An oxindole efflux inhibitor potentiates azoles and impairs virulence in the fungal pathogen Candida auris

Abstract

Candida auris is an emerging fungal pathogen that exhibits resistance to multiple drugs, including the most commonly prescribed antifungal, fluconazole. Here, we use a combinatorial screening approach to identify a bis-benzodioxolylindolinone (azoffluxin) that synergizes with fluconazole against C. auris. Azoffluxin enhances fluconazole activity through the inhibition of efflux pump Cdr1, thus increasing intracellular fluconazole levels. This activity is conserved across most C. auris clades, with the exception of clade III. Azoffluxin also inhibits efflux in highly azole-resistant strains of Candida albicans, another human fungal pathogen, increasing their susceptibility to fluconazole. Furthermore, azoffluxin enhances fluconazole activity in mice infected with C. auris, reducing fungal burden. Our findings suggest that pharmacologically targeting Cdr1 in combination with azoles may be an effective strategy to control infection caused by azole-resistant isolates of C. auris.

Fig. 1. Screen of the BU-CMD library identifies azoffluxin as a fluconazole (FLC) potentiator against C. auris.

a BU-CMD library was screened at 50 μM in the presence or absence of 128 μg/mL of FLC in RPMI medium at 30 °C for 48 h. Growth of C. auris strain VPCI 673/P/12, as determined by optical density at 600 nm (OD₆₀₀), is plotted in the presence of each CMD compound alone and in combination with FLC. Dotted lines represent 7-median absolute deviations from the median for each condition. Red circles indicate compounds that showed significant bioactivity. The shaded quadrant indicates compounds that show enhanced activity in the presence of fluconazole, with azoffluxin shown as a filled red circle. b Checkerboard assays with azoffluxin (CMLD012336) and FLC were performed in RPMI at 30 °C by titering 2-fold dilution series of azoffluxin and FLC. Growth in each well is presented in heat-map format based on the OD₆₀₀ of wells at 48 h relative to the no-drug control (see color bar). The fractional inhibitory concentration index (FICI) was calculated to assess interaction effect, with a value <0.5 indicating synergy. c Structure of CMLD012336 (azoffluxin). d FLC Etest strips in the presence and absence of 50 μM azoffluxin. C. auris cells (1 × 10⁶) C. auris cells were plated on YPD agar, the E-test strip was added, and plates were incubated at 30 °C for 24 h prior to imaging. e Dose-response assay based on 2-fold serial dilution of azoffluxin starting from 50 μM in the absence and presence of indicated concentration of FLC for C. auris Ci6684 (Erg11^Y132F), C. albicans (SN95), C. glabrata (BG2), or S. cerevisiae (BY4741), respectively. The highest FLC concentrations that did not affect growth alone for each species was used. Dose-response assays were incubated for 48 h at 30 °C in RPMI. Growth in each well was quantified by the OD₆₀₀ of treated wells relative to the respective no-drug control (see color bar in b). Source data are provided as a Source Data file.

Fig. 2. Azoffluxin increases intracellular accumulation of fluconazole (FLC) by inhibiting Cdr1-mediated efflux in C. auris.

a Abundance of ergosterol (blue), lanosterol (red), and 14-α-methyl-3,6-diol (yellow) was determined in Ci6684 after compound treatment (• indicates concentrations used in combination treatment) relative to internal cholesterol standard. Growth inhibition (%) caused by each treatment is presented in table. Data are presented as mean ± SD of technical triplicates. Significance was determined by two-sided unpaired Student’s t test of condition compared to untreated; *p-value < 0.05, **p-value < 0.01. Fold-change for each treatment is indicated above the respective bar. b Intracellular concentrations of FLC (green) and azoffluxin (gray) were measured after treatment for 1 h. Data are presented as mean ± SD of technical triplicates. Significance was determined by two-sided unpaired Student’s t test, **p-value = 0.003, and ***p-value > 0.001. c Transcript levels of Ci6684 CDR1 (teal) and CDR4-1 (red) were measured. Cells were treated with indicated concentrations of compound (• indicates concentrations used in combination treatment). Transcript levels were normalized to ACT1 and GPD1 and are relative to the untreated control. Data are presented as mean ± SEM of technical triplicates. Significance of differences between untreated control and treatment was determined by two-sided unpaired Student’s t test; *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001. Fold-change is indicated above each bar. d Ci6684 was treated with azoffluxin, followed by addition of Nile red. Scale bar represents 5 µm. e Cells from Fig. 2d and Fig. S2 were analyzed by flow cytometry. Histograms depict relative fluorescence intensity (PE-A) of events, values depict median fluorescence intensity (MFI). Table displays mean fold-change in MFI of azoffluxin-treated, Nile red stained cells ± SD for biological triplicates. Significance of difference determined by a two-sided unpaired Student’s t test, *p-value < 0.05 compared to parental average MFI. f Dose-response assays were conducted as in Fig. 1e. FLC was applied as a 2-fold dilution series in the absence or presence of azoffluxin (50 µM). Growth was monitored and normalized to no-drug control (see color bar). Source data are provided as a Source Data file.

