The genetic structure of populations of Heterobasidion annosum (Fr.) Bref. from the global to the local scale: Implications for the biology, the epidemiology, and the evolution of a forest pathogen

Hetevobasidion annosum (Fr.) Bref is a pathogen of worldwide distribution found on many woody plants. It is a species complex which includes at least three partially intersterile groups (the S, P, and F ISGs), each with different host preferences. The genetic structure of H. annosum populations at varying geographic scales was studied with aid of molecular techniques. Four studies were based on field sampling, the remaining three on inoculation tests.;Results from the study of H. annosum populations at a broad geographic scale indicated that: (a) a combination of allopatric and sympatric speciation processes can explain the genetic structure of worldwide populations; (b) host specificity of Californian ISGs is a valid concept, but occasionally P-hosts such as pines and junipers may be infected by S isolates; (c) hybridization and intergroup gene flow are occurring between the S and P ISGs in California, presumably in response to changes caused by modern forest uses. Results at a finer geographic scale indicated that H. annosum ISG S is constituted of highly sexual and outcrossing individuals, becoming established mostly through basidiospore infections on live trees. Discrete white fir mortality centers are determined by many unrelated homokaryotic and heterokaryotic genotypes. Identical nuclei can be present in adjacent distinct genotypes, presumably via di-mon mating or by contemporary heterokaryotization of different portions of large homokaryotic isolates. Tree to tree spread is infrequent and more likely for larger trees and in denser stands. Results from a field inoculation study showed that homokaryons can persist and be virulent in nature, and that larger roots are better colonized by H. annosum than smaller roots. In a greenhouse inoculation study, a SP hybrid isolate was less virulent than either parental types on ISG-specific hosts, but was as virulent as the parental types on a universal, non-specific host. These results indicate that: (a) mechanisms of host-pathogen specificity can be better explained by the classical "avirulence model" than by the "pathogenicity model", and (b) selection imposed by the hosts may act as a reinforcement mechanism in a sympatric speciation process.