This Article Abstract Full Text (PDF) All Versions of this Article: 102/6/695    most recent CIRCRESAHA.107.169755v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bossuyt, J. Articles by Bers, D. M. Search for Related Content PubMed PubMed Citation Articles by Bossuyt, J. Articles by Bers, D. M. Related Collections Animal models of human disease Heart failure - basic studies Physiological and pathological control of gene expression

(Circulation Research. 2008;102:695.)
© 2008 American Heart Association, Inc.

Cellular Biology

Ca²⁺/Calmodulin-Dependent Protein Kinase II and Protein Kinase D Overexpression Reinforce the Histone Deacetylase 5 Redistribution in Heart Failure

Julie Bossuyt, Kathryn Helmstadter, Xu Wu, Hugh Clements-Jewery, Robert S. Haworth, Metin Avkiran, Jody L. Martin, Steven M. Pogwizd, Donald M. Bers

From the Departments of Physiology (J.B., K.H., X.W., H.C.-J., D.M.B.) and Medicine (S.M.P.) and Cardiovascular Institute (J.L.M.), Loyola University Chicago, Maywood, Ill; and Cardiovascular Division (R.S.H., M.A.), King’s College London, United Kingdom.

Correspondence to Donald M. Bers, Department of Physiology, Loyola University Chicago, 2160 S First Ave, Maywood, IL 60153. E-mail dbers{at}lumc.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cardiac hypertrophy and heart failure (HF) are associated with reactivation of fetal cardiac genes, and class II histone deacetylases (HDACs) (eg, HDAC5) have been strongly implicated in this process. We have shown previously that inositol trisphosphate, Ca²⁺/calmodulin-dependent protein kinase II (CaMKII), and protein kinase (PK)D are involved in HDAC5 phosphorylation and nuclear export in normal adult ventricular myocytes and also that CaMKII and inositol trisphosphate receptors are upregulated in HF. Here we tested whether, in our rabbit HF model, nucleocytoplasmic shuttling of HDAC5 was altered either at baseline or in response to endothelin-1, which would indicate HDAC5 phosphorylation and transcription effects. The fusion protein HDAC5–green fluorescent protein (HDAC5-GFP) was more cytosolic in HF myocytes (F_nuc/F_cyto 3.3±0.3 vs 7.2±0.4 in control), and HDAC5 was more phosphorylated. Despite this baseline cytosolic HDAC5 shift, endothelin-1 produced more rapid HDAC5-GFP nuclear export in HF versus control myocytes. We also find that PKD and CaMKII_C expression and activation state are increased in both rabbit and human HF. Inhibition of either CaMKII or PKD in HF myocytes partially restored the HDAC5-GFP F_nuc/F_cyto toward control, and simultaneous inhibition restored F_nuc/F_cyto to that in control myocytes. Moreover, adenovirus-mediated overexpression of PKD, CaMKII_B, or CaMKII_C reduced baseline HDAC5 F_nuc/F_cyto in control myocytes (3.4±0.5, 3.8±0.5, and 5.2±0.5, respectively), approaching that seen in HF. We conclude that chronic upregulation and activation of inositol trisphosphate receptors, CaMKII, and PKD in HF shifts HDAC5 out of the nucleus, derepressing transcription of hypertrophic genes. This may directly contribute to the development and/or maintenance of HF.

Key Words: heart failure • histone deacetylase • protein kinase D • Ca²⁺/calmodulin-dependent protein kinase

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cardiac hypertrophy occurs in response to physiological and pathological stimuli. Under prolonged stress, cardiac remodeling frequently becomes maladaptive and leads to heart failure (HF), cardiac arrhythmias, and sudden death.^1–2 This remodeling process has been associated with reactivation of fetal cardiac genes that encode proteins involved in contractility, Ca²⁺ handling, and energetics.³ Recent studies have indicated a key role for class II histone deacetylases (HDACs).^3–5 These HDACs (HDAC4, -5, -7, and -9) are endogenous repressors of transcription (eg, transcription driven by myocyte enhancer factor [MEF]2) and favor condensed DNA. When these HDACs are phosphorylated in response to stress stimuli, they are exported from the nucleus (bound to the chaperone protein 14-3-3), relieving the HDAC repressive function and allowing activation of a hypertrophic program of gene expression.⁴ Indeed, mice lacking these HDACs display pronounced cardiac hypertrophy.⁶ Ca²⁺/calmodulin-dependent protein kinase (CaMK) and protein kinase (PK)D have been identified as type II HDAC kinases.^5,7–8

PKD, previously referred to as the µ isoform of PKC, can act either in parallel with or downstream of PKC.⁹ Based on its structural and enzymatic properties, PKD has been reclassified as a member of the CaMK superfamily.¹⁰ Its proposed functions in various cell types have been as diverse as control of cell growth and survival and Golgi organization and function.^5,9 Although neurohumoral activation of PKD in cardiac myocytes was described some time ago,¹¹ only limited information is available on the role of PKD in the heart. PKD has been shown to phosphorylate cardiac troponin I, reducing myofilament Ca²⁺ sensitivity.^12,13 There is also correlation between hypertrophic agonists that trigger HDAC5 export and those that stimulate PKD activity.^5,14 Moreover, small interfering (si)RNA knockdown of PKD blunted the hypertrophic response in neonatal cardiomyocytes, whereas cardiac-specific expression of constitutively active PKD in vivo resulted in hypertrophy, followed by HF.¹⁴ This suggests a role for PKD in cardiac pathogenesis. Therefore, our first aim was to determine the expression and activity levels of PKD in our rabbit HF model (and in human HF). We found that it was increased.

