Species-identification of dermatophytes Trichophyton, Microsporum and Epidermophyton by PCR and PCR-RFLP targeting of the DNA topoisomerase II genes

Journal of Dermatological Science (2003) 33, 41 /54

www.elsevier.com/locate/jdermsci

Species-identification of dermatophytes Trichophyton, Microsporum and Epidermophyton by PCR and PCR-RFLP targeting of the DNA topoisomerase II genes Toshio Kanbea,*, Yasuhiro Suzukia, Atsushi Kamiyab, Takashi Mochizukic, Masako Kawasakic, Machiko Fujihirod, Akihiko Kikuchia a

Division of Molecular Mycology and Medicine, Department of Advanced Medical Science, Center for Neural Disease and Cancer, Nagoya University Graduate School of Medicine, 65 Tsurumai, Showa-ku, Nagoya 466-8550, Japan b Department of Dermatology, Nagoya University Graduate School of Medicine, 65 Tsurumai, Showa-ku, Nagoya 466-8550, Japan c Department of Dermatology, Kanazawa Medical University, 1-1 Daigaku, Uchinada, Ishikawa 920-0293, Japan d Ibi General Hospital, Ibigawa-cho, Gifu 501-0696, Japan Received 17 March 2003; received in revised form 27 May 2003; accepted 28 May 2003

KEYWORDS Dermatophytes; Identification; DNA topoisomerase II gene; PCR; PCR-RFLP

Summary Background: We have focused on the DNA topoisomerase II genes of pathogenic fungi and have previously applied polymerase chain reaction (PCR)-based identification of several species including the some of the major dermatophyte species. Objective: To identify the dermatophytes (18 species) to a species level by PCR and PCR-restriction fragment length polymorphism (RFLP) techniques, without determining the nucleotide sequence. Methods: The genomic DNAs of the dermatophytes (ten species of Trichophyton , seven species of Microsporum , and Epidermaphyton floccosum ) were amplified by PCR using a common primer set (dPsD1) for the dermatophytes, followed by nested PCR using other primer sets (dPsD2, PsT and PsME) that contained primers specific for the DNA topoisomerase II genes of the dermatophytes. PCRs using PsT and PsME were used for the species-identification of Trichophyton , Microsporum and E. floccosum . The PCR products generated by dPsD2 were digested with restriction enzymes (Hin c II, Hin f, Afl II and Pfl M I), and the restriction profiles were analyzed. Results: Of the eighteen species of dermatophytes, five species (T. rubrum , T. violaceum , M. canis , M. gypseum and E. floccosum ) were specifically identified by the PCR using PsT and PsME to the species level, and the remaining species were identified by the unique restriction profiles for each species in the PCR-RFLP analysis, except that the restriction profile of T. mentagrophytes var. interdigitale was identical to that of T. mentagrophytes var. quinckeanum . Conclusion: PCR and PCR-RFLP techniques targeting the DNA topoisomerase II gene are simple and rapid, and quite useful as tools for the

*Corresponding author. Tel.:
/81-52-744-2460; fax:
/81-52-744-2459. E-mail address: [email protected] (T. Kanbe). 0923-1811/03/$30.00 – 2003 Elsevier Ireland Ltd. on behalf of Japanese Society for Investigative Dermatology. All rights reserved. doi:10.1016/S0923-1811(03)00150-6

42

T. Kanbe et al.

identification of dermatophytes to the species level. – 2003 Elsevier Ireland Ltd. on behalf of Japanese Society for Investigative Dermatology. All rights reserved.

1. Introduction Dermatophyte infections constitute one of the most important groups of fungal infections in the world. Dermatophytes belonging three genera, Trichophyton , Microsporum and Epidermophyton , invade the keratinized tissues and cause infections in skin, nails, or hairs [1]. T. rubrum is the most popular etiologic agent in dermatophytosis, followed by T. mentagrophytes or T. tonsurans , and E. floccosum , M. canis or M. gypseum are also occasionally isolated as etiologic agents from patients with tinea. These dermatophytes are identified to species level by morphological features using conventional morphological techniques such as macroscopic examination of large, mature colonies and slide culture techniques. However, due to the long time required for identification and the lack of significant morphological features, the identification of dermatophyte species by conventional methods has been impeded. Molecular biology-based techniques have been adapted for the identification of dermatophytes to a species level. In these studies, the polymerase chain reaction (PCR), restriction fragment length polymorphism (RFLP) and random amplification of polymorphic DNA (RAPD) methods have been used most frequently as convenient tools for their identification [2 /6]. Sequence analysis of the DNA products is useful for identification of phylogenetically relative species of dermatophytes [7,8]. For identification of many dermatophyte species, ribosomal DNA (rDNA) or mitochondrial DNA (mtDNA) have mainly been used as subject [4,9,10]. We have focused on the DNA topoisomerase II genes of the pathogenic fungi and have reported that this gene is useful as a target not only for the study of phylogenetic relationships but also for the identification of pathogenic Candida and Aspergillus species [11 /14]. For the species-identification of the pathogenic fungi, we have used primer mixes containing several PCR primers and identified the species by the sizes of the amplified DNA fragments, without nucleotide sequencing [13,14]. More recently, we demonstrated nested PCR using primer mixes specific for the DNA topoisomerase II genes of dermatophytes for identification of the major dermatophyte species, and showed that T. rubrum , T. violaceum , M. cani s, M. gypseum and

E. floccosum could be identified by the unique size of the DNA product for each species [15]. However, we could not distinguish T. mentagrophytes from T. tonsurans since the DNA fragment was the same size. Accordingly, if we can develop a method that can identify as many dermatophyte species as possible, including T. mentagrophytes and T. tonsurans , this will be a powerful tool not only for identification of the dermatophytes but also for etiologic study of the dermatophytes. For this purpose, in this article, we designed new PCR primers common to the dermatophytes, and tested the specificity of RFLP targeting the region amplified by these primers to distinguish the dermatophytes to a species level, in addition to the previous PCR. We describe evidence showing that both the PCR and PCR-RFLP techniques targeting the DNA topoisomerase II gene are convenient and potent tools for the identification of dermatophytes and for clinical applications.

