| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Cardiology, Department of Medicine, Johns Hopkins Medical Institutions,
* Department of Physiology, University of Maryland, Baltimore, Maryland, USA;
Department of Pharmacology and Cell Biophysics, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA;
Department of Biochemistry and Molecular Biology, Indiana University, School of Medicine, Indianapolis, Indiana, USA; and
Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS-INSERM-ULP, B.P. 163, 67404 Illkirch-Strasbourg, France
1Correspondence: Halsted 500, Division of Cardiology, 600 N. Wolfe St., Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA. E-mail: dkass{at}jhmi.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-skeletal actin, ß-myosin heavy chain, natriuretic peptides) were similar between groups. However, total and serine-phosphorylated phospholamban protein declined -38 and -64% respectively, and protein phosphatase-1 (PP1) activity increased 44% without increased protein levels (all P<0.01) in CREM-/- vs. controls. These results demonstrate novel involvement of CREM in regulation of PP1 activity and of PLB, likely resulting in a potent frequency-dependent influence on cardiac function.—Isoda, T., Paolocci, N., Haghighi, K., Wang, C., Wang, Y., Georgakopoulos, D., Servillo, G., Della Fazia, M. A., Kranias, E. G., DePaoli-Roach, A. A., Sassone-Corsi, P., Kass, D. A. Novel regulation of cardiac force-frequency relation by CREM (cAMP response element modulator).
Key Words: CRE-responsive transcription factor • mouse • phosphatase • phospholamban • sarcoplasmic reticulum • cardiac function
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. Both CREB and ATF-1 are ubiquitously expressed, whereas CREM has appeared more localized to neuroendocrine tissues (3
4
5
6
7
8)
. Genes containing CRE consensus sequences in their promoter region are involved in a broad array of cellular processes, notably metabolic regulation and neurotransmission, gene transcription, cell cycle/survival/growth factors, immune regulation, and cell signaling (2)
. Their activation is principally coupled to cAMP-dependent protein kinase A phosphorylation. Both CREB and CREM can be phosphorylated by growth factor and stress kinases (1
, 9)
, whereas CREM can be activated in male germ cells by association with ACT, a member of the LIM-only class of proteins (10
, 11)
.
In the heart, CREB is robustly expressed and phosphorylated (12
, 13)
. Although mice harboring deletion of the basic region/leucine zipper domain of CREB (CREB-/-) die at birth from pulmonary atelectasis (14)
, overexpression of a dominant negative CREB results in dilated cardiomyopathy associated with growth arrest (15)
. These hearts display depressed stress fiber shortening relations and diminished dobutamine stimulation responses. CREM mRNA is expressed in the human heart; an isoform (CREM-Ib
C-X) has been described that generates internally translated repressors of CRE-mediated gene transcription (16)
. CREM-/- mutant mice have been generated and survive to adulthood, although males are sterile due to apoptosis of postmeiotic germ cells (17
, 18)
. These animals display delayed liver regeneration after partial hepatectomy (19)
, changes in circadian rhythmicity, and reduced anxiety in behavioral tests (5
, 20)
. Recent data obtained in another murine model of CREM inactivation has suggested impaired basal systolic and diastolic function accompanied by diminished expression of the sarcoplasmic reticular ATPase (21)
—supporting a physiologic role of CREM in the heart. The present study examined the impact of CREM on rest and cardiac reserve function in intact murine hearts employing comprehensive pressure-volume analysis. We demonstrate novel involvement of CREM in the modulation of frequency-dependent cardiac function associated with activation of protein phosphatase with lower phospholamban protein expression and serine-16 phosphorylation. These results provide novel evidence for a functional role of CREM to frequency-dependent function in the intact heart.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. CREM-/- animals (4–6 months) and age-matched wild-type controls (same background) were housed in a 12:12-hr light/dark cycle, with food and water provided ad libitum.
Physiological protocol
In vivo cardiovascular physiology studies were performed using a miniaturized pressure-volume catheter system developed in our laboratory (22)
. Studies were performed in accordance with guidelines of the Animal Care and Use Committee of The Johns Hopkins University. Mice were anesthetized with 1–2% isoflurane, followed by urethane (750–1000 mg/kg, i.p.), etomidate (5–10 mg/kg, i.p.), and morphine (1–2 mg/kg, i.p.). Animals underwent tracheostomy and ventilated with 6–7 µL/g tidal volume at 140 breaths/min. Fluid supplementation (12.5% human albumin) was provided (50–100 µL over 10 min) through a 30G needle cannulating the right external jugular vein. The LV apex was exposed through an incision between the 7th and 8th rib; a pressure-volume catheter (SPR-719, Millar Inst., Houston, TX) was advanced through the apex and out into the aortic root, with the distal electrode placed just within the LV cavity. The volume signal was calibrated by the hypertonic saline method (23)
, with stroke volume matched to that determined from simultaneous ascending aortic flow (AT01RB; Transonic Systems Inc., Ithaca, NY).
