In Vivo Expression of Proinflammatory Mediators in the Adult Heart after Endotoxin Administration: The Role of Toll-Like Receptor–4

Abstract

Tumor necrosis factor (TNF)–α, interleukin (IL)–1β, and nitric oxide (NO) may play a role in lipopolysaccharide (LPS)–induced cardiac depression. Toll-like receptor–4 (TLR-4) mediates the cytokine response to LPS in immune cells. TLR-4 also is expressed in human and murine myocardial tissue. Therefore, the hypothesis that LPS induces proinflammatory cytokines in the heart via TLR-4 was tested. C3H/HeJ (TLR-4 deficient) and C3HeB/FeJ mice were studied. LPS induced a robust increase in myocardial TNF-α and IL-1β mRNA in C3HeB/FeJ mice. The response in C3H/HeJ mice was blunted and delayed. Myocardial TNF-α and IL-1β protein levels were higher in C3HeB/FeJ mice, as were inducible NO synthase protein and NO production. Activation of myocardial NF-κB was observed within 30 min in C3HeB/FeJ mice but not in C3H/HeJ mice. These findings suggest that myocardial TLR-4 is involved in signaling cytokine production within the heart during endotoxic shock

Human sepsis is responsible for >20,000 deaths per year in the United States, and these deaths are caused frequently by shock, which results in multiorgan dysfunction [1]. Myocardial depression is a common and potentially fatal complication of septic shock. Indeed, Parker et al. [2] demonstrated that 40% of patients with sepsis develop both left and right ventricular dysfunction, with a depression in ejection fraction and dilation of the ventricles. Furthermore, in patients who develop cardiovascular impairment, mortality increases from 20% to 70%–90%, which demonstrates a correlation between cardiac function and survival during sepsis [3]. Although the precise endogenous mediators that cause myocardial dysfunction during sepsis remain elusive, animal models [3–5] and in vitro studies strongly support the hypothesis that tumor necrosis factor (TNF)–α, interleukin (IL)–1β, and nitric oxide (NO) mediate the myocardial depression associated with septic shock caused by gram-negative bacteria [6–8]

Kumar et al. [4] were the first to report that the myocardial depressant activity of human serum from patients with septic shock could be eliminated by the immunoprecipitation of TNF-α and IL-1β. Giroir et al. [9] reported that myocardial TNF-α mRNA expression increased after lipopolysaccharide (LPS) administration. Kapadia et al. [10] subsequently demonstrated that cardiac myocytes themselves produce significant amounts of TNF-α after endotoxemia. These findings suggested that the compartmentalized production of TNF-α and other cytokines might play an important role in the pathogenesis of LPS-induced myocardial depression in vivo. Supporting this hypothesis is the finding that the administration of TNF binding proteins preserves myocardial function in endotoxemic rats [11]. To date, however, the signaling pathways that lead to the expression of these proinflammatory mediators in the heart during gram-negative sepsis remain undefined

A recent advance in unraveling the early events in LPS signaling has been the identification of Toll-like receptors (TLRs). Toll is a transmembrane receptor in Drosophila species that is involved in dorsal-ventral patterning in the embryo and in the induction of an antifungal response in the adult fly [12]. Toll and TLRs activate homologous signal transduction pathways that lead to nuclear localization of NF-κB/Rel–type transcription factors. Several lines of evidence suggest that ⩾1 members of the TLR family could be a cell-surface receptor for LPS

Two groups independently reported that overexpression of TLR-2 in human embryonic kidney 293 cells resulted in a cell that responded to LPS by inducing NF-κB activation, thus suggesting that TLR-2 might be the LPS receptor [13, 14]. However, Poltorak et al. [15] and Qureshi et al. [16] later mapped the lps gene, which is responsible for the LPS hyporesponsive phenotype of the C3H/HeJ mouse, to TLR-4. It was found that C3H/HeJ-derived TLR-4 carries a point mutation in the intracytoplasmic region, which leads to the replacement of proline with histidine at codon 712. This proline is highly conserved among TLR family members, which indicates its important role. Despite normal Tlr4 mRNA levels in the C3H/HeJ strain, this mutation could abolish responses to LPS [16]. The importance of TLR-4 as the LPS signal transducer was further strengthened by the observations that the LPS-resistant C57BL/10ScCR strain fails to express Tlr4 mRNA because of a chromosomal deletion of the gene [16] and that genetically engineered TLR-4 knockout mice also were LPS resistant [17]. This genetic evidence strongly suggests that TLR-4 is the primary receptor for LPS

