- Email: [email protected]
Endotoxemia and IL-1β Stimulate Mucosal IL-6 Production in Different Parts of the Gastrointestinal Tract
JOURNAL OF SURGICAL RESEARCH ARTICLE NO.
76, 27–31 (1998)
JR985288
Endotoxemia and IL-1b Stimulate Mucosal IL-6 Production in Different Parts of the Gastrointestinal Tract1 Quan Wang, M.D., Jing Jing Wang, M.D., Steven Boyce, Ph.D., Josef E. Fischer, M.D., and Per-Olof Hasselgren, M.D. Department of Surgery, University of Cincinnati, Cincinnati, Ohio, and Shriners Burns Institute, Cincinnati, Ohio Presented at the Annual Meeting of the Association for Academic Surgery, Dallas, Texas, November 6–8, 1997
matory response to sepsis and endotoxemia [1]. Experiments in our laboratory suggest that mucosal protein synthesis is stimulated during sepsis [2, 3] and that this response at least in part reflects increased protein synthesis in the enterocyte [4]. In other studies, mucosal production of proinflammatory cytokines, including IL-1 [5], TNF [6], and IL-6 [7] was upregulated in septic and endotoxemic rats and mice. Among the cytokines produced in the mucosa, IL6 is particularly important because of its role in the regulation of mucosal protein synthesis during sepsis and endotoxemia [8]. In addition, IL-6 is a significant regulator of acute-phase protein synthesis, mainly in the liver [9] but in intestinal epithelial cells as well [10]. There is evidence that IL-6 may be a mediator of increased mucosal permeability during sepsis and other critical illnesses [11, 12]. The correlation between IL-6 levels and severity of disease further underscores the importance of this cytokine in critical illness [13]. In recent studies from our laboratory, mucosal IL-6 production was increased in the jejunum of septic and endotoxemic mice [7]. It is not known from those studies if other parts of the gastrointestinal tract respond in a similar manner with respect to IL-6 production. In addition, the regulators of mucosal IL-6 production and the cell type(s) accounting for mucosal IL-6 secretion have not been defined. The purpose of the present study was to determine the influence of endotoxemia in mice on mucosal IL-6 production in different parts of the gastrointestinal tract, from the stomach to the colon. In addition, immunohistochemistry was employed to determine in which cell type(s) of the mucosa IL-6 is expressed. Finally, we examined the potential role of TNFa and IL-1b in the regulation of mucosal IL-6 production.
Background. In recent studies, sepsis and endotoxemia were associated with increased IL-6 production in mucosa of the jejunum. We tested the hypothesis that endotoxemia in mice stimulates mucosal IL-6 production in other parts of the gastrointestinal tract as well and that the enterocyte is a source of mucosal IL-6. In addition, we examined the effects of TNFa and IL-1b on mucosal IL-6 production. Materials and Methods. Endotoxin (12.5 mg/kg) was injected subcutaneously in mice. Control mice were injected with a corresponding volume of sterile saline. After 4 h, IL-6 levels were determined in mucosa of stomach, jejunum, ileum, and colon and in plasma and liver. In a second series of experiments, immunohistochemistry was performed of jejunal mucosa to determine in which cell type IL-6 was expressed. Finally, 100 mg/kg of human recombinant TNFa or human recombinant IL-1b was injected intraperitoneally in mice and IL-6 levels were determined in plasma and tissues after 4 h. Results. Endotoxemia resulted in increased mucosal IL-6 levels in small and large bowel but in reduced IL6 levels in gastric mucosa. Immunohistochemistry of jejunal mucosa showed that IL-6 was expressed mainly in the enterocyte and in a few cells of the lamina propria. Treatment of mice with TNFa reduced IL-6 levels in gastric mucosa whereas IL-1b increased IL-6 levels in mucosa of small intestine. Conclusion. Mucosal IL-6 production during endotoxemia is differentially regulated along the gastrointestinal tract. Both TNFa and IL-1b may be involved in the regulation of gastrointestinal IL-6 production during endotoxemia. q 1998 Academic Press Key Words: endotoxemia; intestine; mucosa; cytokines; TNF; IL-1; IL-6.
MATERIALS AND METHODS
INTRODUCTION
Animals and experimental design. Male A/J mice (20–27 g) were purchased from Jackson Laboratories (Bar Harbor, ME) and housed at a temperature of 227C in a room with a 12-h light/dark cycle for 1 week before experiments. The animals had free access to regular chow and water until the time of experiments. Three series of experiments were performed. In the first series of experiments, the influence of endotoxemia on mucosal IL-6 produc-
Recent studies have provided evidence that the gut is an active participant in the metabolic and inflam1 Supported in part by Grant No. 8510 from the Shriners of North America, Tampa, FL.
0022-4804/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
27
AID
JSR 5288
/
6n2f$$$381
06-18-98 22:05:07
srga
28
JOURNAL OF SURGICAL RESEARCH: VOL. 76, NO. 1, APRIL 1998
FIG. 1. IL-6 levels in mucosa of different parts of the gastrointestinal tract, liver, and plasma 4 h after injection of saline or endotoxin in mice. n Å 6 in each group. *P õ 0.05 vs saline.
