Molecular Chaperone Interactions with Steroid Receptors: an Update

Abstract

In the 3 yr since William Pratt and David Toft published their comprehensive review of chaperone interactions with steroid receptors (1 ), progress has been made on several fronts that gives a better appreciation for the range of functions that chaperones serve in mediating steroid signaling. Steroid receptors remain the best characterized examples of an ever growing assortment of cytoplasmic and nuclear proteins—diverse representatives from multiple signal transduction pathways—that rely on molecular chaperones for folding, stabilization, or functional modulation. Chaperone targets include multiple tyrosine and serine/threonine kinases (2 ), the arylhydrocarbon receptor (3, 4, 5, 6, 7, 8 ), the heat shock transcription factor (9, 10, 11 ), common p53 mutants (Ref. 12 and references therein), nitric oxide synthase (13, 14, 15 ), and telomerase (16 )—an impressive list of “hot topic” proteins. Often, investigators have restricted their interpretations of chaperone interactions as a simple reflection of folding insufficiency by the target protein. Supporting this view are two major facts: 1) chaperones generally function in overseeing protein folding processes and 2) it is commonly observed that the target signaling protein will fail to achieve its functional state and is more rapidly degraded by the proteolytic machinery when major chaperone interactions are disrupted. As with other targets, steroid receptors display functional instabilities when deprived of certain chaperones, and thus fit the mold of unstable protein target, but it is hoped that the reader will be convinced by recent findings that chaperone interactions with steroid receptors and, by extension, with unrelated signaling targets, serve a variety of functions that go beyond the simple explanation of folding insufficiency. In the absence of hormone binding or other activating signals, steroid receptors typically exist in heteromeric complexes with heat shock proteins (Hsp) and additional components of the molecular chaperone machinery. In a pioneering study by Toft and his colleagues (17 ), hormone-free chicken progesterone receptor (PR) was found to readily assemble with chaperones in rabbit reticulocyte lysate, and the resulting PR complexes were similar in chaperone composition to native PR complexes (17 ). Taking advantage of reticulocyte lysate as a cell-free assembly medium, further extensive characterizations have been undertaken by the laboratories of Toft and Smith, focusing on PR complexes, and by Pratt’s laboratory, studying glucocorticoid receptor (GR) complexes, to identify assembly components, to delineate biochemical processes, and to understand the functional consequences of receptor-chaperone interactions. As thoroughly reviewed by Pratt and Toft (1 ), amended by more recent studies (18, 19, 20, 21 ), and briefly summarized here, the assembly pathway for steroid receptor-chaperone complexes can involve at least 10 chaperone components, five of which are obligatory in vitro for maturation of hormone binding ability by GR and PR. Three of the obligate factors are constitutively expressed forms of Hsp70, DnaJ/Hsp40, and Hsp90. Two cochaperone proteins are additionally required: the Hsp90-binding protein p23 and Hsp70/Hsp90 organizing protein (Hop), which brings Hsp70 and Hsp90 together in a common complex. Two additional Hsp70 binding cochaperones, Hsp70 interacting protein (Hip) and BAG-1, have been identified in receptor complexes, but these are nonessential in minimal in vitro hormone-binding assays. There are four additional proteins, also nonessential in minimal in vitro assays, that bind Hsp90 and are recovered in native receptor complexes. These are the FK506-binding immunophilins FKBP52 and FKBP51, the cyclosporin-A-binding immunophilin cyclophilin 40 (Cyp40), and the protein phosphatase PP5. Each competes with Hop for binding Hsp90, and, like Hop, each contains a similar tetratricopeptide repeat (TPR) domain that mediates their competitive binding to Hsp90. As in many chaperone processes, ATP is required for assembly of receptor-chaperone complexes. Hsp70 and Hsp90 are both ATP-binding proteins with weak ATPase activity that is necessary for their chaperone functions and influences their interactions with various cochaperone proteins. For Hsp70, ATP hydrolysis and nucleotide exchange regulate its binding and release from misfolded substrates. Furthermore, Hop and Hip only associate with the ADP-bound form of Hsp70. For Hsp90, Hop associates preferentially with its ADP-bound form while p23 binds exclusively to ATP-bound Hsp90. In complete reticulocyte lysate, free receptor initially associates with Hsp70 and Hsp40. Hsp70 and Hsp40 can bind each other and often function in a coordinate manner to facilitate general protein folding processes in eukaryotes (Refs. 22, 23 ; recently reviewed in Ref. 24 ). This early step is dependent on Hsp70 ATPase activity and leads to the ADP-dependent association of Hip and Hop with Hsp70.