Antifungal exposure can enhance Candida glabrata pathogenesis

21 22 Azole antifungal drugs directly inhibit lanosterol 14-ɑ-demethylase and indirectly affect the 23 expression of metabolic, transmembrane transporter, and cell wall organization genes in fungal 24 pathogens. It is not known how these indirect azole effects depend on dose, timing, and specific 25 azole used, or how they influence host interactions. Candida glabrata (recently renamed 26 Nakaseomyces glabratus) is the second leading cause of candidiasis, and clinical strains have 27 high rates of intrinsic resistance to azoles. We investigated the early responses of reference 28 strains BG2 and CBS138 to sub-inhibitory doses of fluconazole and voriconazole, and 29 particularly, how these responses affect host-pathogen interactions. Cell wall profiling and 30 transcriptomic data revealed highly similar responses for each strain to both azoles, including 31 the upregulation of several virulence factors, such as yapsins. We also observed significant 32 increases in CBS138 survival in macrophages and increased virulence in Galleria mellonella 33 after voriconazole exposure. Using a combination of pharmacological inhibition of calcium ion 34 channels and deletion strains, we determined that voriconazole-enhanced virulence requires a 35 yapsin protease, YPS1, and is regulated via the calcineurin pathway and the cell wall integrity 36 pathway, both of which regulate YPS1 expression. We also observed that voriconazole 37 treatment significantly reduced the virulence of the bck1Δ strain in G. mellonella, suggesting 38 that inhibitors of the cell wall integrity pathway might potentiate azole activity by improving 39 susceptibility to host killing. Our study provides new insight into short-term azole adaptation in 40 C. glabrata, and importantly demonstrates that sub-inhibitory azole exposure can induce 41 virulence factors and alter fungal pathogenesis. 42 43 Article summary 44 Antifungal drugs indirectly affect essential fungal cell processes, but we lack an understanding 45 of how drug-induced changes affect fungal pathogenesis. We investigated how Candida 46 glabrata adapts when exposed to azole drugs in terms of cell wall and transcriptional changes. 47

strains had similar transcriptional changes in response to azoles, but azole-treated 48 CBS138 survived better in immune cells and caused more host death than untreated cells, 49 suggesting that short-term azole treatment can significantly affect pathogenesis. Voriconazole-50 enhanced disease requires the calcineurin and cell wall integrity pathways and the virulence 51 factor, YPS1, but could be blocked by a calcium ion channel inhibitor. 52 53 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 3 54

55 56 There is growing awareness and understanding of the importance of antifungal drug resistance 57 and related phenomena in fungal pathogens, including tolerance, persistence and 58 heteroresistance(Amich et al., 2025; Yang & Berman, 2024). These phenotypes have 59 concerning implications for clinical disease management and treatment failure, which has driven 60 investigations to understand the molecular mechanisms that permit pathogenic fungi to adapt to 61 antifungals. Resistance is classically linked to stable genetic mutations that allow survival in 62 high drug concentrations (Marie & White, 2009). However, other adaptive mechanisms, like 63 tolerance and heteroresistance, have proven more difficult to characterize due to their 64 transience within the population and lack of a clear, causal link to genetic modifications (Berman 65 & Krysan, 2020; Rosenberg et al., 2018; Yang & Berman, 2024). Further, while antifungal 66 resistance mutations are known to affect Candida species cell fitness and virulence (Bohner et 67 al., 2022), we have a poor understanding of how other antifungal adaptive processes influence 68 fitness and host-pathogen interactions. 69 70 Candida glabrata (recently renamed Nakaseomyces glabratus) is a major human fungal 71 pathogen and the second leading cause of systemic candidiasis. C. glabrata is categorised as a 72 high priority pathogen by the World Health Organisation (WHO, 2022) due to its serious clinical 73 burden and high rates of antifungal resistance. Consistent with this classification, a recent 74 Public Health England (PHE) surveillance study reported 17% and 21% of C. glabrata 75 bloodstream clinical isolates as resistant to fluconazole and voriconazole, respectively (Budd et 76 al., 2023). In comparison, only 1% of clinical isolates of the leading cause of candidiasis, 77 Candida albicans, were resistant to fluconazole or voriconazole in the same study (Budd et al., 78 2023). Despite these high levels of azole resistance, fluconazole or voriconazole are sometimes 79 still prescribed to patients with suspected fungaemias (Helmstetter et al., 2022). 80 81 Azoles directly inhibit lanosterol 14-ɑ-demethylase (encoded by ERG11 in Candida species) 82 leading to reduced ergosterol production, toxic sterol intermediate accumulation, and altered 83 membrane fluidity. These processes also indirectly affect the expression of carbohydrate 84 metabolism, transmembrane and ion transporters, and cell wall organization genes (Ribeiro et 85 al., 2022). Little is known about the early cellular adaptations that pave the way for C. glabrata 86 survival and drug resistance development in the host. We expect, based on existing 87 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 4 transcriptomics and proteomics datasets, that the indirect effects of azole exposure on cell wall 88 organisation and other biological processes alters host interactions. However, the effects of 89 azole treatment on C. glabrata host-pathogen interactions are unclear. 90 91 In this study, we investigated the early adaptative responses of two C. glabrata reference 92 strains, BG2 and CBS138, to sub-inhibitory doses of fluconazole and voriconazole, and 93 particularly, how these responses affect host-pathogen interactions. We expected, based on our 94 previous review of -omics datasets and the similarities in both drug class and reported 95 resistance rates by PHE, that voriconazole and fluconazole might exert similar effects on cell 96 wall remodeling and host-pathogen interactions (Budd et al., 2023; Ribeiro et al., 2022). While 97 transcriptional and cell wall profiling data highlighted similar responses for each strain to both 98 azoles, we unexpectedly observed that CBS138 survival in macrophages and virulence in 99 Galleria mellonella infection studies was significantly improved after voriconazole exposure. 100 Fluconazole pre-exposure also mildly enhanced both BG2 and CBS138 virulence in 101 G. mellonella. We further demonstrated that voriconazole-enhanced virulence in CBS138 is at 102 least partially dependent on the virulence factor and yapsin, YPS1, and the pathways required 103 for YPS1 expression, including calcium ion channel signaling, the calcineurin pathway, and the 104 Slt2-MAPK (PKC) cell wall integrity pathway. Importantly, our study demonstrates how short-105 term adaptation to antifungals can induce survival strategies that enhance fungal pathogenesis. 106 107

108 109 Strains and Growth Conditions 110 C. glabrata reference strains BG2 and CBS138 (ATCC2001) (Cormack & Falkow, 1999; Dujon 111 et al., 2004; Koszul et al., 2003; Schwarzmuller et al., 2014) were maintained by sub cultivation 112 on YPD plates (2% glucose, 2% bactopeptone, 1% yeast extract) at 37°C from a frozen stock (-113 80°C). 114 115 Before each experiment, yeast cells were conditioned overnight in 5 or 25 mL MOPS-buffered 116 liquid RPMI-1640 medium (final concentration: 2% glucose, MOPS 0.165 mol/L, pH 7) (Sigma 117 R6504) at 37°C, 200 rpm. Yeast cells were back-diluted from overnight cultures (1x106 cells/mL) 118 for a further 4 hours growth in 10, 50 or 100 mL RPMI-1640 with antifungals (MIC50 119 concentrations, Table 1), 50 µg/mL verapamil (Sigma), or DMSO (Sigma) only according to 120 experimental requirements. 121 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 5 122 Minimum Inhibitory Concentration (MIC) Determination 123 MIC testing was performed according to EUCAST guidelines ((The European Committee on 124 Antimicrobial Susceptibility Testing, 2020). Briefly, C. glabrata yeast cells were grown overnight 125 in MOPS-buffered RPMI-1640 at 37°C. Cells were then centrifuged for 5 minutes, 7000 rpm, 126 and the pellet was resuspended in RPMI-1640. 1x105 yeast cells were added to each well in a 127 96 well plate in 90 μL RPMI-1640 and 10 μL of the respective drug dilution. Drug test ranges 128 were 0.125-64 mg/L for fluconazole and 0.0156-8 mg/L for voriconazole. Plates were incubated 129 for 24 hours at 37°C in the dark. The plates were then read on a spectrophotometer (VersaMax, 130 SoftMax® Pro 7 Software), OD530, and the MIC50 and MIC80 for each strain and drug 131 combinations were determined as the lowest concentration of drug needed to inhibit 50% or 132 80%, respectively, of cell growth. 