Heart
Calcium/Calmodulin-Dependent Protein Kinase II Couples
Wnt Signaling With Histone Deacetylase 4 and Mediates
Dishevelled-Induced Cardiomyopathy
Min Zhang, Marco Hagenmueller, Johannes H. Riffel, Michael M. Kreusser, Elmar Bernhold,
Jingjing Fan, Hugo A. Katus, Johannes Backs, Stefan E. Hardt
See Editorial Commentary, pp 287–288
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
Abstract—Activation of Wnt signaling results in maladaptive cardiac remodeling and cardiomyopathy. Recently, calcium/
calmodulin-dependent protein kinase II (CaMKII) was reported to be a pivotal participant in myocardial remodeling.
Because CaMKII was suggested as a downstream target of noncanonical Wnt signaling, we aimed to elucidate the role of
CaMKII in dishevelled-1–induced cardiomyopathy and the mechanisms underlying its function. Dishevelled-1–induced
cardiomyopathy was reversed by deletion of neither CaMKIIδ nor CaMKIIγ. Therefore, dishevelled-1–transgenic mice
were crossed with CaMKIIδγ double-knockout mice. These mice displayed a normal cardiac phenotype without cardiac
hypertrophy, fibrosis, apoptosis, or left ventricular dysfunction. Further mechanistic analyses unveiled that CaMKIIδγ
couples noncanonical Wnt signaling to histone deacetylase 4 and myosin enhancer factor 2. Therefore, our findings indicate
that the axis, consisting of dishevelled-1, CaMKII, histone deacetylase 4, and myosin enhancer factor 2, is an attractive
therapeutic target for prevention of cardiac remodeling and its progression to left ventricular dysfunction. (Hypertension.
2015;65:335-344. DOI: 10.1161/HYPERTENSIONAHA.114.04467.) Online Data Supplement
•
Key Words: calcium-calmodulin-dependent protein kinase type 2 ■ cardiomyopathies
■ histone deacetylase 4 ■ Wnt signaling pathway
C
■
dishevelled proteins
channels ≥3 branches of Wnt signal transduction: the Wnt/βcatenin pathway, the Wnt/Ca2+ pathway, and the Wnt/c-Jun
N-terminal kinase (JNK) pathway.9 Recent reports from our and
other groups have shown that dishevelled expression is upregulated after aortic banding (rat) and in tachycardia-induced heart
failure (porcine).10 Moreover, knockout of dishevelled-1 attenuates pressure-overload–induced left ventricular (LV) remodeling.11 In contrast, transgenic dishevelled-1 overexpression leads
to progressive dilated cardiomyopathy, which is accompanied
by the activation of all branches of Wnt signaling.10 Therefore,
dishevelled-1 is both necessary and sufficient for cardiac remodeling and dysfunction. However, the mechanisms bridging
chronic dishevelled-1 activation or upregulation to expressional
changes and successive cardiac dysfunction after pathological
injuries remain to be elucidated.
Calcium/calmodulin-dependent protein kinase II (CaMKII),
a multifunctional serine/threonine protein kinase with a broad
spectrum of substrates, plays an important role in the Wnt/
Ca2+ pathway.12 Several isoforms of CaMKII are derived from
ongestive heart failure is the leading cause of death in
developed countries.1 There are >23 million patients experiencing worldwide.1–4 Incidence of heart failure, and its associated morbidity and mortality, is still on the rise.5 At present,
therapeutic strategies to attenuate adverse cardiac remodeling
are limited to β-adrenoceptor antagonists and inhibitors of the
renin–angiotensin–aldosterone system. Unfortunately, these
therapies are insufficient in a substantial number of patients.
Given these clinical observations, new therapeutic approaches
are needed and may potentially be directed at signaling pathways that transduce neuroendocrine or stretch-mediated signals.
Pathological injuries reactivate several signaling pathways
in the adult heart that traditionally are thought to be operative
only in the developing heart. The Wnt signaling pathway, an
evolutionarily conserved intracellular signaling pathway, is
the most extensively investigated one of these pathways.6 It is
now evident that Wnt signaling not only is involved in cardiac
development but also is central to the myocardial remodeling
process.6–8 Dishevelled, the hub of Wnt signaling, regulates and
Received August 21, 2014; first decision September 5, 2014; revision accepted September 29, 2014.
From the Department of Cardiology, Angiology, and Pulmology (M.Z., M.H., J.H.R., M.M.K., E.B., J.F., H.A.K., S.E.H.) and Research Unit Cardiac
Epigenetics, Department of Cardiology (M.M.K., J.B.), University of Heidelberg, Heidelberg, Germany; Institute of Cardiology, Union Hospital, Tongji
Medical College of Huazhong University of Science and Technology, Wuhan, China (M.Z.); DZHK (German Center for Cardiovascular Research) (M.H.,
J.H.R., M.M.K., H.A.K., S.E.H., M.M.K., J.B.), Partner Site Heidelberg/Mannheim, Heidelberg, Germany; and Center for Cardiac and Circulatory
Diseases, Bruchsal, Germany (S.E.H.).
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.
114.04467/-/DC1.
Correspondence to Stefan E. Hardt, Department of Cardiology, Angiology, and Pulmology, University of Heidelberg, Im Neuenheimer Feld 410, 69120
Heidelberg, Germany. E-mail [email protected]
© 2014 American Heart Association, Inc.
Hypertension is available at http://hyper.ahajournals.org
DOI: 10.1161/HYPERTENSIONAHA.114.04467
335
336 Hypertension February 2015
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
4 genes (α, β, δ, and γ) that display distinct but overlapping
expression patterns.13 The α and β isoforms are almost exclusively expressed in the brain, whereas the other isoforms are
found more ubiquitously. In the heart, CaMKII δ and γ are
the predominant CaMKII isoforms, with CaMKIIδ displaying
the highest expression level. Upregulation of CaMKII expression and activity is a general characteristic of heart failure in
humans and in animal models.14,15
Although CaMKII is a critical downstream target of dishevelled-1,16 a functional relationship between these proteins in
cardiac remodeling has not been identified to date. In this
study, deletion of neither CaMKII δ nor γ reverses dishevelled-1–induced cardiomyopathy. Interestingly, here we found
that mice lacking both CaMKII δ and γ isoforms are resistant
to dishevelled-1–induced cardiac hypertrophy, fibrosis, and
LV dysfunction, which is mediated by the histone deacetylase 4/myosin enhancer factor 2 (HDAC4/MEF2) complex.
Therefore, this investigation suggests that both the δ and γ
isoforms of CaMKII serve as fundamental links between
dishevelled and consecutive detrimental effects. Our findings
reveal that the dishevelled–CaMKII–HDAC4–MEF2 axis
holds a pivotal role in maladaptive cardiac remodeling and is
a potential novel therapeutic target for the prevention of LV
dysfunction.
Methods
Materials and Methods are available in the online-only Data
Supplement .
Results
CaMKIIδ-Knockout, CaMKIIγ-Knockout,
CaMKIIδγ-Knockout, and Dishevelled-1–
Transgenic Mice Lacking CaMKII Are Generated
We previously reported that dishevelled-1–transgenic (DVL)
mice exhibit a robust increase in both phosphorylated and
total CaMKII levels,10 which are known to play a key role in
cardiac dysfunction. In the same study, we have also found
that Dlv-1 knockdown prevents cardiomyocyte hypertrophy
in response to β-adrenergic stimulation. Using Western blot,
here we further report that isoproterenol (Iso) induced hypertrophy accompanied with increased dishevelled-1, phosphorylated CaMKII, phosphorylated HDAC4 and MEF2, which
could be reversed by dishevelled-1 knockdown (Figure S1
in the online-only Data Supplement). To assess the potential
role of CaMKII in the detrimental effects after dishevelled-1
overexpression, we crossed DVL mice, which display cardiac
hypertrophy and LV dysfunction, with CaMKIIδ-knockout
mice. However, the phenotype of the DVL mice was not rescued by the deletion of CaMKIIδ (Figure S2).
Because we hypothesized that CaMKIIγ may also play a
crucial role in dishevelled-1–induced pathological injuries, we
crossed cardiac-directed CaMKIIγ-knockout and CaMKIIδγknockout animal lines with DVL mice (Figure 1A). All the
mice were viable and developed normally until early adulthood. In the CaMKIIδ-knockout and CaMKIIγ-knockout
mice, as well as in the CaMKIIδγ-knockout animals, no overt
baseline cardiac phenotype was observed.17,18 The phenotype
of the CaMKIIδγ-knockout is described in detail in a separate article.19 Interestingly, CaMKIIδγ-knockout reversed the
phenotype caused by dishevelled-1 overexpression, whereas
CaMKIIγ-knockout did not (Figure S3), indicating that both
the δ and γ isoforms of CaMKII are crucial in dishevelled1–induced cardiac remodeling. To gain further insight into
the mechanisms underlying these outcomes, we selected
4 mouse lines that were generated by crossing DVL mice
with CaMKIIδγ-knockout mice: wild-type mice as control
(CTRL), CaMKIIδγ-deficient, and dishevelled-1 wild-type
mice as CKO, dishevelled-1–transgenic mice with normal
level of CaMKIIδγ as DVL, and dishevelled-1–transgenic
mice lacking CaMKIIδγ as DWC (Figure 1B; Table S1).
