Ergosterol-depleted clinical isolates of Nakaseomyces glabratus can develop multi-drug resistance without apparent fitness and virulence defects

39

Nakaseomyces glabratus (formerly Candida glabrata) is a leading cause 40 of invasive candidiasis and rapidly develops antifungal drug resistance during 41 treatment. An increasing number of clinical isolates shows reduced susceptibility to 42 echinocandins and azoles, leaving amphotericin B (AMB) as a last therapeutic option. 43 Resistance of N. glabratus to this drug is rare and its underlying mechanisms are still 44 not fully understood. Here, we describe two independent multidrug resistant (MDR) 45 bloodstream isolates displaying resistance to AMB and anidulafungin (ANF) as well as 46 a reduced susceptibility to azoles. 47

Whole-genome sequencing and sterol profiling were performed on nine 48 clinical N. glabratus isolates which were resistant to ANF and displayed resistance or 49 low susceptibility to fluconazole (FLU) and AMB. The transcriptional response of 50

strain CBS138 and an AMB R+ANFR isolate was analyzed by RNA -seq. 51 Furthermore, PDR1 was deleted in the latter isolate to examine its influence on efflux 52 pump gene expression. Additionally, fitness and virulence of the AMB R+ANFR isolate 53 were examined in growth assays and a Galleria mellonella infection model. 54

Loss of function mutations in the genes ERG3 and ERG4 is linked to 55 ergosterol depletion and AMB resistance. Ergosterol depletion also contributed to a 56 Pdr1-mediated up- regulation of ERG and ABC transporter genes which was 57 associated with low FLU susceptibility. The AMBR isolates displayed no fitness defects 58 and one of them was fully virulent in a G. mellonella infection model. 59

These findings demonstrate that ergosterol depletion in N. glabratus 60 leads to AMB resistance without affecting fitness or virulence. 61 62 63 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint

64 Invasive candidiasis is a life -threatening fungal infection, caused by yeasts including 65 Candida albicans and Nakaseomyces glabratus , and comprises bloodstream 66 infections, but also dissemination to organs like liver and kidney, accounting for 67 approximately 1.5 million cases per year with attributable mortality of up to 60% [1,2]. 68 While C. albicans is the most virulent and best studied pathogen and still accounts for 69 approximately 50% of systemic Candida infections, non-albicans Candida species are 70 of increasing importance as they often develop antifungal drug resistance or even MDR 71 [1,4]. The recommended first -line treatment for s ystemic Candida infections are 72 echinocandins [5,6]. They bind the catalytic subunit of the ß -1,3-D-glucan synthetase 73 and inhibit the ß- 1,3-D-glucan biosynthesis, leading to disruption of cell wall integrity 74 and osmotic imbalance [7] . Echinocandin r esistance is mainly caused by point 75 mutations in the hot spot regions of the FKS genes which encode the enzyme’s 76 catalytic subunit [8]. Echinocandin resistance in N. glabratus is rare but increased in 77 recent years [9]. Due to the intrinsically low susceptibility of N. glabratus to azoles and 78 a high proportion of azole- resistant isolates , acquisition of echinocandin resistance 79 often results in MDR , leaving liposomal AMB as an indispensable option, despite 80 severe side effects for the patients such as high nephrotoxicity or serum electrolyte 81 changes [1,5,6]. AMB binding to ergosterol leads to either formation of small ion 82 channels or ergosterol extraction from the cell membrane [10-12]. AMB resistance is 83 still very rare among Candida species. Ergosterol depletion seems to be the major 84 resistance mechanism but is often linked to high fitness costs [7,13,14]. 85 Here, we describe AMB resistance in two independent clinical N. glabratus isolates 86 and show that combined mutations in ERG3 and ERG4 are responsible for the 87 resistance but do not result in fitness defects . One of the isolates also displayed 88 decreased susceptibility to anidulafungin and azoles. The latter wa s Pdr1-dependent 89 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint and likely triggered by ergosterol depletion. Thus, ergosterol depletion in the context 90 of AMB resistance can directly result in MDR phenotypes in N. glabratus. 91 92 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint

