|Home | About | Journals | Submit | Contact Us | Français|
Myocardial tumour necrosis factor-α (TNF-α) expression induces cardiac dysfunction in endotoxemia. The aim of this study was to investigate the role of mitogen-activated protein kinase phosphatase-1 (MKP1) pathway in myocardial TNF-α expression and cardiac function during endotoxemia.
Lipopolysaccharide (LPS) increased MKP1 expression in the myocardium in vivo and in cultured neonatal cardiomyocytes in vitro. LPS-induced extracellular signal-regulated kinase (ERK) 1/2 and p38 phosphorylation in the myocardium was prolonged in MKP1−/− mice. Myocardial TNF-α mRNA and protein levels were enhanced in MKP1−/− compared with wild-type (WT) mice in endotoxemia, leading to a further decrease in cardiac function. To study if Rac1/p21-activated kinase 1 (PAK1) signalling regulates MKP1 expression, cardiomyocytes were treated with LPS. Inhibition of Rac1 and PAK1 by a dominant negative Rac1 adenovirus (Ad-Rac1N17) and PAK1 siRNA, respectively, blocked LPS-induced MKP1 expression in cardiomyocytes. PAK1 siRNA also decreased p38 and c-Jun N-terminal kinase (JNK) activation, and TNF-α expression induced by LPS. Furthermore, deficiency in either Rac1 or JNK1 decreased myocardial MKP1 expression in endotoxemic mice.
LPS activates the Rac1/PAK1 pathway, which increases myocardial MKP1 expression via JNK1. MKP1 attenuates ERK1/2 and p38 activation, inhibits myocardial TNF-α expression, and improves cardiac function in endotoxemia. Thus, MKP1 represents an important negative feedback mechanism limiting pro-inflammatory response in the heart during sepsis.
Sepsis is the 10th most common cause of death in the USA.1 Cardiac dysfunction frequently accompanies severe sepsis and septic shock, and is associated with a significant increase in mortality rate (70%) compared with septic patients without cardiovascular impairment (20%).2 Tumour necrosis factor-α (TNF-α) produced by cardiomyocytes is a major contributing factor to cardiac dysfunction.2,3 However, the underlying molecular mechanisms regulating myocardial TNF-α production during sepsis remain elusive.
Mitogen-activated protein kinases (MAPKs), including p38, extracellular signal-regulated kinase (ERK) 1/2, and c-Jun N-terminal kinase (JNK) 1/2 are key signalling molecules regulating inflammatory responses and the expression of pro-inflammatory cytokines.4,5 It has been demonstrated that these MAPKs exhibit different effects on myocardial TNF-α expression during endotoxemia.6–10 Indeed, activation of ERK1/2 and p38 promotes TNF-α expression.6–8,10 In contrast, we recently demonstrated that JNK1 decreases TNF-α expression and improves cardiac function through inhibition of ERK1/2 and p38 activity.9 However, the molecular mechanisms by which JNK inhibits ERK1/2 and p38 are not fully understood.
Inactivation of MAPKs is achieved primarily by MAPK phosphatases (MKPs) that dephosphorylate phosphothreonine and phosphotyrosine residues of MAPKs.11,12 Studies have shown that mitogen-activated protein kinase phosphatase-1 (MKP1) negatively regulates the inflammatory response of the innate immune system by accelerating MAPK inactivation and attenuating the production of pro-inflammatory cytokines, including TNF-α, interleukin (IL)-1β, and IL-6 in macrophages following lipopolysaccharide (LPS) stimulation.13,14 Furthermore, deficiency in MKP1 results in a significantly higher incidence of mortality during endotoxemia, suggesting a protective role of MKP1 in sepsis.14–16 MKP1 is also expressed in cardiomyocytes and is involved in cardiomyocyte apoptosis and cardiac hypertrophy.17–19 However, the role of MKP1 in myocardial TNF-α expression and cardiac dysfunction in endotoxemia remains unknown.
