Endotoxin-Induced Cardiomyopathy and Systemic Inflammation in Mice Is Prevented by Aldose Reductase Inhibition
Background— Sepsis is a systemic inflammatory response syndrome characterized by excessive production of inflammatory cytokines and cardiovascular collapse. Postreceptor signaling events that lead to stress responses and cytokine production are sensitive to redox changes and products of lipid peroxidation.
Methods and Results— We tested the hypothesis that inflammatory signaling and cytokine generation during sepsis depend on the activity of the enzyme aldose reductase, which catalyzes the reduction of lipid peroxidation–derived aldehydes and their glutathione conjugates. The results of the present study show that pharmacological inhibition of aldose reductase by sorbinil or knockdown of the enzyme by small interfering RNA prevents the activation of nuclear factor-κB and the release of tumor necrosis factor-α from lipopolysaccharide-stimulated RAW264.7 or H9c2 cells. Increases in serum and cardiac cytokines in response to lipopolysaccharide challenge were suppressed by inhibition of aldose reductase. Treatment with sorbinil blunted the activation of protein kinase C, c-Jun NH2-terminal kinase, and p38, as well as phosphorylation of interleukin receptor–associated kinase, IκB-α, IκB kinase complex-α/β, and phospholipase-γ1 and -β1. These changes were associated with decreased myocardial nuclear factor-κB and activating protein-1 activity, prostaglandin E2 production, induction of cyclooxygenase 2, and inducible nitric oxide synthase. Sorbinil treatment also induced functional recovery in myocardial fractional shortening in vivo and preserved contractile function of isolated perfused hearts. Inhibition of aldose reductase increased survival in mice injected with lethal doses of lipopolysaccharide.
Conclusions— The present demonstration that aldose reductase mediates endotoxin-induced inflammation and cardiomyopathy suggests that inhibition of this enzyme may be useful to attenuate maladaptive host responses and to treat acute cardiovascular dysfunction associated with endotoxic shock.
Received March 30, 2006; revision received August 23, 2006; accepted August 25, 2006.
Lipopolysaccharide (LPS) is a component of the outer envelope of all Gram-negative bacterial cell walls.1 During infection, LPS is released in the circulation, where it stimulates the pattern-recognizing toll-like receptor (TLR4)-MD2-CD14 receptor complex on circulating cells.2,3 Cells that lack constitutive TLR4 receptors, such as vascular endothelial and smooth muscle cells, respond to LPS by interacting with soluble CD14.4,5 Ligation of LPS with CD14 activates these receptors. This increases the transcription of inflammatory and immune-response genes via mechanisms that depend on dissociation of the IκB–nuclear factor (NF)-κB complex and translocation of NF-κB to the nucleus, where it activates cytokine gene promoters.2,6 Stimulation of the NF-κB pathway leads to increased transcription of several proinflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukin (IL)-1, as well as prostaglandins and nitric oxide (NO). Acting in an autocrine and paracrine manner, these and other cytokines amplify host responses to invading pathogens.7,8 Although activation of the innate immune response during host-pathogen interaction is initially adaptive, the inability to regulate immune responses,9 which causes immunoparalysis,4 leads to sepsis. Clinically, this results in vasodilatory shock, circulatory collapse, and often fatal myocardial depression.10 Although sepsis is associated with a high (30% to 50%) risk of mortality, microbial diagnosis is made in only half the cases.4 Similar symptoms of systemic inflammatory syndrome also develop in patients with severe heart failure who are treated with mechanical assist devices and those who have undergone prolonged cardiopulmonary bypass.10 Currently, few therapeutic options are available, and the mechanisms underlying the pathogenic development of this syndrome remain incompletely understood.4,9,10
Clinical Perspective p 1846
Uncontrolled cytokine production mediates several of the immunopathological features of LPS-induced shock.11 The cytokines released early, usually within 30 to 90 minutes of infection,8 stimulate a broad array of secondary responses that include changes in coagulation, vascular reactivity, cell adhesion, and myocardial contractility. In addition, some cytokines, particularly TNF-α and IL-1β, stimulate gene transcription via the NF-κB pathway, which leads to further increases in NO and cytokine production.4,8,10 Nevertheless, direct interruption of cytokine production has not been efficacious in the clinical management of sepsis.9,11 More effective therapeutic interventions are required, preferably to interrupt regenerative cycles of cytokine synthesis and the uncontrolled progression of inflammation.
