Abstract
The effect of mercaptoethylguanidine (MEG), a selective inhibitor of the inducible nitric oxide synthase and peroxynitrite scavenger, was evaluated in a rat model of colonic injury. A single intracolonic administration of trinitrobenzene sulfonic acid (TNBS, 20 mg/kg) dissolved in ethanol induced a severe colitis in male rats. Rats experienced bloody diarrhea and a significant loss of body weight. At 4 days after TNBS administration, the colon damage was characterized by areas of mucosal necrosis. Activity of myeloperoxidase, a marker of neutrophil infiltration, and levels of the 6-keto-prostaglandin F1α, were also markedly increased, whereas colonic ATP levels were reduced into the damaged tissue. Immunohistochemistry for the inducible nitric oxide synthase and nitrotyrosine, an index of nitrosative stress, showed an intense staining in the inflamed colon. Treatment with MEG (10 mg/kg i.v. b.i.d.) significantly reduced the appearance of diarrhea and the loss of body weight. This was associated with a remarkable amelioration of the disruption of the colonic architecture and suppression of the energetic failure, as well as a significant reduction of colonic myeloperoxidase activity and 6-keto-prostaglandin F1α levels. MEG also reduced the appearance of iNOS and nitrotyrosine immunoreactivity in the colon. The results of this study suggested that administration of MEG may be beneficial for the treatment of inflammatory bowel diseases.
NO has been postulated to play a dual role in the gastrointestinal tract. The continuous release of NO from a constitutive NO synthase (cNOS), which is located in intestinal epithelial and lamina propria cells, neuronal terminals and endothelial cells, is involved in the physiological maintenance of motility, tone, permeability and tissue blood flow (Salzman, 1995). On the other hand, an overwhelming production of NO from an iNOS has been postulated to have a pathological role in IBD. Numerous clinical reports have demonstrated elevated levels of nitrite in rectal dialysates and increased iNOS activity in colonic biopsies of patients affected by Crohn’s diseases or ulcerative colitis (Middleton et al., 1993;Boughton-Smith et al., 1993; Lundberg et al., 1994; Ikeda et al., 1997). Furthermore, a growing body of experimental data obtained with the use of animal models of IBD has demonstrated that colitis is associated with increased expression or activity of iNOS and pharmacological treatment with non-isoform-selective NOS inhibitors exerts beneficial effects (Aiko and Grisham, 1995; Ribbons et al., 1995; Miller et al., 1993; Hogaboam et al., 1995; Mourelle et al., 1996; Rachmilewitz et al., 1995). The salutary action of NOS inhibitors includes protection against the appearance of severe diarrhea and weight loss, reduction of colonic lesions and improvement in colonic contractility and tone. In contrast to these reports, experimental studies in similar models of IBD have also reported that the use of non-isoform-selective NOS inhibitors (Pfeiffer and Qiu, 1995) or the genetic ablation of iNOS (McCafferty et al., 1997) may not exert any therapeutic effect or may even exacerbate the inflammatory process. The discrepancy of these findings may be related to the particular pharmacological regimen (dosage, time of treatment), which may interfere with the activity of cNOS and consequently with the homeostatic role of cNOS-derived NO or with the severity of the specific colitis model.
Recent data suggest that the damaging effects of NO in various forms of inflammation are mediated, at least in part, by peroxynitrite, a potent oxidant produced by the reaction of NO and superoxide anion (Beckman, 1996; Szabó, 1996a). Peroxynitrite is a highly reactive compound which promotes oxidative and nitrosative injury of lipids, proteins, DNA, and other important cellular biomolecules. These cytotoxic effects include lipid peroxidation, inhibition of mitochondrial respiration, inhibition of membrane pumps, and depletion of glutathione (see for review: Szabó, 1996a). The formation of peroxynitrite has been demonstrated in TNBS-induced ileitis in guinea pigs (Miller et al., 1995) and in human colonic mucosa of patients with Crohn’s disease, ulcerative colitis and diverticulites (Singer et al., 1996) by immunohistochemical staining of nitrotyrosine, a marker of peroxynitrite-induced protein modification, in colocalization with iNOS in epithelial cells. Furthermore, Rachmilewitz and colleagues have provided evidence that intracolonic administration of exogenous peroxynitrite induces severe mucosal damage in rats (1993).
