Taxonomy of Hyphodermella: a case study to show that simple phylogenies cannot always accurately place species in appropriate genera

Abstract

The genus is a special and crucial taxonomic rank compared with others above the species level, because a species has to be placed in a certain genus instead of any other higher ranks. With more and more new species being described, the placements of their generic position are sometimes incorrect due to the simple phylogenies resulting from inappropriate sampling. Here, we focus on the taxonomy of a small wood-inhabiting fungal genus Hyphodermella. With the most comprehensive sampling to date, the phylogenetic position of Hyphodermella within Phanerochaetaceae is rearranged by employing the same ITS and nLSU regions as in previous studies and also the ITS, nLSU, rpb1, rpb2 and tef1α regions. Three species are excluded from Hyphodermella: H. poroides is placed in a newly introduced monotypic genus Pseudohyphodermella, while H. aurantiaca and H. zixishanensis are transferred to Roseograndinia. Hyphodermella suiae is described as a new species from South China and Vietnam. Keys to eight species in Hyphodermella and five in Roseograndinia are provided. Beyond solving the taxonomic issue of Hyphodermella itself, the current study also aims to suggest that all fungal taxonomists especially beginners should keep in mind to sample as many comprehensive taxa as possible in phylogenetic analyses.

INTRODUCTION

Despite being one of the most species-rich life forms, Fungi are poorly documented with more than 90% of estimated species (2.2 to 3.8 million species) awaiting formal description (Hawksworth and Lücking 2017). To enlarge the knowledge of fungal diversity, more than one thousand species have been newly introduced each year during the last decades (Dai et al. 2015; Hawksworth and Lücking 2017; Niskanen et al. 2018). Above the species level, genus is a special and crucial taxonomic rank compared with other ranks under the binomial nomenclature system, because a species has to be placed in a certain genus but may be not assigned in any certain higher rank than genus. Although molecular phylogenies are helpful to determine the generic position of fungal species, the placements are sometimes incorrect due to the use of simple phylogenies resulted from inappropriate sampling in a bad practice of phylogenetic analyses. Here, a simple phylogeny is defined to sample only targeted species but not closely related outgroup taxa; in this way, the generic circumscription cannot be reliably delimited (Fig. 1). In contrast, a “good” genus can only be accurately delimited by sampling more related taxa to the targeted species. Indeed, increased taxon sampling has long been known as an efficient method to reduce error signals in phylogenetic analyses (Zwickl and Hillis 2002; Prasanna et al. 2020).

Fig. 1

A schematic illustration of the ‘simple phylogeny’ resulted from inappropriate sampling in a bad practice of phylogenetic analyses. Whatever the statistical support at the node C is high or not, species in Clade A is not always congeneric with species in Clade B

To clearly present the results of simple phylogenetic studies in fungal taxonomy and the resulting incorrect generic placements of taxa, two examples recently dealt with by us are briefly summarized here. One is the incorrect placements at the generic level of two species originally placed in Heteroradulum, viz. H. yunnanense (with the wrong masculine gender as ‘yunnanensis’; Guan et al. 2020) and H. niveum (Li et al. 2022a). In Guan et al. (2020), several taxa of Heteroradulum were selected as the only ingroup and H. yunnanense was placed at the basal position within the so-called Heteroradulum lineage; actually, this simple phylogeny cannot determine whether H. yunnanense should be the member of Heteroradulum or not. With the help of a more comprehensive sampling, a later phylogenetic analysis clearly separated H. yunnanense from Heteroradulum and thus excluded it from this genus (Li et al. 2022b).

Similarly, according to a simple phylogeny (Li et al. 2022a: Fig. 2) adopted from Guan et al. (2020), the new species Heteroradulum niveum was further incorrectly placed in Heteroradulum (Li et al. 2022a). Even worse, the accompanying phylogeny in that paper (Li et al. 2022a: Fig. 1) did not cluster H. niveum with other species of Heteroradulum with reliable statistical support at all. In contrast, Liu et al. (2022b) thoroughly explored the phylogenetic relationships among Heteroradulum and its close genera, which resulted in a new genus Alloexidiopsis for the clade composed of H. yunnanense and H. niveum.

