Current Cancer Drug Targets - Volume 3, Issue 5, 2003

In this issue of Current Cancer Drug Targets we are featuring a series of articles on the subject of Hsp90 Molecular Chaperone Inhibitors: Opportunities and Challenges. This is an exciting and timely topic. As tends to be the case these days in the best translational science, the biology of a hot area is still breaking at the same time as the first attempts at therapeutic exploitation are underway. This has advantages and disadvantages. On the one hand, the fastest possible translation of new science into novel therapies is to be encouraged. The potential for patient benefit can be realised more rapidly - and if the results are disappointing we need to know that quickly too so that the approach can be dropped and other interesting opportunities can be pursued. On the other hand, early entry into a new target area carries with it the risk that we may have insufficient knowledge to fully understand the system in which we are attempting to intervene therapeutically. However, on the whole, the benefits outweigh the downsides. The interdependence and mutual synergies of fundamental and translational research are illustrated very well by the example of Hsp90. The astonishing developments in the basic research on Hsp90 can help us to develop drugs acting on this target more effectively. Equally well, the natural product inhibitors of Hsp90 have proved to be invaluable tools with which to study the function of this vitally important molecular chaperone. As we continue to gain new insights into the role of Hsp90 in evolution, development, chromatin biology and epigenetics, what makes this topic particularly timely is the fact that the Phase I clinical trials with the first-in-class Hsp90 inhibitor, 17AAG, are now coming to completion. The results have proved promising enough for Phase II clinical trials to be planned. In addition, the clinical studies with 17AAG, together with the biological underpinning, have reinforced the enthusiasm of many groups to develop novel Hsp90 inhibitors. The articles in this special issue capture the sense of excitement in the Hsp90 field and the therapeutic opportunities presented by Hsp90 inhibitors. In particular these agents offer the potential of broad spectrum activity against a wide range of different tumours through their combinatorial effects on many different oncogenic proteins and pathways and their action on all of the hallmark traits of malignancy. At the same time there are many challenges ahead to realise the full therapeutic potential of Hsp90 inhibitors. It will require clinicians and scientists from many different disciplines to work together to achieve this objective. The articles in this special issue are designed to encourage this collaborative activity. I would like to thank Chloe Mayo in my office for coordinating the assembly of this special issue and John Buolamwini and his colleagues on the production side for their help and support. I would also like to thank all the contributors for their enthusiastic participation.

The Hsp90 molecular chaperone has emerged as one of the most exciting targets for cancer drug development. Hsp90 is overexpressed in many malignancies, very likely as a result of the stress that is induced both by the hostile cancer microenvironment and also by the mutation and abberant expression of oncoproteins. A particularly attractive feature of Hsp90 as a cancer drug target is that it is required for the conformational stability and function of a wide range of oncogenic ‘client’ proteins, including c-Raf-1, Cdk4, ErbB2, mutant p53, c-Met, Polo-1 and telomerase hTERT. Inhibition of Hsp90 should therefore block multiple mission critical oncogenic pathways in the cancer cell, leading to inhibition of all the hallmark traits of malignancy. This combinatorial blockade of oncogenic targets should give rise to board spectrum antitumour activity across multiple cancer types. The ‘druggability’ of Hsp90 was confirmed by the discovery that the natural products geldanamycin and radicicol, which have anticancer activity, exert their biological effects by inhibiting the essential ATPase activity associated with the N-terminal domain of the protein. The first-inclass Hsp90 inhibitor has entered clinical trial and provided proof of concept that Hsp90 can be inhibited and clinical benefit seen at non-toxic doses. Further development is underway and a related analogue 17DMAG also shows promise in preclinical models. In addition, novel Hsp90 inhibitors have been identified using methods such as high throughput screening and x-ray crystallography. The opportunities and challenges involved in translating the fast moving biology of Hsp90 into patient benefit is discussed.

Understanding the mode of action of Hsp90 requires that molecular detail of its interactions with client proteins and co-chaperones are known. The structure determination of the N-terminal domain of Hsp90 / Hsp90β, proof that it is an ATPase, that this activity is regulated and the identification of cochaperones that facilitate Hsp90 function were landmarks towards understanding conformational changes in Hsp90 brought about by ATP, co-chaperones and client proteins. Sti1 and Cdc37 / p50, which associate with early Hsp90 complexes, were shown to be inhibitors of Hsp90 ATPase activity and therefore promote its ‘open’ state, whereas Sba1 / p23, which associates with mature complexes, inhibits ATPase activity and stabilises the ‘closed’ state. The isolation and characterisation of Aha1, the only known strong activator of Hsp90 ATPase activity, which promotes the ‘closed’ state of Hsp90, will also be of major importance in understanding Hsp90 function. The structure determination of the middle region of Hsp90 has shed further light on the complex ATP-cycle of Hsp90, identifying a catalytic loop, with key residues that are essential for ATP hydrolysis. These studies, together with biochemical ones, suggest that ATP hydrolysis, is dependent on a complex rate-limiting step, involving N-terminal dimerization and association of the middle region, and therefore the catalytic loop, of Hsp90 with the N-terminal domains. The structure of the middle region of Hsp90 will also accelerate our understanding of client protein interactions since this region is implicated in their recognition and in particular their active-site openings.

