内生真菌

RESEARCH New Phytol. (2000), 147, 617–630 Identification of endophytic fungi from Livistona chinensis based on morphology and rDNA sequences L. D. GUO, K. D. HYDE* and E. C. Y. LIEW Centre for Research in Fungal Diversity, Department of Ecology and Biodiversity, The University of Hong Kong, Pokfulam Road, Hong Kong Received 29 November 1999; accepted 15 May 2000 summary A survey of the endophytic fungi in fronds of Livistona chinensis was carried out in Hong Kong. The endophyte assemblages identified using morphological characters consisted of 16 named species and 19 ‘morphospecies ’, the latter grouped based on cultural morphology and growth rates. Arrangement of taxa into morphospecies does not reflect species phylogeny, and therefore selected morphospecies were further identified based on ribosomal DNA (rDNA) sequence analysis. The 5–8S gene and flanking internal transcribed spacers (ITS1 and ITS2) regions of rDNA from 19 representative morphospecies were amplified by the polymerase chain reaction and sequenced. Phylogenetic analysis based on 5–8S gene sequences showed that these morphospecies were filamentous Ascomycota, belonging in the Loculoascomycetes and Pyrenomycetes. Further identification was conducted by means of sequence comparison and phylogenetic analysis of both the ITS and 5–8S regions. Results showed that MS704 belonged to the genus Diaporthe and its anamorph Phomopsis of the Valsaceae. MS594 was inferred to be Mycosphaerella and its anamorph Cladosporium of the Mycosphaerellaceae. MS339, MS366, MS370, MS395, MS1033, MS1083 and MS1092 were placed in the genus Xylaria of the Xylariaceae. MS194, MS375 and MS1028 were close to the Clypeosphaeriaceae. MS191 and MS316 were closely related to the Pleosporaceae within the Dothideales. The other 5 morphospecies, MS786, MS1043, MS1065, MS1076 and MS1095, probably belong in the Xylariales. The value of using DNA sequence analysis in the identification of endophytes is discussed. Key words: DNA identification, fungal morphospecies, ITS, 5–8S gene, sequence analysis. introduction There have been many studies on endophytic fungi in grasses and woody plants in temperate regions (Redlin & Carris, 1996). Several endophyte studies have also recently been undertaken on tropical hosts (Pereira et al., 1993; Fisher et al., 1995; Lodge et al., 1996; Umali et al., 1999), including palms (Rodrigues & Samuels, 1990; Rodrigues, 1994; Southcott & Johnson, 1997; Fro$ hlich et al., 2000). In any study on endophytes it is important that authors define the understanding of the term. In this study we use the term endophyte to mean ‘all organisms inhabiting plant organs that at some time in their life, can colonize internal plant tissue without causing apparent harm to the host ’ (Petrini, 1991), although we only study fungi. This definition has been widely used throughout the literature and uses *Author for correspondence (tel ›825 2975 5632; fax ›852 2517 6082; e-mail kdhyde!hkucc.hku.hk). a traditional approach for the isolation of endophytes (Taylor et al., 1999). This approach involves surface sterilization of material and isolation of fungi from the internal parts. This might include some saprobes, the mycelia of which have penetrated external organs such as stomata, but can still be accommodated using Petrini’s definition. The extent of colonization of endophytes isolated using this method is unknown. This traditional approach to the study of endophytes was used in all the studies mentioned above. Many of the isolates obtained by traditional methods can only be identified to mycelia sterilia and were often grouped as ‘morphospecies’ based on similar cul- tural characters (Umali et al., 1999; Fro$ hlich et al., 2000). This inability to identify mycelia sterilia is com- mon to previous studies of endophytes where various proportions of mycelia sterilia have been reported. Thirteen percent of endophytes obtained from two