Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Quantitative structure-anticonvulsant activity studies of valproic acid analogues Acheampong, Andrew Adu 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A1 A33.pdf [ 10.12MB ]
Metadata
JSON: 831-1.0096534.json
JSON-LD: 831-1.0096534-ld.json
RDF/XML (Pretty): 831-1.0096534-rdf.xml
RDF/JSON: 831-1.0096534-rdf.json
Turtle: 831-1.0096534-turtle.txt
N-Triples: 831-1.0096534-rdf-ntriples.txt
Original Record: 831-1.0096534-source.json
Full Text
831-1.0096534-fulltext.txt
Citation
831-1.0096534.ris

Full Text

QUANTITATIVE STRUCTURE-ANTICONVULSANT ACTIVITY STUDIES OF VALPROIC ACID ANALOGUES By ANDREW ADU ACHEAMPONG B.Sc, The University of Science and Tech., Ghana, 1978 M.Sc, The University of B r i t i s h Columbia, 1982 A THESIS SUBMITTED AS PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Faculty of Pharmaceutical Sciences) Division of Pharmaceutical Chemistry We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1985 © Andrew Adu Acheampong, 1985 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of f k c ^ ^ q . cg_vAJ> i^J^n i_g_J> The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date Q cfcMaJtv" CU f \ Q \ DE-6(3/81) ABSTRACT V a l p r o i c a c i d (2-propylpentanoic acid) i s an a n t i e p i l e p t i c drug widely used f o r treatment of absence s e i z u r e s . V a l p r o i c a c i d has a unique chemical s t r u c t u r e which does not contain the imide s t r u c t u r e found i n most conventional a n t i e p i l e p t i c drugs. An i n vivo study of the antagonism of pentylenetetrazol-induced c l o n i c s e i z u r e s by a l k y l -substituted carboxylic acids and tetrazoles was of interest owing to the known b i o i s o s t e r i s m between the carboxylic and the t e t r a z o l y l moiety. The main objective of t h i s study was to investigate the role played by the l i p o p h i l i c i t y , the electronic properties and the s t e r i c influence of compounds on the i r anticonvulsant potency. Quantitative stru c t u r e - a c t i v i t y relationships of the a l i p h a t i c and a l i c y c l i c substituted carboxylic acids and tetrazoles have been perform-ed using the Hansch linear free-energy relationships model. The study proceeded by synthesis of compounds using known procedures. The d i -unsaturated d e r i v a t i v e s of v a l p r o i c a c i d , 2-[(E)-l'-propenyl]-(E)-2-pentenoic a c i d and 2-[(Z)-l'-propenyl]-(E)-2-pentenoic a c i d were prepared v i a a s t e r e o s e l e c t i v e s y n t h e t i c route. The synthesized d i -unsaturated acids were used i n i d e n t i f i c a t i o n of the major diunsaturated metabolite of valproic acid as 2-[(E)-l'-propenyl]-(E)-2-pentenoic acid. The anticonvulsant potency of test compounds was determined i n mice (CD1 s t r a i n , 20-32g) by the standard subcutaneous p e n t y l e n e t e t r a z o l e seizure threshold test. The pentylenetetrazole clonic seizure test was found to be more s e n s i t i v e to s t r u c t u r a l e f f e c t s than the pentylene-tetrazole mortality assay. The l i p o p h i l i c i t y (octanol-water p a r t i t i o n c o e f f i c i e n t ) of compounds was determined i n d i r e c t l y by reversed phase l i q u i d chromatography employing an octadecylsilane column (Hypersil ODS) i i and mobile phase as 70% methanol : 30% phosphate buffer (pH 3.5). The electronic character of the compounds was monitored by the apparent acid i o n i z a t i o n constant obtained from potentiometric t i t r a t i o n i n 10% raethanol-water system. The ED5Q of 0.70 mmol/kg found for valproic acid was similar to l i t e r a t u r e values. 5-Heptyltetrazole was found to be the most potent compound i n the s e r i e s of analogues s t u d i e d . The c a r b o x y l i c plus tetrazole group gave a low correlation (r = 0.63) between the anticon-vulsant potency and a linear combination of l i p o p h i l i c i t y and apparent ionization constant. However, i n the series of active carboxylic acids, the anticonvulsant a c t i v i t y was noted to be s i g n i f i c a n t l y correlated with l i p o p h i l i c i t y and apparent ionization constant (r = 0.91). The usefulness of the electronic parameters, acid ionization con-stant and dipole moment, were explored i n an extensive set of a l k y l -substituted anticonvulsant compounds with different polar moieties. Addition of the dipole moment term to the l i p o p h i l i c i t y term led to s i g n i f i c a n t l y better correlations (r = 0.81) as compared to that with an added pKa term. The negative dependency of anticonvulsant a c t i v i t y on dipole moment supported previous findings in studies of 1,4-benzo-diazepines and phenyl-substituted anticonvulsant compounds. There were some exceptions to the dependence of anticonvulsant a c t i v i t y on l i p o p h i l i c i t y and d i p o l e moment or pKa. N,N-dibutyl-succinamic acid showed convulsant properties at sublethal doses. The lack of a c t i v i t y of cyclohexylacetic acid and 5-cyclohexylmethyltetra-zole, i n comparison to the active 1-methylcyclohexanecarboxylic a c i d , has some pharmacological significance. I t shows a certain degree of molecular s p e c i f i c i t y i n the anticonvulsant action of valproic acid analogues. The cyclohexylmethyl conformation was suggested, from a proposed model, to be less effective i n hydrophobic binding due to a s t e r i c effect at a stereoselective position on the hydrophobic s i t e of the GABA receptor complex. Thus i t can be concluded that while l i p o -p h i l i c i t y governed access to s i t e s of action, the dependence of a c t i v i t y on the polar character may explain the diverse structures of a n t i -convulsants provided that the s t e r i c requirements of the hydrophobic binding s i t e are met. Steric effects may lead to i n a c t i v i t y or even convulsant properties of alkyl-substituted compounds. i v TABLE OF CONTENTS P aRe Abstract i i Table of Contents v L i s t of Tables x i L i s t of Figures xiv L i s t of Schemes xvi Symbols and Abbreviations x v i i Acknowledgements xix INTRODUCTION 1 Specific Objectives 6 LITERATURE SURVEY 11 1. C l i n i c a l Use and Anticonvulsant Properties of Valproic Acid (VPA) 11 2. Pharmacological Testing 11 a. Experimental Models of Epilepsy 11 b. Mechanism of Action of Pentylenetetrazole 13 c. Pharmacokinetics of Pentylenetetrazole 15 3. Chemistry and Physicochemical Properties of Valproic Acid 16 4. Mechanism of Action of Valproic Acid 16 5. Studies on Anticonvulsant A c t i v i t y of Valproic Acid Analogues 19 6. General Structure-Activity Relationships of Antiepileptic Drugs 23 7. Structural S p e c i f i c i t y of Anticonvulsants 27 8. Pharmacokinetics of Valproic Acid 33 a. Human 33 b. Rodents 34 c. Kinetics of Valproic Acid i n the Central Nervous System 34 v TABLE OF CONTENTS (Contd) Page 9. Metabolism of Valproic Acid 35 10. Toxicity of Valproic Acid 37 11. A Quantitative Structure-Activity Model 37 12. Physicochemical Parameters used i n Quantitative Structure-A c t i v i t y Relationships 40 13. Hydrophobic Parameters 41 a. Determination of L i p o p h i l i c i t y by the Shake-Flask Procedure 41 b. High-Performance Liquid Chromatographic Determination of L i p o p h i l i c i t y 42 c. Determination of L i p o p h i l i c i t y using Substituent Constants 45 EXPERIMENTAL 47 A. Chemicals and Materials 47 1. Synthesis 47 2. Thin Layer Chromatrography 48 3. High-Performance Liquid Chromatography 48 4. Potentiometric Titrimetry 49 5. Gas Chromatography-Mass Spectrometry 49 6. Pharmacological Testing 50 B. Instrumentation 50 1. Nuclear Magnetic Resonance Spectrometry 50 2. Infra Red Spectrometry 50 3. U l t r a v i o l e t Spectrometry 51 4. Gas Chromatography Mass Spectrometry 51 a. Packed Column 51 b. Capillary Column 51 v i TABLE OF CONTENTS (Contd) Page C. Synthesis of A l k y l Carboxylic Acids, Tetrazoles and Succinaraic Acids 52 1. Synthesis of Alpha-Substituted Aliphatic Acids 52 a. Synthesis of 2-Butylhexanoic Acid 52 b. Synthesis of Valproic Acid 53 c. Synthesis of 2-Propyl-(E)-2-Pentenoic Acid 53 d. Synthesis of 2-Propyl-4-0xopentanoic Acid 55 2. Synthesis of Alpha, Alpha-Disubstituted Aliphatic Acids 56 a. Synthesis of 2,2-Dimethylbutyric Acid 56 b. Synthesis of 2,2-Dimethylvaleric Acid 57 3. Synthesis of Beta-Substituted Aliphatic Acids 57 a. Synthesis of 3-Ethylpentanoic Acid 57 b. Synthesis of Cyclohexylacetic Acid 59 c. Synthesis of 3-Methylvaleric Acid 59 d. Synthesis of 3-Methylhexanoic Acid 60 4. Synthesis of 5-Alkyltetrazoles 60 a. Synthesis of 5-Isoamyltetrazole 60 b. Synthesis of 5-Cyclohexylmethyltetrazole 61 c. Synthesis of 5-Heptyltetrazole 62 5. Synthesis of Succinamic Acids 63 a. Synthesis of N,N-Diethylsuccinamic Acid 63 b. Synthesis of N,N-Dibutylsuccinamic Acid 63 6. Synthesis of Diunsaturated Analogues of Valproic Acid 64 l a . Synthesis of Ethyl 2-(l'-Hydroxypropyl) -3-Pentenoate from Ethyl (E)-2-Pentenoate 64 l b . Synthesis of Ethyl 2-(l'-Hydroxypropyl) -3-Pentenoate from Ethyl (Z)-2-Pentenoate 65 I I Dehydration of Ethyl 2-(l'-Hydroxypropyl) -3-Pentenoate 66 a. Phosphorus Pentoxide 66 b. Toluenesulfonyl Chloride-Pyridine 67 c. Methanesulfonyl Chloride-Potassium Hydride 67 I I I Semi-Preparative Argentation Thin Layer Chromatography 69 IV In Vivo Metabolism-Isolation Procedure 70 v i i TABLE OF CONTENTS (Contd) Page. V Derivatization of Acids 70 VI Photochemical Isomerization 71 VII Capillary Gas Chromatography Mass Spectroraetric Resolution of Isomeric 2-(l'-Propenyl)-2-Pentenoic Acid and 2-Propyl-2,4-Pentadienoic Acid 71 D. Subcutaneous Pentylenetetrazole Seizure Threshold Test 72 1. Animals 72 2. Drugs 72 3. Drug Solutions 73 4. Experimental Procedure 73 a. Characterization of Pentylenetetrazole Seizures (85 mg/kg dose) 73 b. Antagonism of Pentylenetetrazole Clonic Seizures 74 c. Mortality Test 75 d. Toxic Effects of Drugs 75 e. Convulsant A c t i v i t y Test 75 E. High Performance Liquid Chromatographic Method for Determination of Octanol-Water P a r t i t i o n Coefficient 76 1. Instrumentation 76 2. Column 76 3. Eluents 76 4. Compounds 76 5. Sample Preparation 77 6. Retention Time Measurements 77 7. Column Void Time 78 F. Determination of Octanol-Water P a r t i t i o n Coefficients by the Shake-Flask Procedure 78 G. Determination of Apparent Ionization Constants (pKa) by Potentiometric T i t r a t i o n 80 v i i i TABLE OF CONTENTS (Contd) Page RESULTS AND DISCUSSION 82 A. Chemistry 82 1. Alpha-Alkyl Substituted Aliphatic Acids 82 2. Alpha-Alpha D i a l k y l Substituted Aliphatic Acids 82 3. Alpha-Alkyl Substituted Aliphatic Acids with Functionality i n the Carbon Chain 82 4. Beta-Substituted Carboxylic Acids 86 5. N,N-Dialkylsuccinamic Acids 86 6. 5-Alkyltetrazoles 90 7. Diunsaturated Derivatives of Valproic Acid -Synthesis and Metabolism Study 92 8. a. Attempted Synthesis of 2-(l'-Propenyl) -3-Pentenoic Acid 94 b. Aldol Condensation Reactions Toward Synthesis of 2-(l'Propenyl)-2-Pentenoic Acid 96 c. Dehydration of B-Hydroxyunsaturated Esters 99 d. Photochemical Isomerization 103 e. GC Elution Order i n Gas Chromatography Mass Spectrometric Analysis 104 f . Argentation Thin Layer Chromatography 104 g. I d e n t i f i c a t i o n of the Major and Minor Diunsaturated Metabolites of Valproic Acid 109 B. High-Performance Liquid Chromatographic Determination of L i p o p h i l i c i t y 113 1. Assay Method 113 2. Void Time and Retention Mechanism i n Reversed-Phase Liquid Chromatography 122 3. Eluent Effects on Capacity Factor 125 4. High-Performance Liquid Chromatographic Log P (Octanol-Water P a r t i t i o n Coefficient) Values of Valproic Acid and Analogues 129 i x TABLE OF CONTENTS (Contd) Page 5. Comparison of L i p o p h i l i c i t y from Reversed Phase Liquid Chromatography and Other Methods 133 6. Intramolecular Bonding Effects of Amic Acids 143 C. Electronic Structural Effects-Determination of Apparent Ionization Constants 147 1. Analytical Method 147 2. Effect of Structural Constitution on Ionization Constants 156 D. Pharmacological Studies 159 1. Evaluation of Anticonvulsant A c t i v i t y 159 2. Toxicity of Compounds 168 E. Structure-Activity Relationships 170 1. Quantitative Structure-Activity Relationships 170 2. Structural Features that Enhance or Diminish Anticonvulsant A c t i v i t y 183 a. Aliphatic Substituents 183 b. A l i c y c l i c Substituents 184 c. Effects of a Polar Functionality i n the A l k y l Chain 187 d. Model to Show Selective Effects of Aliphatic and A l i c y c l i c Substituents at the Hydrophobic Binding Site 189 SUMMARY AND CONCLUSIONS 197 REFERENCES 203 APPENDIX 219 x LIST OF TABLES Composition and chromatographic data of a mixture of synthesized isomeric dienoates NMR (400 MHz) data for diene VPA ethyl esters Retention times of reference compounds using unbuffered mobile phase (CH3CN/H20) Effect of addition of phosphate buffer (pH 3.5), i n mobile phase (Me0H/H20), on the retention times of reference compounds Retention times of seven reference compounds at different percentages of methanol i n the mobile phase (MeOH/O.OlM NaH2P04) Retention times of seven reference compounds at different flow rates a) i n 70% MeOH : 30% 0.01M NaH2P04 mobile phase b) i n 60% MeOH : 40% 0.01M NaH2P04 mobile phase Retention times of seven reference compounds at different percentages of a c e t o n i t r i l e i n the mobile phase Comparison of void times (t ) determined from a) i n j e c t i o n of methanol ana b) dead time i t e r a t i o n of the retention times of the homologous series from C3H7COOH to CyH-^COOH Correlation of log k' and log PQ/W for seven reference compounds at various compositions of the mobile phase (MeOH/O.OlM NaH2P04) Correlation of log k' and log PQ/W for seven reference compounds at various compositions of the mobile phase (CH3CN/O.OIM NaH2P04) Summary of linear regression parameters for log P versus k' HPLC method for determining the l i p o p h i l i c i t i e s of the acidic compounds using 70% MeOH : 30% 0.01M NaH2P04 as mobile phase HPLC method for determining the l i p o p h i l i c i t i e s of the acidic compounds using 50% CH3CN : 50% 0.01M NaH2P04 as mobile phase Calibration curve data of trimethylacetic acid i n 0.1N HC1 Calibration curve data of N,N-dibutylsuccinamic acid i n 0.1N HC1 x i LIST OF TABLES (Contd) Page 16 Calibration curve data of 5-isoamyltetrazole i n 0.1N HC1 136 17 Calibration curve data of 5-cyclohexylmethyltetrazole i n 137 0.1N HC1 18 Octanol-water p a r t i t i o n coefficients of selected compounds 138 determined by the shake-flask procedure 19 Hansch u-values used i n calculating log PQ/W 139 20 Rekker's fragmental values (f) used i n calculating log 140 Po/w 21 L i p o p h i l i c i t i e s (log PQ/W) of the acidic compounds obtained by different methods 141 22 Determination of the ionization constant of a monobasic 148 aci d , valproic a c i d , i n 10% MeOH 23 Determination of the ionization constant of 5-isoamyl- 149 tetrazole i n 10% MeOH 24 Determination of the ionization constant of trimethyl- 150 acetic acid i n 10% MeOH 25 Determination of the ionization constant of dibutylacetic 151 acid i n 50% MeOH 26 Determination of the ionization constant of N,N-diethyl- 152 succinamic acid i n 50% MeOH 27 Determination of the ionization constant of 5-cyclohexyl- 153 methyltetrazole i n 50% MeOH 28 Determination of the ionization constant of N,N-dibutyl- 154 succinamic acid i n 50% MeOH 29 pka values of valproic acid and analogues 155 30 Polar effect of substitution i n an a l i p h a t i c series 158 31 Protection against PTZ-induced seizures i n mice by 161 valproic acid and i t s analogues 32 Anticonvulsant potency of valproic acid and i t s analogues 164 against the clonic phase of PTZ-induced seizures i n mice 33 Anticonvulsant potency of valproic acid and i t s analogues 165 against the clonic phase of PTZ-induced seizures i n mice. Dose range of acids, 0.2-2.0 mmol/kg x i i LIST OF TABLES (Contd) Page 34 Anticonvulsant a c t i v i t y of VPA and i t s analogues on the 167 threshold of PTZ-induced seizures determined by protection against clonic seizures and by percent mortality i n mice 35 Observed toxic effects of test compounds i n mice 169 36 B i o l o g i c a l data and physicochemical properties of compounds 171 37 Equations obtained correlating the anti-PTZ effect of 174 valproic acid and analogues with t h e i r physicochemical parameters 38 Equations obtained correlating the anti-PTZ effects of VPA 175 and analogues (excluding 5-heptyltetrazole) with th e i r physicochemical parameters 39 Anticonvulsant a c t i v i t y of various drugs against clonic 178 seizures induced by PTZ (s.c. 85 mg/kg) i n mice and thei r physicochemical constants 40 Equations correlating anti-PTZ a c t i v i t y and physicochemical 179 properties of alkyl-substituted anticonvulsants x i i i LIST OF FIGURES Page 1 Chemical structures of valproic acid and analogues 7 2 Chemical structures of t r a d i t i o n a l a n t i e p i l e p t i c drugs 24 3 Structurally-related convulsant and anticonvulsant barbiturates 28 4 Model of GABA-Benzodiazepine receptor-chloride ionophore complex . 32 5 Metabolic pathways of valproic acid 36 6 NMR spectra of N,N-diethylsuccinamic acid and N,N-dibutylsuccinamic acid 91 7 Stereoisomers in al k y l a t i o n and aldol reactions of ester enolates 100 8 Capillary GCMS separation of t-BDMS esters of seven 101 isomeric dienoic acid mixture derived from dehydration reaction with phosphorus pentoxide a) Before UV i r r a d i a t i o n b) After 6hr UV i r r a d i a t i o n 9 a. Mass chromatograms of t-BDMS esters of four isomeric 102 dienoic acid mixture derived from dehydration with p-toluenesulfonyl chloride b. Diene-VPA metabolites i n urine extract 10 GCMS analysis of dienoates eluted from TLC plates, using 107 a 3% Dexsil 300 packed column 11 Chemical structures of diunsaturated derivatives of 110 valproic acid investigated as potential metabolites of valproic acid 12 Capillary GCMS separation of the TMS derivatives of 112 a. A mixture of 2-propyl-2,4-pentadienoic acid and 2-(1'-propenyl)-2-pentenoic acid b. Diene-VPA metabolites i n urine extract 13 UV absorption spectra of four acidic compounds 114 14 Superimposed HPLC chromatograms of acidic compounds 117 using an unbuffered mobile phase (20% CH3CN : 80% H20) xiv LIST OF FIGURES (Contd) Page 15 Superimposed HPLC chromatograms of 23 acidic compounds 132 using 70% MeOH : 30% 0.01M NaH2P04 as mobile phase 16 Dose-response curves of valproic acid and analogues using 163 the subcutaneous pentylenetetrazole seizure threshold test i n mice 17 Active and inactive a l i c y c l i c and a l i c y c l i c a l k y l - s u b s t i t u t e d 186 compounds 18 Conformations of (a) GABA, (b) valproic acid amide and 190 (c) 3-ethylpentanoic acid amide 19 Model of pharmacophoric structural features i n carboxylic 193 acids and tetrazoles xv LIST OF SCHEMES Page 1 Synthetic pathway for alpha-substituted a l i p h a t i c acids 83 2 Synthetic route for alpha, alpha-disubstituted a l i p h a t i c 84 acids 3 Outline for synthesis of 2-propyl-(E)-2-pentenoic acid 85 4 Synthetic sequence for preparation of 2-propyl-4-oxo- 87 pentanoic acid (4-Keto VPA) 5 Pathways for synthesis of the beta-substituted carb- 88 oxylic acids 6 Synthetic route for the succinamic acids 89 7 Synthetic pathway for 5-alkyltetrazoles 93 8 Outline for synthesis of dienol ether used i n attempted 95 preparation of 2-(1'-propenyl)-3-pentenoic acid 9 Stereoselective synthetic routes for preparation of 97 2-(1'-propenyl)-2-pentenoic acid 10 Kinetic model proposed by some investigators (183,184) for 145 the hydrolysis of maleamic acids xvi SYMBOLS AND ABBREVIATIONS P octanol-water p a r t i t i o n coefficient pKa negative logarithm of the apparent acid ionization constant effective dose i n 50% of mice TD50 toxic dose i n 50% of mice k' capacity factor retention time of retained solute fco elution time of unretained solute i . p . intraperitoneal s.c. subcutaneous r correlation c o e f f i c i e n t 5 standard error of estimate y dipole moment Eg Taft s t e r i c factor O* polar substituent constant E trans Z c i s m/z mass to charge r a t i o f fragmental constant s singlet t t r i p l e t d doublet dd doublet of doublets q quadruplet m multiplet 6 chemical s h i f t J coupling constant x v i i SYMBOLS AND ABBREVIATIONS (Contd) o/w octanol-water i . d . internal diameter ACN ac e t o n i t r i l e BDZ benzodiazepine C-18 octadecylsilane CNS central nervous system DHP dihydropicrotoxinin diene diunsaturated GABA gamma-aminobutyric acid GABA-T Gamma-aminobutyrate transaminase GAD glutamic acid decarboxylase GCMS gas chromatography mass spectrometry HMPA hexamethylphosphoramide HPLC high-performance l i q u i d chromatography IR i n f r a red LDA lithium diisopropylamide l i t . l i t e r a t u r e MES maximal electroshock seizure test NMR nuclear magnetic resonance ODS octadecylsilane PTZ pentylenetetrazole QSAR quantitative structure-activity relationships RP reversed phase SAR structure-activity relationships s.c. PTZ subcutaneous pentylenetetrazole seizure threshold test SSA Succinic semialdehyde x v i i i SYMBOLS AND ABBREVIATIONS (Contd) t-BDMS te r t i a r y b u t y l d i m e t h y l s i l y l TBPS t-butylbicyclophosphothionate THF tetrahydrofuran TLC thin layer chromatography TMS t r i m e t h y l s i l y l 2,3'-diene VPA 2-(l'-propenyl)-2-pentanoic acid 2,4-diene VPA 2-propyl-2,4-pentadienoic acid 2E-3'E Diene 2-[(E)-l'-propenyl]-(E)-2-pentenoate 2E-3'Z Diene 2-[(Z)-l'-propenyl]-(E)-2-pentenoate 2Z-3'E Diene 2-[(E)-l'propenyl]-(Z)-2-pentenoate 2Z-3'Z Diene 2-[(Z)-l'-propenyl]-(Z)-2-pentenoate 3Z-3'Z Diene 2-[(Z)-l'-propenyl]-(Z)-3-pentenoate 3Z-3'E Diene 2-[(E)-l'-propenyl]-(Z)-3-pentenoate 3E-3'E Diene 2-[(E)-l'-propenyl]-(E)-3-pentenoate 3,3'-Diene 2-(l'-propenyl)-3-pentenoate VPA valproic acid 2- ene VPA 2-propyl-2-pentenoic acid 4-ene VPA 2-propyl-4-pentenoic acid 3- ene VPA 2-propyl-3-pentenoic acid 4- OH VPA 2-propyl-4-hydroxypentanoic acid 3- OH VPA 2-propyl-3-hydroxypentanoic acid 4- Keto VPA 2-propyl-4-oxopentanoic acid xix ACKNOWLEDGEMENTS I would l i k e to express my sincere thanks to Dr. Frank Abbott for his excellent supervision throughout the course of my graduate studies. I greatly appreciate his helpful advice and enthusiastic response to academic and personal a f f a i r s . The helpful discussions with members of my graduate committee, Dr. Terence Brown, Dr. David Godin, Dr. Sid Katz and Dr. Jim Orr, are g r a t e f u l l y acknowledged. I wish to ex p r e s s my g r a t i t u d e to Dr. R.A. Wall and Dr. B. D. Roufogalis for th e i r helpful suggestions. S p e c i a l thanks to Mr. Roland Burton for h i s valuable t e c h n i c a l assistance i n the gas chromatography - mass spectrometric analysis. I appreciate the various assistance by my lab mates and fellow graduate students, Greg S l a t e r , Jeanine Kassam, Sukhbinder Panesar, Kuldeep Singh, Ron Lee and David Kwok. This work would not have been p o s s i b l e without the f i n a n c i a l support provided by a grant from the B.C. Health Care Research F o u n d a t i o n and the k i n d support p r o v i d e d by the F a c u l t y of Pharmaceutical Sciences. F i n a l l y , I wish to thank my parents, brothers and s i s t e r s for t h e i r unflagging support. xx INTRODUCTION The medium branched-chain fatty acid, valproic acid, I,(2-propyl-pentanoic acid) i s a r e l a t i v e l y new a n t i e p i l e p t i c drug, introduced i n North America i n 1978. I t i s used i n treatment of absence seizures and in combination therapy for generalized tonic-clonic seizures. The d i s -covery of the anticonvulsant properties of valproic acid by Meunier et a l . (1) has di r e c t e d some a t t e n t i o n toward s i m i l a r p o t e n t i a l a n t i -convulsant compounds without the imide s t r u c t u r e as found i n most conventional a n t i e p i l e p t i c drugs. Valproic acid presents a good lead compound for structure-activity studies because i t appears to have a novel mechanism of action involving augmentation of GABAergic a c t i v i t y . I t has been reported to increase brain GABA levels i n vivo and i n h i b i t enzymes involved i n GABA degrad-a t i o n pathways (2). E l u c i d a t i o n of the molecular a c t i o n s of v a l p r o i c a c i d that are d i r e c t l y r e l a t e d to i t s anticonvulsant e f f e c t would be f a c i l i t a t e d by the development of structure-activity relationships (SAR) within a series of closely-related analogues of valproic acid. Invest-i g a t i o n of the molecular s p e c i f i c i t y of the anticonvulsant a c t i o n of valproic acid analogues i s also of i n t e r e s t i n determining s t r u c t u r a l requirements at the s i t e of action. E a r l i e r studies on the relationships between structure and a c t i v i t y explored the p o s s i b i l i t y of modifying the carboxylic functional group to enhance anticonvulsant a c t i v i t y . Most of the compounds tested f o r anticonvulsant a c t i v i t y were d i p r o p y l a c e t i c a c i d d e r i v a t i v e s such as amides (3-5), ureides (6), esters (3,5) including an oxazepam derivative (7) and a hydantoin d e r i v a t i v e of v a l p r o i c a c i d (6). Ureas (8), a l c o -1 hols (5,9), carbamates (9) and ketones (9) with the 1-propylbutyl chain, as i n valproic a c i d , were also examined for anticonvulsant a c t i v i t y . A number of a l i p h a t i c and a l i c y c l i c - s u b s t i t u t e d carboxylic acids have been evaluated for anticonvulsant a c t i v i t y but only at a single dose (5,10,11) or b i o l o g i c a l response was determined using a less spec-i f i c t e s t , i . e . protection against pentylenetetrazole (PTZ) -induced mortality (10). However, there have been recent studies which reported the dose-dependency of the anti-PTZ clonic seizure a c t i v i t y of homo-logous straight-chain and alpha-branched fatty acids (12,13,14). Anticonvulsant drugs have diverse chemical structures which suggest they may have different mechanisms of action or that they may interact at a similar s i t e of action by virtue of having some similar pharma-cophore groups. There have been s e v e r a l attempts to uncover the pharmacophoric structural features of the conventional a n t i e p i l e p t i c drugs. Andrews (15) reported that there was no relationship between the anticonvulsant a c t i v i t y and the effective atomic charges at the quater-nary carbon common to the anticonvulsant drugs with the imide or ureide structure. Some investigators have looked at the variety of functional groups that w i l l confer anticonvulsant a c t i v i t y on compounds with a l k y l or phenyl substituents (16). Patrick and Bresee (17) reported that hydrogen-bonding strengths of the major a n t i e p i l e p t i c drugs, measured by the hydrogen-bonding enthalpies with phenol, were the same for the compounds studied and thus unrelated to a c t i v i t y . Camerman and Camerman (18) examined the x-ray structures of some conventional a n t i e p i l e p t i c drugs including phenytoin, diazepam and phenylacylurea and proposed that the s p a t i a l configurations of these compounds allow superposition of the phenyl groups and also the carbonyl groups or an equivalent electron-donor group. 2 Valproic acid has the basic structural features, a polar moiety with an electron-donor group or hydrogen-bonding group and hydrophobic substituents, i n common with conventional anticonvulsant drugs. I t also possesses a carboxylic group and a l k y l chain as i n the structure of gamma-aminobutyric acid (GABA). However, valproic acid does not have a nitrogen function as found i n the structure of GABA. Different a l i -phatic and a l i c y c l i c groups may enhance or have an adverse effect on the interaction of the carboxylic acids at the s i t e of action. This has been shown i n some barbiturates where the presence of an i s o a l k y l , isoalkenyl and 6-cyclohexylidene-ethyl chain at the quaternary carbon results i n convulsant a c t i v i t y (19). From studies on a homologous alpha-branched a l i p h a t i c carboxylic acid s e r i e s , Keane et a l . (13) and Meldrum et a l . (12) reported that there was a si g n i f i c a n t correlation between the anticonvulsant potency and the length of side-chain. They also found good correlations between the increase i n GABA brain levels and anticonvulsant potency. In a different study, Perlman and Goldstein (14) used a fluorescent probe to show that the a b i l i t y of these homologous carboxylic acids to disorder synaptosomal plasma membranes correlated well with their anticonvulsant potency. From the fluorescent polarization studies, they suggested that the anticonvulsant effect of valproic acid i s mediated by nonspecific mechanisms similar to those of general anesthetics. I t seems that a wide variety of structures are required to investigate the structural s p e c i f i c i t y of valproic acid analogues. The present investigation i s concerned with the effect of diverse substituents on anticonvulsant a c t i v i t y of valproic acid analogues. The physicochemical properties, namely l i p o p h i l i c i t y , electronic properties 3 and s t e r i c factors have been determined to find whether a c t i v i t y i n vivo i s determined by a nonspecific property such as l i p o p h i l i c i t y or whether there i s a stereoelectronic factor determining anticonvulsant a c t i v i t y . Multiparametric relationships increase the likeli h o o d of structural s p e c i f i c i t y i n a class of structurally-related compounds. This approach has been used by various investigators to reconcile the high structural s p e c i f i c i t y of drug-receptor interactions and the common physicochemical p r o p e r t i e s of disparate s t r u c t u r e s with a common b i o l o g i c a l a c t -i v i t y (20). Different tests have been used to evaluate the anticonvulsant potency of compounds for development of SAR. Currently two i n vivo experimental models of epilepsy have been widely employed for such purposes. These are the maximal electroshock seizure test (MES) and the subcutaneous pentylenetetrazole seizure threshold t e s t . In t h i s study, the anticonvulsant potency of alkyl-substituted carboxylic acids and tetrazoles have been determined by the subcutaneous pentylenetetrazole seizure threshold t e s t . Several investigators have pointed out the b i o i s o s t e r i s m between the c a r b o x y l i c group and the t e t r a z o l e nuc-leus (21-23). Comparative studies on substituted carboxylic acid and tetrazoles have revealed s i m i l a r , greater or i n f e r i o r b i o l o g i c a l a c t i v -i t y of the tetrazole analogues (23). Kraus (24) found both valproic acid and i t s corresponding tetrazole, 4-tetrazolylheptane, inhibited succinic-semialdehyde dehydrogenase i n the GABA metabolic shunt with inhibit o r y constants (K^) of 0.7 mM and 0.75 mM respectively. Both compounds have a polar acidic group and a l k y l substituents. However, pentylenetetrazole i s a convulsant and has no acidic properties. The effect of acidic properties of compounds on anticonvulsant potency has made i t necessary that the electronic effect of the polar moiety, which 4 i s influenced by the a l k y l substituents, be quantified by physical methods. By expressing physicochemical properties of structural features and the anticonvulsant a c t i v i t y i n quantitative terms, multiple regression analysis i s used to obtain quantitative structure-activity relationships (QSAR). This li n e a r free-energy relationships model was pioneered by Hansch and co-workers (29). The use of s t a t i s t i c a l analysis i n the regression equation appears to allow much more objective establishment of SAR. Several researchers have used the Hansch approach to i d e n t i f y molecular properties that account for the anticonvulsant a c t i v i t y of the str u c t u r a l l y diverse a n t i e p i l e p t i c drugs (15,19,26,27). Lien and co-workers (26,28) found that 1,4-benzodiazepines, which are known to have sp e c i f i c binding s i t e s , could not be included with other a n t i e p i l e p t i c drugs to develop a s i g n i f i c a n t QSAR. There have been suggestions from QSAR studies that CNS-acting drugs have d i f f e r i n g l i p o p h i l i c i t y require-ments which determine their d i s t r i b u t i o n a l l o c a l i z a t i o n and hydrophobic binding at the active s i t e (20). Hansch and co-workers (29) developed quantitative s t r u c t u r e - a c t i v i t y relationships for a series of hypnotic b a r b i t u r a t e s and suggested an optimal octanol-water p a r t i t i o n c o e f f i c i e n t of about 100 (log P = 2.0). This QSAR approach has not been applied to either a l k y l substituted carboxylic acids or tetrazoles which may s p e c i f i c a l l y antagonize PTZ-induced clonic seizures because of the i n s i m i l a r i t y i n structure and the known bioisosterism between carboxylic and tetrazole groups. I t appears that the semi-rigid cycloalkyl-substituted compounds are potential structures to investigate the structural requirements at the s i t e of a c t i o n f o r v a l p r o i c a c i d a n a l o g u e s . A p p a r e n t l y 5 - a l k y l -5 tetrazoles (21,23) and diunsaturated derivatives of valproic acid (2) have not been evaluated for t h e i r anticonvulsant a c t i v i t y . Unidentified diunsaturated metabolites of valproic acid have been reported to be present i n the serum and urine of patients on valproic acid therapy (2). The a v a i l a b i l i t y of synthetic reference material w i l l h e l p c h a r a c t e r i z e the s t e r e o c h e m i c a l c o n f i g u r a t i o n of these diunsaturated metabolites. Their synthesis requires stereoselective methods owing to the number of positional isomers and m u l t i p l i c i t y of the stereoisomers. Compounds with polar groups i n the a l k y l chain, i n addition to the terminal carboxylic function, have also been included i n the study. Specific Objectives 1. Valproic acid has been reported to exhibit selective actions that may mediate i t s anticonvulsant e f f e c t s . A wide variety of struct-ural analogues of valproic acid were to be used to investigate the degree of structural s p e c i f i c i t y of the anticonvulsant actions. The series of valproic acid analogues (Figure 1) include compounds with y- a l k y l substituents; a,a-dialkyl substituents; B-alkyl sub-stituents; a-alkyl substituent, a l i c y c l i c substituents; polar groups i n a l k y l chain; unsaturated a l k y l groups. The f i r s t part of the study was to apply synthetic methods to prepare the substituted carboxylic acids and tetrazoles. 2. In the course of the study, i t was of interest to employ a stereo-selective synthetic method to prepare the diunsaturated analogues of valproic a c i d , 2-[(E)-l'-propenyl]-(E)-2-pentenoic acid and 2-[(Z)-l'-propenyl]-(E)-2-pentenoic a c i d , for evaluation of the anticonvulsant a c t i v i t y and i d e n t i f i c a t i o n of the major d i -6 CH3-CH2-CH2 CH3-CH2-CH2-CH2 CH3 ^CHCOOH \HCOOH ^CHCOOH C H3- C H2- C H £ C H3- C H2- C H2- C H £ C H3 I II III CH3 CH3 CH3-CH2 CH 3-C-COOH CH3-CH2-C-COOH NCH-CH2COOH CH3 CH3-CH2 IV V VI CH3 3 0 •CH2 ^ CHCOOH ChL-CH9-CH9 CHQ-CH=CH r, /C-C00H C^-COOH L M3 L L H2 v CH3-CH2-CH ' CHg-CHg-CI? C H ^ - C H ^ VII VIII IX H H / N CH CH^-CH9-CH9-CH7CH?-CH?-CH9-Cf J CH-CH9-CH9-C^ N 3 2 2 2 2 2 Vn* < 2 2 v > N-N X XI H C H3 -C00H XII XIII XIV CH3-CH2 CH3-CH2-CH2-CH2 Vc-CH9CH9C00H NN-C-CH9-CH9C00H / II 2 2 / II 2 2 CH3-CH2 0 CH3-CH2-CH2-CH2 0 XV XVI Figure 1. Chemical Structures of Valproic Acid and analogues 7 unsaturated metabolites of valproic a c i d . 3. Aside from application of reverse-phase HPLC as an a n a l y t i c a l t o o l , one can relate the observed chromatographic retention parameters of compounds to physicochemical properties such as l i p o p h i l i c i t y . Octanol-water p a r t i t i o n c o e f f i c i e n t s i n d i r e c t l y determined by reverse phase (RP) - HPLC have been shown to be i n close agreement with values obtained by the t r a d i t i o n a l shake-flask procedure. Since chromatographic methods are generally more rapid than the s t a t i c method owing to the higher rate of e q u i l i b r a t i o n of solute between the phases, t h i s study u t i l i z e s a RP-HPLC procedure to determine the octanol-water p a r t i t i o n c o e f f i c i e n t s of valproic acid analogues. The effects of different mobile phases and compounds of diverse chemical structures on the accuracy of the RP-HPLC method were examined. 4. Variation of the substitution pattern i n the carboxylic acids and tetrazoles i s manifested by the resultant electronic effects at the acidic groups of the polar moiety. While the apparent ionization constants (pKa) of the compounds may have been variously reported i n the l i t e r a t u r e , the conditions used for measurement of pKa values usually d i f f e r from one laboratory to the other. I t was i n the interest of accuracy and consistency to employ an appropriate method to determine the pKa of valproic acid and analogues. 5. Valproic acid and alkyl-substituted compounds have been reported to be more effec t i v e against PTZ-induced clonic seizures compared to other chemical-induced seizures or e l e c t r i c a l l y induced seizures. Since there have been r e l a t i v e l y few studies on the dose-dependent anti-PTZ a c t i v i t y of valproic acid analogues, the aim of the study 8 was to determine the anti-PTZ potency of not only the homologous a-branched acids but also compounds with other substitutional char-a c t e r i s t i c s . Since the QSAR approach has not been applied to alkyl-substituted carboxylic acids and tetrazoles, i t was of interest to evaluate the role played by the electronic properties of the polar group and the hydrophobicity of the a l k y l group on the anticonvulsant potencies of test compounds. Additional structural properties considered i n the present study were the conformational or s t e r i c effects and the molecular dipole moments. (5) HC-II (4) N N (3) Tetrazole NH (1) N (2) (8) (7) CH2 CH2 (9) (6) CH2 (5) C II (4) N CH2 (10) 11 (1) N (2) N (3) Pentylenetetrazole (PTZ) Although t h i s study p e r t a i n s s p e c i f i c a l l y to a l k y l s u b s t i t u t e d carboxylic acids and tetrazoles, comparative studies with other a l k y l -substituted heterocyclic compounds were made possible by the available l i t e r a t u r e data on their anti-PTZ potencies and physicochemical prop-e r t i e s . The basis of such comparison i s the p o s s i b i l i t y of a common molecular action of the a l k y l substituted compounds as suggested from biochemical, pharmacological and neurophysiological studies by several researchers (30-34). 10 LITERATURE SURVEY 1. C l i n i c a l Use and Anticonvulsant Properties of Valproic Acid (VPA) The therapeutic efficacy of VPA (Depakene®) has been demonstrated i n several c l i n i c a l studies (35,36). VPA i s now widely used i n primary generalized epilepsies, p a r t i c u l a r l y those of the absence seizure type. Valproic acid i s considered to be at least as effective as ethosuximide i n the treatment of absence seizure (37). I t s broad spectrum of a n t i -e p i l e p t i c effects has, however, proven valuable i n combination therapy for myoclonic epilepsy and generalized tonic-clonic seizures (35,36). VPA shows selective a c t i v i t y against several types of chemically or electrically-induced seizures i n a variety of species. VPA has a weak a c t i v i t y against maximal electroshock seizures i n mice compared to the a c t i v i t y of phenobarbital and phenytoin (38). VPA i s more effective i n p r e v e n t i o n of c l o n i c or t o n i c s e i z u r e s induced by PTZ and picrotoxin (38-40). High doses are reported to block t o n i c - c l o n i c , bicuculline and strychnine-induced seizures (38-40). 