Phylogenetic Classification and Species Identification of Dermatophyte Strains Based on DNA Sequences of Nuclear Ribosomal Internal Transcribed Spacer 1 Regions

Abstract

The mutual phylogenetic relationships of dermatophytes of the genera Trichophyton, Microsporum, and Epidermophyton were demonstrated by using internal transcribed spacer 1 (ITS1) region ribosomal DNA sequences. Trichophyton spp. and Microsporum spp. form a cluster in the phylogenetic tree with Epidermophyton floccosum as an outgroup, and within this cluster, all Trichophyton spp. except Trichophyton terrestre form a nested cluster (100% bootstrap support). Members of dermatophytes in the cluster of Trichophyton spp. were classified into three groups with ITS1 homologies, with each of them being a monophyletic cluster (100% bootstrap support). The Arthroderma vanbreuseghemii-Arthroderma simii group consists of A. vanbreuseghemii, A. simii, Trichophyton mentagrophytes isolates from humans, T. mentagrophytes var. quinckeanum, Trichophyton tonsurans, and Trichophyton schoenleinii. Arthroderma benhamiae, T. mentagrophytes var. erinacei, and Trichophyton verrucosum are members of the Arthroderma benhamiae group. Trichophyton rubrum and Trichophyton violaceum form the T. rubrum group. This suggests that these “species” of dermatophytes have been overclassified. The ITS1 sequences of 11 clinical isolates were also determined to identify the species, and all strains were successfully identified by comparison of their base sequences with those in the ITS1 DNA sequence database.

Dermatophytes (dermatomycetes) have the capacity to invade keratinized tissues of humans and other animals to produce an infection, dermatophytosis (dermatomycosis) (28). The phylogeny of dermatophytes, however, remains unclear because their members are phylogenetically and taxonomically very closely related, their phenotypic features are sometimes poor, and many isolates from medical and veterinary samples have lost their sexual activity (25). From a clinical point of view, for definition of species or for performance of an epidemiological study, it is important to have a reliable method for the identification of dermatophyte species. Molecular biological studies of the phylogeny of the fungi have been performed, primarily by using the G+C content of chromosomal DNA (4), total DNA homology (5), restriction fragment length polymorphism (RFLP) analysis of mitochondrial DNA (mtDNA) (6, 14, 15, 21, 23), random amplification of polymorphic DNA (10, 13, 17, 22), and determination of the base sequence of 18S (11) or 28S (16) rRNA or ribosomal DNA (rDNA). For dermatophytes, however, the phylogenetic relationships of species or species-specific sequences cannot be fully defined by these methods.

Specific DNA sequences of internal transcribed spacer (ITS) 1 (ITS1) of rDNA in the dermatophytes were therefore determined and were analyzed phylogenetically. ITS1 is located between the 18S and the 5.8S rDNAs. As reported previously, the variable ITS regions have been proven to be useful in resolving relationships between close taxonomic relatives (2, 3, 18), and in the field of medical mycology, several phylogenetic studies in which the ITS1 region and the primer system designed by Makimura et al. (20) or White et al. (29) were used were reported on recently (1, 20, 26, 27). We have reported that it is feasible to successfully differentiate between members of the Trichophyton mentagrophytes complex, the major dermatophytes, which are hard to identify by their morphological features, by demonstrating their phylogenetic relationship by comparing the base pair sequences of the ITS1 regions (20). In the present study we determined the phylogeny of the group of dermatophytes, including the genera Trichophyton, Microsporum, and Epidermophyton, and identified the species using the base pair sequences of ITS1.

MATERIALS AND METHODS

Fungal strains.

The 12 standard strains and 11 clinical isolates of dermatophytes used in this study are described in Tables 1 and 2. The clinical strains were isolated in Japan and were identified by their morphological features, but the species of two of them could not be specified because they did not have typical microscopic structures.

TABLE 1.

