Terrence D. Mulhern

In Escherichia coli, SigmaS (σ^S) is the master regulator of the general stress response. The cellular levels of σ^S are controlled by transcription, translation and protein stability. The turnover of σ^S, by the AAA+ protease (ClpXP), is tightly regulated by a dedicated adaptor protein, termed RssB (Regulator of Sigma S protein B)—which is an atypical member of the response regulator (RR) family. Currently however, the molecular mechanism of σ^S recognition and delivery by RssB is only poorly understood. Here we describe the crystal structures of both RssB domains (RssB_N and RssB_C) and the SAXS analysis of full-length RssB (both free and in complex with σ^S). Together with our biochemical analysis we propose a model for the recognition and delivery of σ^S by this essential adaptor protein. Similar to most bacterial RRs, the N-terminal domain of RssB (RssB_N) comprises a typical mixed (βα)₅-fold. Although phosphorylation of RssB_N (at Asp58) is essential for high affinity binding of σ^S, much of the direct binding to σ^S occurs via the C-terminal effector domain of RssB (RssB_C). In contrast to most RRs the effector domain of RssB forms a β-sandwich fold composed of two sheets surrounded by α-helical protrusions and as such, shares structural homology with serine/threonine phosphatases that exhibit a PPM/PP2C fold. Our biochemical data demonstrate that this domain plays a key role in both substrate interaction and docking to the zinc binding domain (ZBD) of ClpX. We propose that RssB docking to the ZBD of ClpX overlaps with the docking site of another regulator of RssB, the anti-adaptor IraD. Hence, we speculate that docking to ClpX may trigger release of its substrate through activation of a “closed” state (as seen in the RssB-IraD complex), thereby coupling adaptor docking (to ClpX) with substrate release. This competitive docking to RssB would prevent futile interaction of ClpX with the IraD-RssB complex (which lacks a substrate). Finally, substrate recognition by RssB appears to be regulated by a key residue (Arg117) within the α5 helix of the N-terminal domain. Importantly, this residue is not directly involved in σ^S interaction, as σ^S binding to the R117A mutant can be restored by phosphorylation. Likewise, R117A retains the ability to interact with and activate ClpX for degradation of σ^S, both in the presence and absence of acetyl phosphate. Therefore, we propose that this region of RssB (the α5 helix) plays a critical role in driving interaction with σ^S at a distal site.

Pore-forming proteins are weapons often used by bacterial pathogens to breach the membrane barrier of target cells. Despite their critical role in infection important structural aspects of the mechanism of how these proteins assemble into pores remain unknown. Streptococcus pneumoniae is the world’s leading cause of pneumonia, meningitis, bacteremia and otitis media. Pneumolysin (PLY) is a major virulence factor of S. pneumoniae and a target for both small molecule drug development and vaccines. PLY is a member of the cholesterol-dependent cytolysins (CDCs), a family of pore-forming toxins that form gigantic pores in cell membranes. Here we present the structure of PLY determined by X-ray crystallography and, in solution, by small-angle X-ray scattering. The crystal structure reveals PLY assembles as a linear oligomer that provides key structural insights into the poorly understood early monomer-monomer interactions of CDCs at the membrane surface.

We developed a new approach to distinguish distinct protein conformations in live cells. The method, exposable tetracysteine (XTC), involved placing an engineered tetracysteine motif into a target protein that has conditional access to biarsenical dye binding by conformational state. XTC was used to distinguish open and closed regulatory conformations of Src family kinases. Substituting just four residues with cysteines in the conserved SH2 domain of three Src-family kinases (c-Src, Lck, Lyn) enabled open and closed conformations to be monitored on the basis of binding differences to biarsenical dyes FlAsH or ReAsH. Fusion of the kinases with a fluorescent protein tracked the kinase presence, and the XTC approach enabled simultaneous assessment of regulatory state. The c-Src XTC biosensor was applied in a boutique screen of kinase inhibitors, which revealed six compounds to induce conformational closure. The XTC approach demonstrates new potential for assays targeting conformational changes in key proteins in disease and biology.

