First records of Tylopilus glutinosus Iqbal Hosen (Boletaceae) from Shorea robusta -dominated forests in tropical India: morphological description and phylogenetic estimation

During the routine survey for exploring the hidden macrofungal wealth of tropical Sal ( Shorea robusta Roth) forests in West Bengal and Jharkhand (India), we found a specimen similar to Tylopilus plumbeoviolaceous (Snell & E.A.Dick) Snell & E.A.Dick. After careful morphological observations and phylogenetic analysis, the species was found to be conspecific with Tylopilus glutinosus Iqbal Hosen, a recently established taxon from Bangladesh. We report T. glutinosus for the first time from India and provide a detailed description, figures, a multigene phylogenetic analysis, and comprehensive comparisons with similar species. A distributional map is also provided.


Introduction
One of the most diverse groups of mushrooms is the family Boletaceae, and a genus in this family, Tylopilus P. Karst., also exhibits a remarkable morphological diversity. Molecular phylogeny has shown that Tylopilus is polyphyletic; a number of species previously assigned to it have been reassigned to newly described genera like Zangia Yan C. Li (Binder and Hibbet 2006;Li et al. 2011;Halling et al. 2012aHalling et al. , 2012bNuhn et al. 2013;Wu et al. 2014). Molecular study of some representatives of Tylopilus s.s. reveals species complexes such as Tylopilus balloui (Peck) Singer or T. plumbeoviolaceous (Snell and E.A. Dick) Singer (Halling et al. 2008;Gelardi et al. 2015Gelardi et al. , 2019Chakraborty et al. 2018).
Tropical Sal forests (Shorea robusta Roth; Dipterocarpaceae) in India harbor a many mushroom species Khatua et al. 2015;Kumar and Atri 2016;Verma and Pandro 2018), but the diversity of boletes are, unfortunately, little explored. Few species of boletes have been reported from India (Parihar et al. 2014(Parihar et al. , 2018Verma and Pandro 2018). During macrofungal surveys to several Shorea robusta-dominated forests in West Bengal and Jharkhand, India, we collected several basidiomata allied to Tylopilus plumbeoviolaceus (Snell & E.A.Dick) Snell & E.A.Dick. Critical macroand micromorphological characterization, coupled with combined nrITS and nrLSU-based phylogenetic studies, revealed our specimens to be T. glutinosus Iqbal Hosen, a recently established taxon from Bangladesh (Iqbal Hosen 2021), which had not been recorded from India until now. Here, we report T. glutinosus in India for the first time, based on a detailed morphological description and a multigene (nrITS + nrLSU) phylogenetic inference.

Methods
We collected fresh young to mature basidiomata during surveys to various areas of Jharkhand and West Bengal during the monsoon season, July and August, in 2019-2021. Macromorphological characters were recorded in the field or at basecamp from fresh and dissected fruitbodies. We photographed basidiomata in the field with a Sony DSC-RX100 camera. Colour codes and terms mostly follow Kornerup and Wanscher (1978). Samples were dried in a field drier. We observed micromorphological characters of freehand sections of dried materials mounted in a solution of 5% KOH, 1% Phloxin, and 1% ammoniacal Congo red with an Olympus CX 41 compound microscope. Drawings of the anatomical features were made with a drawing tube at 1000× magnification. Microscopic photographs were taken with an Olympus BX 53 camera. The basidiospores were measured in lateral view. Basidiospore measurements and length/width ratios (Q) are recorded as: minimum-mean-maximum. Basidium length excludes the length of sterigmata. Herbarium codes follow Thiers (2021). The distributional map ( Fig. 1) was produced in ArcGIS v. 10.5.

DNA extraction, PCR amplification and sequencing.
Genomic DNA was extracted from 100 mg of dried basidiomes with the InstaGeneTM Matrix Genomic DNA isolation kit (Biorad, USA) following the manufacturer's instructions. The nrITS and nrLSU genes regions were amplified with primer pairs ITS-1F and ITS-4 (White et al. 1990;Gardes and Bruns 1993) and LR0R and LR7 (Vilgalys and Hester 1990), respectively. PCR amplification was performed on a thermal cycler (Eppendorf, Germany) programmed for 5 min at 95 °C, 30 cycles of 1 min at 95 °C, 30 s at 52 °C, 2 min at 72 °C, and a final 7 min extension step at 72 °C. The PCR products were purified using the QIAquick PCR Purification Kit (QIAGEN, Germany). Both strands of the PCR fragment were sequenced on the 3730xl DNA Analyzer (Applied Biosystems, USA) using the amplifying primers. The sequence quality was checked using Sequence Scanner Software v. 1 (Applied Biosystems). Sequence alignment and required editing of the obtained sequences were carried out using Geneious v. 5.1 (Drummond et al. 2010). The newly generated sequences in this study were submitted to GenBank (http://www.ncbi.nlm.nih. gov/Genbank/). Accession numbers of species used in the phylogenetic analysis are listed in Table 1.
Sequence alignment and phylogenetic analysis. The newly generated nrITS and nrLSU sequences of T. glutinosus from India and its close relatives were retrieved from the nBLAST search against GenBank (https:// www.ncbi.nlm.nih.gov/genbank, accessed on 2022-2-22), UNITE database (https://unite.ut.ee, accessed on 2022-2-22), and relevant published phylogenies (Gelardi et al. 2015(Gelardi et al. , 2019Chakraborty et al. 2018;Hosen 2021). Two datasets (ITS and LSU) were created separately. Both the datasets were aligned separately using the online version of MAFFT v. 7 with the L-INS-i strategy (Katoh et al. 2019) with default settings and then trailing ends of the alignment trimmed manually with MEGA v. 7 . To eliminate ambiguously aligned positions in the alignment as objectively as possible, Gblocks 0.91b (Talavera and Castresana 2007) was used. The program was run with settings allowing for smaller blocks, gaps within these blocks and less strict flanking positions. Species delimitation was first examined using single locus phylogenies. When significant conflict was not observed among the single locus phylogenies, we concatenated all single locus alignments into one multilocus dataset using BioEdit v. 7.2 (Hall 1999). The concatenated dataset was then phylogenetically analyzed using the maximum-likelihood (ML) and Bayesianinference (BI) methods. For the ML analysis, the concatenated alignment was carried out using raxmlGUI v. 2.0 (Edler et al. 2021) with the GTRGAMMA substitution model. The ML analysis was executed applying the rapid bootstrap algorithm with 1000 replicates to obtain nodal support values. For BI, the 2-loci dataset was divided into four partitions: ITS1, 5.8S, ITS2, and LSU. Parti-tionFinder2 was used to find the appropriate partitioning scheme and substitution models for each partition (SYM+I+G for ITS1, SYM for 5.8S, GTR+G for ITS2 and GTR+I+G for 28S) using the Akaike information criterion (AICc) with a greedy search over all models (Lanfear et al. 2016). BI was computed in MrBayes v. 3.2.6 (Ronquist et al. 2012) with four Markov chain Monte Carlo (MCMC) chains for 1,000,000 iterations until the standard deviation of split frequencies reached below the 0.01 threshold. Trees were sampled every 100 th generation. The first 25% of trees were discarded as burn-in.

