Heart failure and dilated cardiomyopathy develop in mice that lack the muscle LIM protein (MLP) gene (MLP−/−). The character and extent of the heart failure that occurs in MLP−/− mice were investigated using echocardiography and in vivo pressure-volume (P-V) loop measurements. P-V loop data were obtained with a new method for mice (sonomicrometry) using two pairs of orthogonal piezoelectric crystals implanted in the endocardial wall. Sonomicrometry revealed right-shifted P-V loops in MLP−/−mice, depressed systolic contractility, and additional evidence of heart failure. Cellular changes in MLP−/− mice were examined in isolated single cells using patch-clamp and confocal Ca2+ concentration ([Ca2+]) imaging techniques. This cellular investigation revealed unchanged Ca2+ currents and Ca2+ spark characteristics but decreased intracellular [Ca2+] transients and contractile responses and a defect in excitation-contraction coupling. Normal cellular and whole heart function was restored in MLP−/− mice that express a cardiac-targeted transgene, which blocks the function of β-adrenergic receptor (β-AR) kinase-1 (βARK1). These data suggest that, despite the persistent stimulus to develop heart failure in MLP−/− mice (i.e., loss of the structural protein MLP), downregulation and desensitization of the β-ARs may play a pivotal role in the pathogenesis. Furthermore, this work suggests that the inhibition of βARK1 action may prove an effective therapy for heart failure.
- β-adrenergic receptor
- excitation-contraction coupling
- calcium signaling
- β-adrenergic receptor kinase
heart failure isa clinical syndrome in which molecular and cellular changes in cardiac myocytes lead to inadequate myocardial function. Much attention has been focused on the role played by the β-adrenergic receptor (β-AR) signaling pathway in heart failure after the surprising discovery that β-AR blockers, known to have negative inotropic action, provide benefit to patients in heart failure (4-6). Indeed, it has been well established that β-ARs are downregulated and/or desensitized during heart failure (3). Characteristic changes in β-ARs include uncoupling of both β1- and β2-AR subtypes, threefold increase in activity of the β-adrenergic receptor kinase (βARK1), and increased levels of Gi (13, 40). Exploration of the mechanism of action of β-AR modulation in the context of heart failure requires cellular and molecular examination of cardiac myocytes from animals whose myocardial function is well defined.
A useful genetically defined murine model of heart failure was recently described (1). In this model, the muscle LIM protein (MLP) is knocked out to produce the MLP−/− animal. Two genetic interventions have been described that suggest that alterations in either Ca2+ transport (29, 38) or β-AR function (31) play a central role in MLP−/−heart failure. We investigated the nature and extent of heart failure associated with the MLP−/− mouse, and that rescued by the βARK1 inhibitor, by examining the intrinsic myocardial contractile state in vivo using a pressure-volume (P-V) analysis and intracellular Ca2+ concentration ([Ca2+]i) signaling using patch-clamp methods and confocal Ca2+imaging. We present here data showing that MLP−/− animals have a defect in excitation-contraction (EC) coupling that contributes to their contractile dysfunction and that this pathology is significantly more severe than the loss of the MLP per se. Moreover, β-AR dysfunction appears to be a necessary component because restoration of β-AR signaling in these animals is sufficient to prevent the abnormalities of in vivo contractile dysfunction, of poor cellular contraction, of weak Ca2+ signaling, and of defective EC coupling.
Adult MLP−/− and MLP−/−/βARK1 inhibitor (βARKct) littermate mice of either sex, age 5 to 7 mo, were used in this study. The mice were generated by mating transgenic mice with cardiac-targeted overexpression of the βARKct (24) with mice homozygous for the deletion of the MLP gene (MLP−/−) (1) to create MLP−/−/βARKct mice (31). Nonlittermate wild-type mice of either 129Sv/B6 or CD-1 strain were used as controls. The animals in this study were handled according to approved protocols and animal welfare regulations of the authors' Institutional Review Boards.
