- Email: [email protected]
Effects of Celebrex and Zyflo on BOP-induced pancreatic cancer in syrian hamsters
Original Paper Received: February 14, 2001 Accepted: June 25, 2001
Pancreatology 2002;2:54–60
Effects of Celebrex and Zyflo on BOP-Induced Pancreatic Cancer in Syrian Hamsters F.A. Wenger a M. Kilian a P. Achucarro a D. Heinicken a I. Schimke b H. Guski c C.A. Jacobi a J.M. Müller a a Clinic
of General, Visceral, Vascular and Thoracic Surgery, b Clinic of Internal Medicine I and c Institute of Pathology, Charité Campus Mitte, Humboldt University, Berlin, Germany
Key Words Pancreatic cancer W Syrian hamster W Inhibition of eicosanoid synthesis
Abstract Background/Aims: Selective inhibition of eicosanoid synthesis decreases inflammation, however, it is still unknown whether oxidative stress and carcinogenesis might be influenced in ductal pancreatic ductal cancer as well. Methods: 120 male hamsters were randomized into 8 groups (n = 15). While control group 1–4 received 0.5 ml normal saline s.c. weekly for 16 weeks, groups 5–8 were injected 10 mg BOP/kg body weight to induce pancreatic cancer. After establishment of pancreatic cancer, groups 1 and 5 received no therapy, groups 2 and 6 were fed 7 mg Celebrex daily, groups 3 and 7 were given 28 mg Zyflo and groups 4 and 8 received Celebrex and Zyflo orally daily in weeks 17–32. In week 33, all animals were sacrificed, macroscopic size of pancreatic carcinomas was measured, incidence of pancreatic cancer was analyzed histopathologically and activities of antioxidative enzymes and concentration of products of lipid peroxidation in tumor-free and pancreatic intratumoral tissue were determined. Results: Incidence and size of macroscopic pancreatic carcinomas were decreased by single therapy with Zyflo as well as combined therapy
ABC
© 2002 S. Karger AG, Basel and IAP 1424–3903/02/0021–0054$18.50/0
Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com
Accessible online at: www.karger.com/journals/pan
(Zyflo + Celebrex). Activities of antioxidative enzymes were increased and the concentration of products of lipid peroxidation was decreased in tumor-free pancreas. On the other hand, lipid peroxidation was increased in pancreatic tumors. Conclusion: Zyflo alone or in combination with Celebrex reduce tumor growth in pancreatic cancer and thus might be a new therapeutic option in advanced pancreatic cancer. Copyright © 2002 S. Karger AG, Basel and IAP
Introduction
Since the majority of patients with pancreatic cancer suffers from advanced disease at the time of diagnosis, the rate of curative resection is limited to 20–30%. However, even after radical resection, local recurrence is a frequent event [1–5]. Therefore, evaluation of new therapeutic strategies is urgent. Accordingly, the role of free radicals and eicosanoid synthesis in carcinogenesis is stressed by epidemiological and experimental trials [6–11]. Thus, selective inhibition of eicosanoid synthesis is supposed to decrease tumor growth [9, 12–18]. Besides direct effects with a reduction of eicosanoid levels, indirect effects with regard to oxygen radical formation and lipid peroxidation are discussed [9, 12–18].
Dr. F.A. Wenger Clinic of General, Visceral, Vascular and Thoracic Surgery Charité Campus Mitte, Schumannstrasse 20/21 D–10117 Berlin (Germany) Tel. +49 30 2802 2070, Fax +49 30 2802 1323, E-Mail [email protected]
The present trial was designed to investigate the effects of Celebrex, a selective cyclooxygenase-2 (COX-2) inhibitor, and Zyflo, a selective 5-lipoxygenase (5-LOX) inhibitor, on tumor growth and lipid peroxidation in a model of chemically induced ductal pancreatic adenocarcinoma in Syrian hamsters.
Methodology Animals, Tumor Induction, Therapy and Diet One hundred and twenty 8-week-old male Syrian Golden hamsters (Harlan-Winkelmann, Germany) were housed for 32 weeks in single cages with corncob bedding (Bed-O’cobs, Anderson Cob, Maumee, Ohio, USA) under standardized conditions with constant temperature (21 B 5 ° C), humidity (70 B 10%), as well as air exchanges and a 12:12-hour light/dark cycle. The experiment was approved by the public animal welfare committee and was carried out according to the UKCCCR Guidelines for the welfare of animals in experimental neoplasia [19]. Animals were randomized into 4 control groups (groups 1–4) and 4 tumor (groups 5–8) of 15 hamsters each. Control groups: Group 1: no tumor induction (ØBOP), no therapy; group 2: ØBOP, Celebrex; group 3: ØBOP, Zyflo; group 4: ØBOP, combination therapy. Tumor groups: Group 5: BOP, no therapy; group 6: BOP, Celebrex; group 7: BOP, Zyflo; group 8: BOP, combination therapy. While groups 1–4 were injected 0.5 ml 0.9% NaCl s.c. weekly for 16 weeks, groups 5–8 received 10 mg BOP (Ash Stevens Chem., USA) per kg body weight s.c. weekly for the same time. Injections were carried out under ether anesthesia. Therapy started after the induction of pancreatic cancer (week 16) and lasted for 16 weeks. In this period, groups 2 and 6 received 7 mg Celebrex (Pfizer, Zürich, Switzerland) orally daily, while groups 3 and 7 received 28 mg Zyflo (Abbott, Chicago, Ill., USA) orally daily. Groups 4 and 8 were administered Celebrex and Zyflo in the above-mentioned schedules. All animals had free access to water and to a previously described experimental diet (ssniff Soest GmbH, Germany) [20, 21]. In this diet, the raw fat content was elevated to 21.4% (compared to 3.5% in standard hamster diet) with an amount of 2% ·-linolenic acid (ALA) and 11% linoleic acid (LA). In previous studies this diet caused an increase of liver metastasis up to 90% [20, 21].
Sample Preparation for Biochemical Analysis An Ultraturrax homogenizer was used for sample preparation. In order to determine the concentration of lipid peroxide (LPO), tissue was homogenized in 1–10 vol% of ice-cold 140 mmol/l NaCl containing 0.01% butylated hydroxyanisole (pH 7.4). In order to analyze the activity of GSH-Px and SOD, tissue was homogenized in 1– 10 vol% of ice-cold 0.1 mol/l phosphate buffer (pH 7.4). LPO, SOD and GSH-Px Measurement, Protein Determination The concentration of LPO products was measured as thiobarbituric acid-reactive substances (TBARS) according to Ohkawa et al. [23]. Briefly, 20 Ìl homogenate was incubated at 90 ° C for 60 min in 0.8 ml reaction mixture of 1.35 mol/l acetic acid, 0.15 mol/l sodium dodecyl sulfate, 20 mol/l 1-thiobarbituric acid adjusted to pH 3.5 with NaOH. After cooling in ice water, 0.2 ml distilled water and 1 ml n-butanol/pyridine mixture (15/1) were added. This mixture was shaken and centrifuged. The organic layer was taken for fluorimetric measurements of the TBARS fluorescence (excitation 515 nm, emission 553 nm). The TBARS concentration was calculated using 1,1,3,3-tetraethoxypropane dissolved in acetic acid as standard. We are aware that there are some critical aspects concerning the influence of sample collection , storage and preparation as well as the test specificity on the TBARS levels measured in the tissue [24]. Based on previous investigations, we believe that we were able to prevent artifactual changes of the TBARS level during sample collection, storage, homogenization and analytical procedure by strict standardization of all steps (e.g. very fast freezing, storage in liquid nitrogen, tissue homogenization and analytical procedure in the presence of antioxidant). Based on the method of Beauchamp and Fridovich [25], SOD activity was measured in terms of its ability to prevent the formation of formazane from 2-(4-iodophenyl)-3-(4-nitro)-5-phenyltetrazolium chloride (INT) by superoxide radicals generated by xanthine oxidase/ xanthine. The test kit RANSOD (Randox Laboratories Ltd, UK) was used. The formazane formation was recorded spectrophotometrically (Shimadzu UV-2101 PC) at 550 nm. An inhibition by 50% of the INT reduction after addition of the sample (0.01–0.02 mg tissue/ ml test volume) was defined as 1 unit SOD. GSH-Px activity was measured according to Paglia and Valentine [26] using the test kit RANSEL (Randox Laboratories Ltd). The optimal used tissue concentration was 0.05–0.1 mg/ml test volume. Aliquots of the homogenate were used for the measurement of the protein content according to Lowry et al. [27].
