BIOMEDICAL LIBRARY

_idIIIM 80

LO

MYLAN - EXHIBIT 1025

____ Chemical Society Reviews

Vol No 1979

PageTILDEN LECTURE

Concerning Stereochemjcal Choice in Enzymic Reactions

By Overton 447

The Acidity of Solid Surfaces and its Determination

by Amine Titration and Adsorption of Coloured

Indicators

By Atkinson and Curthoys 475

Non-isoprenoid Long Chain Phenols

By Tyman 499

Some New N.M.R Methods for Tracing the Fate of

Hydrogen in Biosynthesis

By Garson and Staunton 539

Isosterism and Molecular Modification in Drug Design

By Thornber563

1979 Indexes581

The Chemical Society London

Chemical Society Reviews

Chemical Society Reviews appears quarterly and comprises approximately 25articles ca 500 pp per annum It is intended that each review article shall be ofinterest to chemists in general and not merely to those with specialist interest

in the subject under review The articles range over the whole of chemistry andits interfaces with other disciplines

Although the majority of articles are intended to be specially commissionedthe Society is always prepared to consider offers of articles for publication Insuch cases short synopsis rather than the completed article should be submitted to The Managing Editor Books and Reviews Section The Chemical

Society Burlington House Piccadilly London W1V OBNMembers of the Chemical Society may subscribe to Chemical Society Reviews

at 6.00 per annum beginning 1980 7.50 per annum they should place their

orders on their Annual Subscription renewal forms in the usual wayNon-members may order Chemical Society Reviews for 16.00 $33 be

ginning 1980 22.00 $50 per annum remittance with order fromThe Publications Sales Officer The Chemical Society Blackhorse Road

Letchworth Herts SG6 1HN England

Note to subscribers We regret the delays in publication of the 1979 issues

CS Reviews The main causes were the long-term effects of shortages in submitted articles and production difficulties during earlier years Steps have beentaken to rectify this distressing situation and the intention is to revert to the

usual quarterly publication dates by the end of 1980 or the beginning of 1981

Copyright reserved by The Chemical Society 1980

ISSN 03060012

Published by The Chemical Society Burlington House London WIV OBNPrinted in England by Eyre Spottiswoode Ltd Thanet Press Margate

Isosterism and Molecular Modffication in Drug Design

By Thornber

IMPERIAL CHEMICAL INDUSTRIES LIMITED PHARMACEUTICALS

DIVISION MERESIDE ALDERLEY PARK MACCLESFIELDCHESHIRE SK1O 4TG

Introduction

The idea of isosterism goes back to Langmuir1 in 1919 At that time the word

isosterism was used to describe the similarity of molecules or ions which have the

same number of atoms and valence electrons e.g 02 Ne Cleariy only

those isosteres with the same nett charge show similar chemical and physical

properties Grimm2 enunciated his hydride displacement law to describe the

similarity between groups which have the same number of valence electrons but

different numbers of atoms For example some similarities are present in the

sequence CH3 NH2 OH Hal

Grimms hydride displacement law points out some similarities of size in

groupings based on elements in the same row of the periodic table Other similar

ities to be found in the periodic table are within the groups where chemical

reactivities are similar but with electronegativity decreasing as atomic weight

increases and lipophilicity and polarizability increasing with the size of the

atom Other relationships exist in diagonal lines across the periodic table where

atoms of similar electronegativity such as nitrogen and sulphur oxygen and

chlorine are found

In trying to relate biological properties to the physical and chemical properties

of atoms groups or molecules many physical and chemical parameters may be

involved and the simple relationships mentioned above are clearly inadequate

for this purpose Friedman3 introduced the term bioisosterism to describe the

phenomenon in which compounds which are related in structure have similaror

antagonistic properties The use of the word isosterism has clearly outgrown its

original meaning when used in medicinal chemistry and loose flexible definition

could be adopted such as Bioisosteres are groups or molecules which have

chemical and physical similarities producing broadly similar biological properties

The term non-classical isosterism is also used interchangeably with bioisoster

ism particularly in connection with isosteres which do not have the same numberof atoms but do produce similarity in some key parameter of importance in

Langmuir Amer Chem Soc 1919 41 868 1543Grimm Elektrochem 1925 31 474 1928 34 430 1934 47 53 594Friedman Influence of Isosteric Replacements upon Biological Activity National

