An efficient gene targeting system using Δku80 and functional analysis of Cyp51A in Trichophyton rubrum

Abstract

Trichophyton rubrum is one of the most frequently isolated fungi in patients with dermatophytosis. Despite its clinical significance, the molecular mechanisms of drug resistance and pathogenicity of T. rubrum remain to be elucidated because of the lack of genetic tools, such as efficient gene targeting systems. In this study, we generated a T. rubrum strain that lacks the nonhomologous end-joining-related gene ku80 (Δku80) and then developed a highly efficient genetic recombination system with gene targeting efficiency that was 46 times higher than that using the wild-type strain. Cyp51A and Cyp51B are 14-α-lanosterol demethylase isozymes in T. rubrum that promote ergosterol biosynthesis and are the targets of azole antifungal drugs. The expression of cyp51A mRNA was induced by the addition of the azole antifungal drug efinaconazole, whereas no such induction was detected for cyp51B, suggesting that Cyp51A functions as an azole-responsive Cyp51 isozyme. To explore the contribution of Cyp51A to susceptibility to azole drugs, the neomycin phosphotransferase (nptII) gene cassette was inserted into the cyp51A 3′-untranslated region of Δku80 to destabilize the mRNA of cyp51A. In this mutant, the induction of cyp51A mRNA expression by efinaconazole was diminished. The minimum inhibitory concentration for several azole drugs of this strain was reduced, suggesting that dermatophyte Cyp51A contributes to the tolerance for azole drugs. These findings suggest that an efficient gene targeting system using Δku80 in T. rubrum is applicable for analyzing genes encoding drug targets.

Key Points

• 1. A novel gene targeting system using Δku80 strain was established in T. rubrum

• 2. Cyp51A in T. rubrum responds to the azole antifungal drug efinaconazole

• 3. Cyp51A contributes to azole drug tolerance in T. rubrum

Introduction

Dermatophytosis is a superficial fungal infection with symptoms such as itching, redness, and nail abnormalities. Tinea pedis (athlete’s foot), a type of dermatophytosis, affects approximately 10% of the world’s population (Havlickova et al. 2008). Trichophyton rubrum, the most common dermatophyte (Zhan and Liu 2017), is a clinically important organism that reduces the quality of life and has a unique life cycle as an anthropophilic dermatophyte that specifically inhabits human surface tissues. A limited class of antifungals, such as azole antifungals, are used in dermatophytosis treatment. Although drug resistance issues in T. rubrum have resulted in a need to elucidate the detailed molecular mechanisms of its drug resistance and to identify and analyze drug targets (Yamada et al. 2017; Monod et al. 2019), these issues have not been completely clarified because of the underdevelopment of genetic methods in T. rubrum.

Homologous recombination (HR), a repair mechanism for DNA double-strand, is one of the most commonly used genetic engineering methods (Smithies et al. 1985). This technique allows not only the precise insertion of any DNA fragment into the desired genomic region, but also the introduction of mutations, deletions, and replacements based on sequence homology. Nevertheless, eukaryotes also possess a nonhomologous end-joining (NHEJ) repair mechanism for double-strand breaks, which competes with HR-mediated insertion of DNA into target regions (Krappmann 2007). To efficiently promote targeted integration via HR, several fungal species have been engineered by disrupting either of the Ku70/Ku80 complexes involved in NHEJ (Yamada et al. 2009; Matsumoto et al. 2021). These strains have demonstrated the effectiveness of improving HR efficiency in various fungi (Yamada et al. 2009; Matsumoto et al. 2021).

