Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992), pp. 1738-1745
Systemic Tumor Necrosis Factor-Alpha Production in Experimental Colitis DAVID R. MACK, MD, ALLAN S. LAU, MD, and PHILIP M. SHERMAN, MD
Tumor necrosis factor-alpha (TNF) is a cytokine released by mononuclear cells in response to inflammation and sepsis. Since the biological effects of TNF are consistent with the systemic and intestinal features of ulcerative colitis, the role of TNF was examined in a rabbit model of chronic colitis. Peripheral blood mononuclear cells were isolated, stimulated with lipopolysaccharide, and cultured supernatants assayed for TNF levels using a cytotoxic assay on mouse fibrosarcoma L929 cells. Basal levels o f TNF production by mononuclear cells from 13 normal rabbits (124.3 units/ml +- 27.1 units/ml, mean +--SE) were not different from nine rabbits with colitis (83.6 units/ml +- 24.4 units/ml, P > 0.05). Treatment with lipopolysaccharide (100 Ixg/ml) induced increased TNF production by mononuclear cells isolated from both normals (672.0 units/ml +- 197.5 units/ml, P < 0.05) and rabbits with colitis (1114.0 units/ml +- 489.6 units/ml, P < 0.05). However, at all lipopolysaccharide concentrations stimulated TNF levels were comparable in experimental and control groups (P > 0.05). In light of the role o f leukotrienes in inflammation, a separate group o f rabbits with colitis was investigated following treatment with an oral leukotriene B 4 receptor antagonist. Serum TNF levels in 15 control rabbits (32.5 units/ml +- 7.6 units/ml, mean +- SE) were not significantly different from rabbits with colitis receiving either leukotriene B 4 receptor antagonist (35.7 units/ml +- 9.2 units/ml, N = 13) or vehicle alone (50.3 units/ml +- 10.2 units/ml, N = 14) (ANOVA, P > 0.05). These data indicate that systemic levels o f TNF are not elevated in this experimental model of chronic colitis. Therefore, other inflammatory mediators with biological functions parallel to those o f T N F are likely to mediate the systemic manifestations o f colitis. KEY WORDS: tumor necrosis factor; inflammatory bowel disease; colitis; intestine; leukotrienes.
Tumor necrosis factor-alpha (TNF) is a cytokine derived from mononuclear leukocytes that is produced in response to inflammation and sepsis. In Manuscript received November 25, 1991; revised manuscript received April 2, 1992; accepted April 16, 1992. From the Divisions of Gastroenterology and Infectious Diseases, Research Institute, The Hospital For Sick Children; and Departments of Pediatrics and Microbiology, University of Toronto, Toronto, Canada. This work was supported by a grant from the Canadian Foundation for Ileitis and Colitis. Dr. Mack was the recipient of a Duncan Gordon .Research Fellowship from The Hospital for Sick Children Foundation. Dr. Sherman is the recipient of a Career Scientist Award from the Ontario Ministry of Health. Address for reprint requests: Dr. Philip Sherman, MD, Division of Gastroenterology (Room 1448), The Hospital For Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G lX8.
1738
addition to its role in the elimination of pathogens, TNF induces necrosis of certain types of tumors in vivo (1) and is cytotoxic for many transformed cell lines in vitro (2). However, high levels of TNF in vivo can be detrimental since they induce metabolic acidosis, wasting, and suppression of hematopoieses. Abnormal expression of this pleiotropic molecule leads to many of the systemic manifestations of chronic disease. These include, for example, appetite suppression, resulting in weight loss (3); suppression of hematopoietic colony formation and increased red blood cell destruction, leading to anemia (4); and acting as an endogenous pyrogen (5). The local and systemic manifestations of inDigestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
0163-2116/92/1100-1738506.50/09 1992PlenumPublishingCorporation
TNF IN EXPERIMENTAL COLITIS f l a m m a t o r y bowel diseases, together with the observation that the intestinal tract contains receptors for T N F (6), suggests that T N F m a y mediate m a n y o f these effects. T o determine if there was a relationship b e t w e e n systemic levels of T N F and chronic colonic inflammation, a well-characterized cytotoxic assay (7) was e m p l o y e d to m e a s u r e serum levels of T N F in a rabbit model of ulcerative colitis. Since recent evid e n c e indicates that lipoxygenase inhibitors suppress the f o r m a t i o n o f T N F both in vitro and in vivo (8), s e r u m T N F levels were m e a s u r e d in rabbits treated with an oral leukotriene B 4 receptor antagonist b o t h prior to and following the induction of colitis.
M A T E R I A L S AND M E T H O D S Measurement of TNF. Serum samples from rabbits were obtained immediately from blood clots and stored at - 7 0 ~ C until assayed for TNF activity. Peripheral blood mononuclear cells were isolated from fresh heparinized blood samples using Ficoll-Hypaque (Pharmacia Lab, Toronto, Ontario, Canada) density gradient centrifugation, as described (9, 10). Briefly, the layer containing mononuclear cells was harvested by aspiration from the density gradient and transferred into sterile PBS at 4~ C in 50 ml polypropylene tubes (#2098 Falcon Plastics, Cockeysville, Maryland). The mixture was then washed four times in PBS to remove residual serum and FicollHypaque. The mononuclear cells were resuspended in RPMI 1640. For induction of TNF synthesis by lipopolysaccharide (LPS), 1 x 106 mononuclear cells cultured in 1.0 ml RPMI 1640 supplemented with fetal calf serum (15%, Bocknek, Rexdale, Ontario, Canada) were treated with 0, 0.1, 1.0, 10 and 100 ixg/ml of Esch erichia coli LPS extract (L-9143, Sigma Chemical Co., St. Louis, Missouri). After incubation with LPS for 18 hr at 37~ C, culture supernatants were assayed for TNF activity. TNF levels were quantitated in a cytotoxicity assay, as described previously (9, 10). Briefly, 5 x 104 mouse fibrosarcoma L929 cells (American Type Culture Collection, Rockville, Maryland), which had previously been seeded into wells of sterile 96-weU microtiter plates (Linbro Flow Lab., McLean, Virginia), were incubated for 24 hr at 37~ C with twofold dilutions of mononuclear cell supernatants or serum in et-MEM medium containing 10% fetal calf serum and 1 ixg/ml of actinomycin D (Boehringer Mannheim Biochemicals, Indianapolis, Indiana). Supernatants of mononuclear cells were then incubated with L929 cells for 24 hr at 37~ C. TNF-induced cytopathic effects were assessed by both light microscopic examination and by staining of the fibrosarcoma cells with a 0.1% solution of crystal violet suspension in 5% ethanol. One unit of TNF was defined as the concentration at which 50% of the L929 cells showed cytopathic effects. In each assay purified recombinant human TNF-a Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
