Activation and potential interactions between the arachidonic acid and L-arginine:nitric oxide pathways in glomerulonephritis

Kidney International, Vol. 53 (1998), pp. 540 –547

PERSPECTIVES IN BASIC SCIENCE

Activation and potential interactions between the arachidonic acid and L-arginine:nitric oxide pathways in glomerulonephritis ELIAS A. LIANOS Nephrology Division, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin, USA

The arachidonic acid metabolites, collectively referred to as eicosanoids, and nitric oxide (NO) originating from the amino acid L-arginine following activation of nitric oxide synthase (NOS), have recently become the focus of intense investigation insofar as the biology of glomerulonepritis is concerned. Three key reasons are: (a) it was realized that certain eicosanoids (those of the arachidonate cyclooxygenase pathway) and NO can be synthesized in a constitutive or inducible manner; (b) while their constitutive release plays a homeostatic role in the kidney, their inducible synthesis can promote cell injury; and (c) regulatory interactions may occur between NO and certain eicosanoids and result in amplification of injury. A number of recent reviews have dealt with the synthesis and origin of eicosanoids and of NO in the normal and diseased kidney [1, 2]. This review emphasizes regulatory interactions that may occur between eicosanoids and NO in glomerulonephritis, and addresses issues regarding the cytotoxic potential of such interactions. PRESENCE AND PHYSIOLOGIC ROLE OF THE L-ARGININE:NITRIC OXIDE PATHWAY IN THE KIDNEY Whereas the role of eicosanoids in renal physiology and pathology has been extensively explored in the last decade, the presence and significance of the L-arginine:NO pathway in the kidney has only recently become apparent. Nitric oxide is a reactive radical gas formed from the terminal guanidino nitrogen atom of the amino acid L-arginine. This reaction is catalyzed by the enzyme nitric oxide synthase (NOS). At least three NOS enzymes have been described. The endothelial and neuronal isoforms are constitutively expressed and are quiescent until activated by increased Ca21 levels in a calmodulin-dependent manner. The inducible NOS (iNOS) is expressed following transcriptional activation by cytokines or lipopolysaccarides (LPS) [3]. Following its induction, iNOS remains active for sustained periods in a Ca21-independent manner. Production of NO from the constitutive forms of NOS is short-lived and regulates a number of homeostatic functions including modulation of renovascular tone, renin secretion, the tubuloglomerular feedback response, and sodium excretion [1, 4]. Key words: glomerulonephritis, nitric oxide, arachidonic acid, L-arginine:NO pathways, eicosanoids, injury, cytotoxicity. Received for publication May 6, 1996 and in revised form September 16, 1997 Accepted for publication September 24, 1997

© 1998 by the International Society of Nephrology

The topography of NOS synthase isoforms in the nephron has recently become ellucidated. Terada et al demonstrated the presence of constitutive NOS in the glomerulus where they also found abundant expression of soluble guanylate cyclase, the well established acceptor of NO [5]. Mohaupt and co-investigators have recently reported the expression and induction of mRNAs encoding two inducible nitric oxide synthases in the rat nephron and found that endotoxemia and cytokines differentially induce their expression [6]. Nitric oxide originating from constitutive NOS plays an important homeostatic role in the glomerular microcirculation. Thus, inhibition of NOS in the normal rat results in glomerular arteriolar vasoconstriction and glomerular hypoperfusion. Efferent arteriolar resistance is disproportionately increased resulting in glomerular capillary hypertension. In addition, the glomerular capillary ultrafiltration coefficient, Kf, drops [7]. These observations have established NO as a basal tonic dilator of the glomerular microcirculation. This effect is further supported by the observation that cultured glomerular endothelial cells release NO [8]. COACTIVATION OF THE ARACHIDONIC ACID AND THE L-ARGININE:NITRIC OXIDE PATHWAYS IN GLOMERULONEPHRITIS Using experimental models of glomerulonephritis investigators have characterized conversion of arachidonic acid to various eicosanoids via the cyclooxygenase and lipoxygenase pathways, and generation of nitric oxide from L-arginine via nitric oxide synthase. In models of glomerulonephritis induced by antibody against the glomerular basement membrane, the glomerular epithelial cell and the glomerular mesangial cell, it was demonstrated that the synthesis of arachidonate cyclooxygenation and lipoxygenation eicosanoids in isolated glomeruli is enhanced [9]. The most abundant arachidonate cyclooxygenation eicosanoids are the prostaglandins (PG)E2 and PGF2a, and thromboxane (Tx)A2 [10]. The most abundant arachidonate lipoxygenation eicosanoids are leukotriene (LT)B4 and 5- and 12-hydroxyeicosatetraenoic acid (HETE) [11]. In the same models of injury it was demonstrated that generation of nitric oxide from L-arginine in isolated glomeruli is also increased [12]. The temporal relationship between glomerular eicosanoid synthesis and NO generation in the course of glomerular injury induced by anti-GBM antibody is diagramatically shown in Figure 1. In the same figure, changes in glomerular filtration rate (GFR), infiltration of glomeruli by platelets and leukocytes, and proliferation and accumulation of extracellular

540

541

Lianos: L-arginine:nitric oxide pathways in GN

Fig. 1. Time course of synthesis of eicosanoids and generation of nitric oxide (NO) in glomeruli isolated following immune injury induced by anti-glomerular basement membrane (GBM) antibody in the rat. Shown in parallel are changes in GFR and in glomerular histopathology (influx of platelets and macrophages, proliferation and accumulation of extracellular matrix). Patterns of eicosanoid synthesis and of NO generation are presented diagramatically and are based on observations reported in references 10 to 16. Enhanced eicosanoid synthesis peaks when GFR falls. While synthesis of prostaglandins (PG’s), thromboxane (Tx)A2 and 12hydroxyeicosatetraenoic acid (HETE) is sustained in the course of immune injury, that of 5-HETE and LTB4 is short-lived and becomes impaired at early stages of injury. GFR subsequently recovers despite sustained production of vasoconstrictor eicosanoids (TxA2). The recovery in GFR coincides with peak generation followed by sustained release of NO.

