TRPC channels are necessary mediators of pathologic
cardiac hypertrophy
Xu Wua,1, Petra Edera,1, Baojun Changa, and Jeffery D. Molkentina,b,2
a b
Department of Pediatrics, Cincinnati Children’s Hospital Medical Center, University of Cincinnati and
45229
Edited by Eric N. Olson, University of Texas Southwestern, Dallas, TX, and approved March 12, 2010 (received for review February 23, 2010)
Pathologic hypertrophy of the heart is regulated through membrane-
bound receptors and intracellular signaling pathways that function, in
part, by altering Ca2+ handling and Ca2+-dependent signaling effec-
tors. Transient receptor potential canonical (TRPC) channels are impor-
tant mediators of Ca2+-dependent signal transduction that can sense
stretch or activation of membrane-bound receptors. Here we gener-
ated cardiac-specific transgenic mice that express dominant-negative
(dn) TRPC3, dnTRPC6, or dnTRPC4 toward blocking the activity of the
TRPC3/6/7 or TRPC1/4/5 subfamily of channels in the heart. Remark-
ably, all three dn transgenic strategies attenuated the cardiac hyper-
trophic response following either neuroendocrine agonist infusion or
pressure-overload stimulation. dnTRPC transgenic mice also were par-
tially protected from loss of cardiac functional performance following
long-term pressure-overload stimulation. Importantly, adult myocytes
isolated from hypertrophic WT hearts showed a unique Ca2+ influx
activity under store-depleted conditions that was not observed in
myocytes from hypertrophied dnTRPC3, dnTRPC6, or dnTRPC4 hearts.
Moreover, dnTRPC4 inhibited the activity of the TRPC3/6/7 subfamily
in the heart, suggesting that these two subfamilies function in coordi-
nated complexes. Mechanistically, inhibition of TRPC channels in trans-
genic mice or in cultured neonatal myocytes significantly reduced
activity in the calcineurin–nuclear factor of activated T cells (NFAT), a
known Ca2+-dependent hypertrophy-inducing pathway. Thus, TRPC
channels are necessary mediators of pathologic cardiac hypertrophy,
in part through a calcineurin–NFAT signaling pathway.
calcium | heart | signaling | calcineurin
P athologic cardiac hypertrophy, an enlargement of the adult
heart caused by disease-inducing stimuli, can cause sudden
death and is a leading predictor for the development of heart
failure (1). The growth of individual myocytes is programmed by
neuroendocrine factors that signal through membrane-bound
receptors leading to activation of signal-transduction pathways
and alterations in gene expression (2). Augmentation in intra-
cellular Ca2+ is thought to be critically involved in signaling car-
diac hypertrophy, in part through the Ca2+-activated protein
phosphatase calcineurin, which leads to activation of the tran-
scription factor, nuclear factor of activated T cells (NFAT), which
induces hypertrophic response genes (3). Transient receptor
potential canonical (TRPC) channels are cation-selective influx
channels that can initiate cardiac hypertrophy when overex-
pressed, in part because of increased Ca2+ influx and calcineurin
activation (4–7). Functional TRPC channels are comprised of
homo- or heterotetramers between either TRPC1/4/5 or TRPC3/
6/7 subfamily members, although overexpression of any one sub-
unit alone can produce enhanced currents (8). In general, TRPC3/
6/7 are activated by diacylglycerol (DAG) generated by G protein-
coupled receptors (GPCR)/Gαq/phospholipase C signaling, and
TRPC1/4/5 can be activated by depletion of intracellular Ca2+
stores (store-operated Ca2+ entry) or by stretch (8, 9). Once
activated, these channels induce signal transduction through
cytoplasmic elevations in Ca2+ and Na+ or through refilling of
endoplasmic reticulum Ca2+ stores to ensure prolonged signaling
events (8, 9).
7000–7005 | PNAS | April 13, 2010 | vol. 107 | no. 15
Howard Hughes Medical Institute, Cincinnati, OH
Overexpression of TRPC3 or -6 was reported to induce cardiac
hypertrophy through calcineurin/NFAT signaling in transgenic
(TG) mice (4, 5). More recently, some data have emerged sug-
gesting that TRPC channels are required for cardiac hypertrophy,
because Trpc1−/− mice and use of the TRPC3 inhibitory compound
Pyr3 showed reduced pressure-overload growth (10, 11). How-
ever, it remains unknown whether endogenous TRPC channels
and associated Ca2+ influx are altered in pathologic cardiac
hypertrophy and which subfamily might be required for pathologic
growth. Here we generated TG mice with inhibition of the TRPC3/
6/7 and TRPC1/4/5 subfamilies, which equivalently inhibited
membrane “leak” of Ca2+ in pathologic hypertrophy and inhibited
growth following agonist stimulation and pressure overload.
Results
Myocytes from Hypertrophic Hearts Have Increased Membrane Ca2+
Influx. Depletion of intracellular Ca2+ in most cell types leads to
activation of store-operated Ca2+ influx through defined channel
complexes that include Orai and possibly TRPCs in the plasma
membrane (12). Even if TRPC channels are not bona fide reg-
ulators of store-operated Ca2+ entry, Ca2+ depletion conditions
in conjunction with GPCR stimulation is often used as a surro-
gate for assessing their activity. Here, adult cardiac myocytes
from WT mouse hearts were bathed in Ca2+-free buffer with the
sarcoplasmic reticulum Ca2+ ATPase (SERCA) inhibitor cyclo-
piazonic acid (CPA) and then immediately were switched to
buffer containing 1 mM Ca2+ to monitor intracellular Ca2+ levels
by Indo-1 fluorescence. WT myocytes showed no Ca2+ influx
under these conditions (Fig. 1A and Fig. S1). However, myocytes
isolated from hypertrophic mouse hearts after transverse aortic
constriction (TAC) showed substantial Ca2+ influx (Fig. 1 B and C
and Fig S1). Most of the myocytes (≈80%) showed modest Ca2+
influx (Fig. 1C), but ≈20% of the myocytes showed robust Ca2+
influx activity (Fig. 1B and Fig. S1). Induction of Ca2+ influx
observed in hypertrophied adult myocytes was not inhibited with
the L-type Ca2+ channel inhibitor verapamil or the Na+/Ca2+
exchanger (NCX) inhibitor KB-R7943 (Fig. 1 D and E) but was
completely inhibited with SKF-96265, a known TRPC-channel
inhibitor (Fig.1F).
Generation of Dominant-Negative TRPC3 TG Mice. To determine if
TRPC channels might underlie the observed induction of sar-
colemmal Ca2+ influx in myocytes from hypertrophic hearts, we
generated TG mice expressing dominant-negative (dn) TRPC3
with the α-myosin heavy chain (αMHC) promoter. Two inde-
pendent lines were generated that each showed abundant over-
Author contributions: X.W., P.E., and J.D.M. designed research; X.W., P.E., and B.C. per-
formed research; and J.D.M. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1
X.W. and P.E. contributed equally to this work.
2
To whom correspondence should be addressed. E-mail: jeff.molkentin@cchmc.org.
This article contains supporting information online at www.pnas.org/cgi/content/full/
1001825107/DCSupplemental.
www.pnas.org/cgi/doi/10.1073/pnas.1001825107
 expression in the heart without affecting endogenous TRPC3
Indo-1 ratio (405/485)
A 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA
B 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA

Indo-1 ratio (405/485)

0.9 0.9 levels (Fig. 2A). Importantly, the overexpressed dnTRPC3 pro-
0.8 100% 0.8 20%
0.7 0.7
tein was localized properly to the sarcolemma and the T-tubular
0.6 Sham 0.6 TAC network of adult myocytes from these hearts, coincident with
0.5 0.5 NCX1 (Fig. 2B). To investigate the effectiveness of the
0.4 0.4
0 5 10 15 20 0 5 10 15 20 dnTRPC3 mutant in vivo, we crossed line 6.6 dnTRPC3 TG mice
Time (min) Time (min)
with mice expressing WT TRPC3 that we described previously
C 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA
D 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA (5). Protein levels were not altered in dn TG mice, indicating no
Indo-1 ratio (405/485)

Indo-1 ratio (405/485)

0.9 0.9 +verapamil promoter competition (Fig. 2C). Adult myocytes isolated from
0.8 80% 0.8
0.7 0.7
WT TRPC3 overexpressors showed a robust Ca2+ influx activity
0.6 TAC 0.6 TAC under store-depleted conditions, as described previously (ref. 5
0.5 0.5 and Fig. 2D). However, this TRPC3-dependent Ca2+ influx
0.4 0.4
0 5 10 15 20 0 5 10 15 20 activity was blocked completely by the presence of the dnTRPC3
Time (min) Time (min)
transgene (Fig. 2E). These data also were quantified from mul-
E tiple myocytes to show the extent of Ca2+ entry with TRPC3 and
1.0 0 Ca2+ CPA 1 mM Ca2+ CPA
F 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA
Indo-1 ratio (405/485)

Indo-1 ratio (405/485)

0.9 +KB-R7943 0.9 +SKF-96265 its complete inhibition with the dnTRPC3 transgene (Fig. S1).
0.8 0.8
0.7 0.7
Adult myocytes from hearts of unstressed sham mice showed
0.6 TAC 0.6 TAC no Ca2+ influx activity in more than 60 myocytes examined from
0.5 0.5 WT or dnTRPC3/6 TG mice (Fig. 2F and Fig. S1). More
0.4 0.4
0 5 10 15 20 0 5 10 15 20 importantly, adult myocytes isolated from hearts of dnTRPC3
Time (min) Time (min)
mice subjected to TAC showed a nearly complete loss of Ca2+
Fig. 1. Pressure overload induces sarcolemmal Ca2+ entry in ventricular influx activity (90% of all myocytes showed no activity), although
myocytes. (A) Ca2+ trace for store repletion from an adult cardiac myocyte ≈10% of myocytes showed a minor Ca2+ influx activity (Fig. 2 G
isolated from a WT sham-operated mouse. (B and C) WT mice subjected to and H and Fig. S1). These results suggest that expression of the
TAC stimulation showed robust Ca2+ entry in 20% of isolated cardiac myo- dnTRPC3 transgene in the heart blocks induction of most
cytes and modest Ca2+ entry in the remaining 80% of myocytes. (D–F) aberrant sarcolemmal Ca2+ influx activity caused by pathological
Verapamil and KB-R7943 did not reduce the Ca2+ entry in 20% of myocytes
cardiac hypertrophy.
after TAC stimulation, but SKF-96265 eliminated all Ca2+ entry in all myo-
cytes. CPA was given throughout to inhibit SR reloading of Ca2+. All data
dnTRPC3 TG Mice Have Reduced Pathologic Cardiac Hypertrophy. We
were collected in multiple myocytes from three to six mice.
hypothesized that the TRPC-dependent Ca2+ influx activity
observed in hypertrophic hearts initiated reactive growth sig-
naling. To examine this hypothesis, we subjected adult dnTRPC3

A B
TG .2
6

NCX1 TRPC3 merge

6.
6

WT
TG

dnTRPC3
dnTRPC3
TRPC3
GAPDH

C D E
TG 1.0
Indo-1 ratio (405/485)

0 Ca2+ CPA 1 mM Ca2+ CPA 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA

TG
Indo-1 ratio (405/485)

