¹National Cancer Institute, NIH, Bethesda, MD, USA

²Department of Medicine, New York Medical College, Valhalla, NY, USA

Correspondence to: M V Blagosklonny, NIH, Bldg 10, R 12 N 226, Bethesda, MD 20892, USA; Fax: 301 402 0172

Geldanamycin (GA), herbimycin A and radicicol bind heat-shock protein-90 (Hsp90) and destabilize its client proteins including v-Src, Bcr-Abl, Raf-1, ErbB2, some growth factor receptors and steroid receptors. Thus, Hsp90-active agents induce ubiquitination and proteasomal degradation of numerous oncoproteins. Depending on the cellular context, HSP90-active agents cause growth arrest, differentiation and apoptosis, or can prevent apoptosis. HSP-active agents are undergoing clinical trials. Like targets of most chemotherapeutics, Hsp90 is not a cancer-specific protein. By attacking a nonspecific target, HSP-90-active compounds still may preferentially kill certain tumor cells. How can this be achieved? How can therapeutic potentials be exploited? This article starts the discussion.

Leukemia (2002) 16, 455-462. DOI: 10.1038/sj/leu/2402415

molecular therapeutics; geldanamycin; oncogenes; heat shock proteins

Introduction

When a half-century ago, anticancer drugs were introduced into clinical practice, their mechanisms of action were not fully elucidated. These chemotherapeutic agents attack DNA, inhibit nucleotide metabolism and supress microtubule function. Yet, conventional chemotherapy can cause remissions and even can cure certain malignancies, such as childhood leukemia and testicular cancer. A conceptual basis for standard chemotherapy was inhibition of cycling and killing of dividing cells.¹ Unrestricted cell cycle is a hallmark of cancer.^2,3,4,5,6 However, the toxicity to normal cells (especially to proliferating cells) limits chemotherapy.^7,8,9

By the beginning of a new millennium, numerous molecular targets of mechanism-based anticancer drugs have been identified. These targets include growth factor (GF) receptors, mitogen-activated kinases, cyclin-dependent kinases and anti-apoptotic kinases such as Bcr-Abl and Akt.^{10,11,12,13,14} Although some of them are etiologic to cancer, they are not cancer-specific. With a few exeptions (eg Bcr-Abl), these oncoproteins also govern life and proliferation of normal cells. Besides, cancer cells usually acquire multiple genetic alterations.¹⁵ Parallel and redundant signaling pathways can support survival and growth of cancer cells.^6,11,12 Therefore, hitting one target may not be sufficient to kill a cancer cell. It has been suggested that in order to reverse the transformed phenotype, it is desirable to identify an agent capable of affecting multiple targets in signal transduction pathways.¹⁶ The concept of the multi-hit modality is emerging.^11,12

In light of the multi-hit concept, geldanamycin (GA) and other agents that target heat shock protein-90 (Hsp90) are 'wonder drugs'. The benzoquinone ansamycins GA and herbimycin A, antibiotics produced by yeast, were initially identified as tyrosine kinase inhibitors. However, as it was determined later, the inhibition of kinases by GA and herbimycin A is indirect. In 1986, it has been shown that benzoquinonoid ansamycins have no direct effect on the Src kinase, but instead 'destroy' the intracellular environment.¹⁷ In other words, GA and herbimycin A are not inhibitors of kinases. They target molecular chaperones: Hsp90 and related Grp94.¹⁸

Heat shock proteins

By definition, heat induces heat shock proteins (Hsps). Heat shock activates the synthesis of only a few proteins and strongly inhibits the synthesis of most others.¹⁹ Hsp with molecular mass 90 (Hsp90) is an abundant cytosilic protein in bacteria and eukaryotes, with homologues of Grp94 in higher eukaryotes. These two proteins are major targets for Hsp-active drugs.²⁰ Hsps are also referred to as 'molecular chaperones'. A chaperone protein helps other proteins to avoid misfolding pathways that produce inactive or aggregated states. Hsp90 acts in concert with other chaperones and partners (Hsp70, p23, HOP, p50/Cdc) to provide maturation and folding, as well as traficking and function of their client proteins (c-Raf, ErbB-2, steroid receptors). The complexity of multi-chaperone complexes has been extensively reviewed.^20,21 What is important for clinical applications is that inactivation of Hsp90 results in inappropriate functioning and rapid degradation of chaperone's client proteins. Several protein kinases, including Raf-1, ErbB-2, and Bcr-Abl depend upon the chaperone Hsp90 for proper function and stability.^20,21,22,23 The benzoquinone ansamycins GA and herbimycin A and the macrocyclic antifungal antibiotic radicicol bind to Hsp90 and specifically inhibit this chaperone's function, resulting in degradation of HSP90-associated proteins.^24,25,26,27

The proteasome and GA-induced degradation

The ubiquitin (Ub)-proteasome pathway is the major non-lysosomal pathway of proteolysis in human cells and accounts for the degradation of most short-lived proteins. Proteins are usually targeted for proteasome-mediated degradation by the covalent addition of multiple units of the 76 amino acid protein ubiquitin (Ub). Ubiquitinylated proteins are degraded by the 26S proteasome, a large protease complex. Normally, many short-lived proteins, such as cyclins or inhibitors of CDK kinases and wild-type p53, are rapidly degraded by ubiquitin-dependent proteolysis.^28,29

In contrast, Hsp90 prevents degradation of Hsp90 client proteins. By inactivating chaperone function, Hsp90-active drugs permit degradation of ErbB-2 and receptors of IGF, insulin, and EGF.^30,31 The enhanced degradation of receptors can be prevented by inhibitors of the 20S proteasome.³⁰ For example, within minutes of exposure to GA, mature ErbB-2 became polyubiquitinated.³¹ One can predict that the inhibition of the proteasome will result in accumulation of ubiquitinated proteins (Figure 1). Indeed, treatment of cells with lactacystin, a proteasome inhibitor, blocked GA-induced degradation of ErbB-2 and enhanced the accumulation of polyubiquitinated ErbB-2. Following GA and lactacystin treatment, a higher molecular weight form of ubiquitin-ErbB-2 conjugates was detected.³¹ Similarly, cotreatment with the proteasome inhibitor PS-341 reduced GA-mediated degradation of Bcr-Abl.³² It should be emphasized that inhibition of the proteasome is extremely toxic for a cell. Due to accumulation of certain proteins, inhibitors of proteasome induce apoptosis in leukemia and many non-leukemia cells.^33,34,35 GA and PS-341 antagonized each other's toxicity in leukemia cells that were transfected with Bcr-Abl.^32,36

Mutant p53 and geldanamycin

Normally, wt p53 is rapidly degraded by the proteasome.^28,37 Wt p53 transcriptionally induces Mdm-2, which in turn targets p53 for degradation by the proteasome.^38,39 Mutant p53 does not induce Mdm-2 and, therefore, mutant p53 is not degraded and is highly overexpressed. In brief, loss of function causes stabilization of mutant p53.⁴⁰ Ectopic Mdm-2 still targets mutant p53 for degradation.⁴¹