matrix are also shown. The following points are emphasized: (a) Enhanced glomerular synthesis of both eicosanoids and NO occurs and peaks at very early stages of the immune injury (within the first 24 hr). (b) The peak of eicosanoid synthesis coincides with a marked drop in GFR. Glomerular filtration rate subsequently recovers partially despite continued synthesis of vasoconstrictor eicosanoids (TxA2). This recovery is coincidental with peak generation of NO. (c) At stages of glomerulonephritis beyond the early phase (24 hr), a dichotomy in glomerular eicosanoid synthesis occurs; whereas synthesis of PGs, TxA2 and 12-HETE remains enhanced, and that of arachidonate 5-lipoxygenation eicosanoids (5-HETE and LTB4) becomes impaired. Following its peak, glomerular generation of NO is sustained and parallels that of TxA2 and 12-HETE. In glomerulonephritis, eicosanoids and NO may originate either from intrinsic glomerular cells or from infiltrating platelets and leukocytes. Using platelet depletion and leukocyte depletion studies, it was shown that activated infiltrating macrophages account to a large extent for the glomerular origin of 5-HETE and LTB4, whereas infiltrating platelets entirely account for the origin of 12-HETE and to a lesser extent for that of TxA2 [13]. Using the same approach it was shown that infiltrating macrophages are the source of NO in glomeruli isolated from rats with experimental membranous nephropathy [14] and mesangioproliferative nephropathy [15]. Thus, the infiltrating activated macrophage has been identified as a key site where synthesis of both eicosanoids and NO occurs in glomerulonephritis. However, a possible contribution of mesangial cells should not be ignored [16]. An important aspect of the arachidonic acid and of the Larginine:NO pathways is that cyclooxygenation of arachidonic acid to prostaglandins or conversion of L-arginine to NO may occur either by constitutive or by inducible isoforms of cyclooxygenase (COX) or nitric oxide synthase (NOS), respectively. This is of major importance in the biology of glomerulonephritis because activation of the inducible isoforms of these enzymes can result in

sustained synthesis of proinflammatory prostaglandins and in sustained high-output generation of NO. Macrophages infiltrating glomeruli in the course of glomerulonephritis are likely to express inducible COX and NOS as a result of their activation by proinflammatory cytokines following adherence to the glomerular capillary endothelium. Indeed, in forms of experimental glomerulonephritis in which glomerular infiltration by macrophages is a prominent feature, iNOS was immunocytochemically localized in these cells [17]. In the same forms, enhanced expression of inducible cyclooxygenase [also referred to as COX-II or prostaglandin H synthase (PGHS-II) ] was also demonstrated [18]. Cytooxygenase (COX)-II expression has not only been detected in isolated glomeruli but also in mesangial cells after exposure to cytokines [19 –21]. PUTATIVE INTERACTIONS BETWEEN THE ARACHIDONIC ACID AND THE L-ARGININE:NITRIC OXIDE PATHWAYS In its capacity as a reactive radical gas, NO binds avidly to iron and iron-containing enzymes. A multitude of effects mediated by NO have been attributed to its high affinity for iron-containing substrates. For example, the ability of NO to inhibit platelet aggregation and to relax vascular smooth muscle has been attributed to NO binding to the heme-Fe11 of the soluble guanylate cyclase leading to stimulation of this enzyme and subsequent increase in the levels of cGMP [22]. Similarly, NO interacts with hemoglobin and with iron-sulfur centers in key enzymes of the respiratory cycle and DNA synthesis [23]. Arachidonate cyclooxygenase may become a target for NO because it contains an iron-heme center at the site that confirms its hydroperoxidase function [24]. Binding of NO to this center may enhance cyclooxygenase activity. Evidence that this may indeed occur has emerged recently. The initial observations pointing to up-regulation of

542

Lianos: L-arginine:nitric oxide pathways in GN

cyclooxygenase activity by NO were based on macrophages stimulated with endotoxin. In these cells, endotoxin stimulated iNOS as well as the inducible isoform of cyclooxygenase (COX-II) resulting in large production of NO and of prostaglandins, respectively. Pharmacologic inhibition of iNOS not only reduced the release of NO but also that of prostaglandins [25]. In these studies the iNOS inhibitors employed were specific in that they did not have an effect on COX activity, nor did they alter the level of induction of COX-II. To further explore the role of NO in up-regulating COX-II activity, interleukin (IL-1B) stimulated human fibroblast cells that did not possess an endogenous Larginine:NO pathway were exposed to exogenous NO. Exposure of IL-1B stimulated fibroblasts to either NO or to two NO donors, sodium nitroprusside and glyceryltrinitrite, increased COX activity by at least fourfold and production of prostaglandins [25]. This was independent of an effect of NO on the soluble guanylate cyclase. A likely mechanism by which NO activates COX is direct stimulation of the enzyme. Indeed, NO directly increases COX activity of microsomal sheep seminal vessels as well as the activity of murine recombinant COX-I and COX-II [25]. The exact molecular mechanism by which NO activates COX remains to be identified. In addition to binding to the heme-iron center of COX, NO interacts with oxygen-derived free radicals such as the superoxide ion (O22), which is also released at sites of inflammatory injury. Interaction between NO and O22 leads to the formation of peroxynitrite (ONOO2), which can then decompose to form the hydroxyl radical, OHz. Thus, NO may augment COX activity by acting as an anti-oxidant (removal of O22) or by generating ONOO2 and OHz that can subsequently activate COX. Indeed, it was shown that O22 and OHz can modulate the COX pathway [26]. Finally, it was recently shown that NO nitrosylates cystein residues in the catalytic domain of COX enzymes [27]. This points to an additional/alternative mechanism whereby NO can stimulate COX. In vivo studies supporting regulatory interactions between the arachidonic acid and the L-arginine:NO pathways are also emerging. In endotoxin-induced septic shock in rats, inhibition of NOS using aminoguanidine reduces the release of prostacycline (PGI2) by approximately 85% in plasma and urine [28]. Rat islet cells or vascular smooth muscle cells stimulated with IL-1B release NO and prostaglandins; inhibition of NO production blocks PGE2 release [29, 30]. In bovine coronary microvascular endothelial cells NO activates constitutive cyclooxygenase leading to increased production of prostacyclin [31]. In a rat model of experimental glomerulonephritis resembling human forms of rapidly progressive crescentic nephritis, iNOS inhibition using a relatively selective pharmacologic inhibitor results in reduced synthesis of the arachidonate cyclooxygenation products, PGE2 and PGI2, in isolated glomeruli without an effect on TxA2 [this issue, 65]. Arachidonate lipoxygenases may also become NO targets because they contain non-heme iron [32]. Indeed, bond formation between NO and the non-heme iron of lipoxygenase was demonstrated using purified soybean 15-lipoxygenase and methods of electron spin resonance spectroscopy [33]. Whether this binding results in up- or down-regulation of lipoxygenase activity is of apparent importance in forms of glomerulonephritis characterized by enhanced arachidonate lipoxygenation to hydroxyeicosatetraenoic acids (HETE) and leukotrienes and enhanced generation of iNOS-derived NO. Indeed, in experimental rapidly