Licuala species in Brunei and Australia, and 11% of isolates from the palm Trachycarpus fortunei in 618 RESEARCH L. D. Guo et al. China were mycelia sterilia (Taylor et al., 1999; Fro$ hlich et al., 2000). Approximately 15% of endophytes isolated from evergreen shrubs in west- ern Oregon (Petrini et al., 1982) and c. 26–9% of endophytes isolated from leaves of Sequoia semper- virens (coastal redwood) in central California were mycelia sterilia (Espinosa-Garcia & Langenheim, 1990). A much higher percentage of mycelia sterilia (54%) was obtained from twigs of Quercus ilex in Switzerland, whereas 18% of isolates from the leaves were mycelia sterilia (Fisher et al., 1994). Various percentages of mycelia sterilia have also been reported in other endophyte studies from different hosts (e.g. Johnson & Whitney, 1992; Fisher et al., 1993, 1995). The high proportion of unidentified endophyte isolates resulting from traditional methodologies has prompted various workers to develop methods to promote sporulation in these mycelia sterilia. Attempts to increase the number of endophytes identified include several approaches, such as the use of different media and inclusion of host tissues in plate cultures to promote sporulation (Matsushima, 1971; Taylor et al., 1999; Fro$ hlich et al., 2000). Guo et al. (1998) increased the initial 48% of sporulating isolates obtained from Livistona chinensis fronds to 59–5%, by the addition of sterile palm leaf tissue onto the agar surface. By inoculating the remaining unidentified isolates onto sterile petiole pieces in conical flasks and incubating them at room tem- perature for 3 months, Guo et al. (1998) further increased the number of identifiable (sporulating) isolates to 83–5% Even with this improved method, 16–5% of isolates obtained would still not sporulate, and could only be classified as mycelia sterilia. These isolates were subsequently grouped into different morpho- species based on culture morphology, that is, colony colour, texture and growth rates (Fro$ hlich et al., 2000). Arrangement of taxa into morphospecies, however, does not reflect species phylogeny, and hence alternative approaches are required for the identification of these fungi. Molecular techniques are commonly used to solve problems in fungal taxonomy (Takamatsu, 1998; Ranghoo et al., 1999) and in several recent studies have also been used in the identification of fungi (Rollo et al., 1995; Ma et al., 1997; Zhang et al., 1997). The basic aim of the present study is to identify as many endophytes as possible from a limited number of palm fronds using traditional morphological techniques and molecular techniques. For the molecular approach, our strategy was to identify these mycelia sterilia by means of rDNA sequence comparison (i.e. assessment of percentage nucleotide similarity with reference sequences), as well as phylogenetic analysis conducted in several stages of varying taxonomic resolutions. The 5–8S gene and ITS regions are used in this study. Because the 5–8S gene is highly conserved, this region is used for the phylogenetic analysis of higher taxonomic levels, whereas the highly variable ITS regions are used for analysis of lower taxonomic levels. materials and methods The Chinese fan palm Livistona chinensis (Jacq.) R. Br. is broadly distributed and cultivated in tropical and temperate regions, such as Hong Kong, southern mainland China, Japan and the Ryukyu Islands. This palm grows up to 20 m high, with fan-shaped fronds up to 1 m in diameter and petioles up to 2 m long. Site and sampling procedure Livistona chinensis plants chosen for this study are localized in mixed woodland at the margins of a permanent stream, adjacent to the campus of The University of Hong Kong in Lung Fu Shan. Three individual mature palms were selected for this investigation in order to minimize variability. The palms were about the same age and c. 20 m apart. One young and one old