Erg11 mutations associated with azole resistance in clinical isolates of Candida albicans

Authors

  • Ming-Jie Xiang,

    1. Department of Immunology, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    2. Department of Laboratory Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    3. Department of Laboratory Medicine, Ruijin Hospital Luwan Branch, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Jin-Yan Liu,

    1. Department of Laboratory Medicine, Ruijin Hospital Luwan Branch, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Pei-Hua Ni,

    1. Faculty of Clinical Laboratory, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Shengzheng Wang,

    1. School of Pharmacy, Second Military Medical University, Shanghai, China
    Search for more papers by this author
  • Ce Shi,

    1. Department of Laboratory Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Bing Wei,

    1. Department of Laboratory Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Yu-Xing Ni,

    1. Department of Clinical Microbiology Laboratory, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • Hai-Liang Ge

    Corresponding author
    • Department of Immunology, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author

Correspondence: Hai-Liang Ge, Department of Immunology, Shanghai Jiao Tong University School of Medicine, No. 280 South Chongqing Road, Shanghai 200025, China. Tel.: +86 021 6445 2974; fax: +86 021 6384 6383; e-mail: gehl@shsmu.edu.cn

Abstract

The widespread use of azoles has led to increasing azole resistance among Candida albicans strains. One mechanism of azole resistance involves point mutations in the ERG11 gene, which encodes the target enzyme (cytochrome P450 lanosterol 14α-demethylase). In the present study, we amplified and sequenced the ERG11 gene of 23 C. albicans clinical isolates. Seventeen mutations encoding distinct amino acid substitutions were found, of which seven (K143Q, Y205E, A255V, E260V, N435V, G472R, and D502E) were novel. We further verified the contribution of the amino acid substitutions to azole resistance using site-directed mutagenesis of the ERG11 gene to recreate these mutations for heterologous expression in Saccharomyces cerevisiae. We observed that substitutions A114S, Y132H, Y132F, K143R, Y257H, and a new K143Q substitution contributed to significant increases (≧ fourfold) in fluconazole and voriconazole resistance; changes in itraconazole resistance were not significant (≦ twofold).

Introduction

The incidence of systemic fungal infections has increased steadily over the past several years. This worldwide health threat is due in part to the growing number of patients who are immunodeficient as a result of organ and bone marrow transplants, cancer treatments, and AIDS (Vandeputte et al., 2012). These opportunistic infections are caused primarily by Candida and Aspergillus species, especially Candida albicans and Aspergillus fumigatus (Tan et al., 2008; Kontoyiannis et al., 2010; Pappas et al., 2010). Invasive fungal infections are now the chief infectious challenge in the practices of hematology, oncology, and intensive care (Richardson, 2005). The availability of effective antifungal agents is thus critical to the treatment of these potentially life-threatening infections.

Triazole antifungals are used as front-line drugs for the treatment and prophylaxis of many Candida infections. However, with long-term treatment, azole-resistant phenotypes of C. albicans appear. The widespread use of azole drugs has led to a rise in fungal strains resistant to one or more of the drugs in this class. The emergence of resistant strains poses a major challenge to treatment and is a matter of great concern.

Azoles inhibit fungal growth by interfering with the synthesis of ergosterol, a necessary component of fungal cell membranes. The ergosterol biosynthetic pathway is interrupted by azoles through inhibition of the enzymatic activity of 14-α-sterol demethylase (also known as CYP51A1), the product of the ERG11 gene. Azoles have a basic nitrogen that coordinates to the iron atom of the heme group located in the active site of 14-α-sterol demethylase. The active site is thus occupied by the azole, which acts as a non-competitive inhibitor (Sanglard et al., 1998; Ji et al., 2000; Podust et al., 2001).

Studies in yeast have revealed several molecular mechanisms of azole resistance (Maebashi et al., 2002; Sanglard, 2002; White et al., 2002; Goldman et al., 2004; Rogers, 2006; Cowen & Steinbach, 2008): (1) decreased affinity of azoles for the target enzyme CYP51A1 caused by point mutations in the ERG11 gene, (2) increases in CYP51A1 copy number through upregulation of ERG11 expression, (3) metabolic modifications, and (4) decreased intracellular azole accumulation by upregulation of multidrug transporters or drug sequestration.

