INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE MAGNITUDE OF HEMODYNAMIC OVERLOAD occurring immediately after myocardial infarction plays a critical role in determining whether the noninfarcted cardiac muscle will develop functionally adaptive hypertrophy or eventually undergo decompensation and failure (28, 31). Although the extant literature indicates that hemodynamic overload affects the myocardium by mechanical stretch primarily (33), the biochemical mechanisms responsible for orchestrating these phenotypically and prognostically different outcomes remain unclear.

Accumulating evidence indicates that a portfolio of endogenous autocrine/paracrine cytokines and growth-promoting factors are promptly synthesized within the myocardium in response to mechanical stretch (14). Although the role that these substances play is not precisely defined, it has been proposed that they may contribute to initiate and modulate critical responses within the overloaded myocardium, such as myocyte growth, apoptotic myocyte death, and reactive fibrosis, which, in turn, are the main determining factors of the final outcome of hemodynamic overload (5).

Among these endogenous molecules, increasing attention has been recently focused on tumor necrosis factor (TNF)-, interleukin (IL)-6, and insulin-like growth factor (IGF)-1. Independent investigators have documented their early upregulation within the myocardium in experimental load-induced cardiac hypertrophy (9, 19, 30). Interestingly, although these peptides share the common ability to activate myocyte growth (15, 18, 40), they exert different effects with regard to apoptotic myocyte death and interstitial compartment. Specifically, whereas IL-6 and IGF-1 possess univocal antiapoptotic properties (8, 38) and preserve the integrity of the interstitial network (7, 15), TNF- appears to serve a dual biological purpose. At "physiological" concentrations, it exerts cytoprotective effects within the myocardium, including antioxidant and antiapoptotic effects (24, 26). In contrast, at "pathophysiological" concentrations, it stimulates myocyte apoptosis (22) and reactives myocardial fibrosis (23, 36).

Given this evidence, in the current study, we examined myocardial TNF-, IL-6, and IGF-1 expression, and the levels of their cognate receptors, in response to different degrees of acute mechanical stretch in the adult rat. To this aim, we used an in vitro isolated, isovolumic, buffer-perfused heart Langendorff preparation to achieve graded levels of mechanical stress by inflating an intraventricular balloon at two different levels of end-diastolic pressures. This model allowed us to eliminate the confounding effects of circulating substances such as neurohormones, which are known to be upregulated under in vivo hemodynamic overload.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adult male normal Wistar rats weighing 300-500 g (Stefano Morini, Reggio Emilia, Italy), fed with normal rat chow and water ad libitum, were used for the whole heart experiments. All methods described conformed to the "Guiding Principles for Research Involving Animals and Human Beings," and the protocol was approved by the Animal Care Committee of the University Federico II of Naples, Italy.

Isolated whole heart preparation. Rats were killed, and the isolated hearts were placed in an isovolumic buffer-perfused preparation according to the Langendorff technique, as previously described (6). Briefly, the rats were anesthetized by an intraperitoneal injection of ketamine (50 mg/kg) and xylazine (10 mg/kg), and 200 IU heparin were injected into the femoral vein. One minute later, the hearts were quickly excised and immersed in ice-cold Krebs-Henseleit solution (see below), weighed, and mounted on a cannula inserted into the ascending aorta. The hearts were retrogradely perfused within 30 s after the thoracotomy using a Krebs-Henseleit solution containing (in mM) 118 NaCl, 4.7 KCl, 1.2 KH₂PO₄, 1.5 CaCl₂, 1.2 MgCl₂, 23 NaHCO₃, and 5.5 dextrose saturated with a 95% O₂-5% CO₂ gas mixture to a pH of 7.4. A constant flow perfusion of 10 ml · min¹ · g heart wt¹ was achieved by means of a roller pump, and coronary perfusion pressure was monitored by a Statham P23 Db transducer (Gould; Cleveland, OH) connected to the perfusion line. This value was chosen according to preliminary experiments with graded ischemia that revealed an aerobic pattern of lactate consumption measured according to Apstein et al. (1). Left ventricular isovolumic pressure was measured by a second Statham P23 DB transducer attached to a fluid-filled latex balloon inserted into the left ventricle via the mitral valve. Thebesian venous return from the left ventricle was emptied via a drain inserted in parallel with the balloon. Cardiac temperature was set at 25°C, measured by a temperature probe inserted into the right ventricle, and the hearts were paced at 3 Hz. The left intraventricular balloon was inflated just enough to obtain a pressure signal to monitor preparation stability.