Regenerative production of cytokines is perpetuated by repeated cycles of NF-κB activation.4,8,11 NF-κB, initially activated by ligation of the TLR4 receptor complex, increases the transcription of TNF-α and IL-β genes. These cytokines in turn cause further activation of NF-κB through paracrine and autocrine effects that lead to amplification of cytokine synthesis.8 Therefore, inhibition of NF-κB represents 1 approach for interrupting regenerative cytokine production and unresolved inflammation. Our previous studies showed that NF-κB activation could be prevented by inhibiting the enzyme aldose reductase (AR), which regulates postreceptor events that lead to stimulation and nuclear translocation of NF-κB.12,13 Thus, inhibition of AR may be 1 method of preventing NF-κB activation during sepsis.
AR is a member of the aldo-keto reductase superfamily classified as aldo-keto reductase 1B2.14 It is a cytosolic protein that catalyzes nicotinamide adenosine dinucleotide phosphate–dependent reduction of a wide range of aldehydes.15 AR displays high catalytic efficiency with aldehyde products of lipid peroxidation16,17 and their conjugates with glutathione.16,18,19 The enzyme also catalyzes the reduction of glucose to sorbitol, and an increase in AR-mediated metabolism of glucose has been linked to the development of secondary diabetic complications.15,20 Our recent studies show that inhibition of AR in vascular smooth muscle cells prevents diacylglycerol synthesis, thereby attenuating the activation of several protein kinase C (PKC) isoforms and downstream signaling events that lead to the activation of NF-κB.21 Inhibition of these signaling events prevents high glucose- and TNF-α–induced vascular smooth muscle growth12,22 and TNF-α–induced vascular endothelial cell apoptosis.13 Inhibitors of AR also prevent NF-κB activation in vivo in arterial lesions, and hence, they decrease neointimal hyperplasia in balloon-injured arteries.12 On the basis of these observations, we tested the hypothesis that inhibition of AR would prevent systemic inflammation and cardiac dysfunction during LPS-induced shock. Our results showing marked antiinflammatory effects of the inhibition of AR suggest that inhibition of AR could represent one approach for keeping renegade inflammatory responses in check.
Cell Culture and Animals
The C57BL/6 mice (weight ≈25 g) were obtained from the Jackson Laboratory (Bar Harbor, Me) and housed in pathogen-free conditions with free access to food and water at the institutional animal care facility. RAW264.7 macrophage cell line and H9C2 rat cardiac myoblasts were obtained from American Type Culture Collection (Manassas, Va) and grown in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum. The animals were maintained in accordance with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health and in accordance with the “Guideline of the Animal Care and Use Committee” of the University of Texas Medical Brach at Galveston and Southwestern Medical Center at Dallas, Tex.
Antisense Ablation of AR
Antisense ablation of AR was performed as described previously.12 Small interfering RNAs (siRNA) were designed to target the coding sequence (AATCGGTGTCTCCAACTTCAA) of mouse AR (GenBank accession No. BC085310). A scrambled oligonucleotide AR siRNA (AAAATCTCCCTAAATCATACA) was used as a control. RAW264.7 cells were grown in DMEM containing 10% fetal bovine serum and 1% penicillin and streptomycin at 37°C and 5% CO2 and seeded on 6-well plates. When the cells reached 60% to 70% confluence (in 24 hours), the medium was replaced with fresh DMEM without serum, and the cells were incubated with the siRNA to a final concentration of 100 nmol/L and with the RNAiFect transfection reagent (Qiagen, Valencia, Calif) as per the supplier’s instructions. After incubation for 15 minutes at 25°C, the medium was aspirated and replaced with fresh DMEM containing 10% serum, added dropwise to the cells. The cells were cultured for 48 hours at 37°C (5% CO2), and changes in AR expression were determined by Western blot analysis with anti-AR antibodies and by measuring AR activity in the total cell lysates.12
