Low level of antifungal resistance of Candida glabrata blood isolates in Turkey: Fluconazole minimum inhibitory concentration and FKS mutations can predict therapeutic failure.

BACKGROUND: Candida glabrata is the third leading cause of candidaemia in Turkey; however, the data regarding antifungal resistance mechanisms and genotypic diversity in association with their clinical implication are limited.
OBJECTIVES: To assess genotypic diversity, antifungal susceptibility and mechanisms of drug resistance of C glabrata blood isolates and their association with patients' outcome in a retrospective multicentre study. PATIENTS/
METHODS: Isolates from 107 patients were identified by ITS sequencing and analysed by multilocus microsatellite typing, antifungal susceptibility testing, and sequencing of PDR1 and FKS1/2 hotspots (HSs).
RESULTS: Candida glabrata prevalence in Ege University Hospital was twofold higher in 2014-2019 than in 2005-2014. Six of the analysed isolates had fluconazole MICs ≥ 32 µg/mL; of them, five harboured unique PDR1 mutations. Although echinocandin resistance was not detected, three isolates had mutations in HS1-Fks1 (S629T, n = 1) and HS1-Fks2 (S663P, n = 2); one of the latter was also fluconazole-resistant. All patients infected with isolates carrying HS-FKS mutations and/or demonstrating fluconazole MIC ≥ 32 µg/mL (except one without clinical data) showed therapeutic failure (TF) with echinocandin and fluconazole; seven such isolates were collected in Ege (n = 4) and Gulhane (n = 3) hospitals and six detected recently. Among 34 identified genotypes, none were associated with mortality or enriched for fluconazole-resistant isolates.
CONCLUSION: Antifungal susceptibility testing should be supplemented with HS-FKS sequencing to predict TF for echinocandins, whereas fluconazole MIC ≥ 32 µg/mL may predict TF. Recent emergence of C glabrata isolates associated with antifungal TF warrants future comprehensive prospective studies in Turkey.

INTRODUCTION

Candida glabrata is the second leading cause of candidaemia in USA, Canada, Australia and some Scandinavian countries , , , , , and the first cause of candidaemia in intensive care units and patients with haematological malignancies and solid tumours. , Compared to the other Candida species, C glabrata is much more tolerant to antifungals, which allows it to rapidly develop antifungal resistance during the course of antifungal therapy. , , , , , , , , , Indeed, C glabrata isolates resistant to azoles or echinocandins and even those demonstrating multidrug resistance are increasingly being identified in clinical settings. , , It was shown that resistance to azoles and increase of virulence in C glabrata is mostly caused by gain‐of‐function mutations in the yeast Zn2Cys6 transcription factor PDR1, which lead to the overexpression of efflux pumps, whereas resistance to echinocandins is associated with non‐synonymous mutations in hotspots (HSs) 1 and 2 of the FKS1 and FKS2 genes encoding glucan synthases. Regarding the clinical prognosis, some studies indicate that sequencing of HSs in FKS1/2 more accurately predicts therapeutic failure (TF) of echinocandins than phenotypic antifungal susceptibility testing (AFST). ,

Candida glabrata is an endogenous opportunistic pathogen normally residing in the human gastrointestinal tract and causing bloodstream infections in immunocompromised hosts. However, some molecular typing studies indicate that a possibility of horizontal transfer, suggesting that clonal enrichment of fluconazole‐resistant C glabrata isolates cannot be excluded. , Furthermore, it has been shown that some C glabrata genotypes are associated with a higher mortality rate, , reinforcing the importance of strain profiling using genotyping techniques in clinical practice.

Candida glabrata is the third leading cause of candidaemia in Turkey, where it shows a low level of antifungal resistance as evidenced by a recent multicentre candidaemia study (1997‐2017). However, the correlation of important clinical parameters and microbiological properties such as genotypic diversity and molecular mechanisms underlying azole and echinocandin resistance has not been investigated. The current retrospective multicentre study was conducted to address these gaps in the knowledge regarding C glabrata blood isolates in Turkey.

