﻿Five new species of Schizoporaceae (Basidiomycota, Hymenochaetales) from East Asia

﻿Abstract Five new wood-inhabiting fungi, Lyomycesalbopulverulentus, L.yunnanensis, Xylodondaweishanensis, X.fissuratus, and X.puerensisspp. nov., are proposed based on a combination of morphological features and molecular evidence. Lyomycesalbopulverulentus is characterized by brittle basidiomata, pruinose hymenophore with a white hymenial surface, a monomitic hyphal system with clamped generative hyphae, and ellipsoid basidiospores. Lyomycesyunnanensis is characterized by a grandinioid hymenial surface, the presence of capitate cystidia, and ellipsoid basidiospores. Xylodondaweishanensis is characterized by an odontioid hymenial surface, a monomitic hyphal system with clamped generative hyphae, and broad ellipsoid-to-subglobose basidiospores. Xylodonfissuratus is characterized by a cracking basidiomata with a grandinioid hymenial surface, and ellipsoid basidiospores. Xylodonpuerensis is characterized by a poroid hymenophore with an angular or slightly daedaleoid configuration, and ellipsoid-to-broad-ellipsoid basidiospores. Sequences of ITS and nLSU rRNA markers of the studied samples were generated and phylogenetic analyses were performed with the maximum likelihood, maximum parsimony, and Bayesian inference methods. The phylogram based on the ITS+nLSU rDNA gene regions (Fig. 1) included six genera within the families Chaetoporellaceae, Hyphodontiaceae, Hymenochaetaceae, and Schizoporaceae (Hymenochaetales)—Fasciodontia, Hastodontia, Hyphodontia, Kneifiella, Lyomyces, and Xylodon—in which the five new species were grouped into genera Lyomyces and Xylodon. The phylogenetic tree inferred from the ITS sequences highlighted that Lyomycesalbopulverulentus formed a monophyletic lineage and was then grouped closely with L.bambusinus, L.orientalis, and L.sambuci; additionally, L.yunnanensis was sister to L.niveus with strong supports. The topology, based on the ITS sequences, revealed that Xylodondaweishanensis was retrieved as a sister to X.hyphodontinus; X.fissuratus was grouped with the four taxa X.montanus, X.subclavatus, X.wenshanensis, and X.xinpingensis; and X.puerensis was clustered with X.flaviporus, X.ovisporus, X.subflaviporus, X.subtropicus, and X.taiwanianus.

Classification of the kingdom of fungi has been updated continuously, based on the frequent inclusion of data from DNA sequences in many phylogenetic studies (Wijayawardene et al. 2020). Based on the early embrace of molecular systematics by mycologists, both the discovery and classification of fungi, among the more basal branches of the tree, are now coming to light from genomic analyses and environmental DNA surveys that have been conducted (James et al. 2020). Hyphodontia s.l. was indicated to be a polyphyletic group, and Xylodon and Kneiffiella P. Karst. included the most species (Yurchenko and Wu 2016;Riebesehl et al. 2019). Given the lack of sequences for a part of the fungal taxa, it is difficult to clearly separate the many genera in this group using molecular data; therefore, a broad concept of Hyphodontia s.l. was accepted (Yurchenko and Wu 2016;Wang and Chen 2017;Riebesehl et al. 2019). On the basis of the nuclear DNA sequence data, six well-distinguished clades-Lagarobasidium clade, Kneiffiella-Alutaceodontia clade, Hyphodontia clade, Hastodontia clade, Xylodon-Lyomyces, Rogersella clade, and Xylodon-Schizopora-Palifer clade-were included based on the phylogenetical studies for Hyphodontia s.l., in which the genera Xylodon and Lyomyces were nested within the Xylodon-Lyomyces-Rogersella clade and Xylodon-Schizopora-Palifer clade, respectively .  revealed that Hyphodontia s.l. was divided into six genera, viz., Hastodontia (Parmasto) Hjortstam & Ryvarden, Hyphodontia J. Erikss, Kneiffiella, Lagarobasidium Jülich, Lyomyces, and Xylodon, in which 35 new combinations were proposed, including fourteen Lyomyces species. Yurchenko et al. (2017) clarified the Lyomyces sambuci complex on the basis of the sequences of the internal transcribed spacer (ITS) and the nuclear large subunit (nLSU) ribosomal DNA gene, in which four new species were described. To confirm the taxonomic relationship of Xylodon species, Viner et al. (2018) proposed two genera, Lagarobasidium and Xylodon, which should be synonymous based on molecular data from the ITS and nLSU regions, and in which the three species were combined into Xylodon. Riebesehl et al. (2019) clarified the generic concept and their phylogenetic reconstruction of Lyomyces, in which L. sambuci was sister to L. crustosus (Pers.) P. Karst. Based on a combination of morphological and molecular evidence, the wood-inhabiting fungal diversity within the family Schizoporaceae of the order Hymenochaetales were analyzed, including Lyomyces fissuratus C.L. Zhao, L. fumosus C.L. Zhao, L. niveus C.L. L. ochraceoalbus C.L. Zhao. (Luo et al. 2021b, 2021c). Viner et al. (2022) described three new species from Africa as Xylodon angustisporus Viner & Ryvarden, X. dissiliens Viner & Ryvarden, and X. laxiusculus Viner & Ryvarden, based on the morphological descriptions and molecular analyses that they conducted. A phylogenetic and taxonomic study on Xylodon showed that three new species of Xylodon were from southern China; in addition, it was also found that it enriched the fungal diversity of these areas .
During investigations on the wood-inhabiting fungi in the Yunnan-Guizhou Plateau of China, samples representing five additional species belonging to genera Lyomyces and Xylodon were collected. To clarify the placement and relationships of the five species, we carried out a phylogenetic and taxonomic study on Lyomyces and Xylodon, based on the ITS and nLSU sequences.