Histology After 32 weeks, hamsters were sacrificed under anesthesia with ursotamine (Serumwerk Bernburg, Germany). Final body weight as well as pancreas and liver weights were estimated. From all animals, slices of macroscopically tumor-free areas (26 B 9 mg) were taken from the duodenal or splenic portion of the pancreas and immediately frozen at –80 ° C for biochemical analysis. Furthermore, in groups 5–8, slices of macroscopically visible pancreatic tumors (24 B 13 mg) were taken and immediately frozen at –80 ° C for biochemical analysis. The rest of the pancreas of BOP-treated animals was embedded in 10% buffered formaldehyde. Pancreatic specimens were cut at 5 Ìm and stained with hematoxylin and eosin for histological examination. Preneoplastic lesions, borderline lesions, carcinomas in situ and ductal adenocarcinomas were classified according to the criteria of Meijers et al. [22].
Statistics Data are given as mean B SD. Continuous data were tested for normal distribution with the Shapiro-Francia test. All data were found to be normally distributed. Accordingly, data between groups were compared using the one-way analysis of variance (ANOVA). Post-hoc tests were performed according to Bonferroni (equal variances assumed) or with Dunnett’s T3 test (unequal variances assumed). Comparison of intrapancreatic vs. intratumoral data in each group was carried out using Wilcoxon rank test. For categorical data, Fisher’s exact test was used, if appropriate. Generally, p values ! 0.05 were considered significant. Statistical analysis was carried out using the PC program SPSS 9.2® for Windows 97®.
Inhibition of Eicosanoid Synthesis in Pancreatic Cancer
Pancreatology 2002;2:54–60
55
Fig. 1. Incidence of pancreatic carcinoma.
PaCa = Pancreatic carcinoma; MPC = macroscopic pancreatic carcinoma. * p ! 0.05 vs. MPC in BOP/SAL.
n = 4) (p ! 0.05) and in group 5 (46.7%, n = 7) (p ! 0.05 vs. all groups). Incidence of Pancreatic Carcinomas As expected, there were no pancreatic carcinomas in groups 1–4. While the histopathological incidence of pancreatic cancer was 100% in groups 5–8, the incidence of macroscopic visible pancreatic carcinomas was 100% in group 5 and 81.8% in group 6 (Celebrex). However, it was significantly decreased in group 7 (53.3%, p ! 0.05 vs. group 5) and in group 8 (63.6%, p ! 0.05 vs. group 5) (fig. 1). All tumors were classified as ductal pancreatic adenocarcinomas.
Fig. 2. Size of macroscopic pancreatic carcinomas (mean B SD). * p ! 0.05 vs. BOP/SAL and BOP/CEL.
Results
Size of Macroscopic Pancreatic Carcinomas The size of macroscopic pancreatic carcinomas was 59.25 B 12.6 mm2 in group 5 and 60.0 B 12.7 mm2 in group 6, however, it was reduced by combined therapy (group 8: 24.86 B 5.52 mm2, p ! 0.05). The lowest size of macroscopic pancreatic carcinomas was found after therapy with Zyflo (group 7: 13.25 B 7.46 mm2, p ! 0.05; fig. 2).
General Characteristics and Lethality
There were no differences according to body weights or pancreas weights between the 8 groups. Furthermore, protein concentration in tumor-free pancreas as well as in pancreatic tumors did not differ. While lethality was 0% in groups 1, 4 and 7, it was 6.7% in group 3 (n = 1), 13.3% in group 2 (n = 2) and increased in groups 6 and 8 (26.7%,
56
Pancreatology 2002;2:54–60
GSH-Px Activity Pancreatic GSH-Px activity did not differ between tumor-free groups 1–4 (group 1: 8.72 B 3.89; group 2: 12.79 B 7.73; group 3: 13.99 B 7.46; group 4: 11.12 B 8.77 W 105 U/mg protein). Comparing tumor-free pancreas tissue of the tumor groups, the lowest GSH-Px activity was
Wenger/Kilian/Achucarro/Heinicken/ Schimke/Guski/Jacobi/Müller
Fig. 3. GSH-Px activity (mean B SD). TFP = Tumor-free pancreas; PaCa = pancreatic carcinoma. * p ! 0.05 vs. all TFP; + p ! 0.05 vs. all TFP, not vice versa; x p ! 0.05 vs. all PaCa.
observed in group 5 (2.76 B 1.24 W 105 U/mg protein) and group 6 (2.44 B 1.68 W 105 U/mg protein) (each p ! 0.05 vs. all other groups). On the other hand, the highest activity was found in group 7 (Zyflo) and group 8 (Zyflo + Celebrex) (33.2 B 6.59 and 25.3 B 3.99 W 105 U/mg protein, each p ! 0.05 vs. all groups; fig. 3). In pancreatic carcinomas, the lowest GSH-Px activity was found in group 5 (0.54 B 0.19 vs. 5.18 B 1.45 vs. 7.16 B 1.18 vs. 5.45 B 0.98 W 105 U/mg protein) (p ! 0.05 vs. groups 6–8; fig. 3). Accordingly, in groups 5, 7 and 8, GSH-Px activity was lower. In groups 5, 7 and 8, GSH-Px activity in pancreatic carcinomas was lower than in tumor-free pancreas (p ! 0.05) (fig. 3). SOD Activity Pancreatic SOD activity did not differ between tumorfree groups 1–4 (79.63 B 44.84, 93.11 B 47.69, 134.31 B 61.37 and 93.88 B 41.44 U/mg protein). However, in tumor-free pancreatic tissue of tumor groups, SOD activity was increased in group 8 (212.10 B 49.87 U/mg protein, p ! 0.05) and group 7 (333.77 B 68.79 U/mg protein, p ! 0.05; fig. 4). While SOD activity in pancreatic carcinomas did not differ between groups 5, 7 and 8 (53.28 B 25.82, 25.03 B 10.65 and 27.77 B 11.79 U/mg protein), it was increased in group 6 (Celebrex) (77.22 B 21.45 U/mg protein, p ! 0.05; fig. 4). In group 6, SOD activity in pancreatic carcinoma was elevated compared
Inhibition of Eicosanoid Synthesis in Pancreatic Cancer
to tumor-free pancreatic tissue, while in groups 7 and 8 we observed higher levels of activity in tumor-free pancreas than in pancreatic cancer (p ! 0.05; fig. 4). TBARS Concentration TBARS concentration in tumor-free pancreas did not differ between groups 1–4, 6 and 8 (35.37 B 27.45, 39.64 B 29.22, 52.35 B 31.88, 33.04 B 27.17, 32.59 B 7.72 and 23.57 B 6.96 nmol/mg protein), however, it was increased in group 5 (84.61 B 18.33 nmol/mg protein, p ! 0.05 vs. groups 1, 2, 4, 6–8). On the other hand, TBARS concentration was decreased in group 7 (12.85 B 3.77 nmol/mg protein) (p ! 0.05 vs. groups 3, 6, 8; fig. 5). In pancreatic carcinomas, TBARS concentration did not differ between groups 5 and 6 (17.15 B 5.23 vs. 14.22 B 4.91 nmol/mg protein), however, it was increased in group 8 (22.96 B 4.94 vs. 14.22 B 4.91 nmol/mg protein, p ! 0.05; group 8 vs. group 6). The highest concentration of TBARS of all tumor groups was found in group 7 (29.93 B 5.27 nmol/mg protein, p ! 0.05; fig. 5). While in group 8 TBARS concentration did not differ between pancreatic carcinomas and tumor-free pancreas, in group 7 it was increased in tumor-free pancreas compared to neoplastic tissue (p ! 0.05). However, it was lower than in tumor-free pancreas of groups 5 and 6 (p ! 0.05; fig. 5).
