Four new Keratinophyton species (Onygenaceae) from Europe

: Four new Keratinophyton species (Ascomycota: Pezizomycotina, Onygenales), K. gollerae , K. lemmensii , K. straussii and K. wagneri , isolated from soil samples originating from Europe (Austria, Italy and Slovakia) are described and illustrated. The new taxa are well supported by phylogenetic analysis of the internal transcribed spacer region (ITS) region, the nuclear large subunit (LSU) rDNA, and their phenotype. Within the Keratinophyton clade, K. lemmensii is clustered with K. durum, K. hubeiense , K. submersum and K. siglerae, while K. gollerae , K. straussii and K. wagneri are resolved in a separate terminal cluster along with K. minutisporosum . All four new species can be well distinguished from the other asexual taxa in the genus Keratinophyton based on phenotypical characteristics alone. Ten new combinations are proposed for all other Chrysosporium asexual morphs which are resolved in the monophyletic Keratinophyton clade.


Introduction
Keratinophyton is a genus of microscopic fungi (Ascomycota, Onygenales, Onygenaceae) comprising teleomorphic (sexual) and anamorphic (asexual) species that live mostly on the remains of hair and feather in soil as saprotrophs (Cano and Guarro 1990;Crous et al. 2016;Sutton et al. 2013;Vidal et al. 2000). Formerly, they were classified in Aphanoascus Zukal mainly based on the presence of ascomata (cleistoperidia) composed of a membranous peridium (Cano and Guarro 1990;Cano et al. 2002). In a review employing a phenotypic and phylogenetic approach, Cano et al. (2002) accepted 18 Aphanoascus species which are all sexual (i.e. teleomorphs -producing ascomata with asci and ascospores). Only recently, the polyphyletic status of Aphanoascus sensu lato has been resolved by Sutton et al. (2013) (Sutton et al. 2013). Within Keratinophyton species, only K. multiporum is connected with a Malbranchea anamorph, while the remaining sexual species hitherto known have a Chrysosporium asexual morph. In addition to the above mentioned sexual (teleomorphic) species, the monophyletic Keratinophyton clade currently encompasses also at least eleven species known only in their anamorphic (asexual) state (Cano and Guarro 1994;Crous et al. 2016;Crous et al. 2017;Liang et al. 2009;Oorschot 1980;Vidal et al. 2000;Vidal et al. 2002;Zhang et al. 2016;Zhang et al. 2017). Recently, Crous et al. (2017) introduced a new asexual species, namely Keratinophyton turgidum Rahul Sharma & Shouche, solely based on the morphology of its chrysosporium-like aleurioconidia. The same authors stated that all asexual species in this monophyletic clade which have a Chrysosporium asexual morph require redisposing in the genus Keratinophyton.
Regarding ecology and distribution, the presence of this large group of ubiquitous and often keratinolytic species is rather common especially in areas with high animal activity that results in transfer of the keratinous material (fur, hairs, etc.) to the soil (Papini et al. 1998;Vidal et al. 2000). The following reports confirm their world-wide distribution and occurrence in different habitats usually associated with soil environments, e.g. soil in city parks (Papini et al. 1998;Vidyasagar et al. 2005), flower pots (Singh et al. 2009), sand in children´s sandpits (Labuda et al. 2008), mud (Zaki et al. 2005), poultry farms (Anbu et al. 2004;Cano and Guarro 1990), marshy meadows, salt pans, desert, cultivated or uncultivated soils (Cano and Guarro 1990;Chmel and Vláčilíková 1977;Deshmukh 2004;Deshmukh et al. 2008;Han et al. 2013;Javorekova et al. 2012;Zhang et al. 2016;Zhang et al. 2017) and river sediments (Ulfig et al. 1997;Vidal et al. 2000;Vidal et al. 2002). In general, these fungi are rarely reported as animal pathogens, and in fact, only two species (C. echinulatum and C. pannicola) have been involved in mycoses (Cabanes et al. 2014;Crous et al. 2016).
During a microbiological survey of environmental samples (soil and compost) in July 2019, interesting Chrysosporium asexual morphs were isolated being phenotypically similar to those ones previously isolated from the same samples in August 2015 by one of the authors (RL).
Phylogenetically informative sequences were obtained from internal transcribed spacer (ITS) region and the nuclear large subunit (LSU) rDNA. Overall, the resulting data revealed that these isolates represent novel asexual species of the genus Keratinophyton, and they are described and illustrated here as Keratinophyton gollerae, Keratinophyton lemmensii, Keratinophyton straussii and Keratinophyton wagneri sp. nov.

