Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies towards a catalytic antibody for an intramolecular diels-alder reaction Yuyitung, Gay M. 1998

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

Item Metadata

Download

Media
831-ubc_1999-39011X.pdf [ 7.35MB ]
Metadata
JSON: 831-1.0059625.json
JSON-LD: 831-1.0059625-ld.json
RDF/XML (Pretty): 831-1.0059625-rdf.xml
RDF/JSON: 831-1.0059625-rdf.json
Turtle: 831-1.0059625-turtle.txt
N-Triples: 831-1.0059625-rdf-ntriples.txt
Original Record: 831-1.0059625-source.json
Full Text
831-1.0059625-fulltext.txt
Citation
831-1.0059625.ris

Full Text

STUDIES TOWARDS A CATALYTIC ANTIBODY FOR AN INTRAMOLECULAR DIELS-ALDER REACTION by Gay M. Yuyitung B.Sc. University of Western Ontario, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1998 © G a y M. Yuyitung, 1998 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 writ ten permission. Department of Cy\t\4 fyuSTVlJ The University of British Columbia Vancouver, Canada Date IX'CeVnW \M" l ^ S DE-6 (2788) II ABSTRACT Hapten 6 was proposed as the transition-state analogue for both an intramolecular Diels-Alder reaction (13b to 46) and an ester hydrolysis (46 to 45). The key step in the synthesis of the hapten was the intramolecular Diels Alder reaction of (1 E,7£)-1-diphenoxyphosphoryl-1,7,9-decatriene which resulted in a 2:1 mixture of isomeric cycloadducts. The stereochemistry of these isomers was determined by comparison of the 1 3 C NMR spectra of the reduced cycloadducts to those of 1-methyl decahydronapthalenes. The c/s-fused Diels-Alder adduct, which is formed via the exo transition state, was formed as the major isomer. Mice immunised with KLH6 led to the generation of high affinity antibodies to 6 as determined by a particle coated fluorescent immunoassay (PCFIA). Further binding studies showed that one monoclonal antibody, F325, had an equilibrium dissociation constant, K D , estimated to be 8.8 x 10"6 M using an indirect competition method with detection by an enzyme-linked immunosrbent assay (ELISA). Kinetic studies of the hydrolysis reaction of 13b showed that F325 followed Michaelis-Menten kinetics with a K m of 265 ± 32 u.M and v m a x of 0.13 ± 0.007 uMmin in the presence of 1.0 u,M F325. The activity of F325 was not affected by the general esterase inhibitor SBTI indicating that the hydrolysis reaction was being catalysed by the antibody and was not due to the presence of contaminating esterases. Studies on substrate specificity of F325 suggested that the conjugated diene moiety was important for the observed catalytic behaviour of the antibody. However, neither the cyclic p-nitrophenyl ester of the IMDA product nor the acid could be detected by an HPLC assay. iv TABLE OF CONTENTS Abstract ii Table of Contents iv List of Figures x List of Schemes xiii List of Tables xv List of Abbreviations xvi Acknowledgements xix 1 INTRODUCTION 1 1.1 INTRODUCTION TO DIELS-ALDER REACTIONS 1 1.1.1 Historical Aspects 3 1.1.2 Mechanism 4 1.1.3 Stereochemical Aspects 6 1.1.4 Diels-Alder Catalysts 8 1.2 INTRODUCTION TO ANTIBODIES 10 1.2.1 The Immune System 10 1.2.2 Antibodies 11 1.2.3 Hybridoma Technology 13 1.2.4 Antigens 15 1.2.5 Antibody-Antigen Interactions 15 1.2.6 Calculation of Binding Constants 19 1.3 INTRODUCTION TO CATALYTIC ANTIBODIES 22 1.3.1 Enzymes 22 1.3.2 Transition-State Theory and Catalysis 23 V 1.3.3 Enzyme Mimics 26 1.3.4 Polyclonal Antibodies 28 1.3.5 Monoclonal Antibodies 29 1.3.6 Diels-Alder Abzyme Examples 33 1.4 RESEARCH PROJECT 36 2 RESULTS AND DISCUSSION 40 2.1 SYNTHESIS OF HAPTEN AND SUBSTRATES 40 2.1.1 Synthesis of Phosphonate Precursor of the Hapten 44 2.1.2 Stereochemistry of the Phosphonate IMDA 46 2.1.3 Synthesis of Hapten-Linker Compound 51 2.1.4 Coupling of Hapten to Proteins 54 2.1.5 Chemical Method 54 2.1.6 Photochemical Method 57 2.1.7 Synthesis of Triene Substrates 60 2.1.8 Synthesis of Diene Substrate 63 2.1.9 Stereochemical Outcome of the IMDA 65 2.2 RABBIT STUDIES 69 2.2.1 Immunisations 69 2.2.2 Immune Response 69 2.2.3 Calculation of the Equilibrium Dissociation Constant 77 2.2.4 Relative Affinity Determination by Chaotropic Elution 81 2.2.5 Test for Catalytic Activity 83 2.2.6 Summary of Rabbit Antibodies Study 85 2.3 MONOCLONAL ANTIBODIES 86 2.3.1 Immune Response 86 vi 2.3.2 Binding Assays 88 2.3.3 Characterisation of Monoclonal Antibodies 90 2.3.4 Large Scale Production of the Monoclonal Antibody 91 2.3.5 Preliminary Catalytic Studies 91 2.3.6 Substrate Specificity of F325 94 2.3.7 Isolation and Determination of the Product from Kinetic runs 97 2.3.8 Michaelis-Menten Kinetics 99 2.4 DISCUSSION 100 2.4.1 Transition-State Theory Analysis 100 2.4.2 Hydrolytic Abzymes 101 2.4.3 Diels-Alder Abzymes 102 2.5 FUTURE WORK 103 2.5.1 Hapten Design 103 2.5.2 Screening for Kinetic Activity 105 2.6 SUMMARY AND CONCLUSION 106 3 EXPERIMENTAL 108 3.1 GENERAL CHEMICAL METHODS 108 3.1.1 Methyl (2E,8E)-2,8,10-Undecatrienoate (4a), Methyl (2E,8Z)-2,8,10-Undecatrienoate (4b), Methyl (2Z )8E)-2,8,10-Undecatrienoate (4c), and Methyl (2Z,8Z)-2,8,10-Undecatrienoate (4d) 110 3.1.2 Methyl 1,2 I3A4aa,5,6,8aa-Octahydronapthalene-5a-carboxylate (5a), Methyl 1 ^.S^^ap^ASaa-Octahydronapthalene-sp-carboxylate (5b) 112 3.1.3 Methyl 1,2,3,4,4aa l5 I6,8aa-Octahydronapthalene-5|^c&rboxylate (5c), Methyl 1,2,3,4 (4a|3,5,6,8aa-Octahydronapthalene-5a-carboxylate (5d) 113 3.1.4 (5-[N-Glutaric Acid-(hydroxy-p-aminophenoxyphosphoryl)]-1 ^ .S^^a^ASa-octahydronapthalene (6) 115 3.1.5 (6£)-6,8-Nonadienal (1 Oa) 116 3.1.6 (62)-6,8-Nonadienal (10b) 117 3.1.7 Ethyl (2E,8E)-2,8,10-Undecatrienoate (12a), Ethyl (2Z,8E)-2,8,10-Undecatrienoate (12c) 117 3.1.8 Ethyl (2E,8Z)-2,8,10-Undecatrienoate (12b) 119 3.1.9 p-Nitrophenyl (2E,8E)-2,8,10-Undecatrienoate (13a) 120 3.1.10 p-Nitrophenyl (2E )8Z)-2,8,10-Undecatrienoate (13b) 121 3.1.11 p-Nitrophenyl (2Z,8E)-2,8,10-Undecatrienoate (13c), p-Nitrophenyl (2Z )8Z)-2,8,10-Undecatrienoate (13d) 122 3.1.12 7'-Oxy-4'-methylcoumaryl (2E,8£)-2,8,10-Undecatrienoate (14) 123 3.1.13 Diphenyl Triphenylphosphoranylidenemethylphosphonate (15) 124 3.1.14 (1E,7E)-1-Diphenoxyphosphoryl-1,7,9-decatriene (16a), (1Z,7E)-1-Diphenoxyphosphoryl-l,7,9-decatriene (16b) 126 3.1.15 Ethyl (2E)-8-Hydroxy-2-octenoate (17) 128 3.1.16 Ethyl (2E)-8-Tetrahydorpyranyloxy-2-octenoate (18) 129 3.1.17 (2E)-8-Tetrahydropyranyloxy-2-octen-1-ol (19) 131 3.1.18 (2E)-8-Tetrahydropyranyloxy-2-octenal (20) 132 3.1.19 (3£)-9-Tetrahydropyranyloxy-1,3-nonadiene (21a) 133 3.1.20 (3Z)-9-Tetrahydropyranyloxy-1,3-nonadiene (21b) 135 3.1.21 (6£)-6,8-Nonadien-1 -ol (22a) 136 3.1.22 (6Z)-6,8-Nonadien-1 -ol (22b) 137 3.1.23 6-Tetrahydropyranyloxy-hexan-1-ol (23) 138 3.1.24 6-Tetrahydropyranyloxy-hexanal (24) 139 3.1.25 5-Diphenoxyphosphoryl-1 ^ .S^^a^ASa-octahydronapthalene (25a/25b) 140 3.1.26 1 -Diphenoxyphosphoryl cis-Decahydronapthalene (26a) 142 3.1.27 1 -Diphenoxyphosphoryl trans-Decahydronapthalene (26b) 143 3.1.28 Benzyl Ester of 4'-Hydroxyglutarinilic Acid (29) 144 VIII 3.1.29 5-Methoxyphenoxyphosphoryl-1 ^.S^^a.S.e.Sa-octahydronapthalene (30), 5-Dimethoxyphosphoryl-1,2,3,4,43,5,6,8a-octahydronapthalene (31) 145 3.1.30 5-Hydroxymethoxyphosphoryl-1,2,3,4,4a,5,6,8a-octahydronapthalene (32) 147 3.1.31 5-[Benzyl N-Glutarate-(methoxy-p-aminophenoxyphosphoryl)]-1,2,3,4,4a,5,6,8a-octahydronapthalene (33) 149 3.1.32 5-[Benzyl N-Glutarate-(hydroxy-p-aminophenoxyphosphoryl)]-I^.S^^a.S.e.Sa-octahydronapthalene (34) 150 3.1.33 5-Hydroxyphenoxyphosphoryl-1 ^.S^^a.S.e.Sa-octahydronapthalene (35) 151 3.1.34 (2E,8E)-2 I8,10-Undecatrienoic Acid (36a) 153 3.1.35 (2E,8Z)-2,8,10-Undecatrienoic Acid (36b) 154 3.1.36 (2Z,8E)-2,8,10-Undecatrienoic Acid (36c), (2Z,8Z)-2,8,10-Undecatrienoic Acid (36d) 154 3.1.37 8-Tetrahydropyranyloxy-octan-1 -ol (37) 156 3.1.38 8-Tetrahydropyranyloxy-octanal (38) 157 3.1.39 9-Tetrahydropyranyloxy-1-nonene (39) 158 3.1.40 8-Nonen-1 -ol (40) 159 3.1.41 8-Nonenal (41) 160 3.1.42 Ethyl (2E)-2,10-Undecadienoate (42) 161 3.1.43 (2£)-2,10-Undecadienoic Acid (43) 162 3.1.44 p-Nitrophenyl (2E)-2,10-Undecadienoate (44) 163 3.1.45 1,2,3,4,4a,5,6,8a-Octahydronapthalene-5-carboxylic acid (45) 164 3.1.46 p-Nitrophenyl 1,2,3,4,4aa,5,6,8aa-Octahydronapthalene-5a-carboxylate (46a), p-Nitrophenyl 1 ^.S^^ap^ASaa-Octahydronapthalene-sp-carboxylate (46b) 165 3.1.47 7'-Oxy-4'-methylcoumaryl 1,2,3,4,4aa,5,6,8aa-Octahydronapthalene-5ct-carboxylate (47a), 7'-Oxy-4'-methylcoumaryl 1,2,3,4,4ap\5,6,8aa-Octahydronapthalene-5p-carboxylate (47b) 166 ix 3.2 GENERAL BIOLOGICAL METHODS 167 3.2.1 Preparation of Protein-Hapten Conjugates 169 3.2.2 Immunisations of Rabbits 171 3.2.3 Polyclonal Antibody Purification.... 172 3.2.4 Screening Assay with an ELISA 173 3.2.5 Immunisations of Mice 175 3.2.6 Hybridoma Technology 176 3.2.7 Selection and Expansion of Hybridomas 178 3.2.8 Freezing Cell Lines 179 3.2.9 Ascites Production 180 3.2.10 Monoclonal Antibody Purification 180 3.3 KINETIC ASSAYS 181 3.3.1 Rabbits 181 3.3.2 Mice 182 References 183 Spectral Appendix 191 X LIST OF FIGURES Figure 1. Retrosynthetic analysis of compactin (1b) and structurally similar derivatives (from ref 4) 2 Figure 2. Stereochemical outcome of the exo and endo transition states for the IMDA reaction of (3E,9E)-1,3,9-decatriene and (3E,9Z)-1,3,9-decatriene. 6 Figure 3. General structure of an antibody 13 Figure 4. The process of hybridoma technology to make monclonal antibodies 14 Figure 5. General procedure for an ELISA or PCFIA binding assay 18 Figure 6. Energy diagram for an uncatalysed versus an enzyme-catalysed reaction for the reaction of substrate (S) to product (P) 23 Figure 7. Thermodynamic cycle for reaction of a single substrate in the presence of an enzyme 25 Figure 8. One of the first successful abzymes reported was for an ester hydrolysis, with Km = 1.9 u.M, kcat = 0.027 s"1, and IWkun = 960 (from ref 50) 31 Figure 9. A Diels-Alder abzyme with K m = 21 mM for the dienophile and an apparent kcat of 4.3 min"1 (from ref 64) 34 Figure 10. A Diels-Alder abzyme with a K m = 1130 u,M and 740 u.M, for diene and dienophile respectively, and kcat = 0.67 s"1 (from ref 65) 35 Figure 11. Dual activity observed from an abzyme, with K m = 8.3 mM for the dienophile and kcat = 0.055 s"1 for the cycloaddition reaction (from ref 67). : 36 Figure 12. The transformation of 7 to 9 may be catalysed by antibodies raised against hapten 6 38 Figure 13. Possible conformations for (a) c/s-26a and (b) frans-26b decalins, where R = P(0)(OPh) 2 50 Figure 14. Change in serum titre overtime, a typical immune response (from ref 20). 70 Figure 15. Binding assays for purified antibodies from rabbits (a) H54 and (b) H55. Antibodies were titrated against BSA6 (squares) or BSA alone (circles). 72 Figure 16. Binding assays for purified antibodies from rabbits (a) H60 and (b) H61. Antibodies were obtained from immunisation with either KLH35 (open xi markers) or KLH6 (closed markers), and were titrated against BSA6 (squares) or BSA alone (circles) 74 Figure 17. Indirect competition coupled with ELISA detection of antibodies from rabbits H54 (diamonds), H55 (squares), H60 (circles), and H61 (triangles). 80 Figure 18. Klotz plot for polyclonal antibodies from rabbit H55 from competitive inhibition with 35. The equilibrium dissociation constant K D was calculated from the slope of the graph to be 7.5 x 10"7 M 81 Figure 19. Antibody affinity profile measured by thiocyanate elution followed by ELISA detection for polyclonal antibodies from rabbits H54 (diamonds), H55 (squares), H60 (circles), H61 (triangles), and H56 (no marker) 82 Figure 20. HPLC trace of PNP triene substrate 13a and PNP IMDA adduct 46 on a Novapak 4p analytical silica column with 4% ethyl acetate in hexanes. Detection of products was carried out by UV absorbance at X, = 205 nm 84 Figure 21. Typical results of PCFIA to test for specific binding to BSA6 (open markers) over BSA (closed markers). The first test bleed (squares) is lower in both the serum titre and hapten affinity relative to the second test bleed (circles) 86 Figure 22. A Klotz plot of the binding of 35 to F325 from absorbance readings of a competitive inhibition assay with ELISA detection. The K D of the impure antibodies was calculated from the slope of the graph 88 Figure 23. Affinity profiles for monoclonal antibodies F325 (squares), F336 (circles), and HIL3 (diamonds) from a chaotropic elution assay using ammonium thiocyanate 90 Figure 24. Comparison of the activity of monoclonal antibodies F325 and HIL3 with substrate 13b. The release of p-nitrophenolate anion was detected at 400 nm at 37 °C on a UNICAM UVA/isible Spectrophotometer. The assays were run in duplicates and gave plots similar to that shown above. The antibody was added to a solution of 630 pM of the PNP ester 13b in 10% DMSO, 0.18% Triton X-100 and 1 x PBS 92 Figure 25. The effect of general esterase inhibitor SBTI on the activity of antibody F325 and a lipase. The assays were run in duplicates at 37 °C, where 0.18 mg/mL lipase or 0.23 mg/mL F325 was added to samples containing 33 pg/mL SBTI, 630 pM substrate 13b, 10% DMSO and 0.18% Triton X-100 in 1 xPBS 93 Figure 26. Inhibition assays with hapten 35 for F325-catalysed hydrolysis of substrate 13b. The reactions were performed in duplicates in the reaction buffers XII described above and the averaged initial rates were plotted. The concentrations of inhibitor 35 were: (•) no inhibitor; (V) 0.10 mM; (0) 0.14 mM; (• ) 0.30 mM 94 Figure 27. Relative rates of hydrolysis of the isomers of p-nitrophenyl (2,8,10)-undecatrienoate in the presence of 0.88 pM F325. Background samples contained 0.88 pM HIL3 95 Figure 28. Comparison of the activity of 0.88 pM F325 with triene substrate 13b versus diene substrate 44. Background samples contained 0.88 pM HIL3. 96 Figure 29. HPLC trace of undecatrienoic acid isomers 36 and the acid adduct 45....98 Figure 30. Plot of initial rates versus varying concentrations of substrate 13b to determine the catalytic activity of 1.0 pM F325 100 XIII LIST OF SCHEMES Scheme 1. Transformation of aldehyde 10a to substrates 13a and 14 or hapten 6....40 Scheme 2. Synthesis of (E)-7,9-nonadienal (10a). Reaction conditions (yields) are as follows: (a) n-BuLi, Dibal-H, THF, -78 °C (70%); (b) DHP, pTsA, CH 2 CI 2 (79%); (c) Dibal-H, CH 2 CI 2 , -78 °C (95%); (d) (COCI)2, DMSO, Et 3N, CH 2 CI 2 , -78 °C (85% and 83% respectively); (e) Ph 3PCH 3Br, n-BuLi, THF, 0 °C (82%); (f) PPTs, EtOH, 70 °C (90%) 42 Scheme 3. Synthesis of (Z)-7,9-nonadienal (10b). Reaction conditions (yields) are as follows: (a) DHP, Dowex H + resin, toluene (95%); (b) (COCI)2, DMSO, Et 3N, CH 2 CI 2 , -78 °C (89% and 97% respectively); (c) t-BuLi, allyldiphenylphosphine, Ti(iPrO) 4, Mel (76%); (d) pTsA, MeOH (77%). ...43 Scheme 4. Reaction of aldehyde 10a with ylide 15 45 Scheme 5. Synthesis of 26a and 26b to determine the stereochemistry of the phosphonate IMDA adducts 47 Scheme 6. Coupling of the linker with aminophenol 52 Scheme 7. Synthesis of hapten-linker 6. Reaction conditions (yields) are as follows: (a) MeOH, n-BuLi, THF, -78 °C (51% of 30, 16% of 31); (b) 3 M KOH, THF-H 2 0, A (63%); (c) TMSBr, CH 2 CI 2 , A (58% and 77% respectively); (d) (COCI)2, then 29, Et 3N, CH 2 CI 2 , A (33%); (e) K 2 C0 3 , THF-H 2 0 (65%). 53 Scheme 8. Photochemical linking of 35 to KLH with sulfo-SANPAH 59 Scheme 9. Synthesis of substrates 13a and 14. Reaction conditions (yields) are as follows: (a) n-BuLi, THF, -78 °C (62% of 10a, 10% of 10c); (b) KOH, THF-H 2 0, A (85%); (c) p-nitrophenol, DCC, CH 2 CI 2 (70%); (d) 7-hydroxy-4-methylcoumarin, DCC, DMAP, CH 2 CI 2 (79%) 61 Scheme 10. Synthesis of substrate 13b. Reaction conditions (yields) are as follows: (a) n-BuLi, THF, -78 °C (76%); (b) KOH, THF-H 2 0, A (77%); (c) p-nitrophenol, DCC, CH 2 CI 2 (72%) 62 Scheme 11. Synthesis of substrates 13c and 13d. Reaction conditions (yields) are as follows: (a) 11b, n-BuLi, THF, 0 °C (85% overall yield for all isomers, isolation of 4c and 4d, which were carried out as a mixture for the next reactions); (b) KOH, THF-H 2 0, A (75%); (c) p-nitrophenol, DCC, CH 2 CI 2 (63%) 63 Scheme 12. Synthesis of substrate 44. Reaction conditions (yields) are as follows: (a) DHP, Dowex H + resin, toluene (98%); (b) (COCI)2, DMSO, Et 3N, xiv CH2CI2, -78 °C (90% and 75% respectively); (c) Ph 3PCH 3Br, n-BuLi, THF, 0 °C (83%); (d) pTsA, MeOH (77%); (e) 11a, n-BuLi, THF, -78 °C (87%); (f) KOH, THF-H 2 0, A (75%); (g) p-nitrophenol, DCC, CH 2 CI 2 (73%). 65 Scheme 13. Stereochemical outcome from the thermal IMDA of alkyl 2,8,10-undecatrienoates, where R = methyl (4 -> 5), PNP (13 -> 46), or coumaryl derivative (14 47) 68 XV LIST OF TABLES Table 1. Comparison of 1 3 C NMR chemical shifts (8 in ppm) for methyl decalins 27 and 28 (from ref 74) versus 26a and 26b. The assignment of the carbons was based on chemical shifts and from the observed 3 1 P - 1 3 C couplings (J in Hz) in the 1 3 C NMR spectra 48 Table 2. Components of various media and buffers used for biological studies. ..168 xvi LIST OF ABBREVIATIONS e molar extinction coefficient Ab antibody ABTS 2,2'-azinobis(3-ethylbenzothiazoline-6-sulphonic acid) diammonium salt Ag antigen Bn benzyl BSA bovine serum albumin Calcd calculated CDR complementary determining regions DCC dicyclohexyldicarbodiimide DCI desorption chemical ionisation DDW doubly distilled water DHP 3,4-dihyro-2H-pyran DIBAL diisobutylaluminum hydride DMAP 4,4-dimethylaminopyridine DMSO dimethyl sulfoxide EDC 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride El electron impact ELISA enzyme-linked immunosorbent assay EWG electron-withdrawing group FCS fetal calf serum FITC fluorescein isothiocyanate GC gas chromatography Hepes N-(2-hydroxyethyl)piperazine-N'-2-ethanesulfonic acid HOMO highest occupied molecular orbital HPLC high performance liquid chromatography HRMS high resolution mass spectrometry HRP horseradish peroxidase Ig immunoglobulin IMDA intramolecular Diels-Alder reaction J coupling constant kcat catalytic rate constant K D equilibrium dissociation constant KLH keyhole limpet hemocyanin Km Michaelis-Menten constant LRMS low resolution mass spectrometry LUMO lowest unoccupied molecular orbital m/z mass-to-charge ratio M + molecular ion Mab monoclonal antibody MRCRB mouse red cell remover buffer MW molecular weight OD optical density PBS phosphate buffered saline PCFIA particle concentration fluorescence immunoassay PNP p-nitrophenyl PPTs pyridinium p-toluene sulfonate Pristane 2,6,10,14-tetramethyldecanoic acid pTsA p-toluenesulfonic acid RIA radioimmunoassay SAM sheep anti-mouse SBTI soya bean trypsin inhibitor sulfo SANPAH sulfosuccinimidyl-6-(4'-azido-2'-nitrophenylamino)hexanoate THF tetrahydrofuran THP tetrahydropyran TLC thin layer chromatography Triton X-100 polyoxyethylene (9-10) p-f-octyl phenol xix ACKNOWLEDGEMENTS First and foremost, I would like to thank my supervisor Professor Larry Weiler for his guidance, support, and encouragement throughout the course of my studies. Many thanks also go to past and present members of the Weiler group for their friendship and useful discussions. In particular, special thanks are due to Dr. Tim Daynard and Dr. Anurag Sharadendu for answering my many questions and for their company during coffee breaks. In addition, the help of Dr. Michael Pungente and Ms. Jennifer Jamieson with this project are gratefully acknowledged. I would also like to thank Professor Hermann Ziltener for his guidance and the use of his laboratory facilities at the Biomedical Research Centre. The assistance of members of the Ziltener group is greatly appreciated, especially to Ms. Helen Merkens for her time and patience with the animal and tissue culture work. Special thanks go to Dr. Sarah Kerrigan for her assistance and advice with the photochemical linking techniques and binding assays. I appreciate the assistance of Professor Steve Withers and members of his laboratory for the use of their resources and helpful suggestions. XX DEDICATION To my parents and family, for all their encouragement and support over the years. 1 1 INTRODUCTION 1.1 INTRODUCTION TO DIELS-ALDER REACTIONS High cholesterol, which leads to various heart related ailments, remains a grossly undertreated condition in North America. It has been estimated that only 30 percent of the 13 million Americans with symptoms of heart disease are on cholesterol-reducing plans, while an additional 16 million with high cholesterol, but no symptoms of heart disease, remain untreated. In 1996, cholesterol-lowering drugs were among the fastest growing in the pharmaceutical industry, with sales of about three billion dollars. Because of this, a family of natural products known as the statins, which are potent cholesterol-lowering compounds, has attracted considerable interest from chemists and biochemists. Lovastatin (1a), also known as mevinolin, compactin (1b) and the dihydro derivatives 2a and 2b have been isolated from various fungi and are competitive inhibitors of an essential enzyme in cholesterol biosynthesis in humans. 1 , 2 Activity studies on 1a and 1b have shown that the lactone is hydrolysed in vivo to a (3-hydroxy acid, 3 and that the bicyclo[4.4.0] ring system is most likely important for binding to the enzyme. A retrosynthetic analysis of these compounds leads to an intramolecular Diels-Alder reaction (IMDA) of the appropriately substituted triene to generate these complex 2 ring systems (Figure 1). 4 In fact, Vederas and co-workers have suggested that the biosynthesis of 1a and related analogues, may involve an enzyme-catalysed IMDA. 5 6 Figure 1. Retrosynthetic analysis of compactin (1 b) and structurally similar derivatives (from ref 4). In addition, this bicyclo[4.4.0] ring system is also found in other biologically active or synthetically interesting targets, including a variety of steroids, terpenoids, and alkaloids. Some examples are shown below in which the ring systems may also be generated via an intramolecular Diels-Alder reaction. 3 HO. NMe O' HO" testosterone a steroid a-cadinene a terpenoid morphine an alkaloid 1.1.1 Historical Aspects The reaction of a 1,3-diene and an alkene to form a cyclohexene has been widely used by chemists for 70 years. The Diels-Alder reaction was named after the two Nobel prize-winning chemists who first reported this 4 + 2 cycloaddition. 7 In terms of atom efficiency in organic synthesis, the Diels-Alder reaction is an ideal reaction, as the only reagents required are the diene and dienophile. Furthermore, often no byproducts are formed in the Diels-Alder reaction. Another advantage of this method is its high regio- and stereoselectivity. These benefits are also observed in the intramolecular Diels-Alder reaction in which the diene and dienophile are tethered together in one molecule, resulting in a bicyclic product in one step. In fact, regioselectivity is generally higher in the IMDA reaction relative to its bimolecular counterpart due to steric constraints imposed by the tether between the diene and dienophile. These constraints may also lead to some ordering in the starting material, thereby lowering the activation energy so that milder conditions and increased reaction rates are observed. 4 Due in part to the lack of suitable synthetic routes to make the appropriate triene precursors, the first example of the deliberate use of an IMDA reaction in natural product synthesis did not appear until 1963. 8 The subsequent two decades witnessed an explosion in the use of the IMDA reaction 9 , 1 0 that provided significant contributions to the mechanism 1 1 and stereoselectivity1 2 of the reaction. 1.1.2 Mechanism The mechanism most frequently proposed for the IMDA reaction is taken from experimental data from the bimolecular process. Extensive studies on the Diels-Alder reaction have confirmed that it is an exothermic reaction with a product-like transition state (TS) that occurs late in the reaction profile. 1 1 The reaction proceeds through a concerted pathway, but may have asynchronous bond formation depending on the substituents on the double bonds. Deuterium labelling experiments in an IMDA reaction support a concerted mechanism and there is no evidence of a two-step mechanism. 1 2 However, the new sigma bonds may be formed at different rates depending on the coefficients of the frontier molecular orbitals of the diene and dienophile. The highly ordered TS gives rise to a large negative activation entropy that is typically in the range of -30 to -45 cal/mol/K for the bimolecular Diels-Alder. 1 1 However, due to the reduction in the degrees of freedom when the diene and dienophile occur in the same molecule, the activation entropy is more favourable in the intramolecular process. The activation entropy for these reactions is generally in the range of -15 to -25 cal/mol/K. 1 3 5 Experimentally, it has been found that the reaction of the unsubstituted decatriene is kinetically controlled, where the activation energy for formation of the frans-product is greater than that for the os-isomer by approximately 0.15 kcal/mol. 1 3 However, theoretical calculations show that the frans-product is 2 kcal/mol more stable relative to the c/s-adduct.1 4 Thus, product stability does not affect the stereocontrol for these IMDA reactions. H more stable Substitution of bulky, electron-donating groups or electron-withdrawing groups (EWG) on the diene and dienophile, respectively, has a rate-enhancing effect on the IMDA reaction. This effect is related to the decrease in energy between the interacting molecular orbitals of the diene and dienophile. Electron-donating substituents raise the energy of the highest occupied molecular orbital of the diene, while the energy of the lowest unoccupied molecular orbital of the dienophile is lowered by an EWG. 6 1.1.3 Stereochemical Aspects The cycloaddition of (3E,9E)-1,3,9-decatriene or (3E,9Z)-1,3,9-decatriene may proceed through two distinct TS in which the tether will preferentially adopt one of two chair conformations (Figure 2). In general, there is a bias for any substituents on the tether to occupy equatorial positions. The terms exo and endo refer to the arrangement of the EWG on the dienophile terminus relative to the diene in the TS. For (3E,9E> 1,3,9-decatrienes, the exo or endo TS leads to a cycloadduct that contains a cis- or transfused ring system, respectively. With the opposite stereochemistry about the dienophile as in the 9Z-isomer, the endo TS will give the c/s-fused adduct, while the exo TS will give the transfused cycloadduct. Figure 2. Stereochemical outcome of the exo and endo transition states for the IMDA reaction of (3E,9E)-1,3,9-decatriene and (3E,9Z)-1,3,9-decatriene. 7 Stabilisation of the TS, by secondary orbital interactions between the diene and the EWG on the dienophile, often gives rise to high stereoselectivity for the endo product. This selectivity, observed in many Diels-Alder reactions, was first noted by Alder and has been termed the Alder or endo rule. The ability to confidently predict the stereochemical outcome of the IMDA reaction is an important factor in making it a widely used synthetic tool to generate complex, polycyclic compounds. In the cycloaddition of decatriene 3, there are four new stereogenic centres in the cycloadduct (shown below). Therefore, disregarding any substituents on the tether, there are 16 (24) possible isomers from the IMDA reaction of triene 3. However, the nature and presence of substituents on the diene, dienophile, or the tether between them, as well as the initial stereochemistry of the triene, can result in high stereoselectivity in the reaction. Since the reaction proceeds via a concerted mechanism, the relative stereochemistry between C-9 and C-10 of triene 3 is retained in the final product, hence only eight of the sixteen isomers are possible. Furthermore, if the endo TS is favoured, then the number of isomers is further limited to four. Finally, steric interactions between any of the substituents may give rise to only one diasteriomer in the reaction. R 3 3 8 Low stereoselectivity is generally observed for reactions of unsubstituted or terminally substituted decatrienes that take place only at high temperatures. The product ratios in these reactions are independent of dienophile geometry and are not affected by secondary orbital interactions. However, high stereoselectivity is often found for cycloadditions of trienes containing bulky substituents on C-3 of the diene or on the tether, due to steric interactions in the chair TS . 1 2 The ring systems in many natural products are often highly substituted and the stereochemistry at those centres may be important for the biological activity of these compounds. Due to the possible control of the stereochemistry during the formation of the desired ring system, the IMDA reaction is often incorporated in the synthetic routes to these compounds. For example, the total synthesis of compactin (1b) followed from the retrosynthetic analysis shown in Figure 1, in which the key step involved an IMDA reaction to give the bicyclo[4.4.0] ring system, with the correct stereochemistry at C-4a, C-7, C-8, and C-8a. 4 Although model studies showed that the reaction should lead to the desired stereochemistry, cycloaddition of the appropriately substituted triene afforded a mixture of isomers. However, 28% of the desired frans-isomer with the correct stereochemistry at all the centres was isolated, illustrating the potential power and drawbacks of the IMDA reaction in natural product synthesis. 1.1.4 Diels-Alder Catalysts A number of catalysts have been found to increase the rates of the Diels-Alder cycloaddition and in certain cases they also provide enhanced stereoselectivity. The most commonly used catalysts are Lewis acids, such as BF 3, ZnCI 2, and EtAICl2. 