Abrogating GRP78 function as a strategy to increase apoptosis of tumour cells

Abstract

The endoplasmic reticulum (ER) is the major site for protein synthesis, folding and trafficking, as well as lipid synthesis and Ca2+ storage. To maintain ER integrity, homeostatic response mechanisms to stimuli which perturb ER function, known as the unfolded protein response (UPR) have evolved. Cancer cells can be exposed to extremes of environmental conditions as a direct consequence of their mechanisms of origin, and may have increased cellular proliferation, coupled with poor vascularisation leading to deficiencies in glucose, oxygen and metabolite requirements for efficient cell growth and survival. When ER function is disrupted, the UPR activates pro-survival mechanisms, such as the induction of chaperones and other folding machinery. However, the UPR also activates pro-death stimuli, to remove cells where the stress is to severe or persists for too long. Glucose regulated protein 78 (GRP78) is the major stress regulator and central hub of the UPR, as well as a key chaperone of the ER. GRP78 bind to and inhibits the activation of; protein kinase-like ER kinase (PERK), inositol requiring element 1 (IRE1) and activating transcription factor 6 (ATF6), known as the UPR activators. Solid tumours, such as melanoma and glioblastoma, may have increased expression of GRP78 correlating with disease stage and resistance to chemotherapy. Conversely, increased GRP78 expression in neuroblastoma has been associated with improved prognosis. To test the hypotheses that abrogating GRP78 function increases apoptosis of tumour cells and that GRP78 is a biomarker for the outcome of ER-stress in differing cancer types, neural-crest derived cancers were compared by stoichiometric analysis of GRP78 and the UPR activators, and the downstream components of the UPR, activating transcription factor 4 (ATF4) and X-box binding protein 1 (XBP-1). To determine the importance of GRP78 across cancer types, changes in sensitivity to the ER stress inducers fenretinide or bortezomib with respect to cell death (propidium iodide stained flow cytometry) or inhibition of cell viability (MTS assay) were assessed in response to siRNA mediated knock-down, GRP78 over-expression and GRP78 inhibition. IV There were differences in cellular concentrations of GRP78 between cell lines representing different cancer types; however the expression of UPR activators did not correlate with GRP78 levels. Stoichiometric analysis of the ratio between GRP78 and the UPR activators demonstrated a significant difference between melanoma and glioblastoma, to that of neuroblastoma. Mapping the activation of the UPR by either fenretinide or bortezomib showed cancer-specific and stress-inducer-dependent responses. Importantly, melanoma and glioblastoma demonstrated significantly greater ATF4 induction whereas neuroblastoma showed prolonged XBP-1 splicing. Testing the effect of altering GRP78 expression on sensitivity to ER-stress-induced cell death demonstrated that high expression of GRP78 in melanoma and glioblastoma correlated with resistance and with neuroblastoma GRP78 over-expression enhanced sensitivity. Investigating the effect of inhibiting GRP78 activity on fenretinide- and bortezomibinduced cell death demonstrated enhanced sensitivity in melanoma and glioblastoma, but no significant enhancement in neuroblastoma. Thus, down-regulating GRP78 or its function increased the death of melanoma and glioblastoma cells in response to ER stress; however there are differences in UPR signalling between cancer types, which ultimately results in contrasting prognosis as demonstrated in neuroblastoma. Although all three cancer types responded to ER-stress induced death, interpretation of the relationship between GRP78 and the UPR activators is essential for determining dependence on GRP78. The data suggest that melanoma and glioblastoma demonstrate increased sensitivity when GRP78 is inhibited due to a shift in the dynamic equilibrium of the UPR, promoting activation and downstream cell death. However cancer types expressing higher UPR activator concentrations compared to that of GRP78 do not respond positively to the inhibition of GRP78.