alcohol in excellent yield. Deprotection of the silyl ether with
TBAF followed by selective monoacylation of the primary
alcohol yielded the allylic acetate cyclization substrate 5.
Cyclization utilized Pd2(dba)3 (1.3 mol %) with Trost’s chiral
(S,S)-N-[2,(2′-diphenylphosphino)benzamido cyclohexyl] (2′-
diphenyl-phosphino) benzamide ligand [(S,S)-DPPBA (5.3
mol %)] to produce 6 (96%) as a single diastereomer.5 With
the third stereocenter in place, oxidative cleavage of the vinyl
group produced the corresponding aldehyde, which was
oxidized under mild conditions with sodium hypochlorite to
afford the N-Cbz-protected amino acid 7.
Scheme 1. Monomer Synthesis
The monomer unit synthesis was completed by protection
of the acid as the ethyl ester followed by deprotection of
the dimethyl N,O-acetal with catalytic CSA in methanol. The
resulting hydroxymethyl side chain was protected as the TBS
ether to provide 9.
With the protected monomer in hand, saponification of
the ethyl ester or hydrogenolysis of the benzyl carbamate
gave the primary coupling partners, acid 10 and amine 11,
respectively. In early experiments, the methyl ester was
employed as the carboxyl protecting group, but the free
amine undergoes competing cyclization to the bicyclic [3.3.1]
lactam under the peptide coupling conditions. The ethyl ester
proved hindered enough to effectively prevent this deleterious
side reaction.
Standard peptide bond formation using diisopropylcarbo-
diimide (DIC) with stoichiometric 1-hydroxybenzotriazole
(HOBt) effected coupling, affording 12 in good yield
(Scheme 2). Hydrogenolysis followed by coupling to another
Scheme 2. Iterative Coupling Sequence
As detailed in Scheme 1, the cyclization substrate would
arise from Grignard addition of the TBS derivative of
6-chlorohex-2-en-1-ol (2)6 to N-Cbz-protected Garner’s al-
dehyde (1), which is readily available from L-serine in four
steps.7 There is precedence for stereoselective addition to
Garner’s aldehyde with reactive nucleophiles, such as
vinylmagnesium bromide or vinyllithium, via Felkin-Ahn
preferred attack to produce the desired erythro alcohol.8
Unfortunately, less reactive nucleophiles, such as the Grig-
nard reagent prepared from 2, react with little stereoselec-
tivity, giving the resulting alcohol as an inseparable mixture
of diastereomers (1:1, determined by HPLC). The desired
alcohol diastereomer 4 was stereoselectively produced via
Swern oxidation followed by chelation-controlled (anti-
Felkin-Ahn) reduction under modified Luche conditions.9
Using a procedure developed and optimized specifically for
this reaction, Zn(BH4)210 with CeCl3 gave the desired erythro
(6) Brennan, J. P.; Saxton, J. E. Tetrahedron 1986, 42, 6719-6734.
(7) (a) Garner, P. Tetrahedron Lett. 1984, 25, 5855-5858. (b) Garner,
P.; Park, J. M. J. Org. Chem. 1987, 52, 2361-2364.
(8) (a) Ojima, I.; Vidal, E. S. J. Org. Chem. 1998, 63, 7999-8003. (b)
Ageno, G.; Banfi, L.; Cascio, G.; Guanti, G.; Manghisi, E.; Riva, R.; Rocca,
V. Tetrahedron 1995, 51, 8121-8134. (c) For alternate chelation theory
see: Coleman, R. S.; Carpenter, A. J. Tetrahedron Lett. 1992, 33, 1697-
1700.
(9) Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103, 5454-
5459.
2274
Org. Lett., Vol. 4, No. 13, 2002