Hydnaceous fungi of China 8. Morphological and molecular identification of three new species of Sarcodon and a new record from southwest China

Abstract Three new stipitate hydnoid fungi, Sarcodon coactus, S. grosselepidotus and S. lidongensis, are described and illustrated, based on morphological characteristics and nuc ITS rDNA + nuc LSU rDNA sequence analyses and a new record, S. leucopus, from China is reported. S. coactus is characterised by ellipsoid to round basidiocarps, reddish-brown to dark brown, felted pileal surface with white and incurved margins, simple-septate and partly short-celled generative hyphae and irregular subglobose, thin-walled, brown basidiospores with tuberculate ornamentation (tuberculi up to 1 μm long). S. grosselepidotus is characterised by infundibuliform to round, occasionally deeply fissured pileus, pale orange to dark ruby pileal surface with ascending and coarse scales, simple-septate generative hyphae and irregular ellipsoid to globose, thin-walled, brown basidiospores with tuberculate ornamentation (tuberculi up to 0.7 μm long). S. lidongensis is characterised by plano-convex to somewhat depressed and regular orbicular pileus, light brown to dark brown pileal surface with adhering squamose and purplish-brown, incurved and occasionally incised margin, cylindrical or broadened below stipe, simple-septate generative hyphae and irregular ellipsoid to subglobose, thin-walled basidiospores with tuberculate ornamentation (tuberculi up to 1 μm long). The absence of the clamp connection is the common morphological characteristic of these three new species; however, S. leucopus, a new record from China, has frequently clamped generative hyphae. Molecular analyses confirm the phylogenetic positions of three new and the new record species. The discriminating characters of these three new species and closely related species are discussed and a key to the species of Sarcodon from China is provided.

Species of Bankeraceae are ectomycorrhizal fungi which associate with many kinds of angiosperm and gymnosperm trees, especially with Pinaceae and Fagaceae, such as Pinus strobus, Picea sitchensis, Fagus grandifolia, Quercus rubra and Castanea sativa (Maas Geesteranus 1975;Harrison 1984;Baird 1986;Baird et al. 2013) and usually occur in natural and comparatively undisturbed forests (Arnolds 1989). They can obtain energy from and transport nutrients to the host plants and are of great ecological significance in promoting forest vegetation recovery (Gardes and Bruns 1996;Erland and Taylor 1999). These fungi are vulnerable to impact due to changes in the environment, such as habitat loss, nitrogen deposition, decrease of host tree species and subsequently increased ground temperatures (Arnolds 1989;Otto 1992;Vesterholt et al. 2000;Newton et al. 2002;Arnolds 2010;Baird et al. 2013). In Europe, stipitate hydnoid fungi have been considered one of the most endangered groups of macrofungi and have been included in Red Data Lists (Hrouda 1999;Walleyn and Verbeken 2000;Hrouda 2005;Nitare 2006;Senn-Irlet et al. 2007), which have been used as indicators that forests need to be protected (Ainsworth 2005;Nitare 2019).
The genus Sarcodon is characterised by solitary to gregarious, stipitate, pileate basidiocarps, hydnaceous hymenophore, the monomitic hyphal system owning inflating or not inflating hyphae, the presence or absence of clamp connections and irregular ellipsoid to globose, tuberculate basidiospores which are brown in mass. Besides, the dry basidiocarps often produce farinaceous to fragrant or acidic odour (Maas Geesteranus 1971;Baird 1986;Arnolds 2003;Baird et al. 2013). In morphology, Sarcodon is closely related to Hydnellum, but the former usually has soft and fleshy basidiocarps and the latter has hard and corky basidiocarps (Maas Geesteranus 1971;Larsson et al. 2019). The macro-morphology of these two genera often depends on their environmental parameters, such as precipitation, temperature or obstructions. Additionally, the variable growth of basidiocarps makes it difficult to distinguish each other. Therefore, it is essential to support and confirm their identities using molecular sequence data (Baird et al. 2013). Recent molecular phylogenetic analyses reveal that Sarcodon and Hydnellum form paraphyletic lineage and suggest using the spore length as the delimitation between the two genera. Hydnellum species had spore lengths in the range 4.45−6.95 μm, while the corresponding range for Sarcodon was 7.4−9 μm (Larsson et al. 2019).
Investigations of hydnaceous fungi in China have been carried out in recent decades and many Sarcodon specimens have been collected. During the study of these specimens, three undescribed species and a new record species have been identified using morphological characters and phylogenetic analyses of nuc rDNA ITS1-5.8S-ITS2 combined with nuc 28S rDNA sequences. Here, we describe them in this paper.

