Mutualistic interactions can strongly influence species invasions, as the inability to form successful mutualisms in an exotic range could hamper a host’s invasion success. This barrier to invasion may be overcome if an invader either forms novel mutualistic associations or finds and associates with familiar mutualists in the exotic range. Here we ask (1) does the community of rhizobial mutualists associated with invasive legumes in their exotic range overlap with that of local native legumes and (2) can any differences be explained by fundamental incompatibilities with particular rhizobial genotypes? To address these questions, we first characterized the rhizobial communities naturally associating with three invasive and six native legumes growing in the San Francisco Bay Area. We then conducted a greenhouse experiment to test whether the invasive legume could nodulate with any of a broad array of rhizobia found in their exotic range. There was little overlap between the Bradyrhizobium communities associated with wild-grown invasive and native legumes, yet the invasive legumes could nodulate with a broad range of rhizobial strains under greenhouse conditions. These observations suggest that under field conditions in their exotic range, these invasive legumes are not currently associating with the mutualists of local native legumes, despite their potential to form such associations. However, the promiscuity with which these invading legumes can form mutualistic associations could be an important factor early in the invasion process if mutualist scarcity limits range expansion. Overall, the observation that invasive legumes have a community of rhizobia distinct from that of native legumes despite their ability to associate with many rhizobial strains, challenges existing assumptions about how invading species obtain their mutualists. These results can therefore inform current and future efforts to prevent and remove invasive species.
La Pierre_invasive shrub_nodulation assay_dryad submission
Data addressing the breadth of rhizobial isolates that three invasive and two native legumes from the San Francisco Bay Area, California, USA could nodulate with in single-isolate inoculations under greenhouse conditions.
We assessed the promiscuity of the three invasive plants in their invasive range by determining their ability to associate with a broad range of 117 rhizobial isolates originally collected from 12 different leguminous hosts (both native and invasive, including hosts not studied here, but all growing in the SF Bay Area; Table S2) in a greenhouse-based nodulation assay. Seeds of each legume species were surface sterilized in full-strength commercial bleach (3% sodium hydroxide) for 30 sec, rinsed five times with sterile water, scarified with sulfuric acid for either 10 min (S. junceum), 30 min (U. europaeus), or 40 min (G. monspessulana), neutralized with a sterile 20% sodium bicarbonate solution and thoroughly rinsed using sterile water. Scarified seeds were germinated in the dark at room temperature in individual wells of 96-well plates filled with 100 µl sterile water. Two weeks later, germinated seedlings were individually planted into 22 mm diameter, 20 cm tall sterile glass 75 ml culture tubes filled with 25 ml sterile vermiculite moistened with sterile water. Tubes were plugged with sterile cotton and kept under shade cloth, which provided indirect natural light, and were provided supplemental artificial light in the Jane Gray Research Greenhouse at the University of California, Berkeley. Twelve days following planting, 1 ml sterile Jensen’s fertilizer (Somasegaran and Hoben 1994) containing 7 ppm nitrogen was added to each tube.
The 117 rhizobial isolates used in the nodulation assay were obtained from two sources: (1) many isolates were obtained from the collection described above prior to genotyping (99 isolates); (2) several isolates were obtained from the investigators’ additional research collections to represent strains associated with other native and invasive legumes common in the San Francisco Bay Area (18 isolates; see Table S2 for a list of the isolates used in the nodulation assay and their sources). Isolates were chosen to span a broad range of host species and collection sites. Inoculum from each isolate was prepared from 50 µl of -80°C glycerol stock prepared from field-collected nodules (see above), grown in YM broth at 25°C shaken at 120 rpm to a density of per ml, as measured by optical density at 600 nm. Each rhizobial isolate was inoculated onto one seedling of each legume species. Seedlings were randomly assigned rhizobial isolates and inoculated 17 days after planting by adding 1 ml of the appropriate inoculum to the base of the plant stem in each tube. An additional ten plants per legume species were inoculated with sterile YM broth as negative controls; none of the control plants were nodulated at harvest. Plants were harvested 47 days after planting (30 days after inoculation), the roots thoroughly washed, and the presence of nodules recorded. Successful association was defined as the formation of at least one robust nodule that appeared to be effectual (i.e., not <1 mm and/or white or clear). Reanalysis of our results increasing the cutoff for defining successful nodulation to two nodules did not qualitatively alter our findings.