Induction of Nitric Oxide Synthase Activity by Cytokines in Ventricular Myocytes Is Necessary but Not Sufficient to Decrease Contractile Responsiveness to β-Adrenergic Agonists
Abstract Recent evidence has documented that increased activity of an inducible nitric oxide synthase (iNOS; type 2 NO synthase) in primary isolates of adult rat ventricular myocytes after exposure to soluble mediators in medium conditioned by lipopolysaccharide-activated macrophages is associated with a decrease in their contractile responsiveness to β-adrenergic agonists. It remained unclear which specific inflammatory cytokines in this medium contribute to the induction of iNOS activity in myocytes and whether induction of iNOS would result in an obligatory decline in contractile function. Interleukin (IL)-1β and tumor necrosis factor-α (TNF-α) were both present in the lipopolysaccharide-activated macrophage-conditioned medium. However, only IL-1 receptor antagonist and not an anti–rat TNF-α antiserum diminished the extent of iNOS induction in myocytes exposed to this medium and prevented a decline in contractile responsiveness to isoproterenol. When recombinant cytokines were used, IL-1β, TNF-α, and IFN-γ each induced iNOS activity in cardiac myocytes at 24 hours. However, only the combination of IL-1β and IFN-γ reproducibly caused contractile dysfunction in cardiac myocytes. Among the constituents of the defined medium routinely used for maintenance of adult rat ventricular myocytes in primary culture, it was noted that insulin (10−7 mol/L) was required for NO production, as detected by nitrite release in cytokine-pretreated myocytes, although insulin had no effect on the extent of induction of iNOS mRNA or maximal enzyme activity in myocyte cell lysates. Insulin was also required for the decrease in contractile responsiveness to isoproterenol to be manifest. Thus, induction of iNOS is necessary but not sufficient to cause inflammatory cytokine-induced contractile dysfunction in cardiac myocytes.
Inappropriate or excessive activation of the immune system has long been implicated in the pathogenesis of contractile dysfunction accompanying myocarditis, cardiac allograft rejection, systemic sepsis, and some forms of idiopathic dilated cardiomyopathy, even in the absence of inflammatory cellular infiltrates and myocyte necrosis. Several studies1 2 3 4 5 have documented that inflammatory mediators in cell-free extracts obtained from the sera of patients with septic shock or from culture medium conditioned by activated immunocytes can depress the contractile function of cardiac myocytes in vitro. Indeed, the presence of a “myocardial depressant factor” in the sera of septic patients and experimental animals had been detected more than 20 years ago by Lefer and Rovetto,6 although the chemical identity of this substance has not been clarified.
The recognition that inflammatory cytokines play an important role in the pathogenesis of many of the clinical signs of septic shock, as well as the recent availability of recombinant cytokines and cytokine antagonists for experimental studies, has led to a rapidly expanding literature examining the role of these agents in the vascular and myocardial complications of systemic sepsis. However, the cellular mechanisms underlying the myocardial depressant effect of specific recombinant cytokines remain unclear. Finkel et al7 reported that recombinant TNF-α, IL-2, and IL-6, but not IL-1α, result in a rapid (ie, minutes) and reversible depression of contractile function of isolated, paced guinea pig papillary muscles, an effect that could be blocked by the addition of an l-arginine analogue that acts as a specific antagonist of nitric oxide (NO) synthase. Increased production of NO by a number of different tissues and cell types in response to LPS and/or inflammatory cytokines has now been reported,8 9 although this typically requires several hours to become apparent, consistent with increased transcription, synthesis, and activation of an inducible isoform of NO synthase (iNOS, or type 2 NO synthase). Pagani et al10 and Natanson et al1 reported that the intravenous injection of recombinant TNF-α in dogs resulted in a depression in myocardial function similar to that observed in systemic sepsis, with a time course consistent with iNOS induction in cellular constituents of cardiac muscle. In contrast, Yokoyama et al11 reported that the addition of TNF-α to physiological buffer perfusing isolated adult rat hearts or to primary isolates of ventricular myocytes from adult rats resulted in a rapid decline in the contractile function of the intact heart and isolated, paced myocyte preparations. This decline, unlike the observations of Finkel et al,7 could not be prevented by the addition of NO synthase antagonists.
The relevance of these immediate pharmacological effects of recombinant cytokines to the pathophysiology of myocardial depression is unclear. Several studies have now appeared that document a decline in cardiac myocyte contractile function12 and an increase in iNOS activity in ventricular muscle13 at least 6 hours after intraperitoneal injection of LPS in experimental animals. A recent report from this laboratory demonstrated that an increase in iNOS activity in primary isolates of adult rat ventricular myocytes, induced by a 24-hour preincubation in a species-specific mixture of inflammatory mediators contained in the cell-free supernatant collected from LPS-activated rat alveolar macrophages, was coincident with a decrease in the contractile responsiveness of these myocytes to β-adrenergic agonists.14 This inotropic response could be completely restored by the NO synthase inhibitor L-NMMA.
In this report, we examine the specific cytokine(s) responsible for the induction of iNOS activity in cardiac myocytes after exposure to LPS-activated macrophage-conditioned medium. We also demonstrate a dissociation between iNOS induction by individual recombinant cytokines and the development of decreased myocyte contractile responsiveness to β-adrenergic agonists, indicating that factors other than the induction of NO synthase per se contribute to myocyte contractile dysfunction.
