Cyclooxygenase-2 (COX-2) is a powerful mediator of hepatic ischemia/reperfusion injury (IRI). While COX-2 depletion significantly ameliorates liver IRI, COX-2 inhibitors present clinical limitations; Therefore, there is a need to identify which COX-2 sources lead to tissue damage. Observations that leukocytes of myeloid origin are major sources of COX-2 support a rationale to dissect the function of myeloid COX-2 in hepatic IRI. Methods: COX-2 floxed mice were crossed to LysM Cre mice to obtain myeloid-specific COX-2 knock-out (COX-2–M/–M; COX-2flox/flox, LysMCre/+) and wild-type (WT; COX-2flox/flox, LysM+/+) mice. LPS-activated COX-2–M/–M macrophages don’t express COX-2. All mice were subject to 90 min of 70% liver ischemia followed by reperfusion. Results: COX-2 expression was depressed in COX-2–M/–M livers (2.2-fold;p<0.02), particularly at 24h post-IRI. All livers developed extensive necrosis and vascular congestion post-IRI; AST (6h:4,146±928 vs. 4,652±1,025; 24h:3,707±453 vs. 3,853±839) and ALT (6h:18,281±4,720 vs. 14,425±1,957; 24h:5,000±2,258 vs. 5,053±1,338) levels (U/L) were similar in COX-2–M/–M and WT mice. We previously showed that total-COX-2 depletion decreases neutrophil infiltration/cytokine release post-IRI. Alternatively, depletion of myeloid COX–2 was ineffective in depressing leukocyte infiltration/activation. Indeed, COX-2-M/-M livers showed an increase in IFN-Α (p<0.02) and IL-6 (p<0.001) post-IRI. COX-2–M/–M and WT livers showed similar Ly-6G-neutrophil (6h:75±18 vs. 81±28; 24h:157±39 vs. 185±30) infiltration post-IRI; Mac-1-macrophage infiltration was similar at 6h (78±12 vs. 76±18) and slightly reduced at 24h (117±15 vs. 170±38;p<0.05). To support that liver damage is independent of myeloid COX-2, COX-2–M/–M mice were treated with a COX-2 inhibitor (celecoxib;100mg/Kg) prior to IRI. Celecoxib significantly improved COX-2–M/–M liver function (AST: 3,985±443 vs. 5,666±831; ALT: 4,788±871 vs. 10,846±2321;p<0.05) and histological preservation, compared to vehicle-treated COX-2–M/–M mice. Conclusions: Our results, using mice in which COX-2 has been selectively ablated in myeloid cells, clearly show that myeloid COX-2 is not responsible for hepatic IRI. Moreover, they provide evidence that COX-2 mediated liver IRI results from COX-2 being predominantly expressed by sources other than myeloid cells. These findings are of particular importance for the development of potential COX-2 targeted therapies in hepatic IRI.

« Back to 2013 American Transplant Congress