbance of
the compounds, the detection limit of the compound is relatively high.
In vitro stability in buffers
The stability of 3a (chosen as a prototype of the series) was studied in
USP buffers (pH 1.5 to 9). 3a exhibited a typical inverted bell shaped pH profile
(Figure 21). The optimum stability was observed in pH range 3.5-4. 3a was
extremely unstable in alkaline solutions.
The degradation pathways were different in acidic and alkaline media
(Figure 22). In acidic medium, 3g was the major product. This compound is
the decarboxylation product of the carboxylate metabolite (2) generated from
the hydrolysis of ethyl ester group of 3a. In alkaline medium, monoethyl
phenylmalonate and quaternized scopine were the major degradation products
with traces of 3g. In alkaline medium, the hydrolytic attack is occurring at
scopine ester bond rather at the ethyl ester bond. This could be due to the
inductive effect of quaternary head group making the scopine ester more
susceptible for nucleophilic attack.
The stability of 3a was studied in pH 4 buffer at 63, 56, 46 and 37C.
An Arrhenius plot was generated by plotting (1000/absolute temperature) vs In
k (degradation rate constant) and a linear relationship was obtained (Figure 23).
The plot was extrapolated and the expected half life of 3a at pH 4 and 25C was
calculated to be 515 days.


86
pH
Figure 21: pH profile of compound 3a in buffers at 63C. k is the degradation
rate constant (per day).


Alkaline pH
och2ch3
OH
3g
Figure 22: Pathways of degradation of 3a in buffers.


Full Text
DESIGN AND EVALUATION OF SOFT ANTICHOLINERGICS
BASED ON METHSCOPOLAMINE
By
GONDI N. KUMAR
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
1992

To my parents, Kondaiah and Sakuntala

ACKNOWLEDGEMENTS
I thank my principal advisor Dr. R.H. Hammer for his excellent guidance
and care throughout this project. I would like to thank my coadvisor Dr. N.S.
Bodor for his example, thoughtful input and constant encouragement. I would
like to thank the member of my supervisory committee, Drs. J. Simpkins, M.O.
James and E. Enholm, for their interest in the project. I would like to thank the
Department of Medicinal Chemistry for the continuous financial support in the
form of an assistantship.
I would like to acknowledge the help I received from Drs. Z. Gunes, L.
Prokai and J. Sastry. The help I received from Dr. W. Wu and L. Prokai needs
special mention. I would to thank the other members at the Center: Laurie
Johnston, Joan Martignago, Julie Berger, Kathy Eberst and Robert Wong.
I would like to thank all friends and colleagues at the Center who made
this tenure pleasant and enjoyable. I cherish the support and encouragement
received from my siblings, Sujatha, Sunitha and Suresh, and my grandparents.
Most importantly, I would like to thank my wife Uma, for her
understanding and love, and for coping with me while contending with her own
dissertation.

TABLE OF CONTENTS
Page
Acknowledgements iii
List of Table vi
List of Figures vii
Key to Symbols x
Abstract xii
Chapters
1 Introduction 1
Drug Design 1
Soft Drugs 3
Autonomic Nervous System 8
Structure and Drug Deliver to the Eye 11
Skin Structure and Drug Delivery 17
In vitro Penetration Studies 20
Sweat Glands - Structure and Pharmacology 24
2 Present Study and its Significance 27
Scopolamine and Methscopolamine 27
Design and Rationale 31
Objectives 34
3 Materials and Methods 36
Materials 36
Methods 37
4 Results and Discussion 53
Syntheses 53
Anticholinergic Activity 63
In Vitro Stability 83
In Vitro Penetration Studies 98
IV

5
Conclusions
112
References 116
Biographical sketch 123
v

LIST OF TABLES
Table Page
1 In vitro anticholinergic activity - guinea pig ileum assay 66
2 In vitro stability of soft drugs 3 (a-f) in biological media 91
3 In vitro stability of soft drugs 6 (a-e) in biological media .... 92
4 The n-Octanol/water partition coefficients and the hairless
mice skin permeability coefficients of 1 and soft drugs 102
VI

LIST OF FIGURES
Figure Page
1 A schematic representation of the metabolic fate of a
conventional drug (D) in vivo 6
2 A schematic representation of the metabolic fate of a
soft drug (SD) in vivo 7
3 A schematic cross section of the eye 12
4 A schematic cross section of the skin showing the
structural elements and three potential routes of
penetration of a diffusant 18
5 A schematic representation of the amount of material
penetrating the skin as a function of time in an in vitro
diffusion experiment 22
6 Structures of methscopolamine, hypothetical metabolites
and the soft drugs of this study 33
7 Synthesis of phenylmalonic acid analogs of
methscopolamine (3 a-f) 39
8 Synthesis of phenylsuccinic analogs of
methscopolamine (6 a-e) 41
9 Representative dose response curves from guinea pig
ileum assay 64
10 A representative Schild plot 65
11 Mydriatic dose response curves in treated rabbit eye 70
12 Mydriatic dose response curves in rabbit treated eye
after unilateral instillation (1, 3 a-e) 71
vii

13 Mydriatic dose response curves in rabbit treated eye
after unilateral instillation (1, 6 a-c) 72
14 AUC24hrs for the treated eye after unilateral instillation
into the rabbit eye 73
15 Duration of mydriatic activity after unilateral instillation
into the rabbit eye 74
16 Time vs mydriatic response curves in the treated eye
after unilateral administration of equieffective doses 75
17 Time vs mydriatic response curves in the untreated eye
after unilateral administration of equieffective doses 77
18 AUC6hrs for the untreated eye after unilateral instillation
of equieffective doses 78
19 Mydriatic activity after intravenous administration of
equipotent doses to rabbits 81
20 Muscarinolytic activity against acetylcholine induced
bradycardia in rats after intravenous administration of
equipotent doses 84
21 pH profile of compound 3a in buffers at 63°C 86
22 Pathways of degradation of 3a in buffers 87
23 Arrhenius plot for compound 3a in pH 4 buffer 88
24 Hydrolytic rates of 3a in rabbit tissues 93
25 Hydrolytic rates of 3a in pure enzyme preparations 94
26 Mydriatic activity in treated rabbit eye after unilateral
administration of equieffective doses of 3a and 3g 97
27 A representative trace of the in vitro penetration of 3c
across hairless mice skin 101
28 Relationship between log Partition Coefficient and
log Permeability Coefficient 103
VIII

29 Network of alternate pathways for percutaneous
absorption 106
30 The time course of penetration of 1, 3a and 6a across
rabbit cornea in vitro 108
31 In vitro permeability coefficients across rabbit cornea 109
IX

KEY TO SYMBOLS
A0
AT
AUC
°C
CDCI3
cm
DMSO-d6
gm
hr
HPLC
IR
iv
k
kg
log P
log Kp
lbs
M +
angstrom
absolute temperature
area under the curve
degrees centigrade
deuterated chloroform
centimeter
deuterated dimethyl sulfoxide
gram
hour
high pressure liquid chromatography
infrared
intravenous
degradation rate constant
kilogram
log partition coefficient
log permeability coefficient
pounds
molecular ion
x

mAChR
mg
min
ml
M
mM
mp
n
NMR
P
QSAR
r2
SEM
*1/2
TLC
vs
v/v
w/v
s
A*
tA
muscarinic acetylcholine receptor
milligram
minutes
milliliter
moles
millimoles
melting point
number of samples or readings
nuclear magnetic resonance
probability
quantitative structure activity relationship
regression coefficient
standard error of mean
half life
thin layer chromatography
versus
volume/volume
weight/volume
parts per million
wave length (micrometer)
microgram
microliter
XI

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
DESIGN AND EVALUATION OF SOFT ANTICHOLINERGICS
BASED ON METHSCOPOLAMINE
By
Gondi N. Kumar
August, 1992
Chairperson: Richard H. Hammer
Cochairperson: Nicholas S. Bodor
Major Department: Medicinal Chemistry
The present study involves the development of a new type of
anticholinergics, called soft anticholinergics, which are based on
methscopolamine. They are designed to act at the site of application but not to
act systemically due to their rapid metabolic inactivation in the systemic
circulation. Two hypothetical metabolites of methscopolamine were chosen as
the lead compounds and they were inactivated by esterification with cyclic and
alicyclic alcohols to yield two series of compounds.
The compounds synthesized were found to be potent anticholinergics in
both in vitro and in vivo assays. The activity was found to decrease with
increasing side chain length. The soft drugs with short side chain (2 or 3
carbons) were found to be very short acting mydriatics in the rabbit eyes after
xii

unilateral topical administration of equieffective doses compared to a very long
duration of activity with methscopolamine. The untreated eye was found to
dilate in methscopolamine treated animals but not in soft drug treated animals,
which indicates a lack of systemic activity of the topically administered soft
drugs. One soft drug (phenylmalonic ethyl analog) exhibited a very short
duration of systemic anticholinergic activity compared to methscopolamine on
intravenous administration in rabbits and rats.
The in vitro hydrolytic rates of soft drugs in biological media were
significantly higher than those of methscopolamine. The soft drugs of
phenylmalonic series degraded, in vitro, to yield an active metabolite which is
the decarboxylated product of the lead hypothetical metabolite. The soft drugs
of the phenylsuccinic series degraded to yield the lead carboxylate metabolite
which is essentially inactive.
The transdermal permeabilities of the soft drugs were found to be directly
dependent on the octanol/water partition coefficients. The transcorneal
permeability of one of the soft drugs tested was found to be higher than
methscopolamine.
The short duration of mydriatic action coupled with a lack of systemic
activity on topical administration would probably increase the therapeutic index
and make these compounds promising candidates for further studies.
xiii

CHAPTER 1
INTRODUCTION
Drug Design
The biological and pharmacological properties of a compound are
determined by its chemical structure. Structural variations should be able to
change the biological properties as well. Numerous skillful variations have been
made by medicinal chemists to yield therapeutically useful drugs. The most
rational route for the design of better drugs would be direct design based on
the biomolecular processes underlying diseased states. Since the biomolecular
processes underlying many diseased states is still not well understood, such an
approach is not always feasible. In fact, many of the presently important types
of drugs have been discovered accidentally (Austel, 1989). Minimization of
chance discovery can be achieved by systematic empirical procedures which
incorporate methods that allow detection and objective expression of
relationships between chemical structures and their biological properties.
Lead Compound
The general scheme for drug discovery starts with identification of a lead
compound or structure. A lead compound is one which has some if not all the
1

2
desirable properties expected of a better drug. The lead compound is then
subjected to structural variations to optimize its biological properties in the
desired direction. A new drug must be able to improve current therapy. The
new drug, therefore, has to fulfill many more and stricter requirements than the
lead compound. Two aspects of this strategy is lead finding and lead
optimization (Testa, 1984).
Lead Finding
Lead finding includes the identification of novel compounds displaying an
interesting biological activity and compounds with a newly discovered biological
activity.
Mass (random) screening, rational approach (based on correct or
incorrect hypothesis), natural products, drug metabolites, unexpected side
effects and the most important of all, serendipity are the methods used to
accomplish lead finding.
Lead Optimization
Lead optimization is the starting point for systematic modification aiming
at increasing the desired activities and decreasing the undesired activities.
Quantitative methods based on biotransformation considerations and
quantitative structure activity relationships (QSAR) are the methods used for
lead optimization. QSAR method advanced by Hansch (1981), which
mathematically correlates physicochemical properties like partition coefficient,
lipophilicity etc. to biological activity, is used extensively in lead optimization.

3
The other approach used for lead optimization takes the metabolic processes of
the body into consideration. Ariens and Simonis (1982) introduced the design
of poorly metabolizable drugs called "hard drugs." Hard drugs are
"metabolically stabilized" and display some advantages, for example: longer
half-lives, decreased intra- and interpatient variability, decreased species
difference and decreased number and significance of active metabolites.
In contrast, the other group "prodrugs" and "soft drugs" are usually
rapidly metabolized due to the incorporation of labile groups into the molecules.
Prodrugs are compounds with little or no activity, but which undergo metabolic
transformation in vivo to yield the active drug (Notari, 1981). Prodrugs
essentially alter the pharmacokinetic rather than the pharmacodynamic profile of
a drug.
Soft Drugs
The therapeutic index (Tl) is the most important characteristic of a drug.
The toxicity of a drug is the result of summation of many effects of not only of
the drug itself but also its metabolites, reactive intermediates, and various
compounds resulting from direct or indirect interaction with cell components. It
is thus apparent that there is a need to include metabolism considerations in
the general drug design process.
Bodor (1981) has suggested an approach for the design of safe and
better drugs which aims to control and direct the metabolism by drug design.

4
This is the concept of "soft drugs," which are defined as "biologically active,
therapeutically useful chemical compounds (drugs) characterized by a
predictable and controllable in vivo destruction (metabolism) to non-toxic
moieties, after they achieve their therapeutic role" (Bodor, 1984, p. 262).
The main objective in the soft drug design is deliberate simplification of
the metabolism. Ideally, a soft drug is metabolized in a single step to an
inactive metabolite. This metabolite may be further metabolized or conjugated
before elimination. This process avoids formation of active metabolites or
reactive intermediates. Since the main objective is an optimized therapeutic
index, the new soft drug does not necessarily have to be the most potent agent
of the series.
There are at least five ways to attempt rational soft drug design (Bodor,
1984).
1. Soft analogs: Bioisosteric/isoelectronic replacement in the known bioactive
compounds to yield structural analogs which are better metabolized.
2. Activated soft compounds: A non-toxic inactive compound is activated by
the incorporation of a group which will provide the desired pharmacological
effect.
3. Active metabolites: A derivative of an active metabolite which is in highest
oxidation state and which then will undergo one-step deactivation.
4. Natural soft drugs: Pro-soft drug approach to deliver endogenous
substances such as steroids and autocoids.

5
5. Inactive metabolite approach: The inactive metabolite approach starts with
the identification of an inactive metabolite of a drug. The inactive metabolite is
then chemically modified to produce a compound with structural resemblance
(isosteric and/or isoelectronic) to the original active drug. This newly
synthesized compound should display sufficient pharmacological activity and
degrade in vivo to yield the original inactive metabolite without the generation of
any toxic intermediates.
This design has been applied successfully to such diverse agents like
DDT (Bodor, 1984), (3-blockers (Bodor et al. 1988), steroids (Druzgala et al.
1991) and anticholinergics (Bodor et al. 1980; Hammer et al. 1988). The
metabolic patterns of a conventional drug (D) and a soft drug (SD) are depicted
in Figures 1 and 2, respectively. A conventional drug (D) has the potential to
generate metabolites (active and/or inactive) and reactive intermediates. In
comparison a soft drug (SD) is metabolized in a single step to the inactive
metabolite, which is then either eliminated directly or metabolized further
(conjugation) before elimination. Thus, the soft drug approach enables the
separation of therapeutic properties from deleterious side effects.
Short Duration of Action
A short duration of action is important if a drug results in undesirable
side effects. If the plasma concentration is high over a long period of time, a
transaxonal transport of drugs may occur across the blood brain barrier
producing central side effects also (Mager et al. 1972). A short duration of

6
D
Figure 1: A schematic representation of the metabolic fate of a conventional drug
(D) in vivo.

7
Delivery
Process
Direct
O
Soft
Drug
SD
Active
(Inactive)
M ,,.M
1
Inactive
n
Elimination
Figure 2: A schematic representation of the metabolic fate of a soft drug (SD) in vivo.

8
action is necessary for diagnostic agents as well. It can be achieved by the
introduction of metabolically labile groups into the molecule (COOR,
OCO(CH2)nR, CH2S03Na etc) (Mager, 1984). If metabolically unstable groups
are placed in an accessible position in the molecule, the resulting entity may be
more rapidly metabolized, conjugated, and rapidly eliminated after fulfilling its
intended pharmacological activity.
Autonomic Nervous System
Acetylcholine acts as the neurotransmitter in all postganglionic
parasympathetic nerves as well as in all preganglionic fibers of peripheral
autonomic ganglia, motor nerves to skeletal muscles, and certain synapses
within the central nervous system. Binding of acetylcholine to the cholinergic
receptors initiates a series of biochemical and/or electrophysiological events
that eventually lead to the final cellular response.
The central and peripheral actions of acetylcholine are exerted at two
main types of receptors. In 1914, Sir Henry Dale classified cholinergic
receptors into two types: nicotinic and muscarinic, to describe the different
physiological effects of acetylcholine that were mimicked by the alkaloids
nicotine and muscarine. Nicotinic and muscarinic receptors differ not only in
their affinities for agonists and antagonists, but also in many other fundamental
respects such as location, function, molecular architecture and receptor-effector
coupling mechanisms (Wess et al. 1990). The muscarinic acetylcholine

9
receptors (mAChR) is an integral membrane protein which is accessible to
muscarinic ligands from the extracellular space. Studies with various probes
has showed the presence in the ligand-binding site of at least one disulfide
bridge, several thiol groups, some strong nucleophilic groups and one or more
tyrosine residues (Sokolovsky, 1984). Studies on antagonist binding showed a
strong correlation between the affinity of the ligands for the mAChR and their
hydrophobicity, indicating that binding of the bulky antagonists involves
hydrophobic interaction with the receptor (Jarv & Bartfai, 1982).
Several pharmacologically distinct subtypes of mAChRs have been
identified, which may represent potential targets for new, therapeutically useful
drugs. Based on their affinity for pirenzepine, the mAChRs have been classified
into two groups, M, and M2. The M1 receptors (high affinity to pirenzepine) are
mainly present in forebrain (thought to be involved in higher brain functions
such as learning and memory as well as locomotor and behavioral effects) and
in peripheral autonomic ganglia such as intramural ganglia of stomach wall
(Wess et al. 1990). The M2 receptors (low affinity to pirenzepine) are located in
brain (implicated in the regulation of vegetative, sensory and motor functions)
and in peripheral effector organs such as heart, glands or smooth muscle
(Mitchelson, 1984).
Muscarinic Antagonists
Muscarinic antagonists competitively block the physiological actions of
acetylcholine mediated by muscarinic acetylcholine receptors. The use of drugs

10
that antagonize the action of acetylcholine long preceded the discovery of
acetylcholine itself. The prototypical antimuscarinic agents are the plant
alkaloids atropine and scopolamine found in several solanaceous species which
have been known to humans for millennia.
The structural elements of cholinergic antagonists (Wess, 1990) are
1. a cationic "head group" which is either a tertiary base which is protonated at
physiological pH or a quaternary ammonium moiety;
2. some "heavy blocking moieties," eg: alicyclic or aromatic rings, for
hydrophobic interaction with the receptor;
3. an interconnecting structural element (ester or amide) of definite length;
4. an "anchoring group," e.g. hydroxyl group(s) are often present at key
positions (Barlow and Ramtoola, 1980).
The major pharmacological effects produced by anticholinergics are:
reduction of lacrimal, salivary, bronchial and gastro-intestinal secretion;
suppression of sweating; increase in heart rate; depression of smooth muscle
activity in the bronchial, gastro-intestinal and genitourinary tract; mydriasis and
cycloplegia.
The variety of pharmacological responses of atropine/ scopolamine has
led to intense efforts to develop antimuscarinic agents which display selectivity
for certain organs or organ systems. But the success has been limited. The
clinical agents available today display essentially the same pharmacological
profile as atropine/scopolamine.