2. Materials and methods 2.1. Fungal species The 18 species and strains of dermatophytes used as subjects for PCR and PCR-RFLP analyses in this study were listed in Table 1. In addition, Trichophyton rubrum (four strains), T. mentagrophytes (four strains), T. tonsurans (four strains), Microsporum canis (two strains), M. gypseum (one strain) and Epidermophyton floccosum (four strains), which were originally isolated from patients with tinea, were also used in this study. T. mentagrophytes var. interdigitale , T. mentagrophytes var. quinckeanum and T. mentagrophytes var. erinacei are referred to as T. mentagrophytes , T. quinckeanum and T. erinacei , respectively, in this article.

2.2. DNA purification The dermatophytes were cultured on GYEP (2% glucose, 1% peptone, 0.3% yeast extract) agar for 10 /14 days at 30 8C, and conidia were suspended in GYEP broth and allowed to grow for 7 /10 days under aerobic conditions by reciprocal agitation. For purification of the genomic DNAs from T.

Identification of dermatophytes by PCR/RFLP

43

Table 1 List of dermatophyte species, strains and the database accession numbers used in this study Dermatophyte species

Database accession numbersa

Arthroderma benhamiae SM103 Arthroderma benhamiae SM166 Trichophyton rubrum CBS303.38 Trichophyton rubrum SM8765 Trichophyton verrucosum KMU4066 Trichophyton violaceum KMU4127 Trichophyton mentagrophytes var. interdigitale KMU4181 Trichophyton mentagrophytes var. interdigitale KMU4179 Trichophyton mentagrophytes var. quinckeanum SM7283 Trichophyton mentagrophytes var. erinacei SM162 Arthroderma vanbreuseghemii SM110 Arthroderma vanbreuseghemii SM111 Trichophyton tonsurans KMU4253 Trichophyton tonsurans KMU4254 Arthroderma simii CBS417.65 Arthroderma simii CBS520.75 Arthroderma obtusum JCM1907 Arthroderma persicolor JCM1911 Arthroderma gypseum RV15250 Arthroderma gypseum RV15251 Arthroderma fulvum SM181 Arthroderma fulvum SM180 Arthroderma incurvatum KMU2981 Microsporum racemosum JCM1913 Microsporum canis KMU4241 Microsporum canis NUM10118 Epidermophyton floccosum CBS240.67 Epidermophyton floccosum CBS358.93 Epidermophyton floccosum NUMK17

AB110286 AB110285 AB096064 (identical AB110283 AB096066 AB096065 (identical AB110279 AB110284 AB110278 (identical AB110282 (identical AB110280 AB110281 AB110273 AB110277 AB096068 AB110275 AB110276 (identical AB110274 AB110272 AB096067 (identical AB096069 (identical (identical

a

to AB096064 sequence)

to AB096065 sequence)

to AB110278 sequence) to AB110282 sequence)

to AB110276 sequence)

to AB096067 sequence) to AB096069 sequence) to AB096069 sequence)

Nucleotide sequences that we have deposited in GenBank/EMBL/DDBJ/International DNA Databases.

tonsurans , T. rubrum , M. canis and E. floccosum , hyphal cells were processed according to the protocol of a DNA purification kit (FastDNA Kit, Bio 101, Joshua Way, Vista, CA) [13,15]. Briefly, cells were suspended in lysis buffer (CLS-Y) with glass beads and the suspensions were vortexed to extract genomic DNA. The DNA bound to the binding matrix was finally eluted with DES, and used as a DNA template for PCR amplification. All reagents for DNA purification were included in the kit. Purification of the genomic DNAs from remaining the dermatophyte species was carried out with the method of Makimura [16]. Briefly, hyphal cells were placed in lysis buffer (200 mM Tris /HCl [pH 8.0]. 0.5% sodium dodecyl sulfate, 250 mM NaCl, 25 mM EDTA) and vortexed. The samples were then heated at 100 8C for 15 min and mixed with 3 M sodium acetate. After centrifugation at 10 000 /g for 5 min, the supernatants were extracted with phenol /chloroform /isoamyl alcohol (25:24:1), and the genomic DNA was finally precipitated by ethanol. All DNA samples were kept at /20 8C before use.

2.3. Primers and primer sets We have determined the genomic sequences of approximately 3600 bp of the DNA topoisomerase II genes from all the dermatophytes species tested in this study. Their accession numbers in the Genbank/EMBL/DDBJ International DNA databases are listed in Tables 1 and 4. These sequences in our collection were used for primer design and choice of appropriate restriction enzymes in PCR/RFLP analyses. Based on these sequences, common primers for dermatophytes species and specific primers for dermatophytes species (21 primers in total) were designed. These primers were divided into four primer sets, and designated dPsD1, dPsD2, PsT and PsME, respectively. The names of the primers contained in each set and their nucleotide sequences are listed in Table 2. All the primers contained in dPsD2, PsT and PsME were designed to be within the region amplified by dPsD1. The numbers of primers contained in each set, the expected sizes of the products generated by each

44

T. Kanbe et al.

Table 2 Primer sets, primers and their nucleotide sequences for identification of dermatophyte species Primer seta

