Objective: Exogenous catecholamine exposure has been associated with p38 mitogen-activated protein kinase (MAPK) and cardiac hypertrophy. In this study, we investigated the regulation of p38 MAPK in cardiac remodeling elicited by endogenous adrenergic mechanisms. Methods: Transgenic male and female mice with fourfold phospholamban (PLB) overexpression exhibited enhanced circulating norepinephrine (NE), as a physiological compensatory mechanism to attenuate PLB's inhibitory effects. This enhanced noradrenergic state resulted in left ventricular hypertrophy/dilatation and depressed function. Results: Male transgenics exhibited ventricular hypertrophy and mortality at 15 months, concurrent with cardiac p38 MAPK activation. Female transgenics, despite similar contractile dysfunction, displayed a temporal delay in p38 activation, hypertrophy, and mortality (22 months), which was associated with sustained cardiac levels of MAP Kinase Phosphatase-1 (MKP-1), a potent inhibitor of p38. At 22 months, decreases in cardiac MKP-1 were accompanied by increased levels of p38 activation. In vitro studies indicated that preincubation with 17-β-estradiol induced high MKP-1 levels, which precluded NE-induced p38 activation. Conclusion: These findings suggest that norepinephrine-induced hypertrophy is linked closely with p38 MAP kinase activation, which can be endogenously modulated through estrogen-responsive regulation of MKP-1 expression.
Time for primary review 21 days.
Progressive increases in circulating plasma catecholamines represent a major prognostic indicator for heart failure, continuously adjusting to correct systolic and diastolic dysfunction. However, such prolonged imbalance within neurohormonal cascades is known to induce cardiac hypertrophy and toxicity [1,2]. Mechanistically, investigations into norepinephrine-induced hypertrophic signaling pathways have implicated the activation of mitogen-activated protein (MAP) kinases . Specifically, PKA and/or PKC may stimulate raf-1 independent of ras, with subsequent activation of p38 and ERK1/2 MAP kinases. Of the known members of the MAPK family, p38 promotes the strongest induction of hypertrophic gene expression and sarcomeric re-organization , suggesting that p38 signaling may be critical to the remodeling process. Furthermore, in response to adrenergic input, recent studies in adult cardiac myocytes have shown that activation of p38 may counterbalance the PKA-mediated positive contractile response by decreasing the myofilament responsiveness to calcium .
The MAPK signaling pathway is tightly regulated by a family of dual-specificity phosphatases . MAP Kinase Phosphatase-1 (MKP-1), which potently deactivates p38 , appears to be a critical determinant of p38-dependent hypertrophic growth in the heart. Indeed, transgenic overexpression of MKP-1 was associated with a suppressed p38 activation and hypertrophic response to either pressure overload or prolonged α-agonist infusion . A more recent study has linked catecholamine production to pressure overload hypertrophy, although the role of MKP-1 in these hearts was not pursued . Furthermore, MKP-1 protein expression may be positively regulated by estrogen in cardiac myocytes , suggesting that, in the context of a hypertrophic stimulus, differential regulation of p38 activity may exist between male and female hearts. Indeed a recent aortic banding pressure-overload model  demonstrated that estrogen replacement is associated with reduced p38 activation and attenuated hypertrophy, but the involvement of MKP-1 was not explored. Thus, it remains unclear whether endogenous regulation of MKP-1 expression can suppress p38-mediated hypertrophic signaling in response to endogenous catecholamine production, and whether gender can modulate this interaction amidst a prolonged adrenergic insult.
To further address the role of MKP-1 in p38 signaling, in vivo, this study explored the remodeling mechanisms in a transgenic mouse model of fourfold PLB overexpression, which develops a physiological, compensatory increase in circulating norepinephrine . Our data link endogenous adrenergic cardiotoxicity to p38 MAP Kinase activation and to reduced MKP-1 protein levels. Moreover, estrogen may protect the heart from structural decompensation through induction of MKP-1 with subsequent suppression of p38 activation.
PLB transgenic animals
Transgenic FVB/N mice (TG) were generated using a 6.4-kb transgene containing the α-myosin heavy chain promoter, PLB cDNA, and the SV40 polyadenylation signal . The investigation conforms with The Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85-23, revised 1996).
Whole hearts were pooled (n=4) and homogenized by age group and gender, prior to quantitative immunoblotting . Primary antibody was detected by peroxidase-conjugated (ECL, Amersham) secondary antibody. Protein concentrations were determined by the Bradford method (Bio-Rad) with a BSA standard.
In vivo echocardiographic measurement of cardiac function
Two-dimensional M-mode and color Doppler echocardiography was carried out as described .
