The Cardiac-specific Nuclear (cid:1) B Isoform of Ca 2 (cid:2) /Calmodulin-dependent Protein Kinase II Induces Hypertrophy and Dilated Cardiomyopathy Associated with Increased Protein Phosphatase 2A Activity*

The (cid:1) isoform of Ca 2 (cid:2) /calmodulin-dependent protein kinase II (CaMKII) predominates in the heart. To investigate the role of CaMKII in cardiac function, we made transgenic (TG) mice that express the nuclear (cid:1) B iso- form of CaMKII. The expressed CaMKII (cid:1) B transgene was restricted to the myocardium and highly concentrated in the nucleus. Cardiac hypertrophy was evidenced by an increased left ventricle to body weight ratio and up-regulation of embryonic and contractile protein genes including atrial natriuretic factor, (cid:3) -myosin heavy chain, and (cid:4) -skeletal actin. Echocardiography revealed ventricular dilation and decreased cardiac function, which was also observed in hemodynamic measurements from CaMKII (cid:1) B TG mice. Surprisingly, phospho- rylation of phospholamban cardiac-specific (cid:1) heavy (MHC) to demonstrate here decreased sarcoplasmic (SR) Ca (cid:1)

Although cardiac hypertrophy is a beneficial adaptive response of the heart to a variety of intrinsic and extrinsic stimuli, chronic hypertrophy often leads to dilated cardiomyopathy and eventually to heart failure if the stimulus is not relieved (1). A hallmark of the pathological transition from hypertrophy to heart failure is decreased cardiac contractility. Cardiac hypertrophy has also been reported to be an independent risk factor for ischemic heart disease, arrhythmia, and sudden death (2). The development of these pathophysiological end points highlights the need to understand the cellular signaling events that regulate cardiac growth and function.
Ca 2ϩ signals have long been known to play a central role in the regulation of cardiac contractility and more recently have been considered as likely regulators of growth and gene expression. A role for Ca 2ϩ signaling in the pathogenesis of cardiac hypertrophy is supported by a growing body of evidence. For example, transgenic (TG) 1 mice with a 3-5-fold overexpression of the Ca 2ϩ -binding protein, calmodulin, develop severe cardiac hypertrophy (3). This chronic elevation of calmodulin in the ventricles of TG mice was more recently shown to increase CaMKII phosphorylation, an index of the Ca 2ϩ -independent activity of CaMKII, and the expression of atrial natriuretic factor (ANF) (4), an established indicator of ventricular hypertrophy. In cultured ventricular myocytes, electrical pacing can elevate intracellular Ca 2ϩ , and this is essential for cardiac myocytes to respond with increased expression of ANF and myosin light chain-2, another marker of hypertrophy (5). In addition, the calmodulin antagonist W-7 can block hypertrophy of primary cultured cardiomyocytes in response to electrical pacing and ␣-adrenergic stimulation (5), further implicating Ca 2ϩ as a mediator of cardiac gene expression in the hypertrophic response. An alternate effector of the Ca 2ϩ /calmodulin complex is calcineurin, a protein phosphatase that has recently attracted attention as a mediator of hypertrophic stimuli in vitro and in vivo (6,7).
Our lab previously showed that transient expression of CaMKII␦ B in neonatal rat ventricular myocytes induced gene expression and resulted in an enhanced response to phenylephrine, as assessed by transcriptional activation of an ANF-luciferase reporter gene (23). The nuclear localization signal of CaMKII␦ B was required for this response because expression of CaMKII␦ C did not result in enhanced ANF expression (23). Activated CaMKI and IV can also induce hypertrophic responses in cultured cardiomyocytes (24) and the CaMKII inhibitor KN-62 can block cardiomyocyte hypertrophy in response to endothelin-1 (25). Of particular significance is the demonstration that hypertrophy develops in TG mice that express increased levels of CaMKIV in the myocardium (24). CaMKIV is expressed at very low levels in the heart relative to CaMKII (8,18), but it has, in common with the predominant cardiac CaMKII␦ B isoform, the ability to enter the nucleus.