In considering the potential central role that TNF-α, IL-1β, and NO might play in LPS-induced myocardial dysfunction and the recent identification of TLR-4 as the LPS receptor, we sought to determine whether TLR-4 is critical to the cardiac response to LPS during endotoxic shock in vivo

Material and Methods

Animal protocolInbred strains of C3H/HeJ (TLR-4 deficient) and C3HeB/FeJ mice were obtained from Jackson Laboratory. The C3H/HeJ strain is hyporesponsive to LPS because of a point mutation in the TLR-4 gene [15, 16]. The C3HeB/FeJ strain is highly sensitive to the effects of LPS [18]. Mice were maintained in microisolator cages and were fed pellet food and water ad libitum. Both groups of animals were injected intraperitoneally with 25 mg/kg Escherichia coli LPS (200 μL) containing 0.075 mg/mL (∼0.02 mg/mouse) of LPS-associated proteins (a phenol extract of serotype 0111:B4; Sigma Chemical) or pyrogen-free PBS (Gibco)

Mice were killed with a lethal injection of sodium pentobarbital (80 mg/kg), and their hearts were harvested 0.5, 1, 2, 4, 8, and 12 h after LPS challenge for use in determining TNF-α and IL-1β gene and protein expression. Blood samples were obtained at the same time intervals for TNF-α measurements. For determination of inducible NO synthase (iNOS) activity and NO production, animals were killed 0, 4, 8, and 12 h after LPS challenge. Hearts were harvested 30 and 60 min after LPS challenge for the assessment of NF-κB activation. Hearts were routinely snap-frozen in liquid nitrogen and were kept at −70°C for later use

Cloning of the murine TLR-4 cDNATotal RNA was isolated as described below. The TLR-4 cDNA was amplified by reverse transcription (RT)–polymerase chain reaction (PCR), using RNA from mouse heart tissue (C3HeB/FeJ). For amplification of the gene-specific fragment, we used a full-length murine sequence for TLR-4 from the National Center for Biotechnology Information database of Expressed Sequence Tags to design primers (GenBank accession no. AF 110133). The sense primer for murine TLR-4 (5′-GCTTACACCACCTCTCAAACTTGAT-3′) was derived from bases 317–341 of the coding region, and the antisense primer (5′-ATTACCTCTTAGAGTCAGTTCATGG-3′) was derived from bases 702–726. The RT reaction was done using the Titan One Tube RT-PCR kit (Boehringer Mannheim), followed by PCR amplification with Taq polymerase. The amplified fragment of the desired length (410 bp) was subcloned into a pSTB1 vector, using the Perfectly Blunt cloning kit (Novagen), according to the manufacturer’s protocol. After the authenticity of the clones was confirmed by automated sequencing, the plasmid was digested with PVU 2 (Promega). The expected fragment size was confirmed by gel electrophoresis

Ribonuclease protection assayTNF-α and IL-1β gene expression was determined by use of a custom-designed multiprobe RNase protection assay system (RiboQuant; Pharmingen), which was used according to the manufacturer’s protocol. The hearts were homogenized, and total RNA was extracted by the guanidinium thiocyanate method [19]. A radiolabeled antisense RNA probe, which was set for TLR-4, TLR-2, TNF-α, IL-1β, and the large ribosomal protein L32, was synthesized. The final reaction mixture contained 10 μL of [γ-³²P] UTP (20 mCi/mL; Amersham Pharmacia Biotech), 1 μL each of GTP, ATP, CTP, and UTP (2.75 mmol each), 2 μL of dithiothreitol (DTT; 100 mmol), 4 μL of transcription buffer (1×), 1 μL of RNasin (40 U), 1 μL of T7 polymerase (20 U), and an equimolar pool of linearized templates (50 ng total)

