Projects

The Plant-Pollinator-Microbe Triangle

Bees are important agents in ecological and agri- cultural systems, mostly through their incommensurable pollination services. It is however not only the abundance of bees that regulates ecosystem benefits, but also a diversity of pollinators is required to maintain effective pollination. Microbial assessments are rare beyond honey- and bumble-bees, although it is of great ecological importance to address these to understand life history and threats.

There is not only a lack of knowledge about pathogens, but bee-microbe associations in general. Mutualists are important for host health as well as nutrition uptake, but their co-evolution with bee species completely unclear. Also in how far foraging strategies, immunity and the chemical composition of body surfaces are linked to such. Beyond bees, we work on identification of patterns in microbiota of flower communities and their effects on pollination. Our group addresses fundamental questions about diversity and community structure, but also aims to understand the ecological functions and the molecular mechanisms. 

Carnivorous Plants: Evolution of Traps and Microbiota

In a pioneer study, we were interested in the difference between an obligatory (N. rafflesiana) and a facultative (N. hemsleyana) insect feeding plant from Borneo. N. hemsleyana is capable of using feces from wooly bats that visit and sleep within pitchers. Astonishingly, bacterial diversity and community structure was not much different between species (excluding gastro-intestinal bacteria), yet pitchers sampled during digestive phases showed increases in the abundance of acidophilic and acid-producing bacteria.

In contrast to pitchers, Dionaea traps close upon insect touch and re-open after digestion. We performed monitoring of dynamics of bacterial communities in time series feeding experiments and were able to show resistances in the natural microbiome against the hard conditions during digestion, whereas those bacteria introduced by prey were extinguished.

Carnivory has evolved independently in different taxonomic plant clades. Our results suggest that there are also convergent patterns in microbial associations. We are currently conducting follow-up studies that include members of all carnivorous genera (within 13 plant families) to find convergent and derived patterns in microbiota and co-evolution of bacterial genomes dependent on plant taxonomy, trap types and biogeography. We aim to identify microbial pathways and genes relevant for prey dissimilation, nutrient uptake and pathogen resistance for the host within the context of the co-evolutionary history.

Molecular and Bioinformatic Tools in Ecology and Evolution

For next-generation sequencing based approaches, most current methods target microbial organisms. It becomes however more and more apparent that such methods are applicable and beneficial also in the investigation of other taxa beyond the microbial world. My workgroup has performed pioneering work in meta-barcoding with eukaryotes. We have developed and applied laboratory methods and bioinformatic strategies to assess the diversity of pollen from bee collections (and other plants) using next-generation sequencing devices without the need for manual separation or light microscopy.

While next-generation sequencing is already a great win for ecologists, it also has it’s downsides. The data needs bioinformatical processing and most algorithms currently available are dedicated to genomics and transcriptomics. My workgroup is thus devel- oping bioinformatic tools to analyse molecular and community data, as e.g. bcgTree, a whole-genome based tree builder or 16s-to-Traits to map genomes to communities inferred through meta-barcoding.

In general, the link between taxa and their ecological function is often not clear in community analyses. For many ecological questions the ecological properties (traits) represented in a community are more informative than just the scientific names of their members. Furthermore, other properties like threat status, invasiveness, or human usage are relevant for many studies, which can not be derived only from taxonomic names. Despite the fact that various public databases collect such trait information, it is still a tedious manual task to enrich existing community tables with this important functional information. We thus develop a browser-based web-plattform, the FENNEC (Functional Exploration of Natural Networks and Ecological Communities)to ease this process by mapping publicly available or private trait data to the user’s community tables in an automated process. Public trait information is still sparse, but what is present already helps in interpretation of community data.