Current approaches to inhibit nuclear receptor (NR) activity target the hormone binding pocket but face limitations. We have proposed that inhibitors, which bind to nuclear receptor surfaces that mediate assembly of the receptor's binding partners, might overcome some of these limitations. The androgen receptor (AR) plays a central role in prostate cancer, but conventional inhibitors lose effectiveness as cancer treatments because anti-androgen resistance usually develops. We conducted functional and x-ray screens to identify compounds that bind the AR surface and block binding of coactivators for AR activation function 2 (AF-2). Four compounds that block coactivator binding in solution with IC 50 ≈ 50 μM and inhibit AF-2 activity in cells were detected: three nonsteroidal antiinflammatory drugs and the thyroid hormone 3,3′,5-triiodothyroacetic acid. Although visualization of compounds at the AR surface reveals weak binding at AF-2, the most potent inhibitors bind preferentially to a previously unknown regulatory surface cleft termed binding function (BF)-3, which is a known target for mutations in prostate cancer and androgen insensitivity syndrome. X-ray structural analysis reveals that 3,3′,5-triiodothyroacetic acid binding to BF-3 remodels the adjacent interaction site AF-2 to weaken coactivator binding. Mutation of residues that form BF-3 inhibits AR function and AR AF-2 activity. We propose that BF-3 is a previously unrecognized allosteric regulatory site needed for AR activity in vivo and a possible pharmaceutical target.

We have studied the cytoskeletal association of intercellular adhesion molecule-1 (ICAM-1, CD54), an integral membrane protein that functions as a counterreceptor for leukocyte integrins (CD11/CD18). A linkage between ICAM-1 and cytoskeletal elements was suggested by studies showing a different ICAM-1 staining pattern for COS cells transfected with wild-type ICAM-1 or with an ICAM-1 construct that replaces the cytoplasmic and transmembrane domains of ICAM-1 with a glycophosphatidylinositol (GPI) anchor. Wild-type ICAM-1 appeared to localize most prominently in microvilli whereas GPI-ICAM-1 demonstrated a uniform cell surface distribution. Disruption of microfilaments with cytochalasin B (CCB) changed the localization of wild-type ICAM-1 but had no effect on GPI-ICAM-1. Some B-cell lines demonstrated a prominent accumulation of ICAM-1 into the uropod region whereas other cell surface proteins examined were not preferentially localized. CCB also induced redistribution of ICAM-1 in these cells. For characterization of cytoskeletal proteins interacting with ICAM-1, a 28-residue peptide that encompasses the entire predicted cytoplasmic domain (ICAM-1,478- 505) was synthesized, coupled to Sepharose-4B, and used as an affinity matrix. One of the most predominant proteins eluted either with soluble ICAM-1,478-505-peptide or EDTA, was 100 kD, had a pI of 5.5, and in Western blots reacted with alpha-actinin antibodies. A direct association between alpha-actinin and ICAM-1 was demonstrated by binding of purified alpha-actinin to ICAM-1,478-505-peptide and to immunoaffinity purified ICAM-1 and by a strict colocalization of ICAM-1 with alpha-actinin, but not with the cytoskeletal proteins talin, tensin, and vinculin. The region of ICAM-1,478-505 interacting with alpha-actinin was mapped to the area close to the membrane spanning region. This region contains several positively charged residues and appears to mediate a charged interaction with alpha-actinin which is not highly dependent on the order of the residues.

An analysis of chemical structures from several commercially available libraries of compounds is presented with a view of acquiring compounds for screening. The Jarvis−Patrick clustering method has been applied, using the MDL “keys” as structural descriptors. The nature of the MDL keys is examined in this context, some features of the clustering algorithm are discussed, and clustering statistics are presented.

Retroviral capsid proteins and replication enzymes are synthesized as polyproteins that are proteolytically processed to the mature products by a virus-encoded proteinase. We have purified the proteinase of human immunodeficiency virus (HIV), expressed in Eseherichia coli, to =90% purity. The purified enzyme at a concentration of "=20 nM gave rapid, efficient, and specific cleavage of an in vitro synthesized gag precursor protein. Purified HIV proteinase also induced specific cleavage of five decapeptide substrates whose amino acid sequences corresponded to cleavage sites in the HIV polyprotein but not of a peptide corresponding to a cleavage site in another retrovirus. Competition experiments with different peptides allowed a ranking of cleavage sites. Inhibition studies indicated that the HIV proteinase was inhibited by pepstatin A with an IC50 of 0.7 ,uM.The capsid and nonstructural proteins of all retroviruses, including human immunodeficiency virus (HIV), are synthesized as polyprotein precursors that are proteolytically processed to the mature viral proteins by a virus-encoded, virus-associated proteinase (for a review see ref. 1). Viral proteinases (PR; for the new nomenclature for common retroviral proteins see ref.2) have been purified from virions and biochemically characterized for a number of avian (3, 4) and mammalian (5-7) retroviruses. These enzymes share limited amino acid sequence homology with members of the aspartic proteinases (8,9) and invariably contain the sequence Asp-Thr(Ser)-Gly corresponding to the catalytic center of cellular aspartic proteinases. However, retroviral enzymes are much smaller than cellular aspartic proteinases and contain only a single homologous catalytic center.