Bacillus subtilis Fnr senses oxygen via a [4Fe-4S] cluster coordinated by three cysteine residues without change in the oligomeric state.

The oxygen regulator Fnr is part of the regulatory cascade in Bacillus subtilis for the adaptation to anaerobic growth conditions. In vivo complementation experiments revealed the essential role of only three cysteine residues (C227, C230, C235) at the C-terminus of B. subtilis Fnr for the transcriptional activation of the nitrate reductase operon (narGHJI) and nitrite extrusion protein gene (narK) promoters. UV/VIS, electron paramagnetic spin resonance (EPR) and Mössbauer spectroscopy experiments in combination with iron and sulphide content determinations using anaerobically purified recombinant B. subtilis Fnr identified the role of these three cysteine residues in the formation of one [4Fe-4S]2+ cluster per Fnr molecule. The obtained Mössbauer parameters are supportive for a [4Fe-4S]2+ cluster with three cysteine ligated iron sites and one non-cysteine ligated iron site. Gel filtration experiments revealed a stable dimeric structure for B. subtilis Fnr which is independent of the presence of the [4Fe-4S]2+ cluster. Gel mobility shift and in vitro transcription assays demonstrated the essential role of an intact [4Fe-4S]2+ cluster for promoter binding and transcriptional activation. An amino acid exchange introduced in the proposed alphaD-helix of B. subtilis Fnr (G149S) abolished its in vivo and in vitro activities indicating its importance for intramolecular signal transduction. The clear differences in the localization and coordination of the [4Fe-4S] cluster and in the organization of the oligomeric state between Escherichia coli and B. subtilis Fnr indicate differences in their mode of action.

The Fnr regulon of Bacillus subtilis.

The Bacillus subtilis transcriptional regulator Fnr is an integral part of the regulatory cascade required for the adaptation of the bacterium to low oxygen tension. The B. subtilis Fnr regulon was defined via transcriptomic analysis in combination with bioinformatic-based binding site prediction. Four distinct groups of Fnr-dependent genes were observed. Group 1 genes (narKfnr, narGHJI, and arfM) are generally induced by Fnr under anaerobic conditions. All corresponding promoters contain an essential Fnr-binding site centered -41.5/-40.5 bp upstream of the transcriptional start point, suggesting their induction by direct Fnr interaction. Group 2 genes (alsSD, ldh lctP, ywcJ, and cydABCD) are characterized by anaerobic repression in the presence of nitrate. Mutational analysis of the Fnr-binding sites found in three of the corresponding promoters excluded their function in Fnr-mediated repression. Genetic evidence showing that group 2 genes are anaerobically repressed by nitrate reductase formation was accumulated. A possible role of the redox regulator YdiH in the regulation of group 2 genes was initially investigated. Group 3 genes are characterized by their Fnr-dependent activation in the presence of nitrate and the lack of an Fnr-binding site in their promoters. The analysis of Group 3 gene transcription (ykuNOP and ydbN) indicated that Fnr induces nitrate reductase production, which leads to the formation of the regulatory compound nitrite from nitrate. Finally, the group 4 operon acoABCL, lacking an Fnr-binding site, requires Fnr-dependent nitrate reductase formation for its general anaerobic induction. A regulatory model for the observed complex Fnr-mediated gene expression was deduced.

Tracking the evolution of porphobilinogen synthase metal dependence in vitro.

Metal ions are indispensable cofactors for chemical catalysis by a plethora of enzymes. Porphobilinogen synthases (PBGSs), which catalyse the second step of tetrapyrrole biosynthesis, are grouped according to their dependence on Zn(2+). Using site-directed mutagenesis, we embarked on transforming Zn(2+)-independent Pseudomonas aeruginosa PBGS into a Zn(2+)-dependent enzyme. Nine PBGS variants were generated by permutationally introducing three cysteine residues and a further two residues into the active site of the enzyme to match the homologous Zn(2+)-containing PBGS from Escherichia coli. Crystal structures of seven enzyme variants were solved to elucidate the nature of Zn(2+) coordination at high resolution. The three single-cysteine variants were invariably found to be enzymatically inactive and only one (D139C) was found to bind detectable amounts of Zn(2+). The double mutant A129C/D139C is enzymatically active and binds Zn(2+) in a tetrahedral coordination. Structurally and functionally it mimics mycobacterial PBGS, which bears an equivalent Zn(2+)-coordination site. The remaining two double mutants, without known natural equivalents, reveal strongly distorted tetrahedral Zn(2+)-binding sites. Variant A129C/D131C possesses weak PBGS activity while D131C/D139C is inactive. The triple mutant A129C/D131C/D139C, finally, displays an almost ideal tetrahedral Zn(2+)-binding geometry and a significant Zn(2+)-dependent enzymatic activity. Two additional amino acid exchanges further optimize the active site architecture towards the E.coli enzyme with an additional increase in activity. Our study delineates the potential evolutionary path between Zn(2+)-free and Zn(2+)-dependent PBGS enyzmes showing that the rigid backbone of PBGS enzymes is an ideal framework to create or eliminate metal dependence through a limited number of amino acid exchanges.