Journal of Surgical Research 91, 154 –158 (2000) doi:10.1006/jsre.2000.5932, available online at http://www.idealibrary.com on
A Novel Immunocytolytic Factor Secreted by Pancreatic Adenocarcinoma 1 Luz P. Angel, M.D.,* Celia M. Divino, M.D.,* ,2 Steven T. Brower, M.D.,* and Shu-Hsia Chen, Ph.D.† *Division of Surgical Oncology, Department of Surgery, and †Institute for Gene Therapy and Molecular Medicine, Mount Sinai School of Medicine, The Mount Sinai Medical Center, New York, New York 10029-6574 Presented at the Annual Meeting of the Association for Academic Surgery, Philadelphia, Pennsylvania, November 18 –20, 1999
Background. We have observed a putative immunocytolytic factor secreted by several pancreatic adenocarcinoma cell lines that mediates a potent cytolytic effect on lymphocytes. We sought to investigate its mechanism of action and determine if it is Fas ligand (FasL)-mediated. Materials and methods. Coincubation assays with murine splenocytes and supernatants from various species of pancreatic adenocarcinoma cell lines were performed. The mechanism of lymphocytic cell death was evaluated by the TUNEL assay. Pancreatic adenocarcinoma supernatant was coincubated with Fassensitive Jurkat cells and Western blotting for FasL was performed. Results. A marked reduction in the viability (%/ control) of target splenocytes was observed after incubation with the conditioned media from hamster PAN-1 (14.7%), PC 1.0 (21.7%), Taka-1 p70 (12.4%), Taka-1 p79 (7.6%), murine PANCO2 (16.1%), and human Capan-1 (14.0%) pancreatic adenocarcinoma cell lines. FACS analysis demonstrated significant lymphoid apoptosis at 16 h. The cytolytic effect appeared to be specific for lymphocytes and was not observed with the conditioned media of other tumor cells or normal pancreatic ductal cells. Pancreatic adenocarcinoma supernatant had no killing effect on Jurkat cells compared with control supernatant of TC-248 cells (87% vs 15%) and immunoblotting did not demonstrate soluble FasL. 1 Supported in part by grants from the National Cancer Institute (R29CA70337 and RO1-75175) to Shu-Hsia Chen, and from the Dean’s Research Incentive Fund, The Mount Sinai Medical Center, New York, to Celia Divino. 2 To whom correspondence should be addressed at the Division of Surgical Oncology, Department of Surgery, Mount Sinai School of Medicine, The Mount Sinai Medical Center, One Gustave L. Levy Place, Box 1259, New York, NY 10029-6574. Fax (212) 534-2654. E-mail: [email protected]
0022-4804/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Conclusions. These findings demonstrate that pancreatic adenocarcinoma cells secrete a potent cytolytic factor that induces apoptosis of lymphocytes and is not FasL-mediated. © 2000 Academic Press Key Words: pancreatic cancer; immunocytolytic factor; apoptosis. INTRODUCTION
The American Cancer Society estimates that 28,600 Americans will be diagnosed with cancer of the pancreas during 1999. An estimated 28,600 Americans will die of pancreatic cancer in 1999, making this type of cancer the fourth leading cause of cancer death among Americans . The overall average 5-year survival rates for patients diagnosed with pancreatic cancer are between 5 and 10% and their resectability rate is approximately 20%, with the majority of patients found to have advanced disease at the time of diagnosis [2, 3]. Immune dysfunction has long been observed in pancreatic cancer patients and this state of immunosuppression may contribute to the aggressiveness of this disease. Recently, there has been intensive research into pancreatic cancer’s immunologic effect on the host. Human pancreatic cancer cells may use the Fas ligand (FasL) “counterattack model” against the host immune defenses [4, 5]. Pancreatic tumors have been shown to express a membrane-bound functional FasL capable of inducing lymphoid apoptosis and a nonfunctional Fas receptor resistant to lymphocyte FasL-mediated apoptosis. Pancreatic cancer cell-derived secreted forms of CA-15.3 and soluble mucin 1 have been shown to inhibit cytolytic effector function and result in downregulation of interferon-␥ synthesis . In this work, we describe a novel immunocytolytic factor specifically secreted in the supernatant of several pancreatic adeno-
ANGEL ET AL.: PANCREAS CANCER SECRETES IMMUNOCYTOLYTIC FACTOR
carcinoma cell lines that induces apoptosis in lymphocytes.
