Antithrombin reduces ischemia/reperfusion-induced liver injury in rats by activation of cyclooxygenase-I

This study was conducted to determine which isoform of cyclooxygenase (COX) is more significantly involved in the antithrombin (AT)-induced increase in prostaglandin production in the liver of rats, subjected to hepatic ischemia/reperfusion (I/R). Hepatic tissue levels of 6-keto-PGF(1alpha), a stable metabolite of prostacyclin (PGl(2)), and PGE(2) were transiently increased 1 hour after reperfusion. Thereafter, hepatic PGE(2) levels were gradually increased until 6 hours after reperfusion, while hepatic 6-keto-PGF(1alpha) levels were decreased to the pre-ischemia levels at 6 hours after reperfusion. AT significantly enhanced increases in hepatic tissue levels of 6-keto-PGF(1alpha) and PGE(2) seen 1 hour after reperfusion, while it inhibited increases in hepatic PGE(2) levels seen 6 h after reperfusion. Neither dansyl-Glu-Gly-Arg-chloromethyl ketone-treated factor Xa (DEGR-Xa), a selective inhibitor of thrombin generation, nor Trp(49)-modified AT which lacks affinity for heparin, showed any effects on these changes. Pretreatment with indomethacin (IM), a non-selective inhibitor of COX, inhibited AT induced increases in hepatic tissue levels of 6-keto-PGF(1alpha) and PGE(2) seen 1 hour after reperfusion, whereas pretreatment with NS-398, a selective inhibitor of COX-2, did not. The increase in hepatic tissue blood flow and inhibition of hepatic inflammatory responses seen in animals given AT were reversed by pretreatment with IM, but were not affected by pretreatment with NS-398. Administration of iloprost, a stable analog of PGl(2), and PGE(2) produced effects similar to those induced by AT Increases in hepatic tissue levels of PGE(2) 6 hours after reperfusion were inhibited by pretreatment with NS-398. Although AT did not affect COX-1 mRNA levels 1 hour after reperfusion, it inhibited the I/R-induced increases in hepatic tissue levels of both PGE(2) and COX-2 mRNA 6 hours after reperfusion. These observations strongly suggested that AT might reduce the I/R-induced liver injury by increasing the production of PGl(2) and PGE(2) through activation of COX-1. Furthermore, since TNF-alpha is capable of inducing COX-2, inhibition of TNF-a production by AT might inhibit COX-2-mediated PGE(2) production. These effects induced by AT might contribute to its anti-inflammatory activity.