ABSTRACT
The fungal Cyp51-specific inhibitors VT-1161 and VT-1598 have emerged as promising new therapies to combat fungal infections, including Candida spp. To evaluate their in vitro activities compared to other azoles, MICs were determined by Clinical and Laboratory Standards Institute (CLSI) method for VT-1161, VT-1598, fluconazole, voriconazole, itraconazole, and posaconazole against 68 C. albicans clinical isolates well characterized for azole resistance mechanisms and mutant strains representing individual azole resistance mechanisms. VT-1161 and VT-1598 demonstrated potent activity (geometric mean MICs ≤0.15 μg/ml) against predominantly fluconazole-resistant (≥8 μg/ml) isolates. However, five of 68 isolates exhibited MICs greater than six dilutions (>2 μg/ml) to both tetrazoles compared to fluconazole-susceptible isolates. Four of these isolates likewise exhibited high MICs beyond the upper limit of the assay for all triazoles tested. A premature stop codon in ERG3 likely explained the high-level resistance in one isolate. VT-1598 was effective against strains with hyperactive Tac1, Mrr1, and Upc2 transcription factors and against most ERG11 mutant strains. VT-1161 MICs were elevated compared to the control strain SC5314 for hyperactive Tac1 strains and two strains with Erg11 substitutions (Y132F and Y132F&K143R) but showed activity against hyperactive Mrr1 and Upc2 strains. While mutations affecting Erg3 activity appear to greatly reduce susceptibility to VT-1161 and VT-1598, the elevated MICs of both tetrazoles for four isolates could not be explained by known azole resistance mechanisms, suggesting the presence of undescribed resistance mechanisms to triazole- and tetrazole-based sterol demethylase inhibitors.
INTRODUCTION
Candida albicans is a dimorphic yeast and opportunistic pathogen that is known to cause a wide range of infections in healthy and immunocompromised patients. In the United States, C. albicans is the leading Candida species identified in oropharyngeal and vulvovaginal infections, where recurrent infections remain problematic (1–5). In more serious systemic disease such as bloodstream infections (BSI), Candida species collectively are the fourth-leading cause of nosocomial BSI in the United States (6). Moreover, resistance to currently available antifungal agents continues to be a problem, particularly given the relatively limited armamentarium against fungal infections (7–11). In particular, azole antifungal resistance in Candida spp. threatens to diminish the efficacy of arguably the most widely used antifungal drug class (12). Appropriate clinical use of available drugs on the market and eventual expansion of the antifungal arsenal is therefore paramount to safeguarding its effectiveness.
Azole antifungal resistance in C. albicans can be attributed to multiple mechanisms. First, efflux pump overexpression, such as the ATP-binding cassette (ABC) transporters Cdr1 and Cdr2, as well as the major facilitator transporter Mdr1, prevents drug accumulation within the yeast cell (13–16). Second, increased production of the azole target 14α-lanosterol demethylase (CYP51) can attenuate the inhibitory effects of the azoles drug class (17–19). Increases in efflux pump and drug target production is often the result of gain-of-function mutations in zinc cluster transcription factors (ZCFs) (Tac1 for CDR1 and CDR2, Mrr1 for MDR1, and Upc2 for ERG11) that regulate their gene expression, though polyploidy of chromosomes in the yeast genome can also result in increased expression of the genes encoding these azole resistance determinants. Third, mutations in ERG11 can confer azole resistance through alteration of the drug target (20–23). Lastly, alternative sterol biosynthesis as a result of changes within the ergosterol biosynthetic pathway allows some C. albicans isolates to circumvent the effects of azole inhibition altogether (24–27).
VT-1161 and VT-1598 are novel tetrazole antifungal agents with high specificity for fungal CYP51 compared to human CYP enzymes (28–30) and thus may have improved adverse effect and drug-drug interaction profiles due to lesser off-target inhibition. In this study, we compare the in vitro activity of the novel tetrazoles VT-1161 and VT-1598 to the current triazole antifungals fluconazole, voriconazole, itraconazole, and posaconazole against a collection of clinical isolates and laboratory strains with known resistance mechanisms.
