Rationally engineered Troponin C modulates in

Full text

Turn on search term navigation

ARTICLE

Received 12 Jun 2015 | Accepted 21 Jan 2016 | Published 24 Feb 2016

DOI: 10.1038/ncomms10794 OPEN

Rationally engineered Troponin C modulates in vivo cardiac function and performance in health and disease

Vikram Shettigar1,*, Bo Zhang1,*,w, Sean C. Little2,*, Hussam E. Salhi1, Brian J. Hansen1, Ning Li1, Jianchao Zhang1, Steve R. Roof3, Hsiang-Ting Ho1, Lucia Brunello1, Jessica K. Lerch4, Noah Weisleder1, Vadim V. Fedorov1, Federica Accornero1, Jill A. Rafael-Fortney1, Sandor Gyorke1, Paul M.L. Janssen1, Brandon J. Biesiadecki1,Mark T. Ziolo1,* & Jonathan P. Davis1,*

Promisingly, our smartly formulated Ca2 -sensitizing TnC (L48Q) enhances heart function without any adverse effects that are commonly observed with positive inotropes. In a myocardial infarction (MI) model of heart failure, expression of TnC L48Q before the MI preserves cardiac function and performance. Moreover, expression of TnC L48Q after the MI therapeutically enhances cardiac function and performance, without compromising survival. We demonstrate engineering TnC can specically and precisely modulate cardiac contractility that when combined with gene therapy can be employed as a therapeutic strategy for heart disease.

1 Davis Heart and Lung Research Institute and Department of Physiology and Cell Biology, Columbus, Ohio 43210, USA. 2 Bristol-Myers Squibb, Department of Discovery Biology, Wallingford, Connecticut 06492, USA. 3 Q-Test Labs, Columbus, Ohio 43235, USA. 4 Center for Brain and Spinal Cord Repair, Department of Neuroscience, The Ohio State University College of Medicine, Columbus, Ohio 43210, USA. * These authors contributed equally to this work.

w Present address: Institute of Organ Transplantation, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China. Correspondence and requests for materials should be addressed to M.T.Z. (email: mailto:[email protected]

Web End [email protected] ) or to J.P.D. (email: mailto:[email protected]

Web End [email protected] ).

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794

The amount of Ca2 bound to Troponin C (TnC), the Ca2 -dependent switch in the heart, is a primary determinant of the strength of cardiac contraction1,2.

Hence, modulating contraction through TnC has been an enticing pharmaceutical goal for many decades. Unfortunately, there are no specic therapeutics that modulate TnC without affecting other systems (for example, phosphodiesterases and hypotension)3,4. A promising strategy to solely modulate the Ca2 sensitivity of TnC is through rationally customizing Ca2 binding to TnC (refs 5,6). By applying the principles that govern

Ca2 binding to TnC, we have smartly formulated TnC variants with a wide range of Ca2 sensitivities5,6. These variants effectively modulate Ca2 sensitivity and force development from in silico to in vitro5,711. Whether TnC engineering can translate in vivo to modulate cardiac function with molecular precision is currently unknown.

There is a dire need to be able to modulate cardiac contraction to better manage cardiovascular disease. For example, a majority of cardiomyopathies present with impaired contraction12. Many strategies to improve cardiac contraction (positive inotropes) do so by increasing the signal for contraction (that is, intracellular Ca2 )13. Owing to the universal nature of Ca2 signalling, increasing Ca2 not only enhances contraction, but also leads to a variety of negative side effects (arrhythmias and cell death) that ultimately increases mortality14. Other inotropic strategies that enhance the sensitivity of the contractile apparatus to Ca2 results in compromised cardiac relaxation and can also raise intracellular Ca2 through off-target interactions4. Thus, there is a great need for a positive inotrope that does not have these confounding issues.

We show that our rationally customized TnC can improve function and performance in a common heart pathology myocardial infarction (MI). We nd that our Ca2 -sensitizing

TnC L48Q expressed before or after an MI, enhances heart function and performance without any adverse effects (that is, slow relaxation, arrhythmias or changing Ca2 ). Thus, smartly formulating TnC combined with gene therapy is a powerful tool to develop unique and novel therapies with molecular precision to combat heart disease.

ResultsEngineered TnCs modify cardiac contraction. We and others have previously shown in vitro that the Ca2 sensitivity of force development can be modulated through our rationally engineered

TnCs (refs 5,7,9,10,15). Excitingly Feest et al.7 recently demonstrated the concept that TnC can enhance contraction in cultured myocytes. To test the efcacy of engineered TnC on in vivo cardiac contractility, we delivered either TnC D73N (Ca2 desensitizer) or TnC L48Q (Ca2 sensitizer) into murine myocardium through recombinant adeno-associated virus serotype 9 (rAAV9). TnCs were selected based on their ability to sensitize or desensitize the contractile apparatus to Ca2 in a variety of biochemical and physiological systems (Fig. 1a)5,7,9. The rAAV9 expression cassette contained a cytomegalovirus (CMV) promoter driving the expression of a C-terminally FLAG-tagged TnC variant (Fig. 1b). To track transduction, an internal ribosome entry site-mCherry sequence was added to the vector. We investigated the transduction efciency of our rAAV9 utilizing several different approaches. At the level of the whole heart, we observed mCherry uorescence throughout the heart (Fig. 1c and Supplementary Fig. 1) and retrieved 228 million vector genomes. Throughout the ventricle, we also observed robust histological mCherry uorescence in neonatal mice that were injected intraperitoneally (IP) or adult mice injected via the thoracic