MKP1 has intrinsic phosphatase activity and is inducible in response to extracellular stimuli. The mechanisms responsible for regulating MKP1 expression are cell type specific.20–23 For example, ERK1/220 and JNK21 are responsible for MKP1 induction in fibroblasts, while ERK, JNK1, and p38 are required for MKP1 expression in macrophages.13,22,24,25 Rac is necessary for cyclic strain stress-induced MKP1 expression in smooth muscle cells.23 We have demonstrated that Rac1 is a critical regulator of TNF-α expression and cardiac dysfunction in endotoxemia.12 Interestingly, p21-activated kinase (PAK), a serine-threonine protein kinase, acts as a downstream effector of Rac.26 Furthermore, activation of the Rac1/PAK pathway increases pro-inflammatory factor expression in macrophages through JNK and p38.27 PAK1 is the main PAK isoform in cardiomyocytes.28 The role of PAK1 in MKP1 and TNF-α expression in cardiomyocytes has not been elucidated.
In the present study, we hypothesized that MKP1 is induced during endotoxemia via the Rac1/PAK1/JNK pathway in cardiomyocytes, leading to inhibition of TNF-α expression and improvement of cardiac function in endotoxemia. To test this hypothesis, cultured cardiomyocytes and MKP1−/− mice were employed. Our results demonstrated for the first time that MKP1 represents an important negative feedback mechanism in limiting pro-inflammatory response in the heart during endotoxemia.
The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication #85-23, revised 1996) and the experimental protocols were approved by Animal Use Subcommittee at the University of Western Ontario. C57BL/6 wild-type (WT), Rac1 floxed (Rac1f/f), and JNK1−/− mice were purchased from the Jackson Laboratory (Bar Harbor, Maine). Cre transgenic mice (CreTG/+), which overexpressed Cre recombinase under the control of α-myosin heavy-chain (MHC) promoter, were provided by Dr E. Dale Abel (University of Utah, UT, USA). The generation of cardiomyocyte-specific Rac1 knockout mice (Rac1CKO) was achieved by breeding Rac1f/f mice with CreTG/+ mice as previously described.29 MKP1−/− mice were kindly provided by Bristol-Myers Squibb Pharmaceutical Research.30 Neonatal mice were sacrificed by cervical dislocation without anaesthesia. Hearts were removed, and cardiomyocytes were isolated and cultured according to methods we have previously described.31,32 Cells were treated with LPS (Sigma, Oakville, Ontario, Canada), p21-activated kinases inhibitor III (IPA-3, EMD Biosciences, San Diego, CA, USA) and SP600125 (Enzo Life Sciences, Plymouth Meeting, PA, USA), infected with adenoviruses or transfected with small-interfering RNAs (siRNAs).
Cardiomyocytes were infected with adenoviruses carrying a dominant-negative form of Rac1 (Ad-Rac1N17, Vector Biolabs, Philadelphia, PA, USA), MKP1 (Ad-MKP1), or LacZ (Ad-LacZ, Vector Biolabs, Philadelphia, PA, USA) at a multiplicity of infection of 10 plaque forming units/cell. Adenovirus-mediated gene transfer was applied as previously described.10 Experiments were performed after 48 h of adenoviral infection.
Cardiomyocytes were treated with PAK1 siRNA (2.5 μM) to knockdown PAK1 expression (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). A scrambled siRNA (Santa Cruz Biotechnology Inc.) was used as a control. The transfection was performed with transfection reagent (Santa Cruz Biotechnology Inc.) according to the manufacturer's instructions. After transfection, cells were maintained in normal culture medium for additional 48 h before LPS treatment.