11
Stereochemistry of Anticholinergics
The most interesting aspect of biologically active compounds such as
hormones is their selective interaction with specific receptors. This interaction is
largely determined by their chemical complimentarity. Another important aspect
of this interaction is the spatial arrangement of atoms in the interacting ligand.
This is well exemplified in nature by the predominance of L-amino acids and D-
sugars. This stereospecific interaction is also true with xenobiotics. More than
75 years ago Cushny observed differences in activity between (-) hyoscyamine
and the corresponding racemate, atropine (Triggle & Triggle, 1976). High
eudismic ratios (eutomer/distomer; ratio of activity between the more active and
less active isomers) have been reported for hyoscyamine (Barlow et al. 1973)
and hyoscine (Triggle & Triggle, 1976). It has been shown that for the
appearance of enantioselectivity that an asymmetric center should be present in
a part of the molecule that actually contributes for binding (Ariens, 1966).
Structure and Drug Delivery to the Eve
A horizontal cross-section of an eye is depicted in Figure 3. The eye is
covered by three layers. The outer layer, sclera, is protective in function. It is
white in color and opaque with a transparent anterior portion (the cornea). The
middle layer is mainly vascular and is made up of the choroid, the ciliary body
and the iris. The innermost layer is the retina, a predominantly nervous tissue.
Within the three coats the eye is divided into two compartments by the lens.

Figure 3: A schematic cross section of the eye (Berman, 1991).

13
The frontal compartment contains aqueous humor, and is itself divided into
anterior and posterior chambers by the iris. The compartment behind the lens
contains the vitreous humor.
The important sites for drug action in the eye are the iris and ciliary
muscle, the blood vessels, the extraocular muscles and the lacrimal gland.
Autonomic Systems in Eve
The iris-ciliary body is composed of smooth muscles which are
innervated by the autonomic nervous system. The iris is a forward extension of
the choroid and arises from the anterior face of the ciliary body. The pupil
forms a central aperture in the iris. Two different types of muscle layers are
contained in the iris. These are dilator and sphincter muscles. The dilator
muscles receive sympathetic innervation. An increase in activity of these
muscles produces mydriasis. The sphincter muscles receive parasympathetic
innervation. An increase in activity of these muscles produces miosis. The
ciliary muscle is composed of smooth muscle units which receive
parasympathetic innervation. Stimulation of these muscles causes contraction,
which in turn results in a lessening of the tension on the suspensory ligaments
which causes lens to change shape, so that it becomes more convex and
allows near objects to be brought into focus at the retina.
Clinical uses of mydriatics include their diagnostic use for accurate
examination of retina and optic disc and in the treatment of acute iritis,
iridocyclitis and keratitis. Complete cycloplegia may be useful in certain clinical

14
conditions such as iridocyclitis and for precise measurement of refractive errors
(Weiner, 1980).
Clinically used anticholinergic agents for ocular use are atropine,
scopolamine, homatropine, cyclopentolate and tropicamide (Bartlett et al. 1989).
Atropine and scopolamine have a very long duration of action (exceeding 3
days). Homatropine and cyclopentolate have durations of about 1-3 days.
Tropicamide has the shortest duration of action lasting about 6 hours.
Tropicamide has been reported to be an unreliable cycloplegic (Gettes, 1961).
Systemic side effects of ocular drugs. Tropicamide syncope has been
reported as a complication of the drug (Schmidt, 1970). Scopolamine has been
reported to produce CNS toxicity (hallucinations and ataxia) (Freund & Merin,
1970). Homatropine has been reported to produce ataxia, hallucinations and
speech difficulty in children (Fraunfelder, 1989). Atropine has been reported to
produce at least 6 deaths involving ocular administration (Gray, 1979).
Cyclopentolate produced CNS side effects including grand mal seizures
(Kennerdell & Wucher, 1972).
Ocular Drug Delivery
Local administration (topical) has been the conventional form of drug
delivery to the eye. The topically applied ophthalmic drugs are in contact with
three absorptive layers: the cornea, the conjunctiva and the nasal mucosa
which results in ocular as well as systemic absorption. It has been shown that

15
about 10% or less of a topical dose is usually absorbed into the eye, the
remaining being systemically available, thereby eliciting unwanted side effects.
Some of the examples with narrow therapeutic indices include B-blockers (Berry
& van Buskirk, 1984) and anticholinergics (Gray, 1979).
Some approaches that have been tried to minimize the systemic
absorption of ocularly applied drugs are
1. manipulation of physical properties of the vehicle (viscosity etc) to increase
the residence time of the drug at the site of administration (Chang et al. 1988);
2. prodrugs to enhance ocular absorption with the consequent reduction in the
dose (Chang et al. 1987);
3. soft drugs, which act locally at the site of application and are metabolized
rapidly to non-toxic moieties in systemic circulation, e.g. soft steroids (Druzgala
et al. 1991);
4. design of chemical delivery systems based on the unique enzymatic systems
of the eye for selective activation in the target tissue (iris-ciliary body), e.g.
alprenolol ketoxime (Bodor & Prokai, 1990); and
5. choosing an oculoselective drug. A new noncardioselective but topically
oculoselective 8-blocker has been reported to be advantageous over the
traditional 8-blockers (Bauer et al. 1991).
Another strategy is to select a less potent analog and then offset the loss
in potency by enhancing its ocular absorption using prodrugs. Some success
has been reported with 8-blockers using this approach (Nathanson, 1985).

16
Ocular Drug Metabolism
Aryl hydroxylase, UDP-glucuronosyltransferase activity (Das & Shichi,
1981) and cytochrome P-450 activity (Shichi & Neber, 1980) have been
demonstrated in the eye. Enzymes for mercapturate synthesis have been
shown to occur in the ciliary body (Das & Shichi, 1981). Abundant presence of
nonspecific carboxylesterases in the iris-ciliary body has been reported (Lee,
1983).
Animal Models
Practical, ethical and economic considerations prevent testing of the
potential pharmacological agents in humans directly. Hence experiments on
living animals are important to study the compounds before they can be
clinically tested. The most frequently used animal model for ocular studies is
the rabbit. A few differences exist between the ocular physiology of rabbits and
humans (Maurice & Mishima, 1984).
1. Pattern of blinking: Rabbits blink 4-5 times per hour compared to much
higher number in humans.
2. Lower corneal permeability in humans than rabbits.
3. Humans have a larger lacrimal gland and different lacrimal drainage system
than rabbits.
Similarities include the similar structure of cornea and similar method of
changing the focal power of the eye. J.H. Prince (1964, p. x) states that "while

17
differences exist, numerous enough similarities justify its continued use of rabbit
in ocular work with due caution."
Corneal Permeability
Transcorneal delivery of pharmacological agents in the treatment of
ocular diseases is of immense importance. The cornea acts as a barrier to
xenobiotics entering the eye. The cornea is a trilaminate structure with
hydrophilic stroma sandwiched between two layers of hydrophobic epithelium.
Thus for hydrophilic compounds, the epithelium acts as a barrier whereas the
stroma acts as a barrier for hydrophobic compounds. Since the permeability of
a certain drug plays as significant a role as its potency, compounds like
epinephrine have been manipulated to increase their permeability by suitable
structural modifications (e.g. dipivalyl ester - dipivefrin) (Mandell et al. 1978).
Skin Structure and Drug Delivery
Skin is composed of two main tissue layers; the dermis and epidermis,
supported by a cushion of subcutaneous fat (Figure 4). Most simply, the skin
may be envisaged as providing a trilaminate membrane barrier to penetration;
(a) a thin, dead, highly resistant and lipophilic outer layer, known as the stratum
corneum, (b) the remainder of the living epidermis, which has the properties of
an aqueous gel, and (c) the dermis, which supports the vasculature and thus
provides a sink. For charged, hydrophilic molecules, the stratum corneum
provides the greatest barrier to skin penetration. This layer may be likened to a

18
ROUTES OF PENETRATION
Figure 4: A schematic cross section of the skin showing the structural elements
and three potential routes of penetration of a diffusant (Barry, 1983).

19
brick wall; the intercellular spaces between the keratinocytes (bricks) are filled
with a heterogeneous, apolar, lipid mixture (mortar). For small organic
molecules, the intercellular route rather than the alternative intracellular (or
shunt) route has been demonstrated to be the more dominant pathway of
penetration. Conversely, for lipophilic compounds, partitioning from the stratum
corneum into the aqueous, viable tissue, is the rate-limiting step.
Dermal and Transdermal Drug Delivery
The traditional forms used for cutaneous application are ointments,
creams, lotions and solution. Transdermal patches of scopolamine and
nitroglycerin (Dasta & Garaets, 1982) and topical solution of minoxidil (Maibach
& Wester, 1984) are some excellent examples of administering drugs by
transdermal and dermal routes, respectively. Bodor and Chikhale (1991) have
experimentally achieved dermal delivery of an acyclovir chemical delivery
system.
Cutaneous Metabolism
Recent studies have shown that certain compounds are indeed
metabolized in skin during the percutaneous absorption process (Kao et al.
1984). It seems likely that, although some compounds are metabolized
extensively during skin absorption, for many, metabolism will be small or
undetectable. However, a small amount of metabolism may be extremely
important, such as in the activation of potential carcinogens or other potent
compounds. Also, a knowledge of skin metabolism is essential for accurate

20
determination of the pharmacokinetics involved in the skin absorption process.
Esterase activity has been demonstrated in skin of humans and animals
(Tauber & Rost, 1986) along with other enzymes responsible for Phase I and
conjugation reactions (Tauber, 1982). The metabolic ability of skin has been
used to advantage in the design and activation of prodrugs (Higuchi, 1977).
Hadgraft (1980) and Bickers (1980) have reviewed this topic.
In Vitro Penetration Studies
In vitro absorption studies, when properly conducted and interpreted,
have proved very useful for the estimation of in vivo human percutaneous and
transcorneal absorption. Accurate determinations of absorption rates can be
made with in vitro diffusion cells, since frequent sampling can be done directly
beneath/across the membranes. For highly toxic compounds, in vitro studies
may be the only ethical way that human absorption data can be obtained.
For many compounds, the primary barriers to absorption are the non¬
living surface layer (stratum corneum) in the case of skin and the hydrophobic
epithelium of cornea (Ashton et al. 1991) in the case of eye. The rationale for
measuring absorption by in vitro techniques is that the absorption rates of
hydrophilic molecules are determined by passive diffusion through the stratum
corneum. In the case of eye, the hydrophilic molecules are known to diffuse
through the aqueous channels that exist in the corneal epithelium (Grass &
Robinson, 1988a).

21
Animal Models
Excised skin from a variety of animals, including rats and mice (normal
and hairless), rabbits, guinea pigs, hamsters, pigs, hairless dogs and primates
have been employed in a variety of types of diffusion cells in efforts to predict
percutaneous absorption in man (Barry, 1983). A popular model has been the
hairless mouse. Permeability of hairless mice skin has been shown to be
similar for some compounds (Stoughton, 1975). The in vitro absorption studies
across the rabbit cornea is the only reported technique so far reported.
Solutions to Fick’s Laws of Diffusion
In an in vitro diffusion experiment, the amount of substance building up in
the receptor phase is measured as a function time over an extended period.
Often, data are collected for a 48-hour period in the case of skin (6 hours in the
case of cornea) and the type of trace obtained is shown schematically in
Figure 5 (Hadgraft, 1989). During the middle section of the experiment the
amount accumulating approximates to a linear function of time; pseudo-steady
state has been established and a linear concentration gradient exists across the
skin. This portion of the overall curve is described by Fick’s first law of
diffusion:
J = DAK (C0 - Cr) / h
where J is the flux of substance across the skin, D is the effective diffusion
coefficient, K is the skin-vehicle partition coefficient, A is the cross-sectional area
of the skin of thickness h, and (C0 - Cr) is the concentration difference between

22
Figure 5: A schematic representation of the amount of material penetrating the
skin as a function of time in an in vitro diffusion experiment (Scott et al. 1989).

23
the donor and receptor phases, respectively. Often, the concentration in the
receptor phase can be considered insignificantly small compared with the donor
concentration (Cd) and the above equation is often simplified to
J = Kp A Cd
where Kp is the permeability coefficient and
Kp = D K / h
It is very difficult to assign precise values to K, D and h, and it is often more
appropriate to quote permeability coefficients generated form flux experiments
rather than to attempt to deconvolute them into their component parameters.
The linear portion of the graph can be extrapolated back to the time axis. This
yields the lag time which, from a solution to Fick’s second law of diffusion, can
be shown to be numerically equal to h2/6D.
Partition Coefficient and Permeability
Membranes like skin and cornea have both hydrophilic and hydrophobic
regions. The currently accepted theory proposes that lipid-soluble substances
pass through membranes because of their lipid content, whereas water-soluble
substances pass through cell membranes because of their hydrated
proteinaceous content. Apart from this, the chemical composition and the
physical characteristics of compounds play an important role in the penetration
process.
The partition coefficient is a measure of the ability of a chemical to
separate between two immiscible phases. This characteristic simulates the

24
partition of a compound in a vehicle and a membrane. The partition coefficient
between water and heptane (Bartek & LaBudde, 1975) or octanol (Roberts &
Anderson, 1977) of some compounds has been used to correlate their
percutaneous absorption. A good correlation has been reported to exist
between in vitro determined permeability and partition coefficients. In the case
of ocularly administered drugs precorneal factors like nasolacrimal drainage and
noncorneal absorption play a significant role in the overall bioavailability of the
drug at the site of action, i.e., the inner compartments of the eye. Hence an
optimal transcorneal absorption is needed to compete with the noncorneal
factors. Partition coefficient has been used to assess the penetration potential
of drugs across the cornea. A parabolic relationship (Schoenwald & Ward,
1979) has been reported for steroids between their log partition coefficient and
log permeability coefficient. Even though relative potency is a significant factor,
a rapid penetration rate can contribute significantly to effectiveness. It is
suggested that Hansch type correlations, linear or parabolic with modifications,
can be used to predict drug transfer across various ocular membranes such as
the cornea, blood-aqueous barrier and blood-vitreous barrier (Lien et al. 1982).
Sweat Glands - Structure and Pharmacology
Schiefferdecker introduced the terms "eccrine" and "apocrine" to describe
the two types of simple tubular exocrine glands of the skin (Greaves & Shuster,

25
1989). Eccrine glands discharge fluid without loss of cytoplasmic material in
contrast to the apocrine glands. Apocrine glands are associated with a hair
follicle/ sebaceous gland unit whereas eccrine glands have an independent
duct opening onto the surface. Eccrine glands are distributed over the entire
body surface while apocrine glands are concentrated in axillary and pubic
areas. Eccrine glands are richly innervated by unmyelinated sympathetic nerve
fibers which are cholinergic. Apocrine glands are innervated by adrenergic
nerves which respond to catecholamines. Very few animal species posses a
mechanism of heat dissipation quite as effective as that of man, who can
secrete as much as 1 liter/hour from eccrine sweat glands. The response of
human eccrine glands to acetylcholine and allied substances and the blocking
action of atropine has been demonstrated (Collins et al. 1959).
Antiperspirants
Hyperhidrosis is a cosmetic problem of great social importance.
Aldehydes (formaldehyde and gfutaraldehyde) have been shown to have
antisudorific effect (Sato & Dobson, 1969). While glutaraldehyde causes
discoloration of skin, formaldehyde causes sensitization. Zinc and aluminum
salts are used extensively as antiperspirants. The mechanism of action of these
salts is still unclear. An aluminum-containing plug of solid material has been
discovered in the duct of eccrine sweat glands following the application of
aluminum chlorhydrate (Hem & White, 1989). These salts may produce
peridural inflammatory changes which result in eccrine duct occlusion and a

26
degenerative change may occur in the apocrine tubular cell following their
repeated use (Kuno, 1956).
Anticholinergics as Antiperspirants
One of the antisecretory effects exerted by anticholinergics is their
inhibition of eccrine sweating. This aspect is usually the side effect of
systemically administered anticholinergics. A number of studies carried out with
anticholinergics show their effectiveness in inhibiting sweating. A detailed study
carried out (MacMillan et al. 1964) with 95 anticholinergics showed that some of
the most active antiperspirants are quaternary compounds. Scopolamine
methyl bromide was found to be more effective on human subjects than
scopolamine hydrobromide. In another study, topical application of a 1%
solution of scopolamine inhibited sweating for six days, while an 18% solution
has an effect lasting over 30 days (Stoughton, 1964).

