INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CARDIAC HYPERTROPHY RESULTS from the interaction between mechanical forces, i.e., hemodynamic overload, and cardiac or humoral growth factors (GFs) (9, 23). Angiotensin (ANG) II, endothelin (ET)-1, and insulin-like growth factor (IGF) I are the GFs most frequently associated with both experimental and human hypertrophy (1, 3, 6, 14, 29, 39, 46). Although the involvement of these GFs has been well documented in established hypertrophy, few studies are available on their formation and time sequence in the early phases of developing hypertrophy. In aortic-banded rats, an increased cardiac expression of mRNA for preproET-1 (ppET-1) has been reported in the first 4 days (17), and a significant increase in the cardiac content of ET-1 has been found after 8 days (1). Enhanced cardiac IGF-I mRNA expression has also been observed on the seventh day after aortic banding in rats (14). Increased levels of mRNAs for IGF-I and transforming growth factor-1 were observed at 1 wk in a myocyte-enriched myocardial fraction from rats with pressure overload but did not occur with volume overload (6). Finally, no changes in mRNAs for cardiac angiotensinogen and ANG receptor AT-1 were found on the first day after the creation of volume overload in rats (19).

In these studies, no hemodynamic measurements were performed, nor was the relationship investigated between cardiac GF production and initial cardiac adaptation responses to the overload. Thus the physiological time sequence of cardiac GF formation and the relationship of these factors to the very early cardiac changes in hemodynamic overload are unknown.

To investigate just these issues, we used a model of pressure- and volume-overload hypertrophy in the pig and periodically measured (from 30 min to 7 days) the cardiac formation of ANG II, ET-1, and IGF-I as well as cardiac contractility and hemodynamic and echocardiographic changes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Study Design

The pigs were randomly divided into three groups: in a first group (n = 27), pressure overload was created by aortic banding; in a second group (n = 27), volume overload was obtained by creating an aorta-cava fistula; and the last group (n = 27) was made up of sham-operated animals. At each set experimental time (0.5, 3, 6, 12, and 24 h and 2, 3, 4, and 7 days after surgery), three aortic-banded, three aorta vena cava-shunted, and three sham-operated animals underwent echocardiographic examination and measurements of hemodynamic parameters. Midline sternotomy was then performed, and a catheter was positioned in the coronary sinus through a purse string on the right atrial appendage for blood sample drawing. Blood samples (10 ml) for the ANG, ET-1, and IGF-I assays were contemporaneously drawn from the aorta and coronary sinus. The heart was then removed, and two transmural left ventricular free-wall specimens, taken midway from base to apex, were immediately placed in liquid nitrogen for RT-PCR studies and in 10% Formalin solution for histology.

Surgical Procedures

1

1

Pressure overload was induced by banding of the ascending aorta. The chest was entered by a median sternotomy, the pericardium was opened, and the aortic root was exposed. Ascending aorta was dissected free from the main pulmonary artery and surrounded by an umbilical tape, and the periaortic fat was dissected to make a small tunnel for the passage of the band. A silicone tube (4-mm diameter, 55 ± 0.5-mm length, depending on the aorta size) was placed around the ascending aorta through the tunnel, and a 3-mm-wide polyester tape was then threaded through the tube to achieve a 60-mmHg transtenotic gradient. The sternotomy was then closed. The silicone tube protected the aortic wall from the edges of the band, which could have constituted a focal point for aortic rupture. After closure of the sternotomy and recovery from anesthesia, the animals received antibiotic and analgesic treatments (600,000 U penicillin daily and 75 mg diclofenac im, respectively) and followed a standard diet under close veterinary supervision.

Volume overload was induced by creating an aorta-cava fistula. Premedication, anesthetic procedure, postoperative analgesia, and antibiotic regimen were identical to those in the previous group. The abdomen was opened via a midline incision, and the inferior vena cava and abdominal aorta distal to the renal arteries were cleaned of fat and adventitia. A side clamp was used after systemic heparinization (3 mg/kg), and a Dacron graft (10-mm diameter) was inserted between the aorta and vena cava by use of a continuous running suture (2). The clamps were then released, hemostasis was obtained, and the abdomen was closed.

Sham animals underwent thoracotomy, instrumentation, and medication in the same way as experimental animals but were not subjected to further surgical procedures.

