2022 Volume 63 Issue 3 Pages 102-117
We describe two new species of resupinate Sistotrema sensu lato (Cantharellales) collected in Japan: S. flavorhizomorphae and S. chloroporum. Both species have urniform basidia with more than four sterigmata and monomitic hyphal system, oil-rich hyphae in subiculum, which is typical for this genus. Sistotrema chloroporum is characterized by poroid hymenophore partly yellowish-green, basidia 4-6-spored, medium-sized basidiospores (4.5-6.5 × 3.5-6 µm), and broadleaf forest habitat. Sistotrema flavorhizomorphae is characterized by hydnoid-irpicoid hymenophore, bright yellowish rhizomorphs, basidia 6-8-spored, small basidiospores (3-3.5 × 2.5-3 µm), and pine forest habitat. Phylogenetic trees inferred from the fungal nrDNA ITS and LSU and the rpb2 sequences supported that both species were distinct and grouped with other ectomycorrhizal Sistotrema and Hydnum species, but their generic boundary was unclear. Mycorrhizae underneath basidiomes of both species were identified and described via molecular techniques. Mycorrhizae of S. chloroporum have similar characteristics to those of other Sistotrema s.l. and Hydnum species, i.e., S. confluens and H. repandum, whereas S. flavorhizomorphae has a distinct morpho-anatomy, for example, a distinct pseudoparenchymatous mantle. Comprehensive characterizations of basidiomes and mycorrhizae improve the taxonomic analysis of mycorrhizal species of Sistotrema s.l.
The genus Sistotrema Fr. belongs to the family Hydnaceae, order Cantharellales, and phylum Basidiomycota (Moncalvo et al., 2006; Hibbett et al., 2014). The genus is morphologically characterized by resupinate or pileate-stipitate, soft basidiomes, smooth, grandinioid, hydnoid, or poroid hymenophore provided with various textures (pellicular, membranaceous, or ceraceous), a monomitic hyphal system with thin-walled and clamped hyphae (or simple septate in few species) with oily inclusions, urniform basidia usually with 6-8 sterigmata, and smooth, thin-walled, and varying-shaped basidiospores (Eriksson, Hjortstam, & Ryvarden, 1984; Bernicchia & Gorjón, 2010). The key characters to identify this genus are the presence of urniform basidia with irregular numbers of sterigmata and hyphae with oily inclusions. Sistotrema confluens Pers. is the type species of the genus (Donk, 1960; Eriksson et al., 1984).
According to the Index Fungorum database (http://www.indexfungorum.org/names/names.asp; accessed on 20 Jul 2021), the genus Sistotrema currently includes more than 60 validated species. Most of them are wood-decaying fungi with resupinate and smooth to grandinioid hymenophores (e.g., Eriksson et al., 1984; Bernicchia & Gorjón, 2010). However, eight species of Sistotrema including the type species have a more developed hymenophore with a hydnoid, irpicoid to a poroid surface. While S. confluens and S. subconfluens L.W. Zhou have pileate-stipitate basidiome and a poroid to irpicoid hymenophores (Eriksson et al., 1984; Ryvarden & Gilbertson, 1994; Zhou & Qin, 2013; Bubner, Morgner, Stark, & Münzenberger, 2014), S. alboluteum (Bourdot & Galzin) Bondartsev & Singer, S. albopallescens (Bourdot & Galzin) Bondartsev & Singer, S. brunneolum Spirin & Zmitr., and S. dennisii Malençon have a poroid-resupinate hymenophore (Eriksson et al., 1984; Malençon, 1977; Spirin & Zmitrovich, 2007; Kout, 2008), and that of S. muscicola (Pers.) S. Lundell and S. raduloides (P. Karst.) Donk have a hydnoid to irpicoid and resupinate hymenophore (Eriksson et al., 1984; Bernicchia & Gorjón, 2010). Among these, DNA sequences of the internal transcribed spacer (ITS) region and 28S large subunit (nrLSU) of fungal nrDNA from a type material of S. subconfluens are available from the NCBI GenBank database (Zhou & Qin, 2013). The ITS and/or nrLSU sequences of untypified specimens of S. alboluteum, S. albopallescens, S. confluens, S. muscicola, and S. raduloides were provided in previous studies (Larsson, Larsson, & Kõljalg, 2004; Nilsson et al., 2006; Münzenberger et al., 2012; Kotiranta & Larsson, 2013; Bubner et al., 2014). No sequence data of S. brunneolum and S. dennisii are currently available.
Recent phylogenetic studies demonstrated the polyphyly of the genus Sistotrema; species of Sistotrema are complexly nested within or between other genera (e.g., Burgella, Clavulina, Membranomyces, and Multiclavula) in the cantharelloid clade, whereas S. alboluteum, S. albopallescens, S. confluens, S. muscicola, S. subconfluens together with all species of Hydnum examined to date form a monophyletic lineage (Larsson et al., 2004; Binder et al., 2005; Moncalvo et al., 2006; Hibbett et al., 2014; Lawrey, Sikaroodi, Gillevet, & Diederich, 2020). Münzenberger et al. (2012) named this monophyletic lineage as “core Sistotrema/Hydnum group” because it contains the type species of both Hydnum and Sistotrema. The main characteristics of core Sistotrema/Hydnum group are a well-developed hymenophore and the ectomycorrhizal status (Moncalvo et al., 2006; Nilsson et al., 2006; Münzenberger et al., 2012). The generic boundary between Hydnum and Sistotrema have yet to be determined, and pileate-stipitate species of Sistotrema (S. confluens and S. subconfluens) form a distinct clade within the core Sistotrema/Hydnum group. Due to this situation, we regarded these two species as Sistotrema sensu stricto (s.s.) and the remaining Sistotrema forming a resupinate hymenophore as “Sistotrema sensu lato (s.l.)” here.
