Activation and Cross-talk between Akt, NF-κB, and Unfolded Protein Response Signaling in 1-LN Prostate Cancer Cells Consequent to Ligation of Cell Surface-associated GRP78*

Binding of activated forms of the proteinase inhibitor α2-macroglobulin (α2M*) to cell surface-associated GRP78 on 1-LN human prostate cancer cells causes their proliferation. We have now examined the interplay between Akt activation, regulation of apoptosis, the unfolded protein response, and activation of NF-κBin α2M*-induced proliferation of 1-LN cells. Exposure of cells to α2M* (50 pm) induced phosphatidylinositol 3-kinase-dependent activation of Akt by phosphorylation at Thr-308 and Ser-473 with a concomitant 60-80% increase in Akt-associated kinase activity. ERK1/2 and p38 MAPK were also activated, but there was only a marginal effect on JNK activation. Treatment of 1-LN cells with α2M* down-regulated apoptosis and promoted NF-κB activation as shown by increases of Bcl-2, p-BadSer-136, p-FOXO1Ser-253, p-GSK3βSer-9, XIAP, NF-κB, cyclin D1, GADD45β, p-ASK1Ser-83, and TRAF2 in a time of incubation-dependent manner. α2M* treatment of 1-LN cells, however, showed no increase in the activation of caspase -3, -9, or -12. Under these conditions, we observed increased unfolded protein response signaling as evidenced by elevated levels of GRP78, IRE1α, XBP-1, ATF4, ATF6, p-PERK, p-eIF2α, and GADD34 and reduced levels of GADD153. Silencing of GRP78 gene expression by RNAi suppressed activation of AktThr-308, AktSer-473, and IκB kinase α kinase. The effects of α2M* on the NF-κB activation, antiapoptotic signaling, unfolded protein response signaling, and proapoptotic signaling were also reversed by this treatment. In conclusion, α2M* promotes cellular proliferation of 1-LN prostate cancer cells by activating MAPK and Akt-dependent signaling, down-regulating apoptotic signaling, and activating unfolded protein response signaling.

(apoptosis signal-regulating kinase 1) is a Ser/Thr kinase that activates both the p38 MAPK and JNK pathways by directly phosphorylating MKK3/MKK6 and MKK4/MKK7, respectively (28, 39 -42). ASK1 is activated by a variety of stimuli, including TNF␣, ER stress, and calcium signaling consequent to ligation of G-protein-coupled receptors (39 -42). The antiapoptotic activity of Bcl-2 is reduced when it is phosphorylated by activated ASK1 (28,41). On the other hand, overexpression of ASK 1 may induce not only apoptosis but cell differentiation and survival, depending on the cell type and cellular context (28,41,42). TNF␣ is a potent activator of ASK1, and this activation is regulated by TRAF2 (TNF receptor-associated factor 2) (28, 40 -42). TRAF2 is an adaptor protein that couples TNF␣ receptor ligation to activation of the ASK1-JNK/p38 MAPK cascades (28, 40 -42).
NF-B regulates the transcription of many genes involved in stress remediation, cell growth, and apoptosis (52)(53)(54)(55). In the inert state, NF-B is present in the cytoplasm in association with the inhibitory protein, IB. Activation of NF-B occurs in response to a variety of stressful conditions, where impaired protein folding in the ER is characteristic (52)(53)(54)(55). These stimuli activate the IB kinase (IKK) ␤ subunit, which phosphorylates NF-B-bound IB and targets it for ubiquitindependent degradation, allowing liberated NF-B dimers to translocate to the nucleus (53). In an alternative pathway, the IKK ␣ subunit of the IKK complex processes NF-B2 precursor protein, which preferentially binds RelB in the cytoplasm, resulting in the release of RelB-p52 dimers (55). Persistent NF-B activation is detected in many solid tumors, and its inhibition increases their sensitivity to chemotherapeutic agents and radiation (54). Overexpression of NF-B suppresses apoptosis, whereas IB-␣ mutants, which can no longer be phosphorylated, are resistant to ubiquitin-mediated degradation and suppress NF-B activation, thus sensitizing cells to apoptotic insults (16,(52)(53)(54)(55). IRE1 is also involved in the activation of NF-B induced by ER stress (52)(53)(54)(55). Activation of NF-B by ER stress-inducing agents is inhibited by dominant negative IRE1 (50) or TRAF2 mutants (56). A recent report demonstrates that UV light-induced activation of NF-B is due to the PERK-induced phosphorylation of eIF2␣, which causes translational inhibition of new IB synthesis (57).
In the present report, we have examined the role of Akt activation, regulation of apoptosis, and the unfolded protein response in ␣ 2 M*induced growth of 1-LN prostate cancer cells. We report an increase in Akt Thr-308 and Akt Ser-473 kinase activities, up-regulation of UPR components, including IREI␣, XBP-1, ATF6, GRP78, p-PERK, p-eIF2␣, GADD34 (growth arrest and DNA damage-inducible 34), and ATF4, as well as upregulation of components of the anti-apoptotic machinery Bcl-2, p-FOXO1, NF-B, GADD45␤, and XIAP in 1-LN prostate cancer cells treated with ␣ 2 M*. We have also studied the role of GRP78 in ␣ 2 M*induced activation of cell survival and antiapoptotic signaling by silencing GRP78 gene expression employing RNAi.
Western Blotting of Activated Akt in ␣ 2 M*-stimulated 1-LN Prostate Cancer Cells-The highly metastatic prostate carcinoma cell line 1-LN, derived from less metastatic PC-3 cells, was a kind gift from Dr. Philip Walther (Duke University Medical Center, Durham, NC). 1-LN cells in 6-well plates (500 ϫ 10 3 cells/well) were allowed to grow to confluence in RPMI 1640 medium containing 10% FBS, penicillin (12.5 units/ml), streptomycin (6.5 g/ml), 2 mM glutamine, and 10 nM insulin at 37°C in a humidified CO 2 (5%) incubator. At about 90% confluence, the medium was aspirated, the monolayers were washed with ice-cold Hepes-buffered Hanks' basic salt solution, pH 7.4, and a fresh volume of medium was added to the monolayers. The cells were exposed to ␣ 2 M* (50 pM) for varying periods of time. The reaction was stopped by aspirating the medium, and a volume of lysis buffer containing 50 mM Tris⅐HCl (pH 7.5), 120 mM NaCl, 1% (v/v) Nonidet P-40, 25 mM sodium fluoride, 1 mM sodium pyrophosphate, 0.1 mM sodium orthovanadate, 1 mM PMSF, 1 mM benzamidine, and leupeptin (20 g/ml) was added. The cells were lysed for 10 min over ice, scraped into tubes, and centrifuged at 8000 ϫ g for 10 min at 4°C, and their protein contents were determined (58). In each case an equal amount of protein was used for electrophoresis. The immunoblottting of membranes with antibodies specific for phosphorylation of Akt in Thr-308 or Ser-473 was performed according to the manufacturer's instructions. The detection and quantification of immunoblots were performed by ECF and phosphorimaging. The membranes were reprobed for the protein loading control actin.