RESULTS
In vitro activity of VT-1161 and VT-1598 against fluconazole-susceptible and fluconazole-resistant clinical isolates.VT-1161 and VT-1598 showed potent in vitro activity against 68 previously described clinical isolates of C. albicans, the majority (57 of 68) of which were fluconazole resistant (MIC ≥8 μg/ml) and possessed multiple known azole resistance mechanisms (see Table S1 in the supplemental material) (19). Both VT-1161 and VT-1598 had lower MIC50 values (0.06 and 0.125 μg/ml, respectively), and VT-1598 had a lower MIC90 value (0.25 μg/ml) compared to the other tested azole antifungals (Table 1). VT-1161 and VT-1598 MICs were ≤0.015 μg/ml against the 11 fluconazole-susceptible isolates within the collection, and the VT-1598 MICs were 0.03 μg/ml against 33% (19 of 57) of the fluconazole-resistant clinical isolates. This suggests that some fluconazole-resistance mechanisms do not affect the in vitro potency of VT-1598. Posaconazole also demonstrated activity against many, but not all, of the same fluconazole-resistant isolates, as posaconazole MICs were within a 2-fold increase (1-dilution difference) to those of the fluconazole-susceptible isolates for 15 of the fluconazole-resistant isolates. Using this same metric, VT-1161 maintained in vitro potency against 8 fluconazole-resistant clinical isolates, which was comparable to that of voriconazole (six isolates) and greater than that of itraconazole (two isolates). Overall, VT-1598 and VT-1161 thus appear to have additional activity against several fluconazole-resistant isolates, and in this respect are at least comparable to commercially available triazoles.
Geometric mean MIC, MIC50, MIC90, and range for each tested compound against 68 clinical isolates of C. albicans
VT-1598 MICs were elevated at least 4-fold (≥0.06 μg/ml; range, 0.06 to >8 μg/ml) against 38 fluconazole-resistant isolates compared to its activity against the fluconazole-susceptible isolates. VT-1161 MICs were elevated at least 4-fold (≥0.06 μg/ml; range, 0.06 to >8 μg/ml) against 49 fluconazole-resistant isolates. Five clinical isolates displayed highly elevated VT-1598 and VT-1161 MICs (range, 4 to >8 μg/ml) and also high fluconazole, voriconazole, itraconazole, and posaconazole MICs. Sequencing and/or relative quantitation of mRNA expression of known resistance genes revealed that four of these isolates overexpressed CDR1 relative to the CDR1 mRNA levels of fluconazole-susceptible clinical isolates (19). The fifth isolate contained a premature stop codon in ERG3, resulting in truncation of the protein after Gly130, which likely explains its significantly elevated resistance not only to VT-1161 and VT-1598 but also to all other tested azole antifungals.
To gain additional insight on the determinants that could confer decreased susceptibility to VT-1161 and VT-1598 in the clinical isolates, a point-biserial correlation between the log2-fold increase in VT-1598 and VT-1161 MICs and the mRNA expression levels of CDR1, MDR1, and ERG11 in the clinical isolates was performed. The log2-fold increase in MICs was compared to the baseline MIC measurement for VT-1598 and VT-1161 against fluconazole-susceptible isolates (MIC ≤ 0.015) and expression levels of either CDR1, MDR1, and ERG11 were measured via RT-qPCR in a previous study (19). The majority of fluconazole-resistant clinical isolates exhibited increased CDR1 expression; however, there was no significant correlation between CDR1 expression and VT-1598 resistance (P = 0.287). In contrast, higher levels of CDR1 expression did positively correlate with increasing VT-1161 MICs (P < 0.01). Similarly, while there was no relationship between MDR1 expression and the VT-1598 MIC (P = 0.105), there was a slight positive correlation between MDR1 expression and an increased VT-1161 MIC (P < 0.05). No significant correlation was established with either drug and ERG11 expression (P = 0.512 and P = 0.355 for VT-1598 and VT-1161, respectively).
