Development of molecular tools to detect and quantify atmospheric CO-oxidizing bacteria and identification of environmental factors regulating their distribution and activity Liliana Quiza, Isabelle Lalonde, Philippe Constant INRS-Institut Armand Frappier, Laval (Québec). Abstract Methodology Soil bacteria scavenging carbon monoxide (CO) are responsible for the biological sink of atmospheric CO. These bacteria exert a significant influence on atmospheric photochemical processes, because the soil uptake of CO mitigates an important fraction of the global emissions of CO from natural and anthropogenic sources. Mitigation of these emissions is of critical importance since CO indirectly regulates the atmospheric lifetime of methane (CH4) - the second most powerful greenhouse gas. Whether the soil uptake of atmospheric CO is vulnerable or not to global change remains to be elucidated. 1) Sampling • Three neighbouring sampling sites - different land-use types: Results A. - Undisturbed deciduous forest (Site F) Table 1. Soil physicochemical properties - Larch monoculture (Site M) * • Triplicate samples (0-10 cm) Fig 4 . Composite soil samples for each triplicate (3 sampling stations per sampling site). ** 5) Soil biological activity, physicochemical properties (Table 1; centered explanatory variables) and coxL gene sequences (Hellinger-transformed OTU frequency data) were integrated into redundancy analysis (RDA) in order to identify clone sequences linked to elevated CO uptake activity and the environmental factors influencing their distribution. The most parsimonious model to explain coxL gene sequence distribution was obtained by stepwise forward selection of environmental variables. Soil water content and CO uptake activity explained >40% of coxL gene distribution in soil (Fig. 9). Introduction Fig 5. Localisation (St-Amable, Qc) and picture of the sampling sites. - “Universal” (OMP + BMS) coxL degenerated primers have been developed from gene sequences available in public database (NCBI) in order to target the as much diversity of coxL as possibly. - Generation of 9 coxL clone library (one per sampling station; 344 clones in total). Studies of microbial CO metabolism unveiled the occurrence of two functional groups of CO-oxidizing bacteria: • “carboxydovore” bacteria unable to grow chemoautotrophically with CO. The CO-dehydrogenase (CODH) is the enzyme catalyzing the CO oxidation reaction in bacteria: Fig 9. Parsimonious RDA triplot of Hellinger-transformed coxL OTUs data explained by soil water content and CO uptake activity. OTU whose distribution was correlated to elevated soil CO uptake activity are highlighted in the triplot as well as the phylogenetic trees (Fig. 6). Most of them belong to the OMP group. 2) Development of molecular tools to detect and quantify coxL in the three environments: Soil bacteria are a significant alternative to sink and oxidize the atmospheric CO which keep it below toxic levels as well as minimize its release to the atmosphere. “carboxydotroph”, bacteria capable of using CO as the sole source of carbon and energy (Fig. 1). ** Fig 8. CO oxidation rate in pmol h-1 g(dw)-1. Work is currently in progress to isolate high affinity CO-oxidizing bacteria and relate coxL gene expression profile to CO uptake activity in soil. • 4) A gas chromatographic assay allowed to measure the potential soil CO uptake activity of the three sampling sites (Fig. 8). CO uptake activity was significantly higher in deciduous forest (Mann-Whitney, P<0.05) and was positively correlated to soil carbon and nitrogen content (P<0.01). - Maize field (Site A) The objective of this investigation was to relate the diversity of coxL gene sequences with CO soil uptake activity, land-use and soil physicochemical properties. Diversity of coxL gene sequence was more important than anticipated and land-use change was shown to influence CO-oxidizing bacterial community structure and activity. CO is an atmospheric trace gas released as a result of terrestrial and aquatic organic mater oxidation, volcanic activity and incomplete combustion. It indirectly contributes to global warming as it strongly competes with greenhouse gases such as methane to react with the hydroxyl radical (OH-) of the atmosphere - the cleansing radical able to destroy CO, CH4 and other reduced trace gases.