progressive nephritis, pharmacologic inhibition of iNOS results in reduced synthesis of 5-HETE and 15-HETE and of LTB4 [65]. Evidence that eicosanoids may regulate generation of NO is also emerging. Arachidonic acid and the leukotrienes (LTB4) stimulate generation of NO in neutrophils and platelets in amounts sufficient to inhibit platelet aggregation [34]. On the other hand, eicosanoids can modulate NOS expression. In this regard, PGE2 negatively modulates iNOS induction in glomerular mesangial cells in response to interleukin-1 [35]. In experimental rapidly progressive glomerulonephritis inhibition of cyclooxygenase or of 5-lipoxygenase enhance expression of iNOS in isolated glomeruli [65]. It becomes clear from the evidence reviewed above that regulatory interactions between arachidonate cyclooxygenation or lipoxygenation pathways and the L-arginine:NO pathway may develop in the glomerular milieu in the course of glomerulonephritis. Such interactions may potentiate adverse hemodynamic and cytotoxic effects thereby amplifying the extent of renal hemodynamic impairment and glomerular cell injury. For example, NO-driven stimulation of proinflammatory eicosanoids (TxA2 and leukotrienes) would promote platelet aggregation (TxA2), leukocyte adhesion to the glomerular capillary endothelium via increased expression of adhesion molecules (TxA2 and leukotrienes), and constriction of the glomerular arterioles (TxA2, LTD4). On the other hand, regulation of iNOS by eicosanoids would modulate NO generation. A diagrammatic presentation of interactions that may occur between the arachidonate cyclooxygenase and lipoxygenase pathways and the L-arginine:NO pathway is shown in Figure 2. RENAL HEMODYNAMIC EFFECTS OF EICOSANOIDS AND NITRIC OXIDE IN GLOMERULONEPHRITIS Eicosanoids and nitric oxide synthesized and released in the glomerular milieu in the course of glomerulonephritis may have opposing or additive effects on the glomerular microvasculature. NO, PGE2 and PGI2 preserve GFR and renal blood flow (RBF) in experimental models of glomerulonephritis via a dilatory effect on the renal vasculature. This was shown in studies assessing the impact of cyclooxygenase (COX) or nitric oxide synthase (NOS) inhibition on changes in GFR and RBF. Cyclooxygenase (COX) inhibition using non-steroidal anti-inflammatory drugs (NSAIDs) reduces glomerular synthesis of PGE2 and PGI2 and markedly decreases GFR and RBF in anti-GBM antibody-induced glomerulonephritis in the rat [36]. In the same model, NOS inhibition using systemic administration of the L-arginine analog SG-monomethyl-L-arginine results in reduced glomerular plasma flow rate, a drop in the glomerular capillary ultrafiltration coefficient, Kf, and an increase in efferent arteriolar constriction resulting in intraglomerular capillary hypertension and exacerbation of proteinuria [37]. These adverse effects may occur due to an unopposed effect of vasoconstrictors by NO as a result of inhibition of renal NO synthesis. Established vasoconstrictors include the eicosanoids thromboxane A2 and LTD4. In anti-GBM or anti-Thy 1 antibody mediated glomerulonephritis, these two eicosanoids mediate the decrease in GFR by increasing the afferent arteriolar resistance and decreasing the glomerular capillary ultrafiltration coefficient [38, 39]. The changes in GFR that occur following anti-GBM antibody-induced glomerulonephritis are shown in Figure 1. The marked decrement in GFR immediately following

Lianos: L-arginine:nitric oxide pathways in GN

543

Fig. 2. Potential interactions between the arachidonic acid and the L-arginine:nitric oxide (NO) pathways in glomerular immune injury. Following injury, macrophages infiltrating glomeruli are activated and generate, (1) prostaglandins (PGs), thromboxane (Tx)A2 and leukotrienes (LTs) from arachidonic acid via cyclooxygenase and lipoxygenases; and (2) nitric oxide from L-arginine via inducible nitric oxide synthase (iNOS). A positive regulatory effect of nitric oxide on arachidonate cyclooxygenase is well established (1). Evidence for a positive regulatory effect of nitric oxide on lipoxygenases is also emerging. Evidence also points to a downregulatory effect of prostaglandins and leukotrienes on expression of iNOS. Solid arrows indicate definitive evidence. Broken arrows indicate emerging evidence.

onset of injury (1 hr) is mediated by TxA2 and LTD4 [38, 39]. Glomerular filtration rate subsequently recovers despite sustained increments in the vasoconstrictor TxA2, indicating that the vasoconstrictor effect of this eicosanoid is opposed. Potential opposing factors are vasodilatory prostaglandins (PGE2 and PGI2) the glomerular synthesis of which is also enhanced at this stage of the immune injury [10], and NO the generation of which peaks at the point of recovery in GFR (Fig. 1) [12]. In other forms of glomerular immune injury, such as the one induced by anti-Thy 1 antibody, enhanced glomerular synthesis of TxA2 also occurs but is not accompanied by increased generation of the vasodilatory prostaglandins PGE2 and PGI2 [13]. Yet, GFR still recovers despite marked increases in glomerular TxA2 levels. This observation questions the role of vasodilatory prostaglandins in opposing the vasoconstrictor effect of TxA2 on the glomerular microvasculature following immune injury and points to NO as the likely vasodilator that may oppose the effect of TxA2. However, direct evidence for such opposing interaction between TxA2 and NO at the level of the glomerular microvasulature is lacking. EICOSANOIDS AND NITRIC OXIDE AS MEDIATORS OF CYTOTOXICITY IN GLOMERULONEPHRITIS A sustained synthesis of arachidonate cyclooxygenation products (PGs and TxA2) and a sustained generation of NO occur in the course of glomerular immune injury (Fig. 1). This is likely due to activation of the inducible form of cyclooxygenase (COX-II) and of the inducible isoform of nitric oxide synthase (iNOS) [16, 17]. Cyclooxygenase (COX) catalyzes the committed step in prostaglandin biosynthesis from arachidonic acid, specifically the oxygenation of arachidonic acid to PGG2 and reduction of PGG2 to PGH2. It has recently become apparent that there are two different isoforms of COX, referred to as COX-I and COX-II, and encoded by separate genes [40]. Both COX-I and COX-II probably catalyze reactions that are mechanistically identical and