Ca²⁺/calmodulin (CaM) and CaMKII have all been implicated in excitation–transcription coupling in cardiac hypertrophy and failure.^15–19 Moreover, CaMKII expression is increased in human HF and animal models of HF,^20–21 and overexpression of either the nuclear CaMKII_B or cytosolic CaMKII_C causes hypertrophy and HF.^18,19 Also, CaMKII inhibition protects against structural heart disease.²² Recently, we showed that in adult rabbit ventricular myocytes, endothelin (ET)-1 elicits a local Ca²⁺ release at the nuclear envelope via the inositol 1,4,5-triphosphate receptor (InsP₃R).²³ This Ca²⁺ signal activates nuclear CaMKII that phosphorylates (in approximately equal measure with PKD) HDAC5 and triggers nuclear export.²³ This novel pathway represents a mechanism by which myocytes can distinguish simultaneous local and global Ca²⁺ signals involved in contraction versus gene expression. Both CaMKII and InsP₃Rs are reportedly upregulated in human HF,^20,24 and this was also true in our arrhythmogenic, nonischemic HF rabbit model.²¹ Therefore, we sought here to determine whether the CaMKII (and PKD) signaling to HDAC5 nuclear export was elevated in HF (in rabbit and human). We demonstrate a tonic activation of this pathway in HF (ie, lower basal HDAC_nuc/HDAC_cyto), which is reversed by kinase inhibition and mimicked in control myocytes by overexpression of PKD or CaMKII. There was also an increased sensitivity of the ET-1–mediated nuclear export of HDAC5 in HF that could be blocked by PKD or CaMK inhibition. These results indicate that the CaMKII and PKD activity in HF may cause or reinforce the HF phenotype. Therefore, suppression of class II HDAC nuclear export by inhibition of PKD or CaMKII may prove beneficial in HF treatment by repressing the genes involved in cardiac remodeling.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
HF Model
HF was induced in New Zealand White rabbits by combined aortic insufficiency and stenosis as previously described.^25,26 Rabbits were studied 11.7±1.9 months later, when end-systolic dimension exceeded 12 mm (measured by 2D echocardiography).²⁷ Myocytes were isolated via Langendorff perfusion as described,²⁶ with backflow across the incompetent aortic valve in HF rabbits blocked by an inflated balloon-tipped catheter. Myocardium pieces were also flash-frozen and stored at –80°C. Protocols were approved by the University of Illinois Chicago and Loyola University Chicago Institutional Animal Care and Use Committees.

Human Left Ventricular Tissue
Left ventricular (LV) tissue was obtained from 22 failing human hearts (20 men, 2 women; aged 50.8±2.8 years; ejection fraction, 17.6±2.2%) at the time of cardiac transplantation at Loyola University Chicago, University of Illinois at Chicago, or University of Pennsylvania Hospitals. Patients had end-stage idiopathic dilated cardiomyopathy (DCM) (n=13) or ischemic cardiomyopathy (ICM) (n=9), 17 were on inotropic drug regimen, 5 patients had LV assist devices, and 7 had implantable defibrillators for documented ventricular tachycardia. Seven nonfailing human hearts (which could not be used for transplantation; 6 men and 1 woman aged 50.5±6.7 years) were obtained from the Regional Organ Bank of Illinois and Gift-of-Life organ procurement organization of Eastern Pennsylvania. Use of human tissue for research was approved by the institutional review boards of each of the participating institutions.

Myocyte Culture and Imaging
Myocytes were cultured as described after infection with HDAC5-GFP adenovirus (10 multiplicities of infection).²³ For PKD and CaMKII overexpression experiments, myocytes were also infected with adenovirus (100 multiplicities of infection) encoding PKD1, CaMKII_C, or CaMKII_B (or dominant-negative counterparts, ie, PKDK618N, CaMKII_CK43A, CaMKII_BK43A) for 24 hours total.^28,29 CaMKII viruses were kind gifts from Dr J. Heller-Brown (University of California, San Diego). Mouse PKD1 cDNA was subcloned into pshuttleCMV vector, and subsequently the adenovirus was generated according to the instructions of the manufacturer (Stratagene). Overexpression of PKD, CaMKII_C, or CaMKII_B was assessed by Western blotting. After 24 hours in culture, myocytes were exposed (or not) for 1 hour to 100 nmol/L ET-1 with or without CaMKII inhibitor (1 µmol/L KN-93) or PKD/PKC inhibitor (10 µmol/L Gö6967) pretreatment for 6 hours before ET-1 exposure. GFP-HDAC5 signals were measured by confocal microscopy. ImageJ software was used for analysis, with the intensity of the regions of interest normalized to area.

Myocyte superfusate contained (in mmol/L): 140 NaCl, 4 KCl, 1 MgCl₂, 2 CaCl₂, 10 Hepes, and 10 glucose at pH 7.4 and 23°C.

Cell Lysate and Homogenate Preparation
Isolated myocytes were rinsed and lysed in ice-cold lysis buffer containing (in mmol/L): 150 NaCl, 10 Tris (pH 7.4), 2 EGTA, 50 NaF, 0.2 NaVO₃, 1% Triton X-100, and protease inhibitor cocktail III (Calbiochem). Cell lysates were flash-frozen and stored at –80°C. Ventricular homogenates were prepared by grinding frozen tissue in a mortar and pestle and polytron and Dounce homogenization in buffer containing (in mmol/L): 150 NaCl, 50 Tris (pH 7.5), 1% Triton X-100, 10 NaF, 5 NaVO₃, 1 Na-pyrophosphate, and protease inhibitor cocktail (Calbiochem). Samples were centrifuged at 2500g for 10 minutes and again at 10 000g for 10 minutes to remove debris, with supernatants flash-frozen and stored at –80°C.

