Introduction

Boletaceae Chevall. is a species-rich fungal family in the order Boletales E.-J. Gilbert. Within the family, many species are important because of their ecological and economical value (Arora 2008; Sitta and Floriani 2008; Sitta and Davoli 2012; Cui et al. 2016). Boletes are attractive to mushroom pickers as they can be easily recognized in the field by their brilliant colors and usually large basidiomata with fleshy context and tubular hymenophore instead of a lamellar configuration (Corner 1972; Wu et al. 2014). Most species of the Boletaceae form ectomycorrhizal symbioses (ECMs) with various plants including Dipterocarpaceae Blume (Hong 1979; Watling and Lee 1995, 1998; Lee et al. 2003; Sirikantaramas et al. 2003; Yuwa-Amornpitak et al. 2006; Peay et al. 2010; Brearley 2012; Hosen and Li 2017; Wu et al. 2018). In Southeast Asia and the Indian subcontinent, Dipterocarpaceae are important ECMs host plants. Shorea robusta C.F. Gaertn., a broad-leaved tree in the plant family Dipterocarpaceae, is a native species on the Indian subcontinent. A considerable number of fungi have previously been reported with an association to S. robusta (Tulloss and Bhandary 1992; Dutta et al. 2015; Hosen et al. 2013, 2015; Hosen and Li 2015, 2017; Parihar et al. 2018a, b); however, Wu et al. (2018) hypothesized that new boletoid taxa await discovery, if more intensive collecting to be done in that region.

Since the development of an excellent phylogenetic framework of the Boletaceae by Wu et al. (2014), ca. 35 new genera have been erected and established in that family, mostly from East Asia (Zhao et al. 2014; Zeng et al. 2014; Orihara et al. 2016; Wu et al. 2016; Zhang and Li 2018), Southeast Asia (Wu et al. 2018; Khmelnitsky et al. 2019; Vadthanarat et al. 2019), and Europe (Gelardi et al. 2014a, b, c; Vizzini 2014a, b). Wu et al. (2014) showed seven major clades at the subfamily levels, Leccinoideae is one of the subfamilies, and includes 14 genera (Wu et al. 2018). A recent study by Kuo and Ortiz-Santana (2020) merged several well-known sequestrate (Chamonixia Rolland, Octaviania Vittad., Rossbeevera T. Lebel & Orihara, and Turmalinea Orihara & N. Maek.) and stipitate-pileate (Leccinum Gray and Leccinellum Bresinsky & Manfr. Binder) genera into a single large genus Leccinum in subfamily Leccinoideae. Orihara et al. (2021) argued against this broad concept of Leccinum because of the lack of backbone resolution in the evolutionary tree, and because several other major clades of stipitate-pileate taxa (Leccinum and/or Leccinellum) within Leccinoideae were poorly resolved. Furthermore, Orihara et al. (2021) pointed that synonymy of these sequestrate and stipitate-pileate genera with Leccinum is premature and may result in the loss of information of evolutionary history. In an earlier study, Lebel et al. (2012) also argued against the collapse of five genera (Chamonixia, Leccinum, Leccinellum, Rossbeevera, and Octaviania) into Leccinum because of the lack of support at the genus level, and the presence of remarkable morphological and molecular characters supporting them as distinct genera within Leccinoideae. In this study, we follow the traditional classification and interpretation of Lebel et al. (2012) and Orihara et al. (2016, 2021).

In Bangladesh, only five putatively ECMs fungi have been reported previously in association with S. robusta (Hosen et al. 2013, 2015; Hosen and Li 2015, 2017). During studies of boletes in Bangladesh in association with S. robusta, the first author found several collections of the same bolete which were superficially similar to a species of Leccinum, Leccinellum, or Spongispora G. Wu et al. Surprisingly, however, the nrLSU sequence derived from the new bolete collection matched only 86.78% with the known species of Leccinum available in GenBank. Molecular phylogenetic analyses of 4-gene, i.e., the nuclear ribosomal large subunit (nrLSU), translation elongation factor-1α (TEF1-α), the largest subunit of RNA polymerase II (RPB1), and the second largest subunit of RNA polymerase II (RPB2) along with detailed morphological observation did not place the new bolete with either Leccinum/Leccinellum or Spongispora. Rather, the evidence supports a separate generic lineage distinct from those genera. Therefore, this new generic lineage is formally described as a new genus in the subfamily Leccinoideae of the family Boletaceae.