tion was determined. Mice were injected subcutaneously with 12.5 mg/kg of endotoxin (Escherichia coli D111:B4, Calbiochem Co., La Jolla, CA). Control mice were injected with a corresponding volume (0.5 ml/mouse) of sterile saline. Drinking water was provided but food was withheld after the injection of endotoxin or saline to avoid any influence on metabolic changes caused by differences in food intake between the two groups of mice. Groups of mice were studied 4 h after injection of endotoxin or saline. With mice under pentobarbital anesthesia (40 mg/kg intraperitoneally), blood was collected by heart puncture for determination of plasma concentrations of IL-6. Ten-centimeter segments of jejunum and ileum and the whole colon and stomach were removed. Mucosa was harvested by opening the intestine along the anti-mesenteric border and by scraping the luminal side with a microscope slide. The mucosa was immediately frozen in liquid nitrogen and stored at 0707C until analysis. The left liver lobe was excised, frozen in liquid nitrogen, and stored at 0707C until analysis. The endotoxin dose used here and the time point for metabolic studies were based on previous experiments in which a maximal increase in plasma and jejunal IL-6 levels were seen 4 h after treatment of mice with 10 mg/kg of endotoxin [7] and a maximal increase in mucosal protein synthesis was seen after treatment of mice with 12.5 mg/kg of endotoxin [8]. The second series of experiments was performed to determine in which cell type of the intestinal mucosa IL-6 is expressed. The experimental protocol was identical to that in the first series of experiments. Four hours after endotoxin or saline injection, animals were anesthetized with pentobarbital, the abdomen was opened, and the lumen of the jejunum was flushed with saline. A 0.5-cm segment was excised from the midportion of the jejunum, immersed in OCT embedding medium (Miles, Inc., Elkhart, IN), and placed on a metal board frozen in liquid nitrogen. The samples were stored at 0707C for subsequent immunohistochemical studies. In the third series of experiments, mice were injected intraperitoneally with 100 mg/kg of either human recombinant TNFa or human recombinant IL-1b (Endogen, Woburn, MA). Control mice were injected intraperitoneally with a corresponding volume (0.5 ml/mouse) of solvent (phosphate-buffered saline, pH 7.4). The doses of cytokines used here were similar to those used in a previous report from this laboratory in which injection of the cytokines resulted in increased mucosal protein synthesis [2]. Four hours after injection of cytokine or solvent, animals were anesthetized with pentobarbital and plasma, mucosa, and liver were harvested and processed as in the first series of experiments. All experiments were performed and the animals were cared for according to the National Research Council’s Guide for Care and Use of Laboratory Animals. The experimental methods were approved by the University of Cincinnati Institutional Animal Care and Use Committee.
AID
JSR 5288
/
6n2f$$$382
06-18-98 22:05:07
Measurement of IL-6. IL-6 levels in plasma and tissues were determined with a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Endogen Inc., Woburn, MA). To determine IL-6 in mucosa and liver, tissues were ultrasonicated for 20 s in 1 ml of phosphate-buffered saline (pH 7.4) containing 2 mg/ml each of the protease inhibitors leupeptin, aprotinin, pepstatin A, and antipain (Sigma, St. Louis, MO) and 2 mM phenylmethylsulfonyl fluoride (Sigma). After centrifugation at 12,000g at 47C for 30 min, the supernatants were used for IL-6 determination. The lower limit for detection of IL-6 was 25 pg/ml. Immunohistochemistry. Immunohistochemistry was performed to determine in which cell type(s) of the mucosa IL-6 was present. Five-micrometer sections were cut from the intestinal segments and embedded on poly-L-lysine coated slides. The staining was performed as described previously [14]. The sections were dried at 457C for 10 min and fixed in 100% methanol at 0207C for 10 min. The sections were then blocked with 0.01% avidin (A-9275, Sigma) and 0.001% d-biotin (B-4501) for 15 min each and further blocked with 5% bovine serum albumin in TBS (pH 7.6) for 30 min at room temperature. The primary antibody (20F3), which was the same as used in the ELISA assay described above, was diluted 1:50 and added to the sections, which were then incubated for 2 h at 377C. Biotin-conjugated goat anti-rat IgG (Cappel, West Chester, PA) and fluorescenceconjugated streptavidin (1055097, Boehringer, Mannheim, Germany) were diluted 1:2000 and 1:400, respectively, and used consecutively for further detection. Propidium iodide (500 mg/ml, Sigma) was used as nuclear counterstaining. During staining, Tris-buffer saline (0.05 M, pH 7.6 and 8.2) was used as a buffering system throughout and each step was followed by washes. Negative controls were performed by omitting the primary antibody. The sections were examined with a Nikon Microphoto-FXA (Nikon Corp., Melville, NY) equipped with an epifluorescence illumination system and photographs were generated using Ektachrome 400 ASA professional daylight film. Statistics. Results are presented as means { SEM. Student’s t test and Tukey’s test were used for statistical comparisons.
RESULTS
Endotoxemic mice started to show signs of illness in the form of piloerection and exudate around the eyes and nostril(s) at the end of the 4-h experimental period. There was no mortality among saline- or endotoxin-injected mice, similar to previous reports from our laboratory in which no mortality was noted 16 –24
srga
WANG ET AL.: MUCOSAL IL-6 PRODUCTION
29
FIG. 2. Immunohistochemistry of jejunal mucosa from saline-injected (A and B) and endotoxin-injected (C and D) mice. Negative controls were obtained by omitting the primary anti-IL-6 antibody (B negative control for A; D negative control for C). The green color in A and C represents IL-6. IL-6 was present mainly in the enterocytes and in a few cells in the lamina propria. Bar represents 100 mm.