Primer

Direction (F, R)b

Sequence (5? /3?)c

dPsD1

dDPF1 dDPR1 MCNR1

F R R

GAACNGAGAARCASATGTGGGTTTA GGTCARRGACCAYAKAGGCATCTGTA GGACCACAAGGCATCTGGGATGTC

dPsD2

dDPF2 dDPR2

F R

GTYTGGAAYAAYGGYCGYGGTATTCC RAAVCCGCGGAACCAKGGCTTCATKGG

PsT

dMF2/86 dMR2/138 TVCF2/34 (TVC)R2/76 TRBF2/253 TRBR2/346 TMTF2/38 MCNR2/138

F R F R F R F R

RCGAGGAGAGGACCCRACHTCTGAC TTCCTTAGTACCRGCYTTG GATCCACAAGGTATGTATTAGTTA GGTGCCAGCCATGTCGTAGAC GCCTGTTGTTCCGCTCATTCTT CGGCTAGGAGGGCGTGGTAGAA GCATGATTTAGAAGTGTAATGCTG TTCCTTGGTACCAGCTTTG

PsME

dTF2/87 dTR2/174 dMR2/80 MCNF2/248 MCNR2/312 MGPF2/35 EFLF213 EFLR2/346

F R R F R F F R

GGAGAGGAYHCCACTTCSGCTTCTG ARGAAGCCRGGTATTTTCAAGAGA CTTKACRGGGATRCTRGTGCCG GCTGGTAAATAACACCGATGATGG TGTATCTGATATGCATACCTTCC GGTATATACCGCCTCCCTGATG CCGATCCATTCCCTCGGTGGTT GATTCAGTTGTGACTAAGTGGACA

a

Primer sets, dPsD1, PsT and PsME were reported in the previous paper [15], and dPsD2 was designed in this study. b F, forward primer; R, reverse primer. c N, A, C, G or T; R, A or G; S, C or G; Y, C or T; K, G or T; H, A, C or T.

set and the corresponding dermatophyte species for each size of product are listed in Table 3. The primer sets dPsD1 and dPsD2 amplified products of 3390 and 2380 bp from all the dermatophytes species, respectively. In the PCR amplification using PsT, two products of 925 and 421 bp were amplified from T. violaceum because this primer set amplified two different regions of the DNA topoisomerase II gene of this fungus.

2.4. PCR conditions and agarose gel electrophoresis In the PCR amplification with the common primer set (dPsD1), the genomic DNA samples were amplified in a reaction mixture (12.5 ml) that contained 0.5 ml of each genomic DNA, 1.25 ml of 10 / buffer, 1.25 ml of dNTPs (dATP, dGTP, dCTP, dTTP, 2mM each), 1.5 ml of dPsD1 or dPsD2 and 0.25 ml of KOD Dash DNA polymerase (2.5 units/ ml) (TOYOBO Co. Ltd, Osaka, Japan). The PCR cycle parameters were as follows: preheating at 96 8C for 2 min; then 30 cycles of 96 8C for 30 s, 63 8C for 3 s and 74 8C for 120 s. The PCR products were diluted 1:200 with distilled water, and 0.5 ml of the products was then used as the DNA template for

the subsequent nested PCR using dPsD2, PsT or PsME. The PCR products generated by dPsD2 were used as substrates for restriction enzymes in the RFLP analysis (see below). The conditions of DNA amplification in the nested PCR were the same as those in the first round PCR except for the primers and the extension time. The extension time in the nested PCR was 90 s for dPsD2 or 60 s for PsT and PsME. The concentration of each primer in the primer mixes was 5 mM. All reaction mixtures were amplified using a thermal cycler (Mastercycler Gradient, Eppendorf, Netheler, Hinz GmbH, Hamburg, Germany). The PCR products generated by the PCR amplifications were analyzed by agarose-gel electrophoresis in TBE (89 mM Tris, 89 mM boric acid, 2 mM EDTA, pH 8.0) at 100 V for approximately 45 /60 min in 1.5% agarose gels (Agarose-ME, Classic Type) (Nacalai Tesque Inc., Kyoto, Japan). For agarosegel electrophoresis of the DNA fragments digested with restriction enzymes (see below), 3% (SeaPlaque GTG) agarose (FMC BioProducts, Rockland, ME) was used. All gels were stained with 0.5 mg/ml of ethidium bromide (Nacalai Tesque Inc.) in distilled water at 21 /25 8C for 20 min, and the gels were then destained at 21 /25 8C for 20 /30 min in

Identification of dermatophytes by PCR/RFLP

45

Table 3 Primer sets, the sizes of expected DNA products generated from each set and corresponding species of dermatophytes to each DNA fragment Primer seta

Size of DNA product (bp)

Corresponding species

dPsD1

3390b

Trichophyton spp. Microsporum spp. E. floccosum

dPsD2

2380c

Trichophyton spp. Microsporum spp. E. floccosum

PsT

925 925
/421d

T. rubrum T. violaceum

392e

T. interdigitale T. quinckeanum A. vanbreuseghemii T. tonsurans A. simii

522f No band

PsME

Microsporum spp. g

T. verrucosum A. benhamiae T. erinacei E. floccosum

464 639 1336 874h

M. gypseum M. canis E. floccosum Trichophyton spp.

No bandi

A. incurvatum A. fulvum M. racemosum A. persicolor A. obtusum

a

The names of primer sets correspond to those in Table 1. dPsD1 amplify a product of 3390 bp from Trichophyton spp., Microsporums pp. and E. floccosum . c dPsD2 amplify a product of 2380 bp from Trichophyton spp,, Microsporums pp. and E. floccosum . d PsT amplify two products of 925 and 421 bp from T. violaceum . e PsT amplify a product of 392 bp from T. interdigitale , T. quinckeanum , A. vanbreuseghemii , T. tonsurans and A. simii . f PsT amplify a common product of all the Microsporum spp. and associated Arthroderma teleomorph spp. g PsT does not amplify any products from T. verrucosum , A. benhamiae , T. erinacei and E. floccosum. h PsME amplify a common product of 874 bp from all the Trichophyton spp. i PsME does not amplify any products from A. incurvatum , A. fulvum , M. racemosum , A. persicolor and A. obtusum . b

distilled water. The DNA products were visualized with a UV transilluminator and photographed.

