﻿Hemiaustroboletus, a new genus in the subfamily Austroboletoideae (Boletaceae, Boletales)

﻿Abstract The present study describes Hemiaustroboletusgen. nov. in the subfamily Austroboletoideae (Boletaceae). Hemiaustroboletus is supported by morphological and molecular data using LSU and RPB2 regions. Additionally, its geographic distribution and intraspecific variation were inferred using ITS sequences. The genus is characterised by pileate-stipitate basidiomata; purple, brown, reddish-brown, orange-brown to dark brown vinaceous pileus; whitish or lilac to vinaceous context and a subclavate stipe. Microscopically, it is characterised by ornamented, slightly verrucose, cracked to perforated brown basidiospores. Two species are described within the genus, Hemiaustroboletusvinaceobrunneussp. nov. and H.vinaceussp. nov.Hemiaustroboletusvinaceussp. nov. is morphologically similar to Austroboletusgracilis, which suggests they may have been confused in the past. This study presents the phylogenetic placement, microscopic structures, detailed morphological descriptions and illustrations of both new species.

In recent years, various authors (Wu et al. 2014;Wu et al. 2016;Gelardi et al. 2020;Kuo and Ortiz-Santana 2020) have recognised the polyphyly of Austroboletus, which is divided into the Austroboletus s.s., Austroboletus s.l. and the A. gracilis s.l. independent clades. This study focuses on the phylogenetic placement and taxonomy of the A. gracilis s.l. clade, placing it in the new genus Hemiaustroboletus with two new species, Hemiaustroboletus vinaceobrunneus and H. vinaceus.

Materials and methods
To resolve the systematics and taxonomy of the new genus Hemiaustroboletus, we conducted an exhaustive sampling of an area with high bolete diversity according to García-Jiménez et al. (2013). The sampling was carried out over the last 10 years including the different biogeographic areas of Mexico: Nearctic, Neovolcanic Axis and Neotropic. The collection trips were conducted in the States of Chiapas, Chihuahua, Estado de Mexico, Jalisco, Michoacan and Oaxaca, in six vegetation types in temperate and subtropical forests during the rainy season from June to October from 2010 to 2019. The samples were characterised at macro-and micromorphological level and three genetic markers were sequenced and analysed.

Morphological study
Morphological characters were described according to Largent (1986) and Lodge et al. (2004). Chemical reactions with KOH and ammonium hydroxide (NH₄OH) were characterised. Photographs of basidiomata were taken in situ, as well as data on the botanical composition of the sites. The colours for taxonomic descriptions were based on Kornerup and Wanscher (1978). Microscopic characters of 30 basidiospores, basidia, pleurocystidia, cheilocystidia, pileipellis cells and stipitipellis were measured by optical microscopy (Carl Zeiss GmbH 37081, Germany). The Q index (length/ width) was estimated for the basidiospores. Ornamentation of basidiospores was observed by scanning electron microscopy (SEM) (Hitachi Su 1510, Hitachi, Japan). The specimens were deposited at the "Herbario Nacional de México" of the "Instituto de Biología, Universidad Nacional Autónoma de México" (MEXU), at the "Herbario José Castillo Tovar del Tecnológico de Ciudad Victoria" (ITCV) and at the "Herbario del Instituto de Botánica, Universidad de Guadalajara" (IBUG).

