Effects of Total Replacement of Atrial Myosin Light Chain-2 With the Ventricular Isoform in Atrial Myocytes of Transgenic Mice
Background—In contrast to their well-known and critical role in excitation-contraction coupling of vascular smooth muscle, the effects of the myosin light chains on cardiomyocyte mechanics are poorly understood. Accordingly, we designed the present experiment to define the cardiac chamber–specific functional effects of the ventricular isoform of the regulatory myosin light chain (MLC2v).
Methods and Results—Postnatal transgenic cardiac-specific overexpression of MLC2v was achieved by use of the α-myosin heavy chain promoter. Enzymatically disaggregated atrial and ventricular mouse myocytes were field-stimulated at multiple frequencies, and mechanical properties and calcium kinetics were studied by use of video edge detection and FURA 2-AM, respectively. MLC2v overexpression resulted in complete replacement of the atrial with the ventricular isoform of the regulatory myosin light chain at the steady-state mRNA and protein levels in the atria of transgenic mice. Mechanical properties of transgenic atrial myocytes were enhanced to the level of ventricular myocytes of control animals in association with modest decreases in the amplitude of the calcium transient.
Conclusions—MLC2v modulates chamber-specific contractility by enhanced calcium sensitivity and/or improved cross-bridge cycling of the thin and thick filaments of the cardiomyocyte.
It has long been recognized that contraction of the heart is dependent on the force generated by the interactions between the thick and thin filaments of the cardiac sarcomere. Detailed structural studies have demonstrated that force generation in muscle cells is due to cross-bridge cycling between thin-filament actin and thick-filament myosin prompted by ATP hydrolysis.1 2 3 Myosin is a hexameric molecule composed of two heavy-chain proteins and two pairs of distinct light-chain proteins. There are two classes of MLCs, and one of each is associated with each heavy chain. Both types of MLC are usually encoded by a multiple gene family, giving rise to a number of isoforms in each class that are regulated in a tissue- and cardiac chamber–specific fashion during development and pathological processes. The two heavy chains each form a head region that contains the ATP binding site and an α-helical tail region, whereas MLC1 and MLC2 are situated in the neck region of the myosin heavy chain proteins. Data from more recent structural studies provide evidence that it is small conformational changes in the light chain binding regions that are responsible for the actual movement of smooth muscle myosin with the release of ADP.1 2 3 Although the structural relationship of the MLC proteins to the contractile apparatus of muscle is becoming clearer, the functional role of the light chain proteins and their isoforms in muscle contraction is incompletely understood. In particular, less is known regarding the role of the cardiac MLCs in excitation-contraction coupling than is the case for these proteins in skeletal and smooth muscle.
MLC2 is also called the regulatory light chain, because phosphorylation of this protein controls contraction in smooth muscle. In skeletal muscle, phosphorylation of MLC2 is thought to have a modulatory role in both the rate and magnitude of force generation.4 5 6 7 8 9 In contrast, little is known about the role of MLC2 and its phosphorylation in cardiac myocyte shortening, although it has been demonstrated that cardiac MLC2 phosphorylation produces a dramatic increase in the sensitivity of tension development to increasing extracellular Ca2+ concentrations.10 In addition, Damron et al11 reported that increased phosphorylation of MLC2 with endothelin or arachidonic acid treatment produced a positive inotropic effect, which they interpreted as being consistent with an increase in calcium sensitivity; however, calcium transient data were not reported in that study. Thus, the role of MLC2 in a phosphorylated or dephosphorylated state in cardiac muscle contraction is unclear. The cardiac regulatory MLCs exist in chamber-specific isoforms for the atria (MLC2a) and ventricles (MLC2v). Although these isoforms arise from distinct genes and are altered in pathological states, assigning different functional roles for the encoded proteins has not been possible.