(1) Results and Discussion - PCR (universal) and qPCR (OMP-specific) assays were developed. 3) Measurement of the potential CO oxidation activity using a gas chromatography assay. 4) Determination of the soil physicochemical properties (i.e. pH, C, N, P, K and water content). B. . Results The aerobic CODH (Fig. 2) belongs to the molybdenum hydroxylases Fe-S family, the enzyme is a dimer of heterotrimers encoded by the genes coxS, coxM and coxL (Fig. 3). CoxL is the large subunit of the CODH. Phylogenetic analysis revealed that coxL gene sequences encompass two main clusters: BMS and OMP but the version conferring a high affinity for CO and the ability to scavenge atmospheric CO is unknown. Conclusion 1) New “Universal” primers targeting coxL (BMS + OMP) were tested by generating coxL clone libraries (800 bp, including CODH active site signature; see Fig. 2). The sequences were aligned, classified into 199 OTU (10% cutoff level) and integrated into a phylogenetic analysis (Fig. 6). Most of the bacteria possessing coxL are unknown. • Land-use change exerts a significant impact on coxL diversity as well as CO oxidation activity (loss of the potential CO soil uptake activity following the conversion of native forest to maize or larch plantation) • Most of the coxL gene sequences retrieved from the soil samples were not affiliated to sequences derived from microbial genome databases, impairing a taxonomic identification of the potential CO-oxidizing bacteria detected in soil. 2) UniFrac analysis demonstrated the correlation between phylogeny and the origin of coxL sequences (unique sequences found in deciduous forest, while sites A and M cannot be discriminated). • Canonical ordination analysis allowed us to identify coxL sequences belonging to potential high affinity CO-oxidizing bacteria. 3) The analysis suggests that conversion of deciduous forest (site F) to larch or maize monocultures have exerted a selective pressure to CO-oxidizing bacteria (Fig. 7). Fig 1. Carboxydotroph metabolic pathway. • Work is currently in progress to isolate, assess the abundance and CO uptake activity of these microorganisms in soil. Taken together, the results of this work will be implemented into molecular models aimed at predicting CO uptake activity in soil. These models will be utilized to predict the response of the biological sink of CO to global change, while determining how land management practices could protect this important ecosystem service. Fig 2. Classification of anaerobic (NiFe) and aerobic (molybdenum) CODH. Bibliography • Weber & King, 2007. Nature Reviews in Microbiology (5): 107-118. • Oelgeschläger & Rother, 2008. Archives of Microbiology (190): 257-269. • Mörsdorf et al., 1992. Biodegradation (3): 61-82. • Borcard et al., 2011, Numerical Ecology with R, Springer, 306 pp. Fig 2. Crystal structure of CODH in Oligotropha carboxidovorans (carboxydotroph bacteria). Hypothesis: Conversion of deciduous forest to monoculture will exert a significant effect on the CO soil uptake activity induced by change in CO-oxidizing bacteria community structure (BMS and OMP groups). Fig 7. UniFrac dendrogram based on coxL gene sequences detected in different sampled ecosystems. Significance of the nodes is shown by the circles (Jackknife resampling, 1000 permutations). Bar = Unifrac units. Fig 6. Maximum likelihood phylogenetic tree based on in silico translated coxL sequences derived from clones libraries and NCBI genome sequence database. (A) Three main clusters were identified (Type I-OMP, Type II-BMS and the novel Type III). (B) Visualization of the OMP group. Color code identify coxL gene sequences for which the occurrence was shown to be correlated to elevated CO uptake activity (see RDA analysis, Fig. 9) • Dobbek et al., 1999. Proceedings of the National Academy of Sciences (96): 8884-8889. Acknowledgements Fonds de recherche du Québec - Nature et technologies (FQRNT) for funding and Prof. Claude Guertin for guidance in the field.