Triazole resistance mediated by mutations of a conserved active site tyrosine in fungal lanosterol 14α-demethylase

Abstract

Emergence of fungal strains showing resistance to triazole drugs can make treatment of fungal disease problematic. Triazole resistance can arise due to single mutations in the drug target lanosterol 14α-demethylase (Erg11p/CYP51). We have determined how commonly occurring single site mutations in pathogenic fungi affect triazole binding using Saccharomyces cerevisiae Erg11p (ScErg11p) as a target surrogate. The mutations Y140F/H were introduced into full-length hexahistidine-tagged ScErg11p. Phenotypes and high-resolution X-ray crystal structures were determined for the mutant enzymes complexed with short-tailed (fluconazole and voriconazole) or long-tailed (itraconazole and posaconazole) triazoles and wild type enzyme complexed with voriconazole. The mutations disrupted a water-mediated hydrogen bond network involved in binding of short-tailed triazoles, which contain a tertiary hydroxyl not present in long-tailed triazoles. This appears to be the mechanism by which resistance to these short chain azoles occurs. Understanding how these mutations affect drug affinity will aid the design of azoles that overcome resistance.

Figure 1

(a) Sequence alignment of fungal CYP51s. Alignment of Histoplasma capsulatum CYP51 (HcCYP51), Zymoseptoria tritici CYP51 (ZtCYP51), Uncinula necator CYP51 (UnCYP51), Aspergillus fumigatus CYP51A (AfCYP51A), Cryptococcus neoformans CYP51 (CnCYP51), Candida albicans CYP51 (CaErg11p), and Saccharomyces cerevisiae CYP51 (ScCYP51). The frequently mutated tyrosine residue homologous to ScErg11p Y140 is highlighted in grey. (b) The chemical structures of triazole antifungals used in this study: fluconazole, voriconazole, itraconazole and posaconazole.

Figure 2. Spectral characterisation of ScErg11p6 × His Y140F/H mutants.

(a) The absolute (continuous line) and CO bound (dotted line) spectra of ScErg11p6 × His Y140F mutant. (b) Type II difference spectra for FLC binding to ScErg11p6 × His wild type and mutant enzymes. (c) Fluconazole binding to wild type ScErg11p6 × His (●) ScErg11p6 × His Y140F (■) and ScErg11p6 × His Y140H (▲). All curves best fit the Hill equation.

Figure 3. Binding of long-tailed azoles ITC and PCZ to ScErg11p6 × His Y140F/H.

(a) ScErg11p6 × His Y140F in complex with ITC (PDB ID: 4ZDY) and (b) PCZ (PDB ID: 4ZE1) and (c) ScErg11p6 × His Y140H in complex with ITC (PDB ID: 4ZE2). Panel (d) depicts the hydrophilic pocket (surface representation) in the substrate channel of ScErg11p6 × His Y140F. The hydrogen bonding interactions are shown between the water molecule (930), the N1 of the piperazine ring and the main chain amide groups of H381 and S382. The ScErg11p6 × His Y140F is shown in lilac cartoon and ScErg11p6 × His Y140H is depicted in yellow cartoon. Hydrogen bonds are shown as yellow dashed lines and water molecules as red spheres. ITC (green carbons), PCZ (yellow carbons), the side chains of residues 140 and 126, the backbone of F384, residues H381 and S382 and the heme (magenta) are shown as sticks.

Figure 4. Binding of short-tailed azoles FLC and VCZ to ScErg11p6 × His.

Water-mediated interactions are shown for binding of (a) FLC and (d) VCZ to wild type ScErg11p6 × His (PDB IDs: 4WMZ and 5HS1, respectively). The modified interactions due to the Y140F/H mutations are depicted in panels (b) and (c) for FLC binding (PDB IDs: 4ZDZ and 4ZE3) and (e) for VCZ binding (PDB ID: 4ZE0). The wild type protein is depicted in grey cartoon (PDB IDs: 4WMZ and 5HS1), the Y140F mutant in lilac and the Y140H in yellow. The hydrogen bonds are shown with dashed lines and water molecules as red spheres. FLC (cyan carbons), VCZ (green carbons), the side chains of residue 140, 126, the backbone of S382 and F384 and the heme (magenta) are shown as sticks.