The mice were preinjected with sorbinil (25 mg/kg body weight IP) or carrier, and after 24 hours, LPS (4 mg/kg body weight) was injected with or without sorbinil. The RAW264.7 cells were preincubated with 10 μg/mL sorbinil for 24 hours followed by incubation with 1 μg/mL LPS. At various time intervals, the animals were killed, and blood and heart tissues were collected. Cytokine (TNF-α, IL-6, IL-12, and interferon [IFN]-γ) and chemokine (monocyte chemoattractant protein-1) levels were measured in the serum and heart homogenates with BD Biosciences mouse inflammation cytometric bead array kits (BD Biosciences, San Jose, Calif). Cyclooxygenase-2, prostaglandin E2, and nitrate/nitrite levels in heart homogenates were determined with commercially available assay kits obtained from Cayman Chemical (Ann Arbor, Mich). The nuclear extracts were prepared, and electrophoretic mobility gel shift assay was performed as described previously.12 The NF-κB activity was determined by electrophoretic mobility gel shift assay or with the colorimetric nonradioactive NF-κB p65 Transcription factor assay kit (Chemicon, Temecula, Calif). Activating protein-1 (AP-1) activity was determined by electrophoretic mobility gel shift assay and PKC activation with the Promega SignaTECT total PKC assay system (Promega, Madison, Wis), as described previously.12
Western Blot Analysis
Equal amounts of cell extracts or heart homogenates were separated on 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis, electroblotted on nitrocellulose membranes, and probed with phospho- and nonphospho-specific antibodies against phospholipase C (PLC)-β1 PLC-γ1, IκB kinase complex (IKK), interleukin receptor–associated kinase (IRAK)-1, IκB-α, p38–mitogen-activated protein (MAP) kinase, c-Jun NH2-terminal kinase (JNK), or extracellular signal-regulated kinase 1/2. Antibody binding was detected by enhanced Pico chemiluminescence (Pierce, Rockford, Ill). Immunopositive bands were quantified with Kodak Image station 2000R loaded with Kodak 1-dimensional image-analysis software (Kodak, Rochester, NY).
Cardiac Dysfunction Assessment by Echocardiography
Echocardiography was performed on 5 groups of mice: (1) control mice (no treatment); (2) mice treated with sorbinil for 5 days, starting 3 days before echocardiography; (3) mice challenged with LPS (4 mg/kg IP in 200 μL of phosphate-buffered saline); (4) mice treated with sorbinil (as in group 2) and LPS challenge (as in group 3) on day 3; and (5) mice treated with sorbinil 2 hours after LPS challenge. Pretreatment with sorbinil (25 mg · kg−1 · d−1) was followed by LPS challenge. Serial echocardiograms to assess cardiac function were performed with M-mode measurements as described previously23 (online-only Data Supplement). M-mode measurement data represent the average of 9 selected cardiac cycles from at least 2 separate scans. End diastole was defined as the maximal left ventricular diastolic dimension, and end systole was defined as the peak of posterior wall motion. Fractional shortening percentage (FS%), a surrogate of systolic function, was calculated from left ventricular dimensions as follows equation
in which LVED indicates left ventricular end-diastolic diameter and LVES is left ventricular end-systolic diameter.
Ex Vivo Cardiac Function Determination by Langendorff
Mouse heart function was determined by the Langendorff assay procedure as described previously.24 Procedural details are described in the online Data Supplement.
Cardiac function determined by the Langendorff preparation (including stabilization data) is expressed as mean±SEM, and separate analyses were performed for each left ventricular pressure, +dP/dtmax, and −dP/dtmax, as a function of treatment group and perfusate Ca2+ concentration (or constant coronary flow for stabilization measurements) with a repeated-measures ANOVA. A multiple-comparison procedure with the Bonferroni correction was used to determine the significance of differences between groups. Cardiac function determined by M-mode echocardiography is expressed as FS% ±SE and was analyzed by 1-way repeated-measures ANOVA. Additional comparisons were performed with the Tukey test to determine significant differences between specific groups. Statistical significance for all analyses was defined as P<0.01. Survival rates were compared by the Kaplan-Meier method and analyzed by log-rank test. All statistical analyses were performed with SigmaStat 2.03 (SPSS, Chicago, Ill) and Microsoft Office Excel (Microsoft, Seattle, Wash).
The authors had full access to the data and take full responsibility for its integrity. All authors have read and agree to the manuscript as written.
Inhibition of AR Prevents LPS-Induced Activation of NF-κB and Cytokine Release in Macrophages
We first examined how inhibition of AR would affect NF-κB activation and cytokine production in LPS-stimulated macrophages. Treatment of RAW264.7 macrophages with LPS led to a 6-fold increase in NF-κB activity within 6 hours of stimulation, and it remained elevated at this level for 12 hours. Treatment with the AR inhibitor sorbinil suppressed the initial increase in NF-κB activity, and after 12 hours, the increase in NF-κB activity was not statistically different from the values obtained before LPS stimulation. Sorbinil treatment also blunted LPS-induced increases in TNF-α and IL-6 (Figure 1B).
