Abstract
The aggressiveness of pancreatic ductal adenocarcinoma (PDA) is characterized by its high metastatic potential and lack of effective therapies, which is the result of a lack of understanding of the mechanisms involved in promoting PDA metastases. We identified Annexin A2 (ANXA2), a member of the Annexin family of calcium-dependent phospholipid binding proteins, as a new molecule that promotes PDA invasion and metastases. We found ANXA2 to be a PDA-associated antigen recognized by post-treatment sera of patients who demonstrated prolonged survival following treatment with a PDA-specific vaccine. Cell surface ANXA2 increases with PDA development and progression. Knockdown of ANXA2 expression by RNA interference or blocking with anti-ANXA2 antibodies inhibits in vitro invasion of PDA cells. In addition, post-vaccination patient sera inhibits in vitro invasion of PDA cells, suggesting that therapeutic anti-ANXA2 antibodies are induced by the vaccine. Furthermore, cell-surface localization of ANXA2 is tyrosine 23 phosphorylation-dependent; and tyrosine 23 phosphorylation is required for PDA invasion. We demonstrated that tyrosine 23 phosphorylation resulting in surface expression of ANXA2 is required for TGFβ-induced, Rho-mediated epithelial-mesenchymal transition (EMT), linking the cellular function of ANXA2 which was previously shown to be associated with small GTPase-regulated cytoskeletal rearrangements, to the EMT process in PDA. Finally, using mouse PDA models, we showed that shRNA knock-down of ANXA2, a mutation at tyrosine 23, or anti-ANXA2 antibodies, inhibit PDA metastases and prolong mouse survival. Thus, ANXA2 is part of a novel molecular pathway underlying PDA metastases and a new target for development of PDA therapeutics.
Introduction
Pancreatic ductal adenocarcinoma (PDA) remains a lethal cancer with an overall 5-year survival rate of <5% [1]. Inability to diagnose early, high metastatic potential, and drug resistance account for its low survival rate. Although it is well-established that the pathogenesis of PDAs follows stepwise stages that display increasing cellular atypia and accumulate clonal mutations or aberrant expression of oncogenes or tumor suppressor genes such as K-Ras, p16, p53, and DPC4/SMAD4 [2], drugs that target these molecular abnormalities have not yet translated into improved clinical responses [3]. The aggressive nature of PDA is featured by its high incidence of metastases at the time of initial diagnosis and high incidence of early metastases following surgical resection. However, little is known about the molecular mechanisms underlying its invasion and metastatic processes. A better understanding of these mechanisms is essential for the development of innovative and improved treatments for PDA.
Cancer immunotherapy treatment approaches are under development for PDA. We developed an allogeneic, granulocyte-macrophage colony-stimulating factor (GM-CSF) secreting PDA vaccine for patients with PDA[4], [5], [6]. Phase I and II trials evaluating this vaccine in patients with resected PDA demonstrated both clinical and immunologic responses[4], [5]. This immunotherapy approach provided immunized lymphocyte reagents for identifying novel PDA antigens that are currently being tested as targets for PDA therapy[7], [8]. The potential therapeutic targets identified so far have also provided important clues for the study of molecular mechanisms underlying PDA development and metastases. Here we report the utilization of a functional proteomic approach that identified Annexin A2 (ANXA2) as a new candidate PDA target of the immune response. In addition, we show that tyrosine 23 phosphorylation-dependent cell-surface localization of Annexin A2 is required for epithelial to mesenchymal transition, invasion, and metastases formation of PDAs.
Results
Identification of ANXA2 as a new candidate PDA tumor-associated antigen and biomarker
We used immunized sera from two subjects who demonstrated both evidence of post-vaccination cellular immune responses and prolonged disease-free survival (DFS) and overall survival (OS) in a phase II study of a GM-CSF secreting whole cell PDA vaccine [4] to screen whole cell extracts from the PDA vaccine cell lines which served as the proteome. Protein extracts were separated by a two-dimensional electrophoresis (2-DE); and immunoblot analysis was performed to compare antigen recognition by pre- and post-vaccination sera. Proteins recognized by post-vaccination sera relative to pre-vaccination sera were identified by mass spectrometry. ANXA2 was a protein identified on the post-vaccination sera immunoblots of both patients evaluated. To further evaluate the prevalence of post vaccination humoral responses to ANXA2, purified recombinant ANXA2 was used to screen pre-vaccination and post-vaccination sera from 16 additional patients treated in this phase II study by both ELISA and western blot (Figure S1). Vaccine induced anti-ANXA2 antibodies, measured by a sandwich ELISA, were detected in 6 of 7 patients who demonstrated a DFS greater than 36 months [4], and only in 1 of the other 9 patients who did not demonstrate long-term DFS ( Table 1 ). These data provide the first evidence that ANXA2 is an antibody target of immune responses against PDA.
Table 1. Correlation between vaccine-induced anti-ANXA2 antibody responses and patients' disease-free survival.
Patient ID* | Disease Status | Disease-Free SurvivalTime (Mo)** | Increased Anti-ANXA2 antibody response(ELISA) |
3.009 | Disease free | >36 | + |
3.010 | Disease free | >36 | - |
3.012 | Disease free | >36 | + |
3.016 | Disease free | >36 | + |
3.027 | Disease free | >36 | + |
3.028 | Disease free | >36 | + |
3.031 | Disease free | >36 | + |
3.041 | Recurrent | <36 | - |
3.023 | Recurrent | <36 | - |
3.032 | Recurrent | <36 | - |
3.025 | Recurrent | <36 | + |
3.033 | Recurrent | <36 | - |
3.039 | Recurrent | <36 | - |
3.001 | Recurrent | <36 | - |
3.004 | Recurrent | <36 | - |
3.037 | Recurrent | <36 | - |
*Patients underwent surgical resection of PDA and were treated with a GM-CSF secreting PDA vaccine in a phase II clinical trial.
**Disease-free survival is defined as the time from surgery until disease recurrence or until the last follow up on April 21, 2008; Mo, month.
ANXA2 is reported to be overexpressed in a variety of cancers including PDA when compared with normal tissues[9]. However, ANXA2 is normally a cytoplasmic and lumenal residing protein in pancreatic tissue, and previous studies[9] have not determined whether cell surface ANXA2 expression is a dominant pattern in PDA tissues. Therefore, we analyzed the location of ANXA2 expression by immunohistochemistry (IHC) in the resected tumors from 52 of the 60 patients treated in our Phase II study for whom specimens were available for staining. We found that normal pancreatic ductal epithelial cells show weak cytoplasmic and lumenal staining by IHC, whereas cell-surface localized ANXA2 increases with progression from PanIN lesions to invasive PDA (Figure S2). Specifically, 39 (75%) of 52 fresh pancreatic tumor tissue samples tested have increased cell surface expression of ANXA2 (Figure S2). These data provide further support that ANXA2 surface expression is associated with PDA development and as such may serve as an immunologic target.
