Introduction

Malaysia, with its warm and moist tropical climate, has been a country with a high diversity of fungi. Many new fungal taxa had been discovered and described during the past decades (Nawawi 1985a, b, 1987; Kuthubutheen and Nawawi 1991, 1994; Lee et al. 2012; Goh et al. 2013, 2014a, 2014b, 2015). During a survey of microfungi occurring on plant litter submerged in a stream in Malaysia, we found a unique hyphomycete producing large, dematiaceous, multi-euseptate, tetrahedral stauroconidia with hyaline filiform appendages at the end of the arms, and enclosed by a thick, hyaline sheath. Superficially the conidia resemble those of Diplocladiella G. Arnaud (Nawawi 1987; Cazau et al. 1993; Lee et al. 1998), but they are produced from brown, short-stalked, bulbose, doliiform to urceolate conidiogenous cells with a terminal pore rimmed with a flared collarette. A literature search has revealed that this fungus on submerged wood has not been previously described (Bhat and Sutton 1985; Goh and Hyde 1996; Goh and Tsui 2003; Seifert et al. 2011; Liu et al. 2016; Lin et al. 2019). As it cannot be suitably placed in any of the known genera of asexual fungi (Seifert et al. 2011), it is described and illustrated in this paper as a new genus. Morphological observation of this unique fungus was supplemented with scanning electron microscopy. The genus is compared with morphologically similar fungi: Adautomilanezia, Anacacumisporium, Bahusutrabeeja, Conioscyphopsis, Craspedodidymum, Cyphellophora, Diplocladiella, Jerainum, Nawawia, Neonawawia, Obeliospora, Phialosporostilbe, Polybulbophiale, Polystratorictus, Pyrigemmula, and Triposporium. We have successfully grown this new fungus in pure culture by single-spore isolation technique. DNA extraction was from these pure cultures and used for molecular studies. Phylogenetic relationship of this new genus was inferred by comparing the nuc rDNA ITS1-5.8S-ITS2 (ITS barcode).

Materials and methods

Sample collection, mycological procedures, and molecular procedures

Plant materials including wood were collected in plastic bags and returned to the laboratory where they were incubated at room temperature under a humid condition in sterile plastic boxes. Materials were examined periodically for the presence of fungal fruiting structures and species were identified primarily based on morphology. Single-spore isolations were made according to the method described in Goh (1999), and the fungi were grown on potato dextrose agar (PDA) slants and plates at 20 °C. Pure cultures from single spores were used for molecular studies. DNA extraction, PCR, and sequencing procedures were similar to the methodology described in Goh et al. (2015). The original specimen of the present fungus previously conserved at the herbarium of the Centre for Biodiversity Research, Faculty of Science, Universiti Tunku Abdul Rahman (UTAR, Perak campus), Kampar, Malaysia, has been recently transferred to Taiwan. Currently, the holotype of this taxon is deposited in the Herbarium (Herbarium Code: TNM) at the National Museum of Natural Science (NMNS), Taichung, Taiwan, whereas an isotype is deposited at the National Chiayi University (NCYU), Taiwan. The ultrastructural features of the present fungus were studied and photographed under the scanning electron microscope (FESEM, Model: JSM-6701F, JEOL, Japan) at UTAR. Air-dried fungal material was directly mounted and sputtered with gold for 60 s for scanning electron microscopy.

Phylogenetic analysis

Sequence data from the ITS region were used to infer phylogenetic placement of the new taxon. DNA sequences were first verified and subjected to BLAST searches to ease phylogenetic taxon sampling. DNA sequences for representative taxa within the Chaetosphaeriaceae retrieved from GenBank were included in our dataset with reference to recent publications (Crous et al. 2012; Liu et al. 2016; Lin et al. 2019). The analysis involved 83 ITS sequences of fungi (Table 1), with Gelasinospora tetrasperma CBS 178.33 (Sordariaceae) being the outgroup taxon. MAFFT was used for DNA alignment (Katoh and Standley 2013). Poorly aligned positions of DNA alignment were manually modified where necessary. There were in total 610 bp in the final dataset including gaps.