Although sorbinil is a relatively specific AR inhibitor, its nonspecific effects cannot be rigorously excluded. Therefore, we examined the effects of ablating the AR transcripts using siRNA. As shown in Figure I in the online Data Supplement, 48 hours after transfection, the level of AR protein in RAW264.7 macrophages transfected with AR siRNA was <5% of the level in untransfected cells or cells transfected with control siRNA. Consistent with the pharmacological data, transfection with AR siRNA, but not with scrambled oligonucleotides, attenuated LPS-induced NF-κB activation and cytokine production (Figures 1C and 1D). Pharmacological inhibition of AR by sorbinil or tolrestat or siRNA knockdown of AR mRNA also significantly prevented LPS-induced NF-κB and AP-1 activation and synthesis of TNF-α and IL-6 in the cardiomyocyte-derived H9c2 cell line (Data Supplement, Figure II). These findings suggest that inhibition of AR suppresses NF-κB–mediated gene transcription in LPS-treated cells of macrophage or cardiac origin.
Inhibition of AR Prevents LPS-Induced Inflammation In Vivo
To determine whether inhibition of AR would also prevent systemic inflammation in vivo, we examined the effects of AR inhibition on NF-κB signaling pathways and myocardial dysfunction in a mouse model of LPS-induced shock. After treatment with sorbinil or vehicle alone, mice were injected intraperitoneally with a sublethal dose of LPS (4 mg/kg body weight). Changes in inflammatory cytokines and chemokines were measured in the serum and the heart. In LPS-treated mice, the levels of TNF-α, IL-6, IL-12, IL-1β, and IFN-γ increased in the serum by 3- to 6-fold after 8 hours of LPS exposure (Figure 2A and 2B). The levels of these cytokines began declining by 12 hours but remained elevated for 24 hours after treatment. In sorbinil-treated mice, the increase in serum cytokine levels at different time intervals was significantly less than that with LPS alone (Figure 2A and 2B). Sorbinil treatment also prevented inflammatory markers in the heart. Within 4 hours of LPS treatment, levels of TNF-α, IL-1β, and monocyte chemoattractant protein-1 in the heart were markedly elevated (Figure 2C and 2D). No significant increase in IL-12 was observed, and the increase in IL-6 was small. The levels of IFN-γ were elevated by 2.5-fold within 4 hours and continued to increase for the next 24 hours. In sorbinil-treated mice, cardiac cytokine production was largely abrogated, and only small increases in TNF-α, IFN-γ, and monocyte chemoattractant protein-1 were noted, which were significantly smaller than in mice treated with LPS alone. Treatment with LPS also increased the myocardial content of cyclooxygenase-2 protein, prostaglandin E2, and nitrate (Figure 2E), which were markedly blunted in sorbinil-treated mice. The efficacy of sorbinil in preventing an LPS-induced increase in serum cytokines was similar in mice that were treated with sorbinil either 3 days before LPS injection or 2 hours after LPS injection (Data Supplement, Table I). Furthermore, comparable changes in TNF-α, IL-6, and IFN-γ were observed in mice treated with 2 other structurally unrelated AR inhibitors, tolrestat and zopolrestat (Data Supplement, Figure III). Collectively, these data suggest that inhibition of AR interrupts and suppresses LPS-induced increases in plasma and cardiac cytokines, even in a therapeutic setting.
To determine whether AR inhibition could rescue cardiac dysfunction associated with systemic inflammation, we performed serial echocardiography. After 4 hours of LPS injection, the FS% decreased from 50% to 25% (Figure 3A). A similar decrease in FS% was initially observed in mice treated with sorbinil and LPS; however, 8 hours after LPS treatment, the sorbinil-treated mice showed significant recovery compared with mice treated with LPS alone. After 48 hours, the FS% in sorbinil-treated mice was >40%, whereas it remained near 25% in those treated with LPS alone (Figure 3A). FS% was not affected in unchallenged controls or in mice treated with sorbinil in the absence of LPS. Similar results were observed in mice given sorbinil 2 hours after LPS challenge (data not shown). The general activity level of LPS-exposed mice was consistent with the echocardiographic findings: Despite LPS injection, the sorbinil-treated mice exhibited normal grooming and other activities within 24 hours, whereas the mice treated with LPS alone remained inactive and huddled close to one another.
