Decreased beta-adrenergic responsiveness following hypertrophy occurs only in cardiomyocytes that also re-express beta-myosin heavy chain

Abstract

Aims

Cardiac hypertrophy is associated with a reduction in the contractile response to beta-adrenergic stimulation, and with re-expression of foetal genes such as beta-myosin heavy chain (MHC). However, whether these two markers of pathology develop concordantly in the same individual cells or independently in different cells is not known.

Methods and results

To answer this question, we examined the beta-adrenergic response of individual beta-MHC expressing and non-expressing myocytes from hypertrophic hearts, using a previously generated mouse model (YFP/beta-MHC) in which a yellow fluorescent protein (YFP) is fused to the native beta-MHC protein allowing easy identification of beta-MHC expressing cells. Yellow fluorescent protein/beta-MHC mice were submitted to 4 weeks of transverse aortic constriction (TAC), and the contractile parameters of isolated individual myocytes in response to the beta-adrenergic agonist isoproterenol were assessed. Our results demonstrate that the decrease in isoproterenol-induced cell shortening that develops in TAC hearts occurs only in those hypertrophic myocytes that re-express beta-MHC. Hypertrophic myocytes that do not express beta-MHC have contractility indices indistinguishable from non-TAC controls.

Conclusion

These data show that the reduction of beta-adrenergic response occurs only in subsets, rather than in all myocytes, and is coincident with re-expression of beta-MHC.

Introduction

Re-expression of beta-myosin heavy chain (MHC) is a well-documented molecular marker of pathological cardiac hypertrophy.¹ Normally, the beta-MHC isoform is expressed in a developmental stage-specific manner in the mouse ventricle. Beta-MHC expression is high during embryonic and foetal stages, but decreases rapidly soon after birth when it is replaced by the expression of alpha isoform.² Beta-MHC is re-expressed in the murine ventricle following cardiac hypertrophy and during cardiac failure.^1–5 Reversal of cardiac failure is associated with beta-MHC expression returning to baseline levels.^3,6 The beta-MHC isoform has a lower ATPase activity, but greater force of contraction than the alpha isoform.⁷ Recent reports have demonstrated that even modest increases in the levels of beta-MHC are accompanied by a significant detriment in pump function of the heart including a decrease in contractility (+dp/dt).^8,9

At the cellular level, a functional hallmark of cardiac hypertrophy and heart failure is a depressed contractile response of individual cardiac myocytes to beta-adrenergic stimulation.^10,11 In myocytes from a healthy adult heart, beta-adrenergic agonists such as isoproterenol induce a short-term hypercontractile response by increasing the degree of cell shortening and by increasing the velocities of contraction and relaxation. However, myocytes sampled from the hearts of mice undergoing hypertrophy and heart failure show a markedly decreased response to beta-adrenergic stimulation, manifested as a smaller increase in cell shortening and in the velocities of contraction and relaxation.^12,13 This effect is a direct consequence of a decrease in the number of beta-adrenergic receptors as well as impaired function of the receptor.^14–18 The decrease in beta-adrenergic response is directly correlated with the severity of the cardiac phenotype, and genetic and therapeutic interventions that maintain normal beta-adrenergic receptor numbers and signalling have a substantial positive impact on heart function.^{10,12,19–23}

At the level of individual myocytes, the relationship between the re-expression of beta-MHC and reduced beta-adrenergic responsiveness is, however, not known. Thus, it is not known whether these two markers of cardiac pathology always develop together in all the cells that change or arise independently in different responding cells. To answer this question, we evaluated the contractile parameters of individual beta-MHC expressing and non-expressing cardiac myocytes isolated from the same heart exposed to a hypertrophic stimulus.

Methods

The generation of YFP/beta-MHC animals, and procedures for morphometrical analyses, confocal image acquisition, transverse aortic constriction (TAC), and myocyte isolation have been described previously.^24,25 The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1985).

