2022 Volume 63 Issue 1 Pages 33-38
Fungi in the genus Termitomyces are external symbionts of fungus-growing termites. The three rhizogenic Termitomyces species T. eurrhizus, T. clypeatus, and T. intermedius, and one species similar to T. microcarpus that lacks pseudorrhiza, have been reported from Ryukyu Archipelago, Japan. In contrast, only two genetic groups (types A and B) of Termitomyces vegetative mycelia have been detected in nests of the fungus-growing termite Odontotermes formosanus. In this study, we investigated the relationships between the mycelial genetic groups and the basidiomata of Termitomyces samples from the Ryukyu Archipelago. We found that all the basidioma specimens and the type B mycelia formed one clade that we identified as T. intermedius. Another clade consisted of the type A mycelia, which showed similarity to T. microcarpus, was identified as T. fragilis. Our results indicate that the Japanese T. eurrhizus and T. clypeatus specimens should re-named as T. intermedius.
Termitomyces is a monophyletic group in the family Lyophyllaceae (order Agaricales, Division Basidiomycota), and its members have only been found in association with the nests of fungus-growing termites (Frøslev, Aanen, Laessøe, & Rosendahl, 2003; Matheny et al., 2006). Fungus-growing termites also belong to a monophyletic group in the subfamily Macrotermitinae (Kambhampati & Eggleton, 2000; Bucek et al., 2019). Termitomyces-Macrotermitinae associations are found in tropical and subtropical regions of Asia and Africa, including the Ryukyu Archipelago, which lies southwest of Japan.
A single species of fungus-growing termite, Odontotermes formosanus Shiraki, along with three pseudorrhiza-forming and one pseudorrhiza-lacking Termitomyces species, have been found in the Ryukyu Archipelago. Termitomyces albuminosus (Berk.) R. Heim was first found on Ishigaki Island and given the common name “Ooshiroaritake” by Otani (1979). It was regarded as the same species as Collybia albuminosa (Berk.) Petch, which was collected on Taiwan Island by Sawada (1919) and named “Ooshiroaritake”. The T. albuminosus sensu Otani was later re-named T. eurrhizus (Berk.) R. Heim by Imazeki and Hongo (1987). The species T. clypeatus R. Heim was found on Iriomote Island and given the common name “Togariarizukatake” by Otani and Shimizu (1981). The third species, T. intermedius Har. Takah. & Taneyama, was described as a novel species with the common name “Shiroarishimeji” by Takahashi et al. (2016), who didn't elucidate its relationships with previously identified species. Recently, a fourth species similar to T. microcarpus (Berk. & Broome) R. Heim that lacks pseudorrhiza was reported with the common name “Ikeharaooshiroaritake” and no scientific name (referred as T. aff. microcarpus here) (Hojo & Shigenobu, 2019). The latter species was recently described as T. fragilis L. Ye, Karun, J.C. Xu, K. D. Hyde & Mortimer (Ye et al., 2019).
Apart from these taxonomic studies, molecular analyses of termite nests have revealed that only two phylogenetic clades of Termitomyces, designated as type A and type B mycelia, exist in the Ryukyu Archipelago (Katoh, Miura, Maekawa, Shinzato, & Matsumoto, 2002; Hojo & Shigenobu, 2019). Termitomyces intermedius corresponds to the type B mycelia, and the pseudorrhiza-lacking T. aff. microcarpus (= T. fragilis) corresponds to the type A mycelia. However, the relationships between genetic types of vegetative mycelia and the T. eurrhizus and T. clypeatus basidiomata remain unclear. To address this knowledge gap, we examined morphological characteritics and investigated the phylogenetic relationships among Termitomyces samples from the Ryukyu Archipelago.
We collected 83 basidiomata and 12 fungal combs of Termitomyces spp. from Iriomote, Ishigaki, Kohama, and Okinawa Islands in the Ryukyu Archipelago (Supplementary Table S1). The basidiomata were collected during the fruiting season (Jun to Aug) from 1993 to 2014. They varied in shape, size, and color. Representative specimens are shown in Fig. 1. Specimen SW-OST1 showed a grayish pileus with a conical perforatorium, white, free, crowded lamella, and a cylindrical stipe with long pseudorrhiza connecting to an underground fungal comb (Fig. 1A). These characteristics matched the description of T. albuminosus sensu Otani (=T. eurrhizus) by Otani (1979). Specimen SW-KGH1 had a distinct, grey to dark brown spiniform protrusion (Fig. 1B) that matched the description of T. clypeatus sensu Otani (Otani & Shimizu, 1981). Specimen SW-TPG1 showed an intermediate shape with a whitish brown color and a black or dark spot in the center (Fig. 1C). The pileal diameters ranged from 40 to 120 mm, and stipe lengths ranged from 30 to 100 mm (Fig. 1A-D). These macroscopic characters were in agreement with taxonomic keys designed to identify T. eurrhizus, T. clypeatus, and T. intermedius based in specimens found previously in Japan (Otani, 1979; Otani & Shimizu, 1981; Imazeki & Hongo, 1987; Pegler & Vanhaecke, 1994; Takahashi et al., 2016).