2. Pharmacological Testing a. Experimental Models of Epilepsy Quantitative effects of structural variants on the pharma-cological a c t i v i t y of the a n t i e p i l e p t i c drugs have been obtained by the use of numerous tests i n experimental animals. The common experimental techniques for inducing seizure i n rodents include electroshock and systemic administration of convulsants such as PTZ. The maximal electroshock seizure test (MES) and the sub-cutaneous pentylenetetrazole seizure threshold test (s.c. PTZ) are 11 widely recognized models for determining the anticonvulsant a c t i v -i t y of compounds. A c t i v i t y i n the MES test has been correlated with a compound's a b i l i t y to modify maximal seizures or i n h i b i t the seizure spread through the brain. In contrast the s.c. PTZ test measures the a b i l i t y of a compound to elevate the degree of seizure threshold (41). The MES test i s thus a model of generalized tonic-c l o n i c s eizure while the s.c. PTZ t e s t i s a model of absence seizures. Other models have been developed to simulate p a r t i a l seizures. Extensive investigations have been carried out to standardize the MES test (41-43) and the s.c. PTZ test (41,44,45). These two methods have been preferred over other tests since they are report-ed to be r a p i d , simple, e a s i l y c o n t r o l l e d and n o n - e r r a t i c i n producing the clonic or tonic seizure component (38). The value of these tests has been shown by good correlation between test results and efficacy i n c l i n i c a l epilepsy (38,46). In the MES t e s t , max-imal seizures are induced by passing high current ( f i v e to seven times threshold value, i . e . 50 mA, 60 Hz) through corneal electrodes for 0.2 sec i n mice (46). In the MES t e s t , active compounds protect against the tonic extension of hind limbs. PTZ i s administered i n mice or r a t s at doses ranging from frank convulsant to nearly l e t h a l doses. In the s.c. PTZ t e s t , threshold c l o n i c s e i z u r e s are produced when PTZ i s a d m i n i s t e r e d subcutaneously i n a dose of 85 mg/kg i n mice. This i s the reported CDoj dose i n mice (46). Protection i n the s.c. PTZ test i s defined as absence of c l o n i c spasms of duration greater than 5 sec. Another PTZ seizure threshold t e s t , the timed i . v . infusion method, 12 has also been used to determine anticonvulsant potency i n rats (45). Whereas there i s a need for the use of i n v i t r o models of neuronal discharge, especially i n id e n t i f y i n g the selective mol-ecular actions of a n t i e p i l e p t i c drugs (31), the empirical i n vivo models have a l s o been d i s c r i m i n a t i v e i n showing d i f f e r e n t i a l actions of a n t i e p i l e p t i c drugs. Ethosuximide and trimethadione, both alkyl-substituted compounds, are effective only for absence seizures and i n the s.c. PTZ t e s t (44,47). Phenytoin i s i n -effective for absence seizures and i n the s.c. PTZ t e s t , but i t prevents generalized tonic-clonic seizures, p a r t i a l seizures and maximal electroshock seizures (42-44). Phenytoin i s thought to suppress the spread of seizures. 1,4-Benzodiazepines, which block the tonic seizure component i n MES test and the clonic spasms i n s.c. PTZ test at approximately the same dose, may act by elevation of seizure threshold (48). Barbiturates are e f f e c t i v e i n the MES test and s.c. PTZ test at separated doses suggesting that the mechanism of action involves both elevation of seizure threshold and i n h i b i t i o n of seizure spread (41,48). The effect of valproic acid i s similar to barbiturates, except that i t i s more effective i n elevation of seizure threshold (39,40). b. Mechanism of Action of PTZ PTZ i s known to act on the whole CNS but there are c o n f l i c t i n g reports as to whether the PTZ-induced convulsions originate i n s p e c i f i c areas such as the cerebral cortex (49). The mechanism of PTZ convulsant actions i s s t i l l not defined. However, the direct effect on c e l l membrane e x c i t a b i l i t y , together with i t s a b i l i t y to 13 s e l e c t i v e l y block the effect of GABA on chloride conductance may account for the convulsant properties of PTZ (49,50). A number of workers have reported that PTZ enhances the e x c i t -atory system through d i r e c t e f f e c t on membrane pr o p e r t i e s to increase spontaneous discharge by a l t e r i n g i o n i c conduct-ance (51,52). Using cultured mouse spinal cord neurons, Macdonald and Barker (53) noted that PTZ reversibly antagonizes i n a dose-dependent manner the conductance produced by iontophoresed GABA, without an effect on the resting membrane properties. The response was selective for GABA since PTZ did not affect the response due to glycine, 8-alanine and glutamic a c i d . The synaptic action of PTZ i s reported to block the increase i n chloride conductance induced by iontophoresed GABA i n neurons of Aplysia Californicus ( 5 4 ) . In the same study ( 5 4 ) , PTZ was observed to show minimal effect on excitatory response of inward sodium current and in h i b i t o r y effect of outward potassium c u r r e n t . Unlike v a l p r o i c a c i d , PTZ i s reported to have no s i g n i f i c a n t effect on brain GABA levels ( 5 5 ) . PTZ inhibited the a c t i v i t y of glutamic acid decarboxylase i n v i t r o at concentrations related to physiological levels ( 5 5 ) . In vivo, PTZ inhibited the a c t i v i t y of glutamic acid decarboxylase at twice the CDgj dose i n mice ( 5 5 ) . I t has been observed that PTZ seizures are similar to those induced by picrotoxin but unlike those of strychnine ( 5 0 ) . Ticku and Olsen (33) reported that PTZ shows a s i g n i f i c a n t a f f i n i t y for the picrotoxin binding s i t e (IC^Q = 30 yM) hence i t i s l i k e l y to act at the benzodiazepine-GABA-chloride ionophore receptor complex. Recently Squires et a l . (56) examined a series of convulsant t e t r a -14 zoles f o r t h e i r potencies i n d i s p l a c i n g S - t - b u t y l b i c y c l o phosphonothionate (TBPS) from the picrotoxin s i t e on the BDZ-GABA-C l ~ ionophore receptor complex. A good c o r r e l a t i o n between r e l a t i v e a f f i n i t i e s f o r JJS-TBPS binding s i t e and convulsant potencies (IC5Q of PTZ = 0.54 mM) was found. Tetrazole derivatives have been investigated for thei r con-vulsant a c t i v i t y . Potent convulsant effects were obtained when 1,5-positions are bridged by a penta to heptamethylene chain (57). Tetrazole derivatives possessing a methyl or ethyl group at pos-iti o n - 5 and a substituent at position 1(N) with four to six carbon atoms (a l i p h a t i c or a l i c y c l i c ) are active as convulsants (58). Depressant effects of some 1,5-disubstituted tetrazoles have been obtained with substitution of aminophenyl groups at position 1 or 5 and a l i p h a t i c groups at the unsubstituted position (59). Ahmad and Shunkla (60) investigated 5-acetamido-substituted tetrazoles and reported anticonvulsant a c t i v i t y of 5-(N-phenylacetamido)-tetrazole with an i. p . dose of 100 mg/kg. c. Pharmacokinetics of PTZ The pharmacokinetic properties of PTZ are attributed to i t s ready s o l u b i l i t y i n both l i p i d s and water. PTZ i s rapidly absorbed from a l l s i t e s of administration i n mice and rats and i s rapidly distributed i n t o t a l body water (61,62). Brain ^C-PTZ uptake i s rapid within 10 min after s.c. administration i n rats and the half -l i f e of PTZ i n brain ranged from 16-21 min (191). The serum h a l f -l i f e of r a d i o l a b e l e d PTZ i n rats was found to be 2-4 hr (61,62). Metabolism i s the major route of PTZ elimination. A low percentage of unchanged drug i s excreted i n urine (61,64). The major metab-15 o l i t e s i n rats are 6-hydroxy PTZ and 8-hydroxy PTZ (23,61). Bind-ing to tissue constituents i s almost negligible and approximately 9% of the drug i s bound to plasma proteins (65). 3. Chemistry and Physicochemical Properties of Valproic Acid Valproic acid (2-propylpentanoic acid) was f i r s t synthesized by Burton (66) i n 1882 by alkaline decomposition of 2,2-dipropylaceto-acetate. The physical and chemical properties of valproic acid have been reviewed by Chang (67). VPA has a s o l u b i l i t y of 1.27 mg/ml i n water. The c r i t i c a l m i c e l l a r concentration of i t s s t r a i g h t chain isomer, octanoic acid i s reported to be 8.0 mmol/L (68). The p a r t i t i o n c o e f f i c i e n t of VPA between various organic solvents and phosphate buffer (pH 7.4) has been reported to be 0.013 for heptane, 0.064 for benzene and 0.21 for chloroform (69). Valproic acid showed a pKa value of 4.56 when determined by potentiometric t i t r a t i o n of the acid i n aqueous medium (70). 4. Mechanism of Action of Valproic Acid The basic mechanism by which valproic acid exerts i t anticonvulsant effect i s not yet c l e a r . Several biochemical and neurophysiological studies have shed some l i g h t on i t s mode of action (2,71,72). The bio-l o g i c a l s p e c i f i c i t y of v a l p r o i c a c i d , l i k e t h a t of oth e r a n t i c o n v u l s a n t s , i s broad. At high doses, v a l p r o i c a c i d shows n o n s p e c i f i c a c t i o n s such as i n h i b i t i o n of mitochondrial o x i d a t i v e phosphorylation (73), actions shared by non-anticonvulsants (73). Some investigators have used the selective actions of anticonvulsants, such as v a l p r o i c a c i d , at lower drug concentrations to pinpoint t h e i r 16 mechanism of a c t i o n (31,74). The p o s s i b i l i t y of a n o n s p e c i f i c anticonvulsant action of valproic acid has received some attention (14) although more direct effects have been observed for valproic a c i d . Four possible modes of action have been postulated from direct effects observed for valproic a c i d . These are elevation of brain GABA l e v e l s , enhancement of GABAergic i n h i b i t i o n at the post-synaptic GABA receptor complex, reduction of e x c i t a t o r y a c t i o n of aspartate and increase i n membrane conductance to potassium ions. Valproic acid has been repeatedly shown to raise l e v e l s of GABA i n the brain (75,76). The increase i n GABA levels induced by valproic acid appears to take place i n nerve terminals (77). In v i t r o studies showed that valproic acid levels up to 1 mM had no effect on the release of pr e l o a d e d r a d i o l a b e l l e d GABA from r a t b r a i n synaptosomal preparations (78). GABA i s synthesized by GAD-catalyzed decarboxylation of glutamate and i s metabolized to s u c c i n i c semialdehyde by the reversible GABA-T catalyzed reaction. Succinic semialdehyde (SSA) i s metabolized either to succinic acid by SSA-dehydrogenase catalyzed oxidation or metabolized by aldehyde reductase-catalyzed reduction to 4-hydroxybutyric a c i d . The increase i n GABA brain levels produced by valproic acid was attributed from early studies to i n h i b i t i o n of GABA-T (75). I t appears that VPA more potently i n h i b i t s SSA-dehydrogenase (Ki f 0.5 mM) compared to i n h i b i t i o n of GABA-T, K± >20 mM (79,80). Thus an increase i n succinic semialdehyde levels might elevate GABA levels by in d i r e c t l y i n h i b i t i n g GABA-T. Valproic acid has also been shown to increase the a c t i v i t y of the GABA biosynthetic enzyme, glutamic acid decarboxylase, i n whole brain and i n brain synaptosomes i n mice (77). Recent studies suggest that VPA may rather act at the post-synaptic 17 membrane of the GABAergic synapse. The administration of valproic acid by iontophoresis potentiates the i n h i b i t i o n effect of GABA on neurons i n the brain (81,82), i n mouse spinal cord neuronal culture (74) and i n cuneate fi b r e preparations (83). Iontophoresed valproic acid had no sig n i f i c a n t effect on responses from either iontophoresed glycine or glutamate ( 7 4 ) . The concentration of v a l p r o i c a c i d a c t i n g on the neuronal membrane during iontophoretic application of valproate was not ascertained i n some of the studies. Harrison and Simmonds (83) reported that valproic acid enhances GABA effects at 3.0 mM i n v i t r o . Benzodiazepines and barbiturates are also reported to potentiate the postsynaptic i n h i b i t o r y effects of GABA (31). This action of a n t i -convulsant barbiturates and valproic acid i s suggested to occur at the picrotoxin binding s i t e , i . e . the chloride ionophore (30,33,34). The GABA receptor appears to be part of a protein complex containing recept-or si t e s for GABA, BDZ and picrotoxin as well as the chloride iono-phore (30). Loscher (84) has reported that valproic acid does not i n h i b i t nor enhance the binding of H-GABA to rat brain membranes at concentrations up to 1 mM. Valproic acid also does not bind to benzo-diazepine receptors at concentrations up to 1 mM (34). The component of the GABA-receptor-chloride ionophore, the picrotoxin "receptor" i s a s i t e at which several convulsants (including pentylenetetrazole) and anticonvulsants appear to act (30,33,34). Ticku and Davis (34) found that valproic acid i n h i b i t s the binding of H-dihydropicrotoxinin to rat brain membrane (IC^Q = 0.5 mM). Valproic acid i s reported to have no a f f i n i t y f o r the binding s i t e of H-phenytoin to rat brain mem-branes (85). In an i n t r a c e l l u l a r study of the effects of valproatec on non-18 mammalian species, Johnson (71) found that concentrations of VPA (5-30 mM), 15-50 times the serum l e v e l of patients on VPA therapy, caused an increase i n potassium membrane conductance to produce hyperpolariz-ation of the resting membrane. Valproic acid i s reported by several researchers to reduce brain aspartate i n vivo (12,75,86). The time course for the decrease i n aspartate concentrations correlated with the period of valproic acid-induced protection against audiogenic seizure i n mice and with the increase i n brain GABA levels (86). So far i t i s not certain whether valproic acid exerts i t s anticon-vulsant effects exclusively through one of the four possible modes of action. Potentiation of GABAergic post-synaptic i n h i b i t i o n may act together with elevated GABA levels resulting from i n h i b i t i o n of SSA-dehydrogenase. Moreover there appear to be i n c o n s i s t e n c i e s as to whether the suggested molecular actions of valproate occur at therapeut-i c a l l y r elevant v a l p r o i c a c i d b r a i n and serum l e v e l s i n mammalian species. In patients receiving therapeutic doses of valproic a c i d , serum levels of 0.3-0.8 mM have been reported to be effective (2). The present study, by investigating the SAR of closely-related analogues of valproic a c i d , may be useful i n determining which of the molecular actions of valproic acid can be correlated with anticonvulsant a c t i v i t y . 5. Studies on Anticonvulsant A c t i v i t y of Valproic Acid Analogues Valproic acid i s not the only active a l i p h a t i c carboxylic acid among i t s congeners. Carraz (10) reported protection of PTZ-induced mortality by some alpha-branched fatty acids of the general form, XVII. These studies were conducted at a single dose, 200 mg/kg i . p . i n mice. 19 R2-CH-COOH Rj . R2 = CH | Rj = R2 = ^2n5 Rj Rj = R2 = C^Hy XVII A c t i v i t y rose with increases i n chain length up to v a l p r o i c a c i d . Dibutylacetic acid was reported to be i n a c t i v e . Recent studies by Chapman et a l . (12) and Keane et a l . (13) showed that the a c t i v i t y of alpha-branched fatty acids continued to increase with elongation of the side-chain. However, they found straight-chain fatty acids (C^-C^) to be inactive up to 4.0 mmol/kg doses. Anticonvulsant a c t i v i t i e s were also noted with t r i s u b s t i t u t e d fatty acids (10,87) of the structural form, XVIII. T a i l l a n d i e r et a l . (5) investigated the anti-PTZ a c t i v i t y of a series of 3-substituted and R3 Rj = ^2 = R3 = ^ ^3 j R^ = R3 = CH3, R2 — ^2^5 R2~C—C00H = R3 = CH3, R2 = ^ 3^7 Rl XVIII a,3-disubstituted acids of the type XIX. R2 ^CH-CH-COOH / I R^ R^  i\2 = — ^ 2 n 7* 1^ = ^2 n5 : A H 9 xix R2 = R3 = CH3, RL R2 = R3 = 02U5' D1 R2 - R3 = C3H7, Rj R2 R3 C3H7, RjR2 = R3 = C3H?, R, R2 = R3 — Rl = C3f l 20 As i n most of these s t r u c t u r e - a c t i v i t y s t u d i e s , the PTZ t e s t was per-formed at a s i n g l e dose i n mice, 0.9 mmol/kg i n t h i s i n s t a n c e . According to the investigators (5), isopentanoic acid, 3-ethylpentanoic acid, 2-ethyl-3-propylhexanoic acid and 2,3-dipropylhexanoic a c i d were inactive while 3-propylhexanoic and 2-butyl-3-propylhexanoic acid were as active as valproic acid. Cyclic analogues of valproic acid have also been studied. Cyclo-heptane c a r b o x y l i c a c i d i s i n a c t i v e compared to v a l p r o i c a c i d when administered at a dose of 200 mg/kg i.p. i n mice (11). Both a-cyclo-p e n t y l p e n t a n o i c a c i d and a - c y c l o h e x y l p e n t a n o i c a c i d are weaker anticonvulsants than valproic acid (5). Two recent studies explored the p o s s i b i l i t y of r i g i d a l i c y c l i c carboxylic compounds revealing the nature of the a c t i v e c o n f i g u r a t i o n of the a l k y l groups that i n t e r a c t at the s i t e of a c t i o n of v a l p r o i c a c i d . Brana and co-researchers (5) found compounds, XX and XXI, to be as a c t i v e as v a l p r o i c a c i d i n the s.c. PTZ t e s t when compounds were administered i.p. at a dose of 200 mg/kg i n mice. Scott et a l . (88) looked at the anti-PTZ a c t i v i t y of supposedly m e t a b o l i c a l l y s t a b l e s p i r o c a r b o x y l i c a c i d s , including XXII and XXIII. Compound XXIII, Spiro[4,6]undecane-2-carboxylic a c i d , the most a c t i v e XX XXI 21 OCT OOH C O O H XXIII XXII compound i n the serie s , was s l i g h t l y more potent than valproic acid Early published studies on anticonvulsant a c t i v i t y of valproic acid analogues focused on various derivatives of the carboxylic acid such as esters, amides and ureas (3-9). Most of the compounds tested were barely a c t i v e except the primary amide of v a l p r o i c a c i d which was equally potent. However, dialkylureides have been previously reported to be active compounds (89). The secondary and t e r t i a r y amides of valproic acid were found to be mostly convulsants (3,4). Conversion of the carboxylic acids to the primary alcohols i n a dialkylalkanoic acid series did not reduce d r a s t i c a l l y their anticonvulsant a c t i v i t i e s while e s t e r i f i c a t i o n of the resulting alcohol destroyed the i r a c t i v i t y (5). The anticonvulsant properties of the primary amide and the primary alcohols were partly attributed to the active carboxylic acid metab-o l i t e s (5). Some workers have evaluated analogues with bifunctional groups. 1,3-Propanediols and dicarbamates such as 2,2-dipropyl-l,3-propanediol and meprobamate are known anticonvulsants (38,90). The malonic acid derivative of the spirocarboxylic compound, XXIII, was observed to be s l i g h t l y l e s s a c t i v e than v a l p r o i c a c i d ( 8 8 ) . Schwartz and co-workers (91) reported that 2-ethyl-2-propylcyanoacetamide and 2-ethyl-2-(ED50 of 0.42 mmol/kg compared to 0.9 mmol/kg for sodium valproate). 22 workers (91) reported that 2-ethyl-2-propylcyanoacetamide and 2-ethyl-2-butylmalondiamide possessed an t i c o n v u l s a n t a c t i v i t y w i t h minimal sedative properties at a dose of 400 mg/kg i n the s.c. PTZ test. A number of v a l p r o i c a c i d metabolites with greater p o l a r i t y possessed a n t i c o n v u l s a n t a c t i v i t y weaker than that of v a l p r o i c acid (92). Among the known metabolites of v a l p r o i c a c i d , the unsat-urated metabolites, namely 4-ene VPA, 3-ene VPA and 2-ene VPA were the most potent. However, these metabolites occur i n serum and urine of human and rodents i n r e l a t i v e l y small quantities (<10%) when compared to levels of valproic acid (92). 3-Keto VPA, the hydroxy metabolites and 2-propylglutaric acid had s l i g h t anticonvulsant properties. There are no data on the anticonvulsant p r o p e r t i e s of the r e c e n t l y described metabolites of v a l p r o i c a c i d , 4-Keto VPA and the diunsaturated com-pounds (93). While no r a t i o n a l b a s i s f o r the p r e d i c t i o n of a n t i c o n v u l s a n t a c t i v i t y of a l k y l s u b s t i t u t e d c a r b o x y l i c a c i d d e r i v a t i v e s has been e s t a b l i s h e d , the anticonvulsant data suggest that the nature of the a l k y l group i s of basic importance to pharmacological a c t i o n . The s i g n i f i c a n c e of determining the QSAR of v a l p r o i c a c i d analogues i s highlighted by the experimental observations that v a l p r o i c a c i d , v a l -proamide, d i p r o p y l u r e i d e and 5 , 5 - d i p r o p y l b a r b i t u r i c a c i d are potent anticonvulsants (4,6) despite differences i n l i p i d s o l u b i l i t y . 6. General SAR of A n t i e p i l e p t i c Drugs Although valproic acid d i f f e r s from the t r a d i t i o n a l a n t i e p i l e p t i c s (Figure 2) which contain a n i t r o g e n f u n c t i o n a l group, there are sim-i l a r i t i e s i n structure. Common structural features of the a n t i e p i l e p t i c 23 (i) C3H7, \_ CH—COOH C 3H 7 ' Valproic acid (Carboxylic acid) II o (ii) C2H5 J , CH3-Y Ethosuximide (Succinimide) ' c=o Tn'methadione (Oxazolidinedione) c=o Phenobarbital (Barbiturate) Di phenylhydantoin (Hydantoin) Phenacemide (acetyl urea) Primidone Clonazepam Carbamazepine (Deoxybarbiturate) (1,4-Benzodiazepine) (Dibenzoazepine-5-Carboxamide) Figure 2. Chemical structures of traditional antiepiletic agents. 24 drugs are embodied in the polar carboxyl or imide functional group and the disubstituted quaternary or t e r t i a r y carbon, i and i i i n Figure 2. Anticonvulsant a c t i v i t i e s of classes of compounds have been compiled from the l i t e r a t u r e and summarized to determine SAR among the a n t i -convulsant drugs (94). Similar structural requirements appear to exist in the conventional anticonvulsant drug groups such as the barbiturates, hydantoins, oxazolidinediones, succinimides and a c y c l i c ureides. Certain distinguishing features in the SAR of these a n t i e p i l e p t i c drugs have been recognized. The introduction of at least one phenyl or similar aromatic group at Rj or R2 (Figure 2) i s required for optimal a c t i v i t y against MES-induced seizures. In contrast, a l k y l substituents at Rj or R2 confer high a c t i v i t y against s.c. PTZ-induced seizures. Increasing the length of the a l k y l chain usually maximizes the hypnotic a c t i v i t y . Most of the c l a s s i c a l a n t i e p i l e p t i c drugs produce maximal a c t i v i t y in the s.c. PTZ test when the sum of carbon atoms i n the sub-stituents i s between Cg and C-^ Q. Diphenylacetic acid i s active i n the MES test (16). 5,5-Diphenyloxazolidinedione, 5,5-diphenylhydantoin, 2,2-diphenylsuccinimide and 5,5-diphenylbarbiturate have minimal a c t i v -i t y (or even i n a c t i v i t y ) i n the s.c. PTZ t e s t ( 9 4 ) . 2-Methyl-2-butylsuccinimide i s inactive in the MES test (94). The nitrogen atom may be substituted preferably by a methyl group, lower a l k y l groups and alkoxyalkyl groups without reducing s i g n i f i c a n t l y the anticonvulsant a c t i v i t y , e.g. m e t h a r b i t a l , N-methyldipropyl-acetamide (5,8), mephenytoin, methsuximide and trimethadione (5,8). Rigid analogues of a l k y l substituents have been tested for anticon-vulsant a c t i v i t y . These compounds can a l t e r the s t e r i c properties and metabolism of the f l e x i b l e a l k y l groups. Thus cyclo C^-C^ spiro com-25 pounds of hydantoin (95), oxazolidinediones (95), succinimides (95), carboxylic acids (88) and barbiturates (94) have been studied. There have been no generalizations as to whether certain spirocycloalkyl groups confer desirable pharmacological properties. The 1,4-benzodiazepines and dibenzazepines have more r i g i d struc-tures with d i f f e r e n t s t r u c t u r e - a c t i v i t y requirements. 5-Phenyl substitution i s considered to be necessary for a l l active 1,4-benzo-diazepines (96). Diazepam and carbamazepine (Figure 2) both have l i p o -p h i l i c groups and the t y p i c a l amido structure containing the electron-donor group or hydrogen-bonding group as i n other a n t i e p i l e p t i c drugs. According to Camerman and Camerman (18), the phenyl groups i n phenytoin overlap with those of diazepam when cr y s t a l structures are examined by x-ray crystallography. Other classes of compounds with anticonvulsant a c t i v i t y are the y-butyrolactones (XXIV) and g l u t a r i m i d e s . Klunk et a l . (47) recently found a,a-disubstituted y-butyrolactones to be potent anticonvulsants when examined i n the s.c. PTZ t e s t . Somers (98) has described the anticonvulsant properties of B-methyl-B-butyl g l u t a r i -mide and convulsant p r o p e r t i e s of B-methyl-B-ethylglutarimide (Bemegride®). Klunk and co-workers (97) have found B,B-dialkyl y-butyrolactones such as B-ethyl-B-methyl-y-butyrolactone to be con-XXIV 26 vulsants. Anticonvulsant or convulsant properties may be obtained by minor stru c t u r a l changes i n the side chains. There are structural features that may confer convulsant properties i n the SAR of anticonvulsant drugs (94). Certain benzylethylbarbiturates (94), dibenzylbarbit-urates (94), branched chain alkylethylbarbiturates (17,99), branched chain alkenylalkylbarbiturates (17,99), N-alkylated and di-N-alkylated compounds, e.g. secondary and t e r t i a r y amides of valproic acid (3,4), 1 , 3 , 5 , 5 - t e t r a a l k y l b a r b i t u r a t e s ( 9 4 ) . 2 , 2 , 3 , 3 - t e t r a a l k y l s u c c i n -imides (97) and 5-methyl-l,4-benzodiazepine (30) are known convulsants. According to studies by Andrews et a l . (19) and Downes et a l . (100), the terminal isopropyl and isopropenyl groups i n 5,5-dialkylbarbiturates (Figure 3) appear to be necessary for convulsant a c t i o n . 7. Structural S p e c i f i c i t y of Anticonvulsants A n t i e p i l e p t i c drugs appear to show low structural and b i o l o g i c a l s p e c i f i c i t y . This may be due to the few pharmacophoric groups i n these compounds and thei r a b i l i t y to modulate GABA e f f e c t s . Although few drugs show absolute b i o l o g i c a l s p e c i f i c i t y , most drugs act by i n t e r -action with s p e c i f i c bioreceptors. S p e c i f i c i t y of b i o l o g i c a l action i s determined by the combination and stereochemical arrangement of chemical groups and thei r physical interaction with s p e c i f i c binding s i t e s or receptors. By virtue of thei r nonspecific perturbation of c e l l membranes and lack of structural s p e c i f i c i t y , general anesthetics have been c l a s s i f i e d as s t r u c t u r a l l y nonspecific drugs (101). General anesthetics have been regarded as non-polar, inert and nonionizable compounds which do not 27 ANTICONVULSANTS CONVULSANTS DIFFERENCES double bond (isopropenyl) methyl group (isopropyl) methyl group (isopropenyl) double bond (cyclohexylidene) methylene group (benzyl) Figure 3. Structurally-related convulsant and anticonvulsant barbiturates. 28 have h i g h l y r e a c t i v e polar groups to interact with s p e c i f i c receptors but p a r t i t i o n into c e l l membranes to secondarily disrupt ionic channels i n the membrane structure (101). Most a n t i e p i l e p t i c drugs, however, show sedative e f f e c t s . I t i s g e n e r a l l y assumed that s e d a t i v e , hypnotic and a n e s t h e t i c a c t i o n s of anticonvulsants occur with increasing the dose of the compound. In some compounds, sedative and a n t i c o n v u l s a n t e f f e c t s overlap although phenytoin i s reported to exert i t s anticonvulsant effect at non-sedative doses (42). I t i s known that anticonvulsants are not the only class of compounds with sedative actions. Most CNS-active agents show sedative properties probably because they have s u f f i c i e n t l i p o p h i l i c i t y to pene-t r a t e the blood b r a i n b a r r i e r to i n t e r a c t n o n s p e c i f i c a l l y w i t h b r a i n membranes. Even certain convulsants have been reported to show sedative properties (59). I t may be premature to conclude that valproic acid and analogues act by n o n s p e c i f i c mechanisms as suggested by Perlman and Goldstein (14). These workers i n d i c a t e d that the a b i l i t y of ethanol, v a l p r o i c a c i d and homologous congeners to d i s o r d e r c e l l membranes correlated well with th e i r sedative and anticonvulsant potencies. Some investigators studied other direct physiological and biochemical effects of v a l p r o i c a c i d analogues and a l s o reported good c o r r e l a t i o n w i t h anticonvulsant potencies (12,13,20,81,82). An t i e p i l e p t i c drugs can be described as more s t r u c t u r a l l y s p e c i f i c than anesthetics. Barbiturates are known to show marked s p e c i f i c i t y . Small changes i n the a l k y l side chain of certain anticonvulsant barbit-urates can produce convulsant p r o p e r t i e s (Figure 3). The s e l e c t i v e nature of a n t i e p i l e p t i c drug therapy with anticonvulsants such as pheny-t o i n , p h e n o b a r b i t a l , v a l p r o i c a c i d and ethosuximide a t t e s t to some degree of structural s p e c i f i c i t y . These findings suggest that anticon-29 vulsants may show d i s c r e t e or s e l e c t i v e e f f e c t s on neuronal c e l l structures. Recent advances i n neurophysiological and biochemical techniques have allowed extensive investigations of the selective synaptic or c e l l u l a r actions of some of the a n t i e p i l e p t i c drugs (74,102). Further-more, r a d i o l i g a n d binding s t u d i e s (30) have added support to the suggestion that different subgroups of anticonvulsants e x i s t , each with a different binding s i t e but i n close association and probably each with i t s own struct u r e - a c t i v i t y relationships. The observation of high a f f i n i t y , saturable and i n some cases stereospecific binding of 1,4-benzodiazepines (103), GABA and analogues (30), dihydropicrotoxinin (33) and phenytoin (85) have added weight to proposals of spe c i f i c molecular mechanisms of a n t i e p i l e p t i c and convulsant actions. 1,4-Benzodiazepines have been documented to enhance GABA binding and also vice-versa (30). Barbiturates are known to enhance the binding of benzodiazepines (EC^Q of pentobarbital 100 uM) and GABA(30). Picro-toxin at concentrations of 1 uM i s reported to i n h i b i t the enhancement of benzodiazepine binding by barbiturates (30). Barbiturates and v a l -proic acid are suggested to act on the same s i t e s as convulsants such as picrotoxin, 1,5-alkyl-substituted tetrazoles and tert-butylbicyclophos-phonothionate (30,33,34,56). The convulsant p i c r o t o x i n has been suggested to exert i t s effects at the s i t e c ontrolling the flux of chloride ions, the chloride ionophore (30). H-Dihydropicrotoxinin (DHP) and "^S-tert-butylbicyclophosphothionate (TBPS) have been used to assay picrotoxin binding s i t e s (30,33,56,104). Barbiturates competed with H-DHP binding with r e l a t i v e potencies that correlated with t h e i r a b i l i t y to enhance H-GABA binding (33,104). Pentobarbital, pheno-30 b a r b i t a l and valproic acid are reported to i n h i b i t the binding of H-DHP with I C5 0 of 50 uM, 400 yM and 500 yM respectively (33,34). Valproic a c i d , i n comparison to barbiturates, did not i n h i b i t the binding of H-GABA (87) or H-diazepam (34) at concentrations up to 1 mM. Studies on 35 other anticonvulsants have shown that ethosuximide inhibi t e d S-TBPS binding by 30% at 1 mM while a-ethyl-a-methylbutylrolactone inhibited 35S-TBPS binding by 23% at 0.5 mM (32). Phenytoin inhibited 3H-DHP binding with an I C5 Q of 100 yM (30). The binding of phenytoin to i t s recognition s i t e i s reported to interact with the GABA-BDZ-receptor C l ~ ionophore complex (85). Using spinal cord neuronal culture, Macdonald and McLean (102) indicated that pheny-to i n augmented postsynaptic GABA response at higher concentrations (>8yM) than i n h i b i t i o n of post-tetanic potentiation (high frequency rep e t i t i v e f i r i n g of neurons) which occurred at therapeutic free serum concentrations (4-8 yM). Phenytoin and carbamazepine, which show si g n i f i c a n t effect on post-tetanic potentiation compared to other a n t i -e p i l e p t i c s , are reported to decrease calcium i n f l u x across synaptosomes and a f f e c t s i g n i f i c a n t l y calcium-calmodulin dependent p r o t e i n phosphorylation (105). Figure 4 depicts a model of the GABA receptor-benzodiazepine-chloride ionophore complex developed by various investigators for s i t e s of action of some anticonvulsants and convulsants. The functional coupling between the GABA receptor and the 1,4-benzodiazepine receptor and picrotoxinin s i t e s have given suggestive evidence of a common mech-anism of a c t i o n i n v o l v i n g modulatory e f f e c t s on the i n h i b i t o r y neurotransmitter, GABA. The model also shows the wide variety of struc-tures which interact on each recognition s i t e . Detailed studies have indicated multiple recognition s i t e s for the GABA and benzodiazepine 31 GABA Receptor -GABA -Muscimol -Bicuculline Chloride Ionophore Benzodiazepine Receptor -Picrotoxin -1,4-Benzodiazepines -t-Butylbicyclophosphonothionate -Purines -Anticonvulsant barbiturates -Nicotinamide -Convulsant barbiturates -B-Carbolines -Valproic acid -Ethosuximide -Convulsant tetrazoles -Purines Figure 4. Model of GABA-benzodiazepine-receptor chloride ionophore complex 32 receptor complex (30,103). A recent study (107) i n v e s t i g a t e d the 35 k i n e t i c s of i n h i b i t i o n of S-TBPS by barbiturates for evidence that there i s one class of binding s i t e s for picrotoxin and barbiturates. I t was suggested that picrotoxin/TBPS binding s i t e s are a l l o s t e r i c a l l y linked to barbiturate s i t e s . There are also reports of differences between a l i p h a t i c - s u b s t i t u t e d b a r b i t u r a t e s and phenyl-substituted barbiturates i n thei r action to augment GABA responses (31,108). 8. Pharmacokinetics of Valproic Acid a. Human V a l p r o i c a c i d i s r a p i d l y and nearly completely absorbed following oral c l i n i c a l doses of 15-60 mg/kg with peak blood levels occurring within 1-4 hr (109). The elimination h a l f - l i f e i s i n the range of 6-18 hr, with shorter h a l f - l i v e s obtained i n the presence of a n t i e p i l e p t i c drugs which induce the metabolism of valproic acid (109). Total plasma clearance i s 0.07-0.14 ml/min/kg with an apparent volume of d i s t r i b u t i o n of 0.1-0.4L/kg (109-111). Plasma protein binding of the drug at therapeutic plasma concentrations (50-100 mg/L) ranges from 80-95% (111). Protein binding s i t e s become saturated at concentrations greater than 80 mg/L (111,112). Plasma clearance of VPA i s dependent on free f r a c t i o n and i n t r i n s i c clearance. These p h a r m a c o k i n e t i c parameters i n d i c a t e t h a t VPA di s t r i b u t i o n i s limited mostly to ext r a c e l l u l a r f l u i d s with only minor tissue uptake. Such d i s t r i b u t i o n a l characteristics are attributed to the low pKa of VPA and the high plasma protein bind-33 i n g . Studies on single dose and multiple dose k i n e t i c s of VPA by Acheampong et a l . (113) and Bowdle et a l . (114) indicate increased plasma clearance at higher doses due to an increase i n free f r a c -t i o n . There i s a tendency for a decrease i n i n t r i n s i c clearance at higher doses probably as a result of saturation of metabolism or autoinhibition of metabolism (113,114). Rodents After oral administration of 200 mg/kg i n mice, valproate i s reported to be rapidly absorbed with maximum serum concentrations reached within 5-15 min (70). There are quantitative differences i n pharmacokinetic properties of the drug i n rodents compared to human, apparently due to differences i n protein binding. The hal f -l i f e of v a l p r o i c a c i d i s 0.8 hr i n mouse and 1-4 hr i n rats (70,115). Volume of d i s t r i b u t i o n i s about 0.66 L/kg i n rat and 0.33 L/kg i n mice with t o t a l body clearance of 4.17 ml/min/kg i n rats and 4.33 ml/min/kg i n mice (110). At plasma concentrations of 50-80 ug/mL, the plasma protein-binding of the drug i s about 90% in human, 63% i n rat and 12% i n mice (110). Kinetics of Valproic Acid i n the CNS Valproic acid has been shown to enter the CNS rapidly with maximal brain concentrations reached at 5-10 min i n mice (70). Several studies have demonstrated that regional d i s t r i b u t i o n of the drug i n the b r a i n of mice and r a t s i s r e l a t i v e l y homo-genous (116,117). Analysis of discrete brain areas, however, showed g r a d u a l a c c u m u l a t i o n of the drug i n o l f a c t o r y b u l b s ( 1 1 6 , 1 1 7 ) . V a l p r o i c a c i d does not b i n d t o b r a i n 34 tissues (118). Studies on the subcellular d i s t r i b u t i o n of the drug by Aly and Abdel-Latif (119) indicated that i t i s mainly associated with the soluble and mitochondrial f r a c t i o n s . I t has also been documented that the drug i s rapidly cleared from the brain (120). The rapid brain uptake of valproic acid i s suggested to be due to an a c t i v e - t r a n s p o r t mechanism (120) or rapi d d i s s o c i a t i o n of protein-bound drug within the brain c a p i l l a r i e s (121). Human brain valproic lev e l s were found to be 7-28% of levels i n plasma (122). In mice, brain drug concentrations (5-60 mg/L) were found to be about 15-20% of those i n serum (20-250 mg/L) after oral admin-i s t r a t i o n of 200 mg/kg of sodium valproate (70,117). 9. Metabolism of Valproic Acid Metabolism i s the major route of valproic acid elimination i n human and rodents (123). Several metabolites have been i d e n t i f i e d i n a number of metabolism studies [reviewed i n (123)]. These studies indicate four main metabolic pathways, namely g l u c u r o n i d a t i o n , B - o x i d a t i o n , w-oxidation and (w-l)-oxidation. The metabolic pathways are summarized i n Figure 5. Glucuronidation and B-oxidation are the major pathways. The major metabolites i n human are the glucuronide and 3-keto VPA (123). 3-Keto VPA and 2-ene VPA are major plasma metabolites i n human, rats and mice (115). Other metabolites that occur i n s i g n i f i c a n t quantities i n serum of patients are 2,3'-diene VPA, 4-0H VPA and 5-0H VPA. Acheampong et a l . (93) have i d e n t i f i e d a new metabolite, 4-keto VPA i n human serum and urine. The urinary compounds, 2-propylsuccinic acid and 2-propyl-malonic acids were confirmed as valproic acid metabolites by stable-isotope and GCMS techniques (93). 35 CH,—CH,—CH OH CH 2-CH 2-CH 2 CH 3-CH 2-CH 2 5-OH VPA HOOC-CH2—CH2 ;CHCOOH CHj-CH 2-CH 2 I 2-Propvlglutarlc a d d 1 1 CHj™ CH^ **"" CH^ "™ C H ^ COOH COOH CH3—C—CH2 / CH3-CH2—CH2 4-Keto VPA CHCOOH 1 HOOC-CH, CH3—CH2—CH2 CHCOOH CH2-=CH-CH2 CHCOOH / CH 3-CH 2-CH 2 4-eneVPA CH,-CH-CH C H j - C H ^ C H ^ CHCOOH 3-ene VPA CH,—CH,—CH v 3 2 ^ CH ; J-CH 2-CH 2/ 2-ene VPA C—COOH i / 0 H C H 3 - C H 2 - C H X CH-COOH CH3—CH2—CH2 3-OH VPA C „ _ C H2- C CH 3-CH 2-CH 2 / CHCOOH CH 2~CH-CH ? CHpCH-CH^ 4,4'-d1ene VPA CHCOOH CHpXH-CH C-COOH (E) 2,4-dl ene VPA CH,—CH=CH COOH CH—CH2— CH 2(E).3'(E)-d1ene VPA 2-Propylmalonlc a d d 2-Propylsucdnic acid 3-Kcto VPA Figure 5. Metabolic pathways of valproic acid (VPA). There have been more recent studies which sought to further d e l i n -eate the metabolic pathways of the drug, with emphasis on the id e n t i t y of d i u n s a t u r a t e d m e t a b o l i t e s d e t e c t e d i n human serum and urine (124,125). Only two diene m e t a b o l i t e s , 2-[2'-propenyl]-4-pentenoic a c i d and 2-propyl-(E)-2,4-pentadenoic a c i d have been i d e n t i f i e d (128,129). I d e n t i f i c a t i o n of the major diene metabolite has been e l u s i v e , p r i m a r i l y due to m u l t i p l i c i t y of the p o s s i b l e diunsaturated acid isomers and also the u n a v a i l a b i l i t y of synthetic reference material. 