Genetic variations of leucine‐rich repeat kinase 2 (LRRK2) are the major cause of dominantly inherited Parkinson disease (PD). LRRK2 protein contains seven predicted domains: a tandem Ras‐like GTPase (ROC) domain and C‐terminal of Roc (COR) domain, a protein kinase domain, and four repeat domains. PD‐causative variations arise in all domains, suggesting that aberrant functioning of any domain can contribute to neurotoxic mechanisms of LRRK2. Determination of the three‐dimensional structure of LRRK2 is one of the best avenues to decipher its neurotoxic mechanism. However, with the exception of the Roc domain, the three‐dimensional structures of the functional domains of LRRK2 have yet to be determined. Based on the known three‐dimensional structures of repeat domains of other proteins, the tandem Roc–COR domains of the Chlorobium tepidum Rab family protein, and the kinase domain of the Dictyostelium discoideum Roco4 protein, we predicted (1) the motifs essential for protein–protein interactions in all domains, (2) the motifs critical for catalysis and substrate recognition in the tandem Roc–COR and kinase domains, and (3) the effects of some PD‐associated missense variations on the neurotoxic action of LRRK2. Results of our analysis provide a conceptual framework for future investigation into the regulation and the neurotoxic mechanism of LRRK2.

C-Src kinase activity is regulated by phosphorylation of Y527 and Y416. Y527 phosphorylation stabilizes a closed conformation, which suppresses kinase activity towards substrates, whereas phosphorylation at Y416 promotes an elevated kinase activity by stabilizing the activation loop in a manner permissive for substrate binding. Here we investigated the correlation of Y416 phosphorylation with c-Src activity when c-Src was locked into the open and closed conformations (by mutations Y527F and Q528E, P529E, G530I respectively). Consistent with prior findings, we found Y416 to be more greatly phosphorylated when c-Src was in an open, active conformation. However, we also observed an appreciable amount of Y416 was phosphorylated when c-Src was in a closed, repressed conformation under conditions by which c-Src was unable to phosphorylate substrate STAT3. The phosphorylation of Y416 in the closed conformation arose by autophosphorylation, since abolishing kinase activity by mutating the ATP binding site (K295M) prevented phosphorylation. Basal Y416 phosphorylation correlated positively with cellular levels of c-Src suggesting autophosphorylation depended on self-association. Using sedimentation velocity analysis on cell lysate with fluorescence detection optics, we confirmed that c-Src forms monomers and dimers, with the open conformation also forming a minor population of larger mass complexes. Collectively, our studies suggest a model by which dimerization of c-Src primes c-Src via Y416 phosphorylation to enable rapid potentiation of activity when Src adopts an open conformation. Once in the open conformation, c-Src can amplify the response by recruiting and phosphorylating substrates such as STAT3 and increasing the extent of autophosphorylation.

The tumour-derived monoclonal IgM antibody PAT-SM6 specifically kills malignant cells by an apoptotic mechanism linked to the excessive uptake of plasma lipids. The mechanism is postulated to occur via the multi-point attachment of PAT-SM6 to the unfolded protein response regulator GRP78, located on the surface of tumour cells, coupled to the simultaneous binding of plasma low density lipoprotein (LDL). We prepared and characterised LDL and oxidized LDL using sedimentation velocity and small-angle X-ray scattering (SAXS) analysis. Enzyme-linked immunosorbent (ELISA) techniques indicated apparent dissociation constants of approximately 20 nM for the binding of LDL or oxidized LDL to PAT-SM6. ELISA experiments showed cross competition with LDL inhibiting PAT-SM6 binding to immobilised GRP78, while, in the reverse experiment, GRP78 inhibited PAT-SM6 binding to immobilized LDL. In contrast to the results of the ELISA experiments, sedimentation velocity experiments indicated relatively weak interactions between LDL and PAT-SM6, suggesting immunoabsorbance to the microtiter plate is driven by an avidity-based binding mechanism. The importance of avidity and the multipoint attachment of antigens to PAT-SM6 was further investigated using antigen-coated polystyrene beads. Absorption of GRP78 or LDL to polystyrene microspheres led to an increase in the inhibition of PAT-SM6 binding to microtiter plates coated with GRP78 or LDL, respectively. These results support the hypothesis that the biological action of PAT-SM6 in tumour cell apoptosis depends on the multivalent nature of PAT-SM6 and the ability to interact simultaneously with LDL and multiple GRP78 molecules clustered on the tumour cell surface.