Full paper
Taxonomy and phylogeny of Exobasidium pentasporium causing witches' broom of Rhododendron species
2022 Volume 63 Issue 6 Pages 247-253
Details
2022 Volume 63 Issue 6 Pages 247-253
Exobasidium pentasporium was first found on Rhododendron kaempferi in Nikko, Tochigi Prefecture, Japan and described only with a brief mentions and illustration of a specimen in 1896. This fungus causes a witches' broom disease of Rhododendron species. To stabilize the concept of this species, the specimen in the protologue was located, carefully examined, and illustrated. In addition, the name was epitypified based on a newly collected topotype specimen. A phylogenetic tree using ITS and LSU sequences showed that our isolates of E. pentasporium grouped with other Exobasidium species on Rhododendron forming a monophyletic clade with strong statistical support and were unrelated to E. nobeyamense, another causal agent of witches' broom disease on Rhododendron species.
Exobasidium pentasporium Shirai found on Rhododendron kaempferi Planch. var. kaempferi in Nikko, Tochigi Prefecture, Japan, was first described by Dr. Mitsutaro Shirai (Shirai, 1896). This species causes witches' broom on several Rhododendron species, such as R. indicum (L.) Sweet, R. kaempferi, R. macrosepalum Maxim., and R. tschonoskii Maxim (Shirai, 1896; Sawada, 1950; Ezuka, 1990; Shibata & Hirooka, 2021). The hymenium of E. pentasporium appears on the abaxial side of living leaves (Ezuka, 1990). Another Exobasidium species causing a witches' broom disease on Rhododendron species, Exobasidium nobeyamense Nagao & Ezuka, has also been recorded in Japan but is morphologically different from E. pentasporium (Nagao Ezuka, Ohkubo, & Kakishima, 2001). Witches' broom often seriously damages Rhododendron species, including the natural monument of R. kaempferi in Yamanashi Prefecture, Japan (Ezuka, 1990; Shibata & Hirooka, 2021).
The description of E. pentasporium was published in the Botanical Magazine, Tokyo (= Journal of Plant Research) Vol. 10, No. 113, in 1896 (Shirai, 1896). There were no clear type descriptions in the text, including illustrations, and descriptions for specific specimens. However E. pentasporium Shirai (1896) was adopted as the official name because a type specimen has been necessary for naming of a new fungal species only after Jan 1, 1958 based on Article 40.1 in the International Code of Nomenclature for algae, fungi, and plants (ICNafp) (Turland et al., 2018). Scientific names enable us to recognize organisms and to communicate scientific knowledge about them, and an unambiguous name is assured by the permanent preservation of physical organisms as type specimens of the name of taxon (Yurkov et al., 2021). Thus, the need for type designation is required because a taxonomic name published later than May 1, 1753, must be typed with a specimen chosen based on Article 7.9. In addition, a name can be epitypified by a newly collected specimen from which a living culture has been derived in order to serve as a modern reference specimen according to Article 9.9 (Turland et al., 2018). Typification of the name E. pentasporium is, therefore, important to stabilize the accurate identification of this pathogenic fungus and to serve as a foundation for applied research on this fungus.
In this study, we found and examined an old specimen preserved in an herbarium that was collected by Shirai in 1895 and a fresh specimen collected in the type locality, in Nikko, Tochigi, Japan. The preserved specimen is herein regarded as the holotype of E. pentasporium and a recently collected, dried specimen is designated as epitype. The morphological characteristics of this species and its molecular phylogenetic position among other Exobasidium species and its relatives were also determined.
We explored dried specimens of Exobasidium spp. collected by Shirai at the mycological herbarium of the National Museum of Nature and Science, Tsukuba, Japan (TNS). A fresh specimen was collected on May 19, 2021, at Nikko city, Tochigi Prefecture, Japan, the type locality cited in the original paper, from a witches' broom on Rhododendron kaempferi. The fresh epitype and additional specimens were deposited in the mycological herbarium of TNS.
2.2. IsolatesTo obtain isolates, infected leaves with sori were cut into small pieces about 5 mm square and the pieces were fixed with plastic tape to the inside of the lid of a sterile Petri dish with 1.8% water agar (WA). The dish was kept at room temperature from 12 to 36 h. Single germinating spores were picked up and transferred onto fresh Potato Dextrose Agar (PDA) plates. Two isolates obtained were deposited in the Genbank, the Genetic Resources Center, NARO (National Agriculture and Food Research Organization) and Microbe Division/Japan Collection of Microorganisms, RIKEN BRC-JCM (RIKEN BioResource Research Center), Tsukuba, Ibaraki, Japan.
