Differential expression of inducible nitric oxide synthase in septic shock

During cardiopulmonary bypass (CPB), the septic patient has markedly decreased peripheral vascular resistance as a consequence of endotoxin release from microorganisms. This decrease in peripheral vascular resistance is the result of endotoxin-induced nitric oxide (NO) produced by inducible nitric oxide synthase (iNOS) and endothelial nitric oxide synthase (eNOS). iNOS and eNOS are responsible for the synthesis of NO because of various stimuli, including the bacterial endotoxin, lipopolysaccharide (LPS). We tested the hypothesis that a differential expression of iNOS among human endothelial cells and murine macrophage is dependent upon exposure to endotoxin and various pro-inflammatory cytokines. Using a human endothelial cell line, ECV-304 and murine macrophage cell line, RAW 264.7, we quantified the expression of iNOS with specific FITC-conjugated antibodies using fluorescence activated cell sorter (FACS) and NO production with a Bioxytech nitric oxide spectrophotometric assay. This in vitro septic model utilized LPS supported with species-specific interferon-gamma, interleukin-1 beta, and tumor necrosis factor-alpha. The cell type were stimulated for 8 hours with combinations of the cytokines mentioned. The FACS data demonstrated a significant stimulus-dependent increase in iNOS expression among the macrophage groups; however, the stimulated endothelial cells showed no significant change in iNOS expression. The nitric oxide production data demonstrated significant increases in NO production among macrophage stimulated groups; whereas, endothelial stimulated groups exhibit no significant change. We conclude that NO secreted during septic shock is the result of activated macrophage, not the endothelium. The clinical relevance is that the more severe the infectious process, the lower the PVR may be during CPB because of increased NO production from activated macrophage.     

Attenuated expression of inducible nitric oxide synthase in lung microvascular endothelial cells is associated with an increase in ICAM-1 expression

BACKGROUND/PURPOSE: The molecular and cellular events that regulate inflammatory lung injury, a major cause of morbidity in surgical patients, remain unclear. The authors hypothesize that nitric oxide (NO) plays an important role in regulating polymorphonuclear cell (PMN)-induced acute lung injury, and further, that attenuated expression of inducible nitric oxide synthase (iNOS) and therefore decreased production of NO by lung microvascular endothelial cells (LMVEC), accelerates inflammation and injury. METHODS: LMVEC and aortic EC (AEC) from rat and human were stimulated with lipopolysaccharide (LPS) and cytokines; changes in iNOS mRNA expression and iNOS activity were determined. The role of NO in mediating inflammatory responses was evaluated by determining PMN adherence to LMVEC and lung tissue slices in the presence and absence of NOS inhibitors and NO donors. Human LMVEC and AEC were assessed by FACS analysis for ICAM-1 expression, because this is thought to be a critical determinant of PMN adherence. RESULTS: When stimulated with endotoxin and cytokines, rat AEC monolayers express nearly 3-fold more iNOS mRNA than rat LMVEC. The low levels of LMVEC iNOS expression are associated with a 4-fold lower nitrite and nitrate production. Similar trends are seen in human endothelial cells. When iNOS activity was blocked, PMN adherence to tumor necrosis factor alpha (TNFalpha)/LPS-stimulated LMVEC was markedly increased. In contrast, adding a nitric oxide donor to endotoxin/cytokine-stimulated LMVEC monolayers reduced PMN adherence to near background levels. Similar responses were observed in vivo. Human lung microvascular endothelial cells show a substantially increased level of ICAM-1 upregulation when compared with similarly stimulated human aortic macrovascular endothelial cells. CONCLUSIONS: These data indicate that LMVEC express less iNOS and produce less NO than AEC. This lower expression and activity of iNOS in LMVEC may be linked to increased expression of ICAM-1. Because ICAM-1 has been shown to be essential for tight PMN adherence, these data suggest that relatively low iNOS expression in LMVEC may contribute to a propensity for the lung to be injured by activated PMNs.     

