Toward a Novel Multilocus Phylogenetic Taxonomy for the Dermatophytes

Type and reference strains of members of the onygenalean family Arthrodermataceae have been sequenced for rDNA ITS and partial LSU, the ribosomal 60S protein, and fragments of β-tubulin and translation elongation factor 3. The resulting phylogenetic trees showed a large degree of correspondence, and topologies matched those of earlier published phylogenies demonstrating that the phylogenetic representation of dermatophytes and dermatophyte-like fungi has reached an acceptable level of stability. All trees showed Trichophyton to be polyphyletic. In the present paper, Trichophyton is restricted to mainly the derived clade, resulting in classification of nearly all anthropophilic dermatophytes in Trichophyton and Epidermophyton, along with some zoophilic species that regularly infect humans. Microsporum is restricted to some species around M. canis, while the geophilic species and zoophilic species that are more remote from the human sphere are divided over Arthroderma, Lophophyton and Nannizzia. A new genus Guarromyces is proposed for Keratinomyces ceretanicus. Thirteen new combinations are proposed; in an overview of all described species it is noted that the largest number of novelties was introduced during the decades 1920–1940, when morphological characters were used in addition to clinical features. Species are neo- or epi-typified where necessary, which was the case in Arthroderma curreyi, Epidermophyton floccosum, Lophophyton gallinae, Trichophyton equinum, T. mentagrophytes, T. quinckeanum, T. schoenleinii, T. soudanense, and T. verrucosum. In the newly proposed taxonomy, Trichophyton contains 16 species, Epidermophyton one species, Nannizzia 9 species, Microsporum 3 species, Lophophyton 1 species, Arthroderma 21 species and Ctenomyces 1 species, but more detailed studies remain needed to establish species borderlines. Each species now has a single valid name. Two new genera are introduced: Guarromyces and Paraphyton. The number of genera has increased, but species that are relevant to routine diagnostics now belong to smaller groups, which enhances their identification.


