- Email: [email protected]
Geographic distribution and diversity in Claviceps purpurea from salt marsh habitats and characterization of Pacific coast populations
Mycol. Res. 109 (4): 439–446 (April 2005). f The British Mycological Society
439
doi:10.1017/S0953756205002467 Printed in the United Kingdom.
Geographic distribution and diversity in Claviceps purpurea from salt marsh habitats and characterization of Pacific coast populations
Alison J. FISHER1*, Thomas R. GORDON2 and Joseph M. DITOMASO1 1
Department of Vegetable Crops, One Shields Avenue, University of California, Davis, CA 95616-8755, USA. Department of Plant Pathology, One Shields Avenue, University of California, Davis, CA 95616-8755, USA. E-mail : afi[email protected]
2
Received 11 June 2004; accepted 4 January 2004.
Claviceps purpurea specific to grasses in salt marsh habitats (Group G3) has previously been identified on Spartina spp. in two locations : New Jersey, USA and southern England. We have identified this subgroup of C. purpurea (G3) in 11 distinct populations including western Europe, South America, and along the Atlantic and Pacific Coasts of the USA. In addition, G3 C. purpurea was discovered on a new host grass genus, Distichlis. Unweighted pair group mean analyses of AFLP and RAPD data reveal distinct structure in G3 C. purpurea populations. Pacific coast populations show little diversity, suggesting they may have been introduced recently in that region. 43 isolates, representing 11 populations were identified as G3 based on the presence of an EcoRI restriction site in the 5.8S ribosomal DNA, and a clear genetic separation from isolates representing the other two C. purpurea subgroups (G1 and G2). In addition, all isolates originating from Spartina densiflora, S. foliosa, S. alterniflora, and S. anglica were identified as belonging to G3. RAPD and AFLP analyses supported the recognition of three discrete groups within C. purpurea and revealed high genetic variability between groups, with only 1.8 % of polymorphic markers shared across all isolates. Similarly, analysis of molecular variation (AMOVA) revealed that genetic variability was mainly due to variations between groups (63.5%) rather than within groups (28.5 %) or within populations (7.96%). G3 isolates were 35 % similar, Pacific coast isolates 83 % similar. Ninety percent similarity among isolates from the San Francisco Bay Area suggests this is a recently introduced population.
INTRODUCTION Claviceps purpurea, the ergot fungus, is a well-known pathogen of cereal grains and forage grasses. Although it is a commonly occurring parasite of grasses in unmanaged landscapes, research has emphasized its impacts on agronomic systems. The species has a global distribution, and a wide host range within the Poaceae. Species in the subfamily Pooideae are the most common hosts, followed by members of the Arundinoideae and Chloridoideae (Bowden 1965, Loveless 1971). This broad host range has led many researchers to seek evidence of intraspecific substructure based on morphology (Loveless 1971), alkaloid production (Eleuterius & Meyers 1977, Kobel & Sanglier 1978), and genetic characteristics (Jungehu¨lsing & Tudzynski 1997). Pazoutova´ et al. (2000) synthesized previous research on C. purpurea morphology, alkaloid chemistry and genetics, and identified three distinct groups within the species, describing them as ‘chemoraces. ’ Rather than * Corresponding author.
a delimitation based on host range, the intraspecific groups were defined by habitat specialization. The largest group, G1, was associated with land grasses, and G2 with grasses in ‘wet and shady ’ environments, whereas G3 was found only on grasses in salt marsh habitats. Group 3 ergot is synonymous with C. purpurea var. spartinae (Duncan et al. 2002) and is referred to herein as G3. Sclerotia of both G2 and G3 C. purpurea float in water while sclerotia from terrestrial C. purpurea (G1) sink, clarifying the significance of earlier reports that sclerotia of some isolates float, and the association of this trait with host habitat (Stager 1922). Lengths of conidia range from 5–8 mm for G1, 7–10 mm for G2, and 10–12 mm for G3. Isolates associated with G1 and G3 share an EcoRI restriction site in the 5.8S ribosomal DNA (rDNA), which G2 isolates lack (Pazoutova´ et al. 2000). The three groups can be differentiated by production of a unique set of alkaloids, and by RAPD analysis (Pazoutova´ et al. 2000). G3 C. purpurea has been isolated exclusively from cordgrass species of the genus Spartina, a member of
Claviceps purpurea in salt marsh habitats the subfamily Chloridoideae, suggesting a very narrow host range compared to G1 and G2 (Pazoutova´ et al. 2000). Two host species to G3 have been identified, each known only from one location: common cordgrass, S. anglica, in the UK and smooth cordgrass, S. alterniflora, from New Jersey on the Atlantic coast of the USA (Pazoutova´ et al. 2000, 2002a). An analysis using unweighted pair group method with arithmetic averaging (UPGMA) placed the US isolates in an outlier position, suggesting a North American origin for the European population. Claviceps purpurea was first found on a Spartina species in North America in 1895 (Eleuterius & Meyers 1974), and was not reported from Europe until the 1970s (Boyle 1976), which is also consistent with a North American origin. Whereas only two populations of G3 have been identified by means of molecular markers, the host to this intraspecific group, Spartina, is a widespread genus including both terrestrial and halophytic salt marsh species. Most members of the genus, including S. alterniflora, share a common native geographic range along the Atlantic coast of North and South America. Introduced populations of S. alterniflora have established on the Pacific coast as well, in the San Francisco Bay, California (Daehler & Strong 1994) and Willapa Bay, Washington (Stiller & Denton 1995). Thus, the potential geographic range of G3 may be greater than is presently recognized and it is reasonable to think that diversity in populations of the pathogen may reflect the diversity of the host species. The objectives of this study were to assess the geographic distribution and ecological host range of G3 C. purpurea, and to evaluate the genetic diversity revealed by this broader sampling of the salt marsh group. In addition, genetic diversity among Pacific coast isolates was evaluated to determine if G3 ergot is likely to have been introduced to that region.
MATERIALS AND METHODS Isolate collection and culturing Isolates included in this study, and their origins, are listed in Table 1. Collections from the same location were taken from inflorescences from host plants at least one meter apart. Intensive sampling was done in the San Francisco Bay area. To examine within population diversity, isolates were obtained from ten inflorescences collected from two S. foliosa marshes: Point Reyes National Seashore (NS) on the outer coast, and Mountain View in the West Bay. In addition, one infected S. foliosa inflorescence was collected at each of three additional marshes : Palo Alto, San Mateo in the West Bay and Bolinas Lagoon on the outer coast. To reduce superficial contaminants, sclerotia were treated by immersion in 70% ethanol for 30 s followed by 1.5 min in 1.3 % sodium hypochlorite and three
440 successive, three min rinses in sterile deionized water. The rind of the sclerotium was removed and fungal tissue plated onto potato dextrose agar (PDA). To ensure that each isolate represented a single genotype, cultures were streaked onto water agar and a hyphal tip was subcultured. Pure cultures were grown on filter paper overlaid on PDA ; fully colonized papers were removed, dried, and refrigerated at 4 xC. Isolates not collected in the field were obtained as pure cultures from S. Pazoutova´, including G1 isolates 374, 428, G2 isolates 236 and 434 (Pazoutova´ et al. 2000), G3 isolates 500 and 538 (Pazoutova´ et al. 2002b), and G1 isolates 165, 204 and 478, which have not been included in a previously published report (Table 1). DNA extraction Tissue for DNA extraction was obtained by growing isolates either on cellophane overlaid on PDA or in potato dextrose broth (PDB). Isolates grown on cellophane were harvested after nine days and stored at x20 x. Fungal biomass recovered from PDB cultures was harvested after seven days by filtration through cheesecloth, lyophilized and stored at x20 x. Mycelia (10–20 mg) in microcentrifuge tubes, were frozen by dipping tubes in liquid nitrogen, and ground with a pestle. Total DNA was extracted using the methods described by Daehler et al. (1999). After DNA was recovered in solution for the final time, 75 ml of supernatant were withdrawn and mixed with 200 ml Tris-EDTA buffer (10 mM Tris, 1 mM EDTA, pH 7.8), and DNA concentration was quantified by spectrophotometry. EcoRI analysis The polymerase chain reaction (PCR) was used to amplify the region containing the 5.8S rDNA, using ITS1 and ITS4 primers (White et al. 1990). Primers were desalted with the aid of an Edge Biosystems SR gel filtration cartridge. Each reaction had a final vol. of 50 ml and contained 0.1 mM of each ITS primer, 0.2 mM of each deoxynucleoside triphosphate (dNTP) (Invitrogen, Carlsbald, CA), 2.5 mM MgCl2, 10% by vol. MgCl2-free 10r reaction buffer A (Promega, Madison, WI), one unit Taq polymerase (Promega), and at least 50 ng genomic DNA. PCR was carried out in a PCT-100 thermocycler (MJ Research, Boston, MA) with the following protocol : an initial denaturing step at 94 xC (2 min) followed by 34 cycles of 94 x (30 s), 55 x (30 s), 72 x (1 min) ; and a final elongation step at 72 x (7 min). For restriction enzyme digestions, 2 ml PCR product were combined with 10 % by vol. 10r REact 3 buffer (Invitrogen), 0.1 mg mlx1 BSA (Promega), and 20 units EcoRI in a total vol. of 20 ml. Reactions were incubated at 37 x for 2 h. Following digestion, products were separated by electrophoresis on 1.5 % agarose gels (SeaKem, Rockland, ME), which were stained with ethidium bromide and photographed under UV illumination.
