Intraspecific variability of C. purpurea 

Sylvie Pazoutova

C. purpurea - wide host spectrum, wide distribution, variable morphology

Claviceps purpurea inhabits Northern temperate regions but it can be found also in subtropic regions. The fungus colonizes mainly grasses of pooid and arundinoid subfamily, but is occassionally found also on chloridoid and panicoid grasses, even on sedges. During 19th and 20th century, many authors attempted to find  formae speciales in C. purpurea based on the host specificity, morphology and size of conidia, morphology of the sclerotium and stroma, even on the type of alkaloid produced (chemoraces).

The problem was, that there was too many exceptions and the sub-groups introduced by different authors were not the same (Campbell 1957; Tanda 1979). The experiments with artificial isolation have shown that all C. purpurea isolates could attack any pooid/arundinoid host provided that sufficiently aggressive method has been used. In our laboratory, we decided to combine all approaches used so far to enable the comparison with the groups postulated in older literature.

We collected over 200 isolates that were defined by:

1. RAPD and AFLP patterns

Examples of RAPD patterns of representative isolates using primer 257

RAPD patterns of all isolates have shared only one band, the species-specific amplicon A, derived from a multicopy sequence. 3' part of this sequence contains a fragment of a pol pseudogene of the Purcopia retroelement. Three clearly separated groups (G1, G2, G3) were found but inside these three groups, no other statistically supported clades were detected. The AFLP fingerprinting gave the same results - three groups with no internal population structure.

2. Eco RI polymorphism

In the sequence of 5.8S rDNA of C. purpurea from selected G1 and G3 isolates, we found a conserved EcoRI site common in Ascomycetes. However, in the sequence from the isolate 162 from G2 group, the site was missing due to transversion that changed GAATTC to GACTTC. The sequence from our isolate lacking the EcoRI site was the same as the sequence (GenBank, accession No. U57669) from C. purpurea isolate collected on Dactylis glomerata (Athens, GA.).

We analyzed all our isolates by PCR amplification and EcoRI digestion of this region. All G1 and G3 turned out to be EcoRI+,  whereas all G2 were EcoRI-. The representative isolates from Secale-Agropyron  group of Jungehülsing and Tudzynski (1997) were EcoRI+, whereas isolates from Molinia-Dactylis group were EcoRI-.

3. Host plants

The table shows, that although some species are preferentially infected (like Lolium and rye with Eco RI+, Molinia and P. annua with Eco RI-), other species are infected by both C. purpurea subgroups. The host preferences, if any, are not strict.


G1, G2


Bromus inermis

Alopecurus pratensis

Alopecurus arundinaceus

Carex sp. 

Ammophila arenaria

Alopecurus myosuroides

Elymus caninus

Arrhenatherum elatior

Agrostis sp.

Elytrigia repens

Dactylis sp.

Calamagrostis sp.

Festuca spp.

Festuca ovina

Glyceria fluitans

Hel. pubescens

Festuca rubra

Hordeum violaceum

Lolium spp. 

Phleum sp.

Milium effusum

Phalaris tuberosa

Poa pratensis

Molinia coerulea

Poa spp.

Phalaroides  arundinacea

Secale cereale

Phragmites communis

Sesleria tatrae

Poa annua

Spartina fusiformis 

Hordeum violaceum

Triticum aestivum


4. Habitats

Therefore we turned our attention to the habitats, where the isolates were found.  G1 isolates were found on sunny and/or dry places, like  fields, open meadows, and along the roads. G2 isolates were found on grasses growing near ponds, rivers, and at the forrest edges.G3 isolates were found exclusively in salt marshes on Spartina alterniflora and S. anglica. Recent results on the distribution of G1-G3 from the United States obtained by Alison Fisher (Paper 1, Paper 2) confirmed this Spartina preference, with Distichlis spicata as the only non-Spartina host.

Usually, from the same locality, only one group was isolated from different host species, except the rare cases of habitats bordering. Other places, where we found both G1 and G2 groups regularly, were the grassland station and the nursery for the collection of Czech grass ecotypes, where  steady introduction of seeds and plants of different origin occured. From more than 200 isolates, only thrice were G1 and G2 found in the same natural habitat.