cavity (to restrict expression to the heart; Fig. 1d and Supplementary Fig. 2). Consistent with these ndings, we found expression of Flag-tagged TnC throughout the ventricles (Fig. 1e and Supplementary Fig. 3). Using high-content microscopy to analyse over 3,000 myocytes, we quantied that 703 percent of these myocytes were positive for mCherry uorescence (Fig. 1f,g). This robust transduction of myocytes does not necessarily mean expression or functional incorporation of TnC. Hence, Fig. 1h demonstrates that the transduced TnCs expressed and incorporated into the sarcomere. At this level of myocyte transduction, using quantitative western blot analysis we determined on average 15 percent of the endogenous TnC was replaced by our virally expressed TnCs (Fig. 1i and Supplementary Fig. 4). We also investigated the transduction of other resident cell types in the heart. We observed no transduction of cardiac broblasts16 and negligible transduction of endothelial and vascular smooth muscle cells (Supplementary Fig. 5a,b). Thus we predict that any functional effects would be due to the transduction of myocytes rather than other cell types.

To determine the cellular inuence of modifying TnC Ca2 sensitivity, we isolated ventricular myocytes from the rAAV9 injected mice. Myocytes isolated from the transduced hearts displayed characteristics consistent with altered Ca2 sensitivity.

TnC D73N signicantly decreased, whereas TnC L48Q increased myocyte shortening compared with TnC wild-type (WT) injected or uninjected (control) myocytes (Fig. 2a, Table 1). The altered shortening was not caused by changes in Ca2 handling (Fig. 2a,

Table 1). TnC L48Q had little effect on the rate of myocyte re-lengthening, whereas TnC D73N signicantly accelerated re-lengthening (Table 1). There was little effect of D73N or L48Q TnC on contractile kinetics (that is, time to peak shortening; Table 1). Thus, under basal conditions, the engineered TnCs were capable of altering contraction amplitude without affecting the Ca2 signal, similar to the results of Feest et al.7. Following isoproterenol superfusion, TnC L48Q myocytes demonstrated a normal contractile beta-adrenergic response, while TnC D73N myocytes showed a blunted response (Table 1). Myocytes isolated from mice transduced with WT TnC did not show any difference in myocyte contraction (Table 1) or heart function and structure 12 weeks after injection (Fig. 2bd).

The depressed myocyte contraction and blunted beta-adrenergic response caused by TnC D73N is indicative of heart disease. In fact, TnC D73N hearts were substantially larger than any of the other injected and uninjected hearts with an abnormal myocyte length-to-width ratio (Fig. 2eg). Additionally, both the systolic and diastolic dimensions of the left ventricle were enlarged (Fig. 2hj) and displayed a reduced ejection fraction (EF; Fig. 2k). TnC D73N mice also exhibited abnormal electrical activity (Fig. 2l). Ultimately, after 8 weeks post injection, 65% of the TnC D73N injected animals died (Fig. 2m). Consistent with transgenic modications that decrease Ca2 sensitivity of cardiac muscle17,18, virally transduced TnC D73N recapitulates a dilated cardiomyopathy phenotype. In contrast, TnC L48Q, TnC WT, GFP or mCherry injected mice did not show any adverse alteration in morphology, brosis, survival (Fig. 2f,g,m and Supplementary Fig. 6) or cardiac function/electrical activities (see below). Consistent with previous studies1821, functional effects can be accomplished with relatively moderate levels of myolament protein exchange (Fig. 1i). These ndings establish a critical principle that engineering TnCs Ca2 sensitivity can modulate in vivo cardiac function.

TnC L48Q is a promising positive inotrope. In isolated myocytes7 (Table 1), TnC L48Q acts as a positive inotrope that does not alter cellular Ca2 or the functional response to beta-

2 NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794 ARTICLE

a b

CMV hGI TnC

Flag IRES mCherry WPRE

% IANBD

fluorescence

100

80 60 40 20

0 9 8 7 6 5 4 pCa

Poly

D73N WT

L48Q

Control

Transduced

Anterior Left lateral Right lateral Endocardial

Posterior

Brightfield

Hoechst

mCherry

Aorta

RA RA

LV LV

Control Transduced mCherry Flag Flag

Heart photo Heart photo Fluorescence Fluorescence

IVS

Aorta

1,500

RA RA RA

RA RA

a.u. a.u.

LV LV LV RV RV RV LV

IVS

LA LA

LV LV

RV RV

LA LA LA

Control

IP Thoracic

WT L48Q D73N

% mCherry

positive myocytes

% Exchange

Control

WT L48Q D73N

Figure 1 | Quantication of rAAV9 transduction efciency and TnC replacement. (a) Summary data of Ca2 sensitivity of rabbit cardiac ventricular myobrils exchanged with engineered TnC variants. (b) Schematic of the gene construct for the expression of the engineered TnC in vivo through rAAV9. HbGI: human beta globin intron; IRES: internal ribosome entry site; WPRE: woodchuck hepatitis virus post-transcriptional response element.

(c) Representative photos (scale bar, 3 mm) and mCherry uorescence from uninjected (n 3) and rAAV9 transduced (n 6) hearts. IVS, interventricular

septum; LA, left atria; LV, left ventricle; RA, right atria; RV, right ventricle. (d) Representative images comparing the rAAV9 transduction of IP versus thoracic cavity injections via histological mCherry uorescence (scale bar, 1,000 mm). (e) Representative images comparing the TnC-FLAG expression via histological anti-FLAG immunouorescence (scale bar, 1,000 mm). (f) Representative images for (g) the quantication of the per cent of cardiac myocytes transduced by rAAV9 via high-content microscopy. n 2 for control hearts and n 3 for rAAV9 transduced hearts (scale bar, 100 mm). (h) Representative

immunouorescence images for sarcomeric incorporation of ag-tagged TnC and (i) quantication of the per cent TnC exchanged post transduction (samples were pooled and averaged from all the non-diseased and diseased models). Error bars are s.e.m.

