Antifungal Susceptibility Profiles and Resistance Mechanisms of Clinical Diutina catenulata Isolates With High MIC Values

Diutina catenulata (Candida catenulata) is an ascomycete yeast species widely used in environmental and industrial research and capable of causing infections in humans and animals. At present, there are only a few studies on D. catenulata, and further research is required for its more in-depth characterization and analysis. Eleven strains of D. catenulata collected from China Hospital Invasive Fungal Surveillance Net (CHIF-NET) and the CHIF-NET North China Program were identified using matrix-assisted laser desorption ionization–time of flight mass spectrometry and internal transcribed spacer sequencing. The antifungal susceptibility of the Diutina catenulata strains was tested using the Clinical and Laboratory Standards Institute broth microdilution method and Sensititre YeastOne™. Furthermore, ERG11 and FKS1 were sequenced to determine any mutations related to azole and echinocandin resistance in D. catenulata. All isolates exhibited low minimum inhibitory concentration (MIC) values for itraconazole (0.06–0.12 μg/ml), posaconazole (0.06–0.12 μg/ml), amphotericin B (0.25–1 μg/ml), and 5-flucytosine (range, <0.06–0.12 μg/ml), whereas four isolates showed high MICs (≥4 μg/ml) for echinocandins. Strains with high MIC values for azoles showed common ERG11 mutations, namely, F126L/K143R. In addition, L139R mutations may be linked to high MICs of fluconazole. Two amino acid alterations reported to correspond to high MIC values of echinocandin, namely, F621I (F641) and S625L (S645), were found in the hot spot 1 region of FKS1. In addition, one new amino acid alteration, I1348S (I1368), was found outside of the FKS1 hot spot 2 region, and its contribution to echinocandin resistance requires future investigation. Diutina catenulata mainly infects patients with a weak immune system, and the high MIC values for various antifungals exhibited by these isolates may represent a challenge to clinical treatment.

Diutina catenulata (Candida catenulata) is an ascomycete yeast species widely used in environmental and industrial research and capable of causing infections in humans and animals. At present, there are only a few studies on D. catenulata, and further research is required for its more in-depth characterization and analysis. Eleven strains of D. catenulata collected from China Hospital Invasive Fungal Surveillance Net (CHIF-NET) and the CHIF-NET North China Program were identified using matrix-assisted laser desorption ionization-time of flight mass spectrometry and internal transcribed spacer sequencing. The antifungal susceptibility of the Diutina catenulata strains was tested using the Clinical and Laboratory Standards Institute broth microdilution method and Sensititre YeastOne™. Furthermore, ERG11 and FKS1 were sequenced to determine any mutations related to azole and echinocandin resistance in D. catenulata. All isolates exhibited low minimum inhibitory concentration (MIC) values for itraconazole (0.06-0.12 mg/ml), posaconazole (0.06-0.12 mg/ml), amphotericin B (0.25-1 mg/ml), and 5flucytosine (range, <0.06-0.12 mg/ml), whereas four isolates showed high MICs (≥4 mg/ ml) for echinocandins. Strains with high MIC values for azoles showed common ERG11 mutations, namely, F126L/K143R. In addition, L139R mutations may be linked to high MICs of fluconazole. Two amino acid alterations reported to correspond to high MIC values of echinocandin, namely, F621I (F641) and S625L (S645), were found in the hot