9 Coordination of the catalyst to the EWG on the dienophile increases the electron withdrawing nature of such a substituent. This coordination leads to a decrease in energy in the lowest unoccupied molecular orbital of the dienophile. As a result a lower temperature is required and a preference for the endo TS is often noted. However, for terminally substituted decatrienes, attempts to use Lewis acid catalysts only led to polymerisation of the starting material. 1 5 More recently, research on macromolecule catalysts has been gaining more interest. Successful catalysis of the Diels-Alder reaction has been reported by using molecular imprinting techniques and catalytic antibodies, as well as preliminary results with RNA based catalysts. 1 6 Molecular imprinting involves the use of polymers to build a cavity complementary to the TS of the desired reaction. This technique has been used with some success to develop stereoselective or catalytic polymers for the bimolecular Diels-Alder react ion. 1 7 1 8 Although the existence of Diels-Alder enzymes has often been alluded to, 6 an isolated and fully characterised Diels-Alderase had not been reported until recently. 1 9 As a result the development of alternative protein based catalysts for this reaction, namely catalytic antibodies, has attracted a great deal of attention. 10 1.2 INTRODUCTION TO ANTIBODIES 1.2.1 The Immune System An animal's immune system is responsible for the protection of the body from foreign molecules. The detection and elimination of potentially harmful substances from the body is regulated by either a nonadaptive or an adaptive immune response. In nonadaptive immunity, macrophages, lacrimal cells and natural killer cells act nonselective^ and indiscriminately to remove foreign molecules. Adaptive immunity is controlled by the activity of specialised cells called lymphocytes that are capable of specific targeting, as well as enhanced activity upon repeated exposure to the foreign substance. Lymphocytes are blood cells produced in the bone marrow and further developed in specialised organs, such as the spleen, thymus, and lymph nodes in mammals; they circulate throughout the body and are promptly activated upon exposure to foreign invaders. Lymphocytes are divided into two broad categories, T cells which originate in the thymus and are involved in cell mediated immunity; and B cells which develop in a thymus independent fashion and mediate the humoral immune response. A cell mediated response involves the binding of cytotoxic T cells to cells infected with a pathogen and lysis of these cells. In humoral immunity, B cells secrete antibodies, which interact with foreign substances such as microorganisms and toxins. Phagocytes (e.g. macrophages) then eliminate the antibody coated foreign substances from the body. Any molecule that binds to an antibody is termed an antigen. 11 As part of the humoral immune response, a library of B lymphocytes circulates throughout the body, with each B cell (one clone) responsible for the production of one antibody (Ab) with a unique specificity. This initial library of lymphocytes has a predisposition of antigen specificity. Upon exposure to a foreign molecule, those B cells that bind specifically to the antigen are activated (clonal selection) to secrete antibodies and the cells divide rapidly (clonal expansion). As the immune response matures through genetic mutations and class switching, the specificity of the antibodies are 'fine-tuned' over time such that the naive B cells produce tightly binding antibodies. Once the pathogen is eliminated, clonal expansion and antibody production is discontinued, apart from some antigen specific B cells that are retained and converted to memory B cells. The initial antibody mediated immune response is slow; however, upon subsequent exposure to the same antigen, the immune system is able to mount a strong and rapid response to destroy the antigen due to this immunological memory. 1.2.2 A n t i b o d i e s Antibodies belong to a family of proteins known as immunoglobulins (Ig). There are five main classes of antibodies (termed isotypes), but all function by binding to antigens to aid in the immune response. Structurally, all antibody isotypes are characterised by four polypeptide chains with a combined molecular weight of 150 kDa. The two identical heavy chains (H) and two identical light chains (L) are connected by disulfide bridges to form a Y-shaped configuration (Figure 3). The antibody isotype is determined by the class of heavy chain, such that a, 8, £, Y and p heavy chains lead to IgA, IgD, IgE, IgG and IgM antibodies, respectively. 12 Primary immune response is generally characterised by the presence of IgM antibodies. These are pentameric and capable of binding up to ten antigens. Levels of IgM in serum are generally low (0.5-2 mg/mL); by contrast, IgG is the most abundant in mammalian systems (8-16 mg/mL). 2 0 These antibodies are monomeric, with only two identical antigen binding sites, and they are present during prolonged or repeated exposure to the antigen. The other isotypes are typically involved in protection of mucous membranes (IgA) or against parasite infections (IgE). These isotypes are found in lower concentrations in the body. Each polypeptide chain of an antibody is made up of a constant ( C ) and a variable (V) region. Amino acid sequencing of the constant regions of light chains (CL) shows that there are only two types, defined as K or X light chains. The constant regions of the heavy and light chains (CH and CL, respectively) make up the Fc (fragment that crystallises) and is responsible for regulation of the immune response. Association of the variable regions of one heavy and light chain (VH and VL) comprises one Fab (fragment containing the antigen binding site) domain. Although the variable regions on either polypeptide chain are made up of approximately 110 amino acid residues, only 20-30 of these are actually involved in the antigen binding site. The heterogeneity in amino acid sequences of this variable region leads to the large library of different active sites in antibodies available for binding to a variety of antigens. 13 Figure 3. General structure of an antibody. There are two identical Fab fragments (arms of the Y) for every one Fc region (trunk of the Y) in each antibody molecule. These regions are connected by the hinge region that allows for flexibility in binding to antigen. 1.2.3 Hybridoma Technology Injection of an immunogen into an animal gives rise to proliferation of B cell clones that have varying binding specificity to the foreign molecule. The heterogeneity of the mixture of antibodies produced is termed a polyclonal response. The ability to study only one of these clones (i.e. a homogeneous antibody preparation or a monoclonal antibody), was impossible until the advent of hybridoma technology. 2 1 Kohler and Milstein were able to successfully fuse myeloma cancer cells with B cells to achieve a hybrid cell that carries characteristics from both parents. The new hybridoma can be maintained in vitro while still producing large quantities of a specific antibody. 14 Figure 4. The process of hybridoma technology to produce monclonal antibodies: (a) Injection of an immunogen into a mouse results in an immune response. (b) A variety of B cells are activated by the presence of an immunogen and the serum from the mouse contains polyclonal antibodies. (c) The antibody producing B cells are isolated from the spleen of a mouse. (d) In vitro fusion of the various B cells with myeloma cells gives rise to hybrid cells between a B cell and a myeloma, known as a hybridoma. (e) Individual hybrid cell lines can be isolated and grown in vitro to provide large quantities of one specific (monoclonal) antibody. 15 1.2.4 Ant igens By definition, an antigen is a molecule that binds to an antibody, while an immunogen is a molecule that induces an immune response. However, the terms are often used interchangeably. The ability of a molecule to induce an immune response is most dependent on the size of the immunogen. Thus, small molecules cannot act as immunogens, but they can be coupled to a large carrier protein such as keyhole limpet hemocyanin (KLH) to form an immunoconjugate that may activate the immune system. Such small molecules that do not by themselves act as immunogens but are antigens are called haptens. The coupling of a hapten to a carrier protein such as KLH generally takes place via a linker arm. The linkers are typically bifunctional molecules with carboxy termini for coupling to amino groups of both the hapten and KLH through amide linkages. The linker acts not only to form the immunoconjugate but also to provide a spacer between the hapten and the larger protein so that the hapten is available for antibody recognition and binding. 1.2.5 Ant ibody-Ant igen Interactions Antibody-antigen binding interactions involve only a portion of the antigen, termed an epitope or antigenic determinant, and the antigen binding site of the antibody. The antigen binding site may also be referred to as the active site or complementary determining site (CDR) of the antibody. Binding interactions between the antibody and antigen are governed by noncovalent forces such as van der Waals forces, hydrogen bonding and electrostatic interactions within this pocket. The strength of the binding 16 energy or affinity of the Ab—Ag interaction therefore is dependent on these attractive forces between the two molecules. The serum of an immunised animal typically contains a mixture of both high and low affinity antibodies (i.e. a polyclonal mixture). However, during an immunisation period (repeated exposure to the same immunogen), the immune systems undergoes clonal selection followed by mutations (genetic manipulation of Ab) and class switching (e.g. from IgM to IgG), leading to the production of higher affinity antibodies. In the case of multivalent antigens (i.e. more than one antigenic site on a molecule), antibodies can bind with both antigen binding sites to the antigen and hence a stronger binding interaction is observed. The binding energy derived from this type of interaction is commonly called the avidity. However, due to their small size, haptens often cannot form more than one binding site to an antibody. The intrinsic affinity is a measure of the interaction between one hapten and one antibody binding site. A polyclonal mixture of antibodies or large antigens that contain more than one epitope will have different Ab-Ag interactions that are described as the functional affinity of the antibody. 2 2 Quantitative and qualitative measurements of Ab-Ag affinity can be determined using an immunoassay. Most commonly used is the enzyme linked immunosorbent assay (ELISA). An analogous system, the particle concentration fluorescence immunoassay (PCFIA) can also be used to measure the relative affinities of different antibodies. Both assays are examples of heterogeneous immunoassays, in that a separation step is performed to remove any unbound antibodies before the addition and incubation of the next immunoreagent. 17 A general procedure for either assay is shown in Figure 5 for a relative affinity study of different mouse antibodies for the same antigen. The antigen is attached by either passive adsorption or covalent linkage to a solid support, such as a polystyrene plate or bead (step a). Incubation with the antibody leads to what is assumed to be an irreversibly bound Ab-Ag complex (step b). A separation step (step c) is carried out to remove any unbound mouse antibodies. Indirect assays require the use of a secondary antibody that recognises the Fc domain of mouse antibodies. This anti-mouse Ab, labelled with either an enzyme (e.g. horseradish peroxidase, HRP) or a fluorescent tag (fluorescein isothiocyanate, FITC), is added to each well (step d). After an incubation period, any excess or unbound secondary Ab is removed with a final washing step (step e). In an ELISA, a substrate that can be cleaved by HRP to a coloured product is added. Finally, spectrophotometric detection of this chromophore or FITC (step r) gives a relative measurement of the antibody binding affinities. 18 ELISA ANTIGEN (a) 1. INCUBATE TO COAT AG IN WELLS 2. WASH TO REMOVE UNBOUND AG PCFIA / * ANTIGEN-BEAD I COMPLEX 1. COVALENT ATTACHMENT OF AG TO BEAD ADDITION OF ANTIBODIES TO BE STUDIED <»> -^ir^isr^r-1. INCUBATION TO FORM AB-AG COMPLEX 2. WASH TO REMOVE UNBOUND AB 1. ADDITION OF SECONDARY AB ENZYME LABELLED a-MOUSE Al I (d) ^ F H ^ L A B E L L E D a-MOUSE AB I 1. WASH TO REMOVE FREE SECONDARY AB 2. ADD ENZYME SPECIFIC SUBSTRATE (O) REACTION OF SUBSTRATE TO PRODUCT CATALYSED BY ENZYME DETECTION OF ABSORBANCE OFCHROMOPHORE(P) (f> DETECTION BY FLUORESCENCE OF FITC Figure 5. General procedure for an ELISA or PCFIA binding assay. 19 1.2.6 Calculation of Binding Constants The simple equilibrium reaction for the association of an antibody (Ab) and an antigen (Ag) to give a bound complex (Ab-Ag), is described by Equation 1. Ab + Ag Ab»Ag (1) kd When the system reaches equilibrium, the association constant (KA) for the reaction is defined as the ratio of the rate of association (ka) over the rate of dissociation (kd) for the complex. The value of K A can also be expressed in terms of concentration of each species involved and is typically in the range of 10 4 to 10 1 4 M" 1 . 2 3 For biochemical systems, an alternate equilibrium constant (KD) is often reported which describes the dissociation of the bound complex and is the inverse of K A (Equation 2) . 2 2 K A = ka / k d = [Ab-Ag] / [Ab][Ag] = 1 / K D (2) The K A can be related in thermodynamic terms as: -AG 0 = RTInKA = A (3) where G is the Gibbs free energy, R is the gas constant and T is the absolute temperature (K). Since the thermodynamic affinity (A) is equal to the negative free energy change of the reaction, then the K A can give a direct measure of the affinity. Antibodies with a large K A (or alternatively, a small K D ) have a high affinity for and are tightly bound to the antigen. 2 2 20 Experimentally, traditional measurements for the equilibrium constant, K D of an antibody involve dialysis, precipitation and filtration methods. 2 3 , 2 4 However, over the last 15 years, competitive indirect ELISA based assays have been used more frequently, due to their relative ease and the inexpensive cost of equipment and reagents. 2 5 , 2 6 The calculated equilibrium constant derived from these types of assays is usually in good agreement with more time intensive methods. 2 7 In general, this technique involves establishing an equilibrium between the antibody (at fixed concentration) and the antigen (at varying concentrations), followed by detection of the free antibody with bound antigen on a solid phase, as in a typical ELISA. The method outlined by Friguet et a l . 2 5 allows for the initial equilibrium reaction to take place in the solution phase, as opposed to interaction of the antibody with antigen bound to the solid phase, as in the assay developed by Goldberg et a l . 2 7 This is proposed to be an important factor since the solid phase may affect the equilibrium reaction. The method of Friguet et al. can be used to give the actual KD while the latter measurement often results in higher values for K D , but this method can be useful for determining relative affinities for various antibodies to the same antigen. The Friguet method can also be used for a solution containing impure antibodies (e.g. from hybridoma supernatant). 2 5 If a polyclonal antibody mixture is used, then there will be many different antibody-antigen interactions and the calculated equilibrium constant is the functional affinity of the antibody mixture. Due to the limits of sensitivity of ELISA detection, values of KD less than 10"9 M cannot be accurately determined with an indirect competition ELISA method. 21 An alternative method for the measurement of antibody affinities can be carried out by chaotropic elution followed by detection using an ELISA. This method has been used to give the affinity distributions of anti-dinitrophenol and anti-viral (rubella, measles, encephalitis) antibodies. 2 8" 3 0 Chaotropic agents that are commonly used in these immunoassays include thiocyanate or chloride ions, diethylamine, and urea. 3 1 The action of thiocyanate ions on antibody binding has been shown to be pH independent, 3 1 and is more likely due to its ability to affect the tertiary structure of proteins by disrupting hydrophobic forces. 3 2 The concentration of thiocyanate required to disrupt the antibody-antigen interaction is related to the strength of the attractive forces within the complex. High affinity antibodies (tight binders) are not as easily removed by ammonium thiocyanate relative to weak binders and so will require higher concentrations of the ion to break up the antibody-antigen complex. Thus, a value known as an affinity index may be calculated from this type of assay, and it can provide insight into the relative binding affinities for various antibody mixtures. 22 1.3 INTRODUCTION TO CATALYTIC ANTIBODIES The development of catalytic antibodies (also referred to as abzymes) provides an interesting example of how collaborative efforts between researchers in various disciplines, such as chemistry and immunology, have increased in recent years. Moreover, the following provides a brief chronological outline of how such basic research can lead to diverse applications. 1.3.1 Enzymes By definition, a catalyst is a compound that increases the rate of a reaction without itself being consumed. Catalysts are often used to increase the yield and rate of reactions, and to enhance selectivity. Desirable characteristics in a catalyst include high turnovers, specificity and large rate enhancements compared to the uncatalysed reaction. In terms of these factors, enzymes are examples of the most efficient catalysts known. An enzyme's ability to stabilise the transition-state (TS) structure of a substrate results in a rate acceleration of the reaction, which is also often highly stereoselective and stereospecific. This notion was first put forth by Pauling forty years ago, and is now the generally accepted view of the critical mode of action for enzymes. 3 3 A chemical reaction proceeds from substrate (S) to product (P) via an unstable, high energy structure, termed the transition state or activated complex (S*) as shown in Figure 6. Enzyme catalysis relies on the protein's ability to stabilise this activated complex, such 23 that the activation energy (Ea) of the enzyme reaction is lowered relative to that of the uncatalysed reaction. S* | E + P Reaction Coordinate Figure 6. Energy diagram for an uncatalysed versus an enzyme-catalysed reaction for the reaction of substrate (S) to product (P). 1.3.2 Transition-State Theory and Catalysis An enzyme-catalysed reaction is often a multistep process even in the case of conversion of a single substrate to product. Although there may be several association and dissociation steps along the reaction pathway, all enzymatic processes involve, at the very least, the binding of the enzyme to substrate to form an initial enzyme-substrate complex (ES) (Equation 4). The rate constants connected with the formation of this complex includes k a , which represents the formation of the complex, as well as k d 24 and kcat, which are associated with the breakdown of the complex to free E and S, or to P, respectively. E + S = ^ = ^ ES — E + P (4) kd The rate (v) for such an enzymatic process is described by the Michaelis-Menten equation (Equation 5). 1 3 3 The rate constant, kcat is often referred to as the catalytic constant of the enzyme and describes the maximum number of turnovers under defined conditions that the enzyme is capable of catalysing per unit of time. m W E K S ] [S] + K m (5) The Michaelis-Menten constant ( K M ) is an apparent dissociation constant for the breakdown of ES to either starting materials or product, and is a ratio of these rate constants (Equation 6). In addition, if the equilibrium for ES is the dominant process and not the breakdown to P, then kcat becomes negligible relative to kd . Under these conditions, K M is approximately equal to the equilibrium dissociation constant KD that was described in Section 1.2.6. K m = ^ + ^ 3 1 m kd_ = K d w n e n ^ > ; > ( 6 ) m k a k a Another important correlation is that at low substrate concentrations (less than Km), the rate equation is a simplified second order equation in both E and S as described by Equation 7. However, at saturating substrate concentrations (much 25 greater than Km), the rate depends only on the concentration of the enzyme (Equation 8). v = (MEP]) / K m for [ S ] « K m (7) v = ME] for [ S ] » K m (8) Since the rate of the enzyme process must be greater than that of the uncatalysed reaction, under high substrate conditions (Equation 8), kcat[E] > k u n[S] = v u n . Therefore, in order to observe catalysis, the enzyme must have a large kcat and/or a small K m . The relationship between rate constants and the equilibrium constants can also be illustrated by the thermodynamic cycle shown in Figure 7, and explained in terms of transition-state theory. E + S E + S* -»> E + P Kcai TS ES ^ = = ^ ES" • E + P Figure 7. Thermodynamic cycle for reaction of a single substrate in the presence of an enzyme. The upper pathway of the thermodynamic cycle describes the uncatalysed reaction of substrate to product, while the catalysed reaction is a multistep process, which involves binding of the enzyme to the substrate to form ES, as described earlier. Transition-state theory assumes that this complex is in equilibrium with the reactants, with a dissociation constant, KD . The transformation of ES to product goes through a 26 TS-enzyme complex (ES*) in an analogous manner to the transition state (S*) formed during the uncatalysed reaction. This transformation is assumed to be the rate-limiting step for the entire process. The thermodynamic box relates the four equilibrium constants as given by Equation 9. Furthermore, if the assumptions above are valid, then the equilibrium constants, K u n* and Kcat* that represent the formation of the transition-state complexes, S* and ES* can be equated to the experimentally determined rate constants, kun and kcat. K D / K T S = Kcat* / K u n* = kcat / k u n (9) Typically, enzyme systems are capable of rate enhancement in the range of 10 1 0 to 10 1 4 , as measured by the ratio of IWkun. 1.3.3 Enzyme Mimics Among his many contributions to our overall knowledge of science, Pauling was one of the first to note the striking similarities between the structure and function of antibodies and enzymes, as they are both large proteins with high specificity in binding to small molecules. 3 4 However, one major difference is that while antibodies display high affinity binding to stable, ground state molecules that they are associated with (antigens); enzymes bind tightly to an unstable, high energy activated complex (transition state). Jencks later hypothesised that if the activity of enzymes is due to the ability of the protein to induce the substrate to take on the shape and structure of the transition state, then antibodies with an active site complementary to a transition-state analogue (TSA) should show similar rate accelerations. 3 5 27 Experimentally, this can be accomplished by immunising an animal with an antigen that has been designed to simulate the proposed TS of a chemical reaction, to produce antibodies that will recognise and bind to the antigen. When exposed to a reaction substrate, the binding forces between the antibody and the substrate will result in a rate acceleration as the reactant takes on the character of the TS required for the chemical transformation. Crystal structures of hapten-antibody complexes have provided a closer look at the antibody active site and support the hypothesis that TS stabilisation is responsible for catalysis. 3 6" 3 8 The catalysis of acyl transfers by enzymes and catalytic antibodies has been one of the most extensively studied reactions. There is a high degree of similarity between the structures found for both serine proteases and the analogous abzyme. The X-ray structures of these hapten-antibody complexes, indicate that the activity of these abzymes are derived from their structure imparted by the hapten. 3 6 The smaller rate enhancements produced by abzymes relative to their enzyme counterparts have been attributed to hapten design flaws and a shorter evolutionary history for abzymes relative to their enzyme counterparts. In the case of the acyl transfer catalysed reactions, the high homology of amino acid sequences between catalytic antibodies and various enzymes suggests that the abzyme may be considered as a primitive serine protease. Despite the limitations of abzymes relative to enzymes in terms of activity, it should also be noted that abzymes provide distinct advantages, the least of which is the ability of abzymes to be selective for a reaction designed by us rather than by nature. 28 Abzymes have been reported to catalyse different reaction pathways or to yield products that are not formed by enzymatic or conventional chemical processes. 5 3 1.3.4 Polyclonal Antibodies Over the subsequent years, the ideas outlined by Pauling and Jencks have inspired others towards the development of abzymes. In one of the initial abzyme studies reported, Raso and Stollar 3 9 immunised rabbits with transaminase inhibitors. The polyclonal antibodies collected showed Schiff base formation occurred in the binding pocket of the protein, however no rate enhancement over background was observed. It was suggested that the rates were too low to be detected and the heterogeneity of the polyclonals in the sera complicated the analysis. These observations were noted in other early studies and only very low rate accelerations were observed if at a l l . 4 0 Some of these problems have been resolved by improvements in hapten design, as well as optimisation of immunisation and purification techniques. In the last decade, successful catalysis of hydrolysis and cycloaddition reactions by polyclonal abzymes have been repor ted. 4 1 4 3 Polyclonal abzymes are advantageous given their lower expense, ease of use and reduced time requirements for their preparation relative to the process to obtain mice monoclonal antibodies. Polyclonal antibodies have been used as fast, reproducible tools to test new hapten designs, 4 4 , 4 5 and to answer questions about the immune response of animals. Polyclonal abzymes have been studied in the hopes of optimising immunological conditions that can be applied to the generation of other abzymes. These studies have included the variability between catalytic activities 29 elicited in different individuals of the same species or different species; types of adjuvants, carrier proteins and linkages used in immunisations; and the correlation between titre or affinity and catalytic activity. 4 4" 4 8 However, the heterogeneity of polyclonal antibodies has long been noted as a limitation in the characterisation of these proteins. It is often assumed that the maximum concentration of specific antibodies in a polyclonal mixture is 10%. Stephens and Iverson 4 3 reported a rough estimate of 12% of the total IgGs in their polyclonal samples were either catalytic or had a high affinity for their hapten. However, there have also been reports of purified polyclonal antibody samples that exhibit a linear plot with respect to their catalytic behaviour, suggesting that the antibody sample may be more homogenous than has been previously thought . 2 9 4 2 1.3.5 Monoclonal Antibodies Using purified polyclonal antibodies can give an overview of the entire immune response and answer questions about a hapten's relative ability to induce catalytic antibodies. In contrast, monoclonal antibodies (MAbs) may be limited by the success of the fusion, and screening for catalytic activity tends to be constrained to only a handful of MAbs due to time restrictions. Nevertheless, the first successful and indeed most of the subsequent reports of successful abzymes have been attributed to the advent of hybridoma technology. The ability to produce large quantities of pure MAbs has reduced both the inability to detect low rates and interference due to contaminating enzymes that may be found in polyclonal mixtures. 30 In 1986, reports of two hydrolytic abzymes from the independent labs of Schultz and Lerner were the first truly successful abzymes. 4 9 , 5 0 Since acyl transferases have been so extensively studied, inhibitors for these proteins are well known and were proposed as stable TSA to be used as a hapten. 5 1 The phosphonate moiety has been a good mimic of the TS of an ester hydrolysis since they both share many common structural (tetrahedral geometry with comparable bond lengths) and electronic (negatively charged oxygen) features. Initial experiments showed that antibodies raised against phosphonate haptens were able to catalyse acyl transfer reactions with rate enhancements of 770 to 960 (Figure 8 ) . 4 9 , 5 0 While this catalytic activity is considerably less than that achieved by enzymes and less than what would be expected given the binding energy possible from antibodies, these reports demonstrated the generation of nonenzyme protein-based catalysts. In the following years a number of hydrolytic abzymes have been prepared showing improved rate enhancements using similar phosphonate haptens. 