The complex role of NO in the physiology and pathology and the cytotoxic potential of peroxynitrite in the gastrointestinal tract have directed research to the investigation of pharmacological tools to neutralize the cytotoxic effects of NO and/or peroxynitrite without interfering with the physiological roles of NO. We have recently demonstrated that MEG is a novel selective inhibitor of iNOS (Southanet al., 1996) with additional unique anti-inflammatory properties that include a modest direct inhibition of cyclooxygenase activity (Zingarelli et al., 1997b) and potent scavenging effects on hydroxyl radical (Kozak and Arient, 1973; Shapira et al., 1957) and peroxynitrite (Szabó et al., 1997b). MEG has been demonstrated to attenuate the inflammatory process of endotoxic shock (Southan et al., 1996), hemorrhagic shock (Zingarelli et al., 1997a) and carrageenan-induced paw edema and pleurisy in rats (Cuzzocrea et al., 1998). In the present study, we have investigated whether MEG affects the course of the inflammatory response in a rat colitis model induced by intracolonic administration of TNBS.
Methods
This investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85–23 revised 1985) and with the approval of the Institutional Animal Care and Use Committee.
Animals.
Male Wistar rats (Charles River Laboratories, Wilmington, MA), weighing 300 to 350 g, were housed in a room with controlled temperature (22°C) and 12-hr light/dark cycle. The animals were food fasted 24 hr before experimentation and allowed food and water ad libitum after the administration of TNBS.
Induction of colitis.
Colitis was induced by using the technique as previously described (Morris et al., 1989;Zingarelli et al., 1993). In fasted rats lightly anaesthetized with isoflurane, a rubber cannula (OD, 2 mm) was inserted into the lumen of the colon via the anus until approximately the splenic flexure (8 cm from the anus). TNBS (20 mg/rat) was dissolved in 50% ethanol (vol/vol) and injected (0.25 ml) into the colon via the rubber cannula. Control animals received 50% ethanol alone. Animals were then maintained in a vertical position for 30 sec and returned to their cages. In a group of rats, MEG was administered intravenously by the dorsal tail vein twice a day at 10 mg/kg. This dosage regimen of MEG has been previously shown to exert anti-inflammatory effects (Southan et al., 1996; Zingarelliet al., 1997a; Cuzzocrea et al., 1998).
Drug treatment was started 24 hr before TNBS intracolonic administration and continued until the end of the experimental period (4 days after TNBS intracolonic administration). After 4 days, the animals were killed by isoflurane anesthesia, and a segment of the colon 8 cm long was excised for macroscopic damage evaluation. Tissue segments 1 cm in length were then fixed in 10% buffered formalin or immediately frozen in liquid nitrogen and stored at −70°C for histological and immunochemical studies and for determination of myeloperoxidase activity, 6-keto-PGF1α and ATP levels.
Assessment of colonic damage.
After removal, the colon was gently rinsed with saline solution, opened by a longitudinal incision and immediately examined under a stereomicroscope. The visible colonic damage was assessed by two independent observers blinded to the experimental protocol using a semiquantitative scoring system (Morriset al., 1989; Zingarelli et al., 1993). The following morphological criteria were considered: score 0, no damage; score 1, localized hyperemia without ulcers; score 2, linear ulcers, with no significant inflammation; score 3, linear ulcers with inflammation at one site; score 4, two or more major sites of ulceration and/or inflammation; score 5, two or more sites of inflammation and ulceration extending >1 cm along the length of the colon; and score 6 through 10, one point is added for each centimeter of ulceration beyond an initial 2 cm.
Histopathological analysis.
For microscopic histological evaluation, formalin-fixed tissues were embedded in paraffin, and 5-μm sections were stained with hematoxylin and eosin and evaluated by light microscopy by a pathologist unaware of the experimental protocol.
Immunohistochemistry for iNOS.