Evaluation of myocyte function

Myocytes were placed in a 0.5 mL chamber with 1.8 mM Ca²⁺ Tyrode's solution at room temperature, and visualized with a Nikon inverted microscope with a solid-state CCD camera attached and displayed on a video monitor. The expression of YFP in individual cells was scored using an epifluorescence apparatus attached to the microscope. Two platinum electrodes connected to a stimulator were used to field stimulate the myocytes with pulse duration of 5 ms and a frequency of 0.5 Hz. Myocyte cell edges are enhanced and processed with a video edge motion detection system (Crescent Electronics) at a sampling rate of 240 Hz. Recordings were performed either under basal conditions or after isoproterenol stimulation (1 µm). Calibrated myocyte lengths were converted from analog to digital on-line (MacLab) and stored on computer. All myocytes were studied within 1–2 h after myocyte isolation. Data from three consecutive contractions were averaged. Approximately 15 myocytes from each category were studied. The recordings for the control group were acquired from untreated hearts which contain only non-yellow myocytes. Those for the TAC group were acquired from yellow and non-yellow myocytes isolated from single hearts. The data for basal and isoproterenol stimulation within each of the three groups (control non-yellow, TAC non-yellow, TAC yellow) were acquired from different cells and analysed with ANOVA with Newman–Keuls correction using Graphpad Prizm software. Student's t-test was used for morphometrical analyses (Figure 1).

Figure 1.

(A) Morphometrical analyses. Cross-sectional areas (sq. microns) of myocytes from animals that were untreated (UT) or that underwent TAC for 4 weeks (TAC) grouped according to their yellow fluorescence (yellow cells, Y or non-yellow cells, NY). A total of 300 cells (100 cells/heart) from three hearts/condition were analysed. Error bars denote SE. (B) Yellow fluorescent images (YFP) and Cy5 fluorescence for WGA lectin (pseudo-coloured blue) acquired from the left ventricular free walls of control (UT) or TAC-treated (TAC) hearts with 20x oil objective. (C) Yellow fluorescent confocal images of ventricular sections from untreated (UT) or TAC-treated (TAC). Overlapping images were acquired using a ×4 objective and stitched together using ImageJ. ***P < 0.0001.

Results

To determine the association between the state of beta-MHC expression and the contractile parameters of individual cells, we used mice that express a yellow fluorescent protein (YFP) fused to the amino terminal of the beta-MHC (YFP/beta-MHC) as described previously.²⁴ This YFP/beta-MHC allele is appropriately regulated during development, and accurately tracks the expression of the wild-type beta-MHC allele in response to cardiac hypertrophy in vivo. In our present experiments, cardiac hypertrophy was induced in 4-month-old heterozygous YFP/beta-MHC animals by 4 weeks of TAC. Echocardiography was conducted on these animals and the data (Table I) show that the TAC-treated animals were in a state of heat failure (%FS 29 vs. 43 for control, P < 0.01), but without overt LV dilation (no significant differences in the levels of end-systolic and end-diastolic LV diameters). Morphometric analyses (Figure 1) of myocytes from the left ventricular free walls of untreated and TAC-treated hearts show that in the TAC hearts the cross-sectional areas of myocytes are larger than myocytes from untreated hearts regardless of whether the TAC myocytes express YFP/beta-MHC or not. Furthermore, the cross-sectional areas of the TAC myocytes that express YFP/beta-MHC (column 3) are not significantly different (P = 0.6) from TAC myocytes from the same hearts that do not express YFP/beta-MHC (column 2), demonstrating that the YFP/beta-MHC expressing cells are as hypertrophic as the non-YFP/beta-MHC expressing cells. This confirms previous observations by ourselves²⁴ and by others⁹ that cellular hypertrophy is not obligatorily coupled to the re-expression of YFP/beta-MHC. Ventricular sections from a TAC or control heart were analysed for YFP fluorescence using confocal microscopy, and Figure 1C illustrates the spatial distribution of YFP expressing cells in the control (UT) or TAC-treated heart.

Table 1.