Material and methods

Specimen collection and deposition

The specimens were collected from tropical Bangladesh in association with Shorea robusta, described and deposited in the Cryptogamic Herbarium of Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, China (KUN-HKAS), and in the private herbarium of the first author (PHI).

Morphological studies

Macromorphological descriptions were based on field notes and documented with photographs. Color codes were derived from Kornerup and Wanscher (1981). Micromorphological observations were made from dried specimens mounted in H2O, 5% aqueous KOH (w/v) and Congo Red. Melzer’s solution was used to check any amyloid reaction of basidiospores and tissues. Line drawings were made freehand. In the description of basidiospore measurements, the notation [n/m/p] was used, in which n basidiospores were measured from m basidiomata of p collections. Dimensions for basidiospores are given as (a–)b–c(–d), in which “b–c” contains a minimum of 90% of the measured values and extreme values “a” and “d” are given in parentheses, whenever necessary. Q denotes the length/width ratio of a measured basidiospore, Qm denotes the average of n basidiospores, and SD is their standard deviation. Results are presented as Qm ± SD. Basidiospores were also observed using a scanning electron microscope (SEM) and following the protocol in Hosen et al. (2013).

DNA extraction, PCR amplification, and sequencing

Protocols for genomic DNA extraction, PCR amplification, and sequencing followed those described in Hosen et al. (2013) and Wu et al. (2014), and references cited therein. In this study, one nuclear ribosomal and three nuclear protein encoding genes were used: 1) the nuclear ribosomal large subunit (nrLSU), 2) the protein encoding gene for translation elongation factor-1α (TEF1-α), 3) the protein encoding gene for the largest subunit of RNA polymerase II (RPB1), and 4) the protein encoding gene for the second largest subunit of RNA polymerase II (RPB2). LR0R/LR5 (Vilgalys and Hester 1990), ef1-983F/ef1-1567R (Rehner and Buckley 2005), rpb1-BF/rpb1-BR (Wu et al. 2014) and rpb2-BF/rpb2-BR (Wu et al. 2014) primer pairs were used for the amplification of nrLSU, TEF1-α, RPB1, and RPB2 fragments, respectively. PCR products were purified with a Gel Extraction & PCR Purification Combo Kit (Spincolumn; Bioteke, Beijing, China). Samples were sequenced using the same primers as in the original PCR amplifications.

DNA sequence alignments and dataset assembly

The newly obtained nrLSU sequences from the new bolete were used as a query in a Blast search to compare with those already deposited in GenBank. The closest matches and reference sequences were then retrieved, mainly from recent phylogenetic studies (Desjardin et al. 2011; Orihara et al. 2012a, 2016; Hosen et al. 2013, 2019; Nuhn et al. 2013; Wu et al. 2014, 2018; Vadthanarat et al. 2018; Khmelnitsky et al. 2019; Kuo and Ortiz-Santana 2020). First, a nrLSU dataset was reconstructed which consisted of 360 nrLSU sequences of Boletales (data not shown). In this dataset, almost all representative genera of Boletaceae were included and the new bolete appeared in the subfamily Leccinoideae with moderate Bootstrap (BS) support. Following the nrLSU analysis, representative species/genera of the subfamily Leccinoideae, within which the new bolete nested, were considered for the 4-gene analyses. For the 4-gene analyses, a total of 67 samples of Boletaceae were retained from the nrLSU taxon set, and subsequently available sequences of TEF1-α, RPB1, and RPB2 were retrieved from GenBank (Table 1). Additionally, five samples from the three genera Binderoboletus T.W. Henkel & M.E. Sm., Ionosporus Khmeln., and Rhodactina Pegler & T.W.K. Young were included in the 4-gene dataset to determine the phylogenetic affinity among genera within Leccinoideae. Each single-gene dataset was aligned in MAFFT v.6.8 using the E-INS-i strategy (Katoh et al. 2005), and manually edited in BioEdit v.7.0.9 (Hall 1999). The single-gene aligned datasets were then concatenated using Phyutility (Smith and Dunn 2008) for combined phylogenetic analyses. MrModeltest 2.3 (Nylander 2004) was used to select the best-fit substitution model of evolution for each aligned dataset separately with Akaike information criterion (AIC). The selected models were GTR+I+G, GTR +I+G, SYM+G, and K80+I+G for nrLSU, TEF1-α, RPB1, and RPB2 datasets, respectively.

