Conformational dynamics in the F/G segment of CYP51 from Mycobacterium tuberculosis monitored by FRET

Abstract

A cysteine was introduced into the FG-loop (P187C) of CYP51 from Mycobacterium tuberculosis (MT) for selective labeling with BODIPY and fluorescence energy transfer (FRET) analysis. Förster radius for the BODIPY-heme pair was calculated assuming that the distance between the heme and Cys187 in solution corresponds to that in the crystal structure of ligand free MTCYP51. Interaction of MTCYP51 with azole inhibitors ketoconazole and fluconazole or the substrate analog estriol did not influence the fluorescence, but titration with the substrate lanosterol quenched BODIPY emission, the effect being proportional to the portion of substrate bound MTCYP51. The detected changes correspond to approximately 10A decrease in the calculated distance between BODIPY-Cys187 and the heme. The results confirm (1) functional importance of conformational motions in the MTCYP51 F/G segment and (2) applicability of FRET to monitor them in solution.

Figure 1

MTCYP51 mutagenesis. A. Location of P187 and internal cysteine residues in the MTCYP51 structure (distal view). Surface accessibilities: P187~65%, C37~16%, C151~7%, C442~18%. Cys 394 is coordinated to the heme iron and can not be modified (prepared in PyMOL). B. SDS -PAGE of MTCYP51 mutants in the absence of β-mercaptoethanol. Lane 1, A197G; lane 2, A197G/C37L/C151A/C446A; lane 3, A197G/C37L/C446A; lane 4, A197G/C37L/C446A/P187C; lane 5, A197G/C37L/C446A/P187C DTT reduced (1:200); lane 6, A197G/C37L/C446A/P187C-BODIPY; lane 7, rainbow marker.

Figure 2

MTCYP51 (A197G/C37L/C442A/P187C) – BODIPY, holo- (red line), apo- (black line); and unlabeled (blue line) protein. A. Absorbance spectra. The practically equal absorption at 280 nm in all three samples indicates that no significant changes occur in the structure of the protein globule upon heme removal. B. fluorescence spectra,. C. BODIPY modeled into the MTCYP51 structure [1H5Z] where P187 is mutated to C. In all calculations the assumption was made that the distance between the fluorophor and the heme in the solution corresponds to the distance between Cys187 and heme in the structure (31.97Å, as in Fig 4. In the model the distance is 31.91 as a result of energy minimization).

Figure 3

Changes upon titration with lanosterol (0.5, 1.5, 2.5, 3.5 and 4.5 μM). A. BODIPY emission. B. Soret band absorbance. Arrows show the directions of changes. A very similar pattern was observed upon titration of 6-IAF-labelled MTCYP51 with lanosterol (not shown) C. Correlation between decrease in BODIPY emission and high-spin content in the MTCYP51 sample. Because the dependence is linear up to the maximal reachable high-spin content for the MTCYP51 A197G/C37L/C442A/P187C-BODIPY (22%, Table 1), to be able to calculate total changes we approximated fluorescence changes to the 100% high-spin P450.

Figure 4

Conformational changes in the FG segment of MTCYP51 calculated using FRET. Upper view of the structure (residues 77–90 of the BC-loop are omitted). Red dotted line shows the calculated distance between the heme and Cys187 in lanosterol bound MTCYP51.