10. Toxicity of Valproic Acid Nausea, vomiting, abdominal cramps and diarrhea are the most commonly reported side effects (35). Drowsiness and sedation occur, especially at high doses and with combination therapy (35). Hyper-ammonemia and hyperglycinemia have been observed to occur during valproic acid therapy (127). Side effects associated with the drug have generally been mild (35). However, i n more recent years more serious side effects have been observed. These include thrombocytopenia (128), pancreatitis (126) and hepatic damage (128-131). The potential hepatic t o x i c i t y i s of major concern. In most serious cases, l e t h a l effects of valproic acid due to hepatic f a i l u r e have been variously reported (128-130). The proposed mechanism for the hepatic t o x i c i t y and/or the Reye-l i k e syndrome (131) induced by valproic acid include direct toxic effect of a metabolite, possibly 4-ene VPA which i s a structural analogue of the hepatotoxin, 4-pentenoic acid (130,132). 11. A Quantitative Structure-Activity Model One appropriate method of examining the relationship between struc-tures of closely-related congeners and anticonvulsant a c t i v i t y was to 37 determine the quantitative relationship between structural properties or s u b s t r u c t u r a l d e s c r i p t o r s and pharmacological a c t i v i t y . Hansch and collaborators (25,29) have a p p l i e d the l i n e a r free energy relationship approach to i n v i t r o and i n vivo conditions where physicochemical i n t e r -a c t i o n s between drugs and biomacroraolecular s i t e s of a c t i o n can be q u a n t i t a t i v e l y expressed i n terms of r e g r e s s i o n equations i n v o l v i n g hydrophobic (TT), electronic (o), s t e r i c (Eg) and other structural para-meters. The Hansch method models the dynamic s i t u a t i o n i n a b i o l o g i c a l system where a drug i s applied at a dose C and proceeds through several biophases to reach the t a r g e t s i t e w i t h p r o b a b i l i t y of access, A and effects response reaction with a rate or equilibrium constant, Kx« Thus the r a t e of b i o l o g i c a l response at the r e a c t i o n s i t e can be expressed mathematically as d(response)/dt = ACKX ( i ) For most drugs the transport process or probability of reaching the s i t e of a c t i o n i s determined l a r g e l y by l i p i d s o l u b i l i t y or p a r t i t i o n co-e f f i c i e n t (TT or log P), i.e. A = f Or) ( i i ) On the b a s i s of an e m p i r i c a l p a r a b o l i c r e l a t i o n s h i p between a drug's b i o l o g i c a l a c t i v i t y and p a r t i t i o n c o e f f i c i e n t , Hansch (25) expresses the function i n the form of a normal Gaussian d i s t r i b u t i o n . 38 _ L _e -(TT-TT0)2/2S2 f = S/2TT . ( i i i ) From equations ( i ) , ( i i ) , ( i i i ) 2 d(response)/d t = AE-(TT-7r0) /b.C.Kx ( i v ) 1 where a = s/2-n, b = 2s , s = standard deviation parameter At a fixed time i n t e r v a l after administration of equieffective doses C (e.g. EDIJQ), a constant b i o l o g i c a l response i s determined under steady s t a t e conditions, i.e. 2 d (response) = constant = ae"^71-7^) /°.KXC dt Taking logarithms, 2 log (constant) = log [ae-(TI'-'n'o) /D«C.KX] and log 1/C = -nlTr2 + n2Tr.Tr0 - n3TT02 + log Kx + 114 (v) The Hansch method r e l i e s on the physicochemical interaction phenom-enon at the c r i t i c a l s i t e of a c t i o n which i s determined by the hydrophobic (IT), electronic (s) and s t e r i c (Eg) factors. log Kx = kTT + k30 + k4E s . . . . . ( v i ) Thus 39 log 1/C = k-j^-rr + k27T + k^O + k^Eg + ks ( v i i ) Equations (vi) and ( v i i ) are linear free energy relationships showing the additive combination of the free-energy related parameters. 2 The f i r s t two terms i n TT describe the transport process. The TT v a r i -able i n equation ( v i i ) i s generally agreed to be essential i n complex b i o l o g i c a l systems, for instance i n whole animal studies. In simpler 9 b i o l o g i c a l systems TT may not be required and TT describes hydrophobic effects (25). On the basis of the additive-constitutive properties of pa r t i t i o n c o efficient (P) revealed by Hansch (133), equation ( v i i ) can also be expressed as log P = ZTT or log P (R-X) = TT(R) + TT(X) L i p o p h i l i c i t y can be considered as the r e l a t i v e a f f i n i t y of a drug for the l i p i d biophase and i s quantitatively defined i n terms of p a r t i t i o n c o e f f i c i e n t (P) which c h a r a c t e r i z e s the e q u i l i b r a t i o n between the aqueous and l i p i d phases. 12. Physicochemical Parameters Used i n QSAR Steric and electronic effects i n drug structures are not as well defined as l i p o p h i l i c i t y . Examination of the structures of valproic acid analogues i n Figure 2 shows that the l i p o p h i l i c parameter describes the hydrophobic character of the a l k y l groups, the electronic parameter evaluates the electronic properties of the polar moiety and the sub-log 1/C = kx log P + k2 (log P )2 + k3a + k4Es + k5 ( v i i i ) where 40 s t i t u t i o n pattern or s t e r i c e f f e c t s of the a l k y l chain can be characterized by the s t e r i c parameter. Since the regression equation ( v i i i ) i s a linea r free energy r e l a t i o n s h i p , other free-energy related parameters can be used. The electronic parameters used frequently i n QSAR are the Hammett's substitution constant ( a * ) , pKa, ApKa and dipole moment (u). Ionization constant (pKa) i s generally considered as a fu n c t i o n of polar e f f e c t s d e s c r i b i n g the electron-withdrawing or electron-releasing effects (134). There are various methods for the determination of pKa values of acidic compounds. These include potentiometric t i t r a t i o n methods, spectrophotometry methods, and conductivity methods. For many drugs such as valproate analogues, s o l u b i l i t y and UV absorption character-i s t i c s appear to determine the methods of choice. In one study (69), the pKa of valproic acid was determined by potentiometric t i t r a t i o n using acetone-water media. Dipole moments of a number of anticonvulsant drugs have also been calculated by measurement of the d i e l e c t r i c con-stant and molar refraction of the drug solutions and solvents (135). L i p o p h i l i c i t y of compounds i s either determined experimentally or c a l -culated using substituent constants. 13. Hydrophobic Parameters a. Determination of l i p o p h i l i c i t y by the shake-flask procedure Determination of the l i p o p h i l i c i t y of a drug usually requires measurement of the equilibrium constant or p a r t i t i o n c o e f f i c i e n t between the non-polar and aqueous phases using shake-flask pro-cedures. The choice of a model for the l i p i d phase i n biomembranes has been a r b i t r a r y , ranging from highly nonpolar solvents to mod-41 erately nonpolar solvents (133). Octan-l-ol i s the most frequently used nonpolar solvent. Apart from preferred advantages of octanol i n terms of a v a i l a b i l i t y , purity and bifunctional properties (133), octanol i s probably as appropriate as other nonpolar solvents for modelling biomembranes. Collander (136) has showed that p a r t i t i o n values can be converted between organic solvent-water systems. I t has been found, however, that for a successful conversion, con-sideration ought to be given to differences i n hydrogen-bonding properties between octanol and other organic solvents as well as hydrogen-donor or acceptor properties of the solutes (133,136). b. HPLC determination of l i p o p h i l i c i t y The t r a d i t i o n a l shake-flask procedure of determining octanol-water p a r t i t i o n c o e f f i c i e n t (log P) has been observed to have many ana l y t i c a l problems (137-141). I t i s tedious and time-consuming. S o l u b i l i t y d i f f i c u l t i e s i n aqueous media, i n s t a b i l i t y of compounds i n l i q u i d phases and loss of material by adsorption or other mech-anisms have been a major problem. Alternative methods, especially chromatography methods (TLC, HPLC) that provide a rapid and accur-ate estimate of l i p o p h i l i c i t y have been sought. In recent years, reverse-phase (RP)-HPLC, with a hydrophobic column consisting of s i l i c a p a r t i c l e s coated with covalent-bonded C-18 and C-8 a l k y l chain, has been used to study hydrophobic effects (137-143) and st r u c t u r e - a c t i v i t y relationships (144). The reverse-phase HPLC method i s based on the determination of retention parameters which are then correlated with log P values. I t i s also required that the retention parameter be reproducible 42 and accounted for by a mechanism to r e f l e c t only p a r t i t i o n i n g behaviour of solute i n l i q u i d phases. The line a r relationships reported between log P and capacity factor of compounds found by numerous investigators (137-150) appear to be a close p a r a l l e l to the l i n e a r regression equation found by Collander (136) to exi s t between p a r t i t i o n c o e f f i c i e n t s determined i n two different nonpolar solvent-aqueous part i t i o n i n g systems. The unique features of the monomeric bonded-phase chromato-graphic column i s the presence of a nonpolar C-18 or C-8 hydrocarbon chain chemically bonded to the s i l i c a support. The various methods for attaching a covalently-bonded phase to the porous or p e l l i c u l a r s i l i c a support rely on the conversion of surface s i l a n o l groups to s i l i c a t e esters (-Si-OR), aminosilanes (-Si-N-) and s i l o x a n e s (-Si-O-Si-R^). The most widely used chemically bonded stationary phases are those based on siloxanes r e l a t i v e l y stable against hydrolysis within the pH range of 2-8 (151). Siloxane bonded phase packings are commercially available with microparticular supports and a monomeric C-18 organic layer such as Hypersil ODS used i n t h i s study. Different reverse phase HPLC procedures of determining log P0/w have emerged over the years of i t s appl i c a t i o n . Both un-treated (138,140,141) and s i l y l a t e d (140) octadecylsilane reverse phase columns have been used as the stationary phase. In a d i f f e r -ent procedure, Mirrlees et a l . (139) coated s i l y l a t e d s i l i c a with a layer of n-octanol and used n-octanol-saturated water as the mobile phase. This was suggested to be closely analogous to the octanol-water p a r t i t i o n i n g system. Unger and co-workers (137) used These phases are known to be 43 octanol-coated C-18 bonded s i l i c a as the stationary phase and octanol-saturated phosphate buffer as the eluent. HPLC methods have been used to demonstrate good correlations between capacity factors and log PQ/W values of straight-chain alkylcarboxylic acids (145,146). In a study of long chain n-a l i p h a t i c a c i d s (C7-C20), d'Amboise and Hanai (145) reported excellent correlations with unacidified 50% acetonitrile-water as the mobile phase among different compositions of methanol, aceto-n i t r i l e and tetrahydrofuran. Tanaka and Thorton (153) studied the hydrophobic effects of various carboxylic acids using a u-Bondapak C^g-coluran and methanol-O.OlM sodium phosphate buffer (pH 3.0-3.5) of different compositions. To suppress ionization (low pH) and minimize adsorption (NH^+) of a series of aromatic acids i n one study (147), 50% methanol-ammonium phosphate (pH 2.15) and Hypersil ODS was used to determine the capacity factors. In t h i s study the HPLC method i s used to determine the capacity f a c t o r s of the compounds as well as the log PQ/W values. This requires use of reference compounds and optimization of the a n a l y t i c a l procedure to be sensitive to structural hydrophobic effects of varied carboxylic acids and tetrazoles. The log P values obtained i n the HPLC method are evaluated i n terms of accuracy by comparison to results from the shake-flask procedure. Although the nature of the stationary phase during the chrom-atographic separations and the mechanism of r e t e n t i o n i n chemically-bonded phases i s s t i l l not completely understood, a partition-adsorption mechanism has been postulated by various i n -vestigators (148-150) for the retention of solutes. According to 44 the proposed mechanism, the stationary phase i s a combination of C-18 hydrocarbon layer, residual s i l a n o l present on the s i l i c a sur-face and associated solvent molecules from the mobile phase. The solute molecules are suggested to adsorb on a modified s i l i c a surface with an adsorbed layer of mobile phase components or part-i t i o n into a l i q u i d phase formed by the s i l i c a bonded-C^g layer with associated solvent molecules from the bulk eluent. The application of l i q u i d chromatography to measurement of physicochemical properties such as log P has been based on chrom-atographic theory. A theoretical treatment of a pure p a r t i t i o n process i n r e l a t i o n to the r e t e n t i o n parameter by Snyder and Kirkland (151) showed, i n the equilibrium s i t u a t i o n , the r e l a t i o n -ship between the observed chromatographic retention parameter and the d i s t r i b u t i o n c o e f f i c i e n t k» = K_Vs Vm where k' i s the capacity factor, Vm i s the t o t a l volume of solvent within the column, Vg i s the t o t a l volume of the stationary phase and K i s the d i s t r i b u t i o n c o e f f i c i e n t which defines the equilibrium d i s t r i b u t i o n of solute between the stationary phase and the mobile phase. c. Determination of l i p o p h i l i c i t y using substituent constants Other methods of estimating log P values are the Hansch TT method (133,152) and Rekker fragmental method (153). Both methods are based on the additive-constitutive properties of p a r t i t i o n co-e f f i c i e n t of compounds, but the approaches are d i f f e r e n t . Hansch TT 45 values are calculated from the r e l a t i o n , TTx = log P (R-X) - log P (R-H). The Hansch method usually requires the octanol-water p a r t i t i o n c o e f f i c i e n t of the parent compound (R-H). Rekker adopted an empirical approach based on the r e l a t i o n log P (R-X) = Z anfn or log P (R-X) = f(R) + f(X) Thus f i s the fragment constant for the molecular fragment, n and an i s the numerical factor indicating the number of times a given fragment n appears i n the structure. The fragment constants are obtained from s t a t i s t i c a l data reduction procedures (153) i n which contributions from the molecular fragment, n, i n various molecular structures are averaged. Thus, the line a r regression method has an as the independent parameter, log P as the dependent variable and the fn values are determined as the regression c o e f f i c i e n t . One result of the Rekker estimation method i s the prediction that the hydrogen atom i n a molecular structure contributes s i g n i f i c a n t l y to the p a r t i t i o n c o e f f i c i e n t . Thus compared to Hansch Tr-values, f-values d i f f e r for C, CH, CH2 and CH3. 46 EXPERIMENTAL A. Chemicals and Materials 1. Synthesis Chemicals were reagent grade and procured from the f o l l o w i n g sources. a. Aldrich Chemical Co. (Milwaukee, Wisconsin) Aluminium chloride (anhydrous), n-Butyl bromide, n-Butyl-lithium (1.6M i n hexane), Calcium hydride, Cyclohexylmethylbromide, Deuterochloroform (gold l a b e l ) , Diisopropylamine, 2-Ethylbutyric a c i d , 2-Ethylhexanoic ac i d , 2-Ethoxyethanol, n-Heptyl bromide, Hexamethylphosphoramide, I s o b u t y r i c a c i d , Lithium aluminium hydride, Magnesium sulfate (anhydrous), N-Methyl-N-nitroso-p-toluenesulfonamide (Diazald®), 1-Methylcyclohexanecarboxylic ac i d , Pentan-2-one, (E)-2-Pentenoic ac i d , Potassium hydride (35% o i l dispersion), n-Propyl bromide, n-Propyl iodide, Sodium hydride (50% o i l dispersion), Tetrahydrofuran, Triethylamine. b. B r i t i s h Drug House (Poole, U.K.) Diethyl malonate, Diethylamine, Pyridine, Sodium cyanide. c. Eastman Kodak Co. (Rochester, New York) Sec-Butyl alcohol, l-Bromo-3-methylbutane, Di-n-Butylamine, Ethyl acetoacetate, Methanesulfonyl chloride, Propionaldehyde, Sodium azide, Succinic anhydride. 47 d. Fisher S c i e n t i f i c Co. (Fairlawn, New Jersey) Bromine, t-Butanol, Hydrogen bromide (48%), Potassium cyanide, Quinoline. e. Mallinkrodt Chemicals (St. Louis, Missouri) Ethyl bromide, Ethyl iodide, Potassium carbonate (anhydrous), Sodium bicarbonate, p-Toluenesulfonyl chloride. f . Matheson Coleman and B e l l Co. (Norward, Ohio) Dimethylsulfoxide (anhydrous), Phosphorus pentoxide, Phos-phorus tribromide. 2. Thin-layer Chromatography Glass plates (20 x 20 cm) - CAMAG, B e r l i n , Germany. Ether solvent - USP grade Petroleum ether, 30°-60°C - USP grade Benzene - USP grade S i l i c a gel G - E. Merck, Darmstadt, FR Germany Sulfu r i c acid (96%) - B r i t i s h Drug House Si l v e r n i t r a t e - Nichols Chemical Co., Vancouver, B.C. TLC Streaking Apparatus - Applied Science Lab Ltd., State College, Pennsylvania Spreader - Desaga, Heidelberg, FR Germany 3. High-Performance Liquid Chromatography Norganic cartridges, membrane f i l t e r s (for preparation of HPLC 48 grade water from deionized d i s t i l l e d water) - M i l l i p o r e Corporation, Bedford, Massachusetts. Sodium dihydrogen orthophosphate, monohydrate - B r i t i s h Drug House (Canada) Ltd., Toronto, Ontario. 2-Ethylbutyric a c i d , 2-Ethylhexanoic a c i d , Trimethylacetic a c i d , n-Valeric acid - Aldrich Chemical Co. n-Heptanoic ac i d , n-Hexanoic acid - Eastman Kodak Co. n-Butyric a c i d , Octanoic acid - N u t r i t i o n a l Biochemicals Corp-oration, Cleveland, Ohio. 4. Potentiometric Titrimetry Potassium hydrogen phthalate - B r i t i s h Drug House. Potassium hydroxide - American S c i e n t i f i c and Chemical Co., Seattle, Washington. Phosphate pH 7.0 buffer - VWR S c i e n t i f i c Inc., San Francisco, C a l i f o r n i a . Potassium dihydrogen orthophosphate - Mallinkrodt Chemicals. 5. Gas Chromatography - Mass Spectrometry 3% Dexsil 300 on 100/120 Supelcoport - Supelco Inc., Bellefonte, Pennsylvania. Fused s i l i c a c a p i l l a r y column (25 m x 0.3 mm i.d.) coated with SE-54 - Hewlett Packard. 49 Fused s i l i c a c a p i l l a r y column (25 m x 0.3 mm i.d.) with bonded 0V-1701 - Quadrex Co. t - B u t y l d i m e t h y l s i l y l c h l o r i d e - Applied Science Lab., State College, Pennsylvania. N-Methyl-N-trimethylsilyltrifluoroacetamide, trimethylanilinium hydroxide (0.2M i n methanol) - Pierce Chemical Co., Rockford, I l l i n o i s . 2-Propyl-(E)-2,4-pentadienoic acid was a g i f t from Dr.T.A. B a i l l i e , School of Pharmacy, University of Washington, Seattle, Washington. 6. Pharmacological Testing Pentylenetetrazole - Aldrich Chemical Co. Sodium chloride - B r i t i s h Drug House. Hydrochloric acid - American S c i e n t i f i c and Chemical Co. B. Instrumentation 1. Nuclear Magnetic Resonance Spectrometry Proton NMR spectra were recorded on a Bruker WP-80, Varian XL-100, Nicolet-Oxford-270 or Bruker WH-400 spectrometer at the NMR f a c i l i t y i n the Department of Chemistry, U.B.C. Spectra were taken with CDClg as solvent and tetramethylsilane as an internal standard. 2. Infra Red Spectrometry IR spectra were obtained with sodium chloride disks either as l i q u i d films or nujol mulls on a Unicam SP1000 spectrometer. 50 3. U l t r a v i o l e t Spectrometry UV spectra were recorded i n 50% methanol on a Beckman Model 24 UV-v i s i b l e spectrophotometer. 4. Gas Chromatography Mass Spectrometry (GCMS) a. Packed Column GCMS analysis was performed on a Hewlett Packard 5700A gas chromatograph interfaced to a Varian MAT-111 mass spectrometer v i a a variable s l i t separator. Electron impact mass spectra were recorded at 70eV, ion source pressure of 5.0 x 10 Torr and emis-sion current of 300yA. Computerized background subtractions were made to plot mass spectra. The scan range was 5 to 500 daltons with one scan taken every 5 sec. Total ion chromatographic plots were based on m/z 50 to 500. Mass chromatograms were plotted i n scan mode. The data were processed by an on-line Varian 620L com-puter system. GCMS analysis was carried out under the following conditions: 3% Dexsil 300 column (1.8 ra x 2 mm i . d . ) ; Oven temperature, 50°C to 280°C at 8°C/min; Helium (c a r r i e r gas) flow rate, 25 mL; Injection port temperature, 250°C; Separator temperature, 250°C; Inlet l i n e temperature, 250°C. b. Fused-silica Capillary Columns Capillary GCMS analysis was done on a Hewlett Packard 5987A instrument equipped with an HP gas chromatograph, mass spectrometer and on-line data system. Electron impact mass spectra were record-ed at 70eV, ion source pressure of 2.0 x 10~^ Torr and emission 51 current of 300 uA. The gas chromatograph was interfaced to the mass spectrometer via an open-split interface. GCMS analysis of t-BDMS derivatives was performed under the following conditions: Dimethylsilicone column (12.5 m x 0.2 mm i.d.), 0V-1701 column (25 m x 0.3 mm i.d.); Oven temperature, 50°C to 100°C at 30°C/min, 100OC to 260OC at 8QC/min; Helium flow rate, 1 mL/min; Splitless mode of injection; Source temperature, 200°C; transfer line temperature, 240°C; injection port temperature, 240°C. C. Synthesis of Alkyl Carboxylic Acids, Tetrazoles and Succinamic Acids  1. Synthesis of Alpha-Substituted Aliphatic Acids a. Synthesis of 2-Butylhexanoic Acid (II) A flame-dried 500 mL three-necked flask, equipped with a graduated separatory funnel with septum inlet and reflux condenser connected to a mercury bubbler, was immersed in an ice-water bath. The reaction vessel was flushed with nitrogen and maintained under a nitrogen atmosphere throughout the reaction. Diisopropylamine (0.13 mol) in 100 mL anhydrous THF was placed in the flask. After cooling the mixture to 0°C, n-butyllithium in hexane (81 mL of 1.6 M, 0.13 mol) was added dropwise and with stirring. The mixture was stirred for 20 min and then n-hexanoic acid (0.055 mol) added at a slow rate to maintain the reaction temperature below 0°C A milky white precipitate formed and then HMPA (0.13 mol) in THF was added. The resulting homogenous mixture was stirred at 0°C for an additional 20 min. n-Butylbromide (0.055 mol) was added rapidly to 52 the dianion of hexanoic acid, during which time the reaction temp-erature approached room temperature. The reaction was allowed to proceed f o r a f u r t h e r 90 min before being quenched wi t h 25% HC1 (80 mL). The organic l a y e r was separated and the aqueous l a y e r extracted twice with petroleum ether, 30°-60°C (100 mL portions). The combined organic l a y e r s were washed i n turn w i t h 10% HC1, saturated NaHC03 and saturated NaCl s o l u t i o n . The organic l a y e r was dried over anhydrous Na2S04» f i l t e r e d and the solvents removed with a rotary evaporator. The residue was then subjected to frac-t i o n a l d i s t i l l a t i o n under reduced pressure to obtain 2-butyl-hexanoic a c i d (70% y i e l d ) , bp 96O-98OC/0.35 mm [ L i t (154), 89°C/0.1 mm]. NMR Spectrum: 6 0.7-1.0 ( t , 6H, 2CH3), 1.1-1.8 (m, 12H, 6CH2>» 2.1-2.5 (m, 1H, CH), 8.3-10 (broad s, 1H, COOH). Synthesis of Valproic Acid (I) V a l p r o i c a c i d was prepared, i n a s i m i l a r fashion to that described above for 2-butylhexanoic acid, by reaction of n-valeric acid (0.084 mol) with n-propylbromide. The f i n a l product, valproic a c i d , was separated from unreacted v a l e r i c a c i d by f r a c t i o n a l d i s t i l l a t i o n [bp of v a l p r o i c a c i d 110°-ll2°C/1.4 mm; l i t -erature (69) bp 221°C at 760 mm]. NMR Spectrum: 60.8-1.1 ( t , 6H, 2CH3), 1.1-1.8 (m, 8H, 4CH2), 2.2-2.5 (m, 1H, CH), 8.0-10 (broad s, 1H, COOH). Synthesis of 2-Propyl-(E)-2-Pentenoic Acid (VII) Valproic acid (0.15 mol), synthesized as described above, was placed i n a 250 mL flask equipped with a reflux condenser the top 53 of which was connected to a gas absorption d e v i c e . Bromine (0.16 mol) was added into the flask followed by 0.7 mL of phos-phorus tribromide. The mixture was s t i r r e d and heated with an o i l bath at 70°C for 30 min and then at 100°C for 4 hours u n t i l a l l the bromine had reacted. The reaction mixture was then d i s t i l l e d under reduced pressure using a water pump to remove residual hydrogen bromide. The d i s t i l l a t i o n assembly was then connected to an o i l pump and the f r a c t i o n containing 2-bromovalproic acid and 2-propyl-2-pentenoic acid collected at 70°-80°C/0.01 mm. The acids were converted to the ethyl ester derivatives by refluxing a mixture of the acids for 12 hours i n an excess of ethyl alcohol i n benzene and catalyzed by a small amount of concentrated s u l f u r i c a c i d . A Dean-Stark apparatus was used to separate out the water. The isolated ethyl ester derivatives and quinoline were placed i n a flask connected to a Vigreux column set for downward d i s t i l l a t i o n . The reaction mixture was heated to 160°C with s t i r -ring during a 15 min period, after which heating was increased rapidly to d i s t i l l the resulting 2-propyl-2-pentenoate. The fra c -tion with bp 188°-196°C was collected and puri f i e d by washing with d i l u t e s u l f u r i c a c i d . The unsaturated ester was hydrolyzed and the unsaturated acid obtained after a c i d i f i c a t i o n . The unsaturated acid was extracted with diethyl ether. The ether extract was washed with water, dr i e d with anhydrous Na2S0^, f i l t e r e d and solvent removed with a r o t a r y evaporator. The product was re c r y s t a l l i z e d from a chloroform solution by keeping the solution at -20°C f o r a prol o n g e d p e r i o d . A f t e r two s u c c e s s i v e r e c r y s t a l l i z a t i o n steps, pure 2-propyl-(E)-2-pentenoic acid was 54 obtained, mp 32°C [ l i t . (155) mp 35°C]. These p u r i f i c a t i o n steps were necessary to remove small amounts of the low-melting Z-isomer. NMR Spectrum: 60.8-1.1 (m, 6H, 2CH3), 1.2-1.7 (m, 2H, CH^-CH?-CH2), 2.1-2.5 (ra, 4H, CH^-CH = and CH2-C=), 6.8-7.1 ( t , 1H, - CH = C, trans). d. Synthesis of 2-Propyl-4-oxopentanoic acid (IX) 2-Bromopentanoic acid was synthesized, i n a s i m i l a r fashion to that described above for synthesis of 2-bromovalproic ac i d , by reaction of pentanoic acid with bromine i n the presence of a c a t a l -y t i c amount of phosphorus t r i b r o m i d e . The crude product was d i s t i l l e d to give 2-bromopentanoic a c i d , bp 102°-105°C/2.5 mm. Mass spectrum: m/z 55 (100%), 138 (30%), 140 (28%), 27 (20%), 41 (18%), 29 (15%), 94 (12%), 43 (10%). The e t h y l e s t e r of 2-bromopentanoic a c i d was prepared by refluxing a mixture of 2-bromopentanoic acid (0.43 mol), ethanol (1.0 mol), benzene (100 mL) and concentrated s u l f u r i c acid (1.5 mL) for 12 hour using a Dean-Stark water separation u n i t . Pure ethyl 2-bromopentanoate was obtained by d i s t i l l a t i o n after the usual work-up. Bp 60°-62°C/3.0 mm. Mass spectrum: m/z 29 (100%), 166 (26%), 168 (24%), 101 (22%), 140 (12%), 138 (10%). For synthesis of 2-propyl-4-oxopentanoic a c i d , anhydrous THF (100 mL) was placed i n a 250 mL f l a s k and sodium hydride (0.084 mol) added. Ethyl acetoacetate (0.08 mol) was added drop-wise over 30 min and the mixture s t i r r e d for an additional 10 min. Ethyl 2-bromopentanoate (0.08 mol) was added drop by drop and the solution refluxed for 5 hours. D i s t i l l e d water (30 mL) was added and the resulting mixture f i l t e r e d under suction. The organic 55 layer was separated and the aqueous phase extracted twice with ether. The combined organic layer was dried over Na2S0^, f i l t e r e d and ether removed by a rotary evaporator. The residue was d i s -t i l l e d to give ethyl 2-propyl-3-acetylsuccinate (75% yield) bp 115°C/0.2 mm. A mixture of the acylsuccinate (0.018 mol) and concentrated HC1 (0.18 mol) was heated under reflux for 10 hours to effect hydrolysis and decarboxylation. The product was extracted three times with ether and the ether extract dried over ^2^0^. The residue l e f t after evaporation of the solvent was d i s t i l l e d to give pure 2-propyl-4-oxopentanoic a c i d , bp 135°-140°C/9 mm [ l i t . (156) bp 165°C/20 mm]. NMR Spectrum: 60.8-1.1 ( t , 3H, CH3), 1.2-1.7 (m, 4H, 2CH2), 2.15 ( s , 3H, CH3-C=0), 2.5-3.1 (complex m, 3H, CH2 - C = 0, CH - C = 0 ) . Synthesis of Alpha, Alpha-Disubstituted A l i p h a t i c Acids Synthesis of 2,2-Dimethylbutyric Acid (V) Anhydrous THF (80 mL) and diisopropylamine (0.1 mol) were added to a dry, nitrogen-flushed flask immersed i n an ice-water bath. n-Butyllithium i n hexane (62 mL of 1.6 M, 0.1 mol) was added drop by drop to the w e l l - s t i r r e d s o l u t i o n . I s o b u t y r i c a c i d (0.045 mol) was then added to the lithium diisopropylamide s o l -u t i o n . The resulting milky white solution was turned into a clear homogenous solution by addition of tetrahydrofuran. S t i r r i n g was continued for 60 minutes at room temperature. E t h y l i o d i d e (0.048 mol) was then added at once into the reaction f l a s k . After 56 2 hours of additional s t i r r i n g at room temperature, the reaction mixture was quenched by addition of 15% HC1 (65 mL) at 0°C. The organic layer was separated and the aqueous layer extracted twice with diethyl ether. The combined organic layers were washed with 10% HC1, and saturated NaCl so l u t i o n . The organic layer was then dried with anhydrous Na2S04, f i l t e r e d and solvent removed using an o i l bath. The residue was d i s t i l l e d to give 2,2-dimethylbutyric ac i d (50% y i e l d ) , bp 183°-185°C/760 mm [ l i t (157) bp 7 9 ° -81°C/11 mm]. NMR Spectrum: 60.8-1.0 ( t , 3H, CH3), 1.2 ( s , 6H, 2CH3), 1.4-1.8 (m, 2H, CH2), 10.5-11.0 (broad s, 1H, COOH). b. Synthesis of 2,2-Dimethylvaleric Acid This compound was prepared using a procedure similar to that described above for 2,2-dimethylbutyric a c i d , by reaction of i s o -butyric acid (0.05 mol) and n-Propyl iodide (0.055 mol). The 2,2-dimethylvaleric acid thus obtained was pu r i f i e d by d i s t i l l a t i o n , bp 203°-204°C/760 mm. [ L i t . (158) bp 110°C/20 mm]. NMR Spectrum: 60.8-1.0 (m, 3H, CH3), 1.2 ( s , 6H, 2CH3), 1.3-1.7 (m, 4H, 2CH2), 10.8 (broad s, 1H, COOH). 3. Synthesis of Beta-Substituted A l i p h a t i c Acids a. Synthesis of 3-Ethylpentanoic Acid (VI) Lithium aluminium hydride (0.48 mol) and sodium-dried ether (700 mL) were introduced into a 2 l i t r e flask equipped with mech-anical s t i r r e r , dropping and reflux condenser. After s t i r r i n g the mixture for 15 min, ethyl 2-ethylbutyrate (0.70 mol) i n anhydrous 57 ether was added such that the ether refluxed gently. On addition of the ester, the mixture was s t i r r e d for 20 min to complete the reaction. Excess hydride was destroyed by addition of water. Following f i l t r a t i o n and separation of the phases, the ethereal solution was dried over MgSO^, f i l t e r e d and the ether removed by f l a s h evaporation. D i s t i l l a t i o n of the residue afforded 2-ethyl-b u t a n - l - o l (80% y i e l d ) , bp 146°-149°C. Mass spectrum (eight intense ions): m/z 43, 70, 71 (M-31), 55, 41, 56, 29, 84 [M-18]. To cold 2-ethylbutan-l-ol (0.54 mol), phosphorus tribromide (0.30 mol) was added dropwise and with s t i r r i n g over 2 hours at -10°C. The solution was then s t i r r e d at ambient temperature for 2 hours before being heated for 20 min with a steam bath. Water was added to the cooled reaction mixture and the product extracted with hexane. The hexane extract was dried over ?2®5 and the solvent removed. D i s t i l l a t i o n of the residue gave l-bromo-2-ethylbutane (50% y i e l d ) , bp 145°-148°C/760 mm. Mass spectrum (eight most i n -tense i o n , r e l a t i v e i n t e n s i t y ) : m/z 43 (100%), 85 (51%), 55 (48%), 71 (27%), 29 (23%), 116 (15%), 135 (12%), 137 (10%). l-Bromo-2-ethylbutane (0.18 mol) i n methanol was added to an aqueous-methanol solution of KCN (0.26 mol). The reaction mixture was refluxed for 24 hours. After d i s t i l l i n g of the alcohol, the product was extracted three times with hexane and the combined organic e x t r a c t s washed s u c c e s s i v e l y with 30% HC1, saturated NaHCO-j, water and dried over anhydrous MgSO^. D i s t i l l a t i o n afford-ed 2-ethylpentanenitrile (50% y i e l d ) , bp 100-102°C/ca 20 mm. Mass spectrum: m/z 43 (100%), 28 (40%), 71 (35%), 41 (30%), 55 (28%), 29 (20%), 54 (16%), 82 (8%). 58 2-Ethylpentanenitrile (0.05 mol) was hydrolyzed to the corres-ponding acid by refluxing with 50% s u l f u r i c acid (25 mL) for 9 hours. On extraction of the acid with ether, the ether extract was dried with anhydrous Na^O^ and the ether d i s t i l l e d o f f . Dis-t i l l a t i o n of the residue produced 3-ethylpentanoic a c i d (80% y i e l d ) , bp 80°-82°C/4 mm [ l i t . (5) bp 87°C/1 mm]. Mass Spectrum: m/z 60 (100%), 43 (64%), 70 (61%), 55 (58%), 61 (23%), 71 (18%), 83 (12%), 101 (10%). NMR Spectrum: 60.8-1.0 (m, 6H, 2CH3), 1.2-1.5 (m, 4H, 2CH2), 1.6-1.9 (m, 1H, CH), 2.2-2.4 (d, 2H, CH2-C00H), 9.5-10.5 (broad s, 1H, C00H). Synthesis of Cyclohexylacetic Acid (XIII) Cyclohexylmethylcyanide (bp 76°-80°C/7 mm) was prepared from cyclohexylraethyl bromide and hydrolyzed to cyclohexylacetic acid i n a similar procedure to that described above for synthesis of 3-ethylpentanoic acid ( a ) . D i s t i l l a t i o n of crude product under re-duced pressure gave c y c l o h e x y l a c e t i c a c i d , bp 136°C/0.15 mm [l i t . ( 1 5 9 ) bp 137°C/0.2 mm] which s o l i d i f i e d under ambient temp-erature. Mass Spectrum: m/z 60 (100%), 82 (75%), 55 (65%), 83 (52%), 61 (51%), 67 (43%), 41 (39%), 27 (17%). NMR Spectrum: 60.8-1.4 (m; 5H; C2, C5 protons on ring and CH), 1.5-2.0 (m, 6H, C3-C5 protons on a l i c y c l i c r i n g ) , 2.2 (d, 2H, CH2-C00H), 10.2 (broad s, 1H, C00H). Synthesis of 3-Methylvaleric Acid Diethyl 1-methylpropylmalonate was synthesized from diethyl 59 malonate and sec-butyl bromide. Ethanol and sodium ethoxide (prepared i n si t u ) were used as the solvent and base, respectively. Saponification of diethyl 1-methylpropylmalonate, followed by a c i d i f i c a t i o n gave upon decarboxylation 3-methylvaleric a c i d , bp 190°-196°C/760 mm [ l i t . (160) bp 187°-200°C/760mm]. NMR Spectrum: 60.8-1.1 (m, 6H, 2CH3), 1.2-1.5 (m, 2H, CH2), 1.6- 2.0 (m, 1H, CH), 2.1-2.5 (m, 2H, CH2C00H), 10.2 (broad s, 1H, C00H). Synthesis of 3-Methylhexanoic Acid Diethyl 1-methylbutylmalonate was synthesized from diethyl malonate and 2-bromopentane (from pentan-2-ol). Ethanol and sodium ethoxide were used as the solvent and base r e s p e c t i v e l y . Saponification of the monoalkylmalonate, followed by a c i d i f i c a t i o n gave upon decarboxylation 3-methylhexanoic acid bp 91°-94°C/5 mm [ l i t . (5) bp 60°-70°C/0.5 mm]. NMR Spectrum: 60.8-1.1 (m, 6H, 2CH3), 1.2-1.5 (m, 4H, 2CH2), 1.7- 2.0 (m, 1H, CH), 2.1-2.5 (m, 2H, CH2C00H), 10.1 (broad s, 1H, C00H). Synthesis of 2-propyl-4-pentenoic acid (4-ene VPA) and 2-propyl-3-hydroxypentanoic acid (3-OH VPA) was accomplished as described i n a previous work (93). 60 4. Synthesis of 5-Alkyltetrazoles a. Synthesis of 5-Isoamyltetrazole (XI) 5-Methylpentanenitrile was prepared i n a similar manner to that described for synthesis of 3-ethylpentanoic a c i d , from 1-bromo-3-methylbutane and an aqueous-methanol solution of potassium cyanide. Fractional d i s t i l l a t i o n gave 5-methylpentanenitrile, bp 150°-155°C/760 mm. Mass Spectrum: m/z 55 (100%), 41 (51%), 43 (47%), 27 (35%), 29 (30%), 54 (21%), 39 (15%), 57 (13%), 82 (M-CH3, 11%). 5-Methylpentanenitrile (0.06 mol) and sodium azide (0.18 mol) i n anhydrous THF (60 mL) were treated at room temperature with an-hydrous aluminium chloride (0.063 mol) dissolved i n cold anhydrous THF. The mixture was refluxed for 24 hours and with s t i r r i n g . The solvent was then d i s t i l l e d off under reduced pressure using a steam bath. Water (80 mL) was added to the residue and the solution acid-i f i e d to pH 2 with concentrated HC1. Evaporation was carried out again under reduced pressure to remove residual hydrazoic a c i d . The solution was further diluted with water to dissolve the alum-inium chloride and sodium chloride formed. The o i l y organic layer was separated and the aqueous phase extracted with ethyl acetate. The organic layers were l e f t i n the refrigerator u n t i l the 5-alkyl-tetrazole c r y s t a l l i z e d out. The c r y s t a l l i n e s o l i d was f i l t e r e d and r e c r y s t a l l i z e d from petroleum ether-diethylether solvent mixture to give 5-isoamyltetrazole, mp 95°-96°C [ l i t . (161) mp 95°-96°C]. NMR Spectrum: 60.95 (d, 6H, 2CH3), 1.5-2.0 (m, 3H, CH-CH2), 3.0-3.3 ( t , 2H, CH2-C=), 11.4 ( s , 1H, NH). 61 Synthesis of 5-Cyclohexylmethyltetrazole (XII) Cyclohexylethanenitrile was prepared, as described i n the synthesis of cyclohexylacetic a c i d , from cyclohexylmethylbromide. The n i t r i l e was converted into 5-cyclohexylmethyltetrazole, i n a similar fashion to that described for synthesis of 5-isoamyl-t e t r a z o l e . This required use of c y c l o h e x y l e t h a n e n i t r i l e and aluminium azide (prepared i n s i t u from sodium azide and anhydrous aluminium c h l o r i d e ) . The isolated crude product was r e c r y s t a l l i z e d from ethyl acetate to give 5-cyclohexylmethyltetrazole, mp 109-110°C [ l i t . (161) mp 109°C]. NMR Spectrum: 60.9-1.4 (m; 5H; C2, CQ protons on a l i c y c l i c ring and CH), 1.5-2.0 (m, 6H, C3-C5 protons on r i n g ) , 2.9-3.2 (d, 2H, CH2C=), 10.5 (broad s, 1H, NH). Synthesis of 5-Heptyltetrazole (X) 1-Bromoheptane (0.32 mol) was added slowly to a warm, w e l l -s t i r r e d solution of sodium cyanide (0.41 mol) i n dimethylsulfoxide (130 mL). S t i r r i n g was continued for 1 hour u n t i l the solution temperature cooled from 130°C to ambient temperature. The reaction mixture was poured into water and the product extracted three times with ether. The combined ether extract was washed with saturated NaCl and dried over anhydrous MgSO^. The l i q u i d l e f t upon evapor-ation of ether was d i s t i l l e d under reduced pressure to give octane-n i t r i l e , bp 85°-88°C/10 mm. Octanenitrile was reacted with aluminium azide i n refluxing THF, employing the procedure for synthesis of 5-isoamyltetrazole. 62 The crude s o l i d product isolated was r e c r y s t a l l i z e d from aceto-n i t r i l e to give 5-heptyltetrazole, mp 41°-42°C [ l i t . (161) mp 41.5°C]. NMR Spectrum: 60.7-1.0 ( t , 3H, CH3), 1.1-1.5 (m, 8H, 4CH2), 1.6-2.0 (m, 2H, CH2 CH2-C=), 2.9-3.3 ( t , 2H, CH2-C=), 10-11 (broad s, 1H, NH). 5. Synthesis of Succinamic Acids a. Synthesis of N,N-Diethylsuccinamic Acid (XV) S u c c i n i c anhydride (0.1 mol) was added with s t i r r i n g to diethylamine (0.1 mol) i n cold absolute ethanol (25 mL). S t i r r i n g was continued u n t i l the reaction mixture attained room temperature. N,N-Diethylsuccinamic acid c r y s t a l l i z e d out of the homogenous s o l -ution on cooling and was f i l t e r e d under suction before r e c r y s t a l -l i z a t i o n from benzene, mp 83°-85°C [ l i t . (162) mp 82-84°C]. IR Spectrum (mull): 1690-1720 cm-1 (C = 0 i n COOH), broad), 1630 (C = 0 i n amide). NMR Spectrum: 61.1-1.5 (distorted t , 6H 2CH3), 2.6 ( s , 0=C-CH2CH2-C=0), 2.8-3.3 (q, 4H, CH2-N-CH2), 5.0-6.0 (broad s, 2H, COOH). b. Synthesis of N,N-Dibutylsuccinamic Acid (XVI) Succinic anhydride (0.15 mol) was added to anhydrous benzene (120 mL) and the solution warmed to p a r t i a l l y dissolve the an-hydride. Dibutylamine (0.15 mol) i n benzene (30 mL) was rapidly added to the mixture at a controlled rate to contain the heat generated. The r e a c t i o n mixture was then r e f l u x e d u n t i l the 63 reaction mixture turned homogenous. The resulting reaction mixture was allowed to cool to ambient temperature and l e f t i n the r e f r i g e r a t o r u n t i l the s o l i d product c r y s t a l l i z e d o u t . N,N-D i b u t y l s u c c i n a m i c a c i d was f i l t e r e d under s u c t i o n and r e c r y s t a l l i z e d with benzene, mp 78°-80°C. IR Spectrum (mull): 1720 cm-1 (C = 0 i n C00H), 1615 (C = 0 i n amide). NMR Spectrum: 60.8-1.1 ( t , 6H, 2CH3), 1.2-1.9 (m, 8H, 2CH2-CH2), 2.6 ( s , 4H, 0 = C-CH2 CH2 C = 0 ) , 2.8-3.1 ( t , 4H, CH2-N-CH2), 8.2-8.9 (broad s, 2H, C00H). 