units/mg; Genentech Inc., San Francisco, California) was used as standard. Induction of Colitis in Rabbits. Colitis was induced in male New Zealand white rabbits (Reimans Fur Ranch, Guelph, Ontario, Canada) weighing approximately 3 kg (11), using a model of ulcerative colitis first described by Rabin and Rodgers (12) and Rabin (13). In this model colitis is due to the induction of cellular immunity by sequential challenge with a chemical hapten 1-chloro-2,4dinitrobenzene (DNCB) (Eastman Kodak Co., Rochester, New York) (13). DNCB was dissolved in acetone at a concentration of 2.0 g/ml, and 0.1 ml of the DNCB solution was applied to a shaved flank area of rabbits. The acetone was then evaporated using a stream of warm air. Ten days later, 100 mg of DNCB, dissolved in 0.5 ml of acetone, was applied onto the mucosa of the distal 12-15 cm of colon using an 8-French rubber catheter inserted into the colon under anesthesia with intramuscular ketamine (Parke-Davis, Scarborough, Ontario, Canada) and xylazine (Bayvet, Etobicoke, Ontario, Canada). In six rabbits, intracolonic application of DNCB was administered without prior skin sensitization. Ten days after intracolonic application of DNCB, peripheral blood was collected by venipuncture for isolation of mononuclear cells and the rabbits were sacrificed using intracardiac Euthanyl (MTC Pharmaceuticals, Cambridge, Ontario, Canada) following sedation with intramuscular ketamine and xylazine. To evaluate the role of leukotrienes, an oral leukotriene B 4 receptor antagonist was administered to a separate group of 13 rabbits with DNCB-induced colitis. These rabbits received 20 mg/kg body weight of SC-41930 (14, 15) in a 5-ml suspension of methylcellulose. Using an 8-French feeding catheter, the leukotriene B4 receptor antagonist was administered intragastrically under anesthesia at 18 hr and 1 hr prior to the intracolonic application of DNCB. SC-41930 was then given every 12 hr until sacrifice. As a control for repeated sedation, another group of 14 rabbits with DNCB-induced colitis received 5 ml of methylcellulose alone orogastrically using the same 10-day dosing schedule. As an untreated control group, samples of heparinized blood were obtained from 15 separate healthy rabbits. Statistics. Results are expressed as means + SE. Comparisons between two groups were determined by the two-tailed, unpaired Student's t test. Comparisons between multiple groups were determined using Fisher's protected least significant difference one-factor analysis of variance at 95% confidence intervals (16). (5 X 10 7
RESULTS Characterization of Colitis. W e previously reported (15) that rabbits with D N C B - i n d u c e d colitis developed blood-streaked diarrhea, anorexia, and a decrease in b o d y weight o v e r t h e 10-day period following intracolonic application of the chemical hapten ( - 0 . 0 1 5 kg --- 0.420 kg). H o u s e d u n d e r identical c o n d i t i o n s , u n t r e a t e d r a b b i t s gained weight (+0.230 kg --- 0.170 kg/10 days). T o confirm
1739
MACK ET AL the development of mucosal inflammation in rabbits following DNCB-induced colitis, histology was reviewed on coded sections. Histopathology of coIonic mucosa after challenge of rabbits with DNCB revealed disruption of normal crypt architecture with a loss of mucosal glands and evidence of a chronic inflammatory cellular infiltrate (Figure 1B). Changes of colitis were confined to the distal colon in areas exposed to prior luminal administration of DNCB. There were no striking differences in histologic appearance of colonic mucosa in rabbits sequentially challenged with skin and colonic applications of DNCB compared with rabbits that received only intracolonic DNCB (Figure 1C). To investigate the role of leukotrienes in the pathogenesis of experimental colitis, rabbits were treated with the leukotriene B 4 receptor antagonist SC-41930 and the distal colonic mucosa examined. The results demonstrated preservation of mucous glands (Figure 1D). However, there was a persisting increase in the number of inflammatory cells contained within the lamina propria compared with colonic histology in the vehicle-treated controls (Figure 1D). Serum TNF Levels. To evaluate whether TNF is involved in the systemic abnormalities associated with chronic colitis, the concentration of TNF in the serum of rabbits with DNCB-induced colitis was examined. These results showed that there were no significant differences in the group with colitis (50.3 units/ml -+ 10.1 units/ml, mean -+ SE) compared with levels of TNF in serum samples derived from controls (32.5 units/ml -+ 7.6 units/ml, P > 0.05), (Figure 2). In addition, TNF in serum from SC-41930 treated rabbits with colitis (35.7 units/ml -+ 9.2 units/ml) was not different from levels in serum obtained from both untreated animals with colitis and controls (P > 0.05). Induction of TNF Synthesis by Peripheral Blood Monocytes. As shown in Figure 3, there was no difference in the basal secretion of TNF by mononuclear cells derived from normal rabbits (124.3 units/ml -+ 27.1 units/ml), and rabbits with DNCBinduced colitis (83.6 units/ml +- 24.4 units/ml, P > 0.05). A dose-dependent rise in TNF levels was observed in response to stimulation of monocytes by LPS using cells isolated from both controls and rabbits with colitis (Figure 4). Compared to basal levels, TNF levels measured in response to monocyte stimulation by 100 ixg/ml LPS were elevated in both cells isolated from normals (672.0 units/ml -+ 197.5 units/ml, P < 0.05) and rabbits with colitis 1740
(1114.0 units/ml - 489.6 units/ml, P < 0.05). However, at each concentration of LPS employed there were no differences between the levels of TNF in stimulated monocytes derived from both normals and rabbits with colitis (P > 0.05, Figure 4). In addition, there was no difference between TNF levels in stimulated mononuclear cells derived from rabbits with colitis induced either by sequential skin and colonic application of DNCB or colonic application of DNCB alone. For example, using 100 ~g/ml of LPS, sequential hapten challenge demonstrated TNF levels of 816 units/ml - 465 units/ml (N = 4) in monocyte supernatants in comparison to colonic DNCB challenge alone with stimulated TNF levels of 1181 units/ml - 748 units/ml (N = 5, P > 0.05). DISCUSSION In this study, levels of biologically active TNF present in the serum of animals with DNCBinduced colitis were not elevated compared to TNF levels in normal rabbit serum. The effects of a leukotriene B 4 receptor antagonist were examined because recent evidence indicates that leukotriene B4 is an important proinflammatory mediator of both human ulcerative colitis (17, 18) and animal models of colitis (19-21). Oral lipoxygenase B4 inhibitors have recently been demonstrated to reduce colonic inflammation in animal models of colitis (14, 22) and in humans with ulcerative colitis (23). In addition, a recent report showed that lipoxygenase inhibitors decrease the production of TNF (8). Therefore, to investigate the role of leukotriene B 4 in DNCB-induced colitis, serum was also collected from rabbits with DNCB-induced colitis that had received the oral lipoxygenase B 4 receptor antagonist SC-41930 (15). However, there was no difference in the level of TNF activity present in these serum samples compared to controls and animals with untreated colitis. A quantitative stimulation of TNF production was present using mononuclear cells derived from both normal and experimental groups of rabbits. This finding supported the reliability and sensitivity of the bioassay employed to measure TNF. However, there were no differences between groups in TNF basal production or stimulation of TNF in response to exogenous lipopolysaccharide. These data suggest that neither up-regulation of TNF production from prior activation of monocytes nor down-regulation (for example, from prior stimulaDigestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
T N F IN E X P E R I M E N T A L COLITIS
Fig 1. Histopathology in dinitrochlorobenzene-induced colitis. Compared with histology of distal colons in normal animals (A), colonic mucosa in rabbits challenged with cutaneous and intracolonic applications of the chemical hapten DNCB showed loss of crypts, branching of the remaining crypts, and a chronic inflammatory cell infiltrate in the lamina propria (B) (15). Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
1741
MACK ET AL
Fig 1. Continued. Similar histologic findings were present in colonic mucosa of rabbits receiving intracolonic application of DNCB alone (hematoxylin and eosin stain) (C). Colonic mucosa derived from rabbits with sequential DNCB-induced colitis and also treated with the oral leukotriene B4 receptor antagonist SC-41930 showed relative preservation of crypts and the absence of crypt branching but continued presence of an inflammatory cell infiltrate in the lamina propria (D).
