植物研究雑誌 J. Jpn. Bot. 83: 88–95 (2008)

Fungal Symbiont and Cultivation Test of elata Blume ()

Gen KIKUCHIa,*, Masami HIGUCHIa, Takashi MOROTAa, Eiji NAGASAWAb and Akira SUZUKIc

aBotanical Raw Materials Research Department, Botanical Raw Materials Division, Tsumura & Co.,3586, Yoshiwara, Ami-machi, Ibaraki, 300-1192 JAPAN; *E-mail: [email protected] bTottori Mycological Institute, The Japan Kinoko Research Center Foundation, Tottori, 689-1125 JAPAN; cFaculty of Education, Chiba University, Chiba, 263-8522 JAPAN (Recieved on October 10, 2007)

Gastrodia elata tubers were collected at eleven sites throughout Honshu, Japan. Fungal symbionts were identified with mating test. Eight isolates were identified as Armillaria gallica, two as A. nabsnona and one as A. cepistipes.Ofthese, A. gallica was the dominant species. A. nabsnona and A. cepistipes are reported here as symbionts of Gastrodia elata for the first time. One strain of the symbiont, A. gallica was used for cul- tivation test. After 32 months cultivation, G. elata tubers had increased from 430 g to 2560 g in weight, and from 18 to 236 in number. Significant differences in the amount of ash, diluted ethanol extract and drying loss were not observed between the cultivated tubers and commercial ones.

Key words: Armillaria, cultivation, Gastrodia elata,mating test, symbiont.

Gastrodia elata Blume is a non- symbiont of non-chlorophyllous orchid spe- chlorophyllous orchid species that is broadly cies (Wargo and Shaw 1985). distributed in temperate and boreal regions Mating tests based on sexual behavior and of , Korea and Japan (Makino 1982, in vitro somatic compatibility of isolates of Lee 1979). Gastrodia elata has a symbiotic Armillaria has been widely used to assess relationship with Armillaria species from interspecific incompatibility in the genus which they obtain nutrition (Kusano 1911). Armillaria. Gastrodia elata is well known as an impor- Species of Armillaria from Europe, North tant substance for Chinese herbal medicine America and Australia have been identified in Asian countries (Hatakoshi 1947). and their morphological characteristics have Armillaria is well known as an important been described in detail (Korhonen 1978, pathogen (Thomas 1934). In Japan, Guillaumin et al. 1993, Anderson and Ullrich Armillaria root rot has been reported in 1979, Kile and Watling 1983). In Japan, at plantations of young Chamaecyparis obtusa least ten species of Armillaria have been re- (Muramoto 1988). The pathogenic role of ported (Nagasawa et al. 1991, Suzuki et al. Armillaria has been controversial histori- 1994, Cha et al. 1992, 1994, Ota et al.1998). cally. The fungus has been described vari- Cha and Igarashi (1995) identified ously as an aggressive killer of healthy trees, Armillaria species obtained from G. elata a secondary pathogen of stressed trees, a tubers in Hokkaido, Japan, but other regions saprophyte decayer of dead trees and a in Japan are not yet identified. Though G.

—88— April 2008 Journal of Japanese Botany Vol. 83 No. 2 89

Fig. 1. Mycelial coils of Armillaria gallica formed in the cortical cells of Gastrodia elata tuber. Bar = 10 m. elata has been produced by cultivation in coils of the mycorrhizal tubers were ob- China, symbiotic Armillaria species are not served within cortical cells of the tubers sufficiently clarified. There were no reports under a microscope (Fig. 1). on field cultivation of G. elata by using sym- biosis with Armillaria in Japan. In order to Isolation of fungal symbiont establish a G. elata cultivation method in The Armillaria symbiont of each G. elata Japan, it is necessary to cultivate G. elata collection was isolated using the method of with Armillaria dominant species in Japan previous report (Kikuchi et al. 2008). and to evaluate the quality of produced tubers. Mating test The objective of this study are to describe Mating test of Armillaria is attempted on the Armillaria species obtained from G. haploid-haploid pairings usually. The isolate elata tubers in Honshu, Japan, to cultivate G. from fruiting body or rhizomorph yield elata using dominant species of Armillaria in depressed, crustose colony, whereas mono- Japan, to evaluate the quality of produced sporous tester strain yield colonies with a tuber preliminarily. fluffy, aerial aspect. In compatible parings, the tester strain of fluffy mycelium is trans- Materials and Methods formed into a crustose colony. This change Collection of Gastrodia elata tubers was often slow as compared with the change Tubers were collected at eleven sites taking place in matings between two com- throughout Honshu, Japan (Table 1). The patible haploids. Because haploid isolates collection method of G. elata tubers and were not available, the diploid isolates from symbiotic Armillaria were followed with our G. elata were paired with haploid testers in previous study (Kikuchi et al. 2008). Hyphal this time. 90 植物研究雑誌 第83巻第2号平成20年4月

Table 1. Origin of vegetative (diploid) isolates of Armillaria from Gastrodia elata.

