Morphological and molecular identification of four new resupinate species of Lyomyces (Hymenochaetales) from southern China

Abstract Four new wood-inhabiting fungal species, Lyomyces bambusinus, L. cremeus, L. macrosporus and L. wuliangshanensis, are proposed based on a combination of morphological and molecular evidence. Lyomyces bambusinus is characterized by resupinate basidiomata with colliculose to tuberculate hymenial surface and broadly ellipsoid, hyaline, slightly thick-walled, smooth basidiospores. Lyomyces cremeus is characterised by resupinate basidiomata with smooth, cream hymenial surface and ellipsoid, hyaline, thin-walled to slightly thick-walled basidiospores. Lyomyces macrosporus is characterized by pruinose basidiomata with reticulate hymenial surface, presence of three kinds of cystidia and larger basidiospores (6.7–8.9 × 4.4–5.4 µm). Lyomyces wuliangshanensis is characterized by coriaceous basidiomata and ellipsoid, hyaline, slightly thick-walled, smooth basidiospores. The phylogenetic analyses based on molecular data of the internal transcribed spacer (ITS) region sequences revealed that the four new species belonged to Lyomyces. Lyomyces bambusinus grouped with L. sambuci. Lyomyces cremeus clade was sister to a clade comprised of L. microfasciculatus. Lyomyces macrosporus was sister to L. allantosporus. Lyomyces wuliangshanensis was closely related to L. mascarensis.

Molecular studies on Lyomyces and related genera have been carried out recently Yurchenko et al. 2017;Viner et al. 2018;Riebesehl et al. 2019). Riebesehl and Langer (2017) indicated that Hyphodontia s.l. should be divided into several genera: Hastodontia (Parmasto) Hjortstam & Ryvarden, Hyphodontia J. Erikss, Kneiffiella (Pers.) Gray, Lagarobasidium Jülich, Lyomyces and Xylodon (Pers.) Gray and thus 35 new combinations were proposed, including fourteen Lyomyces species. The clarification of Lyomyces sambuci complex was conducted based on ITS and 28S sequences analyses and four new species of Lyomyces were described (Yurchenko et al. 2017). Viner et al. (2018) studied the taxonomy of Lagarobasidium and Xylodon, and showed that twelve species clustered into Lyomyces clade and then grouped with Xylodon clade. Phylogenetic and morphological studies on Xylodon showed that Xylodon was distinct from Hastodontia, Hyphodontia, Kneiffiella and Lyomyces and the Lyomyces generic species L. sambuci was sister to L. crustosus (Pers.) P. Karst. formed a single lineage with a high support (Riebesehl et al. 2019).
During investigations on wood-inhabiting fungi in southern China, four additional taxa were found, which could not be assigned to any described species in Lyomyces. In this study, the authors expand samplings from previous studies (Gafforov et al. 2017;Riebesehl and Langer 2017) to examine taxonomy and phylogeny of them within Lyomyces, based on the internal transcribed spacer (ITS) regions sequences.

Morphological studies
The specimens studied have been deposited in the herbarium of Southwest Forestry University (SWFC), Kunming, Yunnan Province, P.R. China. Special color terms fol-low Petersen (1996). Macromorphological descriptions are based on field notes. Micromorphological data were obtained from the dried specimens and observed under a light microscope following Dai (2010) and Cui et al. (2019). The following abbreviations are used: KOH = 5% potassium hydroxide; CB = cotton blue; CB+ = cyanophilous; IKI = Melzer's reagent; IKI-= non-amyloid and non-dextrinoid; L = mean spore length (arithmetic average of all spores); W = mean spore width (arithmetic average of all spores); Q = L/W ratio; n (a/b) = number of spores (a) measured from given number (b) of specimens.

DNA extraction and sequencing
CTAB rapid plant genome extraction kit-DN14 (Aidlab Biotechnologies Co., Ltd, Beijing) was used to obtain genomic DNA from dried specimens, according to the manufacturer's instructions (Han et al. 2016;Song and Cui 2017) . The ITS region was amplified with the primer pair ITS5 and ITS4 (White et al. 1990). The PCR cycling procedure for ITS was as follows: initial denaturation at 95 °C for 3 min, followed by 35 cycles at 94 °C for 40 s, 58 °C for 45 s and 72 °C for 1 min, and a final extension of 72 °C for 10 min followed Shen et al. (2019). The PCR products were purified and directly sequenced at Kunming Tsingke Biological Technology Limited Company, Yunnan Province, P.R.China. All newly generated sequences were deposited in GenBank (Table 1).
Maximum parsimony (MP), Maximum Likelihood (ML) and Bayesian Inference (BI) analyses were applied to the ITS dataset sequences. Approaches to phylogenetic analyses followed Wu et al. (2018) and Zhu et al. (2019) the tree construction procedure was performed in PAUP* version 4.0b10 (Swofford 2002). 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 1000 random sequence additions. Max-trees were set to 5000, branches of zero length were collapsed and all most-parsimonious trees were saved. Clade robustness was assessed using bootstrap (BT) analysis with 1000 replicates (Felsenstein 1985). Descriptive tree statistics tree length (TL), consistency index (CI), retention index (RI), rescaled consistency index (RC) and homoplasy index (HI) were calculated for each mostparsimonious tree generated.
Sequences were also analyzed using Maximum Likelihood (ML) ML analysis was conducted with RAxML-HPC2 through the Cipres Science Gateway (www.phylo.org; Miller et al. 2009). Branch support (BS) for ML analysis was determined by 1000 bootstrap replicates and evaluated under the gamma model. MrModeltest 2.3 (Nylander 2004) was used to determine the best-fit evolution model for the data set for Bayesian Inference (BI). Bayesian Inference was performed with MrBayes 3.1.2 with a general time reversible (GTR) model of DNA substitution and a gamma distribution rate variation across sites (Ronquist and Huelsenbeck 2003). Four Markov chains were used in each of 2 runs from random starting trees for 600,000 generations, with trees and parameters sampled every 100 generations. The first quarter of generations were discarded as burn-in. A majority rule consensus tree of all remaining trees and posterior probabilities was calculated. Branches that received bootstrap support for maximum likelihood (BS) ≥75%, maximum parsimony (BP) ≥75%, and Bayesian posterior probabilities (BPP) ≥0.95 were considered significantly supported.

Phylogeny results
The ITS dataset (Fig. 1) included sequences from 39 fungal specimens representing 18 species. The dataset had an aligned length of 608 characters, of which 277 characters were constant and 242 parsimony-informative. MP analysis yielded 8 equally parsimonious trees (TL = 978, CI = 0.523, HI = 0.478, RI = 0.738, RC = 0.385). The best-fit model for ITS alignment estimated and applied in the BI was GTR+I+G. At the end of the BI runs, the average standard deviation of split frequencies was 0.008676. The tree topology obtained by BI and ML was similar to the MP one.
Basidiomata. Annual, resupinate, subcoriaceous when fresh, becoming coriaceous upon drying, up to 15 cm long and 5 cm wide, 50-150 µm thick. Hymenial surface smooth to more or less tuberculate, white to cream when fresh, turning cream to buff upon drying. Margin narrow, concolorous with hymenial surface.
Hyphodontia s.l. is an extensively studied group of Hymenochaetales (Dai 2012;Viner et al. 2018;Riebesehl et al. 2019), but the Chinese species diversity is still not well known, especially in subtropical and tropical areas. The four new Lyomyces species here described are from the subtropics.