Recently, Palermo et al12 produced a transgenic mouse with cardiac-specific postnatal overexpression of MLC2v. Although large increases in mRNA for MLC2v were seen in both the atria and ventricles of the transgenic mouse heart, no difference was observed in the total MLC2 protein in either compartment. However, ectopic expression of MLC2v in the atria resulted in the total replacement of the atrial isoform of MLC2 by MLC2v. A similar phenomenon has been reported in experimental and clinical cardiomyopathies.13 14 15 16 17 For instance, Kumar et al13 reported mRNA expression of MLC2v in the atria of the spontaneously hypertensive rat, and an increase in MLC2v protein has been reported in the atria of humans with various cardiomyopathies.15 16 Because the atria and ventricles of the heart play much different roles in cardiac function (the ventricles are required to generate much greater forces per beat than the atria), we hypothesized that replacement of the atrial isoform with the ventricular species would result in functional changes at the myocyte level. To test this hypothesis, the present experiments were designed to study the contractile properties and intracellular calcium kinetics of enzymatically isolated atrial and ventricular myocytes derived from a transgenic mouse line that overexpresses MLC2v.
Transgenic mice12 with postnatal cardiac-specific overexpression of MLC2v and nontransgenic control mice were used as sources of isolated myocytes. All mice were studied between 11 and 13 weeks of age, and equal distributions of male and female mice were used in each group. Mouse ventricular myocytes were isolated by methods previously reported for this laboratory unless otherwise noted.18 19 20 Transgenic mice were identified by PCR analysis of DNA isolated from tail clips.
Isolation of Ventricular Myocytes
Mice were anesthetized with methoxyflurane, hearts were quickly excised, and the aortas were cannulated with a blunted 23-gauge needle, flushed with buffer (MEM, Joklik-modified, pH 7.2; Gibco BRL), and mounted onto a Langendorff perfusion apparatus. Ventricular myocytes were isolated from hearts of control and transgenic mice by methods slightly modified from those described previously.18 19 Briefly, all perfusates were maintained at 37°C and continuously bubbled with 95%O2/5%CO2. The coronary arteries were perfused retrogradely at 2.2 mL/min initially with calcium-free Joklik buffer (4 minutes), followed by Joklik buffer supplemented with 0.25 mmol/L Ca2+, 75 U/mL collagenase I (Worthington), 75 U/mL collagenase II (Worthington), 1% BSA (Sigma), and 2% donor calf serum (pH 7.2). After ≈12 to 15 minutes, the heart was removed from the perfusion apparatus and transferred to a glass dish containing Joklik buffer supplemented with 0.25 mmol/L Ca2+ and 2% donor calf serum. The left ventricle was isolated from the rest of the chambers and minced, and isolated cells were washed and resuspended in a physiological buffer (in mmol/L: NaCl 132, KCl 4.8, MgCl2 · 6H2O 1.2, glucose 5, and HEPES 10, pH 7.2) supplemented with 1.8 mmol/L calcium for study.
For atrial myocyte isolation, hearts were perfused by the same methods as described above for ventricular cell isolation. However, once atrial tissue was removed from the rest of the heart, cells were isolated by gentle teasing of the tissue with hypodermic needles in a glass dish with enzyme-free low-calcium Joklik medium. If cells did not easily separate from tissue pieces, the tissue was allowed to soak in enzyme solution for 5 to 10 minutes, as needed. Cells were then allowed to settle in a tube before the Joklik medium was removed and replaced with physiological buffer as described above.
For measurements of morphological and mechanical properties of isolated myocytes, cells were placed in a well on the stage of an inverted microscope and were perfused continuously with oxygenated physiological buffer. Two platinum electrodes connected to a Grass model S9 stimulator were placed on either side of an identified healthy-appearing, rod-shaped myocyte with clearly visible striations and no evidence of blebbing. Myocytes were field-stimulated at varying frequencies (0.25, 0.5, and 1.0 Hz; 5-ms pulse duration) for at least 40 seconds per pacing rate. Cell images were acquired continuously through a charge-coupled device (model GP-CD60) and recorded on videotape. With a video motion edge detector (Crescent Electronics), these videotaped images were then captured on a Gould chart recorder from which percent shortening and rates of shortening (+dL/dt) and relengthening (−dL/dt) were quantified by comparison with a calibrated micrometer on the microscope stage.
