Candida albicans strain SC5314 (Fonzi and Irwin, 1993) and C. glabrata strain BG2 (Kaur et al., 2007) were grown at 30°C on YPD [1% yeast extract, 2% bacteriological peptone (LabM, UK) and 2% glucose (Sigma-Aldrich, USA)] agar plates. Dimethyl sulfoxide (DMSO) (VWR International, Belgium) was used as a solvent to prepare stock solutions of miconazole (MCZ; Sigma-Aldrich), fluorescently labeled imidazole derivative (Benhamou et al., 2017), tetraethylammonium bromide (TCI Europe, Belgium), itraconazole (TCI Europe) and domiphen bromide (DB; Chemische Fabrik Berg GmbH). RPMI 1640 medium (pH 7.0) with L-glutamine and without sodium bicarbonate was purchased from Sigma-Aldrich and buffered with MOPS (Sigma-Aldrich). An ascorbic acid (Sigma-Aldrich) stock was made in Milli-Q water.

To determine whether a treatment is fungicidal against (i) biofilms or (ii) planktonic cultures in stationary phase of C. albicans SC5314 or C. glabrata BG2, a fungicidal activity assay was performed as described in Tits et al. (2020). (i) Briefly, C. albicans or C. glabrata overnight cultures were diluted to an optical density at 600 nm (OD_600nm) of 0.1 in RPMI 1640 medium, followed by biofilm growth in a 100 μL volume at 37°C. Twenty-four hours old biofilms were treated with the appropriate single compound or combination (e.g., MCZ and DB). After an incubation period of 24 h (C. albicans) or 48 h (C. glabrata) at 37°C, the number of CFU was determined. To this end, biofilms were washed with phosphate-buffered saline (PBS) and scraped off from the bottom of the plate. Serial dilutions were plated on YPD agar plates, which were subsequently incubated for 24 h at 37°C. Finally, the number of colonies was determined. (ii) C. albicans or C. glabrata stationary-phase cultures at OD_600nm = 1 in RPMI 1640 medium were treated with the appropriate single compound or combination during 2.5 h shaking at 37°C. To determine the number of CFU, cultures were washed and resuspended in PBS, followed by plating of serial dilutions on YPD agar plates, which were subsequently incubated at 37°C. After an incubation period of 24 h, the number of colonies was determined.

To study the effect of DB on azole internalization in C. glabrata BG2 planktonic cultures, stationary cultures at OD_600nm = 1 in RPMI were treated with a 2-fold dilution series of a fluorescently labeled imidazole derivative (FKD) in presence or absence of 25 μM DB. 1 μM SYTOX Green nucleic acid stain (Thermo Fisher Scientific, USA) was added to the samples as a dead cell stain. After 2.5 h treatment shaking at 37°C, samples were washed with PBS and subjected to a flow cytometric analysis on a BD Influx™ cell sorter. 100,000 cells per samples were analyzed, measuring fluorescence at 530/40 nm (FL2_ λ_ex = 488 nm) and 670/30 nm (FL12_ λ_ex = 633 nm) to detect killing (SYTOX green +) and FKD internalization (FKD +), respectively.

To investigate the subcellular localization of FKD, C. glabrata BG2 planktonic stationary cultures at OD_600nm = 1 in RPMI were treated with 250 μM FKD in the presence or absence of 25 μM DB, followed by an incubation period of 2.5 h shaking at 37°C. Samples were washed and resuspended in PBS and subsequently imaged with Confocal LSM880 airyscan, using ZEN black software. A 100x/1.46 oil immersion objective and a 2x or 8x computer zoom were used. We used the 633 nm laser to visualize FKD. Samples treated with FKD alone and with FKD in the presence of DB were analyzed with the same microscopy settings.

To assess the effects of DB on vacuolar pH, 2′,7′-Bis-(2-Carboxyethyl)-5-(and-6)-Carboxyfluorescein, Acetoxymethyl Ester (BCECF,AM; Thermo Fisher Scientific) staining was used as described previously (Struyfs et al., 2020). Concanamycin A (Sigma-Aldrich) was used as a positive control for an elevated vacuolar pH. C. albicans SC5314 and C. glabrata BG2 planktonic stationary cultures at OD_600nm = 1 in RPMI were treated with a 2-fold dilution series of DB. After 2 h treatment at 37°C, 50 μM BCECF,AM was added to each sample, followed by another 30′ incubation period at 37°C. Subsequently, cells were washed and resuspended in PBS. Of 100 μL cell suspension, optical density at 490 nm (OD490nm) and fluorescence intensity for excitation at 490 nm (I490nm) and 450 nm (I450nm), both measured at an emission wavelength of 535 nm, were measured. [(I490nm/I450nm)/OD490] was calculated and considered as a relative vacuolar pH value of the treated cells.

