CHM."
`
`andthe
`BIOLOGICAL
`ACTIVITY \
`
`<§> Roger Crossley
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 1
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 1
`
`

`

`CHIRALITY
`and the
`BIOLOGICAL
`ACTIVITY
`of DRUGS
`Roger ~rossley
`----
`
`CRC Press
`Boca Raton New York London Tokyo
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 2
`
`

`

`-1/l J3
`RS
`Lf ;)__ q
`.c 1 (;i
`l: ') ;·~~ -c:
`
`'
`
`I
`
`I
`
`.......
`
`Library of Congress Cataloging-in-Publication Data
`
`Crossley, Roger, 1947-
`Chirality and the biological activity of drugs/Roger Cros!iley.
`p.
`em. -
`(New directions in organic and biological
`chemistry)
`Includes bibliographical references and index.
`ISBN 0-8493-9140-7 (alk. paper)
`I. Chiral drugs. 2. Drug receptors.
`RS429.C76 1995
`6 I 5'. I 9-dc20
`
`I. Tille. H. Series.
`
`95-21285
`C!P
`
`This book contains information obtained from authentic and highly regarded sources. Reprinted
`material is quoted with pennission, and sources are indicated. A wide variety of references are listed.
`Reasonable effons have been made to publish reliable data and infonnation, but the author and the
`publisher cannot assume responsibility for the validity of all material!> or for the consequences of their use.
`Neither this book nor any pan may be reproduced or transmitted in any fonn or by any means,
`electronic or mechanical. including photocopying, microfilmlng. and recording, or by any information
`storage or retrieval system. without prior pennission in writing from the publisher.
`CRC Press, foe. 's consent does not extend ro copying for general distribution, for promotion. for
`creating new works. or for resale. Specif1c permission must be obtained in writing from CRC Press for
`such copying.
`Direct all inquiries to CRC Press, Inc., 2000 Corporate Blvd., N.W., Boca Raton, Florida 33431.
`
`© 1995 by CRC Press, Inc.
`
`No claim to original U.S. Government works
`International Standard Book Number 0-8493-9140-7
`Library of Congress Card Number 95-21285
`Printed in the United States of America I 2 3 4 5 6 7 8 9 0
`Printed on acid-free paper
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 3
`
`

`

`36
`
`CHIRALITY AND THE BIOLOGICAL ACTIVITY OF DRUGS
`
`optimization of a series of enantiomer pairs by the comparison of the eutomers and
`distomers in a series of analogs. The ratio of the affinities or potencies of two
`enantiomers is called the eudismic ratio (ER) and the logarithm of this is termed the
`eudismic index (El). These terms are analogous to stereospecific ratio and stereo·
`specific index which are used in a similar way. Because the free energy of binding
`is related to the logarithm of the concentration required to produce a half-maximal
`effect, so the EI can be a direct measure of the difference in free energy of binding
`between enantiomers. Pfeiffer's Rule indicates that the EI should increase as the
`affinity for the eutomer increases and indeed, this is found to be the case in many
`series of enantiomeric pairs (but not all) which have been studied. 12•13•21 ·29
`If, in a series of homologous compounds, a plot is made of the El against the
`logarithm of the potency of the eutomer, a straight line is obtained, the slope of which
`is the rate of change in the El with affinity and is called the eudismic activity
`quotient (EAQ); (Figure 2.6). A positive slope (positive EAQ) provides a validation
`of Pfeiffer's Rule and is found in most of the cases where there is a correlation. Such
`a correlation was obtained in about 60% of datasets examined in a study of over a
`hundred series and accounted for 69% of the data points-" Lack of a correlation
`probably indicates a lack of involvement of the element of chirality in the binding
`process, but it could also be due to other competing factors in a complex system or
`to contamination with small amounts of opposite enantiomers. In general, an EAQ
`of 0.5 is predicted on theoretical grounds 12 to represent an optimal involvement of the
`element of chirality and this is close to the mean (0.4) found in the large dataset
`above.
`On occasions, a slope of zero (EAQ = 0) is observed and this can be used to make
`deductions as to the mode of binding. For example, the EAQ of a series of
`isopropylphosphothionates was found to be 0.62 for acetylcholinesterase inhibition
`but 0 for butyrylcholinesterase inhibition.~' The interpretation 12 being that at least a
`three-point interaction (cf. Easson-Steadman hypothesis) is involved in the first case
`and only a two-point interaction in the second case. It is, however, probably more
`correct to discuss the relative involvement of the element of chirality in the binding
`
`(log)Eu) -log[Dis))
`
`~slope=EAQ
`
`log[EuJ-
`
`FIGURE 2.6 Eudismic analysis.
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 4
`
`