Once left ventricular cells were isolated, half were used for mechanical studies and the other half for the calcium kinetic studies. In contrast, for the assessment of atrial cell function, the entire left atrium was required, so separate mice were used for mechanical and calcium measurements. Disaggregated myocytes were placed in FURA 2-AM (ventricular cells, 7.5 μmol/L and atrial cells, 2.5 μmol/L) and incubated at 37°C for ≈15 minutes in the dark. After FURA loading of cells was completed, the cells were then suspended in physiological buffer as described above. Cytosolic free calcium was measured in mouse myocytes by ratio imaging of 340 to 380 nm fluorescence of FURA 2 (emission wavelength, 510 nm) with a photo scan dual-beam spectrofluorophotometer (Photon Tech, Inc) coupled to an Olympus IMT-2 UV fluorescent microscope with UV transparent optics. Cells underwent a pacing protocol similar to that performed in the mechanical studies, and baseline and peak intracellular calcium transients were measured in response to changes in stimulation frequencies.
At least three cells were examined per mouse, per chamber (atrium and ventricle), and the values were averaged for mechanical parameters and Ca2+ kinetics. Statistical analysis is based on the number of animals rather than the number of cells. Data are expressed as mean±SEM and are analyzed by two-way ANOVA followed by the Student-Newman-Keuls test for individual post hoc comparisons. Morphological data were analyzed by unpaired t test. If data were not normally distributed or failed equal variance tests after log10 transformations, they were analyzed by nonparametric statistics (ie, either Kruskal-Wallis for ANOVA designs or Mann-Whitney rank sum test for comparison between two groups of data). A value of P<.05 was set as the criteria for statistical significance.
Morphological and Mechanical Properties of Mouse Ventricular Myocytes
Ventricular myocytes isolated from transgenic mice were morphologically indistinct from control ventricular myocytes (Table 1⇓). Representative analog tracings of myocyte shortening (Fig 1⇓) demonstrate that the extent of ventricular myocyte shortening measured in these cells was not different between transgenic and control mice at any of the three stimulation frequencies. However, the rates of shortening (+dL/dt) and relengthening (−dL/dt) produced by electrical stimulation of the myocytes at each pacing rate were diminished in the transgenic ventricular myocytes compared with control cells. These findings are confirmed by the composite data as illustrated in Table 1⇓ and Fig 2⇓. Thus, with no detectable difference in total MLC2v protein levels between mice, transgenic ventricular myocytes demonstrated similarities in percent shortening but depressed rates of contraction and relaxation.
Morphological and Mechanical Properties of Mouse Atrial Myocytes
Atrial myocytes isolated and studied from transgenic mice with cardiac-specific ectopic replacement of the atrial with the ventricular isoform of MLC2 in the heart were significantly shorter than atrial myocytes similarly isolated from nontransgenic littermates (Table 1⇑). Compared with ventricular myocytes, atrial myocytes were shorter and thinner in both groups (Table 1⇑). However, no differences were seen in cell width-to-length ratios between either atrial and ventricular myocytes or atrial cells isolated from control versus those from transgenic mice. When electrically stimulated at incremental pacing frequencies, control atrial myocytes exhibited a much attenuated percent shortening compared with transgenic atrial or control ventricular myocytes (Table 1⇑; Figs 1⇑ and 2⇑). Similarly, rates of shortening and relengthening in control atrial myocytes were much less than those in either transgenic atrial or control ventricular cells (Table 1⇑, Fig 2⇑). These findings were similar at all three stimulation frequencies. Furthermore, comparisons of contractile properties within groups were not significantly different with increasing pacing rates. In contrast to control atrial myocytes, the atrial myocytes isolated from the mice in which MLC2v protein had completely replaced the atrial isoform of MLC2 demonstrated contractile properties that were similar to those of nontransgenic ventricular myocytes (Fig 1⇑).
