The EP4 receptor for prostaglandin E2 in glomerular disease: a good receptor turned bad?

Prostanoids generated by the metabolism of arachidonic acid through the cyclooxygenase (COX) pathway have diverse functions in health and disease.1 The kidney is prominent among the physiologic systems affected by these ubiquitous mediators. Along with effects to modulate renal blood flow, GFR, and sodium excretion, prostanoids affect the function of the glomerular filtration barrier. This is illustrated by clinical studies showing that nonsteroidal anti-inflammatory drugs (NSAIDs), which inhibit COX enzymes, attenuate the levels of proteinuria in patients with glomerular diseases.2,3 Because NSAIDs block production of all prostanoids, the specific prostanoids influencing glomerular function in this setting cannot be discerned. Moreover, because NSAIDs frequently cause adverse effects in patients with kidney disease, including acute deterioration in GFR and exacerbation of hypertension, this approach to antiproteinuric therapy has limited applications. In experimental models of glomerular disease, a number of studies have indicated beneficial effects of more specific pharmacologic inhibitors that block individual prostanoid synthetic pathways or receptors. More recently, this issue has been explored using modern molecular genetics. For example, Harris and associates4,5 showed that overexpression of the inducible COX isoform, COX2, in podocytes of transgenic mice enhances susceptibility to glomerular injury caused by adriamycin or puromycin. This was due, at least in part, to exaggerated production of thromboxane A2 and activation of the thromboxane-prostanoid (TP) receptor.6 These findings are consistent with older studies showing efficacy of thromboxane synthase inhibitors and TP receptor antagonists in experimental models such as adriamycin nephropathy7 and murine lupus nephritis.8 In this issue of JASN, Stitt-Cavanagh et al.9 implicate another prostanoid pathway in glomerular disease: Prostaglandin E2 (PGE2) acting through its EP4 receptor. This finding is surprising from at least two perspectives. First, in the studies of Harris and associates, deletion of EP4 receptors from podocytes had no effect on proteinuria and kidney injury in mice also overexpressing COX2.6 Second, the EP4 receptor has long been considered to have beneficial and protective effects in kidney disease and hypertension. Once COX and PGE synthases form PGE2 through the successive metabolism of arachidonic acid, it elicits its biological effects through a family of G protein– coupled receptors. These receptors, by convention designated EP (for E-prostanoid) receptors, are divided into four distinct pharmacologic classes: EP1 through 4.10,11 The EP4 receptor signals through Gs proteins and adenylyl cyclase.11 Because of the linkage between enhanced formation of cAMP and relaxation of bronchial and vascular smooth muscle, the EP4 receptor has been classically considered a relaxant receptor.11 In the kidney, vasodilator actions of PGE2, likely mediated by the EP4 receptor, are linked to protection of renal blood flow during states of compromise, such as congestive heart failure, volume depletion, or renal parenchymal disease. Inhibition of this protective, vasodilator pathway is an underlying mechanism of NSAID-associated acute renal failure. To examine the role of EP4 receptors in glomerular disease, Stitt-Cavanagh et al.9 used two approaches. First, they generated transgenic mice (EP4 ) expressing a modified version of the EP4 receptor in podocytes using nephron-specific promoter. The 5/6 nephrectomy model of renal ablation was used to induce proteinuria and glomerular injury. After partial renal ablation, levels of proteinuria were increased fourfold and mortality was dramatically enhanced in the EP4 mice. Although renal function was not measured, the extent of glomerular pathology in the EP4 mice was also accelerated. Because there were no differences in BP, at least as measured by tail-cuff monometry, the authors attribute the worsening of glomerular injury to consequences of increased activation of EP4 receptors on podocytes. There are, however, some potential interpretive problems with this part of the study. First, because the EP4 mice express a truncated version of the EP4 receptor that is resistant to agonistinduced desensitization, the specific applicability of these findings to the native EP4 receptor is not clear. In addition, concerns about potential nonspecific consequences Published online ahead of print. Publication date available at www.jasn.org.