Azole antifungal drugs used for treating dermatophytosis target the lanosterol demethylase Cyp51, which functions in the ergosterol synthesis pathway. XP_003235929 and XP_003236980 in T. rubrum have been identified as Cyp51A and Cyp51B homologs, respectively (Celia-Sanchez et al. 2022). It has been reported that the addition of azole antifungal drugs induces fungal Cyp51 expression (Henry et al. 2000; Roundtree et al. 2020). This result suggests that Cyp51A functions as a responsible Cyp51 isozyme when ergosterol biosynthesis is hindered, such as during treatment with azole antifungals. Because the cyp51 homolog erg11 is an essential gene in budding yeast (Kalb et al. 1987), a deficiency of dermatophyte cyp51A could cause strong growth defects. In budding yeast, disruption of the natural 3′-untranslated region (UTR) by the insertion of an antibiotic-resistant marker was found to destabilize the corresponding mRNAs, and this strategy has been used to analyze essential genes (Schuldiner et al. 2005; Breslow et al. 2008).

In this study, we established a highly efficient HR system using a ku80-deficient strain of T. rubrum CBS118892 (Martinez et al. 2012), a clinically isolated strain from a patient’s nail. This strain has been used for whole genome analysis (Martinez et al. 2012) and several transcriptome analyses (Persinoti et al. 2014; Mendes et al. 2018; Martins et al. 2019; Cao et al. 2022), as well as to produce genetically modified strains (Lang et al. 2020; Ishii et al. 2023, 2024a,b). Therefore, we used this strain as a parent strain of ku80 deletion strain. Using this established system, we developed a mutant in which the neomycin phosphotransferase (nptII) gene was inserted into the 3′-UTR of cyp51A, which encodes a target for azole antifungals. When the azole antifungal drug efinaconazole was added, the magnitude of increase in cyp51A expression decreased in this mutant, which also exhibited sensitivity to ravuconazole and efinaconazole. This study would accelerate the production of genetically engineered strains to investigate the pathogenicity and drug resistance of T. rubrum and provide novel insights into antifungal targets.

Materials and methods

Fungal and bacterial strains and culture conditions

Trichophyton rubrum CBS118892 was cultured on Sabouraud dextrose agar (SDA; 1% Bacto peptone, 4% glucose, 1.5% agar, pH unadjusted) at 28 °C. The conidia of T. rubrum were prepared as described previously (Uchida et al. 2003). We confirmed the sequence of cyp51A and cyp51B as well as their promoters and terminators.

Plasmid construction

To construct a ku80-targeting vector, pAg1-Δku80-flp, approximately 2.1 and 1.5 kb of the 5′- and 3′-UTR fragments, respectively, of the ku80 open reading frame (ORF) were amplified from T. rubrum genomic DNA by polymerase chain reaction (PCR). The PCR products of the 5′- and 3′-UTR fragments were cleaved by SpeI/ApaI and BglII/KpnI, respectively. The plasmid backbone of pAg1 (Zhang et al. 2003) and the FLP/FRT module (Reuß et al. 2004) of pMRV-TmKu80/T2 were cleaved by SpeI/KpnI and ApaI/BamHI, respectively (Yamada et al. 2014). These fragments were joined using Ligation high version 2 (TOYOBO, Osaka, Japan). To construct a cyp51A 3′-UTR-targeting vector, pAg1-cyp51A-3′-UTR, 1.6 kbp of the cyp51A ORF and 1.5 kbp of the 3′-UTR fragment of cyp51 ORF were amplified from T. rubrum genomic DNA by PCR. The neomycin phosphotransferase gene cassette, which consists of E. coli neomycin phosphotransferase gene (nptII), Aspergillus nidulans trpC promoter (PtrpC), and Aspergillus fumigatus cgrA terminator (TcgrA), was cleaved from pMRV-TmKu80/T2 using ApaI and ClaI. These fragments were joined using an In-Fusion HD Cloning Kit (TaKaRa Bio, Shiga, Japan). The primers used in this study are shown in Table 1.

Table 1 Primers used in this study

Transformation of T. rubrum

Trichophyton rubrum was transformed using the polyethylene glycol (PEG) method as described previously (Yamada et al. 2008). The desired transformants and purified genomic DNA were analyzed by PCR. The ku80 ORF was replaced with a cassette with the nptII and the flippase gene (flp) flanked by flippase recognition sequences (Fig. 1a). As flp was inserted downstream of the copper ion-responsive promoter P_ctr4, nptII and flp were removed from the ku80-deficient genome by adding the copper ion chelator bathocuproinedisulfonic acid to induce FLP recombinase expression (Fig. 1a). Total DNA was extracted using the Quick-DNA Fungal/Bacterial Miniprep Kit (Zymo Research, Irvine, USA). Fungal cells were disrupted by μT-01 (TAITEC, Saitama, Japan) using 5-mm stainless beads.

