HSF1 Pathway Inhibitor Clinical Candidate (CCT361814/NXP800) Developed from a Phenotypic Screen as a Potential Treatment for Refractory Ovarian Cancer and Other Malignancies

CCT251236 1, a potent chemical probe, was previously developed from a cell-based phenotypic high-throughput screen (HTS) to discover inhibitors of transcription mediated by HSF1, a transcription factor that supports malignancy. Owing to its activity against models of refractory human ovarian cancer, 1 was progressed into lead optimization. The reduction of P-glycoprotein efflux became a focus of early compound optimization; central ring halogen substitution was demonstrated by matched molecular pair analysis to be an effective strategy to mitigate this liability. Further multiparameter optimization led to the design of the clinical candidate, CCT361814/NXP800 22, a potent and orally bioavailable fluorobisamide, which caused tumor regression in a human ovarian adenocarcinoma xenograft model with on-pathway biomarker modulation and a clean in vitro safety profile. Following its favorable dose prediction to human, 22 has now progressed to phase 1 clinical trial as a potential future treatment for refractory ovarian cancer and other malignancies.

Abstract 3722: Exploring the role of eIF4E in cancer cells with targeted protein degradation

Initiation of translation is considered the main rate-limiting step of protein synthesis and requires the recognition of 5’ m7G-cap on mature mRNAs and the formation of eukaryotic translation initiation factor 4F (eIF4F) multi-protein mRNA cap-binding complex. Formation of this complex requires the interaction of eIF4E and the scaffold protein eIF4G, and RNA helicase eIF4A. This eIF4F complex along with eIF3 mediate the recruitment of the 40S ribosomal particle to the 5′ cap of mRNA. Activation of eIF4E is a regulatory hub of many major oncogenic pathways, thus, targeting eIF4E has emerged as a potential therapeutic strategy in cancer.

Here we have used a targeted protein degradation approach coupled with genetic rescue to explore the molecular and cellular dependency of cancer cells on eIF4E. Stable H1299 human NSCLC clones expressing FKBP12F36V-tagged eIF4E but lacking endogenous eIF4E were established. Treatment of multiple N- or C-tagged-eIF4E clones with dTAGv-1, an FKBP12F36V selective heterobifunctional molecule that recruits VHL, induced rapid degradation of eIF4E to undetectable levels by 6hr exposure. This also resulted in reduced expression of MCL1, a previously reported biomarker of eIF4E activity. Longer exposures to dTAGv-1 resulted in a cytostasis that was not associated with cell death. A diastereomer negative control of dTAGv-1 that cannot recruit VHL did not elicit loss of eIF4E or the downstream events associated with its loss. Global analysis of protein synthesis initiation by RIBOseq and proteome profiling following dTAGv-1 treatment out to 32hr exposure demonstrated surprisingly few alterations in protein expression despite the significant effect on cancer cell growth.

We also expressed wild-type or eIF4E mutants predicted to disrupt key functions and determined their ability to rescue molecular or cellular phenotype associate with eIF4E-loss following dTAGv-1 treatment. Expression of wild-type eIF4E completely rescued cell growth and MCL1 expression. A W56A mutant predicted to disrupt mRNA-cap binding was unable to rescue eIF4E loss. In contrast, expression of W73F or S290A mutants (predicted to disrupt eIF4G binding or exhibit reduced eIF4E activity, respectively) were able to rescue the loss of eIF4E.

In summary, our rescue experiments show that mRNA-cap binding by eIF4E is required and that eIF4E:eIF4G interaction in cells may be more complex than predicted. This may also explain the challenges associated with developing selective and cellularly potent inhibitors of the eIF4E:eIF4G interaction and that targeting mRNA-cap binding may be a more effective strategy. We predicted that removing eIF4E would impact on the global synthesis of many proteins. However, our data demonstrate that targeting eIF4E leads to limited effects on protein synthesis that remain sufficient to inhibit cancer cell growth. Further experiments are underway to understand which proteins drive this dependency on eIF4E.

Citation Format: Swee Y. Sharp, Marianna Martella, Christopher I. Milton, George Ward, Caroline Richardson, Andrew Woodhead, Paul A. Clarke. Exploring the role of eIF4E in cancer cells with targeted protein degradation. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3722.

Solution structure of the Hop TPR2A domain and investigation of target druggability by NMR, biochemical and in silico approaches

Heat shock protein 90 (Hsp90) is a molecular chaperone that plays an important role in tumour biology by promoting the stabilisation and activity of oncogenic 'client' proteins. Inhibition of Hsp90 by small-molecule drugs, acting via its ATP hydrolysis site, has shown promise as a molecularly targeted cancer therapy. Owing to the importance of Hop and other tetratricopeptide repeat (TPR)-containing cochaperones in regulating Hsp90 activity, the Hsp90-TPR domain interface is an alternative site for inhibitors, which could result in effects distinct from ATP site binders. The TPR binding site of Hsp90 cochaperones includes a shallow, positively charged groove that poses a significant challenge for druggability. Herein, we report the apo, solution-state structure of Hop TPR2A which enables this target for NMR-based screening approaches. We have designed prototype TPR ligands that mimic key native 'carboxylate clamp' interactions between Hsp90 and its TPR cochaperones and show that they block binding between Hop TPR2A and the Hsp90 C-terminal MEEVD peptide. We confirm direct TPR-binding of these ligands by mapping 1H-15N HSQC chemical shift perturbations to our new NMR structure. Our work provides a novel structure, a thorough assessment of druggability and robust screening approaches that may offer a potential route, albeit difficult, to address the chemically challenging nature of the Hop TPR2A target, with relevance to other TPR domain interactors.

Correction to: An in vitro and in vivo study of the combination of the heat shock protein inhibitor 17-allylamino-17-demethoxygeldanamycin and carboplatin in human ovarian cancer models

The corresponding author of this article has informed us of concerns about the immunoblots in Fig. 2 which were carried out in the collaborating laboratory of Professor Ann Jackman.

Abstract 2976: Confirmation of in-cell target engagement using the proteolysis targeting chimeras (PROTACs) against pirin

We recently reported the identification of the original bisamide lead compound CCT251236 as an inhibitor of the HSF1 stress pathway with a high affinity for the putative transcription factor co-regulator, pirin (SPR KD=44nM) (Cheeseman et al., J Med Chem, 60; 180-201, 2017). Pirin is a highly conserved non-heme iron-binding regulatory protein that is a member of the functionally diverse cupin superfamily, but has no known enzymatic function or biomarkers of activity. To understand further this poorly characterized protein and to confirm that CCT251236 binds to pirin within living cells, we conceived and optimized a heterobifunctional protein degradation probe using the proteolysis targeting chimeras (PROTACs; CCT367766) comprising a pirin-binding moiety linked to the cereblon-targeting ligand thalidomide. This PROTAC molecule was designed to recruit pirin to the E3 ubiquitin ligase cereblon resulting in the ubiquitylation and degradation of pirin. Negative control probes lacking binding to pirin (CCT367857) or cereblon (CCT367936) were also designed and synthesized. We demonstrated a concentration-dependent depletion of pirin protein from as low as 0.5nM and as early as 2 hr treatment of SKOV3 human ovarian cancer cells with the PROTAC. The negative controls CCT367857 and CCT367936 exhibited no pirin depletion at equimolar concentrations. At higher concentrations of the active probe, a hook effect is observed, consistent with the formation of a ternary complex. Degradation of pirin by the PROTAC was confirmed to be proteasome-dependent by rescue of depletion following pre-incubation with the proteasome inhibitor MG132. In addition, the PROTAC could not induce pirin degradation in CRISPR/cas9 cereblon knockout SKOV3 cells, confirming dependence on cereblon. Pre-treatment with the bisamide compound CCT251236 or free thalidomide abrogated the PROTAC-induced pirin degradation, consistent with pirin and cereblon engagement. Finally, to estimate the cellular selectivity of the PROTAC to pirin in an unbiased manner, we carried out whole proteome mass spectrometry in SKOV3 cells. From 8547 quantifiable proteins identified, only pirin (2.3-fold reduction) displayed a statistically significant (Padj<0.05) difference in protein expression, indicating impressive selectivity. In summary, we have designed a PROTAC as an intracellular probe against a poorly understood molecular target, pirin. This approach has allowed us to confirm in-cell target engagement of our bisamide lead CCT251236 with pirin and validates CCT367766 as a PROTAC tool to further study this largely unexplored protein. Our results also provide a systematic approach for the use of the powerful PROTAC technology to investigate potential and poorly understood cancer drug targets.