Inhibition of ANXA2 suppresses the in vitro invasion of PDA cells
We next investigated whether the cell surface localization of ANXA2 plays a biologic role in facilitating PDA invasion. ANXA2 has been reported to bind membrane-associated phospholipids and have diverse cellular functions including plasminogen activation, fibrinolysis, membrane transport, cytoskeleton rearrangement, angiogenesis, cell adhesion and migration. ANXA2 also functions as a high-affinity receptor for multiple extracellular ligands that have been implicated in cancer development, invasion, and metastases [10], [11], [12], [13]. To directly test whether ANXA2 is involved in PDA invasion, ANXA2 expression was knocked down in PDA cells by RNA interference ( Figure 1A ). Knock-down of ANXA2 suppressed the in vitro invasion of PDA cells in a Boyden chamber assay ( Figure 1B and Figure S3). The induction of antibodies against ANXA2 that is observed in vaccinated patients with prolonged DFS ( Table 1 ) suggests that anti-ANXA2 antibodies may have a direct anti-tumor effect. We therefore tested both rabbit polyclonal and mouse monoclonal anti-ANXA2 antibodies and found that they can specifically inhibit in vitro invasion of PDA cells ( Figure 1C,D ). Moreover, sera from immunized patients who demonstrated a post-vaccination response to ANXA2 similarly inhibited in vitro invasion of PDA cells ( Figure 1E ). The data presented so far link increasing cell surface expression of ANXA2 with PDA invasion capability and suggests that vaccine-induced antibody responses may inhibit this aspect of PDA progression. However, the mechanism by which ANXA2 mediated PDA invasion occurs has yet to be explored. Interestingly, the invasive capacity of PDA cells is not correlated with their proliferative rate suggesting an independent mechanism (Figure S3). To uncover other regulatory mechanisms that account for the invasion capacity of PDA cells, we further examined the sub-cellular localization of ANXA2 in various PDA cell lines by fluorescent staining with anti-ANXA2 antibodies (Figure S4). ANXA2 is predominantly localized to the cell membrane in all 8 PDA cell lines found to have high invasion capacity, whereas ANXA2 is present predominantly in the cytoplasm of cell lines with low invasion capacity (Figure S4 and Table S1). This data further support a role for ANXA2 translocation from the cytosol to the cell surface/membrane in enhancing PDA cell invasion.
Phosphorylation of ANXA2 at Tyr23 promotes the cell-surface localization of ANXA2 and the invasion capacity of PDA cells
ANXA2 is a substrate for Src kinase, which phosphorylates ANXA2 at Tyr23 both in vivo and in vitro [10], [11], [12], [13], and Tyr23 phosphorylation has been suggested to be important for normal cell scattering and branching morphogenesis[14], [15]. ANXA2 is also reported to be tyrosine-phosphorylated when it localizes to the cell surface under stress [16]. Since malignant cells often mimic normal cells that have been subjected to a variety of stress stimuli, we postulate that ANXA2 is translocated to the cell surface as a tyrosine-phophorylated protein during tumorigenesis as well. To test this, we eluted the cell surface fraction of ANXA2 from Panc10.05 PDA cells, which have cell surface localization of ANXA2 (Figure S4 and Table S1), and found the cell surface fraction of the ANXA2 protein is in fact a tyrosine-phosphorylated protein ( Figure 2A ). In contrast, ANXA2 could not be eluted from the cell surface of Panc 3.11 cells, a PDA cell line that demonstrated cytoplasmic localization of ANXA2 (Figure S4). To test whether phosphorylation of ANXA2 at Tyr23 is important for its localization to the PDA cell surface, we generated a panel of plasmids expressing either wild-type ANXA2 (ANXA2WT), the ANXA2 mutant protein (ANXA2Y23A) in which Tyr23 was altered to an alanine residue making a non-phosphorylatable mutant, or the ANXA2 mutant protein (ANXA2Y23E) in which Tyr23 was altered to a glutamic acid residue mimicking constitutive phosphorylation. When Panc10.05 cells were transfected with these plasmids expressing ANXA2 tagged by GFP [17], ANXA2WT-GFP and ANXA2Y23E-GFP localized predominantly to the cell surface of PDA cells. In contrast, ANXA2Y23A-GFP localized to the cytoplasm ( Figure 2B ). These results were further confirmed by using a lentivirus to constitutively express ANXA2WT, ANXA2Y23A, or ANXA2Y23E in the PDA cells (Figure S4). Taken together, these data demonstrate that phosphorylation at Tyr23 results in the localization of ANXA2 at the cell surface.
To determine whether the change in ANXA2 localization that occurs as a result of Tyr23 phosphorylation affects the invasion capacity of PDA cells, a set of plasmids that express exogenous FLAG-tagged ANXA2 including ANXA2WT-FLAG, ANXA2Y23A-FLAG, and ANXA2Y23E-FLAG were developed. These vectors are RNA interference resistant because of silent mutations within the siRNA target site. Panc10.05 PDA cells transfected with these plasmids were fractionated into cytoplasmic and cell membrane fractions (Figure S4). We first confirmed that only ANXA2 WT-FLAG and ANXA2Y23E-FLAG, but not ANXA2Y23A-FLAG, localize to the cell membrane fraction ( Figure 2C ). As expected, ANXA2 WT-FLAG protein is tyrosine phosphorylated in the cell membrane fraction. Next, we found that co-transfection of the pcDNA plasmid expressing ANXA2WT-FLAG or ANXA2Y23E-FLAG, but not ANXA2Y23A-FLAG, with the siRNA (to inhibit endogenous ANXA2), reversed siRNA-mediated inhibition of invasion of Panc10.05 cells ( Figure 2D ). However, in cells with low invasion capacity and only cytoplasmic localization of ANXA2, such as Panc3.11 (Table S1), co-transfection with ANXA2Y23E-FLAG bypasses the phosphorylation regulatory mechanism by mimicking constitutive phosphorylation and promotes the invasion of Panc3.11 cells ( Figure 2E ). These data suggest that Tyr23 phosphorylated ANXA2 confers PDA invasion capacity.