healthy frond were randomly selected from each tree in February 1997. Samples were processed within 24 h of collection. Isolation and culture of endophytic fungi The sampling regime was designed with the in- tention of isolating as many endophyte species as possible from a few palm fronds. Each selected frond was therefore divided into two parts (leaf and petiole). Five random segments (10‹80 cm) were then chosen from each leaf and 20 discs (6 mm diameter, including 10 vein and 10 intervein discs) were removed from each segment. In total, 600 (300 vein and 300 intervein) discs were cut from six leaves (3 young and 3 old leaves) of the three plants. Twenty-five random segments (3–4 cm long) were also cut from each petiole and one disc (6 mm diameter) was removed from one petiole segment. A total of 150 petiole discs were cut from six petioles (3 young and 3 old petioles) from the three plants. A total of 750 discs from both leaf and petiole were used in this study. Frond surface texture is similar in most palms, and therefore the method of surface sterilization employed in this study was similar to that used in previous studies of palm endophytes (Rodrigues, 1994; Taylor et al., 1999; Fro$ hlich et al., 2000). The discs were surface sterilized by consecutive im- mersion for 1 min in 75% ethanol, 10 min in 65% commercial Chlorox (final concentration 3–25% aqueous sodium hypochlorite) and 30 sec in 75% ethanol. The discs were surface dried with sterile paper towels. Sets of four discs were then evenly placed in each 90 mm Petri dish containing malt extract agar (MEA, 2%; Difco, Detroit, MI, USA) RESEARCH Identification of endophytic fungi from Livistona chinensis 619 supplemented with 1 mg ml−" streptomycin sulphate (Sigma, St Louis, MO, USA). In order to isolate slow growing fungi, rose bengal (0–03 mg ml−") was added to the medium to inhibit fast growing fungi. Petri dishes were sealed, incubated for 2 months at 25°C, and examined periodically. When colonies developed, they were transferred to new Petri dishes with MEA. Subcultures were then incubated on different media which included potato dextrose agar (PDA, 2%, Difco), corn meal agar (CMA, 2%, Difco), and tap water agar (TWA, 0–8%, Difco). In addition, a sterile leaf strip of L. chinensis was also included to promote sporulation (Matsushima, 1971; Taylor et al., 1999). Isolates were incubated at 25°C, with cool white fluorescent light for a light regime of 12:12 h light:dark. Some isolates did not sporulate under these incubation conditions. Attempts were made to promote sporulation using an improved method; that is, sterile petiole fragments of L. chinensis were placed into conical flasks and inocu- lated with sterile isolates, and incubated in the same conditions as already described (Guo et al., 1998). Identification of endophytic fungi using traditional techniques Subcultures on different media were examined periodically and identified when isolates sporulated. Slides were mounted in lactophenol and sealed with nail varnish. These slides were labelled and colonies of sporulating species were dried and stored in the herbarium of The University of Hong Kong (HKU(M)). Living cultures are deposited in The University of Hong Kong Culture Collection (HKUCC). The remaining cultures which failed to sporulate after the treatments were named mycelia sterilia. These mycelia sterilia were divided into 19 different ‘morphospecies’ according to cultural characteristics (i.e. colony colour, texture, habit and growth rate) on MEA (Fro$ hlich et al., 2000). Sources of fungal isolates used in the molecular techniques A representative isolate of each of the 19 morpho- species was included in the molecular part of this study and prefixed MS (Table 1). Reference taxa with DNA sequences obtained from GenBank (National Center for Biotechnology Information, Bethesda, MD, USA) and EMBL (European Mol- ecular Biology