Gene Duplication Associated with Increased Fluconazole Tolerance in Candida auris cells of Advanced Generational Age

Candida auris is an emerging multi-drug resistant yeast that causes systemic infections. Here we show that C. auris undergoes replicative aging (RA) that results from asymmetric cell division and causes phenotypic differences between mother and daughter cells similar to other pathogenic yeasts. Importantly, older C. auris cells (10 generations) exhibited higher tolerance to fluconazole (FLC), micafungin, 5- flucytosine and amphotericin B compared to younger (0–3 generation) cells. Increased FLC tolerance was associated with increased Rhodamine 6G (R6G) efflux and therapeutic failure of FLC in a Galleria infection model. The higher efflux in the older cells correlated with overexpression of the efflux pump encoding gene CDR1 (4-fold). In addition, 8-fold upregulation of the azole target encoding gene ERG11 was noted in the older cells. Analysis of genomic DNA from older cells by qPCR indicates that transient gene duplication of CDR1 and ERG11 causes the observed age-dependent enhanced FLC tolerance in C. auris strains. Furthermore, older cells exhibited a thickened cell wall, decreased neutrophil killing (24% vs 50%), increased epithelial cell adhesion (31.6% vs 17.8%) and upregulation of adhesin protein Als5p. Thus, this study demonstrates that transient gene duplication can occur during RA, causing increased FLC tolerance in old C. auris cells.

Both the FLC-resistant and FLC-susceptible C. auris isolates were subjected to a Rhodamine 6-G (R6G) efflux assay. R6G is a fluorescent dye, the influx and efflux of which is extensively used to study the activities of fungal ABC transporters, which significantly contribute to FLC resistance [17][18][19] . Increased R6G efflux was observed in FLC-resistant isolates (Fig. 1a, red lines) when compared to the R6G efflux in FLC-susceptible isolates (Fig. 1a, blue lines), whereas import was comparable despite the thickened cell wall (Fig. 2a). Next, virulence of all isolates, including seven FLC-resistant (S3, S4, S5, S6, S8, S9 and S10) and four FLC-susceptible (SB, S1, S2, and S7), was assessed in a Galleria melonella infection model. Most C. auris isolates killed more than 50% of the larvae within 4 days at an inoculum dose of 10 5 yeast cells (Fig. 1b). Two isolates (FLC-susceptible isolate S1 and FLC-resistant isolate S3), exhibited decreased virulence in Galleria (Fig. 1b), when compared to the virulence of FLC-resistant isolate S9 (*p < 0.05 by log-rank test). Virulence of yeast is commonly associated with its ability to adhere to the host cells 20 . Consequently, we assessed adherence of all 11 C. auris isolates to HeLa cells. Variable adherence was observed among the C. auris isolates, with isolate SB adhering the least (~3%) and isolate S2 adhering the most (~40%) to HeLa cells (Fig. 1c).
Prior data from our lab showed that Galleria virulence and adherence are altered by replicative aging in C. glabrata and Cryptococcus neoformans 9,11 . To investigate age-related phenotypes in C. auris, the RLS of the 11 clinical isolates had to be first determined. All the isolates showed variable RLS, with median RLS ranging from 18 generations (S5) to 50 generations (S6) (Fig. 1d). Impressive variability of life span, also referred to as stochasticity of RLS, was also noted within same isolates, for example, in strain S10, RLS ranged from 14 to 65 generations. With increase in generational age, the doubling time of the cells increased significantly (Fig. 1e). For example, the doubling time of 1-8 generatiosn old cells from isolate S1 was initially 60 mins but increased to 90 mins in cells aged between 9-30 generations and 150 min in cells above 30 generations old (Fig. 1e). . Error bars represent Standard deviation and One-Way Anova was done to determine the significance. Doubling times significantly changed between Gen 1 and Gen 10 cells, Gen 1 and Gen 20, and Gen 1 and Gen 30 (*p < 0.05).

Figure 2.