To further assess the effect of AR inhibition on cardiac function, we perfused spontaneously beating isolated mouse hearts in the Langendorff mode with the AR inhibitor and challenged them with LPS. Hearts perfused with LPS alone showed diminished contractility, which was more pronounced at high calcium levels and high coronary flow rates, which indicates that LPS treatment depresses the positive ionotropic responses to calcium and coronary flow. The attenuation of calcium and coronary flow–dependent changes in LPS-treated hearts was reflected in both positive and negative dP/dtmax, the velocity of contraction and relaxation, respectively (Figure 3B and 3C). The time to maximal ±dP/dt, coronary perfusion pressure, coronary vascular resistance, and heart rate were unaffected. Treatment with sorbinil significantly prevented the loss of inotropy and enhanced the sensitivity of the heart to calcium and coronary flow (Figure 3B and 3C). These findings demonstrate that inhibition of AR activity attenuates LPS-induced systolic and diastolic dysfunctions.
Inhibition of AR Suppresses LPS Lethality
To determine whether in addition to preventing sublethal effects of LPS, inhibition of AR also affects LPS lethality, we examined whether treatment with sorbinil would affect LPS-induced death. The effects of sorbinil were studied with 2 different protocols. In protocol 1, the mice were pretreated with sorbinil for 3 days before LPS treatment, whereas in protocol 2, the mice were treated with sorbinil 2 hours after injection of LPS. In mice that were pretreated with carrier alone, the lethal dose of LPS for 50% of the population was ≈14 mg/kg body weight over 48 hours. Both protocols of sorbinil treatment, however, resulted in significantly higher survival. As shown in Figure 4, 16 mg of LPS per kilogram of body weight resulted in only 25% survival, whereas in the sorbinil-pretreated group (protocol 1), none of the mice died. In the group in which sorbinil was injected 2 hours after LPS (protocol 2), only 4 mice died (66.6% survival). Even at a dose of 24 mg/kg body weight, at which 100% of the control mice died, 60% of mice survived in the LPS plus sorbinil-pretreatment group, and 40% of mice survived in the sorbinil-posttreatment group. Additional statistical analysis of the survival rate at each dose of LPS compared by the Kaplan-Meier method and analyzed by log-rank test is given as supplementary Figure IV and supplementary Tables II and III. Taken together, these data support the notion that inhibition of AR allows mice to survive a potentially lethal LPS challenge.
Inhibition of AR Extinguishes LPS Signaling
We next asked whether the cardioprotective effects of sorbinil might reflect suppression of the inflammatory signaling cascade. In untreated mice, LPS increased cardiac NF-κB activation 16-fold and AP-1 activation 5-fold within 2 hours (Figure 5A and 5B; supplementary Figure V). The level of these transcription factors remained elevated even 24 hours after LPS challenge. Inhibition of AR severely blunted these responses. In sorbinil-treated mice, LPS-induced activation of NF-κB and AP-1 was decreased by ≈70% at 2 hours compared with mice treated with LPS alone. In the presence of sorbinil, the levels of these transcription factors were close to baseline values by 24 hours. Similar results were observed in mice treated with sorbinil 2 hours after LPS exposure (Data Supplement, Figure VI). Furthermore, subsequent to LPS injection, cardiac PKC activity was increased by 5-fold in 1 hour, and it remained ≈3-fold elevated for the next 24 hours (Figure 5C). In mice pretreated with sorbinil, the increase in PKC activity was significantly less. In addition, the LPS-induced increase in phosphorylation of PLC-β1 and -γ1 was significantly attenuated by AR inhibition (Figure 6). Similarly, cardiac inducible NO synthase levels were increased by ≈3-fold 8 hours after LPS injection and remained elevated for 24 hours (Figure 5D). In sorbinil-treated mice, however, inducible NO synthase levels increased only slightly and returned to baseline levels in 24 hours. Thus, inhibition of AR appears to attenuate NF-κB activation and protects against cardiovascular collapse in the setting of overwhelming sepsis.
