Mycol. Res. 102 (7) : 850–854 (1998) 850 Printed in the United Kingdom Claviceps citrina sp. nov., a parasite of the halophytic grass Distichlis spicata from Mexico S Y L V I E P A ZC O U T O V A; 1, L E O P O L D F U CC I; K O V S K Y; 2, S. G. L E Y V A-M I R3 A N D M I R O S L A V F L I E G E R1 " Institute of Microbiology CAS, VıU den] skaU 1083, 142 20 Prague 4, Czech Republic # Instituto de Fitosanidad, Collegio de Postgraduados, Montecillo, Texcoco, 56230 MeU xico $ Departamento de ParasitologıU a AgrıU cola, Universidad AutoU noma Chapingo, Chapingo, 56230 MeU xico The occurrence of an undescribed ergot species, Claviceps citrina, on the halophytic chloridoid grass Distichlis spicata, in the Texcoco region of central Mexico, is reported. RAPD and ITS1 sequences of this fungus were compared with other Claviceps species and the morphological uniqueness of the fungus was reinforced. Its sclerotia do not contain any alkaloids of the ergoline type. Germination of sclerotia occurred immediately after placing them on humid sand. After removal of the capitula, the remaining stipe was able to regenerate the capitulum. Forty-one species of Claviceps are currently known. Of these, 29 colonize panicoid hosts, 3 occur on chloridoid grasses, 2 on bambusoids, 3 on pooids, 1 on arundinoids and 3 are parasites of rushes and sedges. Some species have a broad geographical distribution, either natural (C. pusilla) or influenced by the human factor (C. paspali, partly C. purpurea and, very recently, C. africana) (Langdon, 1952 ; Frederickson, Mantle & De Milliano, 1991), but species with a monogeneric host spectrum and limited occurrence predominate. When searching for ergot infections, two marked stages can be found in florets. The sphacelial stage is evident by the excretion of honeydew containing conidia and by the formation of whitish mycelial mass replacing the ovaries. Later, dark sclerotia are formed that ensure the survival of the fungus during the period of dry or cold months. Sclerotial germination is triggered by the arrival of a rainy period and}or by the elevated temperatures in the spring. In wet tropical and subtropical regions, however, often the only stage by which ergot can be characterized is the sphacelial} conidial fructification because sclerotium formation is erratic. In 1990, widespread ergotism of ‘ pasto salado ’ (Distichlis spicata (L.) Greene) was observed in the Lago de Texcoco locality, Central Mexico (Matheis-Jimenes & Leyva-Mir, 1993). The presence of elongated sclerotia and the morphological variability of C. purpurea noted in older descriptions (Sprague, 1950), where occurrence on D. spicata was also mentioned, led to the preliminary identification of the Distichlis ergot as this species. The creamy colour of stromata and the small size of elliptical conidia, however, although within the limits given by Sprague (1950), left some doubts. Fresh collections were, therefore, examined on the basis of both morphology and DNA analysis. MATERIALS AND METHODS Strains C. purpurea (Fr.) Tul. Pepty 695}S and C. fusiformis Loveless SD 58 (ATCC 26019) were generously supplied by Professor D. Gro$ ger (Halle}Saale, Germany), C. viridis Padwick & Azmatullah (125.63) and C. grohii J. W. Groves (127.47) were obtained from the CBS, Baarn and Delft, The Netherlands ; C. fusiformis 129 was derived by mutation from the strain 47A. Strains of C. purpurea, C. paspali F. Stevens & J. G. Hall, C. gigantea S. F. Fuentes, Isla, Ullstrup & A. E. Rodr., C. phalaridis J. Walker and C. africana Frede., Mantle & De Milliano were isolated in our laboratory from sclerotia. Isolation of mycelial cultures Sclerotia were surface-sterilized for 2–3 min in 1±3 % sodium hypochlorite (suitably diluted Clorox), then washed in distilled water and placed on T2 agar plates (g l−" sucrose 100 ; asparagine 10 ; yeast extract 0±1 ; KH PO 0±25 ; MgSO . 7H O # % % # 0±25 ; FeSO . 7H O 0±02 ; ZnSO . 7H O 0±015 ; KCl 0±12 ; % # % # Ca(NO ) . 