Immunoblots
Homogenates and cell lysates (30 µg protein/lane) were size-fractionated on 8% or 10% SDS-PAGE. Proteins were transferred to a 0.20-micron nitrocellulose membrane (NaF and NaVO₃ were added to transfer buffer). Immunoblots were blocked with 5% milk in Tris-buffered saline (TBS) Tween. The blots were then incubated overnight at 4°C with primary antibody: CaMKII-pT286 (1:1000 dilution; Affinity Bioreagents), a custom-made CaMKII,³⁰ HDAC5 or HDAC5-pS498 (1:1000, Signal Antibody Technology), PKD, PKD-pS916 or PKD-pSS744/748 (1:1000; Cell Signaling). After incubation with the horseradish peroxidase–labeled secondary antibody, blots were developed using enhanced chemiluminescence (Pierce Supersignal). All signals were recorded using a UVP EpichemII darkroom imaging system for quantification and captured on film for representation. Equal protein loading was ensured by reprobing for GAPDH (1:1000; Research diagnostics), and HF signals were normalized to control sample signals on the same gels. All experiments were performed in duplicate.

PKD Activity Assay
PKD activity was assayed essentially as described.¹¹ Briefly, immunoprecipitated PKD (using a bead-conjugated PKD antibody [Santa Cruz Biotechnology] for 2 hours at 4°C) was washed and resuspended in 40 µL of assay buffer (30 mmol/L Tris–HCl [pH 7.5], 10 mmol/L MgCl₂, and 1 mmol/L dithiothreitol). To initiate reactions, 10 µL of phosphorylation mix was added (containing syntide-2 [2.5 mg/mL] and 100 µmol/L [-³²P]-ATP; 400 to 600 cpm/pmol). The mixture was incubated at 30°C for 5 minutes. The reaction was terminated by addition of 100 µL of ice-cold H₃PO₄, and 100 µL of the reaction mix was spotted onto P-81 phosphocellulose paper. Free [-³²P]-ATP was removed by washing P-81 paper 4 times (5 minutes) in 75 mmol/L H₃PO₄. Radioactivity incorporation was determined by Cerenkov counting. All measurements were performed in duplicate.

MEF2–Luciferase Assay
Cultured myocytes were lysed, and luciferase activity was determined with the ONE-Glo luciferase assay system (Promega).

Immunocytochemistry
Myocytes were fixed with 4% paraformaldehyde in PBS. After incubation in PBS with 0.1% glycine, cells were permeabilized with 1% Triton X-100 and incubated with anti-HDAC5 (Cell Signaling; 1:100 dilution overnight at 4°C). This was visualized with anti-rabbit (Alexa488; Molecular Probes), and the nuclei were costained with Sytox orange.

Data Analysis
All results represent at least 4 separate hearts and are expressed as means±SE where quantified (n is the number of animal or human hearts in the biochemical experiments and the number of myocytes in functional experiments). Unpaired Student’s t test or 1-way ANOVA were used for comparisons (P<0.05 was considered significant).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Depressed Ventricular Function in HF
HF rabbits exhibited significant depression of LV systolic function, based on increases in LV end-diastolic and end-systolic dimension (by 54% and 81%, respectively; P<0.001) and decreases in LV fraction shortening (by 41%, from 37.4±1.3% to 26.4±1.1%; P<0.001) compared with baseline recordings in the same animals. Studies in rabbit myocytes were complemented by experiments in human LV tissue from 14 patients with end-stage HF. Human HF hearts exhibited severely depressed LV function, with a mean LV ejection fraction of 16.9±2.3% for ICM and 18.2±3.4% for DCM hearts, respectively. No apparent differences were seen among LV tissues from the different HF hearts based on the use of mechanical devices or drug treatment; therefore, data from these failing hearts were pooled per etiology.

CaMKII Isoform Expression in HF
Figure 1 shows CaMKII isoform expression in rabbit and human hearts (HF), measured using a custom-made -specific antibody.³⁰ There are 2 splice variants of CaMKII: the cytosolic isoform _C and the nuclear isoform _B (attributable to a nuclear localization sequence), which can be distinguished by different mobility on gels.^18,19 CaMKII activation state was assessed with an autophosphorylation-specific antibody (pT286). Representative blots and densitometric analyses reveal a global CaMKII increase in rabbit and human HF. More specifically CaMKII_B and -_C expression was, respectively, 60% and 112% (P<0.05) higher in rabbit HF myocytes (Figure 1A). Moreover autophosporylated CaMKII was increased in parallel (by 50% and 260%, respectively). A similar trend was observed in human HF (Figure 1B). Note that in ICM and DCM versus nonfailing, CaMKII_B expression was increased by 38% and 15%, respectively (only significant for ICM). CaMKII_C was increased by 57% and 39% in human HF (both statistically significant increases but not different between ICM and DCM). Therefore, in ICM versus DCM hearts, the CaMKII_B overexpression was more pronounced. Autophosphorylation signals were also significantly increased for both CaMKII isoforms but was more pronounced for the nuclear isoform. It is unclear why CaMKII_B was relatively more autophosphorylated than CaMKII_C, but this may reflect stronger chronic stimulation of local nuclear signaling pathways that activate CaMKII in the failing heart.

View larger version (37K): [in this window] [in a new window]   Figure 1. CaMKII isoform expression in HF. CaMKII-specific antibody (and CaMK-pT286–specific antibody) was used to measure CaMKIIC and -B expression (and phosphorylation) in rabbit myocyte lysates (A) and human LV homogenates (B). Blots were quantified and normalized to control hearts (*P<0.05; n=6 for rabbits, n=7 for human nonfailing [NF] and n=9 and 13 for ICM and DCM HF, respectively).

PKD Expression and Activity in HF
As for CaMKII, the activity of PKD can be inferred using phospho-specific antibodies because PKD activation involves phosphorylation of its activation loop (SS744/748) within the catalytic domain by novel PKCs, followed by autophosphorylation near its carboxyl terminus (S916).⁵ Figure 2 shows that in isolated rabbit myocytes from failing hearts, PKD expression was upregulated by 90% (P<0.05; Figure 2A). The signal with phospho-specific antibodies was also increased, but these antibodies (raised against mouse PKD phosphopeptides) gave a weak signal in rabbit versus human; therefore, the rabbit phospho-PKD was not quantified. Human LV homogenates (Figure 2B) show a similar increase in PKD expression, which was more pronounced in DCM than ICM hearts (106% increase versus 35%). The data obtained in isolated rabbit myocytes support the idea that this constitutes an actual myocyte upregulation of the ubiquitously expressed PKD. Phosphorylation of PKD was increased at both SS744/748 (29% and 150% for ICM and DCM, respectively) and S916 (67% and 263%). This is consistent with a higher basal PKD activity in human HF (especially DCM). Complementary data measuring P³² incorporation by immunoprecipitated PKD into the PKD substrate syntide-2 (Figure 2C) also indicate an elevated basal PKD activity in human HF (again higher for DCM hearts; 54% versus 16% [P<0.05]). Thus, PKD expression and activation state are enhanced in both rabbit and human HF.