Left Ventricular Function and Hemodynamic Measurements

Two-dimensional and M-mode echocardiographic studies (2.5/3.5-MHz transducer, ESAOTE spa) were performed from the right parasternal area, and the studies were recorded on videotape. Freeze frames were printed, and wall thicknesses and left ventricular diastolic and systolic internal dimensions were measured according to the recommendations of the American Society of Echocardiography (35). Left ventricular mass (LVM) was calculated with the use of the validated formula (11)
where CSA (cross-sectional area) is calculated as
where D is left ventricular minor axis, and WT is wall thickness. Left ventricular volumes (V) were calculated by use of the formula (40)
where L is left ventricular long-axis dimension.

Calculations of peak systolic (PSS), end-systolic (ESS), and end-diastolic left ventricular meridional wall stress (EDS) were performed from echocardiographic recordings in combination with invasive left ventricular pressure by use of the formula (13)
where P is left ventricular pressure, and LVID is left ventricular internal dimension.

The contractile state was evaluated as the exponential constant (K_sm) of the end-systolic relation between wall stress (ESS) and the natural logarithm of the reciprocal of wall thickness [ln (1/WT)] (representing the strain), solving the equation (28)
where C is a constant and K_sm is the end-systolic stiffness constant of the myocardium, a contractile index that is independent of ventricular size and geometry and insensitive to changes in preload and afterload (28).

Measurements were analyzed independently by two experienced echocardiographers. Interobserver and intraobserver variabilities were 4.1 ± 0.5 and 2.5 ± 0.3% for cavity size and 3.7 ± 0.4 and 2.1 ± 0.3% for wall thickness, respectively.

Estimation of the Cardiac Production of GF Peptides

ANG II assay. Plasma extraction and assay of ANG were performed as previously described in detail (29). ANG II concentrations in plasma were measured with radioimmunoassay by use of specific polyclonal antibody (ITS Technogenetic). Cross-reactivities for ANG I antiserum were 98% with ANG-(2---10) nonapeptide and <1% with ANG II, ANG III, ANG-(3---8) exapeptide, or ANG-(4---8) pentapeptide (<1% for all). Cross-reactivities for ANG II antiserum were 67% with ANG III, 70% with ANG-(3---8) exapeptide, 91% with ANG-(4---8) pentapeptide, 0.1% with ANG I, and 0.2% with ANG-(2---10) nonapeptide. The concentrations were expressed in picograms per milliliter. The lowest detection limit was 1 pg/ml. Overall intra- and interassay variation coefficients were 7.7 and 13.6%, respectively.

ET-1 assay. Plasma extraction of ET-1 was performed as previously described (29). ET-1 was assayed with radioimmunoassay with the use of a specific polyclonal antibody (Peninsula Labs) that cross-reacted 7% with ET-2, 7% with ET-3, and 17% with Big ET. There was no cross-reactivity with Big ET-(22---38), -atrial natriuretic peptide, brain natriuretic peptide, ANG I, ANG II, or ANG III. The minimum detectable concentration was 0.1 pg/ml. The coefficients of intra- and interassay variations were 4 and 10%, respectively. Results were expressed as picograms per milliliter.

IGF-I assay. Plasma extraction and assay of IGF-I were performed as previously described (29). Briefly, IGF-I was extracted by use of Sep-Pak C₁₈ cartridges and measured with radioimmunoassay, using a specific polyclonal antibody (Peninsula Labs). There was no cross-reactivity with IGF-II (0.02%), epidermal growth factor (0%), growth hormone (0%), insulin (0%), or somatostatin (0%). The IGF-I recovery rate was 95 ± 2%. Intra- and interassay variabilities were 3.5 and 10.3%, respectively. The results were expressed as nanograms per milliliter. The minimum detectable concentration was 1 ng/ml.

Quantification of GF mRNA Levels

PCR primers corresponding to each respective GF sequence were purchased from Pharmacia. The sequences of primers and the cDNA sizes (in bp) are summarized in Table 1. To ensure that different amounts of PCRs on myocardial biopsies were not due to markedly different mRNA starting concentrations, PCR analysis was performed for the internal control mRNA (GAPDH) on serial twofold dilutions of cDNA for each sample. The last dilution giving a positive reaction for GAPDH was used to equalize the amount of cDNA used in each PCR. PCR reactions were performed in a DNA thermal cycler (Perkin Elmer Cetus). GAPDH and GF band densities were analyzed with the use of a computer image densitometer (Qwin; Leica). The ratio of the GF to GAPDH was determined. The densitometric GF-to-GAPDH ratio was expressed as a percentage of the values obtained in sham animals.