MATERIALS AND METHODS

Isolates, growth conditions, and identification

Non‐duplicate C glabrata blood isolates (n = 107) recovered from patients with candidaemia (n = 107) were collected in five clinical centres in Turkey including Ege (n = 54), Gulhane (n = 25), Dokuz Eylul (n = 18), Selcuk (n = 6) and Istanbul (n = 4) University Hospitals. Ege University Hospital with 1,800 beds was the largest institution, followed by Istanbul University Hospital, Gulhane Training and Research Hospital and Dokuz Eylul University Hospital (1,100 beds each), and Selcuk University Hospital (900 beds). Hospitals included in the current study typically use echinocandins as the first‐line therapy for treatment of candidaemia due to C glabrata. Isolates from Ege and Gulhane hospitals were collected from 2005– to 2019, whereas those from the other centres were collected from 2015– to 2019.

Isolates were cultured on Sabouraud dextrose agar (Merck, Darmstadt, Germany) plates for 48 h at 35°C and further verified by growth in CHROMAgar Candida medium (CAC, Becton Dickinson) to ensure their purity. Identification was performed by internal transcribed spacer (ITS) rDNA sequencing using ITS1 and ITS4 primers. Persistence of fever and isolation of C glabrata from blood cultures despite antifungal treatment were considered as TF.

AFST

All isolates (n = 107) were tested for drug sensitivity using the broth microdilution protocol suggested by CLSI M27‐A3/S4. , The following drugs were used: fluconazole (FLZ), voriconazole (VRZ), itraconazole (ITZ), amphotericin B (AMB) (all from Sigma‐Aldrich), micafungin (MCF; Astellas, Munich, Germany) and anidulafungin (AND; Pfizer); caspofungin (CSP) was not included in AFST experiments because of interlaboratory variations. Plates were incubated at 35°C for 24 h, and drug minimum inhibitory concentrations (MICs) were determined by visual examination, and Candida parapsilosis (ATCC 22 019) and Candida krusei (ATCC 6258) were used for quality control purposes. FLZ resistance was scored at the MIC ≥ 64 µg/mL, and lower MIC values were considered to define susceptible‐dose dependent isolates. MCF‐ and AND‐resistant isolates were defined at the MICs ≥ 0.25 µg/mL and ≥0.5 µg/mL, respectively. Resistance to VRZ, ITZ and AMB was reported based on epidemiological cut‐off values, and MICs > 0.5 µg/mL, 2 µg/mL and 2 µg/mL, respectively, were considered to indicate non‐wild‐type isolates.

Sequencing of PDR1 and HS1/HS2 of FKS1 and FKS2

PCR amplification and sequencing of the PDR1 gene and HS1/2 regions of the FKS1/2 genes were performed as previously described. , Sequences were assembled and edited using SeqMan Pro software (DNASTAR) and aligned using MEGA 7.0. The genome of C glabrata CBS 138 (http://www.candidagenome.org) was used as a wild‐type reference.

Multilocus microsatellite genotyping (MMT)

The genotypic diversity of C glabrata isolates was evaluated by MMT based on three markers: MT1, RPM2 and ERG3. Isolates that differed in a single locus among the six alleles tested were considered to belong to the same genotype. Bionumerics software v7.6 (Applied Math, Sint‐Martens‐Latem, Belgium) was used for data analysis and dendrogram construction by the unweighted‐pair group method using average linkages.

Statistical analysis

The data were statistically evaluated using SPSS v.24 (SPSS Inc). The two‐tailed chi‐square test and logistic regression were used to analyse the association between patient's outcome (death or survival) and clusters obtained by MMT. As each of the numerous identified genotypes was identified for only few isolates, clusters comprising similar genotypes were used to increase statistical power. To assess the association between clusters and FLZ MIC values, which were not normally distributed, a non‐parametric Kruskal‐Wallis test was used. Statistical significance was defined at α < 0.05.