Introduction
The dermatophytes belong to the oldest groups of microorganisms that have been recognized as agents of human disease. The taxonomy of these fungi was initiated in 1841 with the studies of Robert Remak and David Gruby [1]. Between 1840 and 1875, five of the main species known today, viz. Microsporum audouinii, Epidermophyton floccosum, Trichophyton schoenleinii, T. tonsurans and T. mentagrophytes had already been described; this was several decades before the discovery of Pasteur's invention of axenic culture [2]. The only ubiquitous modern dermatophyte missing from the list is Trichophyton rubrum [3], which has been hypothesized to have emerged in the twentieth century [4].
After Pasteur's time, culturing of dermatophytes and description of new species has taken off enormously. Species were defined on the basis of combined clinical pictures and morphological characters in vitro. Sixteen human-associated species were introduced between 1870 and 1920, with Sabouraud's [5] magistral overview of the dermatophytes setting a new standard. During the decades that followed, application of the new methodological standard led to an explosion of new species and recombined names (Fig. 1). Generic concepts remained confused, leading to repeated name changes with a total of 350 names around the year 1950. Subsequently anamorph nomenclature stabilized by the wide acceptance of Epidermophyton, Microsporum and Trichophyton as the genera covering all dermatophytes.
Culture and microscopic morphology worked well as diagnostic parameters when fresh isolates were used, but were difficult to maintain and reproduce because of rapid degeneration. Standardization with reference strains was therefore difficult, and this led to the introduction of numerous taxa that are now regarded as synonyms of earlier described species. In addition, diverse types of morphological mutants were described as separate taxa, such as Keratinomyces longifusus, which turned out to be Microsporum fulvum with strongly coherent conidia [6]. This misclassification is an unavoidable consequence of a diagnostic system based on the phenotype. Similar misjudgments of mutants of a single species also occurred elsewhere, sometimes unknowingly leading to the description of a separate genus for the mutant: compare, e.g., the genus pairs Bipolaris/Dissitimurus, Scedosporium/Polycytella, Exophiala/Sarcinomyces, or Trichosporon/Fissuricella [7]. In addition, several dermatophytes are known which do not or poorly sporulate in culture and thus show very limited phenotypic characteristics. Classically such species were partly based on clinical symptoms, e.g., T. concentricum or T. schoenleinii, but many more, undescribed species may exist [8].
In the last decades of the twentieth century, it became obvious that morphology had its limitations and could not be used as sole characteristic for classification or identification. Given these problems, Weitzman et al. [9] introduced an additional character set in the form of physiological parameters, so-called trichophyton-agars utilizing the ability of strains to assimilate a panel of essential vitamins, but also growth temperature, gelatin liquefaction, etc. The method now indicated as the 'conventional approach' to dermatophyte taxonomy combines clinical appearance, cultural characteristics, microscopy and physiology. Serology has never really taken off.
Biological species concepts entered the picture with the modern rediscovery of dermatophyte teleomorphs by Dawson and Gentles [10] and Stockdale [11]. Several geophilic and zoophilic dermatophytes, as well as related non-pathogenic species like Trichophyton terrestre and T. ajelloi, were found to produce sexual states, for which the genera Arthroderma and Nannizzia were introduced. This led to a new boom in the number of names ( Fig. 1) and marked the introduction of dual nomenclature for dermatophytes. The delineation of sexual interaction began to take an unusual course when Stockdale [12] discovered that members of many apparently non-mating species could be induced to reveal their mating type in an incomplete mating reaction with testers of Arthroderma simii. Most of the recognized asexual species could be typed in this manner and demonstrated to be descended from a single ancestral mating type. For example, Trichophyton rubrum was shown to be (-) in mating type, while its close relative T. megninii, currently considered to be synonymous, was (?). Just a few important species, such as Epidermophyton floccosum and T. soudanense, a further member of the rubrum series, resisted typing with this system and remained of unknown status. Summerbell [13] 1900 1901-1905 1906-1910 1911-1915 1916-1920 1921-1925 1926-1930 1931-1935 1936-1940 1941-1945 1946-1950 1951-1955 1956-1960 1961-1965 1966-1970 1971-1975 1976-1980 1981-1985 1986-1990 1991-1995 1996-2000 2001-2005 2006 The largest number of new names was created when morphology was added to clinical data as criteria for species distinction. The period 1960-1995 is marked by the addition of teleomorph names, leading to dual nomenclature of the dermatophytes. The bar at the right shows the approximate number of existing anthropophilic species (n = 10), the number of times these have been described (basionyms: n = 103) and the total number of name changes for these 10 species (n = 242). Possible [7] and proven synonyms of Trichophyton rubrum are listed in ocher (n = 48), of which (n = 24) were basionyms, in red Mycopathologia (2017) 182: 5-31 7 out the obvious ecological factor linking the nonsexual species: they all infected animals (including Homo sapiens) without having a terrestrial reservoir allowing the elaborate sexual processes with ascigerous fruit bodies to take place on keratinous debris.

Clinical Significance
Large differences are known to exist between species with respect to their natural habitat. Three broad ecological groups of dermatophyte species are recognized: anthropophilic, zoophilic, or geophilic (Table 1). Sometimes species cannot be clearly attributed to one of these groups due to insufficient data. Anthropophilic species naturally colonize humans, being transmitted between humans and usually cause chronic, mild, noninflammatory infections and often reaching epidemic proportions. Animal-carriage of these species does occur [14] but is exceptional. Zoophilic species live in close association with animals other than humans and transmission to humans usually occurs through their reservoirs. The fungi occur in the fur of particular animal hosts, either symptomatically or asymptomatically, and can become epidemic. Geophilic dermatophytes have their reservoir in the soil around burrows of specific terrestrial mammals, feeding on keratinous debris. They may be carried by these animals in their fur [15]; hence, the difference between geophilic and zoophilic dermatophytes is not always sharp. When transmitted to humans, zoo-and geophilic species cause acute, inflammatory mycoses. Occasionally, humans infected by zoophiles remain contagious, leading to small, self-limiting outbreaks [16], while most infections by geophiles are quickly resolved. Thus, also in the effectivity of human-to-human transmission an increasing trend is observed from geophiles via zoophiles to anthropophiles. No sexual phases are known in truly anthropophilic species, while geophilic species show vigorous mating. By these combined parameters, the three ecological groups, although not sharply separated, are fundamentally different and also have clinical significance (Table 1).