TABLE 1 Biologic Effect of Pancreatic Adenocarcinoma Supernatant on Murine Splenocytes a
MATERIALS AND METHODS Cell lines. Established cell lines, which included PANCO2, a 3-methylcholantrene-induced carcinoma derived from female C57BL/6 mice , 293 (human embryonal renal cells), Jurkat (human acute T leukemia cells), CTLL-2 (murine T cells), Capan-1 or HTB 79 (human metastatic pancreatic adenocarcinoma), MCA-26 (murine colon adenocarcinoma), NC-37 (human lymphoblastoid cells), A20 (murine B lymphoma cells), and K-562 (human chronic myelogenous leukemia cells), were obtained from the American Type Culture Collection (Rockville, MD). The hamster pancreatic adenocarcinoma cell lines (PAN-1, PC1.0, Taka-1 p70, Taka-1 p79), derived from a BOP-induced hamster pancreatic adenocarcinoma tumor model, were provided by Dr. Terry Lawson and Dr. Parviz M. Pour (Eppley Institute, Omaha, NE) [8, 9]; MOD cells (murine breast adenocarcinoma) were provided by Dr. D. Medina (Baylor College of Medicine, Houston, TX); MPEC1 cells (simian virus 40 large T antigen transfected bovine pancreatic duct epithelial cells) were provided by Dr. Calvin U. Cotton ; and TC-248 cells (human Ewing sarcoma) were provided by Dr. Nicholas Mitsiades (NIH, Bethesda, MD) . Cells were maintained in either RPMI, Dulbecco’s modified Eagle’s medium (DMEM), or DMEM/Ham’s F12 supplemented with fetal bovine serum, 2 mM glutamine, 100 U/ml penicillin, and 100 g/ml streptomycin. Antibodies and cytokines. Recombinant murine IL-2 and purified anti-mouse CD3⑀ (0.5 mg/ml) were purchased from Pharmingen (San Diego, CA). Phycoerythrin (PE)-labeled anti-mouse CD45 rat immunoglobulin 2b (IgG2b) isotype and PE-labeled rat IgG2a, k isotype control clone R35-95 (Pharmingen), were used in the terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay. For Western blotting, anti-rabbit IgG-AP (Catalog No. sc-2007) alkaline phosphatase conjugate (200 g/0.5 ml) and FAS-L (c-178, Catalog No. sc-6237) rabbit polyclonal IgG (200 g/ml) were purchased from Santa Cruz Biotechnology. Isolation of murine splenocytes. Splenocytes obtained from naive BALB/C mice were prepared as described previously and seeded in 24-well multiwell dishes at 6 ⫻ 10 6 splenocytes per well in a total volume of 2 ml . Murine recombinant interleukin (IL)-2 was added to a final concentration of 20 U/ml. Splenocytes were washed with lymphocyte separation medium (Organon Teknika Corp., Durham, NC) prior to TUNEL assay and 24 h incubation with purified anti-mouse CD3⑀ (10 g/ml). Production of conditioned media. Cell lines were grown in 180-mm tissue culture dishes to 80% confluency. Their respective media were switched to serum-free media and incubation was continued for 48 h. Supernatants were collected and centrifuged for 20 min at 3000 rpm to remove cell debris. Conditioned medium of TC-248 cells was prepared as described. Cytotoxic bioassays. Target cells (splenocytes, CD3-activated splenocytes, MCA-26, CTLL-2, NC-37, A20) were coincubated with various conditioned media (293, MCA-26, MOD, MPEC1, PAN-1, PC 1.0, Taka-1 p70, Taka-1 p79, PANCO2, and Capan-1) at 1:5 v:v ratio at 37°C for 48 h. RPMI medium was used as a negative control for the conditioned medium. Viability of the splenocytes was determined using the trypan blue exclusion