5. Alkaloids

The isolates of the G1 group were very variable in the type of alkaloid produced. We found (using HPLC) ergotamine strains,  ergocornine/ergocristine strains and ergotamine/ergocristine strains, but also combinations of all the above. No further subgroups based on alkaloid type and/or RAPD pattern were found inside G1.

Sclerotia of  G2 group contained  the mixture of ergosine/ergocristine roughly 1:1, with 5-15 % of ergocryptine. No other alkaloids were found.

G3 isolates from Spartina  anglica produced mixture of ergocristine and ergocryptine.

G2 and G3 also represent chemoraces.

6. Conidia size

Conidia of the G1 isolates  were smaller than those of G2 and G3 strains. However, the overlapping region between 6.5 - 7.5 microns makes the conidia morphology only a supporting marker.

6. Floating of sclerotia

 Stäger (1922) denoted one of his formae speciales as "f.sp. Phalaridis arundinaceae natans", meaning that the sclerotia floated on water. As the host species of his f. sp. were those found for our G2, we repeated his experiment and indeed, the sclerotia of G2 and G3 strains floated for several hours, even days, whereas the sclerotia of G1 roup went down in 30 minutes.
The shape and size of ergot sclerotia are  influenced by the shape of the inflorescence of the host plant and also by the humidity of the season and the position on the flower head. However, G2 and G3 isolates in general have slim sclerotia, often curved and slightly angular.
Sclerotia of G2 and G3 strains
Sclerotia of G1 strains



C. purpurea species is divided to at least three groups, that can be discriminated reliably only on the basis of RAPD markers and the polymorphism of  ITS-5.8S rDNA region. However, the biotope, sclerotial shape, conidial size and the type of alkaloids produced taken together, are good clues what to expect after molecular analysis.

Host specificity of C. purpurea does not exist which proved Campbell's experiments as well as the existence of common hosts for G1 and G2 group. The "preferred hosts" were actually the grass species preferring given habitat. The common hosts are grass species, like Poa pratensis or Dactylis glomerata do not have specific habitat preferences and are encountered on the open dry meadows as well as in the woods.

G2 and G3 probably developed from a single clone and adapted itself to the life in more extreme habitats, as are river banks and salt marshes. Their clonal origin is supported by the fact taht they are also chemoraces. Main survival treat is the peculiar feature of  floating sclerotia, that might prevent the loss of sclerotia fallen in water and at the same time enable their spreading by water current.

G3 was described as a new variety by Duncan et al. (2002).

Baldacci, E. and Forlani, R. (1950) Ricerche sulla spezializzazione della Claviceps purpurea.
Phytopathol. Zschr. 17, 81-84.
Campbell, W.P. (1957) Studies of ergot infection in gramineous hosts. Can. J. Bot. 35, 315-320.
Mastenbroek, C. and Oort, A.J.P. (1941) Het voorkomen van moederkoren (Claviceps) oop granen en grassen en de specialisatie van de moederkorenschimmel. Tijdschr. Plziekt. 47, 165-185.
Stäger R. (1922) Beiträge zur Verbreitungsbiologie der Claviceps-Sklerotien. Zbl. Bakt. Parasitenk. Infektionsk. und Hygiene, Abt. II 56: 329-339

Duncan, Russell A., Jr., Raymond Sullivan, Stephen C. Alderman, Joseph W. Spatafora & James F. White, Jr. Claviceps purpurea var. spartinae var. nov.: an ergot adapted to the aquatic environment. Mycotaxon 81: 11-25. 2002.
Tanda, S. (1979) Mycological studies on the ergot in Japan (6) A physiologic race of C. purpurea Tul. var. alopecuri Tanda collected from Trisetum bifidum Ohwi. Journal of Agric. Sci., Tokyo Nogyo Daigaku 23, 207-214.

Back to Sylvie Pazoutova Homepage