Arsenite oxidase aox genes from a metal-resistant beta-proteobacterium

Abstract

The beta-proteobacterial strain ULPAs1, isolated from an arsenic-contaminated environment, is able to efficiently oxidize arsenite [As(III)] to arsenate [As(V)]. Mutagenesis with a lacZ-based reporter transposon yielded two knockout derivatives deficient in arsenite oxidation. Sequence analysis of the DNA flanking the transposon insertions in the two mutants identified two adjacent open reading frames, named aoxA and aoxB, as well as a putative promoter upstream of the aoxA gene. Reverse transcription-PCR data indicated that these genes are organized in an operonic structure. The proteins encoded by aoxA and aoxB share 64 and 72% identity with the small Rieske subunit and the large subunit of the purified and crystallized arsenite oxidase of Alcaligenes faecalis, respectively (P. J. Ellis, T. Conrads, R. Hille, and P. Kuhn, Structure [Cambridge] 9:125-132, 2001). Importantly, almost all amino acids involved in cofactor interactions in both subunits of the A. faecalis enzyme were conserved in the corresponding sequences of strain ULPAs1. An additional Tat (twin-arginine translocation) signal peptide sequence was detected at the N terminus of the protein encoded by aoxA, strongly suggesting that the Tat pathway is involved in the translocation of the arsenite oxidase to its known periplasmic location.

FIG. 1.

MICs of chromium [Cr(III) and Cr(VI)], manganese (Mn), selenium (Se), nickel (Ni), antimony (Sb), cadmium (Cd), lead (Pb), mercury (Hg), and copper (Cu) for strain ULPAs1. Results are means of duplicate determinations.

FIG. 2.

(A) Genetic organization of the arsenite oxidase cluster in strain ULPAs1. Gene orientations are shown by arrows. Insertion sites of the mini-Tn5::lacZ2 transposon are indicated. The locations of primers a and b, used for RT-PCR, are indicated. Restriction sites KpnI, SphI, and XhoI are shown. (B) Predicted structure of the promoter of the aox cluster. The −35 and −10 regions, ribosome binding sites (RBS), ATG codons, and stop codons are boldfaced.

FIG.3.

Sequence alignment of the two subunits of the arsenite oxidase of A. faecalis and the putative proteins deduced from the DNA sequences of strain ULPAs1 and A. pernix. (A) The small Rieske type subunit of A. faecalis (A.f. rieske) is compared to the putative protein encoded by aoxA (ULPAs1 AoxA) and the predicted protein encoded by A. pernix (A.p.APE2563). Residues belonging to the consensus Tat sequence are boxed. (B) The large subunit of A. faecalis (A.f.largesu) is compared to the putative protein encoded by aoxB (ULPAs1AoxB) and the predicted protein encoded by A. pernix (A.p.APE2556). Dark shading indicates the Fe-S clusters (inside the large and the small subunits); white letters on a solid background represent amino acids interacting with the molybdenum center (identical). Light shading, conserved amino acids; asterisks, positions where amino acids are identical; periods, positions where amino-acids are similar.

FIG. 4.

DNA products derived from RT-PCRs of total RNA from a culture of strain ULPAs1 grown in the presence of arsenite (lanes 1 and 2). In lane 1, the concentration of RNA was 2 μg · μl⁻¹; in lane 2, it was 0.2 μg · μl⁻¹. The control reaction (C) without reverse transcriptase confirmed the absence of products derived from contaminated DNA.