Contributions of both ATP-Binding Cassette Transporter and Cyp51A Proteins Are Essential for Azole Resistance in Aspergillus fumigatus

Abstract

While azole drugs targeting the biosynthesis of ergosterol are effective antifungal agents, their extensive use has led to the development of resistant organisms. Infections involving azole-resistant forms of the filamentous fungus Aspergillus fumigatus are often associated with genetic changes in the cyp51A gene encoding the lanosterol α14 demethylase target enzyme. Both a sequence duplication in the cyp51A promoter (TR34) and a substitution mutation in the coding sequence (L98H) are required for the full expression of azole resistance. A mechanism commonly observed in pathogenic yeast such as Candida albicans involves gain-of-function mutations in transcriptional regulatory proteins that induce expression of genes encoding ATP-binding cassette (ABC) transporters. We and others have found that an ABC transporter protein called Cdr1B (here referred to as AbcG1) is required for wild-type azole resistance in A. fumigatus Here, we test the genetic relationship between the TR34 L98H allele of cyp51A and an abcG1 null mutation. Loss of AbcG1 from a TR34 L98H cyp51A-containing strain caused a large decrease in the azole resistance of the resulting double-mutant strain. We also generated antibodies that enabled the detection of both the wild-type and L98H forms of the Cyp51A protein. The introduction of the L98H lesion into the cyp51A gene led to a decreased production of immunoreactive enzyme, suggesting that this mutant protein is unstable. Our data confirm the importance of AbcG1 function during azole resistance even in a strongly drug-resistant background.

Keywords: ATP-binding cassette transporter; Aspergillus fumigatus; Western blotting; azole resistance; cyp51A; genetic analysis.

FIG 1

AbcG1 is required for wild-type azole resistance. Cells containing or lacking (abcG1Δ) the abcG1 ABC transporter-encoding genes and the indicated cyp51A alleles were tested for their MIC for voriconazole as previously described (33). Strains with MIC values different than those on the ordinate are indicated on the plot above the bars of interest.

FIG 2

Validation of anti-Cyp51A antibodies. Whole-cell protein extracts were prepared from untreated (−) or voriconazole-challenged (+) cells. The two strains used were wild-type AfS35 cells or an isogenic derivative lacking the cyp51A gene. The molecular weight standards are indicated on the right-hand side of the antipeptide 3 Cyp51A Western blot. The arrows denote the location of the specific Cyp51A polypeptide with both antibodies. The top panels represent loading controls in which the blot was stained for total protein (Ponceau S) or subjected to Western blotting using an anti-α tubulin antibody.

FIG 3

Western blot analysis of Cyp51A protein levels. (A) Whole-cell protein extracts were prepared by grinding hyphae in liquid nitrogen. Equal amounts of protein extracts were resolved by SDS-PAGE and were transferred to nitrocellulose membranes. Cultures were grown in the absence (−) or presence (+) of 0.06 μg/ml voriconazole in minimal medium. Equal loading and transfer were ensured via staining the membranes with Ponceau S dye. Membranes were blocked and then probed with rabbit polyclonal antipeptide 3 antibodies prepared against a peptide of Cyp51A corresponding to residues 95 to 108 of the enzyme. Extracts were prepared from strains corresponding to an isogenic wild-type (wt, Afs35) or the same strain containing a TR34 promoter mutation, an L98H substitution form of cyp51A, or the corresponding double mutant strain (TR34 L98H) (left). Extracts were prepared from wild-type cells or an isogenic strain carrying multiple copies of the TR34 version of cyp51A (mcTR34) (right). Estimates by qPCR of the copy number of the cyp51A gene in this multiple copy strain suggest that at least 12 copies of the gene were introduced. These strains were grown as above, either in the absence or presence of voriconazole. (B) The same strains as in panel A were grown in the absence or presence of azole drug and analyzed by Western blotting but with two different antibodies. The top shows blotting with anti-tubulin antibody (α-tubulin Ab) as a loading control. The bottom denotes utilization of the rabbit polyclonal antipeptide 2 antibodies prepared against a peptide of Cyp51A corresponding to residues 478 to 491 of the enzyme. Use of this antibody avoids reactivity issues that might be caused by the L98H form of Cyp51A that is contained within the antigenic region of the antipeptide 3 antibody (see the text for details).

FIG 4

Levels of cyp51A mRNA. Steady-state levels of cyp51A mRNA were analyzed using reverse transcription-qPCR from total RNA recovered from strains containing the indicated forms of the cyp51A gene. (A) Strains were grown to mid-log phase and analyzed for cyp51A mRNA levels. Data are presented after normalization to cyp51A expression in wild-type cells. The bottom shows cyp51A mRNA produced in the strain containing multiple copies of cyp51A containing the TR34 promoter mutation (mcTR34). (B) Voriconazole induction of cyp51A. The strains used above were grown in parallel cultures with voriconazole added to one flask for the final 8 h of growth. Total RNA was prepared from cultures grown in the absence and presence of voriconazole. Data are presented as the ratio of mRNA expression in the presence of voriconazole normalized to expression in the absence of drug for each cyp51A gene.