To investigate the role of CaMKII in the intact heart, we used the well characterized cardiac-specific ␣-myosin heavy chain (MHC) promoter to generate TG mice that express CaMKII␦ B isoform. We demonstrate here that overexpression of wild type CaMKII␦ B in the mouse heart induces cardiac hypertrophy and dilation with decreased ventricular function and that this is associated with changes in protein phosphorylation, phosphatase activity, and sarcoplasmic reticulum (SR) Ca 2ϩ uptake.

EXPERIMENTAL PROCEDURES
Generation of CaMKII␦ B Transgenic Mice-Hemagglutinin (HA)tagged rat wild type CaMKII␦ B cDNA was subcloned into a pBluescriptbased TG vector (a gift from J. Robbins, University of Cincinnati, Cincinnati, OH) between the 5.5-kb murine ␣-MHC promoter and a polyadenylation signal. Purified linear transgene fragments were injected into pronuclei of fertilized mouse oocytes. The resultant pups were screened for the presence of the transgene by PCR as described previously (26), using a CaMKII-specific primer (5Ј-TTGAAGGGTGC-CATCTTGACA-3Ј) and a TG vector-specific primer (5Ј-CGCTCTA-GAACTAGTGGACT-3Ј). Founder mice were bred with C57BL/6 wild type mice. Heterozygous animals from at least the third generation were used for all studies, with their wild type (WT) littermates serving as controls. All procedures were performed in accordance with Guide for the Care and Use of Laboratory Animals (45) and approved by the Institutional Animal Care and Use Committee.
Kinase Assay-Frozen powdered ventricular tissue was resuspended in ice-cold kinase sample buffer (50 mM PIPES, 10 mM EGTA, 20 mM benzamidine, 1 g/ml aprotinin, 1 g/ml leupeptin, 1 g/ml pepstatin, 1 mM phenylmethylsulfonyl fluoride, and 10% glycerol) prior to sonication for 10 s. Protein concentration was measured by Bradford assay, and the homogenates were diluted to 10 mg/ml protein. Kinase activity was measured using a previously published method (27).
Phosphatase Assays-Frozen powdered ventricular tissue was homogenized in ice-cold lysis buffer (20 mM HEPES, pH 7.5, 100 mM NaCl, and protease inhibitors as a mixture from Sigma-Aldrich). The homogenized hearts were centrifuged in an air centrifuge for 10 min in a cold room. The supernatant was considered to be cytosol, and the pellet was treated with lysis buffer containing 1% Triton X-100 to solubilize the proteins. This last extract was centrifuged as before, and the supernatant was considered as particulate. Protein phosphatase 1 (PP1) activity was measured using 32 P-labeled phosphorylase as a substrate, and protein phosphatase 2A (PP2A) activity was measured using a 32 Plabeled synthetic peptide (RRATpVA) that is selective for PP2A as described recently (28).
Immunoprecipitation and Western Blotting-Cardiac homogenates were prepared as described previously (29). Cytosolic and particulate fractions were prepared as stated above. In some experiments, a nuclear fraction was prepared from mouse hearts using a Wheaton Dounce homogenizer as described previously (30). The antibodies used for immunoprecipitation and immunoblotting were as follows: mouse anti-HA (Roche Molecular Biochemicals), CaMKII␦ antibody (rabbit antiserum against a 15-amino acid peptide in the carboxyl-terminal region of CaMKII␦), monoclonal anti-PLB (Upstate Biotechnology, Inc.), phosphorylated PLB (Thr 17 and Ser 16 ) antibodies (Fluorescience, Leeks, UK), CaMKII antibody (Santa Cruz), phosphorylated CaMKII antibody (Affinity Bioreagents), PP2A/C and PP2A/B56␣ antibodies (BD Transduction Laboratories), and PP2A/A antibody (Oxford Biomedical Research). The secondary antibody is a horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse IgG (Sigma-Aldrich). Enhanced chemiluminescence was performed using the SuperSignal Chemiluminescent Detection System (Pierce).