ANXA2 contributes to the Epithelial-Mesenchymal Transition of PDA cells
Phosphorylated ANXA2 plays a role in cell scattering in normal morphogenesis processes[14], [15]. Our data so far support a role for phosphorylated ANXA2 in PDA invasion. The epithelial to mesenchymal transition (EMT) regulates the normal morphogenic process during embryonic development and tissue restructuring, and the initial steps of invasion and metastases are suggested to mimic EMT[18]. Therefore, we sought to determine whether ANXA2 is required for the EMT in PDA cells. EMT is characterized by the suppression of the transcription of epithelial markers such as E-cadherin and induction of mesenchymal markers such as slug and vimentin. ANXA2 has been reported to mediate TGFβ-activated EMT during the process of cardiac valve development [19]. In addition, TGFβ is reported to induce EMT in cultured PDA cells [20], [21]. To examine whether ANXA2 has a direct role in the EMT process of invading PDA cells, a lentiviral vector containing ANXA2 shRNA was used to achieve long-term suppression of ANXA2 in PDA cells (Figure S3). Real-time PCR analysis showed that E-cadherin was suppressed, whereas slug and vimentin were induced during TGFβ-induced EMT in Panc10.05 cells with control shRNA, but not in those infected with ANXA2 shRNA ( Figure 3A ). In addition, E-cadherin protein expression was suppressed in TGFβ-treated cells with control shRNA, but remained unchanged in TGFβ-treated cells that also expressed ANXA2 shRNA ( Figure 3B,C ). As predicted, PDA cells without ANXA2 shRNA lose their cell-cell adhesion phenotype and scatter around the culture slip following TGFβ treatment, reminiscent of an EMT pattern ( Figure 3B ). Although ANXA2 has not yet been shown to be involved in SMAD4-mediated EMT, it has been shown to be involved in Rho (small GTPases)-mediated cell detachment, a trait of EMT [15]. Therefore, we also evaluated whether Rho mediates ANXA2-associated EMT in PDA and found that Rho activation is not detected in PDA cells with shRNA inhibition of ANXA2 following TGFβ treatment ( Figure 3C ). These results demonstrate that loss of ANXA2 expression leads to loss of TGFβ-Rho-mediated EMT in PDA cells.
We next examined whether Tyr23 phosphorylation is important for ANXA2-mediated EMT in PDA cells. We observed that the endogenous ANXA2 no longer localizes to the cell surface in the cells expressing the tyrosine site loss variant ANXA2Y23A (Figure S4). In addition, transfection with ANXA2Y23A-FLAG inhibits the invasion of Panc10.05 cells (Figure S4), suggesting that ANXA2Y23A has a dominant negative effect. Consistent with published data [22], we found that ANXA2Y23A can still bind to its partner S100A10/p11 [23] in the cytosol, but not in the cell membrane (Figure S5). Therefore, overexpressed, cytoplasmic-localized ANXA2Y23A may sequester all of the S100A10/p11 in the cytosol, thereby conferring a dominant negative effect. Therefore, taking advantage of the dominant negative effect of ANXA2Y23A, and employing this ANXA2Y23A mutant, we further demonstrated that EMT is induced by TGFβ in cells expressing ANXA2WT, but not in cells expressing ANXA2Y23A ( Figure 3D ). Thus, these data further demonstrate that the Tyr23 phosphorylation of ANXA2 promotes the EMT of PDA cells and is one conceivable mechanism by which ANXA2 localizes to the PDA cell membrane and confers the potential for PDA cells to invade.
Expression and tyrosine phosphorylation of ANXA2 are required for PDA metastases formation in vivo
Local invasion by tumor cells is a known step in the process of metastasis. Our data demonstrate that ANXA2 facilitates invasion of PDA cells in vitro. We therefore employed a transplantable murine pancreatic cancer model of metastases (Figure S6) to evaluate the role of ANXA2 expression, phosphorylation, and cell surface localization in the PDA metastasis process in vivo. In this model, 100% of the mice die with liver metastases at approximately 4-6 weeks ( Figure 4A ) after splenic injection of 2×106 Panc02 murine PDA cells. Panc02 cells infected with a GFP expressing lentivirus carrying shRNA specific for ANXA2 knockdown or control shRNA were sorted for GFP-positive cells by FACS. After confirming the knockdown of ANXA2 expression (Figure S3), 2×106 cells were injected into the splenic bed prior to performing a hemisplenectomy, and mice were monitored for survival. Mice injected with Panc02 cells expressing the ANXA2 shRNA survived significantly longer than mice injected with Panc02 cells expressing control shRNA (p<0.0001) ( Figure 4B ). Necropsy was performed on all mice that died, as well as those that still survived to day 90 following tumor implantation. Macroscopic inspection showed that 17/17 mice in the control group, but no mice (0/17) in the ANXA2 shRNA group developed liver macro-metastases ( Figure 4C ). Additional pathologic analyses indicated that mice in the ANXA2 shRNA group died due to tumor formed at the splenic bed (local regional tumor growth) (Figure S7). To confirm that PDA cells carrying ANXA2 shRNA are not able to seed the liver, 1×106 Panc02 cells labeled with Qtracker 565 were injected at the splenic bed. Five days later, frozen sections of liver were examined under fluorescent microscope for Qtracker 565-labeled cells ( Figure 4D ). Significantly fewer Qtracker 565-positive cells are seen in the livers of ANXA2 shRNA mice when compared with the control shRNA group (p<0.0001) ( Figure 4E ).
The ANXA2 protein sequence including Tyr23 is conserved in mammals and its phosphorylation can be induced in non-human mammalian cells [14], [15], [24]. Thus, the hemisplenectomy model was again employed to examine whether Tyr23 phosphorylation of ANXA2 is required for ANXA2-mediated murine PDA metastases. Mice injected with Panc02 cells stably expressing exogenous ANXA2Y23A survived significantly longer than mice injected with Panc02 cells stably expressing exogenous ANXA2WT (p = 0.001) ( Figure 4F ). All 12 mice in the control group whereas 0/12 mice in the ANXA2Y23A group developed liver macro-metastases ( Figure 4G ).
We also used a previously established orthotopic model in which PDAs metastasize to the peritoneum [25] to confirm that ANXA2 can initiate metastasis from a primary tumor in addition to facilitating the seeding of PDA metastases in the liver. This study confirmed that ANXA2 knockdown does not affect the growth of primary PDAs significantly (Figure S8). All mice in the control group developed large peritoneal macro-metastases. In contrast, most mice in the ANXA2 shRNA group did not develop macro-metastases. In addition, the mean size of metastatic lesions that formed in ANXA2 shRNA mice, typically manifesting as micro-metastatic foci, was significantly smaller (P = 0.0012) than the control mice (Figure S8). IHC analysis of cell surface ANXA2 demonstrated that these micro-metastatic foci were formed by cells that still expressed ANXA2 due to incomplete knockdown by ANXA2 shRNA (Figure S8). Taken together, our data provide strong evidence that Tyr23 phophorylated and cell surface expressed ANXA2 facilitates the metastatic process, and supports ANXA2 as a novel target for PDA therapy development.