`

`THEORETICAL ASPECTS OF CHIRALITY
`
`37
`
`process and to consider whether a modification to the structure could provide a
`greater level of involvement.
`On rare occasions, a negative EAQ is obtained which, although a violation of the
`Pfeiffer Rule, can also be instructive. One particularly interesting example" is found
`with 32 pairs of auxin analogs comprising sets of aryloxycarboxylic acids,
`~-naphthyloxycarboxylic acids and arylpropionic acids. The arylpropionic acids
`were found to have an EAQ of -D.37, which implies that the stereospecificity
`decreases with an increase in affinity and notably the intercept of the line with the
`abscissa occurs exactly at the point corresponding to the natural plant hormone
`indoleacetic acid." Presumably the system has developed to maximize interactions
`with this achiral ligand, and this may therefore implicate a symmetry in its interac(cid:173)
`tions which are disrupted by chiralligands. These series also provide exemplification
`of another use in drug design of eudismic analysis. The EAQ of the aryloxypropionic
`acids and the ~-naphthyloxypropionic acids taken together as a single series is 0.86
`(r2 = 0.77), but a better correlation is obtained for the subset ofP-naphthyloxypropionic
`acids on their own (EAQ = 1.07, r2 = 0.98). This may or may not be relevant in this
`case as there is a large substituent variation across the whole series and such
`differences can be an indication of a change in mode of binding. It is unlikely that
`different modes of binding to a particular receptor would produce the same EAQ so
`such correlations can be used to divide apparently homologous series into more
`optimal groupings for SAR studies.
`It is also possible, by comparing the eudismic correlations for the same series of
`compounds with different receptor types, to discern the relative involvement of the
`element of chirality with each receptor-binding process. For example, the antihistaminic
`and anticholinergic activities of a series of diphenhydramine derivatives have signifi(cid:173)
`cantly different EAQs (0.52 and 0.76, respectively) and there is also a different rank
`order to the relationships. JJ This indicates that the mode of interaction with each
`receptor is different and further analysis using standard molecular modeling tech(cid:173)
`niques could, therefore, be used to advantage. If the correlations were identical there
`arguably would be little point in trying to optimize this particular series.
`Eudismic-activity correlations can be used to determine the criticality of various
`chiral centers in a binding process and thus lead on to a greater appreciation of the
`mode of binding in a particular series. In other words, the more an element of
`chirality can be seen to influence the activity, the more it is central to that activity and
`implicated in the pharmacophore. One example is the muscarinic agonist potential of
`a series of pyrrolidinones (Figure 2.7, where R' = methyl, propyl and R" = methyl)
`which display a reasonable correlation when a traditional EA is carried out on
`enantiomeric pairs." With two chiral centers it is also possible to study the effect of
`epimerization ofR' orR" separately for this series of compounds. When the mydriatic
`activity was measured there was a good correlation with epimers of R' with EAQ =
`3.8 showing a strong effect of the chirality at this center. For the epimers of R" there
`was also a good correlation but here the EAQ = 0 showed that there was no
`dependence at this site. The obvious conclusion, therefore, is that the principal
`interactions with the receptor are on the right-hand half of the molecule, around the
`basic nitrogen, and that the pyrrolidone is less important as the primary determinant
`
`-·--
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 5
`
`