Figure 1. Generation of dishevelled (Dvl)-1-transgenic (Tg; DVL) mice harboring a calcium/calmodulin-dependent protein kinase IIδγknockout (CaMKIIδγ-KO [CKO]). A, Genetic strategy for depleting CaMKII in DVL mice. B, Western blots of Dvl-1 overexpression in DVL
mice and DVL mice lacking CaMKIIδγ (DWC), as well as a substantial reduction in CaMKII protein levels in heart homogenates from
CKO and DWC mice at 12 weeks, indicating successful deletion of CaMKII in DVL mice. C, mRNA levels of CaMKII isoforms α in control
(CTRL), CKO, DVL, and DWC mouse hearts; CaMKIIα was not detectable. All RNA levels were normalized to internal hypoxanthine
guanine phosphoribosyl transferase (Hprt) expression and presented as fold-change relative to CTRL samples. Values are presented as
mean±SEM. n>10 per group. MHC indicates myosin heavy chain; and ND, not detectable.
Zhang et al Dvl-Induced Cardiomyopathy and CaMKIIδγ 337
CaMKIIα expression was not detected in any of these groups
(Figure 1C).
Deletion of CaMKIIδγ Attenuates Dishevelled1–Induced Severe Cardiac Hypertrophy, Fibrosis,
and Apoptosis
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
In accordance with our previous findings, the size and structure
of the hearts of CKO mice displayed no baseline differences
from the CTRL, whereas the DVL mice exhibited remarkable
cardiac hypertrophy.10 Interestingly, deletion of CaMKIIδγ
nearly normalized dishevelled-1–induced enhancement of
heart size and cross-sectional myocyte area, using hematoxylin and eosin staining, which is as typical as the wheat germ
agglutinin lectin staining, in DWC mice versus DVL mice
(Figure 2A and 2B). DVL mice showed a significant augmentation in heart weight/tibia length ratio and LV weight/
tibia length ratio, whereas in the DWC group, depletion of
CaMKIIδγ inhibited the dishevelled-1–evoked increase in
heart weight/tibia length and LV weight/tibia length ratio
(Figure 2D). This effect is also highlighted by reduced expression of the hypertrophic markers atrial natriuretic factor and
β-myosin heavy chain in DWC mice compared with DVL
mice (Figure 2E). In vivo noninvasive transthoracic echocardiography revealed reduction of the LV posterior wall thickness in systole and end-diastolic diameter, further reflecting
the attenuated hypertrophic response to dishevelled-1 overexpression by CaMKIIδγ silencing (Figure 3A, 3C, and 3D).
Taken together, these data indicate that deletion of CaMKIIδγ
reversed dishevelled-1–induced cardiac hypertrophy.
Apart from cardiac myocyte hypertrophy, it has been recognized that cardiac fibrosis and apoptosis are pivotal to the
development of LV dysfunction. In this study, no significant
difference in fibrosis and apoptosis was observed between
CTRL mice and CKO mice, whereas DVL mice had sharply
augmented interstitial and perivascular fibrosis and apoptosis that were significantly reduced by ablation of CaMKIIδγ
(Figure 2A and 2C; Figure S4).
Deletion of CaMKIIδγ Rescues
Dishevelled-1–Induced LV Dysfunction
Because cardiac hypertrophy, fibrosis, and apoptosis were ameliorated in DWC mice, we used transthoracic echocardiography and hemodynamic measurements to test whether cardiac
function was improved in these animals. Compared with CTRL
Figure 2. Diminished cardiac hypertrophy and fibrosis of dishevelled-1–transgenic (DVL) mice after deletion of calcium/calmodulindependent protein kinase IIδγ (CaMKIIδγ). A, Heart size was assessed to detect reduction of cardiac hypertrophy. Histological sections
were stained with hematoxylin and eosin to measure myocyte size and with picrosirius red to detect fibrosis. B and C, Quantification
of myocyte size (B) and fibrosis area (C) presented as fold-changes in cross-sectional area of myocyte size and fibrosis in each group
compared with control (CTRL) mice. D, Heart weight/tibia length (HW/TL) ratio and left ventricular weight/tibia length
(LVW/TL) ratio reflect blunted cardiac hypertrophy in DWC mice after ablation of CaMKIIδγ. E, Fold-changes in mRNA levels of the
hypertrophic markers atrial natriuretic factor (ANF) and β myosin heavy chain (MHC). Values are presented as mean±SEM. n>10 per
group. *P<0.05 vs CTRL; #P<0.05 vs DWC.
338 Hypertension February 2015
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
Figure 3. Ablation of calcium/calmodulin-dependent protein kinase IIδγ (CaMKIIδγ) rescues cardiac dysfunction in dishevelled-1–
transgenic (DVL) mice. Transthoracic echocardiography revealed normalized dimensions and function in DVL hearts after deletion
of CaMKII. A, Four representative M mode images. B, Pressure volume loops showed the rescue of systolic and diastolic cardiac
dysfunction in the DWC mice. C–F, Quantitative analysis of left ventricular posterior wall thickness (LV PWTs; C), end-diastolic diameter
(EDD; D), ejection fraction (EF; E), and fraction shortening (FS; F). Notably, the EF and FS volumes were completely normalized in DWC
mice by CaMKIIδγ deficiency, strongly suggesting the rescue of cardiac function. Values are reported as mean±SEM. n>10 per group.
*P<0.05 vs control (CTRL); #P<0.05 vs DWC. CKO indicates CaMKII-KO.
mice, DVL mice displayed marked enhancement of end-diastolic volume, end-systolic volume, and end-diastolic pressure,
as well as an apparent reduction in the maximal velocities of
contraction (dp/dt max) and relaxation (dp/dt min and τ), thus
reflecting severely impaired systolic and diastolic cardiac function. However, all of these hemodynamic parameters were completely normalized in DWC mice after depletion of CaMKIIδγ,
suggesting a rescue of LV function (Table; Figure 3B). As a
key observation of this study, we found that depletion of
CaMKIIδγ normalized the reduced ejection fraction and fraction shortening, as measured by echocardiography in DWC
mice (Figure 3A, 3E, and 3F), further supporting the rescue of
LV function in DWC mice via silencing of CaMKIIδγ.
CaMKII Contributes to Noncanonical Wnt
Signaling–Induced Cardiac Hypertrophy by
Mediating MEF2 via Phosphorylation/Export of
HDAC4
Previously, dishevelled-1 overexpression led to robust activation of CaMKII, which induced an increase in the phosphorylation of HDAC4, a repressor of hypertrophy-related
transcription factors, such as MEF2, serum response factor,
and GATA.10,20,21 Furthermore, we found a strong increase
in CaMKII-related HDAC4 phosphorylation in DVL mice
when compared with CTRL littermates via in vitro HDAC4
kinase assay. Importantly, CaMKIIδγ deletion in DWC mice
significantly blocked phosphorylation of HDAC4 (Figure 4A
and 4B), implying that HDAC4 is a downstream target of
CaMKII in response to dishevelled-1 overexpression. To gain
further insight into the links among dishevelled-1, CaMKII,
and HDAC4, adenoviral dishevelled-1 overexpression was
used to stimulate cultured neonatal rat ventricular myocytes.
KN93, a CaMKII inhibitor, effectively prevented dishevelled1–induced phosphorylation of CaMKII and HDAC4, whereas
its analogue KN92 did not (Figure 4C). In addition, pharmacological inhibition of CaMKII also inhibited the dishevelled1–induced increase in myocyte size and atrial natriuretic
factor expression (Figure 5A and 5B), indicating that CaMKII
inhibition blocks cardiomyocyte hypertrophy resulting from
dishevelled-1 overexpression.