Materials and Methods
Isolation and Preparation of Adult Rat Ventricular Myocytes
Calcium-tolerant ARVM were isolated from adult male Sprague-Dawley rats (weight, 225 to 275 g) by the approach originally described by Claycomb and Palazzo15 with modifications to minimize the number of contaminating nonmyocyte cells, as previously described.16 This typically resulted in contamination of myocyte primary isolates with no more than 2% to 4% nonmyocyte cells. Myocytes were cultured in a defined medium that is a modification of that originally described by Volz et al17 and consists of DMEM (GIBCO-BRL), including 2 mmol/L l-glutamine, 2 mmol/L pyruvate, 25 mmol/L HEPES, and NaHCO3 (pH 7.4 at 37°C) (Sigma) supplemented with 2 mg/mL BSA, 2 mmol/L l-carnitine, 5 mmol/L creatine, and 5 mmol/L taurine, with 100 IU/mL penicillin and 100 μg/mL streptomycin (GIBCO-BRL), as described previously.18 This medium supplemented with 0.1 μmol/L insulin and 0.1 nmol/L T3 is referred to as “defined medium.”18 Myocytes were plated at a density of 5×105 cells per 100-mm dish unless stated otherwise.
Isolation and Preparation of Rat Alveolar Macrophages
Alveolar macrophages were obtained by tracheal lavage of sodium pentobarbital–anesthetized male Sprague-Dawley rats (250 to 275 g) by a previously described technique.14 The first cell pellet after washes was resuspended at a concentration of 0.5×106 cells/mL in endotoxin-free DMEM medium containing 0.1% BSA with 100 IU/mL penicillin and 100 U/mL streptomycin and cultured at a density of 2.5×106 cells per 60-mm culture dish in a 95% O2/5% CO2 atmosphere at 37°C. One hour after plating, cells were washed three times in DMEM to remove nonadherent cells. Macrophages were then exposed to either endotoxin (the LPS component of Salmonella typhimurium [Sigma, lot 87F402]) at a concentration of 10 μg/mL or DMEM alone for 24 hours. Macrophage-conditioned medium was harvested, centrifuged at 400g for 10 minutes to remove cell debris, and then stored at −70°C for further use.
Measurement of Myocyte Contractile Function
Measurement of the amplitude and velocity of unloaded ventricular myocyte shortening and relengthening was made on the stage of an inverted phase-contrast microscope (Diavert; E. Leitz, Inc) using an optical-video system in which the analog motion signal was digitized and analyzed by computer, as previously described.16 Laminin-coated coverslips (Thermonox, Nunc Inc) with myocytes plated at a density of 5×103 cells per coverslip were placed in a temperature-controlled chamber at 37°C (total volume, 5 mL) and continuously superfused with a KHB buffer at 0.8 mL/min, with supplements as noted. One cell per coverslip was examined. Cells were depolarized by a 3-ms square-wave pulse delivered through a platinum electrode connected to a Grass S80 stimulator. Cells chosen for contractility experiments displayed clear cross-striations with well-defined edges and without spontaneous contractile twitch activity.
Measurement of NOS Activity
Conversion of [3H]l-Arginine to [3H]l-Citrulline
Myocyte homogenates were prepared by suspending approximately 5×105 cells (ie, the content of one 100-mm dish) in warm HBSS without MgCl2, CaCl2, or MgSO4 (GIBCO-BRL) containing 0.25% trypsin and 1 mmol/L EDTA, centrifuging at 100g at 4°C, washing in ice-cold PBS, and resuspending the final pellet in 0.5 mL of lysis buffer containing 20 mmol/L Tris-HCl (pH 7.4 at 4°C), 0.5 mmol/L EDTA, 0.5 mmol/L EGTA, 1 mmol/L DTT, 1 μmol/L THB4 (Dr B. Schircks Laboratories), 1 μmol/L leupeptin, and 0.2 mmol/L PMSF. The cells were sonicated on ice three times for 10 seconds with a Branson Sonifier 450. Homogenates were centrifuged at 1500g for 15 minutes at 4°C.
To determine NOS activity, 25 μL of total cell homogenate was incubated at 37°C for 2 hours in the presence of 50 mmol/L HEPES (pH 7.4 at 37°C), 1.25 mmol/L CaCl2, 1 mmol/L EDTA, 0.5 mmol/L NADPH, 5 μmol/L FAD, 5 μmol/L THB4, 10 μg/mL calmodulin, and 0.2 nmol/L [3H]l-arginine (Amersham) for a final total volume of 150 μL. The reactions were stopped by the addition of 2 mL of ice-cold 20 mmol/L HEPES (pH 5.5) and 5 mmol/L EDTA, and the total volume was applied to a Dowex-50W X8 column preequilibrated with 20 mmol/L HEPES (pH 5.5). The column retained [3H]l-arginine, while [3H]l-citrulline was eluted with 2 mL of deionized water, and radioactivity was determined by scintillation counting. The protein content of the homogenate was determined by the Bradford technique with a BioRad kit. The data from this NOS activity assay are reported as cpm/mg protein/2 h.