A. J. Fisher, T. R. Grdn. and J. M. DiTomaso
441
Table 1. Origin and host plant of Claviceps purpurea isolates. Isolate
Origina
Year
Host
G1 CMD 1 IA 1 NGE 1 165 204 374b 428b 478
MacDoel, CA, USA Aberdeen, ID, USA L’Anse aux meadows, Newfoundland, Canada Zubri, Czech Republic Lauderdale, AL, USA Altamont, AL, USA Hohenheim, Germany Brˇ ezno, Czech Republic
2001 2000 2001 1994 1996
Secale cereale S. cereale Leymus mollis Poa pratensis Festuca arundinacea F. arundinacea Secale cereale Dactylis glomerata
G2 236b 434b
Vlcˇı´ Pole u Bousova, Czech Republic Phillipsreuth, Germany
1998
Molinia coerulea Dactylis spp.
G3 ADI 1 ARG 1 ARG 2 ARG 3 CDE 1 CDE 2 CDE 3 CDE 4 CDE 5 CDE 6 CDE 7 CDE 8 CDE 9 CDE 10 CDE a1 CDE a3 CDE a4 CNB 21 CPE 10 CSL 1 CSL 2 CSL 3 CSL 4 CSL 5 CSL 6 CSL 7 CSL 8 CSL 9 CSL 10 CSM 10 FSA 1 GML 1 IRE 3 IRE 12 NYF 1 NYH 1 RIH 1 RIH 2 WDI 1 WLT 1 WPR 1 500b 538b
Dolphin Island, AL, USA Argentina Celpa Marsh, Argentina Argentina Celpa Marsh, Argentina Argentina Celpa Marsh, Argentina Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Bolinas Lagoon, CA, USA Palo Alto, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA San Mateo, CA, USA St. Augustine, FL, USA Marsh Landing, GA, USA Dublin, Ireland Dublin, Ireland Flax River, NY, USA Hempstead Bay, NY, USA Rhode Island, USA Rhode Island, USA Willapa River, WA, USA Leadbetter State Park, WA, USA Palix River, WA, USA Southriver, NJ, USA Marchwood, UK
2000 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2001 2001 2002 2002 2002 2002 2002 2002 2002 2002 2002 2001 2001 2000 2000 2001 2001 2001 2001 2001 2001 2002 2001 2002 1998 1999
Spartina alterniflora S. densiflora S. densiflora S. densiflora S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. alterniflora S. alterniflora S. alterniflora S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. alterniflora S. alterniflora S. anglica S. anglica S. alterniflora S. alterniflora S. alterniflora S. alterniflora Distichlis spicata S. alterniflora S. alterniflora S. alterniflora S. alterniflora
a
Numbered isolates were generously supplied by S. Pazoutova´ (Institute of Microbiology, Czech Academy of Sciences, Prague). Representative isolates have been deposited in CCF (Culture Collection of Fungi, Department of Botany, Faculty of Sciences, Charles University, Prague) (Pazoutova´ et al. 2000, 2002b). b
RAPD and AFLP marker development RAPD primers 1CR, 206, 257 (Pazoutova´ et al. 2000) OPA02, OPA03, OPA04, OPA08, OPE01, OPE04, and
OPE14 (Operon Technologies, Alameda, CA) were used. Reaction volumes totalling 15 ml contained 5 mM primer, 0.2 mM of each dNTP (Promega), 2 mM MgCl2, 10 % by vol. 10r MgCl2-free reaction
Claviceps purpurea in salt marsh habitats buffer A (Promega), 0.6 units Taq polymerase, and at least 20 ng genomic DNA. Polymerase chain reaction was carried out on an Eppendorf Mastercycler gradient 5331 (Eppendorf, Cologne) using a protocol with an initial denaturing step at 94 x (4 min) followed by 35 cycles of 93 x (20 s), 46.6 x (1 min), 73 x (1 min) ; and a final elongation step at 73 x (6 min). PCR products were separated by electrophoresis on 1.5 % agarose gels (SeaKem), which were stained with ethidium bromide and photographed under UV illumination. AFLP markers were developed using the method described by Vos et al. (1995), except the restriction enzyme Tru9I (New England Biolabs, Beverly, MA) was used instead of MseI (Tru9I is an isoschizomer of MseI). Three primer combinations with the addition of two selective nucleotides at the 3k end produced ample, reproducible markers. The combinations used for selective amplification were EcoRI (5k-GACTGCGTA CCAATC)+CA/Tru9I (GATGAGTCCTGAGTAA) +AA, EcoRI+AA/Tru9I+GG and EcoRI+AT/ Tru9I+AC. Each 20 ml reaction contained 3.75 mM of each primer (Life Technologies inc., Gaithersburg, MD), 0.2 mM of each dNTP (Promega), 2.5 mM MgCl2, 10r MgCl2-free reaction buffer A (Promega), 0.5 units Taq polymerase (Promega), and 10 ml template. Thermocycler conditions began with an initial denaturing step at 72 x (1 min) followed by 94 x (3 min). Next were nine cycles of 65 x (30 s) decreasing by 1 x each cycle followed by 35 cycles at 56 x (30 s) ; and a final elongation step at 72 x (1 min). A Gene Amp PCR System 9700 (Applied Biosystems, Foster City, CA) was used for all AFLP amplifications. Following amplification, reactions were loaded onto a 4 % acrylamide gel buffered in TBE (1.35 M Tris, 0.45 M boric acid, 25 mM EDTA) for approximately 2 h at 1800–1900 V using a Bio-Rad Sequi-General GT Sequencing Cell (Bio-Rad Laboratories, Hercules, CA), as described by Douhan et al. (2002). Gels were silver-stained following the method described by Bassam et al. (1991) and allowed to dry before visual scoring of bands.
Molecular data analysis All monomorphic and polymorphic RAPD and AFLP markers were scored using a binary system (zero= absent ; and one=present) and the resulting matrix was used to create phenetic cladograms by means of UPGMA algorithms in PAUP 4.0b10. Percent similarity between isolates was calculated by dividing the number of shared markers by the total number of markers and multiplying by 100. Analysis of molecular variance (AMOVA ; Excoffier et al. 1992) was used to partition the total genetic variation among groups (G1, G2 and G3) and within populations of G3 C. purpurea. AMOVA was analysed using Arlequin 2.0 (Schneider et al. 2000).
442 RESULTS This study revealed G3 Claviceps purpurea on three new species, Spartina foliosa, S. densiflora, and Distichlis spicata. G1 and G3 isolates contained an EcoRI restriction site within the 5.8S rDNA, while G2 isolates lacked this site, as reported by Pazoutova´ et al. (2000). RAPD and AFLP protocols produced clear polymorphic bands, repeated in at least two reactions. RAPD primers produced four to eleven bands each, for a total of 62 distinct presence/absence characters. Three AFLP primer combinations produced 30–60 bands each for 159 presence/absence characters. Cluster analyses of RAPD and AFLP data sets each produced trees with similar overall topologies (Fig. 1). A cladogram based on a unified data set (RAPD and AFLP data combined) was congruent with those obtained from each data set separately (Fig. 2). Between-group diversity was high ; only 1.8 % of markers were shared by all members of G1, G2, and G3. Within-group similarity for G3 isolates was 35 %. RAPD and AFLP profiles of G3 isolates collected from all salt marsh grasses were similar to isolates 500 and 538 described in Pazoutova´ et al. (2000b), from Marchwood, UK, and New Jersey, USA respectively. A UPGMA analysis based on Euclidian distance yielded a tree with three main clusters that correspond to the groups described by Pazoutova´ et al. (2000) supported by bootstrap values above 90 % (Figs 1–2). Neighbour-joining analysis produced a tree (not shown) with similar topology that included the same major clusters found in the UPGMA tree. Along with G1 and G2, the G3 cluster was accorded 100 % bootstrap support, evidence for a single origin of this geographically diverse collection. Within G3, isolates from the Pacific coast of the USA clustered together. Markers within the Pacific coast cluster were 83 % similar. In California, isolates from S. alterniflora clustered more closely with isolates from S. foliosa in the same area than with isolates from S. alterniflora in Washington (WPR 1 and WLT 1). Isolates from California were 90% similar ; withinpopulation similarities for Point Reyes and Mountain View were 94 and 93% respectively. Isolates from the northeastern states of New Jersey, New York, and Rhode Island formed a distinct clade that included isolates from Ireland and the UK (Fig. 2). Isolates from Argentina also formed a well supported cluster. AMOVA revealed significant genetic variation between the three habitat specific groups: G1, G2, and G3 (Fst=0.920, P=0.03) (Table 2). A large proportion of the genetic variation was partitioned among (63.5 %), rather than within groups (28.5 %). Considering only G3 isolates, a large proportion of the genetic variation was partitioned among (86 %), rather than within geographic regions (14 %) (Fst=0.859, P=0.00) (Tables 2 and 3).