It has been shown, that GA causes degradation of mutant p53.⁴² Inhibition of Hsp90 leads to depletion of mutant p53, but not of wild-type p53 in leukemia, breast and prostate cell lines.⁴² GA restores p53 polyubiquitination and degradation of mutant p53 by the proteasome.^43,44 However, the mechanism of destabilization of mutant p53 appears to be different from the mechanism of destabilization of most GA-sensitive proteins. For one, there is no evidence that mature p53 binds Hsp-90. In contrast, Hsp90 participates in the achievement of the mutated conformation of nascent p53.⁴⁵ GA stimulates degradation of a newly synthesized protein only. In agreement, a rate of p53 depletion is slower than those for Raf-1 and ErbB2. Interestingly, mechanisms of depletion of mutant p53 and CFTR appears to be similar. Perturbation of Hsp90 interaction with nascent CFTR prevents its maturation and accelerates its degradation by the proteasome.⁴⁶ Although Mdm-2 plays a small role in the degradation of mutant p53 caused by GA, the alternative mechanism of degradation of mutant p53 still involves ubiquitination and the proteasome.⁴⁴

GA does not change levels of wt p53 and does not affect induction of wt p53 by DNA damage,⁴² but it prevents wt p53 accumulation caused by paclitaxel.⁴⁷

Mapping the network of oncoprotein's targets

G1/S cell cycle transition comprise a highly nonlinear network from activation of growth factor receptors to cyclin dependent kinases.^3,5 Folding and stability of numerous signaling proteins depend on the Hsp90 function.^20,21,22 Receptors of GF that are sensitive to GA-mediated degradation include IGF-I, EGF and PDGF receptors, and HER-2.^{30,31,48,49,50} Hsp90-active agents inactivate multiple kinases such as Src, Lyn, Lck, Raf-1 and Cdk-4.^18,51,52,53 Akt is affected either in a direct⁵⁴ or indirect manner (Figure 2). Inhibition of these pathways results in down-regulation of cyclin D1 and functional inactivation of Cdk-4 (Figure 2). In some cell types, cyclin D expression is dependent upon PI3-kinase and Akt which in turn are inhibited by Hsp90-active agents.⁵⁵

Cyclins D are growth factor sensors.⁵⁶ Growth factors regulate cyclin D1 by several mechanisms. (1) Transcriptional induction of cyclin D1 that is dependent on the Ras/Raf-1/Mek/ERK pathway.⁵ (2) Stabilization and accumulation of the cyclin D protein. In the absence of growth factor signaling, cyclin D1 is rapidly degraded by the proteasome. The pathway that sequentially involves Ras/PI-3 kinase/Akt prevents degradation of cyclin D1. (3) Translocation of cyclin D to the nucleus and its assembly with CDK-4 and CDK-6.⁵ All these pathways are blocked by Hsp90-active drugs (Figure 2). In addition, Hsp90-active agents destibilize CDK-4.⁵³

Like GA and its derivites, radicicol (a macrocyclic antifungal antibiotic) suppresses transformation caused by Src, Ras and Mos.^57,58 It has been shown that radicicol can inhibit Ras-induced activation of Erk-2.⁵⁹ Levels of Raf-1 are decreased in radicicol-treated cells, whereas levels of Ras and Erk-2 remain unchanged. Therefore, like GA, radicicol disrupts v-Src- and Ras-activated signaling pathways by selectively depleting the Raf kinase.^59,60 As it was discovered later, radicicol binds Hsp90 with consequent dissociation of the Raf/Hsp90 kinase complex, leading to the attenuation of the Ras/MAP kinase signal transduction pathway.⁶¹

Many other experimental therapeutics are aimed at these signaling network: growth factor receptors and tyrosine kinases, Ras, Mek, PI3-K and CDK.^62,63 However, downstream and/or parallel signaling pathways may render cancer cells resistant to growth inhibition. Hsp90-active agents destabilize multiple signaling oncoproteins (Figure 2). By blocking the upstream signaling, GA and herbimycin A inactivate non-target proteins such as Erk1/2.^32,51,64 Although Hsp90-active drugs do not directly target Ras, they block its upstream and downstream signaling pathways. By blocking the Raf-1/MEK pathway, GA abrogated phorbol ester-induced p21 in SKBr3 breast cancer cells⁶⁵ and in HL60 leukemia cells.⁶⁶

The following example illustrates the importance of targeting parallel pathways. Elevated levels of urokinase plasminogen activator-1 (uPA) and the IGF-I receptor are associated with breast cancer recurrence and decreased survival. IGF-I requires both PI-3K- and MEK-dependent pathways to optimally induce uPA expression. The production of uPA induced by IGF-I was blocked up to 90% by herbimycin A, but was blocked less potently by LY294002 (an inhibitor of PI-3K) or PD98059 (an inhibitor of MEK).⁶⁴

All Hsp90-active drugs inhibit the same signaling pathways. For example, the designed small molecule PU3, which competes with GA for Hsp90 binding, induces degradation of proteins, including HER-2, in a manner similar to GA. Furthermore, PU3 inhibits the growth of breast cancer cells causing Rb hypophosphorylation, G1 arrest and differentiation.⁶⁷

Glucocorticoid receptors

Activities of steroid hormones are mediated by the superfamily of nuclear receptors, which include those for steroid and thyroid hormones and retinoids. These receptors are ligand-dependent transcription factors that can stimulate gene expression and regulate proliferation, differentiation, and specific functions of target tissues. Unligated receptors exist in inactive complexes with chaperone proteins such as heat-shock proteins (HSPs).⁶⁸

The unliganded glucocorticoid receptor is a complex of a ligand-binding protein, HSP90, HSP70, HSP40, p23 and HOP.⁶⁹ Upon binding of glucocorticoids to their receptor, the complex moves along the microtubules to the nucleus. GA disrupts glucocorticoid receptor function.⁷⁰ GA impedes hormone-dependent GR translocation along microtubules.^71,72 A functional antagonism between Hsp90-active agents and glucocorticoids should be taken into account in the therapy of leukemia.

Estrogen and progesterone receptors

Estrogens and progestins control cell proliferation of mammary epithelium. Therefore, anti-estrogens is a tissue-specific therapy in breast cancer. Unstimulated estrogen and progesterone receptors exist as multimolecular complexes consisting of the hormone-binding protein itself and several essential molecular chaperones including Hsp90. Hsp90-active drugs (geldanamycin and radicicol) destabilize these hormone receptors in breast cancer cells.⁷³ In vivo, administration of 17-allylaminogeldanamycin (17-A-GA) to estrogen-supplemented, tumor-bearing mice resulted in marked depletion of progesterone receptor levels in both uterus and tumor. It also delayed the growth of hormone-responsive MCF-7 and T47D human tumor xenografts for up to 3 weeks after the initiation of therapy, suggesting that GA can be used in refractory breast cancer.⁷³

Mechanisms of growth inhibition

The mechanism of cytotoxicity caused by GA is the degradation of Hsp90-associated proteins. There is a perfect correlation between down-regulation of Hsp90's client proteins and growth inhibition caused by analogs of GA. A near-maximal depletion of Raf-1, ErbB2 and mutant p53 is accompanied by near-maximally toxicity to SKBr3 breast cancer cells.¹⁶ For GA, these concentrations (IC_80-90 = 30 nM) are between four and five times greater than an IC₅₀ (below 10 nM).¹⁶ Similarly, 30 nM GA depleted Bcr-Abl in K562 cells. ErbB-2, Raf-1, Cdk-4 and mutant p53 were depleted by GA analogs and KF25706 (a radicicol oxime derivative) at concentrations comparable to those required for the antiproliferative activity.^16,60 Depletion of the Bcr-Abl protein (at concentrations of GA as low as 30 nM) selectively induced apoptosis in Bcr-Abl positive cells and sensitized these cells to standard chemotherapy.³⁶