Pancreatology 2002;2:54–60
57
4
5 Fig. 4. SOD activity (mean B SD). TFP = Tumor-free pancreas; PaCa = pancreatic carcinoma. * p ! 0.05 vs. all TFP; +
p ! 0.05 vs. PaCa in BOP/ZYF and BOP/COM.
Fig. 5. TBARS concentration (mean B SD). TFP = Tumor-free pancreas; PaCa = pancreatic carcinoma. * p ! 0.05 vs.
all TFP without SAL/ZYF; + p ! 0.05 vs. TFP in SAL/ZYF, BOP/CEL and BOP/COM; BOP/CEL; [ p ! 0.05 vs. all PaCa.
Discussion
The model of BOP-induced ductal pancreatic adenocarcinoma in Syrian Golden hamster is a widely accepted experimental model especially because of biological and
58
Pancreatology 2002;2:54–60
P
p ! 0.05 vs. PaCa in
histological similarities to human pancreatic cancer [28, 29]. A dietary modification, the elevation of the content of raw fat, LA and ALA increased the incidence of pancreatic cancer and liver metastasis up to 100 and 90% [20, 21]. While carcinogenesis was stimulated by changes of
Wenger/Kilian/Achucarro/Heinicken/ Schimke/Guski/Jacobi/Müller
oxygen radical metabolism [21] influencing these pathways, tumor growth was decreased by application of antioxidative vitamins [30]. However, oxygen radical metabolism and eicosanoid synthesis seem to have similar mechanisms [6, 8, 13, 31]. Accordingly, eicosanoids possibly play an important role in carcinogenesis as well as in metastasis [32–38]. Thus, protective and therapeutic effects on carcinogenesis and tumor growth by eicosanoid synthesis inhibitors have been demonstrated by several in vitro and in vivo trials [12, 14, 15, 17, 18, 39]. Furthermore, we observed COX-2 expression in intrametastatic hepatic tissue of BOP-induced ductal pancreatic cancer. Therefore, we analyzed the effects of selective inhibition of COX-2 and 5-LOX on tumor growth and pancreatic lipid peroxidation in a model of chemically induced solid ductal pancreatic adenocarcinoma in Syrian hamsters. We did not observe any difference between the control groups without pancreatic cancer concerning pancreatic lipid peroxidation and activity of antioxidative enzymes GSH-Px and SOD after application of Celebrex or Zyflo. In the tumor groups, therapy with high selective COX-2 inhibitor Celebrex had no effect on the size and incidence of macroscopic visible pancreatic carcinomas. In contrast, 5LOX inhibitor Zyflo as well as the combined therapy (Zyflo + Celebrex) decreased post-parameters (size and incidence). Both therapeutic schedules elevated the activities of GSH-Px and SOD in tumor-free pancreas. Furthermore, Zyflo reduced the concentration of TBARS in tumor-free pancreas. In pancreatic carcinoma both schedules increased GSH-Px activity and concentration of TBARS. Accordingly, Zyflo and the combined therapy (Zyflo + Celebrex) seemed to stabilize the antioxidative defense in tumor-free pancreas and possibly protect pancreatic tissue against radicals of the high-fat diet- and BOP-related lipid peroxidation. This mechanism might be responsible for reduced carcinogenesis in our trial. In pancreatic carcinomas, only GSH-Px activity was increased by Zyflo and/or Celebrex, while SOD activity decreased. Since the exclusive rise of GSH-Px activity is not supposed to protect tumorous tissue against lipid peroxidation, one would expect rather an increased lipid peroxidation, because a balance between the various components of the antioxidative system seems to be more important for protection against lipid peroxidation that the single elevation of one enzyme [40–42]. Accordingly, therapy with Zyflo and the combined therapy increased the concentration of TBARS in pancreatic carcinomas. Thus, high levels of lipid peroxidation in pancreatic carcinomas might contribute to membrane damage of tumor cells, leading to loss of integrity and consecutive death of the tumor cells.
In the COX- and LOX-related pathways, a multitude of free radical reactions is involved [6, 8, 31]. Thus, a decreased formation of prostaglandins and leukotrienes was observed after application of antioxidative vitamins [43–45]. Ding et al. [46] demonstrated induction of apoptosis in pancreatic carcinoma cell lines following lipoxygenase inhibition. Furthermore, in vitro they observed an inhibition of proliferation of pancreatic carcinoma cells after inhibition of lipoxygenase [46]. These results were confirmed in our recent trial in a solid tumor model of pancreatic adenocarcinoma. Accordingly, Anderson et al. [47] reported a reduced survival of PANC-11 pancreatic carcinoma cell line according to application of 5-LOX inhibitors. However, contrary to Ding et al. [46], we did not find an effect of COX-2 inhibition on tumor growth in pancreatic cancer. The exact coherence between eicosanoid synthesis and oxidative stress or lipid peroxidation is still unknown. However, there are several reports that the inhibition of one pathway of eicosanoid synthesis might shunt off the remaining pathway [8, 9, 15, 18, 48]. Concerning our data, this possibly means that 5-LOX inhibitor Zyflo might have provoked an increase of COX-2 activity and consecutive overproduction of prostaglandins resulting in a decrease of tumor growth. Thus, the combined inhibition of both pathways showed similar effects. We observed neither synergistic nor antagonistic effects, conclusively this phenomenon cannot be explained at the moment. As tumor induction by subcutaneous injection of BOP was performed between weeks 1 and 6 and therapy with Celebrex and Zyflo was administered between weeks 17 and 33, inhibition of tumor growth does not seem to be caused by change of BOP metabolism. According to our results, treatment reduced the concentration of lipid peroxidation in tumor-free pancreas and increased lipid peroxidation in pancreatic tumors. Furthermore, activities of antioxidative enzymes were increased in tumorfree pancreas. Thus, prevention of free radicals might be a mechanism for reduced tumor growth. Accordingly, we intend to determine in further trials whether the missing effect of COX-2 inhibition is dosedependent. At the moment the clinical indication for Celebrex is rheumatic disease. Thus, results of dosis-finding trials are missing in the literature. Furthermore, the relations between oxidative stress, lipid peroxidation and eicosanoid synthesis should be examined. Possibly, selective inhibition of eicosanoid synthesis is a new approach for therapy of pancreatic cancer.