Table 1 Sources of sequences and spore groups of taxa used in present phylogenetic analysis
Full size table

The sequences were analysed using MEGA 7 (Kumar et al. 2016). The evolutionary history was inferred using the maximum likelihood and Bayesian inference in RAxML v8.2.4 and MrBayes v3.2.6 under UBUNTU 19.10 (64 bit) operating system (Ronquist and Huelsenbeck 2003; Stamatakis 2014). For the RAxML, substitution model was GTR + G. Random seed for rapid bootstrapping and tree inferences were 5566. Analyses were repeated based on 1000 bootstrapped data sets. MrBayes was run for 1,000,000 generations under GTR + G substitution model, with trees sampled every 100 generations. The first 25% of sampled trees were discarded (relburnin).

Results

Taxonomy

Phaeonawawia Goh, gen. nov.

MycoBank: MB 836839

Type species: Phaeonawawia diplocladielloidea Goh, J.H. Ou & C.H. Kuo

Etymology: From Greek, phaeo- (dark grey or dark-coloured), and the generic name Nawawia Marvanová, denotes that this fungus is similar to Nawawia but producing dematiaceous conidia.

Conidial fungi, hyphomycetous. Colonies on natural substratum effuse, brown. Mycelium partly superficial, partly immersed in the substratum, consisting of smooth, brown branched, septate hyphae. Conidiomata none. Setae and hyphopodia absent. Conidiophores absent or rudimentary in the form of a basal stalk under the single conidiogenous cell. Conidiogenous cells phialidic, discrete or integrated, sessile or incorporated terminally in the basal stalk, bulbose, doliiform or ampulliform, with a distinct, rimmed opening at the apex. Conidia enteroblastic, exogenous, solitary, dry, enclosed by a thick hyaline sheath, tetrahedral, staurosporous, arms multi-euseptate, dematiaceous, setulate. Conidial secession schizolytic. Phylogenetic position: Chaetosphaeriaceae.

Note: There are several hyphomycete genera which are similar to Phaeonawawia, morphological characters of which are compared in Table 2. These include genera that have a similar type of conidiogenesis, producing enteroblastic conidia from discrete bulbose or swollen phialides, such as Adautomilanezia, Obeliospora, Conioscyphopsis, Cyphellophora, Polybulbophiale, Polystratorictus, and Pyrigemmula. Examples of genera producing conidia from mononematous conidiophores with integrated phialidic conidiogenous cells include Anacacumisporium, Bahusutrabeeja, and Craspedodidymum. Genera that have setulate conidia include Nawawia, Neonawawia, Obeliospora, and Phialosporostilbe, whereas those producing dematiaceous staurospores are Diplocladiella, Jerainum, and Triposporium. To date, Phaeonawawia is the only known genus in the Chaetosphaeriaceae producing dry, multiseptate, dematiaceous, versicoloured stauroconidia from discrete bulbose phialides.

Table 2 Comparison of hyphomycete genera similar to Phaeonawawia
Full size table

Phaeonawawia diplocladielloidea Goh, J.H. Ou & C.H. Kuo, sp. nov. (Figs. 1, 2, 3, 4, and 5)

Fig. 1
figure 1

Phaeonawawia diplocladielloidea (TNM: F0034163, holotype). a Colonies on the natural substratum (submerged wood). b–f Conidia, each with a thick hyaline sheath (arrowed in d and e). g An ellipsoidal conidium, bearing one hyaline appendage at each end. Scale bars: a = 500 μm, b–g = 20 μm

Full size image
Fig. 2
figure 2

Phaeonawawia diplocladielloidea (TNM: F0034163, holotype). a Squashed mount from the natural substratum showing a stauroconidium and many bulbose conidiogenous cells. b Close-up of a bulbose conidiogenous cell with a terminal opening. c, d Squashed mount of conidia. Arrows point to empty hyaline conidial sheaths. e Four conidia bearing hyaline filiform appendages, one at each arm. f An ellipsoidal conidium, bearing one hyaline appendage at each end. g A developing conidium at the opening of a bulbose conidiogenous cell. h Four tetrahedral or obpyramidal conidia bearing hyaline filiform appendages. i An empty conidial sheath. j Conidia beside empty conidial sheaths. k, l Conidia. Arrows point to a thick hyaline sheath enclosing the conidium. Scale bars: a, c–e = 50 μm, f–l = 20 μm, b = 5 μm

