Bacterial lipopolysaccharide protects gastric mucosa against acute injury in rats by activation of genes for cyclooxygenases and endogenous prostaglandins.

Lipopolysaccharides (LPS) derived from gram-negative bacteria were reported to impair gastrointestinal mucosal integrity, but the results obtained are controversial. This study was undertaken to determine the effects of short-term administration of LPS on gastric secretion and gastric damage induced by 100% ethanol and to assess the role of the gene expression of two isoforms of cyclooxygenase (COX), constitutive (COX-1) and inducible (COX-2), and endogenous prostaglandins (PG) on these effects of LPS. Fasted rats received vehicle (control) or LPS (0.1-40 mg/kg i.g. or i.p.) without or with pretreatment with nonselective inhibitors of COX activity, indomethacin (5 mg/kg i.p.) and meloxicam (2 mg/kg i.g.), or the selective COX-2 inhibitor NS-398 (10 mg/kg i.g.), followed by intragastric application of 100% ethanol. The area of gastric lesions was determined by planimetry, gastric blood flow (GBF) was measured by the H2-gas clearance technique, mucosal PGE2 generation was measured by radioimmunoassay, and expression of COX-1 and COX-1 mRNA was determined by reverse transcription polymerase chain reaction (RT-PCR), quantitative RT-PCR with [32P]dCTP and immunohistochemistry. LPS applied intraperitoneally in various doses (0.1-10 mg/kg), dose dependently inhibited gastric acid and pepsin secretion and significantly reduced the area of gastric lesions induced by ethanol, and this was accompanied by an attenuation of the ethanol-induced fall in GBF and increased mucosal generation of PGE2. LPS applied in higher doses, such as 20 or 40 mg/kg, that caused systemic hypotension failed to protect the mucosa against 100% ethanol. Suppression of mucosal PGE2 generation by indomethacin or meloxicam, significantly reduced the inhibitory action of LPS on gastric secretion and abolished LPS-induced gastroprotection and elevation of GBF. NS-398 did not influence PGE2 generation, but significantly attenuated the protection and hyperemia induced by LPS suggesting that COX-2-derived products play an important role in gastroprotection. The expression of COX-1 mRNA, as determined by RT-PCR, quantitative RT-PCR and immunohistochemistry was found in intact gastric mucosa and after LPS administration. In contrast, the expression of COX-2 mRNA was undetectable in intact gastric mucosa but appeared in this mucosa 2, 4 and 8 h after LPS administration. COX-2 mRNA was not detected in rats treated with ethanol but, when LPS was applied before ethanol, the enhanced expression of COX-2 was detected without affecting COX-1 mRNA expression. We conclude that acute parenteral LPS affords gastroprotection against ethanol-damage through an increase in gastric microcirculation and overexpression of COX-2 and enhanced endogenous PG release.