Two-dimensional (2D) guided M-mode echocardiography was performed using an HDI 5000 echocardiograph (ATL, Bothell, WA) as previously described (11, 31). Mice were first studied in the conscious state using gentle manual restraint after a period of acclimation. A soft plastic collar was fashioned to prevent the animals from biting the probe. After acquisition of satisfactory echocardiograms, the same mice were restudied under anesthesia (Avertin 2.5%, 14 μl/g ip).
Several days after echocardiography, mice were reanesthetized with a mixture of ketamine (100 mg/kg) and xylazine (2.5 mg/kg) and were connected to a rodent ventilator after endotracheal intubation. After bilateral vagotomy, the chest was opened and the pericardium was dissected to expose the heart. A 7-0 suture ligature was placed around the transverse aorta to manipulate loading conditions (see Fig.2 A). A 1.4-Fr (0.46 mm) high-fidelity micromanometer catheter (Millar Instruments, Houston, TX) was inserted into the right carotid and advanced retrograde into the left ventricle (LV). A polyethylene (PE)-50 catheter was inserted into the external jugular vein for drug infusion. Two pairs of miniature omnidirectional piezoelectric crystals (0.7 mm; Sonometrics, London, ON, Canada) were implanted in the endocardium of the LV by inserting the crystals through the epicardial layer into the chamber and then withdrawing until there was resistance. Short-axis dimension was measured with a pair of crystals implanted in the anterior-posterior orientation. Long-axis dimension was measured with a crystal implanted in the apex and one crystal attached to the base of the heart at the level of the aortic valve using cyanoacrylate adhesive (Vetbond, 3M Animal Care Products, St. Paul, MN). The space and time resolution of the sonomicrometry system are 0.015 mm and 0.001 s, respectively.
After a brief period of stabilization, simultaneous LV pressure and LV dimensions were recorded at baseline and during increases in afterload generated by gently pulling on the suture to transiently constrict the transverse aorta. The ventilator was stopped during data acquisition to eliminate effects of positive ventilation. After return of LV pressure to baseline values, the contractile state was increased with a dobutamine infusion (2 μg · kg−1 · min−1). After steady state was reached (∼5 min), recordings of variably loaded pressure-dimension measurements were made as above. All data was recorded digitally at 2,000 Hz and stored on a computer for off-line analysis. At the end of the experiment, animals were killed and proper positioning of the crystals was documented by direct inspection.
The digitized data were analyzed using a computer algorithm. Baseline and dobutamine hemodynamic values were obtained by averaging 10 beats recorded during the steady-state periods. Parameters measured were LV systolic and end-diastolic pressure, maximal (LV dP/dt max) and minimal first derivative of LV pressure, and heart rate. The end-systolic point of each P-V loop was determined by an automated iterative linear regression technique as previously described (21, 30). LV volume was calculated as a modified general ellipsoid using the equation V = (π/6) · D la · (D ap)2where D la is the apex to base long-axis LV dimension and D ap is the anterior to posterior short-axis LV dimension. The end-systolic points were then fitted to a parabolic curvilinear model, Pes =a(Ves − V0)2+ b(Ves − V0), where Pes is end-systolic pressure, Ves is end-systolic volume, V0 is the volume axis intercept,b is the local slope at V0, and a is the curvilinearity coefficient (21). If the 95% confidence interval of a did not include zero, the end-systolic P-V relation (ESPVR) was considered curvilinear (see Table3).
Cell Isolation and Electrophysiology
Adult animals (wild type, n = 5; MLP−/−, n = 6; and MLP−/−/βARKct, n = 7) were killed by intraperitoneal injection of pentobarbital sodium (100 mg/kg). Single mouse ventricular myocytes were isolated (37) and stored at room temperature (22–25°C) in Dulbecco's modified Eagle's medium (JRH Biosciences, Lanexa, KS) until being used. An Axopatch-200A amplifier (Axon Instruments) was used to patch clamp the cells (whole cell configuration) and measure membrane currents. Patch pipettes of nominal resistance of 0.5–3 MΩ were used and were filled with an internal solution of (in mM) 130 CsCl, 20 tetraethylammonium chloride, 5 Mg-ATP, 10 HEPES, and 0.05 fluo 3-K5; pH 7.2 (with NaOH). Some cells were preloaded with fluo 3 by 30-min exposure to 1.5 μM fluo 3-AM at room temperature.