Total RNA was extracted from neonatal cardiomyocytes and myocardium using the Trizol Reagent (Invitrogen, Burlington, ON, Canada), following the manufacturer's instructions. The MKP1 primers were: upstream 5′-GGA GAT CCT GTC CTT CCT GTA-3′ and downstream 5′-CTG ATG TCT GCC TTG TGG TTG-3′. TNF-α and 28S primers were same as in our previous report.10 MKP1 and TNF-α mRNA levels were determined by real-time reverse transcriptase–polymerase chain reaction (RT–PCR).10
TNF-α protein levels were measured using a mouse TNF-α ELISA kit (eBioscience, San Diego, CA, USA), according to the manufacturer's instructions. The measurements were standardized with cell numbers.
Ten to 20 μg of protein lysates were subjected to separation on a 10% SDS–PAGE gel, followed by electrotransfer to nitrocellulose membranes. Blots were probed with specific antibodies against ERK1/2, phospho-ERK1/2, p38, phospho-p38, JNK1/2, phospho-JNK1/2, PAK1 (1:1000), and phospho-PAK1 (1:500, Cell Signaling Technology, Danvers, MA, USA), MKP1 (1:1000, Upstate, Lake Placid, NY, USA), α-actinin, and GAPDH (Santa Cruz Biotechnology Inc.) respectively, as previously reported.10 Signals were detected by the chemiluminescence and quantified by densitometry.
Adult WT and MKP1−/− mice (male, 10 weeks old) were treated with LPS (10 mg/kg, i.p.) or saline. After 4 h, mice were heparinized (5000 U/kg, i.p.) and sacrificed by cervical dislocation under ketamine (50 mg/kg, i.p.) and xylazine (12.5 mg/kg, i.p.) anaesthesia. The adequacy of anaesthesia was monitored by the absence of withdrawal reflex to tail pinch. Hearts were isolated and perfused in a Langendorff system with Kreb's-Henseleit buffer at 3 mL/min constant flow. The Kreb's-Henseleit buffer was kept at 37°C and consistently bubbled with a mixture of 95% O2 and 5% CO2. Myocardial function was detected as previously described.12
Adult male WT and MKP1−/− mice (8–10 weeks old) were treated with LPS (1.5 mg/kg, i.p.) or saline. After 4 and 24 h, mice were anaesthetized with 0.5–1% halothane USP inhalation in a mixture of 95% O2 and 5% CO2. Echocardiography (Sonos 5500, Hewlett-Packard, Andover, MA, USA) was performed as previously described.33,34 A 15 MHz paediatric probe (Agilent Technologies, Santa Clara, CA, USA) was used and M-mode data were captured for each animal. Parameters were determined using the American Society for Echocardiography leading-edge technique in a blinded fashion.33,34
Results are presented as mean ± SEM. Differences between two groups were analysed by a standard Student t-test. For multi-group comparisons, one or two-way ANOVA followed by Student–Newman–Keuls or Bonferroni post-test was performed. P< 0.05 was considered statistically significant.
To examine whether LPS regulates myocardial MKP1 expression, we first measured MKP1 protein levels in cultured cardiomyocytes and found that the MKP1 protein was increased after 1 h of LPS stimulation (Figure 1A). To verify these in vitro results, WT mice were treated with LPS or saline for 1 and 1.5 h. Hearts were then harvested and myocardial MKP1 mRNA and protein levels were detected. Myocardial MKP1 mRNA and protein levels were markedly increased after LPS stimulation (Figure 1B and C). These data show that LPS promotes MKP1 expression in cardiomyocytes in vitro and in the myocardium in vivo.