Experimental Methods
Enabled by the recent publication of whole genome sequences of several dermatophyte species [17], idiomorphs of the mating type loci (alpha domain and HMG domain genes) were detected directly at DNA level. Using partial amplification of each locus, Kano et al. [18] [19,20]. This implies that all anthropophilic and most zoophilic dermatophytes reproduce clonally by asexual propagation in apparently stable environmental niches. In contrast, Anzawa et al. [21] showed mating of a highly competent A. simii tester strain producing a fertile F1 generation with a strain of T. rubrum, challenging the biological species concept, although only a single out of 35 ascospores proved to be a real hybrid of the two species. Apparently, the dermatophytes have held an atavistic ability to undergo genetic exchange via sexual reproduction/hybridization in response, e.g., the stressful conditions of a newly inhabited environment. In practice, due to the different ecological niches of species like the anthropophilic species T. rubrum and the zoophilic species A. simii, they do not have the possibility to meet each other in nature. Like in Pasteur's days, when axenic culture revolutionized microbiology, the application of molecular  [22,23]. In a series of papers, Gräser et al. [6,24] applied the more variable rDNA ITS region and were able to resolve a large number of species. This molecular system has been confirmed several times in later studies [25] and with different molecular markers such as BT2 [26,27] and TEF1 [28]. The main topology of the Arthrodermataceae seems to be molecularly stable, but does not entirely correspond with morphology, as Trichophyton appears to be polyphyletic. As noted in earlier papers by Gräser et al. [6,24], anthropophilic species are confined to some derived clusters, zoophilic species of domesticated mammal hosts are located in the middle of the tree, while geophilic species are located in an ancestral position, and the lower clusters are still unstable due to taxon sampling effects. For reasons of clinical understanding, it is recommendable to formalize these differences in a new taxonomic system, which is one of the aims of the present paper. While the molecular approach was able to resolve the main traits of dermatophyte evolution, it may fail in the details. Several well-established, clinically different species such as Trichophyton rubrum/T. violaceum, T. equinum/T. tonsurans and to a certain extent also M. audouinii/M. canis/M. ferrugineum appeared largely indistinguishable in our multilocus analysis. Small sequencing ambiguities or missing data in this large dataset may blur the small differences very recently emerged species. Therefore, despite the available large body of research on these species, polyphasic studies combining molecular, ecological, phenotypic and life cycle data are needed to establish the validity of these species with certainty.
With the various taxonomic approaches, also nomenclatural rules have evolved over time (Fig. 2). In the nineteenth century, a clinical description was judged sufficient to characterize a fungus. Deposition  of a type specimen became compulsory only in 1957. Today, the reference of a type is essential to stabilize the species' delimitation and nomenclature. Older, long-forgotten names without types are discarded as doubtful, but well-known species names should be maintained by neotypification [6]. During the decades of dual nomenclature, species can have two types, but since 2013 the name, anamorph or teleomorph, always refers to the same, original type specimen. Present-day naming of fungi is according to their gross phylogenetic position. It should be realized, however, that positions in trees are relative, being dependent on the coincidentally selected constituents of the tree. Therefore, polyphasic species remain concepts essential for reliable nomenclature. For a checklist of obsolete names in dermatophytes for which no type material is known to exist, is referred to de Hoog et al. [7]. Numerous laterdescribed species were placed in synonymy, because they proved to not to deviate on the basis of modern characters. de Hoog et al. [7] listed 24 basionyms (with 48 combinations in total) as probably synonymous with Trichophyton rubrum (Fig. 1) (although only 5 basionyms could be proven with extant type materials). Several of the apparent synonyms were only recently segregated from T. rubrum on the basis of physiological parameters, which has shed doubt over usefulness of physiology as a taxonomic parameter.