2.5. Restriction enzymes and PCR-RFLP analysis For RFLP analysis, the major product (2380 bp) amplified by dPsD2 were purified using a MinEluteTM PCR purification kit (QIAGEN GmbH, Hilden, Germany) from each PCR solution, and used as the substrates for restriction enzymes. From the geno-

mic sequences of the regions amplified by dPsD2, we used Hin c II, Hinf I, Afl II or Pfl M I (Bio Labs, Beverly, MA) in order to identify the species by the unique fragment pattern specific to each species. Hin c II and Hinf I were used for all the dermatophytes, Afl II was used to distinguish T. verrucosum from other dermatophytes species, and Pfl M I was used to distinguish A. vanbreuseghemii from T. mentagrophytes and T. quinckeanum . The DNA samples were digested for 6 h at 37 8C for Hinc II, Hin f I, Afl II or Pfl M1. After enzymatic digestion,

46

DNA fragments were analyzed by agarose-gel electrophoresis (see above). The sizes of the fragments generated from each product by the enzymatic digestion are listed in Table 3. These sizes were expected from the nucleotide sequences of the region amplified by dPsD2, and the accession numbers used are listed in Table 4.

2.6. Nucleotide sequencing The amplified DNAs were purified from agarose gels using a QIAEX II Gel Extraction Kit (QIAGEN GmbH). The common primer for the dermatophytes (dDPF2) was used for sequencing the products. The DNA samples were prepared according to the protocol of an ABI PRISM BigDyeTM Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City, CA) and sequenced with a DNA sequencer (ABI PRISMTM 310 Genetic Analyzer, PE Applied Biosystems). The sequences were analyzed with the program GENETIX-MAC version 11.0.0.

2.7. Sensitivity of PCR amplifications Conidia of T. rubrum , T. mentagrophytes or E. floccosum were cultured on GYEP agar plates and cultured for 2 /7 days at 30 8C. Genomic DNAs were purified from each of various sizes of colonies (2 / 15 mm in diameter) using the DNA purification kit (see above). Each DNA sample was amplified by dPsD1, followed by subsequent nested PCR using PsT for T. rubrum and T. mentagrophytes or PsME for E. floccosum . The amplification profiles were compared with those of the direct PCR, in which the genomic DNAs were directly amplified by PsT and PsME.

2.8. Identification of dermatophytes by conventional techniques Scales of patients with tinea, that contained fungal cells under microscopic observation using the KOH technique, were cultured on GYEP agar containing 50 mg/ml of chloramphenicol for 7 /14 days at 30 8C. Subsequently, the fungal species was identified by the colony characteristics on Sabouraud agar plates, and by microscopic observation of the conidia and other structures using a slide culture technique.

T. Kanbe et al.

3. Results 3.1. Amplification profiles of DNA fragments by PCR using PsT or PsME The PCR using dPsD1, in which the genomic DNAs purified from each dermatophyte species were used as the DNA templates, amplified a product of 3390 bp from all the dermatophyte species, but no products from fungi other than the dermatophytes such as Candida , Aspergillus , Penicillium, Cryptococcus and Fusarium species (data not shown). All the products were then used as DNA templates for the subsequent PCR using PsT and PsME. The expected sizes of products amplified by PsT or PsME are shown in Table 3. Using PsT, a product of 925 bp was amplified from T. rubrum (Fig. 1A, arrow a), two products of 925 bp (Fig. 1A, arrow a) and 421 bp (Fig. 1A, white arrow) were amplified from T. violaceum , and a product of 392 bp was commonly amplified from T. mentagrophytes , T. quinckeanum , A. vanbreuseghemii , T. tonsurans and A. simii (Fig. 1A, arrow b). No products were generated from A. benhamiae , T. erinacei and T. verrucosum . For Microsporum species and E. floccosum , PsT commonly amplified a common band of 522 bp among all the Microsporum species, but not from E. floccosum (Fig. 1C, arrow d). On the other hand, PsME amplified species-specific products of 464 bp from A. gypseum , 639 bp from M. canis and 1336 bp from E. floccosum (Fig. 1D, arrows ‘e’ for E. floccosum , ‘f’ for M. canis and ‘g’ for A. gypseum ), but did not amplify any products from the remaining A. obtusum, A. persicolor, A. incurvatum, A. fulvum and A. racemosum (Fig. 1D). For the Trichophyton species, PsME amplified a common product of 874 bp from all the Trichophyton species (Fig. 1B, arrow c). When the PCR products were purified and sequenced, all sequence data coincided with those of the dermatophytes used (data not shown). This indicates that all the products amplified by both PsT and PsME were species-specific.

