Effects of a ciliate protozoa predator on microbial communities in pitcher plant (Sarracenia purpurea) leaves

Abstract

The aquatic communities found within the water filled leaves of the pitcher plant, Sarracenia purpurea, have a simple trophic structure providing an ideal system to study microscale interactions between protozoan predators and their bacterial prey. In this study, replicate communities were maintained with and without the presence of the bactivorous protozoan, Colpoda steinii, to determine the effects of grazing on microbial communities. Changes in microbial (Archaea and Bacteria) community structure were assessed using iTag sequencing of 16S rRNA genes. The microbial communities were similar with and without the protozoan predator, with>1000 species. Of these species, Archaea were negligible, with Bacteria comprising 99.99% of the microbial community. The Proteobacteria and Bacteroidetes were the most dominant phyla. The addition of a protozoan predator did not have a significant effect on microbial evenness nor richness. However, the presence of the protozoan did cause a significant shift in the relative abundances of a number of bacterial species. This suggested that bactivorous protozoan may target specific bacterial species and/or that certain bacterial species have innate mechanisms by which they evade predators. These findings help to elucidate the effect that trophic structure perturbations have on predator prey interactions in microbial systems.

Figure 1. Bar graph of normalized 16S rRNA iTag sequence data.

The most abundant classes are shown. Less abundant classes are summed under “Other.” Samples with C. steinii are referred to as +CS and those without as –CS.

Figure 2. Rarefaction curve of the number of observed OTUs from 16S rRNA iTag sequence data.

Figure 3. Heatmap of OTUs that were statistically significantly more abundant in samples with or without the protozoan predator (only the top ten OTUs that changed the most with protozoa and without protozoa are shown).

Those OTUs that were statistically significantly correlated with an NMDS axis are indicated by an *.

Figure 4. Non-metric multidimensional scaling (NMDS) ordination of normalized 16S rRNA iTag sequence data.

(A) NMDS ordination of the first and third axes showing sample grouping based on the presence or absence of the protozoan predator. (B) NMDS ordination with those OTUs that were statistically significantly correlated with an axis and had a p-value < 0.03 shown by vectors.