adrenergic stimulation, major problems of current inotropes4,14. To further characterize the potential of TnC L48Q, we performed left ventricular pressurevolume relationship analysis to measure

cardiac function in vivo22. Consistent with the myocyte data, TnC L48Q also signicantly enhanced end systolic elastance (Ees) and preload recruitable stroke work (PRSW; Fig. 3ac), which are

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794

a b

%RCL 2

Ejection

fraction %

250 ms

F/F0

0.5

250 ms

10 0 Control WT

Systolic diameter (mm)

Diastolic

diameter (mm)

3 2 1 0

0 Control

WT Control

10 week

control

4 week

D73N

f g

Heat weight/Tibia

length (mg mm1)

Length / width

ratio

* *

12.5

10.0

7.5

5.0

2.5

0.0

Control

D73N

L48Q

0 Control

D73N

L48Q

Control

D73N

i j

Systolic diameter (mm)

2.01.51.00.50.0

2.5 4

Diastolic

diameter (mm)

Control WT D73N

0 Control WT D73N

100

0 GFP/WT/ L48Q

k l m

Dead

Live73/75 19/53

Ejection fraction

(%)

Survival (%)

D73N

Figure 2 | TnC D73N recapitulates a dilated cardiomyopathy. (a) Representative traces of isolated cardiomyocyte contraction and Ca2 transients. (b) Summary data of TnC WT and control mice EF measured 12 weeks after injection (n 4 for control, 5 for TnC WT). Summary data of TnC WT and

control mice left ventricular systolic (c) and diastolic (d) dimensions measured via echocardiography 12 weeks after injection (n 4 for control, 5 for

TnC WT ). (e) Comparison of a 10-week heart from a control mouse and a 4-week heart from a TnC D73N mouse. (f) Summary data of heart weight to tibia length (hypertrophy) of rAAV injected mice 4 weeks post injection compared with uninjected (control) mice (nZ 3 hearts in each group).

(g) Summary data of isolated cardiomyocyte lengthto-width ratio (n 25 in each group). (h) Representative echocardiography traces from a control and

TnC D73N mouse (note the electrical abnormalities in the TnC D73N mouse; lower recordings). (i) Summary data of TnC WT, TnC D73N and control mice left ventricular systolic and (j) diastolic dimensions measured via echocardiography 45 weeks after injection (n 5 for control, 4 for TnC WT and 11 for

TnC D73N ). (k) Summary data of TnC WT, TnC D73N and control mice EF measured 45 weeks after injection (n 5 for control, 4 for TnC WT and 11 for

TnC D73N). (l) Representative ECG recordings of arrhythmias observed in a TnC D73N mouse 4 weeks after injection. (m) Percentage of surviving mice at 8 weeks post injection. The numbers in the column shows the fraction of total mice alive at 8 weeks after injection. *Po0.05 versus all other groups using analysis of variance and NewmanKeuls pairwise analysis. Error bars are s.e.m.

4 NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794 ARTICLE

Table 1 | Summary data of isolated cardiomyocyte contraction and relaxation under basal conditions and beta-adrenergic stimulation (isoproterenol, 10 6 M).

Contraction Relaxation%RCL TTPs DF/F0 TTPc Shortening RT50 Ca Transient RT50

BasalControl 3.50.4 3447 1.40.1 1072 33611 2407 WT 3.50.4 3557 1.60.1 1082 3408 2415 D73N 2.40.2* 3369 1.60.1 992 30211* 2257 L48Q 4.70.4* 3327 1.50.1 1122 32611 2379

ISOControl 7.40.9 2127 4.80.3 932 19210 1064 WT 8.31.4 2287 4.70.2 992 21818 1194 D73N 4.90.7* 2199 3.30.3* 912 18412 1074 L48Q 9.61.2 2215 4.30.2 1042 19715 1234

ANOVA, analysis of variance; F/F , Ca2 transient amplitude; %RCL, per cent change in resting cell length; RT50, relaxation time to 50%; TTPc, time to peak Ca2 transient; TTPs, time to peak

shortening.nZ19 myocytes/group.

*Po0.05 versus other groups using ANOVA and NewmanKeuls pairwise analysis.

load-independent measurements of in vivo cardiac contractility23. To test if TnC L48Q altered relaxation, as has been observed with various Ca2 sensitizing compounds4, we next measured the rate of cardiac relaxation at multiple heart rates along with the end diastolic pressurevolume relationship (EDPVR). The increase in cardiac contractility occurred without delaying the frequency dependent acceleration of relaxation or altering the EDPVR (Fig. 3d,e). Thus, TnC L48Q increases contractility without affecting the diastolic function of the heart.

The major regulator of in vivo contractility is sympathetic stimulation24. Several of the current positive inotropes increase contraction through the beta-adrenergic pathway13. Over stimulating the beta-adrenergic response limits contractile reserve, makes the heart prone to arrhythmias and increases mortality14. To test if this pathway is altered in TnC L48Q mice, we investigated the effects of the beta-adrenergic agonist dobutamine. Consistent with our myocyte data, TnC L48Q mice displayed a normal contractile reserve to beta-adrenergic stimulation, again without affecting the ability of the heart to relax (Fig. 3f,g). Another issue with Ca2 sensitization and sympathetic stimulation is the generation of triggered arrhythmias14. In conscious, unrestrained mice or anaesthetized mice (beta-adrenergic agonist isoproterenol), we did not observe electrocardiography (ECG) abnormalities in the TnC L48Q mice that could be associated with atrial or ventricular electrical dysfunction (Fig. 3h, Supplementary Table 1). Furthermore, TnC L48Q did not change cell death or brosis (Supplementary Fig. 7). Thus, TnC L48Q increases cardiac function without impairing the beta-adrenergic response, electrical function or morphology.