Localization of GFs in the Myocardium

Positive controls were obtained for each sample with the use of a cDNA probe for the housekeeping gene GAPDH to ensure that mRNA in myocardial biopsies was intact. In situ hybridization staining was performed at the same time for all specimens and at least twice on serial sections in each specimen. The presence of mRNA signals was assessed with light microscopy at ×400 magnification.

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pressure Overload

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effects of aortic banding on myocardial strain (A), cardiac stiffness constant (Ksm; B), end-systolic stress (ESS; C), and peak systolic () and end-diastolic () wall stresses (WS; D) in pigs. WT, wall thickness. Values are means ± SD. * P < 0.05 vs. baseline and vs. sham.

Cardiac GF production was quickly activated, and ANG II was the first GF produced (P < 0.01 at 3 h) (Fig. 2). ET-1 was produced soon afterward (P < 0.01 at 6 h). Both ANG II and ET-1 peaked at 12 h and were not significantly different from baseline at 48 and 72 h, respectively (Fig. 2). The early formation of these GFs was confirmed by the increase in the angiotensinogen-to-GAPDH mRNA ratio (16 times at 3 h and 45 times at 6 h vs. sham) and by the almost explosive overexpression of the ppET-1 gene (ppET-1-to-GAPDH mRNA ratio; 112 times at 6 h and 51 times at 12 h vs. sham) (Fig. 3). Neither angiotensinogen nor ppET-1 mRNAs differed from sham at 48 h, but a second overexpression of ppET-1 mRNA and relative peptide was detectable at 7 days (Figs. 2 and 3). Hybridization studies showed that angiotensinogen and ppET-1 mRNAs were exclusively expressed by cardiomyocytes (Fig. 4).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Cardiac formation of angiotensin (ANG) II (), endothelin (ET)-1 (), and insulin-like growth factor (IGF) I () in pigs with aortic banding (A) and aorta-cava fistula (B). Values are means ± SD. * P < 0.05 vs. baseline and vs. sham.

View larger version (36K): [in this window] [in a new window]   Fig. 3.   Top: expression of angiotensinogen (AGTN), preproET-1 (ppET-1), IGF-I, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in heart homogenates from pressure (left)- and volume (right)-overloaded pigs. For representative RT-PCR experiments, L indicates DNA molecular size markers. Bottom: mean changes in the densitometric ratio to GAPDH for AGTN (), ppET-1 (), and IGF-I () in pressure (left)- and volume (right)-overloaded pigs. GF, growth factor. GF/GAPDH, GF-to-GAPDH ratio. * P < 0.05 vs. sham.

View larger version (166K): [in this window] [in a new window]   Fig. 4.   Expression of mRNA for AGTN (top), ppET-1 (middle), and IGF-I (bottom) at in situ hybridization in left ventricular free wall from controls (0 h) and aortic-banded pigs (6 and 48 h). Positive mRNA signals were present in cardiomyocytes at 6 h for AGTN and ppET-1 and at 48 h for IGF-I.

Cardiac IGF-I formation increased later than ANG II and ET-1 formation, because the aorta-coronary sinus gradient for IGF-I increased (P < 0.05) 24 h after surgery and remained high until the end of the observation period (Fig. 2). IGF-I gene expression was upregulated at 12 h (3 times, P < 0.05 vs. sham) and then progressively increased (37, 33, and 30 times; P < 0.01 vs. sham at 24, 48, and 96 h, respectively) (Fig. 3). Hybridization studies showed that IGF-I mRNA was again exclusively expressed by cardiomyocytes (Fig. 4).

Volume Overload

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effects of aorta-cava fistula on myocardial strain (A), Ksm (B), ESS (C), and peak systolic () and end-diastolic () WS (D) in pigs. Values are means ± SD. * P < 0.05 vs. baseline and vs. sham.

In volume overload too, ANG II was the first GF produced. The increase in this factor was, however, lower than in pressure overload, and ANG II significantly increased only at 12 h (P < 0.01) (Fig. 2). At RT-PCR, the angiotensinogen gene was expressed at 6 h (6.7 times, P < 0.01 vs. sham) and disappeared at 12 h (Fig. 3). No significant upregulation of ppET-1 and the relative peptide was detectable at any experimental time. In contrast, cardiac IGF-I production increased at 24 h (P < 0.01 vs. sham) and remained high in the following periods (Fig. 2). The IGF-I gene was overexpressed at 12 h, with an IGF-I-to-GAPDH ratio of 14 times versus sham (P < 0.01), and it remained high up to 1 wk (Fig. 3). Hybridization studies showed that angiotensinogen and IGF-I mRNA were expressed exclusively by cardiomyocytes (Fig. 6).

