Transcriptome analysis of multi-azole–resistant Microsporum canis isolated from a case of feline dermatophytosis

Abstract

The mechanism of multi-azole antifungal resistance in the dermatophyte Microsporum canis was investigated through a transcriptome analysis of the multi-azole–resistant strain LLP25-013. Strain LLP25-013 was cultured with and without itraconazole (ITCZ), after which total RNA was extracted from mycelial samples. Whole-RNA sequencing was performed using a DNBSEQ-G400 system. A total of 8,846 expressed genes were detected in samples of both total RNAs from both ITCZ-treated and ITCZ-free samples. Genes sequenced with more than 10,000 reads were selected from ITCZ-treated and ITCZ-free samples, and the expression levels were compared. Treatment with ITCZ upregulated the expression of 63 genes compared with no ITCZ treatment. By contrast, ITCZ treatment downregulated the expression of 253 identified genes. ITCZ addition reduced the expression of many genes in M. canis, which could be related to the broad-spectrum antifungal drug resistance of this organism. Among the upregulated genes, the ATP-binding cassette (ABC) transporter gene is involved in resistance in other dermatophytes. Co-culture of strain LLP25-013 with milbemycin and other azoles suggested that the isolate acquired azole resistance through high expression of the ABC transporter.

INTRODUCTION

Dermatophytosis attributable to Microsporum canis is a common cutaneous mycosis in small animals [6, 12]. Feline dermatophytosis is typically treated with oral itraconazole (ITCZ) and terbinafine (TBF), compounds that have demonstrated efficacy against both human and animal dermatophytoses in Japan [6, 12]. Azole antifungal drugs have thus been used for many years to treat M. canis infections in small animals; however, concerns have been raised regarding the emergence of azole-resistant strains of M. canis.

In our previous study, we described ITCZ-resistant isolates of Trichophyton rubrum that had been collected from Japanese patients in 2022 [4]. For all three isolates in that study, the minimum inhibitory concentration (MIC) of ITCZ was >32 mg/L, whereas the MICs of ravuconazole (RVCZ), luliconazole, and TBF were <0.03 mg/L. Therefore, strains exhibiting resistance to ITCZ have been isolated from cases of human dermatophytoses [4]. Because oral and topical azole antifungal drugs have been used continuously since the 1990s, it is likely that strains exhibiting acquired resistance strains have arisen in Japan.

We also isolated a multi-azole–resistant M. canis strain from a case of feline dermatophytosis, which was resistant to oral ITCZ therapy. Such cases have increased concerns that the incidence of small animal dermatophytosis involving azole-resistant dermatophytes is increasing. However, the mechanism of azole resistance in M. canis is not as clearly understood as that of anthropophilic dermatophytes such as Trichophyton interdigitale and T. rubrum [2]. In this study, therefore, we performed a transcriptome analysis to elucidate the resistance mechanism of multi-azole–resistant M. canis.

MATERIALS AND METHODS

Case history

A castrated, American curl-breed cat (3 years old) was referred to our animal hospital in Tokyo, Japan, with dermatophytosis of the skin. The cat had been treated with ITCZ at 3–6 mg/kg/, PO, SID for 4 weeks, but after stopping treatment, recurrence was noted. Microsporum canis was isolated from the crust of the lesion, resulting in a diagnosis of dermatophytosis due to M. canis infection.

Strain LLP25-013 of M. canis was isolated from a case of feline dermatophytosis in 2025. The isolate was identified as a M. canis by the colony and microscopic characteristics. Moreover, internal transcribed spacer (ITS) region sequence analysis indicated that 100% identical to that of M. canis (GenBank accession nos. OW988567 and PV422453).

This strain was shown to be ITCZ and RVCZ resistant but susceptible to TBF, voriconazole and luliconazole (Table 1), as assessed according to the Clinical & Laboratory Standards Institute M38-A2 guidelines with modifications, as described previously [3, 5, 9]. The CLSI M38-A2 guideline does not define “clinical resistance breakpoints” for itraconazole or ravuconazole in dermatophytes. Therefore, for convenience, we set the breakpoints for M. canis susceptibility to the ITCZ and RVCZ as more four times higher (ITCZ: >2 mg/L and RVCZ: >1 mg/L) from the MIC₉₀ of both drugs [9].

Table 1. Minimum inhibitory concentrations (μg/mL) of antifungal drugs.

	TBF	ITCZ	RVCZ	VRZ	LLCZ
Antifungal drug alone	0.125	>32	8	0.06	<0.03
Antifungal and milbemycin*		0.25	2	0.06

TBF: terbinafine, ITCZ: itraconazole, RVCZ: ravuconazole, MCZ: miconazole, VRCZ: voriconazole, LLCZ: luliconazole. *Assay was performed in RPMI1640 medium containing 1 μg/mL of milbemycin oxime.