To further assess the extent to which inhibition of AR interrupts signaling events downstream of TLR4 activation, we examined changes in IRAK and MAP kinases in hearts of mice treated with LPS. Previous studies show that TLR4 stimulation leads to phosphorylation of IRAK and the activation of p38 and JNK via a TAK1 (transforming growth factor-β–activated kinase)-dependent pathway that also activates NF-κB signaling.6 In agreement with these observations, we found that LPS stimulated the phosphorylation of IRAK, p38, and JNK (Figure 6). IRAK phosphorylation was increased 3-fold within 15 minutes of stimulation with LPS. Maximal MAP kinase activation (≈8-fold) was observed at 30 minutes, and the activation was sustained up to 120 minutes for p38 and up to 90 minutes for JNK. In contrast, extracellular signal-regulated kinase was only mildly activated (≈2-fold) after 90 to 120 minutes of stimulation. Treatment with sorbinil suppressed LPS-induced increases in IRAK, p38, and JNK, whereas extracellular signal-regulated kinase was only marginally affected (Figure 6). In addition, inhibition of AR also significantly prevented LPS-induced activation of IKK and Iκ-Bα phosphorylation. Inhibition of IKK was particularly profound, and a weak phospho-IKK signal was observed only transiently at 90 minutes. Collectively, these data suggest that inhibition of AR does not directly interfere with NF-κB but interrupts signaling events upstream of IKK activation.
Results of the present study show for the first time that inhibition of AR, an enzyme involved in glucose and aldehyde metabolism, prevents the cytokine production and cardiac dysfunction associated with the systemic inflammatory response syndrome in LPS-treated mice. The present studies suggest that the salutary effects of AR inhibition may be related to the inhibition of inflammatory signaling mediated by transcription factors (NF-κB and AP-1) and stress-activated MAP kinases (JNK and p38). These results raise the possibility that inhibition of AR may be beneficial in treating sepsis and other clinical conditions associated with maladaptive inflammation.
A variety of clinical conditions are associated with a dysregulation of inflammatory responses. Although the most common of these is sepsis, high concentrations of cytokines are also generated by trauma, ischemia-reperfusion, acute rejection, antigen-specific immune responses, and several acute inflammatory states, such as acute hepatitis and pancreatitis.10 Given that tissue dysfunction in these conditions has been linked to excessive inflammation, it appears likely that as in sepsis, increased cytokine production in these conditions could also be mediated by AR. Hence, excessive cytokine production under these conditions could be prevented by the inhibition of AR. A general role of AR in mediating inflammation and cytokine generation is consistent with recent observations showing that inhibition of AR prevents PKC and NF-κB activation by a variety of stimuli such as TNF-α, fibroblast growth factor-2, platelet-derived growth factor-AB, angiotensin II, and high glucose12,13,21,22 and high glucose–induced phosphatidylinositol-3 kinase25 and Janus kinase 2,26 which suggests that AR regulates stress response and the activation of NF-κB and other PKC-sensitive transcription factors.
Activation of NF-κB starts with the assembly of a multiprotein complex comprising IKKα/IKKβ proteins held together with a scaffolding protein, IKKγ (NEMO). This complex phosphorylates IκB-α, which then dissociates from NF-κB and becomes ubiquitinated and degraded.7 The present finding that LPS-stimulated phosphorylation of IKKα/β and IκB-α is prevented by sorbinil (Figure 6) suggests that inhibition of AR interferes with events that occur before the formation of the IKKα/β signalosome. In LPS signaling, the IKKα/β complex is assembled via a TAK1-dependent pathway that also activates JNK and p38.6 The observation that phosphorylation of JNK and the p38 kinases was severely attenuated by sorbinil further indicates that the signals that precede TAK1 activation, such as IRAK activation, are prevented by sorbinil treatment and that inhibition of AR does not directly interfere with NF-κB, stress-activated protein kinase, or their downstream effectors.