Availability of sequencing data

All sequences generated for PDR1 and HSs of FKS1 and FKS2 were submitted to GenBank (https://www.ncbi.nlm.nih.gov/genbank/) under the following accession numbers: MN985836‐MN985942 (PDR1), MN985943‐MN986049 (HS1‐FKS1), MN986050‐MN986156 (HS2‐FKS1), MN986157‐MN986262 (HS1‐FKS2), and MN986263‐MN986369 (HS2‐FKS2).

Ethics statement

The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval (number 20‐2T/30) has been received.

RESULTS AND DISCUSSION

In this study, we performed comprehensive evaluation of C glabrata blood isolates collected in five centres in Turkey; analysis included genotypic diversity, antifungal susceptibility and molecular features defining resistance to azoles and echinocandins. We found a low level of antifungal resistance among isolates, which is consistent with a previous study performed in Turkey, and showed that sequencing together with AFST could provide more reliable data to guide clinicians in their choice of treatment for patients with C glabrata candidaemia. Together with similar studies conducted in China, , USA, South Korea, India and Iran, our study should contribute to better understanding of clinical and microbiological profiles of C glabrata bloodstream isolates.

Candidaemia was equally prevalent in men (n = 53) and women (n = 53) (data missing for one patient). The median age of the patients was 60 years (range, 0‐87 years) (data missing for 12 patients), which is consistent with the fact that elderly patients are more prone to developing C glabrata candidaemia. , The vast majority of the patients were hospitalised in intensive care units (n = 28; 26.2%) and surgical wards (n = 25; 23.4%) followed by other wards (n = 54; 50.4%). The antifungal treatment data were unavailable for 46 patients; based on the available data, echinocandins were used in 39.2% of cases (n = 42; for 11 patients in combination with other antifungals), AMB in 14% (n = 15; for eight patients in combination) and azoles in 8.4% (n = 9; for three patients in combination). This statistics is in contrast with that in Iran and India, where the majority of patients with candidaemia are treated with azoles. The clinical outcome data were unavailable for 11 patients; based on the available data, the mortality rate was calculated as 55.1% (59/107), which is close to those observed in Austria and Germany but much higher than those in Iran and the USA. Interestingly, the number of C glabrata isolates in Ege University Hospital almost doubled during 2015‐2019 (n = 35) compared to 2005‐2014 (n = 19). A similar increasing trend was reported in other studies , and was linked to the disproportionate use of antifungals, which, however, we could not prove because of the scarcity of antifungal treatment data for the 2005‐2014 period.

Antifungal resistance was observed only for FLZ (n = 4, MIC ≥ 64 µg/mL); furthermore, FLZ MIC = 32 µg/mL was observed for two isolates (Table 1 and Table S1). All 107 isolates exhibited the susceptibility of the wild type for VRZ, ITZ and AMB and two isolates (G53 and G103) showed intermediate susceptibility to MCF and AND (0.125 µg/mL and 0.25 µg/mL, respectively). The low rate or apparent absence of antifungal resistance observed in our study is consistent with a previous multicentre candidaemia study conducted in Turkey as well as reports from several other Asian, , , , , South American and European , , , countries; however, it is in contrast with the data from USA, where echinocandin resistance in C glabrata is a major public health problem.

TABLE 1

MICs of antifungal agents used against 107 Candida glabrata isolates

Antifungal agent	Minimum inhibitory concentration (µg/ml)													MIC50	MIC90	GM
	0.016	0.032	0.06	0.12	0.25	0.5	1	2	4	8	16	32	64
Fluconazole						1	23	29	47	1		2	4	8	64	2.86
Voriconazole		106			1									0.032	0.125	0.032
Itraconazole		52	39	11	5									0.125	0.25	0.125
Micafungin	103	2		2										0.016	0.016	0.063
Anidulafungin	1	73	29	2	2									0.03	0.06	0.063
Amphotericin B				2	8	91	5	1						0.5	1	1

Two isolates showing intermediate phenotype for micafungin (0.125 µg/mL) and anidulafungin (0.25 µg/mL) became resistant against both antifungals after 48 incubation (G53; 0.5 µg/mL and G103; 2 µg/mL).

