- Email: [email protected]
5-lipoxygenase, on events associated with NSAID-induced gastrointestinal inflammation
Prostagiandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417423 © PearsonProfessionalLtd 1997
E f f e c t s of t e p o x a l i n , a dual inhibitor of c y c l o o x y g e n a s e / 5 - 1 i p o x y g e n a s e , on events associated with NSAID. induced gastrointestinal inflammation T. Kirchner, B. Aparicio, D. C. Argentieri, C. Y. Lau, D. M. Ritchie Department of Immunopharmacology, The R.W. Johnson Pharmaceutical Research Institute, Department of Immunopharmacology, Route 202, Raritan, NJ 08869, USA
Summary Prostaglandins and thromboxanes are products of arachidonic acid metabolism via the cyclooxygenase (CO) enzyme and are responsible for the pain and swelling common to sites of inflammation. Non-steroidal antiinflammatory drugs (NSAIDs) inhibit the production of these substances and are used in the treatment of inflammatory diseases such as arthritis. However, one of the major side-effects of NSAID therapy is gastric ulceration. It is possible that inhibition of prostaglandin production and a related increase in the formation of leukotrienes via the 5-1ipoxygenase (5-LO) enzymatic pathway are responsible for attracting inflammatory cells, causing local sites of inflammation and producing ulceration. To determine the effects of 5-LO inhibition on this hypothesis, studies were performed in rats to evaluate the effects of tepoxalin, a dual CO/LO inhibitor on leukotriene B 4 levels in gastric mucosa and neutrophil adhesion in mesenteric venules. In rats, chronic oral administration of an NSAID, indomethacin (2 mg/kg daily over 4 days), resulted in 40% mortality, accompanied by intestinal adhesions and perforations when evaluated 24 h after the fourth dose of drug. Additionally, neutrophil adhesion was increased in the mesenteric venules and cell infiltration was evident in the mesenteric interstitium. These gastrointestinal side-effects were inhibited in a separate group of rats administered tepoxalin (20 mg/kg, p.o) 30 min prior to each daily indomethacin treatment. Further studies were performed to determine tepoxalin's effects on early events associated with NSAID-induced gastrointestinal inflammation, including neutrophil adhesion, lipid peroxide generation and LTB 4 production. Indomethacin (100 mg/kg, p.o.) produced elevated levels of LTB4 in rat gastric mucosa 90 min after administration. Additionally, neutrophil adhesion in mesenteric venules was increased at this dose and with the administration of another NSAID, naproxen. No generation of lipid peroxides was evident in the gastric mucosa at this timepoint. Tepoxalin (up to 400 mg/kg, p.o.) did not have an effects on gastric mucosal LTB4 generation and lipid peroxide levels. A decrease in neutrophil adhesion was observed at the highest dose. In another study, pretreatment with tepoxalin (ED50=7.5 mg/kg, p.o.) or the selective 5-LO inhibitor zileuton (100 mg/kg, p.o.) prevented the increases in gastric mucosal LTB 4 levels and neutrophil adhesion induced by indomethacin (100 mg/kg, p.o.). These data suggest that LO inhibition may play a vital role in the prevention of NSAID-induced gastric inflammation, providing insight into the lack of ulcerogenicity with tepoxalin and new approaches to anti-inflammatory therapy which may prevent gastric side effects. INTRODUCTION
Nonsteroidal anti-inflammatory drugs (NSAIDs) are a class of therapeutic agents which can block the pain and Received 3 May 1995 Accepted 4 May 1995 Correspondence to: Thomas Kirchner, Tel. 001 908 704 4337; Fax. 001 908 526 6469; Email: [email protected].
swelling associated with arthritis in some patients by inhibiting the enzyme cyclooxygenase. 1 Cyclooxygenase is the enzyme responsible for converting arachidonic acid, released from cell membranes upon stimulation, into prostaglandins and thromboxanes. These products play an important role in the pathophysiology of arthritis. However, a major side-effect of NSAID therapy is gastrointestinal inflammation and ulceration, presumably due 417
418
Kirchner et al
to the inhibition of the prostaglandin production needed to maintain the integrity of the gastric mucosa. These adverse effects have been associated with high levels of lipid peroxides 2 and leukotriene B4 (LTB4)3-s in the inflamed gastric mucosa. It has also been reported that indomethacin, an NSAID, increases the number of neutrophils roiling and adhering to the endothelium in the microscopic venules located in the intestinal mesentery and that LTB4 may play a mechanistic role in this process. 6-11 The adhesion and diapedesis of inflammatory ceils, probably neutrophils, z,12-~4into sites of gastric irritation can eventually lead to gastric inflammation and ulceration. Tepoxalin is a dual cyclooxygenase/5-1ipoxygenase inhibitor of arachidonic acid metabolism which demonstrates potent anti-inflammatory activity and a favorable gastrointestinal profile. 15 In addition, tepoxalin does not demonstrate activity in a model of neutrophil adhesion to the endothelial lining of rat mesenteric venules. I~,1z Neutrophil adhesion and subsequent diapedesis into tissues is thought to be an important event in the onset of gastric inflammation. This report discusses some of the possible events involved in NSAID-induced gastric irritation in the rat and offers a possible explanation for tepoxalin's lack of gastrointestinal side-effects in the presence of its CO inhibition.