Another example is two species originally placed in Trechispora (Zong et al. 2021) and then in Brevicellicium (Liu et al. 2022c). In Zong et al. (2021), newly describing Trechispora daweishanensis and T. xantha, the first phylogeny did not recover the monophyly of Trechispora with these two species, while the second one simply including taxa only from Trechispora as the ingroup clustered the two species with T. yunnanensis and separated them from additional species of Trechispora. As first noted by Chikowski et al. (2020) and then confirmed by Liu et al. (2022a), the ITS and nLSU sequences from specimens of T. yunnanensis (Xu et al. 2019) actually represent different species from Trechisporales and Hymenochaetales, respectively, and thus the phylogenetic position of T. yunnanensis itself is doubtful. Liu et al. (2022c) recognized the incorrect generic placements of T. daweishanensis and T. xantha by Zong et al. (2021), and transferred these two species to Brevicellicium. However, the phylogeny supporting these transfers was also on the basis of a simple phylogeny (Liu et al. 2022c: Fig. 1), in which these two species also clustered together with species of Brevicellicium but occupied a separated position. By sampling the most comprehensive range of taxa in Trechisporales available to date, the phylogeny in Liu et al. (2022a) clarified these two species placing them outside of both Trechispora and Brevicellicium, and in a new genus, Allotrechispora.

Besides the examples of Heteroradulum, Trechispora, and Brevicellicium having been dealt with (Li et al. 2022b; Liu et al. 2022a, b), similar incorrect placements also exist in other genera. In the current study, we focus on the genus Hyphodermella, in which two recently collected specimens from tropical Asia are identified.

Hyphodermella was erected as a monotypic genus for H. corrugata (Eriksson and Ryvarden 1976). Besides the generic type, another eight species are accepted in this genus within Phanerochaetaceae (Gilbertson et al. 2001; Melo and Hjortstam 2003; Nakasone 2008; Duhem 2010; Telleria et al. 2010; Duhem and Buyck 2011; Zhao et al. 2017; Wang and Zhao 2020; Wang et al. 2021a). Within Hyphodermella, the generic placement of H. poroides is questionable. Hyphodermella poroides was described according to a simple phylogeny that placed this species in a basal position within a clade also comprising H. corrugata and H. rosae (Zhao et al. 2017). Besides the uncertain phylogenetic position, the poroid hymenophoral surface also makes H. poroides distinguished from other species of Hyphodermella (Zhao et al. 2017). Although macrofungal species producing various hymenophoral configurations commonly can be placed in the same genus (Wang et al. 2021b; Li et al. 2022b; Liu et al. 2022a), in this case it is obvious, not as stated in the Abstract by the authors: “Both morphological and molecular evidences confirmed the placement of the new species in Hyphodermella.” (Zhao et al. 2017). Chen et al. (2021) recently performed a much more comprehensive phylogenetic analysis than that of Zhao et al. (2017) which clearly revealed the separation of H. poroides from Hyphodermella (Chen et al. 2021: Fig. 3), but they did not make any taxonomic change possibly due to their focusing mainly on other taxonomic issues. Around the publication time of Chen et al. (2021), another two new species, viz. H. aurantiaca and H. zixishanensis were separately described in Hyphodermella by the same research group, although the related phylogenies never confirmed their close relationship with Hyphodermella (Wang and Zhao 2020; Wang et al. 2021a).

When examining our specimens of Hyphodermella, we also explored the phylogenetic relationship of this genus via the most comprehensive sampling available to date. Accordingly, one genus, one species and three combinations are newly proposed. Beyond the taxonomic issue of Hyphodermella, we also aim to provide a standard to better phylogenies in future taxonomic studies.

