Pivotal Role of gp91phox-Containing NADH Oxidase in Lipopolysaccharide-Induced Tumor Necrosis Factor-α Expression and Myocardial Depression
Background— Lipopolysaccharide (LPS) induces cardiomyocyte tumor necrosis factor-α (TNF-α) production, which is responsible for myocardial depression during sepsis. The aim of this study was to investigate the role of gp91phox-containing NADH oxidase signaling in cardiomyocyte TNF-α expression and myocardial dysfunction induced by LPS.
Methods and Results— In cultured mouse neonatal cardiomyocytes, LPS increased NADH oxidase (gp91phox subunit) expression and superoxide generation. Deficiency of gp91phox or inhibition of NADH oxidase blocked TNF-α expression stimulated by LPS. TNF-α induction was also inhibited by tempol, N-acetylcysteine, or 1,3-dimethyl-2-thiourea. NADH oxidase activation by LPS increased ERK1/2 and p38 phosphorylation, and inhibition of ERK1/2 and p38 phosphorylation blocked the effect of NADH oxidase on TNF-α expression. Isolated mouse hearts were perfused with LPS (5 μg/mL) alone or in the presence of apocynin for 1 hour. Myocardial TNF-α production was decreased in gp91phox-deficient or apocynin-treated hearts compared with those of wild type (P<0.05). To investigate the role of gp91phox-containing NADH oxidase in endotoxemia, mice were treated with LPS (4 mg/kg IP) for 4 and 24 hours, and their heart function was measured with a Langendorff system. Deficiency of gp91phox significantly attenuated LPS-induced myocardial depression (P<0.05).
Conclusions— gp91phox-Containing NADH oxidase is pivotal in LPS-induced TNF-α expression and cardiac depression. Effects of NADH oxidase activation are mediated by ERK1/2 and p38 MAPK pathway. The present results suggest that gp91phox-containing NADH oxidase may represent a potential therapeutic target for myocardial dysfunction in sepsis.
Received July 27, 2004; revision received November 17, 2004; accepted November 19, 2004.
Lipopolysaccharide (LPS) of Gram-negative bacteria induces expression of tumor necrosis factor-α (TNF-α), a proinflammatory cytokine, in cardiomyocytes.1–4 Previous studies have demonstrated that cardiomyocytes are the major local source of TNF-α in the myocardium during sepsis.4 TNF-α is responsible for myocardial depression induced by endotoxemia.4,5 Thus, modulation of local myocardial TNF-α levels produced by cardiomyocytes may be of therapeutic significance in sepsis-induced myocardial dysfunction.4 Mechanisms by which LPS induces TNF-α expression in cardiomyocytes are not fully understood. It is well established that the effects of LPS are mediated via toll-like receptors.6 We have recently demonstrated that activation of p38 mitogen-activated protein kinase (MAPK) is required for LPS-induced TNF-α expression in cardiomyocytes3,4; however, LPS signaling downstream of toll-like receptors and upstream of MAPK remains largely unknown in cardiomyocytes.
NADPH or NADH oxidase is an inducible electron transport system found in cells that transfer reducing equivalents from NADPH or NADH to oxygen, which results in superoxide anion (O2−) generation.7 The enzyme complex comprises 2 membrane subunits (gp91phox and p22phox, which form flavocytochrome b558) and at least 4 cytosolic proteins (p40phox, p47phox, p67phox, and Rac1/2, which form the cytosolic complex). NAD(P)H oxidase is a highly regulated enzyme. In the resting cells, the cytosolic complex is separated from the membrane-bound catalytic core. On stimulation, the cytosolic component p47phox becomes phosphorylated, and the cytosolic complex migrates to the membrane, where it binds to cytochrome b558 to assemble into an active oxidase.7 NAD(P)H oxidase expression has also been demonstrated in cardiomyocytes8,9 and has been considered as a critical determinant of the redox state of the myocardium.8–11 In response to LPS, NAD(P)H oxidase activity and O2− production are markedly increased in the heart12,13; however, the contribution of the NADH oxidase to TNF-α expression in LPS-stimulated cardiomyocytes remains unknown, and the role of NADH oxidase in septic myocardial depression has not been investigated. In addition, the membrane subunit of NADH oxidase, gp91phox (Nox2), has at least 3 other homologs, Nox1, Nox3, and Nox4.14 Both gp91phox and Nox4 are expressed in cardiomyocytes.15 A recent study suggested a differential response of the cardiac Nox isoforms, gp91phox and Nox4, to different pathological stimuli for cardiac hypertrophy15; however, the role of gp91phox in sepsis-induced cytokine induction and myocardial dysfunction has not been investigated.
