Introduction

Molecular systematics and comparative genomics in mycology provide tools to reevaluate large groups of fungi that were previously described on the basis of morphology alone, or with limited molecular data. Additionally, orders that contain medically important species, such as Mucorales (Chibucos et al. 2016; Walther et al. 2019), Chaetothyriales (Quan et al. 2020), Hypocreales (Kepler et al. 2017; Crous et al. 2021), and Onygenales (Kandemir et al. 2022), have been studied by applying a multiphasic approach combining multi-locus phylogenies, morphology, physiology, and ecology of the species. Results of experimental studies tend to lead to numerous taxonomic and nomenclatural changes. These rearrangements inevitably forced researchers to redefine generic and specific concepts, and to revise the criteria for new species (Aime et al. 2021). The use of molecular (e.g., barcoding genes) and morphological (e.g., culture characteristics and sexual and asexual morphs) data to delimit species, genera and families, as addressed in several studies, ultimately should clarify and stabilize nomenclature (Zamora et al. 2018; Kandemir et al. 2020; Lücking et al. 2020; Jiang et al. 2020; Crous et al. 2021).

In a revision of the order Onygenales, Kandemir et al. (2022) provided an overview of the ecological, morphological and phylogenetic characteristics of its families. Several pending nomenclatural and taxonomical issues have been solved in recent papers (Hainsworth et al. 2021; Labuda et al. 2021; Rodríguez-Andrade et al. 2021). However, some issues remain unsolved, among which is Sporendonema and relatives in the family Gymnoascaceae. Molecular data shows that, besides Sporendonema (S.) casei, the genus includes species previously classified in Arachniotus and Sphaerosporium (Kandemir et al. 2022). Sphaerosporium (Sph.) equinum and S. casei were isolated mainly from cheese and dried meat products and are known to be halophilic (Ropars et al. 2012; Scaramuzza et al. 2015), whereas members of Arachniotus (A.) are mainly isolated from dung and agricultural soil. However, A. desertorum was originally isolated from halomorphic soil in Kuwait (Moustafa 1973). In the present study, we aimed to investigate the relationship among the genera Arachniotus, Sphaerosporium, and Sporendonema using nuclear ribosomal internal transcribed spacer (ITS), D1-D2 region of the large subunit (LSU), and partial β-tubulin (TUB) phylogenies, in addition to morphological and physiological data.

Materials and methods

Strains

Strains were obtained from the CBS collection (housed at the Westerdijk Fungal Biodiversity Institute, Utrecht, The Netherlands), the Mycothèque de l’Université Catholique de Louvain (BCCM/MUCL; Louvain-la-Neuve, Belgium), and the University of Alberta Microfungus Collection and Herbarium (UAMH; currently in Toronto, Canada). Additionally, two specimens, ILLS 36355, isotype of Hormiscium (H.) aurantiacum, and ILLS 45141, isoneotype of Torula (T.) equina, were loaned from The Illinois Natural History Survey Herbarium (USA). Strain information is provided in Table 1.

Table 1 Sources and GenBank accession numbers of the strains used in the study

DNA extraction, PCR and sequencing

The strains were grown for 14–21 d on malt extract agar (MEA) were used for DNA extraction using the Wizard® Genomic DNA purification Kit (Promega Corp., Madison, WI, USA), according to the manufacturer’s instructions. Three gene regions, ITS, LSU and TUB, were amplified using the primers ITS4-ITS5 (White et al. 1990, Ward and Adamas 1998), LR0R-LR5 (Vilgalys and Hester 1990) and TUB2Fd-TUB4Fd (Woudenberg et al. 2009), respectively. The PCR conditions were as follows: 35 cycles of 45 s at 94 °C, 45 s at 52 °C and 90 s at 72 °C for the ITS and LSU markers; 35 cycles of 45 s at 94 °C, 45 s at 48 °C and 90 s at 72 °C for the TUB region. PCR amplicons were visualized on 1.5% agarose gels. Sequencing was performed with the same primer pairs as used for PCR amplification using Applied Biosystems BigDye Terminator v.3.1 (Thermo Fisher Scientific).