Aequorin loading. Aequorin loading, performed as previously described (6), was used to monitor qualitative and quantitative change in intracellular calcium, which is a sensitive marker of myocardial ischemia (22). Briefly, 3-5 µl of an aequorin-containing solution (1 µg/ml) were macroinjected into the interstitium of the inferoapical region of the left ventricle. The heart was then positioned in a organ bath with the aequorin-loaded area of the left ventricle directed toward the cathode of a photomultiplier (model 9635QA, Thorn-EMI, Gencom) and submerged in Krebs-Henseleit solution. The organ bath was enclosed in a light-occlusive photographic bellows designed for studies with aequorin-loaded muscles by Blinks et al. (3) and modified for whole heart studies by Kihara et al. (20).

Experimental protocol and signal recording. After 15-30 min at 25°C, the temperature was gradually increased to 37°C and kept constant by regulating the temperature of the perfusate. After a 15-min stabilization period, the balloon was further inflated to achieve an end-diastolic pressure of ~5 mmHg (unstretched control myocardium), 15 mmHg (moderate stretch), or 35 mmHg (severe stretch). The coronary flow rate was adjusted to keep a constant tissue perfusion of 10 ml · min¹ · g heart wt¹. After 10, 20, 40, and 60 min, respectively, the balloon was rapidly deflated to a volume just enough to obtain a pressure signal. After a 5-min stabilization, perfusion was terminated, and the left ventricles, carefully separated from the right ventricles, were quickly cross sectioned in two portions. One portion was snap-frozen in liquid nitrogen for Northern and Western blot analysis (see Northern blotting and Western blotting, respectively), and the other was formalin-fixed for immunohistochemistry (see Immunohistochemical analysis). Before, during, and after the course of experiments, the digital signals of the left ventricular isovolumic pressure, aequorin light signals, and coronary perfusion pressure were simultaneously recorded on a four-channel recorder and averaged in a computer (6), and the left ventricular pressure tracing was further analyzed using customized software (6) to obtain the following parameters: peak left ventricular systolic pressure, left ventricular end-diastolic pressure, left ventricular developed pressure, and maximum and minimum values of the first pressure derivative with respect of time. At 10-min intervals during experiments, a sample of the coronary venous effluent was collected in a calibrated cylinder over a period of 1 min for measuring lactate production (Lactate Reagent, Sigma) and lactate dehydrogenase release (Lactate Dehydrogenase Reagent, Sigma).

Probe generation. Rat-specific cDNA probes for TNF-, IL-6, IGF-1, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were generated by sequential RT-PCR (GIBCO-BRL Life Technologies) according to the manufacturer's procedures. Total RNA from endotoxin-stimulated (10 µg/ml for 8 h) rat macrophage cells and the normal rat liver were used as templates for the RT reactions. For the 540-bp TNF- probe, the following oligonucleotides were used: sense primer (5'-CGCTCTTCTGTCTACTGAAC-3'), corresponding to nucleotides 4,568-4,577; antisense primer (5'-TTCTCCAGCTGGAAGACTCC-3'), corresponding to nucleotides 5,939-5,958 (GenBank Accession No. L00981). For the 650-bp IL-6 probe, the following oligonucleotides were used: sense primer (5'-CTTCCCTACTTCACAAGTCC-3'), corresponding to nucleotides 3,206-3,225; antisense primer (5'-GACCACAGTGAGGAATGTCC-3'), corresponding to nucleotides 7,187-7,206 (GenBank Accession No. M26745). For the 490-bp IGF-1 probe, the following oligonucleotides were used: sense primer (5'-CATGTCGTCTTCACATCTCTTC-3'), corresponding to nucleotides 42-63; antisense primer (5'-GGCTCCTCCTACATTCTGTA-3'), corresponding to nucleotides 415-434 (GenBank Accession No. D00698). For the 428-bp GAPDH probe, the following oligonucleotides were used: sense primer (5'-CACCATCTTCCAGGAGCGAG-3'), corresponding to nucleotides 239-258; antisense primer (5'-ACAGCCTTGGCAGCACCAGT-3'), corresponding to nucleotides 648-667 (GenBank Accession No. AF106860). All specific amplified fragments were purified using QIAquick Spin (QIAgen), labeled with [-³²P]dATP and [-³²P]dGTP (Amersham Pharmacia Biotech) by a random priming procedure (10), and used as probes (see Northern blotting) at the specific activity of at least 1 × 10⁹ counts · min¹ · µg¹.