In the present study, we aimed to investigate the role of gp91phox-containing NADH oxidase signaling in LPS-induced TNF-α expression and myocardial dysfunction induced by endotoxemia. To achieve this goal, experiments were performed on gp91phox-deficient (gp91phox−/−) mice and wild-type mice with cardiomyocyte culture, isolated-heart, and whole-animal approaches. Our results showed that gp91phox-containing NADH oxidase plays an important role in cardiomyocyte TNF-α expression and cardiac dysfunction during LPS stimulation.
Animals and Preparation of Neonatal Mouse Cardiomyocytes
The breeding pairs of C57BL/6 wild-type and gp91phox−/− mice (C57BL/6 background) were purchased from Jackson Laboratory. A breeding program was performed to produce neonates. All animals were used in accordance with the guidelines of the Animal Care Committee at the University of Western Ontario, Canada. The neonatal cardiomyocytes were prepared and cultured according to methods we described previously.16
Small interfering RNAs (siRNAs) for ERK1 and p38α MAPK were purchased from Santa Cruz Biotechnology, and a universal control siRNA was obtained from Qiagen. The transfection of siRNA was performed in cardiomyocytes with TransMessenger transfection reagent (Qiagen) according to the manufacturer’s instructions. One microliter of siRNA (10 μmol/L) was used for each well of cells (48-well plate). After transfection, cells were maintained in normal culture medium for another 48 hours before LPS treatment.
LPS (Salmonella typhosa), diphenyleneiodonium (DPI), apocynin, N-acetylcysteine (NAC), tempol, 1,3-dimethyl-2-thiourea (DMTU), β-NADH, lucigenin, PD98059, and SB203580 were purchased from Sigma.
Measurement of TNF-α Protein
Concentrations of TNF-α protein were determined with a mouse TNF-α ELISA kit (ALPCO Diagnostics) as in our previous reports.3,4
Measurement of Superoxide Generation (O2−)
NADH-dependent O2− generation was measured in cell lysates by lucigenin-enhanced chemiluminescence (20 μg of protein, 100 μmol/L β-NADH, 5 μmol/L lucigenin) with a multilabel counter (Victor3 Wallac).8 Some experiments were performed in the presence of superoxide dismutase (SOD; Sigma). The light signal was monitored for 5 seconds, and counts per second (CPS) were presented as NADH oxidase activity that was SOD inhibitable.
Analysis of TNF-α, gp91phox, and MAPK mRNA by Reverse-Transcriptase–Polymerase Chain Reaction
Total RNA was extracted from cardiomyocytes with the TriZol reagent (Gibco-BRL) according to the manufacturer’s instructions. Semiquantitative reverse-transcriptase–polymerase chain reaction (RT-PCR) for TNF-α was performed as described previously.3 The DNA oligonucleotide primers for gp91phox, ERK1, and p38α MAPK were as follows: gp91phox (Accession No. U43384), 5′ ACGCCCTTTGCCTCCATTCT 3′ (sense) and 5′ GCTTCAGGGCCACACAGG AA 3′ (antisense); ERK1 (Accession No. BC029712), 5′-TCC AAG GGC TAC ACC AAA TC-3′ (sense) and 5′-GCT CCA TGT CGA AGG TGA AT-3′ (antisense); and p38-α (Accession No. NM_011951), 5′-TAC CAC GAC CCT GAT GAT GA-3′ (sense) and 5′-GCC AAG GAC CAT TCA CAA CT-3′ (antisense).
Western Blot Analysis
A total of 30 μg of protein in each sample was subjected to SDS-PAGE with 10% gels, followed by electrotransfer to nitrocellulose membranes. Expression of gp91phox protein was determined by probing the blots with specific antibodies against gp91phox (Signal Transduction, 1/2000), followed by enhanced chemiluminescence detection. Gp91phox protein was detected as a 67-kDa band.