`

`38
`
`CHIRALITY AND THE BIOLOGICAL ACTIVITY OF DRUGS
`
`-OY10rt'18t
`c
`do&tom•r
`
`"
`
`"
`o+-----~----.-----~
`
`'
`
`log[Eu], Iog[Dis]
`
`(a)
`
`0
`
`0
`
`o+----4----.----,
`'
`
`Cl eU\omer
`
`(b)
`
`"
`
`C2 CUIOmcr
`
`'
`
`(c)
`
`FIGURE 2.7 Eudismic analysis of muscarinic receptor antagonists with two chiral centers provides a
`good overall correlation far the eutomers (a) which. can be separated into the effects of epimcrization at
`Cl (b) and C2 (c) showing that most of the recognition at the receptor centers around the Cl carbon.
`(Redrawn from Lehmann, F .. P. A. Trends Pharmacol. Sci. 3: 103-106, 1982. With permission.)
`
`of specificity. The results from a study of the anti tremor activity of these compounds
`also supports this interpretation, although the dependence on stereochemistry is less
`pronounced.
`
`2.5 CHIRALITY COEFFICIENTS
`
`The technique of eudismic analysis provides a description of the increase in
`potency of a molecule as the affinity of the enantiomers with a receptor increases, and
`can therefore be used to analyze a series of compounds, and provide information for
`a molecule design exercise. One limitation though is that the ER is not linked to the
`properties of the actual molecules. In an attempt to overcome this limitation the
`concept of molecular similarity has been applied to enantiomers with the definition
`of a chirality coefficient. Chirality was originally conceived and defined as an
`absolute property of molecules. Nevertheless, it is intuitive to think of degrees of
`chirality and this then begs the question of whether it is a continuous property and
`quantifiable. One attempt to do this has been to define a chirality coefficient30 which
`is equal to 1 if two enantiomers are totally dissimilar and to 0 if the enantiomers are
`identical in the property space under consideration. In this respect it can be seen as
`being equal to ( 1 -molecular similarity) and molecular similarity has been defined
`in several ways. 30·33 The two properties which have the most influence on chiral
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 6
`
`

`

`THEORETICAL ASPECTS OF CHIRALITY
`
`39
`
`recognition and binding processes are the electrostatic potential and the molecular
`shape. The electrostatic similarity of two molecules can be described in terms of the
`Carbo index"·" RAs and the shape similarity in terms of the Meyer index" SAs. The
`electrostatic and shape chirality coefficients are therefore I - RA8 and I - SAB•
`respectively.
`
`(2.6)
`
`The Carbo index RA8 is a function of the electrostatic potentials p of two aligned
`molecules A and B and the Meyer Index S AB is a function of the common volume C
`of the molecules and the volume of each molecule T expressed as numbers of grid
`points.
`A crude approximation to the overall degree of chirality of a molecule can be
`obtained by averaging the electrostatic and shape coefficients, and if this is plotted
`against the ER for the series of phosphonothiolates used in the eudismic analysis
`above a similar correlation is obtained.30 It should be noted that in this example the
`ER rather than the EI was used and the latter would probably have been a better
`correlation between degree of chirality and the free energy of interaction. Neverthe(cid:173)
`less, because the chirality coefficients are calculable properties of the molecules,
`these correlations can then be predictive for molecules that have not been prepared.
`
`Jjp~SR
`Me
`
`O,N
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`a
`·~
`
`"' ·"
`~
`'0
`"
`Ul
`
`0
`0.2
`
`"" t .··
`o···
`.··
`t .···
`q/
`.··
`
`--€1-- BuChE
`... .., ... AChE
`
`.·
`••
`{. ........
`.. ··
`
`.·· 0
`./
`.. ·
`.··
`
`Mo
`I
`f
`
`0.22 0.24 0.26 0.28
`Chirality Coefficient
`
`0.3
`
`0.32
`
`FIGURE 2.8 Linear correlations in a series of phosphonothiolates, R =Me, Et, Pr, Bu, against acetyl(cid:173)
`<::htJlinesterase and butyrylcholinesterase show lhe nature of the recognition processes to be different and
`sterem.elective for the former but not the Iauer. (Reprinted from Seri-Levy, A. and Richards, W. G.,
`Tcrrahedmn Asymm. 4: 1917-1923, 1993 with kind permission from Elsevier Science Ltd.,c The Boule(cid:173)
`vard. Langford Lane. Kidlington OX5 I GB, U.K.
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 7
`
`