5 2 31 HAPTEN °YXP^CO2H HN F3C^N-^ H o, O - OH FoC Ab clone 6D4 Figure 8. One of the first successful abzyme reported was for an ester hydrolysis, with K m = 1.9 pM, kcat = 0.027 s ' \ and IWkun = 960 (from ref 50). Since these initial reports catalytic antibodies have been the subject of many reviews and they have been shown to catalyse a variety of transformations including: peptide bond formation, Claisen rearrangements, retro 2+2 cycloadditions, and a variety of other reactions. 5 3 , 5 4 The variability in the amino acid sequence of the antibody active site leads to the possibility of different binding and catalytic activity. In addition, the shape of the binding site can range from a deep pocket, to clefts and grooves, to flatter surfaces, depending on the antigen used to induce the immune response. 5 5 This 32 versatility, which is vital for normal antibody function within the immune system, is most likely responsible for the wide range of reactions that have been catalysed by abzymes. After the initial reports of successful abzymes, there was a great deal of speculation and hopeful predictions for this new technology. 5 6 Indeed there have been a great number of advances made in the field over the last 15 years. No where is this more evident than in the diversity of the possible applications that are being attempted and realised. The use of abzymes in organic synthesis has been demonstrated in peptide bond formation, 5 7 in the removal of protecting groups, 5 8 and in the hydrolysis of an enol ether which was an important step in the asymmetric synthesis of a natural product. 5 9 Furthermore, there is now an abzyme that is commercially available for use as a catalyst in aldol reactions. Catalytic antibodies are also being used in the design of novel therapeutic agents and in drug delivery systems. This includes research towards a treatment for cocaine overdose in which an abzyme has been shown to hydrolyse cocaine to a nontoxic form in vitro. 6 0 The delivery of chemotherapeutic drugs by antibody directed abzyme prodrug therapy (ADAPT) involves an antitumour antibody-abzyme conjugate and a prodrug (inactive form of the drug). 6 1 The conjugate is administered and due to the targeting ability of the antibody, accumulates at the site of the tumour. The prodrug is then injected and is activated by the abzyme, resulting in high concentrations of the toxic drug only at the tumour site. This approach is analogous to antibody directed enzyme prodrug therapy (ADEPT) which has been successfully used in cancer chemotherapy. 6 2 However, the main disadvantage in ADEPT is that the enzymes used are typically from 33 bacterial sources and hence are immunogenic themselves. Since abzymes can be "humanised", ADAPT should provide a less toxic approach. Another advantage in constructing de novo protein-based catalysts (i.e. artificial enzymes) is in catalysing reactions not catalysed by enzymes. Abzymes that have been generated for disfavoured reactions 6 3 and Diels-Alder reactions 6 4 for example have added a new dimension to the field of abzymes. 1.3.6 Diels-Alder Abzyme Examples The first reported Diels-Alder abzyme was an important step in the development of abzymes since it showed that this methodology could be used to obtain protein-based catalysts for which there were no known natural enzymes at the time. 6 The first example of a Diels-Alderase abzyme by Hilvert et a l . 6 4 was quickly followed by a report by Braisted and Schultz. 6 5 In both of these cases, the abzyme acted as an entropy trap by binding the two substrates to reduce the large negative activation entropy for the bimolecular Diels-Alder reaction, that is typically in the range of -30 to -45 cal/mol. 1 1 In addition to forcing the substrates in close proximity, the antibody active site may also force the substrates to adopt the high energy, boat geometry that has been suggested as the transition state for the Diels-Alder reaction, 6 6 which should lead to a rate enhancement of the reaction. The Diels-Alder reaction between tetrachlorothiophene 1,1-dioxide and N-ethyl maleimide results in the unstable, bicyclic intermediate, which quickly extrudes SO2 to give the final product (Figure 9). The hexachloro compound in Figure 9 was proposed as a stable TSA for this reaction. Antibodies raised against this hapten should have a 34 combining site that is complementary to the shape of the proposed TS. 6 4 Since the final product which has lost SO2 no longer resembles the TS, product inhibition would be reduced in this scheme. A catalytic antibody was isolated from mice immunised with this hapten and found to catalyse the rate of the desired Diels-Alder reaction by a factor of approximately 110 over the uncatalysed reaction. CI C S IP ? CI - (CH 2 ) 5 C 0 2 R CI c i T 0 HAPTEN CI o Vs*° A o r V o + Q CI o N—Et Ab clone p 1E9 ° * crV^ N-Et S 0 5 Figure 9. A Diels-Alder abzyme with K m = 21 mM for the dienophile and an apparent kcat of 4.3 min"1 (from ref 64). The approach taken by Braisted and Schultz for a Diels-Alder abzyme is more general in that it does not rely on a specific product transformation (i.e. the release of SO2) to reduce product inhibition. 6 5 The hapten was once again constrained to a boat TS by an ethano bridge of the bicyclo[2.2.2]octene system shown in Figure 10. The cycloadduct from the reaction of the diene and a maleimide in Figure 10 is expected to be adopt a lower energy conformation than that of the TSA. Thus the product would be released from the active site. A rate enhancement of up to 500 was observed for one of the antibodies raised against this hapten. 35 O Figure 10. A Diels-Alder abzyme with a K m = 1130 uM and 740 pM for diene and dienophile respectively and kcat = 0.67 s (from ref 65). The abzyme isolated by Suckling et a l . 6 7 also catalysed a similar Diels-Alder reaction to that of Braisted and Schultz, but differs from these previous examples in that there was a greater similarity between the hapten and product in this case (Figure 11). In addition, they noted a dual activity by the abzyme; first to transform the substrates into the DA adduct and second to hydrolyse the ester on the side chain. No product inhibition was observed, presumably in part due to the hydrolysis of the intermediate ester. 3 6 Figure 1 1 . Dual activity observed from an abzyme, with K m = 8 . 3 mM for the dienophile and kcat = 0 . 0 5 5 s"1 for the cycloaddition reaction (from ref 6 7 ) . In addition to these examples, there have been other catalytic antibodies that have exhibited Diels-Alderase activity for a hetero Diels-Alder reaction, as well as one that controls the formation of endo/exo products. 6 8 6 9 1.4 RESEARCH PROJECT Despite the high efficiency (yields greater than 8 5 % ) observed in the IMDA reaction of undecatrienoates such as 4c to give the 5-substituted octalins 5c and 5d, the reaction requires harsh conditions ( 1 5 0 °C and high pressure) and is not stereoselective ( 1 : 1 mixture of adducts formed). 1 5 Because bicyclic compounds such as 5c and 5d are found in many natural products and therapeutic agents, and since the reaction is unaffected by Lewis acid catalysis, 1 5 we wished to develop a catalytic antibody for this reaction. 37 C 0 2 M e 5 c 5 d 1:1 mixture of isomers Due to the late transition state for the IMDA reaction, a hapten that resembles the product is a likely candidate for the TSA. However, product inhibition would most likely decrease the efficiency of the abzyme, therefore, a further modification of the resulting product would be necessary. It was proposed that inclusion of a phosphonate in the design of the hapten might also induce estereolytic activity in the antibody. Thus, the active site of an antibody raised against hapten 6 should have charge and shape complementarity to the TSA of both an IMDA reaction and an ester hydrolysis. Such an active site could lead to antibody catalysis of the IMDA reaction of substrate 7, followed by hydrolysis of the ester 8 to give the bicyclic acid 9 as the final product. The difference in charge and shape of the final product relative to the hapten should be sufficient to allow for multiple turnovers of the abzyme. 38 Figure 12. The transformation of 7 to 9 may be catalysed by antibodies raised against hapten 6 . The generation of catalytic antibodies for such an IMDA reaction would require the combination of synthetic chemistry, immunology and biochemistry. Specifically, the procedure for the development and the study of these abzymes would involve: 1. Synthesis of the hapten and coupling with a carrier protein to produce an immunoconjugate. 2. Immunisation with the immunoconjugate into rabbits and mice to produce antibodies. 3. Isolation and/or determination of anti-hapten antibodies, and the study of the binding profiles of these antibodies. 4. Synthesis of substrates to test the catalytic behaviour of the hapten specific antibodies. 39 5. Michaelis-Menten treatment of kinetic data and the study of products from the antibody mediated reaction. 40 2 RESULTS AND DISCUSSION 2.1 SYNTHESIS OF HAPTEN AND SUBSTRATES In order to produce and study the activity of catalytic antibodies for an intramolecular Diels-Alder (IMDA) reaction, it was necessary to synthesise the hapten 6 and substrates such as the trienes 13a and 14. The synthetic routes to both these compounds were designed to originate from aldehdye 10a as shown in Scheme 1. O (EtO) 2P O OEt 11a CHO 10a O P h 3 P ^ P \ O P h 15 o p h OEt 12a 13a R = p-nitrophenol 14 R = coumarin 16a Scheme 1. Transformation of aldehyde 10a to substrates 13a and 14 or hapten 6. Reaction of 10a with either ylide 15 or phosphonate 11a would give the desired precursors for the hapten or substrate, respectively. This route was expected to give mainly the more stable E alkene in the newly formed double bond. In addition, the 41 IMDA reaction of vinyl phosphonates, such as 16a could be investigated with respect to their reactivity and their stereochemical outcome. Synthesis of the desired aldehyde 10a (Scheme 2) began with commercially available s-caprolactone which gave the a,p-unsaturated ester 17 in good yields (65-75%) with high E stereoselectivity.7 0 In one step, the caprolactone was partially reduced to the aldehyde and reacted in situ with the anion of 11a. Due to the slow reduction of lactones and a, p-unsaturated esters at low temperatures (-78 °C), little to no overreduction of the starting lactone or of the resulting ester product was observed. As was reported by Takacs et a l . 7 0 greater than 90% E alkene was formed when the ethyl derivative 11a was used. The stereochemistry of 17 was confirmed by analysis of its 1 H NMR spectrum, which gave a coupling constant for the protons of the vinyl bond as 15.6 Hz. In contrast, it was observed that trimethyl phosphonoacetate (11b) gave much lower or no stereoselectivity (50-70% E alkene) in the reaction. The alcohol 17 was protected as its THP ether 18 in order to differentiate it from the resulting allylic alcohol 19 formed during the subsequent reduction of the oc.p-unsaturated ester with Dibal-H. The alcohol 19 was oxidised to the aldehyde 20 under Swern conditions, followed by a Wittig reaction to produce the diene 21a. The THP ether was cleaved and the unprotected alcohol 22a was reacted in another Swern oxidation to give (E)-6,8-nonadienal (10a) as a very volatile oil. Characterisation of aldehyde 10a by 1 H NMR spectroscopy showed that the E stereochemistry about the diene was retained throughout the series of reactions. The 42 vinyl protons on C-6 and C-7 could be assigned to the multiplets at 5.79 ppm and 6.05 ppm, respectively, and were found to have a large vicinal coupling of 15 Hz. The 1 H NMR spectrum also gave the diagnostic aldehyde peak at 9.8 ppm, in addition to the carbonyl and C-H stretches in the IR spectrum (1732, 2725, 2850 cm"1) that are characteristic for an aldehyde. 22a 10a Scheme 2. Synthesis of (E)-7,9-nonadienal (10a). Reaction conditions (yields) are as follows: (a) n-BuLi, Dibal-H, THF, -78 °C (70%); (b) DHP, pTsA, CH 2 CI 2 (79%); (c) Dibal-H, CH 2CI 2 , -78 °C (95%); (d) (COCI) 2, DMSO, Et 3N, CH 2 CI 2 , -78 °C (85% and 83%, respectively); (e) Ph 3PCH 3Br, n-BuLi, THF, 0 °C (82%); (f) PPTs, EtOH, 70 °C (90%). 43 Alternatively, the Z-diene isomer 10b was also synthesised as shown in Scheme 3. Commercially available hexanediol was monoprotected as its THP ether using Dowex H + resin as the acid catalyst to give good yields of 23, with little to no diprotected diol produced. 7 1 The alcohol was oxidised under Swern conditions and the resulting aldehyde 24 was reacted with a titanium complex of allyldiphenylphosphine to give the Z-isomer 21b, as the only isolated product in 76% yield. 7 2 In the 1 H NMR spectrum of 21b, the vinyl proton on C-6 at 5.4 ppm was shifted upfield relative to isomer 21a (found at 5.65 ppm) and had a vicinal coupling of 10.7 Hz. This data was used to assign the Z-stereochemistry to 21b. Using similar reaction conditions as before, the THP ether was removed and the resulting alcohol was oxidised to the aldehyde 10b. Formation of the aldehyde was confirmed by the presence of the aldehyde signal at 9.8 ppm in its 1 H NMR spectrum. a, b .CHO 10b THPO' 23 R=CH 2OH 24 R=CHO c, d 21b R=THP 22b R=H Scheme 3. Synthesis of (Z)-7,9-nonadienal (10b). Reaction conditions (yields) are as follows: (a) DHP, Dowex H + resin, toluene (95%); (b) (COCI)2, DMSO, Et 3N, CH 2 CI 2 , -78 °C (89% and 97%, respectively); (c) t-BuLi, allyldiphenylphosphine, Ti(iPrO)4, Mel (76%); (d) pTsA, MeOH (77%). 44 2.1.1 Synthesis of Phosphonate Precursor of the Hapten The phosphonate triene precursor 16a was synthesised via a stabilised Wadsworth-Emmons type reaction between the ylide 15 with aldehyde 10a. The ylide 15 was synthesised as described by Jones et a l . 7 3 beginning with commercially available chloromethylphosphonic dichloride. Two equivalents of phenol displaced both chlorides on the phosphorus to give the diphenyl phosphonate derivative in greater than 90% yields. The isolated phosphonate was heated with triphenylphosphine to displace the chloride, followed immediately by treatment of the resulting phosphonate with base to form the desired ylide 15 in an overall yield of 68%. The aldehyde 10a was heated with 15 in toluene for 4 days to give a mixture of the vinyl phosphonates 16a and 16b in a 7:1 ratio of the E/Z isomers about the double bond between C-1 and C-2 (Scheme 4). The stereochemistry of the alkene was determined by 1 H NMR spectroscopy of the purified isomers. The vinyl proton at C-1 of isomer 16a at 5.85 ppm has a large coupling constant of 17.1 Hz that is typical of trans double bonds, relative to the corresponding proton in 16b which is found at 5.77 ppm has a smaller cis coupling of 12.9 Hz. The preferential formation of the more stable E-vinyl phosphonate 16a was anticipated, since the use of stabilised ylides tend to give rise to trans alkenes over the cis isomer. 7 3 45 CHO O I I P h 3 P ^ P \ O P h 10a 15 refluxing toluene OPh 170 oc sealed tube toluene 25 23% OPh 170 oc sealed tube toluene 25a cis fused adduct major product + 25b trans fused adduct minor product EXO PRODUCTS Ratio of exo:endo products 2:1 ENDO PRODUCTS 25d trans fused adduct major product 25c cis fused adduct minor product Scheme 4. Synthesis of bicyclic phosphonates 25. In addition, during the prolonged, high temperature conditions of the Wadsworth-Emmons reaction, some of the triene product proceeded to the give the IMDA adduct 46 25. Spectral data ( 1H and 3 1 P NMR) for the IMDA adducts formed during this reaction were in agreement with those for 25a and 25b, which were produced in the thermal cycloaddition of pure triene 16a. The isomeric cycloadducts 25c and 25d, resulting from the IMDA reaction of 16b were not produced during the reaction of 10a and 15. 2.1.2 Stereochemistry of the Phosphonate IMDA The triene phosphonate 16a was dissolved in toluene and heated in a sealed tube at 170 °C. Good yields (84-93%) of the desired Diels-Alder adducts 25a and 25b were obtained in a 2:1 ratio as determined by HPLC and 3 1 P NMR spectroscopy. In order to determine the stereochemistry of the ring fusion in these compounds, the two isomeric adducts were separated. Although isomers 25a and 25b could not be separated by normal flash chromatography, moderate resolution of the two isomers was achieved by HPLC. A small amount of the mixture was separated by HPLC to give relatively pure samples of the IMDA adducts for further characterisation. The double bond in the adducts 25a and 25b was hydrogenated to give the 1-substituted decalins 26a and 26b as shown in Scheme 5. 47 25a/b 2:1 mixture of isomers separation by HPLC 25a 26a major H 2 Pd/C EtOH 25b 26b minor Scheme 5. Synthesis of 26a and 26b to determine the stereochemistry of the phosphonate IMDA adducts. The 1 3 C NMR spectra of cis and trans decalins and their substituted derivatives have been recorded, and assignment of the chemical shifts have been made by Dalling et a l . 7 4 There appears to be a general trend in the 1 3 C chemical shifts for these decalins that may be used to distinguish between the cis- and frans-fused isomers. The stereochemistry of 26a or 26b was assigned by comparing their 1 3 C NMR spectra to those of the 1-methyl decahydronapthalene isomers 27 and 28 (Table 1). 7 4 48 27 28 Carbon 26a 26b 27 28 Assignment 8(J) 5(J) 5 5 1 33.1 (1J=155) 42.2 (1J=134) 37.2 38.4 2 29.2 (2J=9.4) 43.2 (2J=8.6) 29.5 37.1 3 26.4 (3J=37) 26.3 (3J=38) 27.4 26.8 4 28.9 (br s) 34.5 (4J=2.2) 25.8 35.4 5 36.2 (s) 33.8 (s) 33.6 35.3 6 21.7( 5J=9.5) 32.0( 5J=1.4) 21.9 27.5 7 27.7 (4J=3.4) 27.5 (4J=4.7) 27.4 27.2 8 23.8 (3J=28) 26.1 (3J=16) 20.0 31.0 9 43.4 (2J=0.8) 44.8 (2J=1.1) 43.0 50.6 10 34.7 (3J=32) 43.3 (3J=12) 38.7 44.0 Table 1. Comparison of 1 3 C NMR chemical shifts (6 in ppm) for methyl decalins 27 and 28 (from ref 74) versus 26a and 26b. The assignment of the carbons was based on chemical shifts and from the observed 3 1 P - 1 3 C couplings (J in Hz) in the 1 3 C NMR spectra. From the APT (attached proton test) spectrum of isomer 26b, it was concluded that the shifts at 42.2, 43.3 and 44.8 ppm, corresponded to methine carbons. The 1 3 C NMR spectra for isomers 26a and 26b were complicated by 3 1 P - 1 3 C couplings (J), which 49 due to long range connectivities, gave rise to doublets for the majority of the resonances. The large coupling ( 1J>100 Hz) was indicative of a one bond coupling between carbon and phosphorus and therefore was assigned to C-1. Typically, three bond C-P couplings are in the range of 15-40 Hz, and small couplings (less than 10 Hz) arise from two or four bond connectivities. Thus, the assignment of the 1 3 C NMR spectra for 26a and 26b as shown in Table 1 seemed reasonable. Comparison of the 1 3 C NMR spectra obtained for 26a and 26b, shows that the chemical shifts for C-2, C-4, C-6, and C-8 are downfield for isomer 26b relative to 26a. This downfield shift is a general trend observed for trans decalins versus the cis isomer. In addition, there is good agreement in chemical shifts between 26a and 27; and for 26b and 28. Therefore, 26a was assigned as the cis decalin structure, while 26b has the transfused decalin structure. Furthermore, it was observed that the 1 3 C NMR spectrum of the major isomer 26a simplified when run at higher temperatures (i.e. broad, unresolved or multiple peaks at room temperature were sharper at 49 °C). This would be expected from the c/'s-fused adduct that can interconvert between two chair-chair conformations (Figure 13a). 7 5 Inversion between the two forms of the unsubstituted cis decalin has a free energy of activation for ring interconversion of 10-15 kcal/mol. 7 6 At room temperature, the rate of interconversion is slow and the 1 3 C NMR resonances for the methylene carbons have broad line widths (24.6-26.8 Hz). 7 7 However, at 45.5 °C, the reported 1 3 C NMR spectrum for the unsubstituted cis decalin has narrower line widths (7-7.5 Hz) due to an increased rate of ring inversion. 7 7 50 The 1 3 C NMR spectrum of the minor isomer 26b at room temperature gave sharp signals as would be expected from the rigid conformation of the trans decalin (Figure 13b). Since the lowest energy chair-chair conformation adopted by the trans decalin does not undergo inversion, the sharp resonances observed in its 1 3 C NMR spectrum are unaffected by increasing temperatures. 7 7 Figure 13. Possible conformations for (a) c/s-26a and (b) trans-26b decalins, where R = P(0)(OPh) 2. The major compound 25a results from the exo transition state of the IMDA reaction. Although stabilisation of the transition state by overlap of the secondary orbitals generally leads to endo products, IMDA reactions carried out at high temperature are not greatly affected by this stabilisation and a 1:1 mixture of the two isomers are often formed. 1 5 In addition, steric interactions between the bulky phosphonate and the diene may give rise to a preference for the exo transition state, leading to the slight stereoselectivity observed in the reaction. 51 The Z-vinyl phosphonate 16b was also heated in a sealed tube to 170 °C. The yields of IMDA product for this reaction were lower with a greater recovery of starting triene phosphonate, which suggests that this reaction may have a higher activation energy in comparison to the trans isomer 16a. A mixture of cycloadducts 25c/25d were also obtained in a 2:1 mixture, similar to that found for the cycloaddition of 16a. Only the adducts 25a/25b were used in the following hapten synthesis leading to the antigen used for immunisations. 2.1.3 Synthesis of Hapten-Linker Compound Small molecules or haptens alone do not usually induce an antibody immune response. Thus, it is necessary to attach haptens to larger molecules, such as a protein in order to effect antibody production in the animal. In practice, this can be accomplished by coupling of the transition-state analogue to the carrier protein through a linker or spacer arm. Lowe et a l . 7 8 have reported that the optimal length of the linker arm for attachment of a hapten to beads for affinity chromatography is five or six carbons. A shorter or longer chain length was found to interfere with hapten recognition and binding by the antibody within the column. Assuming that in vitro and in vivo antibody-hapten interactions are similar, then this chain length should also be the optimal length for coupling of hapten to carrier protein for use in immunisations. One of the linkers used in our study was glutaric acid. It is five carbons in length, and can be used as a bifunctional linker by coupling the carboxylic acids with either an alcohol or an amine to make ester or amide linkages, respectively. 52 Glutaric anhydride was reacted with benzyl alcohol to give the mono benzyl protected linker (Scheme 6). Activation of the free carboxylic acid of the linker with DCC and subsequent reaction with aminophenol gave compound 29. The phenol moiety was thought to be important for antibody recognition of the substrate. HO' EtoN O O LINKER DCC DMAP 29 Scheme 6. Coupling of the linker with aminophenol. To complete the synthesis of the hapten-linker compound (Scheme 7), one of the phenoxy groups on the phosphonate esters 25a and 25b was displaced with methoxide to give a mixed phosphonate 30. A small amount of the dimethyl phosphonate 31 was also produced in this reaction. Treatment of 30 with a strong base to hydrolyse the more labile phenoxy group, or reaction of 31 with one equivalent of TMS-Br, produced the mono phosphonic acid 32. The acid was converted to its acid chloride with oxalyl chloride and reacted immediately with 29 to give the protected hapten 33. Deprotection of the methyl phosphonate with TMS-Br, followed by 53 hydrolysis of the benzyl ester using a mild base gave the desired hapten-linker compound 6. O O 33 O O 34 O O Scheme 7. Synthesis of hapten-linker 6. Reaction conditions (yields) are as follows: (a) MeOH, n-BuLi, THF, -78 °C (51% of 30, 16% of 31); (b) 3 M KOH, THF-H 2 0, A (63%); (c) TMSBr, CH 2CI 2 , A (58% and 77%, respectively); (d) (COCI)2, then 29, Et 3N, CH 2 CI 2 ) A (33%); (e) K 2 C0 3 , THF-H 2 0 (65%). 54 2.1.4 Coupling of Hapten to Proteins The final step in the formation of the immunoconjugate required the coupling of 6 to a carrier protein. Keyhole limpet hemocyanin (KLH), a respiratory protein isolated from a shellfish, Megathura crenulata, is a popular carrier protein due to its large size and highly immunogenic nature . 2 0 7 9 Covalent attachment of the protein to the hapten can be achieved by a variety of methods. 8 0 We investigated two different routes, namely a traditional chemical coupling using 1-ethyl-3-(3-dimethylaminopropyl)carbo-diimide hydrochloride (EDC), as well as a photochemical method involving the aryl azide of a bifunctional linker, sulfo-SANPAH. In addition to formation of the immunoconjugate KLH6 that would be used for immunisation into animals, an alternate hapten-protein conjugate was required for the assaying of anti-hapten antibodies. Therefore, a bovine serum albumin (BSA) conjugate was also produced. 2.1.5 Chemical Method The coupling of the hapten to the carrier protein KLH was carried out using a large excess of 6 relative to KLH. The carboxylic acid of 6 was activated with EDC and then reacted with the primary amines of the lysine residues on the protein, linking the acid 6 to the protein via amide bonds. A large excess of hapten relative to the protein is used because there are many sites for possible linkage on either protein. Amino acid analysis has shown that there are 4.6% (molar percent) primary amines from lysine 55 residues on KLH that are available for coupling with carboxylates 7 9 and approximately 30-35 reactive lysines on each BSA molecule. 2 0 It has been reported that an epitope ratio of 5-15:1 of hapten:KLH is ideal, since these epitope ratios give the strongest binding antibodies for small molecules such as LSD, dopamine and dinitrophenyl. 8 1 - 8 3 The epitope density is thought to be important in obtaining highly specific antibodies and in the maturation of the immune response. 8 2 There are many isoforms of KLH and the protein, with a molecular weight (MW) of 450 kDa for KLH-A, is difficult to characterise owing to its large size. In addition, KLH is found as a multimer and has a tendency to form larger aggregates with a MW of greater than 5 x 10 6 Da. 7 9 Our attempt to characterise the KLH6 immunoconjugate by mass spectrometry was unsuccessful due to aggregation which caused precipitation of the sample. Despite this precipitation during cross-linking, KLH is a popular carrier protein, as it is highly immunogenic. The hapten was also coupled with an alternate protein BSA, which was used for subsequent binding assays for the detection of antibodies that bind specifically to the hapten and not to the large KLH. This coupling was carried out in a similar manner as for KLH using a 62:1 molar ratio of hapten:BSA. 56 H 2 N — EDC PBS-MeOH NHR NHR -hNHR Hapten-Protein conjugate R = hapten 6 NHR NH 2 Since BSA is smaller, characterisation of this hapten-protein conjugate could be carried out by mass spectrometry (MS). Adamczyk et a l . 8 4 have reported the use of electrospray ionisation (ESI) MS techniques to determine the distribution of hapten-BSA conjugates formed during coupling. Under ESI-MS conditions, only the parent ion is generally observed and so can be used for the determination of the MW of the conjugated species without interference in the spectra due to fragmentation. Samples of BSA alone and the BSA6 conjugate mixture were introduced as solutions by liquid chromatography (HPLC) into the ESI-MS. The MW of BSA was estimated as 66,446 Da by this method, which is in good agreement with the calculated value of 66,432.9 Da, as determined from the amino acid sequence. 8 4 Separation of the conjugate mixture by HPLC gave low and high MW fractions. The first fraction 57 containing the low MW compounds showed the presence of free hapten (m/z = 422, M ++1). Analysis of the last fraction (m/z = 1200-2100) required conversion of the m/z spectra to the mass domain in order to determine the MW of the species present. The deconvoluted mass spectrum of the BSA6 conjugate suggests that a range of 3 to 18 hapten molecules were covalently attached to each molecule of BSA. 2.1.6 Photochemical Method An alternative approach to the conventional chemical EDC coupling described above was also attempted, following a similar procedure to that used in the successful generation of highly specific anti-LSD polyclonal antibodies. 8 1 In this method, a heterobifunctional linker, containing a succinimidyl ester and an aryl azide was used. Formation of an amide bond between the lysine residues of the protein with the activated ester of the linker is followed by photoactivation of the azide to give a reactive nitrene that can form covalent bonds to the hapten. In our case, commercially available sulfo-SANPAH was used as the linker to be coupled with the transition-state analogue 35 as shown in Scheme 8. The monophosphonic acid 35 was obtained by the base hydrolysis of the phosphonate isomers 25a/25b. The amide linkage between the linker and KLH was performed in the dark so that the azide would remain intact. The excess sulfo-SANPAH was removed by dialysis. Photochemical activation of a freeze-dried sample of KLH-SANPAH in the presence of 35, led to nitrene formation. The nitrene is a highly reactive species that will react rapidly, and nonspecifically, with either the hapten or other protein molecules. It was hoped that this approach would result in the random insertion of the nitrene into 35, 58 thereby providing a variety of epitopes of the hapten for improved antibody recognition. This is especially desirable for the dual activity required of our antibody in which active sites might be formed for both the bicyclic structure as well as the phosphonate moiety that are mimics for the IMDA and ester hydrolysis, respectively. Due to the lack of control in the nitrene insertion, no analysis of the epitope ratio was undertaken for this immunoconjugate KLH35. 