MKP1 inactivates MAPKs by dephosphorylating phosphothreonine and phosphotyrosine residues. To determine the role of MKP1 in cardiac MAPK inactivation, we examined myocardial ERK1/2 and p38 phosphorylation in WT and MKP1−/− mice after LPS (4 mg/kg, i.p.) treatment for 1, 1.5, 2, 2.5, and 3 h. ERK1/2 and p38 phosphorylation was induced by LPS in both WT and MKP1−/− hearts (Figure 2A and B). Phosphorylation of ERK1/2 was diminished in the WT myocardium between 2 and 3 h after LPS stimulation, but was sustained in the MKP1−/− myocardium (Figure 2A). Similarly, myocardial p38 phosphorylation peaked at 1.5 h and then declined between 2 and 3 h after LPS stimulation. Importantly, p38 phosphorylation was significantly higher in MKP1−/− than the WT myocardium (Figure 2B). Thus, MKP1 deficiency enhanced ERK1/2 and p38 phosphorylation, demonstrating an important role for MKP1 in reducing the levels of myocardial ERK1/2 and p38 activation after LPS stimulation.
To investigate the role of MKP1 in myocardial TNF-α expression during endotoxemia in vivo, WT and MKP1−/− mice were treated with vehicle or LPS. Our data showed that LPS increased myocardial TNF-α expression in both WT and MKP1−/− mice. Compared with the WT group, MKP1−/− mice exhibited significantly higher TNF-α mRNA and protein levels (Figure 2C and D). In addition, neonatal cardiomyocytes were infected with Ad-MKP1. Consistently, MKP1 overexpression in neonatal cardiomyocytes inhibited LPS-induced TNF-α mRNA and protein levels (Figure 2E and F). These results suggest that MKP1 limits myocardial TNF-α expression in response to LPS.
Cardiac function was determined using the Langendorff preparation after 4 h of LPS in vivo treatment. The rates of contraction (+dF/dtmax) and relaxation (−dF/dtmin), and heart work were reduced in both WT and MKP1−/− mice relative to vehicle-treated mice (Figure 3). Furthermore, compared with WT mice, rate of contraction and relaxation and heart work were significantly lower in MKP1−/− mice (Figure 3A–C). To confirm these ex vivo data, in vivo cardiac function was determined using echocardiography in WT and MKP1−/− mice at baseline and at 4 and 24 h after LPS treatment. Basal LV dimension, wall thickness, and heart function showed no difference between MKP1−/− and WT mice (Table 1). In response to LPS, cardiac output and fractional shortening showed a trend but not significant decrease in WT mice (Figure 3E and F). However, fractional shortening was significantly decreased in MKP1−/− mice after LPS stimulation (P< 0.05). Furthermore, both cardiac output and fractional shortening were significantly decreased after LPS treatment in MKP1−/− compared with WT mice (P< 0.05). Taken together, these data indicate that MKP1 expression improves cardiac function during endotoxemia.
To assess PAK1 activation, cultured cardiomyocytes from WT mice were treated with LPS (1 μg/mL) for 7, 15, 30, and 60 min. PAK1 phosphorylation in these cells peaked at 15 min and then returned to control levels at 60 min (Figure 4A), indicating that LPS activates PAK1 in cardiomyocytes.
To elucidate the role of PAK1 in LPS-induced TNF-α expression, cardiomyocytes were treated with IPA-3, an inhibitor of PAKs. IPA-3 (1–10 μM) decreased LPS-induced TNF-α mRNA and protein levels in a dose-dependent manner (Figure 4B and C). These results were further verified by using PAK1 siRNA. As shown in Figure 4D, PAK1 mRNA expression was decreased by 56% after PAK1 siRNA treatment. The inhibition of PAK1 expression was associated with decreased TNF-α mRNA and protein levels (Figure 4E and F). These data suggest that the activation of PAK1 promotes cardiac TNF-α expression in response to LPS.
PAK1 is activated by small GTPases, such as Rac.26 We have demonstrated that LPS activates the PI3K/Rac1 pathway, which is critical for cardiac TNF-α expression during endotoxemia.12 To determine whether PI3K acts as an upstream regulator of PAK1, the PI3K inhibitor, LY294002, was employed. LY294002 significantly decreased PAK1 phosphorylation induced by LPS (Figure 5A). To determine the role of Rac1 on PAK1 activity, cardiomyocytes were treated with an adenovirus encoding a dominant negative form of the Rac1 gene (Ad-Rac1N17), which specifically inhibits Rac1 activity. Overexpression of Rac1N17 inhibited PAK1 activity in response to LPS (Figure 5B). Taken together, these data imply that LPS activates PAK1 via PI3K/Rac1 in cardiomyocytes.