Bactericidal/permeability-increasing protein inhibits endotoxin-induced vascular nitric oxide synthesis

BACKGROUND: Endotoxin (lipopolysaccharide, LPS) up-regulates inducible nitric oxide synthase (iNOS) in blood vessels during septic shock. This promotes the production of nitric oxide (NO), leading to dilation of the vessels. The aim of the study was to investigate the effects of the LPS-binding endogenous antibiotic bactericidal/permeability-increasing protein (BPI) on the action of LPS on the blood vessels wall and to identify possible influence on underlying NO-related mechanisms. METHODS: Isolated segments of rat thoracic aorta and cultured primary smooth muscle cells were incubated for 5-48 h in the presence of the following combinations of compounds: (a) LPS; (b) interleukin-1beta (IL-1beta); (c) BPI; (d) BPI + LPS; (e) BPI + IL-1beta or (f) neither BPI, LPS nor IL-1beta (control). After incubation of intact segments, we measured smooth muscle contraction in response to phenylephrine and accumulation of the NO end products nitrate and nitrite in surrounding medium. Western blot was used to assess the levels of inducible nitric oxide synthase (iNOS) in cultured cells. RESULTS: Both LPS and IL-1beta decreased contractility and increased NO production, as well as iNOS. Co-incubation with BPI attenuated all the effects of LPS but only the effects of prolonged exposure to IL-1beta in cultured cells. CONCLUSION: We conclude that BPI attenuates the LPS-induced changes in vascular reactivity by inhibiting the expression of iNOS resulting in decreased NO formation and restored responsiveness to vasoconstrictors. The data suggest that BPI can prevent circulatory disturbances during Gram-negative sepsis.     

Effects of human cathelicidin antimicrobial peptide LL-37 on lipopolysaccharide-induced nitric oxide release from rat aorta in vitro

BACKGROUND: Lipopolysaccharides (LPS), released by Gram-negative bacteria, cause vascular expression of inducible nitric oxide synthase (iNOS) leading to nitric oxide (NO) production and septic shock. Human cathelicidin antimicrobial peptide (LL-37) can bind and neutralize LPS. We wanted to study whether LL-37 affects LPS or interleukin-1beta (IL-1beta)-induced production, release and function of NO in intact rat aorta rings and cultured rat aorta smooth muscle cells. METHODS: Isolated segments of thoracic aorta and cultured cells were incubated in the presence of LPS, LL-37, LPS + IL-37, IL-1beta, IL-1beta + IL-37 or in medium alone. Smooth muscle contraction in response to phenylephrine and accumulation of the sdegradation products of NO, nitrate and nitrite, were measured on aorta segments. Levels of iNOS were assessed by Western blot and cytotoxic effects were detected by measurement of DNA fragmentation in cultured cells. Number of viable cells were determined after Trypan blue treatment. RESULTS: Both LPS and IL-1beta reduced contractility in response to phenylephrine and increased NO production as well as iNOS expression. LL-37 inhibited the LPS depression of vascular contractility induced only by LPS. LL-37 reduced both the LPS- and IL-1beta-induced NO production and iNOS expression. LL-37 at high concentrations induced DNA fragmentation and decreased the number of living cells. CONCLUSION: IL-37 reduces NO production induced by LPS and IL-1beta. The reduction does not seem to result only from neutralization of LPS but also from a cytotoxic effect, possibly via induction of apoptosis.     

Platonin attenuates LPS-induced CAT-2 and CAT-2B induction in stimulated murine macrophages

BACKGROUND: Platonin, a cyanine photosensitizing dye, is a potent immunomodulator that suppresses acute inflammation. Platonin not only inhibits interleukin (IL)-1beta, IL-6, and tumor necrosis factor (TNF)-alpha production but also improves circulatory failure in septic rats. In addition, platonin reduces plasma nitric oxide (NO) formation during sepsis. However, the effects of platonin on inducible NO synthase (iNOS) and cationic amino-acid transporter (including CAT-2, CAT-2 A, and CAT-2B) expressions during sepsis remain uninvestigated. METHODS: Five groups of confluent murine macrophages (RAW264.7 cells) were randomly allocated to receive a 1-h pretreatment of one of five doses of platonin (0.1 microM, 1 microM, 10 microM, 100 microM, or 1000 microM) followed by lipopolysaccharide (LPS; 100 ng ml(-1)). For negative, positive, and platonin control, three other groups of cell cultures were randomly allocated to receive phosphate-buffered saline, LPS, or platonin (1000 microM). The cultures were harvested after exposing them to LPS for 18 h or a comparable duration in those groups without LPS. NO production, L-arginine transport, and expression of the relevant enzymes were then evaluated. RESULTS: Platonin significantly attenuated LPS-induced up-regulation of iNOS expression and NO production in stimulated murine macrophages in a dose-dependent manner. Platonin also significantly inhibited up-regulation of CAT-2 and CAT-2B expression as well as L-arginine transport in LPS-stimulated murine macrophages in a dose-dependent manner. In contrast, CAT-2 A expression in murine macrophages was not affected by LPS and/or platonin. CONCLUSIONS: Platonin attenuates NO production and L-arginine transport in LPS-stimulated murine macrophages possibly through inhibiting iNOS, CAT-2, and CAT-2B expression.     