59 — N H 2 sulfo-SANPAH H?N-dark reaction NH 2 NH 2 NO; O N 3 - •N ^ // H^fs ^ N + H NH 2 NH 2 hv=365 nm O H II ^ o 35 o PhO^ HO' NHR W V V V W ^ -NHR -NHR NHR NH 2 Scheme 8. Photochemical linking of 35 to KLH with sulfo-SANPAH. 60 2.1.7 Synthesis of Triene Substrates The aldehyde 10a was derivatised to give the p-nitrophenyl (PNP) or 7'-oxy-4'-methylcoumaryl (2E,8E)-2,8,10-undecatr ienoates (13a or 14) that were to be used as substrates for testing antibody catalytic activity (Scheme 9). The W a d s w o r t h - E m m o n s reaction of 10a with 11a proceeded to give ethyl (2E,8E)-2,8,10-undecatr ienoate (12a) with 8 6 % stereoselectivity, which was separable from isomer 12c. Determination of the stereochemistry about the newly formed double bond was based on the 1 H NMR spectra of the two isomers. The proton resonance at 6.9 ppm for 12a was assigned to the vinyl proton on C-3 and it had a coupling of 15 Hz consistent with the trans geometry about the double bond. By contrast, the isomer 12c was assigned as the Z-isomer from the upfield shift for the vinyl proton of C-3 to 6.18 ppm, and the smaller vicinal coupling of 11.4 Hz. Hydrolysis of ester 12a with base produced the free acid in 5 3 % yield for the two steps. The acid 36a was converted to the desired substrate esters 13a and 14 by a DCC coupling with either p-nitrophenol or 7-hydroxy-4-methylcoumarin. 61 Scheme 9. Synthesis of substrates 13a and 14. Reaction conditions (yields) are as follows: (a) n-BuLi, THF, -78 °C (62% of 12a, 10% of 12c); (b) KOH, THF-H 2 0, A (85%); (c) p-nitrophenol, DCC, CH 2 CI 2 (70%); (d) 7-hydroxy-4-methylcoumarin, DCC, DMAP, CH 2 CI 2 (79%). The isomeric aldehyde 10b was treated under the same reaction conditions to give p-nitrophenyl (2E,8Z)-2,8,10-undecatrienoate (13b) shown in Scheme 10. 62 Scheme 10. Synthesis of substrate 13b. Reaction conditions (yields) are as follows: (a) n-BuLi, THF, -78 °C (76%); (b) KOH, THF-H 2 0, A (77%); (c) p-nitrophenol, DCC, CH 2 CI 2 (72%). Since trimethyl phosphonoacetate (11b) gives reduced stereoselectivity in the Wadsworth-Emmons reaction as mentioned previously, this reagent was used to make the Z isomers of the alkene between C-2 and C-3 (Scheme 11). A mixture of the isomeric aldehydes 10a and 10b was reacted with 11b to give the four possible isomers of the methyl triene ester 4a, 4b, 4c, and 4d. Purification by flash chromatography resulted in the partial separation of the 2E (4a and 4b) and 2Z (4c and 4d) isomers which had been formed in approximately a 1:2 ratio. This reversal in stereoselectivity relative to that observed for 11a may be due to the prolonged reaction time in which the temperature was allowed to increase during the formation of the esters 4. The methyl esters 4c and 4d were hydrolysed with base and converted to the PNP esters so that all four isomers of the triene ester substrate 13 were available for testing the catalytic activity of our antibodies described below. The antibody kinetic studies on the triene substrates with the Z geometry about the C-2 double bond, were 63 carried out using a mixture of E and Z dienes about C-8, since the isomers 4c and 4d were inseparable by normal chromatographic procedures. CHO 10a 1:1 mixture of aldehydes .CHO 10b *C0 2 Me \ ^ 4a 4c 1:2 mixture of (4a/4b):(4c/4d) •C0 2 Me C 0 2 M e / 4b C 0 2 R 36c/36d R = H 13c/13d R = PNP Scheme 11. Synthesis of substrates 13c and 13d. Reaction conditions (yields) are as follows: (a) 11b, n-BuLi, THF, 0 °C (85% overall yield for all isomers, isolation of 4c and 4d, which were carried out as a mixture for the next reactions); (b) KOH, THF-H 2 0, A (75%); (c) p-nitrophenol, DCC, CH 2 CI 2 (63%). 2.1.8 Synthesis of a Diene Substrate To test substrate specificity for simple ester hydrolysis versus IMDA by the catalytic antibody, it was desirable to have a substrate that contained the ester moiety, but would be unable to undergo the IMDA reaction. Transformation of the conjugated 64 diene portion of compound 13 into a simple alkene would give such a substrate that retains the a,p-unsaturated ester moiety, so that the reactivity of ester hydrolysis is unchanged, but one that can no longer undergo a 4+2 cycloaddition. The synthesis of the substrate 44 (Scheme 12) follows a similar route used previously. Commercially available 1,8-octanediol was monoprotected as its THP ether to give 37 in good yields. The alcohol was oxidised under Swern conditions resulting in aldehyde 38. The alkene 39 was produced by a Wittig reaction with triphenylphosphonium bromide, then deprotected to give alcohol 40. Another Swern oxidation gave 8-nonenal, which was reacted with triethyl phosphonoacetate to give 42. Hydrolysis of the ester and subsequent coupling of the resulting acid with p-nitrophenol gave the final substrate 44 in 47% yield from the aldehyde 38. Compound 44 had a carbonyl peak at 1743 cm"1 in the IR spectrum, which corresponds to an a,p-unsaturated aryl ester. In addition, the aromatic protons of 44 were shifted downfield relative to those of free p-nitrophenol. 65 *0H .OH a,b f ,g 43 R = H 44 R = p-nitrophenyl R ,OTHP 37 R = CH 2 OH 3 8 R = CHO OEt ,OTHP 39 d, b R 40 R = CH 2 OH 41 R = CHO Scheme 12. Synthesis of substrate 44. Reaction conditions (yields) are as follows: (a) DHP, Dowex H + resin, toluene (98%); (b) (COCI)2, DMSO, Et 3N, CH 2 CI 2 , -78 °C (90% and 75%, respectively); (c) Ph 3PCH 3Br, n-BuLi, THF, 0 °C (83%); (d) pTsA, MeOH (77%); (e) 11a, n-BuLi, THF, -78 °C (87%); (f) KOH, THF-H2O, A (75%); (g) p-nitrophenol, DCC, CH 2 CI 2 (73%). 2.1.9 Stereochemical Outcome of the IMDA In the case of the 1-substituted triene substrates such as 4, the IMDA reaction is a powerful synthetic tool, due to the formation of two new carbon-carbon bonds and two new rings, as well as the creation and possible control of three new asymmetric centres. Because of a combination of steric requirements, stereochemistry of the starting triene, and reaction conditions, not all four of the possible diastereomers are produced from the intramolecular cycloaddition of a triene ester. 66 The starting stereochemistry of the dienophile controls the relative stereochemistry of two of the newly formed asymmetric centres in the adduct as shown in Scheme 13. The reaction proceeds in a concerted fashion, therefore, the syn or anti relationship between the ester and the vinyl hydrogen on C-3 depends on the geometry of the C-2 alkene in the starting material. Since this relative stereochemistry is retained in the product, the 2E-trienes 4a or 4b will lead to adducts 5a and 5b in which the ester and H-4a are syn to one another. However, for the (2Z)-2,8,10-undecatrienoates, this translates into an anti relationship of the ester and H-4a as shown in conversion of 4c and 4d to 5c and 5d. The selectivity for cis- or trans-fused bicyclic products is governed by the exo or endo transition state and is independent of dienophile stereochemistry. In certain cases, secondary orbital interactions (endo rule) can lead to high stereoselectivity. However, it has also been reported that these secondary orbital effects do not have a large stabilisation factor in those IMDA reactions that require high temperatures. 1 5 Therefore, the thermal cycloadditions of p-nitrophenyl (2E,8E)-2,8,10-undecatrienoate (13a) or the coumaryl ester analogue 14 resulted in a 1:1 mixture of cis- and frans-fused adducts 46a/46b or 47a/47b, respectively. These results are in agreement with those reported by Roush et a l . 1 5 in their study of simple methyl triene esters 4. The relative stereochemistry of the ring junction of the bicyclic structure depends on the endo or exo transition state of the reaction resulting in the formation of cis- and frans-fused products. However, the cycloaddition of the Z-diene isomers yields only one product because of steric constraints as shown in Scheme 13. In the case of 4b, since the triene cannot adopt an exo transition-state conformation only the endo adduct 67 is produced. Therefore, when a 1:1 mixture of 4a and 4b was heated to 150 °C, a 2:1 product ratio of 5a:5b was formed. Similarly, since isomer 4d leads only to the frans-fused product through the exo transition state, the IMDA reaction of 4c and 4d (as a 1:1 mixture) produced 5c and 5d in a 1:2 ratio, as expected. Product ratios of the adducts were determined by GC analysis. 68 46b R = PNP 47b R = coumarin Scheme 13. Stereochemical outcome from the thermal IMDA of alkyl 2,8,10-undecatrienoates, where R = methyl (4 5), PNP (13 -> 46), or coumaryl derivative (14 -» 47). 69 2.2 RABBIT STUDIES 2.2.1 Immunisat ions The conjugates KLH35 or KLH6 were prepared as an emulsion with Freund's adjuvant, then used to immunise four female New Zealand rabbits (designated as H54, H55, H60, and H61). The immunogen is immobilized by the oil emulsion, which allowed for slow release of the conjugate in the animal thereby increasing its lifetime in the system. Initial immunisations were made with complete Freund's adjuvant, which contains a heat killed mycobacteria that has been found to activate the humoral immune response. Subsequent boost injections were prepared with incomplete Freund's adjuvant that does not contain the bacteria since it would be toxic to the animal on repeated exposure. 2.2.2 Immune Response After the initial injection, the rabbits were immunised once every four weeks. The first test bleed was obtained two weeks after the second injection and subsequent test bleeds were performed every four weeks. The sera from these test bleeds (approximately 10-20 mL) contain polyclonal antibodies and therefore have a wide variety of different binding affinities. The serum was purified by chromatography on sepharose beads coupled with Protein A, which has a high affinity for the constant region of rabbit antibodies. Many of the contaminating proteins were removed by this purification, leaving a solution enriched in the polyclonal antibodies. 70 Test bleeds from rabbits H54 and H55, which were immunised with the immunoconjugate KLH6, were purified and screened for hapten-specific antibodies using an ELISA. In general, studies have shown that there is an increase in the antibody titre with subsequent boosts as shown in Figure 14. 2 0 The sera from rabbits injected with KLH6 were found to follow this pattern and showed the highest titres after the fourth boost, a subsequent boost showed no significant increase in the titres of the antibodies from rabbits H54 and H55. Isotype IgM Valency multi Affinity low Protein A no binding A IgG IgG IgG bi bi bi low-med med-high high yes yes yes A A A Serum Titre t t Primary Secondary t Tertiary t Multiple Time Injections Figure 14. Change in serum titre over time in a typical immune response (from ref 20). Recently, it has been reported that the catalytic activity of polyclonal antibodies changes over the course of the immunisation period. 4 6 Shreder et al. observed an increase in catalytic activity, as measured by k c at, in the serum of rabbits immunised with a phosphate hapten, which corresponded to an increase in hapten affinity through 71 four or five injections. However, over an extended immunisation period, the catalytic activity decreased while the binding affinity was retained. 4 6 Since there was no further increase in the binding affinity of the antibodies from H54 and H55, the rabbits were sacrificed. Body bleeds were performed on the animals to obtain large quantities of the polyclonal antibody mixture. The purified antibodies were found to bind specifically to the conjugate BSA6 and not to BSA alone (Figure 15). Therefore it was concluded that the binding was to hapten 6. A serial dilution of the antisera found that the antibodies exhibited hapten binding at greater than 10,000-fold dilutions. 72 a) 2.0 < 100000 Reciprocal Dilution b) 2 0 N 100 1000 10000 100000 1000000 Reciprocal Dilution Figure 15. Binding assays for purified antibodies from rabbits (a) H54 and (b) H55. Antibodies were titrated against BSA6 (squares) or BSA alone (circles). 73 By contrast, the sera from rabbits H60 and H61 that were injected with KLH35 showed little to no increase in binding affinity above background even after four injections of the immunoconjugate. The antibodies raised against the photochemically linked conjugate were found to bind nonspecifically as shown by the high absorbance readings to both BSA6 and BSA alone, shown by the open squares and open circles respectively in Figure 16. Since KLH35 did not seem to be an effective immunogen, KLH6 was used in the fifth boost in an attempt to use heterologous immunisation to obtain hapten-specific antibodies. The change in immunogen caused an immediate rise in the titres of hapten-specific antibodies. After two immunisations of KLH6, the titres appeared to reach a maximum value at which time the rabbits were sacrificed and body bleeds were performed on the animals. The binding curves of the purified antibodies obtained from these sera showed significant binding to hapten over background (Figure 16). However, the antibodies from H60 and H61 had lower titres than those obtained from rabbits H54 or H55. Positive anti-hapten responses at over 5,000-fold dilutions were observed for antibodies from H60 and H61. Reciprocal Dilution Figure 16. Binding assays for purified antibodies from rabbits (a) H60 and (b) H61 . Ant ibodies were obtained from immunisation with either K L H 3 5 (open markers) or K L H 6 (closed markers), and were titrated against B S A 6 (squares) or BSA alone (circles). 75 The strategy of heterologous immunisations has been used to generate catalytic antibodies for an acyl transfer reaction. 8 5 In contrast to standard immunisation protocols, Tsumuraya et al. used two different but structurally related haptens. An initial immunisation with a positively charged quaternary ammonium alcohol was followed by boosts with a negatively charged phosphonamidate. The host animal was observed to cross-reactively respond to the secondary antigen to produce antisera that had an affinity for the primary, secondary, or both antigens. Larger rate enhancements, as measured by k c at /k u n , were reported for the antibodies derived from the heterologous over the homologous (exposure to only one of the haptens) immunisations. The antibodies from the heterologous immunisations were also found to be tighter binders, as determined by K m .8 5 The initial antigen for rabbits H60/H61 was the photochemically coupled KLH35 using sulfo-SANPAH, which may be considered a heterogeneous mixture due to the method of attachment of the hapten to the linker. Photoactivation of the aryl azide results in a reactive nitrene that can undergo a variety of reactions including insertion in to NH or CH bonds, and addition across carbon-carbon double bonds. The presence of strongly electron withdrawing substituents on the aromatic ring makes the nitrene more electrophilic, resulting in the preferential formation of CH insertion products. 8 6 When the aryl azide was photolysed in the presence of excess hapten 35, it was expected that the nitrene should insert randomly into the hapten forming a wide variety of epitopes. Exposure of many different antigenic determinants to the immune system should increase the probability of obtaining an antibody that would have the desired dual activity. However, alternative reactions of the nitrene are also known to take place, 76 such as insertion into the protein KLH or into the solvent. It has been suggested that the low yields in photoreactive couplings are due to these side reactions. 8 7 Introduction of a different antigen K L H 6 into the animals resulted in an immediate increase in the titres of the H60 and H61 antibodies. It is unknown whether this new binding affinity is due primarily to K L H 6 or may have been induced by the combined effect from both K L H 6 and K L H 3 5 as has been observed in heterologous immunisations. Although the photochemically linked K L H 3 5 antigen did not produce hapten-specific antibodies on its own, it may aid in the development of a different immune response when used in conjunction with another antigen. The linkage site of the spacer arm on the hapten has been found to be important in giving different affinities in the antibodies studied. Yang et al. used three haptens that contained the same TSA for the hydrolysis of cocaine but differed in the tether site, to raise catalytic antibodies. 6 0 Subsequent competition binding assays on the isolated abzymes showed that although the free TSA was bound tightly by the antibodies, only one of the three haptens was recognised. The location of the linker was thought to be important in dictating the specificity of the antibodies produced and can be limiting in the detection of abzymes when binding assays are used. Although specific binding of the antibody to a transition-state analogue does not necessarily translate into catalytic activity, binding assays can provide valuable information, including the confirmation of successful coupling of the hapten to the carrier protein to form the immunoconjugate that was used to induce the immune response. In addition, antibody recognition of the hapten is a necessary requirement for catalytic activity. A tightly binding antibody may aid in transition-state stabilisation, which may be 77 used for catalysis by the antibody, analogous to enzyme activity. 8 8 The strength of the antibody binding was quantified using ELISA-based detection methods to give an estimate of the functional affinity of the polyclonal antibody mixture. 2.2.3 Calculation of the Equil ibrium Dissociation Constant The equilibrium dissociation constant K D is inversely related to the strength of the binding between an antigen and a specific antibody. Among the many techniques to determine this quantity, 2 4 the indirect competition ELISA-based method provides the advantage of both time and cost efficiency. 2 7 In addition, this method does not require the labelling of either the antigen or the antibody, which may affect the binding affinity. Equilibrium dissociation constants calculated by the indirect competition ELISA method on polyclonal antibodies are comparable to those obtained by radiolabelling studies. 8 1 The indirect competition method involves establishing an equilibrium in solution between a fixed concentration of the antibodies and the inhibitor such as 35, at varying concentrations. The amount of free antibody was then determined by an ELISA in which the antigen BSA6 was immobilised on a solid phase. The concentration of the antibody used in the binding equilibrium studies must be lower than K D .2 5 Since this value is unknown, the antibody concentration used in the assay was chosen to be as low as possible. The polyclonal antisera and the inhibitor were incubated in the solution phase for a minimum of 1.5 hours prior to the detection of any unbound antibody using an ELISA. If the binding of the free antibody to an immobilised antigen does not affect the equilibrium between the antibody and inhibitor that was established in solution, then 78 the concentration of free antibody, [Ab], can be correlated to the absorbance (A) measured in an ELISA (Equation 1 0 ) . 2 5 where [Ab]r is the total antibody concentration and Ao is the absorbance measured for the antibody in the absence of the inhibitor. If the total concentration of the inhibitor, [In], is much greater than that of the antibody (i.e. [In] > 10 x [Ab]j), then a modified Klotz equation can be used to determine KD, which is independent of the antibody concentration (Equation 1 1 ) . 2 5 The equilibrium constant of crude antibodies of unknown concentration have been determined from the slope of the plot of Ao/(Ao - A) vs. 1/[ln]. Friguet et al. have shown that there is good correlation between the K D measurements for pure and impure antibodies. 2 5 Therefore, the equilibrium dissociation constants for our polyclonal antibody mixtures were calculated using this method. The inhibitor used was the hapten derivative 35, since it was considered to be a good mimic of the proposed transition state for both the ester hydrolysis and the IMDA cycloaddition. Therefore, if 35 was found to inhibit the binding of antibodies to BSA6, then these antibodies would be likely candidates for catalytic activity. [Ab] / [Ab] T = A / A 0 (10) A o / ( A o - A ) = 1 + K D / [ l n ] (11) 35 79 Since 35 is insoluble in aqueous solvent, it was necessary to make up stock solutions of the inhibitor in DMSO. It was found that at greater than 2.5% D M S O , there was a solvent effect on the antibody-antigen binding. Nevertheless, the solubility of the inhibitor 35 in 2.5% D M S O allowed for an upper concentration limit of 250 LIM. Background control samples contained no inhibitor but equivalent concentrations of D M S O . Antigen and antibody concentrations used to coat the plates with were chosen to give initial absorbance readings (in the absence of inhibitor) in the linear range of the detector, as well as fulfilling the assumptions stated above. It was found that H54 antibodies were only slightly inhibited by the hapten and appeared to contain only weakly binding antibodies for the hapten 35. By contrast, H55 was strongly inhibited by 35 (Figure 17). Antibodies from H60 and H61 initially showed little or no binding to the hapten. However, immunisation with KLH6 lead to an increase in titres as mentioned previously. Unfortunately, these antibodies displayed similar binding inhibition results to that of H54. 80 0) < 0% J , , • r-0 10 20 30 [Inhibitor] x 10"6M Figure 17. Indirect competition coupled with ELISA detection of antibodies from rabbits H54 (diamonds), H55 (squares), H60 (circles), and H61 (triangles). Treatment of the binding data with the modified Klotz equation (Equation 11) gave a linear plot for H55 (Figure 18). The slope calculated by linear regression analysis of this data gives a K D value of 7.5 x 10"7 M for the H55 antibodies. 81 0 1 2 3 4 5 6 7 8 9 10 11 1 / [free Hapten] x 105 M"1 Figure 18. Klotz plot for polyclonal antibodies from rabbit H55 from competitive inhibition with 35. The equilibrium dissociation constant K D was calculated from the slope of the graph to be 7.5 x 10"7 M. 2.2.4 Relative Affinity Determination by Chaotropic Elution The strength of the binding between an antibody and antigen is dependent on the rate of dissociation of the Ab-Ag complex. Functional affinity can be measured from the degree of dissociation resulting from the action of chaotropic agents such as ammonium thiocyanate. Thiocyanate ions can disrupt the Ab-Ag interaction such that low affinity antibodies are readily dissociated in the presence of low concentrations of S C N - . An affinity index is calculated as the concentration of the chaotropic ion that causes a 50% reduction in the initial absorbance reading in the absence of the ion in an ELISA. Therefore a low affinity index indicates low affinity antibodies. 82 A fixed concentration of antibodies from the different rabbits was incubated with the antigen BSA6. Ammonium thiocyanate (0-8 M) was added and incubated for a short period of time at room temperature. Excess chaotropic agent was then removed and the amount of antibody remaining on the plates was detected by an ELISA. Antibodies that have the highest affinity for the hapten will have the greatest resistance to dissociation even in the presence of high concentrations of thiocyanate. By contrast, low affinity antibodies will be readily dissociated and lost by small amounts of the ion. The affinity profiles for the polyclonal mixtures are shown in Figure 19. [Ammonium Thiocyanate] (M) Figure 19. Antibody affinity profile measured by thiocyanate elution followed by ELISA detection for polyclonal antibodies from rabbits H54 (diamonds), H55 (squares), H60 (circles), H61 (triangles), and H56 (no marker). 83 The affinity index was calculated as the concentration of ammonium thiocyanate that is required to remove 50% of the antibodies. It was found that H55 was the most resistant to chaotropic elution, with an affinity index of 7.5 M. The other polyclonal mixtures were less tightly bound to the hapten and so were easily removed by low concentrations of the thiocyanate ion. Affinity indexes for H54, H60 and H61 were determined to be 2.5, 2, and 3 M respectively. A control polyclonal antibody mixture from rabbit H56, which was immunised with a conjugate that is unrelated to 35 or 6 had an affinity index of less than 2 M under the same conditions. 2.2.5 Test for Catalytic Activity All four polyclonal antibodies (H54, H55, H60, and H61) were tested for their catalytic activity against substrate 13a. In these assays, the rates of the reaction were measured by monitoring the release of the p-nitrophenolate anion by spectrophotometric methods. Background samples containing antibodies from rabbit H56 were used as a control. Only slow hydrolysis of the ester was observed under the conditions of the kinetic assay. The rates of p-nitrophenolate release for the samples containing antibodies from any of the rabbits immunised with our hapten were not above background. The possibility that the antibodies were catalysing only the IMDA reaction without hydrolysing the ester was also investigated. Samples from the kinetic runs above were extracted with diethyl ether for analysis of other product formation by HPLC. The triene 13a and the bicyclic product 46 were prepared by chemical means and found to be separable under normal phase HPLC conditions (Figure 20). However, no peak corresponding to adduct 46 was observed in any of the extracted samples. CD o c 03 -Q \ O co .a < CD .> '-*—* _co CD CH 2 n 1 .8 -1 .6 -1 .4 -1 . 2 -1 -0 . 8 -0.6 0.4 0.2 - f 0 0 PNP IMDA adduct PNP Triene 2 4 6 Time (min) i r 8 ~l ' I 10 12 Figure 20. HPLC trace of PNP triene substrate 13a and PNP IMDA adduct 46 on a Novapak A\x analytical silica column with 4% ethyl acetate in hexanes. Detection of products was carried out by UV absorbance at X = 205 nm. 85 2.2.6 Summary of Rabbit Antibodies Study Hapten-specific polyclonal antibodies were raised in rabbits using the EDC coupled KLH-hapten immunoconjugate. The photochemical coupling of KLH to hapten was less successful in generating hapten-specific antibodies. This may be a result of low yield of nitrene insertion into the hapten or it may be due to the lack of immunogenic character of the resulting conjugate. Quantification of the antibody binding affinities for the hapten 35 was carried out using indirect, competitive inhibition with ELISA detection. The best binding antibody mixture was found to be from rabbit H55 with a KD of 7.5 x 10"7 M. The antibodies from H54 were not inhibited by hapten 35, suggesting that the binding of H54 antibodies may be to the linker moiety of 6. Similarly, compound 35 did not inhibit antibodies from H60 and H61 to any great extent. Further binding studies using chaotropic elution assays confirmed these relative affinities. Of the four polyclonal antibody mixtures studied, H55 was shown to be the tightest binder with an affinity index of 7.5 M. Affinity indexes of the other antibodies (H54, H60 and H61) were all in the range of 2-3 M, which is comparable to that of the control antibodies H56. Although the observed hapten-specific binding of H55 did not translate into catalytic activity, immunisation of the rabbits with our antigens was a useful experiment to demonstrate that KLH6 did lead to generation of antibodies specific for the hapten 6. 86 2.3 MONOCLONAL ANTIBODIES 2.3.1 Immune Response Since KLH6 was a successful immunogen for generating hapten specific antibodies in rabbits, the same immunoconjugate was used for injection into Balb/c mice. After immunisation with the KLM6, test bleeds were taken from the mice and screened with a particle concentration fluorescence immunoassay (PCFIA), as described in Section 1.2.5. The antibodies in the sera displayed specific binding to BSA6 relative to BSA alone after the second boost of KLH6. In addition, there was an increase in the serum titre over time (Figure 21). Reciprocal Serum Dilution Figure 21. Typical results of PCFIA to test for specific binding to BSA6 (open markers) over BSA (closed markers). The first test bleed (squares) is lower in both the serum titre and hapten affinity relative to the second test bleed (circles). 