The characterization of mycorrhizae at some point could be useful when trying to determine the ecological status of a fungus, and provides useful information for fungal taxonomic analysis (e.g., Agerer, 2006; Moyersoen & Weiß, 2014; Mrak et al., 2017; Endo et al., 2020). There are more than a thousand descriptions to date (De Roman, Claveria, & De Miguel, 2005; Agerer, 2006; Agerer & Rambold, 2004-2021). The ectomycorrhizal morphologies of only five Sistotrema s.l. species has been determined via molecular identification of field-sampled mycorrhizae or in vitro synthesis (Nilsson et al., 2006; Di Marino, Scattolin, Bodensteiner, & Agerer, 2008; Münzenberger et al., 2012; Bubner et al., 2014). Furthermore, descriptions of mycorrhizae are available for only three Sistotrema s.l. species, including S. aff. albopallescens, S. confluens, and an unidentified species (Di Marino et al., 2008; Münzenberger et al., 2012; Bubner et al., 2014). If due to the basidiomes condition, the Sistotrema s.l. species cannot be morphologically identified, and we obtain morpho-anatomical data on its mycorrhiza, these might enable their identification.
In this study, we collected 15 basidiomes of resupinate Sistotrema s.l. species that would belong to the core of species in the Sistotrema/Hydnum group. The basidiomes of such species have poroid, irpicoid, or hydnoid hymenophore and morphologically resemble S. muscicola or S. alboluteum. Because the basidiomes did not morphologically match any other Sistotrema s.l. species, we conducted phylogenetic analyses with the internal transcribed spacer region (ITS), 28S large subunit (nrLSU) of fungal nrDNA, and fragment gene of the RNA polymerase II second largest subunit (rpb2) to clarify the phylogenetic relationships among Hydnum and mycorrhizal Sistotrema. To characterize the species morphologically and ecologically, we described both, their basidiomes and ectomycorrhizae.
Fifteen basidiomes specimens were collected from 10 sites in Japan during 2019-2021. From 9 specimens, the root system underneath each basidiome was also sampled. Each basidiome was used for culture isolation, air-dried at 45 °C for 1-2 d, and then deposited as voucher specimen in TUMH herbarium from Tottori University (Thiers B. [continuously updated] Index Herbariorum: a global directory of public herbaria and associate staff. New York Botanical Garden`s Virtual Herbarium. http://sweetgum.nybg.org/science/ih/).
To obtain living culture strains monosporous isolations was carried out following to Sugawara et al. (2019). When enough basidiospores were not fallen, we directly isolated cultures from fallen spores on culture medium (multisporous isolation). We used MNC gellun-gum medium (Yamada & Katsuya, 1995; Sugawara et al., 2019) for spore isolations and MNC agar medium for maintaining fungal cultures.
We also used two cultures (TUFC 101478 and TUFC 31972) of mycorrhizal Sistotrema deposited in the Fungus/Mushroom Resource and Research Center (FMRC) in Faculty of Agriculture, Tottori University for phylogenetic analysis (Supplementary Table S1).
2.2. Morphological observations of mycorrhizae and basidiomesFresh basidiomes were observed under 10× or 20× magnification using a dissecting microscope (SMZ1500, Nikon Imaging, Tokyo, Japan) equipped with a digital camera (DS-Fil-L2, Nikon Imaging). The scales of the photographs were calibrated using objective micrometer; then, the macroscopic characters of the basidiomes (e.g., basidiome thickness, spine length, and pore sizes) were measured using PhotoRuler ver. 1.1.3 software (developed by S. Onishi in 2010; http://inocybe.info). Colors were determined based on the Methuen Handbook of Color (e.g., 4A2; Kornerup & Wanscher, 1978). Microscopic features of fresh or dried materials were observed under 3% (w/v) KOH solution, Melzer's reagent (IKI), and cotton blue (CB) using a differential interference contrast (DIC) microscope (Eclipse 80i, Nikon Imaging). Basidiospores, basidia, sterigmata, and hyphal characters were photographed using a digital camera (DS-Fil-L2, Nikon Imaging), and their sizes were measured using PhotoRuler ver. 1.1.3 software. The length, width, and length/width ratio (Q) of 30 typical basidiospores were determined for each specimen. Basidiospore length and width are presented as the 90% ranges of values and infrequent minimum/maximum values were shown in brackets. Q values are represented as (a)b-c(d), where a and d are the 90% ranges of the values and b and c are the mean ± standard deviation. Mean values of Q (Qm) were calculated for each specimen and shown as the broadest ranges. Here, “n = e/f” represents the number of measured specimens (e) and the total number of measured basidiospores (f). The sizes of basidia, sterigmata, hyphae in subiculum and rhizomorphs, and ampullate inflation were described as the range of maximum and minimum values with outliers in brackets. All measurements were made by a single author, R. Sugawara.
Ten mycorrhizae were sampled from each root system, and their morphology was examined following Endo et al. (2020) with minor modifications. Because one basidiome (TUMH 64396) was associated with both an angiosperm (Quercus serrata Murray) and a gymnosperm (Pinus densiflora Siebold & Zucc.), we sampled mycorrhizae from the root system of each host plant. All mycorrhizae were identified by following rhizomorph or emanating hyphae connecting to the basidiome and analyzed after washing with tap water. Morpho-anatomy of fresh mycorrhizae was inspected by dissecting and DIC microscopy, and for the latter, a root tip was sectioned by hand and mounted on a glass slide in lactic acid. The plan view of rhizomorphs, emanating hyphae, mantle, and the cross-section of mycorrhizal root tips were described using terminologies according to Agerer (1991), Agerer (2001), and Agerer and Rambold (2004-2021). Plant species of each mycorrhiza were identified by their morpho-anatomy and forest habitat. If the condition of mycorrhizae was good, the fungal symbiont was isolated from mycorrhiza using the procedure described in Endo et al. (2020). The tissue fragments of inspected mycorrhizae were used for DNA extraction and subsequent molecular identification of fungal species.
2.3. DNA extraction, PCR amplification, and sequencingGenomic DNA was extracted from a small hymenophore fragment of dried basidiomes following Izumitsu et al. (2012). Genomic DNA of mycorrhiza was extracted using the cetyltrimethylammonium bromide (CTAB) method (Gardes & Bruns, 1993). The ITS region of fungal nuclear DNA was amplified from both basidiomes and mycorrhizae via PCR using Dream Taq DNA Polymerase (Thermo Fisher Scientific, Waltham, MA, USA) and the primers ITS1F (forward) and ITS4 or LBW (reverse) (Gardes & Bruns, 1993; White, Bruns, Lee, & Taylor, 1990; Tedersoo et al., 2008). Additionally, the nrLSU region was amplified from each basidiome using the primers CTB6 (forward) and LR5F (reverse) (Garbelotto et al., 1997; Tedersoo et al., 2008). We also amplified a portion of rpb2 gene from the representative specimens using the fRPB2-5F (forward) and RPB2-b7.1R (reverse) (Liu et al., 1999; Matheny, 2005). The PCR products were directly sequenced with the same primers, and bidirectional sequences were assembled with the ClustalW program (Thompson, Higgins, & Gibson, 1994) to form contiguous sequences.