After 1 h at 37°C, the reaction mixture was treated with 2 μL of RNase-free DNase (2 U) for 30 min at 37°C. The probe was purified by extraction with acid phenol–chloroform, precipitated with 100% ethanol and stored for 30 min at −70°C. After centrifugation (14000 g) for 15 min at 4°C, the supernatant was discarded, and the probe was washed with 90% ethanol and was dried at room temperature for 5 min. The pellet was dissolved in 50 μL of hybridization buffer (1×), and the radioactivity was quantitated. The probe then was diluted to the appropriate concentration, according to the manufacturer’s protocol, and 2 μL was added to the tubes containing 20 μg of total RNA. After hybridization, the unprotected RNA was digested with 100 μL of an RNase A and RNase T1 mix. The protected RNA fragments were isolated by extraction and precipitation and were electrophoresed in a 6% polyacrylamide gel (National Diagnostic). The gels were vacuum-dried, were exposed overnight to imaging plates, and were scanned with a phosphoimager (Storm 860; Molecular Dynamics). Signals were quantified by use of software (Image QuaNT; Molecular Dynamics) and were normalized to L32

TLR-4 Western blot analysisTo determine whether TLR-4 protein was expressed in the hearts of the adult mice, we examined TLR-4 protein levels by Western blot analysis. Hearts were homogenized in 2 mL of ice-cold extraction buffer containing 20 mM HEPES (pH 7.4), 20 mM β-glycerophosphate, 20 mM sodium pyrophosphate, 0.2 mM Na₃VO₄, 2 mM EDTA, 20 mM sodium fluoride, 10 mM benzamidine, 1 mM DTT, 20 ng/mL leupeptin, 0.4 mM Pefabloc SC, a serine protease inhibitor (Sigma), and 0.05% Triton X. The homogenate was centrifuged at 14,000 g for 15 min at 4°C. The supernatant was collected, and the protein concentration was determined by use of a bicinchoninic acid assay kit (Pierce, Life Science), with bovine serum albumin as a standard

Protein (150 μg/lane) was separated on 7.5% SDS–polyacrylamide gel under denaturing conditions and was electroblotted to a nitrocellulose membrane (BioRad). The membrane was incubated at 4°C overnight in 5% nonfat milk in Tris-buffered saline containing 0.05% Tween-20 (TBST) and then was immunoblotted for 1 h with a monoclonal antibody (MAb) directed against human TLR-4 (gift of K. Miyake, Saga Medical School, Saga, Japan). This MAb has been characterized elsewhere [20]. The secondary antibody was horseradish peroxidase–conjugated goat anti–mouse polyclonal antibody. Myocardial TLR-4 expression was detected with a Western blotting detection kit (ECL-Plus; Amersham)

Myocardial TNF-α and IL-1β ELISAHearts were harvested at specific times after LPS administration, and homogenates were prepared as described above. Intramyocardial TNF-α and IL-1β protein levels were measured with an ELISA (R&D Systems) used according to the manufacturer’s instructions. All samples were assayed in duplicate. Data are expressed as picograms per milligram of protein

iNOS Western blot analysis and NO measurementsMyocardial proteins were isolated for detection of iNOS protein expression, as described above. Protein (50 μg/lane) was separated on 7.5% SDS–polyacrylamide gel under denaturing conditions and was electroblotted to nitrocellulose membrane (BioRad). After incubation in 5% nonfat milk in TBST at 4°C overnight, the membrane was immunoblotted for 1 h with rabbit anti-iNOS antibody (Santa Cruz Biotechnology). The secondary antibody was horseradish peroxidase–conjugated goat anti-rabbit polyclonal antibody. iNOS protein expression was detected with a Western blotting detection kit (ECL-Plus; Amersham)

Myocardial NO production after LPS administration was quantitated by use of an NO kit (R&D Systems). This assay determines NO on the basis of the enzymatic conversion of nitrate to nitrite by nitrate reductase. To determine the nitrate concentration in each sample, the endogenous nitrite concentration was subtracted from the total nitrite concentration. The myocardial nitrate concentrations are expressed in nanomoles per milligram of protein

Myocardial nuclear extraction and electrophoretic mobility shift assayTo prepare nuclear protein extracts (crude), we minced the hearts and incubated them on ice for 30 min in ice-cold buffer A (10 mM HEPES [pH 7.9], 1.5 mM KCL, 10 mM MgCl₂, 0.5 mM DTT, 0.1 % Nonidet P-40, and 0.5 mM phenylmethylsulfonyl fluoride [PMSF]). The minced tissue was homogenized in a Dounce homogenizer with 1 mL of buffer A and was centrifuged at 5000 g for 10 min at 4°C. The crude nuclear pellet then was suspended in 300 μL of buffer C (20 mM HEPES-NaOH [pH 7.9], 25% glycerol, 420 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.5 mM PMSF, 0.5 mM DTT, and 4 mM leupeptin) and was incubated on ice for 30 min. Samples were centrifuged at 15000 g for 30 min, and supernatants (nuclear proteins) then were collected. The protein concentration was determined by use of a bicinchoninic acid assay (Pierce) kit, according to the manufacturer’s protocol