Steady-state data and pressure-volume relations recorded during transient reduction of venous return were measured. Left ventricular pressure-volume data were recorded at incremental heart rates (400–800 min-1). The If blocker ULFS 49 (BI, 200 µg i.v.) was first infused to lower heart rate < 400 min-1 without influencing LV function; rate was controlled by atrial pacing using an intraesophageal catheter (NuMed, Hopkinton, NY). The decay of systolic potentiation induced by rapid pacing after abrupt restoration to normal sinus rhythm (recirculation fraction, RF) was measured to assess the relative amount of calcium internally recycled through the sarcoplasmic reticulum (24
, 25)
. Last, animals received i.v. isoproterenol (1–100 ng·kg-1·min-1) at a constant atrially paced rate of 600 min-1. After study completion, animals were killed by barbiturate overdose and hearts were excised and prepared for subsequent histological and molecular assays.
Echocardiography
Echocardiograms were performed in six CREM-/- and five WT animals. After induction with 1–2% isoflurane 98% O2, animals were placed on a heated pad and allowed to recover to a near-awake state. Two-dimensionally targeted M-mode LV images (Sonos 5500, Philips, MA) were obtained at the level of the papillary muscles using a 15MHz transducer. LV diastolic and systolic dimensions, percent shortening fraction, and end-diastolic posterior wall thickness were measured.
Histology
Myocardium was embedded in paraffin and (5 µm) sections were stained using hematoxylin/eosin and Masson’s trichrome to assess histology and fibrosis. Two individuals blinded as to tissue source performed the histological analysis.
RNA dot-blot analysis
RNA samples were prepared from snap-frozen hearts using TRIZOL reagent (Life Technology, Gaithersburg, MD) according to the manufacturer’s protocol. RNA dot-blot analysis was performed using a published protocol (26)
with a set of oligonucleotide probes (generously provided by Dr. Gerald Dorn II, University Of Cincinnati). Briefly, 2 µg of total RNA isolated from the hearts of the animals was blotted on the cellulose membrane (S&S) and hybridized with P32 labeled oligonucleotide at various lengths that complement the coding sequence of the target mRNAs. Hybridization signals were quantified using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and presented as relative pixel value normalized against GAPDH.
Quantitative immunoblotting
Hearts were homogenized at 4°C in a buffer containing 10 mM imidazole (pH 7.0), 0.3 M sucrose, 1 mM dithiothreitol, 1 mM sodium metabisulfite, 1 mM sodium metabisulfite, 0.3 mM phenylmethyl-sulfonyl fluoride, 2 mM EDTA, 5 µg/mL leupeptin, 10 µg/mL soybean trypsin inhibitor type II-S (Sigma, St. Louis, MO), and 7 µg/mL pepstatin A. After solubilization of homogenates, SDS-PAGE (13%) and immunoblotting were performed as described (27)
. Membranes were incubated with phospholamban (1:10,000 dilution, ABR), phosphoserine-16 (1:5000 dilution, Badrilla, UK), SERCA (1:500 dilution, homemade), and calsequestrin (1:5000 dilution, ABR) antibodies. Primary antibody binding was visualized by peroxidase-conjugated secondary antibodies and enhanced chemiluminescence (Amersham Pharmacia Biotech, Arlington Heights, IL). Protein concentration was determined by the Bradford method (Bio-Rad, Hercules, CA) with bovine serum albumin as a standard.
Protein phosphatase expression and activity
Protein phosphatase activity was assayed using 32P-labeled rabbit glycogen phosphorylase a as the substrate (28)
. Enzyme activity was determined in the absence or presence of 4 nM okadaic acid, a concentration that, under the assay conditions, inhibits PP2A but has no effect on PP1. The incubation was conducted for 3 min on 1:10 w/vol tissue extracts or for 10 min on 1:40 diluted extracts. Under both conditions, no more that 15% of the substrate was used to assure linearity of the reaction. Phosphatase activity is expressed as nmol phosphate released per minute per milligram protein. Western analysis of PP1c was performed with a monoclonal antibody that recognizes all isoforms of PP1c (Santa Cruz, Santa Cruz, CA) and signals detected as previously reported (28)
.