Echocardiographic parameters

	Untreated	TAC
n	9	10
LVEDD (mm)	3.1 ± 0.2	3.3 ± 0.2
LVESD (mm)	1.8 ± 0.2	2.3 ± 0.2
% FS	43 ± 3	29 ± 3*
IVSD th (mm)	1.1 ± 0.1	1.2 ± 0.1
IVSS th (mm)	1.7 ± 0.1	1.7 ± 0.1
PWD th (mm)	1.2 ± 0.1	1.2 ± 0.1
PWS th (mm)	1.8 ± 0.1	1.5 ± 0.1
HR (b.p.m.)	604 ± 57	506 ± 54

Animals were untreated or TAC treated for 4 weeks.

n, number of animals in each group; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; IVSD th, inter ventricular septum thickness diastole; FS, fractional shortening; IVSS th, inter-ventricular septum thickness systole; PWD th, posterior wall thickness diastole; PWS th, posterior wall thickness systole; HR, heart rate.

*P < 0.01.

To test whether the YFP/beta-MHC expressing and non-expressing myocytes from TAC hearts have different contractile responses to beta-adrenergic stimulation, myocytes from untreated control, and TAC-treated hearts were isolated by collagenase treatment. The contractile parameters were then assessed during basal (unstimulated) conditions or after stimulation with the beta-adrenergic agonist isoproterenol (1 µm). Under basal conditions, the percent cell shortening (%CS, Figure 2A), as well as the rates of relaxation (dl/dt, Figure 2B) and shortening (−dl/dt, Figure 2C) of YFP/beta-MHC expressing cells from TAC hearts are not significantly different from YFP/beta-MHC non-expressing TAC myocytes from the same hearts (P > 0.05), or from control hearts (P > 0.05). Isoproterenol stimulation of myocytes from control hearts induces a marked increase in the %CS (Figure 2A); as well as in the velocities of relaxation (Figure 2B) and shortening (Figure 2C). Isoproterenol stimulation of YFP/beta-MHC non-expressing cells from the TAC hearts likewise induced a marked and significant increase in the %CS, and velocities of relaxation and shortening (Figure 2A–C), and these increases are virtually identical to those of control cells (Figure 2A–C). In marked contrast, isoproterenol stimulation of YFP/beta-MHC expressing cells from the TAC hearts changed the %CS minimally after stimulation (Figure 2A, P > 0.05); the %CS in the YFP/beta-MHC expressing TAC myocytes is significantly lower than that in either the YFP/beta-MHC non-expressing TAC myocytes (P < 0.05) or the control myocytes (Figure 2A, P < 0.01). The changes in rates of relaxation and shortening, of the YFP/beta-MHC expressing TAC myocytes (Figure 2B and C), in response to isoproterenol stimulation, are also significantly lower than those of the YFP/beta-MHC non-expressing TAC myocytes from the same hearts (Figure 2B and C, P < 0.01), or from cells of the control heart (Figure 2B and C, P < 0.01). These results demonstrate that those cardiac myocytes that re-express YFP/beta-MHC are also those in which the beta-adrenergic response is decreased.

Figure 2.

Cell-shortening parameters of myocytes. Fractional cell shortening (% cell shortening, Y-axis) (A), rate of cell relaxation (dl/dt, B), and rate of cell shortening (−dl/dt, C) of myocytes from YFP/beta-MHC animals that were either untreated (columns marked ‘UT’), or treated with TAC for 4 weeks (columns marked ‘TAC’) that either did not express YFP-beta-MHC (open bars,) or that did express YFP-beta-MHC (yellow bars). Myocytes from each group were either unstimulated (−) or stimulated with 1 µM isoproterenol (+). Each experimental group consists of approximately 15 myocytes/heart, and five hearts/condition. Error bars denote SE. *P < 0.05, **P < 0.01, ***P < 0.001, ANOVA with Newman–Keuls correction.

To exclude the possibility that the changes in isoproterenol-induced contractility observed in the YFP/beta-MHC expressing myocytes might be due to the toxic effects of the fluorescent protein expression, we tested the contractile parameters of myocytes from an untreated transgenic animal that expresses GFP under the control of the cardiac alpha MHC promoter at a high level. The data (not shown) show that GFP-expressing myocytes from an untreated transgenic mouse display the same contractile parameters during basal conditions and in response to isoproterenol stimulation as the myocytes from the untreated non-transgenic hearts.