87 A final immunisation of KLH6 was prepared in a 1 x PBS solution and injected intravenously. This boost was expected to lead to the production of a large number of antigen specific B-cells within the lymphoid tissues of the mouse. The antibody-secreting cells were isolated from the spleen and mixed with myeloma cells, then fused with polyethylene glycol (PEG). The newly fused cells were washed to remove the PEG, diluted in selection media, and plated in 96-well tissue culture dishes. Only those cells resulting from a successful fusion between a B cell and a myeloma cell were able to grow in the selection media, which contained hypoxanthine, aminopterine, and thymidine. These hybridomas retained desirable properties from each fusion partner, namely antibody production from the B-cell and immortality from the myeloma cell. Within 10 to 14 days, the hybridoma cells expanded into multicell colonies and there was a sufficient concentration of antibodies in the tissue culture supernatants to be assayed for binding to the hapten using a PCFIA. Initially, each well in the tissue culture dish contained many different colonies of hybridoma cells, with each clone producing a distinct monoclonal antibody. Wells that displayed a positive signal for binding to the hapten over background, as determined by the binding assay were selected. The cells from the well were transferred into a new multiwell tissue culture dish and a limit dilution was performed across the dish. After another 10 to 14 days supernatants from wells containing one cell colonies were again tested for hapten binding by PCFIA. The selection of positive wells followed by a limit dilution of the cells was repeated once more to ensure clonality of the cells. It was assumed that if a positive signal was observed from all the wells of the plate, then all the cells within that plate originate from the same clone and the desired monoclonal antibody had been isolated. Using this method, 13 colonies that initially produced antibodies that bound to 88 BSA6 were subcloned twice, after which two stabilised cell lines producing anti-6 monoclonal antibodies, F325 and F336, were isolated. 2.3.2 Binding Assays In general, as the immune response matures higher levels of specific antibodies will be found. However, a high serum titre does not necessarily translate into a high affinity antibody. A competitive inhibition assay and a chaotropic elution assay, similar to those described in Section 2.2.3 and 2.2.4 respectively, were performed on F325 and F336 to ascertain their relative affinity for the hapten. 0 2 4 6 8 1 / [Inhibitor] (x 10 4 M-1) Figure 22. A Klotz plot of the binding of 35 to F325 from absorbance readings of a competitive inhibition assay with ELISA detection. The KD of the impure antibodies was calculated from the slope of the graph. 89 The binding of F325 to BSA6 was inhibited by free hapten 35, while F336 binding was unaffected under the conditions of this assay. Treatment of the inhibition binding data with the Klotz equation (see Section 2.2.3) gave a linear plot, where the slope of the graph is equal to the equilibrium dissociation constant (Figure 22). A K D value F325 was calculated from this data to be 8.8 x 10"6 M. The binding between F325 and BSA6 was stronger relative to that of F336 and BSA6, as measured by a chaotropic elution assay using ammonium thiocyanate (see Section 2.2.4). The affinity profiles of the monoclonal antibodies F325 and F336 are shown in Figure 23. An affinity index of 3 M and 5 M was calculated for F336 and F325 respectively. By comparison HIL3, which is a monoclonal antibody that was generated against an immunogen unrelated to KLH6, was vulnerable to the thiocyanate ion and had an affinity index of less than 2 M. 90 100% 80% < 40% 20% 0% 1 1 1 1 1 1 1 1 1 1 0 1 2 3 4 5 6 7 8 9 [Ammonium Thiocyanate] (M) Figure 23. Affinity profiles for monoclonal antibodies F325 (squares), F336 (circles), and HIL3 (diamonds) from a chaotropic elution assay using ammonium thiocyanate. 2.3.3 Characterisation of Monoclonal Antibodies From an isotyping assay of the monoclonal antibodies F325 and F336, it was determined that both antibodies had a K light chain. In addition, F325 was found to be an IgGi class antibody, whereas F336 is an IgM class antibody. As mentioned previously, the primary immune response typically consists mainly of IgM class antibodies that are less specific. As the immune response matures, further boosts with the same antigen leads to mutations in the variable region of the antibody resulting in higher affinities. This event is often accompanied by class switching from IgM to IgG class antibodies. However, factors such as the concentration of the immunoconjugate 91 and the time between immunisations may also affect this maturity process and class switching may not take place. In such cases, IgM antibodies are found in the serum even after prolonged exposure to the antigen. The binding studies show that F325 has a greater affinity for the hapten relative to F336 as suggested by the results from both the competitive inhibition assay and from the chaotropic assay. This is in agreement with their respective isotypes since IgM antibodies typically have lower binding affinities than IgG antibodies. 2.3.4 Large Scale Production of the Monoclonal Ant ibody Due to the greater binding affinity exhibited by F325 compared to F336, only F325 was carried on for further studies. Approximately one million F325 cells were injected into a pristane-primed mouse leading to the production of a peritoneal tumour in the animal. The hybridoma cells expand and continue to secrete antibodies within the tumour which is made up of ascitic fluid. The ascites, which is rich in the desired monoclonal antibody, was collected and purified on a sheep anti-mouse affinity column. The antibody samples were dialysed into a 1 x PBS solution and the final antibody concentration was calculated from its absorbance at 280 nm (an absorbance of 1.4 is equal to 1 mg/mL of protein). Subsequent assays on the purified antibodies showed that the binding properties of F325 were retained during this ascites production. 2.3.5 Preliminary Catalytic Studies Kinetic assays were performed on the triene 13b since this substrate would allow us to study the IMDA reaction as well as the hydrolysis of the ester to the acid. If our 92 hapten design generated the appropriate active site within the antibody, then F325 should catalyse both reactions to give acid 45. The rate of the ester hydrolysis was measured spectrophotometrically by monitoring for the release of the p-nitrophenolate anion. Catalysis of the reaction by 1.0 u.M F325 was compared to control samples containing 1.0 \xM HIL3. The rate of hydrolysis was faster in the presence of F325 relative to HIL3 (Figure 24). 0 20 40 60 Time (min) Figure 24. Comparison of the activity of monoclonal antibodies F325 and HIL3 with substrate 13b. The release of p-nitrophenolate anion was detected at 400 nm at 37 °C on a UNICAM UV/Visible Spectrophotometer. The assays were run in duplicates and gave plots similar to that shown above The antibody was added to a solution of 630 u.M of the PNP ester 13b in 10% DMSO, 0.18% Triton X-100 and 1 x PBS. 93 The activity of porcine lipase at a concentration of 0.18 mg/mL was comparable to that of F325 for the hydrolysis of the ester 13b. However, a general esterase inhibitor, soya bean trypsin inhibitor (SBTI), affected the activity of the lipase sample while the rate of F325 was not inhibited by SBTI (Figure 25). CD o c CO . Q k_ o w -O < 0.05 0.04 J 0.03 - 0.02 CD CT) c CO F325, no inhibitor F325 + SBTI Lipase, no inhibitor o 0.01 I / / ^ Lipase + SBTI ~ l 20 30 40 Time (min) Figure 25. The effect of general esterase inhibitor SBTI on the activity of antibody F325 and a lipase. The assays were run in duplicates at 37 °C, where 0.18 mg/mL lipase or 0.23 mg/mL F325 was added to samples containing 33 ug/mL SBTI, 630 uM substrate 13b, 10% DMSO and 0.18% Triton X-100 in 1 x PBS. The results from both of these experiments were strong evidence that the observed hydrolytic activity was derived from F325 and was not due to the presence of contaminating esterolytic enzymes. Inhibition assays suggest that hapten 35 was an uncompetitive inhibitor for the activity of F325 on substrate 13b (Figure 26). In uncompetitive inhibition, the inhibitor affects the catalytic activity by binding to the antibody-substrate complex. Typically, 94 uncompetitive inhibitors do not resemble the substrate and are rarely observed for single substrate catalysts with the exception of small inhibitors such as protons. Although 35 was found to inhibit binding of F325 in the competitive binding assay described previously (see Section 2.3.2), this binding did not affect substrate affinity. This would suggest that the active site is different from the antibody-binding site for 35. 3 ^ E " E CD CO 1— lo '•4—' 0 3 1 / [Substrate] (mM-1) Figure 26. Inhibition assays with hapten 35 for F325-catalysed hydrolysis of substrate 13b. The reactions were performed in duplicates in the reaction buffers described above and the averaged initial rates were plotted. The concentrations of inhibitor 35 were: (•) no inhibitor; (V) 0.10 mM; (0) 0.14 mM; ( • ) 0.30 mM. 2.3.6 Substrate Specificity of F325 Antibody F325 was also tested for substrate specificity by studying the four possible isomers of the undecatrienoate. The 2E-esters 13a or 13b were hydrolysed at a faster rate relative to the mixture of 22-isomers 13c/13d (Figure 27). 95 13a C0 2 PNP OPNP OPNP O OPNP O 13b 13c 13d CD O c 03 \ O to .0 < c cu 0.06 - | CD C as .c O F325 + 1 3 b F325 +13a HIL3 + 13b HIL3 +13a F325 + 13C/13 HIL3 + 13c/13d 40 60 80 Time (min) 1 r 100 120 Figure 27. Relative rates of hydrolysis of the isomers of p-nitrophenyl (2,8,10)-undecatrienoate in the presence of 0.88 u.M F325. Background samples contained 0.88 pM HIL3. However, the difference in rates are most likely due to the fact that 2Z-isomers are more sterically hindered in hydrolysis reactions relative to 2E-isomers. This is in agreement with the significantly lower background rate of hydrolysis for the sample containing substrates 13c/13d. The activity of F325 with the partially reduced substrate 44 was also studied to test if the antibody was catalytic for only the hydrolysis reaction. A comparison of the 96 relative rates for the release of the p-nitrophenolate anion from these esters showed that F325 hydrolysed 13b faster than 44 (Figure 28). This suggested that the conjugated diene moiety was important for the activity of F325 and indicated that the antibody may be catalysing both the IMDA and the hydrolysis reaction. 13b 44 Figure 28. Comparison of the activity of 0.88 u.M F325 with triene substrate 13b versus diene substrate 44. Background samples contained 0.88 u.M HIL3. 97 2.3.7 Isolation and Determination of the Product f rom Kinetic runs Although it appeared that F325 was catalysing the release of p-nitrophenolate anion, it remained to be seen whether the IMDA reaction was also being catalysed by the antibody. The above results using the diene substrate indicated that the conjugated diene moiety was important for the rate enhancement observed (see Figure 28). However, it is necessary to isolate the IMDA adduct to prove that F325 was catalysing the cycloaddition reaction. The samples from the kinetic runs were pooled and concentrated by rotary evaporation. These samples were acidified and analysed by HPLC with UV detection at X = 205 nm. Under reverse phase conditions on an analytical 10 u. C-18 column eluting with water and acetonitrile, it was found that the acids of the triene isomers 36 and the IMDA adduct 45, which were obtained by hydrolysis of the corresponding methyl esters 4 or 5, could be separated and detected. The four possible isomers of the triene acids 36 were separable, however, the isomers of the Diels-Alder adduct 45 were not. Co-injection of the four isomers of the triene acids with the IMDA acids (at least two isomers) gave only five peaks on the HPLC trace (Figure 29). I C 0 2 H I ^ ^ C 0 2 H C 0 2 H 36a/36b 36c/36d 45 98 0 .32- . 0 10 20 30 40 Time (min) Figure 29. HPLC trace of undecatrienoic acid isomers 36 and the Diels-Alder adducts 45. Concentrated samples were obtained from the kinetic runs from the HIL3 (negative control antibody), lipase (positive control for hydrolysis activity), or F325 (possible catalytic antibody) runs. Analysis of these samples by HPLC for the acid products showed that all these samples gave similar HPLC traces. Comparison of the relative intensities of the triene acid 36 relative to the Triton peak showed that the HIL3 samples had less hydrolysed product than that for either of the lipase and F325 samples. This agreed with the kinetic data that showed little to no hydrolysis in the HIL3 samples. However, no peak corresponding to the IMDA acids 45 was detected for any of the samples under our HPLC assay conditions. In addition, the samples from kinetic runs were acidified and extracted with diethyl ether, and again no IMDA ester adduct 46 was detected under the normal phase HPLC conditions described earlier. 99 2.3.8 Michael is-Menten Kinetics Since no IMDA adduct was detected, it was assumed that only the hydrolysis of the ester was being catalysed by F325. An approximate value of K m and v m a x for this reaction was determined by measuring the initial reaction rates at different concentrations of the substrate ranging from 52.5 to 630 uM. A more accurate value of K m can be obtained by measuring the initial reaction rates at substrate concentrations that are 0.3 to 5 times Km. However, this was not possible for our system due to the limited solubility of the triene 13. The molar extinction coefficient (e) for the p-nitrophenolate anion in our reaction buffer system was determined by measuring the absorbance of a 24 uM solution of p-nitrophenol. Absorbance measurements (A) were performed on a Unicam UV/Vis spectrophotometer at 37 °C. Application of the Beer-Lambert law A = sci, where c is the concentration of the solute and I is the path length of the cuvette, gave a molar extinction coefficient of 16,000 M"1cm"1 at A,m a x = 400 nm for the p-nitrophenolate anion. The initial rates, taken from the slope of the absorbance versus time graphs, were converted in units of M"1min"1 by dividing the slope by the molar extinction coefficient. The initial rates for the hydrolysis of 13b in the presence of F325 were plotted against the substrate concentration (Figure 30). The kinetic parameters were calculated by fitting the data to a weighted non-linear regression of the Michaelis-Menten equation using the computer program GraFit. The K m for 13b was calculated to be 265 + 32 uM, with a Vmax of 0.13 ± 0.007 u,M/min in the presence of 1.0 pM F325. The k c a t was calculated as 0.13 min"1. The k u n was determined from the background runs and found 100 to be 7.8 x 10"5 min"1 which corresponds to an approximate rate enhancement of 1.7 x 10 3. 1 0 - , [Substrate] (uM) Figure 30. Plot of initial rates versus varying concentrations of substrate 13b to determine the catalytic activity of 1.0 u,M F325. 2.4 DISCUSSION 2.4.1 Transition-State Theory Analysis To gain a better understanding of the scope and limitations of antibody catalysis, transition-state theory has been employed to study abzyme-catalysed reactions. 8 9 9 1 If the hapten is an accurate mimic of the TSA, then the observed rate accelerations (kcat/kun) can be used to predict the theoretical value of KTs- From the thermodynamic cycle of the uncatalysed and antibody-catalysed reactions, the ratio of equilibrium constants for the reaction substrate and the transition state (Ks/KTs) is equal to k c a t / k U n 101 (see Section 1.3.2). A small value of KTs corresponds to tight transition-state binding, and thus efficient catalysis. The calculated KTs for F325 is 1.6 x 10"7 M. The observed dissociation constant for the representative TSA was measured by a competitive inhibition binding assay, where K D was found to be 8.8 x 10"6 M (see Section 2.3.2). Typically, the calculated KTs is much smaller than the observed dissociation constant. 8 8 This is not surprising since a stable TSA would not be expected to mimic all aspects of the true transition state. 2.4.2 Hydrolytic Abzymes A comparison of the X-ray structures of hydrolytic abzymes show that there is a high degree of homology in the amino acid sequence for many of the abzymes that were generated by immunisation with phosphonate haptens. 9 2 The TSA is typically buried deep within a hydrophobic pocket of the antibody-binding site with up to 95% of the hapten surface contained within the cavity. 9 3 A large proportion of the amino acids within the cavity have aromatic residues and the p-nitrophenyl group of the hapten is in a very hydrophobic environment at the bottom of the antibody-combining site pocket. 9 3 , 9 4 The phosphonate moiety is often positioned at the opening of the active site. 9 3 This is consistent with the phosphonate being close to the attachment site of the inducing hapten to the carrier protein. But in our immunogen KLH6, the linking arm is connected to the aryl group that is the mimic of the leaving group of the substrate. Therefore, the carrier protein may block this epitope site, such that the antibody cannot build an active site that is as complementary to the aromatic ring as in the more successful examples of hydrolytic abzymes. This may account for the low catalytic activity of F325. 102 2.4.3 Diels-Alder Abzymes Abzymes which catalyse the Diels-Alder reaction have been studied by X-ray crystallography, NMR spectroscopy, and amino acid analysis. 9 5" 9 8 The antibody active site consists of a hydrophobic pocket that contains approximately 80% of the hapten. 9 8 Successful Diels-Alderases are thought to act by the restriction of both rotational and translational entropy in the substrates. 9 5 9 6 In addition, the X-ray structure of the antibody-hapten complex shows that there is a specific hydrogen bonding interaction which likely serves to reduce the electron density on the dienophile, thereby accelerating the Diels-Alder reaction. 9 5 Abzymes for both the Diels-Alder reaction and an acyl hydrolysis reaction typically require a hydrophobic binding pocket to give the desired catalytic activity. I propose that the lack of dual activity by our antibody may be due to a competition for the hydrophobic pocket by both the bicyclic adduct and the aryl moiety. Since no Diels-Alder adduct was detected by the HPLC assays, it is suggested that only the aryl group of the hapten is involved in antibody recognition and therefore only the hydrolysis of the ester is catalysed by F325. However this does not explain the greater rate of p-nitrophenolate release from the triene 13 over the diene 44, since F325 should hydrolyse either of these substrates at comparable rates if the active site was only binding the p-nitrophenyl moiety. Alternatively, even if the binding pocket includes both the triene moiety as well as the aryl group, the success of our proposed IMDA pathway may depend on the cycloaddition reaction occurring before the hydrolysis to the acid. Therefore, if the ester hydrolysis reaction were faster, then IMDA products would not be formed. The ester is 103 a better EWG than the carboxylate making the energy barrier for the IMDA of the carboxylate of acid 36 too great to be overcome by the antibody. One way to test this hypothesis would be to use a more electron-withdrawing substituent, such as a ketone, that can not be hydrolysed. 2.5 FUTURE WORK Despite the many advances in the field of catalytic antibodies, there are many obstacles still to overcome. The low success rate in obtaining catalytic antibodies may be attributed in part to poor hapten design and limitations in the screening for catalysis. 2.5.1 Hapten Design Bimolecular Diels-Alder reactions have been successfully catalysed by antibodies. A major contribution to this catalysis is in the ability to bind both the diene and the dienophile in close proximity in a hydrophobic environment, thereby lowering the entropy of activation for the reaction. In the IMDA case, since the reaction is already unimolecular, this stabilisation is not as great. Therefore, there must be an additional transition-state stabilisation for an antibody to catalyse an IMDA reaction. Perhaps a hapten that better mimics the TS of the IMDA reaction may lead to a catalytic antibody for the cycloaddition reaction. Incorporation of the strained boat moiety such as proposed hapten 48 may be such a transition-state analogue. 104 R 48 The activation energy for the IMDA of simple trienes such as 13 may be too large for catalysis by an antibody. Substituents on the tether joining the diene and dienophile have been shown to lower the activation energy of IMDA carried out under thermal conditions. Thus, a substrate such as 49 may be a better target for antibody catalysis. The hapten 50 might be used to generate antibodies to recognise substrate 49 and catalyse its cycloaddition. Our antibody did catalyse the hydrolysis reaction. This catalytic activity may be improved by altering the placement of the linker. The majority of successful esterase-type abzymes have the phosphonate of the hapten close to the surface of the antibody active site. However, since the site of attachment to the carrier protein in our hapten is at the aryl group rather than directly on the phosphonate, the induced active site is unlikely to be an ideal fit. Our attempt at using the photolinker to make 49 50 105 immunoconjugate KLH35 was unsuccessful, but this is a potential way to obtain many linkage sites, thereby exposing different epitopes for antibody binding. In addition, placement of the linker at an alternative site would allow us to incorporate a p-nitrophenyl phosphonate in the hapten such as 50, since the nitro residue has been found to be important for antigen binding." 2.5.2 Screening for Kinetic Activity Other active abzymes may be missed in the screening since selection is primarily based on the results of binding assays. As can be seen with our antibodies, binding does not necessarily correlate with catalytic activity. Assays have been developed to allow direct screening for catalysis, but most techniques are not general for any reaction and these assays may require generation of other antibodies that can detect product. 1 0 0 " 1 0 5 A more general assay to screen libraries of catalytic antibodies based on the analysis of the reaction with acridone-tagged substrates by thin layer chromatography can detect products at levels as low as 10" 1 2 moles. 1 0 2 This approach might be used as a kinetic screening method for the detection of IMDA catalytic antibodies using a modified substrate such as 51. However, only the IMDA reaction is assayed by this procedure and product inhibition may be a problem with the catalytic activity of the antibody. 106 2.6 SUMMARY AND CONCLUSION A synthetic route to both the hapten 6 and substrate 13 was carried out. The IMDA reactions of vinyl phosphonates such as triene 15a were studied and resulted in a mixture of the Diels-Alder adducts in good yields. The hapten was coupled to KLH for immunisation in rabbits and mice. KLH6 was an effective immunoconjugate and led to the generation of anti-6 antibodies as determined by binding assays. Competitive inhibition assays and chaotropic elution assays were carried out on the sera obtained from the rabbits to determine the functional affinity of the antibodies. From these assays, the polyclonal antibodies from rabbit H55 were found to exhibit a significantly higher affinity relative to the other sera. A K D of 7.5 x 10"7 M was determined for H55 with an affinity index of 7.5 M. Unfortunately this binding energy did not result in any observable catalytic activity for either the hydrolysis of the ester or for the IMDA of the triene. However, the use of polyclonal antibodies showed that the coupling between KLH and the hapten was successful and immunisation of KLH6 generates anti-6 antibodies. Compared to the widely used monoclonal antibody approach, polyclonal antibodies are advantageous because generation of the latter is 107 significantly faster and less costly. Polyclonal antibodies can be directly isolated from the serum of the animal by affinity chromatography and are easier to obtain since there is no need for tissue culture manipulations. In addition, because the polyclonal antibody mixture is composed of the complete distribution of all antibodies elicited by a hapten, the catalytic results obtained with polyclonals are more representative of an animal's entire immune response. This aspect is especially advantageous in our attempts to catalyse two consecutive reactions. However, the heterogeneous nature of polyclonal antibodies may prevent the detection of moderately efficient Diels-Alderase catalysts that are present in small amounts. The immunoconjugate KLH6 was also used to immunise mice and hybridoma technology was used to produce monoclonal antibodies. Two monoclonal antibodies were isolated for binding studies. F325 was found to be a better binder with a K D of 8.8 x 10"6 M and exhibited catalytic activity in preliminary assays. Studies of the hydrolysis reaction of the p-nitrophenyl ester 13b showed that F325 followed Michaelis-Menten kinetics with a K m of 265 ± 32 u.M and v m a x of 0.13 ± 0.007 pM/min. Esterase inhibitors confirmed that this activity was due to the antibody and not contaminating esterases. However, hapten 35 appeared to be an uncompetitive inhibitor, suggesting that F325 contains both an esterase active site and a hapten binding site. Although we obtained some evidence that the conjugated diene moiety was important for the observed activity, neither the cyclic p-nitrophenyl ester of the IMDA product nor the acid cycloadduct could be detected by an HPLC assay. 108 3 EXPERIMENTAL 3.1 GENERAL CHEMICAL METHODS All moisture-sensitive reactions were performed under a N 2 atmosphere, using flame-dried or oven-dried glassware. Cold temperature baths were prepared as either a dry ice/acetone bath for -78 °C or ice/water for 0 °C. Anhydrous solvents (THF, ether, CH2CI2, toluene, methanol) used in reactions were dried according to literature procedures. 1 0 6 Reagents were supplied by Aldrich Chemical Co. and distilled prior to use if required. Standardisation of n-BuLi was carried out by titration against 2,2-diphenylacetic acid in THF at 0 °C until a pale yellow solution persisted. Solvents used in the work-up of reactions or chromatography were reagent grade and were used as received. The petroleum ether used was the low boiling (35-60 °C) fraction. Where possible, reactions were monitored by thin layer chromatography (TLC) using Merck silica gel 60 F254 precoated aluminum plates. Detection was carried out by irradiation with UV light at X = 254 nm and/or by reaction with TLC dipping reagent (1 mL p-anisaldehyde, 5 mL cone. H2SO4, 10 mL glacial acetic acid, 90 mL methanol) and heat. Work-up involving washing with a brine solution refers to a saturated solution of NaCl. Removal of solvent or concentration of the sample under reduced pressure refers to the use of a rotary evaporator equipped with a water aspirator and heating with a water bath. Further removal of solvent or water from the purified compounds prior to analysis or reaction was accomplished by drying on a vacuum pump. 109 Flash chromatography was performed using 230-400 mesh ASTM silica gel supplied by E. Merck Co. Plates for radial chromatography were prepared using silica gel 60 PF254 containing gypsum supplied by EM Science. Infrared (IR) spectra were recorded on a Bomem Michelson 100 FT-IR spectrophotometer as a neat sample of the liquid or as a CDCI3 solution in NaCI plates. Chemical shifts (8) for all nuclear magnetic resonance (NMR) spectra are reported in parts per million (ppm). Proton (1H) NMR spectra were recorded on either a Bruker AC-200 (200 MHz), Bruker WH-400 (400 MHz) or Bruker AMX-500 (500 MHz) spectrometer as deuteriochloroform (CDCI3) or deuterated methanol (CD3OD) solutions. All spectra were calibrated to the solvent peak (CDCI3 at 7.24 ppm or C D 3 O D at 3.3 ppm), with coupling constants (J) reported in Hertz (Hz). Carbon ( 1 3C) NMR spectra, with proton decoupling, were recorded on a Bruker AC-200 MHz (50 MHz) or a Varian XL-300 (75 MHz) as CDCI3 solutions and calibrated to the solvent peak at 77.0 ppm as an internal standard. Phosphorus ( 3 1P) NMR spectra, with proton decoupling, were recorded on a Bruker AC-200 (81.75 MHz) as CDCI 3 solutions, and calibrated versus 85% phosphoric acid as an external standard. Low and high resolution mass spectrometry (LRMS or HRMS, respectively) were performed on a Kratos-AEI model MS 50 for electron-impact (El) using an ionisation energy of 70 eV. Desorption chemical ionisation (DCI) spectra, using NH 3 gas, were obtained on a Delsi Nermag R10-10C for LRMS or a Kratos MS 80 RFA for HRMS. Microanalyses were carried out on a Carlo Erba Elemental Analyser 1106 at the UBC Microanalytical Lab by Mr. Peter Borda. 110 3.1.1 Methyl (2E,8£)-2,8,10-Undecatrienoate (4a), Methyl (2E,8Z)-2,8,10-Undecatrienoate (4b), Methyl (2Z,8£)-2,8,10-Undecatrienoate (4c), and Methyl (2Z,8Z)-2,8,10-Undecatrienoate (4d) To a cooled solution (0 °C) of trimethyl phosphonoacetate (2.17 g, 11.9 mmol) in 20 mL of THF was added n-BuLi (1.60 M, 7.50 mL, 12.0 mmol). The colourless solution was stirred for 30 min before the addition of a 1:1 mixture of aldehydes 10a and 10b (1.10 g, 7.97 mmol) via a cannula. After 7 h the reaction was quenched with water and extracted with 3 x 20 mL of diethyl ether. The combined organic extracts were washed with brine, dried over MgSCu, filtered and concentrated under reduced pressure. Purification by flash chromatography using 3:1 petroleum ether/ethyl acetate resulted in 1.32 g (85%) of a mixture of all four isomers of compound 4 as a colourless oil. Further purification by radial chromatography allowed for the partial separation of the isomers to give a mixture of 4a/4b or 4c/4d. The 2E:2Z isomers were formed in a 1:2 ratio determined by GC analysis. The 1 H NMR spectra for all isomers were assigned on the basis of calculated coupling constants and by comparison to the spectra for the analogous ethyl esters 12a, 12b, and 12c reported below. The ratio of both isomeric mixtures 4a:4b and 4c:4d were 1:1 by integration of the vinyl protons in the 1 H NMR spectra. 