Cow mastitis has a major negative impact on dairy industries, causing significant economic losses to farmers. Pathogenically, a wide variety of microorganisms have been identified as causative agents of cow mastitis, mainly bacteria and fungi (1). In cases of fungal infection of the mammary gland, yeasts of Candida genus are the most reported fungal pathogens in cow mastitis (2).

Historically, the mycotic mastitis caused by fungi of the Candida genus was first described by Fleischer as early as 1930 (3). Although the infection of C. albicans was considered as the most common cause of mastitis, cases of cow mastitis caused by the infection of non-albicans Candida spp. (NAC), such as C. krusei, C. parapsilosis, C. glabrata, and C. tropicalis, have increased significantly during the last decade (4). Among these NACs, C. krusei ranked as the fifth most common cause of cow mycotic mastitis (5, 6). C. krusei was reported as the causative agent of bovine mastitis and bronchopneumonia in Canada, Mexico, Japan, the United Kingdom, Turkey, Poland, and Algeria (7, 8). Our previous investigation also suggested that C. krusei was one of the most important pathogens in mycotic mastitis in dairy farms of the Yinchuan region in Ningxia, China. This result was in accordance with the report of Erbaş et al. (9).

Candida krusei has been regarded as a multidrug-resistant fungal pathogen because of its intrinsic resistance to fluconazole (FLC) (10, 11), with more than 96% of human clinical and veterinary isolates being fluconazole-resistant (12). Azole is one of the most common antifungal drugs in agricultural practices (13), including itraconazole, ketoconazole, and tetraconazole. Although it has been reported that multiple mechanisms are involved in drug resistance in Candida spp., mechanisms involved in alterations of target enzymes and upregulation of multidrug resistance (MDR)-related proteins are the common mechanisms of Candida resistant to azoles. In this regard, 14α-lanosterol demethylase (14-DM) is the target enzyme of azoles, which is responsible for the production of an ergosterol precursor and is encoded by the ERG11 gene. In C. albicans and C. parapsilosis, the efflux pump genes CDR1, CDR2, and MDR1 are also associated with azole resistance (14). Nowadays, although transporter genes ABC1 and ABC2 were involved in drug resistance in C. krusei (15, 16), increased lines of evidence suggested that changes in the expression of activity of target enzyme and upregulation of MDR were the main mechanisms of drug resistance in C. krusei (11, 16). To date, the study on mechanisms of Candida in azole resistance has mainly focused on C. albicans, C. glabrata, and C. tropicalis, but studies on azole resistance mechanisms in C. krusei, especially the C. krusei isolates from cow mastitis resistant to itraconazole, are limited, and the involvement of ERG11, ABC1, and ABC2 genes in the drug resistance of C. krusei has not been determined (17).

In the present study, we evaluated the profile of the susceptibility of C. krusei to itraconazole and investigated the potential alterations of the ERG11 gene and the differential expression of ERG11, ABC1, and ABC2 genes of 15 clinical C. krusei isolates that were isolated from cow mastitis in Yinchuan, Ningxia, China.

Our previous study demonstrated that C. krusei was a predominant pathogen isolated from mycotic mastitis of cows in Yinchuan, Ningxia, China (12), suggesting that it may be an important fungal pathogen of mycotic mastitis of cows in this area. The crucial roles of drug-resistant genes ERG11, ABC1, and ABC2 in FLC-resistant clinical isolates of C. krusei from human have been well-established (15, 24). However, the pathogenic molecular mechanism of C. krusei isolated from cow mastitis remains unclear. In the present report, we evaluated the susceptibility profiles and mutations in the ERG11 gene in 15 clinical C. krusei isolates. The expression of drug-resistant genes ERG11, ABC1, and ABC2 between isolates susceptible, susceptible dose-dependent, and resistant to ITR was also analyzed. We identified three synonymous mutations and one missense mutation in the ERG11 gene in these clinical C. krusei isolates, as previously described in human C. krusei. Furthermore, the A756T was only presented in ITR-resistant strains, suggesting that it might be correlated with drug resistance in C. krusei, while mutations T642C and T939C were presented in all these 15 C. krusei isolates. The expression of drug-resistant ERG11 and ABC2 was also significantly higher in ITR-resistant C. krusei isolates compared to ITR-susceptible and susceptible dose-dependent isolates, suggesting a correlation of mutations of the ERG11 gene with the resistance to antifungal agents in C. krusei.