Nitric oxide, sepsis, and the kidney

Although excess nitric oxide (NO) production plays a major role in the hypotension characteristic of sepsis, concurrent constitutive NO generation in the kidney during sepsis is essential for preservation of renal perfusion and prevention of glomerular thrombosis. The authors have shown that although all nitric oxide synthase (NOS) inhibitors restore normal blood pressure in lipopolysaccharide (LPS) treated rats, only selective inducible NOS (iNOS) inhibition prevents the reductions in glomerular filtration rate (GFR), whereas nonselective inhibition of NOS further decreases GFR. Glomerular endothelial NOS (eNOS) activity was found to be inhibited by LPS. The decrease in eNOS activity was completely prevented by selective iNOS inhibition in vivo and in vitro. The adverse renal outcomes after LPS administration correlated with decreased glomerular eNOS activity rather than elevated NO production. These findings suggest that the decrease in GFR after LPS is caused by local inhibition of eNOS by iNOS possibly via NO autoinhibition. Selective inhibition of iNOS could represent a substantially superior approach for the treatment of the sepsis syndrome.     

Endotoxemic renal failure in mice: Role of tumor necrosis factor independent of inducible nitric oxide synthase

BACKGROUND: Renal failure is a frequent complication of sepsis with a high mortality. Tumor necrosis factor (TNF) has been suggested to be a factor in the acute renal failure in sepsis or endotoxemia. Recent studies also suggest involvement of nitric oxide (NO), generated by inducible NO synthase (iNOS), in the pathogenesis of endotoxin-induced renal failure. The present study tested the hypothesis that the role of TNF in endotoxic renal failure is mediated by iNOS-derived NO. METHODS: Renal function was evaluated in endotoxemic [Escherichia coli lipopolysaccharide (LPS), 5 mg/kg IP] wild-type and iNOS knockout mice. The effect of TNF neutralization on renal function during endotoxemia in mice was assessed by a TNF-soluble receptor (TNFsRp55). RESULTS: An injection of LPS to wild-type mice resulted in a 70% decrease in glomerular filtration rate (GFR) and in a 40% reduction in renal plasma flow (RPF) 16 hours after the injection. The results occurred independent of hypotension, morphological changes, apoptosis, and leukocyte accumulation. In mice pretreated with TNFsRp55, only a 30% decrease in GFR without a significant change in RPF in response to LPS, as compared with vehicle-treated mice, was observed. Also, the serum NO concentration was significantly lower in endotoxemic wild-type mice pretreated with TNFsRp55, as compared with untreated endotoxemic wild-type mice (260 +/- 52 vs. 673 +/- 112 micromol/L, P < 0.01). In LPS-injected iNOS knockout mice and wild-type mice treated with a selective iNOS inhibitor, 1400W, the development of renal failure was similar to that in wild-type mice. As in wild-type mice, TNFsRp55 significantly attenuated the decrease in GFR (a 33% decline, as compared with 75% without TNFsRp55) without a significant change in RPF in iNOS knockout mice given LPS. CONCLUSIONS: These results demonstrate a role of TNF in the early renal dysfunction (16 h) in a septic mouse model independent of iNOS, hypotension, apoptosis, leukocyte accumulation, and morphological alterations, thus suggesting renal hypoperfusion secondary to an imbalance between, as yet to be defined, renal vasoconstrictors and vasodilators.     

内源性一氧化氮在人前交叉韧带(HACL)细胞增殖的表达水平

目的检测人前交叉韧带(HACL)细胞生长增殖过程中内源性一氧化氮(NO)表达水平。方法分离和培养HACL细胞,用脂多糖(LPS)作诱导培养HACL细胞表达iNOS,用L-硝基精氨酸(LN-MMA)作为内源性NO的阻断剂,间接测定加入单用和联用LPS和LN-MMA后的内源性NO的变化。结果LPS可促进HACL产生内源性NO,与对照组比较差异有显著性意义(P<0.05),LN-MMA可抑制HACL产生内源性NO,与对照组比较差异有显著性意义(P<0.05),两者联用与对照组比较差异有显著性意义(P<0.05)。结论在LPS的诱导下HACL细胞的生长增殖过程中有内源性NO参与,LN-MMA可抑制HACL细胞产生内源性NO,NO使HACL的增殖减慢。     