6. Synthesis of Diunsaturated Analogues of Valproic Acid The diunsaturated derivatives of valproic acid were prepared by dehydration of the 8-hydroxy unsaturated esters followed by hydrolysis of the diunsaturated esters. The 3-hydroxy unsaturated esters were prepared by deconjugative aldol condensation of the lithium dienolates of ethyl 2-pentenoate and propionaldehyde. The dienolate of ethyl (E)-2-pentenoate was expected to give mainly the B-hydroxy-(Z)-3-pentenoate whereas 8-hydroxy-(E)-3-pentenoate was expected from reaction with ethyl (Z)-2-pentenoate. I a . Synthesis of Ethyl 2-(l'-hydroxypropyl)-3-Pentenoate from Ethyl (E)  -2-Pentenoate E s t e r i f i c a t i o n of (E)-2-pentenoic a c i d ( E t I , K2C03, THF, r e f l u x , 6 hr) gave ethyl-(E)-2-pentenoate, bp 36-38°C/0.07 mm. NMR Spectrum: 61.0 ( t , 3H, CH3), 1.3 ( t , 3H, CH3), 2.2 (q, 2H, CH2-C=), 4.2 (q, 2H, 0CH2), 5.6-6.0 (d, 1H, = CH, J = 16Hz), 6.8-7.3 (dt, 1H, HC ==, J = 16 Hz). 64 n-Butyllithium i n hexane (69 mL of 1.6M, 0.11 mol) was added dropwise with s t i r r i n g to a solution of diisopropylamine (0.11 mol) i n anhydrous THF (120 mL) at 0°C. The solution was s t i r r e d for 20 min at 0°C and then cooled to -78°C. Hexamethylphosphoramide (0.11 mol) was added to the solution followed by dropwise addition of ethyl-(E)-2-pentenoate (0.10 mol). After 30 min, propionalde-hyde (0.10 mol) was added. The reaction mixture was s t i r r e d for a further 30 min at -78°C before the reaction was quenched with 15% HC1. The product was extracted with ether and the ether extract washed successively with 10% HC1, saturated NaHC03 and saturated NaCl. The organic layer was dried over Na2S04, and solvents re-moved with a rotary evaporator. The residue after f r a c t i o n a l d i s t i l l a t i o n under reduced pressure gave the major product, ethyl 2-(l'-hydroxypropyl)-3-pentenoate, bp 85-90°C/0.25 mm (48% y i e l d ) . IR Spectrum: 3400 cm-1 (0-H), 1735 cm-1 (C=0), 1635 cm-1 (C=C), 1175 cm-1 (C-0), 985 cm-1 (medium i n t e n s i t y , =CH, t r a n s ) , 745 cm-1. NMR Spectrum: 60.95 ( t , 3H, CH3), 1.25 ( t , 3H, OC^CHj), 1.55 (m, 2H, CH2), 1.75 (d, 3H, CH3-CH=), 2.55 (broad s, 1H, OH), 3.40 (m, 1H, CH-C=0), 3.8 (m, 1H, CH-0), 4.2 (q, 2H, 0CH2CH3), 5.3-5.9 (m, 2H, CH = CH). Synthesis of Ethyl 2-(lf-hyrodxypropyl)-3-pentenoate from Ethyl-(Z)  -2-Pentenoate Ethyl (Z)-2-Pentenoate was prepared i n three stages, ( i ) A mixture of 2-pentanone (0.5 mol) and precooled 48% HBr (50 mL) was c h i l l e d to 0°C and bromine (1.0 mol) added dropwise. After add-i t i o n of bromine, water (200 mL) was added and the heavier organic 65 layer d i s t i l l e d to give a fract i o n with bp 84-89°C/10 mm. GCMS analysis showed 1,3-dibromo-2-pentanone to be the main product, ( i i ) 1,3-dibromo-2-pentanone (0.3 mol) was added to a solution of KHCOg (2.1 mol) i n 1200 mL water. The reaction mixture was s t i r r e d for 24 hours. Nonacidic by-products i n basic solution were d i s -carded a f t e r e x t r a c t i o n with e t h e r . The basic s o l u t i o n was a c i d i f i e d to pH 2 with d i l u t e HC1 and the product extracted with ether. The ether phase was i n turn dried over MgSO^, f i l t e r e d and concentrated with a rotary evaporator. D i s t i l l a t i o n of the residue gave (Z)-2-pentenoic a c i d , bp 39°-41°C/0.4 mm. ( i i i ) E s t e r i -f i c a t i o n of the acid ( E t I , K2C03, THF, r e f l u x , 6 hr) gave ethyl (Z)-2-pentenoate, bp 50°-52°C/12 mm. NMR Spectrum; 61.0 ( t , 3H, CH3), 1.3 ( t , 3H, ( X ^ C H j ) , 2.65 (q, 2H, CH2-C=), 4.2 (q, 2H, 0CH2-), 5.6-5.9 (d, 1H, CH=, J = 10 Hz), 6.0-6.4 (dt, 1H, CH=, J = 10 Hz). . The lithium dienolate of ethyl (Z)-2-pentenoate (0.05 mol) was prepared at -78°C using lithium diisopropylamide (0.055 mol) i n 60 mL THF and hexamethylphosphoramide (0.055 mol). Proprion-aldehyde (0.05 mol) was added dropwise to the solution and the reaction s t i r r e d for 30 min and f i n a l l y quenched with 10% HC1. The major product, ethyl 2-(l'-hydroxypropyl)-3-pentenoate, was i s o -l a t e d , bp 95-100°C/l mm (50% y i e l d ) . IR Spectrum: 3400 cm-1 (0-H), 1735 cm-1 (C=0), 1640 cm-1 (C=C), 985 cm-1 (strong i n t e n s i t y , = CH, trans), 745 cm-1. 66 I I . Dehydration of ethyl 2-(l'-hydroxypropyl)-3-pentenoate a. Phosphorus Pentoxide Phosphorus pentoxide (36 mmol) was added to the B-hydroxy unsaturated ester (39 mmol, from ethyl (E)-2-pentenoate) i n 60 mL of anhydrous benzene and the mixture refluxed for 4 hours. The diunsaturated esters isolated were saponified with d i l u t e NaOH and subsequently a c i d i f i e d with d i l u t e H2S04 to give a mixture of seven isomeric dienoic acids, bp 105°-110°C/2.5 mm. b. Toluenesulfonyl chloride-pyridine Dehydration of the S-hydroxyunsaturated ester (33 mmol, from e t h y l (E)-2-pentenoate) i n 20 mL of p y r i d i n e with p-toluene-sulfonylchloride (45 mmol) was carried out by f i r s t s t i r r i n g the reaction mixture at room temperature for two days, d i l u t i n g with ice-water and extracting the p-toluenesulfonate derivative with CHCI3. The isolated p-toluenesulfonate was refluxed with d i l u t e NaOH and a c i d i f i e d to give a mixture of four isomeric dienoic acids, bp 120°-128°C/6 mm. NMR Spectrum (270 MHz) of mixture: 60.95-1.1 ( t , CH3), 1.55 (dd, CH3-C=, 2E-3'Z), 1.64 (dd, CH3-O, 3Z-3'Z), 1.70 (dd, CH3-C=, 2Z-3'E), 1.85 (d, CH3-C=, 2E-3'E), 2.16 (m, CH2-C=, 2E-3'Z), 2.35 (m, CH2-C=, 2E-3'E), 2.55 (ra, CH2-C=, 2Z-3'E), 4.05 ( t , CH, 3Z-3'Z), 5.65 (m, CH=CH, 3Z-3'Z), 5.73 (m, CH=CH, 2E-3'Z), 5.85 (m, CH=CH, 2Z-3'E), 5.95 ( t , CH=, 2Z-3'E), 6.03 (m, CH=, 2E-3'E), 6.14 (d, CH=, 2E-3'E), 6.83 ( t , CH=, 2E-3'E), 7.00 ( t , CH=, 2E-3'Z). IR Spectrum of mixture: 1690 cm-1 (C=0), 1633 cm-1 (C=C), 965 cm-1 (=CH, trans), 935 cm-1, 735 cm-1. 67 NMR assay of the four dienoic acid isomeric mixture, using appropriate integrated areas, gave the r a t i o of 2-(l'-propenyl)-2-pentenoic acid and 2-(l'-propenyl)-3-pentenoic acid isomers 3Z-3'Z, 2Z-3'E, 2E-3'Z, 2E-3'E as approximately 12 : 14 : 56 : 18. Methanesulfonyl chloride - Potassium hydride Dehydration of the 6-hydroxy unsaturated ester (0.05 mol), derived from ethyl (E)-2-pentenoate, was carried out by forming the sulfonate with methanesulfonyl chloride (0.06 mol) i n the presence of triethylamine (0.08 mol) and methylene chloride (40 mL) at 0°C. KH (0.10 mol) was added at 0°C to the i s o l a t e d s u l f o n a t e and reaction mixture s t i r r e d for 12 hours at room temperature. Excess hydride was decomposed by addition of t-butanol and water. Follow-ing the usual work-up, the i s o l a t e d product mixture, bp 75-80°C/2 mm, contained three isomeric dienoates. IR Spectrum of mixture: 1716 cm-1 (C=0), 1636 cm-1 (C=C), 968 cm-"'' (=CH, trans), 756 cnT^, 690 cm-''" (=CH, c i s more intense than =CH, trans). NMR assay of the ethyl esters of the three dienoic acid mix-ture, using appropriate integrated areas, gave the r a t i o of isomers 2Z-3'E: 2E-3'Z: 2E-3'E as 7:71:22. NMR Spectrum of isomeric mixture (270 MHz): 61.0-1.1 (m, CH3), 1.3 (m, CH3), 1.55 (dd, CH3-C=, 2E-3'Z), 1.7 (dd, CH3-C=, 2Z-3'E), 1.83 (d, CH3-C=, 2E-3'E), 2.12 (ra, CH2-C=, 2E-3'Z), 2.3 (m, CH2-C=, 2E-3'E), 2.52 (m, CH2-C=, 2Z-3'E), 4.2 (q, CH20), 5.75 (m, CH =, 2E-3'Z), 5.98 (d, CH=, 2E-3'Z), 6.04 (m, CH=, 2E-3'E), 6.13 (d, CH=, 2E-3'E), 6.55 ( t , CH=, 2E-3'E), 6.8 ( t , CH=, 2E-3'Z). 68 The isomeric dienoates mixture was saponified with d i l u t e NaOH and subsequently a c i d i f i e d with d i l u t e H2S0^ to give the isolated product mixture of the corresponding acids of two major isomers 2E-3'Z, 2E-3'E and trace amounts of the isomer 2Z-3'E. NMR Spectrum (400 MHz) of the mixture: 61.0-1.1 (m, CH3), 1.56 (dd, CH3-C=, 2E-3'Z), 1.73 (dd, CH3-C=, 2Z-3'E), 1.84 (d, CH3~ C=, 2E-3'E), 2.15 (m, CH2-C=, 2E-3'Z), 2.23 (m, CH2-C=, 2E-3'E), 2.50 (m, CH2-C=, trace, 2Z-3'E), 5.8 (m, CH=, 2E-3'Z), 5.96 (d, CH=, 2E-3'Z), 6.07 (m, CH=, 2E-3'E), 6.13 (d, CH=, 2E-3*E), 6.78 ( t , CH=, 2E-3'E), 6.97 ( t , CH=, 2E-3'Z). ( i i ) D e h y d r a t i o n of the 8 - h y d r o x y u n s a t u r a t e d e t h y l e s t e r (0.02 mol), derived from ethyl (Z)-2-pentenoate, was carried out by treatment of the mesylate derivative with KH (0.04 mol). The isolated product mixture, bp 65-70°/0.1 mm contained three isomeric dienoates, 3Z-3'Z, 2E-3'Z, 2E-3'E i n the r a t i o of approximately 44:8:48 (determined by NMR an a l y s i s ) . IR Spectrum: 1733 cm-1 (C=0), 1716 cm-1 (C=0), 1655-1640 cm"1 (C=C), 985-965 cm-1 (CH=, trans), 745 cm-1. NMR Spectrum of the isomeric mixture (270 MHZ): 61.0-1.1 (m, CH3), 1.2-1.25 (m, CH3), 1.53 (d, CH3-C=, 2E-3'Z), 1.64 (dd, CH3~ C=, 3Z-3'Z), 1.68 (dd, CH3-C=, 3Z-3'Z), 1.82 (d, CH3-C=, 2E-3'E), 2.1 (m, CH2-C=, 2E-3'Z), 2.35 (m, CH2~C=, 2E-3'E), 3.5 (m, CH, 3Z-3'Z), 5.5-5.7 (m, CH=CH, 3Z-3'Z), 6.04 (m, CH=, 2E-3'E), 6.13 (d, CH=, 2E-3E), 6.57 ( t , CH=, 2E-3'E), 6.77 ( t , CH=, 2E-3'Z). I I I . Semi-Preparative Argentation TLC S i l i c a gel G was impregnated with AgN03 (20% w/w) and TLC 69 p l a t e s prepared with 0.5 mm t h i c k n e s s . A mixture of i s o m e r i c dienoates (20-40 mg) i n CHCI3 was applied across the plates (20 x 20 cm) i n a narrow band. Benzene: petroleum ether, 30°-60°: d i e t h y l ether (80 : 20 : 5) was s e l e c t e d as the mobile phase a f t e r experimenting w i t h d i f f e r e n t solvent systems. The p l a t e s were developed i n a rectangular glass tank. The solvent f r o n t was allowed to run to 15 cm on the plates, from the o r i g i n . The plates were then sprayed w i t h 50% H2SO4 and heated at 200°C f o r 10 min. values of three charred bands were determined. Another series of p l a t e s was developed under the same c o n d i t i o n s . With pre-knowledge of band p o s i t i o n s , s t r i p s of support m a t e r i a l corresponding to the three bands were scraped o f f the unheated plates into centrifuge tubes. The ethyl esters were extracted with methanol. F o l l o w i n g c e n t r i f u g a t i o n and f i l t r a t i o n of the methanolic s o l u t i o n s , the three f r a c t i o n s were concentrated and reconstituted i n CDCI3 for GCMS and NMR analysis. IV. In vivo metabolism - i s o l a t i o n procedure a. Serum. A serum sample (2.0 mL) from a p a t i e n t on VPA, was a c i d i f i e d to pH 2.0 with 4N HC1. Extraction was carried out twice with 2.0 mL of ethyl acetate, each time c e n t r i f u g i n g at 2000 rpm f o r 20 min to separate the phases. The e t h y l acetate l a y e r s were combined, dried over anhydrous Na2S04 and then concentrated at room temperature using a gentle stream of dry n i t r o g e n . The concen-trated extracts were derivatized and aliquots injected into the gas chromatograph mass spectrometer. b. Urine. To 5 mL of a urine sample was added 3N NaOH to b r i n g 70 the pH to 13.0 and hydrolysis of conjugates was effected by heating the solution at 60°C for 1 hr. After cooling, the solution was a c i d i f i e d to pH 2.0 with AN HC1. The a c i d i f i e d urine was extracted twice with ethyl acetate. The combined ethyl acetate layers were dried, concentrated and derivatization carried out. V. Derivatization of Acids t-BDMS derivatives were prepared by adding 30-50 uL of t-BDMS reagent to synthesized sample (3-5mg) or urine extraction samples (see above) and warming at 60°C for 5 min. The t-BDMS derivatives were extracted with 100-150 u l of solvent (ethyl acetate-hexane) and aliquots injected into the GCMS. TMS-derivatives were formed by treating the concentrated urine extraction samples or synthesized samples with 30-50 u l of N-methyl-N-trimethylsilyl-trifluoroacetamide at 50°C for 5 min. Ethyl esters of acids were formed using appropriate volumes of triraethylanilinium hydroxide solution and ethyl iodide. The mix-ture was warmed at 50°C for 5 min. Methyl esters of acids were prepared by reaction of the acid sample with diazoraethane generated from Diazald® (dissolved i n ether/2-ethoxyethanol) and 60% K0H. VI. Photochemical Isomerization A mixture of seven isomeric dienoic acids ( l g ) , derived from dehydration of ethyl 2-(l'-hydroxypropyl)-3-pentenoate with P2^5» was dissolved i n 150 mL of hexane and added to a pyrex w e l l . 2-Propyl-(E)-2-pentenoic acid (2-ene VPA) was added as an inter n a l 71 standard. A quartz immersion well with a 450W unfi l t e r e d Hanovia lamp was set into place and i r r a d i a t i o n carried out for 6 hours. Aliquots of the reaction mixture were removed pe r i o d i c a l l y and the t-BDMS esters of photolyzed acids analyzed by c a p i l l a r y GCMS. VII . Capillary GCMS resolution of 2,3'-diene VPA and 2,4-diene VPA Samples of synthetic diene acids, 2-propyl-(E)-2,4-penta-dienoic acid and 2-[(Z)-l'-propenyl]-(E)-2-pentenoic a c i d , or the urine extract from one patient on VPA therapy, were made with ethyl acetate and derivatives formed. Urine extracts were also spiked with synthetic diene acids prior to TMS or t-BDMS d e r i v a t i z a t i o n . Capillary GCMS analysis of the TMS derivatives was performed using a 25 m x 0.3 mm i . d . SE 54 column at oven temperatures from 50°C to 90°C at 30°/min and then held at 90°C for 10 min. Capillary GCMS analysis of the t-BDMS derivatives was performed using either the SE-54 column or 0V-1701 column. Oven temperatures were programmed from 50°C to 100°C at 30°/min and then 100°C to 200°C at 8°/min. D. Subcutaneous Pentylenetetrazole (PTZ) Seizure Threshold Test 1. Animals Adult male Swiss mice (CD1 s t r a i n , 20-32g, Charles River, Quebec, Canada) were used as experimental animals. A l l animals were allowed free access to food and water except during the time of t e s t . Mice were housed after a r r i v a l for at least 24 hours before use. 72 Drugs Isobutyric a c i d , trimethylacetic acids, 1-Methylcyclohexane-l-carboxylic acid and pentylenetetrazole (PTZ) were obtained from Aldrich Chemical Company, Milwaukee, USA. Valproic acid was pur-chased from Abbott Laboratories, I l l i n o i s , USA. 2-Butylhexanoic a c i d , 2,2-Dimethylbutyric a c i d , 3-Ethylpentanoic a c i d , Cyclohexyl-acetic a c i d , 2-Propyl-A-oxopentanoic acid (4-keto VPA), 2-Propyl-(E)-2-pentenoic acid (2-ene VPA), 2-(l'-Propenyl)-2-pentenoic acid (2,3'-diene VPA), as a mixture of 2E-3'Z and 2E-3'E i n a r a t i o of 7 : 3, N,N-Diethylsuccinamic ac i d , N,N-Dibutylsuccinamic ac i d , 5-Cyclohexylraethyltetrazole, 5-Isoamyltetrazole and 5-Heptyltetrazole were synthesized as described i n the experimental section. The purity and i d e n t i t y of these compounds were v e r i f i e d by IR, NMR spectroscopy and GCMS analysis. Drug Solutions Pentylenetetrazole was dissolved i n 0.9% sodium chloride to make a concentration of 1.7%. The succinamic acids, which are soluble in water, were dissolved i n 0.9% sodium chloride. The s o l u t i o n s of the f a t t y a c i d s and t e t r a z o l e s were prepared by neutralizing a known amount of the drug with IN NaOH and the pH of the resulting solution adjusted to 7.4 with 1.0N HC1. A l l drugs were administered to mice i n concentrations which permitted optimal accuracy i n dosage. Drugs were administered i . p . i n a volume not more than 7 mL/kg body weight i n mice and PTZ injected s.c. i n a volume of 5 mL/kg. 73 4. Experimental Procedure a. Characterization of PTZ Seizures (85 mg/kg dose) Mice received a subcutaneous i n j e c t i o n of PTZ, 85 mg/kg, i n a loose f o l d of skin, on the back of the neck. This i s the CDgy dose of PTZ i n mice (46). Control mice, i n a group of 12, were injected s.c. w i t h PTZ, 85 mg/kg. A f t e r PTZ a d m i n i s t r a t i o n , animals were placed i n individual cages for observation. Observation time was 30 min following the s.c. PTZ i n j e c t i o n . The r e a c t i o n of each mouse to PTZ (85 mg/kg) f o l l o w e d a def-i n i t e p a t t e r n , c h a r a c t e r i z e d by a b r i e f episode of body tw i t c h e s which passed into a series of clonic jerkings usually accompanied by audible squeaks and Straub t a i l phenomenon. The clonic spasms occurred between resting phases. The convulsive syndrome usually ended with the swimming phase, characterized by coordinated flexor-extensor movement of the l e g s . The post s e i z u r e phase u s u a l l y ended f a t a l l y with tonic extension of hind limbs or was character-ized by a stupor phase after which mice recovered. During the 30 minute observation period, sustained synchronous c l o n i c j e r k i n g s occurred i n 100% of c o n t r o l mice. The onset to c l o n i c episodes (episodes of 5 second duration or longer) was between 3-9 minutes. Seventy-five percent of c o n t r o l mice died within the 30 min observation period. b. Antagonism of PTZ clonic seizures Mice i n groups of eight per dose of drug, received i.p. i n j e c -tions of test drug i n a maximum of four doses between 0.2 mmol/kg and 2.0 mmol/kg. For each drug, a f i x e d time i n t e r v a l of 15 min 74 after i . p . administration was chosen before mice received a s.c. 85 mg/kg PTZ i n j e c t i o n . Each mouse was in d i v i d u a l l y caged for observation after the s.c. PTZ i n j e c t i o n . Each mouse was observed for a maximum time of 30 min after PTZ administration but clonic seizures rarely occurred after times longer than 20 min. During the period of observation, the incidence and timing of clonic convulsions was recorded with absence of sustained clonic spasms (5 sec duration or longer) being defined as protection. Eight mice were used for each point on the dose-effect curves. At least three points were established between the l i m i t s of com-plete protection and no protection. The probit of the percentage protected was plotted against the log dose. The data were analyzed by the s t a t i s t i c a l method of L i t c h f i e l d and Wilcoxon (192). The ED^Q values, slope of the curves, and their 95% confidence l i m i t s were then recorded for each anticonvulsant drug. Mortality Test The percentage of mice which survived within 30 minutes after the s.c. 85 mg/kg PTZ in j e c t i o n was determined at the 1.0 mmol/kg i.p . dose l e v e l of test compound. Test drugs were administered i . p . 15 minutes before PTZ administration. Toxic effects of drugs Specific t o x i c i t y tests were not conducted. However, observ-ations of toxic signs were made during the 15 minute time i n t e r v a l before administration of PTZ. Neurotoxic effects were indicated by 75 sedation, abnormal gait or ataxia, hyperactivity, r a p i d - c i r c l i n g a c t i v i t y , abnormal body posture, abnormal spread of the limbs, prostration (immobility or resting on b e l l y ) , f a s c i c u l a t i o n (move-ment of muscles on the back), tremors, and convulsions. e. Convulsant A c t i v i t y Test Compounds which induced tremors, hyperexcitability and convul-sions i n mice were tested for convulsant a c t i v i t y by i . p . administration of sublethal doses and the mice observed for 30 minutes without administration of PTZ. E. HPLC Method for Determination of Octanol-Water P a r t i t i o n Coefficients 1. Instrumentation Analyses were performed with a Hewlett-Packard Model 1082B HPLC equipped with an HP 79850B LC terminal. Detection of the acidic compounds was at 210 nm with a Hitachi Model 155-10 variable wave length spectrophotometer. The Absorbance Units F u l l Scale (AUFS) was set at 0.05 AU. 2. Column The analyses were carried out using a 20 cm x 4.6mm i . d . column packed with 5 um, Hypersil 0DS (Hewlett-Packard). 3. Eluents HPLC-grade a c e t o n i t r i l e and methanol were used i n the prep-a r a t i o n of the mobile phases. D i s t i l l e d and deionized water ( M i l l i - Q system, M i l l i p o r e Corp., Bedford, MA) was used to prepare 76 0.01M NaH2P04 buffer, pH 3.5. The eluents were 0.01M NaH2POA buffer containing different percentages of methanol or aceto-n i t r i l e . The solvents were mixed and degassed with s t i r r i n g for 5 min under a water-aspirator vacuum. Eluent flow rates were between 1.0-1.5 mL/rain requiring a pressure of 900-1500 p s i . 4. Compounds Seven a l i p h a t i c acids whose octanol-water p a r t i t i o n co-e f f i c i e n t s had been determined i n the same laboratory served as reference compounds. The log PQ/W of these acids ranged from 0.98 to 3.20. Straight-chain fatt y acids obtained commercially, pro-vided a homologous series for estimation of void time. Isobutyric ac i d , trimethylacetic acid and 1-methylcyclohexane-l-carboxylic acid were purchased from Aldrich Chemical Company, Milwaukee, USA. The other a c i d i c compounds used were synthesized as described i n the experimental section. 5. Sample Preparation Standards of acidic compounds were prepared i n either methanol or a c e t o n i t r i l e to give concentrations of saturated fatt y acids of 0.2-0.6 Ug/uL, unsaturated fatt y acids of 0.01-0.05 ug/yL, N,N-dibutylsuccinamic acid of 0.1-0.3 Ug/uL, N,N-diethylsuccinaraic acid of 0.8-1.2 ug/uL and tetrazoles of 0.02-0.07 ug/uL. 6. Retention Time Measurements The acidic compounds were studied i n d i v i d u a l l y . Sample i n j e c -t i o n volumes were 20 uL. The retention times were expressed i n 77 terras of log (capacity factors) or log k'. Log k' = log (tR-tQ) fco where t-g represents the retention time of the compound and tQ, the elution time of unretained peak generated by methanol or aceto-n i t r i l e . The recorder chart speed was 1.0 cm/min and the flow rate of the eluent was between 1.0 mL/min and 1.5 mL/min. Retention times were measured by the integrator and the average of two r e p l i -cate injections was used. 7. Column Void Time While the dead volume (v ) i s independent of flow rate, the column void (or dead) time i s dependent on the eluent composition and flow rate. I t was therefore determined by i n j e c t i o n of a substance that i s expected to be unretained, i . e . methanol. The void time was compared to the void time determined by dead time i t e r a t i o n of the retention times of homologous n-alkylcarboxylic acids (181). The equation i s expressed by tp = atg N ~ to (a-1). By plotting t j ^ + 1 (retention time of the N+l = carbon homologue) versus tp ^ (retention time of the N-carbon homologue), t can be calculated from the slope of the regression l i n e , a ( r e l a t i v e retention) and the intercept. Values of tQ and t ^ were not corrected for the small time lag between column and detector. F. Determination of Octanol-Water P a r t i t i o n Coefficients by the Shake-Flask Procedure  The p a r t i t i o n c o e f f i c i e n t of four compounds, trimethylacetic a c i d , 5-isoamyltetrazole, 5-cyclohexylmethyltetrazole and N,N-dibutylsuccinaraic a c i d , were determined i n l-octanol-0.IN HC1. For the par t i t i o n i n g studies, 1-octanol saturated with 0.1N HC1 and 78 0.1N HC1 saturated with 1-octanol were used. The amount of compound and volume of the aqueous phase were chosen such that the i n i t i a l concentrations of compounds i n 0.1N HC1 saturated with 1-octanol were i n the range, 1.0 x 10-3M to 5.0 x 10 M. Trimethylacetic acid and N,N-dibutylsuccinaraic were s o l -uble i n the volume of aqueous phase used. The weighed amount of the tetrazoles were rendered soluble i n the aqueous phase by f i r s t dissolving i n a small volume of MeOH (2%, v/v) and the calculated volume of 0.1N HC1 added. The rat i o s of the volumes of the aqueous and octanol phase were 2:1 for trimethylacetic acid and 10:1 for the other compounds. The rat i o s were chosen so that a n a l y t i c a l errors can be decreased. The separatory funnel (50-120 mL capacity) was shaken gently and inverted several times for 15 min. The aqueous phase was collected into centrifuge tubes and centrifuged at 2000 rpm for 20 min. The concentration of the compounds i n the aqueous phase after p a r t i t i o n i n g was determined by HPLC analysis. Calibration curves for concentrations i n the range (0.01 mg/mL to 0.6 mg/mL) were obtained for each compound studied. Duplicate injections were made for standard and sample solutions. The concentration of compounds i n the aqueous phase was deduced from the c a l i b r a t i o n curves. HPLC analyses were conducted with a Hewlett Packard Model 1082B, HPLC. Detection of the ac i d i c compounds was accomplished at 210 nm with a Hitachi Model variable wave length spectrophotometer. The spectrophotometer was set at 0.05 absorbance units f u l l s cale. A 20 cm x 4.6 mm i . d . column packed with 5 ym H y p e r s i l 0DS (Hewlett-Packard) was used i n the analysis. Mobile phase of 70% MeOH : 30% 0.01M NaH2P0A was used for trimethylacetic a c i d , 5-79 isoamyltetrazole and N,N-dibutylsuccinamic acid analyses. Mobile phase of 50% CH3CN : 50% 0.01M NaH2P04 was used for 5-cyclohexyl-m e t h y l t e t r a z o l e a n a l y s i s . Eluent flow rate was 1 mL/min and volumes of 20 uL were injected. Quantitation was by peak height or peak area measurements. HPLC analyses were conducted at room temperature. The p a r t i t i o n experiment for each compound was done i n quad-r u p l i c a t e . The p a r t i t i o n c o e f f i c i e n t (p) of a compound was calculated from the r e l a t i o n s h i p . v= J where X^ " and Xgq are the respective amount of compound i n the aqueous phase before and after p a r t i t i o n i n g and VQq and VQ are the volumes of the aqueous and organic phases respectively. G. Determination of apparent ionization constants (pKa) by potentio-metric t i t r a t i o n  The pKa values of the a c i d i c compounds were determined by the method of A. A l b e r t and E. Serjeant (185). Potentiometric t i t r a t i o n was c a r r i e d out with an automatic potentiometric t i t r a t i o n instrument containing a motor-driven microburet and a recorder (Beckman). The electrode system was a glass electrode i n combination with a saturated calomel e l e c t r o d e . Before each t i t r a t i o n the instrument was calibrated against potassium hydrogen phthalate (0.05M, pH 4.00) and phosphate buffer (pH 7.00). Sol-vents were deionized water and HPLC-grade methanol. 80 The apparent ionization constants were determined i n aqueous methanol (10% and 50% MeOH, v/v) by potentiometric t i t r a t i o n with standard KOH (standardized with potassium hydrogen phthalate), delivered from a microburet. Samples of compounds were weighed out into beakers and dissolved i n the solvent mixture (47.5 mL) such that the mean concentration of acids during the t i t r a t i o n (or concentration at half-neutralization) was approximately 0.002M for low molecular weight acids and 0.001M for high molecular weight (>130 Daltons) acids. Values of pKa were calculated at several points i n each t i t r a t i o n by applying the Henderson-Hasselbalch equation and a mean pKa determined. An average of 6 pKa values was used. 81 RESULTS AND DISCUSSION A. Chemistry 1. Alpha-alkylsubstituted a l i p h a t i c acids Valproic a c i d , I , and 2-butylhexanoic acid, I I , were prepared according to the method of Pfeffer and co-workers (154). The syn-thetic sequence i s shown i n Scheme 1. I t i s a one-pot synthetic method involving treatment of alkylcarboxylic acids with the strong base, lithium diisopropylamide (LDA) to generate the dianions. Alkylation of the dianions with the appropriate a l k y l bromide occurred r e g i o s p e c i f i c a l l y at the a-anionic s i t e to produce the a-substituted a l i p h a t i c acids i n good y i e l d s . 2. a.q-Dialkylsubstituted a l i p h a t i c acids The synthetic procedure of Creger (163) was used to make the a,a-disubstituted acids, 2,2-dimethylvaleric acid (XXV) and 2,2-di-methylbutyric a c i d (V) (Scheme 2 ) . The r e a c t i v e centres are usually less accessible with other bases (e.g. NaH) or require multiple reaction steps i n other synthetic methods such as the c l a s s i c a l malonic ester synthesis. 3. a-Alkylsubstituted a l i p h a t i c acids with functionality i n the carbon chain  2-Propyl-(E)-2-pentenoic acid was prepared employing standard procedures (Scheme 3). The f i r s t step involves bromination of v a l -proic acid, I , i n the a-position. The bromo-acid was converted to the ester and dehydrobromination carried out successfully with 82 a) C3H7-CH^COOH 2LDA THF/HMPA/0°C C3H7-CHCOO 2Li C3H7. (i) C3H7Br ( i i ) H+ C3H7 CHCOOH b) (^ Hg-Ch^ COOH 2LDA THF/HMPA/0°C C4HG-CHC00 C4H9. 2Li (i) C4HgBr ( i i ) H+ C4H9 > HCOOH II Scheme 1. Synthetic pathway for alpha-substituted aliphatic acids. 83 CH. CH-:HCOOH (i) C 3H ?I ( i i ) H+ CH. CH3CH2CH2-C-COOH CH, 2LDA THF/0°C e e ( C H 3 ) 2 C-COO 2Li (i) C 2H 5I (11) H+ CH, I 3 CH,-CH9-C-COOH 3 2 , CH 0 XXV Scheme 2. Synthetic route for alpha,alpha-disubstituted aliphatic acids 84 C3H7 Br9 C3H7 \ 2 CHCOOH . CBr — COOH C3H7 3 C3H7 EtOH/H+ CH3_CH2-CH2 \ (i)Qui no!ine, A C-COOH C2H5~C|' ( i i ) OH" ( i i i ) HH C3H. C3H? :Br - COOEt VII Scheme 3. Outline for synthesis of 2-propyl-(E)-2-pentenoic acid (VII) 85 quinoline. The synthetic product which occurs as a mixture of E:Z geometric isomers (~5:1) was purified by f r a c t i o n a l c r y s t a l l i z a t i o n to give the thermodynamically stable E isomer. The synthesis of 4-keto VPA (Scheme 4), IX, as reported else-where (93) involved application of the c l a s s i c a l acetoacetic acid synthesis. B-substituted carboxylic acids The malonic ester synthesis was employed to synthesize 3-methylvaleric acid and 3-methylhexanoic a c i d . The familiar path-ways of the synthesis involved a l k y l a t i o n of malonate with secondary halides, followed by base and acid hydrolysis to convert the diester to a monoacid through deesterification and decarboxyl-ation . 3-Ethylpentanoic acid VI, and cyclohexylacetic acid X I I I , were synthesized using a one-carbon homologation method as outlined i n Scheme 5, starting from available chemical reagents. N,N-Dialkylsuccinamic acids Formation of N,N-diethylsuccinamic acid XV, and N,N-dibutyl-succinamic acid XVI, proceeded readily by reaction of succinic anhydride and the secondary bases (Scheme 6). Various solvents, ether (162), ethanol (164) and benzene have been used for these reactions. In t h i s study, dissolution of succinic anhydride and i s o l a t i o n of c r y s t a l l i n e product proceeded more r e a d i l y with ethanol than benzene. The IR spectroscopic data of the t e r t i a r y succinamic acids are 86 0 II CH3-C-CH2-C00Et (i) Na OEt ( i i ) C,H,CHCOOEt 7 Br 0 II -•CH3-C-CH-C00Et / C3H7CH-C00Et Cone. HC1/A 0 II CH3-C-CH2-CHC00H C3H? IX Scheme 4. Synthetic sequence for preparation of 2-propyl-4-oxopentanoic acid (IX) 87 XIII Scheme 5. Pathways for Synthesis of the beta-substituted Carboxylic acids 88 C2H5 a) \ NH + C2H5 CH2-CO V C 2 H 5 . 0 0 - - ^N-C-CH 2CH 2C00H C H 2 - C 0X E T 0 H XV b) C 4H G \ NH + CH2-CO CH2-C0' \ C4H9, Benzene C 4H G \ 0 \ M N-C-CH2CH2COOH XVI Scheme 6. Synthetic route for preparation of the succinamic acids 89 i n agreement with l i t e r a t u r e data (165). The IR spectra of the succinamic acids show two characteristic bands at 1690-1725 cm-1 and 1610-1640 cm-1 corresponding to the carboxylic and t e r t i a r y amide carbonyl respectively. The NMR spectra (Figure 6) of the succinamic acids show differences between N,N-diethylsuccinamic acid (XV) and N,N-dibutylsuccinamic acid (XVI). Both compounds have two units between the amide and carboxylic functional groups. While N,N-dibutylsuccinamic acid gives a singlet at 2.66 corresponding to four equivalent hydrogens of the two CH^ units (Figure 6b), N,N-diethylsuccinamic acid spectrum shows a singlet at 2.66 equivalent to less than 4 protons. The ^ H-NMR spectrum of N,N-dibutylsuccinamic acid shows a broad peak at 8.56, for the hydrogen-bonded carboxylic proton, equivalent to two protons. In contrast, the •'•H-NMR of N,N-diethylsuccinamic acid shows a broad peak at 5.56 for the hydrogen-bonded carboxylic proton, equivalent to two protons. I t appears l i k e l y that the amide and carboxylic group interact intraraolecularly, especially i n N,N-diethylsuccin-amic acid. In a study of molecular interactions of N,N-diethyl-succinamic acid and N,N-diisobutylsuccinamic acid by IR spectro-scopy, Antonenko (165) reported that, i n dilut e solutions, the amides probably exist i n a c y c l i c form through interaction of the acidic function with the amide group. Other workers (162) have also suggested the existence of a c y c l i c structure i n N-alkyl-succinamic acid and N,N-dialkylsuccinamic acids. 6. 5-Alkyltetrazoles The synthesis of 5-isoamyltetrazole (XI) 5-heptyltetrazole (X) and 5-cyclohexylmethyltetrazole (XII) has been reported i n the 90 (a) 6(ppm) Figure 6. NMR spectra of (a) N,N-diethyl succinamic acid and (t>) N,N-dibutylsuccinamic acid. 91 l i t e r a t u r e (161) and involved reaction of alkylcyanide with hydra-zoic acid (from NaN3 and acetic acid) i n benzene at 150°C for 96-120 hrs in a sealed tube. The most frequently employed open-reaction method requires reaction of n i t r i l e s with ammonium azides i n dimethylformamide (166). Attempts to apply this method for the synthesis of the 5 - a l k y l t e t r a z o l e s proved u n s u c c e s s f u l . The reaction products were apparently not acidic and possessed d i f f e r -ent melting points from expected values (161). In addition, •'•H-NMR data did not conform to expectations. I t i s known that electron-r e l e a s i n g a l k y l groups attached to the cyano f u n c t i o n a l i t y decreases the r e a c t i v i t y of the a l k y l n i t r i l e towards hydrazoic acid. An alternative procedure (167), requiring reaction of a l k y l -n i t r i l e with aluminium azide (prepared in s i t u from anhydrous aluminium chloride and sodium azide) i n tetrahydrofuran (Scheme 7) was successful i n producing desired products. The disadvantages associated with t h i s method include the excess of sodium azide used and the presence of inorganic aluminium s a l t i n the crude product. As a result of the known d i f f i c u l t i e s (161,166) encountered i n i s o l a t i o n and p u r i f i c a t i o n of the 5-alkyltetrazole, the yields of products were not determined. 7. Diunsaturated derivatives of valproic acid-synthesis and metabolism study  The synthesis of selected analogues of valproic acid have been reported i n the l i t e r a t u r e except for the synthesis of 2 - ( l ' -propenyl)-2-pentenoic acid (VI) and 2-(l'-propenyl)-3-pentenoic acid (XXVI). These two dienoic acids have been previously proposed as possible structures for the major diunsaturated metabolite of 92 CH. a) CH. > ,CHCH2CH2CN NaN3/AlC13 THF, A CH. CH. H 3 N CHCH,CH9-C / 2 2 \\. // • N XII NaN,7AlCl, c) C7H1(.-CN -1 5 THF, A H N C7H1 5-C ( ^ \ II N — N X Scheme 7. Synthetic pathway for 5-alkyltetrazoles 93 CH3-CH=CH CH3—CH2—CH C-COOH VI CH3-CH=CH CH3-CH=CH CH-COOH XXVI valproic acid (93). There are four geometric isomers of VI and three stereoisomers of XXVI. Due to the m u l t i p l i c i t y of the stereoisomers, stereoselective synthetic methods are necessary to characterize these dienoic acid isomers. a. Attempted synthesis of 2-(1'-propenyl)-3-pentenoic acid (XXVI) In attempts to synthesize 2-(1'-propenyl)-3-pentenoic acid, the procedure of Normant (168) was followed. Ethyl a-methoxy-acetate was treated with two equivalents of the Grignard reagent, propenylmagnesium bromide to afford the d i - o l e f i n t e r t i a r y alcohol, XXVII (Scheme 8)- whose structure was established by MS, 1H-NMR and IR spectroscopic data. Mass spectrum of dienol ether, XXVII: m/z 111 (M-CH2OCH3, 100%), 43 (35%), 55 (33%), 41 (27%), 77 (22%), 91 (20%), 45 (17%), 123 (14%), 138 (M-HOH, 9%). NMR spectrum of dienol ether, XXVII: 61.8 (m, 6H, 2CH3-C=), 2.45 (broad peak, 1H, OH), 3.43 (m, 5H, CH20CH3), 5.4-5.8 (m, 4H, 2CH=CH). However, dehydration of the dienol ether followed by 94 CH30CH2C00C2H5 + 2 CH3-CH = CHMgBr • THF/ether (CH3 - CH = CH)2> C-CH20CH3 OH HCOOH or (COOH), XXVII (CH3- CH = CH)2 =C=CH0CH3 CH3-CH CH3-CH CH \ / CHCOOH AgNO, NaOH CH (CH3-CH=CH)2==C = CHOH CH3-CH CH3-CH = CH = CH \ / H-CHO XXVI Scheme 8. Outline for synthesis of dienol ether in an attempt to prepare 2-(l-propenyl)-3-pentenoic acid. 95 hydrolysis with either anhydrous oxalic acid or formic acid (168) yielded a dark viscous polymeric product. The polymolecular pro-duct was presumed to r e s u l t from polymerization of the diene aldehydes, although t h i s problem was not reported i n s i m i l a r conditions for synthesis of monounsaturated aldehydes (168). b. Aldol Condensation reactions towards synthesis of 2-(l'-propenyl)- 2-pentenoic ac i d , VI A recent study by Kochen et a l . (124) attempted synthesis of 2-(l'-propenyl)-2-pentenoic acid via nucleophilic acylation of 1-bromopropane with "umpolung" or carbonyl carbanion equivalent of crotonaldehyde, followed by saponification. They reported, how-ever, that the s y n t h e t i c products consisted of a mixture of diunsaturated acid isomers of unknown stereochemistry. The synthetic strategic sequence used for the synthesis of 2-(1'-propenyl)-2-pentenoic acid proceeded v i a an aldol condensation of e t h y l 2-pentenoate (E or Z isomer) with propionaldehyde, followed by dehydration of the 6-hydroxy-81.y'-unsaturated ester with dehydration agents of varied stereoselectivity (Scheme 9). NMR and IR s p e c t r a l a n a l y s i s demonstrated that under the k i n e t i c a l l y - c o n t r o l l e d reaction procedure, propionaldehyde adds regioselectively at the a-position of the a,6-unsaturated ester enolates to afford ethyl 2-(l'-hydroxypropyl)-3-pentenoate (XXVIII) as the major product. Addition of aldehydes to enolates of a, 8-unsaturated esters i s reported to occur at either the y- or a-carbon of the esters (169-173). The geometric aspects of the a l d o l condensation r e a c t i o n have been subjected to s e v e r a l i n v e s t -igations (170-173). 