Many studies have identified point mutations in the ERG11 gene in azole-resistant C. albicans isolates. Such mutations can alter the affinity of CYP51A1 for an azole if the resultant amino acid substitutions lead to changes in the tertiary structure of the enzyme. Mutations in the ERG11 gene encoding more than 160 distinct amino acid substitutions have been reported (Feng et al., 2010; Morio et al., 2010; Manastır et al., 2011). However, only 10 of these amino acid substitutions have been confirmed to cause fluconazole (FLZ) resistance. Of these, four substitutions (T315A, Y118A, Y18F, and Y118T) were created in the laboratory and have not yet been detected in clinical isolates (Lamb et al., 1997; Chen et al., 2007). The knowledge of further specific point mutations associated with azole resistance could be used to identify resistant strains, to adjust treatments accordingly, and for the rational design of new drugs less prone to resistance.

The aim of the present study was to identify mutations in the C. albicans ERG11 gene associated with clinical cases of azole-resistant candidiasis. To this end, we amplified and sequenced the ERG11 genes from clinical isolates of C. albicans and identified mutations that might be related to FLZ resistance. To determine the contribution of each individual mutation to the resistant phenotype, we then used site-directed mutagenesis of the ERG11 gene to recreate these mutations for heterologous expression in Saccharomyces cerevisiae.

Materials and methods

Strains and plasmids

Between 1 November 2006 and 31 December 2008, 23 clinical isolates of C. albicans were collected from non-AIDS patients with vulvovaginal candidiasis, mucocutaneous candidiasis, and candidal infection of the respiratory, skin, and digestive tracts, at four hospitals: Shanghai First Maternity and Infant Hospital at Tongji University School of Medicine, Renji Hospital at Shanghai Jiao Tong University School of Medicine, Huashan Hospital at Fudan University, and Ruijin Hospital at Shanghai Jiaotong University School of Medicine.

Candida albicans strain ATCC 90028 and Candida parapsilosis strain ATCC 22019 were used as controls for antifungal susceptibility tests. Identification procedures, including germ tube formation in serum-containing medium, morphology analysis on CHROMagar Candida (CHROMagar, Paris, France) and carbohydrate assimilation tests (API 20C AUX BioMérieux, Marcy l'Etoile, France), were performed on each strain.

Saccharomyces cerevisiae strain INVsc-1 (MATα his3△1/his3△1 leu2/leu2trp1-289/trp-289ura3-52/ura3-52, His-, Leu-, Trp-, Ura-) and plasmid pYES2 were purchased from Invitrogen (Carlsbad, CA). The plasmid pMD18-T (Takara, Shiga, Japan) was used for point mutation constructs. Escherichia coli DH5a was used for propagation of the plasmids constructed in this study.

Drug susceptibility testing

Testing to determine the susceptibility of isolates to FLZ, itraconazole (ITZ), and voriconazole (VOR) was performed using the broth microdilution method according to the standards of the Clinical and Laboratory Standards Institute (CLSI) M27-A2 (CLSI, 2002). The minimum inhibitory concentrations (MIC) of these azole antifungal drugs were determined for C. albicans clinical isolates in RPMI 1640 medium and S. cerevisiae transformants in selective Yeast Nitrogen Base (YNB) medium containing galactose and raffinose (Genmed, Minneapolis, MN). Candida albicans ATCC 90028 and C. parapsilosis ATCC 22019 were used as controls. Changes in susceptibility to FLZ, ITZ, or VOR ≧ fourfold were considered significant.

