The total syntheses of bleomycin A2 (1) by two routes are described. The final step in the synthesis
of bleomycin A2 involves methylation of bleomycin demethyl A2 (2). This bleomycin derivative is of interest
mechanistically, and can also provide access to other bleomycins via its known chemical conversion to
bleomycinic acid. Accordingly, the synthetic strategy presented represents a particularly versatile approach
for the elaboration of a wide variety of BLM congeners. Bleomycin was constructed from five key intermediates,
the syntheses of which are described. 1,6-Di-O-acetyl-3,4-di-O-benzyl-2-O-[2,4,6-tri-O-acetyl-3-O-(N-acetylcarbamoyl)-α-d-mannopyranosyl]-β-l-gulopyranose (3) was converted quantitatively to its disaccharide
chloride (4), the latter of which was condensed with N
α,N
im-bis(t-Boc)-(S)-erythro-β-hydroxyhistidine (7) to
provide α-O-glycosidated product 16. The subsequent couplings with benzyl valerate 8, threonylbithiazole 9,
and N
α
-t-Boc-pyrimidoblamic acid (10) afforded access to bleomycin demethyl A2 (2) and decarbamoyl
bleomycin demethyl A2 (26). Although it was obtained as a byproduct of the synthesis of BLM, the synthesis
of decarbamoyl bleomycin demethyl A2 nonetheless constitutes the first report of synthetic access to this BLM
congener that is of particular importance from a mechanistic perspective. This BLM can be used to resolve
the issue of the participation of the carbamoyl group as a metal ligand in metallobleomycins. Also reported
is a new route to bleomycin demethyl A2 that employed an unprotected carbamoyl group throughout the
synthesis. In conjunction with the use of the uncharged methylthiopropylamide C-substituent, the latter strategy
permitted improved functional group manipulation and thereby afforded intermediates that could be purified
with greater facility. Because the synthesis of bleomycin is limited by the efficiency of elaboration of the
carbohydrate moiety, a study was carried out to define improved methods for the preparation of this constituent
of BLM. Three new routes were explored, and a route involving the intermediacy of disaccharide activated
as a glycosyl bromide was found to be particularly efficient and convenient. Also of importance was the
finding that activation of carbohydrate intermediates as their glycosyl trichloroacetimidates permitted the requisite
couplings to be carried out conveniently and in good yields. Synthetic BLM demethyl A2 was shown to have
the same potency as a naturally derived sample in a plasmid DNA relaxation assay. The sequence selectivity
of DNA cleavage by synthetic and authentic Fe(II)·BLMs was also shown to be the same. Also established
for the first time for any synthetic bleomycin was the actual chemistry of DNA degradation, which is the same
as that for naturally derived bleomycins.