Sample collection and isolation of the fungi
A sample of a garden soil in Vieste (Italy) was collected in July, 2004. A sample of a forest soil in Tatranská Lomnica (The Slovak Republic) was collected in August 2011. A sample of compost from an agricultural base at the Institute of Agrobiotechnology (IFA Tulln, Austria) was collected in August 2015. All three samples were collected from the surface layer (3-5 cm). The samples were dried and stored in plastic bags in a fridge (5-8 °C) till the time of analysis (August 2015 and July 2019). Isolation of the keratinophilic fungi was done as described previously (Javorekova et al. 2012). Each sample was divided into 10 subsamples.
The subsamples (20 g each) were poured into Petri dishes and soaked with antibiotic solution containing 500 ppm cycloheximide and 100 ppm chloramphenicol. Sterile defatted horse hair fragments (10 pieces of ca 2.0 cm per plate) were used as baits. The Petri dishes were then incubated at laboratory temperature (23-25 ± 1 °C), under ambient daylight, for a period of 2-3 months and remoistened with sterile deionized water whenever necessary. The Petri dishes were checked weekly for the presence of fungi, and isolates were cultured on Sabouraud 4% dextrose agar (SDA, VWR) supplemented with 500 ppm cycloheximide and 50 ppm chloramphenicol. Pure cultures were then transferred onto potato dextrose agar (PDA, Fluka). The preliminary identification of the resulting keratinophilic fungi (Chrysosporium asexual morphs) was based on their phenotypic characteristics according to van Oorschot (1980) and Vidal et al. (2000 and.

Cultivation of a strain, media and morphological analysis
For phenotypic determination, the strains were transferred (three-point inoculation with a needle) on potato dextrose agar (PDA), malt extract agar (MEA, Merck), Sabouraud 4% dextrose agar (SDA) and incubated for 14 days in the dark at 25°C. Christensen´s urea agar (Sigma-Aldrich) was used for additional physiological and biochemical characteristics (25°C, 14 days, in the dark). Corn meal agar (CMA, Oxoid), potato carrot agar (PCA, (Samson et al. 2010)) and Emerson YpSs agar (Atlas, 1946) were used for stimulation of sexual reproduction (20°C, 25°C, and 28°C, up to 3 months, in the dark).
Colony size (in mm), colony structure and characteristics were noted after 14 days (PDA, MEA, SDA, PYE, YpSs, CMA and PCA), however, the cultivation was prolonged up to 3 months in order to observe and record changes in pigmentation of the colonies as well as to determine the onset of sexual reproduction. In order to determine the optimal and minimal/maximal temperatures for growth, PDA,MEA and SDA at 5,8,10,12,15,18,20,25,28,29,30,31,32,35 and 37 °C were used and measured at 14 th day of cultivation. For comparative description of the macroscopic and microscopic characteristics, PDA was used according to Vidal et al. (Vidal et al. 2002) and Crous et al. (Crous et al. 2016;Crous et al. 2017).
For determination of microscopic traits, PDA was used after 14-18 days. Conidiophore structures and conidia formation were observed in situ under low magnification (50 -100x).
Details of conidiophores, conidia (aleurioconidia) and other microscopic structures, such as width of hyphae, were observed in mounts with Melzer´s reagent and lactic acid with cotton blue and were used also as mounting media for microphotography. The photomicrographs were taken using phase and Nomarski contrast on the Olympus BX51 microscope with Olympus DP72 camera and QuickPHOTO Micro 3.0 software. Photographs of the colonies were taken with a Sony DSC-RX100.
Scanning electron microscopy (SEM) was performed on a JEOL JSM-6380 LV microscope (JEOL Ltd. Tokyo, Japan). Fungal samples were prepared according to a simplified method (Samson et al. 1979). Pieces of colonies (ca. 3x5 mm) growing on PDA were fixed in 6% glutaraldehyde overnight in the refrigerator (ca. 20 h), then dehydrated in 2methoxyethanol for 10 min. This was followed by drying at a critical point and gold coating in

