Solution 1H NMR of the Active Site of Substrate-bound, Cyanide-inhibited Human Heme Oxygenase

Abstract

The majority of the active site residues of cyanideinhibited, substrate-bound human heme oxygenase have been assigned on the basis of two-dimensional NMR using the crystal structure of the water-ligated substrate complex as a guide (Schuller, D. J., Wilks, A., Ortiz de Montellano, P. R., and Poulos, T. L. (1999) Nat. Struct. Biol. 6, 860 -867). The proximal helix and the N-terminal portion of the distal helix are found to be identical to those in the crystal except that the heme for the major isomer (ϳ75-80%) in solution is rotated 180°a bout the ␣-␥-meso axis relative to the unique orientation in the crystal. The central portion of the distal helix in solution is translated slightly over the heme toward the distal ligand, and a distal four-ring aromatic cluster has moved 1-2 Å closer to the heme, which allows for strong hydrogen bonds between the hydroxyls of Tyr-58 and Tyr-137. These latter interactions are proposed to stabilize the closed pocket conducive to the high stereospecificity of the ␣-meso ring opening. The determination of the magnetic axes, for which the major axis is controlled by the Fe-CN orientation, reveals a ϳ20°tilt of the distal ligand from the heme normal in the direction of the ␣-meso bridge, demonstrating that the close placement of the distal helix over the heme exerts control of stereospecificity by both blocking access to the ␤, ␥, and ␦-meso positions and tilting the axial ligand, a proposed peroxide, toward the ␣-meso position. Heme oxygenase (HO)1 is a membrane-bound protein that carries out the NADPH-, O 2 -, and cytochrome P450 reductasedependent regiospecific catabolism of heme to iron, ␣-biliverdin, and carbon monoxide (1). There are two well-established mammalian isoforms, the 288-residue inducible HO1 (2, 3) and the 316-residue constitutive HO2 (4), that share significant identity and have a common substrate, mechanisms, and products. The sequence of HO fails to exhibit significant identity to any previously structurally characterized enzyme, and hence, in the absence of a molecular model, the early research on the soluble enzyme focused on functional (5-13), mutational (5, 11), and spectroscopic (5, 14 -18) investigations. The current understanding of HO is largely based on investigation of a recombinant, truncated 265-residue portion of HO1 from which the membrane-binding domain has been deleted, but which retains full activity (13). The mechanism of HO resembles that of cytochrome P450 in its ability to oxidize unactivated C-H bonds (19). However, in contrast to cytochrome P450 and the heme peroxidases, the action of HO does not pass through a ferryl intermediate (13,20) but appears rather to involve a peroxo ligand capable of attacking the ␣-meso bridge.Spectroscopic investigations agree on a normal iron-His (5-7, 14, 15) bond that more closely resembles that of Mb, with a neutral imidazole ligand, than that of the peroxidases, in which the axial ligand has significant imidazolate character (21). Mutational work identified His-25 as the axial ligand (5, 22). Whereas there...

Solution NMR study of environmental effects on substrate seating in human heme oxygenase: Influence of polypeptide truncation, substrate modification and axial ligand

“…The resolved portion of the 1 H NMR spectra of equilibrated hHO-PH-CN [12,43] and hHO-DMDH-CN [13] in 1 H 2 O at 25°are illustrated in Fig. 3(d) and (e), respectively.…”

“…Hence only limited 2D data are provided in the Supporting Information. The chemical shifts for the heme, strongly dipolar shifted active site residues and labile proton involved in strong H-bonds of hHO*-PH-CN isomer, are listed in Table 1, and compared to the same data of the 265-residue hHO-PH-CN [12,31,43] and hHO-DMDH-CN complexes [13,44]. The chemical shift for other assigned residues are provided in the Supporting Information.…”

“…3(d) and (e), respectively. The complete heme assignment of the latter unique complex [13], and the major isomer of the former complex [12,31,43], have been presented. The significant differences in the heme methyl hyperfine [27] (largely contact) shifts for PH and DMDH (for all but the propionate C b Hs) arise from variable in-plane electronic influences of 2,4-vinyl versus 2,4-methyl substituents [27,45].…”

“…The detailed procedure for definitive assignment of the heme, axial His, and the hyperfine shifted portions of the proximal and distal helices, as well as the members of the distal H-bond network, have been described in detail for both the major isomers of hHO-PH-CN [12,31,43] and the unique form of hHO-DMDH-CN [13]. The obvious similarities of the position of the assigned resolved signal for the major or A isomer of hHO*-PH-CN in Fig.…”

“…3(a)-(c), respectively. Peaks are labeled by upper case M i , H i for methyls and single protons for the major (A) or sole isomer, as assigned previously [12,13,31,43], where i designates the position on the heme in the Fischer notation ( Fig. 1), and the sequence number of a residue.…”