Because we could not identify the genus level of four angiosperm-mycorrhiza, we molecularly identified them based on the plastid DNA barcode gene ycf1 (Dong et al., 2015; Pang et al., 2019). The primers ycf1bF (forward) and ycf1bR (reverse) (Dong et al., 2015) were used for PCR amplification and subsequent sequencing analysis. Each assembled sequence was compared to the NCBI GenBank database using NBLAST and the best taxonomic hit per sequence was used for molecular identification.
2.4. Phylogenetic analysesThe phylogenetic analyses of the ITS and nrLSU datasets were performed independently. The nrLSU phylogeny was analyzed to reveal higher phylogenetic relationships of the investigated fungi. The nrLSU dataset included 17 sequences from this study and 83 sequences labeled as “Sistotrema” in the NCBI GenBank database. Taxon sampling of other genera in the order Cantharellales was done following Lawrey et al. (2020) and Masumoto and Degawa (2020): species of the genera Cantharellus and Craterellus were excluded due to their distinct divergence, and Cerinomyces crustulinus (Bourdot & Galzin) G.W. Martin, Platygloea disciformis (Fr.) Neuhoff, and Tilletiaria anomala Bandoni & B.N. Johri were selected as the outgroup. The nrLSU sequences list is shown in Supplementary Table S1 and they were downloaded from GenBank. The dataset was aligned using MAFFT online (Madeira et al., 2019) and then manually corrected using MEGA 7 software (Kumar, Stecher, & Tamura, 2016). The substitution model was selected using PartitionFinder Ver. 2.1.1 software (Lanfear, Frandsen, Wright, Senfeld, & Calcott, 2017). A maximum likelihood (ML) tree was inferred using raxmlGUI Ver. 2.0 software (Stamatakis, 2014; Edler, Klein, Antonelli, & Silvestro, 2021), using a GTRGAMMAI substitution model and a nonparametric bootstrap analysis with 1000 replicates (MLBS). The Bayesian inference posterior probability (BPP) of each branch was generated using MrBayes Ver. 3.2.7 software (Ronquist et al., 2012). Two runs with four Markov chain Monte Carlo (MCMC) iterations were performed for 2,000,000 generations and their convergence was confirmed using Tracer Ver. 1.7.2 software (Rambaut, Drummond, Xie, Baele, & Suchard, 2018). Trees were kept for every 100 generations, and the last 75% of trees were used to calculate the 50% majority-rule consensus topology and to determine the BPP for individual branches. The resulting phylogenies were plotted using FigTree Ver. 1.4.4 software (http://tree.bio.ed.ac.uk/software/figtree/).
The ITS dataset was built to delimitate phylogenetic species among related taxa within the core Sistotrema/Hydnum group. The generated mycorrhizal sequences in this study were added to the dataset to phylogenetically identify them. Taxon sampling was focused on complete (ITS1 + 5.8S + ITS2) or partial (5.8S + ITS2) ITS sequences in the NCBI GenBank and UNITE database and included the following: voucher specimens and their mycorrhizae of mycorrhizal-resupinate Sistotrema s.l. (S. alboluteum, S. albopallescens, and S. muscicola), a hydnoid fungus of saprotrophic Sistotrema s.l. (S. raduloides), type sequences of several representatives of Hydnum; mycorrhizae closely related to novel species (> 90% on NBLAST search), MH854646 [Minimedusa polyspora (Hotson) Weresub & P.M. LeClair] and LC507463 [Multiclavula vernalis (Schwein.) R.H. Petersen] as an outgroup. Although S. confluens is the type species of the genus Sistotrema, we excluded its sequences from the dataset to avoid long-branch attraction caused by distinct genetic divergence from other Sistotrema/Hydnum species (see Supplementary Fig. S1). The ITS sequences list is shown in Supplementary Table S2 and they were downloaded from GenBank. The sequence alignment and phylogenetic analyses of the ITS dataset were performed as described for nrLSU dataset. The alignments and phylograms were submitted in TreeBASE (http://treebase.org/: S28838).
The phylogenetic relationships of the core Sistotrema/Hydnum group were analyzed using a fragment of the rpb2 gene. We used sequences of species of Hydnaceae, i.e., Cantharellus, Clavulina, Craterellus, Hydnum, and Sistotrema as ingroups, and Botryobasidium as an outgroup (Supplementary Table S1). We did not set the partition schemes among codon positions according to the analysis of these taxa in Moncalvo et al. (2006). Other settings for alignment, model selection, and phylogenetic analyses matched those described for the nrLSU and ITS dataset.
Sistotrema chloroporum R. Sugaw., N. Maek., Sotome & N. Endo, sp. nov. Figs. 1, 2, 3.
MycoBank: MB 841341.
Diagnosis: This species morphologically resembles S. alboluteum but differ by its greenish hymenophore, slightly elliptic basidiospores (4.5-6.5 × 3.5-6 µm), and basidia with 4-6 sterigmata.
Type: JAPAN, Tottori Pref., Tottori City, Kakuji, 100 m a.s.l., on decayed wood, bark, and leaves of broadleaf trees in broadleaf forest of Quercus serrata, Q. acutissima Carruth., Q. variabilis Blume, Castanopsis sieboldii (Makino) Hatus., and Carpinus tschonoskii Maxim., 19 Nov 2020, R. Sugawara, SuR20201119-206 (holotype, TUMH 64399); ex-holotype culture, TUFC 101896 (monosporic strain)]; ex-holotype slides, mycorrhiza SuR20201119-206A (mycorrhiza with Q. serrata), plate nos. 2021-RS-SC01, SC02, and SC03.
Gene sequences ex-holotype: LC642034 (ITS), LC642057 (nrLSU), LC667369 (rpb2) from basidiomes; LC642035 (fungal ITS), LC642071 (plant ycf1) from mycorrhiza.
Etymology: From chlorus and porum, referring to the greenish color of the poroid hymenophore.
Japanese name: midoriana-tsubo-take.
