521 Mycoscience 41: 521-526, 2000 Cylindrocladium angustatum sp. nov., a new leaf spot pathogen of Tillandsia capitata from Florida, U . S . A . Pedro W . Crous 1)*, Conrad L. Schoch 11, Nabih EI-Gholl 2), T i m S. Schubert 2} and Robert M . Leahy 2) 1) Department of Plant Pathology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa 21Florida Department of Agriculture and Consumer Services, Division of Plant industry, P.O. Box 147100, Gainesville, FL 3 2 6 1 4 - 7 1 0 0 , U.S.A. Accepted for publication 9 June 2000 Cylindrocladium angustaturn is described as a new species from Tillandsia introduced with plant material into the U.S.A. from Central America. Koch's postulates are established to prove its pathogenicity to this host. The species is compared with and distinguished from C. heptaseptatum and C. rumohrae based on morphology, cultural characteristics and phylogenetic analysis of sequence data of the beta-tubulin gene. Key Words Cylindrocladium; foliar fungal pathogen; Hypocreales; Tillandsia. During March 1999, a new Cylindrocladium disease was encountered on Tillandsia capitata Griseb. plants at a large commercial containerized plant production greenhouse in Sarasota, FL in the U.S.A. The plants, which were showing symptoms of crown, basal rot and leaf spots, were collected by nursery inspectors of the Florida Department of Agriculture and Consumer Services--Division of Plant Industry, who submitted them to the Gainesville plant disease clinic for diagnosis (log number P99-0454). The Cylindrocladium isolate appeared unique from the first time it was observed in the clinic. The Tillandsia plants at this nursery were imported exclusively from Central America. The disease was detected a second time (log number P99-1321) in a subsequent visit to the same nursery on 11 June 1999. As part of a collaborative research programme on Cylindrocladium diseases, isolates were subsequently sent to the Department of Plant Pathology at the University of Stellenbosch, South Africa, for further characterisation. The aim of the present study was thus to report this new disease of Tillandsia from the U.S.A., to identify and characterise the Cylindrocladium species involved, prove Koch's postulates, and discuss its morphological similarities to other species in the genus. up to 1 L with deionized water. After cooling to 50~ 1.4ml of 5 0 ~ lactic acid was added per litre of autoclaved medium to obtain a pH of 4.3. Single conidi- Materials and Methods Morphology Leaf spots and blotches were surfacesterilised with l~ NaOCI for 3 m i n and plated on acidified potato dextrose agar (APDA, pH 4.3) to inhibit bacterial growth. APDA was prepared from the broth of 200 g of freshly peeled, diced, and boiled Irish potatoes (Solanum tuberosum L.) supplemented with 2 0 g dextrose, 1 g KH2PO4, and 18 g Difco Bacto agar and made * Corresponding author J Fig. 1. Cylindrocladiurn angustatum. Clavate vesicles, macroconidia and conidiophores formed on carnation leaf agar. Subverticillate conidiophores were rarely observed. The variation in vesicle shape is given in the lower right corner, with the solid line indicating the dominant shape. Bars= 10 f~m. 522 P.W. Crous et al. um isolates were obtained, and subsequently transferred to divided plates containing 2 ~ malt extract agar (MEA) (Biolab, Midrand, Johannesburg, South Africa), and carnation leaf agar (CLA) (Fisher et al., 1982; Crous et al., 1992). Plates were incubated at 25~ under continuous near-ultraviolet light. For microscopic examination fungal structures were mounted in lactophenol and measurements made at 1 0 0 0 x magnification. Thirty observations were made of each structure, the 95~00 confidence intervals determined, and the extremes given in parentheses. Reference cultures are maintained in the culture collection of the Department of Plant Pathology at the University of Stellenbosch (STE-U) and the American Type Culture Collection, 10801 University Blvd., Manassas, Virginia 2 0 1 1 0 - 2 2 0 9 , U.S.A. Sequencing Single conidial isolates selected for