INTRODUCTION
In recent years, Candida infections have been on the rise worldwide (Sanguinetti et al., 2015;Pristov and Ghannoum, 2019). Although Candida albicans remains the main causative agent of these infections, the rate of non-C. albicans candidial infections is increasing (Sanguinetti et al., 2015). Candida infections are not usually multidrug-resistant, but there are notable cases of drug resistance among non-albicans Candida (NAC) species. Candida auris shows in vitro multidrug resistance and is associated with outbreaks. Candida glabrata is the most common cause of candidemia and is resistant to azoles and echinocandins. Candida parapsilosis is another notorious pathogen isolated from patients, which causes outbreaks and multidrug resistance (Arastehfar et al., 2021). These Candida species, though rare, are clinically common and necessitate a better understanding of their pathogenic mechanisms. Diutina catenulata (C. catenulata), an ascomycete, can colonize the digestive tract of animals and humans and cause superficial or deep infections (Crozier, 1977;Radosavljevic et al., 1999;Ha et al., 2018;Cafarchia et al., 2019). Diutina catenulata is usually utilized in the production of industrial products (Joo et al., 2008;Subramanya et al., 2017;Babaei and Habibi, 2018;Cafarchia et al., 2019). Diutina catenulata belongs to the CTG-Ser clade and is closely related to Saccharomyces cerevisiae (O'Brien et al., 2018). The first time describing D. catenulata associated with disease in humans in 1977, and D. catenulata was recovered from the nail of a 50-year-old male Australian patient (Crozier, 1977). Subsequently, the first case of candidemia was diagnosed in a patient with cancer (Radosavljevic et al., 1999).
There are few reports of human infection with D. catenulata, and only a small number of studies are available to guide clinical treatment. Diutina catenulata generally exhibits antifungal sensitivity (Radosavljevic et al., 1999;Ha et al., 2018); however, some strains showing resistance to azoles and echinadines have been isolated from eggs and feces (Glushakova et al., 2021). Thus, it is important to consider the potential antifungal resistance of D. catenulata. However, the resistance mechanism of D. catenulata is poorly understood. Amino acid alternations in Erg11p and Fks1p are responsible for causing azole and echinocandin resistance in pathogenic Candida spp. (Berkow and Lockhart, 2017;Toutounji et al., 2019). However, the complete open reading frames of both ERG11 and FKS1 genes have not been revealed, and the gene polymorphisms of ERG11 and FKS1 in clinical D. catenulata isolates remain unknown.
Despite the three cases reported worldwide, the occurrence of D. catenulata in clinical specimens of the Chinese population remains unelucidated. Thus, we studied the antifungal susceptibility and potential drug resistance mechanisms of the clinical isolates of D. catenulata obtained in China over a period of 9 years.

Ethics Statement
This study was approved by the Human Research Ethics Committee of Peking Union Medical College Hospital (No. S-263). A written informed consent was obtained from all study participants to examine the isolates cultured from them for scientific research.

Isolates
During the period from 2010 to 2018 (Table 1), 11 D. catenulata isolates were collected from seven different hospitals in five provinces from the China Hospital Invasive Fungal Surveillance Net (CHIF-NET) and the CHIF-NET North China Program. These isolates were mainly from invasive fungal infection specimens. Most specimens were obtained from patients with bloodstream infections, which indicates the seriousness of the situation for these patients.

Identification
All isolates were identified at the species level using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) conducted with Autof MS 1000 (Autobio, Zhengzhou, China) and Vitek MS (Bio Merieux, Marcy-l'Étoile, France). The species identification was confirmed via the sequencing of the rDNA internal transcribed spacer (ITS) region (ABI 3730XL, Thermo Fisher Scientific, Cleveland, OH, USA). PCR and sequencing of the amplicons were performed using the forward primers, V9G and ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′), and the reverse primers, LS266 and ITS4 (5′-TCCTCCGCTT ATTGATATGC-3′) (Zhang et al., 2014;Hou et al., 2016). The phylogenetic tree of D. catenulata was constructed by alignment with the ITS gene sequences of the common Candida species in the NCBI gene library. Maximum-likelihood phylogenetic trees were constructed with IQ-TREE using an ultrafast bootstrap approximation approach with 1,000 replicates (Trifinopoulos et al., 2016).

Antifungal Susceptibility Testing
Minimum inhibitory concentrations (MICs, mg/ml) of nine antifungal agents, namely, anidulafungin, micafungin, caspofungin, fluconazole, posaconazole, voriconazole, itraconazole, amphotericin B, fluorocytosine, were determined for all isolates using the Sensititre YeastOne ™ system (SYO, Trek Diagnostic Systems, Thermo Fisher Scientific, Cleveland, OH, USA) according to the manufacturer's instructions. MICs of micafungin (Astellas, Japan), caspofungin (CAS; Merck, USA), fluconazole [(National Institute for Food and Drug Control (NIFDC), China], voriconazole (NIFDC, China), amphotericin B (NIFDC, China), and flucytosine (NIFDC, China) were determined using the CLSI broth microdilution method. MICs were determined after 36 h of culture. Candida krusei ATCC 6258 and C. parapsilosis ATCC 22019 were used as quality control strains. When the MIC obtained by the two methods fell within a twofold dilution gradient, the essential agreement (EA) between Sensititre ™ YeastOne ™ and CLSI was considered for D. catenulata. The clinical breakpoints of antifungals against D. catenulata in vitro have not yet been established by the Clinical and Laboratory Standards Institute (CLSI) or the European Committee on Antimicrobial Susceptibility Testing (EUCAST). Consequently, the interpretative criteria were followed to test the in vitro susceptibility of Candida spp. according to the CLSI M59 guidelines (CLSI, 2017).

