Background Although concern remains about the athero-thrombotic risk posed by COX-2-selective inhibitors (COXIBs), the recent PRECISION trial demonstrated non-inferiority of moderate dose celecoxib when compared to naproxen and ibuprofen with respect to cardiovascular safety, with fewer actual CV events recorded for celecoxib. Moreover, celecoxib proved significantly safer than either comparator in regard to gastrointestinal events1. Given the markedly different cardiovascular risk associated with celecoxib and rofecoxib, we investigated the hypothesis that, in addition to cyclo-oxygenase inhibition, celecoxib specifically activates COX-2-independent AMP kinase (AMPK) signalling to exert protective effects in the vascular endothelium.
Objectives To investigate COX-2-independent vasculoprotective signalling pathways activated by celecoxib in human endothelium.
Methods In vitro studies of celecoxib, rofecoxib, ibuprofen and naproxen were performed on human umbilical vein and human aortic endothelial cells (HUVEC and HAEC). Inhibition of signalling pathways was achieved using siRNA. The vascular effects of celecoxib in vivo were studied in C57Bl/6 mice fed celecoxib (1000 ppm) or control chow (48 hrs). Aortic tissue was snap-frozen and sections studied by immunofluorescence confocal microscopy.
Results At therapeutically relevant concentrations celecoxib (1–10μ M) induced the vasculoprotective protein heme oxygenase-1 (HO-1) in HUVEC and HAEC (EC) (p<0.01). In contrast, rofecoxib and the commonly used non-selective NSAIDs ibuprofen and naproxen failed to induce HO-1. Celecoxib derivative 2,5-dimethyl-celecoxib (DMC), which lacks COX-2 inhibition, also upregulated HO-1, implicating a COX-2-independent mechanism. Immunoblotting demonstrated that celecoxib and DMC induce AMPKα(Thr172) and CREB-1(Ser133) phosphorylation leading to Nrf2 nuclear translocation (p<0.05). These responses were not seen with ibuprofen or naproxen, while siRNA depletion of AMPKα abrogated celecoxib-mediated CREB and Nrf2 activation (p<0.05). Acting via the same pathway, celecoxib induced additional cytoprotective genes including H-ferritin. In vivo, celecoxib similarly increased HO-1 and H-ferritin in murine aortic endothelium when compared to control-fed mice (p<0.05). Functionally, celecoxib treatment inhibited TNF-α-induced NF-κB p65(Ser536) phosphorylation by increasing AMPK activity. This attenuated VCAM-1 upregulation via induction of HO-1, as revealed by HO-1 siRNA (p<0.05). Similarly, celecoxib prevented the IL-1-mediated increase of IL-6 mRNA (p<0.01). These responses were not seen with ibuprofen or naproxen.
Conclusions Celecoxib induces anti-inflammatory, anti-oxidant proteins HO-1 and H-ferritin in human vascular endothelium via a novel AMPK-CREB-Nrf2-dependent pathway. This mechanism may contribute to the important and marked differences in cardiovascular risk between celecoxib and rofecoxib. Understanding mechanisms underlying NSAID heterogeneity may ultimately lead to the development of safer anti-inflammatory drugs.
Nissen SE et al. Cardiovascular Safety of Celecoxib, Naproxen, or Ibuprofen for Arthritis. N Engl J Med. 2016;375:2519–29.
Disclosure of Interest None declared