View larger version (169K): [in this window] [in a new window]   Fig. 6.   In situ hybridization for AGTN (top), ppET-1 (middle), and IGF-I (bottom) mRNAs in left ventricular free-wall specimens from controls (0 h) and volume-overloaded pigs (6 and 48 h). Positive mRNA signal was present in cardiomyocytes at 6 h for AGTN and at 48 h for IGF-I. No ppET-1 mRNA expression was detectable at any experimental time.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present study describes for the first time the time sequence of gene activation of some cardiac GFs, i.e., ANG II, ET-1, and IGF-I, and their relationship with cardiac adaptation responses to hemodynamic overload in the initial phase of cardiac hypertrophy due to acute overload.

Cardiac and hemodynamic changes after aortic banding or aorta-cava fistula creation were substantially similar to those reported in previous studies (5, 7, 42). The abrupt increase in pressure overload resulted in the immediate decrease (at 0.5, 3, and 6 h) in myocardial contractility, with increased systolic and diastolic wall stresses, whereas heart rate increased poorly (+13 and +9% at 3 and 6 h, respectively). The decrease in contractility was associated with overexpression of angiotensinogen mRNA (at 3 h) and ppET-1 mRNA (at 6 h) and with increased formation of the relative peptides. In contrast, myocardial contractility did not significantly change in volume overload, and cardiac adaptation to hemodynamic overload was mainly supported by the increase in heart rate (+49 and +42% at 3 and 6 h, respectively), with consequent augmented cardiac output. Diastolic stress increased, whereas systolic stress did not. These mild cardiac and hemodynamic changes were associated with only marginal increases (at 6 h) in angiotensinogen mRNA expression, there being no changes in ppET-1 mRNA expression or in ET-1 formation. Thus the different cardiac adaptation responses to abrupt pressure or volume overload were associated with, and probably affected by, the different formations and time sequences of cardiac GFs.

The functional meaning of the increased ANG II and ET-1 formation in this very early phase of myocardial hypertrophy development remains to be determined. Participation of ANG II and ET-1 in the development of hypertrophy appears unlikely, both because the increased ANG II and ET-1 formation was limited to several hours and because a number of recent studies have indicated that ANG II is not an essential factor for the development of hypertrophy (15, 20, 30, 33, 45). The increase in ET-1 formation occurring in this phase is also unlikely to play a direct role in the development of hypertrophy, because it was short lasting, and a second increase in ET-1 formation occurred at 6-7 days, i.e., from 2 to 3 days after the return to baseline of the first early peak of ppET-1 mRNA expression. Both ANG II and ET-1 are endowed with potent inotropic activity (4, 12, 25, 27); hence, they may play an important role in supporting cardiac contractility and preventing stroke volume failure as a consequence of the abruptly increased afterload. This hypothesis is suggested by the notable and short-lasting increase in ANG II and ET-1 formation, by the close temporal relationship of their increased formation with the recovery of myocardial contractility, and by the absence of angiotensinogen and ppET-1 mRNA overexpression in volume overload, where cardiac contractility did not decrease. Experimental studies also bear out an inotropic functional significance of cardiac ANG II and ET-1 production. Reduced cardiac contractility has been reported in response to intracoronary administration of the renin inhibitor remikeren in intact pigs (41). Likewise, acute application of an endothelin receptor antagonist decreased myocardial contractility in rats with depressed contractility, but did not in normal rats (36).

ESS seems to be the major cause of the early angiotensinogen gene activation and ANG II increase. In the present study, wall stress measurements were obtained by cardiac catheterization under anesthesia, and the measurements of volume, radius, and wall thickness of the left ventricle were performed by well-trained personnel; consequently, the wall stress values may be taken as reliable. The close temporal relationship between ESS increase and angiotensinogen mRNA overexpression together with the simultaneous return of ANG II and ESS to baseline value, despite PSS and EDS remaining high, indicates that ESS plays a major role in inducing ANG II formation. In vitro studies have demonstrated that mechanical stretching of myocytes resulted in acute autocrine ANG II secretion; upregulation of angiotensinogen, renin, and ANG-converting enzyme (ACE) genes; and increased expression of angiotensinogen, renin protein, and ACE-like activity (24, 26, 34).