3.2. Sensitivity of the PCR amplifications The genomic DNAs purified from the various sizes of colonies of T. rubrum and E. floccosum were amplified by PsT and PsME, in order to determine the sensitivities of the direct and nested PCRs. In both species, the intensity of the major band generated from the DNA samples derived from a 2 mm-colony in the direct PCR, in which the genomic DNAs were amplified by the primer sets, was much

Identification of dermatophytes by PCR/RFLP

47

Fig. 1 Amplification profiles of DNA products from the dermatophytes by nested PCR using PsT or PsME. The PCR products generated by dPsD1 from the Trichophyton species (A and B), or Microsporum species and E. floccosum (C and D) were used as DNA templates for the nested PCR using PsT (A and C) or PsME (B and D). PsT amplified speciesspecific products (A, arrows ‘a’ for T. rubrum , and both ‘a’ and ‘white arrow’ for T. violaceum ), a common band from five Trichophyton species (A, arrow ‘b’), and a genus-specific band for Microsporum (C, arrow ‘d’). PsME amplified species-specific products (D, arrow ‘f’ for M. canis , ‘g’ for M. gypseum and ‘e’ for E. floccosum ) and a genus-specific fragment for Trichophyton (C, arrow ‘c’). M indicates marker DNAs. The numbers on the left side of each panel indicate the sizes of the molecular markers (bp).

lower than that in the nested PCR, however, their band intensities were sufficient to determine the dermatophyte species by the size of the product (Fig. 2A for T. rubrum , B for E. floccosum ). In the direct PCR, several faint bands were generated along with the major band (Fig. 2B, left panel), but these were clearly less prominent than the major band. Similar results were obtained in T. mentagrophytes (data not shown).

3.3. PCR-RFLP analysis of dermatophytes The dPsD2 exclusively amplified a product of 2380 bp from the PCR products amplified by dPsD1 of all the dermatophyte species (Fig. 3, arrow). The major PCR products were purified and used as substrates for the RFLP analysis using Afl II, Hin c II, Hin f I and Pfl M I. The DNA substrates of the Trichophyton species were digested with Hin c II

or Hin f I, and the restriction profiles were analyzed by agarose-gel electrophoresis (Fig. 4A for Hin c II and 4B for Hin f I). Using Hinc II, A. benhamiae SM166, T. rubrum and T. violaceum were distinguished, and on the other hand, using Hin f I, A. benhamiae SM103, T. rubrum , T. violaceum , T. tonsurans and A. simii were distinguished. Thus, using Hin c II and Hin f I, six of the ten Trichophyton species were identified by RFLP. For the remaining four Trichophyton species, Afl II and Pfl M I were used. RFLP using Afl II was useful for the identification of T. verrucosum because Afl II sites were only present in this fungus (Table 4). As expected, two fragments (1601 and 783 bp) were generated from T. verrucosum , but not from other any dermatophytes (Fig. 4C). Furthermore, using Pfl M, A. vanbreuseghemii was distinguished from T. mentagrophytes and T. quinckeanum (Fig. 4D). T. mentagrophytes was not distinguished from T.

48

T. Kanbe et al.

Fig. 2 Sensitivities of PCR amplifications using PsT or PsME. Genomic DNAs purified from various sizes of colonies (2 to 15 mm in diameter) of T. rubrum and E. floccosum , that had been cultured on GYEP plates for 2 /7 days at 30 8C, were amplified by dPsD1, followed by nested PCR using PsT for T. rubrum and PsME for E. floccosum (Nested PCR). Alternatively, the genomic DNAs were directly amplified by PsT or PsME (Direct PCR). In both species, all cases and both PCR primer sets, the major product specific for each species was generated (A, arrow for T. rubrum ; B, arrow for E. floccosum ). M indicates marker DNAs. The numbers on each lane indicate the diameters of colonies (mm). The numbers on the left side of each panel indicate the sizes of the molecular markers (bp).

quinckeanum by any of the restriction enzymes tested in this study.

For the identification of Microsporum species, associated Arthroderma teleomorph species and E. floccosum , RFLP analysis using Hin c II, Hin f I or

Fig. 3 Specific amplification of a DNA fragment from the dermatophyte species by dPsD2. The genomic DNAs were amplified by dPsD1, and then amplified by dPsD2. A product of 2380 bp was generated from all the species (arrow). The major products were purified and used for the RFLP. M indicates marker DNAs. The numbers on the left side of the panel indicate the sizes of the molecular markers (bp).

Identification of dermatophytes by PCR/RFLP

49

Fig. 4 RFLP-based identification of Trichophyton to a species level. The PCR products generated by dPsD2 were purified and digested with Hin c II (A), Hin f I (B), Afl II (C) or Pfl M I (D). Using Hin c II and Hin f I, almost all the dermatophyte species were identified to the species level by the unique fragment patterns (A for Hin c II and B for Hin f I). T. verrucosum and A. vanbreuseghemii were identified by Afl II (C) and Pfl M I (D), respectively. M indicates marker DNAs. The numbers on the left side of each panel indicate the sizes of the molecular markers (bp).

Xcm I was performed. Using Hinc II, all the species tested were identified to a species level (Fig. 5A). Similarly, using Hin f I, all the Microsporum species were distinguished to the species level by the unique restriction profile specific to each species (Fig. 5B).

3.4. Identification of clinical isolates by nested PCR and RFLP In order to assess the stability and clinical application of the PCR using PsT or PsME, and RFLP targeting the DNA topoisomerase II genes of dermatophytes, genomic DNAs were purified from T. rubrum , T. mentagrophytes , T. tonsurans , M. canis , M. gypseum and E. floccosum , which had been isolated from clinical specimens and identified by conventional morphological techniques. All the genomic DNAs were amplified with dPsD1 and then used as DNA templates for nested PCR using PsT or PsME. Using PsT, the product of 925 bp was amplified from T. rubrum and the product of 392 bp was amplified from both T. mentagrophytes and T. tonsurans (Fig. 6, PCR/PsT, left panel). The product of 522 bp was amplified from both M. canis and M. gypseum but not from E. floccosum (Fig. 6, PCR/PsT, right panel). On the other hand, using

PsME, the product of 874 bp was amplified from all the Trichophyton species (Fig. 6, PCR/PsME, left panel), and the products of 639, 464 and 1336 bp were amplified from M. canis , M. gypseum and E. floccosum , respectively, (Fig. 6, PCR/PsME, right panel). The PCR products generated by dPsD1 were amplified by dPsD2, and the PCR products were digested with Hin f I. All the species tested were identified to a species level by the restriction profiles unique to each species (Fig. 6, RFLP/Hin f I). All of the restriction profiles of each species coincided with those of the reference strains (SM8765 for T. rubrum , KMU4178 for T. mentagrophytes , KMU4253 for T. tonsurans , KMU4241 for M. canis , RV15250 for M. gypseum and CBS358.93 for E. floccosum ).

