ApoFnr binds as a monomer to promoters regulating the expression of enterotoxin genes of Bacillus cereus

Abstract

Bacillus cereus Fnr is a member of the Crp/Fnr (cyclic AMP-binding protein/fumarate nitrate reduction regulatory protein) family of helix-turn-helix transcriptional regulators. It is essential for the expression of hbl and nhe enterotoxin genes independently of the oxygen tension in the environment. We studied aerobic Fnr binding to target sites in promoters regulating the expression of enterotoxin genes. B. cereus Fnr was overexpressed and purified as either a C-terminal His-tagged (Fnr(His)) fusion protein or an N-terminal fusion protein tagged with the Strep-tag (IBA BioTAGnology) ((Strep)Fnr). Both recombinant Fnr proteins were produced as apoforms (clusterless) and occurred as mixtures of monomers and oligomers in solution. However, apoFnr(His) was mainly monomeric, while apo(Strep)Fnr was mainly oligomeric, suggesting that the His-tagged C-terminal extremity may interfere with oligomerization. The oligomeric state of apo(Strep)Fnr was dithiothreitol sensitive, underlining the importance of a disulfide bridge for apoFnr oligomerization. Electrophoretic mobility shift assays showed that monomeric apoFnr, but not oligomeric apoFnr, bound to specific sequences located in the promoter regions of the enterotoxin regulators fnr, resDE, and plcR and the structural genes hbl and nhe. The question of whether apoFnr binding is regulated in vivo by redox-dependent oligomerization is discussed.

FIG. 1.

Gel filtration and DLS chromatograms of purified Fnr proteins. Fnr_His (A), _StrepFnr (B) and reduced _StrepFnr (C) were injected (∼300 μg in 100 μl) into a Superdex 200 column (HR 10/30) with 50 mM Tris-HCl (pH 8.3)-120 mM NaCl as the eluant at a flow rate of 0.5 ml/min. DTT (10 mM) was added to the elution buffer to determine the oligomeric state of reduced _StrepFnr in panel C. The black and gray lines correspond to the light scattering (LS) signal and the UV signal recorded at 280 nm, respectively. These signals were normalized as a ratio from 0.0 to 1.0 for comparison (left y axis). The molecular mass estimates of the major peaks are also indicated by thick black broken lines (right y axis).

FIG. 2.

SDS-PAGE analysis of the oligomeric nature of _StrepFnr and Fnr_His. (A and B) Effect of DTT on _StrepFnr (A) and Fnr_His (B) oligomerization. Purified proteins were incubated with 0, 10, 50, 100, or 200 mM DTT (lanes 2 to 6, respectively). Recombinant proteins were then subjected to nonreducing SDS-PAGE. The arrows show the positions of monomers (m), dimers (d), and higher oligomers (o). Lane 1 contains molecular mass standard proteins. (C) SDS-PAGE profile of Fnr_His cross-linked with EDC. Fnr_His (5 μM) was cross-linked with EDC. Products were visualized by immunoblotting with anti-His antibody. Lane 1, cross-linked Fnr_His; lane 2, untreated Fnr_His. (D) Nondenaturing SDS-PAGE profile of Fnr_His cross-linked with diamide. Lane 1, molecular mass standard proteins; lane 2, untreated Fnr_His; lanes 3 and 4, disulfide-linked Fnr_His with 1 mM and 10 mM diamide, respectively. The arrows show the positions of monomers (m), dimers (d), trimers (t), and higher oligomers (o).

FIG. 3.

Western blot detection of endogenous Fnr species from B. cereus cells. Lysates of wild-type B. cereus F4430/73 and fnr mutant were probed with polyclonal Fnr antiserum. Both strains were grown in regulated batch culture (pH 7.2) under aerobiosis (4). Proteins were separated by nonreducing SDS-PAGE. Lane 1, _StrepFnr purified from E. coli; lane 2, fnr mutant; lane 3, wild-type strain. The putative identities shown on the right were determined for the wild-type strain on the basis of results obtained with both recombinant Fnr and fnr mutant strains. The arrows show the positions of monomer (m) and dimer (d) forms. The positions and masses (in kilodaltons) of molecular mass markers are given to the left of the gel.

FIG. 4.

Potential Fnr-binding sites in the 5′ untranslated regions of fnr, resDE, plcR, hbl, and nhe. All numbering is relative to the transcription start site at position +1. (A) Potential Fnr-binding sites are shown relative to the transcription start site as gray boxes. PlcR boxes are shown as black boxes. (B) Genetic organization of the resDE promoter region. The transcriptional start site (+1) determined by 5′ RACE PCR is in bold type. The putative −35 and −10 motifs are underlined. Putative Crp/Fnr boxes are indicated by a gray background.

FIG. 5.

Binding of apoFnr to 5′UTRs of fnr, resDE, plcR, hbl, and nhe genes determined by EMSAs. DNAs corresponding to fnr (A), resDE (B), plcR (C), hbl1 (D), hbl2 (E), nhe (F), and a negative control (G) were bound with increasing concentrations of apoFnr as indicated by the height of the triangle below the gel. The results presented are representative examples of an experiment performed in triplicate with either purified Fnr_His or with reduced _StrepFnr (purified _StrepFnr plus 200 mM DTT). Lanes 1 to 11 contain 0, 0.2, 0.4, 0.6, 0.8, 1, 2, 3, 4, 5, and 6 μM apoFnr protein, respectively.

FIG. 6.

Proposal for the regulation of apoFnr activity by a thiol-disulfide redox switch. Brackets indicate that one or more disulfide bonds may be involved.