We next tested whether this increase in cardiac function translated into increased cardiovascular performance. During exercise tolerance testing, a predictor of cardiovascular tness and health25,26, the TnC L48Q mice displayed enhanced cardiovascular performance as evidenced by increased VO2max

and distance run before exhaustion (Fig. 3i,j). Thus, TnC L48Q has properties of a promising positive inotrope that can increase cardiac function and cardiovascular performance without adversely affecting cardiac morphology, electrical activity, beta-adrenergic response, relaxation, diastolic function, intracellular Ca2 or survival. Unlike other positive inotropes, there appears to be no detrimental impact of long-term (1 year) TnC L48Q transduction on cardiac function, remodelling or survival (Fig. 3k,l). These are the desired characteristics of a positive inotrope that could be used to aid and improve the diseased heart.

TnC L48Q protects cardiac function and performance after MI. To test if TnC L48Q preserves cardiac function and cardiovascular performance after MI, TnC L48Q and TnC WT expressing mice underwent surgical induction of MI 5 weeks after injection (Fig. 4a). TnC L48Q did not alter infarct size (3 days post MI; Fig. 4b; Supplementary Fig. 8) or reduce scar formation compared with that of TnC WT (8 weeks post MI; Fig. 4c). Even after MI, we observed robust transduction and sarcomeric TnC exchange (Fig. 4d-f).

To test the short-term and long-term effects of TnC L48Q, we measured cardiac function beginning at 3 days and up to 8 weeks post MI. Even though 440% of the left ventricle was damaged by

MI, TnC L48Q mice were able to resist the MI induced left ventricular dysfunction. At 3 days after MI, TnC L48Q mice presented signicantly higher EF and fractional shortening (Fig. 5a,b) and pressurevolume analysis revealed a faster rate of pressure development (dp/dtmax) across multiple heart rates (Fig. 5c). Moreover, the dp/dtmax generated by TnC L48Q myocardium was not signicantly different from that of the sham operated mice myocardium (Supplementary Fig. 9). Consistent with increased contractility, Ees and PRSW were also signicantly higher in the TnC L48Q mice (Fig. 5d,e). TnC L48Q mice also had lower left ventricular dilation (Fig. 5f,g) and showed no arrhythmia or differences in heart rate variability in conscious and unrestrained TnC L48Q mice (Fig. 5h). Additionally, TnC L48Q mice did not have compromised relaxation (Fig. 5i). Thus, TnC L48Q has positive inotropic properties that provide protective support immediately after MI.

Over the 8 week time course, TnC L48Q mice maintained signicantly higher EF and fractional shortening (Fig. 5a,b). Pressurevolume analysis demonstrated the signicant enhancement of cardiac contractility was also maintained, as evidenced by higher dp/dtmax, Ees and PRSW (Fig. 5ce). This increased

contractility occurred without impairing the ability of the heart to relax across a wide range of heart rates (Fig. 5j). To test whether this increase in cardiac function after MI translated into increased cardiovascular performance, we subjected the mice to exercise tolerance testing. In line with the maintained cardiac function, TnC L48Q mice displayed higher VO2max and ran a signicantly

longer distance before reaching exhaustion (Fig. 5k,l). Consistent with protected cardiac function and performance, the lung weight of the TnC L48Q mice did not show signs of congestive heart failure, which was evident in TnC WT mice (Fig. 5m). These ndings strongly suggest that TnC L48Q is a promising positive inotrope that can protect cardiac contractility and performance

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794

Cont

L48Q

b c

120

18 16 14 12

140

130

120

110

100

Control

WT * *

ControlWT L48Q

Pressure (mm Hg)

100

80 60

5 80

0.5 350

300 250 200 150 100

L48Q

Ees (mm Hg l1 )

EDPVR (mm Hg l1 )

PRSW (mm Hg l1 )

40 20

30 40 50

Volume (l)

60 70

e f

Control WT

L48Q

ControlWT L48Q

0.4

0.3

0.2

0.1

0.0

Tau (ms)

8 6

200 300 400 500 600BPM

dP/dt max/EDV (mm Hg ms1 ml1 )

Baseline

Dobutamine

Baseline

Dobutamine

Baseline

Dobutamine

0 Cont WT

L48Q

12 10

8 6

Control basal

WT basal

WT ISO

* *

Tau (ms)

V0 2max(ml kg

Control ISO

420 Cont WT

Baseline

Dobutamine

Baseline

Dobutamine

Baseline

Dobutamine

L48Q

1 min1 )

Distance (metres)

Control WT

L48Q

ControlWT L48Q

L48Q basal

L48Q ISO

j k l

350300 60

Control L48Q

250 200 150 100

Ejection fraction (%)

0 Control L48Q

Diameter (mm)

0 Diastolic

Systolic

Figure 3 | TnC L48Q increases cardiac contractility and performance. (a) Representative pressurevolume loops from TnC L48Q and control mice during inferior vena cava occlusion. The thick solid lines are ts for end systolic pressurevolume relationship (ESPVR), whereas the dashed lines are ts for the EDPVR (b) Summary data of TnC L48Q, TnC WTand control mice Ees and (c) PRSW (n 8 for control, 11 for TnC L48Q and 4 for TnC WT). (d) Summary

data of TnC L48Q, TnC WT and control mice frequency dependent acceleration of relaxation and (e) EDPVR (n 8 for control, 11 for TnC L48Q and 4 for