Molecular identification and phylogenetic analyses

Sequences for each marker were edited and assembled in Geneious R11 v.2022.0.1 (Kearse et al. 2012) and deposited in GenBank (Table 1). The sequences were aligned in MAFFT v.7 (Katoh et al. 2019) and combined with Sequence Matrix v.1.8 (Vaidya et al. 2011). In total, 54 isolates were included from which 54 ITS, 47 LSU, and 28 TUB sequences were analysed. The best-fitting model for each gene was found using ModelFinder (Kalyaanamoorthy et al. 2017) on the IQ-TREE web server (http://iqtree.cibiv.univie.ac.at/) (Nguyen et al. 2015) according to the Bayesian Information Criterion (BIC). Phylogenetic trees were constructed using MrBayes v.3.2.7 analyses on the CIPRES website (http://www.phylo.org) (Ronquist and Huelsenbeck 2003) and maximum likelihood (ML) methods implemented on the W-IQ-TREE web server (Trifinopoulos et al. 2016; Minh et al. 2020). All alignments and phylogenetic trees were deposited in TreeBASE (http://treebase.org; TB2:S29250) and figshare (https://doi.org/10.6084/m9.figshare.23284661) repositories.

Morphology

Morphology of the colonies was observed on potato dextrose agar (PDA) and yeast powder-soluble starch agar (YpSs) after 21 d of incubation at 24 °C in the dark. Since S. casei and H. aurantiacum strains were not able to grow at 24 °C, these strains were incubated at 15 °C in the dark for 21 days. Colony details and morphology of the fungarium specimens were observed using a Nikon SMZ1500 microscope, and microphotographs were taken by Nikon Eclipse80i equipped with a Nikon DSRi2 camera.

Physiology

Physiological tests were performed in five categories for the strains classified in Sporendonema. The growth rates were recorded at 4, 10, 15, 27, 30 and 36 °C. Growth was also compared using oat meal agar (OA) and OA supplemented with streptomycin and penicillin (OA/PS). Salt tolerance was evaluated using MEA containing 3, 10, 17 and 25% NaCl at 15 °C for Sporendonema and Hormiscium strains and at 24 °C for the remaining strains. Lipolytic activity was evaluated on Tween-80 agar (Ates et al. 2008), casein hydrolysis on home-made skim milk agar and cellulolytic ability on cellulose Congo-Red agar (CCA) (Gupta et al. 2012). Lipolytic, proteolytic and cellulolytic abilities were evaluated after incubation for 21 d by the formation of transparent halos and a lipolysis zone around the colonies (Fig. 1). All physiological tests were performed twice.

Fig. 1
figure 1

Interpretation of the lipolytic, proteolytic and cellulolytic activities. Positive results are shown on A Tween-80 agar (S. casei CBS 206.35). B Skim milk agar (Sph. equinum MUCL 40625). C Cellulose Congo-red agar (A. ruber CBS 592.71)

Results

Phylogeny

The best-fitting model was TNe + G4 for ITS and TUB, and TN + F + I + G4 for LSU. Compared to the analyses with ITS, LSU, and TUB data, ITS + LSU alone yielded better resolution among clusters (Fig. 2). In the two-marker analysis, all known supported relationships in Gymnoascaceae, i.e., those of Arachniotus, Gymnascella, Narasimhella and Gymnoascus designated previously, were fully confirmed by the Bayesian analysis (Fig. 2).

