Cyclooxygenases, Prostanoids, and Glomerular Injury

he association of prostanoid production with glomerular injury has been appreciated for more than 20 yr, well in advance of more recent studies that have characterized pathways that are associated with phospholipid hydrolysis and metabolism of free arachidonic acid to prostanoids. Release of arachidonic acid from phospholipids by phospholipase A2 gener- ally is the rate-limiting step in the synthesis of prostanoids (1). Free arachidonic acid is metabolized by cyclooxygenases (COX), fol- lowed by prostaglandin or thromboxane synthases. There are two COX genes, which result in transcription of COX-1 and COX-2 proteins, as well as COX-3, a splice variant of COX-1 (2). In many tissues, COX-1 is expressed constitutively, whereas COX-2 is in- duced in disease states, but in certain cells within the kidney, COX-2 also is expressed constitutively (3). Therefore, the view that COX-1 produces prostanoids that are necessary for normal phys- iologic functions whereas COX-2-derived prostanoids play a pathologic role does not necessarily apply to kidney physiology. Effects of prostanoids are mediated via specific receptors, and many of these have been identified in renal cells (4). In experimental animal models, administration of nephritogenic toxins or deposition of antibody and complement within glomeruli may induce changes in prostanoid generation by intrinsic glomerular cells. Alternatively, production of glomerular prostanoids may be altered after recruitment of leukocytes. A functional role for prosta- noids in glomerular injury first was suggested by studies that showed that nonsteroidal anti-inflammatory drugs can alter glomer- ular hemodynamics and urinary protein excretion in various protein- uric states. Studies in the 1980s and early 1990s used pharmacologic manipulation of prostanoid synthesis to study changes in glomerular permselectivity, and some but not all of these studies showed that administration of inhibitors of COX or thromboxane synthase can reduce proteinuria (reviewed in reference (5)). For example, these drugs lowered proteinuria in Adriamycin nephropathy in rats (a model of minimal-change disease/focal glomerulosclerosis), nephro- toxic nephritis in rats and mice (anti-glomerular basement mem- brane disease), passive Heymann nephritis (PHN) in rats (membra- nous nephropathy), and rat anti-Thy-1 nephritis (mesangiopro- liferative nephritis). When studied, reduction of proteinuria occurred independent of changes in glomerular hemodynamics. Another ap- proach to modulating prostanoid effects has been to shift production of dienoic prostanoids to inactive trienoic metabolites with diets that are rich in omega-3 fatty acids (e.g., fish oils). These diets have shown beneficial effects on renal function and/or proteinuria in animals with lupus nephritis, nephrotoxic nephritis, PHN, and focal glomer- ulosclerosis. Fish oil may alter humoral or cell-mediated immune responses or may affect glomerular cell function and permselectivity directly. Together, a number of studies that used distinct experimen- tal approaches support a role for prostanoids in exacerbating protein- uria in various glomerulopathies. It also should be noted that indo- methacin was shown to reduce proteinuria in human nephrotic syndrome (6), and a beneficial effect of fish oil diet on the progression of renal failure was reported in IgA nephropathy in humans (7). Since the late 1990s, a newer generation of studies has addressed the role of COX and prostanoids in glomerular injury. These studies followed the discovery of COX-2, characterization of prostanoid re- ceptors, and the advent of selective inhibitors. For example, glomeruli from rats with PHN express significantly more COX-1 and COX-2 and produce more prostanoids than normal rat glomeruli (8). The increase in prostanoids was attenuated with a COX-2 selective inhib-