Morphology
The studied specimens were deposited at the Herbarium of Southwest Forestry University (SWFC), Kunming, Yunnan Province, P.R. China. Macromorphological descriptions are based on field notes and photos captured in the field and lab. Color terminology follows Petersen (Petersen 1996). Micromorphological data were obtained from the dried specimens when observed under a light microscope following Dai (2012). The following abbreviations are used: KOH = 5% potassium hydroxide water solution, CB = Cotton Blue, CB-= acyanophilous, IKI = Melzer's Reagent, IKI-= both inamyloid and indextrinoid, L = mean spore length (arithmetic average for all spores), W = mean spore width (arithmetic average for all spores), Q = variation in the L/W ratios between the specimens studied and n = a/b (number of spores (a) measured from given number (b) of specimens).

Molecular phylogeny
The CTAB rapid plant genome extraction kit-DN14 (Aidlab Biotechnologies Co., Ltd, Beijing) was used to obtain genomic DNA from the dried specimens following the manufacturer's instructions (Zhao and Wu 2017). The nuclear ribosomal ITS region was amplified with the primers ITS5 and ITS4 (White et al. 1990). The nuclear ribosomal LSU gene was amplified with the primers LR0R and LR7 (Vilgalys and Hester 1990;Rehner and Samuels 1994). The PCR procedure for ITS and nLSU followed previous study (Zhao and Wu 2017). All newly-generated sequences were deposited in NCBI GenBank (Table 1).
The sequences were aligned in MAFFT version 7 (Katoh et al. 2019) using the G-INS-i strategy. The alignment was adjusted manually using AliView version 1.27 (Larsson 2014). Each dataset was aligned separately at first and then the ITS and nLSU regions were combined with Mesquite version 3.51. The combined dataset was deposited in TreeBASE (submission ID 29868). Sequences of Hymenochaete ochromarginata P.H.B. Talbot and H. rubiginosa (Dicks.) Lév. retrieved from GenBank were used as an outgroup in the ITS+nLSU analysis (Fig. 1); Sequences of Xylodon quercinus and X. ramicida Spirin & Miettinen retrieved from GenBank were used as an outgroup in the ITS analysis (Fig. 2); Lyomyces bambusinus C.L. Zhao and L. sambuci were selected as outgroup (Fig. 3) as inspired by a previous study (Luo et al. 2021c). Maximum parsimony analysis in PAUP* version 4.0a169 (http://phylosolutions. com/paup-test/) was applied to ITS and the combined ITS+nLSU datasets following a previous study (Zhao and Wu 2017). All characters were equally weighted and gaps were treated as missing data. Trees were inferred using the heuristic search option with TBR branch swapping and 1,000 random sequence additions. Max-trees were set to 5,000, branches of zero length were collapsed and all parsimonious trees were saved. Clade robustness was assessed using bootstrap (BT) analysis with 1,000 pseudo replicates (Felsenstein 1985). Descriptive tree statistics -tree length (TL), composite consistency index (CI), composite retention index (RI), composite rescaled consistency index (RC) and composite homoplasy index (HI) -were calculated for each maximum parsimonious tree generated. The combined dataset was also analysed using Maximum Likelihood (ML) in RAxML-HPC2 through the CIPRES Science Gateway (Miller et al. 2012). Branch support (BS) for the ML analysis was determined by 1000 bootstrap pseudoreplicates.
MrModeltest 2.3 (Nylander 2004) was used to determine the best-fit evolution model for each dataset for the purposes of Bayesian inference (BI). which was performed using MrBayes 3.2.7a with a GTR+I+G model of DNA substitution and a gamma distribution rate variation across sites (Ronquist et al. 2012). A total of four Markov chains were run for two runs from random starting trees for 8 million generations for ITS+nLSU (Fig. 1); 0.5 million generations for ITS (Fig. 2) and 9.5 million generations for ITS (Fig. 3) with trees and parameters sampled every 1,000 generations. The first quarter of all of the generations were discarded as burn-ins. A majority rule consensus tree was computed from the remaining trees. Branches were considered as significantly supported if they received a maximum likelihood bootstrap support value (BS) of > 70%, a maximum parsimony bootstrap support value (BT) of > 70% or a Bayesian posterior probability (BPP) of > 0.95.