1742
Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
TNF IN EXPERIMENTAL COLITIS
1500
/
E
"E
1ooo
.~_
i S00(b
o r
h
Z
0
Normal
m
Colitis (untreated)
Colitis (treated)
Fig 2. Serum levels of tumor necrosis factor-alpha. Levels of biologically active TNF in serum samples derived from control rabbits (N = 15), rabbits with DNCB-induced colitis (N --- 14), and animals with colitis and treated with the oral leukotriene B 4 receptor antagonist SC-41930 (N = 13). All of the rabbits had skin sensitization followed by intracolonic application of DNCB. Results are expressed as units of TNF activity (--+SE). There were no significant differences between levels of TNF in the three groups (ANOVA, P > 0.05).
tion of TNF-producing cells) had occurred in rabbits in response to the DNCB-induced colitis. Lipopolysaccharide was also employed because it is known to stimulate monocyte production of TNF, but it does not stimulate production of a structurally related glycoprotein produced by lymphocytes that is variously referred to as lymphotoxin and tumor necrosis factor-beta (7). Production of TNF can be determined using bioassays that evaluate cytotoxic activities, assays which measure suppression of lipoprotein lipase activity in adipocyte cultures, immunoassays, and
150E E 7
g lOOE E
~
50-
U_ Z
o
Normal
Colitis
Fig 3. Levels of tumor necrosis factor-alpha produced by peripheral blood monocytes. Basal levels of TNF in supernatants of mononuclear cells purified from blood of normal rabbits (N = 13) and rabbits following DNCB-induced colitis (N = 9). Five of nine rabbits with colitis received intracolonic DNCB not preceded by skin sensitization. Results are expressed as units of TNF activity (--+$E). There was no significant difference between the two groups (Student's t test, P > 0.05). Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
0
l
i
0.1 1 1'0 logLPSConcentration(mg/ml)
i
100
Fig 4. Peripheral blood monocyte production of tumor necrosis factor-alpha in response to lipopolysaccharide, and TNF levels in mononuclear cell supernatants following stimulation of monocytes with varying concentrations of exogenous lipopolysaccharide (LPS). LPS stimulated a dose-dependent rise in TNF release in mononuclear cells derived from both rabbits with colitis (open squares) and normals (solid squares). However, at each concentration of LPS there was no difference in levels of TNF in the supernatants of mononuclear cells derived from normal rabbits compared to animals with DNCB-induced colitis (Student's t test, P > 0.05).
by measurements of mRNA using both Northern blotting and polymerase chain reactions (7). Whereas immunoassays and RNA assays may recognize inactive, denatured, aggregated, or fragmented TNF, a cytotoxicity assay measures only intact, biologically active T N F (7, 24). The cytotoxicity assay employed in this study used recombinant human TNF as a positive control because there is 80% homology in the amino acid sequence of TNF of human and lapine origins (2). As serial twofold dilutions were made to compare levels of TNF activity, only fourfold or greater increases were considered different. In this study an animal model of colitis was examined at a time point when there is histological evidence of alterations in the colonic mucosa that are consistent with chronic inflammation (15). In addition, both the intestinal and systemic features that followed the induction of colitis in rabbits resembled many of the symptoms that are associated with chronic idiopathic ulcerative colitis in humans (25). These included anorexia, weight loss, and blood-streaked semiformed stools. For this study serum and mononuclear cells were collected for TNF determinations. TNF levels were not measured in animals sacrificed at earlier time points as our interest was in the systemic, chronic features of colitis. It might be possible that T N F has a role in
1743
MACK ET AL the initial phases of inflammation, but this study shows that TNF is not produced at sustained levels. For this study two methods were employed to induce colitis in rabbits. As in previous studies (13-15), some animals received an initial skin application of DNCB, followed 10 days later by the intracolonic application of DNCB. In other rabbits colitis was induced by a single intracolonic application of DNCB. The histology of distal colons that had come into direct luminal contact with the DNCB showed identical features in both groups. Thus, the initial skin sensitization with DNCB is not an essential step in either the induction of colitis in rabbits or the development of histologic evidence of chronic mucosal inflammation. Similarly, Morris et al (26) described the induction of a chronic colitis in rats after a single colonic application of trinitrochlorobenzene sulfonic acid suspended in ethanol. Serum levels of TNF have been measured by immunoassay in children with chronic inflammatory bowel diseases with variable results (27, 28). Recent studies have also examined the presence of TNF in intestinal tissue and levels of TNF in supernatants of mononuclear cells extracted from intestinal mucosa of patients with chronic inflammatory bowel diseases. Using both biological assays and specific oligonucleotide probes for TNF activity, these studies failed to reveal an elevation in TNF levels in inflamed tissue (29, 30). In contrast, levels of immunologic activity and mRNA for other cytokines, such as interleukin-I and interleukin-6, were elevated compared to control intestine (29, 31). Measured by both bioassay of supernatants from intestinal-derived mononuclear cells and Northern blot analysis of mRNA extracted from colonic mucosa, induction of colitis in rodents using enemas containing trinitrochlorobenzene sulfonic acid suspended in ethanol results in increased TNF levels (32). Trinitrobenzene sulfonic acid is a compound related to dinitrochlorobenzene. It also appears to act as a chemical hapten in mediating induction of experimental colitis (26). The timing of cytokine determination during the course of intestinal mucosal inflammation may be a critical factor in whether or not alterations in TNF levels are detected. For example, Beagley et al (32) measured levels of cytokines in the acute phase of inflammation after trinitrochlorobenzene sulfonic acid challenge. In contrast, in the present study, determination of TNF levels was undertaken at a time point when histologic examination of colonic mucosa revealed evidence of chronic inflammatory changes (15).
1744
In summary, using a lapine model of DNCBinduced chronic colitis we have shown that TNF is not a mediator of the systemic features of colitis. We postulate that another proinflammatory mediator with biological functions closely related to those of TNF--for example, interleukin-1 (33-35)--may be responsible for the systemic manifestations of chronic colitis. ACKNOWLEDGMENTS The authors thank Dr. Timothy Gaginella and Searle Pharmaceuticals, Skokie, Illinois, for providing the leukotriene B4 receptor antagonist SC-41930.