View larger version (44K): [in this window] [in a new window]   Figure 2. PKD expression and activity in HF. PKD expression and phosphorylation were examined in rabbit myocyte lysates (A) and human LV homogenates (B). Shown are representative Western blots (using global and phospho-specific antibodies) and quantified protein signals (normalized to control hearts; *P<0.05; n=6 for rabbits, n=7, 9, and 13 for NF, ICM, and DCM, respectively). C, Basal PKD activity in control and HF following immunoprecipitation using phosphorylation of syntide-2 as indicator of activity (*P<0.05; n=7, 9, and 13 for NF, ICM, and DCM, respectively).

HDAC5 Expression and Subcellular Location in HF
Given the upregulation of these key kinases in HF, we next tested whether there is a tonic shift of the relative activation state of the InsP₃-CaMKII/PKD-HDAC5 system in HF. HDAC5 expression was unchanged in both rabbit and human HF (Figure 3A). However, HDAC5 was more heavily phosphorylated in LV homogenates of failing human hearts (by 58% and 120% in ICM and DCM, respectively; Figure 3A, right), which would tend to reduce the nuclear versus cytosol HDAC5 levels. Indeed, measurement by immunostaining with an HDAC5 antibody, which is indifferent to phosphorylation (Figure 3B), showed that the basal HDAC5_nuc/HDAC5_cyto ratio decreases from 8.4±0.3 to 1.8±0.6 in rabbit HF cells. This is similar to the shift measured using adenovirally expressed HDAC5-GFP in HF versus control rabbit myocytes (Figure 3C). Thus, although total HDAC5 expression does not change in HF, it is more phosphorylated and thus more cytosolic (which would derepress MEF2-driven transcription).

View larger version (43K): [in this window] [in a new window]   Figure 3. HDAC5 expression and subcellular location. A, HDAC5 expression and phosphorylation at Ser498 in failing human (right) and rabbit (left) hearts (n=6 for rabbits; n=7, 9, and 13 for NF, ICM, and DCM, respectively). Western blot signals were normalized to NF hearts. B, Immunocytochemistry of HDAC5 in control and HF rabbit. The small images show the colocalization (left) of the HDAC5 signal (middle) with the nucleus stained with Sytox orange (right). HDAC5 distribution was quantified as Fnuc/Fcyto (n=40 to 60). C, GFP-HDAC5 in short-term cultured rabbit myocytes mimics the native HDAC5 distribution (n=40). *P<0.05.

PKD and CaMKII Modulation of HDAC5 Distribution
Next we tested whether inhibition of CaMKII or PKD in HF myocytes could restore the nuclear to cytosol ratio toward the normal basal level. Figure 4A shows that inhibition of CaMKII (with KN-93) or of PKD (with Gö6976) partially restored basal F_nuc/F_cyto levels in HF myocytes toward that in nonfailing myocytes (6.3±0.3 and 5.6±0.4, respectively). Inhibition of both kinases fully restored F_nuc/F_cyto in HF myocytes to that of control (F_nuc/F_cyto 7.7±0.7). In addition, when we coexpressed HDAC5-GFP with dominant-negative kinases for CaMKII and PKD in HF myocytes, HDAC5 shifted significantly back into the nucleus (toward the normal baseline; Figure 4B).

View larger version (41K): [in this window] [in a new window]   Figure 4. PKD and CaMKII isoforms modulate HDAC5 nuclear export. A, Inhibition of PKD or CaMKII (with 1 µmol/L KN-93 or 10 µmol/L Gö6976) separately or simultaneously for 6 hours restores the GFP-HDAC5 distribution in HF myocytes to that of nonfailing (n=20). B, Overexpression of dominant-negative PKD or CaMKII isoforms also restores HDAC5 to the nucleus (n>30). C, Overexpression of PKD or CaMKII isoforms drives GFP-HDAC5 nuclear export (n=20). Western blots show the extent of kinase overexpression. D, Basal MEF2–luciferase activity is increased by overexpression of PKD or CaMKII isoforms (n=4). *P<0.05.

The converse experiment (Figure 4C) tested whether adenoviral overexpression of PKD1, CaMKII_B, or CaMKII_C in control rabbit myocytes would shift F_nuc/F_cyto HDAC5 levels toward those in HF myocytes. Adenovirus-mediated expression of these kinases in myocytes (from control hearts) resulted in approximately 10- and 5-fold increases in PKD and CaMKII isoforms, respectively (Western blots in Figure 4C). Overexpression of PKD or the nuclear CaMKII_B isoform robustly drove HDAC5 from the nucleus, whereas the cytosolic CaMKII_C had a weaker, yet significant effect (F_nuc/F_cyto were 3.8±0.5, 3.4±0.5, and 5.2±0.5, respectively). Therefore, overexpression of these kinases partially mimics the HDAC5 HF phenotype. Overexpression of PKD and CaMKII_B together (data not shown) further reduced the F_nuc/F_cyto HDAC5 levels (even lower than that in HF). To confirm that this cytosolic shift of HDAC5 on kinase overexpression reflects transcriptional activation, we used a MEF2–luciferase reporter construct. Figure 4D shows that overexpression of PKD or either CaMKII isoform increased basal MEF2-driven transcription (presumably via HDAC5 shifts, although other mechanisms cannot be excluded). Similarly, Zhang et al²⁹ showed that overexpression of either CaMKII isoform in cultured neonatal rat ventricular myocytes increases MEF2–luciferase activity in response to the hypertrophic agonist phenylephrine.