Standard strains of dermatophytes used in this study

Species	Straina	ITS1 accession no.
Trichophyton mentagrophytes var. quinckeanum	SM7283←IMI140690	AB017171
Trichophyton mentagrophytes var. quinckeanum	SM7284←IMI163337
Trichophyton tonsurans	TIMM1254←IP133-73	AB017172
Trichophyton schoenleinii	TIMM3395	AB017173
Trichophyton violaceum	TIMM3396	AB017174
Trichophyton terrestre	TIMM3394←CBS307.65	AB017175
Trichophyton verrucosum	TIMM3328	AB017176
Microsporum gypseum	TIMM3397←CBS161.69	AB017177
Microsporum audouinii	TIMM0757←IP941	AB017178
Microsporum canis	TIMM1502	AB017179
Microsporum cookei	TIMM3398	AB017180
Epidermophyton floccosum	TIMM0431	AB017181

^aArrows are used to indicate the history of the strain. CBS, Centraalbureau voor Schimmelcultures, Barn, The Netherlands; IMI, International Mycological Institute, Surrey, United Kingdom; Institut Pasteur, Paris, France; RV, Institute de Medicine Tropicale, Antwerp, Belgium; SM, Department of Dermatology, Shiga University of Medical Science, Otsu, Japan; TIMM, Teikyo University Institute of Medical Mycology, Tokyo, Japan. 

TABLE 2.

Clinical isolates used in this study

Strain	Source	Morphological identification	Identification according to ITS1 sequence
CN9708001	Tinea pedis	Trichophyton rubrum	Trichophyton rubrum
CN9708002	Tinea pedis	Trichophyton rubrum	Trichophyton rubrum
CN9708003	Tinea pedis	Trichophyton rubrum	Trichophyton rubrum
CN9708004	Tinea pedis	Trichophyton mentagrophytes	Trichophyton mentagrophytes
CN9708005	Tinea pedis	Trichophyton mentagrophytes	Trichophyton mentagrophytes
CN9708006	Tinea pedis	Trichophyton mentagrophytes	Trichophyton mentagrophytes
CN9708007	Tinea pedis	Trichophyton mentagrophytes	Trichophyton mentagrophytes
CN9710001	Tinea capitis	Trichophyton violaceum	Trichophyton violaceum
CN9703001	Tinea corporis	Microsporum canis	Microsporum canis
CN9708001	Tinea corporis	Unidentifiablea	Trichophyton rubrum
CN9802001	Kerion celsi	Unidentifiable	Trichophyton violaceum

^aUnidentifiable, strains in which the species could not be specified because they did not show the typical microscopic structures. 

Preparation of DNA from fungal cells.

All fungal strains were grown on Sabouraud dextrose agar (peptone, 1% [wt/vol]; glucose, 1% [wt/vol]; agar, 1.5% [wt/vol]) at 27°C for 5 days. Rapid preparation of DNA from strains was performed by the method described by the authors (20). A small amount of mycelium grown on Sabouraud dextrose agar was placed in lysis buffer (200 mM Tris-HCl [pH 8.0], 0.5% [wt/vol] sodium dodecyl sulfate, 250 mM NaCl, 25 mM EDTA) and crushed with a conical grinder. It was then incubated at 100°C for 15 min and mixed with 150 μl of 3.0 M sodium acetate, kept at −20°C for 10 min, and then centrifuged at 10,000 × g for 5 min. The supernatant was extracted once with phenol-chloroform-isoamyl alcohol (25:24:1 [vol/vol]) and was subsequently extracted once with chloroform. The DNA was precipitated with an equal volume of isopropanol at −20°C for 10 min, washed with 0.5 ml of 99% ethanol, dried, and suspended in 50 μl of ultrapure water (Milli-Q Synthesis A10; Millipore). One microliter of solution was used as the template for PCR. The total time required to prepare the DNA was 80 min.

Oligonucleotides.

The oligonucleotide primers, designed by the authors (20), (18SF1, 5′-AGGTTTCCGTAGGTGAACCT-3′; 58SR1, 5′-TTCGCTGCGTTCTTCATCGA-3′) were made by Amersham Pharmacia Biotech Co., Ltd. (Tokyo, Japan).

PCR.

Each PCR mixture contained 10 μl of 10× reaction buffer (Pharmacia), 100 μM (each) dATP, dCTP, dGTP, and dTTP (Pharmacia), 2.5 U of Taq polymerase (Pharmacia), 30 pmol of each primer, and DNA template solution. Ultrapure water was added to increase the volume to 100 μl. Each reaction mixture was heated to 94°C for 5 min, and PCR was performed under the following conditions: 94°C for 1 min, 60°C for 15 s, and 72°C for 15 s for 25 cycles. The thermal cycles were terminated by polymerization at 72°C for 10 min. The products were detected as a single band of 0.3 kbp by agarose gel electrophoresis and UV irradiation.