DNA Extraction, PCR and Sequencing
Samples of dehydrated basidiomata were used for DNA extraction. The DNA was extracted using the DNeasy Power-Soil kit (QIAGEN). Cell lysis was performed by grinding samples in mortar with liquid nitrogen. Three nuclear loci (ITS, LSU and RPB2) were amplified with Platinum Taq DNA Polymerase (Invitrogen-Thermo Fisher Scientific) and Taq & Load PCR Mastermix (MP Biomedicals) in a thermocycler (BIO-RAD). The PCR parameters were as follows: 95 °C initial denaturation for 4 min; 35 cycles of denaturation at 94 °C for 60 s, alignment at 54 °C for 60 s, extension at 72 °C for 60 s and a final extension at 72 °C for 10 min. The primers ITS1/ITS4 (White et al. 1990) were used for the ITS region; LROR/LR5 (Vilgalys and Hester 1990) for LSU; and RPB2-B-F2/RPB2-B-R (Wu et al. 2014) for the partial RPB2 gene. The amplification was examined by 1% agarose gel electrophoresis; gels were stained with GelRed (Biotium) and observed under an UVP Multidoc-It transilluminator (Analytikjena). Only PCR products generated with Taq-Platinum required LB loading buffer. PCR products with successful amplification were cleaned with ExoSAP-IT (Thermo Fisher Scientific) diluted 1:1 with ddH 2 O and incubated at 37 °C for 45 min and 80 °C for 15 min. Sanger sequencing was performed at the "Laboratorio de secuenciación genómica de la biodiversidad y la salud, Instituto de Biología, Universidad Nacional Autónoma de México". Samples were sequenced in both directions with PCR primers using BigDye Terminator v.3.1 (Thermo Fisher Scientific).
We conducted two sets of phylogenetic analyses, the first one to reconstruct the phylogenetic relationships of Hemiaustroboletus gen. nov. and the second one to complement its taxonomic concept with biogeographic and ecological information. The first analysis used the LSU and RPB2 markers in a concatenated matrix, while the second used ITS in order to leverage GenBank data.
The best-fit evolutionary model was estimated with JMODELTEST 2 (Darriba et al. 2012) using CIPRES SCIENCE GATEWAY V. 3.3 (Miller et al. 2010) for each marker separately. For all three markers, the best model was GTR+G+I. We used the LSU-RPB2 dataset to make evolutionary inferences within Austroboletoideae and the ITS dataset to make biogeographic/ecological inferences for Hemiaustroboletus.
The phylogenetic hypotheses (LSU-RPB2) were constructed with Bayesian Inference (BI) and Maximum Likelihood (ML) on a partitioned alignment with same evolutionary model for both markers. Bayesian posterior probability phylogeny was performed using MrBayes algorithm (Ronquist et al. 2012) using two separate Monte Carlo four chains starting from random trees for 10 million generations each (final standard deviation ± 0.224), trees were sampled every 100 generations. The first 25% of samples were discarded as burn-in. ML analyses were performed using the RAxML algorithm (Stamatakis 2014) with 1000 bootstrap replicates. For both analyses, members of subfamilies Boletoideae and Xerocomoideae were used as outgroup. The second analysis (ITS) was performed with the same parameters including Veloporphyrellus and Austroboletus without outgroup. The resulting phylogenetic trees were edited with FIGTREE V.1.4.3 (Rambaut 2009).
Average intrageneric and intergeneric nucleotide similarities between the genera within Austroboletoidae were obtained separately for RPB2, LSU and ITS alignments as follows. For each alignment a nucleotide similarity matrix was computed in GE-NEIOUS 10.2.6 (Biomatters Ltd). Sequences belonging to genera outside Austroboletoidae were removed and then the mean nucleotide similarity was calculated amongst all pairwise comparisons between sequences of each pair of genera.

MycoBank No: 838460
Diagnosis. Hemiaustroboletus is characterised by small and medium basidiomata with slightly ornamented pileus surface, stipe fibrillose to striated without veil, slightly verrucose or cracked to pitted basidiospores and pileipellis formed by an ixotrichoderm or trichoderm.  Etymology. From the Latin hemi "almost or half ", Austroboletus the generic epithet refers to the morphological affinities with this genus.
Distribution. Canada, China, Japan, Mexico, South Korea and United States. Ecology. Temperate and subtropical forests, with conifers and broadleaf trees (Abies spp., Quercus spp., Pinus spp.) from 2000 to 3000 m alt.
Etymology. The name refers to the colour of the pileus, from the Latin "vinosus" vinaceous when young and "brunneus" brown when mature.
Etymology. The name refers to the colour of the pileus from the Latin "vinosus" vinaceous.