TnC WT). (f) Summary data of TnC L48Q, TnC WT and control mice to beta-adrenergic stimulation (dobutamine, 5 mg kg 1) measured as dp/dtmax/end diastolic volume (EDV) and (g) tau. (h) Representative ECG traces of anaesthetized mice at baseline and after isoproterenol injection (n 6 for TnC L48Q

and control, 4 for TnC WT). (i) Summary data of VO2max and (j) distance run for mice during exercise tolerance testing (n 12 for TnC L48Q and control, 6

for TnC WT). Summary data of EF (k) and chamber dimensions (l) 1 year post injection (n 7 for control and TnC L48Q). *Po0.05 versus control (for f

and g *Po0.05 versus corresponding baseline) using two-tailed Students t-test or analysis of variance and NewmanKeuls pairwise analysis as appropriate. Error bars are s.e.m.

after MI. Furthermore, TnC L48Q long-term expression does not cause any adverse cardiac effects. Specically, TnC L48Q mice resisted ventricular dilation and hypertrophy (Fig. 5f,g,n). Most

importantly, there was no increase in mortality after MI commonly observed by long-term use of other inotropic strategies previously tested27,28 (Fig. 5o). All these functional changes were

6 NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794 ARTICLE

Injection

Time (weeks)

L48Q 3 days post MI L48Q 8 weeks post MI

WT 3 days post MI

5 0

WT 3 weeks post MI

TTC

H&E

mCherry Flag Flag

Figure 4 | Histological analysis of protective MI hearts. (a) Timeline of rAAV9 injection, induction of myocardial infarction and terminal measurements. The gradient of red in the timeline shows the approximate timing of TnC expression. Representative images of infarct size 3 days post-MI (b) and 8 weeks post-MI (c) for TnC L48Q and TnC WT mice (scale bar, 1,000 mm). Representative images for TnC L48Q and TnC WT-3-days and 8 weeks post MI showing: (d) mCherry uorescence (scale bar, 1,000 mm); (e) anti-Flag immunouorescence (scale bar, 1,000 mm); and (f) sarcomeric anti-Flag immunouorescence (scale bar, 25 mm).

observed without any difference in cell death, brosis or inammation (Fig. 6ac). Thus, TnC L48Q protects cardiac function and performance after an MI. We believe our MI data is consistent with TnC improving the contractility of the surviving myocytes rather than salvaging dying tissue or decreasing inammation.

Therapeutic role of TnC L48Q. To assess the therapeutic potential of TnC L48Q as an inotropic intervention, we modied our experimental strategy so that expression of TnC L48Q would begin B12 weeks after the infarct (Fig. 7a). Eight weeks after the

MI, we observed robust TnC expression and sarcomeric exchange (Fig. 7bd). Over the 8 weeks (56 days) of observation, TnC L48Q resulted in a steady and signicant improvement in EF (Fig. 8a). Pressurevolume analysis at 8 weeks revealed that the TnC L48Q mice had enhanced cardiac contractility (Ees and PRSW; Fig. 8b,c) with dP/dtmax comparable to that developed by the sham operated mice (Fig. 8d). The improvement in cardiac contractility resulted in signicantly less remodelling (that is, ventricular dilation and hypertrophy, Fig. 8eg). Consistent with the improved contractility and function, TnC L48Q also increased the exercise tolerance of the mice as demonstrated by signicantly enhanced VO2max and distance run before reaching exhaustion (Fig. 8h,i). Furthermore, TnC L48Q mice did not have compromised relaxation and showed no sign of congestive heart failure (Fig. 8j,k). Unlike other positive inotropes, long-term

transduction of TnC L48Q lessened mortality (Fig. 8l). In this instance, we observed a larger number of infarcted TnC WT mice die (compared with Fig. 5o). Considering the majority of mice die within 2 weeks after MI, we injected mice with TnC L48Q at this time point and still observed a progressive improvement in EF (Supplementary Fig. 10). All these functional changes were observed without any difference in cell death, brosis or inammation (Fig. 9ac). These ndings demonstrate that TnC L48Q is capable of improving the contractility of the surviving myocytes rather than salvaging dying tissue or decreasing inammation of an infarcted heart to act as a therapeutic intervention.

DiscussionHere, we provide the rst data that smartly formulating sarcomeric proteins can be used to improve cardiac function and performance in vivo. Previous in vitro work in simplied physiological systems showed that altering Ca2 binding to TnC has a strong inuence on contraction9,15. However, in these systems, Ca2 levels were typically held at steady-state. With the dynamic changes to Ca2 in the in vivo heart, it has been presumed that altered Ca2 binding to TnC would be ineffective (that is, TnC would be subservient to the Ca2 signal). However, we believe that the response of the contractile apparatus is nely tuned to the Ca2 signal2,29,30. That is, these systems work in concert to ultimately dictate the level of contraction. Thus,

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 7

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794

a b c

70 60 50 40 30 20 10

WT L48Q

1 )

Ejection fraction (%)

Fractional shortening

(%)

* *

* * * * * * * *

dP/dt max(mm Hg ms

8,000

6,000

4,000

2,000

WT (3 days) WT (8 weeks) L48Q (3 days) L48Q (8 weeks)

2 4 6 8 0 3d 2 4 6 8 300 400 500 600

Weeks

0 0 3d

Weeks BPM

d e

Ees PRSW

L48Q WT L48Q

Systolic diameter (mm)

* *

** **

mm Hg l1

** ** ** **

* *

Diastolic diameter (mm)

mm Hg l1

0 0 3d 8W 0 3d 8W

Sham

3 days

8 weeks

Sham

3 days

8 weeks

L48Q

h i j

30 16

350

300

250

200

100

10 0 0 10 20 30 40 50 60

WT (3 day)

Tau (ms)