To test whether HDAC4 inhibits dishevelled-1–related
cardiac hypertrophy further, neonatal rat ventricular myocytes were treated with adenovirus encoding a phosphorylation-resistant HDAC4 mutant, HDAC4-S246,467,632A
Zhang et al Dvl-Induced Cardiomyopathy and CaMKIIδγ 339
Table. Hemodynamic Parameters Demonstrate the Rescue
of Cardiac Dysfunction in Dishevelled-1–Transgenic Mice by
Calcium/Calmodulin-Dependent Protein Kinase IIδγ-Knockout
Parameters
CTRL
CKO
DVL
DWC
Heart rate, bpm
493±12
End-systolic
volume, μL
6.7±0.8
487±19
451±14
499±15
6.9±0.5
12.7±1.5*
7.3±0.8
End-diastolic
volume, μL
19.6±1.8
17.8±2.9
31.4±4.7*
18.7±2.2
End-systolic
pressure, mm Hg
85.9±2.8
84.6±2.2
79.8±1.9
82.9±2.9
End-diastolic
pressure, mm Hg
5.1±0.5
5.0±0.5
Stroke volume, μL
12.1±0.9
12.3±1.6
9.8±1.2*
11.0±1.8
4.7±0.8
12.7±1.0
dP/dt max, mm Hg/s
6511±388
6283±404
4194±296*
6242±359
dP/dt min, mm Hg/s
-5823±401
-5672±354
-3853±289*
-5457±312
12.1±0.8
12.8±0.9
15.5±1.3*
13. 0±0.6
τ, ms
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
Values are demonstrated as mean±SEM. CKO indicates calcium/calmodulindependent protein kinase IIδγ-knockout; CTRL, control, DVL, dishevelled-1–
transgenic; DWC, dishevelled-1–transgenic with calcium/calmodulin-dependent
protein kinase IIδγ-knockout; and τ, tau (a parameter reflecting diastolic
cardiac function).
*P<0.05 vs CTRL; P<0.05 vs DWC. n≥14 per group.
(HDAC4-3S/A), which localizes predominantly to the nucleus.
Interestingly, HDAC4-3S/A completely reversed dishevelled-1
overexpression-induced enhancement of expression of the
hypertrophic marker atrial natriuretic factor and cardiomyocyte
size (Figure 4D; Figure 5A and 5B), suggesting that HDAC4
is central to dishevelled-1–evoked cardiomyocyte hypertrophy.
Interestingly, dishevelled-1 overexpression resulted in a
remarkable increase in MEF2 activity, which was blocked
by CaMKII inhibition and by HDAC4-3S/A overexpression
(Figure 5C). Meanwhile, we observed Wnt5a-induced cardiomyocyte hypertrophy, complying with MEF2 activation. Both
of them could be inhibited by CaMKII inhibition (Figure S5).
Taken together, these observations confirm that CaMKII and
HDAC4 are critical to noncanonical Wnt signaling–induced
cardiac hypertrophy, providing an important link to MEF2 as
the end point for noncanonical Wnt signaling–induced hypertrophy in cardiomyocytes.
Dishevelled-1–Induced Activation of β-Catenin,
Protein Kinase C, and JNK Is Not Affected by
CaMKII Deletion
CaMKII was previously shown to interact with the Wnt/βcatenin signaling pathway,22 raising the question of whether
the reduction in dishevelled-1–induced cardiomyopathy after
genetic deletion or pharmacological inhibition of CaMKII
occurs via suppression of Wnt/β-catenin signaling. These 2
approaches of cardiomyopathy reduction resulted in neither
the inhibition of β-catenin accumulation nor the downregulation of the β-catenin–dependent Wnt target gene Axin2
(Figure 6A–6C). Thus, attenuation of dishevelled-1–induced
cardiomyopathy by CaMKII suppression is not mediated
through interaction with the canonical Wnt/β-catenin signaling pathway. Furthermore, CaMKII-knockout in the DWC
mice did not prevent the activation of protein kinase C and
JNK, as assessed by Western blot (data not shown). Figure 7
illustrates the proposed mechanism by which dishevelled-1
Figure 4. Implication of the histone deacetylase 4/myosin enhancer factor 2 (HDAC4/MEF2) complex in dishevelled (Dvl)-1–induced
cardiac hypertrophy. A, Western blot of increased levels of phosphorylated calcium/calmodulin-dependent protein kinase II (CaMKII)
after Dvl-1 overexpression, phosphorylated HDAC4 at the CaMKII phosphorylation site Ser-632, and total HDAC4 levels. B, HDAC4
kinase activity assayed with GST-HDAC4 as substrate. HDAC4 phosphorylation was significantly increased in ventricular extracts from
Dvl-1-transgenic (DVL) mice. C, The use of an adenovirus harboring wild-type Dvl-1 (Dvl-1-wt) stimulated cardiomyocytes in vitro and
phosphorylated the Ser-632 site of HDAC4. These processes were inhibited by addition of KN93, a CaMKII inhibitor, but not by the
inactive analogue KN92, as revealed by Western blot. D, Western blot of the change in concentration of adenoral Flag-HDAC4-3S/A.
Values are reported as mean±SEM. n>10 per group. *P<0.05 vs control (CTRL; mice) or AdC KN92 (cardiomyocytes); #P<0.05 vs DWC
(mice), or AdDvl KN93 (cardiomyocytes). All experiments with cardiomyocytes were performed ≥3×. AdC indicates control adenovirus;
AdDvl, adenovirus harboring the Dvl-1-wt gene; and HDAC4-3S/A, HDAC4-S246,467,632A.
340 Hypertension February 2015
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
Figure 5. Histone deacetylase 4/myosin enhancer factor 2 (HDAC4/MEF2) are involved in cardiomyocyte hypertrophy. A and B,
Representative images of the inhibition of augmentation of dishevelled (Dvl-1-wild-type (wt)–induced cardiomyocyte size, and fetal
atrial natriuretic factor (ANF) expression by treating cardiomyocytes with KN93 and HDAC4-3S/A, a phosphorylation-resistant HDAC4
mutant. C, MEF2-dependent luciferase activity was markedly upregulated by Dvl-1 overexpression; KN93 and HDAC4-3S/A inhibited
Dvl-1-induced upregulation of luciferase activity. Values are reported as mean±SEM. *P<0.05 vs AdC KN92; #P<0.05 vs AdDvl KN93.
All experiments with cardiomyocytes were performed ≥3×. AdC indicates control adenovirus; AdDvl, adenovirus harboring the Dvl-1-wt
gene; and HDAC4-3S/A, HDAC4-S246,467,632A.
activates CaMKII, phosphorylates HDAC4, and mediates
MEF2-related gene transcription, which consequently leads
to cardiac hypertrophy.
Discussion
Wnt signaling is centrally involved in myocardial remodeling
after pathological injuries. Here, we show that combined deletion of CaMKII δ and γ rescues dishevelled-1–induced cardiac hypertrophy, fibrosis, and apoptosis, and also completely
reverses dishevelled-1–induced LV dysfunction. Further
analyses of signaling events indicated that cardioprotection in
DWC mice via CaMKIIδγ deletion was mediated by downregulation of MEF2 transcription through phosphorylation/
export of HDAC4 from the nucleus. This is the first report to
demonstrate a crucial role for the CaMKII axis in association
with cardiomyopathy induced by activation of Wnt signaling
components. Wnt5a-induced CaMKII activities are mediated
via HDAC4 and MEF2 further downstream functionally linking Wnt signaling to HDAC4 for the first time.
Previous studies revealed that dishevelled-1 is necessary
and sufficient to induce cardiac remodeling and dysfunction,10,11 but the main effecter acting after pathological activation of dishevelled-1 has not yet been determined. Although
some studies support the notion that dishevelled activation
is specific to the noncanonical Wnt pathway,23 other data
indicate that dishevelled activation is critical for all 3 Wnt
branches.6,10,24 During the process of cardiac remodeling, we
observed that dishevelled-1 activation involves all 3 Wnt
branches, with predominant signaling to the noncanonical
pathway. In addition, we show that the dishevelled–CaMKII
axis is more important in this context than protein kinase C
and JNK, a conclusion supported by multiple lines of evidence. In the DWC group, deletion of CaMKIIδγ blocked
dishevelled-1–induced increases in heart size, myocyte area,
heart weight/tibia length, LV weight/tibia length ratio, hypertrophic marker gene expression, posterior wall thickness, and
end-diastolic diameter in response to dishevelled-1 overexpression. Cardiac fibrosis and apoptosis were ameliorated by
Zhang et al Dvl-Induced Cardiomyopathy and CaMKIIδγ 341
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
Figure 6. Genetic deletion and pharmacological inhibition of calcium/calmodulin-dependent protein kinase II (CaMKII) do not affect βcatenin accumulation. A and B, Western blots and densitometry of active and total β-catenin levels revealed β-catenin accumulation after
dishevelled (Dvl-1) overexpression, even after genetic deletion and pharmacological inhibition of CaMKII. C, No significant changes in the
levels of the Wnt/β-catenin target gene Axin2 were detected relative to Dvl-1 overexpression both in vivo and in vitro, indicating that the
Wnt/β-catenin signaling pathway was not altered by CaMKII suppression. Values are reported as mean±SEM. n>10 per group. *P<0.05
vs control (CTRL; mice) or AdC KN92 (cardiomyocytes); #P<0.05 vs DWC (mice), or AdDvl KN93 (cardiomyocytes). All experiments with
cardiomyocytes were performed ≥3×. AdC indicates control adenovirus; AdDvl, adenovirus harboring the Dvl-1-wt gene; CKO, CaMKIIknockout; DVL, Dvl-1-transgenic; DWC, DVL mice lacking CaMKIIδγ; and HDAC4-3S/A, HDAC4-S246,467,632A.
the loss of CaMKIIδγ, and more importantly, cardiac dysfunction induced by dishevelled-1 was completely rescued
by CaMKIIδγ deletion. Dishevelled-1, CaMKII, and HDAC4
were activated simultaneously after aortic banding at 72 hours
and 7 days. CaMKIIδγ-knockout attenuated aortic bandinginduced cardiac dysfunction and interstitial fibrosis but not
hypertrophy.19 Whereas after aortic banding and Iso stimulation, cardiac expression of Rcan1-4 as a measure for calcineurin activity was highly upregulated in CaMKIIδγ-knockout
mice19; this was not the case in DVL mice (data not shown).