2.3. MorphologyTo describe the morphological characteristics in nature, six specimens including the preserved specimen (TNS-F-4617) of the host R. kaempferi were observed. Also five isolates derived from the newly collected specimens (TNS-F-97534 = HM21-486, HM21-488, HM21-586, HM21-595 and HM21-602) were used for cultural observation. The isolates were incubated on PDA at 20 °C for 30 d in the dark. A light microscope (OLYMPUS BX51, Tokyo, Japan) was used to examine morphological characters including the size and shape of basidia, basidiospores, and conidia. Spores or thin sections of sori from dry specimens were mounted in a drop of water on glass slides. The slide preparations were examined and photographed using a digital camera (Olympus DP21, Tokyo, Japan). Approximately 30 spores from each specimen were randomly chosen, and their length and width were measured using ‘ImageJ’ software (free download available at http://rsbweb.nih.gov/ij/). Measurements represent mean values plus/minus one standard deviation and minimum and maximum values are given in parentheses. For observation of colony morphology, the two isolates were incubated on PDA at 20 °C for 5 d in the dark.
2.4. Molecular phylogeny 2.4.1. DNA extraction, amplification, sequencing, and sequence alignmentTotal DNA was extracted from cultures using Tris-EDTA (TE) Buffer. Tubes including fungal tissue and 50 µL of TE buffer were heated at 95 °C for 10 min and then were centrifuged at 10,000 rpm for 3 min. The supernatant was used as template for the PCR reaction. PCR amplification was performed in a VeritiPro Thermal Cycler (Applied Biosystems, USA) with a program consisting of an initial denaturing step at 94 °C for 4 min; 35 cycles of denaturation at 94 °C for 1 min, annealing at 47-52 °C for 1 min, and extension at 72 °C for 1 min; and a final extension step at 72 °C for 8 min. The ribosomal DNA internal transcribed spacer (ITS) and D1/D2 region of the large-subunit rDNA (LSU rDNA) were amplified by PCR using V9G/LR5 and LR0R/LR5, respectively (de Hoog & Gerrits van den Ende, 1998; Rehner & Samuels, 1994; Vilgalys & Hester, 1990). The amplicons were treated with ExoSAP-IT (Affymetrix, USA) and used as the templates for the sequencing reaction with a BigDye Terminator v3.1 Cycle Sequencing Kit (Life Technologies, USA) with M-ITS1/ITS4 (Stoll, Pienenbring, Begerow, & Oberwinkler, 2003; White, Bruns, Lee, & Taylor, 1990) and NL1/NL4 (O'Donnell, 1993). Sequence data were collected with an ABI 3130 Genetic Analyzer (Applied Biosystems, USA). The sequences were deposited in DDBJ (DNA Data Bank of Japan).
2.4.2. Phylogenetic analysesTo construct a phylogenetic tree, the sequences were aligned using MAFFT version 7 with E-INS-i (Katoh & Standley, 2013). Ambiguously aligned portions of the alignments were manually removed using Mesquite version 3.2 (Maddison & Maddison, 2017). The optimum substitution models for each data set were estimated using jModeltest 2.1.10 (Darriba, Taboada, Doallo, & Posada, 2012; Guindon & Gascuel, 2003) based on the Akaike information criterion (AIC; Akaike, 1974) for the ML analyses. To analyze the taxonomic relationships between the isolates and its relatives, maximum likelihood (ML) was performed using the program RAxML-NG (Kozlov, Darriba, Flouri, Morel, & Stamatakis, 2019) based on the models selected with the GTR+I+G for ITS and SYM+I+G for LSU. Bootstrap proportions (BPs) were obtained using 100 bootstrap replications. A total of 66 taxa including 54 species of the genus Exobasidium and its relatives were included in our tree. Clinoconidium cinnamomi (Accession No. KX196602, KX196604 for KUN-F 94646) and Laurobasidium hachijoense (Accession No. LC146733 for NBRC 31857) were used as outgroups. These alignments were shown in Supplementary alignment S1.