ITS1 DNA sequencing and phylogenetic analysis.

Both strands of the PCR products were directly sequenced with a DNA Sequencing Kit (Perkin-Elmer) with primers 18SF1 and 58SR1 and an automatic sequencer (Genetic Analyzer 310; Perkin-Elmer), according to the manufacturer’s instructions.

The ITS1 sequences of the standard strains used in this study and of members of the T. mentagrophytes complex (Arthroderma vanbreuseghemii, DDBJ/EMBL/GenBank accession no. AB011488; Arthroderma benhamiae [Americano-European race], accession no. AB011457; Arthroderma benhamiae [African race], accession no. AB011454; Arthroderma simii, accession no. AB011461; T. mentagrophytes isolates from humans, accession no. AB011463; T. mentagrophytes var. erinacei, accession no. AB011455; Trichophyton rubrum, accession no. AB011453), as reported by the authors (20), were aligned by using the Clustal W computer program (12) and GENETYX-MAC 10.1 software (Software Development Co., Ltd., Tokyo, Japan). Phylogenetic trees were then constructed by the DNA maximum-likelihood (ML) method in the PHYLIP program (Phylogeny Inference Package), version 3.5c (8), and the neighbor-joining (NJ) (24) method in the NJPLOT program (9). Bootstrap (7) analysis with the Clustal W program was performed by taking 1,000 random samples from the multiple alignment. This provided a measure of how well supported parts of the tree are, given the data set and the method used to construct the tree. The tree was rooted with Epidermophyton floccosum as an outgroup, because it was shown that this species is phylogenetically distant from other dermatophytes by the RFLP analysis of mtDNA (15). The evolutionary distance between organisms is indicated by the horizontal branch length, which reflects the number of nucleotide substitutions per site along that branch from the node to the endpoint. In the NJ tree, the percentage of bootstrap samplings that support the interior branches is noted.

Nucleotide sequence accession numbers.

The nucleotide sequence data reported in this paper will appear in the DDBJ/EMBL/GenBank nucleotide sequence database with the accession numbers presented in Table 1.

RESULTS

Each strain of the dermatophytes tested was shown to have unique ITS1 base sequences, and the two strains of T. mentagrophytes var. quinckeanum were found to be identical. Phylogenetic trees were prepared by the NJ (Fig. 1) and ML (Fig. 2) methods and were constructed from data for 18 species of dermatophytes. The sequences of their ITS1 regions are presented in Fig. 3; the sizes of these regions ranged from 175 to 293 bp. In the NJ tree, Trichophyton spp. and Microsporum spp. form cluster A, and all Trichophyton spp. except Trichophyton terrestre form cluster B (100% bootstrap support). The members of the dermatophytes in cluster B were classified into three groups with ITS1 homology (groups a, b, and c) according to their ITS1 DNA sequences, and each of them is a monophyletic cluster (100% bootstrap support). ITS1 homology group a (A. vanbreuseghemii-A. simii group) consists of A. vanbreuseghemii, A. simii, T. mentagrophytes isolates from humans, T. mentagrophytes var. quinckeanum, Trichophyton tonsurans, and Trichophyton schoenleinii. Both races of A. benhamiae, T. mentagrophytes var. erinacei, and Trichophyton verrucosum are members of ITS1 homology group b (A. benhamiae group). T. rubrum and Trichophyton violaceum form a cluster of ITS1 homology group c (T. rubrum group). The phylogenetic relationships mentioned above were also supported by the ML tree.

FIG. 1.

NJ tree of dermatophytes on the basis of their ITS1 sequences. The NJ tree was constructed with data for standard strains of dermatophytes (see Table 1 and the text). The numbers above the branches indicate the percentage of bootstrap samplings. Branches without numbers have frequencies of less than 70%. A, cluster of Trichophyton spp. and Microsporum spp.; B, cluster of all Trichophyton spp. except T. terrestre; a, Arthroderma vanbreuseghemii-A. simii group; b, A. benhamiae group; c, T. rubrum group. K nuc, thousands of nucleotides.

FIG. 2.