Basidiomes resupinate, effused, loosely adnate, 1.5-2(-2.5) mm thick, soft, cottony, fragile. Hymenophore poroid with angular to round, more or less labyrinthiform, 2-4 pores/mm, maximum pore size of each specimen 0.3-0.5(-0.7) mm diam; dissepiment thin, sometimes lacerate when mature; white to pale cream (4A1-3) when young, pale yellow to pale ocher (4A5-6) in older parts, partly stained yellowish green to deep green (28C8-E8), not reacted with 3% KOH solution, on dried specimen mostly pale ocher to ocher (4A6-8), white to pale cream (4A1-3) on thin hymenophore, partly greenish (28C8-E8). Subiculum thin, up to 0.2 mm thick, fragile, soft, somewhat arachnoid, white (4A1); margin fragile, thinning out, byssoid, when young arachnoid, white, partly stained yellowish green to deep green. Rhizomorphs absent or few, if present very loose, subhyaline to white (4A1), non-continuous between hymenophore to mycorrhizae.
Hyphal system monomitic; hyphae thin-walled, regularly clamped, 2.5-4 µm wide, infrequently inflated near septa, 5-7 µm wide, frequently containing hyaline to slightly yellowish oily droplets, oily droplets cyanophilous with CB; subhymenial hyphae densely branched and intertwined. Basidia usually urniform, but sometimes pleural near the pore surface, ovoid when young, 15.5-27(-31) × 7-9(-12) µm [excluding pleural basidia, n = 5/52], shortest width 4.5-6.5(-10) µm wide, clamped, containing hyaline oily droplets, with 4 to 6 (mainly 6) sterigmata measuring 2.5-5 µm long. Basidiospores subglobose to broadly ellipsoid, hyaline, smooth, thin-walled, 4.5-6.5 × (3-)3.5-6 µm, Q = 1.00-1.45(-1.60), Qm = 1.14-1.24 [on average, 5.3 × 4.4 µm, Q = 1.20; n = 6/180], CB -, IKI -. Cystidium absent.
Mycorrhizae present in organic- to A-layer of clay soil, within whitish hyphal mat, abundant. Mycorrhizal system pyramidal to irregularly pinnate, up to 3-ordered; main axis 4 mm long, 0.25-0.3 mm wide; ramified tips up to 1 mm long, side axis 1.4-1.8 mm long, 0.2-0.25 mm wide. Mycorrhizal surface cream to pale ocher, slightly cottony, more or less hydrophobic, emanating hyphae abundant, subhyaline to white; apex of tip few encompassed by mantle, more whitish. Rhizomorphs absent or rarely present, if present, woolly, loose, white, not connected to mantle. Exploration type short-distance (Agerer, 2001). Rhizomorphs 40-80 µm wide, composed of collapsed hyphae sometimes ampule-like inflated near septum, without differentiation of vessel hyphae. Emanating hyphae cylindric, thin-walled, regularly clamped, frequently containing hyaline oily droplets, 2.5-3.5 µm wide, frequently inflated at septa (ampullate inflation), 6-7 µm wide, anastomoses of hyphae frequent, closed by clamp. Mantle layers composed of plectenchymatous hyphae, Type B or A (Agerer, 1991, 2006), up to 25 µm thick; outer layer loose plectenchymatous, with occasionally ring-like hyphal arrangement, hyphae cylindric, thin-walled, clamped, containing hyaline oily droplets, 2.5-4 µm wide; middle layer plectenchymatous, frequently branching as Y-shape, hyphae cylindric to slightly irregularly-shaped, thin-walled, clamped or simple septate, containing hyaline oily droplets, 2.5-5.5 µm wide; inner layer densely plectenchymatous with pseudoparenchymatous cells, hyphae complexly nested, cylindric to irregularly-shaped, thin-walled, more or less containing hyaline oily droplets, 2.5-6 µm wide, partly inflated up to 9 µm wide. Hartig net present, palmetii, 1.5-3.5 µm wide between epidermal cells. Cystidium absent in all sections.
Ecology and distribution: Substrate humus of broadleaves, decayed wood, ground on slope, or small tunnel with clay soil. Occurring in broadleaf-dominated forests with Quercus, Castanopsis, Castanea, and/or Carpinus, in Oct to Dec, in temperate climate. Mycorrhizae underneath basidiomes were associated with Castanopsis sieboldii, Quercus serrata, or Carpinus sp. (Table S3). This species also formed mycorrhizae with Pinus densiflora root nearby those with Q. serrata. JAPAN: Nagano, Tottori, and Okayama Pref.
Additional specimens/mycorrhizae examined: JAPAN. Nagano Pref.: Kamiina-gun, Tatsuno Town, Yokokawa, 980 m a.s.l., on humus near individuals of Fagus crenata Blume, Castanea crenata Siebold & Zucc., Q. mongolica subsp. crispula (Blume) Menitsky, and Carpinus sp., 22 Nov 2020, R. Sugawara SuR20201122-002 [basidiome TUMH 64400; mycorrhiza SuR20201122-002A (with Carpinus sp.)]; Okayama Pref.: Tomata-gun, Kagamino Town, Onbarakogen, 760 m a.s.l., on fallen oak-leaves under P. densiflora in mixed forest dominated by Q. serrata and C. crenata, 31 Oct 2020, R. Sugawara SuR20201031-105 [basidiome TUMH 64396; mycorrhizae SuR20201031-105A (with Q. serrata), SuR20201031-105B (with P. densiflora)]; Tottori Pref.: Tottori City, Kokufu Town, Okamasu, 60 m, on slope of clay soil in broadleaf forest dominated by C. sieboldii and Q. glauca Thunb. including few individuals of P. densiflora, 14 Nov 2020, R. Sugawara SuR20201114-302 [basidiome TUMH 64398; mycorrhiza SuR20201114-302A (with C. sieboldii)]: Higashi-machi, Kyushosan, 150 m a.s.l., on humus in broadleaf forest dominated by C. sieboldii and Carpinus tschonoskii Maxim., 12 Dec 2020, K. Nagamune NaK20201212-08 (TUMH 64395): Kurayoshi City, Utsubukiyama, 130 m, on slope of clay soil in mixed forest of C. sieboldii, Q. serrata and P. densiflora, 7 Nov 2020, R. Sugawara SuR20201107-407 (TUMH 64397).