Armillaria Gastrodea elata collection isolate Date Locality Host Habitat Ar18 05/06/95 O¯ hito-cho, Takata-gun, Shizuoka Pref. unidentified trees Cryptomeria japonica plantation Ar28 31/01/96 Ryûgasaki City, Ibaraki Pref. Diospyrtos kaki summer-green forest Ar30 31/01/96 Ushiku City, Ibaraki Pref. Quercus acutissima summer-green forest Ar31 20/06/96 O¯ hito-cho, Takata-gun, Shizuoka Pref. Quercus acutissima summer-green forest Ar36 15/11/96 Ushiku City, Ibaraki Pref. Quercus acutissima summer-green forest Ar38 30/07/97 Katashina-mura, Tone-gun, Gunma Pref. unidentified trees summer-green forest Ar40 28/07/97 Nikkou City, Tochigi Pref. Quercus. mongolica summer-green forest Ar41 30/07/97 Katashina-mura, Tone-gun, Gunma Pref. Salix vulpina summer-green forest Ar42 12/08/97 Azumi-mura, Minamiazumi-gun, Nagano Betula ermanii summer-green forest Pref. Ar43 12/08/97 Azumi-mura, Minamiazumi-gun, Nagano Betula ermanii summer-green forest Pref. Ar52 07/07/03 Ushiku City, Ibaraki Pref. Quercus acutissima summer-green forest

Host tree identification was determined with tracing Armillaria rhizomorphal connection.

The unknown Armillaria diploid isolates Table 2. Stock number of haploid tester strains of from G. elata were mated with haploid tester Armillaria. strains of several Armillaria species. Tester Number Stock No. of isolates strains were listed in Table 2. Mating experi- ments followed the method of previous study 1 A. ostoyae TMIC 30966-1 (Kikuchi et al. 2008). 2 A. ostoyae TMIC 31491-9 3 A. gallica TMIC 30984-11 4 A. gallica TMIC 31063-3 Cultivation in the field 5 A. cepistipes TMIC 30970-2 One strain of G. elata symbiont, 6 A. cepistipes TMIC 31069-3 Armillaria gallica isolated from Ryugasaki 7 A. nabsnona TMIC 30981-3 City, Ibaraki Prefecture was used for the 8 A. nabsnona TMIC 30987-2 9 A. ISG E TMIC 31067-10 cultivation test in Ami-machi, Ibaraki Pre- 10 A. ISG E TMIC 31515-8 fecture. To obtain inocula, small wood logs (length diameter = 20 7–10 cm) were placed in 2 Liter polypropylene bottles and autoclaved for 120 minutes at 121ºC then were laid on the layer of fallen leaves with 5 inoculated with the fungal symbiont. The cm spacing from each wood log. Five small inoculated wood logs were kept 3 months at wood logs colonized by rhizomorphs were 25ºC in the dark until rhizomorph formed ac- inoculated between large wood logs and then tively. To produce tuber of G. elata, cultiva- filled with fallen broad leaves in the space tion site was dug out to the size of 100 100 between wood logs. With repetition of this 40 cm (width length depth) in the field. way, the inoculation was conducted by two The bottom of the site was covered with a layers of wood logs. The inoculated wood layer of fallen leaves about 5 cm thick. Five logs were covered with fallen broad leaves large wood logs (length diameter = 90  and then covered with soil layer about 10 cm 7–10 cm) as nutrient source of A. gallica and incubated for one year. After incubation, April 2008 Journal of Japanese Botany Vol. 83 No. 2 91

Fig. 2. Cultivation method of Gastrodia elata tubers. gt. Inoculated Gastrodia elata tuber. bl: Fallen broad leaves. lw. Large wood logs. sl. Soil layer. sw. Small wood logs colonized by rhizomorphs of Armillaria gallica.