Anti-ANXA2 antibodies suppress PDA metastases in vivo and prolong survival in a mouse model of PDA
Next, we evaluated whether an inhibitory ANXA2-targeted monoclonal antibody can suppress Panc02 hepatic metastases using the above described liver metastases forming mouse model. Specifically, Panc02 injected mice were treated twice weekly with a mouse monoclonal anti-ANXA2 antibody (shown to inhibit Panc02 invasion in vitro in Figure 1 ) until death. All mice (10/10) in the control IgG group whereas only 1/9 mice in the ANXA2 antibody group developed liver macro-metastases ( Figure 5A ). Furthermore, mice treated with the anti-ANXA2 antibody survived significantly longer than mice treated with isotype control IgG (p = 0.02). Similar to the mice receiving ANXA2 shRNA, mice in the ANXA2 antibody group died due to tumor progression at the splenic bed, suggesting this anti-ANXA2 antibody therapy does not inhibit the growth of primary PDAs. Predictably, mice implanted with Qtracker 565-labeled Panc02 cells and treated with a single intraperitoneal injection of anti-ANXA2 antibody were found to have significantly (p<0.0001) fewer Qtracker 565-positive cells in their livers as compared with mice given the control mAb ( Figure 5B,C ). These data further support ANXA2 as a target for therapeutic intervention of metastases.
Discussion
ANXA2 was brought to our attention as a target of vaccine induced immune responses identified on a proteomic screen of PDA proteins using immunized sera from patients who demonstrated prolonged DFS. A previously published study links ANXA2 overexpression in pancreatic cancer tissue with rapid recurrence after gemcitabine adjuvant chemotherapy in postoperative patients [26]. In this study, we demonstrate three new findings that elucidate the role of ANXA2 in PDA invasion and metastases. First, localization of ANXA2 expression on the cell surface is required for PDA invasion in vitro and metastases formation in vivo. Second, Tyr23-phosphorylation is required for localization of ANXA2 on the PDA cell surface and subsequent PDA invasion and metastases formation. Importantly, knockdown of ANXA2 or inhibition with ANXA2 antibody therapy inhibits the metastatic process. Third, loss of ANXA2 expression or phosphorylation at Tyr23 leads to loss of TGFβ-Rho-mediated EMT in PDA cells. Taken together, these findings identify a new role for phosphorylated ANXA2 in mediating PDA cell invasion via Rho-regulated EMT and facilitating PDA metastases.
Our study shows for the first time that cell surface localization of ANXA2, mediated by Tyr23 phosphorylation, enhances PDA cell invasion in vitro and PDA metastases to liver in vivo. Relevant to our findings, a recent report showed that knockdown of ANXA2 in a PDA cell line inhibits cell migration in vitro [27]. We also evaluated a panel of PDA tumor cell lines and 52 fresh tissue specimens and showed the frequent expression of cell surface ANXA2 in primary PDAs, which further supports a role for ANXA2 in PDA progression. Another report demonstrated that the reduction in cell surface ANXA2 expression in response to changes in interferon gamma levels leads to reduced prostate cancer cell invasion[28], suggesting that the cell surface translocation of ANXA2 may be an important cellular process for multiple types of cancers[10].
We also show that ANXA2 phosphorylation is required for TGFβ-induced and Rho-mediated EMT phenotypes of PDA cells. A previous study showed that loss of Smad4 expression in PDA can lead to the aberrant activation of STAT3, which may contribute to the switch in function of TGFβ from a tumor-suppressive to a tumor-promoting EMT pathway in PDA [20]. Currently, knowledge of the EMT process in cancer development is largely limited to its characteristic transcription circuit. It is still unknown how cancer cells functionally undergo cytoskeletal rearrangement to exhibit mesenchymal features, interact with extracellular matrix, and migrate to distant locations when the EMT process is initiated. One of the cellular functions of ANXA2 is its involvement in cytoskeletal rearrangement, which is thought to be mediated by its tyrosine phosphorylation and its interaction with small GTPases [14], [15]. Thus, our data, linking ANXA2 and EMT in PDA, provide a new pathway that may regulate the EMT process that occurs in cancer invasion and metastases.
However, it is still not clear which tyrosine kinases phosphorylate ANXA2 in PDA. A number of tyrosine kinases including Src, insulin receptor, and insulin-like growth factor (IGF)-1 receptor have previously been implicated in regulating the cellular function and cell surface translocation of ANXA2 in normal morphogenesis processes[14], [15], [24]. Further investigation of these tyrosine kinases as mediators of ANXA2 phosphorylation in PDA is therefore warranted to further delineate the mechanism by which ANXA2 becomes phosphorylated in PDA. In addition, inhibitors of Src and IGF-1 receptor are under clinical development for the treatment of various cancers including PDA [29]. Therefore, ANXA2 phosphorylation or its cell surface translocation may also serve as a biomarker for assessing treatment response to these agents. Importantly, blocking ANXA2 by anti-ANXA2 antibodies can inhibit PDA invasion and metastases, suggesting that ANXA2 and its associated pathways are viable therapeutic targets.
We have used two mouse models of PDA that have provided complementary data supporting a role for ANXA2 in PDA progression and metastases formation in vivo. Data employing the PDA orthotopic model suggests that ANXA2 confers PDA cells with the invasion and metastasis potential when the PDA cells originate from the primary tumor. Data from the liver metastasis model suggests that ANXA2 also facilitates PDA cell metastasis formation once the PDA cells gain access to the blood supply supporting a metastastic site. Thus, ANXA2 appears to be important for the initial invasion process and for the survival of PDA cells at their metastatic destination. Accumulated evidence has supported the “seed-and-soil” hypothesis that the metastatic “seeds” of a given cancer only take root and grow in certain organs with a welcoming microenvironment or “soil” [30]. Our data support this hypothesis and suggest that ANXA2 facilitates PDA cell invasion and motility locally, but also provides an important signaling pathway linking the PDA cell to the growth signals in the new metastatic environment (the “soil”).
In summary, this study provides in vivo evidence supporting ANXA2 as a mediator of PDA invasion and metastases. Additional studies are needed to further elucidate the mechanisms by which ANXA2 interacts with PDA stroma at both the primary and metastatic tumor sites to accomplish these processes. Importantly, this study demonstrated that blocking ANXA2 by anti-ANXA2 antibodies can inhibit PDA metastases, providing the rationale for developing therapeutic agents that target ANXA2.