REFERENCES i. Carswell EA, Old LJ, Kassel RL, Green S, Fiore N, Williamson B: An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci USA 72:3666-3670, 1975 2. Tracey KJ, Vlassara H, Cerami A: Cachectin/tumour necrosis factor. Lancet 1:1122-1126, 1989 3. Beutler B, Cerami A: The history, properties, and biological effects of cachectin. Biochemistry 27:7575-7582, 1988 4. Tracey KJ, Wei H, Manogue KR, Fong Y, Hesse DG, Nguyen HT, Kuo GC, Beutler B, Cotran RS, Cerami A, Lowry SF: Cachectin/tumour necrosis factor induces cachexia, anemia and inflammation. J Exp Med 167:12111227, 1988 5. Dinarello CA, Cannon JG, WolffSM, Bernheim HA, Beutler B, Cerami A, Figari IS, Palladino MA Jr, O'Connor JV: Tumour necrosis factor (cachectin) is an endogenous pyrogen and induces the production of interleukin- 1. J Exp Med 163:1433-1450, 1986 6. Beutler BA, Milsark IW, Cerami A: Cachectin/tumor necrosis factor: Production, distribution and metabolic fate in vivo. J Immunol 135:3972-3977, 1985 7. Meager A, Leung H, Woolley J: Assays for tumour necrosis factor and related cytokines. J Immunol Methods 116:1-17, 1989 8. Schade UF, Ernst M, Reinke M, Wolter DT: Lipoxygenase inhibitors suppress formation of tumor necrosis factor in vitro and in vivo. Biochem Biophys Res Commun 159:748754, 1989 9. Lau AS, Read SE, Williams BRG: Downregulation of interferon alpha but not gamma receptor expression in vivo in the acquired immunodeficiency syndrome. J Clin Invest 82:1415-1421, 1988 10. Lau AS, Livesey JS: Endotoxin induction of tumor necrosis factor is enhanced by acid-labile interferon-alpha in acquired immunodeficiency syndrome. J Clin Invest 84:738-743, 1989 11. Mack D, Sherman P: Mucin isolated from rabbit colon inhibits in vitro binding of Escherichia coli RDEC-1. Infect Immun 59:1015-1023, 1991 12. Rabin BS, Rogers SJ: A cell-mediated immune model of inflammatory bowel disease in the rabbit. Gastroenterology 75:29-33, 1978 13. Rabin BS: Immunologic model of inflammatory bowel disease. Am J Pathol 99:253-256, 1980 Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
T N F IN E X P E R I M E N T A L COLITIS 14. Fretland DJ, Levin S, Tsai BS, Djuric SW, Widomski DL, Zemaitis JM, Shone RL, Bauer RF: The effect of leukotriene B4 receptor antagonist SC-41930 on acetic acid-induced colonic inflammation. Agents Actions 27:395-397, 1989 15. Mack DR, Gaginella TS, Sherman PM: Effect of colonic inflammation on mucin inhibition of Escherichia coli RDEC-1 binding in vitro. Gastroenterology 102:1199-12!1, 1992 16. Winer BJ: Statistical Principles in Experimental Design. New York, McGraw-Hill, 1971 17. Lauritsen K, Laursen LS, Bukhave K, Rask-Madsen J: In vivo profiles of eicosanoids in ulcerative colitis, Crohn's colitis, and Clostridium d ~ c i l e colitis. Gastroenterology 95:11-17, 1988 18. Sharon P, Stenson WF: Enhanced synthesis of leukotriene B 4 by colonic mucosa in inflammatory bowel disease. Gastroenterology 86:453-460, 1984 19. Sharon P, Stenson WF: Metabolism of arachidonic acid in acetic acid colitis in rats. Similarity to human inflammatory bowel disease. Gastroenterology 88:55-63, 1985 20. Zipser RD, Nast CC, Lee M, Kao HW, Duke R: In vivo production of leukotriene B 4 and leukotriene C4 in rabbit colitis. Relationship to inflammation. Gastroenterology 92:33-39, 1987 21. Rachmilewitz D, Simon PL, Schwartz LW, Griswold DE, Fondacaro JD, Wasserman MA: Inflammatory mediators of experimental colitis in rats. Gastroenterology 97:326-337, 1989 22. Wallace JL, Keenan CM: An orally active inhibitor of leukotriene synthesis accelerates healing in a rat model of colitis. Am J Physiol 258:G527-G534, 1990 23. Laursen LS, Naesdal J, Bukhave K, Lauritsen K, RaskMadsen J: Selective 5-1ipoxygenase inhibition in ulcerative colitis. Lancet 335:683-685, 1990 24. Fomsgaard A, Worsaae H, Bendtzen K: Detection of tumor necrosis factor from lipopolysaccharide-stimulated human mononuclear cells by enzyme-linked immunosorbent assay and cytotoxicity bioassay. Scand J Immunol 27:143-147, 1988
Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
25. Kirsner JB, Shorter RG: Recent developments in "nonspecific" inflammatory bowel disease. N Engl J Med 306:775785, 1982 26. Morris GP, Beck PL, Herridge MS, Depew WT, Szewczuk MR, Wallace JL: Hapten-induced model of chronic inflammation and ulceration in the rat colon. Gastroenterology 96:795-803, 1989 27. Hyams JS, Treem WR, Eddy E, Wyzga N, Moore RE: Tumor necrosis factor-a is not elevated in children with inflammatory bowel disease. J Pediatr Gastroenterol Nutr 12:233-236, 1991 28. Murch SH, Lamkin VA, Savage MD, Walker-Smith JA, MacDonald I"1": Serum concentrations of tumour necrosis factor a in childhood chronic inflammatory bowel disease. Gut 32:913-917, 1991 29. Isaacs K, Sartor RB, Wang A, Haskill JS: Profiles of cytokine activation in inflammatory bowel disease tissue: Measurement by cDNA amplification. Gastroenterology 98:A455, 1990 30. Mahida YR, Wu K, Lamming CED, Jewell DP, Hawkey CJ: Human colonic tumor necrosis factor-a (TNF-a) production. Gastroenterology 98:A313, 1990 31. Stevens C, Walz G, Zanker B, Singaram C, Lipman M, Strom TB: Interleukin-6 (IL-6), interleukin-I beta (IL-lb) and tumor necrosis factor alpha (TNF-a) expression in inflammatory bowel disease (IBD). Gastroenterology 98:A475, 1990 32. Beagley KW, Cummings OW, Elson CO: Experimentally induced colitis in mice. Gastroenterology 98:A438, 1990 33. Le J, Vilcek J: Tumour necrosis factor and interleukin 1: Cytokines with multiple overlapping biological activities. Lab Invest 56:234-248, 1987 34. Larrick JW, Kunkel SL: The role of tumor necrosis factor and interleukin 1 in the immunoinflammatory response. Pharm Res 5:129-139, 1988 35. Ligumsky M, Simon PL, Karmeli F, Rachmilewitz D: Role of interleukin 1 in inflammatory bowel disease-enhanced production during active disease. Gut 31:686-689, 1990
1745
Systemic Tumor Necrosis Factor-Alpha Production in Experimental Colitis DAVID R. MACK, MD, ALLAN S. LAU, MD, and PHILIP M. SHERMAN, MD
Tumor necrosis factor-alpha (TNF) is a cytokine released by mononuclear cells in response to inflammation and sepsis. Since the biological effects of TNF are consistent with the systemic and intestinal features of ulcerative colitis, the role of TNF was examined in a rabbit model of chronic colitis. Peripheral blood mononuclear cells were isolated, stimulated with lipopolysaccharide, and cultured supernatants assayed for TNF levels using a cytotoxic assay on mouse fibrosarcoma L929 cells. Basal levels o f TNF production by mononuclear cells from 13 normal rabbits (124.3 units/ml +- 27.1 units/ml, mean +--SE) were not different from nine rabbits with colitis (83.6 units/ml +- 24.4 units/ml, P > 0.05). Treatment with lipopolysaccharide (100 Ixg/ml) induced increased TNF production by mononuclear cells isolated from both normals (672.0 units/ml +- 197.5 units/ml, P < 0.05) and rabbits with colitis (1114.0 units/ml +- 489.6 units/ml, P < 0.05). However, at all lipopolysaccharide concentrations stimulated TNF levels were comparable in experimental and control groups (P > 0.05). In light of the role o f leukotrienes in inflammation, a separate group o f rabbits with colitis was investigated following treatment with an oral leukotriene B 4 receptor antagonist. Serum TNF levels in 15 control rabbits (32.5 units/ml +- 7.6 units/ml, mean +- SE) were not significantly different from rabbits with colitis receiving either leukotriene B 4 receptor antagonist (35.7 units/ml +- 9.2 units/ml, N = 13) or vehicle alone (50.3 units/ml +- 10.2 units/ml, N = 14) (ANOVA, P > 0.05). These data indicate that systemic levels o f TNF are not elevated in this experimental model of chronic colitis. Therefore, other inflammatory mediators with biological functions parallel to those o f T N F are likely to mediate the systemic manifestations o f colitis. KEY WORDS: tumor necrosis factor; inflammatory bowel disease; colitis; intestine; leukotrienes.