Cells were superfused with one of three solutions. Solution 1 contained (in mM) 140 NaCl, 5 KCl, 0.5 MgCl2, 1.8 CaCl2, 0.33 NaH2PO4, 5.5 glucose, and 5 HEPES. Solution 2 contained (in mM) 140 NaCl, 5 CsCl, 0.5 MgCl2, 1.8 CaCl2, 0.33 NaH2PO4, 5.5 glucose, and 5 HEPES.Solution 2 was used to measure Ca2+ current (I Ca). Solution 3 contained (in mM) 145 CsCl, 0.5 MgCl2, 5.5 glucose, and 5 HEPES. This solution was used to block the Na+/Ca2+exchanger when the sarcoplasmic reticulum (SR) Ca2+ content was tested by applying 20 mM caffeine. The pH of all extracellular solutions was kept at 7.4 with the temperature at 34–37°C.
Confocal Microscopy and [Ca2+]i Measurements
Confocal microscopy was used to measure [Ca2+]i (7-9) and to investigate cellular function as previously described in rat and mouse heart cells (16, 36, 37). A robust preloading protocol was used to attain similar SR Ca2+ loads (35, 37). We were therefore able to compare Ca2+ signaling in cells from wild-type, MLP−/−, and MLP−/−/βARKct animals because they had a constant amount of Ca2+ in the SR (15). This procedure permits meaningful comparison of EC coupling gain (35, 37).
Data are expressed as means ± SE. Two-way repeated ANOVA was used to evaluate the hemodynamic measurements, the ESPVR variables under basal conditions and with dobutamine stimulation, and the echocardiographic parameters between anesthetized and conscious states. When appropriate, post hoc analysis was performed with a Newman-Keuls test. Bland-Altman (2) analysis (coefficient of variability) was used to show agreement between the crystal and echocardiographic measurements of LV diastolic dimension. For all analyses, P < 0.05 was considered significant. For the data from single-cell experiments, two-sample comparisons were performed using Student's t-test, and multigroup comparisons were made using a one-way ANOVA and Tukey's test.
Heart Failure in MLP−/− Mice
Evidence from echocardiographic studies.
Investigations into the nature and extent of heart failure associated with the MLP−/− mouse are shown in Fig.1. Echocardiograms performed in mice under both conscious and anesthetized conditions for wild-type (Fig. 1,left), MLP−/− (Fig. 1, middle), and a genetically crossed hybrid mouse (Fig. 1, right) are shown. From Fig. 1, it is clear that the MLP−/− mice have hearts with significantly enlarged internal chamber dimensions [end-diastolic dimension (EDD) and end-systolic dimension (ESD)] and reduced fractional shortening compared with wild-type mice (see Fig. 1and Table 1). Such LV dilation and reduced cardiac function are apparent in both anesthetized and conscious animals. Although echocardiography is valuable to measure chamber size and systolic performance in heart failure in vivo, it is limited in the assessment of the intrinsic contractile state of the ventricle because of the dependence on afterload (26). To examine the murine myocardial contractile state in vivo, a new method was developed for use in mice: sonomicrometry pressure-volume analysis (See Fig. 2 A).
Cardiac function assessed using P-V loop measurements.
With two pairs of endocardial implanted piezoelectric crystals (seemethods and Fig. 2 A) and a high-fidelity micromanometer in the LV, in vivo P-V relationships were obtained. These P-V relationships were obtained at different “loading” conditions and in the presence and absence of β-AR stimulation. P-V relationships provide a powerful approach to examine contractile function in vivo (20, 33), and these measurements were used to investigate in vivo contractile dysfunction in the MLP−/− mice.