Citation Format: Swee Y. Sharp, Nicola E. Chessum, John J. Caldwell, Marissa V. Powers, A Elisa Pasqua, Birgit Wilding, Ian Collins, Bugra Ozer, Martin Rowlands, Mark Stubbs, Rosemary Burke, Rob L. van Montfort, Matthew D. Cheeseman, Paul A. Clarke, Paul Workman, Keith Jones. Confirmation of in-cell target engagement using the proteolysis targeting chimeras (PROTACs) against pirin [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2976.

Demonstrating In-Cell Target Engagement Using a Pirin Protein Degradation Probe (CCT367766)

Demonstrating intracellular protein target engagement is an essential step in the development and progression of new chemical probes and potential small molecule therapeutics. However, this can be particularly challenging for poorly studied and noncatalytic proteins, as robust proximal biomarkers are rarely known. To confirm that our recently discovered chemical probe 1 (CCT251236) binds the putative transcription factor regulator pirin in living cells, we developed a heterobifunctional protein degradation probe. Focusing on linker design and physicochemical properties, we generated a highly active probe 16 (CCT367766) in only three iterations, validating our efficient strategy for degradation probe design against nonvalidated protein targets.

Abstract LB-304: Discovery of chemical probe CCT251236: An orally bioavailable efficacious pirin ligand from an HSF1 phenotypic screen

Heat shock factor 1 (HSF1) was originally identified as a master regulator of the classical ‘cytoprotective’ heat shock response. However, a large body of evidence has now verified the importance of HSF1 to tumorigenesis and cancer progression. HSF1 is activated by various elements of the cancer state, reprogramming the transcriptome in a way that is overlapping with, but distinct from, the canonical heat-shock response. Also, there is a strong correlation between the expression of activated HSF1 in tumors and adverse clinical outcomes. This evidence indicates that the inhibition of HSF1-mediated transcription could be a viable strategy in cancer treatment. Inhibiting the HSF1 stress pathway represents an attempt at targeting non-oncogene addiction and proteotoxic stress, which has been proposed to be advantageous. However, HSF1 is a ligandless transcription factor and is unlikely to be amenable to standard drug discovery strategies and direct inhibition with small molecules. Therefore, we proposed that inhibitors of HSF1-mediated transcription, which antagonize the HSF1 pathway but without necessarily binding directly to HSF1, could be discovered and developed via a cell-based phenotypic screen. We carried out a high throughput Arrayscan assay of 200,000 compounds to measure the inhibition of HSF1-mediated HSP72 expression stimulated by pre-treatment with an HSP90 inhibitor. We identified a singleton hit with a bisamide core, CCT245232. This compound showed potent growth inhibition in a range of human cancer cell lines but had poor physicochemical properties leading to an unacceptable pharmacokinetic profile. Improvement of the physicochemical properties of CCT245232 whilst maintaining potency versus our cell-based assays led to the orally bioavailable tool compound CCT251236. This compound shows potent growth inhibition (GI50 values in low nanomolar range) of human ovarian cancer cell lines in vitro and good efficacy against human ovarian cancer xenografts in nude mice in vivo. We applied chemo-proteomic strategies to identify the molecular target using a probe based on CCT251236 and discovered pirin as a high affinity molecular target. Binding of CCT251236 to recombinant pirin was confirmed in biophysical assays. CCT251236 recapitulates the reported anti-migratory phenotype for a pirin ligand although binding to pirin alone does not explain the cellular phenotype observed with our chemical tool. We are currently using CCT251236 as a chemical probe while further optimizing its properties to identify a clinical candidate.

Citation Format: Matthew D. Cheeseman, Nicola E. Chessum, Carl S. Rye, Elisa A. Pasqua, Michael J. Tucker, Birgit Wilding, Lindsay E. Evans, Susan Lepri, Meirion Richards, Swee Y. Sharp, Salyha Ali, Martin Rowlands, Lisa O'Fee, Asadh Miah, Angela Hayes, Alan T. Henley, Marissa Powers, Robert te Poele, Emmanuel De Billy, Loredana Pellegrino, Florence Raynaud, Rosemary Burke, Robert L. van Montfort, Suzanne A. Eccles, Keith Jones, Paul Workman. Discovery of chemical probe CCT251236: An orally bioavailable efficacious pirin ligand from an HSF1 phenotypic screen [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-304. doi:10.1158/1538-7445.AM2017-LB-304

Discovery of a Chemical Probe Bisamide (CCT251236): An Orally Bioavailable Efficacious Pirin Ligand from a Heat Shock Transcription Factor 1 (HSF1) Phenotypic Screen

Phenotypic screens, which focus on measuring and quantifying discrete cellular changes rather than affinity for individual recombinant proteins, have recently attracted renewed interest as an efficient strategy for drug discovery. In this article, we describe the discovery of a new chemical probe, bisamide (CCT251236), identified using an unbiased phenotypic screen to detect inhibitors of the HSF1 stress pathway. The chemical probe is orally bioavailable and displays efficacy in a human ovarian carcinoma xenograft model. By developing cell-based SAR and using chemical proteomics, we identified pirin as a high affinity molecular target, which was confirmed by SPR and crystallography.

Discovery of 4,6-disubstituted pyrimidines as potent inhibitors of the heat shock factor 1 (HSF1) stress pathway and CDK9

Heat shock factor 1 (HSF1) is a transcription factor that plays key roles in cancer, including providing a mechanism for cell survival under proteotoxic stress. Therefore, inhibition of the HSF1-stress pathway represents an exciting new opportunity in cancer treatment. We employed an unbiased phenotypic screen to discover inhibitors of the HSF1-stress pathway. Using this approach we identified an initial hit (1) based on a 4,6-pyrimidine scaffold (2.00 μM). Optimisation of cellular SAR led to an inhibitor with improved potency (25, 15 nM) in the HSF1 phenotypic assay. The 4,6-pyrimidine 25 was also shown to have high potency against the CDK9 enzyme (3 nM).

Fused imidazoles as potential chemical scaffolds for inhibition of heat shock protein 70 and induction of apoptosis. Synthesis and biological evaluation of phenanthro[9,10-d]imidazoles and imidazo[4,5-f][1,10]phenanthrolines

Fused imidazoles inhibit growth of human cancer cell lines, and the Hsp70 pathway in cells, and induce apoptosis.

Correction: Discovery of 4,6-disubstituted pyrimidines as potent inhibitors of the heat shock factor 1 (HSF1) stress pathway and CDK9. See DOI: 10.1039/c6md00159a

Correction for ‘Discovery of 4,6-disubstituted pyrimidines as potent inhibitors of the heat shock factor 1 (HSF1) stress pathway and CDK9’ by Carl S. Rye et al., Med. Chem. Commun., 2016, 7, 1580–1586.

Effects of HSP90 inhibitor 17-allylamino-17-demethoxygeldanamycin (17-AAG) on NEU/HER2 overexpressing mammary tumours in MMTV-NEU-NT mice monitored by Magnetic Resonance Spectroscopy

BACKGROUND

The importance of ERBB2/NEU/HER2 in the response of breast tumours to the heat shock protein 90 (HSP90) inhibitor 17-allylamino-17-demethoxygeldanamycin (17-AAG; tanespimycin) has been demonstrated in the clinic. ERBB2 is an oncoprotein client that is highly dependent on HSP90. This and other oncogenic client proteins (e.g. B-RAF, C-RAF, ALK and CDK4) are depleted by 17-AAG in both animal tumours and patients. Here we investigate by Magnetic Resonance Spectroscopy (MRS) the metabolic response of 17-AAG in spontaneous, NEU/HER2 driven mammary tumours in transgenic MMTV-NEU-NT mice and in cells isolated and cultured from these tumours.

METHODS

Mammary tumours were monitored by 31P MRS in vivo and in tumour extracts, comparing control and 17-AAG treated mice. A cell line derived from NEU/HER2 mammary tumours was also cultured and the effect of 17-AAG was measured by 31P MRS in cell extracts. Molecular biomarkers were assessed by immunoblotting in extracts from cells and tumours. For comparison of tumour volume, metabolite concentrations and Western blot band intensities, two-tailed unpaired t-tests were used.