h after injection of 10 or 12.5 mg/kg of endotoxin in mice [7, 8]. In the first series of experiments, endotoxemia in mice was associated with an approximately sixfold increase in IL-6 levels in mucosa of the jejunum and an approximately twofold increase in IL-6 levels in ileum and colon (Fig. 1). In contrast, IL-6 levels in gastric mucosa were significantly reduced during endotoxemia. Measurements of IL-6 levels in plasma were included to provide a ‘‘positive control’’ for the present experiments because in previous reports sepsis and endotoxemia resulted in increased circulating IL-6 [15, 16]. As expected, pronounced increases in liver and plasma levels of IL-6 were noted in the endotoxemic mice (Fig. 1). Of note is the finding that IL-6 was present in mucosa and liver of saline-injected control mice indicating that IL-6 is constitutively expressed in these tissues. Because IL-6 was measured in mucosal scrapings, it
is not known from the first series of experiments in which cell type(s) of the mucosa IL-6 was present. In the next series of experiments, therefore, immunohistochemistry was performed in order to define the mucosal cell type(s) expressing IL-6. Jejunal mucosa was examined because this part of the gastrointestinal tract showed the most prominent increase in IL-6 levels during endotoxemia (see Fig. 1). Immunohistochemistry showed that IL-6 was present predominantly in the enterocytes although some lamina propria cells also stained for IL-6 (Fig. 2). Endotoxemia did not seem to alter the distribution of IL-6 between different cell types in the mucosa. Although the staining for IL-6 seemed to be stronger in enterocytes of endotoxemic mice than in those of control mice (compare Figs. 2A and 2C), this difference needs to be interpreted with caution because immunohistochemistry is not a quantitative method. In the next series of experiments, mice were injected
AID
srga
JSR 5288
/
6n2f$$$382
06-18-98 22:05:07
30
JOURNAL OF SURGICAL RESEARCH: VOL. 76, NO. 1, APRIL 1998
FIG. 3. IL-6 levels in mucosa of different parts of the gastrointestinal tract, liver, and plasma 4 h after intraperitoneal injection of solvent, TNFa, or IL-1b. n Å 6 in each group. *P õ 0.05 vs solvent.
with TNFa or IL-1b and mucosal IL-6 levels were measured 4 h later. Injection of TNFa did not influence mucosal IL-6 levels in the small or large bowel but significantly reduced IL-6 levels in mucosa of the stomach (Fig. 3). Injection of TNFa increased IL-6 levels in the liver. Surprisingly, plasma IL-6 levels were not significantly affected by TNFa treatment. One reason for this may be the relatively long interval (4 h) between injection of TNFa and measurement of IL-6. In previous reports, peak plasma levels of IL-6 were seen 1 h after TNFa injection in mice [17]. Treatment of mice with IL-1b resulted in increased IL-6 levels in mucosa of jejunum and ileum with no changes observed in mucosa of stomach or colon (Fig. 3). Liver and plasma levels of IL-6 were increased approximately 2- and 200-fold, respectively, following injection of IL-1b.
The result in the present study of increased mucosal IL-6 levels in endotoxemic mice supports the concept that the intestine becomes a source, rather than a target only, of cytokines during sepsis and endotoxemia [1, 5–7]. The present report expanded previous observations of increased IL-6 production in jejunal mucosa [7] by determining the influence of endotoxemia on IL6 levels in different parts of the gastrointestinal tract. We found that endotoxemia in mice was associated with reduced IL-6 levels in gastric mucosa, a marked increase of IL-6 in jejunal mucosa, and a somewhat less pronounced increase in ileum and colon. These observations suggest that mucosal IL-6 production is differentially regulated along the gastrointestinal tract during endotoxemia. Interestingly, a similar differential regulation of mucosal protein synthesis was observed in a recent study from our laboratory in which sepsis in rats resulted in increased mucosal protein synthesis in the small and large intestine with a con-
comitant decrease in protein synthesis in gastric mucosa [3]. The connection between these observations, if any, is not known at present. It may be speculated that the differential regulation of mucosal protein synthesis in different parts of the gastrointestinal tract [3] was secondary to changes in mucosal IL-6 levels because in recent studies we found evidence that IL-6 regulates mucosal protein synthesis during sepsis and endotoxemia [8]. Alternatively, the different responses in the various parts of the gastrointestinal tract of both protein synthesis and IL-6 levels may be caused by the same underlying mechanism. More studies are needed to clarify these questions. It should be noted that although the increased mucosal IL-6 levels noted in the present study were interpreted as indicating increased mucosal production of IL6, this interpretation needs to be done with caution for several reasons. First, it is possible that the increased mucosal IL-6 levels reflected deposition of blood-born IL-6 rather than local production of the cytokine. This is less likely, however, because in a recent study, we found that the increase in IL-6 levels in jejunal mucosa during endotoxemia in mice was associated with increased expression of IL-6 mRNA in the mucosa [7]. In addition, reduced IL-6 levels in gastric mucosa would argue against deposition of blood-born IL-6 as being the predominant mechanism of the changes in mucosal IL-6 levels. Second, actual production rates of IL-6 were not determined in the present study: changes in degradation and/or secretion of IL-6 may also account for changes in mucosal IL-6 steady state levels. Further studies are needed to clarify the exact mechanisms of increased mucosal IL-6 levels during endotoxemia. Results from the second series of experiments, employing immunohistochemistry, suggest that the enterocyte is an important (albeit not the only) source of mucosal IL-6. This observation is in line with recent studies in which IL-6 was expressed in human enterocytes in vivo [18] and was produced by cultured entero-
AID
srga
DISCUSSION
JSR 5288
/
6n2f$$$382
06-18-98 22:05:07
WANG ET AL.: MUCOSAL IL-6 PRODUCTION
cytes in vitro [19]. Because immunohistochemistry was only performed in jejunal mucosa, it is not known if the epithelial cells in the other parts of the gastrointestinal tract express IL-6. It should also be noted that immunohistochemistry was only performed at one time point (4 h) after induction of endotoxemia and it is possible that other cell types in the mucosa produce IL-6 earlier or later in the endotoxemic course. In recent studies we found evidence that treatment of cultured Caco-2 cells, a human intestinal epithelial cell line, with IL-1b stimulated the production of IL-6 [19]. The result in the present study of increased mucosal levels of IL-6 in the jejunum and ileum following treatment of mice with IL-1b suggests that this cytokine may regulate enterocyte IL-6 production in vivo as well. More studies are needed, however, to further test that notion. For example, it will be important to determine if the increased IL-6 levels in mucosa of IL1b-treated mice reflect increased enterocyte production of IL-6. It will also be important to measure mucosal IL-6 levels in endotoxemic mice following treatment with an inhibitor of IL-1, such as IL-1 receptor antagonist. In addition, it is not known if the increased intestinal IL-6 levels following treatment with IL-1b represented a direct or indirect effect of the cytokine. The role of other mediators released by IL-1, such as glucocorticoids [20], in the regulation of mucosal IL-6 production needs to be determined. Finally, even if IL-1b regulates mucosal IL-6 production during endotoxemia, it is not known if this effect is caused by IL-1b locally produced in the mucosa or by circulating IL-1b. In light of previous reports of increased mucosal IL-1 production in endotoxemic mice [5] it is possible that mucosal IL-1b stimulates enterocytes in a paracrine fashion. From a theoretical standpoint, IL-6 produced in the intestine may exert local effects in the mucosa or peripheral effects, including effects in the liver following release of IL-6 into the portal vein. The regulation of mucosal IL-6 production during sepsis and endotoxemia, therefore, has important clinical implications. The results in the present study are significant because they improve our understanding of where in the gastrointestinal tract IL-6 is produced during endotoxemia. It will be important in future studies to determine the role of IL-6 in the regulation of mucosal metabolism and integrity in different parts of the gastrointestinal tract during sepsis and endotoxemia.
