Hydnophanerochaete and Odontoefibula, two new genera of phanerochaetoid fungi (Polyporales, Basidiomycota) from East Asia

Abstract Two new genera with phylogenetic affinities to Phanerochaete s.l. are presented, namely Hydnophanerochaete and Odontoefibula. The generic type of Hydnophanerochaete is Phanerochaeteodontoidea. Odontoefibula is established based on a new species: O.orientalis (generic type). Both genera have effused basidiocarps with odontioid hymenial surface, simple-septate generative hyphae, cystidia lacking, clavate basidia and ellipsoid basidiospores that are smooth, thin-walled, inamyloid, non-dextrinoid and acyanophilous. Hydnophanerochaete is additionally characterised by a compact texture in the subiculum with thick-walled generative hyphae and quasi-binding hyphae. Odontoefibula has a dense texture of subiculum with thin- to slightly thick-walled hyphae and further a dark reddish reaction of basidiocarps when treated with KOH. Multi-marker phylogenetic analyses based on sequences, inferred from the ITS+nuc 28S+rpb1+rpb2+tef1 dataset, indicate that Hydnophanerochaete and Odontoefibula are placed in the Meruliaceae and Donkia clades of Phanerochaetaceae, respectively. Phanerochaetesubodontoidea is a synonym of P.odontoidea, according to morphological and molecular evidence.


Introduction
The genus Phanerochaete P. Karst., typified by P. alnea (Fr.) P. Karst., belongs to Polyporales Gäum of the Basidiomycota R.T. Moore and is one of the largest genera of corticoid fungi, including over 150 names according to Index Fungorum (http://www.indexfungorum.org/). Basidiocarps are typically membranaceous, effused, with various hymenial surfaces (i.e. smooth, tuberculate, odontioid, hydnoid, merulioid or poroid). Microscopically, Phanerochaete has a monomitic hyphal system, ordinarily simple-septate generative hyphae (rare clamp connections can be found in the subiculum), ellipsoid to cylindrical thin-walled basidiospores and clavate basidia. Phanerochaete is widespread and grows on diverse woody substrates (i.e. twigs and branches or trunks of angiosperms or gymnosperms), causing a white rot. Phanerochaete s.l. has attracted increasing study interest due to its abundant taxonomic diversity and potential applications in the field of biodegradation and bioconversion (Sánchez 2009).
Phanerochaete odontoidea Sheng H. Wu and P. subodontoidea Sheng H. Wu were described from Taiwan (Wu 2000). Both species have ceraceous basidiocarps with odontioid to hydnoid hymenial surface, compact subiculum, but no cystidia. These species have been shown to be phylogenetically far from the core Phanerochaete clade Ghobad-Nejhad et al. 2015;Wu et al. 2018) and were placed by Justo et al. (2017) in Meruliaceae. In this study, we evaluate the generic placement of P. odontoidea and P. subodontoidea, as well as morphologically similar species. To accommodate our target taxa, we found it necessary to introduce two new genera placed within Meruliaceae and Phanerochaetaceae, respectively.
When Phanerochaete odontoidea and P. subodontoidea were described, they were separated by basidiospore width (Wu 2000). After 2000, we have accumulated more collections identified as P. odontoidea and P. subodontoidea from China, Japan, Taiwan and Vietnam. To better reflect their morphological variations, this study provides updated morphological and molecular evidence for revising their species concepts.

Morphological studies
The specimens used for illustrations and descriptions are deposited at the herbarium of National Museum of Natural Science of ROC (TNM, acronym according to Index Herbariorum; http://sweetgum.nybg.org/science/ih/). Free-hand thin sections of basidiocarps were mounted in three mounting media for microscopic studies: 5% (w/v) KOH with 1% (w/v) phloxine was used for observation and measurements; Melzer's reagent (IKI) was utilised to check amyloidity and dextrinoidity; and Cotton Blue (CB, Fluka 61335) was employed to determine cyanophily. Sections were studied with a Leica DM2500 (Leica, Wetzlar) microscope. Drawings were done with the aid of a drawing tube. We followed the method for measurements of microscopic characters by Wu (1990). The abbreviations below were used when presenting statistical measurements of basidiospores: L = mean basidiospore length, W = mean basidiospore width, Q = variation in L/W ratio, n = number of measured spores. The terminology of microscopic characters follows Wu (1990).

DNA extraction and sequencing
Dried specimens or mycelia grown on MEA were used for isolating genomic DNA.

Phylogenetic analyses
Two datasets were compiled for phylogenetic analyses: the ITS+nuc 28S+rpb1+rpb2+tef1 dataset was analysed to confirm the generic placement of target species within the phlebioid clade of Polyporales. The ITS dataset was used to get better resolutions on species level within the Hydnophanerochaete clade of Meruliaceae. The selection of strains and species for the 5-marker dataset was based on Binder et al. (2013), Flou-das and Hibbett (2015), Kuuskeri et al. (2015), Justo et al. (2017), Miettinen et al. (2016), Moreno et al. (2017), Papp and Dima (2017), Yuan et al. (2017) and Zhao et al. (2017). Alignment was done with MAFFT v. 7 using two strategies: Q-INS-I for ITS and FFT-NS-I for nuc 28S, rpb1, rpb2 and tef1 (Katoh and Standley 2013). The resulting alignments were manually adjusted in Mega 7 (Kumar et al. 2016 (Ronquist et al. 2012). The Maximum Likelihood (ML) method was carried out for the 5-marker dataset using RAxML BlackBox (Stamatakis 2014). For the BI analyses, jModeltest 2.1.10 (Darriba et al. 2012) was first used to estimate separate models for each of the markers in both datasets, based on Akaike information criterion (AIC). The Markov chain Monte Carlo (MCMC) search was run for ten million generations, with four chains and trees sampled every 100 generations. The first twenty-five percent of trees were discarded as burn-in while the remaining trees were used to construct the fifty percent majority-rule consensus phylogram with posterior probabilities (PP). For the ML analysis, the best-scoring tree with proportional values of bootstrap (BS) was computed under a GTRGAMMA model with one thousand bootstrap replicates, followed by a thorough ML search. Gaps were treated as missing data. Branches were regarded as having statistical support if values of PP and/or BS were equal to or over 0.9 and 70%, respectively. Both BI and ML analyses were performed at the CIPRES Science Gateway (Miller et al. 2010; http://www.phylo.org/). Phylograms were visualised and edited in TreeGraph 2 (Stöver and Müller 2010) and Adobe Illustrator (Adobe Systems, Inc).