133 134 Flow Cytometry 135 To analyse cell wall carbohydrate exposure, cells grown with and without antifungals were 136 inactivated overnight in 50 mM thimerosal (Sigma). Cells were then washed three times with 137 PBS and counted by haemocytometer. 2.5x106 cells were stained with 0.5 µg/mL Fc-Dectin-1 138 (kindly provided by Gordon Brown, MRC-CMM) and 1:200 diluted goat anti-human IgG antibody 139 conjugated to Alexa Fluor 488 (Invitrogen), 50 µg/mL Wheat Germ Agglutinin (WGA) conjugated 140 to Alexa Fluor 680 (Invitrogen), and 25 µg/mL Concanavalin A (ConA) conjugated to Texas Red 141 (Invitrogen). Data were acquired for a minimum of 20,000 events on the Attune NxT (Thermo 142 Fisher) and analysed using FlowJo v10 software (TreeStar Inc.) and gated as previously 143 described (Ribeiro et al., 2025). 144 145 Transmission Electron Microscopy (TEM) 146 Yeast cells were grown overnight in 5 mL RPMI-1640 (37°C, 200 rpm), counted, and 1x108 cells 147 were back-diluted and grown for a further 4 hours in 100 mL RPMI-1640 at 37°C, 200 rpm, 148 containing voriconazole (MIC50) or DMSO (solvent control). Cells were then centrifuged at 4,000 149 rpm for 5 minutes. The concentrated pellet was placed between the sides of a small copper 150 holder, enough to fill up the required space. Cells were then frozen in a high-pressure freezer 151 and rapid transport system (Leica EMPACT2). Freeze substitution was carried out following the 152 program detailed in Supplemental Table 2. Samples were then removed and placed in 10% 153 Spurr’s (TAAB):Acetone for 72 hours – 30% Spurr’s overnight; 50% Spurr’s for 8 hours; 70% 154 Spurr’s overnight; 90% Spurr’s for 8 hours. Subsequently, samples were embedded in Spurr’s 155 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 6 resin at 60°C for at least 24 hours. Then 90nm sections were prepared using a diamond knife 156 (Diatome Ltd, Switzerland) onto copper grids (EMResolutions) using a Leica UC6; and with 157 Uranyless (TAAB) and Lead Citrate in a Leica AC20. Samples were viewed on the 158 Transmission Electron Microscope JEM 1400 plus (JEOL) and captured using an AMT 159 UltraVUE camera (AMT). Image J (Fiji) was used to measure the thickness of the inner (chitin 160 and glucan) and outer (mannan) cell wall of 19-30 cells/group and 10-13 measurements/cell. 161 162 BMDM Challenge 163 Bone Marrow-Derived Macrophages (BMDMs) were isolated from the femurs and tibias of male 164 12-weeks old C57BL/6 mice as previously described (Davies & Gordon, 2005; Gonçalves & 165 Mosser, 2015). Mice were a kind gift from Gordon Brown and were randomly selected from in-166 house breeding colonies housed under specific-pathogen-free conditions at University of 167 Aberdeen. Mice were not subjected to any regulated procedures prior to cervical dislocation and 168 femur removal in accordance with ethical regulations approved by the University of Aberdeen 169 Animal Welfare and Ethical Review body and the ARRIVE guidelines. BMDMs were maintained 170 and differentiated in Dulbecco’s modified Eagle’s medium (DMEM; Sigma) supplemented with 171 10% heat inactivated Fetal Calf Serum (Gicbo), 15% L929 cell conditioned medium, 1% L-172 glutamine (Sigma), and 1% Penicillin/Streptomycin (Sigma). For macrophage interaction 173 studies, 3x104 BMDMs were plated on flat bottom 96-well plates and incubated overnight at 174 37°C, 5% CO2. BG2 and CBS138 wild-type cells were grown overnight in RPMI-1640, counted 175 and 1x107 cells were grown for a further 4 hours with MIC50 fluconazole, voriconazole, or <1% 176 DMSO in 10 mL RPMI-1640 (2% Glucose, pH 7) at 37°C. Yeast cells were then washed with 177 PBS and added to the 96-well plates in duplicate wells at a multiplicity of infection (MOI) of 3:1 178 (yeast cells to macrophages). Two hours post challenge the supernatant was removed, each 179 well was washed with DMEM, and new media was added to remove unengulfed yeast cells. For 180 the 2-hour timepoint 100 μL of 0.02% chilled SDS (Sodium Dodecyl Sulphate, Melford) was 181 added to each well, its contents were scraped, serially diluted, spotted on YPD agar plates and 182 incubated at 37°C. The same BMDM lysis and yeast recovery procedure was performed after 183 24 hours co-culture to determine CFU/mL and fold change in yeast cell recovery. 184 185 Galleria mellonella Infection, Survival and Melanization 186 G. mellonella larvae were purchased from Livefood UK Ltd. (Axbridge, UK) and stored in wood 187 shavings in the dark at room temperature prior to infection. C. glabrata yeast cells were grown 188 overnight in 6 mL RPMI-1640 at 37°C, 200 rpm, and back-diluted to 1x108 yeast cells in 100 mL 189 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 7 RPMI-1640 with the indicated antifungals at MIC50 concentration or equivalent volume solvent, 190 for a further 4 hours growth at 37°C, 200 rpm. Cells were washed and resuspended in sterile 191 PBS. Larvae (~250 mg weight) were randomly allocated into groups (specific sample sizes 192 indicated on figure legends) and infected in the last left proleg with 5x106 cells in a 50 μL/larvae 193 suspension using a U-100 30G Micro-fine syringe (BD). Control groups were injected with 50 μL 194 sterile saline only. Larvae were incubated at 37°C in the dark and survival and melanisation 195 were assessed daily for a period of 6 days (144 hours). Larvae were scored for melanization as 196 described previously (Usher et al., 2023). Briefly, larvae were considered partially melanised 197 when their natural colour had been visibly altered, however they still did not present a fully 198 darkened body. Larvae were considered fully melanised when their colour had been completely 199 altered to a dark grey/brown pigmentation. 200 201 Statistical Analyses 202 Statistical analyses were performed using GraphPad Prism v5.0 software (GraphPad Software) 203 and IBM SPSS Statistics v27.0 (IBM Corp.). Specific experimental analyses described on figure 204 legends. Macrophage-yeast survival and flow cytometry were analysed by Two-way ANOVA 205 with Dunnett’s multiple comparisons test. TEM measurements were analysed by Two-Way 206 ANOVA with Sidak’s multiple comparisons test. G. mellonella survival was analysed by Kaplan-207 Meier, Log-Rank pairwise over strata. A p value of <0.05 was considered to be significant, and 208 the results are shown as mean ± standard error of the mean (SEM). 209 210 RNA sequencing 211 Strains were streaked to single colonies on YPD agar plates for 2 days, then a single colony 212 inoculated for each biological replicate into 5 mL RPMI-1640 with 2% glucose (Sigma R6504) 213 and grown overnight. The next day, 108 cells were pelleted and inoculated into 50 mL of RPMI-214 1640 media with 2% glucose pre-warmed to 37°C with drug or DMSO (mock) treatment. For 215 drug addition, stock solutions of VCZ or fluconazole were prepared in DMSO, and the solution 216 mixed with RPMI media immediately before inoculation. Antifungals were added at MIC50 217 concentration: VCZ at 0.25 ug/mL or 0.125 ug/mL for BG2 and CBS138, respectively; FCZ at 16 218 ug/mL or 8 ug/mL for BG2 and CBS138, respectively. Mock-treatment was performed using 219 0.004% DMSO, and DMSO was added to all VCZ and FCZ treatments to the final concentration 220 of 0.004%. 221 222 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 8 After inoculation, cells were grown for 4 hours at 37°C with 200 rpm shaking, and harvested by 223 pelleting cells for 3 min at 4000 rpm, the supernatant was decanted, and the cells were pelleted 224 for 2 min at 3000 g to remove all supernatant, and the pellet frozen in liquid nitrogen, and stored 225 at -80°C. Three biological replicates were prepared on successive days. 226 227 RNA was extracted using a modified silica column protocol following bead-beating with zirconia 228 beads. The yeast pellets were thawed briefly on ice, then transferred to a screw cap tube and 229 200 μL of zirconia beads were added. 400 μL of RNA binding buffer (R1013, Zymo Research) 230 was added, and the mixture was kept on ice for 1 minute. The tubes were transferred to 231 PreCellys homogenizer (Bertin Technologies) and lysed using the following protocol: 10 232 seconds vortexing, followed by 10 seconds of waiting, repeated 3 times. The cells were then 233 transferred to ice for 1 minute. Vortexing and incubation on ice were repeated a total of 6 times. 234 The tubes were then centrifuged at 12,000 × g for 2 minutes. The supernatant was transferred 235 to a Zymo Spin IIICG column (C1006, Zymo Research) and centrifuged at 12,000 g for 1 236 minute. 