Inhibition of Eicosanoid Synthesis in Pancreatic Cancer
Pancreatology 2002;2:54–60
59
References 1 Gloor B, Todd KE: Diagnostic workup of patients with suspected pancreatic carcinoma. Cancer 1997;79:1780–1786. 2 Warshaw AL, Fernandez-del Castillo C: Pancreatic carcinoma. N Engl J Med 1992;326: 455–465. 3 Amikura K, Kobari M: The time of occurrence of liver metastasis in carcinoma of the pancreas. Int J Pancreatol 1995;17:139–146. 4 Böhming M, Wiedenmann B, Rosewicz S: Therapie des Pankreasadenokarzinoms. Med Klin 1999;94:614–625. 5 Hawes RH, Xiong Q, Waxman I, Chang KJ, Evans DB, Abbruzzese JL: A multispecialty approach to the diagnosis and management of pancreatic cancer. Am J Gastroenterol 2000; 95:17–31. 6 Rice-Evans C, Burdon R: Free radical lipid interactions and their pathological consequences. Prog Lipid Res 1993;32:71–110. 7 Ondrey FG: Arachidonic acid metabolism: A primer for head and neck surgeons. Head Neck 1998;20:334–349. 8 Davies GR, Rampton DS: Eicosanoids: Role in gastrointestinal inflammation and cancer. Eur J Gastroenterol Hepatol 1997;9:1033–1044. 9 Karmali RA: Eicosanoids in neoplasia. Prev Med 1987;16:493–502. 10 Benzie IF: Lipid peroxidation: A rewiev of causes, consequences, measurement and dietary influences. Int J Food Sci Nutr 1996;47: 233–261. 11 Hollan S: Free radicals in health and disease. Haematologia (Budap) 1995;26:177–189. 12 Vainio H, Morgan G: Non-steroidal anti-inflammatory drugs and the chemoprevention of gastrointestinal cancers. Scand J Gastroenterol 1998;33:785–789. 13 Damtew B, Spagnuolo PJ: Tumor cell-endothelial cell interactions: Evidence for roles for lipoxygenase products of arachidonic acid in metastasis. Prostaglandins Leukot Essent Fatty Acids 1997;56:295–300. 14 Subbaramaiah K, Zakim D, Weksler BB, Dannenberg AJ: Inhibition of cyclooxygenase: A novel approach to cancer prevention. Proc Soc Exp Biol Med 1997;216:201–210. 15 Marks F, Fürstenberger G, Müller-Decker K: Metabolic targets of cancer chemoprevention: Interruption of tumor development by inhibitors of arachidonic acid metabolism. Recent Res Cancer Res 1999;151:45–67. 16 Lupulescu AP: Prostaglandins, their inhibitors and cancer. Prostaglandins Leukot Essent Fatty Acids 1996;54:83–94. 17 Ara G, Teicher BA: Cyclooxygenase and lipoxygenase inhibitors in cancer therapy. Prostaglandins Leukot Essent Fatty Acids 1996;54:3– 16. 18 Vane JR, Botting RM: Mechanism of action of anti-inflammatory drugs. Int J Tissue React 1998;20:3–15. 19 UKCCCR Guidelines for the welfare of animals in experimental neoplasia. Cancer Metastasis Rev 1989;8:82–88.
60
20 Wenger FA, Kilian M, Jaccobi CA, Schimke I, Guski H, Müller JM: Does ·-linolenic acid in combination with linolenic acid influence liver metastasis and hepatic lipid peroxidation in BOP-induced pancreatic cancer in Syrian hamsters? Prostaglandins Leukot Essent Fatty Acids 2000;62:329–334. 21 Wenger FA, Jacobi CA, Kilian M, Zieren J, Zieren HU, Müller JM: Does dietary ·-linolenic acid promote liver metastasis pancreatic carcinoma initiated by BOP in Syrian hamsters? Ann Nutr Metabol 1999;43:121–126. 22 Meijers M, van Garderen-Hoetmer A, Lamers CB, Rovati LC, Jansen JB, Woutersen RA: Role of cholecystokinin in the development of BOP-induced pancreatic lesions in hamsters. Carcinogenesis 1990;11:2223–2226. 23 Ohkawa H, Ohishi N, Yagi K: Assay for lipid peroxides in animals tissue by thiobarbituric acid reaction. Anal Bioechem 1979;95:351– 358. 24 Janero DR: Malondialdehyde and thiobarbituric acid reactivity as diagnostic indices of lipid peroxidation and peroxidative tissue injury. Free Radical Biol Med 1990;9:515–540. 25 Beauchamp C, Fridovich I: Superoxide dismutase: Improved assay and an assay applicable to acrylamide gels. J Anal Biochem 1971;44:276– 287. 26 Paglia DE, Valentine WN: Studies on the quantitative and qualitative characterization of erythrocytes glutathione peroxidase. J Lab Clin Med 1967;29:143–148. 27 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265–275. 28 Longnecker DS: Experimental cancer of the pancreas. Curr Opin Gastroenterol 1991;7: 731–738. 29 Pour PM, Runge RG, Birt D, Gingell R, Lawson T, Nagel D, Wallave L, Salmasi SZ: Current knowledge of pancreatic carcinogenesis in the hamster and its relevance to the human disease. Cancer 1981;47:1573–1587. 30 Wenger FA, Kilian M, Neumann A, Peter FJ, Jacobi CA, Guski H, Schimke I, Müller JM: Effects of antioxidative vitamins A, C, E on liver metastasis and intrametastatic lipidperoxidation in BOP-induced pancreatic cancer in Syrian hamsters. Carcinogenesis 2001, in press. 31 Arntzen KJ, Brekke OL, Vatten L, Austgulen R: Reduced production of PGE2 and PGF2· from decidual cell cultures supplemented with n-3 polyunsaturated fatty acids. Prostaglandins Lipid Mediat 1998;56:183–195. 32 DuBois RN, Smalley WE: Cyclooxygenase, NSAIDs, and colorectal cancer. J Gastroenterol 1996;31:898–906. 33 Yip-Schneider MT, Barnard DS, Billings SD, Cheng L, Heilmann DK, Lin A, Marshall SJ, Crowell PL, Marshall MS, Sweeney CJ: Cyclooxygenase-2 expression in human pancreatic adenocarcinomas. Carcinogenesis 2000;21: 139–146.
Pancreatology 2002;2:54–60
34 Uefuji K, Ichikura T, Mochizuki H, Shinomiya N: Expression of cyclooxygenase-2 protein in gastric adenocarcinoma. J Surg Oncol 1998;69: 168–172. 35 Wolff H, Saukkonen K, Anttila S, Karjalainen A, Vainio H, Ristimäki A: Expression of cyclooxygenase-2 in human lung carcinoma. Cancer Res 1998;58:4997–5001. 36 Hwang D, Scollard D, Byrne J, Levine E: Expression of cyclooxygenase-1 and cyclooxygenase-2 in human breast cancer. J Natl Cancer Inst 1998;90:455–460. 37 Shiota G, Okubo M, Noumi T, Noguchi N, Oyama K, Takano Y, Yashima K, Kishimoto Y, Kawasaki H: Cyclooxygenase-2 expression in hepatocellular carcinoma. Hepatogastroenterology 1999;46:407–412. 38 Myers CE, Ghosh J: Lipoxygenase ihhibition in prostate cancer. Eur Urol 1999;35:395–398. 39 Tsuji M, Kawano S, Tsuji S, Sawaoka H, Hori M, DuBois RN: Cyclooxygenase regulates angiogenesis induced by colon cancer cells. Cell 1998;93:705–716. 40 De Haan JB, Christiano F, Janello R, Bladier C, Kelner MJ, Kola J: Elevation in the ratio of Cu/Zn superoxide dismutase to glutathione peroxidase activity induced features of cellular senescence and this effect is mediated by hydrogen peroxide. Hum Mol Genet 1996;5:283– 292. 41 Oberley TD, Oberley LW: Invited review: Antioxidant enzyme levels in cancer. Histol Histopathol 1997;12:525–535. 42 Amstadt P, Moret R, Cerutti P: Glutathione peroxidase compensates for the hypersensitivity of Cu/Zn superoxide dismutase overproducers to oxidant stress. J Biol Chem 1994;269: 1606–1609. 43 Hamelin S, Chan AC: Modulation of platelet thromboxane and malondialdehyde by dietary vitamin E and linoleate. Lipids 1983;18:267– 269. 44 Chan AC, Tran K, Pyke DD, Powell WS: Effects of dietary vitamin E on the biosynthesis of 5-lipoxygenase products by rat polymorphonuclear leukocytes PMNL. Biochim Biophys Acta 1989;1005:265–269. 45 Sakamoto W, Fujie K, Nishihira J, Handa H: Effect of vitamin E on arachidonic acid release in rat peritoneal macrophages. Biochim Biophys Acta 1993;1170:296–300. 46 Ding XZ, Kuszynski CA, El-Metwally TH, Adrian TE: Lipoxygenase inhibition-induced apoptosis, morphological changes, and carbonic anhydrase expression in human pancreatic cancer cells. Biochem Biophys Res Commun 1999;266:392–399. 47 Anderson KM, Seed T, Meng J, Ou D, Alrefai WA, Harris JE: 5-Lipoxygenase inhibitors reduce PANC-1 survival: The mode of cell death and synergism of MK886 with Á-linolenic acid. Anticancer Res 1998;18:791–800. 48 Reich R, Martin GR: Identification of arachidonic acid pathways required for the invasive and metastatic activity in malignant tumor cells. Prostaglandins 1996;51:1–17.