3 P C3 0.9 0.9
PC TR G
WT TR dn DT 0.8 0.8
dnTRPC3 0.7 0.7 TRPC3 x dnTRPC3 TG
TRPC3 TG
TRPC3 0.6 0.6 +PE
+PE
GAPDH 0.5 0.5
0.4 0.4
0 5 10 15 20 0 5 10 15 20
Time (min) Time (min)

F G H
1.0 0 Ca2+ CPA 1 mM Ca2+ CPA 1.0 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA
Indo-1 ratio (405/485)

Indo-1 ratio (405/485)

Indo-1 ratio (405/485)

0 Ca2+ CPA 1 mM Ca2+ CPA

0.9 0.9 0.9
MEDICAL SCIENCES

10% 90%
0.8 0.8 0.8
0.7 dnTRPC3 TG 0.7 0.7 dnTRPC3 TG
dnTRPC3 TG
0.6 Sham 0.6 0.6 TAC
TAC
0.5 0.5 0.5
0.4 0.4 0.4
0 5 10 15 20 0 5 10 15 20 0 5 10 15 20
Time (min) Time (min) Time (min)

Fig. 2. Overexpression of dnTRPC3 in the hearts of TG mice inhibits TAC-induced Ca2+ entry. (A) Western blots for the dnTRPC3 truncation protein and
endogenous TRPC3 protein from hearts of WT and two dnTRPC3 TG lines. (GAPDH was used as a loading control.) (B) Immunocytochemistry from a dnTRPC3 TG
myocyte reacted with an anti-TRPC3 antibody (green) and NCX1 (red). (C) Western blots for endogenous TRPC3, overexpressed dnTRPC3, and GAPDH from hearts
of WT, TRPC3 TG, dnTRPC3 TG, and double transgenic (DTG) mice. (D) Ca2+ influx tracing in an adult ventricular myocyte isolated from TRPC3 TG mice with PE
addition (50 μM). (E) Ca2+ influx tracing in an adult ventricular myocyte isolated from TRPC3 × dnTRPC3 DTG mice with PE. (F) Ca2+ influx tracing in an adult
ventricular myocyte isolated from dnTRPC3 TG mice subjected to a sham surgical procedure. (G and H) Ca2+ influx tracings in adult ventricular myocytes isolated
from dnTRPC3 TG mice subjected to a TAC surgical procedure to induce hypertrophy. All data were collected in multiple myocytes from three to six mice.

Wu et al. PNAS | April 13, 2010 | vol. 107 | no. 15 | 7001

 A Vehicle
B Sham 2wks ulation with 21 days of swimming (Fig. 3H). These results suggest
6
*
PE/AngII
7 * TAC 2wks that blockade of the TRPC3/6/7 subfamily antagonizes neuro-
HW/BW (mg/g)
# #
*

HW/BW (mg/g)
5 * 6 endocrine-like and pressure-overload–induced pathologic car-
4 5
3 4 diac hypertrophy in vivo, as well as transition to heart failure, but
3
2 2
has no involvement in physiologic hypertrophy.
1 7 7 7 6 1 8 7 8 8
5
0 0
WT dnTRPC3 WT dnTRPC3 dnTRPC6 Transgenic Mice Show Reduced Pathologic Cardiac Hypertrophy.
To substantiate further our conclusion that the TRPC3/6/7 subclass
C 4.5 * Sham 2wks
D 5
* Sham 2wks is necessary for mediating cardiac hypertrophy, we generated two

Rel. MHC mRNA

TAC 2wks
Rel. BNP mRNA

TAC 2wks
4 lines of cardiac-specific dnTRPC6 TG mice (Fig. 4C). These mice
3.0 3
#
* showed a significant reduction in TAC-induced Ca2+ influx across
1.5
*# 2 the sarcolemma in adult myocytes, similar to dnTRPC3 TG mice
1 (Fig. 4 A and B and Fig. S1). The dnTRPC6 protein is not a
0 0 truncation like the dnTRPC3 protein but instead contains muta-
WT dnTRPC3 WT dnTRPC3
tions that disable pore functionality (13). Compared with WT mice,
E 40
Sham 8wks
TAC 8wks
F 10 Sham 8wks
dnTRPC6 TG mice showed complete inhibition of PE/AngII-
* induced cardiac hypertrophy after 2 weeks of infusion (Fig. 4D)
LW/BW (mg/g)

#
30 * * 7.5
TAC 8wks
and a significant inhibition of hypertrophy following 2 weeks of
FS (%)

20 5.0 TAC stimulation (Fig. 4E). Even after 8 weeks of TAC stimulation,
10
8 8
2.5
8 8 8
dnTRPC6 TG mice continued to show less cardiac hypertrophy
9 8 9
0
WT dnTRPC3
0
WT dnTRPC3
than WT mice (Fig. 4F). Interestingly, compared with WT mice,
dnTRPC6 TG mice showed hyperfunctionality at baseline and
were partially protected from a loss of cardiac ventricular per-
G 20
Sham 8wks H 6
*
Rest
Swim * formance after 8 weeks of TAC stimulation (Fig. 4G). Consistent
HW/BW (mg/g)

*
Fibrosis (%)