Fig. 1

ku80 locus targeting and nptII marker excision. a Schematic representation of the ku80 locus before and after excision of the copper ion-responsive promoter P_ctr4, nptII and flp in T. rubrum. Site-specific recombination between the flanking FRT sequences (black box) was performed by the conditional expression of flp. b–e PCR analysis of total DNA samples from transformants. WT was used as a control. b Fragments were amplified with primer pairs (Primers 1 and 2). c Fragments were amplified with primer pairs (Primers 3 and 4). d Internal fragments of the ku80 ORF were amplified with primer pairs (Primers 5 and 4). e Internal fragments of nptII were amplified with primer pairs (Primers 6 and 7). The nptII-harboring strain (Δcla4) was used as a positive control. f Southern blot analysis of genome DNA samples from wild-type and Δku80 strains. g Mycelial growth of WT and Δku80 strains on SDA at 28 °C for 16 days

Antifungal susceptibility assay

Conidia (2 × 10³) were incubated with two-fold serial dilutions of antifungal agents in 200 μl 3-Morpholinopropanesulfonic acid (MOPS)-buffered Roswell Park Memorial Institute (RPMI)1640 medium (pH 7.0) at 28 °C for 7 days, and the minimum inhibitory concentration MIC₁₀₀ (minimal concentration required to inhibit growth by 100%) was determined. Efinaconazole was purchased from BLD Pharmatech Ltd, Shanghai, China, and ravuconazole was purchased from Merck, Darmstadt, Germany.

Quantitative reverse transcription-PCR (qRT-PCR)

Total RNAs were purified using NucleoSpin RNA (Macherey–Nagel, Düren, Germany) and reverse-transcribed into cDNAs using ReverTra Ace (Toyobo, Osaka, Japan) according to the manufacturers’ instructions. qRT-PCR was performed using TB Green Premix Ex Taq II (TaKaRa Bio, Shiga, Japan) on a StepOne Real-time PCR (Thermo Fisher Scientific, Waltham, USA). The relative mRNA expression level was determined using the 2^−∆∆Ctcsh1) as an endogenous control to normalize the samples (Jacob et al. 2012). The primers used in this study are listed in Table 1.

Statistical analysis

Mean values of three or more groups with two variables were compared using two-way ANOVA with Šidák correction and Tukey’s post hoc test, according to the recommendation of Prism 10 (GraphPad, Boston, USA). The difference in the efficiency of HR in wild-type (WT) and Δku80 strains was analyzed by two-sided Fisher’s exact test using Prism 10 (GraphPad, Boston, USA). Differences were considered significant at P < 0.05.

Results

To increase gene targeting efficiency, we attempted to delete the gene encoding Ku80. WT strain was transformed using the disruption cassette and 14 of the 261 transformants obtained (5.4%) were found to be deficient in ku80 gene. The cloned fungi were cultivated under conditions in which bathocuproinedisulfonic acid was incorporated into the medium to induce FLP recombinase and facilitate the removal of the nptII gene cassette. Subsequent cloning was performed to obtain the deficient strain candidates. To confirm that the ku80-deficient strain (Δku80) was generated as designed, PCR was performed using genomic DNA purified from WT and Δku80 strains (Fig. 1a, top and bottom, respectively) as templates. PCR performed using WT genomic DNA and primers designed for the 5′- and 3′-UTR of ku80 (Primers 1 and 2 in Fig. 1a, respectively) amplified the PCR products with the expected size (6.6 kbp; Fig. 1b, left lane). The size of PCR products in Δku80 was reduced as expected (3.8 kbp; Fig. 1b, right lane). In contrast, PCR performed using primers designed against sequences in the 5′-UTR (Primer 3 in Fig. 1a) and the ORF of ku80 (Primers 4 and 5 in Fig. 1a) yielded PCR products of the expected size for WT (Fig. 1c, d, left lanes) but not for Δku80 (Fig. 1c, d, right lanes) strain. The deletion of nptII from the genome of Δku80 + nptII strain (Fig. 1a, middle) was confirmed by PCR using primers designed against the sequences in the promoter and terminator of nptII (Primers 6 and 7 in Fig. 1a, respectively, Fig. 1e). The deletion of Δku80 was also confirmed by Southern blot analysis of genomic DNA from WT and Δku80 strains (Fig. 1f). These data indicated that the Δku80 strain was successfully generated with no reduction in the number of available drug markers. To ascertain the extent of the impact of Ku80 protein on growth, we compared mycelial growth between WT and Δku80 strains, which revealed comparable mycelial growth (Fig. 1g).