Tumor necrosis factor-alpha (TNF) is a cytokine derived from mononuclear leukocytes that is produced in response to inflammation and sepsis. In Manuscript received November 25, 1991; revised manuscript received April 2, 1992; accepted April 16, 1992. From the Divisions of Gastroenterology and Infectious Diseases, Research Institute, The Hospital For Sick Children; and Departments of Pediatrics and Microbiology, University of Toronto, Toronto, Canada. This work was supported by a grant from the Canadian Foundation for Ileitis and Colitis. Dr. Mack was the recipient of a Duncan Gordon .Research Fellowship from The Hospital for Sick Children Foundation. Dr. Sherman is the recipient of a Career Scientist Award from the Ontario Ministry of Health. Address for reprint requests: Dr. Philip Sherman, MD, Division of Gastroenterology (Room 1448), The Hospital For Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G lX8.
1738
addition to its role in the elimination of pathogens, TNF induces necrosis of certain types of tumors in vivo (1) and is cytotoxic for many transformed cell lines in vitro (2). However, high levels of TNF in vivo can be detrimental since they induce metabolic acidosis, wasting, and suppression of hematopoieses. Abnormal expression of this pleiotropic molecule leads to many of the systemic manifestations of chronic disease. These include, for example, appetite suppression, resulting in weight loss (3); suppression of hematopoietic colony formation and increased red blood cell destruction, leading to anemia (4); and acting as an endogenous pyrogen (5). The local and systemic manifestations of inDigestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
0163-2116/92/1100-1738506.50/09 1992PlenumPublishingCorporation
TNF IN EXPERIMENTAL COLITIS f l a m m a t o r y bowel diseases, together with the observation that the intestinal tract contains receptors for T N F (6), suggests that T N F m a y mediate m a n y o f these effects. T o determine if there was a relationship b e t w e e n systemic levels of T N F and chronic colonic inflammation, a well-characterized cytotoxic assay (7) was e m p l o y e d to m e a s u r e serum levels of T N F in a rabbit model of ulcerative colitis. Since recent evid e n c e indicates that lipoxygenase inhibitors suppress the f o r m a t i o n o f T N F both in vitro and in vivo (8), s e r u m T N F levels were m e a s u r e d in rabbits treated with an oral leukotriene B 4 receptor antagonist b o t h prior to and following the induction of colitis.
M A T E R I A L S AND M E T H O D S Measurement of TNF. Serum samples from rabbits were obtained immediately from blood clots and stored at - 7 0 ~ C until assayed for TNF activity. Peripheral blood mononuclear cells were isolated from fresh heparinized blood samples using Ficoll-Hypaque (Pharmacia Lab, Toronto, Ontario, Canada) density gradient centrifugation, as described (9, 10). Briefly, the layer containing mononuclear cells was harvested by aspiration from the density gradient and transferred into sterile PBS at 4~ C in 50 ml polypropylene tubes (#2098 Falcon Plastics, Cockeysville, Maryland). The mixture was then washed four times in PBS to remove residual serum and FicollHypaque. The mononuclear cells were resuspended in RPMI 1640. For induction of TNF synthesis by lipopolysaccharide (LPS), 1 x 106 mononuclear cells cultured in 1.0 ml RPMI 1640 supplemented with fetal calf serum (15%, Bocknek, Rexdale, Ontario, Canada) were treated with 0, 0.1, 1.0, 10 and 100 ixg/ml of Esch erichia coli LPS extract (L-9143, Sigma Chemical Co., St. Louis, Missouri). After incubation with LPS for 18 hr at 37~ C, culture supernatants were assayed for TNF activity. TNF levels were quantitated in a cytotoxicity assay, as described previously (9, 10). Briefly, 5 x 104 mouse fibrosarcoma L929 cells (American Type Culture Collection, Rockville, Maryland), which had previously been seeded into wells of sterile 96-weU microtiter plates (Linbro Flow Lab., McLean, Virginia), were incubated for 24 hr at 37~ C with twofold dilutions of mononuclear cell supernatants or serum in et-MEM medium containing 10% fetal calf serum and 1 ixg/ml of actinomycin D (Boehringer Mannheim Biochemicals, Indianapolis, Indiana). Supernatants of mononuclear cells were then incubated with L929 cells for 24 hr at 37~ C. TNF-induced cytopathic effects were assessed by both light microscopic examination and by staining of the fibrosarcoma cells with a 0.1% solution of crystal violet suspension in 5% ethanol. One unit of TNF was defined as the concentration at which 50% of the L929 cells showed cytopathic effects. In each assay purified recombinant human TNF-a Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
units/mg; Genentech Inc., San Francisco, California) was used as standard. Induction of Colitis in Rabbits. Colitis was induced in male New Zealand white rabbits (Reimans Fur Ranch, Guelph, Ontario, Canada) weighing approximately 3 kg (11), using a model of ulcerative colitis first described by Rabin and Rodgers (12) and Rabin (13). In this model colitis is due to the induction of cellular immunity by sequential challenge with a chemical hapten 1-chloro-2,4dinitrobenzene (DNCB) (Eastman Kodak Co., Rochester, New York) (13). DNCB was dissolved in acetone at a concentration of 2.0 g/ml, and 0.1 ml of the DNCB solution was applied to a shaved flank area of rabbits. The acetone was then evaporated using a stream of warm air. Ten days later, 100 mg of DNCB, dissolved in 0.5 ml of acetone, was applied onto the mucosa of the distal 12-15 cm of colon using an 8-French rubber catheter inserted into the colon under anesthesia with intramuscular ketamine (Parke-Davis, Scarborough, Ontario, Canada) and xylazine (Bayvet, Etobicoke, Ontario, Canada). In six rabbits, intracolonic application of DNCB was administered without prior skin sensitization. Ten days after intracolonic application of DNCB, peripheral blood was collected by venipuncture for isolation of mononuclear cells and the rabbits were sacrificed using intracardiac Euthanyl (MTC Pharmaceuticals, Cambridge, Ontario, Canada) following sedation with intramuscular ketamine and xylazine. To evaluate the role of leukotrienes, an oral leukotriene B 4 receptor antagonist was administered to a separate group of 13 rabbits with DNCB-induced colitis. These rabbits received 20 mg/kg body weight of SC-41930 (14, 15) in a 5-ml suspension of methylcellulose. Using an 8-French feeding catheter, the leukotriene B4 receptor antagonist was administered intragastrically under anesthesia at 18 hr and 1 hr prior to the intracolonic application of DNCB. SC-41930 was then given every 12 hr until sacrifice. As a control for repeated sedation, another group of 14 rabbits with DNCB-induced colitis received 5 ml of methylcellulose alone orogastrically using the same 10-day dosing schedule. As an untreated control group, samples of heparinized blood were obtained from 15 separate healthy rabbits. Statistics. Results are expressed as means + SE. Comparisons between two groups were determined by the two-tailed, unpaired Student's t test. Comparisons between multiple groups were determined using Fisher's protected least significant difference one-factor analysis of variance at 95% confidence intervals (16). (5 X 10 7