Northern blotting. Total RNA was extracted from the homogenized left ventricular myocardium using TRIzol Reagent (GIBCO-BRL Life Technologies) according to the manufacturer's procedures. Northern blots (Hybond-N+, Amersham Pharmacia Biotech) were performed using 1.2% agarose gel electrophoresis under denaturing conditions according to standard procedures (35). Hybridizations were carried out at 65°C in Rapid-hyb buffer (Amersham Pharmacia Biotech) with the specific cDNA probes, followed by washing at various final stringencies until radioactive background was negligible, according to the manufacturer's procedures. Specifically, blots were sequentially hybridized, stripped, and reprobed with the TNF-, IL-6, and IGF-1 probes and finally, to correct for potential differences in the amount of RNA loaded and transferred, with the cDNA probe for GAPDH. Thus exposure time for each hybridization (70°C with intensifying screens) was chosen within the linear response range of the radiographic film (Kodak XAR), and quantitative evaluation was approached by normalizing the scanning densitometric intensity of the specific autoradiograms for TNF-, IL-6, and IGF-1 to the hybridization signal obtained with GAPDH from the same lane (imaging densitometer model GT-670, Bio-Rad). The sizes of the hybridized messengers were estimated using the 28S and the 18S rRNA bands as standards. Total RNAs from endotoxin-stimulated (10 µg/ml for 8 h) rat macrophage cells and the normal rat liver were used as positive controls for the TNF-, IL-6, and IGF-1 cDNA probes, respectively.

Western blotting. Western blotting experiments were performed according to standard procedures (13). Briefly, the powdered left ventricular myocardium was homogenized in JS lysis buffer [50 mM HEPES (pH 7.5), 150 mM NaCl, 1% glycerol, 1% Triton X-100, 1.5 mM MgCl₂, and 5 mM EGTA] containing 1 mM phenylmethylsulfonyl fluoride and protease inhibitor cocktail (complete Protease Inhibitors Cocktail Tablets, Roche Molecular Biochemicals). Lysates were clarified by centrifugation at 10,000 g, and protein concentration was estimated by a modified Bradford assay (Bio-Rad). Western blots (Protran, Schleichen and Schuell) were performed using 15% SDS-PAGE under reducing conditions. High-range Rainbow molecular weight markers (Amersham Pharmacia Biotech) were run simultaneously with protein homogenates. Ponceau red staining of blots was used to assess total protein quality. Primary antibodies were rat reactive affinity-purified polyclonal IgG specific for TNF- (1:250), IL-6 (1:200), IGF-1 (1:100), TNF receptor (TNF-R)1 (1:150), IL-6 receptor (IL-6-R) (1:50), and IGF-1 receptor (IGF-1-R) (1:100) (Santa Cruz Biotechnology). Immunoblots were stained with the appropriate secondary antibodies (Santa Cruz Biotechnology) and revealed with the enhanced chemiluminescence system (Amersham Pharmacia Biotech). Primary antibody for extracellular signal-regulated kinase (ERK)1 (1:500, Santa Cruz Biotechnology) was used to correct for potential differences in the amount of total protein loaded and transferred. Thus quantitative evaluation of specific protein expression was approached by scanning densitometry (imaging densitometer model GT-670, Bio-Rad) of the exposed bands normalized to the expression of ERK1 from the same lane. Antibody positive controls were total proteins from endotoxin-stimulated (10 µg/ml for 8 h) rat macrophage cells (for TNF-/TNF-R1 and IL-6/IL-6-R) and the normal rat liver (IGF-1/IGF-1-R).