CHAPTER 2
PRESENT STUDY AND ITS SIGNIFICANCE
Scopolamine and Methscopolamine
Scopolamine (Hyoscine) is an alkaloid found chiefly in shrubs,
Hyoscyamus niger and Scopolia carnolica. It is an antimuscarinic agent due to
its competitive antagonism of acetylcholine at muscarinic receptors. It is used
therapeutically in both tertiary form as hydrobromide salt and quaternary forms
as methobromide (1) (Figure 6), methonitrate and butylbromide (Extra
Pharmacopoeia, 1989).
Scopolamine has both central and peripheral anticholinergic actions. Its
peripheral effects include increase in heart rate at higher doses, decreased
production of saliva, sweat, bronchial, nasal and intestinal secretions, decreased
intestinal motility and inhibition of micturition. Its ocular effects include dilation
of pupil, paralysis of accommodation and photophobia. Scopolamine has found
wide application in therapy. In its tertiary form it is used as an adjunct to
anaesthetic medication, for prevention of motion sickness (usually as a
transdermal patch) and in its quaternary form as an antispasmodic for gastro¬
intestinal disorders and as a mydriatic/cycloplegic. Scopolamine ointment was
used experimentally to prevent gustatory sweating with success (Bailey &
27

28
Pearce, 1985). Topical application of 1% solution of scopolamine was found to
prevent thermal and pilocarpine induced sweating (Shelley & Horvath, 1951;
Brun & Hunziker, 1955). MacMillan et al (1964) have tested a series of O-
esters of scopolamine for antiperspirant activity and have reported high efficacy
in preventing thermally induced sweating.
Metabolism of Scopolamine
After transdermal administration of scopolamine to humans, 79%
appeared in urine as conjugates of glucuronic and/or sulfuric acid and 21% as
unchanged drug (Scheurlen et al. 1984). In the mouse the following
metabolites were found after administration of scopolamine hydrobromide:
scopolamine 9’-glucuronide, 6-hydroxyhyoscyamine, scopine, aposcopolamine,
nor-scopolamine, nor-scopolamine 9’-glucuronide and unchanged scopolamine
(Werner & Schmidt, 1968). In rats the following metabolites were found after
administration of methscopolamine bromide : methaposcopolamine, quaternary
scopine, p-hydroxy methscopolamine, p-methoxy methscopolamine,
methscopolamine glucuronide and two other unidentified metabolites (Sano &
Hakusui, 1974). In an in vitro stability study of scopolamine in different animal
tissues, human serum did not hydrolyze scopolamine whereas rabbit serum
exhibited the highest capacity of hydrolysis. The enzyme which hydrolyses
scopolamine has been shown to be different from cholinesterase (Otorii, 1969).
Incubation of scopolamine with guinea-pig liver microsomal preparations
resulted in the formation of the corresponding nor-alkaloids and N-oxide

29
(Phillipson et al. 1976). In case of atropine, an analogous tropane alkaloid
which differs from scopolamine by the absence of epoxy moiety, a toxic
aldehyde metabolite, atropanal, was detected in rabbits (Matsuda, 1966).
Side Effects of Scopolamine
Systemic administration of scopolamine in therapeutic doses normally
causes dry mouth, drowsiness, euphoria, amnesia and fatigue. Occasionally it
also causes excitement, restlessness, hallucinations or delirium (Extra
Pharmacopoeia, 1987). Transdermal administration of scopolamine as anti¬
motion-sickness agent produced dilation of pupil (Johnson & Moore, 1983).
Administration of scopolamine eye drops resulted in visual hallucinations,
strange behavior and restlessness (Hamborg et al. 1984; Birkhimer et al. 1984).
The side effects after ocular administration are due to the drainage of the
applied drug into the nasolacrimal duct and subsequent systemic distribution.
In view of the above cited systemic side effects after local administration
of scopolamine, it is advisable to design soft drugs based on scopolamine for
achieving a local action without systemic side effects.
Soft Anticholinergics
Bodor et al. (1980) have reported soft ester open chain analogs of
anticholinergics synthesized from cyclopentyl-phenylacetic acid, phenylacetic
acid and branched aliphatic carboxylic acids. In this class of soft
anticholinergics the ester oxygen and quaternary head are separated by only
one carbon whereas the classical anticholinergics contain a 2-3 carbon bridge.

30
The quaternary head is destroyed on hydrolysis with consequent loss of
potency with the formation of formaldehyde and other inactive moieties. The
soft anticholinergics so designed were found to be potent and hydrolytically
unstable. This concept was also applied for the design of soft anticholinergics
based on propantheline (Brouillette, 1987). The ethylene bridge of
propantheline was shortened by one carbon to produce a new series of
compounds with retention of anticholinergic activity and decreased hydrolytic
stability.
Hammer et al. (1988) have applied the inactive metabolite approach of
designing soft anticholinergics of atropine. They have chosen a hypothetical
metabolite of atropine, an oxidation product of the primary hydroxyl group as
the lead compound. This lead compound was reactivated by esterification with
aliphatic and cycloaliphatic alcohols of varying chain length. In both in vitro
(guinea pig ileum assay) and in vivo (mydriatic activity in rabbit eye) tests, these
compounds were shown to be active and comparable to atropine. Compared
to atropine, these compounds are reported to be less stable in biological media
yielding the inactive metabolite. The soft anticholinergics have been shown to
have short duration of mydriatic activity upon topical administration into rabbit
eye (Hammer et al. 1991) and an ultra-short duration of muscarinolytic activity
on systemic administration in rats (Bodor et al. 1990) compared to atropine.
Soft anticholinergics so far synthesized essentially have two
characteristics:

31
1. retention of anticholinergic activity and
2. increased rates of hydrolysis in biological media.
Tropicamide and tropane alkaloids, viz: atropine and scopolamine are
used in ophthalmic procedures as mydriatic/cycloplegic agents for the
examination of the interior of the eye. With tropane alkaloids, the effect is
usually accompanied by systemic side effects due to drainage of the ocularly
administered drug into the systemic circulation through the nasolacrimal canal.
The mydriatic effect of tropane alkaloids usually lasts for more than two days.
The duration of action with tropicamide is about six hours (Weiner, 1980).
Anticholinergics have been tested as antiperspirants experimentally and
have been found to be active. But prolonged use of traditional anticholinergics
as antiperspirants may not be feasible due to the systemic absorption and
consequent accumulation in the systemic circulation resulting in unwanted side
effects.
Design And Rationale
In view of the above cited undesirable systemic side effects due to the
locally administered drug and extremely long duration of mydriatic action, the
design of soft drugs based on scopolamine is desirable. It is also desirable to
develop a topical antisecretory agent with a selective local action on the eccrine
sweat glands with loss of potency on entering the systemic circulation. The
inactive metabolite approach for the design of soft drugs advanced by Bodor

32
(1984) and successfully adopted for the design of soft drugs of atropine by
Hammer et al. (1988) is applicable to scopolamine/ methscopolamine as well.
A hypothetical carboxylate metabolite of methscopolamine (2) is chosen as the
lead metabolite for the design of the soft drugs. Even though this metabolite
has not been detected in either animals or humans, it is a logical choice as a
lead compound since it is the oxidation product of primary alcoholic group of
methscopolamine. Another hypothetical metabolite (2a) chosen as the lead
compound is designed with a methylene group in the side chain. This
extension of side chain will expose the ester moiety even further and possibly
lessen the steric hinderance making it more amenable for enzymatic hydrolysis.
The quaternary form of scopolamine is chosen as the lead compound instead
of the tertiary form to minimize the central side effects on systemic absorption
(Malatray & Simon, 1972). The hypothetical metabolites are expected to be
highly polar (pKa: ca. phenylmalonate = 2.65 and phenylsuccinate = 4.0) and
ionized at physiological pH, and thus be subject to facile elimination from the
systemic circulation either directly or after conjugation. The hypothetical
metabolites are activated by esterification with suitable aliphatic and
cycloaliphatic alcohols to yield soft drugs (3 or 6) (Figure 6) which will degrade
in vivo, in a single step, to the more polar metabolite (2 or 2a). Strong
nucleophilic groups are shown to be present at the muscarinic receptor site
(Sokolovsky, 1984). There is a possibility that the carboxylate metabolite (2 or
2a) that results on the hydrolysis of the soft drugs will have an unfavorable

33
1
2 (Hypothetical metabolite)
3 (Soft drugs)
2a (Hypothetical metabolite)
Figure 6: Structures of methscopolamine, hypothetical metabolites and the soft
drugs of this study.

34
interaction with the receptor site, making it less active than the original soft
drug. The hydrolysis is expected to be facile in the possible target tissues for
the soft drugs due to the abundant presence of non-specific esterases in ocular
tissues (Lee, 1983) and skin (Tauber, 1982). Thus a shorter local action with
potentially reduced systemic side effects can be visualized with these soft
drugs.
Objectives
The aims of this study are the development and evaluation of soft
anticholinergics based on methscopolamine. These putative soft
anticholinergics are expected to be active locally at the site of application but
are hydrolyzed in a facile manner in systemic circulation to an inactive polar
metabolite. The net result is expected to be an increase in the therapeutic
index with localization of response and avoidance of systemic side effects. A
decrease in the duration of mydriatic activity is expected due to the
incorporation of a labile ester moiety into the molecule. The experimental
protocol is as follows:
1. Syntheses of phenylmalonic and phenylsuccinic acid analogs, and their
respective metabolites.
2. Development of a suitable analytical system to evaluate the stability and
the metabolic pathways of the soft drugs synthesized.

35
3. Evaluation of physical stability and degradation pathways of a selected
soft drug in buffers.
4. Evaluation of in vitro stability in various biological media.
5. Evaluation of in vitro pharmacological activity by guinea-pig ileum assay.
6. Evaluation of in vivo activity in suitable animal models.
7. Studies on penetration characteristics of soft drugs through rabbit cornea
and hairless-mice skin, and correlating the permeabilities with the
partition coefficients.

CHAPTER 3
MATERIALS AND METHODS
Materials
All chemicals used were reagent grade. Scopine hydrochloride,
acetylcholinesterase, pseudocholinesterase, porcine liver esterase, carbachoi
and hexamethonium bromide were obtained from Sigma Chemical Company.
Other chemicals were obtained from Aldrich Chemical Company and solvents
from Fisher Scientific. All melting points were recorded using Fisher-Johns
melting point apparatus and are uncorrected. NMR data were recorded with
Varían T-90 NMR spectrometer and are reported in parts per million (5) relative
to tetramethylsilane. All quaternary compounds were dissolved in DMSO-d6 and
other compounds were dissolved in CDCI3. Infrared spectra were recorded with
Perkin Elmer 1420 Ratio Recording Infrared Spectrophotometer. The elemental
analyses were carried out at Atlantic Microlab. Inc., Atlanta, Ga. FAB mass
spectrometry (Kratos MFC 500) of the quaternary compounds and metabolites
for characterization and identification in in vitro stability studies was performed.
Thin layer chromatography was carried out using EM Science DC-plastic foil
plates coated to a thickness of 0.2 mm with silica gel 60 containing Florescent
(254) indicator. The mobile phase consisted of toluene:methanol or
36

37
hexanes:acetone in various proportions. Column chromatography was
performed with silica gel (70-230 mesh) with appropriate mobile phases. All the
animal studies were conducted in accordance with the guidelines set forth in the
Declaration of Helsinki and The Guiding Principles in the Care and Use of
Animals (DHEW Publication, NIH 80-23). The following strains of animals were
used in the studies:
1. Male New Zealand albino rabbits weighing 3 kg (Kel Farm, Alachua, Florida),
2. Male Sprague Dawley rats weighing 300 gms,
3. Male Hartley guinea-pigs weighing 400 gms, and
4. 4-5 weeks old female hairless mice (NIHS-bg-nu-xid).
Rats, guinea pigs and mice were obtained from Harlan Sprague Dawley
Inc., Indianapolis.
Methods
Synthesis of Phenvlmalonic Analogs (3 a-f) (Figure 7)
Synthesis of monoalkvl 2-phenvlpropanedioic acids 4(a-f)
Phenylmalonic acid (9.00 gms, 0.05 moles) of phenylmalonic acid
dissolved in 100 ml of diethyl ether was treated with 5.95 gms (0.05 moles) of
thionyl chloride and 2 drops of dimethyiformamide. The mixture was refluxed at
40-50°C for 2 hours. The solvent was removed under vacuum. The acid
chloride, dissolved in 50 ml of dry benzene, was treated with 0.055 moles of the
relevant alcohol and the mixture was stirred overnight at room temperature.

38
The mixture was then washed with three 50 ml portions of water to remove any
traces of unreacted phenylmalonic acid. The monoester was then extracted
into 100 ml saturated solution of sodium bicarbonate. The bicarbonate solution
was washed with 50 ml of diethyl ether and neutralized with 10% of hydrochloric
acid. The oily monoester was extracted into 50 ml of diethyl ether, dried over
anhydrous sodium sulfate and crystallized from diethyl ether/petroleum ether.
The purity of the phenylmalonic acid monoesters was confirmed by thin layer
chromatography and their structures were confirmed by NMR spectroscopy.
Synthesis of (2R.2S1 1-alkvl. 3-r3a-f8-methvl-6B.73-epoxv-8-azabicvclo [3.2.1]
oct-3-vll 2-phenvlpropanedioic acid 5fa-fi
Scopine hydrochloride was dissolved in methanol and neutralized with an
equimolar quantity of methanolic potassium hydroxide. The precipitated
potassium chloride was filtered out, the filtrate was evaporated and the oily
residue was redissolved in chloroform. The filtrate was evaporated and dried
under vacuum. Scopine base was recrystallized from ether/pet.ether to yield
needle shaped crystals.
Phenylmalonic acid monoester (0.01 moles) dissolved in 25 ml of diethyl
ether was treated with 11 millimoles of thionyl chloride and two drops of
dimethylformamide. The mixture was refluxed for 2 hrs at 40-50°C. Solvent and
excess thionyl chloride were evaporated under vacuum. The oily acid chloride
was dissolved in dry benzene and to it was added, dropwise, a solution of 0.01
moles of scopine in benzene. After overnight stirring at room temperature, the
reaction mixture was washed with three quantities of 50 ml each of saturated

39
3 (a-f) 5 (a-f)
a : ethyl
b : n-propyl
c : n-butyl
d : n-pentyl
e : c-hexyl
f : ethyl-c-hexyl
Figure 7: Synthesis of phenylmalonic acid analogs of methscopolamine (3 a-f).

40
sodium bicarbonate solution and separated. The scopine ester was then
extracted into 50 ml of 10% hydrochloric acid. The acidic solution of scopine
ester was washed with 50 ml of diethyl ether and neutralized with sodium
bicarbonate. The oily scopine ester was extracted into 50 ml of chloroform,
which was dried over anhydrous sodium sulfate and evaporated to yield a
yellowish brown viscous liquid. The structure of the compound was confirmed
with NMR (Solvent: CDCI3) and purity with thin layer chromatography.
Quaternization
Quaternization of scopine ester (5 a-f) was done by reacting 0.005 moles
of scopine ester and 0.006 moles of dimethyl sulfate in diethyl ether. The
precipitated quaternary compound (3 a-f) was filtered and recrystallized from
methanol/ether. The compounds were identified by NMR/Mass spectra and
the purity was confirmed by elemental analysis.
Synthesis of metabolite (3g)
Essentially similar procedure adopted for 5 was followed for the synthesis
of metabolite 3g. Phenylacetic acid was coupled with scopine followed by
quaternization with dimethyl sulfate.
Synthesis of Phenvlsuccinic analogs (6a-e1 (Figure 8)
Synthesis of bromoacetate esters 7(b-e1
Ethyl bromoacetate was purchased from Aldrich Chemical Company.
Bromoacetic acid (8.34 gms, 0.06 moles) was dissolved in 75 ml of anhydrous

41
O
OH
ROH
Benzene
â–º
O
R
6a
ethyl
6b
n-propyl
6c
n-butyl
6d
c-hexyl
6e
benzyl
2a
H
2a <4-
Pd/C, Hydrogen
COOH
2 eq LDA
Phenylacetic
acid
1. Thionyl
chloride
2. Scopine
3. Dimethyl
sulfate
6 (a-e)
Figure 8: Synthesis of phenylsuccinic analogs of methscopolamine (6 a-e).