Hsp90-active drugs induce both G1 and G2/M phase arrest of the cell cycle. Herbimycin A down-regulates cyclin D, causing an Rb-dependent growth arrest in the G1 phase of the cell cycle.^55,74,75 In breast cancer cells, a G1 arrest was accompanied by differentiation and followed by apoptosis. The differentiation was characterized by specific changes in morphology and induction of milk fat proteins and lipid droplets. In cells lacking Rb, neither G1 arrest nor differentiation occurs. Instead, they undergo apoptosis during mitosis.⁷⁶

In K562 leukemia cells, GA induces both G1 and G2/M arrests.^36,77 In these cells, GA down-regulated the expression of cyclin B1 and inhibited phosphorylation of p34Cdc2, causing G2/M arrest.⁷⁸ Effects of GA are cell-type dependent. By arresting MCF-7 cells, GA prevents paclitaxel-induced mitotic arrest and Bcl-2 phosphorylation in MCF-7 cells,⁷⁹ but not in HL60 cells.⁸⁰ Therefore, GA can either decrease or increase the cytotoxicity of paclitaxel, depending on cellular context, that potentially could be exploited therapeutically. Both sequence of drugs and tumor cell biology matters in combining cytotoxics with Hsp90-active drugs.⁸¹ For example, in a subset of breast cancer cell lines, addition of 17-A-GA to cells after exposure to paclitaxel increased apoptosis. In breast cancer cells with intact Rb, such as SKBr3 cells, exposure to 17-A-GA before paclitaxel resulted in growth arrest and abrogated apoptosis.⁸² Such a schedule dependence was not seen in BT-549 and MDA-468 breast cancer cells with mutated Rb. Exposure to 17-A-GA before paclitaxel rendered lung cancer cells with low ErbB-2 levels refractory to paclitaxel cytotoxicity.⁸³ 17-A-GA sensitized breast cancer cells to doxorubicin, in a schedule- and Rb-independent manner.⁸² In HL60 leukemia cells, Hsp90-active agents diminished the cytotoxicity of doxorubicin.³⁶

In a cell-type dependent manner, Hsp90-active agents cause apoptosis. For example, 17-A-GA induces cytosolic accumulation of cytochrome C, activates caspase-9 and caspase-3, triggering apoptosis in HL-60/Bcr-Abl and K562 cells.³²

Crucial targets of Hsp90-active drugs vary in different cell types. For example, in Bcr-Abl-expressing K562 leukemia cells, Bcr-Abl appears to be the crucial target which depletion causes cytotoxicity. ErbB-2 is likely an important target in SKBr3 breast cancer cells, which overexpress ErbB-2. Context-dependent effects of GA is one of the basis for cancer-specific cytotoxicity.

Bcr-Abl-expressing leukemia

Chronic myelogenous leukemia (CML) is characterized by a reciprocal (t9;22) chromosomal translocation, known as the Philadelphia chromosome, that fuses the truncated Bcr gene to the truncated c-Abl.⁸⁴ Bcr-Abl is an active tyrosine protein kinase, which renders cells resistant to apoptosis.^85,86,87,88 Bcr-Abl exists in a complex with Hsp90, and GA causes degradation of Bcr-Abl after 3-5 h of treatment.^77,89 GA down-regulates Bcr-Abl in natural Ph¹-positive K562 cells and in HL60 cells transfected with Bcr-Abl.^32,36,89 In both cell lines, a depletion of the Bcr-Abl protein and a near-maximal toxicity were achieved at concentrations of GA as low as 30 nM.³⁶ Furthermore, GA sensitized Bcr-Abl-expressing cells to doxorubicin and, albeit to a lesser degree, to paclitaxel.³⁶

Specific inhibitors of the Abl kinase, such as STI 571, are very effective in the therapy of Bcr-Abl-positive leukemia,^90,91 but resistance to STI 571 develops.^91,92,93 Hsp90-active drugs, especially in combination with doxorubicin or STI571, may be a second-line therapy of Bcr-Abl-positive leukemias.

FLT3-expressing leukemias

A somatic mutation of the FLT3 gene, in which the juxtamembrane domain has an internal tandem duplication, is found in 20% of human acute myeloid leukemias. Transfection of mutant FLT3 gene into an IL3-dependent murine cell line, 32D, abrogated the IL3-dependency, and caused leukemia in addition to subcutaneous tumors in mice.⁹⁴ Herbimycin A, a Hsp90-active agent, inhibited the growth of the transformed 32D cells, but it was ineffective in parental 32D cells. Herbimycin A suppressed the constitutive tyrosine phosphorylation of the mutant FLT3, but not the phosphorylation of the ligand-stimulated wild-type FLT3.⁹⁵ In mice transplanted with the transformed 32D cells, the administration of herbimycin A prolonged the latency of disease or completely prevented leukemia, depending on the number of cells inoculated and schedule of drug administration. These results suggest that mutant FLT3 is a promising target for Hsp90-active drugs in the treatment of leukemia.⁹⁵

Selectivity against cancer cells

Two previous examples illustrate the basis of selective killing of ceratin oncoprotein-expressing cells. In addition to Bcr-Abl and FLT3, Raf-1 may be an important target in leukemias.⁶³ As shown in Bcr-Abl-expressing cells, acute destabilization of the kinase results in cell death. In breast cancer, simultaneous targeting of ErbB-2, the IGF receptor, the Akt kinase, and estrogen receptors by Hsp90-active drugs may be effective. If a cell particularly depends on a certain oncoprotein, it may fall apart following a brief exposure to Hsp90-active drugs.

Secondly, Hsp90 is overexpressed in tumor cells,⁹⁶ indicating that these cells are highly dependent on the Hsp90 function. Mutant oncoproteins may depend on the full function of Hsp90 as a conformational buffer to maintain full activity. Overloading of the Hsp90 capacity with mutated oncoproteins under treatment with Hsp90-active agents could cause death of cancer cells. For example, maintenance of wild-type Hck (a Src-family kinase) and its constitutively active counterpart, Hck499F, requires Hsp90. Hck499F had a greater requirement for on-going support from Hsp90 than did mature wild-type Hck.⁹⁷

Thirdly, Hsp90-active agents can selectively sensitize oncoprotein-overexpressing cells to chemotherapy. Overexpression of ErbB-2 contributes to chemoresistance. 17-A-GA treatment efficiently depleted ErbB-2 in lung cancer cells. Induction of apoptosis was observed after treatment of cells with the combination of paclitaxel and 17-A-GA.⁸³ Furthermore, Hsp90-active agents can protect certain cells against chemotherapy. This could be exploited for selective killing of cancer cells.