As cardiac hypertrophy in CaMKIIδγ-knockout mice after
aortic banding and Iso stimulation was clearly dependent on
the calcineurin overacitvation,19 the lack of that in DVL mice
could explain why these animals were protected from cardiac
hypertrophy. However, it remains unclear by which pathways
aortic banding and Iso treatment, but not dishevelled-transgenic, lead to an overactivation of calcineurin in CaMKIIδγknockout mice.
In Figure 6, we demonstrated that β-catenin and Axin2
were increased by dishevelled-1 overexpression and
not inhibited by CaMKIIδγ-knockout, which indicates
that the canonical Wnt signaling is still activated in the
dishevelled-1–transgenic with CaMKIIδγ-knockout mice.
Notably, CaMKIIδγ deletion also did not result in significant changes in the levels of protein kinase C, or JNK
because both of these 2 proteins increased similarly in DVL
mice in the presence or absence of CaMKIIδγ expression.
Therefore, we conclude that dishevelled-1 upregulation
activates all 3 branches of the Wnt signaling pathway, with
CaMKII acting as the most pivotal component in the process of myocardial remodeling.
Apart from the implication of CaMKII in excitation–contraction coupling and gene transcription in the heart, CaMKII
has also been associated with the development of cardiac
remodeling and dysfunction.18,25–27
Interestingly, in this study, we observed that deletion of
CaMKIIδ or CaMKIIγ alone is not sufficient to rescue the phenotype of DVL mice, but simultaneous knockout of CaMKIIδ
and CaMKIIγ does rescue the phenotype, indicating that
CaMKIIγ, which is not the most highly expressed CaMKII
isoform in the heart, may also function after pathological
injuries. Therefore, when developing new pharmacological
342 Hypertension February 2015
Figure 7. Model of the molecular
mechanism underlying cardiac
hypertrophy. Dishevelled (Dvl)-1
predominantly transduces signal to
calcium/calmodulin-dependent protein
kinase II (CaMKII), which phosphorylates
histone deacetylase 4 (HDAC4), a
repressor of myosin enhancer factor 2
(MEF2). Without repression of HDAC4,
MEF2-mediated gene expression leads
to cardiac remodeling. No interaction
between Wnt/CaMKII signaling and βcatenin/T-cell factor transcription was
observed in this investigation. Targetspecific inhibition of Dvl, CaMKII, or MEF2
may be attractive strategies to block
pathological cardiac remodeling and its
progression to left ventricular dysfunction.
LEF indicates lymphoid-enhancing factor;
and TCF, T-cell factor.
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
compounds to treat cardiac remodeling, both isoforms of
CaMKII δ and γ should be taken into account.
Although the loss of CaMKIIδγ considerably ameliorated
cardiac hypertrophy and fibrosis in response to dishevelled-1
overexpression, 2 interesting questions arise from our observations of dishevelled-1 transgenic mice deficient in CaMKIIδγ.
First, why is atrial natriuretic factor expression in DWC mice
still significantly increased when compared with CTRL animals, but much less than DVL mice (Figure 2E)? Second, why
is cardiac fibrosis not completely normalized (Figure 2C)? We
here suggest the involvement of the β-catenin–dependent Wnt
signaling rather than the Wnt/JNK pathway. This hypothesis
is in line with our recent findings that Dapper-1–induced activation of canonical Wnt signaling results in mild myocardial
remodeling.28 And, it is also supported by previous findings
that cardiomyocyte-specific β-catenin loss-of-function mutations lead to attenuated LV remodeling and improved cardiac
function after pressure overload, chronic angiotensin II stimulation, or experimental infarction.29–31 Conversely, augmented
mortality and dilated cardiomyopathy occur in conditional
β-catenin gain-of-function mutations.30,31 Therefore, although
CaMKII signaling predominates, canonical Wnt/β-catenin
signaling may also partially contribute to dishevelled-1–induced cardiac remodeling. Further investigations are necessary to gain additional insight into this issue.
To examine the downstream signaling components of Wnt/
CaMKII further, we examined the class II HDACs, which can
be phosphorylated by CaMK and protein kinase D. HDAC4,
a class II HDAC, has recently been reported to have a specific
CaMKII docking site that is not present in other HDACs.18,21,32
Zhang et al33 reported that HDAC4 (but not HDAC5) kinase
activity is apparently increased in CaMKII transgenic mouse
hearts when compared with wild-type controls. In comparison, ablation of CaMKIIδ inhibits cardiac hypertrophy.18
Mechanistically, depletion of CaMKIIδ suppresses phosphorylation/export of HDAC4 from the nucleus,34 leading to
downregulation of MEF2 transcription, which is both necessary and sufficient for pathological cardiac remodeling.18,32,35
We observed that total and phosphorylated HDAC4 levels are
comparably augmented in the DVL mouse heart versus the
CTRL mouse heart, and furthermore, ablation of CaMKII
blocked HDAC4 phosphorylation in DWC mice. We note
that the inclusion of HDAC4-3S/A, a signal-resistant HDAC4
mutant that localizes predominantly to the nucleus, reduced
hypertrophic marker gene expression evoked by dishevelled-1,
indicating that HDAC4 functions as a downstream target of
CaMKII after dishevelled-1 overexpression. Interestingly,
CaMKIIδB is among the splicing variants that are upregulated
in cardiac hypertrophy.36 We found previously that CaMKII
selectively signals to HDAC4 via binding to a unique docking
site and phosphorylation of Ser-467 and Ser-632.32
The transcription factor MEF2 has been suggested to serve
as a common end point for hypertrophic signaling pathways
in cardiomyocytes.37 HDAC4 is known to be associated with
MEF2 and represses its activation; signal-dependent dissociation of HDAC4 from MEF2 results in activation of MEF2
and induction of cardiac hypertrophy.32,33,37 CaMKII is able to
control MEF2 activation and hypertrophic gene expression
selectively both in vitro and in vivo by regulating phosphorylation and dissociation of HDAC4 from MEF2. In the present
study, pharmacological inhibition and genetic elimination of
CaMKII blocked dishevelled-1 upregulation-induced cardiac
hypertrophy and was accompanied by decreased HDAC4
phosphorylation. Interestingly, knockdown of dishevelled-1
abrogated Iso-induced activation of CaMKII, HDAC4, and
MEF2. Accordingly, we hypothesized that MEF2 serves as the
end point of the dishevelled-1–induced hypertrophic signaling pathway, a conjecture strongly supported by the observation that dishevelled-1 overexpression induced a remarkable
increase in MEF2 activity. This increase was completely
inhibited by CaMKII antagonists and HDAC4-3S/A overexpression, reflecting that the dishevelled–CaMKII–HDAC4
axis converges at MEF2. Furthermore, we found that Wnt5a
leads to cardiomyocyte hypertrophy and the activation of
MEF2, and these could be reversed by KN93, a CaMKII
inhibitor. This confirms that the noncanonical Wnt signal can
Zhang et al Dvl-Induced Cardiomyopathy and CaMKIIδγ 343
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
activate MEF2 via CaMKII. Certainly, KN93 is a nonselective CaMKII inhibitor; however, it should be emphasized
here that the α and β isoforms of CaMKII are almost exclusively expressed in the brain. In the heart, CaMKII δ and γ
are the predominant CaMKII isoforms. In Figure 1, we also
confirmed this. Therefore, even though the KN compounds
are nonselective, it will not affect the outcome in this specific
context.