To determine whether PAK1 regulates p38, ERK1/2, and JNK phosphorylation, cultured cardiomyocytes were treated with PAK1 siRNA. Interestingly, PAK1 siRNA blocked p38 and JNK phosphorylation in response to LPS, but had no significant effect on ERK1/2 phosphorylation (Figure 5C–E).
To further explore the downstream effectors of the Rac1/PAK1/JNK pathway, MKP1 expression was studied. Overexpression of Rac1N17 in cardiomyocytes significantly decreased MKP1 protein levels (Figure 6A). This result was further confirmed in vivo. As shown in Figure 6B, LPS enhanced MKP1 protein levels in Rac1f/f but not in Rac1CKO hearts. To examine whether PAK1 mediates MKP1 expression, cultured cardiomyocytes were treated with IPA-3 or PAK1 siRNA. LPS-induced MKP1 protein expression was blocked by either IPA-3 or PAK1 siRNA (Figure 6C and D). To determine whether JNK regulates MKP1 expression, cardiomyocytes were treated with SP600125, an inhibitor of JNK. LPS-induced MKP1 protein expression was significantly reduced by SP600125 (Figure 6E). To verify this result in vivo, WT and JNK1−/− mice were treated with LPS (4 mg/kg, i.p.) or saline for 1.5 h. LPS significantly increased myocardial MKP1 protein levels in WT mice but had no apparent effect in JNK1−/− mice (P< 0.05, Figure 6F). Taken together, these data indicate that the Rac1/PAK1/JNK pathway mediates LPS-induced MKP1 expression.
The present study demonstrated for the first time that myocardial MKP1 is induced by LPS via the PI3K/Rac1/PAK1/JNK pathway. Enhanced MKP1 expression attenuates ERK1/2 and p38 phosphorylation, leading to inhibition of myocardial TNF-α expression and improvement of cardiac function during endotoxemia (Figure 6G).
MKP1 dephosphorylates MAPKs and impedes their cellular functions. Recent studies have shown that MKP1 expression is up-regulated in macrophages in response to LPS.13 Deficiency of MKP1 enhances p38 and JNK phosphorylation and pro-inflammatory cytokine production, including TNF-α, IL-1β, and IL-6 in macrophages.13,14 Consistent with this notion, MKP1−/− mice exhibit significantly higher serum cytokine concentrations and higher mortality rate after LPS stimulation.14–16 In the present study, we showed that LPS induced myocardial MKP1 expression. Furthermore, overexpression of MKP1 decreased TNF-α expression induced by LPS. MKP1 deficiency enhanced myocardial ERK1/2 and p38 phosphorylation and increased cardiac TNF-α expression. Most importantly, MKP1 deficiency further decreased cardiac function during endotoxemia. Thus, both in vitro and in vivo evidence demonstrated that LPS induces MKP1 expression, which inhibits ERK1/2 and p38 activation and myocardial TNF-α expression, and improves cardiac function during endotoxemia.
Rac is one of the factors that are responsible for MKP1 expression.23 We recently demonstrated that PI3K-mediated Rac1 activation promotes TNF-α expression and cardiac dysfunction in endotoxemia.12 PAK proteins are downstream effectors of Rac and regulate many cellular events, including cell motility, survival, proliferation, and gene expression.26 PAK1 is the main isoform of this enzyme in the myocardium and plays an important role in cardiac contractility and hypertrophy.35,36 Studies have shown that the PAK protein is activated and contributes to IL-1 expression through MAPKs upon ligand stimulation in macrophages.27 In the present study, LPS enhanced PAK1 activity in cardiomyocytes. Inhibition of PAK1 activity by the pharmacological inhibitor, IPA-3, or PAK1 expression by PAK1 siRNA decreased LPS-induced TNF-α expression in cardiomyocytes. Furthermore, inhibition of PI3K and Rac1 decreased PAK1 activity in response to LPS in cardiomyocytes. These results indicate that PI3K/Rac1-mediated PAK1 activation plays a key role in LPS-induced TNF-α expression in cardiomyocytes.