Systolic blood pressure measurements
Tail-cuff measurements of systolic blood pressure were assessed in male and female, wild-type (WT) and TG animals at 3 months of age. Ten repetitive cycle measurements were averaged daily for five consecutive days to obtain average systolic pressures for each mouse (n=5 animals per group).
Plasma and tissue catecholamine levels
Norepinephrine, epinephrine, and dopamine levels were determined as previously described  (n=6 animals for each group).
MAP Kinase activation in cardiomyocyte preparations
Wild-type female mouse left ventricular cardiomyocytes were isolated  and resuspended in 1.8 mM (physiological) Ca2+-Tyrode buffer (pH 7.3). Pooled myocyte preparations (two hearts) were required for detection of activated MAP kinase levels. Equal volumes of cardiomyocytes were treated with either 1 μM norepinephrine (NE), or 1 nM 17β-estradiol (E2) at 37 °C. A third group was preincubated with 1 nM E2 for 30 min at 37 °C and treated with 1 μM NE. Aliquots were removed on ice and solubilized in equal volumes of deionized water and sample buffer (20% glycerol, 2% β-mercaptoethanol, 4% SDS, 0.001% bromophenol blue, and 130 mmol/l Tris–HCl (pH 6.8)) for 30 min. Samples were subjected to SDS–PAGE on 10% polyacrylamide gels and transferred to nitrocellulose membranes (0.2 μm), as described  (n=4 per group, two determinations per n).
Antibodies: PLB, monoclonal (Affinity BioReagents); PS-16 and PT-17, polyclonal (PhosphoProtein Research); Calsequestrin polyclonal (Affinity Bioreagents); SERCA polyclonal ; p38 and p44/42 MAPK phospho-forms (active and total) (New England Biolabs); MKP-1 (sc-1199, Santa Cruz Biotechnology).
Data are presented as mean±S.E.M. The number (n) of mice used is indicated with each figure. Statistical analyses were carried out using Student's t-test or two-way ANOVA.
Early mortality in male and female transgenic mice
Phospholamban overexpression is associated with compensatory increases in adrenergic drive to alleviate its inhibitory effects. However, chronic beta-adrenergic receptor activity results in cardiac hypertrophy, depressed left ventricular function and premature mortality . TG males displayed an early mortality between 15 and 18 months (Fig. 1A), which was associated with cardiac hypertrophy and fibrosis. Pulmonary congestion was not evident as lung-to-body mass ratios were not significantly different between aging TG and WT mice (mg/g: WT, 5.12±0.41; TG, 5.41±0.23). The deaths of the TG mice were sudden, suggesting that the cause of death may be related to the arrhythmic activity. Notably, female TGs exhibit significantly (P<0.05) longer average lifespans than male TGs (survival in months: males, 16±0.4; females, 21.3±0.7). There are no apparent differences in the survival rates between WT females and males (Fig. 1A). The delayed mortality of female TGs prompted inquiries into gender-dependent alterations in: (a) PLB expression levels; (b) adrenergic drive; (c) left ventricular function; and/or (d) chamber remodeling, each of which could contribute to the improved survival of female TGs.
Phospholamban overexpression in male and female transgenic mice
Calcium handling in the cardiomyocyte is largely dependent upon SR function. To assess the degree of PLB overexpression and the consequent SR functional effects, the protein levels of PLB, SERCA2, and calsequestrin (CSQ) were determined by quantitative immunoblotting. Three-month-old WT males and females expressed similar levels of PLB and SERCA2. Age-matched male and female TG hearts exhibited fourfold increases in PLB, whereas SERCA2 protein levels were unchanged, thereby increasing the relative PLB/SERCA2 ratio to 4:1 in both female (arbitrary units: WT, 1.13±0.22; TG, 4.05±0.42; n=4 hearts per group, P<0.05) and male (WT, 1.05±0.20; TG, 4.00±0.40; n=4 hearts per group, P<0.05) TG hearts.