Measurement of Phosphorylation of Akt at Thr-308 and Ser-473 and Their Kinase Activities in Cells Stimulated with ␣ 2 M*-The p-Akt Thr-308 and p-Akt Ser-473 kinase activities in Akt immunoprecipitates of 1-LN prostate cancer cells stimulated with ␣ 2 M* (50 pM/15 min) were assayed essentially according to Hill and Jennings (59). Briefly, 1-LN cells (in two 6-well plates, 4 ϫ 10 6 cells/well) were incubated as above until confluence. The monolayers were washed in Hepes-buffered Hanks' basic salt solution, pH 7.4, twice, and a fresh volume of incubation medium was added to each well. The monolayers were stimulated with buffer or ␣ 2 M* (50 pM/15 min) in triplicate. The reactions were stopped by aspirating the medium, and a volume of lysis buffer containing 50 mM Tris⅐HCl (pH 7.5), 120 mM NaCl, 1% (v/v) Nonidet P-40, 25 mM sodium fluoride, 1 mM sodium pyrophosphate, 0.1 mM sodium orthovanadate, 1 mM PMSF, 1 mM benzamidine, and leupeptin (20 g/ml) was then added to each incubation. The cells were lysed for 10 min over ice, scraped into tubes, and centrifuged at 8000 ϫ g for 10 min at 4°C, and their protein contents were determined (58). Equal amounts of lysate proteins were immunoprecipitated with Akt antibodies (1:50) at 4°C overnight with gentle rotations, according to the manufacturer's instruction. Akt immunoprecipitates were washed sequentially with lysis buffer supplemented with 0.5 M NaCl, lysis buffer, and 50 mM Tris⅐HCl (pH 7.4) supplemented with 1 mM dithiothreitol, 1 mM PMSF, and 1 mM benzamidine with centrifugation at 8000 ϫ g for 5 min at 4°C employed between each wash. To each immunoprecipitate, 40 l of cold kinase buffer containing 50 mM Tris⅐HCl (pH 7.5), 10 mM MgCl 2 , 1 mM dithiothreitol, 1 mM PMSF, 1 mM benzamidine, and 20 g/ml leupeptin was added, followed by the addition of 30 M Akt Thr-308 kinase substrate peptide (NH 2 -RRPHFPQFSYSA-COOH) in respective tubes. Zak3tide (NH 2 -GGEEEEYFELVKKKK-COOH) was used as the control peptide. The reaction was initiated by adding 50 M ATP and 2 Ci of [␥-33 P]ATP in each tube, and the tubes were incubated for 30 min at 30°C with shaking. The reaction was stopped by adding 5 l of 0.5 M EDTA to each tube, the tubes were centrifuged at 3000 rpm for 3 min, and 40 l of each supernatant was applied on p81 phosphocellulose paper (Whatman). Supernatants were allowed to dry, and the papers were washed four times each by immersing them in a liter of 1 N phosphoric acid for 3 min. The papers were rinsed with acetone, and their radioactivity was counted in a liquid scintillation counter (22).   , and TRAF2 in ␣ 2 M*stimulated 1-LN Prostate Cancer Cells-1-LN cells in 6-well plates (4 ϫ 10 6 cells/well) were allowed to grow to confluence in RPMI 1640 medium containing 10% FBS, penicillin (12.5 units/ml), streptomycin (6.5 g/ml), 2 mM glutamine, and 10 nM insulin at 37°C in a humidified CO 2 (5%) incubator as described above. At about 90% confluence, the medium was aspirated, the monolayers were washed with ice-cold Hepes-buffered Hanks' basic salt solution, pH 7.4, and a fresh volume of the medium was added to the monolayers. The cells were exposed to ␣ 2 M* (50 pM) for varying periods of time. The reaction was stopped by aspirating the medium, and a volume of lysis buffer containing 50 mM Tris⅐HCl (pH 7.5), 120 mM NaCl, 1% (v/v) Nonidet P-40, 25 mM sodium fluoride, 1 mM sodium pyrophosphate, 0.1 mM sodium orthovanadate, 1 mM PMSF, 1 mM benzamidine, and leupeptin (20 g/ml) was added. The cells were lysed for 10 min over ice, scraped into tubes, and centrifuged at 8000 ϫ g for 10 min at 4°C, and their protein contents were determined. In each case, an equal amount of protein was used for electrophoresis. The immunoblotting of the respective membranes was done with antibodies specific for p-ERK1/2, p-MKK3/6, p-MKK4, p-MKK7, p-p38 MAPK, p-JNK, p-ASK1, and TRAF2, respectively, according to the manufacturer's instructions. The respective membranes were reprobed for unphosphorylated ERK1/2, MKK3/6, MKK4, MKK7, p38 MAPK, JNK, and ASK1 as well as the protein loading control actin according to the manufacturer's instruction. The detection and quantification of immunoblots were performed by ECF and phosphorimaging. The membranes were reprobed for protein loading control actin.
Measurement of the Effects of Treatment of 1-LN Prostate Cancer Cells with ␣ 2 M* on Activation of Caspases by Western Blotting-The experimental details of incubating 1-LN prostate cancer cells with ␣ 2 M*, their lysis, and electrophoresis were identical to that described above. The respective membranes were immunoblotted with antibodies specific for procaspase-12, procaspase-9, and cleaved caspase-3, respectively. Since the antibody is specific for mouse procaspase-12, we performed a study to demonstrate that this antibody cross-reacts with the human protein. For these studies, murine macrophages were compared with human 1-LN and PC-3 prostate cancer cells to demonstrate that the anti-murine antibody recognizes human procaspase-12. This study is shown as a control in Fig. 3. The membranes were reprobed for actin, a protein loading control. The detection of immunoblots was performed by ECF and phosphorimaging.