Materials and Methods
Cell lines and tissue culture
The human pancreatic ductal adenocarcinoma (PDA) cell lines were previously described [31], [32]. The human fibroblast cell line was established from human PDA paracancerous tissues. All human PDA cell lines were maintained in RPMI1640 media supplemented with 10% fetal bovine serum (FBS) in a humidified incubator at 37°C and 5% CO2. The mouse Panc02 cells were maintained in DMEM media supplemented with 10% FBS in a humidified incubator at 37°C and 10% CO2. When indicated, TGFβ1 (R&D Systems) was added in the culture medium at a final concentration of 400 pM. Cells were treated with TGFβ1 for 48 hours before harvest unless otherwise noted. The Panc02 cell line is an established chemical carcinogen induced mouse PDA cell line that originated from the C57Bl/6 mice [33].
Antibodies
The rabbit polyclonal anti-ANXA2 antibody (H50), mouse monoclonal anti-ANXA2 antibody (clone: ZO14), mouse monoclonal anti-p11 antibody, and the mouse monoclonal anti-E-cadherin antibody were obtained from Santa Cruz Biotechnology, Invitrogen, BD Transduction, and Zymed Laboratories, respectively.
Human serum
Human sera and pancreatic tumor tissue samples were obtained from patients enrolled in a phase II study of an allogeneic GM-CSF secreting whole cell PDA vaccine. Written informed consent was obtained in compliance with the Johns Hopkins Medical Institution Institutional Review Board (IRB)-approved J9988 protocol with approval number #00-01-13-02 [4]. Serum was collected and stored according to standard procedures.
Enzyme-Linked Immunosorbent Assay (ELISA)
A previously published ELISA assay was used to detect anti-ANXA2 antibody responses in patients [34]. The calibration curve was generated with the post-vaccination sera from patient 3.009.
DNA cloning and plasmid constructions
The full-length human ANXA2 cDNA was obtained by reverse transcription of total RNA purified from Panc10.05 cells, followed by high-fidelity PCR amplification with ANXA2 primers. The non-complementary region of the reverse primer also contained the FLAG tag sequence. The resultant PCR product of ANXA2 cDNA was then cloned into the pCR vector (Invitrogen) and was sequenced to confirm no introduction of missense or nonsense mutations. The ANXA2 cDNA fragment with a C-terminal FLAG tag was further subcloned into the lentiviral vector (LV). In this lentivirus, ANXA2 is expressed under the control of the EF-1α promoter. For the co-transfection of both plasmids and siRNA, the resultant PCR product of ANXA2 cDNA with a C-terminal FLAG tag was cloned into the pcDNA3.3 vector (Invitrogen) directly. Y23A and Y23E mutations were created by site-directed mutagenesis according to the manufacturer's manual (Stratagene).
Plasmid transfection, lentiviral infection, and RNA interference
For plasmid transfection and RNA interference, cells were seeded in multiple 6-well plates to 80% confluence. For each well, 2 µg of pcDNA-based plasmid and/or 40 pmol siRNA duplex, were transfected with lipofectamine 2000 in serum-containing medium according to the manufacturer's manual (Invitrogen). For protein expression analysis, cells were harvested at 48 hours. The ANXA2 siRNA was synthesized by Ambion, Inc.; the scramble siRNA was purchased from Ambion.
To produce lentivirus expressing ANXA2, the plasmid with lentiviral constructs was co-transfected with packaging plasmids into 293T cells as previously described [35]. Lentivirus supernatant was collected at 48 hours. For infection, cells were seeded in multiple 6–well plates to 80% confluence. For each well, 2 ml lentivirus supernatant was added and incubated for 48 hours before cells were harvested.
The lentivirus expressing human ANXA2 hairpin shRNA was obtained from Open Biosystems. Lentivirus was produced according to the manufacturer's manual. For infection of Panc10.05 cells, 6 ml of viral supernatant was added to adherent cells plated in each 75 cm flask and incubated for 48 hours. Cells from two flasks were sorted by GFP in a FACS cell sorter at 72–96 hours after infection. The cells infected with lentivirus expressing GFP alone were sorted. Total RNA was immediately extracted after cell sorting. The lentivirus expressing mouse ANXA2 hairpin shRNA was obtained from GeneCopoeia.
Cell invasion assays
Cell invasion assays were carried out using 96-well Transwell plates with 8-µm pores and reagents in the Cultrex BME Cell Invasion Assay system according to the manufacturer's manual with modification (R&D Systems). For all invasion experiments, 0.5xBME (Basement Membrane Extract) was used to coat the top well and 10% serum containing media was added to the bottom well. To score the cells across the transwell, MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assays were used as previously described [36]. Relative MTT units representing invasion capacity were measured and normalized by total cell numbers. To assess the spontaneous leakage of cells through the BME-coated transwell, invasion specific controls were also performed by adding serum-free media in the bottom wells. To exclude the effect of such leakage, relative MTT units in the invasion experiments were adjusted by subtracting the MTT values of leaked cells in matched invasion specific controls.
PDA cells were plated at 1×104 cells per well in triplicate. PDA cells were transfected with either ANXA2-targeted siRNA or control siRNA prior to plating. Invasion was measured at 48 hours following plating using an MTT readout system.
Fluorescent immunostaining
Panc10.05 cells were grown on cover slips to 90% confluence and were fixed in 4% paraformaldehyde for 15 min. Cover slips were then treated with PBS containing 0.1% Triton X-100 for 5 minutes followed by washing with PBS. After, cover slips were blocked with 10% normal goat sera in PBS for 1 hour and were then incubated with rabbit anti-ANXA2 antibodies or mouse anti-E-cadherin mAb at a 1∶100 dilution in 10% normal goat sera overnight at 4°C. Following a PBS wash, they were further incubated with FITC-conjugated goat anti-rabbit IgG or PE-conjugated goat anti-mouse IgG (Santa Cruz Biotechnology) at a 1∶200 dilution in 10% normal goat sera at room temperature for 1 hour. They were subsequently washed with PBS containing 0.5% NP-40. Cover slips were mounted in a medium containing DAPI (4′,6-diamidino-2-phenylindole) (Vector Labs) and examined by a fluorescent microscope.
