Two new species of the genus Verruconis, V. hainanensis and V. pseudotricladiata, were described using combined morphological and DNA sequence data. The DNA sequences of respective strains including nuclear ribosomal DNA genes (nuSSU, ITS, nuLSU) and fragments of three protein-coding genes (ACT1, BT2, TEF1) were sequenced and compared with those from closely-related species to genera Ochroconis and Verruconis (Family Sympoventuriaceae, Order Venturiales). Morphologically, both species showed typical ampulliform conidiophores and conidiogenous cells, features not seen in other species of Verruconis. The conidia of V. hainanensis are fusiform and those of V. pseudotricladiata are Y or T shaped, similar to old members of a closely-related genus Scolecobasidium. The addition of these two new species provides a new perspective on the heterogeneity of Scolecobasidium.

Aquatic hyphomycetes, dematiaceous fungi, phylogenetic placement, new taxon

The genus Verruconis Samerp. et al. was proposed for the neurotropic opportunist Ochroconis gallopava (W.B. Cooke) de Hoog (Samerpitak et al. 2014). The thermophilic characteristic of this genus is remarkable because all three proposed species of Verruconis can grow at 35–42 °C. In addition to the difference in growth temperature, Verruconis and Ochroconis de Hoog & Arx also differed in conidia colour (Samerpitak et al. 2014). However, a recent molecular phylogenetic analysis placed the mesophilic V. panacis T. Zhang & Y. Zhang into Verruconis, a result suggesting that both genera are more heterogeneous in their morphological and growth requirements than previously thought (Zhang et al. 2018).

Besides V. panacis, other three Verruconis species were transferred from other genera. The type species, V. gallopava (W.B. Cooke) Samerp. & de Hoogs [≡ Dactylaria gallopava (W.B. Cooke) G.C. Bhatt & W.B. Kendr., ≡ Ochroconis gallopava (W.B. Cooke) de Hoog] was transferred from Diplorhinotrichum Höhn.; V. verruculosa (R.Y. Roy et al.) Samerp. & de Hoog (≡ Scolecobasidium verruculosum R.Y. Roy et al.) was transferred from Scolecobasidium and V. calidifluminalis (Yarita et al.) Samerp. & de Hoog (≡ Ochroconis calidifluminalis Yarita et al.) was transferred from Ochroconis. These reclassifications suggested that genera Ochroconis, Verruconis and Scolecobasidium E.V. Abbott are closely related and that both morphological and molecular data are needed in order to derive robust classifications. Ochroconis, typified by O. constricta (E.V. Abbott) de Hoog & Arx, transferred from Scolecobasidium, was set up to comprise species with unbranched, subspherical to cylindrical or clavate conidia. Based on these criteria, many Scolecobasidium species were transferred to Ochroconis, while species in the genus Scolecobasidium were restricted to those with T- or Y-shaped or bi-lobed, two- to many-celled conidia and ampulliform conidiogenous cells, possessing one to three conidium-bearing denticles at the apex of the conidiogenous cells (de Hoog and von Arx 1973). However, there is a significant disagreement amongst mycologists about whether the genus Ochroconis should be established and some researchers still placed species with unbranched conidia under Scolecobasidium (Ellis 1976; Matsushima 1980, 1985, 1987, 1993, 1996; Punithalingam and Spooner 2011; Lu et al. 2013; Ren et al. 2013; Xu et al. 2014).

Samerpitak et al. (2014) revised the genera Ochroconis and Scolecobasidium using DNA sequences of the nuclear ribosomal RNA gene clusters and three protein-coding genes (actin: ACT1, β-tubulin: BT2, translation elongation factor 1-α: TEF1). They found that the type species of Scolecobasidium, S. terreum E.V. Abbott, ex-type strain CBS 203.27, originally described as having the T-shaped conidia, had lost the ability to produce conidia. Interestingly, this strain was phylogenetically distant from other strains with Y-shaped conidia as described for S. terreum in all analyses. Consequently, type strain S. terreum CBS 203.27 is now regarded as a non-representative strain of the species and, indeed, the validity of this species has been questioned and Scolecobasidium is considered to be of doubtful identity.

