Plant Invasions and Microbes: The Interactive Effects of Plant-Associated Microbes on Invasiveness of Phragmites australis

Abstract

The mechanisms driving biological invasions are important for predicting range expansion and developing effective invasive species management strategies but are often difficult to disentangle. One driver of plant invasions may be through differential interactions with belowground microbes, whereby invasive plants gain a disproportionate advantage over natives either through a relatively stronger interaction with mutualists or a weaker interaction with pathogens. I aimed to examine whether invasive Phragmites australis, a clonal wetland plant, gains a performance advantage over a related native lineage through interactions with belowground microbial communities. I explored bacterial, fungal, and oomycete communities associated with native and non-native Phragmites in the Great Lakes region and the impacts of those microbial communities on invasiveness. I used a combination of field surveys of natural populations and controlled environment experimental manipulations combined with next-generation sequencing to thoroughly examine whether invasiveness in Phragmites is facilitated by interactions wih belowground microbes and which microbial players are most influential. My results were very consistent among all chapters in this dissertation, finding no strong link between invasiveness and belowground microbial communities, and therefore suggesting that belowground microbes are not fostering invasion of Phragmites australis in the Great Lakes region. Field surveys provided evidence that belowground microbial communities did not differ between Phragmites lineages in roots or rhizospheres of natural populations. Root communities did differ in fungal colonization and in oomycete richness, but both of those differences were weak and inconsistent among different environmental conditions. In addition, the few differences that were found between lineages were consistently opposite of my expectation that non-native Phragmites would be associated with more mutualists and/or fewer pathogens than native. The rhizosphere largely followed the same patterns with one exception: the rhizosphere bacterial communities differed by lineage in large, dense patches of Phragmites, but not elsewhere. Given the small magnitude of the observed differences in bacterial communities, and the fact that they only existed in dense, mature patches of Phragmites, no differences in functional potential could be attributed to the community differences observed. Taken together, the evidence that I have obtained strongly suggests that any observed differences in soil microbial communities between Phragmites lineages may be a consequence rather than a driver of invasiveness. Consistent with results from natural populations, I also observed that experimentally-conditioned soils differed only slightly between lineages in bacterial community composition and even less so in fungal community composition. The plant response to those slightly different microbial communities was more significant, but again opposite of expectations if microbes were driving invasiveness. Non-native Phragmites was overall negatively impacted by the total soil microbiome, whereas native was unaffected by the total soil microbiome, regardless of which lineage conditioned the soil, with bacterial pathogens likely playing a significant role in the negative plant-soil interaction. These findings on lineage-specific plant responses are counter to our expectation that if microbial communities are driving invasiveness, non-native Phragmites should derive disproportionate benefits from microbial communities over native. Given the preponderance of data suggesting that belowground microbes are not drivers of invasiveness in Phragmites, it is reasonable to assume that the non-native’s invasiveness is primarily derived from other sources. However, importantly, differential response to similar microbes in native and non-native lineages suggests that microbial manipulation could be a reasonable tool for lineage-specific biocontrol.