method and expressed as a percentage of control. For the cytotoxic assay against Jurkat cells, TC-248and PANCO2-conditioned media were concentrated 30-fold and then coincubated with a Jurkat cell suspension at a 1:1 (v:v) ratio and with a final Jurkat cell concentration of 1.25 ⫻ 10 5 cells/ml. After 48 h, viability of the Jurkat cells was determined using the trypan blue exclusion method. Apoptosis assay. Apoptosis was measured using the In Situ Cell Death Detection Kit (Boehringer-Mannheim, Indianapolis, IN). Splenocytes were incubated with PANCO2- and 293-conditioned me-
RPMI 293 MCA-26 MOD MPEC1 PAN-1 PC 1.0 Taka-1 p70 Taka-1 p79 PANCO2 Capan-1
100 93.6 96.5 99.8 98.5 14.68 21.68 12.38 7.58 16.07 14.0
0 0.13 0.13 0.20 0.20 0.004 0.021 0.007 0.0148 0.004 0.007
a Splenocytes were coincubated for 48 h with conditioned media from various species of pancreatic adenocarcinoma cells (PAN-1, PC 1.0, Taka-1 p70, Taka-1 p79, PANCO2, and Capan-1) and other cell types (293, MCA-26, MOD, MPEC1).
dia for 16 and 24 h. The cells were then analyzed with the FACScan (Becton Dickinson, San Jose, CA). Immunoblot analysis of FasL expression. For preparation of protein extracts, 2.0 ⫻ 10 7 K562 cells and a confluent PANCO2 180-mm cell culture plate were rinsed in phosphate-buffered saline (PBS) and lysed in RIPA buffer [1⫻ PBS, 1% Triton X-100, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (SDS)] with the following inhibitors: 10 mg/ml phenylmethylsulfonyl fluoride (PMSF, 10 l/ ml), aprotinin (30 l/ml), and 100 mM sodium orthovanadate (10 l/ml). The cells were further disrupted by passage through a 21gauge needle and an additional 10 l of 10 mg/ml PMSF. Equal amounts (70 g) of protein (measured with the Bio-Rad Protein Assay Reagent, Bio-Rad Laboratories, Hercules, CA) were loaded on a 12% SDS–polyacrylamide gel followed by transfer onto a polyvinyl difluoride (PVDF) membrane. The FasL protein was detected using an anti-FasL polyclonal rabbit IgG and a secondary anti-rabbit IgG alkaline phosphatase and a BCIP/NBT chromogenic solution (Catalog No. B-1911, Sigma). For detection of the Fas ligand in the supernatant, conditioned medium of the PANCO2 tumor cells was concentrated 700-fold by centrifugation with the Speed Vac-SC110 (Savant). Western blots were performed as previously described.
Cytotoxic Effect of Pancreatic Adenocarcinoma Cell Supernatants on Splenocytes When 48-h conditioned media from various species of pancreatic adenocarcinoma cell lines were added to freshly isolated splenocytes, a marked cytotoxic effect was observed with all species of pancreatic adenocarcinoma cell lines (Table 1). Cell viability, expressed as a percentage of control, was reduced markedly after incubation with hamster PAN-1 (15%, P ⬍ 0.001), PC1.0 (22%, P ⬍ 0.001), Taka-1 p70 (12%, P ⬍ 0.001), Taka-1 p79 (8%, P ⬍ 0.001), murine PANCO2 (16%, P ⬍ 0.001), and human Capan-1 (14%, P ⬍ 0.001). Supernatants from other tumor cell lines including MCA-26 (97%, P ⫽ 0.022), MOD (100%, P ⫽ 0.855), and 293 (94%, P ⫽ 0.07), as well as super-
JOURNAL OF SURGICAL RESEARCH: VOL. 91, NO. 2, JUNE 15, 2000
FIG. 1. Flow cytometric analysis of PANCO2 pancreatic adenocarcinoma supernatant-induced apoptosis in murine splenocytes at 16 h. Supernatant from 293 cells was used as control.
