The possibility that NO+ could bind to one of the two thioether sulfur atoms that covalently link the heme group of cytochrome c to the polypeptide chain, and that the resulting nitrososulfonium ion might undergo S–C cleavage to form a carbocation and a nitrosothiol were investigated computationally, by HF/3-21G(∗) geometry optimizations. The molecules studied were relatively small cytochrome model compounds with only the iron(II) porphyrin unit, and histidine and Me2S molecules coordinated as axial ligands to the heme iron, and R–S+(NO)–CHMe–porphyrin (R=cysteine–histidine, 1; n-butyl, 2; n-pentyl, 3; methyl, 4Me; n-propyl, 4Pr; CH2CH(NH2)CO–NH–CH(Me)–COOH, 5; and 1 without the iron and the Me2S, 6. Molecules 1, 2, 3 and 5 were stationary points on the potential energy surface (they exist at the HF/3-21G(∗) level); on attempted optimization of 4 and 6 the S–CHMe bond breaks tending to give a nitrosothiol RSNO and a delocalized cation +CHMe–porphyrin. This behavior was rationalized by idealizing the iron porphyrin moiety as Fe++ and porphyrin−−; electron density on the porphyrin ring, when not drained away into the iron, can assist with expulsion of the RSNO leaving group (equivalently, departure of RSNO gives a highly delocalized cation +CHMe–porphyrin). Besides iron, a minimum size for R (n-C4H9 or bigger) is needed to hold R in place, perhaps as a manifestation of the ponderal effect. From calculations on smaller molecules, rough CBS-4 water solvation-corrected reaction energies were estimated for loss of NO+ from 1, 2, and 3 (171, 154 and 154kJmol−1, respectively; binding is thermodynamically favored), and for S–CHMe bond breaking of 2 and 3 (−63 and −63kJmol−1, respectively; cleavage is thermodynamically favored).
Read full abstract