CoChat

Candida auris skin tropism and antifungal resistance are mediated by carbonic anhydrase Nce103

Open access

The DOI matches its Crossref record; 87 of the 87 references it cites were found in a registry or an index; 51 of the 71 sources its claims cite were checked for support.

Abstract

The pronounced skin tropism and pan-antifungal resistance of Candida auris pose a serious global health threat. A key question in C. auris biology is how clinical isolates acquire amphotericin B resistance. Here we demonstrate that a carbonic sensing pathway (CSP) contributes to amphotericin B resistance by modulating mitochondrial energy functions in clinical C. auris isolates. Integrated transcriptomics and proteomics identify the carbonic anhydrase Nce103 and its transcription factors Rca1 and Efg1 as important regulatory components of the CSP. The conversion of CO 2 into bicarbonate sustains energy metabolism required for colonization and fitness on human skin and in nutrient-limited microenvironments. We also show that bacterial skin colonizers engage urease to release CO 2 that sustains C. auris fitness and skin colonization. These findings highlight therapeutic options to re-sensitize C. auris to antifungal treatments, as well as to prevent skin colonization by blocking the CSP.

Overview

CoChat Summary

TL;DR

This study identifies a carbonic sensing pathway (CSP) centered on Nce103, Rca1, and Efg1 as a driver of both amphotericin B resistance and skin colonization in *Candida auris*. The authors show that CO2/bicarbonate-dependent metabolism supports fungal fitness, and that CO2 released by urease-positive skin bacteria can promote *C. auris* growth.

Key findings

  • In clinical *C. auris* isolates, deleting RCA1 reduced amphotericin B MIC by 2-fold to 4-fold, while deleting EFG1 reduced MIC by 2-fold; the rca1∆efg1∆ double mutant did not further worsen susceptibility, consistent with an epistatic CSP.
  • Deleting NCE103 caused severe fitness defects in ambient air, and the nce103∆ mutant showed the same amphotericin B MIC as rca1∆ in 1.5% CO2; 5.5% CO2 fully restored amphotericin B resistance in all strains tested.
  • In phenotype assays, deleting RCA1, EFG1, or NCE103 increased sensitivity to SDS, NaCl, H2O2, and antimycin A, while NCE103 overexpression reduced susceptibility to NaCl, H2O2, antimycin A, MMS, and rapamycin.
  • In skin colonization models, nce103∆ showed 10- to 100-fold lower burden than wild type on mouse skin after 48 h and on human skin ex vivo after 24 h; rca1∆ and efg1∆ also showed 10- to 100-fold reduced colonization of mouse and human skin.
  • Urease-positive bacteria, specifically Proteus mirabilis and Klebsiella pneumoniae, significantly enhanced growth of CSP mutants in vitro, but this effect was lost with urease-negative bacterial controls.

Methods

  • The study combined shotgun proteomics and RNA-seq/transcriptomics across resistant and sensitive clinical *C. auris* isolates to identify pathways linked to antifungal resistance.
  • Functional validation used gene deletions, overexpression constructs, CRISPR–Cas9-based NCE103 deletion, spotting assays, MIC and checkerboard assays, lipidomics, and mouse/human skin colonization models.
  • Mouse work included intradermal infection, skin colonization, and systemic infection assays in C57BL/6 mice; human skin assays used abdominal skin samples from three independent donors.

Limitations

  • The paper states that the effect of skin-dwelling bacteria is supported by in vitro co-culture assays and suggests a role for microbiome-derived CO2, but the causal contribution in vivo is not directly established.
  • The authors note that Nce103-mediated changes in lipid composition were not clearly detectable by their lipidomics, and some lipid/fluidity findings were variable across independent replicates.
  • Evident from the text: the study relies on a limited number of human skin donors (three independent donors) and mouse groups (5–11 mice per group), which may constrain generalizability.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41564-025-02189-z