4. Discussion Recently, we demonstrated PCR targeting of the DNA topoisomerase II gene for the identification of dermatophyte species, and reported that this PCR distinguished T. rubrum , T. violaceum , M. canis , M. gypseum and E. floccosum by the unique sizes of products for each species, however, T. menta-

50

T. Kanbe et al.

Table 4 Restriction enzymes and the expected sizes of DNA fragments generated by enzymatic digestion Dermatophyte speciesa

A. benhamiae I

Restriction enzyme and DNA fragment (bp)b Afl II

Hin c II

NCSd

1879, 159, 151, 129, 62 1209, 482, 261, 192, 178, 58 1879, 310, 129, 62 954, 482, 370, 261, 255, 58 1879, 159, 153, 129, 62 1209, 482, 370, 263, 58 1879, 159, 151, 129, 66 1213, 482, 370, 261, 58

A. benhamiae II T. erinacei T. verrucosum

NCS NCS 1601, 783 T. rubrum NCS T. violaceum NCS T. mentagrophytes NCS T. quinckeanum

NCS

A. vanbreuseghemii T. tonsurans

NCS NCS

A. simii I

NCS

A. simii II

NCS

A. gypseum I

NCS

A. gypseum II

NCS

A. fulvum

NCS

A. incurvatum

NCS

A. obtusum

NCS

A. persicolor

NCS

M. canis A. racemosum

NCS NCS

E. floccosum

NCS

Hinf I

1670, 400, 159, 152 1267, 482, 370, 262 1670, 400, 159, 141 1267, 482, 370, 251 1479, 400, 310, 129, 62 1209, 482, 233, 166, 137, 95, 58 1479, 400, 310, 129, 62 1209, 482, 233, 166, 137, 95, 58 1479, 400, 310, 129, 62 1209, 482, 233, 166, 137, 95, 58 1479, 400, 159, 151, 1209, 482, 233, 166, 95, 129, 62 70, 67, 58 1479, 400, 159, 151, 1209, 482, 261, 178, 137, 129, 62 58, 55 1479, 400, 159, 151, 1209, 482, 261, 178, 137, 129, 62 58, 55 1444, 394, 307, 164, 62 954, 482, 258, 255, 232, 132, 58 1444, 394, 307, 164, 62 954, 482, 258, 255, 232, 132, 58 1146, 462, 394, 312, 62 902, 313, 307, 263, 186, 178, 169, 58 1494, 394, 311, 114, 66 637, 317, 313, 262, 259, 186, 178, 169, 58 1444, 394, 159, 150, 1209, 364, 313, 260, 169, 129, 62, 35 58 1001, 443, 358, 164, 1012, 482, 261, 255, 178, 159, 151, 62, 36 131, 55 1160, 706, 346, 164 846, 572, 564, 263, 73, 58 1306, 394, 159, 153, 1267, 536, 263, 213, 97 138, 129, 62, 35 1444, 394, 164, 159, 954, 482, 260, 255, 186, 150, 62 178, 58

Accession number usedc PflM I 1625, 755

AB110286

1625, 755 1627, 755 1625, 759

AB110285 AB110284 AB110283

1626, 755 1615, 755 1625, 755

AB096064 AB096066 AB096065

1625, 755

AB110279

1185, 755, AB110278 440 1625, 755 AB110282 1625, 755

AB110280

1625, 755

AB110281

1616, 141 1616, 141 1621, 141 1761,

AB110274

614, AB096068 614, AB110275 614, AB110276 618

NCS

AB110273

1619, 755

AB110277

1621, 755 1621, 755

AB096067 AB110272

1618, 755

AB096069

a A. benhamiae I for A. benhamiae SM103, A. benhamiae II for A. benhamiae SM166, A. simii I for A. simii CBS417.65, A. simii II for A. simii CBS520.75, A. gypseum I for A. gypseum RV15250 and A. gypseum II for A. gypseum RV15251. b The sizes of DNA fragments generated from each PCR product by digestion with each restriction enzyme. c Accession numbers (GenBank/ EMBL/DDBJ International DNA databases) used for RFLP analysis. d NCS indicates no cutting site.

grophytes was not distinguished specifically from T. tonsurans by the size of the product [15]. In the present study, the PCR using PsT and PsME, in which the genomic DNA’s of some other Trichophyton and Microsporum species were amplified, showed that a product of similar size was generated not only from T. mentagrophytes and T. tonsurans but also from T. quinckeanum , A. vanbreuseghemii , and A.

simii , and no bands were generated from A. benhamiae , T. erinacei or T. verrucosum . These results indicate that the PCR using PsT and PsME is specific for the five species, T. rubrum , T. violaceum , M. canis , M. gypseum and E. floccosum , but not for other Trichophyton species. Similarly, the PCR using PsME is specific for M. canis , M. gypseum and E. floccosum , but not for other Microsporum

Identification of dermatophytes by PCR/RFLP

51

Fig. 5 RFLP-based identification of Microsporum and E. floccosum to a species level. The PCR products generated by dPsD2 were purified and digested with Hin c II (A) or Hin f I (B). All Microsporum species and E. floccosum were identified by RFLP using Hin c II or Hin f I. M indicates marker DNAs. The numbers on the left side of each panel indicate the sizes of the molecular markers (bp).