96 /O O E« OH (ll)CH3CH2CHO (jj)KH,THF f = I Z.mainr S < (E) 2, joCH 3 C H^sC CH3 CH=C ^ C H C O O E t ^ ^ ^ OOEt 3Z-3'E+3Z-3'Z+3E-3'E 3Z-3'Z (~12%) + CH-j CHj—Cf ™XOOEt COOEt CH 3 C H S S C H ^ + 2Z-3'Z+2Z-3,E+2E-3'Z+2E-3'E 2Z-3'E (-14%) + 2Z-3'E (-7%) ;ooEt 2E-3'Z (-71%) + COOEt 2E-3'E (-22%) y ^ ^ —C O O E t + ^ " V - C O O E t 2E-3'Z (-56%) 2E-3'E(-18%) OOEt OH (i) LDA,HMPA.THF,-78^ \ ^ £ - C O O E t 3CH 2CHO ' E,major (XXVIII) (II) CH 3 2(Z) Ai)MsCI,Et3N,CH2CI2 («)KH,THF r X ^ ^ C O O E t + ^ ^ - C O O E t • ^ - C O O E t 3Z-3'Z (-44%) 2E-3'E (-48*) 2E-3'Z ("8X) Scheme 9. Stereoselective synthetic routes for preparation of 2-(11-propenyl)-2-pentenoic acid. 97 CH3 - CH - CH = CH - COOEt CH3 - CH = CH - CH - COOEt X-anion a-anion In the present study, NMR and IR analysis of the aldol con-densation products indicated that ethyl 2-(l'-hydroxypropyl)-(Z)-3-pentenoate predominated over the corresponding (E)-isomer when ethyl (E)-2-pentenoate was the starting material (Scheme 9). On the other hand, ethyl 2-(l'-hydroxypropyl)-(E)-3-pentenoate pre-dominated over the (Z)-isomer when ethyl (Z)-2-pentenoate was the starting reagent. In p a r t i c u l a r , the (Z)-isomer showed a medium intensity IR absorption band at 985 cm-1 and a strong band at 690 cm-1 corresponding to the Z-configuration at the 6,y-ethylene group i n XXVIII. The E-isomer showed a strong IR absorption band at 985 cm- 1 c h a r a c t e r i s t i c of an E - c o n f i g u r a t i o n at the 8,y-ethylenic group (see Experimental 5-Ib). These results are i n accord with similar findings reported for aldol condensation and al k y l a t i o n reactions (170,172,173). The mechanism of the inversion of the geometrical ( o l e f i n i c ) c o n f i g u r a t i o n from (Z)-precursor to (E)-product i s probably analogous to that proposed by Kende and Toder (170) as shown i n Figure 6. The stereochemical course of the reaction has been based on product s t e r e o s e l e c t i v e l y and to f a c t o r s s t a b i l i z i n g the incip i e n t intermediates. Kende and Toder (170) have asserted that the base-induced deprotonation of the 2Z-precursor occurs preferably from the conformation XXIXa to afford the more stable carbanion, XXXIa and subsequently the 3E-product. The carbanion 98 intermediate XXXIIa was considered to be less stable due to 1,3-a l l y l i c s t e r i c repulsion between R and COOEt (Figure 7). The geometry of the product from the 2E-precursor was considered to arise from the greater s t a b i l i t y of the carbanion XXXIb although 1 , 3 - a l l y l i c non-bonded interactions are not apparent i n XXXIIb. While i t seems less ob vious that the carbanion XXXIb i s more s t a b l e than the c a r b a n i o n XXXIIb as put forward by the authors (170), the p o s s i b i l i t y of s l i g h t differences i n s t a b i l i t y between XXXIb and XXXIIb could explain reports (172,173) that the reaction with 2Z-precursor i s more stereoselective than that with 2E-precursor. Moreover, t h i s study showed the presence of 3E-product together with the major 3Z-product using ethyl (E)-2-pentenoate as s t a r t i n g m a t e r i a l (see IR s p e c t r a l data i n Experimental 6-Ia). c Dehydration of B-hydroxyunsaturated esters (XXVIII) Var ious dehydration agents were used for dehydration of the 8— hydroxyunsaturated ester formed from the a l d o l condensation reaction. Phosphorus pentoxide dehydration of ethyl 2-(l'-hydroxy-propyl)-3-pentenoate, derived from ethyl (E)-2-pentenoate, gave the seven possible isomeric acids of 2,3'-diene VPA (VI) and 3,3'-diene VPA, XXVI (Scheme 9). Figure 8a shows the t o t a l ion chromatogram of the t-BDMS esters of the dienoic acid isomeric mixture after conversion of the ethyl esters to the dienoic acids. Dehydration of ethyl 2-(1'-hydroxypropyl)-3-pentenoate, der-ived from ethyl (E)-2-pentenoate, with p-toluenesulfonyl chloride afforded four diunsaturated isomeric esters (Scheme 9). Figure 9a shows the mass chromatogram of the [M-57]+ ion of the t-BDMS esters of the four diene VPA isomers i n the synthetic product mixture. 99 Conformation Resulting Carbanion Ester Favored Disfavored Favored Disfavored Product (E) XXIXb XXXb XXXIb XXXIIb (Z) R = C H 3 ; R = C H C H 2 C H j | R = C H 3 R = C H 3 ( C H 2 ) n -Figure 7. Stereoisomers in alkylation and aldol reactions of ester enolates. 100 5" ' * CH, - CH - CH CH • COW 5 « 3 / CH, - CH - CH 3,3-diene VPA [3Z-* ; 3Z-3TZ ; 3E-3**c) s- / r CHj - CH • CH V C - COOH 3^ CHj - CH2 - CH 23,-<JieneVPA (2Z (^fe: 22-3*2; 2E-3TZ; 2E-3"E) — i 1 i i i i i — 52 60 6.8 76 TIME{nfa) 84 Figure 8„ Capillary GCMS separation of t-BDMS esters of isomeric diene-VPA. a) before UV irradiation, b) after 6 hr UV irradiation. Peak numbers correspond to 1: ( Z,E)-3,3'-diene; 2: (Z,E)-3,3'-diene; 3: (Z,Z)-3,3'-diene; 4: ( E,E)-3,3'-diene; 5: (Z,E)-2,3'-diene; 6: (E)-2-ene VPA; 7: (E,Z)-2,3'-diene; 8: (Z,Z)-2,3,-diene; 9: (E,E)-2,3'-diene. 101 2 I u m/z 197 -» 1 i — % i 6.0 7.0 8.0 9.0 10.0 11.0 TIME (min) B. r I m/z 197 6.0 70 8.0 ' . 9>0 TIME (min) 10.0 11.0 Figure 9. Mass chromatograms of t-BDMS esters of a) four diene-VPA isomeric mixture prepared using p-TsCl, b) diene-VPA metabolites in urine extract. Peak 1: (Z,Z)-3,3'-diene; 2: (Z,E)-2,3'-diene; 3: ( E,Z)-2,3'-diene; 4: (E,E)-2,3'-diene. 102 The combination of methanesulfonyl chloride and potassium hydride i n the dehydration of the B-hydroxyunsaturated esters gave the minimum number of isomeric dienoic acids (Scheme 9) . These r e s u l t s , along with the studies of Kende and Toder (170), demon-strate the highly stereoselective nature of CH2SO2CI-KH i n the dehydration reactions. d. Photochemical Isomerization Photochemical isomerization of the unsaturated acids provided an indirect method of determining the positional isomers and their c a p i l l a r y GC retention times. Cis and trans a,B-unsaturated acids have been reported to isomerize photochemically to c i s and/or trans 8,y-unsaturated acids (174,175). a,8-Unsaturated acid esters usually possess a strong absorption band i n the 220-250 nm region compared to l e s s strong absorption of t h e i r isomeric B , y -unsaturated compounds. UV i r r a d i a t i o n of the seven isomeric acid mixture (from dehydration of B-hydroxyunsaturated esters with P2O5) resulted i n a s t r i k i n g build-up of four peaks (peaks 1, 2, 3, 4 i n F i g . 7) which were assigned to B,y-B',y'-diunsaturated acids ( V I I ) . Peaks 1 and 2 can be described as diastereoisomers of 3Z-3'E-diene VPA or one of the peaks may be due to a diene VPA with a terminal double bond. The peaks whose height decreased with UV i r r a d i a t i o n (peaks 7, 9 i n F i g . 8) were described as trans a,B-diunsaturated acids. The peaks whose height did not show any s i g n i f i c a n t change (peaks 5, 8) were described as c i s a,B-diunsaturated esters since under the conditions of photoisomerization they can be formed from the corresponding trans a,B-isomers and i n turn isomerize to the B,y-isomers (174,175). 103 e. GC Elution Order i n GCMS Analysis A d d i t i o n a l support f o r a t e n t a t i v e assignment of stereo-chemical configuration of the dienoic acid peaks in Figure 8 was obtained from the GC retention data. Shorter retention times for the cis-isomer of a given position compared to the trans-isomer on non-polar columns have been frequently observed (176,177). The two peaks before peak 1 i n Figure 8b, which were confirmed using authentic samples to be 3-ene VPA and c i s 2-ene VPA, were produced by photochemical isomerization of trans 2-ene VPA (peak 6) added as an internal standard. With a tentative description of the GC elution order (12.5m dimethylsilicone column) of the seven isomeric acids, peaks 3 and 4 i n Figure 9a were due to trans a,8-8,y'-diunsaturated acids. This was supported by NMR spectra of the four diene VPA isomeric mixture obtained from dehydration of the 8-hydroxyunsaturated ester with p-toluenesulfonyl chloride. The intensity of the •''H-NMR t r i p l e t s at 66.83 and 67.0 (see Experimental l i b ) due to the trans B-vinyl proton indicated the high proportion of 2E-3'Z and 2E-3'E isomers i n the mixture (Scheme 9) . f . Argentation TLC The synthesized isomeric mixtures of dienoates obtained from dehydration of the major products, ethyl 2-(l'-hydroxypropyl)-(Z)-3-pentenoate and ethyl 2-(l'-hydroxypropyl)-(E)-3-pentenoate with CH3SO2CI-KH were subjected to p u r i f i c a t i o n by preparative argentation t h i n layer chromatography. The o b j e c t i v e was to 104 i s o l a t e the isomers from the bands on the TLC plates, characterize the NMR spectral data for each isomer and confirm stereochemical designation of the isomeric dienoates and dienoic acids. Table 1 shows the i d e n t i t y , proportion and Rf values of the isomers i n the three bands, following argentation TLC. The prop-ortion of isomers i n the mixture of dienoates before and after argentation TLC was determined by NMR analysis. Figure 10 shows the p r o f i l e of components i n the three bands following argentation TLC and GCMS analysis of the ethyl esters on a Dexsil 300 packed column. The order of elution of isomeric dienoates on the Dexsil 300 column and the mobility of the isomers on the TLC plate (Table 1) agree with the chromatographic properties of the stereoisomers. Trans-trans diunsaturated esters are reported to be less polar (high Rf) than their corresponding c i s - c i s , trans-cis or cis-trans isomers (177-178). Moreover conjugated dienoates have been reported to have higher Rf values than their non-conjugated congeners using similar eluents (181,182). Mass spectra of the four dienoic acid ethyl esters show the intense M+, m/z 168 of the 2,3'-dienes compared to the 3,3'-dienes (Table 1) indicating the effect of conjugation i n 2,3'-dienes. Diene ethyl esters eluted from the TLC bands could be isolated either as a single isomer or a mixture of isomers (Figure 10). P u r i f i e d synthesized products were characterized by NMR spectro-scopy and Table 2 summarizes the NMR data for the dienoate isomers. The chemical s h i f t s assigned the o l e f i n i c protons, 8-vinyl proton, methylene and methyl protons adjacent to the double bands are characteristic of the stereochemical features of the dienoate 105 Table 1 Composition of chromatographic data of a mixture of synthesized isomeric dienoates Diene VPA Ethyl Ester Product-ratio with reactant, ethyl 2-pentenoatea E Z TLCD  Rf tRC (min) Mass spectrum (70ev) m/z ( r e l . intensity) 3Z-3»Z 0.44 0.42-0.49 (Band la) 9.37 95 (100%), 168 (7%), 139 (2%), 122(1%) 2Z-3»E 0.07 0.53-0.62 (Band I l l b ) 10.0 95 (100%), 168 (50%), 140 (41%), 122 (24%) 2E-3'Z 0.71 0.08 0.45-0.53 (Band l i b ) 10.5 95 (100%), 168 (57%), 140 (41%), 122 (28%), 153 (4%) 2E-3'E 0.22 0.48 0.53-0.62 (Band I l i a , b) 11.1 95 (100%), 168 (54%), 140 (35%), 122 (32%), 153 (4%) r a t i o of isomers i n s y n t h e t i c product mixture before TLC using either ethyl (Z) or (E)-2-pentenoate as reactant. band i n which isomer i s concentrated. GCMS retention time of isomeric peaks analyzed with 3% Dexsil 300 column (1.8m x 2 mm i.d.) Helium, 25 ml/min. Column temp of 50°C to 280°C at rate of 8°C/min. 106 Ia 1 I Ilia 3 u nb 11 1Mb 0 TO" m/z168 —1% TIME (min) Figure 10. GCMS analysis of dienoates eluted from TLC plates. Total ion chromatograms I, II, III correspond to ethyl esters in bands I, II, III respectively, a and b refer to esters synthesized from(Z) and (E)-2-pentenoate respectively. Peak 1: (Z,Z)-3,3'-diene; 2: (Z,E)-2,3'-diene; 3: (E,Z)-2,3'-diene; 4: (E,E)-2,3'-diene0 107 Table 2 NMR (400 MHz) data for diene VPA ethyl esters Dienote CH3 CH3-C= CH2 CH H(3')a H(4')a H(3)a 2E-3'E 1.04(t) 1.84(d) 2.31(m) 6.17(d) J=16Hz 6.08(dq) J=16Hz 6.59(t) 2E-3»Z 1.0A(t) 1.54(dd) 2.11(m) 6.01(d) J=11.4Hz 5.79(dq) J=11.4Hz 6.79(t) 2Z-3'E 1.04(t) 1.70(d) 2.44(m) b b 5.92(t) 3Z-3'Z 1.66(dd) 1.69(d) 3.5(t) 5.5-5.6 (m) 5.6-5.7 (dq) aPosition of hydrogen in the branched-carboxylic acid ester (Scheme 9). ^Resonance peaks were very weak due to small amounts of isomer obtained. 108 isomers and agree with l i t e r a t u r e precedence (170,179,180). The structures of the 2,3'-dienes and 3,3'-dienes are shown i n Scheme 9. The stereochemistry of the isomers i s c l a r i f i e d by the observ-ation that the 2E-3'Z and the 3Z-3'Z isomers have the less deshielded methyl (CH3~C=) doublets while the 2E-3'E and 2Z-3'E isomers have the more deshielded methyl (CH3-C=) doublets. Similarly the 2E-3'Z has the least deshielded methylene (CH2-C=) multiplets i n comparison with those of 2E-3'E and 2Z-3'E. The 2E-3'E and 2E-3'Z have the more deshielded 8-vinyl proton occurring as a t r i p l e t at 66.59 and 66.79 respectively. 2E-3'E and 2E-3'Z can also be differentiated by the coupling constants obtained for the coupling of the o l e f i n i c protons, H(3') and H(4') i n Table 2. g. I d e n t i f i c a t i o n of the Major and Minor Diene VPA Metabolites The i d e n t i f i c a t i o n of one of the minor diene VPA metabolites has been reported to be 2-propyl-(E)-2,4-pentadienoic acid (124,125). In t h i s study, the t-BDMS derivatives of 2E-3'Z-diene VPA and 2E-3'E diene VPA were found to have i d e n t i c a l retention times with the minor and major diene VPA metabolites respectively i n human urine (Figures 8 and 9). With the acquisition of a synthetic sample of 2-propyl-(E)-2,4-pentadienoic aci d , i t was included i n the GCMS analysis of synthetic dienoic acid derivatives and urine metabolites. Separation of the t-BDMS and TMS derivatives was accomplished using a 25m long SE-54 and 0V-1701 columns. Figure 11 shows the structures of the diunsaturated derivatives of valproic acid which were considered i n the i d e n t i f i c a t i o n of the diunsaturated metab-o l i t e s . The retention times of the t-BDMS derivatives of 2-propyl-109 5 ' 5 4' CH 4 CH 3' CH V •/ CH - COOH 5' CH 5 CH: 4' 3' 3 - CH = C 2 t 4 3 / CH2 - CH - COOH 3,3'-DIENE VPA 2,3'-DIENE VPA •3'E ; 3Z-3'Z ; 3E-3'E) (2Z-3'E ; 2Z-37 ; 2E-37 ; 2E-3'E) CH3 - CH2 - CH2 \ - COOH CH2 = CH - D T CH2 = CH - CH CH - COOH CH2 = CH - CH2 2,4-EIENE VPA (2E j 2Z) M'-DIENE VPA Figure 1 1 . Chemical structures of diunsaturated derivatives of valproic acid investigated as the potential metabolites of valproic acid. 110 (E)-2,4-pentadienoic acid and 2-[(Z)-l'-propenyl]-(E)-2-pentenoic acid as well as the minor diene VPA metabolite were a l l i d e n t i c a l . To resolve the problem of which of the two dienoic acids can be ascribed to the minor metabolite, TMS derivatives were analyzed by c a p i l l a r y GCMS. Separation of the TMS derivatives of the dienoic acids was accomplished (Figure 12) and indicated that one of the two minor diene VPA metabolites and 2-propyl-(E)-2,4-pentadienoic acid had i d e n t i c a l retention times which was different from that of 2-[(Z)-l'-propenyl]-(E)-2-pentenoic acid (peak 3 i n Figure 12a). The major diene VPA metabolite (peak 5 i n Figure 12b) had the same retention time as 2-[(E)-l'-propenyl]-(E)-2-pentenoic a c i d . This confirmed e a r l i e r results obtained with the t-BDMS derivatives of dienoic acids i n a synthesized product sample and i n a urine ex-tract (Figures 8, 9). The small peak i n Figure 12b had a r e l a t i v e retention time similar to that of 2-[(E)-l-propenyl]-(Z)-2-pentenoic acid (peak 4 i n Figure 12a), a by-product i n the synthesis of 2-[(Z)-l'-propenyl]-(E)-2-pentenoic a c i d . This diene has been tentatively suggested to be another diene VPA metabolite. On examination of the stereochemical outcome of the aldol condensation and subsequent dehydration reactions, i t i s clear that there i s an inversion of the geometric configuration i n transform-ing the a8-unsaturated ester to the 6'y'-unsaturated B-hydroxy ester while the 2E-isomer i s favoured over the 2Z-isomer i n the dehydration reactions with the more selective dehydration agents. However, the presence of s i g n i f i c a n t amounts of the 3Z-3'Z-diene VPA i n the dehydration of 2-[-l'-hydroxypropyl]-(E)-3-pentenoate (XXVIII) i s i n contrast to the established direction of the 111 (a) m/z197 TIME (min) I • I •» • i • 4.0 4B i -1 • i i 1 I•I 5.6 64 TIME (min) Figure 12. Capillary GCMS separation of TMS derivatives of (a) synthesized 2,3'-diene VPA and 2,4-diene VPA; (b) diene VPA metabolites in urine extract. Peak 1: (Z)-2,4-diene VPA; 2: ( 2,4-diene VPA; 3: (E,Z)-2,3'-diene VPA; 4: (Z,E)-2,3'-diene VPA; 5: (E,E)-2,3'-diene VPA. 112 dehydration reaction of 2-[-l'-hydroxypropyl]-(Z)-3-pentenoate. This unexpected by-product could arise from deconjugation and inversion of geometric configuration during elimination of mesylate with KH. I t i s evident from the "''H-NMR spectra of synthesized dienoates that the chemical s h i f t s are consistent with the stereochemical assignment of the 2,3'-diene VPA isomers. This was not the case with the 3,3'-diene VPA isomers. The geometry about the double bond could not be determined from the NMR pattern alone except when combined with expected GC elution order (Figures 8 and 9) and photoisomerization results (Figure 8 ) . HPLC determination of l i p o p h i l i c i t y Assay method The RP-HPLC procedure was developed systematically by i s o -c r a t i c chromatographic a n a l y s i s of reference compounds under different conditions. Detection of the compounds was accomplished by UV absorption. The selection of the wave length of 210 nm was based on the absorption spectra of those compounds which show maximum absorption i n the range of 205-215 nm (Figure 13). Methanol and a c e t o n i t r i l e were chosen as the organic co-solvents, i n anticipation of s e l e c t i v i t y differences between these two s o l -vents and th e i r individual effects on the octanol-water p a r t i t i o n c o e f f i c i e n t values as determined by RP-HPLC. RP-HPLC i s generally considered to be a combination of part-i t i o n (dynamic e q u i l i b r i u m c o n d i t i o n s ) and adsorptive (non-equilibrium conditions) processes. Minimization of the adsorption 113 UV spectrum of tiN-Obutytouccinwnic add UV spectrum of 2,3'-Dtone VPA 300 Wavetength,nm Uttraviotet spectrum of Valproic acid Wsvslertgttvwn UV spectrum of SHsoamyltetrazole 200 - i 1 r- 300 WavalengtMm Figure 13. UV absorption spectra of four acidic compounds. 114 of compounds on residual s i l a n o l s i t e s ensures hydrophobic i n t e r -actions as the sole process i n retention of compounds evidenced by the correlation between retention factors and log P. The seven reference compounds selected for the correlation of retention f a c t -ors and log P are valproic acid and several analogues. The log P values for these compounds have been determined from the shake-flask procedure by Keane et a l . (13). d'Amboise and Hanai (145) used unbuffered mobile phases to study hydrophobic effects of medium chain and long chain normal al i p h a t i c acids i n RP-HPLC. In t h i s study, the retention times (tg) of the polar and ionizable valproic acid analogues were very close using a low percentage of organic modifier i n unbuffered mobile phase (Table 3). Figure 14 shows asymmetric peak shapes or even double peaks for the ionizable compounds using the Hypersil ODS column and the unbuffered acetonitrile-water mobile phase. Reducation of solute adsorption with a higher percentage of organic co-solvent i n unbuffered water (higher eluting strength) caused them to elute closely to each other and to the unretained solvent (Table 4). Sodium phosphate buffer (pH 3.5) was used to suppress form-ation of ionized forms which interacted strongly with the ODS stationary phase. By selecting the appropriate flow rates and varying the composition of the methanol-buffer and a c e t o n i t r i l e -buffer mobile phases, the retention times of the reference com-pounds were determined (Tables 5-7). Longer retention times and broad, t a i l i n g peaks were observed for the highly l i p o p h i l i c com-pound 2-butylhexanoic acid i n methanol-buffer compositions below 115 Table 3 Retention times of reference compounds using unbuffered mobile phase (CH3CN/H20) Compounds Retention Time (min) % CH3CN (v/v) 20% 22% 25% MeOH 1.68 - 2.23 1. Butyric acid 1.66 - 2.03 2. Valeric acid 1.76 1.72 2.09 3. 2-Ethylbutyric acid 1.87 1.79 2.16 4. Hexanoic acid 1.99* 1.94 2.38 5. Valproic acid 2.87* 2.35 -6. 2-Ethylhexanoic acid - 2.42 -7. 2-Butylhexanoic acid 3.55* — — *Broad peak 116 6 2 4 § § 10 Time(min) Figure 14. Superimposed HPLC chromatograms of acidic compounds using unbuffered mobile phase. Column; Hypersil ODS reverse phase column (20cm x 4.6mm). Mobile phase; 20% acetonitrile in water (v/v). Flow rate, 1.0 mL/min. a) solvent, b) valproic acid in solvent, c) heptanoic acid in solvent, d) valeric acid in sol vent. 117 Table 4 Effect of addition of phosphate buffer (pH 3.5), i n mobile phase (MeOH/r^O), on the retention times of reference compounds. Flow rate i s 1.0 mL/min Compounds Retention Time (min) Presence of Buffer Absence of Buffer 60% MeOH 70% MeOH 60% MeOH 70% MeOH Methanol 2.40 2.39 2.28 2.39 1. Butyric acid 3.23 2.94 1.92 2.09 2. Valeric acid 4.36 3.37 2.19 2.27 3. 2-Ethylbutyric acid 5.50 3.91 2.29 2.34 4. Hexanoic acid 6.28 4.18 2.30 2.39 5. Valproic acid 13.85 6.53 3.24* 2.65* 6. 2-Ethylhexanoic acid 13.96 6.58 - 2.95* 7. 2-Butylhexanoic acid >30.0 13.70 - 2.89* *Broad and t a i l i n g peaks. 118 Table 5 Retention times of seven reference compounds at different percentages of MeOH i n the mobile phase (MeOH/O.OlM NaH2P04). Flow rate of mobile phase i s 1.0 mL/min. Compound Retention Time (min) % MeOH (v/v) 50% 60% 70% Methanol 1.54 2.40 2.39 1. Butyric acid 2.49 3.23 2.94 2. Valeric acid 3.80 4.36 3.37 3. 2-Ethylbutyric acid 5.54 5.50 3.91 4. Hexanoic acid 6.85 6.28 4.18 5. Valproic acid 20.45' 13.85 6.53 6. 2-Ethylhexanoic acid 19.45 13.96 6.58 7. 2-Butylhexanoic acid >50.0 >30.0 13.70 119 Table 6 Retention times of seven reference substances at different flow rates (a) In 70% MeOH: 30% 0.01M NaH2P0A mobile phase Compound Retention Time (min) Flow Rates 1.OmL/min 1.2mL/min 1.5mL/min Methanol 2.39 2.03 1.54 1. Butyric acid 2.94 2.45 1.90 2. Valeric acid 3.37 2.83 2.21 3. 2-Ethylbutyric acid 3.91 3.27 2.54 4. Hexanoic acid 4.18 3.50 2.73 5. Valproic acid 6.53 5.49 4.37 6. 2-Ethylhexanoic acid 6.58 5.49 4.40 7. 2-Butylhexanoic acid 13.70 11.46 9.50 (b) In 60% MeOH: 40% 0.01M NaH2P0A mobile phase Compound Retention Flow Time (min) Rates 1.OmL/min 1.5mL/min Methanol 2.40 1.63 1. Butyric acid 3.23 2.19 2. Valeric acid 4.36 2.94 3. 2-Ethylbutyric acid 5.50 3.71 4. Hexanoic acid 6.28 4.25 5. Valproic acid 13.85 9.26 6. 2-Ethylhexanoic acid 13.96 9.32 7. 2-Butylhexanoic acid >30.0 >23.0 120 Table 7 Retention times of seven reference compounds at different percentages of ac e t o n i t r i l e i n the mobile phase Retention Time (min) Compounds (Flow % Rate=1.0mL/min) CH3CN (v/v) (Flow Rate=1.5mL/min) % CH3CN (v/v) 50% 55% 60% 40% 45% 50% Methanol 2.00 2.00 2.00 1.47 1.42 1.37 1. Butyric acid 2.72 2.78 2.70 2.06 1.93 1.85 2. Valeric acid 3.17 3.16 3.00 2.67 2.33 2.18 3. 2-Ethylbutyric acid 3.65 3.52 3.25 3.24 2.72 2.47 4. Hexanoic acid 3.96 3.73 3.40 3.73 3.02 2.69 5. Valproic acid 6.32 5.51 4.63 7.71 5.31 4.30 6. 2-Ethylhexanoic acid 6.34 5.50 4.63 7.77 5.35 4.33 7. 2-Butylhexanoic acid 13.51 10.25 7.88 23.23 13.39 9.40 121 2. 60% and acetonitrile-buffer compositions below 40%. Void time and retention mechanism i n RP-HPLC The unretained elution time or void time ( tQ) i n the RP-HPLC method was determined by the elution time of the mobile phase component, methanol or a c e t o n i t r i l e . Table 8 shows the values of tQ obtained by organic co-solvent and linear regression of homo-logous straight-chain carboxylic acids (^HyCOOH-CyH-j^COOH) accord-ing to the method of Berendsen et a l . (181). For the same flow rate, the void times i n the two methods appear to be similar and stay reasonably constant with increases i n the volume percentage of a c e t o n i t r i l e i n the buffered mobile phase. For the same flow rate, anomalous values are obtained for the methanol-buffer mobile phase. The void times determined by either of the two methods do not appear to decrease or stay constant with increases i n the prop-ortion of methanol i n the mobile phase. Similar findings of void times i n methanol-water mixtures, especially i n the region of 60-70% me t h a n o l , have been r e p o r t e d by s e v e r a l r e -searchers (149,150,181). However, for the same mobile phase, the void time decreases with an increase i n the flow r a t e . Berendsen et a l . (181), i n a study of retention mechanisms i n RP-HPLC, have proposed that the normal effect of decrease i n void time with increase i n volume percentage of methanol or a c e t o n i t r i l e could be explained by solvophobic e f f e c t s . Thus, with a decrease i n volume percentage of organic modifier, the solvation of the C-18 monomeric layer by the organic modifier decreases. Solvation was suggested to be minimal for pure water, hence interaction between 122 the hydrocarbon chain and water eluent i s minimal. According to the authors, minimal solvation implied a maximal internal porosity of the microparticulate supports and hence a maximal hold-up or void time. In comparison of methanol-water and acetonitrile-water effects on void time, Yonker et a l . (149,150) explained the anomalous values of void time i n methanol-water mixtures as due to a station-ary phase phenomenon. They also proposed a model of the stationary phase composed of the C-18 monomeric layer, s i l i c a support and a solvation layer from mobile phase components. With increases i n the volume of percentage of methanol up to 70%, they postulated that the mobile phase components, methanol or water, hydrogen-bond to the exposed s i l i c a support between the C-18 chains and t h i s phenomenon predominates over that of solvation. They explained the plateau with approximately 70% methanol-water (instead of a decrease i n void time) to be due to the dominating effect of solv-ation layer on C-18 hydrocarbon chain. Void times of the mobile phase component, methanol i s then held up by interaction with the solvation layer on the C-18 chain. On the other hand, the void time determined i n acetonitrile-water mixtures decreases s l i g h t l y or stays nearly constant with an increase i n volume percentage of a c e t o n i t r i l e (Table 8 ) . The effects of acetonitrile-water mobile phase i n t h i s study could be explained accordingly by the lower hydrogen-bonding properties of a c e t o n i t r i l e with the effects of the solvation layer i n opposition to the higher eluting strength of a c e t o n i t r i l e . In view of the d i f f e r i n g stationary phase conditions, void times were determined experimentally for each mobile phase compos-123 Table 8 Comparison of void times ( tQ) determined from (a) i n j e c t i o n of methanol, and (b) dead time i t e r a t i o n of the retention times of the homologous series from C3H7COOH to C7H15COOH Mobile Phase using 0.01M NaH^ PO/ as co-solvent (v/v) Flow Rate (mL/min) tQ(Me0H) (min) tQ( i t e r a t i o n ) a (min) 50% MeOH 1.5 1.54 1.38 60% MeOH 1.0 2.40 2.13 70% MeOH 1.0 2.39 2.33 70% MeOH 1.2 2.03 1.86 70% MeOH 1.5 1.54 1.51 60% MeOH 1.5 1.63 1.33 50% CH3CN 1.0 2.00 2.06 50% CH3CN 1.5 1.37 1.43 55% CH3CN 1.0 2.00 2.02 60% CH3CN 1.0 2.00 1.80 40% CH3CN 1.5 1.47 1.41 45% CH3CN 1.5 1.42 1.38 determined using equation 1 LR,N+1 = atR,N- - t 0 ( a - l ) 124 i t i o n . Capacity factor (k') for each reference compound i n a different mobile phase was then calculated. 3. Eluent effects on capacity factor Mobile phases containing various percentages of organic co-solvent were used to determine which mobile phase provides a s i g n i f i c a n t high correlation for the linear regression equation log P = a + b log k'. Tables 9-11 show the log k' values and the correlation parameters obtained for the various percentages of organic modifier i n buffered mobile phases. Several interesting points emerge from the regression data. For the same organic co-solvent and flow rate, the slope of the plot of log P versus log k' decreased with a decrease i n the per-centage of organic co-solvent i n the mobile phase. Thus s e n s i t i v i t y to changes i n the l i p o p h i l i c character increased with the more aqueous mobile phase, suggesting that the hydrophobic effect i s a function of the p o l a r i t y difference between the mobile and stationary phases. On the other hand, there was v i r t u a l l y no change i n the slope with the same mobile phase but d i f f e r i n g flow rates between 1.0 ml/min and 1.5 ml/min. The regression equations with methanol-buffer mixture between 50-70% methanol and 40-60% acet o n i t r i l e - b u f f e r mixture as mobile phases gave s i g n i f i c a n t l y high correlation c o e f f i c i e n t s . The line a r model explained greater than 97% of the variations observed. The correlation c o e f f i c i e n t increased at lower organic modifier volume percentage due to the greater s e n s i t i v i t y to changes i n l i p o p h i l i c character. However, lower percentages could not elute 125 Table 9 Correlation of log k' and log PQ/W for seven reference compounds at various compositions of the mobile phase (MeOH/O.OlM NaH2P04). log PQ/W = a + b log k' Compound 50% MeOH (FR*=1.0) log k' 60% MeOH (FR=1.0) log k' 70% MeOH (FR=1.0) log k' 70% MeOH (FR=1.2) log k' 70% MeOH (FR=1.5) log k' ** L O8 Po/w 1. Butyric acid -0.2098 -0.4611 -0.6380 -0.6842 -0.6312 0.98 2. Valeric acid 0.1666 -0.0879 -0.3872 -0.4044 -0.3614 1.51 3. 2-Ethylbutyric acid 0.4145 0.1111 -0.1965 -0.2141 -0.1875 1.68 4. Hexanoic acid 0.5376 0.2086 -0.1255 -0.1402 -0.1120 1.93 5. Valproic acid 1.089 0.6786 0.2386 0.2316 0.2643 2.75 6. 2-Ethylhexanoic acid 1.065 0.6827 0.2438 0.2316 0.2688 2.64 7. 2-Butylhexanoic acid >1.498 >1.061 0.6751 0.6670 0.7134 3.20 a 1.2336 1.6279 2.1955 2.2223 2.1576 n = 6 b 1.3348 1.5218 1.9461 1.8822 1.9179 (excluding compound #7) r 0.9957 0.9944 0.9930 0.9928 0.9944 s 0.070 0.080 0.090 0.091 0.080 F 467 355 281 273 357 a 2.1463 2.1755 2.1098 b 1.7618 1.7222 1.726 n = 7 r 0.990 0.991 0.990 s 0.122 0.117 0.120 F 246 270 255 * FR i s Flow Rate (mL/min) of eluent , s i s standard e r r o r of estimate, FQ QJ = 16.0, r i s c o r r e l a t i o n c o e f f i c i e n t , F s t a t i s t i c from analysis of variance ** Values obtained from P.E. Keane et a l . ( 1 3 ) . Table 10 Correlation of log k' and log PQ/W for seven reference compounds at various compositions of the mobile phase (CH3CN/O.OIM NaH2P04). L O8 Po/w = a + b log k* Compound 50% (FR*=1.0) log k' 55% (FR=1.0) log k' 60% (FR=1.0) log k' 40% (FR=1.5) log kf 45% (FR=1.5) log k» 50% (FR=1.5) log k' ** L O8 Po/w 1. Butyric acid -0.4437 -0.4089 -0.4559 -0.3965 -0.4447 -0.4555 0.98 2. Valeric acid -0.2328 -0.2366 -0.3010 -0.0881 -0.1932 -0.2282 1.51 3. 2-Ethylbutyric acid -0.0835 -0.2744 -0.2041 0.0806 0.03834 -0.09533 1.68 4. Hexanoic acid -0.00877 -0.063 -0.1549 0.1870 0.05183 -0.01615 1.93 5. Valproic acid 0.3345 0.2443 0.1189 0.6279 0.4377 0.3301 2.75 6. 2-Ethylhexanoic acid 0.3365 0.2430 0.1189 0.6320 0.4421 0.3346 2.64 7. 2-Butylhexanoic acid 0.7600 0.6154 0.4683 1.1703 0.9258 0.7680 3.20 a 1.9178 2.0623 2.2436 1.6309 1.7941 1.9266 b 1.9104 2.1190 2.4780 1.4793 1.6945 1.8879 n=7 r 0.9880 0.9798 0.9834 0.9895 0.9836 0.9877 s 0.133 0.173 0.157 0.125 0.156 0.135 F 205 120 146 235 148 200 a 1.9506 2.1153 2.3472 1.6236 1.8003 1.9623 b 2.1840 2.4244 2.9534 1.6765 1.9276 2.1740 n=6 r 0.9958 0.9822 0.9968 0.9967 0.9875 0.9968 (excluding compound #7) s 0.069 0.143 0.061 0.062 0.120 0.061 F 478 109 616 607 157 613 * FR i s Flow Rate (mL/min) of eluen t , s i s standard e r r o r of estimate, FQ QJ = 16.0, r i s c o r r e l a t i o n c o e f f i c i e n t , F s t a t i s t i c from analysis of variance ** Values obtained from P.E. Keane e t ' a l . ( 1 3 ) . Table 11 Summary of linear regression parameters for log P versus log k1 Log P = a + b log k' n* = 6 Mobile Phase (Aqueous buffer-organic solvent) Flow Rate (mL/min) a b r s 50% MeOH 1.0 1.2336 1.3348 0.9957 0.070 60% MeOH 1.0 1.6279 1.5218 0.9944 0.080 70% MeOH 1.0 2.1955 1.9461 0.9930 0.090 -70% MeOH 1.2 2.2223 1.8822 0.9928 0.091 70% MeOH 1.5 2.2389 2.0530 0.9944 0.080 50% AcN 1.0 1.9506 2.1840 0.9958 0.069 55% AcN 1.0 2.1153 2.4244 0.9822 0.143 60% AcN 1.0 2.3472 2.9534 0.9968 0.961 40% AcN 1.5 1.6236 1.6765 0.9967 0.062 45% AcN 1.5 1.8327 1.9334 0.9965 0.120 50% AcN 1.5 1.9623 2.1740 0.9968 0.961 n* = 7 70% MeOH 1.0 2.1463 1.7618 0.990 0.122 70% MeOH 1.2 2.1755 1.7222 0.991 0.117 70% MeOH 1.5 2.1561 1.7278 0.990 0.120 50% AcN 1.0 1.9178 1.9104 0.9880 0.133 55% AcN 1.0 2.0623 2.1190 0.9798 0.173 60% AcN 1.0 2.2436 2.4780 0.9834 0.157 40% AcN 1.5 1.6309 1.4793 0.9895 0.125 45% AcN 1.5 1.8144 1.6842 0.9879 0.156 50% AcN 1.5 1.9266 1.8879 0.9877 0.135 •number of reference compounds (solutes) used; r i s correlation c o e f f i c i e n t , s i s standard error of estimate. 128 the highly l i p o p h i l i c 2-butylhexanoic acid without peak broadening. Consequently when t h i s compound was included, the correlation co-e f f i c i e n t decreased. Thus reduction of the extent of adsorption i n the p a r t i t i o n process using optimum flow rate and eluting strength produced high correlation c o e f f i c i e n t s for the set of compounds. A decrease i n the adsorption phenomenon then ensures optimum rate of exchange and mass transfer of solute between the l i q u i d phases. 4. HPLC log P values of valproic acid and analogues Mobile phases selected for the determination of log P values were 70% methanol-buffer and 50% a c e t o n i t r i l e - b u f f e r . The two mobile phases gave high correlation c o e f f i c i e n t s i n the regression equation and were able to elute the highly l i p o p h i l i c 2-butyl-hexanoic acid without excessive peak broadening. Having selected the mobile phases and experimental conditions the retention times of reference compounds and remaining compounds were determined (Table 12, 13). Figure 15 shows the HPLC chromatograms of the compounds, superimposed for comparison, using 70% methanol-buffer mobile phase. Twenty-three compounds were analyzed including the reference compounds. A l l the compounds showed symmetrical and sharp peaks. Values of log P were calculated from the respective regression equation below and are shown i n Tables 12, 13. log P = 2.1463 + 1.7618 log k' n = 7, r = 0.990, s = 0.122, F = 246 (Mobile phase: 70% MeOH-30% NaH2P04, 0.01M) log P = 1.9506 + 2.1840 log k1 n = 7, r = 0.9958, s = 0.069, F = 478 (Mobile phase: 50% CH3CN - 50% NaH2P0A, 0.01M) 129 Table 12 HPLC method for determining the l i p o p h i l i c i t i e s of the ac i d i c compounds using 70% MeOH: 30% 0.01M NaH2P04 as mobile phase3 Compounds tR (min) log k' l oS PHPLC l o8 pbo/w 1. Isobutyric acid 3.00 -0.5931 1.10 2. Butyric acid 2.94 -0.6380 1.02 0.98 3. Trimethylacetic acid 3.47 -0.3450 1.54 4. Valeric acid 3.37 -0.3872 1.46 1.51 5. 2-Ethylbutyric acid 3.91 -0.1965 1.80 1.68 6. 2,2-Dimethylbutyric acid 4.12 -0.1403 1.90 7. Hexanoic acid 4.18 -0.1255 1.93 1.93 8. Heptanoic acid 5.49 0.1130 2.35 9. 3-Ethylpentanoic acid 4.93 0.02644 2.19 10. Cyclohexylacetic acid 5.35 0.09289 2.31 11. 1-Methylcyclohexane-l-carboxylic acid 5.65 0.1348 2.38 12. Valproic acid (VPA) 6.53 0.2386 2.57 2.75 13. 4-Keto VPA 3.19 -0.4753 1.31 14. 2-Ene VPA 5.89 0.1658 2.44 15. 2,3'-Diene VPA 5.21 0.07185 2.27 16. 2-Ethylhexanoic acid 6.58 0.2438 2.58 2.64 17. Octanoic acid 7.66 0.3434 2.75 18. 2-Butylhexanoic acid 13.70 0.6751 3.34 3.20 19. 5-Isoamyltetrazole 3.17 -0.4863 1.29 20. 5-Heptyltetrazole 4.39 -0.07737 2.01 21. 5-Cyclohexylmethyl-tetrazole 3.54 -0.3177 1.59 22. N,N-Diethylsuccinamic acid 5.20 0.07031 2.27 23. N,N-Dibutylsuccinamic acid 5.31 0.08699 2.30 aFlow rate i s 1.0 mL/min, tQ(Me0H) i s 2.39 min. 'Used i n l i n e a r c o r r e l a t i o n of log PQ/W versus log k'. Values taken from P.E. Keane et a l . (13). 130 • Table 13 HPLC method for determining the l i p o p h i l i c i t i e s of the acidic compounds using 50% CH3CN: 50% 0.01M NaH2P0A as mobile phase3 Compounds tR (min) log k' l o§ PHPLC ! ° 8 pbo/w 1. Isobutyric acid 2.72 -0.4437 1.07 2. Butyric acid 2.72 -0.4437 1.07 0.98 3. Trimethylacetic acid 3.24 -0.2076 1.52 4. Valeric acid 3.17 -0.2328 1.47 1.51 5. 2-Ethylbutyric acid 3.65 -0.0835 1.76 1.68 6. 2,2-Dimethylbutyric acid 3.82 -0.04096 1.84 7. Hexanoic acid 3.96 -0.00877 1.90 1.93 8. Heptanoic acid 5.17 0.2000 2.30 9. 3-Ethylpentanoic acid 4.61 0.1156 2.14 10. Cyclohexylacetic acid 4.97 0.1717 2.25 11. 1-Methylcyclohexane-l-carboxylic acid 5.24 0.2095 2.32 12. Valproic acid (VPA) 6.32 0.3345 2.56 2.75 13. 4-Keto VPA 3.03 -0.2882 1.37 14. 2-Ene VPA 6.08 0.3096 2.51 15. 2,3'-Diene VPA 4.99 0.1746 2.25 16. 2-Ethylhexanoic acid 6.34 0.3365 2.56 2.64 17. Octanoic acid 7.31 0.4241 2.72 18. 2-Butylhexanoic acid 13.51 0.7600 3.37 3.20 19. 5-Isoamyltetrazole 3.00 -0.2967 1.35 20. 5-Heptyltetrazole 4.09 0.01912 1.95 21. 5-Cyclohexylmethyl-tetrazole 3.52 -0.1192 1.69 22. N,N-Diethylsuccinamic acid 4.92 0.1651 2.23 23. N,N-Dibutylsuccinamic acid 5.35 0.2240 2.35 aFlow rate i s 1.0 mL/min, tQ(Me0H) i s 2.0 min. Taken from P.E. Keane et a l . (13). Values used i n l i n e a r c o r r e l a t i o n of log PQ/W versus log k'. 131 4 6 § 10 12 14 16~ Time(min) Superposed HPLC chromatograms of a c i d i c compounds. Column; hypersil ODS reverse phase column (20cm x 4.6mm). Mobile phase; 70% Methanol:30% 0. 01M NaH2P04 (pH3.5). Flow r a t e , 1.Oml/min. S,solvent; 1. Butyric acid 2. Isobutyric acid 3. 5-Isoamyltetrazole 4. 4-Keto VPA 5. Valeric acid 6. Trimethylacetic acid 7. 5-Cyclohexylmethyltetrazole 8. 2-Ethylbutyric acid 9. 2,2-Dimethylbutyric acid 10. Hexanoic acid 11. 5-Heptyltetrazole 12. 3-Ethylpentanoic acid 13. N.N-Diethylsuccinamic acid 14. 2.3-Diene VPA 15. N.N-Dlbutylsuccinamic acid 16. Cyclohexylacetic acid 17. Heptanoic acid IB. 1-Methylcyclohexane carboxylic acid 19. 2-Ene VPA 20. Valproic a d d (VPA) 21. 2-Ethylhexanoic acid 22. Octanoic acid 23. 2-Butylhexanoic acid 132 5. Comparison of l i p o p h i l i c i t y from RP-HPLC and other methods Retention of the test compounds on the RP-HPLC column appeared to be based on the hydrocarbon structure, the presence of polar functional groups, the type of a l k y l substitution, the number of double bonds, the presence of a ring structure and molecular size (Tables 12, 13). As shown i n Tables 12 and 13, the HPLC log P values determined from both mobile phases are i n good agreement with an average difference of ± 0.06 log u n i t s . The shake-flask method was used to determine the octanol-water p a r t i t i o n c o e f f i c i e n t of four compounds of diverse structures i n order to ve r i f y the accuracy of the HPLC log P values. Tables 14-17 show the c a l i b r a t i o n data for the four compounds determined using HPLC analysis. High correlation c o e f f i c i e n t s were obtained for the c a l i b r a t i o n curves. The log P values of the four compounds determined by the shake-flask method are shown i n Table 18. The log P values are for the unionized form of the acidic compounds since 0.01N HC1 was used as the aqueous phase. The precision of the log P values was high with average standard deviations less than 0.05 log u n i t s , except for N,N-dibutylsuccinamic acid with a deviation of 0.08 log u n i t s . Probable sources of error include low s e n s i t i v i t y for detection of these compounds by HPLC and propensity of N,N-dibutylsuccinamic acid to be unstable i n an ac i d i c aqueous phase. In addition, dissolution of tetrazoles i n the 0.01N HC1 phase was d i f f i c u l t and had to be effected using small volumes of methanol. 133 Table 14 Calibration curve data of trimethylacetic acid i n 0.1N HC1 Concentration mg/mL Mean Peak Height Linear Regression Parameters3 0.05 193 0.1 392 a0 = 22.81 0.2 767 a1 = 3639 0.4 1465 r = 0.9999 0.6 2210 r2 = 0.9998 i s the c o e f f i c i e n t of determination, a-1- i s the slope and a° i s the i n t e r c e p t . Equation f o r the l i n e i s y = a*x + a0 where y i s the peak height and x i s concentration of compound. 134 Table 15 Calibration curve data of N,N-Dibutylsuccinamic acid i n 0.1N HC1 Concentration mg/mL Mean Peak Height Linear Regression Parameters3 0.1 128 a° = 3.14 0.2 260 a1 = 1321 0.4 530 r = 0.99994 0.6 787 r2 = 0.99988 r i s the coe f f i c i e n t of determination, a i s the slope and a i s the intercept. Equation for the l i n e i s y = a x + a where y i s the peak height and x i s concentration of compound. 135 Table 16 Calibration curve data of 5-Isoamyltetrazole i n 0.1N HC1 Concentration mg/mL Mean Peak Height Linear Regression Parameters3 0.01 1099 a° = 235.9 0.02 2153 a1 = 92742 0.05 4889 r = 0.9999 0.1 9496 r2 = 0.9998 r2 i s the c o e f f i c i e n t of de t e r m i n a t i o n , a l i s the slope and a° i s the i n t e r c e p t . Equation f o r the l i n e i s y = a^x + a° where y i s the peak height and x i s concentration of compound. 136 Table 17 Calibration curve data of 5-cyclohexylmethyltetrazole i n 0.1N HC1 Concentration mg/mL Mean Peak Height Linear Regression Parameters3 0.01 455 a° = -72.67 0.02 974 a1 = 52782 0.05 2581 r = 0.99999 0.1 5200 r2 = 0.99998 r2 i s the c o e f f i c i e n t of determination, a l i s the slope and a° i s the i n t e r c e p t . Equation f o r the l i n e i s y = a^x + a° where y i s the peak height and x i s concentration of compound. 