natant from a pure culture of immortalized bovine pancreatic ductal cells (MPEC1) (99.7%, P ⫽ 0.095), failed to have a significant effect on the splenocytes. To determine if this cytotoxic effect is specific for lymphocytes, pancreatic adenocarcinoma supernatant was coincubated with different target cells. Significant reduction in cell viability, expressed as a percentage of control, was observed with all lymphocytic cell types including CTLL-2 (50%, P ⫽ 0.008), NC-37 (41%, P ⬍ 0.001), and A20 (39%, P ⫽ 0.010) cell lines after incubation with pancreatic adenocarcinoma-conditioned medium. This effect also was demonstrable in a primary culture of CD3activated lymphocytes (42%, P ⫽ 0.002) from murine splenocytes but not with MCA-26 cells (98%, P ⫽ 0.423) or human embyronal kidney cells (100% P ⫽ 0).
Cytotoxic Effect of PANCO2-Conditioned Medium is not FasL-Mediated To determine whether the biologic effect of pancreatic adenocarcinoma supernatant is FasL-mediated, the pancreatic adenocarcinoma supernatants were tested against Jurkat cells which are known to constitutively express soluble FasL and are sensitive to FasL-induced apoptosis . Conditioned medium of TC-248 cells, which is a human Ewing sarcoma cell line known to contain soluble FasL, was used as a control.
Lymphocyte Cell Death Is Mediated by Apoptosis After 16 h of incubation with PANCO2-conditioned medium, flow cytometric analysis showed 9% of the total lymphocytes had undergone apoptosis, compared with 1% with 293-conditioned medium (Fig. 1). At longer times, greater lymphoid cell apoptosis with the PANCO2-conditioned medium was observed compared with control. Overall, flow cytometric studies illustrated greater nuclear staining of apoptotic cells after incubation with pancreatic adenocarcinoma supernatant.
FIG. 2. Effect of PANCO2- and TC-248-conditioned media on Fas-sensitive Jurkat cells. Media were concentrated 30⫻ before incubation. Bars represent means ⫾ SD.
ANGEL ET AL.: PANCREAS CANCER SECRETES IMMUNOCYTOLYTIC FACTOR
FIG. 3. Western blot analysis of pancreatic adenocarcinoma supernatant for FasL.
After 48 h, pancreatic adenocarcinoma supernatant did not have a significant effect on the viability of Jurkat cells compared with control supernatant of TC248 cells (87%, P ⫽ 0.237, vs 15% P ⬍ 0.001). These results strongly suggest that the immunocytolytic factor in the pancreatic adenocarcinoma supernatants is not FasL (Fig. 2). Immunoblotting for FasL Western blot analysis for FasL expression in pancreatic adenocarcinoma supernatant was performed. K-562, a human chronic myelogenous leukemia cell line, which is known to contain the 40-kDa FasL, was used as control (Fig. 3). Although the membrane-bound 40,000-kDa FasL was present in the pancreatic adenocarcinoma cell lysate, the 27,000-kDa soluble FasL was undetectable in the supernatant. These data are consistent with other recent studies in the literature demonstrating the absence of the soluble 27,000-kDa FasL in the supernatant of pancreatic adenocarcinoma cells [4, 5]. DISCUSSION
Immune dysfunction has long been observed in cancer patients and this has prompted investigations into tumor mechanisms of immunosuppression and evasion. Tumors have been shown to elaborate numerous factors such as transforming growth factor ␤ (TGF-␤), protein p15E, interleukin-10, prostaglandin E 2, mucins, adhesion molecules, growth factors, chemotactic molecules, and proteases that can suppress the host immune system [14, 15]. In renal cell carcinoma patients, tumor-derived soluble mucins can result in altered T-cell function by impairing activation of transcription factor NF-B . Fas ligand, expressed by several human cancers such as esophageal and head and neck cancer, promotes apoptosis of tumor-