L48Q (3 day) WT (3 day) WT sham WT 8W MI

L48Q (3 day)L48Q 8W MI

0 0 3d 8W 0 3d 8W WT L48Q

0 300 400 500 600BPM

300

6 200 400 500 600 BPM

140

130

120

110

100 0

WT L48Q

1 min1 )

** **

V02 max(ml kg

Distance (metres)

8 weeks

Sham 8 weeks

Sham

m n o

WT L48Q

Lung weight / Tibia lenght

(mg mm1 )

0.009

0.006

0.003

0.000

Heart weight / Tibia lenght

(mg mm1 )

Survival (%)

Sham

WT L48Q

3 days

Sham 8 weeks

0 Sham 3 days Days

8 weeks

Figure 5 | TnC L48Q protects cardiac function and performance after MI. Summary data of sham or before MI (day 0), TnC L48Q and TnC WT mice (a) EF and (b) fractional shortening before and up to 8 weeks after MI. (c) Summary data of dP/dtmax, (d) Ees and (e) PRSW 3 days or 8 weeks after MI.

(f) Summary data of systolic and (g) diastolic left ventricular dimensions before, 3 days and 8 weeks post-MI. (h) Representative ECG traces in conscious, unrestrained TnC L48Q and TnC WTmice 3 days post-MI. Summary data of frequency dependent acceleration of relaxation (i) 3 days and (j) 8 weeks post-MI. Summary data of (k) VO2max and (l) distance run during exercise tolerance testing 8 weeks post-MI. (m) Summary data of lung weight to tibia length and (n) heart weight to tibia length 3 days and 8 weeks post-MI. (o) KaplanMeier survival curves (n 12 in each group). (For data sets (ao) n 7 for TnC

WT 3 day MI, 9 for TnC L48Q-3 day MI, 6 for TnC WT-8-weeks MI, 8 for TnC L48Q-8 weeks MI,6 for sham TnC WTand 6 for sham TnC L48Q). *Po0.05 versus sham TnC WT, **Po0.05 versus corresponding TnC WT using analysis of variance and NewmanKeuls pairwise analysis. Error bars are S.E.M.

8 NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794 ARTICLE

L48Q 3 day

L48Q 8 week

WT 3 day WT 8 week

Injection

Time (days) 2 0

IgG Col cd45

L48Q 8 week WT 8 week

mCherry Flag Flag

Scar Border zone Remote

WT 3 day

WT8 week

Figure 7 | Histological analysis of therapeutic MI hearts. (a) Timeline of rAAV9 injection, induction of myocardial infarction and terminal measurements. The gradient of red in the timeline shows the approximate timing of TnC expression. Representative images for TnC L48Q and TnC WT-8-weeks post MI showing: (b) mCherry uorescence (scale bar, 1,000 mm), (c) anti-Flag immunouorescence (scale bar, 1,000 mm); and (d) sarcomeric anti-Flag immunouorescence (scale bar, 25 mm).

Figure 6 | Similar cell death, brosis and inammation after MI in TnC L48Q and TnC WT mice. Representative images of 3 days and 8 weeks post-MI for TnC L48Q and TnC WT mice showing (a) IgG immunouorescence (scale bar, 1,000 mm) and (b) brosis via collagen immunouorescence (scale bar, 400 mm). (c) Representative images of 3 days and 8 weeks post-MI for TnC L48Q and TnC WT mice showing inammation of scar, border zone and remote myocardium via CD45 immunouorescence (green) and mCherry uorescence (red; scale bar, 100 mm).

changing either system (the Ca2 trigger or the response) will inuence contractility2. Our data strongly suggests that altering the Ca2 sensor (TnC) has a signicant impact on in vivo heart function and performance. Thus, modulating TnC is a logical approach to regulate cardiac contraction in health and a possible therapeutic approach for disease, without the confounding issues of increasing intracellular Ca2 .

While generating new myocytes31 and/or limiting cell death32 (which we did not observe) are viable treatment options, prevailing thought is that positive inotropes are whipping a sick horse to death33 and only used as a last resort (decompensated heart failure)34. We are resurrecting an old idea1 that increasing contractility will aid a diseased heart since our approach does not have detrimental effects. Using a model of impaired heart function (MI), the Ca2 sensitized TnC L48Q was able to aid the diseased heart. In fact, the in vivo performance of TnC L48Q suggests that it may be a promising positive inotrope. That is, transducing the myocardium with TnC L48Q increases cardiac function and performance without adversely affecting cardiac morphology,

electrical activity, beta-adrenergic response, relaxation, diastolic function, intracellular Ca2 or survival. Furthermore, TnC L48Q was not only protective but also therapeutic and benecial chronically, unlike other positive inotropes27. These positive effects can be achieved with only a modest replacement of the endogenous TnC, increasing the likelihood of gene therapy approaches for sarcomeric proteins. Since we did not observe any changes in cell death, brosis or inammation, we propose a new paradigm that smartly formulating TnC may be a viable therapeutic approach for many cardiomyopathies.

On the other hand, there are several cardiomyopathies that manifest with hypercontractility35. Our previous in vitro work has designed specic TnC constructs to correct these hypercontractile disorders (that is, TnC D73N), as well as restore the aberrant responses that cause different types of familial and acquired cardiomyopathies5. Hence, we are able to customize a variety of TnC sensors to tune the contractile response of various cardiac diseases. Combining gene therapy36 with designer proteins opens the door for unprecedented personalized medicines against the plethora of diseases.

Methods

Animal care and use. All the experiments were performed using C57Bl/6 J mice (Jackson Laboratories). For the intraperitoneal injection studies, a mixed population of males and females were injected at 23 days of age and experimented at 48 weeks of age. For the intra-thoracic cavity injection studies, male mice were injected at 1214 weeks of age, and serially studied as indicated. All the mice were housed in an animal facility with a 12-h light/dark cycle and free access to food and water. All experimental procedures were approved and met the guidelines set by the Institutional Animal Care and Use Committee at The Ohio State University.