A. J. Fisher, T. R. Grdn. and J. M. DiTomaso
RAPD
94
100
99
98
0.01 changes
CDE 1 CDE 2 CDE 3 CDE 4 CDE 5 CDE 6 CDE 8 CDE 9 CDE 10 CDE a1 CDE a3 CDE a4 CSL 1 CSL 3 CSL 4 CSL 5 CSL 6 CSL 7 CSL 8 CSL 9 CSL 10 CPE 10 CSM 10 CSL 2 CDE 7 CNB 21 WPR 1 WLT 1 WDI 1 ARG 1 ARG 2 ARG 3 ADI 1 GML 1 FSA 1 RIH 1 NYF 1 NYH 1 RIH 2 IRE 3 IRE 12 538 500 NGE 1 CMD 1 165 204 374 ia 1 428 236 434
443
California (G3)
Washington (G3) Argentina (G3) Southeast US (G3)
Northeast US + Europe (G3)
(G1) (G1)
CDE 1 CDE 4 CDE 7 CDE 2 CDE 9 CNB 21 CPE 10 CSL 1 CSL 4 CSL 7 CSL 9 CDE 3 CDE 5 CDE 6 CDE 8 CDE a1 CDE 10 CSL 2 CSL 6 CSM 10 CDE a3 CSL 3 CSL 5 CSL 8 CSL 10 CDE a4 WPR 1 WLT 1 WDI 1 ADI 1 FSA 1 GML 1 ARG 1 ARG 2 ARG 3 IRE 3 IRE 12 NYH 1 NYF 1 RIH 1 RIH 2 500 538 CMD 1 IA 1 165 204 374 428 NGE 1 236 434
AFLP
100
98
94 100
100
100
100
97 100 97
0.01 changes
Fig. 1. Comparison of cladograms from RAPD analysis (left) and from amplified fragment length polymorphism (AFLP) fingerprinting (right).
DISCUSSION The salt marsh group of Claviceps purpurea (G3) has been distinguished from other sub-groups in this species by having an EcoRI site in the 5.8S ribosomal DNA, and a unique banding pattern based on RAPD analysis of nuclear DNA (Pazoutova´ et al. 2000). These criteria, and AFLP analysis, indicate the isolates collected for this study from Spartina spp. and Distichlis spicata are representative of G3 C. purpurea. A high degree of genetic variation among, compared to within groups, as shown by AMOVA, offers further support for three intraspecific groups within C. purpurea. The results of this study revealed G3 C. purpurea on two new Spartina species, S. foliosa and S. densiflora. The presence of C. purpurea on S. foliosa has been reported from herbarium samples (Alderman, Halse & White 2004), however genetic similarity to known G3 samples has not been established. California cordgrass (S. foliosa) is native to the Pacific coast of North America, from California south to the Baja Peninsula in Mexico (Mobberley 1956). In the San Francisco Bay, the range of S. foliosa is threatened by hybridization with a backcrossing swarm resulting from crosses between S. foliosa and S. alterniflora (Daehler & Strong
1997, Ayres, Strong & Baye 2003, Ayres et al. 2004). S. alterniflora was intentionally introduced in to the San Francisco Bay as seed, supplied by a consulting firm in Maryland (Faber 2000), as part of a wetland restoration program in the 1970s (Daehler & Strong 1994, Ayres et al. 2003). S. alterniflora was unintentionally introduced to Willapa Bay, Washington in the late 19th century (Stiller & Denton 1995, Davis et al. 2004) and it is from this estuary that the Washington isolates used in this study were collected (WLT 1, WPR 1, and WDI 1). AMOVA pairwise comparisons revealed that populations from California and Washington were significantly different (Table 3). However, the Pacific coast cluster in the UPGMA generated cladogram (Fig. 2), including isolates from California and Washington, was accorded 100% bootstrap support. 83% similarity among Pacific coast isolates, is consistent with the low variability of a founder population and suggests this population may have originated from a single introduction. It is therefore possible that G3 C. purpurea reached the Pacific coast in association with introduced S. alterniflora. Interestingly, in the San Francisco Bay, S. alterniflora and S. alterniflorarS. foliosa hybrids have a far lower frequency of infection by C. purpurea
Claviceps purpurea in salt marsh habitats
444 Region
82
100 70 100
100
93 100 100 100 0.01 changes
CDE 1 CDE 4 CDE 7 CDE 2 CDE 9 CNB 21 CSL 1 CSL 4 CSL 7 CPE 10 CSL 9 CDE 3 CDE 5 CDE 6 CDE 8 CDE a1 CDE 10 CSL 2 CSL 6 CDE a3 CSM 10 CSL 3 CSL 5 CSL 8 CS CDE a4 WPR 1 WLT 1 WDI 1 ADI 1 FSA 1 GML 1 ARG 1 ARG 2 ARG 3 IRE 3 IRE 12 NYF 1 NYH 1 RIH 1 RIH 2 500 538 CMD 1 165 IA 1 204 374 478 428 NGE 1 236 434
Group
California
G3
Washington Southeast US Argentina Ireland Northeast US United Kingdom
G1 G2
Fig. 2. Unweighted pair group method with arithmetic average algorithm tree based on 221 AFLP and RAPD markers. Bootstrap support is shown only where values >70 %. Twenty-two isolates were analysed, including representatives of three previously described subspecific groups within C. purpurea. The six G1 isolates were collected from Europe, the USA, and Canada, two G2 isolates were collected in Europe, and G1 isolates were collected in the USA, Argentina, and western Europe.
than S. foliosa (A. Fisher, unpubl.). Estuaries in California and Washington contain multiple susceptible species and therefore allow testing of the relative importance of geographic association versus host preference. In California, isolates from S. alterniflora and S. foliosa form a minor cluster with 95% bootstrap support that does not include S. alterniflora isolates from Washington. This result, along with very low genetic diversity and results from AMOVA pairwise comparisons, suggests the San Francisco Bay population is reproductively isolated. Spartina anglica, host to G3 C. purpurea on the coast of the British Isles and now identified as a host in Ireland, also has a hybrid origin, being derived from native S. maritima and introduced S. alterniflora (Raybould et al. 1991). The first incidence of C. purpurea on S. anglica and S.rtownsendii, a sterile hybrid of S. anglica and S. alterniflora, was reported in Ireland in 1975 (Boyle 1976). While unsure of its origin, Boyle argued that this and a herbarium specimen of C. purpurea on S. anglica collected in 1971 were the first definitive reports of C. purpurea on Spartina spp. in Great Britain and Ireland. Distichlis spicata has been previously identified as a host to C. purpurea (Sprague 1950), and is currently the only known host to G3 outside the genus Spartina.
Perhaps not surprisingly, Distichlis and Spartina are closely related genera, belonging to the same subfamily (Chloridoideae) and tribe (Cynodonteae) (Peterson et al. 2001). This suggests a phylogenetic correspondence between host and pathogen. The range of D. spicata includes the Atlantic coast of North America, the Gulf coast states, Cuba, and the Pacific coast of North America, from British Columbia south to Mexico, and South America (Hitchcock 1971). D. spicata is typically found in salt marshes and seashores on moist and alkaline soils, and borders some S. alterniflora marshes in Willapa Bay, WA. A new host to G3 C. purpurea identified in this study was S. densiflora. Isolates from Argentina formed a well supported cluster by UPGMA, consistent with their apparent geographic isolation relative to other isolates in this collection. However, AMOVA pairwise comparisons showed that isolates from Argentina were not significantly different from isolates from western Europe (UK and Ireland) (P=0.154) and the Northeast USA (New Jersey, New York, and Rhode Island) (P=0.077). Spartina densiflora is native to the Atlantic and Pacific coasts of southern South America (Mobberley 1956). Based on alkaloid profile, Samuelson & Gjerstad (1966) proposed that C. purpurea from Spartina spp. in coastal Argentina was a
A. J. Fisher, T. R. Grdn. and J. M. DiTomaso
445
Table 2. Variation among and within groups of Claviceps purpurea based on analysis of molecular variance. D .F .
Sources of variation
Sum of squares
I. Grouping based on habitat specificity (G1, G2, G3) Among groups 2 531.16 Among populations Within groups 5 316.56 Within populations 45 144.79 II. G3 groupings based on geographic locations Among populations 5 316.56 Within populations 38 73.07
Variance
% Total
Fst
P value
25.676
63.51
0.920
0.03
11.536 3.217
28.53 7.96
11.785 1.923
85.97 14.03
0.00 0.00 0.859
0.00 0.00
Table 3. Pairwise genetic distances (below diagonal) and significance (above diagonal) between G3 Claviceps purpurea populations. Populationa
California
Argentina
south-east USA
western Europe
north-east USA
Washington
California Argentina south-east USA Western Europe north-east USA Washington
– 0.932* 0.883* 0.930* 0.901* 0.636*
0.000 – 0.738* 0.773 0.713 0.850*
0.000 0.000 – 0.723 0.687* 0.718
0.000 0.154 0.077 – 0.385* 0.852*
0.000 0.077 0.000 0.000 – 0.780*
0.000 0.000 0.154 0.000 0.000 –
a
South-east USA includes isolates from Alabama, Florida and Georgia; western Europe includes isolates from UK and Ireland; north-east USA isolates include New Jersey, New York and Rhode Island. Significant differences at P=0.01 (*).