PCR amplification and sequencing of the ERG11 gene

The ERG11 genes from isolates were amplified by PCR using the following two pairs of primers: ERG1A (5′-ATGGCTATTGTTGAAACTGTCATT-3′) with ERG1B (5′-GGATCAATATCACCACGTTCTC-3′), and ERG2A (5′-ATTGGAGACGTGATGCTGCTCAA-3′) with ERG2B (5′-CCAAATGATTTCTGCTGGTTCAGT-3′). Two fragments were produced, one 815 bp in length extending from 1 to 815 bp and the other 833 bp in length, extending from 728 to 1560 bp. The 25-μL PCR reaction mixture contained 2.5 μL of 10× PCR Buffer (with Mg2+), 4 μL of 200 ng genomic DNA, 0.2 μL of 25 mmol L−1 dNTP, 0.2 μL of 50 μmol L−1 each primer and 1 μL of 1 U μL−1 Taq DNA platinum polymerase high fidelity (Invitrogen). PCR conditions were as follows: 95 °C for 5 min; 95 °C for 40 s, 55 °C for 40 s, and 72 °C for 90 s, 35 times; followed by an extension step at 72 °C for 5 min. The PCR products were purified and sequenced using an ABI 3730 sequencer (ABI, Foster City, CA). The sequencing results were analyzed by blast and compared with the published GenBank sequence for AY856352 available at http://www.ncbi.nlm.nih.gov.

Mutagenesis of the C. albicans ERG11 gene

Site-directed mutagenesis of ERG11 to introduce the point mutations observed in the clinical FLZ-resistant isolates was performed using the Takara MutanBest Kit (Takara) using a PCR-based approach. Although the full length ERG11 open reading frame (ORF) in C. albicans is 1587 bp, we used only the first 1560 bp in our constructs because previous studies indicate that several mutations between 1560 and 1587 bp are not associated with FLZ resistance.

The first 1560 bp of the ERG11 ORF of C. albicans ATCC 90028 were amplified using the primers ERG11-F (5′-ATGGCTATTGTTGAAACTGTCATT-3′) and ERG11-R (5′-AAACATACAAGTTTCTCTTTTTTC-3′). This ampl-ified ORF was then ligated into pMD18-T, creating the plasmid pMD18-T-ERG11.

The pMD18-T-ERG11 mutants were created using the mutagenic primers shown in Table 1. The ERG11 genes were then inserted into the pYES2 plasmid to create wild-type pYES2-ERG11 and pYES2-ERG11-mutant constructs as follows. The wild-type and various pMD18-T-ERG11-mutant constructs were used as templates for amplification of the ERG11 gene. PCR was carried out with high fidelity pyrobest DNA polymerase (Takara) using the primers ERG11-A (5′-ATATGGTACCATGGCTATTGTTGAAACTGTC-3′) and ERG11-B (5′-CGCCTCGAGGAAACATACAAGTTTCTCTTTT-3′), flanked with KpnI and XhoI restriction sites to allow subcloning of the amplified ERG11 fragments into pYES2 precut with the same enzymes.

Table 1. Sequences of primer used in this study
NameSequences (5′-3′)
ERG1AATGGCTATTGTTGAAACTGTCATT
ERG1BGGATCAATATCACCACGTTCTC
ERG2AATTGGAGACGTGATGCTGCTCAA
ERG2BCCAAATGATTTCTGCTGGTTCAGT
ERG11-A5′-ATATGGTACCATGGCTATTGTTGAAACTGTC-3′
ERG11-B5′-CGCCTCGAGGAAACATACAAGTTTCTCTTTT-3′
ERG11-FATGGCTATTGTTGAAACTGTCATT
ERG11-RAAACATACAAGTTTCTCTTTTTTC
A114STTATCTGATGTTTCTTCTGAA
TTTAGCATTGAAAACAAATTC
Y132FGGGGTTATTTTTGATTGTCCA
TGTACCGAAAACTGGAGTAGT
Y132HGGTACAGGGGTTATTCATGAT
GAAAACTGGAGTAGTTAAATG
K143QTTAATGGAACAACAAAAATTTGCT
TCTAGAATTTGGACAATCATAAAT
Y143RATGGAACAAAGAAAATTTGCT
TAATCTAGAATTTGGACAATC
Y257HTCTGCTACTCATATGAAAGAA
GATTTTCTTTTGAGCAGCATC
E260VGCTACTTATATGAAAGTAATT
AGAGATTTTCTTTTGAGCAGC
N435VGCCAAAGCTGTTTCTGTTTCA
AGCAGCAGTATCCCATCTAGT
G448EGTTGATTATGAGTTTGGGAAA
TTCATCAGAAGAGTTAAATGA
G472RAGATGTATTAGGGAACAATTT
ATGTCTACCACCACCAAATGG
D502EAAAGTGCCTGAACCTGATTAT
ATAACCATCAATAGTCCATCT