Increased Virulence of 10 Gen Old C. auris Cells from FLC-Susceptible S1: (a) Cell size and Cell wall thickness of the old (10 Gen) cells were significantly greater compared to young (0-3 Gen) cells. Cell wall thickness (n = 23) and Cell Size (n = 150) were measured by ImageJ software from the microscopic images of individual young and old cells. Student's t-test was performed to analyze the significance (*p < 0.001). Transmission Electron Microscopic (TEM) image of the old (10 Gen) cells of C. auris showing increased cell wall thickness compared to the young (0-3 Gen) cells. (b) Percent Neutrophil-mediated Killing of young (0-3 Gen: black bar) and old (10 Gen: red bar) cells of C. auris. The assay was done in triplicate and error bars signify standard deviation. Student's t-test with Welch's correction was performed to determine the significance; *p = 0.0266. (c) Percent adhesion of young (0-3 Gen: black bar) and old (10 Gen: red bar) cells of C. auris. The assay was done in triplicate and error bars signify standard deviation. Student's t-test with Welch's correction was performed to determine the significance; *p = 0.0291. (d) qPCR analysis to study the expression of ALS5 in young and old cells. ACT1 gene was used as an internal control and the data was normalized to gene expression in young cells. The assay was done in triplicate and error bars signify standard deviation. Multiple t-test was performed using Holm-Sidak method to analyze the significance; *p = 0.016. (e) Virulence of young (0-3 Gen) and old (10 Gen) cells of C. auris in Galleria. 20 larvae were infected separately with 10 5 young and old cells. Log rank test was used to determine the statistical significance.
www.nature.com/scientificreports www.nature.com/scientificreports/ phenotypic variations associated with replicative aging in C. auris. C. auris isolate S1 was chosen to study age-related phenotypic variations between mother and daughter cells. The isolate S1 has a median RLS of 27.5 generations (range 14-42 generations). Important replicative age related phenotypic variations are thickened cell wall and increased cell size that were observed in older generation cells of C. neoformans 10 , C. glabrata cells 9 and in S. cerevisiae 21 . To test whether these occur in C. auris, C. auris clinical isolate S1 cells were aged to 10 generations (defined here as "old cells") and transmission electron microscopy was performed. Images of old cells demonstrated significant increase in cell wall thickness as well as significant increase in cell size (Fig. 2a). To test whether this increased thickness and size shields old cells from neutrophil-mediated killing, we performed a neutrophil-mediated killing assay using young C. auris and old C. auris cells. Young cells were killed significantly more than 10 generation old cells (50.17% vs 24.45% respectively, Fig. 2b). Further, old cells (10 generation) adhered more to the HeLa cells when compared to young cells (31.6% vs 17.83%, Fig. 2c). Increased adherence may result from increased expression of C. albicans homolog agglutinin-like sequence protein-encoding gene ALS5, which was 6-fold up-regulated in old C. auris cells (Fig. 2d). Als5p is an adhesin protein present in C. auris that was previously reported to aid in biofilm adhesion in C. auris 22 . Additionally, increased adherence is often associated with increased virulence, and we observed this in comparing survival of old cells versus young cells in the context of Galleria infection (Fig. 2e). Most importantly, old C. auris cells from FLC-susceptible isolate S1 also acquired significant tolerance to antifungals AMB, MF, 5-FC and FLC ( Fig. 3a-d).
Possible molecular mechanisms of increased FLC tolerance in old C. auris cells. It is conceivable that increased cell wall thickness in old C. auris cells could affect FLC import. However, old C. glabrata were shown to have increased efflux pump activities and FLC resistance, suggesting age-associated increase in efflux may also play a role in C. auris FLC resistance 13 . We employed the R6G assay to assess efflux pump activities in young and old cells. These data demonstrated comparable influx of RG6 (data not shown), but significant increase in efflux of R6G in the old cells when compared to the young cells (Fig. 3e). R6G is a substrate of ABC transporter Cdr1p, which is linked with azole resistance in Candida sp. The expression of both CDR1 along with the FLC target encoding ERG11 was observed to be upregulated by 4-fold and 8-fold respectively in old C. auris cells isolated from FLC-susceptible strain S1 (Fig. 3f). Increased expression of both CDR1 (16-fold) and ERG11 (16-fold) was also observed in the old cells isolated from FLC-resistant C. auris isolate S9 (Fig. 4a). Consistent with these upregulations old cells of S9 also exhibited even more pronounced increased R6G efflux compared to young population (Fig. 4b).