PDR1 sequencing showed that only two isolates (1.9%) were wild type; the rest harboured mutations leading to changes in the protein sequence (Figure 1, Table 2, and Table S1), most of which (87.6%, 348/397) occurred between the inhibitory and middle homology domains of Pdr1p (Figure 1). Among the changes exclusively found in FLZ‐resistant isolates, the KLTQAVN insertion between residues 226 and 227 has been previously reported, whereas mutations F439I + D554E + P927R, P695R and N768I were detected for the first time (Figure 1, Tables 2 and 3, and Table S1). One of the two isolates with the FLZ MIC = 32 µg/mL harboured a unique novel mutation (L281V) (Figure 1, Tables 2 and 3, and Table S1). Recent studies indicate that UPC2, a zinc cluster transcription factor, may contribute to FLZ resistance in C glabrata, , which may explain the high FLZ MIC for isolate G51 (32 µg/mL) without any exclusive amino acid changes in Pdr1p expected for this phenotype (Tables 2 and 3, and Table S1). Although AFST did not reveal echinocandin resistance in vitro, previously reported mutations, including S629T (in isolate G29 susceptible to both MCF and AND) and S663P (in isolates G53 and G103 intermediate for susceptibility to MCF and AND), were detected in HS1‐Fks1 and HS1‐Fks2, respectively (Table 3). As S633P has been associated with high MIC values for echinocandins, we repeated the AFST of MCF and AND for the three isolates harbouring FKS1 and FKS2 mutations and obtained consistent results, suggesting that the S663P substitution could be present in isolates not resistant to echinocandins. Interestingly, one FLZ‐resistant isolate (G103) simultaneously harboured the KLTQAVN insertion in Pdr1p and the S663P mutation in HS1‐Fks2p (Tables 2 and 3, and Table S1). None of the isolates carried mutations in HS2 regions of FKS1/2.

FIGURE 1

Mutations in the Pdr1p protein of Candida glabrata isolates. Amino acid changes observed in the isolates with high FLZ MIC values (≥32 µg/mL) are highlighted in orange colour. BD, DNA‐binding domain; MHD, middle homology domain; ID, inhibitory domain; and AD, activator domain

TABLE 2

FLZ MICs for isolates carrying Pdr1p mutations, MICs ≥ 64 µg/mL was considered to indicate FLZ resistance

Amino acid change	Resistance %	Fluconazole minimum inhibitory concentration (µg/mL)								Total
		0.5	1	2	4	8	16	32	64
Wild type	0			1	1					2
S76P, V91I, L98S, T143P, 226‐Ins_KLTQAVN‐227 a	100								1	1
S76P, V91I, L98S, T143P, P695R	100								1	1
S76P, V91I, L98S, T143P, N768I	100								1	1
S76P, V91I, L98S, T143P, F439I, D554E, E590D, P927R	100								1	1
L98S, V91I, D243N, L281V, E590D	0							1		1
S76P, V91I, L98S, T143P	0		11	9	25			1		46
S76P, V91I, L98S, T143P, V582A, E590D	0					1				1
S76P, V91I, L98S, T143P, E590D	0		5	4	4					13
V91I, L98S, T143P, E590D	0			1	2					3
T143P, E590D	0		3	4	2					9
T143P	0	1		5	2					8
V91I, L98S, D243N	0		1	1	6					8
T143P, D243N, E590D	0				1					1
S76P, V91I, L98S, T143P, I380L, K704N	0				1					1
S76P, V91I, L98S, T143P, E590D, N791Y	0				1					1
S76P, V91I, L98S, T143P, D810E, Y811N	0				1					1
S76P, V91I, T143P	0				1					1
S76P, V91I, L98S, T143P, S316I	0			1						1
S76P, V91I, L98S, T143P, M774I, V775L	0			1						1
V91I, T143P, E590D	0			1						1
T143P, E590V	0			1						1
T143P, D243N	0			2						2
T143P, E590D, R593P	0		1							1
S76P, V91I, L98S, T143P, E590V	0		1							1
Total	4/107	1	22	30	47	1		2	4	107

Seven amino acids were inserted between amino acids 226‐227.