RNA Dot Blot Analysis-RNA was prepared from ventricular tissue using Trizol reagent (Invitrogen), and dot blot analysis was performed as described previously (26,31).
Hemodynamic Measurements-Mice were anesthetized with a mixture of ketamine (100 mg/kg) and xylazine (5 mg/kg) intraperitoneally. After endotracheal intubation, the mice were connected to a volumecycled rodent ventilator. A PE-50 catheter was placed in the left jugular vein and used for intravenous access. Through the right carotid artery a 1.8-French high fidelity catheter tip micromanometer was inserted via a small incision, and the tip was manipulated across the aortic valve into the left ventricle (LV). A bilateral vagotomy was performed. When LV pressure and heart rate became stable, dobutamine was given intravenously by an infusion pump, at a rate at 0.75, 2, 4, 8, and 12 g/kg/min. Atrial pacing was accomplished using a guide wire placed in the right atrium via the right jugular vein. Prior to pacing, the heart rate was controlled at a slow, stable rate using UL-FS 49, a selective sinus node inhibitor, at a dosage of 0.05 mg given intravenously to achieve a rate of 150 -250 beats/min. Heart rate was then increased by atrial pacing in steps of 50 beats/min. There was 30 s between two steps. Then dobutamine was given intravenously at 2 g/kg/min for 3 min, and the pacing was repeated. Data analysis was done by computer, and 12 beats were averaged.
Transthoracic Echocardiography-Mice were anesthetized by intraperitoneal injection of 2.5% Avertin (15 l/g body weight). The current echocardiographic system is an Agilent Technologies, Sonos 5500 with a 15 MHz linear probe that utilizes ultraband technology. M-mode and Doppler tracing were recorded at a sweep speed of 150 mm/s. At least three independent M-mode measurements/animal were obtained by an examiner blinded to the genotype of the animals.
Measurements of SR Ca 2ϩ Uptake-Ventricular tissue from TG and littermate control mouse hearts was homogenized at 4°C in 1.5 ml of homogenizing solution (25 mM imidazole, pH 7.0) with a Teflon glass Thomas tissue grinder. SR Ca 2ϩ uptake assays were performed in ventricular homogenates at room temperature based on a protocol modified from that of Pagani and Solaro (32). Aliquots (350 l) of homogenates were transferred into tubes containing 2.8 ml of uptake buffer (100 mM KCl, 10 mM potassium oxalate, 40 mM imidazole, 10 mM sodium azide, 4.5 mM MgCl 2 ) and 45 Ca-EGTA buffer containing 0.185 Ci/ml 45 Ca (Amersham Biosciences, Inc.) and a given amount of free Ca 2ϩ (20 and 200 nM), which was calculated on the basis of the amount of added EGTA. After 5 min of preincubation, the uptake reaction was initiated by the addition of 2.5 mM sodium ATP. Ca 2ϩ uptake was terminated at various times (1, 3, and 5 min for 200 nM free Ca 2ϩ ; 1, 10, and 20 min for 20 nM free Ca 2ϩ ) by filtering 500-l aliquots on 0.45-m nitrocellulose membranes (Millipore-type HA), followed by two washes (5 ml) with uptake buffer without Ca 2ϩ or ATP. The radioactivity remaining on the nitrocellulose filters was determined by liquid scintillation spectroscopy. Protein concentration in the ventricular homogenates was assayed with a Bradford reagent. Ca 2ϩ uptake was calculated from the slope of the linear regression analysis relating 45 Ca 2ϩ uptake/milligram of protein to reaction time.
Immunocytochemical Staining-Ventricular myocytes were isolated from WT and CaMKII␦ B TG mice following recently published methods (33) and cultured on laminin-coated (3.5 g/cm 2 ; Upstate Biotechnology, Inc.) chamber slides overnight. Indirect immunofluorescence stainings were performed as described previously (34). The CaMKII␦ B transgene was detected using rabbit anti-HA antibody (Santa Cruz, 1:100 dilution) followed by fluorescein isothiocyanate-conjugated goat antirabbit IgG antibody (Cappel, 1:100 dilution). The cells were observed by a Zeiss Axiovert 135 fluorescence microscope and photographed using a CCD camera.