Increased adrenergic drive in male and female transgenic mice
Another possibility for the prolonged lifespan in TG females involved a possible delay in adrenergic signaling enhancement, an important compensatory mechanism in this model . As previously reported, male TG mice exhibited high PLB phosphorylation levels (∼ninefold increase), which were linked to increased sympathetic drive . To determine whether similar PLB overexpression in female TG hearts might elicit comparable compensation, PLB phosphorylation was quantified, using site-specific phosphorylation antibodies. The phosphorylation levels of both serine-16 (∼10-fold) and threonine-17 (∼fourfold) sites were substantially higher in female TG hearts (Fig. 1B), similar to their male counterparts . This increased phosphorylation at a PKA-dependent site prompted measurements of adrenergic drive in female hearts as a logical source for such increases. Myocardial norepinephrine (NE) was depleted by 18% (pg/mg: WT females, 1480±80; TG females, 1100±60; n=6 hearts per group, P<0.05) and plasma NE elevated by 50% in TG female hearts, compared to WT females (ng/ml: WT females, 2.8±0.3; TG females, 4.2±0.6; n=5 hearts per group, P<0.05). The relative adrenergic alterations were similar to the 18% decrease and 52% increase in male TG myocardial and plasma NE, respectively , indicating that female TG mice possessed a comparable degree of increased catecholaminergic drive. Interestingly, propanolol treatment for 10 weeks exerted no effect on left ventricular fractional shortening (FS%) and the velocity of circumferential fiber shortening (Vcfc) in WTs, while these parameters were significantly decreased compared to non-treated TG mice. This depressed function was associated with reduced PLB phosphorylation and increased inhibition of SERCA .
Depressed left ventricular and SR function in male and female transgenics
The findings above indicate that the prolonged survival in female TG mice existed despite similar increases in PLB protein expression and catecholamine excess. We then investigated whether left ventricular function might be preserved in female hearts, thereby influencing their survival. M-Mode and Doppler echocardiography was carried out at specific time-points (3, 15, and 22 months). Three-month-old male and female TG mice possessed similar left ventricular function. Surprisingly, upon aging to 15 months—the beginning of the mortality period in males—female TGs exhibited nearly identical depressions in contractile parameters as their male TG counterparts. Upon direct comparison, LV fractional shortening (FS%) and the velocity of circumferential fiber shortening (Vcfc) values were not significantly different (P0.05) between female and male TGs. At 22 months, further reductions in these parameters were observed in TG females (Table 1).
M-Mode and Doppler echocardiography; EDD, end-diastolic dimension; ESD, end-systolic dimension; FS, fractional shortening percentage; Vcfc, velocity of circumferential fiber shortening, corrected for heart rate differences. Data represent mean±S.E.M.
*P<0.05 versus age-matched, same-sex WT. ‡P<0.05 versus month counterparts.
Left ventricular functional perturbations were previously found to correlate with alterations in SR calcium-handling proteins . Indeed, SERCA2 levels and PLB phosphorylation were decreased in aging female TG hearts. PLB phosphorylation at Thr-17 showed a trend to be slightly higher than age-matched male TGs, but it was not statistically different (Fig. 1C). Thus, although comparable protein and functional alterations were evident in TG females at 15 months, an early mortality did not accompany this functional decline.
Left ventricular hypertrophy and dysfunction in male and female transgenics
One salient feature of the cardiomyopathy observed in male TG hearts was the development of left ventricular hypertrophy and dilation at 15 months . In this regard, echocardiography revealed a clear departure in 15-month TG female hearts. Whereas 15-month male TG hearts developed a 32% increase in gravimetric LVM/BM ratio, 15-month females displayed no increase in either index of hypertrophy (Fig. 2A). Furthermore, left ventricles from 15-month TG females were not dilated, as evidenced by a preserved end-diastolic dimension (EDD, Fig. 2B), unlike their 15-month male counterparts. Upon aging to 22 months, female TGs developed substantial left ventricular hypertrophy (33%) and chamber dilatation, which was concurrent with their premature mortality (Fig. 2A). Hence, a transition was apparent from 15 months onward, which precipitated the development of pathology in aging female TG hearts.
Blood pressure measurements
To address whether the disparate hypertrophic patterns in this model were due to gender-dependent differences in hemodynamic loads, systolic tail-cuff blood pressures were measured in young male and female animals. No significant differences among male (mmHg: WT, 117±2; TG, 116±6) or female (mmHg: WT, 118±3; TG, 120±2) mice were observed with regard to systolic blood pressures. Although gender-dependent hemodynamic differences might develop later in life, and thereby contribute to hypertrophic growth, our data suggest that hemodynamic loads were similar at 3–4 months.