RESULTS

The NEU/HER2 mammary tumours were very sensitive to 17-AAG and responded in a dose-dependent manner to 3 daily doses of 20, 40 and 80mg/kg of 17-AAG, all of which caused significant regression. At the higher doses, 31P MRS of tumour extracts showed significant decreases in phosphocholine (PC) and phosphoethanolamine (PE) whereas no significant changes were seen at the 20mg/kg dose. Extracts of isolated cells cultured from the mammary carcinomas showed a significant decrease in viable cell number and total PME after 17-AAG treatment. Western blots confirmed the expected action of 17-AAG in inducing HSP72 and significantly depleting HSP90 client proteins, including NEU/HER2 both in tumours and in isolated cells.

CONCLUSIONS

The data demonstrate the high degree of sensitivity of this clinically relevant NEU/HER2-driven tumour model to HSP90 inhibition by 17-AAG, consistent with the clinical data, and suggest that the metabolic signature of choline phospholipids obtained by MRS could be useful both as a preclinical and clinical tool for investigating surrogate markers of response to treatment.

Abstract 4749: Insights into the molecular mechanism of HSP90 binding of methoxy-substituted resorcinylic isoxazole amide inhibitors reveal different isoform selectivity profiles

The molecular chaperone Heat Shock Protein 90 (HSP90) regulates the conformation, stability and activity of several oncogenic client proteins (eg ERBB2, ALK, BRAF, CRAF and AKT). Inhibition of HSP90 results in the degradation of these clients, leading to cell cycle arrest and apoptosis. The resorcinylic isoxazole amide HSP90 inhibitor NVP-AUY922 is currently in Phase I/II clinical trials and other related agents are in development. We previously reported that one or both of the phenolic groups on the resorcinol ring are important for HSP90 inhibitory activity. To further characterise the importance of the phenol groups on protein binding and biological activity, a series of mono- and dimethoxy-substituted resorcinylic 5′-ethyl isoxazole amides were synthesized and compared with the 2′,4′-dihydroxy compound by X-ray crystal structure and biological techniques. The most potent HSP90 inhibitors in terms of binding to the HSP90α target was the 2′,4′-dihydroxy-5′-ethyl isoxazole CCT239215 (Kd=11nM), followed by the 2′-hydroxy-4′-methoxy analogue CCT078722 (Kd=36nM). In contrast, the 2′-methoxy-4′-hydroxy and 2′,4′-dimethoxy derivatives showed no binding to HSP90. X-ray crystal structures showed that for CCT078722, the 4′-methoxy group of the resorcinol ring disrupts the hydrogen bonding that is seen with 2′,4′-dihydroxy CCT239215 and NVP-AUY922, and instead is involved in van der Waals contacts. CCT239215 also showed the greatest cellular growth inhibitory potency (GI50=5-16nM in a panel of human cancer cell lines), followed by CCT078722 (GI50=59-588nM). As predicted, CCT078723 and CCT078721 were inactive in the cells (GI50>10µM). This structure-activity relationship was also maintained with respect to biomarkers of HSP90 target inhibition, as measured by depletion of HSP90 client proteins and induction of HSP72. In HCT116 human colon cells, CCT239215 and CCT078722 caused a loss of HSP90 from immunoprecipitates of the HSP90 co-chaperone, demonstrating dissociation between P23 and HSP90, further confirming their on-target effects. No such dissociation was observed with the inactive analogs. These results explain how the hydroxy group at the 2′-position on the resorcinol ring is crucial for HSP90 inhibitory activity. Interestingly, the 2′-hydroxy-4′-methoxy-5′-ethyl isoxazole CCT078722 exhibited 55-fold less affinity for GRP94 when compared to HSP90α/α, indicating that these compounds show potential for different isoform selectivity profiles. Modelling studies suggest that IIe247 in GRP94 negatively impacts on the binding of CCT078722. Our findings offer insights into the molecular mechanism of HSP90 binding by resorcinylic isoxazole inhibitors, provide active/inactive probe compounds for mechanistic studies and suggest potential approaches to isoform selectivity.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4749. doi:1538-7445.AM2012-4749

Targeting the Hsp90 molecular chaperone with novel macrolactams. Synthesis, structural, binding, and cellular studies

A series of resorcylic acid macrolactams, nitrogen analogues of the naturally occurring macrolactone radicicol, have been prepared by chemical synthesis and evaluated as inhibitors of heat shock protein 90 (Hsp90), an emerging attractive target for novel cancer therapeutic agents. The synthesis involves, as key steps, ring opening of an isocoumarin intermediate, followed by a ring-closing metathesis reaction to form the macrocycle. Subsequent manipulation of the ester group into a range of amides allows access to a range of new macrolactams following deprotection of the two phenolic groups. These new resorcylic acid lactams exhibit metabolic stability greater than that of related lactone counterparts, while co-crystallization of three macrolactams with the N-terminal domain ATP site of Hsp90 confirms that they bind in a similar way to the natural product radicicol and to our previous synthetic lactone analogues. Interestingly, however, in the case of the N-benzylamide, additional binding to a hydrophobic pocket of the protein was observed. In biological assays, the new macrocyclic lactams exhibit a biological profile equivalent or superior to that of the related lactones and show the established molecular signature of Hsp90 inhibitors in human colon cancer cells.

Modulation of melanoma cell phospholipid metabolism in response to heat shock protein 90 inhibition

Molecular chaperone heat shock protein 90 (Hsp90) inhibitors are promising targeted cancer therapeutic drugs, with the advantage that they deplete multiple oncogenic client proteins and modulate all the classical hallmarks of cancer. They are now in clinical trial and show potential for activity in melanoma and other malignancies. Here we explore the metabolic response to Hsp90 inhibition in human melanoma cells using magnetic resonance spectroscopy. We show that, concomitant with growth inhibition and re-differentiation, Hsp90 inhibition in human melanoma cells is associated with increased glycerophosphocholine content. This was seen with both the clinical geldanamycin-based Hsp90 drug 17-AAG and the structurally dissimilar Hsp90 inhibitor CCT018159. The effect was noted in both BRAF mutant SKMEL28 and BRAF wildtype CHL-1 melanoma cells. Elevated content of the -CH2+CH3 fatty acyl chains and cytoplasmic mobile lipid droplets was also observed in 17-AAG-treated SKMEL28 cells. Importantly, the phospholipase A2 inhibitor bromoenol lactone prevented the rise in glycerophosphocholine seen with 17-AAG, suggesting a role for phospholipase A2 activation in the Hsp90 inhibitor-induced metabolic response. Our findings provide a basis for using metabolic changes as non-invasive indicators of Hsp90 inhibition and potentially as biomarkers of anticancer activity with Hsp90 drugs in malignant melanoma and possibly in other cancers.

MGMT-independent temozolomide resistance in pediatric glioblastoma cells associated with a PI3-kinase-mediated HOX/stem cell gene signature

Sensitivity to temozolomide is restricted to a subset of glioblastoma patients, with the major determinant of resistance being a lack of promoter methylation of the gene encoding the repair protein DNA methyltransferase MGMT, although other mechanisms are thought to be active. There are, however, limited preclinical data in model systems derived from pediatric glioma patients. We screened a series of cell lines for temozolomide efficacy in vitro, and investigated the differential mechanisms of resistance involved. In the majority of cell lines, a lack of MGMT promoter methylation and subsequent protein overexpression were linked to temozolomide resistance. An exception was the pediatric glioblastoma line KNS42. Expression profiling data revealed a coordinated upregulation of HOX gene expression in resistant lines, especially KNS42, which was reversed by phosphoinositide 3-kinase pathway inhibition. High levels of HOXA9/HOXA10 gene expression were associated with a shorter survival in pediatric high-grade glioma patient samples. Combination treatment in vitro of pathway inhibition and temozolomide resulted in a highly synergistic interaction in KNS42 cells. The resistance gene signature further included contiguous genes within the 12q13-q14 amplicon, including the Akt enhancer PIKE, significantly overexpressed in the KNS42 line. These cells were also highly enriched for CD133 and other stem cell markers. We have thus shown an in vitro link between phosphoinositide 3-kinase-mediated HOXA9/HOXA10 expression, and a drug-resistant, progenitor cell phenotype in MGMT-independent pediatric glioblastoma.