Results
Phylogenetic inferences. Both ML and BI analyses produced the same topology; therefore, only the Bayesian trees with both MLBS and BPP values are shown (Fig. 2). The ITS data matrix comprised a total of 26 sequences; the alignment comprised 446 characters. The LSU matrix consisted of 50 sequences; the alignment comprised 831 characters. The 2-loci (ITS + LSU) final dataset consisted of 58 sequences including our consensus sequence for each species. The final alignment comprised 1277 characters including gaps. Our phylogenetic analysis using nrITS and nrLSU genes shows that our Indian Tylopilus glutinosus  are nested (with strong support; MLBS = 100%, BPP = 1) Identification. Pileus 6-110 mm in diameter, initially subglobose then convex; surface dry but viscid when wet, matte to subvelvety, greyish brown (7D3), brown to purple brown (14D3-D2), paler towards margin, fading to ash grey (1B-C2) with age; margin entire, decurved with a narrow flap of tissue. Pore surface greyish orange (6B3-B2) when young, greyish brown (6C3) with age, unchanging on bruising; pores angular, stuffed when young. Tubes adnate, 3-4 mm long, grey whitish (4C1), unchanging on bruising. Stipe 14-140 × 8-37 mm, mostly subclavate with wider base or broadly cylindrical, solid, greyish magenta to dark purple (14E3-F4), white towards base; surface with faint longitudinal striations, without reticulum. Veil absent. Basal mycelium white. Context up to 16 mm thick in the pileus, milky white (1A2), unchanging when exposed; no color change with 5% KOH, FeSO 4 , and 10% NH 4 OH. Odour mushroomoid. Taste bitter. Spore print not obtained.
Habit and habitat. Mostly gregarious to subcaespitose, growing in association with Shorea robusta in tropical deciduous forests.

Discussion
Our collections of Tylopilus glutinosus from India morphologically alike to the species, as recently described from Bandladesh (Iqbal Hosen 2021). However, our material differs from the type collection in having longer pleurocystidia, 35-60 versus 30-45 µm according to Hosen (2021). Our 2-loci phylogenetic results confirm sequences from our Indian samples are nested within the T. glutinosus clade (Fig. 2,  glutinosus, but all of them differ by their mycorrhizal association with members of the Fagaceae. Although these species may be close relatives of T. glutinosus, they are not contaxic, as T. glutinosus associates with Shorea in tropical regions of Bangladesh and India. Our phylogenetic reconstruction places T. glutinosus sister to the Chinese T. plumbeoviolaceoides, which morphologically differs by its pale pink to flesh-pink or pale vinaceous tan-colored tubes and pore surfaces and the subpruinose stipe (Li et al. 2002). Another morphologically similar taxon, the American T. plumbeoviolaceus, can also be separated from this Asian lookalike by its larger basidiomata (pileus 3-15 cm in diameter), the amber-orange and pale brownish amber reactions with KOH and NH 4 OH on the pileus surface, respectively, and longer basidiospores (10-13 µm long) (Bessette et al. 2010). Tylopilus violaceobrunneus can also be distinguished from this species by its reddishbrown to brownish-violet pileus and distinctly reticulate stipe, especially at apex (Wu et al. 2016). Tylopilus vinosobrunneus is separated by its reticulate stipe at apex and pallid tubes that turns wood-brown when bruised (Chen et al. 2004). Similarly, T. atroviolaceobrunneus is another morphologically related species which differs from T. glutinosus in turning reddish in the context and tubes on injury and in having longer basidiospores (10-13 µm long) (Wu et al. 2016). Tylopilus alboater can easily be distinguished in the field by its robust pileus (3-15 cm in diameter), purple-black to dark purple pileus, and mild taste (Bessette et al. 2010;Wu et al. 2016). Tylopilus atripurpureus differs from T. glutinosus by its purpleblack to dark purple pileus, longer pleurocystidia (75-100 µm long), and palisadoderm pileipellis (Horak 2011;Gelardi et al. 2015). Tylopilus neofelleus Hongo, which  has also been reported from India, is distinguished from T. glutinosus by the pale violet reticulum on the upper part of the stipe (Chakraborty et al. 2018;Gelardi et al. 2019).