Protein quality control and the maintenance of proteome homeostasis are critical to cellular and overall health of an organism. Deficiencies in protein homeostasis have been shown to cause or progress numerous diseases, such as neurodegeneration and dementia, type 2 diabetes, peripheral amyloidosis, lysosomal storage disease, cystic fibrosis, cancer and cardiovascular disease. A network of several hundred proteins, notably molecular chaperones and their regulators, assist in de novo folding or re-folding.
Molecular chaperones are an evolutionarily conserved class of proteins that prevent aggregation and assist in the conformational maturation of other cellular proteins (referred to as client proteins). Heat shock proteins (Hsps) are a group of molecular chaperones that are ubiquitously expressed under non-stressed conditions and upregulated upon exposure to cellular stress, including elevated temperature. These chaperone proteins comprise 5-10% of all proteins in the cell. HSPs function as components of a complex that also include other chaperones, cochaperones, modulators of ATPase activity and various accessory proteins. While HSPs are vital for preventing promiscuous interactions between proteins and to ensure normal protein folding. Misfolded proteins are targeted for ubiquitin-proteasomal degradation. Under stress, the heat shock response enhances cell survival by preventing the accumulation of misfolded and aggregated proteins, and protects them from targeted degradation. In addition to their role in facilitating protein folding, HSPs are also required for the translocation of proteins across membranes, quality control in the endoplasmic reticulum, and normal protein turnover. They are also involved in the posttranslational regulation of signaling molecules, the assembly/disassembly of transcriptional complexes, and the processing of immunogenic peptides by the immune system.
Hsp90 is the most abundant heat shock protein and represents 1-2% of total cellular proteins in unstressed cells. It is unique from the other chaperones in that it does not play a major role in de novo polypeptide folding. Instead, it regulates post-translational maturation of many conformationally unstable substrates or client proteins, many of which are involved in oncogenesis. There are four human isoforms of Hsp90; the cytosolic isoforms Hsp90α and β, Grp94 (localized to the endoplasmic reticulum) and TRAP1 (localized to the mitochondria). Hsp90 facilitates the conformational maturation of Hsp90-dependent proteins via the Hsp90 chaperone cycle, in which the Hsp90 homodimer forms a large, multiprotein complex that relies upon co-chaperones, immunophilins, and partner proteins to fold nascent polypeptides, as well as the rematuration of denatured proteins. The Hsp90 heteroprotein complex folds these substrates through a series of conformational transitions at the middle and N-terminal domains of Hsp90 that are facilitated by ATP hydrolysis at the N-terminus. Inhibition of the Hsp90 protein folding machinery results in client protein ubiquitinylation and subsequent degradation via the proteasome, which can ultimately result in cell death.
HSP90 client proteins include steroid hormone receptors, receptor tyrosine kinases, cytosolic signaling proteins, and cell cycle regulators, some of which are involved in apoptosis and cell cycle regulation. Many Hsp90-dependent client proteins (e.g. ErbB2, B-Raf, Akt, steroid hormone receptors, mutant p53, HIF-1, survivin, telomerase, etc.) are associated with the six hallmarks of cancer. Therefore, oncogenic client protein degradation via Hsp90 inhibition represents a promising approach toward anticancer drug development.
Two alternative strategies for inhibiting the function of Hsp90 include disruption of the Hsp90 heteroprotein complex and disruption of the Hsp90 C-terminal dimerization domain. Disruption of the Hsp90 heteroprotein complex has emerged as an effective strategy to prevent client protein maturation without induction of the HSR. More specifically, disruption of interactions between Hsp90 and co-chaperones, such as Cdc37, or direct inhibition of cochaperones and immunophilins, such as p23, F1F0 ATP synthase and FKBP52, prevent the maturation of Hsp90 clients at concentrations that do not induce the HSR. Originally, small molecule inhibitors of Hsp90 were designed to perturb the ATPase activity located at the N-terminus and include derivatives of geldanamycin, radiciol and purine. N-terminal Hsp90 inhibitors are effective at inhibiting Hsp90 function and lead to anti-proliferative activity through client protein degradation; however, Hsp90 N-terminal inhibition also leads to induction of the Heat Shock Response (HSR). N-terminal inhibitors displace the Hsp90-bound transcription factor, Heat Shock Factor-1 (HSF-1). It has been observed that these inhibitors are selective at inhibiting HSP90 in cancer cells rather than in normal cells. In cancer cells, HSP90 is thought to participate in a multi-chaperone complex in which it is activated and has high ATPase activity, whereas HSP90 in normal cells is in an inactive, noncomplexed form. Whether HSP90 is involved in a complex likely determines the cell affinity for HSP90 inhibitors- cancer cells have an approximately 100-fold greater affinity for its inhibitors than normal cells, which leads to the accumulation of the drug within tumors. HSP90 inhibitors may be more effective at treating breast and other forms of cancer due to HSP90 being a highly conserved, constitutively expressed protein that is also overexpressed in breast cancer cells.
Geldanamycin and radicicol, two unrelated natural products, were first shown to bind with higher affinity than natural nucleotides to prevent the chaperone from cycling between its ADP-and ATP-bound conformations, but these agents are either highly toxic or inactive in vivo. This led to the development of synthetic derivatives of geldanamycin demonstrating less toxicity that act as HSP90 inhibitors. In more recent years, several inhibitors have been developed. These include both ansamycin and non-ansamycin-derived inhibitors, including some of which are orally bioavailable. To overcome the toxicities and poor solubilities of past HSP90 inhibitors, numerous inhibitors have since been developed which are non-ansamycin based and display less toxicity. A number of ansamycin derivatives, gamitrinibs, have been designed to accumulate in the mitochondria. The rationale for these drugs is that current HSP90 inhibitors do not reach a pool of HSP90 in the mitochondria of tumor cells. These protected proteins can still function to promote cancer cell growth. Gamitrinibs have been shown to be well tolerated and less toxic in vivo compared to previous HSP90 inhibitors. It is interesting to note that HSP90 is heavily acetylated and that HSP90 hyperacetylation can inhibit its chaperone function. In addition, HSP90 is heavily phosphorylated. It is not known if such post-translational modifications of HSP90 can be therapeutic targets.