4H O 1 ; agar 20 ; pH 5±2) (Spalla, 1973) supple$# # mented with 100 µg ml−" of ampicillin or 20 µg ml−" of tetracycline. Germination of sclerotia Sclerotia were laid on moist sand, in aluminium-foil covered 600 ml beakers, which were placed on a laboratory shelf and watered weekly with distilled H O. The ambient temperature # was 20–22 °C. Sylvie Paz) outova! and others 851 Microscopy RESULTS Vertical sections of mature capitula (30 µm thick), obtained using a cryomicrotome, were stained with aniline blue. Conidial mounts were prepared using spores washed from sclerotia as the laboratory cultures did not sporulate. Ascospores were collected from germinated sclerotia glued to the bottom of an inverted Petri dish and then stained with aniline blue. Morphology Alkaloid analysis Powdered sclerotia were defatted by extraction with hexane. The powder was further extracted with the mixture CH Cl –MeOH (1 : 1) with 1 % of ammonia, the extract # # volume reduced and analysed by hplc (Flieger et al., 1993). DNA isolation Washed mycelium from 10-day-old shaken cultures of C. citrina in T2 medium (250 rpm, 24°), or from 2-wk cellophane cultures of other species on T2 agar plates, was powdered in liquid nitrogen. To 1 g of mycelium, 3 ml of extraction buffer (Triton X-100 2 %, SDS 1 %, NaCl 0±25 , Tris-HCl pH 7±8 0±1 , EDTA pH 8±2 0±1 ) and of phenol–chloroform were added. The suspension was gently agitated for 15 min, centrifuged and the supernatant precipitated 2 h with 0±7 vol. of isopropanol at 0°. Pelleted DNA was dissolved in 3 ml 1¬TE pH 8±2, treated with 50 µg of RNAse, extracted 2–3¬ with chloroform, precipitated with 0±7 vol. isopropanol}0±3  sodium acetate and dissolved in 1¬TE pH 7±2. RAPD The reaction mixture (20 µl) contained : DNA 100 ng, dNTP 2 m, MgCl 2±5 m, primer 206 (TCAACAATGTCGGC# CTCCGT) 20 pmol, DynaZyme-buffer, DynaZyme 1 U (Finnzyme, Oy). The mixture was overlaid with mineral oil and put in the thermocycler GeneE (Techne) when the plate reached over 80°. The cycling was as follows : 94° 3 min, 32° 1 min, 72° 2 min, 1¬ ; 94° 1 min, 32° 1 min, 72° 2 min, 43¬ ; 94° 1 min, 32° 1 min, 72° 6 min, 1¬. The amplified bands were separated on 2 % agarose gel in 1¬TBE. Claviceps citrina sclerotia were collected in September 1996 on D. spicata in a sandy, saline area once covered by Lake Texcoco. The type locality is the seed production station of the Comissio! n del Lago de Texcoco, 6 km from Texcoco, near the highway to Mexico City. C. citrina colonies on T2 agar were compact, nonsporulating, creamy to beige with brown agar pigmentation (Fig. 1). Sclerotia (Fig. 2) were clavate, straight or slightly curved, capped with the remnants of sphacelial tissue, with reticulate surface, brown–grey, lighter on base and apex, 4–12 (xa ¯ 6±8) mm long¬1–2 mm wide at the base (measured from 42 sclerotia). The sclerotia, stored at laboratory temperature, were laid on moist sand at the beginning of December 1996. About 50 % of sclerotia germinated within 10 d of incubation and all germinated within 30 d. No special pretreatment was required. On some sclerotial tips, Cladosporium sp. and}or Fusarium sp. contamination developed, but this did not inhibit germination. First, the capitulum appeared, white and covered with hyphae, later it became smooth and light lemon yellow (Fig. 3). The stipe was slim (max. 0±7 mm at the base), whitish with a yellow tinge, later pale yellow. The mature capitulum was of intense lemon colour which prompted the species name. Ostioles appeared as dark yellow dots, that later enlarged and gave the capitulum a characteristically papillate appearance (Fig. 4). Old ‘ wilting ’ stromata became dark creamy to ochrecoloured. The stroma development from capitulum emergence to septate ascospore discharge took about 25 d. Ascospores (Fig. 5) readily germinated in water. Conidia of C. citrina were rather small (Fig. 5), ovoid to elliptical, and found mostly on the proximal part of the sclerotium. From one sclerotium, the formation of up to six stromata from different sites was observed. The length of the stipe depended on the air humidity – in uncovered beakers only short stipes about 7 mm long were formed, whereas in covered beakers their usual length was 15 mm. The diameter of the capitula was 1–1±2 mm. When the capitulum was cut off, the stipe developed a tuft of mycelium from which a new stipe and a capitulum