Abstract
Summary: Geochemistry often reveals unexpected (anti)correlations. Arsenic (As) and selenium (Se) are cases in point. We explore the hypothesis that bacteria living in an As‐replete environment recruited a biological process involving Se and sulfur to fulfil their need for As detoxification. In analogy with the formation of arsenolipids and arsenosugars, which are common non‐toxic As metabolites derived from microbial and plant metabolism, we attempt to explain the prevalence of novel sulfur‐containing As derivatives, in particular monothioarsenate, in the aqueous environment. Thiolated‐As species have been overlooked so far mainly because of the difficulty of their identification. Based on comparative genomics, we propose a scenario where SelD and SelU proteins, commonly used to make selenophosphate and modify transfer RNA, have been recruited to make monothioarsenate, a relatively innocuous arsenical. This hypothesis is discussed in terms of the relative geochemical distribution of Se and As.

This publication has 95 references indexed in Scilit: