Our group uses X-ray crystallography to study the structures of integral membrane proteins.  About 30 percent of the human genome codes for membrane proteins, and a similar distribution is found in lower organisms.  Membrane proteins are difficult to work with, however, due to their low abundance in cells and to their preference to reside in a lipid bilayer.  Although a large number of pharmaceutical targets are membrane proteins, they currently represent less than 1 percent of all solved protein structures.

We study transporters embedded in the outer membranes of Gram-negative bacteria, which are surface accessible and therefore have the potential to be good vaccine or drug targets against infectious diseases.

One area of particular interest is the transport of small molecules and large proteins across the outer membrane by a single family of membrane proteins.  We focus on iron transporters from several bacterial pathogens. Iron is essential for bacterial proliferation: If iron uptake could be blocked, an infection could be eradicated.

So far, our structures have shown how iron transporters specifically recognize Fe3+ bound to small molecules such as enterobactin (a siderophore synthesized by Escherichia coli) and citrate.  Each transporter has a unique binding pocket for its preferred small molecule.  When the correct substrate binds, the transporter undergoes conformational changes that send a signal across the outer membrane and prepare the system for transport.

However, transport into the periplasm is complicated and involves another protein complex and energy in the form of protonmotive force.  We are still working to understand the actual transport process.

Even without knowing exactly how they function, we believe that these iron transporters may make good vaccine or drug targets because they are surface exposed and often antigenic.  We are currently testing this idea using an iron transporter from Yersinia pestis.  Y. pestis causes plague, and deletion of the gene encoding an iron transporter abolishes virulence in a mouse model of bubonic plague.  We recently solved the structure of the Y. pestis iron transporter in two states, alone and in complex with its cognate Fe3+-siderophore.

These structures allowed us to precisely define the binding pocket for the substrate.  The next step is to use computational methods to screen for small molecules that effectively compete with the natural substrate for binding.  This could lead to the design of novel antibiotics.

We also are collaborating with Joe Hinnebusch (NIAID), who is evaluating this protein and several others for a protective immune response in rat and mouse models of bubonic plague.  We hope that this work will identify new vaccine targets.

Recently, we extended our work on small-molecule transporters to ask how proteins are ferried across the outer membrane.  Some of the iron transporters that we study also facilitate the uptake of large protein toxins called colicins. Whether the transport mechanism is the same as found for small molecules or entirely different, we hope that our crystal structures will suggest answers.

We also have begun to study protein export through collaboration with Harris Bernstein (NIDDK).  Together we have solved the structure of the transporter domain of an autotransporter from O157:H7 E. coli.  This protein forms a transport channel similar to those found in iron transporters, but the secreted protein domain is very large.  Exactly how it gets to the bacterial cell surface is still a mystery.

Our next major goal is to solve outer membrane protein structures from mitochondria.  Most proteins residing in mitochondria are nuclear encoded and must be imported through a general protein-import channel.  We aim to solve structures of this channel and others to see how similar these proteins are to bacterial outer membrane proteins, and how the passage of proteins across membranes varies between these systems.