ET-1–Induced HDAC5 Nuclear Export in HF
In addition to the tonic HDAC5 nucleocytoplasmic shift in the basal state, we assessed whether the ability of ET-1 to trigger HDAC5 export dynamically was altered in HF. As in control myocytes, ET-1 exposure induced a time-dependent export of GFP-HDAC5 from the nucleus (Figure 5A). The export was analyzed as a decrease of F_nuc/F_cyto, normalized to the initial ratio. The control group was treated the same without ET-1 exposure. Despite the smaller fraction of HDAC5 in the nucleus initially, there was a larger percentage shift of HDAC5 in HF myocytes. Longer exposure to ET-1 could completely deplete the nucleus (as in control myocytes). This enhanced nuclear signaling could be blunted by inhibition of either PKD or CaMK (with Gö6976 and KN-93, respectively; Figure 5B). Moreover, inhibition of both kinases completely blocked HDAC5 nuclear export. In control adult rabbit myocytes overexpressing PKD, CaMKII_B, or CaMKII_C (via adenovirus), the ET-1–induced shift in F_nuc/F_cyto for HDAC5-GFP was significantly increased (34±2% decrease in normal myocytes versus 52±2%, 45±3%, and 43±2% for PKD-, CaMKII_B-, and CaMKII_C-overexpressing myocytes, respectively; Figure 5C). This demonstrates that the more rapid HDAC5 translocation induced by ET-1 in HF myocytes can be mimicked, in part, by overexpression of PKD or CaMKII.

View larger version (27K): [in this window] [in a new window]   Figure 5. ET-1–induced nuclear export of GFP-HDAC5. A, ET-1 (100 nmol/L) was applied for 60 minutes to quiescent rabbit ventricular myocyte expressing HDAC5-GFP. Mean individual myocyte measures of nuclear and cytosolic fluorescence (Fnuc/Fcyto) normalized to the initial value before ET-1 exposure (n=6 rabbits). Controls (Ctl) were treated the same, except without ET-1 application. B, Adv-GFP-HDAC5–infected myocytes were pretreated with 1 µmol/L KN-93 or 10 µmol/L Gö6976, separately or simultaneously for 6 hours, followed by ET-1 application (n=2 rabbits). C, Average end point (at 60 minutes) of HDAC5 nuclear export with and without ET-1 treatment (n=10 rabbits). *P<0.05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Class II HDACs are well established as nodal points in the transmission of cardiac stress to hypertrophic gene expression,^2–5 the paradigm being that signal-dependent phosphorylation of these HDACs induce their nuclear export, thereby relieving their repressive effect on transcription. In previous work,²³ we identified a novel pathway for ET-1–induced transcriptional regulation in adult ventricular myocytes. Namely, ET-1 stimulates InsP₃ production, which elicits a local Ca²⁺ release at the nuclear envelope by InsP₃Rs there, causing activation of nuclear CaMKII and consequent HDAC5 phosphorylation and nuclear export, with resultant derepression of MEF2-dependent transcription. Small molecule inhibitors suggested that CaMKII and PKD contributed about equally to HDAC5 phosphorylation in this process,²³ but only recently has there been evidence for Ca²⁺ dependence of PKD activation.³¹ In the present study, we examine how ET-1–induced HDAC5 nuclear export is altered in HF and how altered CaMKII and PKD expression and activation are functionally involved (using HDAC5 nuclear export to indicate HDAC5 transcriptional regulation). We find, for the first time, that there is increased PKD expression and activity in failing human and rabbit hearts (in addition to both CaMKII forms). Chronic activation of PKD and CaMKII in HF caused both a higher basal shift of HDAC5 out of the nucleus (implying chronic transcriptional activation) and an increased sensitivity of the system to acute ET-1 stimulation. Furthermore, acute PKD and CaMKII overexpression in normal myocytes were additive in mimicking the HF phenotype.

Upregulation of PKD and CaMKII Isoforms in HF
Upregulation of CaMKII in animal models and human HF has been reported previously,^18–21 but little information is available on the relative contribution of the cytosolic and nuclear CaMKII isoforms. This is of particular interest because nuclear signaling is usually attributed to the nuclear _B splice variant (containing a nuclear localization sequence). The cytosolic _C variant, on the other hand, is thought to mediate the effects on excitation–contraction coupling in HF, ie, the reduction in SR Ca²⁺ content and systolic function, and contribute to arrhythmogenesis by diastolic SR Ca²⁺ leak and Ca²⁺ present changes.³² Our data indicate that both CaMKII forms are upregulated and more active in HF (Figure 1). In rabbit HF, the cytosolic CaMKII_C was more strongly expressed and activated, whereas in human ventricle, the nuclear CaMKII_B increase was more prominent. Notably, acute overexpression of either CaMKII_B or CaMKII_C enhances HDAC4 phosphorylation and MEF2 reporter activity in cultured neonatal rat myocytes²⁹ or COS cells.³³ Additionally, transgenic mice overexpressing either nuclear or cytosolic CaMKII exhibit hypertrophy and HF (more dramatically with cytosolic CaMKII_C).^18,19 Thus, both CaMKII forms mediate HDAC nuclear export and hypertrophic transcriptional signaling, and their upregulation and activation must be functionally contributing in HF.