15 TAC 8wks
10
*#
5
0
WT dnTRPC3
Fig. 3. Overexpression of dnTRPC3 inhibits pathological cardiac hyper-
trophy. (A) Ratio of heart weight to body weight (HW/BW) in WT and
dnTRPC3 TG mice after 2 weeks of PE/Ang II infusion versus vehicle treat-
ment with PBS. *, P < 0.05 vs. vehicle; #, P < 0.05 vs. WT PE/AngII. (B) HW/BW
ratio in WT and line 6.6 dnTRPC3 TG mice after 2 weeks of TAC stimulation.
*, P < 0.05 vs. sham; #, P < 0.05 vs. WT TAC. (C and D) RT-PCR for relative B-
type natriuretic peptide (BNP) and β-myosin heavy chain (βMHC) mRNA
levels from hearts of the indicated groups. *, P < 0.05 vs. sham; #, P < 0.05 vs.
WT TAC. (E) Fractional shortening (FS) by echocardiography in WT and
dnTRPC3 TG mice after 8 weeks of TAC or sham treatment. (F) Lung weight
to body weight (LW/BW) ratio in WT and dnTRPC3 TG mice after 8 weeks of
TAC stimulation. *, P < 0.05 vs. sham. (G) Ventricular fibrosis after 8 weeks of
TAC in the indicated groups, measured from histologically stained sections.
*, P < 0.05 vs. sham; #, P < 0.05 vs. WT TAC. (H) HW/BW ratios in WT and
dnTRPC3 TG mice after 21 days of swimming exercise. *, P < 0.05 vs. sham.
The number of mice analyzed in each group is shown in the bars.
TG mice to coinfusion of phenylephrine (PE) and angiotensin II
(AngII) to model a neuroendocrine-GPCR–stimulated hyper-
trophy response. Remarkably, dnTRPC3 TG mice showed sig-
nificantly less cardiac hypertrophy than WT control mice of the
same strain infused with PE/AngII for 2 weeks (Fig. 3A). To
extend these results, two lines of adult dnTRPC3 TG mice also
were subjected to pressure-overload stimulation by TAC; both
lines showed a significant reduction in cardiac hypertrophy
compared with WT controls over 2 weeks of stimulation (Fig. 3B
and Fig. S2A). During this 2-week time course, neither WT nor
TG groups showed a reduction in fractional shortening from
echocardiograms (Fig. S2B). Moreover, dnTRPC3 TG mice
continued to show less cardiac hypertrophy than WT control
mice after 8 weeks of TAC (Fig. S2C). dnTRPC3 TG mice also
showed less induction of hypertrophic marker gene expression
after TAC than WT mice (Fig. 3 C and D). Consistent with this
reduction in the hypertrophic program, dnTRPC3 TG mice
showed less of a reduction in fractional shortening after 8 weeks
of TAC than WT controls, were protected from lung edema that
characterizes heart failure, and showed less ventricular fibrosis
than WT controls (Fig. 3 E–G). However, dnTRPC3 TG mice
hypertrophied normally following physiologic exercise stim-
5
4 with these results, dnTRPC6 TG mice showed less ventricular
3
2
fibrosis than WT mice after 8 weeks of TAC (Fig. 4H). These
1 results further indicate that the TRPC3/6/7 subclass is necessary for
8 9 8 7
0
WT dnTRPC3
mediating the full extent of pathologic cardiac hypertrophy and
transition to failure.
dnTRPC4 Transgenic Mice Show Reduced Pathologic Cardiac Hypertrophy.
Although both dnTRPC3 and dnTRPC6 attenuated cardiac
hypertrophy, it was uncertain if the TRPC1/4/5 subfamily was
similarly involved in the hypertrophic response. Therefore we
generated dnTRPC4 TG mice with cardiac-specific expression.
The dnTPRC4 truncated protein was overexpressed robustly in the
heart without altering endogenous TRPC4 expression (Fig. 5A).
Immunocytochemistry showed that, similar to dnTRPC3, overex-
pressed dnTRPC4 and NCX1 colocalized to the sarcolemma and
T-tubules in isolated adult myocytes (Fig. 5B). Importantly, the
induction of Ca2+ influx that occurs in adult myocytes from
hypertrophied WT hearts was significantly reduced in dnTRPC4
TG hearts subjected to TAC (Fig. 5C, WT TAC: 0.7 ± 0.14,
dnTRPC4 TAC: 0.58 ± 0.13, P < 0.05). Associated with this
inhibition of Ca2+ influx activity, dnTRPC4 TG mice also showed
less cardiac hypertrophy after 2 weeks of TAC, similar to
dnTRPC3 and dnTRPC6 TG mice (Fig. 5D). dnTRPC4 TG mice
also showed less induction of hypertrophic marker gene expression
and less fibrosis after 6 weeks of TAC than did WT TAC mice (Fig.
S3 A and B). These results suggest that the TRPC1/4/5 subfamily
also is involved in regulating cardiac hypertrophic signaling.
Although TRPC3/6/7 and TRPC1/4/5 subfamily members gen-
erally prefer self-oligomerization, examples of cross-oligomerization
between the subfamilies have been observed (14–17). Thus we
crossed dnTRPC4 mice with WT TRPC3 TG mice to assess Ca2+
influx in isolated myocytes, which showed significant inhibition of
TRPC3-dependent Ca2+ influx under store-depleted conditions
(Fig. 5E). Moreover, immunoprecipitation of TRPC3 protein from
double transgenic (DTG) hearts identified the dnTRPC4 protein,
suggesting that these two channels could coassociate across these two
subfamilies (Fig. 5F). However, this “promiscuity” was not complete,
because crossing the dnTRPC4 and dnTRPC3 transgenes together
resulted in a 100% inhibition of all TAC-induced Ca2+ entry in iso-
lated myocytes (Fig. 5G). As a control, we showed that the dnTRPC4
and dnTRPC3 truncation proteins could immunoprecipitate with
their respective homotypic, full-length counterparts in neonatal
myocytes and even that dnTRPC6 could interact with TRPC3 (Fig.
S4 A–C). Taken together, these results suggest that TRPC channels

7002 | www.pnas.org/cgi/doi/10.1073/pnas.1001825107 Wu et al.