Selective iNOS inhibition prevents hypotension in septic rats while preserving endothelium-dependent vasodilation

Nitric oxide (NO) derived from inducible nitric oxide synthase (iNOS) mediates hypotension and metabolic derangements in sepsis. We hypothesized that selective iNOS-inhibition would prevent hypotension in septic rats without inhibiting endothelium-dependent vasodilation caused by the physiologically important endothelial NOS. Rats were exposed to lipopolysaccharide (LPS) for 6 h and the selective iNOS-inhibitor L-N6-(1-iminoethyl)-lysine (L-NIL), the nonselective NOS-inhibitor N:(G)-nitro-L-arginine methyl ester (L-NAME), or control. Mean arterial pressure (MAP) and vasodilation to acetylcholine (ACh, endothelium-dependent), sodium nitroprusside (SNP, endothelium-independent), and isoproterenol (ISO, endothelium-independent beta agonist) were determined. Exhaled NO, nitrate/nitrite-(NOx) levels, metabolic data, and immunohistochemical staining for nitrotyrosine, a tracer of peroxynitrite-formation were also determined. In control rats, L-NAME increased MAP, decreased the response to ACh, and increased the response to SNP, whereas L-NIL did not alter these variables. LPS decreased MAP by 18% +/- 1%, decreased vasodilation (ACh, SNP, and ISO), increased exhaled NO, NOx, nitrotyrosine staining, and caused acidosis and hypoglycemia. L-NIL restored MAP and vasodilation (ACh, SNP, and ISO) to baseline and prevented the changes in exhaled NO, NOx, pH, and glucose levels. In contrast, L-NAME restored MAP and SNP vasodilation, but did not alter the decreased response to ACh and ISO or prevent the changes in exhaled NO and glucose levels. Finally, L-NIL but not L-NAME decreased nitrotyrosine staining in LPS rats. In conclusion, L-NIL prevents hypotension and metabolic derangements in septic rats without affecting endothelium-dependent vasodilation whereas L-NAME does not. IMPLICATIONS: Sepsis causes hypotension and metabolic derangements partly because of increased nitric oxide. Selective inhibition of nitric oxide produced by the inducible nitric oxide synthase enzyme prevents hypotension and attenuates metabolic derangements while preserving the important vascular function associated with endothelium-dependent vasodilation in septic rats.     

NF-kappaB inhibitors stabilize the mRNA of high-affinity type-2 cationic amino acid transporter in LPS-stimulated rat liver

BACKGROUND: Induction of inducible nitric oxide synthase (iNOS) results in nitric oxide (NO) overproduction during endotoxemia. Cellular uptake of L-arginine, modulated by the isozymes of type-2 cationic amino acid transporters (CAT), including CAT-2, CAT-2A and CAT-2B, has been reported to be a crucial factor in the regulation of iNOS activity. We sought to elucidate the expression of CAT-2 isozymes and the role of nuclear factor-kappaB (NF-kappaB) in this expression in lipopolysaccharide (LPS)-treated rat liver. METHODS: Adult male Sprague-Dawley rats were randomly given intravenous (i.v.) injections of normal saline (N/S), LPS, LPS preceded by an NF-kappaB inhibitor (PDTC, dexamethasone or salicylate) or an NF-kappaB inhibitor alone. After injection, rats were sacrificed at different times and enzyme expression and liver injury were examined. Hepatic and systemic NO production were also measured. RESULTS: CAT-2, CAT-2A and CAT-2B were constitutively expressed in un-stimulated rat liver. LPS stimulation not only significantly increased iNOS mRNA and NO concentrations but also decreased the mRNA concentrations of CAT-2 and CAT-2B, but not CAT-2A, in a time-dependent manner. LPS-induced hepatic and systemic NO overproduction was associated with hepatocellular injury. Pre-treatment with NF-kappaB inhibitors significantly attenuated LPS-induced iNOS induction as well as CAT-2/CAT-2B mRNA destabilization, which was associated with significant inhibition of NO biosynthesis and less liver injury. CONCLUSION: NF-kappaB inhibitors stabilize CAT-2 and CAT-2B mRNA in LPS-stimulated rat liver. The hepatic CAT-2/CAT-2B pathway may be a constitutive part of cytoprotective mechanisms against sepsis.     

