INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

PRESSURE OVERLOAD produces a concentric hypertrophy in which cardiac myocytes display an increase in cell diameter with the addition of new myofibrils in parallel. In contrast, volume overload induces eccentric hypertrophy with cardiac chamber dilatation. Myocytes derived from dilated hearts exhibit an increase in cell length, reflecting the addition of new sarcomeric units in series.

Cardiotrophin-1 (CT-1) is a unique cytokine that induces longitudinal cardiac myocyte hypertrophy with the assembly of sarcomeric units in series in in vitro studies (15, 21). On the other hand, other growth factors such as endothelin-1, angiotensin II, and -adrenergic agonists promote a uniform enlargement of cell size, resulting from the assembly of new sarcomeric units in parallel. Recently, we and others (1, 12) reported that CT-1 expression is increased in the left ventricle of pacing-induced congestive heart failure in dogs or in the ventricles after myocardial infarction in rats. These two models show eccentric cardiac hypertrophy with ventricular dilatation.

To examine whether CT-1 participates in the progression from left ventricular hypertrophy (LVH) to congestive heart failure (CHF), we used Dahl salt-sensitive (DS) rats, which develop systemic hypertension with a high-salt diet and show concentric LVH at the age of 11 wk and CHF at the age of 16-18 wk (10). We defined the expression levels of CT-1, leukemia inhibitory factor (LIF), and gp130 during transition from LVH to CHF using this novel animal model and examined the correlation of the expression levels of CT-1 with the left ventricular dimension in these two different phases of cardiac hypertrophy.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Experimental animals and assessment of left ventricular geometry and function. Male inbred DS and Dahl salt-resistant (DR) rats were used for the experiments. They were obtained from Eisai (Tokyo, Japan) and were fed an 8% NaCl (high salt) diet after 6 wk of age.

Isolation of rat ventricular myocytes. Single left ventricular myocytes were obtained from rats using enzymatic digestion as previously described (17). Briefly, the dissected heart was mounted on a Langendorff apparatus and then perfused with Tyrode solution for ~2 min, followed by nominally Ca²⁺-free Tyrode solution for ~3 min at 37°C. The heart was then perfused with Ca²⁺-free Tyrode solution containing 0.4 mg/ml collagenase (type I, Sigma; St. Louis, MO) and 0.15 mg/ml protease (type X, Sigma) for 25 min and rinsed with a high-K, low-Cl "KB solution." The left ventricle was cut into small pieces, and the dispersed cells were filtered through a 105-µm stainless mesh and stored in DMEM containing 1.8 mM Ca²⁺ at room temperature (22-25°C) until use. Cell isolation routinely yielded 40-70% of viable rod-shaped cells. The image of a single ventricular myocyte was visualized under an inverted microscope (Axiovert 135, Carl Zeiss; Jena, Germany) using standard bright-field illumination, and the cell size was measured.