Immunohistochemical analysis. Immunohistochemistry was performed on 4-µm-thick sections from formalin-fixed paraffin-embedded blocks of the left ventricle cross sectioned perpendicularly to their major axis. Briefly, tissue immunoreactivity was intensified by microwave treatment in 10 mM sodium citrate buffer, pH 6.0, and endogenous peroxidase activity was quenched with 0.3% H₂O₂ in 90% methanol. Sections were incubated 1 h at 37°C with rat reactive affinity-purified polyclonal IgG specific for TNF- (1:150), IL-6 (1:100), IGF-1 (1:100), TNF-R1 (1:50), IL-6-R (1:25), and IGF-1-R (1:150) (Santa Cruz Biotechnology), followed by 30-min incubation at room temperature with the appropriated secondary biotinylated antibody (Santa Cruz Biotechnology). The presence of the specific protein was revealed by incubating slides with streptavidin-horseradish peroxidase complex (30 min) and then by adding diaminobenzidine (DAB) chromogen as peroxidase substrate for 1 min (Dako LSAB kit, Dako A/S). Specifically, the duration of incubation with DAB was empirically chosen as the time resulting in the optimal ratio of specific protein signal to unspecific staining as determined by monitoring the peroxidase reaction under a light microscope. Slides were weakly counterstained with Harris's hematoxylin and permanently mounted with a synthetic mounting medium. Control negative sections were obtained for each left ventricular cross section by the same procedure described above except for the omission of the primary antibody incubation. All sections were examined with a light microscope at ×250 and ×500 magnification.

Statistical analysis. The data are presented as means ± SE of three independent experiments for each time and level of stretch. One-way ANOVA followed by the Newman-Keuls post hoc test was used for statistical comparisons of the differences in aequorin light signals, lactate production, and changes in developed pressure. Two-way ANOVA followed by the Newman-Keuls post hoc test was used for statistical comparisons of the differences in mRNA and protein expression for each of the candidate molecules at each time and level of stretch. The threshold for statistical significance was set at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Compared with hearts perfused under control conditions (unstretched myocardium, 5 mmHg), no significant changes were detected during 60 min of moderate (15 mmHg) and severe (35 mmHg) stretch in epicardial diastolic [0.07 ± 0.05, 0.09 ± 0.02, and 0.08 ± 0.03 nA, respectively, P = not significant (NS)] and systolic (0.40 ± 0.09, 0.49 ± 0.07, and 0.42 ± 0.08 nA, respectively, P = NS) aequorin light signals or in lactate production and lactate dehydrogenase release (data not shown). Moreover, compared with unstretched hearts, no significant changes were measured before or after 60 min of moderate and severe stretch in developed pressure (percent difference: 2.86, 4.17, and 10.14, respectively, P = NS). Taken together, these findings indicated that in our experimental settings neither relevant ischemia nor mechanical tissue damage have occurred.

Figure 1 shows representative autoradiograms (A) and corresponding densitometry (B) of total myocardial RNA by Northern blotting for TNF-, IL-6, and IGF-1. In unstretched/unperfused hearts (data not shown) and throughout control perfusion (unstretched myocardium), TNF- was undetectable, whereas both IL-6 (~1.5 kb) and IGF-1 (~7.5 kb) were constitutively and stably expressed. Induction of moderate stretch was accompanied by de novo TNF- (~1.7 kb) expression and significant upregulation of both IL-6 and IGF-1 levels, which appeared maximal after 10 min and remained near identical up to 60 min. In comparison, under severe stretch, a further and progressive increase in TNF-, IL-6, and IGF-1 levels was observed up to 20 min, although, at that time, the enhancement of gene expression was less evident for IL-6 (~1.2-fold increase, P = NS) than for both TNF- and IGF-1 (~3.1- and ~2.5-fold increase, respectively, both P < 0.005). After 20 min of severe stretch, whereas TNF- levels continued to increase, approaching a plateau between 40 and 60 min (~4.7-fold increase vs. the corresponding moderate stretch, P < 0.005), both IL-6 and IGF-1 levels progressively decreased. In particular, at 60 min of severe stretch, both IL-6 and IGF-1 levels were significantly lower than those observed under moderate stretch (P < 0.005 and P < 0.05, respectively), and IL-6 levels were even significantly lower than those observed under control conditions (P < 0.005).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   A: representative Northern blots for tumor necrosis factor (TNF)-, interleukin (IL)-6, insulin-like growth factor (IGF)-1, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in control (unstretched myocardium: 5 mmHg), moderately (15 mmHg), and severely (35 mmHg) stretched myocardium (see METHODS for details). Twenty-five micrograms of total RNA were loaded for each lane. The first and second lane of each panel display signals obtained after 60 min in control and moderately stretched myocardium, respectively, and they are representative of near-identical signals observed after 10, 20, and 40 min. Probe control lane refers to endotoxin-stimulated (10 µg/ml for 8 h) rat macrophage cells (for TNF- and IL-6) and the normal rat liver (for IGF-1). B: densitometric analysis of autoradiographic bands, expressed as means ± SE of 3 independent experiments for each value of time and stretch (ST). The degree of induction (fold induction) was referred to as an arbitrary number, defined as 1, assigned to the level of expression estimated in control (C; for IL-6 and IGF-1) or in 10-min moderately stretched (for TNF-) myocardium. *P < 0.05 and **P < 0.005 vs. control; P < 0.05 and P < 0.005 vs. the corresponding 15-mmHg group.

