A Carbonyl Ylide Cycloaddition Approach to Platensimycin

Abstract

Platensimycin (1) is a novel broad-spectrum antibiotic (against Gram-positive bacteria) which was isolated from Streptomyces platensis by scientists from Merck: it inhibits bacterial growth by selectively inhibiting the condensing enzyme FabF of the bacterial fatty acid synthesis pathway. [1] Platensimycin (1) shows no cross-resistance to methicillinresistant Staphylococcus aureus, vancomycin-intermediate S. aureus, and vancomycin-resistant enterococci. As a result of its remarkable biological profile and challenging structure, platensimycin (1) has been the focus of intense synthetic activity, [2] and herein we describe the results of our recent efforts towards the synthesis of this intriguing compound.At the outset, a synthesis of the pivotal tetracyclic intermediate 2 from the cagelike ketone A was envisioned (Scheme 1). Rhodium(II)-catalyzed decomposition of a diazoketone D would lead to the formation of A and/or B through [3+2] cycloaddition [3] of the corresponding carbonyl ylide with conformations C and/or C'. This type of cycloaddition in the presence of rhodium(II) acetate is known to favor the formation of B (R = H; A/B/cyclopropanes = 6:41:36), [4] and a reversal of this product distribution seemed necessary to ensure the success of our synthetic approach.In practice, preparation of the quaternary-substituted diazoketone D was problematic.[5] After considerable experimentation, we found that the reaction sequence was successful with R = CN. Thus, diazoketone 6 was prepared from ethyl cyanoacetate (3) in a straightforward manner by sequential alkylation (Scheme 2). In the presence of rhodium(II) acetate, decomposition of diazoketone 6 proceeded smoothly to yield the cagelike ketone 8, accompanied by only a trace amount of the desired product 7, and a small amount of cyclopropane products 9. The use of rhodium(II) trifluoroacetate led to a clean conversion of 6 into 8.At first glance these results were disappointing, but we recognized that the overall cyclization-cycloaddition process was more competitive than the cyclopropanation reaction, and that the LUMO-dipole/HOMO-dipolarophile (type III) interaction [6] clearly dominated in the cycloaddition process because of the introduction of a nitrile group. The desired regioselectivity would be attained through the reversal of the olefin (dipolarophile) HOMO coefficient, for which halogen substitution emerged as an appropriate possibility. Accordingly, the halogenated substrates 10-13 were prepared from 4 for further regioselectivity studies.Rhodium(II)-catalyzed decomposition of 10 led to the relatively clean production of ketone 14 (Scheme 3). Under similar conditions, the (Z)-bromide 11 produced a relatively complex mixture from which 15 and 16 [7] were isolated in low yields. Gratifyingly, the (E)-bromide 12 was converted into Scheme 1. Retrosynthetic analysis of platensimycin (1).Scheme 2. Prototype carbonyl ylide [3+2] cycloaddition. a) NaOMe, MeCOCH 2 Cl, MeOH; b) NaH, CH 2 CHCH 2 Br, THF; c) 1 n KOH, MeOH; d) ClCO 2 iBu, TEA, diethyl ether, 0 8C; ...

Fortschritte in der Chemie und Biologie natürlicher Antibiotika

“…Retrosynthetische Analyse (a) und Synthese (b) von 310 über Ketyl-Olefin-Cyclisierung (Nicolaou, Chen et al, 2007). [246] K. C. Nicolaou [246] Der letzte bislang beschriebene Ansatz zur Platensimycinsynthese stammt von Eun Lee et al [248] und basiert auf einer Carbonyl-Ylid-Cycloaddition. [249] Die Autoren zerlegten den Enonring in 310 durch Umkehrung einer Aldolkondensation und einer Olefinierung, was über 408 zum Ketonitril 409 zurückführte (Abbildung 32).…”

“…Oxidation von 407, TMS-Enol-Etherbildung und wieder Oxidation lieferten 310 als 2:1-Regioisomerengemisch, was die Selektivität der Silylierung widerspiegelt. [246] Der letzte bislang beschriebene Ansatz zur Platensimycinsynthese stammt von Eun Lee et al [248] und basiert auf einer Carbonyl-Ylid-Cycloaddition. [249] [248] Totalsynthesen von Antibiotika Angewandte Chemie mere Keton bildete sich ebenfalls, und zwar in 23 % Ausbeute, was die Selektivität der Hydrosilylierung widerspiegelt.…”

“…[246] Der letzte bislang beschriebene Ansatz zur Platensimycinsynthese stammt von Eun Lee et al [248] und basiert auf einer Carbonyl-Ylid-Cycloaddition. [249] [248] Totalsynthesen von Antibiotika Angewandte Chemie mere Keton bildete sich ebenfalls, und zwar in 23 % Ausbeute, was die Selektivität der Hydrosilylierung widerspiegelt. Die formale asymmetrische Synthese wurde durch eine effiziente säurekatalysierte Aldolkondensation abgeschlossen, die 310 in 96 % Ausbeute lieferte.…”

“…Die formale asymmetrische Synthese wurde durch eine effiziente säurekatalysierte Aldolkondensation abgeschlossen, die 310 in 96 % Ausbeute lieferte. [248] Zwei bioaktive Analoga von Platensimycin wurden von der Nicolaou-Gruppe beschrieben. (À)-Adamantaplatensimycin (425, Schema 60) [253] wurde durch eine Rhodiumcarben-C-H-Insertion [254] in 422 zu 423 hergestellt (422 wurde aus einer kommerziell erhältlichen Adamantanvorstufe synthetisiert).…”

Dialkylamino‐2,4‐dihydroxybenzoic Acids as Easily Synthesized Analogues of Platensimycin and Platencin with Comparable Antibacterial Properties

“…It is just over four years since a group at Merck reported the identification of a novel class of antibiotics called platensimycin [1] or platencin, [2] which are FabF/H inhibitors and therefore inhibit the synthesis of lipids used by bacteria to make cell membranes. In only four years since the seminal report by the Merck group, over 20 groups worldwide have reported the total [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] or enzymatic syntheses [20,21] of platensimycin/platencin or analogues [22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38] thereof, and about 12 papers [22,23,[28][29][30][31][32][33][34][36]…”