RESULTS Characterization of Colitis. W e previously reported (15) that rabbits with D N C B - i n d u c e d colitis developed blood-streaked diarrhea, anorexia, and a decrease in b o d y weight o v e r t h e 10-day period following intracolonic application of the chemical hapten ( - 0 . 0 1 5 kg --- 0.420 kg). H o u s e d u n d e r identical c o n d i t i o n s , u n t r e a t e d r a b b i t s gained weight (+0.230 kg --- 0.170 kg/10 days). T o confirm
1739
MACK ET AL the development of mucosal inflammation in rabbits following DNCB-induced colitis, histology was reviewed on coded sections. Histopathology of coIonic mucosa after challenge of rabbits with DNCB revealed disruption of normal crypt architecture with a loss of mucosal glands and evidence of a chronic inflammatory cellular infiltrate (Figure 1B). Changes of colitis were confined to the distal colon in areas exposed to prior luminal administration of DNCB. There were no striking differences in histologic appearance of colonic mucosa in rabbits sequentially challenged with skin and colonic applications of DNCB compared with rabbits that received only intracolonic DNCB (Figure 1C). To investigate the role of leukotrienes in the pathogenesis of experimental colitis, rabbits were treated with the leukotriene B 4 receptor antagonist SC-41930 and the distal colonic mucosa examined. The results demonstrated preservation of mucous glands (Figure 1D). However, there was a persisting increase in the number of inflammatory cells contained within the lamina propria compared with colonic histology in the vehicle-treated controls (Figure 1D). Serum TNF Levels. To evaluate whether TNF is involved in the systemic abnormalities associated with chronic colitis, the concentration of TNF in the serum of rabbits with DNCB-induced colitis was examined. These results showed that there were no significant differences in the group with colitis (50.3 units/ml -+ 10.1 units/ml, mean -+ SE) compared with levels of TNF in serum samples derived from controls (32.5 units/ml -+ 7.6 units/ml, P > 0.05), (Figure 2). In addition, TNF in serum from SC-41930 treated rabbits with colitis (35.7 units/ml -+ 9.2 units/ml) was not different from levels in serum obtained from both untreated animals with colitis and controls (P > 0.05). Induction of TNF Synthesis by Peripheral Blood Monocytes. As shown in Figure 3, there was no difference in the basal secretion of TNF by mononuclear cells derived from normal rabbits (124.3 units/ml -+ 27.1 units/ml), and rabbits with DNCBinduced colitis (83.6 units/ml +- 24.4 units/ml, P > 0.05). A dose-dependent rise in TNF levels was observed in response to stimulation of monocytes by LPS using cells isolated from both controls and rabbits with colitis (Figure 4). Compared to basal levels, TNF levels measured in response to monocyte stimulation by 100 ixg/ml LPS were elevated in both cells isolated from normals (672.0 units/ml -+ 197.5 units/ml, P < 0.05) and rabbits with colitis 1740
(1114.0 units/ml - 489.6 units/ml, P < 0.05). However, at each concentration of LPS employed there were no differences between the levels of TNF in stimulated monocytes derived from both normals and rabbits with colitis (P > 0.05, Figure 4). In addition, there was no difference between TNF levels in stimulated mononuclear cells derived from rabbits with colitis induced either by sequential skin and colonic application of DNCB or colonic application of DNCB alone. For example, using 100 ~g/ml of LPS, sequential hapten challenge demonstrated TNF levels of 816 units/ml - 465 units/ml (N = 4) in monocyte supernatants in comparison to colonic DNCB challenge alone with stimulated TNF levels of 1181 units/ml - 748 units/ml (N = 5, P > 0.05). DISCUSSION In this study, levels of biologically active TNF present in the serum of animals with DNCBinduced colitis were not elevated compared to TNF levels in normal rabbit serum. The effects of a leukotriene B 4 receptor antagonist were examined because recent evidence indicates that leukotriene B4 is an important proinflammatory mediator of both human ulcerative colitis (17, 18) and animal models of colitis (19-21). Oral lipoxygenase B4 inhibitors have recently been demonstrated to reduce colonic inflammation in animal models of colitis (14, 22) and in humans with ulcerative colitis (23). In addition, a recent report showed that lipoxygenase inhibitors decrease the production of TNF (8). Therefore, to investigate the role of leukotriene B 4 in DNCB-induced colitis, serum was also collected from rabbits with DNCB-induced colitis that had received the oral lipoxygenase B 4 receptor antagonist SC-41930 (15). However, there was no difference in the level of TNF activity present in these serum samples compared to controls and animals with untreated colitis. A quantitative stimulation of TNF production was present using mononuclear cells derived from both normal and experimental groups of rabbits. This finding supported the reliability and sensitivity of the bioassay employed to measure TNF. However, there were no differences between groups in TNF basal production or stimulation of TNF in response to exogenous lipopolysaccharide. These data suggest that neither up-regulation of TNF production from prior activation of monocytes nor down-regulation (for example, from prior stimulaDigestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
T N F IN E X P E R I M E N T A L COLITIS
Fig 1. Histopathology in dinitrochlorobenzene-induced colitis. Compared with histology of distal colons in normal animals (A), colonic mucosa in rabbits challenged with cutaneous and intracolonic applications of the chemical hapten DNCB showed loss of crypts, branching of the remaining crypts, and a chronic inflammatory cell infiltrate in the lamina propria (B) (15). Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
1741
MACK ET AL
Fig 1. Continued. Similar histologic findings were present in colonic mucosa of rabbits receiving intracolonic application of DNCB alone (hematoxylin and eosin stain) (C). Colonic mucosa derived from rabbits with sequential DNCB-induced colitis and also treated with the oral leukotriene B4 receptor antagonist SC-41930 showed relative preservation of crypts and the absence of crypt branching but continued presence of an inflammatory cell infiltrate in the lamina propria (D).