Measurement of ERK1/2 and p38 MAPK Phosphorylation
Assessment of the phosphorylation status of ERK1/2 and p38 MAPK in cardiomyocytes was accomplished by Western blotting (see above) with antibodies against ERK1/2/ phospho-ERK1/2 and p38/phospho-p38 MAPK (New England BioLabs, 1/1000), respectively, as described previously.3
Isolated Mouse Heart Preparations
Adult mouse (male aged 2.5 months) hearts were isolated and perfused in a Langendorff-system with Krebs-Henseleit buffer at 2 mL/min constant flow. The perfusion buffer was maintained at 37°C and bubbled continuously with a mixture of 95% O2 and 5% CO2. Myocardial function was assessed by a previously described method with modifications.17,18 Briefly, a 6-0 silk suture was placed through the apex of the left ventricle and threaded through a light-weight rigid coupling rod, which was connected to a force-displacement transducer (FT03) to record tension and heart rate. The heart work was calculated by multiplying the force (g) by the heart rate (bpm). Maximal and minimal first derivatives of force (+dF/dtmax and −dF/dtmin) as the rate of contraction and relaxation were analyzed by PowerLab Chart program (AD Instruments).
All data are given as mean±SD. Differences between 2 groups were compared by unpaired Student t test. For multigroup comparisons, ANOVA followed by Student-Newman-Keuls test was performed. A value of P<0.05 was considered statistically significant.
Effects of LPS on NADH Oxidase Expression and NADH Oxidase Activity in Cardiomyocytes
To examine the role of gp91phox-containing NADH oxidase activity in LPS-induced oxidative stress, we measured gp91phox expression and NADH oxidase activity in cardiomyocytes. As shown in Figure 1, LPS treatment increased gp91phox mRNA and protein expression. The upregulation of NADH oxidase expression induced by LPS was associated with an increase in O2− generation, which was SOD-inhibitable. These data showed that LPS increases NADH oxidase activity in cardiomyocytes.
LPS-Induced TNF-α Expression in gp91phox−/− Cardiomyocytes
To investigate the role of gp91phox-containing NADH oxidase in TNF-α expression in LPS-stimulated cardiomyocytes, cardiomyocytes from gp91phox−/− mice were used. In response to LPS, TNF-α protein and mRNA expression in gp91phox−/− cardiomyocytes were decreased by 49% and 39%, respectively, compared with wild-type cardiomyocytes (P<0.05; Figures 2A and 2B). This was associated with a significant decrease in O2− generation in gp91phox−/− cardiomyocytes (P<0.05; Figure 2C).
Effects of NADH Oxidase Inhibitor on LPS-Induced TNF-α Expression in Cardiomyocytes
The contribution of NADH oxidase to TNF-α expression in response to LPS was also examined by using its pharmacological inhibitors, DPI and apocynin. The inhibitory effects of DPI and apocynin on NADH oxidase were confirmed by measuring NADH-dependent O2− generation (data not shown). As shown in Figure 3, either DPI or apocynin abolished LPS-induced TNF-α protein and mRNA expression. These data suggest that LPS induces TNF-α expression via NADH oxidase activation in cardiomyocytes.
We further determined the contribution of O2− in LPS-induced TNF-α expression. A scavenger of O2−, the membrane-permeable SOD mimetic tempol, was used in LPS-stimulated cardiomyocytes. Tempol decreased LPS-induced TNF-α production in a dose-dependent manner (Figure 4A). To confirm this finding, another antioxidant, NAC, was used. Similarly, NAC dose-dependently suppressed TNF-α expression in LPS-stimulated cardiomyocytes (Figure 4B). Superoxide can form hydroxyl radicals via the Haber-Weiss reaction.19 To examine whether hydroxyl radicals played a role in LPS-induced TNF-α expression, a hydroxyl radical scavenger, DMTU, was used. As shown in Figure 4C, incubation with DMTU dose-dependently blocked LPS-induced TNF-α expression. Thus, NADH oxidase-produced O2−, hydroxyl radicals, and oxidative stress contributed to TNF-α expression in LPS-stimulated cardiomyocytes.
