To test the hypothesis that host-related differences in the genotypic composition of populations of the late blight pathogen Phytophthora infestans can be explained by differential pathogenicity, the aggressiveness of isolates of the pathogen collected in France from potato and tomato was measured on detached leaflets of potato (cv. Bintje) and tomato (cv. Marmande). A preliminary trial with four isolates (two each from potato and tomato) showed that lesion appearance and development were similar for each isolate in detached leaflets and in whole plant tests in growth cabinets. Isolates collected from tomato were more pathogenic to tomato than isolates collected from potato. This was particularly the case for isolates belonging to the A2 mating type. Isolates originating from potato had a higher infection efficiency and a higher sporulation capacity on this host, but they induced lesions that generally spread more slowly than those caused by isolates from tomato. Extensive variation for components of aggressiveness on potato, and to a lesser extent on tomato, was observed in collections of isolates from each of the two hosts. Competition experiments between one potato isolate and one tomato isolate in field plots of the susceptible potato cv. Bintje clearly demonstrated the higher competitive fitness of the potato isolate on its host of origin. Therefore, differential pathogenicity to potato and tomato certainly contributes to the differentiation between P. infestans populations present on potato and tomato in France; however, additional factors, possibly related to survival ability or random genetic drift, are probably also involved and may explain the persistence of weakly pathogenic isolates in these populations.

A method was developed to assess the genetic structure of Gaeumannomyces graminis var. tritici (Ggt) populations and test the hypothesis of an association between disease level in the field with changes in pathogen populations. A long-term wheat monoculture experiment, established since 1994, generated different take-all epidemics with varying the number of wheat crop successions in the 1999-2000 cropping season. Genetic polymorphism in Ggt populations was investigated over natural, local epidemics. Four populations of 30 isolates were isolated from necrotic wheat roots in a first, third, fourth, and sixth wheat crop in the same year. Each Ggt isolate was characterized with RAPD (Random Amplification Polymorphism DNA) markers and AFLP (Amplified Fragment Length Polymorphism) fingerprinting. Seventeen multilocus genotypes based on the combination of RAPD and AFLP markers were identified among all these populations. The 120 isolates were divided into two main groups, G1 and G2, according to bootstrap values higher than 86%, except for an unique isolate from the third wheat crop. Within each group, populations ranged between 93 and 100% similarity. Both groups included isolates collected from the first, third, fourth or sixth wheat crop. However, G1 group profiles dominated amongst isolates sampled in the first and the sixth wheat crops, whereas G2 group profiles largely dominated amongst isolates collected from the third and fourth wheat crops. Aggressiveness of group G2 (38%) was significantly greater than that of G1 (29.5%). These results suggest that changes in Ggt population structure occur during continuous wheat cropping. The distinction of two Ggt groups provides a simple basis for further spatio-temporal analysis of Ggt population during polyetic take-all decline.

The temporal dynamics of rhizosphere and root microbiota composition was compared between healthy and infected Chinese cabbage plants by the pathogen Plasmodiophora brassicae . When inoculated with P . brassicae , disease was measured at five sampling dates from early root hair infection to late gall development. The first symptoms of clubroot disease appeared 14 days after inoculation (DAI) and increased drastically between 14 and 35 DAI. The structure of microbial communities associated to rhizosphere soil and root from healthy and inoculated plants was characterized through high-throughput DNA sequencing of bacterial (16S) and fungal (18S) molecular markers and compared at each sampling date. In healthy plants, Proteobacteria and Bacteroidetes bacterial phyla dominated the rhizosphere and root microbiota of Chinese cabbage. Rhizosphere bacterial communities contained higher abundances of Actinobacteria and Firmicutes compared to the roots. Moreover, a drastic shift of fungal communities of healthy plants occurred between the two last sampling dates, especially in plant roots, where most of Ascomycota fungi dominated until they were replaced by a fungus assigned to the Chytridiomycota phylum. Parasitic invasion by P . brassicae disrupted the rhizosphere and root-associated community assembly at a late step during the root secondary cortical infection stage of clubroot disease. At this stage, Flavisolibacter and Streptomyces in the rhizosphere, and Bacillus in the roots, were drastically less abundant upon parasite invasion. Rhizosphere of plants colonized by P . brassicae was significantly more invaded by the Chytridiomycota fungus, which could reflect a mutualistic relationship in this compartment between these two microorganisms.

Phytophthora infestans isolates, collected in the two most important potato production regions of France (Brittany during 1992–96 and northern France during 1994–96), were analysed for mating type, Gpi and Pep allozyme genotypes, mitochondrial DNA haplotypes, response to metalaxyl and virulence patterns in order to determine the structure of populations. ‘New’ populations, similar to those observed in northern Europe and characterized by the presence of both mating types, Gpi 90/100 Pep 83/100, Gpi 90/100 Pep 100/100, Gpi 100/100 Pep 100/100 and Gpi 100/100 Pep 83/100 allozyme genotypes, Ia and IIa mitochondrial haplotypes and common races 1.3.4.7, 1.4.10.11, 1.3.4.7.11 and 1.3.4.7.10.11 were present in both areas. A new allozyme genotype (Gpi 90/90 Pep 83/100), never previously detected in Europe, was discovered in Brittany in 1992, but disappeared in later years. Despite the appearance of the A2 mating type in Brittany in 1996 and in northern France since 1995, both populations were highly homogeneous and probably consisted of a few related clones. No evidence for sexual recombination was obtained. Although Brittany, isolated from other potato production areas, and northern France, situated close to north European production areas where the A2 mating type was discovered some years ago, are two contrasted regions, no geographic substructuring was apparent between the two regions. However, variations in genotypic frequencies and in race composition were observed from one year to the next at a single site, as well as between several local sites during the same year. The role of possible evolutionary mechanisms implicated in these evolutions, especially founder effects, is discussed.