Recombinant plasmids extracted from E. coli transformants were sequenced from both directions using an ABI 3730 sequencer (ABI) to confirm the presence of each mutation.

Heterologous expression of C. albicans ERG11 in S. cerevisiae

The constructed wild-type expression plasmid pYES-ERG11 and mutant expression plasmids (pYES-ERG11M) were transformed into S. cerevisiae INVsc-1 using a Frozen EZ Yeast Transformation Kit (ZYMO Research, Orange, CA) in selective YNB medium with glucose as a carbon source (Genmed). The S. cerevisiae transformants were then suspended in YNB-inducing medium (with 2% galactose and 1% raffinose as a carbon source) for inducing the expression of various ERG11 alleles. Recombinant plasmids were extracted from S. cerevisiae transformants using a Zymoprep Yeast Plasmid Miniprep Kit (ZYMO Research) and confirmed using restriction analysis with the endonucleases KpnI and XhoI.

Results

Antifungal susceptibility testing of clinical isolates

The MIC values obtained for the 23 C. albicans isolates with the three different azoles (FLZ, ITZ, and VOR) are summarized in Table 2. Of the 23 C. albicans strains isolated, two were resistant to FLZ (MIC ≥ 64 μg mL−1), nine were susceptible to FLZ in a dose-dependent manner (S-DD for MICs of 16 and 32 μg mL−1), and 12 were susceptible to FLZ (MICs ≤ 8 μg mL−1). The data in Table 2 indicate that all 11 isolates that were S-DD or resistant to FLZ were also S-DD (n = 5) or resistant (n = 6) to ITZ. Only five of the 11 isolates that were S-DD or resistant to FLZ exhibited susceptibility to VOR.

Table 2. Results of in vitro azole susceptibility testing and ERG11 sequence analysis for 23 clinical isolates of Candida albicans
No. of strainSite of isolationMIC (μg mL−1)Amino acid change(s) in Erg11Missense mutation/Hot spot
FLZITZVOR
  1. The newly observed substitutions are shown in bold.

  2. The sequence of C. albicans isolates was compared with GenBank AY856352.

  3. Amino acid abbreviations: A, (Alanine, Ala); D, (aspartic acid, Asp); E, (glutamic acid, Glu); F, (phenylalanine, Phe); G, (glycine, Gly); H, (histidine, His); I, (isoleucine, Ile); K, (lysine, Lys); N, (asparagine, Asn); Q, (glutamine); R, (arginine, Arg); S, (serine, Ser); T, (threonine, Thr); Q, (glutamine, Gln); V, (valine, Val); Y, (tyrosine, Tyr).