Competitive RT-PCR. Total cellular RNA was extracted from the left ventricle using the acid guanidinium thiocyanate-phenol-chloroform method and was quantified spectrophotometrically by absorbance at 260 nm. Total cellular RNA from each sample was reverse-transcribed into cDNA with the following components: 1.5 µg total RNA, 10 mM Tris · HCl (pH 8.3), 5 mM MgCl₂, 50 mM KCl, 1 mM each dNTP, 1 µM antisense primer, 0.5 units RNase inhibitor, and 0.25 units avian myeloblastosis virus reverse transcriptase XL (Takara Shuzo; Kyoto, Japan). The mixture was incubated at 42°C for 30 min, heated to 95°C for 5 min, and flash-cooled to 4°C. The cDNA samples were subjected to PCR amplification using primers specific for CT-1 (sense, 5'-TCT ATG GCG AGT GGG TGA GC-3'; antisense, 5'-AGC AAG CAA GCA AAG AAA GA-3') (11), gp130 (sense, 5'-CAT CAA CAG AAC GGC ATC CAG-3'; antisense, 5'-TCA CTT TAT CCA CGG GGT CAA-3') (20), and LIF (sense, 5'-TGC CCT CTT TAT TTC CTA TT-3'; antisense, 5'-CAC CGC ACT AAT GAC TTG-3') (18). PCR amplification of constitutively expressed glyceraldehyde-3-phosphate dehydrogenase cDNA was used as a measure of the amount of input RNA (6). DNA fragments that shared the same primer template sequence with the target cDNA but contained a completely different smaller intervening sequence were prepared and used as DNA internal standards (mimics) (13). Aliquots of sample cDNA mixed together with serial dilutions of DNA mimics were coamplified as templates in the presence of primer pairs. PCR was performed in a total volume of 25 µl containing cDNA made from 0.25 µg total RNA, diluted DNA mimics, 10 mM Tris · HCl (pH 8.3), 50 mM KCl, 2.5 mM MgCl₂, 0.02 µM primers, and 0.5 units Taq polymerase in a PCR Thermal Cycler (Takara Shuzo). The amplification step involved 30 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s, and extension at 72°C for 1 min. The PCR reaction products (20 µl) were analyzed via electrophoresis through 1.5% agarose gels containing 0.5 µg/ml ethidium bromide, and the results were quantified by scanning densitometry using NIH Image computer software (Bethesda, MD).

Western blot analysis. Lysates (60 µg) from rat ventricles were separated by 12% SDS-PAGE for CT-1 and 7.5% SDS-PAGE for gp130, and the separated proteins were transferred to a polyvinylidene difluoride membrane (Immobilion-P, Millipore) with transfer buffer (25 mM Tris, 192 mM glycine, and 10% methanol). The membranes were blocked with 5% skim milk and immunoblotted with anti-CT-1 (1) or gp130 antibody (Santa Cruz Biotechnology) using an enhanced chemiluminescence detection system (Amersham) according to the manufacturer's instructions. The results were quantified by scanning densitometry as described above.

Statistics. Values are presented as means ± SE. Statistical comparisons among the four groups were performed by one-factor ANOVA with post hoc comparisons by Fisher's protected least-significant difference test. Relationships between two variables were tested by linear regression analysis. Significance was taken as P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Somatic and cardiac growth at 11 and 17 wk. The left ventricular weights of DS rats were significantly increased compared with those of DR rats at both 11 and 17 wk, whereas the body weights of DS rats were significantly decreased compared with DR rats at both times. The ratio of left ventricular weight to body weight or tibial length of DS rats was significantly increased compared with that of DR rats at 11 and 17 wk. The systolic blood pressure of DS rats was significantly higher than that of DR rats at LVH and CHF stages (Table 1). However, the blood pressure of DS rats tended to decrease along with progression of CHF after 17 wk in this model (10). DS rats showed labored respiration and loss of activity at 17 wk. Autopsy of these rats revealed massive pulmonary congestion.

Echocardiographic study. Left ventricular geometry and function were examined by echocardiography (Table 2). End-diastolic and end-systolic left ventricular dimension were decreased, and posterior wall thickness and relative wall thickness were increased in DS rats compared with those of DR rats at 11 wk. Fractional shortening and wall stress were maintained within the normal range in DS rats at 11 wk. On the other hand, left ventricular diameter in both the end diastole and end systole in DS rats were increased at 17 wk compared with those of DR rats. Relative wall thickness decreased in DS rats from 11 to 17 wk. Fractional shortening was decreased and wall stress was significantly increased in DS rats at 17 wk. The left ventricular mass of DS rats at 17 wk was highest among the four groups of rats. These findings are consistent with our previous report (10) and show the transition from compensatory pressure-overloaded LVH to CHF between 11 and 17 wk in this model.