For the electrophoretic mobility shift assay, the NF-κB oligonucleotide (5′-AGTTGAGGGGACTTTCCCAGGC-3′ [Santa Cruz Biotechnology]) was end-labeled with [γ-³²P]ATP (Amersham). Binding reactions (25 μL total) were done by incubating 20 μg of nuclear extracts for 30 min at room temperature with 4 mM Tris-Cl (pH 7.9), 12 mM HEPES, 1 mM DTT, 60 mM KCL, 10% glycerol, 1 mM EDTA, 2 μg poly(dI-dC)-poly(dI-dC), and 20,000 cpm of the labeled NF-κB oligonucleotide. The specificity of the DNA-protein binding was determined by competition with a 50-fold molar excess of the unlabeled NF-κB oligonucleotide. For supershift assays, the nuclear extracts were incubated with 2 μg of polyclonal anti-p50 or anti-p65 antibody (Santa Cruz Biotechnology) before the addition of the labeled probe to the reaction mix. Nuclear extracts used in supershift and competition experiments were harvested 1 h after LPS challenge. The DNA-protein complexes were electrophoresed for 2 h at 30 mA in a 4% polyacrylamide gel in 0.5× Tris-borate-EDTA running buffer. The gels were dried for 1 h, were exposed overnight to imaging plates, and were scanned with a phosphoimager (Storm 860; Molecular Dynamics)

StatisticsAll values are expressed as mean±SE. Two-way analysis of variance followed by Bonferroni-corrected t tests was used to determine significant differences in LPS-induced cytokine gene and protein expression and NO production between C3HeB/FeJ and C3H/HeJ mice at each time point. Significant differences were considered to exist at P⩽.05

Results

Clinical manifestations of sepsis and survivalC3HeB/FeJ and C3H/HeJ (TLR-4 deficient) mice were challenged intraperitoneally with 25 mg/kg LPS and were monitored for lethal effects. About 2 h after LPS challenge, C3HeB/FeJ mice developed shock-like symptoms, including ruffled hair, diarrhea, eye exudates, and lethargy. In contrast, C3H/HeJ mice remained healthy throughout the observation period. Mortality at 24 h was 20% in C3HeB/FeJ mice (2/10) and 0% in C3H/HeJ mice (0/10). These findings are in agreement with those reported by other researchers [21] and confirm the LPS hyporesponsive phenotype of the C3H/HeJ strain and the LPS responsiveness of the C3HeB/FeJ strain

Role of TLR-4 in the induction of myocardial TNF-α and IL-1β after LPS challengePrevious studies have shown that LPS administration induces intramyocardial production of TNF-α [6, 7, 10]. To determine whether TLR-4 might play a role in the in vivo induction of myocardial proinflammatory cytokines, TNF-α and IL-1β gene and protein expression were measured in C3HeB/FeJ and C3H/HeJ mice after LPS administration. Figure 1A illustrates the time course of LPS-induced intramyocardial TNF-α and IL-1β mRNA expression in C3HeB/FeJ and C3H/HeJ mice. There was no evidence of TNF-α or IL-1β mRNA expression by RNA protection assay in the myocardium of PBS-treated animals. In striking contrast, LPS induced a robust increase in TNF-α and IL-1β mRNA transcripts (figure 1A 1B) in the hearts of C3HeB/FeJ mice. TNF-α mRNA up-regulation was maximal 30 min after LPS challenge, whereas peak IL-1β mRNA expression occurred 2 h after injection. LPS challenge also induced the expression of these 2 cytokine genes (figure 1A 1B) in the myocardium of C3H/HeJ mice; however, the magnitude of the response was significantly attenuated, compared with that in the LPS-responsive strain. A delay in gene expression after LPS challenge was noted only for TNF-α, with peak expression occurring 30 min earlier in the C3HeB/FeJ strain