400 μL of ethanol was added to the flow through, mixed, transferred to a Zymo Spin IIC 237 column (C1011, Zymo Research) and centrifuged for 1 minute. The column was washed with 238 DNA/RNA Prep buffer (D7010-2, Zymo Research) and centrifuged for 1 minute, and then 239 washed with DNA/RNA Wash buffer (D7010-3, Zymo Research) and centrifuged for 1 minute 240 twice. The column was then transferred to a clean 1.5 mL tube, and RNA was eluted by adding 241 30 μL of water and centrifuging at 10,000 × g for 1 minute. The concentration of the samples 242 was measured using Denovix. The quality and integrity of RNA was assessed on Fragment 243 Analyzer (Agilent) using High Sensitivity RNA Kit (DNF-472-1000, Agilent). RQN and 28S/18S 244 ratio were used to determine the quality of the sample. 245 246 RNA sequencing libraries were prepared from 500 ng total RNA using QuantSeq 3′ mRNA-Seq 247 V2 Library Prep Kit REV (Lexogen, Vienna, Austria), a method that sequences a single 248 fragment per mRNA, at the 3′ end proximal to the poly(A)-tail. Libraries were sequenced on 249 NextSeq 2000 (Illumina, San Diego, USA) 250 251 RNA-Seq FASTQ files were processed using a Nextflow pipeline for QuantSeq data which is 252 available online in GitHub (https://github.com/DimmestP/nextflow_paired_reads_pipeline). 253 Software versions used were (Nextflow 3.1, FastQC 0.12.1, Cutadapt 4.3, HISAT2 2.2.1, 254 SAMtools 1.17, MultiQC 1.14, BEDTools 2.30.0, Subread /FeatureCounts 3.11.3, DESeq2 255 1.40.2, Python 3.11.3, R 4.3.2, NCBI Datasets CLI 16.0.0) (Danecek et al., 2021; Di Tommaso 256 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 9 et al., 2017; Ewels et al., 2016; Kim et al., 2019; Liao et al., 2014; Love et al., 2014; Martin, 257 2011; O'Leary et al., 2024; Quinlan & Hall, 2010; Wingett & Andrews, 2018). The reads were 258 aligned to the genome sequence, for strain BG2: GCA_014217725.1; For CBS138, 259 GCF_000002545.3. For assigning 3′ fragments to mRNAs, we used stranded alignment with an 260 annotation file including 300nt added to the 3′ end of the annotated CDS. Differential gene 261 expression was performed using DESeq2 (Love et al., 2014), R (R Core Team, 2021) and 262 packages from the tidyverse (Wickham et al., 2019), and code is shared in GitHub 263 (https://github.com/ewallace/cglab_rnaseq/). Genes were called as differentially expressed if 264 they showed at least 2-fold difference at an adjusted p-value of 0.05 (5% false discovery rate), 265 unless otherwise stated. 266 267 268

269 270 Cell wall polysaccharide exposure is affected by sub-inhibitory azole treatment 271 We first investigated inhibitory concentrations of fluconazole (FCZ) and voriconazole (VCZ) for 272 C. glabrata with the aim of identifying concentrations that impose significant stress while 273 mimicking treatment failure (i.e. failure to completely inhibit or kill cells). We performed minimum 274 inhibitory concentration (MIC) testing in accordance with EUCAST guidelines (The European 275 Committee on Antimicrobial Susceptibility Testing, 2020). The concentration required to inhibit 276 50% of growth (MIC50) for CBS138 in FCZ and VCZ was 8 mg/L and 0.125 mg/L, respectively 277 (Table 1). In comparison, the MIC50 for BG2 was 16 mg/L FCZ and 0.25 mg/L VCZ (Table 1), 278 suggesting CBS138 is mildly more susceptible than BG2 to azole inhibition. This susceptibility 279 was more pronounced for the MIC80 (concentration of drug required to inhibit at least 80% of 280 growth compared to the control), where BG2 was four times more resistant to FCZ (64 mg/L 281 versus 16 mg/L) and twice as resistant to VCZ (4 mg/L versus 2 mg/L) compared to CBS138 282 (Table 1). 283 284 Previous studies demonstrated that antifungal treatment leads to differential cell wall gene 285 expression and significant changes in the cell wall that can have paradoxical effects on survival 286 in C. albicans mammalian infections (Hopke et al., 2018; Lee et al., 2012; Walker et al., 2008). 287 Therefore, we next tested whether short-term (4-hour) MIC50 antifungal exposure affected yeast 288 cell wall polysaccharide detection among the C. glabrata reference strains, BG2 and CBS138. 289 Cell wall features in BG2 were not majorly affected by FCZ or VCZ pre-treatment (Fig. 1a-c). In 290 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 10 CBS138, there was a minor increase in β- glucan (~1.21 and 1.27 fold-change for FCZ and 291 VCZ, respectively) and chitin (~1.23 and ~1.16 fold-change for FCZ and VCZ, respectively) 292 exposure levels in response to both azoles (Fig. 1a and 1c) compared to untreated cells. 293 Mannan exposure was also significantly higher for CBS138 compared to BG2 (~1.45 and ~1.48 294 fold-change for FCZ and VCZ, respectively) (Fig. 1b). 295 296 Cell wall layer measurements by TEM further show that BG2 yeast cells pre-exposed to VCZ 297 had slight differences in inner and outer wall thickness. BG2 cells had a slight, but significant, 298 increase in inner (~1.12 fold-change) layer thickness, but similar outer layer size (~1.05 fold-299 change) compared to its control (Fig. 1d and 1e). However, and in contrast to BG2, CBS138 300 cells pre-treated with VCZ showed significantly reduced inner (~0.6 fold change; p<0.05) but 301 similar outer (~1.03 fold-change) layer thickness compared to controls (Fig. 1d and 1f). 302 303 Taken together, our flow cytometry data showed minimal changes in β- glucan and chitin 304 exposure after azole treatment which was unexpected given the changes in inner layer 305 thickness by TEM. However, both azoles increased mannan exposure in CBS138 compared to 306 BG2 (~1.9-fold; Fig. 1b), and our TEM measurement data indicates that CBS138 cells generally 307 had a thicker outer cell wall layer compared to BG2 (Fig. 1d-f). 308 309 Voriconazole exposure enhances CBS138 pathogenesis 310 311 We observed above that short-term FCZ and VCZ exposure differentially impacted some 312 carbohydrate exposure and the gross cell wall architecture of the two C. glabrata reference 313 strains, CBS138 and BG2. Cell wall composition plays an important role in modulating host 314 responses, including fungal clearance by immune cells (Gow et al., 2017). Therefore, we next 315 tested how azole pre-exposure affected yeast survival in macrophages by measuring yeast 316 colony forming unit (CFU) recovery following macrophage challenge (Ribeiro et al., 2025). 317 318 As before, yeast cells were treated with or without MIC50 FCZ or VCZ prior to co-incubation with 319 bone marrow-derived macrophages. After 2 hours of yeast-macrophage challenge, we observed 320 no statistically significant differences in yeast recovery from macrophages between azole-321 treated and untreated groups for either strain, though there was a trend toward a greater 322 percentage recovery of the CBS138 inoculum from groups pre-exposed to azoles, especially 323 VCZ (Fig. 2a and 2b). After 24 hours of co-incubation, we still observed no significant 324 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 11 differences in yeast recovery between treatment groups for BG2 (Fig. 2c). However, at 24 hours 325 we recovered significantly higher CFUs of VCZ-treated CBS138 cells compared to FCZ-treated 326 cells or the control, suggesting that VCZ-treated CBS138 cells were able to replicate better 327 within macrophages than FCZ-treated and untreated yeast cells (Fig. 2c). As expected, based 328 on the CFUs recovered at each time point, the fold change in yeast recovery between 2 and 24 329 hours showed no variance for BG2 between groups and a trend of increased survival for VCZ-330 treated CBS138 yeast cells compared to FCZ-treated and untreated cells (Figure 2d). 331 332 We next tested whether azole pre-treatment affected C. glabrata virulence in the G. mellonella 333 systemic infection model. Consistent with our macrophage interaction data, we observed no 334 significant differences in G. mellonella survival during infection with azole-treated and untreated 335 BG2 yeast cells (Fig. 2e). For CBS138, infection with FCZ-treated yeast induced slightly faster 336 larval death than the control, but VCZ-treated yeast killed larvae significantly faster than 337 untreated cells (Fig. 2f; p40% between control and VCZ-338 exposed infection groups. 339 340 Altogether, our findings suggest that azole pre-treatment has minimal effects on BG2 host 341 interactions, but azoles, and especially VCZ, trigger enhanced survival and virulence in 342 CBS138. 343 344 Transcriptomic responses to azole drugs are broadly similar across strains 345 We designed an RNA-seq experiment to identify transcriptomic changes that might explain 346 differences in azole-enhanced virulence between BG2 and CBS138. As above, we treated yeast 347 for 4 hours with either FCZ or VCZ at MIC50 concentration or a mock-treated DMSO-only 348 control. We prepared 3 biological replicates and made libraries using a 3′ mRNA-Seq approach. 