Gene duplication in old cells may cause increase expression of drug resistant genes.
Although gain-of-function mutations in their respective transcription factors could possibly increase expression of CDR1 and ERG11 23 , this mechanism could not explain upregulation of these genes in old cells, since age-related phenotypes are not inherited by the young cells. Another, mechanism by which CDR1 and ERG11 expression could be induced in old C. auris cells is transient duplication of gene copies in the old cell's genome. To test this hypothesis, genomic DNAs from both young and old cells from a FLC-sensitive and FLC resistant strains were isolated and subjected to qPCR. Three sets of oligonucleotides (Table S1) were used to span the entire CDR1 and ERG11 gene. The oligos were designed in such a manner that they span the entire genes as shown in Fig. 5a. qPCR analysis with these oligos demonstrated that the copy numbers of the CDR1 and ERG11 gene doubled in both FLC-sensitive and FLC-resistant strains as evident in Fig. 5b,c that show the results from one oligonucleotide set. Amplification with two other oligonucleotide sets yielded comparable results in both the strains (data not shown). To further investigate the cause of the observed gene duplication, we compared ploidy and DNA content of old and young cells by flow cytometry. No change in total DNA content of young and old cells were observed (Fig. S2).
Besides increased expression of ERG11 and CDR1, increased expression of adhesin encoding gene ALS5 was also observed in old cells when compared to young cells (Fig. 2d). ALS5 is an adhesion protein whose function is not associated with ERG11 and CDR1. Of note is that although there was an increased expression of ALS5, duplication of this gene in the genome was not observed in older C. auris cells (Fig. S1).
Most importantly, gene duplication was limited to the older mother cells and disappeared when the old cells were allowed to replicate into a younger cell population (Fig. S3). The old cells were grown into fresh media for 3 h (two doublings). After growth, young and old cells were isolated, and qPCR was performed by obtaining genomic DNA from the young cells obtained by re-growing the 10 generation old cell.
The relevance of this finding was further explored in the host environment. Specifically, Galleria was infected with old and young C. auris cells and treated with FLC or PBS. Galleria infected with young C. auris population when treated with FLC survived longer than the Galleria infected with young C. auris and treated with PBS only (Fig. 6a). In contrast, no difference in survival was observed between FLC treated or untreated Galleria when Galleria was infected with old C. auris. Here treated or untreated C. auris exhibited comparable survival (Fig. 6b). Additionally, we assessed whether old cells were resilient in Galleria. Galleria larvae were infected with either young or old cells from isolate S1, and haemolymph was extracted from the infected larvae at different time points to assess initial clearance. Initially, larvae infected with young cells were able to clear a portion of the young cells from haemolymph whereas there was no clearance seen in the haemolymph of larvae infected with old cells. In fact, old cells were able to replicate faster than they were cleared from the haemolymph (Fig. 6c).