LPS-triggered signaling events further upstream of IKK activation are mediated by the activation of PKC. Macrophage PKC activity is increased by LPS stimulation, and PKC inhibitors prevent LPS-induced NF-κB activation and the release of NO, TNF-α, and IL-1β.27,28 In agreement with a central role of PKC, we found a marked PKC activation within 2 hours of stimulation with LPS (Figure 5C). Sorbinil treatment prevented LPS-induced activation of PKC, which supports our previous observation that inhibition of AR prevents PKC activation.12,21,22 The present results, however, do not rule out the possibility that inhibition of AR could prevent inflammatory signaling by preventing PKC-independent signaling pathways. The present in vivo experiments show that treatment with AR inhibitors prevented cytokine production even when the drug was administered 2 hours after LPS challenge, although in the in vitro experiments, PKC and NF-κB activations were early events (<2 hours). Hence, either sustained activation of PKC and NF-κB is required for in vivo effects or other non–PKC- or NF-κB–dependent mechanisms may be responsible for the effects of AR inhibitors. Nevertheless, even though the mechanism by which AR facilitates PKC signaling remains unclear, we propose that inhibition of AR prevents events that lead to the activation of phosphatidylcholine-PLC, which is activated by LPS.29 We have recently reported that inhibition of AR prevents phosphatidylinositol-PLC–dependent synthesis of diacylglycerol.21 A similar mechanism may be responsible for the effects of AR inhibition on LPS-induced PLC-β1, PLC-γ1, PKC, and NF-κB activation. Alternatively, inhibition of AR could affect signaling owing to products of lipid peroxidation or their glutathione conjugates. In this regard, it is interesting that the oxidized phospholipids such as 1-palmitoyl, 2-oxovaleryl phosphocholine (POVPC), which is a substrate of AR,17 inhibit NF-κB activation and increase survival in mice injected with lethal doses of LPS.30 Thus, inhibition of AR could potentially prevent inflammation by allowing oxidized phospholipids to accumulate. However, oxidized phospholipids bind LPS and thereby prevent its binding to CD14 and LPS-binding protein and its presentation to TLR4.30 These lipids do not inhibit TNF-α–induced NF-κB activation,30 which is sensitive to AR inhibition.12 Hence, changes in phospholipid metabolism are unlikely to account fully for the antiinflammatory effects of AR inhibition, although the possibility that multiple mechanisms contribute to the in vivo efficacy of AR inhibition in preventing sepsis cannot be ruled out.
Regardless of the specific mechanisms involved, the present data showing prevention of NF-κB and AP-1 activation by sorbinil in myocardial tissue of LPS-treated mice suggest that this is an in vivo phenomenon and not restricted to isolated cells in culture. This view is further reinforced by the observation that sorbinil prevented the rise in myocardial cytokines (Figure 2) and the activation of cyclooxygenase-2 (Figure 2E) and inducible NO synthase (Figure 5D) activity in the heart, an observation that inhibition of AR prevents not only the activity of inflammatory transcription factors but also the products of the genes they activate.
Our observation that inhibition of AR decreases cytokine release and promotes recovery of cardiac function suggests that this enzyme is a critical modifier of LPS toxicity. Because excessive cytokine production is a significant problem in sepsis, treatment with AR inhibitors may be useful in attenuating both cytokine generation and their ability to induce tissue dysfunction. That inhibition of AR could affect tissue responses directly is suggested by our previous observation that inhibition of AR blocks TNF-α–induced NF-κB activation in endothelial cells and smooth muscle cells.13 The present data showing that inhibition of AR preserves cardiac function in an isolated perfused heart preparation further underscores the significance of AR in mediating oxidative stress signals. Hence, inhibition of AR could diminish not only the extent to which cytokines are synthesized but also the intensity with which tissues respond to cytokine challenge.
It is significant that treatment with sorbinil was able to prevent inflammation even when the therapy was initiated after LPS challenge. One of the reasons the clinical efficacy of anticytokine production drugs cannot be tested is that cytokines are released in the early stages of infections, ie, patients present too late for the therapy to be effective.4 Hence, we propose that the inhibition of metabolic pathways, which simultaneously decreases both cytokine production and toxicity, is likely to be a more tractable approach because it could treat secondary tissue inflammation even if it is too late to prevent early systemic release of cytokines. Moreover, because mortality is not primarily due to the infection but to the secondary tissue dysfunction and multiorgan failure,4,10 it appears likely that sepsis deaths could be prevented by AR inhibitors. In summary, the present results provide the first line of evidence for an unanticipated role of an aldehyde-metabolizing enzyme in mediating acute inflammatory responses and provide a new concept that inhibition of AR might be useful in preventing the clinical sequelae of sepsis or acute hemorrhagic or cardiogenic shock.
Sources of Funding
This work was supported by the National Institutes of Health grants GM71036 (to Dr Ramana), ES-11860 (to Dr Bhatnagar), and DK36118 and EY01677 (to Dr S.K. Srivastava). Dr D. Srivastava was supported by grants from the National Heart, Lung, and Blood Institute/National Institutes of Health, American Heart Association, and March of Dimes. We are also grateful for the use of National Institute of Environmental Health Sciences grant-supported core facilities (ESO6676-11).
Landmann R, Zimmerli W, Sansano S, Link S, Hahn A, Glauser MP, Calandra T. Increased circulating soluble CD14 is associated with high mortality in gram-negative septic shock. J Infect Dis. 1995; 171: 639–644.