Representative P-V loop families are shown for a typical wild-type mouse in Fig. 2 B. At baseline the end-diastolic volume (Ved) is 32 μl, which increases to 55 μl as afterload is increased by aortic constriction, as shown by the shift to the right and upwards as expected for a normal heart (Fig. 2 B and Table 2). The addition of the β-AR agonist dobutamine (2 μg · kg−1 · min−1, red P-V loops) slightly increases the LV end-systolic pressure (LVESP) with low afterload but has an even more dramatic action as afterload is increased. More importantly, however, there is an increase in the slope of the relation describing end-systolic pressure and volume as afterload increases (Figs. 2 B and3 A). This pattern of changes in cardiac function is typical for a normal heart whether it be from a mouse, dog, pig, or human (19, 20, 22, 33) and is a fingerprint for the normal myocardial response to increased afterload in the presence and absence of β-AR stimulation.
The P-V relationships observed in animals lacking the muscle LIM protein (MLP−/−) are dramatically different from those observed in wild-type mice. The Ves and Ved of the hearts are dramatically increased (7-fold and 4-fold, respectively), and fractional shortening is significantly reduced from 47–55% in control animals to 26–29% in MLP−/−animals (P < 0.001). The responses to increases in afterload are less steep in MLP−/− animals (Fig.3 A, Table 3) compared with wild-type hearts, and there is little change in the slope of the ESPVR in response to the β-AR agonist dobutamine (Figs. 2 C and3 A). Together, these data indicate that MLP−/−mice are in functional heart failure, a finding consistent with clinical observation of these animals. These data also support the large body of published work indicating that the β-AR signaling system is poorly responsive or unresponsive in many models of heart failure (3). It is also worth noting that although LV dP/dt max is mildly reduced in the MLP−/− mice compared with wild-type mice, it does not have the sensitivity to detect the marked abnormalities in intrinsic contractility as shown by the P-V loop analysis (12, 27).
To test whether this measure of cardiac function (P-V analysis) in mice agrees with an established standard (echocardiography), we compared cardiac dimensions determined by sonomicrometry and by echocardiography in the same mice. Figure 4 Ashows that the LVEDD measured by the two methods was similar, with the sonomicrometry giving only a slightly larger dimension. Furthermore, a Bland-Altman analysis (Fig. 4 B) showed a good agreement between the two in vivo measurements of LV diastolic dimension.
EC Coupling Defect in MLP−/− Mice
The cellular cause of the contractile dysfunction in MLP−/− mice is unknown. However, Arber et al. (1) suggested that the absence of the MLP, a structural protein in striated muscle, might be responsible. Although this hypothesis seems reasonable, Ca2+ signaling defects in heart failure have been detected in other models of heart failure even when alternative reasonable causes were found (16, 41). Consequently, we decided to test the possibility that Ca2+signaling may be altered and may contribute to the heart failure phenotype in MLP−/− mice. We examined membrane currents, [Ca2+]i, and cellular shortening and relaxation in patch-clamped heart cells (whole cell mode). Figure5 A shows the voltage protocol, the [Ca2+]i measurement as fractional fluorescence (F/Fo), the shortening record, and the membrane current density (pA/pF) in a single control heart cell. It is clear from the data that the cardiomyocytes from MLP−/− mice produce significantly smaller [Ca2+]i changes and reduced contractions for similar Ca2+ currents (see Fig. 5 B). These differences are demonstrated more clearly when the voltage dependence for these data is shown in Fig. 6. There is also a clear reduction in the rate of cellular shortening shown in Fig. 5. Figure 6 D shows the voltage dependence decrease in shortening and relaxation in the cardiac myocytes taken from MLP−/− animals.