Role of NADH Oxidase in LPS-Induced ERK1/2 and p38 MAPK in Cardiomyocytes
We have shown that p38 MAPK activation is essential for LPS-induced TNF-α expression in cardiomyocytes.3,4 This was also demonstrated in the present study in which treatment with the p38 inhibitor SB203580 completely blocked LPS-induced TNF-α production (Figure 5A). To further investigate the role of ERK1/2 activation in LPS-induced TNF-α expression, cardiomyocytes were treated with PD98059, which blocks ERK1/2 signaling. PD98059 abrogated TNF-α expression in LPS-stimulated cardiomyocytes (Figure 5A). The effect of p38 and ERK1/2 MAPK on TNF-α expression was further confirmed by specific downregulation of MAPK with siRNAs against p38α and ERK1. siRNA treatment selectively suppressed p38α and ERK1 expression, as demonstrated by semiquantitative RT-PCR (data not shown), and decreased LPS-induced TNF-α expression by 24% and 50.8%, respectively (Figure 5B). The data showed that both p38 and ERK activation are required for TNF-α expression. To determine whether NADH oxidase activation leads to p38 and ERK1/2 activation in response to LPS, phosphorylation of p38 and ERK1/2 MAPK was measured after inhibition of NADH oxidase by DPI. As shown in Figure 5C, DPI treatment significantly blunted phosphorylation of p38 and ERK1/2 stimulated by LPS. These results suggest that NADH oxidase signaling-induced TNF-α expression is mediated through ERK1/2 and p38 MAPK activation.
Role of NADH Oxidase in LPS-Induced TNF-α Expression in Isolated Hearts
To demonstrate the role of gp91phox-containing NADH oxidase in LPS-induced TNF-α expression at the organ level, mouse hearts isolated from gp91phox−/− and wild-type mice were perfused with LPS in the presence or absence of apocynin for 1 hour. The perfusates were collected and assayed for TNF-α production. Deficiency of gp91phox and apocynin treatment significantly decreased TNF-α production by 53% and 74% (Figure 6), respectively, which indicates that gp91phox-containing NADH oxidase also played an important role in adult hearts in terms of TNF-α production induced by LPS.
Role of NADH Oxidase in Myocardial Dysfunction of Endotoxemia
To explore the role of gp91phox-containing NADH oxidase in myocardial depression induced by endotoxemia, gp91phox−/− and wild-type mice were given vehicle or LPS (4 mg/kg IP). Four or 24 hours later, cardiac function was assessed in isolated hearts to avoid systemic reflex influences. Although there was no change in heart rate, heart work and rate of contraction were significantly reduced in endotoxemic mice compared with sham animals at both time points (Figures 7 and 8⇓), which indicates myocardial depression. Lack of gp91phox restored heart work and rate of contraction and relaxation without affecting heart rate in endotoxemic mice (Figures 7 and 8⇓). These results demonstrated that gp91phox contributed to myocardial dysfunction at both early (4 hours) and late (24 hours) stages of endotoxemia in mice.
The present study used cardiomyocytes, isolated hearts, and an in vivo model of endotoxemia to investigate the role of NADH oxidase in LPS-induced TNF-α expression. We demonstrated for the first time that gp91phox, a subunit of NADH oxidase, plays an important role in myocardial depression induced by endotoxemia and that gp91phox-containing NADH oxidase signaling contributes to LPS-stimulated TNF-α expression in cardiomyocytes. The effect of gp91phox-containing NADH oxidase was mediated through p38 and ERK1/2 MAPK signaling pathway. The present study provides definitive evidence that LPS-induced myocardial dysfunction is mediated by gp91phox-containing NADH oxidase. Activation of NADH oxidase represents a novel signaling pathway by which LPS induces p38 and ERK1/2 MAPK, leading to TNF-α expression in cardiomyocytes.
NADH Oxidase Activation in Sepsis
NAD(P)H oxidase has been shown to be a major source of oxidative stress in the myocardium.8–11 The upregulation of NAD(P)H oxidase activity has been observed in human failing hearts.20 NAD(P)H oxidase-derived O2− production has been demonstrated to contribute to various cardiovascular diseases, including cardiac hypertrophy,11 hypertension,21 and atherosclerosis.22,23 In sepsis, NAD(P)H oxidase activation in neutrophils plays a major role in mediating sepsis-induced acute lung injury.24,25 O2− production by NAD(P)H oxidase is involved in the pathogenesis of endothelial dysfunction,26 platelet-endothelial cell adhesion in intestinal venules,27 and production of peroxynitrite and prostaglandin E2 in activated rat microglia.28,29 Recent studies have shown that NAD(P)H oxidase plays an important role in LPS-induced TNF-α expression and neurotoxicity in microglia and astrocytes.30,31 In the heart, NAD(P)H oxidase activity and O2− production are increased in response to LPS stimulation.12,13 However, a causal relationship between NAD(P)H oxidase activation and myocardial dysfunction during sepsis has not been reported. In the present study, LPS-induced myocardial dysfunction was restored in gp91phox−/− mice. The present results demonstrated that gp91phox-containing NADH oxidase activity contributes to myocardial depression in endotoxemia. Mechanisms by which gp91phox-containing NADH oxidase mediates myocardial dysfunction are not fully understood. Data from the present study suggest that LPS activates gp91phox-containing NADH oxidase, which increases oxidative stress and TNF-α production in cardiomyocytes, leading to myocardial dysfunction.