VT-1598 and VT-1161 MICs against the clinical isolates were plotted directly against those of fluconazole, voriconazole, itraconazole, and posaconazole to visualize relative susceptibility differences (Fig. 1). As previously noted, both VT-1161 and VT-1598 retained potency against several fluconazole-resistant isolates, and all isolates with reduced VT-1598 or VT-1161 potency were resistant to fluconazole. By comparison, VT-1161 MICs were disproportionately higher against some clinical isolates compared to those of the other tested azoles. One voriconazole-susceptible isolate (0.125 μg/ml) that contained a K143R ERG11 mutation and exhibited increased CDR1 expression had an 8-fold increase in the MIC of VT-1161 compared to that observed against fluconazole-susceptible isolates (≤0.015 μg/ml). In addition, a single itraconazole-susceptible isolate (0.125 μg/ml) demonstrated a 32-fold increase in the VT-1161 MIC (0.5 μg/ml). This isolate contained three Erg11 amino acid substitutions (F126L, Y132F, and H283R), but lacked other obvious azole resistance mechanisms. Against isolates with low posaconazole MICs (range ≤0.03 to 0.25 μg/ml), seven displayed ≥16-fold increases in VT-1161 MIC over fluconazole-susceptible isolates. Among these seven isolates, all contained various ERG11 mutations, four overexpressed CDR1 by at least 2-fold, one overexpressed both ERG11 and CDR1, and one overexpressed MDR1. In contrast, there were no strong outliers for VT-1598 MICs compared against those of the triazoles.
Comparison of the MICs of VT-1161 and VT-1598 against the MICs of fluconazole (a), voriconazole (b), itraconazole (c), and posaconazole (d) in a collection of C. albicans clinical isolates. Plotted points represent the MICs of clinical isolates, with darker points representative of multiple, superimposed points. Concentration of points to the lower right of each plot represent favorable activity (low MICs relative to susceptible isolates) for VT-1161 or VT-1598 versus the comparator azole. Conversely, points concentrated to the top left of each plot represent isolates with high MICs of VT-1161 and VT-1598 relative to the comparator azole. Solid vertical lines represent the resistant clinical breakpoint for fluconazole and voriconazole, while dotted vertical lines represent the epidemiological cutoff values for posaconazole.
In vitro activity of VT-1161 and VT-1598 against strains with known azole resistance mechanisms.To identify determinants of VT-1161 and VT-1598 resistance, we evaluated the influence of specific azole resistance mechanisms on VT-1161 and VT-1598 MICs when placed in the fluconazole-susceptible isolate SC5314 (Fig. 2). Increased CDR1 and CDR2 expression through artificial activation of the TAC1 gene increased the VT-1161 MICs more than 8-fold compared to the susceptible parent strain. This increase in VT-1161 MIC was diminished, but not completely abolished, when the CDR1 gene was deleted, suggesting that the overexpression of CDR1, as well as other Tac1 target genes (most likely CDR2), was responsible for the decreased susceptibility to VT-1161. On the other hand, Tac1 activation did not result in reduced susceptibility to VT-1598, as opposed to its effect on VT-1161 fluconazole and voriconazole MICs, which increased between 4- and 16-fold compared to that of the parental strain SC5314. Posaconazole and itraconazole MICs were only slightly elevated (2-fold) by the hyperactive Tac1. Thus, it appears that while drug efflux via Cdr1 plays a role in VT-1161 resistance, Tac1 activation and the approximately 10-fold increase in CDR1 expression are not sufficient to alter MICs of VT-1598.
MICs of tested azole compounds against strains with individual known azole resistance mechanisms. Tested strains include those containing the artificially activated transcription factors Tac1 and Mrr1 in strains SCTAC1GAD1A and -B and SCMRR1GAD1A and -B, respectively, as well as Δcdr1 derivatives of SCTAC1GAD1A (SCΔcdr1TAC1GAD1A and -B), Δmdr1 derivatives of SCMRR1GAD1A and -B (SCΔmdr1MRR1GAD1A and -B), and SCUPC2R14A and -B containing the G648D gain-of-function mutation in UPC2. The MICs for the strains with artificially activated Tac1, Mrr1, and for the UPC2G648D gain-of-function mutation are displayed as the highest MIC value of both independently created A- and B- strains for each respective transcription factor. The relative fold change in expression compared to the parent strain SC5314 of CDR1 for SCTAC1GAD1A and -B (A) and MDR1 for SCMRR1GAD1A and -B (B) is shown on the left of the figure. (C) Antifungal MICs of the UPC2G648D homozygous strains SCUPC2R14A and -B.