Full size image
Fig. 3
figure 3

Phaeonawawia diplocladielloidea (TNM: F0034163, holotype). a Several bulbose conidiogenous cells, each with a terminal opening. b Vertical view of a conidiogenous cell with a flared collarette at the terminal opening. Arrow points to the short basal stalk. c Two bulbose conidiogenous cells, one view from the top and the other from the side. d Vertical view of a conidiogenous cell with a short neck and flared collarette (arrowed) at the terminal opening. e–h Developing conidia at the opening of conidiogenous cells. i Two conidia and several stalked conidiogenous cells. Arrow points to the hyaline conidial sheath. j A stalked conidiogenous cell. k, l Stalked conidiogenous cells showing percurrent regenerations (highlighted by an asterisk in l). Scale bars: a–c = 10 μm, d–l = 20 μm

Full size image
Fig. 4
figure 4

Phaeonawawia diplocladielloidea (TNM: F0034163, holotype). Scanning electron micrographs. a Colony on the natural substratum. Arrows point to an ellipsoidal conidium. b, c Clumps of conidia. The asterisk denotes an ellipsoidal conidium. d Two tetrahedral (obpyramidal) conidia and two conidiogenous cells. Filiform appendages are visible (arrowed). e–g Conidiogenous cells. h–j Conidia. Arrows in h point to filiform appendages. Large pores in i are ends of conidial arms lacking filiform appendages. Arrow in j points to the basal hilum of an obpyramidal conidium. Scale bars: a = 100 μm; b = 50 μm; c = 20 μm; d, h–j = 10 μm, e-g = 5 μm

Full size image
Fig. 5
figure 5

Phaeonawawia diplocladielloidea. Diagrammatic representation of conidiogenesis. a A discrete phialide situated on a basal stalk. b–e Sequential steps of conidiogenesis. f A mature stauroconidium bearing a setula at the end of each arm. g A mature ellipsoidal conidium bearing a setula at each end. h An old phialide showing percurrent regeneration. i Conidium ontogeny from regenerated phialide. j Percurrent regeneration of phialides. Scale bar = 20 μm

Full size image

MycoBank: MB 837328

Etymology: From Greek, −oides (resembling), and the generic name Diplocladiella G. Arnaud, the epithet “diplocladielloidea” denotes that this fungus is similar to Diplocladiella in producing dematiaceous stauroconidia.

Colonies on natural substratum effuse, brown, somewhat glistening. Mycelium partly superficial, partly immersed in the substratum, consisting of smooth, brown branched, septate hyphae. Conidiophores absent or rudimentary in the form of a basal stalk (10–20 × 4–6 μm) under the single conidiogenous cell. Conidiogenous cells phialidic, discrete or integrated, sessile or incorporated terminally in the basal stalk, bulbose, doliiform or urceolate, (26.5)30–35.5 × 10–11.5 μm, with a distinct, rimmed opening (3.5–6 μm wide) at the apex, monoblastic. Sometimes regenerating percurrently. Conidia staurosporous, with 3–4 arms, broadly rounded or sometimes truncated at the tip of the arms, enteroblastic, exogenous, solitary, dry, smooth-walled, enclosed by a hyaline sheath (2–2.5 μm thick), tetrahedral to obpyramidal or occasionally ellipsoidal, mostly an equilateral triangle in surface view, with a length of (53)60–70(76.5) μm at each side and a height of 50–61 μm, multicellular, arms 2–4-euseptate, septa thick and sometimes banded, not constricted at the septa, olivaceous to medium brown and versicoloured, with the central cell darker and the arms paler, each arm bears a hyaline, aseptate, filiform appendage (25–48.5 × 2–2.5 μm); ellipsoidal conidia 75–85 × 25–30 μm, with two arms, each bearing a setula. Conidial secession schizolytic.

Teleomorph: Unknown

Conidiogenesis: Conidium ontogeny is enteroblastic, monoblastic, begins as a spherical, hyaline blown-out at the opening of the bulbose phialide. The blown-out then enlarged and becomes more or less ellipsoidal, positioning horizontally at the opening of the phialide, with its proximal side more tapering. The young conidium during the early development stages is hyaline and subsequently become more pigmented, while it continues to enlarge and finally becoming septate, setulate, and enclosed by a thick hyaline sheath. The conidium either develops into a horizontally oriented, ellipsoidal and bisetulate form, or into an obpyramidal to tetrahedral, 3–4-armed stauroform, with a hyaline setula at the end of each arm. Occasionally, the phialides may regenerate by forming a new phialide inside or outside the old ones (Fig. 3i). The process of conidiogenesis and phialide regeneration is depicted in Fig. 5.