Inhibition of tumor necrosis factor-alpha and inducible nitric oxide synthase correlates with the induction of IL-10 in septic rats undergoing laparotomy and laparoscopy

Proinflammatory mediators are implicated in the mediation of host response to surgical stress. Greater inflammatory response has been reported after open surgery than after laparoscopic surgery in animal models. This study investigated the inflammatory response of tumor necrosis factor alpha (TNF) and inducible nitric oxide synthase (iNOS) and the anti-inflammatory response of interleukin (IL)-10 after laparotomy and laparoscopy in a rat endotoxic shock model. Rats received lipopolysaccharide (LPS) intraperitoneally and underwent laparotomy (n = 5), laparoscopy (n = 5), or no surgical intervention (n = 5). A control group received anesthesia only (n = 5). Serum TNF levels peaked at 2 hours after LPS injection and were significantly suppressed in animals undergoing laparotomy and laparoscopy ( < 0.05). Serum IL-10 levels were higher at 2 hours in the laparotomy and laparoscopy groups but were higher only in the laparotomy group at 4 hours after LPS injection ( < 0.05). Hepatic iNOS mRNA and protein were significantly inhibited at 4 and 8 hours in the laparotomy and laparoscopy groups in comparison with the animals receiving LPS only ( < 0.05). The induction of IL-10 correlated with the suppression of TNF and iNOS suggests that IL-10 may play a role in downregulating TNF and iNOS in septic rats undergoing laparotomy and laparoscopy.     

Catecholamines decrease nitric oxide production by cytokine-stimulated hepatocytes

BACKGROUND: Catecholamines are significantly elevated in inflammatory responses and play a regulatory role in sepsis. Nitric oxide (NO), also a key inflammatory mediator in sepsis, is produced in large amounts by the inducible nitric oxide synthase (iNOS) in the liver. The purpose of this study was to test the hypothesis that catecholamines play a role in the regulation of NO production by hepatocytes. METHODS: Primary hepatocytes were isolated from healthy male Sprague-Dawley rats and either cultured with normal medium or stimulated with cytomix (interleukin-1 beta, interferon-gamma, and tumor necrosis factor-alpha) in the presence or absence of epinephrine or norepinephrine at varying concentrations. Total RNA was isolated 6 hours after treatment and analyzed by Northern blotting for iNOS mRNA. Protein extracts were obtained at 12 hours and were analyzed by Western immunoblotting for iNOS. Cell culture supernatants were analyzed for NO, determined as the stable end-product NO(2)(-), at 24 hours. RESULTS: Epinephrine and norepinephrine significantly decreased NO(2)(-) levels in stimulated hepatocytes but had no effect on iNOS mRNA or protein levels. The decrease in NO(2)(-) was reproduced by the adenylate cyclase stimulator, forskolin. The catecholamine-induced decrease in NO(2)(-) was completely reversed by the protein kinase A inhibitor Rp-8-Br-cyclic adenosine monophosphate. CONCLUSIONS: Catecholamines decrease hepatocyte production of NO in response to cytokine stimulation. This effect seems to be due to post-translational events and appears to be mediated in part by cyclic adenosine monophosphate.     

High glucose increases inducible NO production in cultured rat mesangial cells. Possible role in fibronectin production.

