SUMMARY

Phytophthora root and stem rot (PRR) of soybean, caused by , has been managed traditionally through breeding for resistance. Breeding efforts have focused on incorporating major Rps gene resistance into soybean lines but the pathogen is rapidly and continuously evolving, the complexity of pathotypes is increasing, and the durability of the Rps in the field is short.

It has been suggested that a large reservoir of genetic diversity exists in populations of , however, only a few studies have tried to characterize this diversity using genetic markers. In this study, we assessed diversity of in Iowa using microsatellite (SSR) markers (Objective 1). Forty MGLs (multilocus SSR genotypes) and 19 pathotypes were found, with a genotypic and pathotype diversity (D) of 0.96 and 0.86, respectively. Low observed heterozygosity (Ho= 0.027) and a high inbreeding coefficient (F=0.944) were found as expected in this homothallic species. Data suggest that multiple pathotypes and genotypes of may infect a single plant in the field. Furthermore, at least one of the seven populations of recovered from a commercial soybean field showed some level of outcrossing.

It has also been hypothesized that the deployment of specific resistant genes result in changes in the pathotype structure of the populations. An alternative way to prolong the life of a resistance gene in the field is to generate disruptive selection by rotating major resistant genes through time and space. To test this hypothesis, we monitored pathotype and genetic shifts in populations in a four year soybean cultivar rotation (Objective 2). We recovered new pathotypes and MGLs from rotation treatments at the second sampling date, indicating P. sojae has the ability to evolve quickly. We found cultivar rotation affected the genetic structure of the P. sojae population, but pathotype shift was not a function of cultivar rotation.

A third strategy used to control PRR is through the use of partial resistance (PR), which allows the host to tolerate infection by better when compared to a fully susceptible host. PR is also effective against all physiological races of the pathogen. Thus, tolerance should be more stable than single-gene resistance because of lack of selection pressure imposed on the pathogen. Few cultivars with high levels of PR are currently available, because this type of resistance is polygenic, and thus harder to recognize and introgress in a good agronomical type. Our goal was to develop an easier and more objective method that could be used by soybean breeders to screen for PR (Objective 3). The inoculation method we propose uses rice infested with as an inoculum source and the evaluation is done by dry root weight instead of the customary 1-10 scale used in the standard layer test. For the variables evaluated, the rice method did not differ statistically from the layer test, and it has the advantage that it is cheaper, set up is less time consuming, and more than one pathotype can be used in a single test ensuring that no Rps gene masks the partial resistance while screening. The six soybean cultivars used in the study ranked as expected for PR based on the published information and on the information supplied by company. Partial resistance ratings were negatively and significantly correlated to corrected dry root weight (r= -0.975, P < 0.0001).