Discussion
This study is the first to investigate replicative aging in multi-drug resistant C. auris clinical isolates and yields important results that inform our understanding about the emergence of azole resistance in this important pathogenic yeast. C. auris constitutes a threat to the medical community as the majority of clinical isolates obtained from patients are resistant to multiple anti-fungals 4 . Indeed 70% of the isolates used in this study exhibited elevated MICs to FLC. Additionally, FLC resistance emerges rapidly in patients being treated with azoles 24 . Why this www.nature.com/scientificreports www.nature.com/scientificreports/ species rapidly emerged as multidrug resistant is not understood and may not only be the result of exposure to azole prophylaxis. Resistant C. auris isolates have increased R6G efflux when compared to the susceptible isolates in our collection. R6G is frequently used to study efflux activities of ABC-transporters, and increased efflux significantly corelates with azole resistance in various Candida sp. 18,19,25 . Similarly, our data is supported by other data that also suggests that increased efflux activities cause the observed FLC resistance in C. auris 2 . The assays were done in triplicate and error bars signify standard deviation. Multiple t-Test was performed using Holm-Sidak method to analyze the significance; *p < 0.05; (e) R6G efflux assay performed on young (black line) and old (red line) in triplicate and error bars signify standard deviation. Student's t-test with Welch's correction was performed to determine the significance; *p < 0.05. (f) qPCR analysis to study the expression of CDR1 and ERG11 in young and old cells. ACT1 gene was used as an internal control. The data was normalized to the gene expression in young cells. The assay was done in triplicate and error bars signify standard deviation. Multiple t-test was performed using Holm-Sidak method to analyze the significance; *p < 0.05. (2019) 9:5052 | https://doi.org/10.1038/s41598-019-41513-6 www.nature.com/scientificreports www.nature.com/scientificreports/ Besides FLC, C. auris isolates exhibit high MICs to 5-FC, an antifungal which is at times used synergistically with FLC to treat candidiasis 26 . In this study we found that C. auris undergoes replicative aging similar to Cryptococcus neoformans and C. glabrata with increasing doubling times over time 9 . As expected, C. auris strains of different clonal backgrounds also exhibit variable RLS, consistent with the observed RLS variability of other yeast strains 9,11 .
During the process of replicative aging in Cryptococcus neoformans and C. glabrata, phenotypic traits emerge that promote selection of older cells and cause them to accumulate and persist in the hosts during chronic   Table S1. Genomic DNA was isolated from both old (10 generation) cells and young (0-3 generation) cells of FLC-sensitive strain S1 (b) and FLC-resistant strain S9 (c). Data shows the gene copy numbers of CDR1 and ERG11 using oligo sets P7-P8 (CDR1) and P5-P6 (ERG11) ACT1 was used as a control for the experiment and the data was normalized to the gene copy number of young cells. The assay was performed in triplicate and error bars signify standard deviation. Multiple t-test was performed using Holm-Sidak method to analyze the significance; *p < 0.05. www.nature.com/scientificreports www.nature.com/scientificreports/ infections. These older cells are more resilient to antifungal and neutrophil killing 9 . Hence, we were interested in investigating if replicative-aging-related resilient phenotypes also emerge in C. auris. Older C. auris cells exhibit thickened cell wall and increased adhesion to HeLa cells. C. auris adhesin protein Als5p shares 34% homology with N-terminal end of C. albicans Als5p, which is the extracellular binding portion of the protein 27 . Our data demonstrates 8-fold increased expression of C. auris ALS5 in the older population. Als5p in C. auris has recently been described as an important adhesin protein that aids in biofilm adhesion 22 . Als5p belongs to agglutin-like sequence family of protein and exhibits high homology to Als3p, the latter is an important adhesin protein, which has been incorporated in anti-candida vaccine 28 , where antibodies against Als3p predict vaccine efficacy in vivo 29 . Besides this, increased adherence in older C. auris cells to host cells supports the concept that aging involves remodeling of the cell wall, which contributes to the increased virulence of older C. auris cells in vivo in the Galleria infection model.
We found 80% of 10 generation old cells from FLC-sensitive isolates survived at FLC concentrations of 256 µg/ml and 64 µg/ml respectively, although the 0-3 generation young daughter cells of the FLC-sensitive isolates exhibited FLC MICs around 1 µg/ml. This is consistent with our observation that FLC treatment failed to cure in the Galleria infection model experiment when older C. auris cells were administered. Increased tolerance to FLC in the older C. auris cells may result from two mechanisms. First, old cells are retained significantly more in the haemolymph of Galleria, and second, the combination of overexpression of drug resistant genes CDR1 and ERG11, both of which were observed. Older C. auris cells exhibited increased efflux of R6G, which is a substrate for Cdr1p ABC transporter. This is consistent with the increased ABC transporter mediated efflux observed in old C. glabrata cells 13 . Therefore, drugs such as Sir2p agonist that modify the RLS of a C. auris population should be investigated because studies suggest that they constitute a potential class of novel antifungal drug targets, which enhance antifungal efficacy by shifting lifespan and vulnerability of the fungal population 16 .