species. Furthermore, the amplification of the common band indicates that PsT and PsME identify all the Trichophyton and Microsporum species at the genus level. We tried to design other primers specific for each dermatophyte species, but could not obtain any other primers due to the difficulty in designing species-specific primers for the highly conserved sequences. Sequencing of the PCR products is the most powerful method for the correct identification of dermatophytes to a species level [4,16], however, this is not convenient for processing large numbers of samples with regard to cost and time. Liu et al. [17] reported that PCR using random primers identified 20 dermatophyte species by the distribution patterns of DNA bands. In the case of identification by the distribution patterns of DNA bands amplified by random primers, it is generally accepted that RAPD is generally considered to be a poorly reproducible method for technical reasons. The presence of other DNA such as bacterial DNA or host DNA may influence the band patterns of the amplified DNAs, especially when DNA samples prepared directly from clinical samples were used as DNA templates. As compared with RAPD, the PCR using PsT or PsME specifically amplified the species-specific DNA

fragment from each species, even in the presence of other DNAs [15]. In this study, the RFLP analysis targeting the DNA topoisomerase II gene was quite powerful for the identification of Microsporum species because the RFLP using Hin c II or Hinf I identified all the Microsporum species including the species that were not identified by the PCR using PsME. Almost all of the Trichophyton species were also identified by the RFLP analysis using Hinc II, Hin f I, Afl II and Pfl M I. Of the restriction enzymes, Afl II specifically identified T. verrucosum because the sites for this enzyme were only present in T. verrucosum . T. mentagrophytes var. interdigitale was not distinguished from T. mentagrophytes var. quinckeanum by RFLP analysis using any of the restriction enzymes. Unfortunately, this may be a limitation of our RFLP analysis because the nucleotide sequence of the DNA topoisomerase II gene of T. mentagrophytes is extremely similar to that of T. quinckeanum (the difference in the nucleotides of the DNA topoisomerase II gene is only six per 4000 bp). PCR identification using random primers may distinguish these two species [18]. RFLP-based identification has been used for species-identification of Trichophyton , Microsporum and E. floccosum [19,20]. Furthermore, the results obtained

52

T. Kanbe et al.

Fig. 6 Identification of clinical isolates by PCR and RFLP techniques. The genomic DNAs purified from clinical isolates (19 strains), which were originally isolated from patients with tinea, were amplified by dPsD1. The PCR products were then amplified by nested PCR using PsT (A: PCR/PsT) or PsME (B: PCR/PsME). In all cases, the species or genus-specific product was amplified from each sample. The products generated by dPsD2 were digested with Hin f I, and the RFLP patterns were analyzed by agarose gel electrophoresis (C: RFLP/HIn f I). The RFLP using Hin f I distinguished all the species of the clinical isolates, including T. tonsurans , by the unique RFLP pattern for each species. No differences in the RFLP profiles within the same species were found. The reference strains used in this experiment were the following: T. rubrum SM8765 for T. rubrum , T. mentagrophytes KUM4181 for T. mentagrophytes , T. tonsurans KUM4253 for T. tonsurans , M. canis KMU4241 for M. canis , A. gypseum RV15250 for M. gypseum and E. floccosum CBS358.93 for E. floccosum . M indicates marker DNAs. The numbers on the left side of each panel indicate the sizes of the molecular markers (bp).

from this study showed that the RFLP is valuable for the identification of Trichophyton , Microsporum and E. floccosum . For the major dermatophyte species that we have generally isolated from patients, the PCR using PsT and PsME, and PCRRFLP using Hin f I are available. rDNA or mtDNA have been used for RFLP studies on the phylogenetic relationships or identification of dermatophytes to a species level, however, large amounts of DNAs were necessary for the analysis of the DNA fragment patterns. In this study, we have used the PCR products, which were generated by

dPsD2 as substrates for the restriction enzymes. dPsD2 is common to all the dermatophyte species used, and stably generates the DNA product of 2380 bp from each dermatophyte species used. Furthermore, in the experiment to test the sensitivities of the PCR using PsT and PsME, the species-specific DNA bands were amplified from colonies of approximately 2 mm in diameter. The experiments using several clinical isolates showed that the PCR and PCR-RFLP showed high stability and reproducibility. These results indicate that our PCR and PCR-RFLP are able to identify almost all of the

Identification of dermatophytes by PCR/RFLP

dermatophyte species within a few days (or from 2mm colonies). Recently, Mochizuki et al. reported that PCR-RFLP was available for the rapid identification of T. tonsurans [21]. When the reproducibility of the DNA fragments in the PCR using PsT and PsME, and the restriction profiles in RFLP using Hin f I were tested using clinical isolates (300 strains of T. rubrum and 100 strains of T. mentagrophytes ), no differences in the amplification and restriction profiles were observed in either species (unpublished data). This indicates that these species are homogenous at the nucleotide level of the DNA topoisomerase II gene, and both PCR and RFLP targeting the DNA topoisomerase gene II are quite stable as tools for identification of dermatophytes to the species level. Jackson et al. [7,22] reported the presence of subrepeat elements in the rDNA nontranscribed spacer region of T. rubrum , and described that these elements were useful as targets for strain typing by PCR because T. rubrum has a variety of sizes of the element. In addition, Gupta et al. [20]found genotype variations in T. rubrum and T. mentagrophytes , but not T. tonsurans , which had been serially isolated from patients, by digestion of the genomic DNA with Eco RI. It is very interesting and important to correctly identify T. rubrum and T. mentagrophytes at the strain level for etiologic study of this fungus with regard to surveillance of dermatophyte dissemination in our environment. Prior to the strain or subtype identification, a method that can correctly and rapidly identify many clinical isolates to a species level is necessary. For this purpose, we believe that the PCR and PCR-RFLP targeting the DNA topoisomerase II genes of the dermatophytes demonstrated here are quite useful and potent tools for identification of the dermatophyte species usually isolated from patients with tinea, and are adaptable for a large scale survey of the dermatophytes. Epidemiological studies of the dermatophytes at the strain and species levels in patients with tinea are under investigation.