While a hyperactive Mrr1 did not result in increased resistance to VT-1598, it caused a 4-fold increase in the MIC of VT-1161. This increase was abolished upon MDR1 deletion, suggesting that the Mdr1 transporter is involved in VT-1161 resistance. Fluconazole and voriconazole were the only tested azole drugs against the MDR1-overexpressing strain that showed a >2-fold increase in MIC (32- and 4-fold, respectively) over SC5314. By comparison, itraconazole showed a minimal 2-fold increase (1-dilution difference) in MIC, and posaconazole MICs were not affected by MDR1 overexpression. Strangely, there was a 4-fold increase in posaconazole MIC when MDR1 was deleted in the hyperactive Mrr1 strain. However, this is consistent with variability observed for posaconazole MICs in other published strains and fluconazole-susceptible clinical isolates (31).
Upregulated expression of ERG11 via artificial activation of the Upc2 transcription factor also did not affect VT-1161 or VT-1598 MICs. However, despite an approximately 4- to 8-fold increase in ERG11 expression (data not shown), the activated Upc2 strain failed to demonstrate a relevant change in voriconazole, posaconazole, and itraconazole MICs. Surprisingly, this strain also exhibited no change in fluconazole MIC, as has previously been reported. We therefore decided to also test azole susceptibilities in a strain homozygous for the G648D amino acid substitution, previously shown to be the strongest clinically derived gain-of-function mutation in Upc2 (19, 32). Strains SCUPC2R14A and -B overexpressed ERG11 relative to the parent strain SC5314 by 6.4-fold (previously published) and 4.5-fold (unpublished data), respectively (data not shown) (19). A modest 2-fold increase in fluconazole MIC in these two strains was observed compared to SC5314, whereas no changes were observed for the MICs of any of the other antifungal agents.
To compare the effects of different alterations in the azole target enzyme on the susceptibility of C. albicans to VT-1161 and VT-1598, twelve single Erg11 amino acid substitutions and four double substitutions in Erg11 were tested (Table S2, Fig. 3). The Y132F single substitution caused a substantial (16-fold) increase inVT-1161 MIC. Surprisingly, the double amino acid substitutions Y132F&K143R and Y132F&F145L had a lesser impact on VT-1161 MIC (8- and 4-fold increases, respectively) than the single Y132F substitution alone. Other amino acid substitutions did not have an appreciable effect on VT-1161 MICs, and none of the tested ERG11 mutants showed greater than a 2-fold increase in the MIC of VT-1598. The F145L and S405F single mutants and the double substitution mutants D278N&G464S and Y132F&F145L showed a slight 2-fold increase in the VT-1598 MIC compared to that against SC5314.
Relative fold change compared to SC5314 in MIC of various azole antifungal agents against strains containing single and double ERG11 mutations. Open blue circles represent VT-1598 MICs, while open black circles represent VT-1161. Open gray diamonds represent fluconazole. Solid green triangles represent voriconazole. Solid orange diamonds represents itraconazole, and solid inverted purple triangles represent posaconazole.
DISCUSSION
VT-1598 has previously demonstrated a broad spectrum of activity in vitro against yeasts such as Candida and Cryptococcus spp., molds (including Aspergillus spp.), and endemic fungi (33) and has shown improved survival and reduced fungal burden in murine models of central nervous system coccidioidomycosis (34) and cryptococcosis (35). Pertinent to the present study, VT-1598 has also recently shown potent in vitro and in vivo antifungal activity against fluconazole-sensitive and -resistant Candida spp. isolated from chronic mucocutaneous candidiasis patients (36). Structurally, while the tetrazole moiety has lower affinity for interaction with the heme iron of CYP51, other structural modifications have made the drug more fungus specific. For example, a critical H-bond between VT-1598 and the CYP51 enzyme of many fungi likely confers its broad activity (37). This greater selectivity may decrease the potential for undesirable adverse effects and drug interactions that occur with the triazoles through inhibition of human cytochrome P-450 enzymes.
Our study supports the previous finding that VT-1598 has potent activity against C. albicans isolates. Overall, VT-1598 displayed the lowest MIC50 and MIC90 values compared to fluconazole, voriconazole, posaconazole, itraconazole, and VT-1161 against the clinical isolates tested. More importantly, VT-1598 MICs often remained unchanged from its baseline measurement against SC5314 and other fluconazole-susceptible clinical isolates even against isolates containing known resistance mechanisms, indicating that this tetrazole may retain activity against isolates that are normally less susceptible to other azole antifungals. This included multiple fluconazole-resistant isolates with various combinations of CDR1, MDR1, and ERG11 expression increases and mutations in the ERG11 gene.