Cites

87 of 87 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Proteomics unravels multidrug resistance mechanisms in C. auris. a, Heat map showing MIC (µg ml−1) values for multidrug-resistant (R1, R2) and drug-sensitive (S) clinical strains following the CLSI method. 5FC, 5-fluorocytosine; FLC, fluconazole; VRC, voriconazole; ITC, itraconazole; KTC, ketoconazole; CAS, caspofungin. b, Proteomic workflow and data analysis. c, Heat map shows top transporters that are of high abundance in strain R1. FC, fold change. d, Volcano plot of proteomics data shows DAPs between R1 and S strains. Statistical analyses were performed using moderated two-sided t-statistics with the limma-trend method, followed by multiple test correction using the Benjamini–Hochberg procedure. Differential expression analysis filters proteins with log2 fold change (log2FC) ± 0.58 and adjusted P < 0.05 indicated as dashed line. e, Gene Ontology term enrichment analysis for DAPs between R1 and S strains. The dot plot presents data from Gene Ontology enrichment analysis using clusterProfiler with P value calculated by hypergeometric distribution and Benjamini–Hochberg for multiple test correction. f, Scheme reveals a CSP controlling a link between metabolism and morphogenesis in fungal species. Panel b created with BioRender.com. Panel f adapted with permission from ref. 54, Springer Nature Limited. Source data
Fig. 1 Proteomics unravels multidrug resistance mechanisms in C. auris. a, Heat map showing MIC (µg ml−1) values for multidrug-resistant (R1, R2) and drug-sensitive (S) clinical strains following the CLSI method. 5FC, 5-fluorocytosine; FLC, fluconazole; VRC, voriconazole; ITC, itraconazole; KTC, ketoconazole; CAS, caspofungin. b, Proteomic workflow and data analysis. c, Heat map shows top transporters that are of high abundance in strain R1. FC, fold change. d, Volcano plot of proteomics data shows DAPs between R1 and S strains. Statistical analyses were performed using moderated two-sided t-statistics with the limma-trend method, followed by multiple test correction using the Benjamini–Hochberg procedure. Differential expression analysis filters proteins with log2 fold change (log2FC) ± 0.58 and adjusted P < 0.05 indicated as dashed line. e, Gene Ontology term enrichment analysis for DAPs between R1 and S strains. The dot plot presents data from Gene Ontology enrichment analysis using clusterProfiler with P value calculated by hypergeometric distribution and Benjamini–Hochberg for multiple test correction. f, Scheme reveals a CSP controlling a link between metabolism and morphogenesis in fungal species. Panel b created with BioRender.com. Panel f adapted with permission from ref. 54, Springer Nature Limited. Source dataCC BY · PubMed Central
Fig. 2 A CSP controls AMB susceptibility. a, Mutants in a CSP were subjected to MIC assays using the CLSI protocol (three biological replicates at indicated CO2 concentrations). The deletion mutants nce103∆ and rca1∆nce103∆ fail to grow in ambient air (~0.04% CO2). b, Protein levels of Rca1, Nce103 and Efg1 in strains R1 and S indicated by proteomics data (n = 3 biological replicates per group). Adjusted P values of DAPs from Limma analysis (Fig. 1d) are shown, and error bars indicate mean ± s.d. c, A synteny scheme shows conservation of carbonic anhydrase genes across fungal pathogens. d, Integrated heat map between proteomics and RNA-seq datasets reveal the upregulation of NCE103 in resistant strains. Cut-off: log2FC = ±0.58. e, CO2 enhances AMB tolerance in a subset of clinical C. auris strains (highlighted in bold pink). Spot dilution assays were performed using 5-fold serial dilutions of C. auris cells spotted onto YPD agar with or without AMB. f, Two-sided Pearson correlation between NCE103 mRNA expression (log2FC (NCE103/ACT1)) and AMB resistance among clinical C. auris isolates. AMB resistance indexes were semi-quantified using spot dilution assays on AMB plates. Two AMB concentrations (1 µg ml−1 and 2 µg ml−1) were used to capture a range of susceptibilities. Two biological replicates were performed, yielding consistent results. g, NCE103 mRNA levels in C. auris mutants under low (0.04%) and high (5.5%) CO2 conditions show that both Rca1 and Efg1 modulate NCE103 expression (n = 3 biological replicates per group). Approximately 200 c.f.u. of C. auris were plated onto YPD agar and incubated at 37 °C under either 5.5% CO2 or ambient air conditions for 2 days. Colonies were then collected for mRNA isolation. Two-way ANOVA followed by Dunnett’s test was applied with R1 used as control group. Each data point is shown, and error bars indicate mean ± s.d. Orthologues and gene order information were retrieved from the Candida Gene Order Browser (cgob.ucd.ie) and fungi.ensembl.org. Source data
Fig. 2 A CSP controls AMB susceptibility. a, Mutants in a CSP were subjected to MIC assays using the CLSI protocol (three biological replicates at indicated CO2 concentrations). The deletion mutants nce103∆ and rca1∆nce103∆ fail to grow in ambient air (~0.04% CO2). b, Protein levels of Rca1, Nce103 and Efg1 in strains R1 and S indicated by proteomics data (n = 3 biological replicates per group). Adjusted P values of DAPs from Limma analysis (Fig. 1d) are shown, and error bars indicate mean ± s.d. c, A synteny scheme shows conservation of carbonic anhydrase genes across fungal pathogens. d, Integrated heat map between proteomics and RNA-seq datasets reveal the upregulation of NCE103 in resistant strains. Cut-off: log2FC = ±0.58. e, CO2 enhances AMB tolerance in a subset of clinical C. auris strains (highlighted in bold pink). Spot dilution assays were performed using 5-fold serial dilutions of C. auris cells spotted onto YPD agar with or without AMB. f, Two-sided Pearson correlation between NCE103 mRNA expression (log2FC (NCE103/ACT1)) and AMB resistance among clinical C. auris isolates. AMB resistance indexes were semi-quantified using spot dilution assays on AMB plates. Two AMB concentrations (1 µg ml−1 and 2 µg ml−1) were used to capture a range of susceptibilities. Two biological replicates were performed, yielding consistent results. g, NCE103 mRNA levels in C. auris mutants under low (0.04%) and high (5.5%) CO2 conditions show that both Rca1 and Efg1 modulate NCE103 expression (n = 3 biological replicates per group). Approximately 200 c.f.u. of C. auris were plated onto YPD agar and incubated at 37 °C under either 5.5% CO2 or ambient air conditions for 2 days. Colonies were then collected for mRNA isolation. Two-way ANOVA followed by Dunnett’s test was applied with R1 used as control group. Each data point is shown, and error bars indicate mean ± s.d. Orthologues and gene order information were retrieved from the Candida Gene Order Browser (cgob.ucd.ie) and fungi.ensembl.org. Source dataCC BY · PubMed Central