DNA comparison were grown on MEA plates. Mycelial mats were removed from the plates by means of a sterile scalpel and ground to a powder using liquid nitrogen and a mortar and pestle. Approximately 40rag of ground mycelium was added to 2 m l microtubes containing 600 f~l of extraction buffer. The extraction buffer consisted of l~ SDS, 5 0 m M Tris-HCI (pH8.0), 1 5 0 m M NaCI and 100 mM EDTA. Subsequently, the protocol was followed as suggested for the Wizard Genomic DNA Purification kit (Promega, Madison, U.S.A.). A 600 bp fragment of the ~-tubulin gene was amplified with primers T1 (O'Donnell and Cigelnik, 1997) and Bt2b (Glass and Donaldson, 1995). Additional amplification and sequencing conditions were the same as described by Schoch et al. (1999). Representative sequences used for comparison (C. quinqueseptatum Boedijn & Reitsma, C. multiseptatum Crous & M . J . Wing., C. rumohrae EI-Gholl & Alfenas, C. heptaseptatum Sobers, Alfieri & Knauss and C. theae (Petch) Subram.) were taken from Crous et al. (1999). Phylogenetic analysis Phylogenetic analysis of aligned DNA sequences was performed using PAUP* Version 4 . 0 b l (Swofford, 1999) and printed with the help of Treeview Version 1.5 (Page, 1996). Gaps were treated as missing. A single most parsimonious tree was obtained through branch and bound analysis in PAUP ~ and clade stability was assessed after 1000 bootstrap repetitions. The sequence of Fusarium subglutinans (Wollenw. & Reinking) Nelson et al. (NRRL 22016) deposited by O'Donnell et al. (1998), was obtained (GenBank accession number/%tubulin: U34417), and used as outgroup. New sequence data of the Cylindrocladium sp. were also deposited at GenBank (AF 207543, 207544). Pathogenicity To confirm Koch's postulates, 30 ml of a conidial suspension (123,500 conidia/ml) of the Cylindrocladium sp. was used to spray inoculate three cultivars of Tillandsia capitata ('Guatemala'-10 cm tall; 'Hondurensis'-5 cm tall; and 'Red Mexico'-8 cm tall). Two plants per cultivar were inoculated. Two plants were used for control, and sprayed with sterile water. Plants were individually maintained in a moist chamber at room temperature (25~ _+2~ Plants were removed from the moist chambers after 3d. Lesions were removed from infected leaves by means of a sterile scalpel blade. Leaf disks were surface sterilised by immersion in 70~ ethanol for 30 s, 1~ NaOCI for 1 min and again in 70~ ethanol for 30 s. Leaf disks were placed in Petri dishes containing 1.5~ water agar and incubated for seven days at 25~ Cultures were isolated and identified as discussed above. Results Morphology Isolates sporulated well on CLA, and produced penicillate macroconidiophores. Stipe extensions terminating in narrowly clavate vesicles were infrequently observed. Macroconidia were predominantly 7 10-septate, and ( 9 0 - ) 1 0 0 - 1 2 0 ( - 1 3 0 ) x 9 - 1 1 ( - 1 2 ) / ~ m (Figs. 1-6). No other conidial forms could be induced in culture. Morphologically this species was most similar to C. heptaseptaturn and C. rumohrae (Figs. 7-10). It's cultural morphology was quite distinct, however, in that isolates rarely produced chlamydospores, and relatively few microsclerotia formed scattered throughout the medium. Sequencing DNA sequences amplified from the 5' end of ~-tubulin contained several introns and consisted of 543 characters with 127 being informative. The single most parsimonious tree obtained from this data set consisted of 311 steps (Fig. 11). The t w o isolates of the Cylindrocladium species from Tillandsia clustered together, and were most similar to the clade represented by isolates of C. rumohrae (56~ bootstrap support). Pathogenicity Symptoms appeared within 3 d on all three cultivars, and varied from leaf spots to foliar blight. No crown rot symptoms developed on inoculated plants. On T. capitata 'Guatemala', very few leaf spots formed. Spots had tan centers with reddish-brown borders. On the cv. 'Red Mexico', leaf spots appeared with tan centers surrounded by reddish-brown water-soaked borders (Fig. 5). Symptoms on cv. 