Frozen sections (5 μm) were treated with 0.3% H2O2 for 15 min to block endogenous peroxidase activity and then rinsed briefly in PBS. Nonspecific binding was blocked by incubating the slides with a blocking solution (0.1 M phosphate-buffered saline containing 0.1% Triton X-100 and 2% normal goat serum) for 2 hr. To detect iNOS, rabbit polyclonal anti-iNOS antibody was applied in a dilution of 1:500 at 4°C overnight. Control sections included buffer alone or nonspecific purified rabbit IgG. Immunoreactivity was detected with a biotinylated goat anti-rabbit secondary antibody and the avidin-biotin-peroxidase complex (Vectastain Elite ABC kit, Vector Laboratories). Color was developed using diaminobenzidine (Milleret al., 1995).
Immunohistochemistry for nitrotyrosine.
Tyrosine nitration was detected in colonic sections by immunohistochemistry (Miller et al., 1995). Frozen sections 5 μm thick were fixed in 4% paraformaldehyde and incubated for 2 hr with a blocking solution (0.1 M phosphate-buffered saline containing 0.1% Triton X-100 and 2% normal goat serum) to minimize nonspecific adsorption. Sections were then incubated overnight with 1:500 dilution of primary anti-nitrotyrosine antibody or with control solutions. Controls included buffer alone or nonspecific purified rabbit IgG. Some sections were also incubated with the primary antibody in the presence of excess nitrotyrosine (10 mM) to verify the binding specificity. Specific labeling was then detected by incubating for 30 min with a biotin-conjugated goat anti-rabbit IgG and amplified with avidin-biotin peroxidase complex (Vectastain Elite ABC kit, Vector Laboratories) after quenching endogenous peroxidase with 0.3% H2O2 in 100% methanol for 15 minutes. Diaminobenzidine was used as a chromogen.
Assay of myeloperoxidase activity.
Myeloperoxidase activity was determined as an index of neutrophil accumulation (Krawisz et al., 1984). Colonic tissues were homogenized in a solution containing 0.5% hexa-decyl-trimethyl-ammonium bromide dissolved in 10 mM potassium phosphate buffer (pH 7) and centrifuged for 30 min at 20,000 × g at 4°C. An aliquot of the supernatant was then allowed to react with a solution of tetra-methyl-benzidine (1.6 mM) and 0.1 mM H2O2. The rate of change in absorbance was measured by spectrophotometry at 650 nm. Myeloperoxidase activity was defined as the quantity of enzyme degrading 1 μmol of peroxide/min at 37°C and was expressed in milliunits per 100 mg weight of tissue.
Measurement of colonic ATP.
ATP levels were measured by HPLC. Tissues were homogenized in 2 N perchloric acid, neutralized with 1 M KH2PO4 and centrifuged at 3500 × g at 4°C. Supernatants were then assayed using a Beckman Model 231 HPLC instrument equipped with a Waters Wisp autoinjector. For determinations of ATP, aliquots of the samples were analyzed on a Whatman SAX-10 ion exchange column, using a mobile phase consisting of 0.16 M KH2PO4and 0.1 M KCl in deionized water (pH 6.5) at a flow rate of 1.5 ml/min. ATP were detected by ultraviolet absorption spectrophotometry at 254 nm, and values were expressed as nmol/mg protein.
Measurement of colonic 6-keto-PGF1α.
Colonic samples were homogenized on ice in a 50 mM Tris buffer (pH 7.4) and centrifuged for 15 min at 4000 × g. The supernatant was then stored in microfuged tubes containing indomethacin (10 μg/ml) and frozen at −20°C. The stable metabolite of prostacyclin, 6-keto-PGF1α, was measured by radioimmunoassay (Zingarelli et al., 1997b).
Materials.
Primary anti-iNOS and anti-nitrotyrosine antibody was purchased from Upstate Biotech (Saranac Lake, NY). MEG was prepared as previously described (Southan et al., 1996). [3H]6-keto-PGF1αwas purchased from DuPont-NEN (Boston, MA). Reagents, secondary and nonspecific IgG antibodies for immunohistochemical analysis were obtained from Vector Laboratories (Burlingame, CA). All other chemicals were from Sigma/Aldrich (St. Louis, MO).
Statistical analysis.