NADH Oxidase Signaling in TNF-α Expression in LPS-Stimulated Cardiomyocytes
TNF-α has been shown to be a major factor responsible for myocardial depression during endotoxemia4,5; however, mechanisms by which LPS induces TNF-α expression in cardiomyocytes are not fully understood. The present study aimed to investigate the role of NADH oxidase in LPS-induced TNF-α expression in cardiomyocytes. Using cultured neonatal mouse cardiomyocytes and isolated heart preparations, we demonstrated the following. First, LPS increased gp91phox expression and O2− generation. Second, deficiency of gp91phox blunted TNF-α production in response to LPS. Third, the blocking of NADH oxidase activity abrogated LPS-induced TNF-α expression. The effect of NADH oxidase was mediated through O2− generation, hydroxyl radicals, and related oxidative stress, because either the O2− scavenger tempol, the hydroxyl radical scavenger DMTU, or the antioxidant NAC dose-dependently inhibited LPS-induced TNF-α expression. These results strongly support a pivotal role of gp91phox-containing NADH oxidase signaling in TNF-α expression in LPS-stimulated cardiomyocytes.
A deficiency of gp91phox blunted LPS-induced TNF-α production, whereas NADH oxidase inhibitors abrogated TNF-α production in cardiomyocytes, which suggests the presence of an alternative Nox isoform. Indeed, a previous study showed that the heart expresses at least 2 Nox isoforms, gp91phox and Nox4.15 Nox4 isoform was also detected in both wild-type and gp91phox−/− cardiomyocytes in the present study (data not shown). Further studies are required to investigate the contribution of different Nox isoforms in myocardial TNF-α expression during LPS stimulation.
NADH Oxidase in MAPK Activation by LPS
We have demonstrated that p38 MAPK is required for TNF-α expression in LPS-stimulated cardiomyocytes.3,4 MAPK is sensitive to oxidative stress. In the present study, our working hypothesis was that the signaling pathway downstream of NADH oxidase involves ERK1/2 and p38 MAPK activation. Consistent with this hypothesis, we showed that inhibition of NADH oxidase attenuated ERK1/2 and p38 MAPK phosphorylation in LPS-stimulated cardiomyocytes. Selective inhibition of ERK1 and p38α with their respective siRNAs significantly attenuated LPS-induced TNF-α expression. The present study suggests that LPS increases NADH oxidase activity, which increases oxidative stress and hydroxyl radicals through O2− generation. Increased oxidative stress and hydroxyl radicals subsequently lead to activation of ERK1/2 and p38 MAPK, which upregulates TNF-α expression in cardiomyocytes (Figure 9).
In summary, the present study provided definitive evidence that gp91phox-containing NADH oxidase activity is pivotal in LPS-induced TNF-α expression in the heart. This signaling pathway involves O2− generation and activation of ERK1/2 and p38 MAPK. Moreover, gp91phox-containing NADH oxidase contributes to myocardial dysfunction during endotoxemia (Figure 9). The present study suggests that NADH oxidase may represent a novel therapeutic target for TNF-α expression and myocardial dysfunction in sepsis.
This study was supported by grants awarded to Dr Qingping Feng from the Canadian Institutes of Health Research (MOP-64395) and the Heart & Stroke Foundation of Ontario (T-4923). Dr Feng is a recipient of a Premier’s Research Excellence Award (PREA) from the Province of Ontario. We thank Xuemei Xu for her expert technical assistance in cardiomyocyte culture.
Grandel U, Fink L, Blum A, Heep M, Buerke M, Kraemer HJ, Mayer K, Bohle RM, Seeger W, Grimminger F, Sibelius U. Endotoxin-induced myocardial tumor necrosis factor-alpha synthesis depresses contractility of isolated rat hearts: evidence for a role of sphingosine and cyclooxygenase-2–derived thromboxane production. Circulation. 2002; 102: 2758–2764.
Peng T, Lu X, Lei M, Feng Q. Endothelial nitric-oxide synthase enhances lipopolysaccharide-stimulated tumor necrosis factor-alpha expression via cAMP-mediated p38 MAPK pathway in cardiomyocytes. J Biol Chem. 2003; 278: 8099–8105.