Microscopic characteristics of the basidiomata were examined under compound microscopes BX51 and BX61 with UPlanSApo 10×, 20×, 40×, 60×/water and 100×/oil objective lenses and a stereomicroscope SZX7 (Olympus, Tokyo, Japan). Dried specimens were re-moistened and softened in a 3% KOH (Wako, Osaka, Japan) solution. The specimens were mounted in 3% agarose (Wako, Osaka, Japan) and sectioned using a DSK LinearSlicer PRO7 (Dosaka EM, Kyoto, Japan). Basidiospores of specimen SW-OST1 were 5.9-7.0 × 4.1-5.1 µm (average 6.5 × 4.6 µm), hyaline, ellipsoidal with a hilar appendix, and non-amyloid (Fig. 2A). Basidia were 18.6-29.8 × 5.6-7.4 µm (average 24.2 × 6.5 µm), clavate, non-amyloid, with four sterigmata (Fig. 2B, C). Other specimens, including SW-KGH1 and SW-TPG1, also showed similar characteristics in both the basidiospores and basidia. Pleurocystidia were 32.6-54.4 × 14.3-32.1 µm (average 43.5 × 23.2 µm), infrequent, clavate to napiform, and non-amyloid (Fig. 2D). The pileipellis was cutis, and cells between the outermost layer and the pigmented layer were usually inflated, but the degree of such inflation differed among specimens (Fig. 2E, F). Where the inflation was high (Fig. 2E), the structure resembled Takahashi et al. (2016)'s description, and where the inflation was low (Fig. 2F), the structure resembled Otani (1979)'s description. The inflated layer was often absent, even within similar specimens that showed comparable but distinct inflated layers (Fig. 2G). Although this can be caused by exfoliation of the outermost and inflated layer, the structures resembled the description by Otani and Shimizu (1981). No clamp connections were found on any hyphae. The stipe trama of specimen 20120609-3 was dextrinoid (Fig. 2H) whereas that of specimen 20120609-2 was not (Fig. 2I). Thus, dextrinoid stipe trama is not a common trait of these basidiomata. This description does not match the characteristics of T. intermedius as described by Takahashi et al. (2016). Similarly, dextrinoidity of the pileitrama also differed among the specimens.
The fungal combs were collected from Nov 2013 to Oct 2014. We isolated hyphae from nodules on the fungal combs by culturing them on potato dextrose agar (PDA; Nissui, Tokyo, Japan) for 1 mo at 28 °C in the dark. The hyphal colonies were used for microscopy and DNA analysis. Calcofluor white (Fluorescent Brightener 28, Wako) was used to stain the cell walls of the cultured mycelia, and nuclei were stained with 10 µg/mL DAPI (Dojindo, Osaka, Japan) for 30 min. A filter unit with a 360-370 nm excitation filter, a 400 nm dichromatic mirror, and a 420 nm emission filter was used for fluorescent imaging of the calcofluor white- and the DAPI-stained samples.
The strains isolated from nodules on the fungal combs formed cream-colored mycelial colonies with nodule-like structures on the PDA plates (Fig. 3A). However, nodule-like structures became rare after extended subculture (Fig. 3B). No clamp connections were observed in the cultured hyphae (Fig. 3C). The nodule-like structures consisted of highly inflated hyphae (Fig. 3D). These characteristics of the cultured hyphae were different from those described by Otani (1979), who reported sac-like cells between filamentous cells. We checked the numbers of nuclei in the cells, because multinuclearization was previously reported for a Termitomyces species (Henrik, Andersen, & Aanen, 2005). We found both dikaryotic and multinucleate cells in the cultured hyphae (Fig. 3E, F), and inflated cells were frequently multinucleate (Fig. 3F). Although there are few reports on the karyotic status of Termitomyces species, we observed that multinucleate cells is a common trait of Termitomyces.
