Contact info:
University of California, Berkeley
Department of Molecular and Cell Biology
229 Stanley Hall # 3200
Berkeley, California 94720-3275
USA

Location: 418B Stanley Hall
Phone: (510) 642-8758
Lab Phones: (510) 642-8766/2-8797
Fax: (510) 643-9290
Email: tom@ucxray6.berkeley.edu

Web Site: UC Berkeley Department of Molecular and Cell Biology

Research Emphasis
Our goal is to elucidate the principles of protein interactions. The primary tools we use are physical biochemistry, molecular genetics, and X-ray crystallography. Our main biological targets are regulatory systems.

Coiled coils. Subunit oligomerization in numerous proteins is mediated by coiled coils. These structures, which are helical ropes of two or more strands, are especially simple models for studying protein interactions. Coiled coils occur in 3-4% of proteins that encompass a wide variety of physiological functions. Our previous studies of the structures of leucine zipper mutants led to the discovery of amino acid sequence patterns that distinguish two-, three- and four-helical coiled coils. The conformational switch is driven by the complementarity between the amino acid sequence and the characteristic packing spaces in the three alternate structures. We used these ideas to write computer programs that accurately predict oligomerization state from the amino acid sequence and to design coiled coils with new folds and binding functions. Our current work focuses on two simple questions: how do structural variations correlate with the diverse functions of coiled coils, and what determines the partners of coiled-coil sequences?

We initially tackled the problem of heterospecificity by analyzing natural heterodimer sequences to derive a set of coiled-coil pairing rules. We used these rules to design peptides complementary to the coiled coil of the medically important APC tumor suppressor, a protein that is mutated in most colon tumors. The designed peptides show pM affinity for the target, and they can be used in lieu of antibodies for Western blots and affinity purification. Many applications can be envisioned for this new technology. A current challenge is to generalize this approach to obtain sequences that bind other interesting targets. We are developing computational and genetic approaches to meet this design challenge. We are exploring the diversity of structure and function of coiled coils by determining the structures of a variety of coiled-coil proteins. To speed this work, we are creating a complete set of crystallographic automation tools.

Protein crystallography. To better understand specific regulatory processes, we are determining new protein structures. Current studies focus on the bifunctional transcriptional coactivator DCoH, the dimerization domain of the tumor suppressor APC, the allosteric enzyme ATCase, proteins involved in signaling through the TNF receptor class, and the RNA-binding protein Sex Lethal. The X-ray crystal structure of DCoH, for example, yielded the first high-resolution view of a transcriptional co-activator. Our cocrystal structure of DCoH bound to the dimerization domain of its target transcription factor, HNF-1, showed that the regulatory interaction involves a change in DCoH oligomerization. The structure also showed that the coactivator has a remarkable autoinhibitory architecture.

In another study of protein recognition and signaling, we determined the cocrystal. structure of a fragment of the TNF receptor associated factor, TRAF2, bound to a peptide from the CD40 receptor. This work revealed the unique, trimeric TRAF fold and the mode of receptor recognition. The structure also supported an elegant transmembrane signaling mechanism in which trimeric, extracellular ligands preorganize the receptors to bind simultaneously to three sites on the TRAFs. These and other structures are being determined to define the general mechanisms of biological regulation.

Publications