Nomenclature
A search for possible generic names in Arthrodermataceae was limited to members of the order Onygenales. Candidate generic names were those type species in the family according to the Index Fungorum (www.indexfungorum.org). Obsolete generic names were taken from species synonyms and list of doubtful species in the Atlas of Clinical Fungi [7]. For every taxon to be accepted as a potential name or synonym, permanently inactivated (dried or under liquid nitrogen) holotype material had to be necessary. Holotypes as well as living strains connected with the holotypes were indicated as type (T). In heterothallic species, mating partners needed to obtain the teleomorph were listed as syntypes (ST). Taxa without types were discarded as doubtful, or, when these concerned wellknown clinical taxa described without deposition of type material, were neotypified. Neotypes (NT) in the present article have a single CBS (Centraalbureau voor Schimmelcultures) number, which refers to dried holotype material, or to metabolically inactivated samples under liquid nitrogen of which the original batch will remain unopened. In case the original holotype may not be interpretable, epitypes (ET) were indicated. If no type was indicated in the original protologue, but strains from the describing author(s) were available, these were listed as authentic (AUT). If none of these applies, but strains were used by authoritative authors, they were listed as reference strains. The latter two categories do not have official nomenclatural status.

Strains Analyzed
Strains preserved in the reference collection of Centraalbureau voor Schimmelcultures (CBS-KNAW Fungal Biodiversity Centre) were used for the multilocus phylogenetic analysis of members of the family Arthrodermataceae. In total, 261 strains were analyzed. Strains were cultured on Sabouraud's glucose agar (SGA) plates using lyophilized, cryo-preserved or fresh mycelial material for inoculation. Most of the cultures were incubated for 7 to 14 days at the temperature of 24°C, with some exceptions for very slow-growing species, while some others grew within a few days.

DNA Extraction, PCR and Sequencing
Genomic DNA was isolated from either preserved material or material harvested from living cultures. The DNA extraction was performed using MasterPure TM Yeast DNA Purification Kit from Epicentre. Five gene regions were amplified: ITS and LSU loci of the rDNA operon [29] and two protein coding genes. The universal fungal locus ITS1-5.8-ITS2 of the rDNA was amplified with ITS5 [30] and ITS4 [31] according to standard protocols [32]. The D1-D2 region of LSU was amplified using primers LR0R and LR5 [33] according to conditions as for ITS except for a longer extension time (90 s). Partial b-tubulin (TUB) was amplified with primers TUB2Fd and TUB4Fd [34]. PCR had an annealing temperature of 58°C for one min and elongation time of 70 s. 60S ribosomal protein L10 was amplified with 60S-908R and 60-S506F [35].
All PCRs were done in 12.5 lL final PCR volume (CBS-KNAW barcoding lab protocol), using 2.5 lL of the DNA extract, 1.25 lL PCR buffer (Takara, Japan, incl. 2.5 mM MgCl 2 ), 1 lL dNTPs (1 mM stock; Takara, Japan), 0.6 lL dimethylsulfoxide (DMSO; Sigma, The Netherlands), forward-reverse primer 0.25 lL each (10 mM stock), 0.06 lL (5 U) Takara HS Taq polymerase, 7.19 lL MilliQ water [32,36]. PCR products were visualized on 1 % agarose gel. Positive PCR products were sequenced in cycle-sequencing reaction using ABI big dye terminator v.3.1 using only one quarter of the suggested volume (modified manufacturer's protocol). Bidirectional sequencing was performed in a capillary electrophoresis system (Life Technologies 3730XL DNA analyser). The obtained sequences were manually edited, and consensus sequences were stored in a Biolomics database [37].

Sequence Alignment and Phylogenetic Analysis
Sequences were aligned with MAFFT v. 6.850b using default settings except for the 'genafpair' option [38]. The datasets for the five loci were assembled in one multilocus dataset using sequence matrix software and deposited in Genbank. Alignments were compared manually and via the Gblocks server (http://molevol. cmima.csic.es/castresana) with stringency settings 'allow gaps positions within the final blocks' and 'do not allow many contiguous nonconserved positions'. For both ITS and multilocus dataset Maximum likelihood phylogeny was inferred using RAxML v. 8.0.0 employing GTRCAT model and 1000 bootstrap replicates. Bootstrap branch supports above 80 % are shown. A general rDNA ITS and several more detailed multilocus single-genus trees are provided in summary (Figs. 3, 4).

