Matrix Metalloproteinase (MMP) Inhibition Selectively Decreases Type II MMP Activity in a Murine Model of Pancreatic Cancer

Matrix Metalloproteinase (MMP) Inhibition Selectively Decreases Type II MMP Activity in a Murine Model of Pancreatic Cancer

Journal of Surgical Research 81, 65– 68 (1999) Article ID jsre.1998.5447, available online at http://www.idealibrary.com on Matrix Metalloproteinase ...

81KB Sizes 0 Downloads 16 Views

Journal of Surgical Research 81, 65– 68 (1999) Article ID jsre.1998.5447, available online at http://www.idealibrary.com on

Matrix Metalloproteinase (MMP) Inhibition Selectively Decreases Type II MMP Activity in a Murine Model of Pancreatic Cancer Emmanuel E. Zervos, M.D., Alexis E. Shafii, B.S., and Alexander S. Rosemurgy, M.D. Department of Surgery, University of South Florida, and James A. Haley Veteran’s Administration Medical Center, Tampa, Florida Presented at the 22nd Annual Symposium of the Association of Veterans Administration Surgeons, Baltimore, Maryland, April 26 –28, 1998

INTRODUCTION Background. Basement membrane degradation is a critical component of tumor invasion and metastasis that is facilitated by the family of enzymes known as matrix metalloproteinases (MMPs). MMP-2 and MMP-9 are two subtypes that have specifically been identified in tumors of gastrointestinal origin. We have previously shown that broad inhibition of these enzymes with the MMP inhibitor BB-94 improves survival in a murine model of pancreatic cancer. The purpose of this study was to determine MMP-2 and MMP-9 activity in orthotopic tumors from mice treated with and without BB-94. Methods. Ten million cells of a moderately differentiated pancreatic cancer cell line (HPAC) were implanted orthotopically into Balb/c nu/nu mice. The mice were treated with BB-94 or vehicle control for 70 days or until death. At necropsy, tumors were harvested, total protein was extracted, and MMPs were purified from 400 mg of crude protein extract by gelatin–Sepharose affinity chromatography. Relative enzyme levels and activity of MMP-2 and MMP-9 were determined by Western blot and gelatin zymography. Results. Tumors from treated animals were significantly smaller than those from nontreated animals. MMP-2 was present in greater amounts in both treated and nontreated animals than MMP-9. Active MMP-2 was present in both groups but significantly decreased in animals treated with BB-94. Active MMP-9 was absent in both groups, whereas levels of latent MMP-9 appeared lower than those of MMP-2 in all samples. Conclusions. Activated MMP-2 and not MMP-9 in HPAC cells grown in nude mice suggests that this MMP subtype is more critical in the phenotypic behavior of such tumors. Furthermore, attenuated levels of active MMP-2 in animals treated with the enzyme inhibitor BB-94 suggest that previously observed improvements in survival correlate with the level of MMP-2 activity. © 1999 Academic Press Key Words: pancreas; neoplasm; matrix metalloproteinases.

The poor prognosis associated with pancreatic cancer is a recognized consequence of the advanced stage of the disease at the time of diagnosis [1, 2]. Surgical resection, the only reasonable chance for cure, is precluded in the majority of patients by local invasion and contiguous spread to surrounding vital vascular structures, biologic activity that is presumed to be facilitated by the family of enzymes known as matrix metalloproteinases (MMPs). The critical role of such enzymes in tumor growth and spread was first suspected after relative increases in their levels were noted in cancerous as compared with normal tissue [3–5]. Through enzymatic degradation of type IV collagen and other components of the basement membrane, tumors expressing high levels of MMPs have demonstrated increased invasion, a higher likelihood to metastasize, and, ultimately, increased lethality [6 –9]. Currently, 14 separate metalloproteinases have been described which are categorized into subgroups based on structural similarities and substrate specificity. Type IV collagenases include MMP-2 and MMP-9 and have been implicated in cancers of gastrointestinal origin presumably because of their ability to degrade type IV collagen [10 –12]. These enzymes are secreted by tumor cells as proenzymes or in latent form and then are activated by cleavage of an NH 2-terminal peptide that is responsible for maintaining the latency of the enzyme [13]. These complex series of interactions leading to cleavage and subsequent activation of MMP-2 and MMP-9 are not fully understood but are also believed to act as a secondary regulatory system determining the relative invasiveness of tumors expressing these enzymes. Further evidence supporting the critical role of MMPs in cancer is derived from the observation that inhibition of such enzymes, either through pharmacologic blockade or genetic or biologic manipulation of their naturally occurring antagonists (TIMP-1 and -2), has been shown to decrease the tumorigenicity of human cancers in animal models [14 – 17].