The currently used Hsp90 inhibitors, geldanamycin, herbimycin A and radicicol, were isolated many years ago from Streptomyces and fungi originally for their antiprotozoal activity, herbicidal activity and antifungal activity, respectively. In the mid 1980s, it was found that the benzoquinone ansamycin antibiotics (herbimycin A, geldanamycin, and macbecin) reversed v-Src transformed cells to normal phenotypes, and Bcr-abl was subsequently suggested to be the molecular target for the treatment of chronic myelogenous leukemia through a study using herbimycin A for its selective antioncogenic activity. In 1994, these ansamycins were found to bind to Hsp90 and to cause the degradation of client proteins including Src kinases; further efforts to develop anticancer drugs were made using geldanamycin analogs, and 17AAG was chosen as the best candidate for clinical trials. The number of novel natural products isolated from microbial origins is continuing to increase and is doubling every 10 years. Thus, screening of bioactive substances from natural origins, using assays including defined targets, and developing leads toward drugs via optimized derivatization is a conventional but still promising strategy for drug discovery and development.

HSP90 inhibitors such as 17AAG have the major therapeutic advantage that they exert downstream inhibitory effects on multiple oncogenic client proteins. They therefore block several mission critical cancercausing pathways and have the potential to modulate all of the hallmark biological features of malignancy. Consistent with this combinatorial anti-oncogenic profile, 17AAG exhibits broad-spectrum antitumour activity against cultured cancer cell lines and in vivo animal models. However, there are clear differences in sensitivity between various cancer cell lines and it is quite possible that some tumour types or individual patients will be more responsive in the clinic than others. We describe the methods used to investigate the genes and proteins involved in the mechanism of action of HSP90 inhibitors and discuss the significance of these for cellular sensitivity. Methods used involve the conventional cell and molecular biology techniques, together with the more recent application of high throughput global technologies such as gene expression microarrays and proteomics. Selected examples that seem to play a role in sensitivity to HSP90 inhibitors are highlighted and the potential relevance to the response of cancer patients is discussed. Important determinants of response include: 1) Dependence upon key HSP90 client proteins such as ERBB2, steroid hormone receptors and AKT / PKB; 2) Levels of HSP90 family members and co-chaperones, such as HSP70 and AHA1; and 3) expression of various cell cycle and apoptotic regulators. In the case of 17AAG, metabolic enzymes such as NQO1 and membrane efflux pumps are also important for sensitivity.

The molecular chaperone heat shock protein 90 (Hsp90) is required for stability and function of multiple mutated, chimeric, and over-expressed signaling proteins that promote cancer cell growth and / or survival. It is also critical for the function of many normally expressed proteins, including protein kinases, steroid receptors and other transcription factors, and it may protect the cell from incapacitating or deleterious mutations. The recent identification of a nucleotide binding pocket within the first 220 amino acids of the protein, together with the discovery that at least two structurally distinct classes of antibiotic can replace nucleotide at this site and alter chaperone activity, has deservedly focused attention on Hsp90's amino terminus as an important regulator of function. However, data continue to accumulate pointing to the Cterminal half of the chaperone as an equally important regulator of activity, and small molecules that bind to this portion of Hsp90 have been identified.

In their role as molecular chaperones, heat shock proteins serve as central integrators of protein homeostasis within cells. As part of this function, they guide the folding, assembly, intracellular disposition and proteolytic turnover of many key regulators of cell growth, differentiation and survival. Not surprisingly then, heat shock proteins are over expressed in many types of cancer, and induction of the stress response may actually be required for cells to tolerate the genetic disarray characteristic of malignant transformation. Regulation of heat shock protein levels via the stress response is complex, but recent data indicate that the molecular chaperone Hsp90 plays a key role. Specifically, Hsp90 inhibitors alter the multi-chaperone complexes associated with Heat Shock Factor 1 (HSF1), the dominant transcription factor controlling induction of the stress response, and stimulate HSF1-activated heat shock gene expression. Induction of this heat shock response has now emerged as an important consideration in the further clinical development of Hsp90 inhibitors for several reasons. First, tumors in which the stress response is compromised appear particularly sensitive to Hsp90 inhibition. Second, induction of the stress response by Hsp90 inhibitors provides a sensitive pharmacodynamic endpoint with which to monitor drug action in individual patients. Third, Hsp90 inhibitors display important therapeutic interactions with both conventional DNA-targeted chemotherapeutics and newer molecularly targeted agents. These interactions are, at least in part, due to modulation of the stress response by these drugs. Lastly, stress response induction by Hsp90 inhibitors may have therapeutic benefits in non-neoplastic disorders such as heart disease, stroke and neurodegenerative diseases. These benefits are just beginning to be explored.