 Indo-1 ratio (405/485)
A B

Indo-1 ratio (405/485)

1.0 0 Ca2+ CPA 1 mM Ca2+ CPA 1.0 0 Ca2+ CPA 1 mM Ca2+ CPA
0.9 0.9
0.8 25% 0.8 75%
0.7 0.7
0.6 TAC 0.6 TAC
0.5 0.5
0.4 0.4
0 5 10 15 20 0 5 10 15 20
Time (min) Time (min)

C D vehicle
6 PE/AngII

HW/BW (mg/g)
*

6
PC

PC
#
5 *

TR

TR
4

dn
WT WT

dn
TRPC6 3
2
GAPDH 1 7
8 7 8
0
WT dnTRPC6

E Sham 2wks F Sham 8wks

TAC 2wks
7 * #
8 * TAC 8wks
*

HW/BW (mg/g)
7 #
HW/BW (mg/g)

6
5 6 *
5
4 4
3 3
2 2
1 8 7 7 8 1 5 5 6 6
0 0
WT dnTRPC6 WT dnTRPC6

G 50 Sham 8wks H 20 Sham 8wks

TAC 8wks * TAC 8wks
Fibrosis (%)

40 15 *
FS (%)

30 *# 10
20 * *#
5
10 5 5
5 6
0 0
WT dnTRPC6 WT dnTRPC6
2+
Fig. 4. dnTRPC6 inhibits pathological cardiac hypertrophy and heart failure. (A and B) Ca influx tracings in adult ventricular myocytes isolated from
dnTRPC6 TG mice subjected to TAC. (C) Western blots of TRPC6 and GAPDH protein in hearts from WT and dnTRPC6 TG mice. (D) HW/BW ratio in WT and
dnTRPC6 TG mice after PE/AngII infusion for 2 weeks. *, P < 0.05 vs. vehicle; #, P < 0.05 vs. WT PE/AngII. (E and F) HW/BW in WT and dnTRPC6 TG mice after
2 and 8 weeks of TAC stimulation. *, P < 0.05 vs. sham; #, P < 0.05 vs. WT TAC. (G) FS in WT and dnTRPC6 TG mice after 8 weeks of TAC stimulation. *, P < 0.05
vs. sham; #, P < 0.05 vs. WT TAC. (H) Ventricular fibrosis after 8 weeks of TAC in the indicated groups, measured from histologically stained sections. *, P < 0.05
vs. sham; #, P < 0.05 vs. WT TAC. The number of mice analyzed in each group is shown in the bars.

can form complexes across the subfamilies in cardiac myocytes so that also was reduced with dnTRPC6 overexpression in PE-stimulated
overexpression of any dnTRPC family member renders many po- myocytes (Fig. 6 D and E). Finally, we also showed that dnTRPC3
tential TRPC tetrameric channel assemblies inactive. and dnTRPC4 overexpression could reduce PE-enhanced NFAT
activity caused by TRPC3 or TRPC4 overexpression in NRVMs
dnTRPC Inhibits Calcineurin–NFAT Signaling. Because calcineurin– (Fig. S5). These results indicate that prohypertrophic GPCR
NFAT has been implicated in mediating TRPC-dependent signaling utilizes a TRPC-dependent Ca2+ signal in activating the
hypertrophy, we evaluated the ability of our dnTPRC proteins to calcineurin-NFAT circuit.
affect this signaling pathway. First, TG mice containing an NFAT-
Discussion
MEDICAL SCIENCES

luciferase reporter transgene (18) were crossed with dnTRPC3

TG mice and subjected to TAC stimulation. NFAT-luciferase TG
mice alone showed a 6-fold induction in NFAT activity in the
heart following 2 weeks of TAC, but this activation was inhibited
by ≈50% in DTG mice that also contained the dnTRPC3 trans-
gene (Fig. 6A). Similar results also were obtained in cultured
neonatal rat ventricular myocytes (NRVM). NRVMs were
infected with an NFAT-luciferase reporter adenovirus and
stimulated with PE, which produced robust NFAT activity that
was reduced by ≈60% with an adenovirus encoding dnTRPC6
(Fig. 6B). Similarly, AdTRPC3- or AdTRPC4-mediated over-
expression significantly enhanced PE-induced NFAT activity in
NRVMs (Fig. 6C). Calcineurin association with calmodulin,
which indicates the fraction of calcineurin in the activated state,
Ca2+-dependent signaling effectors are present in cardiac myo-
cytes, where they influence the cardiac hypertrophic response (2,
3), although, given the excitable nature of this cell type and the
dynamic fluxing of Ca2+ that bathes the entire cytoplasm during
each contractile cycle, specifically how these effectors might be
activated remains a mystery. One hypothesis is that Ca2+-activated
signaling effectors are compartmentalized in membrane micro-
domains in direct proximity or even attached to Ca2+ influx
channels (3). For example, calmodulin-dependent protein kinase
II is regulated, in part, by a perinuclear Ca2+ pool associated with
the inositol triphosphate receptor that controls translocation of
histone deacetylase 5 out of the nucleus, presumably to permit hy-
pertrophic gene expression (19). T-type Ca2+ channels, which are

Wu et al. PNAS | April 13, 2010 | vol. 107 | no. 15 | 7003

 A B A B
dnTRPC4 TG

TG
NFAT-Luc TG
NRVM - AdNFAT-luc
NCX1 TRPC4

4
*

Rel. NFAT activity

Sham 2wks
7 5 *

Rel. NFAT activity

PC
6 TAC 2wks Vehicle

TR
4
5 PE

dn
WT #
TRPC4
4 * 3
* #
3 2
dnTRPC4 2
1
1 5 6 6 6
4 5 5 6
GAPDH 0 0
Ad gal
Luc-TG Luc-TG x Ad-dnTRPC6
dnTRPC3 TG
C D 25
12 Sham C
Indo-1 ratio (405/485)

