Sanghuangporus toxicodendri sp. nov. (Hymenochaetales, Basidiomycota) from China

Abstract Sanghuangporus toxicodendri (Hymenochaetales) is described as new based on collections made from Shennongjia Forestry District, Hubei Province, China. All studied basidiocarps grew on living trunks of Toxicodendron sp. This new species is characterized by having perennial, effused-reflexed to pileate basidiocarps; pore surface brownish yellow or yellowish brown, pores 7–9 per mm; context 1–5 mm thick or almost invisible; setae ventricose, dark brown, 26–42 × 7–10 μm; basidia 4-sterigmate or occasionally 2-sterigmate; basidiospores broadly ellipsoid, smooth, brownish yellow, slightly thick-walled, mostly 3.5–4 × 2.8–3 μm. Maximum likelihood and Bayesian inference phylogenies inferred from internal transcribed spacer (ITS) region of rDNA indicated that Sanghuangporus spp. formed a monophyletic clade and resolved as a sister to Tropicoporus spp., and six strains of S. toxicodendri formed a monophyletic group which is sister to S. quercicola. An identification key to known species of Sanghuangporus is provided.


Introduction
Sanghuangporus Sheng H. Wu et al. and Tropicoporus L.W. Zhou et al. were recently segregated from the broad generic concept of Inonotus P. Karst (Zhou et al. 2016). The former two genera differ from Inonotus s. str. chiefly in having dimitic hyphal system. Sanghuangporus is characterized by perennial and effused-reflexed to pileate basidiomata, occurring in a variety of climate environment, whereas Tropicoporus is distinguished by annual to perennial basidiomata, and a tropical distribution (Zhou et al. 2016). Zhu et al. (2019) showed the molecular phylogeny strongly supports the monophyly of Sanghuangporus spp.; they also indicated that the maximum crown age of Sanghuangporus is approximately 30.85 million years, and East Asia is the likely ancestral area. Sanghuangporus spp. usually have host-specificity relationships with their host trees. Sanghuangporus accommodates some important medicinal fungal species generally are called "Sanghuang" (means yellow organism grows on Morus) in China and Korea, and "Meshimakobu" in Japan. Sanghuangporus sanghuang (Sheng H. Wu et al.) Sheng H. Wu et al., the generic type, was detected by Wu et al. (2012) as the genuine Sanghuang species growing exclusively on Morus in the wild. Before this study, 13 species of Sanghuangporus were known (Ghobad-Nejhad 2015;Tomsovsky 2015;Zhou et al. 2016;Zhu et al. 2017). In this study, we present a new species of Sanghuangporus sp. growing on Toxicodendron sp. collected from Shennongjia Forestry District, Hubei Province of China.

Morphological studies
All studied specimens are deposited in the herbarium of National Museum of Natural Science, ROC (TNM). The description is based on dried basidiocarps. Freehand and thin sections of fruiting bodies were prepared in three media for microscopic studies: 5% (w/v) potassium hydroxide (KOH) with 1% (w/v) phloxine was used for observation and measurement of microscopic characters; Melzer's reagent was applied to check amyloidity and dextrinoidity; Cotton blue was used to test cyanophily. The abbreviations in the text were used as followed: L = mean spore length (arithmetical average for all spores), W = mean spore width (arithmetical average for all spores), n = total number of spores measured from a specimen, Q = variation in the L/W ratio between the studied specimens. When presenting the variation in the dimensions of spores, 5% of the measurements were rejected from each edge of the range and were given in parentheses.

DNA extraction and sequencing
Genomic DNA were extracted from dried samples with the Plant Genomic DNA Extraction Miniprep System (Viogene-Biotek Corp., New Taipei, Taiwan) following the manufacturer's protocol. Nuclear ribosomal internal transcribed spacer (ITS) region was amplified with primer pair ITS1/ITS4 (White et al. 1990). The PCR protocols for ITS regions were as follows: initial denaturation at 95 °C for 5 min, followed by 40 cycles at 94 °C for 45 s, 53 °C for 45 s and 72 °C for 45 s, and a final extension of 72 °C for 10 min. PCR products were purified and sequenced by the MB Mission Biotech Company (Taipei, Taiwan). Newly obtained sequences were assembled and manually adjusted when necessary using BioEdit (Hall 1999). The sequences were then submitted to Genbank. Zhu et al. (2017) conducted ITS-based phylogenetic analysis for all previously known 13 species of Sanghuangporus. The ingroup strains of the Sanghuangporus spp. and Tropicoporus spp. employed in their analysis were basically adopted in the present analysis. We added newly generated sequences of six strains of the new species (Table 1). Inonotus rickii (Pat.) D.A. Reid, the outgroup in Zhu et al.'s analysis was not adopted, as this root failed to separate all Sanghuangporus spp. from the Tropicoporus spp. We consulted the study of Zhou et al. (2016) and chose Inocutis tamaricis (Pat.) Fiasson & Niemelä as the outgroup, which was successful in constructing the tree with a satisfactory result. The dataset was aligned using MAFFT 7 with Q-INS-i strategy. The aligned sequences were manually adjusted in BioEdit (Hall 1999) when necessary. Parsimony informative sites were calculated using MEGA 7 (Kumar et al. 2016). Phylogenetic trees were inferred from Bayesian inference (BI) and Maximum Likelihood (ML) methods using MrBayes v. 3.2.6. (Ronquist et al. 2012) at the CIPRES Science Gateway (http://www.phylo.org/) and PhyML 3.0 (Guindon et al. 2010), respectively. The best fit model for both algorithms was estimated by jModelTest2 (Darriba et al. 2012) using the Bayesian information criterion (BIC). For ML analysis, bootstrap (BS) values were calculated after running 1000 replicates. The BI analysis was conducted with 10 million generations initiated from random starting trees. Trees were sampled every 1000 generations, and the first 2500 trees were discards as burn-in. The Posterior Probability (PP) values were calculated from the remaining trees. Only the phylogram inferred from ML analysis was shown because both BI and ML analyses yield similar topologies. The statistical supports were shown on nodes of the ML tree when BS ≥ 70 and PP ≥ 0.7. The final phylogenetic trees and alignment were submitted to TreeBASE (submission number 24234; http://www.treebase.org).