ATCC 90028 (Erg11 wild type)0.250.03130.0313NoneNone
141Vagina16> 160.5A114S, Y205E, Y257H, V437IG340T/I, T613G or C 615A/–, T769C/–, G1309A/III
201Vagina16161A114S, Y205E, Y257H, V437IG340T/I, T613G or C 615A/–, T769C/–, G1309A/III
205Vagina160.52Y132H, Y205E, V437I, G448ET394C/I, T613G or C 615A/– G1309A/III, G1343A/III
206Vagina16162Y132H, Y205E, V437I, G472RT394C/I, T613G or C 615A/–, G1309A/III, G1414A/III
208Vagina32161Y132H, Y205E, N435V, G448E、D502ET394C/I, T613G or C 615A/–, A1303G/III, G1343A/III, C1506A/–
509Vagina320.250.25D116E, K128T, Y205E, V437IT348A/I, A383C/I, T613G or C 615A/–, G1309A/III
210Vagina160.254Y132H, Y205E, Y257H, E260V, V437I, G448ET394C/I, T613G or C 615A/–,T769C/–, A779T/–, G1309A/III, G1343A/III
24308Skin320.50.25D116E, K128T, K143R, Y205E, V437IT348A/I, A383C/I, A428G/I, T613G or C 615A/–, G1309A/III
75045Vagina160.252Y132H, Y205E, V437I, G448ET394C/I, T613G or C 615A/–, G1309A/III, G1343A/III
13139Skin> 64> 168D116E, Y132F, K143Q, Y205E, Y257HT348A/I, A395T/I, A427C/I, T613G or C 615A/–, T769C/–
21897Skin> 64> 1616D116E, Y132F, K143Q, Y205E, V437IA395T/I, A427C/I, T613G or C 615A/–, G1309A/III
c701Sputum0.250.0313< 0.0313Y205E, V437IT613G or C 615A/–, G1309A/III
c837Sputum0.250.0313< 0.0313D116ET348A/I
c924Sputum0.125< 0.0313< 0.0313Y205E, E266D, V437IT613G or C 615A/–, A789C/II, G1309A/III
c289Throat0.250.03130.0313Y205E, V437IT613G or C 615A/–, G1309A/III
c273Sputum0.25< 0.03130.0313D116E, V437IT348A/I, G1309A/III
c923Sputum0.250.03130.0313Y205E, A255V, V437IT613G or C 615A/–, C764T/–, G1309A/III
c592Sputum0.250.0313< 0.0313D116E, Y205ET348A/I, T613G or C 615A/–
c271Stool0.25< 0.0313< 0.0313D116E, V437IT348A/I, G1309A/III
c286Sputum0.250.03130.0313Y205E, V437IT613G or C 615A/–, G1309A/III
c944Sputum10.03130.0313D116ET348A/I
c963Throat0.250.03130.0313Y205E, V437IT613G or C 615A/–,G1309A/III
c827Sputum0.125< 0.0313< 0.0313Y205E, V437IT613G or C 615A/–, G1309A/III

Mutations in the ERG11 gene of clinical isolates

By comparing the 1560 bp of the ERG11 coding re-gion of C. albicans clinical isolates with that of the published wild-type sequence, we identified 33 mutations. As expected from previous investigations of clinical isolates, frequent silent mutations that do not change the protein sequence were identified (data not shown). The remaining ERG11 mutations and their resultant amino acid changes are listed in Table 2. Among 17 missense mutations, 10 had been reported previously and seven were new (K143Q, Y205E, A255V, E260V, N435V, G472R, and D502E). All of these mutations were homozygous, even in isolates with multiple mutations. Interestingly, more point mutations were identified in S-DD and resistant isolates than in susceptible isolates. The data in Table 2 show that S-DD and resistant isolates contained an average four to five missense mutations; two missense mutations were detected among the susceptible isolates. As observed in previous studies, the majority of missense mutations were located in three diffuse hot-spot regions: amino acids 105–165, 266–287, and 405–488 (Marichal et al., 1999; Table 2).

The corresponding amino acid substitution in the ERG11 protein alter azole susceptibility

The mutations detected in the ERG11 genes from each of the C. albicans isolates investigated here contained multiple mutations. To reveal the effect of each individual amino acid substitution on changes in azole susceptibility, each mutation was introduced into the wild-type ERG11 gene (susceptible strain ATCC 90028) by site-directed mutagenesis and expressed in S. cerevisiae. All clones were homozygous for each mutation introduced. The susceptibility of the transformants to FLZ, ITZ, and viroconazole was tested by dilution MIC assay (Table 3). All single mutations affected the susceptibility to azole derivatives, with the exception of the E260V and G472R mutations, which had the same azole MIC value as the wild-type pYES2-ERG11 transformant.