1.0 0 Ca2+ CPA 1 mM Ca2+ CPA Vehicle

Rel. NFAT activity

*

HW/TL (mg/mm)
TAC # PE
0.9 10
* 20

0.8 8 15 *#
0.7 dnTRPC4 TG 6 10 *#
4
*
0.6 TAC 5
0.5 2 0 6 6 6 6 6 6
5 9 6 7 Ad gal AdTRPC3 AdTRPC4
0.4 0
0 5 10 15 20 WT dnTRPC4
Time (min)
D E
*

CnA:CaM pull down

0.9 Vehicle
E F control dnTRPC6 PE #
1.0 0 Ca CPA 1 mM Ca CPA *
Indo-1 ratio (405/485)

2+ 2+ IP: CaM - PE - PE 0.6

0.9 IP CnB
0.3
0.8 C3 IgG input
TRPC3 CaM 4 4 4 4
0.7 dnC4 x C3 TGs 0
Ad gal Ad-dnTRPC6
0.6 +PE dnTRPC4
0.5 Fig. 6. dnTRPCs attenuate calcineurin–NFAT signaling in cardiac myocytes.
0.4 (A) NFAT-luciferase activity from hearts of NFAT-Luc single TG mice versus
0 5 10 15 20 NFAT-Luc x dnTRPC3 DTG mice subjected to TAC. *, P < 0.05 vs. sham; #, P <
Time (min)
0.05 vs. Luc-TG TAC. Number of mice analyzed is shown in the bars of the
graph. (B) NFAT luciferase activity in Adßgal and Ad-dnTRPC6 coinfected
Indo-1 ratio (405/485)

G 1.0
0.9
0 Ca2+ CPA 1mM Ca2+ CPA NRVM with or without PE treatment for 48 h. *, P < 0.05 vs. vehicle; #, P <
0.05 vs. Adßgal PE. Number of plates of myocytes used to sum the results is
0.8 shown in the bars of the graph. (C) NFAT luciferase activity in NRVM with or
0.7 dnC4 x dnC3 TGs without PE, infected with the indicated viruses. *, P < 0.05 vs. vehicle; #, P <
0.6 TAC 0.05 vs. Adßgal PE. (D and E) Western blots (D) and quantitation after
0.5 immunoprecipitation of calmodulin (CaM) (E) to pull down calcineurin B
0.4 (CnB) or calcineurin A (CnA) from NRVMs infected with control or Ad-
0 5 10 15
Time (min) dnTRPC6, with or without PE treatment for 48 h.

Fig. 5. dnTRPC4 inhibits pathologic cardiac hypertrophy in TG mice. (A)

Western blot for endogenous TRPC4 and the dn deletion mutant of TRPC4
For example, the TRPC3 chemical inhibitor Pyr3 administered to
from WT and dnTRPC4 TG mouse heart protein extracts. (GAPDH was used as a
loading control.) (B) Immunocytochemistry of a myocyte from a dnTRPC4 TG
mice at 0.1 mg/kg/d was shown to attenuate the cardiac hypertrophic
heart reacted with an anti-TRPC4 (green) or NCX1 (red) antibody. (C) Ca2+ response following pressure-overload stimulation (11). Moreover,
influx tracing in an adult ventricular myocyte isolated from a dnTRPC4 TG Trpc1−/− mice were shown to develop less cardiac hypertrophy in
mouse subjected to TAC. (D) Heart weight to tibia length (HW/TL) ratios in WT response to 8 weeks of pressure-overload stimulation or 4 weeks of
and dnTRPC4 TG mice after 2 weeks of TAC stimulation. *, P < 0.05 vs. sham; AngII infusion (10). Although our study is in agreement, we iden-
#
, P < 0.05 vs. WT TAC. The number of animals examined is shown in the bars of tified several concepts that further establish an overarching Ca2+
the graph. (E) Ca2+ influx tracing in an adult ventricular myocyte treated with regulatory paradigm in the heart. First, we observed that the sar-
PE isolated from TRPC3 × dnTRPC4 DTG mice. (F) Western blot for TRPC3 and
colemma from hypertrophic cardiac myocytes is altered profoundly
dnTRPC4 after immunoprecipitation of TRPC3 (IgG was used as a control) from
TRPC3 × dnTRPC4 DTG mouse hearts. (G) Ca2+ influx tracing in an adult ven-
and effectively is “leaky” to Ca2+ through non–L-type or non–T-type
tricular myocyte from dnTRPC3 × dnTRPC4 DTG mice after TAC stimulation. channels (non-NCX1). Second, this enhanced Ca2+ influx profile
observed under store-depleted conditions was inhibited with either
dnTRPC3, dnTRPC6, or dnTRPC4 overexpression in vivo. Third,
induced in hypertrophic hearts, provide a local Ca2+ signal to NOS3 the TRPC3/6/7 and TRPC1/4/5 subclasses appeared to function
to generate an antihypertrophic and protective effect through interdependently in the heart to mediate pathologic hypertrophy.
cGMP-dependent protein kinase type I (20). Similarly, induction of The observation that both dnTRPC4 and dnTRPC3/6 could
TRPC channel activity during cardiac hypertrophy is hypothesized inhibit pathologic cardiac hypertrophy equally when overex-
to generate a distinct Ca2+ signaling microdomain that can impact pressed in the mouse heart was unexpected, because they are from
calcineurin signaling and the hypertrophic response directly (21). different functional and oligomerization subfamilies (8, 9).
Simple overexpression of TRPC3 or TRPC6 in the mouse heart However, TRPC1 and TRPC3 were reported to coassemble in
was sufficient to induce Ca2+ entry and enhance the cardiac generating a DAG-sensitive Ca2+ channel in HEK293 cells (14).
hypertrophic response (4, 5). Mechanistically, we and others have In the brain, TRPC1/4/5 also were identified in complexes with
shown that TRPC channels engage calcineurin–NFAT signaling in TRPC3 and TRPC6, and a dnTRPC5 mutant was capable of
the heart, a well-known prohypertrophic pathway that is both quenching TRPC3 currents when TRPC1 was co-overexpressed,
necessary and sufficient for growth (4–7). Indeed, deletion of but not TRPC3 alone (15). Moreover, TRPC3 and TRPC4 were
calcineurin Aβ reduced hypertrophic inducibility by TRPC3 shown to coassemble in forming a redox-sensitive cation channel
overexpression in the heart (5). Although TRPC channels can in endothelial cells and HEK293 cells by FRET analysis, immu-
induce pathologic cardiac growth when overexpressed, the noprecipitation, and direct measurement of channel current (22).
necessity of these channels was not investigated until very recently. More provocatively, TRPC channels probably coexist with TRPM

