The spatial and temporal analysis of Phytophthora infestans genetic diversity and its influence on late blight epidemics at a regional scale

The temporal and spatial population genetics of Phytophthora infestans, the causal agent of the potato and tomato late blight, was analyzed in a mixed potato and tomato production area in the Del Fuerte Valley, Sin., Mexico. Isolates of P. infestans were characterized by mating type, allozyme analysis at the glucose-6-phosphate isomerase (GPI) and peptidase (PEP) loci, restriction fragment length polymorphism (RFLP) with probe RG57, metalaxyl sensitivity, and aggressiveness to tomato and potato. Spatial patterns of P. infestans genotypes were analyzed using Geographical Information Systems (GIS) and Geostatistics during the seasons 1994-1995 to 1996-1997. A temporal analysis of the P. infestans genetic structure from 1990-1991 to 1996-1997 suggests an asexual or clonal population with frequent introductions from outside the valley. In the period from 1990-1991 to 1994-1995, the A2 mating type was predominant in both tomato and potato crops, with a very low frequency of the A1 mating type occurring either on tomato or on potato. Conversely, by the 1995-1996 season the predominant mating type was the A1, with low frequency of the A2 on tomato. By 1996-1997 only the A1 mating type was found. This suggests sexual reproduction is unlikely to be occurring in this area. Genotype variation, based on mating type, allozymes, and RFLP was, in general, very low with one predominant genotype affecting both crops each year. These predominant genotypes were highly aggressive to both tomato and potato in an in vitro detached leaf aggressiveness test. Other genotypes found on either potato or tomato, but not on both hosts, were non-aggressive to either tomato or potato. Data on metalaxyl sensitivity indicates that allozyme analysis can accurately discriminate between sensitive and resistant isolates. RFLP analysis showed that, in 1995-1996, there was greater diversity than could be determined by allozyme analysis alone. Spatial analysis of the genetic structure of P. infestans indicates that geographic substructuring of this pathogen does occur in this area. Maps displaying the probabilities of occurrence of mating types and genotypes of P. infestans, and of disease severity in a regional scale were obtained. Some genotypes, which exhibited differences in epidemiologically important features such as metalaxyl sensitivity and aggressiveness to tomato and potato, had a restricted spread and were localized in separated areas. Analysis of late blight severity demonstrates recurrent patterns such as the early onset of the disease in the area where both potato and tomato are growing, strengthening the hypothesis that infected potato tubers are the main source of primary inoculum. The information that geostatistics can provide together with the power of GIS and molecular biology techniques can help improve management programs for late blight in the Del Fuerte Valley