93 94 Strains and media 95 All strains used in this study are listed in Table S1. They were routinely grown in YPD 96 medium (20g/L glucose, 10g/L yeast extract, 20g/L peptone) at 37°C unless otherwise 97 indicated. 98 99 Antifungal drug susceptibility testing (AFST) 100 AFST was either performed with EUCAST-based broth microdilution [15] or with Etests 101 (Biomérieux) according to manufacturer’s instructions. 102 103 Plasmid and strain construction 104 The procedures of plasmid and strain construction can be found in the supplement 105 (Data S4). Oligonucleotide primers and plasmids are listed in Table S2 and S3. 106 107 DNA Isolation and whole-genome sequencing (WGS) 108 The ZR Fungal/Bacterial DNA MiniPrep kit (Zymo Research, Irvine, CA, USA) was 109 used to extract fungal genomic DNA. Genomic libraries were constructed and 110 barcoded using the NEBNext Ultra DNA Library Prep kit for Illumina (New England 111 Biolabs, Ipswich, MA, USA) and then sequenced using the Illumina platform. Further 112 details of the data analysis are available in the supplement (Data S4). 113 114 Measurement of sterol components 115 The sterol composition of clinical N. glabratus isolates was determined by gas 116 chromatography- mass spectrometry (GC-MS) as previously described [16]. 117 118 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Cell wall composition analysis 119 Details of the staining and measurement of N. glabratus cell wall components by flow 120 cytometry are listed in the supplement (Data S4). 121 122 Transcriptome analysis 123 Fungal cells were grown overnight in YPD at 37°C. 1x10 6 cells / mL were then added 124 to prewarmed YPD with or without 1 µg/mL AMB and grown for 1 h at 37°C. Cells were 125 harvested by centrifugation and total RNA was isolated as previously described [17]. 126 Library preparation and RNA -sequencing was performed by the Core Unit SysMed 127 Würzburg. Details of the data analysis are available in the supplement (Data S4). 128 129 Data availability 130 Sequencing data are available in the National Center for Biotechnology (NCBI) 131 Sequence Read Archive (SRA) under BioProject PRJNA1299776. 132 133 Gene expression analysis 134 100 ng/µl total RNA from the same conditions as used for the transcriptome analysis 135 were the template for RT-qPCR using the Luna Universal One-Step RT-qPCR Kit with 136 SYBR Green (New England Biolabs). Table S2 lists all used oligonucleotide primers. 137 Gene expression was calculated with the ∆∆Ct method [18]. RDN5.8 and a control 138 RNA (5h YPD, 37°C) were used for normalization. Data from independent biological 139 triplicates were compared with a two- tailed, unpaired student’s t-test and p values ≤ 140 0.05 were regarded as statistically significant. 141 142 Assessment of in vivo pathogenicity in Galleria mellonella larvae 143 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Prior to the inoculation with N. glabratus, Galleria mellonella larvae weighing 220-280 144 mg fasted for 24 hours. Larvae were wiped with 70% ethanol, clustered into groups of 145 15 and placed in individual petri dishes for observation. N. glabratus CBS138 and NRZ-146 2016-252 were grown overnight at 37 °C on Sabouraud dextrose agar . Fungal cells 147 were then harvested and suspended in sterile phosphate-buffered saline (PBS). Each 148 larva within a group was injected with 10 µL of the yeast cell suspension with 2x109 149 colony-forming units /mL, resulting in the inoculation of 2x10 7 yeast cells into each 150 larva. The control group larvae were injected with sterile PBS. The infected larvae were 151 incubated at 37°C. Larvae of the intervention groups were injected with AMB (5 mg / 152 kg body weight) one hour after inoculation. Larval survival, indicated by melani sation 153 and mobility, was periodically monitored during the next seven days . Survival rates 154 were visualized using Kaplan-Meier plots. 155 156 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint

157 ERG3 mutations are enriched among MDR isolates 158 We have previously analyzed the emergence of echinocandin-resistant clinical isolates 159 of N. glabratus in Germany [9]. Based on this work, we further examined strains, which 160 displayed (i) resistance to the echinocandin ANF without harboring FKS hot spot 161 mutations and (ii) additional resistance to either FLU or AMB (=MDR isolates, Figure 162 1 A). Whole-genome sequencing was performed for all six identified MDR isolates and 163 three isolates with isolated ANF resistance (ANFR) and one control isolate (AMBS, 164 ANFS, FLUI). The MDR strains formed no clear cluster but showed an enrichment of 165 ERG3 mutations (Figure 1 A, B). The two AMB R strains were genetically unrelated. 166 Their respective closest relatives were AMB S and displayed a high genetic variation 167 compared to the AMBR strains (Figure 1 B, C). We identified putative loss of function 168 mutations in the ERG3 genes of the strains NRZ -2017-099 (M1*), NRZ -2016-252 169 (Q26*) and NRZ-2016-150 (K133del) (Figure 1 A). The latter two strains also displayed 170 putative loss of function mutations in the ERG4 genes: T158fs and Y327* (Figure 1 A). 171 Despite identifying several FKS mutations in our strains, only the isolate NRZ-2017-172 099 harbored two mutations which might explain ANF resistance: K1323N in FKS1 and 173 T970fs in FKS2 (Figure 1 A). Despite displaying ANF resistance, no FKS mutations 174 were identified in NRZ-2016-252, NRZ-2016-191 and NRZ-2017-475 (Figure 1 A). 175 176 Cell wall composition of MDR isolates 177 Therefore, we used flow cytometry to measure the amounts of chitin, glucan and 178 mannan within the isolates as an altered composition is sometimes associated with 179 echinocandin resistance. There was no clear link to ANF resistance, but the 180 AMBR+ANFR isolate NRZ-2016-252 and the ANF R strain NRZ-2016-191 which both 181 harbor no FKS mutation, tend to have more of chitin, glucose and mannan (Figure S1). 182 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint 183 Ergosterol depletion correlates with resistance to AMB 184 We additionally examined the sterol composition of the strains which exhibited 185 resistance to AMB or azoles. With a percentage of 77%, ergosterol was the main sterol 186 in the reference strain CBS138 (Figure 2 A , Table S 5). The ERG3 mutations M1*, 187 V101del and C223F in the isolates NRZ-2017-099, NRZ-2017-128 and NRZ-2018-172 188 were associated with decreased ergosterol levels and the accumulation of ergosta-189 7,22-dien-3β-ol and ergost -7-en-3β-ol, indicating that the function of the sterol C5-190 desaturase Erg3 is disturbed or lost (Figure 2 A). The AMBR + ANFR isolates NRZ-191 2016-252 and NRZ-2016-150 had extremely low ergosterol concentrations (0.1-1%) 192 while ergosta-7,22,24(28)-trien-3β-ol increased up to 85% (Figure 2, Table S5). This 193 is only possible if neither Erg3 nor Erg4 work properly, fitting to the identified mutations 194 in both strains (Figures 1A and 2 B). Strains with low ergosterol concentration of 4-6% 195 are still susceptible to AMB, indicating that a complete depletion is required for AMB 196 resistance (Figure 2 B). 197 198 Absence of functional ERG3 and ERG4 induces resistance to AMB and ANF 199 We hypothesized that the simultaneous loss of function in Erg3 and Erg4 enzymes 200 cause AMB resistance and also lead to resistance against ANF and low susceptibility 201 to FLU. To confirm this, we replaced the ERG3Q26* allele in NRZ-2016-252 with the wild 202 type allele of CBS138. The resulting mutant became susceptible to AMB and ANF 203 (Figure 3 A ). Surprisingly, the strain was FLU R, maybe caused by the still present 204 ERG4T158fs mutation (Figure 3 A). Additionally, we deleted ERG3 and ERG4 in the 205

strain HTL. The resulting double mutant displayed AMB and ANF 206 resistance and a low susceptibility to FLU, similar to NRZ-2016-252 (Figure 3 B). 207 208 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Pdr1 links ergosterol depletion and low susceptibility to FLU 209 To better understand the fungal cell response to AMB, we examined the transcriptomes 210 of the AMBS strain CBS138 and the AMBR isolate NRZ-2016-252 after 1 h incubation 211 in YPD with or without 1 µg/ml AMB at 37°C. 1203 genes were differentially expressed 212 in CBS138 in response to AMB (Figure 4 A, dataset S6). Interestingly, 68% of them 213 were differentially expressed in the AMB R strain NRZ-2016-252 in absence of AMB , 214 indicating that this strain is already well -adapted to AMB (Figure 4 A, dataset S 6). 215 Among the upregulated genes in the AMBR strain were stress-related genes (UPC2B, 216 TYE7, ICL1, ICL2, HSP12, RAD27), efflux pumps (CDR1, PDH1, FLR1 and FLR2) and 217 ergosterol biosynthesis genes (Figure 4 B and C, dataset S6). The two latter groups 218 are known targets of the transcription factor Pdr1 [19,20]. Especially the up-regulation 219 of the efflux pump genes might explain the low FLU susceptibility of NRZ -2016-252. 