349 Extracted RNA was high-quality and the aligned reads passed all relevant quality checks, 350 including high correlations between replicate samples (Figs S1, S2). 351 352 Transcriptome profiles clustered both by strain and by drug treatment, as revealed by principal 353 component analysis of the regularized log-counts (Fig 3a). Comparing principal components 1 354 and 2 shows that differences between strain and drug are almost orthogonal (Fig 3a) and 355 contain almost 70% of the variance (Fig 3b). Transcriptome profiles from treatment by VCZ and 356 FCZ were very similar within each strain, both in the principal component plot (Fig 3a) and by 357 correlation analysis (Figs S1, S2). Differential gene expression analysis confirmed these 358 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 12 findings: hundreds of genes were significantly differentially expressed between strains in each 359 growth condition, and differentially expressed between drug-treated cells and mock-treated cells 360 for each strain (Fig S3). However, within-strain no significantly differentially expressed genes 361 were found between FCZ and VCZ (Fig S3). We conclude that as each strain received the same 362 “subjective dose” of VCZ and FCZ, at the strain-specific MIC50 concentration, the transcriptomic 363 responses to these two azoles were practically indistinguishable. 364 365 Thus, we break down the analysis into three main components: baseline differences between 366 strains (Fig. 3c), common drug-regulated transcripts in both strains (Fig. 3d), and transcripts that 367 were differentially induced in one strain compared to the other, i.e. drug-strain interactions (Fig. 368 3e). 369 370 The baseline differences between BG2 and CBS138 are extensive (Fig 3c). In the control 371 samples, 194 genes had significantly higher expression in CBS138 compared to 265 in BG2 372 (Fig. 3c). Genes with higher expression in CBS138 are enriched in GO categories including 373 those associated with cell wall assembly, cell-cell adhesion, and cell aggregation (i.e. GAS3, 374 SWM1, FKS3, EPA6, EPA3, ZAP1, KSS1, and several uncharacterized genes), and some 375 involved in amino acid biosynthetic processes including lysine and other amino acid 376 biosynthesis (i.e. LYS9, ARG1, IDP1, MET13, LYS12, LEU2, LYS21, STR3, and several 377 uncharacterized genes). Genes with higher expression in BG2 are enriched in a variety of 378 categories related to metabolism including trehalose metabolism (TPS2, ATH1, UGP1, 379 CAGL0H02387g, CAGL0K03421g), stress responses (including GCN4, MSN4, YHB1, TUP11, 380 NUC1, KRE29, TDH3, HSP12, SSA3, HSP78), and translation (including FRS2, TIF1, EFT2). 381 We did not find a clear picture here about how baseline transcriptomic differences between 382 strains could explain their different phenotypes, so focused on azole responses subsequently. 383 384 Common azole-regulated targets are extensive (Fig 3d) and consistent with previous datasets 385 (Ribeiro et al., 2022). The 258 significantly azole-induced upregulated genes are enriched in GO 386 terms such as lipid metabolism, organelle organization, response to chemical, and vesicle-387 mediated transport. Consistent with azoles targeting ergosterol production, ergosterol 388 biosynthesis pathway genes were induced in both strains including ERG1, ERG2, ERG3, 389 ERG5, ERG7, ERG11, ERG24, and ERG25 (Fig 4). Azole upregulates the multidrug resistance 390 transcription factor PDR1, along with target transporters involved in drug resistance, CDR1 and 391 PDH1, but not the homolog SNQ2 (Fig 4a). Several yapsins, proteases which are important 392 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 13 virulence factors that suppress host immune responses (Rasheed et al., 2018), were 393 upregulated in both strain backgrounds (Fig 4). The 355 significantly azole-downregulated 394 genes include genes involved in ribosomal biogenesis and translation, consistent with drug 395 treatment having a negative impact on growth. Full GO results are included in the online 396 supplementary data. 397 398 We expected that transcripts that are differentially induced by azoles in CBS138 compared to 399 BG2 might explain the increase in virulence in azole-treated CBS138. Surprisingly, very few 400 genes fall into this category: only 9 are more induced in CBS138 than BG2, and 22 vice versa, 401 at a false discovery rate of 5% and minimal 2-fold expression change (Fig 3e). The genes 402 induced more in CBS138 include the YAP6 transcription factor, that has roles in stress 403 responses (Merhej et al., 2016), and eight uncharacterized genes. The 22 genes induced more 404 in BG2 are largely associated with transport and metabolic processes (full list available in online 405 supplemental). 406 407 The calcineurin pathway and its transcription factor, CRZ1, are important regulators of azole 408 resistance in C. glabrata (Vu et al., 2023) and provide a critical stress response to combat 409 azole-mediated membrane disruption in C. albicans (Onyewu et al., 2004). In our datasets, we 410 observed differential expression of CRZ1-dependent genes, including induction of the 411 calcineurin negative feedback regulator, RCN2, in response to azole treatment. Given the 412 diverse roles calcineurin plays in cell wall maintenance, stress responses and host interaction, 413 we hypothesized that the enhanced virulence of azole-treated CBS138 requires calcineurin 414 activity and induction of CRZ1-dependent targets, like yapsins. 415 416 Calcium ion channel inhibition suppresses voriconazole-enhanced virulence 417 418 The calcineurin pathway is typically known for its role in calcium signaling, and blocking calcium 419 channels alongside azole treatment synergistically inhibits the growth of drug-resistant 420 C. albicans strains (Liu et al., 2016). We therefore tested the importance of calcium for azole-421 enhanced virulence using the drug verapamil to inhibit calcium-importing ion channels (Fig. 5a) 422 (Teng et al., 2008; Yu, Q. et al., 2013). 423 424 CBS138 and BG2 cells were untreated or treated for 4 hours with 50 µg/mL verapamil, MIC50 425 VCZ, or a combination of both verapamil and VCZ prior to infecting G. mellonella (Fig. 5). 426 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 14 Similar to our earlier observations (Fig. 2e), there were no significant differences in virulence for 427 larvae infected with either VCZ-treated or untreated BG2 cells (Fig. 5b). Verapamil treatment 428 alone also did not significantly alter BG2 virulence compared to untreated yeast. BG2 cells pre-429 treated with the combination of verapamil and VCZ resulted in slightly better overall larval 430 survival (~40% survival versus 25% for the control), but this was not significant compared to 431 infection with untreated cells. 432 433 We again observed significantly enhanced virulence for CBS138 cells pre-treated with VCZ 434 compared to untreated controls (p<0.01, Fig. 5c) Treatment with verapamil alone did not 435 significantly alter CBS138 virulence (Fig. 5c). However, the combination of verapamil and VCZ 436 rescued larval survival back to control levels (Fig. 5c), suggesting that calcium ion channels are 437 required for voriconazole-enhanced virulence. Indirectly, azoles may be inadvertently triggering 438 calcium signaling and other pathways in a way which promotes CBS138 virulence. 439 440 Cell wall integrity, calcineurin and YPS1 are necessary for voriconazole-enhanced 441 virulence in CBS138 442 Our verapamil study suggested that target genes downstream of the calcineurin pathway may 443 be important for voriconazole-enhanced virulence in CBS138. One target of this pathway that 444 was upregulated in our RNA-Seq dataset is the yapsin, YPS1, which requires both calcineurin 445 and cell wall integrity pathway (Slt2-MAPK) signaling for its expression (Fig. 6a) (Miyazaki et al., 446 2011). We used available mutants in a published gene deletion collection (Schwarzmuller et al., 447 2014) to test our hypothesis that calcineurin, the cell wall integrity pathway, and virulence factor 448 YPS1 contribute to voriconazole-enhanced virulence. 449 450 cna1Δ, cnb1Δ, crz1Δ, bck1Δ and slt2Δ were grown for 4 hours with or without VCZ at MIC50 451 concentration prior to infecting G. mellonella (Fig. 6b-f). Larvae infected with either treatment 452 group for cna1Δ, cnb1Δ, crz1Δ and slt2Δ showed no significant differences in survival at 168 453 hours post-infection. The only slight, but still statistically insignificant difference was for crz1Δ 454 during early infection (18 hours), where we observed reduced survival for larvae infected with 455 untreated cells compared to VCZ-treated cells (~40% survival versus ~70% survival, 456 respectively; Fig. 6d). Surprisingly, VCZ-treated bck1Δ cells were attenuated for virulence 457 compared to untreated cells. Untreated bck1Δ cells killed all larvae within 144 hours while 458 infection with VCZ-treated cells resulted in ~40% survival (Fig. 6f). Overall, these data support 459 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 15 our hypothesis that Slt2-MAPK and the calcineurin pathway are necessary for the CBS138 460 voriconazole-enhanced virulence phenotype. 461 462 We next tested yps1Δ, which we grew with and without VCZ for 4 hours prior to infecting 463 G. mellonella (Fig. 6g). Larvae infected with VCZ-treated or untreated yps1Δ cells died at similar 464 rates up to 48 hours post-infection. After 48 hours, larvae infected with VCZ-treated yps1Δ cells 465 died at a slower rate than larvae infected with untreated cells in a trend similar to the bck1Δ 466 strain. 467 468 Altogether, our data suggest that the voriconazole-enhanced virulence we observed for CBS138 469 requires both the cell wall integrity and calcineurin signaling pathways and their downstream co-470 regulated virulence factor, YPS1. 471 472