TABLE 3

Clinical characteristics of patients infected with Candida glabrata isolates showing FLZ MIC values ≥ 32 µg/mL and/or harbouring HS1‐Fks1/2 mutations

Patient #	Age/sex (y)	Underlying diseases	Risk factors	Prophylaxis/Empiric	MAF	AAF	Outcome	C/G	Pdr1p AAC	Fksp AAC	MIC (µg/mL)
											FLZ	MCF	AND	AMB
G6	NA (2006)	NA	NA	NA	NA	NA	Died	C5/G23	P695R	WT	64	0.016	0.032	0.5
G29	56/M (2017)	Chronic viral hepatitis B, diabetes mellitus, atrophic left kidney	Abdominal and liver abscesses, CVC, BSAT	FLZ (first dosage 800 mg/d followed by 400 mg/d for 32 d), AND (unknown dosage for 7 d) → TF and PF	Ambisome (3 mg/kg for 20 d)	NO	Died	C2/G11	NSAAC	S629T (HS1‐Fks1)	1	0.016	0.064	0.5
G51	60/M (2019)	Pancreatic cancer and chronic gastritis	CVC, BSAT	FLZ (first dosage 800 mg/d followed by 400 mg/d for 13 d) → TF and PF	AND (200 mg/d for 4 d)	NO	Died	C7/G28	NSAAC	WT	32	0.016	0.064	0.5
G53	63/M (2019)	Diabetes mellitus, chronic obstructive pulmonary disease, hypertension, acute atrial fibrillation	Pleural puncture, PICVC, SVC, BSAT	MCF (100 mg/d for 3 d) → PF and TF but MCF was not changed	MCF (100 mg/d for 82 d)	NO	Died	C8/G30	NSAAC	S663P (HS1‐Fks2)	4	0.125	0.25	0.5
G55	76/M (2016)	Ovarian cancer, cardiac problems, hypertension, chronic renal failure and bacterial infection	Cholecystectomy, CVC, BSAT	FLZ (200 mg/d for 31 d) → TF and PF	Ambisome (3 mg/kg) + AND (100 mg/d) for 13 d	NO	Survived	C8/G32	N768I	WT	64	0.016	0.064	0.5
G98	13/F (2017)	Acute myeloid leukaemia	Bone marrow transplantation, CVC, BSAT	Posaconazole (200 mg/d, for 3 d)	Caspofungin (50 mg/d for 40 d)	NO	Died	C8/G30	L281V	WT	32	0.016	0.032	0.5
G103	22/F (2018)	Acute myeloid leukaemia	JVC, BSAT	Caspofungin (70 mg/d for 6 d) → TF and PF	FLZ (200 mg d for 4 d) → TF and PF	AMB (200 mg/d first dosage and continued with 3 mg/kg for 7 d) → TF and PF → AMB (3 mg/kg) + voriconazole (dosage unknown) for 7 d	Died	C1/G1	226‐Ins_KLTQAVN‐227	S663P (HS1‐Fks2)	64	0.125	0.25	0.5
G107	78/M (2019)	Acute renal failure, upper gastrointestinal bleeding, and pneumonia	FC, UC, PICVC, BSAT	NO	FLZ (400 mg d for 3 d) → TF and PF	AMB (200 mg/d for 2 d)	Died	C3/G15	F439I, D554E, P927R	WT	64	0.016	0.064	0.5

Main treatment was defined as the first‐choice antifungal therapy followed by blood culture; persistent fever was defined as TF despite antifungal therapy (prophylactic or main). Alternative antifungal treatments was provided in case of TF of the main treatment. Pdr1p mutations included only those occurring in the isolates with FLZ MIC ≥ 32 µg/mL. Risk factors do not include previous exposure to antifungals, which is mentioned separately.

