(corrin synthesis/vitamin B12 biosynthesis/sulfide contractions/nucleop

Crystalline nickel(II) and dicyanocobalt(III) complexes of a racemic 1,2,2,7,7,12,12,17,17,20-decamethyl-20-hy, droxy-1,2,3,7,8,12,13,20-octahydro-17H-porphyrin rearrange to the corresponding complexes of racemic 19-acetyl-l,2,2,7,7,12,12,17,17-nonamethyl-trans-corrin on melting; (approximately 290?C and 260'C, respectively). The nickel(II) 19-acetylcorrinate formed in this way is shown to deacetylate.to racemic nickel(II) 1,2,2,7,7,12,12,17,17-nonamethyl-trans-corrinate on treatment with 2 M KOH. These reactions are being studied as potentially biomimetic chemical models for the elusive ring contraction step in vitamin B12 biosynthesis. The crucial and still.unknown step in the biosynthesis of vitamin B12 from uroporphyrinogen (1, 2) is the generation of the characteristic A/D-ring junction of the corrinoid ligand. A major stride towards a solution of this problem was made in 1979 when it was discovered (3-5) that the third methyl group in the consecutive methyl group transfers from adenosylmethionine to hydroporphinoid intermediates most surprisingly becomes attached at the meso position C-20 between rings A and D (cf. structure 1 in Fig. 1), only to leave the scene again, probably together with carbon atom C-20, at some unknown later stage of the uroporphyrinogen -corrin transformation. It was this important finding combined with the results of recent chemical studies on the mechanistic spectrum of A/D-secocorrin -> corrin ring closures (6, 7) that led us to envisage and to test in a model system the chemical feasibility and characteristics of the hypothetical dihydrocorphinolt -> corrin rearrangement depicted in Fig. 2.? In a hexahydroporphinoid structure of type 2 protonation of the enaminoid double bond of ring D generates the structure 2a containing an a-tert-hydroxyiminium function, which in principle should be prone to undergo two isomeric rearrangements, one of which constitutes a ring contraction to the 19-acetylcorrin. structure 3. Such a transformation is formally analogous to the known semibenzoin rearrangement of a-tert-hydroxyketones and is in fact-as is best.recognized in its stepwise mechanistic variant (see structure 2b)-very closely related to our previously described (7) A/D-secocorrin-> corrin cycloisomerization of 1-methylidene-19-formyl-1, 19-secocorrinates to 19-formylcorrinates. A process of type 2 -> 3 implies the isomerization of an allylic alcohol system to an unconjugated carbonyl function and may well, therefore, be thermodynamically feasible in the free ligand; on the other. hand, in metal complexes the position of a corresponding equilibrium may reflect the preference of a given metal ion for either the hydroporphinoid or the (smaller) corrinoid coordination sphere. Kinetically, the rearrangement is expected to respond to acid/base catalysis and it may also offer a thermally inducible pathway via an intramolecular proton shift The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. ?1734 solely to indicate this fact. 16 id to corrinoid complexes hilic carbenes/photochemistry) FKYt, AND ALBERT ESCHENMOSER* versitAitstrasse 16, CH-8092 Zulrich, Switzerland; and tInstitute of Physical'