To further understand the relationships among the Japanese Termitomyces specimens, we extracted DNA and performed phylogenetic analyses based on the internal transcribed spacer (ITS) region, the translation elongation factor 1α gene (TEF1), and RNA polymerase II subunit b (RPB2). Genomic DNA was extracted using the cetyltrimethylammonium bromide (CTAB) method. PCR was performed using Takara ExTaq DNA polymerase (Takara Bio, Kusatsu, Japan) according to the standard protocol (https://catalog.takara-bio.co.jp/PDFS/RR001A_DS_j.pdf). Primers ITS1 and ITS4 were used for amplification and sequencing of the ITS region (White, Bruns, Lee, & Taylor, 1990). Primers were also constructed to amplify and sequence regions of the TEF1 gene and RPB2 (Supplementary Table S2). After PCR, excess primers were removed using ExoSAP-IT (GE Healthcare, Connecticut, USA), and the samples were sequenced using the BigDye Terminator v3.1 kit (Thermo Fisher Scientific, Massachusetts, USA) and an ABI 3130xl capillary sequencer (Thermo Fisher Scientific, Massachusetts, USA). The determined sequences were deposited in the DDBJ, EMBL, and GenBank databases (GenBank accession nos. LC578316-LC578463).
We obtained reference sequences for Termitomyces, including 36 ITS and 5 RPB2 sequences, from the NCBI nucleotide database. Sequences that appeared in previous studies (Rouland-Lefevre, Diouf, Brauman, & Neyra, 2002; Katoh et al., 2002; Hojo & Shigenobu, 2019; Ye et al., 2019) and sequences that were highly homologous to our samples or annotated as related species were used as references. No sequences related to TEF1 were found in the database. Sequences for Lyophyllum shimeji (Kawam.) Hongo, L. decastes (Fr.) Singer, and Tricholoma matsutake (S.Ito & Imai) Singer were also included as outgroups, since Lyophyllum is another member of the Lyophyllaceae (Bellanger et al., 2015; Endo et al., 2019) and Tricholoma is a member of Tricholomataceae, which is a family related to and morphologically resembling Lyophyllaceae (Hofstetter et al., 2014; Sánchez-García, Matheny, Palfner, & Lodge, 2014; Endo et al., 2019). The L. shimeji sequences were obtained using DNA isolated from cultivated mycelia of strain AT787 (Furukawa & Katagiri, 2020), which was provided by Dr. Akiyoshi Yamada (Shinshu University). The TEF1 and RPB2 sequences for T. matsutake were extracted from the T. matsutake whole genome ver3.0 archived in the JGI genome database (http://genome.jgi.doe.gov/).
Our phylogenetic analyses compared 137 ITS sequences (550-700bp), 29 partial TEF1 sequences (800-900 bp), and 34 partial RPB2 (550-650 bp) sequences. Sequence alignments and phylogenetic tree construction were carried out using MEGA X (version 10.1.8) software (Kumar, Stecher, Li, Knyaz, & Tamura, 2018; Stecher, Tamura, & Kumar, 2020). The sequences were aligned using the MUSCLE algorithm in MEGA X. All positions containing gaps and missing data were eliminated. In the final dataset, 400, 761, and 442 positions were used in the construction of the phylogenetic trees for ITS, TEF1, and RPB2, respectively. The phylogenetic relationships were inferred using the Maximum Likelihood method according to BIC criteria calculated with the “Find Best DNA/Protein Models” tool of MEGA X. We used the Tamura 3-parameter model (Tamura, 1992) with a discrete Gamma distribution (T92 + G) for the ITS tree, the Kimura 2-parameter model (Kimura, 1980) with a discrete Gamma distribution (K2 + G) for the TEF1 tree, and the Kimura 2-parameter model with the rate variation model of evolutionarily invariable (K2 + I) for the RPB2 tree. Initial trees for the heuristic search were obtained automatically by applying the Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood values. The posterior probability from the Bayesian analysis was calculated using MrBayes v3.2.7 (Ronquist et al., 2012). We used K2 + G for ITS and TEF1 and K2 + I for RPB2. Markov Chain Monte Carlo iterations were performed for 1,000,000 generations. The alignment files were deposited in TreeBase (Accession URL: http://purl.org/phylo/treebase/phylows/study/TB2:S28075).
