The Evolution of Azole Resistance in Candida albicans Sterol 14α-Demethylase (CYP51) through Incremental Amino Acid Substitutions

Abstract

Recombinant Candida albicans CYP51 (CaCYP51) proteins containing 23 single and 5 double amino acid substitutions found in clinical strains and the wild-type enzyme were expressed in Escherichia coli and purified by Ni²⁺-nitrilotriacetic acid agarose chromatography. Catalytic tolerance to azole antifungals was assessed by determination of the concentration causing 50% enzyme inhibition (IC₅₀) using CYP51 reconstitution assays. The greatest increase in the IC₅₀ compared to that of the wild-type enzyme was observed with the five double substitutions Y132F+K143R (15.3-fold), Y132H+K143R (22.1-fold), Y132F+F145L (10.1-fold), G307S+G450E (13-fold), and D278N+G464S (3.3-fold). The single substitutions K143R, D278N, S279F, S405F, G448E, and G450E conferred at least 2-fold increases in the fluconazole IC₅₀, and the Y132F, F145L, Y257H, Y447H, V456I, G464S, R467K, and I471T substitutions conferred increased residual CYP51 activity at high fluconazole concentrations. In vitro testing of select CaCYP51 mutations in C. albicans showed that the Y132F, Y132H, K143R, F145L, S405F, G448E, G450E, G464S, Y132F+K143R, Y132F+F145L, and D278N+G464S substitutions conferred at least a 2-fold increase in the fluconazole MIC. The catalytic tolerance of the purified proteins to voriconazole, itraconazole, and posaconazole was far lower and limited to increased residual activities at high triazole concentrations for certain mutations rather than large increases in IC₅₀ values. Itraconazole was the most effective at inhibiting CaCYP51. However, when tested against CaCYP51 mutant strains, posaconazole seemed to be the most resistant to changes in MIC as a result of CYP51 mutation compared to itraconazole, voriconazole, or fluconazole.

Keywords: Candida albicans CYP51; azole; mutations.

FIG 1

Absolute spectra of the wild-type CaCYP51 protein and CaCYP51 proteins with double amino acid substitutions. The purified CaCYP51 proteins were diluted 20-fold with 0.1 M Tris-HCl buffer (pH 8.1) and 20% glycerol, and the absolute spectra in the resting oxidized state were determined between 300 and 700 nm. The absolute spectra for the CaCYP51 proteins with single amino acid substitutions can be found in Fig. S1 in the supplemental material.

FIG 2

Triazole IC₅₀ determinations for wild-type CaCYP51 and CaCYP51 proteins containing Y132F, Y132H, K143R, and F145L amino acid substitutions. IC₅₀ values were determined for fluconazole (filled circles), voriconazole (hollow circles), itraconazole (filled triangles), and posaconazole (crosses). IC₅₀ assays were performed in duplicate. Mean data points together with standard deviations (as error bars) are shown.

FIG 3

Triazole IC₅₀ determinations for CaCYP51 proteins containing D278N, G307S, G450E, and G464S amino acid substitutions. IC₅₀ values were determined for fluconazole (filled circles), voriconazole (hollow circles), itraconazole (filled triangles), and posaconazole (crosses). IC₅₀ assays were performed in duplicate. Mean data points together with standard deviations (as error bars) are shown.

FIG 4

Triazole IC₅₀ determinations for CaCYP51 proteins containing S279F, S405F, Y447H, G448E, R467K, and I471T amino acid substitutions. IC₅₀ values were determined for fluconazole (filled circles), voriconazole (hollow circles), itraconazole (filled triangles), and posaconazole (crosses). IC₅₀ assays were performed in duplicate. Mean data points together with standard deviations (as error bars) are shown.

FIG 5

Fluconazole binding properties of CaCYP51 proteins. Type II difference spectra for 4 μM wild-type CaCYP51 and the five CaCYP51 proteins with double substitutions are shown along with the ligand binding saturation curves for the wild-type protein (filled circles) and proteins with the Y132F+K143R (hollow circles), Y132H+K143R (filled triangles), Y132F+F145L (crosses), D278N+G464S (hollow triangles), and G307S+G450E (hollow squares) substitutions. Ligand binding determinations were performed in triplicate, although only one replicate is shown. Ligand saturation curves were fitted using the modified Morrison equation (54).