Table 3. Changes in azole susceptibility of Saccharomyces cerevisiae strain INVSc1 expressing mutated ERG11 genes
ConstructAmino acid substitutionMIC (μg mL−1)
FluconazoleVoriconazoleItraconazole
  1. W, wild-type; M, mutant.

pYES-W 160.250.5
pYES-MA114S6411
pYES2-MY132F25641
pYES2-MY132H6420.5
pYES2-MK143Q> 25641
pYES2-MK143R> 25621
pYES2-MY257H12821
pYES2-ME260V160.250.5
pYES2-MN435V320.50.5
pYES2-MG448E1280.250.5
pYES2-MG472R160.250.5
pYES2-MD502E320.50.5

To investigate the relationship between azole resistance and mutations in the ERG11 gene of C. albicans, we mapped these mutations to the 3D model of the target enzyme interacting with an azole (Fig. 1). The A114S, Y132F, Y132H, K143Q, and K143R substitutions are near the substrate channel. The G448E substitution is located just before the heme-binding site and near the end of helix I of the enzyme. Although the G472R substitution is under the porphyrin ring of the heme group, it is located further from the ring than is Cys470, which is the fifth ligand of the heme group. The substitutions located in the G helix (Y257H and E260V) and near the L helix (N435V and D502E) were far from the active center and substrate access channel of the protein.

Figure 1.

Homology model of CA-CYP51A structure. α-Helices and β-pleated sheets are shown in blue and pink, respectively. The heme cofactor is shown in yellow. Amino acid substitutions are shown in red.

Discussion

In the present study, we searched for mutations in the ERG11 gene of C. albicans in 11 clinical isolates with reduced susceptibility to azoles and 12 azole-susceptible isolates using PCR amplification and gene sequencing. We found 17 amino acid substitutions, of which 10 (A114S, D116E, K128T, Y132F, Y132H, K143R, Y257H, E266D, V437I, and G448E) had been reported previously and seven were novel (K143Q, Y205E, A255V, E260V, N435V, G472R and D502E).

The amino acid substitutions D116E, V437I, and the new Y205E observed here were found in both azole-susceptible and azole-resistant strains, strongly suggesting that these substitutions were not associated with the azole-resistant phenotype. In addition, the amino acid substitutions K128T and E266D reported previously and the new substitution A255V discovered here in one azole-susceptible strain were also determined not to confer resistance.

We then focused on the relationship between the remaining 11 missense mutations and azole resistance. The A114S and Y257H substitutions previously were reported to occur simultaneously in 14 isolates resistant to FLZ (Xu et al., 2008); here, we found the same phenomenon in isolates 141 and 201. Chau et al. (2004) explored the changes in azole susceptibility resulting from expressing C. albicans PCR-amplified ERG11 alleles in S. cerevisiae and demonstrated that Y257H combined with G464S could significantly increase the FLZ and VOR MIC. In this study, we observed that A114S and Y257H alone did confer some level of resistance to azoles; relative increases in the MICs of FLZ, VOR, and ITZ were four-, four-, and twofold, respectively, for A114S and eight-, eight-, and twofold, respectively, for Y257H. As a fourfold increase in MICs was considered significant, mutations A114S and Y257H significantly increased the resistance to FLZ and VOR but not ITZ.

The Y132H substitution was found in five isolates (205, 206, 208, 210, and 75045), four of which also had the G448E substitution (205, 208, 210, and 75045). In several reports, the Y132H substitution was determined to be sufficient to confer FLZ resistance based on heterologous expression of S. cerevisiae with site-direct mutagenesis of the wild-type ERG11 gene or affinity of the CYP51A1 CO for azole (Sanglard et al., 1998; Marichal et al., 1999; Kakeya et al., 2000; Chau et al., 2004). Sanglard et al. (1998) further observed that the combination of Y132H with S405F or R467K led to measureable increases in the MICs for FLZ, ketoconazole, and ITZ. However, Bellamine et al. (2004) suggested that F145L substitution rather than Y132H substitution was associated with FLZ resistance. In our study, it was noted that one single Y132H substitution did result in reduced susceptibility to FLZ and VOR, with MIC increases of four- and eightfold, respectively, but had no effect on ITZ susceptibility.