We also demonstrate upregulation of PKD expression and activity in both human and rabbit HF (Figure 2). This is consistent with PKD involvement in the "reactivation of fetal genes" because myocyte PKD expression is developmentally regulated (and much more expressed in fetal and neonatal versus adult hearts).¹¹ Also, in rodent hearts, PKD was elevated in response to chronic hypertension and pressure overload mediated by aortic constriction.¹⁴ Likewise, overexpression of constitutively active PKD (like CaMKII) resulted in cardiac hypertrophy (and PKD was suggested to have a role in cardiac fibrosis).⁵ All of this suggests that increased PKD activity may be causally related to adverse cardiac remodeling in HF. Besides these long-term effects, PKD may directly depress contractility by phosphorylating myofilament proteins and reducing Ca²⁺ sensitivity.^12,13 This may partly explain the attenuated inotropic response to ET-1 in failing human myocardium,³⁴ where PKD is more active. Moreover, PKD-dependent troponin I phosphorylation (at the same sites as PKA) reduces myofilament Ca²⁺ sensitivity^12,13 but without the prominent inotropic and lusitropic effects of PKA on Ca²⁺ current and SR Ca²⁺ uptake, which allow PKA to enhance contractility (despite myofilament desensitization). Thus, PKD effects on myofilaments may contribute to reduced systolic function in HF.

Enhanced Nuclear Signaling by HDAC5 in HF
Our hypothesis was that nuclear signaling to HDAC5 would be elevated in HF, given previously reported increases in neurohumoral signals (such as ET-1) and CaMKII and InsP₃R expression in HF²¹ and our previous delineation of this InsP₃-CaMKII-HDAC5 pathway.²³ We found unaltered HDAC5 expression in both human and rabbit HF (Figure 3), consistent with previous reports in disease states.^6,35 Instead, HDAC5 was more highly phosphorylated in human HF, indicating a higher cytosolic distribution (Figure 3). This was shown directly in rabbit myocytes with immunostaining and expression of GFP-tagged HDAC5 (basal F_nuc/F_cyto was 50% lower in HF). Note that this was not an all-or-none response; a reasonable fraction of HDAC5 remained nuclear. Moreover the system appears to be finely tuned, highly dynamic, and reversible because long exposures to ET-1 could drive HDAC5 from the nucleus completely and acute inhibition of the key kinases CaMKII or PKD could reverse the HF phenotype. Conversely, overexpression of these kinases mimicked the HF phenotype (Figure 4). We also found a higher sensitivity to ET-1 in HF, ie, a greater and faster nuclear export in response to the stimulus, despite the lower initial nuclear HDAC5 localization (Figure 5). This may result from enhanced sensitivity of the local nuclear InsP₃R-CaMKII-PKD-HDAC5 axis attributable to upregulation and partial basal activation or depressed β-AR/PKA signaling in HF (because PKA can inhibit ET-1–induced PKD activation in adult rat ventricular myocytes³⁶). Furthermore, the higher circulating ET-1 levels and ET_A receptor density in HF³⁷ would synergize with the more sensitive InsP₃R-CaMKII-PKD-HDAC5 axis to further reinforce the HF phenotype (and other neurohumoral agents in HF may differ). Notably, sensitization to ET-1 in HF could be blunted by CaMKII and PKD inhibitors or mimicked in normal myocytes by overexpression of these kinases. These experiments illustrate the redundancy of signaling to HDAC5 because either kinase alone had only a partial effect (and intervention of both kinases was required for maximal effect). In addition, the overexpression data suggest that nuclear CaMKII_B is more potent for the long-term transcriptional effects of CaMKII. A smaller effect on nuclear signaling was observed for the cytosolic isoform. A possible explanation is that cytosolic CaMKII blocks nuclear import of HDACs by rephosphorylating cytosolic HDAC (whereas nuclear CaMKII stimulates nuclear export).^34,29

Our findings in adult ventricular myocytes here and elsewhere²³ differ from recent reports in COS cells and cultured neonatal rat ventricular myocytes.^8,33,38 First, in COS cells, only PKD, but not constitutively active CaMKII, caused export of HDAC5. Second, CaMKII selectively signals to HDAC4 by means of a unique CaMKII-docking domain that was not found in HDAC5, -7, or -9.³³ Backs et al³³ did indicate a possible explanation for our findings. That is, HDAC4 can complex with HDAC5 and phosphorylation of HDAC4 by CaMKII could result in cotranslocation of HDAC5. However, in adult myocytes, we found that a nonphosphorylatable HDAC5 mutant abolished the CaMKII-dependent HDAC5 nuclear export,²³ proving that CaMKII-dependent HDAC5 (and not HDAC4) phosphorylation was required for ET-1–induced HDAC5 nuclear export. Conceivably, HDAC5 phosphorylation differs in adult rabbit and mouse ventricular myocytes (versus COS cells or cultured neonatal rat ventricular myocytes). Developmental downregulation of PKD¹¹ and possible enhanced access of CaMKII to HDAC5 may explain more prominent CaMKII-dependent HDAC5 regulation in adult ventricular myocytes.

In summary, we found increased expression and activity of the type II HDAC kinases CaMKII and PKD in HF. This contributed to more cytosolic versus nuclear HDAC5 in HF, less basal repression of transcription, and an increased sensitivity to stimulation by the hypertrophic agonist ET-1. Therefore, functional upregulation of PKD and CaMKII in HF may contribute to and reinforce altered gene expression in HF. CaMKII and/or PKD activation also have acute excitation–contraction coupling effects that could contribute to diminished contractility and arrhythmias in HF.^12,21,32 Therefore, interference with CaMKII or PKD phosphorylation may be a useful therapeutic strategy to limit both hypertrophic signaling and acute dysfunction in HF. A deeper understanding of the pleiotropic functions of these kinases in heart, especially for the underexplored PKD, will be important to assess therapeutic strategies for the treatment of HF.

	Acknowledgments

 
We thank Dr Kenneth B. Margulies for providing human tissue samples and Dr Xun Ai, Jodi Jeanes, Georgia Acuna, and Karl Hench for their contributions.

Sources of Funding

Supported by NIH grants R01-HL64724 and P01-HL80101 (to D.M.B.) and R01-HL46929 (to S.M.P.).

Disclosures

None.