Notes: Description of mycorrhizae were based on oak root tip (Table S3) collected from underneath of specimen TUMH 64399 (holotype) with Q. serrata. Additional descriptions from mycorrhizae with other specimens were shown in Tables 1 and 2. The result of the NBLAST search in plant DNA obtained from each mycorrhizal tip was shown in Supplementary Table S3.
| Fungal species | ECM code | Host plant | Systems | Main color of mycorrhiza | Hydrophobicity | Texture | Emanating hyphae | Rhizomorphs |
| Sistotrema chloroporum | SuR20201031-105A | Q. serrata | Pyramidal | Deep cream to pale orange | Hydrophobic | Cottony | White: Mat-like | Absent |
| SuR20201031-105B | P. densiflora | Dichotomous | Cream to pinkish orange | Hydrophobic | Cottony-stringy | White: Mat-like | Absent | |
| SuR20201114-302A | C. sieboldii | Irregularly ramified | Deep orange | Hydrophobic | Cottony | White: Mat-like | Present: White, non-continuous | |
| SuR20201119-206A | Q. serrata | Irregularly ramified | Deep cream | Hydrophobic | Cottony | White: Mat-like | Present: White, non-continuous | |
| SuR20201122-002A | Carpinus sp. | Irregularly ramified | Greyish cream | Hydrophobic | Cottony | White: Mat-like | Absent | |
| S. flavorhizomorphae | NS191113-2A | P. thunbergii | Unramified | Deep orange to brownish orange | Hydrophobic | Slightly woolly | Yellow | Present: Yellow |
| SuR20201025-015A | P. densiflora | Dichotomous | Deep orange to brownish orange | Hydrophobic | Woolly | Yellow | Present: Yellow | |
| SuR20201110-003A | P. thunbergii | Unramified | Deep orange to brownish orange | Hydrophobic | Woolly | Yellow | Present: Yellow | |
| SuR20201212-002A | P. thunbergii | Dichotomous | Reddish brown | Hydrophobic | Slightly woolly | Yellow | Present: Yellow |
Bold shows ex-type material.
| Fungal species | ECM code | Rrhizomorphs and emanating hyphae characteristics | Mantle characteristicsa | ||||||||
| Rhizomorph type | Encrustation of hyphae | Hyphal width | Ampullate width | Outer Mantle type | Outer mantle | Middle to inner mantle | Mantle thickness | ||||
| Hyphal arrangement | width | Hyphal arrangement | width | ||||||||
| Sistotrema chloroporum | SuR20201031-105A | - | Absent | 2.3-4.5 µm | 5-6.5 µm | Type A | PL: Ring-like | 2-4 µm | PL & PS | 2-5 µm | 20-25 µm |
| SuR20201031-105B | - | Absent | 2.5-4 µm | 5-7.5 µm | Type A | PL: Ring-like | 2.5-4 µm | PL & PS | 2.5-4.5 µm | 25-32 µm | |
| SuR20201114-302A | Type C | Absent | 2.5-3.5 µm | 5.5-7.0 µm | Type B | PL: Ring-like | 2.5-3.5 µm | PL & PS | 2.5-5.5 µm | 20-35 µm | |
| SuR20201119-206A | Type C | Absent | 2-3.5 µm | 6-7 µm | Type A | PL: Ring-like | 2.5-5.5 µm | PL & PS | 2.5-6 µm | 20-25 µm | |
| SuR20201122-002A | - | Absent | 2-3.5 µm | 5.5-8 µm | Type B | PL | 2-4 µm | PL & PS | 2.5-4.5 µm | 25-30 µm | |
| S. flavorhizomorphae | NS191113-2A | Type C | Present: Yellow | 2-3.5 µm | 5.5-9 µm | Type P | PL | 2.5-4 µm | PS: Angular | < 12 µm | 18-20 µm |
| SuR20201025-015A | Type C | Present: Yellow | 2-3 µm | 5-8 µm | Type P | PL | 2-3.5 µm | PS: Angular | < 11 µm | 23-30 µm | |
| SuR20201110-003A | Type C | Present: Yellow | 2-2.5 µm | 5.5-6.5 µm | Type P | PL | 2-2.5 µm | PS: Angular | < 12 µm | 20-27 µm | |
| SuR20201212-002A | Type C | Present: Yellow | 2-4.5 µm | 4.5-8 µm | Type P | PL | 2.5-4 µm | PS: Angular | < 13 µm | 25-30 µm | |
a: PL plectenchymatous; PS pseudoparenchymatous. Rhizomorph type and outer mantle type are following to Agerer (1991). Bold shows ex-type material.
R. Sugaw., N. Shiras., N. Maek. & N. Endo, sp. nov. Figs. 4, 5, 6.
MycoBank MB 841343.
Diagnosis: This species is characterized by hydnoid to irpicoid hymenophoral surface, subglobose, small basidiospores (3-3.5 × 2.5-3 µm), and yellowish rhizomorphs composed of hyphae encrusted with yellowish crystalloid materials.
Type: JAPAN, Tottori-Pref., Tottori City, Hamasaka, near Tottori Sand Dune, ca. 50 m a.s.l., on dead bark of Pinus thunbergii in P. thunbergii pine forest, 12 Dec 2020, R. Sugawara SuR20201212-002 (holotype, TUMH 64409); ex-holotype cultures, TUFC 101894 (monosporic strain), TUFC 101895 (secondary mycelial strain isolated from mycorrhiza)]; ex-holotype slides, SuR20201212-002A (mycorrhiza with P. thunbergii), plate nos. 2021-RS-SF01, SF02, SF03, and SF04.
Gene sequences ex-holotype: LC642049 (ITS), LC642067 (nrLSU), LC667371 (rpb2) from basidiome; LC642050 (fungal ITS) from mycorrhiza.
Etymology: From flavus and rhizomorphae, referring to the yellowish rhizomorph.
Japanese name: kihari-takotsubo-take.
Basidiomes resupinate, effused, pellicular, easily detachable, soft, fragile. Hymenophore hydnoid to irpicoid; spines conical to spathulate, white, in maximum 0.6-1.2 mm long, crowded, usually 10-14 spines/mm2; white to pale cream (4A1-2), when dried cream to pale ocher (4A3-5). Subiculum thin, white; margin smooth, white (4A1), thinning out, fibrillose to byssoid, when young arachnoid. Rhizomorphs cylindric, stipe-like, 0.1-0.4 mm wide, vertically producing through the hymenophore or margin, directly connected to the subicular hyphae and mycorrhiza, white, pale cream to cream (4A1-3) in upper side, pale yellow to bright yellow (3A7-8, 4A8) in lower side in humus or soil.