the two enzymes exhibit the same Km values for their substrate, arachidonic acid. The biological basis for the existence of two distinct cyclooxygenases is not well understood; COX-I is constitutively expressed in most tissues and is believed to mediate “housekeeping” functions. Surprisingly, however, COX-I deficient mice are generally healthy and develop normally. Their kidneys show minimal abnormalities, mainly scarce foci of immature tubules [41]. In contrast, COX-2 deficient mice die early (by eight weeks). Pathologically, their tissues are normal except for kidney tissue, the maturation of which stops early after only a small number of nephrons have developed. The majority of glomeruli and tubules are immature while the residual nephrons develop glomerular sclerosis, interstitial inflammation and progression to end stage disease [42]. Different patterns of expression of COX-I and COX-II have been identified in the kidney, that is, COX-I is mainly expressed in the collecting tubule while COX-II in the macula densa [43]. The most dramatic difference between the two COX isoenzymes relates to regulation of their expression. While expression of COX-I is maintained at constant levels, that of COX-II can be dramatically increased during inflammation [44], an observation that has led to the proposal that activation of COX-II results in generation of proinflammatory prostaglandins. The induction of COX-II by proinflammatory cytokines and during inflammatory responses in cells that mediate inflammation, that is, the macrophage, and the reduced expression of this enzyme by antiinflammatory agents such as corticosteroids, provides circumstantial evidence to support this role. It was originally presumed that COX-I is the site of action of non-steroidal anti-inflammatory drugs (NSAIDS). However, the discovery of the COX-II isoenzyme and its association with inflammation has suggested that the therapeutic target of NSAIDS functioning in their anti-inflammatory capacity is COXII. This observation has advanced the attractive concept that use of non-steroidal anti-inflammatory agents that can selectively

544

Lianos: L-arginine:nitric oxide pathways in GN

inhibit COX-II would result in a beneficial anti-inflammatory effect without the adverse consequences of COX-I inhibition, such as peptic ulcer formation and renal functional impairment (decreased RBF and GFR). This attractive concept has not yet been tested in inflammatory processes involving the kidney, such as glomerulonephritis. Preliminary evidence does indicate that in experimental glomerulonephritis there is up-regulation of glomerular COX-II expression [17]. Whether this promotes glomerular inflammatory injury is unknown. For example, in mice lacking COX-II, the severity of inflammatory responses to tetradecanoyl phorbol acetate or to arachidonic acid is not diminished in a model of acetone-induced ear inflammation [42]. As mentioned earlier, NO can up-regulate the activity of both COX-I and COX-II. It is, therefore, quite possible that in the course of glomerular immune injury the sustained high-output generation of NO originating from activation of iNOS can “drive” synthesis of proinflammatory prostaglandins and TxA2 by simulating COX-II. This could amplify the inflammatory injury to intrinsic glomerular cells. Whether sustained generation of proinflammatory eicosanoids in the glomerular milieu in the course of glomerulonephritis is injurious to glomerular cells is debatable. Solid evidence that proinflammatory eicosanoids including PGs, TxA2, and the leukotrienes can directly effect cell injury is lacking. Indeed, in other inflammatory conditions such as osteoarthritis, inhibition of cyclooxygenase using NSAIDS reduces synthesis of PGs and TxA2 and alleviates symptoms. Yet, the destruction of joints by the inflammatory process is not prevented. In contrast, iNOS-derived NO can directly effect cell injury. This can occur either directly by NO itself or by oxidation products of NO. Examples of direct toxic effects mediated by NO itself include: (a) inhibition of cytochrome P450 [45]; (b) reversible inhibition of ribonucleotide reductase, which is a critical enzyme for synthesis of DNA precursors [46]; and (c) irreversible inhibition of cytochrome C oxidase, which may promote leakage of superoxide from the mitochondrial electron transport chain [47]. A key mechanism of injury attributed to production of nitric oxide is its reaction with superoxide (O22) to form peroxynitrite (ONOO2). Peroxynitrite is not a free radical and its formation is irreversible. In infiltrative forms of glomerulonephritis, both NO and superoxide can be released in the glomerular milieu from infiltrating leukocytes. Therefore, formation of peroxynitrite is likely to occur. However, the concentration of superoxide is generally kept low by its scavenger superoxide dismutase the intracellular concentration of which is quite high (estimated 4 to 10 mM in brain and liver). Thus, very high concentrations (mM levels) of NO must be attained in order for NO to effectively compete with superoxide dismutase and react with superoxide to form peroxynitrite. This may only occur following activation of the inducible NOS, which, as reviewed above, results in sustained high-output generation of NO. Whether peroxynitrite is indeed generated in the glomerular milieu in the course of glomerulonephritis is unknown. One indirect approach to detect its formation is to search for an in vivo “fingerprint” of interaction between peroxynitrite and protein tyrosine residues. Nitration of protein tyrosine residues occurs following formation of peroxynitrite and its subsequent cleavage to produce hydroxyl anion and nitronium ion (NO21). This cleavage is catalyzed by transition metals including metal centers of superoxide dismutase and myeloperoxidase [48]. The nitronium ion formed in this manner can nitrate protein tyrosine residues to form 3-nitrotyrosine which is regarded

as the fingerprint of peroxynitrite in vivo. Detection of nitrotyrosine-containing proteins has recently become possible using antibodies against these proteins. Using these antibodies, it was demonstrated that extensive nitration of protein tyrosine residues occurs in human atherosclerotic lesions [49]. Once formed as a result of the reaction between NO and superoxide, peroxynitrite can be particularly toxic. Thus, protonation of peroxynitrite results in its isomerization to peroxynitrous acid. This can subsequently yield hydroxyl radical and nitrogen dioxide. In addition to generating these two toxic radicals, peroxynitrous acid can itself oxidize iron/sulfur centers, zinc fingers and protein thiols. Examples of peroxynitrite-induced cytotoxicity include: (a) inactivation of a-1-anti-protease inhibitor [50], which is a protease inhibitor that plays an important protective role in the lung against neutrophile-derived proteases; (b) damage of amiloride-sodium sensitive channels [51]; (c) mitochondrial injury the extent of which is more severe than that induced by NO [52]; and (d) oxidation of low density lipoproteins to a form that facilitates their uptake by activated macrophages [53]. IS INHIBITION OF NITRIC OXIDE A REASONABLE GOAL IN GLOMERULONEPHRITIS? As reviewed above, generation of NO in the course of glomerulonephritis is important in preserving GFR. Yet, NO is most likely derived from activation of iNOS [14, 15]. Therefore, its generation can be sustained and of high output and can cause cell injury by the mechanisms reviewed above. A critical dilemma thus emerges: should generation of NO (particularly of iNOS-derived NO) be left unopposed to preserve GFR or should it be inhibited in order to protect glomerular cells from oxidative injury at the risk of causing renal vasoconstriction and intraglomerular or systemic hypertension? An additional risk of inhibiting glomerular generation of NO in infiltrative forms of glomerulonephritis is that this maneuver may enhance platelet and leukocyte adherence to the glomerular capillary endothelium thereby promoting platelet and leukocyte-induced injury. A number of recent studies support both the argument of leaving NO generation unopposed and that of inhibiting NO generation. In anti-GBM antibodyinduced glomerulonephritis, inhibition of NO synthesis using systemic administration of nonselective NOS inhibitors or systemic NOS substrate (L-arginine) depletion, results in intraglomerular hypertension [37] and worsening of proteinuria [37, 54]. Contrasting these findings are those in murine lupus nephritis in which long-term pharmacologic NOS inhibition improved proteinuria and renal histopathology [55]. Careful experiments are needed to clarify whether inhibition of NO will have a net beneficial or detrimental effect in glomerulonephritis. The design of such experiments should include attempts to determine whether NO generation in immunologically injured glomeruli reaches biologically relevant levels. One should not underestimate the efficiency with which nitric oxide can be removed and neutralized immediately after its generation at the site of inflammatory injury, and the fact that nitric oxide itself is far less reactive and toxic at biologically relevant concentrations than in vitro experiments suggest. Most studies in the literature in general and in the nephrology literature in particular, have not measured the biologically relevant concentrations of NO in glomeruli and may have erroneously assumed that NO is being continuously generated because of its short half-life. However,

