|
| |||||
|
| |||||
|
|
|
|
|
|
|
Research Interests
Our interests focus on defining the molecular logic of regulatory circuits in physiology and disease. An immediate challenge is to understand how protein interactions control biochemical reactions. To investigate the fundamental problems of molecular recognition and signaling, the primary tools we use are X-ray crystallography, molecular biology, genomics and physical biochemistry.
Current Projects
- Our attack on the principles of regulation employs two distinct strategies. To explore the diversity of regulatory mechanisms, we are determining the crystal structures and biochemical properties of signaling proteins, allosteric enzymes and transcriptional enzymes. To explore the fundamental aspects of intramolecular signaling and protein interactions, we are developing new methods to model structural ensembles from crystallographic data.
Regulatory mechanisms in infectious disease. Structural studies lie at the heart of our exploration of regulatory mechanisms. To speed crystallographic work, we developed a program called Elves, which automates the computational steps of X-ray structure determination (Alber & Holton, 2004). In addition, we built collaboratively a shared X-ray beamline at the Lawrence Berkeley National Laboratory. We routinely use this facility for studies of protein complexes and for an international collaboration to explore the structural genomics of tuberculosis.
A major goal is to understand the functions of Ser/Thr/Tyr protein phosphorylation in Mycobacterium tuberculosis (Mtb). This pathogenic bacterium infects one third of the world's population and kills over two million people annually, more than any other infectious disease. Through reversible protein phosphorylation, protein kinases and phosphates provide the fundamental machinery for environmental sensing and physiological signaling. In addition to the traditional two-component signaling systems Mtb expresses 11 "eukaryotic-like" Ser/Thr protein kinases (STPKs) and four protein phosphates that are candidate mediators of developmental changes and host interactions. The biological functions of these systems are largely unknown.
We used a combination of structural and biochemical methods to establish a framework for understanding the Mtb systems for Ser/Thr/Tyr protein phosphorylation and dephosphorylation. We determined the first sturctures of eight distinct signaling modules, which in concert with biochemical studies, support the central idea that universal mechanisms of regulation and substrate recognition govern the functions of prokaryotic and eukaryotic STPKs (Young et al., 2003; Good et al., 2004; Gay et al., 2006). These studies also revealed novel mechanisms of regulation, including dimerization of STPK domains and protection of a protein phosphatase active site from oxidative inactivation (Grundner et al., 2005). Studies of the structure and activitiy of the Mtb PstP phosphatase, which antagonizes the kinases, provide a new basis for understanding PP2C phosphatases (Pullen et al., 2004). Analysis of mutants of the STPK, PknB, show that catalytic activity is under dynamic control and suggest that the STPKs provide tractable targets for chemical inhibitors. Future work focuses on the function and localization of the Mtb STPKs and protein phosphatases in vivo, as well as studies of substrate interactions and microarrays to probe physiological functions.
Protein polymorphism. Protein recognition and function depend on a balance between structural rigidity and flexibility. We have developed two new methods for defining conformational polymorphism in protein structures. Mathematical analysis of dozens of structures in the Protein Data Bank suggests that native proteins adopt a large number of specific, alternate conformations that previously have escaped detection. Systematic sampling of low level electron density in the neighborhood of protein side chains in high-resolution structures reveals patterns corresponding to the presence of small populations of additional rotamers of most side chains. Methyl groups also showed a tendency to adopt specific, "rotameric" conformations that reflected the presence of barriers to rotations. Our results indicate that native protein structures generally are much less unique than currently believed. The alternate native conformations provide specific reservoirs of structural polymorphism that can facilitate folding, function and evolution. Coupled shifts may reflect intramolecular signaling pathways while uncoupled conformations contribute to the residual entropy of the native state. Modeling these ensembles and exploring their functional significance represents a fruitful direction for ongoing work.