G. elata tubers were inoculated between given in Table 4. wood logs where rhizomorph formed ac- tively (Fig. 2). Results Identification of species of Armillaria iso- Quality evaluation of cultivated tubers lates Tubers cultivated in the field were dried The results of mating tests of the eleven at 30ºC for 1 week. The commercial samples Armillaria isolates of diploids are detailed in of G. elata were purchased from Yamamoto Table 1. Among these isolates, eight isolates Yakuhin Co., Ltd., Japan. These plant mate- were identified as A. gallica, two were iden- rials deposited in our laboratory were identi- tified as A. nabsnona and one was identified fied by Dr. M. Higuchi of our institute. as A. cepistipes (Table 3). These isolates These dried tubers and five commercial sam- were clearly compatible with the haploid ples from China were powdered. testers of the corresponding species. The powder of the G. elata tuber was ex- tracted with ethanol at room temperature for Cultivation in the field 10 minutes. The extract was concentrated Inoculation of Gastrodia elata tubers was and analyzed with TLC with chloroform- performed at Ami-machi, Ibaraki Prefecture methanol-water (13:7:2). The spots were de- in September 1997. After five month incuba- tected under UV (254 nm). Then, the plate tion, rhizomorphs had adhered to the tubers. was sprayed with a p-anisaldehyde-sulfuric These tubers formed many buds. Eight G. acid solution and heated at 130ºC for ap- elata tubers sprouted in May and flowered in proximately 1 min. June, 2000. After 32 months cultivation, G. The amount of ash, diluted ethanol extract elata tubers were harvested. The weight of and drying loss of the tubers were measured the tubers had increased from 430 g to 2,560 according to the methods of the Pharma- g (yield increase was 5.95 times), and the copoeia of Japan. Averages of three meas- number of tubers had increased from 18 to urements with the standard deviation are 236. There were two types of tubers. One 92 植物研究雑誌 第83巻第2号平成20年4月

Table 3. Results of mating tests between Armillaria vegetative (diploid) isolates from Gastrodia elata and haploid testers

Tester Isolate A. ostoyae A. gallica A. cepistipes A. nabsnona A. ISG E (diploid) 12 34 56 78 910 Ar18 – – + + – – – – – – Ar28 – – + + – – – – – – Ar30 – – + + – – – – – – Ar31 – – + + – – – – – – Ar36 – – + + – – – – – – Ar38 – – + + – – – – – – Ar40 – – + + – – – – – – Ar41 – – – – + + – – – – Ar42 – – – – – – + + – – Ar43 – – – – – – + ? – – Ar52 – – + + – – – – – –

?, uncertain pairing, +, compatible pairing; –, incompatible pairing.

was adhered to by rhizomorphs of A. gallica on the surface and the other was not. A part of the tubers with adhered rhizomorphs formed Mycorrhiza. The rhizomorphs of A. gallica were developed well around the tubers, between soil surface and 10 cm of soil depth, and also colonized favorably on the surface of the wood logs. The soil was colonized to a depth of about 30 cm, rhizo- morph development was rarely observed and colonized poorly on the surface of the wood logs.

Quality evaluation of cultivated tubers No significant difference in the TLC fluo- rescent spot pattern was observed between the dried tubers of cultivated G. elata and commercial samples. After p-anisaldehyde- sulfuric acid solution spraying no significant difference was observed in the TLC profiles, and all the samples had the spot of the pur- plish red color at about Rf value 0.4 either Fig. 3. TLC pattern of cultivated tuber and commer- (Fig. 3). cially obtained Gastrodia elata tubers from China. Lane 1. China. Lane 2. Honan Prov., The ash content of the cultivated G. elata China. Lane 3. Guizhou Prov., China. Lane 4. tubers was 4.3, diluted ethanol extract was Anhwei Prov., China. Lane 5. Korea. Lane 6. 23.6 and drying loss value of the tubers Cultivated. April 2008 Journal of Japanese Botany Vol. 83 No. 2 93

Table 4. Amounts of ash, drying loss value and diluted ethanol extract of Gastrodia elata tubers cultivated and markets (n = 3)

