Protein sensor turns itself inside out and back again
New research has uncovered the molecular details of how bacteria deal with oxidative stress.
Colin Kleanthous in the department, in collaboration with Jennifer Potts at the University of York and colleagues in Oxford, has published the work in Nature Communications (1). By elucidating the structure and dynamic behaviour of a key protein in a stress-response pathway, they have revealed a new way in which stress sensors in general may function.
Streptomyces colonies. Credit: Mark Buttner, John Innes Centre
All bacteria have sigma factors, proteins that initiate transcription by binding to RNA polymerase. Sigma factors can regulate how bacteria respond to extracellular stresses, which can be diverse – from nutritional cues, to changes in oxidative state or temperature. For some sigma factors, the regulation is imposed by a group of proteins called zinc-binding anti-sigma factors (ZAS). In this system, the ZAS protein rather than the sigma factor responds to stress. ZAS protein and sigma factor are bound in a tight complex in the resting state. When a cue is received, the ZAS protein releases the sigma factor, enabling it to switch on genes that will neutralize the stress.
Different bacteria use ZAS proteins to respond to different stresses. In Mycobacterium tuberculosis, a ZAS-sigma factor complex is required for pathogenesis. In Bacillus subtilis, the system is used to protect the bacteria against antibiotics. The ZAS-sigma system is important in many bacteria for dealing with oxidative stress, as Professor Kleanthous explains: 'All organisms have to be able to deal with oxidative stress, which comes in different types. The ZAS-sigma system is one of the main regulators for dealing with disulfide stress, a toxic consequence of reactive oxygen species.'
The ZAS sensor protein RsrA from the soil microorganism Streptomyces coelicolor enables it to detect and respond to disulfide stress. When its sigma factor (σR) is released, the factor can activate transcription of anti-oxidant genes that dampen the redox change. The system is fast and reversible - within 40 minutes, it has re-established redox homeostasis in the cell.
ZAS proteins share similarities at their zinc-binding site, in particular, a sequence of cysteines and histidines. In RsrA, the three zinc-binding cysteines are essential for redox sensing. All three are reactive cysteines, two of which can form a 'trigger' disulfide bond upon oxidation of RsrA. Together with loss of the zinc ion, this results in the complex falling apart, leaving σR free to activate transcription. The details of how this process takes place, though, were not clear, says Professor Kleanthous. 'We've known that the complex prevents sigma factor from activating genes and that zinc is involved, but we haven't known how the redox-sensing mechanism works.'
Scheme showing the redox homeostasis loop for the RsrA-σR complex. It highlights the zinc coordination residues in reduced (resting) RsrA (RsrAred.Zn2+) from Streptomyces coelicolor. Disulfide stress results in the loss of zinc and formation of a trigger disulfide bond in RsrAox.
Now he and his colleagues have used a combination of kinetic and thermodynamic methods to look at the impact of zinc and oxidation on the complex. Their research, driven in particular by postdoc Karthik Rajasekar in Oxford, has uncovered a new form of redox sensing in bacteria.
The group used NMR, carried out in York by postdoc Konrad Zdanowski in Professor Potts' lab, to determine the structure of RsrA in its reduced, zinc-bound, state and in its oxidised state in which it has lost zinc and formed an internal disulfide bond. This was the first time these structures had been described. They combined this with modeling of the σR-RsrA complex, using cross-linking based docking, in collaboration with Carol Robinson and Shabaz Mohammed at Oxford. A further approach was to develop rapid reaction kinetics methods to look at the association/dissociation of the complex in real-time, so that they could follow how the complex responds to oxidation.
The studies reveal RsrA senses redox state using a mechanism involving the protein's hydrophobic core that has not been described previously. The group found that RsrA changes its structure significantly to bind σR. In doing so, it exposes its hydrophobic core, using these residues to keep hold of σR and stabilize the complex.
'RsrA opens a bit like a clam shell, embracing the sigma factor so that it can't bind RNA polymerase,' explains Professor Kleanthous. 'As part of the structural re-organisation of RsrA, one of its helices undergoes a telescopic extension, essentially doubling its length.'
Zinc plays an essential role in stabilizing the RsrA-σR interaction. Whilst it is not directly involved in RsrA binding σR, it coordinates the way in which RsrA holds on to σR. When zinc is released upon oxidation, the complex is weakened because of the loss of this coordination, and so the complex begins to dissociate. The previously zinc-bound cysteines are now free to form a trigger disulfide, which causes the structure to collapse to a compact state where all the σR-binding residues are pulled back into the hydrophobic core.
RsrAred.Zn2+(a) uses hydrophobic core residues to bind σR (b), which are sequestered to the interior in RsrAox (c) following oxidation. Hydrophobic residues that contribute to RsrA's hydrophobic core in all three of its structural states are coloured green, while those that also interact with σR are coloured red.
The discovery of a new paradigm for redox stress sensing in which zinc plays a central role in holding the ZAS protein-sigma factor complex together, has wider implications for the field, says Professor Kleanthous. 'More than a thousand ZAS proteins have been found in bacterial genomes, which can respond to a range of different cues, not just redox stress. Researchers can use our findings to understand how ZAS proteins function as generic stress sensors.'
1. The anti-sigma factor RsrA responds to oxidative stress by reburying its hydrophobic core. Rajasekar KV, Zdanowski K, Yan J, Hopper JTS, Francis M-LR, Seepersad C, Sharp C, Pecqueur L, Werner JM, Robinson CV, Mohammed S, Potts JR and Kleanthous C. (2016) Nat. Commun. 7:12194 doi:10.1038/ncomms12194