473 474 Azole treatment stalls fungal growth by directly inhibiting ergosterol biosynthesis, but also 475 indirectly affects cellular processes such as cell wall biogenesis (Ribeiro et al., 2022). We have 476 a poor understanding of how the direct and indirect effects of azole treatment contribute to 477 fungal fitness and persistence in the host. Most studies investigate antifungal responses using a 478 single reference strain or drug. This approach is perfectly reasonable given potential issues with 479 study feasibility and cost, but makes systematic comparisons between drugs, strains and 480 species difficult. In our study, we address some of these issues by exploring the early 481 adaptation of two C. glabrata reference strains to two azole drugs. 482 483 Published transcriptomics and proteomics datasets indicated that cell wall biogenesis processes 484 are differentially regulated by multiple antifungal drug classes, including azoles (Ribeiro et al., 485 2022). The cell wall is the first point of contact between fungi and host cells, and cell wall 486 carbohydrates are an important mediator of host innate immune responses. However, we 487 observed few changes in cell wall carbohydrate exposure after 4 hours of FCZ or VCZ treatment 488 (Fig 1a-c), though we did observe differences in inner cell wall thickness in VCZ-treated cells 489 compared to controls and in CBS138 mannan exposure in response to both drugs (Fig 1d-f). 490 Our flow cytometry probes include lectins specific for the inner cell wall carbohydrates chitin and 491 β-1,3-glucan (Allen et al., 1973; Palma et al., 2006), therefore it is possible that changes in the 492 inner cell wall architecture are related to β-1,6-glucan abundance, which we are currently unable 493 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 16 to detect. The differences in cell wall thickness and CBS138 mannan exposure also suggest 494 potential changes in cell wall protein abundance, though there are limited data available 495 characterizing the fungal proteome in response to antifungal treatment (Pais et al., 2016; 496 Ribeiro et al., 2022). Pais et al. found that clotrimazole treatment altered the abundance of 37 497 C. glabrata membrane proteins; however, 25 of these proteins had decreased abundance. In 498 C. albicans, ketoconazole treatment increased the abundance of 32 proteins, but the protein 499 isolation procedure was not specific to the cell wall (Hoehamer et al., 2010). Our RNA-Seq 500 analysis shows evidence for differential regulation of cell wall-associated genes (Fig. 4a), but no 501 clear bias towards mechanisms that would be consistent with our flow cytometry profiling, such 502 as general upregulation of genes encoding mannoproteins. 503 504 Azole exposure and associated minor cell wall changes did not appear to affect host 505 interactions for strain BG2. Azole-treated BG2 cells had no major differences in inoculum uptake 506 by macrophages, intracellular survival, or virulence in G. mellonella versus untreated controls. 507 However, we observed that sub-inhibitory azole treatment altered CBS138-host interactions. 508 FCZ-treated cells had a mild trend towards decreased intracellular replication over 24 h and 509 slightly increased virulence in G. mellonella compared to control cells. VCZ-treated cells had 510 higher intracellular inoculum recovery at 2 h post-macrophage co-incubation, were recovered at 511 significantly higher CFU after 24 h co-incubation with macrophages and were significantly more 512 virulent in G. mellonella than control cells. While we do not yet know the reason behind these 513 drug-specific effects in CBS138, others have shown that azoles can localize to multiple 514 subcellular compartments, including the mitochondria (Benhamou et al., 2017; Elias et al., 2019; 515 Koren et al., 2024), and may have off-target effects on heme production or other processes 516 which could explain the differences in virulence phenotypes between drugs. 517 518 We performed RNA-Seq to determine what transcriptional changes might drive the strain- and 519 drug-dependent differences we observed in C. glabrata virulence. We were surprised to 520 discover that both azoles induced nearly indistinguishable transcriptional differences for BG2 521 and CBS138, with only 31 genes showing altered differential expression between strains in 522 response to azoles. Most of these 31 genes are uncharacterized, and several genes 523 upregulated in BG2 during azole exposure are associated with metabolic processes such as 524 sterol uptake (TIR3), the tricarboxylic acid cycle (CAGL0L02079g, ACO2, CAGL0K11616g), and 525 amino acid biosynthesis for methionine (MET15), lysine (CAGL0J06402g, CAGL0K07788g, 526 LYS9, LYS12, LYS21) and arginine (ARG8, CAGL0I08987g). Notably, the azole-induced lysine 527 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 17 biosynthetic genes are homologous to S. cerevisiae enzymes that play a vital role in cellular 528 tolerance against oxidative stress (O'Doherty et al., 2014; Olin-Sandoval et al., 2019). Their 529 upregulation in BG2 suggests that cells mount an oxidative stress defense early during azole 530 treatment, which is consistent with studies demonstrating that FCZ and other azoles induce 531 reactive oxygen species (ROS) in multiple Candida species (Gonzalez-Jimenez et al., 2023; 532 Mahl et al., 2015). While azole-induction of the lysine biosynthetic pathway appeared to be 533 BG2-specific, it is important to note that 4 out of 5 of these genes already had significantly 534 elevated baseline expression in CBS138 under control conditions compared to BG2 (online 535 supplemental). 536 537 Our RNA-Seq data also indicated that azoles induced the expression of several virulence 538 factors, particularly the yapsins, in both reference strains. Yapsins are 539 glycosylphosphatidylinositol (GPI)-linked aspartyl proteases that participate in cell wall 540 remodeling, immune evasion, and virulence. YPS1 and YPS5 expression is regulated by CRZ1 541 via the calcineurin pathway (Chen et al., 2012), and YPS1 expression requires additional input 542 from the cell wall integrity (Slt2-MAPK) signaling pathway (Miyazaki et al., 2011). The 543 calcineurin and cell wall integrity pathways play important roles in drug and stress tolerance 544 across fungal pathogens. In particular, the calcineurin pathway is required for azole and 545 caspofungin tolerance (Yu, S. et al., 2015), thermotolerance (Chen et al., 2012), and maximizes 546 C. glabrata survival in response to micafungin and manogepix treatment (Pavesic et al., 2024). 547 We observed that several C. glabrata CRZ1-regulated targets were induced by both FCZ and 548 VCZ, including the calcineurin negative feedback regulator, RCN2 (Chen et al., 2012). The loss 549 of VCZ-enhanced virulence for genetic mutants in the calcineurin pathway as well as after 550 pharmacological inhibition of calcium ion channels using verapamil (Scorzoni et al., 2020; Teng 551 et al., 2008) strongly suggest that the calcineurin pathway is an essential component of the 552 CBS138 VCZ-enhanced virulence phenotype. Further, genes in the cell wall integrity pathway 553 and YPS1 were also necessary for CBS138 VCZ-enhanced virulence in G. mellonella. 554 Unexpectedly, VCZ-treated bck1Δ was significantly less virulent than control cells, and VCZ-555 treated yps1Δ also had reduced virulence compared to the control, though this was not 556 statistically significant. This virulence attenuation is not likely due to differences in drug cidality 557 because we obtained similar CFU compared to the controls when inocula were plated to verify 558 cell counts (data not shown). However, this virulence defect was not observed with slt2Δ. The 559 VCZ-induced virulence defect for bck1Δ and yps1Δ suggests that BCK1 and YPS1 contribute to 560 compensatory processes that are necessary for C. glabrata azole adaptation independently of 561 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 18 SLT2. Further, previous work has shown that YPS1 expression was not fully repressed in the 562 slt2Δ background in response to thermal stress (Miyazaki et al., 2011), suggesting that low 563 levels of YPS1 expression are sufficient to maintain virulence post-VCZ treatment in the slt2Δ 564 background. 