1742
Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
TNF IN EXPERIMENTAL COLITIS
1500
/
E
"E
1ooo
.~_
i S00(b
o r
h
Z
0
Normal
m
Colitis (untreated)
Colitis (treated)
Fig 2. Serum levels of tumor necrosis factor-alpha. Levels of biologically active TNF in serum samples derived from control rabbits (N = 15), rabbits with DNCB-induced colitis (N --- 14), and animals with colitis and treated with the oral leukotriene B 4 receptor antagonist SC-41930 (N = 13). All of the rabbits had skin sensitization followed by intracolonic application of DNCB. Results are expressed as units of TNF activity (--+SE). There were no significant differences between levels of TNF in the three groups (ANOVA, P > 0.05).
tion of TNF-producing cells) had occurred in rabbits in response to the DNCB-induced colitis. Lipopolysaccharide was also employed because it is known to stimulate monocyte production of TNF, but it does not stimulate production of a structurally related glycoprotein produced by lymphocytes that is variously referred to as lymphotoxin and tumor necrosis factor-beta (7). Production of TNF can be determined using bioassays that evaluate cytotoxic activities, assays which measure suppression of lipoprotein lipase activity in adipocyte cultures, immunoassays, and
150E E 7
g lOOE E
~
50-
U_ Z
o
Normal
Colitis
Fig 3. Levels of tumor necrosis factor-alpha produced by peripheral blood monocytes. Basal levels of TNF in supernatants of mononuclear cells purified from blood of normal rabbits (N = 13) and rabbits following DNCB-induced colitis (N = 9). Five of nine rabbits with colitis received intracolonic DNCB not preceded by skin sensitization. Results are expressed as units of TNF activity (--+$E). There was no significant difference between the two groups (Student's t test, P > 0.05). Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
0
l
i
0.1 1 1'0 logLPSConcentration(mg/ml)
i
100
Fig 4. Peripheral blood monocyte production of tumor necrosis factor-alpha in response to lipopolysaccharide, and TNF levels in mononuclear cell supernatants following stimulation of monocytes with varying concentrations of exogenous lipopolysaccharide (LPS). LPS stimulated a dose-dependent rise in TNF release in mononuclear cells derived from both rabbits with colitis (open squares) and normals (solid squares). However, at each concentration of LPS there was no difference in levels of TNF in the supernatants of mononuclear cells derived from normal rabbits compared to animals with DNCB-induced colitis (Student's t test, P > 0.05).
by measurements of mRNA using both Northern blotting and polymerase chain reactions (7). Whereas immunoassays and RNA assays may recognize inactive, denatured, aggregated, or fragmented TNF, a cytotoxicity assay measures only intact, biologically active T N F (7, 24). The cytotoxicity assay employed in this study used recombinant human TNF as a positive control because there is 80% homology in the amino acid sequence of TNF of human and lapine origins (2). As serial twofold dilutions were made to compare levels of TNF activity, only fourfold or greater increases were considered different. In this study an animal model of colitis was examined at a time point when there is histological evidence of alterations in the colonic mucosa that are consistent with chronic inflammation (15). In addition, both the intestinal and systemic features that followed the induction of colitis in rabbits resembled many of the symptoms that are associated with chronic idiopathic ulcerative colitis in humans (25). These included anorexia, weight loss, and blood-streaked semiformed stools. For this study serum and mononuclear cells were collected for TNF determinations. TNF levels were not measured in animals sacrificed at earlier time points as our interest was in the systemic, chronic features of colitis. It might be possible that T N F has a role in
1743
MACK ET AL the initial phases of inflammation, but this study shows that TNF is not produced at sustained levels. For this study two methods were employed to induce colitis in rabbits. As in previous studies (13-15), some animals received an initial skin application of DNCB, followed 10 days later by the intracolonic application of DNCB. In other rabbits colitis was induced by a single intracolonic application of DNCB. The histology of distal colons that had come into direct luminal contact with the DNCB showed identical features in both groups. Thus, the initial skin sensitization with DNCB is not an essential step in either the induction of colitis in rabbits or the development of histologic evidence of chronic mucosal inflammation. Similarly, Morris et al (26) described the induction of a chronic colitis in rats after a single colonic application of trinitrochlorobenzene sulfonic acid suspended in ethanol. Serum levels of TNF have been measured by immunoassay in children with chronic inflammatory bowel diseases with variable results (27, 28). Recent studies have also examined the presence of TNF in intestinal tissue and levels of TNF in supernatants of mononuclear cells extracted from intestinal mucosa of patients with chronic inflammatory bowel diseases. Using both biological assays and specific oligonucleotide probes for TNF activity, these studies failed to reveal an elevation in TNF levels in inflamed tissue (29, 30). In contrast, levels of immunologic activity and mRNA for other cytokines, such as interleukin-I and interleukin-6, were elevated compared to control intestine (29, 31). Measured by both bioassay of supernatants from intestinal-derived mononuclear cells and Northern blot analysis of mRNA extracted from colonic mucosa, induction of colitis in rodents using enemas containing trinitrochlorobenzene sulfonic acid suspended in ethanol results in increased TNF levels (32). Trinitrobenzene sulfonic acid is a compound related to dinitrochlorobenzene. It also appears to act as a chemical hapten in mediating induction of experimental colitis (26). The timing of cytokine determination during the course of intestinal mucosal inflammation may be a critical factor in whether or not alterations in TNF levels are detected. For example, Beagley et al (32) measured levels of cytokines in the acute phase of inflammation after trinitrochlorobenzene sulfonic acid challenge. In contrast, in the present study, determination of TNF levels was undertaken at a time point when histologic examination of colonic mucosa revealed evidence of chronic inflammatory changes (15).
1744
In summary, using a lapine model of DNCBinduced chronic colitis we have shown that TNF is not a mediator of the systemic features of colitis. We postulate that another proinflammatory mediator with biological functions closely related to those of TNF--for example, interleukin-1 (33-35)--may be responsible for the systemic manifestations of chronic colitis. ACKNOWLEDGMENTS The authors thank Dr. Timothy Gaginella and Searle Pharmaceuticals, Skokie, Illinois, for providing the leukotriene B4 receptor antagonist SC-41930.