Intracellular Ca2+ Measurements in Mouse Ventricular Myocytes
The differences in rates of contraction and relaxation between transgenic and control ventricular myocytes could not be explained by differences in intracellular Ca2+ kinetics (Table 2⇓). Baseline and peak Ca2+ levels obtained during electrical pacing of myocytes were not different between ventricular cells isolated from control and transgenic mice. In addition, the times of 50% (T50) and 80% (T80) Ca2+ signal decay were similar between groups. Altering the pacing frequencies affected neither the intergroup group relationships nor intragroup group comparisons of the Ca2+ kinetics.
Intracellular Ca2+ Measurements in Mouse Atrial Myocytes
Baseline Ca2+ levels and Ca2+ signal amplitudes for atrial myocytes paced at three pacing rates are shown in Table 2⇑. As was seen with ventricular myocytes isolated from transgenic mice, atrial myocytes from these mice demonstrated baseline Ca2+ signals similar to atrial cells taken from nontransgenic mice. The amplitude of the Ca2+ signals was slightly but significantly lower in the transgenic atrial myocytes than in control cells. Furthermore, although baseline signals were not different between ventricular and atrial myocytes, the amplitude of the Ca2+ signals produced by electrical pacing of the cells was significantly lower in the atrial than in the ventricular myocytes in both groups of mice.
The present studies report, for the first time, the mechanical properties and calcium kinetics of atrial myocytes derived from the mouse heart. These data demonstrate that isolated unloaded mouse atrial myocytes contract to a lesser extent and at slower rates than do isolated ventricular cells. However, total replacement of the atrial isoform of MLC2 by the ventricular isoform in the atria of the mouse results in atrial cells that contract and relax at greater rates and to a greater extent than do isolated atrial cells from control mice. In fact, the transgenic atrial myocytes demonstrate contractile properties similar to normal ventricular cells. These studies also demonstrate that although baseline Ca2+ transients are not different between control and transgenic cells, normal atrial myocytes exhibit lower Ca2+ signal amplitudes than do ventricular myocytes and transgenic atrial cells have lower amplitudes than do atrial cells from nontransgenic mice. These data suggest that the regulatory MLCs can play a major role in the differentiation of cardiac compartment muscle mechanics and calcium signaling and that chamber-specific isoforms have unique functional properties.
The atrial-to-ventricular switch of MLC2 occurs during postnatal development of the ventricle in response to the accompanying changes in intra-atrial pressures.13 17 On the basis of the present studies, the postnatal MLC isoform switch in the left ventricle of the neonate may facilitate ejection against the increased systemic arterial pressure that occurs at parturition. In addition, several studies have demonstrated expression of the ventricular isoform of MLC2 in the atria of hypertrophied hearts of both humans and experimental animals in response to pathological conditions.13 14 15 16 17 Wanker et al15 found MLC2v in atrial samples from patients with a variety of cardiomyopathies, and the level of ventricular isoform expression correlated with the severity of heart failure. Likewise, Cummins16 reported that the degree of pressure-overload hypertrophy in humans is the most significant factor influencing ventricular MLC2v isoform expression in the human atria. Just as in the developing neonatal ventricle, it was hypothesized that the changes in atrial chamber pressures were responsible for this isoform switch in myopathic atria. However, Kumar et al13 demonstrated that the atria from the spontaneously hypertensive rat had greater levels of MLC2v mRNA expression than did atria from age-matched normotensive Wistar-Kyoto rats that preceded the development of both hypertension and cardiac hypertrophy. These studies are inconsistent with the hypothesis that the atrial-to-ventricular switch of MLC2 in the atria occurs solely as a result of hemodynamic factors. Therefore, although there may be some relationship between cardiomyopathy and the atrial-to-ventricular MLC2 switch in the atria, it remains unknown whether this phenomenon plays a role in or is a consequence of the pathological condition. Furthermore, it is unknown how this switch affects atrial as well as ventricular function. Data from the present studies suggest that enhanced atrial expression of MLC2v in the diseased heart may be a compensatory mechanism to maintain and enhance the left atrial contribution to ventricular filling.
