Soil microbiota can colonise plant roots via a two-step selection process, which involves the recruitment of microbiota first from bulk soil into plant rhizospheres, then into root endospheres. This process is poorly understood in all but a few model species, which is surprising given its fundamental role in plant and soil ecology. Here we examined the microbial assembly processes across the rhizospheres and root endospheres in eight natural populations of the pan-palaeotropical C4 grass, Themeda triandra, in southern Australia. We assessed whether root endosphere colonisation patterns aligned with the two step-selection process. We also assessed the degree to which the assembly patterns of these rhizospheres and endospheres were influenced by deterministic processes. We show that two-step selection was the dominant recruitment dynamic across these natural T. triandra populations, and present clear evidence that host plants influenced microbial assembly via deterministic pressures that produced strong convergence of endospheres. Both endospheres and rhizospheres were influenced by local environmental filtering, including aridity. Our study improves our understanding of assembly processes of root endospheres, which is central to plant-soil interactions yet poorly understood in non-model species. We show that endospheres of native populations of a widely distributed, keystone grass (T. triandra) were strongly shaped by the host plant and displayed patterns consistent with the two-step selection process. These findings raise intriguing questions about the functions of this ‘core’ microbial endosphere, but our limited understanding of their ecology hinders our ability to harness these important relationships to, for example, improve plant propagation and revegetation practices.

Soil microbiota can colonise plant roots via a two-step selection process, which involves the recruitment of microbiota first from bulk soil into plant rhizospheres, then into root endospheres. This process is poorly understood in all but a few model species, which is surprising given its fundamental role in plant and soil ecology. Here we examined the microbial assembly processes across the rhizospheres and root endospheres in eight natural populations of the pan-palaeotropical C4 grass, Themeda triandra, in southern Australia. We assessed whether root endosphere colonisation patterns aligned with the two step-selection process. We also assessed the degree to which the assembly patterns of these rhizospheres and endospheres were influenced by deterministic processes. We show that two-step selection was the dominant recruitment dynamic across these natural T. triandra populations, and present clear evidence the host plants influenced microbial assembly via deterministic pressures that produced strong convergence of endospheres. Both endospheres and rhizospheres were influenced by local environmental filtering, including aridity. Our study improves our understanding of assembly processes for root endospheres, central to plant-soil interactions yet poorly understood in non-model species. We show that endospheres of native populations of a widely distributed, keystone grass (T. triandra) were strongly shaped by the plant host and displayed patterns consistent with the two-step selection process. These findings raise intriguing questions about the functions of this ‘core’ microbial endosphere, but our limited understanding of their ecology hinders our ability to harness these important relationships to, for example, improve plant propagation and revegetation practices.

A document by Riley Hodgson. Click on the document to view its contents.

Soil bacterial taxa have important functional roles in ecosystems (e.g., nutrient cycling, soil formation, plant health). Many factors influence their assembly and regulation, with land cover type (e.g., remnant vegetation, agriculture, urban parks) and plant-soil feedbacks being two well studied factors. However, changes in soil bacterial communities in situ over light-dark cycles have received little attention, despite plants and some bacteria having endogenous circadian rhythms that could influence soil bacterial communities. We sampled surface soils in situ across 24-hour light-dark cycles (at 00:00, 06:00, 12:00, 18:00) at two land cover types (remnant vegetation vs. cleared, grassy areas) and applied 16S rRNA amplicon sequencing to investigate changes in bacterial communities. We show that land cover type strongly affected soil bacterial diversity, with soils under native vegetation expressing 15.41-16.42% lower alpha diversity but 4.92-10.67% greater heterogeneity than soils under cleared vegetation. In addition, we report time-dependent and site-specific changes in bacterial network complexity and between 598-922 ASVs showing significant changes in relative abundance across times. Native site node degree (bacterial interactions) at phylum level was 16.0% higher in the early morning hours compared to the afternoon/evening. Our results demonstrate for the first time that light-dark cycles have subtle yet important effects on the composition of soil bacterial communities in situ and that land cover influences these dynamics. We provide a new view of soil microbial ecology and suggest that future studies should consider the time of day when sampling soil bacteria.

Toxoplasma infection in intermediate host species closely associates with inflammation. This association has led to suggestions that the behavioural changes associated with infection may be indirectly driven by the resulting sustained inflammation rather than a direct behavioural manipulation by the parasite. If this is correct, sustained inflammation in chronically infected rodents should present as widespread changes in the gastrointestinal microbiota due to the dependency between the composition of these microbiota and sustained inflammation. We conducted a randomized controlled experiment in rats that were assigned to a Toxoplasma-treatment, placebo-treatment or negative control group. We sacrificed rats during the chronic phase of infection, collected their cecal stool samples and sequenced the V3-V4 region of the 16S rRNA gene to characterise the bacterial community in these samples. Toxoplasma infection did not induce widespread changes in the bacterial community composition of the gastrointestinal tract of rats. Rather, we found sex differences in the bacterial community composition and only minor changes in Toxoplasma infected rats. We conclude that it is unlikely that sustained inflammation is the mechanism driving the highly specific behavioural changes observed in Toxoplasma-positive rats.