| 19484273 |
Hofmann M: Density functional theory study of model complexes for the revised nitrate reductase active site in Desulfovibrio desulfuricans NapA. J Biol Inorg Chem. 2009 Sep;14(7):1023-35. Epub 2009 May 30.
[Mo (SSCH3)(S2C2 (CH3) 2) 2](x) complexes with charges x between -3 and +3 were investigated by density functional theory computations as minimal nitrate reductase active-site models. The strongly reduced species (x = -2, -3) exist preferentially as pentacoordinate sulfo complexes separated from a thiolate anion. The oxidized extremes (x > 0) clearly prefer hexacoordinate complexes with an eta (2)-MeSS ligand. Among the neutral and especially for the singly negatively charged species structures with eta (2)-MeSS and eta (1)-MeSS ligands are energetically close to the sulfo methyl sulfide complex without SS bonding. For x = -1 the three isomers lie in a 1.5 kcal mol (-1) energy range. Putative mechanistic pathways for nitrate reduction from the literature were investigated computationally: (1) reduction at a pentacoordinate sulfo complex, (2) reduction at the ligand, and (3) reduction at the molybdenum center with an R-S-S ligand. All three pathways could be traced at least for some overall charges but no definite conclusion can be drawn about the mechanism. Complexes with larger dithiolato ligands were also computed in order to model the tricyclic metallopterin framework more accurately: the first heterocyclus (5,6-dihydro-2H-pyran) stabilizes the nitrate complex and the molybdenum oxo product complex by approximately 10 kcal mol (-1) and also reduces the activation barrier (by approximately 5 kcal mol (-1)). The effect of the second (1,2,3,4-tetrahydropyrazin) and third heterocyclus (2-amino-3H-pyrimidin-4-one) on the relative energies is relatively small. For bigger models derived from an experimental protein structure, nitrate reduction at a persulfo molybdenum (IV) complex fragment (mechanism 3) is clearly favored over the oxidation of a molybdenum-bound sulfur atom (mechanism 2). Mechanism 1 could not be investigated for the big models but seems the least favorable on the basis of the results from smaller models. |
1(0,0,0,1) |