Probing regulon of ArcA in Shewanella oneidensis MR-1 by integrated genomic analyses

Abstract

Background: The Arc two-component system is a global regulator controlling many genes involved in aerobic/anaerobic respiration and fermentative metabolism in Escherichia coli. Shewanella oneidensis MR-1 contains a gene encoding a putative ArcA homolog with ~81% amino acid sequence identity to the E. coli ArcA protein but not a full-length arcB gene.

Results: To understand the role of ArcA in S. oneidensis, an arcA deletion strain was constructed and subjected to both physiological characterization and microarray analysis. Compared to the wild-type MR-1, the mutant exhibited impaired aerobic growth and a defect in utilizing DMSO in the absence of O2. Microarray analyses on cells grown aerobically and anaerobically on fumarate revealed that expression of 1009 genes was significantly affected (p < 0.05) by the mutation. In contrast to E. coli ArcA, the protein appears to be dispensable in regulation of the TCA cycle in S. oneidensis. To further determine genes regulated by the Arc system, an ArcA recognition weight matrix from DNA-binding data and bioinformatics analysis was generated and used to produce an ArcA sequence affinity map. By combining both techniques, we identified an ArcA regulon of at least 50 operons, of which only 6 were found to be directly controlled by ArcA in E. coli.

Conclusion: These results indicate that the Arc system in S. oneidensis differs from that in E. coli substantially in terms of its physiological function and regulon while their binding motif are strikingly similar.

Figure 1

Characteristics of S. oneidensis MR-1 and mutant strains under various conditions. MR-1 (blue Diamond), JZ3988K (purple Square). (A) Growth under aerobic conditions, Complementation of JZ3988K with pBBR-ARCA (light green triangle) was also shown. (B) Dissolved oxygen concentration in MR-1 and JZ3988K cultures was plotted against OD₆₀₀ values. (C) Survival rates during the stationary phase under aerobic conditions in shaken (blue diamond and purple Square) and still cultures (yellow diamond and green square) were shown.

Figure 2

Differentially expressed genes grouped by functional classification according to the TIGR S. oneidensis genome database. A, Amino acid biosynthesis; B, Biosynthesis of cofactors, prosthetic groups, and carriers; C, Cell envelope; D. Cellular processes; E, Central intermediary metabolism; F, Disrupted reading frame; G, DNA metabolism; H, Energy metabolism; I, Fatty acid and phospholipid metabolism; J, Hypothetical proteins; K, Mobile and extrachromosomal element functions; L, Protein fate; M, Protein synthesis; N, Purines, pyrimidines, nucleosides, and nucleotides; O, Regulatory functions; P, Signal transduction; Q, Transcription; R, Transport and binding proteins; S, Unclassified; T, Unknown function. Bars in black are the genes that showed decreased expression in the presence of ArcA; bars in gray are the genes that showed increased expression in the presence of ArcA.

Figure 3

Hierarchical clustering of selected genes. All these genes are listed in Table S2 (in additional file 3). Expression differences (ΔarcA/MR-1) were represented by colors: red, induced, yellow, insignificant, and green, repressed. Each pattern is identified by different colors on the dendrogram and by numbers that correspond to the gene expression patterns. +O₂, aerobic conditions; -O₂, anaerobic conditions.

Figure 4

Expression changes of genes in TCA cycle and glyoxylate pathway under aerobic conditions. Changes were recorded as the ratio of expression in ΔarcA to that in MR-1, "--" represents unaffected by the mutation, "↑" represents up-regulated in the arcA^- strain.

Figure 5

ArcA(-P) Binding to selected promoters by EMSA. (A). Overproduced and purified recombinant S. oneidensis His₆-ArcA from E. coli BL21 cells. (B). Interaction of so1661 promoter DNA with S. oneidensis His₆-ArcA. The probe was prepared by PCR with SO1661-EMSA-F (³³P end-labeled) and SO1661-EMSA-R primers (Table S4 in additional file 5). The EMS assay was performed with 2 nM ³³P end-labeled probes and various amounts of ArcA (left panel) or ArcA-P proteins (right panel). The protein concentrations for lanes 1–9 are 0, 0.125, 0.25, 0.5, 1.0, 2.0, 4.0, 4.0, 4.0 μM, respectively. Non-specific competitor DNA, (2 μg poly dI·dC), was added (lane 8) and specific competitor (10 μM unlabeled SO1661 probe) was added (lane 9). (C). The binding assay was performed in the presence of 0, 1, or 2 μM ArcA-P and 2–5 nM radiolabeled promoter DNA 0.2 μg/μl poly(dI·dC) was used in all these binding reactions to block non-specific interactions. Promoter region of so0011 (gyrB) was included as negative control. The phosphorylation of the ArcA protein was done with carbamoyl phosphate.

Figure 6

Sequence logo for the ArcA-P recognition matrix in S. oneidensis. The sequences used were listed in Table 1. The sequence conservation, measured in bits, is shown as the height of a stack of letters at each base position. Sequence logo for the ArcA-P recognition matrix in E. coli was also shown as comparison. The E. coli sequences used were from the previous report by Liu and De Wulf [12].

Figure 7

Common genes in S. oneidensis and E. coli genomes and in ArcA regulons of these two bacteria. Common genes in S. oneidensis and E. coli genomes are obtained from [68] using default similarity cutoffs (maximum E-value, 1e-5; minimum percent identity, 30).