Taken together, our results demonstrate that the decrease in beta-adrenergic responsiveness in individual myocytes isolated from TAC treated hearts is co-incident with YFP/beta-MHC expression. Myocytes that do not express YFP/beta-MHC isolated from the same TAC treated hearts have a normal beta-adrenergic response, even though the YFP/beta-MHC non-expressing cells are as hypertrophic as those that express YFP/beta-MHC.

Discussion

We have previously demonstrated that, during cardiac hypertrophy, beta-MHC re-expression occurs in subsets (approximately. 38%) rather than in all hypertrophic myocytes, indicating that the pathways that induce cellular hypertrophy are not obligatorily coupled to the pathways that induce beta-MHC.^24,26 These YFP-expressing myocytes are mostly found in the left ventricle, generally in the mid-ventricular wall, and surround areas of fibrosis. In our current report, we demonstrate that decreased contractile response to beta-adrenergic stimulation also occurs in subsets of myocytes and is co-incident with beta-MHC re-expression; beta-MHC non-expressing cells from the same TAC hearts show a normal contractile response to beta-adrenergic stimulation even though they are equally hypertrophic.

Although we have demonstrated that the reduction of beta-adrenergic response is coincident with beta-MHC re-expression in individual cells in vivo, our data do not show whether the reduction of beta-adrenergic response and beta-MHC re-expression are linked in a cause-and-effect relationship or are the coincidental outcomes of some other change. However, genetic means of rescuing pathological heart failure by decreasing beta-adrenergic stimulation decrease beta-MHC expression.^3,6 Similarly, beta-blocker therapy of hearts in failure shows a decrease in the levels of beta-MHC mRNA along with the improvement of cardiac function.^27,28 Furthermore, in neonatal rat myocytes, isoproterenol induces both a decrease in beta-adrenergic responsiveness, and an increase in beta-MHC expression.²⁹ These observations indicate that agents which increase or decrease beta-adrenergic responsiveness also lead to changes in beta-MHC regulation, and that pathogenic stimulation that promote beta-adrenergic receptor downregulation/desensitization will also cause beta-MHC re-expression. Future studies addressing whether the beta-MHC expressing myocytes also show an altered response to other agonists would be instrumental in dissecting the pathways that link beta-MHC re-expression to other pathological stimuli.

The reduction of beta-adrenergic response in myocytes during TAC and heart failure has been well documented.^{10–13,30,31} Our present results demonstrate that the reduction in beta-adrenergic response in myocytes following hypertrophy induced by TAC occurs in subsets rather than in all myocytes. It has been proposed that the decrease in beta-adrenergic responsiveness of cardiac myocytes during heart failure may be a consequence of increase in circulating catecholamines.³² However, different regions of the same heart have different levels of catecholamines, and myocytes from the right ventricle of animals with increased circulating catecholamines do not show signs of decreased beta-adrenergic response.³³ These results and our current finding that only subsets of myocytes within the heart display decreased beta-adrenergic responsiveness support a hypothesis that local, rather than systemic, increases in catecholamine levels cause the reduction in beta-adrenergic responsiveness and increase in beta-MHC expression that we observe in individual myocytes.

Some comment is required on the apparent discrepancy between our results that show normal basal contractile behaviour in beta-MHC expressing cells and the results of Tardiff et al.⁹ that showed a decrease in baseline contractile parameters in beta-MHC expressing cells. Important differences exist between the two studies. The Tardiff study used a mouse model in which beta-MHC was transgenically over-expressed in all cardiac myocytes (12% of total MHC). In contrast, in our studies, beta-MHC was re-expressed in response to pressure overload and occurred in subsets of cells. Furthermore, while the Tardiff study conducted contractility studies (measuring dp/dt) on the whole heart using retrograde perfusion on a Langendorff apparatus, our contractility studies (measuring cell shortening) were carried out at the level of individual isolated cardiac myocytes.

Funding

This work was supported by grants from NIH [HL71266 and HL49277 (O.S.); HL56687 (H.A.R.)]; and the American Heart Association [0325655U and 0525594U (K.P.)]. Authors declare no financial conflict of interest.