Graphical illustration of the potential role of prostaglandin formation for dermal extravasation induced by activation of bradykinin type 2 receptors (B2). B2 translates the bradykinin signal via the Gq transduction pathway including generation of diacylglycerol (DAG) and inositol 1,4,5-trisphosphate (IP₃) resulting in an increase of the intracellular concentration of Ca²⁺. This ion binds and activates a variety of intracellular proteins. Among those, binding to calmodulin is an essential step to activate endothelial NO synthase (eNOS), while binding to phospholipase A2 (PLA₂) directly stimulates this enzyme to release arachidonic acid (AA) from the cell membrane and initiates the generation of PGH₂ by cyclooxygenases (COX). This prostaglandin functions as a substrate for a variety of prostaglandin synthases (PG-S) finally resulting in formation of specific prostaglandins. Once synthetized, these lipid mediators are released by the producing endothelial cell into small dermal blood vessel and activate their receptors located either in the producing cell (autocrine), or in adjacent cells (paracrine) or both. This activation of endothelial prostaglandin receptors might contribute to bradykinin induced extravasation, i.e. the increase of fluid transfer to the interstitial space (red intravascular dots indicate red blood cells).

Non-allergic angioedema is largely driven by increased plasma levels of bradykinin and over-activation of bradykinin receptor type II (B2), but the specific downstream signalling pathways remain unclear. The aim of this study was to identify signal transduction events involved in bradykinin-induced dermal extravasation.

Quantification of dermal extravasation was accomplished following intradermal (i.d.) injection of bradykinin or the B2 agonist labradimil in mice with endothelial NO-synthase (eNOS) deficiency and in C57BL/6J mice pre-treated with vehicle, NO-synthase or cyclooxygenase (COX) inhibitors. In the multicentre clinical study ABRASE, 38 healthy volunteers received i.d. bradykinin injections into the ventral forearm before and after oral treatment with the COX inhibitor ibuprofen (600 mg). The primary endpoint of ABRASE was the mean time to complete resolution of wheals (TTCR) and the secondary endpoint was the change of maximal wheal size.

Neither NOS inhibitors nor eNOS deficiency altered bradykinin-induced extravasation. In striking contrast, the COX inhibitors ibuprofen, diclofenac, SC560 and celecoxib significantly diminished this extravasation when given before injection. As for diclofenac, a similar but significantly lower effect was observed when given after i.d. injection of bradykinin. Similar results were obtained when bradykinin was replaced by labradimil. In volunteers, ibuprofen significantly reduced TTCR (P < 0.001) and maximal wheal size (P = 0.0044).

These data suggest that COX activity contributes to bradykinin-induced dermal extravasation in mice and humans. In addition, our findings may open new treatment options and point to a potential activity of drugs interfering with the release of the COX substrate arachidonic acid, e.g. glucocorticoids.