regenerated. Description rDNA region sequencing and analysis The region containing ITS1 was amplified using ITS1 and ITS2 primers (White et al., 1990). The mixture (20 µl) contained 50 ng of genomic DNA, 10 pmol of each primer, 0±2 m dNTP and 0±5–1 U of DynaZyme with the respective buffer (Finnzyme, Oy). The cycling conditions were 95° 5 min, 55° 1 min and 72° 1 min, for 35 cycles with hot start. Amplified fragments were purified by Wizard DNA Clean-Up System (Promega, Madison, WI) and subjected to automatic AmpliTaq polymerase cycle sequencing process on ITS1 and ITS2 primers with dye terminator labelling on an ABI 373A sequencer (Perkin Elmer). Claviceps citrina Paz) outova! , Fuc) ı! kovsky! , Leyva-Mir & Flieger, sp. nov. (Figs 1–6) Sclerotia oblonga vel obclavata, recta vel subcurvata, ultra floris partes protrudentia, brunneo-grisea, superficie reticulata, usque ad 12 mm longa (mediet. 6±8 mm), 1–2 mm lata, cum medulla plectenchymatis albi, apice sphacelioidibus pileata. Stromata 1–6, plerumque 4, circa 25 dies post germinationem matura. Stipites plerumque 15 mm longi, primum albi, deinde subflavi, apex stipitis in base capituli ad 40 % immersus. Capitula globosa, 1–1±2 mm diam., citrina, in maturitate papillosa. Perithecia ovato-pyriformia, 238–350 µm longa (mediet. 287 µm), 92–146 µm (mediet. 107 µm) lata. Ascosporae filiformes, septatae, 78–136 µm (mediet. 112 µm) longae, diam. New species of Claviceps 852 1 3 2 Figs 1–3. Claviceps citrina. Fig. 1. Colony on T2 agar (bar, 3 cm) (Photo S. Paz) outova! ). Fig. 2. Sclerotia (bar, 1 mm) (Photo P. Br) icha! c) ek). Fig. 3. Germinated sclerotium with mature stromata (bar, 1 mm) (Photo P. Br) icha! c) ek). 4 5 6 Figs 4–6. Claviceps citrina. Fig. 4. Vertical section of capitulum showing perithecia (bar, 100 µm). Fig. 5. Germinating ascospore. Arrows indicate septa (bar, 10 µm). Fig. 6. Conidia (bar, 10 µm) (Photos Dr Z. ZC iz) ka). Sylvie Paz) outova! and others 853 Pepty 695/S Australia GAM12885 109 C. citrina Pepty 695/S Australia GAM12885 109 C. citrina ia at ro a iC m al ba sp la pa li A C. spa taly co pa li I xi C. spa Me pa tea C. gan ndia gi s I da ia C. ridi ana tral vi i C us C. ohi is A gr rid ico i C. ala ex our rica ph a M iss Af C. trin M 58 a ci rea D ric C. rpu is S Af pu rm 29 C. sifo is 1 fu rm 27 C. sifo is F alia fu rm str C. sifo Au fu na C. ica r af C. Fig. 7. RAPD patterns of nine Claviceps species from different localities. Size marker is λDNA cut with Bgl I. The RAPD was done with the primer 206. 0±45–0±6 µm. Conidia hyalina, ovoidea vel elliptica, 3±65–7±2 µm (mediet. 5±5 µm) longa, 2±5–2±7 µm lata, aseptata. Hab. in flosculis Distichlidis spicatae (L.) Greene, Mexico, Edo de Mexico, Texcoco. Holotypus : PRM 842 966 ex Distichlis spicata, Texcoco, Mexico, 13 Sep. 1996, S. Paz) outova! . Sclerotia oblong to clavate, straight or slightly curved, protruding beyond floral parts, brown-grey, with reticulate pattern on the surface ; up to 12 mm long (mean 6±8 mm), 1–2 mm wide, with white plectenchymatic medulla. The tip is capped with sphacelial tissue. Stromata 1–6, mostly 4, arising from various points on the sclerotial surface, maturing in 25 d. Stipes usually 15 mm long, initially whitish then becoming light yellow, immersed to 40 % of the capitulum diameter. Capitula globose, 1–1±2 mm diam., lemon yellow, papillate when mature. Perithecia ovate-pyriform, 238–350 µm long (mean 287 µm), 92–146 µm (mean 107 µm) wide. Ascospores filiform, septate, 78–136 µm (mean 112 µm) long, diam. 0±45–0±6 µm. Conidia hyaline, non-septated, oval to elliptical, 3±65–7±2 µm (mean 5±5 µm) long, 2±5–2±7 µm wide. Holotype : PRM 842 966 ex Distichlis spicata, Texcoco, Mexico, 13 Sep. 1996, leg. S. Paz) outova! . Deposited in Herbarium of the Mycological Department, National Museum, Prague, Czech Republic. Secondary metabolites The sclerotia were analysed for alkaloid content by hplc. Among the detected substances, no traces of clavines or lysergic acid or its peptide derivatives were found, only a series of unknown metabolites whose uv spectra differed from those of ergoline-derived compounds. DNA analyses DNA of Distichlis ergot was subjected to the RAPD fingerprinting with