In the present study, based on the CLSI BMD method's susceptibility to ITR, the 15 C. krusei clinical isolates from cows with clinical mastitis could be divided into three groups: ITR-susceptible (2 isolates), ITR-susceptible dose-dependent (8 isolates), and ITR-resistant (5 isolates). The antifungal testing showed that 13.3, 53.3, and 33.3% were susceptible, susceptible dose-dependent, and resistant to itraconazole among these C. krusei clinical isolates, respectively. Notably, 13 of the 15 C. krusei isolates (73.4%) were also resistant to FLC and KET, 10 of the 15 C. krusei isolates (66.7%) were resistant to flucytosine, and 3 of the 15 isolates (20%) were resistant to amphotericin B. This finding was consistent with a study by Namvar et al. (25), but was different from reports by others (8, 9, 26). Of interest, the rate of resistance to antifungal agents in these C. krusei isolates was lower than our previous findings (12), which might be attributed to the reduction in the use of antifungal drugs during breeding. Consistent with our previous studies, C. krusei isolates were double-resistant and multidrug-resistant to antifungal drugs. It is strongly recommended that ketoconazole and other azole antifungal agents should not be used in the treatment of C. krusei infection in dairy cows in Ningxia, China, owing to high drug resistance.

It is worth noting that CK2, CK5, and CK12 showed resistance to amphotericin B, but such cases are rare (27). However, several lines of studies evidenced an increased minimum inhibitory concentration (MIC) of amphotericin B in C. krusei isolates. In Candida spp., resistance to amphotericin B was found to be associated with a decreased ergosterol content of cell membrane (28–30). In addition, an inactivation of ERG3 could substitute 14α-methylfecosterol with ergosterol, thus reducing ergosterol levels, which, in turn, resulted in deficient ergosterol to counter the function of amphotericin B (31). Moreover, several other mutations in the ergosterol biosynthetic genes such as ERG2, ERG5, ERG6, and ERG24 can also result in C. albicans and C. glabrata resistant to amphotericin B (32–34). However, such mutations have not been reported in C. krusei and further studies are needed in future studies.

A number of studies have focused on ERG11 gene mutation in Candida species (35–37). In order to fully understand ERG11 gene mutation of C. krusei, the whole open reading frame of the ERG11 gene was amplified for sequencing analysis, and three synonymous mutations and one missense mutation were identified in this study. C44T has one missense mutation and was found in all 15 isolates and the reference strain, suggesting that it had no impact on the itraconazole resistance of C. krusei. Molecularly, the C44T mutation resulted in the alteration of alanine to valine in the 15th amino acid of 14α-lanosterol demethylase (14-DM); this mutation might occur outside the active site of the ERG11 gene, which might not affect the mutual interaction of azole and 14-DM, or a single missense mutation might not be sufficient to change the affinity of the 14-DM to azole (15). Moreover, synonymous mutations (C642T, A756T, and T939C) in the 15 isolates were consistent with a previous report (24). Synonymous mutations can affect transcription, splicing, mRNA transport, and translation, any of which could change phenotype, rendering the synonymous mutation non-silent (38). He et al. (15) also reported that C642T, T1389C, and G1536C mutations occurred in all the experimental strains. In addition, the T1389C mutation was also reported by Ricardo et al. (16). Tavakoli et al. (35) revealed a heterozygous polymorphism at position T939C in the ERG11 coding region and speculated that this polymorphism might play a key role in the transcriptional regulation of genes and be involved in the processes of ergosterol biosynthesis. Of note, several previously reported mutations, including the T418C missense mutation (16), and C51T, T1389C, and G1536C synonymous mutations (15), have not been found in this study; thus, limited clinical isolates (15) were analyzed. Mechanistically, previous studies on C. albicans and C. tropicalis have demonstrated that the missense mutation was associated with resistance to azole, which was partly through changing the conformation of the target enzyme 14-DM, which, in turn, decreased its drug affinity and influenced the enzyme's function in ergosterol biosynthesis (39, 40). The resistance mechanism of these resistant strains may be due to one or multiple mutations in these genes, which needs further investigation.