Keratinolytic activity
The keratinolytic activity (Oorschot 1980) was tested by placing a few sterilized blond hairs of a five years old child on the PDA plate 1 cm away from the point of inoculation. Ability to digest keratin was observed after 21 days after incubation at 25 °C, in darkness. In addition, a hair perforation test according to de Hoog et al., (2000) using 25 mL water containing 2-3 drops 10% yeast extract (YEW) was used as well (De Hoog et al. 2000). The children hairs were examined microscopically after 14 and 21 days after inoculation at 25 °C, in darkness. By the end of the cultivation a few pieces of hairs were taken out from the testing media (PDA and YEW). Overgrown fungus was deactivated with 70% ethanol and consequently discarded from the hair surface mechanically in the stream of a tap water. A degree of hair digestiondegradation (keratinolytic activity) was observed in a light microscope under 100x and 400x magnification. For the observation and microphotography of the hairs water was used as mounting fluid. Intensity of attack on the hair was estimated on a scale of 0 to 4 (Marchisio et al. 1994): 0, no attack; 0-1, light attack on the cuticle; 1, moderate attack on the cuticle and/or rare formation of boring hyphae; 2, attack on cuticle and cortex, with about 20% destruction of the hair; 3, attack on cuticle and cortex, with about 50% destruction of the hair; 4, attack on cuticle and cortex, with about 80 % destruction of the hair. The photomicrographs of the hairs were taken using a Motic BA 310 microscope with Motic Image Plus 3.0 software. Final microscopic pictures were black and white inverted.

DNA extraction, PCR amplification and sequencing
DNA was extracted using a standard cetyltrimethyl ammonium bromide (CTAB) procedure, as described previously (Doyle and Doyle 1987). The internal transcribed spacer (ITS) region with primers ITS1-F (Gardes and Bruns 1993) and ITS4 (White et al. 1990) was amplified with Taqpolymerase. The D1/D2 domains of the large-subunit (28S) rRNA gene (LSU) were amplified and sequenced using the primer pair ITS1/TW14 (White et al., 1990;Mori et al., 2000). All reactions were performed in an Eppendorf Gradient MasterCycler (Eppendorf, Hamburg The PCR products were sequenced with the same primers used for the PCR amplifications (Microsynth AG, Balgach, Switzerland). All sequences obtained in this study were deposited in GenBank. For information on fungal strains used in this study see Table 1. This table provides GenBank accession numbers to ITS sequences for all accepted species in the genus Keratinophyton (except K. multiporum) and for selected ones in the genus Aphanoascus. The LSU sequences were provided to all new species and to majority of Keratinophyton and Aphanoascus species used for phylogeny.

Phylogenetic analysis
For phylogenetic analysis, sequences were aligned with ClustalX (Larkin et al. 2007).
Phylogenetic analysis was done with SEAVIEW 4.6 (Gouy et al. 2010)
Negative urease activity (after 14 days of incubation).
Micromorphology (Fig. 8b- Distinguishing characteristics. Robust and coarsely roughed conidia produced from non-swollen conidiogenous cells, none to very limited growth at 30°C, and production of pink pigment on PCA after prolonged incubation (3 weeks).
The main distinguishing phenotypic characteristics of the four new species compared with the other asexual members of the genus Keratinophyton are listed in the Table 2.     Marchisio et al., (1994)) substantially differed from species to species, being very strong in K.

