|Home | About | Journals | Submit | Contact Us | Français|
Shewanella oneidensis is a metal reducer that uses the cyclic AMP receptor protein, CRP, to regulate anaerobic respiration. In addition, ArcASo is required for anaerobic growth with dimethyl sulfoxide (DMSO) and plays a role in aerobic respiration. The sensor kinase that activates ArcASo in S. oneidensis is not known. ArcB1So, a homolog of the Escherichia coli sensor kinase ArcBEc, was identified and found to be required for DMSO reductase gene expression. In combination with HptA, ArcB1So complemented an E. coli arcBEc mutant. ArcASo, ArcB1So, and HptA appear to constitute a two-component signal transduction system that regulates DMSO reduction in S. oneidensis.
Shewanella oneidensis MR-1 is a metal reducer that belongs to the Gammaproteobacteria. This bacterium can use several soluble electron acceptors, such as oxygen, fumarate, dimethyl sulfoxide (DMSO), trimethylamine N-oxide (TMAO), and nitrate, in addition to insoluble metal oxides (16). In many bacteria, anaerobic respiration requires the activity of the oxygen sensor FNR (8, 19). In S. oneidensis MR-1 and the related species Shewanella sp. ANA-3, anaerobic respiration is regulated by the cyclic AMP receptor protein CRP (15, 17). Global transcriptome analysis and phenotypic studies of an S. oneidensis crp mutant indicated that several anaerobic terminal reductase genes are regulated by CRP (3). Activation of CRP under anaerobic conditions in S. oneidensis MR-1 depends to a large extent on the membrane-bound adenylate cyclase CyaC (3). The mechanism by which CyaC and its homolog in Shewanella ANA-3 (15) are activated is not known.
In addition to CRP, two other systems that control respiration have been identified in S. oneidensis MR-1. The TorSTR phosphorelay system regulates the expression of the TMAO reductase (1, 6), and ArcASo regulates the expression of genes involved in aerobic and anaerobic respiration (7). The S. oneidensis ArcASo exhibits 81% identity with its E. coli counterpart, and arcASo complements the aerobic growth defect and dye sensitivity in the E. coli arcAEc mutant (7). Similarly, arcAEc complements the aerobic growth defect in the MR1 arcASo mutant (7).
In E. coli, the response regulator ArcAEc is activated by the sensor kinase ArcBEc (9, 10). Sensor histidine kinases that activate ArcASo in S. oneidensis MR-1 have not been identified. It was recently suggested that hptA (SO_1327), which encodes a protein with ~58% similarity to the Hpt (histidine phosphotransfer) domain of the E. coli ArcBEc, may be involved in ArcASo activation (7). In E. coli, the Hpt domain is required for ArcAEc phosphorylation (11).
To identify additional genes involved in regulating anaerobic respiration in S. oneidensis, we analyzed the role of SO_0577, which encodes a protein with 28% identity and 47% similarity, over 528 amino acids, to the Vibrio cholerae ArcB (locus tag Vc2369). SO_0577, which we designated arcB1So, is predicted to encode a histidine sensor kinase of 1,188 amino acids. Sequence analysis using BLASTP and MIST (http://genomics.ornl.gov/mist/) (20) indicated that ArcB1So contains two putative PAS and two regulatory receiver (REC) domains but lacks an Hpt domain (Fig. (Fig.1).1). The role of ArcB1So in respiration was compared to that of ArcASo. S. oneidensis mutants with chromosomal deletions of arcB1So and arcASo were generated using previously published methods (3) and the primers listed in Table S1 of the supplemental material. Wild-type and mutant strains were grown anaerobically in minimal medium supplemented with 50 mM lactate, 0.02% Casamino Acids, and 10 mM fumarate, 10 mM TMAO, or 10 mM DMSO as described previously (17). In agreement with previous reports (7), ΔarcASo was deficient in anaerobic growth with DMSO (Fig. (Fig.22 A) but grew similar to the wild type when TMAO and fumarate were used as electron acceptors (data not shown). The ΔarcB1So mutant exhibited a similar phenotype to ΔarcASo. It grew with fumarate and TMAO (data not shown) but was deficient in growth with DMSO (Fig. (Fig.2A2A).