Lianos: L-arginine:nitric oxide pathways in GN

NO is an uncharged molecule that can diffuse through most cells and tissues with little consumption or direct reaction [56]. A very efficient removal mechanism of NO is generated at the site of inflammation and is provided by red blood cells which rapidly destroy NO following its avid binding to red blood cell oxyhemoglobin. Thus, red blood cells can rapidly reduce intracellular concentrations of NO in NO-producing cells or extracellular concentrations of NO at a site of inflammatory injury. Glomeruli are typically hyperemic in glomerulonephritis. Therefore, a quite efficient removal of NO generated within the glomerular milieu can be expected. Whether NO concentrations in nephritic glomeruli reaches levels sufficient to cause cell injury via formation of peroxynitrite is unknown. Most studies have measured ex vivo production of NO in nephritic glomeruli by assessing concentration of nitrite/nitrate. In these studies, prolonged (up to 48 hr) incubations are usually required in the presence of the NOS substrate, L-arginine [12–14]. This approach may not provide an estimate of the biologically relevant concentrations of NO attained within nephritic glomeruli. For example, NO concentrations of about 5 nM cause vasodilation by activating guanylate cyclase and can, therefore, be assumed as “homeostatic.” In contrast, mM concentrations of NO are required to compete with superoxide dismutase and react with superoxide to generate peroxynitrite. Are such concentrations of NO achieved in nephritic glomeruli? Newer direct NO detection methods using chemiluminescence detectors may provide accurate estimates of “glomerular” NO concentrations. The demonstration that expression of iNOS is enhanced in glomerulonephritis raises the obvious question of whether selective inhibition of this NOS isoform while leaving the constitutive isoform intact would have a beneficial anti-inflammatory effect. Increasingly selective iNOS inhibitors are now becoming available and may serve as powerful tools to definitively address the role of iNOS-derived NO in the pathobiology of glomerulonephritis. Such inhibitors have shown promising results in other forms of inflammation, specifically in adjuvant-induced arthritis in rats and in carrageenan-induced subcutaneous inflammation. In these models, it was shown that iNOS inhibition attenuates nitrite accumulation, leukocyte infiltration and the histopathological severity of the lesions [57, 58]. Moreover, these beneficial effects were not associated with systemic hypertension as a result of NO synthesis inhibition. The efficacy of iNOS inhibition in ameliorating the extent of glomerular injury and proteinuria in glomerulonephritis is unknown. Selective inhibition of iNOS would still reduce NO generation thereby promoting renal vasoconstriction, platelet aggregation and leukocyte adherence to the glomerular capillary endothelium. This could offset the potential benefit of selective iNOS inhibition. Indeed, in a rat model of rapidly progressive crescentic nephritis, iNOS inhibition using a relatively selective inhibitor worsened proteinuria [65]. Although increasingly selective iNOS inhibitors are being developed, they are not entirely without an inhibitory effect on constitutive NOS and this may be sufficient to cause renal vasoconstriction. The recent development of transgenic mice lacking the iNOS gene [59] may provide a definitive tool to assess the role of iNOS in the biology of glomerulonephritis. Little is known about activation of inducible cyclooxygenase or nitric oxide synthase in human forms of glomerulonephritis. This is particularly important insofaras activation of iNOS is concerned as this form of iNOS may not be induced in human macrophages

545

infiltrating immunologically injured glomeruli. Indeed, the circumstances required for the release of iNOS-derived NO in human macrophages are controversial. That human monocytes/ macrophages contain and can express the iNOS gene is well established. The accumulation of iNOS mRNA and protein has been documented in interferon-gamma (IFN-g) and TNF-stimulated human monocytes, and in monocytes cultured with HIV-1 [60, 61]. However, the actual production of NO by human monocytes/macrophages differs from that of rodent macrophages. While treatment of rodent macrophages with established iNOS inductants such as interferon gamma (IFN-g), tumor necrosis factor (TNF) or lipopolysaccharide (LPS) results in rapid and abundant accumulation of NO, the amount of NO metabolites released by human macrophages is modest when compared with production in rodent cells. Also, induction of iNOS requires a period of stimulation of several days [62]. This has lead to the proposal that the human macrophage iNOS is a relatively lowoutput system. However, notable exceptions to this slow induction/slow output system also exist. Human monocytes produce large amounts of NO22 when cultured with M. avium [62]. Unstimulated cultured human peripheral blood monocytes produce relatively high levels of NO22 (26 nmol/106 cells/day) during the first day of culture and production increases linearly or abruptly after six days in culture [63]. Finally, cross-linking of the CD69 cell surface receptor in human monocytes results in rapid production of NO22 [64]. Collectively, these observations point to limitations of the animal (in particular rodent) models of inflammation in studying activation of the L-arginine:NO system. They also suggest a more sophisticated and tightly regulated system for iNOS-derived NO in human as compared to rodent forms of macrophage-dependent forms of inflammation. CONCLUDING REMARKS In glomerulonephritis the arachidonic acid pathway to cyclooxygenation and lipoxygenation eicosanoids and the L-arginine to NO pathway are coactivated. Regulatory interactions between the two pathways may occur as a result of NO binding to iron containing centers of cyclooxygenase and lipoxygenases. Evidence that NO can up-regulate cyclooxygenase activity is firm while evidence that NO may also regulate activity of lipoxygenases is emerging. In experimental forms of glomerulonephritis, inducible isoforms of cyclooxygenase and nitric oxide synthase are expressed. This may lead to sustained generation of proinflammatory prostaglandins and TxA2, and in high output generation of NO. Whereas evidence that eicosanoids have a direct cytotoxic effect is lacking, the cytotoxic effect of NO is well established particularly when NO reacts with superoxide to form the strong and relatively stable oxidant peroxynitrite. Whether the concentrations of NO or peroxynitrite attained in the glomerular milieu in the course of glomerulonephritis are sufficient to cause cytotoxicity by this mechanism is unknown. Detection of nitrotyrosinecontaining proteins in glomeruli using immunohistochemical methods may provide a means of assessing the extent of peroxynitrite-induced cell injury. Without such demonstration, the argument that iNOS-derived NO causes cell injury in the course of glomerulonephritis remains unfounded. Obtaining such evidence is important in order to justify use of selective iNOS inhibitors to reduce generation of iNOS-derived NO in glomerulonephritis. A serious adverse effect of NO inhibition is renal vasoconstriction with intraglomerular and/or systemic hypertension. Therefore,