Table 1 List of collections included in the molecular phylogeny
Full size table

Phylogenetic analyses

The maximum likelihood (ML) method was conducted with RAxML v.7.2.6 (Stamatakis 2006). As RAxML only supports the GTR model, GTRGAMMAI was used for analysis of the 4-gene dataset. The phylogenetic tree was inferred by a single analysis with four partitions (one for each gene). All parameters in the ML analysis were set as default, and BS support values were obtained with 1000 replicates. Support values from bootstrapping runs (MLB) were mapped on the best ML tree using the “-f a” option of RAxML and “-x 12345” as a random seed to invoke the novel rapid bootstrapping algorithm.

Bayesian inference (BI) analysis was conducted with MrBayes 3.2 (Ronquist and Huelsenbeck 2003). Individual best-fitting substitution models were assigned to four different partitions. Partitioned Bayesian analysis with four chains was conducted by setting 6 million generations under the selected evolutionary models. Trees were sampled every 100 generations and posterior probabilities (PP) calculated after discarding the first 25% of the samples as the burn-in. At the end of the run, the average deviation of split frequencies was 0.006.

Results

Molecular phylogenetic results

A total of nine sequences from three collections of the bolete species was obtained and submitted to GenBank (Table 1). The final 4-gene dataset consisted of 181 sequences (nrLSU: 67, TEF1-α: 61, RPB1: 18, RPB2: 35; Table 1) from 72 bolete samples, and Boletus edulis Bull. as the outgroup taxon. The aligned dataset consisted of 3017 characters with gaps (nrLSU: 927, TEF1-α: 642, RPB1: 784, RPB2: 664) of which 1754 were constant, 270 were parsimony uninformative and 999 are parsimony informative, and was submitted to TreeBASE (S25339). Both ML and BI analyses produced almost the same topologies; thus, only the ML tree is presented (Fig. 1). In the phylogenetic analyses, the new bolete appears on a long branch in the subfamily Leccinoideae, and clusters with the stipitate-pileate genera Leccinum, Leccinellum, and Spongispora, as well as the sequestrate genera Chamonixia, Octaviania, Rossbeevera, and Turmalinea with moderate support (70% ML BS, PP = 0.97). Although the new bolete is nested in a well-supported leccinoid clade, its closest sister relationship with other genera remains unclear. However, the new bolete lineage always forms a clade basal/sub-basal to Chamonixia, Octaviania, Leccinum, Leccinellum, Turmalinea, and Rossbeevera. The ML analysis of the 4-gene matrix resulted in the phylogenetic tree shown in Fig. 1.

Fig. 1
figure 1

Phylogenetic relationships of the genera of subfamily Leccinoideae inferred from 4-gene (nrLSU, TEF1-α, RPB1, and RPB2) dataset of 72 samples of Boletaceae using maximum likelihood (ML). Bootstrap (BS) values derived from ML (BS > 50%) and posterior probabilities (PP) from Bayesian Inference (BI) (PP = 0.90) analyses are shown above or beneath the branches at nodes. Kaziboletus rufescens gen. et sp. nov. is highlighted in bold. Voucher specimen number for each species is provided after the species name followed by country name. An asterisk next to the voucher specimen number indicates holotype material

Full size image

Taxonomy

Kaziboletus Iqbal Hosen & Zhu L. Yang, gen. nov.

MycoBank No.: MB 833238.