MATERIALS and METHODS

Morphological examination

The studied specimens are preserved at the Fungarium, Institute of Microbiology, Chinese Academy of Sciences (HMAS), Beijing, China. The hymenophoral surfaces of basidiomes were examined with a Leica M125 stereomicroscope (Wetzlar, Germany) at a magnification of up to 100 × . The microscopic characters were observed with an Olympus BX43 light microscope (Tokyo, Japan) at magnifications up to 1000 × . The microscopic procedure followed Yu et al. (2021). Basidiome sections were prepared with Cotton Blue (CB), Melzer’s reagent, and 5% potassium hydroxide (KOH). All measurements were made from sections in CB. When presenting the variation of basidiospore sizes, 5% of the measurements were excluded from each end of the range and are given in parentheses. Drawings were made with the aid of a drawing tube. In the morphological description, L = mean basidiospore length (arithmetic average of all measured basidiospores), W = mean basidiospore width (arithmetic average of all measured basidiospores), Q = variation in the L/W ratios between the studied specimens, and (a/b) = the number of measurements (a) from a given number (b) of specimens.

Molecular sequencing

A small piece of basidiome was taken for DNA extraction using a CTAB rapid plant genome extraction kit-DN14 (Aidlab Biotechnologies, Beijing). Then, the crude DNA was used as templates for PCR amplifications of ITS, nLSU, rpb1, rpb2 and tef1α regions with the primer pairs ITS5/ITS4 (White et al. 1990), LROR/LR7 (Gardes and Bruns 1993), RPB1-Af/RPB1-Cr (Matheny et al. 2002), RPB2-f5F/RPB2-b7.1R (Liu et al. 1999; Matheny 2005) and 983F/1567R (Rehner and Buckley 2005), respectively. The PCR procedure was as follows: initial denaturation at 95 °C for 3 min, followed by 34 cycles at 94 °C for 40 s, 57.2 °C for 45 s and 72 °C for 1 min, and a final extension at 72 °C for 10 min for ITS and tef1α regions; initial denaturation at 94 °C for 1 min, followed by 35 cycles at 94 °C for 30 s, 48 °C 1 min and 72 °C for 1.5 min, and a final extension of 72 °C for 10 min for nLSU region; initial denaturation at 94 °C for 2 min, followed by 10 cycles at 94 °C for 45 s, 60 °C for 45 s (minus 1 °C per cycle) and 72 °C for 1.5 min, then followed by 36 cycles at 94 °C for 45 s, 53 °C for 1 min and 72 °C for 1.5 min, and a final extension of 72 °C for 10 min for rpb1 and rpb2 regions. The PCR products were sequenced with the same primers as those used in PCR amplification at the Beijing Genomics Institute, Beijing, China. All newly generated sequences were deposited in GenBank (https://www.ncbi.nlm.nih.gov/genbank/; Table 1).

Table 1 Species and sequences used in phylogenetic analyses

Phylogenetic analyses

Besides the newly generated sequences, additional molecular sequences were downloaded from GenBank for the phylogenetic analysis (Table 1). Two datasets were assembled to explore the phylogenetic position of our specimens in Hyphodermella and, more importantly, the phylogenetic relationship among Hyphodermella and related genera within Phanerochaetaceae. For the dataset of the combined ITS and nLSU regions, genera represented mostly by generic types in Phanerochaetaceae as well as Irpicaceae and Meruliaceae were comprehensively sampled as ingroup taxa. Hyphoderma litschaueri, H. mutatum and Candelabrochaete africana were selected as outgroup taxa (Chen et al. 2021). For the dataset of combined ITS, nLSU, rpb1, rpb2 and tef1α regions, genera phylogenetically close to our specimens were further sampled as ingroup taxa and Gelatinofungus brunneus was selected as the outgroup taxon according to the topology resulting from the previous two-locus dataset. ITS, nLSU, rpb1, rpb2 and tef1α regions were separately aligned using MAFFT 7.110 (Katoh and Standley 2013) under the G-INS-i option (Katoh et al. 2005), and the ambiguous regions of the alignments were trimmed using trimAl v1.2 under default parameters (Capella-Gutiérrez et al. 2009). Firstly, the resulting alignments for each locus were separately subjected to phylogenetic analyses, and no conflict in main lineages of our targeted taxonomic groups was observed from each other (data not shown). Then, the resulting alignments were concatenated as two alignments corresponding to the two datasets (Additional file 1: Alignment S1, Additional file 2: Alignment S2). The ITS region in these two alignments were further divided into ITS1, 5.8S and ITS2 subregions using ITSx 1.1.2 (Bengtsson-Palme et al. 2013) for separate model selection of phylogenetic analyses.