1. Deitch, E. A. Multiple organ failure: Pathophysiology and potential future therapy. Ann. Surg. 216: 117, 1992. 2. von Allmen, D., Hasselgren, P. O., Higashiguchi, T., Frederick, J., Zamir, O., and Fischer, J. E. Increased intestinal protein synthesis during sepsis and following the administration of tumor necrosis factor a or interleukin-1a. Biochem. J. 286: 585, 1992. 3. Higashiguchi, T., Noguchi, Y., O’Brien, W., Wagner, K., Fischer, J. E., and Hasselgren, P. O. Effect of sepsis on mucosal protein
JSR 5288
/
6n2f$$$382
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
REFERENCES
AID
4.
06-18-98 22:05:07
19.
20.
31
synthesis in different parts of the gastrointestinal tract in rats. Clin. Sci. 87: 207, 1994. Higashiguchi, T., Noguchi, Y., Meyer, T., Fischer, J. E., Hasselgren, P. O. Protein synthesis in isolated enterocytes from septic or endotoxemic rats: Regulation by glutamine. Clin. Sci. 89: 311, 1995. Mester, M., Tomkins, R. G., Gelfand, J. A., Dinarello, C. A., Burke, J. F., and Clark, B. D. Intestinal production of interleukin-1a during endotoxemia in the mouse. J. Surg. Res. 54: 584, 1993. Nathens, A. B., Ding, J. W., Marshall, J. C., Dibeiro, M., and Rotstein, O. D. The gut as a cytokine generating organ: Small bowel TNF-a production during systemic endotoxemia. Surgical Infection Society 14th Annual Meeting, 1994. [abstract] Meyer, T. A., Wang, J., Tiao, G. M., Ogle, C. K., Fischer, J. E., and Hasselgren, P. O. Sepsis and endotoxemia stimulate intestinal interleukin-6 production. Surgery 118: 336, 1995. Wang, Q., Fischer, J. E., and Hasselgren, P. O. Treatment of endotoxemic mice with anti-interleukin-6 antibody paradoxically increases interleukin-6 levels and stimulates mucosal protein synthesis. Arch. Surg. 132: 82, 1997. Gauldie, J., Richards, C., and Baumann, H. IL-6 and the acute phase reaction. Res. Immunol. 143: 755, 1992. Molmenti, E. P., Ziambaras, T., and Perlmutter, D. H. Evidence for an acute phase response in human intestinal epithelial cells. J. Biol. Chem. 268: 14,116, 1993. Langkamp-Henken, B., Donavan, T. B., Pate, L. M., Maull, C. D., and Kudsk, K. A. Increased intestinal permeability following blunt and penetrating trauma. Crit. Care Med. 23: 660, 1995. Janu, P., Minard, G., and Kudsk, K. Systemic interleukin-6 (IL6) correlates with intestinal permeability. Surg. Forum 47: 7, 1996. Ohzato, H., Yoshizaki, K., Nishimoto, N., Ogata, A., Tagoh, H., Monden, M., Gotoh, M., Kishimoto, T., and Mori, T. Interleukin6 is a new indicator of inflammatory status: detection of serum levels of interleukin-6 and C-reactive protein after surgery. Surgery 111: 201, 1992. Goretsky, M. J., Harriger, D. M., Supp, A. P., Greenhalgh, D. G., and Boyce, S. T. Expression of IL-1a, IL-6, and basic fibroblast growth factor by cultured skin substitutes before and after grafting to full-thickness wounds in athymic mice. J. Trauma 40: 894, 1996. Damas, P., Ledoux, D., Nys, M., Vrindts, Y., DeGroot, D., Franchimont, P., and Lamy, M. Cytokine serum levels during severe sepsis in human: IL-6 as a marker of severity. Ann. Surg. 215: 356, 1992. Barke, R. A., Roy, S., Chapin, R. B., Charbonneau, R., Brady, P. S., and Brady, L. J. Sepsis-induced release of interleukin6 may activate the immediate-early gene program through a hypothalamic–hypophyseal mechanism. Surgery 116: 141, 1994. Shalaby, M. R., Waape, A., Aarden, L., and Espevik, T. Endotoxin, tumor necrosis factor-a and interleukin-1 induce interleukin-6 production in vivo. Clin. Immunol. Immunopathol. 53: 488, 1989. Jones, S. C., Trejdosiewicz, L. K., Bardis, R. W., Howdle, P. D., Axon, A. T. R., Dixon, M. F., and Whicher, J. T. Expression of interleukin-6 by intestinal enterocytes. J. Clin. Pathol. 46: 1097, 1993. Parikh, A. A., Salzman, A. L., Kane, C. D., Fischer, J. E., and Hasselgren, P. O. IL-6 production in human intestinal epithelial cells following stimulation with IL-1b is associated with activation of the transcription factor NFkB. J. Surg. Res. 69: 139, 1997. Sapolsky, R., Rivier, C., Yamamoto, G., Pitsky, P., and Vale, W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin-releasing factor. Science 238: 522, 1991.