MATERIALS
AND METHODS
Intravital microscopic measurement of neutrophil adhesion in rat mesenteric venules
Fasted female Sprague-Dawley rats (250-350 g) (Charles River, Wilmington, MA) were orally administered test compound (suspended in 0.5% methyl cellulose) or vehicle in a volume of 4 ml/kg. Sixty minutes following compound administration, the rats were anesthetized with Nembutal (Abbott Laboratories, North Chicago, IL) (250 mg/kg s.c.), and the trachea cannulated to ventilate the animal via a miniature Starling pump. The animals were placed on a heated (37°C) plexiglas platform (University of Calgary) used to adapt a microscope for intravital microscopyfi A midline laparotomy was performed and a segment of mid-jejunum was exteriorized and placed over an optically clear saline-coated cover slip in such a way that a clear area of mesentery was within the focal area. The tissue was suffused with heated (37°C) phosphate buffered saline (PBS) and saline moistened gauze was placed over the exposed intestine to prevent the tissue from dehydrating and to hold the tissue in position. Single unbranched post-capillary venules approximately 25-35 gm in diameter were used for these studies, since they are transparent enough to observe blood flow and leukocyte rolling along the endothelial wall of the
vessel. The preparation was allowed to stabilize for 510 min and at 90 rain post-compound administration, the mesenteric microcirculation was observed and recorded for 5 min using an inverted microscope (Nikon DiaphotTMD) with a 40X objective and a focal length of 14.9 mm (Nikon M Plan 40). A videocamera (Optronics LX-450A) mounted on the microscope projected the image onto a color monitor (Sony 1343MD), and images were recorded with a videocassette recorder (Panasonic AG-1960). Leukocyte adhesion was measured using video imaging analysis of videocassettes with Bioscan Optimas software (Imaging Processing Solutions, Woburn MA). Adherent leukocytes are defined as those white blood cells which are stationary or rolling at a velocity less than that of erythrocytes in the same stream. Adherent cells are expressed as number per 25 gm 2 area of venule. Measurement of lipid peroxides and LTB 4 in rat gastric mucosa
Female Sprague-Dawley rats (250-300 g) were administered vehicle (0.5% methyl cellulose) or test compounds by gavage as described in the leukocyte adhesion model. The animals were then killed in a CO2 chamber 90 rain post-treatment. A fluorometric assay for the measurement of lipid peroxides in gastric mucosa TM was performed. LTB4 concentrations were determined by RIA analysis. The gastric mucosa was removed from each stomach and placed in a tared glass tissue homogenizer. Tissue homogenates were prepared as a ratio of 1 g of wet tissue to 9 ml 1.15% KC1. To measure lipid peroxide concentration, the tissue homogenates were vortexed and 0.2 ml was added to 0.2 ml 8.1% sodium dodecyl sulfate (SDS), 1.5ml of 20% acetic acid adjusted to pH3.5 with 0.1N NaOH, and 1.5 ml 0.8% aqueous solution of thiobarbituric acid (TBA). The mixtures were then brought up to a volume of 4 ml by the addition of 0.6 ml distilled water. A standard solution was prepared by reacting 0.5 nmole tetraethoxypropane with 1 ml 0.8% aqueous TBA in 4 ml distilled water. The reaction mixtures were then heated at 90°C in an oil bath for 60 min. After cooling with tap water, they were treated with 1 ml distilled water and 5 ml of a n-butanol/pyridine mixture (15:1, v/v) and vortexed vigorously. The mixtures were then centrifuged at 4 0 0 0 r p m for 15 min at room temperature. The top organic layers were removed and placed in cuvettes. The absorbance of each sample was measured at 532 nm. Measurement of LTB4 concentration was performed by placing 3 ml of the previously described homogenated gastric mucosa suspended in 1.15% KC1 into tubes. The sample tubes were placed in an ice bath until they were centrifuged at 2 700 g for 10 min. The supernatants were removed and frozen until they were analyzed by RIA for
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
© Pearson Professional Ltd 1997
Effects of tepoxalin on events associated with NSAID-induced gastrointestinal inflammation
Control
Indomethacin (100 mg/Kg, P.O.)
Tepoxalin (100 mg/Kg, P.O.)
Tepoxatin (100 mg/Kg, P.O.) Pretreatment 30 Minutes Prior to Indomethacin (100 mg/Kg, P.O.)
Chronic Treatment (4 Days)
Indomethacin (2 mg/Kg, P.O.)
419
Tepoxalin (20 mg/Kg, P.O.) Pretreatment 30 Minutes Prior to Indomethacin (2 mg/Kg, P.O.)
Fig. 1 Intravitalmicroscopic video images of rat mesenteric venules taken 90 min post-treatment. Each image is representative of its treatment group. Neutrophils are observed as white spheres within the venule.
© Pearson Professional Ltd 1997
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
420
Kirchner et al
LTB4 concentration (Perceptive Diagnostics, Cambridge, M_a).
30
~E =25
.~
"
-~20 ~-15=
RESULTS Effects of chronic administration with anti-inflammatory doses of indomethacin or a combination of tepoxalin and indomethacin on neutrophil adhesion in mesenteric venules and on the gross anatomy of the gastrointestinal tract
To determine the effects of an NSAID after repeated oral treatment, 10 rats were dosed with indomethacin (2 rag/ kg p.o.) once a day for 4 consecutive days, and the number of adherent neutrophils in the mesenteric venules was assessed on the fifth day. By the fifth day, four of the rats died from the chronic treatment. Intestinal adhesions or perforations were observed in all of the rats. The intestines were flexible enough for intravital microscopy in only three of the 10 rats. The number of adherent neutrophils observed in these three rats was increased from 7.2 + 1.4/25 gm 2 area of venule in normal rats to 31.4 + 9.4. Also, a striking difference was observed in the mesenteric interstitium (Fig. 1). Normally, the interstitium is clear and devoid of cells, but in rats chronically adimnistered indomethacin, a marked cellular infiltrate was observed. In contrast, pretreatment with tepoxalin (20 mg/kg p.o.) 30 min prior to the indomethacin treatments completely inhibited neutrophil emigration into the mesentery. Neutrophil adhesion in the venules was also decreased in the tepoxalin/indomethacin group, although the difference was not statistically significant due to the small sample size in the indomethacin group (Table). Since tepoxalin is a dual CO/5-LO inhibitor, these results may indicate an important role for 5-LO in the etiology of NSAID-induced gastrointestinal inflammation.
Intravital microscopic evaluation of the early effects of CO, 5-LO, or dual CO/5-LO inhibition on neutrophil adhesion in rat mesenteric venules
To further investigate the role of 5-LO inhibition in this model, studies were performed to determine the effects Table Effects of tepoxalin pretreatment on deily doses of indomethacin in rats Parameter
Indomethacin
Tepox/Indo
Death Intestinal perforations Intestinal adhesions Neutrophils adherence (neutrophils/25 p.m2)
40% 100% 100% 31.42 -+ 9.38
0% 0% 40% 19.19 +_4.90
¢
/ L
/
/~If"
-
e
Indomethacin Naproxen
-
I
±
Tepoxalin
o
Z
0
I
0
100
i
;
200 300 mg/kg p.o.
