Influence of Grassland Management and Herbivory on Diversity and Ecology of plant-associated Bacterial Communities

Abstract

In the last years, plant-associated bacterial communities caught the attention of investigators due to their importance for both plant health and the environmental balance. Despite the increasing number of studies, there is still a knowledge gap about the influence of management regimes on the diversity of plant-associated bacteria in grassland ecosystems. In this study, we gained new and interesting insights into the diversity of plant-associated bacteria in grassland ecosystems. All investigations in this study were carried out in the same area, the GrassMan experimental field in the Solling Uplands, central Germany. The GrassMan project was set up in 2008. It was conducted in a matrix of meadow plots at a permanent grassland site. The full-factorial design of GrassMan included two mowing frequencies (mowing once per year in July vs. mowing thrice per year in May, July, and September) and two fertilization treatments (no vs. fertilization with NPK). A third factor, the gradient of species richness, was manipulated by selective herbicide applications targeting either dicots or monocots. The first aim of this study was to investigate the effect of the different mowing and fertilization regimes onto the bacterial endophytic community in three grass species, Festuca rubra, Lolium perenne, and Dactylis glomerata, respectively. Therefore, tiller samples were taken from the dicot-reduced plots in September 2010 as well as in April, July, and September 2011. Total DNA was extracted from the collected samples and subjected to 16S rRNA gene PCRs. Community structures were assessed by DGGE-based analysis of the generated PCR products. We found differences in bacterial endophyte community structures with respect to the grassland management regimes investigated. While fertilizer application had a high impact onto endophytic diversity in both F. rubra and L. perenne, the endophytic community structure in D. glomerata was not influenced by this management regime. Moreover, tillers of L. perenne derived from unfertilized plots grouped in distinct clusters indicating a more similar bacterial community composition in these plots when analyzing for the influence of the mowing treatment. We also recorded a strong seasonal effect on community composition. As a consequence, both the season and the host plant have to be regarded in further studies as they might alter the effects of different grassland management regimes on endophytic bacterial community structures. The second aim of this study was to investigate the effect of above-ground herbivory on the bacterial community composition in the rhizosphere. Therefore, a lysimeter experiment was established in autumn 2010. Following a two-week exposure to herbivory by grasshoppers and snails, soil samples were collected from the lysimeters in summer 2011. To gain insights into the composition of the plant-associated bacterial communities in the rhizosphere, total DNA was extracted from the collected samples and subjected to 16S rRNA gene PCRs. Community structure were assessed either by DGGE analysis or pyrosequencing of the obtained PCR products. Whereas herbivory did not seem to affect the bacterial richness, slight changes in the relative abundances of certain bacterial groups were recorded. For example, an uncultured Acidobacterium was significantly affected by herbivory. As part of the lysimeter experiment, we also investigated the influence of sward composition and the different management regimes on the bacterial communities in the rhizosphere. Both the herbicide application and lower mowing frequencies decreased the bacterial richness in the rhizosphere. Moreover, no differences in bacterial richness between fertilized and unfertilized plots were recorded. Further analyses revealed that a variety of distinct bacterial groups and species in the rhizosphere do respond to the treatments studied. For example, the abundance of the Acidobacteria was significantly reduced in fertilized plots. The opposite was observed for the Actinobacteria. In conclusion, plant-associated bacteria in the endosphere and in the rhizosphere are affected by management regimes. Evaluating the impact of different grassland management regimes and above-ground herbivory onto plant-associated bacteria may results in a better understanding of the multitrophic interaction between plant species, bacterial communities, and above-ground herbivores. Furthermore, the results of this study will help to predict the impact of different grassland management regimes onto plant-associated bacterial communities and related effects on soil ecosystems.