DNA Extraction and Sequencing of ERG11 and FKS1
On comparing published amino acid sequences of ERG11 (3641571, 1466526) and FKS1 (3639844, 856398) with D. catenulata CBS 565 whole-genome sequences (PJEZ00000000.1) and conducting gene annotation, we identified the ERG11 and FKS1 homologous proteins in C. catenulata. Amino acid alignment of Erg11p and Fks1p from published data and whole-genome sequencing of analyzed data is shown in Supplementary Figures 1, 2. ERG11 and FKS1 Sequencing Genomic ERG11 and FKS1 were amplified by PCR using specific primers and sequenced as previously described. The forward (5′-ACATTATTTATTGCCCCATG-3′) and reverse primers (5′-GCAAGTATCCCGCTTTTCCC-3′) for ERG11 and those for FKS1 are shown in Supplementary Table 1.

Nucleotide Sequence Accession Numbers
The ITS region sequences of strains found in this study were deposited in GenBank with accession numbers MW624477 to MW624482.

Literature Review
This literature review considered the available data regarding the susceptibility of the D. catenulata species to antifungals. The literature search was performed on September 3, 2021, using the following three databases: PubMed (https://pubmed. ncbi.nlm.nih.gov), Web of Science (https://webofknowledge. com), and Embase (https://www.embase.com). The terms "Diutina catenulates" or "Candida catenulate" or "Candida ravautii" were entered in the category of "Title/Abstract" in the PubMed Advanced Search Builder, and "TS=(Diutina catenulata)" or "TS=(Candida catenulata)" or "TS=(Candida ravautii)" was entered in the Web of Science databases. The search in Embase was conducted in the advanced search area, including the terms "'diutina catenulate':ti,ab,kw" or "'candida catenulata':ti,ab,kw" or "'candida ravautii':ti,ab,kw." Studies describing the patients infected with D. catenulata were selected and summarized.

RESULTS
Species Identification of D. catenulata Using MALDI-TOF and DNA Sequencing All 11 clinical isolates were identified as D. catenulata by the Autof MS 1000 and Vitek MS. The ITS sequences of the study isolates exhibited 100% sequence identity to the corresponding ITS sequences from the reference D. catenulata isolates in GenBank (C. catenulata CBS 565). We performed a phylogenetic analysis using ITS sequences (Figure 1).

Antifungal Susceptibility
The quality control strains (C. krusei ATCC 6258 and C. parapsilosis ATCC 22019) showed MICs within the expected ranges. Aggregated MIC distributions of the nine antifungal agents tested against D. catenulata isolates by YeastOne ™ are shown in

Agreement Between the CLSI Method and Sensititre YeastOne ™
The EA values of the MICs between the CLSI method and YeastOne ™ for a majority of the antifungal drugs tested were >90%. In triazoles, 100% EA values were obtained for amphotericin B and 5-flucytosine. EA values for caspofungin and micafungin were 81.9 (9/11) and 91% (10/11), respectively (Supplementary Table 2).

Sequence Analysis of the ERG11 Gene in the Clinical Isolates
The complete open reading frame of D. catenulata ERG11 gene, predicted using bioinformatics analysis, showed that ERG11 comprises 1,563 base pairs ( Table 3). The ERG11 gene was sequenced for all 11 D. catenulata isolates. These isolates displayed amino acid substitutions in Erg11p compared with the sequences in GenBank. The SNP F126L (C378G) and synonymous mutations C681T, C888A, C999T, and C1164T were found in two isolates with high fluconazole MICs (>32 mg/ml). Amino acid substitutions, F126L (C378G) and G417K (L139F), and synonymous mutations, C681T, C888A, C999T, and C1164T, were found in three isolates with fluconazole MICs >64 mg/ml. Additionally, one isolate with K143R (A428G) and synonymous mutations, C681T, C888A, C999T, and C1164T, showed high resistance to fluconazole (MIC >256 mg/ml). Only one resistant strain had the K143R (A428G) mutation. In addition, synonymous mutations G306A and C852T were found in three isolates ( Table 4). By aligning the amino acid sequences of Erg11P from C. albicans and S. cerevisiae, we found that Erg11p of D. catenulata shared 69.75 and 65.25% identity with that of C. albicans and S. cerevisiae, respectively. Notably, the C. catenulata Erg11p contained residues at three locations (F126, L139, K143) that corresponded with three residues (F126, L139, K143, respectively) in C. albicans ERG11 (Supplementary Figure 1) whose alterations (F126L, K143R) have been reported to be associated with azole resistance (Berkow and Lockhart, 2017).