separate species, C. maritima, or ‘feather ergot. ’ The host at the time was misidentified as S. maritima (Eleuterius & Meyers 1977). Eleuterius & Meyers (1977) proposed it was more likely to have been S. alterniflora, based on floristic surveys of the region, but since S. densiflora is now known to be a host in South America, it could be a candidate as well. Invasive and naturalized populations of S. densiflora persist in the San Francisco Bay and Humboldt Bay in northern California. In 2002, S. densiflora was sampled in the north San Francisco Bay along Corta Madera Creek and at four sites in Humboldt Bay. C. purpurea was not detected at either location (A. Fisher, unpubl.). Additional collections are needed to determine if the Argentina population is the result of an introduction or, perhaps more likely, if natural populations of G3 extend along the Atlantic coastline from North America through the South American coastal salt marshes. There is evidence of C. purpurea on Spartina in this range. For example, a 1943 herbarium specimen of C. purpurea collected in Uruguay from S. brasiliensis, a synonym of S. alterniflora, is in the US National Fungus Collection (BPI) (Duncan et al. 2002). The collection also contains specimens collected from S. alterniflora along the east coast of the USA from as early as the 1930s. Isolates collected from New York, Rhode Island, and New Jersey on S. alterniflora formed a major cluster using UPGMA with isolates on S. anglica from Ireland and the UK. However, AMOVA pairwise comparisons showed populations from these geographic regions to be significantly different (P=0.385). Interestingly, the origin of S. alterniflora in Europe is thought to be somewhere between Boston, MA and
Newfoundland, Canada (Hubbard 1965). As previously discussed, S. anglica is a hybrid between the native S. maritima and the introduced S. alterniflora. While isolates from Ireland clustered together with 100 % bootstrap support, isolate 538 from the UK was not part of that minor cluster. Therefore, it is unclear if G3 established in Western Europe following one, or multiple, introduction events. Our results support the recognition of three discrete groups within C. purpurea. Results of RAPD and AFLP analysis showed high intergroup variability between G1, G2, and G3. Using a collection of isolates from a different set of hosts, Jungehu¨lsing & Tudzynski (1997) also found high intraspecific variability using RAPD markers. In their study, parsimony analysis grouped samples from the same host species together, suggesting some degree of host specificity. More sampling is necessary to reveal the ecological and physiological range of G3, although there is evidence that the pathogenicity profile of G3, following greenhouse inoculations, is not limited to salt marsh species (Pazoutova´ et al. 2002a). Laboratory experiments notwithstanding, our results imply a degree of specificity in the ecological range of G3. Our results confirm the three intraspecific groups within C. purpurea and support from both RAPD and AFLP profiles strongly suggests a singular origin of G3 C. purpurea followed by widespread geographic dispersal.
ACKNOWLEDGEMENTS We thank Sylvie Pazoutova´ for providing Claviceps purpurea isolates; Alejandro Bartolis, Kennebla L. Heck, Dino Garcia-Rossi, Tracy Buck, Alexander Kolker, Steve Orloff, Michael Burzynski, Darrell
Claviceps purpurea in salt marsh habitats Wesenberg, and E. Landy for field samples; and B. Aegerter, Debra Ayres, Ravindra Bhat, Greg Douhan, and David Rizzo for technical assistance and use of laboratory equipment. The manuscript was improved by comments from Donald Strong and two anonymous reviewers. This study was financially supported by The Garden Club of America, Jastro Shields and Humanities awards from the University of California, Davis and NSF Biocomplexity DEB 0083583 to A.H.
REFERENCES Alderman, S. C., Halse, R. R. & White, J. F. (2004) A reevaluation of the host range and geographical distribution of Claviceps species in the United States. Plant Disease 88 : 63–81. Ayres, D. R., Smith, D. L., Zaremba, K., Klohr, S. & Strong, D. R. (2004) Spread of exotic cordgrasses and hybrids (Spartina spp.) in the tidal marshes of San Francisco Bay, CA, USA. Biological Invasions 6: 221–231. Ayres, D. R., Strong, D. R. & Baye, P. (2003) Spartina foliosa (Poaceae) – a common species on the road to rarity? Madron˜o 53: 209–213. Bassam, B. J., Caetano-Anolles, G. & Gresshoff, P. M. (1991) Fast and sensitive silver staining of DNA in polyacrylamide gels. Annals of Biochemistry 196 : 80–83. Bowden, B. N. (1965) Modern grass taxonomy. Outlook on Agriculture 4: 243–253. Boyle, P. J. (1976) Ergot epiphytotic on Spartina spp. in Ireland. Irish Journal of Agricultural Research 15 : 419–424. Daehler, C. C., Anttila, C. K., Ayres, D. R., Strong, D. R. & Bailey, J. P. (1999) Evolution of a new ecotype of Spartina alterniflora (Poaceae) in San Francisco Bay, California, USA. American Journal of Botany 86 : 543–546. Daehler, C. C. & Strong, D. R. (1994) Variable reproductive output among clones of Spartina alterniflora (Poaceae) invading San Francisco Bay, California: the influence of herbivory, pollination, and establishment site. American Journal of Botany 81: 307–313. Daehler, C. C. & Strong, D. R. (1997) Hybridization between introduced smooth cordgrass (Spartina alterniflora, Poaceae) and native California cordgrass (Spartina foliosa) in San Francisco Bay, California, USA. American Journal of Botany 84: 607–611. Davis, H. G., Taylor, C. M., Civille, J. C. & Strong, D. R. (2004) An allee effect at the front of a plant invasion: Spartina in a Pacific estuary. Journal of Ecology 92: 321–327. Douhan, G. W., Peever, T. L. & Murray, T. D. (2002) Multilocus population structure of Tapesia yallundae in Washington State. Molecular Ecology 11: 2229–2239. Duncan, R. A., Sullivan, R., Alderman, S. C., Spatafora, J. W. & White, J. F. (2002) Claviceps purpurea var. spartinae var. nov.: an ergot adapted to the aquatic environment. Mycotaxon 81: 11–25. Eleuterius, L. N. & Meyers, S. P. (1974) Claviceps purpurea on Spartina in coastal marshes. Mycologia 66: 979–986. Eleuterius, L. N. & Meyers, S. P. (1977) Alkaloids of Claviceps from Spartina. Mycologia 69 : 838–840. Excoffier, L., Smouse, P. E. & Quattro, J. M. (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: applications to human mitochondrial DNA restriction data. Genetics 131: 479–491. Faber, P. (2000) Good intentions gone awry: why would anyone bring an alien cordgrass to San Francisco ? Coast and Ocean 16: 14–17.
446 Hitchcock, A. S. (1971) Manual of the Grasses of the United States. Dover, New York. Hubbard, J. C. E. (1965) Spartina marshes in southern England. VI. Pattern of invasion in Poole Harbour. Journal of Ecology 53: 799–813. Jungehu¨lsing, U. & Tudzynski, P. (1997) Analysis of genetic diversity in Claviceps purpurea by RAPD markers. Mycological Research 101: 1–6. Kobel, H. & Sanglier, J. J. (1978) Formation of ergotoxine alkaloids by fermentation and attempts to control their biosyntheses. FEMS Symposium 5: 233–242. Loveless, A. R. (1971) Conidial evidence for host restriction in Claviceps purpurea. Transactions of the British Mycological Society 56: 419–434. Mobberley, D. G. (1956) Taxonomy and distribution of the genus Spartina. Iowa State Journal of Science 30: 471–574. Pazoutova´, S., Cagas, B., Kolı´ nska´, R. & Honza´tko, A. (2002a) Host specialization of different populations of ergot fungus (Claviceps purpurea). Czech Journal of Genetics and Plant Breeding 38: 75–81. Pazoutova´, S., Olsovska, J., Linka, M., Kolı´ nska´, R. & Flieger, M. (2000) Chemoraces and habitat specialization of Claviceps purpurea populations. Applied and Environmental Microbiology 66: 5419–5425. Pazoutova´, S., Raybould, A. F., Honza´tko, A. & Kolı´ nska´, R. (2002b) Specialized populations of Claviceps purpurea from salt marsh Spartina species. Mycological Research 106: 210–214. Peterson, P. M., Soreng, R. J., Davidse, D., Filgueiras, T. S., Zuloaga, F. O. & Judziewicz, E. J. (2001) Catalogue of New World Grasses (Poaceae) : II. Subfamily Chloridoideae. Contributions of the US National Herbarium 41: 1–255. Raybould, A. F., Gray, A. J., Lawrence, M. J. & Marshall, D. F. (1991) The evolution of Spartina anglica C. E. Hubbard (Gramineae) ; the origin and genetic variability. Biological Journal of the Linnean Society 43: 111–126. Samuelson, R. J. & Gjerstad, G. (1966) Intermediary metabolism of ergot XIII. Feather ergot, a new, promising Claviceps species. Meddelelser Fra Norsk Farmaceutisk Selskap 28: 229–237. Schneider, S., Roessli, D. & Excoffier, L. (2000) Arlequin: A software for population genetics data analysis. Version 2.001. Genetics and Biometry Laboratory, University of Geneva. Sprague, R. (1950) Diseases of Cereals and Grasses in North America. Ronald Press: New York. Stager, R. (1922) Beitrag zur Verbreitungsbiologie der ClavicepsSklerotien. Zentralblatt Fur Bakteriologie Parasitenkunde Infefektionskranheiten Und Hygiene 56: 329–339. Stiller, J. W. & Denton, A. L. (1995) One hundred years of Spartina alterniflora (Poaceae) in Willapa Bay, Washington: random amplified polymorphic DNA analysis of an invasive population. Molecular Ecology 4: 355–363. Vos, P., Hogers, R., Bleeker, M., Reijans, M., Van De Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M. & Zabeau, M. (1995) AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407–4444. 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 (M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White, eds): 315–322. Academic Press, San Diego.