Fig. 2
figure 2

Maximum likelihood Phylogenetic tree from the combined ITS and LSU data. Type species of the genera are bold, epitype and the new species described in the present study are shown in red, newly proposed names are shown at the left side. Values of ≥ 0.95 pp for Bayesian probability and ≥ 80% for maximum likelihood are shown with thickened branches. Arthroderma ciferrii and Arthroderma crocatum were used as outgroup species

Species previously described in Arachniotus were found in different clusters based on combined data analyses of the ITS + LSU and ITS + LSU + TUB loci (Fig. 2; Fig. S1). Hormiscium aurantiacum and S. casei strains were clustered together, while another cluster was formed by Sph. equinum strains, the latter segregating into two groups compatible with their sources of isolation, i.e., cheese vs. other sources. Arachniotus ruber (CBS 352.90) showed 99% ITS homology with S. casei (CBS 543.75) and 98% with Sph. equinum (MUCL 46080). The ITS similarity between S. casei (CBS 543.75) and Sph. equinum (MUCL 46080) was 97%. Both H. auranticaum strains were identical to S. casei (100% of the ITS and LSU sequences; 525 bp and 803 bp, respectively).

Physiology

All strains showed growth at 10 °C and 15 °C. Sporendonema casei, H. aurantiacum and two of the A. ruber strains did not grow at 27 °C and 30 °C. Sporendonema casei CBS 207.27, H. aurantiacum and A. confluens did not grow at 4 °C. In contrast to the remaining soil strains, strains that were isolated from agricultural soils were able to grow at 4 °C, but unable to grow at 27 °C.

All strains were able to grow in the presence of 3% and 10% NaCl. A zone of lipolysis was detected on Tween-80 agar for all strains except A. ruber CBS 351.66 and Sph. equinum MUCL 46080. Even though all strains showed growth on skim milk agar, only A. aurantiacus CBS 603.67, A. confluens CBS 352.66 and CBS 634.72, A. ruber CBS 351.66, H. aurantiacum CBS 206.35, S. casei CBS 207.27 and CBS 355.29 and Sph. equinum MUCL 40625 and MUCL 46080 were able to hydrolyse caseins. Additionally, almost all Arachniotus strains (7/8) were found to be able to grow on OA/PS, while the cheese isolates and strain H. aurantiacum CBS 111.18 isolated from stockfish did not grow on this medium. The results on the CCA were found to be inconsistent in repeated analyses. Only in A. confluens CBS 352.66 and A. ruber CBS 351.66 halos around their colonies on CCA remained consistently absent. The results of the physiological tests are shown in Table 2.

Table 2 Source information of the strains used for the physiology tests and the results after 21 days of incubation

Taxonomy

Based on the above phylogenetic and phenotypic results, the treated species previously described as Arachniotus, and Sphaerosporium are considered congeneric with Sporendonema, since this genus has historical nomenclatural priority. Differences in micromorphology of Sporendonema species are provided in Table 3.

Table 3 Summary of the morphological differences among Sporendonema species

Sporendonema Desm. – Annls Sci. Nat., Sér. 1 11: 246 (1827).

 = Coprotrichum Bonord.—Handb. Allgem. Mykol. (Stuttgart): 76 (1851).

 = Arachniotus Schröt.—Krypt.-Fl. Schlesien (Breslau) 3.2(1–2): 210 (1893) [1908].

Type species: Sporendonema casei Desm.

Sporendonema aurantiacum (Kamyschko) Kandemir & de Hoog, comb. nov.

Figure 3A–D, M, N

Fig. 3
figure 3

A‒D, M, N Sporendonema aurantiacum CBS 603.67. E‒H, O, P Sporendonema confluens CBS 352.66. I‒L, Q Sporendonema rubrum CBS 352.90. A, B, E, F, I, J Colony surface and reverse on PDA after 21 d at 24 °C. C, D, G, H, K, L Colony surface and reverse on YpSs agar after 21 d at 24 °C. M‒Q Asci and ascospores. Scale bars M‒Q = 10 μm

MycoBank number: MB842801.

Pseudoarachniotus aurantiacus Kamyschko – Nov. Sist. Niz. Rast. 4: 224 (1967) ≡ Arachniotus aurantiacus (Kamyschko) v. Arx - Persoonia 6(3): 373 (1971).