Mechanistic evaluation of the novel HSP90 inhibitor NVP-AUY922 in adult and pediatric glioblastoma

The dismal prognosis of glioblastoma (GB) indicates the urgent need for new therapies for these tumors. Heat shock protein 90 (HSP90) inhibitors induce the proteasome-mediated degradation of many oncogenic client proteins involved in all of the hallmark characteristics of cancer. Here, we explored the mechanistic potential of the potent synthetic diarylisoxazole amide resorcinol HSP90 inhibitor, NVP-AUY922, in adult and pediatric GB. In vitro antiproliferative potency (nanomolar range) was seen in both adult and pediatric human GB cell lines with different molecular pathologies. A cytostatic effect was observed in all GB lines; more apoptosis was observed at lower concentrations in the SF188 pediatric GB line and at 144 hours in the slower growing KNS42 pediatric GB line, as compared with the adult GB lines U87MG and SF268. In vitro combination studies with inhibitors of phosphoinositide 3-kinase/mammalian target of rapamycin (PI-103) or mitogen-activated protein/extracellular signal-regulated kinase (ERK) kinase (PD-0325901) supported the hypothesis that sustained inhibition of ERK up to 72 hours and at least temporary inhibition of AKT were necessary to induce apoptosis in GB lines. In athymic mice bearing established s.c U87MG GB xenografts, NVP-AUY922 (50 mg/kg i.p x 3 days) caused the inhibition of ERK1/2 and AKT phosphorylation and induced apoptosis, whereas 17-AAG used at maximum tolerated dose was less effective. NVP-AUY922 antitumor activity with objective tumor regression resulted from antiproliferative, proapoptotic, and antiangiogenic effects, the latter shown by decreased microvessel density and HIF1alpha levels. Our results have established a mechanistic proof of concept for the potential of novel synthetic HSP90 inhibitors in adult and pediatric GB, alone or in combination with phosphoinositide 3-kinase/mammalian target of rapamycin and mitogen-activated protein/ERK kinase inhibitors.

Abstract 2677: A detailed analysis of protein binding and biological activity of methoxy-substituted resorcinylic isoxazole amide HSP90 inhibitors

The molecular chaperone Heat Shock Protein 90 (HSP90) has emerged as an exciting biological target in cancer therapy. HSP90 regulates the conformation, stability and activity of several client proteins such as ERBB2, BRAF, CRAF, AKT and mutant p53, many of which are associated with the six hallmarks of cancer. Inhibition of HSP90 results in the degradation of these clients by the 26S proteasome, leading to cell cycle arrest and apoptosis. Resorcinylic isoxazole amide HSP90 inhibitors show considerable promise and NVP-AUY922 is currently in Phase I clinical trials. Our previous studies have reported that one or both of the phenolic groups on the resorcinol ring are important for HSP90 inhibitory activity (Brough et al J Med Chem 51 196-218 2008). To explore the mechanisms underlying the importance of the phenol groups on protein binding and biological activity, a series of monomethoxy- and dimethoxy-substituted resorcinylic 5′-ethyl isoxazole amides were synthesized and compared with the 2′,4′-dihydroxy compound by protein X-ray crystal structure and biological techniques. The most potent HSP90 inhibitors in this series in terms of binding to the HSP90 target and inhibiting cancer cell proliferation was the 2′,4′-dihydroxy-5′ethyl isoxazole CCT239215, followed by the 2′-hydroxy-4′methoxy-5′ethyl isoxazole CCT078722. In contrast, 2′,4′-dimethoxy and 2′-methoxy, 4′-hydroxy derivatives showed no binding to HSP90 or antiproliferative activity. This structure-activity relationship was also maintained with respect to biomarkers of HSP90 target inhibition, as measured by depletion of HSP90 client proteins and induction of HSP72. In HCT116 human colon cells treatment with 350nM of CCT239215 and CCT078722 for 4hr caused a loss of HSP90 from P23 immunoprecipitates, demonstrating dissociation between P23 and HSP90, thereby confirming on-target effects. No dissociation was observed with the inactive 2′-methoxy, 4′-hydroxy or 2′,4′-dimethoxy analogs. These results confirm that the hydroxy group at the 2′ position on the resorcinol ring is crucial for HSP90 inhibitory activity. X-ray crystal structures showed that for CCT078722, the 4′-methoxy group of the resorcinol ring disrupts the hydrogen bonding that is seen with 2′,4′-dihydroxy CCT239215 and with the clinical candidate NVP-AUY922, and instead is involved in van der Waals contacts. Thus the results provide a structural explanation for the reduced but still significant activity of the 2′-methoxy-subtituted analog and the lack of activity of the 2′,4′-dimethoxy-substituted compound.

Note: This abstract was not presented at the AACR 101st Annual Meeting 2010 because the presenter was unable to attend.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2677.

Combining hit identification strategies: fragment-based and in silico approaches to orally active 2-aminothieno[2,3-d]pyrimidine inhibitors of the Hsp90 molecular chaperone

Inhibitors of the Hsp90 molecular chaperone are showing considerable promise as potential molecular therapeutic agents for the treatment of cancer. Here we describe novel 2-aminothieno[2,3-d]pyrimidine ATP competitive Hsp90 inhibitors, which were designed by combining structural elements of distinct low affinity hits generated from fragment-based and in silico screening exercises in concert with structural information from X-ray protein crystallography. Examples from this series have high affinity (IC50 = 50-100 nM) for Hsp90 as measured in a fluorescence polarization (FP) competitive binding assay and are active in human cancer cell lines where they inhibit cell proliferation and exhibit a characteristic profile of depletion of oncogenic proteins and concomitant elevation of Hsp72. Several examples (34a, 34d and 34i) caused tumor growth regression at well tolerated doses when administered orally in a human BT474 human breast cancer xenograft model.

Acquired resistance to 17-allylamino-17-demethoxygeldanamycin (17-AAG, tanespimycin) in glioblastoma cells

Heat shock protein 90 (HSP90) inhibitors, such as 17-allylamino-17-demethoxygeldanamycin (17-AAG, tanespimycin), which is currently in phase II/phase III clinical trials, are promising new anticancer agents. Here, we explored acquired resistance to HSP90 inhibitors in glioblastoma (GB), a primary brain tumor with poor prognosis. GB cells were exposed continuously to increased 17-AAG concentrations. Four 17-AAG-resistant GB cell lines were generated. High-resistance levels with resistance indices (RI = resistant line IC(50)/parental line IC(50)) of 20 to 137 were obtained rapidly (2-8 weeks). After cessation of 17-AAG exposure, RI decreased and then stabilized. Cross-resistance was found with other ansamycin benzoquinones but not with the structurally unrelated HSP90 inhibitors, radicicol, the purine BIIB021, and the resorcinylic pyrazole/isoxazole amide compounds VER-49009, VER-50589, and NVP-AUY922. An inverse correlation between NAD(P)H/quinone oxidoreductase 1 (NQO1) expression/activity and 17-AAG IC(50) was observed in the resistant lines. The NQO1 inhibitor ES936 abrogated the differential effects of 17-AAG sensitivity between the parental and resistant lines. NQO1 mRNA levels and NQO1 DNA polymorphism analysis indicated different underlying mechanisms: reduced expression and selection of the inactive NQO1*2 polymorphism. Decreased NQO1 expression was also observed in a melanoma line with acquired resistance to 17-AAG. No resistance was generated with VER-50589 and NVP-AUY922. In conclusion, low NQO1 activity is a likely mechanism of acquired resistance to 17-AAG in GB, melanoma, and, possibly, other tumor types. Such resistance can be overcome with novel HSP90 inhibitors.

NVP-AUY922: a novel heat shock protein 90 inhibitor active against xenograft tumor growth, angiogenesis, and metastasis

We describe the biological properties of NVP-AUY922, a novel resorcinylic isoxazole amide heat shock protein 90 (HSP90) inhibitor. NVP-AUY922 potently inhibits HSP90 (K(d) = 1.7 nmol/L) and proliferation of human tumor cells with GI(50) values of approximately 2 to 40 nmol/L, inducing G(1)-G(2) arrest and apoptosis. Activity is independent of NQO1/DT-diaphorase, maintained in drug-resistant cells and under hypoxic conditions. The molecular signature of HSP90 inhibition, comprising induced HSP72 and depleted client proteins, was readily demonstrable. NVP-AUY922 was glucuronidated less than previously described isoxazoles, yielding higher drug levels in human cancer cells and xenografts. Daily dosing of NVP-AUY922 (50 mg/kg i.p. or i.v.) to athymic mice generated peak tumor levels at least 100-fold above cellular GI(50). This produced statistically significant growth inhibition and/or regressions in human tumor xenografts with diverse oncogenic profiles: BT474 breast tumor treated/control, 21%; A2780 ovarian, 11%; U87MG glioblastoma, 7%; PC3 prostate, 37%; and WM266.4 melanoma, 31%. Therapeutic effects were concordant with changes in pharmacodynamic markers, including induction of HSP72 and depletion of ERBB2, CRAF, cyclin-dependent kinase 4, phospho-AKT/total AKT, and hypoxia-inducible factor-1alpha, determined by Western blot, electrochemiluminescent immunoassay, or immunohistochemistry. NVP-AUY922 also significantly inhibited tumor cell chemotaxis/invasion in vitro, WM266.4 melanoma lung metastases, and lymphatic metastases from orthotopically implanted PC3LN3 prostate carcinoma. NVP-AUY922 inhibited proliferation, chemomigration, and tubular differentiation of human endothelial cells and antiangiogenic activity was reflected in reduced microvessel density in tumor xenografts. Collectively, the data show that NVP-AUY922 is a potent, novel inhibitor of HSP90, acting via several processes (cytostasis, apoptosis, invasion, and angiogenesis) to inhibit tumor growth and metastasis. NVP-AUY922 has entered phase I clinical trials.