Figure 2 shows representative autoradiograms (A) and corresponding densitometry (B) of total myocardial protein by Western blotting for TNF-, IL-6, and IGF-1. In unstretched/unperfused hearts (data not shown) and throughout the control perfusion (unstretched myocardium), TNF- was undetectable, whereas both IL-6 and IGF-1 were constitutively and stably expressed. Moderate stretch resulted in de novo TNF- production, which was maximal after 10 min and remained stable up to 60 min, whereas it had only a marginal effect on IL-6 and IGF-1 expressions. When the intraventricular balloon was inflated at the diastolic pressure of 35 mmHg, the effect of mechanical stretch on myocardial TNF-, IL-6, and IGF-1 protein production was more pronounced, and the patterns of protein expression paralleled those of the corresponding mRNAs over time. Myocardial immunohistochemistry showed that the peptides were focally expressed at the cardiomyocyte level (cytoplasmatic staining) and mostly within the subendocardial wall layer (Fig. 3).

View larger version (28K): [in this window] [in a new window]   Fig. 2.   A: representative Western blots for TNF-, IL-6, IGF-1, and extracellular signal-regulated kinase (ERK)1 in control (unstretched myocardium: 5 mmHg), moderately (15 mmHg), and severely (35 mmHg) stretched myocardium (see METHODS for details). Fifty micrograms of total protein were loaded for each lane. The first and second lane of each panel display the signals obtained after 60 min in control and moderately stretch myocardium, respectively, and they are representative of near-identical signals observed after 10, 20, and 40 min. Antibody control lane refers to endotoxin-stimulated (10 µg/ml for 8 h) rat macrophage cells (for TNF- and IL-6) and the normal rat liver (for IGF-1). B: densitometric analysis of autoradiographic bands, expressed as means ± SE of 3 independent experiments for each value of time and stretch. The degree of induction was referred to as an arbitrary number, defined as 1, assigned to the level of expression estimated in control (for IL-6 and IGF-1) or in 10-min moderately stretched (for TNF-) myocardium. *P < 0.05 and **P < 0.005 vs. control; P < 0.05 and P < 0.005 vs. the corresponding 15-mmHg group.

View larger version (86K): [in this window] [in a new window]   Fig. 3.   Representative photomicrographs showing myocardial TNF-, IL-6, and IGF-1 immunoreactivity (see METHODS for details). A: focal cytoplasmatic cardiomyocyte TNF- protein immunostaining in 60-min severely stretched myocardium (subendocardial wall layer). B: focal cytoplasmatic cardiomyocyte IL-6 protein immunostaining in 20-min severely stretched myocardium (subendocardial wall layer). C: focal cytoplasmatic cardiomyocyte IGF-1 protein immunostaining in 20-min severely stretched myocardium (subendocardial wall layer). Peroxidase/diaminobenzidine (DAB) staining; original magnification of all photomicrographs, ×500.