The maximum likelihood (ML) algorithm was performed using IQ-tree v2.1.2 (Minh et al. 2020), which implements automatic substitution model selection for each locus in ModelFinder (Kalyaanamoorthy et al. 2017) assessing nodal support determined by ultrafast bootstrapping (BS) with 10,000 replicates. The Bayesian inference (BI) algorithm was performed using MrBayes 3.2 (Ronquist et al. 2012). jModelTest 2 was used to estimate the best-fit evolutionary models of all loci separately for the BI algorithm under the corrected Akaike information criterion (Guindon and Gascuel 2003; Posada 2008). A discrete gamma distribution was used to model evolutionary rate differences among sites (four categories, + G). In the BI algorithm, two independent runs, each with four chains of one million generations and starting from random trees, were employed; trees were sampled every 1000th generation, of which the first 25% were removed as burn-in and the other 75% were retained for constructing a 50% majority consensus tree and calculating Bayesian posterior probabilities (BPPs). Tracer 1.5 (http://tree.bio.ed.ac.uk/software/tracer/) was used to judge whether chains converged.

RESULTS

Seven new sequences were generated from our specimens for this study (Table 1). The concatenated alignment of ITS and nLSU regions included 1643 characters with 311 parsimony-informative ones from 87 collections representing 68 species. For the ML algorithm, the best-fit partitioned models were determined as TVM + F + I + I + R4 for ITS1, GTR + F + I + I + R3 for both 5.8S and nLSU, and GTR + F + R4 for ITS2. For the BI algorithm, K80 + G, JC, JC and GTR + I + G were estimated as the best-fit partitioned models for the partitions of ITS1, 5.8S, ITS2 and nLSU, respectively. All chains in BI converged after ten million generations, which is indicated by the effective sample sizes of all parameters above 200 and the potential scale reduction factors close to 1.000. ML and BI algorithms construct similar topologies that differed only at several poorly supported nodes. The topology resulted from the ML algorithm is shown along with BS values more than 50% and BPPs more than 0.8 at the nodes (Fig. 2). In this phylogeny, the sampled species of Hyphodermella are separated in three lineages within the Donkia clade of Phanerochaetaceae (Fig. 2). The core lineage comprises the generic type Hyphodermella corrugata, H. pallidostraminea and H. rosae (BS = 92%, BPP = 0.99). In addition, two newly sequenced specimens, viz. LWZ 20190613-54 from Guangdong, China and LWZ 20191208-13 from Malaysia fall within the core lineage of Hyphodermella, and are separated from other species in this lineage. Hyphodermella aurantiaca and H. zixishanensis grouped together with two species of Roseograndinia (BS = 99%, BPP = 0.95). Hyphodermella poroides forms an independent lineage from other genera and species (BS = 100%, BPP = 1).

Fig. 2

Phylogenetic relationships among Hyphodermella and related genera inferred from ITS and nLSU regions. The topology was generated from the maximum likelihood algorithm, and bootstrap values and Bayesian posterior probabilities simultaneously above 50% and 0.8, respectively, are presented at the nodes. Phanerochaetaceae is indicated by the background in blue color, and the three genera related to Hyphodermella in darker blue color. The generic type species are indicated by the blue character T at the end of tip labels