Nitrite Release in ARVM-Conditioned Medium
ARVM were plated on laminin-coated tissue culture plates (Costar) in defined medium and then for 24 hours in defined medium without phenol red and with additional reagents as indicated. After a 24-hour incubation, the medium was collected and centrifuged once at 1500g for 15 minutes at 4°C to remove cellular debris, and 150 μL of this supernatant was added to a 1:1 (vol/vol) mixture of Griess reagent (0.75% sulfanilamide [final concentration] in 0.5N HCl/0.075% naphthylethylenediamine; Sigma), and absorbance at 543 nm was determined spectrophotometrically. A standard curve was constructed by use of known concentrations of sodium nitrite over the linear range of the assay (0.01 to 50 μmol/L nitrite).
Cell Respiration Assay
To verify myocyte viability after exposure to cytokines or to LPS-activated macrophage-conditioned medium, 6×104 ARVM were plated in 12-well plates (Costar) in defined medium with and without recombinant cytokines or a 1:1 (vol/vol) dilution of LPS-activated macrophage-conditioned medium. After a 24-hour incubation, the medium was supplemented with 0.2 mg/mL MTT (Sigma). At successive time intervals, the culture medium was removed and myocytes were solubilized in 1 mL of DMSO. The extent of reduction of MTT to formazan within cells, a measure of cellular respiration, was quantified by measurement of the ratio of absorbances at 550 and 630 nm.19 The results are expressed as a percentage of the 550/630-nm absorbance ratio of cells incubated for the same time duration in defined medium alone.
Western Analysis of iNOS Protein in ARVM
An equal number of myocytes per experimental condition were lysed directly in each well by application of 500 μL of twice-concentrated sample buffer containing 25% (vol/vol) 4×Tris-HCl/SDS at pH 6.8 (1×Tris/SDS contains 6.05 g Tris base and 0.4 g SDS in 100 mL), 20% glycerol, 4% SDS (wt/vol), and 2% (vol/vol) β-mercaptoethanol, and the mixture was boiled for 5 minutes. The denatured proteins (ie, 40 μL per well; approximately 100 μg of protein) were separated on a precast 12% polyacrylamide gel and Miniprotean II system (BioRad) and then transferred to a nitrocellulose membrane (Millipore HATF 20200) in 25 mmol/L CAPSO buffer (pH 10) overnight at 4°C. Reversible staining with Ponceau red was used to verify equal loading and transfer efficiency for each lane. The membrane was blocked for 2 hours at room temperature with 1% BSA in TBST and probed for 2 hours with an anti–murine macrophage iNOS–specific antiserum at a 1:000 dilution in TBST. After three 10-minute washes with TBST, the membranes were incubated for 1 hour with an iodinated goat anti-rabbit secondary antibody (NEN DuPont) at a specific activity of 100 000 cpm/mL. The membranes were washed three times with TBST, dried, and exposed for at least 72 hours to XAR Kodak film at −80°C. The intensity of each lane was quantified by densitometric analysis.
Northern Analysis of Myocyte iNOS mRNA
Northern blot hybridizations were performed with the 217-bp cDNA probe representing a portion of the ARVM iNOS mRNA sequence identified by this laboratory using RT-PCR techniques on mRNA isolated from cytokine-pretreated myocytes, as reported elsewhere.20 Hybridizations were performed by electrophoresis of 15 μg of total RNA through a 1.5% formaldehyde-agarose gel and blotting onto a nylon membrane overnight by capillary transfer. cDNA probes were radiolabeled with [32P]dCTP by random-primer labeling. After 4 hours of prehybridization at 42°C, the blots were hybridized overnight at 42°C and then washed with 2×SSC/0.1% SDS (SSC is 0.15 mol/L NaCl and 0.015 mol/L sodium citrate) for 30 minutes at room temperature, followed by 1×SSC/0.1% SDS at 37°C and 0.2×SSC/0.1% SDS at 65°C. The blots were prepared for autoradiography at −70°C for at least 6 hours.
Reversal of MacLPS(+) Medium–Induced iNOS Activity in Cardiac Myocytes by IL-1 Receptor Antagonist
Since we previously detected both IL-1 and TNF-α in MacLPS(+),14 specific antagonists to both cytokines were tested for their ability to prevent both the induction of iNOS activity and the hyporesponsiveness to β-adrenergic agonists that are characteristic of adult cardiac myocytes after a 12- to 24-hour exposure to this conditioned medium.14 In preliminary experiments, 100- and 200-ng/mL concentrations of rhIL-1RA and serial dilutions of a rabbit polyclonal anti–TNF-α antiserum were tested to determine the peptide concentration and antiserum dilution sufficient to inhibit all endogenous rat IL-1 and TNF-α bioactivities, respectively, in typical batches of MacLPS(+) medium using cytokine-specific bioassays as previously described.14 A 1:1000 dilution of polyclonal anti–rat TNF-α antisera blocked the TNF-α–dependent WEHI clone 164 cytotoxicity of undiluted MacLPS(+)-conditioned medium (TNF-α concentration, ≈40 to 60 ng/mL). One hundred nanograms per milliliter rhIL-1RA was sufficient to completely inhibit endogenous IL-1 activity in MacLPS(+) medium (concentration, ≈2 ng/mL) as assayed by stimulation of D10.64.1 T-cell proliferation. IL-1RA had no nonspecific effect on myocyte viability or cellular respirations (data not shown).