65

0022-4804/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

66

JOURNAL OF SURGICAL RESEARCH: VOL. 81, NO. 1, JANUARY 1999

We have previously shown that the synthetic inhibitor of metalloproteinases, BB-94 (British Biotech, Oxford, Cowley, U.K.) increases survival, decreases tumor burden, and limits metastases of the moderately differentiated pancreatic cancer cell line HPAC in an orthotopic murine model [17]. While this study documented the efficacy of BB-94 in the setting of pancreatic cancer, we did not determine the breadth or the magnitude of BB-94’s effect on MMPs expressed by such tumors. The purpose of the current study, therefore, was twofold. First, we sought to characterize the type IV collagenase/gelatinase profile of HPAC cells grown in vivo and, second, to quantify enzyme activity in tumors harvested from animals treated with BB-94 and those treated with vehicle control. Our hypothesis in undertaking this study was that MMP-2 and MMP-9 would be equally expressed in treated and nontreated animals but their relative activity would be attenuated in animals treated with the enzyme inhibitor BB-94. METHODS Xenografts. Ten million moderately differentiated human pancreatic adenocarcinoma (HPAC) cells (ATCC CRL-2119) [18] were surgically implanted into the pancreatic head of Balb-c nu/nu mice (Life Sciences, Inc., St. Petersburg, FL) as previously described [17]. One week after implantation, mice were randomized to receive daily (0.1 cc) intraperitoneal injections of either BB-94 (30 mg/kg) or vehicle control [phosphate-buffered saline (PBS), pH 7.4, containing 0.01% Tween 80]. At the time of death or 70 days, the mice were sacrificed and orthotopic tumors were harvested. The tumors were weighed and processed for subsequent protein analysis. Protein extraction. One hundred milligrams of tumor tissue from each mouse was individually homogenized at 4°C in 0.1 M Tris buffer (pH 8.1) containing protease inhibitors [aprotinin 100 mg/ml, phenylmethylsulfonyl fluoride (PMSF) 200 mM, leupeptin 10 mg/ml; Sigma, St. Louis, MO]. The homegenate was sonicated for 1 min at 4°C and then spun in an ultracentrifuge at 100,000g for 45 min at 4°C. The supernatant was collected and protein concentrations were determined with the Bio-Rad Protein Assay Reagent (Bio-Rad, Hercules, CA). Affinity chromatography. Gelatin–Sepharose (Pharmacia Biotech Inc, Sweden) was washed three times with equilibration buffer (pH 7.5: 50 mM Tris, 150 mM NaCl, 5mM CaCl 2, 0.02% Tween 20, 0.07% Brij 35, 10 mM EDTA). Twenty microliters of equilibrated gelatin–Sepharose was added to 1600 mg of tumor protein extracts and diluted to a final volume of 1 ml with PBS. Samples were then placed on an end-over-end shaker overnight at 4°C to allow the binding of gelatinases to the gelatin–Sepharose. Following overnight shaking, nonspecifically bound proteins were washed from gelatin– Sepharose with salted equilibration buffer (50 mM Tris, 200 mM NaCl, 5mM CaCl 2, 0.02% Tween 20, 0.07% Brij 35, 10 mM EDTA) and 15 ml of nonreducing Laemmli was added to the gelatin– Sepharose-purified samples. Western blot. To determine whether the human xenografts produced the metalloproteinases of interest, Western blots were probed for MMP-2 and MMP-9. Affinity-purified samples as described above, were electrophoretically separated on an 8% SDS–polyacrylamide gel. The separated proteins were then transferred to a nitrocellulose membrane (Amersham, New Chalfont, U.K.) using the Mini-Genie blotting system (Idea Scientific, Minneapolis, MN). The membrane was blocked in 5% milk in TBS-T 0.1% for 1 h at room temperature and then incubated with a 1:500 dilution of antibody against human MMP-2 (Biogenesis, Poole, U.K.) and a 1:750 dilution of antibody against human MMP-9 (Biogenesis) in TBS at room temperature for 2 h. The membrane was then washed in TBS-T 0.5% for 1 h at room temperature, and probed with anti-rabbit IgG conjugated with

