The hsp90 family of molecular chaperones are key players in the conformational maturation and folding of a variety of cellular client proteins, including steroid receptors, kinases, Toll-like receptors, and G-proteins. Inhibitors of hsp90s, such as gendanamycin, are potent anti-tumor compounds because of their inhibitory effect on client protein maturation, and are the subject of intense pharmacological interest. The mechanism by which hsp90 chaperones act to mature client proteins is poorly understood, however, our group has recently determined the structure of the N-terminal domain of the endoplasmic recticulum hsp90, called GRP94, in complex with a variety of inhibitory and naturally-occuring ligands. These structures have revealed a ligand-dependent conformational switch that is unique to GRP94 and which may explain the mechanism by which the chaperone responds to the changes in ATP levels that result from cellular stress. The structure of GRP94 also identified a novel ligand binding pocket that allows the binding of ligands that are selective for GRP94 and not the 3 other cellular hsp90 paralogs. This pocket is currently being exploited in the design and synthesis of novel GRP94 inhibitors. These compounds would not only have therapeutic potential but should also prove useful for dissecting the cellular roles of the individual hsp90s.

X-ray Crystallographic Studies of Androgen Receptors

The androgen receptor (AR) is the key cellular mediator of prostate cancer, the most common form of cancer to afflict western men after lung cancer. Late stage prostate cancers are often associated with androgen insensitivity, which renders standard anti-androgen therapies ineffective. The androgen receptor is a transcription factor that binds not only androgens but specific DNA targets as well. The AR belongs to the steroid subclass of hormone receptors, whose other members include the glucocorticoid, mineralocorticoid, and progesterone receptors. One longstanding puzzle about these receptors has centered on the fact that their consensus DNA targets are identical. This raises the question of how the androgen receptor is able to selectively activate androgen-responsive genes. Recently androgen specific DNA targets that differ in their geometry from the canonical steroid response element have been identified, and our group has now determined the structure of the AR DBD bound to a selective DNA target. The structure explains how AR forms the dimeric interactions that allow it to bind to these elements while the other steroid receptors cannot. Compounds that specifically interfere with these protein-protein interactions may constitute a new approach to anti-androgen therapy. Ongoing studies are now examining the N-terminal activation domain of the receptor, which may form tertiary interactions with the other domains of the receptor and allow constituitive gene activation. This analysis should explain the hormone independent activity of the AR and may lead to the development of novel anti-androgens that do not target the mutagenically sensitive hormone binding domain.

X-ray Crystallographic Studies on Nuclear Hormone Receptors

Nuclear hormone receptors are transcription factors that play a central role in human growth and development. The vitamin D receptor (VDR) is a member of this family and is classically responsible for the regulation of genes involved in calcium homeostasis. Recent work, however, has also identified a role for VDR in the expression of proteins that detoxify some cancer-triggering chemicals that are released during the digestion of high-fat foods. This correlates with recent reports that vitamin D treatment lowers the incidence of colon cancers. A more general role for VDR in the etiology of cancer is suggested by earlier reports that treatment with 1,25 dihydroxyvitamin D, the functional ligand for VDR, leads to the differentiation of neoplastic cells. Our lab has determined the crystal structure of the VDR DNA binding domain bound to a series of VDR response elements. Our 2002 study revealed the structural basis for selective DNA binding by this receptor. We have also identified the C-terminal extension (CTE) of the receptor as a key player in DNA binding and target selection. Current work from this group now seeks to identify the role of the CTE in receptor activity, as a prelude to the determination of the structure of the intact, full-length receptor. These studies should lead to a fuller understanding of the mechanism by which VDR regulates genes that are important to fighting cancer, and to the eventual development of VDR-directed anti-cancer therapies.