I
400
I
500
Fig. 2 Comparison of neutrophil adhesion induced by indomethacin, naproxen, or tepoxalin measured by intravital microscopy. Fasted rats were treated with test compound 90 min prior to the intravital microscopy procedure. Adhesion is defined as leukocytes which are stationary or rolling at a velocity less than that of other blood cells in the stream. Each point represents the mean _+SEM of 10 rats.
of tepoxalin on some of the early events which lead to NSAID-induced gastrointestinal inflammation. Oral administration of the CO inhibitors, indomethacin or naproxen, 90 min prior to measurement of neutrophil adhesion resulted in a dose dependent increase in the number of neutrophils rolling or adhering to the endothelial wall of venules. High doses of these compunds were necessary to achieve effects on neutrophil adhesion at this early time point. All NSAIDs do not demonstrate the same profile on gastrointestinal side effects, therefore it is interesting to note that naproxen is less active than indomethacin in causing neutrophil adhesion, which correlates to their relative activity as ulcerogenic agents in humans. 19 Tepoxalin does not induce these effects when administered up to 400 mg/kg p.o. (Fig. 2). In a second study used to determine the effects of 5-1ipoxygenase inhibition on indomethacin-induced neutrophil adhesion, rats were pretreated 30 min prior to indomethacin (100 mg/kg, p.o.) with 100 mg/kg, p.o. of naproxen, another CO inhibitor, zfleuton, a 5-LO inhibitor, or tepoxalin, a dual CO/5-LO inhibitor. Neutrophil adhesion in the mesenteric venules was observed 90 rain post-indomethacin. The results of this study (Fig. 3) demonstrate that a combination of NSAIDs (naproxen/ indomethacin) produces slightly greater amounts of adhesion than observed with indomethacin alone. However, pretreatment with a 5-1ipoxygenase inhibitor (zfleuton) results in a response similar to that of the control group. Neutrophil adhesion in the tepoxalin pretreated animals is significantly lower than controls. The EDso demonstrated by tepoxalin in this model (7.5 mg/kg p.o.) (Fig. 4) is in the range of its reported lipoxygenase activity.19
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
© Pearson Professional Ltd 1997
Effects of tepoxafin on events associated with NSAID-induced gastrointestinal inflammation
*
~22~
60-
-> 2o 1j
50-
~18~ E16~
[] • [] [] k'q
Control Indomethacin Naproxen/Indo Zileuton/Indo Tepoxalin/Indo
~20b
]
0
* = S i g n i f i c a n t F r o m C o n t r o l P < 0 . 0 5 S t u d e n t ' s t Test
* = Significant from Control P<.O5 Student's t Test
-il00v 8O 2
60 =- 4O ~, 20~
•
~ y
3
6
0
= -5.96 + 63.9log(x) R= 0.93057 ED50 = 7.53 mg/Kg P.O. I
10
I
30
100
Tepoxalin Dose (mg/kg p.o.)
Fig. 4 Dose/Response effect of tepoxalin on indomethacininduced neutrophil adhesion in rat mesenteric venules. Rats were pretreated P.O. with various doses of tepoxalin 30 min before indomethacin (100 mg/kg, p.o.). Each point represents the mean ___SEM of 10 rats.
Effects of orally administered indomethacin, tepoxalin, or a combination on lipid peroxide and LTB4 concentrations in the gastric mucosa
Evaluation of lipid peroxide and LTB4levels in the gastric mucosa 90 min after rats were treated with 100 mg/kg p.o. of either indomethacin, tepoxalin, or a tepoxalin pretreatment followed by indomethacin revealed no effect on lipid peroxide generation. However, LTB4 levels were elevated in the indomethacin group, but not in the tepoxalin or tepoxalin pretreated groups (Fig. 5). This follows the pattern observed with neurotrophil adhesion studies except that tepoxalin decreased the amount of neutrophil adhesion observed in the control group. © Pearson Professional L td 1997
Effect of indomethacin, tepoxalin, or tepoxalin/ indomethacin (Tep/Indo) combination on LTB4 concentration in gastric mucosa 90 minutes post compound administration (100 mg/kg, p.o.). Each bar represents the mean _+SEM of 10 rats.
Fig. 5
Effect of pretreatment with CO or LO inhibitors on indomethacin-induced neutrophil adhesion. All compounds were administered at 100 mg/kg p.o., 30 min prior to indomethacin (Indo). Intravital microscopy was conducted 90 rain after admistration of indomethacin. Each point represents the mean _+SEM of 10 rats.
{
;302
2
Fig. 3
O
¢4044
Untreated Indomethacin Tepoxalin Tep/Indo
10-
~ 6
=if=
[] [] [] []
E
~144
~
421
DISCUSSION
NSMD-induced gastric inflammation has been associated with increased neutrophil adhesion in mesenteric venules ~-8 along with elevated levels of lipid peroxides 2 and LTB43,4,8-1°in the gastric mucosa. The rat models described in this report examine the effects of chronic NSMD therapy and some early events in the etiology of NSMD-induced gastric inflammation which are assessed at 90 min post-treatment Daily administration of indomethacin (2 mg/kg p.o.) over 4 days resulted in intestinal lesions, perforations and mortality. Intravital microscopic evaluation of the surviving rats demonstrated increased neutrophil adhesion and cell infiltration into the mesenteric interstitium. These effects were reduced in another group of rats pretreated each day with a dual CO/5-LO inhibitor, tepoxalin (20 rag/ kg p.o.), which may suggest that 5-1ipoxygenase plays an important role in the development of NSMD-induced gastrointestinal inflammation. To further examine some early mechanisms involved in the etiology of NSAID-induced gastrointestinal inflammation, studies were performed to determine the effects of CO and 5-LO inhibition occurring 90 min post-indomethacin treatment. Oral administration of 100 mg/kg indomethacin in the rat increased neutrophil adhesion in mesenteric post-capillary venules. Tepoxalin, a dual inhibitor of CO and 5-LO, did not produce the neutrophil adhesion observed with specific CO inhibitors when tested up to 400 mg/kg, p.o. In a separate study to determine the effects of tepoxalin on the early events of indomethacin-induced gastrointestinal inflammation, rats were dosed orally with tepoxalin 30 min prior to indomethacin to determine if the increases in neutrophil adhesion in mesenteric veules could be reversed b y 5-LO inhibition. The results indicate that tepoxalin inhibits the effects of indomethacin on
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
422
Kirchner et al
neutrophfl adhesion. The EDs0 established in this experim e n t is 7.5 mg/kg, p.o., w h i c h is within the range of its reported lipoxygenase activity. 15 Zileuton, a specific 5-LO inhibitor, also inhibited indomethacin-induced neutrophil adhesion w h e n tested at a high dose (100 mg/kg p.o.), further implicating 5-LO products as major contributors in the NSAID-induced sequence of events leading to gastric ulceration. It is interesting that w h e n tepoxalin was tested at 100 mg/kg, it significantly decreased neutrophfl adhesion from baseline levels. This effect might be explained b y additional activity observed at higher doses, such as suppression of IL-2 activity. 2° Since elevations in LTB4 and lipid peroxide levels have b e e n reported in inflammed gastric m u c o s a and m a y play a role in the chemotaxis of neutrophils into sites of inflammation, 8,1° a separate experiment was performed where the gastric m u c o s a was r e m o v e d 90 min after administration of indomethacin or the combination of tepoxalin/indomethacin and analyzed to determine concentrations of LTB4 and lipid peroxides. LTB4 levels in the gastric m u c o s a were e n h a n c e d b y inhibition of CO. However, i n d o m e t h a c i n treatment did not alter lipid peroxide levels in the gastric m u c o s a at this time, indicating that lipid peroxides m a y not play a role in the early stages of NSAID-induced gastric inflammation, b u t m a y appear later as products of tissue damage and possibly add to the o n g o i n g inflammatory response. Tepoxalin (100 mg/kg, p.o.) did not significantly elevate LTB4levels in the mucosa. W h e n it was administered 30 min prior to indomethacin, it inhibited this p r o d u c t i o n of LTB4 in the gastric mucosa. These data m a y further indicate that 5-LO products contribute to neutrophil adhesion and gastrointestinal inflammation. Recent reports indicate that 5-LO inhibition alone does n o t prevent NSAID-induced gastric side effects. 21,22 However, tepoxalin appears to inhibit the ulcerogenic events described in these studies within the dose range of its reported 5-LO activity, suggesting that CO and 5-LO products m a y have to be in a certain balance in order to maintain the integrity of the gastrointestinal system. A marginal shift in the production of 5-LO metabolites, such as LTB4, as a result of CO inhibition m a y signal the expression of adhesion molecules. The r e s u k a n t diapedesis of inflammatory cells could lead to ulceration observed during NSAID therapy in some patients. It could that tepoxalin m a y have u n i q u e properties yet to be determined and that the inhibition of indomethacini n d u c e d neutrophil adhesion b y zileuton (100mg/kg, p.o.) m a y also be due to other effects p r o d u c e d at s u c h a high dose. In these studies, tepoxalin appears to have a beneficial effect on b o t h the early events and final pathology of gastric inflammation induced by NSAID treatment w h e n tested at doses within its reported range of 5-LO activity. These studies also suggest that a possible
therapeutic approach to prevent the gastrointestinal side effects associated with current NSAID treatment m a y be to design new agents, such as tepoxalin, which prevent the deleterious effects of 5-LO metabolites while maintaining their inhibition of CO activity.