Superoxide detection was performed as described previously (De Cremer et al., 2016). Briefly, 24-h-old biofilms of C. albicans SC5314 or C. glabrata BG2 were treated with 150 or 500 μM MCZ, respectively, 37.5 μM DB or a combination of both in the presence or absence of 16 mM ascorbic acid (Sigma-Aldrich). After an incubation period of 24 h (C. albicans) or 48 h (C. glabrata) biofilms were washed and diluted in PBS. Then, biofilms were scraped off from the bottom of the microplate and 10 μL of this cell suspension was separated for CFU determination. To the remaining cells, a final concentration of 20 μM dihydroethidium (Thermo Fisher Scientific) was added, followed by an incubation period of 20 min at room temperature in the dark. Finally, cells were washed with PBS and fluorescence at λ_ex 510 nm and λ_em 595 nm was measured. Values were corrected for fluorescence detected in blank wells.

To determine the effects of DB on the growth inhibitory potential of a triazole (e.g., itraconazole) against C. albicans, we performed a (i) planktonic growth inhibition assay and a (ii) biofilm inhibition assay. (i) The planktonic growth inhibition assay was performed according to the standard Clinical and Laboratory Standards Institute (CLSI) protocol M27-A3 (CLSI, 2008) as described in Vriens et al. (2015). (ii) A biofilm inhibition assay was performed as described in Cools et al. (2017) with some modifications. C. albicans overnight cultures were diluted to an OD_600nm = 0.1 in RPMI and added to the wells of a microplate, containing itraconazole, DB or a combination of both. After a 1-h adhesion phase (37°C), biofilm cells were washed with PBS and the same treatments were added. After 24 h of biofilm growth, biofilm formation was quantified with Cell-Titer Blue (CTB; Promega, USA).

To analyze the experimental data, GraphPad Prism (version 6) was used. Significant CFU differences were assessed based on the mean log values. Furthermore, significant CFU differences as well as differences in fluorescence units per 1,000 CFU upon dihydroethidium staining were determined using a two-way analysis of variance (ANOVA), followed by the appropriate multiple-comparison test. Significant differences in (I490nm/I450nm)/OD490nm were assessed by means of a one-way ANOVA and the appropriate multiple-comparison test. FlowJo™ 10 software and FIJI software (Schindelin et al., 2012) were used to analyze flow cytometry data and Airyscan images, respectively.

The effects of DB on azole internalization and cell death in C. glabrata BG2 planktonic stationary cultures were studied via flow cytometry using two fluorescent dyes, FKD and SYTOX green, respectively. SYTOX green enters the cell through compromised membranes of dead cells, subsequently binds to DNA and fluoresces (Roth et al., 1997). A tight negative correlation (correlation coefficient r = −0.74) between the number of CFU and the number of SYTOX green fluorescent cells was observed upon FKD treatment in the presence of DB, indicating that SYTOX green can indeed be used to assess yeast cell death induced by uptake of FKD (Supplementary Figure 3). Cultures were treated with a series of FKD concentrations in presence or absence of a fixed DB concentration, with two conditions shown in Figure 1. At low FKD concentrations, significantly more cells had internalized FKD in the presence of 25 μM DB as compared to FKD treatment alone; while a minor fraction (22%) of the cell population was killed upon treatment with 25 μM DB alone, approximately all cells that internalized FKD were dead (SYTOX green positive) (Figure 1). This indicates that DB greatly augmented the cidality of MCZ, e.g., 0.95 μM MCZ and 25 μM DB (33% killed).

At high concentrations of FKD, it was internalized into all cells in the population, even in the absence of DB. However, while only 6% of the cell population was dead in the presence of FKD alone, 93% of the population was dead (SYTOX green positive) in the presence of both drugs (Figure 1). Furthermore, 7% of the cell population internalized FKD in the presence of DB, yet had no compromised membranes and was not dead (SYTOX green negative). In contrast to treatment with low FKD concentrations where increased FKD levels were observed inside the cells upon addition of DB, the FKD levels inside the C. glabrata cells treated with high FKD concentrations changed only minimally in the presence of DB compared to single FKD treatment (Supplementary Figure 4). However, the addition of DB significantly increased the number of dead cells at high FKD concentrations. This finding cannot be explained by increased FKD internalization alone and requires further investigation. How a small subpopulation (7%) survived the combination of FKD+DB also remains to be determined.

Next we investigated the subcellular localization of FKD in presence and absence of DB, using confocal microscopy (Airyscan). Consistent with previous studies (Benhamou et al., 2017, 2018), FKD when treated alone localized to the mitochondria. By contrast, the combination of FKD and DB resulted in cells staining brightly and homogeneously throughout the cytoplasm (Figure 2), indicating that DB alters the cytoplasmic distribution of FKD in C. glabrata cells.