Corresponding Editor: J. K. Stone
439
doi:10.1017/S0953756205002467 Printed in the United Kingdom.
Geographic distribution and diversity in Claviceps purpurea from salt marsh habitats and characterization of Pacific coast populations
Alison J. FISHER1*, Thomas R. GORDON2 and Joseph M. DITOMASO1 1
Department of Vegetable Crops, One Shields Avenue, University of California, Davis, CA 95616-8755, USA. Department of Plant Pathology, One Shields Avenue, University of California, Davis, CA 95616-8755, USA. E-mail : afi[email protected]
2
Received 11 June 2004; accepted 4 January 2004.
Claviceps purpurea specific to grasses in salt marsh habitats (Group G3) has previously been identified on Spartina spp. in two locations : New Jersey, USA and southern England. We have identified this subgroup of C. purpurea (G3) in 11 distinct populations including western Europe, South America, and along the Atlantic and Pacific Coasts of the USA. In addition, G3 C. purpurea was discovered on a new host grass genus, Distichlis. Unweighted pair group mean analyses of AFLP and RAPD data reveal distinct structure in G3 C. purpurea populations. Pacific coast populations show little diversity, suggesting they may have been introduced recently in that region. 43 isolates, representing 11 populations were identified as G3 based on the presence of an EcoRI restriction site in the 5.8S ribosomal DNA, and a clear genetic separation from isolates representing the other two C. purpurea subgroups (G1 and G2). In addition, all isolates originating from Spartina densiflora, S. foliosa, S. alterniflora, and S. anglica were identified as belonging to G3. RAPD and AFLP analyses supported the recognition of three discrete groups within C. purpurea and revealed high genetic variability between groups, with only 1.8 % of polymorphic markers shared across all isolates. Similarly, analysis of molecular variation (AMOVA) revealed that genetic variability was mainly due to variations between groups (63.5%) rather than within groups (28.5 %) or within populations (7.96%). G3 isolates were 35 % similar, Pacific coast isolates 83 % similar. Ninety percent similarity among isolates from the San Francisco Bay Area suggests this is a recently introduced population.
INTRODUCTION Claviceps purpurea, the ergot fungus, is a well-known pathogen of cereal grains and forage grasses. Although it is a commonly occurring parasite of grasses in unmanaged landscapes, research has emphasized its impacts on agronomic systems. The species has a global distribution, and a wide host range within the Poaceae. Species in the subfamily Pooideae are the most common hosts, followed by members of the Arundinoideae and Chloridoideae (Bowden 1965, Loveless 1971). This broad host range has led many researchers to seek evidence of intraspecific substructure based on morphology (Loveless 1971), alkaloid production (Eleuterius & Meyers 1977, Kobel & Sanglier 1978), and genetic characteristics (Jungehu¨lsing & Tudzynski 1997). Pazoutova´ et al. (2000) synthesized previous research on C. purpurea morphology, alkaloid chemistry and genetics, and identified three distinct groups within the species, describing them as ‘chemoraces. ’ Rather than * Corresponding author.
a delimitation based on host range, the intraspecific groups were defined by habitat specialization. The largest group, G1, was associated with land grasses, and G2 with grasses in ‘wet and shady ’ environments, whereas G3 was found only on grasses in salt marsh habitats. Group 3 ergot is synonymous with C. purpurea var. spartinae (Duncan et al. 2002) and is referred to herein as G3. Sclerotia of both G2 and G3 C. purpurea float in water while sclerotia from terrestrial C. purpurea (G1) sink, clarifying the significance of earlier reports that sclerotia of some isolates float, and the association of this trait with host habitat (Stager 1922). Lengths of conidia range from 5–8 mm for G1, 7–10 mm for G2, and 10–12 mm for G3. Isolates associated with G1 and G3 share an EcoRI restriction site in the 5.8S ribosomal DNA (rDNA), which G2 isolates lack (Pazoutova´ et al. 2000). The three groups can be differentiated by production of a unique set of alkaloids, and by RAPD analysis (Pazoutova´ et al. 2000). G3 C. purpurea has been isolated exclusively from cordgrass species of the genus Spartina, a member of
Claviceps purpurea in salt marsh habitats the subfamily Chloridoideae, suggesting a very narrow host range compared to G1 and G2 (Pazoutova´ et al. 2000). Two host species to G3 have been identified, each known only from one location: common cordgrass, S. anglica, in the UK and smooth cordgrass, S. alterniflora, from New Jersey on the Atlantic coast of the USA (Pazoutova´ et al. 2000, 2002a). An analysis using unweighted pair group method with arithmetic averaging (UPGMA) placed the US isolates in an outlier position, suggesting a North American origin for the European population. Claviceps purpurea was first found on a Spartina species in North America in 1895 (Eleuterius & Meyers 1974), and was not reported from Europe until the 1970s (Boyle 1976), which is also consistent with a North American origin. Whereas only two populations of G3 have been identified by means of molecular markers, the host to this intraspecific group, Spartina, is a widespread genus including both terrestrial and halophytic salt marsh species. Most members of the genus, including S. alterniflora, share a common native geographic range along the Atlantic coast of North and South America. Introduced populations of S. alterniflora have established on the Pacific coast as well, in the San Francisco Bay, California (Daehler & Strong 1994) and Willapa Bay, Washington (Stiller & Denton 1995). Thus, the potential geographic range of G3 may be greater than is presently recognized and it is reasonable to think that diversity in populations of the pathogen may reflect the diversity of the host species. The objectives of this study were to assess the geographic distribution and ecological host range of G3 C. purpurea, and to evaluate the genetic diversity revealed by this broader sampling of the salt marsh group. In addition, genetic diversity among Pacific coast isolates was evaluated to determine if G3 ergot is likely to have been introduced to that region.
MATERIALS AND METHODS Isolate collection and culturing Isolates included in this study, and their origins, are listed in Table 1. Collections from the same location were taken from inflorescences from host plants at least one meter apart. Intensive sampling was done in the San Francisco Bay area. To examine within population diversity, isolates were obtained from ten inflorescences collected from two S. foliosa marshes: Point Reyes National Seashore (NS) on the outer coast, and Mountain View in the West Bay. In addition, one infected S. foliosa inflorescence was collected at each of three additional marshes : Palo Alto, San Mateo in the West Bay and Bolinas Lagoon on the outer coast. To reduce superficial contaminants, sclerotia were treated by immersion in 70% ethanol for 30 s followed by 1.5 min in 1.3 % sodium hypochlorite and three
440 successive, three min rinses in sterile deionized water. The rind of the sclerotium was removed and fungal tissue plated onto potato dextrose agar (PDA). To ensure that each isolate represented a single genotype, cultures were streaked onto water agar and a hyphal tip was subcultured. Pure cultures were grown on filter paper overlaid on PDA ; fully colonized papers were removed, dried, and refrigerated at 4 xC. Isolates not collected in the field were obtained as pure cultures from S. Pazoutova´, including G1 isolates 374, 428, G2 isolates 236 and 434 (Pazoutova´ et al. 2000), G3 isolates 500 and 538 (Pazoutova´ et al. 2002b), and G1 isolates 165, 204 and 478, which have not been included in a previously published report (Table 1). DNA extraction Tissue for DNA extraction was obtained by growing isolates either on cellophane overlaid on PDA or in potato dextrose broth (PDB). Isolates grown on cellophane were harvested after nine days and stored at x20 x. Fungal biomass recovered from PDB cultures was harvested after seven days by filtration through cheesecloth, lyophilized and stored at x20 x. Mycelia (10–20 mg) in microcentrifuge tubes, were frozen by dipping tubes in liquid nitrogen, and ground with a pestle. Total DNA was extracted using the methods described by Daehler et al. (1999). After DNA was recovered in solution for the final time, 75 ml of supernatant were withdrawn and mixed with 200 ml Tris-EDTA buffer (10 mM Tris, 1 mM EDTA, pH 7.8), and DNA concentration was quantified by spectrophotometry. EcoRI analysis The polymerase chain reaction (PCR) was used to amplify the region containing the 5.8S rDNA, using ITS1 and ITS4 primers (White et al. 1990). Primers were desalted with the aid of an Edge Biosystems SR gel filtration cartridge. Each reaction had a final vol. of 50 ml and contained 0.1 mM of each ITS primer, 0.2 mM of each deoxynucleoside triphosphate (dNTP) (Invitrogen, Carlsbald, CA), 2.5 mM MgCl2, 10% by vol. MgCl2-free 10r reaction buffer A (Promega, Madison, WI), one unit Taq polymerase (Promega), and at least 50 ng genomic DNA. PCR was carried out in a PCT-100 thermocycler (MJ Research, Boston, MA) with the following protocol : an initial denaturing step at 94 xC (2 min) followed by 34 cycles of 94 x (30 s), 55 x (30 s), 72 x (1 min) ; and a final elongation step at 72 x (7 min). For restriction enzyme digestions, 2 ml PCR product were combined with 10 % by vol. 10r REact 3 buffer (Invitrogen), 0.1 mg mlx1 BSA (Promega), and 20 units EcoRI in a total vol. of 20 ml. Reactions were incubated at 37 x for 2 h. Following digestion, products were separated by electrophoresis on 1.5 % agarose gels (SeaKem, Rockland, ME), which were stained with ethidium bromide and photographed under UV illumination.