Study groups. Control represents non-injected, normal mice. WT represents mice injected with rAAV9 containing a ag-tagged WT TnC. L48Q represents

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 9

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794

a b c

WT L48Q

L48Q

WT L48Q

Ejection fraction (%)

dP/dt max(mm Hg ms

Ees (mm Hg l1 )

PRSW (mm Hg l1 )

8,000

0 0 2 4

Weeks

6 8

Sham 8 weeks

6 5 4 3 2

1 )

6,000

4,000

2,000

0 300 400 500 600

BPM

systolic diameter (mm)

1 0

Diastolic diameter (mm)

1 0

6 5 4 3 2

L48Q

Sham

0 8 0 8

WT L48Q

0 8 0 8

WT L48Q

g h i

10 WT L48Q

WT MI

* **

WT L48Q

1 min1 )

Heat weight / Tibia

length (mg mm1 )

** **

V02 max(ml kg

125

115

105

Distance (metres)

Sham 8 weeks

135 300

275

250

225

200

0 Sham 8 weeks

Sham 8 weeks

Sham

L48Q MI

WT L48Q

Tau (ms)

0 300 400 500 600BPM

Lung weight / Tibia

length (mg mm1 )

0.009

0.006

0.003

0.000

Sham 8 weeks

100

0 0 30 60 Days

Survival (%)

L48Q

Sham

Figure 8 | Therapeutic effects of TnC L48Q after an MI. (a) Summary data of TnC L48Q and TnC WT mice EF before and up to 8 weeks post-MI (n 12

per group). Summary data of (b) Ees and (c) PRSW, (d) dp/dtmax 8 weeks post-MI. (e) Summary data of systolic and (f) diastolic left ventricular dimensions before and 8 weeks post-MI. (g) Summary data of heart weight to tibia length 8 weeks post-MI and sham. (h) Summary data of VO2max and

(i) distance run during exercise tolerance testing 8 weeks post-MI and sham. (j) Summary data of TnC L48Q and TnC WT mice frequency dependent acceleration of relaxation 8 weeks post-MI and sham. (k) Summary data of lung weight to tibia length. (l) KaplanMeier survival curves. n 12 per group.

For data sets (b-l) n 5 for TnC WT, 9 for TnC L48Q, 4 for sham TnC WT and 4 for sham TnC L48Q. *Po0.05 versus WT sham, **Po0.05 versus

corresponding TnC WT using analysis of variance and NewmanKeuls pairwise analysis. Error bars are s.e.m.

mice injected with rAAV9 containing a ag-tagged L48Q TnC. D73N represents mice injected with rAAV9 containing a ag-tagged D73N TnC.

Myobril Ca2 sensitivity. Ventricular cardiac muscle was obtained from male New Zealand White rabbits (23 months old)37. For obtaining ventricular myobrils, ventricular muscle was dissected in a KrebsHenseleit solution (in mM, 137 NaCl, 5 KCl, 1.2 MgSO4, 1.2 NaH2PO4, 20 NaHCO3, 10 glucose, 0.25 CaCl2 and 20 2,3-butanedione monoxime). Isolated ventricular tissue was minced with scissors in KrebsHenseleit solution and homogenized with 10 s bursts of a Polytron homogenizer. The suspension was passed through cheesecloth and then further Dounce-homogenized. The myobrils were collected by centrifugation and resuspended in buffer A (in mM, 10 MOPS, 150 KCl, 3 MgCl2, 1 dithiothreitol and0.02% Tween 20 (pH 7.0)) containing N-((2-(iodoacetoxy)ethyl)-N-methyl)amino-7-nitrobenz-2-oxa-1,3-diazole (IANBD)-labelled Tn (6 mM stock in buffer A) and stored at 4 C overnight. After the overnight Tn exchange, the myobrils were washed three times with buffer A to remove the un-exchanged Tn. Steady-state

Ca2 sensitivity was performed using a Perkin Elmer LS55 spectrouorimeter at 15 C with excitation of 470 nm for IANBD uorescence. Microliter amounts of CaCl2 were added with the [Ca2 ]free measured using a computer programme developed by Robertson and Potter6.

Virus production and injection. Adeno-Associated Virus serotype 9 was produced in HEK293 cells and puried using an iodixanol gradient purication38. The rAAV9 titre was obtained via TaqMan (LifeTechnologies) based quantication using a CMV sequence specic probe. For neonatal mice, 3050 ml containing1 1011 viral genomes were injected at 23 days of age through intraperitoneal

injection of both male and female pups. Adult (1214 weeks old) male mice were injected through the intra-thoracic cavity with 100 ml containing 1 1011 viral

genomes. Mice were lightly anaesthetized (1% isourane) and kept in supine position with clear access to the chest area. A 29.5 gauge needle was then inserted at an angle halfway between the ribs and B7.5 mm left of sternum. Care was taken not to insert the needle into the lungs or heart. We developed this technique to

10 NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794 ARTICLE

L48Q 8 week WT 8 week

IgG Col

Scar

Border zone Remote

L48Q

8 week

cd45

Figure 9 | Similar cell death, brosis and inammation after MI in TnC L48Q and TnC WT mice. Representative images for TnC L48Q and TnC WT mice 8 weeks post-MI showing (a) IgG immunouorescence (scale bar, 1,000 mm) and (b) brosis via collagen immunouorescence (scale bar, 400 mm). (c) Representative images for TnC L48Q and TnC WT mice 8 weeks post-MI showing inammation of scar, border zone and remote myocardium via CD45 immunouorescence (green) and mCherry uorescence (red; scale bar, 100 mm).

restrict the viral transduction to the adult heart. We chose this technique over

straight intracardiac injection because: (1) its very simple and straightforward;(2) can be performed extremely quickly; (3) no additional instrumentation or surgery is required; (4) it is highly reproducible; (5) requires only a single injection and (6) most importantly, there is signicantly less potential for damaging the rapidly beating heart.

