Mode of cell death induced by the HSP90 inhibitor 17-AAG (tanespimycin) is dependent on the expression of pro-apoptotic BAX

Marissa V Powers,Melanie Valenti,Paul Workman,Suzanne A Eccles,George Thomas,Susana Miranda,Alison Maloney,Paul A Clarke

doi:10.18632/oncotarget.1419

Abstract

Inhibitors of the molecular chaperone heat shock protein 90 (HSP90) are of considerable current interest as targeted cancer therapeutic agents because of the ability to destabilize multiple oncogenic client proteins. Despite their resulting pleiotropic effects on multiple oncogenic pathways and hallmark traits of cancer, resistance to HSP90 inhibitors is possible and their ability to induce apoptosis is less than might be expected. Using an isogenic model for BAX knockout in HCT116 human colon carcinoma cells, we demonstrate the induction of BAX-dependent apoptosis at pharmacologically relevant concentrations of the HSP90 inhibitor 17-AAG both in vitro and in tumor xenografts in vivo. Removal of BAX expression by homologous recombination reduces apoptosis in vitro and in vivo but allows a lower level of cell death via a predominantly necrotic mechanism. Despite reducing apoptosis, the loss of BAX does not alter the overall sensitivity to 17-AAG in vitro or in vivo. The results indicate that 17-AAG acts predominantly to cause a cytostatic antiproliferative effect rather than cell death and further suggest that BAX status may not alter the overall clinical response to HSP90 inhibitors. Other agents may be required in combination to enhance tumor-selective killing by these promising drugs. In addition, there are implications for the use of apoptotic endpoints in the assessment of the activity of molecularly targeted agents.

Highlights

The molecular chaperone heat shock protein 90 (HSP90) is of particular interest as a therapeutic target in cancer owing to its role in maintaining the correct conformation and stability of a number of key oncogenic client proteins such as receptor and non-receptor tyrosine kinases (e.g. ERBB2, ALK, ABL) and serine/threonine kinases (e.g. CRAF, BRAF, AKT, CDK4), including especially those with oncogenic abnormalities and drugresistant alleles [1, 2].The natural product HSP90 inhibitors radicicol, geldanamycin and its derivative 17-allylamino-17demethoxygeldanamycin (17-AAG) exert their effects by interacting with the N-terminal ATP binding domain and inhibiting the intrinsic ATPase activity of HSP90 which is critical for its chaperone function [3,4,5]
This was true for HCT116 cells in vitro and the corresponding solid tumor xenografts growing in immune-compromised mice
To determine the role of BAX in 17-AAG induced apoptosis we used an isogenic pair of HCT116 human colon adenocarcinoma cell lines, as previously described [11]

Summary

Introduction

The molecular chaperone heat shock protein 90 (HSP90) is of particular interest as a therapeutic target in cancer owing to its role in maintaining the correct conformation and stability of a number of key oncogenic client proteins such as receptor and non-receptor tyrosine kinases (e.g. ERBB2, ALK, ABL) and serine/threonine kinases (e.g. CRAF, BRAF, AKT, CDK4), including especially those with oncogenic abnormalities and drugresistant alleles [1, 2].The natural product HSP90 inhibitors radicicol, geldanamycin and its derivative 17-allylamino-17demethoxygeldanamycin (17-AAG) exert their effects by interacting with the N-terminal ATP binding domain and inhibiting the intrinsic ATPase activity of HSP90 which is critical for its chaperone function [3,4,5]. Based on that observation we hypothesized that BAX is required for the induction of apoptosis in response to 17-AAG To test this we used an isogenic pair of HCT116 colon cancer cells, in one member of which BAX was knocked out using homologous recombination [11]. We found that following 17-AAG treatment BAX is required for the induction of cell death via the intrinsic apoptotic pathway This was true for HCT116 cells in vitro and the corresponding solid tumor xenografts growing in immune-compromised mice. The results indicate that cytostatic effects on cell proliferation may represent the dominant therapeutic response in this cancer model.

Methods

Results

Conclusion