42
benzene and to it was added 0.05 moles of the relevant alcohol. The mixture
was refluxed using a Dean-Starke tube until no more water generated from the
reaction. The mixture was then washed with 100 ml of saturated sodium
bicarbonate solution to remove excess acid, treated with anhydrous sodium
sulfate and evaporated to yield a yellowish viscous liquid. The esters were then
purified by vacuum distillation at 10 mm of Hg.
Synthesis of monoalkvl 2-phenvlbutanedioic acids Sfa-el
Phenylacetic acid (1.63 gms, 0.012 moles) was dissolved in 50 ml of
anhydrous tetrahydrofuran and cooled to -70°C in acetone-dry ice bath under a
brisk stream of dry nitrogen. Lithium diisopropyl amide monotetrahydrofuran
(17 ml, 1.5 M solution in hexanes- Aldrich Chemical Company, 0.025 moles)
was injected through stopple slowly. The mixture was stirred for 1 hour to yield
a thick white suspension of the dianion. Then 0.012 moles of the bromoacetate
ester dissolved in 5 ml of tetrahydrofuran was injected at once through stopple
and the mixture was stirred for 1 hour to yield a reddish brown liquid which was
allowed to warm to room temperature. The reaction was then stopped by the
addition of 1 ml of water. The mixture was evaporated and the residue was
suspended in 50 ml of ethyl ether. The ether layer was then washed with 100
ml of dilute hydrochloric acid and the reaction products were extracted into 100
mi of saturated sodium bicarbonate solution. The bicarbonate layer was
separated, washed with 50 ml of diethyl ether, neutralized with dilute
hydrochloric acid and the precipitated products were extracted into 50 ml of

43
diethyl ether. The ether layer was treated with anhydrous sodium sulfate and
evaporated to yield an amorphous solid. If the TLC (hexanes:acetone = 2:1)
showed the presence of phenylacetic acid, the product was purified by silica gel
chromatography using hexanes-acetone. The product was crystallized in
ether/petroleum ether.
Synthesis of (2R.2S1 4-alkvl 1-[8-methvl-6B.7B-epoxv-8-azabicvclo [3.2.1] oct-3-
vl] 2-phenvlbutanedioic acid 9fa-e1
4-alkyl 2-phenylbutanedioic acid (0.05 moles) was dissolved in 25 ml of
diethyl ether. To it was added 1 ml of thionyl chloride and the mixture was
refluxed at 50°C for 2 hours. The solvent was removed under vacuum. Three
portions of 10 ml each of dry benzene were added and removed under reduced
pressure. The reddish brown oily acid chloride was redissolved in 20 ml of dry
benzene and to it was added dropwise a solution of 0.8 gms of scopine (0.005
moles) in benzene. The mixture was stirred overnight at room temperature.
Then it was washed with saturated sodium bicarbonate and the scopine ester
was extracted into 25 ml of dilute hydrochloric acid. The acid layer was washed
with ether, neutralized with sodium bicarbonate and the scopine ester was
extracted into 25 ml of chloroform. The solvent was removed under vacuum to
yield a viscous brownish liquid. TLC was performed with toluene:acetone (3:1).
Quaternization
Phenylsuccinic diester (0.002 moles) was dissolved in 20 ml of
chloroform and to it was added 0.5 gms of dimethyl sulfate. The mixture was
stirred at room temperature overnight. The white precipitate (6 a-e) was

44
collected on a filter, washed with diethyl ether and crystallized from
ethanol/diethyl ether.
Carboxvlate metabolite (2a)
Debenzylation of 6e was done by subjecting a mixture of 0.85 gms of 6e
and 1 gm of 5% palladium on carbon in 20 ml of acetonitrile to hydrogenolysis
at 30 Ibs/sq.in. in Parr apparatus for 1 hour. The mixture was filtered and the
filtrate was evaporated to yield a white powder, which was crystallized from
methanol/ether.
In Vitro Stability
Analytical method
A high pressure liquid chromatographic (HPLC) method was developed
to assay the soft drugs and metabolites in buffer media and biological fluids.
The system consisted of SP 8810 precision isocratic pump, SP 4290
autosampler with 20 loop, reverse phase Waters Nova-Pak CN 5mm X 10cm
radial-pak column with a guard column, SP 8450 UV/visible detector and
SP4290 integrator. The mobile phase consisted of 2.5 mM potassium
dihydrogen phosphate buffer containing 5 mM of 1-octane sulfonic acid sodium
and acetonitrile in different proportions. The flow rate was 1 ml/min. The
detection was made at 254 nm. The area under the peak was used as a
measure of the concentration. The concentration vs. area under the peak plot
showed linearity (r = 0.999) for the range of 0.2-1 ¡jq of the injected

45
compounds with a detection limit of 0.2 ¡jg (0.4 nm) of injected sample. The
retention times (minutes) for soft drugs and metabolites with 40% v/v
acetonitrile are as follows: 1 - 5.6, 3a - 7.5, 3b - 8.6, 3c - 9.7, 3d - 11.2, 3e -
10.2, 3f - 15.2, 3g - 6.5, 6a - 8.2, 6b - 8.5, 6c - 8.7, 6d - 9.4, 6e - 9.2 and 2a -
6.8.
In Vitro Stability in Buffers
USP standard buffer solutions (0.2 M) (USP XXI, 1985) in the pH range
of 1.5 to 9.5 were used in the study. Solutions of 3a (0.05 M) in buffers were
made. The solutions were kept in glass vials with screw caps with septa to
facilitate withdrawal of sample without opening the cap. The solutions were
stored at 25°C, 37°C, 46°C, 56° and 63°C. A 100 ¡j\ sample was withdrawn at
various time points and was analyzed by HPLC for 3a and degradation
products. Atropine sulfate monohydrate, 0.5% w/v in pH 3.8 USP buffer was
used as internal standard. Atropine solution was stored at 5°C and showed no
degradation during the period of study.
In vitro metabolism in biological media
The samples were analyzed by HPLC as described above. The samples
were either analyzed immediately or stored frozen until analysis.
Rat, rabbit and human plasma. The plasma was obtained by
centrifugation of freshly obtained heparinized blood. The stability studies were
carried out by adding aliquot of the drug stock solution to plasma to obtain a

46
final concentration of 0.25 mM. The plasma was kept at 37°C while shaking.
Samples of 0.1 ml were withdrawn at appropriate time intervals and mixed with
0.4 ml of ice cold acetonitrile to stop enzymatic reaction. The sample was then
centrifuged and the supernatant was analyzed by HPLC.
Rabbit liver homogenate. A suspension of freshly obtained rabbit liver
was made in isotonic pH 7.4 phosphate buffer. The protein content was
adjusted to 50 mg per ml. The liver suspension was equilibrated at 37°C. An
aliquot of soft drug stock solution was added to obtain a final concentration of
0.25 mM. Sampling, deproteination and analysis were done as described
above.
Pure enzymes. Drug solution (0.25 mM) in phosphate buffer of
appropriate pH was prepared and equilibrated at 37°C. An aliquot of pure
enzyme or reconstituted enzyme was added. Sampling, deproteination and
analysis were done as described above.
Ocular tissues. Rabbits were sacrificed by rapid i.v. injection of
pentobarbital (100 mg/kg) into the marginal ear vein. Eyes were enucleated
and rinsed in cold phosphate buffer (pH 7.4). The aqueous humor was
obtained by making a single puncture at limbus. Cornea was excised and iris-
ciliary body was isolated. Whole aqueous humor was used in the study without
further dilution. The homogenates of cornea and iris-ciliary body were made in
phosphate buffer (pH 7.4) using a Tekmer SDT tissuemizer for 2 minutes and
then centrifuged at 6000 rpm for 10 mins. The supernatant was used for the

47
study. Protein content was adjusted to 1.25 mg/ml. An aliquot of stock
solution of 3a was added to the homogenates to yield 0.1 mM solution. The
sampling and analysis was performed as described above.
Determination of protein content. Protein content was determined by
Peterson modification of the micro-Lowry method (Protein Assay Kit, Procedure
No. P 5656, Sigma Chemical Company) using bovine serum albumin as the
standard.
In Vitro Activity
Determination of anticholinergic activity
A guinea pig weighing about 400 gms was sacrificed by decapitation and
the terminal ileum was isolated. The ileum was cut into one inch pieces and
each was suspended in a 30 ml jacketed organ bath containing Tyrode’s
solution containing 0.1 mM hexamethonium bromide and constantly aerated
with 5:95 mixture of carbon dioxide and oxygen. One end of the ileum strip
was attached to a fixed support at the bottom of the organ bath and the other
end to a isometric force transducer (Model TRN001, Kent Scientific Corp.,
Connecticut) operated at 10 gms range. The ileum was kept at 0.5 gms
tension. Carbachol was used as an agonist and soft drugs and scopolamine
methyl bromide were used as antagonists. The ileum responded by contracting
upon the addition of carbachol. The contractions were recorded on a
physiograph (Desktop Model DMP 4B, Narco Biosystems Inc. Hounston).

48
Cumulative dose response curves of longitudinal contractions to the
addition of carbachol in the absence and presence of antagonists were
obtained following the method of Van Rossum et al. (1963). The pA2 value, an
empirical parameter which defines the negative logarithm of the molar
concentration of the antagonist which produces a two-fold shift to the right of a
concentration-response curve, was used as a measure of comparison between
the affinities of scopolamine and the soft drugs to the muscarinic receptor. In a
plot, the -Log (antagonist) vs the log (Dose ratio - 1), called the Schild plot, the
X-axis intercept is the pA2 value.
In Vivo Activity
Mydriatic activity
Twelve male New Zealand albino rabbits weighing 2.5-3 kg were used in
the study. The study was performed in a light and temperature controlled room
with minimum noise disturbance. The rabbits were divided into two groups of
six each. Drug solution in normal saline (50 /jl) was administered into one eye
of the one group of the animals. To the other group 50 /jL of normal saline was
administered into one eye. The pupil diameters of both the treated eye and the
untreated eye were measured with locally fabricated Haab’s scale (Alexandridis,
1985) before the administration of the drug and at appropriate time intervals
until the diameter reaches the time zero value. The difference between the
diameter at a particular time interval and the diameter at time zero is reported

49
as the mydriatic response. The net mydriatic response was obtained by
deducting the normal saline treated value from the corresponding drug treated
value at a particular time interval for the same animal. The experiment was
repeated after a period of 48 hours by switching the groups. The mydriatic
activity was compared using two parameters viz: recovery time (i.e., time taken
for the pupil to return to 0.75 mm above the time zero value) and 24 hour area
under time vs mydriatic response curve (AUC24hrs).
The mean and the standard error of the mean (SEM) were calculated for
the twelve animals tested. Student t-test was performed to determine the
statistical significance at 4 hours for the treated eye and 30 minutes for the
untreated eye. The AUC was calculated by approximate trapezoids and
triangles method for different time intervals.
Mydriatic effect in rabbits on intravenous administration
Groups of four male New Zealand albino rabbits weighing 3-3.5 kg were
used in the study. Pupil diameters of both the eyes were noted. 200 ¡j\ of
equipotent doses methscopolamine or soft drugs (3a or 6a) dissolved in normal
saline was then injected into the marginal ear vein. Control group received 200
fj\ of saline. Pupil diameter was recorded at various time intervals until it
reached the basal reading.
Effect on the resting heart rate of rats
Groups of four male Sprague-Dawley rats weighing 250-300 gms were
used. The animal was anesthetized with 50 mg/kg of pentobarbital given

50
intraperitoneally. The heart rate was recorded using a physiograph (Projector
Model Type PMP-4B, Narco Biosystems, Inc). After a 10 minute equilibration
period, a dose of drug dissolved in saline was injected as a bolus into the left
jugular vein. The control group received saline. The heart rate was recorded at
periodic intervals.
Effect on acetylcholine induced bradycardia
Groups of four male Sprague-Dawley rats weighing 250-300 gms were
used. The animal was anesthetized with 50 mg/kg of pentobarbital given
intraperitoneally. The heart rate was recorded using a physiograph (Projector
Model Type PMP-4B, Narco Biosystems, Inc.). After a 10 minute equilibration
period, an equipotent dose of methscopolamine or soft drug (3a) or metabolite
(3g) dissolved in saline was injected into the left jugular vein. Control group
received saline. At appropriate time intervals 50 /l/I of 0.1% w/v solution of
acetylcholine in saline was injected into right jugular vein and the heart rate was
recorded.
In Vitro Penetration Studies
Octanol/water partition coefficient
The solution of drug in water (5 ml, 1 mM) was vigorously shaken with 5
ml of n-octanol at 25°C in a sealed container for 24 hours (Grass and Robinson,
1984). The content of the drug in water and octanol was quantitated by HPLC.

51
Transdermal penetration
Female hairless-mice were used for the study. Two-compartment,
vertical diffusion cells (Kresco Engineering, Palo Alto) with a surface area of 7.1
cm2 were used in the study. The animals were sacrificed by cervical
dislocation. Whole skin was dissected carefully from the abdomen and back.
The underlying fat tissue was gently peeled off with a forceps. The skin was
immediately mounted in the cell with dermal side facing the receptor chamber.
The receptor chamber was then filled with 40 ml of isotonic phosphate buffered
saline of pH 7.1 containing 0.1% v/v of formaldehyde. Drug solution (1 ml, 0.1
M) in isotonic phosphate buffered saline was placed on the top of the skin. The
entire system was stirred continuously at 200 rpm and incubated at 32°C. At
appropriate time intervals, 500 /jI sample was withdrawn and the receptor was
replenished with an equal quantity of buffer. The sample withdrawn was diluted
with 500 /jI of acetonitrile and vortexed. The mixture was analyzed by HPLC for
drug and metabolites.
Transcorneal penetration
Male New Zealand albino rabbits weighing 2.5-3.5 kg were used in the
study. Transcorneal permeation system of Crown Glass Company (Somerville,
NJ) was used in this study. This design is a two equal sized side-by-side cell
assembly which keeps the mounted cornea wrinkle free while maintaining
corneal curvature. The animals were sacrificed by a rapid injection of
pentobarbital (100 mg/kg) into marginal ear vein. The corneas were isolated by

52
making a small incision in the sclera approximately 2 mm from limbus and then
cutting circumferentially. The isolated cornea were washed with bicarbonated
Ringer and immediately mounted on the receptor cell. The donor cell was then
placed properly in position and the two cells were clamped securely. First the
receptor cell was filled with 4 ml of bicarbonated Ringer followed by the donor
cell with 4 ml of 0.025 M solution of drug in normal saline. The system was
kept at 37°C by connecting to a circulating water bath and continuously stirred.
Oxygen:Carbon dioxide (95:5) was continuously bubbled into both receptor and
donor chambers. Samples (200 /jI) were withdrawn at appropriate time intervals
form receptor cell and the chamber was replenished by addition of 200 fj\ of
bicarbonated Ringer. The samples were analyzed by HPLC without further
treatment.
Statistical analysis
All the values reported are mean _+ SEM (Standard error of mean =
Standard deviation/v(n-1)). Student "t" test was performed to assess the
significance of the results.

CHAPTER 4
RESULTS AND DISCUSSION
Syntheses
The physical and spectral characteristics of the compounds synthesized
are enumerated below:
Monoalkvl 2-phenylpropanedioic acids (4 a-fl: The compounds were
identified by NMR spectroscopy and the purity was tested with TLC. HOOC-
CH(Phenyl)-COOR 4a ethyl: M.P. 80°C, Yield 69%, 4b n-Propyl: 65°C, 71%, 4c
n-Butyl: 66°C, 72%, 4d n-Pentyl: 75°C, 66%, 4e c-Hexyl: 80°C, 53% and 4f ethyl-
c-hexyl: 73°C, 60%. The melting points of 4a and 4e coincided with the values
reported In the literature (Hammer et al. 1988).
(2R.2S1 ( + 11-alkvl. 3-[3a-i8-methyl-6i3.7B-epoxv-8-azabicvclo [3.2.1] oct-3-vO] 2-
phenvlpropanedioic acids (5 a-fi
5a. (2R,2S) (±)1-ethyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylpropanedioic acid: Yield, 81%, brownish viscous liquid, 1H
NMR (CDCIg): 6 1.2 (t, 3H, CH3), 2.0 (m, 4H, bicyclic), 2.5 (s, 3H, N-CH3), 3.2
(m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s,
1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic).
53

54
5b. (2R,2S) (±)1-n-propyl, 3-[3a-(8-methyl-68,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 76.5%, viscous liquid, 1H NMR
(CDCI3): <5 0.9 (t, 3H, CHg), 1.4-2.2 (m, 6H, bicyclic and CH2), 2.5 (s, 3H, N-
CHg), 3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, O-
CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H,
aromatic).
5c. (2R,2S) (±)1-n-butyl, 3-[3a-(8-methyl-68,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 81%, viscous liquid, 1H NMR
(CDCI3): S 0.9 (t, 3H, CH3), 1.4-2.2 (m, 8H, bicyclic and CH2-CH2), 2.5 (s, 3H, n-
chg), 3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2),
5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic).
5d. (2R,2S) (±)1-n-pentyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 79%, viscous liquid, 1H NMR
(CDCI3): 5 0.9 (t, 3H, CH3), 1.2-2.2 (m, 10H, bicyclic and CH2-CH2-CH2), 2.5 (s,
3H, N-CHg), 3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H,
0-CH2), 5.0 (s, 1H,Ar-CH), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H,
aromatic).
5§. (2R,2S) (±)1-c-hexyl, 3-[3o:-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 65%, brownish viscous liquid, 1H
NMR (CDCI3): S 1.1-2.2 (m, 14H, bicyclic and cyclohexyl), 2.5 (s, 3H, N-CH^,
3.2 (m, 2H, CH-N-CH), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 1H, CO-O-

55
CH,cyclohexyl), 5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m,
5H, aromatic).
5f. (2R,2S) (±)1-ethyl-c-hexyl, 3-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid: 75%, viscous liquid, 1H NMR
(CDCI3): S 1.1-2.2 (m, 17H, bicyclic and cyclohexyl ethyl), 2.5 (s, 3H, N-CH3),
3.2 (m, 2H, CH-N-CH), 3.3 - 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7 (m, 2H,
CH-O-CH epoxide), 4.1 (q, 1H, CO-0-CH2,cyclohexylethyl), 4.9 (s, 1H, Ar-CH),
5.1 (m, 1H, CH-O-CO, scopyl) and 7.3 (m, 5H, aromatic).
(2R.2S1 (4-11-alkvl. 3-[3a-(8.8-dimethvl-6B.7B-epoxv-8-azabicvclo [3.2.1] oct-3-
vll] 2-Dhenvlpropanedioic acid methvl sulfates (3 a-f)
(2R,2S) (±)1-ethyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 82%. m.p.
104-106°C. 1H NMR (DMSO-d6): 5 1.2 (t, 3H, CH3), 2.0 (m, 4H, bicyclic), 3.2
(m, 2H, CH-N-CH), 3.3 - 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7 (m, 2H, CH-
O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m, 1H, CH-O-
CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 360. Infrared spectrum (KBr pellet):
5.9 ¿/m - C=0 stretching, 7.9 fjm - C-N stretching, 8 fjm - epoxy ring breathing,
and 12.2 /jm - "12 band’ of epoxide. The ultraviolet spectrum (in acetonitrile)
showed three maxima at 264 nm (e 45.4), 258 nm (e 85.8) and 252 nm (e
45.4). Elemental analysis: calculated/found; C, 51.79/52.01; H, 6.01/6.09; and
N, 2.86/2.92.
3b. (2R,2S) (±)1-n-propyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 81.5%.