Fig. 3 Different mechanisms contribute to CSP-mediated antifungal resistance. a, Semiquantitative spotting assays with different stress agents. Serial 5-fold dilutions of fungal cell suspensions were spotted onto YPD agar containing chemical agents as indicated. Semiquantitative data were collected by comparing relative growth properties of tested strains. b, Potential mechanisms implicated in CSP-mediated AMB resistance deduced from data shown in a. c, Gene Ontology enrichment for differentially expressed genes in rca1∆ and efg1∆ mutants compared to WT. Data were filtered with cut-off log2FC of ±0.58. Common Gene Ontology terms of rca1∆ and efg1∆ were visualized. d, Spotting assays for CSP mutants on media containing 3 µg ml−1 AMB and 5 mM VitC. e, MIC assays for AMB alone and in combination with the mitochondrial inhibitor AA. The AA concentration ranged from 0 to 1 µg ml−1 (top to bottom) for each strain. Three biological replicates yielded similar results in a, d and e. Credit: icons in b, Bioicons.com under a Creative Commons license CC BY 3.0. Source data
Fig. 3 Different mechanisms contribute to CSP-mediated antifungal resistance. a, Semiquantitative spotting assays with different stress agents. Serial 5-fold dilutions of fungal cell suspensions were spotted onto YPD agar containing chemical agents as indicated. Semiquantitative data were collected by comparing relative growth properties of tested strains. b, Potential mechanisms implicated in CSP-mediated AMB resistance deduced from data shown in a. c, Gene Ontology enrichment for differentially expressed genes in rca1∆ and efg1∆ mutants compared to WT. Data were filtered with cut-off log2FC of ±0.58. Common Gene Ontology terms of rca1∆ and efg1∆ were visualized. d, Spotting assays for CSP mutants on media containing 3 µg ml−1 AMB and 5 mM VitC. e, MIC assays for AMB alone and in combination with the mitochondrial inhibitor AA. The AA concentration ranged from 0 to 1 µg ml−1 (top to bottom) for each strain. Three biological replicates yielded similar results in a, d and e. Credit: icons in b, Bioicons.com under a Creative Commons license CC BY 3.0. Source dataCC BY · PubMed Central
Fig. 4 Targeting mitochondrial cytochrome bc1 enhances AMB susceptibility. a, Deletion of RIP1 encoding the active subunit of the cytochrome bc1 complex increases AMB susceptibility. MIC assays were performed for C. auris in RPMI, YPD and YP–2% glycerol (YPG). b, The cytochrome bc1 inhibitor Inz-5 partially phenocopies the sensitivity of the rip1∆ mutant in a serial dilution spotting assay, in both the presence and absence of AMB. c, Inz-5 reduces the AMBMIC by about 2- to 4-fold across multiple fungal pathogens in checkerboard assays. Two to three biological replicates were performed in a–c. Source data
Fig. 4 Targeting mitochondrial cytochrome bc1 enhances AMB susceptibility. a, Deletion of RIP1 encoding the active subunit of the cytochrome bc1 complex increases AMB susceptibility. MIC assays were performed for C. auris in RPMI, YPD and YP–2% glycerol (YPG). b, The cytochrome bc1 inhibitor Inz-5 partially phenocopies the sensitivity of the rip1∆ mutant in a serial dilution spotting assay, in both the presence and absence of AMB. c, Inz-5 reduces the AMBMIC by about 2- to 4-fold across multiple fungal pathogens in checkerboard assays. Two to three biological replicates were performed in a–c. Source dataCC BY · PubMed Central
Fig. 5 Rca1 and Efg1 are required for skin colonization and fungal fitness. a, Growth curves were recorded at OD600nm for 72 h in MM, YP and YPD medium. b, Nce103 protein levels under different conditions. The NCE103 locus in the R1 strain was tagged with dTomato; fluorescence signals were quantified by flow cytometry. c, Growth properties of C. auris from five clades are promoted by low and high CO2 on MM. Tenfold serial dilutions of C. auris cell suspensions were spotted on MM agar and incubated at 37 °C for 5 days. d, Competitive fitness assays of mutants and the WT strain (R1) following intradermal infections after 3 days (n = 6 or 8 per group). The competitive index was determined using qPCR assays for gDNA isolated from a mixture of C. auris strains (left). For nce103∆ mutants, fitness was assessed by counting of colony-forming units after culturing in ambient air (control nce103∆ does not grow) and in 5.5% CO2. e, Fungal burden in murine biopsies after colonizing intact back skin for 14 days (n = 6 for R1, rca1∆ and efg1∆; n = 10 or 11 for other groups). f, Determination of the fungal burden in skin biopsies after 2 days of colonization on back skin (n = 5 or 6 per group). g, Fungal burden on human skin (9 biopsies from 3 human donors). Fungal suspensions of the WT strain (R1) and the mutants were applied on human skin biopsies. After 24 h, biopsies were washed in PBS, and combined colony-forming units were quantified from washing supernatants and skin tissues, with the total representing the sum of colony-forming units in both fractions. Statistical analyses were performed on average colony-forming units data from each donor. h,i, Representative images of SEM (h) and periodic acid–Schiff staining (i) of human skin topically infected with C. auris for 24 h. Scale bar, 10 µm in h and 100 µm in i. P values were obtained by two-tailed paired t-test in d, one-way ANOVA in e, two-way ANOVA in f, and one- or two-way ANOVA for wash/tissue or total colony-forming units, respectively, in g, followed by Bonferroni or Tukey’s multiple comparisons tests. Box plots indicate the median and interquartile range, with whiskers representing minimum and maximum values using data pooled from at least two independent experiments in d–g. Each data point and exact P < 0.05 are shown in d–g. Panel g (right) created with BioRender.com. Credit: icons in d–g, Bioicons.com under a Creative Commons license CC BY 3.0. Panel g (left) adapted with permission from ref. 53, Springer Nature Limited. Source data