Figure 1

Impaired lipopolysaccharide (LPS)–induced myocardial tumor necrosis factor (TNF)–α and interleukin (IL)–1β mRNA expression in C3H/HeJ mice. A C3HeB/FeJ and C3H/HeJ mice (n=7/time point) were injected intraperitoneally with Escherichia coli O111:B4 LPS (25 mg/kg body weight) or PBS. Hearts were harvested at the indicated time points, and expression of TNF-α and IL-1β was assessed by use of an RNase protection assay. A representative autoradiograph is shown. B and C Relative TNF-α and IL-1β mRNA levels after LPS challenge. Data are mean±SE TNF-α or IL-1β mRNA:L32 (large ribosomal mRNA) ratio. *P<.05

We also compared TNF-α and IL-1β protein levels in homogenates of myocardial tissue from mice after LPS injection. Figure 2 shows that the kinetics of TNF-α and IL-1β protein production correlated with the appearance of TNF-α and IL-1β mRNA transcripts in the heart after LPS administration. Although TNF-α and IL-1β protein were measured in myocardial tissue from both groups, the levels were significantly higher (P<.01) in C3HeB/FeJ mice 1 and 2 h after LPS challenge. That is, both myocardial TNF-α and IL-1β production increased by 80% (from baseline) within 0.5 h of LPS challenge in C3HeB/FeJ mice. In contrast, myocardial TNF-α and IL-1β protein increased by only 40% and 33% (from baseline), respectively, in C3H/HeJ mice. Similarly, serum TNF-α levels after LPS challenge were significantly higher in the C3HeB/FeJ strain (table 1). Despite the presence of TNF-α in the serum samples and myocardia of C3H/HeJ mice, these animals did not develop signs consistent with septic shock

Figure 2

Myocardial tumor necrosis factor (TNF)–α (A) and interleukin (IL)–1β (B) protein production after lipopolysaccharide (LPS) challenge. C3HeB/FeJ and C3H/HeJ mice (n=5/time point) were injected intraperitoneally with LPS. Hearts were harvested at the indicated time, and the production of TNF-α and IL-1β was measured by ELISA (R&D Systems). Data are mean±SE from 5 different animals at each time point. *P<.05. All samples were assayed in duplicate

Table 1

Serum tumor necrosis factor (TNF)–α levels in C3H/HeJ and C3HeB/FeJ mice after lipopolysaccharide challenge

LPS administration also induced the up-regulation of myocardial TLR-4 transcripts in both groups of animals (figure 3A). Despite the observed increase in TLR-4 mRNA, no significant changes in TLR-4 protein expression were detected in myocardial tissue (figure 3B). These findings are in agreement with those reported by Frantz et al. [22] in neonatal rat ventricular myocytes and coronary endothelial cells stimulated with LPS in vitro

Figure 3

Time course of mouse Toll-like receptor–4 (TLR-4) expression in response to lipopolysaccharide (LPS) challenge. A C3HeB/FeJ and C3H/HeJ mice (n=5/time point) were challenged with LPS (25 mg/kg), and hearts were harvested at the indicated times. TLR-4 expression was assessed by RNase protection assay. Data have been normalized to large ribosomal (L32) mRNA content. B Western blot analysis of myocardial TLR-4 protein levels during endotoxemia. Proteins (150 μg of whole-heart homogenate) were separated on an SDS–7.5% polyacrylamide gel and were transferred to nitrocellulose membrane. For the detection of TLR-4 protein, a mouse anti–human TLR-4 monoclonal antibody (HTA125) was used. The blot is representative of 3 independent experiments (mean±SE)

Induction of myocardial iNOS and NO production during endotoxic shockEndogenous release of TNF-α and IL-1β in response to LPS is responsible for the expression of the Ca⁺²-independent, inducible isoform of NO synthase in a variety of cells and tissue in vivo [8]. Enhanced formation of NO by iNOS also has been implicated in the cardiovascular alterations associated with septic shock [23]. We therefore determined the level of expression of myocardial iNOS protein after LPS challenge in both groups of mice. Western blot analysis (figure 4A) showed an increase in iNOS protein levels in myocardial homogenates from C3HeB/FeJ mice 8 and 12 h after LPS provocation. In striking contrast, this increase in iNOS expression was not observed in myocardial tissues from C3H/HeJ mice. Figure 4B also shows that NO production was significantly greater 12 h after LPS challenge in myocardial samples from C3HeB/FeJ mice