The characteristic decline in contractile responsiveness to isoproterenol in MacLPS(+)-pretreated cardiac myocytes and its reversal by the l-arginine analogue L-NMMA are illustrated in Fig 1A⇓. IL-1RA, at concentrations as high as 500 ng/mL, did not affect baseline or isoproterenol-stimulated myocyte contractile function in myocytes preincubated for 24 hours in defined medium alone (data not shown). Addition of rhIL-1RA (200 ng/mL) to a 50% dilution (vol/vol) of MacLPS(+) medium in defined medium used for the 24-hour myocyte incubation prevented the abnormal contractile responsiveness to isoproterenol (Fig 1A⇓). As expected from these data, IL-1RA also prevented the induction of iNOS activity in MacLPS(+)-pretreated cardiac myocytes, as shown in Fig 1B⇓. Concentrations of IL-1RA <200 ng/mL did not reproducibly prevent the induction of iNOS activity in cardiac myocytes exposed to a 50% dilution of MacLPS(+) medium (data not shown). As anticipated from the results of the iNOS enzymatic activity assay, IL-1RA also diminished the amount of a 130-kD band detected by anti–murine macrophage iNOS antibodies on Western analysis (Fig 1C⇓) compared with myocytes exposed to MacLPS(+) medium in the absence of the cytokine antagonist. Note that L-NMMA does reduce NOS activity to levels below that observed in control myocytes, implying the presence of a constitutive NOS activity in these cells, as previously reported.21
In contrast to IL-1RA, dilutions of a polyclonal anti–rat TNF-α antiserum (ie, 1:250) in excess of that determined to completely inhibit endogenous TNF-α bioactivity in MacLPS(+) medium had no effect on the appearance of decreased contractile responsiveness to isoproterenol in myocyte-pretreated medium from LPS-activated macrophages (data not shown). Similarly, anti–TNF-α antisera did not significantly attenuate the increase in iNOS enzyme activity induced by activated macrophage-conditioned medium. Three dilutions of TNF-α antiserum had no effect on myocyte respiration or on baseline myocyte contractile function (data not shown).
IL-1β and iNOS Induction in Cardiac Myocytes
The ability of IL-1RA to prevent the induction of iNOS activity, as well as the negative inotropic effect of MacLPS(+) medium, suggested that IL-1β was necessary and perhaps sufficient for the pharmacological effect of the activated macrophage-conditioned medium on myocyte responsiveness to isoproterenol. rhIL-1β (4 ng/mL) alone was sufficient to induce iNOS mRNA (Fig 2A⇓), as we have reported previously,20 and a synergistic increase in iNOS protein levels was observed only with the combination of IL-1β and IFN-γ, as shown in Fig 2A⇓. IL-1β alone did cause a readily detectable increase in iNOS enzyme activity that was ≈10-fold higher than basal levels of enzyme activity measured in myocyte cell lysates (data not shown). This increase was detectable at 12 to 16 hours after addition of cytokines and maximal within 24 hours (data not shown). The increase in iNOS activity in IL-1β–exposed myocytes was abolished by pretreatment with cycloheximide and was not affected by addition of 5 mmol/L EGTA (Fig 2B⇓), consistent with the reported relative insensitivity of this NOS isoform to changes in Ca2+ activity within the physiological range in the cytosol of most cells.
As expected from reports that THB4 is a necessary cofactor for NOS activity,19 removal of THB4 and addition of the GTP cyclohydrolase 1 inhibitor DAHP (2,4-diamino-6-hydroxypyrimidine; Sigma) effectively inhibited nitrite release into cytokine-pretreated myocyte-conditioned medium after 24 hours of exposure to IL-1β and IFN-γ (data not shown). This GTP cyclohydrase 1 inhibitor also did not affect the cytokine-induced increase in iNOS protein levels in these cells (Fig 2A⇑).
Dissociation Between iNOS Induction and Diminished Myocyte Responsiveness to β-Adrenergic Agonists
Despite the important role of increased NOS activity in mediating the negative inotropic effect of MacLPS(+) medium and the data reported above that IL-1RA could entirely abrogate the effect of MacLPS(+) medium on myocyte responsiveness to β-agonists, induction of iNOS activity in myocytes by IL-1β alone did not affect either basal contractile function (data not shown) or the positive inotropic response of these cells to isoproterenol (Fig 3A⇓). Similarly, TNF-α (100 ng/mL), which also results in an induction of iNOS activity in cardiac myocytes, did not affect the contractile responsiveness to isoproterenol either alone for 24 hours (data not shown) or in combination with IL-1β (Fig 3A⇓). Similarly, rmIFN-γ also induced iNOS activity in cytokine-pretreated myocytes that was maximal at 500 U/mL, although again, this cytokine alone did not affect myocyte contractile responsiveness to isoproterenol (data not shown). In contrast, the combination of rhIL-1β (4 ng/mL) and rmINF-γ (500 U/mL) both induced iNOS activity (Fig 4D⇓) and diminished myocyte contractile responsiveness to isoproterenol (Fig 4C⇓).