REFERENCES i. Carswell EA, Old LJ, Kassel RL, Green S, Fiore N, Williamson B: An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci USA 72:3666-3670, 1975 2. Tracey KJ, Vlassara H, Cerami A: Cachectin/tumour necrosis factor. Lancet 1:1122-1126, 1989 3. Beutler B, Cerami A: The history, properties, and biological effects of cachectin. Biochemistry 27:7575-7582, 1988 4. Tracey KJ, Wei H, Manogue KR, Fong Y, Hesse DG, Nguyen HT, Kuo GC, Beutler B, Cotran RS, Cerami A, Lowry SF: Cachectin/tumour necrosis factor induces cachexia, anemia and inflammation. J Exp Med 167:12111227, 1988 5. Dinarello CA, Cannon JG, WolffSM, Bernheim HA, Beutler B, Cerami A, Figari IS, Palladino MA Jr, O'Connor JV: Tumour necrosis factor (cachectin) is an endogenous pyrogen and induces the production of interleukin- 1. J Exp Med 163:1433-1450, 1986 6. Beutler BA, Milsark IW, Cerami A: Cachectin/tumor necrosis factor: Production, distribution and metabolic fate in vivo. J Immunol 135:3972-3977, 1985 7. Meager A, Leung H, Woolley J: Assays for tumour necrosis factor and related cytokines. J Immunol Methods 116:1-17, 1989 8. Schade UF, Ernst M, Reinke M, Wolter DT: Lipoxygenase inhibitors suppress formation of tumor necrosis factor in vitro and in vivo. Biochem Biophys Res Commun 159:748754, 1989 9. Lau AS, Read SE, Williams BRG: Downregulation of interferon alpha but not gamma receptor expression in vivo in the acquired immunodeficiency syndrome. J Clin Invest 82:1415-1421, 1988 10. Lau AS, Livesey JS: Endotoxin induction of tumor necrosis factor is enhanced by acid-labile interferon-alpha in acquired immunodeficiency syndrome. J Clin Invest 84:738-743, 1989 11. Mack D, Sherman P: Mucin isolated from rabbit colon inhibits in vitro binding of Escherichia coli RDEC-1. Infect Immun 59:1015-1023, 1991 12. Rabin BS, Rogers SJ: A cell-mediated immune model of inflammatory bowel disease in the rabbit. Gastroenterology 75:29-33, 1978 13. Rabin BS: Immunologic model of inflammatory bowel disease. Am J Pathol 99:253-256, 1980 Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
T N F IN E X P E R I M E N T A L COLITIS 14. Fretland DJ, Levin S, Tsai BS, Djuric SW, Widomski DL, Zemaitis JM, Shone RL, Bauer RF: The effect of leukotriene B4 receptor antagonist SC-41930 on acetic acid-induced colonic inflammation. Agents Actions 27:395-397, 1989 15. Mack DR, Gaginella TS, Sherman PM: Effect of colonic inflammation on mucin inhibition of Escherichia coli RDEC-1 binding in vitro. Gastroenterology 102:1199-12!1, 1992 16. Winer BJ: Statistical Principles in Experimental Design. New York, McGraw-Hill, 1971 17. Lauritsen K, Laursen LS, Bukhave K, Rask-Madsen J: In vivo profiles of eicosanoids in ulcerative colitis, Crohn's colitis, and Clostridium d ~ c i l e colitis. Gastroenterology 95:11-17, 1988 18. Sharon P, Stenson WF: Enhanced synthesis of leukotriene B 4 by colonic mucosa in inflammatory bowel disease. Gastroenterology 86:453-460, 1984 19. Sharon P, Stenson WF: Metabolism of arachidonic acid in acetic acid colitis in rats. Similarity to human inflammatory bowel disease. Gastroenterology 88:55-63, 1985 20. Zipser RD, Nast CC, Lee M, Kao HW, Duke R: In vivo production of leukotriene B 4 and leukotriene C4 in rabbit colitis. Relationship to inflammation. Gastroenterology 92:33-39, 1987 21. Rachmilewitz D, Simon PL, Schwartz LW, Griswold DE, Fondacaro JD, Wasserman MA: Inflammatory mediators of experimental colitis in rats. Gastroenterology 97:326-337, 1989 22. Wallace JL, Keenan CM: An orally active inhibitor of leukotriene synthesis accelerates healing in a rat model of colitis. Am J Physiol 258:G527-G534, 1990 23. Laursen LS, Naesdal J, Bukhave K, Lauritsen K, RaskMadsen J: Selective 5-1ipoxygenase inhibition in ulcerative colitis. Lancet 335:683-685, 1990 24. Fomsgaard A, Worsaae H, Bendtzen K: Detection of tumor necrosis factor from lipopolysaccharide-stimulated human mononuclear cells by enzyme-linked immunosorbent assay and cytotoxicity bioassay. Scand J Immunol 27:143-147, 1988
Digestive Diseases and Sciences, Vol. 37, No. 11 (November 1992)
25. Kirsner JB, Shorter RG: Recent developments in "nonspecific" inflammatory bowel disease. N Engl J Med 306:775785, 1982 26. Morris GP, Beck PL, Herridge MS, Depew WT, Szewczuk MR, Wallace JL: Hapten-induced model of chronic inflammation and ulceration in the rat colon. Gastroenterology 96:795-803, 1989 27. Hyams JS, Treem WR, Eddy E, Wyzga N, Moore RE: Tumor necrosis factor-a is not elevated in children with inflammatory bowel disease. J Pediatr Gastroenterol Nutr 12:233-236, 1991 28. Murch SH, Lamkin VA, Savage MD, Walker-Smith JA, MacDonald I"1": Serum concentrations of tumour necrosis factor a in childhood chronic inflammatory bowel disease. Gut 32:913-917, 1991 29. Isaacs K, Sartor RB, Wang A, Haskill JS: Profiles of cytokine activation in inflammatory bowel disease tissue: Measurement by cDNA amplification. Gastroenterology 98:A455, 1990 30. Mahida YR, Wu K, Lamming CED, Jewell DP, Hawkey CJ: Human colonic tumor necrosis factor-a (TNF-a) production. Gastroenterology 98:A313, 1990 31. Stevens C, Walz G, Zanker B, Singaram C, Lipman M, Strom TB: Interleukin-6 (IL-6), interleukin-I beta (IL-lb) and tumor necrosis factor alpha (TNF-a) expression in inflammatory bowel disease (IBD). Gastroenterology 98:A475, 1990 32. Beagley KW, Cummings OW, Elson CO: Experimentally induced colitis in mice. Gastroenterology 98:A438, 1990 33. Le J, Vilcek J: Tumour necrosis factor and interleukin 1: Cytokines with multiple overlapping biological activities. Lab Invest 56:234-248, 1987 34. Larrick JW, Kunkel SL: The role of tumor necrosis factor and interleukin 1 in the immunoinflammatory response. Pharm Res 5:129-139, 1988 35. Ligumsky M, Simon PL, Karmeli F, Rachmilewitz D: Role of interleukin 1 in inflammatory bowel disease-enhanced production during active disease. Gut 31:686-689, 1990
1745