220 Deletion of the PDR1 gene caused a dramatic decline of CDR1, FLR1 and PDH1 221 transcription in NRZ -2016-252 (Figure 5 A). Compared to NRZ-2016-252, the pdr1 ∆ 222 derivate was extremely susceptible to fluconazole, indicating that up regulation of 223 CDR1 and PDH1 in NRZ-2016-252 was required for the low FLU susceptibility of NRZ-224 2016-252 (Figure 5 B ). NRZ-2016-252 pdr1∆ remained AMB and ANF resistant, 225 illustrating that Pdr1 was not required for resistance to these drugs (Figure 5 B). 226 227 AMBR isolates display no severe fitness defects or attenuated virulence 228 As AMB resistance is often linked to profound fitness defects, we examined the growth 229 dynamics of the AMBR isolates under different conditions. No apparent growth defects 230 were observed in YPD medium at 37°C where both strains showed similar dynamics 231 as the reference strain CBS138 (Figure 6 A). We then tested the growth of these 232 strains, the isolate NRZ -2016-191 and the reference strain CBS138 under several 233 stress conditions. Despite the ergosterol depletion, both AMB R isolates showed good 234 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint stress resilience at 37°C and robust growth at 42°C (Figure 6 B). Only NRZ-2016-252 235 displayed a growth delay under osmotic stress conditions (1.5 M NaCl) and was also 236 more susceptible to combined stressors such as 42°C and 1.5 M NaCl or 0.0125% 237 SDS (Figure 6 B, C). 238 Finally, we analyzed the virulence of the AMBR strain NRZ-2016-252 in Galleria 239 mellonella. The isolate displayed no attenuated virulence compared to CBS138 (Figure 240 7). After treatment with AMB, 86% of the larvae (12/14) survived the infection with 241 CBS138 after 7 days (Figure 7) which was not the case for NRZ -2016-252-infected 242 larvae which died within the first 4 days, independent from the addition of AMB (Figure 243 7). 244 245 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint

246 Our study showed that loss of function mutations in the ERG3 and ERG4 genes led to 247 ergosterol depletion and consequently AMB resistance, similar to previous findings for 248 Clavispora lusitaniae and Candida auris (syn. Candidozyma auris) [21,22]. Absence of 249 functional Erg3 and Erg4 resulted in a massive shift from ergosterol to ergosta-250 7,22,24(28)-trien-3ßol in the two AMBR isolates which was also reported for AMB R 251 C. auris strains [22]. This understudied sterol has a similar binding affinity to liver X 252 receptors as ergosterol and is often enriched in ergosterol lacking fungal strains [23-253 25]. The primary effect of ergosterol depletion is AMB resistance as the drug can no 254 longer bind to its target ergosterol. Additionally, it induced a transcriptional adaptation 255 against AMB even in its absence, including the up-regulation of Pdr1-controlled genes 256 CDR1 and PDH1. The increased expression of these ABC transporter genes led to an 257 increased efflux pump activity and therefore to reduced susceptibility to FLU which is 258 in accordance with previous observations [19,20]. The observed association between 259 AMBR and ANF R without underlying FKS hot spot mutations could not be fully 260 explained. Changes in the cell wall and the cell membrane caused by the absence of 261 ergosterol may reduce the accessibility of the ß -1,3-D glucan synthase for 262 anidulafungin. Similar to previous works [26], our strains were susceptible to 263 micafungin, either caused by a higher affinity or less interaction with the altered cell 264 membrane. 265 Strikingly, AMB R isolate NRZ -2016-252 was fully virulent in a Galleria mellonella 266 infection model. Its overall in -vitro and in-vivo fitness contradicts the hypothesis that 267 acquisition of AMB resistance is associated with high fitness costs [14, 22]. Especially, 268 the bloodstream was previously discussed as an environment too hostile for the 269 survival of AMB resistant strains [14], however our two AMBR isolates were obtained 270 from blood cultures. We presume that the acquisition of AMB resistance alone made 271 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint the strains not more susceptible to the harsh conditions within the bloodstream. 272 Suppressor mutations might have bypassed some defects caused by ergosterol 273 depletion. 274 We described N. glabratus bloodstream isolates with stable AMB resistance without 275 apparent fitness and virulence defects. In combination with the intrinsically low 276 susceptibility to azoles and the emerging echinocandin resistance, these findings 277 underline the threat of an increasing MDR and extensively drug resistance (XDR) in 278 this major human fungal pathogen. A close resistance monitoring is therefore urgently 279 needed. 280 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint

281 This work was supported by the German Research Foundation (DFG) through the TRR 282 124 FungiNet “Pathogenic fungi and their human host: Networks of Interaction”, DFG 283 project number 210879364, project C3 (O. K.) and NIH NIAID grant U19AI110818 to 284 the Broad Institute (C.A.C.). We want to thank Elke Huprich, Ina Gaube, Barbara 285 Conrad, Sabrina Speiser and Margarete Göbel for their excellent technical support to 286 this project. 287 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint

288 289 [1] Lass-Florl C, Kanj SS, Govender NP, Thompson GR, 3rd, Ostrosky -Zeichner L, Govrins MA. 290 Invasive candidiasis. Nat Rev Dis Primers. 2024;10(1):20. 291 [2] Holmes CL, Albin OR, Mobley HLT, Bachman MA. Bloodstream infections: mechanisms of 292 pathogenesis and opportunities for intervention. Nat Rev Microbiol. 2025;23(4):210- 24. 293 [3] Denning DW. Global incidence and mortality of severe fungal disease. Lancet Infect Dis. 294 2024;24(7):e428-e38. 295 [4] Arendrup MC, Patterson TF. Multidrug- Resistant Candida: Epidemiology, Molecular 296 Mechanisms, and Treatment. J Infect Dis. 2017;216(suppl_3):S445-S51. 297 [5] Pappas PG, Kauffman CA, Andes DR, Clancy CJ, Marr KA, Ostrosky-Zeichner L, et al. Clinical 298 Practice Guideline for the Management of Candidiasis: 2016 Update by the Infectious Diseases 299 Society of America. Clin Infect Dis. 2016;62(4):e1-50. 300 [6] Cornely OA, Sprute R, Bassetti M, Chen SC, Groll AH, Kurzai O, et al. Global guideline for the 301 diagnosis and management of candidiasis: an initiative of the ECMM in cooperation with ISHAM 302 and ASM. Lancet Infect Dis. 2025;25(5):e280-e93. 303 [7] Lee Y, Puumala E, Robbins N, Cowen LE. Antifungal Drug Resistance: Molecular Mechanisms 304 in Candida albicans and Beyond. Chem Rev. 2021;121(6):3390-411. 305 [8] Perlin DS. Resistance to echinocandin- class antifungal drugs. Drug Resist Updat. 306 2007;10(3):121-30. 307 [9] Aldejohann AM, Herz M, Martin R, Walther G, Kurzai O. Emergence of resistant Candida 308 glabrata in Germany. JAC Antimicrob Resist. 2021;3(3):dlab122. 309 [10] Ermishkin LN, Kasumov KM, Potzeluyev VM. Single ionic channels induced in lipid bilayers by 310 polyene antibiotics amphotericin B and nystatine. Nature. 1976;262(5570):698-9. 311 [11] Mouri R, Konoki K, Matsumori N, Oishi T, Murata M. Complex formation of amphotericin B in 312 sterol-containing membranes as evidenced by surface plasmon resonance. Biochemistry. 313 2008;47(30):7807-15. 314 [12] Anderson TM, Clay MC, Cioffi AG, Diaz KA, Hisao GS, Tuttle MD, et al. Amphotericin forms an 315 extramembranous and fungicidal sterol sponge. Nat Chem Biol. 2014;10(5):400- 6. 316 [24] Barrero AF, Oltra JE, Robinson J, Burke PV, Jimenez D, Oliver E. Sterols in erg mutants of 317 Phycomyces: metabolic pathways and physiological effects. Steroids. 2002;67(5):403- 9. 318 [13] Ellis D. Amphotericin B: spectrum and resistance. J Antimicrob Chemother. 2002;49 Suppl 1:7-319 10. 320 [14] Vincent BM, Lancaster AK, Scherz -Shouval R, Whitesell L, Lindquist S. Fitness trade- offs 321 restrict the evolution of resistance to amphotericin B. PLoS Biol. 2013;11(10):e1001692. 322 [15] The European Committee on Antimicrobial Susceptibility Testing. Overview of antifungal 323 ECOFFs and clinical breakpoints for yeasts, moulds and dermatophytes using the EUCAST 324 E.Def 7.3, E.Def 9.4 and E.Def 11.0 procedures. Version 3, 2022. http://www.eucast.org 325 [16] Muller C, Binder U, Bracher F, Giera M. Antifungal drug testing by combining minimal inhibitory 326 concentration testing with target identification by gas chromatography -mass spectrometry. Nat 327 Protoc. 2017;12(5):947-63. 328 [17] Martin R, Moran GP, Jacobsen ID, Heyken A, Domey J, Sullivan DJ, et al. The Candida 329 albicans-specific gene EED1 encodes a key regulator of hyphal extension. PLoS One. 330 2011 ;6(4):e18394. 