GRP-78 does not interact with bortezomib or affect proteasome activity. (A) Endothelial cells were stimulated with or without 10 ng/mL bortezomib in the presence of mannitol (260nM), a control protein (10nM), or GRP-78 (10nM). In contrast to controls, GRP-78 blocked the action of bortezomib in the WST-1 cell viability assay. This effect was significantly inhibited by the addition of a polyclonal antibody (PAB) directed against GRP-78 (10nM). The control antibody (10nM) did not show any effect. (B) PC-3 cells were transfected either with control siRNA (si control) or siRNA mix down-regulating GRP78 gene expression (si GRP-78). After 2 days, respective supernatants were collected. Moreover, PC-3 cell supernatant was spiked with 10 ng/mL bortezomib (PC-3 SN) and then GRP-78 removed by immunoprecipitation using a polyclonal antibody directed against GRP-78 (IP GRP-78). A rabbit polyclonal antibody isotype not binding GRP-78 served as control (IP-control). Absorption and knockdown were controlled by Western blot analysis. (C) Immunoprecipitation experiments testing the hypothesis of a direct interaction of GRP-78 as being responsible for bortezomib inhibition. Bortezomib was incubated together with GRP-78 in PC-3 tumor cell supernatants. After immunoprecipitation using a GRP-78 binding antibody and Sepharose beads in excess amounts, viability of HUVECs was determined to test for biologically active bortezomib present in the supernatant. (D) All collected supernatants were tested on HUVECs in a WST-1 cell viability assay. In PC-3 supernatants containing no GRP-78 resulting from knockdown or immunoabsorption, viability of HUVECs was significantly decreased in the presence of 10 ng/mL bortezomib. (E) GRP-78 did not show any influence on proteasome activity, either in the presence or absence of bortezomib. The 26S proteasome activity was quantified after cleavage of a synthetic substrate. Bars represent mean ± SEM. *P < .05.