137 Table 18 Octanol-water p a r t i t i o n c o e f f i c i e n t s of selected acidic compounds determined by the shake-flask procedure Compound log PQ/W (+SDb) 1. Trimethylacetic acid 2. N,N-Dibutylsuccinamic acid 3. 5-Isoamyltetrazole 4. 5-Cyclohexylmethyltetrazole 1.54 + 0.01 2.27 + 0.08 1.38 + 0.03 1.61 + 0.04 a Shake-flask method used with 1-octanol as organic phase and 0.01N HC1 as aqueous phase. k Standard d e v i a t i o n i n log u n i t s , from four separate experimental measurements. 138 Table 19 Hansch-ir - values3 used i n calculating log PQ/W A l i p h a t i c Group 1. CH3 0.5 2. CH2 0.5 3. -OH -1.16 4. -0- -0.98 5. C00H -0.65 6. C=0 -1.21 7. C0NH2 -1.71 8. NH2 -1.19 9. -CON- -2.27 10. N -1.32 11. Tetrazole -1.04b 12. NMe2 -0.32 13. Double bond -0.30 (ATT) 14. Chain branch (single) -0.20 (ATT) 15. Branching i n ring closure -0.09 (Air) 16. Ring closure (per bond) -0.09 (ATT) 17. Intramolecular H-bonding -0.65 (ATT) aTaken from A. Leo et a l . (133). \ ( t e t r a z o l e ) obtained from C. Hansch and A. Leo (152). 139 Table 20 Rekker's fragmental values ( f )a used i n calculating log PQ/W Fragment f ( a l i p h a t i c ) 1. CH3 0.702 2. CH2 0.527 3. CH 0.236 4. C 0.14 5. H 0.175 6. CH2 = CH 0.93 7. COOH -1.003 8. C = 0 -1.69 9. NH2 -1.38 10. NH -1.864 11. -N- -2.133 12. CON -2.894 13. C0NH2 -1.99 14. Tetrazole -2.93 15. NH (heterocyclic) -0.70 16. N (heterocyclic) -1.06 17. C (heterocyclic) 0.157 18. H (heterocyclic) 0.199 19. CH (heterocyclic) 0.35 20. Single conjugated pattern 0.314 (Af) 21. Proximity effect for 1 C separation 0.80 22. Proximity effect of electronegative group 0.46 for 2C separation 23. -N = C-NH -0.79 24. -N = N -2.14 aTaken from R. F. Rekker (153). 140 Table 21 L i p o p h i l i c i t i e s (log PQ/W) of the acidic compounds obtained by different methods Compounds l o8 Po/w HPLC3 (MeOH) HPLCb (CH3CN) Hansch0 Rekkerd Shake-Flask 1. Isobutyric acid 1.10 1.07 0.65 0.64 2. Butyric acid 1.02 1.07 0.85 0.76 0.98e 3. Trimethylacetic acid 1.54 1.52 0.95 1.24 1.54f 4. Valeric acid 1.46 1.47 1.35 1.28 1.51e 5. 2-Ethylbutyric acid 1.80 1.76 1.65 1.69 1.68e 6. 2,2-Dimethylbutyric 1.90 1.84 1.45 1.45 acid 7. Hexanoic acid 1.93 1.90 1.85 1.81 1.93e 8. Heptanoic acid 2.35 2.30 2.35 2.33 9. 3-Ethylpentanoic 2.19 2.14 2.15 2.22 acid 10. Cyclohexylacetic 2.31 2.25 2.22 2.39 acid 11. 1-Methylcyclohexane- 2.38 2.32 2.13 2.47 1-carboxylic acid 12. Valproic acid (VPA) 2.57 2.56 2.65 2.75 2.75e 13. 4-keto VPA 1.31 1.37 1.14 0.99 14. 2-Ene VPA 2.44 2.51 2.35 2.56 15. 2,3-Diene VPA 2.27 2.25 2.05 2.26 16. 2-Ethylhexanoic 2.58 2.56 2.65 2.75 2.64e acid 17. Octanoic acid 2.75 2.73 2.85 2.86 18. 2-Butylhexanoic acid 3.34 3.37 3.65 3.80 3.20e 19. 5-Isoamyltetrazole 1.29 1.35 1.26 -1.03 1.38f 20. 5-Heptyltetrazole 2.01 1.95 2.46 0.83 21. 5-Cyclohexylmethyl- 1.56 1.69 1.85 0.50 1.61f tetrazole 22. N,N-Diethylsuccinamic 2.27 2.23 0.08 0.08 acid 23. N,N-Dibutylsuccinamic 2.30 2.35 2.08 2.18 2.27f acid 3 determined using 70% MeOH: 30% 0.01M NaHoPO, b determined using 50%CHoCH: 50% 0.01M NaH2P04  0 calculated using Hansch ir-approach calculated using Rekker's fragment constant ^ experimentally determined by P. E. Keane et a l . (13) octanol-water p a r t i t i o n c o e f f i c i e n t determined i n t h i s study 141 Hansch-rr and Rekker-f values (Tables 19, 20) have been used to predict log P values of the compounds studied (Table 21). Com-parison of the HPLC log P values with Hansch and Rekker values i n Table 21 shows that there i s good agreement for the homologous s t r a i g h t - c h a i n and alpha-branched a l i p h a t i c a c i d s , C^HgCOOH-CyH-j^COOH. The discrepancies i n log P values for the HPLC and calculated methods are, however, for highly substituted compounds (e.g. trimethylacetic a c i d ) , substituted a l i c y c l i c compounds (e.g. 1-methylcyclohexanecarboxylic a c i d ) , tetrazoles, intramolecular bonded compounds e.g. N,N-diethylsuccinamic acid (165) and 4-keto VPA (162). The shake-flask log P values of the highly-substituted t r i -m ethylacetic a c i d , h i g h l y l i p o p h i l i c 2-butylhexanoic a c i d , tetrazoles and less l i p o p h i l i c butyric acid agree better with HPLC log P values than the calculated log P values (Table 21). This indicates that the Hansch and Rekker log P values may lead to less accurate values for compounds showing constitutive effects on hydrophobicity. Obviously, a d d i t i v i t y of group contribution does not hold f o r the t e t r a z o l e s and the log P value of the un-substituted tetrazole would have to be used, as noted i n Table 19. There i s close agreement among the various methods with the shake-flask value for N,N-dibutylsuccinamic acid (XVI). Lower values from a d d i t i v i t y methods were obtained for N,N-diethylsuccinamic acid (XV) which may show s i g n i f i c a n t intramolecular effects to increase the log P value. 142 6. Intramolecular bonding effects of amic acids The HPLC log P value for N,N-diethylsuccinamic acid (XV) was considerably higher than predicted (Table 21). This could be explained by a greater intramolecular bonding for N,N-diethyl-succinamic acid than for N,N-dibutylsuccinamic a c i d . In addition, the NMR spectra of N,N-diethylsuccinamic appear to indicate an interaction between the carboxylic group and the amido group com-pared to that of N,N-dibutylsuccinamic ac i d (Figure 6). A study (165) of NMR spectra of N,N-dialkylsuccinamic acids also showed evidence of an interaction between the amido and carboxyl functional groups i n N,N-diethylsuccinamic acid through hydrogen bonding i n d i l u t e solutions. Another aspect of amic acids i n v o l v e s t h e i r s t a b i l i t y i n acidic media. Both N,N-dibutylsuccinamic and N,N-diethylsuccinamic are soluble i n water. N,N-dibutylsuccinaraic acid was r e l a t i v e l y s t a b l e i n 0.1N HC1 and 0.01M NaH2P04 as shown by the high correlation c o e f f i c i e n t of i t s linear c a l i b r a t i o n curves used to determine the concentration i n the aqueous phase of the octanol-0.1N HC1 p a r t i t i o n system. The long-term s t a b i l i t y of N,N-dibutyl-succinaraic acid was not investigated. On the other hand, N,N-diethylsuccinamic was not stable i n 0.1N HC1 and i t s p a r t i t i o n c o e f f i c i e n t i n the octanol/O.lN HC1 system could not be determined. In order to observe chromatographic peaks, amounts of 20ug or greater of N,N-diethylsuccinamic acid i n methanol or a c e t o n i t r i l e had to be injected. There i s l i t e r a t u r e precedence (183,184) for the i n s t a b i l i t y of some amic acids i n acidic media. I t has been documented that 143 the presence of a carboxylic group adjacent to an amide within the molecular structure usually leads to hydrolysis of the amic a c i d . The degree of hydrolysis appears to depend on the type of amic acid and N - a l k y l s u b s t i t u e n t s (183,184). Thus the h y d r o l y s i s i s reported to be even greater i n N-alkylmaleamic acids (182,183) where the cis-configuration favours intramolecular interactions. In a k i n e t i c study of N-alkylmalearaic acid hydrolysis, Kluger and Chin (184) proposed a mechanism where for compounds with more basic leaving groups, the amine group elimination XXXIV-XXXV i s the rate-determining step i n formation of the internal anhydride, XXXV (Scheme 10). Aldersley et a l . (183) proposed a similar mechanism to account for the rapid hydrolysis of the N-alkylmaleamic acids. However, the rate-determining step was suggested to be cleavage of the C-N bond, XXXIII-XXXV. The HPLC chromatograms, following i n -jection of N,N-diethylsuccinamic acid, indicated a bigger than expected solvent peak. This was probably due to co-elution of the degraded products, succinic a c i d , diethylamine and methanol. The amide group has also been reported to give underestimated log P values, calculated by the Hansch method, i n compounds with multiple functional groups (182). In comparative studies, both HPLC and shake-flask methods have advantages and l i m i t a t i o n s . The shake-flask method i s laborious, time-consuming, subject to errors from i n s o l u b i l i t y and i n s t a b i l i t y of compounds i n the octanol-O.lN HC1 p a r t i t i o n system and requires sensitive a n a l y t i c a l techniques to determine the concentration of solute i n the aqueous phase. Advantages associated with the shake-flask procedure include high s e n s i t i v i t y to hydrophobic effects 144 Scheme 10. Kinetic model proposed by some investigators (182, 183) for the hydrolysis of maleamic acids. 145 since aqueous solutions were used instead of mixed solvents. The shake-flask method can also be used for compounds of widely d i f f e r -ent chemical structures. The HPLC method, on the other hand, i s a r a p i d technique. I t o f f e r s high r e p r o d u c i b i l i t y of log P values since the only parameter, r e t e n t i o n time, can be determined a c c u r a t e l y . I t does not r e q u i r e q u a n t i t a t i v e a n a l y s i s of com-pounds. I t may be appropriate for compounds prone to degradation i n octanol-water systems or for compounds i n less pure samples. The RP-HPLC method, however, requires reference compounds with log P values determined by the standard shake-flask method. The range of l i p o p h i l i c i t y i s such that different chromatographic con-d i t i o n s have to be employed f o r compounds of h i g h or low l i p o p h i l i c i t y . In t h i s study the HPLC method under i s o c r a t i c conditions was l i m i t e d to compounds wi t h log P of about 0.8-3.2. One notable shortcoming of the HPLC method observed i n t h i s study i s that the order of r e t e n t i o n time between s t r u c t u r a l isomers, f o r example b u t y r i c a c i d and i s o b u t y r i c a c i d or v a l p r o i c a c i d and 2-ethy-hexanoic acid, can change depending on the composition and type of mobile phase. Some workers (141,147) have tackled t h i s problem by extrapolation of the plot of log k' versus mobile phase composition to obtain log k' f o r water. Apparently t h i s was not p o s s i b l e f o r the compounds studied since the mobile phase composition needed to elute a l l the test compounds while maintaining dynamic equilibrium conditions was l i m i t e d to a range of 40-60% acetonitrile-buffer and 60-70% methanol-buffer. Nevertheless, the accuracy of HPLC log P values was high, and values were comparable to shake-flask log P values. The HPLC method could a l s o be a p p l i e d to a v a r i e t y of 146 chemical structures unlike the a d d i t i v i t y methods of Hansch and Rekker. C. Electronic Structural Effects - Determination of Apparent Ioniz-ation Constants  1. Analytical method The apparent ionization constants (pKa) of test compounds were determined by potentiometric t i t r a t i o n i n aqueous-methanol media due to the limited s o l u b i l i t y of most of the compounds i n water. Potentiometric and UV spectrophotometric methods have been frequently used to determine the ionization constants of acidic compounds (185,187). The conductivity method has been used to determine the i o n i z a t i o n constants of a l i p h a t i c acids (188). Spectrophotometric methods were not appropriate for the compounds studied since there i s no change i n the UV spectra upon i o n i z a t i o n . In the potentiometric procedure, solutions of the compounds were progressively neutralized with quantities of standard KOH and the pH recorded. Tables 22-28 show t y p i c a l results obtained i n the potentiometric t i t r a t i o n procedure. The correction factor i n the Henderson-Hasselbalch equation, which i s reported (185) to be more si g n i f i c a n t at pH values lower than 5 and higher than 9, was uni-formly applied at a l l pH values. As shown i n Tables 23, 26, 28, values of pKa at the beginning of the t i t r a t i o n o c c a s i o n a l l y deviated from expected values. Table 29 shows the pKa values of 23 compounds determined i n either 10% methanol-water or 50% methanol-water. The ionization constants were determined at 23°C + 1.0°C. Values of pKa were determined to within a precision of _+ 0.05 pKa 147 Table 22 Determination of the ionization constant of a monobasic a c i d , valproic acid i n 10% MeOH Temperature - 24°C Titrant 0.0105M KOH (mL) pH [HA] x 103 (mol. L"1) [A~] x 103 (mol. L- 1) THAI - fH+] [A"] + {H-M pKa 0.5 4.24 0.9218 0.105 5.325 4.97 1.0 4.47 0.8178 0.210 2.930 4.94 1.5 A.6A 0.7128 0.315 2.042 4.95 2.0 4.82 0.6078 0.420 1.362 4.95 2.5 4.99 0.5028 0.525 0.9204 4.95 3.0 5.18 0.3978 0.630 0.6147 4.97 148 Table 23 Determination of the ionization constant of 5-Isoamyltetrazole i n 10% MeOH Temperature - 23°C Titrant 0.0105M KOH (mL) pH [HA] x 103 (mol. L- 1) [A-] x 103 (mol. L- 1) THA] - {H+} [A"] + {IF} pKa 0.5 4.36 0.888 0.105 5.7043 5.12 1.0 4.69 0.783 0.210 3.3099 5.21 1.5 4.91 0.678 0.315 2.0339 5.22 2.0 5.12 0.543 0.420 1.3223 5.24 2.5 5.28 0.468 0.525 0.8729 5.22 3.0 5.46 0.363 0.630 0.5672 5.21 3.5 5.65 0.258 0.735 0.3470 5.19 149 Table 24 Determination of the ionization constant of trimethylacetic acid i n 10% MeOH Temperature - 23°C Titrant 0.01997M KOH (mL) pH [HA] x 103 (mol. L- 1) [A~] x 103 (raol. L_ 1) [HA] - {H+} [A"] + {H"1"} pKa 0.5 4.27 1.8983 0.1997 7.2794 5.10 1.0 4.52 1.6986 0.3994 3.8836 5.10 1.5 4.77 1.4989 0.5991 2.4053 5.15 2.0 4.91 1.2992 0.7988 1.5866 5.11 2.5 5.08 1.0995 0.9985 1.0838 5.11 3.0 5.22 0.8998 1.1982 0.7422 5.09 3.5 5.38 0.7001 1.3979 0.4963 5.08 150 Table 25 Determination of the ionization constant of dibutylacetic acid i n 50% MeOH Temperature - 23°C Titrant 0.0105M KOH (mL) pH [HA] x 103 (mol. L- 1) [A~] x 103 (mol. L- 1) [HA] - {H+} [A"] + {H+} pKa 0.5 5.11 0.8776 0.105 7.7110 6.00 1.0 5.38 0.7726 0.210 3.5873 5.93 1.5 5.58 0.6676 0.315 2.0938 5.90 2.0 5.74 0.5626 0.420 1.3295 5.86 2.5 5.98 0.4576 0.525 0.8681 5.92 3.0 6.29 0.3526 0.630 0.5584 6.04 3.5 6.53 0.2476 0.735 0.3363 6.06 151 Table 26 Determination of the ionization constant of N,N-diethylsuccinamic acid i n 50% MeOH Temperature - 2A°C Titrant 0.0098M KOH (mL) pH [HA] x 103 (mol. L_ 1) [A-] x 103 (mol. L- 1) [HA] - {H+} [A"] + {H+} pKa 0.5 6.29 0.896 0.098 9.030A 7.2A 1.0 6.39 0.792 0.196 A.0305 7.00 1.5 6.50 0.69A 0.29A 2.3571 6.87 2.0 6.61 0.596 0.392 1.5191 6.79 2.5 6.76 0.A98 0.A90 0.0155 6.77 3.0 6.91 0.A00 0.588 0.6800 6.7A 3.5 7.12 0.302 0.686 O.AAOA* 6.76 4.0 7.A3 0.20A 0.78A 0.2607* 6.85 [HA] + {0H~} •calculated using the correction factor [A~] - {OH-} 152 Table 27 Determination of the ionization constant of 5-cyclohexylmethyltetrazole i n 50% MeOH Temperature - 23°C Titrant 0.0105M KOH (mL) pH [HA] x 103 (mol. L"1) [A~] x 103 (mol. L"1) THAI - {H+} [A~] + {H"n pKa 0.5 4.75 0.9552 0.105 7.658 5.63 1.0 5.05 0.8502 0.210 3.843 5.63 1.5 5.27 0.7452 0.315 2.309 5.63 2.0 5.45 0.6402 0.420 1.503 5.62 2.5 5.62 0.5352 0.525 1.010 5.62 3.0 5.79 0.4302 0.630 0.6786 5.61 3.5 5.97 0.3252 0.735 0.4403 5.57 153 Table 28 Determination of the ionization constant of N,N-dibutylsuccinamic acid i n 50% MeOH Temperature - 24°C Titrant 0.0098M KOH (mL) pH [HA] x 103 (mol. L- 1) [A~] x 103 (mol. L_ 1) [HA] - {H+} [A"] + {H+} pKa 0.5 6.21 0.9060 0.0980 9.1825 7.17 1.0 6.33 0.8080 0.1960 4.1094 6.94 1.5 6.42 0.7100 0.2940 2.4103 6.80 2.0 6.56 0.6120 0.3920 1.5593 6.75 2.5 6.69 0.5140 0.4900 1.0481 6.71 3.0 6.82 0.4160 0.5880 0.7069 6.67 3.5 7.01 0.3180 0.6860 0.4635 6.68 4.0 7.23 0.2200 0.7840 0.2809* 6.68 4.5 7.58 0.1220 0.8820 0.1388* 6.72 5.0 8.29 0.0240 0.9800 0.0265* 6.71 [HA] + {OH-} *calculated using the correction factor [A-] - {OH-} 154 Table 29 pKa Values of valproic aid and analogues Compound pK;3 t +_ s.d. (n=6) ( i n 10% MeOH) (i n 50% MeOH) 1. Isobutyric acid 4.96 0.02 5.85 + 0.06 2. Trimethylacetic acid 5.11 + 0.02 5.94 + 0.06 3. 3-Methylpentanoic acid 4.85 + 0.03 5.76 + 0.05 4. Diethylacetic acid 4.83 + 0.03 6.02 ± 0.02 5. 3-Methylhexanoic acid 5.02 + 0.01 5.72 + 0.05 6. 2,2-Dimethylbutyric acid 5.26 + 0.03 6.52 + 0.02 7. 2,2-Dimethylpentanoic acid 5.26 + 0.05 6.36 ± 0.01 8. Valproic acid (VPA) 4.95 + 0.01 6.03 + 0.01 9. 3-OH VPA 4.39 + 0.02 5.56 + 0.04 10. 4-keto VPA 4.58 + 0.01 5.28 + 0.06 11. 4-Ene VPA 4.63 + 0.01 5.51 + 0.06 12. 2-Ethylhexanoic acid 5.05 + 0.03 6.14 + 0.13 13. Dibutylacetic acid 5.05 + 0.06 5.95 + 0.08 14. 1-Methylcyclohexane-l-carboxylic acid 5.17 + 0.05 6.58 + 0.01 15. 3-Ethylpentanoic acid 4.78 + 0.08 5.99 + 0.02 16. Cyclohexylacetic acid 4.78 + 0.02 5.77 + 0.04 17. N,N-Diethylsuccinamic acid 5.58 + 0.05 6.80 + 0.05 18. N,N-Dibutylsuccinamic acid 5.58 + 0.05 6.72 + 0.03 19. 5-Isoamyltetrazole 5.22 + 0.02 5.60 + 0.01 20. 5-Cyclohexylmethyltetrazole 5.37 + 0.03 5.62 + 0.01 21. 5-Heptyltetrazole 5.31 + 0.05 5.60 + 0.01 22. 2,3'-Diene VPA 4.02 + 0.02 5.47 + 0.06 23. 2-Ene VPA 4.36 + 0.05 5.51 + 0.06 155 u n i t s . Standard deviations of the pKa values were i n a few cases higher than +0.06 pKa. The pKa of valproic a c i d , 4.95, determined i n 10%-methanol-water i s higher than the reported values of 4.82 i n 5% methanol-water (189) and 4.6 as the aqueous value from extrapolated acetone-water mixtures (67). The pKa values i n Table 29 are higher i n 50% methanol-water than i n 10% methanol-water due to the lower d i e l e c t r i c constant of 50% methanol-water. However, the order of ionization strength i n 50% methanol-water did not r e f l e c t the same order i n 10% methanol-water. Similar findings on the non-linearity between pKa and the composition of organic solvent have been reported f o r some compounds (185-187). There have als o been reports that the conventional approach of extrapolating values of pKa to aqueous values may be useful when lower solvent compositions are used (185,186). The pKa values i n 10% methanol-water were considered to be more r e l i a b l e than values i n 50% methanol-water. Compared to pKa values of a l i p h a t i c acids reported by Dippy (188) i n conductivity methods (pKa of trimethylacetic acid = 5.05, pKa of diethylacetic acid = 4.75, pKa of isobutyric acid = 4.86), the pKa values i n 10% methanol-water are 0.1-0.2 pKa u n i t s above the aqueous values. 2. Effects of structural constitution on ionization constants To the extent that the pKa values i n 10% methanol are most l i k e l y to be d i f f e r e n t from the aqueous values by a constant amount, the electronic effects can be determined by the pKa values. This i s supported by the observation that the pKa values i n Table 29 are consistent with the inductive effects of substituent groups 156 expressed i n Table 30 by the polar s u b s t i t u e n t constants. Electron-withdrawing substituents have a positive polar substituent constant while electron-repelling groups have negative values. The ionization constants of the tetrazoles (Table 29) conform to the generalization (21) that they are about a tenth to one-half as large as their corresponding carboxylic acid values. The pKa values of the tetrazoles are similar to results obtained by Mihina and Herbst (161) i n 25% by weight methanol-water. The influence of substitution pattern, polar groups, a l i c y c l i c groups, unsaturation and intramolecular bonding are reflected i n the pKa values. Beta-substitution i n the a l k y l chain, exemplified by 3-ethylpentanoic a c i d and c y c l o h e x y l a c e t i c a c i d , leads to r e l a t i v e l y low pKa values compared to alpha-substituted compounds. Diethylacetic and valproic a c i d , both alpha-branched acids, have comparable or s l i g h t l y higher i o n i z a t i o n strength than isobutyric a c i d , i n agreement with results of Dippy (188). Compounds such as a,a-dimethylvaleric a c i d , trimethylacetic acid and 1-methylcyclo-hexane carboxylic acid have r e l a t i v e l y high pKa values due to combined inductive effects of the methyl groups. The decrease i n pKa of the unsaturated compounds compared to t h e i r saturated analogues i s known to be due to polarization of the double bond. Values of the polar substituent constants indicate that cyclohexyl-methyl and c y c l o h e x y l s u b s t i t u t i o n diminishes the i o n i z a t i o n strength of the unsubstituted carboxylic acids and the tetrazoles (Table 30). The presence of the keto group or hydroxy group i n the a l k y l chain increases the ionization strength i n accordance with the 157 Table 30 Polar effect of substitution i n an a l i p h a t i c s e r i e s , R-Y where Y i s the functional group. R a * (Polar Substituent Constant)3 C13C 2.65 CH3CO 1.65 CH3C0CH2 0.60 H0CH2 0.55 H 0.49 CH3-CH=CH 0.36 CH3—CH=CH—CH2 0.13 CH3 0 C2H5 -0.1 i - C4H9 -0.125 n - C4H9 -0.13 n - C3Hy -0.115 i - C3H7 -0.19 Cyclo - C6HnCH2 -0.06 Cyclo - C6Hn -0.15 s - C4H9 -0.21 ((^2^5) 2 CH -0.225 t - C4H9 -0.30 ataken from R. W. Taft (134). 158 polar substituent constants. The presence of an amido group close to a carboxyl group usually leads to a decrease i n pKa r e l a t i v e to an a l k y l substituent because of the inductive e f f e c t . The opposite results are obtained for the N,N-dialkylsuccinamic acids. The r e l a t i v e l y high pKa values could be explained by hydrogen-bonding between the carboxylic proton and the amido group. Intramolecular bonding i s even stronger i n N-propylmaleamide, pKa of 10.53 (184) where the c i s configuration favours such an i n t e r a c t i o n . The pKa values for most of the compounds studied are available i n the l i t e r a t u r e but the values are usually determined under different conditions and by different workers. This consideration prompted the determination of the pKa values of test compounds by the potentiometric t i t r a t i o n method using the same solvent-mixture for a l l the compounds. Pharmacological Studies Evaluation of anticonvulsant a c t i v i t y The anticonvulsant a c t i v i t y of each t e s t compound was determined using the i n vivo s.c. PTZ seizure threshold t e s t . The test appears to be suitable for valproic acid and analogues. Val-proic acid and alkyl-substituted a n t i e p i l e p t i c drugs are known to be more effective i n the s.c. PTZ threshold test than i n maximal electroshock seizure t e s t s . In addition, valproic acid and PTZ appear to exert opposing a c t i o n s at the GABA post-synaptic s i t e (53,54,81). Recent studies (30,31,33,34,57) have suggested that valproic acid and PTZ may act at the picrotoxin s i t e of the 159 GABA-Benzodiazepine-receptor-chloride ionophore complex. The spec-i f i t y of the s.c. PTZ t e s t has been r e c e n t l y r e - i n v e s t i g a t e d and suggested to be r e l a t e d to the dosage l e v e l of PTZ used (190). C l o n i c threshold s e i z u r e s were found to be induced reproducibly with 85 mg/kg s.c. PTZ (CDg7) i n mice (190). The c o n t r o l t e s t i n t h i s study showed that 100% of the mice responded to the s.c. i n j e c t i o n of 85 mg/kg PTZ with well-defined clonic spasms. The time of onset of clonic seizures was between 3-9 minutes which i s consistent with previous studies on the onset of clonus at d i f f e r e n t PTZ doses (191). PTZ was i n j e c t e d s.c. 15 minutes a f t e r i.p. i n j e c t i o n of each t e s t compound. The sel e c t e d time i n t e r v a l was based on l i t e r a t u r e reports where 2 min (14), 15 min (5,12,13) and 30 min (10,11,88) have been chosen as s u i t a b l e elapsed times before s.c. i n j e c t i o n of PTZ. The i.p. dose of t e s t compounds ranged from 0.2 mraol/kg to 2.0 mmol/kg. The observation time of 30 min a f t e r s.c. PTZ i n j e c t i o n was chosen since c l o n i c seizures rarely occurred after times longer than 20 min with i.p. doses of compounds that prevent the i n i t i a l series of clonic act-i v i t y . The dose-response data f o r t e s t compounds are presented i n Table 31. Nine compounds prevented the clonic seizures induced by the threshold dose of PTZ i n a dose-dependent fashion. Isobutyric a c i d , 4-keto VPA, c y c l o h e x y l a c e t i c a c i d , 5-cyclohexylmethyl-t e t r a z o l e and 5 - i s o a m y l t e t r a z o l e were inactive i n the dose range studied. N,N-Diethylsuccinamic acid was inactive at doses between 0.2-2.0 mmol/kg but induced hyperactivity ( r a p i d - c i r c l i n g a c t i v i t y ) when injected alone i n mice at doses of 3.0 mmol/kg. On the other hand, N,N-dibutylsuccinamic acid exhibited convulsant a c t i v i t y at 160 Table 31 Protection against PTZ-induced seizures3 i n mice by valproic acid and i t s analogues Compounds % Protected against clonic seizures Dose, mmol/kg 0.2 0.3 0.5 1.0 1.5 2.0 1. Valproic acid (VPA) - - 25 87.5 87.5 100 2. 2-Butylhexanoic acid - 12.5 37.5 87.5 100 -3. 1-Methylcyclohexane-l-carboxylic acid - - 12.5 37.5 75 87.5 4. 2,2-Dimethylbutyric acid - - 0 50 37.5 62.5 5. 3-Ethylpentanoic acid - - - 12.5 25 62.5 6. Trimethylacetic acid - - 0 25 50 37.5 7. 5-Cyclohexylmethyltetrazole - - - 12.5 12.5 0 8. Cyclohexylacetic acid - - - 12.5 12.5 0 9. Isobutyric acid - - - 0 - 12.5 10. 5-Isoamyltetrazole - - - 12.5 - 0 11. 2,3'-Diene VPA - - - 50 37.5 62.5 12. 2-Ene VPA - - - 37.5 50 62.5 13. 4-keto VPA - - - 12.5 - 0 14. 5-Heptyltetrazole 25 37.5 87.5 100 - -15. N,N-Diethylsuccinamic acid - - - 12.5 - 0 16. N,N-Dibutylsuccinamic acid - - Convi llsant - -Anticonvulsant a c t i v i t y of compounds was evaluated 15 min after i . p . administration. Mice were observed for clonic spasms (>5 sec duration) within 30 min after 85 mg/kg s.c. PTZ administration. Adult male Swiss mice (CD1 s t r a i n , 20-32g) were used. Eight mice were used per dose of test compounds. 161 sub-lethal doses of 1.0 mmol/kg or l e s s . The dose-response curves of active compounds are depicted i n Figure 1 6 . The dose-response curves of the most potent drugs, valproic acid, 2-butylhexanoic a c i d , 5-heptyltetrazole and 1-methylcyclohexane-l-carboxylic acid are r e l a t i v e l y steep compared with the moderately active drugs. C a l c u l a t i o n of ED^g values was based on the method of L i t c h f i e l d and Wilcoxon ( 1 9 2 ) . Goodness-of-fit of the estimated curve to the observed dose-response data was tested using the Chi-Square t e s t . Tables 32 and 33 show the ED^Q values, slopes of dose-response curves and chemical structures of active compounds. The slope values i n Tables 32 and 33 were determined by the method of L i t c h f i e l d and Wilcoxon ( 1 9 2 ) and are the antilogarithm of the inverse of the slope constant i n the dose-response curves. The dose-response curves did not deviate s i g n i f i c a n t l y from p a r a l l e l -ism. The estimated ED^Q of valproic a c i d , 0.70 mmol/kg i s i n the range of values of 0.57 mmol/kg ( 1 4 ) and 0.90 mmol/kg ( 4 0 ) reported i n previous studies using the s.c. PTZ t e s t . The anticonvulsant a c t i v i t y of f i v e a l i p h a t i c acids studied have previously been examined i n a single-dose study ( 1 0 ) where the endpoint used was the protection of mortality. The results of the mortality study are summarized i n Table 34 along with the results of t h i s study at 1 mmol/kg i . p . dose. As shown i n Table 3 4 , the mortality endpoint does not reveal the quantitative differences i n a c t i v i t y of the f i v e a l i p h a t i c acids compared to the discriminative clonic seizure t e s t . Moreover, dibutylacetic acid was reported to be inactive at 1.39 mmol/kg, which i s contrary to the results of the present 162 TrtnaltiylacaMc acid 10 DoMftranoto/kg) 1-M«ttiylcyelohexan«cartooxy»c acid 10 Do*a{mmola/kg) Valproic acid 10 Poaaftwiufci/fcg) 150 I tt 2 96 BO SO | 30 I » 2 8.2-Oim«ttiytX«yrie »cld 10 Do—(mmota/ko) 3-Ettiyk»ntanole acid 10 Doaa<imio4a/kg) <80 cl ho * 8-Butytwxanofc: add 10 DoaaOimoai/ka) 6-Heptyltatraiole 03 10 Doaeinmola/kQ) i 160 | 130 * Figure 16. Dose-response curves of valproic acid and analogues using the subcutaneous pentylenetetrazole seizure threshold test in mice. 163 Table 32 Anticonvulsant potency of valproic acid and i t s analogues against the clonic phase of PTZ-induced seizures i n mice Compounds3 EDCQ, mmol/kg, i . p . (95X confidence l i m i t s )b Slope (95% confidence l i m i t s )b 1. 5-Heptyltetrazole 0.31 (0.23-0.42) 1.56 (1.22-2.00) 2. 2-Butylhexanoic acid 0.57 (0.33-0.97) 1.72 (1.02-2.91) 3. Valproic acid (VPA) 0.70 (0.50-0.98) 1.63 (1.14-2.33) 4. 1-Methylcyclohexane-l-carboxylic acid 1.08 (0.71-1.64) 1.83 (1.06-3.16) 5. 2,2-Dimethylbutyric acid 1.43 (0.56-2.58) 2.84 (0.72-11.2) 6. 2,3'-Diene VPA 1.45 (0.82-2.57) 2.76 (0.29-25.9) 7. 2-Ene VPA 1.46 (0.78-2.74) 3.06 (0.24-38.8) 8. 3-Ethylpentanoic acid 1.91 (1.43-2.56) 1.69 (0.81-3.52) 9. Trimethylacetic acid 2.02 (1.11-3.68) 2.88 (0.77-10.8) 10. Isobutyric acid inactive -11. Cyclohexylacetic acid inactive -12.5-Cyclohexylmethyltetrazole inactive -13. 5-Isoamyltetrazole inactive -14. 4-keto VPA inactive -15. N,N-Diethylsuccinamic acid i n a c t i v e0 -16. N,N-Dibutylsuccinamic acid convulsant -aDrugs administered 15 min before s.c. 85 mg/kg PTZ i n j e c t i o n . Dose range of 0.2-2.0 mmol/kg. bData analyzed by the method of L i t c h f i e l d and Wilcoxon (192). cR a p i d - c i r c l i n g a c t i v i t y of test drug at 3.0 mmol/kg. 164 Table 33 Anticonvulsant potency of valproic acid and i t s analogues against the clonic phase of PTZ-induced seizures i n mice. Dose range of acids, 0.2-2.0 mmol/kg Compounds ED^Q, mmol/kg, i . p . (95% confidence l i m i t s ) Slope (95% confidence l i m i t s ) H CH3CH2CH2CH2CH2CH2CH2C*N_N 0.31 (0.23-0.42) 1.56 (1.22-2.00) c $ p ™ ~  C 0 0 H 0.57 (0.33-0.97) 1.72 (1.02-2.91) §H^CH - COOH 0.70 (0.50-0.98) 1.63 (1.14-2.33) / VC H 3 X /^COOH 1.08 (0.71-1.64) 1.83 (1.06-3.16) CHo 1 CH3 CH2 - C - COOH 1.43 (0.56-2.58) 2.84 (0.72-11.2) CH3 CHo - CH = CH ^ CH3 - CH2 - CH**0 " 0 0 0 1 1 1.45 (0.82-2.57) 2.76 (0.29-25.9) CHo - CH2 - CHo. o^t t CH3 - CH2 - CH^C - COOH 1.46 (0.78-2.74) 3.06 (0.24-38.8) °2 H 5)CH - CH2 - COOH 1.91 (1.43-2.56) 1.69 (0.81-3.52) CH3 CH3 - C - COOH 2.02 (1.11-3.68) 2.88 (0.77-10.8) CH3 CH^CH - COOH inactive 165 Table 33 (Cont'd) Compounds ED5Q. mmol/kg, i . p . (95% confidence l i m i t s ) Slope (95% confidence l i m i t s ) <^  CH2_COOH inactive -H / \ /N _J} \ /> _ C H2 - C^N-N inactive -C H3K H ^CH - CHo - CHo - C —N-N C H3 N—-N inactive CH3 - ^ - CHo ^CH - COOH CH3 - CH2 - CH^ inactive -^ N-C - CH2CH2 COOH C2H5^ 0 inactive -CAH9 ^ N-C - CH2_CH2-C00H C4H9^ 0 convulsant -166 Table 34 Anticonvulsant a c t i v i t y of VPA and i t s analogues on the threshold for PTZ-induced seizures determined by protection against clonic seizures and by percent mortality i n mice % Protected Against (n = 8/dose) Compounds Clonic Seizures3 M o r t a l i t yb M o r t a l i t y0 1. Valproic acid (VPA) 87.5 100 100 2. 2-Butylhexanoic acid 87.5 100 0 3. 2,2-Dimethylbutyric acid 50 100 80 4. Trimethylacetic acid 25 100 80 5. Isobutyric acid 0 75 20 6. 3-Ethylpentanoic acid 12.5 100 7. 1-Methylcyclohexane-l-carboxylic acid 37.5 100 8. Cyclohexylacetic acid 12.5 100 9. 2-Ene VPA 37.5 62.5 10. 2,3'-Diene VPA 50 100 11. 4-keto VPA 12.5 25 12. 5-Isoamyltetrazole 12.5 62.5 13. 5-Cyclohexylmethyltetrazole 12.5 75 14. 5-Heptyltetrazole 100 100 15. N,N-Diethylsuccinamic acid 12.5 62.5 16. N,N-Dibutylsuccinamic acid 0 0 3A c t i v i t y at 1.0 mmol/g i . p . of test compounds administered 15 min before 85 mg/kg s.c. PTZ i n j e c t i o n . ^Conditions same as i n (a) above with mortality determined within 30 min after PTZ administration. cTaken from G. Carraz (4) where 1.39 mmol/kg i . p . administered 30 min before 80 mg/kg s.c. PTZ administration. 167 study. Table 34 also shows that compounds inactive against clonic seizures would have been described as active by the mortality test. 2. Toxicity of compounds The acute t o x i c e f f e c t s of the t e s t compounds were noted during the 15 min i n t e r v a l before i n j e c t i o n of PTZ (Table 3 5 ) . Sedation was a common side effect of the anticonvulsant compounds. Valproic acid showed remarkable sedation at 2 mmol/kg doses. The TD^Q of valproic acid i s reported to be 2.56 mmol/kg i n mice (40). The unsaturated analogues of valproic acid and dibutylacetic acid exhibited marked sedation even at 1 mmol/kg doses while the most potent anticonvulsant 5 - h e p t y l t e t r a z o l e showed sedation at much lower doses. Dibutylacetic acid appeared to have the lowest pro-t e c t i v e index w i t h i t s l e t h a l e f f e c t s o c c u r r i n g at 2.0 mmol/kg. Compared to the carboxylic acids, the tetrazoles possessed greater toxic properties with ataxia occurring at low doses. N,N-Dibutyl-succinamic a c i d had convulsant p r o p e r t i e s at s u b l e t h a l doses of 0.5-1.0 mmol/kg. Among the compounds s t u d i e d , v a l p r o i c a c i d appeared to have the most desirable pharmacological p r o p e r t i e s since i t possessed high a n t i c o n v u l s a n t a c t i v i t y w i t h marked sedation appearing at doses greater than 2.0 mmol/kg. I t i s apparently not yet c l e a r whether anticonvulsant and sedative effects occur at the same s i t e i n the CNS (14,30). I t has been suggested that the d i s t r i b u t i o n a l l o c a l i z a t i o n of anticonvulsant drugs may be the predominating f a c t -or to determine the sedative side effects (20,29). The unsaturated analogues of valproic acid showed more sedative effects and less anticonvulsant a c t i v i t y than valproic acid (Table 35). 168 Table 35 Observed toxic effects of test compounds i n mice Compounds Observed Toxic Effects A. Isobutyric acid 4-keto VPA Trimethylacetic acid 2,2-Diraethylbutyric acid 3-Ethylpentanoic acid Cyclohexylacetic acid Minimal sedative effects at doses not greater than 2.0 mmol/kg. B. Valproic acid (VPA) 1-Methylcyclohexane-l-carboxylic acid Sedative effects at 2.0 mmol/kg dose. C. 2,3'-Diene VPA 2-Ene VPA Sedative effects at doses greater than 1.0 mmol/kg. D. 2-Butylhexanoic acid Sedation and ataxia at doses greater than 1.0 mmol/kg. Lethal dose at 2.0 mmol/kg. E. 5-Cyclohexylmethyltetrazole 5-Isoamyltetrazole Ataxia at doses greater than 1.0 mmol/kg. Prostration and ab-normal spread of hind limbs at 2.0 mmol/kg dose. F. 5-Heptyltetrazole Sedative effects and ataxia at doses greater than 0.2 mmol/kg. Abnormal body posture. Occasion-a l f a s c i c u l a t i o n at 1.0 mmol/kg dose. G. N,N-Diethylsuccinamic acid Abnormal spread of hind limbs at doses greater than 0.5 mmol/kg. Hyperactivity ( r a p i d - c i r c l i n g a c t i v i t y ) at doses of 3.0 mmol/kg H. N,N-Dibutylsuccinamic acid Tremors, jumping or hopping move-ments and convulsions at doses tested (0.5-1.0 mmol/kg). Tonic-clonic spasms. Lethal dose at 2.0 mmol/kg. 169 E. Structure-Activity Relationships 1. Quantitative Structure-Activity Relationships The investigation of the molecular s p e c i f i t y of the a n t i -convulsant action of valproic acid analogues was one of the major objectives of t h i s study. Towards t h i s end, the Hansch li n e a r free-energy model was applied to study the SAR among the structurally-diverse valproate analogues. Since the c l a s s i c a l work of Meyer and Overton, i t has become evident that l i p o p h i l i c i t y plays a predominant role i n determining the i n vivo a c t i v i t y of many CNS-active drugs. According to the Hansch model, the prob-a b i l i t y that CNS-active drugs reach t h e i r s i t e of a c t i o n i s determined by l i p o p h i l i c i t y (log P ) . In addition, pharmacokinetic factors such as protein-binding, d i s t r i b u t i o n , metabolism and ex-cretion which play a major role i n determining the efficacy of these drugs are at least governed by l i p o p h i l i c i t y (193,194). On the other hand, s t e r i c and electronic factors appear to be c r i t i c a l for the pharmacodynamic action of drugs that arise by interaction with s p e c i f i c receptors. In t h i s study, the l i p o p h i l i c i t y of compounds was described by the octanol-water p a r t i t i o n c o e f f i c i e n t (log P) values, obtained from HPLC studies using 70% methanol-phosphate buffer as the mobile phase. The electronic effects were represented by the pKa values of the compounds as determined i n 10% methanol-water. The pKa values express the r e l a t i v e i o n i z a t i o n strength of the compounds. The b i o l o g i c a l a c t i v i t y evaluated was the antagonism of PTZ-induced clonic seizures i n mice. Table 36 shows the structure, a n t i -170 Table 36 Biol o g i c a l data and physicochemical properties of compounds tested Compounds ED5?/w mmol/kg i. p . E S(R) * log 1 ED5FJ log P (HPLC) pKa (10% MeOH) E S ( R ) ** H CH3CH2CH2CH2CH2CH2CH2C^v.j^_j^ 0.31 -0.35 0.509 2.01 5.31 -1.54 C ^ > C H - COOH 0.57 -2.23 0.243 3.34 5.05 -3.47 r3l7^cn - COOH 0.70 -2.11 0.155 2.57 4.95 -3.35 / - VC H3 V X^-COOH 1.08 -2.03 -0.0334 2.38 5.17 -3.27 C H3 CHo-CH9 - C - COOH 3 ! CH3 1.43 -1.60 -0.156 1.90 5.26 -3.41 CH3 - CH2 -CCH*C _ C 0 0 H 1.45 -0.161 2.27 4.02 -cl33: S*: Ssc - < w » 1.46 -0.164 2.44 4.36 -CoHc ^CH - CH2 - COOH C2H5/ 1.91 -2.00 -0.281 2.19 4.78 -2.17 CHo 1 3 CHo - C - COOH 3 1 CH3 2.02 -1.54 -0.305 1.54 5.11 -2.78 CH ^3^CH - COOH inactive -0.47 1.10 4.96 -1.71 171 Table 36 (Cont'd) Compounds ED59/w mmol/kg i. p . * log 1 ED5 Q log P (HPLC) pKa (10% MeOH) ** ^ yCH2-COOH inactive -0.98 2.31 4.78 -3.22 H ( /"C H2 _ C-N-N inactive -0.98 1.59 5.37 -3.22 •^CH - CHo - CHo - C ^N~N CHf 2 2 \\ II N — N inactive -0.35 1.29 5.22 -1.59 9 CHo — C — CHo ^CH - COOH CH^ — CH2 — CH2 inactive 1.31 4.58 2 5^ N-C - CH0CH0-COOH C H ' " inactive 2.27 5.58 C4H9 ^ N-C - CH0-CH0-COOH C4H9 0 convul-sant 2.30 5.58 * Taken from R. W. Taft (134). ** Taken from C. Hansch and A. Leo (152). 172 convulsant potency and physicochemical properties of valproic acid and the other analogues tested. The log P values range from 1.0 to 3.A and pKa values from A.O to 5.6. The s t e r i c bulk of the a l k y l substituents was described by the Taft s t e r i c constant, Eg* or Hansch s t e r i c constant, E**. In accordance with the linea r free-energy relationship model, log I/ED5Q was correlated with log P, pKa and E' using multiple re-gression analysis (Michigan Interactive Data Analysis System, MIDAS). In Table 37 are l i s t e d the regression equations along with the multiple correlation c o e f f i c i e n t , r (from residual and predict-ed values), the standard error of the regression, (S) and the attained significance l e v e l , p (from analysis of variance for the regression). From the correlation matrix i n Table 35, i t i s ob-served that a l l interdependence among the variables of the nine active compounds are i n s i g n i f i c a n t except between log P and Eg (r = -0.5A). This i s expected from the increase i n log P with molecular size which i n turn correlates with Eg (195). For the nine active compounds, i t can be seen from Table 37 that the regression equations 1-A are not s i g n i f i c a n t . The a n t i -PTZ potency could not be described by either log P (equation 1) or pKa (equation 2) alone. The correlation improved i n equation 3 where both log P and pKa are i n c l u d e d . However, equation 3 accounts for only A0% (r ) of the variations i n anticonvulsant potency. Examination of the graphic plot of log l/ED^g versus log P or pKa revealed that 5-heptyltetrazole was actually more active than anticipated. When 5-heptyltetrazole was deleted from the QSAR, the potency was better correlated with log P (Table 38, equation 5 ) . 173 Table 37 Equations obtained correlating the anti-PTZ effect of valproic acid and analogues with their physicochemical parameters Equation na rb Sc F Pd 1. -log EDCQ = 0.223 log P - 0.534 (+0.186)e (+0.435) 9 0.413 0.264 1.442 0.269 2. -log ED50 = 0.254 pKa - 1.263 (+0.215) (+1.05) 9 0.408 0.265 1.398 0.276 3. -log EDrn = 0.263 log P + (+0.173) 0.300 pKa - 2.09 (+0.199) (+1.11) 9 0.631 0.243 1.986 0.218 4. -log ED5 Q = -0.265 Eg - 0.517 7 -0.291 anuraber of compounds studied, ^correlation c o e f f i c i e n t , cstandard error of estimate, l e v e l of significance of F-value, enumber i n parentheses gives standard error of c o e f f i c i e n t . Correlation matrix for the physicochemical parameters with nine observations pKa Es 1 0.298 1 log P pKa E_ log P 1 -0.153 -0.540 174 Table 38 Equations obtained correlating the anti-PTZ effects of VPA and analogues (excluding 5-heptyltetrazole) with their physicochemical parameters Equation n r S F p 5. -log ED50 = 0.322 log P - 0.837 (+0.079) (+0.188) 8 0.856 0.110 16.48 0.0067 6. -log ED50 = 0.102 pKa - 0.582 8 0.225 0.207 0.321 0.59 (+0.180) (+0.876) 7. -log ED50 = 0.331 log P + 0.136 pKa 8 0.907 0.098 11.56 0.013 . (+0.171) (+0.086) -1.516 (+0.461) 8. -log EDsn = 0.065 log P (+0.622) +0.052 (log P )2 - 0.536 (+0.126) (+0.751) 9. -log EDcn = 0.478 log P J U (+0.647) -0.0297 (log ?) z + 0.147 pKa (+0.130) (+0.107) -1.743 (+1.12) 0.861 0.908 0.118 0.109 7.190 6.263 For meaning of n, r , s, F, P: refer to Table 35. 