Fig. 5 Rca1 and Efg1 are required for skin colonization and fungal fitness. a, Growth curves were recorded at OD600nm for 72 h in MM, YP and YPD medium. b, Nce103 protein levels under different conditions. The NCE103 locus in the R1 strain was tagged with dTomato; fluorescence signals were quantified by flow cytometry. c, Growth properties of C. auris from five clades are promoted by low and high CO2 on MM. Tenfold serial dilutions of C. auris cell suspensions were spotted on MM agar and incubated at 37 °C for 5 days. d, Competitive fitness assays of mutants and the WT strain (R1) following intradermal infections after 3 days (n = 6 or 8 per group). The competitive index was determined using qPCR assays for gDNA isolated from a mixture of C. auris strains (left). For nce103∆ mutants, fitness was assessed by counting of colony-forming units after culturing in ambient air (control nce103∆ does not grow) and in 5.5% CO2. e, Fungal burden in murine biopsies after colonizing intact back skin for 14 days (n = 6 for R1, rca1∆ and efg1∆; n = 10 or 11 for other groups). f, Determination of the fungal burden in skin biopsies after 2 days of colonization on back skin (n = 5 or 6 per group). g, Fungal burden on human skin (9 biopsies from 3 human donors). Fungal suspensions of the WT strain (R1) and the mutants were applied on human skin biopsies. After 24 h, biopsies were washed in PBS, and combined colony-forming units were quantified from washing supernatants and skin tissues, with the total representing the sum of colony-forming units in both fractions. Statistical analyses were performed on average colony-forming units data from each donor. h,i, Representative images of SEM (h) and periodic acid–Schiff staining (i) of human skin topically infected with C. auris for 24 h. Scale bar, 10 µm in h and 100 µm in i. P values were obtained by two-tailed paired t-test in d, one-way ANOVA in e, two-way ANOVA in f, and one- or two-way ANOVA for wash/tissue or total colony-forming units, respectively, in g, followed by Bonferroni or Tukey’s multiple comparisons tests. Box plots indicate the median and interquartile range, with whiskers representing minimum and maximum values using data pooled from at least two independent experiments in d–g. Each data point and exact P < 0.05 are shown in d–g. Panel g (right) created with BioRender.com. Credit: icons in d–g, Bioicons.com under a Creative Commons license CC BY 3.0. Panel g (left) adapted with permission from ref. 53, Springer Nature Limited. Source dataCC BY · PubMed Central
Fig. 6 C. auris can scavenge CO2 released by bacterial skin colonizers. a, A schematic illustration depicts how urease-positive bacteria supply CO2 to the CSP of C. auris. b, Hospital-acquired bacteria can promote growth of C. auris by releasing CO2 through the urease pathway. Experiments were performed on 12-well plates, where bacterial cell suspensions were spotted in the two middle columns and fungal cell suspensions in both outer columns. Each fungal strain was tested at three levels of 10-fold dilutions. The plates were covered, sealed in plastic bags and incubated at 37 °C for 24 h (Extended Data Fig. 7d). Representative data from at least three independent experiments. Source data
Fig. 6 C. auris can scavenge CO2 released by bacterial skin colonizers. a, A schematic illustration depicts how urease-positive bacteria supply CO2 to the CSP of C. auris. b, Hospital-acquired bacteria can promote growth of C. auris by releasing CO2 through the urease pathway. Experiments were performed on 12-well plates, where bacterial cell suspensions were spotted in the two middle columns and fungal cell suspensions in both outer columns. Each fungal strain was tested at three levels of 10-fold dilutions. The plates were covered, sealed in plastic bags and incubated at 37 °C for 24 h (Extended Data Fig. 7d). Representative data from at least three independent experiments. Source dataCC BY · PubMed Central
Extended Data Fig. 1 Proteomics reveals potential drug-resistant mechanisms. a. PCA analysis of proteomics data from resistant and sensitive strains with / without AMB treatment (3 biological replicates for each strain). b. group GO analysis for differentially abundant proteins (DAPs) using the clusterProfiler package suggests that ion transport (32/92 DAPs) proteins are elevated in the resistant strain. c. The IGV identifies a hotspot S639F mutant variant (green) in Fks1 responsible for caspofungin resistance in R1, R2, 471/P/14R15. d. Volcano plot showing comparative proteomic data between R1 and S strains. e. Integrated heatmap showing the overlap of DAPs with highly expressed genes15,17 between resistant and sensitive C. auris strains. f. Volcano plots showing comparative proteomics data of R1 and S strains under AMB-treatment. Genes in blue bold were functionally validated in this study or by the available literature. Among them, PDR16 and NCE103 contribute to AMBR. Deletion of HPA2, CFL4, SMF12, GCN4, and B9J08_000008 did not alter AMB MICs. Statistical analyses were performed using moderated two-sided t-statistics with the limma-trend method, followed by multiple test correction using the Benjamini–Hochberg (BH) method (d, f). Differential expression analysis of proteomics and RNA-seq datasets used a cut-off ± 0.58 for log2FC and an adjusted p-value < 0.05 (d-f). Source Data
Extended Data Fig. 1 Proteomics reveals potential drug-resistant mechanisms. a. PCA analysis of proteomics data from resistant and sensitive strains with / without AMB treatment (3 biological replicates for each strain). b. group GO analysis for differentially abundant proteins (DAPs) using the clusterProfiler package suggests that ion transport (32/92 DAPs) proteins are elevated in the resistant strain. c. The IGV identifies a hotspot S639F mutant variant (green) in Fks1 responsible for caspofungin resistance in R1, R2, 471/P/14R15. d. Volcano plot showing comparative proteomic data between R1 and S strains. e. Integrated heatmap showing the overlap of DAPs with highly expressed genes15,17 between resistant and sensitive C. auris strains. f. Volcano plots showing comparative proteomics data of R1 and S strains under AMB-treatment. Genes in blue bold were functionally validated in this study or by the available literature. Among them, PDR16 and NCE103 contribute to AMBR. Deletion of HPA2, CFL4, SMF12, GCN4, and B9J08_000008 did not alter AMB MICs. Statistical analyses were performed using moderated two-sided t-statistics with the limma-trend method, followed by multiple test correction using the Benjamini–Hochberg (BH) method (d, f). Differential expression analysis of proteomics and RNA-seq datasets used a cut-off ± 0.58 for log2FC and an adjusted p-value < 0.05 (d-f). Source DataCC BY · PubMed Central