565 566 In conclusion, our study provides fundamental insights into the baseline and azole-induced 567 differences of two key C. glabrata reference strains. We demonstrated how azole-exposure 568 alters CBS138 host interactions, which can be blocked by calcium ion channel inhibition (i.e. 569 verapamil) or deletion of key components in the calcineurin and cell wall integrity pathways. Our 570 study establishes that these pathways, in addition to contributing to cell survival and drug 571 resistance development, also contribute to maintaining virulence in the host after azole 572 treatment and offer the prospect that interfering with cell wall integrity signaling could potentiate 573 azole-induced fitness defects in the host. 574 575 576 577 Author contributions: G.F.R - experimental design, data acquisition and analysis, writing & 578 editing; W.D. - experimental design, data acquisition and analysis, funding acquisition, writing & 579 editing; L.T. - data acquisition and analysis, editing; E.W.J.W. - experimental design, data 580 analysis, funding acquisition, writing & editing; D.S.C. - experimental design, data acquisition 581 and analysis, funding acquisition, writing & editing 582 583 Data availability 584 RNA-seq data are available on Gene Expression Omnibus (GEO), accession number 585 GSE273379. Supplemental code and GO tables are available at: 586 https://github.com/ewallace/cglab_rnaseq/. 587 588 Acknowledgments 589 We are grateful to our colleagues at the University of Aberdeen Institute of Medical Sciences 590 core facilities and wish to acknowledge Andrea Holme and the Iain Fraser Cytometry Centre 591 and Debbie Wilkinson, Gillian Milne and Lucy Wight in the Microscopy and Histology Facility for 592 training and assistance with cytometry and microscopy. We thank our Aberdeen Fungal Group 593 colleagues, especially Donna MacCallum, for helpful discussions. We thank Wallace lab 594 members for helpful discussions. We thank Richard Clarke, Angie Fawkes, and Lee Murphy for 595 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 19 performing RNA-seq at the Genetics Core of the Edinburgh Wellcome Trust Clinical Research 596 Facility. We also thank Jane Usher at University of Exeter for helpful comments and discussion. 597 598 Study funding 599 The authors are supported by the following funding sources. L.T. and G.F.R. received a PhD 600 studentship from the University of Aberdeen and G.F.R. received the Elphinstone scholarship. 601 D.S.C. received funding from the Academy of Medical Sciences (SBF006\1128). W.D. was 602 funded by the Medical Research Council (grant number MR/N013166/1). E.W.J.W. received 603 funding from the Wellcome Trust (208779/Z/17/Z). 604 605 Conflict of interest 606 The authors declare no known conflicts of interest. 607 608

609 Allen, A. K., Neuberger, A., & Sharon, N. (1973). The purification, composition and specificity of 610 wheat-germ agglutinin. The Biochemical Journal, 131(1), 155–162. 10.1042/bj1310155 611 Amich, J., Bromley, M., Goldman, G. H., & Valero, C. (2025). Toward the consensus of 612 definitions for the phenomena of antifungal tolerance and persistence in filamentous fungi. 613 mBio, 16(4), e0347524–24. Epub 2025 Feb 25. 10.1128/mbio.03475-24 614 Benhamou, R. I., Bibi, M., Steinbuch, K. B., Engel, H., Levin, M., Roichman, Y., Berman, J., & 615 Fridman, M. (2017). Real-Time Imaging of the Azole Class of Antifungal Drugs in Live 616 Candida Cells. ACS Chemical Biology, 12(7), 1769–1777. 10.1021/acschembio.7b00339 617 Berman, J., & Krysan, D. J. (2020). Drug resistance and tolerance in fungi. Nature 618 Reviews.Microbiology, 18(6), 319–331. 10.1038/s41579-019-0322-2 619 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 20 Bohner, F., Papp, C., & Gacser, A. (2022). The effect of antifungal resistance development on 620 the virulence of Candida species. FEMS Yeast Research, 22(1), foac019. doi: 621 10.1093/femsyr/foac019. 10.1093/femsyr/foac019 622 Budd, E., Guy, R., Borman, A., Johnson, E., Manuel, R., & Brown, C. (2023). Laboratory 623 surveillance of fungaemia due to yeasts in England: 2023. Health Protection Report, 18(10) 624 Chen, Y., Konieczka, J. H., Springer, D. J., Bowen, S. E., Zhang, J., Silao, F. G. S., Bungay, A. 625 A. C., Bigol, U. G., Nicolas, M. G., Abraham, S. N., Thompson, D. A., Regev, A., & 626 Heitman, J. (2012). Convergent Evolution of Calcineurin Pathway Roles in 627 Thermotolerance and Virulence in Candida glabrata. G3 (Bethesda, Md.), 2(6), 675–691. 628 10.1534/g3.112.002279 629 Cormack, B. P., & Falkow, S. (1999). Efficient Homologous and Illegitimate Recombination in 630 the Opportunistic Yeast Pathogen Candida glabrata. Genetics, 151(3), 979–87. 631 10.1126/science.285.5427.578 632 Danecek, P., Bonfield, J. K., Liddle, J., Marshall, J., Ohan, V., Pollard, M. O., Whitwham, A., 633 Keane, T., McCarthy, S. A., Davies, R. M., & Li, H. (2021). Twelve years of SAMtools and 634 BCFtools. GigaScience, 10(2), giab008. doi: 10.1093/gigascience/giab008. 635 10.1093/gigascience/giab008 636 Davies, J. Q., & Gordon, S. (2005). Isolation and Culture of Murine Macrophages. Current 637 Protocols in Immunology (3rd ed., pp. 91–103)10.1385/1-59259-838-2:091 638 Di Tommaso, P., Chatzou, M., Floden, E. W., Barja, P. P., Palumbo, E., & Notredame, C. 639 (2017). Nextflow enables reproducible computational workflows. Nature Biotechnology, 640 35(4), 316–319. 10.1038/nbt.3820 641 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 21 Dujon, B., Sherman, D., Fischer, G., Durrens, P., Casaregola, S., Lafontaine, I., De Montigny, 642 J., Marck, C., Neuvéglise, C., Talla, E., Goffard, N., Frangeul, L., Aigle, M., Anthouard, V., 643 Babour, A., Barbe, V., Barnay, S., Blanchin, S., Beckerich, J., . . . Souciet, J. (2004). 644 Genome Evolution in Yeasts. Nature, 430(6995), 35–44. 10.1038/nature02579 645 Elias, R., Benhamou, R. I., Jaber, Q. Z., Dorot, O., Zada, S. L., Oved, K., Pichinuk, E., & 646 Fridman, M. (2019). Antifungal activity, mode of action variability, and subcellular 647 distribution of coumarin-based antifungal azoles. European Journal of Medicinal Chemistry, 648 179, 779–790. 10.1016/j.ejmech.2019.07.003 649 Ewels, P., Magnusson, M., Lundin, S., & Kaller, M. (2016). MultiQC: summarize analysis results 650 for multiple tools and samples in a single report. Bioinformatics (Oxford, England), 32(19), 651 3047–3048. 10.1093/bioinformatics/btw354 652 Gonçalves, R., & Mosser, D. M. (2015). The Isolation and Characterization of Murine 653 Macrophages. Current Protocols in Immunology, 111(1), 14.1.1–14.1.16. 654 10.1002/0471142735.IM1401S111 655 Gonzalez-Jimenez, I., Perlin, D. S., & Shor, E. (2023). Reactive oxidant species induced by 656 antifungal drugs: identity, origins, functions, and connection to stress-induced cell death. 657 Frontiers in Cellular and Infection Microbiology, 13, 1276406. 10.3389/fcimb.2023.1276406 658 Gow, N. A. R., Latge, J., & Munro, C. A. (2017). The Fungal Cell Wall: Structure, Biosynthesis, 659 and Function. Microbiology Spectrum, 5(3)10.1128/MICROBIOLSPEC.FUNK-0035-2016 660 Helmstetter, N., Chybowska, A. D., Delaney, C., Da Silva Dantas, A., Gifford, H., Wacker, T., 661 Munro, C., Warris, A., Jones, B., Cuomo, C. A., Wilson, D., Ramage, G., & Farrer, R. A. 662 (2022). Population genetics and microevolution of clinical Candida glabrata reveals 663 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 22 recombinant sequence types and hyper-variation within mitochondrial genomes, virulence 664 genes, and drug targets. Genetics, 221(1), iyac031. doi: 10.1093/genetics/iyac031. 665 10.1093/genetics/iyac031 666 Hoehamer, C. F., Cummings, E. D., Hilliard, G. M., & Rogers, P. D. (2010). Changes in the 667 proteome of Candida albicans in response to azole, polyene, and echinocandin antifungal 668 agents. Antimicrobial Agents and Chemotherapy, 54(5), 1655–1664. 10.1128/AAC.00756-669 09 670 Hopke, A., Brown, A. J. P., Hall, R. A., & Wheeler, R. T. (2018). Dynamic Fungal Cell Wall 671 Architecture in Stress Adaptation and Immune Evasion. Trends in Microbiology, 26(4), 672 284–295. 10.1016/J.TIM.2018.01.007 673 Kim, D., Paggi, J. M., Park, C., Bennett, C., & Salzberg, S. L. (2019). Graph-based genome 674 alignment and genotyping with HISAT2 and HISAT-genotype. Nature Biotechnology, 37(8), 675 907–915. 10.1038/s41587-019-0201-4 676 Koren, V., Ben-Zeev, E., Voronov, I., & Fridman, M. (2024). Chiral Fluorescent Antifungal Azole 677 Probes Detect Resistance, Uptake Dynamics, and Subcellular Distribution in Candida 678 Species. JACS Au, 4(8), 3157–3169. 