Abbreviations: AA, Amino acid change; AAF, Alternative antifungal used due to therapeutic failure; AMB, Amphotericin B; AND, Anidulafungin; BSAT, Broad‐spectrum antiobiotic therapy; C/G, Cluster/Genotype; CVC, Central venous catheter; FC, Femoral catheter; FLZ, Fluconazole; JVC, Jugular venous catheter; MAF, Main antifungal; MCF, Micafungin; MIC, Minimum inhibitory concentration; NA, Not available; NSAAC, No specific amino acid change; PF, Persistent fever; PICVC, Peripherally inserted central venous catheter; SVC, Subclavian venous catheter; TF, Therapeutic failure; UC, Urine catheter; WT, Wild type.

We next evaluated factors potentially associated with TF in patients infected with isolates harbouring mutation in HS1 of FKS1/FKS2 and those with FLZ MIC ≥ 32 µg/mL. Our data indicate that TF occurred in patients with isolates carrying HS1‐FKS mutations and those with FLZ MIC ≥ 32 µg/mL (n = 8; Table 3). Although S663P and S629T were not associated with echinocandin resistance in vitro, both of them corresponded to TF of MCF, AND and CSP (Table 3). This finding confirmed the notion that FKS sequencing is a more reliable approach to predict treatment outcome than phenotypic assays and that AFST alone may be misleading in the selection of appropriate antifungal therapy. However, some echinocandin‐resistant C glabrata isolates harbour mutations outside of the HS regions ; therefore, the combination of AFST and HS‐FKS sequencing may more accurately predict echinocandin TF than either techniques alone. Consistent with previous studies, , our results indicated that development of abscesses and empiric/prophylactic treatment with echinocandins were associated with mutations in HS regions and echinocandin TF. Furthermore, we found that the isolates with FLZ MIC ≥ 32 µg/mL, which is below the clinical breakpoint of 64 µg/mL recommended by CLSI and their respective mutations, could be associated with FLZ TF (Table 3). Considering that diverse mutations were detected throughout the entire Pdr1p sequence (Figure 1) and that one of the isolates with FLZ MIC = 32 µg/mL did not harbour any mutations in Pdr1p, AFST was more efficient in predicting FLZ TF compared to PDR1 sequencing. Isolate #G103, which simultaneously harboured mutations in PDR1 and FKS2, was responsible for TF with all azoles and echinocandins used. Among the eight isolates associated with TF, seven were detected in Ege (n = 4) and Gulhane (n = 3) hospitals; among these isolates, six were recovered between 2016– and 2019, including three recovered in 2019. Out of the eight patients with TF who were infected with isolates showing FLZ MIC ≥ 32 µg/mL and/or carrying FKS mutations, seven (87.5%) died (Table 3 and Appendix S1). Collectively, these data indicate the predictive potential of FLZ MIC ≥ 32 µg/mL for FLZ TF and of FKS sequencing combined with AFST data for echinocandin TF. However, it should be noted that TF cannot be exclusively attributed to microbiological characteristics of the isolates; other factors may be involved, including serum concentration of the antifungal which shows patient‐to‐patient variations, highlighting the importance of therapeutic drug monitoring to attain a favourable clinical outcome. Moreover, considering that all patients with azole/echinocandin‐resistant C glabrata isolates had a catheter inserted (except one patient infected with isolate #G6 without clinical data), it is plausible that catheter removal may have implications on clinical outcome.