The mRNA expression of cyp51A in T. rubrum was upregulated by the addition of the azole antifungal drug efinaconazole, but that of cyp51B was not upregulated (Fig. 2a). We attempted to insert the nptII cassette into the downstream of cyp51A ORF of T. rubrum, as demonstrated in budding yeast studies (Schuldiner et al. 2005; Breslow et al. 2008). Using the obtained Δku80 strain, we inserted the nptII cassette into cyp51A 3′-UTR (hereinafter termed the insertional mutant; Fig. 2b, c). Homologous recombinant strains were obtained in 12 of 26 strains (46.2%; Table 2) in which the insertion of the drug resistance gene within target region was confirmed by PCR using primers designed within the ORF and 3′-UTR of cyp51A (Primers 8 and 9, respectively; Fig. 2b, c). The HR efficiency of the Δku80 strain was 46 times higher than that of the WT strain (1/98; 1.0%; Table 2). These data demonstrated that a highly efficient HR method had been established in T. rubrum.

Fig. 2

Production and characterization of the cyp51A 3′-UTR insertional mutants of T. rubrum. a The mRNA expression of cyp51A and cyp51B with or without 1 ng/ml efinaconazole in WT. Data are expressed as mean ± SD. The dots on the graph represent biological replicates (n = 4). n.s., not significant. ****, P < 0.0001. Two variables were compared using two-way ANOVA with Šidák correction. b Schematic representation of the cyp51A locus of WT and insertional mutant. c PCR analysis of total DNA samples from the independently isolated insertional mutant #1 and #2. The fragments were amplified with primer pairs (Primer 8 and 9). Δku80 was used as a control. d The mRNA expression of cyp51A in Δku80 and insertional mutant #1 and #2 with or without 1 ng/ml efinaconazole. The bars represent the standard deviation of the data obtained from three independent experiments. Data are expressed as mean ± SD. The dots on the graph represent biological replicates (n = 4–10). n.s., not significant. ****, P < 0.0001. Two variables were compared using two-way ANOVA with Tukey’s post hoc test. (e) Mycelial growth of Δku80 and insertional mutant #1 and #2 on SDA at 28℃ for 13 days

Table 2 Gene targeting efficiency of WT and Δku80 strains

Under efinaconazole-free conditions, the mRNA level of cyp51A in the two independently isolated insertional mutants, which were derived from the Δku80 strain, were comparable to that in the parent strain Δku80 (Fig. 2d). Nevertheless, efinaconazole-induced elevation of cyp51A mRNA level decreased in the insertional mutants (Fig. 2d). These findings suggest that the insertion of the nptII gene cassette into the 3′-UTR of cyp51A causes mRNA perturbation at least under the condition of cyp51A induction in T. rubrum. The insertional mutants exhibited similar mycelial growth as that of the parent strain Δku80 (Fig. 2e), but it showed increased sensitivity to the azole antifungals efinaconazole and ravuconazole (Table 3). However, the MICs of itraconazole and luliconazole remained unchanged in the insertional mutants. These findings suggest that Cyp51A functions as a factor for azole antifungal tolerance in T. rubrum.

