Antibiotic resistance has become a major clinical problem worldwide. Our lab is interested in the structure-based design of inhibitors that target antibiotic resistance mechanisms or novel targets essential to bacterial pathogenesis. To illuminate the molecular basis of antibiotic resistance and pathogenicity afforded by the proteins involved, we use a multidisciplinary structural biology approach that relies on x-ray crystallography, NMR, and electron microscopy complemented by phenotypic analysis in vivo. Many of the bacterial proteins and protein complexes we study are membrane associated, adding to the challenge of achieving their structural, biochemical, and in vivo characterization. Collectively, the information we glean from our work provides the molecular foundation for the design of compounds that specifically inhibit resistance- or pathogenicity-inducing bacterial targets.
We are focusing on several bacterial mechanisms that facilitate broad-spectrum resistance to the beta-lactam antibiotics such as penicillin and cephalosporins, mechanisms that operate via hydrolysis, efflux, or acquisition of altered and drug-insensitive targets. For example, the key determinant of broad-spectrum beta-lactam resistance in the superbug methicillin-resistant Staphylococcusaureus (MRSA) is the membrane-spanning, penicillin-binding protein 2a (PBP2a), a transpeptidase that is required for the production of peptide cross-links that give the bacterial cell wall its necessary strength and rigidity. Because of its low affinity for beta-lactams, PBP2a provides cross-linking transpeptidase activity at beta-lactam concentrations that inhibit the other cell wall transpeptidases; such peptidases are normally produced by S. aureus and other pathogenic bacteria. We determined, to 1.8 Å resolution, the crystal structures of native MRSA PBP2a as well as those of acyl-enzyme complexes with various beta-lactam antibiotic substrates. Analysis of PBP2a's active site reveals the structural basis of its broad-spectrum resistance to the clinically utilized beta-lactam antibiotics and identifies features important for high-affinity binding. This information was used in structure-based inhibitor design strategies aimed at combating MRSA resistance.
We are also pursuing a remarkable signaling system that controls expression of PBP2a in MRSA in an antibiotic-dependent manner. MecR, the responsible membrane-spanning receptor, senses beta-lactam antibiotic concentrations by selective binding to its periplasmic domain, and it signals that binding to a putative metalloprotease domain on its cytoplasmic face via its transmembrane components. The protease domain is believed to relieve repression of the gene that encodes the antibiotic-resistant PBP2a through a proteolytic cleavage event. We are using a combination of structural and in vitro and in vivo mutagenesis studies to elucidate the molecular details of this resistance factor in MRSA and other Gram-positive bacterial pathogens. Recently, we also determined the structure of the first bifunctional cell wall–cross-linking enzyme PBP2, a class A PBP, also from S. aureus, that cross-links the peptide and sugar building blocks of the cell wall. We analyzed the structure of this membrane-anchored enzyme in the presence of the substrate analogue inhibitor moenomycin, a highly potent antibiotic commonly used in animal feed. Our structures provide the first insight into the molecular basis of cell wall cross-linking by these transpeptidase/glycosyltransferase bifunctional enzymes as well as the essential features of moenomycin inhibition. This structure-based information is being used to create modified forms of moenomycins and new antibiotic leads for treating human infections.
In terms of novel targets, our laboratory has made significant progress in the structural elucidation of essential proteins and subcomplexes that comprise the type III secretion apparatus (T3SS) common to many Gram-negative pathogens, including enteropathogenic E. scherichia coli, Salmonella, Shigella, Bordetella, Chlamydia, and Pseudomonas. The "needle-like" T3SS allows specific and direct injection of bacterial virulence proteins into the cytoplasm of human host cells, where they mediate a wide range of pathogenic effects by manipulating host cytoskeletal proteins, signaling machinery, and other core cellular processes. Whereas the injected virulence proteins differ from bacterial pathogen to pathogen, the T3SS apparatus is highly conserved and thus presents an excellent target for the design of antimicrobials that disable only pathogenic bacteria, leaving "good bacteria" to continue to colonize in the patient. The T3SS is composed of approximately two dozen proteins that created an oligomerized set of membrane-spanning rings and connecting hollow filaments reaching from the bacterial cytoplasm to the host cytoplasm. Our laboratory has provided some of the first high-resolution structures and a biochemical characterization of the proteins that make up the T3SS complex, including the x-ray crystallographic and cryo–electron microscopy models of the EspA translocation tube, a 90-nm hollow filament that bridges the gap between bacterial outer membrane and host membrane and ensures selective secretion of bacterial virulence proteins into the host cell. Our structural data show that the filament has an inner dimension of about 25 Å, suggesting that the proteins passing through are in an unfolded or partially unfolded state. We have also obtained x-ray crystallographic, isothermal titration calorimetric, and in vivo localization, secretion, and virulence data on native and mutant forms of the secretion pilot protein, a barrel-shaped lipoprotein that facilitates correct targeting, oligomerization, and insertion of the large outer-membrane secretion ring of the T3SS into the bacterial outer membrane.
An inner membrane ATPase is also a hallmark of the T3SS, and we recently determined, for the first time, the structure of this enzyme at high resolution. In the presence of ATP analogues, the hexameric structure shows remarkable similarity to aspects of the functionally distinct energy-transducing F1-ATPase. In addition to answering many questions, solving the structure raises new ones about the unique modifications that allow the T3SS ATPase to perform its role in the specific secretion of bacterial virulence proteins. We also obtained high-resolution structural information on the 24-subunit inner-membrane polymeric ring that is thought to act as the initial "platform" upon which the ATPase and other type III components are thought to assemble. Using mass spectrometry and targeted labeling of isolated needle complexes in native and mutant forms, we are beginning to unravel the specifics of the many protein-protein interactions that stabilize the needle complex in vivo. In conjunction with the ongoing pursuit of higher-resolution cryo–electron microscopy images of the intact needle complex, biochemical and in-vivo data provide the foundation for understanding the molecular details of this pathogenic process as well as for the design of novel antimicrobials that combat the clinically notorious bacteria that encode them.
Our studies are also supported by grants from the Canadian Institutes of Health Research, the Canadian Foundation for Innovation, and the Michael Smith Foundation for Health Research.
Last updated March 2007
