n of eIF2a in the peptide expressing cells. It is likely that another kinase is responsible for the increase phosphorylation of eIF2a Indeed we could show that the upstream kinase, general control nonderepressible 2 is upregulated in the peptide expressing cells. The enhanced phosphorylation of eIF2a phosphorylation and activation of ATF4 in the peptide expressing cells are correlated with an increased expression of the pro-apoptotic factor CHOP indicating that the Bag-1 peptide induces apoptosis. Further evidence that the peptide is involved in the induction of apoptosis is shown in experiment in which we treated 22Rv.1 cells with thapsigargin for 24 h and compared the CHOP expression in 11904527 control cells with cells expressing the Bag-1 peptide. Twenty-four hours treatment resulted in induced CHOP expression accompanied by an increase in PARP cleavage even in the control cells. However the level of CHOP expression and PARP cleavage was further enhanced in the cells expressing the Bag-1 peptide indicating that the peptide sensitizes the cells to the apoptosis inducing effect of thapsigargin. Following siRNA knockdown of Proapoptotic Action of a GRP78/BiP Peptidic Ligand 7 Proapoptotic Action of a GRP78/BiP Peptidic Ligand micron sections of formalin-fixed, paraffin-embedded tumor tissues were subjected to immunohistochemistry. Nuclei are stained blue-purple with hematoxylin while apoptotic cells detected with the TUNEL assay are stained brown. Representative sections of xenografts obtained from each 22Rv.1 stable clone were acquired with an Axioscop microscope. V: stands for empty vector expressing clone and P: peptide expressing clone. doi:10.1371/journal.pone.0045690.g004 of the Bag-1 peptide led to an increase in the PARP cleavage even in the absence of ER stress. However following the knock-down of GRP78/BiP, PARP cleavage was enhanced in the control 22Rv.1 cells but no difference in the cleavage pattern was further observed in the cells expressing the Bag-1 peptide. Bag-1 Specific Sequence Inhibits Prostate Tumor Cell Growth To better analyze the effect of the Bag-1 peptide on prostate cell growth, we determined the Fenoterol (hydrobromide) proliferation of single 22Rv.1 cell clones expressing either the Bag-1 peptide or an empty expression vector using CellTiter-BlueTM proliferation assay. The peptide clones, even in the absence of ER stress, showed a significantly reduced proliferation compared to the vector controls . 21187674 When these stably transfected cells were injected into athymic nude mice, cells containing the empty expression vector developed tumors with an average allowable volume of 1.2 cm3 in 48 weeks. In contrast, tumors generated by cells expressing the peptide only reached 0.4 cm3 in 9 weeks. This difference in tumor volume was also reflected in the weight of the tumors determined at the end of the experiment. Note that in some cases, the mice were euthanized earlier than 9 weeks because the tumors were necrotic. Similar results on the inhibition of tumor cell growth in athymic mice were obtained when the peptide was stably expressed in the prostate tumor cell line LNCaP. Immunohistological studies were 
carried out on the tumors generated by the 22Rv.1 cell clones using a TUNEL assay. Consistent with the growth profile, tumors derived from the control clones showed very few TUNEL positive cells. In contrast, tumors derived from the peptideexpressing clones showed strong signals indicative of massive apoptosis particularly in the cells expressing hig