Findings of Kühl et al38 support the link between intracellular Ca2+ release and the activation of CaMKII. Taken
together previous observations39 with our findings, we think
that there is no interaction between dishevelled and CaMKII.
In the heart, the binding of the noncanoncial Wnt ligand to
cognate Frizzled receptor activates dishevelled. Receptor
ligand interaction and activation of dishevelled further activate phospholipase C, which leads to a short-lived increase in
the concentration of inositol 1,4,5-triphosphate and 1,2 diacylglycerol. Inositol 1,4,5-triphosphate diffuses through the
cytosol and interacts with the calcium channels present on the
membrane of endoplasmic reticulum resulting in release of
calcium ions. Calcium ions along with ubiquitously expressed
eukaryotic protein calmodulin activate CaMKII.
Perspectives
Our present work demonstrates that dishevelled-1 predominantly signals to CaMKII, which then induces phosphorylation/dissociation of HDAC4 from MEF2. Without repression
of HDAC4, MEF2-related gene transcription contributes to
cardiac hypertrophy and LV dysfunction. Therefore, our study
yields insight into the mechanism of pathogenesis of cardiac
dysfunction and has substantial implications for the development of novel interventions for preventing cardiac remodeling and its progression to heart dysfunction by targeting the
dishevelled–CaMKII–HDAC4–MEF2 axis.
Acknowledgments
We thank Silvia Harrack for her excellent technical assistance.
Sources of Funding
This work was partially supported by the Deutsche
Forschungsgemeinschaft (DFG) (HA 2959/3-1 and HA 2959/32 within the DFG Forschergruppe1036), (S.E. Hardt), and by the
China Scholarship Council (M. Zhang). J. Backs was supported
by the Deutsche Forschungsgemeinschaft (BA 2258/2-1 within
the Emmy Noether Program), by the European Commission (FP7Health-2010; MEDIA-261409), by the ADUMED foundation and the
Klaus-Georg-und-Sigrid-Hengstberger-Forschungsstipendium.
Disclosures
None.
References
1. Vitali E, Colombo T, Bruschi G, Garatti A, Russo C, Lanfranconi M,
Frigerio M. Different clinical scenarios for circulatory mechanical support
in acute and chronic heart failure. Am J Cardiol. 2005;96(12A):34L–41L.
2. Bui AL, Horwich TB, Fonarow GC. Epidemiology and risk profile of
heart failure. Nat Rev Cardiol. 2011;8:30–41.
3. Daskalopoulos EP, Janssen BJ, Blankesteijn WM. Targeting Wnt signaling to improve wound healing after myocardial infarction. Methods Mol
Biol. 2013;1037:355–380.
4. Liu L, Eisen HJ. Epidemiology of heart failure and scope of the problem.
Cardiol Clin. 2014;32:1–8, vii.
5. Klein L, O’Connor CM, Gattis WA, Zampino M, de Luca L, Vitarelli A,
et al. Pharmacologic therapy for patients with chronic heart failure and
reduced systolic function: review of trials and practical considerations.
Am J Cardiol. 2003; 91:18F–40F.
6.Bergmann MW. WNT signaling in adult cardiac hypertrophy and
remodeling: lessons learned from cardiac development. Circ Res.
2010;107:1198–1208.
7.Daskalopoulos EP, Hermans KC, Janssen BJ, Matthijs Blankesteijn W.
Targeting the Wnt/frizzled signaling pathway after myocardial infarction: a new
tool in the therapeutic toolbox? Trends Cardiovasc Med. 2013;23:121–127.
8. van de Schans VA, Smits JF, Blankesteijn WM. The Wnt/frizzled pathway in cardiovascular development and disease: friend or foe? Eur J
Pharmacol. 2008;585:338–345.
9. Gao C, Chen YG. Dishevelled: The hub of Wnt signaling. Cell Signal.
2010;22:717–727.
10. Malekar P, Hagenmueller M, Anyanwu A, Buss S, Streit MR, Weiss CS,
Wolf D, Riffel J, Bauer A, Katus HA, Hardt SE. Wnt signaling is critical
for maladaptive cardiac hypertrophy and accelerates myocardial remodeling. Hypertension. 2010;55:939–945.
11. van de Schans VA, van den Borne SW, Strzelecka AE, Janssen BJ, van der
Velden JL, Langen RC, et al. Interruption of Wnt signaling attenuates the
onset of pressure overload-induced cardiac hypertrophy. Hypertension.
2007; 49:473–480.
12. Means AR. Regulatory cascades involving calmodulin-dependent protein
kinases. Mol Endocrinol. 2000;14:4–13.
13.Tombes RM, Faison MO, Turbeville JM. Organization and evolution
of multifunctional Ca(2+)/CaM-dependent protein kinase genes. Gene.
2003;322:17–31.
14. 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.
15.Bossuyt J, Helmstadter K, Wu X, Clements-Jewery H, Haworth RS,
Avkiran M, Martin JL, Pogwizd SM, Bers DM. Ca2+/calmodulin-dependent protein kinase IIdelta and protein kinase D overexpression reinforce the histone deacetylase 5 redistribution in heart failure. Circ Res.
2008;102:695–702.
16.Sheldahl LC, Slusarski DC, Pandur P, Miller JR, Kühl M, Moon RT.
Dishevelled activates Ca2+ flux, PKC, and CamKII in vertebrate embryos.
J Cell Biol. 2003;161:769–777.
17. Backs J, Backs T, Neef S, Kreusser MM, Lehmann LH, Patrick DM,
Grueter CE, Qi X, Richardson JA, Hill JA, Katus HA, Bassel-Duby R,
Maier LS, Olson EN. The delta isoform of CaM kinase II is required for
pathological cardiac hypertrophy and remodeling after pressure overload.
Proc Natl Acad Sci U S A. 2009;106:2342–2347.
18. Backs J, Stein P, Backs T, Duncan FE, Grueter CE, McAnally J, Qi X,
Schultz RM, Olson EN. The gamma isoform of CaM kinase II controls
mouse egg activation by regulating cell cycle resumption. Proc Natl Acad
Sci U S A. 2010;107:81–86.
19. Kreusser MM, Lehmann LH, Keranov S, Hoting MO, Kohlhaas M, Reil
JC, et al. The cardiac CaMKII genes delta and gamma contribute redundantly to adverse remodeling but inhibit calcineurin-induced myocardial
hypertrophy. Circulation. 2014;130:1262–1273.
20.Davis FJ, Gupta M, Camoretti-Mercado B, Schwartz RJ, Gupta MP.
Calcium/calmodulin-dependent protein kinase activates serum response
factor transcription activity by its dissociation from histone deacetylase,
HDAC4. Implications in cardiac muscle gene regulation during hypertrophy. J Biol Chem. 2003;278:20047–20058.
21. Backs J, Worst BC, Lehmann LH, Patrick DM, Jebessa Z, Kreusser MM,
Sun Q, Chen L, Heft C, Katus HA, Olson EN. Selective repression of
MEF2 activity by PKA-dependent proteolysis of HDAC4. J Cell Biol.
2011;195:403–415.
22. Ishitani T, Kishida S, Hyodo-Miura J, Ueno N, Yasuda J, Waterman M,
Shibuya H, Moon RT, Ninomiya-Tsuji J, Matsumoto K. The TAK1NLK mitogen-activated protein kinase cascade functions in the Wnt-5a/
Ca(2+) pathway to antagonize Wnt/beta-catenin signaling. Mol Cell Biol.
2003;23:131–139.
23.Grigoryan T, Wend P, Klaus A, Birchmeier W. Deciphering the function of canonical Wnt signals in development and disease: conditional
loss- and gain-of-function mutations of beta-catenin in mice. Genes Dev.
2008;22:2308–2341.
24. Rao TP, Kühl M. An updated overview on Wnt signaling pathways: a prelude for more. Circ Res. 2010;106:1798–1806.
25. Ling H, Zhang T, Pereira L, Means CK, Cheng H, Gu Y, Dalton ND,
Peterson KL, Chen J, Bers D, Brown JH, Heller Brown J. Requirement
344 Hypertension February 2015
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
for Ca2+/calmodulin-dependent kinase II in the transition from pressure
overload-induced cardiac hypertrophy to heart failure in mice. J Clin
Invest. 2009;119:1230–1240.
26. Zhang R, Khoo MS, Wu Y, et al. Calmodulin kinase II inhibition protects
against structural heart disease. Nat Med. 2005;11:409–417.
27. Erickson JR, Pereira L, Wang L, Han G, Ferguson A, Dao K, Copeland RJ,
Despa F, Hart GW, Ripplinger CM, Bers DM. Diabetic hyperglycaemia activates
CaMKII and arrhythmias by O-linked glycosylation. Nature. 2013;502:372–376.