Insulin: Effects on iNOS Induction and Diminished Myocyte Responsiveness to β-Adrenergic Agonists
The defined medium preparation routinely used in this laboratory for adult rat ventricular myocyte primary cultures (called “ACCITT” in Reference 1818 ) contains 0.1 μmol/L insulin and 0.1 nmol/L T3, both of which increase myocyte survival beyond 48 hours in culture while maintaining selected aspects of the phenotype these cells exhibit in fresh primary isolates. Removal of T3 from the culture medium did not affect the extent of induction of iNOS mRNA in myocytes (data not shown), nor did it affect the increase in iNOS protein content (Fig 4A⇑) in myocytes exposed to the combination of IL-1β and IFN-γ for 24 hours.
Similarly, selective removal of insulin from the defined medium had little effect on the extent of induction of iNOS mRNA transcript. As shown in Fig 4B⇑, iNOS mRNA abundance was typically minimally affected by the presence or absence of insulin in the preincubation medium with cytokines. Similarly, there was no consistent difference in protein levels (data not shown) or maximal iNOS protein activity in cell lysates in cytokine-treated cardiac myocytes (Fig 4D⇑). However, cytokine-stimulated release of nitrite from intact myocytes was negligible in the absence of insulin and increased 10-fold when insulin was added along with cytokines to the defined medium (Fig 4E⇑). Insulin was also necessary for the manifestation of L-NMMA–inhibitable abnormal contractile responsiveness to isoproterenol after a 24-hour preincubation of myocytes in medium containing IL-1β and IFN-γ. As shown in Fig 4C⇑, in the absence of insulin, the contractile response of cytokine-pretreated myocytes to isoproterenol was not different from that of control cells. Addition of 0.1 μmol/L insulin to the defined medium resulted in the characteristic hyporesponsiveness to β-adrenergic agonists of cytokine-pretreated myocytes.
iNOS Contractile Dysfunction in ARVM Exposed to MacLPS(+) Medium
The focus of this report is on the relation between iNOS (ie, type II NO synthase) induction in cardiac myocytes by specific cytokines or combinations of cytokines and decreased inotropic responsiveness to β-adrenergic agonists. Blockade of IL-1β activity but not TNF-α activity in the activated macrophage-conditioned medium abrogated both iNOS induction and the contractile dysfunction induced by MacLPS(+) in ARVM. rhIL-1β, rhTNF-α, and rmIFN-γ individually induced iNOS activity at 24 hours in ARVM, but only the combination of IL-1β and IFN-γ reproduced the contractile dysfunction. Moreover, insulin was required for the hyporesponsiveness to β-adrenergic agonists to be manifested, even though it did not affect the extent of iNOS induction, as measured in cell lysates.
In preliminary experiments, we had noted that individual recombinant cytokines had little or no effect on either baseline or β-adrenergic agonist–stimulated myocyte contractile function at 24 hours, when iNOS induction should have been maximal. For this reason, we adopted an experimental model similar to that used by other researchers3 4 5 using the species-specific mixture of soluble inflammatory mediators present in medium conditioned by LPS-activated rat alveolar macrophages [ie, MacLPS(+) medium]. As we reported,14 incubation of adult cardiac myocytes in MacLPS(+) medium for 24 hours resulted in increased NO release into myocyte-conditioned medium, as detected by a rise in nitrite content in myocyte-conditioned medium and by an increase in cGMP levels in a reporter cell line (RFL-6 cells) exposed to this medium. The increase in nitrite accumulation after exposure to MacLPS(+) medium corresponded temporally to a decline in myocyte contractile responsiveness to isoproterenol. Addition of the l-arginine analogue L-NMMA prevented the rise in nitrite content in myocyte-conditioned medium and restored to normal the inotropic responsiveness of these cells to β-agonists, linking iNOS induction in cardiac myocyte primary cultures to contractile dysfunction. In that report, we noted that another major cellular constituent of ventricular muscle, microvascular endotheli-al cells, also exhibited a marked increase in iNOS activ-ity after a 24-hour exposure in primary culture to MacLPS(+) medium.14 Although it is possible that these microvascular endothelial cells or other cell types could be contributing to the iNOS activity detected in adult cardiac myocyte primary isolates, the number of nonmyocyte cells present when myocytes were maintained in defined medium was always <4%. Importantly, we also reported that release of NO can be detected from single isolated cytokine-pretreated cardiac myocytes by use of an NO-selective porphyrinic/Nafion microsensor.20
To determine whether IL-1β or TNF-α, two cytokines present in relatively high concentrations in MacLPS(+) medium14 that are known to induce iNOS in several cell types, played a role in the induction of iNOS in MacLPS(+)-exposed cardiac myocytes, we used rhIL-1RA and an anti–rat TNF-α antiserum as selective cytokine antagonists. As shown in Fig 1⇑, only IL-1RA prevented the induction of iNOS activity and the decline in myocyte contractile responsiveness to isoproterenol when added at the initiation of the myocyte incubation with MacLPS(+) medium. It is unclear why the TNF-α antiserum was ineffective, since TNF-α induces iNOS at 24 hours in cardiac myocytes and, as shown in Fig 3⇑, results in roughly an additive increase in iNOS activity with IL-1 at 24 hours in these cells, although neither cytokine, alone or in combination, affects myocyte contractile responsiveness to isoproterenol. The ineffectiveness of the anti–rat TNF-α antiserum in this model cannot be ascribed to the presence of other endogenous cytokine antagonists in the MacLPS(+) medium, since the same antiserum was effective in reducing iNOS induction in confluent, low-passage primary cultures of microvascular endothelial cells, also isolated from adult rat ventricular muscle, that had been exposed for 24 hours to the same dilution of MacLPS(+) medium.22 However, endogenous TNF-α levels in the MacLPS(+) medium determined by bioassay14 were only 40 to 60 ng/mL, well below the concentration that leads to maximal iNOS induction in adult cardiac myocytes by 500 ng/mL rhTNF-α. In contrast, endogenous IL-1β was present at levels that were sufficiently high to result in maximal iNOS induction in these cells (ie, 1 to 2 ng/mL). This may explain in part the difference in efficacy of the two cytokine antagonists.