Wenger/Kilian/Achucarro/Heinicken/ Schimke/Guski/Jacobi/Müller
Pancreatology 2002;2:54–60
Effects of Celebrex and Zyflo on BOP-Induced Pancreatic Cancer in Syrian Hamsters F.A. Wenger a M. Kilian a P. Achucarro a D. Heinicken a I. Schimke b H. Guski c C.A. Jacobi a J.M. Müller a a Clinic
of General, Visceral, Vascular and Thoracic Surgery, b Clinic of Internal Medicine I and c Institute of Pathology, Charité Campus Mitte, Humboldt University, Berlin, Germany
Key Words Pancreatic cancer W Syrian hamster W Inhibition of eicosanoid synthesis
Abstract Background/Aims: Selective inhibition of eicosanoid synthesis decreases inflammation, however, it is still unknown whether oxidative stress and carcinogenesis might be influenced in ductal pancreatic ductal cancer as well. Methods: 120 male hamsters were randomized into 8 groups (n = 15). While control group 1–4 received 0.5 ml normal saline s.c. weekly for 16 weeks, groups 5–8 were injected 10 mg BOP/kg body weight to induce pancreatic cancer. After establishment of pancreatic cancer, groups 1 and 5 received no therapy, groups 2 and 6 were fed 7 mg Celebrex daily, groups 3 and 7 were given 28 mg Zyflo and groups 4 and 8 received Celebrex and Zyflo orally daily in weeks 17–32. In week 33, all animals were sacrificed, macroscopic size of pancreatic carcinomas was measured, incidence of pancreatic cancer was analyzed histopathologically and activities of antioxidative enzymes and concentration of products of lipid peroxidation in tumor-free and pancreatic intratumoral tissue were determined. Results: Incidence and size of macroscopic pancreatic carcinomas were decreased by single therapy with Zyflo as well as combined therapy
ABC
© 2002 S. Karger AG, Basel and IAP 1424–3903/02/0021–0054$18.50/0
Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com
Accessible online at: www.karger.com/journals/pan
(Zyflo + Celebrex). Activities of antioxidative enzymes were increased and the concentration of products of lipid peroxidation was decreased in tumor-free pancreas. On the other hand, lipid peroxidation was increased in pancreatic tumors. Conclusion: Zyflo alone or in combination with Celebrex reduce tumor growth in pancreatic cancer and thus might be a new therapeutic option in advanced pancreatic cancer. Copyright © 2002 S. Karger AG, Basel and IAP
Introduction
Since the majority of patients with pancreatic cancer suffers from advanced disease at the time of diagnosis, the rate of curative resection is limited to 20–30%. However, even after radical resection, local recurrence is a frequent event [1–5]. Therefore, evaluation of new therapeutic strategies is urgent. Accordingly, the role of free radicals and eicosanoid synthesis in carcinogenesis is stressed by epidemiological and experimental trials [6–11]. Thus, selective inhibition of eicosanoid synthesis is supposed to decrease tumor growth [9, 12–18]. Besides direct effects with a reduction of eicosanoid levels, indirect effects with regard to oxygen radical formation and lipid peroxidation are discussed [9, 12–18].
Dr. F.A. Wenger Clinic of General, Visceral, Vascular and Thoracic Surgery Charité Campus Mitte, Schumannstrasse 20/21 D–10117 Berlin (Germany) Tel. +49 30 2802 2070, Fax +49 30 2802 1323, E-Mail [email protected]
The present trial was designed to investigate the effects of Celebrex, a selective cyclooxygenase-2 (COX-2) inhibitor, and Zyflo, a selective 5-lipoxygenase (5-LOX) inhibitor, on tumor growth and lipid peroxidation in a model of chemically induced ductal pancreatic adenocarcinoma in Syrian hamsters.
Methodology Animals, Tumor Induction, Therapy and Diet One hundred and twenty 8-week-old male Syrian Golden hamsters (Harlan-Winkelmann, Germany) were housed for 32 weeks in single cages with corncob bedding (Bed-O’cobs, Anderson Cob, Maumee, Ohio, USA) under standardized conditions with constant temperature (21 B 5 ° C), humidity (70 B 10%), as well as air exchanges and a 12:12-hour light/dark cycle. The experiment was approved by the public animal welfare committee and was carried out according to the UKCCCR Guidelines for the welfare of animals in experimental neoplasia [19]. Animals were randomized into 4 control groups (groups 1–4) and 4 tumor (groups 5–8) of 15 hamsters each. Control groups: Group 1: no tumor induction (ØBOP), no therapy; group 2: ØBOP, Celebrex; group 3: ØBOP, Zyflo; group 4: ØBOP, combination therapy. Tumor groups: Group 5: BOP, no therapy; group 6: BOP, Celebrex; group 7: BOP, Zyflo; group 8: BOP, combination therapy. While groups 1–4 were injected 0.5 ml 0.9% NaCl s.c. weekly for 16 weeks, groups 5–8 received 10 mg BOP (Ash Stevens Chem., USA) per kg body weight s.c. weekly for the same time. Injections were carried out under ether anesthesia. Therapy started after the induction of pancreatic cancer (week 16) and lasted for 16 weeks. In this period, groups 2 and 6 received 7 mg Celebrex (Pfizer, Zürich, Switzerland) orally daily, while groups 3 and 7 received 28 mg Zyflo (Abbott, Chicago, Ill., USA) orally daily. Groups 4 and 8 were administered Celebrex and Zyflo in the above-mentioned schedules. All animals had free access to water and to a previously described experimental diet (ssniff Soest GmbH, Germany) [20, 21]. In this diet, the raw fat content was elevated to 21.4% (compared to 3.5% in standard hamster diet) with an amount of 2% ·-linolenic acid (ALA) and 11% linoleic acid (LA). In previous studies this diet caused an increase of liver metastasis up to 90% [20, 21].
Sample Preparation for Biochemical Analysis An Ultraturrax homogenizer was used for sample preparation. In order to determine the concentration of lipid peroxide (LPO), tissue was homogenized in 1–10 vol% of ice-cold 140 mmol/l NaCl containing 0.01% butylated hydroxyanisole (pH 7.4). In order to analyze the activity of GSH-Px and SOD, tissue was homogenized in 1– 10 vol% of ice-cold 0.1 mol/l phosphate buffer (pH 7.4). LPO, SOD and GSH-Px Measurement, Protein Determination The concentration of LPO products was measured as thiobarbituric acid-reactive substances (TBARS) according to Ohkawa et al. [23]. Briefly, 20 Ìl homogenate was incubated at 90 ° C for 60 min in 0.8 ml reaction mixture of 1.35 mol/l acetic acid, 0.15 mol/l sodium dodecyl sulfate, 20 mol/l 1-thiobarbituric acid adjusted to pH 3.5 with NaOH. After cooling in ice water, 0.2 ml distilled water and 1 ml n-butanol/pyridine mixture (15/1) were added. This mixture was shaken and centrifuged. The organic layer was taken for fluorimetric measurements of the TBARS fluorescence (excitation 515 nm, emission 553 nm). The TBARS concentration was calculated using 1,1,3,3-tetraethoxypropane dissolved in acetic acid as standard. We are aware that there are some critical aspects concerning the influence of sample collection , storage and preparation as well as the test specificity on the TBARS levels measured in the tissue [24]. Based on previous investigations, we believe that we were able to prevent artifactual changes of the TBARS level during sample collection, storage, homogenization and analytical procedure by strict standardization of all steps (e.g. very fast freezing, storage in liquid nitrogen, tissue homogenization and analytical procedure in the presence of antioxidant). Based on the method of Beauchamp and Fridovich [25], SOD activity was measured in terms of its ability to prevent the formation of formazane from 2-(4-iodophenyl)-3-(4-nitro)-5-phenyltetrazolium chloride (INT) by superoxide radicals generated by xanthine oxidase/ xanthine. The test kit RANSOD (Randox Laboratories Ltd, UK) was used. The formazane formation was recorded spectrophotometrically (Shimadzu UV-2101 PC) at 550 nm. An inhibition by 50% of the INT reduction after addition of the sample (0.01–0.02 mg tissue/ ml test volume) was defined as 1 unit SOD. GSH-Px activity was measured according to Paglia and Valentine [26] using the test kit RANSEL (Randox Laboratories Ltd). The optimal used tissue concentration was 0.05–0.1 mg/ml test volume. Aliquots of the homogenate were used for the measurement of the protein content according to Lowry et al. [27].