ERK1/2, p38, and JNK MAPKs are important signalling molecules regulating TNF-α expression in cardiomyocytes.6–10 ERK1/2 and p38 are positive regulators of TNF-α expression.6–8,10 In contrast, JNK1 inhibits TNF-α expression via inactivating LPS-induced ERK1/2 and p38 MAPKs.9 In the present study, inhibition of PAK1 activity blocked LPS-induced p38 and JNK phosphorylation. These results suggest that PAK1 may have both stimulatory and inhibitory effects on LPS-induced TNF-α expression in cardiomyocytes. Its activation on p38 promotes TNF-α expression. On the other hand, PAK1-mediated JNK activation inhibits LPS-induced TNF-α expression.
To address the molecular mechanisms by which JNK inhibits LPS-induced TNF-α expression, the role of MKP1 was examined in the present investigation. It has been shown that MAPKs are involved in regulating MKP1 expression. For example, ERK1/2 and JNK activation is responsible for MKP1 induction in fibroblasts.20,21 However, the effects of ERK1/2, p38, and JNK on MKP1 expression in macrophages are controversial. Several studies have shown that ERK1/2 and p38 are required for MKP1 expression in macrophages. 13,24,25 On the other hand, Sanchez-Tillo et al.22 reported that JNK1 is necessary for MKP1 expression in macrophages which consequently decreases ERK1/2 and p38 phosphorylation levels. In the present study, we demonstrated that LPS-induced MKP1 expression was blocked by the inhibition of Rac1, PAK1, and JNK1 in cardiomyocytes, suggesting that the Rac1/PAK1/JNK1 pathway is required for LPS-induced MKP1 expression in cardiomyocytes. MKP1 mediates the inhibitory effect of JNK1 on ERK1/2 and p38 activity. Therefore, Rac1/PAK1/JNK1-mediated MKP1 expression provides a key negative feedback mechanism to limit myocardial TNF-α production and improve cardiac function during endotoxemia. These results do not exclude other pathways through which Rac1 promotes MKP1 expression. For example, we have previously shown that Rac1 activation leads to TNF-α production through reactive oxygen species (ROS), which is also a regulator of MKP1.12,37 Thus, the Rac1/ROS pathway could be an alternative mechanism responsible for MKP1 expression.
In conclusion, the present study showed that PAK1 is activated by the PI3K/Rac1 pathway and increases p38 and JNK activity. PAK1 increases TNF-α expression via p38 activation in response to LPS. On the other hand, the Rac1/PAK1/JNK1 pathway enhances MKP1 expression, which inactivates ERK1/2 and p38, limits TNF-α expression, and improves cardiac function during endotoxemia (Figure 6G). Thus, the Rac1/PAK1/JNK1/MKP1 signalling pathway represents a novel negative feedback mechanism in regulating TNF-α expression and cardiac function in endotoxemia. Our study suggests that MKP1 may have therapeutic potential in the clinical treatment of sepsis. Further studies are required to assess its efficacy and safety in large animal models and in patients with sepsis.
This study was supported by operating grants from the Canadian Institutes of Health Research (MOP-14653 to Q.F.) and the National Institutes of Health (R01AI068956 to Y.L. and Q.F.). T.Z. was supported by an Ontario Graduate Scholarship. Q. F. is a Career Investigator of the Heart and Stroke Foundation of Ontario.
Conflict of interest: none declared.