Notes
All four new species are directly distinguishable from the other taxa in the genus Keratinophyton, known only as Chrysosporium asexual morphs, based on phenotypical characteristic alone. From the sexual Keratinophyton species, they are principally differing by inability to form ascomata in a culture (Cano et al. 2002), but also by overall phenotypic characters, such as growth at high temperature and/or conidia morphology (Cano and Guarro 1990;Cano and Guarro 1994;Currah 1985). The most solid species-specific phenotypic distinguishing characteristics are morphology of conidia (shape, surface and dimensions) and growth response to 30 °C exposure after 14 days on PDA.
Phenotypically, K. lemmensii is characteristic and differs from the relatives in the cluster (K. lemmensii can be directly distinguished from the phylogenetically closest K. durum asexual morph also by the presence of numerous arthroconidia which are completely missing in the later species (Cano and Guarro 1990;Currah 1985).
Keratinophyton gollerae can be readily distinguished from the two other species in their joint cluster, i.e. K. straussii and K. wagneri, by its inability to grow at 30 °C, by narrower and mostly smooth to finally roughened conidia, and by its slower growth at 25 °C on PDA. Moreover, in comparison with K. straussii, K. gollerae grows substantially better at 15°C (on PDA and SDA) and its spores germinate at 8 °C (see Table S2a-c). The remaining two species in the cluster, K.
As pointed out by Crous et al., (2017) (Sutton et al. 2013) and 15 asexual morphs (species), including recently described K. turgidum (Crous et al. 2017) and four novel taxa described here. Each of these Chrysosporium asexual morphs is characteristic by its particular combination of the morphological traits (the colony color and growth rate, growth response on higher/lower temperature as well as morphology of conidia) and also easily distinguished from the four novel taxa on these phenotypic bases.
Hubalek (2000) provided a list of keratinolytic fungi associated with free-living mammals and birds in which Keratinophyton pannicola (as Chrysosporium evolceanui) has been isolated from a variety of animals, namely from different species of rodents in Australia, Czechia, England, Germany, former Yugoslavia, from a rabbit in Canada, and birds in Czechia, Queensland, former Yugoslavia and India (Hubalek 2000). Keratinophyton durum (as Aphanoascus durus) has been isolated from a hedgehog in Ivory Coast, and Keratinophyton terreum (as Aphanoascus terreus) has been associated with a variety of rodents in Romania, Germany, India, Czechia, Yugoslavia and Nigeria, and birds in Czechia, former Yugoslavia, India, Queensland and USA (Hubalek 2000). To the best of our knowledge, only a single report dealing with a human clinical isolation is connected to the ex-type strain K. echinulatum (CCF 4652=CBS 141178) from the sole of the foot of a 35-years-old woman in the Czech Republic (Crous et al. 2016). However, these same authors indicated that the etiological significance of the fungus is unclear, and they concluded that the infection was in fact caused by a dermatophyte, which was not isolated or overgrown by this K. echinulatum isolate. A few other cases have been published in a small range of animals including Keratinophyton pannicola (as Chrysosporium pannicola) from affected skin of a dog in former Yugoslavia (Cabanes et al. 2014;Oorschot 1980) and from a case of keratomycosis in a horse (Cabanes et al. 2014). In her review on Chrysosporium and related genera in Onygenaceae, Sigler (2003) stated that some reports concerning Chrysosporium species as etiological agents must be viewed with caution, however, since the isolated organism has neither been identified to species nor documented well enough to confirm the etiology. In the follow up list of species of medical relevance, there is no species mentioned being currently affiliated within the genus Keratinophyton (Sigler 2003). On the other side, according to Papini et al., (1998), every keratinophilic fungus can be considered a potential pathogen. Thus, soil can be regarded as an epidemiological link, probably evolutionary as well, that relates geophilic, zoophilic, and anthropophilic keratinophilic fungi. In fact, during a mycological investigation of the samples (data not shown), there was a huge prevalence of geophilic dermatophytes such as Nannizzia gypsea (as Microsporum gypseum) in a soil sample from Italy (2004), Arthroderma uncinatum (as Trichophyton ajelloi) with co-occurrence of Aphanoascus keratinophilus (as Chrysosporium keratinophilum) in a soil sample from Slovakia (2011), and finally, Trichophyton terrestre along with abundant A. uncinatum co-population in a compost sample from Austria (2015).

Significance
As members of the genus Keratinophyton are considered as typical soilborne fungi (Cano and Guarro 1990;Cano et al. 2002;Sutton et al. 2013) and there is not any solid evidence of pathogenicity reported, it is likely that those above mentioned animal-associated cases reflect just a simple environmental pollution of the animals with soil particles present around their habitats and dwellings and ability to persistent in dormant stages on fur or feather of the animals. The ability of these fungi to persist and survive in soils in dormancy was observed also during the present study, as in case of K. straussii, the type strain (BiMM-F78) was isolated 11 years after the sample collection in 2004, and two more strains representing this taxon (RL-05 and RL-06 obviously clonal) were isolated in a repeated study even after 15 years since the sample collection. Likewise, a second strain (RL-07) used for description of K. wagneri and the type strain of K. gollerae (BiMM-F250) were both isolated after 8 years since the time of collection. In this study, the degree of keratin digestion by the tested strains varied and was found to be very strong in both K. gollerae and K. straussii evoking attack on cuticle and cortex with about 50-80% destruction of the hair. According to Marchisio et al., (1994) and Mitola et al., (2002) keratinolytic fungi share common properties with dermatophytes. Potential pathogenicity to humans and homoiotherm vertebrates (mammals or birds) by theses fungi is highly unlikely as they do not grow at temperatures above 32°C. Instead, their strong keratinolytic ability might be providing a competitive advantage in nature to acquire nutrients and may potentially be used in industry for production of proteolytic enzymes applied in degradation processes of keratinous material (hairs, fur, feather etc.). Furthermore, these fungi represent yet undiscovered source of new bioactive secondary metabolites as there is not much known from literature about these properties of the genus (e.g. Kushwaha and Guarro 2000). Hence, all these newly described fungi are currently under investigation in our research facilities as for their ability to produce prospective bioactive secondary metabolites.