Cdr1p and Erg11p mediated drug resistance is extensively studied in C. albicans 18,19,30 . Both Cdr1p and Erg11p can be induced by acquiring gain-of-function mutations in the transcription factors Tac1p and Upc2p respectively in response to antifungal stress 31,32 . Our results strongly suggest that a transient gene duplication occurs www.nature.com/scientificreports www.nature.com/scientificreports/ and the increased copy number of CDR1 and ERG11 results in the increased expression of CDR1 and ERG11 in the older C. auris cells. Gene copy numbers were analyzed by qPCR, which can reliably quantify gene copy numbers 33 . Three sets of oligonucleotides were used in two independent strains in this study to span the entire CDR1 and ERG11 genes. These data show that whole CDR1 and ERG11 genes are duplicated in older cells and more importantly only transiently, as the gene duplication is lost when the older cells are allowed to replicate to produce young daughter cells. It is conceivable that the stress of continuous replicative aging in cells of advanced generational age may select for the observed duplication event. Consistent with that interpretation is the data in S. cerevisiae, which demonstrates that the RLS in young cells obtained from old cells is reset and that young cells do not inherit materials from old mother cells 34 . Whether this gene duplication is a result of a chromosomal duplication event cannot be concluded from our data. At this point the available whole genome sequencing data in C. auris is not granular enough to determine chromosome numbers and assign genes to individual chromosome locations. However, it is known that aneuploidy can, despite being a severe burden on the cell, confer a selective advantage. Specifically, in C. glabrata as well as in C. neoformans aneuploidy has been associated with hetero-resistance to FLC 35 . In C. albicans, CDR1 and ERG11 are located in two different chromosomes (chromosomes 3 and 5 respectively). Without the precise knowledge of chromosomal locations of CDR1 and ERG11 we cannot know if these two genes are on the same chromosome in C.auris. It is noteworthy that ALS5 gene copy does not change with aging. Causal relationship between genomic instability and replicative aging has been suggested [36][37][38] , but still remains incompletely understood. Most data to date has investigated genomic instability at the ribosomal DNA (rDNA) locus in the process of aging 39 . Chromatin structure constitutes a key modulator of aging in yeast 40 and previous data indicates global loss of histones from all regions of the yeast genome during replicative aging. Loss of histones results in an open chromatin conformation that is associated with globally increased gene expression and accumulation of DNA breaks 41 . Defects in homologous recombination can lead to elevated levels of DNA breaks, translocations, loss of heterozygosity (LOH), and amplifications, in older yeasts 42 . Further studies need to be done in C. auris to establish chromosome numbers, and extent of genomic instability and gene duplications that occurs in replicative aging, and it will be important to determine at what generational age it occurs, and how many generations it takes to correct the duplications. Azole resistance via chromosomal duplication are extensively studied in C. albicans, C. glabrata and Cryptococcus neoformans [43][44][45] , but they have not been linked to replicative aging. For example, the left arm of chromosome 5 that contains TAC1 and ERG11 duplicates causing increased azole resistance in C. albicans 43 .
Besides increased FLC resilience, old C. auris cells showed increased tolerance to MF similar to the observed phenotype in old C. glabrata population 13 . MF targets cell wall, and the thickened cell wall in the older population of C. auris cells may increase the target drug ratio and explain the only slightly increased MF resistance, which may not be clinically relevant. Additionally, increased tolerance to AMB and 5FC was also observed in old C. auris cells consistent with the findings in the old C. glabrata population 13 . In C. glabrata, old cells exhibit decreased ergosterol content, which may explain increased AMB resilience in old cells 9,13 and hence future studies will have to determine, the ergosterol content in old C. auris cells.