G448E substitution was first reported by Löffler et al. (1997), but its correlation with resistance to azole antifungals is still unknown. Here we found that G448E substitutions were associated with an eightfold increase in resistance to FLZ, with less of an effect on VOR and ITZ resistance.

Two isolates with high resistance to FLZ, VOR, and ITZ had Y132F and K143Q substitutions. The relationship of the Y132F substitution to FLZ resistance was observed by Perea et al. (2001), who studied the functional expression of C. albicans PCR-amplified ERG11. In the present study, we found that Y132F substitution significantly increased FLZ and VOR MICs 16-fold. The newly identified K143Q substitution displayed similar resistance.

The K143R, E260V, N435V, G472R, and D502E substitutions were only recovered in one single isolate, which had an S-DD to FLZ. In agreement with a previous report (Chau et al., 2004), expression of the K143R mutant protein in S. cerevisiae led to an increase in the MICs of FLZ, VOR, ITZ by factors of 16-, 8-, and 2-fold, respectively. The newly observed N435V and D502E substitutions affected azole resistance to a lesser extent. Transformants demonstrated twofold increases in the MICs of FLZ and VOR. No contribution to azole resistance was seen in the newly observed E260V and G472R substitutions.

Overall, we observed that all but one of the mutations tested conferred changes to FLZ and VOR susceptibility at the same time and had less effect on ITZ resistance. This observation is consistent with previous reports (Chau et al., 2004; Li et al., 2004). The high incidence of cross-resistance between FLZ and VOR presumably results from their structural similarity and the absence of the long side chain present in ITZ, which is predicted to make extensive hydrophobic contacts with CYP51A1. Only the A61V and P230L substitutions are predicted to allow interaction with posaconazole and ITZ side chains; these substitutions were observed to confer resistance to these antifungal drugs when combined with other amino acid changes (Chau et al., 2004; Li et al., 2004).

To investigate the relationship between azole resistance and mutations in the ERG11 gene of C. albicans, we mapped these mutations to the 3D model of the target enzyme interacting with an azole (Sheng et al., 2009ab, 2010). The A114S, Y132F, Y132H, K143Q, and K143R substitutions are near the substrate channel in CYP51A1 and thus may interfere with entry of the inhibitor/substrate or its binding to the active site. The G448E substitution is located just before the heme-binding site and near the end of helix I of the enzyme. This mutation might affect the function of the active center. Although the G472R substitution is under the porphyrin ring of the heme group, it is located further away from the ring than is Cys470, which is the fifth ligand of the heme group. The G472 residue is small and cannot interact with the heme group by hydrogen bonding or coordination and thus might not be involved in azole resistance. Surprisingly, the substitutions located in the G helix (Y257H and E260V) and near the L helix (N435V and D502E) were far from the active center and the substrate access channel of the protein and thus might not directly change the affinity of the Erg11 protein to an azole. However, according to our results, the Y257H mutation caused some changes in the susceptibility of transformants to FLZ and VOR. The contribution of these mutations to azole resistance must therefore be further verified using site-directed mutagenesis.

In summary, all Erg11 point mutations that resulted in amino acid substitutions conferred resistance to azoles, particularly FLZ and VOR. Our observations are clinically relevant, as these point mutations were originally derived from clinical isolates. These mutations could potentially be used to identify resistant strains, to adjust treatments accordingly, and for the rational design of new drugs less prone to resistance.

Based on the results of previous studies and the data reported here, we suggest using the following amino acid substitutions as predictive markers of azole resistance: A61V, A114S, Y132F, Y132H, K143Q, K143R, Y257H, S405F, G448E, F449S, G464S, R467K, and I471T.

Acknowledgements

This work was supported by grants from the Program of Shanghai Municipal Health Bureau of China (#2009239) and the Program of Shanghai Jiao Tong University School of Medicine of China (#09XJ21036).

Ancillary