Bacterial biocontrol agents (BCAs) open up the possibility of controlling plant pathogens in an environmentally friendly way. Although they are naturally occurring microbes, some of them can cause diseases in humans. For successful registration it is necessary to test potentially adverse effects on the human health of at-risk candidates. Existing pathogenicity assays are cost-intensive, time-consuming and furthermore they are often inappropriate for facultative pathogens. We developed a new, fast and inexpensive bioassay on the basis of the nematode Caenorhabditis elegans, which is a well-accepted model organism to study bacterial pathogenicity. A selection of eight strains from clinical and environmental origin as well as potential and commercial BCAs from the genera Bacillus, Pseudomonas, Serratia and Stenotrophomonas were screened for their potential to kill the nematode in an in vitro agar plate assay. Furthermore, the motility and reproductive behaviour of nematodes exposed to strains were tested in comparison with those fed by the human pathogen Pseudomonas aeruginosa QC14-3-8 (positive control) and the negative control Escherichia coli OP50. Commercial as well as potential biocontrol strains did not display any adverse effects in all tests. In contrast, the C. elegans assay showed slight effects for clinical and environmental Stenotrophomonas strains. Results showed that the nematode C. elegans provides a model system to indicate the pathogenic potential of BCAs in a very early stage of product development.

We simultaneously assessed the metabolic activity and viability of individual bacterioplankton cells in the coastal and open North Sea. Three different techniques were applied to determine cell features related to the physiological status of the cell. SYBR Green I was used to estimate the nucleic acid content of the cell. Propidium iodide (PI) stains cells with a compromised cell membrane, commonly interpreted as indicative of dead cells. Microautoradiography (MA) with radiolabeled glucose and leucine was applied to indicate metabolically active cells. The relative abundance of metabolically active cells determined by MA was usually < 20% of the total abundance of bacteria. In contrast, the percentage of PI-positive cells in the total bacterial community was generally high (~80%). However, the overwhelming majority (97%) of cells taking up glucose and leucine were also PI-positive. Apparently, the uptake of radiolabeled substrate is related to PI accumulation in cells, indicating that PI is not a reliable stain to indicate non-active or dead bacteria. We suggest that several methods should be combined to assess the physiological status of individual cells in natural bacterioplankton communities.