Statistics
Data are presented as mean ± SE. Between-group baseline comparisons were analyzed by unpaired t test. Force-frequency and isoproterenol dose-response relations were analyzed by repeated measures ANOVA (multivariate linear regression model) with a grouping factor.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and Fig. 1
. Pressure-volume relations (Fig. 1A
) display normal chamber volumes and systolic and diastolic function in CREM-/- animals at rest. The only hemodynamic difference was a slightly lower LV end-systolic pressure in CREM-/- that could reflect differences in systemic resistance. Echocardiography confirmed normal chamber size wall thickness, and fractional shortening (Fig. 1B
); histological examination showed normal myocardial architecture without interstitial fibrosis or fiber disarray (Fig. 1C
).
|
|
The presence or absence of an altered cardiac failure genotype was further probed by analyzing fetal genes commonly recapitulated in hypertrophic and dilated cardiac failure. As shown in Fig. 1D
, dot-blot assays of -skeletal actin, and ß-myosin heavy chain, and atrial or brain natriuretic peptide revealed no differences between WT and CREM-/- hearts.
Force-frequency relationship
In contrast to basal function, the enhancement of systolic function and diastolic relaxation observed with incremental heart rate (FFR) in control (WT) hearts was diminished in CREM-/- mice. Systolic function indexed by load-independent parameters (e.g., peak rate of change of pressure at a matched end-diastolic volume) rose 62 ± 11% in WT controls vs. only 22 ± 9% in CREM-/- (P<0.05 for group interaction). Similar results were obtained for preload-adjusted maximal power, another load-independent systolic index (29)
(Fig. 2
A). Even greater frequency-dependent disparities were observed in diastolic relaxation rates. The exponential time constant of pressure decline and pressure half-time both declined by 
20% in WT animals at faster heart rates, whereas neither was significantly altered by frequency change in CREM-/- mice (P<0.01, Fig. 2B
).
|
To further probe abnormalities of force-frequency modulation, the rate of systolic potentiation decay after abrupt cessation of rapid pacing was determined. The geometric time constant reflecting this decay is the recirculation fraction, an index of the relative proportion of calcium internally recycled via the sarcoplasmic reticulum. The RF in CREM-/- was 0.97 ± .016, nearly identical to that in WT (0.96±.02) (Fig. 3
). Both are compatible with values recently reported in a variety of normal mouse strains (30)
.
|
Response to ß-adrenergic stimulation
An alternative contributor to a blunted FFR is down-regulation of ß-adrenergic signaling (30)
. This was potentially relevant in that ß-receptors and downstream signaling components have CRE consensus sequences in their promoters. We therefore examined inotropic and lusitropic responses to ß-adrenergic stimulation at a fixed heart rate. However, unlike the FFR response, the dose response of systolic (Fig. 4
A) and diastolic function (Fig. 4B
) to isoproterenol was virtually identical between CREM-/- and WT animals.
|
SERCA and PLB expression and phosphorylation
Reduced FFR behavior is frequently linked to abnormal calcium uptake into the SR as mediated by SERCA2 and PLB. Although recirculation fraction was similar in WT and CREM-/- hearts, suggesting similar total recycled calcium, abnormal frequency-dependent enhancement of SR uptake could depend on either protein. SERCA2 and PLB gene expression (Fig. 5
A) were similar in WT and CREM-/-, but there were important post-translational changes, with a 38% decline in total PLB protein expression and 64% decline in ser(16)-phosphorylated PLB (both P<0.01; Fig. 5B, C
). These changes were similar (if not slightly greater) when total and ser(16)-PLB were normalized to calsequestrin protein levels measured in each heart (-45, -70%, respectively, P<0.01; Fig. 5C
, bottom). In contrast, SERCA2 protein and calsequestrin were identical between groups.