Measurement of NF-B1 (p65) and NF-B2 (p52) in Nuclei of 1-LN Cells Stimulated with ␣ 2 M*-1-LN cells (6 ϫ 10 6 cells) were incubated and stimulated with ␣ 2 M* (50 pM/15 min) as described above. Nuclei from 1-LN cells were isolated, and their purity was determined as described previously (18,19). Nuclei were lysed in lysis buffer as described above, and their protein contents were determined. Equal amounts of protein were immunoprecipitated with antibodies against NF-B1 and NF-B2 as described above. The NF-B1 and NF-B2 immunoprecipitates were processed for electrophoresis and Western blotting as described above, and the p65 and p52 protein bands were visualized by ECF and phosphorimaging.
Assay of IKK Activation in 1-LN Cells Stimulated with ␣ 2 M*-The IKK assay in 1-LN cells stimulated with ␣ 2 M (50 pM/15 min) was performed essentially according to Takada et al. (52). 1-LN cells (4 ϫ 10 6 /well in 6-well plates) were stimulated with ␣ 2 M* (50 pM/15 min) and incubated as above. The reaction was terminated by aspirating the medium. The cells in the respective wells were lysed in lysis buffer as described above, and lysate protein contents were determined. Equal amounts of lysate protein from buffer and ␣ 2 M*-stimulated cells were immunoprecipitated with 1 g of IKK␣ antibodies as described above. The protein A-Sepharose beads were washed three times with lysis buffer and then three times with kinase assay buffer containing 50 mM Hepes (pH 7.5), 20 mM MgCl 2 , and 2 mM dithiothreitol, by centrifugation at 2500 rpm for 5 min at 4°C. To the IKK␣ immunoprecipitate was added 20 l of kinase buffer containing 10 M ATP, 10 Ci of [␥-33 P]ATP, and 2 g of substrate glutathione S-transferase-IB-␣ (amino acid residues 1-54), and the samples were incubated for 30 min at 30°C with shaking. The reaction was terminated by adding 10 l of 4ϫ SDS sample buffer, and the samples were boiled for 5 min. The samples were electrophoresed on 10% gels, protein bands were transferred to membranes, the membranes were autoradiographed, and radioactive bands were visualized and quantified by phosphorimaging.

Measurement of the Effects of ␣ 2 M* Treatment of 1-LN Prostate Cancer Cells on Up-regulation of the Components of the UPR Signaling Pathway by Western
Blotting-The experimental details of incubating 1-LN prostate cancer cells with ␣ 2 M*, their lysis, and electrophoresis were identical to those described above. The respective membranes were immunoblotted with antibodies specific for GRP78, IRE1␣, XBP-1, ATF6, p-PERK Thr-980 , p-eIF2␣ Ser-51 , ATF4, GADD34, and GADD153, respectively (37). The membranes were reprobed for eIF2␣ and actin as the protein loading control. The detection of immunoblots was performed by ECF and phosphorimaging.
Chemical Synthesis of dsRNA Homologous in Sequence to the Target GRP78 Gene-The chemical synthesis of dsRNA homologous in sequence to the target GRP78 peptide sequence 370 KIQQLVK 376 , mRNA sequence 5Ј-AAA ATA CAG CAA TTA GTA AAG-3Ј (Swiss-Prot GRP primary sequence accession number P11021), was performed by Ambion (Austin, TX). For making dsRNA of the sense 5Ј-AAU ACA GCA AUU AGU AAA GTT-3Ј and antisense 5Ј-CUU UAC UAA UUG CUG UAU UTT-3Ј, oligonucleotides were annealed according to the manufacturer's instructions. Throughout the studies, handling of reagents was performed in an RNasefree environment. The protocol is described in detail elsewhere (20,22). Briefly, equal amounts of sense and antisense oligonucleotides were mixed in annealing buffer and heated at 90°C for 1 min and then maintained for 1 h at 37°C in an incubator. The dsRNA preparation was stored at Ϫ20°C before use.
Transfection of 1-LN Cells with dsRNA Homologous in Sequence to the GRP78 Gene-Silencing of GRP78 gene expression was performed as previously described (18 -22). We have used the following protocol for silencing the expression of the GRP78 gene in macrophages and 1-LN cells. This protocol results in 60 -70% suppression of GRP78 gene expression as quantified by GRP78 mRNA and protein levels. This suppression of GRP78 gene expression also causes a concomitant loss of cell surface-anchored GRP78 as measured by receptor binding, as well as  MAY 12, 2006 • VOLUME 281 • NUMBER 19 abrogation of ␣ 2 M*-induced 1,4,5-trisphosphate synthesis, elevation of [Ca 2ϩ ] i , DNA synthesis, and mitogenic signaling (19 -22). Confluent 1-LN cell monolayers (1.5 ϫ 10 6 /well in 6-well plates) in quadruplicate, incubated as described above, were washed twice with Hanks' balanced salt solution, and 2 ml of DMEM containing 10% of FBS and antibiotics was added. The cells were incubated as above for 16 h. Just before each transfection, 25 g of GRP78 dsRNA was diluted to 100 l of serumand antiobiotic-free DMEM in a tube. In another tube, 10 l of Lipofectamine was diluted into 100 l of serum-and antiobiotic-free medium. The two solutions were combined, mixed gently, and incubated for 45 min at room temperature, followed by the addition of 800 l of serum-and antibiotic-free medium to each tube. The monolayers were washed twice with serum-antibiotic-free DMEM, layered in each well with 1 ml of Lipofectamine-DMEM or lipid dsRNA mixtures containing 25 g of dsRNA of GRP78 target mRNA gently mixed, and incubated for 5 h at 37°C in a humidified CO 2 incubator in separate experiments. At the end of the incubation, 1 ml of antibiotic-free DMEM containing 10% FBS was added to each well, and the cells were incubated for 16 h. The medium was replaced with DMEM containing antibiotics and 10% FBS 24 h following the start of the transfection. The monolayers were incubated for a further 24 h as above. At the end of the incubation, the medium was aspirated, and the monolayers were washed with the above medium once. A volume of the same medium was added, and the cells were used for the experiment outlined below. To demonstrate that the transfection of 1-LN prostate cancer cells with dsRNA homologous in sequence to the target GRP78 gene does not produce any nonspecific effects on target gene expression, the 1-LN cells were transfected with equimolar concentrations of scrambled small interference RNA (Silencer TM negative control, catalog number 4610; Ambion) under identical conditions as described above for transfection with GRP78 dsRNA. At the end of the transfection period (48 h), the medium was aspirated, and a volume of DMEM was added. The cells were stimulated either with buffer or ␣ 2 M* (50 pM) for 15 min. The reaction was stopped by aspirating the medium and adding a volume of lysis buffer as described above. As stated above, this protocol suppresses the expression of GRP78 gene expression by 60 -70% as measured by GRP78 mRNA and protein levels (19 -22). Equal amounts of lysate protein as determined by Bradford (58) were employed for the Akt kinase assay and IKK␣ kinase assays as well as electrophoresis. We have previously demonstrated that this protocol does not significantly alter cell viability (20,22).