Figure 4 shows representative autoradiograms (A) and corresponding densitometry (B) of total myocardial protein by Western blotting for TNF-R1, IL-6-R, and IGF-1-R. In unstretched/unperfused hearts (data not shown) and throughout the control perfusion (unstretched myocardium), TNF-R1 and IGF-1-R were constitutively and stably expressed. No significant changes in their expression occurred after myocardial stretch regardless of the time and the level of the balloon inflation, although TNF-R1 levels tended to be progressively increased under severe stretch over time (P = 0.072 vs. control at 60 min). IL-6-R was not detected in unstretched/unperfused hearts (data not shown) or throughout the control perfusion (unstretched myocardium), nor it was expressed in the myocardium from moderately and severely stretched hearts until 60 min. Myocardial immunohistochemistry showed that TNF-R1 and IGF-1-R were focally expressed at the cardiomyocyte level (membrane staining) and diffusely within the myocardial wall (Fig. 5). In parallel with Western blot analyses, TNF-R1 tended to have a more intense immunostaining after 60 min of severe stretch, which mainly localized within subendocardial wall layer at cardiomyocyte levels. No specific myocardial IL-6-R immunoreactivity was found in control nor it was detected in the stretched myocardium.

View larger version (32K): [in this window] [in a new window]   Fig. 4.   A: representative Western blots for TNF receptor (TNF-R)1, IL-6 receptor (IL-6-R), IGF-1 receptor (IGF-1-R) and ERK1 in control (unstretched myocardium: 5 mmHg), moderately (15 mmHg), and severely (35 mmHg) stretched myocardium (see METHODS for details). Fifty micrograms of total protein were loaded for each lane. The first and second lane of each panel display the signals obtained after 60 min in control and moderately stretch myocardium, respectively, and they are representative of near-identical signals observed after 10, 20, and 40 min. Antibody control lane refers to endotoxin-stimulated (10 µg/ml for 8 h) rat macrophage cells (for TNF-R1 and IL-6-R-) and the normal rat liver (for IGF-1-R). B: densitometric analysis of autoradiographic bands (TNF-R1 and IGF-1-R), expressed as means ± SE of 3 independent experiments for each value of time and stretch. The degree of induction was referred to as an arbitrary number, defined as 1, assigned to the level of expression estimated in the control myocardium.

View larger version (90K): [in this window] [in a new window]   Fig. 5.   Representative photomicrographs showing myocardial TNF-R1, IL-6-R, and IGF-1-R immunoreactivity (see METHODS for details). A: focal, membrane, cardiomyocyte TNF-R1 protein immunostaining in 60-min severely stretched myocardium (subendocardial wall layer). B: absence of IL-6-R protein immunostaining in 60-min severely stretched myocardium (subendocardial wall layer). Inset, IL-6-R protein immunostaining in rat transitional cell carcinoma, used as a positive control (25). C: focal, membrane, cardiomyocyte IGF-1-R protein immunostaining in 60-min severely stretched myocardium (subendocardial wall layer). Peroxidase/DAB staining; original magnification of all photomicrographs, ×500.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Major findings of the present study. The results of the present study provide the first evidence of rapid, coordinate, and differential changes in myocardial TNF-, IL-6, and IGF-1 mRNA and protein expression in response to moderate and severe acute mechanical stress in the normal rat independent of ischemia and neurohormonal interference. The divergent pattern of myocardial TNF-, IL-6, and IGF-1 expression by graded mechanical stretch, in the presence of substantially unchanged GAPDH and ERK1 levels (used as internal standards for Northern and Western blot analyses, respectively), suggests that the differential gene expression occurred as a specific result of the different magnitudes of mechanical stimulation. In particular, the similarity between the mRNA and protein time courses would be consistent with the involvement of specific transcriptional and/or posttranscriptional mechanisms. To a certain extent, this specificity is also supported by the immunohistochemical findings demonstrating that, in response to graded mechanical stretch, myocardial TNF-, IL-6, and IGF-1 expression were mainly modulated within the subendocardial wall layer. This is consistent with the expected transmural gradient of "normal" strains (fibers extension along the circumferential, longitudinal, and radial axes), as described by Omens et al. (29) in an isolated, potassium-arrested dog heart model. Specifically, three-dimensional myocardial normal strains, measured by biplane radiography of transmural sets of radiopaque beads implanted in the midanterior left ventricular free wall, increase in proportion to left ventricular end-diastolic pressure changes and from the subepicardium to subendocardium. Taken together, our observations suggest a differential relationship between mechanical forces acting on cardiomyocytes and TNF-, IL-6, and IGF-1 expression, thus complementing and extending previous observations in similar ex vivo or in vitro models of myocardial/cardiomyocyte stretch (9, 20, 30).