Specimen examined: MALAYSIA. PERAK. Menglembu, Bukit Kledang, 4.58027–101.02528, 253 m a.s.l., on submerged wood, 9 January 2014, leg. Wai-Yip Lau and Teik-Khiang Goh, UTAR-G1, Herb. no. TNM: F0034163 (holotype), NCYU-UTAR-G1 (isotype); GenBank: ITS = MT946684

Note: Phaeonawawia diplocladielloidea is unique in producing dematiaceous staurospores from discrete bulbose phialides. Its conidia resemble those of Diplocladiella species the most, especially those of D. alta R. Kirschner & Chee J. Chen; D. aquatica O.H.K. Lee, Goh & K.D. Hyde; D. cornitumida F.R. Barbosa, Gusmão & R.F. Castañeda; and D. tricladioides Nawawi, which are Y-shaped or triangular in face view, and bear filiform appendages at the end of the conidial arms. However, these species differ in having distinct conidiophores which are geniculate, cicatriced, with integrated polyblastic sympodial conidiogenous cells (Nawawi 1985b; Lee et al. 1998; Kirschner and Chen 2004; Barbosa et al. 2007). Two other fungi also have stauroconidia which are comparable with those of P. diplocladielloidea. In Jerainum triquetrum Nawawi & Kuthub., conidia are triangular in face view, muriform, bearing a single hyaline appendage at the base and several others at the distal ends. However, the conidiogenous cells are not phialides, instead, they are holoblastic, monotretic, and percurrently extending, although they are doliiform or ellipsoidal reminiscent of phialides (Nawawi and Kuthubutheen 1992). The stauroconidia of Triposporium elegans Corda, also resemble those of P. diplocladielloidea but lack filiform appendages, and also differ in conidiogenesis, being holoblastic and non-phialidic (Corda 1837; Ellis et al. 1951). Several other hyphomycetes have discrete and/or integrated bulbose phialides similar to those of Phaenawawia, especially those of Craspedodidymum spp. (Yanna et al. 2000; Pinruan et al. 2004; Ma et al. 2011; Mel’nik et al. 2014), Obeliospora spp. (Nawawi and Kuthubutheen 1990; Wu and McKenzie 2003; Cantillo-Pérez et al. 2018), and Polybulbophiale palmicola (Goh and Hyde 1998b), but they all differ in conidial morphology. Detailed comparison of conidiogenous cells and conidial morphology is given in Table 2.

Phylogeny

Phylogenetic tree (TreeBASE TB2:S27242) inferred from aligned ITS sequences of 83 representative fungal taxa showing evolutionary relationships of Phaeonawawia diplocladielloidea with other fungi in the Chaetosphaeriaceae using the maximum likelihood and Bayesian inference statistical methods, with Gelasinospora tetrasperma CBS 178.33 (Sordariaceae) being the outgroup, is shown in Fig. 6. Molecular data revealed several branches and small clusters of taxa (designated alphabetically from A1 to R in Fig. 6) representing most of the anamorphic groups with phialidic conidiogenous cells in the Chaetosphaeriaceae. Phaeonawawia diplocladielloidea (O) did not cluster with any of the existing anamorphic groups and represents a distinct taxon in the Chaetosphaeriaceae with high bootstrap support and Bayesian inference. The tree also shows 12 major groups of fungal taxa in the family according to the spore types they produced (Table 1), which are indicated by the Roman numbers I–XII, but mostly without bootstrap support. Group I comprised three genera producing hyaline amerospores, including Neonawawia (A1) and Nawawia (A2) with setulate tetrahedral conidia, and Zanclospora (B) with conidia lacking appendages. Group II included Cryptophiale (C1) and Cryptophialoidea (C2), which are similar genera producing hyaline falcate or fusoid spores from groups of conidiogenous cells on the shaft of setiform conidiophores. Group III and group IV included the two major groups of anamorphic fungi: the acervular coelomycetes (D1–D5) and the hyphomycetes (E1 and E2). All taxa in these two groups produce hyaline setulate unicellular conidia (amerospores). The hyphomycete genera in these two groups belonged to the Menispora groups or the Dictyochaeta complex, which were composed of several morphologically similar genera, namely Codinaeopsis, Dictyochaeta, and Menispora. Group V comprised a mixture of spore types produced by various taxa. These include Infundibulomyces (D6), a cupulate coelomycete genus producing hyaline setulate amerospores, the Chlorodium complex (F1 and F2) producing subhyaline amerospores in mass, and Sporoschisma (G) producing catenate phaeophragmospores. Group VI comprised several similar genera belonging to the Menispora groups or the Dictyochaeta complex, producing hyaline setulate conidia (E3-E6), and the genus Anacacumisporium (H) which produces coloured multiseptate conidia (phaeophragmospores). Group VII comprised two morphologically distinct genera: the synnematous hyphomycete Phialosporostilbe (I) producing setulate hyaloconidia, and the sporodochial hyphomycetes Adautomilanezia (J) producing solitary phaeophragmospores. Group VIII comprised three species of Menisporopsis (K) producing setulate hyaloconidia from synnematous conidiophores. Group IX was represented by Pyrigemmula aurantiaca (L), a hyphomycete producing distoseptate phragmospores. Group X was composed of a mixture of taxa producing various spore types, including Phaeonawawia (O) producing solitary versicoloured staurospores, Catenularia (N) producing coloured cuneiform amerospores in chains, Kylindria complex (M1 and M2) producing septate hyalospores or phaeospores in mass, and Craspedodidymum species (P) producing coloured amerospores. Group XI was composed of Chloridium-like taxa representing the Gongromeriza complex (Q). Group XII was represented by two species of Kionochaeta (R) producing subhyaline amerospores from setiform conidiophores.

Fig. 6
figure 6

Phylogenetic tree inferred from aligned ITS sequences of 83 fungal taxa showing relationships of Phaeonawawia diplocladielloidea and other fungi in the Chaetosphaeriaceae using the maximum likelihood and Bayesian inference in RAxML v8.2.4 and MrBayes v3.2.6. The tree was rooted with Gelasinospora tetrasperma (Sordariaceae). Bootstrap support values (greater than 50%) from the maximum likelihood analysis and posterior probabilities (greater than 0.7) from the Bayesian analysis are shown near each node. The tree shows branches and small clusters of taxa (designated alphabetically from A1, A2, B to R) representing most of the anamorphic groups in the Chaetosphaeriaceae. The Roman numbers I–XII indicate the major groups of fungal taxa in the family according to the spore types they produced. The anamorphic names of known Chaetosphaeria species were in brackets. The result shows that P. diplocladielloidea (in bold red) did not cluster with any of the existing anamorphic groups and represents a distinct taxon in Chaetosphaeriaceae

Full size image

Discussion

Reflection on the hyaline conidial sheath

An outstanding feature in Phaeonawawia diplocladielloidea is the thick, hyaline sheath enclosing its stauroconidia. Several technical terms, namely ectosporium, endosporium, episporium, exosporium, and perisporium, have been used to describe different layers of spore walls seen in certain ascospores (Goh and Hanlin 1999; Kuhnert et al. 2016), basidiospores (Halbwachs and Bässler 2015), and teliopores (Khanna and Payak 1968).

The perisporium has been referred to be the “sheath” by several authors. It is the outermost layer of the spore and is usually thin, hyaline and sometimes fugacious (Kirk et al. 2008). Detachable perisporic sheaths have been reported in the ascospores of Annulohypoxylon species (Kuhnert et al. 2016) and the teliospores of certain smut fungi (Khanna and Payak 1968). Likely, the thick, hyaline, sheath-like outer covering of the conidia in P. diplocladielloidea is the perisporium. Empty sheaths were frequently observed in squashed mounts of the present collection (Figs. 2a–j and 3i), and they were probably the dehisced and detached perispores of the conidia. Such isolated sheaths might also be immature conidia lacking cellular content or perhaps due to an undetermined artefact of shrinking cytoplasm. However, the evidence is lacking, and these explanations remain hypothetical. A similar detachable outer coating (described as the “episporic sheath”) has been reported in the didymoconidia of Cordana abramovii var. seychellensis K.D. Hyde & Goh (Hyde and Goh 1998). Such detachable sheath is, however, absent in Cordana abramovii var. abramovii Seman & Davydkina, a species which has been commonly collected worldwide (Seman and Davydkina 1983; Rao and de Hoog 1986; Zelski et al. 2014; Santos et al. 2018; Luo et al. 2019). Besides these two species, there are currently no other conidial fungi reported to have such a dehiscent or detachable “perisporic” sheath.