infiltrating lymphocytes [17, 18]. Expression of HLA-G on the surface of melanoma cells enables them to escape cytotoxic lysis by natural killer cells . Downregulation of tumor antigen and costimulatory molecules, loss of functional ␤ 2-microglobulin, and deletion of specific ␣-chain alleles result in poor T-cell-mediated rejection [20, 21]. Advanced stages of cancer in patients have been correlated with tumor-induced immunosuppression. Whiteside demonstrated that the decreased expression of the chain in CD4 and CD8 T cells and natural killer cells in patients with stage III or IV squamous cell head and neck carcinoma correlated with a shorter 5-year survival and a poorer response to biologic therapies . Investigators have even demonstrated the reversibility of immune dysfunction by removal of the tumor, suggesting that some factor related to the tumor causes immunosuppression . Pancreatic cancer patients are especially immunosuppressed and recent molecular elucidations have been made into the causative mechanism. Human pancreatic tumors have been shown to express a membrane-bound functional FasL capable of inducing lymphoid apoptosis and a nonfunctional Fas receptor resistant to lymphocyte FasL-mediated apoptosis [4, 5]. Pancreatic tumor-derived secreted forms of CA-15.3 and soluble mucin 1 have been shown to inhibit cytolytic effector function and result in downregulation of interferon ␥ synthesis . We have shown that pancreatic adenocarcinoma cells secrete a distinctive immunocytolytic factor which seems to be specific for lymphocytes and does not correspond to any of the known immunosuppressive factors. This factor is unique to pancreatic adenocarcinoma cells since supernatants from other tumor cell lines or normal pancreatic ductal cells failed to exhibit the same cytolytic effect. The immunocytolytic effect was present among the different species of pancreatic cancer cells (human, hamster, and murine) and the mechanism of lymphocytic cell death is apoptosis. Pancreatic adenocarcinoma supernatant can be diluted up to fivefold with preservation of its immunocytolytic effect. Coincubation studies with Jurkat cells and Western blot analysis of the pancreatic adenocarcinoma supernatant excluded the soluble FasL as the immunocytolytic factor. Isolation and further characterization of this immunocytolytic factor will have numerous implications in pancreatic cancer immunotherapy and immunomodulatory gene therapy. It will facilitate the earlier detection of pancreatic cancer and measurement of tumor progression and response to therapies. ACKNOWLEDGMENTS We thank Dr. Simon Hall for his helpful discussion, Dr. Cotton and Dr. Mitsiades for their generous gifts of cells, Mr. Khiem Pham-
JOURNAL OF SURGICAL RESEARCH: VOL. 91, NO. 2, JUNE 15, 2000
Nguyen for his technical help, and Dr. Morris D. Kerstein and Ms. Gae O. Decker-Garrad for editorial assistance.
REFERENCES 1. 2. 3.
American Cancer Society Facts and Figures 1999. Gold, E. B. Epidemiology of and risk factors for pancreatic cancer. Surg. Clin. North Am. 75: 819, 1995. Conlon, K. C., Klimstra, D. S., and Brennan, M. F. Long term survival after curative resection for pancreatic ductal adenocarcinoma. Ann. Surg. 223: 273, 1996. Ungefroren, H., Voss, M., Jansen, M., Roeder, C., HenneBrunds, D., Kremer, B., and Kalthoff, H. Human pancreatic adenocarcinomas express Fas and Fas ligand yet are resistant to Fas-mediated apoptosis. Cancer Res. 58: 1741, 1998. Bernstorff, W., Spanjaard, R. A., Chan, A. K., Lockhart, D. C., Sadanaga, N., Wood, I., Peiper, M., Goedegebuure, P. S., and Eberlein, T. J. Pancreatic cancer cells can evade immune surveillance via nonfunctional Fas (APO-1/CD95) receptors and aberrant expression