Recent transcriptome analysis identified genes that may be regulated by ArcASo in S. oneidensis (5). These included the DMSO reductase operon (dmsEFAB; SO_1427-SO_1430), the fumarate reductase gene (fccA; SO_0970), and the cco operon that is predicted to encode a cbb3-type cytochrome c oxidase (ccoNOQP; SO_2364 to SO_2361). If ArcASo and ArcB1So were components of the same signal transduction pathways, they should exert a similar effect on the expression of the above reductase genes. To test this, we generated the reporter plasmid pMC10 by modification of the IncP plasmid pJBC1 (2). The 0.48-kb fragment that contains the lac promoter, a polycloning region, and the α-lacZ fragment coding region in pJBC1 was replaced with a promoterless lacZ amplified from pPR9TT (18). DNA fragments upstream of the dms, fccA, and cco genes were amplified by PCR using the primers listed in Table S1 of the supplemental material and Phusion polymerase (New England Biolabs). The fragments were digested with HindIII and BamHI and cloned into pMC10, and the resulting constructs were transferred from E. coli β2155 (4) into S. oneidensis wild-type and mutant strains by conjugation. The recombinant strains were incubated for 3 h in minimal medium supplemented with 10 mM fumarate (for cells harboring the fccA or cco promoter-fusion plasmids) or 10 mM DMSO (for cells carrying a dms promoter-fusion plasmid), then assayed for β-galactosidase activity as described previously (14). Expression from the dms promoter was very low in the ΔarcASo and ΔarcB1So backgrounds compared to the wild type (Fig. (Fig.2B).2B). Expression from the fccA promoter was slightly reduced in both mutants, but not to the extent observed for the dms promoter. In contrast, expression from the cco promoter increased in both mutants compared to its level in the wild type under anaerobic conditions (Fig. (Fig.2B).2B). To determine if ArcASo and ArcB1So play a role in gene regulation under aerobic conditions, we measured Pcco-lacZ expression levels in wild-type and mutant strains grown aerobically to early log phase. β-Galactosidase activity in ΔarcASo and ΔarcB1So was slightly higher in these mutants (1,368 ± 23 and 1,085 ± 23 Miller units, respectively [means ± standard deviations]) compared to the wild type (813 ± 19 Miller units). The gene expression results in aerobic and anaerobic cells, in combination with the phenotypic similarities of the mutants, suggest that ArcB1So and ArcASo may be components of the same regulatory pathway, with the sensor kinase ArcB1So playing a role in ArcASo activation.
The prediction that ArcB1So may activate ArcASo was tested by complementation of ΔarcB1So with E. coli arcBEc. Introduction of arcBEc into the S. oneidensis mutant restored anaerobic growth with DMSO similar to the wild type and to the mutant complemented with arcB1So (Fig. (Fig.33 A). The reciprocal experiment was also performed where we tested the ability of the S. oneidensis arcB1So to complement ΔarcBEc. We also tested the role of hptA, which encodes a protein similar to the ArcBEc Hpt domain (7). E. coli BW29859 (ΔarcBEc; The Coli Genetic Stock Center, Yale University) was transformed with plasmids that carry E. coli arcBEc, S. oneidensis arcB1So, hptA, or arcB1So and hptA (Fig. (Fig.3B).3B). The parent and resulting recombinant strains were transformed with a plasmid that carries the E. coli succinate dehydrogenase (sdh) promoter fused to gfp (Psdh-gfp) (21). Under anaerobic conditions, ArcBEc negatively regulates the expression of the succinate dehydrogenase operon, and deletion of arcBEc increases sdh expression (9). E. coli BW29859 recombinant cells were grown anaerobically as described previously (9), and fluorescence was measured using a Synergy HT multimode microplate reader (Biotek). Relative fluorescence in E. coli ΔarcBEc/Psdh-gfp was reduced when either arcBEc or S. oneidensis arcB1So and hptA were introduced into the mutant (Fig. (Fig.3B),3B), indicating complementation and repression of the sdh promoter activity. Introduction of S. oneidensis arcB1So or hptA alone did not complement the ΔarcBEc mutant (Fig. (Fig.3B).3B). Since ArcB1So lacks the Hpt domain that is required for phosphotransfer to ArcAEc (11), these results were not surprising. The finding that complementation of E. coli ΔarcBEc is accomplished when both ArcB1So and HptA are present strongly suggests that these two proteins possess functions combined in the E. coli ArcBEc protein.