4a/4b 4c/4d 111 Isomer 4a IR (neat): 2939, 2864, 1733, 1670, 1442, 1284, 1209, 1004, 908 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.25-1.7 (m, 4H), 2.08 (m, 2H), 2.2 (m, 2H), 3.7 (s, 3H), 4.93 (dd, J=2, 10.3 Hz, 1H), 5.07(dd, J=1.7, 16.8 Hz, 1H), 5.66, (dt, J=15.2, 6.8 Hz, 1H), 5.8 (dt, J=15.6, 0.7 Hz, 1H), 6.03 (dd, J=10.2, 15.2 Hz, 1H), 6.29 (ddd, J=10.2, 10.2, 16.9 Hz, 1H), 6.95 (dt, J=15.6, 7.1 Hz, 1H); Isomer 4b 1 H NMR (200 MHz, CDCI 3) 8: 1.5 (m, 4H), 2.2 (m, 4H), 4.96 (dd, J=1.2, 10 Hz, 1H), 5.05 (dd, J=1.2, 15.4 Hz, 1H), 5.4 (dt, J=10, 7 Hz, 1H), 5.8 (dt, J=15, 1.2 Hz, 1H), 6.05 (dd, J=10, 10 Hz, 1H), 6.6 (ddd, J=10, 10, 15.1 Hz, 1H), 6.93 (dt, J=15, 6.8 Hz, 1H); Isomer 4c IR (neat): 2939, 2858, 1730, 1652, 1452, 1409,1207, 1182, 1000, 910, 837 cm" 1; 1 H NMR (200 MHz, CDCI3) 8: 1.5-1.7 (m, 4H), 2.2 (m, 2H), 2.6 (m, 2H), 3.69 (s, 3H), 5.05 (d, J= 10.3 Hz, 1H), 5.15 (dd, J=1.2, 16.8 Hz, 1H), 5.65 (dt, J=15.1, 7.5 Hz, 1H), 5.76 (dt, J=11.4, 1.8 Hz, 1H), 6.05 (m, 1H), 6.2 (dt, J=11.5, 7.6 Hz, 1H), 6.3 (m, 1H); Isomer 4d 1 H NMR (200 MHz, CDCI 3) 8: 1.3-1.7 (m, 4H), 2.2 (m, 2H), 2.6 (m, 2H), 3.69 (s, 3H), 5.05 (d, J=10 Hz, 1H), 5.15 (dd, J=1.2, 16.8 Hz, 1H), 5.4 (m, 1H), 5.76 (dt, 112 J=11.4, 1.8 Hz, 1H), 5.98 (dd, J=11, 11 Hz, 1H), 6.2 (ddd, J=11.5, 7.6 Hz, 1H), 6.6 (ddd, J=10.2, 10, 16.9 Hz, 1H); Mixture of all isomers: LRMS (El) m/z (relative intensity): 194 (20), 162 (14), 135 (100), 100 (16), 93 (17), 79 (21), 67 (24); HRMS Calcd for C i 2 Hi 8 0 2 :194.13068; found: 194.13070. 3.1.2 Methyl 1,2,3,4,4aa,5,6,8aa-Octahydronapthalene-5a-carboxylate (5a), Methyl 1,2,3,4,4ap\5,6,8aa-Octahydronapthalene-5p-carboxylate (5b) A 1:1 mixture of the methyl esters 4a and 4b (95.0 mg, 0.490 mmol) was dissolved in toluene (1 mL) and heated to 150 °C for 3 days in a sealed tube. The solvent was removed by rotary evaporation. Purification by radial chromatography, eluting with 3:1 petroleum ether/diethyl ether, afforded a mixture of the cycloadducts 5a and 5b (86.2 mg, 91%). The ratio of isomers 5a:5b was found to be 2:1 (cis:trans) by GC analysis. The 1 H NMR spectra were assigned by comparison to literature values. 1 5 IR (neat): 3020, 2925, 2854, 1737, 1444, 1371, 1311, 1257, 1189, 1161, 1037 cm" 1; 5a 5b 113 1 H NMR (200 MHz, CDCI 3) 5: 0.9-1.8 (m, 9H), 2-2.48 (m, 3H from 5a, 4H from 5b), 2.7 (ddd, J=5, 5, 7.3 Hz, 1H from 5a), 3.64 (s, 3H from 5b), 3.65 (s, 3H from 5a), 5.45 (br d, J=10 Hz, 1H from 5b), 5.55 (m, 2H from 5a), 5.6 (m, 1H from 5b); LRMS (El) m/z (relative intensity): 194 (20), 162 (8), 135 (100), 84 (27), 67 (5); HRMS Calcd for C i 2 Hi 8 0 2 :194.13068; found: 194.13075. 3.1.3 Methyl 1,2,3,4,4aa,5,6,8aa-Octahydronapthalene-58-carboxylate (5c), Methyl 1,2,3,4,4aB,5,6,8aa-Octahydronapthalene-5a-carboxylate (5d) A 1:1 mixture of the methyl esters 4c and 4d (100 mg, 0.515 mmol) was dissolved in toluene (1 mL) and heated to 150 °C for 3 days in a sealed tube. The solvent was removed by rotary evaporation. Purification by radial chromatography, eluting with 3:1 petroleum ether/diethyl ether, afforded a mixture of the cycloadducts 5c and 5d (68.2 mg, 68%), with 18.6 mg (19%) of recovered starting triene 4c and 4d. The ratio of isomers 5c and 5d was found to be 1:2 (cis:trans) by GC analysis. These isomers were separable by radial chromatography yielding a pure sample of 5c for further characterisation. The 1 H NMR spectrum for 5d was assigned by subtraction of 5c from the spectrum of the mixture. 5c 5d 114 Isomer 5c 1 H NMR (200 MHz, CDCI 3) 8: 1-1.8 (m, 8H), 2-2.4 (m, 3H), 2.42-2.58 (m, 1H), 2 . 6 1 -2.75 (m, 1H), 3.68 (s, 3H), 5.4 (br d, J=10 Hz, 1H), 5.6-5.7 (m, 1H); 1 3 C NMR (50 MHz): 8: 21.4, 21.8, 23.3, 25.5, 31.8, 36.1, 36.2, 44.0, 51.5, 125.5, 130.8, 175.2; Isomer 5d 1 H NMR (200 MHz, CDCI 3) 8: 1.1-2.75 (m, 13H), 3.64 (s, 3H), 5.3-5.6 (m, 2H); 1 3 C NMR (50 MHz): 8: 26.5, 27.0, 28.4, 31.2, 33.3, 37.0, 42.0, 42.8, 51.0, 123.6, 132.3, 174.9; Mixture of both isomers: IR (neat): 3024, 2923, 2854, 1737, 1442, 1216, 1159 cm' 1 ; LRMS (El) m/z (relative intensity): 194 (31), 162 (58), 134 (100), 91 (26), 79 (18); HRMS Calcd for C12H18O2:194.13068; found: 194.13048. 115 3.1.4 5-[N-Glutaric Acid-(hydroxy-p-aminophenoxyphosphoryl) ]-1,2,3,4,4a,5,6,8a-octahydronapthalene (6) OH o o 6 A solution of 34 (22 mg, 0.043 mmol) was dissolved in 3 mL of THF and 10 mL of 2 M K2CO3, and the reaction mixture was stirred at room temperature for 3 days. The reaction mixture was then quenched with 10% HCI, diluted and extracted with diethyl ether. The combined organic extracts were washed with brine, dried over MgS04, filtered and concentrated under reduced pressure. Purification by radial chromatography using 3:1 CH 2Cl2/ethyl acetate, followed by treatment with Dowex H + resin, afforded 11.8 mg (65%) of hapten 6 as an off-white solid. IR (CDCI3): 3315, 2977, 2931, 2858, 1730, 1679, 1608, 1508, 1444, 1409, 1207, 1112, 979, 838 cm" 1; 1 H NMR (200 MHz, CD 3OD) 8:1.2-1.8 (m, 13H), 1.95 (m, 2H), 2.4 (m, 4H), 5.5 (m, 2H), 7.15 (d, 2H), 7.45 (d, 2H); 3 1 P NMR 8: 31.9, 32.4 (minor); 32.9, 33.4 (major); LRMS (El) m/z (relative intensity): 421; (DCI, NH 3): 439 (M++18); (LSIM): 422; HRMS Calcd for C 2 iH 2 8N0 6 P (M ++1): 422.17325; found: 422.1748. 116 3.1.5 (6£)-6,8-Nonadienal (10a) 10a The alcohol 22a (2.23 g, 0.0159 moi) was oxidised under Swern conditions as described for compound 20. Purification by Krughelor distillation afforded 1.82 g (83%) of 10a as a volatile, colourless oil. IR (neat): 3091, 2939, 2850, 2725, 1732, 1653, 1607, 1460, 1423, 1143, 1010, 964, 906 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.45-1.85 (m, 4H), 2.1 (dt, J=7.1, 7.3 Hz, 2H), 2.43 (dt, J=1.8, 7.2 Hz, 2H), 4.95 (dd, J=1.9, 10 Hz, 1H), 5.08 (dd, J=1.9, 16.9 Hz, 1H), 5.79 (dt, J=15.1, 6.7 Hz, 1H), 6.05 (dd, J=10.7, 15 Hz, 1H), 6.3 (ddd, J=10, 10.7, 16.9 Hz, 1H), 9.8 (t, J=1.7 Hz, 1H); 1 3 C N M R (50 MHz) 8:21.5, 28.6, 32.1,43.7, 115.0, 131.4, 134.4, 137.1,202.5; LRMS (El) m/z (relative intensity): 138 (4), 93 (11), 84 (100), 79 (40), 67 (64); HRMS Calcd for C9H140:138.1045; found: 138.1049; Anal. Calcd for C 9 H 1 4 O : C, 78.21; H, 10.21. Found: C, 78.28; H, 10.04. 117 3.1.6 (6Z)-6,8-Nonadienal (1 Ob) 10b The alcohol 22b (0.239 g, 1.71 mmol) was oxidised under Swern conditions as described for compound 20 to obtain a volatile sample of aldehyde 10b (22.9 mg, 97%) as a colourless oil. 1 H NMR (200 MHz, CDCI 3) 5: 1.3 (m, 2H), 1.5 (m, 2H), 2.1 (m, 2H), 2.4 (t, J=7 Hz, 2H), 4.95 (dd, J=1.2, 10 Hz, 1H), 5.1 (dd, J=1.2, 15 Hz, 1H), 5.4 (dt, J=10, 7 Hz, 1H), 6.0 (dd, J=10, 10 Hz, 1H), 6.6 (ddd, J=10, 10, 15 Hz, 1H), 9.8 (s, 1H). 3.1.7 Ethyl (2£,8£)-2,8,10-Undecatrienoate (12a), Ethyl (2Z,8E)-2,8,10-Undecatrienoate (12c) 12a 12c To a cooled solution (0 °C) of triethyl phosphonoacetate (403 mg, 1.77 mmol) in 10 mL of THF was added n-BuLi (1.56 M, 1.15 mL, 1.79 mmol). The colourless solution was stirred for 1 h before the addition of aldehyde 10a (226 mg, 1.64 mmol) via a cannula. After 3.5 h, the reaction was quenched with water and extracted with 3 x 20 mL of diethyl ether. The combined organic extracts were washed with 1 M HCI and brine, dried over MgSCu, filtered and concentrated under reduced pressure. Purification by radial chromatography using 10:1 petroleum ether/diethyl ether resulted in the isolation of 209 mg (62%) of compound 12a and 34.2 mg (10%) of isomer 12c as colourless oils (86:14 ratio of 2E2Z). Isomer 12a IR (neat): 2987, 2937, 2867, 1728, 1658, 1454, 1369, 1270, 1186, 1101, 1056, 1016, 904 cm" 1; 1 H NMR (200 MHz, CDCI 3) 5: 1.25 (t, J=7 Hz, 3H), 1.3-1.5 (m, 4H), 1.9-2.2 (m, 4H), 4.15 (q, J=7 Hz, 2H), 4.96 (dd, J=1.2, 10 Hz, 1H), 5.05 (dd, J=1.2, 16.4 Hz, 1H), 5.7 (dt, J=15, 7 Hz, 1H), 5.8 (dt, J=15, 1.2 Hz, 1H), 6.05 (dd, J=10, 15 Hz, 1H), 6.3 (ddd, J=10, 10, 16. Hz, 1H), 6.9 (dt, J=15, 6.8 Hz, 1H); 1 3 C NMR (50 MHz) 8: 14.3, 27.5, 28.6, 32.0, 32.2, 60.1, 114.9, 121.4, 131.2, 134.8, 137.2, 149.0, 167.3; LRMS (El) m/z (relative intensity): 208 (10), 162 (36), 134 (100), 114 (78), 93 (68), 81 (86), 67 (76); HRMS Calcd for C13H20O2: 208.14633; found: 208.14572. Isomer 12c 1 H NMR (200 MHz, CDCI 3) 8: 1.25 (t, J=7 Hz, 3H), 1.43 (m, 4H), 2.08 (m, 2H), 2.63 (m, 2H), 4.15 (q, J=7 Hz, 2H), 4.92 (dd, J=1.2, 10 Hz, 1H), 5.05 (dd, J=1.3, 16.8 Hz, 119 1H), 5.7 (dt, J=15, 6.9 Hz, 1H), 5.73 (dt, J=11.7, 1.7 Hz, 1H), 6.05 (dd, J=10.2, 15.1 Hz, 1H), 6.18 (dt, J=11.4, 7.4 Hz, 1H), 6.3 (ddd, J=10, 10, 16.1 Hz, 1H). 3.1.8 Ethyl (2E,8Z)-2,8,10-Undecatrienoate (12b) This compound was prepared following the procedure outlined for compound 12a using aldehyde 10b (157 mg, 1.14 mmol) to afford 180 mg (76%) of ester 12b as a colourless oil. 1 H NMR (200 MHz, CDCI 3) 5: 1.3 (t, J=7 Hz, 3H), 1.45 (m, 4H), 2.25 (m, 4H), 4.15 (q, J=7 Hz, 2H), 5.05 (dd, J=1.5, 10 Hz, 1H), 5.15 (dd, J=1.5, 15 Hz, 1H), 5.4 (dt, J=10, 7 Hz, 1H), 5.8 (dt, J=15, 1.2 Hz, 1H), 6.0 (dd, J=10, 10 Hz, 1H), 6.6 (ddd, J=10, 10, 15 Hz, 1H), 6.95 (dt, J=15, 6.8 Hz, 1H); 1 3 C NMR (50 MHz) 8: 14.2, 27.4, 27.5, 29.3, 32.4, 60.1, 117.0, 121.4, 129.5, 132.1, 132.8, 149.3, 166.7. 12b 120 3.1.9 p-Nitrophenyl (2£,8£)-2,8,10-Undecatrienoate (13a) O 13a A solution of the acid 36a (44 mg, 0.24 mmol), p-nitrophenol (49.3 mg, 0.354 mmol) and DCC (65 mg, 0.32 mmol) in CH2CI2 was stirred at room temperature for 5.5 h. The resulting yellow solution was evaporated. The residue was taken up in a minimal amount of CH 2Cl2/diethyl ether and filtered through a silica plug to remove the precipitate. Purification by radial chromatography using 4:1 petroleum ether/ethyl acetate afforded 51.4 mg (70%) of ester 13a as a pale yellow, viscous oil. IR (neat): 2929, 2858, 1743, 1652, 1595, 1523, 1346, 1213, 1122 cm" 1; 1 H NMR (200 MHz, CDCI3) 5: 1.5 (m, 4H), 2.1 (m, 2H), 2.3 (m, 2H), 4.98 (dd, J=1.6, 10 Hz, 1H), 5.1 (dd, J=1.6, 17 Hz, 1H), 5.7 (dt, J=15, 7 Hz, 1H,), 6.0 (d, J=15.7 Hz, 1H), 6.1 (dd, J=10, 15.1 Hz, 1H), 6.3 (ddd, J=10, 10, 17 Hz, 1H), 7.1 (dt, J=15.7, 6.8 Hz, 1H), 7.3 (d, J=9.3 Hz, 2H), 8.3 (d, J=9.3 Hz, 2H); 1 3 C NMR (50 MHz) 5: 27.3, 28.6, 32.2, 32.4, 115.1, 119.8, 122.5, 125, 125.2, 131.4, 136, 139, 154, 157, 166; LRMS (El) m/z (relative intensity): 301 (17), 271 (3), 163 (26), 135 (100), 93 (43), 67 (92); 121 HRMS Calcd for C17H19NO4: 301.13141; found: 301.13126; Anal. Calcd for C17H19NO4: C, 67.76; H, 6.36; N, 4.65. Found: C, 68.00; H, 6.45; Esterification of acid 36b (90 mg, 0.50 mmol) with p-nitrophenol was carried out as described for 13a to afford 108 mg (72%) of compound 13b. 1 H NMR (200 MHz, CDCI 3) 6: 1.4-1.7 (m, 4H), 2.2 (m, 2H), 2.3 (m, 2H), 5.1 (dd, J=1.2, 10 Hz, 1H), 5.2 (dd, J=1.3, 15 Hz, 1H), 5.42 (dt, J=10, 7 Hz, 1H), 6.05 (dt, J=15, 1.2 Hz, 1H), 6.1 (dd, J=10, 10 Hz, 1H), 6.6 (ddd, J=10, 10, 15 Hz, 1H), 7.2 (dt, J=15, 7 Hz, 1H), 7.3 (d, 2H), 8.25 (d, 2H); LRMS (El) m/z (relative intensity): 301(7), 163 (41), 139 (37), 135 (100), 109 (15), 93 (23), 78 (31). N, 4.67. 3.1.10 p-Nitrophenyl (2E,8Z)-2,8,10-Undecatrienoate (13b) 13b 122 3.1.11 p-Nitrophenyl (2Z,8£)-2,8,10-Undecatrienoate (13c), p-Nitrophenyl (2Z,8Z)-2,8,10-Undecatrienoate (13d) 13c 13d Esterification of acids 36c/36d (40.1 mg, 0.223 mmol) with p-nitrophenol was carried out as outlined for compound 13a to afford 42 mg (63%) of isomers 13c/3d. Purification of the mixture of diastereomers by radial chromatography using 5:1 petroleum ether/diethyl ether afforded a small amount of pure 13d. The 1 H NMR spectrum of compound 13c was assigned by subtraction of 13d from the 1 H NMR spectrum of the mixture of the two isomers. Isomer 13c 1 H NMR (200 MHz, CDCI 3) 8: 1.4 (m, 4H), 2.1 (m, 2H), 2.6 (m, 2H), 4.9 (dd, J=1.2, 10.3 Hz, 1H), 5.05 (dd, J=1.2, 15.6 Hz, 1H), 5.67 (dt, J=15.2, 6.8 Hz, 1H), 5.97 (dt, J=11.3, 1.5 Hz, 1H), 6.03, (dd, J=10.2, 15.2 Hz, 1H), 6.3 (dt, J=10.2, 10.2, 15.9 Hz, 1H), 6.5 (dt, J=11.5, 7.8 Hz, 1H), 7.2 (d, 2H), 8.3 (d, 2H); Isomer 13d 1 H NMR (200 MHz, CDCI 3) 8: 1.4 (m, 4H), 2.2 (m, 2H), 2.6 (m, 2H), 4.9 (dd, J=1.2, 10.3 Hz, 1H), 5.05 (dd, J=1.2, 15.6 Hz, 1H), 5.4 (dt, J=10, 7.8 Hz, 1H), 5.97 (dt, J=11.3, 1.5 Hz, 1H), 6.03 (dd, J=10.2, 15.2 Hz, 1H), 6.5 (dt, J=11.5, 7.8 Hz, 1H), 6.6 (m, 1H), 7.2 (d, 2H), 8.3 (d, 2H); 123 Mixture of both isomers: IR ( C D C I 3 ) : 3086, 2932, 2858, 1741, 1640, 1615, 1592, 1525, 1491, 1415, 1347, 1265 c m 1 ; 1 3 C NMR (50 MHz) 8: 27.3, 28.2, 29.1, 32.1, 118, 122, 125, 129.5, 131.2, 132.1, 134.7, 137.1, 145.1, 155.1, 163.3; LRMS (El) m/z (relative intensity): 301(0.7), 224 (4), 195 (7), 163 (40), 139 (37), 135 (100), 109 (15), 93 (46), 79 (30), 67 (98); HRMS Calcd for C17H19NO4: 301.13141; found: 301.13096. 3.1.12 7 -Oxy-4 -methylcoumaryl (2E,8£)-2,8,10-Undecatrienoate (14) A solution of acid 36a (100 mg, 0.555 mmol), 7-hydroxy-4-methylcoumarin (170 mg, 0.966 mmol), DCC (150 mg, 0.728 mmol) and DMAP (catalytic amount) in 10 mL of CH2CI2 was stirred at room temperature for 12 h. The resulting precipitate was filtered off and the filtrate was concentrated to give a white solid. Purification by radial chromatography using 5:1 petroleum ether/ethyl acetate afforded 147 mg (79%) of ester 14 as a white solid. 124 IR (CDCI3): 2937, 2862, 1739, 1662, 1614, 1438, 1276, 1236, 1139, 1029 cm" 1; 1 H NMR (200 MHz, CDCI3) 8: 1.4-1.7 (m, 4H), 2.1 (m, 2H), 2.28 (m, 2H), 2.41 (d, J=0.97 Hz, 3H), 4.94 (dd, J=1.5, 10 Hz, 1H), 5.1 (dd, J=1.7, 16.6 Hz, 1H), 5.7 (dt, J=15.2, 7.1 Hz, 1H), 6.05 (dt, J=15.6, 1.5 Hz, 1H), 6.3 (s, 1H), 6.3 (ddd, J=10, 10, 16.9 Hz, 1H), 6.2-6.4 (m, 1H), 7.1 (s, 1H), 7.12 (d, J=8.3 Hz, 1H), 7.2 (dt, J=15.6, 7.1 Hz, 1H),7.6 (d, J=8.3 Hz, 1H); LRMS (El) m/z (relative intensity): 338 (38), 177 (19), 163 (12), 148 (15), 135 (58), 93 (45), 67 (100); HRMS Calcd for C21H22O4: 338.15181; found: 338.15121; Anal. Calcd for C 2 iH 2 20 4 : C, 74.54; H, 6.55. Found: C, 74.19; H, 6.68. 3.1.13 Diphenyl Tr iphenylphosphoranyl idenemethylphosphonate (15) O P h 3 P V ^ \ 0 P h OPh 15 The ylide 15 was prepared according to the procedure outlined by Jones et a l . 7 3 To a cooled solution (0 °C) of chloromethylphosphonic dichloride (3.04 g, 18.2 mmol) and phenol (3.74 g, 39.8 mmol) in 75 mL of CH 2 CI 2 was added Et 3N (5.60 mL, 40.4 mmol). The mixture was stirred for 30 min before the ice bath was removed and reaction was allowed to continue at room temperature for an additional 3 h. The solvent was 125 removed under reduced pressure and the resulting white solid was taken up in a 1:1 petroleum ether/diethyl ether mixture and filtered through a silica plug. The filtrate was concentrated under reduced pressure and purified by flash chromatography with 5:1 petroleum ether/ethyl acetate to afford 4.61 g (91%) of chloromethyl diphenylphosphonate. The phosphonate was then combined with triphenylphosphine (4.42 g, 16.9 mmol) and heated to 170 °C for 8 h. The reaction mixture was cooled and the resulting glassy brown solid was taken up in THF in which the impurities were soluble. The insoluble white salt of the desired product was filtered and collected, then dissolved in a 1:1 water/CH2Cl2 mixture. A 3 M NaOH solution was added slowly to the salt until the solution was basic. After 1 h the layers were separate and the aqueous phase was extracted with 3 x 50 mL of CH2CI2. The combined organic layers were dried with MgSCu, filtered and concentrated to give a creamy white solid. Recrystallisation from ethyl acetate yielded 6.22 g (75%) of the Wittig reagent 15 as shiny, white crystals. IR (CDCI 3): 3090, 2930, 1599, 1463, 1410, 1205, 1080 cm"1; 1 H NMR (200 MHz, CDCI3) 8: 1.45 (s, 1H), 7-7.6 (m, 25H); 1 3 C NMR (50 MHz) 8:121.3, 123.5, 128.4, 128.6, 129.0, 131.7, 131.8, 132.9, 133.1; 3 1 P NMR 8: 20.7, 27.3; LRMS (El) m/z (relative intensity): 508 (10), 507 (22), 415 (38), 352 (100), 321 (18), 277 (53), 199 (24), 183 (86), 165 (38), 152 (31), 94 (63), 77 (54); HRMS Calcd for C 3 i H 2 6 0 3 P 2 : 508.13126; found: 508.13571. 126 3.1.14 (1E,7£)-1-Diphenoxyphosphoryl-1,7,9-decatriene (16a), (1Z,7E)-1-Diphenoxyphosphoryl-1,7,9-decatriene (16b) A solution of the aldehyde 10a (1.29 g, 9.35 mmol) and Wittig reagent 15 (6.58 g, 13.0 mmol) in 50 mL of toluene was refluxed for 4 days. The reaction was cooled and the solvent was removed under reduced pressure. The crude, brown oil was taken up in diethyl ether and washed with 2 x 20 mL of 3 M NaOH and 1 x 30 mL of brine. The organic layer was dried over MgSCu, filtered and concentrated under reduced pressure to give an orange solid. The crude material was taken up in a minimum amount of 10:1 petroleum ether/diethyl ether solution, and the resulting slurry was filtered through a silica plug, eluted with 5:1 petroleum ether/diethyl ether. The yellow oil was further purified by flash chromatography, followed by radial chromatography to afford pure samples of phosphonates 16a (1.12 g, 33%), 16b (0.189 g, 5%) and 25 (0.782 g, 23%) with an overall yield of 67% for all three products which were formed in a 7:1:4 ratio of 16a :16b 25. Isomer 16a IR (neat): 3068, 3016, 2931, 1627, 1591, 1488, 1456, 1270, 1213, 1189, 1070, 1026, 1006, 929, 833, 769 cm" 1; 127 1 H NMR (200 MHz, C D C I 3 ) 8: 1.25-1.5 (m, 4H), 2.05 (m, 2H), 2.25 (m, 2H), 4.97 (dd, J=1.2 Hz, 10, 1H), 5.1 (dd, J=1.2, 16.7 Hz, 1H), 5.63 (dt, J=15.4, 6.6 Hz, 1H), 5.85 (ddt, J=22.9, 17.1, 1.7 Hz, 1H), 6.02 (dd, J=10.2, 15.3 Hz, 1H), 6.3 (ddd, J=10 10, 16.9 Hz, 1H), 6.95 (m, 1H), 7.15-7.4 (m, 10H); 1 3 C NMR (50 MHz) 8: 27.0, 28.3, 32.1, 33.8, 113.8, 115.0, 117.6, 120.6, 125.0, 129.6, 131.3, 134.5, 137.1, 156.7; 3 1 P N M R 8 : 1 1 . 6 ; LRMS (El) m/z (relative intensity): 368 (52), 274 (100), 235 (6), 193 (9), 134 (68), 117 (38), 94 (74), 77 (45); HRMS Calcd for C 2 2 H 2 5 03P: 368.15414; found: 368.15465; Anal. Calcd for C 2 2 H 2 5 0 3 P : C, 71.72; H, 6.84. Found: C, 71.55; H, 6.83. Isomer 16b IR (neat): 3068, 3018, 2925, 2854, 1591, 1488, 1456, 1270, 1215, 1189, 1162, 1110, 1070, 1026, 925, 765 cm" 1; 1 H NMR (400 MHz, CDCI 3) 8: 1.35-1.5 (m, 4H), 2.05 (d, J=7.8 Hz, 2H), 2.55 (m, 2H), 4.94 (dd, J=1.5, 10 Hz, 1H), 5.06 (dd, J=1.5, 16.9 Hz, 1H), 5.62 (dt, J=15.2, 7 Hz, 1H), 5.77 (ddt, J=21.7, 12.9, 1.5 Hz, 1H), 6.0 (dd, J=10.3, 15.1 Hz, 1H), 6.28 (ddd, J=10.2, 10.2, 16.9 Hz, 1H), 6.6 (ddt, J=57, 13, 7.8 Hz, 1H), 7.15-7.4 (m, 10H); 128 1 3 C NMR (50 MHz) 8: 27.4, 28.7, 30.8, 32.2, 116.9, 120.3, 125.0, 129.5, 131.2, 132.2, 134.8, 137.2, 150.3, 156.7; 3 1 P NMR 8: 9.6; LRMS (El) m/z (relative intensity): 368 (34), 362 (12), 287 (18), 274 (100), 193 (12), 134 (30), 117 (64), 94 (86), 77 (59); HRMS Calcd for C22H25O3P: 368.15414; found: 368.15439; Anal. Calcd for C22H25O3P: C, 71.72; H, 6.84. Found: C, 71.38; H, 6.7. 3.1.15 Ethyl (2E)-8-Hydroxy-2-octenoate (17) 17 This compound was made following the procedure of Takacs et a l . 7 0 To a cooled solution (-78 °C) of triethyl phosphonoacetate (14.7 g, 0.0659 moi) in 200 mL of THF was added n-BuLi (1.60 M, 41.2 mL, 0.0659 moi) over 30 min. The colourless solution was stirred for 10 min before addition of e-caprolactone (7.49 g, 0.0658 moi) solution in 10 mL of THF via cannula. The reaction mixture was stirred for a further 15 min before the dropwise (1 mL/min) addition of Dibal-H (1.0 M, 66 mL, 0.066 moi). The reaction was stirred for 20 h at -78 °C and then quenched with Na2SO4-10H2O. The resulting 129 cloudy, thick liquid was filtered through a silica plug and eluted with diethyl ether, then concentrated under reduced pressure to give a yellow oil. Purification by flash chromatography using 3:1 petroleum ether/ethyl acetate afforded 8.59 g (70%) of pure compound 17 as a colourless oil. IR (neat): 3417 (br), 2943, 2855, 1722, 1656, 1467, 1375, 1267, 1186, 1058 cm' 1 ; 1 H NMR (200 MHz, CDCI 3) 5: 1.15 (t, J=7 Hz, 3H), 1.3-1.6 (m, 7H), 2.2 (ddt, J=1.5, 7.1, 7 Hz, 2H), 3.6 (dt, J=6.1, 4.7 Hz, 2H), 4.1 (q, J=7 Hz, 2H), 5.7 (dt, J=15.6, 1.7 Hz, 1H), 6.93 (dt, J=15.6, 7.1 Hz, 1H); 1 3 C NMR (50 MHz) 8: 14.2, 25.3, 27.8, 32.1, 32.4, 60.1, 62.6, 121.4, 149.0, 166.7; LRMS (El) m/z: 187 (M ++1). 3.1.16 Ethyl (2£)-8-Tetrahydropyranyloxy-2-octenoate (18) C 0 2 E t J JDTHP A solution of 17 (8.57 g, 0.0460 moi), DHP (7.75 g, 0.0923 moi) and PPTs (1.15 g, 0.00461 moi) in 250 mL of CH2CI2 was stirred at room temperature for 4 h. The solvent was then removed by rotary evaporation and the residue was dissolved in 150 mL of diethyl ether. The organic layer was washed with 4 x 60 mL of half-saturated NaCl 18 solution, dried over MgS0 4 and filtered. The solvent was removed under reduced pressure to give a yellow oil which was purified by flash chromatography using 5:1 petroleum ether/ethyl acetate to afford 9.85 g of 18 (79%) as a colourless oil. IR (neat): 2937, 2867, 1720, 1662, 1263, 1200, 1128, 1041, 975 cm' 1 ; 1 H NMR (200 MHz, CDCI 3) 8: 1.25 (t, J=7.2 Hz, 3H), 1.3-1.9 (m, 12H), 2.2 (m, 2H), 3.2-3.9 (m, 4H), 4.1 (q, J=7.1 Hz, 2H), 4.55 (m, 1H), 5.8 (dt, J=15.1, 1.5 Hz, 1H), 6.9 (dt, J=15.9, 6.8 Hz, 1H); 1 3 C NMR (50 MHz) 8: 14.4, 20.2, 25.2, 25.5, 28.0, 30.7, 31.1, 32.9, 60.5, 63.2, 68.1, 99.7, 122.5, 149.2, 168.3; LRMS (DCI, NH 4+ ) m/z (relative intensity): 288 (M ++18, 25), 271 (M + +1, 15), 187 (100), 102 (60), 85 (90); HRMS Calcd for C i 5 H 2 6 0 4 : 288.21748 (M ++18, NH 4 + ) ; found: 288.21720 (M ++18, NH 4 + ) ; Anal. Calcd for C i 5 H 2 6 0 4 : C, 66.64; H, 9.69. Found: C, 66.69; H, 9.86. 131 3.1.17 (2£)-8-Tetrahydropyranyloxy-2-octen-1-ol (19) OTHP 19 A solution of ester 18 (9.76 g, 0.0361 moi) in 200 mL of CH 2 CI 2 was cooled to -78 °C before the addition of Dibal-H (1.0 M, 90 mL, 0.090 moi) via an addition funnel over 90 min. The reaction was stirred for an additional 2 h, then diluted with ethyl acetate (1 L) and slowly quenched with 1 M HCI (200 mL). The mixture was stirred at room temperature for 2 h. The two clear layers were separated followed by successive washes of the organic layer with 1 x 200 mL of 1 M HCI, 1 x 200 mL of saturated NaHCC>3 and 1 x 200 mL of brine. The organic layer was dried over MgSC»4, filtered and concentrated under reduced pressure. The crude oil was purified by flash chromatography with 5:1 petroleum ether/ethyl acetate to yield 7.84 g of 19 (95%) as a colourless oil. IR (neat): 3413 (br), 2937, 2866, 1448, 1360, 1205, 1124, 1084, 1026, 974, 916, 867 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.25-1.9 (m, 13H), 2.0 (m, 2H), 3.25-3.9 (m, 4H), 4.05 (m, 2H), 4.54 (t, J=3 Hz, 1H), 5.05 (m, 2H); 132 1 3 C NMR (50 MHz) 5: 19.7, 25.5, 25.8, 29.4, 30.7, 32.6, 33.0, 62.4, 64.8, 67.3, 99.5, 130.1, 133.8; LRMS (DCI, NH 4+ ) m/z: 246 (M ++18); HRMS Calcd for C13H24O3 . N H 4 (M ++18): 246.20691; found: 246.20627; Anal. Calcd for Ci 3H 2 40 3 : C, 68.4; H, 10.6. Found: C, 68.37; H, 10.49. 3.1.18 (2E)-8-Tetrahydropyranyloxy-2-octenal (20) O OTHP 20 To a cooled (-78 °C) solution of oxalyl chloride (3.30 mL, 0.0378 moi) in 85 mL of CH2CI2 was added DMSO (5.40 mL, 0.0756 moi) in 15 mL of CH 2 CI 2 over 15 min. The mixture was stirred for 5 min before dropwise addition via a cannula of 19 (7.84 g, 0.0344 moi) in 35 mL of CH 2 CI 2 over 20 min. After 15 min, Et 3N (24.0 mL, 0.171 moi) was added via a syringe. The ice bath was removed and the reaction was allowed to slowly warm to room temperature and stirred for an additional 3 h. The reaction mixture was quenched with water (80 mL) and extracted with 3 x 50 mL of CH2CI2. The organic layer was washed with brine, dried with MgS04, filtered and concentrated under reduced pressure to give a yellow liquid with a white precipitate. The crude material 133 was purified by flash chromatography using 4:1 petroleum ether diethyl ether to afford 6.61 g (85%) of 20 as a colourless oil. IR (neat): 2937, 2865, 2736, 1689, 1637, 1463, 1446, 1359, 1137, 1033, 985 cm" 1; 'H NMR (200 MHz, CDCI 3) 5:1.4-1.8 (m, 12H), 2.35 (dt, J=7.8, 6.9 Hz, 2H), 3.3-4.0 (m, 4H), 4.6 (m, 1H), 6.1 (dd, J=15.7, 7.8 Hz, 1H), 6.9 (dt, J=15.7, 6.9 Hz, 1H), 9.5 (d, J=7.8 Hz, 1H); 1 3 C NMR (50 MHz) 8: 19.7, 25.5, 25.8, 27.7, 29.4, 30.7, 32.6, 62.4, 67.3, 98.9, 133.0, 158.7,194.1; LRMS (DCI, NH 4+ ) m/z (relative intensity): 244 (M ++18, 79), 102 (100); HRMS Calcd for C13H22O3 NH 4 : 244.19126 (M++18); found: 244.19062; Anal. Calcd for C i 3 H 2 2 0 3 : C, 68.99; H, 9.80. Found: C, 68.93; H, 9.65. 3.1.19 (3£)-9-Tetrahydropyranyloxy-1,3-nonadiene (21a) 21a A suspension of methyltriphenylphosphonium bromide (22.4 g, 0.0628 moi) in 200 mL of THF was cooled to 0 °C before the addition of n-BuLi (1.39 M, 46.8 mL, 0.0651 moi) 134 via a syringe over 1 h. The mixture was stirred for an additional hour and more n-BuLi (1.20 M, 20.0 mL, 0.0240 moi) was added to give a clear, dark orange solution. Finally, aldehyde 20 in 50 mL of THF was added over 30 min and the reaction was stirred at 0 °C for 2 h. The reaction was quenched with 200 mL of water, and extracted with 4 x 75 mL of diethyl ether. The combined organic layers were washed with 1 x 100 mL dilute HCI and 1 x 100 mL brine, dried with MgSC»4, filtered and concentrated under reduced pressure to give a brown solid. The crude solid was taken up in petroleum ether and the resulting slurry was filtered through a silica plug and eluted with additional petroleum ether. The solvent was removed to afford 5.40 g (82%) of 21a as a colourless oil. IR (neat): 2937, 2863, 1654, 1608, 1463, 1434, 1359, 1141, 1083, 1033, 898, 879 cm' 1 ; 1 H NMR (200 MHz, CDCI 3) 5: 1.3-1.85 (m, 12H), 2.1 (dt, J=6.9, 6.3 Hz, 2H), 3.3-3.95 (m, 4H), 4.5 (t, J=3 Hz, 1H), 4.95 (dd, J=2, 10.2 Hz, 1H), 5.08 (dd, J=2, 16.9 Hz, 1H), 5.65 (dt, J=15.2, 6.8 Hz, 1H), 6.05 (dd, J=15.2, 10.2 Hz, 1H), 6.3 (ddd! J=16.8, 10.3, 10 Hz, 1H); 1 3 C NMR (50 MHz) 8: 19.7, 25.5, 25.7, 29.0, 29.5, 30.8, 32.4, 62.4, 67.5, 98.9, 114.7, 131.8, 135.3, 137.3; LRMS (DCI, NH 4) m/z (relative intensity): 242 (M++18, 100), 225 (M + +1, 19), 158 (5), 119 (11), 102 (100), 85 (67); HRMS Calcd for C14H24O2: 225.18545 (M ++1); found: 225.18527; Anal. Calcd for C14H24O2: C, 74.95; H, 10.78. Found: C, 74.79; H, 10.65. 135 3.1.20 (3Z)-9-Tetrahydropyranyloxy-1,3-nonadiene (21 b) OTHP 21b To a cooled solution (-78 °C) of allyldiphenylphosphine (1.36 g, 6.02 mmol) in THF (20 mL), was added t-BuLi (1.50 M, 4.00 mL, 6.00 mmol). The resulting red solution was stirred at 0 °C for 30 min before the addition of Ti(iPrO) 4 (1.80 mL, 6.00 mmol). After 10 min, aldehyde 24 (1.05 g, 5.25 mmol) in THF (10 mL) was added via cannula and the reaction was stirred for 1 h at 0 °C before the addition of Mel (0.40 mL, 6.0 mmol). The reaction mixture was stirred further for 1 h at 0 °C before warming to room temperature. After 2.5 h the reaction was complete by TLC analysis and it was quenched with water, resulting in an immediate white precipitate. The layers were separated and the organic layer was washed with 1 M HCI, saturated NaHCOa and brine, dried over MgSC»4, filtered and concentrated under reduced pressure. Purification by flash chromatography using 10:1 petroleum ether/diethyl ether afforded compound 21b (0.851 g, 76%) as a colourless oil. IR (neat): 2939, 2858, 1654, 1606, 1463, 1357, 1130, 1043, 1008, 902 cm" 1; *H NMR (200 MHz, CDCI 3) 8: 1.2-1.9 (m, 12H), 2.2 (m, 2H), 3.3-3.9 (m, 4H), 4.5 (t, J=3.7 Hz, 1H), 5.02 (d, J=10 Hz, 1H), 5.15 (dd, J=1.7, 16.8 Hz, 1H), 5.4 (dt, J=10.5, 7.6 Hz, 1H), 5.95 (dd, J=10, 10.7 Hz, 1H), 6.8 (ddd, J=10, 10, 17 Hz, 1H); 136 1 3 C NMR (50 MHz) 8: 19.7, 25.5, 25.8, 27.6, 29.4, 29.5, 30.7, 62.3, 67.5, 98.8, 116.7, 129.2, 132.3, 132.7; LRMS (El) m/z (relative intensity): 225 (M + +1, 12), 224 (1), 122 (3), 85 (100). 3.1.21 (6E)-6,8-Nonadien-1-ol (22a) To a solution of compound 21a (5.34 g, 0.0238 moi) in 200 mL of ethanol was added PPTs (1.81 g, 0.00722 moi) and the resulting mixture was gently heated to 70 °C for 3.5 h. The reaction mixture was cooled and concentrated under reduced pressure. The oily residue was taken up in diethyl ether and the insoluble PPTs was removed by filtration through a silica plug. Further purification by flash chromatography using 4:1 petroleum ether/diethyl ether gave 3.00 g (90%) of 22a as a colourless oil. IR (neat): 3353 (br), 3016, 2939, 2858, 1656, 1464, 1437, 1379, 1247, 1058, 999 c m 1 ; 1 H NMR (200 MHz, CDCI 3) 8: 1.3-1.7 (m, 7H), 2.1 (dt, J=6.8, 6.6 Hz, 2H), 3.6 (dt, J=5.4, 6.3 Hz, 2H), 4.95 (dd, J=1.2, 10 Hz, 1H), 5.08 (dd, J=1.2, 16.8 Hz, 1H), 5.65 (dt, J=15.1, 6.8 Hz, 1H), 6.05 (dd, J=10.3, 15.2 Hz, 1H), 6.3 (ddd, J=10, 10.1, 16.9 Hz, 1H); 22a 137 1 3 C NMR (50 MHz) 8: 25.2, 28.1,32.4, 32.6, 62.9, 114.7, 131.0, 135.1, 137.2; LRMS (El) m/z (relative intensity): 140 (8), 122 (35), 107 (19), 93 (38), 79 (100), 67 (93); HRMS Calcd for C 9 Hi 6 0:140.12012; found: 140.11957; Anal. Calcd for C 9 H i 6 0 : C, 77.09; H, 11.05. Found: C, 77.34; H, 11.28. 3.1.22 (6Z)-6,8-Nonadien-1-ol (22b) To a solution of compound 21b (0.854 g, 3.81 mmol) in 20 mL of methanol was added pTsA (1.82 g, 7.28 mmol) and the reaction mixture was stirred overnight. The solvent was removed under reduced pressure and the residue was taken up in diethyl ether and washed with saturated NaHCC»3 and brine, dried over MgSCM, filtered and concentrated under reduced pressure. Further purification by flash chromatography using 5:1 petroleum ether/diethyl ether gave 0.409 g (77%) of 22b as a colourless oil. IR (CDCI 3): 3620, 3460 (br), 3087, 3010, 2935, 2858, 1650, 1593, 1434, 1286, 1049, 22b 1006 cm 138 1 H NMR (200 MHz, CDCI 3) 8:1.32 (m, 5H), 1.55 (m, 2H), 2.2 (m, 2H), 3.6 (t, J=7 Hz, 2H), 4.95 (dd, J=1.2, 10 Hz, 1H), 5.1 (dd, J=1.2, 15 Hz, 1H), 5.4 (dt, J=10, 7 Hz, 1H), 6.0 (dd, J=10, 10 Hz, 1H), 6.6 (ddd, J=10, 10, 15 Hz, 1H); L R M S (El) m/z (relative intensity): 140 (19), 122 (55), 107 (21), 93 (55), 79 (100); H R M S Calcd for C 9 Hi 6 0:140.12012; found: 140.11971. 3.1.23 6-Tetrahydropyranyloxyhexan-1 -ol (23) Protection of hexanediol was carried out following a literature procedure. 