28.Hagenmueller M, Riffel JH, Bernhold E, Fan J, Zhang M, Ochs M,
Steinbeisser H, Katus HA, Hardt SE. Dapper-1 induces myocardial
remodeling through activation of canonical Wnt signaling in cardiomyocytes. Hypertension. 2013;61:1177–1183.
29. Zhou J, Qu J,Yi XP, Graber K, Huber L, Wang X, Gerdes AM, Li F. Upregulation
of gamma-catenin compensates for the loss of beta-catenin in adult cardiomyocytes. Am J Physiol Heart Circ Physiol. 2007;292:H270–H276.
30.Baurand A, Zelarayan L, Betney R, Gehrke C, Dunger S, Noack C,
Busjahn A, Huelsken J, Taketo MM, Birchmeier W, Dietz R, Bergmann
MW. Beta-catenin downregulation is required for adaptive cardiac remodeling. Circ Res. 2007;100:1353–1362.
31.Zelarayán LC, Noack C, Sekkali B, Kmecova J, Gehrke C, Renger
A, Zafiriou MP, van der Nagel R, Dietz R, de Windt LJ, Balligand JL,
Bergmann MW. Beta-Catenin downregulation attenuates ischemic cardiac
remodeling through enhanced resident precursor cell differentiation. Proc
Natl Acad Sci U S A. 2008;105:19762–19767.
32. 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.
33. Zhang T, Kohlhaas M, Backs J, Mishra S, Phillips W, Dybkova N, et al.
CaMKIIdelta isoforms differentially affect calcium handling but similarly
regulate HDAC/MEF2 transcriptional responses. J Biol Chem. 2007;
282:35078–35087.
34. Backs J, Olson EN. Control of cardiac growth by histone acetylation/
deacetylation. Circ Res. 2006;98:15–24.
35. Kim Y, Phan D, van Rooij E, Wang DZ, McAnally J, Qi X, Richardson
JA, Hill JA, Bassel-Duby R, Olson EN. The MEF2D transcription factor mediates stress-dependent cardiac remodeling in mice. J Clin Invest.
2008;118:124–132.
36. Colomer JM, Mao L, Rockman HA, Means AR. Pressure overload selectively up-regulates Ca2+/calmodulin-dependent protein kinase II in vivo.
Mol Endocrinol. 2003;17:183–192.
37. Lu J, McKinsey TA, Nicol RL, Olson EN. Signal-dependent activation of
the MEF2 transcription factor by dissociation from histone deacetylases.
Proc Natl Acad Sci U S A. 2000;97:4070–4075.
38.Kühl M, Sheldahl LC, Malbon CC, Moon RT. Ca(2+)/calmodulindependent protein kinase II is stimulated by Wnt and Frizzled homologs and promotes ventral cell fates in Xenopus. J Biol Chem. 2000;275:
12701–12711.
39.De A. Wnt/Ca2+ signaling pathway: a brief overview. Acta Biochim
Biophys Sin (Shanghai). 2011;43:745–756.
Novelty and Significance
What Is New?
• Here, we crossed dishevelled-1–transgenic mice with calcium/calmodu-
lin-dependent protein kinase IIδγ double-knockout mice. Interestingly, the
dishevelled-1–transgenic mice lacking calcium/calmodulin-­
dependent
protein kinase IIδγ displayed a normal cardiac phenotype without cardiac
hypertrophy, fibrosis, or left ventricular dysfunction. Therefore, we report
that dishevelled-1 upregulation activates all 3 branches of the Wnt signaling pathway, with calcium/calmodulin-dependent protein kinase II acting
as the most pivotal component in the process of myocardial remodeling.
Furthermore, mechanistic analyses revealed that calcium/calmodulindependent protein kinase IIδγ couples noncanonical Wnt signaling with
histone deacetylase 4/myosin enhancer factor 2.
What Is Relevant?
• In our previous study, we generated dishevelled-1–transgenic mice with
strong expression of dishevelled-1 and confirmable activation of all the
3 branches of Wnt signaling pathways. Some doubt that the overexpression of dishevelled-1 is too strong to exclude the nonspecific effect in
this model.
Summary
All in all, our study demonstrates for the first time that
­dishevelled–calcium/calmodulin-dependent protein kinase II–histone deacetylase 4–myosin enhancer factor 2 axis might hold a
critical role in cardiac remodeling and its progression to left ventricular ­dysfunction.
Calcium/Calmodulin-Dependent Protein Kinase II Couples Wnt Signaling With Histone
Deacetylase 4 and Mediates Dishevelled-Induced Cardiomyopathy
Min Zhang, Marco Hagenmueller, Johannes H. Riffel, Michael M. Kreusser, Elmar Bernhold,
Jingjing Fan, Hugo A. Katus, Johannes Backs and Stefan E. Hardt
Downloaded from http://hyper.ahajournals.org/ by guest on June 18, 2017
Hypertension. 2015;65:335-344; originally published online December 8, 2014;
doi: 10.1161/HYPERTENSIONAHA.114.04467
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2014 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/65/2/335
Data Supplement (unedited) at:
http://hyper.ahajournals.org/content/suppl/2014/12/08/HYPERTENSIONAHA.114.04467.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Hypertension can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/
Online Supplement
CAMKII COUPLES WNT SIGNALING TO HDAC4 AND MEDIATES DISHEVELLEDINDUCED CARDIOMYOPATHY
Min Zhang1,2, Marco Hagenmueller1,3, Johannes H. Riffel1,3, Michael M. Kreusser1,3,4,
Elmar Bernhold1, Jingjing Fan1, Hugo A. Katus1,3, Johannes Backs 4, and Stefan E.
Hardt1,3,5
1. Department of Cardiology, Angiology and Pulmology, University of Heidelberg, Im
Neuenheimer Feld 410, 69120 Heidelberg, Germany
2. Institute of Cardiology, Union Hospital, Tongji Medical College of Huazhong
University of Science and Technology, Wuhan 430022, China.
3.
DZHK
(German
Center
for
Cardiovascular
Research),
partner
site
Heidelberg/Mannheim, Heidelberg, Germany
4.
Research Unit Cardiac Epigenetics, Department of Cardiology, University of
Heidelberg, 69120 Heidelberg, Germany; DZHK (German Centre for Cardiovascular
Research), partner site Heidelberg, Germany
5. Center for Cardiac and Circulatory Diseases, Hoheneggerstr. 9, 76646 Bruchsal,
Germany
Supplemental Information
Methods
Chemical reagents and adenovirus production.
KN92 and KN93 (1µmol/L) were obtained from Calbiochem. Adenovirus harbouring
myosin enhancer factor 2 (MEF2) luciferase and HDAC4-S246,467,632A (HDAC43S/A) were produced as described previously[1, 2]. Recombinant adenovirus was
generated with the AdMax™ Vector Creation Kit (Microbix) containing full-length cDNAencodingDvl-1-wt (AdDvl). Empty adenovirus was used as control (AdC). After
generation the adenoviruses were amplified, purified with the Adeno-X Purification Kit
(BD) and its infectious units per µl were determined with the Adeno-X Rapid Titer Kit
(BD). An infectious unit of 30 per µl was maintained throughout all experiments as well
as the infection period of 48 hours, unless otherwise noted. Luciferase reporter assay kit
was obtained from Promega.
Animals.
This study was performed in accordance with the Guide for the Care and Use of
Laboratory Animals published by the United States National Institutes of Health (NIH
Publication No. 85-23, revised 1996) and was approved by the local animal ethics
review board (approval number: 40631) and the Government Bureau of Karlsruhe
(project number: 35-9185.81/G-131/06). The generation of heart-directed DVL mice was
described previously[2]. CaMKIIδ-KO, CaMKIIγ knockout (CaMKIIγ-KO) and CaMKIIδγ
double knockout (CaMKIIδγ-KO) animal lines were produced with the protocol in
Supplemental Material. To generate the homozygous CaMKIIδγ-KO mice, heterozygous
CaMKIIδγ-KO offspring carrying the Dvl-1 transgene were inbred with a line lacking the
transgene. The C57BL/6 background DVL mice were crossed with
129SvEv/C57BL6/CD1 background CaMKIIδγ-KO mice, leading to four genotypes: Dvl1 wild-type mice with normal levels of CaMKIIδγ (CTRL), CaMKIIδγ-deficient with nonDvl-1-Tg mice (CKO), Dvl-1-Tg mice with normal CaMKIIδγ levels (DVL), and Dvl-1-Tg
mice deficient for CaMKIIδγ (DWC). DVL mice were also crossed with CaMKIIδ-KO
mice and CaMKIIγ-KO mice using the same mating plan. There is no genetic variability
between the groups. All mice used in the present investigation were 12 weeks old.