Extended Data Fig. 2 Deletion of genes encoding highly expressed cell surface proteins in resistant strains does not alter susceptibility to antifungal drugs. a. Dose response assays for 5-fluorocytosine (5FC), amphotericin B (AMB), voriconazole (VRC) and caspofungin (CAS) using the CLSI protocol. b. Multiple sequence alignments (MSA) reveal the evolutionary conservation of the bZip domain in Rca1 across yeast species. C. auris genomic sequences of RCA1 were subjected to blastn searches against the indicated species. A reciprocal blastn search was performed to confirm orthologous genes. bZip domains for MSA were predicted with ScanProsite tool (prosite.expasy.org/scanprosite). c. Synteny maps illustrate orthologous relationships of RCA1 and EFG1, including their proximally neighboring genes in C. auris compared to other fungal pathogens. Orthologues and gene order information were retrieved from the Candida Gene Order Browser (cgob.ucd.ie) and fungi.ensembl.org. Source Data
Extended Data Fig. 2 Deletion of genes encoding highly expressed cell surface proteins in resistant strains does not alter susceptibility to antifungal drugs. a. Dose response assays for 5-fluorocytosine (5FC), amphotericin B (AMB), voriconazole (VRC) and caspofungin (CAS) using the CLSI protocol. b. Multiple sequence alignments (MSA) reveal the evolutionary conservation of the bZip domain in Rca1 across yeast species. C. auris genomic sequences of RCA1 were subjected to blastn searches against the indicated species. A reciprocal blastn search was performed to confirm orthologous genes. bZip domains for MSA were predicted with ScanProsite tool (prosite.expasy.org/scanprosite). c. Synteny maps illustrate orthologous relationships of RCA1 and EFG1, including their proximally neighboring genes in C. auris compared to other fungal pathogens. Orthologues and gene order information were retrieved from the Candida Gene Order Browser (cgob.ucd.ie) and fungi.ensembl.org. Source DataCC BY · PubMed Central
Extended Data Fig. 3 The Rca1-Nce103-Efg1 axis controls AMB susceptibility. a, c, d, h. CLSI-based dose-response assays (n = 3 biological replicates) of C. auris strains for AMB (Fig. a, h), CAS, VRC, 5FC (Fig. c) and acetazolamide (ACZ; Fig. d). b. Flocculation phenotypes of C. auris WT and mutants in dH2O after 15 minutes at room temperature. Cells were cultured overnight in liquid YPD, washed twice with dH2O, and adjusted to an OD600nm of 10 prior to the assay. e. Overlay of Nce103 homology models generated using the coordinates for C. albicans Nce103 (PDB: 6GWU), and the Coccomyxa (PDB: 3UCJ) carbonic anhydrase crystal structures. f. The homology model of C. auris Nce103 shows the conservation of the binding pocket for the ACZ carbonic anhydrase inhibitor. g. Relative expression of NCE103 to the WT S strain (n = 3 biological replicates). mRNAs isolated from fungal cells in the exponential growth phase were quantified with qPCR (ACT1 as loading control). Statistical analysis used the one-way ANOVA with Dunnett’s multiple comparisons test, and p values < 0.05 are shown. Each data point is shown, error bars indicate mean ± SD. Source Data
Extended Data Fig. 3 The Rca1-Nce103-Efg1 axis controls AMB susceptibility. a, c, d, h. CLSI-based dose-response assays (n = 3 biological replicates) of C. auris strains for AMB (Fig. a, h), CAS, VRC, 5FC (Fig. c) and acetazolamide (ACZ; Fig. d). b. Flocculation phenotypes of C. auris WT and mutants in dH2O after 15 minutes at room temperature. Cells were cultured overnight in liquid YPD, washed twice with dH2O, and adjusted to an OD600nm of 10 prior to the assay. e. Overlay of Nce103 homology models generated using the coordinates for C. albicans Nce103 (PDB: 6GWU), and the Coccomyxa (PDB: 3UCJ) carbonic anhydrase crystal structures. f. The homology model of C. auris Nce103 shows the conservation of the binding pocket for the ACZ carbonic anhydrase inhibitor. g. Relative expression of NCE103 to the WT S strain (n = 3 biological replicates). mRNAs isolated from fungal cells in the exponential growth phase were quantified with qPCR (ACT1 as loading control). Statistical analysis used the one-way ANOVA with Dunnett’s multiple comparisons test, and p values < 0.05 are shown. Each data point is shown, error bars indicate mean ± SD. Source DataCC BY · PubMed Central
Extended Data Fig. 4 Deletion of NCE103 has marginal effects on lipidomes. Fungal cells in the exponential growth phase were cultured in RPMI medium for 5 or 16 hours, followed by treatment with 1 µg/ml AMB for an additional two hours. Cell pellets were then processed for lipid and sterol analysis. a. Sterol composition analysis with GC-MS from C. auris cells collected in the early and late logarithmic growth phases in RPMI liquid medium. b. Fold-change of lanosterol between mutants and WT strains with/without AMB treatment. Statistical comparisons were conducted between strains within each treatment, and separately between treatments for each strain. Statistics used two-sided Welch’s t-test for each comparison with multiple test correction by Benjamini-Hochberg; exact p values < 0.05 are shown. Error bars indicated mean ± SD based on three biological replicates (A, B). c, d, f. Global lipid profiling of C. auris cultured in RPMI medium with or without AMB treatment (biological replicates from two independent experiments: n = 5 for AMB-nce103∆; n = 7 for no-drug-nce103∆; n = 6 for other groups). Fungal cells were grown in RPMI for 5 hours prior to treatment with 1 µg/mL AMB for 2 hours, followed by lipidomic analysis (see Methods for details). Data represents the relative abundance of each lipid species relative to total lipids. c. Heatmap displays lipid species across samples, scaled as Z-scores. d. Proportion of sphingolipid classes within the total lipid pool. Certain classes were summed up data from their related species. Cer: ceramide; HexCer: hexosylceramide; IPC: inositol phosphorylceramide; MIPC: mannosylinositol phosphorylceramide. The box plot indicates the median and interquartile range, with whiskers representing minimum and maximum values. Two-way ANOVA with Šídák’s correction was used for multiple comparisons across strains and treatments, and exact p values are shown. e. Susceptibility of C. auris to AMB and aureobasidin A (0–1 µg/mL) in RPMI at 37 °C. n = 2 biological replicates. f. The lipidomics dataset was subjected to PCA analysis using the prcomp function in R. Source Data