Recombinant Cytokines, iNOS Induction, and Myocyte Contractile Dysfunction
The results of the experiments with the IL-1 receptor antagonist suggested that IL-1β present in the MacLPS(+) medium was responsible for a portion of the induction of iNOS activity in cytokine-pretreated myocytes and was necessary for the decreased contractile responsiveness of these cells studied after 24 hours of incubation in the activated macrophage-conditioned medium. However, IL-1β is clearly not sufficient to induce contractile dysfunction, since this cytokine, at a concentration and time point that resulted in maximal induction of iNOS, had no effect on the inotropic responsiveness of cardiac myocytes to isoproterenol. A reproducible decline in myocyte contractile responsiveness to β-adrenergic agonists with recombinant cytokines was achieved only with the combination of IL-1β and IFN-γ. IFN-γ alone also induced iNOS activity in ARVM at 24 hours but did not affect myocyte inotropic responsiveness to isoproterenol. The addition of IFN-γ to IL-1 results in a synergistic increase in iNOS protein content and iNOS activity (Fig 2⇑), as has been reported for other cells, including rat cardiac microvascular endothelial cells23 and (with LPS) murine macrophages.24 25
As described briefly above, exposure to soluble inflammatory mediators either in vivo or in vitro has been associated with a decrease in cardiac myocyte contractile function, with a time course consistent with iNOS induction.3 4 12 14 Schulz et al13 demonstrated iNOS induction in TNF-α– and IL-1β–pretreated adult rat ventricular myocytes by measuring the rate of conversion of [3H]l-arginine to [3H]l-citrulline in myocyte cell lysates, although these investigators did not measure contractile function. Other protocols that have implicated iNOS in mediating cardiac or isolated myocyte contractile dysfunction in animals injected with LPS or infused with specific recombinant cytokines are likely to involve induction of multiple cytokine and other autacoid signaling pathways. It is unclear from the data reported here or by others whether cytokine-induced decreased basal or isoproterenol-stimulated myocyte contractile responsiveness can be explained solely on the basis of the magnitude of iNOS induction and continuous NO release or whether other factors, including the release of autacoids such as platelet-activating factor26 or cyclooxygenase products27 28 could contribute to the contractile dysfunction we observed in isolated myocytes.
Insulin: Effects on iNOS Induction and NO-Dependent Myocyte Contractile Dysfunction
There is a rapidly growing literature on the regulation of NOS activity by a number of peptide signaling factors, autacoids, drugs, and intracellular second messengers, including NO itself.20 26 29 30 31 32 33 34 In the case of iNOS, this regulation occurs not only at the transcriptional level and at the level of mRNA half-life but also posttranslationally by several reported mechanisms. We observed that one of the components of the defined medium we routinely use to maintain adult cardiac myocytes in short-term (ie, 1- to 3-day) primary culture appeared to modify iNOS activation in cardiac myocytes in response to cytokines, as well as the extent of myocyte hyporesponsiveness to β-adrenergic agonists after cytokine pretreatment. As shown in Fig 4⇑, neither T3 nor insulin affected the extent of iNOS induction by IL-1β and IFN-γ in cardiac myocytes at 24 hours. However, addition of insulin to the defined medium during myocyte exposure to cytokines resulted in a marked (10-fold) increase in nitrite release into the medium by intact cells (Fig 4E⇑). Importantly, insulin was required also for the reduced contractile responsiveness to isoproterenol to become apparent (Fig 4C⇑).
Schini et al35 reported that insulin-like growth factors (ie, IGF-I and IGF-II), as well as insulin itself, decrease maximal iNOS activity in cell lysates and nitrite production by confluent, serum-starved rat aortic smooth muscle cells after exposure to either IL-1β or TNF-α when added concurrently with the cytokines. The greatest effect was observed with IGF-I, while insulin, at the same concentration used in the experiments reported here (ie, 0.1 μmol/L), only modestly reduced iNOS activity in cytokine-pretreated smooth muscle cells. However, there were important differences in the response to IGF-I in cytokine-pretreated rat aortic smooth muscle cells and the effects we observed with insulin in adult rat cardiac myocytes. IGF-I markedly diminished the functional effects of iNOS induction in cytokine-treated rat aortic smooth muscle cells (eg, inhibition of thrombin-induced platelet aggregation). Also, the responsiveness of aortic smooth muscle cells to IGF-I was dependent on the amount of time the cells had been in serum-free medium, declining rapidly after 24 hours, an observation that these authors attributed to changes in cell phenotype with withdrawal from the cell cycle.35
Although the concentration of insulin used in our defined medium (0.1 μmol/L) is sufficiently high that some cross talk with cardiac myocyte IGF-I or IGF-II receptors is possible, we have no evidence for a decline in maximal iNOS activity in insulin-treated cells. Indeed, insulin was necessary to obtain maximal rate of nitrite production in cytokine-pretreated cardiac myocytes. Insulin may have been necessary to facilitate l-arginine synthesis by argininosuccinate synthase or l-arginine transport, although the cationic amino acid transporters that have been cloned to date that constitute the “y+” transport activity in most cells are not thought to be regulated directly by insulin.36 37 38 Insulin also could directly or indirectly affect levels of other cofactors necessary for maximal iNOS activity in intact cells. Whereas all these cofactors are added exogenously in excess concentrations in the assay for iNOS activity in cell lysates, limiting concentrations of substrate, cofactors, or both could explain the markedly reduced nitrite release observed from cytokine-pretreated myocytes.