Hyphal system monomitic; hyphae cylindric, thin-walled, regularly clamped, 2-4.5 µm wide, rarely inflated at septa, 5-9 µm wide, frequently containing hyaline to slightly yellowish oily droplets, oily contents cyanophilous with CB; subhymenial hyphae densely branched and intertwined; apex of spines composed of cylindric hyphae. Rhizomorphs on hymenophore dense, composed of clamped hyphae corresponding to subiculum, without basidia and cystidium. Basidia urniform, 10.5-20 × 5-6.5(-8) µm [n = 7/74], shortest width (2.5-)3-5(-6) µm wide, clamped, containing hyaline oily droplets, with 6 to 8 sterigmata, ovoid when young, sterigmata 2-3.5(-5) µm long. Basidiospores globose to ovoid, hyaline, smooth, thin-walled, 3-3.5 × (2-)2.5-3 µm, Q = (1.00-)1.05-1.35(-1.40), Qm = 1.13-1.23 [on average, 3.2 × 2.8 µm, Q = 1.18; n = 7/210], CB -, IKI -. Cystidium absent.
Mycorrhizae present in organic- to A-layer of sand soil, in small numbers. Mycorrhizal system dichotomous, 1-3-ordered; main axis between ramification 0.5-1 mm long, 0.2-0.3 mm wide. Mycorrhizal surface ocher to brown, partly silver-white, somewhat woolly, strongly hydrophobic, emanating bright-yellowish hyphae and rhizomorphs; apex of tip 0.1-0.2 mm wide, few encompassed by mantle, clear- to greyish-colored. Rhizomorphs woolly, bright yellow to orangish yellow. Exploration type medium-distance, fringe-subtype (Agerer, 2001). Rhizomorphs dense, 30-140 µm wide, without differentiation of vessel hyphae, ramarioid (Type C; Agerer, 1999), yellow to orange, partly hyaline, hyphae cylindric, thin-walled, regularly clamped, frequently containing hyaline to slightly yellowish oily droplets, encrusted with yellowish crystalloid materials, 2-4.5 µm wide, frequently inflated at septa (ampullate inflation), 4.5-8 µm wide, anastomoses of hyphae frequent, closed by clamp. Emanating hyphae has similar characters to rhizomorphs. Mantle layers composed of both plectenchymatous and pseudoparenchymatous hyphae, Type P (Agerer, 1991, 2006), up to 30 µm thick; outer layer loose plectenchymatous, without hyphal arrangement, hyphae cylindric, thin-walled, clamped, containing hyaline to slightly yellowish oily droplets, heavily encrusted with yellowish crystalloid materials, 2.5-4 µm wide; middle to inner layer pseudoparenchymatous composed of angular cells, hyphal cells thin-walled, containing few hyaline oily droplets, less than 20 µm long and 13 µm wide, without yellowish encrustation. Hartig net present, palmetii, 1-2.5 µm wide between cortical cells. Cystidium absent in all sections.
Ecology and distribution. Substrate humus of needle-leaves and dead bark of pine, decayed wood, pinecone, sometimes on ground with sandy or cray soil. Occurring in pine forest or mixed forest dominated with P. densiflora or P. thunbergii Parl., in Oct to Dec, in temperate climate. Mycorrhizae underneath basidiomes were associated with pine (P. densiflora or P. thunbergii). JAPAN: Tottori Pref. DNA sequences of uncultured ectomycorrhizae potentially being this species have been reported from Iwate Pref., Japan (GenBank accession no. AB251813; AB211250) and China (GenBank accession no. MN549483).
Additional specimens/mycorrhizae examined: JAPAN. Tottori Pref.: Tottori City, Mount Jubo, 630 m a.s.l., on sandy soil under P. densiflora in mixed forest of P. densiflora and Q. serrata, 25 Oct 2020, R. Sugawara SuR20201025-015 (basidiome TUMH 64402; mycorrhiza SuR20201025-015A): Hamasaka, Arid land Research Center, 20 m a.s.l., on decayed wood of red pine on sandy soil in P. thunbergii forest, 13 Nov 2019, T. Yamamoto NS191113-2 (basidiome TUMH 64401; mycorrhiza NS191113-2A): on humus of pine spines, 10 Nov 2020, R. Sugawara SuR20201110-003 (basidiome TUMH 64404; mycorrhiza SuR20201110-003A): on humus, dead bark, and spines of red pine, 19 Nov 2020, R. Sugawara SuR20201119-008 (TUMH 64406): near Tottori Sand Dunes, 50 m a.s.l., on humus of pine spines on sandy soil of P. thunbergii forest, 10 Nov 2020, R. Sugawara SuR20201110-103 (TUMH 64405): Mochigase Town, Misumiyama, on humus in mixed forest of P. densiflora and Q. serrata, 6 Dec 2020, R. Sugawara SuR20201206-102 (TUMH 64407): on dead bark of red pine in P. densiflora forest including several individuals of broadleaved trees, 6 Dec 2020, R. Sugawara SuR20201206-106 (TUMH 64408): Kurayoshi City, Utsubukiyama, 130 m a.s.l., on slope of clay soil in mixed forest of P. densiflora, Q. serrata and C. sieboldii, 7 Nov 2020, R. Sugawara SuR20201107-405 (TUMH 64403).
Notes: Description of mycorrhizae was based on P. thunbergii root tip collected from underneath of specimen TUMH 64409 (holotype) with P. thunbergii. Additional descriptions from mycorrhizae with other specimens were shown in Tables 1 and 2.
3.2. Phylogenetic analysesForty two new sequences of nrLSU (LC642053-LC642067, LC667373-LC667374), ITS (LC642027-LC642050), and rpb2 (LC667369-LC667372) were obtained. The respective number of distinct alignment patterns and the percentages of gaps and completely undetermined characters used in the RAxML analyses for each gene were as follows: 510 and 11.76% for nrLSU, 410 and 21.32% for ITS, and 431 and 6.12% for rpb2. The final ML optimization likelihood value (-lnL) in RAxML was 9978.10 for nrLSU, 4708.99 for ITS, and 9641.57 for rpb2. For the substitution models in MrBayes, we selected GTR+G+I for nrLSU and rpb2 and TVM+G+I for ITS. Each MrBayes analysis was completed with smaller values (< 0.01) of the average standard deviation of the split frequencies (ASDSF), and the -lnL values of two runs were 10118.84 and 10125.69 for nrLSU, 4793.44 and 4795.31 for ITS, and 9640.34 and 9640.19 for rpb2.