The results described above indicate that complementation of E. coli ΔarcBEc by the S. oneidensis genes requires both arcB1So and hptA. If, by analogy to ArcBEc (11), ArcB1So phosphorylates HptA and HptA transfers the phosphate to ArcASo, then we expect mutants deficient in either protein to be phenotypically similar to ArcASo mutants. An S. oneidensis mutant deficient in hptA was previously generated and analyzed (7). This mutant was slightly deficient in anaerobic growth with DMSO, but not to the same extent as an arcASo mutant (7). To further explore the role of hptA and arcB1So, we generated the mutant strains ΔhptA and ΔhptA ΔarcB1So and tested them for anaerobic growth with DMSO. The single and double mutants exhibited growth deficiencies with DMSO (Fig. (Fig.44 A and B) identical to that of the ΔarcASo mutant (Fig. (Fig.2A),2A), and growth was completely restored by the introduction of the corresponding wild-type genes (Fig. (Fig.4A4A and data not shown). Furthermore, complementation of either ΔarcB1So or ΔarcB1So ΔhptA was achieved by the introduction of E. coli arcBEc (Fig. (Fig.3A3A and and4B),4B), indicating that the combination of ArcB1So and HptA is functionally similar to the E. coli ArcBEc.
In addition to anaerobic growth with DMSO, ArcASo and HptA are reported to be involved in aerobic respiration (7). To determine if ArcB1So plays a role in aerobic respiration similar to ArcASo and HptA, mutant and wild-type strains were grown in 50 ml of Luria-Bertani medium in 1-liter flasks with vigorous agitation. Under the conditions of our experiments, ΔarcASo exhibited a more pronounced deficiency in aerobic growth than ΔhptA or ΔarcB1So (Fig. (Fig.4C).4C). This suggests that although ArcB1So and HptA may play a role in aerobic growth, other proteins or mechanisms may be involved in ArcASo activation.
Regulation of anaerobic respiration in S. oneidensis MR-1 is unusual in that it relies on CRP and not on FNR (12, 17). Furthermore, the expression of the DMSO reductase operon in S. oneidensis requires the activity of both CRP and ArcASo (3, 7). In E. coli, ArcAEc plays a major role in the transition to anaerobic growth (13) but is not known to directly regulate the expression of terminal anaerobic reductase genes. We identified ArcB1So, a homolog of the E. coli sensor kinase ArcBEc, that appears to play a role in ArcASo activation and in the regulation of the DMSO reductase operon in S. oneidensis. ArcB1So is predicted to be a hybrid sensor kinase that lacks an Hpt domain. This domain appears to be encoded by a separate gene, hptA, and exhibits a high degree of similarity to the ArcBEc Hpt domain (7). Based on the results presented here we propose that ArcB1So/HptA/ArcASo constitute a signal transduction pathway that regulates anaerobic respiration with DMSO in S. oneidensis. It is interesting that ArcASo also plays a more significant role in aerobic respiration than ArcB1So or HptA, suggesting the presence of some other activating mechanism that may be involved in aerobic activation of ArcASo.
This work was supported by National Science Foundation grant MCB 0543501.
We thank M. McBride and S. Kuchin for helpful discussions and critical reading of the manuscript.
Published ahead of print on 16 April 2010.
†Supplemental material for this article may be found at http://jb.asm.org/.