Skip to main content

Community Dynamics and Phenotypic Changes of Limnic Bacteria During Experimental Manipulation of Bottom-up and Top-down Factors

Final Report Summary - COMPLEX (Community dynamics and phenotypic changes of limnic bacteria during experimental manipulation of bottom-up and top-down factors)

The Intra European Fellow COMPLEX had three main objectives:
(i) test the impact of predation on the composition and metabolic versatility of model bacterial communities;
(ii) to investigate bacteria isolates obtained from cultivation systems, and, based on the results from these first two steps;
(iii) to evaluate the response of isolates at close to natural conditions with and without grazing.

Research was first focused on experiments with a natural microbial community from Lake Zurich, Switzerland, and to its adaptation to lab conditions. Once a state of stability in the overall bacterial abundances was reached in lake water cultures, they were tested for objective (i) by adding a protistan predator, a grazing-resistant competitor, or both of them at the same time. By 16S rRNA sequencing it was possible to identify treatment-specific trends in richness, diversity and species composition: the allochtonous strain, even if favoured by the presence of the predator, did not succeed in invading the community. The presence of the predator diminished the total biomass and strongly reduced the bacterial abundance, but supported the overall diversity of the bacterial community.

Moreover, by applying flow cytometry counting and quantitative discrimination of the large phylogenetic groups of bacteria it was possible to detect a large number of co-aggregating strains that were able to escape predation by this strategy. This result allowed selecting for strains able to co-aggregate that were particularly interesting for further research. For the isolation of the co-aggregating strain a specific setup of flow cytometric cell sorting was designed. By sorting single events of the size of a large aggregate from the natural communities exposed to grazing on petri dishes, it was possible to isolate up to 50 different co-aggregating bacterial strains. The strains were then growth to purity, and tested for their ability to escape grazing. They were phylogenetically described (16S rRNA gene sequencing) and tested for their metabolic activity on BIOLOG phenotype arrays, to distinguish between metabolic specialists and generalists.

The aggregating and non-aggregating strains were then tested for possible co-aggregation when exposed to predation. These experiments were performed in batch, semi-continuous, and continuous cultures. Due to the huge amount of samples it was necessary to develop specific analysis software for flow cytometric data. This software was programmed by a master student supervised by the fellow (J. Villiger). It will be made available for free to researchers in the first half of 2011.

In these experiments different behaviour of strains and interesting interactions were observed: strains unable to produce active defence against predation could join aggregates built by other strains and use them as core for the construction of a large aggregate, finding this way to escape grazing. Finally we could identify strains able to grow in very dense populations (up to 50-60 x 106 bacteria ml-1 in the very nutrient limiting media provided) and to escape from predation by such efficient growth.

One finding from these co-cultures was that the supported number of predators was higher, possibly because they could modulate the presence of edible and inedible strains at the same time. In a co-cultivation of the strains Brevundimonas and Arthrobacter co-aggregates (up to 100 %) were mainly composed of Brevundimonas, which is unable to aggregate when under predation in pure culture. This cooperation between bacterial strains, observed for the first time in an experimental system based on isolates common in natural waters, can be considered an efficient system for the preservation of rare bacteria but potentially important in condition of predation stress.

Contact details:
Prof. Jakob Pernthaler
Institute for Plant Biology
University of Zurich
pernthaler@limnol.uzh.ch