Consistency of the membrane-bound models of CYP2C9 with experimental data.

1. For experiments based on site-directed antibodies and Trp fluorescence quenching, the experimental observable is shown as the residue numbers mapped on CYP2C9 based on sequence alignment with the CYP used in the experiment.2. The structures corresponding to the maximum peaks of the orientation histograms (Fig. 3) were compared with experiments.3. The location of protein residues with respect to the membrane is classified as: (i) M: lipid tail region of the membrane, (ii) HG: the head group region, (iii) C: the cytosol. For each peptide, the location in the models was assessed starting from the N-terminal residue (e.g. a peptide with its location identified as ‘M-HG-C’ has its N-terminal residues in the membrane and the C-terminal residues in the cytosol).4. Peptides for which the location in the models does not agree with the location inferred from the experiments.5. The peptides were either accessible (A) or inaccessible (IA) when site-directed antibodies were tested on microsomes.6. From the experiments based on TRP fluorescence quenching, it was inferred that residue 380 is the deepest buried in the membrane, residues 36 and 69 are in the membrane but not as deep as 380, while residues 80 and 225 are in the region of the head groups or do not contact the membrane (these 2 options are indistinguishable in the experiments). In the simulations, residue 380 is either in the head group region or lies shallow in the membrane, while residue 225 is either positioned shallow in the membrane or in the head group region.b. The peptide is not in contact with the membrane but is buried in the protein globular domain.*The agreement with experiment was assessed as follows: (i) ‘+’ = good, (ii) ‘0’ = possible, (iii) ‘-’ = disagreement.7. (a) [2]; (b) [8]; (c) [11]; (d) [10].