Discussion
According the phylogenetic analysis, our collections are nested within the Austroboletoideae close to Veloporphyrellus. Recognising the Hemiaustroboletus genus contributes to solving the systematics within Austroboletoideae since previous works have shown that Austroboletus and Veloporphyrellus, as currently morphologically circumscribed, are polyphyletic (Wu et al. 2016;Gelardi et al. 2020;Kuo and Ortiz-Santana 2020). For example, Wu et al. (2016) found two clades of Austroboletus, Austroboletus. s.s. and a second clade where Austroboletus gracilis s.l. (strain, 112/96) is nested with Veloporphyrellus gracilioides, this species being separated from the Veloporphyrellus s.s. clade. Gelardi et al. (2020) also recovered Austroboletus as polyphyletic with Austroboletus s.s. containing most of the species and other samples divided into four more clades. Particularly, in their analyses, most A. gracilis samples nested close to Veloporphyrellus; this is the clade we are erecting now as Hemiaustroboletus.
Our analyses show that Hemiaustroboletus is related to Veloporphyrellus (Fig. 1). This is supported by previous analyses (Gelardi et al. 2020;Kuo and Ortiz-Santana, 2020); indeed, they differ in several morphological characteristics. Veloporphyrellus has a veil which often embraces the apex of the stipe in younger basidiomata; hymenophoral surface white when young becoming pinkish to pink when mature; basidiospores smooth subfusiform to oblong. In contrast, Hemiaustroboletus has furfuraceous, tomentose to minutely areolate pileus surface; whitish, pink-purple, lilac, magentagrey to brown-violet hymenophoral surface; and slightly verrucose, cracked to pitted ornamented basidiospores (Table 2). Even while the phylogenetic relations between both genera are not statistically supported, nucleotide similarity demonstrated that   (Table 3). These amounts of variation in the three markers also support the conclusion of recognising both genera. Hemiaustroboletus gen. nov. accomplishes the guidelines for the establishment of new genera proposed by Vellinga et al. (2015). It is a monophyletic group supported by morphological data and phylogenetic analyses (BPP = 0.98) (Fig. 1). When Hemiaustroboletus is recognised, the related clade Austroboletus s.s. (the clade including A. dictyotus, the genus type) becomes monophyletic. Additionally, the DNA sequence sampling is broad in taxonomic and geographic terms and uses ribosomal markers and protein coding genes. Indeed, holotypes for both species described are represented with the three markers included in the phylogenetic analyses.
Hemiaustroboletus is proposed as a new genus with two species H. vinaceobrunneus and H. vinaceus, including several of the revised material being previously identified as A. gracilis by Singer et al. (1991), Ayala-Vásquez et al. (2018) and Saldivar et al. (2021). The genus has at least one more known clade (Fig. 1) containing samples originally identified as A. gracilis (TM03-434) from Canada, A. gracilis var. gracilis (CFMR BOS-547) and A. gracilis var. flavipes (CFMR BOS-562) from USA. As found in our analyses and previous works (Wu et al. 2016;Gelardi et al. 2020;Kuo and Ortiz-Santana 2020), A. gracilis is a name widely applied to several clades. In our analysis, the sample A. gracilis 112/96 belongs to Austroboletus (maybe because it lacks RPB2 locus), while the rest of the sequences with this epithet belong to Hemiaustroboletus. As this species is polyphyletic, establishing the true identity of A. gracilis s.s. requires the sequencing of its type specimen, a task beyond the objectives of this study.
Finally, A. gracilis, described by Ortiz-Santana et al. (2007) from Central America, is probably Hemiaustroboletus vinaceus or a close species, because they match the description presented here. Further analysis of these collections and others, labelled as A. gracilis in subtropical regions of Central America and eastern Asia, are needed to fully understand the diversity and distribution of Hemiaustroboletus. Wu