Interestingly, when tested against laboratory strains containing individual azole resistance mechanisms, VT-1598 MICs changed relatively little. Traditional azole resistance mechanisms, such as efflux pump overexpression (Cdr1 and Mdr1) and overexpression of the azole target (Erg11), did not alter VT-1598 MICs within the concentration ranges tested here. While it is possible that testing lower concentrations might reveal differences in MIC, the clinical relevance at such low concentrations is questionable. Our current finding suggests that these mechanisms individually are not sufficient to confer resistance to VT-1598. Previously, the Tyr132 and Lys143 substitutions in Erg11 were reported to have the strongest individual effects on fluconazole and voriconazole MICs (20). The combination substitutions Y132F&K143R and Y132F&F145L, which have been shown to have some of the strongest increases in fluconazole and voriconazole MICs, respectively, did not appreciably change the MIC of VT-1598. Against VT-1161, both these double substitutions showed less of an effect than the single amino acid substitution Y132F. The K143R substitution is thought to alter the H-bond strength of the heme ring propionates of the C. albicans Erg11 protein, which may possibly interfere with the coordination of the azole ring nitrogen and the iron of the CYP51 heme group, and the F145L amino acid substitution is located on the Erg11 proximal surface, allowing possible interactions with NADPH-cytochrome P450 reductase (20, 38). Based on the crystal structure of the C. albicans CYP51 enzyme complexed with VT-1161, the Y132F substitution is thought to altogether abolish one of six H-bonds between VT-1161 and the protoporphyrin IX propionates on Erg11 (38). It is possible that the Y132F substitution is more important to VT-1161 resistance than either K143R or F145L and that combination mutations might interfere with the primary Y132F substitution, thus leading to the differences in the observed VT-1161 MICs.
Against a collection of predominantly fluconazole-resistant clinical isolates, VT-1161 showed good activity, though its individual MIC50 and MIC90 were higher compared to VT-1598. In contrast to VT-1598, the potency of VT-1161 appeared to be more affected by the presence of CDR1 and MDR1 overexpression and ERG11 mutations. This was supported by the significant positive correlation established between CDR1/MDR1 expression and VT-1161 MIC in C. albicans clinical isolates. In addition, susceptibility testing in strains containing individual mechanisms of azole resistance, wherein the CDR1-overexpressing strains SCTAC1GAD1A and -B and the MDR1-overexpressing strains SCMRR1GAD1A and -B demonstrated increased VT-1161 MICs, further supports Mrr1 and Tac1 as mediators of resistance to VT-1161, at least in part through increased production of the transporters Mdr1 and Cdr1, respectively. The recent work by Monk et al. also demonstrated that both the Cdr1 and Mdr1 efflux pumps reduced the effectiveness of VT-1161, and activity against Cdr1- and Mdr1-overexpressing isolates could be restored via the Cdr1- and Mdr1-specific inhibitors RC21v3 and MCC1189, respectively (39). The Erg11 amino acid substitutions Y132F, Y132F&K143R, and Y132F&F145L also resulted in shifts in VT-1161 MIC and may contribute to VT-1161 resistance. However, VT-1161 retained activity against a number of isolates with known azole resistance mechanisms. The tested ERG11 mutant strains containing the Y132H, K143R, F145L, E266D, D278N, S405F, G448E, F449V, G450E, G464S, and D466E single substitutions and the D278N&G464S and G450E&I483V double substitutions showed little change in VT-1161 MICs compared to the susceptible parent strain SC5314. Thus, VT-1161 has potential to be a future treatment option of azole-resistant C. albicans infections or recurrent infections previously treated with older members of the azole class.
Within the five clinical isolates that displayed greatly reduced susceptibility to VT-1161, VT-1598, and the other commercially available triazoles tested, one isolate contained an early stop codon at Trp131 within the ERG3 gene, which encodes sterol Δ5,6-desaturase and is critical for ergosterol biosynthesis in C. albicans. It has been previously reported that mutations in ERG3 can result in azole resistance and alternative sterol biosynthesis by avoidance of accumulation of toxic sterol intermediates through defective desaturase enzyme (26, 27, 40). The inhibition of the azole target 14α-lanosterol demethylase causes accumulation of the toxic sterol intermediate, 14α-methylergosta-8,24(28)-dien-3β,6α-diol, which is thought to be the source of the fungistatic effect seen with azole antifungal class (24, 25, 41). However, dysfunctional Erg3 results in alternative sterol usage and an inability to produce this toxic intermediate. Therefore, mechanisms that result in a nonfunctional Erg3 might render an isolate resistant to the azole antifungal drug class, as is seen in the case of the isolate containing the premature stop codon in ERG3.