4,5-diarylisoxazole Hsp90 chaperone inhibitors: potential therapeutic agents for the treatment of cancer

Inhibitors of the Hsp90 molecular chaperone are showing considerable promise as potential chemotherapeutic agents for cancer. Here, we describe the structure-based design, synthesis, structure-activity relationships and pharmacokinetics of potent small-molecule inhibitors of Hsp90 based on the 4,5-diarylisoxazole scaffold. Analogues from this series have high affinity for Hsp90, as measured in a fluorescence polarization (FP) competitive binding assay, and are active in cancer cell lines where they inhibit proliferation and exhibit a characteristic profile of depletion of oncogenic proteins and concomitant elevation of Hsp72. Compound 40f (VER-52296/NVP-AUY922) is potent in the Hsp90 FP binding assay (IC50 = 21 nM) and inhibits proliferation of various human cancer cell lines in vitro, with GI50 averaging 9 nM. Compound 40f is retained in tumors in vivo when administered i.p., as evaluated by cassette dosing in tumor-bearing mice. In a human colon cancer xenograft model, 40f inhibits tumor growth by approximately 50%.

Inhibition of the heat shock protein 90 molecular chaperone in vitro and in vivo by novel, synthetic, potent resorcinylic pyrazole/isoxazole amide analogues

Although the heat shock protein 90 (HSP90) inhibitor 17-allylamino-17-demethoxygeldanamycin (17-AAG) shows clinical promise, potential limitations encourage development of alternative chemotypes. We discovered the 3,4-diarylpyrazole resorcinol CCT018159 by high-throughput screening and used structure-based design to generate more potent pyrazole amide analogues, exemplified by VER-49009. Here, we describe the detailed biological properties of VER-49009 and the corresponding isoxazole VER-50589. X-ray crystallography showed a virtually identical HSP90 binding mode. However, the dissociation constant (K(d)) of VER-50589 was 4.5 +/- 2.2 nmol/L compared with 78.0 +/- 10.4 nmol/L for VER-49009, attributable to higher enthalpy for VER-50589 binding. A competitive binding assay gave a lower IC(50) of 21 +/- 4 nmol/L for VER-50589 compared with 47 +/- 9 nmol/L for VER-49009. Cellular uptake of VER-50589 was 4-fold greater than for VER-49009. Mean cellular antiproliferative GI(50) values for VER-50589 and VER-49009 for a human cancer cell line panel were 78 +/- 15 and 685 +/- 119 nmol/L, respectively, showing a 9-fold potency gain for the isoxazole. Unlike 17-AAG, but as with CCT018159, cellular potency of these analogues was independent of NAD(P)H:quinone oxidoreductase 1/DT-diaphorase and P-glycoprotein expression. Consistent with HSP90 inhibition, VER-50589 and VER-49009 caused induction of HSP72 and HSP27 alongside depletion of client proteins, including C-RAF, B-RAF, and survivin, and the protein arginine methyltransferase PRMT5. Both caused cell cycle arrest and apoptosis. Extent and duration of pharmacodynamic changes in an orthotopic human ovarian carcinoma model confirmed the superiority of VER-50589 over VER-49009. VER-50589 accumulated in HCT116 human colon cancer xenografts at levels above the cellular GI(50) for 24 h, resulting in 30% growth inhibition. The results indicate the therapeutic potential of the resorcinylic pyrazole/isoxazole amide analogues as HSP90 inhibitors.

Heat shock protein inhibition is associated with activation of the unfolded protein response pathway in myeloma plasma cells

Plasma cells producing high levels of paraprotein are dependent on the unfolded protein response (UPR) and chaperone proteins to ensure correct protein folding and cell survival. We hypothesized that disrupting client-chaperone interactions using heat shock protein 90 (Hsp90) inhibitors would result in an inability to handle immunoglobulin production with the induction of the UPR and myeloma cell death. To study this, myeloma cells were treated with Hsp90 inhibitors as well as known endoplasmic reticulum stress inducers and proteasome inhibitors. Treatment with thapsigargin and tunicamycin led to the activation of all 3 branches of the UPR, with early splicing of XBP1 indicative of IRE1 activation, upregulation of CHOP consistent with ER resident kinase (PERK) activation, and activating transcription factor 6 (ATF6) splicing. 17-AAG and radicicol also induced splicing of XBP1, with the induction of CHOP and activation of ATF6, whereas bortezomib resulted in the induction of CHOP and activation of ATF6 with minimal effects on XBP1. After treatment with all drugs, expression levels of the molecular chaperones BiP and GRP94 were increased. All drugs inhibited proliferation and induced cell death with activation of JNK and caspase cleavage. In conclusion, Hsp90 inhibitors induce myeloma cell death at least in part via endoplasmic reticulum stress and the UPR death pathway.

In vitro Biological Characterization of a Novel, Synthetic Diaryl Pyrazole Resorcinol Class of Heat Shock Protein 90 Inhibitors

The molecular chaperone heat shock protein 90 (HSP90) has emerged as an exciting molecular target. Derivatives of the natural product geldanamycin, such as 17-allylamino-17-demethoxy-geldanamycin (17-AAG), were the first HSP90 ATPase inhibitors to enter clinical trial. Synthetic small-molecule HSP90 inhibitors have potential advantages. Here, we describe the biological properties of the lead compound of a new class of 3,4-diaryl pyrazole resorcinol HSP90 inhibitor (CCT018159), which we identified by high-throughput screening. CCT018159 inhibited human HSP90beta with comparable potency to 17-AAG and with similar ATP-competitive kinetics. X-ray crystallographic structures of the NH(2)-terminal domain of yeast Hsp90 complexed with CCT018159 or its analogues showed binding properties similar to radicicol. The mean cellular GI(50) value of CCT018159 across a panel of human cancer cell lines, including melanoma, was 5.3 mumol/L. Unlike 17-AAG, the in vitro antitumor activity of the pyrazole resorcinol analogues is independent of NQO1/DT-diaphorase and P-glycoprotein expression. The molecular signature of HSP90 inhibition, comprising increased expression of HSP72 protein and depletion of ERBB2, CDK4, C-RAF, and mutant B-RAF, was shown by Western blotting and quantified by time-resolved fluorescent-Cellisa in human cancer cell lines treated with CCT018159. CCT018159 caused cell cytostasis associated with a G(1) arrest and induced apoptosis. CCT018159 also inhibited key endothelial and tumor cell functions implicated in invasion and angiogenesis. Overall, we have shown that diaryl pyrazole resorcinols exhibited similar cellular properties to 17-AAG with potential advantages (e.g., aqueous solubility, independence from NQO1 and P-glycoprotein). These compounds form the basis for further structure-based optimization to identify more potent inhibitors suitable for clinical development.

Inhibition of Hsp90 with synthetic macrolactones: synthesis and structural and biological evaluation of ring and conformational analogs of radicicol

A series of benzo-macrolactones of varying ring size and conformation has been prepared by chemical synthesis and evaluated by structural and biological techniques. Thus, 12- to 16-membered lactones were obtained by concise routes, involving ring-closing metathesis as a key step. In enzyme assays, the 13-, 15-, and 16-membered analogs are good inhibitors, suggesting that they can adopt the required conformation to fit in the ATP-binding site. This was confirmed by cocrystallization of 13-, 14-, and 15-membered lactones with the N-terminal domain of yeast Hsp90, showing that they bind similarly to the "natural" 14-membered radicicol. The most active compounds in the ATPase assays also showed the greatest growth-inhibitory potency in HCT116 human colon cancer cells and the established molecular signature of Hsp90 inhibition, i.e., depletion of client proteins with upregulation of Hsp70.

