For more than a decade, it has been recognized that arsenate [H2AsO4 1-; As(V)] can be used by microorganisms as a terminal electron acceptor in anaerobic respiration. Given the toxicity of arsenic, the mechanistic basis of this process is intriguing, as is its evolutionary origin. Here we show that a two-gene cluster (arrAB; arsenate respiratory reduction) in the bacterium Shewanella sp. strain ANA-3 specifically confers respiratory As(V) reductase activity. Mutants with in-frame deletions of either arrA or arrB are incapable of growing on As(V), yet both are able to grow on a wide variety of other electron acceptors as efficiently as the wild-type. Complementation by the wild-type sequence rescues the mutants' ability to respire As(V). arrA is predicted to encode a 95.2-kDa protein with sequence motifs similar to the molybdenum containing enzymes of the dimethyl sulfoxide reductase family. arrB is predicted to encode a 25.7-kDa iron-sulfur protein. arrA and arrB comprise an operon that contains a twin arginine translocation (Tat) motif in ArrA (but not in ArrB) as well as a putative anaerobic transcription factor binding site upstream of arrA, suggesting that the respiratory As(V) reductase is exported to the periplasm via the Tat pathway and under anaerobic transcriptional control. These genes appear to define a new class of reductases that are specific for respiratory As(V) reduction.T he consumption of arsenic (As)-tainted surface waters and ground waters has created a public health crisis in many countries (1, 2). Although much of the As contamination derives from natural weathering and dissolution of As-bearing minerals, recognition that microorganisms can alter the mobility of As in natural waters through redox transformations (3) drove the discovery of the first arsenate [As(V)]-respiring bacterium nearly a decade ago (4). Since then, many more microorganisms that can reduce As(V) to arsenite [H 3 AsO 3 , As(III)] have been discovered (5-8), but a mechanistic understanding of this metabolism has lagged. To date, only three studies have described the biochemistry of arsenate respiration (5, 9, 10), and detailed biochemical analyses have not been performed. In part, the limitations of these studies can be attributed to the fact that the As(V)-respiring organisms being studied were not genetically tractable. To quantify the geochemical impact of As(V)-respiring microorganisms in a given locale, we must be able to predict when these organisms will be active, and how rapidly they will transform As(V). Identification of the gene(s) that control this process, elucidation of their regulation, and determination of the kinetics of their protein products, are necessary steps toward understanding the specific contribution of As(V)-respiring bacteria to As-cycling in the environment.In response to this need, a new As(V)-respiring species, Shewanella strain ANA-3 that is amenable to genetic analysis, was recently isolated (8). This organism contains two systems for reducing As(V). One is similar to the well conserved...