Diversity of anamorphic genera in Chaetosphaeriaceae

In a recent review of chaetosphaeriaceous fungi, Lin et al. (2019) recognised 49 genera in the family Chaetosphaeriaceae, among which 5 are teleomorphic names (ascomycetes) and 44 are anamorphic names (i.e. 9 are coelomycetes, 35 are hyphomycetes). The compilation of hyphomycete genera by Seifert et al. (2011) has contributed to the understanding of the diverse asexual forms of fungi, including those known to be anamorphs of Chaetosphaeria. Since then, several new asexual genera were added to the family (Lin et al. 2019) over the last few years. These include four coelomycete genera (Crous et al. 2012; Hashimoto et al. 2015; Hernández-Restrepo et al. 2016), namely Brunneodinemasporium, Calvolachnella, Neopseudolachnella, and Pseudodinemasporium, and six hyphomycete genera (Magyar et al. 2011; Crous et al. 2016, 2017; Ma et al. 2016; Yang et al. 2018), namely Adautomilanezia, Anacacumisporium, Eucalyptostroma, Neonawawia, Pyrigemmula, and Verhulstia. The description and illustration of Phaeonawawia diplocladielloidea in this paper added a unique hyphomycete genus to this family.

Addition of Phaeonawawia to Chaetophaeriaceae with minimal phylogenetic support

In this paper, we included a simple phylogenetic analysis to support our proposal of Phaeonawawia as a new genus in Chaetosphaeriaceae. Not all the known anamorphic genera in Chaetosphaeriaceae were included in the present phylogenetic study. We have excluded taxa that lack available sequences in GenBank and a few of those that are morphologically distinct from Phaeonawawia so that the tree was not too large lest it complicated the analysis. Since we only used the ITS sequence data to infer evolutionary relationships of taxa, some anamorphic taxa in the family that currently do not have available ITS sequences in the GenBank, such as Exserticlava, Morrisiella, and Stanjehughesia, were excluded from the present phylogenetic analysis. Extensive phylogenetic analyses of chaetosphaeriaceous teleomorphs and associated anamorphs are beyond our focus for the present study. This is because the specimen of Phaeonawawia diplocladielloidea was collected in January 2014, and we had only got its ITS sequenced. It is indeed a pity that a living ex-type culture of this new taxon is no longer available today, and therefore, no other gene sequences from it could be obtained for further phylogenetic studies. Due to this limitation in the selection of ITS gene segment for the present study, it is not suitable for large trees covering extensive phylogenetic studies of multiple taxonomic groups. However, the anamorphic genus Phaeonawawia is phylogenetically distinct, with outstanding morphological features in the family Chaetosphaeriaceae.

