Azole affinity of sterol 14α-demethylase (CYP51) enzymes from Candida albicans and Homo sapiens

Abstract

Candida albicans CYP51 (CaCYP51) (Erg11), full-length Homo sapiens CYP51 (HsCYP51), and truncated Δ60HsCYP51 were expressed in Escherichia coli and purified to homogeneity. CaCYP51 and both HsCYP51 enzymes bound lanosterol (K(s), 14 to 18 μM) and catalyzed the 14α-demethylation of lanosterol using Homo sapiens cytochrome P450 reductase and NADPH as redox partners. Both HsCYP51 enzymes bound clotrimazole, itraconazole, and ketoconazole tightly (dissociation constants [K(d)s], 42 to 131 nM) but bound fluconazole (K(d), ~30,500 nM) and voriconazole (K(d), ~2,300 nM) weakly, whereas CaCYP51 bound all five medical azole drugs tightly (K(d)s, 10 to 56 nM). Selectivity for CaCYP51 over HsCYP51 ranged from 2-fold (clotrimazole) to 540-fold (fluconazole) among the medical azoles. In contrast, selectivity for CaCYP51 over Δ60HsCYP51 with agricultural azoles ranged from 3-fold (tebuconazole) to 9-fold (propiconazole). Prothioconazole bound extremely weakly to CaCYP51 and Δ60HsCYP51, producing atypical type I UV-visible difference spectra (K(d)s, 6,100 and 910 nM, respectively), indicating that binding was not accomplished through direct coordination with the heme ferric ion. Prothioconazole-desthio (the intracellular derivative of prothioconazole) bound tightly to both CaCYP51 and Δ60HsCYP51 (K(d), ~40 nM). These differences in binding affinities were reflected in the observed 50% inhibitory concentration (IC(50)) values, which were 9- to 2,000-fold higher for Δ60HsCYP51 than for CaCYP51, with the exception of tebuconazole, which strongly inhibited both CYP51 enzymes. In contrast, prothioconazole weakly inhibited CaCYP51 (IC(50), ~150 μM) and did not significantly inhibit Δ60HsCYP51.

Fig 1

Chemical structures of medical azole antifungal agents. The chemical structures of clotrimazole (molecular weight [MW], 345), fluconazole (MW, 306), voriconazole (MW, 349), ketoconazole (MW, 531), and itraconazole (MW, 706), which were used in this study, are shown.

Fig 2

Chemical structures of agricultural azole antifungal agents. The chemical structures of epoxiconazole (molecular weight [MW], 330), prochloraz (MW, 377), propiconazole (MW, 342), prothioconazole (MW, 344), prothioconazole-desthio (MW, 312), tebuconazole (MW, 308), and triadimenol (MW, 296), which were used in this study, are shown.

Fig 3

Spectral properties of CaCYP51, Δ60HsCYP51, and HsCYP51. (A and B) Absolute oxidized absorption spectra (A) between 700 and 300 nm and reduced carbon monoxide difference spectra (B) between 500 and 400 nm were determined using 5 μM CaCYP51 (line 1), Δ60HsCYP51 (line 2), and HsCYP51 (line 3), with matched quartz semi-micro cuvettes with 10-mm light paths. (C) Type I difference spectra were obtained by progressive titration of lanosterol against 5 μM solutions of the three CYP51 proteins using quartz semi-micro cuvettes with 4.5-mm light paths. (D) Lanosterol saturation curves were constructed for the CaCYP51 (●), Δ60HsCYP51 (○), and HsCYP51 (×) proteins. All spectral determinations were performed in triplicate, although the results from only one replicate are shown.

Fig 4

Itraconazole and fluconazole binding with CaCYP51, Δ60HsCYP51, and HsCYP51. (A and B) Type II difference spectra were obtained for 5 μM solutions of the three CYP51 proteins by progressive titration with itraconazole (A) and fluconazole (B). (C and D) Azole saturation curves were constructed for itraconazole (C) and fluconazole (D) as the change in absorbance (ΔA_{peak − trough}) against azole concentration using a rearrangement of the Morrison equation (40) for the tight ligand binding observed with CaCYP51 (●), Δ60HsCYP51 (○), and HsCYP51 (×).

Fig 5

Epoxiconazole and propiconazole binding with CaCYP51 and Δ60HsCYP51. (A and B) Type II difference spectra were obtained for 5 μM CaCYP51 and Δ60HsCYP51 by progressive titration with epoxiconazole (A) and propiconazole (B). (C and D) Azole saturation curves were constructed for epoxiconazole (C) and propiconazole (D) as the change in absorbance (ΔA_{peak − trough}) against azole concentration using a rearrangement of the Morrison equation (40) for the tight ligand binding observed with CaCYP51 (●) and Δ60HsCYP51 (○).

Fig 6

Prothioconazole and prothioconazole-desthio binding with CaCYP51 and Δ60HsCYP51. (A and B) Type II difference spectra were obtained for 5 μM CaCYP51 and Δ60HsCYP51 by progressive titration with prothioconazole (A) and prothioconazole-desthio (B). (C and D) Azole saturation curves were constructed for prothioconazole (C) and prothioconazole-desthio (D) as the change in absorbance (ΔA_{peak − trough}) against azole concentration. Prothioconazole-desthio saturation curves were fitted using a rearrangement of the Morrison equation (40) for the tight ligand binding observed with CaCYP51 (●) and Δ60HsCYP51 (○). The Michaelis-Menten equation was used to fit the weak binding observed with prothioconazole.

Fig 7

Azole IC₅₀ determinations with CaCYP51 and Δ60HsCYP51. (A and B) IC₅₀s were determined with 1 μM CaCYP51 (A) and 0.4 μM Δ60HsCYP51 (B) for the medical azoles fluconazole (●), itraconazole (⊙), and ketoconazole (■). (C and D) IC₅₀s were also determined with 1 μM CaCYP51 (C) and 0.4 μM Δ60HsCYP51 (D) for the agricultural azoles epoxiconazole (●), prochloraz (■), propiconazole (▲), tebuconazole (⧫), triadimenol (○), prothioconazole (□), and prothioconazole-desthio (△) with itraconazole (⊙) as a control. Relative velocities of 1.00 correspond to actual velocities of 3.9 ± 0.3 min⁻¹ for CaCYP51 and 22.7 ± 4.8 min⁻¹ for Δ60HsCYP51.