Sample No. Location Drying loss value Amount of ash Diluted ethanol extract Date Cultivated Ibaraki Pref, Japan 9.170.25 4.030.02 23.580.09 26/06/2000 THS 82191 Korea 8.300.30 3.120.10 21.411.45 26/03/1976 THS 82192 Guizhou Prov., China 8.000.20 4.020.04 21.790.17 09/10/1990 THS 82193 China 8.200.26 4.150.14 22.600.12 05/01/1984 THS 82194 Anhwei prov., China 8.130.55 2.100.21 23.170.58 26/03/1980 THS 82195 Henan Prov., China 7.300.26 4.530.02 23.650.05 13/09/1996 were 9.2. Four commercial samples are f. Our test results cause us to conclude that given in Table 4. saprophytic or weak-pathogenic Armillaria species are the main symbionts of G. elata. Discussion In all the inoculated tubers, a part of A total of 11 Armillaria isolates from them was adhered to and penetrated by the Gastrodia elata were identified as three spe- rhizomorphs of A. gallica. Then the fungal cies by using mating test. All isolates paired mass of rhizomorph was digested with the with tester strains showed clear reaction of tubers as nutrition. The inoculated tubers did compatibility. not grow themselves but produced some new Identified species associated with G. elata tubers. The tubers adhered to and penetrated in this study were A. gallica, A. nabsnona by rhizomorphs digested fungal mass and and A. cepistipes.Ofthese, A. gallica was provided nutrition to new daughter tubers. the most common (72.7). Armillaria This growth process corresponds to the result nabsnona and A. cepistipes were 18.2 and of in vitro two-member culture (Kikuchi 9.1, respectively. and et al. 2008). A. cepistipes are reported here as symbionts Mycorrhizal tubers formed many new of Gastrodia elata for the first time. Cha and tubers spreading at the depths of 5–15 cm in Igarashi (1995) reported that the Armillaria soil. And these new daughter tubers last for species isolates from G. elata in Hokkaido long time without decomposition (data not were A. gallica, A. ostoyae, A. jezoensis, shown). This facilitates eventual meeting A. sinapina, and A. singula. Among them, with Armillaria.IfGastrodia tuber promotes A. gallica was the most common. Of these mycelial growth and rhizomorph formation species, A. jezoensis, A. sinapina and as the Galeola roots, it will encourage spread A. singula have not been reported from of new tubers to form mycorrhiza with Honshu, and A. ostoyae was not isolated Armillaria.Asamatter of fact, the new from G. elata in this study. Moreover tubers were not adhered to by rhizomorphs at A. cepistipes and A. nabsnona were associ- first. Rhizomorph extended from mother ated with G. elata. Ota et al. (1998) reported tuber to daughter tubers and adhered to them. that A. gallica seems to be a weak pathogen In many cases, this rhizomorph was elon- most commonly found on dead hardwoods gated from ancestor tuber in the fields. After and the soil surface. Terashita and Chuman a while, the new tubers were separated from (1989) concluded that A. gallica and A. the old tuber. This study was the first report cepistipes were saprophytic or weak- of G. elata cultivation using dominant spe- pathogenic and they were the main cies of Armillaria gallica in Japan. symbionts of Galeola septentrionalis Rchb. According to the Pharmacopeia of Japan 94 植物研究雑誌 第83巻第2号平成20年4月