Besides, anti-HER-2 monoclonal antibody can be used to deliver GA, which depletes HER-2, by coupling GA to these monoclonal antibodies.⁹⁸ This might enhance the capacity of the antibody to down-regulate HER-2 and also to avoid side-effects of GA.⁹⁹

In a special case, cancer cells that overexpress quinone-metabolizing enzyme (NQO1), which increases in 17-A-GA growth-inhibition activity, can be more sensitive to this HSP90-active compound.¹⁰⁰

Protection against chemotherapy-induced apoptosis

Hsp90-active drugs can protect some cells against apoptosis caused by other anticancer drugs. For example, cultured dorsal root ganglion (DRG) neurons from chick embryos were extremely susceptible to the antineoplastic drugs, cisplatin, vincristine and paclitaxel even in the presence of the neurotrophins.¹⁰¹ The neurotoxic effects of these anticancer drugs were completely prevented by the addition of low doses of radicicol (20 nM) or GA (2 nM), but higher doses of GA (>5 nM) had severe cytotoxic effects on neurons. Higher doses of radicicol (500 nM), however, still promoted neurites and prevented apoptosis in the absence of neurotrophins. Slightly different cellular effects of the two antibiotics are not explained.

While potentiating doxorubicin-mediated cytotoxicity in Bcr-Abl-expressing cells, GA protected parental HL60 cells against doxorubicin-induced apoptosis. Combinations of doxorubicin with low concentrations of GA might eliminate Ph-positive cells, while protecting cells that do not express Bcr-Abl. In theory, this can decrease side-effects and increase the therapeutic index.

How can Hsp90-active drugs protect cells from doxorubicin-induced apoptosis? GA and herbimycin A, Hsp90 inhibitors, induce synthesis of Hsp70 and Hsp90.^102,103 In turn, HSPs can inhibit apoptosis.²⁰ Induction of Hsp70 by herbimycin A was observed in several cell lines, including A431 human epidermoid carcinoma cells, HeLa S3 cells, chicken embryo fibroblasts, NIH3T3 cells and Rous sarcoma virus-transformed NIH3T3 cells.¹⁰⁴ In cardiac cells, herbimycin A induces Hsp70. Moreover, Hsp's induction correlated with the ability of herbimycin A to protect cells against severe stress. These results indicate the possibility of a pharmacological approach to Hsp70 induction and cardiac protection, which may ultimately be of clinical relevance.¹⁰⁵

Effective concentrations in vivo

Concentrations of Hsp90-active drugs that deplete Hsp90-associated oncoproteins can be achieved and tolerated in vivo. Given that GA displayed hepatotoxicity in dogs which appears to be unrelated to its Hsp90 antagonism, 17-allylamino,17-demethoxygeldanamycin (17-A-GA), a geldanamycin analog, is the first inhibitor of Hsp90 that enters a phase I clinical trial in cancer.⁸¹ In mice, bolus i.v. delivery of 60 mg/kg 17-A-GA produced 'peak' plasma 17-A-GA concentrations between 5.8 and 19.3 g/ml 5 min after injection. After i.v. bolus delivery to mice, 17-A-GA distributed rapidly to all tissues, except the brain. Substantial concentrations of 17-A-GA were measured in each tissue. A 60 mg/kg dose of 17-A-GA, caused no changes in appearance, appetite, waste elimination, or survival of treated animals.¹⁰⁶

Phase I clinical trials

Although few data are available, one can certainly expect that Hsp90-active agents should be toxic to normal cells. It has been shown, for example, that by inhibiting the Raf-1/MAPK pathway, GA induced apoptosis in luteinized granulosa cells.¹⁰⁷

In clinical trials, one can retrospectively analyze side-effects due to toxicity to normal cells, thus revealing dose-limiting targets in proliferating (eg mucosa and bone marrow cells) and non-proliferating normal cells. This in turn can help to design rational therapeutic modalities. In a clinical trial at the National Cancer Institute (NIH), 17-allylamino,17-demethoxygeldanamycin (17-A-GA) was administered daily by 1-h infusion for 5 days every 3 weeks in adult patients with various solid tumors.¹⁰⁸ Dose-limiting toxicity was reversible grade III hepatotoxicity. Other grade I/II toxicities included fever, emesis, anemia and fatigue. The maximum tolerated dose was 40 mg/m² and plasma maximal concentrations levels were 1860 ± 660 nM. The recommended 17-A-GA phase II dose on this schedule is 40 mg/m², at which inhibition of the target Hsp90 occurs.¹⁰⁸ In the clinical trial at Memorial Sloan Kettering in New York, the drug was administered daily for 5 days and repeated every 3 weeks. At 80 mg/m² dose (with peak plasma levels of the drug: 2700 nM), limiting toxicities were diarrhea, thrombocytopenia and transient transaminitis.¹⁰⁹ This study suggests that 17-AAG can be administered to patients in concentrations exceeding those that were effective in pre-clinical models.¹⁰⁹ In a phase I trial at Royal Marsden Hospital (UK), with weekly administrations at doses of 80 mg/m², no hematological or biochemical toxicity has been observed. In this study, the evidence of biological activity has been demonstrated by induction of Hsp70 in PBLs at 6 h. In the data available by 2001, no objective responses were seen,^109,110 however four of 13 patients had stable disease beyond 3 months.¹⁰⁹

Two conclusions could be drawn. Firstly, a Hsp90-active drug is not a 'magic bullet' against cancer. Secondly, by shortening the duration of a treatment (1 day instead of 5 days), one may eliminate side-effects. This could be expected. Unlike DNA-damaging drugs, Hsp90-active drugs do not cause irreversible damage, if the time of exposure is brief. Therefore, cells that have not undergone apoptosis potentially may recover.

Although Hsp90-active drugs are introduced in the therapy of solid tumors, preclinical data indicate great promise for these drugs in therapy of leukemia. In striking symmetry, STI571, which was designed to treat Bcr-Abl-expressing leukemia, is underway to explore its utility in solid tumors harboring c-kit and PDGFR abnormalities.

Unexpected twist: instead of conclusion

Thus, Hsp90-active drugs entered clinical trials. But what if Hsp90-active drugs had already been used in therapy. Would it have been a disappointment? Or a useful lesson?

The drug novobiocin has clinically been used for a decade. Coumarin antibiotics, including novobiocin, are known as inhibitors of topoisomerase II. In addition, novobiocin is used clinically as a modulator of alkylating agents.¹¹¹ Besides, novobiocin reverses drug resistance and increases intracellular accumulation of etoposide.¹¹²

Novobiocin (300 M) induces granulocytic differentiation in HL-60 cells.¹¹³ Finally, at doses of 300-800 M, novobiocin depleted cells of Raf-1, ErbB2, mutant p53 and v-Src.¹¹⁴ Although novobiocin binds to a site on Hsp90 that is different from the geldanamycin-binding site, it, like GA, is able to inhibit the chaperone function of Hsp90 and to deplete tumor cells of Hsp90-dependent proteins.¹¹⁴ Novobiocin therapy decreased levels of Raf-1 in murine splenocytes, indicating that mechanism-based concentration is achievable.¹¹⁴ Following administration of novobiocin to patients, the serum drug concentration vary between 100 and 400 g/ml,¹¹⁵ which corresponds to doses that down-regulate Hsp90-associated proteins in vitro.

Thus, the Hsp90-active agent novobiocin is already used in cancer therapy. Side-effects and maximum tolerated dose of novobiocin are difficult to evaluate because the drug was administered in combination with other cytostatics.¹¹¹ Given that the severity of mucositis correlated with the plasma levels of novobiocin,¹¹¹ one can conclude that novobiocin inhibits proliferation of normal epithelial cells.