Histology After 32 weeks, hamsters were sacrificed under anesthesia with ursotamine (Serumwerk Bernburg, Germany). Final body weight as well as pancreas and liver weights were estimated. From all animals, slices of macroscopically tumor-free areas (26 B 9 mg) were taken from the duodenal or splenic portion of the pancreas and immediately frozen at –80 ° C for biochemical analysis. Furthermore, in groups 5–8, slices of macroscopically visible pancreatic tumors (24 B 13 mg) were taken and immediately frozen at –80 ° C for biochemical analysis. The rest of the pancreas of BOP-treated animals was embedded in 10% buffered formaldehyde. Pancreatic specimens were cut at 5 Ìm and stained with hematoxylin and eosin for histological examination. Preneoplastic lesions, borderline lesions, carcinomas in situ and ductal adenocarcinomas were classified according to the criteria of Meijers et al. [22].
Statistics Data are given as mean B SD. Continuous data were tested for normal distribution with the Shapiro-Francia test. All data were found to be normally distributed. Accordingly, data between groups were compared using the one-way analysis of variance (ANOVA). Post-hoc tests were performed according to Bonferroni (equal variances assumed) or with Dunnett’s T3 test (unequal variances assumed). Comparison of intrapancreatic vs. intratumoral data in each group was carried out using Wilcoxon rank test. For categorical data, Fisher’s exact test was used, if appropriate. Generally, p values ! 0.05 were considered significant. Statistical analysis was carried out using the PC program SPSS 9.2® for Windows 97®.
Inhibition of Eicosanoid Synthesis in Pancreatic Cancer
Pancreatology 2002;2:54–60
55
Fig. 1. Incidence of pancreatic carcinoma.
PaCa = Pancreatic carcinoma; MPC = macroscopic pancreatic carcinoma. * p ! 0.05 vs. MPC in BOP/SAL.
n = 4) (p ! 0.05) and in group 5 (46.7%, n = 7) (p ! 0.05 vs. all groups). Incidence of Pancreatic Carcinomas As expected, there were no pancreatic carcinomas in groups 1–4. While the histopathological incidence of pancreatic cancer was 100% in groups 5–8, the incidence of macroscopic visible pancreatic carcinomas was 100% in group 5 and 81.8% in group 6 (Celebrex). However, it was significantly decreased in group 7 (53.3%, p ! 0.05 vs. group 5) and in group 8 (63.6%, p ! 0.05 vs. group 5) (fig. 1). All tumors were classified as ductal pancreatic adenocarcinomas.
Fig. 2. Size of macroscopic pancreatic carcinomas (mean B SD). * p ! 0.05 vs. BOP/SAL and BOP/CEL.
Results
Size of Macroscopic Pancreatic Carcinomas The size of macroscopic pancreatic carcinomas was 59.25 B 12.6 mm2 in group 5 and 60.0 B 12.7 mm2 in group 6, however, it was reduced by combined therapy (group 8: 24.86 B 5.52 mm2, p ! 0.05). The lowest size of macroscopic pancreatic carcinomas was found after therapy with Zyflo (group 7: 13.25 B 7.46 mm2, p ! 0.05; fig. 2).
General Characteristics and Lethality
There were no differences according to body weights or pancreas weights between the 8 groups. Furthermore, protein concentration in tumor-free pancreas as well as in pancreatic tumors did not differ. While lethality was 0% in groups 1, 4 and 7, it was 6.7% in group 3 (n = 1), 13.3% in group 2 (n = 2) and increased in groups 6 and 8 (26.7%,
56
Pancreatology 2002;2:54–60
GSH-Px Activity Pancreatic GSH-Px activity did not differ between tumor-free groups 1–4 (group 1: 8.72 B 3.89; group 2: 12.79 B 7.73; group 3: 13.99 B 7.46; group 4: 11.12 B 8.77 W 105 U/mg protein). Comparing tumor-free pancreas tissue of the tumor groups, the lowest GSH-Px activity was
Wenger/Kilian/Achucarro/Heinicken/ Schimke/Guski/Jacobi/Müller
Fig. 3. GSH-Px activity (mean B SD). TFP = Tumor-free pancreas; PaCa = pancreatic carcinoma. * p ! 0.05 vs. all TFP; + p ! 0.05 vs. all TFP, not vice versa; x p ! 0.05 vs. all PaCa.
observed in group 5 (2.76 B 1.24 W 105 U/mg protein) and group 6 (2.44 B 1.68 W 105 U/mg protein) (each p ! 0.05 vs. all other groups). On the other hand, the highest activity was found in group 7 (Zyflo) and group 8 (Zyflo + Celebrex) (33.2 B 6.59 and 25.3 B 3.99 W 105 U/mg protein, each p ! 0.05 vs. all groups; fig. 3). In pancreatic carcinomas, the lowest GSH-Px activity was found in group 5 (0.54 B 0.19 vs. 5.18 B 1.45 vs. 7.16 B 1.18 vs. 5.45 B 0.98 W 105 U/mg protein) (p ! 0.05 vs. groups 6–8; fig. 3). Accordingly, in groups 5, 7 and 8, GSH-Px activity was lower. In groups 5, 7 and 8, GSH-Px activity in pancreatic carcinomas was lower than in tumor-free pancreas (p ! 0.05) (fig. 3). SOD Activity Pancreatic SOD activity did not differ between tumorfree groups 1–4 (79.63 B 44.84, 93.11 B 47.69, 134.31 B 61.37 and 93.88 B 41.44 U/mg protein). However, in tumor-free pancreatic tissue of tumor groups, SOD activity was increased in group 8 (212.10 B 49.87 U/mg protein, p ! 0.05) and group 7 (333.77 B 68.79 U/mg protein, p ! 0.05; fig. 4). While SOD activity in pancreatic carcinomas did not differ between groups 5, 7 and 8 (53.28 B 25.82, 25.03 B 10.65 and 27.77 B 11.79 U/mg protein), it was increased in group 6 (Celebrex) (77.22 B 21.45 U/mg protein, p ! 0.05; fig. 4). In group 6, SOD activity in pancreatic carcinoma was elevated compared
Inhibition of Eicosanoid Synthesis in Pancreatic Cancer
to tumor-free pancreatic tissue, while in groups 7 and 8 we observed higher levels of activity in tumor-free pancreas than in pancreatic cancer (p ! 0.05; fig. 4). TBARS Concentration TBARS concentration in tumor-free pancreas did not differ between groups 1–4, 6 and 8 (35.37 B 27.45, 39.64 B 29.22, 52.35 B 31.88, 33.04 B 27.17, 32.59 B 7.72 and 23.57 B 6.96 nmol/mg protein), however, it was increased in group 5 (84.61 B 18.33 nmol/mg protein, p ! 0.05 vs. groups 1, 2, 4, 6–8). On the other hand, TBARS concentration was decreased in group 7 (12.85 B 3.77 nmol/mg protein) (p ! 0.05 vs. groups 3, 6, 8; fig. 5). In pancreatic carcinomas, TBARS concentration did not differ between groups 5 and 6 (17.15 B 5.23 vs. 14.22 B 4.91 nmol/mg protein), however, it was increased in group 8 (22.96 B 4.94 vs. 14.22 B 4.91 nmol/mg protein, p ! 0.05; group 8 vs. group 6). The highest concentration of TBARS of all tumor groups was found in group 7 (29.93 B 5.27 nmol/mg protein, p ! 0.05; fig. 5). While in group 8 TBARS concentration did not differ between pancreatic carcinomas and tumor-free pancreas, in group 7 it was increased in tumor-free pancreas compared to neoplastic tissue (p ! 0.05). However, it was lower than in tumor-free pancreas of groups 5 and 6 (p ! 0.05; fig. 5).