(2006) the ash content of G. elata tuber Hatakoshi M. 1947. Illustrations of Medicinal . should be less than 6. The ash content of Tokyonougyousyoin, Tokyo (in Japanese). the dried tuber in cultivation was 4.03. Kikuchi G., Higuchi M., Yoshimura H., Morota T. and Diluted ethanol extract, drying loss value as Suzuki A. 2008. In vitro Symbiotic Examination with Gastrodia elata Blume (Orchidaceae) and well as TLC profiles of the present dried Armillaria Kummer (Tricholomataceae) Species tubers of cultivation were all about the same Isolated from the Orchid tuber. J. Jpn. Bot. 83: as those of commercial samples. These re- 00–00. sults indicate that dried tubers cultivated Kile G. A. and Watling R., 1983. Armillaria species with A. gallica have same quality as com- from south-eastern Australia. Trans. Br. Mycol. Soc. 81: 129–140. mercial samples. Korhonen K. 1978. Interfertility and clonal size in the complex. Karstenia 18: 31–42. We are grateful to Dr. Y. Ota (Forestry Kusano S. 1911. Gastrodia elata and its symbiotic as- and Forest Products Research Institute) for sociation with Armillaria mellea.J.Coll. Agric., valuable information and supplying litera- Imp. Univ. Tokyo, 4: 1–65. ture. We are also grateful to Dr. M. Okada Lee T. B. 1979. Illustrated Flora of Korea. Hyang- Moon-Sa Co., Seoul. (The Kochi Prefectural Makino Botanical Makino T. 1982. Tokyo. In: Hamada M. (ed.), Garden) and Dr. S. Terabayashi (Yokohama Makino’s Illustrated Flora in Colour. Hokuryukan College of Pharmacy) for helpful discus- Co., Tokyo (in Japanese). sions. We also thank to Mr. S. Minoura, Mr. Muramoto M. 1988. Occurrence of Armillaria root rot N. Igari, Mr. M. Tanaka and the members of of Chamaecyparis obtusa. Forest Pest 37: 65–70. Nagasawa E., Komatsu M. and Maekawa N. 1991. Botanical Raw Materials Division, Tsumura Taxonomic reassessment of Armillaria mellea in & Co., for useful information and their sug- Japan. Report for a Grant-in-Aid for Scientific gestions. Research No. 63560155, Ministry of Education, Science and Culture of Japan. References Ota Y., Matsushita N., Nagasawa E., Terashita T., Fukuda K. and Suzuki K. 1998. Biological species Anderson J. B. and Ullrich R. C. 1979. Biological spe- of Armillaria in Japan. Plant Disease 82: 537–543. cies of Armillaria mellea in North America. Suzuki K., Fukuda K., Shiga Y. and Matsushita N. Mycologia 71: 402–414. 1994. Identification of Japanese biological species Cha J. Y. and Igarashi T. 1995. Armillaria species as- of Armillaria by isozyme patterns. pp. 364–370. sociated with Gastrodia elata in Japan. Eur. J. For. In: Johansson M. and Stenlid J. (eds.), Proceedings Path. 25: 319–326. of the 8th Inter. Conference on Root and Butt Rots, Cha J. Y., Sung J. M. and Igarashi T. 1992. Biological IUFRO Working Party, Wik and Haikko. species and morphological characteristics of Terashita T. and Chuman S. 1989. Armillarias, isolated Armillaria mellea complex in Hokkaido: from the wild orchid, Galeola septentrionalis. In: A. ostoyae and A. bulbosa. Res. Bull. Exp. For. Morrison D. J. (ed.), Proceedings of the 7th Hokkaido Univ. 49: 185–194. Interational Conference on Root and Butt Rots, Cha J. Y., Sung J. M. and Igarashi T. 1994. Biological IUFRO Working Party. pp. 364–370. Forestry species and morphological characteristics of Canada, Pacific Forestry Centre, Victoria. Armillaria mellea complex in Hokkaido: The Society of Japanese Phamacopoeia (ed.). 2006. A. sinapina and two new species, A. jezoensis and Japanese Pharmacopoeia 15th ed. 2006 pp. A. singula. Mycoscience 35: 39–47. Hirokawa Shoten, Tokyo. Guillaumin J. J., Mohammed C., Anselmi N., Thomas H. E. 1934. Studies on Armillaria mellea Courtecuisse R., Gregory S. C., Holdenrieder O., (Vahl) Quel., infection, parasitism, and host resis- Intini M., Lung B., Marxmüller H., Morrison D., tance. J. Agric. Res. 48: 187–218 Rishbeth J., Termorshuizen A. J., Tirró A. and Van Wargo P. M. and Shaw C. G. 1985. Armillaria Root Dam B. 1993. Geographical distribution and ecol- Rot: The puzzle is being solved. Plant Disease 69: ogy of the Armillaria species in western Europe. 826–832. Eur. J. For. Pathol. 23: 321–341. April 2008 Journal of Japanese Botany Vol. 83 No. 2 95

菊地 原a,樋口正視a,諸田隆a,長澤栄史b,鈴 木彰c:オニノヤガラ (ラン科) の共生菌の同 定と栽培実験 頻度の高い A. gallica を用いて栽培試験を実施し 本州の11ヶ所よりオニノヤガラ塊茎を採取した. た. その結果, 32ヶ月間の培養で接種した430 g この塊茎より共生菌であるナラタケを分離し, ナ から2,560 g, 18個から236個に増殖した. 灰分, ラタケテスター菌株との交配試験を用いて種を同 希エタノールエキス, 乾燥減量を計測したところ 定した. その結果, 分離した11菌株のうち 8 種が 栽培したオニノヤガラ塊茎と市場品 4 種の間に差 Armillaria gallica, 2 種が A. nabsnona, 1 種が A. 異は認められなかった. cepistipes であることが判明した. これらのうち A. (a株式会社ツムラ生薬本部生薬研究部, gallica が優占種であった. また, A. nabsnona と b財団法人日本きのこセンター菌蕈研究所, A.cepistipes が新たにオニノヤガラに共生するこ c千葉大学教育学部) とが明らかになった. これら共生菌のうち, 最も