The present studies demonstrate that mechanical properties of mouse atrial myocytes that ectopically express MLC2v are similar to ventricular myocytes. It appears that the only biochemical difference between the nontransgenic and transgenic left atria is the total replacement of the atrial MLC2 isoform by the ventricular MLC2 isoform. Transgenic atrial MLC2v has a higher basal phosphorylation level than MLC2v in the ventricles. In addition, there was also no difference in the degree of phosphorylation between the wild-type and transgenic atrial MLC2.21 Thus, the altered mechanical properties of the transgenic atrial myocytes appear to be unrelated to any change in the level of phosphorylation brought about by the atrial-to-ventricular MLC2 isoform switch. In addition, there are no differences in the myosin heavy chain isoforms, the major determinant of myosin ATPase activity, in the calcium-cycling proteins (the sarcoplasmic reticulum ATPase and phospholamban) or in α-actin isoform composition between the atria of wild-type and transgenic mice.21 We therefore consider it a reasonable hypothesis that the morphological differences (shorter atrial myocytes) as well as mechanical differences (greater percent shortening and faster rates of shortening and relengthening) in the transgenic atrial myocyte are a direct consequence of the regulatory MLC isoform replacement. The cell length differences could not have been observed or predicted from previously performed in vitro motility assays and are difficult to explain on the basis of current understanding of structural relationships of myosins in cardiac muscle. However, these data imply that MLC2 plays an important role in determining the contractile properties of the cardiac chambers.
Depressed mechanical function of nontransgenic atrial myocytes compared with ventricular cells might be predicted from the calcium kinetic studies. Increased intracellular Ca2+ levels and lower T50 and T80 in the ventricular myocytes compared with atrial myocytes support the mechanical data that unloaded ventricular cells contract faster and to a greater extent than do atrial cells. However, compared with nontransgenic atrial cells, transgenic atrial myocytes exhibit slightly lower electrically stimulated increases in intracellular Ca2+, with no differences in T50 or T80. Thus, the increase in atrial myocyte contractility in transgenic mice compared with wild-type atrial myocyte shortening resulting from replacement of the atrial with the ventricular isoform of the regulatory MLC cannot be explained on the basis of altered calcium kinetics. It appears that the ventricular MLC isoform switch enhances atrial cardiomyocyte calcium sensitivity of the myofilament and/or facilitates more effective actin-myosin cross-bridge development and cycling.
On the basis of the biochemical analyses of the transgenic ventricles (ie, no difference in MLC2v protein expression between control and transgenic),12 no difference in mechanical properties would be predicted between the left ventricular myocytes of these groups. There was no difference in the gross morphology of the ventricular myocytes. However, modest but statistically significant slower rates of shortening and relengthening of transgenic ventricular myocytes were observed. The reasons for depressed mechanical function in the transgenic ventricular myocytes are not readily apparent. These differences cannot be explained on the basis of altered phosphorylation status, myosin heavy chain isoform composition, calcium-cycling proteins, or α-actin isoform composition between wild-type and transgenic ventricles.21 It has been postulated that heart rate plays a role in the activity of MLC kinase, the enzyme responsible for phosphorylation of MLC2.7 10 However, neither conscious heart rates nor phosphorylation status differed between wild-type and transgenic mice.21 One possibility for the mildly depressed function of the transgenic ventricular myocytes is that there may be other biochemical changes not yet established, either as a result of the enhanced mechanical properties of the atria or simply endogenous to this transgenic line. What is clear from the present experiments is that the depressed mechanical properties of the transgenic ventricular myocytes are not due to changes in intracellular calcium handling.
