J. Amer. Chem. Soc. 1991, 113, 9877.

 

Oxidation of Molybdenum Dithiolene Complexes Yields Thiophene

Analogs of Urothione and Molybdopterin Form B

Cheryl L. Soricelli, Veronika A. Szalai, Sharon J. Nieter Burgmayer*

Contribution from the Department of Chemistry, Bryn Mawr College, Bryn Mawr, Pennsylvania 19010

Molybdopterin is the pterin component of the molybdenum cofactor [Mo-co] required by oxo-molybdenum enzymes such as xanthine oxidase, sulfite oxidase and nitrate reductase. Destruction of Mo-co by various oxidative methods yields many different pterin products and several of these products are unique to Mo-co. The identification of two degradation products and the metabolic product of Mo-co, Form A, Form B and urothione, respectively, ultimately led to a proposed structure for molybdopterin. A dithiolene group is one unusual feature of this structure and its presence has been supported by subsequent chemical and spectroscopic means.

The initial report of the molybdopterin structure suggested that the dithiolene group served to bind the Mo atom. Results from recent Resonance Raman studies on Mo-dithiolene model complexes and on Mo-co from flavin-free dimethylsulfoxide reductase support this idea.

The alkyne group in Form A suggests a strategy for a chemical synthesis of Mo-bound molybdopterin. Dithiolene ligands are known products from reactions of molybdenum tetrasulfide complexes and electron-deficient acetylenes. We and others have studied reactions of metal tetrasulfides with acetylenes substituted by nitrogen heterocycles to investigate whether N-heterocycles are sufficiently electron-deficient to activate the acetylene bond. Acetylenes substituted by both quinoxaline and pterin are successful in these reactions and form dithiolene complexes in good yields. We now report that oxidative degradation of Mo-(quinoxalyl)dithiolene complexes produces quinoxalylthiophene molecules structually related to both urothione and Form B. Quinoxalylthiophene formation from these new Mo (quinoxalyl)dithiolene complexes is the first evidence to support the proposed transformation of a dithiolene ligand of Mo-co to the thiophenes urothione and Form B.

[TEA]2[Mo(S9)] reacts with phenylethynylquinoxaline (PEQO) under conditions similar to other literature syntheses for Mo-dithiolenes to give [TEA]2[Mo(S2PEQO)3] 1 in ca. 60 %.

Like other Mo (tris)dithiolene complexes, compound 1 exhibits two reversible couples at +210 mV and -230 mV in acetonitrile vs Ag/AgCl . These potentials indicate that the corresponding Mo(V)- and Mo(VI)-tris(quinoxalyl)dithiolene complexes should be chemically accessible using mild oxidants. Iodine oxidation of 1 yields both [TEA][MoV(S2PEQO)3] and MoVI(S2PEQO)3 in amounts that depend on the reaction conditions. In addition to oxidized molybdenum compounds, a third product, compound 2, was isolated from I2 oxidation. 2 is yellow and has an infrared spectrum

very similar to the S2PEQO dithiolene ligand but it contains no molybdenum. The 1H NMR spectrum of 2 shows that all protons of S2PEQO are intact with the exception of H3. This information led us to speculate that compound 2 was a thiophene derivative of the [-S2PEQO]2- ligand , a hypothesis proven correct by an x-ray crystal structure analysis.

A view of the molecular structure of 2-(phenyl)thieno[3,2-g]quinoxaline, 2, in the Figure reveals the fate of S2PEQO ligand oxidation. A thiophene ring fused to quinoxaline is formed from cyclization of the b-thiolate at quinoxaline C3. Oxidation of the a-thiolate causes formation of a disulfide bond to a second quinoxalylthiophene moiety. A crystallographic C2 axis passes through the midpoint of the disulfide bond and relates one quinoxalylthiophene plane to its molecular partner. Bond distances and angles within this molecule are unremarkable since they reproduce values previously reported.

Compound 2 is obtained from 1 using a variety of oxidants (I2, CeIV, O2 and S8) as well as from solutions of the Mo(V) and Mo(VI) tris-dithiolene complexes after long exposure to the atmosphere. In fact, we have not yet accomplished an oxidation of [TEA]2[Mo(S2PEQO)3] that does not yield some of compound 2. Our continued study of these reactions seeks to determine if formation of 2 proceeds through a particular Mo oxidation state and if quinoxaline N-coordination aids a dithiolene cis-trans isomerization that must precede thiophene ring closure.

A second degradation product, compound 3, has been obtained in small amounts from recrystallization attempts using impure Mo-(quinoxalyl)dithiolene complexes. Our preliminary report on this material presents its structure determined from 1H and 13C NMR and x-ray analysis. As depicted schematically below, 3 is also a thiophene derivative of quinoxaline where the exocyclic sulfur is bridged to a second 3-thio(quinoxalyl)thiophene by a methylene group.

Decomposition products 2 and 3 isolated from Mo complexes having -S2PEQO dithiolene ligands demonstrate for the first time that thiophene cyclization is a likely decomposition result from such dithiolene complexes. The isolation of both the disulfide-bridged and the S-alkylated products indicates multiple decomposition pathways as has been observed for Mo-co decomposition leading to two pterinylthiophene compounds, urothione and Form B. These results provide the needed experimental support to link the known structures of urothione and Form B to the proposed pterinyl-dithiolene unit in Mo-co.

Acknowledgements. This work was supported by NIH grant GM-42049-01 and a grant from the Pew Science Consortium. We thank Patrick J. Carroll of the University of Pennsylvania for the collection of x-ray data and we are grateful to R. Pilato and E. I. Stiefel for private communication of unpublished results.

Supplementary Material Available: a listing of analytical and physical properties (1 page), crystallographic collection and solution data, atom positions, thermal parameters, bond distances and angles ( 5 pages), and tables of Fobs and Fcalc (19 pages). Ordering information is given on any current masthead page.