Figure 4

Myocardial inducible nitric oxide (NO) synthase (iNOS) protein expression after lipopolysaccharide (LPS) injection. A Western blot analysis of myocardial iNOS protein levels during endotoxemia. Proteins (50 μg of whole-heart homogenate) were separated on an SDS–7.5% polyacrylamide gel and were transferred to nitrocellulose membrane. For the detection of mouse iNOS protein, a rabbit anti–mouse iNOS antibody was used. The blot is representative of 3 independent experiments. B Myocardial NO production after LPS administration. Hearts were harvested at the indicated times, and homogenates were assayed for nitrate content, as described in Materials and Methods. Data are mean±SE (n=5–8 animals/time point). *P<.05

Myocardial NF-κB activation after LPS administrationThe transcription factor NF-κB signals the activation of several proinflammatory cytokine genes whose products are thought to play an important role in the pathogenesis of LPS-induced myocardial dysfunction [24, 25]. We therefore studied the in vivo effect of TLR-4 on LPS-induced myocardial NF-κB activation. Figure 5 shows the time course of LPS-induced myocardial NF-κB binding activity in 3 animals per time point. NF-κB–DNA binding activity was detectable at low levels in myocardial extracts from untreated mice. However, a robust increase in myocardial NF-κB–DNA binding activity was observed in C3HeB/FeJ mice within 30 min of LPS provocation. In contrast, the NF-κB response to LPS was delayed and attenuated in myocardial extracts from C3H/HeJ mice. This attenuation paralleled the retardation in myocardial TNF-α and IL-1β expression in the C3H/HeJ strain. In both groups, the composition of the LPS-activated NF-κB complex consisted predominantly of p65/p50 heterodimers, as determined by supershift assays. The specificity of the NF-κB–DNA complexes were confirmed by displacement of the NF-κB complex in the presence of 50-fold excess unlabeled NF-κB oligonucleotide. Thus, a defective myocardial TLR-4 results in decreased NF-κB activation after in vivo LPS administration

Figure 5

Myocardial NF-κB activation is impaired in lipopolysaccharide (LPS)–challenged C3H/HeJ mice. Crude nuclear extracts were prepared from myocardial tissue isolated from C3HeB/FeJ and C3H/HeJ mice at the indicated times (minutes) after LPS challenge. Nuclear extracts were tested for their ability to retard the electrophoretic mobility of a ³²P-radiolabeled oligonucleotide containing a NF-κB binding site. Nuclear extracts used for supershift and competition experiments were harvested 1 h after LPS challenge. Results are representative of 3 independent experiments

Discussion

The results of these experiments suggest that myocardial TLR-4 signaling is important in the induction of the proinflammatory response in the heart during endotoxic shock. Two sets of observations support this conclusion. First, we demonstrated that the in vivo induction of intramyocardial TNF-α and IL-1β gene expression in C3H/HeJ mice was significantly attenuated, compared with that in C3HeB/FeJ mice, after LPS provocation (figure 1A). In addition, LPS-induced myocardial TNF-α and IL-1β protein expression was significantly lower in C3H/HeJ mice. We also have demonstrated significantly lower intramyocardial iNOS expression and NO production in C3H/HeJ mice after LPS challenge. Second, after LPS administration, the activation of myocardial NF-κB, an important transcription factor in proinflammatory cytokine production, was impaired in C3H/HeJ mice. Liu et al. [24, 25] recently demonstrated that activation of NF-κB is a critical in vivo regulatory mechanism mediating LPS-induced TNF-α and iNOS expression in various organs, including the heart. Our observations extend these findings and establish an in vivo link between myocardial TLR-4, NF-κB activation, and the expression of proinflammatory mediators implicated in the pathogenesis of sepsis-induced myocardial depression

A second potentially important observation of this study was that LPS-induced proinflammatory cytokine expression was blunted rather than abrogated in the hearts of the C3H/HeJ mice. Of interest, we recently observed that C3H/HeJ mice are protected from LPS-induced myocardial depression in a high-dose model of endotoxic shock (J.G.V., unpublished data). Thus, although inflammatory mediators are expressed in the heart of C3H/HeJ mice, the amount is not sufficient to depress myocardial function. These findings also suggest that LPS-induced signaling in the heart may be mediated through ⩾1 alternative receptors