the primer 206. This primer gave species specific patterns identical for the isolates from different localities ; in our laboratory it is routinely used for preliminary species verification of new isolates. The banding pattern of C. citrina was compared with patterns of isolates from different localities belonging to the species C. paspali, C. phalaridis, C. gigantea, C. viridis, C. africana, C. grohii, C. fusiformis and C. purpurea (Fig. 7). The banding pattern of C. citrina clearly Pepty 695/S Australia GAM12885 109 C. citrina Fig. 8. Differences in ITS1 spacer sequence among C. citrina and C. purpurea isolates from different regions. Dot, base identical with Pepty 695}S ; dash, indel. C. purpurea isolates – Pepty 695}S, European ; Australia, Australian, from Phalaris sp. ; GAM 12885, D. glomerata, Georgia, U.S.A. ; 109, F. arundinacea, Kentucky, U.S.A. differed from all other species and exhibited none of the C. purpurea-specific bands. Its strongest band, about 180 bp, was absent in all other species. The ITS1 spacers (Fig. 8) of C. citrina, C. purpurea Pepty 695}S (Schumann et al., 1982) and our isolate from Phalaris sp. (Australia) were sequenced, and the sequences deposited in EMBL Nucleotide Sequence Database (C. citrina AJ 000068 ; C. purpurea 695}S AJ 000069 ; C. purpurea AU AJ 000070). The published sequences of American strains C. purpurea GAM 12885 (D. glomerata, Georgia, U.S.A.) (Glenn & Bacon, 1996 ; GenBank accession No. U57669) and C. purpurea 109 (F. arundinacea, Kentucky, U.S.A.) (Schardl et al., 1991) were used for further comparison. The sequences of the Pepty 695}S and Australian isolates were identical. The American isolates GAM 12885 and 109 differed from them in 5 and 9 positions respectively, whereas C. citrina differed in 29 positions. Moreover, it had a deletion 32 bp long. The absence of C. purpurea specific bands in the RAPD profiles of the Distichlis ergot, together with the marked difference in the ITS1 spacer sequence, excluded the possibility of its classification as C. purpurea. DISCUSSION The important criteria in the classification of ergot species are the identity of the host plant, the locality where the infection occurred and the morphological characters of the parasite (Loveless, 1964). Distichlis spicata is a halophytic chloridoid grass species. Distichlis contains five species that are found in America and one in Australia. It can be expected, therefore, that the Mexican Distichlis ergot would be an American indigenous species. Most ergot species are monogeneric or parasitize related genera (Langdon, 1952 ; Loveless, 1964). Species that colonize two or more genera, and possess yellow stromata, are found mainly on the panicoid and andropogonoid grasses of Africa, Asia and Australia (e.g. C. hirtella Langdon, C. pusilla Ces., C. sulcata Langdon and C. microspora Tanda) (Langdon, 1952 ; Loveless, 1964 ; Tanda, 1991). None of these ergots is reported to occur on chloridoid grasses. The only other chloridoid parasites known to date are C. cynodontis Langdon from Cynodon, C. yanagawaensis Togashi from Zoysia and C. cinerea Griffiths from Hilaria mutica. New species of Claviceps Claviceps yanagawaensis was found in Japan and occasionally is transferred to U.S.A. with seeds. Its dark purple stromata and blackish violet sclerotia are very different from C. citrina. Claviceps cynodontis is found exclusively on Cynodon in southern Europe, Africa, India and Burma. Its occurrence on other chloridoid genera was not observed. Claviceps cynodontis, although it may be introduced to America with the seeds of Bermuda grass, differs from C. citrina in the smaller sclerotia, rounded perithecia not exceeding 170 µm and typical reniform conidia more than twice as long as the C. citrina oval ones. Moreover, the sclerotia need a 6-month resting period prior to germination (Loveless, 1965). Some relatedness could, however, be observed between C. citrina and another American ergot, C. cinerea. The appearance and coloration of C. citrina sclerotia (although about half the size), and their quick germination induced by humidity, resemble the properties of sclerotia of C. cinerea. The capitulum of C. cinerea