Literature Review
Only three of the sourced articles had reported clinical cases of infection with D. catenulata, including one case, in which the D. catenulata was isolated from the nails and two cases of D. catenulata isolated from blood culture (Crozier, 1977;Radosavljevic et al., 1999;Ha et al., 2018). One of the patients was diagnosed with gastric cancer and the other with endocarditis. One of the isolates showed low MICs for amphotericin B, miconazole, econazole, tioconazole, ketoconazole (1 mg/ml), traconazole (0.25 mg/ml), and fluconazole (16 mg/ml). Other isolates showed low MICs for fluconazole and micafungin (2 and 0.06 mg/ml, respectively). In addition, based on the patients' clinical information, we can conclude that broad-  spectrum antibiotic therapy may be a risk factor for infection with D. catenulata (Supplementary Table 4).

DISCUSSION
With the development of identification technologies, such as MALDI-TOF MS and high-throughput sequencing, many hitherto unknown species of microorganisms have been precisely identified (Qiu et al., 2019). Diutina catenulata is a rare yeast that causes infections in humans (Radosavljevic et al., 1999;Ha et al., 2018;Çakır et al., 2019). Therefore, we determined the proportion of D. catenulata isolates among clinical strains recovered from patients using CHIF-NET and the CHIF-NET North China Program. DNA sequencing and MALDI-TOF MS can be used to identify D. catenulata at the species level, and the proportions of the different species in the sample can be compared. ITS sequencing identified all 11 isolates, and the consistency of these clinical isolates with CBS 565 was 100%. Currently, ITS sequencing is considered a reliable marker for fungal identification (Wang et al., 2012). Most Candida species are pathogens that cause infections in immunocompromised patients (Cernakova et al., 2019). In this study, D. catenulata infection mainly occurred in elderly male admitted ICU patients, and nearly half of the infections were bloodstream infections. There are differences in drug resistance rates among Candida species (Sanguinetti et al., 2015). Candida auris has attracted much attention worldwide due to its high drug resistance and high fatality rates (Lone and Ahmad, 2019). Despite the low frequency of the species, the high MIC values observed for D. catenulata potentially could be important. In vitro antifungal susceptibility results showed that all D. catenulata showed low MICs for itraconazole, posaconazole, amphotericin B, and 5flucytosine, but four isolates showed high MICs (≥4 mg/ml) for the echinocandins. More than 36.4% of the isolates showed high MIC values for two echinocandins. High MICs of caspofungin may predict high MICs of micafungin and anidulafungin. More than 72.7% of the isolates showed high MICs for fluconazole. Thus, we suggest that D. catenulata with high MICs for fluconazole should be evaluated for its susceptibility to other antifungals. Micafungin has excellent effects in vivo against D. catenulata in infants (Çakır et al., 2019). However, in the present study, our in vitro antifungal susceptibility test showed that 36.4% of isolates showed high MICs (≥4 mg/ml) for echinocandins.
To analyze the molecular mechanisms of the antifungal resistance of D. catenulata, we sequenced the ERG11 and FKS1 genes. In agreement with the azole drug susceptibility profiles, we found three mutations in ERG11 of D. catenulata. Interestingly, F126L and K143R are the most prevalent residue substitutions in ERG11 (Berkow et al., 2017), but the L139F substitution is an unreported new observation. The mutations in FKS1 showed polymorphism. We identified two mutations in FKS1 hot spot 1 [F621I (F641) and S625L (S645)] and one near FKS1 hot spot 2 [I1348S (I1368)] associated with high MICs of caspofungin (Toutounji et al., 2019). S625L (S645) has been reported to cause failure of micafungin treatment against C. albicans (Slater et al., 2011). G3367A was found in isolates showing high and low MIC values against caspofungin and therefore may not be a potential of high MIC values against this antifungal. Of note, caspofungin was found to have a high interlaboratory variation (Espinel-Ingroff et al., 2013) and therefore the high MIC observed for this antifungal may not necessarily be a cause of concern as long as the MIC of the micafungin and anidulafungin are included and remain low. Polymorphisms of D. catenulata may also be used as a basis for typing.
Compared with the three articles have been described, we have noticed that the MICs of fluconazole and echinocandin of some of our strains were higher than 64 and 4 µg/mL, respectively, which are significantly higher than previous studies. This obvious difference also shows that there may be some evolutionary differences between Chinese and foreign isolates. In addition, our ICU patients also are treated with antibiotics. Therefore, patients maybe at high risk of D. catenulata infection. Similar to the case with most studies, lack of detailed clinical data on antifungal treatment is a major limitation of our study, which precludes us from reaching a conclusive statement about the choice of antifungal treatment and their potential outcome.
In conclusion, because of the low isolation rates, epidemiological knowledge of rare species is critical for clinical treatment. To the best of our knowledge, this study is the first to report the occurrence and distribution of D. catenulata in China, and its findings are expected to guide clinical treatment of D. catenulata infection in the future.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.