546

Lianos: L-arginine:nitric oxide pathways in GN

beneficial effects expected as a result of iNOS inhibition should be carefully weighed against such adverse sequelae. Oxy-hemoglobin in red blood cells provides a very efficient system of removing NO following its release. On the other hand, superoxide dismutase provides an efficient system to remove O22. These two systems prevent reaction of O22 with NO to form the potent oxidant ONOO2. This, and the ability of NO to diffuse through cells and tissues with little consumption or direct reaction, raises skepticism and urges carefully conducted studies to confirm the cytotoxic potential of NO in glomerulonephritis. This is particularly true in humans because human macrophages infiltrating nephritic glomeruli may be unable to up-regulate activity of the inducible NOS. Reprint requests to Elias A. Lianos, M.D., Nephrology Division, Department of Medicine, Medical College of Wisconsin, 9200 W. Wisconsin Avenue, Milwaukee, Wisconsin 53226, USA.

APPENDIX Abbreviations used in this article are: COX, cyclooxygenase; GBM, glomerular basement membrane; GFR, glomerular filtration rate; HETE, hydroxyeicosatetraenoic acid; IL, interleukin; INF-g, interferon-gamma; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide; LT, leukotriene; Kf, ultrafiltration coefficient; NO, nitric oxide; NO21, nitronium ion; NOS, nitric oxide synthase; NSAIDs, non-steroidal anti-inflammatory drugs; O22, superoxide; OHz, hydroxyl radical; ONOO2, peroxynitrite; PG, prostaglandin; PGI2, prostacycline; RBF, renal blood flow; TNF, tumor necrosis factor; Tx, thromboxane.

REFERENCES 1. RAIJ L, BAYLIS C: Glomerular actions of nitric oxide. (Editorial Review) Kidney Int 48:20 –32, 1995 2. BADR KF, LAKKIS FG: Lipoxygenase products in normal and diseased glomeruli, in Annals of New York Academy of Sciences. Cellular Generation, Transport and Effects of Eicosanoids. Biological Roles and Pharmacological Intervention (vol 744), edited by GOETZL EJ, LEWIS RP, ROLA-PLESZCZYNSKI M, New York, New York Academy of Sciences, 1994, pp 276 –228 3. MONCADA S, PALMER RJ, HIGGS EA: Nitric oxide, physiology, pathophysiology and pharmacology. Pharmacol Rev 42:109 –142, 1991 4. ROMERO JC, LAHERA V, SALOM MG, BIONDI ML: Role of endothelium-dependent relaxing factor nitric oxide on renal function. J Am Soc Nephrol 2:1371–1387, 1992 5. TERADA Y, TOMIDA K, NONOGUCHI H, MARUMO F: Polymerase chain reaction localization of constitutive nitric oxide synthase and soluble guanylase cyclase messenger RNA’s in microdissected rat nephron segments. J Clin Invest 90:659 – 665, 1992 6. MOHAUPT MG, ELZIE LJ, NAHN KY, CLAPP WL, WILCOX CS, KONE BC: Differential expression and induction of mRNA’s encoding two inducible nitric oxide synthases in rat kidney. Kidney Int 46:653– 665, 1994 7. ZATZ R, DENUCCI G: Effects of acute nitric oxide inhibition on rat glomerular microcirculation. Am J Physiol 261:F360 –F363, 1991 8. MARSDEN PA, BROCK TA, BALLERMANN BJ: Glomerular endothelial cells respond to calcium-mobilizing agonists with release of EDRF. Am J Physiol 258:F1295–F1303, 1990 9. LIANOS EA: Modulation of renal immune injury by eicosanoids. Am J Therapeutics 2:829 – 836, 1995 10. LIANOS EA, ANDRES G, DUNN MJ: Glomerular prostaglandin and thromboxane synthesis in rat nephrotoxic serum nephritis. Effects on renal hemodynamics. J Clin Invest 74:1439 –1448, 1983 11. LIANOS EA: Synthesis of hydroxyeicosatetraenoic acids and leukotrienes in rat nephrotoxic serum glomerulonephritis: Role of antiglomerular basement membrane antibody dose, complement and neutrophils. J Clin Invest 82:427– 435, 1988 12. CATTELL V, COOK HT: The nitric oxide pathway in glomerulonephritis. Curr Opin Nephrol Hypertens 4:4, 1995