56
m.p.116-118°C. 1H NMR (DMSO-d6): S 0.9 (t, 3H, CH3), 1.4-2.2 (m, 6H, bicyclic
and CH2), 3.2 (m, 2H, CH-N-CH), 3.3 - 3.5 (3s, 9H, 2 X N-CH3 and CH3S04),
3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m,
1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 374. Calculated /found;
C, 53.44/53.64; H, 6.68/6.57; and N, 2.83/2.92.
3c. (2R,2S) (±)1-n-butyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate: Yield 84%. m.p.100-
102°C. 1H NMR (DMSO-d6): 5 0.9 (t, 3H, CH3), 1.4-2.2 (m, 8H, bicyclic and
CH2-CH2), 3.2 (m, 2H, CH-N-CH), 3.3 - 3.5 (3s, 9H, 2 X N-CH3 and CH3S04),
3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H, Ar-CH-), 5.1 (m,
1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 388. Calculated/found;
C, 55.31/55.21; H, 6.61/6.69; and 2.81/2.76.
3d. (2R,2S) (±)1-n-pentyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 79%. m.p.106-
107°C. 1H NMR (DMSO-d6): 5 0.9 (t, 3H, CH3), 1.2-2.2 (m, 10H, bicyclic and
CH2-CH2-CH2), 3.2 (m, 2H, CH-N-CH), 3.3 -3.5 (3s, 9H, 2 X N-CH3 and
CH3S04), 3.7 (m, 2H, CH-O-CH epoxide), 4.2 (q, 2H, 0-CH2), 5.0 (s, 1H,Ar-CH),
5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M + , 402.
Calculated/found; C, 56.14/55.99; H, 6.82/6.89; and N, 2.73/2.64.
3e. (2R,2S) (±)1-c-hexyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylpropanedioic acid methyl sulfate : Yield 75%.
m.p.183°C. 1H NMR (DMSO-d6): S 1.1-2.2 (m, 14H, bicyclic and cyclohexyl),

57
3.2 (m, 2H, CH-N-CH), 3.3 - 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7 (m, 2H,
CH-O-CH epoxide), 4.2 (q, 1H, CO-O-CH,cyclohexyl), 5.0 (s, 1H, Ar-CH-), 5.1
(m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 414. Calculated/
found; C, 56.18/55.96; H, 6.74/6.67; and N, 2.62/2.60.
3f. (2R,2S) (±)1-ethyl-c-hexyl, 3-[3a-(8,8-dimethyl-6B,7B-epoxy-8-
azabicyclo [3,2,1] oct-3-yl)] 2-phenylpropanedioic methyl sulfate : Yield 85%.
m.p.145°C. 1H NMR (DMSO-d6): 6 1.1-2.2 (m, 17H, bicyclic and cyclohexyl
ethyl), 3.2 (m, 2H, CH-N-CH), 3.3 - 3.5 (3s, 9H, 2 X N-CH3 and CH3S04), 3.7
(m, 2H, CH-O-CH epoxide), 4.1 (q, 1H, CO-0-CH2,cyclohexylethyl), 4.9 (s, 1H,
Ar-CH), 5.1 (m, 1H, CH-0-CO,scopyl) and 7.3 (m, 5H, aromatic). M+, 442.
Elemental analysis: calculated/found; C, 58.58/58.49; H, 7.05/7.11; and N,
2.53/2.54.
3a. [3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo [3,2,1] oct-3-yl)]
benzeneacetic acid dimethyl sulfate: White crystalline powder, yield : 82%, M.P.
153°C, m+: 288. NMR data: (DMSO-d6) 1.2-2.2 (m, 4H) bicyclic, 2.4-2.7 (3s,
9H) N + (CH3)2 and CH3S04, 2.7-3.7 (m, 8H) N + (CH)2, 0(CH)2 (epoxy), CH2-COO
and Ar-CH2, 4.9 (t, 1H) bicyclic CH-OCO and 7.3 (m, 5H) aromatic. Elemental
analysis: calculated/found; C, 54.14/53.95; H, 6.27/6.31; and N, 3.51/3.46.
The syntheses of the phenylmalonic acid analogs (3 a-f) was
accomplished by sequential esterification of phenylmalonic acid by the relevant
alcohol followed by scopine. The diester obtained was then quaternized.
Dimethylformamide was used as a catalyst in the preparation of acid chlorides.

58
Dimethylformamide is known to form a highly activated complex on reacting
with thionyl chloride which collapses on reaction with an acid to yield acid
chloride (Zaugg et al. 1960). In the first esterification step, the yields were not
usually very high due to the formation of diester of phenylmalonic acid instead
of the monoester alone. In the second esterification step and quaternization
step the yields were high. The quaternary compounds are extremely
hygroscopic and were stored in vacuo.
Several attempts were made to synthesize the hypothetical carboxylate
metabolite (2).
1. esterification of phenylmalonic acid with scopine;
2. esterification of scopine with monobenzyl phenylmalonate and debenzylation
by hydrogenation (in methylene chloride or diethyl carbonate);
3. debenzylation of quaternized scopine ester of monobenzyl phenylmalonate
in dilute acetic acid or acetone;
4. esterification of mono-TBDMS phenylmalonate with scopine and
deprotection with FN(t-Bu)4; and
5. enzymatic hydrolysis of 3a with porcine liver esterase in buffers.
The end product was 3g in all the above reactions instead of 2.
Compound 2 seems to be unstable in reaction conditions. This is interesting
since there are no reports in the literature about decarboxylation of carbenicillin,
the only other derivative of phenylmalonic acid in therapy.

59
Alkvl bromoacetates (7 a-e): The structures were confirmed by NMR
spectroscopy. Br-CH2-COO-C2H5 was obtained from Aldrich Chemical
Company. Br-CH2-COO-n-propyl: Yield=84%, B.P. 69°C at 10 mm of Hg. Br-
CH2-COO-n-butyl: 90%, 56°C. Br-CH2-COO-cyclohexyl: 89%, 71 °C. Br-CH2-
COO-benzyl: 84%, 129°C.
4-alkvl 2-phenvlbutanedioic acids (8 a-e): The structures were confirmed
by NMR spectroscopy and the purity was tested with TLC. HOOC-CH(Phenyl)-
CH2-COO-R: ethyl: M.P. 97°C, Yield 65%. n-propyl: 58°C, 66%, n-butyl: 60°C,
68%, c-hexyl: 63°C, 60% and benzyl: 100°C, 62%. The melting point of 8a
coincided with the value reported in literature (Gunes et al 1992), but differs
from the value reported in another report, 87-89°C (Kofron and Wideman,
1971).
(2R.2S) f+H-alkvl 1-f3a-f8-methvl-6B.7B-eDoxv-8-azabicvclo f3.2.1] oct-3-vh] 2-
phenvlbutanedioic acids (9a-e1:
9a. (2R,2S) (+)4-ethyl 1-[3a-(8-methyl-68,78-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: Viscous dark brownish liquid, Yield, 90%,
NMR data: (CDCI3) 1.2 (t, 3H, CH3), 1.3-2.2 (m, 4H, bicyclic), 2.4 (s, 3H, N-
CH3), 2.8-3.5 (m, 6H, N(CH)2, 0(CH)2 (epoxy), CH2-COO), 4.1 (m, 3H, COO-
CH2), Ar-CH-COO, 5 (t, 1H, bicyclic CH-OCO), 7.3 (m, 5H, aromatic).
9b. (2R,2S) (±)4-n-propyl 1-[3a-(8-methyl-66,78-epoxy-8-azabicyclo
[3,2,1] oct-3-yl)] 2-phenylbutanedioic acid: viscous liquid, 85%, NMR data:
(CDCIg) 0.9 (t, 3H, CH3), 1.5 (m, 2H, -CH2-), 2.2 (m, 4H, bicyclic), 2.5 (s, 3H, N-

60
CH3), 2.8-3.5 (m, 6H, CH2-COO, N(CH)2, 0(CH)2 epoxy), 4.1 (m, 3H, COO-CH2)
and Ph-CH, 4.9 (t, 1H, bicyclic CH) and 7.3 (s, 5H, aromatic).
9c. (2R,2S) (±)4-n-butyl 1-[3a-(8-methyl-66,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: viscous liquid, 81%, NMR data: (CDCI3) 0.9
(t, 3H, CH3), 1.5 (m, 4H, CH2-CH2-), 2.2 (m, 4H, bicyclic), 2.5 (1s, 3H, N-CH3),
2.8-3.5 (m, 6H, CH2-COO, N(CH)2, 0(CH)2 epoxy), 4.1 (m, 3H, COO-CH2) and
Ph-CH, 4.9 (t, 1H, bicyclic CH) and 7.3 (s, 5H, aromatic).
9cj. (2R,2S) (+)4-c-hexyl 1-[3a-(8-methyl-6B,7B-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: viscous liquid, 75%, NMR data: (CDCI3)
1.2-1.7 (m, 10H, Cyclohexyl), 2.2 (m, 4H, bicyclic), 2.5 (s, 3H, N-CH3), 2.8-3.5
(m, 6H, CH2-COO, 0(CH)2 epoxy), 4.1 (q, 1H, Ph-CH), 4.8 (m, 1H, COO-CH,
cyclohexyl), 4.9 (t, 1H, bicyclic CH) and 7.3 (s, 5H, aromatic).
9e. (2R,2S) (±)4-benzyl 1-[3a-(8-methyl-66,78-epoxy-8-azabicyclo [3,2,1]
oct-3-yl)] 2-phenylbutanedioic acid: dark brownish viscous liquid, 88%, NMR
data: (CDCI3) 1.3-2.2 (m, 4H, bicyclic), 2.4 (s, 3H, N-CH3), 2.7-3.7 (m, 6H,
N + (CH)2, 0(CH)2 epoxy, CH2-COO), 4.1 (m, 1H, Ar-CH-COO), 4.9 (t, 1H,
bicyclic CH-OCO), 5.0 (s, 2H, Ar-CH2) and 7.3 (m, 5H, aromatic).
(2R.2S1 í + 14-alkvl 1-[3a-(8.8-dimethyl-6B.7B-epoxv-8-azabicvclo [3.2.1] oct-3-vH]
2-phenvlbutanedioic acid methvl sulfates (6 a-el:
6a. (2R,2S) (±)4-ethyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1] oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: White crystals, Yield =
93%, M.P. 118-120°C, m + , 374. NMR data: (DMSO-d6) 1.2 (t, 3H) CH3, 1.3-2.2
(m, 4H) bicyclic, 2.4-2.7 (3s, 9H) N + (CH3)2, CHgSO^, 2.7-3.7 (m, 6H) N + (CH)2,

61
0(CH)2 (epoxy), CH2-COO, 4.1 (m, 3H) COO-CH2, Ar-CH-COO, 5 (t, 1H)
bicyclic CH-OCO, 7.3 (m, 5H) aromatic. Elemental analysis: calculated/found;
C, 54.44/54.63; H, 6.39/6.49; and 2.89/2.96.
6£. (2R,2S) (±)4-n-propyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: Extremely hygroscopic
white powder, 94%, M.P. 114°C, m + , 388, NMR data: (DMSO-d6) 0.9 (t, 3H)
CH3, 1.5 (m, 2H) -CH2-, 2.2 (m, 4H) bicyclic, 2.5-2.7 (3s, 9H) N(CH3)2 and
CHgSO/, 2.8-3.5 (m, 6H) CH2-COO, 4.1 (m, 3H) COO-CH2 and Ph-CH, 4.9 (t,
1H) bicyclic CH and 7.3 (s, 5H) aromatic. Calculated/found; C, 57.38/57.37; H,
6.70/6.67; and N, 2.70/2.78.
,§c. (2R,2S) (±)4-n-butyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: Extremely
hygroscopic white powder, 87%, M.P. 120°C, m+: 402, NMR data: (DMSO-d6)
0.9 (t, 3H) CH3, 1.5 (m, 4H) CH2-CH2-, 2.2 (m, 4H) bicyclic, 2.5-2.7 (3s, 9H)
N(CH3)2 and CH3S04‘, 2.8-3.5 (m, 6H) CH2-COO, 4.1 (m, 3H) COO-CH2 and Ph-
CH, 4.9 (t, 1H) bicyclic CH and 7.3 (s, 5H) aromatic. Calculated/found; C,
54.24/54.41; H, 6.82/6.77; and N, 2.64/2.65.
6d. (2R,2S) (±)4-c-hexyl l-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1]oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: Extremely
hygroscopic white powder, 87%, M.P. 118°C, m+: 428, NMR data: (DMSO-d6)
1.2-1.7 (m, 10H) Cyclohexyl, 2.2 (m, 4H) bicyclic, 2.5-2.7 (3s, 9H) N(CH3)2 and
CH3S04', 2.8-3.5 (m, 6H) CH2-COO, 4.1 (q, 1H) Ph-CH, 4.8 (m, 1H) COO-CH

62
cyclohexyl, 4.9 (t, 1H) bicyclic CH and 7.3 (s, 5H) aromatic. Calculated/found;
C, 57.90/57.71; H, 6.86/6.88; and N, 2.60/2.59.
6§. (2R,2S) (±)4-benzyl 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo
[3,2,1] oct-3-yl)] 2-phenylbutanedioic acid methyl sulfate: White crystals, 96%,
M.P. 98°C, m+: 436. NMR data: (DMSO-d6) 1.3-2.2 (m, 4H) bicyclic, 2.4-2.7
(3s, 9H) N+(CH3)2 and CH3S04', 2.7-3.7 (m, 6H) N + (CH)2, 0(CH)2 (epoxy), CH2-
COO, 4.1 (m, 1H) Ar-CH-COO, 4.9 (t, 1H) bicyclic CH-OCO, 5.0 (s, 2H) Ar-CH2
and 7.3 (m, 5H) aromatic. Calculated/found; C, 57.44/57.27; H, 6.21/6.17; and
N, 2.48/2.45.
2a. (2R,2S) (±) 1-[3a-(8,8-dimethyl-6B,7B-epoxy-8-azabicyclo [3,2,1] oct-
3-yl)] 2-phenylbutanedioic acid methyl sulfate: White crystals, 80%, M.P. 80°C,
m + : 346. NMR data: (DMSO-dg) 1.3-2.2 (m, 4H) bicyclic, 2.4-2.7 (3s, 9H)
N + (CH3)2 and CH3S04‘, 2.7-3.7 (m, 6H) N + (CH)2, 0(CH)2 (epoxy), CH2-COO,
4.1 (m, 1H) Ar-CH-COO, 4.9 (t, 1H) bicyclic CH-OCO, 7.3 (m, 5H) aromatic and
10.5 (s, 1H) COOH. Calculated/found; C, 50.52/50.61; H, 6.11/6.15; and N,
2.95/3.01.
The esterification of bromoacetic acid with the relevant alcohols by
azeotropic distillation in benzene gave high yields of bromoacetate esters. The
next alkylation step of phenylacetic acid is extremely moisture sensitive.
Precautions were taken to exclude moisture from the reaction setup. Dry
nitrogen was bubbled continuously to maintain a positive pressure inside the

flask. The yields of the alkylated products were in the range of 60 - 70% and
are in agreement yields reported in the literature (Kofron and Wideman, 1972).
Coupling of scopine and subsequent quaternization with dimethyl sulfate
proceeded smoothly with cumulative yields ranging from 65 - 90%. Catalytic
hydrogenation of 6e gave the carboxylate metabolite 2a.
Anticholinergic Activity
In Vitro Activity
All the values are means of four determinations. The dose response
curves in the absence and presence of increasing concentrations of antagonist
were generated by van Rossum’s method (1963). This is the most widely used
method for determination of the affinity of agonists and antagonists to receptors
in isolated organs. In this experiment, hexamethonium was added to the
medium to prevent any interaction of carbachol with the nicotinic receptors. So
the response obtained was exclusively from the action of the compounds on the
muscarinic receptors.
Methscopolamine and all the soft drugs showed reversible competitive
antagonism with parallel dextral shifts of dose response curve with increasing
concentrations of antagonist without depression of the maximal response
(Figure 9). Washing off the antagonist completely restored the original
sensitivity to carbachol clearly indicating the competitive nature of the
antagonism. The pA2 values were calculated according to Schild’s method

64
% of maximal reaponaa
1.000E-07 1.OOOE-O0 1.000E-06
Agonist (moles)
— 0 -I- 0.12 0.4S -a- O.s 0.06 2.4 -A- 4.0
Antagonist oonoantratlon (mM)
Figure 9: Representative dose response curves from guinea pig ileum assay.
Agonist: Carbachol. Antagonist: 3c. Each curve represents the response
obtained to increasing concentrations of agonist in the presence of a certain
concentration of antagonist.

Log (Dose Ratio
65
5.00 5.50 6.00 6.50 7.00 7.50 8.00
-Log (Antagonist)
Figure 10: A representative Schild plot. Each value of dose ratio (DR) is the
mean of four determinations at a particular concentration of the antagonist (3c).
X axis intercept is pA2.

66
Compound
9^2
r2
Slope
1
9.50
0.992
-1.067
3a
7.40
0.915
-0.920
3b
7.85
0.935
-1.010
3c
7.40
0.974
-1.003
3d
7.00
0.982
-0.876
3e
5.85
0.967
-1.039
3f
5.90
0.996
-1.503
6a
7.20
0.985
-0.967
6b
6.80
0.950
-0.895
6c
6.03
0.972
-1.080
6d
5.50
0.971
-0.934
6e
5.40
0.994
-0.928
2a
5.12
0.966
-1.102
3g
7.60
0.994
-1.000
Tropicamide
6.02
0.992
-1.028
Table 1: In vitro anticholinergic activity - guinea pig ileum assay. All values are
means of four determinations.

