Methylation of arsenic by single-species and soil-derived microbial cultures

Arsenic (As) is simultaneously a ubiquitous and a toxic element. Arsenic is subject to bio-transformations catalyzed by microorganisms constituting the As biogeochemical cycle. The primordial Earth was devoid of oxygen, exposing life to the highly mobile and toxic arsenite. This early contact with As is imprinted in the tree of life as arsenic resistance genes, e.g. arsenite efflux by transmembrane transporters. Arsenic bio-transformations includes As methylation catalyzed by the ArsM enzyme. ArsM attaches methyl groups to inorganic As, generating organic As in the trivalent (highly toxic species) and pentavalent (relatively innocuous species) forms. Worldwide concern about exposure to As has been raised due to the high As levels in groundwater in South-East Asia used for consumption and rice cultivation. Rice plants are grown under flooded conditions which are conducive to arsenite mobilization and uptake by roots, resulting in the presence of inorganic and methylated As in grains. Compared to inorganic As, methylarsenicals are more easily accumulated in rice grains and are implicated in rice plant disorders resulting in sterility. Methylated species are not synthesized by the plant but by soil microorganisms. To date, mainly aerobic microbial species able to methylate As as an apparent detoxification strategy, have been isolated from paddies. Thus, despite the fact that methylarsenical synthesis appears to be enhanced during anoxic conditions, the drivers and the physiological role of As methylation remain unknown in the absence of oxygen.

The thesis focuses on the identification of active As-methylating microorganisms as single species and as members of anaerobic microbiomes from rice paddy soils. In the first part, five bacterial strains and two methanogenic archaea, belonging to genera identified in paddy soils, were evaluated for active arsenite methylation. While aerobic bacterial strains were sensitive to increasing As concentrations, they efficiently methylated As. Conversely, anaerobic bacterial strains were resistant to increasing As concentrations but methylated poorly. Suspecting that As detoxification was occurring through As efflux and outcompeting methylation under anaerobic conditions, we proceeded with the deletion of the arsenite transmembrane transporter in the anaerobic strain Clostridium pasteurianum. The As efflux was disrupted in the mutant leading to higher intracellular concentrations, higher arsM transcription, and increase in the methylation efficiency. Based on the results, we hypothesize that under anoxic conditions, the efficient arsenite efflux systems from anaerobic microorganisms preclude efficient As methylation. In the second part, meta-omic approaches (metagenomics, metatranscriptomics and metaproteomics) were used to identify active As methylators in two soil-derived microbiomes shown to methylate arsenite. The metagenomic data was used to reconstruct the microbial genomes present as metagenome-assembled genomes (MAG). The metabolic capacity for each MAG was identified by the functional annotation of genes while active metabolisms were deciphered from mRNA transcripts (metatranscriptome) and protein expression (metaproteome). Annotation and expression of arsenic resistance genes pointed to fermenting microorganisms as the main drivers of As methylation in both microbiomes. This work is a contribution to the understanding of the linkages between microbial diversity and arsenic bio-transformation.