INRA Station de Pathologie Végétale, Domaine de la Motte, BP 29, F-35653 Le Rheu Cedex, France

Background and objectives
Phytophthora infestans. the cause of late blight of Solanaceae. is a major problem for potato and tomato production world-wide. Designing effective, yet environment- and consumer-friendlv control strategies requires a better understanding of the diversity and evolutionary patterns of the pathogen. Preliminary analyses of the diversity of P. infestans on potato and tomato in France showed that the nature and frequencies of the genotypes of the pathogen markedly differed according to the host plant [1]. These data suggested a major influence of the host on population structures. The objectives of the present work were therefore (1) to characterize P. infestans genotypes from both hosts, including an extended collection of isolates from tomato, and (2) to test the hypothesis of a preferential adaptation of the isolates to the host from which they had been isolated.

Materials and methods
A collection of 72 isolates collected in France in 1995-96 (40 from potato and 32 from tomato) was characterized for mating type, isozyme genotypes at the Gpi(glucose phosphate isomerase 1. EC and Pep (peptidase 1. EC loci. Virulence to the differential set of potato clones (genes R1 to R11, except R9). mitochondrial DNA haplotypes and total genomic DNA RFLPs after hybridization with the moderately repetitive probe RG57 cold-labelled with digoxygenin [2].

The aggressiveness of 43 isolates (27 from potato and 16 from tomato) was measured on detached leaflets of tomato cv. Marmande and/or potato cv. Bintje, taken from plants grown in the greenhouse for 6-8 ;weeks. Leaflets were infected with a drop of a sporangial suspension ( 104 spores/ml). and the leaflet area diseased was assessed daily until complete colonization of the most susceptible leaflets. The competitive ability of two isolates, one from potato and one from tomato, was measured in field plots of the susceptible potato cv. Bintje. Among other traits. these isolates differed by their peptidase genotype (83/100 for the potato isolate and 100/100 for the tomato one). Infector plants (cv. Bintje) were grown in pots and inoculated in the greenhouse with a sporangial suspension of one or the other isolate. Plots (12 rows of 20 tubers) were infected by depositing sporulating plants in two or four points (one or two points per isolate). Epidemic spread was monitored by scoring each plant for disease severity in the plot twice a week for two weeks. Isolates recovered from each plant newly infected were assigned to the 'potato' or the 'tomato' types after electrophoresis of peptidases.

Results and conclusions
Isolates collected on potato were all of the A1 mating type. Most of them belonged to a single clone. characterized by allozyme genotype Gpi90/100 ;Pep83/100, mitochondrial DNA haplotype Ia and a single RFLP profile; most if not all of these isolates belonged to complex pathotypes, the most frequent being Conversely, isolates from tomato included both mating types and frequently belonged to allozyme genotype Gpi100/100 ;Pep 100/100- they included mtDNA haplotypes Ia and IIa, and usually consisted of simple pathotypes (races 1.4 and No single lineage predominated in that group, based on RFLP fingerprints. A few genotypes were found on both hosts but at very different frequencies, These data support the hypothesis of the existence of two, host-related. populations of P. infestans in France with limited gene flow between them. Aggressiveness measurements on detached leaflets indicated that all isolates were pathogenic on both potato and tomato, but were most pathogenic on the host from which they had been isolated. In field plots, only the potato isolate spread , although the tomato isolate was successfully recovered from the original infector plants. Preferential adaptation to one or the other host may thus be a major factor limiting gene flow between the two groups of isolates of P. infestans coexisting in France.

1. Lebreton L, Duvauchelle S, Andrivon D, 1996. Abstracts, 13th Triennial Conference EAPR, pp. 262-263.
2. Pipe ND, Shaw DS, 1997. Molecular Plant Patholology On Line. http://www.bspp.org.uk/mmppol/1997/1124pipe