|
Protein phosphatase activity and expression
PLB phosphorylation is regulated by protein phosphatases 1 and 2. Since ser(16)-PLB was reduced in CREM-/-, we examined whether PP1 or PP2A were differentially activated in this model. The results (Fig. 6
A) revealed a + 44% increase in PP activity (P<0.01). When the assay was performed in the presence of 4 nM okadaic acid, the total PP activity was decreased, consistent with inhibition of PP2A, but the increase in activity remained, indicating that activation was due principally to PP1. Data were after 3 min incubation; similar results were obtained after a 10 min assay (not shown). PP1 protein expression was determined by Western blot, but there was no difference between WT and CREM-/-. Thus, enhanced PP1 activity was due to enzyme activation rather than increased levels.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Earlier studies of CREM have demonstrated its pivotal role in the hypothalamic-pituitary-gonadal axis (1
, 3
, 5)
. CREM is highly expressed in male postmeiotic cells, where it undergoes a switch from repressor to activator by association with the testes-specific ACT (activator of CREM) protein (1
, 11)
. Male CREM-/- animals are sterile with complete block at the first step of spermiogenesis (17)
. CREM is important to circadian rhythms within the pineal gland (5)
, coupling transcriptional modulation to ß-sympathetic-cAMP diurnal oscillation (20
, 31)
.
CREM mRNA is expressed abundantly in the human heart. Müller et al. (16)
first reported the presence of CREM-IbC-X mRNA in failing right ventricular tissue taken from patients undergoing cardiac transplantation for end-stage dilated cardiomyopathy. The 9 and 11 kDa internal translation products (putative HIbI and HIbII) demonstrated inhibitory signaling properties, blunting cAMP-stimulated luciferase activity (CRE-mediated transcription) by 55% in cells overexpressing CREM-IbC-X. The authors speculated that ß-adrenergic hyperstimulation in the failure state might activate this inhibitory protein and, much like a dominant negative CREB model, contribute to the development of cardiac dysfunction. However, whether expression of CREM-IbC-X is indeed restricted to failing vs. normal hearts remains unclear, as data for normal myocardium have not been reported. Whereas in vitro overexpression of CREM-IbC-X had potent inhibitory effects on CRE-mediated transcription (16)
, such data did not identify basal CREM regulation. More recently, the same investigators studied basal and adrenergic stimulated cardiac function in a murine model of CREM inactivation (21)
, and reported impaired function accompanied by depressed SERCA2 expression.
As with the recent study of Muller et al (21)
, we also observed a slight reduction in systolic pressure in CREM-/- animals; however, this was not associated with basal depression of systolic function when examined by comprehensive and load-insensitive pressure-volume relations. Adrenergic responses were similar between CREM-/- and controls (measured in the present study at constant heart rate), whereas the frequency response was blunted. Furthermore, SERCA2 expression was unchanged at both mRNA and protein levels in the current study, whereas total PLB (and ser(16)-PLB) declined, in contrast to the prior study. The reasons for these disparities remain unclear, however differences between the murine models themselves as well as in the experimental methods and procedures (e.g. anesthesia, basal function, heart rates) used to assess cardiac hemodynamics may have played a role.
Absence of CREM did not appear to trigger excessive CRE-mediated signaling as such a change would likely trigger hypertrophic responses, including myosin isoform shifts and altered adrenergic signaling; yet neither was observed. Inhibition of PP1 by either up-regulation of the intrinsic I-1 inhibitor (33)
or pharmacologic agents, enhances CREB phosphorylation (34)
. The increased PP1 in CREM-/- suggests a novel mechanism whereby CREM may interact with CREB to modulate transcriptional activity; in this regard, the present data may extend beyond the heart.
Activation of PP1 has been reported in experimental and human heart failure and is thought to contribute to contractile dysfunction by enhancing dephosphorylation of calcium handling and contractile proteins (35
36
37)
. Inhibition of PP with cantharidin shortens relaxation time in association with enhanced ser(16)-PLB and enhances inotropy associated with increased thr(17)-PLB and troponin-I phosphorylation (38)
. Mice expressing a threefold increase in PP1 display dilated cardiomyopathy and premature death (39)
associated with marked elevation of ANP, -skeletal actin, and ß-MHC gene expression. More modest increases in PP1 activity (23%) achieved by ablation of an intrinsic PP1-specific inhibitor led to modest cardiac depression at baseline and impaired ß-adrenergic responsiveness, but no frank cardiac failure (39)
. In the present study, PP1 activity was modestly elevated in CREM-/- mice, yet basal and ß-adrenergic stimulated function appeared normal. This disparity may relate to differentially affected downstream PP1 targets or concomitant changes in other enzymes associated with the lack of CREM, which was not induced in the earlier model.