FIG. 1. Western blot of purified protein samples in tumors harvested from animals treated with vehicle control (lane 1), the MMPI, BB-94 (lane 2), and HT-1080 cells (known producers of MMP, lane 3). Purified proteins were blotted with the M r 72,000 type IV collagenase antibody and the M r 92,000 type IV collagenase antibody. HPAC tumors secreted similar levels of MMP, regardless of treatment with BB-94; MMP-2 was more pronounced than MMP-9. horseradish peroxidase (Bio-Rad) 1:5000 in TBS for 1 h at room temperature. Following the secondary antibody incubation, the membrane was washed in TBS-T 0.5% for 1 h. Immunoblots were detected by enhanced chemiluminescence (Amersham). The conditioned medium from HT 1080 cells (ATCC CCL-121), known producers of MMP-2 and MMP-9, served as the positive control [19]. Gelatin zymography. The purified samples were diluted 1:4 and electrophoretically separated on 8% SDS–polyacrylamide gel impregnated with gelatin (1 mg/ml). After incubation, the gels were rinsed twice in 2.5% Triton X-100 and three times in double-distilled H 2). The gels were then incubated at 37°C for 4 h in 200 mM NaCl, 10 mM CaCl 2, 0.07% Brij 35, 50 mM Tris–HCl buffer, pH 7.4. The gels were stained with 0.05% Coomassie brilliant blue and destained in 10% acetic acid in H 2O. Gelatinolytic enzymes were detected as transparent bands on the background of the Coomassie blue-stained gel. Relative enzyme activity was quantified by densitometric analysis of the negatively stained bands. Latent MMP-2 (M r 72,000), activated MMP-2 (M r 62,000), latent MMP-9 (M r 92,000), and activated MMP-9 (M r 86,000) were identified by comparison with known gelatinolytic activities from the conditioned medium of HT 1080 cells. Data and statistical analysis. Tumor weights for the selected samples were compared using Student’s t test. Relative band densities were determined using the UVP GDS 8000 gel documentation system (UltraViolet Products, Upland, CA). Peak areas were determined using GDS Image Analysis Software (UV Products) and compared using Student’s t test.

RESULTS

Tumor Isolates Orthotopic tumors suitable for metalloproteinase analysis were isolated at necropsy from eight animals treated with BB-94 and eight animals treated with vehicle control. Average tumor weights in treated animals were significantly less than those of controls. (0.41 1 0.09 g vs 0.71 1 0.09 g, P 5 0.02). Western Blot MMP-2 and MMP-9 were present in all tumor samples. MMP-2 appeared to be more uniformly secreted and conserved than MMP-9 in both treated and nontreated animals. Treatment with BB-94 had no demonstrable effect on the expression of MMP-2 or MMP-9 as detected by Western blot (Fig. 1).

ZERVOS, SHAFII, AND ROSEMURGY: METALLOPROTEINASE ACTIVITY IN PANCREATIC CANCER

67

FIG. 2. Representative zymogram of purified MMP isolated from internal control (HT-1080 cells, lane 1) expressing latent MMP-9 (92 kDa), activated MMP-9 (86 kDa), latent MMP-2 (72 kDa), and active MMP-2 (62 kDa). Tumors harvested from control animals are represented in lanes 2–5 and demonstrate latent MMP-9 only and both latent and active MMP-2. Tumors harvested from animals treated with BB-94 are shown in lanes 6 –9 and demonstrate minimal MMP-9 (latent or active) and significantly attenuated active MMP-2.

Gelatin Zymography MMP-9. Relative levels of latent or pro-MMP-9 were not statistically different in tumors from treated versus nontreated animals. While latent MMP-9 was constitutively expressed by all tumors from both treated and untreated mice, relative levels were clearly less than those of latent MMP-2. Activated MMP-9 was absent in eight of eight tumors from mice treated with BB-94 and only one nontreated animal had a tumor with barely detectable levels of active MMP-9 (Figs. 2 and 3). MMP-2. Latent MMP-2 and active MMP-2 were expressed to a significantly higher degree in both treated and nontreated mice as compared with MMP-9. Latent MMP-2 was universally secreted in each tumor isolate; treatment with BB-94 had no effect on levels of latent MMP-2. Each tumor isolate also demonstrated active MMP-2, though these levels were lower than

FIG. 3. Relative band densities of negatively stained bands shown in Fig. 2 for MMP-9. No difference in latent MMP-9 could be detected between treated and nontreated animals, while the presence of active MMP-9 was detectable in only one tumor coming from a nontreated animal.