Insulin could also facilitate or modify downstream signaling pathways in cardiac myocytes. For example, insulin has been shown to affect the activation of a cGMP-inhibited cAMP phosphodiesterase in adipocytes39 and in platelets.40 41 The observation that forearm basal levels of NO release and vascular smooth muscle responsiveness to nitroprusside are diminished in diabetic patients is consistent with several of these mechanisms.42 These possibilities are the subject of ongoing research in this laboratory.
In summary, induction of iNOS in cardiac myocytes is necessary for the delayed onset of contractile hyporesponsiveness to β-adrenergic agonists characteristic of cardiac myocytes exposed to soluble inflammatory mediators produced by activated macrophages in vitro. Among these mediators, IL-1β is a necessary but not sufficient factor for both iNOS induction and contractile dysfunction to occur. The combination of iNOS induction and decreased myocyte responsiveness to β-adrenergic agonists could be reproduced only by the combination of IL-1β and IFN-γ together and required the presence of insulin in the culture medium. Therefore, induction and activation of iNOS in ventricular myocytes in vitro appears to be regulated by signal transduction pathways activated by specific combinations of cytokines (eg, IL-1β and IFN-γ) and requires the presence of additional myocyte trophic factors.
Selected Abbreviations and Acronyms
|ARVM||=||adult rat ventricular myocytes|
|CMEC||=||cardiac microvascular endothelial cells|
|DMEM||=||Dulbecco’s minimum essential medium|
|IGF||=||insulin-like growth factor|
|IL-1RA||=||IL-1 receptor antagonist|
|MacLPS(+) medium||=||LPS-activated rat alveolar macrophage–conditioned medium|
|RT-PCR||=||reverse transcriptase–polymerase chain reaction|
|TBST||=||Tris-buffered saline with 0.05% (vol/vol) Tween 20|
|TGF-β||=||transforming growth factor-β|
|TNF-α||=||tumor necrosis factor-α|
This work was supported by grants R37-HL-36141 (to Dr Smith) and R29-HL-46457 (to Dr Michel) from the National Institutes of Health. Dr Michel is the recipient of an Established Investigator Award from the American Heart Association. Dr Balligand and Dr Ungureanu-Longrois were supported by fellowship awards from the Massachusetts Affiliate of the American Heart Association. Dr Simmons is supported by a fellowship award from the Medical Research Council of Canada. Dr Okada was the recipient of a fellowship award from the Japanese Heart Foundation. rhIL-1RA was a gift of Synergen (Boulder, Colo).
This manuscript was sent to Dr Leslie A. Leinwand, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
- Received March 14, 1994.
- Accepted April 20, 1995.
- © 1995 American Heart Association, Inc.
Natanson C, Eichenholz PW, Danner RL, Eichacker PQ, Hoffman WD, Kuo GC, Banks SM, MacVittie TJ, Parillo JE. Endotoxin and tumor necrosis factor challenges in dogs simulate the cardiovascular profile of human septic shock. J Exp Med. 1989;169:823-832.
Chung MK, Gulick TS, Rotondo RE, Schreiner GF, Lange LG. Mechanisms of cytokine inhibition of β-adrenergic stimulation of cyclic AMP in rat cardiac myocytes: impairment of signal transduction. Circ Res. 1990;67:753-763.
Lange LG, Schreiner GF. Immune cytokines and cardiac disease. Trends Cardiovasc Med. 1992;2:145-151.
Lefer A, Rovetto M. Influence of a myocardial depressant factor on physiologic properties of cardiac muscle. Proc Soc Exp Biol. 1970;134:269-273.
Finkel MS, Oddis CV, Jacob TD, Watkins SC, Hattler BG, Simmons RL. Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science. 1992;257:387-389.
Dinerman JL, Lowenstein CJ, Snyder SH. Molecular mechanisms of nitric oxide regulation: potential relevance to cardiovascular disease. Circ Res. 1993;73:217-222.
Pagani FD, Baker LS, Hsi C, Knox M, Fink MP, Visner MS. Left ventricular systolic and diastolic dysfunction after infusion of tumor necrosis factor-α in conscious dogs. J Clin Invest. 1992;90:389-398.
Yokoyama T, Vaca L, Rossen RD, Durante W, Hazarika P, Mann DL. Cellular basis for the negative inotropic effects of tumor necrosis factor-α in the adult mammalian heart. J Clin Invest. 1993;92:2303-2312.
Brady AJB, Poole-Wilson PA, Harding SE, Warren JB. Nitric oxide production within cardiac myocytes reduces their contractility in endotoxemia. Am J Physiol. 1992;263:H1963-H1966.