Immunohistochemistry (IHC)
IHC staining for both human and mouse ANXA2 was performed using a standard protocol on an automated stainer from Leica Microsystems. After deparaffinization and hydration of tissue, heat induced antigen retrieval was performed with EDTA buffer (pH 9.0) for 20 minutes. Incubation with the H50 rabbit anti-AnxA2 antibody (Santa Cruz Biotechnology) at a 1∶100 dilution was followed by secondary antibody incubation from the bond polymer refine detection kit (Leica Microsystems). The reaction was developed using substrate 3,3′-Diamino-benzidine hydrochloride (DAB). All slides were counterstained with hematoxylin. Each area of PDA cells on the entire slide was scored from 0 to 3+ by two clinical pathologists (P.I. and R.A.) independently. Scores of 0 to 3+ measure the different intensities of cell-surface staining of AnxA2, with a score of 0 representing no staining and a score of 3+ representing the strongest staining. The percentage of PDA cells at each score level was estimated. The average score of cell-surface AnxA2 expression was calculated as follows:
(a%, b%, c%, and d% are the percentages of PDA cells with score 0 to 3, respectively.)
Whole cell extract and cell fractionation
PDA whole cell extracts were obtained as previously described [37]. In brief, cell pellets were resuspended in Lysis 250 buffer followed by a freeze and thaw process that was performed 3 times. The cell lysate was spun at 15,000 rpm for 10 minutes and supernatant was removed. The protocols to separate membrane and cytoplasmic fractions were adapted from those previously published [38]. EGTA (Ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid) elution of cell surface AnxA2 followed a previously established procedure [16].
Immunoprecipitation and immunoblot analysis
Anti-phosphotyrosine antibody conjugated sepharose (P-Tyr-100, Cell Signaling Technology) was used to immunoprecipate tyrosine-phosphorylated proteins. Anti-AnxA2 antibodies were first conjugated to sepharose beads according to the manufacturer's manual (Pierce) prior to being used for immunoprecipitation. Anti-FLAG M2 antibodies-conjugated beads were obtained from Sigma. For anti-p11 immunoprecipitation, cell fractionations were first incubated with anti-p11 antibodies and then with protein A sepharose beads. All immunoprecipitations were done at 4°C overnight, followed by washing with Lysis 250 buffer [37].
After whole cell extracts, cell fractions, or immunoprecipitants were boiled in SDS-sampling buffer, they were loaded on 10% gradient SDS-PAGE (BioRad). The gel was then transferred to a nitrocellulose membrane and blotted with the rabbit anti-AnxA2 polyclonal antibody at a 1∶1000 dilution followed by HRP-conjugated goat anti-rabbit IgG (Amershan Pharmacia) at a 1∶3000 dilution.
Recombinant His6-tagged ANXA2 was expressed in TOP10 E. coli and purified on a High-Trap Ni column according to the manufacturer's manual (Amershan Pharmacia). One microgram of purified His6-tagged ANXA2 was loaded on each well of a 10% gradient SDS-PAGE. After transferring to a nitrocellulose membrane, each individual lane was blotted with either pre-vaccination serum or post-vaccination serum at a 1∶1000 dilution. Mouse anti-human IgG antibody (Sigma) was used at a 1∶5000 dilution as the secondary antibody.
Total cellular Rho, or activated Rho which was isolated by incubating the cell extract with an activated Rho affinity binding column (GST-tagged Rho-binding domain of Rhotekin), were detected by western blot with mouse anti-Rho A,B,C mAbs provided by the Rho Activation Assay Kit according to the manufacturer's manual (Upstate).
Reverse transcription and real-time PCR
RNA was isolated from cells using the RNAEasy kit (Qiagen), and reverse transcribed using the first strand cDNA synthesis kit (Invitrogen). Quantitative real-time reverse transcription-PCR (qRT-PCR) was performed with gene-specific fluorescent TaqMan probes (Applied Biosystems) using an ABI PRISM 7500 Sequence Detection System Instrument and the associated software (Applied Biosystems) following the manufacturer's instructions. Each reaction was performed in triplicate at 2 cDNA dilutions. The standard human β-actin gene (BACT; Applied Biosystems) was used to normalize variations in the quantities of input cDNA.
GFP-sorted, shRNA-lentiviral infected cells were subjected to real-time RT-PCR analysis to measure the mRNA expression of E-cadherin, slug, and vimentin. As a negative control, the lentivirus expressing only GFP was also used to infect a control population of Panc10.05 cells. These cells were also sorted for GFP-positive cells for this analysis. One pair of Panc10.05 cell lines, with and without ANXA2 shRNA, were treated with TGFβ1 for 48 hours before they were harvested for mRNA expression analysis.
Mouse models of PDA
All animal experiments conformed to the guidelines of the Animal Care and Use Committee of the Johns Hopkins University and animals were maintained in accordance to guidelines of the American Association of Laboratory Animal Care (Approval# MO08M142).
The mouse liver metastasis model was established using a previously described hemispleen injection technique [39], [40]. The transplantable tumor model allows for the assessment of in vivo tumor growth following shRNA knockdown of targeted PDA expressed proteins. The spleens of anesthetized female C57Bl/6 mice ages 8 to 10 weeks were divided into two halves and the halves were clipped. Mouse PDA cells (2×106) were injected into the splenic bed (splenic artery and veins) through one hemispleen followed by a flush with the HBSS buffer. After 30 seconds, the splenic vessels draining the injected hemispleen was clipped and the hemispleen was removed. The abdominal wall was sutured, and the skin adapted using wound clips. All mice were followed three times a week for survival. At necropsy, mice were examined macroscopically; and multiple sections of livers and splenic bed injection sites were examined microscopically with H&E staining and, as indicated, anti-ANXA2 IHC.
For survival analysis, mice were treated intraperitoneally with 10 µg anti-ANXA2 monoclonal antibody ZO14 antibodies (Invitrogen) or isotype control IgG1 in 100 µl PBS twice a week starting the day following the hemisplenectomy procedure and continued until death. For the analysis of Qtracker-labeled cells, mice in each group were treated intraperitoneally either with 100 µg anti-ANXA2 monoclonal antibody ZO14 or with 100 µg isotype control IgG1 on the day following the hemisplenectomy procedure and sacrificed 5 days later.
Statistics
Statistical analysis was performed using software GraphPad Prism 5. Two tailed Mann-Whitney test was used to compare differences between treatment groups. Mouse survival was analyzed by the Kaplan-meier curve and the Log-Rank test. Mouse experimental data from independent experiments were shown in a combined analysis because of the consistency of the results.
Supporting Information
Acknowledgments
We thank Dr. Kiyoshi Yoshimura for his technical advice and Xiaoyu Pan for her technical assistance in performing the mouse hemisplenectomies and thank Dr. Stephen Moss for providing the GFP-ANXA2 construct.
Footnotes
Competing Interests: I have read the journal's policy and have the following conflicts: Under a licensing agreement between BioSante and the Johns Hopkins University, the University is entitled to milestone payments and royalty on sales of the vaccine product described in this manuscript. We do not have any other relevant conflicts of interest to be disclosed. P. Illei has given a lecture sponsored by Leica Microsystems. This does not alter our adherence to all the PLoS ONE policies on sharing data and materials.