REFERENCES
1. Vane J. R. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature (London) New Bio11971; 231: 232-235. 2. Yoshikawa T., Naito Y., Kishi A., et al. Role of active oxygen, lipid peroxidation, and antioxidants in the pathogenesis of gastric mucosal injury induced by indomethacin in rats. Gut 1993; 34: 732-737. 3. Ohara A., Sugiyama S., Hoshino H., et al. Reduction of adverse effects of indometacin by anti-allergic drugs in rat stomachs. Arzne#n#teIforschung 1992; 42:1115-1118. 4. Hudson N., Balsitis M., Everitt S., Hawkey C. J. Enhanced gastric mucosal leukotriene B4synthesis in patients taking nonsteroidal anti-inflammatory drugs. Gut 1993; 34: 742-747. 5. Rainsford K. D. Leukotrienes in the pathogenesis of NSA1Dinduced gastric and intestinal mucosal damage. Agents Actions 1993; 39: C24-C26. 6. Bienvenu K., Hernandez L., Granger D. N. Leukocyte adhesion and emigration in inflammation. Ann N YAcad Sci 1992; 664: 388-399. 7. Wallace J. L., Keenan C. M., Granger D. N. Gastric ulceration induced by nonsteroidal anti-inflammatory drugs is a neutrophil-dependent process. Am J Physiol 1990; 259:(3 Part 1), G462-G467. 8. Asako H., Kubes P., Wallace J., Gaginella T., Wolf R. E., Granger D. N. Indomethacin-induced leukocyte adhesion in mesenteric venules: role of lipoxygenase products. Am ] Physiol 1992; 262:(5 Part 1), Gg03-Gg08. 9. Anthony A., Dhillon A. P., Nyg/ird G., et aI. Early histological features of small intestine injury induced by indomethacin. AlimentPharmacoI Ther 1993; 7: 29-40. 10. Nyg~rd G., Anthony A., Piasecki C., et al. Acute indomethacininduced jejunal injury in the rat: early morphological and biochemical changes. Gastroenterology 1994; 106: 567-575. 11. Wallace J. L., McKnight W., Miyasaka M., et al. Role of endothelial adhesion molecules in NSAID-induced gastric mucosal injury. AmJPhysio11993; 265:(5 Part 1), G993-998. 12. Trevethick M. A., Clayton N. M., Strong P., Harman I. M. Do infiltrating neutrophfls contribute to the pathogenesis of indomethacin induced ulceration of the rat gastric antrum? Gut 1993; 34: 156-160. 13. Lawrence M. B. Springer T. A. Neutrophils roll on E-Selectin. ] Immuno11993; 151: 6338-6346. 14. Wallace J. L. Gastric ulceration: critical events at the neutrophilendothelium interface. Can J Physiol Pharmacol 1993; 71: 98-102. 15. Argentieri D. C., Ritchie D. M., Ferro M. P., et al. Tepoxalin: a dual cyclooxygenase/5-1ipoxygenase inhibitor of arachidonic acid metabolism with potent anti-inflammatory activity and a favorable gastrointestinal profile. J Pharmacol Exp Ther 1994; 271: 1399-1408. 16. Wallace J. L., Cirino G., Cicala C., Anderson D. W., Argentieri D., Capetola R. J. Comparison of the ulcerogenic properties of tepoxalin with those of non-steroidal anti-inflammatory drugs (NSAIDs).Agents Actions 1991; 34: 247-250. 17. Wallace J. L., McCafferty D., Carter L., McNight W., Argentieri
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
© Pearson Professional Ltd 1997
Effects of tepoxalin on events associated with NSAID-induced gastrointestinal inflammation
D. C. Tissue selective inhibition of prostaglandin synthesis in rat by tepoxalin: anti- inflammatory without gastropathy. Gastroenterology 1993; 10N: 1630-1636. 18. Ohkawa H., Ohishi N., Yagi K. Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal Biochem 1979; 9N: 351-358. 19. Pemberton R. E., Strand L.J. A review of upper gastrointestinal effects of the newer nonsteroidal anti-inflammatory agents. Am y Dig Dis 1979; 24: 53-64. 20. Zhou L., Ritchie D., Wang E. Y., Barbone A. G., Argentieri D., Lau C. Y. Tepoxalin, a novel immunosuppressive agent with a
© Pearson Professional Ltd 1997
423
different mechansim of action from cyclosporin Am J lmmunol 1994; 1~3: 5026-5037. 21. Ford-Hutchinson A. W., Tagari P., Ching S. V., Anderson C. A., Coleman J. B., Peter C. P. Chronic leukotriene inhibition in the rat fails to modify the toxicological effects of a cyclooxygenase inhibitor. Can J Physiol Pharmacol 1993; 71: 806-810. 22. Trevethick M. A., Clayton N. M., Bahl A. K., Strong P., Harman I. W. Leukotrienes do not contribute to the pathogenesis of indomethacin-induced ulceration of the gastric antrum in the re-fed rat. Agents AcEons 1994; 41: 179-183.