Statistical Analysis-All of the data are reported as the means Ϯ S.E. The statistical significance of difference between the WT mice and CaMKII␦ B TG mice was determined using unpaired Student's t test. A p value of Ͻ0.05 was considered statistically significant.

Generation and Identification of CaMKII␦ B Transgenic
Mice-TG mice expressing HA-tagged rat wild type CaMKII␦ B under the control of the cardiac-specific ␣-MHC promoter were generated as described under "Experimental Procedures." Three TG founders showed germline transmission of the transgene. Significant expression of the transgene was seen only in the heart based on examination with the anti-HA antibody (Fig. 1A). The CaMKII␦ B TG line studied in detail here showed at least a 10-fold overexpression of CaMKII␦ in the heart, based on Western blots using the anti-CaMKII␦ antibody (Fig.  1B). Enzymatic activity of CaMKII measured in ventricular homogenates was ϳ4-fold higher in the TG animals than in the littermate controls (Fig. 1C). The discrepancy between CaMK activity and expression may be due at least in part to the fact that the activity assay includes other isoforms of CaMKII. The phosphorylation state of CaMKII reflects its Ca 2ϩ -independent activity. Experiments using a phospho-CaMKII antibody revealed a ϳ1.5-fold increase in the CaMKII␦ B TG mice (Fig. 1D). Immunocytochemical staining of cardiomyocytes isolated from the TG animals using a HA antibody confirmed that the CaMKII␦ B transgene was, as predicted based on its nuclear localization signal, present and highly concentrated in the nucleus (Fig. 2).
Cardiac Overexpression of CaMKII␦ B Induces Cardiac Hypertrophy-Most CaMKII␦ B TG mice showed significantly enlarged hearts at 3ϳ4 months of age (that shown in Fig. 3A is a typical one). On average, CaMKII␦ B TG mice exhibited a 22% increase in heart weight to body weight ratio and a 27% increase in left ventricle to body weight ratio at 12 weeks of age (Fig. 3B). Body weights in the TG and WT mice were equivalent; thus the heart to body weight ratio increase in the TG mice is due to increased ventricular mass.
To determine whether specific alterations in cardiac gene expression were associated with CaMKII␦ B overexpression, RNA was isolated from mouse ventricles, and a selected panel of hypertrophic genes was examined by dot blot analysis. As shown in Fig. 3C, ANF, ␤-MHC, and ␣-skeletal actin mRNA levels were significantly increased in TG ventricles, and there was a modest but significant decrease in ␣-MHC, SERCA, and PLB mRNA levels.
Cardiac Overexpression of CaMKII␦ B Causes Ventricular Dilation and Decreased Contractile Function-To assess chamber size and cardiac function in CaMKII␦ B TG mice, we performed echocardiography on 3-4-month-old mice. As shown in Fig. 4, left ventricular end diastolic diameter and left ventricular end systolic diameter in TG mice were increased by 14 and 31%, respectively. The calculated left ventricular mass was increased by 20%, consistent with changes described above. Fractional shortening and velocity of circumferential shortening, which reflect left ventricular contractile function, also decreased significantly in TG mice compared with the control mice.
Hemodynamic measurements of cardiac contractility and relaxation demonstrated that both ϩdP/dt and ϪdP/dt responses to pacing were blunted in hearts of TG mice (Fig. 5, A and B). Stimulation with dobutamine, a ␤-adrenergic receptor agonist, should increase Ca 2ϩ , either directly or indirectly and might therefore activate CaMKII␦ B in the TG mice. However, dose response curves to this ␤-adrenergic agonist were not significantly altered in TG versus WT mouse hearts (Fig. 5, C and D).