BACKGROUND/AIM: Increased nitric oxide (NO) generation and action have been suggested to be associated with glomerular hyperfiltration and increased vascular permeability early in diabetes. However, previous studies have primarily focused on the constitutive nitric oxide synthase (cNOS) pathway present in endothelial cells, and the role of the inducible NOS (iNOS) pathway in diabetic nephropathy has remained unclear. This study examined whether high glucose modulates NO synthesis by the iNOS pathway in rat mesangial cells. In addition, the effect of inhibition of the iNOS pathway on fibronectin production was determined to examine the role of the iNOS pathway in high glucose-induced extracellular expansion by mesangial cells. METHODS: NO synthesis by the iNOS pathway was evaluated by nitrite and iNOS mRNA and protein productions. The effects of protein kinase C (PKC) inhibitor and aldose reductase inhibitor on the iNOS mRNA expression and aminoguanidine, a relatively specific inhibitor of the iNOS on fibronectin protein production were examined. RESULTS: High 30 mM glucose concentration led to significant increases in nitrite production of rat mesangial cells upon stimulation with lipopolysaccharide (LPS) plus interferon-gamma (IFN-gamma) compared with control 5.6 mM glucose concentration. Mesangial iNOS mRNA expression and protein production also increased significantly in response to high glucose. The addition of calphostin C, a PKC inhibitor, and 6-bromo-1,3-dioxo-1H-benz[d,e]isoquinoline-2(3H)-acetic acid, an aldose reductase inhibitor, significantly suppressed the enhancement of iNOS mRNA expression in high glucose concentration. High glucose also significantly increased fibronectin protein production of mesangial cells upon stimulation with LPS plus IFN-gamma compared to control glucose. Aminoguanidine reversed this high glucose-induced fibronectin production at dose inhibiting iNOS mRNA expression. CONCLUSIONS: These results indicate that high glucose enhances cytokine-induced NO production by rat mesangial cells, and that the activation of PKC and aldose reductase pathway may play a role in this enhancement. In addition, high glucose-induced NO production by the iNOS pathway may promote extracellular matrix accumulation by mesangial cells under certain condition.     

Nitric oxide synthase is downregulated, while haem oxygenase is increased, in patients with septic shock

Background. The vasodilatation characteristic of human septicshock is conventionally attributed to increased nitric oxideproduction, primarily by extrapolation of animal and human invitro studies. There are no conclusive studies of human disease,and the cellular source of nitric oxide in human sepsis is notknown. Haem oxygenase is upregulated by oxidative stress, butlittle is known about haem oxygenase expression in human sepsis.Haem oxygenase may modulate nitric oxide production, and mayalso have a direct effect on vascular tone. Methods. Mesenteric arterial smooth muscle (ASM) (obtained duringlaparotomy) and peripheral blood mononuclear cells (PBMCs) wereobtained from patients with early septic shock and from controlpatients. mRNA levels were determined by real-time RT-PCR. Results. mRNA for inducible and endothelial nitric oxide synthasewas reduced in both ASM and PBMCs from septic patients. In contrast,inducible haem oxygenase mRNA was increased in sepsis in bothcell types. Conclusions. These results suggest that, rather than being induced,the enzymes which produce nitric oxide are reduced at this timepoint in human septic shock. Thus many of the in vitro modelsof sepsis studied to date may not fully replicate human disease.The increase in haem oxygenase expression confirms that thesecells have been subjected to oxidative stress in sepsis. Theactivity of induced haem oxygenase may limit nitric oxide production,while possibly causing vasodilation through production of carbonmonoxide.     

Glutathione regulates nitric oxide synthase in cultured hepatocytes.

OBJECTIVE: The authors determine the relationship between glutathione and nitric oxide (NO) synthesis in cultured hepatocytes. SUMMARY BACKGROUND DATA: Glutathione is a cofactor for a number of enzymes, and its presence is essential for maximal enzyme activity by the inducible macrophage nitric oxide synthase (iNOS), which produces the reactive nitric oxide radical. Hepatocytes contain substantial quantities of glutathione, and this important tripeptide is decreased in hepatocytes stressed by ischemia/reperfusion or endotoxemia. Endotoxemia also induces the synthesis of inflammatory cytokines that result in the production of nitric oxide from hepatocytes by iNOS, suggesting that hepatocytes may be attempting to synthesize nitric oxide at times when intracellular glutathione is reduced. METHODS: Hepatocytes were cultured with buthionine sulfoximine and 1,3-bis(chloroethyl)-1-nitrosourea (BCNU) to inhibit glutathione. After exposure to cytokines, NO synthesis was assessed by supernatant nitrite levels, cytosolic iNOS enzyme activity, and iNOS mRNA levels. RESULTS: Inhibition of glutathione synthesis with buthionine sulfoximine or inhibition of glutathione reductase activity with BCNU inhibited nitrite synthesis. Both buthionine sulfoximine and BCNU inhibited the induction of iNOS mRNA, as detected by Northern blot analysis. Exogenous glutathione increased cytokine-stimulated iNOS induction, overcame the inhibitory effects of BCNU, and increased nitrite production by intact hepatocytes, induced hepatocyte cytosol, and partially purified hepatocyte iNOS. CONCLUSIONS: In cultured hepatocytes, adequate glutathione levels are required for optimal nitric oxide synthesis. This finding is predominantly due to an effect on iNOS mRNA levels, although glutathione also participates in the regulation of iNOS enzyme activity.