Extended Data Fig. 4 Deletion of NCE103 has marginal effects on lipidomes. Fungal cells in the exponential growth phase were cultured in RPMI medium for 5 or 16 hours, followed by treatment with 1 µg/ml AMB for an additional two hours. Cell pellets were then processed for lipid and sterol analysis. a. Sterol composition analysis with GC-MS from C. auris cells collected in the early and late logarithmic growth phases in RPMI liquid medium. b. Fold-change of lanosterol between mutants and WT strains with/without AMB treatment. Statistical comparisons were conducted between strains within each treatment, and separately between treatments for each strain. Statistics used two-sided Welch’s t-test for each comparison with multiple test correction by Benjamini-Hochberg; exact p values < 0.05 are shown. Error bars indicated mean ± SD based on three biological replicates (A, B). c, d, f. Global lipid profiling of C. auris cultured in RPMI medium with or without AMB treatment (biological replicates from two independent experiments: n = 5 for AMB-nce103∆; n = 7 for no-drug-nce103∆; n = 6 for other groups). Fungal cells were grown in RPMI for 5 hours prior to treatment with 1 µg/mL AMB for 2 hours, followed by lipidomic analysis (see Methods for details). Data represents the relative abundance of each lipid species relative to total lipids. c. Heatmap displays lipid species across samples, scaled as Z-scores. d. Proportion of sphingolipid classes within the total lipid pool. Certain classes were summed up data from their related species. Cer: ceramide; HexCer: hexosylceramide; IPC: inositol phosphorylceramide; MIPC: mannosylinositol phosphorylceramide. The box plot indicates the median and interquartile range, with whiskers representing minimum and maximum values. Two-way ANOVA with Šídák’s correction was used for multiple comparisons across strains and treatments, and exact p values are shown. e. Susceptibility of C. auris to AMB and aureobasidin A (0–1 µg/mL) in RPMI at 37 °C. n = 2 biological replicates. f. The lipidomics dataset was subjected to PCA analysis using the prcomp function in R. Source DataCC BY · PubMed Central
Extended Data Fig. 5 Mitochondrial function regulates AMB susceptibility. a-b. Dose-response assays of C. auris strains (three biological replicates) for antimycin A (a) and AMB (b) at indicated CO2 concentrations using the CLSI protocol. The nce103∆ and rca1∆ nce103∆ mutants fail to grow in ambient air ( ~ 0.04% CO2). c. Checkerboard assay for AMB and antimycin A for representative strains from clades I to V. FICI: fractional inhibitory concentration index. At least two biological replicates yielded similar results. Source Data
Extended Data Fig. 5 Mitochondrial function regulates AMB susceptibility. a-b. Dose-response assays of C. auris strains (three biological replicates) for antimycin A (a) and AMB (b) at indicated CO2 concentrations using the CLSI protocol. The nce103∆ and rca1∆ nce103∆ mutants fail to grow in ambient air ( ~ 0.04% CO2). c. Checkerboard assay for AMB and antimycin A for representative strains from clades I to V. FICI: fractional inhibitory concentration index. At least two biological replicates yielded similar results. Source DataCC BY · PubMed Central
Extended Data Fig. 6 Vitamin C (VitC) only marginally restores AMBMIC. Dose response assays for AMB in RPMI containing VitC. The nce103∆ mutant fails to grow in ambient air ( ~ 0.04% CO2); pink circles indicate the MIC reading where VitC exhibited protective effects (AR387). Two independent replicates yielded similar results. Source Data
Extended Data Fig. 6 Vitamin C (VitC) only marginally restores AMBMIC. Dose response assays for AMB in RPMI containing VitC. The nce103∆ mutant fails to grow in ambient air ( ~ 0.04% CO2); pink circles indicate the MIC reading where VitC exhibited protective effects (AR387). Two independent replicates yielded similar results. Source DataCC BY · PubMed Central
Extended Data Fig. 7 Carbohydrate metabolism affects fitness of C. auris rca1∆ and efg1∆ mutants. a. Gene set enrichment analysis (GSEA) using gseGO (clusterProfiler) shows upregulation of carbohydrate uptake upon deletion of RCA1 or EFG1. Enrichment Scores (ES) indicate upregulated genes in null mutants. b. Growth kinetics within the first 8 hours of C. auris in different growth media. MM: minimal medium, SCD: synthetic complete dextrose medium; YP: yeast extract-peptone. Detailed media formulations are described in Supplementary Table 3. c. Standard growth curves of C. auris were recorded in RPMI medium after inoculation at either high (OD600nm = 0.2) or low (OD600nm = 0.001) starting cell densities. d. A representative picture illustrates the experimental setup for a bacterial-fungal co-culture assay in a 12-well plate. e. C57BL/6 mice were injected with 2×106 fungal cells through a lateral tail vein; fungal loads in organs were assessed on day five post-infection (n = 5 or 6 or 7 mice per group). f. C57BL/6 female mice (n = 6) were infected via lateral tail vein injection with fungal cells. AMB was administered intraperitoneally every 24 hours on day 1 and 2 post-infection. p-values were calculated by Two-way ANOVA with Tukey’s multiple comparisons test and exact p values < 0.05 are shown (e-f). The box plot indicates the median and interquartile range, with whiskers representing minimum and maximum values; each data point is shown (e-f). g. Number of putative transcription factors regulated by the Rca1 and Efg1; data shows all DEGs with adjusted p-value < 0.05 from RNA-seq datasets of rca1∆ and efg1∆. Source Data