All data are expressed as mean ± S.E.M.; n refers to the number of rats. Statistical differences between groups were calculated by one- and two-way ANOVA followed by the Dunnett’s post hoc test. Differences were considered significant at a P value of < .05.
Results
Colonic damage.
Four days after intracolonic administration of TNBS, the colon appeared flaccid and filled with liquid stool. The cecum, colon and rectum all had evidence of mucosal congestion, erosion and hemorrhagic ulcerations (see fig. 1for damage score). The histopathological features included a transmural necrosis and edema and a diffuse leukocyte cellular infiltrate in the submucosa (fig. 2). In vivotreatment with MEG resulted in a significant decrease in the extent and severity of the colonic damage (figs. 1 and 2).
Changes of body weight.
The inflammatory changes of the intestinal tract were accompanied by a significant loss in body weight in comparison to control rats (fig. 3). Treatment with MEG significantly reduced the loss in body weight, which correlated well with the amelioration of colonic injury.
Immunohistochemistry for iNOS and nitrotyrosine.
Expression of iNOS, evaluated by immunoreaction, was negative in colonic sections from control alcohol-administered rats (fig.4A). However, in the area of ulcer formation in colonic sections from TNBS-treated rats, a positive immunostaining staining for iNOS was found, mainly localized in the infiltrated inflammatory cells and in disrupted epithelial cells (fig.4B).
Furthermore, to determine the tissue localization of peroxynitrite and/or other nitrogen derivatives produced during colitis, nitrotyrosine, a specific marker of nitrosative stress (Beckman, 1996), was measured by immunohistochemical analysis in the distal colon. Sections of colon from control alcohol-administered rats did not reveal any immunoreactivity for nitrotyrosine within the normal architecture (fig. 5A). A positive brown color for immunoreactive nitrotyrosine was evident throughout the colon wall in TNBS-induced colitis (fig. 5B). The specificity of the nitrotyrosine immunoreactivity in injured sections was demonstrated by the fact that incubation of anti-nitrotyrosine antiserum with excess of nitrotyrosine (10 mM) largely eliminated the positive staining (fig. 5C).
In TNBS-treated rats administered with MEG, there was no evidence of iNOS expression and nitrotyrosine formation (figs. 4C and 5D).
Colonic myeloperoxidase activity and prostaglandin levels.
Colonic injury by TNBS administration was also characterized by an increase of myeloperoxidase activity, indicative of neutrophil infiltration in inflamed tissue (fig.6A), confirming the enhanced leukocytes infiltration seen at the histological inspection. Levels of 6-keto-PGF1α indicative of activation of the inflammatory response by cyclooxygenase were also elevated in tissue samples from TNBS-administered rats compared with control vehicle-intracolonic injected rats (fig. 6B). Treatment with MEG reduced leukocyte infiltration and the elevation of 6-keto-PGF1α (fig. 6).
Colonic levels of ATP.
The morphological changes were also compared with changes in ATP levels. The concentration of ATP in the colon of TNBS-treated rats was markedly decreased by 44% in comparison to control vehicle-treated rats. AMP and ADP levels were not significantly altered (data not shown). MEG significantly reduced the ATP depletion (fig. 6C).
Discussion
Formation of nitrogen derived-oxidants in TNBS-induced colitis.
Among the proinflammatory mediators involved in IBD, the free radical NO and its related metabolite peroxynitrite, formed from the reaction with superoxide anion, have been proposed to promote a pathological pathway (Miller et al., 1995; Singer et al., 1996; Rachmilewitz et al., 1993). Several reports from clinical research or experimental animal studies support this concept. Monocytes from patients with Crohn’s disease (Kitahoraet al., 1988) and polymorphonuclear leukocytes from patients with ulcerative colitis (Shiratora et al., 1989) have an increased capacity to generate free radicals. Furthermore, advanced stages of bowel inflammation in humans (Middleton et al., 1993; Boughton-Smith et al., 1993Lundberg et al., 1994; Ikeda et al., 1997) and animals (Aiko and Grisham, 1995; Ribbons et al., 1995; Miller et al., 1993; Hogaboam et al., 1995; Mourelle et al., 1996; Rachmilewitz et al., 1995) are characterized by elevated formation of NO after expression of iNOS. Under these conditions, the formation of peroxynitrite is also promoted and may have a pathological role (see Introduction). There is also evidence that peroxynitrite may not be the only oxidant to produce nitrotyrosine. It is possible that other reactions, such as the reaction of nitrite with hypochlorous acid or, in the presence of neutrophils, the reaction of nitrite with myeloperoxidase and hydrogen peroxide, may yield nitrotyrosine (Eiserich et al., 1998). These findings suggest that during the inflammatory process, nitrotyrosine formation is the result of a concerted action of several oxidants and may be considered a sign of a sustained nitrosative stress. Accordingly, in the present study, we found a positive immunohistochemical staining for iNOS and nitrotyrosine in the inflamed colon of TNBS-treated rats suggesting the formation of peroxynitrite and other NO-derived oxidants.