Peng T, Lu X, Lei M, Moe GW, Feng Q. Inhibition of p38 MAPK decreases myocardial TNF-alpha expression and improves myocardial function and survival in endotoxemia. Cardiovasc Res. 2003; 59: 893–900.
Mohazzab HK, Kaminski PM, Wolin MS. Lactate and PO2 modulate superoxide anion production in bovine cardiac myocytes: potential role of NADH oxidase. Circulation. 1997; 96: 614–620.
Xiao L, Pimentel DR, Wang J, Singh K, Colucci WS, Sawyer DB. Role of reactive oxygen species and NAD(P)H oxidase in alpha1-adrenoceptor signaling in adult rat cardiac myocytes. Am J Physiol Cell Physiol. 2002; 282: C926–C934.
Li JM, Gall NP, Grieve DJ, Chen M, Shah AM. Activation of NADPH oxidase during progression of cardiac hypertrophy to failure. Hypertension. 2002; 40: 477–484.
Bendall JK, Cave AC, Heymes C, Gall N, Shah AM. Pivotal role of a gp91phox-containing NADPH oxidase in angiotensin II–induced cardiac hypertrophy in mice. Circulation. 2002; 105: 293–296.
Khadour FH, Panas D, Ferdinandy P, Schulze C, Csont T, Lalu MM, Wildhirt SM, Schulz R. Enhanced NO and superoxide generation in dysfunctional hearts from endotoxemic rats. Am J Physiol Heart Circ Physiol. 2002; 283: H1108–H1115.
Byrne JA, Grieve DJ, Bendall JK, Li JM, Gove C, Lambeth JD, Cave AC, Shah AM. Contrasting roles of NADPH oxidase isoforms in pressure-overload versus angiotensin II–induced cardiac hypertrophy. Circ Res. 2003; 93: 802–805.
Song W, Lu X, Feng Q. Tumor necrosis factor-alpha induces apoptosis via inducible nitric oxide synthase in neonatal mouse cardiomyocytes. Cardiovasc Res. 2000; 45: 595–602.
Tada H, Thompson CI, Recchia FA, Loke KE, Ochoa M, Smith CJ, Shesely EG, Kaley G, Hintze TH. Myocardial glucose uptake is regulated by nitric oxide via endothelial nitric oxide synthase in Langendorff mouse heart. Circ Res. 2000; 86: 270–274.
Rajagopalan S, Kurz S, Munzel T, Tarpey M, Freeman BA, Griendling KK, Harrison DG. Angiotensin II–mediated hypertension in the rat increases vascular superoxide production via membrane NADH/NADPH oxidase activation: contribution to alterations of vasomotor tone. J Clin Invest. 1996; 97: 1916–1923.
Warnholtz A, Nickenig G, Schulz E, Macharzina R, Brasen JH, Skatchkov M, Heitzer T, Stasch JP, Griendling KK, Harrison DG, Bohm M, Meinertz T, Munzel T. Increased NADH-oxidase–mediated superoxide production in the early stages of atherosclerosis: evidence for involvement of the renin–angiotensin system. Circulation. 1999; 99: 2027–2033.
Gao XP, Standiford TJ, Rahman A, Newstead M, Holland SM, Dinauer MC, Liu QH, Malik AB. Role of NADPH oxidase in the mechanism of lung neutrophil sequestration and microvessel injury induced by Gram-negative sepsis: studies in p47phox−/− and gp91phox−/− mice. J Immunol. 2002; 168: 3974–3982.
Brandes RP, Koddenberg G, Gwinner W, Kim D, Kruse HJ, Busse R, Mugge A. Role of increased production of superoxide anions by NAD(P)H oxidase and xanthine oxidase in prolonged endotoxemia. Hypertension. 1999; 33: 1243–1249.
Cerwinka WH, Cooper D, Krieglstein CF, Ross CR, McCord JM, Granger DN. Superoxide mediates endotoxin-induced platelet–endothelial cell adhesion in intestinal venules. Am J Physiol Heart Circ Physiol. 2003; 284: H535–H541.
Qin L, Liu Y, Wang T, Wei SJ, Block ML, Wilson B, Liu B, Hong JS. NADPH oxidase mediates lipopolysaccharide-induced neurotoxicity and proinflammatory gene expression in activated microglia. J Biol Chem. 2004; 279: 1415–1421.