Measurement of the Effects of Transfecting 1-LN Cells with GRP78 dsRNA on Activation of NF-B1, NF-B2, pIKK-␣, and p-IB-␣ after ␣ 2 M* Stimulation-Equal amounts of lysate protein from 1-LN cells treated with Lipofectamine plus buffer, Lipofectamine plus ␣ 2 M* (50 pM/15 min), GRP78 dsRNA plus ␣ 2 M* (50 pM/15 min), or GRP78 dsRNA plus ␣ 2 M* were electrophoresed, and protein bands were transferred to membranes. After immunoblotting with the respective antibodies, the immunoblots were visualized and quantified by ECF and phosphorimaging. The immunoblots were reprobed for actin as a protein loading control.
Measurements of the Effects of Transfection of 1-LN Cells with GRP78 dsRNA on Expression of p-FOXO, p-GSK3␤, XIAP, Cyclin D1, and p27 kip upon ␣ 2 M* Stimulation-Equal amounts of lysate protein from the respective experimental groups were electrophoresed, and protein bands on the gels were transferred to membranes. The immunoblots were probed with the respective antibodies, and the immunoblots were visualized and quantified by ECF and phosphorimaging as described above. The immunoblots were reprobed for actin as a protein loading control.

␣ 2 M* Up-regulates Activation of Akt in 1-LN Cells in a PI 3-Kinasedependent
Manner-Activation of Akt is dependent upon PI 3-kinase activation, and disregulation of Akt activation is implicated in the behavior of malignant tumors (5-7). Upon stimulation, cytosolic Akt is recruited to the plasma membrane through the binding of its NH 2terminal plekstrin homology domain to the lipid product of PI 3-kinase activation (5)(6)(7). Akt is then activated by phosphorylation on two residues, namely Thr-308 in the activation loop and Ser-473 in the hydrophobic motif of the COOH-terminal tail. Phosphorylation of both residuesisrequiredforitsfullactivation (60 -62).3Ј-Phosphatidylinositoldependent kinase 1 is a plekstrin homology domain-containing kinase that phosphorylates Akt at Thr-308 (5, 6); however, the identity of the kinase responsible for phosphorylating Akt at Ser-473 is disputed (60 -62). We have previously demonstrated that expression of GRP78 on the cell surface is essential for ␣ 2 M*-dependent signal transduction in both macrophages and human prostate cancer (18 -22). We have characterized the identity of cell surface synthesis, cellular proliferation, and mitogenic signaling upon binding of ␣ 2 M* to GRP78 in a variety of prostate cancer cells of differing metastatic potential as well as in macrophages (18 -22). These postreceptor events are drastically inhibited by prior treatment of cells with antibodies against GRP78 or silencing GRP78 gene expression by RNA interference (18 -22). Human prostate cancer cells lacking GRP78 on their cell surface do not demonstrate ␣ 2 M*-dependent signal transduction (18,20). This includes PC-3 cells, the parent line for 1-LN cells. Thus, PC-3 cells do not express GRP78 on the cell surface and do not demonstrate ␣ 2 M*-dependent activation of signaling cascades. We therefore focused the present studies on the 1-LN cell line. Treatment of 1-LN prostate cancer cells with ␣ 2 M* (50 pM) elevated both p-Akt Thr-308 and p-Akt Ser-473 at about 5 min, and these levels remained elevated during 60 min of incubation (Fig. 1A). The kinetics of activation of both p-Akt Thr-308 and p-Akt Ser-473 were comparable in ␣ 2 M*-treated 1-LN prostate cancer cells (Fig. 1A). ␣ 2 M*induced phosphorylation of Akt at Thr-308 and Ser-473 was greatly reduced by prior treatment of cells with LY294002, a specific inhibitor ofPI3-kinase (Fig.1B),demonstratingthatactivationofAktisPI3-kinasedependent. We further demonstrated ␣ 2 M*-induced phosphorylation of Akt at Thr-308 and Ser-473 in 1-LN cells by assaying their kinase activities using specific kinase peptide substrates (Fig. 1C). ␣ 2 M* treatment caused a nearly equal activation of both p-Akt Thr-308 kinase and p-Akt Ser-473 kinase which corroborates the Western blotting data (Fig. 1A). The results presented show that ␣ 2 M*, like growth factors, induces mitogenic and cell survival signaling in 1-LN prostate cancer cells.
Suppression of GRP78 Gene Expression Inhibits ␣ 2 M*-induced Activation of Akt Thr-308 and Akt Ser-473 Kinases in 1-LN Cells-In the next series of experiments, we examined the role of ligating 1-LN prostate cancer cell surface-associated GRP78 with ␣ 2 M* on activation of Akt kinase activities and downstream signaling by silencing GRP78 gene expression. We have shown that silencing GRP78 gene expression reduces its mRNA and protein levels by about 50 -60% in macrophages and 1-LN prostate cancer cells (20,22). This level of suppression disproportionately reduces signal transduction events in these cells (20 -22), consistent with our observation that there is a threshold level of GRP78 that must be present on the cell surface before the cells can respond to ␣ 2 M* (63). Silencing of GRP78 gene expression inhibited the activities of Akt Thr-308 and Akt Ser-473 kinases in ␣ 2 M*-treated 1-LN cells to nearly basal levels (Fig. 2). Under identical experimental conditions, transfection of 1-LN cells with scrambled dsRNA showed negligible effects on ␣ 2 M*-induced Akt kinase activities (Fig. 2). These studies demonstrate that binding of ␣ 2 M* to cell surface-associated GRP78 is involved in activating downstream cell survival signaling in 1-LN prostate cancer cells.
␣ 2 M* Up-regulates Activation of ERK1/2 in 1-LN Prostate Cancer Cells-The activation of surface receptors leads to the activation of Ras, a membrane-resident GTPase, which recruits Raf kinase from the cytosol to the cell membrane. Most tumors demonstrate sustained and elevated activation of the Raf-MEK-ERK pathway. This pathway promotes cell survival by activating antiapoptotic mechanisms. In the MEKdependent pathway, activated ERK activates 90-kDa ribosomal S6 kinase, which phosphorylates and inactivates proapoptotic Bad (16, 33, 35, 64 -70). 90-kDa ribosomal S6 kinase also activates the transcription factor CREB, which promotes cell survival by up-regulating Bcl-2 expression (33, 35, 64 -70). In addition, Raf up-regulates the expression of the antiapoptotic transcription factor NF-B. Treatment of murine macrophages with ␣ 2 M* causes activation of Ras⅐GTP and up-regulates activation of MEK1 and ERK (20,71). In 1-LN prostate cancer cells, ␣ 2 M* treatment promotes cell proliferation as evidenced by increased DNA synthesis and increased cell numbers (20). To assess the role of the MEK-ERK pathway in 1-LN cancer cells, we determined the effect of ␣ 2 M* treatment of these cells on activation of ERK1/2. ␣ 2 M*-induced the phosphorylation of ERK1/2 as early as 5 min of incubation, which remained nearly sustained up to 60 min of incubation (Fig. 1A). These data suggest that ␣ 2 M*-induced cell growth is partly mediated by the ERK signaling pathway.
