Enhanced gastric and duodenal platelet-activating factor and leukotriene generation in duodenal ulcer patients

Platelet-activating factor induces TLR4 expression in intestinal epithelial cells: implication for the pathogenesis of necrotizing enterocolitis

Necrotizing enterocolitis (NEC) is a leading cause of morbidity and mortality in neonatal intensive care units, however its pathogenesis is not completely understood. We have previously shown that platelet activating factor (PAF), bacteria and TLR4 are all important factors in the development of NEC. Given that Toll-like receptors (TLRs) are expressed at low levels in enterocytes of the mature gastrointestinal tract, but were shown to be aberrantly over-expressed in enterocytes in experimental NEC, we examined the regulation of TLR4 expression and signaling by PAF in intestinal epithelial cells using human and mouse in vitro cell lines, and the ex vivo rat intestinal loop model. In intestinal epithelial cell (IEC) lines, PAF stimulation yielded upregulation of both TLR4 mRNA and protein expression and led to increased IL-8 secretion following stimulation with LPS (in an otherwise LPS minimally responsive cell line). PAF stimulation resulted in increased human TLR4 promoter activation in a dose dependent manner. Western blotting and immunohistochemical analysis showed PAF induced STAT3 phosphorylation and nuclear translocation in IEC, and PAF-induced TLR4 expression was inhibited by STAT3 and NFκB Inhibitors. Our findings provide evidence for a mechanism by which PAF augments inflammation in the intestinal epithelium through abnormal TLR4 upregulation, thereby contributing to the intestinal injury of NEC.

The effect of single-dose naproxen on eicosanoid formation in human gastroduodenal mucosa

BACKGROUND

In animal studies, aspirin and non-aspirin non-steroidal anti-inflammatory drugs contribute to gastroduodenal damage via cyclo-oxygenase inhibition and consecutive leucotriene formation (COX-LOX eicosanoid shunt).

AIM AND METHODS

Ten Helicobacter pylori-negative healthy volunteers received a single dose of 500 mg naproxen to address two questions: (i) is there a crosstalk between eicosanoids before medication in the human gastroduodenal mucosa and (ii) can we demonstrate a COX-LOX shunt following single-dose naproxen?

RESULTS

Significant correlations in the stomach mucosa before medication were obtained between leucotriene B4 (LTB4) and thromboxane B(2) (TxB(2); r = -0.38, P = 0.05), as well as LTB4 and prostaglandin E(2) (PGE(2); r = 0.71, P < 0.0001). In serum, a >90% inhibition of TxB(2) and PGE(2) occurred within 30 min of naproxen administration. In gastric mucosa, a significant decrease of TxB(2) occurred already at 15 min and preferably in the antrum. For LTB4 there was a non-significant trend towards a transient increase. Mucosal PGE(2) was unchanged in all regions; transcript levels of both cyclo-oxygenases/5-lipoxygenase were unaffected (except for a trend of increasing cyclo-oxygenase-2 in the corpus).

CONCLUSIONS

Baseline correlations between LTB4-TxB(2) and LTB4-PGE(2) reflect a crosstalk between these eicosanoids. A COX-LOX shunt; however, cannot be demonstrated following single-dose naproxen in a low-risk population.

Mechanosensitive properties of gastric vagal afferent fibers in the rat

Single, teased fiber recordings were made from the decentralized right cervical vagus nerve (hyponodal) of the rat. A total of 67 afferent fibers that responded to gastric distension (GD) were studied: 9 fibers were stimulated by phasic balloon GD, 58 by more natural fluid GD. All balloon GD-responsive fibers had resting activity (3.1 imp/s), and 57/58 fluid GD responsive fibers had resting activity (1.3 imp/s). All balloon GD-responsive fibers exhibited a dynamic response to phasic distension followed by slow adaptation, whereas fluid GD-responsive fibers exhibited increasing responses as intragastric pressure increased, followed typically by slow adaptation. Responses to graded GD were studied in all fibers, and all gave increasing responses to increasing pressures (5-60 mmHg). Thresholds for response varied between 0 and 18 mmHg. Mean response thresholds for two durations of fluid GD (30 and 60 s) were 5.6 and 3.9 mmHg; the mean response threshold to phasic balloon GD (30 s duration) was 5.3 mmHg. The potential sensitizing effect of platelet activating factor (PAF, 50 or 100 ng. kg(-1). min(-1) for 20 min) infused into the gastric artery was studied in 20 fibers. Fifteen fibers exhibited an increase in spontaneous activity; intragastric pressure also slightly increased during PAF infusion. The increase in activity produced by PAF was attenuated in the presence of the PAF receptor antagonist WEB 2086. After PAF-induced acute inflammation of the stomach, three of five fibers studied did not exhibit any change in response to graded GD. The present study characterized distension-sensitive afferent fibers in the right cervical vagus innervating the stomach of the rat by balloon GD and fluid GD. The results document that all distension-sensitive gastric vagal afferent fibers encoded the intensity of GD, but none had response thresholds in what might be considered the noxious range. PAF infusion activated mechanosensitive gastric vagal afferent fibers, but acute inflammation produced by PAF did not sensitize responses to GD.