'Hondurensis' were light reddish-brown leaf spots with tan centers, advancing to foliar blighting. Within 6 d of inoculation, the plants were dead and sporulation of the pathogen was evident on the necrotic tissues. The pronounced symptoms on 'Hondurensis' implied extreme susceptibility of this cultivar. The Cylindrocladium sp. was recovered from symptomatic leaf tissue of all inoculated hosts. Taxonomy Cylindrocladium angustatum Crous & EI-Gholt, sp. nov. Figs. 1-4, 6 Filum septatum, hyalinum, (200-)250-300/~m, in vesiculam clavatam 2 - 3 p r o diam terminans. Rami primarii non vel 1-septati, 3 3 - 7 0 x 4 - 6 fire. Rami secundarii aseptati, 2 0 - 3 0 x 4-5/~m. Rami tertiarii vel quarti raro visi, aseptati, 1 5 - 2 0 x 4-5 ~m. Phialides ad extremum ramorum 1-4 exorientes, cylindricae, rectae vel leviter curvatae, hyalinae, aseptatae, 2 0 - 4 0 x 4 - 5 / ~ m . Conidia cylindrica, hyalina, (1-)7-10(-12)-septata, apice utrinque obtusa, interdum basi truncata, recta, saepe in medio leviter inflata, (90-) 1 0 0 - 1 2 0 ( - 1 3 0 ) x 9-11 (-12) Cylindrocladium disease of Tillandsia 523 Figs. 2-6. Cylindrocladium angustatum. 2. Conidiophore with cylindrical phialides. 3, 4, 6. Multiseptate conidia that taper slightly at their ends. 5. Leaf spot symptomson Tillandsiacapitatacv. 'Red Mexico',confirming Koch's postulates. B a r = l O ~ m . 524 P.W. Crous et al. pm. HOLOTYPE: U.S.A. FLORIDA: Sarasota nursery, leaf spots of Tillandsia capitata, 1 Mar. 1999, R. M. Leahy, PREM 56546, culture ex-type P99-0454, STE-U 2347. Etymology: Refers to the very thin stipe extensions. Macroconidiophores comprised of a stipe, a sterile elongation and a penicillate arrangement of fertile branches. Stipe extension septate, (200-)250-300 pm long, terminating in a narrowly clavate vesicle, 2-3 f~m diam; primary branches aseptate or 1-septate, 3 3 - 7 0 • 6 ftm; secondary branches aseptate, 20030• 4-5/~m, tertiary and quaternary branches rarely observed, aseptate, 15-20 x 4-5/~m, each terminal branch producing 14 phialides; phialides cylindrical, straight to slightly curved, hyaline, aseptate, 2 0 - 4 0 x 4-5/~m, apex with minute periclinal thickening and inconspicuous collarette. Conidia cylindrical, bluntly rounded at both ends, or truncate at the base, straight, frequently slightly swollen in the middle, (90-)100-120(-130) • 9-11 (-12) ~m, (1-)710(-12)-septate, lacking a visible abscission scar, held in cylindrical clusters by colorless slime. Megaconidiophores and Microconidiophores not observed. Chalamydospores rarely observed, dark brown, thickened, sparse, occurring in aggregated clusters to form Figs. 7-10. microsclerotia. Cultures: Colony color (underneath and surface) buff 15'd, with moderate white aerial mycelia. Colony margin regular, with slight to no sporulation on aerial mycelia. Colonies reaching 33 mm diam on MEA after 6 d in the dark at 25~ Cardinal temperatures for growth: Min above 10~ max below 35~ opt 25~ This is a moderate temperature species, not growing below 10~ or above 30~ Substrate: Tillandsia capitata Griseb. Distribution: Florida, U.S.A. Additional culture examined: U.S.A. FLORIDA, Sarasota nursery, leaf spots of Tillandsia capitata. 11 Jun. 1999, R. M. Leahy, culture P99-1321, STE-U 3152. Notes. Cylindrocladium heptaseptatum has (1-)7(8)-septate macroconidia, (80 o) 110-130(-144) x (6-)78(-9)/~m in size, and clavate vesicles that are (3-)4-5/Lm wide. Macroconidia of C. rumohrae are (1-)5{-7)-seprate, (70-) 1000120(-130) x (8-) 10(-12)/~m, and its vesicles are clavate, (3-)4-5/~m wide (Figs. 7-10). Both species also form extensive chlamydospores in culture. Macroconidia of C. angustatum are (1-)7-10(-12)septate, (90-)100-120(-130) x 9-11(-12) /~m, thus Conidiophores and macroconidia of Cylindrocladium spp. 7-9. C. heptaseptaturn. 10. C. rumohrae. Bar= 10/~m. Cylindrocladium disease of Tillandsia // 525 Fusarium subglutinansNRRL 22016 i Cylindrocladium quinqueseptatum A T C C 16550 68 ! 