Radicicol, a macrocyclic antibiotic produced by fungi, was originally isolated many years ago, and was described as tyrosine kinase inhibitor. We also rediscovered radicicol as an inhibitor of signal transduction of oncogene products, such as K-ras and v-Src, using yeast and mammalian cell-based assays. In a study of mechanisms of action, it was revealed that radicicol depletes the Hsp90 client signaling molecules in cells, and thus inhibit the signal transduction pathway. In addition, direct binding of radicicol to the Nterminal ATP / ADP binding site of Hsp90 was shown, and thus radicicol has been recognized as a structurally unique antibiotic that binds and inhibits the molecular chaperone Hsp90. Although radicicol itself has little or no activity in animals because of instability in animals, its oxime derivatives showed potent antitumor activities against human tumor xenograft models. Hsp90 client proteins were depleted and apoptosis was induced in the tumor specimen treated with radicicol oxime derivatives. Taken together, these results suggest that the antitumor activity of radicicol oxime derivatives is mediated by binding to Hsp90 and destabilization of Hsp90 client proteins in the tumor. Among Hsp90 clients, we focused on ErbB2 and Bcr-Abl as examples of important targets of Hsp90 inhibitors. Radicicol oxime showed potent antitumor activity against ER negative / ErbB2 overexpressing breast cancer and Bcr-Abl expressing CML. Putative mechanisms of action and future directions of radicicol oxime against these kinds of tumor are discussed.

The Hsp90 chaperones play a key role in regulating the physiology of cells exposed to environmental stress and in maintaining the malignant phenotype in tumor cells. Agents that interfere with the function of the chaperone may thus be beneficial in the treatment of cancers. The ansamycins (geldanamycin and herbimycin) and the unrelated natural product radicicol were found to bind to the N-terminal pocket of Hsp90 and inhibit its function. However, translation of these compounds to the clinic was impeded by stability and hepatoxicity issues. 17AAG, a derivative of geldanamycin, was found to be less hepatotoxic and is currently undergoing Phase I clinical trial. Unfortunately, 17AAG is insoluble, difficult to formulate and it is not yet clear if therapeutically effective doses can be administered without escalating non-Hsp90 associated toxicities. Additionally, for reasons not yet completely understood, a subset of tumor cells are insensitive to the action of the drug. The development of novel agents that lack the drawbacks of the natural products is thus necessary. Here we present an overview of such efforts with focus on a new class of purine-scaffold Hsp90 inhibitors developed by rational design.

17-allylamino, 17-demethoxygeldanamycin (17AAG; NSC 330507) is the first modulator of heat shock protein 90 (Hsp90) to enter clinical trials. Hsp90 serves a chaperone role to properly fold and deliver client proteins to appropriate intracellular locations. Interest in Hsp90 modulators for the experimental therapeutics of cancer has arisen based on pre-clinical evaluations suggesting that Hsp90 client proteins regulate signaling pathways critical to the molecular economy of many types of tumors, including oncogene signaling, cyclin-dependent kinase activation, steroid hormone receptors, and mediators of invasion and metastasis. Thus, Hsp90-directed agents could affect molecules upon which tumors depend for their proliferation and survival. Initial clinical studies have therefore sought to incorporate assessment of these endpoints into initial clinical evaluations. Three schedules of administration have been supported for initial evaluation in Phase I studies sponsored by the National Cancer Institute (NCI) or supported by NCI and sponsored by Cancer Research UK. In the daily times five schedule, a recommended Phase II dose (RPTD) of 40 mg / m2 has been reached, while once weekly or three of four weekly schedules are defining RPTDs of 295 and 308 mg / m2. Toxicity is tolerable and appears dominated by hepatic, gastrointestinal, and constitutional symptoms. Concentrations of drug at peak of ∼1700-3000 nM are concordant with concentrations predictive of useful outcomes in pre-clinical model systems. Evidence of modulation of Hsp90 partner molecules has been obtained in both surrogate and some tumor compartments. These very early results encourage additional clinical evaluations of 17AAG and related molecules.

The potential clinical applications of the prototype first-in-class Hsp90 inhibitor 17AAG and other emerging Hsp90 drugs are very exciting. Rigorously planned and executed clinical trials, incorporating measurement of appropriate biomarkers and pharmocodynamic endpoints are critical for selecting the optimal dose and schedule. A detailed understanding of the molecular mode of action of Hsp90 inhibitors alongside the elucidation of the molecular pathology of individual cancers will help us to identify tumour types and individual patients that will benefit most from treatment. Careful in vitro and in vivo experiments are needed to choose the most potentially advantageous combination studies. It is important to construct a pharmacologic audit trail linking molecular biomarkers and pharmacokinetic and pharmacodynamic parameters to tumour response endpoints. Phase I clinical studies with 17AAG have shown that the drug can be given at does that are well tolerated and that also achieve active pharmacokinetic exposures and elicit molecular signatures of gene and protein expression that are indicative of Hsp90 inhibition. Furthermore, examples of disease stabilisation have been documented, consistent with the generally cytostatic responses that are seen in animal models. Selecting tumour types for Phase II clinical trials must involve balancing 1) our knowledge of molecular response determinants, such as the expression of and dependence upon key client proteins and 2) more pragmatic evidence of antitumour activity in the relevant preclinical models. Examples of likely disease targets include chronic myeloid leukaemia, melanoma, breast, ovarian, brain, thyroid, colorectal and prostate cancer.