Culture and adenoviral infection of NRVMs.
NRVMs were isolated from 1- to 3-day-old BR-Wistar rats (Charles River
Laboratories) and purified using a discontinuous Percoll gradient. We consistently
obtained cell populations containing more than 95% cardiomyocytes using this protocol.
For adenoviral infection or siRNA treatment, NRVMs were infected 24 hours after
changing to serum-free medium, grown for an additional 24 hours, and if necessary,
then stimulated with Wnt5a conditional medium for another 24 hours. Before Wnt5a
treatment, CaMKII inhibitor was applied. Cells were then harvested for western blot,
quantitative real-time PCR, and/or fixed for immunocytochemical staining. For
cardiomyocyte hypertrophy, the myocyte cell surface area was measured as
previously[3]. For the MEF2 reporter assay, adenovirus harboring MEF2 luciferase was
applied to cultured cardiomyocytes 8 hours prior to all the above stimulation. After
treatment, cellular extracts were prepared to determine luciferase expression using a
1
luciferase
reporter assay kit (Promega).
Dvl-1 knockdown and isoproterenol stimulation.
siRNAs were synthesized by MWG (Ebersberg, Germany). For siRNA transfection,
cells were cultured in serum free medium for 40 hours and followed at least 5 hours in
both serum and antibiotic free myomedium or OptiMEM (Invitrogen). Then, they were
incubated for 24 hours at 37°C with siRNA treatment. After this, the medium was
changed to serum-free myomedium for another 24 hours with isoproterenol (50nM)
stimulation. siRNAs used were Dvl-1 5´-AGA UCA CCA UUG CCA AUG C-3´ and nonspecific control siRNA 5´-AGG UAG UGU AAU CGC CUU G-3´. In all siRNA mediated
knockdown experiments myocytes transfected with non-specific siRNA were used as
controls.
Echocardiography.
Mice were lightly anesthetized by isofluorane inhalation (~2% isofluorane) and kept
warm on a heating platform in a supine position. In vivo non-invasive transthoracic
echocardiography was performed in a modified setting as previously described[2]. In
brief, to evaluate cardiac function, a two-dimensional parasternal short axis view and Mmode tracings of the LV were obtained with an ATL 5000 echocardiogram. LV
dimensions were determined on the M-mode tracings and averaged from at least three
consecutive cardiac cycles. LV fractional shortening was calculated as (LVIDd LVIDs)/LVIDd and expressed as a percentage. The observer was unaware of the
genotype of the mice.
Hemodynamic measurement.
Before sacrificing the mice, cardiac performance was detected as described
previously[2]. Briefly, a 1.4F micromanometer-tipped catheter (Millar Instruments) was
inserted retrogradely into the LV through the right carotid artery. Data were recorded
using the software package Chart 5 (AD Instruments). The following indices were
measured with the PVAN 3.5 software (Millar Instruments Inc.) and averaged from 10
sequential beats: heart rate, LV end-systolic volume, LV end-diastolic volume, stroke
volume, peak rates of LV pressure development (dP/dt max) and relaxation (dP/dt min),
LV end-systolic pressure, LV end-diastolic pressure, and the time needed for relaxation
of 50% maximal LV pressure to baseline value (τ), which has been reported to be a
useful indicator of LV diastolic dysfunction.
HDAC4 kinase activity assay.
HDAC4 kinase activity assay was performed in ventricular homogenates as
previously[1, 4]. Briefly, GST-HDAC4 fusion proteins (amino acids 419–670 of HDAC4)
were used as substrates. The GST-HDAC proteins were conjugated to glutathione–
agarose beads. GST-HDAC-bound beads were incubated with ventricular lysates (100
µg) in lysis buffer for 4 hours at 4 °C. Beads were washed one time with the same buffer.
Then beads resuspended in kinase reaction buffer (containing 12.5 µM ATP and 5 µCi
of [32P]ATP and reactions) were allowed to proceed for 30 min at room temperature.
Samples were boiled. Phosphoproteins were resolved by SDS/PAGE, visualized by
autoradiography, and quantified by using a PhosphorImager.
Heart tissue preparation and histopathology.
Mice were euthanized and heart was excised in diastole by injection of saturated
potassium chloride solution. After dissecting LV, part of myocardial samples were snap2
frozen with liquid nitrogen for protein as well as mRNA analysis and the other part was
fixed for 24 hours in 4% formalin dissolved in 0.1 M PBS (pH 7.4), subsequently
embedded in paraffin, and transversely cut into 5 μm sections for further histological
analysis.
Sections were stained with haematoxylin and eosin (H&E) to determine the
myocyte size[5]. To quantify cardiac fibrosis and apoptosis, we stained sections with
picrosirius red (PSR)[6] and tunnel staining, respectively. Images were captured with a
digital camera and analyzed using Image J program (myocyte size and fibrosis) and
Image 2 (apoptosis) in a blind manner.
Western blotting.
Total ventricular extracts were prepared from the LV myocardium and cultured
cardiomyocytes on ice as described previously[2]. Protein concentrations were
measured by the BCA assay. Equal amounts of protein extracts were separated with
SDS-PAGE and transferred to a nitrocellulose membrane (Millipore). Primary antibodies
used were anti Dvl-1 (Santa Cruz ), anti β-catenin (Santa Cruz), anti active β-catenin
(Millipore), anti CaMKII (BD Bioscience), anti active CaMKII (Thr286) (Promega), anti
HDAC4 (Cell Signaling), anti p-HDAC4 (Ser632) (Santa Cruz ), anti PKCα (Cell
Signaling), anti p-PKCα/βΙΙ (Thr638/641) (Cell Signaling), anti PKCδ (Cell Signaling),
anti p-PKCδ (Tyr311) (Cell Signaling), anti JNK1 (Santa Cruz), anti p-JNK (Santa Cruz),
anti p-CREM (Ser133) (Cell Signaling) and anti α-actin (Sigma-Aldrich). Anti-rabbit IgG
and anti-mouse IgG horseradish peroxidase-conjugated antibodies (Santa Cruz) were
used as secondary antibodies. Bands were detected by ECL and quantified by
densitometry using the Image J program.
Quantitative real-time PCR.
Total RNA was extracted from the LVs of the animals and cultured cardiomyocytes
using TRIzol (Invitrogen, Germany). First-strand cDNA was synthesized with the
RevertAid First Strand cDNA Synthesis Kit (Fermentas). Quantitative real-time PCR
was performed as described previously[2]. For rat, HPRT 5´-gtcaagggggacataaaag-3´
and 5´-tgcattgttttaccagtgtcaa-3´, probe #22; ANF 5´-cccgacccagcatgg-3´, and 5´caactgctttctgaaaggggtg-3´, probe #25; Axin-2 5´-gagagtgagcggcagagc-3´ and 5´cggctgactcgttctcct-3´, probe #116. For mouse, HPRT 5´-gtcaagggggacataaaag-3´ and
5´-tgcattgttttaccagtgtcaa-3´, probe #22; ANF 5´-cccgacccagcatgg-3´, and 5´caactgctttctgaaaggggtg-3´, probe #25; βMHC 5´-caaggtcaatactctgaccaagg-3´, and 5´ccatgcgcactttcttctc-3´, probe #78; Axin2 5´-ctgctggtcaggcaggag-3´, and 5´tgccagtttctttggctctt-3´, probe #50; CaMKIIα 5´-gctgccaagattatcaacacc-3´, and 5´cacgctccagcttctggt-3´, probe #106; CaMKIIβ 5´-gccatcctcaccactatgct-3´, and 5´ctccatctgctttcttgttgagt-3´, probe #66; CaMKIIδ 5´- agttcacagggacctgaagc-3´, and 5´cgccttgaacttctatggcta-3´, probe #68; and CaMKIIγ 5´-gtgccatcctcacaaccat-3´, and 5´catctgacttcttgttcaataggc-3´, probe #80.
Statistics.
Data are summarized as mean ± standard error of the mean. Statistical analysis
was performed with the Graph-Pad Prism Software Package Version 5.0 (GraphPad,
Inc.). Differences between groups were compared by one-way ANOVA followed by
Student-Newman-Keuls post hoc testing for the analysis of multiple groups. A value of
P<0.05 in a two-tailed distribution was considered statistically significant unless
otherwise stated.
3
Discussion
Recently, several studies have highlighted the role of CaMKIIδ in heart disease.
Cardiac-specific overexpression of CaMKIIδ induces hypertrophy and dilated
cardiomyopathy with ventricular dysfunction, loss of intracellular Ca2+ homeostasis, and
premature death, as compared to wild-type littermates[7-9]. In contrast, CaMKII
inhibition by pharmacological or genetic approaches reverses heart failure-associated
changes such as arrhythmias, hypertrophy, and dysfunction in animal models of
structural heart disease[10-12]. Interestingly, the use of mice lacking CaMKIIδ，which is
the dominant isoform of CaMKII in the heart significantly attenuates the development of
pressure overload-induced heart failure and improves survival, concomitant with
decreases in cardiac dysfunction, apoptosis, and fibrosis[11, 12].