Morphological studies
Specimens are deposited at the herbarium of the Institute of Applied Ecology, Chinese Academy of Sciences (IFP). Microscopic procedures follow Mu et al. (2019). Microscopic studies used sections mounted in Cotton Blue (CB): 0.1 mg aniline blue dissolved in 60 g pure lactic acid; CB− = acyanophilous. Amyloid and dextrinoid reactions were tested in Melzer's reagent (IKI): 1.5 g KI (potassium iodide), 0.5 g I (crystalline iodine), 22 g chloral hydrate, 20 ml distilled water; IKI− = neither amyloid nor dextrinoid reaction. Sections were mounted in 5% KOH (potassium hydroxide) and studied at magnifications up to 1000× using a Nikon Eclipse E600 microscope (Tokyo, Japan) with phase contrast illumination. Dimensions were estimated subjectively with an accuracy of 0.1 μm. In presenting basidiospore size ranges, 5% of the measurements at each end of the range are given in parentheses. The following abbreviations are used in the text: L m = mean spore length, W m = mean spore width, Q = range of length/width ratios for specimens studied and n = total number of basidiospores measured from a given number of specimens. The surface morphology for the basidiospores was observed with a Phenom Prox scanning electron microscope (ESEM, Phenom Prox, FEI, The Netherlands) at an accelerating voltage of 10 kV. A thin layer of gold was coated on the samples to avoid charging. Special colour terms are from Rayner (1970) and Munsell (2015).
Nuclear ribosomal RNA genes were used to determine the phylogenetic position of the new species. After PCR amplification, the products were sequenced in both directions and the sequences were assembled using DNAMAN 8.0. DNA sequences were aligned with MUSCLE in MEGA7 (Kumar et al. 2016). Alignments were manually adjusted to allow maximum alignment and minimise gaps. Maximum parsimony and Bayesian analysis were applied to the ITS + LSU dataset. All characters were weighted and gaps were treated as missing data. Maximum parsimony analysis (PAUP* version 4.0b10) was used (Swofford 2002). Trees were inferred using the heuristic search option with tree bisection reconnection (TBR) branch swapping and 1,000 random sequence additions. Max-trees were set to 5000 and no-increase, branches of zero length were collapsed and all parsimonious trees were saved. Clade stability was assessed using a bootstrap (BT) analyses with 1,000 replicates (Gaget et al. 2017). Descriptive tree statistics, tree length (TL), consistency index (CI), retention index (RI), rescaled consistency index (RC) and homoplasy index (HI), were calculated for all trees generated under different optimality criteria. Maximum Likelihood (ML) analysis was performed in RAxML v8.2.4 with GTR+I+G model (Stamatakis 2014). The best tree was obtained by executing 1000 rapid bootstrap inferences and thereafter a thorough search was undertaken for the most likely tree using one distinct model/data partition with joint branch length optimisation (Stamatakis et al. 2008). Bayesian analyses with MrBayes 3.2.4 (Cannatella 2015) implementing the Markov  (Posada and Crandall 1998;Nylander 2004) were performed and the parameters in MrBayes were set as follows: lset nst = 6, rates = invgamma. Four simultaneous Markov chains were run starting from random trees, keeping one tree every 100 th generation until the average standard deviation of split frequencies was below 0.01. The value of burn-in was set to discard 25% of trees when calculating the posterior probabilities. Bayesian posterior probabilities were obtained from the 50% majority rule consensus of the trees kept. Then we used the FigTree v1.3.1 or Treev32 to visualise the resulting trees.