In summary, the in vitro activity of VT-1161 and VT-1598 against azole-resistant C. albicans clinical isolates and strains with known azole resistance mechanisms suggests that they may prove useful against resistant C. albicans infections. Furthermore, given their low and relatively unchanged MICs against many azole-resistant strains, it is possible that VT-1161 and VT-1598 may fill some of the gaps in coverage against azole-resistant isolates. This, in combination with the potential for favorable safety and drug interaction profiles, makes VT-1161 and VT-1598 attractive options as alternative therapies for azole-resistant C. albicans infections. However, the presence of strongly resistant isolates, such as the five clinical isolates with greatly increased MICs to all azoles tested here, suggests the existence of azole resistance determinants that can provide obstacles to the successful utilization of all azoles, including these new tetrazoles. Further investigation should be undertaken to identify the mechanism(s) responsible for resistance to these agents.
MATERIALS AND METHODS
Isolate and strain growth conditions.Sixty-eight clinical C. albicans isolates were obtained from the University of Iowa. C. albicans isolates and strains were cultured from –80°C freezer stock (40% glycerol in yeast extract-peptone-dextrose [YPD] media) onto YPD-agar plates overnight at 30°C. Colonies from YPD-agar plates were then either streaked onto Sabouraud dextrose agar for azole susceptibility testing or grown in liquid YPD media and incubated overnight at 30°C for preparation of genomic DNA.
Construction of C. albicans strains.Table 2 lists the constructed strains used in this study. Ten single ERG11 mutations and four double mutations were selected from previous studies (20). Two additional strains expressing the Y132H and D278N ERG11 gene mutations were created in a previous study utilizing the SAT flipper cassette (42, 43). Briefly, to create the mutant strain SCERG11R1S1C1, ERG11 gene fragments were generated by primers pairs CaERG11_1F with CaERG11SOE-3R_Y132H and CaERG11SOE-2F_Y132H with CaERG11_4R using SC5314 template genomic DNA (Table S3). Short-overlapping extension PCR was used to fuse the resulting ERG11 gene fragments using nested primers CaERG11-AF_(ApaI) and CaERG11-BR_(XhoI). For mutant strain SCERG11R3S3C1, ERG11 gene fragments were generated by primers pairs CaERG11_1F with CaERG11SOE-6R using template genomic DNA from clinical isolate 43 and CaERG11SOE-5F with CaERG11_4R using SC5314 genomic DNA. The resulting fragments were again fused into a single fragment containing either the D278N-containing mutant ERG11 open reading frame (ORF) using nested primers CaERG11-AF_(ApaI) and CaERG11-BR_(XhoI). In constructing the plasmids used in the transformation of both strains, the ERG11 3′ flanking sequence was amplified from SC5314 genomic DNA and primers CaERG11_C-F′ and CaERG11_D-R′. The inserts 3′ of the ERG11 ORF were digested with restriction enzymes NotI and SacII and cloned into the pSFS2-derived plasmid pBSS2 previously described by Vasicek et al. containing the SAT1 flipper cassette from Reuss et al. (42) and Vasicek et al. (44) to create plasmid pERG11CD. The ERG11 ORF-containing fragments with either the Y132H or D278N mutations were digested using restriction enzymes ApaI and XhoI, gel excised, and cloned into the plasmid pERG11CD to generate plasmids pERG11A1 and pERG11A3. Plasmids were digested with restriction enzymes ApaI and SacII and used to transform strain SC5314 by electroporation to generate heterozygous ERG11 mutants. Recycling of the nourseothricin selection marker through 24 h of growth in YPD and repeat transformation of the resultant strains generated the homozygous ERG11 allele replacements SCERG11R1S1C1 and SCERG11R3S3C1, confirmed via Southern hybridization and Sanger sequencing. The artificially activated Tac1, Mrr1, and Upc2 mutants used in this study, as well as SCΔcdr1TAC1GAD1A and -B, containing the artificially activated TAC1 allele in a cdr1Δbackground, were described in a previous study (45). Strains SCΔmdr1MRR1GAD1A and -B were constructed by introducing the artificially activated MRR1 allele from plasmid pMRR1-GAD1 (32) into the mdr1Δ mutants SCMDR1M4A and -B (46), respectively.