Holotype Russia, Republic of Kalmykia, from semi-desert (slightly loam) soil, Culture 4–1/2, (Kamyschko 1967), was preserved Institute of Antibiotics, Saint-Petersburg (Leningrad). Ex-holotype culture CBS 603.67. Alternative collection numbers BKM F-1140, ATCC 22394, NRRL A-18287, BKM F-1140, and UAMH 3529.

Notes: Sporendonema aurantiacum has globose, and smooth ascospores without a prominent equatorial rim similar to that of S. confluens. Sporendonema aurantiacum can be differentiated from S. confluens by its darker ascospores that has discoid from the side view. A detailed description has been provided by von Arx (1970).

Sporendonema casei Desm. – Annls Sci. Nat., Sér. 1, 11: 246 (1827).

Figure 4A–M

Fig. 4
figure 4

A‒M Sporendonema casei A, B, I‒K CBS 543.75, E‒H, L, M CBS 206.35. N Hormiscium aurantiacum type specimen ILLS 36355. A, B, E, F Colony surface and reverse on PDA after 21 d at 15 °C. C, D, G, H Colony surface and reverse on YpSs agar after 21 d at 15 °C. I, J Mass of thallic-arthric conidiophores, slides from MEA. K Mass of hyphae, conidiophore, and conidia, slide from PDA. L Hyphae and conidiophores. M, N Conidiophore developing conidia. Scale bars I‒N = 10 μm

 =  Torula sporendonema Berk. & Broome – Ann. Mag. Nat. Hist., Ser. 2, 5: 460 (1850).

Holotype material is not known to be preserved. Lectotype (designated here, MBT 10017580), drawings in Desmazières (1827) plate 21A, Fig. 1. Epitype (designated here, MBT 10017581) CBS 543.75, isolated from cheese, by Sochal, 1975, preserved in metabolically inactive state.

Notes: Sporendonema casei is a well-known cheese-inhabiting fungus that produces orange-red spots on cheese. This slow-growing and xerotolerant fungus produces cubical conidia with rounded corners from club-shaped hyphae by enteroarthric conidiogenesis. A detailed description of S. casei has been provided by Sigler and Carmichael (1976).

Sporendonema confluens (Sartory & Bainier) Kandemir & de Hoog, comb. nov.

Figure 3E–H, O, P

MycoBank number: MB842802.

Gymnoascus confluens Sartory & Bainier – Bull. Soc. Mycol. Fr. 29: 261 (1913) ≡ Arachniotus confluens (Sartory & Bainier) Apinis – Mycol. Pap. 96: 37 (1964) ≡ Gymnascella confluens (Sartory & Bainier) Currah – Mycotaxon 24: 75 (1985).

Neotype UK, London, Birbeck College, from dung, 1959, dry culture BDUN 375, designated by Apinis (1964). Alternative collection numbers ATCC 22220, CBS 352.66, IMI 100873, NRRL 5979, Orr O-3559, and UAMH 3565.

Notes: See the notes under the S. auranticum section. Detailed description has been provided by Currah (1985) and Apinis (1964)

Sporendonema equinum (Desm.) Kandemir, Decock & de Hoog, comb. nov.

Figure 5

Fig. 5
figure 5

A‒L Sporendonema equinum A‒F, I‒L MUCL 46080, G, H type specimen ILLS 45141. A, B Colony surface and reverse on PDA after 21 d at 24 °C. C Colony details on YpSs agar 21 d at 24 °C. D, E Colony surface and reverse on PDA after 21 d at 24 °C. F Spore chains, conidiophores and hyphae. G Colonies on horse hooves. H Thick-walled spores obtained from the type specimen. I Catenate conidia. J Conidiophores and conidia. K, L Conidia. Scale bars F, H‒L = 10 μm

MycoBank number: MB842804.

Torula equina Desm. – Annls Sci. Nat., Bot., Sér. 4, 4: 126 (1855) ≡ Oospora equina (Desm.) Sacc. & Voglino – Syll. Fung. (Abellini) 4: 22 (1886) ≡ Sphaerosporium equinum (Desm.) Crane & Schokn. – Mycologia 78(1): 86 (1986).