Cardiac Overexpression of CaMKII␦ B Results in Reduced Ca 2ϩ Uptake and Decreased Phosphorylation of Phospholamban-The decreased relaxation properties observed in the hemodynamic measurements led us to assess changes in SR Ca 2ϩ uptake. Experiments examining Ca 2ϩ uptake in SR containing cardiac homogenates revealed that SR Ca 2ϩ uptake was significantly decreased in ventricles from 3-4-month-old TG mice (Fig. 6A). Because SR Ca 2ϩ uptake is regulated by PLB phos-FIG. 1. Identification of CaMKII␦ B transgenic mice. A, protein isolated from different TG mouse organs was immunoblotted with anti-HA antibody. B, protein derived from whole hearts isolated from WT and TG mice was immunoblotted with anti-HA or anti-CaMKII␦ antibodies. C, CaMKII activity was measured in ventricular homogenates from WT and TG mice (n ϭ 5 for each group). The data are presented as the means Ϯ S.E. *, p Ͻ 0.001 versus WT. D, phosphorylated CaMKII was measured in ventricular homogenates by Western blots (n ϭ 6 for each group). The data are presented as the means Ϯ S.E. *, p Ͻ 0.05 versus WT.
FIG. 2. Immunocytochemical staining of ventricular myocytes isolated from WT and CaMKII␦ B TG mice. Ventricular myocytes were isolated from WT and TG mice and cultured on laminin-coated chamber slides overnight. Indirect immunofluorescence stainings were performed. Transgene was detected by rabbit anti-HA antibody (1:100 dilution) followed by fluorescein isothiocyanate-conjugated goat antirabbit IgG antibody (1:100 dilution). The transgene is present and highly concentrated in the nucleus. 100 cells of each group were examined. Cytoplasmic staining was not observed in any cells.
phorylation and by the SERCA/PLB ratio, we asked whether the decrease in SR Ca 2ϩ uptake could be the result of altered PLB phosphorylation or an altered SERCA/PLB ratio in the CaMKII␦ B TG mice. PLB is regulated by phosphorylation of Thr 17 by CaMKII and of Ser 16 by PKA. Interestingly, both Thr 17 -and Ser 16 -phosphorylated PLB were significantly decreased in extracts prepared from 3-4-month-old TG mouse hearts (Fig. 6B). There was no change in total PLB protein or SERCA protein (data not shown). Because the basal state of PLB phosphorylation is generally low, we also examined PLB phosphorylation in hearts rapidly frozen after stimulation of ␤-adrenergic receptors with dobutamine. The levels of Thr 17and Ser 16 -phosphorylated PLB remained significantly lower in TG versus WT mice even following stimulation (Fig. 6C). Thus maximal phosphorylation of neither the PKA nor the CaMKII site could be achieved in the TG animals.
PP2A Is Activated in CaMKII␦ B Transgenic Mice-The decreased phosphorylation of PLB was unexpected and led us to hypothesize that phosphatases might be increased in CaMKII␦ B TG mice. We therefore assayed PP1 and PP2A phosphatase activity in WT and TG mouse hearts. As shown in Fig.  7A, PP2A activity was selectively increased in the particulate fraction from TG mouse hearts. There was no change in PP2A activity in the cytosol (Fig. 7A), nor was PP1 activity altered in either fraction (data not shown). Western blots examining phosphatase expression levels also showed no change in PP1 or PP2A in the cytosol from TG hearts (data not shown). However, in the particulate fraction from TG hearts, increases in expression of the catalytic as well as the A subunit and the B56␣ targeting subunit of PP2A were observed (Fig. 7B). This did not occur at the transcriptional level, because Northern blot analysis using a 32 P-labeled PP2A catalytic subunit cDNA fragment as a probe revealed no difference in PP2A mRNA levels in TG and WT hearts (data not shown).