Acknowledgements This study was supported by grants from the Ministry of Education, Science, Sports and Culture of Japan (Grant-in-Aid for Scientific Research on Priority Areas (B), (to AK).

References [1] Weitzman I, Summerbell RC. The dermatophytes. Clin Microbiol Rev 1995;8:240 /59.

53 [2] Faggi E, Pini G, Campisi E, Bertellini C, Difonzo E, Mancianti F. Application of PCR to distinguish common species of dermatophytes. J Clin Microbiol 2001;39:3382 /5. [3] Mochizuki T, Sugie N, Uehara M. Random amplification of polymorphic DNA is useful for the differentiation of several anthropophilic dermatophytes. Mycoses 1997;40:405 /9. [4] Mochizuki T, Kawasaki M, Ishizaki H, Kano R, Hasegawa A, Tosaki H, et al. Molecular epidemiology of Arthroderma benhamiae , an emerging pathogen of dermatophytes in Japan, by polymorphisms of the non-transcribed spacer region of the ribosomal DNA. J Dermatol Sci 2001;27:14 / 20. [5] Liu D, Coloe S, Baird R, Pedersen J. Molecular determination of dermatophyte fungi using the arbitrarily primed polymerase chain reaction. Br J Dermatol 1997;137:351 /5. [6] Graser Y, el Fari M, Presber W, Sterry W, Tietz HJ. Identification of common dermatophytes (Trichophyton, Microsporum, Epidermophyton ) using polymerase chain reactions. Br J Dermatol 1998;138:576 /82. [7] Mochizuki T, Kawasaki M, Ishizaki H, Makimura K. Identification of several clinical isolates of dermatophytes based on the nucleotide sequence of internal transcribed spacer 1 (ITS 1) in nuclear ribosomal DNA. J Dermatol 1999;26:276 / 81. [8] Makimura K, Mochizuki T, Hasegawa A, Uchida K, Saito H, Yamaguchi H. Phylogenetic classification of Trichophyton mentagrophytes complex strains based on DNA sequences of nuclear ribosomal internal transcribed spacer regions. J Clin Microbiol 1998;36:2629 /33. [9] Jackson CJ, Barton RC, Evans EG. Species identification and strain differentiation of dermatophyte fungi by analysis of ribosomal-DNA intergenic spacer regions. J Clin Microbiol 1999;37:931 /6. [10] Kawasaki M, Aoki M, Ishizaki H, Nishimura K, Miyaji M. Phylogeny of Epidermophyton floccosum and other dermatophytes. Mycopathologia 1996;134:121 /8. [11] Kim K-H, Kanbe T, Akashi T, Mizuguchi I, Kikuchi A. Identification of a single nuclear localization signal in the C-terminal domain of an Aspergillus DNA topoisomerase II. Mol Genet Genomics 2002;268:287 /97. [12] Kato M, Ozeki M, Kikuchi A, Kanbe T. Phylogenetic relationship and mode of evolution of yeast DNA topoisomerase II gene in the pathogenic Candida species. Gene 2001;272:275 /81. [13] Kanbe T, Horii T, Arishima T, Ozeki M, Kikuchi A. PCR-based identification of pathogenic Candida species using primer mixes specific to Candida DNA topoisomerase II genes. Yeast 2002;19:973 /89. [14] Kanbe T, Yamaki K, Kikuchi A. Identification of the pathogenic Aspergillus species by nested PCR using a mixture of specific primers to DNA topoisomerase II gene. Microbiol Immunol 2002;46:841 /8. [15] Kanbe T, Suzuki Y, Kamiya A, Mochizuki T, Fujihiro M, Kikuchi A. PCR-based identification of common dermatophyte species using primer sets specific for the DNA topoisomerase II genes. J Dermatol Sci 2003;32:151 /61. [16] Makimura K, Tamura Y, Mochizuki T, Hasegawa A, Tajiri Y, Hanazawa R, et al. Phylogenetic classification and species identification of dermatophytes strains based on DNA sequences of nuclear ribosomal internal transcribed spacer I regions. J Clin Microbiol 1999;37:920 /4. [17] Liu D, Coloe S, Baird R, Pedersen J. Application of PCR to the identification of dermatophyte fungi. J Med Microbiol 2000;49:493 /7. [18] Liu D, Coloe S, Baird R, Pedersen J. PCR identification of Trichophyton mentagrophytes var. interdigitale and T.

54 mentagrophytes var. mentagrophytes dermatophytes with a random primer. J Med Microbiol 2000;46:1043 /6. [19] Blanz P, Buzina W, Ginter G, Graser Y. Molecular biological methods and their consequences in taxonomy and diagnosis of dermatophytes. Mycoses 2000;43(Suppl 1):11 /6. [20] Gupta AG, Kohli Y, Summerbell RC. Variation in restriction fragments length polymorphisms among serial isolates from patients with Trichophyton rubrum infection. J Clin Microbiol 2001;39:3260 /6.

T. Kanbe et al. [21] Mochizuki T, Tanabe H, Kawasaki M, Ishizaki H, Jackson CJ. Rapid identification of Trichophyton tonsurans by PCR-RFLP analysis of ribosomal DNA regions. J Dermatol Sci 2003;32:25 /32. [22] Jackson CJ, Barton RC, Kelly SL, Evans EGV. Strain identification of Trichophyton rubrum by specific amplification of subrepeat elements in the ribosomal DNA nontranscribed spacer. J Clin Microbiol 2000;38:4527 /34.