10.1021/jacsau.4c00479 679 Koszul, R., Malpertuy, A., Frangeul, L., Bouchier, C., Wincker, P., Thierry, A., Duthoy, S., Ferris, 680 S., Hennequin, C., & Dujon, B. (2003). The Complete Mitochondrial Genome Sequence of 681 the Pathogenic Yeast Candida (Torulopsis) glabrata. FEBS Lett, 534(1-3), 39–48. 682 10.1016/S0014-5793(02)03749-3 683 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 23 Lee, K. K., Maccallum, D. M., Jacobsen, M. D., Walker, L. A., Odds, F. C., Gow, N. A., & Munro, 684 C. A. (2012). Elevated cell wall chitin in Candida albicans confers echinocandin resistance 685 in vivo. Antimicrobial Agents and Chemotherapy, 56(1), 208–217. 10.1128/AAC.00683-11 686 Liao, Y., Smyth, G. K., & Shi, W. (2014). featureCounts: an efficient general purpose program 687 for assigning sequence reads to genomic features. Bioinformatics (Oxford, England), 30(7), 688 923–930. 10.1093/bioinformatics/btt656 689 Liu, S., Yue, L., Gu, W., Li, X., Zhang, L., & Sun, S. (2016). Synergistic Effect of Fluconazole 690 and Calcium Channel Blockers against Resistant Candida albicans. PloS One, 11(3), 691 e0150859. 10.1371/journal.pone.0150859 692 Love, M. I., Huber, W., & Anders, S. (2014). Moderated estimation of fold change and 693 dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550–8. 694 10.1186/s13059-014-0550-8 695 Mahl, C. D., Behling, C. S., Hackenhaar, F. S., de Carvalho e Silva, M. N., Putti, J., Salomon, T. 696 B., Alves, S. H., Fuentefria, A., & Benfato, M. S. (2015). Induction of ROS generation by 697 fluconazole in Candida glabrata: activation of antioxidant enzymes and oxidative DNA 698 damage. Diagnostic Microbiology and Infectious Disease, 82(3), 203–208. 699 10.1016/j.diagmicrobio.2015.03.019 700 Marie, C., & White, T. C. (2009). Genetic Basis of Antifungal Drug Resistance. Current Fungal 701 Infection Reports, 3(3), 163–169. 10.1007/s12281-009-0021-y 702 Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing 703 reads. EMBnet.Journal, 17(1)https://doi.org/10.14806/ej.17.1.200 704 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 24 Merhej, J., Thiebaut, A., Blugeon, C., Pouch, J., Ali Chaouche, M. E. A., Camadro, J., Le Crom, 705 S., Lelandais, G., & Devaux, F. (2016). A Network of Paralogous Stress Response 706 Transcription Factors in the Human Pathogen Candida glabrata. Frontiers in Microbiology, 707 7, 645. 10.3389/fmicb.2016.00645 708 Miyazaki, T., Izumikawa, K., Yamauchi, S., Inamine, T., Nagayoshi, Y., Saijo, T., Seki, M., 709 Kakeya, H., Yamamoto, Y., Yanagihara, K., Miyazaki, Y., Yasuoka, A., & Kohno, S. (2011). 710 The glycosylphosphatidylinositol-linked aspartyl protease Yps1 is transcriptionally regulated 711 by the calcineurin-Crz1 and Slt2 MAPK pathways in Candida glabrata. FEMS Yeast 712 Research, 11(5), 449–456. 10.1111/j.1567-1364.2011.00734.x 713 O'Doherty, P. J., Lyons, V., Tun, N. M., Rogers, P. J., Bailey, T. D., & Wu, M. J. (2014). 714 Transcriptomic and biochemical evidence for the role of lysine biosynthesis against linoleic 715 acid hydroperoxide-induced stress in Saccharomyces cerevisiae. Free Radical Research, 716 48(12), 1454–1461. 10.3109/10715762.2014.961448 717 O'Leary, N. A., Cox, E., Holmes, J. B., Anderson, W. R., Falk, R., Hem, V., Tsuchiya, M. T. N., 718 Schuler, G. D., Zhang, X., Torcivia, J., Ketter, A., Breen, L., Cothran, J., Bajwa, H., Tinne, 719 J., Meric, P. A., Hlavina, W., & Schneider, V. A. (2024). Exploring and retrieving sequence 720 and metadata for species across the tree of life with NCBI Datasets. Scientific Data, 11(1), 721 732–y. 10.1038/s41597-024-03571-y 722 Olin-Sandoval, V., Yu, J. S. L., Miller-Fleming, L., Alam, M. T., Kamrad, S., Correia-Melo, C., 723 Haas, R., Segal, J., Pena Navarro, D. A., Herrera-Dominguez, L., Mendez-Lucio, O., 724 Vowinckel, J., Mulleder, M., & Ralser, M. (2019). Lysine harvesting is an antioxidant 725 strategy and triggers underground polyamine metabolism. Nature, 572(7768), 249–253. 726 10.1038/s41586-019-1442-6 727 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 25 Onyewu, C., Wormley, F. L. J., Perfect, J. R., & Heitman, J. (2004). The calcineurin target, Crz1, 728 functions in azole tolerance but is not required for virulence of Candida albicans. Infection 729 and Immunity, 72(12), 7330–7333. 10.1128/IAI.72.12.7330-7333.2004 730 Pais, P., Costa, C., Pires, C., Shimizu, K., Chibana, H., & Teixeira, M. C. (2016). Membrane 731 Proteome-Wide Response to the Antifungal Drug Clotrimazole in Candida glabrata: Role of 732 the Transcription Factor CgPdr1 and the Drug:H+ Antiporters CgTpo1_1 and CgTpo1_2. 733 Molecular & Cellular Proteomics : MCP, 15(1), 57–72. 10.1074/mcp.M114.045344 734 Palma, A. S., Feizi, T., Zhang, Y., Stoll, M. S., Lawson, A. M., Diaz-Rodriguez, E., Campanero-735 Rhodes, M. A., Costa, J., Gordon, S., Brown, G. D., & Chai, W. (2006). Ligands for the 736 beta-glucan receptor, Dectin-1, assigned using "designer" microarrays of oligosaccharide 737 probes (neoglycolipids) generated from glucan polysaccharides. The Journal of Biological 738 Chemistry, 281(9), 5771–5779. 10.1074/jbc.M511461200 739 Pavesic, M. W., Gale, A. N., Nickels, T. J., Harrington, A. A., Bussey, M., & Cunningham, K. W. 740 (2024). Calcineurin-dependent contributions to fitness in the opportunistic pathogen 741 Candida glabrata. mSphere, 9(1), e0055423–23. Epub 2024 Jan 3. 742 10.1128/msphere.00554-23 743 Quinlan, A. R., & Hall, I. M. (2010). BEDTools: a flexible suite of utilities for comparing genomic 744 features. Bioinformatics (Oxford, England), 26(6), 841–842. 10.1093/bioinformatics/btq033 745 R Core Team. (2021). R: A Language and Environment for Statistical Computing 746 Rasheed, M., Battu, A., & Kaur, R. (2018). Aspartyl proteases in Candida glabrata are required 747 for suppression of the host innate immune response. The Journal of Biological Chemistry, 748 293(17), 6410–6433. 10.1074/jbc.M117.813741 749 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 26 Ribeiro, G. F., Denes, E., Heaney, H., & Childers, D. S. (2022). What 'Omics can tell us about 750 antifungal adaptation. FEMS Yeast Research, 21(8), foab070. doi: 751 10.1093/femsyr/foab070. 10.1093/femsyr/foab070 752 Ribeiro, G. F., Priest, E. L., Heaney, H., Richardson, J. P., & Childers, D. S. (2025). Mannan is a 753 context-dependent shield that modifies virulence in Nakaseomyces glabratus. Virulence, 754 16(1)10.1080/21505594.2025.2491650 755 Rosenberg, A., Ene, I. V., Bibi, M., Zakin, S., Segal, E. S., Ziv, N., Dahan, A. M., Colombo, A. 756 L., Bennett, R. J., & Berman, J. (2018). Antifungal tolerance is a subpopulation effect 757 distinct from resistance and is associated with persistent candidemia. Nature 758 Communications, 9(1), 2470–x. 10.1038/s41467-018-04926-x 759 Schwarzmuller, T., Ma, B., Hiller, E., Istel, F., Tscherner, M., Brunke, S., Ames, L., Firon, A., 760 Green, B., Cabral, V., Marcet-Houben, M., Jacobsen, I. D., Quintin, J., Seider, K., Frohner, 761 I., Glaser, W., Jungwirth, H., Bachellier-Bassi, S., Chauvel, M., . . . Kuchler, K. (2014). 762 Systematic phenotyping of a large-scale Candida glabrata deletion collection reveals novel 763 antifungal tolerance genes. PLoS Pathogens, 10(6), e1004211. 764 10.1371/journal.ppat.1004211 765 Scorzoni, L., Menezes, R. T. D., Pereira, T. C., Oliveira, P. S., Ribeiro, F. D. C., Santos, E. L. D. 766 S., Fugisaki, L. R. O., Oliveira, L. D. D., & Amorim, J. B. O. (2020). Antifungal and anti-767 biofilm effect of the calcium channel blocker verapamil on non-albicans Candida species. 768 Anais Da Academia Brasileira De Ciencias, 92(4), e20200703. 10.1590/0001-769 3765202020200703 770 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 27 Teng, J., Goto, R., Iida, K., Kojima, I., & Iida, H. (2008). Ion-channel blocker sensitivity of 771 voltage-gated calcium-channel homologue Cch1 in Saccharomyces cerevisiae. 772 Microbiology (Reading, England), 154(Pt 12), 3775–3781. 10.1099/mic.0.2008/021089-0 773 The European Committee on Antimicrobial Susceptibility Testing. (2020). Method for the 774 determination of broth dilution minimum inhibitory concentrations of antifungal agents for 775 yeasts, EUCAST DEFINITIVE DOCUMENT E.DEF 7.3.2. Retrieved 7/10 2020, from 776 https://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/AFST/Files/EUCAST_E_777 Def_7.3.2_Yeast_testing_definitive_revised_2020.pdf 778 Usher, J., Ribeiro, G. F., & Childers, D. S. (2023). The Candida glabrata Parent Strain Trap: 779 How Phenotypic Diversity Affects Metabolic Fitness and Host Interactions. Microbiology 780 Spectrum, 11(1)10.1128/SPECTRUM.03724-22/SUPPL_FILE/SPECTRUM.03724-22-781 S0003.PDF 782 Vu, B. G., Simonicova, L., & Moye-Rowley, W. S. (2023). Calcineurin is required for Candida 783 glabrata Pdr1 transcriptional activation. mBio, 14(6), e0241623–23. Epub 2023 Nov 9. 