`

`40
`
`CHIRALITY AND THE BIOLOGICAL ACTIVITY OF DRUGS
`
`1
`
`In addition, the correlations achieved in this example confirm that the drug-receptor
`interaction involves the most stable conformations of the molecules, as those were
`used in determining the coefficients. They can, therefore, provide information on the
`actual conformations which molecules adopt in their receptor-bound conformation.
`
`2.6 MULTIPLE EUDISMIC ACTIVITY CORRELATIONS
`
`The normal use of eudismic analysis is to take a series of enantiomeric pairs and
`obtain correlations against the activity of a single receptor type and sometimes to
`extend this correlation of the series to another receptor, as in the case of the
`diphenhydramine derivatives above. An alternative method is to take a single pair of
`well-characterized enantiomers and to compare their EI against the logarithm of the
`activity of the eutomer (log[Eu)) over a range of receptor types. This so-called
`multiple eudismic activity correlation (MEA C) can be used to quantify the activi(cid:173)
`ties of promiscuous drugs and can, analogously to EAQ, generate multiple eudismic
`affinity quotients (MEAQ) values. 36 The exact relevance of these values is difficult
`to ascertain, but they do seem to reflect some function of the binding process for
`drugs and, over a large series, the MEAQ were mostly found to be between 0.3 and 0.7.
`An important aspect of the use of multiple eudismic correlations is in the
`pharmacological characterization of receptors. For example, the correlation of EI
`against the activity p!C50(Eu) of propranolol enantiomers at serotonin receptors37 (5-
`HT2, 5-HT18, and 5-HT, from rat frontal cortex) has been used to differentiate them
`(Figure 2.9) with a single set of experiments.36 Normally, such a differentiation
`would require the testing of a wide range of more or less specific ligands. Following
`on from the characterization in this way such correlations can then be used to
`determine whether a particular effect measured in vivo is likely to be due to the
`activity of such a promiscuous ligand at a particular receptor type. The differential
`stereoselectivity of the methotrimeprazine (a combined analgesic and neuroleptic)
`
`1.6
`
`1.4
`
`1.2
`
`0.8
`
`0.6
`
`w
`
`5-HT1
`
`0
`
`0
`
`S-HT1A
`
`0.4
`
`0
`
`5-HT2
`
`0.2
`
`5
`
`5.5
`
`6
`pJC50(Eu)
`
`6.5
`
`7
`
`FIGURE 2.9 The MEAC of propranolol enantiomers against 5-HT receptor subtypes. (Redrawn from
`Lehmann, F., P. A .. Quant. Struc:r. Act. Relat. 6: 57-65, L987. With permission.)
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 8
`
`