Funding: This study was supported by the NCI SPORE in Gastrointestinal Cancers P50 CA062924‐14 (E.M.J.), Lustgarten Foundation (E.M.J.), Broad Foundation (E.M.J.), NCI R01 CA88058 (E.M.J.), Viragh Foundation, NIH 1K23 CA93566‐01A1 (D.L.), ASCO Young Investigator Award (L.Z.), NIH 5T32 CA0090701‐28 Training Grant (L.Z.), the Sol Goldman Pancreatic Cancer Center (L.Z.). Dr. Jaffee is the first recipient of the Dana and Albert "Cubby" Broccoli Endowed Professorship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References
- 1.Pierantoni C, Pagliacci A, Scartozzi M, Berardi R, Bianconi M, et al. Pancreatic cancer: progress in cancer therapy. Crit Rev Oncol Hematol. 2008;67:27–38. doi: 10.1016/j.critrevonc.2008.01.009. [DOI] [PubMed] [Google Scholar]
- 2.Goggins M, Kern SE, Offerhaus JA, Hruban RH. Progress in cancer genetics: lessons from pancreatic cancer. Ann Oncol. 1999;10(Suppl 4):4–8. [PubMed] [Google Scholar]
- 3.Strimpakos A, Saif MW, Syrigos KN. Pancreatic cancer: from molecular pathogenesis to targeted therapy. Cancer Metastasis Rev. 2008;27:495–522. doi: 10.1007/s10555-008-9134-y. [DOI] [PubMed] [Google Scholar]
- 4.Lutz E, Yeo CJ, Lillemoe KD, Biedrzycki B, Kobrin B, et al. A Lethally Irradiated Allogeneic Granulocyte-Macrophage Colony Stimulating Factor-Secreting Tumor Vaccine in Pancreatic Adenocarcinoma: A Phase II Trial of Safety, Efficacy, and Immune Activation. Annals of Surgery [Epub ahead of print] 2010 doi: 10.1097/SLA.0b013e3181fd271c. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jaffee EM, Hruban RH, Biedrzycki B, Laheru D, Schepers K, et al. Novel allogeneic granulocyte-macrophage colony-stimulating factor-secreting tumor vaccine for pancreatic cancer: a phase I trial of safety and immune activation. J Clin Oncol. 2001;19:145–156. doi: 10.1200/JCO.2001.19.1.145. [DOI] [PubMed] [Google Scholar]
- 6.Laheru D, Lutz E, Burke J, Biedrzycki B, Solt S, et al. Allogeneic granulocyte macrophage colony-stimulating factor-secreting tumor immunotherapy alone or in sequence with cyclophosphamide for metastatic pancreatic cancer: a pilot study of safety, feasibility, and immune activation. Clin Cancer Res. 2008;14:1455–1463. doi: 10.1158/1078-0432.CCR-07-0371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Thomas AM, Santarsiero LM, Lutz ER, Armstrong TD, Chen YC, et al. Mesothelin-specific CD8(+) T cell responses provide evidence of in vivo cross-priming by antigen-presenting cells in vaccinated pancreatic cancer patients. J Exp Med. 2004;200:297–306. doi: 10.1084/jem.20031435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hassan R, Ho M. Mesothelin targeted cancer immunotherapy. Eur J Cancer. 2008;44:46–53. doi: 10.1016/j.ejca.2007.08.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Esposito I, Penzel R, Chaib-Harrireche M, Barcena U, Bergmann F, et al. Tenascin C and annexin II expression in the process of pancreatic carcinogenesis. J Pathol. 2006;208:673–685. doi: 10.1002/path.1935. [DOI] [PubMed] [Google Scholar]
- 10.Sharma MC, Sharma M. The role of annexin II in angiogenesis and tumor progression: a potential therapeutic target. Curr Pharm Des. 2007;13:3568–3575. doi: 10.2174/138161207782794167. [DOI] [PubMed] [Google Scholar]
- 11.Kwon M, MacLeod TJ, Zhang Y, Waisman DM. S100A10, annexin A2, and annexin a2 heterotetramer as candidate plasminogen receptors. Front Biosci. 2005;10:300–325. doi: 10.2741/1529. [DOI] [PubMed] [Google Scholar]
- 12.Kim J, Hajjar KA. Annexin II: a plasminogen-plasminogen activator co-receptor. Front Biosci. 2002;7:d341–348. doi: 10.2741/kim. [DOI] [PubMed] [Google Scholar]
- 13.Rescher U, Gerke V. Annexins–unique membrane binding proteins with diverse functions. J Cell Sci. 2004;117:2631–2639. doi: 10.1242/jcs.01245. [DOI] [PubMed] [Google Scholar]
- 14.de Graauw M, Tijdens I, Smeets MB, Hensbergen PJ, Deelder AM, et al. Annexin A2 phosphorylation mediates cell scattering and branching morphogenesis via cofilin Activation. Mol Cell Biol. 2008;28:1029–1040. doi: 10.1128/MCB.01247-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rescher U, Ludwig C, Konietzko V, Kharitonenkov A, Gerke V. Tyrosine phosphorylation of annexin A2 regulates Rho-mediated actin rearrangement and cell adhesion. J Cell Sci. 2008;121:2177–2185. doi: 10.1242/jcs.028415. [DOI] [PubMed] [Google Scholar]
- 16.Deora AB, Kreitzer G, Jacovina AT, Hajjar KA. An annexin 2 phosphorylation switch mediates p11-dependent translocation of annexin 2 to the cell surface. J Biol Chem. 2004;279:43411–43418. doi: 10.1074/jbc.M408078200. [DOI] [PubMed] [Google Scholar]
- 17.Merrifield CJ, Rescher U, Almers W, Proust J, Gerke V, et al. Annexin 2 has an essential role in actin-based macropinocytic rocketing. Curr Biol. 2001;11:1136–1141. doi: 10.1016/s0960-9822(01)00321-9. [DOI] [PubMed] [Google Scholar]
- 18.Yang J, Weinberg RA. Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell. 2008;14:818–829. doi: 10.1016/j.devcel.2008.05.009. [DOI] [PubMed] [Google Scholar]
- 19.Krishnan S, Deora AB, Annes JP, Osoria J, Rifkin DB, et al. Annexin II-mediated plasmin generation activates TGF-beta3 during epithelial-mesenchymal transformation in the developing avian heart. Dev Biol. 2004;265:140–154. doi: 10.1016/j.ydbio.2003.08.026. [DOI] [PubMed] [Google Scholar]
- 20.Zhao S, Venkatasubbarao K, Lazor JW, Sperry J, Jin C, et al. Inhibition of STAT3 Tyr705 phosphorylation by Smad4 suppresses transforming growth factor beta-mediated invasion and metastasis in pancreatic cancer cells. Cancer Res. 2008;68:4221–4228. doi: 10.1158/0008-5472.CAN-07-5123. [DOI] [PubMed] [Google Scholar]