It was believed that inhibitors of DNA repair (eg novobiocin) can overcome resistance to alkylators which damage DNA.¹¹¹ Ironically, inhibition of DNA repair facilitates development of resistance to alkylating agents.¹¹⁶ It is tempting to suggest that, by inhibiting Hsp90 functions (rather than DNA repair), novobiocin sensitizes certain tumors to chemotherapy.

1 Ernst P, Killmann SA. Effect of anti-leukemic drugs on cell cycle of human leukemic blast cells in vivo. Acta Med Scand 1969; 186: 239-240. MEDLINE

2 Pardee AB. G1 events and regulation of cell proliferation. Science 1989; 246: 603-608. MEDLINE

3 Bartek J, Lukas J, Bartkova J. Perspective: defects in cell cycle control and cancer. J Pathol 1999; 187: 95-99. Article MEDLINE

4 Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100: 57-70. MEDLINE

5 Sherr CJ. The Pezcoller lecture: Cancer cell cycle revisited. Cancer Res 2000; 60: 3689-3695. MEDLINE

6 Evan GI, Vousden KH. Proliferation, cell cycle and apoptosis in cancer. Nature 2001; 411: 342-348. Article MEDLINE

7 Fisher DE. Apoptosis in cancer therapy: crossing the threshold. Cell 1994; 78: 539-542. MEDLINE

8 Lowe SW, Lin AW. Apoptosis in cancer. Carcinogenesis 2000; 21: 485-495. Article MEDLINE

9 Blagosklonny MV, Pardee AB. Exploiting cancer cell cycling for selective protection of normal cells. Cancer Res 2001; 61: 4301-4305. MEDLINE

10 Shapiro GI, Harper JW. Anticancer drug targets: cell cycle and checkpoint control. J Clin Invest 1999; 104: 1645-1653. MEDLINE

11 Kaelin WGJ. Choosing anticancer drug targets in the postgenomic era. J Clin Invest 1999; 104: 1503-1506. MEDLINE

12 Kaelin WG. Taking aim at novel molecular targets in cancer therapy. J Clin Invest 1999; 104: 1495-1506. MEDLINE

13 Gibbs JB. Mechanism-based target identification and drug discovery in cancer. Science 2000; 287: 1969-1973. Article MEDLINE

14 Buolamwini JK. Cell cycle molecular targets and drug discovery. In: Blagosklonny MV (ed.). Cell Cycle Checkpoints and Cancer Landes Bioscience: Austin, TX, 2002, 235-246.

15 Vogelstein B, Kinzler KW. The multistep nature of cancer. Trends Genet 1993; 9: 138-141. MEDLINE

16 An WG, Schnur RC, Neckers LM, Blagosklonny MV. Depletion of ErbB2, Raf-1 and mutant p53 proteins by geldanamycin derivatives correlates with antiproliferative activity. Cancer Chemother Pharmacol 1997; 40: 60-64. Article MEDLINE

17 Uehara Y, Hori M, Takeuchi T, Umezawa H. Phenotypic change from transformed to normal induced by benzoquinonoid ansamycins accompanies inactivation of p60src in rat kidney cells infected with Rous sarcoma virus. Mol Cell Biol 1986; 6: 2198-2206. MEDLINE

18 Whitesell L, Mimnaugh EG, De Costa B, Myers CE, Neckers LM. Inhibition of HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation. Proc Natl Acad Sci USA 1994; 91: 8324-8328. MEDLINE

19 Morimoto RI, Santoro MG. Stress-inducible responses and heat shock proteins: new pharmacologic targets for cytoprotection. Nat Biotechnol 1998; 16: 833-838. MEDLINE

20 Creagh EM, Sheehan D, Cotter TG. Heat shock proteins - modulators of apoptosis in tumour cells. Leukemia 2000; 14: 1161-1173. MEDLINE

21 Young JC, Moarefi I, Hartl FU. Hsp90: a specialized but essential protein-folding tool. J Cell Biol 2001; 154: 267-273. MEDLINE

22 Richter K, Buchner J. Hsp90: chaperoning signal transduction. J Cell Physiol 2001; 188: 281-290. Article MEDLINE

23 Jolly C, Morimoto RI. Role of the heat shock response and molecular chaperones in oncogenesis and cell death. J Natl Cancer Inst 2000; 92: 1564-1572. Article MEDLINE

24 Stebbins CE, Russo AA, Schnieder C, Rosen N, Hartl FU, Pavletich NP. Crystal structure of an Hsp90-geldanamycin complex: targeting of a protein chaperone by an antitumor agent. Cell 1997; 89: 239-250. MEDLINE

25 Grenert JP, Sullivan WP, Fadden P, Haystead TAJ, Clark J, Mimnaugh E, Krutzsch H, Ochel HJ, Schulte TW, Sausville E, Neckers LM, Toft DO. The amino-terminal domain of heat shock protein 90 (hsp90) that binds geldanamycin is an ATP/ADP switch domain that regulates hsp90 conformation. J Biol Chem 1997; 272: 23843-23850. Article MEDLINE

26 Bohen SP. Genetic and biochemical analysis of p23 and ansamycin antibiotics in the function of Hsp90-dependent signaling proteins. Mol Cell Biol 1998; 18: 3330-3339. MEDLINE

27 Roe SM, Prodromou C, O'Brien R, Ladbury JE, Piper PW, Pearl LH. Structural basis for inhibition of the Hsp90 molecular chaperone by the antitumor antibiotics radicicol and geldanamycin. J Med Chem 1999; 42: 260-266. Article MEDLINE

28 Maki CG, Huibregtse JM, Howley PM. In vivo ubiquitination and proteasome-mediated degradation of p53. Cancer Res 1996; 56: 2649-2654. MEDLINE

29 DeSalle LM, Pagano M. Regulation of the G1 to S transition by the ubiquitin pathway. FEBS Lett 2001; 490: 179-189. MEDLINE

30 Sepp-Lorenzino L, Ma Z, Lebwohl DE, Vinitsky A, Rosen N. Herbimycin A induces the 20 S proteasome- and ubiquitin-dependent degradation of receptor tyrosine kinases. J Biol Chem 1995; 270: 16580-16587. MEDLINE

31 Mimnaugh EG, Chavany C, Neckers L. Polyubiquitination and proteasomal degradation of the p185(c-erbB-2) receptor protein-tyrosine kinase induced by geldanamycin. J Biol Chem 1996; 271: 22796-22801. MEDLINE

32 Nimmanapalli R, O'Bryan E, Bhalla K. Geldanamycin and its analogue 17-allylamino-17-demethoxygeldanamycin lowers Bcr-Abl levels and induces apoptosis and differentiation of Bcr-Abl-positive human leukemic blasts. Cancer Res 2001; 61: 1799-1804. MEDLINE

33 Drexler HC. Activation of the cell death program by inhibition of proteasome function. Proc Natl Acad Sci USA 1997; 94: 855-860. Article MEDLINE

34 Adams J, Palombella VJ, Sausville EA, Johnson J, Destree A, Lazarus DD, Maas J, Pien CS, Prakash S, Elliott PJ. Proteasome inhibitors: a novel class of potent and effective antitumor agents. Cancer Res 1999; 59: 2615-2622. MEDLINE