The almost explosive overexpression of ppET-1 mRNA, which peaked at 6 h, and the close temporal relationship between angiotensinogen and ppET-1 mRNA overexpression indicate the combined action of the increases in ESS and ANG II generation as the mechanism that might be responsible for ET-1 gene activation. There is evidence that mechanical stretching of isolated myocytes associated with ANG II quickly induces ppET-1 mRNA overexpression and ET-1 formation (17, 21, 22, 43). Moreover, indirect evidence that ET-1 formation is attributable to ESS and ANG II increases was also provided by the absence of an ET-1 increase in volume overload, where ANG II generation was significantly lower than in pressure overload and ESS did not increase.

IGF-I and ET-1 appear to be important for the development of hypertrophy but are selectively induced by the different hemodynamic overloads. IGF-I mRNA overexpression was detectable at 24 h, remained persistently high, and preceded the increase in posterior wall and septum thickness. The growth-promoting activity of IGF-I has been well documented in experimental and human studies. IGF-I directly induces hypertrophy in isolated myocytes (18) and enhances ventricular hypertrophy and myocyte function with no or only a mild increase in myocardial fibrosis in adult rat (8, 10). IGF-I mRNA upregulation has been reported in pressure- and volume-overloaded hypertrophy both in humans (29) and in experimental models of hypertrophy (6, 14, 16). The second overexpression of ppET-1 mRNA occurred later than that of IGF-I mRNA, being detectable on the seventh day, at the end of the experimental observation period. Although we did not follow cardiac ET-1 formation beyond day 7, there are numerous studies demonstrating the involvement of ET-1 in both human (29) and experimental hypertrophy due to pressure overload (1, 37, 46).

IGF-I and ppET-1 mRNA overexpression occurred when contractility recovered, ESS normalized, and the increased ANG II levels and early ET-1 peak returned to baseline values, thus pointing to the persistent increase in workload rather than ESS or ANG II as the main cause of the IGF-I upregulation and the delayed ET-1 gene activation. Both the evaluation of IGF-I and ppET-1 mRNAs and the measurements of the active peptides in coronary sinus blood showed that IGF-I formation had increased in both volume and pressure overload, unlike ppET-1 gene activation and ET-1 formation, which were selectively induced only by pressure overload. It is noteworthy that IGF-I gene overexpression preceded the second ET-1 gene increase, suggesting that the IGF-I formation increase is the primary, nonselective cardiac response to the increase in workload, whereas a more selective stimulus, such as pressure overload, is required to enhance ET-1 formation. These findings confirm that myocardial load is a fundamental factor in the characterization of GF synthesis and hence of myocyte growth response (6, 24, 31, 32, 44).

Hybridization studies showed that ppET-1 and IGF-I mRNAs as well as angiotensinogen mRNA were expressed exclusively on myocytes. It is important to note that ppET-1 and IGF-I mRNAs were overexpressed by myocytes, notwithstanding the ESS increase, thus indicating that this increase does not in itself inhibit ET-1 and IGF-I formation by myocytes. This finding contributes to clarification of the relationship between the increased ESS and the decreased or even absent ET-1 and IGF-I production by myocytes in human hypertrophy with high ESS (29); it demonstrates that reduced ET-1 and IGF-I formation is due to the incapacity of myocytes to maintain sufficiently high ET-1 and IGF-I formation, which has been found to contribute to compensatory hypertrophy (29). Thus reduced ET-1 and IGF formation, resulting in decreased contractility, appears to be the cause of the ESS increase in inadequate human hypertrophy. However, it should be noted that what is described in the present study is due to a sudden-onset insult to the heart, characteristic of these models, and not to a slow cardiac overload, as usually occurs in the clinical situation.

In conclusion, present results show a close relationship between acute cardiac adaptation responses to sudden hemodynamic overload and cardiac GF formation. The comparison between pressure and volume overloads, both able to increase left ventricular mass after a week, reveals that in a very early phase of developing hypertrophy, different GFs (ANG II and ET-1) are produced that mainly contribute to supporting myocardial contractility. Only in a later phase are GFs (IGF-I and ET-1) formed that contribute to increasing ventricular mass, and their formation is selectively induced by the type of hemodynamic overload applied.