Extended Data Fig. 7 Carbohydrate metabolism affects fitness of C. auris rca1∆ and efg1∆ mutants. a. Gene set enrichment analysis (GSEA) using gseGO (clusterProfiler) shows upregulation of carbohydrate uptake upon deletion of RCA1 or EFG1. Enrichment Scores (ES) indicate upregulated genes in null mutants. b. Growth kinetics within the first 8 hours of C. auris in different growth media. MM: minimal medium, SCD: synthetic complete dextrose medium; YP: yeast extract-peptone. Detailed media formulations are described in Supplementary Table 3. c. Standard growth curves of C. auris were recorded in RPMI medium after inoculation at either high (OD600nm = 0.2) or low (OD600nm = 0.001) starting cell densities. d. A representative picture illustrates the experimental setup for a bacterial-fungal co-culture assay in a 12-well plate. e. C57BL/6 mice were injected with 2×106 fungal cells through a lateral tail vein; fungal loads in organs were assessed on day five post-infection (n = 5 or 6 or 7 mice per group). f. C57BL/6 female mice (n = 6) were infected via lateral tail vein injection with fungal cells. AMB was administered intraperitoneally every 24 hours on day 1 and 2 post-infection. p-values were calculated by Two-way ANOVA with Tukey’s multiple comparisons test and exact p values < 0.05 are shown (e-f). The box plot indicates the median and interquartile range, with whiskers representing minimum and maximum values; each data point is shown (e-f). g. Number of putative transcription factors regulated by the Rca1 and Efg1; data shows all DEGs with adjusted p-value < 0.05 from RNA-seq datasets of rca1∆ and efg1∆. Source DataCC BY · PubMed Central
Extended Data Fig. 8 Differential expression of cell wall- and membrane-related genes in rca1∆ and efg1∆ mutants. a. Volcano plots showing comparative transcriptomics data between R1, rca1∆, efg1∆ strains. Dash lines indicate a cut-off ± 0.58 for log2FC and adjusted p-value < 0.05. b. Heatmap showing the overlap of differentially expressed transporters between WT and mutant strains. Dots (●) indicate DEGs with adjusted p-value < 0.05. Differential expression analysis was conducted using EdgeR, including the two-sided quasi-likelihood F-test (glmQLFTest) approach (a-b). c. Membrane lipid permeability was assessed by an FDA uptake assay. Two-way ANOVA (two-sided) was performed for time × strain, followed by Tukey’s multiple comparison tests using R1 as control group, with exact p values reported (based on data from 3 biological replicates in 1 independent experiment). Error bars indicate mean ± SD. See detail method in Supplementary Information. d. CO2 supports fitness of most Candida pathogens but not of Cryptococcus neoformans (Cryp. neo). Strains used in this experiment: R1::nce103∆; C. albicans SC5314; C. glabrata ATCC 2001; C. glabrata KK01; C. glabrata KK04; C. glabrata KK05; C. krusei ATCC6258; C. parapsilosis ATCC22019; C. dubliniensis KK0908899; C. lusitaniae DSM 70102; C. neoformans NEQAS strain. Source Data
Extended Data Fig. 8 Differential expression of cell wall- and membrane-related genes in rca1∆ and efg1∆ mutants. a. Volcano plots showing comparative transcriptomics data between R1, rca1∆, efg1∆ strains. Dash lines indicate a cut-off ± 0.58 for log2FC and adjusted p-value < 0.05. b. Heatmap showing the overlap of differentially expressed transporters between WT and mutant strains. Dots (●) indicate DEGs with adjusted p-value < 0.05. Differential expression analysis was conducted using EdgeR, including the two-sided quasi-likelihood F-test (glmQLFTest) approach (a-b). c. Membrane lipid permeability was assessed by an FDA uptake assay. Two-way ANOVA (two-sided) was performed for time × strain, followed by Tukey’s multiple comparison tests using R1 as control group, with exact p values reported (based on data from 3 biological replicates in 1 independent experiment). Error bars indicate mean ± SD. See detail method in Supplementary Information. d. CO2 supports fitness of most Candida pathogens but not of Cryptococcus neoformans (Cryp. neo). Strains used in this experiment: R1::nce103∆; C. albicans SC5314; C. glabrata ATCC 2001; C. glabrata KK01; C. glabrata KK04; C. glabrata KK05; C. krusei ATCC6258; C. parapsilosis ATCC22019; C. dubliniensis KK0908899; C. lusitaniae DSM 70102; C. neoformans NEQAS strain. Source DataCC BY · PubMed Central
Extended Data Fig. 9 The skin niche allows for Candida auris colonization and growth. When C. auris colonizes the nutrient-limited skin surface, it encounters ~0.04% CO2 in ambient air. Under such low CO2 levels, C. auris generates HCO3− requiring the carbonic anhydrase Nce103, which in turn is regulated by Rca1 as well as Efg1. The resulting bicarbonate feeds mitochondrial energy metabolism through the tricarboxylic acid cycle (TCA) cycle, as well as signaling processes through the cAMP/PKA/Efg1 pathway. The CSP pathway controls drug resistance and skin colonization by sustaining mitochondrial functions, thus improving fitness and antifungal tolerance. Importantly, the microbiome containing urease-positive bacteria offers additional sources for CO2 to sustain metabolism and C. auris persistence as well as fitness. Figure adapted with permission from ref. 53, Springer Nature Limited, and ref. 54, Springer Nature Limited. Source Data
Extended Data Fig. 9 The skin niche allows for Candida auris colonization and growth. When C. auris colonizes the nutrient-limited skin surface, it encounters ~0.04% CO2 in ambient air. Under such low CO2 levels, C. auris generates HCO3− requiring the carbonic anhydrase Nce103, which in turn is regulated by Rca1 as well as Efg1. The resulting bicarbonate feeds mitochondrial energy metabolism through the tricarboxylic acid cycle (TCA) cycle, as well as signaling processes through the cAMP/PKA/Efg1 pathway. The CSP pathway controls drug resistance and skin colonization by sustaining mitochondrial functions, thus improving fitness and antifungal tolerance. Importantly, the microbiome containing urease-positive bacteria offers additional sources for CO2 to sustain metabolism and C. auris persistence as well as fitness. Figure adapted with permission from ref. 53, Springer Nature Limited, and ref. 54, Springer Nature Limited. Source DataCC BY · PubMed Central