To better understand the reduced contractility found in vivo using a P-V analysis, we examined the dependence of cellular shortening on [Ca2+]i during the contractile cycle at representative depolarizations (−20, 0, and +20 mV), as shown in Fig.7. There is a clear hysteresis in the relationship, as illustrated in Fig. 7 A, middle. The arrows along the dashed line reveal the trajectory of the plot. The hysteresis reflects the rapid release of Ca2+ by the EC coupling mechanism (Ca2+-induced Ca2+release, CICR) and the relatively slower responsiveness of the contractile machinery. Hysteresis in wild-type mouse heart cells is expected because of the known kinetics of contraction. In MLP−/− animals, the heart cells show a characteristic flattened hysteresis loop. This can be seen best if one compares Fig.7 A, left, with Fig. 7 B,middle. These examples have a similar elevation of [Ca2+]i, but the extent of the shortening is considerably decreased (∼5%) in the MLP−/− animals compared with controls (∼10%). The data in Fig. 7 B show the consistent pattern of reduced contractile responsiveness to [Ca2+]ichanges. The reduced extent of contraction leads to the flattened appearance of each of the trajectories shown in Fig. 7 B. Together, the data suggest that, in addition to the decreased shortening due to the reduction in the [Ca2+]i transients (as shown in Fig.6 C), there is a further reduction in contractility due to an additional factor that has not yet been defined. In MLP−/− heart failure, the additional factor may be alterations in troponin or other contractile proteins that have been observed in heart failure (28, 39) or may be directly related to the chronic loss of MLP.
The contractile defects revealed in Figs. 5, 6, and 7 are further examined in Fig.8 A, where peak shortening is plotted against peak [Ca2+]i. There is the clear trend that the steep relationship seen in cells from wild-type animals is less steep in cells from the MLP−/− animals. This reinforces the data shown in Fig. 7 but does not answer the question of considerable interest. Why is the [Ca2+]i transient reduced in the cells from the MLP−/− mice?
To determine why the [Ca2+]i transients are smaller in MLP−/− mice, experiments were done under conditions that produce similar SR loading (16, 35, 37,41). This control is important because changes in SR Ca2+ load alone can account for changes in EC coupling gain (35, 37). Under such conditions, there are at least two clear ways by which the [Ca2+]i transient can be reduced. First, the elementary units of SR Ca2+ release, the Ca2+ sparks, can be altered. They could be smaller or shorter in duration, for example. Figure 8 B shows typical Ca2+ sparks from cells taken from control and MLP−/− animals. The Ca2+ sparks are found to be indistinguishable, as shown in the sample records and the signal-averaged Ca2+ sparks. Second, EC coupling gain could be altered (16, 34, 37, 41). The EC coupling gain function indicates how well I Ca activates SR Ca2+ release. At negative potentials (i.e., negative to −20 mV), the single-channel current is largest, and the opening of a single L-type Ca2+ channel can activate an elementary SR Ca2+-release unit (9), a Ca2+spark (9, 34). Thus it is at these negative potentials where EC coupling gain reflects the efficacy of the coupling between Ca2+ influx through L-type Ca2+ channels and the triggered response known as CICR (9, 34). Thus we note at negative potentials in Fig. 8 C that there is a significant reduction in EC coupling gain in MLP−/−myocytes compared with control myocytes. At more positive potentials (0 and +20 mV), the single-channel current of L-type Ca2+channels is reduced, and the opening of a single L-type Ca2+ channel is not sufficient to activate a Ca2+ spark. Thus, as test depolarizations reach more positive potentials, local [Ca2+]i becomes relatively less important and global (i.e., cell wide) [Ca2+]i becomes more important. For that reason, EC coupling gain declines and differences in EC coupling gain between wild-type and MLP−/− cells vanish (see Ref.34). It is important to remember that in this analysis, SR Ca2+ load is being artificially kept constant to permit comparison of EC coupling gain in cells from the different mouse types. The reduction of EC coupling gain in MLP−/− heart cells indicates that the efficacy of activation of Ca2+ sparks by Ca2+ influx through the L-type Ca2+ channel is reduced in MLP−/− heart cells (16). This important observation provides us with the primary evidence that there is a defect in EC coupling in MLP−/− myocardial cells.