This study demonstrates for the first time that replicative aging results in transient duplication of genes that are relevant for FLC resistance. It will be important to acquire complete chromosome maps and further investigate this age-related genomic instability and assess if it remains permanent in some daughter cells. As other traits also promote selection of older cells this epigenetic phenotype of drug resistance could be highly relevant and may have contributed to the rapid emergence of FLC resistance in this species even without extensive exposure to azoles.

Materials and Methods
strains and culture conditions. 11 C. auris strains were obtained from CDC with proper Material Transfer Agreement (MTA). The strains were cultured in Yeast Peptone Dextrose (YPD) media (Difco) and incubated at 37 °C. 30% glycerol stocks were made for each strain and stored in −80 °C.
Antifungal susceptibility testing. Minimum Inhibition Concentration (MIC) assay was performed on all the 11 C. auris strains following previously published protocol 46 . Briefly, the C. auris strains were cultured in YPD at 37 °C for overnight, and 0.1 OD of the cultured cells were used as inoculum for MIC the next day. Antifungals fluconazole (FLC), 5-flucytosine (5-FC), amphotericin B (AMB), and micafungin (MF) were two-fold serially diluted in RPMI 1640 media (GE Life Sciences) with 25 mM HEPES (pH 7.0) in a flat bottom 96-well plate (Costar). The plates were incubated for 48 h at 37 °C without shaking. In the plate, one column without any antifungal and one column without cells were used as controls for the assay. MICs were performed in biological triplicate and the average value of the triplicate is reported.
Replicative life span (RLS). RLS was performed as described previously in C. glabrata and Saccharomyces cerevisiae 9,47 . Briefly, 20 virgin cells were arrayed in a straight line in YPD plates, and new buds from the virgin cells were separated using a 25 µm needle (CoraStyles) under a tetrad dissection microscope (Zeiss) at 250X magnification. The plates were incubated at 37 °C at the end of each budding event, and budding was monitored every 1 h. The cells that did not bud for 24 h were considered dead.

Isolation of old and young cells.
Isolation of old and young cells were performed as previously described 9,11 . Briefly, C. auris cells from isolates S1 and S9 were grown in Synthetic Dextrose (SD) media (6.7 g Bacto Yeast Nitrogen Base without aminoacids and 20 g glucose/litre) overnight. Next day, 0.1 OD of overnight cells were exponentially grown for 8 h, and 10 8 cells were washed three times with 1X PBS and labelled with 4 mg/ml of Biotin (Sulfo-NHS-LC-Biotin, Pierce, Rockford, IL) for 30 mins at room temperature. Next, the cells were washed three times in 1X PBS, and were grown for 7 generations (~7-8 h) in 50 ml SD media in 250 ml www.nature.com/scientificreports www.nature.com/scientificreports/ flask. After growth, the cells were washed three times with 1X PBS, and counted in hemocytometer. All the cells were resuspended at a concentration of 10 7 cells per 90 µl of 1X PBS and incubated with 10 µl of streptavidin conjugated magnetic microbeads (Miltenyi Biotec, Auburn, CA) per 10 7 cells for 30 mins at 4 °C. After incubation, the cells were washed three times in 1X PBS and resuspended at a concentration of 10 8 cells per 500 µl of 1X PBS. The cells were subjected to LS magnetic column (Miltenyi Biotec), which retained the cells conjugated with biotin-streptavidin in presence of a magnetic field and allowed the non-labelled cells to pass through. The biotin-streptavidin labelled cells were recovered by removing the magnetic field. This fraction containing the 7 generation old cells, was then washed three times in 1X PBS and grown in SD media for three more generations (~3-4 h). After growth, the cells were subjected to LS column as described above to isolate 10 generation old cells. In this publication old cells refers to 10 generation old cells. At 10 generations S1 cells have completed about 36% of their median RLS (27.5 generations). The purity of the old cell population was confirmed by staining the streptavidin labelled cells with Fluorescein Isothiocyanate (FITC), and the population was at least 90% pure. transmission electron microscopy. Young and old cells were isolated from C. auris isolate S1 and resuspended in 5% glutaraldehyde and fixed for 2 h at room temperature. Imaging was done by Stony Brook University microscope facility following previously published protocol 9 . Cell wall thickness (n = 23) and Cell Size (n = 150) were measured by ImageJ software from the microscopic images of individual young and old cells.