The marked reduction in ser(16)-PLB in CREM-/- is compatible with elevated PP1 activity and, relative to the decline in PLB, indicates net PLB hypophosphorylation that could have contributed to the FFR deficit observed. Depressed force-frequency relations and diminished PLB phosphorylation are observed with cardiac failure and are thought to be mechanistically linked (35
, 40
41
42
43)
. Total PLB was reduced in CREM-/- despite similar levels of SERCA2, lowering the PLB/SERCA2 ratio. Although this ratio itself can alter the FFR (44
, 45)
, levels of protein-phosphorylation are very important to the net effect. In the present model, a greater decline in ser(16)-PLB over total PLB with preserved SERCA2 reduced FFR behavior, most strikingly with respect to relaxation. Additional targets of PP1 (e.g., TnI) might have played a role in this response, but clarifying this pathway will require further study.
PLB has two phosphorylation sites—Ser(16) and Thr(17)—the former coupled to protein kinase activation, and the latter to Ca2+/calmodulin-dependent protein kinase II. Frequency encoded stimulation has been proposed to principally stimulate Thr(17)-PLB whereas ß-adrenergic stimulation principally couples with Ser(16)-PLB (46)
, the latter being sufficient to achieve a maximal ß-agonist response (47)
. However, Thr(16)-PLB is synergistically stimulated when frequency-dependent signaling is combined with adrenergic stimulation (46)
as present in vivo. Although we could not selectively identify Thr(17)-PLB in the mouse hearts, diminution of Ser-16-PLB could certainly modulate ambient adrenergic-coupled calcium uptake and thus force-frequency behavior.
The importance of the FFR to normal cardiac reserve has been firmly established, with studies showing a potent involvement of SR calcium cycling proteins in mediating its depression in heart failure (42
, 43)
. Such insights have led to gene transfer and murine genetic cross experiments confirming the potency of restoring SERCA2 (42)
or elevating phosphorylated-PLB (48)
. The present data suggest that CREM can further contribute to this process, but that it does so in a manner that preserves ß-adrenergic response and without inducing failure phenotype or genotype. The involvement of PP1 stimulation suggests CRE-responsive signaling of PP1 inhibitors, which may have implications to other organs systems. Whether increased CREM expression has the opposite effects remains to be determined. Future studies using microarrays and proteinomics in normal and diseased cardiac states should help clarify the nature of CRE/CREM/CREB expression changes, their potential role in phosphorylation cascades, and their importance to the heart.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 6, 2002. Accepted for publication October 24, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. S. Huggins, J. J. Lepore, S. Greytak, R. Patten, R. McNamee, M. Aronovitz, P. J. Wang, and G. L. Reed The CREB leucine zipper regulates CREB phosphorylation, cardiomyopathy, and lethality in a transgenic model of heart failure Am J Physiol Heart Circ Physiol, September 1, 2007; 293(3): H1877 - H1882. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Matus, G. Lewin, F. Stumpel, I. B. Buchwalow, M. D. Schneider, G. Schutz, W. Schmitz, and F. U. Muller Cardiomyocyte-specific inactivation of transcription factor CREB in mice FASEB J, June 1, 2007; 21(8): 1884 - 1892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ukropec, R. P. Anunciado, Y. Ravussin, M. W. Hulver, and L. P. Kozak UCP1-independent Thermogenesis in White Adipose Tissue of Cold-acclimated Ucp1-/- Mice J. Biol. Chem., October 20, 2006; 281(42): 31894 - 31908. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Tyson-Capper, J. Bailey, A. R. Krainer, S. C. Robson, and G. N. Europe-Finner The Switch in Alternative Splicing of Cyclic AMP-response Element Modulator Protein CREM{tau}2{alpha} (Activator) to CREM{alpha} (Repressor) in Human Myometrial Cells Is Mediated by SRp40 J. Biol. Chem., October 14, 2005; 280(41): 34521 - 34529. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. U. Muller, G. Lewin, H. A. Baba, P. Boknik, L. Fabritz, U. Kirchhefer, P. Kirchhof, K. Loser, M. Matus, J. Neumann, et al. Heart-directed Expression of a Human Cardiac Isoform of cAMP-Response Element Modulator in Transgenic Mice J. Biol. Chem., February 25, 2005; 280(8): 6906 - 6914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A Movsesian Altered cAMP-mediated signalling and its role in the pathogenesis of dilated cardiomyopathy Cardiovasc Res, June 1, 2004; 62(3): 450 - 459. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Palmer, E. A. Mokelke, A. M. Thayer, and R. L. Moore Mild renal hypertension alters run training effects on the frequency response of rat cardiomyocyte mechanics J Appl Physiol, November 1, 2003; 95(5): 1799 - 1807. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