Isoneotype France, from old and humid horse hoofs, collected and identified by Desmazieres, ILLS 45141, collector number H. G. 1510, designated by Crane and Schoknecht (1986).

Notes: Sphaerosporium was introduced by von Schweinitz (1834) based on morphology of the type species Sph. lignatile found growing on dead wood in the USA. The holotype for Sph. lignatile was designated as #3036 (PH, Paris Herbarium). Later, Sph. equinum, originally described as Torula equina, was added to the genus (Crane and Schoknecht 1986). Partridge and Morgan-Jones (2002) reviewed Sphaerosporium and provided descriptions for both Sph. lignatile and Sph. equinum. Authors noted that despite their substrate differences, these two taxa share morphological similarities suggesting a close relationship (Partridge and Morgan-Jones 2002).

However, molecular analyses do not support any relationship between Sph. lignatile and S. equinum (Song et al. 2019). In the current study, micromorphology of the type specimen of Torula equina ILLS 45141 was examined. The conidia were abundant, arranged in basipetal chains, globose, with thick, and smooth walls (Fig. 5).

Additionally, we examined MUCL 46080, which was isolated from the rind of a sheep cheese in France, as a reference strain to evaluate morphological characteristics and physiology of Sph. equinum. Since we were not able to obtain a pure culture from the type specimen, we could not compare the type and the cheese isolates phylogenetically. Nevertheless, we propose a new combination for the cheese isolates of the Sph. equinum since they are classified within Sporendonema, Onygenales, Eurotiomycetidae (Kandemir et al. 2022), while Sph. lignatile is classified in Pezizales, Pezizomycetidae (Song et al. 2019).

The strains UAMH 11516 (= MUCL 58097), obtained from the skin of bat wings, and MUCL 54024, from insect pupa were found to be phylogenetically related to the cheese isolates of Sph. equinum (Fig. 2). However, the growth rate on PDA, OA, OA/PS and MEA, the conidial shape and size, and caseinase activity were different between the two groups. Therefore, a new species was described to accommodate MUCL 54024 and MUCL 58097.

Sporendonema isthmoides Decock, Kandemir, Hern.-Rest. & de Hoog, sp. nov.

Figure 6

Fig. 6
figure 6

A‒L Sporendonema isthmoides MUCL 58097. A, B Colony surface and reverse on PDA after 21 d at 24 °C. C Colony details on OA/PS. D, E Colony surface and reverse on YpSs agar after 21 d at 24 °C. F Mass of hyphae and conidia. G Arthric conidiophore. H, I Enteroarthric conidiophore. J Hyphal remnants on catenate conidia. K Fertile hyphae with warts. L Thick-walled, single-cell conidia. Scale bars F‒L = 50 μm

MycoBank number: MB842809.

Etymology In Greek “isthmus” means “neck”, and “isthmoides” is used for “resembling isthmus”, referring to the narrow conjunction between conidia in chains.

Holotype Canada, New Brunswick, Berryton Cave, from swab sample of living female little brown bat (Myotis lucifugus) skin, 2010, isolated by K. J. Vanderwolf, dried culture UAMH 11516, preserved in a metabolically inactive state. Alternative collection number MUCL 58097; GenBank numbers ITS: OM468607, LSU: OM515118, TUB: OM616026. Additional specimen Belgium, insect pupa in the attic of a house, 2012, C. Decock, MUCL 54024; Genbank numbers ITS: OK255531, LSU: OK255535.

Vegetative hyphae hyaline, septate, smooth, 2.5–4.5 µm wide; fertile hyphae mostly smooth and some ornamented with warts (Fig. 6); conidiogenesis thallic-enteroarthric; conidia hyaline to pale yellow, yellow-orange in mass, 1-celled, lemon-shaped in chains and becoming globose when separated, truncated at one or both ends, smooth- and thick-walled, occasionally with warts; 6–8.5 × 3.5–5 µm. Sexual morph not observed.