Antifungal killing assay. Antifungals FLC, 5-FC, AMB, and MF were serially diluted in 100 µl RPMI 1640 media at concentrations of 256 to 1 µg/ml, 8 to 0.5 µg/ml, 0.15 to 0.01 µg/ml, and 0.15 to 0.01 µg/ml respectively. Both young and old C. auris cells from isolate S1 were isolated as described above and resuspended in RPMI media. 100 µl of 10 4 cells/ ml were plated in flat bottom 96-well plate containing 100 µl of serially diluted antifungals, to achieve a final cell concentration of 5 × 10 3 cells/ml in each well. The plates were incubated for 2 h at 37 °C without shaking. After incubation 100 µl of drug treated cells were plated in 25 ml YPD agar plates and incubated at 37 °C for 48 h. Cells without antifungal treatment were used as controls. The percent killing was calculated by comparing the number of colony forming units (CFUs) of antifungal treated cells and the number of CFUs in the control (non-treated cells). The experiment was performed in triplicate.
Neutrophil-mediated killing assay. The neutrophils were isolated as described previously 48 . The neutrophil-mediated killing assay was performed as before 49 . Briefly, 10 5 neutrophils were diluted in RPMI 1640 media and incubated with 1% human serum for opsonization and were incubated for 1 h at 37 °C with 5% CO 2 in flat-bottom 96 well plate. Young and old (10 gen) cells from C. auris isolate S1 were isolated and added to the wells containing neutrophils at a fungus to neutrophil ratio of 1:10. The wells with no neutrophils were used as controls. The plates were incubated for 1 h at 37 °C with 5% CO 2 to allow phagocytosis and then were washed three times with RPMI 1640 media. One set of neutrophil-fungal cells after phagocytosis was immediately lysed with sterile water and plated in YPD media to use as a phagocytic control. These cells were subjected to no killing by neutrophils. Other set of neutrophil-fungal cells were incubated for 2 h at 37 °C with 5% CO 2 to initiate neutrophil killing. After 2 h, the neutrophils were lysed and 100 µl of the reaction mixture was plated in YPD media and incubated at 37 °C for 48 h. Percent killing was calculated by comparing the CFUs between 2 h neutrophil killed cells and non-killed cells. The experiment was performed in triplicate.
Infection model. Galleria mellonella infection model was carried out as previously described using Hamilton Syringe 9,50 . G. mellonella caterpillars in the final instar larval stage were obtained (Vanderhorst Wholesale Inc., St. Mary's, Ohio) and were stored in the dark. The larvae were used within 7 days from the day of shipment. Larvae employed in all assays were randomly chosen. Briefly, 10 5 young and old (10 gen) from isolate S1, or exponentially growing C. auris cells from all isolates were injected into each larva. For each condition, 20 larvae were used, and larvae injected with only 1X PBS was used as control. Additionally, in a separate experiment, infected larvae were treated with 1 µg/ml of FLC, on the same day of infection to study the effects of FLC treatment in vivo. In an independent experiment, retention of young and old C. auris cells in haemolymph of Galleria larvae was analyzed as previously described 9 by infecting the larvae with 5 × 10 4 young and old cells from isolate S1. Haemolymph was extracted after 0, 0.5, 2, 4, 8 and 24 h of infections, plated on YPD agar plates, and CFUs were counted after 48 h incubation at 37 °C. For each time point, haemolymph from 15 (n = 15) larvae was collected.