Affect on Caspase-12, Caspase-9, and Caspase-3 Activation by ␣ 2 M* in 1-LN Prostate Cancer Cells-In response to intrinsic and extrinsic signals, caspases are activated through cleavage by upstream caspases, which leads to apoptotic cell death (see Ref. 16 and references therein). Activation of caspase-12 is specific to ER stress. Once activated during  MAY 12, 2006 • VOLUME 281 • NUMBER 19 ER stress, the catalytic subunit of caspase-12 is released into the cytosol, where it activates the caspase-9 cascade (73). To assess the role of proapoptotic caspases in ␣ 2 M*-induced cell proliferation in 1-LN prostate cancer cells, we quantified levels of caspase-12, caspase 9, and cleaved caspase-3 (Fig. 3) by Western blotting. Under our experimental conditions, ␣ 2 M* did not activate the activation of caspase-12, caspase-9, and caspase-3.

Up-regulation of Bcl-2 and p-Bad Ser-136 in 1-LN Prostate Cancer Cells Treated with ␣ 2 M*-In the next series of experiments, we examined the effects of treating 1-LN prostate cancer cells with ␣ 2 M* on levels of
Bcl-2 and Bad phosphorylated at Ser-136 by Western blotting (Fig. 4). Bcl-2 is the prototype for a large family of structurally related proteins that regulate cell death in mammalian cells. Bcl-2 family members bind to apoptotic proteinase-activating factor 1 and inhibit caspase-9, and this binding is antagonized by Bax (43). Bad binds to Bcl-2, neutralizing the antiapoptotic effects of Bcl-2 (43); however, Akt phosphorylates Bad at Ser-136, which causes its sequestration with 14-3-3 protein (5). In ␣ 2 M*-stimulated 1-LN cells, an approximately 2-fold increase in Bcl-2 was noted at about 10 -20 min of incubation, which declined at longer periods of incubation compared with controls (Fig. 4A). This was accompanied by an increase in protein 14-3-3 (Fig. 4A). A similar but sustained increase in p-Bad Ser-136 was observed at about 10 -20 min of incubation of 1-LN prostate cancer cells with ␣ 2 M* (Fig. 4B). That Akt is the mediator of Bad phosphorylation at Ser-136 in 1-LN cells is demonstrated by their treatment with the PI 3-kinase inhibitor LY2904002 (20 M/20 min) before ␣ 2 M* stimulation (Fig. 4C).

␣ 2 M* Up-regulates the Phosphorylation of FOXO1 in 1-LN Cancer
Cells-Akt activation positively regulates G 1 /S cell cycle progression through inactivation of GSK3␤, leading to increased cyclin D1, inhibition of forkhead transcription factors including the FOXO subfamily, and the subsequent reduction of p27 kip (5,74,75). Phosphorylation of FOXO1 by activated Akt promotes its export from the nucleus to the cytosol, thus preventing FOXO1 interaction with DNA, which otherwise would up-regulate transcription factors involved in the apoptotic pathway. FOXO1 also interacts with the 14-3-3 protein, which serves to localize the p-FOXO1 in the cytoplasm, and it also facilitates nuclear export of FOXO1 (5,74,75). Treatment of 1-LN cancer cells with ␣ 2 M* (50 pM/15 min) elevated p-FOXO1 by about 50 -70% at about 10 min of incubation, and these levels remained elevated up to 60 min of incubation as compared with control 1-LN cells (Fig. 4B). Treatment of 1-LN cells with LY294002 (20 M/20 min), before ␣ 2 M* addition markedly reduced p-FOXO1 levels (Fig. 4C). An increased up-regulation of 14-3-3 parallel to that of p-FOXO1 was also observed in 1-LN cells treated with ␣ 2 M* (Fig. 4A), which suggests Akt-mediated inactivation of proapoptotic FOXO1.

Activation of NF-B1 and NF-B2 in 1-LN Prostate Cancer Cells Treated with ␣ 2 M*-NF-B is constitutively activated in many cancers
where ER stress is a common occurrence. Treatment of 1-LN cancer cells with ␣ 2 M* (50 pM) caused a severalfold increase in the NF-B1 as determined by Western blotting (Fig. 5A). The upstream activators of NF-B1 (namely IKK␣/␤ and IB-␣) also exhibited a 1.5-2-fold transitory activation at about 10 -20 min of incubation (Fig. 5A), subsequently returning to the basal state. The Western blotting data with respect to p-IKK␣ are also corroborated by the IKK␣ kinase activity studies, which

GRP78-induced Proliferation of 1-LN Prostate Cancer Cells
demonstrate a nearly parallel increase in ␣ 2 M*-stimulated 1-LN cells, as compared with buffer-treated cells (Fig. 5B). Under these experimental conditions, the levels of phosphorylated IB-␤ were not affected (data not shown). Activation of NF-B is also affected by NIK in the cytoplasm as well as in the modulation of its transactivation potential in the nucleus (55). NIK associates with TRAF2 and promotes the phosphorylation of IKK␣ and IKK␤. Enhanced activity of NIK occurs in various tumors, which suggests its association with enhanced levels of p52 in breast cancer (53,54). We therefore studied the effect of ␣ 2 M* in 1-LN cancer cells with respect to activation of NF-B2 by the noncanonical pathway (Fig. 5C). Exposure of 1-LN cells to ␣ 2 M* caused a transient increase in both TRAF2 (Fig. 5A) and NIK (Fig. 5C) at about 10 -20 min of incubation. ␣ 2 M* treatment of 1-LN cancer cells also caused a 50 -60% sustained increase in NF-B2 (Fig. 5C). Furthermore, whereas activation of NF-B1 via the canonical pathway was PI 3-kinase-dependent, activation of NF-B2 via the noncanonical pathway was PI 3-kinaseindependent (Fig. 5D). Incubation of 1-LN cells with LY294002 (20 M/20 min) before the addition of ␣ 2 M* inhibited the increase in the protein levels of NF-B1 and p-IB-␣ but had no effect on NF-B2 and NIK protein levels compared with ␣ 2 M*-treated cells (Fig. 5D). That activated NF-B1 and NF-B2 translocate to nuclei in ␣ 2 M*-stimulated 1-LN cells is shown in Fig.  4E, where these proteins were present at concentrations about 2-fold higher than the buffer-treated controls.