Teleomorph-anamorph connections

Species of Chaetosphaeria teleomorphs are generally simple and relatively homogeneous; however, their anamorphs are morphologically distinctive, complex, and diverse. Because teleomorphs of chaetosphaeriaceous fungi are hardly distinguishable, species identification is therefore based primarily on characters of the anamorphs. Attempts to solve in part the natural status of Chaetosphaeria and its anamorphs have been ongoing since the 1970s (Gams and Holubová-Jechová 1976; Fernández et al. 1998; DiCosmo et al. 1983; Réblová and Gams 1999; Réblová 2000; Réblová and Winka 2000; Réblová and Seifert 2003; Fernández and Huhndorf 2005; Huhndorf and Fernández 2005; Fernández et al. 2006). Previous phylogenetic studies had revealed that species groupings within Chaetosphaeria are concordant with groupings based on morphological characters of their anamorphs. Certain general morphological patterns indicative of phylogenetic relationships were discerned within the family. Although Chaetosphaeria species appear homogeneous in morphology, phylogenetic analyses (Réblová and Winka 2000; Réblová and Seifert 2003) reveal that the genus is not monophyletic. Similar to previous findings, the result of the present phylogenetic study shows that some of the anamorphic genera associated with Chaetosphaeria are monophyletic, each clade with strong bootstrap support, such as Craspedodidymum, Menispora, Menisporopsis, Sporoschisma, and Thozetella, whereas others are polyphyletic and complex, such as Chloridium and Dictyochaeta. Réblová (2000) distinguished some of these complex anamorphs of Chaetosphaeria and divided them into four natural groups of taxa based on morphological, cultural, and molecular studies, namely the Chloridium group, the Gongromeriza group, the Kylindria group, and the Menispora group. However, these groupings are polyphyletic. The present phylogenetic tree (Fig. 6) inferred from aligned ITS sequences shows a similar result of groupings: the Chloridium group comprises species of Chloridium (F1) and Gonytrichum (F2); the Gongromeriza group (Q) comprises species of Chloridium, Dictyochaeta, and Phialophora; the Kylindria group comprises species of Chloridum (M1) and Cylindrotrichum (M2); and the Menispora group comprises species of Codinaeopsis (E1), Dictyochaeta (E1, E3, E5), and Menispora (E2). Based on molecular data and cultural characters, Huhndorf and Fernández (2005) recognised a group of Chaetosphaeria species that has teleomorph-anamorph connections with some Craspedodidymum species and rarely with Chloridium-like synanamorphs. The present phylogenetic study shows the same result, with high bootstrap support on the Craspedodidymum group (P).

Reflection on “one fungus = one name”

The adoption of a dual nomenclatural system for fungal species has been a tradition in mycology. When pleomorphism in fungi is encountered, confusion and frustration experienced by many practitioners of mycology and plant pathology are reckoned. The existence of synanamorphs in certain fungi further enhances the confusion with multiple fungal names. With the advent of DNA techniques and the era of molecular phylogeny in fungal systematics, mycologists nowadays have a better understanding of pleomorphism in fungi. Taylor (2011) proposed a “one fungus one name” of nomenclatural system to solve the confusion. This system has been followed by many mycologists and has particularly welcomed by the plant pathologists, as they recognised that many important plant pathogens produce the asexual forms of spores or propagules to facilitate disease dissemination and the sexual forms for overwintering (Wingfield et al. 2011; Rossman et al. 2016). Based on the concept of “one fungus one name”, Réblová et al. (2016) recommended the adoption of either the sexual names or the asexual names for some taxa in the Sordariomycetes. These include the preference of adopting the anamorphic names over their teleomorphic names for several taxa in the Chaetosphaeriaceae, namely Chloridium (instead of Melanopsammella Höhn.), Menispora (instead of Zignoëlla Sacc.), Menisporopsis (instead of Menisporopascus Matsush.), Sporoschisma (instead of Melanochaeta E. Müll., Harr & Sulmont), and Stanjehughesia (instead of Umbrinosphaeria Réblová). We concur with Réblová et al. (2016) to keep these anamorphic names, as we realise that chaetosphaeriaceous fungi are relatively homogeneous in their teleomorphic forms but quite diverse in their anamorphic forms. The recognition of the various asexual forms in Chaetosphaeriaceae and the conservation of these anamorphic names may be helpful in species identification for the time being. Both the sexual and asexual names are therefore cited wherever possible in this paper (Fig. 6) to facilitate identification and examination of these fungi.