7 1 To a solution of 1,6-hexanediol (1.49 g, 12.6 mmol), DHP (4.80 g, 57.2 mmol) in 50 mL of toluene, was added Dowex H + resin (2.52 g) at room temperature. After 3 h, the resin beads were filtered and the mixture was diluted with 50 mL of diethyl ether. The organic layer was washed with 2 x 50 mL of saturated NaHC03, 1 x 50 mL of brine, dried with MgSCu, filtered and concentrated under reduced pressure. Purification by flash chromatography with 3:1 petroleum ether/ethyl acetate resulted in 2.45 g (95%) of compound 23 as a colourless oil. IR (neat): 3419 (br), 2937, 2875, 1461, 1353, 1220, 1147, 1076, 1045 cm" 1; 1 H NMR (200 MHz, CDCI3) 8: 1.2-1.9 (m, 13H), 3.3-3.9 (m, 6H), 4.55 (t, J=3 Hz, 1H); 139 1 3 C NMR (50 MHz) 5: 19.6, 25.4, 25.5, 26.2, 29.6, 30.7, 32.6, 62.3, 62.7, 67.5, 98.8. 3.1.24 6-Tetrahydropyranyloxyhexanal (24) The alcohol 23 (3.88 g, 0.0192 moi) was oxidised under Swern conditions as described for compound 20. Purification by flash chromatography with 10:1 petroleum ether/ethyl acetate afforded 3.42 g (89%) of aldehyde 24 as a pale yellow oil. IR (neat): 2939, 2867, 2713, 1726, 1444, 1357, 1114, 1087, 1026, 908, 871 cm" 1; 1 H NMR (200 MHz, CDCI 3) 5: 1.2-1.9 (m, 12H), 2.4 (t, J=7 Hz, 2H), 3.2-3.9 (m, 4H), 4.55 (t, J=3 Hz, 1H), 9.7 (s, 1H); 1 3 C NMR (50 MHz) 8:19.5, 21.8, 25.3, 25.7, 29.3, 30.6, 43.7, 62.2, 67.1, 98.7, 202.5; LRMS (DCI, NH3) m/z (relative intensity): 218 (M ++18, 4), 201 (M + +1, 5), 200 (1), 183 (15), 134 (18), 99 (74), 85(100). 24 140 3.1.25 5 - D l p h e n o x y p h o s p h o r y M ,2 ,3 ,4 ,4a ,5 ,6 ,8a-oc tahydronaptha lene (25a/25b) 25a 25b A solution of triene 16a (0.288 g, 0.783 mmol) in 5 mL of toluene was heated in a sealed tube for 4 days at 170 °C. The solvent was removed under reduced pressure and residue was purified by radial chromatography to afford 0.252 g (88%) of adducts 25a:25b as a 2:1 mixture, by integration of the 1 H NMR spectrum. A small amount of each isomer was isolated by HPLC for further characterisation. Conditions for HPLC separation were as follows: 4 u. Novapak silica column, elution with 4% ethyl acetate in hexanes, flow rate = 2 mL/min, with detection by UV absorbence at X = 265 nm. The retention times of the isomers under these conditions were 25b at 9.46 min and 25a at 10.08 min with partial overlap of the two peaks. Isomer 25a (c/s-fused) I R (neat): 3066, 3018, 2925, 2852, 1593, 1492, 1444, 1270, 1215, 1191, 925, 763 cm" 1; 1 H N M R (400 MHz, CDCI 3) 6: 1.15-2.1 (m, 8H), 2.3-2.6 (m, 5H), 5.62 (m, 2H), 7.05-7.4 (m, 10H); 141 1 3 C NMR (50 MHz) 8: 23.4, 23.9, 26.5 (J=37 Hz), 28.7 (J=2.9 Hz), 30.6, 33.5 (J=3.4 Hz), 34.4 (J=19.8 Hz), 40.9 (J=4.3 Hz), 120.4, 120.4 (J=10 Hz), 123.5 (J=7.6 Hz), 124.7, 124.8, 129.5, 129.6, 131.9, 150.5 (J=4.2 Hz), 150.6 (J=4.8 Hz); 3 1 P NMR 8: 26.7; Isomer 2 5 b (transfused) 1 H NMR (400 MHz, CDCI 3) 8: 1-1.9 (m, 9H), 2.2-2.6 (m, 4H), 5.4-5.7 (m, 2H), 7.05-7.4 (m, 10H); 1 3 C NMR (50 MHz) 8: 26.4 (J=38 Hz), 27.2 (J=3.8 Hz), 30.6, 31.4 (J=3.3 Hz), 33.3 (J=2.8 Hz), 33.5, (J=3.4 Hz), 37.8 (J=10.1 Hz), 42.3 (J=15.7 Hz), 120.4, 120.4 (J=10 Hz), 123.8 (J=16.7 Hz), 124.7, 124.8, 129.5, 129.6, 132.5 (J=2.4 Hz), 150.3 (J=9.5 Hz), 150.6 (J=10 Hz); 3 1 P NMR 8: 25.5; Mixture of both isomers: L R M S (El) m/z (relative intensity): 368 (15), 235 (7), 134 (100), 91 (42), 77 (22); H R M S Calcd for C22H25O3P: 368.15414; found: 368.15509; Anal. Calcd for C22H25O3P: C, 71.72; H, 6.84. Found: C, 71.81; H, 6.71. 142 3.1.26 1-Diphenoxyphosphoryl c/'s-Decahydronapthalene (26a) 26a A solution of phosphonate 25a (38 mg, 0.10 mmol) in 5 mL of ethanol was treated with catalytic amounts of 5% Pd/C under a H2 atmosphere and stirred at room temperature. After 8 h, the reaction mixture was filtered through Celite and concentrated under reduced pressure to give 35 mg (92%) of the reduced product 26a as a colourless oil. IR (CDCI3): 3074, 3043, 2931, 2860, 1593, 1494, 1448, 1261, 1240, 1193, 1116, 1070, 1026, 1006, 962 cm' 1 ; 1H NMR (400 MHz, CDCI3) 8: 1.0-2.5 (m, 17H), 7.0-7.4 (m, 10H); 13C NMR (100 MHz) 8: 21.7 (5J=9.5 Hz), 23.8 (3J=28 Hz), 26.4 ( 3J=37 Hz), 27.7 (4J=3.4 Hz), 28.9 (brs), 29.2 (2J=9.4 Hz), 33.1 (J=155 Hz), 34.7 ( 3J=32 Hz), 36.2 (s), 43.4 (2J=0.8 Hz), 120.5, 124.6, 129.6, 150.5; 31P NMR 8: 27.0; LRMS (El) m/z (relative intensity): 370 (100), 261 (36), 248 (31), 167 (30), 157 (31), 95 (20), 94 (98), 77 (36); HRMS Calcd for C2 2H 270 3P: 370.16977; found: 370.16914. 143 3.1.27 1-Diphenoxyphosphoryl frans-Decahydronapthalene (26b) 26b A solution of phosphonate 25a (17 mg, 0.046 mmol) was hydrogenated following the procedure outlined for compound 26a to give 15 mg (88%) of the reduced product 26b as a colourless oil. IR (CDCIa): 2927, 2856, 1593, 1492, 1446, 1265, 1213, 1195, 1162, 1026 cm" 1; 1 H NMR (400 MHz, CDCI3) 5: 1.0-1.9 (m, 15H), 2.15-2.3 (m, 1H), 2.42 (m, 1H), 7.0-7.2 (m, 10H); 1 3 C NMR (100 MHz) 8: 26. 1 (3J=16 Hz), 26.3 (3J=38 Hz), 27.5 (4J=4.7 Hz), 32 (5J=1.4 Hz), 33.8 (s), 34.5 (4J=2.2 Hz), 42. 2 ( 1J=134 Hz), 43. 2 (2J=8.6 Hz), 43.3 ( 3J=12 Hz), 44.8 (2J=1.1), 120.5, 124.7, 129.6, 150.5 3 1 P NMR 8: 26.2; LRMS (El) m/z (relative intensity): 370 (100), 261 (17), 248 (15), 157 (25), 94 (12); HRMS Calcd for C22H27O3P: 370.16977; found: 370.16917. 3.1.28 Benzyl Ester of 4'-Hydroxyglutarinil ic acid (29) 144 O 29 To a solution of glutaric anhydride (3.39 g, 29.8 mmol) and benzyl alcohol (3.20 g, 29.6 mmol) in 60 mL of CH 2 CI 2 was added Et 3N (4.20 mL, 30.3 mmol). After 4 h the reaction was quenched with 1 M HCI and extracted with CH 2 CI 2 . The combined organic extracts were washed with brine, dried over MgS04, filtered and concentrated under reduced pressure to give 6.34 g (96%) of the monobenzyl protected glutaric acid. The crude acid was carried on without further purification. Aminophenol (1.51 g, 13.9 mmol) was added to the acid prepared above (3.00 g, 13.5 mmol) and dissolved in THF (80 mL). To the resulting clear, brown solution was added DCC (2.85 g, 13.9 mmol) and DMAP (catalytic amount). The reaction flask was covered in foil and stirred at room temperature for 12 h. The precipitate was filtered off and the filtrate was concentrated under reduced pressure. Purification by flash chromatography using 6:1 CH 2CI 2/ethyl acetate afforded 2.21 g (52%) of compound 29 as a light brown solid. Further purification on a small amount by recrystallisation from hot CHCIa and hexanes gave an elementally pure sample of the linker as an off-white solid powder. IR (CDCI3): 3178, 3068, 3035, 2954, 2885, 1737, 1722, 1454, 1415, 1317, 1243, 1209, 1153 cm' 1 ; 145 1 H NMR (200 MHz, CD 3OD) 8: 2.0 (m, 2H), 2.2-2.4 (m, 4H), 5.1 (s, 2H), 6.7 (d, 2H), 7.2 (d, 2H), 7.25 (s, 5H); 1 3 C NMR (50 MHz) 8: 35.2, 48.2, 49.5, 66.3, 115, 122.0, 128.0, 128.2, 128.5, 130.2, 135.6, 153.5, 171.1, 173.5; LRMS (El) m/z (relative intensity): 313 (5), 224 (38), 143 (25), 109 (30), 99 (33), 91 (15), 56 (100); HRMS Calcd for C i 8 H i 9 N 0 4 : 313.13141; found: 313.13127; Anal. Calcd for C i 8 H i 9 N 0 4 : C, 68.99; H, 6.11, N, 4.47. Found: C, 68.74; H, 6.13, N, 4.63. 3.1.29 5-Methoxyphenoxyphosphoryl-1,2,3,4,4a,5,6,8a-octahydronapthalene (30), 5-Dimethoxyphosphoryl-1,2,3,4,4a,5,6,8a-octahydronapthalene (31) 30 31 To a cooled solution (-78 °C) of methanol (0.20 mL, 4.8 mmol) in 30 mL THF, was added n-BuLi (1.35 M, 3.50 mL, 4.75 mmol). After 20 min the phosphonate isomers 25a/25b (1.59 g, 4.32 mmol) in 15 mL of THF was added and the reaction was allowed to continue at room temperature for 12 h. The reaction was quenched with water and extracted with 3 x 30 mL of diethyl ether. The combined organic extracts were washed with 1 x 30 mL of brine, dried with M g S C u , filtered and concentrated under reduced pressure. Purification by flash chromatography using 5:1 petroleum ether/ethyl acetate afforded 0.67 g (51%) of compound 30 and 0.17 g (16%) of phosphonate 31 as colourless oils. Analysis of the 3 1 P NMR spectra showed that diastereomers of each compound were formed in a 2:1 ratio. Compound 30 IR (neat): 3066, 3018, 2925, 2854, 1593, 1490, 1454, 1251, 1215, 1195, 1162, 1072, 1026, 923, 765 cm" 1; 1 H NMR (200 MHz, CDCI3) 8: 1.25-2 (m, 9H), 2.2-2.55 (m, 4H), 3.78 (d, J=11 Hz, 3H), 5.37-5.77 (m, 2H), 7.1-7.4 (m, 5H); 1 3 C NMR (50 MHz) 8: 23.4, 23.6, 23.9, 28.5, 28.6, 28.8, 28.9, 30.6, 30.7, 32.0, 33.4, 33.5, 33.6, 34.4, 52.8, 52.9, 120.3, 120.4, 122.9, 123.6, 123.7, 124.0, 129.6, 131.7, 131.8, 150.1; 3 1 P NMR 8: 30..4 and 30.5 (minor), 31.6 and 31.7(major); LRMS (El) m/z (relative intensity): 306 (10), 173 (8), 134 (100), 105 (15), 91 (51), 77 (24); HRMS Calcd for C17H23O3P: 306.13849; found: 306.13834; Anal. Calcd for C17H23O3P: C, 66.65; H, 7.57. Found: C, 66.64; H, 7.8. 147 Compound 31 IR (neat): 3018, 2927, 2854, 1591, 1490, 1454, 1247, 1209, 1162, 1043, 923, 800 cm" 1; 1 H NMR (200 MHz, CDCI 3) 5: 1.1-1.8 (m, 8H), 2.0-2.5 (m, 5H), 3.7 (d, J=10.5 Hz, 6H), 5.6 (m, 2H); 1 3 C NMR (50 MHz) 8: 23.3, 23.6, 26.2, 26.6, 28.4, 28.6, 30.4, 31.0, 33.3, 34.2, 35.4, 38.2, 40.5, 40.6, 42.0, 42.3, 52.0, 52.2, 123.9, 124.1, 131.6, 132.3; 3 1 P NMR 8: 34.8 (minor), 35.9 (major); LRMS (El) m/z (relative intensity): 244 (26), 134 (100), 119 (15), 111 (28), 91 (75), 79 (48); HRMS Calcd for C i 2 H 2 i 0 3 P : 244.12283; found: 244.12232; Anal. Calcd for C i 2 H 2 i 0 3 P : C, 59.01; H, 8.67. Found: C, 59.18; H, 8.88. 3.1.30 5-HydroxymethoxyphosphoryM ,2,3,4,4a,5,6,8a-octahydronapthalene (32) A solution of phosphonate 30 (0.667 g, 2.18 mmol) in 5 mL of THF and 20 mL of 3 M KOH was refluxed for 24 h. The reaction was acidified with concentrated HCI and 148 extracted with diethyl ether. The combined ether extracts were dried over M g S C u , filtered and concentrated under reduced pressure. Purification by radial chromatography using 5:2 petroleum ether/ethyl acetate, followed by treatment of the acid with Dowex H + resin, afforded 0.316 g (63%) of acid 32 as a yellow oil. Alternatively, to a solution of phosphonate 31 (130 mg, 0.53 mmol) in 10 mL of CH2CI2 was added TMS-Br (0.12 mL, 0.80 mmol). After refluxing for 6 h, the reaction mixture was cooled and quenched with 1 M HCI, then extracted with diethyl ether. The combined ether extracts were washed with brine, dried over MgSO-t, filtered and concentrated under reduced pressure. Purification as above gave 70 mg (58%) of acid 32. IR (CDCI3): 3380, 2937, 2854, 1667, 1450, 1203, 1027 cm' 1 ; 1 H NMR (400 MHz, CDCI3) 6: 1.0-1.95 (m, 9H), 2.05-2.45 (m, 4H), 3.7 (d, J=10 Hz, 3H), 5.55 (m, 2H), 9.15 (br s, 1H); 3 1 P NMR 8: 37.0 (minor), 37.9 (major); LRMS (El) m/z (relative intensity): 230(14), 134 (100), 91 (90), 74 (38); HRMS Calcd for C11H19O3P: 230.10718; found: 230.10700; Anal. Calcd for C11H19O3P: C, 57.38; H, 8.32. Found: C, 56.98; H, 8.19. 149 3.1.31 5-[Benzyl N-Glutarate-(methoxy-p-aminophenoxyphosphoryl)] 1,2,3,4,4a,5,6,8a-octahydronapthalene (33) 33 To a solution of acid 32 (0.88 mg, 0.38 mmol) in 5 mL of CH2CI2 was added oxalyl chloride (0.32 mL, 3.8 mmol). After 12 h, the solvent was removed under reduced pressure and the residue was dried under vacuum for 2 h. The resulting acid chloride was combined with compound 29 and dissolved in 7 mL of CH2CI2 and Et3N (0.13 mL, 0.94 mmol) to give a clear, brown solution. The mixture was refluxed for 10 h, then quenched with water and extracted with ethyl acetate. The combined organic extracts were washed with brine, dried over MgSG*4, filtered and concentrated under reduced pressure. Purification by radial chromatography using 5:1 CH^CVethyl acetate afforded 65.2 mg (33%) of phosphonate 33 as a pale brown oil. IR (neat): 3272, 2929, 2854, 1733, 1681, 1608, 1542, 1506, 1454, 1407, 1305, 1238, 1203, 1047, 919, 838, 732 cm" 1; 1 H NMR (200 MHz, CDCI 3) 5: 1.2-2.0 (m, 8H), 2.05 (m, 2H), 2.1-2.5 (m, 9H), 3.7 (d, J=10.8 Hz, 3H), 5.1 (s, 2H), 5.6 (m, 2H), 7.0 (d, 2H), 7.26 (s, 5H), 7.3 (d, 2H), 8.15 (brs, 1H); 3 1 P NMR 8: 31.0 (minor), 32.1 (major); 150 LRMS (El) m/z (relative intensity): 525 (3), 391 (2), 179 (8), 134 (30), 109 (36), 91 (100); HRMS Calcd for CzgHaeNOeP: 525.22803; found: 525.22826; Anal. Calcd for CjsHaeNOeP: C, 66.27; H, 6.9, N, 2.66. Found: C, 65.99; H, 7.17, N, 2.58. 3.1.32 5-[Benzyl N-Glutarate-(hydroxy-p-aminophenoxyphosphoryl)]-1,2,3,4,4a,5,6,8a-octahydronapthalene (34) 34 To a solution of compound 33 (65.2 mg, 0.124 mmol) in 5 mL of CH2CI2 was added TMS-Br (0.17 mL, 1.2 mmol) and the reaction mixture was refluxed for 5 h. The reaction was quenched with water and HCI, then extracted with CH2CI2. The combined organic extracts were washed with brine, dried with MgSC>4, filtered and concentrated under reduced pressure to give a white solid. The solid was dissolved in methanol and treated with Dowex H + resin, filtered and concentrated to yield 49 mg (77%) of acid 34 as a brown, viscous oil. 151 IR (CDCIa): 3330, 3024, 2931, 2856, 1731, 1687, 1608, 1537, 1512, 1444, 1409, 1265, 1228, 1205, 1151,1016, 979, 838 cnf 1 ; 1 H NMR (200 MHz, CDCI3) 5: 1.3-2.0 (m, 13H), 2.05 (m, 2H), 2.35 (m, 2H), 2.45 (m, 2H), 5.1 (s, 2H), 5.5 (m, 2H), 6.95 (d, 2H), 7.27 (s, 5H), 7.3 (d, 2H), 7.7 (br s, 1H), 8.2 (brs, 1H); 3 1 P NMR 8: 32.9 (minor), 33.9 (major); LRMS (LSIMS) m/z (relative intensity): 510 (M + -1, 0.4), 312 (1), 222 (100); HRMS Calcd for C 2 8H34N0 6P: 510.20456 (M+-1); found: 510.20263; Anal. Calcd for C 2 8 H 3 4 N O 6 P : C, 65.74; H, 6.7, N, 2.74. Found: C, 65.78; H, 6.83, N, 2.75. 3.1.33 5-Hydroxyphenoxyphosphoryl-1,2,3,4,4a,5,6,8a-octahydronapthalene (35) A solution of the phosphonates 25a/25b (200 mg, 0.543 mmol) in THF (5 mL) and 20 mL of 3 M KOH was refluxed for 24 h. The reaction was acidified with concentrated HCI and extracted with diethyl ether. The combined ether extracts were dried over MgS04, filtered and concentrated under reduced pressure. Purification by radial 152 chromatography with 5:1 diethyl ether/methanol gave 108 mg (68%) of the acid 3 5 as a colourless oil. IR (neat): 3022, 2929, 2856, 1593, 1492, 1446, 1207, 1162, 1026, 981 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.0-2.4 (m, 13H), 5.4 (m, 2H), 7.1 (m, 5H), 10.2 (br s, 1H); 1 3 C NMR (75 MHz) 8: 23.5, 23.9, 26.4, 26.6, 26.9, 28.7, 28.8, 30.5, 31.2, 31.7, 33.4, 33.7, 34.7, 40.7, 42.1, 42.3, 120.7, 123.6, 123.7, 124.6, 129.5, 131.8, 132.4, 150.1, 150.2; 3 1 P NMR 8: 32.0 (minor), 33.1 (major); LRMS (El) m/z (relative intensity): 292 (35), 134 (100), 91 (33), 77 (8); HRMS Calcd for C i 6 H 2 i 0 3 P : 292.12254; found: 292.12283; Anal. Calcd for C i 6 H 2 i 0 3 P : C, 65.74; H, 7.27. Found: C, 65.48; H, 7.27. 153 3.1.34 (2£,8£)-2,8,10-Undecatrienoic Acid (36a) 36a A solution of ester 12a (0.802 g, 3.85 mmol) in 5 mL of THF and 15 mL of 3 M KOH was refluxed for 2 days. The reaction was cooled in an ice bath and acidified slowly with concentrated HCI. The mixture was extracted with 4 x 10 mL of diethyl ether and then the combined organic extracts were washed with brine, dried over MgS04, filtered and concentrated under reduced pressure. Purification by flash chromatography afforded 0.588 g (85%) of the acid 36a as a colourless oil. IR (neat): 3091, 3016, 2937, 2854, 1697, 1650, 1417, 1292, 1226, 1006, 983, 950, 902 cm" 1; 1 H NMR (200 MHz, CDCI3) 8: 1.2-1.5 (m, 4H), 1.85-2 (m, 2H), 2-2.2 (m, 2H), 4.94 (dd, J=1.5, 10 Hz, 1H), 5 (dd, J=1.5, 15 Hz, 1H), 5.5 (dt, J=15, 7 Hz, 1H), 5.7 (dt, J=15, 1.2 Hz, 1H), 5.9 (dd, J=10, 15 Hz, 1H), 6.2 (ddd, J=10, 10, 15 Hz, 1H), 6.95 (dt, J=15, 7 Hz, 1H), 11.0 (br.s, 1H); 1 3 C NMR (50 MHz) 8: 27.3, 28.6, 32.1, 32.2, 114.9, 120.7, 131.3, 134.6, 137.1, 152.1, 172.1; 154 3.1.35 (2E,82)-2,8,10-Undecatrienoic Acid (36b) C 0 2 H 36b Hydrolysis of ester 12b (120 mg, 0.576 mmol) was carried out as described for 36a to give 79.8 mg (77%) of 36b. IR (neat): 3087, 3020, 2941, 2860, 1703, 1658, 1610, 1431, 1350, 1288 cnf 1 ; 1 H NMR (200 MHz, CDCI 3) 5: 1.3-1.6 (m, 4H), 2-2.3 (m, 4H), 4.98 (dd, J=1.2, 10 Hz, 1H), 5.08 (dd, J=1.3, 17 Hz, 1H), 5.4 (dt, J=10, 7 Hz, 1H), 5.8 (dt, J=15, 1.4 Hz, 1H), 5.95 (dd, J=10, 10 Hz, 1H), 6.6 (ddd, J=10, 10, 17 Hz, 1H), 7.05 (dt, J=15, 7 Hz, 1H). 3.1.36 (2Z,8£)-2,8,10-Undecatrienoic Acid (36c), (2Z,8Z)-2,8,10-Undecatrienoic Acid (36d) 36d Hydrolysis of the methyl esters 4c and 4d (200 mg, 1.03 mmol) was carried out as described for 36a to afford 140 mg (75%) of a mixture of acids 36c and 36d, in a 1:1 ratio. Assignment of the 1 H NMR resonances for each isomer was based on a 155 combination of coupling constants and by comparison to the spectra from the assigned methyl and ethyl esters 4 and 1 2 , respectively. Isomer 36c 1 H NMR (200 MHz, CDCI 3) 8: 1.3-1.6 (m, 4H), 2.1 (m, 2H), 2.6 (m, 2H), 4.95 (dd, J=1.7, 10 Hz, 1H), 5.08 (dd, J=1.7, 16.9 Hz, 1H), 5.7 (dt, J=15.1, 6.8 Hz, 1H), 5.78 (dt, J=11.8, 1.4 Hz, 1H), 6.05 (dd, J=11, 16 Hz, 1H), 6.3 (ddd, J=10.2, 10, 16.9 Hz, 1H), 6.34 (dt, J=11.5, 7.6 Hz, 1H); 1 3 C NMR (50 MHz) 8: 28.4, 28.7, 29.0, 32.2, 114.8, 119.2, 131.2, 134.9, 137.2, 153.2, 171.8. Isomer 36d 1 H NMR (200 MHz, CDCI 3) 8: 1.3-1.6 (m, 4H), 2.2 (m, 2H), 2.6 (m, 2H), 4.95 (dd, J=1.7, 10 Hz, 1H), 5.08 (dd, J=1.7, 16.9 Hz, 1H), 5.4 (dt, J=10.3, 8.3 Hz, 1H), 5.78 (dt, J=11.8, 1.4 Hz, 1H), 5.99 (dd, J=8, 11 Hz, 1H), 6.34 (dt, J=11.5, 7.6 Hz, 1H), 6.6 (ddd, J=10.3, 10, 16.8 Hz, 1H); 1 3 C NMR (50 MHz) 8: 27.4, 28.4, 29.0, 29.1, 116.9, 119.2, 129.5, 132.2, 132.4, 153.2, 171.8. Mixture of all isomers (36a/36b/36c/36d): IR (CDCI3): 3058, 2935, 2866, 1695, 1640, 1526, 1440, 1294, 1241, 1003, 902, cm" 1; LRMS (El) m/z (relative intensity): 180 (21), 151 (6), 135 (87), 120 (15), 107 (10), 95 (46), 81 (61), 67 (100); 156 H R M S Calcd for CnHi 6 0 2 :180.11504; found: 180.11471; Anal. Calcd for C n H i 6 0 2 : C, 73.3; H, 8.95. Found: C, 73.46; H, 8.99. 3.1.37 8-Tetrahydropyranyloxyoctan-1-ol (37) Monoprotection of 1,8-octanediol (3.54 g, 24.2 mmol) was carried out as outlined in the procedure for alcohol 23. Purification by flash chromatography with 3:1 petroleum ether/ethyl acetate resulted in 5.48 g (98%) of compound 37 as a colourless oil. I R (neat): 3406 (br), 2933, 2858, 1452, 1357, 1200, 1120, 1074, 1042, 983 cm" 1; 1 H N M R (200 MHz, CDCI 3) 6:1.2-1.9 (m, 19H), 3.3-3.9 (m, 6H), 4.55 (t, J=3 Hz, 1H). 157 3.1.38 8-Tetrahydropyranyloxyoctanal (38) 38 The alcohol 37 (5.45 g, 23.7 mmol) was oxidised under Swern conditions as outlined for compound 20. Purification by flash chromatography using 4:1 petroleum ether/ethyl acetate afforded 4.45 g (90%) of aldehyde 38 as a pale yellow oil. IR (neat): 2939, 2858, 2723, 1730, 1460, 1359, 1120, 1089, 1035 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.2-1.9 (m, 16H), 2.4 (dt, J=2, 7.5 Hz, 2H), 3.3-3.9 (m, 4H), 4.55 (t, J=3 Hz, 1H), 9.7 (d, J=2 Hz, 1H); 1 3 C NMR (50 MHz) 8: 19.5, 21.8, 25.3, 25.7, 29.1, 29.3, 29.6, 30.6, 43.7, 62.2, 67.1, 98.7, 202.5. 158 3.1.39 9-Tetrahydropyranyloxy-1-nonene (39) 39 The Wittig reaction with aldehyde 38 (3.71 g, 16.3 mmol) was carried out following the procedure outlined for compound 21a to afford 3.06 g (83%) of alkene 39 as a colourless oil. IR (neat): 3078, 2939, 2858, 1645, 1448, 1359, 1201, 1132, 1076, 1031, 987, 918 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8:1.2-1.45 (m, 8H), 1.45-1.9 (m, 8H), 2.0 (dt, J=5.8, 6.6 Hz, 2H), 3.3-3.9 (m, 4H), 4.55 (t, J=3 Hz, 1H), 4.9 (ddt, J=2.2, 10.3, 2 Hz, 1H), 4.97 (ddt, J=2, 17, 2 Hz, 1H), 5.78 (ddt, J=10.3, 17.1, 6.6 Hz, 1H); 1 3 C NMR (50 MHz) 8: 19.7, 25.5, 26.2, 28.9, 29.0, 29.3, 29.7, 30.8, 33.7, 63.4, 67.7, 98.8, 114.1, 139.2; LRMS (DCI, NH 3) m/z (relative intensity): 244 (M++18, 1), 227 (M + +1, 9), 185 (4), 169 (4), 101 (34), 85 (100). 159 3.1.40 8-Nonen-1-ol (40) 40 To a solution of compound 39 (3.06 g, 13.5 mmol) in 80 mL of methanol was added pTsA (0.50 g, 2.0 mmol). After 8 h, the solvent was removed under reduced pressure and the residue was taken up in diethyl ether and washed with saturated NaHC03 and brine, dried over MgSC-4, filtered and concentrated under reduced pressure. Further purification by flash chromatography using 4:1 petroleum ether/diethyl ether gave 1.47 g (77%) of alcohol 40 as a colourless oil. IR (neat): 3300 (br), 2938, 2858, 1645, 1450, 1180, 1031, 918, 1076 cm' 1 ; 1 H NMR (200 MHz, CDCI 3) 8 1.2-1.45 (m, 9H), 1.8 (m, 2H), 2.05 (dt, J=5.8, 6.6 Hz, 2H), 3.1 (t, J=7.3 Hz, 2H), 4.9 (ddt, J=2.2, 10.3, 2 Hz, 1H), 4.97 (ddt, J=2.0, 17, 2 Hz, 1H), 5.78 (ddt, J=10.3, 17.1, 6.6 Hz, 1H); 1 3 C NMR (50 MHz) 8: 25.6, 28.8, 29.0, 29.2, 32.7, 33.7, 63.4, 114.1, 139.1. 160 3.1.41 8-Nonenal (41) The Swern oxidation of alcohol 40 (1.19 g, 8.38 mmol) was carried out as outlined for compound 20, followed by purification on silica using 4:1 petroleum ether/diethyl ether to afford 0.88 g (75%) of aldehyde 41 as a pale yellow oil. IR (neat): 3076, 2929, 856, 2733, 1726, 1649, 1465, 1411, 1390, 993, 910 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.2-1.4 (m, 6H), 1.6 (m, 2H), 2.05 (dt, J=5.8, 6.6 Hz, 2H), 2.3 (dt, J=7.3, 1.3 Hz, 2H), 4.9 (ddt, J=2.2, 10.3, 2 Hz, 1H), 4.97 (ddt, J=2, 17, 2 Hz, 1H), 5.78 (ddt, J=10.3, 17.1,6.6 Hz, 1H), 9.8 (s, 1H); 1 3 C NMR (50 MHz) 8: 21.9, 28.6, 28.7, 28.9, 33.6, 43.8, 114.2, 138.8, 202.7. 161 3.1.42 Ethyl (2E)-2,10-Undecadienoate (42) O 42 A Wittig reaction was carried out with aldehyde 41 (0.88 g, 6.3 mmol) and triethyl phosphonoacetate as outlined for compound 12a. Purification by flash chromatography using 4:1 petroleum ether/ethyl acetate resulted in 1.15 g (87%) of compound 42 as a colourless oil. IR (neat): 3076, 2979, 2927, 2856, 1722, 1654, 1463, 1444, 1367, 1315, 1267, 1182, 1043, 983, 910 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.3 (t, J=7 Hz, 3H), 1.2-1.6 (m, 8H), 2.1 (m, 2H), 2.2 (m, 2H), 4.2 (q, J=7 Hz, 2H), 4.9 (ddt, J=2.2, 10.3, 2 Hz, 1H), 4.97 (ddt, J=2.0, 17, 2 Hz, 1H), 5.78 (ddt, J=10.3, 17.1, 6.6 Hz, 1H), 5.8 (dt, J=15.7, 1.2 Hz, 1H), 6.95 (dt, J=15.6, 7 Hz, 1H). 162 3.1.43 (2£)-2,10-Undecadienoic Acid (43) 43 O The ester 42 (1.15 g, 5.48 mmol) was hydrolysed according to the procedure outlined for compound 36a. Purification by flash chromatography using 4:1 petroleum ether/diethyl ether afforded 0.75 g (75%) of the acid 43 as a colourless oil. IR (neat): 3076, 2977, 2929, 2856, 1697, 1651, 1421, 1309, 1284, 4234, 985, 910 cm" 1; 1 H NMR (200 MHz, CDCI 3) 8: 1.2-1.6 (m, 8H), 1.95-2.1 (m, 2H), 2.2 (m, 2H), 4.9 (ddt, J=2.2, 10.3, 2 Hz, 1H), 4.97 (ddt, J=2.0, 17, 2 Hz, 1H), 5.79 (ddt, J=10.3, 16.9, 6.6 Hz, 1H), 5.8 (dt, J=15.6, 1.7 Hz, 1H), 7.1 (dt, J=15.7, 7 Hz, 1H), 9.2 (br s, 1H); 1 3 C NMR (50 MHz) 8: 27.8, 28.7, 28.8, 28.9, 32.3, 33.7, 114.3, 120.6, 138.9, 152.4, 172.1; LRMS (El) m/z (relative intensity): 182 (1), 164 (2), 153 (5), 139 (8), 122 (34), 95 (45), 81 (66), 67(62), 55(100); HRMS Calcd fo rCnHi 8 0 2 :182 .13068; found: 182.13129. 163 3.1.44 p-Nitrophenyl (2£)-2,10-Undecadienoate (44) 44 O The esterification of acid 43 (100 mg, 0.549 mmol) was carried out as outlined for compound 13a to afford 122 mg (73%) of compound 44 as a pale yellow oil. IR (neat): 3078, 2929, 2856, 1743, 1650, 1593, 1525, 1490, 1346, 1211, 1161, 1135, 1116, 968, 912, 858, 730 cm" 1; 1 H NMR (200 MHz, CDCI3) 8: 1.35-1.7 (m, 8H), 2.1 (m, 2H), 2.42 (m, 2H), 4.95 (dd, J=1.5, 10 Hz, 1H), 5.08 (dd, J=1.7, 16.9 Hz, 1H), 5.78 (m, 1H), 6.05 (dd, J=1.7, 15 Hz, 1H), 7.2 (m, 1H), 7.3 (m, 2H), 8.3 (m, 2H); 1 3 C NMR (50 MHz) 8: 27.8, 28.7, 28.8, 28.9, 32.5, 33.6, 114.3, 119.7, 122.4, 125.1, 138.9, 145.1, 153.6, 155.6, 163.8; LRMS (El) m/z (relative intensity): 303 (0.2), 273 (0.4), 165 (100), 147 (5), 123 (5), 109 (5), 95 (25), 81 (33), 67 (19), 55 (40); HRMS Calcd for C17H21NO4: 303.14706; found: 303.14817; Anal Calcd for C i 7 H 2 i N 0 4 : C , 67.29; H, 6.98; N, 4.62. Found: C , 67.38; H, 7.14; N, 4.61. 164 3.1.45 1,2,3,4,4a,5,6,8a-Octahydronapthalene-5-carboxylic acid (45) Methyl esters 5a/5b (35.2 mg, 0.181 mmol) were hydrolysed following the procedure outlined for compound 36a. The oily residue was purified by radial chromatography using 1:1 petroleum ether/diethyl ether to give a mixture of acid isomers 45 (24.3 mg, 74%) as colourless oils. IR (CDCIa): 2977, 2939, 2852, 1700, 1442, 1110 cm" 1; *H NMR (200 MHz, CDCI3) 5: 1.15-1.75 (m, 8H), 2.1-2.35 (m, 3H), 2.55 (m, 1H), 2.7-2.78 (m, 1H), 5.4 (d, J=10.1 Hz, 1H), 5.62-5.68 (m, 1H); LRMS (El) m/z (relative intensity): 180 (10), 162 (17), 135 (100), 119 (14), 107 (21), 93 (51), 91 (57), 79 (59), 67(41); HRMS Calcd for CnHi 6 0 2 :180.11504; found: 180.11477. 45 165 3.1.46 p-Nitrophenyl 1,2,3,4,4aa,5,6,8aa-Octahydronapthalene-5o> carboxylate (46a), p-Nitrophenyl 1,2,3,4,4aB,5,6,8ao> Octahydronapthalene-5B-carboxylate (46b) COgPNP H, COgPNP 46a 46b A solution of triene 13a (32.7 mg, 0.109 mmol) in 1 mL of toluene was heated to 150 °C in a sealed tube for 4 days. The solvent was removed under reduced pressure and purification by radial chromatography afforded 29.8 mg (91%) of cycloadducts 46a/46b as a pale yellow solid. The isomers were formed in a 1:1 mixture as determined by GC analysis. IR (neat): 3018, 2929, 2862, 1764, 1598, 1533, 1496, 1348, 1209, 1112, 871 cm" 1; 1 H NMR (200 MHz, CDCI3) 5: 1.0-2.0 (m, 10H), 2.0-2.8 (m, 8H), 5.5 (m, 2H), 7.25 (d, 2H), 8.25 (d, 2H); LRMS (El) m/z (relative intensity): 301 (2), 195 (4), 163 (69), 135 (100), 93 (16), 67 HRMS Calcd for C17H19NO4: 301.13141; found: 301.13105; Anal. Calcd for C17H19NO4: C, 67.76; H, 6.36; N, 4.65. Found: C, 67.74; H, 6.19; (10); N, 4.50. 166 3.1.47 7-Oxy-4'-methylcoumaryl 1,2,3,4,4aa,5,6,8aa-Octahydronapthalene-5a-carboxylate (47a), 7-Oxy-4'-methylcoumaryl 1,2,3,4,4aB,5,6,8ao> Octahydronapthalene-5p-carboxylate (47b) A solution of triene 14 (22.2 mg, 0.0657 mmol) in 1 mL of toluene (not readily soluble at room temperature) was heated to 150 °C in a sealed tube for 4 days. The solvent was removed under reduced pressure and purification by radial chromatography afforded 10.8 mg (49%) of the cycloadducts 47a and 47b as a white solid. The isomers were formed in a 1:1 mixture by GC analysis. IR (CDCI 3): 2927, 2858, 1764, 1737, 1618, 1450, 1390, 1269, 1135 cm"1; 1 H NMR (200 MHz, CDCI 3) 5: 1.1-2.0 (m, 10H), 2.2-2.8 (m, 3H), 2.4 (s, 3H), 5.6 (m, 2H), 6.3 (s, 1H), 7.05 (d, 1H), 7.1 (s, 1H), 7.6 (d, 1H); LRMS (El) m/z (relative intensity): 338 (6), 177 (34), 163 (29), 135 (100), 107 (4), 93 (11), 67 (5); HRMS Calcd for C21H22O4: 338.15181; found: 338.15092. 167 3.2 GENERAL BIOLOGICAL METHODS Deionized water was used in the preparation of all buffers. Keyhole limpet hemocyanin (KLH) and bovine serum albumin (BSA) were supplied by the Sigma Chemical Co. Slide-A-Lyzers and KLH used for the photochemical couplings were supplied by Pierce Chemicals. Porcine lipase type II crude used in kinetic assays was supplied by Sigma Chemical Co. ELISA assays were performed using Falcon 3911 Microtest III Flexible Assay Plates supplied by Becton Dickinson Labware. The absorbance was measured using an EL 309 Microplate Autoreader reading at X = 405 and 490 nm. Rabbits designated as H54, H55, H60 and H61 were housed at the Animal Care Facility at UBC. All immunisations and bleeds were carried out by the animal care technicians at the facility. Immunisations, antibody purification, hybridoma fusion procedures and cell tissue culture work were carried out following the protocols outlined by Ms. Helen Merkens and Mr. Michael Williams in Dr. Hermann Ziltener's lab of the Biomedical Research Centre at UBC. 168 MEDIA COMPONENTS Citrate Buffer 1.29 g citric acid/100 mL DDW 0.077 M Na 2 HP0 4 DMEM High Glucose 1 M Hepes buffer DMEM 4.5 mg/mL glucose 0.37% HC0 3 " 2 mM L-glutamine ELISA Blocking Solution 0.5% skim powder (w/v) 1 xPBS Freeze Media 90% FCS 10% DMSO Fusion Media RPMI 1640 2 mM L-glutamine 50 u.M 2B-mercaptoethanol 2% 10 x 3T3 conditioned medium 100 u.g/mL streptomycin 100 Units/mL penicillin HT Media Fusion Media 20% FCS 0.1 mM hypoxanthine 16 pM thymidine Hybridoma Media Fusion Media 10% FCS Mouse Red Cell Removal 0.15 M NH4CI Buffer (MRCRB) 0.015 M Tris-HCl Myeloma Growth Media DMEM 20% heat inactivated FCS PCFIA Buffer 1 xPBS 2% FCS 0.2% sodium azide PEG solution 50% (w/w) polyethylene glycol (Merck PEG 4000) DMEM Phosphate Buffered 8.2 mM NaHP0 4 Saline (1 x PBS) pH 7.4 137mMNaCI 2.7 mM KCI 1 . 5 m M H 2 P 0 4 Selection HAT Media Fusion Media 20% FCS 0.1 mM hypoxanthine 0.4 |iM aminopterine 16 |iM thymidine Table 2. Components of various media and buffers used for biological studies. 169 3.2.1 Preparation of Protein-Hapten Conjugates Immunoconjugate KLH6 Hapten 6 (3.92 mg, 9.3 u.mol) in methanol (0.4 uL) was combined with EDC (3.9 mg, 20 (irnol) in 1 x PBS (1.0 mL) and stirred for 20 min. A suspension of KLH (10 mg, 0.022 |imol of KLH, MW = 450 kDa) in 1 x PBS (2.6 mL) was added to the activated ester and the mixture was stirred at room temperature overnight. Final concentration of the conjugate was assumed to be 2.5 mg/mL (by weight of protein) and 0.98 mg/mL of hapten 6. This solution was stored in 0.5 mL aliquots at -70 °C. Immunoconjugate KLH35 Photochemical coupling of 35 to KLH was carried out following the procedure developed by Kerrigan. 8 1 A solution of KLH (3.0 mg, 0.0067 u.mol) in doubly distilled water (DDW) was mixed with sulfo-SANPAH (6.7 mg, 14 u,mol) in 1 x PBS in a Slide-A-Lyzer wrapped in foil, and placed on a shaker for 8 h. The mixture was dialysed overnight against DDW, then transferred into a covered round bottom flask containing 35 (6.8 mg, 23 u.mol) in DMSO and the mixture was freeze-dried in the dark. The aromatic azide was activated by 5 to 10 camera flashes at a distance of 20 cm, followed by 20 min under a UV lamp {X = 365 nm). The conjugate was resuspended in DDW and dialysed against DDW in the Slide-A-Lyzer and stored at -70 °C. Binding Conjugate BSA6 Hapten 6 (3.9 mg, 9.3 u.mol) in methanol (0.4 u.L) was combined with EDC (4.6 mg, 24 jimol) in 1 x PBS (1 mL) and stirred for 20 min. A suspension of BSA (10 mg, 170 0.15 pmol) in 1 x PBS (2.6 mL) was added to the activated ester and the mixture was stirred at room temperature overnight. Final concentration of the conjugate was assumed to be 2.4 mg/mL or 37 pM (by weight of BSA, MW=66,445 Da) with a hapten concentration of 2.3 mM. This stock BSA6 conjugate solution was used for binding assays and was stored in smaller aliquots at -70 °C. Analysis of the BSA6 conjugate by liquid chromatography electrospray ionisation mass spectrometry (LC-ESI-MS) was carried out as described by Adamczyk et a l . 