In an attempt to further explore the nuclear mediator of Wnt/CaMKII in cardiac
remodeling and dysfunction, we first monitored activating transcription factor-1 (ATF1)
and cAMP response element-binding protein (CREM), since they are transcription
factors mediated by Wnt/CaMKII signaling[13], and gene regulation by CREM is an
important mechanism of Wnt-directed mammalian myogenic gene expression[14] and
β1-adrenoceptor-induced cardiac damage in mice[15]. Although these observations
raise the possibility that ATF1 or CREM may be a potential downstream target in Dvl-1
overexpression-induced cardiac remodeling, we did not detect any significant changes
in the activation of ATF1 and CREM in DVL mice (data not shown).
4
References
1.
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.
2.
Malekar P, Hagenmueller M, Anyanwu A, Buss S, Streit MR, Weiss CS, et al. Wnt
signaling is critical for maladaptive cardiac hypertrophy and accelerates myocardial remodeling.
Hypertension. 2010; 55:939-945.
3.
Hagenmueller M, Malekar P, Fieger C, Weiss CS, Buss SJ, Wolf D, et al. Depletion of
mammalian target of rapamycin (mTOR) via siRNA mediated knockdown leads to stabilization
of beta-catenin and elicits distinct features of cardiomyocyte hypertrophy. FEBS Lett. 2010;
584:74-80.
4.
Backs J, Backs T, Neef S, Kreusser MM, Lehmann LH, Patrick DM, et al. The delta
isoform of CaM kinase II is required for pathological cardiac hypertrophy and remodeling after
pressure overload. Proc Natl Acad Sci U S A. 2009; 106:2342-2347.
5.
Malekar P, Hagenmueller M, Anyanwu A, Buss S, Streit MR, Weiss CS, et al. Wnt
signaling is critical for maladaptive cardiac hypertrophy and accelerates myocardial remodeling.
Hypertension. 55:939-945.
6.
Hadi AM, Mouchaers KT, Schalij I, Grunberg K, Meijer GA, Vonk-Noordegraaf A, et al.
Rapid quantification of myocardial fibrosis: a new macro-based automated analysis. Cell Oncol
(Dordr). 2011; 34:343-354.
7.
Zhang T, Maier LS, Dalton ND, Miyamoto S, Ross J, Jr., Bers DM, et al. The deltaC
isoform of CaMKII is activated in cardiac hypertrophy and induces dilated cardiomyopathy and
heart failure. Circ Res. 2003; 92:912-919.
8.
Maier LS, Zhang T, Chen L, DeSantiago J, Brown JH, Bers DM. Transgenic
CaMKIIdeltaC overexpression uniquely alters cardiac myocyte Ca2+ handling: reduced SR
Ca2+ load and activated SR Ca2+ release. Circ Res. 2003; 92:904-911.
9.
Zhang T, Kohlhaas M, Backs J, Mishra S, Phillips W, Dybkova N, et al. CaMKIIdelta
isoforms differentially affect calcium handling but similarly regulate HDAC/MEF2
transcriptional responses. J Biol Chem. 2007; 282:35078-35087.
10.
Zhang R, Khoo MS, Wu Y, Yang Y, Grueter CE, Ni G, et al. Calmodulin kinase II
inhibition protects against structural heart disease. Nat Med. 2005; 11:409-417.
11.
Backs J, Stein P, Backs T, Duncan FE, Grueter CE, McAnally J, et al. The gamma
isoform of CaM kinase II controls mouse egg activation by regulating cell cycle resumption.
Proc Natl Acad Sci U S A. 2009; 107:81-86.
12.
Ling H, Zhang T, Pereira L, Means CK, Cheng H, Gu Y, et al. Requirement for
Ca2+/calmodulin-dependent kinase II in the transition from pressure overload-induced cardiac
hypertrophy to heart failure in mice. J Clin Invest. 2009; 119:1230-1240.
13.
Kuhl M. The WNT/calcium pathway: biochemical mediators, tools and future
requirements. Front Biosci. 2004; 9:967-974.
14.
Chen AE, Ginty DD, Fan CM. Protein kinase A signalling via CREB controls
myogenesis induced by Wnt proteins. Nature. 2005; 433:317-322.
15.
Lewin G, Matus M, Basu A, Frebel K, Rohsbach SP, Safronenko A, et al. Critical role of
transcription factor cyclic AMP response element modulator in beta1-adrenoceptor-mediated
cardiac dysfunction. Circulation. 2009; 119:79-88.
5
Supplemental table
Supplemental table S1: Detailed genotype of each group.
Groups
CTRL
CKO
DVL
DWC
Genotypes
Dvl1-/Cre-δFF-γFF
Dvl1-/Cre+
-δFF-γFF
Dvl1+/Cre-δFF-γFF
Dvl1+/Cre+
-δFF-γFF
CTRL equals control (Dvl-1 wildtype mice with normal levels of CaMKIIδγ), CKO
CaMKII-KO (Dvl-1 wildtype mice without CaMKIIδγ expression), DVL Dvl-1-Tg (Dvl-1-Tg
mice with normal levels of CaMKIIδγ), and DWC (Dvl-1-Tg with CaMKIIδγ-KO). The
third to sixth generation of offspring was used in this study. There is no genetic
variability between the littermates.
6
Supplemental Figures and Figure Legends
Figure S1: Dvl-1 knockdown inhibits ISO-induced activation of CaMKII, HDAC4
and MEF2. (A, B) Western blots and densitometry of increased levels of Dvl-1,
phosphorylated CaMKII, phosphorylated HDAC4 and MEF2 following ISO stimulation,
and all of them were inhibited by Dvl-1 knockdown. ISO, isoproterenol; Dvl-KD, Dvl-1
knockdown; Values are reported as mean ± standard error of the mean. *P < 0.05 vs.
CTRL; #P < 0.05 vs. ISO + KD. All experiments with cardiomyocytes were carried out at
least three times.
7
Figure S2: Dvl-1-induced cardiac hypertrophy and LV dysfunction was not
rescued Deletion of CaMKIIδ alone. (A–D) Demonstrating quantitative analysis of left
ventricular PWTs, EDD, EF and FS. EDD means end-diastolic diameter, PWTs
posterior wall thickness in systole, EF ejection fraction, and FS fraction shortening.
CTRL equals control, CKOδ CaMKIIδ knockout, DVL Dvl-1 transgenic, and DWCδ Dvl-1
transgenic with CaMKIIδ knockout. Values are demonstrated as mean ± SEM. *
denotes P < 0.05 vs. CTRL, and # denotes P < 0.05 vs. DWCδ. n > 4 per group.
8
Figure S3: Dvl-1-induced cardiac hypertrophy and LV dysfunction was also not
rescued Deletion of CaMKIIγ. (A–D) Demonstrating quantitative analysis of left
ventricular PWTs, EDD, EF and FS. EDD means end-diastolic diameter, PWTs
posterior wall thickness in systole, EF ejection fraction, and FS fraction shortening.
CTRL equals control, CKOγ CaMKIIγ knockout, DVL Dvl-1 transgenic, and DWCγ Dvl-1
transgenic with CaMKIIγ knockout. Values are demonstrated as mean ± SEM. *
denotes P < 0.05 vs. CTRL, and # denotes P< 0.05 vs. DWCγ. n > 4 per group.
9
Figure S4: Reduced cardiac apoptosis of DWC mice by deletion of CaMKIIδγ. (A)
As determined by TUNEL assay, frequency of apoptosis in the LV was robustly
enhanced in DVL mice, whereas deletion of CaMKIIδγ significantly reduced the
increase of apoptosis in the DWC mice. (B) Bar graph summarized changes of the
apoptosis frequency in all groups. Values are presented as mean ± standard error of the
mean. *P < 0.05 vs. CTRL; #P < 0.05 vs. DWC. n = 12 per group.
10
Figure S5: The role of Wnt5a on cardiac hypertrophy. (A) Conditional Wnt5a
stimulation resulted in cardiomyocytes hypertrophy. This could be rescued by CaMKII
inhibition. (B) MEF2-dependent luciferase activity was markedly upregulated by
conditional Wnt5a stimulation, and KN93 inhibited the upregulation of luciferase activity.
Values are reported as mean ± standard error of the mean. . *P < 0.05 vs. KN92; #P <
0.05 vs Wnt5a with KN93. All experiments with cardiomyocytes were carried out at least
three times.
11