331 [18] Pfaffl MW. A new mathematical model for relative quantification in real -time RT-PCR. Nucleic 332 Acids Res. 2001;29(9):e45. 333 [19] Vu BG, Thomas GH, Moye- Rowley WS. Evidence that Ergosterol Biosynthesis Modulates 334 Activity of the Pdr1 Transcription Factor in Candida glabrata. mBio. 2019;10(3). 335 [20] Vu BG, Stamnes MA, Li Y, Rogers PD, Moye- Rowley WS. The Candida glabrata Upc2A 336 transcription factor is a global regulator of antifungal drug resistance pathways. PLoS Genet. 337 2021;17(9):e1009582. 338 [21] Kannan A, Asner SA, Trachsel E, Kelly S, Parker J, Sanglard D. Comparative Genomics for the 339 Elucidation of Multidrug Resistance in Candida lusitaniae. mBio. 2019;10(6). 340 [22] Massic L, Doorley LA, Jones SJ, Richardson I, Siao DD, Siao L, et al. Acquired Amphotericin B 341 Resistance Attributed to a Mutated ERG3 in Candidozyma auris. bioRxiv. 2025. 342 [23] Renaud RL, Subden RE, Pierce AM, Oehlschlager AC. Sterol composition ofNeurospora 343 crassa. Lipids. 1978;13(1):56-8. 344 [25] Taank Y, Randhawa V, Agnihotri N. Ergosterol and its metabolites as agonists of Liver X 345 receptor and their anticancer potential in colorectal cancer. J Steroid Biochem Mol Biol. 346 2024;243:106572. 347 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint [26] Healey KR, Katiyar SK, Castanheira M, Pfaller MA, Edlind TD. Candida glabrata mutants 348 demonstrating paradoxical reduced caspofungin susceptibility but increased micafungin 349 susceptibility. Antimicrob Agents Chemother. 2011; 55(8):3947-9. 350 351 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure Legends 352 Figure 1 . Resistance profile and genetic characteristics of MDR N. glabratus 353 isolates 354 (A). Summary of antifungal susceptibility testing (AFST) and underlying mutations in 355 possible resistance genes. AFST was performed with broth microdilution and based 356 on EUCAST clinical breakpoints, the isolates were determined as susceptible (S), 357 increased-dose-dependent (I) or resistant (R). Shown mutations in the indicated genes 358 were identified by whole genome sequencing (WGS). (B) WGS- based phylogenetic 359 tree of the examined N. glabratus isolates. (C) Genetic diversity of AMBR isolates NRZ-360 2016-252 and NRZ-2016-150 and their closest relatives NRZ -2016-191 and NRZ-361 2017-476. 362 363 Figure 2. Sterol composition analysis of CBS138 and the clinical isolates. 364 (A) Sterol composition of the indicated strains was analyzed by gas chromatography- 365 mass spectrometry (GC-MS). Shown are the percentages of the single sterols in the 366 indicated strains. (B) Illustration of the effects of the absence of either Erg3 and/ or 367 Erg4 in AMBS and AMBR strains. Shown are branches of the ergosterol biosynthesis 368 pathway in absence of Erg3 and/ or Erg4 leading to the production of non-physiological 369 predecessors. 370 371 Figure 3. Absence of ERG3 and ERG4 is required for resistance to AMB and ANF. 372 (A) Integration of a wild type ERG3 allele into the AMB R isolate NRZ -2016-252 373 increased susceptibility to AMB and ANF but not to FLU. The indicated strains were 374 plated onto RPMI1640 medium and E-test stripes for AMB, ANF and FLU were applied. 375 The plates were then grown for 48h at 37°C before pictures were taken. (B) ERG4 and 376 ERG3 were deleted in the N. glabratus HTL background strain. HTL and the 377 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint erg4∆/ erg3∆ double mutants were plated and antifungal drug susceptibility was 378 determined with E-tests as described in (A). 379 380 Figure 4. Transcriptional response of AMBS and AMBR strains to AMB. 381 The AMBS strain CBS138 and the AMB R strain NRZ-2016-252 were grown for 1h in 382 YPD with 1 µg/ml AMB (AMB) or without 1 µg/ml AMB (YPD) prior to RNA isolation. 