Fig. 3. Azoffluxin potentiates intracellular acting compounds against C. auris, to a similar degree as deletion of CDR1.

Dose-response assays were conducted with a C. auris Ci6684 parental strain in the absence and presence of 25 µM azoffluxin where indicated, as well as with a strain with the efflux pump gene CDR1 deleted. Indicated compounds were titered in a 2-fold serial dilution. Growth was measured after 24 h in YPD as described in Fig. 1e (see color bar). Source data are provided as a Source Data file.

Fig. 4. Synergistic activity of azoffluxin is clade specific.

Checkerboard assays were performed fluconaozle (FLC) and azoffluxin as described in Fig. 1b with isolates from each major clade of C. auris. CDC identification number is followed by the clade number to which the isolate belongs. Relative growth was measured in YPD medium after 24 h using OD₆₀₀ and normalized to a no-drug control well (see color bar). The FICI calculated for each checkerboard is shown in the top right of each plot, with values <0.5 indicating synergy, values >0.5 indicating no interaction, and N/A indicating an FICI that could not be calculated due to a lack of growth inhibition. Source data are provided as a Source Data file.

Fig. 5. Azoffluxin does not potentiate fluconazole (FLC) against most clade III isolates despite intracellular FLC accumulation and increased CDR1 expression.

a Relative transcript levels of MDR1 (B9J08_003981) and b relative transcript levels of CDR1 (B9J08_000164) were measured in clade I isolate Ci6684 (gray) and clade III isolates B11221 (red) and B11222 (blue) (• indicates concentrations of FLC and azoffluxin used in combination (combo) treatment). Data are presented as mean ± SEM of technical triplicates. A two-sided unpaired Student’s t test was performed to evaluate significance of differences between Ci6684 and each clade III isolate *p-value < 0.05, **p-value < 0.01, and ***p-value < 0.001. c Flow cytometry was used to measure the Nile red accumulation in C. auris clade III strains as described in Fig. 2e. Values in histogram plots depict median fluorescence intensity (MFI) and table shows the mean fold-change in MFI ± SD for biological triplicates. d The relative intracellular azoffluxin abundance and e the relative intracellular FLC abundance was quantified by LC-MS as described in Fig. 2b in Ci6684 (gray), B11221 (red), and B11222 (blue). Data are presented as mean ± SD of technical triplicates. Significance of differences between azoffluxin and the combination treatment for each strain was determined by two-sided unpaired Student’s t test, *p-value < 0.05, and ***p-value < 0.001 comparing. f Checkerboard assay as described in Fig. 1b using parental clade III isolate B12037 and the strain with CDR1 deleted, in YPD medium. Relative growth was measured after 24 h as described in Fig. 1b (see color bar). The FICI for each checkerboard is shown as described in Fig. 4. Source data are provided as a Source Data file.

Fig. 6. Azoffluxin enhances fluconazole (FLC) activity against azole-resistant C. albicans isolates.