13. LIANOS EA, BRESNAHAN BA, PAN C: Mesangial cell immune injury: Synthesis, origin and role of eicosanoids. J Clin Invest 88:623– 631, 1991 14. CATTELL V, LARGEN P, DE HEER E, COOK TH: Nitrite synthesis in active Heymann nephritis is macrophage dependent, in Biology of Nitric Oxide (vol 2), edited by MONCADA S, MARLETTA MA, HIBBS JB, HIGGS EA, Portland, Portland Press Limited, 1992, pp 257–261 15. CATTELL V, LIANOS EA, LARGEN P, COOK HT: Glomerular NO synthase activity in mesangial cell immune injury. Exp Nephrol 1:36 – 40, 1993 16. TEJANI A, SULLIVAN EK, FINE RN, HARMON W, ALEXANDER S: Steady improvement in renal allograft survival among North American children: A five year appraisal by the North American Pediatric Renal Transplant Cooperative Study. Kidney Int 48:550 –560, 1995 17. JANSEN A, COOK T, TAYLOR GM, RIVEROS-MORENO V, MONCADA S, CATTELL V: Induction of nitric oxide synthase in rat immune complex glomerulonephritis. Kidney Int 45:1215–1219, 1994 18. CAREY T, CRUET E, MUNGER K, SERHAN CN, BADR KF: Cyclooxygenase II (COX-II) expression and lipoxin (LX) biosynthesis in passive (NTS) and accelerated (aNTS) nephrotoxic serum nephritis in the rat: Modulation by aspirin (ASA) therapy. (abstract) J Am Soc Nephrol 6:753, 1995 19. RZYMKIEWICZ D, LEINGANG K, BAIRD N, MORRISON AR: Regulation of prostaglandin endoperoxide synthase gene expression in rat mesangial cells by interleukin-1b. Am J Physiol 266:F39 –F45, 1994 20. STROEBEL M, GOPPELT-STRUEBE M: Signal transduction pathways responsible for serotonin-mediated prostaglandin G/H synthase expression in rat mesangial cells. J Biol Chem 269:22952–22957, 1994 21. KLEIN T, NUSING RM, PFEILSCHIFTER J, ULLRICH V: Selective inhibition of cyclooxygenase 2. Biochem Pharmacol 48:1605–1610, 1994 22. IGNARRO LJ: Signal transduction mechanisms involving nitric oxide. Biochem Pharmacol 41:485– 490, 1991 23. NATHAN C: Nitric oxide as a secretory product of mammalian cells. FASEB J 6:3051–3064, 1992 24. KALYANARAMAN B, MASON RP, TAINER B, ELING TE: The free radical formed during the hydroperoxide-mediated deactivation of ram seminal vesicles is hemoprotein. J Biol Chem 257:4764 – 4768, 1982 25. SALVEMINI D, MISKO TP, MASFERRER JL, SEIBERT K, CURRIE MG, NEEDLEMAN P: Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci USA 90:7240 –7244, 1993 26. EGAN RW, PAXTON J, KUEHL FA JR: Mechanism of irreversible self-deactivation of prostaglandin-derived synthase. J Biol Chem 251: 7329 –7335, 1976 27. HAJJAR DP, LANDER HM, PEARCE SFA, UPMACIS RK, POMERANTZ KB: Nitric oxide enhances prostaglandin-H synthase-1 activity by a heme-independent mechanism: Evidence implicating nitrosothiols. J Am Chem Soc 117:3340 –3346, 1995 28. SALVEMINI D, SETTLE SL, MASFERRER JL, MISKO TP, CURRIE MC, NEEDLEMAN P: Regulation of prostaglandin production by nitric oxide; an in vivo analysis. Br J Pharmacol 114:1171–1178, 1995 29. CORBETT JA, KWON G, TURK J, MCDANIEL ML: IL-1b induces the coexpression of both nitric oxide synthase and cyclooxygenase by islets of Langerhans: Activation of cyclooxygenase by nitric oxide. Biochem J 32:13767–13770, 1993 30. INOUE T, FUKUO K, MORIMOTO S, KOH E, OGIHARA T: Nitric oxide mediates interleukin-1-induced prostaglandin E2 production by vascular smooth muscle cells. Biochem Biophys Res Comm 194:420 – 424, 1993 31. DAVIDGE ST, BAKER PN, MCLAUGHLIN MK, ROBERTS JM: Nitric oxide produced by endothelial cells increases production of eicosanoid through activation of prostaglandin H synthase. Circ Res 77:274 –283, 1995 32. RIEF DW, SIMMONS RD: Nitric oxide mediates iron release from ferritin. Arch Biochem Biophys 283:537–541, 1990 33. RUBBO H, PARTHASARATHY S, BARNES S, KIRK M, KALYANARAMAN B, FREEMAN BA: Nitric oxide inhibition of lipoxygenase-dependent liposome and low-density lipoprotein oxidation: Termination of radical chain propagation reactions and formation of nitrogen-containing oxidized lipid derivatives. Arch Biochem Biophys 324:15–25, 1995 34. MCCALL TB, BOUGHTON-SMITH RK, PALMER RMJ, WHITTEL BJR, MONCADA S: Synthesis of nitric oxide from L-arginine by neutrophils. Release and interaction with superoxide anion. Biochem J 261:293– 296, 1989

Lianos: L-arginine:nitric oxide pathways in GN 35. TETSUKA T, DAPHNA-IKEN D, SRIVASTAVA SK, BEIER LD, DUMAINE J, MORRISON AR: Cross-talk between cyclooxygenase and nitric oxide pathways: Prostaglandin E2 negatively modulates induction of nitric oxide synthase by interleukin-1. Proc Natl Acad Sci USA 91:12168 – 12172, 1994 36. STORK JE, DUNN MJ: Hemodynamic roles of thromboxane A2 and prostaglandin E2 in glomerulonephritis. J Pharmacol Exp Ther 233: 672– 678, 1985 37. FERRARIO R, TAHASHI K, FOGO A, BADR KF, MUNGER KA: Consequences of acute nitric oxide synthase inhibition in experimental glomerulonephritis. J Am Soc Nephrol 4:1847–1854, 1994 38. BADR KF, SCHREINER GF, WASSERMAN M, ICHIKAWA I: Preservation of the glomerular capillary ultrafiltration coefficient during rat nephrotoxic serum nephritis by a specific leukotriene D4 receptor antagonist. J Clin Invest 81:1702–1709, 1988 39. BRESNAHAN BA, WU S-H, FENOY FJ, ROMAN RJ, LIANOS EA: Mesangial cell immune injury: Hemodynamic role of leukocyte- and platelet-derived eicosanoids. J Clin Invest 90:2304 –2312, 1992 40. KRAEMER SA, MEADE EA, DEWITT DL: Prostaglandin endoperoxide synthase gene structure: Identification of the transcriptional start site and 59-flanking regulatory sequences. Arch Biochem Biophys 293:391– 400, 1992 41. LANGENBACH R, MORHAM SG, TIANO HF, LOFTIN CD, GHANAYEM BI, CHULADA PC, MAHLER JF, LEE CA, GOULDING EH, KLUCKMAN KD, KIM HS, SMITHIES O: Prostaglandin synthase 1 gene disruption in mice reduces arachidonic acid-induced inflammation and indomethacin-induced gastric ulceration. Cell 83:483– 492, 1995 42. MORHAM SG, LANGENBACH R, LOFTIN CD, TIANO HF, VOULOUMANOS N, JENNETTE JC, MAHLER JF, KLUCKMAN KD, LEDFORD A, LEE CA, SMITHIES O: Prostaglandin synthase 2 gene disruption causes severe renal pathology in the mouse. Cell 83:473– 482, 1995 43. HARRIS RC, MCKANNA JA, AKAI JA, JACOBSON HR, DUBOIS RN, BRYER MD: Cyclooxygenase-2 is associated with the macula densa of rat kidney and increases with salt restriction. J Clin Invest 94:2504 – 2510, 1994 44. SANO H, HLA T, MAIER JAM, CROFFORD LJ, CASE JP, MACIAG T, WILDER RL: In vivo cyclooxygenase expression in synovial tissues of patients with rheumatoid arthritis and osteoarthritis and rats with adjuvant and streptococcal cell wall arthritis. J Clin Invest 89:97–108, 1992 45. WINK DA, OSAWA Y, DARBYSHIRE F, JONES CR, ESHENAUR SC, NIMS RW: Inhibition of cytochrome P450 by nitric oxide and nitric oxidereleasing agents. Arch Biochem Biophys 300:115–123, 1993 46. LEPOIVRE M, FLAMAN JM, HENRY Y: Early loss of the tyrosyl radical in ribonucleotide reductase of adenocarcinoma cells producing nitric oxide. J Biol Chem 267:22994 –23000, 1992 47. CLEETER MWJ, COOPER JM, DARLEY-USMAR EM, MONCADA S, SCHAPIRA HV: Reversible inhibition of cytochrome C oxidase, the terminal enzyme of the mitochrondrial respiratory chain, by nitric oxide. FEBS Lett 345:50 –54, 1994 48. ISCHIROPOULOS H, ZHU L, CHEN J, TSAI HM, MARTIN JC, SMITH CD, BECKMAN JS: Peroxynitrite-mediated tyrosine nitration is catalyzed by superoxide dismutase. J Arch Biochem Biophys 298:431– 437, 1992 49. BECKMAN JS, YE YZ, ANDERSON P, CHEN J, ACCAVETTI MA, TARPEY MN, WHITE CR: Extensive nitration of protein tyrosines observed in human atherosclerosis detected by immunohistochemistry. Biol Chem Hoppe Seyler 375:81– 88, 1994 50. MORENO JJ, PRYOR WA: Inactivation of a-1-antiprotease inhibitor by peroxynitrite. Chem Res Toxicol 5:425–531, 1992 51. BAUER M, BECKMAN JS, BRIDGES R, MATLON S: Peroxynitrite inhibits sodium transport in rat colonic membrane vessicles. Biochem Biophys Acta 1104:84 – 87, 1992