	Footnotes

 
Original received August 9, 2007; resubmission received December 11, 2007; revised resubmission received January 9, 2008; accepted January 16, 2008.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Frey N, Olson EN. Cardiac hypertrophy: the good, the bad, and the ugly. Annu Rev Physiol. 2003; 65: 45–79.[CrossRef][Medline] [Order article via Infotrieve]

2. Chien KR. Stress pathways and heart failure. Cell. 1999; 98: 555–558.[CrossRef][Medline] [Order article via Infotrieve]

3. Backs J, Olson EN. Control of cardiac growth by histone acetylation/deacetylation. Circ Res. 2006; 98: 15–24.[Abstract/Free Full Text]

4. Olson EN, Schneider MD. Sizing up the heart: development redux in disease. Genes Dev. 2003; 17: 1937–1956.[Free Full Text]

5. McKinsey TA. Derepression of pathological cardiac genes by members of the CaM kinase superfamily. Cardiovasc Res. 2007; 73: 667–677.[Abstract/Free Full Text]

6. Zhang CL, McKinsey TA, Chang S, Antos CL, Hill JA, Olson EN. Class II histone deacetylases act as signal-responsive repressors of cardiac hypertrophy. Cell. 2002; 110: 479–488.[CrossRef][Medline] [Order article via Infotrieve]

7. McKinsey TA, Zhang CL, Lu J, Olson EN. Signal-dependent nuclear export of a histone deacetylase regulates muscle differentiation. Nature. 2000; 408: 106–111.[CrossRef][Medline] [Order article via Infotrieve]

8. Vega RB, Harrison BC, Meadows E, Roberts CR, Papst PJ, Olson EN, McKinsey TA. Protein kinases C and D mediate agonist-dependent cardiac hypertrophy through nuclear export of histone deacetylase 5. Mol Cell Biol. 2004; 24: 8374–8835.[Abstract/Free Full Text]

9. Rozengurt E, Rey O, Waldron RT. Protein kinase D signaling. J Biol Chem. 2005; 280: 13205–13208.[Free Full Text]

10. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the human genome. Science. 2002; 298: 1912–1934.[Abstract/Free Full Text]

11. Haworth RS, Goss MW, Rozengurt E, Avkiran M. Expression and activity of protein kinase D/protein kinase Cµ in myocardium: evidence for ₁-adrenergic receptor- and protein kinase C-mediated regulation. J Mol Cell Cardiol. 2000; 32: 1013–1023.[CrossRef][Medline] [Order article via Infotrieve]

12. Cuello F, Bardswell SC, Haworth RS, Yin X, Lutz S, Wieland T, Mayr M, Kentish JC, Avkiran M. Protein kinase D selectively targets cardiac troponin I and regulates myofilament Ca²⁺ sensitivity in ventricular myocytes. Circ Res. 2007; 100: 864–873.[Abstract/Free Full Text]

13. Haworth RS, Cuello F, Herron TJ, Franzen G, Kentish JC, Gautel M, Avkiran M. Protein kinase D is a novel mediator of cardiac troponin I phosphorylation and regulates myofilament function. Circ Res. 2004; 95: 1091–1099.[Abstract/Free Full Text]

14. Harrison BC, Kim MS, van Rooij E, Plato CF, Papst PJ, Vega RB, McAnally JA, Richardson JA, Bassel-Duby R, Olson EN, McKinsey TA. Regulation of cardiac stress signaling by protein kinase D1. Mol Cell Biol. 2006; 26: 3875–3888.[Abstract/Free Full Text]

15. Frey N, McKinsey TA, Olson EN. Decoding calcium signals involved in cardiac growth and function. Nat Med. 2000; 6: 1221–1227.[CrossRef][Medline] [Order article via Infotrieve]

16. Colomer JM, Means AR. Chronic elevation of calmodulin in the ventricles of transgenic mice increases the autonomous activity of calmodulin-dependent protein kinase II, which regulates atrial natriuretic factor gene expression. Mol Endocrinol. 2000; 14: 1125–1136.[Abstract/Free Full Text]

17. Zhu W, Zou Y, Shiojima I, Kudoh S, Aikawa R, Hayashi D, Mizukami M, Toko H, Shibasaki F, Yazaki Y, Nagai R, Komuro I. Ca²⁺/calmodulin-dependent kinase II and calcineurin play critical roles in endothelin-1-induced cardiomyocyte hypertrophy. J Biol Chem. 2000; 275: 15239–15245.[Abstract/Free Full Text]

18. Zhang T, Maier LS, Dalton ND, Miyamoto S, Ross J Jr, Bers DM, Brown JH. The deltaC isoform of CaMKII is activated in cardiac hypertrophy and induces dilated cardiomyopathy and heart failure. Circ Res. 2003; 92: 912–919.[Abstract/Free Full Text]

19. Zhang T, Johnson EN, Gu Y, Morisette MR, Sah VP, Gigena MS, Belke DD, Dillmann WH, Rogers TB, Schulman H, Ross J Jr, Brown JH. The cardiac-specific nuclear delta(B) isoform of Ca²⁺/calmodulin-dependent protein kinase II induces hypertrophy and dilated cardiomyopathy associated with increased protein phosphatase 2A activity. J Biol Chem. 2002; 277: 1261–1267.[Abstract/Free Full Text]

20. Hoch B, Meyer R, Hetzer R, Krause EG, Karczewski P. Identification and expression of delta-isoforms of the multifunctional Ca²⁺/calmodulin-dependent protein kinase in failing and nonfailing human myocardium. Circ Res. 1999; 84: 713–721.[Abstract/Free Full Text]

21. Ai X, Curran JW, Shannon TR, Bers DM, Pogwizd SM. Ca²⁺/calmodulin-dependent protein kinase modulates cardiac ryanodine receptor phosphorylation and sarcoplasmic reticulum Ca²⁺ leak in heart failure. Circ Res. 2005; 97: 1314–1322.[Abstract/Free Full Text]

