Experimental colitis in rats induces low-grade endotoxinemia without hepatobiliary abnormalities

Bile synthesis in rat models of inflammatory bowel diseases

BACKGROUND

A broad spectrum of hepatobiliary disorders are found in patients with inflammatory bowel diseases. The aim of the present work was to study interactions between gut and liver in experimental rat models of colitis and small bowel inflammation.

MATERIALS AND METHODS

Colitis was induced either by trinitrobenzene sulphonic acid or dextran sodium sulphate. Small-bowel inflammation was induced by indomethacin. Bile acid secretion, bile acid pool, and cholesterol 7-alpha hydroxylase were studied. Cholesterol 7-alpha hydroxylase protein expression was analysed in the microsomal liver fraction. As portal mediators released form the inflamed gut we measured lipopolysaccharide, tumour necrosis factor-alpha and interleukin-1beta in portal serum. The hepatic inflammatory response was evaluated by binding activity of nuclear factor-kappaB, activator protein-1 and alpha-2-macroglobulin.

RESULTS

Increased bile acid secretion, total bile acid content in gut and liver (bile acid pool size), and hepatic cholesterol 7-alpha hydroxylase protein and mRNA levels were found in the two colitis models associated with only a minor hepatic acute phase and cytokine response. In contrast, during indomethacin-induced small-bowel inflammation bile acid secretion, pool size, and cholesterol 7-alpha hydroxylase decreased in parallel to a strong hepatic cytokine and acute phase response.

CONCLUSIONS

Colitis without portal cytokine release and acute phase reaction shows an induction of bile acid secretion, pool size, and cholesterol 7-alpha hydroxylase. In contrast, intestinal inflammation after indomethacin treatment is associated with an acute phase response and a repression of bile acid synthesis.

Influence of modest endotoxemia on postoperative antithrombin deficiency and circulating secretory immunoglobulin a levels

OBJECTIVE

To evaluate the influence of modest endotoxemia on postoperative antithrombin deficiency and cholestasis.

SUMMARY BACKGROUND DATA

It has not been determined whether endotoxin translocation in small amounts is a physiological phenomenon or whether it is a potential health hazard.

METHODS

Blood endotoxin, antithrombin III (ATIII), secretory immunoglobulin A (sIgA), which was selected as a marker of cholestasis, C-reactive protein (CRP), and alpha-1-antitrypsin (AAT) concentrations were measured from the 20 patients undergoing curative gastrectomy for gastric cancer preoperatively and postoperatively. Portal and systemic blood samples were taken for the analysis of endotoxin and interleukin-6 (IL-6) concentrations during surgery in these patients.

RESULTS

Although plasma endotoxin levels showed a significant increase during surgery, we did not find a correlation with ATIII, sIgA, CRP, and IL-6 levels. Systemic blood endotoxin levels during surgery correlated with a postoperative rise of serum AAT levels. Plasma ATIII levels transiently decreased on the first and third postoperative day, and sIgA levels were shown to increase on the seventh postoperative day. There was a weak relationship between the extent of postoperative endotoxemia and a reduction in ATIII concentrations.

CONCLUSIONS

The influence of modest endotoxemia on postoperative antithrombin deficiency and cholestasis was limited, and increased translocational endotoxemia during abdominal surgery may be a physiological phenomenon to trigger off an acute-phase protein response.

Experimental colitis attenuates development of toxin-induced cholangitis in rats

Primary sclerosing cholangitis frequently occurs in association with ulcerative colitis. This close association may be due to colitis predisposing patients to bile ductular injury. Therefore, we determined the susceptibility of rats with experimental colitis to toxin-induced cholangitis. Sprague-Dawley rats received 2,4,6-trinitrobenzene-sulfonic-acid (TNBS) or ethanol vehicle intracolonically. Seven days later, rats received either the biliary epithelial cell toxin alpha-naphthylisothiocyanate (ANIT) or vehicle and were killed 24 hr later. Liver histology, serum biochemistries and tumor-necrosis factor-alpha (TNF-alpha), and hepatic interleukin-10 (IL-10) mRNA were determined. TNBS-treated rats showed extensive macroscopic colonic damage and a 10-fold increase in myeloperoxidase activity compared to ethanol-treated controls. ANIT-treated noncolitic rats showed portal inflammation centered on damaged bile ducts (cholangitis), which was markedly attenuated in ANIT-treated colitic rats. Hepatic IL-10 mRNA was twofold higher in colitic compared to noncolitic rats, with no difference in serum TNF-alpha. In conclusion, experimental colitis attenuates the development of toxin-induced cholangitis in rats, possibly by up-regulating hepatic IL-10 expression.