0.034 0.054 log P (log P): pKa Correlation Matrix log P (log P )2 1 0.991 1 -0.0841 -0.0206 pKa 175 The addition of a pKa term to equation 5, gave a s i g n i f i c a n t and better correlation (equation 7 ) . The positive c o e f f i c i e n t for log P and pKa i n equation 7 indicated that an increase i n log P and pKa enhanced the anticonvulsant a c t i v i t y . The same conclusion was apparent i n equation 3 for Table 36. The relationship expressed by equation 7 agrees with e a r l i e r studies (196) which showed that l i p i d s o l u b i l i t y and ionization constants of acidic and basic drugs are of greater importance i n drug entry into the CNS. Since a n t i -convulsant a c t i v i t y of the carboxylic acids increases with high log P values, r e a l i z a t i o n of an optimum log P value could be obtained from the parabolic relationship between i n vivo b i o l o g i c a l a c t i v i t y and log P suggested by Hansch and co-workers (19,199). However, as shown i n equation 9 (Table 38), the parabolic relationship i n log P was not s i g n i f i c a n t , probably because the log P values of the active compounds are less than the optimum log P value but f a l l into the linear portion of the curve. The f a i l u r e of log P and pKa to e x p l a i n the v a r i a t i o n i n antagonism of PTZ by the tested compounds including 5-heptyl-tetrazole could be due to the absence of a physicochemical para-meter to characterize the i n t r i n s i c a c t i v i t y of compounds a r i s i n g from interaction with the recognition s i t e . The Hansch model has been extended to replace the Hammet constant, a, with other elec-tronic parameters such as pKa, dipole moment and net atomic charges (205). However, the electronic effects expressed by the pKa values have been noted to have a dual function (197). The pKa affects the proportion of unionized drug which penetrates the blood brain b a r r i e r . I t also describes the electronic effects of the a l k y l 176 substituent on the charge density of the carboxylic group or the other polar moiety. Dipole moments have been shown to be useful i n characterizing the polar moiety which may interact e l e c t r o s t a t i c a l -l y at the recognition s i t e (26-28,182). Due to the p o s s i b i l i t y that alkyl-substituted anticonvulsant compounds act v i a interaction at the picrotoxin s i t e of the GABA receptor complex, the physicochemical properties that determine anti-PTZ a c t i v i t y were investigated. The anti-PTZ potency follow-ing standard procedures and physicochemical properties of the other a l k y l - s u b s t i t u t e d compounds, i n c l u d i n g the molecular d i p o l e moments, were obtained from l i t e r a t u r e sources and are summarized i n Table 39. The results of the QSAR are presented i n Table 40. I t can be seen that the regression equations 1-7 are i n s i g n i f i c a n t . Equations 1-3 show that log P, pKa or dipole moment (M) alone, do not account for the variations i n anticonvulsant potency. Addition of a term i n (log P) to equation 1 gives a sl i g h t improvement i n the correlation (equation 4 ) . The combin-ation of log P and pKa (equation 5) or log P and u (equation 6) also s l i g h t l y improves the correlation i n equation 1. Equation 8 gave the best c o r r e l a t i o n . Comparison of equations 7 and 8, demon-s t r a t e s that equation 8 i s superior i n terms of s t a t i s t i c a l significance (p = 0.01) and reduction of the standard error of the estimate. Thus the regression equation with the added dipole moment term accounts better for the variation i n anticonvulsant a c t i v i t y than that with the added pKa term. The optimum log P value i n equation 8 below was 1.45. The negative c o e f f i c i e n t of 177 Table 39 Anticonvulsant a c t i v i t y of various drugs against clonic seizures induced by PTZ (s.c. 85 mg/kg) i n mice and their physicochemical constants Compound ED5 0 mmol/kg ED50 l o8 Po/w pKa a.b y» ' (debyes) 1. Valproic acid 0.70c 0.155 2.57c 4.95° 1.15 2. 2-Butylhexanoic acid 0.57c 0.243 3.34c 5.05c 1.15 3. Ethosuximide 0.922d 0.035 0.016b 9.1e 1.47 4. a,a-Dimethyl-succinimide 4.2e -0.623 -0.49f 9.1 1.47 5. Pentobarbital 0.057d 1.244 2.198 7.938 1.13 6. Barbital 0.293h 0.533 0.688 7.758 1.13 7. Metharbital 0.051h 2.973 1.21b 8.451 1.13 8. Butobarbital 0.085h 1.070 1.70§ 7.818 1.13 9. 5-Heptyltetrazole 0.31c 0.509 2.01° 5.31c 2.65 10. Dimethadione 5.75b -0.760 -0.93b 6.131 1.74 11. Trimethadione 2.083d -0.319 -0.37b - 1.74 12. Paramethadione 0.399d 0.399 0.13b - 1.69 13. a,a-Dimethyl-y-butyrolactone 3.51e -0.545 1.01f - 4.13 14. a,a-Ethylmethy1-Y-butyrolactone 1.17e -0.157 1.51f - 4.13 aTaken from A. L. McClellan (198). bTaken from E. J . Lien et a l . (28). °Taken from t h i s study. dTaken from R. L. K r a l l et a l . (38). ^Taken from W. E. Klunk et a l . (47). fCalculated using Hansch 7T-parameters, or fragraental constants (152). BTaken from S. Toon and M. Rowland (194). bTaken from A. Raines et a l . (200). xTaken from D. M. Woodbury et a l . (46). 178 Table 40 Equations correlating anti-PTZ activity and physicochemical properties of anticonvulsants in Table 39 Equations na sb rc F Pd 1. -log ED5Q = 0.2961 log P + 0.0327 14 0.925 0.387 2.12 0.17 2. -log ED50 = 0.150 pKa - 0.537 10 1.10 0.232 0.454 0.52 3. -log ED5Q = -0.37u + 1.03 14 0.917 -0.406 2.37 0.15 4. -log ED5Q = -0.873 log P - 0.262 (log P)2 + 0.103 14 0.859 0.573 2.70 0.11 5. -log ED50 = 0.586 log P + 0.431 pKa - 3.263 10 0.900 0.668 2.81 0.13 6. -log ED5Q = 0.286 log P - 0.360u + 0.710 14 0.874 0.552 2.41 0.14 7. -log ED50 = 1.11 log P - 0.302 (log P)2 + 0.205 pKa - 1.31 10 0.830 0.772 2.95 0.12 8. -log ED5Q = 1.14 log P - 0.392 (log P)2 -0.559u + 1.19 14 0.643 0.811 6.40 0.01 number of compounds used, ^standard error of estimate, ccorrelation coefficient, level of significant of F-value and degrees of freedom. Correlation matrix log P (log P) pKa V 1 0.8723 -0.5691 -0.0336 1 -0.7342 -0.1971 1 -0.3076 log P (log P) pKa U log 1/ED50 = 1.14 log P - 0.392 (log P )2 - 0.559 y + 1.19 (n = 14, r = 0.811, s = 0.643, log PQ = 1.45) the dipole moment variable i n equation 8 implied that an increase i n dipole moment without the necessary log P value would decrease the anticonvulsant potency. Similar findings, indicating the role of dipole moment i n anticonvulsant a c t i v i t y , have been reported i n the l i t e r a t u r e . Lien et a l . (26) reported that the best equation for phenyl- and alkyl-substituted heterocyclic a n t i e p i l e p t i c drugs was log 1/ED50 = 0.852 log P - 0.301 (log P )2 - 0.629u + 4.139 (n = 12, r = 0.915, s = 0.227, log PQ = 1.42) As i n t h i s study, highly active compounds i n different classes of an t i e p i l e p t i c drugs were used i n the regression analysis. B l a i r and Webb (27) i n v e s t i g a t e d the r e l a t i o n s h i p between anti-PTZ a c t i v i t y and physicochemical properties i n a set of 1,4-benzo-diazepines. The best equation was indicated to be log 1/ED5Q = -0.50y + 3.26 (n = 52, r = 0.626, s = 0.866) The equation showed that i n a series of 1,4-benzodiazepines with similar log P values or with almost equal access to the CNS s i t e s of action, dipole moment played the major role i n determining the anticonvulsant a c t i v i t y . 180 The dependence of a c t i v i t y on dipole moment may explain the low structural s p e c i f i c i t y of anticonvulsant compounds i n which s t e r i c effects are not apparent. Thus there i s a continuous spec-trum of st r u c t u r a l l y diverse compounds with varying molecular dipole moments. The physical meaning of the negative dependence of anticonvulsant a c t i v i t y on dipole moment i s not c l e a r . Those compounds with a low molecular dipole moment are not necessarily the most potent anticonvulsant compounds. The 1,4-benzodiazepines are reported to have higher dipole moments and log P values than other t r a d i t i o n a l a n t i e p i l e p t i c drugs (26). The dipole moment of diazepam was found to be 2.65 debyes (26). However, the benzo-diazepines show very high a c t i v i t y against PTZ-induced clonic seizures. The effectiveness of these compounds may be due to favourable conformations for hydrophobic binding and el e c t r o s t a t i c interaction at target s i t e s . Since the benzodiazepines could not be included in a series of anticonvulsants to develop QSAR (26), i t appears that d i f f e r e n t c l a s s e s of anticonvulsant drugs have characteristic molecular dipole moments that determine their elec-t r o s t a t i c interaction at speci f i c binding s i t e s . I t has been suggested that i n whole animal studies, the significance of dipole moment in QSAR of anticonvulsant drugs may be indicative of dipole-dipole or charge-dipole interaction not only at si t e s of action but also at si t e s of nonspecific binding (20). Pentylenetetrazole and 5-heptyltetrazole show opposing pharma-cological e f f e c t s , and have different physicochemical properties. 5-Alkyltetrazoles have an average dipole moment of 2.65 whereas 1,5-dialkyltetrazoles have an average value of 5.30 (198). Both 181 compounds are l i p i d - s o l u b l e while PTZ alone i s readily soluble i n water. PTZ, however, lacks acidic properties. I t i s obvious that i n such comparisons, differences exist i n h y d r o p h i l i c i t y , dipole moment values and aci d i c properties (pKa). Also s t e r i c differences cannot be ruled out. The dominating influence of dipole moment i n the preceding regression equation does not rule out the role of pKa i n determin-ing anticonvulsant a c t i v i t y . I t would be expected that among struc t u r a l congeners such as the alkylcarboxylic acids that there w i l l be high c o l l i n e a r i t y between pKa and u as reported for homo-logous alkyl-substituted compounds (134). However, as shown i n Table 40, for compounds with diverse polar groups the correlation between pKa and u i s low (r = 0.3076). In t h i s study, approximately 66% (r ) of the variance i n a n t i -convulsant a c t i v i t y of the s t r u c t u r a l l y diverse alkyl-substituted anticonvulsants could be accounted for by l i p o p h i l i c i t y and dipole moment (equation 8, Table 38). I t appears unlikely that a near-perfect correlation could be obtained since the anticonvulsant a c t i v i t y and physicochemical p r o p e r t i e s were determined from different sources. In addition, the correlation i n the QSAR study i s limited by the complex nature of the bi o l o g i c a l test system, whole animal. The v a r i a b i l i t y i n the anticonvulsant potency could be influenced by structural effects on absorption, metabolism, d i s t r i b u t i o n and e x c r e t i o n . These processes have a l s o been variously reported to be governed by physicochemical parameters such as l i p o p h i l i c i t y and pKa (193,194). Furthermore, s t e r i c factors could explain the residual variance unaccounted for by regression equations. 182 S t r u c t u r a l features that enhance or di m i n i s h anticonvulsant a c t i v i t y  A l iphatic substituents The lack of a c t i v i t y of isobutyric acid and reported i n a c t i v -i t y of non-branched C^HyCOOH to C^jCOOH (12,13,20) may be due to thei r low l i p o p h i l i c i t y . I t appears that a or 3-branching may be as effective as straight-chain a l k y l substitution i n conferring anticonvulsant a c t i v i t y . 5-Heptyltetrazole, with a straight-chain a l k y l substituent possessed potent anticonvulsant a c t i v i t y . From QSAR studies i t i s known that a l i p h a t i c substituents increase the l i p i d s o l u b i l i t y of compounds f o r entry i n t o the CNS. The anticonvulsant a c t i v i t y of t r i m e t h y l a c e t i c a c i d and 2,2-dimethylbutyric acid probably r e f l e c t the s t a b i l i z i n g effect of m u l t i p l e a-branching on o x i d a t i v e metabolism or increased hydrophobic groups for binding to a hydrophobic portion of the recognition s i t e . Increased hydrophobic binding by the Cy-chain may also account for the high a c t i v i t y of 5-heptyltetrazole. Valproic acid and i t s analogues probably bind to a protein complex i n the GABA-metabolizing enzyme system or the reported (30) protein complex of the GABA-benzodiazepine picrotoxin receptor complex. In v i t r o studies have shown that valproic acid (112) and other medium-chain f a t t y acids (199) bind with high a f f i n i t y to bovine or human serum albumin. Bovine or human serum albumin has been used as a convenient model to study the molecular basis of specific-ligand protein interactions (112,199), since many drug receptors have recognition s i t e s on protein subunits. Valproic acid and other medium chain f a t t y acids have been suggested, from i n v i t r o protein-binding studies, to bind with high a f f i n i t y to a 183 s i t e on human a l b u m i n i d e n t i c a l or c l o s e t o t h e indole/benzodiazepine s i t e (112,199). In a recent study, Brown et a l . (199) i n v e s t i g a t e d the albumin binding of normal a l i p h a t i c a c i d s (pentanoic a c i d up to nonanoic acid) using u l t r a f i l t r a t i o n techniques. A s i n g l e high a f f i n i t y s i t e was observed f o r each fatty acid with an increase i n number of secondary s i t e s with chain elongation. From the binding a f f i n i t i e s and competitive binding data, the authors suggested that there are d i s t i n c t albumin binding sit e s for the short-chain « C?) and the medium chain (C8-C9) fa t t y a c i d s . On the other hand, the high a f f i n i t y s i t e s of long chain fatty acids are reported to be different from that of medium-chain fatty acids (112,199). 5-Isoamyltetrazole was inactive. The carboxylic b i o i s o s t e r i c analogue, isohexanoic a c i d has a l s o been reported to be i n -active (5). This could be due to the low l i p o p h i l i c i t y of these compounds. Thus the effect of the isopropyl terminus could not be investigated i n t h i s series. The convulsant properties of barbit-urates with the isopropyl and isopropenyl terminus (Figure 3) have been attributed to either a s t e r i c influence from conformation of these groups at the r e c o g n i t i o n s i t e or to s t e r i c i n f l u e n c e on oxidative metabolism at the (co-l) position (19,99,100). b. A l i c y c l i c - s u b s t i t u e n t s C y c l o h e x y l a c e t i c a c i d and i t s t e t r a z o l e analogue, 5-cyclo-hexylmethyltetrazole were found to be inactive i n the dose range studied (Table 29). However, 1-methylcyclohexanecarboxylic acid was r e l a t i v e l y active even though the combined effect of log P and 184 pKa values i s similar to that of the cyclohexylmethyl-substituted compounds. Owing to the enormous difference i n a c t i v i t y between the cyclohexyl and cyclohexylmethyl-substituted compounds, i t i s apparent that s u b t l e s t e r i c e f f e c t s were re s p o n s i b l e f o r the difference i n a c t i v i t y . The Taft s t e r i c parameter, E_, has been frequently found to be inadequate i n describing s t e r i c properties that influence b i o l o g i c a l a c t i v i t y (195). Literature E_ values showed covariance with log P values (Table 37). The uniqueness of the cyclohexylmethyl group i n reducing a n t i -convulsant a c t i v i t y was evident i n other classes of anticonvulsants although i t appears to have been overlooked. 5-Cyclohexylmethyl barbiturate, XXXVI, has been reported to have limited b i o l o g i c a l a c t i v i t y compared to ba r b i t a l (201), while 5-ethyl-5-benzylbarb-i t u r a t e , XXXVII, has convulsant properties (201). The urea deriv-a t i v e , XXXVIII ( 8 ) , acylurea d e r i v a t i v e , XL ( 8 9 ) , and s p i r o -carboxylic acid derivative, XLI (98), have been reported to have limited a c t i v i t y against PTZ-induced seizures. On the other hand, 1-methylcyclohexylurea, XXXIX (205), and the spiro carboxylic derivative, XXIII (88), have been found to be active against PTZ-induced seizures at comparative doses. The lack of a c t i v i t y of cyclohexylmethyl-substituted compounds may be due to thei r i n a b i l i t y to assume a required conformation to interact with binding s i t e s . I t seems possible that t h i s may be due to i n vivo metabolic e f f e c t s . However, i t i s d i f f i c u l t to see how metabolic effects can uniformly apply to a l l the different classes and also not to those with cyclohexyl groups. The effects of structural features such as the cyclohexylmethyl group, benzyl group, i s o p r o p y l or i s o p r o p e n y l group may due to s t e r i c 185 Figure 17. Active and inactive a l i c y c l i c and alicyclicalkyl-substituted compounds. 186 interactions at the s i t e of action. The molecular conformation of the convulsant b a r b i t u r a t e s with an i s o p r o p y l or isopropenyl terminus i n the butyl side chain (Figure 3) has been investigated u s i n g NMR s p e c t r o s c o p y (100) and m o l e c u l a r o r b i t a l calculations (99). I t was suggested from these studies that the convulsant properties may be due to the s t e r i c influence of the isopropyl or isopropenyl terminus. Confirmation of the s t e r i c influence of these structural features w i l l have to await results of further i n v i t r o and i n vivo s t r u c t u r e - a c t i v i t y studies. Recent studies on spiro analogues of valproic acid (88) and r i g i d c ycloalkyl compounds (11) have been aimed at revealing con-figurations which would confer high a c t i v i t y of valproic acid analogues. In analogy to structures of the GABA agonists, i s o -guvacine and THIP (4,5,6,7-tetrahydro-isoxazolopyridin-3-ol), 1-methylcyclohexanecarboxylic acid derivatives w i l l be of interest for future studies. c. Effects of a polar f u n c t i o n a l i t y i n the a l k y l chain N,N-dibutylsuccinamic acid was found to possess convulsant properties while N,N,-diethylsuccinamic acid appeared to induce h y p e r a c t i v i t y at the dose range t e s t e d . Both compounds have r e l a t i v e l y high pKa and log P values when compared to the other compounds studied. Secondary and t e r t i a r y amides of a l i p h a t i c acid amides (3,4,50), succinic acid-bis-dialkylamides, t e t r a a l k y l -ureas (49), and polymethylene lactams (26) have been reported to have stimulant and/or convulsant properties. The N,N-dialkyl-succinamic acids and these compounds share a common functional 187 group, the substituted amido group. According to Lien et a l . (26) the convulsant a c t i v i t y of some alkyl-substituted ureas and lactams increased with higher molecular dipole moments. The two succinamic acids studied appear to show some structur-a l s i m i l a r i t i e s to the putative excitatory transmitters, aspartic and glutamic acids. I t has been proposed that glutamic-like act-i v i t y required a cationic and two anionic centres i n the molecular structure (204). Due to amide resonance i n the succinamic a c i d s there could be two centres of negative charge and a positive centre but with charge separation different from that of glutamic acid or asp a r t i c acid. 0 _ II 0-C-CH2-CH2-CH-C00~ glutamic acid NH3+ 0 _ II 0-C-CH2-CH-C00~ aspartic acid NH3+ 0 0 _ II _ II 0-C-CH2-CH2-C-06- — 0-C-CH2-CH2-C=0 -(jj-06- — 0-C-CH2-CH2-C=C N,N-dialkylsuccinamic acid L i k e PTZ, the two succinamic acids are water-soluble and li p i d - s o l u b l e . I t has been pointed out that an optimal balance of water and l i p i d s o l u b i l i t y may also be important i n the development 188 of convulsant a c t i v i t y due to d i s t r i b u t i o n a l l o c a l i z a t i o n of com-pounds i n the CNS (49). Succinamic acids and glutaramic acid are the open-chain forms of the anticonvulsant succinimides and glut-arimides, respectively. They are reported to be formed i n minor amounts from i n vivo transformation of the succinimides and glut-arimides (49,98). Published studies have indicated the lack of anticonvulsant a c t i v i t y of a series of N-alkyl-2-phenylsuccinamic acids (203) and s l i g h t convulsant a c t i v i t y of 3,3-ethylmethyl-glutaramic a c i d , the open-chain form of the convulsant beme-grid eR (49). d. Model showing selective effects of a l i p h a t i c and a l i c y c l i c  substituents at the hydrophobic binding s i t e The results of x-ray or theoretical analysis on GABA structure have indicated the presence of two preferential conformers, one planar extended conformation, the other a non-planar folded con-formation (Figure 18a). Ferrandes et a l . (205) also determined the conformation of c r y s t a l l i n e amides of valproic acid and pointed out that the molecular conformation of valproic acid may overlap with the GABA conformers. The dipropyl branched-chain was shown to adopt a planar extended conformation which i s symmetrical with respect to the amido group (Figure 18b). Cohen-Addad (206) deter-mined the molecular conformation of c r y s t a l l i n e amides of 3-ethyl-pentanoic acid and indicated that the 6,8-diethyl branched chain was not oriented symmetrically with respect to the amido group as i n valproic acid (Figure 18c). However, the conformation of a segment of 3-ethylpentanoic acid could be compared to that observed for valproic a c i d . As shown i n Figure 18, the geometry of the 189 -617A GABA extended planar conformation 1*4-1 4.22A -5-58 A (a) GABA folded conformation Figure 18. Conformation of (a) GABA, (b) Valproic acid amide, and (c) 3-ethylpentanoic amide. 190 atoms [0^02,03,C"pC"2] of 3-ethylpentoic acid was shown to be equivalent to that of [CpC2,C^.C^.C'g] from valproic a c i d , and also the folded conformation of GABA. However, the ethyl groups i n 3-ethylpentanoic are not stereochemically i d e n t i c a l . In a previous single dose study (5), 3-ethylpentanoic acid was found to be i n -a c t i v e . This result i s contrary to the present study which has shown i t s effectiveness i n antagonizing PTZ-induced seizures (Table 31). From the conformations shown i n Figure 18, i t i s l i k e l y that the more stable conformations of the anticonvulsant compounds i n t h i s study w i l l have segments of the a l k y l c h a i n almost indistinguishable from the folded or extended conformations of GABA. The f l e x i b l e normal a l i p h a t i c side chain i n 5-heptyl-tetrazole would l i k e l y adopt an equilibrium conformation that superimposes with the extended GABA conformation. Convulsant tetrazoles have been reported to interact at the picrotoxin s i t e of the GABA receptor complex l i k e valproic acid ( 7 ) . Kraus (24) has also reported that valproic acid and i t s tetrazole analogue, 4-tetrazolylheptane inhibited succinic semi-aldehyde dehydrogenase, an enzyme i n the GABA metabolism shunt, with i n h i b i t o r y constants of 0.7 mM and 0.75 mM respectively. The conformation of the amide of the highly active dibutylacetic acid was also reported to be similar to that of valproic acid with the dibutyl a l k y l chain oriented symmetrically to the amide group (206). Similarly the conformations of the a l k y l chain i n the unsaturated a c t i v e compounds may overlap with either the extended or folded GABA conformation. The a l k y l configurations i n 2,2-dimethylbutyric acid and trimethylacetic acid would l i k e l y overlap with either the 191 extended or the folded GABA conformation. In the active a l i c y c l i c - s u b s t i t u t e d compound, 1-methylcyclo-hexanecarboxylic a c i d , the semi-rigid cyclohexyl chain conformation (Figure 19) would l i k e l y conform to the GABA extended or folded conformation with overlap of the atoms [C^.C^.C^] i n Figure 18a. The cyclohexylmethyl group i n the inactive compounds, 5-cyclohexyl-methyltetrazole and cyclohexylacetic a c i d , appears to have segments which overlap with atoms [C2, C3, C^] of the GABA extended or folded conformation (Figure 18a). However, i n comparison with 1-methylcyclohexanecarboxylic a c i d , the presence of the methylene group between the polar functional group and the a l i c y c l i c ring may cause the terminal portion of the cyclohexyl ring to appear i n the position occupied by the nitrogen group of GABA. Thus, unlike the 3,8-diethyl chain i n 3-ethylpentanoic a c i d , the presence of two methylene groups connecting C2" and (Figure 18c) may explain the unexpected i n a c t i v i t y of cyclohexylacetic acid and 5-cyclohexyl-raethyltetrazole. The structural model (Figure 19) proposed for the pharma-cophore features i n valproic analogues i s based on the i n a c t i v i t y and anticonvulsant a c t i v i t y of the carboxylic acids and tetrazoles studied. Both alkylsubstituted carboxylic acids and tetrazoles were included i n the pharmacophoric model because of reports of po s s i b l e s i m i l a r binding s i t e s of v a l p r o i c a c i d and t e t r a -zoles (33,34,55,57,104). Although the evidence that v a l p r o i c a c i d and convulsant tetrazoles enhance or diminish GABA-mediated postsynaptic inhib-192 Position of N in GABA conformations C" \ C Q electron-donor Q \ C 9r 0 UPs " ' ~ - C — \ / OH hydrogen-C bonding \ groups / I Steric Effect C , IK Cyclohexylacetic acid CH. 1-Methylcyclohexane-carboxylic acid h. electron-donor groups w H hydrogen-bonding groups Zone for alpha alkyl groups H 5-Cyclohexylmethyltetrazole Pentylenetetrazole (PTZ) Figure 19. Model of pharmacophoric structural features in carboxylic acids and tetrazoles. 193 i t i o n may not be s t r o n g or complete (compared t o 1,4-benzodiazepines and barbiturates), there has been a growing body of evidence implying i n h i b i t i o n of GABA metabolism by valproic acid at synaptic sit e s (2). Valproic acid may be a potential substrate of GABA-transaminase or succinic semialdehyde dehydrogenase. Valproic acid and i t s tetrazole analogue, 4-tetrazolylheptane were shown to i n h i b i t succinic semialdehyde dehydrogenase (24). In the structural model (Figure 19), the a l k y l chain probably binds to a hydrophobic s i t e on the GABA recognition s i t e . This hydrophobic s i t e i s probably between binding s i t e s for the nitrogen and carboxylic groups in GABA. The p o s s i b i l i t y that GABA exists i n either the extended or folded conformation has been considered i n the proposed structural model. The lack of anticonvulsant a c t i v i t y of c y c l o h e x y l a c e t i c a c i d and 5-cyclohexylraethyltetrazole i s suggested to be due to s t e r i c effects at the hydrophobic s i t e , close to the GABA nitrogen binding s i t e on the GABA-receptor com-p l e x . The zone of a l k y l groups at the a - p o s i t i o n i n f l u e n c e s anticonvulsant a c t i v i t y as shown by the a c t i v i t y of a-cyclohexyl-pentanoic acid which was, however, reported to be less than that of VPA (5 ) . Thus i f metabolism effects on the cyclohexylmethyl-substituted and cyclohexylsubstituted compounds are s i m i l a r , then the difference i n a c t i v i t y i s suggested to be due to conformational effects at the s i t e of action. The polar moiety plays a si g n i f i c a n t role' i n determining the anticonvulsant a c t i v i t y of compounds as shown by the presence of the dipole moment term i n the QSAR. I t i s considered that favour-able orientation of the electron-donor or hydrogen-bonding groups 194 i n the c a r b o x y l i c and t e t r a z o l e m o i e t i e s , i n a d d i t i o n to hydro-phobic binding of a l k y l chain, i n f l u e n c e anticonvulsant a c t i v i t y . V a l p r o i c a c i d i s s t r o n g l y bound to the serum p r o t e i n , albumin (112). A theoretical evaluation of the interaction between valproic acid and human serum albumin has been made by Andrews et a l . (207). V a l p r o i c a c i d was i n d i c a t e d to meet the geometric requirements of the s p e c i f i c binding s i t e s and the carboxylic group was implicated i n the high strength of the non-covalent bonding. The presence of polar f u n c t i o n s i n the a l k y l chain of the succinamic acids may imply different dipole-dipole or dipole-charge interactions compared with that of valproic acid. I t i s l i k e l y that the convulsant or s t i m u l a t o r y succinamic acids i n t e r a c t at a different s i t e from that of alkylsubstituted carboxylic acids and tetrazoles. The i n vivo antagonism between valproic acid and N,N-dibutylsuccinamic acid was, however, not investigated. The structure-activity c o r r e l a t i o n s i n t h i s study i m p l i c a t e both the carboxylic and tetrazole analogues of valproic acid i n the structural pharmacophore model shown i n Figure 19. X-ray analysis of c r y s t a l s of pentylenetetrazole-iodine monochloride complex by Baenziger et a l . (208) indicated that the tetrazole ring i s planar with the C^-Cy-Cg-Cg-Cio a l i c y c l i c ring i n the chair conformation (Figure 19). The electron-donor group, bound to the iodine-mono-c h l o r i d e was found to be the nitr o g e n atom at p o s i t i o n - 4 of the tetrazole r i n g . I t has been reported that s u b s t i t u t i o n of a methyl or i s o -propyl group at C8 °f pentylenetetrazole increased the stimulatory a c t i v i t y while substitution of s i m i l a r a l k y l groups at Cg and C^Q led to a decrease i n stimulatory a c t i v i t y (58,59). I t i s conceiv-195 able that substitution of a l k y l groups i n the pentylene chain of PTZ exerts stereochemical effects similar to that proposed for the carboxylic and tetrazole analogues of valproic acid (Figure 19). However, these p o s s i b i l i t i e s can only be evaluated adequately when the x-ray structures of these compounds have been determined. I t i s also worth investigating i f the preferential conformations i n solution phase correspond to those i n the s o l i d s t a t e . I t appears that further s t r u c t u r e - a c t i v i t y correlations of analogues of GABA, valproic acid,picrotoxin and pentylenetetrazole may provide supporting evidence for the active conformations and structural basis for interactions at the GABA-receptor complex. This could lead to compounds with greater s p e c i f i c i t y i n t h e i r anticonvulsant actions. 196 SUMMARY AND CONCLUSIONS A wide range of struct u r a l l y - r e l a t e d analogues of valproic acid have been synthesized and examined for anticonvulsant a c t i v i t y . The s e r i e s of analogues included compounds with a - a l k y l s u b s t i t u e n t s , 6,6-dialkyl substituents, B-alkyl substituents, Y-alkyl substituents, a l i c y c l i c and a l i c y c l i c a l k y l groups, an unsaturated a l k y l chain and the tetrazole functional group as a bi o i s o s t e r i c function of the carboxylic group. The compounds were synthesized using known procedures. A stereoselective method was used to prepare the diunsaturated analogues of valproic a c i d , 2-[(E)-l'-propenyl]-(E)-2-pentenoic acid and 2-[(Z)-l'-propenyl]-(E)-2-pentenoic a c i d . The stereochemical course of the deconjugative a l d o l - t y p e r e a c t i o n of e t h y l 2-pentenoate with propionaldehyde was investigated using the (E) or (Z)-isomer of ethyl 2-pentenoate. The addition reaction was shown to occur with high regio-s e l e c t i v i t y and deconjugation occurred with high s t e r e o s e l e c t i v i t y . The 6'-hydroxy-B',y'-unsaturated products were dehydrated with various de-hydration agents. The combination of methanesulfonyl chloride and KH resulted i n the minimum number of diunsaturated VPA ethyl ester isomers. NMR spectroscopic data f o r each dienoate isomer was ch a r a c t e r i z e d following p u r i f i c a t i o n of the product mixture by argentation TLC. Using synthesized reference diunsaturated acids, the unidentified major d i -unsaturated VPA metabolite has been assigned the chemical structure, 2-[(E)-l'-propenyl]-(E)-2-pentenoic a c i d . L i p o p h i l i c and electronic properties of test compounds have been determined experimentally to ide n t i f y the molecular properties relevant to th e i r anticonvulsant a c t i v i t y . L i p o p h i l i c i t y was described by the 197 octanol-water p a r t i t i o n c o e f f i c i e n t . The p o s s i b i l i t y of estimating octanol-water p a r t i t i o n c o e f f i c i e n t s (P) of compounds of diverse struc-tures by RP-HPLC was explored. A 5ym Hypersil ODS column was used. Different compositions of methanol-phosphate buffer (pH 3.5) or aceto-nitrile-phosphate buffer were investigated for t h e i r s u i t a b i l i t y as mobile phase to determine hydrophobic e f f e c t s . Analysis of the linear regressions of known log P values of refer-ence substances and their log [capacity factors] for the different mobile phases culminated i n the selection of 70% methanol-phosphate buffer and 50% acetonitrile-phosphate buffer as the mobile phases to be used i n determining the HPLC log P values. The two mobile phases showed high correlation c o e f f i c i e n t s , good s e n s i t i v i t y to hydrophobic effects and eluted h i g h l y l i p o p h i l i c compounds with minimization of non-equilibrium adsorption phenomena. HPLC log P values, determined from the regression equations, were found to be better than Hansch or Rekker predicted log P values i n determining accurate log P values for highly substituted a l i p h a t i c and a l i c y c l i c carboxylic acids, heterocyclic t e t -razoles and intramolecularly-bonded compounds. The accuracy of the HPLC log P values was v e r i f i e d by determining the shake-flask log P values of four compounds of diverse structure. An HPLC method was described for measuring compounds i n aqueous solutions. Owing to i t s speed, high r e p r o d u c i b i l i t y , and a b i l i t y to determine the log P values of unstable compounds, the RP-HPLC method complements and provides an alternative to the conventional shake-flask procedure. The electronic properties of the test compounds were described by the apparent ionization constants. The ionization strength was deter-mined with good precision by a potentiometric t i t r a t i o n method. Mixed aqueous-methanol solvents were used because of the limited aqueous 198 s o l u b i l i t y of the compounds. pKa Values i n 10% methanol-water were found to be more meaningful than values i n 50% methanol-water. The isdsterism between carboxylic acid and tetrazole was indicated by the pKa values of t e t r a z o l e s which were j u s t s l i g h t l y higher than the corresponding carboxylic acids. The standardized s.c. PTZ seizure threshold was used to establish the anticonvulsant potency of test compounds i n mice. Anti-PTZ a c t i v i t y of each compound was determined at i . p . doses between 0.2 and 2.0 mmol/kg. The r e l i a b i l i t y of the pharmacological testing was shown by the close agreement of the measured ED^Q of valproic acid to l i t e r a t u r e values. The s.c. PTZ seizure test was noted to be more selective i n discrimination of the anticonvulsant potency than the PTZ mortality test used i n previous studies of some a l i p h a t i c acids. 5-Heptyltetrazole was the most potent compound tested. However, valproic acid had the most desirable pharmacological properties i n terms of high anticonvulsant a c t i v i t y and minimum qual i t a t i v e toxic e f f e c t s . QSAR studies showed that the dependency of anticonvulsant potency on a combination of log P and pKa was s t a t i s t i c a l l y s i g n i f i c a n t when 5-heptyltetrazole was deleted from the series of active compounds. The stated relationship suggested that the i n vivo a c t i v i t y was governed by the proportion and l i p o p h i l i c i t y of the unionized form for entry into the CNS. The f a i l u r e of log P and pKa to account for the variation i n potency of the nine active compounds including 5-heptyltetrazole was attributed to lack of an effective parameter to measure the electronic properties of the polar functional group which e l e c t r o s t a t i c a l l y i n t e r -acts with the recognition s i t e i A comparison between pKa and dipole moment i n describing the elec-199 tron properties was undertaken using highly active alkyl-substituted compounds from different classes of anticonvulsant drugs. These a l k y l substituted compounds have been suggested by various workers to have a common mechanism of action. The physicochemical properties and anti-PTZ ED^Q values were obtained from l i t e r a t u r e sources. Regression analysis of the QSAR showed that for the different classes of alkyl-substituted compounds, the dependence of a c t i v i t y on log P and u terms was stat -i s t i c a l l y s i g n i f i c a n t compared to dependence of a c t i v i t y on log P and pKa. Dipole moment values showed low covariance with pKa va l u e s . Similar findings of negative dependence of anticonvulsant a c t i v i t y on dipole moment have been reported for 1,4-benzodiazepines and other groups of anticonvulsants. Further studies may be required for e l u c i d -ation of the negative dependence of a c t i v i t y on dipole moment. A major finding of the st r u c t u r e - a c t i v i t y studies was the unique structural property of the cyclohexylmethyl configuration i n greatly reducing the anticonvulsant a c t i v i t y of cyclohexylacetic acid and 5-cyclohexylmethyltetrazole. A model of the structural requirement at the s i t e of action has been proposed to account for the a c t i v i t y and i n -a c t i v i t y of v a l p r o i c a c i d analogues. This model i s based on the reported X-ray structure of GABA, c r y s t a l l i n e amides of valproic a c i d , 3-ethylpentanoic ac i d , dibutylacetic a c i d , pentylenetetrazole and the hypothesized mechanism of a c t i o n of v a l p r o i c a c i d and pentylene-tetrazole. The modulatory influence of the different a l i p h a t i c and a l i c y c l i c substituents probably results from binding to a hydrophobic s i t e on the GABA-receptor complex or the GABA metabolizing enzyme system. The cyclohexylmethyl conformation was suggested to be less effective i n hydrophobic bonding due to i t s s t e r i c effect at a stereo-selective position on the hydrophobic recognition s i t e . 