Pancreatology 2002;2:54–60
57
4
5 Fig. 4. SOD activity (mean B SD). TFP = Tumor-free pancreas; PaCa = pancreatic carcinoma. * p ! 0.05 vs. all TFP; +
p ! 0.05 vs. PaCa in BOP/ZYF and BOP/COM.
Fig. 5. TBARS concentration (mean B SD). TFP = Tumor-free pancreas; PaCa = pancreatic carcinoma. * p ! 0.05 vs.
all TFP without SAL/ZYF; + p ! 0.05 vs. TFP in SAL/ZYF, BOP/CEL and BOP/COM; BOP/CEL; [ p ! 0.05 vs. all PaCa.
Discussion
The model of BOP-induced ductal pancreatic adenocarcinoma in Syrian Golden hamster is a widely accepted experimental model especially because of biological and
58
Pancreatology 2002;2:54–60
P
p ! 0.05 vs. PaCa in
histological similarities to human pancreatic cancer [28, 29]. A dietary modification, the elevation of the content of raw fat, LA and ALA increased the incidence of pancreatic cancer and liver metastasis up to 100 and 90% [20, 21]. While carcinogenesis was stimulated by changes of
Wenger/Kilian/Achucarro/Heinicken/ Schimke/Guski/Jacobi/Müller
oxygen radical metabolism [21] influencing these pathways, tumor growth was decreased by application of antioxidative vitamins [30]. However, oxygen radical metabolism and eicosanoid synthesis seem to have similar mechanisms [6, 8, 13, 31]. Accordingly, eicosanoids possibly play an important role in carcinogenesis as well as in metastasis [32–38]. Thus, protective and therapeutic effects on carcinogenesis and tumor growth by eicosanoid synthesis inhibitors have been demonstrated by several in vitro and in vivo trials [12, 14, 15, 17, 18, 39]. Furthermore, we observed COX-2 expression in intrametastatic hepatic tissue of BOP-induced ductal pancreatic cancer. Therefore, we analyzed the effects of selective inhibition of COX-2 and 5-LOX on tumor growth and pancreatic lipid peroxidation in a model of chemically induced solid ductal pancreatic adenocarcinoma in Syrian hamsters. We did not observe any difference between the control groups without pancreatic cancer concerning pancreatic lipid peroxidation and activity of antioxidative enzymes GSH-Px and SOD after application of Celebrex or Zyflo. In the tumor groups, therapy with high selective COX-2 inhibitor Celebrex had no effect on the size and incidence of macroscopic visible pancreatic carcinomas. In contrast, 5LOX inhibitor Zyflo as well as the combined therapy (Zyflo + Celebrex) decreased post-parameters (size and incidence). Both therapeutic schedules elevated the activities of GSH-Px and SOD in tumor-free pancreas. Furthermore, Zyflo reduced the concentration of TBARS in tumor-free pancreas. In pancreatic carcinoma both schedules increased GSH-Px activity and concentration of TBARS. Accordingly, Zyflo and the combined therapy (Zyflo + Celebrex) seemed to stabilize the antioxidative defense in tumor-free pancreas and possibly protect pancreatic tissue against radicals of the high-fat diet- and BOP-related lipid peroxidation. This mechanism might be responsible for reduced carcinogenesis in our trial. In pancreatic carcinomas, only GSH-Px activity was increased by Zyflo and/or Celebrex, while SOD activity decreased. Since the exclusive rise of GSH-Px activity is not supposed to protect tumorous tissue against lipid peroxidation, one would expect rather an increased lipid peroxidation, because a balance between the various components of the antioxidative system seems to be more important for protection against lipid peroxidation that the single elevation of one enzyme [40–42]. Accordingly, therapy with Zyflo and the combined therapy increased the concentration of TBARS in pancreatic carcinomas. Thus, high levels of lipid peroxidation in pancreatic carcinomas might contribute to membrane damage of tumor cells, leading to loss of integrity and consecutive death of the tumor cells.
In the COX- and LOX-related pathways, a multitude of free radical reactions is involved [6, 8, 31]. Thus, a decreased formation of prostaglandins and leukotrienes was observed after application of antioxidative vitamins [43–45]. Ding et al. [46] demonstrated induction of apoptosis in pancreatic carcinoma cell lines following lipoxygenase inhibition. Furthermore, in vitro they observed an inhibition of proliferation of pancreatic carcinoma cells after inhibition of lipoxygenase [46]. These results were confirmed in our recent trial in a solid tumor model of pancreatic adenocarcinoma. Accordingly, Anderson et al. [47] reported a reduced survival of PANC-11 pancreatic carcinoma cell line according to application of 5-LOX inhibitors. However, contrary to Ding et al. [46], we did not find an effect of COX-2 inhibition on tumor growth in pancreatic cancer. The exact coherence between eicosanoid synthesis and oxidative stress or lipid peroxidation is still unknown. However, there are several reports that the inhibition of one pathway of eicosanoid synthesis might shunt off the remaining pathway [8, 9, 15, 18, 48]. Concerning our data, this possibly means that 5-LOX inhibitor Zyflo might have provoked an increase of COX-2 activity and consecutive overproduction of prostaglandins resulting in a decrease of tumor growth. Thus, the combined inhibition of both pathways showed similar effects. We observed neither synergistic nor antagonistic effects, conclusively this phenomenon cannot be explained at the moment. As tumor induction by subcutaneous injection of BOP was performed between weeks 1 and 6 and therapy with Celebrex and Zyflo was administered between weeks 17 and 33, inhibition of tumor growth does not seem to be caused by change of BOP metabolism. According to our results, treatment reduced the concentration of lipid peroxidation in tumor-free pancreas and increased lipid peroxidation in pancreatic tumors. Furthermore, activities of antioxidative enzymes were increased in tumorfree pancreas. Thus, prevention of free radicals might be a mechanism for reduced tumor growth. Accordingly, we intend to determine in further trials whether the missing effect of COX-2 inhibition is dosedependent. At the moment the clinical indication for Celebrex is rheumatic disease. Thus, results of dosis-finding trials are missing in the literature. Furthermore, the relations between oxidative stress, lipid peroxidation and eicosanoid synthesis should be examined. Possibly, selective inhibition of eicosanoid synthesis is a new approach for therapy of pancreatic cancer.
Inhibition of Eicosanoid Synthesis in Pancreatic Cancer
Pancreatology 2002;2:54–60
59
References 1 Gloor B, Todd KE: Diagnostic workup of patients with suspected pancreatic carcinoma. Cancer 1997;79:1780–1786. 2 Warshaw AL, Fernandez-del Castillo C: Pancreatic carcinoma. N Engl J Med 1992;326: 455–465. 3 Amikura K, Kobari M: The time of occurrence of liver metastasis in carcinoma of the pancreas. Int J Pancreatol 1995;17:139–146. 4 Böhming M, Wiedenmann B, Rosewicz S: Therapie des Pankreasadenokarzinoms. Med Klin 1999;94:614–625. 5 Hawes RH, Xiong Q, Waxman I, Chang KJ, Evans DB, Abbruzzese JL: A multispecialty approach to the diagnosis and management of pancreatic cancer. Am J Gastroenterol 2000; 95:17–31. 6 Rice-Evans C, Burdon R: Free radical lipid interactions and their pathological consequences. Prog Lipid Res 1993;32:71–110. 7 Ondrey FG: Arachidonic acid metabolism: A primer for head and neck surgeons. Head Neck 1998;20:334–349. 8 Davies GR, Rampton DS: Eicosanoids: Role in gastrointestinal inflammation and cancer. Eur J Gastroenterol Hepatol 1997;9:1033–1044. 9 Karmali RA: Eicosanoids in neoplasia. Prev Med 1987;16:493–502. 10 Benzie IF: Lipid peroxidation: A rewiev of causes, consequences, measurement and dietary influences. Int J Food Sci Nutr 1996;47: 233–261. 11 Hollan S: Free radicals in health and disease. Haematologia (Budap) 1995;26:177–189. 12 Vainio H, Morgan G: Non-steroidal anti-inflammatory drugs and the chemoprevention of gastrointestinal cancers. Scand J Gastroenterol 1998;33:785–789. 13 Damtew B, Spagnuolo PJ: Tumor cell-endothelial cell interactions: Evidence for roles for lipoxygenase products of arachidonic acid in metastasis. Prostaglandins Leukot Essent Fatty Acids 1997;56:295–300. 14 Subbaramaiah K, Zakim D, Weksler BB, Dannenberg AJ: Inhibition of cyclooxygenase: A novel approach to cancer prevention. Proc Soc Exp Biol Med 1997;216:201–210. 15 Marks F, Fürstenberger G, Müller-Decker K: Metabolic targets of cancer chemoprevention: Interruption of tumor development by inhibitors of arachidonic acid metabolism. Recent Res Cancer Res 1999;151:45–67. 16 Lupulescu AP: Prostaglandins, their inhibitors and cancer. Prostaglandins Leukot Essent Fatty Acids 1996;54:83–94. 17 Ara G, Teicher BA: Cyclooxygenase and lipoxygenase inhibitors in cancer therapy. Prostaglandins Leukot Essent Fatty Acids 1996;54:3– 16. 18 Vane JR, Botting RM: Mechanism of action of anti-inflammatory drugs. Int J Tissue React 1998;20:3–15. 19 UKCCCR Guidelines for the welfare of animals in experimental neoplasia. Cancer Metastasis Rev 1989;8:82–88.