srga
76, 27–31 (1998)
JR985288
Endotoxemia and IL-1b Stimulate Mucosal IL-6 Production in Different Parts of the Gastrointestinal Tract1 Quan Wang, M.D., Jing Jing Wang, M.D., Steven Boyce, Ph.D., Josef E. Fischer, M.D., and Per-Olof Hasselgren, M.D. Department of Surgery, University of Cincinnati, Cincinnati, Ohio, and Shriners Burns Institute, Cincinnati, Ohio Presented at the Annual Meeting of the Association for Academic Surgery, Dallas, Texas, November 6–8, 1997
matory response to sepsis and endotoxemia [1]. Experiments in our laboratory suggest that mucosal protein synthesis is stimulated during sepsis [2, 3] and that this response at least in part reflects increased protein synthesis in the enterocyte [4]. In other studies, mucosal production of proinflammatory cytokines, including IL-1 [5], TNF [6], and IL-6 [7] was upregulated in septic and endotoxemic rats and mice. Among the cytokines produced in the mucosa, IL6 is particularly important because of its role in the regulation of mucosal protein synthesis during sepsis and endotoxemia [8]. In addition, IL-6 is a significant regulator of acute-phase protein synthesis, mainly in the liver [9] but in intestinal epithelial cells as well [10]. There is evidence that IL-6 may be a mediator of increased mucosal permeability during sepsis and other critical illnesses [11, 12]. The correlation between IL-6 levels and severity of disease further underscores the importance of this cytokine in critical illness [13]. In recent studies from our laboratory, mucosal IL-6 production was increased in the jejunum of septic and endotoxemic mice [7]. It is not known from those studies if other parts of the gastrointestinal tract respond in a similar manner with respect to IL-6 production. In addition, the regulators of mucosal IL-6 production and the cell type(s) accounting for mucosal IL-6 secretion have not been defined. The purpose of the present study was to determine the influence of endotoxemia in mice on mucosal IL-6 production in different parts of the gastrointestinal tract, from the stomach to the colon. In addition, immunohistochemistry was employed to determine in which cell type(s) of the mucosa IL-6 is expressed. Finally, we examined the potential role of TNFa and IL-1b in the regulation of mucosal IL-6 production.
Background. In recent studies, sepsis and endotoxemia were associated with increased IL-6 production in mucosa of the jejunum. We tested the hypothesis that endotoxemia in mice stimulates mucosal IL-6 production in other parts of the gastrointestinal tract as well and that the enterocyte is a source of mucosal IL-6. In addition, we examined the effects of TNFa and IL-1b on mucosal IL-6 production. Materials and Methods. Endotoxin (12.5 mg/kg) was injected subcutaneously in mice. Control mice were injected with a corresponding volume of sterile saline. After 4 h, IL-6 levels were determined in mucosa of stomach, jejunum, ileum, and colon and in plasma and liver. In a second series of experiments, immunohistochemistry was performed of jejunal mucosa to determine in which cell type IL-6 was expressed. Finally, 100 mg/kg of human recombinant TNFa or human recombinant IL-1b was injected intraperitoneally in mice and IL-6 levels were determined in plasma and tissues after 4 h. Results. Endotoxemia resulted in increased mucosal IL-6 levels in small and large bowel but in reduced IL6 levels in gastric mucosa. Immunohistochemistry of jejunal mucosa showed that IL-6 was expressed mainly in the enterocyte and in a few cells of the lamina propria. Treatment of mice with TNFa reduced IL-6 levels in gastric mucosa whereas IL-1b increased IL-6 levels in mucosa of small intestine. Conclusion. Mucosal IL-6 production during endotoxemia is differentially regulated along the gastrointestinal tract. Both TNFa and IL-1b may be involved in the regulation of gastrointestinal IL-6 production during endotoxemia. q 1998 Academic Press Key Words: endotoxemia; intestine; mucosa; cytokines; TNF; IL-1; IL-6.
MATERIALS AND METHODS
INTRODUCTION
Animals and experimental design. Male A/J mice (20–27 g) were purchased from Jackson Laboratories (Bar Harbor, ME) and housed at a temperature of 227C in a room with a 12-h light/dark cycle for 1 week before experiments. The animals had free access to regular chow and water until the time of experiments. Three series of experiments were performed. In the first series of experiments, the influence of endotoxemia on mucosal IL-6 produc-
Recent studies have provided evidence that the gut is an active participant in the metabolic and inflam1 Supported in part by Grant No. 8510 from the Shriners of North America, Tampa, FL.
0022-4804/98 $25.00 Copyright q 1998 by Academic Press All rights of reproduction in any form reserved.
27
AID
JSR 5288
/
6n2f$$$381
06-18-98 22:05:07
srga
28
JOURNAL OF SURGICAL RESEARCH: VOL. 76, NO. 1, APRIL 1998
FIG. 1. IL-6 levels in mucosa of different parts of the gastrointestinal tract, liver, and plasma 4 h after injection of saline or endotoxin in mice. n Å 6 in each group. *P õ 0.05 vs saline.