There is growing evidence for association of phosphatases and kinases in signaling complexes. In particular, CaMKIV has been shown to interact with PP2A catalytic subunit (35). To determine whether there was a direct interaction between PP2A and CaMKII in cardiomyocytes, we first examined association of these molecules in neonatal rat ventricular myocytes. Immunoprecipitation with either CaMKII or PP2A catalytic subunit antibody followed by Western blots with the opposite antibody demonstrated that CaMKII associates with PP2A in neonatal rat ventricular myocytes (data not shown). In addition, we demonstrated that there was an increase in PP2A catalytic subunit that coimmunoprecipitated with CaMKII in the particulate fraction (Fig. 7C) as well as in the nuclear fraction (Fig. 7D) prepared from TG mouse ventricles. DISCUSSION A variety of studies performed throughout the last decade suggest that Ca 2ϩ signaling is a central mechanism triggering hypertrophic growth (3)(4)(5)(6). A likely sensor for the effects of elevated calcium is CaMK, an enzyme suggested to mediate changes in cell growth and gene expression in a number of neuronal systems (36,37). Most studies linking CaMK to control of gene expression have focused on the effects of CaMKIV because it is a momoneric enzyme that, in contrast to the multimeric CaMKII, is readily able to enter the nucleus. Recently published work demonstrated that CaMKIV could induce hypertrophic responses in cardiomyocytes in vitro and that CaMKIV expression can cause cardiac hypertrophy in TG mice (24). In contrast, there is no information concerning the effects of CaMKII␦, the CaMK subtype that predominates in FIG. 3. Cardiac hypertrophy in CaMKII␦ B TG mice at 3-4 months of age. A, whole hearts from WT and littermate TG mice at 15 weeks of age. B, heart weight/body weight (ϫ1000) and LV/body weight (ϫ1000) ratios were measured at 12 weeks of age. The data are presented as the means Ϯ S.E. *, p Ͻ 0.001 versus WT. C, ventricular gene expression in CaMKII␦ B TG mice. RNA isolated from ventricular tissue of WT and TG mice (n ϭ 6 for WT, n ϭ 10 for TG) at 12 weeks of age was subjected to dot blot analysis using gene transcript-specific antisense oligonucleotide probes as indicated. GAPDH was used as the normalizing control in each experiment. SK.Actin, ␣-skeletal actin; CA.Actin, ␣-cardiac actin. The data are presented as the means Ϯ S.E. *, p Ͻ 0.05 versus WT; **, p Ͻ 0.01 versus WT.  6. SR Ca 2؉ uptake and Western blots of phosphorylated PLB in WT and CaMKII␦ B TG mice. A, Ca 2ϩ uptake in ventricular homogenates from WT and CaMKII␦ B TG mice. SR Ca 2ϩ uptake assays were performed in ventricular homogenates at room temperature. Ca 2ϩ uptake was calculated from the slope of the linear regression analysis relating 45 Ca 2ϩ uptake per milligram of total protein to reaction time (n ϭ 5 for each group). *, p Ͻ 0.05 versus WT. B, Thr 17 -and Ser 16phosphorylated PLB were significantly decreased in TG hearts (n was between 5 and 12 for all groups). *, p Ͻ 0.05 versus WT; **, p Ͻ 0.01 versus WT. Total PLB was unchanged (data not shown). C, ventricles were removed and rapidly frozen for 3 min following infusion of dobutamine (12 g/kg/min) (n ϭ 6 for each group). PLB phosphorylation at Thr 17 and Ser 16   the heart, on myocardial cell growth in vivo.
The findings reported here utilized the ␣-MHC promoter to drive the expression of the wild type isoform of CaMKII␦ B . This approach has been applied to a number of genes and has been shown to lead to considerable increases in mRNA and generally also in the protein of interest in the TG mouse heart. In the present report we assessed not only increases in protein expression but also examined concomitant changes in enzyme activity and in the extent to which there is active enzyme in the myocardium in vivo. Interestingly, these further measures allow one to "rationalize" the use of 10-fold overexpression by demonstrating that this leads to only a 4-fold increase in the total cellular CaMKII pool and only a 1.5-fold increase in the amount of active CaMKII in vivo. The modest increase in the amount of active kinase associated with overexpression of the wild type form of CaMKII probably accounts for the relatively mild phenotype. On the other hand, the increase is clearly within a range that would reflect physiological increases in CaMKII activity.