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
E f f e c t s of t e p o x a l i n , a dual inhibitor of c y c l o o x y g e n a s e / 5 - 1 i p o x y g e n a s e , on events associated with NSAID. induced gastrointestinal inflammation T. Kirchner, B. Aparicio, D. C. Argentieri, C. Y. Lau, D. M. Ritchie Department of Immunopharmacology, The R.W. Johnson Pharmaceutical Research Institute, Department of Immunopharmacology, Route 202, Raritan, NJ 08869, USA
Summary Prostaglandins and thromboxanes are products of arachidonic acid metabolism via the cyclooxygenase (CO) enzyme and are responsible for the pain and swelling common to sites of inflammation. Non-steroidal antiinflammatory drugs (NSAIDs) inhibit the production of these substances and are used in the treatment of inflammatory diseases such as arthritis. However, one of the major side-effects of NSAID therapy is gastric ulceration. It is possible that inhibition of prostaglandin production and a related increase in the formation of leukotrienes via the 5-1ipoxygenase (5-LO) enzymatic pathway are responsible for attracting inflammatory cells, causing local sites of inflammation and producing ulceration. To determine the effects of 5-LO inhibition on this hypothesis, studies were performed in rats to evaluate the effects of tepoxalin, a dual CO/LO inhibitor on leukotriene B 4 levels in gastric mucosa and neutrophil adhesion in mesenteric venules. In rats, chronic oral administration of an NSAID, indomethacin (2 mg/kg daily over 4 days), resulted in 40% mortality, accompanied by intestinal adhesions and perforations when evaluated 24 h after the fourth dose of drug. Additionally, neutrophil adhesion was increased in the mesenteric venules and cell infiltration was evident in the mesenteric interstitium. These gastrointestinal side-effects were inhibited in a separate group of rats administered tepoxalin (20 mg/kg, p.o) 30 min prior to each daily indomethacin treatment. Further studies were performed to determine tepoxalin's effects on early events associated with NSAID-induced gastrointestinal inflammation, including neutrophil adhesion, lipid peroxide generation and LTB 4 production. Indomethacin (100 mg/kg, p.o.) produced elevated levels of LTB4 in rat gastric mucosa 90 min after administration. Additionally, neutrophil adhesion in mesenteric venules was increased at this dose and with the administration of another NSAID, naproxen. No generation of lipid peroxides was evident in the gastric mucosa at this timepoint. Tepoxalin (up to 400 mg/kg, p.o.) did not have an effects on gastric mucosal LTB4 generation and lipid peroxide levels. A decrease in neutrophil adhesion was observed at the highest dose. In another study, pretreatment with tepoxalin (ED50=7.5 mg/kg, p.o.) or the selective 5-LO inhibitor zileuton (100 mg/kg, p.o.) prevented the increases in gastric mucosal LTB 4 levels and neutrophil adhesion induced by indomethacin (100 mg/kg, p.o.). These data suggest that LO inhibition may play a vital role in the prevention of NSAID-induced gastric inflammation, providing insight into the lack of ulcerogenicity with tepoxalin and new approaches to anti-inflammatory therapy which may prevent gastric side effects. INTRODUCTION
Nonsteroidal anti-inflammatory drugs (NSAIDs) are a class of therapeutic agents which can block the pain and Received 3 May 1995 Accepted 4 May 1995 Correspondence to: Thomas Kirchner, Tel. 001 908 704 4337; Fax. 001 908 526 6469; Email: [email protected].
swelling associated with arthritis in some patients by inhibiting the enzyme cyclooxygenase. 1 Cyclooxygenase is the enzyme responsible for converting arachidonic acid, released from cell membranes upon stimulation, into prostaglandins and thromboxanes. These products play an important role in the pathophysiology of arthritis. However, a major side-effect of NSAID therapy is gastrointestinal inflammation and ulceration, presumably due 417
418
Kirchner et al
to the inhibition of the prostaglandin production needed to maintain the integrity of the gastric mucosa. These adverse effects have been associated with high levels of lipid peroxides 2 and leukotriene B4 (LTB4)3-s in the inflamed gastric mucosa. It has also been reported that indomethacin, an NSAID, increases the number of neutrophils roiling and adhering to the endothelium in the microscopic venules located in the intestinal mesentery and that LTB4 may play a mechanistic role in this process. 6-11 The adhesion and diapedesis of inflammatory ceils, probably neutrophils, z,12-~4into sites of gastric irritation can eventually lead to gastric inflammation and ulceration. Tepoxalin is a dual cyclooxygenase/5-1ipoxygenase inhibitor of arachidonic acid metabolism which demonstrates potent anti-inflammatory activity and a favorable gastrointestinal profile. 15 In addition, tepoxalin does not demonstrate activity in a model of neutrophil adhesion to the endothelial lining of rat mesenteric venules. I~,1z Neutrophil adhesion and subsequent diapedesis into tissues is thought to be an important event in the onset of gastric inflammation. This report discusses some of the possible events involved in NSAID-induced gastric irritation in the rat and offers a possible explanation for tepoxalin's lack of gastrointestinal side-effects in the presence of its CO inhibition.