tion was determined. Mice were injected subcutaneously with 12.5 mg/kg of endotoxin (Escherichia coli D111:B4, Calbiochem Co., La Jolla, CA). Control mice were injected with a corresponding volume (0.5 ml/mouse) of sterile saline. Drinking water was provided but food was withheld after the injection of endotoxin or saline to avoid any influence on metabolic changes caused by differences in food intake between the two groups of mice. Groups of mice were studied 4 h after injection of endotoxin or saline. With mice under pentobarbital anesthesia (40 mg/kg intraperitoneally), blood was collected by heart puncture for determination of plasma concentrations of IL-6. Ten-centimeter segments of jejunum and ileum and the whole colon and stomach were removed. Mucosa was harvested by opening the intestine along the anti-mesenteric border and by scraping the luminal side with a microscope slide. The mucosa was immediately frozen in liquid nitrogen and stored at 0707C until analysis. The left liver lobe was excised, frozen in liquid nitrogen, and stored at 0707C until analysis. The endotoxin dose used here and the time point for metabolic studies were based on previous experiments in which a maximal increase in plasma and jejunal IL-6 levels were seen 4 h after treatment of mice with 10 mg/kg of endotoxin [7] and a maximal increase in mucosal protein synthesis was seen after treatment of mice with 12.5 mg/kg of endotoxin [8]. The second series of experiments was performed to determine in which cell type of the intestinal mucosa IL-6 is expressed. The experimental protocol was identical to that in the first series of experiments. Four hours after endotoxin or saline injection, animals were anesthetized with pentobarbital, the abdomen was opened, and the lumen of the jejunum was flushed with saline. A 0.5-cm segment was excised from the midportion of the jejunum, immersed in OCT embedding medium (Miles, Inc., Elkhart, IN), and placed on a metal board frozen in liquid nitrogen. The samples were stored at 0707C for subsequent immunohistochemical studies. In the third series of experiments, mice were injected intraperitoneally with 100 mg/kg of either human recombinant TNFa or human recombinant IL-1b (Endogen, Woburn, MA). Control mice were injected intraperitoneally with a corresponding volume (0.5 ml/mouse) of solvent (phosphate-buffered saline, pH 7.4). The doses of cytokines used here were similar to those used in a previous report from this laboratory in which injection of the cytokines resulted in increased mucosal protein synthesis [2]. Four hours after injection of cytokine or solvent, animals were anesthetized with pentobarbital and plasma, mucosa, and liver were harvested and processed as in the first series of experiments. All experiments were performed and the animals were cared for according to the National Research Council’s Guide for Care and Use of Laboratory Animals. The experimental methods were approved by the University of Cincinnati Institutional Animal Care and Use Committee.
AID
JSR 5288
/
6n2f$$$382
06-18-98 22:05:07
Measurement of IL-6. IL-6 levels in plasma and tissues were determined with a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Endogen Inc., Woburn, MA). To determine IL-6 in mucosa and liver, tissues were ultrasonicated for 20 s in 1 ml of phosphate-buffered saline (pH 7.4) containing 2 mg/ml each of the protease inhibitors leupeptin, aprotinin, pepstatin A, and antipain (Sigma, St. Louis, MO) and 2 mM phenylmethylsulfonyl fluoride (Sigma). After centrifugation at 12,000g at 47C for 30 min, the supernatants were used for IL-6 determination. The lower limit for detection of IL-6 was 25 pg/ml. Immunohistochemistry. Immunohistochemistry was performed to determine in which cell type(s) of the mucosa IL-6 was present. Five-micrometer sections were cut from the intestinal segments and embedded on poly-L-lysine coated slides. The staining was performed as described previously [14]. The sections were dried at 457C for 10 min and fixed in 100% methanol at 0207C for 10 min. The sections were then blocked with 0.01% avidin (A-9275, Sigma) and 0.001% d-biotin (B-4501) for 15 min each and further blocked with 5% bovine serum albumin in TBS (pH 7.6) for 30 min at room temperature. The primary antibody (20F3), which was the same as used in the ELISA assay described above, was diluted 1:50 and added to the sections, which were then incubated for 2 h at 377C. Biotin-conjugated goat anti-rat IgG (Cappel, West Chester, PA) and fluorescenceconjugated streptavidin (1055097, Boehringer, Mannheim, Germany) were diluted 1:2000 and 1:400, respectively, and used consecutively for further detection. Propidium iodide (500 mg/ml, Sigma) was used as nuclear counterstaining. During staining, Tris-buffer saline (0.05 M, pH 7.6 and 8.2) was used as a buffering system throughout and each step was followed by washes. Negative controls were performed by omitting the primary antibody. The sections were examined with a Nikon Microphoto-FXA (Nikon Corp., Melville, NY) equipped with an epifluorescence illumination system and photographs were generated using Ektachrome 400 ASA professional daylight film. Statistics. Results are presented as means { SEM. Student’s t test and Tukey’s test were used for statistical comparisons.
RESULTS
Endotoxemic mice started to show signs of illness in the form of piloerection and exudate around the eyes and nostril(s) at the end of the 4-h experimental period. There was no mortality among saline- or endotoxin-injected mice, similar to previous reports from our laboratory in which no mortality was noted 16 –24
srga
WANG ET AL.: MUCOSAL IL-6 PRODUCTION
29
FIG. 2. Immunohistochemistry of jejunal mucosa from saline-injected (A and B) and endotoxin-injected (C and D) mice. Negative controls were obtained by omitting the primary anti-IL-6 antibody (B negative control for A; D negative control for C). The green color in A and C represents IL-6. IL-6 was present mainly in the enterocytes and in a few cells in the lamina propria. Bar represents 100 mm.
h after injection of 10 or 12.5 mg/kg of endotoxin in mice [7, 8]. In the first series of experiments, endotoxemia in mice was associated with an approximately sixfold increase in IL-6 levels in mucosa of the jejunum and an approximately twofold increase in IL-6 levels in ileum and colon (Fig. 1). In contrast, IL-6 levels in gastric mucosa were significantly reduced during endotoxemia. Measurements of IL-6 levels in plasma were included to provide a ‘‘positive control’’ for the present experiments because in previous reports sepsis and endotoxemia resulted in increased circulating IL-6 [15, 16]. As expected, pronounced increases in liver and plasma levels of IL-6 were noted in the endotoxemic mice (Fig. 1). Of note is the finding that IL-6 was present in mucosa and liver of saline-injected control mice indicating that IL-6 is constitutively expressed in these tissues. Because IL-6 was measured in mucosal scrapings, it
is not known from the first series of experiments in which cell type(s) of the mucosa IL-6 was present. In the next series of experiments, therefore, immunohistochemistry was performed in order to define the mucosal cell type(s) expressing IL-6. Jejunal mucosa was examined because this part of the gastrointestinal tract showed the most prominent increase in IL-6 levels during endotoxemia (see Fig. 1). Immunohistochemistry showed that IL-6 was present predominantly in the enterocytes although some lamina propria cells also stained for IL-6 (Fig. 2). Endotoxemia did not seem to alter the distribution of IL-6 between different cell types in the mucosa. Although the staining for IL-6 seemed to be stronger in enterocytes of endotoxemic mice than in those of control mice (compare Figs. 2A and 2C), this difference needs to be interpreted with caution because immunohistochemistry is not a quantitative method. In the next series of experiments, mice were injected
AID
srga
JSR 5288
/
6n2f$$$382
06-18-98 22:05:07
30
JOURNAL OF SURGICAL RESEARCH: VOL. 76, NO. 1, APRIL 1998
FIG. 3. IL-6 levels in mucosa of different parts of the gastrointestinal tract, liver, and plasma 4 h after intraperitoneal injection of solvent, TNFa, or IL-1b. n Å 6 in each group. *P õ 0.05 vs solvent.