Academy of SciencesNational Research Council Publication No 206 WashingtonD.C 1951 295

563

isosterism and Molecular Modfication in Drug Design

that series For example4 the two /3-adrenergic stimulants compounds andhave similar activity

CHOHCH2NHMe Cl-IOHCH2NHMe

HO3MeSO2N

pKa 9.6pKa 9.1

The concept of bioisosterism has been described in reviews by Burgersa

Schatz5b Foye6 Korolkovas7 Ariens8 and Hansch.9 This present review

collates and extends the earlier observations with more recent reports from the

literature and suggests new techniques for exploiting the conceptThe classical isosteres as defined by Burger5 and Korolkovas7 are given in

Table

Table

Univalent atoms and groups

OH NH2 Me Cl

SH PH2But

Br Pr1

Bivalent atoms and groups

Se CH2CO2R COSR COCH2R CONHR

Tervalent atoms and groups CHAsQuadrivalent atoms

C- -Si

Ring equivalentsCHCH e.g benzene thiophen

N.- e.g benzene pyridine

S---- CH2 NHLarson and Lish Nature 1964 203 1283

Burger in Medicinal Chemistry 3rd Edn ed Burger Wiley-Interscience New York1970

5b Schatz in Medicinal Chemistry 2nd Edn ed Burger Wiley-Interscience NewYork 1960

Foye Principles of Medicinal Chemistry Lea and Febiger Philadelphia 1970Korolkovas Essentials of Molecular Pharmacology Background for Drug Design

Wiley 1970Ariens in Drug Design ed Ariens Academic Press New York 1971 Vol

Hanscb Jntra-Sclcnce Chem Rep 1974 17

564

Thornber

Bioisosterism in Molecular Modification

In the process of developing lead compound an antagonist to knownagonist or an anti-metabolite from known substrate large number of

systematic molecular modifications will be made The modern concept of

bioisosterism can be an aid to the design of such modifications In makingbioisosteric replacement the following parameters of the group being changedcould be considered

Size

Shape bond angles hybridization

Electronic distribution polarizability inductive effects charge dipoles

Lipid solubility

Water solubility

pKa

Chemical reactivity including likeithood of metabolism

Hydrogen bonding capacity

It is unlikely that any bioisosteric replacement will leave all these parametersundisturbed The extent to which the replacement is useful will depend uponwhich of these parameters is important and which ones the bioisostere can best

mimic

The element of molecule being modffied may have one or more of the follow

ing roles

Structural If the moiety has structural role in holding other function

alities in particular geometry parameters such as size and bond angle will be

important The moiety may be buried deep in the molecule and have little

contact with the external medium

ii Receptor interactions If the moiety to be replaced is concerned with

specific interaction with receptor or enzyme its size shape electronic

properties pICa chemical reactivity and hydrogen bonding will be the

important parameters

iii Pharmacokinetics The moiety to be replaced may be necessary for the

absorption transport and excretion of the compound In this case lipophili

city hydrophilicity hydrogen bonding and pKa are likely to be important

iv Metabolism The moiety may be involved in blocking or aiding metabolism In this case chemical reactivity will be an important parameter Forexample chloro and methyl substituents on benzene ring may be inter

changeable for certain purposes but the toluene derivative can be metabolized

to benzoic acid and may therefore have shorter half-life or unexpectedside effects

Usually one will not know which roles the various parts of the molecule

plays in its action and this determination will be part of the structureactivity

study However from the simple considerations listed above it is clear that

565

Isosterism and Molecular Modification in Drug Design

given molecular modification may allow some but probably not all of

the parameters ah to be kept the same

Whether the same or different biological activity results from the replace

ment will be governed by the roles which that moiety fulfils in the molecule

and whether parameters affecting that role have been disturbed

From and it follows that what proves to be good bioisosteric

replacement in one series of compounds will not necessarily be useful in

another

Completely identical properties are rarely sought and will in any case be

difficult if not impossible to achieve What we are more likely to be seeking is

subtle change in the molecule which will leave some properties the same and

some different in order to improve potency selectivity absorption duration and

toxicity Bioisosteric replacements allow molecular modifications in which the

number of variables changed are limited Ariens8 and Korolkovas7 have tried to

introduce the idea of partial bioisosteric groups as those which turn an agonist

into an antagonist Although their lists of groups may be suggestive to the drug

designer the idea is probably incorrect because of the statement above Anantagonist group in one molecule will only antagonize similaragonist group