99 F Cylindrocladiumquinqueseptatum S T E - U L Cylindrocladium quinqueseptatum S T E - U ?9 759 516 f Cylindrocladiummultiseptatum S T E - U 100 Cylindrocladium multiseptatum S T E - U 1589 1602 Cylindrocladium rumohrae S T E - U 1603 100 98 r Cylindrocladiumrumohrae UFV 218 I Cylindrocladium rumohrae UFV 215 56 100 I Cylindrocladiumangustatum P99-0454 96 Cylindrocladium angustatum P99-1321 Cylindrocladium heptaseptatum F T C C 1003 99 98 Cylindrocladium heptaseptatumF T C C 1002 Cylindrocladium heptaseptatumP99-031 100 F Cylindrocladiumtheae A T C C 48895 / 1 ICylindrocladium theae UFV 16 Fig. 11. Most parsimonious tree (307 steps, C1=0.827, R1=0.859, RC=0.711) generated by the branch and bound algorithm in PAUP* version 4.0 bl based on sequences of the 5' end of the/3-tubulin gene. Clade stability was assessed with 1000 bootstrap replications (values above 50~ are shown) and F. subglutinans was used as outgroup. being w i d e r and d e v e l o p i n g more conidial septa than C. heptaseptatum. A l t h o u g h m a c r o c o n i d i a of C. angustaturn are similar in size to those of C. rurnohrae, t h e y tend t o develop more septa. A further difference b e t w e e n these species can also be found in their terminal vesicles. A l t h o u g h those of C. heptaseptaturn and C. rumohrae are clavate, ( 3 - ) 4 - 5 / L m w i d e , those of C. angustaturn are n a r r o w l y clavate, 2 - 3 / L m w i d e , and rarely develop in culture. A final distinguishing feature in this species c o m p l e x can be f o u n d in their g r o w t h on MEA. Both C. heptaseptaturn and C. rumohrae are observed to f o r m extensive c h l a m y d o s p o r e s in this m e d i u m , w h i l e isolates of C. angustaturn rarely f o r m c h l a m y d o s p o r e s in culture, but rather tend t o f o r m a f e w scattered microsclerotia. These m o r p h o l o g i c a l differences are also supported by their ~-tubulin sequence data (Fig. 1 1), w h i c h s h o w e d these three m o r p h o l o g i c a l species to also by p h y l o g e n e t i c a l l y distinct, w i t h C. angustaturn being more closely related to C. rumohrae than to C. heptaseptaturn. No Calonectria t e l e o m o r p h s are k n o w n for C. angustaturn and C. heptaseptatum, and all mating studies have thus far proven unsuccessful, A c k n o w l e d g e m e n t - - P W C acknowledges financial support from the South African National Research Foundation. Literature cited Crous, P.W., Phillips, A . J . L . and Wingfield, M.J. 1992. Effects of cultural conditions on vesicle and conidium morphology in species of Cylindrocladium and Cylindrocladiella. Mycologia 84: 497-504. Crous, P.W., Kang, J.-C., Schoch, C. L. and Mchau, G. R. A. 1999. Phylogenetic relationships of Cylindrocladiumpseudogracile and Cylindrocladium rumohrae with morphologically similar taxa, based on morphology and DNA sequences of internal transcribed spacers and/~-tubulin. Can. J. Bot. 77: 1813-1820. Fisher, N. L., Burgess, L. W., Toussoun, T. A. and Nelson, P. E. 1982. Carnation leaves as a substrate and for preserving cultures of Fusarium species. Phytopathology 72: 151153. Glass, N. L. and Donaldson, G. 1995. Development of primer sets designed for use with PCR to amplify conserved genes from filamentous ascomycetes. Appl. Environ. Microbiol. 61: 1323-1330. 526 P.W. Crous et al. O'Donnell, K. and Cigelnik, E. 1997. Two divergent intragenomic rDNA ITS2 types within a monophyletic lineage of the fungus Fusarium are nonorthologous. Mol. Phylogenet. Evol. 7 : 1 0 3 - 1 1 6 . O'Donnell, K., Cigelnik, E. and Nirenberg, H. 1998. Molecular systematics and phylogeography of the Gibberella fujikuroi species complex. Mycologia 90: 465-493. Page, R. D.M. 1996. TREEVlEW: An application to display phylogenetic trees on personal computers. Comp. Appl. Biosci. 12: 357-358. Schoch, C.L., Crous, P.W., Wingfield, B.D. and Wingfield, M.J. 1999. The Cylindrocladium candelabrum species complex includes four distinct mating populations. Mycologia 91 : 286-298. Swofford, D. L. 1999. PAUP~: Phylogenetic analysis using parsimony (~ and other methods), ver. 4. Sinauer Associates, Sunderland, Massachussetts, U.S.A.