βARKct Transgene Expressed in MLP−/− Mice
In vivo function.
We examined the effect of overexpression of the βARKct transgene on cardiac function. The expression of this transgene largely prevents the downregulation and desensitization of the β-AR by inhibiting the action of βARK1 to phosphorylate the receptor (31). Figure 1, right, shows that cardiac function measured by echocardiography has largely returned to normal, a finding similar in principle to our earlier report (31) but now refined by improved anesthesia and echocardiography techniques. Most importantly, however, the P-V analysis shown in Fig. 2 D is remarkable. It shows that the most dramatic change in the MLP−/−/βARKct animals is the reduction in end-diastolic and end-systolic cardiac volumes, which have become near normal. Just as important is the increased steepness of the ESPVR with increasing afterload, which approaches that observed in wild-type animals. Furthermore, the increased slope in the ESPVR after the application of the β-AR agonist dobutamine seen in wild-type animals is restored in the MLP−/−/βARKct animals. This functional restoration of heart function in the MLP−/−/βARKct animals is also clearly shown in Fig. 3. These data are also supported by the LV/body weight data showing normalization of cardiac mass in the MLP−/−/βARKct mice (wild type 3.28 ± 0.03 mg/g, MLP−/− 5.13 ± 0.21 mg/g, MLP−/−/βARKct 3.55 ± 0.33 mg/g;P < 0.005 MLP−/− vs. either wild type or MLP−/−/βARKct).
Cellular function in cells from MLP−/−/βARKct animals.
The decrease in [Ca2+]i transient observed in cells from the hearts of the MLP−/− mice is profound. There is, however, a remarkable change in the [Ca2+]i transients of the heart cells in the MLP−/−/βARKct animals, as shown in Fig. 5 C. Similarly, the cellular contractions are restored to the control levels as are rates of shortening and relaxation. These alterations in [Ca2+]i and cell shortening are observed widely over the voltage range −40 to +50 mV, as shown in Fig. 6, B andC. Importantly, the Ca2+ current density (pA/pF) is identical for cells from control, MLP−/−, and MLP−/−/βARKct animals. The dependence of cell shortening as a function of [Ca2+]i for the cells from MLP−/−/βARKct animals, as shown in Fig.7 C, is very close to that of the wild-type animals. Compare, for example, the similarity of the shape in the sample records from wild-type and MLP−/−/βARKct cells at −20 mV [Fig. 7,A and C, left]. The plot of peak cell shortening vs. [Ca2+]i shown in Fig.8 A for cells from the MLP−/−/βARKct animals has returned toward control (see discussion). In our investigation of why the [Ca2+]i transients returned to normal, we examined Ca2+ sparks and the EC coupling gain function in the cells from the MLP−/−/βARKct animals. We found that the Ca2+ sparks in MLP−/−/βARKct myocytes were similar to those in control cells. Because Ca2+ sparks in the MLP−/− myocytes were also identical to those in control cells, we learn about the functional similarity of the SR/T-tubular junction in the myocytes of the three types of mice we examined. We would conclude, therefore, that there is no significant difference in the ryanodine receptor cluster organization at SR/T-tubular junctions. Furthermore, because Ca2+ sparks in the MLP−/− myocytes were identical to those in the control myocytes, it also shows that there is no defect in SR Ca2+ uptake in the MLP−/− cells. However, as in other models of heart failure (8, 16, 34, 37, 41), there is a clear difference in the EC coupling gain function. In the experiments with MLP−/−/βARKct mice, however, we find that the EC coupling gain function in cells from the MLP−/−/βARKct animals has returned to normal.