However, Gams thought that an ex-type culture was not so important to decide if a genus is retained, because there are other cultures of S. terreum available all over the world, which clearly define the identity of this characteristic fungus. He even thought that CBS 510.71, the ex-type of Humicola minima Fassat., a species with characteristic Y-shaped conidia, may replace S. terreum (Gams 2015). However, in Samerpitak’s analysis, many Scolecobasidium species were scattered outside the Family Sympoventuriaceae. Consequently, the genus Scolecobasidium has been questioned (Samerpitak et al. 2014). Since then, several new Ochroconis species have been described under Ochroconis (Giraldo et al. 2014; Samerpitak et al. 2015a; 2015b; 2017; Crous et al. 2016; 2017), while the number of Scolecobasidium species has not increased since 2014 (Index Fungorum 2018). Species with forked conidia, similar to S. terreum, were also added to Ochroconis based on phylogenetic relationships amongst members of Sympoventuriaceae (Giraldo et al. 2014). The strict morphological characters to demarcate Scolecobasidium were abandoned in favour of the molecular phylogenetic approach. Subsequent analyses based on combined molecular sequence information, ecological and physiological traits and morphological differences resulted in the establishment of the genus Verruconis.

Hainan Province, China is a centre of biodiversity for aquatic hyphomycetes. Since 2015, we have reported several new aquatic hyphomycetes from this area (Guo et al. 2015; Qiao et al. 2017, 2018). During further studies of aquatic hyphomycetes on submerged decaying leaves collected from a stream in Hainan Province, we encountered two fungi which resembled species of Scolecobasidium. Based on phylogenetic analyses, we identified that the fungi belonged to Verruconis. In this paper, we describe the two fungi as new species and determined their phylogenetic placement based on the combined sequences of SSU, ITS, LSU, BT2, TEF1 and ACT1.

The Index Fungorum currently lists 66 names in Scolecobasidium. However, 22 of these 66 names have been transferred into genera Dactylaria Sacc., Paradendryphiella Woudenb. & Crous, Ochroconis, Trichoconis Clem., Neta Shearer & J.L. Crane and Verruconis (Index Fungorum 2018). Of the remaining 42 species, the majority lacks authentic culture materials and DNA sequence data, making the revision of Scolecobasidium very difficult. However, since 2014, the number of Scolecobasidium species has not increased, while many new species have been reported under Ochroconis, including species with forked conidia (Giraldo et al. 2014). Although Scolecobasidium is still listed as an accepted genus of Ascomycota (Wijayawardene et al. 2017), this genus will likely be phased out. Thus, we have placed our strains into Verruconis based on phylogenetic analysis.

Morphologically, the two new species resemble some members of the genus Scolecobasidium. Conidiophores composed of 2–5 globose serial cells are very typical in old members of Scolecobasidium, such as S. alabamense Matsush., S. amazonense Matsush., S. cateniphorum Matsush. and S. lanceolatum Matsush. However, amongst these species, only the LSU sequence of S. cateniphorum was available. Further, Y- branched conidia of V. pseudotricladiata was previously only described in S. tricladiatum, while T-shaped branched conidia appeared in four species, including the type species S. terreum, O. minima (Fassat.) Samerp. & de Hoog, O. ramosa Samerp. et al. and O. icarus Samerp. et al. In the molecular phylogenetic tree, inferred from the combined sequences of six marker loci, except for the type species, three species with T-shaped branched conidia form a single clade with high support within Ochroconis. In the combined analysis of SSU and LSU, S. tricladiatum strain P051 is closely related to V. pseudotricladiata and S. terreum 043 fell into Ochroconis, nested with other species with T-shaped branched conidia (data not shown). The phylogenetic analysis is partly consistent with the morphological comparison. The article, comprising sequences of S. tricladiatum strain P051 and S. terreum 043, has not been published and we do not know if two species have been identified correctly. Anyhow, molecular data for our strains will help improve the taxonomy and revision of Scolecobasidium.

When the genus Verruconis was established, the thermophilic character was one of the main characteristics distinguishing this genus from Ochroconis. The first three species included in this genus all have a high optimal growing temperature of 35–42 °C and maximum growing temperature of 47–50 °C (Samerpitak et al. 2014). However, both our species and their close relative V. panacis are mesophilic, which blurred a major distinguishing feature between Verruconis and Ochroconis. Morphologically, Verruconis is characterised by poorly differentiated, flexible, mostly cylindrical to acicular, with 0(−1) thin septa conidiophores, sometimes without conidiophores (Samerpitak et al. 2014). However, conidiophores of two new species are distinct, only occasionally reducing to conidiogenous cells. Ampulliform conidiogenous cells also appeared in O. minima and O. icarus, but no species in Verruconis and Ochroconis have similar conidiophores to those of the two new species, which were composed of 2–5 globose serial cells. Based on the phylogenetic relationships amongst the species, the distributions of morphological features indicate that conidiophores and conidiogenous cells are important features for defining these two related genera. Our results suggest that the analyses of more sequences and more cultures in this group of fungi are needed to provide a robust revision of the three genera Verruconis, Ochroconis and Scolecobasidium.

This work was financed by the National Natural Science Foundation of PR China (31770026, 31760012). We are grateful to two reviewers for critically reviewing the manuscript and for providing helpful suggestions to improve this paper.