Disturbed bile secretion and cytochrome P450 function during the acute state of experimental colitis in rats

BACKGROUND/AIMS

A variety of hepatobiliary abnormalities has been described in patients with inflammatory bowel diseases. However, the pathophysiological mechanisms leading to these liver alterations are poorly understood. The aim of the present study was to investigate parameters of liver function in a trinitrobenzenesulfonic acid (TNB)-induced rat colitis model.

METHODS

Glucose output, bile acid secretion, bile acid uptake, and the cytochrome P-450 metabolic capacity during TNB-colitis were studied in the perfused liver model. Furthermore, hepatic bile acid- and glycogen content was measured. To evaluate the inflammatory response in the colon and liver, NF-kappaB/Rel was quantified by electrophoretic mobility shift assays. As an NF-kappaB/Rel regulated gene the inducible NO-synthase (NOS2) was evaluated by Western blot analysis. As possible mediators released from the inflamed colon into the portal vein, endotoxin and the stable metabolite of prostaglandin I2 (6-keto-prostaglandin-F1alpha) were determined.

RESULTS

Glucose output, bile acid secretion, bile acid uptake, and cytochrome P-450 metabolic capacity decreased on the first and second day of TNB-colitis. Hepatic bile acid content increased at day 14 of colitis. Glycogen content was reduced, most likely due to an inadequate chow intake of these animals. A low level of portal endotoxin was detectable during the first 2 days of colitis. In addition, 6-keto-prostaglandin-F1alpha was clearly increased in portal blood. NF-kappaB/Rel binding activity and inducible NOS2 were strongly positive in the colon during colitis. Although low levels of portal endotoxin were measured during the first 2 days of colitis, no significant NF-kappaB/Rel activity and NOS2 induction were detected in the liver.

CONCLUSION

Our results indicate that during the acute state of the TNB-colitis, bile acid secretion and cytochrome P-450 function are disturbed in the absence of distinct inflammatory changes in the liver.

Varying cecal bacterial loads influences colitis and gastritis in HLA-B27 transgenic rats

BACKGROUND & AIMS

Recent data support an important role of resident luminal bacteria in experimental colitis. We determined how altered cecal bacterial loads influence colitis and gastritis.

METHODS

A cecal self-filling blind loop (SFBL) was created or the cecum was excluded from the fecal stream in specific pathogen-free HLA-B27 transgenic (TG) rats with early colitis and in nontransgenic (nonTG) littermates; controls underwent sham operation (SHAM). Luminal bacterial concentrations were determined by culture and counting chamber.

RESULTS

TG rats with SFBL had more severe cecal inflammation and leukocytosis than TG SHAM controls. TG excluded rats with low cecal bacterial loads had no cecal inflammation and less colitis and gastritis than SHAM controls, despite having normal distal colonic and gastric bacterial concentrations. Metronidazole attenuated cecal inflammation and eliminated Bacteroides in SFBL TG rats. NonTG SFBL rats had mild cecal inflammation and no gastritis and colitis. The ratio of total anaerobic to aerobic bacteria was 1000-fold greater in SFBL than in SHAM rats, with a 10,000-fold increased ratio of Bacteroides spp. to aerobes.

CONCLUSIONS

The luminal bacterial load and composition determines the activity of cecal inflammation in genetically susceptible hosts. Lowering cecal bacterial concentrations can diminish inflammation in remote organs.