35 An WG, Hwang SG, Trepel JB, Blagosklonny MV. Protease inhibitor-induced apoptosis: accumulation wt p53, p21WAF1/CIP1, and induction of apoptosis are independent markers of proteasome inhibition. Leukemia 2000; 14: 1276-1283. MEDLINE

36 Blagosklonny MV, Fojo T, Bhalla KN, Kim J-S, Trepel JB, Figg WD, Rivera Y, Neckers LM. The Hsp90 inhibitor geldanamycin selectively sensitizes Bcr-Abl-expressing leukemia cells to cytotoxic chemotherapy. Leukemia 2001; 15: 1537-1543. MEDLINE

37 Blagosklonny MV, Wu GS, Omura S, El-Deiry WS. Proteasome-dependent regulation of p21WAF1/CIP1 expression. Biochem Biophys Res Comm 1996; 227: 564-569. MEDLINE

38 Kubbutat MHG, Jones SN, Vousden KH. Regulation of p53 stability by Mdm2. Nature 1997; 387: 299-303. MEDLINE

39 Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature 1997; 387: 296-299. MEDLINE

40 Blagosklonny MV. Loss of function and p53 stabilization. Oncogene 1997; 15: 1889-1893. MEDLINE

41 Midgley CA, Lane DP. p53 protein stability in tumour cells is not determined by mutation but is dependent on Mdm2 binding. Oncogene 1997; 15: 1179-1189. MEDLINE

42 Blagosklonny MV, Toretskey J, Neckers LM. Geldanamycin selectively destabilizes and conformationally alters mutated p53. Oncogene 1995; 11: 933-939. MEDLINE

43 Whitesell L, Sutphin P, An WG, Schulte T, Blagosklonny MV, Neckers L. Geldanamycin-stimulated destabolization of mutated p53 is mediated by the proteasome in vivo. Oncogene 1997; 14: 2809-2816. MEDLINE

44 Nagata Y, Anan T, Yoshida T, Mizukami T, Taya Y, Fujiwara T, Kato H, Saya H, Nakao M. The stabilization mechanism of mutant-type p53 by impaired ubiquitination: the loss of wild-type p53 function and the hsp90 association. Oncogene 1999; 18: 6037-6049. MEDLINE

45 Blagosklonny MV, Toretskey J, Bohen S, Neckers LM. Conformation of mutated p53 requires functional HSP90. Proc Natl Acad Sci USA 1996; 93: 8379-8383. MEDLINE

46 Loo MA, Jensen TJ, Cui L, Hou Y, Chang XB, Riordan JR. Perturbation of Hsp90 interaction with nascent CFTR prevents its maturation and accelerates its degradation by the proteasome. EMBO J 1998; 17: 6879-6887. Article MEDLINE

47 Blagosklonny MV, Schulte TW, Nguyen P, Mimnaugh EG, Trepel J, Neckers L. Taxol induction of p21Waf1 and p53 requires c-raf-1. Cancer Res 1995; 55: 4623-4626. MEDLINE

48 Miller P, DiOrio C, Moyer M, Schnur RC, Bruskin A, Cullen W, Moyer JD. Depletion of the erbB-2 gene product p185 by benzoquinoid ansamycins. Cancer Res 1994; 54: 2724-2730. MEDLINE

49 Tikhomirov O, Carpenter G. Geldanamycin induces ErbB-2 degradation by proteolytic fragmentation. J Biol Chem 2000; 275: 26625-26631. MEDLINE

50 Supino-Rosin L, Yoshimura A, Yarden Y, Elazar Z, Neumann D. Intracellular retention and degradation of the epidermal growth factor receptor, two distinct processes mediated by benzoquinone ansamycins. J Biol Chem 2000; 275: 21850-21855. MEDLINE

51 Schulte TW, Blagosklonny MV, Romanova L, Mushinski JF, Monia BP, Johnston JF, Nguyen P, Trepel J, Neckers LM. Destabilization of Raf-1 by geldanamycin leads to disruption of the Raf-1-MEK-Mitogen-activated protein kinase signalling pathway. Mol Cell Biol 1996; 16: 5839-5845. MEDLINE

52 Hartson SD, Barrett DJ, Burn P, Matts RL. Hsp90-mediated folding of the lymphoid cell kinase p56lck. Biochemistry 1996; 35: 13451-13459. Article MEDLINE

53 Stepanova L, Leng X, Parker SB, Harper JW. Mammalian p50Cdc37 is a protein kinase-targeting subunit of Hsp90 that binds and stabilizes Cdk4. Genes Dev 1996; 10: 1491-1502. MEDLINE

54 Sato S, Fujita N, Tsuruo T. Modulation of akt kinase activity by binding to hsp90. Proc Natl Acad Sci USA 2000; 97: 10832-10837. MEDLINE

55 Muise-Helmericks RC, Grimes HL, Bellacosa A, Malstrom SE, Tsichlis PN, Rosen N. Cyclin D expression is controlled post-transcriptionally via a phosphatidylinositol 3-kinase/Akt-dependent pathway. J Biol Chem 1998; 273: 29864-29872. MEDLINE

56 Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 1999; 13: 1501-1512. MEDLINE

57 Kwon HJ, Yoshida M, Fukui Y, Horinouchi S, Beppu T. Potent and specific inhibition of p60v-src protein kinase both in vivo and in vitro by radicicol. Cancer Res 1992; 52: 6926-6930. MEDLINE

58 Zhao JF, Nakano H, Sharma S. Suppression of RAS and MOS transformation by radicicol. Oncogene 1995; 11: 161-173. MEDLINE

59 Soga S, Kozawa T, Narumi H, Akinaga S, Irie K, Matsumoto K, Sharma SV, Nakano H, Mizukami T, Hara M. Radicicol leads to selective depletion of Raf kinase and disrupts K-Ras-activated aberrant signaling pathway. J Biol Chem 1998; 273: 822-828. MEDLINE

60 Soga S, Neckers LM, Schulte TW, Shiotsu Y, Akasaka K, Narumi H, Agatsuma T, Ikuina Y, Murakata C, Tamaoki T, Akinaga S. KF25706, a novel oxime derivative of radicicol, exhibits in vivo antitumor activity via selective depletion of Hsp90 binding signaling molecules. Cancer Res 1999; 59: 2931-2938. MEDLINE

61 Sharma SV, Agatsuma T, Nakano H. Targeting of the protein chaperone, HSP90, by the transformation suppressing agent, radicicol. Oncogene 1998; 16: 2639-2645. MEDLINE

62 Gibbs JB. Anticancer drug targets: growth factors and growth factor signaling. J Clin Invest 2000; 105: 9-13. MEDLINE

63 Weinstein-Oppenheimer CR, Blalock WL, Steelman LS, Chang FM, McCubrey JA. The Raf signal transduction cascade as a target for chemotherapeutic intervention in growth factor-responsive tumors. Pharmacol Ther 2000; 88: 229-279. MEDLINE

64 Dunn SE, Torres JV, Oh JS, Cykert DM, Barrett JC. Up-regulation of urokinase-type plasminogen activator by insulin-like growth factor-I depends upon phosphatidylinositol-3 kinase and mitogen-activated protein kinase kinase. Cancer Res 2001; 61: 1367-1374. MEDLINE

65 Blagosklonny MV. The mitogen-activated protein kinase pathway mediates growth arrest or E1A-dependent apoptosis in SKBr3 human breast cancer cells. Int J Cancer 1998; 78: 511-517. Article MEDLINE