Phylogeny results
The final ITS+nuc 28S+rpb1+rpb2+tef1 dataset consisted of 126 sequences and 7253 characters (of which 43.7% were parsimony-informative) including gaps and the ITS dataset comprised 12 sequences and 887 characters (of which 7.7% were parsimonyinformative) including gaps. In the BI analyses, since the GTR+G+I model was selected as the best model of nucleotide substitution for each of the five markers in the 5-marker dataset, it was used for the entire alignment with five partitions. The HKY+I+G model was selected as the best model of nucleotide substitution for the ITS dataset. The fifty percent majority-rule consensus phylogram with PP support values was reconstructed after the average standard deviation of split frequencies fell below 0.001. The best-scoring ML tree with BS support values was built. Phylogenetic trees of the 5-marker dataset, inferred from BI and ML algorithms, shared similar topologies and thus only the ML tree was shown (Fig. 1).  In the 5-marker analyses (Fig. 1), six main clades with high statistic supports (BS = 96-100%, PP = 1) could be recognised in the ingroup: the antrodia clade, the core polyporoid clade, the gelatoporia clade, the phlebioid clade, a residual clade and the skeletocutis-tyromyces clade. The phlebioid clade, which is the focus of this study, The tree inferred from the ITS dataset (Fig. 2) showed that sequences of holotype (CWN00776) and paratype (Wu 911206-38) of Phanerochaete subodontoidea were clustered with sequences of P. odontoidea within a monophyletic lineage (PP = 1).
Remarks. Hydnophanerochaete is morphologically similar to the genus Hydnophlebia (Telleria et al. 2017). Both genera have resupinate basidiocarps with odontioid to hydnoid hymenial surface, a monomitic hyphal system, ordinarily simple-septate hyphae and similar basidiospore shape. However, we note three distinguishing differences. First, Hydnophlebia has membranaceous basidiocarps usually with rhizomorphic margin, while Hydnophanerochaete has ceraceous basidiocarps with fairly determinate margin. Second, occasional single or multiple clamp connections are present in sub- icular or aculei hyphae of Hydnophlebia, whereas they are lacking in hyphae of Hydnophanerochaete. Third, Hydnophlebia occasionally bears tubular to ventricose leptocystidia, which are lacking in Hydnophanerochaete.
Little morphological differences exist between Hydnophanerochaete and Odontoefibula: both genera have monomitic hyphal system with simple-septate hyphae and are lacking cystidia. However, Hydnophanerochaete is distinguished from Odontoefibula by its basidiocarps without colour change in KOH; additionally, its subiculum is compact, not dense.
Remarks. Phanerochaete subodontoidea morphologically resembles Phanerochaete odontoidea, whereas they were distinguished merely based on the width of basidiospores [P. odontoidea: 2.6-3 μm vs. P. subodontoidea: 3-3.7 μm, Wu (2000)]. However, after carefully measuring the basidiospore size of available specimens of these two species, we found basidiospore ranges are highly overlapping (Table 2). Additionally, the ITS sequences of the holotype of P. subodontoidea (CWN 00776) is almost identical to the ITS sequences of the paratype of P. odontoidea . We failed to obtain sequences from the holotype of P. odontoidea (Wu 910807-11), but Wu 9310-8 was confirmed as conspecific with the holotype by morphological comparison. Thus, based on morphological and molecular evidence (Fig. 2), P. subodontoidea is treated as a synonym of P. odontoidea. A paratype specimen named P. odontoidea  probably belongs to the genus Flavodon Ryvarden based on preliminary BLAST results of nuc 28S sequences. However, this specimen was not included in this study. Etymology. From odonto (= tooth-like) + efibula (= without clamp connection), referring to the odontioid hymenial surface and simple-septate hyphae of the genus.
Description. Basidiocarps annual, resupinate, effused, adnate, membranaceous to ceraceous. Hymenial surface at first honey yellow, becoming ochraceous to pale brown with age, turning dark reddish in KOH, initially smooth to slightly tuberculate, becoming grandinioid to odontioid with age. Aculei conical to cylindrical, separate or fused, up to 0.3 mm long.
Habitat. On fallen trunk of angiosperm (e.g. Amygdalus). Distribution. Hitherto known from China (Beijing), Japan and Taiwan

Discussion
Our 5-marker phylogenetic analyses ( Fig. 1) provided an updated taxonomic framework for evaluating generic placements of the target taxa of the phlebioid clade. The tree topologies are consistent with previous results Floudas and Hibbett 2015;Justo et al. 2017;Papp and Dima 2017). Within the phlebioid clade, we recovered two monophyletic lineages of phanerochaetoid fungi (Fig. 1), which supports the status of the two genera erected here: Hydnophanerochaete, typified by P. odontoidea, is accommodated in Meruliaceae; Odontoefibula, typified by O. orientalis, is placed in Donkia clade of Phanerochaetaceae.