Phialides and conidial morphology in Chaetophaeriaceae

Majority of the asexual genera in Chaetosphaeriaceae have phialidic conidiogenesis (Liu et al. 2016; Lu et al. 2016; Lin et al. 2019). The phialides in these genera differ in shapes (e.g. lageniform, ampulliform, and bulbose), in conidiogenous loci (monophialidic or polyphialidic), in conidium ontogeny, and in development (e.g. solitary, catenulate, or in slimy mass). More comprehensive studies of phialides and conidial development have been given by Hughes (1953), Tubaki (1958), and Minter et al. (1982, 1983). The mechanisms of regeneration in conidiogenous cells (i.e. how a no-longer functional conidiogenous cell is replaced) are discussed in Minter et al. (1982). In certain phialidic fungi, their phialides undergo intermittent regenerations between conidiogenous episodes (Minter et al. 1983), either percurrently, as in species of Catenularia (Hughes 1965) and Nawawia (Goh et al. 2014b), or sympodially, as in species of Codinaea and Dictyochaeta (Luo et al. 2019). In the present paper, percurrent regenerations of phialides that are bulbose or urceolate in shape were observed in Phaeonawawia diplocladielloidea (Fig. 3). The same manner of regeneration has also been observed in other hyphomycetes with bulbose phialides, such as Polybulbophiale palmicola (Goh and Hyde 1998b) and Obeliospora microappendiculata (Cantillo-Pérez et al. 2018). A detailed study of phialides and conidial development in Phaeonawawia is out of the scope of the present paper. Among the various anamorphic forms in Chaetosphaeriaceae, several genera, such as Morrisiella Saikia & A.K. Sarbhoy, Stanjehughesia Subram., and probably Multiguttulispora C.G. Lin & J.K. Liu, however, are not phialidic, instead, they produce holoblastic conidia from mono- or polyblastic conidiogenous cells.

Conidia of chaetosphaeriaceous anamorphs come in diverse forms, but mostly they bear hyaline appendages (Crous et al. 2012; Liu et al. 2016; Lin et al. 2019). Shenoy et al. (2006) reported some Sporidesmium-like taxa phylogenetically positioned in the Chaetosphaeriales, namely Ellisembia brachypus, Linkosia sp., Morrisiella indica, and Stanjehughesia vermiculata, but these hyphomycetes produce dark, obclavate or rostrate, non-setulate phragmoconidia from holoblastic conidiogenous cells. To date, Phaeonawawia is the only known asexual genus in the Chaetosphaeriaceae that produces versicolored setulate stauroconidia, although the setulate conidia in Nawawia or Neonawawia may be regarded as staurosporous, they are unicellular and hyaline. This genus was collected from decaying wood submerged in freshwater streams. Such stauroconidia appear to be adapted to dispersal by water and attachment to submersed substrata (Goh and Hyde 1996).

On the contrary, with evidence from many phylogenetic studies of asexual fungi based on multigene analyses in recent decades, some phialidic hyphomycetes that had been considered to belong to the Chaetosphaeriales in the past were inferred to belong to other ordinal lineages. Examples of non-chaetosphaeriaceous fungi that produce phialoconidia (Cai et al. 2009; Réblová et al. 2011; Maharachchikumbura et al. 2018) include Monilochaetes species (Australiascaceae, Glomerellales), the Chalara & Exochalara complex (Helotiales, Leotiomycetes), and the Kylindria & Cylindrotrichum complex (Reticulascaceae, Glomerellales).

Conclusions

This paper describes and illustrates Phaeonawawia diplocladielloidea from decaying wood submerged in freshwater. It is a new anamorphic taxon belonging to Chaetosphaeriaceae with phialoconidia of a unique morphology. Despite multiple efforts to study species of Chaetosphaeria and their anamorphs, there exist some unresolved problems in their taxonomy. One of the limitations is that until today, not all chaetosphaeriaceous taxa have their DNA sequenced. There also exist several polyphyletic taxon groups which have some of their members scattered in the Chaetosphaeriaceae and also among other fungal lineages, such as the Chalara complex (Cai et al. 2009), the Chloridium complex (Gams and Holubová-Jechová 1976), the Dictyochaeta complex (Wei et al. 2018; Liu et al. 2016; Lin et al. 2019) the Kylindria & Cylindrotrichum complex (DiCosmo et al. 1983; Maharachchikumbura et al. 2018); the Phialophora complex (Gams 2000; Réblová et al. 2011), and the Sporidesmium complex (Shenoy et al. 2006). Another interesting aspect for further detailed studies of chaetosphaeriaceous fungi is the phialidic conidiogenous cells of their anamorphs which exist in various forms, especially it has been over 30 years without further extensive research on the developmental biology of phialides since the contributions of Hughes (1953), Tubaki (1958), and Minter et al. (1982, 1983). Chaetosphaeriaceous fungi remain complex, especially the biology and phylogeny of the anamorphs, and await more work to resolve in the future.