Constructed strains used in this study
ERG11 amplification and sequencing.Table S3 lists the primers used for ERG11 amplification and sequence verification. The ERG11 ORF of each isolate was PCR amplified from genomic DNA using the primers CaERG11_F_Amp and CaERG11_R_Amp. PCR products were purified using a QIAquick PCR purification kit (Qiagen), and the product was sequenced on an ABI 3130XL genetic analyzer using sequencing primers. Sequencing was accomplished in duplicate in independently grown isolates.
Relative gene expression by real-time PCR.Expression levels of the genes CDR1, MDR1, and ERG11 in clinical C. albicans isolates were measured in a previous study, and CDR1, MDR1, and ERG11 expression of laboratory strains were measured similarly using previously described methods (19). Briefly, first-strand cDNA was generated from 1 μg of extracted RNA for each strain using the SuperScript VILO master mix (Invitrogen) reaction kit. Quantitative PCRs were performed on the StepOnePlus real-time PCR system (Applied Biosystems) in technical and biological triplicates for each sample. SYBR green PCR master mix (Applied Biosystems) was used for amplification detection of candidate genes against the CaACT1 normalizing gene. Calculation of the relative quantitation values of CDR1, ERG11, and MDR1 gene expression was accomplished using the StepOne Software v2.3 (Applied Biosystems). Primers used in the amplification of genes measured via qPCR are listed in Table S3.
Susceptibility testing.The MICs of VT-1161, VT-1598, fluconazole, voriconazole, posaconazole, and itraconazole were measured using broth microdilution methods in accordance with the Clinical and Laboratory Standards Institute (47, 48). Microtiter plates (96 well) containing RPMI 1640 medium (0.165 M morpholinepropanesulfonic acid with l-glutamine and without sodium bicarbonate [pH 7.0]) were used to incubate strains across serially diluted concentrations of the each azole. Concentrations ranged from 0.015 to 8 μg/ml for VT-1161 and VT-1598, 0.125 to 64 μg/ml for fluconazole, and 0.03 to 16 μg/ml for voriconazole, posaconazole, and itraconazole. The MICs were visually read at 24 h postincubation at 35°C as the minimum concentration required to reduce growth of cells by approximately 50% or greater compared to drug-free control wells. MICs were determined in duplicate for clinical isolates, ERG11 mutant strains, and laboratory strains SCTAC1GAD1A and -B, SCΔcdr1TAC1GAD1A and -B, SCMRR1GAD1A and -B, SCΔmdr1MRR1GAD1A and -B, and SCUPC2GAD1A and -B. When reporting MICs for strains and isolates, the higher of the MICs was used in this analysis (Tables S1 and S2), though 98% (592/606) of the MIC duplicate readings were identical or within a single dilution of each other. The geometric mean MIC (GM MIC), MIC50, and MIC90 values were reported for clinical C. albicans isolates for each triazole and tetrazole agent used in this study. The MIC50 and MIC90 values reported for VT-1161, VT-1598, fluconazole, voriconazole, itraconazole, and posaconazole were defined as the minimum drug concentrations required to inhibit 50 and 90% of the clinical C. albicans isolates tested, respectively.
Statistical analysis.A point-biserial correlation or phi coefficient was used for all continuous and dichotomous variables, respectively, to identify possible predictors of azole resistance. For all statistical tests, a P value of <0.05 was considered significant. Statistical calculations were performed using IBM SPSS analytical software, version 23.
ACKNOWLEDGMENTS
We thank Daniel Diekema of the University of Iowa for graciously providing the C. albicans clinical isolates.
We acknowledge the generous support of this work by National Institutes of Health grant R01 AI058145 to P.D.R.
N.P.W. has received research support to the UT Health San Antonio from Astellas, bioMérieux, Cidara, F2G, Merck, and Viamet, and has served on advisory boards for Merck, Astellas, Toyama, and Viamet. C.M.Y., R.J.S., and E.P.G. are employees of Viamet Pharmaceuticals, Inc. All other authors have no conflicts of interest.
FOOTNOTES
- Received 20 February 2019.
- Returned for modification 10 March 2019.
- Accepted 16 March 2019.
- Accepted manuscript posted online 25 March 2019.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.00341-19.
- Copyright © 2019 American Society for Microbiology.