Hypertrophic signaling pathway evaluation
The absence of cardiac hypertrophy in 15-month-old female TG mice prompted inquiry into the mechanisms of the observed hypertrophic phenotype. In this transgenic model, the dominant hypertrophic stimulus appeared to be chronically-enhanced catecholaminergic input to the myocardium, which has been linked to MAPK activation. To investigate this possibility, we measured the levels of phosphorylated (active) p38 MAPK in aging cardiac homogenates (Fig. 3). Quantitative immunoblotting revealed a progressive increase in activated p38 from 3 to 15 months in male TG hearts, compared to WT males (Fig. 3A). In contrast, 3-month female TG mice showed no MAPK activation, although increasing p38 activation was evident from 15 to 22 months. Equally intriguing was the reduction in the levels of total MKP-1 protein at 15 (male TG) and 22 (female TG) months (Fig. 3B), the respective ages at which male and female TG mice had developed cardiac hypertrophy. Thus, decreases in MKP-1 protein and increases in p38 MAPK activation were temporally associated with the development of hypertrophy and premature mortality in this model of enhanced adrenergic drive. Evaluation of ERK phosphorylation, a second member of the MAPK signaling family and a potential hypertrophic mediator, revealed significant (P0.05) increases in both male and female TG hearts (with respect to same-sex WTs). However, there was no significant evidence of differential ERK activation between the genders (Fig. 3A).
Effects of norepinephrine and estrogen in isolated cardiomyocytes
To test the hypothesis that estrogen modulates hypertrophic signaling, left ventricular myocytes from wild-type female mice were isolated and treated with norepinephrine (NE) and 17β-estradiol (E2). The treated myocytes were harvested and subjected to quantitative immunoblotting to determine the degree of MAPK activation. Both NE and E2 were able to activate ERK1/2 (Fig. 4). However, estrogen (alone and in combination with NE) achieved a significant (P<0.05) prolongation of ERK1/2 activation (∼1 h) compared to NE alone (Fig. 4A). Recent studies have implicated ERK-dependent phosphorylation of MKP-1 to a rapid stabilization of the phosphatase in vitro . When MKP-1 levels were assessed in E2-treated myocyte preparations, a significant, sustained increase in MKP-1 protein levels was observed, compared to NE alone (Fig. 4B). Immunoblots were also probed for levels of activated p38 MAP kinase. NE-treated cardiomyocytes exhibited a maximal activation of p38 MAP kinase at 15 min; however, E2 preincubation was able to suppress this NE-induced p38 activation (Fig. 5). It is critical to note that the in vitro suppression of p38 activation was consistent with E2's stimulation of MKP-1 expression, which has been shown to potently deactivate p38 .
This study represents, to our knowledge, the first in vivo evidence that endogenous elevations in adrenergic drive may lead to cardiac remodeling through p38 activation, which is regulated, at least partially, by MKP-1 levels in a gender-dependent manner. Our data also suggest that hypertrophy and left ventricular dysfunction may arise distinctly, secondary to hormonal influence over cardiac remodeling, and that gender may influence the progression of left ventricular hypertrophy in the context of chronically elevated adrenergic drive.
The development of cardiac hypertrophy involves molecular signaling pathways that are responsive to both mechanical (i.e. pressure overload) and neurohormonal (i.e. catecholamines, endothelin, angiotensin II) imbalances . A recent study of aortic constriction in catecholamine-deficient mice emphasized the dependence of pressure-overload hypertrophy upon adrenergic stimulation . In our model, the absence of blood pressure differences supports at least a partial role of catecholamines in inducing hypertrophy through the MAP kinase pathway, and specifically, p38 MAP Kinase. Recent in vitro findings in isolated cardiomyocytes suggested that the stimulatory effects of β2-AR on p38 MAPK are mediated by a PKA-dependent pathway . These data imply that, in failing human hearts, the selective downregulation of β1 receptors might allow norepinephrine to continue to activate p38 through both α and β2 signaling. Similarly in this model, desensitization of adrenergic signaling  did not preclude increasing activation of p38 MAP Kinase. Future investigations into altered β1 versus β2 signaling components may provide insight into p38 activation in adrenergically-desensitized hearts.
It is important to note that recent work  appears to indicate that p38 activation alone may not be sufficient to induce hypertrophy, suggesting that other molecular mediators may be involved in this phenotype. Calcium cycling defects have also been shown to trigger the onset of hypertrophic signaling, specifically through calcineurin, a calcium-dependent phosphatase . The signaling activities of calcineurin require perturbations in baseline calcium levels/cycling, and abnormalities of calcium-handling in the failing cardiomyocyte  are presumed to be critical to calcineurin's activation. Assessment of calcineurin levels indicated ∼2.0-fold increase in both aged male and female TG mice (data not shown), consistent with the nearly identical SR protein and functional alterations in both genders.