Results and Discussion
A phylogenetic tree was constructed for all species discussed in this paper using the ITS rDNA region only, since this gene was comparable and alignable over the entire set of strains (Fig. 3). Seven clades were distinguishable. The upper clade (A) in this figure comprised anthropophilic and zoophilic Trichophyton species. This clade is shown in more detail with multilocus data in Fig. 4. Four 100 % bootstrap-supported species or species series were recognizable: (A-1) Trichophyton mentagrophytes and related anthropophilic and zoophilic species including some strictly anthropophilic clonal offshoots, with Trichophyton interdigitale and T. tonsurans as most common species; (A-2) Trichophyton benhamiae series with T. schoenleinii and T. verrucosum; (A-3) The zoophilic species Trichophyton bullosum; (A-4) Trichophyton rubrum series in which no individual species could be distinguished. The next, well-supported clade (B) in Fig. 1 contained a single species, Epidermophyton floccosum, which is paraphyletic to clade (C). Clade (C) contained zoophilic and geophilic species of which Microsporum gypseum was the most common one. Clades (D) and (E) were two groups of largeconidial, heterothallic species. Clade (F) comprised the Microsporum canis series, which is shown in more detail with multilocus data in Fig. 5. Clade (G) was highly diverse, containing well-resolved geophilic species only, many of which are currently known under their Arthroderma teleomorph name because of heterothallic mating. The anamorphs were characterized by large, multi-celled, thickand rough-walled macroconidia and abundant microconidia.
Data were also generated for additional partial genes LSU, 60S L10, and TUB (Figs. 4,5). Clades (A) and (F), containing the great majority of species that are relevant in clinical and veterinary settings, were partially resolved. A number of classical species in medical and veterinary mycology proved to be indistinguishable, possibly due to the fact that the large number of SNPs overshadowed consistent differences. The application of the Gblocks tool, reducing ambiguously aligned positions, led to inclusion of only 39 % of the original 830 positions in ITS and reduced the resolution between species. For this reason, we maintained manually aligned datasets and used additional phenotypic and ecological data for species delimitation. This did not always yield expected results; further detailed studies with mating tests remain necessary. In this study we differentiate 'species series', which are larger clusters of taxa which unite at the ITS level, and 'species complexes'. Chen et al. [39] defined a complex as a number of populations that are doubtfully distinct. In our species series, some of the taxa were unambiguously different when multilocus data were applied, while neighboring taxa could not properly be distinguished and thus might be regarded as species complexes. For precise species delimitation, data on natural hosts, virulence on non-optimal hosts, growth and sporulation, metabolite production and mating behavior are needed in addition to more detailed molecular studies. In the present overview, we prefer to be conservative in the maintenance of the number