The concatenated alignment of ITS, nLSU, rpb1, rpb2 and tef1α regions included 4550 characters with 882 parsimony-informative ones from 22 collections representing 18 species. For the ML algorithm, the best-fit partitioned models were determined as TPM2u + F + I + I + R2 for ITS1, TN + F + R2 for both 5.8S and nLSU, TIM + F + I + I + R2 for ITS2, GTR + F + I + I + R3 for both rpb1 and rpb2, and TIM2 + F + I + I + R2 for tef1α. For the BI algorithm, SYM + G, K80 and HKY + I + G were estimated as the best-fit partitioned models for the partitions of ITS1, 5.8S and ITS2, respectively, and GTR + I + G for all of nLSU, rpb1, rpb2 and tef1α. All chains in BI converged after one million generations, which is indicated by the effective sample sizes of all parameters above 200 and the potential scale reduction factors close to 1.000. ML and BI algorithms construct similar topologies that differed only at several poorly supported nodes. The topology resulted from the ML algorithm is shown along with BS values more than 50% and BPPs more than 0.8 at the nodes (Fig. 3). Like the phylogeny inferred from the dataset of combined ITS and nLSU regions (Fig. 2), this five-locus based phylogeny also recovered the sampled species of Hyphodermella in three independent lineages and the distinct position of the two newly sequenced specimens within the core lineage (Fig. 3).

Fig. 3

Phylogenetic relationships among Hyphodermella and related genera inferred from ITS, nLSU, rpb1, rpb2 and tef1α regions. The topology was generated from the maximum likelihood algorithm, and bootstrap values and Bayesian posterior probabilities simultaneously above 50% and 0.8, respectively, are presented at the nodes. Pseudohyphodermella is indicated by the background in blue color. The generic type species are indicated by the blue character T at the end of tip labels

In association with morphological characters, the two newly sequenced specimens are described as a new species of Hyphodermella, a new genus is erected for H. poroides, and H. aurantiaca and H. zixishanensis are transferred to Roseograndinia.

TAXONOMY

Hyphodermella suiae Shan Shen, S.L. Liu & L.W. Zhou, sp. nov. (Figs. 4, 5)

Fig. 4

Basidiomes of Hyphodermella suiae (a–d) in general and detailed views. a, b LWZ 20190613-54 (holotype); c, d LWZ 20191208-13 (paratype). Bars: a, c = 1 cm; b, d = 2 mm

Fig. 5

Microscopic structures of Hyphodermella suiae (drawn from LWZ 20190613-54, holotype). a Basidiospores. b Basidia. c Cystidioid hyphal ends. d Basidioles. e A vertical section through basidiomes. Bars: 10 μm

MycoBank: MB 848641

Etymology: suiae (Lat.), in memory of the Chinese mycologist Hong-Yan Su (苏 鸿雁), who was a professor in Dali University and kindly helped the corresponding author in many ways; she passed away on 3 May 2022 during the preparation of the current paper at the age of 55 years.

Diagnosis: Distinguished from other species of Hyphodermella by the small basidiospores.

Type: China: Guangdong: Ruyuan County, Nanling National Forest Park, on fallen angiosperm twig, 13 Jun 2019, Li-Wei Zhou, LWZ 20190613-54 (HMAS 287394—holotype).

Description: Basidiomes annual, resupinate, adnate, adherent, without odor or taste, leathery when fresh, up to 2.5 cm wide, 15 cm long and 100–150 μm thick. Hymenophoral surfaces smooth to tuberculate, shaped with the substrate shape partly, white to pale buff when fresh, becoming darker buff pale and cracking when drying. Margin distinct, white.

Hyphal system monomitic; generative hyphae with simple septa, thin-walled, 2.5–4 μm (n = 40/2) diam, branched, acyanophilous, inamyloid, indextrinoid, interwoven in subhymenium, more or less regularly arranged in subiculum; tissue unchanged in KOH. Basidia clavate, with four sterigmata and a basal simple septum, 20–25 × 5.5–6.5 μm (n = 40/2); basidioles dominant, in shape similar to basidia, but slightly smaller. Cystidia and cystidioles absent; cystidioid hyphal ends occasionally present, narrow clavate, thin-walled. Crystals present among hyphae, rhomboidal. Basidiospores ellipsoid, hyaline, thin-walled, smooth, inamyloid, indextrinoid, acyanophilous, (4.1–)4.2–5.2(–5.3) × 3.1–3.9(–4) µm, L = 4.81 μm, W = 3.42 μm, Q = 1.39–1.44 (n = 60/2).

