–O2C HN SH O NH O H3 +N –O2C HN O N CO2H O H3 +N S Me Me H H 2 1 O2 2H2O IPNS CO2H S O COR S O HN O CO2R¢ 13 R = OH 14 R = Cl 7a,b R¢ = But R¢HN NHR CO2R¢¢ 8 R = R¢ = R¢¢ = H 10 R = Z, R¢ = R¢¢ = H 9 R = Z, R¢ = Boc, R¢¢ = H 11 R = Z, R¢ = Boc, R¢¢ = But 12 R = H, R¢ = Boc, R¢¢ = But HN ButO O NH CO2But O O HN Boc SH HN ButO O NH CO2But O O HN Boc S-)2 20a,b 21 –O2C HN NH CO2H O O +H3N S-)2 22 –O2C HN NH O SH CO2H O H3 +N 6a,b 200 J. CHEM. RESEARCH (S), 1997 J. Chem. Research (S), 1997, 200–201 J.Chem. Research (M), 1997, 1401–1411 Studies on Side Chain Interactions during the Isopenicillin N Synthase Catalysed Biosynthesis of Penicillins Florine Cavelier, Andrew T. Russell, Christopher J. Schofield and Jack E. Baldwin* The Dyson Perrins Laboratory and the Oxford Centre for Molecular Sciences, South Parks Road, Oxford OX1 3QY, UK The role of the L-d-(a-aminoadipoyl)–L-cysteinyl amide bond in isopenicillin N synthase catalysis is probed by the synthesis of two analogues of its tripeptide substrate L-d-a-aminoadipoyl-L-cysteinyl-D-valine.Isopenicillin N synthase (IPNS) catalyses the reaction of dioxygen and the tripeptide L-d-a-aminoadipoyl-L-cysteinyl- D-valine (ACV, 1) to give isopenicillin N (IPN, 2) and two water molecules (Scheme A).1,2 The L-d-a-aminoadipoyl side chain of IPNS may be replaced with a variety of analogues,1,2 but for efficient conversion to a penicillin a linear 6-carbon chain (or equivalent) is required.4 The puckered nature of the penicillin products and presumably intermediates during IPNS catalysis may place the L-d-a-aminoadipoyl–cysteinyl amide link proximate to the reactive iron centre and it was considered possible that this amide plays a catalytic role.The analogue 6 in which the NH of the amide link is 1,3-transposed into the side chain of ACV may thus act as a mechanistic probe for the involvement of the L-d-a-aminoadipoyl-L-cysteinyl amide link in catalysis. Diaminobutyric acid (8) was sequentially N-protected with benzyloxycarbonyl (Z) and tert-butyloxycarbonyl (Boc) groups, on its d- and a-amino groups respectively to give 9 via 10.8 tert-Butyl ester 11 formation using tert-butyl alcohol and dimethylformamide neopentyl acetal,9,10 followed by hydrogenolysis of the d-amino protecting group gave 12.Racemic thioparaconic acid (13), synthesised7 from itaconic acid, was activated as its acid chloride 14 and reacted with D-valine tertbutyl ester to give the lactones 7a,b (70%).Diprotected L-diaminobutyric acid (12) was treated (EtOH, Cairos tube, 110 °C, overnight)13 with the lactones 7a,b to give the epimeric thiols 20a,b in low yield (16%). Thiols 20a,b were oxidised [PhI(OAc)2] to a mixture of disulfides 21 (36%) in order to prevent intramolecular attack of the thiol on the side chain amide link resulting in reformation of thiolactones 7a, b and 12 (Scheme B). Acid mediated deprotection gave the desired epimeric peptides as a mixture of disulfides 22 (98%), which were subsequently reduced using dithiothreitol to give crude epimeric thiols 6a,b which were purified by HPLC.Neither 6a nor 6b were found to be substrates or inhibitors of IPNS. The influence of the substrate side chain linkage in IPNS catalysis was also investigated by removal of the carbonyl of the side chain amide link, i.e. by the synthesis and incubation of 24. Since the amino group of the L-d-a-aminoadipoyl side chain is not required for IPNS turnover,4 the synthetic target was simplified to 25.Thus, aldehyde 2614 and S-benzhydrylcysteine (27) were reacted under reductive amination conditions (NaBH3CN, NaOH, MeOH) to give acid 28 (40%), which was coupled with D-valine tert-butyl ester to give 29 (62%) (Scheme C). Deprotection via basic cleavage of the methyl ester, followed by acid mediated removal of the benzhydryl and ester groups gave tripeptide 25 (97% prior to HPLC). Incubation of 25 with IPNS led again only to recovered starting material with no evidence for the formation of products (by 1H NMR, bioassay, or HPLC analyses).Preincubation experiments of 25 with IPNS did not lead to any increased inactivation relative to controls with ACV 1. However, preliminary kinetic studies indicated that amine 25 *To receive any correspondence. Scheme A Scheme BMeO H O O 26 +H3N O– O SBzh 27 MeO NH OH O O SBzh 28 MeO NH HN O O SBzh 29 Bzh = CHPh2 CO2Bu HO2C HN NH O SH R CO2H 24 R = NH2 25 R = H J.CHEM. RESEARCH (S), 1997 201 reversibly inhibited IPNS with an IC50 of between 35 and 45 mM. Techniques used: IR, 1H and 13C NMR, MS, TLC, polarimetry, elemental analysis References: 14 Schemes: 1 Received, 29th January 1997; Accepted, 11th March 1997 Paper E/7/00662D References cited in this synopsis 1 J. E. Baldwin and C. J. Schofield, in The Chemistry of b-lactams, ed. M. I. Page, Blackie, London, 1992, ch. 1, pp. 1–78 and references cited therein. 2 J. E. Baldwin and M. Bradley, Chem. Rev., 1990, 90, 1079 and references cited therein. 4 J. E. Baldwin, E. P. Abraham, R. M. Adlington, G. A. Bahadur, B. Chakravarti, B. P. Domayne-Hayman, L. D. Field, S. I. Flitsch, G. S. Jayatilake, A. Spakovskis, H.-H. Ting, N. J. Turner, R. L. White and J. J. Usher, J. Chem. Soc., Chem. Commun., 1984, 1225. 7 B. Holmberg and E. Schjånberg, Ark. Kemi, Min. Geol., 1940, 14A, 1. 8 J. Leclerc and L. Benoiton, Can. J. Chem., 1968, 46, 1047. 9 J. E. Baldwin, C. N. Farthing, A. T. Russell, C. J. Schofield and A. C. Spivey, Tetrahedron Lett., 1996, 37, 3761. 10 T. R. Welter, US Pat., 5 087 288, 1992. 13 J. Megnan, M. Colin and G. Lang, Eur. Pat., 368 763A1, 1990. 14 A. W. Burgstahler, L. O. Weigeland and C. G. Sheifer, Synthesis, 1976, 767. Scheme C