A PRELIMINARY 3D MODEL FOR CYTOCHROME-P450 2D6 CONSTRUCTED BY HOMOLOGY MODEL-BUILDING

A homology model building study of cytochrome P450 2D6 has been carried out based on the crystal structure of cytochrome P450 101. The primary sequences of P450 101 and P450 2D6 were aligned by making use of an automated alignment procedure. This alignment was adjusted manually by matching alpha-helices (C, D, G, I, J, K and L) and beta-sheets (beta3/beta4) of P450 101 that are proposed to be conserved in membrane-bound P450s (Ouzounis and Melvin [Eur. J. Biochem., 198 (1991) 307]) to the corresponding regions in the primary amino acid sequence of P450 2D6. Furthermore, alpha-helices B, B' and F were found to be conserved in P450 2D6. No significant homology between the remaining regions of P450 101 and P450 2D6 could be found and these regions were therefore deleted. A 3D model of P450 2D6 was constructed by copying the coordinates of the residues from the crystal structure of P450 101 to the corresponding residues in P450 2D6. The regions without a significant homology with P450 101 were not incorporated into the model. After energy-minimization of the resulting 3D model of P450 2D6, possible active site residues were identified by fitting the substrates debrisoquine and dextrometorphan into the proposed active site. Both substrates could be positioned into a planar pocket near the heme region formed by residues Val370, Pro371, Leu372, Trp316, and part of the oxygen binding site of P450 2D6. Furthermore, the carboxylate group of either Asp100 or Asp301 was identified as a possible candidate for the proposed interaction with basic nitrogen atom(s) of the substrates. These findings are in accordance with a recently published predictive model for substrates of P450 2D6 [Koymans et al., Chem. Res. Toxicol., 5 (1992) 211].