The ML phylogenetic analyses were conducted using an aligned sequence dataset composed of 675 nucleotides from ITS and 593 from LSU for the analyses. The alignment contained a total of 66 taxa which consisted of 47 taxa (71.2%) in ITS and 66 (100%) in LSU (Supplementary Table S1). This combined dataset provided higher confidence values than those of the individual locus trees (data not shown). Of the 1268 characters included in the alignment, 616 were variable and 570 were conserved. The ML tree with the highest log likelihood (-11437.5635) is shown in Fig. 1. Monophyly of E. pentasporium in the Rhododendron clade of the genus Exobasidium was resolved as a strongly supported clade (100 % ML BP, Fig. 1). Exobasidium nobeyamense, another causal agent of witches' broom disease on Rhododendron species, was not closely related to E. pentasporium. Arcticomyces warmingii and Muribasidiospora indica were phylogenetically clustered in Exobasidium clade with low bootstrap support.
Exobasidium pentasporium Shirai, Bot. Mag., Tokyo 10: 53 (1896) Figs. 2, 3.
MycoBank no.: MB 111481.
≡ Microstroma pentasporium (Shirai) Imazeki, In Asahina et al., Nippon Inkwasyokubutu Dukan p. 377 (1939)
Holotypus: JAPAN, Tochigi Pref., Nikko city, Chuzen-ji, on leaves of witches' broom of Rhododendron kaempferi (‘Yamatsutsuji’ in Japanese), Jun. 1895, M. Shirai (TNS-F- 4617)
Epitypus designated here MBT 10007689 : JAPAN, Tochigi Pref., Nikko city, on leaves of witches' broom of Rhododendron kaempferi, 19 May 2021, S. Shibata [TNS-F-97534 = HM21-486, culture MAFF 247707 = JCM 39281 = HM21-486C; gene sequence ITS (LC711023), LSU (LC711026)]
Additional specimens examined: JAPAN, Tochigi Pref., Nikko city, on leaves of witches' broom of R. kaempferi, 19 May 2021, S. Shibata (HM21-488, derived culture MAFF 247708 = JCM 39282 = HM21-488C); Yamanashi Pref., Hokuto city, on leaves of witches' broom of R. kaempferi, 10 Jun 2021, S. Shibata & Hirooka (HM21-586, HM21-595, HM21-602, derived culture MAFF 247709, 247710, 247711 = HM21-586C, HM21-595C, HM21-602C)
Description: Hymenium composed of basidia with 3-5 sterigmata and conidia. Hyphae not developing on surface of epidermis. Basidia clavate to cylindrical, (13.4-)22.8-35.7(-42.1)×(2.8-)3.4-6.6(-9.2) μm, apices obtuse. Basidia emerging directly from leaf surface or through stomata, not fasciculate. Sterigmata (1.0-)1.5-3.7(-5.7) μm, emerging outwardly and tapering toward tip. Basidiospores oblong to reniform, rarely clavate, (11.3-)12.2-15.3(-17.5) × (2.0-)2.8-3.7(-4.0) μm, hyaline, smooth, one-celled when formed, becoming septate with 1(-2) septa. Two types of conidia: large conidia abundant, clavate (3.9-)5.4-8.6(-9.7)×(0.5-)0.7-1.2(-1.6) μm; small conidia rare, ellipsoidal to cylindrical conidia. Germ tubes or conidia emerged from both ends of basidiospores. Conidia budding to produce daughter cells at each end and also developing hyphae.
Colonies on PDA slow growing, wrinkled irregularly around periphery. Surface pale yellow, later brown. Colonies gelatinous, partially sub-immersed in agar surface. Colonies composed of branching, intricate hyphae, and blastoconidia. Reverse pale yellow with dark pigmentation on PDA. Blastoconidia hyaline, smooth, bacilliform, clavate, (6.8-)7.5-10.9(-15.4)×(0.9-)1.1-1.8(-2.1) μm, not powdery.
Notes : A specimen deposited as TNS-F-4617 labeled as Exobasidium pentasporium collected by Dr. Shirai in Jun, 1895, was found and serves as the holotype specimen (Fig. 2). This holotype specimen and the epitype specimen designated herein agree with the illustrations in the protologue of E. pentasporium, in terms of some features of symptoms and basidia (Shirai, 1896, Figs. 18 and 19). Because some morphological characters of E. pentasporium are not present for the holotype specimen, the lacking information was complemented based on the epitype specimen (TNS-F-97534 = HM21-486) collected from the same locality and host (Supplementary Table S2).
