Understanding strain to strain interactions that affect the bioaugmentation success of a pesticide degrading bacterium in a microbial community

In Europe, pesticide residue concentrations in freshwater used for drinking water production although present at trace concentrations of below-μg/L levels (designated as micropollutant concentrations), often exceed the threshold concentration of 0.1 μg/L for an individual compound. As a consequence, expensive treatment steps are included in drinking water treatment plants (DWTPs). Microbial degradation is considered as a sustainable economically-sound alternative for removing pesticides. 2,6-dichlorobenzamide (BAM), a transformation product of the herbicide dichlobenil, is a common groundwater micropollutant in Europe and creates major problems for drinking water production. Aminobacterniigataensis MSH1 mineralizing BAM, is a candidate for bioaugmentation of sand filters in DWTP to avert BAM-contamination. Long-term functionality of the organism is though jeopardized by loss of cell density and activity in the oligotrophic system. Bioaugmentation involves an invasion process requiring the establishment and activity of a foreign microbe in the resident community of the target environment. Interactions with resident micro-organisms, either antagonistic or cooperative, are believed to impact invasion. However, few studies have examined the interactions between an invader and resident species of its target environment, and none of them considered a bioremediation context. The overall aim of this thesis was to improve our knowledge about the interactions between BAM degrading strain MSH1 and sand filter residents and to characterize cooperative interactions that support BAM biodegradation.

In the first part of the study, the objective was to examine the nature of the interactions between MSH1 and sand filter resident bacteria in dual- and triple-species assemblies in sand microcosms, especially with respect to the behavior of the resident strains. To this end, sand microcosm experiments mimicking the oligotrophic DWTP sand filter environment were performed. BAM mineralization, the target functionality, was measured as well as the cell densities of MSH1 and of resident strains after a competition phase to score the interaction. As shown previously (Vandermaesen et al., 2017), the residents affected MSH1 mediated BAM mineralization without always impacting MSH1 cell densities, indicating effects on cell physiology rather than cell number. Exploitative competition explained most effects (70%) but indications of interference competition were also found. Interestingly, two residents, Piscinibacter sp. K169 and Brachybacterium sp. S51, were identified that improved BAM mineralization in dual-species assemblies and overruled negative effects from antagonist residents in triple-species systems. Moreover, the beneficial effects of Piscinibacter sp. K169 and Brachybacterium sp. S51 on MSH1 functionality was accompanied by increased cell numbers of K169 and S51 suggesting a mutualistic interaction between MSH1 and these two so-called “benefactor” strains. Co-inoculation of MSH1 with benefactor strains in which the benefactor strain - adapted to the target environment - supports the fitness and activity of MSH1, was proposed as a new type of strategy to improve bioaugmentation.

In the second part of the thesis, we studied the role of the initial population size and the available organic carbon resources in the mutualistic interaction between Piscinibacter sp. K169 and A.niigataensis MSH1 in sand microcosm experiments. The interaction between MSH1 and K169 was only disturbed when the densities of one of the strains were not higher than 104 cells/mL. Supplying acetate as a carbon source was found redundant for improving BAM mineralization. Instead, the organic carbon present on the sand was indispensable as the effect on BAM mineralization disappeared and none of the two strains showed growth in sand devoid from organic carbon. The sand matrix itself was not required. The result suggests that intrinsic organic carbon present on the sand thrives the mutualistic interaction between K169 and MSH1 and that the mutualism remains when both strains are inoculated at similar cell densities that are higher than 104 cells/mL. These findings support the use of K169 for assisting bioaugmentation with MSH1 in sand filters exploited in DWTPs since no additional carbon needs to be added and relatively low initial cell densities might be used.

In the third part of this study, we questioned and examined whether this mutualistic interaction extends to other pesticide degrading bacterial strains of the same or a different genus of MSH1. Moreover, we examined the persistence of the cooperation by scoring mineralization and cell densities after 14 days in addition to 7 days of co-habitation. In addition to BAM mineralization in MSH1, K169 stimulated BAM and 2,4-D mineralization by respectively Aminobacterniigataensis LG1 and Cupriavidus necator JMP134 while linuron mineralization by Variovorax sp. SRS16 and carbofuran mineralization by Novosphingobium sp. KN65.2, were not affected. On the other hand, growth of K169 was stimulated in the presence of all pesticide degraders except JMP134. After 2 weeks, the beneficial effects of K169 on MSH1, LG1 and JMP134 functionality diminished or even reversed, likely because of nutrient depletion. In contrast, cell densities of K169 in all dual-species systems remained higher than in the K169 mono-culture system. These findings extend the beneficial effect of the sand filter isolate Piscinibacter sp. K169 towards other pesticide degraders and opens doors for K169 assisted bioaugmentation of multiple pesticide degraders in drinking water treatment.

In the final part of this study, the aim of was to elucidate the underlying molecular mechanisms of the mutualistic interaction between Piscinibacter sp. K169 and A.niigataensis MSH1. To this end, hypotheses related to the mechanisms were proposed by analyzing the full genome sequences of the two individual organisms, identifying complementing metabolisms that could suggest cross-feeding and/or cell signaling features. To examine these hypotheses, differential transcriptomic analysis was performed on mono- versus dual- species systems but unfortunately, sufficient quality RNA was only obtained from two K169 mono-species systems and one dual-species system. Cross-feeding of amino acids, carbohydrates, cofactors, aromatic degradation products and energy were identified as a potential mechanism for supporting growth of K169 by MSH1. Moreover, genes involved in chemotaxis and two-component system were upregulated in K169 in the dual-species system compared to the K169 mono-species system. Additional RNAseq experiments are required to verify our preliminary results.