60
20 Wenger FA, Kilian M, Jaccobi CA, Schimke I, Guski H, Müller JM: Does ·-linolenic acid in combination with linolenic acid influence liver metastasis and hepatic lipid peroxidation in BOP-induced pancreatic cancer in Syrian hamsters? Prostaglandins Leukot Essent Fatty Acids 2000;62:329–334. 21 Wenger FA, Jacobi CA, Kilian M, Zieren J, Zieren HU, Müller JM: Does dietary ·-linolenic acid promote liver metastasis pancreatic carcinoma initiated by BOP in Syrian hamsters? Ann Nutr Metabol 1999;43:121–126. 22 Meijers M, van Garderen-Hoetmer A, Lamers CB, Rovati LC, Jansen JB, Woutersen RA: Role of cholecystokinin in the development of BOP-induced pancreatic lesions in hamsters. Carcinogenesis 1990;11:2223–2226. 23 Ohkawa H, Ohishi N, Yagi K: Assay for lipid peroxides in animals tissue by thiobarbituric acid reaction. Anal Bioechem 1979;95:351– 358. 24 Janero DR: Malondialdehyde and thiobarbituric acid reactivity as diagnostic indices of lipid peroxidation and peroxidative tissue injury. Free Radical Biol Med 1990;9:515–540. 25 Beauchamp C, Fridovich I: Superoxide dismutase: Improved assay and an assay applicable to acrylamide gels. J Anal Biochem 1971;44:276– 287. 26 Paglia DE, Valentine WN: Studies on the quantitative and qualitative characterization of erythrocytes glutathione peroxidase. J Lab Clin Med 1967;29:143–148. 27 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265–275. 28 Longnecker DS: Experimental cancer of the pancreas. Curr Opin Gastroenterol 1991;7: 731–738. 29 Pour PM, Runge RG, Birt D, Gingell R, Lawson T, Nagel D, Wallave L, Salmasi SZ: Current knowledge of pancreatic carcinogenesis in the hamster and its relevance to the human disease. Cancer 1981;47:1573–1587. 30 Wenger FA, Kilian M, Neumann A, Peter FJ, Jacobi CA, Guski H, Schimke I, Müller JM: Effects of antioxidative vitamins A, C, E on liver metastasis and intrametastatic lipidperoxidation in BOP-induced pancreatic cancer in Syrian hamsters. Carcinogenesis 2001, in press. 31 Arntzen KJ, Brekke OL, Vatten L, Austgulen R: Reduced production of PGE2 and PGF2· from decidual cell cultures supplemented with n-3 polyunsaturated fatty acids. Prostaglandins Lipid Mediat 1998;56:183–195. 32 DuBois RN, Smalley WE: Cyclooxygenase, NSAIDs, and colorectal cancer. J Gastroenterol 1996;31:898–906. 33 Yip-Schneider MT, Barnard DS, Billings SD, Cheng L, Heilmann DK, Lin A, Marshall SJ, Crowell PL, Marshall MS, Sweeney CJ: Cyclooxygenase-2 expression in human pancreatic adenocarcinomas. Carcinogenesis 2000;21: 139–146.
Pancreatology 2002;2:54–60
34 Uefuji K, Ichikura T, Mochizuki H, Shinomiya N: Expression of cyclooxygenase-2 protein in gastric adenocarcinoma. J Surg Oncol 1998;69: 168–172. 35 Wolff H, Saukkonen K, Anttila S, Karjalainen A, Vainio H, Ristimäki A: Expression of cyclooxygenase-2 in human lung carcinoma. Cancer Res 1998;58:4997–5001. 36 Hwang D, Scollard D, Byrne J, Levine E: Expression of cyclooxygenase-1 and cyclooxygenase-2 in human breast cancer. J Natl Cancer Inst 1998;90:455–460. 37 Shiota G, Okubo M, Noumi T, Noguchi N, Oyama K, Takano Y, Yashima K, Kishimoto Y, Kawasaki H: Cyclooxygenase-2 expression in hepatocellular carcinoma. Hepatogastroenterology 1999;46:407–412. 38 Myers CE, Ghosh J: Lipoxygenase ihhibition in prostate cancer. Eur Urol 1999;35:395–398. 39 Tsuji M, Kawano S, Tsuji S, Sawaoka H, Hori M, DuBois RN: Cyclooxygenase regulates angiogenesis induced by colon cancer cells. Cell 1998;93:705–716. 40 De Haan JB, Christiano F, Janello R, Bladier C, Kelner MJ, Kola J: Elevation in the ratio of Cu/Zn superoxide dismutase to glutathione peroxidase activity induced features of cellular senescence and this effect is mediated by hydrogen peroxide. Hum Mol Genet 1996;5:283– 292. 41 Oberley TD, Oberley LW: Invited review: Antioxidant enzyme levels in cancer. Histol Histopathol 1997;12:525–535. 42 Amstadt P, Moret R, Cerutti P: Glutathione peroxidase compensates for the hypersensitivity of Cu/Zn superoxide dismutase overproducers to oxidant stress. J Biol Chem 1994;269: 1606–1609. 43 Hamelin S, Chan AC: Modulation of platelet thromboxane and malondialdehyde by dietary vitamin E and linoleate. Lipids 1983;18:267– 269. 44 Chan AC, Tran K, Pyke DD, Powell WS: Effects of dietary vitamin E on the biosynthesis of 5-lipoxygenase products by rat polymorphonuclear leukocytes PMNL. Biochim Biophys Acta 1989;1005:265–269. 45 Sakamoto W, Fujie K, Nishihira J, Handa H: Effect of vitamin E on arachidonic acid release in rat peritoneal macrophages. Biochim Biophys Acta 1993;1170:296–300. 46 Ding XZ, Kuszynski CA, El-Metwally TH, Adrian TE: Lipoxygenase inhibition-induced apoptosis, morphological changes, and carbonic anhydrase expression in human pancreatic cancer cells. Biochem Biophys Res Commun 1999;266:392–399. 47 Anderson KM, Seed T, Meng J, Ou D, Alrefai WA, Harris JE: 5-Lipoxygenase inhibitors reduce PANC-1 survival: The mode of cell death and synergism of MK886 with Á-linolenic acid. Anticancer Res 1998;18:791–800. 48 Reich R, Martin GR: Identification of arachidonic acid pathways required for the invasive and metastatic activity in malignant tumor cells. Prostaglandins 1996;51:1–17.
Wenger/Kilian/Achucarro/Heinicken/ Schimke/Guski/Jacobi/Müller