Verification

Claims the paper makes

33 Valid · 4 Not valid · 9 Not checked · 123 Own results and interpretations

83 more claims · 4 with their cited sources checked

Create a free account to see all 169 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 36
  • It may state two separate claims as one. 55
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Show all 87 references

Research works printed but not linked: 2

  1. Trinh, P.-C., Duc-Minh, N.-L., Phuc-Loi, L., Narakorn, K. & Karl, K. Rapid in vitro evolution of flucytosine resistance in Candida auris. mSphere 0, e0097724 (2025).
  2. Trinh, P.-C., Tamires, B. & Karl, K. Gene dosage of PDR16 modulates azole susceptibility in Candida auris. Microbiol. Spectr. 0, e0265924 (2025).

Cited by

See all 16

Metadata

Authors
Phan-Canh, T., Coman, C., Lackner, M., Troppmair, N., Müller, C., Cerbu, D., Seiser, S., Penninger, P., Tsymala, I., Khunweeraphong, N., Bitencourt, T., Knarr, A., Jenull, S., Arzani, H., Zenz, L.-M., Ianiri, G., Chen, W., Chowdhary, A., Mobley, H. L. T., Hartl, M., Moser, D., Ahrends, R., Elbe-Bürger, A., & Kuchler, K.
Affiliations
  • Max Perutz Labs (AT)
  • University of Vienna (AT)
  • Innsbruck Medical University (AT)
  • Ludwig-Maximilians-Universität München (DE)
  • Medical University of Vienna (AT)
  • University of Molise (IT)
  • University of Delhi (IN)
  • University of Michigan (US)
Corresponding author
Kuchler, Karl
Venue
Nature Microbiology
Published
2025-12-23
Volume
11
Issue
2
Pages
461-475
ISSN
2058-5276 (ISSN-L)
OpenAlex
W7116827782
Cited by
16
References
72
Citations by year
2026: 16
FWCI
12.21
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Microbiology (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • Archivio Istituzionale della Ricerca - Università degli Studi del Molise (submittedVersion)
Topics
  • Health Sciences › Medicine › Infectious Diseases › Antifungal resistance and susceptibility
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Fungal and yeast genetics research
  • Health Sciences › Medicine › Infectious Diseases › Fungal Infections and Studies
Keywords
Candida auris, skin colonization, amphotericin B resistance, bicarbonate, mitochondrial bioenergetics, EFG1, urease, antifungal resistance, resensitization
MeSH terms
Candida auris, Amphotericin B, Antifungal Agents, Candidiasis, Carbon Dioxide, Carbonic Anhydrases, Fungal Proteins, Humans, Skin, Tropism, Drug Resistance, Fungal

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-09, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-09 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Candida auris skin tropism and antifungal resistance are mediated by carbonic anhydrase Nce103. https://app.cochat.ai/papers/10.1038/s41564-025-02189-z
The paper
Phan-Canh, T., Coman, C., Lackner, M., Troppmair, N., Müller, C., Cerbu, D., Seiser, S., Penninger, P., Tsymala, I., Khunweeraphong, N., Bitencourt, T., Knarr, A., Jenull, S., Arzani, H., Zenz, L.-M., Ianiri, G., Chen, W., Chowdhary, A., Mobley, H. L. T., . . . Kuchler, K. (2025). Candida auris skin tropism and antifungal resistance are mediated by carbonic anhydrase Nce103. Nature Microbiology. https://doi.org/10.1038/s41564-025-02189-z

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.