Etymology: The genus name “Kaziboletus (Lat.)” honors Dr. Kazi M. Badruddoza, a national emeritus scientist and founder of modern agriculture of Bangladesh.

Diagnosis: Pileus 30–60 mm broad, epigeous, stipitate-pileate with a tubular hymenophore. Pileus glabrous to rimose, red to dark reddish brown when young, becoming off white to cream-white or gray to grayish white with age. Hymenophore tubular, free, depressed around the stipe, white to cream white. Stipe central, cylindrical, covered with reddish brown scabrous squamules when young, becoming grayish brown at maturity, longitudinal striations with fine cross lines present from the apex almost to the base, the ridges somewhat anastomosing but not reticulate; basal mycelium whitish. Context white, turning pale red to pale reddish orange in some patches when exposed. Basidiospores light yellow to deep yellow in H2O and 5% KOH, smooth, elongated to cylindrical, somewhat fusoid. Cystidia mostly fusoid. Pileipellis an epithelium. Clamp connections absent.

Type species: Kaziboletus rufescens Iqbal Hosen & Zhu L. Yang

Kaziboletus rufescens Iqbal Hosen & Zhu L. Yang, sp. nov. Figs. 2, 3, and 4

Fig. 2
figure 2

Basidiomata of Kaziboletus rufescens. a, b HKAS 74706, holotype; PHI-13, isotype; c, d, f PHI-14. Bars = 1 cm

Full size image
Fig. 3
figure 3

Microscopic features of Kaziboletus rufescens (HKAS 74706). a Basidia with basidioles; b Basidiospores; c Hymenial cystidia (Cheilo- and pleurocystidia); d Epithelium pileipellis; e Caulocystidia

Full size image
Fig. 4
figure 4

SEM of smooth basidiospores of Kaziboletus rufescens (a–d HKAS 74706, holotype). Bars = a, b 5 μm. c, d 10 μm

Full size image

MycoBank No.: MB 833239.

Etymology: The epithet “rufescens” (Lat.) refers to the context which changes from white to pale red or reddish orange in some patches when exposed.

Typification: Bangladesh: Dhaka division, Gazipur, Bhawal National Park, latitude 24°45'N 90°50'E, elevation 20 m, 29 July 2011, Iqbal 304 (HKAS 74706, holotype; PHI-13, isotype).

Description: Basidiomata small to medium-sized. Pileus 30–60 mm, convex to hemispherical, sometimes with an irregular to wavy margin; surface dry, covered with pale reddish brown (9D7–8, 9C8) to pale brown (10C7–8) squamules when young, gray to grayish white (10B1, 10C1) at maturity, brownish gray when wet, becoming rimose to rugulose with age or during some stage of development; context up to 8 mm thick, solid, white to cream white, turning slightly pale red (8A4–5) to pale reddish orange (7A6–7) in some patches on exposure. Hymenophore tubulose, free, depressed around the stipe, white to creamy white, turning slightly brown when injured; tubes 7–10 mm long, pores angular or round up to 1 mm broad. Stipe 40–75 × 8–12 mm, central, cylindrical, attenuated upwards, scabrous, covered with reddish brown (when young) to gray brown (at maturity) squamules; conspicuous longitudinal striations present with fine cross lines from the apex to just above the base, the ridges somewhat anastomosing, sometimes sub-reticulate at the apex, the basal half more scabrous than the upper half; context 8 mm thick, solid, white to dull white, staining pale red to reddish orange in some patches, pale blue at stipe base in one specimen (Iqbal 807). Basal mycelium white. Odor and taste not distinctive.