Ramana KV, Chandra D, Srivastava S, Bhatnagar A, Aggarwal BB, Srivastava SK. Aldose reductase mediates mitogenic signaling in vascular smooth muscle cells. J Biol Chem. 2002; 277: 32063–32070.
Ramana KV, Bhatnagar A, Srivastava SK. Inhibition of aldose reductase attenuates TNF-alpha-induced expression of adhesion molecules in endothelial cells. FASEB J. 2004; 18: 1209–1218.
Srivastava S, Spite M, Trent JO, West MB, Ahmed Y, Bhatnagar A. Aldose reductase-catalyzed reduction of aldehyde phospholipids. J Biol Chem. 2004; 279: 53395–53406.
Dixit BL, Balendiran GK, Watowich SJ, Srivastava S, Ramana KV, Petrash JM, Bhatnagar A, Srivastava SK. Kinetic and structural characterization of the glutathione-binding site of aldose reductase. J Biol Chem. 2000; 275: 21587–21595.
Ramana KV, Friedrich B, Tammali R, West MB, Bhatnagar A, Srivastava SK. Requirement of aldose reductase for the hyperglycemic activation of protein kinase C and formation of diacylglycerol in vascular smooth muscle cells. Diabetes. 2005; 54: 818–829.
Ramana KV, Friedrich B, Srivastava S, Bhatnagar A, Srivastava SK. Activation of nuclear factor-kappaB by hyperglycemia in vascular smooth muscle cells is regulated by aldose reductase. Diabetes. 2004; 53: 2910–2920.
Garner LB, Willis MS, Carlson DL, DiMaio JM, White MD, White DJ, Adams GA, Horton JW, Giroir BP. Macrophage migration inhibitory factor is a cardiac-derived myocardial depressant factor. Am J Physiol Heart Circ Physiol. 2003; 285: H2500–H2509.
Campbell M, Trimble ER. Modification of PI3K- and MAPK- dependent chemotaxis in aortic vascular smooth muscle cells by protein kinase C βII. Circ Res. 2005; 96: 197–206.
Shaw S, Wang X, Redd H, Alexander GD, Isales CM, Marrero MB. High glucose augments the angiotensin II-induced activation of JAK2 in vascular smooth muscle cells via the polyol pathway. J Biol Chem. 2003; 278: 30634–30641.
Bakouche O, Moreau JL, Lachman LB. Secretion of IL-1: role of protein kinase C. J Immunol. 1992; 148: 84–91.
Shapira L, Takashiba S, Champagne C, Amar S, Van Dyke TE. Involvement of protein kinase C and protein tyrosine kinase in lipopolysaccharide-induced TNF-alpha and IL-1 beta production by human monocytes. J Immunol. 1994; 153: 1818–1824.
Monick MM, Carter AB, Gudmundsson G, Mallampalli R, Powers LS, Hunninghake GW. A phosphatidylcholine-specific phospholipase C regulates activation of p42/44 mitogen-activated protein kinases in lipopolysaccharide-stimulated human alveolar macrophages. J Immunol. 1999; 162: 3005–3012.
Bacterial infections can result in overwhelming septic shock, which equates to a mortality rate of 30% despite antibiotic therapy. Antibiotic-induced lysis of bacteria results in a concentrated release of cell-wall lipopolysaccharide (LPS) that induces reactive oxygen species, which stimulate a nuclear factor (NF)-κB–dependent activation of cytokines and chemokines. This uncontrolled host response is deleterious and results in cardiovascular collapse from vascular and myocardial dysfunction. Here, we show that the enzyme aldose reductase (AR) mediates bacterial endotoxin LPS-induced signaling cascade involving phosphorylation of phospholipase C, protein kinase C, and IκB, which ultimately results in nuclear translocation of NF-κB. Inhibition of AR in macrophages disrupted this signaling cascade and blunted the cytokine response to LPS. Correspondingly, AR inhibition prevented LPS-induced NF-κB activation and cytokine release in vivo and prevented the cardiovascular collapse and mortality typically associated with bacterial sepsis. The present study describes a novel role for AR in blunting a subset of deleterious NF-κB–dependent inflammatory processes; hence, the inhibition of AR may be a useful approach to attenuate maladaptive host responses and treat acute cardiovascular collapse due to endotoxic shock.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.106.630830/DC1.