`

`THEORETICAL ASPECTS OF CHIRALITY
`
`41
`
`enantiomers over six receptor types produced an excellent correlation (r2 = 0.96) and
`was used to show that it does not produce analgesia by a direct action on opiate
`receptors. 38
`
`2.7 ENANTIOMERIC PURITY
`
`Apart from the normal variability of the process of bioassay, one factor that has not
`been well controlled in the past is enantiomeric purity. Part of the problem has been
`the inability to determine this with any accuracy, especially in earlier work where an
`optical rotation has been the only guide. The variation of the [a] 0 values with the
`conditions of measurement has already been discussed and more recently developed
`techniques such as chiral NMR and especially chiral HPLC should be used wherever
`possible. [a] 0 is a poor measure of optical purity when the enantiomeric contamina(cid:173)
`tion is less than a few percent, although the estimate of purity can be improved by
`measurement at several frequencies. It is also less well appreciated how the degree
`of resolution affects the apparent stereospecificity of enantiomers, especially at high
`values of the ER.'9
`With two pairs of enantiomers the observed affinity constant of the eutomer is
`given by
`
`K *-y ·K +(1-y )K
`eu
`-
`eu
`~ ills
`
`~
`
`(2.7)
`
`where K'" is the true affinity constant for the eutomer, and K,;, is the true affinity
`constant for the distomer, and y'" is the fraction of the eutomer present. A similar
`equation can be derived for the distomer and the apparent eudismic ratio ER* can be
`expressed as
`
`E * y
`· K + (1- y )K
`ER*=~= eu
`eu
`cu
`dis
`Kdis Y dis' Kdis + ( 1- Y diJKeu
`
`(2.8)
`
`At high ERs K'" li> K,;, and at reasonable purity, the apparent affinity constant
`of the eutomer tends to y'".K'" because the term (I - y'")K,;, becomes neg!ip,ible.
`Rearranging the equation gives
`
`(2.9)
`
`and, assuming the degree of resolution is approximately the same for the eutomer and
`distomer (y = y'" = y ,;,), the equation further reduces to
`
`(1-y)
`I
`I
`--~-+--
`ER* ER
`y
`
`(2.10)
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 9
`
`

`

`42
`
`CHIRALITY AND THE BIOLOGICAL ACTIVITY OF DRUGS
`
`sao
`
`EA=-inlinire
`
`ER..300
`
`Degree of Resolution (y)
`
`FIGURE 2.10 The relation between the apparent ER and the true ER is highly dependent in a nonlinear
`manner on the degree of resolution. (Redrawn from Barlow et al., 1. Pharm. Pharmacal. 24: 753-761,
`1972. With permission.)
`
`and the function has a form plotted in Figure 2.10 from which it can be seen that there
`is great sensitivity of the apparent ER to the degree of resolution. Indeed, if the true
`stereospecific index was infinity, corresponding to inactivity of the distomer, and the
`degree of resolution was 95%, probably the limit of detection by optical rotation for
`a weakly rotating compound, then the ER would be 19 (EI = 1.3). Likewise, an ER
`of 100 (El ~ 2) implies that the optical purity should be at least 99% to be meaningful
`and when ER ~ 1000 (El = 3), the optical purity should be 99.9%.40·"
`Failure to compensate for the optical purity may lead to a great variation in
`measured ERs. For example, the ERs for the effects of salbutamol in relaxing guinea
`pig tracheas have been variously reported as 70 and 300.41•43 A particularly striking
`illustration of the need to obtain the best possible optical purity is illustrated with the
`1)2-adrenoceptor agonist formoterol which has two chiral centers. An improvement in
`optical purity from 99 to 99.9% increases the ER for the relaxation of guinea pig
`tracheas from 50 to 850 for the RR/SS pair and eliminates the difference entirely for
`the RS/SR pair. 39 Indeed, an early report of the pharmacology in this tissue produced
`a ratio of only 14 which implies that the optical purity was perhaps 95 to 97%.44 The
`effects with formoterol are so striking that some ;n!eraction between the enantiomers
`may also be suspected, but the simplified theoretical treatment above does not allow
`for this kind of effect.
`
`2.8 USE OF CHIRALITY AS A TOOL IN DRUG DESIGN
`
`The discussion thus far has served to highlight several areas where drug development
`may be affected by the chirality of molecules. That problems exist cannot be denied,
`but every problem also presents an opportunity and chirality can also be used
`aggressively as a tool in drug design. The rest of this book contains many examples
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 10
`
`