Culture characteristics on PDA reaching 40 mm diam after 21 d at 24 °C; flat, elevated in the center; margin regular; obverse color orange, dirty white-beige at the periphery (Fig. 6A); reverse dark brown at the center and orange-yellow at the periphery (Fig. 6B). Colonies on YpSs agar reaching 38 mm diameter after 21 d at 24 °C, flat, slightly elevated at the margin, texture velvety, obverse color orange with cream-white edges (Fig. 6D); reverse orange with a cream-colored periphery (Fig. 6E).

Growth temperatures minimum 4 °C and maximum 27 °C.

Physiology Casein not hydrolysed. Growth present at NaCl concentrations of 3, 10 and 17 but not 25% (w/w).

Notes: Based on ITS and LSU data analyses, the phylogenetically closest species to S. isthmoides is S. equinum. Sporendonema isthmoides and S. equinum differ morphologically in conidiogenesis (thallic-enteroarthric vs holoblastic; Figs. 5 and 6), shape of conidia in chains (lemon-shaped vs. globose) and size (9.5 × 13 µm vs. 4.0 × 7.5 µm). The conidia and hyphae of S. equinum are smooth-walled, whereas some conidia and hyphae of S. isthmoides are warted. Sporendonema isthmoides grows faster than S. equinum on almost all tested media (PDA, OA, and MEA supplemented with 3% and 10% NaCl at 24 °C; Table 2), can grow on OA/PS medium and lacks caseinase activity. Differences in the micromorphology of S. casei, S. equinum and S. isthmoides are illustrated in Fig. 7.

Fig. 7
figure 7

Line drawing of conidiophores developing conidia. A Sporendonema isthmoides. B S. equinum. C S. casei. Bar = 10 μm

Sporendonema rubrum (Tiegh.) Kandemir & de Hoog, comb. nov.

Figure 3I–L, Q

MycoBank number: MB842803.

Gymnoascus ruber Tiegh. – Bull. Soc. Bot. Fr. 24: 159 (1877) ≡ Arachniotus ruber (Tiegh.) Schröt. – Krypt. -Fl. Schlesien (Breslau) 3.2(1–2): 210 (1893) [1908].

Neotype UK, from soil, IMI 92796, designated by Kuehn and Orr (1964). Alternative collection numbers CBS 352.90 and ATCC 15315.

Notes: Arachniotus ruber was described from coyote dung as type species of the genus Arachniotus and was outstanding with its low temperature (5 °C) requirement for isolation (Currah 1985). It has hyaline asci, orange-yellow and smooth ascospores with two equatorial lines (Fig. 3Q). A detailed description of the fungus is given by Kuehn and Orr (1964).

Discussion

Nomenclaturally, Sporendonema casei is the oldest described species in the family Gymnoascaceae. It was introduced by Desmazières (1827) for an orange-red fungus growing on cheese. After several disagreements on the nomenclature and the taxonomic position of this “red mould” (Corda 1838; Berkeley and Broome 1850; Saccardo 1882; Bainier 1907), the name S. casei became widely accepted (Hammer and Gilman 1944; von Arx 1970). Ropars et al. (2012) and Kandemir et al. (2022) confirmed its placement in the Gymnoascaceae, Onygenales. No type specimen of S. casei is known to be preserved, and although the species was included in several subsequent studies (Hammer and Gilman 1944; Sigler and Carmichael 1976; Ropars et al. 2012), no type culture has been indicated. To stabilize the nomenclature, we therefore proposed the strain CBS 543.75, isolated from cheese, as epitype.

Sphaerosporium equinum was originally described from a keratinous source (Desmazières 1855, Fig. 5G). However, all strains that were subsequently analysed under this name were isolated from cheese. This could be a result of a lack of sampling from different keratinous substrates. Such that, horse hooves contain beta (β) keratin similar to that of reptiles and birds which is different from that of other mammals containing alpha (α) keratin as the major component (Greenwold et al. 2014; Kakkar et al. 2014). Possibly cheese isolates had been misidentified in the past.