FIG. 4. Relative band densities of negatively stained bands shown in Fig. 2 for MMP-2. Both latent MMP-2 and active MMP-2 were secreted in tumors from both treated and nontreated animals to a higher degree than MMP-9. No differences in latent MMP-2 levels could be detected between treatment and no treatment, but BB-94 was associated with significantly (*P , 0.05) reduced levels of activated MMP-2.

those of the latent form in each treatment group. Levels of active MMP-2 in tumors from animals treated with BB-94 were significantly lower than those of active MMP-2 levels in tumors from animals given vehicle alone (Figs. 2 and 4). DISCUSSION

The exact mechanism through which pancreatic cancers become locally invasive and ultimately fatal remains a point of speculation. Compelling evidence implicating MMPs in this process exists, though the specific MMP subtypes and the series of biochemical and physiologic events leading to their activation are still poorly understood. These events are thought to be critical in rendering such enzymes capable of performing their reputed destructive processes. This study was conceived to determine the role of two MMP subtypes that are known to facilitate tumor invasion and spread with other human cancers, using an established orthotopic animal model of pancreatic cancer. We further sought to determine what effect a known inhibitor of these enzymes has an observed MMP activity and to what degree this impacts clinically relevant endpoints. In this study, we have shown that both MMP-2 and MMP-9 are present in human pancreatic tumors grown orthotopically in athymic mice. Though both enzymes are present, we surmise from the data that MMP-2 plays a more critical role in determining the phenotypic behavior of HPAC cells than MMP-9. Western blot analysis identifies a pronounced and more consistent presence of MMP-2 than MMP-9 and zymograms examining their relative activity demonstrate little or no active MMP-9. In contradistinction, active MMP-2

68

JOURNAL OF SURGICAL RESEARCH: VOL. 81, NO. 1, JANUARY 1999

was found in all tumor isolates. These findings are consistent with those of other investigators [20] and suggest that MMP inhibition is potentially a clinically relevant therapeutic strategy. BB-94, the MMP inhibitor used in this study, is known, from previous work done in this laboratory, to be associated with an improved outcome and less invasive tumor. In light of the current findings, the salutary effects of such inhibition may be achieved through more than one mechanism. First, BB-94 may decrease absolute levels of active enzymes. Second, the activation of the latent to active form of MMP, a critical step to ultimate degradation of membrane components, is also associated with MMP inhibition. Inhibition of the enzymes and inhibition of their activation may serve to regulate their activity through more than one mechanism. Levels of active MMP-2 were significantly reduced in animals treated with BB-94 while active MMP-9 was demonstrable in only one tumor sample in a nontreated animal. It appears that this attenuation is probably the result of direct antagonism and/or inhibition of activation (rather than as a secondary effect from some undefined growth-inhibiting mechanism resulting in less tumor volume) simply because the latent forms of MMP-2 do not seem to be affected by BB-94. Multiple factors contribute to the lethality of pancreatic cancer. Any therapy aimed at slowing or containing the growth of a neoplastic process originating in this anatomically vexing location has significant potential to improve quality of life and possibly survival. Broad-spectrum matrix metalloproteinase inhibition shows promise in both animal and early clinical trials. This study provides additional data, specifically regarding MMP-2 and its activation, that may help to focus future laboratory and clinical investigations. Focused inhibition or genetic deletion of this enzyme in particular may serve to achieve a more potent and prolonged effect while maintaining the salutary effects of the remaining spectrum of naturally occurring MMPs.

6.

Tryggvason, K., Hoyhtya, M., and Pyke, C. Type IV collagenases in invasive tumors. Breast Cancer Res. Treat. 24: 209, 1993. (Review).

7.

Waxman, J. The architecture of cancer: The tissue metalloproteinases and their inhibitors Br. Med. J. 305: 1306, 1992. (Review).

8.

Van der Stappen, J. W., Hendriks, T., and Wobbes, T. Correlation between collagenolytic activity and grade of histologic differentiation in colorectal tumors. Int. J. Cancer 45: 1071, 1990.

9.

Davies, B., Miles, D. W., Happerfield, M. L. C., Naylor, M. S., Bobrow, L. G., Rubens, R. D., and Balkwill, F. R. Activity of type IV collagenases in benign and malignant breast disease. Br. J. Cancer 67: 1126, 1993.