A. J. Fisher, T. R. Grdn. and J. M. DiTomaso
441
Table 1. Origin and host plant of Claviceps purpurea isolates. Isolate
Origina
Year
Host
G1 CMD 1 IA 1 NGE 1 165 204 374b 428b 478
MacDoel, CA, USA Aberdeen, ID, USA L’Anse aux meadows, Newfoundland, Canada Zubri, Czech Republic Lauderdale, AL, USA Altamont, AL, USA Hohenheim, Germany Brˇ ezno, Czech Republic
2001 2000 2001 1994 1996
Secale cereale S. cereale Leymus mollis Poa pratensis Festuca arundinacea F. arundinacea Secale cereale Dactylis glomerata
G2 236b 434b
Vlcˇı´ Pole u Bousova, Czech Republic Phillipsreuth, Germany
1998
Molinia coerulea Dactylis spp.
G3 ADI 1 ARG 1 ARG 2 ARG 3 CDE 1 CDE 2 CDE 3 CDE 4 CDE 5 CDE 6 CDE 7 CDE 8 CDE 9 CDE 10 CDE a1 CDE a3 CDE a4 CNB 21 CPE 10 CSL 1 CSL 2 CSL 3 CSL 4 CSL 5 CSL 6 CSL 7 CSL 8 CSL 9 CSL 10 CSM 10 FSA 1 GML 1 IRE 3 IRE 12 NYF 1 NYH 1 RIH 1 RIH 2 WDI 1 WLT 1 WPR 1 500b 538b
Dolphin Island, AL, USA Argentina Celpa Marsh, Argentina Argentina Celpa Marsh, Argentina Argentina Celpa Marsh, Argentina Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Point Reyes NS, CA, USA Bolinas Lagoon, CA, USA Palo Alto, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA Mountain View, CA, USA San Mateo, CA, USA St. Augustine, FL, USA Marsh Landing, GA, USA Dublin, Ireland Dublin, Ireland Flax River, NY, USA Hempstead Bay, NY, USA Rhode Island, USA Rhode Island, USA Willapa River, WA, USA Leadbetter State Park, WA, USA Palix River, WA, USA Southriver, NJ, USA Marchwood, UK
2000 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2002 2001 2001 2002 2002 2002 2002 2002 2002 2002 2002 2002 2001 2001 2000 2000 2001 2001 2001 2001 2001 2001 2002 2001 2002 1998 1999
Spartina alterniflora S. densiflora S. densiflora S. densiflora S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. alterniflora S. alterniflora S. alterniflora S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. foliosa S. alterniflora S. alterniflora S. anglica S. anglica S. alterniflora S. alterniflora S. alterniflora S. alterniflora Distichlis spicata S. alterniflora S. alterniflora S. alterniflora S. alterniflora
a
Numbered isolates were generously supplied by S. Pazoutova´ (Institute of Microbiology, Czech Academy of Sciences, Prague). Representative isolates have been deposited in CCF (Culture Collection of Fungi, Department of Botany, Faculty of Sciences, Charles University, Prague) (Pazoutova´ et al. 2000, 2002b). b
RAPD and AFLP marker development RAPD primers 1CR, 206, 257 (Pazoutova´ et al. 2000) OPA02, OPA03, OPA04, OPA08, OPE01, OPE04, and
OPE14 (Operon Technologies, Alameda, CA) were used. Reaction volumes totalling 15 ml contained 5 mM primer, 0.2 mM of each dNTP (Promega), 2 mM MgCl2, 10 % by vol. 10r MgCl2-free reaction
Claviceps purpurea in salt marsh habitats buffer A (Promega), 0.6 units Taq polymerase, and at least 20 ng genomic DNA. Polymerase chain reaction was carried out on an Eppendorf Mastercycler gradient 5331 (Eppendorf, Cologne) using a protocol with an initial denaturing step at 94 x (4 min) followed by 35 cycles of 93 x (20 s), 46.6 x (1 min), 73 x (1 min) ; and a final elongation step at 73 x (6 min). PCR products were separated by electrophoresis on 1.5 % agarose gels (SeaKem), which were stained with ethidium bromide and photographed under UV illumination. AFLP markers were developed using the method described by Vos et al. (1995), except the restriction enzyme Tru9I (New England Biolabs, Beverly, MA) was used instead of MseI (Tru9I is an isoschizomer of MseI). Three primer combinations with the addition of two selective nucleotides at the 3k end produced ample, reproducible markers. The combinations used for selective amplification were EcoRI (5k-GACTGCGTA CCAATC)+CA/Tru9I (GATGAGTCCTGAGTAA) +AA, EcoRI+AA/Tru9I+GG and EcoRI+AT/ Tru9I+AC. Each 20 ml reaction contained 3.75 mM of each primer (Life Technologies inc., Gaithersburg, MD), 0.2 mM of each dNTP (Promega), 2.5 mM MgCl2, 10r MgCl2-free reaction buffer A (Promega), 0.5 units Taq polymerase (Promega), and 10 ml template. Thermocycler conditions began with an initial denaturing step at 72 x (1 min) followed by 94 x (3 min). Next were nine cycles of 65 x (30 s) decreasing by 1 x each cycle followed by 35 cycles at 56 x (30 s) ; and a final elongation step at 72 x (1 min). A Gene Amp PCR System 9700 (Applied Biosystems, Foster City, CA) was used for all AFLP amplifications. Following amplification, reactions were loaded onto a 4 % acrylamide gel buffered in TBE (1.35 M Tris, 0.45 M boric acid, 25 mM EDTA) for approximately 2 h at 1800–1900 V using a Bio-Rad Sequi-General GT Sequencing Cell (Bio-Rad Laboratories, Hercules, CA), as described by Douhan et al. (2002). Gels were silver-stained following the method described by Bassam et al. (1991) and allowed to dry before visual scoring of bands.
Molecular data analysis All monomorphic and polymorphic RAPD and AFLP markers were scored using a binary system (zero= absent ; and one=present) and the resulting matrix was used to create phenetic cladograms by means of UPGMA algorithms in PAUP 4.0b10. Percent similarity between isolates was calculated by dividing the number of shared markers by the total number of markers and multiplying by 100. Analysis of molecular variance (AMOVA ; Excoffier et al. 1992) was used to partition the total genetic variation among groups (G1, G2 and G3) and within populations of G3 C. purpurea. AMOVA was analysed using Arlequin 2.0 (Schneider et al. 2000).
442 RESULTS This study revealed G3 Claviceps purpurea on three new species, Spartina foliosa, S. densiflora, and Distichlis spicata. G1 and G3 isolates contained an EcoRI restriction site within the 5.8S rDNA, while G2 isolates lacked this site, as reported by Pazoutova´ et al. (2000). RAPD and AFLP protocols produced clear polymorphic bands, repeated in at least two reactions. RAPD primers produced four to eleven bands each, for a total of 62 distinct presence/absence characters. Three AFLP primer combinations produced 30–60 bands each for 159 presence/absence characters. Cluster analyses of RAPD and AFLP data sets each produced trees with similar overall topologies (Fig. 1). A cladogram based on a unified data set (RAPD and AFLP data combined) was congruent with those obtained from each data set separately (Fig. 2). Between-group diversity was high ; only 1.8 % of markers were shared by all members of G1, G2, and G3. Within-group similarity for G3 isolates was 35 %. RAPD and AFLP profiles of G3 isolates collected from all salt marsh grasses were similar to isolates 500 and 538 described in Pazoutova´ et al. (2000b), from Marchwood, UK, and New Jersey, USA respectively. A UPGMA analysis based on Euclidian distance yielded a tree with three main clusters that correspond to the groups described by Pazoutova´ et al. (2000) supported by bootstrap values above 90 % (Figs 1–2). Neighbour-joining analysis produced a tree (not shown) with similar topology that included the same major clusters found in the UPGMA tree. Along with G1 and G2, the G3 cluster was accorded 100 % bootstrap support, evidence for a single origin of this geographically diverse collection. Within G3, isolates from the Pacific coast of the USA clustered together. Markers within the Pacific coast cluster were 83 % similar. In California, isolates from S. alterniflora clustered more closely with isolates from S. foliosa in the same area than with isolates from S. alterniflora in Washington (WPR 1 and WLT 1). Isolates from California were 90% similar ; withinpopulation similarities for Point Reyes and Mountain View were 94 and 93% respectively. Isolates from the northeastern states of New Jersey, New York, and Rhode Island formed a distinct clade that included isolates from Ireland and the UK (Fig. 2). Isolates from Argentina also formed a well supported cluster. AMOVA revealed significant genetic variation between the three habitat specific groups: G1, G2, and G3 (Fst=0.920, P=0.03) (Table 2). A large proportion of the genetic variation was partitioned among (63.5 %), rather than within groups (28.5 %). Considering only G3 isolates, a large proportion of the genetic variation was partitioned among (86 %), rather than within geographic regions (14 %) (Fst=0.859, P=0.00) (Tables 2 and 3).