Laboratory, Heidelberg, Germany), along with their accession numbers, are also listed in Table 1. In addition to the 19 morphospecies, we also sequenced an isolate of Xylaria hypoxylon, and four Xylaria species isolated as endophytes in this study. Because these four Xylaria species only formed immature stromata covered with asexual spores in culture and did not produce sexual spores, they could not be identified to species level and are named Xylaria sp. 1–4 (Rodrigues & Samuels, 1990; Rodrigues et al., 1993). DNA extraction from mycelia A hyphal tip was obtained from each fresh culture using a dissecting microscope and was grown on MEA in the dark at 25°C for 3–30 d. Genomic DNA was extracted from fresh cultures using a modified protocol of Doyle & Doyle (1987) and Lee & Taylor (1990). Fresh fungal mycelia (c. 50 mg) was scraped from the surface of the agar plate and transferred into a 1–5 ml microcentrifuge tube with 700 ll of preheated (60°C) 2X CTAB extraction buffer (2% (w}v) CTAB, 100 mM Tris-HCl, 1–4 M NaCl, 20 mM EDTA, pH 8–0), and c. 0–2 g sterilized quartz sand (Sigma). Fungal mycelia were ground using a glass pestle for 5–10 min and then incubated in a 60°C water bath for 30 min with occasional gentle swirling. 500 ll of phenol :chloroform (1:1) was added into each tube and mixed thoroughly to form an emulsion. The mixture was spun at 11 900 g for 15 min at room temperature in a microcentrifuge and the aqueous phase was removed into a fresh 1–5 ml tube. The aqueous phase containing DNA was re- extracted with chloroform:isoamyl alcohol (24:1) until no interface was visible. 50 ll of 5 M KOAc was added into the aqueous phase followed by 400 ll of isopropanol and inverted gently to mix. The genomic DNA was precipitated at 9200 g for 2 min in a microcentrifuge. The DNA pellet was washed with 70% ethanol twice and dried using SpeedVac2 (AES 1010; Savant, Holbrook, NY, USA) for 10 min or until dry. The DNA pellet was then resuspended in 100 ll TE buffer (10 mM Tris-HCl, 1 mM EDTA). PCR amplification of the 5–8S gene and flanking ITS Primers ITS5 and ITS4 (White et al., 1990) were used to amplify the 5–8S and flanking ITS regions of the morphospecies. The DNA fragment was ampli- fied in an automated thermal cycler (PTC-1004, MJ Research, Inc., Watertown, MA, USA). Ampli- fication was performed in a 50 ll reaction volume which contained PCR buffer (10 mM KCl, 10 mM (NH % ) # SO % , 20mM Tris-HCl, pH 8–8, 0–1% Triton X-100), 1–5 mM MgCl # , 200 lM of each deoxy- ribonucleotide triphosphate, 15 pmols of each primer, c. 100 ng template DNA, and 2–5 units of Taq DNA polymerase (Promega, Madison, WI, USA). The thermal cycling program was as follows: 3 min initial denaturation at 95°C, followed by 35 cycles of 40 sec denaturation at 94°C, 50 s primer annealing at 52°C, 1 min extension at 72°C, and a final 10 min extension at 72°C. A negative control using water instead of template DNA was included 620 RESEARCH L. D. Guo et al. Table 1. Taxa and GenBank sequences used in the study Taxa GenBank}EMBL accession number* Ascomycota Alternaria alternata (Fr. : Fr.) Keissl. U05195 Amphisphaeria umbrina (Fr.) De Not. AF009805 Atrotorquata lineata Kohlm. and Volkm.-Kohlm. AF009807 Bipolaris oryzae (Breda de Hann) Schoemaker Y10856 Botryosphaeria dothidea (Moug.: Fr.) Ces. & De Not. AF027752 Capsulospora sp. AF009819 Cladosporium cladosporioides (Fresen.) G. A. De Vries L25429 Cladosporium fulvum Cooke L25430 Cladosporium herbarum (Pers. : Fr.) Link L25431 Cladosporium oxysporum Berk. & M. A. Curtis L25432 Cladosporium sphaerospermum Penz. L25433 Cladosporium tenuissimum Cooke Y15966 Cladosporium sp. AJ222808 Clypeosphaeria mamillana (Fr.) Lamb. AF009808 Cryphonectria cubensis (Bruner) C. S. Hodges AF046900 Cryphonectria gyrosa (B. et Br.) Sacc. AF046905 Cryphonectria parasitica (Murrill) Barr AF046903 Dendryphion penicillatum (Corda) Fr. AF102889 Diaporthe ambigua Nitschke AF046906 Diaporthe meridionalis