784 10.1128/mbio.02416-23 785 Walker, L. A., Munro, C. A., de Bruijn, I., Lenardon, M. D., McKinnon, A., & Gow, N. A. R. 786 (2008). Stimulation of chitin synthesis rescues Candida albicans from echinocandins. PLoS 787 Pathogens, 4(4), e1000040. 10.1371/journal.ppat.1000040 788 WHO. (2022). WHO fungal priority pathogens list to guide research, development and public 789 health action. Geneva: World Health Organization, Licence: CC BY-NC-SA 3.0 IGO. 790 Wickham, H., Averick, M., Bryan, J., Chang, W., D'Agostino McGowan, L., Francois, R., 791 Grolemund, G., Hayes, A., Henry, L., Hester, J., Kuhn, M., Pedersen, T. L., Miller, E., 792 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 28 Bache, S. M., Muller, K., Ooms, J., Robinson, D., Seidel, D. P., Spinu, V., . . . Yutani, H. 793 (2019). Welcome to the Tidyverse. The Journal of Open Source Softaware, 794 4(43)10.21105/joss.01686. 795 Wingett, S. W., & Andrews, S. (2018). FastQ Screen: A tool for multi-genome mapping and 796 quality control. F1000Research, 7, 1338. 10.12688/f1000research.15931.2 797 Yang, F., & Berman, J. (2024). Beyond resistance: antifungal heteroresistance and antifungal 798 tolerance in fungal pathogens. Current Opinion in Microbiology, 78, 102439. 799 10.1016/j.mib.2024.102439 800 Yu, Q., Ding, X., Xu, N., Cheng, X., Qian, K., Zhang, B., Xing, L., & Li, M. (2013). In vitro activity 801 of verapamil alone and in combination with fluconazole or tunicamycin against Candida 802 albicans biofilms. International Journal of Antimicrobial Agents, 41(2), 179–182. 803 10.1016/j.ijantimicag.2012.10.009 804 Yu, S., Chang, Y., & Chen, Y. (2015). Calcineurin signaling: lessons from Candida species. 805 FEMS Yeast Research, 15(4), fov016. 10.1093/femsyr/fov016 806 807 808 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 29 Figure Legends 809 Figure 1. Pre-exposure to fluconazole and voriconazole impacts yeast cell wall 810 architecture for C. glabrata reference strains BG2 and CBS138. We analyzed Median 811 Fluorescence Intensities (MFI), based on Median Absolute Deviation of flow cytometry data, for 812 C. glabrata reference strains BG2 and CBS138, pre-exposed or not to MIC50 fluconazole (FCZ) 813 or voriconazole (VCZ), for β-glucan (Fc-Dectin-1) (a), mannan (Concanavalin A, ConA) (b) and 814 chitin (Wheat Germ Agglutinin, WGA) exposure (c). Data represents two independent 815 experiments, n = 4-6 biological replicates/group plotted as mean ± standard error of the mean 816 (SEM) and normalized to their respective controls (DMSO only). * p ≤ 0.05 between indicated 817 groups and their respective controls. Statistical analyses were done by Two-Way ANOVA with 818 Dunnett’s multiple comparisons test. (d) Transmission Electron Microscopy (TEM) comparison 819 of the cell wall of BG2 and CBS138. (e, f) TEM measurements of inner and outer cell wall 820 thickness for BG2 (e) and CBS138 (f), pre-exposed or not to FCZ or VCZ. Scale bars represent 821 100 nm. Arrow indicates separation of inner and outer layers for CBS138 pre-treated with VCZ. 822 n = 19-28 cells/group, 10-13 measurements/cell (maximum of 256 values plotted). Data plotted 823 as mean ± SEM. * p ≤ 0.05 between indicated groups. Statistical analyses were done by Two-824 Way ANOVA with Sidak’s multiple comparisons test. CT, control (DMSO only). 825 826 Figure 2. VCZ pre-exposure improves CBS138 yeast cell recovery after 24-hour 827 challenge with BMDMs and enhances virulence in G. mellonella. (a-d) BMDMs were 828 challenged in technical duplicate at an MOI of 3:1 C. glabrata cells to macrophages. Internalized 829 yeast cells at 2 hours post-challenge are presented as percent of initial inoculum (a) and 830 CFU/mL (b). (c) Internalized yeast cells were also determined at 24 hours post-challenge. (d) 831 The fold change of yeast survival was determined by the ratio of recovered cells at 24 hours vs 832 2 hours post-challenge. The mean and SEM are indicated by the line and whiskers on each plot. 833 Data represents six independent experiments, n = 4-8 biological replicates per group for 834 macrophage yeast survival, mean of technical replicates. Statistical analyses were done by 835 Two-Way ANOVA with Dunnett’s multiple comparisons test. * p ≤ 0.05 between indicated 836 groups. (e and f) G. mellonella larvae were injected with 5x106 BG2 (e) or 837 CBS138 (f) yeast cells that had been exposed to no treatment, FCZ or VCZ for 4 hours. Survival 838 was monitored for up to 144 hours post-infection. Data represents five independent experiments 839 n = 10-15 larvae per group per experiment. * p ≤ 0.05 between the indicated group versus 840 control. Statistical analyses were done by Kaplan-Meier. CT, DMSO only control. 841 842 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 30 Figure 3. Azole drugs induce a consistent transcriptomic response, overlaid on strain-843 dependent baseline gene expression. (a) Principal Component Analysis shows that RNA-seq 844 samples cluster by strain and by azole drug treatment. Principal components 1 and 2 of the 845 regularized logarithm of counts per gene are shown for all samples; see methods for details. (b) 846 Between-sample variance is concentrated in principal components 1 and 2, that panel a shows 847 cluster by strain and azole treatment. (c) There are extensive baseline gene expression 848 differences between strains BG2 and CBS138. (d) A consistent azole-dependent transcriptomic 849 response is identified by pooling azole-dependent differential expression across both FCZ and 850 VCZ drugs across both strains BG2 and CBS138. (e) There are minimal strain-dependent 851 differences in drug-induced gene expression, detected using the interaction term in a DESeq2 852 analysis with both factors (`design = ~ Strain * Drug`). See methods for details of differential 853 gene expression analysis across 3 biological replicates using DESeq2, and supplementary 854 figure S3 for additional pairwise differential expression plots. 855 856 857 Figure 4. Azole drugs induce differential expression of specific genetic pathways. (a) 858 Azoles induce expression of select multidrug transporters (PDR1 transcription factor, CDR1 859 transporter, PDH1 transporter, but not the SNQ2 transporter), along with multiple ergosterol 860 biosynthesis genes. Azoles also induce expression of the yapsin family of aspartyl proteases. 861 Azoles further induce differential expression of different cell wall genes. (b) Azoles induce 862 expression of multiple genes in the CRZ1 calcineurin-responsive transcription factor pathway, 863 and other key stress response genes. 864 865 866 Figure 5. Verapamil inhibits voriconazole-enhanced virulence. (a) Diagram of the 867 calcineurin pathway. (B-C) G. mellonella larvae were injected with 5x106 BG2 (b) or CBS138 868 HTL (c) yeast cells that had been pre-exposed to no treatment, MIC50 VCZ, 50 µg/mL verapamil, 869 or both VCZ and verapamil for 4 hours. Survival was monitored for up to 120 hours post-870 infection. Data represents three independent experiments n = 10 larvae per group per 871 experiment. ** p ≤ 0.01 between the VCZ group versus control and VCZ+Verapamil. Statistical 872 analyses were done by Kaplan-Meier. CT, DMSO only control. 873 874 875 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 31 Figure 6. YPS1, calcineurin pathway, and PKC pathway components are required for 876 voriconazole-enhanced virulence. (a) Diagram of calcineurin and MAPK pathways 877 coordinating YPS1 expression in C. glabrata. (b-g) G. mellonella larvae were injected with 5x106 878 cells of the indicated strain that were untreated or pre-treated with MIC50 VCZ for 4 hours. 879 Survival was monitored for up to 168 hours post-infection. Data represents two independent 880 experiments n = 10 larvae per group per experiment. * p ≤ 0.05 between the VCZ group versus 881 control. Statistical analyses were done by Kaplan-Meier. CT, DMSO only control. 882 883 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 32 Graphical Abstract .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 33 Figure 1 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 34 Figure 2 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 35 Figure 3 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 36 Figure 4 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 37 Figure 5 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 38 Figure 6 .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint 39 Table 1. Minimum inhibitory concentrations for indicated antifungals and strains determined by broth microdilution method. Fluconazole Voriconazole Strain MIC50 MIC80 MIC50 MIC80 BG2 16mg/L >64mg/L 0.25mg/L 4mg/L CBS138 8mg/L >16mg/L 0.125mg/L >2mg/L .CC-BY 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 31, 2025. ; https://doi.org/10.1101/2025.07.31.667834doi: bioRxiv preprint