Basidiospores [60/3/3] (15–)17–19(–21) × 5.5–6.5(–7) μm, elongated to cylindro-fusoid, thin-walled, inamyloid, smooth under light microscope and SEM, pale yellow (1A2–3) to vivid yellow (2A6–8) in H2O and 5% KOH. Basidia 30–36 (–45) × (9–)10–12 μm, narrowly clavate, clavate or broadly clavate, colorless in H2O and 5% KOH, thin-walled, 4-spored, occasionally 2- or 3-spored; sterigmata up to 3.5 μm long. Hymenophoral trama 100–130 μm wide, boletoid, hyphae cylindrical, 4–10 μm wide, hyaline. Cheilo- and pleurocystidia 50–70 × 10–15 μm, scattered, not abundant, fusoid to subfusoid, mucronate, somewhat lanceolate, hyaline, thin-walled, smooth, without encrustation. Pileipellis an epithelium, 100–130 μm thick, consisting of 2–3(–5) broadly clavate to pyriform to broadly ellipsoid inflated cells in chains; terminal cells 15–30 (–50) × 10–16 (–20) μm, with yellowish brown to brown vacuolar pigments; sometimes mixed with filamentous hyphae 3–5 μm wide, with brown vacuolar pigments. Stipe trama composed of parallel, cylindrical, hyaline hyphae up to 13 μm wide. Stipitipellis covered by a hymenium-like structure, composed of clavate to broadly clavate caulocystidia 16–35 × 6–11 μm, with yellowish brown to pale brown vacuolar pigmentation. Clamp connections absent in all tissues.

Habit and habitat: Mostly solitary in pure stands of S. robusta, and putatively forming ectomycorrhizae with S. robusta.

Known distribution: Currently known only from tropical Bangladesh, and possibly Malaysia (based on sequence data).

Additional specimens examined: Bangladesh. Dhaka division: Gazipur, Bhawal National Park, elevation 20 m, 5 Jul 2011, Iqbal 154 (HKAS 74705). Rangpur division: Dinajpur, Birol, Kaliganj Sal Baghan, elevation 60–70 m, 7 Aug 2013, Iqbal 807 (PHI-14).

Discussion

The presence of a coarsely scaly or scurfy stipe surface, smooth basidiospores, and fusoid cheilocystidia in the new genus Kaziboletus are similar to those of Leccinum, Leccinellum and Spongispora. However, the phylogenetic analyses presented here confirm that Kaziboletus cannot be placed in any of these genera. Furthermore, Leccinum and Leccinellum are most diverse in North America and Europe with most species host specific to the members of Betulaceae Gray, Ericaceae Durande, Fagaceae Dumort., Pinaceae Spreng. ex F. Rudolphi, and Salicaceae Mirb. (Corner 1972; Engel 1978; Singer 1986; Bresinsky and Besl 2003; den Bakker and Noordeloos 2005), while Kaziboletus is distributed in tropical South Asia with putative host specificity to S. robusta. Although some species of Leccinum have been reported from the tropics, for example, in Costa Rica by Halling (1999), Halling and Mueller (2003), Ortiz-Santana and Halling (2009); and in Southeast Asia by Corner (1972) and Horak (2011), their phylogenetic placements largely need to be explored. Furthermore, Kaziboletus differs from Leccinum and Leccinellum by the context that becomes reddish to pale reddish orange in patches when injured (except L. rugosiceps (Pk.) Singer and L. intusrubens (Corner) Høil), and the presence of longitudinal striations from the apex to just above the base. Spongispora, a monotypic genus originally described from Singapore, is easy to separate from the new genus by a coarsely reticulate stipe, a pileus context that turns pale brown on exposure, an interwoven trichodermial pileipellis, broadly ellipsoid to ovoid ornamented basidiospores, and putative association with Hopea odorata Roxb. A key is provided for convenience in comparison among the four morphologically closely related genera Kaziboletus, Leccinum, Leccinellum, and Spongispora.

Kaziboletus rufescens resembles Leccinum rugosiceps because of its rugose pileus, pink or pale red context when injured, and a hymeniform or epithelium pileipellis (Halling and Mueller 2003). However, the latter taxon, originally described from eastern USA, has a large pileus up to 150 mm broad, bright yellow to wax yellow tubes, non-staining stipe context, and association with Quercus sp. (Smith and Thiers 1971; Halling and Mueller 2003). Furthermore, phylogenetically, L. rugosiceps forms a distinct lineage apart from the Kaziboletus clade. Morphologically, Leccinellum albellum (Peck) Bresinsky & Manfr. Binder also resembles K. rufescens, but the species is distinguished from the latter taxon by its context that never turns pale reddish to reddish orange when exposed, abundant hymenial cystidia, and distribution predominantly within the USA (Smith and Thiers 1971).