Phylogeny results
The ITS dataset consisted of 48 taxa and 1117 sites including gaps, of which 306 sites were parsimony informative. The HKY+G was selected as the best fit model for both the ML and BI analyses. The BI analysis was terminated when the average standard deviation of split frequencies fell to 0.009547. The ML tree shows that Sanghuangporus spp. formed a monophyletic clade (BS = 93%, PP = 1) and resolved as a sister to Tropicoporus spp. (BS = 92%, PP = 1) (Fig. 1). Six strains of Sanghuangporus toxicodendri formed a monophyletic group with statistical supports (BS = 78%, PP = 1), which was sister to S. quercicola L. Zhu & B.K. Cui with significant support (BS = 98%, PP = 1) (Fig. 1).    Etymology. The epithet refers to the host genus. Description. Basidiocarps perennial, effused-reflexed to pileate, applanate, semicircular, adaxially slightly concave, woody hard. Pilei projecting 4-6 cm, up to 18 cm wide and up to 6 cm thick at base. Pileal surface grayish black to blackish brown, glabrous, occasionally cracked, concentrically zonate and sulcate; margin generally obtuse, concolorous or brownish yellow. Pore surface brownish yellow, yellowish brown, brownish or rusty brown, somewhat glancing, darkening in KOH; pores 7-9 per mm, circular. Context homogeneous, 1-5 mm thick or almost invisible, brownish yellow or brownish, with blackish crust at pileus parts. Tubes concolorous with pore surface, 1-5 cm thick, usually with several growth layers.

Discussion
The present phylogenetic study indicated that S. toxicodendri is sister to S. quercicola with significant support (Fig. 1). Both species are distributed in central China; the former grows on Toxicodendron, while the latter occurs on Quercus. However, two morphological features can separate these species. The yellow or brownish-yellow wide marginal zone on the pileus surface of S. quercicola (Zhu et al. 2017: figs A, B) is lacking in S. toxicodendri. Secondly, the basidiospores of S. toxicodendri are mostly longer than 2.8 μm, but are generally shorter than 2.8 μm in S. quercicola.

toxicodendri. Sanghuangporus vaninii grows on
Populus and also resembles S. quercicola in having a wide marginal yellow zone on pileus surface, but it has larger basidiospores (3.8-4.4 × 2.8-3.7 μm; Dai 2010) than S. toxicodendri. Sanghuangporus zonatus is a tropical species distributed in southern China and differs from S. toxicodendri in having thicker context and shorter setae (Tian et al. 2013).
Several Sanghuangporus spp. are used for medicinal application in China, Korea, Japan, and South Asian countries. Wu et al. (2012) indicated that S. sanghuang, the only Sanghuangporus sp. growing on Morus in the wild, is the genuine Sanghuang species. Comparing health-care effectiveness among the so-called Sanghuang species, Lin et al. (2017) proved that S. sanghuang has better medicinal properties than two other commercial species: S. baumii (Pilát) L.W. Zhou & Y.C. Dai and S. vaninii. Sanghuangporus vaninii grows on Populus davidiana in the wild and is widely cultivated in China, Korea, and Japan as a medicinal fungus. Sanghuangporus baumii, which grows on Syringa in the wild, is also served as medicinal fungus in China. The medicinal properties of many Sanghuangporus spp. are not understood. It is noted that S. toxicodendri and the recently described S. quercicola are closely related to the medicinal species S. sanghuang and S. vaninii (Zhu et al. 2019; this study, Fig. 1). The medicinal properties of these two species are worth studying.