In conclusion, these data demonstrate that total replacement of the atrial isoform of MLC2 with the ventricular isoform in the left atrium of the mouse results in atrial myocytes with mechanical and morphological properties similar to those of ventricular cells, despite modestly diminished intracellular calcium transients. These studies suggest that MLC2 isoforms in cardiac tissue are central to the differential contractility of compartmentalized heart muscle. Further study of these transgenic mice might lead to an even greater understanding of the role of MLC2v in cardiac muscle contraction under normal conditions, as well as the role of its expression in the atria observed in cardiomyopathies.
Selected Abbreviations and Acronyms
|MLC||=||myosin light chain|
|MLC1||=||essential myosin light chain|
|MLC2||=||regulatory myosin light chain|
|MLC2a||=||atrial isoform of regulatory myosin light chain|
|MLC2v||=||ventricular isoform of regulatory myosin light chain|
This work was supported in part by SCOR in Heart Failure grant HL-52318 from the National Heart, Lung, and Blood Institute.
Reprint requests to Richard A. Walsh, MD, Division of Cardiology, University of Cincinnati, PO Box 670542, Cincinnati, OH 45267-0542.
- Received August 22, 1997.
- Revision received October 29, 1997.
- Accepted November 7, 1997.
- Copyright © 1998 by American Heart Association
Rayment I, Rypniewski WR, Schmidt-Base K, Smith R, Tomchick DR, Benning MM, Winkelmann DA, Wesenberg G, Holden HM. Three-dimensional structure of myosin subfragment-1: a molecular motor. Science. 1993;261:50–58.
Rayment I, Holden HM, Whittaker M, Yohn CB, Lorenz M, Holmes KC, Milligan RA. Structure of the actin-myosin complex and its implication for muscle contraction. Science. 1993;261:58–65.
Sweeney HL, Bowman BF, Stull JT. Myosin light chain phosphorylation in vertebrate striated muscle: regulation and function. Am J Physiol. 1993;264:C1085–C1095.
Trybus KM, Chatman TA. Chimeric regulatory light chains as probes of smooth muscle myosin function. J Biol Chem. 1993;268:4412–4419.
Damron DS, Darvish A, Murphy LA, Sweet W, Moravec CS, Bond M. Arachidonic acid-dependent phosphorylation of troponin I and myosin light chain 2 in cardiac myocytes. Circ Res. 1995;76:1011–1019.
Palermo J, Gulick J, Colbert M, Fewell J, Robbins J. Transgenic remodeling of the contractile apparatus in the mammalian heart. Circ Res. 1996;78:504–509.
Kumar C, Saidapet C, Delaney P, Mendola C, Siddiqui MAQ. Expression of ventricular-type myosin light chain messenger RNA in spontaneously hypertensive rat atria. Circ Res. 1988;62:1093–1097.
Cummins P. Transitions in human atrial and ventricular myosin light-chain isoenzymes in response to cardiac-pressure-overload-induced hypertrophy. Biochem J. 1982;205:195–204.
Henkel RD, Kammerer CM, Escobedo LV, VandeBerg JL, Walsh RA. Correlated expression of atrial myosin heavy chain and regulatory light chain isoforms with pressure overload hypertrophy in the non-human primate. Cardiovasc Res. 1993;27:416–422.
Dorn GW, Robbins J, Ball N, Walsh RA. Myosin heavy chain regulation and myocyte contractile depression after LV hypertrophy in aortic-banded mice. Am J Physiol. 1994;267:H400–H405.
Gunteski-Hamblin AM, Song G, Walsh RA, Frenzke M, Boivin GP, Dorn GW II, Kaetzel MA, Horseman ND, Dedman JR. Annexin VI overexpression targeted to heart alters cardiomyocyte function in transgenic mice. Am J Physiol. 1996;270:H1091–H1100.
Fewell JG, Osinska H, Klevitsky R, Ng W, Sfyris G, Bahrehmand F, Robbins J. A treadmill exercise regimen for transgenic mice. Am J Physiol. 1997;273:H1595–H1605.