Attenuation of ␣ 2 M*-induced Activation of IKK␣ Kinase and NF-B in 1-LN Cells
Transfected with GRP78 dsRNA-Above, we have shown that ␣ 2 M* treatment of 1-LN prostate cancer cells with ␣ 2 M* increases activation of IKK␣ and up-regulates both NF-B1 and NF-B2 in an Akt-dependent manner (Fig. 5). Transfection of 1-LN prostate cancer cells with GRP78 dsRNA nearly abolished ␣ 2 M*-induced activation of IKK␣ compared with cells treated with Lipofectamine and ␣ 2 M* or cells treated with scrambled dsRNA and ␣ 2 M* (Fig. 2). Since silencing GRP78 gene expression profoundly inhibited the activation of p-Akt Thr-308 and p-Akt Ser-473 kinases caused by ␣ 2 M* stimulation (Fig. 2), we suggest that activation of IKK␣, hence NF-B activation, is mediated by activated Akt. These observations are further supported by inhibited expression of p-IKK␣/␤, p-IB␣, NF-B1, and NF-B2 in 1-LN cells transfected with GRP78 dsRNA and stimulated with ␣ 2 M* as compared with cells treated with Lipofectamine and ␣ 2 M* or cells transfected with scrambled dsRNA and treated with ␣ 2 M* (Fig. 2).
␣ 2 M* Treatment of 1-LN Cancer Cells Up-regulates the Expression of GADD45␤-GADD45␤ is a pivotal mediator of the cell-protective effects of NF-B against TNF␣ and Fas-induced apoptosis (78,79). NF-B induces GADD45␤ promoter activity, which down-regulates proapoptotic JNK. GADD45␤ associates tightly with MKK7, a selective activator of JNK, and inhibits its activation and apoptosis (80,81). Treatment of 1-LN cancer cells within ␣ 2 M* showed little or negligible activation of either MKK7 or JNK phosphorylation as determined by Western blotting (Fig. 6). Under our experimental conditions, JNK activation appears to be suppressed by Akt-dependent phosphorylation of ASK1 (Fig. 1B) and/or NF-B activation (Fig. 5, A and C). In the next series of experiments, we evaluated the role of GADD45␤ in JNK activation in 1-LN cancer cells treated with ␣ 2 M* by Western blotting (Fig. 5C). ␣ 2 M* treatment of 1-LN cells elevated the expression of GADD45␤ by about 2-fold at 10 -20 min of incubation but returned to basal levels at 60 min (Fig. 5C). The results indicate that the very modest effect of JNK in ␣ 2 M*-treated 1-LN cells occurs at the levels of Akt-dependent inhibition of ASK1 and GADD45␤ inhibition of MKKK7.
Up-regulation of Cyclin D1 and Down-regulation of p27 kip in 1-LN Cells Exposed to ␣ 2 M*[med]-During the G 1 to S transition, p27 kip is phosphorylated by kinases, including Akt, resulting in its nuclear exclu-sion and dissociation from E-CDK2 complexes and processing for degradation (82,83). As noted above, FOXO-induced cell cycle arrest is in part due to up-regulation of p27 kip . Increased levels of p27 kip cause G 0 /G 1 arrest by inhibiting cyclin-cyclin-dependent kinase complexes necessary for S-phase entry and progression (82,83). FOXO factors directly up-regulate p27 kip transcription and also prolong its halflife. The subcellular localization of p27 kip is also tightly regulated. Through the cell cycle, p27 kip shuttles between nucleus and cytoplasm. During the G 0 phase, p27 kip localization is completely nuclear, and mitogenic stimulation causes cytoplasmic redistribution, where it probably fails to inhibit events involved in the G 1 /S transition (82, 83).

GRP78-induced Proliferation of 1-LN Prostate Cancer Cells
Thus, in proliferating cells, the levels of p27 kip should be low, and that of cyclin D1 should be high, and the reverse should be true in cells undergoing cell arrest and apoptosis. 1-LN prostate cancer cells exposed to ␣ 2 M* showed low and unchanged levels of p27 kip protein in contrast to the levels of cyclin D1 protein, which were elevated at early periods of incubation (Fig. 4B). These results indeed support the growth-promoting effect of ␣ 2 M* on 1 LN cells.
␣ 2 M* Up-regulates the Expression of XIAP in 1-LN Prostate Cancer Cells-The IAPs are a family of intracellular antiapoptotic proteins that play a key role in cell survival by modulating death signaling pathways at a postmitochondrial level (84). These proteins characteristically contain a caspase recruitment domain, an NH 2 terminus repeat motif, which is necessary for its biological activity, and a RING finger domain at their COOH terminus, which is responsible for ubiquitination and degradation of IAP after an apoptosis stimulus (84). Among IAPs, XIAP is the most potent inhibitor of caspases and apoptosis. XIAP directly inhibits caspase-9 (84). XIAP is a physiological substrate of Akt, which phosphorylates it on Ser-87 (85), thereby inhibiting ubiquitination. These effects reduce XIAP degradation, and the increased levels of XIAP are associated with decreased caspase activity and programmed cell death (84). ␣ 2 M* induced a severalfold increase in XIAP (Fig. 4A), and treatment of the cells with PI 3-kinase inhibitor LY294002 before ␣ 2 M* exposure significantly reduced levels of XIAP (Fig. 4C). It is therefore suggested that in addition to down-regulating the activities of several proapoptotic components of programmed cell death, Akt also aids in the attenuation of cell death by preventing XIAP degradation, thus increasing its effective concentration.

Silencing of GRP78 Gene Expression Inhibits the Expression of ␣ 2 M*induced Prosurvival Factors in 1-LN Prostate Cancer Cells-Since both
Akt and NF-B promote cell survival by up-regulating the expression of antiapoptotic factors, one would expect that in 1-LN cells transfected with GRP78 dsRNA, the expression of these antiapoptotic factors would be down-regulated and those of proapoptotic factors would be up-regulated. Indeed, this is correct (Fig. 6). Silencing GRP78 gene expression down-regulated levels of p-FOXO, p-GSK3␤, XIAP, and cyclin D1, whereas levels of p27 kip were up-regulated in 1-LN prostate cancer cells stimulated with ␣ 2 M* (Fig. 6).