- 21.Gordon KJ, Dong M, Chislock EM, Fields TA, Blobe GC. Loss of type III transforming growth factor beta receptor expression increases motility and invasiveness associated with epithelial to mesenchymal transition during pancreatic cancer progression. Carcinogenesis. 2008;29:252–262. doi: 10.1093/carcin/bgm249. [DOI] [PubMed] [Google Scholar]
- 22.Morel E, Gruenberg J. Annexin A2 binding to endosomes and functions in endosomal transport are regulated by tyrosine 23 phosphorylation. J Biol Chem. 2009;284:1604–1611. doi: 10.1074/jbc.M806499200. [DOI] [PubMed] [Google Scholar]
- 23.Bianchi R, Pula G, Ceccarelli P, Giambanco I, Donato R. S-100 protein binds to annexin II and p11, the heavy and light chains of calpactin I. Biochim Biophys Acta. 1992;1160:67–75. doi: 10.1016/0167-4838(92)90039-g. [DOI] [PubMed] [Google Scholar]
- 24.Zhao WQ, Chen GH, Chen H, Pascale A, Ravindranath L, et al. Secretion of Annexin II via activation of insulin receptor and insulin-like growth factor receptor. J Biol Chem. 2003;278:4205–4215. doi: 10.1074/jbc.M210545200. [DOI] [PubMed] [Google Scholar]
- 25.Feldmann G, Dhara S, Fendrich V, Bedja D, Beaty R, et al. Blockade of hedgehog signaling inhibits pancreatic cancer invasion and metastases: a new paradigm for combination therapy in solid cancers. Cancer Res. 2007;67:2187–2196. doi: 10.1158/0008-5472.CAN-06-3281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Takano S, Togawa A, Yoshitomi H, Shida T, Kimura F, et al. Annexin II overexpression predicts rapid recurrence after surgery in pancreatic cancer patients undergoing gemcitabine-adjuvant chemotherapy. Ann Surg Oncol. 2008;15:3157–3168. doi: 10.1245/s10434-008-0061-5. [DOI] [PubMed] [Google Scholar]
- 27.Nedjadi T, Kitteringham N, Campbell F, Jenkins RE, Park BK, et al. S100A6 binds to annexin 2 in pancreatic cancer cells and promotes pancreatic cancer cell motility. Br J Cancer. 2009;101:1145–1154. doi: 10.1038/sj.bjc.6605289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hastie C, Masters JR, Moss SE, Naaby-Hansen S. Interferon-gamma reduces cell surface expression of annexin 2 and suppresses the invasive capacity of prostate cancer cells. J Biol Chem. 2008;283:12595–12603. doi: 10.1074/jbc.M800189200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mackenzie RP, McCollum AD. Novel agents for the treatment of adenocarcinoma of the pancreas. Expert Rev Anticancer Ther. 2009;9:1473–1485. doi: 10.1586/era.09.109. [DOI] [PubMed] [Google Scholar]
- 30.Hart IR, Fidler IJ. Role of organ selectivity in the determination of metastatic patterns of B16 melanoma. Cancer Res. 1980;40:2281–2287. [PubMed] [Google Scholar]
- 31.Jones S, Zhang X, Parsons DW, Lin JC, Leary RJ, et al. Core signaling pathways in human pancreatic cancers revealed by global genomic analyses. Science. 2008;321:1801–1806. doi: 10.1126/science.1164368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Jaffee EM, Schutte M, Gossett J, Morsberger LA, Adler AJ, et al. Development and characterization of a cytokine-secreting pancreatic adenocarcinoma vaccine from primary tumors for use in clinical trials. Cancer J Sci Am. 1998;4:194–203. [PubMed] [Google Scholar]
- 33.Wang B, Wei D, Crum VE, Richardson EL, Xiong HH, et al. A novel model system for studying the double-edged roles of nitric oxide production in pancreatic cancer growth and metastasis. Oncogene. 2003;22:1771–1782. doi: 10.1038/sj.onc.1206386. [DOI] [PubMed] [Google Scholar]
- 34.Salle V, Maziere JC, Smail A, Cevallos R, Maziere C, et al. Anti-annexin II antibodies in systemic autoimmune diseases and antiphospholipid syndrome. J Clin Immunol. 2008;28:291–297. doi: 10.1007/s10875-008-9188-1. [DOI] [PubMed] [Google Scholar]
- 35.Zhou X, Cui Y, Huang X, Yu Z, Thomas AM, et al. Lentivirus-mediated gene transfer and expression in established human tumor antigen-specific cytotoxic T cells and primary unstimulated T cells. Hum Gene Ther. 2003;14:1089–1105. doi: 10.1089/104303403322124800. [DOI] [PubMed] [Google Scholar]
- 36.Imamura H, Takao S, Aikou T. A modified invasion-3-(4,5-dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide assay for quantitating tumor cell invasion. Cancer Res. 1994;54:3620–3624. [PubMed] [Google Scholar]
- 37.Chen Y, Riley DJ, Chen PL, Lee WH. HEC, a novel nuclear protein rich in leucine heptad repeats specifically involved in mitosis. Mol Cell Biol. 1997;17:6049–6056. doi: 10.1128/mcb.17.10.6049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Abrams HD, Rohrschneider LR, Eisenman RN. Nuclear location of the putative transforming protein of avian myelocytomatosis virus. Cell. 1982;29:427–439. doi: 10.1016/0092-8674(82)90159-3. [DOI] [PubMed] [Google Scholar]
- 39.Jain A, Slansky JE, Matey LC, Allen HE, Pardoll DM, et al. Synergistic effect of a granulocyte-macrophage colony-stimulating factor-transduced tumor vaccine and systemic interleukin-2 in the treatment of murine colorectal cancer hepatic metastases. Ann Surg Oncol. 2003;10:810–820. doi: 10.1245/aso.2003.10.006. [DOI] [PubMed] [Google Scholar]
- 40.Yoshimura K, Jain A, Allen HE, Laird LS, Chia CY, et al. Selective targeting of antitumor immune responses with engineered live-attenuated Listeria monocytogenes. Cancer Res. 2006;66:1096–1104. doi: 10.1158/0008-5472.CAN-05-2307. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.