MATERIALS
AND METHODS
Intravital microscopic measurement of neutrophil adhesion in rat mesenteric venules
Fasted female Sprague-Dawley rats (250-350 g) (Charles River, Wilmington, MA) were orally administered test compound (suspended in 0.5% methyl cellulose) or vehicle in a volume of 4 ml/kg. Sixty minutes following compound administration, the rats were anesthetized with Nembutal (Abbott Laboratories, North Chicago, IL) (250 mg/kg s.c.), and the trachea cannulated to ventilate the animal via a miniature Starling pump. The animals were placed on a heated (37°C) plexiglas platform (University of Calgary) used to adapt a microscope for intravital microscopyfi A midline laparotomy was performed and a segment of mid-jejunum was exteriorized and placed over an optically clear saline-coated cover slip in such a way that a clear area of mesentery was within the focal area. The tissue was suffused with heated (37°C) phosphate buffered saline (PBS) and saline moistened gauze was placed over the exposed intestine to prevent the tissue from dehydrating and to hold the tissue in position. Single unbranched post-capillary venules approximately 25-35 gm in diameter were used for these studies, since they are transparent enough to observe blood flow and leukocyte rolling along the endothelial wall of the
vessel. The preparation was allowed to stabilize for 510 min and at 90 rain post-compound administration, the mesenteric microcirculation was observed and recorded for 5 min using an inverted microscope (Nikon DiaphotTMD) with a 40X objective and a focal length of 14.9 mm (Nikon M Plan 40). A videocamera (Optronics LX-450A) mounted on the microscope projected the image onto a color monitor (Sony 1343MD), and images were recorded with a videocassette recorder (Panasonic AG-1960). Leukocyte adhesion was measured using video imaging analysis of videocassettes with Bioscan Optimas software (Imaging Processing Solutions, Woburn MA). Adherent leukocytes are defined as those white blood cells which are stationary or rolling at a velocity less than that of erythrocytes in the same stream. Adherent cells are expressed as number per 25 gm 2 area of venule. Measurement of lipid peroxides and LTB 4 in rat gastric mucosa
Female Sprague-Dawley rats (250-300 g) were administered vehicle (0.5% methyl cellulose) or test compounds by gavage as described in the leukocyte adhesion model. The animals were then killed in a CO2 chamber 90 rain post-treatment. A fluorometric assay for the measurement of lipid peroxides in gastric mucosa TM was performed. LTB4 concentrations were determined by RIA analysis. The gastric mucosa was removed from each stomach and placed in a tared glass tissue homogenizer. Tissue homogenates were prepared as a ratio of 1 g of wet tissue to 9 ml 1.15% KC1. To measure lipid peroxide concentration, the tissue homogenates were vortexed and 0.2 ml was added to 0.2 ml 8.1% sodium dodecyl sulfate (SDS), 1.5ml of 20% acetic acid adjusted to pH3.5 with 0.1N NaOH, and 1.5 ml 0.8% aqueous solution of thiobarbituric acid (TBA). The mixtures were then brought up to a volume of 4 ml by the addition of 0.6 ml distilled water. A standard solution was prepared by reacting 0.5 nmole tetraethoxypropane with 1 ml 0.8% aqueous TBA in 4 ml distilled water. The reaction mixtures were then heated at 90°C in an oil bath for 60 min. After cooling with tap water, they were treated with 1 ml distilled water and 5 ml of a n-butanol/pyridine mixture (15:1, v/v) and vortexed vigorously. The mixtures were then centrifuged at 4 0 0 0 r p m for 15 min at room temperature. The top organic layers were removed and placed in cuvettes. The absorbance of each sample was measured at 532 nm. Measurement of LTB4 concentration was performed by placing 3 ml of the previously described homogenated gastric mucosa suspended in 1.15% KC1 into tubes. The sample tubes were placed in an ice bath until they were centrifuged at 2 700 g for 10 min. The supernatants were removed and frozen until they were analyzed by RIA for
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
© Pearson Professional Ltd 1997
Effects of tepoxalin on events associated with NSAID-induced gastrointestinal inflammation
Control
Indomethacin (100 mg/Kg, P.O.)
Tepoxalin (100 mg/Kg, P.O.)
Tepoxatin (100 mg/Kg, P.O.) Pretreatment 30 Minutes Prior to Indomethacin (100 mg/Kg, P.O.)
Chronic Treatment (4 Days)
Indomethacin (2 mg/Kg, P.O.)
419
Tepoxalin (20 mg/Kg, P.O.) Pretreatment 30 Minutes Prior to Indomethacin (2 mg/Kg, P.O.)
Fig. 1 Intravitalmicroscopic video images of rat mesenteric venules taken 90 min post-treatment. Each image is representative of its treatment group. Neutrophils are observed as white spheres within the venule.
© Pearson Professional Ltd 1997
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
420
Kirchner et al
LTB4 concentration (Perceptive Diagnostics, Cambridge, M_a).
30
~E =25
.~
"
-~20 ~-15=
RESULTS Effects of chronic administration with anti-inflammatory doses of indomethacin or a combination of tepoxalin and indomethacin on neutrophil adhesion in mesenteric venules and on the gross anatomy of the gastrointestinal tract
To determine the effects of an NSAID after repeated oral treatment, 10 rats were dosed with indomethacin (2 rag/ kg p.o.) once a day for 4 consecutive days, and the number of adherent neutrophils in the mesenteric venules was assessed on the fifth day. By the fifth day, four of the rats died from the chronic treatment. Intestinal adhesions or perforations were observed in all of the rats. The intestines were flexible enough for intravital microscopy in only three of the 10 rats. The number of adherent neutrophils observed in these three rats was increased from 7.2 + 1.4/25 gm 2 area of venule in normal rats to 31.4 + 9.4. Also, a striking difference was observed in the mesenteric interstitium (Fig. 1). Normally, the interstitium is clear and devoid of cells, but in rats chronically adimnistered indomethacin, a marked cellular infiltrate was observed. In contrast, pretreatment with tepoxalin (20 mg/kg p.o.) 30 min prior to the indomethacin treatments completely inhibited neutrophil emigration into the mesentery. Neutrophil adhesion in the venules was also decreased in the tepoxalin/indomethacin group, although the difference was not statistically significant due to the small sample size in the indomethacin group (Table). Since tepoxalin is a dual CO/5-LO inhibitor, these results may indicate an important role for 5-LO in the etiology of NSAID-induced gastrointestinal inflammation.
Intravital microscopic evaluation of the early effects of CO, 5-LO, or dual CO/5-LO inhibition on neutrophil adhesion in rat mesenteric venules
To further investigate the role of 5-LO inhibition in this model, studies were performed to determine the effects Table Effects of tepoxalin pretreatment on deily doses of indomethacin in rats Parameter
Indomethacin
Tepox/Indo
Death Intestinal perforations Intestinal adhesions Neutrophils adherence (neutrophils/25 p.m2)
40% 100% 100% 31.42 -+ 9.38
0% 0% 40% 19.19 +_4.90
¢
/ L
/
/~If"
-
e
Indomethacin Naproxen
-
I
±
Tepoxalin
o
Z
0
I
0
100
i
;
200 300 mg/kg p.o.
I
400
I
500
Fig. 2 Comparison of neutrophil adhesion induced by indomethacin, naproxen, or tepoxalin measured by intravital microscopy. Fasted rats were treated with test compound 90 min prior to the intravital microscopy procedure. Adhesion is defined as leukocytes which are stationary or rolling at a velocity less than that of other blood cells in the stream. Each point represents the mean _+SEM of 10 rats.