A. J. Fisher, T. R. Grdn. and J. M. DiTomaso
RAPD
94
100
99
98
0.01 changes
CDE 1 CDE 2 CDE 3 CDE 4 CDE 5 CDE 6 CDE 8 CDE 9 CDE 10 CDE a1 CDE a3 CDE a4 CSL 1 CSL 3 CSL 4 CSL 5 CSL 6 CSL 7 CSL 8 CSL 9 CSL 10 CPE 10 CSM 10 CSL 2 CDE 7 CNB 21 WPR 1 WLT 1 WDI 1 ARG 1 ARG 2 ARG 3 ADI 1 GML 1 FSA 1 RIH 1 NYF 1 NYH 1 RIH 2 IRE 3 IRE 12 538 500 NGE 1 CMD 1 165 204 374 ia 1 428 236 434
443
California (G3)
Washington (G3) Argentina (G3) Southeast US (G3)
Northeast US + Europe (G3)
(G1) (G1)
CDE 1 CDE 4 CDE 7 CDE 2 CDE 9 CNB 21 CPE 10 CSL 1 CSL 4 CSL 7 CSL 9 CDE 3 CDE 5 CDE 6 CDE 8 CDE a1 CDE 10 CSL 2 CSL 6 CSM 10 CDE a3 CSL 3 CSL 5 CSL 8 CSL 10 CDE a4 WPR 1 WLT 1 WDI 1 ADI 1 FSA 1 GML 1 ARG 1 ARG 2 ARG 3 IRE 3 IRE 12 NYH 1 NYF 1 RIH 1 RIH 2 500 538 CMD 1 IA 1 165 204 374 428 NGE 1 236 434
AFLP
100
98
94 100
100
100
100
97 100 97
0.01 changes
Fig. 1. Comparison of cladograms from RAPD analysis (left) and from amplified fragment length polymorphism (AFLP) fingerprinting (right).
DISCUSSION The salt marsh group of Claviceps purpurea (G3) has been distinguished from other sub-groups in this species by having an EcoRI site in the 5.8S ribosomal DNA, and a unique banding pattern based on RAPD analysis of nuclear DNA (Pazoutova´ et al. 2000). These criteria, and AFLP analysis, indicate the isolates collected for this study from Spartina spp. and Distichlis spicata are representative of G3 C. purpurea. A high degree of genetic variation among, compared to within groups, as shown by AMOVA, offers further support for three intraspecific groups within C. purpurea. The results of this study revealed G3 C. purpurea on two new Spartina species, S. foliosa and S. densiflora. The presence of C. purpurea on S. foliosa has been reported from herbarium samples (Alderman, Halse & White 2004), however genetic similarity to known G3 samples has not been established. California cordgrass (S. foliosa) is native to the Pacific coast of North America, from California south to the Baja Peninsula in Mexico (Mobberley 1956). In the San Francisco Bay, the range of S. foliosa is threatened by hybridization with a backcrossing swarm resulting from crosses between S. foliosa and S. alterniflora (Daehler & Strong
1997, Ayres, Strong & Baye 2003, Ayres et al. 2004). S. alterniflora was intentionally introduced in to the San Francisco Bay as seed, supplied by a consulting firm in Maryland (Faber 2000), as part of a wetland restoration program in the 1970s (Daehler & Strong 1994, Ayres et al. 2003). S. alterniflora was unintentionally introduced to Willapa Bay, Washington in the late 19th century (Stiller & Denton 1995, Davis et al. 2004) and it is from this estuary that the Washington isolates used in this study were collected (WLT 1, WPR 1, and WDI 1). AMOVA pairwise comparisons revealed that populations from California and Washington were significantly different (Table 3). However, the Pacific coast cluster in the UPGMA generated cladogram (Fig. 2), including isolates from California and Washington, was accorded 100% bootstrap support. 83% similarity among Pacific coast isolates, is consistent with the low variability of a founder population and suggests this population may have originated from a single introduction. It is therefore possible that G3 C. purpurea reached the Pacific coast in association with introduced S. alterniflora. Interestingly, in the San Francisco Bay, S. alterniflora and S. alterniflorarS. foliosa hybrids have a far lower frequency of infection by C. purpurea
Claviceps purpurea in salt marsh habitats
444 Region
82
100 70 100
100
93 100 100 100 0.01 changes
CDE 1 CDE 4 CDE 7 CDE 2 CDE 9 CNB 21 CSL 1 CSL 4 CSL 7 CPE 10 CSL 9 CDE 3 CDE 5 CDE 6 CDE 8 CDE a1 CDE 10 CSL 2 CSL 6 CDE a3 CSM 10 CSL 3 CSL 5 CSL 8 CS CDE a4 WPR 1 WLT 1 WDI 1 ADI 1 FSA 1 GML 1 ARG 1 ARG 2 ARG 3 IRE 3 IRE 12 NYF 1 NYH 1 RIH 1 RIH 2 500 538 CMD 1 165 IA 1 204 374 478 428 NGE 1 236 434
Group
California
G3
Washington Southeast US Argentina Ireland Northeast US United Kingdom
G1 G2
Fig. 2. Unweighted pair group method with arithmetic average algorithm tree based on 221 AFLP and RAPD markers. Bootstrap support is shown only where values >70 %. Twenty-two isolates were analysed, including representatives of three previously described subspecific groups within C. purpurea. The six G1 isolates were collected from Europe, the USA, and Canada, two G2 isolates were collected in Europe, and G1 isolates were collected in the USA, Argentina, and western Europe.
than S. foliosa (A. Fisher, unpubl.). Estuaries in California and Washington contain multiple susceptible species and therefore allow testing of the relative importance of geographic association versus host preference. In California, isolates from S. alterniflora and S. foliosa form a minor cluster with 95% bootstrap support that does not include S. alterniflora isolates from Washington. This result, along with very low genetic diversity and results from AMOVA pairwise comparisons, suggests the San Francisco Bay population is reproductively isolated. Spartina anglica, host to G3 C. purpurea on the coast of the British Isles and now identified as a host in Ireland, also has a hybrid origin, being derived from native S. maritima and introduced S. alterniflora (Raybould et al. 1991). The first incidence of C. purpurea on S. anglica and S.rtownsendii, a sterile hybrid of S. anglica and S. alterniflora, was reported in Ireland in 1975 (Boyle 1976). While unsure of its origin, Boyle argued that this and a herbarium specimen of C. purpurea on S. anglica collected in 1971 were the first definitive reports of C. purpurea on Spartina spp. in Great Britain and Ireland. Distichlis spicata has been previously identified as a host to C. purpurea (Sprague 1950), and is currently the only known host to G3 outside the genus Spartina.
Perhaps not surprisingly, Distichlis and Spartina are closely related genera, belonging to the same subfamily (Chloridoideae) and tribe (Cynodonteae) (Peterson et al. 2001). This suggests a phylogenetic correspondence between host and pathogen. The range of D. spicata includes the Atlantic coast of North America, the Gulf coast states, Cuba, and the Pacific coast of North America, from British Columbia south to Mexico, and South America (Hitchcock 1971). D. spicata is typically found in salt marshes and seashores on moist and alkaline soils, and borders some S. alterniflora marshes in Willapa Bay, WA. A new host to G3 C. purpurea identified in this study was S. densiflora. Isolates from Argentina formed a well supported cluster by UPGMA, consistent with their apparent geographic isolation relative to other isolates in this collection. However, AMOVA pairwise comparisons showed that isolates from Argentina were not significantly different from isolates from western Europe (UK and Ireland) (P=0.154) and the Northeast USA (New Jersey, New York, and Rhode Island) (P=0.077). Spartina densiflora is native to the Atlantic and Pacific coasts of southern South America (Mobberley 1956). Based on alkaloid profile, Samuelson & Gjerstad (1966) proposed that C. purpurea from Spartina spp. in coastal Argentina was a
A. J. Fisher, T. R. Grdn. and J. M. DiTomaso
445
Table 2. Variation among and within groups of Claviceps purpurea based on analysis of molecular variance. D .F .
Sources of variation
Sum of squares
I. Grouping based on habitat specificity (G1, G2, G3) Among groups 2 531.16 Among populations Within groups 5 316.56 Within populations 45 144.79 II. G3 groupings based on geographic locations Among populations 5 316.56 Within populations 38 73.07
Variance
% Total
Fst
P value
25.676
63.51
0.920
0.03
11.536 3.217
28.53 7.96
11.785 1.923
85.97 14.03
0.00 0.00 0.859
0.00 0.00
Table 3. Pairwise genetic distances (below diagonal) and significance (above diagonal) between G3 Claviceps purpurea populations. Populationa
California
Argentina
south-east USA
western Europe
north-east USA
Washington
California Argentina south-east USA Western Europe north-east USA Washington
– 0.932* 0.883* 0.930* 0.901* 0.636*
0.000 – 0.738* 0.773 0.713 0.850*
0.000 0.000 – 0.723 0.687* 0.718
0.000 0.154 0.077 – 0.385* 0.852*
0.000 0.077 0.000 0.000 – 0.780*
0.000 0.000 0.154 0.000 0.000 –
a
South-east USA includes isolates from Alabama, Florida and Georgia; western Europe includes isolates from UK and Ireland; north-east USA isolates include New Jersey, New York and Rhode Island. Significant differences at P=0.01 (*).