22. Zhang R, Khoo MS, Wu Y, Yang Y, Grueter CE, Ni G, Price EE Jr, Thiel W, Guatimosim S, Song LS, Madu EC, Shah AN, Vishnivetskaya TA, Atkinson JB, Gurevich VV, Salama G, Lederer WJ, Colbran RJ, Anderson ME. Calmodulin kinase II inhibition protects against structural heart disease. Nat Med. 2005; 11: 409–417.[CrossRef][Medline] [Order article via Infotrieve]

23. Wu X, Zhang T, Bossuyt J, Li X, McKinsey TA, Dedman JR, Olson EN, Chen J, Brown JH, Bers DM. Local InsP3-dependent perinuclear Ca²⁺ signaling in cardiac myocyte excitation-transcription coupling. J Clin Invest. 2006; 116: 675–682.[CrossRef][Medline] [Order article via Infotrieve]

24. Go LO, Moschella MC, Watras J, Handa KK, Fyfe BS, Marks AR. Differential regulation of two types of intracellular calcium release channels during end-stage heart failure. J Clin Invest. 1995; 95: 888–894.[Medline] [Order article via Infotrieve]

25. Pogwizd SM, Schlotthauer K, Li L, Yuan W, Bers DM. Arrhythmogenesis and contractile dysfunction in heart failure: roles of sodium-calcium exchange, inward rectifier potassium current and residual β-adrenergic responsiveness. Circ Res. 2001; 88: 1159–1167.[Abstract/Free Full Text]

26. Pogwizd SM, Qi M, Yuan W, Samarel AM, Bers DM. Upregulation of Na⁺/Ca²⁺ exchanger expression in an arrhythmogenic rabbit model of heart failure. Circ Res. 1999; 85: 1009–1019.[Abstract/Free Full Text]

27. Pogwizd SM. Nonreentrant mechanisms underlying spontaneous ventricular arrhythmias in a model of nonischemic heart failure in rabbits. Circulation. 1995; 92: 1034–1048.[Abstract/Free Full Text]

28. Haworth RS, Sinnett-Smith J, Rozengurt E, Avkiran M. Protein kinase D inhibits plasma membrane Na⁺/H⁺ exchanger activity. Am J Physiol. 1999; 277: C1202–C1209.[Medline] [Order article via Infotrieve]

29. Zhang T, Kohlhaas M, Backs J, Mishra S, Philips W, Dybkova N, Chang S, Ling H, Bers DM, Maier LS, Olson EN, Brown JH. CaMKII isoforms differentially affect calcium handling but similarly regulate HDAC/MEF2 transcriptional responses. J Biol Chem. 2007; 282: 35078–35087.[Abstract/Free Full Text]

30. Huke S, Bers DM. Temporal dissociation of frequency-dependent acceleration of relaxation and protein phosphorylation by CaMKII. J Mol Cell Cardiol. 2007; 42: 590–599.[CrossRef][Medline] [Order article via Infotrieve]

31. Kunkel MT, Toker A, Tsien RY, Newton AC. Calcium-dependent regulation of protein kinase D revealed by a genetically encoded kinase activity reporter. J Biol Chem. 2007; 282: 6733–6742.[Abstract/Free Full Text]

32. Maier LS, Bers DM. Role of Ca²⁺/calmodulin-dependent protein kinase (CaMK) in excitation-contraction coupling in the heart. Cardiovasc Res. 2007; 73: 631–640.[Abstract/Free Full Text]

33. Backs J, Song K, Bezprozvannaya S, Chang S, Olson EN. CaM kinase II selectively signals to histone deacetylase 4 during cardiomyocyte hypertrophy. J Clin Invest. 2006; 116: 1853–1864.[CrossRef][Medline] [Order article via Infotrieve]

34. Pieske B, Beyermann B, Breu V, Löffler BM, Schlotthauer K, Maier LS, Schmidt-Schweda S, Just H, Hasenfuss G. Functional effects of endothelin and regulation of endothelin receptors in isolated human nonfailing and failing myocardium. Circulation. 1999; 99: 1802–1809.[Abstract/Free Full Text]

35. Chang S, McKinsey TA, Zhang CL, Richardson JA, Hill JA, Olson EN. Histone deacetylases 5 and 9 govern responsiveness of the heart to a subset of stress signals and play redundant roles in heart development. Mol Cell Biol. 2004; 24: 8467–8476.[Abstract/Free Full Text]

36. Haworth RS, Roberts NA, Cuello F, Avkiran M. Regulation of protein kinase D activity in adult myocardium: Novel counter-regulatory roles for protein kinase Cepsilon and protein kinase A. J Mol Cell Cardiol. 2007; 43: 686–695.[CrossRef][Medline] [Order article via Infotrieve]

37. Schneider MP, Boesen EI, Pollock DM. Contrasting actions of endothelin ET(A) and ET(B) receptors in cardiovascular disease. Annu Rev Pharmacol Toxicol. 2007; 47: 731–759.[CrossRef][Medline] [Order article via Infotrieve]

38. Little GH, Bai Y, Williams T, Poizat C. Nuclear Calcium/calmodulin dependent protein kinase II preferentially transmits signals to histone deacetylase 4 in cardiac cells. J Biol Chem. 2007; 282: 7219–7231.[Abstract/Free Full Text]

This article has been cited by other articles:

				S. Wagner, E. Hacker, E. Grandi, S. L. Weber, N. Dybkova, S. Sossalla, T. Sowa, L. Fabritz, P. Kirchhof, D. M. Bers, et al. Ca/Calmodulin Kinase II Differentially Modulates Potassium Currents Circ Arrhythmia Electrophysiol, June 1, 2009; 2(3): 285 - 294. [Abstract] [Full Text] [PDF]

				Highlights From The Literature Physiology, August 1, 2008; 23(4): 183 - 186. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 102/6/695    most recent CIRCRESAHA.107.169755v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bossuyt, J. Articles by Bers, D. M. Search for Related Content PubMed PubMed Citation Articles by Bossuyt, J. Articles by Bers, D. M. Related Collections Animal models of human disease Heart failure - basic studies Physiological and pathological control of gene expression