with TNFa or IL-1b and mucosal IL-6 levels were measured 4 h later. Injection of TNFa did not influence mucosal IL-6 levels in the small or large bowel but significantly reduced IL-6 levels in mucosa of the stomach (Fig. 3). Injection of TNFa increased IL-6 levels in the liver. Surprisingly, plasma IL-6 levels were not significantly affected by TNFa treatment. One reason for this may be the relatively long interval (4 h) between injection of TNFa and measurement of IL-6. In previous reports, peak plasma levels of IL-6 were seen 1 h after TNFa injection in mice [17]. Treatment of mice with IL-1b resulted in increased IL-6 levels in mucosa of jejunum and ileum with no changes observed in mucosa of stomach or colon (Fig. 3). Liver and plasma levels of IL-6 were increased approximately 2- and 200-fold, respectively, following injection of IL-1b.
The result in the present study of increased mucosal IL-6 levels in endotoxemic mice supports the concept that the intestine becomes a source, rather than a target only, of cytokines during sepsis and endotoxemia [1, 5–7]. The present report expanded previous observations of increased IL-6 production in jejunal mucosa [7] by determining the influence of endotoxemia on IL6 levels in different parts of the gastrointestinal tract. We found that endotoxemia in mice was associated with reduced IL-6 levels in gastric mucosa, a marked increase of IL-6 in jejunal mucosa, and a somewhat less pronounced increase in ileum and colon. These observations suggest that mucosal IL-6 production is differentially regulated along the gastrointestinal tract during endotoxemia. Interestingly, a similar differential regulation of mucosal protein synthesis was observed in a recent study from our laboratory in which sepsis in rats resulted in increased mucosal protein synthesis in the small and large intestine with a con-
comitant decrease in protein synthesis in gastric mucosa [3]. The connection between these observations, if any, is not known at present. It may be speculated that the differential regulation of mucosal protein synthesis in different parts of the gastrointestinal tract [3] was secondary to changes in mucosal IL-6 levels because in recent studies we found evidence that IL-6 regulates mucosal protein synthesis during sepsis and endotoxemia [8]. Alternatively, the different responses in the various parts of the gastrointestinal tract of both protein synthesis and IL-6 levels may be caused by the same underlying mechanism. More studies are needed to clarify these questions. It should be noted that although the increased mucosal IL-6 levels noted in the present study were interpreted as indicating increased mucosal production of IL6, this interpretation needs to be done with caution for several reasons. First, it is possible that the increased mucosal IL-6 levels reflected deposition of blood-born IL-6 rather than local production of the cytokine. This is less likely, however, because in a recent study, we found that the increase in IL-6 levels in jejunal mucosa during endotoxemia in mice was associated with increased expression of IL-6 mRNA in the mucosa [7]. In addition, reduced IL-6 levels in gastric mucosa would argue against deposition of blood-born IL-6 as being the predominant mechanism of the changes in mucosal IL-6 levels. Second, actual production rates of IL-6 were not determined in the present study: changes in degradation and/or secretion of IL-6 may also account for changes in mucosal IL-6 steady state levels. Further studies are needed to clarify the exact mechanisms of increased mucosal IL-6 levels during endotoxemia. Results from the second series of experiments, employing immunohistochemistry, suggest that the enterocyte is an important (albeit not the only) source of mucosal IL-6. This observation is in line with recent studies in which IL-6 was expressed in human enterocytes in vivo [18] and was produced by cultured entero-
AID
srga
DISCUSSION
JSR 5288
/
6n2f$$$382
06-18-98 22:05:07
WANG ET AL.: MUCOSAL IL-6 PRODUCTION
cytes in vitro [19]. Because immunohistochemistry was only performed in jejunal mucosa, it is not known if the epithelial cells in the other parts of the gastrointestinal tract express IL-6. It should also be noted that immunohistochemistry was only performed at one time point (4 h) after induction of endotoxemia and it is possible that other cell types in the mucosa produce IL-6 earlier or later in the endotoxemic course. In recent studies we found evidence that treatment of cultured Caco-2 cells, a human intestinal epithelial cell line, with IL-1b stimulated the production of IL-6 [19]. The result in the present study of increased mucosal levels of IL-6 in the jejunum and ileum following treatment of mice with IL-1b suggests that this cytokine may regulate enterocyte IL-6 production in vivo as well. More studies are needed, however, to further test that notion. For example, it will be important to determine if the increased IL-6 levels in mucosa of IL1b-treated mice reflect increased enterocyte production of IL-6. It will also be important to measure mucosal IL-6 levels in endotoxemic mice following treatment with an inhibitor of IL-1, such as IL-1 receptor antagonist. In addition, it is not known if the increased intestinal IL-6 levels following treatment with IL-1b represented a direct or indirect effect of the cytokine. The role of other mediators released by IL-1, such as glucocorticoids [20], in the regulation of mucosal IL-6 production needs to be determined. Finally, even if IL-1b regulates mucosal IL-6 production during endotoxemia, it is not known if this effect is caused by IL-1b locally produced in the mucosa or by circulating IL-1b. In light of previous reports of increased mucosal IL-1 production in endotoxemic mice [5] it is possible that mucosal IL-1b stimulates enterocytes in a paracrine fashion. From a theoretical standpoint, IL-6 produced in the intestine may exert local effects in the mucosa or peripheral effects, including effects in the liver following release of IL-6 into the portal vein. The regulation of mucosal IL-6 production during sepsis and endotoxemia, therefore, has important clinical implications. The results in the present study are significant because they improve our understanding of where in the gastrointestinal tract IL-6 is produced during endotoxemia. It will be important in future studies to determine the role of IL-6 in the regulation of mucosal metabolism and integrity in different parts of the gastrointestinal tract during sepsis and endotoxemia.