Synthetic ansamycins prepared by a ring-expanding Claisen rearrangement. Synthesis and biological evaluation of ring and conformational analogues of the Hsp90 molecular chaperone inhibitor geldanamycin

A series of ansa-quinones has been prepared by chemical synthesis, and evaluated by biological techniques. Thus, 19-membered ansa-lactams, simplified analogues of the naturally occurring Hsp90 molecular chaperone inhibitor geldanamycin, were obtained by concise routes, the key steps being the combination of a ring-closing metathesis to give a 17-membered ring followed by Claisen rearrangement to effect ring expansion. The methodology was also used to prepare an "unnatural" 18-membered ring analogue. In ATPase enzyme assays, the synthetic ansa-quinones were weak inhibitors of Hsp90.

The identification, synthesis, protein crystal structure and in vitro biochemical evaluation of a new 3,4-diarylpyrazole class of Hsp90 inhibitors

High-throughput screening identified the 3,4-diarylpyrazole CCT018159 as a novel and potent (7.1 microM) inhibitor of Hsp90 ATPase activity. Here, we describe the synthesis of CCT018159 and a number of close analogues together with data on their biochemical properties. Some initial structure-activity relationships are discussed, as well as the crystal structure of CCT018159 bound to Hsp90.

Retention of activity by the new generation platinum agent AMD0473 in four human tumour cell lines possessing acquired resistance to oxaliplatin

Four models of acquired resistance to the clinically-used platinum drug, oxaliplatin, have been established using human tumour cell lines in vitro; two colon (HCT116 and HT29) and two ovarian (A2780 and CH1). Levels of acquired resistance ranged from 3.0- to 15.8-fold with levels of resistance higher in the colon relative to the ovarian carcinoma cell lines. Notably, the platinum analogue, AMD0473, currently undergoing clinical evaluation, exhibited superior circumvention of acquired oxaliplatin resistance in comparison to either cisplatin or the trinuclear platinum BBR3464. Resistance in the two colon cell lines was unique to oxaliplatin itself among the platinum drugs studied. Acquired oxaliplatin resistance was not due to either reduced drug membrane transport or increased levels of glutathione in any of the four resistant lines. Following exposure to oxaliplatin, a lower level of platinum-DNA adducts was present in acquired oxaliplatin-resistant HT29 cells. In the remaining resistant lines, there was no change in the levels of platinum-DNA adducts relative to the parent lines. There was no change in hMLH1 DNA mismatch repair gene status in any of the four cell line pairs. However, in an A2780 subline where loss of hMLH1 and a p53phe172 mutation occurred, 5-fold resistance to cisplatin was observed, but only 1.7-fold resistance to oxaliplatin and no resistance to AMD0473 were observed. Re-introduction of hMLH1 into these cells caused no significant change in the sensitivity to cisplatin, oxaliplatin or AMD0473. These data show that acquired resistance to oxaliplatin may occur in cell lines (and therefore probably in the clinic) and in the four independent cell lines studied this was circumvented by AMD0473. Alongside previously described models of acquired resistance to cisplatin, these oxaliplatin-resistant cell line models may be useful in the evaluation of further novel platinum agents.

Establishment of an isogenic human colon tumor model for NQO1 gene expression: application to investigate the role of DT-diaphorase in bioreductive drug activation in vitro and in vivo

Many tumors overexpress the NQO1 gene, which encodes DT-diaphorase (NADPH:quinone oxidoreductase; EC 1.6.99.2). This obligate two-electron reductase deactivates toxins and activates bioreductive anticancer drugs. We describe the establishment of an isogenic human tumor cell model for DT-diaphorase expression. An expression vector was used in which the human elongation factor 1alpha promoter produces a bicistronic message containing the genes for human NQO1 and puromycin resistance. This was transfected into the human colon BE tumor line, which has a disabling point mutation in NQO1. Two clones, BE2 and BE5, were selected that were shown by immunoblotting and enzyme activity to stably express high levels of DT-diaphorase. Drug response was determined using 96-h exposures compared with the BE vector control. Functional validation of the isogenic model was provided by the much greater sensitivity of the NQO1-transfected cells to the known DT-diaphorase substrates and bioreductive agents streptonigrin (113- to 132-fold) and indoloquinone EO9 (17- to 25-fold) and the inhibition of this potentiation by the DT-diaphorase inhibitor dicoumarol. A lower degree of potentiation was seen with the clinically used agent mitomycin C (6- to 7-fold) and the EO9 analogs, EO7 and EO2, that are poorer substrates for DT-diaphorase (5- to 8-fold and 2- to 3-fold potentiation, respectively), and there was no potentiation or protection with menadione and tirapazamine. Exposure time-dependent potentiation was seen with the diaziquone analogs methyl-diaziquone and RH1 [2, 5-diaziridinyl-3-(hydroxymethyl)-6-methyl-1,4-benzoquinone], the latter being an agent in preclinical development. In contrast to the in vitro potentiation, there was no difference in the response to mitomycin C when BE2 and BE vector control were treated as tumor xenografts in vivo. This isogenic model should be valuable for mechanistic studies and bioreductive drug development.

Mechanisms of drug resistance to the platinum complex ZD0473 in ovarian cancer cell lines

Acquired drug resistance to the sterically hindered platinum drug ZD0473 (formerly known as JM473 and AMD473) and currently being tested in phase I clinical trials, has been studied in two human ovarian carcinoma cell lines (CH1 and A2780) where previously, acquired cisplatin resistance has been described. Common mechanisms of resistance were observed in A2780 acquired cisplatin and ZD0473R (resistant) lines (including reduced drug transport and DNA platination, increased glutathione (GSH) and loss of the MLH1 DNA mismatch repair gene). However, contrasting mechanisms were observed in the CH1 sublines. While ZD0473 retained activity against the acquired cisplatin resistant sublines, cisplatin did not circumvent acquired ZD0473 resistance. The trans platinum complex JM335 circumvented resistance in CH1cisR and A2780ZD0473R lines, but not in A2780cisR or CH1ZD0473R cells. Overexpression of metallothionein (MT) in A2780 cells by stable gene transfection resulted in protection from the growth-inhibitory effects of cadmium chloride (3. 8-fold) and a range in protection with platinum drugs (from 7-fold with cisplatin, but only 1.3-fold with ZD0473). Overall, the results show that some mechanisms of resistance to ZD0473 are shared with those previously described in the same parental lines for cisplatin (e.g. in A2780), but in the CH1 lines, differing mechanisms were apparent. Moreover, ZD0473 possesses distinct cellular pharmacological properties in comparison with cisplatin with respect to reduced interactions with MTs, a thiol-containing species associated with tumour resistance to cisplatin.

BCL-2 family protein expression and platinum drug resistance in ovarian carcinoma

The expression of the BCL-2 family proteins, BCL-2, BAX, BCL(XL) and BAK have been determined in a panel of 12 human ovarian carcinoma cell lines encompassing a wide range in sensitivity to cisplatin. Whereas BAX, BCL(XL) and BAK levels did not correlate with sensitivity, there was a statistically significant inverse correlation (r = -0.81; P = 0.002) between growth inhibition by cisplatin and BCL-2 levels. In sublines possessing acquired resistance to various platinum-based drugs or across a panel of human ovarian carcinoma xenografts, there was no consistent pattern of BCL-2 expression. Two relatively sensitive lines (A2780 and CH1) have been stably transfected with bcl-2 and bcl(XL) respectively and two relatively resistant lines (A2780cisR and SKOV-3) stably transfected with bax. Overexpression of BCL-2 in A2780 cells led to resistance to cisplatin compared to the vector control when assayed at 48 h post-drug incubation but a significant increase in sensitivity at 96 h. Relative rates of apoptosis at 48- and 96-h post-cisplatin exposure mirrored the growth inhibition. There was no significant difference in sensitivity of the pair of lines by clonogenic assay. No significant changes in chemosensitivity to a variety of DNA-damaging or tubulin-interactive agents were observed in the remaining transfected lines. Taken together, these results suggest that, in human ovarian carcinoma cells, high BCL-2 levels (either naturally occurring or through gene transfection) confers a trend towards sensitivity not resistance to platinum drugs.