`

`THEORETICAL ASPECTS OF CHIRALITY
`
`43
`
`OH
`
`~~~
`HO~ Me ~OMe
`
`NHCHO
`
`salbutamol
`
`fonnotcrol
`
`FIGURE 2.11 Salbutamol and formoterol have variable ERs reported in the literature because of a
`failure to compensale for chiral purity.
`
`where it has been used, sometimes unwittingly, to obtain better drugs. It has been
`suggested that there is an overemphasis on the quest for chirally pure drugs and even
`a hint that to do so may damage the worldwide pharmaceutical industry." On the
`contrary, application of chiral design principles can provide a competitive edge.
`The importance of chiral interaction is that it provides a tool to discover which
`parts of molecules are involved in primary receptor interactions. Armed with this
`knowledge it is then possible to use it in drug design to increase both selectivity and
`potency. Biological systems are very complicated and the final result of interaction
`with a drug may not be easily determined a priori. Pairwise testing of enantiomers
`can help to elucidate cause and effect relationships, as demonstrated with NK 1
`receptor antagonists (Chapter 4). Many receptor systems, for example, the muscar(cid:173)
`inic receptor (Chapter 4), are highly sensitive to chirality which can be used to obtain
`subtype specificity. In others, such as the ion channel receptors (Chapter 3), chirality
`can be used to distinguish between the same receptor in different states or, as with
`antiviral agents (Chapter 5), provide selectivity over organisms with different bio(cid:173)
`chemistry.
`Pfeiffer's Rule provides an opportunity to increase both selectivity and potency
`at the same time. As we have seen, the binding forces are highly cooperative, so the
`better the fit of the molecule the more the increase in the binding energy and, in an
`asymmetric world, this implies the use of chiral molecules. How to find that optimum
`structure is the next problem.
`Faced with any lead compound the first approach of a medicinal chemist is
`usually to consider as many ways as possible to prepare it. At the same time it is
`jud;c;ous to ask which of these ways are amenable to the use of chiral pool meterials
`or which naturally would enable a chiral separation step. Unless it is unavoidable and
`predictable, resolution by diastereomeric salt formation should not be used in a
`research phase but deferred until a development phase. The general synthetic design
`should come before deciding on a specific target and be as convergent as possible.
`It is undoubtedly possible to develop a good stereospecific synthesis to almost any
`specific target in time, but the aim at this stage should be to introduce versatility.
`Highly efficient stereospecific reactions are also good synthetic reactions but the
`converse is not the case, so it may be worth introducing a potentially stereospecific
`reaction and using it in both an achiral and chiral manner. The synthesis may then
`suggest alternative substituent patterns which would not be obvious with achiral
`chemistry.
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 11
`
`