Table 3 MIC values (μg/ml) of efinaconazole and ravuconazole

Discussion

Trichophyton rubrum is an anthropophilic dermatophyte specialized for human parasitism, whereas several other dermatophytes are zoophilic or geophilic (Reiss Errol et al. 2011). The nature of this fungus is of great interest from not only a medical but also biological point of view. In recent years, transcriptomic, proteomic, and immunological studies of this fungus have been conducted extensively (Xu et al. 2018, 2022; Burstein et al. 2020; Peres et al. 2022; Galvão-Rocha et al. 2023). Nevertheless, molecular and cellular biological studies of T. rubrum have been limited partially due to a lack of genetic tools for this organism. In this study, we generated a ku80-deficient strain of this fungus and demonstrated that this strain can be applied in efficient HR methods, similar to a system established in a zoophilic dermatophyte, Trichophyton mentagrophytes (formerly Arthroderma vanbreuseghemii) (Yamada et al. 2009). The method established in this study might serve as a fundamental technique to promote research that will advance the findings of previous comprehensive analyses and immunological analyses observed on the host side.

The insertional mutant, in which the expression induction of cyp51A by efinaconazole was attenuated, exhibited increased sensitivity to efinaconazole and ravuconazole. Considering that cyp51A expression was upregulated in response to efinaconazole addition, we speculated that T. rubrum Cyp51A is an inducible Cyp51 isozyme crucial for tolerance to azole antifungals. Indeed, it has been reported that itraconazole treatment also induces an increase in cyp51A expression (Diao et al. 2009). In A. fumigatus, loss or suppression of cyp51A expression enhances sensitivity to the azole antifungal fluconazole (Hu et al. 2007). Conversely, cyp51B deficiency does not significantly alter fluconazole sensitivity (Hu et al. 2007). This difference may be partially explained by the lower binding affinity of Cyp51A for fluconazole than for Cyp51B (Andrew et al. 2010). Nevertheless, a difference in the induction of the expression of each cyp51 gene in response to azoles may also contribute to this disparity in sensitivity. Regarding T. rubrum, no studies have investigated the contribution of Cyp51A and Cyp51B isozymes to the resistance to azole antifungal drugs. It has been reported that strains of T. mentagrophytes with a deficiency in Cyp51B exhibit a 2–3-fold reduction in MICs and enhanced susceptibility to itraconazole and voriconazole (Yamada et al. 2022). This observation highlights the necessity for further investigation into the susceptibility of the Cyp51 isozymes in dermatophytes to azole drugs. In the future, it is important to generate T. rubrum strains that are deficient in cyp51A and cyp51B, followed by analyzing their involvement in growth and resistance to azole antifungal drugs.

The induction of Cyp51 expression has been extensively studied in Aspergillus species. It is known that in response to ergosterol depletion by azole treatment, the membrane-bound transcription factor SrbA is cleaved by proteases and activated (Dhingra et al. 2016; Bat-Ochir et al. 2016), translocating into the nucleus and upregulating the expression of enzymes involved in the ergosterol synthesis pathway, including Cyp51A (Zhang et al. 2021). A deficiency in SrbA in A. fumigatus has been demonstrated to markedly enhance the azole susceptibility of the azole-sensitive and -resistant fungi (Willger et al. 2008; Hagiwara et al. 2016). Moreover, the SrbA-binding region located upstream of the cyp51A gene has been shown to contribute to drug resistance by forming tandem repeats (Gsaller et al. 2016; Kühbacher et al. 2022). As the regulatory mechanism of cyp51 expression in T. rubrum is anticipated to represent a novel drug target for azole drug susceptibility and a cornerstone of research for elucidating the mechanism of azole resistance. Consequently, it is of great interest to elucidate the detailed molecular mechanism of the regulation of cyp51A expression by utilizing genetic tools, such as Δku80, which can be employed to elucidate the molecular mechanisms underlying the regulation of cyp51A expression in T. rubrum.