Cardiovascular disease variably affects different cohorts of the population, particularly with respect to gender. Female patients with similar degrees of advanced, non-ischemic left ventricular dysfunction exhibited improved survival rates over males . In addition, female patients with aortic stenosis exhibited better preservation of systolic function and increased left ventricular hypertrophy versus males [22–24]. Similarly, aortic banding studies in rats revealed earlier transition to left ventricular chamber dilatation, eccentric hypertrophy, and diastolic dysfunction in males versus females [25,26]. However, it is currently unclear whether estrogen is directly responsible for these disparities in left ventricular function and structural remodeling/hypertrophy . In this study, male TG mice exhibited increased p38 MAPK activation at 3 months, reflecting chronically enhanced adrenergic stimulation. Although p38 activation in female transgenics was not detected at 3 months, it was present at 15 and 22 months, suggesting that a possible molecular ‘brake’ on p38 activation by MKP-1 was lifted, and that hypertrophic signaling had ensued. Interestingly, gender appeared to influence not only left ventricular hypertrophy, but also chamber dilatation, indicating effects on both the degree and the nature of hypertrophic growth. Indeed, it may be important for females to regulate the development of cardiac hypertrophy, since there is some evidence that female patients with a similar degree of non-ischemic hypertrophy as males exhibit higher mortality rates .
Notably, concurrent MKP-1 protein decline and p38 activation in females occurred following the reported menopausal period for mice (14–15 months) . In this regard, rapid induction of MKP-1 by estrogen in isolated myocytes was observed in this and other studies . Interestingly, a recent study  indicated that estrogen may mediate p38-kinase inactivation via MKP-1 induction in glomerular mesangial cells, suggesting a protective effect of estrogen in various target organs. However, other factors contributing to the delayed p38 activation in females may also exist, since there was no effect of gender upon baseline MKP-1 levels at 3 months of age, despite selective p38 activation in transgenic males. These may include PKA activation of a tyrosine phosphatase , or even Inhibitor-1 of Protein Phosphatase-1, which may lead to increased MAPK activation . Moreover, we cannot rule out the possibility that other hormonal influences are acting in this model. Notably, the beneficial effects of estrogen against cardiac hypertrophy may be mediated by interactions between this hormone and the renin–angiotensin system [32–34].
The progressive functional impairment was predictably mirrored by decreases in SERCA2 protein and PLB phosphorylation. These subcellular protein alterations, similar to previous observations upon hyperadrenergic stimulation , were nearly identical between male and female TGs. A non-significant trend existed for aging female TGs to possess higher Thr-17 PLB phosphorylation levels. However, transgenic mice expressing mutant PLB, which lacks a Thr-17 phosphorylation site (T17A), exhibited no alterations in basal contractile parameters or their β-adrenergic responses, indicating that phosphorylation of this residue may not serve a critical role . In light of similar SR and functional profiles between genders, it was unexpected to find the synchronous decline in left ventricular performance dissociated from the cardiac hypertrophic response between genders. Although left ventricular functional decline and hypertrophic growth may stem from the same precipitating insult (i.e. excess adrenergic drive), they may diverge into distinctly modulatable processes. Consistent with this observation, recent work by DiPaola et al.  captured a unique window, demonstrating that human hearts possessing left ventricular hypertrophy in the absence of functional impairment did not yet exhibit alterations in SR calcium-handling protein levels, whereas functionally-impaired hypertrophic hearts did. Interestingly, the early mortality in our transgenic model coincided with the development of left ventricular hypertrophy and dilatation, but not with SR protein alterations. These observations provide some insight into the important transition from compensation (i.e. hypertrophy without SR/functional alterations) to decompensation (i.e. hypertrophy with SR/functional alterations) in human heart failure .
Together, our findings implicate MKP-1-dependent regulation of p38 activation as an integral component of catecholamine-induced hypertrophy. Moreover, gender-dependent modulation of MKP-1 may contribute to the improved cardiac remodeling and survival rates observed in females with left ventricular dysfunction. A more detailed analysis of the temporal expression of p38 in early- and late-stage hypertrophy may yield important insights into the pathophysiology of this process. Furthermore, future studies may be designed to utilize direct infusion of catecholamines in non-genetically modified animals to elucidate the regulation of p38 activation by estrogen and MKP-1.
This work was supported by an American Federation for Aging Research Grant (R.D.), NIH grants: HL-26057, HL-64018, HL-52318 and P40RR12358 (E.G.K.). We would like to thank Dr J.D. Molkentin for valuable discussions.
- cardiac myocyte
- animals, transgenic
- left ventricular hypertrophy
- depressive disorders
- dilatation, pathologic
- mice, transgenic
- mitogen-activated protein kinases
- muscle contraction
- phosphoric monoester hydrolases
- depressed mood
- protein overexpression
- mitogen-activated protein kinase p38