Based on the results of antifungal susceptibility testing of the clinical isolate, the treatment was changed to TBF (19 mg/kg, PO, SID) for 4 weeks. After 4 weeks of treatment with TBF, the lesion was cured.

Next-generation sequence analysis of whole-RNA samples

Strain LLP25-013 was cultured in Sabouraud’s dextrose broth (SDB: 1% peptone and 2% glucose) at 28°C with shaking for 5 days. The culture medium containing mycelial cells was incubated with and without ITCZ at a concentration of 0.2 µg/mL for an additional 3 hr at 28°C. Samples of mycelia (approximately 100 to 200 mg) were flash-frozen in liquid nitrogen and then ground until homogeneous. Total RNA was extracted using an RNeasy Total RNA kit (QIAGEN, Tokyo, Japan). Approximately 1 µg/µL of total RNA from ITCZ-treated and ITCZ-free cultures was stored at −80°C until further analysis.

Whole-RNA (300 ng of RNA) sequencing was performed using a DNBSEQ-G400 system (MGI Tech Co., Ltd., Tokyo, Japan), and the resulting data were analyzed by Bioengineering Lab Co., Ltd. (Yokohama, Japan; https://gikenbio.com). cDNA libraries were synthesized using a MGIEasy Fast RNA Library Prep Set (MGI Tech) and transformed into circular libraries using a MGIEasy Dual Barcode Circularization kit (MGI Tech) for sequencing analysis. For next-generation sequencing analysis of the cyclic cDNA libraries, a DNBSEQ-G400RS High-throughput Sequencing kit (MGI Tech) and High-throughput Paired-End Sequencing Primer kit (App-D) (MGI Tech) were used to synthesize samples for DNA nanoballs. Whole-RNA sequencing (MGI DNBSEQ-G400RS) and data analysis for the ITCZ-treated and ITCZ-free samples were performed by Bioengineering Lab Co., Ltd.

Real-time polymerase chain reaction analysis of heat shock protein and ATP-binding cassette (ABC) transporter genes

The expression levels of heat shock protein and ABC transporter genes were analyzed using real-time quantitative polymerase chain reaction (RT-qPCR), as follows. First, cDNA samples synthesized from 0.5 µg of RNA were amplified by RT-qPCR (Thermal Cycler Dice; Takara, Kyoto, Japan) and analyzed as described in our previous report [7].

The following primer pair was used to analyze the ABC transporter gene (302-bp fragment): forward primer 5′-GGAACAACAGCCTTCAATCCACC-3′ and reverse primer 5′-CACCGGGAATTTCTCCGTCG-3′, corresponding to nucleotides 47–69 and 315–340, respectively, of the M. canis ABC transporter gene (Microsporum canis CBS 113480 ABC transporter [MCYG_02928], partial mRNA, GenBank accession no. XM_002847376).

RT-qPCR assays and data analysis (ΔΔCT method) for relative quantitation were carried out using Thermal Cycler Dice Real-Time System software, version 5.11C (Takara). Basal expression levels were estimated based on normalization to the level of actin-encoding transcripts in RNA from cultures under the assumption that an equivalent total RNA input and observed equal PCR efficiency would provide comparable threshold cycle (C_t) values. Gene expression levels were normalized to that of the actin-encoding gene, and values were compared between strains cultured in SDB with and without 0.2 µg/mL ITCZ. All RT-qPCR experiments were performed in duplicate.

Susceptibility of the isolate to a drug efflux pump inhibitor and azoles

The susceptibility of the isolate to azoles and milbemycin (a drug efflux pump inhibitor) was determined using the broth microdilution assay based on CLSI M38-A2 [10]. To screen for changes in azole susceptibility, fungal growth was tested on Roswell Park Memorial Institute 1640 medium (Cytiva, Tokyo, Japan) containing 1 µg/mL of milbemycin oxime (Fujifilm Wako Pure Chemical Corp., Osaka, Japan). This concentration of milbemycin reportedly does not inhibit fungal growth [10].

RESULTS

Comparison of gene expression with and without ITCZ treatment

Approximately 7,500 Mb were sequenced for both samples, with approximately 24 million paired reads for each. After annotation, the data were mapped based on Microsporum canis CBS 113480 (assembly ASM15114v1), which is registered in GenBank, confirming approximately 98% identity between RNA sequences for the ITCZ-treated and ITCZ-free samples.