7004 | www.pnas.org/cgi/doi/10.1073/pnas.1001825107 Wu et al.

and TRPP subclasses of channels in even larger channel com-
plexes (16, 17). Because all TRPC family members have been
detected in the heart, some of which are up-regulated by hyper-
trophy (4–7, 23), it is likely that many combinations of tetramers
are possible and can generate unique cation influx properties.
However, combined inhibition of both subfamilies in dnTRPC4,
dnTRPC3 DTG mice completely eliminated all TAC-associated
Ca2+ entry, suggesting that not all TRPC complexes show cross-
oligomerization in the heart.
The cardiac myocyte is perhaps one of the few cell types in the
body that does not require store-operated Ca2+ entry to reload the
endoplasmic reticulum/sarcoplasmic reticulum (ER/SR) com-
partment. Indeed, Ca2+ loading of the ER/SR occurs during each
contractile cycle and can be explained fully by the action and
equilibrium between the L-type Ca2+ channel, the NCX, the rya-
nodine receptor, and SERCA2 (24). Thus, the relatively high
prevalence of TRPC channel expression in cardiac myocytes
probably mediates other modes of Ca2+ entry for putative sig-
naling functions. The results of our study, in conjunction with the
evolving literature in this area, suggest that pharmacologic inhib-
itors of TRPC channels might be a strategy for attenuating local
Ca2+ signals involved in pathologic cardiac hypertrophy or failure.
Methods
Myocytes Isolation and Ca2+ Measurements. Adult myocytes from WT, dnTRPC,
and TRPC3 TG mice were isolated as previously described (25). Myocytes were
loaded with Indo-1 AM (10 μM) for 12–45 min in normal Tyrode solution (in
mmol/L: CaCl2 1, NaCl 140, KCl 4, MgCl2 1, Hepes 5, glucose 10; pH 7.4) or MEM
at room temperature. Cells were bathed in Ca2+-free normal Tyrode solution in
the presence of 10 μM CPA for 30 min to deplete intracellular Ca2+ stores. The
solution then was switched to 1 mM Ca2+ normal Tyrode solution with CPA to
evoke store-operated Ca2+ entry at baseline or in the presence of PE (50 μM).
Other inhibitors included verapamil (10 μM), KB-R7943 (5 μM), and SKF-96365
(5 μM). NRVMs for NFAT activity assays were isolated and cultured as described
previously (18).
1. Levy D, Garrison RJ, Savage DD, Kannel WB, Castelli WP (1990) Prognostic implications
of echocardiographically determined left ventricular mass in the Framingham Heart
Study. N Engl J Med 322:1561–1566.
2. Heineke J, Molkentin JD (2006) Regulation of cardiac hypertrophy by intracellular
signalling pathways. Nat Rev Mol Cell Biol 7:589–600.
3. Houser SR, Molkentin JD (2008) Does contractile Ca2+ control calcineurin-NFAT
signaling and pathological hypertrophy in cardiac myocytes? Sci Signal 1 (25):pe31.
4. Bush EW, et al. (2006) Canonical transient receptor potential channels promote
cardiomyocyte hypertrophy through activation of calcineurin signaling. J Biol Chem
281:33487–33496.
5. Nakayama H, Wilkin BJ, Bodi I, Molkentin JD (2006) Calcineurin-dependent cardiomyop-
athy is activated by TRPC in the adult mouse heart. FASEB J 20:1660–1670.
6. Kuwahara K, et al. (2006) TRPC6 fulfills a calcineurin signaling circuit during
pathologic cardiac remodeling. J Clin Invest 116:3114–3126.
7. Onohara N, et al. (2006) TRPC3 and TRPC6 are essential for angiotensin II-induced
cardiac hypertrophy. EMBO J 25:5305–5316.
8. Nilius B, Owsianik G, Voets T, Peters JA (2007) Transient receptor potential cation
channels in disease. Physiol Rev 87:165–217.
9. Abramowitz J, Birnbaumer L (2009) Physiology and pathophysiology of canonical
transient receptor potential channels. FASEB J 23:297–328.
10. Seth M, et al. (2009) TRPC1 channels are critical for hypertrophic signaling in the
heart. Circ Res 10:1023–1030.
11. Kiyonaka S, et al. (2009) Selective and direct inhibition of TRPC3 channels underlies
biological activities of a pyrazole compound. Proc Natl Acad Sci USA 106:5400–5405.
12. Wang Y, Deng X, Hewavitharana T, Soboloff J, Gill DL (2008) Stim, ORAI and TRPC
channels in the control of calcium entry signals in smooth muscle. Clin Exp Pharmacol
Physiol 35:1127–1133.
13. Hofmann T, Schaefer M, Schultz G, Gudermann T (2002) Subunit composition of
mammalian transient receptor potential channels in living cells. Proc Natl Acad Sci
USA 99:7461–7466.
14. Lintschinger B, et al. (2000) Coassembly of Trp1 and Trp3 proteins generates
diacylglycerol- and Ca2+-sensitive cation channels. J Biol Chem 275:27799–27805.
15. Strübing C, Krapivinsky G, Krapivinsky L, Clapham DE (2003) Formation of novel TRPC