of tepoxalin on some of the early events which lead to NSAID-induced gastrointestinal inflammation. Oral administration of the CO inhibitors, indomethacin or naproxen, 90 min prior to measurement of neutrophil adhesion resulted in a dose dependent increase in the number of neutrophils rolling or adhering to the endothelial wall of venules. High doses of these compunds were necessary to achieve effects on neutrophil adhesion at this early time point. All NSAIDs do not demonstrate the same profile on gastrointestinal side effects, therefore it is interesting to note that naproxen is less active than indomethacin in causing neutrophil adhesion, which correlates to their relative activity as ulcerogenic agents in humans. 19 Tepoxalin does not induce these effects when administered up to 400 mg/kg p.o. (Fig. 2). In a second study used to determine the effects of 5-1ipoxygenase inhibition on indomethacin-induced neutrophil adhesion, rats were pretreated 30 min prior to indomethacin (100 mg/kg, p.o.) with 100 mg/kg, p.o. of naproxen, another CO inhibitor, zfleuton, a 5-LO inhibitor, or tepoxalin, a dual CO/5-LO inhibitor. Neutrophil adhesion in the mesenteric venules was observed 90 rain post-indomethacin. The results of this study (Fig. 3) demonstrate that a combination of NSAIDs (naproxen/ indomethacin) produces slightly greater amounts of adhesion than observed with indomethacin alone. However, pretreatment with a 5-1ipoxygenase inhibitor (zfleuton) results in a response similar to that of the control group. Neutrophil adhesion in the tepoxalin pretreated animals is significantly lower than controls. The EDso demonstrated by tepoxalin in this model (7.5 mg/kg p.o.) (Fig. 4) is in the range of its reported lipoxygenase activity.19
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
© Pearson Professional Ltd 1997
Effects of tepoxafin on events associated with NSAID-induced gastrointestinal inflammation
*
~22~
60-
-> 2o 1j
50-
~18~ E16~
[] • [] [] k'q
Control Indomethacin Naproxen/Indo Zileuton/Indo Tepoxalin/Indo
~20b
]
0
* = S i g n i f i c a n t F r o m C o n t r o l P < 0 . 0 5 S t u d e n t ' s t Test
* = Significant from Control P<.O5 Student's t Test
-il00v 8O 2
60 =- 4O ~, 20~
•
~ y
3
6
0
= -5.96 + 63.9log(x) R= 0.93057 ED50 = 7.53 mg/Kg P.O. I
10
I
30
100
Tepoxalin Dose (mg/kg p.o.)
Fig. 4 Dose/Response effect of tepoxalin on indomethacininduced neutrophil adhesion in rat mesenteric venules. Rats were pretreated P.O. with various doses of tepoxalin 30 min before indomethacin (100 mg/kg, p.o.). Each point represents the mean ___SEM of 10 rats.
Effects of orally administered indomethacin, tepoxalin, or a combination on lipid peroxide and LTB4 concentrations in the gastric mucosa
Evaluation of lipid peroxide and LTB4levels in the gastric mucosa 90 min after rats were treated with 100 mg/kg p.o. of either indomethacin, tepoxalin, or a tepoxalin pretreatment followed by indomethacin revealed no effect on lipid peroxide generation. However, LTB4 levels were elevated in the indomethacin group, but not in the tepoxalin or tepoxalin pretreated groups (Fig. 5). This follows the pattern observed with neurotrophil adhesion studies except that tepoxalin decreased the amount of neutrophil adhesion observed in the control group. © Pearson Professional L td 1997
Effect of indomethacin, tepoxalin, or tepoxalin/ indomethacin (Tep/Indo) combination on LTB4 concentration in gastric mucosa 90 minutes post compound administration (100 mg/kg, p.o.). Each bar represents the mean _+SEM of 10 rats.
Fig. 5
Effect of pretreatment with CO or LO inhibitors on indomethacin-induced neutrophil adhesion. All compounds were administered at 100 mg/kg p.o., 30 min prior to indomethacin (Indo). Intravital microscopy was conducted 90 rain after admistration of indomethacin. Each point represents the mean _+SEM of 10 rats.
{
;302
2
Fig. 3
O
¢4044
Untreated Indomethacin Tepoxalin Tep/Indo
10-
~ 6
=if=
[] [] [] []
E
~144
~
421
DISCUSSION
NSMD-induced gastric inflammation has been associated with increased neutrophil adhesion in mesenteric venules ~-8 along with elevated levels of lipid peroxides 2 and LTB43,4,8-1°in the gastric mucosa. The rat models described in this report examine the effects of chronic NSMD therapy and some early events in the etiology of NSMD-induced gastric inflammation which are assessed at 90 min post-treatment Daily administration of indomethacin (2 mg/kg p.o.) over 4 days resulted in intestinal lesions, perforations and mortality. Intravital microscopic evaluation of the surviving rats demonstrated increased neutrophil adhesion and cell infiltration into the mesenteric interstitium. These effects were reduced in another group of rats pretreated each day with a dual CO/5-LO inhibitor, tepoxalin (20 rag/ kg p.o.), which may suggest that 5-1ipoxygenase plays an important role in the development of NSMD-induced gastrointestinal inflammation. To further examine some early mechanisms involved in the etiology of NSAID-induced gastrointestinal inflammation, studies were performed to determine the effects of CO and 5-LO inhibition occurring 90 min post-indomethacin treatment. Oral administration of 100 mg/kg indomethacin in the rat increased neutrophil adhesion in mesenteric post-capillary venules. Tepoxalin, a dual inhibitor of CO and 5-LO, did not produce the neutrophil adhesion observed with specific CO inhibitors when tested up to 400 mg/kg, p.o. In a separate study to determine the effects of tepoxalin on the early events of indomethacin-induced gastrointestinal inflammation, rats were dosed orally with tepoxalin 30 min prior to indomethacin to determine if the increases in neutrophil adhesion in mesenteric veules could be reversed b y 5-LO inhibition. The results indicate that tepoxalin inhibits the effects of indomethacin on
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
422
Kirchner et al
neutrophfl adhesion. The EDs0 established in this experim e n t is 7.5 mg/kg, p.o., w h i c h is within the range of its reported lipoxygenase activity. 15 Zileuton, a specific 5-LO inhibitor, also inhibited indomethacin-induced neutrophil adhesion w h e n tested at a high dose (100 mg/kg p.o.), further implicating 5-LO products as major contributors in the NSAID-induced sequence of events leading to gastric ulceration. It is interesting that w h e n tepoxalin was tested at 100 mg/kg, it significantly decreased neutrophfl adhesion from baseline levels. This effect might be explained b y additional activity observed at higher doses, such as suppression of IL-2 activity. 2° Since elevations in LTB4 and lipid peroxide levels have b e e n reported in inflammed gastric m u c o s a and m a y play a role in the chemotaxis of neutrophils into sites of inflammation, 8,1° a separate experiment was performed where the gastric m u c o s a was r e m o v e d 90 min after administration of indomethacin or the combination of tepoxalin/indomethacin and analyzed to determine concentrations of LTB4 and lipid peroxides. LTB4 levels in the gastric m u c o s a were e n h a n c e d b y inhibition of CO. However, i n d o m e t h a c i n treatment did not alter lipid peroxide levels in the gastric m u c o s a at this time, indicating that lipid peroxides m a y not play a role in the early stages of NSAID-induced gastric inflammation, b u t m a y appear later as products of tissue damage and possibly add to the o n g o i n g inflammatory response. Tepoxalin (100 mg/kg, p.o.) did not significantly elevate LTB4levels in the mucosa. W h e n it was administered 30 min prior to indomethacin, it inhibited this p r o d u c t i o n of LTB4 in the gastric mucosa. These data m a y further indicate that 5-LO products contribute to neutrophil adhesion and gastrointestinal inflammation. Recent reports indicate that 5-LO inhibition alone does n o t prevent NSAID-induced gastric side effects. 21,22 However, tepoxalin appears to inhibit the ulcerogenic events described in these studies within the dose range of its reported 5-LO activity, suggesting that CO and 5-LO products m a y have to be in a certain balance in order to maintain the integrity of the gastrointestinal system. A marginal shift in the production of 5-LO metabolites, such as LTB4, as a result of CO inhibition m a y signal the expression of adhesion molecules. The r e s u k a n t diapedesis of inflammatory cells could lead to ulceration observed during NSAID therapy in some patients. It could that tepoxalin m a y have u n i q u e properties yet to be determined and that the inhibition of indomethacini n d u c e d neutrophil adhesion b y zileuton (100mg/kg, p.o.) m a y also be due to other effects p r o d u c e d at s u c h a high dose. In these studies, tepoxalin appears to have a beneficial effect on b o t h the early events and final pathology of gastric inflammation induced by NSAID treatment w h e n tested at doses within its reported range of 5-LO activity. These studies also suggest that a possible
therapeutic approach to prevent the gastrointestinal side effects associated with current NSAID treatment m a y be to design new agents, such as tepoxalin, which prevent the deleterious effects of 5-LO metabolites while maintaining their inhibition of CO activity.