383 (A) Comparison of upregulated genes in CBS138 and NRZ -2016-252 in AMB and of 384 genes upregulated in NRZ-2016-252 compared to CBS138 in YPD. Example genes 385 up-regulated in NRZ-2016-252 compared to CBS138 in presence of absence of AMB 386 are shown in the box. (B) Vulcano plot of differentially expressed genes in NRZ-2016-387 252 compared to CBS138 after 1 h growth in YPD at 37°C. (C ) Comparison of ERG 388 gene expression in NRZ-2016-252 and CBS138 in either YPD or YPD + AMB. 389 390 Figure 5. Deletion of PDR1 led to downregulation of efflux pump genes in NRZ -391 2016-252 and increased susceptibility to FLU. 392 (A) Expression of efflux pump genes in NRZ-2016-252 and NRZ-2016-252 pdr1∆ after 393 1 h growth in YPD medium at 37°C. Gene expression was normalized against CBS138 394 (1h YPD, 37°C) and the RDN5.8 gene. Asterisks indicate significant changes in gene 395 expression between the two N. glabratus strains (two-tailed, unpaired student’s t-test, 396 p ≤ 0.05). (B) The indicated strains were plated onto RPMI1640 medium, and E- test 397 stripes for AMB, ANF and FLU were applied. Pictures were taken after an incubation 398 for 48h at 37°C. 399 400 Figure 6. AMB R isolates show no fitness defects under single but under 401 combinatory stress conditions. 402 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint (A) Growth curve of CBS138, NRZ -2016-252 and NRZ -2016-150 grown in YPD 403 medium at 37°C for 11 h. The optical density of the three strains was measured by 404 photometry at 600 nm. (B) The indicated N. glabratus strains were grown on YPD with 405 or without 1.5 M sodium chloride or 100 µM menadione to induce osmotic and oxidate 406 stress. The plates were incubated at 37°C for 3 days prior to photography, except one 407 plate which was incubated at 42°C. (C) The same N. glabratus strains were grown on 408 YPD at 42°C with or without 200 µg/ml Congo Red to initiate cell wall stress, 1.5 M 409 NaCl for osmotic stress, 100 µM menadione for oxidative stress and 0.0125% SDS for 410 cell membrane stress. The plates were grown for 3 days prior to photography. 411 412 Figure 7. N. glabratus NRZ-2016-252 shows resistance against AMB under in 413 vivo conditions. 414 Galleria mellonella larvae were infected with 2x10 7 N. glabratus cells. The infected 415 larvae were treated with 5 mg AMB per kg body weight and incubated for up to 7 days 416 at 37°C. The evaluation of survival of the larvae, indicated by melanisation and mobility, 417 was first monitored 12 h post infection and then periodically each 24 h. Survival rates 418 were visualized by Kaplan-Meier plots. 419 420 Figure S1. Cell wall composition of MDR N. glabratus isolates. 421 The indicated N. glabratus strains were grown in YPD at 37°C, harvested and then 422 stained for chitin, ß- 1,3-D-glucan and mannan. The amounts of the three cell wall 423 components were measured by flow cytometry. The values were normalized against 424 the values of the control strain CBS138. 425 426 427 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 1 428 429 430 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 2 431 432 433 434 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 3 435 436 437 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 4 438 439 440 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 5 441 442 443 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 6 444 445 446 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure 7 447 448 449 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Figure S1 450 451 452 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Table S1. N. glabratus strains used in this study. 453 454 Strain Genotype Source CBS138 wild type Gillum et al, 1984 NRZ-2015-067 wild type Aldejohann et al., 2021 NRZ-2016-150 wild type Aldejohann et al., 2021 NRZ-2016-191 wild type Aldejohann et al., 2021 NRZ-2016-252 wild type Aldejohann et al., 2021 NRZ-2017-099 wild type Aldejohann et al., 2021 NRZ-2017-128 wild type Aldejohann et al., 2021 NRZ-2017-475 wild type Aldejohann et al., 2021 NRZ-2017-476 wild type Aldejohann et al., 2021 NRZ-2018-032 wild type Aldejohann et al., 2021 NRZ-2018-172 wild type Aldejohann et al., 2021 HTL his3::FRT, trp1::FRT, leu2::FRT Jacobsen et al., 2010 erg3∆ HTL, erg3::ScHIS3 This work. erg4∆ HTL, erg4::ScLEU2 This work. erg4∆ erg3∆ HTL, erg4::ScLEU2, erg3::ScHIS3 This work. NRZ-2016-252 ERG3WT NRZ-2016-252, erg3Q26*::ERG3 This work. NRZ-2016-252 pdr1∆ NRZ-2016-252, pdr1::NAT1 This work. 455 456 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Table S2. Oligonucleotide primers used in this study. 