547

52. RADI R, RODRIGUEZ M, CASTRO L, TELLERI R: Inhibition of mitochondrial electron transport by peroxynitrite. J Arch Biochem Biophys 308:89 –95, 1994 53. GRAHAM A, HOGG N, KALYANARAMAN B, O’LEARY V, DARLEYUSMAR V, MONCADA S: Peroxynitrite modification of low-density lipoprotein leads to recognition by the macrophage scavenger receptor. FEBS Lett 330:181–185, 1993 54. WADDINGTON S, COOK HT, REAVELEY D, JANSEN A, CATTELL V: L-arginine depletion inhibits glomerular nitric oxide synthesis and exacerbates rat nephrotoxic nephritis. Kidney Int 49:1090 –1096, 1996 55. WINEBURG JB, GRAINGER DL, PISETSKI DS, SELDAN MF, MISUKONIS MA, MASON SM, PIPPEN AM, RUIZ P, WOOD ER, GILKESON GS: Role of nitric oxide in the pathogenesis of spontaneous murine autoimmune disease: Increased nitric oxide production and nitric oxide synthase expression in MRL-lpr/mice, and reduction of spontaneous glomerulonephritis and arthritis by orally administered NG-monomethyl-L-arginine. J Exp Med 179:651– 660, 1994 56. LANCASTER JR: Stimulation of the diffusion and reaction of endogenously produced nitric oxide. Proc Natl Acad Sci USA 91:8137– 8141, 1994 57. CONNOR J, MANNING PT, SETTLE SL, MOORE WM, JEROME GM, WEBBER RK, TJOENG FS, CURRIE MG: Suppression of adjuvantinduced arthritis by selective inhibition of inducible nitric oxide synthase. (abstract) Eur J Pharmacol 273:1524, 1995 58. SALVEMINI D, MANNING PT, ZWEIFEL BS, SEIBERT K, CONNOR J, CURRIE MG, NEEDLEMAN P, MASFERRER JL: Dual inhibition of nitric oxide and prostaglandin production contributes to the anti-inflammatory properties of nitric oxide synthase inhibitors. J Clin Invest 96:301–308, 1995 59. MACMICKING JD, NATHAN C, HOM G, CHARTRAIN N, FLETCHER DS, TRUMBAUER M, STEVENS K, XIE Q-W, SOKOL K, HUTCHINSON N, CHEN H, MUDGETT JS: Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase. Cell 81:641– 650, 1995 60. WEINBERG JB, MISUKONIS MA, WOOD ER, SMITH GL, SHAMI PL, MASON SN, GRANGER DL: Human mononuclear phagocyte nitric oxide synthase: Evidence for induction of NOS mRNA and protein without detectable capacity for nitric oxide production. (abstract) Blood 82(Suppl 1):186a(730), 1993 61. BUKRINSKY MI, NOTTET HSLM, SCHMIDTMAYEROVA H, DUBROVSKY L, FLANAGAN CR, MULLINS ME, SIPTON SA, CENDELMAN HE: Regulation of nitric oxide synthase activity in human immunodeficiency virus type 1 (HIV-1)-infected monocytes: Implications for HIV-associated neurological disease. J Exp Med 181:735–745, 1995 62. DENNIS M: Tumor necrosis factor and granulocyte macrophagecolony stimulating factor stimulate human macrophages to restrict growth of virulent Mycobacterium avium and to kill a virulent M. avium: Killing effector mechanism depends on the generation of reactive nitrogen intermediates. J Leuk Biol 49:380 –387, 1991 63. MARTIN JHJ, EDWARDS SW: Changes in mechanisms of monocyte/ macrophage-mediated cytotoxicity during culture: Reactive oxygen intermediates are involved in monocyte-mediated cytotoxicity, whereas reactive nitrogen intermediates are employed by macrophages in tumor cell killing. J Immunol 150:3478 –3486, 1993 64. DEMARIA R, CIFONE MG, TROTTA R, RIPPO MR, FESTUCCIA C, SANTONI A, TESTI R: Triggering of human monocyte activation through CD69, a member of the natural killer cell gene complex family of signal transducing receptors. J Exp Med 180:1999 –2004, 1994 65. LIANOS EA, GUGLIELMI K, SHARMA M: Regulatory interactions between inducible nitric oxide synthase and eicosanoids in glomerular injury. Kidney Int 53:645– 653, 1998.