Possible mechanisms for changes in myocardial TNF-, IL-6, and IGF-1 expression by graded mechanical stretch. It is known that once the mechanical stimulus is received by specific mechanosensors (integrins, cytoskeleton, and sarcolemmal proteins), it is converted into three major intracellular cross-talking signal transduction pathways, i.e., the mitogen-activated protein kinase (MAPK), Janus kinase/STAT, and calcineurin-dependent pathways, which ultimately modulate gene expression through activation of disparate downstream nuclear transcription factors (14, 33). Interestingly, it has been recently reported that the activation of p38 (a member of the MAPK superfamily belonging to the stress-activated protein kinases subfamily) leads to the activation of the transcription factor nuclear factor-B (8), which is required for the induction of most cytokine genes, including TNF- and IL-6 (2). The above signal transduction pathways may be also activated in response to stretch-induced endogenous autocrine/paracrine cytokines and growth-promoting factors, which may act in a synergistic, antagonistic or permissive manner (14, 33). Given this intricate scenario, it is possible that the differential gene expression reported in the current study could have occurred as a result of multiple levels of integration between the above cross-talking signal-transduction pathways.

Potential implications. Extrapolation of the present acute in vitro observations to the chronic in vivo process of cardiac hypertrophy/remodeling and failure requires extreme caution. The two magnitudes of acute myocardial stretch used in the current report (15 and 35 mmHg of end-diastolic pressure) were chosen as representative left ventricular loads that may occur after mild-to-moderate and severe myocardial infarction commonly associated with phenotypically and prognostically different outcomes (28, 31, 32). In this scenario, our results might be relevant for two reasons. First, they suggest that cytokines and growth factors, such as TNF-, IL-6, and IGF-1, may play an important role in the orchestration and timing of stretch-induced responses within the myocardium. In particular, the hypothesis could be put forward that the initial response of the myocardium to moderately increased hemodynamic load may be characterized by the contemporary activation, among others, of endogenous autocrine/paracrine TNF-, IL-6, and IGF-1 aimed at promoting functionally adaptive cardiac hypertrophy to match the increased workload. Indeed, these peptides have shown to activate hypertrophic growth in the cardiac myocyte (15, 18, 41) and to exert cytoprotective effects within the myocardium as well (8, 24, 26, 38). Conversely, under conditions of excessive hemodynamic stimulation, mechanical stress might subsequently promote and sustain an unbalanced milieu of these peptides within the myocardium, with enhanced generation of TNF- accompanied by a simultaneous reduction of IL-6 and IGF-1. This divergent expression, by altering the local balance between growth and death signals and interfering with extracellular matrix composition, might critically contribute to the development of the heart failure phenotype. Importantly, increased TNF- and decreased IGF-1 expression within the stretched myocardium might also contribute to depress myocardial contractility. In fact, whereas IGF-1 displays distinct positive inotropic properties by sensitizing the myofilaments to intracellular [Ca²⁺] and/or by increasing the availability of intracellular [Ca²⁺] to the myofilaments (6, 21), TNF- decreases the levels of peak intracellular [Ca²⁺] during the systolic contraction sequence and/or myofilaments intracellular [Ca²⁺] responsiveness, thus depressing myocardial contractility (11, 39). This hypothesis is in line with the current thesis that the phenotypic outcome induced by cardiac hemodynamic overload (i.e., compensatory hypertrophy or failure) would be dependent on interactive changes in myocardial expression of several cytokines and growth factors and on the balance between their relatively distinct pattern of myocardial responses (16). Interestingly, our hypothesis is congruent with the recent demonstration by Neri Serneri et al. (27) of specific qualitative and quantitative changes in growth factors biosynthesis during the transition from compensated, normalizing wall stress to decompensated hypertrophy with elevated wall stress in humans with valvular heart disease. A second potentially relevant aspect of the current study is that it may serve the heuristic purpose of focusing future studies on the mechanisms that are responsible for the differential myocardial production of these peptides by graded mechanical stress, which would be of considerable theoretical and perhaps therapeutic interest as well.