457 458 Name Sequence in 5’ to 3’ direction Usage ScHIS3p-CgERG3-OL aacagcactaagcttttatacaaaaacaaataAACACAGTC CTTTCCCGCAATTTTCTTTTT plasmid construction ScHIS3t-CgERG3-OL cgagacaccggtgtttcctgtAATATGAAATGCTTTTCTT GTTGTTCTTACG plasmid construction 5'CgERG3-ScHIS3-OL AAAAAGAAAATTGCGGGAAAGGACTGTGTTtatt tgtttttgtataaaagcttagtgctgtt plasmid construction 3'CgERG3-ScHIS3-OL CGTAAGAACAACAAGAAAAGCATTTCATATTac aggaaacaccggtgtctcg plasmid construction 5'CgERG3-SacII-OL GGGGATCCACTAGTTCTAGAGCGGCCGCCAgt aaagtcagtgttggcgacca plasmid construction 3'CgERG3-SacII-OL TCACTAAAGGGAACAAAAGCTGGAGCTCCAtgc tagtcagcagccgtgggt plasmid construction ScLEU2p-CgERG4-OL ccaagacagacactttttttgagatcaacaATCTATTACATT ATGGGTGGTATGTTGGAA plasmid construction ScLEU2t-CgERG4-OL gggttataatgcatcttttctttatggcatGTGTTTTTTATTTGT TGTATTTTTTTTTTTTTAG plasmid construction 5'CgERG4-ScLEU2p- OL TTCCAACATACCACCCATAATGTAATAGATtgttg atctcaaaaaaagtgtctgtcttgg plasmid construction 3'CgERG4-ScLEU2t- OL CTAAAAAAAAAAAAATACAACAAATAAAAAACA Catgccataaagaaaagatgcattataaccc plasmid construction 5'CgERG4-PstIoverlap GACGGTATCGATAAGCTTGATATCGAATTCagc gcctgctgctaaaacactg plasmid construction 3'CgERG4-PstIoverlap GGCCGCTCTAGAACTAGTGGATCCCCCGGGg gaaggtcgtctataccaagttga plasmid construction ScNAT1p-CgPDR1-OL gtcattctttagctacgttattgagagaatCATAGCTTCAAAA TGTTTCTACTCCTTTTT plasmid construction ScNAT1t-CgPDR1-OL tgagagatattgtagtgttatcgctaGCAAATTAAAGCCTT CGAGCGTCCCAAAAC plasmid construction 5'CgPDr1-ScNAT1-OL AAAAAGGAGTAGAAACATTTTGAAGCTATGattc tctcaataacgtagctaaagaatgac plasmid construction 3'CgPDR1-ScNAT1- OL GTTTTGGGACGCTCGAAGGCTTTAATTTGCtag cgataacactacaatatctctca plasmid construction 5'CgPDR1-SacII-OL GGGGATCCACTAGTTCTAGAGCGGCCGCCAta catcgtaacaaacatttcctcatagatc plasmid construction 3'CgPDR1-SacII-OL TCACTAAAGGGAACAAAAGCTGGAGCTCCAag agttacagacgaccaacgtg plasmid construction pSK forward_2 GATGTGCTGCAAGGCGATTAAGTTG deletion cassette amplification pSK revers ACACAGGAAACAGCTATGACCATGA deletion cassette amplification X2-NAT1 CTGTGCTTGGGTGTTTTGAAGTGGTAC verification X3-NAT1 TACGACGGCACCGCCTCGGA verification X2-ScHIS3 GAGTGTACTAGAGGAGGCCAAGA verification X3-ScHIS3 TGTGGTGATAGGTGGCAAGTGG verification X2-ScLEU2 GCGTCATCTTCTAACACCGTATATG verification X3-ScLEU2 ACAAGGAGGAGGGCACCACA verification G1-CgERG3 CTACGAGAACAAGAGCTAAGAGTAT verification G4-CgERG3 GATGTAGGAAAAGTAATGTGTGCG verification G1-CgERG4-NEB GAAGGAGAATGCGGGTCCAG verification G4-CgERG4-NEB GCTGCTTCTGCTGCTGGTTATG verification G1-CgPDR1 TGATTGTACCCATACAGAAGAAAACTTAGA verification G4-CgPDR1 ATGACTGATTCTTTTGGTAATTATTTGATTCAG verification R1-CgERG2 ATGTCATCCTATTTGGTACCGCAG gene expression R2-CgERG2 GTTTTGATCCATAGCGTATTGCTTTG gene expression R1-CgERG3 CACTCCATTCGCCTCCCAC gene expression R2-CgERG3 GATGTAGGAAAAGTAATGTGTGCG gene expression R1-CgERG4 ACGGTTGGTACAGATATGCCAG gene expression .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint R2-CgERG4 TGCAGTATTCAACCCAGTCCTTG gene expression R1-CgERG5 GACTACCAAGCTCCAAAGGGTTC gene expression R2-CgERG5 TGGAGTGACCTTGTGCTTGAAGTC gene expression R1-CgERG6 ATGAAGAGCACCGTAAGATCGCTTA gene expression R2-CgERG6 CATACAGTTAGTGAATTTTCTACCGAAG gene expression R1-CgERG11 AGTCTCCCCAGGTTACACTCAC gene expression R2-CgERG11 ACACCCAATTGACAGTAAGCGAAC gene expression R1-CgCDR1 CCAGGTGGCAGAAGCAGCA gene expression R2-CgCDR1 ATGGTCCCAAGTACTCGCCAC gene expression R1-CgFLR1 AGCATCAAAGTCGCAGCTAAGAG gene expression R2-CgFLR1 GACTGAAGCAACATACTTTGGATAG gene expression R1-CgFLR2 GTGTTATCCAGAATACGTTGCATC gene expression R2-CgFLR2 TCTGGACTAAATCTTGATCTTGCTC gene expression R1-CgPDH1 TGTGGTGTGATGGCTACTCCAG gene expression R2-CgPDH1 AGTACCTGCTACATTCAGATAAGGAG gene expression R1-CgSNQ2 TGTGGTGTTGTTCAGCCCGTTTC gene expression R2-CgSNQ2 AGTTTGTCCAGCTGGGGGATC gene expression 459 low case: restriction sites 460 461 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint Table S3. Plasmids used in this study. 462 463 Strain Features Source pSK Bluescript Beta lactamase gene Agilent pSK-SCH9-NAT1 Beta lactamase gene, NAT1 gene with 1000bp homology regions for the integration into the CgSCH9 locus Pohlers et al., 2017 pSK-CgERG3-ScHIS3 Beta lactamase gene, ScHIS3 gene with 1000bp homology regions for the integration into the CgERG3 locus This work. pSK-CgERG4-ScLEU2 Beta lactamase gene, ScLEU2 gene with 1000bp homology regions for the integration into the CgERG4 locus This work. pSK-CgPDR1-NAT1 Beta lactamase gene, NAT1 gene with 1000bp homology regions for the integration into the CgPDR1 locus This work. pSK-CgERG3WT- NAT1 Beta lactamase gene, ERG3WT (from CBS138, including promoter and terminator) gene with 1000bp homology regions for the integration into the CgERG3 locus This work. 464 .CC-BY 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 28, 2025. ; https://doi.org/10.1101/2025.08.28.672802doi: bioRxiv preprint