The Species Problem
In the T. mentagrophytes series (Clade A-1) in Fig. 4 showing a multilocus tree, T. mentagrophytes was close to T. interdigitale. The latter species was exclusively isolated from humans, while T. mentagrophytes preponderantly originated from animals but also contained clinical strains. Trichophyton equinum could as yet not be distinguished from T. tonsurans. This touches on an essential question in medical mycology, as the species couples are known as zoophilic and anthropophilic, respectively, and a human infection by a zoophile is believed to be more inflammatory than when there is no host change. These questions cannot be solved in the present overview due to lack of clinical data of the strains examined. In the T. benhamiae series (Clade A-2), Trichophyton benhamiae, T. concentricum, T. erinacei and T. verrucosum could all be separated with multilocus data. Trichophyton quinckeanum is very close to T.  schoenleinii. The Trichophyton rubrum complex (Clade A-4) showed some diversity, but this did not entirely match with observed differences in phenotype and clinical predilection. In clade (F), when analyzed with multilocus data (Fig. 5), Microsporum canis, M. audouinii and M. ferrugineum were difficult to distinguish, particularly because the (?) and (-) mating partners showed a mutual distance that spanned the diversity of nearly the entire genus. With distance, a gradational loss of sporulation is observed via an 'M. distortum phenotype', concomitant with adaptation to the human host, which is in accordance with current species concepts.
A major taxonomic problem, frequently encountered in environmental fungi in general, is unexpected phylogenetic diversity of groups that previously seemed to be phenotypically monomorphic. Species with similar microscopic appearance sometimes even prove to belong to entirely different orders. Dermatophytes, in contrast, have consistently been found to belong to a single lineage, i.e., the family Arthrodermataceae. This shared phylogeny has been explained by their keratinophilic character, which is a rare property in the fungal kingdom. Evolution within the family shows a strong coherence with the animal hosts providing the keratin, as already noted in classical literature [40].
A second, current taxonomic problem is the molecular species concept. Almost everywhere in the fungal kingdom the number of molecular species appears to be much larger than what was earlier be recognized by conventional methods, see, for example, the fragmentation of Aspergillus fumigatus [41], Candida parapsilosis [42], or Aureobasidium pullulans [43]. Again, the dermatophytes seem to be exceptional. In the course of 150 years medical mycology mainly focusing on Caucasians in Europe, and with a wide diversity of diseases from different body parts, an exhaustive amount of pheno-and genotypes has been investigated in numerous publications. About 10 species can be categorized as common anthropophilic dermatophytes on the Eurasian and North-American continents. However, in the Atlas of Clinical Fungi [7], 103 basionyms, with 242 synonymous names in total, have been extracted from the literature to describe these same &10 species. It appears that the diversity seen with conventional approaches is much higher than the existing genetic diversity. We may conclude that the anthropophilic and perhaps also the zoophilic dermatophytes have been over-classified. Similar phenomena of over-classification are apparent in other fungal groups of practical importance and which have therefore been studied in extenso. For example, Rhizopus species are easy to grow in culture, and their culturing has started immediately after Pasteur's time because of their role in fermentation processes of soy-based Asian foodstuffs. By 1920, 43 species were described in Rhizopus microsporus and R. arrhizus, which today are reduced to just two on molecular grounds [44,45]. Another example is the ubiquitous saprobe Alternaria alternata, where the large number of morphological taxa mainly distinguished previously on the basis of conidial shape and three-dimensional conidiophore branching patterns were reduced to synonymy on the basis of genomic data [46].