In the nrLSU phylogram (Fig. 7), S. chloroporum and S. flavorhizomorphae clustered within the family Hydnaceae with strong support (MLBS/BPP = 99/1). Species of mycorrhizal Sistotrema (S. alboluteum, S. albopallescens, S. chloroporum, S. confluens, S. flavorhizomorphae, S. muscicola, and S. subconfluens) and Hydnum formed a monophyletic clade labeled “core Sistotrema/Hydnum” with strong support (MLBS/BPP = 99/1). Two mycorrhizal Sistotrema sp. (TUFC 101478 and TUFC 31978) were also positioned in this clade. The monophyletic S. chloroporum and S. flavorhizomorphae clades were each strongly supported (MLBS/BPP = 99/1 and 99/0.99, respectively).
In the ITS phylogram (Fig. 8), mycorrhizal Sistotrema s.l. and Hydnum species formed a monophyletic clade with moderate support (MLBS/BPP = 71/0.91). Each mycorrhiza in this study clustered with conspecific basidiomes. Sistotrema chloroporum formed a monophyletic clade with S. alboluteum, S. aff. alboluteum, and unidentified mycorrhizal root tips (MLBS/BPP = 95/1). Sistotrema flavorhizomorphae formed a monophyletic clade with S. aff. muscicola and unidentified root tips (MLBS/BPP = 98/1). The monophyly of S. chloroporum and S. flavorhizomorphae was each well-supported (MLBS/BIP = 100/1; 100/1) and delimited by small intraspecific variation (3%).
The rpb2 phylogeny (Fig. 9) included a well-supported monophyletic clade containing mycorrhizal Sistotrema and Hydnum (MLBS/BIP = 99/1). This clade was phylogenetically distinct from other ectomycorrhizal taxa in Hydnaceae, i.e., Cantharellus, Clavulina, and Craterellus. Within the core Sistotrema/Hydnum clade, Hydnum species are monophyletic (MLBS/BIP = 83/0.98), while Sistotrema s.l. species are divided into at least three clades, where S. chloroporum and Sistotrema sp. TUFC 31972 (MLBS/BIP = 78/0.99), S. flavorhizomorphae and Sistotrema sp. TUFC 101478 (MLBS/BIP = 98/1), and Swedish S. confluens (DQ381837) each forming independent monophyletic groups.
We describe two new species of Sistotrema s.l.: S. chloroporum and S. flavorhizomorphae. In the ITS phylogeny, S. alboluteum, S. albopallescens, and S. muscicola were each polyphyletic. We could not determine which clade was the true one for each because taxonomically reliable DNA sequences have never been determined from authentic specimens of mycorrhizal Sistotrema s.l. except for S. subconfluens. Furthermore, we could not predict phylogenetic positions of S. brunneolum and S. dennisii because no DNA sequences of them are available. Thus, we unfortunately cannot discuss whether the two novel species described here differ from previously described species based on phylogenetic relationships alone. However, both novel species were well-defined based on the morphology of their basidiomes and mycorrhiza, and distinguished from other Sistotrema s.l. species. Sistotrema chloroporum and S. flavorhizomorphae were phylogenetically positioned into the core Sistotrema/Hydnum group and were related to known mycorrhizal species of both lineages. We further observed ectomycorrhizal formations of S. flavorhizomorphae and S. chloroporum in their habitats.
Basidiomes of S. flavorhizomorphae are characterized morphologically by their resupinate, hydnoid to irpicoid hymenophore, bright-yellowish rhizomorphs, and small and subglobose basidiospores (3-3.5 × 2.5-3 µm). Sistotrema muscicola and S. raduloides also form hydnoid hymenophores. However, S. muscicola has larger, ellipsoid basidiospores (all cited descriptions outreach 4 µm long), basidia normally with six sterigmata, and no typical rhizomorphic bundle (Bourdot & Galzin, 1914; Rogers, 1944; Lundell & Nannfeldt, 1947; Eriksson et al., 1984; Gilbertson & Ryvarden, 1987; Ryvarden & Gilbertson, 1994). Sistotrema raduloides differs in longer basidiospores (6-9 × 2.5-3.5 µm), lack of a rhizomorphic bundle, and adnate basidiomes on wood impressing wood-decaying fungus (Eriksson et al., 1984).
Basidiomes of S. chloroporum are chracterized morphologically by poroid, partly yellowish-green hymenophores, basidia with 4-6 sterigmata, and medium-sized basidiospores (4.5-6.5 × 3.5-6 µm). Currently, six valid poroid species besides S. chloroporum are known in the genus Sistotrema: S. albopallescens, S. alboluteum, S. brunneolum, S. confluens, S. dennisii, and S. subconfluens. Of these, only S. chloroporum has green hymenophores. In addition, S. alboluteum differs slightly in its thick-walled, roundish basidiospores (4.5-5-6 × 4-5 µm; Bourdot & Galzin, 1925, or 4.5-6 µm; Eriksson et al., 1984) and basidia producing 2-4 sterigmata (Bourdot & Galzin, 1925; Eriksson et al., 1984; Ryvarden & Gilbertson, 1994; Spirin & Zmirovich, 2007; Gorjón & Hallenberg, 2008). Sistotrema albopallescens has thinner hymenophores, shorter basidia (< 20 µm long), and smaller basidiospores (< 4.5 µm long) (Bourdot & Galzin, 1925; Eriksson et al., 1984). Sistotrema brunneolum has “resinous brown” hymenophores and more ellipsoid, thick-walled, cyanophilous basidiospores (Spirin & Zmirovich, 2007). Sistotrema confluens and S. subconfluens have pileate-stipitate basidiomes and more elliptic basidiospores (Eriksson et al., 1984; Ryvarden & Gilbertson, 1994; Zhou & Qin, 2013; Bubner et al., 2014).