Briefly, 10 8 young and old (10 gen) cells from isolate S1 or exponentially growing cells of all isolates were starved for 2 h in presence of 5 mM deoxy glucose (Sigma-Aldrich) at 37 °C with shaking and were washed three times with 1X PBS. R6G (Sigma-Aldrich) was then added to the washed cells at a final concentration of 10 µM and incubated for 30 mins at 37 °C with shaking to initiate R6G import. After incubation, the cells were washed three time in 1X PBS, and efflux was initiated by adding 2% glucose. Samples were collected every 30 mins, and the fluorescence of the supernatants were measured at excitation and emission wavelengths of 525 nm and 555 nm respectively. The assay was performed in triplicate.
Analysis of gene duplication and gene expression with qPCR. RNA was isolated from young and old (10 gen) C. auris cells from both isolates S1 and S9 using Qiagen RNEASY Kit following manufacturer's protocol and quantified in a nano drop (BioSpectometer, Eppendorf). A260/280 ratio of 2.0 or above was considered as pure RNA.
For gene expression quantification, 450 ng of RNA was subjected to DNAse (ThermoScientific) treatment following manufacturer's protocol. qPCR was performed using Power SYBR ® Green RNA-to-CT ™ 1-Step Kit (Applied Biosystem) following manufacturer's protocol. Oligonucleotides used for qPCR are listed in Table S1.
www.nature.com/scientificreports www.nature.com/scientificreports/ Housekeeping gene ACT1 was used as a control for qPCR, and the data was normalized to the gene expression in the young cells. 2 −ΔΔCt method was used to calculate the fold change 51 .
For gene copy number quantification, genomic DNA was isolated from both young and old C. auris cells from both isolates S1 and S9 by using previously published protocol 52 . Only 50 ng of DNA was used to analyze gene copy number with qPCR using Power SYBR Green (Applied Biosystem) following manufacturer's guidelines. ACT1 gene was used as a control. 2 −ΔΔCt method was used to calculate the change in copy number 51 . Cell adhesion assay. Cell adhesion assay was performed as previously described 53 . Briefly, HeLa cells were grown in DMEM (GE Life Sciences) media supplemented with 10% Fetal Bovine Serum (FBS), and 10 4 cells were added into flat-bottom 96 well plate and were incubated at 37 °C with 5% CO 2 for 1 h for the HeLa cells to adhere to the wells. After incubation, 10 4 young and old (10 gen) from isolate S1 or exponentially growing C. auris cells from all isolates were added on to the HeLa cells and incubated at 37 °C with 5% CO 2 for 1 h for the yeast cells to adhere to the HeLa cells. The non-adherent C. auris cells were removed by washing five times with fresh DMEM media. After washing, the adherent C. auris cells were recovered by lysing the HeLa cells with 0.1% Triton X100 and plating in YPD media. The plates were incubated at 37 °C for 48 h and CFUs were counted. As control, 10 4 C. auris cells were directly plated in the YPD media and incubated at 37 °C for 48 h and CFUs were counted. Percent adherence was calculated by comparing the CFUs of the HeLa treated cells with the control. The assay was done in triplicate.
Flow Cytometry (FACs) Analysis of DNA Content. DNA content of both young and old C. auris cells (S1) were analyzed as described before 54 . Briefly, the young and the old cells were subjected to permeabilization and fixation via addition of 1 ml chilled 70% ethanol for 2 h at 4 °C. The fixed cells were then washed three times with FACs buffer (0.2 M Tris-HCl (Sigma-Aldrich), pH 7.4, 20 mM EDTA (Sigma-Aldrich) and incubated for 2 h at 37 °C with 1 mg/ml RNase A (Sigma-Aldrich). After RNase treatment, the cells were washed three times with 1X PBS and stained overnight at 4 °C with 100 µl of Propidium Iodide (PI, Sigma Aldrich), at a final concentration of 50 µg/ml of PI. After adding 900 µl of 1X PBS the next day, FACs analysis of the DNA content was done using a FACSCalibur analyzer (Becton-Dickinson). A total of 50,000 cells of both young and old population were analyzed.
statistics. All statistics were done in Graph Pad Prism 6.0. The specific statistical test performed are listed in the figure legends.

Data Availability
All data required to evaluate and understand the article are included. If additional data is required, the data are available on request from the correspondence author.