Cardiomyocyte length and width. The myocyte length and width in the left ventricles of DS rats at 11 wk were greater and the ratio of length to width of cardiomyocytes was decreased compared with DR rats. The cell surface area of the myocytes was also increased in DS rats compared with that in DR rats at 11 wk (Table 3). This type of myocyte change is compatible with LVH evoked by pressure overload. At 17 wk, the myocyte length in DS rats was further increased compared with that of DS rats at 11 wk. Because cell surface area in DS rats at 17 wk was increased compared with DR rats, there still was concentric hypertrophy, which did not compensate for the decreased cardiac function.

Quantitative RT-PCR of CT-1, LIF, and gp130. CT-1 mRNA levels determined by competitive RT-PCR in the left ventricles of DS rats at 11 wk tended to be higher (P = 0.062) than those of DR rats. Moreover, mRNA levels of CT-1 in the left ventricles of DS rats at 17 wk were significantly increased compared with those of DS rats at 11 wk (Fig. 1, A and B). In contrast, there were no quantitative differences in the expression levels of LIF mRNA in the left ventricles among the four groups of rats (Fig. 1, C and D). The mRNA levels of gp130 in the ventricles of DS rats at 11 wk and 17 wk were significantly increased compared with those of age-matched DR rats, respectively. There was no difference in the expression levels of gp130 mRNA in DS rats between 11 and 17 wk (Fig. 1, E and F).

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Representative competitive RT-PCR of cardiotrophin-1 (CT-1; A), leukemia inhibitory factor (LIF; C) and gp130 (E). Densitometric analyses showed that the CT-1 mRNA level of Dahl salt-sensitive (DS) rats at 17 wk (17w) was increased compared with DS rats at 11 wk (11w) and Dahl salt-resistant (DR) rats at 17 wk (B). There was no difference in LIF mRNA levels among the 4 groups of rats (D). gp130 mRNA levels of DS rats at 11 and 17 wk were significantly higher compared with age-matched DR rats (F). n = 6 rats/group. NS, not significant.

Western blotting of CT-1 and gp130. Because the mRNA levels of CT-1 and gp130 were increased in the ventricles of DS rats, we next determined the protein levels of both proteins. In the six DR rats, the mean value for the density of CT-1 or gp130 protein band was defined as onefold. The protein level of CT-1 in DS rats at 17 wk was significantly higher than that of DS rats at 11 wk and DR rats at 17 wk (Fig. 2, A and B). The gp130 protein levels in DS rats were significantly increased compared with age-matched DR rats, and there was no difference in gp130 protein levels in DS rats between 11 and 17 wk (Fig. 2, C and D). These protein levels were parallel to the mRNA levels in each protein.

View larger version (40K): [in this window] [in a new window]   Fig. 2.   Representative Western blotting of CT-1 (A) and gp130 (C). Densitometric analyses showed that the CT-1 protein level of DS rats at 11 wk tended to be increased compared with DR rats. The CT-1 protein level was highest in the ventricle of DS rats at 17 wk (B). The protein levels of gp130 of DS rats at 11 and 17 wk were increased compared with age-matched DR rats (D). n = 6 rats/group.

Correlation between the amount of CT-1 and gp130, and left ventricular dimension. We plotted the relationships between the CT-1 and gp130 protein levels and the left ventricular end-diastolic diameter for each animal. As shown in Fig. 3, there was a strong correlation between the CT-1 protein level and left ventricular end-diastolic diameter (y = 0.472x  2.396, r = 0.79, P < 0.001). However, there was no correlation between the gp130 protein level and ventricular diameter.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   The relationship between the CT-1 and gp130 protein level and left ventricular end-diastolic dimension (LVDd). There was a significant correlation between the protein level of CT-1 and LVDd (A). In contrast, there was no correlation between gp130 protein level and LVDd (B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In DS rats, salt-sensitive hypertension causes concentric LVH at the age of 11 wk, followed by CHF at 16-18 wk of age (10). The data by echocardiography presented in this study were compatible with this model.