Molecular phylogeny
The ITS+nLSU dataset (Fig. 1) comprised sequences from 43 fungal specimens representing 31 taxa. The dataset had an aligned length of 2,100 characters, of which 1,323 characters were constant, 156 were variable and parsimony-uninformative and 621 (35%) were parsimony-informative. Maximum parsimony analysis yielded 1 equally parsimonious tree (TL = 2,867, CI = 0.4423, HI = 0.5577, RI = 0.6488 and RC = 0.2869). The best model of nucleotide evolution for the ITS+nLSU dataset estimated and applied in the Bayesian analysis was found to be GTR+I+G. Bayesian analysis and ML analysis resulted in a similar topology as in the MP analysis. The Bayesian analysis had an average standard deviation of split frequencies = 0.008603 (BI) and the effective sample size (ESS) across the two runs is double the average ESS (avg. ESS) = 1,623. The phylogram based on the ITS+nLSU rDNA gene regions (Fig. 1) include six genera within Schizoporaceae (Hymenochaetales), which are Fasciodontia, Hastodontia, Hyphodontia, Kneiffiella, Lyomyces, and Xylodon-in which five new species were grouped into the genera Lyomyces and Xylodon.
The ITS dataset (Fig. 2) comprised sequences from 47 fungal specimens representing 29 taxa. The dataset had an aligned length of 661 characters, of which 316 characters were constant, 53 were variable and parsimony-uninformative and 292 (35%) were parsimony-informative. Maximum parsimony analysis yielded 1 equally parsimonious tree (TL = 1,371, CI = 0.4136, HI = 0.5864, RI = 0.6984 and RC = 0.2888). The best model of nucleotide evolution for the ITS dataset estimated and applied in the Bayesian analysis was found to be GTR+I+G. Bayesian analysis and ML analysis resulted in a similar topology as in the MP analysis. The Bayesian analysis had an average standard deviation of split frequencies = 0.006564 (BI) and the effective sample size (ESS) across the two runs is double the average ESS (avg. ESS) = 359. The phylogenetic tree (Fig. 2), inferred from the ITS sequences, highlighted that L. albopulverulentus formed a monophyletic lineage. It was then grouped closely with L. bambusinus, L. orientalis Riebesehl, Yurch. & Langer, and L. sambuci. In addition, L. yunnanensis was found to be the sister to L. niveus with strong supports.
The ITS dataset (Fig. 3) comprised sequences from 72 fungal specimens representing 65 taxa. The dataset had an aligned length of 702 characters, of which 283 characters were constant, 96 were variable and parsimony-uninformative and 323 (35%) were parsimony-informative. Maximum parsimony analysis yielded 5,000 equally parsimonious trees (TL = 2,726, CI = 0.2748, HI = 0.7252, RI = 0.4280 and RC = 0.1176). The best model of nucleotide evolution for the ITS dataset estimated and applied in the Bayesian analysis was found to be GTR+I+G. Bayesian analysis and ML analysis resulted in a similar topology as in the MP analysis. The Bayesian analysis had an average standard deviation of split frequencies = 0.02518 (BI) and the effective sample size (ESS) across the two runs is double the average ESS (avg. ESS) = 1,440. The topology (Fig. 3), based on ITS sequences, revealed that X. daweishanensis was retrieved as a sister to X. hyphodontinus (Hjortstam & Ryvarden)    Etymology. Albopulverulentus (Lat.): referring to the white and pruinose hymenial surface.
Description. Basidiomata annual, resupinate, adnate, brittle, without odor or taste when fresh, up to 12 cm long, 1.5 cm wide, and 150 µm thick. Hymenial surface pruinose, white when fresh and drying. Sterile margin indistinct, white, and up to 2 mm wide.
Additional specimens examined ( Etymology. Fissuratus (Lat.): referring to the cracking hymenial surface. Description. Basidiomata annual, resupinate, adnate, coriaceous, without odor or taste when fresh, up to 12 cm long, 2.5 cm wide, and 150 µm thick. Hymenial surface grandinioid, and white when fresh, white to slightly cream on drying, cracking. Sterile margin indistinct, white, and up to 1 mm wide.