10.

Levy, J. T., Cioce, V., Sobel, M. E., Garbisa, S., Grigioni, W. F., Liotta, L. A., and Stetler-Stevenson, W. G. Increased expression of the Mr 72,000 type IV collagenase in human colonic adenocarcinoma. Cancer Res. 51: 439, 1991.

11.

Koshiba, T., Wada, M., Fujimot, K., Lee, J. U., Kato, M., Doi, R., Shimada, Y., Hosotani, R., Arii, S., and Imamura, M. Activation of MMP-2 and MMP-9 in human pancreatic cancer tissues and human pancreatic cancer cell lines. Gastroenterology 110: A545, 1996.

12.

Liabakk, N. B., Talbot, I., Smith, R. A., Wilkinson, K., and Balkwill, F. Matrix metalloprotease 2 (MMP-2) and matrix metalloprotease 9 (MMP-9) type IV collagenases in colorectal cancer. Cancer Res. 56: 190, 1996.

13.

DeClerck, Y. A., and Imren, S. Protease inhibitors: Role and potential therapeutic use in human cancer. Eur. J. Cancer 14: 2170, 1994.

14.

Imren, S., Kohn, D. B., Shimada, H., Blavier, L., and DeClerck, Y. A. Overexpression of tissue inhibitor of metalloproteinases-2 retroviral-mediated gene transfer in vivo inhibits tumor growth and invasion. Cancer Res. 56: 2891, 1996.

15.

Wang, X., Fu, X., Brown, P. D., Crimmin, M. J., and Hoffman, R. M. Matrix metalloproteinase inhibitor BB-94 (batimastat) inhibits colon tumor growth and spread in a patient like orthotopic model in nude mice. Cancer Res. 54: 4726, 1994.

16.

Eccles, S. A., Box, G. M., Court, W. J., Bone, E. A., Thomas, W., and Brown, P. D. Control of lymphatic and hematogenous metastasis of a rat mammary carcinoma by the matrix metalloproteinase inhibitor batimastat (BB-94). Cancer Res. 56: 2815, 1996.

17.

Zervos, E. E., Norman, J. G., Gower, W. R., Franz, M. G., and Rosemurgy, A. S. Matrix metalloproteinase inhibition attenuates human pancreatic cancer growth in vitro and decreases mortality and tumorigenesis in vivo. J. Surg. Res. 69: 367, 1997.

18.

Gower, W. R., Risch, R. M., Godellas, C. M., and Fabri, P. J. HPAC, a new human glucocorticoid-sensitive pancreatic ductal adenocarcinoma cell line. In Vitro Cell Dev. Biol. 30A: 151, 1994.

19.

Brown, P. D., Bloxidge, R. E., Stewart, N. S., Gatter, K. C., and Carmichael, J. Association between expression of activated 72 kD gelatinase and tumor spread in non-small cell lung carcinoma. J. Natl. Cancer Inst. 85: 574, 1993.

20.

Hosotani, R., Koshiba, M., Miyamoto, K., Fujimoto, J. U., Lee, Doi, R., Arii, S., Tamaki, K., Tanzawa, K., and Imamura, M. C l i n i c a l a n d t h e r a p e u t i c s i g n i fi c a n c e o f m a t r i x metalloproteinase-2 (MMP-2) activation in pancreatic cancer. In Proceedings of the 31st Annual Meeting of the Pancreas Club, Inc., May 1997, p. 44.

REFERENCES 1.

2. 3.

4. 5.

Yeo, C. J., Cameron, J. L., and Lillemoe, K. D. Pancreaticoduodenectomy for cancer of the head of the pancreas. Ann. Surg. 221: 721, 1995. Lillemoe, K. D. Current management of pancreatic carcinoma. Ann. Surg. 221: 133, 1995. Stetler-Stevenson, W. G., Aznavoorian, S., and Liotta, L. A. Tumor cell interactions with the extracellular matrix during invasion and metastasis. Annu. Rev. Cell. Biol. 9: 541, 1993. Birkedal-Hansen, H. Proteolytic remodeling of extracellular matrix. Curr. Opin. Cell. Biol. 7: 728, 1995. Powell, W. C., and Matrisian, L. M. Complex roles of matrix metalloproteinases in tumor progression. Curr. Top. Microbiol. Immunol. 213: 1, 1996.