Abstract
Exposure of myeloma cells to Hsp90 inhibitors (eg 17-AAG) or proteasome inhibitors (eg Bortezomib) results in a time-dependent induction of heat shock protein 72 (Hsp72), protecting cells from drug-induced apoptosis. Hsp72 is a member of the heat shock protein 70 (Hsp70) family and enables cells to cope with harmful aggregations of proteins during and after stress, protecting cells from apoptosis in addition to acting as a chaperone for proteins such as immunoglobulin. We show that inhibition of Hsp72 upregulation induced by these anti-myeloma drugs, by either inhibitors of Hsp70 induction or siRNA, increases susceptibility of myeloma cells to drug-induced apoptosis. This provides a rationale for using such inhibitors in combination with anti-myeloma drugs. A number of commercially available inhibitors of Hsp70 induction, including KNK437 and triptolide, were used to inhibit stress-induced upregulation of Hsp72. Combining Hsp90 inhibitors (Hsp90i) with these compounds increased cell death to 73% from 33% when Hsp90i were used alone. Cell cycle analysis indicated that combining these compounds with Hsp90 inhibitors overcame the G2 growth arrest induced by Hsp90 inhibitors and increased the proportion of cells in sub G1, indicative of apoptosis. Analysis of their anti-proliferative effects demonstrated that, in H929 cells, Hsp90 inhibitor/triptolide combinations had additive effects whilst in U266 cells their effects were synergistic. To determine the underlying mechanisms responsible for the effects of the combination, key proteins involved in endoplasmic reticulum (ER) stress, Unfolded Protein Response and caspase-dependent and independent signalling pathways were analysed. We show that ER stress signalling through IRE1 and PERK pathways, measured by cleavage of XBP1 and induction of CHOP, were initiated earlier and sustained longer, when cells were treated with the combination. Similarly, analysis of caspase-dependent and independent signalling demonstrated that the combination induced early cleavage of caspases 9, 8 and 3 which was sustained. Nuclear translocation of AIF, the caspase-independent cell death signal, was strongly induced suggestive of strong activation of this pathway. We demonstrate that death induced by triptolide or its combination with 17-AAG is p53-independent using of a panel of myeloma cell lines. To confirm our findings, we used lentiviral siRNA constructs targeting the inducible forms of Hsp70 and then analysed response to Hsp90 inhibitors. These experiments confirm the results generated using the chemical inhibitors of Hsp70 expression. In conclusion, inhibiting drug-induced Hsp72 expression appears to overcome cellular resistance to Hsp90 inhibitors, supporting the clinical development of Hsp70 inhibitors and their combination with Hsp90 and proteasome inhibitors.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author