The tropical Asian Leccinum spp., L. borneense (Corner) E. Horak and L. intusrubens, originally described from Malaysia (Corner 1972), can also be distinguished from K. rufescens. Leccinum borneense has an ixocutis as a pileipellis, a pale yellow context and tubes that quickly become blue on bruising, and basidiospores 13–15 × 4–4.5 μm in size. Leccinum intusrubens has a context that quickly turns blood red or saffron on exposure and then slowly turns blackish, a pileipellis composed of 2–3 septate moniliform hyphal end cells, and proportionally shorter and narrower basidiospores 11–14 × 5–5.5 μm. No molecular data are available for these Malaysian species of Leccinum, and their phylogenetic position can therefore not be assessed in the current study.

It is interesting to note that two nrLSU sequences generated either from mycorrhizal root tips (GQ268714, labelled as uncultured bolete) or basidioma (KY091033, labelled as Boletaceae sp.) from Sarawak, Malaysia are identical to those of K. rufescens (Fig. 1), suggesting that Kaziboletus may also occur in Southeast Asia. Surprisingly, however, apart from these two nrLSU sequences from Malaysian samples, the nrLSU sequences of K. rufescens vary significantly (i.e., matched only 85–86.78%) with the known species of Leccinum and other boletoid genera available in GenBank. Prior to proposing the new genus Kaziboletus, we carefully consulted the literature of Malaysian boletes (Corner 1972, 1974; Horak 2011), and determined that neither of those Corner boletes appear similar to the new bolete described here. Finally, it may be concluded that though several species of Leccinum are described from tropical regions such as Southeast Asia, Africa and South America or Costa Rica, none of these are morphologically identical to the South Asian K. rufescens. In accordance with the guidelines set out by Vellinga et al. (2015), Kaziboletus has enough morphological differences, host preference, and phylogenetic support to be considered as a distinct genus in Boletaceae.

The molecular phylogenetic results of Wu et al. (2018) distinguished 14 genera in the subfamily Leccinoideae. Shortly after, Khmelnitsky et al. (2019) erected another new Leccinoideae genus, Ionosporus, which is phylogenetically closely related to the monotypic genus Borofutus Hosen & Zhu L. Yang (Fig. 1). The erection of Kaziboletus gen. nov. in this study raises the number of genera in the subfamily Leccinoideae to 16, with all genera appearing to be monophyletic except Leccinum and Leccinellum (Fig. 1).

Key to the genera morphologically similar to Kaziboletus

  1. 1.

    With tropical and temperate distribution .…..……………2

    1* With tropical distribution only....................................…… 3

  2. 2.

    Hymenophoral surface yellow when young, pileus context yellowish to white then quickly blackeing when injured; pileipellis a palisade trichoderm; ectomycorrhizal symbioses mainly with Fagaceae and Betulaceae.....……...Leccinellum

    2* Hymenophoral surface diverse colors but not yellow when young; pileus context whitish to light gray to grayish; pileipellis diverse- trichoderm, hymeniform/epithelium; ectomycorrhizal symbioses with Pinaceae, Fagaceae, Betulaceae, Caesalpinoid legumes and Dipterocarpaceae……....………………………...Leccinum

  3. 3.

    Pileus context whitish but faintly staining pale brown when injured; stipe almost reticulate and covered by coarse scurfy squamules, basidiospores not smooth, with irregular clefts and warts under SEM; pileipellis a trichoderm composed of cylindrical cells; ectomycorrhizal symbioses with Hopea odorata Roxb., known from Singapore….............…......………..... Spongispora

    3* Pileus context white to creamy white but turning slightly pale red to pale reddish orange when exposed; stipe surface with sparse coarse, scurfy squamules; basidiospores smooth; pileipellis an epithelium composed of 2–3(5) inflated cells; putatively ectomycorrhizal symbiosis with Shorea robusta; known from Bangladesh and possibly Malaysia..……….........….......….... Kaziboletus