Phylogenetic analyses
The combined ITS-LSU dataset represented 97 taxa and 1328 characters long after being trimmed. Amaurodon aquicoerule was used as the outgroup. The data matrix comprised 800 constant characters, 81 parsimony uninformative variable characters and 447 parsimony informative positions. Maximum parsimony analysis was performed and a strict consensus tree was obtained (TL = 2351, CI = 0.376, RI = 0.728, RC = 0.273, HI = 0.624). Bayesian analysis ran for 8 million generations and resulted in an average standard deviation of split frequencies of 0.004708. The same dataset and alignment were analysed using the ML method and a similar topology was generated.
The ML tree is shown in Figure 1. In the phylogenetic tree, nine sampled specimens formed three single clades with high to full support (100% in ML, 99% or 100% in MP and 1.00 BPP) and clustered in the clade that comprised most species of    Description. Basidiocarps annual, solitary to gregarious, soft and fleshy when fresh, becoming firm and light in weight upon drying; taste none, odour farinaceous when dry. Pileus planar, ellipsoid when young, later round with age, up to 35 mm across and 4-8 mm thick at centre. Pileal surface reddish-brown (8D5) to dark brown (8F8), azonate, pubescent, floccose to felted when fresh, becoming smooth, rugose, scrobiculate when dry; margin white (7A1) when fresh, greyish-brown (7D3) with age, incurved, rarely lobed. Spine surface white (4A1) to yellowish-white (4A2) when fresh, brownishorange (5C5) to yellowish-brown (5F6) when dry; spines up to 2.1 mm long, base up to 0.3 mm diam., conical, 3-5 per mm, decurrent on stipe, without spines at pileus margin, brittle when dry. Context not duplex, up to 6 mm thick, light brown (5D5), firm; Stipe central, up to 5.5 cm long and 1.3 cm diam., fleshy, greyish-brown (8D3) to violet brown (10F7) when fresh, becoming hollow with age, greyish-orange (5B3) to dark brown (7F7) upon drying, rugous, columniform or attenuate below with bulbous base when old.
Additional specimen examined -China. Etymology. Grosselepidotus (Lat.), from the Latin word grosse and lepidotus, in reference to the coarsely scaled pileal surface.
Description. Basidiocarps annual, solitary to gregarious, soft and freshy when fresh, becoming fragile and light in weight upon drying; taste none, odour mildly fragrant when dry. Pileus infundibuliform or circular when young, later planar and ellipsoid to round with age, occasionally deeply fissured, up to 75 mm diam. and 4-8 mm thick at centre. Pileal surface pale orange (6A3) to dark ruby (12F8), azonate, glabrous with ascending, broad and dark brown (9F5) scales when fresh, becoming scabrous, rugose when dry; margin inflexed and wavy, sometimes lobed with age. Spine surface white (4A1) to pale yellow (4A3) when fresh, light brown (6D6) to dark brown (6F8) when dry; spines up to 1.4 mm long, base up to 0.3 mm diam., conical, 4-6 per mm, strongly decurrent on stipe, without spines at pileus margin, brittle when dry. Context not duplex, up to 5 mm thick, greyish-orange (5B5), firm; Stipe central to lateral, up to 9.5 cm long and 2 cm diam., fleshy when fresh, firm upon drying, brownish-yellow (5C7) to dark brown (7F7), creased, inside solid, cylindrical or attenuate below with bulbous base when old. Hyphal structure. Hyphal system monomitic; generative hyphae with simple-septa, CB-, IKI-; tissues olivaceous in KOH.
Etymology. Lidongensis, refers to Lidong County, where the specimens were collected. Description. Basidiocarps annual, simple to concrescent, soft and freshy when fresh, becoming firm and light in weight upon drying; taste bitterish, odour farinaceous when dry. Pileus planar and circular when young, later plano-convex to somewhat depressed and regular orbicular with age, up to 35 mm across and 5-8 mm thick at centre. Pileal surface light brown (6D7) to brown (7E8), azonate, velutinate, then matted, appressed squamose to rimose when fresh, and purplish-brown at the pileal margin, dark brown in centre, becoming scrobiculate and verrucose when dry; margin incurved and occasionally incised with age. Spine surface greyish-orange (6B3) to brown (6E6) when fresh, light brown (6D5) to brown (6E7) when dry; spines up to 1 mm long, base up to 0.2 mm diam., conical, 4-6 per mm, more or less decurrent on stipe, with spines at pileus margin, brittle when dry. Context not duplex, up to 6 mm thick, orange white (5A2) to yellowish-brown (5D6), firm; stipe central, up to 4.5 cm long and 1 cm diam., fleshy when fresh, rigid upon drying, light brown (6D6) to dark brown (6F6), fibrillose, inside solid, cylindrical or broadened below with bulbous base when old.

Discussion
Three new species of Sarcodon were described, based on the morphological characteristics and molecular data and were the first new species described from China. Phylogenetic analyses of the nuc ITS rDNA + nuc LSU rDNA dataset by ML, MP and Bayes in this study showed a low level of support in the deeper nodes of the topology, but high support at the species level. The result is in keeping with previous reports (Baird et al. 2013;Larsson et al. 2019).
Sarcodon coactus and S. grosselepidotus are closely related in the phylogenetic tree and share similar morphological and anatomical characteristics: solitary to gregarious basidiocarps with round pileus, central and columniform stipe, decurrent spines, context tissue becoming olivaceous in KOH and isolated or grouped tuberculi. However, S. grosselepidotus can be differentiated by infundibuliform basidiocarps, fissured pileus, coarse and scaly pileal surface, shorter spines (up to 1.4 mm vs. 2.1 mm in S. coactus) and slightly shorter tuberculi (up to 0.7 μm vs. 1 μm in S. coactus).
The specimens, involved in this study, were collected from the forests dominated by Fagaceae trees such as Quercus acutissima, Lithocarpus dealbatus, Castaopsis orthacantha and a small portion of coniferous trees, for instance, Pinus armandii. We speculated that these species may form an ectomycorrhizal association with Fagaceae trees. The new record sample was fully identical with S. leucopus described by Mleczko (2011) in morphology and molecular analysis and pine and spruce were primary ectomycorrhizal companions of this fungus.