Pythium phragmitis sp. nov., a new species close to P. arrhenomanes as a pathogen of common reed (Phragmites australis) During a study on the occurrence and pathogenicity of oomycetes in the reed-belt (Phragmites australis) of Lake Constance (Germany), a new Pythium resembling the important cereal pathogen species complex P. arrhenomanes/P. graminicola was consistently isolated from necrotic mature reed leaves and reed rhizosphere samples. The new species proved to be significantly more aggressive towards reed leaves and seedlings in vitro than related species. It is characterised by filamentous, inflated sporangia and plerotic oospores with usually more than one antheridium. ITS and cox II sequence data indicate this new species shares a common ancestor with P. arrhenomanes, but the sequence differences are clearly consistent with a divergence of the two taxa and with P. phragmitis being a distinct species. ITS 1 and 2 of 15 isolates of the taxon consistently differed from P. arrhenomanes by 13 positions. Sequence analyses of the cox II gene confirmed the new species' phylogenetic position. This paper gives a formal description of the taxon as P. phragmitis sp. nov., providing information on morphology, ecology and pathogenicity in comparison to related species. As indicated by the close association to Phragmites australis, the high aggressiveness towards reed leaves and seedlings, and the abundance in the investigated stands, Pythium phragmitis might act as a reed pathogen of considerable importance, in particular under flooding situations.

Natural ecosystem Phytophthora Pythium Reed dieback a b s t r a c tWe studied the diversity of oomycetes in a declining reed belt (Phragmites australis) of Lake Constance, Germany, using conventional baiting with specific reed and standard oak baits, accompanied by molecular techniques. Apart from an Aphanomyces sp. and a Phytophthora sp., baiting from reed rhizosphere samples from flooded, as well as drier, littoral sites revealed only Pythium spp. A total of 67 oomycete isolates was classified according to PCR-RFLP banding patterns and ITS sequencing, and 18 different sequence types could be separated. The majority of these seemed previously unknown species, as indicated by the degree of similarity to those deposited in nucleotide databases. Species communities in both flooded and drier habitats or both reed and oak baits clearly differed from one another, and only few species occurred in both dry and flooded sites, or in both oak and reed baits. A frequently occurring group of related Pythium species appeared to be specifically associated with reed, and these were the only species that proved pathogenic towards this host in vitro. Our study proved that unexplored natural ecosystems harbour diverse communities of oomycete species with specific habitat and host preferences within close-by, but ecologically contrasting, sites. Among the species isolated, those associated with the predominating plant might accumulate and thus may be reed pathogens of considerable importance.

Pythium species are economically significant soilborne plant pathogens with worldwide distribution, causing seedling damping-off or root rot diseases. Pythium phragmitis is a newly described pathogen of common reed (Phragmites australis), widespread in the reed-belt of Lake Constance, Germany. It is highly aggressive towards reed leaves and seedlings, but obviously does not affect roots. In the context of 'reed decline' phenomena, P. phragmitis infection of reed inundated during flooding events may be of particular significance. We could show that flooding itself is not necessarily detrimental for reed plants. In the presence of the pathogen, however, most submerged leaves and plants were killed within several weeks. Clipped plants did not show regrowth in the Pythium infested treatments. Significant losses in assimilating leaf area of reeds could, thus, be the result of Pythium infection rather than of flooding alone. Therefore, we suggest that the combination of extended flooding and the presence of P. phragmitis might considerably contribute to 'reed decline' at Lake Constance. In parallel, we could show that pathogenicity and spread of this species are considerably favoured by rising temperatures. Since an increase in average water temperature has been found for Lake Constance, we propose that P. phragmitis could be an important factor in the dieback of reed stands likely to be promoted by predicted climate change phenomena.

In a 3-year-study, we analysed the population dynamics of the reed pathogen Pythium phragmitis and other reed-associated oomycetes colonizing fresh and dried reed leaves in the littoral zone of a large lake. Oomycete communities derived from internal transcribed spacer clone libraries were clearly differentiated according to substrate and seasonal influences. In fresh leaves, diverse communities consisting of P. phragmitis and other reed-associated pathogens were generally dominant. Pythium phragmitis populations peaked in spring with the emergence of young reed shoots, and in autumn after extreme flooding events. In summer it decreased with falling water levels, changing water chemistry and rising temperatures. Another Pythium species was also highly abundant in fresh leaves throughout the year and might represent a new, as-yet uncultured reed pathogen. In dried leaves, reed pathogens were rarely detected, whereas saprophytic species occurred abundantly during all seasons. Saprophyte communities were less diverse, less temperature sensitive and independent of reed development. In general, our results provide evidence for the occurrence of highly specialized sets of reed-associated oomycetes in a natural reed ecosystem. Quantitative analyses (clone abundances and quantitative real-time PCR) revealed that the reed pathogen P. phragmitis is particularly affected by changing water levels, water chemistry and the stage of reed development.

No abstract