67
(Figure 10) (Table 1). The pA2 values unambiguously quantify the potency of
an antagonist for a receptor. The slopes of Schild plots were around unity for
all compounds tested except for compound 3f. If the linear regression
coefficients obtained in the Schild plot are close to unity, the agents are
considered to be pure antagonists (Kenakin, 1982). The deviation of slope from
unity for compound 3f could be due to insufficient equilibration time of the
antagonist for the receptors or the activation of other receptors at high
concentration of the compound which might produce a physiological
antagonism of the agonist responses (Kenakin, 1982). An equilibration time of
ten minutes was allowed for the antagonists before the addition of carbachol in
all determinations. Probably this time was not sufficient for compound 3f to
equilibrate.
The pA2 value of methscopolamine determined in this study is identical to
the value reported in the literature (Triggle & Triggle, 1976). The pA2 values
were found to decrease with increasing chain length. The importance of
hydrophobic forces in cholinergic drug receptors have long been recognized
(van Rossum and Ariens, 1957). In a study carried out with straight chain
esters of phenylacetic acid, the following conclusions have been drawn
(Banerjee & Lien 1990).
CH(R)-COO-CH2-CH2-N(C2H5)2.HCl

68
1. substitution of benzylic carbon with 1 or 2 carbon alkyl groups increased
anticholinergic activity, while further increase in chain length decreased activity,
probably due to steric interactions,
2. unsubstituted phenylacetic acid derivative had very low activity,
3. substitution of benzylic carbon with hydroxymethyl group markedly
enhanced the activity, suggesting the possibility of a weak hydrogen bonding
site at the receptor site. In the present study, -CH2OH group of
methscopolamine is substituted with either -COOR or CH2COOR group. The
most potent compounds of both the series are about 2 log units less potent
than methscopolamine. By design, this compromise of potency is not a matter
of concern as long as the compounds are deactivated, in vivo, to less potent
and less toxic moieties. The pA2 values of both the phenylmalonic and
phenylsuccinic series have been found to decrease with increasing chain length
(Table 1). Contrary to the observation that the unsubstituted phenylacetic acid
esters had very low activity (Benerjee & Lien, 1990), 3g, a phenylacetic acid
derivative of scopine is active. The pA2 values of the phenylmalonic analogs are
about one half to one log unit higher than the corresponding phenylsuccinic
analogs. This could be due to the increased steric hinderance due to the
introduction of a methylene group. 2a, the hypothetical lead metabolite of
phenylsuccinic analogs, is about 100 times less potent than the most active
compound (6a) of the series suggesting that the soft drugs (6 a-e) on
hydrolysis, in vivo, would yield a metabolite which is practically inactive.

69
In Vivo Activity
In vivo activity was studied by three methods viz: mydriatic activity on
unilateral topical instillation into rabbit eye and on systemic administration into
rabbits, and muscarinolytic activity against acetylcholine induced bradycardia on
systemic administration into rats.
Mydriatic activity
The mydriatic dose response curves were obtained by administering
increasing concentrations of the drugs until the maximum dilation was achieved
(Figures 11, 12 & 13). Due to the mechanical restriction, the pupil will not dilate
infinitely. Hence the lowest dose which produces the maximum achievable
dilation was used as the dose for comparison. Methscopolamine (0.05% w/v),
tropicamide (0.33% w/v), 3a (0.75% w/v), 3b (1% w/v), 3c (1.25% w/v), 3d (1%
w/v), 6a (5% w/v), 6b (5% w/v) and 6c (1.5% w/v) produced equieffectiveness.
3e exhibited only 89% of the maximum effect even at higher concentrations. 3f
(0.5% w/v) and 6d (2% w/v) produced intense irritation with erythema, tears
and secretions, and were not studied further. Moderate irritation was noticed
with tropicamide, 3d, 3e and 6c, whereas none to mild irritation was noticed
with methscopolamine, 3a, 3b, 3c, 6a and 6b. The maximal dilation was
achieved about 30 minutes after instillation without any significant difference
between 1, tropicamide and soft drugs.
At equieffective doses, the percentages of AUC24hr for the drugs to that
of methscopolamine (Figure 14) are: 3a - 23.2%, 3b - 45.6%, 3c - 60.6%, 3d -

70
Concentration (w/v)
Concentration (% w/v)
Concentration (% w/v) Concentration (% w/v)
Figure 11: Mydriatic dose response curves in treated rabbit eye. Each value is the mean of
twelve readings and error bar represents SEM. The arrows denote the equieffective doses.

71
% of maximum dilation
120 -T
o H—
0.0001
t i i i i i M i i i i i rm i i i i r
0.0010 0.0100
Log molar concentration
3* 3b -B- 3o Trop 3d -^3*
Mil
0.1000
Figure 12: Mydriatic dose response curves in rabbit treated eye after unilateral
instillation. Each value is the mean of twelve readings.

72
% of maximum dilation
Log molar concentration
+ 1 ^Trop — 6b -£“6c
Figure 13: Mydriatic dose response curves in rabbit treated eye after unilateral
instillation. Each value is the mean of twelve readings.

73
AUC(24hrs)
Trop 3a
Figure 14: AUC24hrs for the treated after unilateral instillation into the rabbit eye.
Each value is the mean of twelve readings. Error bar represents SEM.

74
Hours
Figure 15: Duration of mydriatic activity after unilateral instillation into the rabbit
eye. Each value is the mean of twelve readings. Error bar represents SEM.

75
Pupil dilation (mm)
^ 3« -H 3b 3c ■S" So Me Br ^ Trop ^ 3d ^ 3e
1 Control
Figure 16: Time vs mydriatic response curves in the treated eye after unilateral
administration of equieffective doses. Each reading is the mean of twelve
readings.

76
187%, 3e - 49.1%, 6a - 22.4%, 6b - 32.7%, 6c - 60% and tropicamide - 58.3%.
At equieffective doses the recovery periods for the drugs (Figure 15) are as
follows: methscopolamine - 20 hrs, tropicamide - 5.7 hrs, 3a - 3.7 hrs, 3b - 6.8
hrs, 3c - 9.8 hrs, 3d - 36 hrs, 3e - 5.9 hrs, 6a - 3.0 hrs, 6b - 4.5 hrs and 6c - 6.9
hrs. The time vs mydriatic response curves for the drugs at equieffective doses
are summarized in Figure 16. The control eye into which only vehicle was
administered did not show any significant dilation. At 4 hours, 3a and 6a, and
methscopolamine (0.05%) exhibited significant difference in dilation (P < 0.001).
At 4 hours, 3a and 6a, and tropicamide (0.33%) also exhibited significant
difference in dilation (P < 0.005).
The time course of mydriatic activity in the untreated eyes (at
equieffective doses) is depicted in Figure 17. At equieffective doses, the
percentages of AUC6hr to that of methscopolamine (Figure 18) are:
tropicamide - 17.6%, 3a - 2.3%, 3b - 15.7%, 3c - 10.9%, 3d - 17.5%, 3e - 9.8%,
6a - 10.4%, 6b - 8.1% and 6c - 13.3%. At 30 minutes, the difference in dilation
between 3a and 6a, and methscopolamine was found to be significant (P <
0.001). At 30 minutes, the difference in dilation between 3a and tropicamide
was also found significant (P < 0.05).
At equieffective doses all the soft drugs tested, except 3d, showed
significantly shorter duration of mydriatic action in the treated eye than
methscopolamine. Compounds 3a and 6a exhibited significantly shorter
duration than tropicamide. The short duration of mydriatic action could possibly

77
Pupil dilation (mm)
Time (hrs)
-©- 3« 3b -fc- 3c "EM Trop 3d 3e Control
Figure 17: Time vs mydriatic response curves in the untreated eye after
unilateral administration of equieffective doses. Each reading is the mean of
twelve readings.

78
1 Trop 3a 3b So 3d 3o Control 6a 6b 6o
Figure 18: AUC6hrs for the untreated eye after unilateral instillation of equieffective
doses. Each value is the mean of twelve readings. Error bar represents SEM.

79
be due to rapid hydrolysis of soft drugs in the eye. A parabolic relationship has
been reported between chain length and in vitro hydrolytic rates of ester
prodrugs by ocular esterases (Chang and Lee, 1983); compounds with 4-5
carbon chain exhibiting fastest hydrolytic rates. In this study, compounds
synthesized with a 2 carbon alcohol (ethyl, 3a and 6a) and 3 carbon alcohol
(propyl, 3b and 6b) were the shortest acting. In fact, 3d with an n-pentyl ester
was even longer acting than 1. This may be due to binding to ocular pigments
through hydrophobic interactions (Koneru et al. 1986).
The extent of mydriasis is an instantaneous response of the quantity of
the drug residing in the biophasic tissue (iris musculature) and hence the time
course of the pupil response directly reflects the change of drug concentration
in the iris (Mishima, 1981). More than 90% of topically administered drugs have
been reported to be drained into systemic circulation through nasolacrimal duct
without entering the interior of the eye. This results in a high incidence of
systemic side effects after ocular administration of drugs. In this study, the
drugs were administered into only one eye of the rabbit. The other eye served
as an indicator of systemic absorption of the drug and its subsequent side
effects. The untreated eye was observed to dilate in methscopolamine treated
animals but not with soft drug treated animals. This could be due to the
persistence of methscopolamine in the systemic circulation in comparison to the
rapid systemic inactivation of soft drugs.

80
Clinically mydriatics are used more for diagnostic rather than for
therapeutic purposes. Hence persistence of the mydriatic action after a certain
required period of time is both unnecessary and inconvenient. One of the aims
of the present study was to reduce the duration of mydriatic action by
incorporation of a metabolically labile moiety (ester) into the molecule. The
esterases present in the interior of the eye, especially in the iris-ciliary body, will
aid in the metabolism of the drug to the more polar metabolite. The net result
will be a reduction in the duration of activity as shown by the mydriatic activity
studies with soft drugs containing a 2 or 3 carbon ester chain (3a, 3b, 6a and
6b). Further increase in ester chain (3d) gradually increased duration of
mydriatic action. This could be due to two reasons: inefficient hydrolysis by
carboxylesterases and non-specific binding of the drug to ocular pigments by
hydrophobic interactions. The soft drugs with shorter chain length are found to
have no or very low irritation potential on topical administration into the eye.
The irritation potential is found to increase with increasing chain length as was
seen with compounds 3f and 6d. There are reports in the literature which
suggest the relationship between hydrophobic side chains and irritation
potential of pharmaceutical compounds (Fraunfelder, 1989).
Mydriatic activity of intravenous administration to rabbits
Equipotent doses of methscopolamine and soft drugs 3a and 6a were
administered into rabbit ear vein and the mydriatic activity was followed (Figure
19). At all the doses tested, 1 exhibited an extremely long duration of mydriatic

81
Pupil dilation (mm)
-S- 3a 6a
Figure 19: Mydriatic activity after intravenous administration of equipotent doses
to rabbits. Each value is the mean of eight readings (from four animals).

82
action indicating the persistence of the compound in systemic circulation for
prolonged periods of time. Comparatively, 3a and 6a exhibited significantly
shorter duration of mydriatic action indicating facile metabolism and elimination
from the circulation. At a dose of 100 /jg of 1 and equipotent doses of 3a and
6a, mydriatic activity lasted for about 30 hours with 1 compared to 4.5 hours
with 3a and 8 hours with 6a.
The long duration of mydriatic action after systemic administration of 1
into rabbits is consistent with the result reported for humans. Prolonged
suppression of resting pulse rate, decreased secretion of saliva and long lasting
mydriatic activity have been reported in humans after intramuscular
administration of 1 (Brand, 1969).
Muscarinolytic activity on intravenous administration into rats
Effect on the resting heart rate. All the compounds tested (1, 3a and 6a)
showed no significant change in the resting heart rate during the one hour
period after the administration.
Effect on acetylcholine induced bradycardia. The muscarinolytic activity
of 1, 3a and 3g were studied by intravenous administration of equipotent doses
into anaesthetized rats followed by bolus intravenous injections of acetylcholine
at different time intervals. The blockade of cholinergic activity was studied by
recording the heart rate (Figure 20). Methscopolamine completely inhibited
activity of acetylcholine even one hour after the administration compared to 10
minutes with 3a and 40 minutes with the metabolite 3g. Saline control did not

83
exhibit any muscarinoiytic activity. From the in vitro metabolism results
described in later sections it can be concluded that rat metabolizes 1 less
efficiently than the soft drug 3a or the metabolite 3g. This study shows that
compared to methscopolamine, the soft drug 3a undergoes more facile
metabolism probably producing an inactive metabolite. The activity profile of
compound 3g is discussed in a later section.
In Vitro Stability
Analytical procedure
An accurate and reproducible qualitative and quantitative method of
analysis is needed for the estimation of compounds and their degradation and
metabolic products. The High Pressure Liquid Chromatography (HPLC) has
been the most widely used method for such purposes. The main advantages
of HPLC are its reproducibility, accuracy of quantitation and economy of
operation. Various types of columns and detectors have been developed for
HPLC. In the present study, a reverse phase CN column was used. The
detection was made with the UV absorbance detector set at 254 nm. The
mobile phase consisted of varying proportions of acetonitrile and phosphate
buffer. Octane sulfonic acid was used as an ion-pairing agent. The inclusion of
an ion-pairing agent has greatly improved the resolution of the peaks. The
concentration vs area under the peak plot showed linearity (r = 0.999) for the
range of 0.2 to 2 ¡jg of injected sample with a detection limit of 0.2 fjg

84
O
-10
-20
-30
-40
-60
-60
-70
0 10 20 30 40 50 60
Time (mins)
— Control H— 3a 3g --B-1
% decrease in heart rate
70
Figure 20: Muscarinolytic activity against acetylcholine induced bradycardia in rats
after intravenous administration of equipotent doses. Each value is the mean of
four determinations.

85
(approximately 0.4 nanomoles) of Injected sample. Due to low absorbance of
the compounds, the detection limit of the compound is relatively high.
In vitro stability in buffers
The stability of 3a (chosen as a prototype of the series) was studied in
USP buffers (pH 1.5 to 9). 3a exhibited a typical inverted bell shaped pH profile
(Figure 21). The optimum stability was observed in pH range 3.5-4. 3a was
extremely unstable in alkaline solutions.
The degradation pathways were different in acidic and alkaline media
(Figure 22). In acidic medium, 3g was the major product. This compound is
the decarboxylation product of the carboxylate metabolite (2) generated from
the hydrolysis of ethyl ester group of 3a. In alkaline medium, monoethyl
phenylmalonate and quaternized scopine were the major degradation products
with traces of 3g. In alkaline medium, the hydrolytic attack is occurring at
scopine ester bond rather at the ethyl ester bond. This could be due to the
inductive effect of quaternary head group making the scopine ester more
susceptible for nucleophilic attack.
The stability of 3a was studied in pH 4 buffer at 63°, 56°, 46° and 37°C.
An Arrhenius plot was generated by plotting (1000/absolute temperature) vs In
k (degradation rate constant) and a linear relationship was obtained (Figure 23).
The plot was extrapolated and the expected half life of 3a at pH 4 and 25°C was
calculated to be 515 days.

86
pH
Figure 21: pH profile of compound 3a in buffers at 63°C. k is the degradation
rate constant (per day).

87
3g
Figure 22: Pathways of degradation of 3a in buffers.

88
1000/ÍAT)
Figure 23: Arrhenius plot for compound 3a in pH 4 buffer, k is the degradation
rate constant (per day) and AT is absolute temperature.

89
A drug entity, whatever may be its activity and potency, will not be useful
for therapeutic purposes, if it is unstable at ambient temperature and storage
conditions. A high stability of drug entities in aqueous solutions is needed to
make stable pharmaceutical formulations. The soft drugs, as exemplified by 3a,
fulfil this stability requirement.
Stability in biological media
In vitro stability studies in biological matrices may not result in absolute
values but will give a relative picture of the stability profile of the compounds.
They also give clues to the possible metabolic profile in in vivo conditions. All
the soft drugs and compound 1 exhibited first order degradation kinetics. The
extraction efficiency with acetonitrile was consistently higher than 95% in all the
biological matrices.
Soft drugs of phenvlmalonic series (3 a-f) (Table 2). Soft drugs 3 a-f are
highly unstable in rat plasma. Only traces of the compounds could be detected
after 30 seconds, while 1 has a half of 900 minutes. Soft drugs 3 a-f exhibited
half lives ranging from 49 minutes to 181 minutes in rabbit plasma while 1 has a
half life of 1980 minutes. Soft drugs 3 a-f were highly unstable in rabbit liver
homogenate compared to the relative stability of 1 (t1/2 = 580 mins). The
stability of 3 a-f in human plasma is higher than in rabbit plasma, but when
compared with the stability of 1 in human plasma they are much lower. Soft
drugs with straight chain alcohols (3 a-d) are more susceptible to hydrolysis by

90
porcine liver esterase than 3e. The general trend is longer half lives with
increasing side chain length.
Tropic acid and quaternized scopine were the metabolites of 1. 3g was
the metabolite of the soft drugs 3 a-f.
Soft drugs of Dhenvlsuccinic acid series (6 a-el (Table 3) The half lives
of 6 a-e in 25% rat plasma ranged from 3 to 59 minutes compared to 260
minutes for 1. In rabbit plasma, half lives of 6 a-e ranged from 6.35 hours to
12.25 hours compared to 33 hours for 1. The hydrolytic rates for soft drugs 6
(a-e) were significantly higher than 1 in rabbit liver homogenate. The hydrolytic
rates in human plasma were comparable to rabbit plasma. The half lives
increased with the increasing chain length. 6e was extremely unstable in rat
plasma, but the stability was comparable to the other compounds in the rest of
the media.
Tropic acid and quaternized scopine are metabolites of 1. The
carboxylate metabolite, 2a, was the metabolite of 6 a-e.
Stability of 3a in rabbit ocular tissues (Figure 24) Rabbit aqueous humor
and iris-ciliary body exhibited higher hydrolytic rates (nmoles
hydrolyzed/min/mg of protein) than rabbit plasma while cornea exhibited about
equal hydrolytic rate.
Stability of 3a in pure enzyme preparations (Figure 25) Carboxylesterase
from porcine liver exhibited extremely high hydrolytic rate (nmoles/min/mg of
protein) compared to acetyl cholinesterase and pseudocholinesterase.

Compound
Rat
plasma
Rabbit
plasma
Rabbit
liver
homogenate
Human
plasma
Porcine
liver
esterase
Half life (minutes)
1
900
1980
580
2170
—
(0.996)
(0.988)
(0.977)
(0.986)
3a
< 1
81
16
279
61
(0.989)
(0.995)
(0.991)
(0.998)
3b
< 1
49
6.4
505
77
(0.990)
(0.990)
(0.979)
(0.991)
3c
< 1
137
16
195
82
(0.993)
(0.992)
(0.985)
(0.989)
3d
< 1
147
24
281
94
(0.996)
(0.995)
(0.975)
(0.994)
3e
< 1
147
27
558
402
(0.997)
(0.993)
(0.988)
(0.993)
3f
< 1
181
32
812
—
(0.991)
(0.994)
(0.977)
Table 2: In vitro stability of soft drugs 3 (a-f) in biological media. Each value is the mean of three
determinations. Figures in parenthesis are regression coefficients.
CO

92
Compound
25% rat
plasma
(mins)
Rabbit
plasma
(hours)
Rabbit
liver
homogenate
(mins)
Human
plasma
(hours)
1
280
33
540
36
(0.996)
(0.998)
(0.977)
(0.986)
6a
28
6.35
29
6.5
(0.990)
(0.998)
(0.978)
(0.991)
6b
23
10
15
9.8
(0.988)
(0.976)
(0.979)
(0.987)
6c
25
8
33
8
(0.975)
(0.989)
(0.973)
(0.994)
6d
60
12.3
115
13.3
(0.987)
(0.985)
(0.994)
(0.985)
6e
4
7.3
40
(0.981)
(0.988)
(0.972)
Table 3: In vitro stability of soft drugs 6 (a-e) in biological media. Each value is the
mean of three determinations. Figures in parentheses are regression coefficients.