in another molecule if the agonist groups in both series are performing the samefunction If an isosteric replacement results in molecule which has some

properties similar to the parent molecule but some important property has

changed it may be possible to compensate for this undesirable change by modifi

cations elsewhere in the molecule For example molecular modification mayreduce the lipid solubility of the molecule thereby affecting its absorption

transport and apparent potency Optimum activity may be regained by inserting

lipophilic groups into the molecule at some sterically undemanding site Consequently the best compounds in this parallel series of isosteres such as for

example furans and thiophens are likely to have different substituent patterns

The Mathematical Formulation

The arguments used above can be expressed in the mathematical form used byHansch for the case where simple substituent is being varied for example onbenzene ring If the potency of drug is function of several parameters of the

substituent then

log A7r Bor CE8

where Hanschs ir value is used for the lipophilic character Hammetts value

for the electronic property Tafts steric parameter to denote the size of the

group and is the concentration of drug required to achieve given effect

If such relationship were found for drug series in which the constants

and were zero then the potency would be function of IT only In this context

groups would be bioisosteric if they have similar IT values independent of their

10 Hansch Accounts Chem Res 1969 232

566

Thornber

and values If however the three constants and are all significantmuch more limited range of equivalent groups will be available

If series of compounds has more than one property as is usual then morethan one equation will be needed to describe the effects of changing the substituent

log Air Bar CE8

Desired activity

log Dir Ea FESide effects

Clearly if and etc no selectivity can be found withinthis limited series If however then for the desired activity E8 is not important and ir and may be optimized while reducing the value of therebyreducing the side effects This phenomenon of increasing selectivity by bioisosteric replacement relies upon the fact that some desirable properties in the

molecule can be retained when unimportant parameters can be varied An unimportant parameter for the biological activity desired may be key parameter in

the side effect

Thus bioisosteric replacements are useful in searching for potency selectivity

absorption and duration Following the Hansch treatment one could producemodem definition of bioisosterism based upon measurable parameters such as

ir E8 hydrogen bonding properties pK etc and Hansch9 has used the term

isolipophilic for groups with the same value

Table shows some functional groups with similar electron-withdrawingproperties If electronic effects alone influence the biological activity in series ofdrugs then these groups would be equivalent If however the lipophilicity andsteric factors are important then absolute identity cannot be achieved

Table

Functional Group IT Es

0.34 0.14 0.78

Cl 0.37 0.71 9.27

Br 0.39 0.86 0.08

0.35 1.12 0.16CF3 0.43 0.88 1.16SCF 0.40 1.44

COMe 0.31 0.55CHO 0.36 0.65CO2Me 0.32 0.01CHCHN02 0.32 0.11

567

Isosterism and Molecular Modification in Drug Design

Extensive tables of IT and E5 values are now available These can be usedto gain more quantitative idea of some aspects of isosterism using the better

known functional groups

Chemical Reactivity

Biological effects are generally produced by weak interactions between the drugand the receptor but covalent bonding does occasionally play part series of

aspirin isosteres was reported in 1975.12 The nitrogen sulphur and carbon

MeCOX

3x NHor Cl-i2

isosteres were all totally inactive despite the classical purity of the replacementstried Now that it is known that aspirin is an acetylating agent for prostaglandinsynthetase this result is more readily understood.3 The agents are widely differ

ent in their ability to act as acylating agents unless other substantial modifications are made in the molecules

Non-classical Isosteres Some Further Points

In considering bioisosterism in its widest sense it should be noted that similar

effects in two functional groups need not imply atom upon atom overlapEdwards4 has pointed out that common enzyme or receptor interaction involves hydrogen bonding to carbonyl group Strong hydrogen bonds may beformed to the carbonyl oxygen by hydrogen atoms within cone having an angleof about 60 at its apex Two molecules RXH and RAXH where is an additional atom may be able to bind to the active site without identical positioningof the or In addition the conformational mobility in both the drug and the

receptor molecule will allow essentially similarbinding of two drugs without the

need to consider that the binding groups on the drugs are positioned in spacein an identical manner

Examples of Non-classical Isosteres.The list shown in Table is drawn fromearlier reviews59 and from the examples given in Table at the end of this

1Tables of substituent constants can be found in the following papers HanschRockwell Jow Leo and Steller Med Chem 1977 20 304Topliss Med Chem 1972 15 1006 and 1977 20463 Hansch Leo UngerKi Hwan Kim Nikaitoni and Lien Med Chem 1973 16 1207