A variety of end points confirm that CaMKII␦ B TG mice develop cardiac hypertrophy by 3-4 months of age. These include statistically significant increases in heart and left ventricle to body weight ratios, up-regulation of embryonic and contractile protein genes including ANF, ␤-MHC, and ␣-skeletal actin, and down-regulation of ␣-MHC, SERCA, and PLB genes. These hypertrophic changes in ventricular mass and gene expression were associated with development of a dilated cardiomyopathy in the CaMKII␦ B TG mice. Specifically, these mice showed ventricular dilation and decreased contractile function, as documented by echocardiography, and impairment of the force-frequency response, as assessed by hemodynamic measurements. The observation that overexpression of CaMKII␦ B can induce cardiac hypertrophy and dilated cardiomyopathy indicates that the phenotype previously reported for CaMKIV TG mice (24) mirrors that elicited by expression (and presumably activation) of the endogenous cardiac CaMK.
PLB is a substrate for CaMKII. Although we did not anticipate enhanced phosphorylation of this cytosolic substrate by the nuclear CaMKII␦ B , we were surprised to observe that PLB phosphorylation was actually decreased. Notably phosphorylation of both the CaMKII and the PKA site were decreased by 33ϳ50% under conditions of both basal activity and adrenergic stimulation. Furthermore, functional effects of the diminished phosphorylation were evident as a decrease in contractility and relaxation in vivo and a decrease in SR Ca 2ϩ uptake in vitro. The decreased PLB phosphorylation associated with heart failure has been suggested to result from ␤-adrenergic receptor desensitization (38,39); however, hemodynamic measurements did not reveal significantly diminished ␤-adrenergic responsiveness in the CaMKII TG mice. The basis for the global decrease in PLB phosphorylation seen in the CaMKII␦ B TG mice would therefore appear to be the increase in phosphatase activity. Our preliminary observation (data not shown) that PLB phosphorylation is not decreased, and the fact that phosphatase activity is not significantly elevated in 6-week-old TG mice further suggests that PLB phosphorylation is diminished secondary to increases in phosphatase activity.
Decreased basal levels of Ser 16 and Thr 17 phosphorylated PLB and prolonged relaxation were recently reported in postinfarction remodeled myocytes and partially explained by increased PP1 activity (40). Although there is evidence that PP1 is mainly responsible for the dephosphorylation of PLB (41,42), PP2A can also dephosphorylate this substrate (41,43). We observed no change in PP1 activity in CaMKII TG mice. However, hearts from CaMKII TG mice had increased PP2A activity and increased expression of the catalytic, and regulatory A, and B56␣ subunits in the particulate fraction. The particulate fraction is comprised of membranes from various cellular organelles. Increased PP2A is evident in nuclei in association with CaMKII (see below); however, the changes in PLB phosphorylation suggested that PP2A might also be increased in the SR membrane. This was supported by the observation that PP2A protein level increased ϳ60% in the cytoplasmic membrane fraction, which would include SR (data not shown).
The observation that the increase in PP2A was confined to the membrane-associated particulate fraction suggested the possibility that PP2A was localized in a complex of signaling molecules. Accordingly we asked whether there was a direct association between PP2A and CaMKII. Studies using neonatal rat ventricular myocytes demonstrated that endogenous CaMKII coimmunoprecipitated with endogenous PP2A and vice versa. Furthermore, there was an increase in PP2A associated with CaMKII in the particulate fraction as well as in the nuclear fraction from TG mice. Although the mechanism for the increase in PP2A activity and its association with CaMKII is not known, we suggest that this alteration contributes to the development of the cardiomyopathy, particularly to the decreased contractile function observed in CaMKII␦ B TG mice. In this regard, it is of interest that a recent study reported that TG mice expressing a mutant A subunit of PP2A exhibit a dilated cardiomyopathy (44).
The generation and analysis of TG mice overexpressing CaMKII␦ B support the conclusion that physiologically relevant levels of activation of the predominant cardiac isoform of CaMK II can function within the nucleus to induce cardiac gene expression and hypertrophy. In addition, kinase increases in both the amount and the activity may be associated with compensatory increases in phosphatase activity, and these may contribute to changes in SR function and Ca 2ϩ handling that underlie the impairment of ventricular function.