Additional specimen examined: Malaysia: Kuala Lumpur: KL Forest Eco park, on fallen angiosperm twig, 8 Dec 2019, Li-Wei Zhou, LWZ 20191208-13 (HMAS 287395).

Notes: Hyphodermella suiae is similar to H. brunneocontexta in the smooth to tuberculate hymenophoral surface and the size of basidiospores. However, the hyphae of H. brunneocontexta in subiculum are thick-walled and brown (Duhem and Buyck 2011), while H. suiae has thin-walled, hyaline hyphae. In addition, H. suiae differs in having smaller basidiospores than the three species of Hyphodermella sampled in the current phylogenetic analysis, viz. H. corrugate (7–10 × 4–6 μm, Eriksson and Ryvarden 1976), H. pallidostraminea (5.4–6.6 × 3–3.5 μm, Crous et al. 2021), and H. rosae (6–8 × 4.3–5 μm, Nakasone 2008).

Pseudohyphodermella Shan Shen, S.L. Liu & L.W. Zhou, gen. nov.

MycoBank: MB 848651

Etymology: Pseudohyphodermella (Lat.), referring to the incorrect placement of the generic type in Hyphodermella.

Diagnosis: Distinguished from other genera in Phanerochaetaceae by the annual, resupinate basidiomes, a poroid hymenophore configuration, tissues unchanged in KOH, absence of cystidia, and broadly ellipsoid basidiospores.

Type: Pseudohyphodermella poroides (Y.C. Dai & C.L. Zhao) Shan Shen et al. 2023.

Description: Basidiomes annual, resupinate, effused. Hymenophoral surface poroid, cream to orange. Hyphal system monomitic; generative hyphae with simple septa, hyaline, thin-walled, wider in subiculum than in trama. Cystidia absent. Basidia clavate, hyaline, thin-walled, with four sterigmata and a basal simple septum. Basidiospores broadly ellipsoid, hyaline, thin-walled, smooth, inamyloid, indextrinoid, acyanophilous.

Notes: Within the Donkia clade of Phanerochaetaceae, the poroid hymenophoral surface makes Pseudohyphodermella and Geliporus distinct from other genera. Moreover, the tissues of Pseudohyphodermella do not change in KOH and the basidiospores are broadly ellipsoid (Zhao et al. 2017), while Geliporus has tissues that darken in KOH and cylindric to oblong-ellipsoid basidiospores (Yuan et al. 2017). In addition, Phanerina and Riopa fall outside the Donkia clade but within Phanerochaetaceae but also resemble Pseudohyphodermella in having resupinate basidiomes with a poroid hymenophoral surface; however, these two genera differ in the presence of cystidia and curved cylindrical to narrow ellipsoid basidiospores (Miettinen et al. 2016).

Pseudohyphodermella poroides (Y.C. Dai & C.L. Zhao) Shan Shen, S.L. Liu & L.W. Zhou, comb. nov.

MycoBank: MB 848652

Basionym: Hyphodermella poroides Y.C. Dai & C.L. Zhao, Mycoscience 58: 454 (2017).

Notes: Pseudohyphodermella poroides was originally described in Hyphodermella with a simple phylogeny as reference (Zhao et al. 2017). Although this species shares some morphological characters with Hyphodermella, such as a monomitic hyphal system with simple-septate generative hyphae and absence of cystidia, its poroid hymenophoral surface makes it distinguished from other species of Hyphodermella. Chen et al. (2021) first revealed the separation of H. poroides from Hyphodermella from a phylogenetic perspective. The current phylogeny (Fig. 2) further confirms the independence of H. poroides from all known genera and species. Therefore, a new genus Pseudohyphodermella is erected for this species, and H. poroides is accordingly transferred as P. poroides.

Roseograndinia aurantiaca (C.L. Zhao) Shan Shen, S.L. Liu & L.W. Zhou, comb. nov.

MycoBank: MB 848653

Basionym: Hyphodermella aurantiaca C.L. Zhao, Ann. bot. fenn. 58: 65 (2020).

