Carbon-13 and oxygen-17 chemical shifts, (¹⁶O/¹⁸O) isotope effects on¹³C chemical shifts, and vibrational frequencies of carbon monoxide in various solvents and of the Fe-C-O unit in carbonmonoxy heme proteins and synthetic model compounds

¹³C shieldings, δ(¹³C), ¹⁷O shieldings, δ(¹⁷O), and ¹⁸O isotope effects on ¹³C shieldings, ¹Δ¹³C(¹⁸O/¹⁶O), of carbon monoxide (99.7% ¹³C, 0.9% ¹⁷O, and 11.8% ¹⁸O enriched) in a variety of solvents and of the Fe-C-O unit of several carbonmonoxy hemoprotein models with varying polar and steric effects of the distal organic superstructure and constraints of the proximal side are reported. This enables, first, comparisons with hemoproteins, C-O vibrational frequencies, v(C-O), and X-ray structural data to be made; second, to investigate whether polarizable CO is an adequate model for distal ligand effects in carbonmonoxy heme proteins and synthetic model compounds; third, to investigate the effect of electronic perturbation within the heme pocket and pocket deformation on δ(¹³C), δ(¹⁷O), ¹Δ¹³C(¹⁸O/¹⁶O), and v(C-O). A variety of solvents with varying dielectric constants and solvation abilities appears to have negligible effect on δ(¹⁷O), δ(¹³C), and ¹Δ¹³C(¹⁸O/¹⁶O) and little direct effect on v(C-O) of dissolved carbon monoxide. On the contrary, ¹³C and ¹⁷O shieldings of several carbonmonoxy hemoprotein models vary widely and an excellent correlation was found between the infrared C-O vibrational frequencies and ¹³C shieldings and a reasonable correlation with ¹⁸O isotope effects on ¹³C shieldings. The ¹³C shieldings of heme models cover a 4.0 ppm range which is extended to 7.0 ppm when several HbCO and MbCO species at different pHs are included. The latter were found to obey a similar linear δ(¹³C) vs v(C-O) relationship. v(C-O), δ(¹³C), and ¹Δ¹³C(¹⁸O/¹⁶O) parameters of heme model compounds reflect similar interaction which is primarily the modulation of π back-bonding from Fe d_π to CO π* orbital by the distal pocket polar interactions. Our results suggest that, contrary to earlier claims, polarizable carbon monoxide is not an adequate model for distal ligand effects in carbonmonoxy hemoproteins and synthetic model compounds. Very probably this is caused by the large effect of the electric field on the back-bonding and the large polarizability of the π subsystem of the Fe-C-O unit. The ¹⁷O shieldings of heme models cover a range of 17 ppm which is extended to 24 ppm when selected heme proteins are included. The lack of correlation between δ(¹³C) and δ(¹⁷O) suggests that the two probes do not reflect a similar type of electronic and structural perturbation. δ(¹⁷O) is not primarily influenced by the local distal field interactions and does not correlate with any single structural property of the Fe-C-O unit; however, atropisomerism and deformation of the porphyrin geometry appear to play a significant role.