`

`44
`
`CHIRALITY AND THE BIOLOGICAL ACTIVITY OF DRUGS
`
`1
`
`Once the chemistry has been fully cvalunted, a possibly pragmatic decision
`should be made as to how to introduce a chiral element into the molecule. If this can
`be readily introduced homochirally, then that should be done, otherwise, a racemate
`should be prepared and tested. If reasonably active, the racemate should be examined
`by chiral chromatography and if possible separated by that means, as this is a
`reasonably efficient way of producing small quantities of enantiomers. If the sepa(cid:173)
`ration looks to be difficult or if the compound is inactive, another racemic derivative
`should be produced. The aim at this stage should be to produce a few enantiomeric
`pairs as quickly and efficiently as possible so as to discover whether a particular
`element of chirality is critical to the receptor interaction. If it is not, then the process
`should be repeated with the aim of discovering where on the molecule it is possible
`to put groups that will produce chiral interactions.
`The benefit of this approach is that three-dimensional information is available at
`an early stage to feed back into a parallel modeling exercise and also that the data on
`single enantiomers is more reliable and easily used in pharmacophore definition or
`molecular field analysis. At the same time eudismic analysis enables a rapid quan(cid:173)
`tification of selectivity and affinity. Once the approximate requirements for the chiral
`molecule have been established, they can then be refined by substituent variation and,
`again, eudismic analysis gives some indication when an optimum ER has been
`reached. It is because the data on single enantiomers are more reliable and more
`easily interpretable that the process should be more efficient. At the same time many
`of the "problems" of stereochemistry which confront more traditional approaches
`have been tackled at an early stage. There is also a better chance of developing second
`generation or back-up compounds by finding extra interactions and thus extending
`the element of chirality. In this respect it more closely mimics the approach of nature
`in the development of highly chiral and specific ligands.
`One good example of how to produce high potencies and selectivities by using
`chirality as a tool rather than by treating it as a problem, albeit more fortuitously in
`the early stages and with some diversions on the way, is the development of some
`ion channel facilitators (Chapter 3) from Rhone-Poulenc Rorer. The association of
`some cardiovascular activity with an early series of anti-ulcer compounds related to
`picartamide led to the development of the potassium channel opener aprikalim the
`active enantiomer of RP-49356. Aprikalim had an IC90. against the relaxation of rat
`aorta induced by 20 mM potassium chloride, of 0.4 ).I.M and the distomer was
`essentially inactive. In a search for analogs, the sulfoxide was replaced by carbonyl
`and chemistry developed using (+)-R-1-phenylcthylamine as a chiral auxiliary
`(Figure 2. 12) to enable an efficient synthesis of analogs. The cyclohexanones again
`had the same R-configuration at the 2-position and were approximately equipotent
`with aprikalim.46.47 Further work with oximes and other derivatives indicated that
`another lipophilic binding site could be found by extending a chain from the !-position
`of the cyclohexanone. The natural extension to this work was then stereospecifically
`to reduce the imine in (I) to provide the amine (2) with three chiral centers,
`introduced stereospecifically, which reduces the IC90 down to a remarkable
`0.00003 ).I.M.
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 12
`
`

`

`THEORETICAL ASPECTS OF CHIRALITY
`
`45
`
`(\csNHMe
`..... ,AI ~
`so=-'_
`[!._ __ _)
`N
`
`IC90 = 0.4~M
`
`picartamide
`
`aprikalim
`
`I. BuLi
`2. Mc:NCS
`
`~YYv
`
`Ph
`
`IC90 = 0.00003~M
`
`FIGURE 2.12 Development of a series of potassium channel facilitators.
`
`2.9 CHIRALITY AND RECEPTORS
`
`Biological receptors come in many shapes and sizes. Some, namely, the ion
`channel and G-protein-coupled receptors modulate the transmission of nervous
`impulses by interacting with the signal or the neurotransmitter. Others may be
`involved in translocation processes and control the passage of ions and organic
`molecules across cell membranes or, as enzymes, participate in metabolic or cata(cid:173)
`bolic processes. In addition, drugs not only interact with well-classified receptors but
`also interact with membranes themselves and with other proteins which also may be
`treated as receptors. The localization of the receptors determines their shape and
`ability to interact with drug molecules, and often receptors within these classes fall
`into superfamilies of proteins, related by sequence homology. There are great simi·
`larities in families in their interactions with drug molecules, their localization, and in
`their transduction mechanisms. Many drugs owe their therapeutic effect to the
`interaction with more than one receptor. When it is considered that the eventual
`usefulness of a drug is largely governed by the rate of onset and duration of action,
`which involve distribution and metabolic processes, selectivity between receptors is
`as important a consideration as activity.
`Interactions with all these systems involve stereospecific processes, some of
`which have been extensively studied. If enantiomers discriminate between particular
`types of interaction with one receptor system then there will usually be found similar
`
`DR. REDDY’S LABS., INC. EX. 1066 PAGE 13
`
`