Animal models of intestinal and joint inflammation

Recent rodent models have been exploited to explore mechanisms of intestinal and joint inflammation. HLA-B27 transgenic rats develop colitis, gastritis, and arthritis when raised in a conventional environment, but have no evidence of inflammation under germfree (sterile) conditions. Metronidazole treatment attenuates gastrointestinal inflammation, suggesting that anaerobic bacteria are important. Experimental bacterial over-growth of predominantly anaerobic bacteria reactivates arthritis in Lewis rats which have been previously injected intra-articularly with bacterial cell wall polymers. Reactivation arthritis is mediated by interleukin-1, tumour necrosis factor-alpha, and can be blocked by metronidazole. Intramural injection of the bacterial cell wall polymer, peptidoglycan-polysaccharide, leads to biphasic, chronic granulomatous enterocolitis and peripheral arthritis in Lewis rats, but only transient intestinal inflammation and no arthritis in Buffalo or MHC-matched Fischer rats. Chronic granulomatous inflammation is mediated by T lymphocytes and interleukin-1 and is dependent on persistent antigenic stimulation by poorly biodegradable bacterial polymers. Results in these models firmly incriminate resident normal enteric flora (especially anaerobes), bacterial products, and host genetic susceptibility in the pathogenesis of spondyloarthropathies. We suggest that increased uptake of luminal bacterial components across the inflamed mucosa leads to systemic distribution of these arthropathic products. The genetically susceptible host develops reactive arthritis due to defective downregulation of inflammation in response to immunologically active bacterial components.

Endotoxaemia and cytokine production in experimental colitis

Systemic endotoxaemia is a well recognized feature of inflammatory bowel disease but its pathogenic role remains uncertain. This study examined plasma endotoxin and cytokine concentrations and the acute-phase protein response in a hapten-induced model of experimental colitis. On days 2, 8 and 14 after induction of colitis with trinitrobenzenesulphonic acid in ethanol (TNBS-E), plasma endotoxin, immunoglobulin (Ig) G and IgM endotoxin-core antibody (EndoCAb), tumour necrosis factor (TNF), interleukin (IL) 6 and alpha 2-macroglobulin (alpha 2M) concentrations and colon macroscopic inflammation score were determined. At all time points there was significant colonic inflammation when compared with control values (P < 0.0001). Animals treated with TNBS-E had raised concentrations of endotoxin at all time points (P < 0.04). In TNBS-E-treated animals EndoCAb concentrations were reduced on day 2 (P < 0.0001) and later increased. There were increases in IL-6 and alpha 2M concentrations in TNBS-E-treated animals but no significant change in TNF concentrations. Endotoxin concentrations correlated with macroscopic inflammation score, IL-6 and alpha 2M concentrations. There was a less consistent correlation between EndoCAb concentrations and these parameters. These results suggest that endotoxin is a mediator of the systemic response in this model of experimental colitis.

Role of endogenous enteric organisms in the reactivation of arthritis

Reactive arthritis is an acute form of arthritis apparently caused by a combination of bacterial infection and genetic influences. Recent experiments using an animal model suggest that certain bacterial cell wall polymers originating from endogenous enteric bacteria may be responsible for the condition.

A microcholangiographic study of liver disease models in rats

RATIONALE AND OBJECTIVES

Rats develop hepatobiliary injury due to small bowel bacterial overgrowth (SBBO) that, at specimen, resembles cholangiography sclerosing cholangitis. To better visualize the smaller bile ducts, we used microcholangiography to determine the spectrum of biliary lesions in this and five other models of liver disease.

METHODS

The models studied were as follows: (1) Surgically created jejunal, self-filling blind loops induce SBBO. (2) Intraperitoneal injection of a bacterial cell wall polymer, peptidoglycan-polysaccharide (PG-PS), causes granulomatous hepatitis. (3) Intraperitoneal injection of endotoxin (lipopolysaccharide) causes sinusoidal congestion and shock. (4) Bile duct ligation induces bile duct proliferation. (5) Alpha-naphthyl-isothiocyanate (ANIT) induces bile duct proliferation. (6) Carbon tetrachloride (CCl4) causes fibrosis and cirrhosis. Warmed barium sulfate, gelatin, and saline were injected in the extrahepatic bile duct. Liver slices (2 mm) underwent microradiographic techniques, and images were correlated with histology.

RESULTS

Rats with SBBO had irregular and dilated extrahepatic bile ducts with thickened walls. Rats treated with endotoxin and CCl4 had normal microcholangiograms. Bile duct proliferation was identified following ANIT and bile duct ligation. Rats given PG-PS demonstrated irregular intrahepatic bile ducts. Microcholangiograms following SBBO and PG-PS showed similarities including focal ductal dilatation, narrowing, proliferation, and destruction.

CONCLUSION

Various models of liver injury induce characteristic cholangiographic appearances. Microcholangiography is useful in examining biliary tract lesions and complements histology.