To better understand the pathogenesis of heart failure and to explore novel treatments, we have examined the cellular and molecular defects that develop with well-defined models of heart failure (16, 41). Here we study a distinct model of heart failure that occurs when MLP is “knocked out” (i.e., the MLP−/− mouse) (1, 31). Echocardiographic abnormalities consistent with dilated cardiomyopathy and heart failure (1, 29, 31) are known, but intrinsic contractile and cellular defects have not been investigated until now (see Figs.2-8). We have identified functional changes during in vivo measurements of cardiac pressure and volume consistent with an abnormality in intrinsic contractility. We have also observed an impaired β-AR response and a major defect in EC coupling that we identified as a decreased sensitivity of the SR Ca2+release mechanism to triggering Ca2+. There is an additional defect in Ca2+ signaling consistent with reduced sensitivity of the contractile proteins to Ca2+. These findings in the MLP−/− animals are interesting because similar changes have been identified in other animal models of heart failure that are attributed to unrelated instigating insults [i.e., hypertension (16) or viral myocardiopathy (41)]. The most interesting and provocative finding, however, is that the heart failure pathology in MLP−/−animals can be almost completely recovered by the expression of a single transgene that functionally inhibits the action of βARK1. The most critical questions raised by these findings are discussed below.
Cellular and Molecular Mechanisms of Heart Failure
MLP−/− dilated cardiomyopathy.
Because the MLPs are reported to link actin filaments together at the Z line, it has been argued that the absence of MLP may contribute to the development of dilated cardiomyopathy by the removal of its structural function (10) . The data presented here appear to argue against that simple conclusion. The reasoning is that there is virtually full recovery of cardiac contractile function in the MLP−/−/βARKct animals and yet the MLP remains absent. Our conclusion is supported by Minamisawa et al. (29), who found that MLP−/− cardiomyopathy appears to be rescued by knocking out a different protein, phospholamban (PLB).
Initiating insult in MLP−/− animals.
The data to date do not permit us to identify which feature of the MLP−/− pathology leads to the dilated cardiomyopathy, and thus allows the conclusion that some of the structural role of MLP is important. MLP, however, subserves many functions in addition to its putative structural role. MLP contributes to muscle development and serves as a nuclear transcription regulator (10). There are no compelling data to date to indicate which of these many functions of MLP is critical to the development of dilated cardiomyopathy.
Why is the MLP−/− mouse rescued by cardiac-targeted expression of βARKct?
We postulate that the MLP−/− dilated cardiomyopathy arises as the result of a pathological spiral that begins with the yet-unidentified initiating insult. The insult is inadequate to produce enormous contractile dysfunction on its own but can lead to sufficient dysfunction so that the β-AR system is activated in compensatory response. We reason that activation of the β-AR system is the primary means to increase cardiac output when it must be increased transiently. However, tonic activation of the β-ARs is not a normal response and becomes maladaptive because tonic β-AR occupancy by agonist increases β-AR desensitization and downregulation mediated in part by phosphorylation of receptors by βARK1. This hypothesis, first suggested by us (32), is supported by our findings presented here, where we show that there is virtually normal in vivo cardiac function with normalization of β-AR responsiveness in the MLP−/−/βARKct mouse. This can be better understood in light of the restoration of cellular function in myocytes from these animals. Presumably, the main benefit of expressing the βARKct transgene in the hearts of the MLP−/− mice is the protection of β-ARs from chronic desensitization and downregulation.
We propose that normal β-AR function is important for the normal function of the heart. The β-AR, like other heptahelical receptors, not only serves to signal specific G protein-coupled effectors such as adenylyl cyclase but also signals a host of additional effectors through diverse intracellular proteins (17). Chronic downregulation and desensitization of β-ARs will blunt all cAMP-dependent protein kinase-mediated dependent signaling but will activate other intracellular pathways, such as those linked to cell growth (e.g., the mitogen-activated protein kinases) (25). We suggest that overexpression of the βARKct transgene reverses these maladaptations.
MLP−/−/PLB−/− rescue of MLP−/−.
How effective is the MLP−/−/PLB−/− rescue (see Ref. 29) compared with the MLP−/−/βARKct rescue shown here? How does this rescue come about? Although we would like to know the answer to these questions, there is no way to address them properly at this time because neither P-V relations nor patch-clamp and confocal cellular [Ca2+]i signals were studied in the MLP−/−/PLB−/− mouse (29).