Activation of MKK3/6, ASK1, p38 MAPK, JNK, and TRAF2 in 1-LN Prostate Cancer Cells Exposed to ␣ 2 M*-The ability of cells to react to environmental changes is dependent on the cooperation of intracellular signal transduction pathways to coordinate cellular responses. In most cases, p38 MAPK is simultaneously activated with JNK. ASK1 activates both the p38 MAPK and JNK pathways by directly phosphorylating MKK4/MKK7 and MKK3/MKK6. In the next series of experiments, we examined the effect of ␣ 2 M* on the activation of stress-activated MAPKs, namely p38 MAPK and JNK in 1-LN prostate cancer cells. Exposure of 1-LN cells to ␣ 2 M* (50 pM) elevated levels of TRAF2 (Fig.  5A) by about 1.5-2-fold transiently and elevated the level of ASK1 at about 10 min of incubation, which remained elevated until the time of incubation (Fig. 5A). Stimulation of 1-LN cells with ␣ 2 M* elevated p-MKK3/6 and p-MKK4 at about 10 min of incubation but showed no effect on the phosphorylation of MKK7, the specific kinase, which activates JNK (Fig. 7B). The results show that ␣ 2 M*-induced receptor activation in 1-LN cells up-regulates ASK1 activity, which activates kinases upstream of p38 MAPK and JNK activation. Akt-mediated phosphorylation of ASK1 inhibits its activity (8). To understand the possible role of ASK1 phosphorylation in suppression of JNK activation in 1-LN cells stimulated with ␣ 2 M*, we determined levels of p-ASK1 (Fig. 7B). Levels of p-ASK were profoundly elevated under our experimental conditions (Fig. 7B). Incubation of cells with the PI 3-kinase inhibitor LY294002 before ␣ 2 M* addition profoundly reduced levels of p-ASK1 compared with controls (Fig. 1B). This indicates a possible role of Akt in suppressing JNK activation (Fig. 7B). An increase in the levels of activated p38 MAPK was observed at about 20 -40 min of incubation (Fig. 7A). In contrast, ␣ 2 M* treatment of 1-LN prostate cancer cells only slightly affected levels of p-JNK (Fig. 7B). ␣ 2 M* showed a differential effect on the activation of p38 MAPK and JNK as assessed by Western blotting. An increase in the levels of activated p38 MAPK could be observed at about 20 -40 min of incubation (Fig. 7A). In contrast, ␣ 2 M* treatment of 1-LN prostate cancer cells only slightly affected the levels of p-JNK (Fig. 7B).
Up-regulation of UPR Signaling Cascade in 1-LN Cells Exposed to ␣ 2 M*-An increased expression of GRP78 protein is part of the UPR required to alleviate ER stress, maintain ER functions, and protect cells against cell death (14 -17). GRP78 binds to three mammalian UPR transducers (namely, IRE1, PERK, and ATF6) and maintains them in an inactive state in the absence of ER stress (14 -17). Prolonged activation of IRE1 promotes binding to TRAF2 and subsequent recruitment of caspase-12, an ER-specific inducer of apoptosis (16, 29, 49 -51). The transient inhibition of protein synthesis during UPR is achieved by the activation of PKR-like ER kinase (PERK), which phosphorylates the eukaryotic translation initiator factor-2␣. The transient inhibition of protein synthesis during UPR allows the transcription of ATF4 transcription factor. ATF4 induces GADD34, which in turn dephosphorylates eIF2␣, allowing protein synthesis to continue (14 -17). The block in protein synthesis that limits the accumulation of unfolded proteins in the ER also induces a G 1 cell cycle arrest, activation of NF-B, and induction of the apoptotic transcription factor GADD153 (14 -17). In view of several reports on the up-regulation of UPR signaling in a variety of cancer cells, we have examined the expression of various components of UPR in 1-LN cells treated with ␣ 2 M* for varying periods of time by Western blotting in the following parameters: GRP78, IRE1␣, XBP-1, ATF6, ATF4, PERK, eIF2␣, GADD153, and GADD34 (Figs. 8 and 9). ␣ 2 M*-induced expression of the ER stress biomarker, GRP78, showed a steady increase up to 60 min of incubation (Fig. 8). The increase in GRP78 protein was ␣ 2 M* concentration-dependent, and the maximal increase occurred at 50 pM of ␣ 2 M* (data not shown). ␣ 2 M* treatment of 1-LN cancer cells also caused a sustained increase in IRE1-␣ and XBP-1 (Fig. 8). Under these experimental conditions, levels of ATF6, which have been reported to increase the transcription of GRP78 and XBP1, also showed a 2-fold increase at about 20 min of incubation (Fig. 8). Thus, activation of ATF6 signaling provides a positive feedback for UPR activation.
ER stresses that activate the transcriptional components of UPR also transiently attenuate protein synthesis, a response that is coupled to phosphorylation of PERK and eIF2␣ (14 -17). We next determined the levels of p-PERK, p-eIF2␣, and eIF2␣ in 1-LN cells exposed to ␣ 2 M* by Western blotting (Fig. 9). 1-LN cancer cells exposed to ␣ 2 M* (50 pM) showed an increase in p-PERK and p-eIF2␣ at about 20 min of incubation, which remained elevated up to the time of incubation (Fig. 9A). ␣ 2 M* treatment of 1-LN cancer cells elevated the protein levels of both ATF4 and GADD34 at about 10 -20 min of incubation, and these levels remained elevated until the time of incubation (Fig. 9B). The prime function of UPR expression is to protect cells against ER stress-induced damage to the cells; therefore, one would expect that under these conditions the cell will exhibit maximally the antiapoptotic and minimally the proapoptotic signaling. The block in protein synthesis, which limits the accumulation of unfolded proteins in the ER, also induces a G 1 cell cycle arrest, activation of the antiapoptotic protein NF-B, and induc- tion of the proapoptotic transcription factor GADD153 (86). As expected, exposure of 1-LN cancer cells treated with ␣ 2 M* (50 pM) showed no affect on the levels of proapoptotic GADD153 (Fig. 9). Our results indeed suggest that 1-LN prostate cancer cells are under ER stress and that these cells take remedial steps by up-regulating UPR signaling, which protects them against adverse effects of ER stress. PERK-induced phosphorylation of eIF2␣ has recently been shown to activate NF-B by causing translational inhibition of new IB synthesis (57).