66 Blagosklonny MV, Chuman Y, Bergan RC, Fojo T. Mitogen-activated protein kinase pathway is dispensable for microtubule-active drug-induced Raf-1/Bcl-2 phosphorylation and apoptosis in leukemia cells. Leukemia 1999; 13: 1028-1036. MEDLINE

67 Chiosis G, Timaul MN, Lucas B, Munster PN, Zheng FF, Sepp-Lorenzino L, Rosen N. A small molecule designed to bind to the adenine nucleotide pocket of Hsp90 causes Her2 degradation and the growth arrest and differentiation of breast cancer cells. Chem Biol 2001; 8: 289-299. MEDLINE

68 Tsai M-J, O'Malley BW. Molecular mechanisms of action of steroid/thyroid hormone receptor superfamily members. Ann Rev Biochem 1994; 63: 451-486. MEDLINE

69 Pratt WB, Silverstein AM, Galigniana MD. A model for the cytoplasmic trafficking of signalling proteins involving the hsp90-binding immunophilins and p50cdc37. Cell Signal 1999; 11: 839-851. MEDLINE

70 Whitesell L, Cook P. Stable and specific binding of heat shock protein 90 by geldanamycin disrupts glucocorticoid receptor function in intact cells. Mol Endocrinol 1996; 10: 705-712. MEDLINE

71 Czar MJ, Galigniana MD, Silverstein AM, Pratt WB. Geldanamycin, a heat shock protein 90-binding benzoquinone ansamycin, inhibits steroid-dependent translocation of the glucocorticoid receptor from the cytoplasm to the nucleus. Biochemistry 1997; 36: 7776-7785. Article MEDLINE

72 Galigniana MD, Scruggs JL, Herrington J, Welsh MJ, Carter-Su C, Housley PR, Silverstein AM, Pratt WB. Heat shock protein 90-dependent (geldanamycin-inhibited) movement of the glucocorticoid receptor through the cytoplasm to the nucleus requires intact cytoskeleton. Mol Endocrinol 1998; 12: 1903-1913. MEDLINE

73 Bagatell R, Khan O, Paine-Murrieta G, Taylor CW, Akinaga S, Whitesell L. Destabilization of steroid receptors by heat shock protein 90-binding drugs: a ligand-independent approach to hormonal therapy of breast cancer. Clin Cancer Res 2001; 7: 2076-2084. MEDLINE

74 Yen A, Soong S, Kwon HJ, Yoshida M, Beppu T, Varvayanis S. Enhanced cell differentiation when RB is hypophosphorylated and down-regulated by radicicol, a SRC-kinase inhibitor. Exp Cell Res 1994; 214: 163-171. Article MEDLINE

75 Srethapakdi M, Liu F, Tavorath R, Rosen N. Inhibition of Hsp90 function by ansamycins causes retinoblastoma gene product-dependent G1 arrest. Cancer Res 2000; 60: 3940-3946. MEDLINE

76 Munster PN, Srethapakdi M, Moasser MM, Rosen N. Inhibition of heat shock protein 90 function by ansamycins causes the morphological and functional differentiation of breast cancer cells. Cancer Res 2001; 61: 2945-2952. MEDLINE

77 Shiotsu Y, Neckers LM, Wortman I, An WG, Schulte TW, Soga S, Murakata C, Tamaoki T, Akinaga S. Novel oxime derivatives of radicicol induce erythroid differentiation associated with preferential G(1) phase accumulation against chronic myelogenous leukemia cells through destabilization of Bcr-Abl with Hsp90 complex. Blood 2000; 96: 2284-2291. MEDLINE

78 Kim HR, Lee CH, Choi YH, Kang HS, Kim HD. Geldanamycin induces cell cycle arrest in K562 erythroleukemic cells. IUBMB Life 1999; 48: 425-428. Article MEDLINE

79 Blagosklonny MV, Schulte TW, Nguyen P, Trepel J, Neckers L. Taxol-induced apoptosis and phosphorylation of Bcl-2 protein involves c-raf-1 and represents a novel c-Raf-1 signal transduction pathway. Cancer Res 1996; 56: 1851-1854. MEDLINE

80 Ibrado AM, Liu L, Bhalla K. Bcl-xL overexpression inhibits progression of molecular events leading to paclitaxel-induced apoptosis of human AML HL-60 cells. Cancer Res 1997; 57: 1109-1115. MEDLINE

81 Sausville EA. Combining cytotoxics and 17-allylamino, 17-demethoxygeldanamycin: sequence and tumor biology matters. Clin Cancer Res 2001; 7: 2155-2158. MEDLINE

82 Munster PN, Basso A, Solit D, Norton L, Rosen N. Modulation of Hsp90 function by ansamycins sensitizes breast cancer cells to chemotherapy-induced apoptosis in an RB and schedule-dependent manner. Clin Cancer Res 2001; 7: 2228-2236. MEDLINE

83 Nguyen DM, Chen A, Mixon A, Schrump DS. Sequence-dependent enhancement of paclitaxel toxicity in non-small cell lung cancer by 17-allylamino 17-demethoxygeldanamycin. J Thor Cardiovascular Surg 1999; 118: 908-915.

84 Deininger MWN, Goldman JM, Melo JV. The molecular biology of chronic myeloid leukemia. Blood 2000; 96: 3343-3356. MEDLINE

85 Bedi A, Barber JP, Bedi GC, el-Deiry WS, Sidransky D, Vala MS, Akhtar AJ, Hilton J, Jones RJ. BCR-ABL-mediated inhibition of apoptosis with delay of G2/M transition after DNA damage: a mechanism of resistance to multiple anticancer agents. Blood 1995; 86: 1148-1158. MEDLINE

86 Dubrez L, Eymin B, Sordet O, Droin N, Turhan AG, Solary E. BCR-ABL delays apoptosis upstream of procaspase-3 activation. Blood 1998; 91: 2415-2422. MEDLINE

87 Amarante-Mendes GP, Naekyung Kim C, Liu L, Huang Y, Perkins CL, Green DR, Bhalla K. Bcr-Abl exerts its antiapoptotic effect against diverse apoptotic stimuli through blockage of mitochondrial release of cytochrome C and activation of caspase-3. Blood 1998; 91: 1700-1705. MEDLINE

88 McCubrey JA, May WS, Duronio V, Mufson A. Serine/threonine phosphorylation in cytokine signal transduction. Leukemia 2000; 14: 1060-1079. MEDLINE

89 An WG, Schulte TW, Neckers LM. The HSP90 antagonist geldanamycin alters chaperone association with p210BCR-ABL and v-src proteins prior to their degradation by the proteasome. Cell Growth Diff 2000; 11: 355-360. MEDLINE

90 Druker BJ, Sawyers CL, Kantarjian H, Resta DJ, Reese SF, Ford JM, Capdeville R, Talpaz M. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med 2001; 344: 1038-1042. MEDLINE

91 Druker BJ, Talpaz M, Resta DJ, Peng B, Buchdunger E, Ford JM, Lydon NB, Kantarjian H, Capdeville R, Ohno-Jones S, Sawyers CL. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001; 344: 1031-1037. MEDLINE