ML tree of dermatophytes on the basis of their ITS1 sequences. The ML tree was constructed with data for standard strains of dermatophytes (see Table 1 and the text). A, cluster of Trichophyton spp. and Microsporum spp.; B, cluster of all Trichophyton spp. except T. terrestre; a, A. vanbreuseghemii-A. simii group; b, A. benhamiae group; c, T. rubrum group. K nuc, thousands of nucleotides.

FIG. 3.

Alignment of ITS1 sequences of dermatophytes. The sequences of 18 species of dermatophytes (see Table 1 and the text) were aligned by using the Clustal W (12) and GENETYX-MAC 10.1 (Software Development Co., Ltd.) computer programs. Hyphens designate gaps that were added to permit alignment. T. men. quinckeanum, T. mentagrophytes var. quinckeanum; T. men. erinacei, T. mentagrophytes var. erinacei; A. benhamiae A/E and Af, A. benhamiae Americano-European race and African race, respectively.

The ITS1 sequences of 11 clinical isolates were also determined. Each of the morphologically identified strains of dermatophytes (three strains of T. rubrum, four strains of T. mentagrophytes, one strain of T. violaceum, and one strain of Microsporum canis) was shown to have ITS1 base pair sequences identical to that of the respective standard strain tested (Table 2). Two morphologically unidentifiable strains were then subjected to ITS1 sequencing; one was revealed to be T. rubrum, and the other was shown to be T. violaceum on the basis of the ITS1 DNA sequence database constructed by the authors.

DISCUSSION

Using ITS1 rDNA sequences from 12 newly sequenced and 7 previously reported strains of fungi, we have described the phylogeny of members of the dermatophytes. The phylogenetic relationship based on the ITS1 DNA sequence alignment of meiosporic (perfect) and mitosporic (imperfect) states of the strains agreed with the proposed taxonomic connection in their sexual compatibility (25), RFLP analysis of mtDNA (14, 15, 21, 23), and a phylogenetic study based on 18S rDNA (11) or 28S rDNA (16) sequences. In particular, the last two papers dealt with the base sequences of the rDNA region, but they omitted A. benhamiae, T. verrucosum, and T. mentagrophytes var. erinacei, which together formed a unique cluster, cluster B-b, in the phylogenetic trees that appear in Fig. 1 and 2. Thus, because of their highly variable ITS1 sequences, the phylogenetic analysis of the members of the dermatophytes was achieved in more accurate detail.

We stated earlier that there were three ITS1 homology groups (groups a, b, and c) in the cluster of Trichophyton spp. The clusters of ITS1 homology group a (A. vanbreuseghemii-A. simii group), group b (A. benhamiae group), and group c (T. rubrum group) were 100% supported by bootstrap analysis. A. vanbreuseghemii, A. simii, A. benhamiae, and anamorphic species of T. mentagrophytes constitute the T. mentagrophytes complex (20, 25) because they are difficult to distinguish from each other by their morphological features. By using sexual compatibility (25), RFLP analysis of mtDNA (21), or DNA sequence analysis of ITS1 (20), the T. mentagrophytes complex was shown to be monophyletic, and each of the members was identified. The present study showed that the species of dermatophytes pathogenic for humans or animals, T. tonsurans, T. schoenleinii, and T. verrucosum, are members of the A. vanbreuseghemii-A. simii group (Fig. 1, cluster a) or the A. benhamiae group (Fig. 1, cluster b). This suggests that these medically important dermatophytes have been overclassified. However, since each of the “species” has a unique phenotype, pathogenicity, and host-specific affinity, it is reasonable to retain their “species” identifications in order to identify the pathogen.

In addition to establishing the significance of the ITS1 region from a taxonomic standpoint, we also identified these clinically important species using the ITS1 DNA sequence database. With this system, not only the morphologically identified strains of T. mentagrophytes, T. rubrum, T. violaceum, and M. canis but also the two strains of morphologically unidentifiable strains, T. rubrum and T. violaceum, were successfully identified. This ITS1-based identification system saves time (it takes 2 to 3 days) and is accurate and applicable even to strains with atypical morphological features.

Further evaluation of the phylogenetic analysis and identification system, both of which are based on ITS1 rDNA sequences, is under way in our laboratory with other species and strains. Moreover, a new approach to the detection and identification of pathogenic fungi from clinical specimens, i.e., skin, nail, or hair samples, with this database, along with previously reported molecular diagnostic systems (19), is also in progress in our laboratory.