Attentuation of ␣ 2 M*-induced UPR Signaling in 1-LN Prostate Cancer Cells Transfected with GRP78 dsRNA-Above, we have shown that ␣ 2 M* stimulation of 1-LN prostate cancer cells up-regulates UPR signaling. Since GRP78 is critically involved in the regulation of UPR signaling during ER stress, we have modulated levels of GRP78 by silencing GRP78 gene expression by RNAi in 1-LN prostate cancer cells and studying its effects on the expression of various components of the UPR signaling cascades dependent on ␣ 2 M* stimulation (Fig. 10). As expected, the limitations imposed on the availability of intracellular GRP78 by inhib-iting its new synthesis resulted in the down-regulation of IRE1␣, XBP-1, ATF6, p-PERK, and p-eIF2␣ compared with control cells treated with Lipofectamine and ␣ 2 M* (Fig. 10). However, levels of GADD153, an apoptosis-inducing protein, were elevated by 2-3-fold, and those of ATF4 remained nearly unaffected in cells transfected with GRP78 dsRNA compared with control cells treated with Lipofectamine and ␣ 2 M* (Fig. 10). The results demonstrate that restrictions imposed on the availability of GRP78 cause impairments of cell proliferative and cell survival signaling triggered by the binding of ␣ 2 M* to cell surface-associated GRP78 and UPR signaling resulting in the induction of proapoptotic signaling.

DISCUSSION
Highly metastatic 1-LN prostate cancer cells were derived from less metastatic PC3 prostate cancer cells. 1-LN prostate cancer cells differ from PC-3 cancer cells in one very important aspect; the former express GRP78 on their cell surface (19 -22). GRP78 is a member of the HSP70 family, and its expression is greatly induced upon pharmacologically and/or pathophysiologically induced ER stress, which causes the elicitation of UPR signaling to promote cell survival against the adverse effects of ER stress (see Ref. 14 and references therein). ␣ 2 M* binds to cell surface-associated GRP78 as well as to the low density lipoprotein receptor-related protein (25). However, the number of binding sites, binding affinities, and signaling events elicited after receptor binding of ␣ 2 M* to GRP78 and low density lipoprotein receptor-related protein are very different (87,88). In contrast to low density lipoprotein receptorrelated protein, the binding of ␣ 2 M* to GRP78 elicits mitogenic signaling and promotes cell proliferation in murine peritoneal macrophages (89 -92). ␣ 2 M* binding to 1-LN cells, but not PC-3 cells, also promotes such events (20 -22). One might also expect that in 1-LN prostate cancer cells, the ER-resident pool and cell surface resident pool of GRP78 would protect and promote the cell survival and proliferation of these cells by both up-regulation and cross talk between UPR signaling and receptor-activated mitogenic and proliferative signaling, respectively. In this paper, we report that this indeed happens. The major points of this study are as follows. 1) Stimulation of 1-LN prostate cancer cells with ␣ 2 M* up-regulates the activation of proproliferative and antiapoptotic signaling mediated by ERK1/2 and PI 3-kinase. As a consequence, p-Akt Thr-308 and p-Akt Ser-473 kinases are activated, as evidenced by the inhibition of caspase-12, caspase-9, and caspase-3 activation; phosphorylation of BAD at residue Ser-136; suppression of ASK1 activity by its PI 3-kinase-dependent phosphorylation; up-regulation of Bcl-2, XIAP, p-FOXO1, p-GSK3␤, and cyclin D1; and down-regulation of p27 kip . 2) ␣ 2 M* treatment of 1-LN cells up-regulates ASK1 and transient activation of MKK3/6, MKK4, and p38 MAPK with little or no effect on the activation of MKK7 and JNK. 3) ␣ 2 M* treatment caused up-regulation of TRAF2 and GADD45␤. ␣ 2 M* exposure activated NF-B by both the canonical (NF-B1) and noncanonical (NF-B2) pathways, as evidenced by the increased activation of IKK␣/␤, increased phosphorylation of IB-␣ in a PI 3-kinase-dependent manner, and activation of NIK in a PI 3-kinase-independent manner. 4) ␣ 2 M* treatment of 1-LN cells activated the unfolded protein response by up-regulating the expression of GRP78, IRE1␣, XBP-1, ATF6, p-PERK, p-eIF2␣, ATF4, and GADD34 and down-regulating the expression of GADD153 (5). Silencing GRP78 gene expression by RNAi suppressed cell survival signaling triggered by ␣ 2 M* binding to cell surface-associated GRP78 as well as cell survivalpromoting UPR signaling regulated by ER-associated GRP78. The molecular chaperone GRP78 regulates the IRE1, ATF6, and PERKtransduced signaling pathways of UPR; however, these signaling path- *, values significantly different from corresponding buffer-treated cells and GRP78 dsRNA-transfected and ␣ 2 M*-treated cells; **, significantly different from buffer-treated controls at the 5% level. Actin was used as a protein loading control in all experiments, but only a representative actin immunoblot is being shown.
ways may have distinctive sensitivities to fluctuations of the free GRP78 pool (34,72).
Taken together, these studies demonstrate that the effect of ␣ 2 M* on prostate cancers that express high levels of GRP78 on their cell surface is to activate multiple mechanisms that directly promote cellular proliferation but also to block mechanisms triggering programmed cell death. As noted earlier in this report, the occurrence of antibodies against GRP78 in the plasma of patients with prostate cancer correlates with a very poor prognosis (23,24). Moreover, histologic examination of human prostate cancers clearly demonstrates that GRP78 is dramatically up-regulated both intracellularly and on the cell surface of aggressive tumors (24). In the current report, we provide a mechanistic basis for these clinical observations. It is of interest that anti-GRP78 antibodies may appear in the plasma of patients with prostate cancer (24). Clearly, these are antibodies against "self." Why such autoantibodies should appear is unclear, although it is a frequent observation in patients bearing malignant tumors that altered immune states exist and may contribute to tumor progression. Whether the antibodies have any effect on the behavior of prostate cancer is as yet unknown. Such antibodies might themselves function as an ␣ 2 M* agonist, or they might antagonize the function of ␣ 2 M* by blocking binding of this ligand to the cell surface. In any event, ultimately the presence of a high concentration of these antibodies in the plasma is a harbinger of a poor outcome.
In summary, here we show that treatment of 1-LN cells with ␣ 2 M* promotes their proliferation by activating ERK1/2, p38 MAPK, and PI 3-kinase signaling cascades and their cell survival by activating Akt and NF-B signaling. To overcome the ER stress induced by their aggressive growth, these cells trigger UPR. There is an extensive cross-talk between these signaling cascades that contributes to their growth and survival. We have schematically presented these events in Fig. 11.