Two species treated as nomenclatural synonyms of poroid Sistotrema s.l. are morphologically similar to S. chloroporum: Sistotrema eluctor Donk (synonym of S. alboluteum: Eriksson et al., 1984; Ryvarden & Gilbertson, 1994) and Poria albolutea var. stenospora Bourdot & Galzin (synonym of S. dennisii: Eriksson et al., 1984). Sistotrema eluctor has thin-walled basidiospores and basidia with 4-6 sterigmata (Rogers, 1944; Donk, 1967) similar to S. chloroporum but differs in its shorter basidium (< 22 µm long in both descriptions) and hymenophores without greenish color. Poria albolutea var. stenospora Bourdot & Galzin has yellowish-green or fawn yellow hymenophores, urniform basidia (15-31 × 6-8 µm) producing 4(-6) sterigmata, and medium-sized basidiospores (4-5-7 × 3-4.5 µm) (Bourdot & Galzin, 1925), similar to those of S. chloroporum. We did not treat this taxon as S. chloroporum because P. albolutea var. stenospora illustrated in Bourdot and Galzin (1928) has elliptic basidiospores. Furthermore, these two species are European species but there is no evidence of European distribution of S. chloroporum in the NCBI GenBank and UNITE databases (see below).
An ecological difference can serve as a taxonomic character to distinguish novel species. Although the host preference/specificity of mycorrhizal fungi is complex and needs both field surveys and in vitro mycorrhizal synthesis assays using mycelia and basidiospores to evaluate accurately (Lofgren et al., 2018; Pérez-Pazos et al., 2021), our observations and molecular analyses of mycorrhizae demonstrated that S. flavorhizomorphae prefers pines whereas S. chloroporum prefers broadleaved trees in Betulaceae (Carpinus) and Fagaceae (Castanopsis and Quercus). This suggests that different species in this genus have different host preferences. However, there are few studies of the forest habitats of mycorrhizal Sistotrema s.l. because most species in this genus have saprotrophic-like habitats on decaying-wood or humus (e.g., Eriksson et al., 1984). The sequences generated by the mycorrhizal community analyses can be used for ecogeographical comparisons. We recognized several unidentified lineages within the sequences of mycorrhizal samples. These lineages have no morphological data for basidiomes; however, host preference and geographic distribution can be estimated (see Fig. 8). Sistotrema flavorhizomorphae showed affinity with pine according to our study and previously detected mycorrhizal samples, whereas its sister clade is associated with Populus tremula L. (Bahram et al., 2011). Furthermore, S. chloroporum and S. flavorhizomorphae were estimated to have narrow distributions in Asian countries, whereas their sister lineages showed limited distribution in Europe (S. alboluteum: Larsson et al., 2004; Nilsson et al., 2006) or North America (S. aff. alboluteum: Rosenthal et al., 2017) and wider Holarctic distribution (S. aff. muscicola: Rosenthal et al., 2017; Suz et al., 2017).
Both S. chloroporum and S. flavorhizomorphae showed stable mycorrhizal characters regardless of their host plant species (Tables 1, 2). Mycorrhizal morpho-anatomy of S. chloroporum resembles previous mycorrhizal descriptions of Sistotrema spp. (Di Marino et al., 2008; Münzenberger et al., 2012; Bubner et al., 2014), however S. flavorhizomorphae showed distinctive characters. Similar to other mycorrhizal descriptions of Sistotrema/Hydnum, both new species have a hydrophobic root surface, oil-rich hyphae with clamps, ampullate inflation and closed-anastomoses in emanating hyphae (Agerer, Kraigher, & Javornik, 1996; Harrington & Mitchell, 2002; Agerer, 2006; Di Marino et al., 2008; Sugawara, Sotome, Maekawa, Nakagiri, & Endo, 2021). By contrast, S. flavorhizomorphae mycorrhizae show typical pseudoparenchymatous cells in the middle mantle and bright yellowish rhizomorphs attributing to yellowish incrustation surrounding hyphae, which have been never reported among Sistotrema/Hydnum species. Therefore, mycorrhiza of Sistotrema exhibit not only common character traits within closely related species but also diagnostic character traits within specific taxa. While Sistotrema s.l. and Hydnum have a high phylogenetic species diversity (Münzenberger et al., 2012; Feng et al., 2016; Niskanen et al., 2018), only a few mycorrhizal characterizations are available; hence, a comprehensive study of mycorrhizal morpho-anatomy is needed for accurate and precise taxonomic classification. Note that even if the ectomycorrhiza is well-defined and has great novelty in its characteristics, no one can describe new species in this genus based on its mycorrhiza alone, because species of the genus Sistotrema have traditionally been described based on basidiome morphology and accurate identification strongly depends on this.
The genus Sistotrema in the traditional sense is polyphyletic. Sistotrema confluens (type species) and S. subconfluens formed a monophyletic clade in the nrLSU phylogeny and homologous morphotypes of pileate-stipitate basidiomes. The rpb2 phylogeny also suggested polyphyly of Sistotrema which was divided into more than four lineages within the core Sistotrema/Hydnum lineages: S. confluens, Hydnum, and at least two clades of resupinate Sistotrema. Therefore, the clade consisting of S. confluens and S. subconfluens should be treated as Sistotrema s.s. Other mycorrhizal and wood-inhabiting Sistotrema s.l. species having resupinate basidiomes need genus-level taxonomic re-evaluation. However, the generic boundary of Sistotrema s.s. is unclear due to the long-branched, unstable phylogenetic position of S. confluens in rDNA (see Fig. 7; Supplementary Fig. S1). Although protein-coding genes give better phylogenetic solution, currently there are limited sequence data from mycorrhizal Sistotrema s.l. Furthermore, numerous environmental DNA sequences suggest that mycorrhizal Sistotrema s.l. may have much greater species diversity than described to date. To elucidate generic relationships among mycorrhizal-resupinate Sistotrema s.l., pileate-stipitate Sistotrema s.s., and Hydnum species and to establish new genera for mycorrhizal-resupinate Sistotrema s.l., further sampling and taxonomic studies of mycorrhizae as well as basidiomes of Sistotrema s.l. is necessary.
The authors declare no conflicts of interest. All the experiments undertaken in this study comply with the current laws of the country where they were performed.
This study was supported by a Grant-in-Aid for Scientific Research (no. JP20J20884) from the the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. We thank Fasmac Co., Ltd for technical support regarding DNA sequencing.