Amelioration of colonic injury by MEG treatment.
Under these conditions, we have found that daily treatment with MEG markedly suppressed tissue injury, attenuated leukocyte infiltration and prostaglandin production and preserved the energetic pool of ATP. The amelioration of the TNBS-induced colitis symptomatology corresponded with the absence of the nitrotyrosine staining in colonic tissue.
Although it is difficult to establish from the in vivo data what is the prevalent protective action of MEG in the amelioration of TNBS-induced colitis, several mechanisms of action may contribute to its therapeutic action.
Inhibition of iNOS expression and activity.
As a member of the mercaptoalkylguanidines, MEG has been identified as a potent inhibitor of the catalytic activity of NOS, with selectivity toward iNOS in rodent models and with protective effects in endotoxic (Southanet al., 1996) and hemorrhagic (Zingarelli et al., 1997a) shock and carrageenan-induced inflammation (Cuzzocrea et al., 1998). Furthermore, MEG and related compounds have been shown to reduce expression of iNOS in macrophages, lungs and hearts of endotoxin-shocked rats (Ruetten and Thiemermann, 1996). In the present study, we found a reduction of iNOS-like immunoreactivity in the colon. This effect may be directly correlated to inhibition of iNOS expression and indirectly to the reduction of the influx of inflammatory cells expressing iNOS into the inflamed colon. Nevertheless, regardless of the mechanism on iNOS activity and/or expression, MEG provides a therapeutic effect in our model of colitis by inhibiting the generation of iNOS-derived NO and subsequent nitrogen-centered oxidants. Our hypothesis is in agreement with multiple studies demonstrating that pharmacological inhibition of NO synthesis seems to exert beneficial effects in several experimental models of colitis in animals (Ribbonset al., 1995; Miller et al., 1993; Hogaboamet al., 1995; Mourelle et al., 1996; Rachmilewitzet al., 1995).
Scavenging effect on peroxynitrite and hydroxyl radical.
In vitro and in vivo studies have shown that MEG has a potent scavenging effect on hydroxyl radical (Shapira et al., 1957) and peroxynitrite (Szabó et al., 1997b). The ability of directly scavenging peroxynitrite represents a therapeutic advantage in comparison to the sole inhibitory action on iNOS activity and expression. In this regard, it is important to mention that under conditions of oxidant stress peroxynitrite can be also formed from the cNOS-derived NO (Szabó, 1996a). Furthermore, nonenzymatic sources of NO generation, such as the reduction of nitrite under acidic conditions (Zweier et al., 1995), can eventually lead to peroxynitrite formation. In agreement with this concept, we have found that nitrotyrosine is still formed in aortas of iNOS-deficient mice subjected to endotoxin shock (Zingarelli et al., 1998). The possibility of iNOS-independent sources of NO and related oxidants may also explain why genetic deficiency of iNOS did not provide protection against acetic acid-induced colitis (McCaffertyet al., 1997). Therefore, MEG limits the cytotoxicity of peroxynitrite while maintaining the physiological functions of the constitutive production of NO, such as modulation of fluid and electrolyte secretion, motor activity, mucosal blood flow and inhibition of inflammatory cells infiltration. Experimental evidence shows, in fact, that acute nonselective blockade of constitutive NO synthesis augments leukocyte adhesion to the microvascular endothelium (for review, see Salzman, 1996) and exacerbates the damage associated with TNBS intracolonic administration (Pfeiffer and Qiu, 1995). In addition, considering the toxic effect of peroxynitrite on endothelial function (Szabó et al., 1997a), the mechanism of protection by MEG may involve the preservation of endothelial cell integrity. This latter effect, associated with the maintenance of constitutive NO release, is responsible for the reduction of neutrophil accumulation in the injured zone, as demonstrated by histology and tissue myeloperoxidase activity. The fact that activated neutrophils play a role in the pathogenesis of TNBS-induced colitis in rats is supported by the finding that the colonic alterations associated with TNBS colitis are attenuated by prior neutrophil depletion with anti-neutrophil antiserum (Buell and Berin, 1994).