MMT analysis revealed 34 genotypes and 10 clusters (Figure 2, Figure S1, and Table S1). As some isolates were clonal and/or belonged to the same genotype, horizontal transfer could be suggested; however, this hypothesis requires experimental confirmation by performing whole genome sequencing and environmental screening, which are beyond the scope of our study. Nowadays, various next‐generation sequencing platforms have been increasingly employed to assess genotypic diversity as well as to identify mutations responsible for antifungal resistance, which may not be used for a particular gene but rather for numerous genes scattered throughout the genome. , , In contrast to a previous study, in our study we did not observe the phenomenon of clonal enrichment for FLZ‐resistant C glabrata isolates as evidenced by the lack of statistical association between FLZ MIC values and cluster and MMT patterns (Figure 2). Moreover, statistical analysis did not reveal any link between genotype/cluster and mortality (Appendix S1, statistical analysis section), which, however, was detected in previous studies. ,

FIGURE 2

Minimum spanning tree illustrating the lack of clonal enrichment for isolates with FLZ MIC ≥ 32 µg/mL and/or mutations in HS1 of FKS1/2. All eight isolates, except for one without clinical data (marked red), showed azole and/or echinocandin TF

In conclusion, we performed the first analysis of clinical and microbiological characteristics of C glabrata isolates from Turkish patients with candidaemia and updated the AFST data on a multicentre scale. Although the rate of antifungal resistance in vitro was low, TF was common and mostly observed in recent years. Fks mutations and FLZ MIC ≥ 32 µg/mL were predictive of echinocandin and FLZ TF, respectively.

This study was limited by its retrospective nature, which accounted for incomplete clinical data. Moreover, although it was a multicentre study, almost 50% of the isolates were from one institution (Ege University Hospital). Therefore, prospective comprehensive multicentre studies should be conducted in the future to more accurately determine the burden of antifungal resistance and its association with the clinical profile of C glabrata‐infected patients in Turkey. It should also be noted that there were no repetitive isolates, which may ultimately result in underestimation of antifungal resistance. The same is true for mutations found in PDR1, which warrants future studies that should examine the expression of efflux pumps and determine if the identified mutations directly confer azole resistance. Other potentially relevant genes such as MSH2 could be sequenced in azole/echinocandin‐resistant and susceptible isolates to clarify their role in the sensitivity of C glabrata to antifungal drugs.

CONFLICT OF INTEREST

The authors report no conflicts of interest. The authors alone are responsible for the content and writing.

AUTHOR CONTRIBUTIONS

Amir Arastehfar: Conceptualization (equal); data curation (equal); formal analysis (equal); investigation (equal); methodology (equal); project administration (equal); supervision (equal); validation (equal); visualization (equal); writing‐original draft (equal); writing‐review and editing (equal). Farnaz Daneshnia: Data curation (equal); investigation (equal); methodology (equal); project administration (equal); supervision (equal); validation (equal); writing‐review and editing (equal). Mohamad Salehi: Data curation (equal); formal analysis (equal); investigation (equal); resources (equal); writing‐review and editing (equal). Melike Yaşar: Data curation (equal); investigation (equal); resources (equal); writing‐review and editing (equal). Tuğrul Hoşbul: Data curation (equal); investigation (equal); writing‐review and editing (equal). Macit Ilkit: Conceptualization (equal); data curation (equal); investigation (equal); project administration (equal); resources (equal); supervision (equal); validation (equal); writing‐review and editing (equal). Weihua Pan: Conceptualization (equal); funding acquisition (equal); investigation (equal); methodology (equal); project administration (equal); supervision (equal); validation (equal); writing‐review and editing (equal). Ferry Hagen: Formal analysis (equal); visualization (equal); writing‐review and editing (equal). Nazlı Arslan: Data curation (equal); resources (equal); writing‐review and editing (equal). Hatice Türk‐Dağı: Data curation (equal); resources (equal); writing‐review and editing (equal). Süleyha Hilmioglu‐Polat: Conceptualization (equal); data curation (equal); investigation (equal); methodology (equal); project administration (equal); resources (equal); supervision (equal); validation (equal); writing‐review and editing (equal). David S. Perlin: Conceptualization (equal); funding acquisition (equal); methodology (equal); project administration (equal); supervision (equal); writing‐review and editing (equal). Cornelia Lass‐Floerl: Conceptualization (equal); data curation (equal); investigation (equal); methodology (equal); project administration (equal); resources (equal); supervision (equal); validation (equal); writing‐review and editing (equal).

Supplementary Material

Click here for additional data file.