Notes: Hyphodermella aurantiaca was recently described as a new species; however, the original simple phylogenies inferred from the nLSU region and a combination of ITS and nLSU regions did not provide reliable statistical support for the taxonomic position of this species in Hyphodermella (Wang and Zhao 2020). With our more comprehensive sampling, the current phylogeny strongly supports H. aurantiaca being separated from H. corrugata the type species of Hyphodermella and grouping together with species of Roseograndinia (BS = 99%, BPP = 0.95; Fig. 2). Morphologically, the combination of rose-colored basidiomes with a smooth to tuberculate hymenophoral surface, absence of cystidia and ellipsoid basidiospores makes H. aurantiaca consistent with the concept of Roseograndinia sensu Chen et al. (2021). Accordingly, H. aurantiaca is transferred as Roseograndinia aurantiaca.

Roseograndinia zixishanensis (C.L. Zhao) Shan Shen, S.L. Liu & L.W. Zhou, comb. nov.

MycoBank: MB 848654

Basionym: Hyphodermella zixishanensis C.L. Zhao, Nordic Jl Bot. 38(8): e03329, 4 (2021).

Notes: Hyphodermella zixishanensis was described as a new species in Hyphodermella (Wang et al. 2021a, b) soon after the publication of H. aurantiaca (Wang and Zhao 2020). Like Wang and Zhao (2020), the simple phylogenies in Wang et al. (2021ab) also did not reliably support the taxonomic position of this species in Hyphodermella. Instead, H. zixishanensis and H. aurantiaca formed a strongly supported clade (Wang et al. 2021a, b). The current phylogeny with a more comprehensive sampling strongly supports a close phylogenetic relationship between these two species and Roseograndinia (Fig. 2). Morphologically, H. zixishanensis is characterized by reddish, ceraceous basidiomes with a tuberculate hymenophoral surface and the absence of cystidia, which fits the concept of Roseograndinia sensu Chen et al. (2021). Therefore, H. zixishanensis is transferred as Roseograndinia zixishanensis.

A key to all eight known species in Hyphodermella

1 Basidiospores > 8 μm in length	2
Basidiospores < 8 μm in length	3
2 (1) Hymenophore surface orange to yellow orange; basidia > 35 μm in length	H. corrugata
Hymenophore surface ochraceous; basidia < 35 μm in length	H. ochracea
3 (1) Cystidia present	H. maunakeaensis
Cystidia absent	4
4 (3) Hymenophore surface odontioid	5
Hymenophore surface smooth to tuberculate	6
5 (4) Basidia suburniform to cylindric, 18–25 × 5–6.5 μm	H. densa
Basidia more or less clavate, 24–35 × 6–8 μm	H. rosae
6 (4) Hyphae thin-walled	H. suiae
Hyphae thick-walled, especially in subiculum	7
7 (6) Basidiomes pale yellowish; generative hyphae hyaline	H. pallidostraminea
Basidiomes pale grey or olive or brown; generative hyphae brown to dark brown in subiculum	H. brunneocontexta

A key to all five known species in Roseograndinia

1 Hymenophoral surface grandinioid to odontioid	2
Hymenophoral surface smooth to tuberculate	3
2 (1) Basidiospores < 3.1 μm in width, each with 1–2 oil drops	R. jilinensis
Basidiospores > 3.1 μm in width, without oil drops	R. rosea
3 (1) Basidia > 20 μm in length	R. zixishanensis
Basidia < 20 μm in length	4
4 (3) Basidiomes to 130 μm thick; basidiospores > 4 μm in length	R. minispora
Basidiomes 300–500 μm thick; basidiospores < 4 μm in length	R. aurantiaca

DISCUSSION

With the most comprehensive sampling to date in the current phylogenetic analyses (Figs. 2, 3), our specimens were identified as definitely belonging in Hyphodermella, being described as a new species, H. suiae. Although the current two-locus based phylogeny (Fig. 2) and the previous phylogenies related to Hyphodermella (Zhao et al. 2017; Wang and Zhao 2020; Chen et al. 2021; Wang et al. 2021a) are all inferred from the ITS and nLSU regions, the relationship at the generic level will be more accurate with sampling more comprehensive taxa in phylogenetic analyses (Fig. 2; Chen et al. 2021: Fig. 3). Furthermore, the five-locus based phylogenetic analysis we performed and the resulting phylogeny (Fig. 3) further confirmed the accuracy of phylogenetic relationships among sampled species of Hyphodermella inferred from the ITS and nLSU regions. Accordingly, Hyphodermella aurantiaca, H. poroides and H. zixishanensis are all excluded from Hyphodermella.