Stimulated human gastric tumor cells (HGT) fail to synthesize eicosanoids

HGT cells are a human gastric tumor cell line. Preliminary data have shown that HGT cells incorporate exogenous arachidonic acid (AA) in their membrane lipids. However, we found that HGT cells are unable to produce significant amounts of AA metabolites after stimulation with calcium ionophore A23187. Furthermore, no lipoxygenase activity was detected in crude HGT cell extracts by employing an assay monitoring the in vitro utilization of linoleic acid. The meaning of these results is discussed in respect of the role of eicosanoids during cell proliferation.

Enhanced gastric mucosal leukotriene B4 synthesis in patients taking non-steroidal anti-inflammatory drugs

The effects of longstanding non-steroidal anti-inflammatory drug (NSAID) treatment on gastric mucosal synthesis of leukotriene B4 (LTB4), leukotriene C4 (LTC4), and prostaglandin E2 (PGE2) was studied. Gastric antral biopsies in 65 patients with arthritis taking NSAIDs and 23 control patients were taken and eicosanoid concentrations, stimulated by vortex mixing or calcium ionophore, were measured by radioimmunoassay. Median gastric mucosal synthesis of LTB4 was increased in patients taking NSAIDs compared with non-users: (0.9(0.2-2.5) pg/mg v 0 (0-0.6) pg/mg (p < 0.001)). These differences persisted when subgroups of patients were analysed according to Helicobacter pylori colonisation or degree of mucosal injury. Synthesis of LTB4 was strongly associated with the presence of type C (chemical) gastritis. Increased synthesis of LTC4 was associated with Helicobacter pylori colonisation but not NSAID use. Synthesis of PGE2 was decreased in patients taking NSAIDs compared with control patients (p < 0.001). Enhanced gastric mucosal synthesis of LTB4 in patients taking NSAIDs may represent a primary effect of these drugs and could be implicated in the pathogenesis of gastritis and ulceration associated with NSAIDs.

Incorporation of arachidonic acid in a human cancer gastric tumor cell line (HGT) at various stages of cell proliferation

Human cancer gastric cells (HGT) in culture were labeled with [14C]arachidonic acid. The kinetic of acylation of arachidonic acid was similar using cells after 3, 5 and 7 days of culture. However, the duration of the proliferation enhanced the acyltransfer to triglycerides and inhibited the transfer to phospholipids. After 8 h of incubation with [14C]arachidonic acid the radioactivity was predominantly recovered in the phosphatidylcholine and phosphatidylethanolamine species. Prelabeling-chase experiments indicated a preferential incorporation of arachidonic acid into phosphatidylcholine followed by a transfer of arachidonate to phosphatidylethanolamine.

Effect of duodenal ulcer healing induced by omeprazole and ranitidine on the generation of gastroduodenal eicosanoids, platelet-activating factor, pepsinogen A, and gastrin in duodenal ulcer patients

The effect of duodenal ulcer healing induced by omeprazole on gastroduodenal generation of eicosanoids, platelet-activating factor (PAF), pepsinogen A, and gastrin was evaluated. Sixty patients with endoscopically proven duodenal ulcer were randomized to receive 20 mg omperazole once daily or 300 mg ranitidine at bedtime for 2 weeks. Patients whose ulcers did not heal were treated for an additional 2 weeks. Endoscopic biopsy specimens and serum samples were obtained before and after treatment. There was no significant difference in the healing rate between the two treatment modalities. At 2 weeks healing rates were 60% and 56% in the omperazole and ranitidine groups, respectively, whereas at 4 weeks the respective healing rates were 96% and 86%. Ulcer healing induced by omeprazole and ranitidine was not accompanied by significant changes in mucosal leukotriene B4 or C4 generation. Mucosal PAF significantly decreased in patients treated with omeprazole for 4 weeks. In omperazole-treated patients there was a trend towards increase in mucosal prostaglandin E2 generation which was significant in the fundus after 4 weeks of treatment. After 2 weeks of omeprazole treatment, serum gastrin and pepsinogen A levels almost doubled when compared with their pretreatment levels. In conclusion, duodenal ulcer healing with 20 mg omeprazole daily is not superior to healing rates with 300 mg ranitidine at bedtime after both 2 and 4 weeks of treatment. In omeprazole-treated subjects ulcer healing was accompanied by a significant decrease in mucosal PAF generation and increased levels of serum gastrin and pepsinogen A.