92 Gorre ME, Mohammed M, Ellwood K, Hsu N, Paquette R, Rao PN, Sawyers CL. Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 2001; 293: 876-880. Article MEDLINE

93 Sirulink A, Silver RT, Najfeld V. Marked ploidy and BCR-ABL gene amplification in vivo in a patient treated with STI571. Leukemia 2001; 15: 1795-1797. MEDLINE

94 Tse KF, Mukherjee G, Small D. Constitutive activation of FLT3 stimulates multiple intracellular signal transducers and results in transformation. Leukemia 2000; 14: 1766-1776. MEDLINE

95 Zhao M, Kiyoi H, Yamamoto Y, Ito M, Towatari M, Omura S, Kitamura T, Ueda R, Saito H, Naoe T. In vivo treatment of mutant FLT3-transformed murine leukemia with a tyrosine kinase inhibitor. Leukemia 2000; 14: 374-378. MEDLINE

96 Ferrarini M, Heltai S, Zocchi MR, Rugarli C. Unusual expression and localization of heat-shock proteins in human tumor cells. Int J Cancer 1992; 51: 613-619. MEDLINE

97 Scholz GM, Hartson SD, Cartledge K, Volk L, Matts RL, Dunn AR. The molecular chaperone Hsp90 is required for signal trans- duction by wild-type Hck and maintenance of its constitutively active counterpart. Cell Growth Diff 2001; 12: 409-417. MEDLINE

98 Mandler R, Wu C, Sausville EA, Roettinger AJ, Newman DJ, Ho DK, King CR, Yang D, Lippman ME, Landolfi NF, Dadachova E, Brechbiel MW, Waldmann TA. Immunoconjugates of geldanamycin and anti-HER2 monoclonal antibodies: antiproliferative activity on human breast carcinoma cell lines. J Natl Cancer Inst 2000; 92: 1573-1581. MEDLINE

99 Mendelsohn J. Use of an antibody to target geldanamycin. J Natl Cancer Inst 2000; 92: 1549-1551. MEDLINE

100 Kelland LR, Sharp SY, Rogers PM, Myers TG, Workman P. DT-diaphorase expression and tumor cell sensitivity to 17-allylamino,17-demethoxygeldanamycin, an inhibitor of heat shock protein 90. J Natl Cancer Inst 1999; 91: 1940-1949. MEDLINE

101 Sano M. Radicicol and geldanamycin prevent neurotoxic effects of anti-cancer drugs on cultured embryonic sensory neurons. Neuropharmacology 2001; 40: 947-953. Article MEDLINE

102 Kim HR, Kang HS, Kim HD. Geldanamycin induces heat shock protein expression through activation of HSF1 in K562 erythroleukemic cells. IUBMB Life 1999; 48: 429-433. Article MEDLINE

103 Bagatell R, Paine-Murrieta GD, Taylor CW, Pulcini EJ, Akinaga S, Benjamin IJ, Whitesell L. Induction of a heat shock factor 1-dependent stress response alters the cytotoxic activity of hsp90-binding agents. Clin Cancer Res 2000; 6: 3312-3328. MEDLINE

104 Murakami Y, Uehara Y, Yamamoto C, Fukazawa H, Mizuno S. Induction of hsp 72/73 by herbimycin A, an inhibitor of transformation by tyrosine kinase oncogenes. Exp Cell Res 1991; 195: 338-344. MEDLINE

105 Morris SD, Cumming DV, Latchman DS, Yellon DM. Specific induction of the 70-kD heat stress proteins by the tyrosine kinase inhibitor herbimycin-A protects rat neonatal cardiomyocytes. A new pharmacological route to stress protein expression? J Clin Invest 1996; 97: 706-712.

106 Egorin MJ, Zuhowski EG, Rosen DM, Sentz DL, Covey JM, Eiseman JL. Plasma pharmacokinetics and tissue distribution of 17-(allylamino)-17-demethoxygeldanamycin (NSC 330507) in CD2F1 mice. Cancer Chemother Pharmacol 2001; 47: 291-302. Article MEDLINE

107 Khan SM, Oliver RH, Dauffenbach LM, Yeh J. Depletion of Raf-1 protooncogene by geldanamycin causes apoptosis in human luteinized granulosa cells. Fertil Steril 2000; 74: 359-365. MEDLINE

108 Wilson RH, Takimoto CH, Agnew EB, Morrison G, Grollman F, Thomas RR, Saif MW, Hopkins J, Allegra C, Grochow L, Szabo E, Hamilton JM, Brian P, Monahan BP, Neckers L, Grem JL. Demethoxygeldanamycin (AAG) in adult patients with advanced solid tumors. Proc Am Soc Clin Oncol 2001; (Abstr. 325).

109 Munster PN, Tong W, Schwartz L, Larson S, Kenneson K, De La Cruz A, Rosen N, Scher H. Phase I trial of 17-(allylamino)-17-demethoxygeldanamycin (17-AAG) in patients (Pts) with advanced solid malignancies. Proc Am Soc Clin Oncol 2001; (Abstr. 327).

110 Banerji U, O'Donnell A, Scurr M, Benson C, Hanwell J, Clark S, Raynaud F, Turner A, Walton M, Workman P, Judson I. Phase I trial of the heat shock protein 90 (HSP90) inhibitor 17-allylamino 17-demethoxygeldanamycin 17aag). Pharmacokinetic (PK) profile and pharmacodynamic (PD) endpoints. Proc Am Soc Clin Oncol 2001; (Abstr. 326).

111 Kennedy MJ, Armstrong DK, Huelskamp AM, Ohly K, Clarke BV, Colvin OM, Grochow LB, Chen TL, Davidson NE. Phase I and pharmacologic study of the alkylating agent modulator novobiocin in combination with high-dose chemotherapy for the treatment of metastatic breast cancer. J Clin Oncol 1995; 13: 1136-1143. MEDLINE

112 Murren JR, DiStasio SA, Lorico A, McKeon A, Zuhowski EG, Egorin MJ, Sartorelli AC, Rappa G. Phase I and pharmacokinetic study of novobiocin in combination with VP-16 in patients with refractory malignancies. Cancer J 2000; 6: 256-265. MEDLINE

113 Stocker U, Schaefer A, Marquardt H. DMSO-like rapid decrease in c-myc and c-myb mRNA levels and induction of differentiation in HL-60 cells by the anthracycline antitumor antibiotic aclarubicin. Leukemia 1995; 9: 146-154. MEDLINE

114 Marcu MG, Schulte TW, Neckers L. Novobiocin and related coumarins and depletion of heat shock protein 90-dependent signaling proteins. J Natl Cancer Inst 2000; 92: 242-248. MEDLINE

115 Eder JP, Wheeler CA, Teicher BA, Schnipper LE. A phase I clinical trial of novobiocin, a modulator of alkylating agent cytotoxicity. Cancer Res 1991; 51: 510-513. MEDLINE

116 Breivik J. Don't stop for repair in a war zone: darwinian evolution unites genes and environment in cancer development. Proc Natl Acad Sci USA 2001; 98: 5379-5381. MEDLINE

Figure 1 Mechanism of action of Hsp90-active drugs. Geldanamycin (GA) causes degradation of Hsp90-client proteins (eg Src). See text for details.

Figure 2 Hsp90-active drugs disrupt proliferative and antiapoptotic signaling pathways. Molecular targets of Hsp90-active drugs are in bold.