of functional Fas ligand. Surgery 125: 73, 1999. Plate, J. M. D., Peterson, K. S., and Harris, J. E. Cellular immunity in pancreatic cancer and mechanisms involved in its regulation. Proc. Am. Assoc. Cancer Res. 40: 77, 1999. Corbett, T. H., Roberts, B. J., Leopold, W. R., Peckham, J. C., Wilkoff, L. J., Griswold, D. P. J., and Schabel, F. M. J. Induction and chemotherapeutic response of two transplantable ductal adenocarcinomas of the pancreas in C57BL/6 mice. Cancer Res. 44: 717, 1984. Pour, P., Kruger, F. W., Althoff, J., Cardesa, A., and Mohr, U. Cancer of the pancreas induced in the Syrian golden hamster. Am. J. Pathol. 76: 349, 1974. Egami, H., Takiyama, Y., Cano, M., Houser, W. H., and Pour, P. M. Establishment of hamster pancreatic ductal carcinoma cell line (PC-1) producing blood group antigens. Carcinogenesis 10: 861, 1989. Cotton, C. U., and Al-Nakkash, L. Isolation and culture of bovine pancreatic duct epithelial cells. Am. J. Physiol. 272: G1328, 1997. Mitsiades, N., Poulaki, V., Kotoula, V., Leone, A., and Tsokos, M. Fas ligand is present in tumors of the Ewing’s sarcoma family and is cleaved into a soluble form by a metalloproteinase. Am. J. Pathol. 153: 1947, 1998. Pham-Nguyen, K. B., Yang, W., Saxena, R., Thung, S. N., Woo,
S. L. C., and Chen, S. Role of NK and T cells in IL-12 induced anti-tumor response against hepatic colon carcinoma. Int. J. Cancer 81: 813, 1999. Martinez-Lorenzo, M. J., Alava, M. A., Anel, A., Pineiro, A., and Naval, J. Release of preformed Fas ligand in soluble form is the major factor for activation-induced death of Jurkat T cells. Immunology 89: 511, 1996. Elgert, K. D., Alleva, D. G., and Mullins, D. W. Tumor-induced immune dysfunction: The macrophage connection. J. Leukoc. Biol. 64: 275, 1998. Botti, C., Seregni, E., Ferrari, L., Martinetti, A., and Bombardieri, E. Immunosuppressive factors: Role in cancer development and progression. Int. J. Biol. Markers 13: 51, 1998. Ling, W., Rayman, P., Uzzo, R., Clark, P., Kim, H. J., Tubbs, R., Novick, A., Bukowski, R., Hamilton, T., and Finke, J. Impaired activation of NF-B in T cells from a subset of renal cell carcinoma patients is mediated by inhibition of phosphorylation and degradation of the inhibitor, I-B␣. Blood 92: 1334, 1998. Gastman, B. R., Atarshi, Y., Reichert, T. E., Saito, T., Balkir, L., Rabinowich, H., and Whiteside, T. L. Fas ligand is expressed on human squamous cell carcinomas of the head and neck, and it promotes apoptosis of T lymphocytes. Cancer Res. 59: 5356, 1999. O’Connell, J., Bennett, M. W., O’Sullivan, G. C., Collins, J. K., and Shanahan, F. Resistance to Fas (APO-1/CD95)-mediated apoptosis and expression of Fas ligand in esophageal cancer: The Fas counterattack. Dis. Esophagus 12: 83, 1999. Adrian Cabestre, F., Moreau, P., Riteau, B., Ibrahim, E. C., Le Danff, C., Dausset, J., Rouas-Freiss, N., Carosella, E. D., and Paul, P. HLA-G expression in human melanoma cells: Protection from the NK cytolysis. J. Reprod. Immunol. 43: 183, 1999. Restifo, N. P., Kawakami, Y., Marincola, F., Shamamian, P., Taggarse, A., Esquivel, F., and Rosenberg, S. A. Molecular mechanisms used by tumors to escape immune recognition: Immunotherapy and the cell biology of major histocompatability complex class I. J. Immunother. 14: 182, 1993. Zheng, P., Sarma, S., Guo, Y., and Liu, Y. Two mechanisms for tumor evasion of preexisting cytotoxic T-cell responses: Lessons from recurrent tumors. Cancer Res. 59: 3461, 1999. Whiteside, T. L. Signaling defects in T lymphocytes of patients with malignancy. Cancer Immunol. Immunother. 48: 346, 1999. Salvadori, S., Martinelli, G., and Zier, K. Resection of solid tumors reverses T cell defects and restores protective immunity. J. Immunol. 164: 2214, 2000.