200 The significance of th i s research l i e s i n the delineation of the dependence of anticonvulsant a c t i v i t y on the physicochemical properties: l i p o p h i l i c i t y , apparent acid ionization constant, dipole moment and s t e r i c f a c t o r s . While the proportion and l i p o p h i l i c i t y of unionized species governed the i r access to CNS si t e s of action, the dependence of a c t i v i t y on dipole moment may explain the diverse structures of a n t i -convulsants. There appears to be a provision that s t e r i c requirements of the hydrophobic binding s i t e are accommodated i n active compounds. S t e r i c e f f e c t s , as suggested to be present i n cyclohexylmethyl-substituted compounds, may lead to i n a c t i v i t y or probably convulsant properties as i n barbiturates with an isopropyl terminus or benzyl sub-s t i t u e n t s . A survey of the available l i t e r a t u r e data indicated that there has been no report of the high potency of a 5-alkyltetrazole (e.g. 5-heptyl-tetrazole) i n antagonizing the convulsant effects of a closely-related analogue, pentylenetetrazole. These findings have pharmacological significance i n elucidating the nature of the active s i t e of valproic acid and i t s analogues. I t can be postulated from the results of t h i s s t r u c t u r e - a c t i v i t y study that the a c t i v i t y of valproic a c i d , 5-heptyl-te t r a z o l e , cyclohexylacetic a c i d , 5-cyclohexylmethyltetrazole, pentyl-enetetrazole and the other t e t r a z o l e and c a r b o x y l i c a c i d s may be governed by common structural requirements at a similar s i t e of action. This study also reports the stereochemical course of a stereo-selective synthesis of 2-[(Z)-l'-propenyl]-(E)-2-pentenoic acid and 2-[(E)-l ' - p r o p e n y l ] - ( E ) - 2 - p e n t e n o i c a c i d . The major diunsaturated metabolite of v a l p r o i c a i d was i d e n t i f i e d . Studies of the t o x i c properties of the diunsaturated analogues of valproic acid could be 201 undertaken with due regard to their importance as metabolites of v a l -proic a c i d . The i n vivo s t r u c t u r e - a c t i v i t y studies are useful, despite the li m i t a t i o n s of whole animal studies, for establishing correlations between i n vivo and i n v i t r o actions of valproic acid analogues. 202 REFERENCES 1. H. Meunier, G. Carraz, Y. Meunier, P. Eymard and M. Aimard. P r o p r i e t e s pharmacodynamique de l ' a c i d e d i p r o p y l a c e t i q u e . Therapie, 18, 435 (1963). 2. A. Chapman, P.E. Keane, B.S. Meldrum, J . S i m i a n d and J.C. Vernieres. Mechanism of anticonvulsant a c t i o n of v a l p r o a t e . Prog. Neurobiol., 19, 315 (1982). 3. J . L. Benoit-Guyod, A. Boucherle, M. Benoit-Guyod, R. Rupp, P. Eymard, G. Carraz, M. B o i t a r d , S. Lebreton, H. B e r i e l and H. Meunier. Derives de l'acide d i p r o p y l a c e t i q u e . I I I . Nouveaux amides et e s t e r s . Eur. J . Med. Chem., _3» 336 (1968). 4. G. Carraz, A. Boucherle. S. Lebreton, J.L. Benoit-Guyod and M. B o i t a r d . Le neurotropisme de l a s t r u c t u r e d i p r o p y l a c e t i q u e . Therapie 19, 917 (1964). 5. G. T a i l l a n d i e r , J.L. Benoit-Guyod, A. Boucherle, M. B r o i l and P. Eymard. Recherches dans l a serie d i p r o p y l a c e t i q u e X I I . Acid et alcool aliphatiques ramifies anticonvulsants. Eur. J . Med. Chem., jLO, 453 (1975). 6. G. Carraz and N. Emin. A c t i o n anticonvulsante du monooureide de l'acide d i p r o p y l a c e t i q u e et du derive hydanoinique de cet acide. Therapie, 22, 641 (1967). 7. F. De Marchi and M. V. T o r r i e l l i . Synthese et proprietes a n t i -convulsantes de l a chloro-7 dihydro-1,3 dipropylacetoxy-3 phenyl-5(H) benzodiazepine -1,4 one-2. Eur. J . Med. Chem., 3.. 430 (1968). 8. M. Benolit-Guyod, J.L. Benoit-Guyod, A. Boucherle, M. B r o i l and P. Eymard. Recherches dans l a s e r i e dipropylacetique. V I I I . Structures homologues: amides et urfes de l a propyl-2 pentylamine. Eur. J . Med. Chem., 7, 393 (1972). 9. M. Benoit-Guyod, J.L. Benoit-Guyod, M. B r o i l , A. B o u c h e r l i e , J.P. Werbenec and P. Eymard. Recherches dans l a serie d i p r o p y l -acetique. X. Cetones et a l c o o l secondaries a chaine propyl-1 b u t y l e . A c t i o n sur l e systeme nerveux c e n t r a l . Eur. J . Med. Chem., 8, 419 (1973). 10. G. Carraz. Approaches pour une theorie sur l ' a c t i v i t e de l a struc-ture dipropylacetique. Agressologie, j5, 13 (1967). 11. M. F. Brana, M. M a r t i n e z , J . Garrido and CM. Roldan, S i n t e s i s y a c t i v i d a d farmacologica de homogos c i c l i c o s d e l acido d i p r o p i l -a c e t i c o . An. Quim., 79, 47 (1983). 12. A.G. Chapman, B. S. Meldrum and E. Mendes. Acute anticonvulsant a c t i v i t y of s t r u c t u r a l analogues of v a l p r o i c a c i d and changes i n brain GABA and aspartate content. L i f e Sci., 32, 2023 (1983). 203 13. P.E. Keane, J . Simiand, E. Mendes, V. Santucci and M. Morre. The effect of analogues of valproic acid on pentylenetetrazole-induced s e i z u r e s and b r a i n GABA content i n mice. Neuropharmacology, 22, 875 (1983). 14. B.J. Perlman and D.B. Goldstein. Membrane disordering potency and anticonvulsant action of valproic acid and other short-chain f a t t y a c i d s . Mol. Pharmacol., 26, 83 (1984). 15. P.R. Andrews. Molecular o r b i t a l c a l c u l a t i o n s on anticonvulsant drugs. J . Med. Chem., 12, 761 (1969). 16. T.J. Putnam and H.H. M e r r i t t . Chemistry of anticonvulsant drugs. Arch. Neurol. Psychiat., 43, 505 ( 1940). 17. T.B. P a t r i c k and R.R. Bresee. E n t h a l p i e s of hydrogen bonding i n pscyhotropic drugs. J . Pharm. S c i . , 65, 1066 (1976). 18. A. Camerman and N. Camerman i n "Antiepileptic Drugs: Mechanism of Ac t i o n " , G.H. Glaser, J.K. Penry and D.M. Woodbury, Eds., Raven Press, New York, p.223 (1980). 19. P.R. Andrews, G.P. Jones and D. Lodge. Convulsant, a n t i c o n v u l s a n t and anesthetic b a r b i t u r a t e s . 5-Ethyl-5-(3'-methylbut-2,-enyl)-barbituric acid and related compounds. Eur. J . Pharmacol., 55, 115 (1979) . 20. G.L. Jones and D.M. Woodbury. Anticonvulsant S t r u c t u r e - A c t i v i t y R e l a t i o n s h i p s . H i s t o r i c a l Development and Probable Causes of F a i l u r e . Drug. Dev. Res., 2, 333 (1982). 21. R.M. Herbst. T e t r a z o l e s as c a r b o x y l i c a c i d analogs. "Essays i n Biochemistry". J . Wiley, New York, p.141 (1956). 22. P.F. Juby, T.W. Hudyma and M. Brown. Preparation and a n t i -inflammatory properties of some 5-(2-Anilinophenyl)tetrazoles. J . Med. Chem., 1_1, 111 (1968). 23. H. Singh, A.S. Chawla, V.K. Kapoor, D. Paul and R.K. Mahhotru. M e d i c i n a l chemistry of t e t r a z o l e s . Progr. i n Med. Chem., 17, 151 (1980) . 24. J.L. Kraus. Isosterism and molecular modification i n drug design. New dipropylacetate analogs as i n h i b i t o r s of succinic semialdehyde dehydrogenase. Pharmacol. Res. Commun., 15, 119 (1983). 25. C. Hansch and T. F u j i t a . A method for the correlation of bi o l o g i c -a l a c t i v i t y and chemical s t r u c t u r e . J . Amer. Chem. S o c , 86, 1616 (1964). 26. E.J. L i e n , R.C.H. Liao and H.G. Shinouda. Q u a n t i t a t i v e s t r u c t u r e -a c t i v i t y relationships and dipole moments of an t i c o n v u l s a n t s and CNS depressants. J . Pharm. S c i . , 68, 463 (1979). 204 27. T. B l a i r and G.A. Webb. E l e c t r o n i c f a c t o r s i n the s t r u c t u r e -a c t i v i t y r e l a t i o n s h i p of some l,4-benzodiazepine-2-ones. J . Med. Chem., 20, 1206 (1977). 28. E.J. L i e n , G.L. Tong, J.T. Chou and L.L. L i e n . Structure r e q u i r e -ments for centrally acting drugs. J . Pharm. Sci., 62, 246 (1973). 29. C. Hansch and J.M. Clayton. L i p o p h i l i c character and b i o l o g i c a l a c t i v i t y of drugs. The p a r a b o l i c case. J . Pharm. S c i . , 62, 1 (1973). 30. R.W. Olsen. GABA-Benzodiazepine-Barbiturate receptor interactions. J . Neurochem., 37, 1 (1981). 31. R.L. Macdonald and J.L. Barker. Enhancement of GABA-mediated post-synaptic i n h i b i t i o n i n cultured spinal cord neurons. A common mode of anticonvulsant action. Brain Res., 167, 323 (1979). 32. B.A. Weisraan, T.R. Burker, K.C. Rice and P. Sko l n i c k . A l k y l - s u b -s t i t u t e d gamma-butyrolactones i n h i b i t [3^S]-TBPS binding to GABA linked chloride ionophore. Eur. J . Pharmacol., 105, 195 (1984). 33. M.K Ticku and R.W. Olsen. I n t e r a c t i o n of b a r b i t u r a t e s w i t h d i -h y d r o p i c r o t o x i n i n binding s i t e s r e l a t e d to the GABA-receptor ionophore system. L i f e Sci., 22, 1643 (1978). 34. M.K Ticku and W.C. Davis. E f f e c t of v a l p r o i c a c i d on ^ H-diazepam and 3H-d ihydropicrotoxinin binding s i t e s at the benzodiazepine-GABA receptor-ionophore complex. Brain Res., 223, 218 (1981). 35. D. Simon and J.K. Penry. Sodium d i p r o p y l a c e t a t e i n the treatment of epilepsy. A review. E p i l e p s i a , 16, 549 (1975). 36. R.M. Pinder, R.N. Brogden, T.M. Speight and G.S. Avery. Sodium valproate: A review of i t s pharmacological p r o p e r t i e s and th e r a -peutic efficacy i n epilepsy. Drugs, 13, 81 (1977). 37. S. Sato, B.G. White, J.K. Penry, F.E. D r e i f u s s , J.C. S a c k e l l a r e s and H.J. Kupferberg. V a l p r o i c a c i d versus ethosuximide i n the treatment of absence seizures. Neurology, 32, 157 (1983). 38. R.L. K r a l l , J.K. Penry, B.G. W h i t e , H.J. K u p f e r b e r g and E.A. Swinyard. An t i e p i l e p t i c drug development. I I . Anticonvuls-ant drug screening. E p i l e p s i a , 19, 409 (1978). 39. H.H. Frey and W. Loscher. D i p r o p y l a c e t i c a c i d - p r o f i l e of a n t i -convulsant a c t i v i t y i n mice. Arzneim. Forsch., 26, 299 (1976). 40. A.K. Shenoy, J.T. Miyahara, E.A. Swinyard and H.J. Kupferberg. Comparative anticonvulsant a c t i v i t y and neurotoxicity of clobazam, diazepam, phenobarbital and valproate i n mice and ra t s . E p i l e p s i a , 23, 399 (1982). 205 41. E.A. Swinyard, W.C. Brown and L.S. Goodman. Comparative assays of a n t i e p i l e p t i c drugs i n mice and r a t s . J . Pharmacol. Exp. Ther., 106, 319 (1952). 42. H.H. M e r r i t t and T.J. Putnam. A new series of anticonvulsant drugs tes t e d by experiments on animals. Arch. Neurol. Psychiat., 39, 1003 (1938). 43. J.E.P. Toman, E.A. Swinyard and L.S. Goodman. P r o p e r t i e s of max-i m a l s e i z u r e s and t h e i r a l t e r a t i o n by anticonvulsant drugs and other agents. J . Neurophysiol., 9, 231 (1946). 44. G.M. Everett and R.K. Richards. Comparative anticonvulsive action of 3,5,5-trimethyloxazolidine-2,4-dione (Tridione), d i l a n t i n and phenobarbital. J . Pharmacol. Exp. Ther., 81, 402 (1944). 45. M.J. O r l o f f , H.L. W i l l i a m s and C.C. P f e i f f e r . Timed intravenous i n f u s i o n of metrazol and strychnine f o r t e s t i n g anticonvulsant drugs. Proc. Soc. Exp. B i o l . Med., 70, 254 (1949). 46. E.A. Swinyard and J.H. Woodhead i n " A n t i e p i l e p t i c Drugs", D.M. Woodbury, J.K. Penry and C.E. Pippenger, Eds., Raven Press, New York, p . l l l (1982). 47. W.E. Klunk, D.F. Corey and J.A. F e r r e n d e l l i . S t r u c t u r e - a c t i v i t y relationship of a l k y l substituted gamma-butyrolactones andsuccin-imides. Mol. Pharmacol., 22, 444 (1982). 48. R. Testa, L. Graziani and G. Graziani. Do different anticonvulsant tests provide the same information concerning the p r o f i l e of a n t i -e p i l e p t i c a c t i v i t y ? Pharmacol. Res. Commun., 15, 765 (1983). 49. F. Hahn. Analeptics. Pharmacol. Rev., 12, 447 (1960). 50. D.M. Woodbury i n " A n t i e p i l e p t i c Drugs: Mechanism of A c t i o n " . G.H. G l a s e r , J.K. Penry and D.M. Woodbury, Eds., Raven Press, New York, p.249 (1980). 51. R.J. David, W.A. Wilson and A.V. Escueta. Voltage clamp a n a l y s i s of pentylenetetrazole effects on Aplysia neurons. Brain Res., 67, 549 (1974). 52. M. Kleen, D. Faber and W. Heiss. Strychnine and pentylene-t e t r a z o l e - i n d u c e d changes of e x c i t a b i l i t y i n A p l y s i a Neurons. Science, 179, 1133 (1973). 53. R.LJ^acdonald and J.L. Barker. P e n i c i l l i n and p e n t y l e n e t e t r a z o l e s e l e c t i v e l y antagonize GABA-mediated postsynaptic i n h i b i t i o n of cultured mammalian neurones. Neurology, 27, 337 (1979). 54. T.C. Pellman and W.A. Wilson. Synaptic mechanism of pentylene-t e t r a z o l e : S e l e c t i v i t y f o r c h l o r i d e conductance. Science, 197, 912 (1977). 206 55. W. Loscher and H.H. Frey. Effect of convulsant and anticonvulsant agents on l e v e l and metabolism of gamma-aminobutyric acid i n mouse b r a i n . N.S. Arch. Pharmacol., 196, 263 (1977). 56. R.F. S q u i r e s , E. Saederup, S.N. Crawley, P. Skolnick and S.M. P a u l . Convulsant potencies of t e t r a z o l e s are h i g h l y c o r r e l a t e d w i t h a c t i o n on GABA/Benzodiazepine/Picrotoxin receptor complexes i n b r a i n . L i f e S c i . , 35, 1439 (1984). 57. W.E. Stone. Convulsant action of t e t r a z o l e d e r i v a t i v e s . Pharma-cology, 3, 367 (1970). 58. E.G. Gross and R.M. Featherstone. Studies w i t h t e t r a z o l e d e r i v -atives. I. Some pharmacologic properties of a l i p h a t i c substituted pentaraethylene t e t r a z o l e d e r i v a t i v e s . J . Pharmacol. Exp. Ther., 87, 291 (1946). 59. E.G. Gross and R.M. Featherstone. V. Some pharmacologic properties of aminophenyl t e t r a z o l e s . J . Pharmacol. Exp. Ther., 92, 330 (1948). 60. I . Ahmad and J.S. Shunkla. Studies on the neuropharmacological action of some newer substituted tetrazoles and quinazolones. Ind. J . P h y s i o l . Pharmacol. 26, 289 (1982). 61. H.W. Vohland and W. Koransky. Effect of a-hexachlorocyclohexane on metabolism and e x c r e t i o n of p e n t y l e n e t e t r a z o l e i n the r a t . N.S. Arch. Pharmacol., 273, 99 (1972). 62. D.W. E s p l i n and D.M. Woodbury. The f a t e and e x c r e t i o n of C^-labeled pentylenetetrazole i n the r a t , with comments on a n a l y t i c a l methods for pentylenetetrazole. J . Pharmacol. Exp. Ther. 118, 129 (1956). 63. J . J . B r i n k , H. C a r i g l i a , D.G. S t e i n and L.A. Galipeau. Uptake of [l^C]pentylenetetrazole by developing r a t b r a i n . Brain Res., 19, 445 (1970). 64. S.G. Rowles, G.S. Born, H.T. R u s s e l l , W.V. K e s l l e r and J.E. Christian. B i o l o g i c a l disposition of pentylenetetrazole-10-C i n r a t s and humans. J . Pharm. S c i . , 60, 725 (1971). 65. J.T. Stewart and J.L. Story. In v i t r o binding of pentylenetetrazol to plasma p r o t e i n s . J . Pharm. S c i . , 61, 652 (1972). 66. R.S. Burton. On the propyl derivatives and decomposition products of. e t h y l a c e t o a c e t a t e . Am. Chem. J . , 3, 383 (1982). 67. Z.L. Chang. Sodium valproate and v a l p r o i c a c i d . In " A n a l y t i c a l p r o f i l e s of drug substances", V o l . 8, K. F l o r e y , ed., Academic Pr e s s , New York, p.529 (1979). 68. M.A. Taylor and L.H. P r i n c e n . Fatty acids i n s o l u t i o n . In "Fatty acids". American O i l Chemists Society, I l l i n o i s , p.195 (1979). 207 69. W. Loscher and H.H. Frey. Kinetics of penetration of common a n t i -e p i l e p t i c drugs i n t o c e r e b r o s p i n a l f l u i d . E p i l e p s i a , 25, 346 (1984). 70. W. Loscher and H. Esenwein. Pharmacokinetics of sodium valproate i n dog and mouse. Arzneim. Forsch„ 28, 782 (1978). 71. D. Johnston. V a l p r o i c a c i d : Update on i t s mechanism of a c t i o n . E p i l e p s i a , 25 (Sup p l . l ) , SI (1984). 72. R.W. Kerwin and P.V. Toberner Minireview: The mechanism of action of sodium valproate. Gen. Pharmacol., 12,71 (1981). 73. R. Haas, D.A. Stumpf, J.K. Parks and L. Eguren. I n h i b i t o r y e f f e c t s of sodium valproate on oxidative phosphorylation. Neurology, 31, 1473 (1981). 74. R.L. Macdonald and G.K. Bergey. V a l p r o i c a c i d augments GABA-mediated postsynaptic i n h i b i t i o n i n c u l t u r e d mammalian neurons. Br a i n Res., 170, 558 (1979). 75. Y. Godin, L. Heiner, J . Mark and P. Mandel. E f f e c t s of d i p r o p y l -a c e t a t e , an anticonvulsant compound, on GABA metabolism. J . Neurochem., 16, 869 (1969). 76. S. S i m l e r , L. G i e s i e l s k i , M. M a i t r e , H. Randrianarisoa and P. Mandel. E f f e c t s of sodium d i p r o p y l a c e t a t e on audiogenic s e i z u r e s and b r a i n y - a m i n o b u t y r i c a c i d l e v e l s . Biochem. Pharmacol., 22, 1701 (1973). 77. W. Loscher. Valproate,induced changes i n GABA metabolism at the subcellular l e v e l . Biochem. Pharmacol., 30, 1364 (1981). 78. A.S. Abdul-Ghani, J . Continho-Netto, D. Druce and H.F. Bradford. E f f e c t s of ant i c o n v u l s a n t s on the i n vivo and i n v i t r o release of GABA. Biochem. Pharmacol., 30, 363 (1981). 79. S.R. Whittle and A.J. Turner. Effects of the anticonvulsant sodium valproate on aminobutyrate and aldehyde metabolism i n ox brain. J . Neurochem., 31, 1453 (1978). 80. J.W. Van Der Laan, T. De Boer and J . B r u i n v e l s . D ipropylacetate and GABA degradation. P r e f e r e n t i a l i n h i b i t i o n of s u c c i n i c semi-aldehyde dehydrogenase and in d i r e c t i n h i b i t i o n of GABA-transminase. J . Neurochem., 32, 1769 (1979). 81. J.P. Gent and N.I. P h i l i p s . Sodium d i p r o p y l a c e t a t e (valproate) p o t e n t i a t e s responses to GABA and muscimol on s i n g l e c e n t r a l neurons. B r a i n Res., 197 (1980). 82. R.W. Kerwin, H.R. Olpe and M. S c h u l t z . The e f f e c t of sodium d i -propylacetate on gamma-aminobutyric acid dependent i n h i b i t i o n i n the rat cortex and substantia nigra i n r e l a t i o n to i t s anticonvuls-ant a c t i v i t y . Br. J . Pharmacol., 71, 545 (1980). 208 83. N.L. Harrison and M.A. Simmonds. Sodium valproate enhances res-ponses to GABA receptor activation only at high concentrations. Brain Res., 250, 201 (1982). 84. W. Loscher. Effect of i n h i b i t o r s of GABA transmission on the synthesis, binding, uptake and metabolism of GABA. J . Neurochem., 34, 1603 (1980). 85. W.M. Burnhanu L. Spero, M.M. Okasaki and B.K. Madras. Saturable binding of H-phenytoin to rat brain membrane f r a c t i o n . Can. J . P h y s i o l . Pharmacol., 59, 402 (1981). 86. P.J. Schecter, Y. Tranier and J . Grove. Effect of dipropylacetate on amino acid concentrations i n mouse brain: correlations with anticonvulsant a c t i v i t y . J . Neurochem., 31, 1325 (1978). 87. A. Lespagnol, J . Mercier, F. Erb-Deruyne and S. Dessaigne. Sur les proprietes neurosedatives et antiepileptiques d'acides organiques a chaines ramifiees. Ann. Pharm. Fr., 30, 193 (1972). 88. K.R. Scott, J.A. Moore, I.B. Lalusky, J.M. Nicholson, A.M. Lee and C.N. Hinko. Spiro [4,5] and Spiro [4,6] carboxylic acids: c y c l i c analogues of valproic a c i d . Synthesis and anticonvulsant eval-uation. J . Med. Chem., 28, 413 (1985). 89. M.A. Spielman, A.0. Geiszler and W.J. Close. Anticonvulsant drugs. Some Acylureas. J . Amer. Chem. S o c , 70, 4189 (1948). 90. F.M. Berger and B.J. Ludwig. The anticonvulsant action of 2,2-diethyl 1,3-propanediol and some of i t s homologues and esters. J . Pharmacol., Exp. Ther., 99,27 (1951). 91. H.F. Schwartz, L.F. Worrell and J.N. Delgado. Synthesis of some disubstituted cyanoacetamides as potential anticonvulsants. J . Pharm. S c i . , 56, 80 (1967). 92. W. Loscher. Anticonvulsant a c t i v i t y of metabolites of valproic a c i d . Arch. I n t . Pharmacodyn., 249, 58 (1981). 93. A. Acheampong, F. Abbott and R. Burton. I d e n t i f i c a t i o n of valproic acid metabolites i n human serum and urine using hexadeuterated valproic acid and GCMS analysis. Biomed. Mass Spectrom., 10, 586 (1983). 94. J.A. Vida and E.H. Gerry i n "Anticonvulsants", Academic Press, New York, p.151 (1977). 95. M.A. Davis, S.0. Winthrop, R.A. Thomas, F. Herr, M. Charest and R.0. Gandry. Anticonvulsants. 11. Spiro compounds Dibenzo [a,d] cycloheptadiene-5,5'-hydantoins,-5,5'-oxazolidinediones and -5,2'-succinimides. J . Med. Chem., 7, 439 (1964). 96. L.H. Sternbach i n "The Benzodiazepines", S. Ganattini, E. Mussini and L.O. Randall, eds., Raven Press, New York, p.l (1973). 209 97. W.E. Klunk, D.F. Corey and J.A. F e r r e n d e l l i . Comparison of epileptogenic properties of unsubstituted and B-alkyl substituted gamma-butyrolactones. Mol. Pharmacol., 22, 431 (1982). 98. T.C. Somers. New analeptics and hypnotics related to the barbit-urate antagonist, 'Bemigride'. Nature, 178, 996 (1956). 99. P.R. Andrews and G.P. Jones. Convulsant and anticonvulsant barbit-urates. Molecular o r b i t a l c a l c u l a t i o n s . Eur. J . Med. Chem., 16, 139 (1981). 100. G.D. Davies, R.B. Belshee and W.R. Anderson. Solution conformations of ethyl-1'-methylbutylbarbituric a c i d . Implications for drug receptor s i t e interactions. Mol. Pharmacol., 11, 470 (1975). 101. P. Seeman. The membrane actions of anesthetics and t r a n q u i l i z e r s . Pharmacol. Rev., 24, 583 (1972). 102. R.L. Macdonald and M.J. McLean. C e l l u l a r basis of barbiturates and phenytoin anticonvulsant drug action. E p i l e p s i a , 23 (Suppl. 1), S7 (1983). 103. R.F. Squires and C. Braestrup. Benzodiazepine receptors i n rat brain. Nature, 266, 732 (1977). 104. R.F. Squires, J.E. Casida, M. Richardson and E. Saederup. [3- % ] - t -butylbicyclophosphorothionate binds with high a f f i n i t y to brain s p e c i f i c s i t e s coupled to y-aminobutyric acid-A and ion recognition s i t e s . Mol. Pharmacol., 23, 326 (1983). 105. J.A. Ferrendelli and S. Daniels-McQueen. Comparative actions of phenytoin and other anticonvulsant drugs on potassium and vera-t r i d i n e stimulated calcium uptake i n synaptosomes. J . Pharmacol. Exp. Ther., 220, 29 (1982). 106. R.J. DeLorenzo. Phenytoin: calcium- and calmodulin-dependent pro-t e i n p h o s p h o r y l a t i o n and n e u r o t r a n s m i t t e r r e l e a s e . I n " A n t i e p i l e p t i c Drugs. Mechanism of A c t i o n " . G. H. G l a s e r , J.K. Penry and D.M. Woodbury, Raven Press, New York, p.399 (1980). 107. R. T r i f u l e t t i , A.M. Snowman and S.H. Snyder. Barbiturate recog-n i t i o n s i t e on the GABA/Benzodiazepine receptor complex i s d i s t i n c t from the picrotoxin/TBPS recognition s i t e . Eur. J . Pharmacol., 106, 44 (1984). 108. M.J. Iadarola, R.F. F a n e l l i , J.0. McNamara and W.A. Wilson. Com-parison of the effects of diphenylbarbituric a c i d , phenobarbital, pentobarbital and secobarbital on GABA-mediated i n h i b i t i o n and benzodiazepine binding. J . Pharmacol. Exp. Ther., 232, 127 (1984). 109. U. Kotz and R.H. Antonin. Pharmacokinetics and b i o a v a i l a b i l i t y of sodium valproate. C l i n . Pharmacol. Ther., 21, 736 (1977). 210 110. W. Loscher. Serum protein binding and pharmacokinetics of v a l -proate i n man, dog, rat and mouse. J . Pharmacol. Exp. Ther., 204, 255 (1978). 111. R. Gugler and G. Mueller. Plasma protein binding of valproic acid i n healthy subjects and i n patients with renal disease. Br. J . C l i n . Pharmacol., 5, 441 (1978). 112. A. Kober, Y. Olsen and T. Sjoholm. Binding of drugs to human serum albumin: XIV. The theoretical basis for the interaction between phenytoin and valproate. Mol. Pharmacol., 18, 237 (1980). 113. A. Acheampong, F.S. A b b o t t , J.M. O r r , S.M. Ferguson and R.W. Burton. Use of hexadeuterated valproic acid and GCMS to determine the pharmacokinetics of valproic a c i d . J . Pharm. S c i . , 73, 489 (1984). 114. T.A. Bowdle, I.H. P a t e l , R.H. Levy and H.J. Wilensky. Valproic acid dosage and plasma protein binding and clearance. C l i n . Pharm-a c o l . Ther., 28, 486 (1980). 115. H. Nau and W. Loscher. Valproic acid and metabolites: Pharma-cological and tox i c o l o g i c a l studies. E p i l e p s i a , 25 (Suppl. 1), S14 (1984). 116. F. Schobben, E. Van der K l e i j n and T.B. Vree. Therapeutic monitor-ing of valproic a c i d . Ther. Drug Monit., 2, 61 (1980). 117. E. Mesdjian, L. C i e s i e l s k i , M. V a l l i , B. Bruguerolle, G. Jadot, P. Bouyard and P. Mandel. Sodium valproate: k i n e t i c p r o f i l e and effects on GABA levels i n various brain areas of the r a t . Prog. Neuro-Psychopharmacol. B i o l . Psychiat., 6, 223 (1982). 118. M.A. Goldberg and T. Todoroff. Brain binding of anticonvulsants: carbamazepine and valproic a c i d . Neurology, 30, 826 (1980). 119. M.L. Aly and A.A. Abdel-Latif. Studies on d i s t r i b u t i o n and metab-olism of valproate i n rat brain, l i v e r and kidney. Neurochem. Res., 5, 1231 (1980). 120. H.H. Frey and W. Loscher. Dis t r i b u t i o n of valproate across the interface between blood and cerebrospinal f l u i d . Neuropharmacol., 17, 637 (1978). 121. E.M Cornford. Blood-brain barrier permeability to anticonvulsant drugs. In "Metabolism of A n t i e p i l e p t i c Drugs", R.H. Levy, W.H. P i t l i c k , M. Eichelbaum and J . Meijer, eds., Raven Press, New York, p.129 (1984). 122. R.J.E. Vadja, G.A. Donnan, J . P h i l l i p s and P.F. Bladin. Human brain, plasma and cerebrospinal f l u i d concentrations of sodium v a l -proate after 72 hours of therapy. Neurology, 31, 486 (1981). 123. R. Gugler and G.E. Von Unruh. C l i n i c a l pharmacokinetics of v a l -proic a c i d . C l i n . Pharmacok., 4, 433 (1980). 211 124. W. Kochen, H.P. Sprunck, B. Tauscher and M. Klemens. Five doubly unsaturated metabolites of valproic acid i n urine and plasma of patients on valproic acid therapy. J . C l i n . Chem. C l i n . Biochem., 22, 309 (1984). 125. G.R. Granneraan, S.I. Wang, J.M. Machinist and J.W. Kesterson. As-pects of the metabolism of valproic a c i d . Xenobiotica, 14, 375 (1984). 126. D.L. Coulter and R.J. A l l e n . Pancreatitis associated with valproic acid therapy for epilepsy. Ann. Neurol., 7, 92 (1980). 127. J . Jaekens, P. Casear and L. C o r b e l l . Hyperammonemia and hyper-glycinemia during sodium valproate therapy. Lancet 2, 260 (1980). 128. D.A. Winfield, P.B. Benton and M.L.E. Espir . Sodium Valproate and thrombocytopenia. Br. Med. J . , 2, 981 (1976). 129. L.J. Wilmore, B.J. Wilder, J . Bruni and H.J. V i l l a r e a l . Effect of valproic acid on hepatic function. Neurology, 28, 961 (1978). 130. H.J. Zimmerman and K.G. Ishak. Valproate-induced hepatic injury: analysis of 23 f a t a l cases. Hepatology, 2, 591 (1982). 131. N. Gerber, R.G. Dickinson, R.C. Harland, R.K. Lynn, D. Houghton, J . I . Antonias and J.C. Schimschock. Reye-like syndrome associated with valproic acid therapy. J . Pediatr., 95, 142 (1979). 132. A.W. Rettenmeier, K.S. P r i c k e t t , W.R. Gordon, S.M. Bjorge, S.L. Chang, R.H. Levy and T.A. B a i l l i e . Studies on the biotrans-formation i n the perfused rat l i v e r of 2-propyl-4-pentenoic a c i d , a metabolite of the a n t i e p i l e p t i c drug, valproic a c i d . Evidence for the formation of chemically reactive intermediates. Drug Metab. Disp., 13, 81 (1985). 133. A. Leo, C. Hansch and D. E l k i n s . P a r t i t i o n Coefficients and th e i r uses. Chem. Rev., 71, 525 (1971). 134. R.W. Taft i n "Steric Effects i n Organic Chemistry", M.S. Newman, ed., Wiley, London, p.559 (1956). 135. CM. Lee and W.D. Kumler. The dipole moment and structure of the imide group. I . Five- and six-raembered c y c l i c imides, lactams and the carbamate group. J . Amer. Chem. S o c , 83, 4586 (1961). 136. R. Collander. The p a r t i t i o n of organic compounds between higher alcohols and water. Acta. Chem. Scand., 5, 774 (1951). 137. S.H. linger, J.R. Cook and J.S. Hollenberg. Simple procedure for determining octanol-aqueous p a r t i t i o n , d i s t r i b u t i o n and io n i z a t i o n c o e f f i c i e n t by reversed-phase HPLC. J . Pharm. S c i . , 67, 1364 (1978). 212 138. R.M. Carlson, R.E. Carlson and H.L. Kopperman. Determination of pa r t i t i o n c o efficient by l i q u i d chromatography. J . Chromatogr., 107, 210 (1975). 139. M.S. Mirrlees, S.J. Moulton, C.T. Murphy and P.J. Taylor. Direct measurement of octanol-water p a r t i t i o n c o e f f i c i e n t by HPLC. J . Med. Chem., 19, 615 (1976). 140. J.M. McCall. L i q u i d - l i q u i d p a r t i t i o n c o e f f i c i e n t s by high-pressure l i q u i d chromatography. J . Med. Chem., 18, 549 (1975). 141. M. Harnisch, H.S. Mockel and G. Schultze. Relationship between log P shake-flask values and capacity factor derived from reversed phase HPLC for n-alkylbenzenes and some OECD reference substances. J . Chromatogr., 282, 315 (1983). 142. R.E. Koopmans and R.F. Rekker. HPLC of alkylbenzenes as determined from octanol-water p a r t i t i o n c o e f f i c i e n t or calculated from hydro-phobic fragmental data and connectivity indices. J . Chromatogr., 285, 267 (1984). 143. S. Toon, T. Mayer and R.M. Rowland. LC determination of l i p o p h i l -i c i t y with application to a homologous series of barbiturates. J . Pharm. S c i . , 73, 625 (1984). 144. R. Kaliszan. Chromatography i n studies of quantitative structure-a c t i v i t y r elationships. J . Chromatogr., 220, 71 (1981). 145. M. d'Amboise and T. Hanai. Hydrophobicity and retention i n revers-ed phase l i q u i d chromatography. J . L i q . Chromatogr., 5, 229 (1982). 146. N. Tanaka and E.R. Thornton. Structural and isotopic effects i n hydrophobic binding measured by HPLC. A stable and highly precise method for hydrophobic interaction i n biomembranes. J . Amer. Chem. S o c , 99, 7300 (1977). 147. T.L. Hafkenscheid and E. Toralinson. C o r r e l a t i o n between alkane/water and octanol/water d i s t r i b u t i o n c o e f f i c i e n t . Isocratic reversed phase LC capacity factors of acids, bases and neutrals. I n t . J . Pharmaceutics, 16, 225 (1983). 148. C. Horvath, W. Melander and I . Molnar. Solvophobic interactions i n l i q u i d chromatography with nonpolar stationary phases. J . Chroma-togr., 125, 129 (1976). 149. C.R. Yonker, T.A. Zwier and M.F. Burke. Comparison of stationary phase formation i n reverse phase for methanol-water systems. J . Chromatogr., 241, 257 (1982). 150. C.R. Yonker, T.A. Zwier and M.F. Burke. Investigation of stat i o n -ary phase formation for RP-18 using various organic modifiers. J . Chromatogr., 241, 269 (1982). 213 151. L.R. Snyder and J . J . Kirkland. "Introduction to Modern Chromato-graphy". Wiley, New York, p.22 (1979). 152. C. Hansch and A.J. Leo. "Substituent Constants for Correlation Analysis i n Chemistry and Biology". J . Wiley, New York (1979). 153. R.F. Rekker. "The Hydrophobic Fragmental Constant". E l s e v i e r , Amsterdam (1977). 154. P.E. P f e f f e r , L.S. S i l b e r t and J.M. Chirinko. Alpha-anions of carboxylic acids I I . The formation and al k y l a t i o n of alpha-metal-ated a l i p h a t i c acids. J . Org. Chem., 37, 451 (1972). 155. G. T a i l l a n d i e r , J.L. Benoit-Guyod, C. Laruelle and A. Boucherle. Investigation i n the dipropylacetic acid s e r i e s , C8 and C9 branched chain ethylenic acids and amides. Arch. Pharm. (Weinheim), 310, 394 (1977). 156. N. Thoai. Isomerization of 2,5-dimethyl-3-propyl-2-pentyl-2,3-dihydrofuran. B u l l . S o c Chim., 2, 225 (1964). 157. T. R e i c h s t e i n , H.R Rosenberg and R. Eberhadt. Eine einfache methode zur gewinning gesattigter f e r t i a r e r carbonsauren. Helv-Chim. Acta., 18, 721 (1935). 158. S. Pawlenko. Carboxylic acids. Chem. Abstracts, 68, 2055 (1968). 159. K. Bott. Carboxylic acid synthesis with 1,1-dichloroethylene. Preparation of sec. and t e r t . a l k y l a c e t i c a c i d . Chem. Ber., 100, 978 (1967). 160. K.B. Wiberg and T.W. Hutton. The stereochemistry of the Wolff rearrangement. J . Amer. Chem. S o c , 78, 1640 (1956). 161. J.S. Mihina and R.M. Herbst. The reaction of n i t r i l e s with hydra-zoic a c i d . Synthesis of monosubstituted tetrazoles. J . Org. Chem., 15, 1082 (1950). 162. D. Preissman, J.H. Bryder and L. Pauling. The reactions of a n t i -serum homologous to the p-Azosuccinilate ion group. J . Amer. Chem. S o c , 70, 1352 (1948). 163. P.L. Creger. Metalated carboxylic acids I . A l k y l a t i o n . J . Amer. Chem. S o c , 89, 2500 (1967). 164. A.W.D. Avison. The application of lithium aluminium hydride to the preparation of some amino-alkanols. J . Appl. Chem., 1, 469 (1951). 165. N.S. Antonenko. Study of the molecular interaction of amic acids by IR spectroscopy. Zhur. Obskich. Khim., 35, 425 (1965). 166. W.G. Finnegan, R.A. Henry and R. Lofquist. An improved synthesis of 5-substituted tetrazoles. J . Amer. Chem. S o c , 80, 3908 (1958). 214 167. H. Behringer and K. Kohl. Uber tetrazole die synthese des 5-6-aminoathyl-tetrazoles und einiger tetrazolylmethyl derivative. Ber., 89, 2648 (1956). 168. H. Normant. Recherches sur les magnesien vinyliques V I I . Synthese d'aldehyde a',B-ethyleniques. Comptes Rendus, 240, 1435 (1955). 169. A. Kajikawa, M. Morasaki and N. Ikekawa. Exclusive y-coupling i n the aldol reaction of a8-unsaturated esters. Tett. L e t t . , 4135 (1975). 170. A.S. Kende and B.H. Toder. Stereochemistry of deconjugative a l k y l -ation of ester dienolates. Stereospecific t o t a l synthesis of the l i t s e n o l i d s . J . Org. Chem., 47, 163 (1982). 171. M.W. Rathke and D. S u l l i v a n . The preparation and reaction of enolate anions derived from aB-unsaturated esters. Tett. L e t t . , 4249 (1972). 172. P.E. Pf e f f e r , L.S. Si l b e r t and E. K i n s e l . A re-examination of unsaturated carboxylate dianion reactions. Evidence for a- and y -substitution i n alkenoic acids. Tett. L e t t . , 1163 (1973). 173. P.E. Pfeffer and L.S. S i l b e r t . y-Anions. IV. Position and stereo-chemical isomerization of 2- and 3- unsaturated carboxylic acid dianions. J . Org. Chem., 36, 3290 (1971). 174. R.R. Rando and W. von E. Doering. 6,y-unsaturated acid and esters by photochemical isomerization of Y,B-congeners. J . Org. Chem., 33,1671 (1968). 175. R.M. Duhaime, D.A. Lombardo, I.A. Skinner and A.C. Weedon. Con-version of a,6-unsaturated esters to their 8,y-unsaturated isomers by photochemical deconjugation. J . Org. Chem., 50, 873 (1985). 176. J.A. Barve, F.D. Gunstone, F.R. Jacobsberg and P. Winlow. Fatty A c i d s . Behaviour of a l l the methyl o c t a d e c e n o a t e s and octadecynoates i n argentation chromatography and gas l i q u i d chromatography. Chem. Phys. L i p i d s , 18, 117 (1972). 177. H.B.S. Conacher. Chromatographic determination of cis-trans mono-ethylenic unsaturation i n fats and o i l s - A review. J . Chromatogr. S c i . , 14, 405 (1976). 178. M.S.F. Lie Ken J i e and C H . Lam. The thin-layer chromatographic behaviour of a l l the c i s , c i s and trans,trans - dimethylene i n t e r -rupted methyl octadecadienoates and methyl octadecadiynoates. J . Chromatogr., 124, 147 (1976). 179. K.C. Chan, R.A. Jewell, W.H. Nutting and H. Rapoport. Synthesis and stereochemical assignment of c i s and trans 2-methyl-2-pentenoic acid and the corresponding esters, aldehydes and alcohols. J . Org. Chem., 33, 3382 (1968). 215 180. M. Anteneunis, A. De Bruyn, H. De Proter and G. Verhegge. Cis and trans-isomers of ethyl heptenoates. « B u l l . Soc. Chim., 11, 371 (1968). 181. G.E. Berendsen, P.J. Schoenmakers, L. De Galan, G. Bigh, Z. Varga-Puchony and J . Inczedy. On the determination of the hold-up time i n reversed phase l i q u i d chromatography. J . L i q . Chromatogr., 3, 1669 (1980). 182. M.S. Tute. Pri n c i p l e s and practice of Hansch analysis: A guide to structu r e - a c t i v i t y correlation for the medicinal chemist. Adv. Drug Res., 6, 1 (1971). 183. M.F. Aldersley, A.J. Kirby, P.W. Lancaster, R.S. McDonald and C.R. Smith. Intramolecular c a t a l y s i s of amide hydrolysis by the carboxyl group. Rate-determining proton transfer from external general acids i n the hydrolysis of substituted maleamic acids. J . Chem. S o c , Perkin I I , 1487 (1974). 184. R. Kluger and J . Chin. Carboxylic acid participation i n amide hydrolysis. Evidence that separation of a non-bonded complex can be rate determining. J . Amer. Chem. S o c , 104, 2891 (1982). 185. A. Albert and E.P. Serjeant. "The determination of ionization constants". Chapman and H a l l , London (1971). 186. R.F. Cookson. The determination of a c i d i t y constants. Chem. Rev., 74,5 (1974). 187. L.Z. Benet and J.E. Goyan. Potentiometric determination of d i s -sociation constants. J . Pharm. S c i . , 56, 665 (1967). 188. J.F.J. Dippy. Chemical constitution and the dissociation constants of monocarboxylic acids. Part X. Saturated a l i p h a t i c acids. J . Chem. S o c , 1222 (1938). 189. A.E. Hershey, J.R. Patton and K.H. Dudley. Gas chromatographic method for the determination of valproic acid i n human plasma. Ther. Drug Monit., 1, 217 (1979). 190. S.G. Piredda, J.H. Woodhead and E.A. Swinyard. Effect of stimulus intensity on the p r o f i l e of anticonvulsant action of phenytoin, ethosuximide and valproate. J . Pharmacol. Exp. Ther., 232, 741 (1985). 191. W.D. Yonekawa, H.J. Kupferberg and D.M Woodbury. Relationship between pentylenetetrazole-induced seizures and brain pentylene-t e t r a z o l e l e v e l s i n mice. J . Pharmacol. Exp. Ther., 214, 59 (1980). 192. J.T. L i t c h f i e l d and F. Wilcoxon. A sim p l i f i e d method of evaluating dose-effect experiments. J . Pharmacol., Exp. Ther., 96, 99 (1949). 193. J.K. Seydel and K.J. Schaper. Structure-pharmacokinetic r e l a t i o n -ship and drug design. Pharmacol. Ther., 15, 131 (1982). 216 194. S. Toon and M. Rowland. Structure-pharmacokinetic relationships among the barbiturates i n the r a t . J . Pharmacol. Exp. Ther., 225, 752 (1983). 195. P.N. Craig. Interdependence between physical properties and selec-ti o n of substituent groups for correlation studies. J . Med. Chera., 14, 680 (1971). 196. B.B. Brodie, H. Kurtz and L.S. Schanker. The importance of d i s -sociation constant and l i p i d - s o l u b i l i t y i n influencing the passage of drugs into the CSF. J . Pharmacol. Exp. Ther., 130, 120 (1960). 197. Y.C. M a r t i n i n " P h y s i c a l Chemical P r o p e r t i e s of Drugs", 5. H. Yalkowsky, A.A. Sinkula and S.C. Valvani, eds., Marcel Dekker, p.49 (1980). 198. A.L. McClellan. "Tables of Experimental Dipole Moments:, V o l . 1. Freeman and Co., San Francisco (1963). 199. N.A. Brown, A.G.E. Wilson and J.W. Bridges. Chain-length dependen-cy of fatty acid and carbamate binding to serum albumin. Biochem. Pharmacol., 31, 4019 (1982). 200. A. Raines, G.J. Blake, B. Richardson and M.B. G i l b e r t . Differen-t i a l s e l e c t i v i t y of several barbiturates on experimental seizures and neurotoxicity i n the mouse. Ep i l e p s i a , 20, 105 (1979). 201. F.F. B l i c k e and M.F. Z i e n t r y . Acid amides as h y p n o t i c s . IV. Barbituric acids. J . Amer. Chem. S o c , 63, 2891 (1984). 202. E.S. Echague and R.K.S. Lim. Anticonvulsant a c t i v i t y of some carbinylureas. J . Pharmacol. Exp. Ther., 138, 224 (1962). 203. C.A. M i l l e r and L.M. Long. Anticonvulsants. An investigation of N-R-a-RT-a-Phenylsuccinimides. J . Amer. Chem. S o c , 73, 4895 (1951).1 204. D.R. Curtis and J.C. Watkins. The excitation and depression of spinal neurons by st r u c t u r a l l y related amino acids. J . Neurochem., 6, 117 (1960). 205. B. Ferrandes, C. Cohen-Addad, J.L. Benoit-Guyod and P. Eymard. Etude des r e l a t i o n entre l a structure c r i s t a l l i n e et l ' a c t i v i t e biologique de molecules de series d i - et t r i - a l c o y l acetiques. Biochem. Pharmacol., 23, 3363 (1974). 206. C. Cohen-Addad, G. d'Assenza, G. Ta i l l a n d i e r and J.L. Benoit-Guyod. Structures c r i s t a l l i n e s de derives des acides dipropylace"tiques et tripropylacetique. 111. Diethylpropionamide and dipropylpropion-amide. Acta Cryst., 31, 835 (1975). 207. P.R. Andrews, D.J. Craik and J.L. Martin. Function group contrib-u t i o n s to drug-receptor i n t e r a c t i o n s . J . Med. Chem., 27,1648 (1984). 217 208. N.C. B a e n z i g e r , A.D. N e l s o n , A. T u l i n s k y , J.H. B l o o r and A.I. Popov. Two independent determinations of the c r y s t a l and molecular s t r u c t u r e s of the iodine monochloride complex of pentylenetetrazole. J . Amer. Chem. S o c , 89, 6463 (1967). 218 APPENDIX NMR spectra of some of the synthesized compounds. 219 H-NMR ( 400 MHz) spectrum of isomeric mixture of ethyl 2-(Z-l'-propenyl)-E-2 pentenoate and ethyl 2-(E-l'-propenyl)-E-Z-pentenoate. H-NMR (400 MHz) magnified s E-2-pentenoate and ethyl 2-( pectrum of isomeric mixture of ethyl 2-(Z-l'-propenyl)-E-l'-propenyl)-E-2-pentenoate. 5 4 3 2 1 0 6(ppm) H-NMR (400 MHz) spectrum of ethyl 2-(Z-l'-propenyl)-Z-3-pentenoate with trace amount of ethyl 2-(Z-l'-propenyl)-E-2-pentenoate. r S3 4> 1 i -A. 6(ppm) *H-NMR ( 400 MHz) spectrum of 2-(Z-l'-propenyl )-E-2-pentenoic acid with trace amount of 2-(E-l'-propenyl)-Z-2-pentenoic acid. H-NMR ( 400 MHz) spectrum of isomeric mixture of 2-(Z-l'-propenyl)-E-2-pentenoic acid, 2-(E-l'-propenyl)-E-2-pentenoic acid, 2-(E-l'-propenyl)-Z-2-pentenoic acid, and 2-(Z-l1-propenyl)-Z-3-pentenoic acid. 227 H-NMR( 80 MHz) spectrum of N,N-diethylsuccinamic acid from the same synthesized product shown in Figure 6a. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096534/manifest

Comment

Related Items