1. Deitch, E. A. Multiple organ failure: Pathophysiology and potential future therapy. Ann. Surg. 216: 117, 1992. 2. von Allmen, D., Hasselgren, P. O., Higashiguchi, T., Frederick, J., Zamir, O., and Fischer, J. E. Increased intestinal protein synthesis during sepsis and following the administration of tumor necrosis factor a or interleukin-1a. Biochem. J. 286: 585, 1992. 3. Higashiguchi, T., Noguchi, Y., O’Brien, W., Wagner, K., Fischer, J. E., and Hasselgren, P. O. Effect of sepsis on mucosal protein
JSR 5288
/
6n2f$$$382
5.
6.
7.
8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
REFERENCES
AID
4.
06-18-98 22:05:07
19.
20.
31
synthesis in different parts of the gastrointestinal tract in rats. Clin. Sci. 87: 207, 1994. Higashiguchi, T., Noguchi, Y., Meyer, T., Fischer, J. E., Hasselgren, P. O. Protein synthesis in isolated enterocytes from septic or endotoxemic rats: Regulation by glutamine. Clin. Sci. 89: 311, 1995. Mester, M., Tomkins, R. G., Gelfand, J. A., Dinarello, C. A., Burke, J. F., and Clark, B. D. Intestinal production of interleukin-1a during endotoxemia in the mouse. J. Surg. Res. 54: 584, 1993. Nathens, A. B., Ding, J. W., Marshall, J. C., Dibeiro, M., and Rotstein, O. D. The gut as a cytokine generating organ: Small bowel TNF-a production during systemic endotoxemia. Surgical Infection Society 14th Annual Meeting, 1994. [abstract] Meyer, T. A., Wang, J., Tiao, G. M., Ogle, C. K., Fischer, J. E., and Hasselgren, P. O. Sepsis and endotoxemia stimulate intestinal interleukin-6 production. Surgery 118: 336, 1995. Wang, Q., Fischer, J. E., and Hasselgren, P. O. Treatment of endotoxemic mice with anti-interleukin-6 antibody paradoxically increases interleukin-6 levels and stimulates mucosal protein synthesis. Arch. Surg. 132: 82, 1997. Gauldie, J., Richards, C., and Baumann, H. IL-6 and the acute phase reaction. Res. Immunol. 143: 755, 1992. Molmenti, E. P., Ziambaras, T., and Perlmutter, D. H. Evidence for an acute phase response in human intestinal epithelial cells. J. Biol. Chem. 268: 14,116, 1993. Langkamp-Henken, B., Donavan, T. B., Pate, L. M., Maull, C. D., and Kudsk, K. A. Increased intestinal permeability following blunt and penetrating trauma. Crit. Care Med. 23: 660, 1995. Janu, P., Minard, G., and Kudsk, K. Systemic interleukin-6 (IL6) correlates with intestinal permeability. Surg. Forum 47: 7, 1996. Ohzato, H., Yoshizaki, K., Nishimoto, N., Ogata, A., Tagoh, H., Monden, M., Gotoh, M., Kishimoto, T., and Mori, T. Interleukin6 is a new indicator of inflammatory status: detection of serum levels of interleukin-6 and C-reactive protein after surgery. Surgery 111: 201, 1992. Goretsky, M. J., Harriger, D. M., Supp, A. P., Greenhalgh, D. G., and Boyce, S. T. Expression of IL-1a, IL-6, and basic fibroblast growth factor by cultured skin substitutes before and after grafting to full-thickness wounds in athymic mice. J. Trauma 40: 894, 1996. Damas, P., Ledoux, D., Nys, M., Vrindts, Y., DeGroot, D., Franchimont, P., and Lamy, M. Cytokine serum levels during severe sepsis in human: IL-6 as a marker of severity. Ann. Surg. 215: 356, 1992. Barke, R. A., Roy, S., Chapin, R. B., Charbonneau, R., Brady, P. S., and Brady, L. J. Sepsis-induced release of interleukin6 may activate the immediate-early gene program through a hypothalamic–hypophyseal mechanism. Surgery 116: 141, 1994. Shalaby, M. R., Waape, A., Aarden, L., and Espevik, T. Endotoxin, tumor necrosis factor-a and interleukin-1 induce interleukin-6 production in vivo. Clin. Immunol. Immunopathol. 53: 488, 1989. Jones, S. C., Trejdosiewicz, L. K., Bardis, R. W., Howdle, P. D., Axon, A. T. R., Dixon, M. F., and Whicher, J. T. Expression of interleukin-6 by intestinal enterocytes. J. Clin. Pathol. 46: 1097, 1993. Parikh, A. A., Salzman, A. L., Kane, C. D., Fischer, J. E., and Hasselgren, P. O. IL-6 production in human intestinal epithelial cells following stimulation with IL-1b is associated with activation of the transcription factor NFkB. J. Surg. Res. 69: 139, 1997. Sapolsky, R., Rivier, C., Yamamoto, G., Pitsky, P., and Vale, W. Interleukin-1 stimulates the secretion of hypothalamic corticotropin-releasing factor. Science 238: 522, 1991.
srga