Hyphodermella poroides, occupying an independent phylogenetic position (Figs. 2, 3), is placed in a newly introduced monotypic genus Pseudohyphodermella. This new genus forms a weakly supported clade with Geliporus and Odontoefibula in the two-locus based phylogeny (BS = 85%, BPP = 0.89; Fig. 2), and has no close relationship with these two genera or any other genera in the five-locus based phylogeny (Fig. 3). Therefore, the alternative options of generic delimitation instead of erecting the new monotypic genus as suggested by Vellinga et al. (2015) cannot be supported according to the current phylogenies.

Roseograndinia was erected as a monotypic genus for R. rosea (Hjortstam and Ryvarden 2005). Due to a lack of molecular sequences from the type species of the genus, R. rosea, the phylogenetic independence of this genus in Phanerochaetaceae was recovered by two morphologically similar species R. jilinensis and R. minispora (Chen et al. 2021) and we follow the taxonomic proposal by Chen et al. (2021). The current phylogenies (Figs. 2, 3) strongly support the clade comprising H. aurantiaca, H. zixishanensis, R. jilinensis, and R. minispora. Moreover, morphologically H. aurantiaca and H. zixishanensis also fit well with the concept of Roseograndinia sensu Chen et al. (2021). Therefore, H. aurantiaca and H. zixishanensis are transferred as R. aurantiaca and R. zixishanensis here.

We note that in the current five-locus based phylogenetic analysis, only ITS and nLSU regions are used for the Pseudohyphodermella lineage. That is because additional gene regions were not published when P. poroides was originally described (Zhao et al. 2017), and moreover, the type specimens are also unavailable for molecular sequencing as they appear to be missing from the collections of the Institute of Microbiology, Beijing Forestry University, where the types were originally deposited. Even then, according to the separation of Hyphodermella and Roseograndinia in both the two-locus and five-locus based phylogenies (Figs. 2, 3), and the separation of Pseudohyphodermella from Hyphodermella and Roseograndinia in the two-locus based phylogeny (Fig. 2), it is reasonable to postulate that Pseudohyphodermella is a bona fide distinct lineage from others. Taking previous phylogenies of Hyphodermella (Zhao et al. 2017; Wang and Zhao 2020; Chen et al. 2021; Wang et al. 2021a) into consideration together, our study indicates that the ITS and nLSU regions are enough to delimit generic circumscriptions if the related genera are comprehensively sampled in phylogenetic analyses. Namely, sampling more taxa prior to employing more genes is more crucial to explore phylogenetic relationships among genera, at least those related to Hyphodermella. Normally, it is better to sample all known genera in a certain family, but we recognize that sometimes this is quite difficult, if possible, when the targeted genera belong to a phylogenetically not well-resolved family. So, we suggest comprehensively sampling at least closely related genera with targeted genera in taxonomic studies in these fungi.

CONCLUSION

In conclusion, species originally belonging to Hyphodermella are placed in three genera, including Hyphodermella, a new genus Pseudohyphodermella, and Roseograndinia, and H. suiae is described as a new species. Beyond resolving the taxonomy of Hyphodermella itself, this study further clarified that simple phylogenies cannot always accurately place species in appropriate genera. This is an obvious but sometimes omitted phylogenetic practice in recent years (Guan et al. 2020; Zong et al. 2021; Li et al. 2022a; Liu et al. 2022c). We suggest that all fungal taxonomists especially beginners should keep in mind to sample as many comprehensive taxa as possible in phylogenetic, and for that matter morphological analyses (Hawksworth 2020).