One of the proposed inflammatory mechanisms induced by peroxynitrite, hydroxyl radical, oxidants and other free radicals involves DNA damage and activation of the nuclear enzyme poly(ADP-ribose) synthetase. The occurrence of this process has been demonstrated to lead to a rapid depletion of the intracellular NAD+ and ATP energetic pools in human epithelial cells (Kennedy et al., 1998). Nitrogen-centered oxidants may also degrade the iron clusters of mammalian mitochondrial NADH-coenzyme Q reductase, succinate dehydrogenase, aconitase and ATPase (e.g.,Gardner et al., 1997), therefore altering intracellular energetics. In the present study, the inflammatory process after TNBS administration was associated with a marked decrease in the concentrations of ATP, thus suggesting the presence of an altered balance between energy-consuming and energy-producing processes in the intestinal cells. Treatment with MEG prevented the loss of colonic ATP levels. The effect of the drug may be related to an effect on cellular energy generation or may simply reflect the higher proportion of intact/viable cells (as opposed to necrotic and apoptotic cells) in the tissue samples.
Inhibition of cyclooxygenase activity.
Mucosal eicosanoids are suggested to play a role in the pathogenesis of inflammatory bowel disease (Zingarelli et al., 1993; Hendel and Nielsen, 1997). In our experiments, colonic concentration of 6-keto-PGF1α, considered a marker of cyclooxygenase activation, were found to be elevated after TNBS-induced colitis and significantly reduced by in vivo treatment with MEG. The inhibition by MEG of prostaglandin production in TNBS-induced colitis is likely secondary to a direct and an indirect mechanism. MEG may exert an anti-inflammatory effect through a direct inhibition on cyclooxygenase activity, as shown in recent in vitro studies (Zingarelli et al., 1997b). In addition, because NO has been proposed to amplify tissue damage by increasing the catalytic activity of cyclooxygenase (Salvemini et al., 1993), MEG may promote the resolution of injury by inhibiting the synthesis of the proinflammatory NO.
Conclusion.
Our data demonstrate that MEG suppresses the course of TNBS-induced colitis. The pharmacological effects of MEG may be multiple. In the current study, we provide evidence that MEG inhibits expression of iNOS and nitrotyrosine (as a marker of formation of peroxynitrite and other nitrogen-derivatives). Considering also the ability of MEG to block other pathways of inflammation (such as cyclooxygenase) and to scavenge oxyradicals, MEG may represent an efficacious anti-inflammatory drug for IBD treatment.
Acknowledgments
The authors thank Mr. Alvin Denenberg for technical assistance.
Footnotes
-
Send reprint requests to: Basilia Zingarelli, MD, PhD, Children’s Hospital Medical Center, Division of Critical Care, 3333 Burnet Ave, Cincinnati, OH 45229.
-
↵1 Funding for this study was provided in part by National Institutes of Health Grant 1R01-HL59352–01 to Dr. Andrew L. Salzman.
- Abbreviations:
- NO
- nitric oxide
- cNOS
- constitutive nitric oxide synthase
- iNOS
- inducible nitric oxide synthase
- TNBS
- 2,4,6-trinitrobenzene sulfonic acid
- MEG
- mercaptoethylguanidine
- IBD
- inflammatory bowel disease
- 6-keto-PGF1α
- 6-keto-prostaglandin F1α
- ANOVA
- analysis of variance
- Received March 19, 1998.
- Accepted July 9, 1998.
- The American Society for Pharmacology and Experimental Therapeutics