Two strains with superficial similarity to Sph. equinum were derived from other sources than cheese: UAMH 11516 (= MUCL 58097) isolated from a bat wing and MUCL 54024 from an insect pupa. These two strains were also phylogenetically different from the cheese isolates (Fig. 2). In addition, their growth rate, caseinase activity, and ascospore size were also found to be different. Therefore, these two strains were introduced here as a new species. In bat wings, sensory hairs were made of α-keratin (Khan et al. 2014). Insect pupa cocoon structure contains silk which has a different form of β-keratin (Palmer and Bonner 2011). It was also reported that insects contain high quantity of fatty acids in their pupal life stage (Meetali et al. 2014; Smets et al. 2020) which might be a source of nutrition for the fungi grown on this substrate.

Two strains identified as H. aurantiacum were preserved in the CBS collection: CBS 111.18 and CBS 206.35, both originating from salted environments, i.e., cheese and stockfish. These strains produced red–orange colonies (Fig. 4E–H) similar to those of the S. casei strains in the present study. In addition to colony morphology, these two H. aurantiacum strains share the similar habitat and micromorphology as well as the identical ITS, LSU, and TUB sequences with S. casei. Therefore, the strains CBS 111.18 and CBS 206.35 were regarded as previously misidentified and corrected here as S. casei.

Soil, dung, and, fluvial sediments are common sources for onygenalean fungi, and xerophilic and halophilic capacities are characteristic for certain families, such as Ascosphaeraceae and Spiromastigoidaceae (Kandemir et al. 2022, Torres-Garcia et al. 2023). In contrast, species of Sporendonema, Sphaerosporium and Arachniotus are classified in Gymnoascaceae, and are able to grow on substrates with low water activity, such as cheese, dried meat products, and desert soil (Ropars et al. 2012; Scaramuzza et al. 2015). The cheese rind, the prevalent source of isolation of Sporendonema, has high free fatty acid, protein, and salt contents (Kandemir et al. 2022), and in line with this, all cheese isolates were able to tolerate 17% NaCl and showed lipolytic activity in the present study. The only strain lacking lipolysis, MUCL 46080, was isolated from sheep cheese, which forms a soft and bloomy rind different from those of hard cheeses.

Morphologically, S. casei strains yielded hyphae that produce enteroarthric conidia with thick walls and rounded corners (Fig. 4). Spharosporium equinum showed holoblastic conidia with thick walls and was mostly smooth and oblate; S. isthmoides yielded thick-walled, oblate, lenticular conidia produced by thallic-enteroarthric conidiogenesis and showing a distinct point of attachment (Fig. 6). Ascus formation together with arthroaleuriospore production was observed only for A. aurantiacum, A. confluens, and A. ruber (Fig. 3). Nevertheless, these morphological variations did not interfere with the phylogenetic classification of the species in a single genus.

In general, multilocus sequencing data are applied to delimitate fungal species (Giraldo et al. 2014; Kandemir et al. 2020; Crous et al. 2021; Geiser et al. 2021; Hainsworth et al. 2021). In the current dataset, the combined data of ITS + LSU + TUB did not reveal significant differences from those obtained with only ITS + LSU data. Similarly, Ropars et al. (2012) also did not find a major difference between the phylogenetic trees constructed with TEF1 + TUB loci and ITS + LSU sequences of Arachniotus, Sporendonema and Sphaerosporium strains. As all genes yield a similar, stable phylogenetic topology, ITS alone is sufficient to identify Sporendonema species (Fig. S2).

Conclusion

Based on phylogenetic data, species previously described as Arachniotus aurantiacus, A. confluens, A. ruber, Sphaerosporium equinum and Sporendonema casei are congeneric. These fungi represent halophilic, psychrophilic, and xerotolerant members of the Gymnoascaceae. Differences in conidial morphology, cellulolytic and lipolytic ability, casein degradation and maximum temperature of growth are variable between species and even among strains of the same species, but insufficient for accommodating these species in different genera. The individual species within the genus can be recognized by rDNA ITS as a primary barcode.