Balligand J-L, Ungureanu D, Kelly RA, Kobzik L, Pimental D, Michel T, Smith TW. Abnormal contractile function due to induction of nitric oxide synthesis in rat cardiac myocytes follows exposure to activated macrophage-conditioned medium. J Clin Invest. 1993;91:2314-2319.
Berger H-J, Prasad SK, Davidoff AJ, Pimental D, Ellingsen Ø, Marsh JD, Smith TW, Kelly RA. Continual electric field stimulation preserves contractile function of adult ventricular myocytes in primary culture. Am J Physiol. 1994;266:H341-H349.
Ellingsen Ø, Davidoff AJ, Prasad SK, Berger H-J, Springhorn JP, Marsh JD, Kelly RA, Smith TW. Adult rat ventricular myocytes cultured in defined medium: phenotype and electromechanical function. Am J Physiol. 1993;265:H747-H754.
Gross SS, Levi R. Tetrahydrobiopterin synthesis: an absolute requirement for cytokine-induced nitric oxide generation by vascular smooth muscle. J Biol Chem. 1992;267:25722-25729.
Balligand J-L, Ungureanu-Longrois D, Pimental D, Malinski TA, Kapturczak M, Taha Z, Lowenstein CJ, Davidoff AJ, Kelly RA, Smith TW, Michel T. Cytokine-inducible nitric oxide synthase (iNOS) expression in cardiac myocytes: characterization and regulation of iNOS expression and detection of iNOS activity in single cardiac myocytes in vitro. J Biol Chem. 1994;269:27580-27588.
Balligand J-L, Kelly RA, Marsden PA, Smith TW, Michel T. Control of cardiac muscle cell function by an endogenous nitric oxide signalling system. Proc Natl Acad Sci U S A. 1993;90:347-351.
Ungureanu-Longrois D, Balligand J-L, Okada I, Simmons WW, Kobzik L, Lowenstein CJ, Kunkel SL, Michel T, Kelly RA, Smith TW. Contractile responsiveness of ventricular myocytes to isoproterenol is regulated by induction of nitric oxide synthase activity in cardiac microvascular endothelial cells in heterotypic primary culture. Circ Res. 1995;77:486-493.
Balligand J-L, Ungureanu-Longrois D, Kobzik L, Lowenstein CJ, Simmons WW, Lamas S, Kelly RA, Smith TW, Michel T. Induction of NO synthase in cardiac microvascular endothelial cells: IL-1β and IFN-γ promote transcription of an endothelial iNOS. Am J Physiol. 1995;268:H1293-H1303.
Lorsbach RB, Murphy WJ, Lowenstein CJ, Snyder SH, Russell SW. Expression of the nitric oxide synthase gene in mouse macrophages activated for tumor cell killing: molecular basis for the synergy between interferon-γ and lipopolysaccharide. J Biol Chem. 1993;268:1908-1913.
Lowenstein CJ, Alley EW, Raval P, Snowman AM, Snyder SH, Russell SW, Murphy WJ. Macrophage nitric oxide synthase gene: two upstream regions mediate induction by interferon γ and lipo-polysaccharide. Proc Natl Acad Sci U S A. 1993;90:9730-9734.
Szabo C, Wu C-C, Mitchell JA, Gross SS, Thiemermann C, Vane JR. Platelet-activating factor contributes to the induction of nitric oxide synthase by bacterial lipopolysaccharide. Circ Res. 1993;73:991-999.
Salvemini D, Misko TP, Masferrer JL, Seibert K, Currie MG, Needleman P. Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci U S A. 1993;90:7240-7244.
Koide M, Kawahara Y, Nakayama I, Tsuda T, Yokoyama M. Cyclic AMP-elevating agents induce an inducible type of nitric oxide synthase in cultured vascular smooth muscle cells: synergism with the induction elicited by inflammatory cytokines. J Biol Chem. 1993;268:24959-24966.
Radomski MW, Palmer RMJ, Moncada S. Glucocorticoids inhibit the expression of an inducible, but not the constitutive, nitric oxide synthase in vascular endothelial cells. Proc Natl Acad Sci U S A. 1990;87:10043-10047.
Buga GM, Griscavage JM, Rogers NE, Ignarro LK. Negative feedback regulation of endothelial cell function by nitric oxide. Circ Res. 1993;73:808-812.
Schini VB, Catovsky S, Schray-Utz B, Busse R, Vanhoutte PM. Insulin-like growth factor I inhibits induction of nitric oxide synthase in vascular smooth muscle cells. Circ Res. 1994;74:23-32.
Closs EI, Lyons CR, Kelly C, Cunningham JM. Characterization of the third member of the MCAT family of cationic amino acid transporters: identification of a domain that determines the transport properties of the MCAT proteins. J Biol Chem. 1993;268:20796-20800.
Smith CJ, Vasta V, Degerman E, Belfrage P, Manganiello VC. Hormone-sensitive cyclic GMP-inhibited cyclic AMP phosphodiesterase in rat adipocytes: regulation of insulin- and cAMP-dependent activation by phosphorylation. J Biol Chem. 1991;266:13385-13390.
Calver A, Collier J, Vallance P. Inhibition and stimulation of nitric oxide synthesis in human forearm arterial bed of patients with insulin-dependent diabetes. J Clin Invest. 1992;90:2548-2554.