12 Thompkins and Lee Pharm Sd 1975 64 760Roth Stanford and Majerus Proc Nat Acad Sd U.S.A 1975 72 3073Edwards IC. Pharmaceuticals Division personal communication

568

Thornber

review In addition few proposals5-7 which have not yet been realized in

medicinal chemical work are included

Table

arbonyl group

-SO2L

ref 15

Carboxylie acid group

CO2H SO2NHR SO3H POOHNH2 POOHOEt

CON HCN HO....15

ref 16

Ilydroxy-group

OH NHCOR NHSO2R CH0H NHCONH2

NHCN C-1CN2

ref 16 ref 16

Catechol

N20 HOO HOXXNR

Halogen

Halogen CF3 CN NCN CCN3 ref 16 17

Wallenfels Friedrich Rieser Ertel and Thieme Angew Chem Internat

Edn 1976 15 261

von Kohier Eichler and Salewski anorg Chem 1970 379 183 also includes

other possibilities in the sulphur and phosphorus and nitro acid series

von Wallenfels Chimia 1966 20 303

569

Isosterism and Molecular Modification in Drug Design

Table continued

Thioet herCN CN CN\/

/N\ /S\ /C\ref 16

Thiourea

NCN CHNO2Ii

IIIINHCNH2 NHCNH2 NHCNH2

CNAzomethine

ref 17

Pyridine

NO9

RNR3Spacer groups

CH3

In addition ring-opened forms of molecules may be considered to be isostericwith the corresponding ring-closed forms although the conformation of theseco form will be unlike the parent molecule However if in ring opening anatom is removed conformation similar to the parent molecule may be possible

Subsfructure Searching and Bioisosterism

Although the classical Hansch approach is used largely for optimization within

series molecular modifications based on bioisosterism principles can generatenew series or even develop new leads if an agonist is used as the starting point forthe design of an antagonist One aid to this process is the use of compoundcollection and computer techniques for doing substructure searches e.g the

570

Thornber

Crossbow suite of programmes.8 For example suppose that random screening

has turned up the lead One may consider bioisosteric replacements for

the ring the oxygen the polymethylene chain or the amidic moiety and design

substructure search for compounds of type vast number of permutations

are possible and from these compounds may be available for tests which result

in new leads which have properties worth exploiting such as perhaps

01OC1LNHCOMe

CHCH CHN or NR

SO SO2 Se NCN or NCOR

NHCONHMe n23or4or alkylincluding forming ring

Me COR CO2R SOR SO2R or CONHR

AB defined substituents

Examples.The literature of medicinal chemistry is rich in examples of the use

of the concept of bioisosterism and the reader is referred to the reviews mentioned5-8 and the references quoted therein for examples reported before 1970There follows brief discussion of bioisosteres of some indoleamines which has

some useful lessons and Table lists examples culled from the literature since

1970 Only the structures are given in this Table as an illustration of the kinds

of change which have been useful The reader is referred to the original papersfor the full details of biological activity and selectivity The list is not comprehensive but represents some uses of more novel non-classical types Rudinger9has reviewed isosteric replacements in the field of peptide chemistry up to 1971

and some further discussions2 have been published recently

Indole-amines..-Campaigne2 has studied and reviewed the work on bioisosteres

of 5-hydroxytryptamine and one or two details of the work are instructive

Whereas was inactive as an agonist or antagonist on the rat uterus preparation

the corresponding tryptophan analogue had weak activity as an enzymeinhibitor for 5-hydroxytryptamine decarboxylase.22 This type of bioisostere

Townsley and Warr Chemical and Biological Data An Integrated On-Line

Approach in Retrieval of Medicinal Chemical Information ed Howe Mime and Pennell

Symposium Series No 84 American Chemical Society Washington D.C1J Rudinger in ref Vol II Chapter20 Further discussion of peptide backbone replacement is found in ref 19 and Soudyn and

van Wijngaarden in Biological Activity and Chemical Structure ed KeverlingBuisman Elsevier Holland 1977 peptide link isostere CH2S has been reportedby Yankeelov Kam-Fook Fok and Carothers Org Chem 1978 43 1623

Campaigne Maichel and Bosin Medicinal Chemistry Specialist Contributions 3rd International Symposium 1972 Butterworths 1973 65

