NONRANDOM GENOTYPIC ASSOCIATIONS IN A LEGUME— BRADYRHIZOBIUM MUTUALISM

Abstract

Abstract.-Genetically divergent lineages often coexist within populations of the annual legume Amphicarpaea braeteata. At one site dominated by two such lineages (termed biotypes "C" and "S"), isolates of root-nodule bacteria (Bradyrhizobium sp.) were sampled from both hosts and analyzed by enzyme electrophoresis. Symbiont populations on the two plant biotypes were highly distinct. Out of 15 bacterial multilocus genotypes detected (among 51 isolates analyzed), only one was shared in common by the two plant biotypes. Cluster analysis revealed three bacterial lineages (designated I, II, and III), with lineage I found exclusively on biotype C plants, and the two other lineages almost completely restricted to biotype S hosts. Laboratory inoculation tests indicated that lineage I bacteria were strictly specialized on biotype C hosts, forming few or no nodules on plants of the other host biotype. Bacterial lineages II and III were capable of forming nodules on both kinds of plants, but nodule numbers were often significantly higher on biotype S hosts. The nonrandom association between plant and bacterial lineages at this site implies that genetic diversity of hosts is an important factor in the maintenance of polymorphism within the symbiont population.Key words.-Bacteria, Bradyrhizobium, coevolution, legumes, linkage disequilibrium, mutualism, nitrogen fixation, polymorphism, symbiosis.Received April 12, 1994. Accepted September 21, 1994 Although it is widely accepted that host-parasite interactions should generate genetic polymorphism (Haldane 1949;Clarke 1976;Hamilton 1982), the situation is much less clear in mutualistic interactions among species. In host-parasite interactions, parasite specialization on hosts with a particular defense mechanism can result in frequency-dependent disease impact on hosts, leading to stable or cyclic polymorphisms in both species (Lewis 1981;Seger 1988). Law (1985) argued that quite different evolutionary processes should occur in mutualisms. Selection on each species in a mutualism will favor adaptation to the most prevalent phenotype in the other species, because individuals adapted only to rare phenotypes will receive fewer mutualistic benefits given their smaller chance of encountering a suitable partner. As a result, Law suggested that the evolutionary trend toward genotypic specialization thought to typify host-parasite coevolution should generally be lacking in mutualisms. Because polymorphism due to frequency-dependent disease impact requires parasite specialization (Lewis 1981;Seger 1988; Parker 1992a), the absence of such specialization in mutualisms would also imply the absence of any inherent tendency toward genetic polymorphism.However, this argument ignores the fact that different mutualist partners may vary widely in the fitness benefits that they can provide, with the optimal partner unlikely to be identical for all conspecific individuals. Thus, selection in mutualisms may favor intrapopulation specialization, if mechanisms exist to exclude incompatible partners and...

Divergence in Symbiotic Compatibility in a Legume-Bradyrhizobium Mutualism

“…In cases where a site was represented in more than one experiment (Table 1), different parents furnished seeds in each experiment. In a previous study of one site, symbiotic phenotypes were found to be highly consistent for different plant families in a given lineage (Spoerke et al 1996), which is not surprising in view of the limited withinlineage genetic diversity in local A. bracteata populations (Parker 1988(Parker , 1991.…”

“…Because reproduction occurs predominantly with obligately self-fertilized cleistogamous flowers, A. bracteata populations tend to be highly inbred (Parker 1988). Root nodule bacteria associated with A. bracteata are in the genus Bradyrhizobium, but appear to be distinct from the two species of Bradyrhizobium described to date (Spoerke et al 1996). Direct hereditary transmission of symbionts from parent plants to their offspring does not occur in the legume-rhizobial mutualism (Sprent and de Faria 1988).…”

“…Bacteria were sampled by collecting a single nodule from each of several individual plants at each site. A single bacterial isolate per nodule was purified as described in Spoerke et al (1996). The isolates used for experiments were selected randomly from the several available for each site.…”

“…Seedlings were replanted in a 1:2:3:3 mix of perlite, potting soil, milled peat, and sand in 8 X 6 X 5 em containers. At 24 h after potting, each plant was inoculated with 10 9 cells of a particular bacterial isolate grown in yeast-mannitol broth (Spoerke et al 1996).…”

Microevolution of symbiotic Bradyrhizobium populations associated with soybeans in east North America

“…Support for this proposition was provided by our finding that A. bracteata and D. canadense were associated with distinct core, and to some extent, symbiotic gene lineages of soybean-nodulating bacteria. Moreover, Spoerke et al (1996) reported that distinct sub-populations of A. bracteata in a natural woodland environment were associated with different lineages of Bradyrhizobium defined on the basis of enzyme electrophoresis. Further investigation is obviously needed to elucidate the role of native legume biogeography (at local and regional scales) with regard to the spatial structuring of Bradyrhizobium populations associated with crop legumes.…”