A total of 8,846 expressed genes were identified in ITCZ-treated and ITCZ-free samples, and the expression levels were then compared. Among the compared genes, those for which the expression changed by a factor of >10,000 or <10,000 in the presence versus absence of ITCZ were plotted in the order of highest to lowest expression level (Supplementary Tables 1 and 2). For 63 genes, the expression increased following treatment with ITCZ, with the ABC transporter gene (GenBank accession no. XM_002847376.1) the highly expressed and thus presumed to be involved in resistance (Supplementary Table 1). By comparison, the expression of 253 genes declined following ITCZ treatment (Supplementary Table 2).

ITCZ-induced increase in gene expression

Transcript levels of the ABC transporter gene in azole-resistant M. canis strain LLP25-013 were compared between samples cultured in SDB containing 0.2 µg/mL ITCZ versus samples cultured in the absence of ITCZ. The results of RT-qPCR analyses indicated that transcript levels of the ABC transporter gene were approximately 2 times higher in the presence of ITCZ than in its absence (Fig. 1).

Fig. 1.

Transcript levels of the actin and ABC transporter genes in a multi-azole–resistant Microsporum canis strain (LLP25-013). The strain was cultured in Sabouraud’s dextrose broth containing 0.2 µg/mL itraconazole for 3 hr at 28°C. Expression levels of all genes were normalized to that of the actin-encoding gene and shown as relative expression.

MICs of milbemycin oxime and various azoles

The MICs against M. canis strain LLP25-013 for azoles in milbemycin-containing medium were 0.25 µg/mL for ITCZ, 2 µg/mL for RVCZ and 0.06 µg/mL for voriconazole (Table 1).

Data summary

Updated sequences were submitted to GenBank under the following accession numbers: for data from the transcriptome analysis of ITCZ-treated/non-treated LLP25-0013, BioProject: PRJDB35742/ PRJDB35742, and Run: DRR706902/ DRR706901.

DISCUSSION

This is the first report of a transcriptome analysis conducted to clarify the multi-azole–resistance mechanism of a strain of M. canis isolated from an infection. Unfortunately, the publicly available genome information for M. canis is quite limited, particularly in terms of a large number of unidentified genes (uncharacterized proteins) being registered. Therefore, many of the unidentified genes in this transcriptome analysis could be involved in azole resistance. Further analysis is needed to specifically identify these genes. Among upregulated genes, the ABC transporter gene is reportedly involved in resistance in other dermatophytes. Resistance to azoles has been linked to overexpression of genes encoding ABC transporter proteins in T. rubrum [8, 11]. To determine whether the ABC transporter is involved in azole resistance in strain LLP25-013, we first confirmed that expression of the gene increased by at least 2-fold in the presence of ITCZ using RT-PCR. Next, we investigated whether sensitivity to azoles increased during co-treatment with milbemycin, which inhibits the ABC transporter in dermatophytes. Milbemycin is a macrocyclic lactone produced as a fermentation product of Streptomyces hygroscopicus subsp. aureolacrimosus and exhibits broad-spectrum activity against nematodes that commonly infect animals, such as heartworms in dogs. Milbemycin is also known to block the drug efflux activity of the ABC transporter, thereby increasing the susceptibility of T. rubrum to azoles [8, 11]. In a previous study, we investigated the effect of milbemycin on the azole susceptibility of azole-resistant strains of T. rubrum and found that they became azole susceptible following milbemycin treatment [8]. In the present study, milbemycin counteracted the ITCZ and RVCZ resistance of the M. canis isolate, suggesting that the strain acquired resistance through high expression of the ABC transporter.

In contrast to the number of upregulated genes, ITCZ stimulation downregulated the expression of a greater number of genes. With regard to human dermatophytosis, Carmo et al. reported the in vitro induction of adaptation to ITCZ in a strain of Trichophyton interdigitale; notably, this adaptation was correlated with resistance to other antifungal drugs and to host immunity [1]. They hypothesized that successive exposure to ITCZ affects the emergence of adapted strains and leads to the recalcitrance of dermatophytosis to treatment with antifungal agents [1]. The results of our transcriptome analysis also suggested that cellular metabolic activity was attenuated in our M. canis isolate. This decline in cellular metabolism may play a role in the mechanism of antifungal drug resistance in a broad sense. The results of our study also suggest that multiple intracellular functions are involved in the mechanism of resistance to azole antifungal drugs in M. canis. However, further research is needed to fully elucidate the mechanism of resistance.

In conclusion, we report for the first time the isolation of a multi-azole–resistant strain of M. canis from a case of feline dermatophytosis that had been treated with oral ITCZ for 5 months. Co-culture of this isolate with milbemycin and other azoles suggested that the isolate acquired azole resistance through upregulated expression of the ABC transporter.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.