a Checkerboard assays as described in Fig. 1b were performed in YPD with isolates of C. albicans. Strains CaCi-2 and CaCi-17 represent early and late clinical isolates in which FLC resistance evolved over time. Growth was measured after 24 h using OD₆₀₀ and normalized to a no-drug control well (see color bar). The FICI calculated for each checkerboard is shown in the top right of each plot, with values <0.5 indicating synergy and >0.5 indicating no interaction. b Dose-response assays were conducted in YPD medium with a C. albicans parental strain, and strains with gain of function mutations in TAC1 as indicated. FLC was titered in a 2-fold dilution on the x-axis in the absence and presence of 50 µM azoffluxin. Growth was measured at 24 h using OD₆₀₀ and normalized to a no-drug control well (see color bar). c Flow cytometry was used to measure relative Nile red accumulation in C. albicans strains as described in Fig. 2e. Values in histogram plots depict median fluorescence intensity (MFI) and table shows the mean fold-change in MFI ± SD for biological triplicates. d Relative intracellular levels of FLC and e relative intracellular azoffluxin were measured by LC-MS in C. albicans strains SN95 (gray) and CaCi-17 (blue) after treatment (combo treatment: 6.25 µM azoffluxin, 8 µg/mL FLC) for 1 h as described in Fig. 2b. Data are presented as mean ± SD of technical triplicates. Significance of differences was determined by a two-sided unpaired Student’s t test comparing fluconazole alone to the combination, *p-value = 0.013 and **p-value >  0.011. Source data are provided as a Source Data file.

Fig. 7. Preliminary characterization of the in vivo potential of azoffluxin.

a Human cells (293T) expressing luciferase were grown in DMEM medium overnight. 24 h later the indicated concentrations of compounds (• indicates concentrations of fluconazole (FLC) and azoffluxin used in combination (combo) treatment) were added to cells alone (gray) or those infection with C. auris (blue). Co-cultures were incubated for 48 h at 37 °C followed by measurement of luminescence. Data are presented as mean ± SD of quadruplicate wells. Significance of differences between 293T cells alone versus co-cultures was determined by two-sided unpaired Student’s t test, (***p-value < 0.001 b Periodic-Acid Schiff (PAS) staining was used to visualize cells in co-culture. Light purple staining identifies 293 T cells and the bright pink signal indicates C. auris. Scale bar represents 50 µm. c Plasma stability of azoffluxin and relevant control compounds, gepinacin (GPN) and caspofungin (CF). Samples were incubated in 100% mouse plasma at either at 37 °C with 5.5% CO₂ (maroon) or on ice (gray), or in the absence of serum in YPD (black). The drug-plasma mixtures were diluted 1:10 into C. auris Ci6684-inoculated YPD medium. Relative growth was measured after 48 h at 30 °C by OD₆₀₀. Data are presented as mean ± SD between technical triplicates, two-sided unpaired Student’s t test was used to determine significance of difference between 37 °C with 5.5% CO₂ condition compared to ice condition for each treatment, ***p-value < 0.001. d Plasma concentrations of azoffluxin in mice following intraperitoneal (IP) bolus administration of compound (10 mg/kg). Azoffluxin was quantitated in mouse blood (n = 3) by LC-MS/MS. Data are presented as mean ± SD of three mice. Pharmacokinetic properties shown in the table below were evaluated using Analyst software (AB Sciex.) and the noncompartmental analysis tool in WinNonlin (Certara, Corp.). e Tolerability of azoffluxin in mice was evaluated by treating neutropenic ICR (CD-1) mice with azoffluxin 10 mg/kg IP twice daily for 4 days and monitoring the health and survival of treated (blue) and untreated (black) mice (n = 5) for 21 days. Source data are provided as a Source Data file.

Fig. 8. Azoffluxin increases the antifungal activity of fluconazole (FLC) in mice.

a Checkerboard assays were performed as described in Fig. 1b with C. auris clade IV isolate B11801. Relative growth was measured after 24 h using OD₆₀₀ and normalized to no-drug control wells (see color bar). The FICI is shown in the top right of each plot, with values <0.5 indicating synergy. b Kidney fungal burden (CFU) in mice from each treatment group that had been infected with C. auris B11801. Input is the CFU recovered in an aliquot of the fungal suspension used to inoculate mice. All other values are the CFU recovered from kidney homogenates after 4 days of treatment. Fluconazole was administered at 32 mg/kg intraperitoneally twice daily and azoffluxin at 10 mg/kg subcutaneously four-times daily. Data are presented as mean ± SD of n = 3 mice per treatment group. Experiment was performed in two independent replicates (purple and blue). The significance of differences between combination treatment and treatment with each compound alone was determined for each replicate by two-sided unpaired Student’s t test, **p-value < 0.01 and ***p-value < 0.001. Source data are provided as a Source Data file.