Fractional shortening in wild-type animals investigated by Minamisawa et al. (29) was quite low, as were heart rates, suggesting that the level of anesthesia may have been high and thus may have affected these measurements. We adjusted the level of anesthesia in our experiments to reduce potential toxicity and performed echocardiography under conscious conditions. Furthermore, heart rates measured during conscious echocardiography were similar to those obtained by telemetry in ambulatory mice, indicating little stimulation by the procedure (18). It is also interesting to note that the MLP−/− mice rescued by the PLB knockout did not regain β-AR responsiveness but remained in a highly stimulated state. This is similar to the result we found with cardiac-targeted expression of the β2-AR, which we showed to be deleterious in the MLP−/− mouse (31).
It is widely accepted that LV dP/dt max provides useful information on relative changes in contractile behavior when all conditions are controlled in the same animal, (33); however, it has limited sensitivity when groups of different animals are compared (12, 27). In this regard, one of the best measures of cardiac contractile function in vivo is the P-V loop (23, 33), as we have used here.
P-V loops in the mouse.
The analysis of P-V relations with the ESPVR has long been established as a standard method to evaluate cardiac function in larger species. This study in the mouse shows the nonlinear nature of the LV ESPVR (21, 30) and demonstrates how the shift of this relation can detect changes in contractility in the same heart but, most importantly, can identify differences in inotropic state between groups of genetically altered hearts. Recently, it was reported that ESPVRs were obtained by miniaturized conductance micromanometry (14). This study nicely demonstrates the utility of obtaining load-independent measures of ventricular performance to better understand the biology of disease using gene-targeted animals. Conversion to absolute volumes can be achieved using the hypertonic saline dilution method, but this can result in marked hemodynamic changes (42). In our study, we placed two orthogonal pairs of miniature piezoelectric crystals in the endocardial wall of the mouse heart. Because the output of the dimension crystal is a calibrated signal, we were able to obtain internal dimensions in two planes with high accuracy and therefore calculated absolute LV volume throughout the cardiac cycle. An important advantage of this technique is the ability to measure instantaneous wall thickness by placing an additional crystal on the epicardial surface juxtaposed to the endocardial crystal. This will allow for the calculation of wall stress throughout the cardiac cycle and will provide a method for the in-depth analysis of cardiac mechanics in gene-targeted murine hearts. It should be noted that the sonomicrometer technique requires open-chest instrumentation, which is technically demanding and will affect the measurement of basal hemodynamic parameters. For the assessment of murine cardiac function, this technique is complementary to that of closed-chest catheter-based hemodynamic and echocardiography measurements and provides an assessment of intrinsic in vivo contractile function in a load-independent manner.
In summary, a model of dilated cardiomyopathy with heart failure (MLP−/−) was used to investigate cellular defects in heart failure and possible new therapies. We have extended our preliminary result (31) by examining the in vivo and cellular behavior of the relevant hearts. Importantly, we show that by expressing the cardiac-targeted βARKct transgene in MLP−/− animals a nearly complete restoration of contractile function can be achieved. The compelling restoration of function by transgenic expression of βARKct suggests that modeling the mechanism of action of the βARKct is an attractive therapeutic approach.
We gratefully acknowledge Debbie Colpitts for expert secretarial assistance.
↵* G. Esposito, L. F. Santana, and K. Dilly contributed equally to this work.
This work was supported in part by National Heart, Lung, and Blood Institute Grants HL-61558 (H. A. Rockman) and HL-36974, HL-61602, and HL-25675 (W. J. Lederer). H. A. Rockman is a recipient of a Burroughs Wellcome Fund Clinical Scientist Award in Translational Research.
Address for reprint requests and other correspondence: H. A. Rockman, Dept. of Medicine, Duke Univ. Medical Center, DUMC 3104, Durham, NC 27710 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society