separate species, C. maritima, or ‘feather ergot. ’ The host at the time was misidentified as S. maritima (Eleuterius & Meyers 1977). Eleuterius & Meyers (1977) proposed it was more likely to have been S. alterniflora, based on floristic surveys of the region, but since S. densiflora is now known to be a host in South America, it could be a candidate as well. Invasive and naturalized populations of S. densiflora persist in the San Francisco Bay and Humboldt Bay in northern California. In 2002, S. densiflora was sampled in the north San Francisco Bay along Corta Madera Creek and at four sites in Humboldt Bay. C. purpurea was not detected at either location (A. Fisher, unpubl.). Additional collections are needed to determine if the Argentina population is the result of an introduction or, perhaps more likely, if natural populations of G3 extend along the Atlantic coastline from North America through the South American coastal salt marshes. There is evidence of C. purpurea on Spartina in this range. For example, a 1943 herbarium specimen of C. purpurea collected in Uruguay from S. brasiliensis, a synonym of S. alterniflora, is in the US National Fungus Collection (BPI) (Duncan et al. 2002). The collection also contains specimens collected from S. alterniflora along the east coast of the USA from as early as the 1930s. Isolates collected from New York, Rhode Island, and New Jersey on S. alterniflora formed a major cluster using UPGMA with isolates on S. anglica from Ireland and the UK. However, AMOVA pairwise comparisons showed populations from these geographic regions to be significantly different (P=0.385). Interestingly, the origin of S. alterniflora in Europe is thought to be somewhere between Boston, MA and
Newfoundland, Canada (Hubbard 1965). As previously discussed, S. anglica is a hybrid between the native S. maritima and the introduced S. alterniflora. While isolates from Ireland clustered together with 100 % bootstrap support, isolate 538 from the UK was not part of that minor cluster. Therefore, it is unclear if G3 established in Western Europe following one, or multiple, introduction events. Our results support the recognition of three discrete groups within C. purpurea. Results of RAPD and AFLP analysis showed high intergroup variability between G1, G2, and G3. Using a collection of isolates from a different set of hosts, Jungehu¨lsing & Tudzynski (1997) also found high intraspecific variability using RAPD markers. In their study, parsimony analysis grouped samples from the same host species together, suggesting some degree of host specificity. More sampling is necessary to reveal the ecological and physiological range of G3, although there is evidence that the pathogenicity profile of G3, following greenhouse inoculations, is not limited to salt marsh species (Pazoutova´ et al. 2002a). Laboratory experiments notwithstanding, our results imply a degree of specificity in the ecological range of G3. Our results confirm the three intraspecific groups within C. purpurea and support from both RAPD and AFLP profiles strongly suggests a singular origin of G3 C. purpurea followed by widespread geographic dispersal.
ACKNOWLEDGEMENTS We thank Sylvie Pazoutova´ for providing Claviceps purpurea isolates; Alejandro Bartolis, Kennebla L. Heck, Dino Garcia-Rossi, Tracy Buck, Alexander Kolker, Steve Orloff, Michael Burzynski, Darrell
Claviceps purpurea in salt marsh habitats Wesenberg, and E. Landy for field samples; and B. Aegerter, Debra Ayres, Ravindra Bhat, Greg Douhan, and David Rizzo for technical assistance and use of laboratory equipment. The manuscript was improved by comments from Donald Strong and two anonymous reviewers. This study was financially supported by The Garden Club of America, Jastro Shields and Humanities awards from the University of California, Davis and NSF Biocomplexity DEB 0083583 to A.H.
REFERENCES Alderman, S. C., Halse, R. R. & White, J. F. (2004) A reevaluation of the host range and geographical distribution of Claviceps species in the United States. Plant Disease 88 : 63–81. Ayres, D. R., Smith, D. L., Zaremba, K., Klohr, S. & Strong, D. R. (2004) Spread of exotic cordgrasses and hybrids (Spartina spp.) in the tidal marshes of San Francisco Bay, CA, USA. Biological Invasions 6: 221–231. Ayres, D. R., Strong, D. R. & Baye, P. (2003) Spartina foliosa (Poaceae) – a common species on the road to rarity? Madron˜o 53: 209–213. Bassam, B. J., Caetano-Anolles, G. & Gresshoff, P. M. (1991) Fast and sensitive silver staining of DNA in polyacrylamide gels. Annals of Biochemistry 196 : 80–83. Bowden, B. N. (1965) Modern grass taxonomy. Outlook on Agriculture 4: 243–253. Boyle, P. J. (1976) Ergot epiphytotic on Spartina spp. in Ireland. Irish Journal of Agricultural Research 15 : 419–424. Daehler, C. C., Anttila, C. K., Ayres, D. R., Strong, D. R. & Bailey, J. P. (1999) Evolution of a new ecotype of Spartina alterniflora (Poaceae) in San Francisco Bay, California, USA. American Journal of Botany 86 : 543–546. Daehler, C. C. & Strong, D. R. (1994) Variable reproductive output among clones of Spartina alterniflora (Poaceae) invading San Francisco Bay, California: the influence of herbivory, pollination, and establishment site. American Journal of Botany 81: 307–313. Daehler, C. C. & Strong, D. R. (1997) Hybridization between introduced smooth cordgrass (Spartina alterniflora, Poaceae) and native California cordgrass (Spartina foliosa) in San Francisco Bay, California, USA. American Journal of Botany 84: 607–611. Davis, H. G., Taylor, C. M., Civille, J. C. & Strong, D. R. (2004) An allee effect at the front of a plant invasion: Spartina in a Pacific estuary. Journal of Ecology 92: 321–327. Douhan, G. W., Peever, T. L. & Murray, T. D. (2002) Multilocus population structure of Tapesia yallundae in Washington State. Molecular Ecology 11: 2229–2239. Duncan, R. A., Sullivan, R., Alderman, S. C., Spatafora, J. W. & White, J. F. (2002) Claviceps purpurea var. spartinae var. nov.: an ergot adapted to the aquatic environment. Mycotaxon 81: 11–25. Eleuterius, L. N. & Meyers, S. P. (1974) Claviceps purpurea on Spartina in coastal marshes. Mycologia 66: 979–986. Eleuterius, L. N. & Meyers, S. P. (1977) Alkaloids of Claviceps from Spartina. Mycologia 69 : 838–840. Excoffier, L., Smouse, P. E. & Quattro, J. M. (1992) Analysis of molecular variance inferred from metric distances among DNA haplotypes: applications to human mitochondrial DNA restriction data. Genetics 131: 479–491. Faber, P. (2000) Good intentions gone awry: why would anyone bring an alien cordgrass to San Francisco ? Coast and Ocean 16: 14–17.
446 Hitchcock, A. S. (1971) Manual of the Grasses of the United States. Dover, New York. Hubbard, J. C. E. (1965) Spartina marshes in southern England. VI. Pattern of invasion in Poole Harbour. Journal of Ecology 53: 799–813. Jungehu¨lsing, U. & Tudzynski, P. (1997) Analysis of genetic diversity in Claviceps purpurea by RAPD markers. Mycological Research 101: 1–6. Kobel, H. & Sanglier, J. J. (1978) Formation of ergotoxine alkaloids by fermentation and attempts to control their biosyntheses. FEMS Symposium 5: 233–242. Loveless, A. R. (1971) Conidial evidence for host restriction in Claviceps purpurea. Transactions of the British Mycological Society 56: 419–434. Mobberley, D. G. (1956) Taxonomy and distribution of the genus Spartina. Iowa State Journal of Science 30: 471–574. Pazoutova´, S., Cagas, B., Kolı´ nska´, R. & Honza´tko, A. (2002a) Host specialization of different populations of ergot fungus (Claviceps purpurea). Czech Journal of Genetics and Plant Breeding 38: 75–81. Pazoutova´, S., Olsovska, J., Linka, M., Kolı´ nska´, R. & Flieger, M. (2000) Chemoraces and habitat specialization of Claviceps purpurea populations. Applied and Environmental Microbiology 66: 5419–5425. Pazoutova´, S., Raybould, A. F., Honza´tko, A. & Kolı´ nska´, R. (2002b) Specialized populations of Claviceps purpurea from salt marsh Spartina species. Mycological Research 106: 210–214. Peterson, P. M., Soreng, R. J., Davidse, D., Filgueiras, T. S., Zuloaga, F. O. & Judziewicz, E. J. (2001) Catalogue of New World Grasses (Poaceae) : II. Subfamily Chloridoideae. Contributions of the US National Herbarium 41: 1–255. Raybould, A. F., Gray, A. J., Lawrence, M. J. & Marshall, D. F. (1991) The evolution of Spartina anglica C. E. Hubbard (Gramineae) ; the origin and genetic variability. Biological Journal of the Linnean Society 43: 111–126. Samuelson, R. J. & Gjerstad, G. (1966) Intermediary metabolism of ergot XIII. Feather ergot, a new, promising Claviceps species. Meddelelser Fra Norsk Farmaceutisk Selskap 28: 229–237. Schneider, S., Roessli, D. & Excoffier, L. (2000) Arlequin: A software for population genetics data analysis. Version 2.001. Genetics and Biometry Laboratory, University of Geneva. Sprague, R. (1950) Diseases of Cereals and Grasses in North America. Ronald Press: New York. Stager, R. (1922) Beitrag zur Verbreitungsbiologie der ClavicepsSklerotien. Zentralblatt Fur Bakteriologie Parasitenkunde Infefektionskranheiten Und Hygiene 56: 329–339. Stiller, J. W. & Denton, A. L. (1995) One hundred years of Spartina alterniflora (Poaceae) in Willapa Bay, Washington: random amplified polymorphic DNA analysis of an invasive population. Molecular Ecology 4: 355–363. Vos, P., Hogers, R., Bleeker, M., Reijans, M., Van De Lee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J., Kuiper, M. & Zabeau, M. (1995) AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 4407–4444. 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 (M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White, eds): 315–322. Academic Press, San Diego.
Corresponding Editor: J. K. Stone