channels by complex subunit interactions in embryonic brain. J Biol Chem 278:
39014–39019.
Wu et al.
Animal Models and Procedures. Animal procedures were approved by the
Cincinnati Children’s Hospital Institutional Animal Care and Use Committee.
NFAT-luciferase and heart-specific TRPC3 TG mice were described previously
(5, 18). TG mice were generated with the αMHC promoter to drive expres-
sion of dnTRPC3 (26), dnTRPC4 (27), and dnTRPC6 (13).
Echocardiography and Mouse Procedures. Echocardiography under 2% iso-
flurane has been described previously (28). Pathologic hypertrophy was
induced by TAC as described previously (18). Doppler echocardiography was
used to confirm equal pressure gradients across the aortic constrictions in all
TAC procedures. Alzet miniosmotic pumps (model 1002; Durect Corp.) con-
taining a mixture of PE (100 mg/kg/d) and AngII (432 μg/kg/d), or PBS (vehicle
control) were surgically inserted s.c. dorsally in mice under isoflurane anes-
thesia. Swimming exercise to induce hypertrophy has been described pre-
viously (18).
Western Blotting, Immunoprecipitation, and Immunocytochemistry. Western
blotting and immunoprecipitation were done as described previously (28). The
antibodies used in this study were anti-TRPC3 polyclonal antibody (Abcam),
anti-TRPC6 and anti-TRPC3 polyclonal antibodies (Alomone Labs), anti-TRPC4
(gift from M. Zhu, Center of Molecular Neurobiology, Ohio State University),
anti-GAPDH antibody (Fitzgerald), anti-calcineurin A and B (Sigma-Aldrich), and
anti-calmodulin (Zymed Laboratories). Immunocytochemistry was performed as
described previously, and primary antibodies were used at 1:100 (19).
Statistics. All data are expressed as mean ± SEM. Differences between
experimental groups were evaluated for statistical significance using the
Student’s t test for unpaired data or for multiple groups using both two-
tailed t test and two-way ANOVA. The data were distributed normally in all
cases. P < 0.05 was considered statistically significant.
ACKNOWLEDGMENTS. This work was supported by grants from the National
Institutes of Health, the Fondation Leducq, and the Howard Hughes Medical
Institute (J.D.M). X.W. was supported by a postdoctoral fellowship from the
American Heart Association, and P.E. was supported by the Austrian Science
Fund Award J 2775-B12.
16. Park JY, et al. (2008) TRPM4b channel suppresses store-operated Ca2+ entry by a
novel protein-protein interaction with the TRPC3 channel. Biochem Biophys Res
Commun 368:677–683.
17. Zhang P, et al. (2009) The multimeric structure of polycystin-2 (TRPP2): structural-
functional correlates of homo- and hetero-multimers with TRPC1. Hum Mol Genet 18:
1238–1251.
18. Wilkins BJ, et al. (2004) Calcineurin/NFAT coupling participates in pathological, but
not physiological, cardiac hypertrophy. Circ Res 94:110–118.
19. Wu X, et al. (2006) Local InsP3-dependent perinuclear Ca2+ signaling in cardiac
myocyte excitation-transcription coupling. J Clin Invest 116:675–682.
20. Nakayama H, et al. (2009) α1G T-type Ca2+ current antagonizes cardiac hypertrophy
through a NOS3-dependent mechanism. J Clin Invest 119:3787–3796.
21. Watanabe H, Murakami M, Ohba T, Ono K, Ito H (2009) The pathological role of
transient receptor potential channels in heart disease. Circ J 73:419–427.
22. Poteser M, et al. (2006) TRPC3 and TRPC4 associate to form a redox-sensitive cation
channel. Evidence for expression of native TRPC3-TRPC4 heteromeric channels in
endothelial cells. J Biol Chem 281:13588–13595.
23. Ohba T, et al. (2007) Upregulation of TRPC1 in the development of cardiac
hypertrophy. J Mol Cell Cardiol 42:498–507.
24. Bers DM (2008) Calcium cycling and signaling in cardiac myocytes. Annu Rev Physiol
70:23–49.
25. Wu X, et al. (2009) Plasma membrane Ca2+-ATPase isoform 4 antagonizes cardiac
MEDICAL SCIENCES
hypertrophy in association with calcineurin inhibition in rodents. J Clin Invest 119:
976–985.
26. Balzer M, Lintschinger B, Groschner K (1999) Evidence for a role of Trp proteins in the
oxidative stress-induced membrane conductances of porcine aortic endothelial cells.
Cardiovasc Res 42:543–549.
27. Schindl R, et al. (2008) The first ankyrin-like repeat is the minimum indispensable key
structure for functional assembly of homo- and heteromeric TRPC4/TRPC5 channels.
Cell Calcium 43:260–269.
28. Kaiser RA, et al. (2005) Genetic inhibition or activation of JNK1/2 protects the
myocardium from ischemia-reperfusion-induced cell death in vivo. J Biol Chem 280:
32602–32608.
PNAS | April 13, 2010 | vol. 107 | no. 15 | 7005