ITS-RFLP optimization for dermatophyte identification from clinical sources in Alagoas (Brazil) versus phenotypic methods

Introduction: Dermatophytoses are superficial mycoses, and the identification of their etiological agents is routinely performed by culture and microscopic features, which is time-consuming and relies on personnel expertise. Molecular approaches have been developed to provide faster and reliable results; therefore, this study aimed to identify dermatophytes isolated from Alagoas state patients, employing phenotypical and molecular methods. Methodology: Clinical samples for morphological identification were collected from private and public laboratories and cultivated on Sabouraud dextrose agar. DNA extraction was followed by ITS amplicon analysis after restriction enzyme digestion Dde I (ITS-RFLP). Results: Out of fourteen representative strains, ITS-RFLP with Dde I efficiently identified Microsporum canis , Nannizzia gypsea , and Trichophyton rubrum, while species of the complex T. tonsurans / T. mentagrophytes presented the same restriction pattern. After genotyping, 2 T. tonsurans and 1 Microsporum sp. strain were reclassified as T. rubrum . Conclusions: RFLP of ITS-region followed by Dde I digestion produced faster and relatively reliable results than classic methods; however, this method has not been as efficient for closely related dermatophytes cryptic species.


Introduction
Dermatophytoses are superficial infections diagnosed frequently in the dermatological clinical routine, caused by filamentous fungi (dermatophytes) which attack keratinized tissues in humans and animals [1]. Advances in phylogenetic taxonomy culminated in the adoption of the polyphasic nomenclature, with the recognition of nine holomorphic genera of dermatophytes, of which seven had species that could cause infections in humans, including the anamorphic genera (Trichophyton, Microsporum, and Epidermophyton), Nannizzia, Paraphyton, Lophophyton, and Arthroderma [2][3][4]. Dermatophytes thrive under humid conditions and temperatures ranging from 25-28 ºC, making skin and its annexes excellent environments for their growth and spread [5].
According to their ecological habitat, dermatophytes are classified as anthropophilic, geophilic, or zoophilic, and can cause dermatophytosis in humans, as well as in small and large animals, configuring an important zoonotic and public health problems due to the potential risk of transmission from animals to humans [6,7], which can be amplified with the transmission between domestic animals. Etiological agent identification is essential for epidemiological purposes and accurate infection management since there is clinical significance: anthropophilic species cause more chronic and non-inflammatory dermatophytosis, while zoo-geophilic species cause acute and inflammatory mycoses [2]. The clinical routine of dermatophyte identification is based on phenotypical observation of culture and direct microscopic examination, low-cost methods that rely on technical expertise and diagnosis, often timeconsuming and inconclusive at the species level [8].
The main fungal etiological agents on the American continent belong to the Trichophyton genus [9][10][11][12][13], as found in studies from northeastern Brazil, confirming the high prevalence of the genus [6,14]. Although they affect approximately one-quarter of the population worldwide [15], dermatophytoses are often neglected due to their non-invasive nature, with limited or no medical care whatsoever, increasing the pathogen spread and chronic mycoses cases [16,17].
Restriction enzyme analysis (REA) and restriction fragment length polymorphism (RFLP) have shown satisfactory results in dermatophyte species identification for faster and reproducible diagnosis [18]. Molecular approaches based on ribosomal DNA (rDNA) internal transcribed spacer (ITS) sequence have been employed as the best method to close the gap in dermatophyte identification [18][19][20]. Once PCR-based methods increase the accuracy of the identification, the present study aimed to track patients diagnosed with dermatophytosis by phenotypic methods in a public hospital and private laboratories for molecular identification of etiological agents, comparing the phenotypic with molecular results to elucidate the most common species in Alagoas.

Clinical strains
This research was approved by the Research Ethics Committee (No. 23065.017665/2011-64) (SisGen No. A5B1165), according to resolution 466/2012 (Brazil). Patients who underwent medical evaluation due to lesions with suspected dermatophyte infection were referred for laboratory diagnosis in a public hospital and 3 private laboratories in Maceió (Alagoas/Brazil), and clinical specimens were collected and processed using phenotypic methods for fungal structure identification via direct microscopic examination with lactophenol cotton blue and cultivated on Sabouraud dextrose agar (SDA; 14 days, 37 ºC). Among positive samples, representative strains were cultured in our laboratory, and stored in distilled water for reactivation on potato dextrose agar or SDA with chloramphenicol (0.05 g/L) (Sigma-Aldrich, Burlington, USA) at 37 ºC.

Discussion
Superficial mycotic infections, including dermatophytosis, are considered the most common fungal infections worldwide [16,21]. Classic laboratory diagnosis relies upon morphological features under culture on solid media and microscopy (macro and micromorphology), which can vary according to outside factors such as temperature incubation, medium, and chemotherapy [6,22]. Since dermatophytes can be similar or frequently vary into the same species, phenotypical methods are subject to misidentifications between species and even genera [6] and molecular methods have been frequently applied for proper identification.
Here we identified 12 isolates among 14 by ITS-RFLP as belonging to the Trichophyton genus, an expected finding since it has been reported as the most prevalent in Latin America, Brazil, and in Alagoas State [6,9,11,13]. It is important to note that the reference strain ATCC 9533, previously T. mentagrophytes var. interdigitale, has recently been reclassified as T. interdigitale and based on the knowledge that T. tonsurans, T. mentagrophytes, and T. interdigitale have similar ITS sequences and share a taxonomic complex [23], strains phenotypically identified as the first two but presenting the same restriction pattern as T. interdigitale was considered as a match. Isolate MF07, phenotypically identified as T. tonsurans, did not match  any of the available ATCC strains, but it was considered as a mismatch since it lacks the > 400 bp fragment after DdeI digestion reported in other works for this species [24]. ITS-RFLP was shown to be a reproducible, fast, and reliable method to identify important dermatophytes species from clinical samples [19,25]; however, for some strains this technique was not sufficiently precise for the identification of species level [26,27]. Regarding enzyme choice, many enzymes were tested for the digestion of dermatophyte ITS amplicon, such as MvaI, HinfI, DdeI, and HaeIII, with MvaI being the most widely tested and chosen by researchers as the best-performing enzyme [26,28]. DdeI restriction patterns for T. interdigitale and T. rubrum found in our study were confirmed by another report [28], as well as the M. canis and N. gypsea digestion patterns [26]. Although DdeI is a suitable enzyme for the differentiation of some dermatophyte species, the distinction between closely related species such as T. rubrum/T. raubitschenkii or M. canis/M. audouiini using DdeI has proven to be difficult [26]. When we compared the digestion patterns of the strains by ITS-RFLP/DdeI with the results of phenotypic methods, 4 isolates previously misidentified were reclassified, as reported in other studies concerning molecular identification of dermatophytes [19,25,29].
Ghojoghi et al. [30] emphasized the importance of accurate identification of dermatophytes for appropriate treatment and control of potential environmental sources of infection after misidentifying 58 samples by phenotypic methods, re-identifying as T. interdigitale and T. rubrum after ITS-RFLP. PCR-based methods have brought advances in the distinction between species and strains but are not yet applied in the routine clinical diagnosis of dermatophytosis due to the high cost of equipment, reagents, and specialized professionals [28].

Conclusions
This was the first molecular approach to dermatophytes in our state. ITS-RFLP analysis of the ITS-rDNA followed by DdeI treatment produced faster and more reproducible results than classical culturebased methods. However, the definitive molecular identification for less frequent dermatophyte species can only be achieved through DNA sequencing.