CT-1 is a potent stimulator of myocyte hypertrophy, and CT-1-stimulated myocytes reveal the assembly of sarcomeric units in series rather than in parallel in in vitro studies (15, 21). It has been suggested that this type of myocyte hypertrophy promotes the eccentric form of hypertrophy with cardiac chamber dilatation. In the present study, we demonstrated that CT-1 but not LIF expression in the left ventricle was markedly increased during the transition from LVH to CHF in the hypertensive rat model. CT-1 was positively correlated with myocyte length and left ventricular end-diastolic diameter. Recently, we (1) reported that the expressions of CT-1 and gp130 were increased in the ventricles of rats with large myocardial infarction in which the left ventricle was rapidly and progressively dilated. Jougasaki et al. (12) reported that cardiac production of CT-1 was augmented in a canine model of pacing-induced experimental congestive heart failure and that ventricular CT-1 mRNA was correlated with LVH. Together with the results of these two reports, our findings suggest that CT-1 plays an important role in the structural remodeling that characterizes CHF. On the other hand, it has been reported that CT-1 mRNA expression was augmented in the ventricles of 12-wk-old SHRSP/Izm rats compared with age-matched control rats by Ishikawa et al. (11). Pan et al. (14) also reported that CT-1 and interleukin (IL)-6 mRNA expression were transiently increased 1 h after pressure overload by ligation of the rodent abdominal aorta. In the present study, CT-1 mRNA and protein expression also tended to be increased at the LVH stage. It has been reported that CT-1 increased angiotensinogen mRNA expression in cardiac myocytes. CT-1-induced myocyte hypertrophy was partially attenuated by an angiotensin II type 1 receptor antagonist (7). These results reported by Fukuzawa et al. (7) suggest that CT-1-induced myocyte hypertrophy is partially dependent on angiotensin II production. CT-1 may in part contribute to LVH by its direct action on myocytes or by producing angiotensin II.

gp130 is a common receptor component of IL-6-related cytokines such as IL-6, CT-1, LIF, oncostatin M (OSM), ciliary neurotrophic factor, and IL-11 (9). Among these cytokines, CT-1 and LIF induce myocyte hypertrophy at concentrations of 0.1 nM or lower. Because the expression level of gp130 was increased at the LVH stage, the expressions of IL-6-related cytokines other than CT-1 and LIF are likely to be increased at the LVH stage. Pan et al. (14) reported that IL-6 mRNA was increased after pressure overload by abdominal aortic banding in the murine model. IL-6 was also increased in the viable border zone of reperfused infarctions (8) and in the ventricles during myocardial ischemia-reperfusion (3). These lines of evidence suggest that IL-6 could be increased at the LVH stage in DS rats. However, IL-6 did not induce longitudinal myocyte hypertrophy. IL-6 may act as a cardiac depressant via production of nitric oxide, which reduces myocardial contraction (5). Timmermann et al. (19) reported that the binding epitopes of gp130 for IL-6 and IL-11 are different from those for OSM and LIF (19). These data imply the possibility that the binding epitopes of gp130 for CT-1 and the intracellular signaling pathway of CT-1 are different from those of IL-6. Thus, although gp130 is equally increased at both LVH and CHF stages, the physiological consequences may differ between these two stages.

Two structural mechanisms have been identified in the pathogenesis of ventricular dilatation in the heart. One is side-to-side slippage of myocytes, and the other is the elongation of myocytes (2). There was a tendency for wall thickness to decrease from 11 to 17 wk in DS rats, which suggested the myocyte slippage. Recently, several lines of evidence have suggested that increased matrix metalloproteinase (MMP) activities contribute to the degradation of collagen struts between myocytes or muscle cell layers, resulting in slippage of myocytes or muscle layers and ventricular dilatation (4, 16). In this rat model, MMP activities in the ventricles of DS rats were increased at 17 wk compared with those at 11 wk (Y. Iwanaga, T. Aoyama, and Y. Kihara, unpublished data). However, the myocytes were longest in DS rats in the CHF stage. Thus, in addition to myocyte slippage, myocyte elongation may partially contribute to ventricular dilatation in this model.