Pigini Gianella Gualtieri Meichiorne Bolle and Angelucci EuropeanMed Chem 1975 10 29 33

571

Isosterism and Molecular Modification in Drug Design

5-hydroxytryptamine

10 C021-1 5-hydroxytryptophan CO2H

loses all affinity for the 5-hydroxytryptamine 5-HT receptor but retains it in

part for an enzyme system Similarly in the series of compounds 5-HT 11 12and 13 activity has been measured against the rat fundic strip preparation andon the enzyme caeruloplasmin.23 Whereas 5-HT is substrate for the enzymecompound 11 inhibited caeruloplasmins oxidation of 5-HT and noradrenaline

Rat Fwzdic Strzp

Intrinsic PD2

activity

5HT0.96

ti. jj 12 0.84 4.6

Compound 12 inhibits only 5-HT oxidation and compound 13 was inactive as

substrate or an antagonist This would appear to demonstrate that for the

enzyme system the imino grouping at the 1-position of the ring is essential

On the rat fundic strip however all the analogues have full agonist activity

though with reduced potency demonstrating that the 5-HT receptor has

greater tolerance for loss of the imino nitrogen These simple experiments demonstrate the role of bioisosteric replacements in exploring selectivity betweendifferent receptors and enzymes

Barrass Goult Pinder and Sheds Blochem Pharmacol 1973 222891

572

Thornber

Table Some recent examples of bioisosterism

Dihydroxyphenylalanine analogues

xxTo2H H.co2HDopa Mimosine ref 24

HO/Z2 HO12ref 25 ref 26 ref 27

Histamine H-2 antagonists

IjI

MeCH7SCH2CHNNHMC

HNNor NCN ref 28

NHMeMe2NCH2

NCN or CHNO2 ref 29

24 Haguchi Mo Pharmacol 1977 13 36225 natural product from Streptomyces species Inoue Shamura Tsurvoka

Ogawa Watanabe Yoshidea and Nuda Chem and Pharm Bull Japan 197523 2669

26Synthesized as mimosine analogue Harris and Teitei AustraL Chem1977 30 649

27 Norton and Sanders Med Chem 1967 10 96128 Brimblecombe Duncan Durant Emmett Gannelin and

Parsons mt Med Res 1975 86 See also Suiphurmethylene isosterism in the

development of metiamide Black Durant Emmett and GannelinNature 1974 248 65 and Gannellin App Chem Biotechnol 1978 28 183

26 and Hanbury U.S.P 128 658

573

isosterism and Molecular Modification in Drug Design

Table continued

Neuroleptics

CH 23X

II or CHCN ref 30

Anthelmintics

PhS PhS

NHCO2Me NHCO2Me

ref 31

PhyXSorSe ref 32

fl-A drenergic blockers

R_ç--

ref 33

Boehringer Sohn U.S.P 085 216Fisher and Lusi Med Chem 1972 15 982 Bochis Dybas EskolaKulsa Linn Lusi Mutzner Milkowski Mrozik Olen

Peterson Tolman Wagner Waksmunski Egerton and

Osteind Med Chem 1978 21 23532 Hanson Giese Davis and Costello Med Chem 1978 21

49633T Jen Frazee Schwartz Erhard Kaiser Colella and

Wardell Med Chem 1977 20 1263

574

Thornber

$-Adrenergic stimulants

01-i OH

1_.NHR Cl..LNHBut

H01 H2NCl

Me 01-1 Adrenaline Clenbuterol ref 37

But CH2OHSalbutamol ref 34

But NHCONH2Carbuterol ref 35

Pr NHSO2MeSoterenol ref 36

NJ..NHPr1 HOX NHPr

ref 38 ref 39

HZIIL_

OH

ref 39Hk.1ç

ref 40

Hartley Jack Lunts and Ritchie Nature 1968 219 861 CollinHartley Jack Lunts Press Ritchie and ToonJ Med Chem

1970 13 674

Kaiser Med Chem 1974 17 4936A Larsen Gould Roth Corner Uloth Dungan and

Lish Med Chem 1967 10 462Keck Kruger Noll and Machleidt Arzneimittelforsch 1972 22 861