Phylogenetic Overview
It may be concluded that the taxonomy of common anthropophilic dermatophytes is now mature enough to be stabilized at the benefit of clinical routine. Taxa that are recognized today are not likely to be subject to drastic change in the near future. Trees do not suffer from taxon sampling effects, and nomenclatural stability is within reach. Additional species on the human host are to be expected only among rare taxa, such as Trichophyton eriotrephon, degenerate and difficult to identify species, such as Microsporum aenygmaticum, species from geographically remote areas, such as Trichophyton concentricum, or from coincidental infections of otherwise zoo-or geophilic species. Particularly, the geophilic dermatophytes have insufficiently been studied compared to their large number of potential host animals and environmental habitats, and in these groups a larger number of taxonomic novelties can be expected, which however have limited clinical relevance.
The current main genera Epidermophyton, Microsporum, and Trichophyton in their classical circumscription are based on morphology of macroconidia. This corresponds only partly with phylogeny in that species fulfilling the morphological criteria of Trichophyton partly cluster in derived anthropophilic clades, and partly in ancestral clades of prevalently geophilic species [24]. Consequently, a number of geophilic species which are phylogenetically remote from anthropophilic Trichophyton and hardly ever cause human infection are now included in routine identification panels [7]. From ecological and clinical viewpoints, the difference between the two groups is immense, because anthropophilic species are considered to be real pathogens in that they have evolutionary advantage of being transmitted between human hosts, whereas an overwhelming number of geophilic species are opportunistic and are acquired from a natural habitat in the environment. The combination of such highly diverse fungi in a single genus is not optimal and might lead to inefficient use of hospital resources when pathogenic species have to be distinguished from numerous non-human taxa. Molecular phylogeny using 5 genes clearly separated the preponderantly geophilic species from the remainder, comprising several zoophilic and a preponderantly anthropophilic clade, which confirms previously published topologies based on ITS [6], TEF1 [28] and CAL [47]. Most zoophilic species compose clusters that are clearly separate from the preponderantly anthropophilic clades of Trichophyton and Epidermophyton. Now is the time to draw final conclusions and formulate the dermatophyte system in a modern sense, based on molecular phylogeny, supported by polyphasic data, and providing better tools for identification. This leads to a novel phylogenetic taxonomy and genus delimitation as outlined below. Main sets of criteria for species delimitation optimally should be based on the biological species concept, i.e., random mating with fertile progeny among members of the same species, and absence of mating between species. However, in microbiological practice, this criterion is often not easily applicable. Mating experiments and observation of fertile cleistothecia were particularly helpful to delineate species of the M. gypseum and of the T. mentagrophytes series [19,20,[48][49][50]. However, sexual reproduction is often not known because the conditions under which teleomorphs are produced are unknown, or perhaps they may not exist at all. Inter-sterile populations may exist within what we regard as a single species. In dermatophytes, preponderance of a single mating type-which may have mating typeassociated properties-may lead to asexual offshoots, explaining the clonal genetic composition of many species or other entities [51]. An alternative approach is genealogical concordance, i.e., the biological species concept expressed in silico. In the present study, this approach was adopted using four genes: LSU, ITS, 60S, and TUB. Different levels of resolution of clades were obtained with these genes. Listing the number of clades supported by bootstrap values [80 %, we observe ITS [ TUB [ 60S [ LSU, yielding 44, 37, 32, and 17 clades, respectively (data not shown). For routine diagnostics, ITS is optimal, although for distinction of individual members of species complexes additional genes like TUB are necessary.
Once species have been delimited, the entities should be named according to the new rules of fungal nomenclature where Art. 59 of the ICBN regulating the pleomorphic naming system was abandoned. In principle the oldest name stands. From January 1, 2013, onwards, teleomorph names that are added later are considered as new combinations of the original basionym rather than as separate names. For older publications, the pleomorphic nomenclature still stand, in the sense that the different phases of the fungus are treated as facultative synonyms, even if they are introduced in the same paper and when based on the same type specimen. Often these types date back before 1958 since when explicit deposition was required (Art. 40 ICBN); in such cases the type of the teleomorph was selected as neotype of the species. In this way the currently accepted species is closely approached. The oldest, best known and widely used species names were mostly introduced even culture methods were available, and most of the nineteenth century names were based on clinical appearance only. Original materials are available of only a small selection of much younger taxa and synonyms. In order to maintain species names in current circumscriptions, widely used names are fixed by neotypes. In contrast, obsolete names for which no type materials are available are regarded as of doubtful identity and are thus permanently discarded.

Nomenclature
Clades (A-G) in Fig. 3 are judged to represent genera. Table 2 summarizes and evaluates all genera described in dermatophyte taxonomy since 1841, and Table 3 provides the distribution of extant type species of each of these genera over the phylogenetic tree of Fig. 3. The oldest legitimate generic names      [52], occasionally other animals [53]. A white and a yellow phenotype are known, the yellow genotype containing MT-strains only [20]; a hybridization depression is noted with the remaining lineages. Contet-Andonneau & Leyer [54] invalidly introduced Trichophyton erinacei var. porcellae (without indication of type specimen, Art. 52 ICBN) matching the yellow phenotype. Note that with multilocus sequencing the mating types deviate slightly from remaining strains (Fig. 4).
Kamyszek, Med. Weteryn. 24: 146, 1945 Ctenomyces is a gymnothecial genus of terrestrial fungi with chrysosporium-like conidia and is classified in the Gymnoascaceae. Several species have been classified in the genus. For a description, see Böhme [71].
terrestre-Trichophyton terrestre Durie & Frey, Mycologia 49: 401, 1957. Type UAMH was not available for study. In literature the species has been listed as the anamorph of different Arthroderma species which on molecular grounds appear to be remote from each other. Trichophyton terrestre needs to be reevaluated.

Epilogue
The present paper provides an evaluated list of currently accepted species in Arthrodermataceae, but is by no means exhaustive. Many groups require more detailed polyphasic studies with mating experiments to determine exact borderlines between species. Some extant types could not be acquired during the course of this study. New, genomic and proteomic studies will provide understanding of the observed clinical differences in predilection between closely related species. It is expected that among the geo-and zoophilic groups numerous species are yet to be discovered in undersampled habitats; our review means to provide a new starting point for these subsequent studies.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.