is grey, the stipe is white and the perithecia are sunken, not protruding as in mature C. citrina capitula (Sprague, 1950). Claviceps cinerea was observed in the northern Mexican semi-arid States of Chihuahua, Coahuila, Durango and in Arizona, U.S.A. (Zenteno-Zevada, 1958 ; Sprague, 1950). In Central Mexico, as well as in the above localities, there is no rain during winter and both Claviceps species apparently developed rapid sclerotial germination as an adaptive mechanism. The only ergot described on D. spicata to date is C. purpurea (Sprague, 1950), but the intense lemon-yellow colour of C. citrina capitula, the immediate germination of sclerotia without any requirement for the dormancy period, and the small conidia and complete absence of any traces of ergot alkaloids in the sclerotia distinguish it from C. purpurea. Claviceps citrina sclerotia do not have any trace of the purple black colour that is typical of C. purpurea. The average length of perithecia and ascospores exceeds the highest values for C. purpurea. The sequence difference between ribosomal DNA of C. citrina and various C. purpurea isolates is final evidence that Distichlis ergot is not C. purpurea. The RAPD patterns of C. citrina, with the species-specific primer 206 designed in our laboratory, are markedly different from eight other Claviceps species. This short survey of ergot species that are known to occur on D. spicata or other chloridoids confirms that the Mexican Distichlis ergot merits the status of new species. (Accepted 16 September 1997) 854 Thanks are due to Professor Gareth Morgan-Jones (Auburn University, Alabama, U.S.A.) for the critical reading of the manuscript and valuable suggestions and to Professor Douglas G. Parbery (University of Melbourne, Victoria, Australia) for helpful discussions. REFERENCES Flieger, M., Sedmera, P., Havlı! c) ek, V., Cvak, L. & Stuchlı! k, J. (1993). 10-OHcis and 10-OH-trans paspalic acid amide – new alkaloid from Claviceps paspali. Journal of Natural Products (Lloydia) 56, 810–814. Frederickson, D. E., Mantle, P. G. & De Milliano, W. A. J. (1991). Claviceps africana sp. nov. ; the distinctive ergot pathogen of sorghum in Africa. Mycological Research 95, 1101–1107. Langdon, R. F. N. (1952). Studies on ergot. Ph.D. thesis, University of Queensland, Australia. Loveless, A. R. (1964). Use of the honeydew state in the identification of ergot species. Transactions of the British Mycological Society 47, 205–213. Loveless, A. R. (1965). Studies on Rhodesian ergots 4. Claviceps cynodontis Langdon. Kirkia 5, 25–29. Matheis-Jimenes, E. & Leyva-Mir, S. G. (1993). C. purpurea (Fr.) Tul. causante del ergot o cornezuelo en pasto salado Distichlis spicata (L.) Greene en el Lago de Texcoco, Me! xico. Revista Mexicana de FitopatologıU a 11, 98–102. Schardl, C. L., Liu, J.-S., White, J. F., Finkel, R. A., An, Z. & Siegel, M. R. (1991). Molecular phylogenetic relationships of nonpathogenic grass mycosymbionts and clavicipitaceous plant pathogens. Plant Systematics and Evolution 178, 27–41. Schumann, B., Erge, D., Maier, W. & Gro$ ger, D. (1982). A new strain of C. purpurea accumulating tetracyclic clavine alkaloids. Planta Medica 45, 11–14. Spalla, C. (1973). Genetic problems of production of ergot alkaloids in saprophytic and parasitic conditions. In Genetics of Industrial Microorganisms (ed. Z. Vane) k, Z. Hos) t) a! lek & J. Cudlı! n), pp. 393–403. Elsevier : Amsterdam. Sprague, R. (1950). Diseases of Cereals and Grasses in North America. Ronald Press : New York. Tanda, S. (1991). Mycological studies on the ergot in Japan. 24. A new variety of C. microspora on Eccoilopus cotulifer. Transactions of the Mycological Society of Japan 33, 343–348. White, T. J., Bruns, T., Lee, S. & Taylor, J. W. (1990). Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In PCR Protocols : A Guide to Methods and Applications (ed. M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White), pp. 315–322. Academic Press : San Diego. Zenteno-Zevada, M. (1958). Estudios sobre hongos parasitos de Gramineas de la repu! blica Mexicana. II. Memorios del Primer Congreso Nacional de EntomologıU a y FitopatologıU a, pp. 501–519. Chapingo : Mexico.