38Arnett Wright and Zenker Med Chem 1978 21 72

30Williams Canad .1 Chem 1976 54 3377

Yoshizakj K. Tarimura Tamada Yabuuchi and Nakagawa Med Chem1976 19 1138

575

Isosterism and Molecular Modification in Drug Design

Table continued

Vasodilators

-LXRor SO2 ref 41XO YCO ref.42XS YCO.ref.43

MeO2CJJUL

Me Me

CO2Me ref 44

SO2Me ref 45

Androgens OAc

SorNCN ref.46

SmithKline Corp U.S.P 117 12842 Vaughan Williams and Polster European Pharmacol 1974 25 241 Unlisted

Drugs 1971 23 110Claeys Goldenberg Wandestrick Devay Descamps Delaunois

Bauthier and Charlier Chim Ther 1972 377Bossert and Vater Naturwiss 1971 58 578 Drugs of Today 1975 11 154

Ciba-Geigy B.P 464 324W.-H Chiu Klein and Wolff Med Chem 1979 22 119

576

Thornber

Anti-inflammatory

MeOS%CH2_X

Cl N-NCO2H or ref 49

N-Nref

CO2H ref 48

Ornithine decarboxylase inhibitor

H2NCH23CHNN

NH2

ref 50

Gabergic agents

HO2C.NH2 HO Ji OH

HOS NH2 11NH2 H2Nref 53 ref 51 ref 52

Juby and Hudyma Med Chem 1969 12 396Shen Ellis Windholz Matzuk Rosegay Lucas WitzelStammer Wilson Holly Willet Sarett Holtz

Risley Nuss and Winter Amer Chem Soc 1963 85 488Drain Davy Horlington Howes Scruton and Seiway .1Pharm Pharmaco/ 1971 23 857

Bey Danzin van Dorsselaer Mamont Jung and Tardiff Med Chem1978 21 50

51Atkinson Giraud Rokach Rooney McFarlane Rackham andShare Med Chem 1979 22 99

52Curtis Duggan Felix and Johnston Brain Res 1971 32 69Curtis and Watkins Nature 1961 191 1010

577

Isosterism and Molecular Modification in Drug Design

Table continued

Prostaglandin ring system

L_ref 54 ref 54 ref 55 ref 55

II

MeN MeY

ref 56 ref 57 ref 58

MeN HNAN H2NAN_

Lre

.0 00

Me Meref 60 ref 60 ref 61

Zoretic Soja and Shiah Prostaglandins 1978 16 555Zoretic Soja and Shiah Med Chem 1978 21 1330

Harris Whittaker Higgs Armstrong and Reed Prostaglandins1978 16 773

Jones Holtz Bicking Cragoe Mandel and Kuehl MedChem 1977 20 1299

58 Bicking Robb Smith Cragoe Kuehi and MandelMed Chem 1977 20 35

Jones Hoitz Bicking Cragoe Mandel and KuehiMed Chem 1977 20 44

60 Smith Bicking Gould T.-J Lee Robb Kuehi Mandeland Cragoe Med Chem 1977 20 540

Eggelte de Koning and Huisman Rec Tray chim 1977 96 271

578

Thornber

HN1N HN1NL% L%ref 62 ref 63 ref 63

1LN RNJL

ref.64 ref 65ref 63

RNAN/

HOref 66 ref 67 ref 68

oa

HO HOref 69 ref 70 ref 71 ref 72

eaZoretic Branchard and Sirka Org Chem 1977 42 3201 Bruin

de Koning and Huisman Tetrahedron Letters 1975 4599 Bollinger andMuchowski Tetrahedron Letters 1975 2931Smith T.-J Lee Gould Cragoe Olen and Kuehi Med

Chem 1977 20 129264 Merck U.S.P 087 435

Beechams Belgian 861 95666 Beechams Belgian 861 957

Miles U.S.P 27 61268

Pfizer U.S.P 132 84769 Vlattas and Dellavecchja Tetrahedron Letters 1974 4459

Vlattas and Dellavecchia Tetrahedron Letters 1974 445571 Tanabe Seijaku G.P 229 225 Hauser and Huffman Tetrahedron Letters

1974 90572

Harrison Taylor and Fried Tetrahedron Letters 1975 1165

579

sosterism and Molecular Modification in Drug Design

Table continued

Prostaglandin ring system continued

Oyref 73 ref 74 ref 75 ref 76

Harrison Fletcher and Fried Tetrahedron Letters 1974 2733du Pont de Nemours B.P 428 431Harrison and Fletcher Tetrahedron Letters 1974 2729Bender Med Chem 1975 18 1094

580