8 4 A sample of B S A 6 was taken from stock solution and introduced onto the ESI-MS by HPLC using a linear gradient of 90:10 0.2% formic acid/methanol to 100% methanol over 20 min at a flow rate of 100 pL/min. Ion source conditions: ion spray 4800 V, orifice 30 V, nebulising gas 1.46 L/min, and TurboProbe jet 350 °C with air flow of 4 L/min. The sample gave three distinct LC fractions, of which the first two contained low MW peaks (m/z=400 to 900) and the last fraction contained high MW peaks (m/z=1200 to 2100). The molecular weight of the low molecular weight compounds was determined directly from the m/z spectra. High MW conjugates were resolved by converting the high m/z scan to the mass domain using BioReconstruct (deconvolution software from manufacturer). The low MW fractions of the conjugate mixture contained free hapten 6 (MW = 422, M ++1). Deconvolution of the high m/z spectrum indicates a hapten density in the range of 3 to 18 molecules of 6 per BSA molecule. In addition, a sample of BSA alone in PBS (1 mg/mL) was analysed under the same conditions and was calculated to have a MW of 66,446 Da. 171 Carboxyl Polystyrene Particles Coated with BSA6 An aliquot of BSA6 conjugate (0.5 mL) was diluted in 4 mL of 1 x PBS, then mixed with EDC (2.5 mg, 13 pmol) and Fluoricon carboxy beads (0.84 mm pore size, 5% w/v, 0.5 mL) in a 15 mL siliconized Corex glass centrifuge tube. The milky white solution was incubated overnight at room temperature and then centrifuged at 6000 rpm for 20 min on a Sorval HB4 rotor. The supernatant was carefully removed and discarded, and the pellet was washed twice with PCFIA buffer before the beads were resuspended in a 20 mL volume of PCFIA buffer and stored at -4 °C. Final concentration of the coated particles was taken to be 63 pg/mL protein and 0.13% (w/v) particles. Prior to use in assays, the beads were further buffered by the addition of Hepes (final concentration 20 mM) to maintain a neutral pH (7.4) in the presence of the selection (HAT) media. 3.2.2 Immunisat ions of Rabbits New Zealand white, female rabbits were immunised with KLH6 or KLH35 as per the schedule outlined below: WEEK 0: KLH6 (0.4 mL of stock.) was diluted with 1 x PBS (0.6 mL) and then emulsified with complete Freund's adjuvant (2 mL), which was injected intramuscularly. Each rabbit received 1 mL of this emulsion, containing 0.19 mg of 6 coupled to 0.5 mg of KLH for the initial immunisation. WEEK 4: KLH6 (0.2 mL of stock) was diluted with 1 x PBS (0.8 mL) and mixed with incomplete Freunds' adjuvant (2 mL) to give a thick emulsion, of which 1 mL was 172 injected intramuscularly. Each rabbit received 95 u.g of 6 coupled to 0.25 mg of KLH for the primary boost. WEEKS 8, 12, 16, 20: KLH6 (0.1 mL of stock) was diluted with 1 x PBS (0.9 mL) and mixed with incomplete Freunds' adjuvant (2 mL) to give a thick emulsion, of which 1 mL was injected intramuscularly. Each rabbit received 48 u.g of 6 coupled to 0.13 mg of KLH for the second boost. WEEKS 10, 14, 18: An ear bleed was taken from each rabbit to give approximately 30 mL of fluid. The blood was collected without heparin so that the blood cells would clot and be easy to separate from the sera. The blood was stored at 4 °C overnight, then centrifuged at 4000 rpm for 5 min. The sera were transferred into a new Falcon tubes and centrifuged once more to remove any remaining red blood cells from the sample. The sera were incubated at 56 °C for 35 min to kill the complement enzymes and then stored at -20 °C. WEEK 22: The rabbits were sacrificed and a whole body bleed was performed. The sera were treated as described above. 3.2.3 Polyclonal Ant ibody Purif ication Affinity column purification was carried out on a column containing Protein A coupled to Sepharose beads. Prior to use, the Protein A column was cleaned with successive washes of 5 bed volumes of 1 x PBS and 0.1 M glycine (pH 2.5). An additional 5 bed volumes of 1 x PBS was used so that the pH of the column was approximately 7.2 before the addition of the crude antibody sample. 173 Rabbit serum was pre-filtered through Syril-MF filters of pore size 0.45 u,m and 0.22u,m, prior to loading onto the column. The sample was loaded onto the column and the column was washed with 10 bed volumes of 1 x PBS. Finally, the bound antibodies were eluted with 0.1 M glycine, pH 2.5, collected in 1 mL portions and neutralised with 50 uL of saturated Tris buffer. Fractions containing purified antibody were combined and dialysed against 1 x PBS buffer overnight. The final concentration was determined by measuring its absorbance at 280 nm, where O.D./1.4 = concentration of antibody in mg/mL. 3.2.4 Screening Assay wi th an ELISA Detection of hapten-specific antibodies was carried out by an ELISA as follows. 1. Polystyrene plates were coated with either BSA6 conjugate or BSA alone (50 u.L/well), and incubated for 3 to 24 h at room temperature. 2. Plates were washed 3x with a 0.5% (w/v) skim milk powder in 1 x PBS to remove excess antigen and to block open sites on the plate. 3. Test serum was diluted in the blocking solution to give absorbance measurements in the linear range and added to each well (50 u.l_/well). The serum was incubated for 3 h at room temperature. 4. Plates were washed and blocked as in step 2. 5. Horse radish peroxidase (HRP) labelled secondary antibody at a 1:2,000 dilution of a 1 mg/mL stock solution was added (50 pL/well) and incubated for 1 h at room temperature. 6. Plates were washed and blocked as in step 2, as well as an additional wash of 6 x DDW. 7. A solution of ABTS (1 mg/mL) and 0.006% H2O2 in citrate buffer was added (50 pL/well) and incubated at 37 °C for 20 min. The absorbance was immediately measured with an ELISA reader at 405 and 490 nm. ELISA VARIATIONS ISOTYPING: A rabbit anti-mouse isotype (IgGi, lgG 2 a , lgG2t>, lgG 3 > IgM, IgA, IgE, Kand X light chains) antibody was added (50 pL/well) after step 4. The mixture was incubated for 2 h, then the plates were washed and the ELISA was carried out as described above. CHAOTROPIC ELUTION: Ammonium thiocyanate at concentrations from 0 to 8 M were added (50 pL/well) after step 4. The mixture was incubated for 15 min, then the plates were washed thoroughly and the ELISA was carried out as described above. INDIRECT COMPETITIVE INHIBITION: Inhibitor 35 at various concentrations were mixed with a fixed concentration of the test serum and allowed to equilibrate for 3 h, then added (50 pL/well) to plate bound conjugate as in step 3. The mixture 175 was incubated for 2 h, then the plates were washed and the ELISA was carried out as described above. 3.2.5 Immunisat ions of Mice Balb/c mice (6 to 8 week old females) were immunised with the KLH6 immunoconjugate as per the schedule outlined below: Day 1: KLH6 (0.3 mL of stock.) was mixed with complete Freund's adjuvant (0.6 mL) to give a thick emulsion, of which 100 u.L was injected subcutaneously per mouse. Each mouse received approximately 50 u.g of 6 coupled to 0.15 mg of KLH. Day 28: KLH6 (0.1 mL of stock) was diluted with 1 x PBS (0.2 mL) and mixed with incomplete Freund's adjuvant (0.5 mL) to give a thick emulsion, of which 100 uL/mouse was injected subcutaneously. Each mouse received 19.6 u.g of 6 coupled to 50 u.g of KLH for the primary boost. Day 42: Boost #2, as described above. Day 49: The mouse was warmed under a heat lamp (approximately 18 inches from the cage) to dilate blood vessels in tail. In a mouse restraint, the tail was wiped with 70% ethanol to enhance vasodilation. A tail vein was nicked with a razor blade and 20 u.L of blood collected with a sterile heparinized capillary tube and transferred into an eppendorf tube. The blood was incubated at 37 °C for 20 min and centrifuged at 4000 rpm for 4 min to separate the blood cells from the serum. 176 The serum was carefully drawn off and transferred into a new tube and stored at -20 °C. Day 84: A PBS boost containing 30 pL KLH6 in 1 x PBS (300 pL) was injected intraperitoneal^. Day 88: The mouse was sacrificed using CO2 and the spleen was harvested. 3.2.6 Hybridoma Technology The myeloma fusion partner P3-X63-Ag8.653 (C-653) was expanded as described below such that the cells were healthy and growing exponentially at the time of fusion. Approximately 2 x 10 7 cells with a viability of 99% were required per fusion. Day 1: The C-653 cells were expanded into 4 x 30 mL of myeloma growth medium at a concentration of approximately 10 3 cells/mL. Day 4: The cells were counted and each dish was split into four new dishes and diluted to a volume of 30 mL, each with myleloma growth medium at a concentration of 10 5 cells/mL. Day 5: The live cell count was approximately 2-5 x 10 s cells/mL. Counting Cells To a 10 pL aliquot of a suspended cell sample was added 10 pL of eosin (a dye that stains dead cells only). The resulting solution was mixed thoroughly and a sample 177 was pipetted into the sample groove of a hemocytometer. The hemocytometer was placed on an inverted phase microscope and the cells in each quadrant were counted. The formula used to calculate the number of cells/mL is as follows: (# of cells x 10 5) / (5 x # of quadrants) = # cells/mL Harvesting Spleenocytes All dissecting instruments were sterilised in 70% ethanol prior to use. The mouse was anaesthetised to death with CO2 and immediately bled by cardiac puncture to collect blood to be used as a positive polyclonal control in the screening assays. The abdominal area was soaked with ethanol prior to incision into the peritoneal cavity. The spleen was removed and transferred into a 10 mL sterile tube containing 2% FCS Hanks solution. The spleen was cut into pieces and mashed through a sterile wire mesh into fresh sterile media in a petri dish using the end of a 10 mL syringe plunger. The cells were centrifuged at 300g for 5 min and the supernatant was removed. A warmed solution (37 °C) of MRCRB was added slowly (5 mL over 1 min) to the pellet of cells and the resuspended mixture was placed in a 37 °C water bath for 7 min. The MRCRB was immediately neutralised with 2% FCS Hanks solution (5 mL) and the cells were centrifuged at 300g for 5 min. The supernatant was discarded and the cells were resuspended in 10 mL of 2% FCS Hanks and counted. 178 Fusion using Polyethylene Glycol (PEG) Spleen and myeloma cells were combined in a 5:1 ratio in a 50 mL Falcon tube and centrifuged at 300g for 5 min. The cells were resuspended in serum free RPMI 1640 media washed 3x to remove all serum from the medium. Finally, the medium was removed and the cell pellet was liquefied by knocking the bottom of the tube on a hard surface. A warmed solution (37 °C) of PEG (0.8 mL) was added dropwise to the pellet over one min with gentle mixing. The mixture was allowed to stand for 1 min, followed by the slow addition of RPMI 1640 (1 mL) over one min. An additional 20 mL of RPMI 1640 was added over 5 min with gentle swirling of the tube. The cells were centrifuged at 300g for 5 min and resuspended in 40 mL of HAT media. The cell suspension was plated into four 96-well flat-bottomed plates (100 u.l_/well). Final volumes in each well were made up to 200 u.L with additional HAT media and incubated at 37 °C. Wells were maintained by removal of half the media via a sterile Pasteur pipette and suction, and replaced with fresh HAT media as required (approximately on days 1, 3, 6 ,10,14) . 3.2.7 Selection and Expansion of Hybridomas Initial screening of the hybridomas was carried out by a PCFIA. Carboxyl polystyrene particles coated with BSA6 (25 uL) were plated into 96-well Fluoricon acetate plates. Supernatant from fusion plates (50 uL) was added and the mixture was incubated for 30 min at 37 °C. Using the PANDEX, the plates were suctioned and 179 washed to remove nonbinding antibodies, followed by detection with a goat anti-mouse FITC-labelled secondary antibody at 535 nm (excitation at 485 nm). Wells containing anti-6 antibodies (those that gave positive signals in the binding assay) were chosen and cloned by limit dilution in HT media. The cells were transferred and titrated 1:2 step dilution down the first column of a new 96-well plate, followed by another titration (1:2 step dilution) of the column across the entire plate. Fresh HT media was added to give a final volume of 200 uL in each well. Cells were allowed to grow and were maintained as described earlier. After 10 days, the supernatants from the wells were assayed as above. Clones producing hapten specific antibodies were again expanded by limit dilution in hybridoma media. After 2 or 3 limit dilutions, all wells that contained cells on a single limit dilution plate gave positive signals, indicating the presence of only one stable clone producing a monoclonal antibody. 3.2.8 Freezing Cell Lines Copies of clones were expanded and frozen for long term storage. The cells were grown in 5 mL tissue culture dishes in hybridoma media and allowed to reach confluence. Cells were centrifuged at 300g for 5 min, taken up in fresh RPMI media and counted. The cells were resuspended in freezing medium at a concentration of 1-2 x 10 s cells/mL and transferred to Nunc® freezing vials (1 mL). The vials were cooled on ice for 20 min, then cooled in a freezing cone at a rate of 1 °C/min to reach a 180 final temperature of -195 °C. Finally the vials were stored in a liquid nitrogen storage tank. 3.2.9 Ascites Production Balb/c mice (6 to 8 weeks old, females) were primed by injection of 200 pL of pristane (2, 6, 10, 14-tetramethylpentadecane) into the peritoneum. After 7 days, the hybridoma cells (0.5 mL of 2 x 10 6 cells/mL) were injected into the peritoneal cavity. In 10 to 14 days, a liquid tumour was visible and ready to be harvested. The mouse was anaesthetised to death with CO2, and a small incision was made in the skin above the abdomen to expose the abdomen muscles. The muscles were pinched upwards with tweezers and a small incision in the abdominal cavity allowed for the removal of the ascites fluid via a disposable liquipette. 3.2.10 Monoclonal Ant ibody Purification Ascites fluid containing the desired monoclonal antibody was centrifuged at 800g for 5 min and the top layer of pristane oil was removed. The remaining supernatant was separated from the cell pellet and diluted 1:2 with 1 x PBS before subsequent filtrations with 0.45 pm and 0.22 pm filters. The ascites were then loaded onto a SAM affinity column. The column was washed with 10 bed volumes of 1 x PBS. Antibodies were eluted from the column with 0.1 M glycine (pH 2.5) and collected in 1 mL portions, which were then neutralised with 50 pL of saturated Tris buffer. 181 Fractions containing purified antibody were combined and dialysed against 1 x PBS buffer overnight. The final concentration was determined by measuring the absorbance at 280 nm. 3.3 KINETIC ASSAYS All kinetic assays were done with disposable cuvettes using a UNICAM UV/Visible spectrophotometer, equipped with a circulating water bath and thermostatted cuvette holders. Assays were carried out in 600 pL volumes at 37 °C and the cuvettes were covered to prevent loss of solvent through evaporation and to allow for proper mixing of the final solutions. Initial rates were determined by measuring the release of 4-nitrophenolate anion (e=15,968 L moi"1 cm"1) at 400 nm. 3.3.1 Rabbits Kinetic assays on the purified rabbit antibodies typically contained final concentrations (v/v) of 12.5% DMSO, 0.25% Triton X-100 in 1 x PBS buffer, pH 7.4 to 7.6. Final concentrations of the substrates used ranged from 400 to 600 pM. Rate measurements were carried out using working antibody concentrations of 10 to 20 pM H54/H55/H60/H61 (based on an IgG with molecular weight of 150 kDa). The background rate of hydrolysis of the substrate was measured at the same time under the same conditions using either a control antibody (nonspecific antibody such as H56 or H57) or no antibody at all. 182 3.3.2 Mice Kinetic assays typically contained final concentrations (v/v) of 10% DMSO, 0.18% Triton X-100 in 1 x PBS buffer, pH 7.6 to 8.0. Final concentrations of the substrates used ranged from 50 to 650 u M Rate measurements were carried out using working antibody concentrations of 0.88 u.M to 1.0 u.M F325 (based on IgG with molecular weight 150 kDa). The background rate of hydrolysis of the substrate was measured at the same time under the same conditions in the presence of a control antibody (nonspecific monoclonal antibody, HIL3) or porcine lipase. REFERENCES 183 1) Brown, A. G.; Smale, T. C.; King, T. J.; Hasenkamp, R.; Thomson, R. H. J. Chem. Soc, Perkin Trans. 11976,1165. 2) Endo, A. J. Antibiot. 1979, 32, 852. 3) Alberts, A. W.; Chen, J.; Kuron, G.; Hunt, V.; Huff, J.; Hoffman, C ; Rothrock, J.; Lopez, M.; Joshua, H.; Harris, E.; Patchett, A.; Monaghan, R.; Currie, S.; Stapley, E.; Albers-Schonberg, G.; Hensens, O.; Hirshfield, J.; Hoogsteen, K.; Liesch, J.; Springer, J. Proc. Natl. Acad. Sci. USA 1980, 77, 3957. 4) Hirama, M.; Uei, M. J. Am. Chem. Soc. 1982, 104, 4251. 5) Moore, R. N.; Bigam, G.; Chan, J. K.; Hogg, A. M.; Nakashima, T. T.; Vederas, J. C. J. Am. Chem. Soc. 1985, 107, 3694. 6) Witter, D. J.; Vederas, J. C. J. Org. Chem. 1996, 61, 2613. 7) Diels, O.; Alder, K. Justus Liebigs Ann. Chem. 1928, 460, 98. 8) Klemm, L. H.; Gopinath, K. W. Tetrahedron Lett. 1963, 1243. 9) Ciganek, E. Org. React. 1984, 32,1. 10) Brieger, G.; Bennett, J. N. Chem. Rev. 1980, 80, 63. 11) Sauer, J.; Sustmann, R. Angew. Chem., Int. Ed. Engl. 1980, 19, 779. 12) Craig, D. Chem. Soc. Rev. 1987, 16,187. 13) Diedrich, M. K.; Klarner, F.-G.; Beno, B. R.; Houk, K. N.; Senderowitz, H.; Still, W. C. J. Am. Chem. Soc. 1997, 119, 10255. 14) Lin, Y.-T., Houk, K. N. Tetrahedron Lett. 1985, 26, 2269. 184 15) Roush, W. R.; Hall, S. E. J. Am. Chem. Soc. 1981, 103, 5200. 16) Tarasow, T. M.; Tarasow, S. L ; Eaton, B. E. Nature 1997, 389, 54. 17) Kang, J.; Rebek Jr., J. Nature 1997, 385, 50. 18) Liu, X . -C; Mosbach, K. Macromol. Rapid Commun. 1997, 18, 609. 19) Oikawa, H.; Katayama, K.; Suzuki, Y.; Ichihara, A. J. Chem. Soc, Chem. Commun. 1995, 1321. 20) Harlow, E.; Lane, D. Antibodies: A Laboratory Manual; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, 1988. 21) Kohler, G.; Milstein, C. Nature 1975, 256, 495. 22) Steward, M. W.; Steensgaard, J. Antibody Affinity: Thermodynamic Aspects and Biological Significance; CRC Press: Boca Raton, 1983, pp 1—12. 23) Steward, M. W.; Steensgaard, J. Antibody Affinity: Thermodynamic Aspects and Biological Significance; CRC Press: Boca Raton, 1983, pp 59-69. 24) Attie, A. D.; Raines, R. T. J. Chem. Ed. 1995, 72, 119. 25) Friguet, B.; Chaffotte, A. F.; Djavadi-Ohaniance, L.; Goldberg, M. E. J. Immunol. Methods 1985, 77, 305. 26) Nieto, A.; Gaya, A.; Jansa, M.; Moreno, C ; Vives, J. Moi. Immunol. 1984, 21, 537. 27) Goldberg, M. E.; Djavadi-Ohaniance, L. Curr. Opin. Immunol. 1993, 5, 278. 28) Pullen, G. R.; Fitzgerald, M. G.; Hosking, C. S. J. Immunol. Methods 1986, 86, 83. 29) Luxton, R. W.; Thompson, E. J. J. Immunol. Methods 1990, 131, 277. 185 30) Macdonald, R. A.; Hosking, C. S.; Jones, C. L. J. Immunol. Methods 1988, 106, 191. 31) Goldblatt, D.; van Etten, L ; van Milligen, F. J.; Aalberse, R. C ; Turner, M. W. J. Immunol. Methods 1993, 166, 281. 32) Levison, S. A.; Kierszenbaum, F.; Dandliker, W. B. Biochemistry1970, 9, 322. 33) Kraut, J. Science 1988, 242, 533. 34) Pauling, L. Am. Sci. 1948, 36, 51. 35) Jencks, W. P. Catalysis in Chemistry and Enzymology, McGraw-Hill: New York, NY, 1969, pp 283-288. 36) Haynes, M. R.; Stura, E. A.; Hilvert, D.; Wilson, I. A. Science 1994, 263, 646. 37) Hsieh-Wilson, L C.; Schultz, P. G.; Stevens, R. C. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5363. 38) Jahangiri, G. K.; Reymond, J.-L. J. Am. Chem. Soc. 1994, 116, 11264. 39) Raso, V.; Stollar, B. D. Biochemistry 1975, 14, 591. 40) Kohen, F.; Kim, J.-B.; Barnard, G.; Lindner, H. R. Biochim. Biophys. Acta 1980, 629, 328. 41) Hu, Y.-J.; Ji, Y.-Y.; Wu, Y. -L; Yang, B.-H.; Yeh, M. Bioorg. Med. Chem. Lett. 1997,7,1601. 42) Shreder, K.; Harriman, A.; Iverson, B. L. J. Am. Chem. Soc. 1995, 117, 2673. 43) Stephens, D. B.; Iverson, B. L. Biochem. Biophys. Res. Commun. 1993, 192, 1439. 44) Wilmore, B. H.; Iverson, B. L. J. Am. Chem. Soc. 1994, 116, 2181. 186 Wallace, M. B.; Iverson, B. L. J. Am. Chem. Soc. 1996, 118, 251. Shreder, K.; Thomas, R.; Wallace, M.; Helms, E.; Iverson, B. Isr. J. Chem. 1996,36,215. Schoneberg, T.; Erhard, M. H.; Kellner, J.; Kuhn, S. O.; Losch, U. J. Immunoassay1995, 16, 17. Gallacher, G.; Jackson, C. S.; Searcey, M.; Goel, R.; Mellor, G. W.; Smith, C. Z.; Brocklehurst, K. Eur. J. Biochem. 1993, 214, 197. Tramontane A.; Janda, K. D.; Lerner, R. A. Science 1986, 234, 1566. Pollack, S. J.; Jacobs, J. W.; Schultz, P. G. Science 1986, 234,1570. Jacobsen, N. E.; Bartlett, P. A. J. Am. Chem. Soc. 1981, 103, 654. Tawfik, D. S.; Lindner, A. B.; Chap, R.; Eshhar, Z.; Green, B. S. Eur. J. Biochem. 1997, 244, 619. Lerner, R. A.; Benkovic, S. J.; Schultz, P. G. Science 1991, 252, 659. Thomas, N. R. Nat. Prod. Rep. 1996, 479. Hilvert, D. Curr. Opin. Struct. Biol. 1994, 4, 612. Baum, R. M. Chem. Eng. A/ews1987, 75(14) 30. Jacobsen, J. R.; Schultz, P. G. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 5888. Li, T.; Hilton, S.; Janda, K. D. J. Am. Chem. Soc. 1995, 117, 2123. Sinha, S. C ; Keinan, E. J. Am. Chem. Soc. 1995, 117, 3653. Yang, G.; Chun, J.; Arakawa-Uramoto, H.; Wang, X.; Gawinowicz, M. A.; Zhao, K.; Landry, D. W. J. Am. Chem. Soc. 1996, 118, 5881. 187 Wentworth, P.; Datta, A.; Blakey, D.; Boyle, T.; Partridge, L J.; Blackburn, G. M. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 799. Bagshawe, K. D. Br. J. Cancer1987, 56, 531. Janda, K. D.; Shevlin, C. G.; Lerner, R. A. Science 1993, 259, 490. Hilvert, D.; Hill, K. W.; Nared, K. D.; Auditor, M.-T. M. J. Am. Chem. Soc. 1989, 7 7 7,9261. Braisted, A. C ; Schultz, P. G. J. Am. Chem. Soc. 1990, 772, 7430. Brown, F. K.; Houk, K. N. Tetrahedron Lett. 1984, 25, 4609. Suckling, C. J.; Tedford, M. C ; Bence, L. M.; Irvine, J. I.; Stimson, W. H. J. Chem. Soc, Perkin Trans. 11993, 1925. Meekel, A. A. P.; Resmini, M.; Pandit, U. K. J. Chem. Soc, Chem. Commun. 1995, 571. Gouverneur, V. E.; Houk, K. N.; de Pascual-Teresa, B.; Beno, B.; Janda, K. D.; Lerner, R. A. Science 1993, 262, 204. Takacs, J. M.; Helle, M. A.; Seely, F. L. Tetrahedron Lett. 1986, 27, 1257. Nishiguchi, T.; Kuroda, M.; Saitoh, M.; Nishida, A.; Fujisaki, S. J. Chem. Soc, Chem. Commun. 1995, 2491. Ukai, J.; Ikeda, Y.; Ikeda, N.; Yamamoto, H. Tetrahedron Lett. 1983, 24, 4029. Jones, G. H.; Hamamura, E. K.; Moffatt, J. G. Tetrahedron Lett. 1968, 55, 5731. Dalling, D. K.; Grant, D. M.; Paul, E. G. J. Am. Chem. Soc. 1973, 95, 3718. Bastiansen, O.; Hassel, O. Nature 1946, 757, 765. 188 76) Gerig, J. T.; Roberts, J. D. J. Am. Chem. Soc. 1966, 88, 2791. 77) Dalling, D. K.; Grant, D. M.; Johnson, L. F. J. Am. Chem. Soc. 1971, 93, 3678. 78) Lowe, C. R.; Harvey, M. J.; Craven, D. B.; Dean, P. D. G. Biochem. J. 1973, 133, 499. 79) Swerdlow, R. D.; Ebert, R. F.; Lee, P.; Bonaventura, C ; Miller, K. I. Comp. Biochem. Physiol. B: Comp. Biochem. 1996, 113B, 537. 80) Brinkley, M. Bioconjugate Chem. 1992, 3, 2. 81) Kerrigan, S. Ph.D. Thesis, University of British Columbia, 1997. 82) Klaus, G. G. B.; Cross, A. M. Cell. Immunol. 1974, 14, 226. 83) Malaitsev, V. V.; Azhipa, O. Y. Bull. Exp. Biol. Med. 1993, 115, 726. 84) Adamczyk, M.; Gebler, J. C.; Mattingly, P. G. Bioconjugate Chem. 1996, 7, 475. 85) Tsumuraya, T.; Suga, H.; Meguro, S.; Tsunakawa, A.; Masamune, S. J. Am. Chem. Soc. 1995, 117, 11390. 86) Schrock, A. K.; Schuster, G. B. J. Am. Chem. Soc. 1984, 106, 5228. 87) Das, M.; Fox, C. F. Annu. Rev. Biophys. Bioeng. 1979, 8,165. 88) Mader, M. M.; Bartlett, P. A. Chem. Rev. 1997, 97, 1281. 89) Benkovic, S. J.; Napper, A. D.; Lerner, R. A. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 5355. 90) Jacobs, J. W. Biotechnology 1991, 9, 258. 91) Stewart, J. D.; Benkovic, S. J. Nature 1995, 375, 388. 189 MacBeath, G.; Hilvert, D. Chem. Biol. 1996, 3, 433. Charbonnier, J.-B.; Carpenter, E.; Gigant, B.; Golinelli-Pimpaneau, B.; Eshhar, Z.; Green, B. S.; Knossow, M. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 11721. Zemel, R.; Schindler, D. G.; Tawik, D. S.; Eshhar, Z.; Green, B. S. Moi. Immunol. 1994, 31, 127. Driggers, E. M.; Cho, H. S.; Liu, C. W.; Katzka, C. P.; Braisted, A. C ; Ulrich, H. D.; Wemmer, D. E.; Schultz, P. G. J. Am. Chem. Soc. 1998, 120, 1945. Haynes, M. R.; Lenz, M.; Taussig, M. J.; Wilson, I. A.; Hilvert, D. Isr. J. Chem. 1996,36,151. Heine, A.; Stura, E. A.; Yli-Kauhaluoma, J. T.; Gao, C ; Deng, Q.; Beno, B. R.; Houk, K. N.; Janda, K. D.; Wilson, I. A. Science 1998, 279, 1934. Romesberg, F. E.; Spiller, B.; Schultz, P. G.; Stevens, R. C. Science 1998, 279, 1929. Benkovic, S. J.; Adams, J. A.; Borders Jr., C. L.; Janda, K. D.; Lerner, R. A. Science 1990, 250, 1135. Tawfik, D. S.; Green, B. S.; Chap, R.; Sela, M.; Eshhar, Z. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 373. Lane, J. W.; Hong, X.; Schwabacher, A. W. J. Am. Chem. Soc. 1993, 115, 2078. Reymond, J. -L; Koch, T.; Schroer, J.; Tierney, E. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 4251. Janda, K. D.; Lo, L.-C; Lo, C.-H. L; Sim, M.-M.; Wang, R.; Wong, C.-H.; Lerner, R. A. Science 1997, 275, 945. Hua, T. D.; Rolland-Fulcrand, V.; Lazaro, R.; Viallefont, P.; Lefranc, M.-P.; Weill, M. Tetrahedron Lett. 1996, 37, 175. 190 105) Fenniri, H.; Janda, K. D.; Lerner, R. A. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 2278. 106) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals; 3rd ed.; Permagon: New York, 1988. SPECTRAL APPENDIX 192 O 4a 1 OO H 2 o : 1 0 - ' ° ~l 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 193 194 1 OO -\ 1 I 1 1 1 1 1 1 1 1 1 1 1 — ~ 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m 1 — 1 ) 195 H 'H 5a/5b ' C 0 2 M e 1 o o H 1 o H o -I ! ! 1 1 1 1 1 1 1 1 1 1 1 1 1 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 196 7 6 5 4 3 2 1 (ppm) 197 1 1 1 1 1 1 1 1 1 1 1 1 1 — 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 198 1 o o n 2 0 - I 1 0 - I o -3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W o v e n u m b e r ( c m — 1) 199 201 (ppm) 202 (ppm) 203 (ppm) 205 2 0 - ' I 1 O -o -1 1 1 1 1 1 1 1 1 1 1 1 I I I 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 206 J 5 5 + j i 207 7 6 5 4 3 2 1 (ppm) W a v e n u m b e r ( c m — 208 1 oo -\ 30 -20 -1 0 - " ° "I 3 2 0 0 24-00 1 600 800 W o v e n u m b e r ( c m — 1 ) 2 0 9 7 6 5 4 3 2 1 (ppm) 210 i i i i i i i i i i i i 'i i i I . . . 1 1 i i i i 'i i 1 i ' . i i 7 6 5 4 3 2 ) (ppm) 1 o o -\ o -1 1 1 1 1 1 1 1 1 1 i i i i i 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 211 CD O cz o " E o 6^ 3 2 0 0 2 4 0 0 1 6 0 0 W a v e n u m b e r ( c m — 1 ) S O O 212 7 6 5 4 3 2 1 (ppm) 1 o o -\ — i 1 1 1 1 1 1 1 1 1 1 1 1 — 3 2 0 0 2 4 0 0 1 6 0 0 B O O W a v e n u m b e r ( c m — 1 ) 213 214 • ' i ' 1 1, 'i i' 'i i 'i 11 9 8 7 6 5 4 3 2 (ppm) 215 1 oo H 3 2 0 0 ' 2-40 0 ' 1 6 0 0 ' ' 8 0 0 W a v e n u m b e r ( c m — 1 ) 216 217 1 oo H W a v e n u m b e r ( c m — 1 ) 218 220 1 oo H o -1 , 1 1 1 1 1 1 1 1 1 1 i i i 3 2 OO 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 221 6 ' 5 V 3 2 1 (ppm) 222 7 6 5 4 3 2 1 (ppm) 223 Ml 26a 4 (ppm) CD CO cz o I — 3 2 0 0 2 4 0 0 1 6 0 0 W a v e n u m b e r ( c m — 1) 8 0 0 224 l i , - 1 , 1 1 1 , - i - r - i - i — i - | - i i i i i i i | i i ' I I i i i i i i i i i ' I I i 1 7 6 5 4 3 2 ) (ppm) 1 oo H 1 o H o - ,„_^_^^_ 7__ r____^_^__-__ n__ r™ 3 2 0 0 ' ' 2 4 - 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 225 226 7 6 5 4 3 2 1 (ppm) 1 oo -\ ° H 4 0 0 0 3 2 0 0 2 4 0 0 1 6 0 0 80( Wav^e n u m b e r ( c m — 1 ) 227 ° i \ — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 2 2 8 o -1 1 1 1 1 1 1 1 1 1 i i i i i 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 229 1 o H o -1 1 1 1 ; 1 1 1 1 1 1 1 1 1 1 — 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1) 230 3 2 0 0 ' 2 4 0 0 ' 1 6 0 0 ' ' ' 8 0 0 W a v e n u m b e r ( c m — 1 ) 231 o -1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 — 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 232 233 7 * 5 4 a 2 1 IPPTO 1 oo H 3 0 -2 0 -1 o -o -3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 234 1 OO H ° 1 l 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 — 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1) 235 236 237 1 oo -\ .o -1 1 1 1 1 1 1 1 1 i i i i i i 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W a v e n u m b e r ( c m — 1 ) 2 3 8 7 6 5 4 3 2 1 (ppm) 1 OO H 3 2 0 0 2 4 0 0 1 6 0 0 8 0 0 W o v e n u m b e r ( c m — 1 ) 239 240 r1 7 t S * i (ppn) 241 CD CD CZ O ' i c: o 3 2 0 0 2 4 0 0 1 6 0 0 W a v e n u m b e r ( c m — 1 ) 8 0 0 242 8 7 6 5 4 3 2 ) (ppm) 243 2 4 4 H 46a/46b CD o o Ul c z o 1 OO 9 0 S O H 7 0 6 0 5 0 4 0 H 3 0 2 0 1 O o H 3 2 0 0 2 4 0 0 1 6 0 0 W a v e n u m b e r ( c m — 1 ) iOO 245 1 o o -\ o -3 2 0 0 2 4 0 0 1 6 0 0 S O O W a v e n u m b e r ( c m — 1 ) 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059625/manifest

Comment

Related Items