REFERENCES
1. Vane J. R. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature (London) New Bio11971; 231: 232-235. 2. Yoshikawa T., Naito Y., Kishi A., et al. Role of active oxygen, lipid peroxidation, and antioxidants in the pathogenesis of gastric mucosal injury induced by indomethacin in rats. Gut 1993; 34: 732-737. 3. Ohara A., Sugiyama S., Hoshino H., et al. Reduction of adverse effects of indometacin by anti-allergic drugs in rat stomachs. Arzne#n#teIforschung 1992; 42:1115-1118. 4. Hudson N., Balsitis M., Everitt S., Hawkey C. J. Enhanced gastric mucosal leukotriene B4synthesis in patients taking nonsteroidal anti-inflammatory drugs. Gut 1993; 34: 742-747. 5. Rainsford K. D. Leukotrienes in the pathogenesis of NSA1Dinduced gastric and intestinal mucosal damage. Agents Actions 1993; 39: C24-C26. 6. Bienvenu K., Hernandez L., Granger D. N. Leukocyte adhesion and emigration in inflammation. Ann N YAcad Sci 1992; 664: 388-399. 7. Wallace J. L., Keenan C. M., Granger D. N. Gastric ulceration induced by nonsteroidal anti-inflammatory drugs is a neutrophil-dependent process. Am J Physiol 1990; 259:(3 Part 1), G462-G467. 8. Asako H., Kubes P., Wallace J., Gaginella T., Wolf R. E., Granger D. N. Indomethacin-induced leukocyte adhesion in mesenteric venules: role of lipoxygenase products. Am ] Physiol 1992; 262:(5 Part 1), Gg03-Gg08. 9. Anthony A., Dhillon A. P., Nyg/ird G., et aI. Early histological features of small intestine injury induced by indomethacin. AlimentPharmacoI Ther 1993; 7: 29-40. 10. Nyg~rd G., Anthony A., Piasecki C., et al. Acute indomethacininduced jejunal injury in the rat: early morphological and biochemical changes. Gastroenterology 1994; 106: 567-575. 11. Wallace J. L., McKnight W., Miyasaka M., et al. Role of endothelial adhesion molecules in NSAID-induced gastric mucosal injury. AmJPhysio11993; 265:(5 Part 1), G993-998. 12. Trevethick M. A., Clayton N. M., Strong P., Harman I. M. Do infiltrating neutrophfls contribute to the pathogenesis of indomethacin induced ulceration of the rat gastric antrum? Gut 1993; 34: 156-160. 13. Lawrence M. B. Springer T. A. Neutrophils roll on E-Selectin. ] Immuno11993; 151: 6338-6346. 14. Wallace J. L. Gastric ulceration: critical events at the neutrophilendothelium interface. Can J Physiol Pharmacol 1993; 71: 98-102. 15. Argentieri D. C., Ritchie D. M., Ferro M. P., et al. Tepoxalin: a dual cyclooxygenase/5-1ipoxygenase inhibitor of arachidonic acid metabolism with potent anti-inflammatory activity and a favorable gastrointestinal profile. J Pharmacol Exp Ther 1994; 271: 1399-1408. 16. Wallace J. L., Cirino G., Cicala C., Anderson D. W., Argentieri D., Capetola R. J. Comparison of the ulcerogenic properties of tepoxalin with those of non-steroidal anti-inflammatory drugs (NSAIDs).Agents Actions 1991; 34: 247-250. 17. Wallace J. L., McCafferty D., Carter L., McNight W., Argentieri
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423
© Pearson Professional Ltd 1997
Effects of tepoxalin on events associated with NSAID-induced gastrointestinal inflammation
D. C. Tissue selective inhibition of prostaglandin synthesis in rat by tepoxalin: anti- inflammatory without gastropathy. Gastroenterology 1993; 10N: 1630-1636. 18. Ohkawa H., Ohishi N., Yagi K. Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal Biochem 1979; 9N: 351-358. 19. Pemberton R. E., Strand L.J. A review of upper gastrointestinal effects of the newer nonsteroidal anti-inflammatory agents. Am y Dig Dis 1979; 24: 53-64. 20. Zhou L., Ritchie D., Wang E. Y., Barbone A. G., Argentieri D., Lau C. Y. Tepoxalin, a novel immunosuppressive agent with a
© Pearson Professional Ltd 1997
423
different mechansim of action from cyclosporin Am J lmmunol 1994; 1~3: 5026-5037. 21. Ford-Hutchinson A. W., Tagari P., Ching S. V., Anderson C. A., Coleman J. B., Peter C. P. Chronic leukotriene inhibition in the rat fails to modify the toxicological effects of a cyclooxygenase inhibitor. Can J Physiol Pharmacol 1993; 71: 806-810. 22. Trevethick M. A., Clayton N. M., Bahl A. K., Strong P., Harman I. W. Leukotrienes do not contribute to the pathogenesis of indomethacin-induced ulceration of the gastric antrum in the re-fed rat. Agents AcEons 1994; 41: 179-183.
Prostaglandins, Leukotrienes and Essential Fatty Acids (1997) 56(6), 417-423