93
Plasma
Aqueous humor
ICB homogenate
Corneal homogenate
Liver homogenate
0 1 2 3 4 5
Hydrolysis rate (nmoles/min/mg of protein)
Figure 24: Hydrolytic rates of 3a in rabbit tissues. The rates are represented as
the nanomoles of 3a hydrolyzed per min per mg of protein. Each value is the
mean of three determinations. P < 0.001 (liver homogenate vs plasma or ocular
tissues). P < 0.05 (plasma vs aqueous humor or iris-ciliary body homogenate).

94
Caboxyle8terase
Acetylcholinesterase
Pseudocholinesterase
0 20 40 60 80 100 120 140
Hydrolysis rate (nmoles/min/mg of protein)
Figure 24: Hydrolytic rates of 3a in pure enzyme preparations. The hydrolytic
rate are represented as the nanomoles of 3a hydrolyzed per min per mg of
protein. Each value is the mean of three determinations.

95
When hydrolytic rates of analogous compounds of phenylmalonic and
phenylsuccinic series were compared, the compounds of the latter series were
more stable in the plasma of all animals tested. The stability of analogous
compounds of both series was comparable in rabbit liver homogenate. The
higher lability of phenylmalonic compounds might be due to the proximity of the
ester group to the aromatic ring which possibly stabilizes the reaction
intermediate by resonance and inductive interactions.
When the hydrolytic rates (nmoles of substrate hydrolyzed/min/mg of
protein) of 3a in various rabbit tissues were compared, liver homogenate
exhibited the highest capacity. The hydrolysis rate was slightly higher in ocular
tissues compared to plasma. The high instability of all the compounds in liver
homogenate is very significant since the liver is the main metabolizing organ in
the body. When the ocularly administered drug is drained into the systemic
circulation, it passes through the liver where it will be metabolized efficiently.
This probably explains the absence of dilation of the untreated eye on unilateral
administration of the soft drugs. Another significant aspect is the moderate
stability of 3a in ocular tissues. Thus at the site of action i.e., eye, 3a is able to
exert its pharmacological effect for a certain duration of time before being
metabolized and excreted out of the eye.
The in vitro stability studies were conducted with an intention to
understand the kinetic and metabolic behavior of soft drugs in biological media.
Ideally, soft drugs should not decompose at the site of action instantaneously.

96
In contrast to a prodrug, a soft drug is a pharmacologically active moiety which
on decomposition yields an inactive metabolite. If the soft drug decomposes at
the site of action rapidly, it will not be able to exert its action. Thus a soft drug
should not be too unstable at the site of action but should be easily
metabolizable in the detoxifying organs of the body (e.g. liver).
Significance of metabolite 3a
The metabolite 3g, as shown by guinea pig ileum assay and mydriatic
activity is a potent anticholinergic. At equieffective doses, 3g has a longer
duration of mydriatic action than 3a in rabbit eye (Figure 26). It also has a
longer half life in biological media than 3a. At equivalent concentrations, 3g has
a half life of 2511 mins compared to 10 mins of 3a in 20% rat liver homogenate.
It is the chemical degradation product of 3a in buffers as well as the enzymatic
degradation product of 3a in in vitro studies in biological media such as plasma.
The only way 3g could form from 3a is by an initial hydrolysis to the carboxylate
metabolite (2), which then undergoes decarboxylation to 3g.
Malonic acids are known to undergo thermal decarboxylation to the
corresponding acetic acids. There are no reports of such decarboxylation
occurring with carbeniciliin, a semisynthetic B-lactam antibiotic containing
phenylmalonic acid moiety in its structure. A recent article (Naylor et al. 1992)
dealing with the hepatic microsomal metabolism of cimetropium, an N-
methylene-cyclopropyl quaternary derivative of scopolamine, indicated the
formation of a metabolite similar to 3g. The formation of this dehydroxymethyl

97
Pupil dilation (mm)
Time (hours)
3a 3g
Figure 26: Mydriatic activity in treated rabbit eye after unilateral administration of
equieffective doses of 3a (0.75%) and 3g (0.5%). Each value is the mean of twelve
readings.

98
metabolite of cimetropium could be due to the oxidation of hydroxymethyl
group to a carboxyl group, analogous to the lead compound 2, which then
decarboxylates.
The in vitro metabolic data of 3a indicates that it degrades to form
metabolite 3g, which is as potent as the parent compound itself. The
pharmacodynamic data does not correlate with this finding. 3a has a very short
duration of mydriatic as well as muscarinolytic action on acetylcholine induced
bradycardia (Figure 20) compared to 3g. If 3g is the quantitative metabolic
product of 3a in vivo, 3a should also have a longer lasting pharmacological
effect. But it is not the case as clearly shown by the pharmacodynamic studies
of 3a. The situation could be different in in vivo conditions. The carboxylate
metabolite (2) generated on hydrolysis of soft drugs 3 a-f, due to its polar
nature could possibly be conjugated rapidly and eliminated. More detailed
metabolic investigations are needed to confirm this hypothesis.
In Vitro Penetration Studies
Partition coefficients (PI
The n-Octanol/water partition coefficients of 1, 3 a-e and 6a were
determined (Table 4). The value determined for 1 (2.98 X 10'2) coincides with
the value reported in the literature (2.6 X 10'2) (Michales et al. 1975). The
partition coefficients for soft drugs ranged from 1.5 X 10'2 to 12.5 X 10'2. The
partition coefficients are found to increase with increasing chain length for n-

99
alkyl compounds (3 a-d). For compound 3e which has a branched side chain
(c-hexyl), the increase in partition coefficient is not consistent with the increase
in the number of carbon atoms in the side chain. The inconsistent increase in
partition coefficient between straight chain and branched chain analogs of
atropine Is in agreement with this result (Kumar et al. 1992). The difference in
the partition coefficients between n-alkyl side chain compounds and branched
side chain compounds could be due to the masking of the cationic head by the
wrapped over hydrophobic n-alkyl side chain thereby making the whole
molecule more lipophilic (Hada et al. 1983). The same masking is not possible
with branched chain compounds.
In vitro transdermal penetration
The in vitro penetration of 1, 3 a-e and 6a across hairless mice skin was
studied. The time vs cumulative amount penetrated curves show a linear profile
after an initial lag period. A representative curve is depicted in Figure 27. The
permeability coefficients (Kp) are listed in Table 4. The lag periods ranged from
1 to 4 hours without significant differences between the compounds and 1. The
soft drugs were found to metabolize to a great extent during their penetration
across the skin (Figure 27). In in vitro studies carried out with transdermal
penetration of steroid esters across fuzzy rat skin, considerable retention of
enzymatic activity was shown even after prolonged periods of study (Bronaugh
et al. 1989). No noticeable metabolism was seen with 1. The Kp of 1
determined in this study (5.3 X 10'5 cm/hr) is one order of magnitude higher

100
than the value reported for human skin permeability (3 X 1CT6 cm/hr) (Michaels
et al. 1975). The permeability coefficients ranged from 2.9 X 10'5 cm/hr to
21.9 X 10’5 cm/hr, with the most lipophilic compound, 3d, penetrating to the
highest extent. The amount that accumulated in the receptor chamber in all the
studies accounted for less than ten percent of the applied drug. Thus
essentially sink conditions were maintained throughout the period of study. The
maintenance of sink conditions in the receptor is essential in order to maintain
zero-order flux conditions. A linear correlation was obtained (r2= 0.85) between
log partition coefficient (log Kp) and log permeability coefficient (log P) for the
compounds tested (Figure 28).
Skin acts as a good barrier preventing the entry of xenobiotics into the
body. Unfortunately, the topically applied therapeutic agents also encounter
these barrier properties of the skin. Due to the accessible nature of the skin,
various attempts have been made to administer drugs noninvasively into the
body through the skin. Some well known examples of successful transdermal
administration for systemic action include scopolamine to treat motion sickness,
nitroglycerin for angina and nicotine for tobacco addiction. Some attempts
have also been made to localize the effect of the drug in skin itself i.e., dermal
delivery. The examples include chemical delivery system of acyclovir for the
treatment of herpetic infections (Bodor and Prashant, 1991) and the prodrugs of
nucleotide antimetabolites as antiproliferatives (Waranis and Sloan, 1987). The
general principle behind these designs has been to increase the lipophilicity of

101
+ 3c O 3g A 3c+3g
Figure 27: A representative trace of the in vitro penetration of 3c across
hairless mice skin. 3c denotes the intact drug, 3g denotes the metabolite and
3c+3g denotes the total amount drug and metabolite in the receiver cell. Each
value is the mean of three determinations and the error bar represents the
SEM. Error bars not appearing in the figure are smaller than the markers.

102
Compound
Octanol/Water
Partition
Coefficient
(X 10'2)
Hairless mice skin
Permeability
Coefficient (cm/hr)
(X 10'5)
Lag period
(hours)
1
2.98
5.30 _± 0.65
2.4
3a
1.49
2.90 _+ 0.45
1.9
3b
2.78
7.00 _± 0.68
2.7
3c
5.09
14.7 ± 0.95
3.1
3d
12.5
21.9 ± 1.75
3.4
3e
3.39
4.52 ± 0.52
2.8
6a
6.33
9.20 _± 0.70
1.5
Table 4: The n-Octanol/water partition coefficients and the hairless mice skin
permeability coefficients of 1 and soft drugs. The permeability coefficients are the
means of three determinations + SEM.

103
Permeability Coefficient (cm/hr)
Partition Coefficient
Figure 28: Relationship between log Partition Coefficient and log Permeability
Coefficient (hairless mice skin).

104
the compounds so that they can traverse the hydrophobic stratum corneum to
reach the inner layers of the skin. Some comprehensive studies have been
reported in which percutaneous permeability has been correlated to the
corresponding oil-water partition coefficients (Guy and Hadgraft, 1989). In
general a parabolic relationship has been shown to exist between log P and log
Kp with the maximum permeability attained typically at a log P in the range 2.0-
2.5. Thus an essentially linear relationship is shown to exist at lower P values
with log Kp increasing with log P (Yano et al. 1986). In the present study, the P
values of the compounds ranged from 1.49 X 10'2 to 12.5 X 10'2, which are well
below the optimal P values reported for maximal permeability and are in the
ascending arm of the parabola. Hence as expected a linear correlation exists
between log P and log Kp. The permeabilities of soft drugs which are very
hydrophilic are comparable to the permeabilities reported for other polar
compounds. It has been postulated that the phospholipid bilayer, which
constitutes the stratum corneum, and is comprised largely of fatty acid
zwitterions, may solubilize the organic electrolytes thus facilitating their
absorption through the skin (Michaels et al. 1975).
The inhibition of eccrine sweating by scopolamine and its derivatives has
been reported (MacMillan et al, 1964). The soft anticholinergics synthesized in
the present study could have implications for use as topical antiperspirants. A
topically applied antiperspirant should traverse through the sweat duct to the
site of action i.e., sweat gland to exert its effect. The antiperspirants that are in

105
use today act as astringents rather than by any specific receptor mediated
mechanism (Kuno, 1956). Hence no detailed studies have appeared in the
literature so far about the physicochemical characteristics that are ideal for an
antisecretory agent. In the case of an antiperspirant drug, a balance should be
achieved between the localization of the drug in the interior of the skin and its
diffusion out into the systemic circulation. Figure 29 illustrates the pathways of
percutaneous absorption. Katz and Poulsen (1973) have discussed extensively
the importance of the three possible routes of transcorneal penetration viz:
through the hair follicle, via the sweat ducts and across the continuous stratum
corneum. Even though the permeability rates of polar molecules are reported
to be several orders of magnitude higher through the sweat ducts and hair
follicles than the stratum corneum, their contribution is very insignificant since
their fractional diffusional volume is low compared to stratum corneum
(Scheuplein & Blank, 1971). The transdermal studies of the soft drugs
synthesized in this study were conducted to determine their permeability
characteristics.
In vitro transcorneal penetration
The in vitro transcorneal penetration of 1, 3a and 6a across rabbit cornea
was studied. The time course of permeation is depicted in Figure 30.
Compound 3a (1.6 X 10-3 cm/hr) was found to have slightly higher permeability
than 1 (1.1 X 10-3 cm/hr). Compound 6a was found to have significantly higher
permeability (2.9 X 10-3 cm/hr) than 1 (Figure 31). The values determined in

106
Figure 29: Network of alternate pathways for percutaneous absorption (Barry,
1983, p. 97).

107
this study are in general agreement with the values reported for polar
compounds in the literature (Grass & Robinson, 1988a). No significant
hydrolysis of either 1 or soft drugs 3a or 6a occurred during the penetration
process. The lag periods for the three compounds are about 1 to 2 hours.
There is no significant difference in lag periods for 1 and the soft drugs 3a or
6a.
The bioavailable fraction of ocularly administered therapeutic agents at
the site of action has been shown to be extremely low (Burstein and Anderson,
1985). This is due to the poor penetration of the compounds through the
cornea. Hence higher amounts of drugs are needed to elicit the required
quantum of response, which sometimes might produce adverse systemic
effects. The movement of hydrophobic drugs across corneal epithelium has
been hypothesized to involve several partitioning steps. In contrast, the
movement of hydrophilic compounds is not well understood. The transcellular
movement by partition controlled process and the paracellular movement
through pores by diffusion controlled process are the two major routes of
absorption through corneal epithelium (Grass & Robinson, 1988a). For
hydrophilic compounds, molecular size and hydrogen bonding ability are
proposed to be the factors which influence the paracellular transport. The
permeability is shown to decrease with increasing molecular size for small
hydrophilic molecules. As the number of potential hydrogen bonding sites
increases, permeability through cornea has been shown to decrease (Grass &

108
Timo (hours)
+ 1 o 3a a 6a
Figure 30: The time course of penetration of 1, 3a and 6a across rabbit cornea
in vitro. Each value is the mean of four readings and the error bars denote
SEM. The error bars not appearing in the figure are smaller than the markers.
The X-axis intercept of the regression line is the lag period.

109
Permeability coefficient (E-4) (cm/hr)
Figure 31: In vitro permeability coefficients across rabbit cornea. Each value is
the mean of four determinations and the error bar represents S.E.M. P < 0.05 (1
vs 6a)

110
Robinson, 1988b). A model depicting the corneal epithelium as a sieve with
aqueous diffusional pathways or actual pores within the cell membranes has
been proposed by Tonjum (1974). The experiments conducted with horse
radish peroxidase have indicated the size of these pores to be less than 3 nm
in diameter (Tonjum, 1974). In another study, the limiting molecular dimensions
of both the corneal endothelium and epithelium were found to be 1-3 nm
(Maurice, 1953). The ultrastructural analysis of the cornea conducted by
scanning electron microscopy also suggested the existence of aqueous
pathways in the cornea of rabbit (Grass & Robinson, 1988c).
In the present study, the hydroxymethyl group of methscopolamine is
substituted with an alkoxycarbonyl group to get a new series of compounds
which are expected to be more lipophilic than 1. This holds true for all the
compounds except 3a. When the corneal penetration characteristics were
compared, compound 6a has slightly higher permeability than 1. The
permeability of compound 3a does not differ significantly from 1, even though
the partition coefficient of 3a is less than that of 1. The lag periods do not differ
significantly for the three compounds tested.
The compounds of the present study are extremely hydrophilic. Hence
the transcellular transport may not be the significant route for their absorption.
The three compounds tested in this study (1, 3a and 6a) do not differ
significantly in their molecular dimensions, but 1 has a free primary hydroxyl
group which can contribute to the hydrogen bonding capacity of the molecule.

111
Probably this explains the conflicting partition and permeability results obtained.
Compound 1 because of its ability to form hydrogen bonds is able penetrate
the corneal epithelium less efficiently than the soft drugs 3a and 6a.