Indeed, previous studies have documented that C3H/HeJ mice manifest LPS-induced gene and protein expression in various tissues, which suggests that redundancy in LPS signaling exists [25, 26]. Nill et al. [26] compared the effect of LPS administration on TNF-α and IL-1α gene expression in lung tissue of C3H/HeN (an LPS-responsive strain) and C3H/HeJ mice: 15–30 min after challenge, the magnitude of LPS-induced TNF-α mRNA expression was 40% less in C3H/HeJ than in C3H/HeN mice. However, 1–2 h after LPS challenge, the levels of TNF-α mRNA in lung tissue from C3H/HeN and C3H/HeJ mice were comparable. In contrast, we noted significant differences in myocardial TNF-α mRNA expression when we compared C3HeB/FeJ with C3H/HeJ mice as long as 4 h after LPS challenge

Evans et al. [27] also reported that levels of hepatic NO synthase were greater in C3H/HeN than in C3H/HeJ mice after intravenous challenge with LPS. These latter findings are similar to what we observed in myocardial iNOS expression. Thus, previous studies and the data presented here underscore the fact that tissue microenvironments may react uniquely to LPS administration in vivo

Although the nature of the additional LPS-responsive receptor(s) in the heart remains speculative for the present, it is possible that TLR-2 and/or CD14 may play a role. Comstock et al. [28] have shown that CD14 is expressed in rat myocardium and that this receptor is necessary for the induction of TNF-α by LPS in cardiac myocytes. Although the role of TLR-2 in LPS signaling remains to be defined, Lien et al. [29] have suggested that TLR-2 may be a low-affinity LPS receptor. Indeed, Underhill et al. [30] have shown that a MAb directed against TLR-2 inhibits LPS-induced IL-12 production in human monocytes and that a dominant negative mutant of TLR-2 inhibits LPS induction of iNOS promoter activity in RAW 264.7 cells. In parallel studies, we observed that TLR-2 expression is up-regulated in the murine heart after LPS challenge (J.G.V., unpublished data)

The results reported here provide compelling evidence for the presence of a functional LPS-responsive TLR-4 system in the adult mammalian heart. Apart from the novelty of these findings, the results of the present study may be important for several reasons. As one example, the observation that LPS-induced cytokine and NO expression was significantly delayed or blunted in the hearts of mice deficient in TLR-4 signaling suggests that the TLR-4 receptor pathway may be a novel therapeutic target in clinical settings, such as gram-negative sepsis [31], or after certain forms of cardiac surgery [32–35], in which excess endotoxin has been implicated in the pathogenesis of myocardial depression

Although the above discussion focused attention on the potential deleterious effects of TLR-4 in the heart, the observation that TLRs have been highly conserved in both vertebrate and invertebrate species suggests that the TLR system may, in some way, provide a selective survival benefit to the host. Janeway et al. [36] hypothesized that pattern-recognition receptors can interact with microbial structures and elicit a danger signal to the host cell. Indeed, an increasing body of evidence suggests that TLR proteins may serve this purpose. Although the potential salutary effects of the TLR-4 pathway in the heart remain largely unknown, it has been suggested recently that TLR-4, and perhaps other TLRs, may contribute to the activation of an innate immune response in injured cardiac tissue [22]. Given our observation that TLR-4 is tightly coupled to NF-κB activation and proinflammatory cytokine expression in the heart and that both NF-κB and proinflammatory cytokines (physiologic levels) are cytoprotective in the heart [36–38], it will be important in future studies not only to delineate the full spectrum of TLRs and ligands in the heart, but also to elucidate the adaptive and maladaptive signaling pathways that are downstream from TLR-mediated signaling in the adult heart

Acknowledgments

We thank Carol J. Baker and Sheldon L. Kaplan for critically reviewing the manuscript and Claire M. Skeeter (Baylor College of Medicine, Houston) for technical expertise. The authors are indebted to Carol J. Baker for her past and present support

Presented in part: 73d annual meeting of the American Heart Association, New Orleans, 13 November 2000 (abstract 354)

All experiments were done in compliance with guidelines of the Institutional Animal Care and Use Committee and with the National Institutes of Health regulations on animal handling and usage

Financial support: Minority Medical Faculty Development Program of the Robert Wood Johnson Foundation (029212 to J.G.V.); Deutsche Herzstiftung (to G.B.); Deutsche Forschungsgemeinschaft (KN521/1-1 to P.K.)