DT-Diaphorase expression and tumor cell sensitivity to 17-allylamino, 17-demethoxygeldanamycin, an inhibitor of heat shock protein 90

BACKGROUND

To our knowledge, 17-allylamino,17-demethoxygeldanamycin (17AAG) is the first inhibitor of heat shock protein 90 (Hsp90) to enter a phase I clinical trial in cancer. Inhibition of Hsp90, a chaperone protein (a protein that helps other proteins avoid misfolding pathways that produce inactive or aggregated states), leads to depletion of important oncogenic proteins, including Raf-1 and mutant p53 (also known as TP53). Given its ansamycin benzoquinone structure, we questioned whether the antitumor activity of 17AAG was affected by expression of the NQO1 gene, which encodes the quinone-metabolizing enzyme DT-diaphorase.

METHODS

The antitumor activity of 17AAG and other Hsp90 inhibitors was determined by use of a sulforhodamine B-based cell growth inhibition assay in culture and by the arrest of xenograft tumor growth in nude mice. DT-diaphorase activity was determined by use of a spectrophotometric assay, and protein expression was determined by means of western immunoblotting.

RESULTS

In two independent in vitro human tumor cell panels, we observed a positive relationship between DT-diaphorase expression level and growth inhibition by 17AAG. Stable, high-level expression of the active NQO1 gene transfected into the DT-diaphorase-deficient (by NQO1 mutation) BE human colon carcinoma cell line resulted in a 32-fold increase in 17AAG growth-inhibition activity. Increased sensitivity to 17AAG in the transfected cell line was also confirmed in xenografts. The extent of depletion of Raf-1 and mutant p53 protein confirmed that the Hsp90 inhibition mechanism was maintained in cells with high and low levels of DT-diaphorase. 17AAG was shown to be a substrate for purified human DT-diaphorase.

CONCLUSION

These results suggest that the antitumor activity and possibly the toxicologic properties of 17AAG in humans may be influenced by the expression of DT-diaphorase. Careful monitoring for NQO1 polymorphism and the level of tumor DT-diaphorase activity is therefore recommended in clinical trials with 17AAG.

Mini-review: discovery and development of platinum complexes designed to circumvent cisplatin resistance

The discovery and development of new platinum-containing anticancer drugs have represented an integral part of anticancer drug development at the Institute of Cancer Research, Sutton, over almost 20 years. As part of a collaboration with chemists at Johnson Matthey, later AnorMED, four major new classes of platinum drug have been discovered, three of which have entered clinical trial. Earlier studies led to the clinical development of the less toxic analogue carboplatin and JM216, the first orally administerable platinum drug. In recent years, the focus has been on two lead complexes designed to overcome the major mechanisms of tumour resistance to cisplatin: JM335 (trans-ammine (cyclohexylaminedichlorodihydroxo) platinum(IV)), an active trans platinum complex; and ZD0473 (cis-amminedichloro(2-methylpyridine) platinum(II)), a sterically hindered complex shown to be less reactive towards thiol-containing molecules than cisplatin. JM335 shows some circumvention of acquired cisplatin resistance in vitro and exhibits unique cellular pharmacological properties in comparison to cisplatin or its cis-isomer in terms gene-specific repair of adducts on DNA and the rate of induction of apoptosis. ZD0473 is now in phase I clinical trial. Myelosuppression is the dose-limiting toxicity at a dose of 130 mg/m2 given i.v. every 3 weeks and there has been evidence of antitumour activity. ZD0473-resistant human ovarian carcinoma cell lines have been established in vitro. Some mechanisms of resistance common to those described for cisplatin (decreased drug uptake, increased glutathione) have been observed plus, in one cell line, increased BCL2 levels and loss of the DNA mismatch repair protein MLH1.

A molecular cytogenetic approach to studying platinum resistance

The technique of comparative genomic hybridisation (CGH) has until recently been used to screen for common genomic abnormalities in fresh tumour material; it has identified previously unrecognised regions of amplification associated with poor prognosis subtypes of breast cancer and lymphoma. Our group has applied this technique to resistant cell lines and their sensitive counterparts in order to define chromosomal abnormalities associated with acquired drug resistance. We have demonstrated the applicability of this technique to the study of drug resistance using cell lines with known mechanisms of resistance. The ability to detect novel genomic alterations in cell lines with novel mechanisms of resistance was also demonstrated. We subsequently examined the CGH profiles of seven different cell lines made resistant to three platinum analogues and showed the most consistent abnormalities to involve over-representation of regions 4q and 6q. More recently, we have applied the CGH technique to a series of testicular germ cell tumours (TGCTs) collected as formalin-fixed paraffin-embedded biopsy specimens from patients, both pre- and post-therapy using a platinum-based regimen (POMB/ACE). Previous reports have shown over-representation of X, 7q, 8q and 12p and loss of 13q to occur in 25% of primary TGCTs. Over-representation of 12p was confirmed in the majority of these biopsy samples; deletion of 13q was noted in the initial biopsies of several patients. We also demonstrated alterations of 4p, 4q, 5q and 6q in this series of patients. Newly acquired deletions of 2q and 18q and amplifications of 8q were frequently observed in post-chemotherapy samples from resistant tumours. The CGH studies on these patients with TGCT will not only enable us to correlate our observations on clinical material with those from long-term cell lines, but should also identify sites of key genes involved in clinical platinum resistance.

Lack of a role for MRP1 in platinum drug resistance in human ovarian cancer cell lines

The level of expression of the multidrug resistance-associated protein (MRP1) in a panel of human ovarian carcinoma cell lines and their variants with acquired cisplatin resistance was determined using Western blotting. No overexpression of MRP1 was detected in any of the cell lines. In addition, we have transfected the MRP1 gene into an intrinsically cisplatin-resistant cell line SKOV3, previously shown to have elevated levels of glutathione (GSH). The MRP1-transfected line SKOV3-S2 was shown to be cross-resistant to doxorubicin, vincristine and etoposide but not to paclitaxel, vinblastine and platinum agents, such as cisplatin, JM216 [bis-acetato-ammine-dichloro-cyclohexylamine platinum (IV)] and AMD473 [cis-ammine dichloro (2-methyl-pyridine) platinum (II)]. No cross-resistance to any of the platinum agents was observed in a MRP1-overexpressing human lung cancer cell line with acquired doxorubicin resistance. Reduction of GSH levels (80-90%) by buthionine sulphoximine (BSO) produced significant potentiation in cisplatin sensitivity in the parental SKOV3, the vector-alone control SKOV3-puro and the MRP1-transfected line SKOV3-S2. The degree of sensitization was similar in all cell lines (1.6-fold). However, selective sensitization by BSO to vincristine was observed in the MRP1-transfected line (4.1-fold) but not in the vector control. No significant differences were observed in cisplatin accumulation in the SKOV3-puro and the SKOV3-S2 cells, although both these transfected lines accumulated significantly more than the parental line. Our results suggest that MRP1 does not play a significant role in platinum resistance in the human tumour cell lines investigated in this study.

In vitro circumvention of cisplatin resistance by the novel sterically hindered platinum complex AMD473

A novel sterically hindered platinum complex, AMD473 [cis-aminedichloro(2-methylpyridine) platinum (II)], has been selected for phase I clinical trials due to commence in 1997. AMD473 was rationally designed to react preferentially with nucleic acids over sulphur ligands such as glutathione. This report documents the in vitro circumvention of acquired cisplatin resistance mechanisms in human ovarian carcinoma (HOC) cell lines by AMD473. In a panel of 11 HOC cell lines, AMD473 showed intermediate growth inhibition potency (mean IC50 of 8.1 microM) in comparison to cisplatin (mean IC50 of 2.6 microM) and carboplatin (mean IC50 of 20.3 microM). AMD473 showed only a 30.7-fold increase in IC50 value from the most sensitive to the most resistant HOC cell line, whereas for cisplatin it was 117.9-fold and for carboplatin 119.7-fold. AMD473 also showed significantly (P < 0.05) reduced cross-resistance to cisplatin in a panel of three cell lines with known acquired platinum drug resistance mechanisms (mean RF for AMD473 was 1.9, for cisplatin 9.1). Cellular accumulation of AMD473 was not reduced in two HOC cell lines (A2780cisR and 41McisR), in which reduced cisplatin accumulation is a major mechanism of acquired cisplatin resistance. AMD473 naked-DNA binding was significantly less affected (P < 0.05) than that of cisplatin by the presence of 5 mM glutathione. Also, AMD473 almost completely circumvented acquired cisplatin resistance in a cell line (A2780cisR) with fivefold elevated intracellular glutathione levels compared with the parent A2780 cell line when measured by clonogenic assay (RF 4.5 for AMD473 vs RF 18 for cisplatin). AMD473 also showed a lower increase in IC50 than cisplatin in an A2780 cell line model with artificially elevated glutathione levels. AMD473 DNA binding was slower than that of cisplatin on both naked and cellular DNA. AMD473 also formed DNA interstrand cross-links (ICLs) at a slower rate than cisplatin (peak ICL formation was at 5 h for cisplatin vs > or = 14 h for AMD473) after equitoxic doses were exposed to HOC cells for 2 h. AMD473 ICLs in the CH1 HOC cell line were slowly formed and showed no visible signs of being repaired 24 h after removal of drug. This was paralleled by a slower, longer lasting induction of p53 protein by equitoxic doses of AMD473 in HOC cell lines with wild-type p53. This new class of sterically hindered platinum compound, selected for clinical trial in 1997, may therefore elicit improved clinical response in intrinsically and acquired cisplatin-resistant tumours in the clinic.