CHAPTER 5
CONCLUSIONS
The anticholinergics on topical application (e.g. as eye drops) are known
to elicit systemic side effects due to their drainage from the site of application.
The present study involves the development of a new type of anticholinergics,
called soft anticholinergics, which are based on methscopolamine. They are
designed to act at the site of application but not to act systemically due to their
rapid metabolic inactivation in the systemic circulation. A hypothetical
carboxylate metabolite of methscopolamine (an oxidation product of the primary
hydroxyl group) was chosen as the lead compound and it was reactivated by
esterification with cyclic and aiicyclic alcohols to yield a series of compounds
(phenylmalonic series). Another series of compounds were also designed by
including a methylene group into the side chain of the phenylmalonic series
(phenylsuccinic series).
The soft drugs of phenylmalonic and phenylsuccinic series are active as
anticholinergics as shown in the guinea pig ileum assay. Compound 3b (n-
propyl) in the phenylmalonic series and compound 6a (ethyl) In the
phenylsuccinic series are the most potent of the respective series. The potency
has been found to decrease with increasing chain length. The anticholinergic
112

113
activity of the carboxylate metabolite (2a) of phenyisuccinic series is about 100
times less than the most potent of the series (6a). Compound 3g, which was
the in vitro metabolite in biological media of phenylmalonic acid analogs (3 a-f)
was equipotent with that of the most potent of the series.
The in vitro stability study in buffers carried out on compound 3a showed
that the soft drugs are extremely stable in aqueous solutions at pH 3.5-4. The
stability study carried out in the biological media showed the extreme instability
of 3 a-f in rat plasma and rabbit liver homogenate and shorter half lives than 1
in rabbit and human plasma. This clearly indicates the increased metabolic
lability of soft drugs in biological media compared to 1. Compound 3g has
been identified as the metabolic degradation product of 3 a-f. The stability
studies carried out with soft drugs 6 a-e in various biological media indicated
higher metabolic lability of these compounds compared to 1. Compounds 6 a-e
degraded in the biological media to yield the carboxylate metabolite 2a which is
inactive.
The mydriatic activity of 1 and the soft drugs was tested in rabbit eyes.
At equieffective doses, all soft drugs tested, except 3d, exhibited shorter
duration of mydriatic action. Compounds 3a and 6a showed significantly
shorter duration of mydriatic action than tropicamide, the shortest acting
mydriatic available in the market. A significant dilation of untreated eye after
unilateral topical administration of 1 was observed. In the soft drug treated
animals no dilation of the untreated eye was observed. The absence of dilation

114
in the untreated eye of soft drug treated animals indicates the facile metabolism
of soft drugs in systemic circulation compared to the persistence of 1. This was
further confirmed by testing mydriatic activity after intravenous administration of
equipotent doses of 1, 3a and 6a into rabbits. While the mydriatic activity, an
indicator of the presence of the drug in the body, in 1 treated animals persisted
for over 24 hours, in 3a and 6a treated animals it lasted for 4.5 and 8 hours
respectively.
In another experiment performed to compare the in vivo muscarinoiytic
activity in rat, 1 completely abolished the muscarinic effects of acetylcholine on
heart for more than 1 hour, while 3a and 3g, abolished the effects of
acetylcholine for 10 and 40 minutes respectively indicating the high metabolic
lability of soft drug 3a in in vivo conditions.
Compound 3g, the in vitro metabolic degradation product of 3 (a-f), is a
potent anticholinergic, more stable in biological media than 3a and longer acting
as a mydriatic in rabbits. These results are in conflict with the short duration of
anticholinergic activity on both topical and systemic administration, and highly
metabolically labile nature of the soft drugs, especially 3a. Probably, 3a and
other soft drugs of phenylmalonic series are first metabolized to the carboxylate
metabolite (2) in vivo, which is then rapidly conjugated and eliminated, thus
avoiding the formation of metabolite 3g. Further detailed metabolic
investigations are needed to confirm this hypothesis.

115
The in vitro/in vivo activity and stability studies clearly show that the soft
drugs 6 a-c are fairly potent anticholinergics which degrade into an essentially
inactive metabolite (2a) in vivo. The carboxylate metabolite 2a fits into the
category of inactive metabolites. When the soft drugs 6 a-c, are applied
topically to eye or other peripheral tissues/organs, they will exert an
anticholinergic effect locally for a certain duration of time before being
metabolized to the inactive polar carboxylate metabolite (2a) which will then be
eliminated. Whatever the fraction of soft drugs (6 a-c) that enters systemic
circulation will enter the liver through the hepatic circulation where it will be
efficiently metabolized apparently to the carboxylate metabolite and excreted.
Thus a local action possibly without systemic side effects is achieved.
The partition coefficients (P) were found to increase with increasing chain
length. The permeability coefficients (K ) across hairless mice skin were found
to increase with increasing chain length. A linear correlation (r2= 0.875) was
found to exist between log P and log Kp. The soft drugs were found to
metabolize during the penetration process. The transcorneal permeability of 6a
was higher than 1, while the permeability of 3a did not differ significantly from 1.

REFERENCES
Alexandras, E., (1985) in The Pupil. Springer Verlag, Berlin, 30.
Ashton, L, Wang, W„ & Lee, V.H., (19911 J. Pharmacol. Exp. Therp.. 259. 719.
Ariens, E.J., (1966) Adv. Drug. Res.. 3, 325.
Ariens, E.J., & Simonis, A.M., (1982) in Strategy in Drug Research. 3. Ed: J.A.
Keveriing, Elsevier, Amsterdam, 165.
Austel, V., (1989) in Modern Drug Research. Eds: Y.C. Martin, E. Kutter & V.
Austel, Marcel Dekker, New York, 243.
Bailey, B.M.W., & Pearce, D.E., (19851 Br. Dent. J.. 158. 17.
Barlow, R.B., Franks, F.M., & Pearson, J.D., (19731 J. Med. Chem.. 16,439.
Barlow, R.B., & Ramtoola, S., (19801 Br. J. Pharmacol.. 71. 31.
Bartek, M.J., & LaBudde, J.A., (19751 Animal Models in Dermatology. Ed: H.l.
Maibach, Churchill Livingston, New York, 103.
Barry, D.W., (1983) in Dermatological Formulations: Percutaneous Absorption.
Marcel Dekker, New York, 240.
Bartlett, J.D., Jaanus, S.D., & Pagano, V.T., (1989) in Clinical Ocular
Pharmacology. 2, Eds: J.D. Bartlett & S.D. Jaanus, Butterworths,
Stoneham, 69.
Bauer, K., Brunner, F.F., Distlerath, L.M., Lippa, E.A., Binkowitz, B., Till, P., &
Kaik, G., (19911 Clin. Pharmacol. Ther.. 49. 658.
Benerjee, S., & Lien, E.J., (19901 Pharm. Res.. 7. 746.
116

117
Berman, E.R., (1991) in Biochemistry of the Eve. Plenum press, New York, ix.
Berry, D.P. Jr, van Buskirk, E.M., & Shields, M.B., (1984) Arch. Ophthalmol..
102. 42.
Bickers, D.R., (1980) in Current Topics in Cutaneous Toxicity. Eds: V.A. Drill &
P. Lazar, Academic Press, New York, 95.
Birkhimer, L.J., Jacobson, P.A., Olson, J., & Goyette, D.M., (1984) J. Fam.
Prac.. 18, 464.
Bodor, N., (1981) in Drugs of the Future. 6. 165.
Bodor, N., (1984) in Advances in Drug Research. 13, Ed: B. Testa, Academic
Press, Orlando, 255.
Bodor, N., & Chikhale, P., (1991) J. Pharm. Sci.. 80. 402.
Bodor, N., Elkoussi, A., Kano, M., & Khalifa, M., (1988) J. Med. Chem.. 31.
1651.
Bodor, N., Elkoussi, A., & Hammer, R.H., (1990) J. Biopharm, Sci.. 1. 215.
Bodor, N., Woods, R., Raper, C., Kearney, P., & Kaminski, J., (1980) J. Med.
Chem.. 23, 474.
Brand, J.J., (1969) Br. J. Pharmac.. 35. 202.
Bronaugh, R.L., Collier, S.W., Strom. J.E., Nathan, D.L., & Stewart, R.F., (1989)
in Prediction of Percutaneous Absorption. Eds: R.C. Scott, R.H. Guy & J.
Hadgraft, IBC Technical Services Ltd., London, 58.
Brouiilette, G., (1987) Ph. D. dissertation. University of Florida, 27.
Brun, R., & Hunziker, H., (1955) Dermatológica. 110. 245.
Burstein, N.F., & Anderson, J.A., (1985) J. Ocular. Pharmacol.. 1. 309.
Chang, S.C., Bundgaard, H., Buur, A., & Lee, V.H., (1987) Invest. Ophthamol.
Vis. Sci.. 28, 487.
Chang, S.C., Chien, D.S., Bundgaard, H., & Lee, V.H.L., (1988) Exp. Eve. Res..
46, 59.

118
Chang, S.C., & Lee, V.H., (1982/831 Curr. Eve. Res.. 2. 651.
Collins, K.J., Sargent, F., & Weiner, J.S., (19591 J. PhvsioL 148. 592.
Das, N.D., & Shichi, H., (19811 Exp. Eve. Res.. 33, 525.
Dasta, D.F., & Garaets, D.R., (19821 Am. Pharm.. NS22. 29.
Druzgaia, P., Hochhaus, G., & Bodor, N., (19911 J. Steroid. Biochem.. 38. 149.
Extra Pharmacoeooeia. Martindale’s. (1989) 29, The Pharmaceutical Press,
London, 522.
Fraunfelder, F.T., (1989) in Drug Induced Ocular Side Effects & Drua
Interactions. 3, Lea & Febiger, New York, 445.
Freund, M., & Merin, S., (19701 Am. J. Ophthalmol.. 70. 637.
Gettes, B.C., (19611 Arch. Ophthalmol.. 65. 632.
Grass, M.G., & Robinson, J.R., (19841 J. Pharm. Sci.. 73. 1021.
Grass, M.G., & Robinson, J.R.. J. Pharm. Sci.. 77. (1988a) 3; (1988b) 15; and
(1988c) 24.
Gray, L.G., (19791 J. Am. Optom. Assoc.. 50. 465.
Greaves, M.W., & Shuster, S., (19891 Handb. Exp. Pharm.. 87/1. Springer
Verlag, Berlin, 193.
Gunes, Z., Hammer, R., Wu, W., Kumar, G., & Bodor, N., (1992), Unpublished
results.
Guy, R.H., & Hadgraft, J., (1989) in Percutaneous Absorption. 2. Eds: R.L.
Bronaugh & H.l. Maibach, Marcel Dekker, New York, 95.
Hada, S., Neya, S., & Funasaki, N., (19831 J. Med. Chem.. 26. 686.
Hadgraft, J., (19801 Int. J. Pharmaceutics. 4. 229.
Hadgraft, J., (1989) in Prediction of Percutaneous Penetration. Eds: R.C. Scott,
R.H. Guy & J. Hadgraft, IBC Technical Services Ltd., London, 254.

119
Hamborg, P.B., Neilsen, M.M., & Thordal, C., (1984) Arch. Ophthalmol.
Copenh.. 62, 485.
Hammer, R.H., Amin, K., Gunes, Z.E., Brouiilette, G., & Bodor, N., (1988) Drug,
Des. Del.. 2, 207.
Hammer, R.H., Wu, W., Sastry, J., & Bodor, N., (1991) Curr. Eve. Res.. 10. 565.
Hansch, C., (1981) Drug. Dev. Res.. 1. 267.
Hem. S.L., & White, J.L., (1989) in Aluminum & Health. A Critical Review. Ed:
H.J. Giteiman, Marcel Dekker, New York, 275.
Higuchi, T., (1977) in Pesian of Biopharmaceutical Properties through Prodruos
and Analogs. Ed: B. Roche, A.Ph.A., Washington D.C., 409.
Jarv, J., & Bartfai, T., (1982) Acta. Chem. Scand. (B], 36. 487.
Johnson, S.F., & Moore, R.J., (1983) Ann. Neurol.. 13. 583.
Kao, J., Hall, J., Shugart, L.R., & Holland, J.M., (1984) Toxicol. AddI.
Pharmacol.. 75, 289.
Katz, M., & Poulsen, B.J., (1971) in Handbook of Experimental Pharmacology.
Eds: B.B. Brodie & J. Gillette, 28, Pt:1, Springer-Verlag, Berlin, 103.
Kenakin, T.P., (1982) Cañad. J. Phvsiol. Pharmacol.. 60. 249.
Kennerdell, J.S., & Wucher, F.P., (1972) Arch. Ophthalmol.. 87. 634.
Kofron, W.G., & Wideman, L.G., (1972) J. Ora. Chem.. 37. 555.
Kumar, G.N., Hammer, R.H., & Bodor, N.S., (1992) Unpublished results.
Koneru, P.B., Lien, E.J., & Koda, R.T., (1986) J. Ocul. PhamacoL 2. 385.
Kuno, Y., (1956) in Human Perspiration. Charles C. Thomas, Springfield, IL,
353.
Lee, V.H., (1983) J. Pharm. Sci.. 72. 239.
Lien, E.J., Alhaider, A.A., & Lee, V.H., (1982) J. Parent. Sci. Tech.. 36. 86.

120
MacMillan, F.S.K., Reller, H.H., & Snyder, F.H., (1964) J. Invest. Dermatol.. 42.
363.
Mager, P., (1984) in Multidimentional Pharmacochemistry - Design of Safer
Drugs. Academic Press, Orlando, 36.
Mager, P„ Theil. S„ & Kling, B„ (1972) Acta. Biol. Med. Ger.. 28. 387.
Malatray, J., & Simon, P., (1972) Therapie. 27, 153.
Mandell, A.I., Stetz, F., & Kitabchi, A.E., (1978) Ophthalmology. 85. 268.
Matsuda, K., (1966) Niiaata laakki Zasshi . 80. 53.
Maurice, D.M., & Mishima, S., (1984) in Pharmacology of the Eve. Ed: M.L.
Sears, Springer Verlag, Berlin, 25.
Michaels, A.S., Chandrasekharan, S.K., & Shaw, J.E., (1975) A. I. Ch. E. J.. 21.
985.
Mishima, S., (1981) Invest. Ophthalmol. Vis. Sci.. 21. 504.
Mitchelson, F., (1984) Tr. Phamacol. Sci.. 5. 12.
Nathanson, J.A., (1985) Curr. Eve. Res.. 4. 191.
Naylor, S., Kajbaf, M., Lamb, J.H., Jahanshahi, M., &
Garrod, J.W., (1992) Biol. Mass. Spectro,. 21. 165.
Notari, R.E., (1981) Pharmacol. Then. 14. 25.
Ottori, T., (1969) Acta. Med. Biol.. 17. 145.
Phillipson, J.D., Handa, S.S., & Garrod, J.W., (1976) J. Pharm. Phamacol.. 28.
687.
Prince, J.H., (1969) in The Rabbit in Eve Research. Charles C. Thomas,
Springfield, X.
Roberts, M.S., Anderson, R.A., & Swarbrick., J., (1977) J. Pharm. Pharmacol..
29, 677.
Sano, M., & Hakusui, H., (1974) Chem. Pharm. Bull. Tokyo. 22. 696.

121
Sato, K., & Dobson, R.L., (1969) Arch. Dermatol., 100. 564.
Scheuplein, R.J., & Blank, I.H., (1971) Phvsiol. Rev.. 51. 702.
Scheurlen, M., Bittiger, H., & Ammann, B., (1984) J. Pharm. Sci.. 73. 561.
Schmidt, I., (1970) J. Am. Qptom. Arch. Am. Acad. Qptom.. 47. 312.
Schoenwald, R.D., & Ward, R.L., (1979) J. Pharm. Sci.. 68. 673.
Scott, R.C., (1989) in Prediction of Percutaneous Penetration. R.C. Scott, R.H.
Guy, & J. Hadgraft (Eds), IBC Technical Services, London, 254.
Shelley, W.B., & Horvath, P.N., (1951) J. Invest. Dermatol.. 16. 267.
Shichi, H., & Neber, D.W., (1980) in Enzymatic Metabolism of Drugs & other
Foreign Compounds. Ed: T.E. Gram, S.P. Medical & Scientific, New York,
333.
Sokolovsky, M., (1984) Int. Rev. NeurobioL 25. 139.
Stoughton, R.B., (1975) in Animal Models in Dermatology. Ed: H.l. Maibach,
121.
Stoughton, R.B., Chiv, F., Fritsch, W., & Nurse, D., (1964) J. Invest. Dermatol..
42, 151.
Tauber, U., (1982) in Dermal & Transdermal Absorption. Eds: L. Brandau,
Wissenschafttiche, Stuttgart, 133.
Tauber, U., & Rost, K.L., (1986) in Skin Pharmacokinetics. Eds: B. Shroot & H.
Shaefer, Larger, Basel, 170.
Testa, B., (1984) in Advances in Drug Research. 13, Academic Press, New
York, 39.
Tonjum, A.M., (1974) Acta. Ophthalmol.. 52. 650.
Triggle, D.J., & Triggle, C.R., (1976) in Chemical Pharmacology of Synapse.
Academic Press, New York, 291.
van Rossum, J.M., & Ariens, E.J., (1957) Arch. Int. Pharmacodvn.. 110. 349.

122
van Rossum, J.M., Hurkmans, H., & Wolter, C.J., (19631 Internat. Pharmacodvn.
Therap.. 143. 299.
Waranis, R.P., & Sloan, K.B., (19871 J. Pharm. ScL, 76. 587.
Weiner, N., (1980) in The Pharmacological Basis of Therapeutics. Eds. A.G.
Gilman, L.S. Goodman & A. Gilman, 6, Macmillan, New York, 120.
Werner, G., & Schmidt, K.H., (19681 Hoppe Sevier’s Z. Phvsiol. Chem.. 349.
741.
Wess, J., Buhl, T., Lambrecht. G., & Mustscher, E., (1990) in Comprehensive
Medicinal Chemistry. 3, Ed: C. Hansch, Pergammon, New York, 423.
Wester, R.C., Maibach, H.I., Guy, R.H., & Novak, E., (19841 J. Invest. Dermatol..
82, 515.
Yano, T., Nakagawa, A., Tsuji, M., & Noda, K., (19861 Life. Sci.. 39. 1043.
Zaugg, H.E., Horrom, B.W., & Borgwardt, S., (19601 J. Am. Chem. Soc.. 2895.

BIOGRAPHICAL SKETCH
Gondi N. Kumar was born on March 4, 1960 in Anantapur, India. He
graduated with a B.S. in pharmacy in 1980 from Kakatiya University, Warangal,
India. He then obtained M.S. in pharmacy in fermentation technology from
Andhra University, Visakhapatnam, India in 1983. He worked as a
manufacturing/ development pharmacist in the pharmaceutical industry in India
for five years. In 1988 he joined the graduate program of the Department of
Medicinal Chemistry, College of Pharmacy, University of Florida and graduated
with a Ph.D. in August 1992. He married Uma Devi in August 1991.
123

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the
Professor of Medicinal Chemistry
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
\a-\/^-rvrhr
Nicholas S. Bodor, Cochairman
Graduate Research Professor of
Medicinal Chemistry
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
degree of Doctor of Philosophy.
Richard H. Hammer,
i ^ i /a —a
Margaret ÓJ James
Professorof Medicinal Chemistry
I certify that i have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
S'
James Simpkins
fbfessor of Pharmacodynamics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
Eric J. Enholm
Assistant Professor of Chemistry

This dissertation was submitted to the Graduate Faculty of the College of
Pharmacy and to the Graduate School and was accepted as partial fulfillment of
the requirements for the degree of Doctor of Philosophy.
August, 1992
Dean, College oTPharmacy
Dean, Graduate School