A Biomimetic Synthesis of a Porphobilinogen Precursor Using a Mukaiyama Aldol Reaction** Andre  R. Chaperon, Thomas M. Engeloch, and Reinhard Neier* Dedicated to Professor Dieter Seebach on the occasion of his 60th birthday The tetrapyrrolic ªpigments of lifeº fulfill many different functions and therefore have a special position among natural pigments. [1] The structures of the intermediates (d-amino- levulinic acid (ALA, 1) and porphobilinogen (PBG, 2)) leading to uroporphyrinogen (Uro) III (3), the precursor of all tetrapyrroles, were determined in the early 1950s (Scheme 1). [2] Even early on, the second and third steps of this elegant and convergent biosynthesis could be mimicked chemically. [3] N HOOC COOH NH 2 NH 2 O COOH HN NH HN NH HOOC HOOC COOH COOH HOOC COOH COOH HOOC (EC 4.2.1.24) 2 (ALA) 2) (E.C. 4.2.1.75) PBGS (Uro III) 8 x 1) (E.C. 4.3.1.8) 3 (PBG) 4 x 1 H Scheme 1. Biosynthesis of uroporphyrinogen III (3). PBGS porphobi- linogen synthetase. [*] Prof. Dr. R. Neier, Dr. A. R. Chaperon, Dipl.-Chem. T. M. Engeloch Institut de Chimie, Universite  de Neucha Ãtel Av. Bellevaux 51, CH-2000 Neucha à tel (Switzerland) Fax: Int. code  (41)32 718-2511 E-mail: reinhard.neier@ich.unine.ch [**] This work was supported by the Swiss National Science Foundation and the Fonds der Basler Chemischen Industrie. This work contains parts of the Ph.D. theses of A. R. C. (The Á se de Doctorat, Universite  de Neucha Ãtel, 1996) and T. M. E. (The Á se de Doctorat, in preparation). We thank H. Bursian and Dr. S. Claude (Neucha à tel) for measuring the NMR spectra. In this context the question arose whether the biosynthesis of porphobilinogen (2), which formally corresponds to a Knorr synthesis, could be imitated in the test tube. [4] Despite efforts in different laboratories the process could not be imitated satisfactorily so far. We are examining whether one can imitate the mechanism for the biosynthesis of 2 proposed by Shemin and Nandi [5] and use it for the synthesis of pyrroles. [6] We report here on the synthesis of an N-protected derivative of PBG, which relies on the Mukaiyama aldol reaction. [7] Since the structure determination of 2 40 years ago, six different synthetic strategies have been developed. [2a, 8] Despite the simplicity of the structure, the synthesis of this natural product in large quantities has remained difficult. In recent years several groups have developed novel approaches to porphobilinogen (2). [9] The starting point for our synthesis was the preparation of alkyl-substituted pyrroles using the two-step procedure con- sisting of a Mukaiyama aldol reaction followed by the reduction of the azido function to give the amino group. [6] The silyl enol ether 4, obtained from 5-phthalimidomethyl- levulinate in 93 % yield, was required as starting material for our synthesis. [10] Because 4 is not very nucleophilic, we were unable to couple it under standard conditions with the acetal of 5-azidomethyllevulinate. [6] At temperatures below  408C TiCl 4 was not active enough to catalyze the aldol reaction. At temperatures above  408C we only could observe decom- position of the starting materials. When we used Lewis acids like TMSOTf [11] or the ªsuper-Lewis acidº (TMS)B(OTf) 4 described by Davis [12] we could induce the aldol reaction with the dimethyl acetal of methyllevulinate (5). We applied the conditions described by Noyori et al. [11a] with 0.11 equiv TMSOTf as catalyst and were able to isolate 30 % of the pure diastereoisomer rac-6 (Scheme 2). Even with these Lewis acids the crucial C±C bond formation could not be achieved when protected precursors of 5-aminolevulinate were employed. Increasing the reactivity of the carbonyl component seemed the most promising way to solve the problem. When 4 was allowed to react with acylcyanide 7 , the cyanohydrin could be detected in the crude reaction product. However, after aqueous extraction and purification by column chromatography the hydrolysis prod- uct rac-8 was obtained in 35 % yield (Scheme 2). COOCH 3 OCH 3 H 3 CO H 3 COOC COOCH 3 OCH 3 N O O OSi(CH 3 ) 3 O H 3 COOC COOCH 3 O O COOCH 3 NC O H 3 COOC N O O N O O 5 4 0.1 TMSOTf, CH2Cl2 – 63°C, 23 h rac-8 (35%) 7 1) 1.1 TiCl4, CH2Cl2 –78°C, 25 h 2) SiO2 rac-6 (30%) Scheme 2. Aldol reaction with silyl enol ether 4. Published in Angewandte Chemie (International Edition) 37, issue 3, 358-360, 1998 which should be used for any reference to this work 1