Transport of cisplatin and bis-acetato-ammine-dichlorocyclohexylamine Platinum(IV) (JM216) in human ovarian carcinoma cell lines: identification of a plasma membrane protein associated with cisplatin resistance

The mechanisms by which cis-diamminedichloroplatinum(II) (cisplatin) is transported across the plasma membrane (i.e., passive diffusion versus active transport) were investigated in the 41M and CH1 human ovarian carcinoma cell lines and their acquired cisplatin-resistant variants 41McisR6 and CH1cisR6, respectively. Intracellular cisplatin accumulation was significantly reduced (4.0 +/- 1.7-fold) in the parental 41M line at 4 degrees C when compared to incubations at 37 degrees C. However, no significant differences in platinum uptake were observed in the 41McisR6 and in the CH1 pair of lines at 4 degrees C versus 37 degrees C. Similarly, in the presence of ouabain (an inhibitor of Na+,K+-ATPase), there was a marked reduction (2.0 +/- 0.4-fold) in drug accumulation in the sensitive 41M cells only, and no changes in drug uptake were observed in the other cell lines in the absence or presence of ouabain. Platinum accumulation was significantly enhanced in all cell lines in the presence of metabolic inhibitors (NaF and NaN3). These results suggest that in the parental 41M cell line, cisplatin transport may occur via passive diffusion and active/facilitated transport, whereas in the resistant 41McisR6 variant, cisplatin enters cells by passive diffusion only. The orally active drug bis-acetato-ammine-dichloro-cyclohexylamine platinum(IV) (JM216) is a lipophilic platinum(IV) complex that has been shown to circumvent cisplatin resistance in the 41McisR6 by increasing drug uptake. Across the entire range of concentrations used (5-50 microm), intracellular accumulation of JM216 was significantly reduced in 41M and 41McisR6 cells (3.5 +/- 0.7-fold; P < 0.01), and in CH1 and CH1cisR6 cells (14.2 +/- 6.0-fold; p < 0.01) at 4 degrees C when compared to incubations at 37 degrees C. No significant difference in JM216 uptake was observed in the 41M pair of lines in the absence or presence of ouabain. Additional studies have revealed that the fold reduction observed in cis-ammine(cyclohexylamine)dichloroplatinum(II) (JM118) accumulation in the 41M and 41McisR6 cells at 4 degrees C (3.7 +/- 1.9) reflects similar fold reductions to those observed with JM216 uptake at 4 degrees C. These results suggest that the mechanism of JM216 transport across cell membranes is through passive diffusion, predominantly as a result of its enhanced lipophilicity. Notably, an overexpression of a Mr 36,000 plasma membrane protein was observed in the 41McisR variants when compared to the sensitive 41M line. Increased levels of this Mr 36,000 protein may relate to the observed reduction in active transport of cisplatin in the 41McisR6 variant. Tyrosine phosphorylation of the Mr 36,000 protein appeared to be greater in the resistant 41McisR6 variant than in the parental 41M line. In addition, the constitutive levels of the Mr 36,000 protein in the CH1 pair of lines and in two acquired JM216-resistant variants (41M/JM216R and CH1/JM216R), where resistance in these cell lines is not mediated through reduced drug uptake, were similar to those observed in their respective parental lines. These results suggest that the overexpression of this Mr 36,000 protein in the acquired cisplatin-resistant subline 41McisR6 may play a significant role in cisplatin uptake in resistant cells exhibiting reduced drug accumulation as a major mechanism of cisplatin resistance.

Effects of a new antioestrogen, idoxifene, on cisplatin- and doxorubicin-sensitive and -resistant human ovarian carcinoma cell lines

Pyrrolidino-4-iodotamoxifen (idoxifene) is a new non-steroidal antioestrogen currently undergoing phase I clinical evaluation. Using idoxifene and tamoxifen and two additional analogues of tamoxifen (3-hydroxytamoxifen and 4-iodotamoxifen) and the imidazole-based calmodulin inhibitor, calmidazolium, a strong positive correlation (r2 > 0.95) was observed between cytotoxicity and inhibition of calmodulin-dependent cyclic AMP phosphodiesterase (e.g. mean IC50 across four human ovarian carcinoma cell lines of 4.5 microM for idoxifene and 6.3 microM for tamoxifen). Using two parent human ovarian carcinoma cell lines (41M and CH1; both oestrogen receptor negative) in which acquired resistance to doxorubicin or cisplatin has been generated, we have determined the ability of idoxifene to overcome resistance in these lines. At a non-toxic concentration of 2 microM, idoxifene appeared at least as effective as the clinically used multidrug resistance modifiers verapamil and tamoxifen in overcoming doxorubicin resistance in two acquired resistant cell lines shown to overexpress the P-170 efflux glycoprotein. Non-cross-resistance between cisplatin and idoxifene was observed in two acquired resistant cell lines possessing contrasting mechanisms of resistance to cisplatin (41McisR6 reduced drug transport and CH1cisR6 resistance mediated at the level of DNA). In one of four cell lines (CH1), synergism between idoxifene and cisplatin was observed by median effect analysis. However, with the 41M and its 6-fold cisplatin-resistant variant, antagonism was observed. These observations made by median effect analysis appeared to be unrelated to platinum uptake or removal of platinum-induced DNA interstrand cross-links. These in vitro data suggest that idoxifene may be usefully combined with doxorubicin in the clinical setting, but caution should be exercised in combining it with cisplatin in the treatment of certain tumours.

Selective potentiation of platinum drug cytotoxicity in cisplatin-sensitive and -resistant human ovarian carcinoma cell lines by amphotericin B

Resistance to the clinically used platinum-based drugs cisplatin and carboplatin represents a major limitation to their clinical effectiveness. Using cisplatin-sensitive and -resistant human ovarian carcinoma cell lines previously characterized in terms of their major underlying mechanisms of resistance, we attempted to potentiate the cytotoxic effects of cisplatin and carboplatin using the clinically used antifungal agent amphotericin B (AmB). Using non-toxic concentrations of AmB (up to 15 micrograms/ml) and continuous exposure to cisplatin, a concentration-dependent selective potentiation (maximum of 3.2-fold) of cisplatin cytotoxicity was observed in two cisplatin-resistant cell lines (41McisR6, acquired resistant, and HX/62, intrinsically resistant). In both these cisplatin-resistant cell lines, previous studies have shown resistance to be due primarily to reduced platinum uptake. Notably, no significant potentiation was observed in the parent 41M cell line, in the intrinsically resistant SKOV-3 cell line (where reduced drug accumulation plays only a partial role in determining resistance) or in a pair of cell lines (CH1 and its acquired-resistant variant CH1cisR6) were reduced drug uptake does not play any role in determining resistance. The potentiating effect of AmB was lower with carboplatin and not significant in all cell lines. Platinum uptake following a 2-h exposure of cells to cisplatin was enhanced 3.5-fold in 41McisR6 cells (producing platinum levels similar to those obtained in the parental line) and 1.7-fold in 41M cells by the concomitant exposure to AmB. These data indicate that the potentiation of cisplatin (and carboplatin) cytotoxicity by AmB is not due to a generalized membrane disruption, as effects were observed only in resistant lines where reduced drug transport was apparent. Moreover, AmB did not increase the cytotoxicity of JM216 [bis-acetatoammine(cyclohexylamine)dichloroplatinum (II)], a recently developed, more lipophilic orally active platinum drug, in the 41M/41McisR6 lines. JM216 has previously been shown to circumvent acquired cisplatin resistance due to decreased drug uptake. In vivo, however, using the HX/62 xenograft. AmB (at its maximum tolerated dose of 20 mg/kg; q7d x 4 schedule) did not enhance the antitumour effect of carboplatin (at its maximum tolerated dose of 80 mg/kg; q7d x 4 schedule.