Kishi lactam

F. Luzzio and co-workers devised a total synthesis for both antipodes of the (-)-Kishi iactam, which is a versatiie intermediate for the synthesis of the perhydrohistrionicotoxin (pHTX) aikaioids. in the final stages of the synthesis of the (-)-Kishi lactam, it was necessary to remove one of the secondary alcohol groups. The Barton radical deoxygenation protocol was utilized for this operation. 

The double Henry reaction was examined by Luzzio and Fitch for the synthesis of a key intermediate 21 of perhydrohistrionicotoxin (22) [3]. Reaction of nitroalkane 15 and glutaraldehyde 16 using TMG (3) in dry tetrahydrofiiran (THF) proceeded via a double nitroaldol reaction to give meso-17 in 80-87% yield. After conversion of the diol to meso lactamdiacetate 19, esterase mediated hydrolysis gave optically active 20 in 87% yield with 93% ee (Scheme 7.2). This hydroxyacetate was successfully led to the Kishi lactam (21) [4], a key intermediate for 22. 

Luzzio, F.A. and Fitch, R.W. (1999) Formal synthesis of (+)- and (—)-perhydrohistrionicotoxin a double Henry /enzymatic desymmetrization route to the Kishi lactam. The Journal of Organic Chemistry, 64, 5485-5493. 

Wardrop, D.J., Zhang, W., and Landrie, C.L (2004) Stereoselective nitrenium ion cyclizations asymmetric synthesis of the (-l-)-Kishi lactam and an intermediate for the preparation of fiisicularin. Tetrahedron Lett., 45, 4229-4231. 

Hydroxy-lactam 26 (n=2) gave spiroisomer 27 (n=2), albeit in lower yield, under similar conditions. The same authors 191 have also reported the successful cyclization of hydroxy-lactam 29 into spirolactam 30. Analogous results were obtained by Evans and Thomas (10) who found that the cyclization of a 9 1 mixture of enamides 31 and 32 in anhydrous formic acid gave the spirocompound 30. This compound is a key intermediate in Kishi s total synthesis of perhydrohistrionicotoxin (11, 12). 

The first highly stereocontrolled total synthesis of a natural penicillin was reported 2 years later by Baldwin et al. <1976JA3045>. In this case, the methodology relies on the formation of the fi-lactam ring before the thiazolidine ring closure, via the sulfenic acid intermediate 97 (Rz = OH), which gives electrophilic attack on the double bond to produce a penam sulfoxide 98 (see Section 2.03.5.3) (Scheme 52). A similar route has been developed independently by Kishi for the total synthesis of 6cr-methoxy penicillin derivatives <1975JA5008>. 

In the studies of the synthesis of the ansamycin antibiotic rifamycin S (13S), Corey and Clark [76] found numerous attempts to effect the lactam closure of the linear precursor 132 to 134 uniformly unsuccessful under a variety of experimental conditions, e.g. via activated ester with imidazole and mixed benzoic anhydride. The crux of the problem was associated with the quinone system which so deactivates the amino group to prevent its attachment to mildly activated carboxylic derivatives. Cyclization was achieved after conversion of the quinone system to the hydroquinone system. Thus, as shown in Scheme 45, treatment of 132 with 10 equiv of isobutyl chloroformate and 1 eqtuv of triethylamine at 23 °C produced the corresponding mixed carbonic anhydride in 95% yield. The quinone C=C bond was reduced by hydrogenation with Lindlar catalyst at low temperature. A cold solution of the hydroquinone was added over 2 h to THF at 50 °C and stirred for an additional 12 h at the same temperature. Oxidation with aqueous potassium ferricyanide afforded the cyclic product 134 in 80% yield. Kishi and coworkers [73] gained a similar result by using mixed ethyl carbonic anhydride. 

The next perhydrohistrionicotoxin synthesis we will examine was reported by the Keck group (Utah). This synthesis featured chemistry of N-acylnitroso compounds, just as did the Keck synthesis of pyrrolizidines (Chapter 4). Rather than approaching this in a retrosynthetic manner , suffice it to say that this work intersects with the Kishi synthesis at the point of keto-lactam 46. Now we will jump right into the synthesis. 

---