In the protologue of E. pentasporium, Shirai (1896) listed R. indicum as the host of E. pentasporium. Rhododendron indicum corresponds to the current scientific name of ‘Satsuki’ in Japanese. An annotation card in Shirai's handwriting included in this specimen indicates that ‘Yamatsutsuji’ (the Japanese name for R. kaempferi) was erroneously named as R. indicum (Fig. 2B). It is therefore assumed that the host plant in the protologue of E. pentasporium was R. kaempferi, not R. indicum.
Exobasidium pentasporium have been reported on various Rhododendron species in China, Japan, and Russia (Farr & Rossman, 2021) and is known as a plant pathogen causing persistent epiphytic symptoms in the host, resulting in debilitation and death (Ezuka, 1990). Shibata and Hirooka (2021) recently reported serious damage caused by this fungus on the natural monument tree of R. kaempferi in Yamanashi Prefecture, Japan. Our findings of the type specimens have thus contributed to the accurate diagnosis of this disease. The host plants of the fungus, Rhododendron species, are widely planted as a horticultural plant all over the world (Watanabe & Takahashi, 2018). To know the taxonomic position, host range and the genetic diversity of E. pentasporium as well as the other species of the genus Exobasidium, is, therefore, important for plant protection.
According to Ito (1955) and the database of plant diseases in Japan (https://www.gene.affrc.go.jp/databases-micro_pl_diseases_en.php), Elliottia paniculata (Siebold & Zucc.) Benth. & Hook.f. (as Tripetaleia paniculata Siebold & Zucc., ‘Hotsutsuji’ in Japanese) has been recorded as a host plant of E. pentasporium. We suggest that this plant might not be a host of the fungus. This report may be based on a specimen for which the host was misidentified in the sense of host range of E. pentasporium. Our phylogenetic tree clearly showed that E. pentasporium isolates fall into the Rhododendron clade of the genus Exobasidium. Also, the genus Elliottia is phylogenetically distant from the genus Rhododendron (Gillespie & Kron, 2010). Ito (1955), in fact, wrote the scientific name of ‘Hotsutsuji’ as “R. weyrichii Makino”. Ezuka (1990) then pointed out that there is no plant species corresponding to “R. weyrichii Makino”, but Rhododendron weyrichii Maxim. exists. We speculate that the specimen reported by Ito (1955) was obtained from E. pentasporium on R. weyrichii Maxim. To clarify the host name, it is necessary to find and examine the specimen that Ito (1955) observed.
Our molecular phylogenetic tree with 54 species of Exobasidium including the sequence derived from an epitype-related culture of E. pentasporium showed that most of the branches were not strongly supported. In addition, our tree did not show the clear correlation between phylogenetic relatedness and symptoms of the host. Exobasidium nobeyamense, another causal agent of witches' broom disease on Rhododendron species, was phylogenetically unrelated to E. pentasporium. It may be because of the low number of registered sequences of Exobasidium species. Approximately 130 species of Exobasidium have been described, but less than half of them (41 species/128 species) are found in the Taxonomy Database of NCBI (http://www.ncbi.nlm.nih.gov/taxonomy). This may result from the fact that (1) most species of Exobasidium were described before molecular tools were available, and (2) the classification of Exobasidium species has been conducted mainly based on the characteristics of symptoms, host range and morphology on plants. In the current analysis, insertion of Arcticomyces and Muribasidiospora species into the Exobasidium clade was observed with low bootstrap support. The phylogenetic relationships of Exobasidium species are difficult to determine because of the lack of sequence data derived from type specimens. The type of the genus, Exobasidium vaccinii (Fuckel) Woronin, has no available authentic DNA sequence data in public databases. In order to resolve these problems, it is crucial to clarify the holotype specimen and designate an epitype specimens of Exobasidium species and to deposit the strains and sequence data obtained from them in public institutions.
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.
We sincerely thank Dr. Amy Y. Rossman and Dr. Scott Redhead for valuable discussions and comments. We also thank Dr. Tsuyoshi Hosoya as a curator of the National Museum of Nature and Science TNS-F 4617 for loans of specimens. Our thanks are extended to staff of the Nikko Botanical Garden, the University of Tokyo, for sampling witches' broom of R. kaempferi.