Sacc. AF001015 Diaporthe phaseolorum (Cke. & Ell.) Sacc. AF001023 Diaporthe phaseolorum var. caulivora Athow & Caldwell AF000567 Discostroma tosta (Berk. and Broome) Brockmann AF009814 Dothidea hippophaeos (Passerini) Fuckel AF027763 Drechslera avenae (Eidam) Scharif X78125 Ellurema indica (Punith.) Nag Raj and Kendr. AF009816 Elsinoe australis Bitancourt & Jenk. U28057 Endothia eugeniae (F. J. Nutman & F. M. Roberts) J. Reid & C. Booth AF046904 Eurotium rubrum U18357 Gelasinospora nigeriensis AJ002400 Glomerella graminicola Politis AF059676 Hypocrea rufa (Persoon) Fr. X93980 Lepteutypa cupressi (Nattrass, Booth and Sutton) Swart AF009817 Leptosphaeria bicolor D. Hawksw., W. J. Kaiser & Ndimande U04203 Metschnikowia bicuspidata (Metschnikov) Kamenski U51436 Morchella crassipes AF008232 Mycosphaerella graminicola (Fuckel) J. Schr u ot. in Cohn U77363 Ophiosphaerella herpotricha (Fr. : Fr.) J. C. Walker U04861 Ophiostoma ulmi (Buisman) Nannf. U23423 Pestalosphaeria elaeidis (C. Booth and J. S. Robertson) H. A. van der Aa AF009815 Phaeocryptopus gaeumannii (T. Rohde) Petr. AF013225 Phaeosphaeria avenaria (G. F. Weber) O. Eriksson U77357 Phomopsis longicolla Hobbs U97658 Phomopsis sp. 1 U94898 Phomopsis sp. 2 U91617 Pleospora herbarum (Pers. : Fr.) Rabenh. U05202 Pyrenophora teres Drechs. Y08746 Rhabdocline parkeri U92297 Saccharomyces cerevisiae Meyen ex E. C. Hansen Z95941 Xylaria hypoxylon (L.: Fr.) Grev. AF194027 Xylaria sp. 1 AF153724 Xylaria sp. 2 AF153725 Xylaria sp. 3 AF153726 Xylaria sp. 4 AF153727 Basidiomycota Lentinula edodes (Berk.) Pegler U33093 Pisolithus tinctorius (Pers.) Coker et Couch AF004737 Pleurotus pulmonarius (Fr.) Que! l. U60648 Puccinia carcina de Candolle U88234 Tremella foliacea Pers. : Fr. AF042452 Zygomycota Glomus mosseae (T. H. Nicolson & Gerd.) Gerd. & Trappe U31996 RESEARCH Identification of endophytic fungi from Livistona chinensis 621 Table 1 (cont.) Taxa GenBank}EMBL accession number* Morphospecies MS191 AF153728 MS194 AF153729 MS316 AF153730 MS339 AF153731 MS366 AF153732 MS370 AF153733 MS375 AF153734 MS395 AF153735 MS594 AF153736 MS704 AF153737 MS786 AF153738 MS1028 AF153739 MS1033 AF153740 MS1043 AF153741 MS1065 AF153742 MS1076 AF153743 MS1083 AF153744 MS1092 AF153745 MS1095 AF153746 *GenBank (NCBI, National Centre for Biotechnology Information, Bethesda, MD, USA); EMBL (European Molecular Biology Laboratory, Heidelberg, Germany). in the amplification process. From each PCR reaction 4 ll of PCR products were examined by electrophoresis at 75 V for 2 h in a 0–8% (w}v) agarose gel in 1‹TAE buffer (0–4 M Tris, 50 mM NaOAc, 10 mM EDTA, pH 7–8) and visualized under UV light after staining with ethidium bromide (0–5 lg ml−"). DNA sequencing PCR products were purified using minicolumns (Wizard PCR Preps DNA Purification System, Promega) according to the manufacturer’s protocol. Purified PCR products were directly sequenced in an automated sequencer (ALFexpress, Pharmacia- Biotech, Piscataway, NJ, USA) following the manu- facturers instructions. Primers ITS5, ITS2, ITS3 and ITS4 (White et al., 1990) were used in the sequencing reactions. Both DNA strands were sequenced. Sequence data analysis The ITS and 5–8S sequences were aligned using ClustalW (Thompson et al., 1994) and the alignment results were adjusted manually where necessary. Each sequence of the 19 morphospecies was used as query sequence to search for similar sequences in GenBank and EMBL using FASTA in the GCG program (Pearson & Lipman, 1988). The most similar reference sequences with query sequences were obtained and used for subsequent phylogenetic analysis along with selected taxonomic reference sequences. Maximum parsimony analysis was per- formed using PAUP version 3–1.1 (Swofford, 1993). To understand the general placement of the 19 morphospecies in the Kingdom Fungi, represen- tative reference sequences of Ascomycota, Basidio- mycota and Zygomycota obtained from GenBa