As vacuolar sequestration has been seen as an azole tolerance strategy in both S. cerevisiae and C. albicans (Khandelwal et al., 2019), we investigated whether DB itself affects vacuolar integrity using BCECF,AM, a fluorescent dye indicator specifically reporting vacuolar pH and hence, vacuolar function (Plant et al., 1999). More specifically, the ratio of fluorescence intensity of BCECF,AM from excitation at 490 nm (I490nm) to fluorescence intensity from excitation at 450 nm (I450nm), both measured at an emission wavelength of 535 nm, was calculated and normalized for the cell culture biomass (OD490nm). An increase in (I490nm/I450nm)/OD490nm indicates an elevated vacuolar pH and, accordingly, vacuolar dysfunction. Concanamycin A, a vacuolar H⁺-ATPase inhibitor, was used as a positive control for elevated vacuolar pH (Dröse and Altendorf, 1997). Upon treatment of both C. albicans SC5314 and C. glabrata BG2 cultures with sublethal DB concentrations, an increased (I490nm/I450nm)/OD490nm value was observed (Figure 3), suggesting that DB indeed negatively affects vacuolar function. Interestingly, quaternary ammonium compounds that cannot potentiate MCZ, such as tetraethylammonium bromide (Tits et al., 2020), did not affect vacuolar function (Supplementary Figure 5). These data support the hypothesis that DB enables release of vacuole-sequestered azoles, probably via its action on the vacuolar membrane. DB thereby increases intracellular availability of MCZ, resulting in cell killing as MCZ is characterized by fungicidal action (François et al., 2006). However, whether FKD in the absence of DB also localizes to the vacuoles of C. glabrata cells remains to be determined as well as the possible causal link between DB's negative effects on vacuolar function and the altered cytoplasmic FKD distribution.

Although FKD-DB was not characterized by increased fungicidal activity against stationary cultures of C. albicans as compared to treatment with FKD or DB alone, we also investigated FKD internalization and subcellular localization in this pathogen. In line with the absence of a potentiator action by DB in this setup, flow cytometry data indicated that FKD was internalized in approximately all C. albicans cells, irrespective of whether DB was present or not, but this did not result in cell killing: <1% of the cells was killed (data not shown). In addition, confocal microscopy data indicated that the addition of DB did not result in an altered cytoplasmic distribution of FKD in C. albicans cells as compared to single FKD treatment (data not shown). These data point to the inability of DB to affect FKD's intracellular distribution in C. albicans, which is linked to the absence of a potentiator action by DB in this setup.

In contrast to most azoles, MCZ is characterized by fungicidal action against planktonic Candida cultures and, at very high supra-therapeutic concentrations (5 mM), against biofilms (François et al., 2006; Vandenbosch et al., 2010). In addition to its inhibition of ergosterol biosynthesis, MCZ induces the accumulation of reactive oxygen species (ROS) in planktonic fungal cultures (Kobayashi et al., 2002; François et al., 2006). Here, we investigated whether MCZ-DB treatment results in increased ROS, more specifically superoxide, accumulation in C. albicans biofilms, by staining treated cells with dihydroethidium (DHE). We previously revealed that MCZ induces the accumulation of superoxide in C. albicans biofilm cells and that this accumulation is essential for MCZ's fungicidal activity against C. albicans biofilms (De Cremer et al., 2016). Because of the dependence of DHE conversion on the number of living cells within a biofilm, fluorescence values were normalized for the amount of viable cells/biofilm as assessed by CFU plating (De Brucker et al., 2013; De Cremer et al., 2016; Tits et al., 2020). Treatment of C. albicans biofilms with MCZ-DB resulted in significantly increased superoxide accumulation as compared to both compounds alone and the DMSO control treatment (Figure 4). This increase in superoxide accumulation, relative to each compound alone, was also observed when treating C. glabrata biofilms with the combination (Figure 5). CFU determination of the treated biofilms indicated that the MCZ-DB combination was fungicidal against biofilms of C. albicans (Figure 4) as well as against those of C. glabrata (Figure 5), which is according to our previous data (Tits et al., 2020).

To determine whether the fungicidal activity of MCZ-DB resulted from the observed elevated superoxide levels upon combination treatment, C. albicans biofilms were treated with MCZ, DB or a combination in the presence of the antioxidant ascorbic acid, and stained with DHE. As shown in Figure 4, the presence of the antioxidant abolished the accumulation of superoxide upon treatment with the MCZ-DB combination (Figure 4), as well as its fungicidal activity. Hence, these data suggest a causal link between the accumulation of ROS in C. albicans biofilms treated with MCZ-DB and the combination's fungicidal activity against C. albicans biofilm cells.