Umeå universitets logga

The dynamics of a plant-pathogen interaction vary both within and among species. Both spatial structuring and specific genetic and life-history characteristics will affect the interaction and the outcome of a potential co-evolution between the two organisms. In this thesis I have studied the interaction between the wild perennial herb Silene dioica and its automictic, obligate anther smut Microbotryum violaceum MvSd. From the plant perspective, I have examined different aspects of biochemical resistance in S. dioica to M. violaceum MvSd. From the pathogen perspective, I have focused on the breeding system of M. violaceum MvSd and its connection to fitness and distribution of genetic diversity. I have used varying methods; glasshouse trails involving inoculation of plants with the pathogen, classical Mendelian analysis involving controlled crosses between plants, microscopic studies of spores and molecular DNA-analysis.

With the results I demonstrate that resistance to M. violaceum MvSd in S. dioica can be specific to the attacking pathogen strain and also spatially highly diverse both within and among populations within a metapopulation. Together, these factors are likely to delay the establishment of the disease within host populations and reduce the spread and amount of disease, once it has been established. The results also suggest that the specific resistance expressed against two different M. violaceum MvSd strains were determined by separate gene systems and that, in both cases, the resistance was simply inherited. This implies a potential for relatively rapid response to M. violaceum-induced selection in S. dioica populations variable for resistance.

My results also show that automixis clearly is the predominating breeding system of M. violaceum MvSd, similarly to what earlier has been shown for M. violaceum MvSl. Furthermore, I found lower levels of neutral genetic diversity in M. violaceum MvSd in the northern parts of Sweden, compared to what has been found in populations in more southern Europe. This result is consistent with predictions that populations in the outer regions of a species distribution have lower levels of genetic variation. Moreover, populations were highly differentiated in northern Sweden, which could have been generated by high selfing rates, genetic drift and high population turnover rates, all factors that coincide with life-history and ecology of M. violaceum MvSd. However, despite the general low variability in neutral genetic markers, I did find variation in fitness related traits, both within and among populations, as well as differences in infection ability between strains, suggesting there is a potential for co-evolution between S. dioica and M. violaceum MvSd in the area.

To summarize, this thesis reflect a plant-pathogen system that is highly influenced by constant colonisation-extinction dynamics, which is likely to have influenced both the genetics of resistance in the plant and the breeding system of the pathogen and thus also the interaction between the two organisms.