We obtained a phylogenetic tree based on the ITS sequences from all the collected samples (GenBank accession nos. LC578316-LC578410) and from the public databases (Supplementary Table S3). The tree showed two clades among the fungal comb samples, which were previously designated as genetic types A and B by Katoh et al. (2002) (Fig. 4A). All of our basidioma samples were included in the type B clade, where T. intermedius (Takahashi et al., 2016) was also located. Therefore, all our basidioma specimens were phylogenetically identified as T. intermedius. The phylogenetic trees based on TEF1 (Fig. 4B) (GenBank accession nos. LC578411-LC578437; Supplementary Table S3) and RPB2 (Fig. 4C) (GenBank accession nos. LC578438-LC578463; Supplementary Table S3) showed similar topologies to the ITS tree, distinguishing the type A and type B clades in the same way. The type A ITS sequences were substantially longer than the type B ITS sequences. The type A ITS1 and ITS2 regions were approx. 250 bp and 220bp, respectively, while the type B ITS1 and ITS2 regions were approx. 155 bp and 195 bp, respectively.
Thus, even though we found macroscopic and microscopic variations in the basidiomata of rhizogenic Termitomyces collected in the Ryukyu Archipelago, the specimens were all identified as a single species, T. intermedius, in the phylogenetic analyses. In addition, our phylogenetic analyses confirmed the existence of two genetic types (A and B) among the Termitomyces specimens, as was reported by Katoh et al. (2002) and Hojo and Shigenobu (2019). Both phylogroups were commonly found on isolated islands of the Ryukyu Archipelago (Fig. 4).
Termitomyces intermedius from the Ryukyu Archipelago, Japan, was described by Takahashi et al. (2016). The species has also been found in Henan and Guandong Provinces, China (Huang et al., 2017). Our phylogenetic tree showed that T. intermedius also exists in Xichang (accession# KF302100) and Yunnan Provinces (FN812869) of China (Fig. 4). Thus, T. intermedius occurs extensively throughout the southern part of East Asia. Previously, Termitomyces specimens from the Ryukyu Archipelago were reported as “T. eurrhizus” or “T. clypeatus” (Otani, 1979; Otani & Shimizu, 1981; Imazeki & Hongo, 1987; Kinjo, Anucha, & Miyagi, 2005). However, our results show T. intermedius in a different clade than the true T. eurrhizus from China (KC414254). Also, the Japanese “T. eurrhizus” and the Sri-Lankan T. eurrhizus are morphologically different (Takahashi et al., 2016). Therefore, we infer that the Japanese “T. eurrhizus” is actually T. intermedius (Fig. 4). Furthermore, the Japanese “T. clypeatus” is not the true T. clypeatus that was originally found in Africa and described by Heim (1951), because the Asian “T. clypeatus” is genetically distant from the African T. clypeatus (Frøslev et al., 2003). Thus, it is likely that the type B specimens, including those Japanese specimens previously designated as “T. eurrhizus” or “T. clypeatus”, are all T. intermedius. Whereas, the common name “Ooshiroaritake”, given to the Japanese “T. eurrhizus” (=T. albuminosus sensu Otani) (Otani, 1979; Imazeki & Hongo, 1987; Kinjo et al., 2005), is given precedence when considering what the common name for this species is. Therefore, the common name “Shiroarishimeji”, which was given to T. intermedius by Takahashi et al. (2016) should be treated as an alias of “Ooshiroaritake”.
The type A clade, T. aff. microcarpus, is a pseudorrhiza-lacking species recently reported from Ryukyu Archpelago by Hojo and Shigenobu (2019), and described as T. fragilis from China (Ye et al., 2019). Our phylogeny (Fig. 4A, C) shows the type-A clade as related to T. microcarpus with some distance, whereas Japanese type-A specimens clustered with T. fragilis (Ye et al., 2019).
Overall, our results show that two species of Termitomyces are found in Japan: the pseudorrhiza-forming T. intermedius (common name “Ooshiroaritake”), and a pseudorrhiza-lacking T. fragilis (common name “Ikeharaooshiroaritake”).
The authors declare no conflict of interest. All the experiments undertaken in this study comply with the current laws of Japan.
This work was supported by a grant from the JSPS KAKENHI (Number 15K07798) awarded to M.H. & S.S. We thank the NIBB Core Research Facilities for technical support. We thank Mr. Haruki Takahashi & Ms. Atsuko Hadano for help in sample collection. We thank Dr. Akiyoshi Yamada for providing the L. shimeji sample used for phylogenetic analysis. We also thank Dr. Glen Cowan for advice on English expressions.