In addition, the expression of ABC2 and ERG11 genes was significantly upregulated in C. krusei veterinary clinical isolates resistant to itraconazole. Previous studies demonstrated that resistance to azole was also due to the increased expression of ERG11. This results in insufficient azole activity owing to the overproduction of the target enzyme (41). Although ERG11 overexpression has been reported in C. krusei, the mechanism behind the overexpression remains unclear (15, 35). Another mechanism of resistance to azole is via the decreased intracellular accumulation of azole. This can be due to efflux pump activity or changes in the cell membrane. In this regard, drug efflux pumps belong to either the ATP-binding cassette (ABC) family of transporters or the major facilitator super family (MFS) class. These proteins can pump out fungicidal compounds across the cell membrane, and their overexpression results in multidrug resistance phenotype in pathogenic fungus (17). In contrast to members of the MFS class that are actuated by electrochemical proton-motive force, ABC family members depend on the hydrolysis of ATP for energy (42, 43). Indeed, along with the decrease of susceptibility to itraconazole, the ABC2 gene mRNA expression in the isolates appeared to increase. Although it was considered that Abc1p played an important role in the innate FLC resistance of C. krusei (16, 17), the ABC2 gene could be activated slower than the ABC1 gene in the presence of voriconazole (16). Given voriconazole tolerance, the Abc1p efflux pump is supposed to be more efficient in expelling drug and plays a late role in the development of resistance, and the accumulation of itraconazole in drug-susceptible C. krusei was higher than that of resistant strains (44). Interestingly, Venkateswarlu et al. found that two isolates highly resistant to fluconazole showed a different sensitivity to itraconazole, suggesting that itraconazole and fluconazole in C. krusei may have different resistance mechanisms (44). The authors stated that C. krusei resistant to itraconazole was due to decreased drug accumulation in cells and speculated that there might exist more efflux pumps contributing to itraconazole resistance of C. krusei, which could be well-explained in this study. The overexpression of both ERG11 and ABC2 has been reported to be involved in itraconazole resistance in C. krusei (15, 35). However, an unusual transient or stable resistance of C. krusei to voriconazole has also emerged. Overexpression of ABC2 and ERG11, as observed in itraconazole resistance, imparts a transient resistance to voriconazole, while a more stable resistance was observed due to the overexpression of ABC1 and point mutation in ERG11 (16). Taking this into account, we speculate that the resistance mechanisms of itraconazole and voriconazole in C. krusei clinical isolates may be different. Moreover, other genes encoding ATP-dependent efflux transporters may occur in C. krusei, such as a CgSNQ2 homologous gene that was verified as an azole-associated resistance gene in C. glabrata (45). Although the C. krusei genome has been sequenced, it is not completely annotated yet; thus, other transporter genes were not assessed. Our mRNA expression data showed that Abc2p may play a more important role in itraconazole resistance of C. krusei, instead of Abc1p.

This study has enriched our knowledge in the veterinary clinical C. krusei resistance gene expression and mutation data by comparing the difference between the veterinary clinical and the human clinical C. krusei, and further deepened our understanding of the resistance mechanism of C. krusei in veterinary clinics. The limitation of this study is that the sample size was small and no drug susceptibility test and resistance mechanism research related to echinocandin has been included, which require further investigations.

In conclusion, in this study, we found that C. krusei veterinary clinical isolates exhibited a different susceptibility to antifungal agents. Mechanistically, the A756T mutation in the ERG11 gene resulted in an upregulation of drug-resistant genes ERG11 and ABC2, substantially enhancing the resistance to itraconazole of C. krusei. Although we have identified four point mutations in the ERG11 gene associated with itraconazole resistance and have already described their role on the itraconazole resistance of C. krusei, it is necessary to confirm the effect of these mutations by site-directed mutagenesis of the C. krusei strain in the future. Nevertheless, this study may thus provide an insight into the mechanism of the resistance of C. krusei to antifungal agents, which warrants for further investigation.

The datasets generated for this study are available on request to the corresponding author.

The animal study was reviewed and approved by Ethics Committee of Animal Research of Ningxia University.

JD, YW, and XZ conceived and designed the experiments. JD and JF analyzed the data and drafted the manuscript. JD, WM, JF, and XL performed experiments and acquired data. XL and XZ interpreted data and critically revised the manuscript. All authors read and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.