The role of soil biota in the invasion success of legumes in Australia

"It has been widely acknowledged that soil microbial communities may play a significant role in the processes that regulate successful invasion by exotic plant species. Species within the genus Acacia have been extensively introduced for horticultural and ornamental purposes into novel regions within Australia, and many of these have become invasive. These invasive acacias may have a significant negative effect on the co-occurring native flora and induce changes to belowground microbial composition. This thesis examines the role of soil biota in the invasion success of four Acacia species (A. cyclops, A. longifolia, A. melanoxylon, A. saligna) and their close relative Paraserianthes lophantha introduced into novel areas within Australia. Acacia cyclops, A. saligna and P. lophantha are native to Western Australia but have become invasive or naturalised in the eastern states, whereas A. longifolia and A. melanoxylon are, vice versa, native to the eastern states of Australia but have become invasive in Western Australia. For the purpose of this thesis a biogeographic approach was applied and soil and seed material was sampled for each species in both ranges (native and non-native) and from multiple populations and individuals within a population. The field collected material was then used for experiments described in Chapters 2, 3, 4 and 5. In Chapter 1 (Introduction) the published literature on the role of soil biota in the invasion success of plant species introduced to novel environments is discussed. In addition, the five study species and their invasion ecology, history and introduction pathways are described in the local as well as global context. In Chapter 2 the net role of soil microbial communities on the invasion success of the five legume study species was investigated. This Chapter describes the results from a common garden experiment that employs the plant-soil feedback approach. Plant-soil feedback experiments have been widely applied in invasion ecology to test for plant-soil reciprocal interactions. The results from this experiment showed that there was no significant effect of the soil origin on plant growth, however there was a significant effect of the seed origin on plant performance. This suggests that invasion success of these legumes in Australia is not limited by mutualistic soil biota or facilitated by the absence of soil pathogens in novel environments, but rather genetic adaptation to novel environments and human-mediated artificial selection could have influenced the invasion success of these acacias in Australia. In Chapter 3 multiple complementary approaches, including a glasshouse experiment to assess plant growth in non-native compared to native range soils, estimation of abundance of nitrogen fixing bacteria (i.e. rhizobia) and molecular analysis of the rhizobial community composition across both native and non-native ranges, were used to comprehensively assess the role of rhizobia in the invasion success of the five study species. Only one of the species (A. longifolia) had greater biomass when grown in its non-native range soils, rhizobial abundance was similar for all five study species across both ranges, and rhizobial community composition was significantly different between the ranges for western natives A. cyclops, A. saligna and P. lophantha but similar for eastern natives A. longifolia and A. melanoxylon. This study has revealed that overall invasive success of these five species is unlikely to be constrained by the absence of compatible rhizobia in novel ranges, although there appears to be variation in rhizobial communities across the ranges for some of the host species. In Chapters 4 and 5 the diversity of free-living and nodulating nitrogen fixing bacterial and fungal communities, respectively, associating with the five study species were investigated using next-generation sequencing. The results from Chapter 4 revealed that free-living nitrogen fixing bacterial communities in the soils were different across the continent, while similar across the ranges in the nodules of the host legumes. These results indicate that despite fundamental differences in the bacterial communities across the continent these legumes are unlikely to be constrained by the absence of compatible symbionts in the introduced range since they appear to associate with the same common subsets of bacteria in eastern and western populations. In Chapter 5 the fungal communities in the rhizospheres of the study species in their non-native compared to native ranges were described and analysed. Similarly to the results from Chapter 4, I found that the fungal communities in the rhizosphere of these legumes were different across the continent. Overall, these results indicate that it is unlikely that these legumes are constrained by novel fungal communities or have been released from harmful pathogens in their non-native ranges in Australia. Chapter 6 (Discussion) summarizes the original findings of this thesis and places them in the broader context of the plant invasion literature." -- Abstract.