Whole-heart uorescence imaging. A central pin was placed from the atria through the apex of the heart to standardize rotational positions. The entire epicardial surface of each heart was imaged by consequent 90 rotation around the longitudinal axis to capture emission light from anterior, left lateral, posterior and right arterial projections in a bath of 4 C Tyrode solution39. Next, a cut was made at the anterior ventricular septum to open the right and left ventricles and expose the endocardial surface. Excitation light generated by four halogen lamps was passed through an excitation lter (52525 nm, BrightLine). The uorescent light emitted from the heart was bandpass-ltered (58520 nm, BrightLine) before reaching the MiCAM Ultima-L CMOS camera (SciMedia) with high spatial(100 100 pixels, 120 mm per pixel) resolution. The total uorescent signal from

the heart at each projection was averaged, excluding tissue within 0.5 mm of the tissue edge due to light scattering, using customized software (Matlab)40.

Histology. Hearts were frozen in liquid nitrogen cooled isopentane using Optimum Cutting Temperature Compound (OCT) (Tissue-Tek). 8 mm cryosections were performed for histology. Hematoxylin and eosin and triphenyl tetrazolium chloride staining were performed. In brief, the hearts were excised and cut into 2-mm-thick transverse slices. The transverse slices were incubated with 0.75% triphenyl tetrazolium chloride solution (15 min, 37 C) and xed in 4% formalin solution (24 h at room temperature). The infarct area was measured using ImageJ (NIH). mCherry uorescence was obtained directly from that of the virally transduced protein. Immunoglobulin G (IgG) immunouorescence was performed as a measure of cell death4146 via an anti-mouse IgG goat antibody (1:200, Jackson Immunoresearch). Flag immunouorescence was performed using a rabbit anti-ag antibody (1:400, Cell Signaling Technology). Collagen immunouorescence

was performed using a rabbit anti-collagen antibody (1:200, Abcam). Endothelial cells were identied via CD31 immunouorescence using a rat anti-CD31 antibody (1:150, Abcam)47. Inammation was measured via CD45 immunouorescence using rat anti-CD45 antibody (1:50, Abcam)48.

Qualitative and quantitative expression of exogenous TnC. Samples were prepared by homogenizing mouse ventricular tissue in urea sample buffer(8 M Urea, 2 M Thiourea, 75 mM dithiothreitol, 50 mM Tris-HCl, pH6.8)49.Qualitative detection of rAAV expressed TnC WT from ventricular homogenates was determined via SDSpolyacrylamide gel electrophoresis silver stain. The quantitative amount of rAAV expressed TnC WT or TnC D73N in the cardiac myocyte was determined by western blot50. Homogenates were fractionated by SDSpolyacrylamide gel electrophoresis on a 15% (29:1) Laemmli gel and transferred to 0.45 mm polyvinylidene diuoride (PVDF) membrane. Western blot of the transferred membranes was conducted by incubation with the anti-TnC monoclonal antibody 7B9 (Fitzgerald, 1:1000) followed by a uorescent labelled secondary antibody (Jackson ImmunoResearch, 1:3000) for detection on a Typhoon imager 9410 (GE). The presence of a C-terminal FLAG in the rAAV expressed TnC slows its migration compared with that of the endogenous TnC resulting in distinct exogenous and endogenous TnC Western bands. The TnC D73N band was conrmed via anti-Flag (Sigma, 1:1000). Fluorescent quantication of the 7B9 bands were conducted by ImageQuant TL (GE) in arbitrary units and the amount of exogenous TnC was reported as the per cent exogenous TnC of the total myolament TnC (for example, exogenous TnC / (exogenous TnC endogenous TnC) 100).

We determined that the 7B9 TnC antibody demonstrates B1020-fold decreased afnity to TnC L48Q compared with the exogenous TnC WT or TnC D73N. This was the case for three different commercially available TnC antibodies tested. The decreased antibody afnity against the TnC L48Q likely results from the L48Q amino acid disruption of the TnC antibody epitope that is located in a different region from D73N. Due to the difference in antibody afnity, we were unable to quantify the rAAV exogenous expression of the TnC L48Q by the above method. Instead, using standard curves, we determined the concentration of both the expressed Flag-tagged TnC L48Q and endogenous TnC. To minimize cross-reactivity, parallel gels were run containing identically loaded ventricular samples. The gels also contained a standard curve for known amounts of recombinant Flag-tagged TnC L48Q and control TnC. Anti-TnC 7B9 (Fitzgerald, 1:1000) was used to probe for the endogenous TnC and anti-Flag (Sigma, 1:1000) was used to probe for Flag-tagged TnC L48Q. Based on the standard curves, each tissue samples endogenous TnC and ag-tagged TnC L48Q was quantied. The per cent of FLAG-tagged TnC L48Q expressed was determined by dividingthe concentration of TnC L48Q by the total TnC (mg of expressed Flag-tagged

TnC L48Q/(mg of expressed Flag-tagged TnC L48Q mg of endogenous

TnC) 100). NOTE: Ca2 binding proteins can migrate as two bands dependent

upon free Ca2 in SDS51. There is different free Ca2 in the heart samples compared to the puried proteins, causing the high afnity TnC L48Q to be most affected.

Cardiac myocyte isolation and broblast culture. Cardiomyocytes were isolated by Langendorff perfusion52. Briey, the heart was cannulated and suspended from the Langendorff apparatus to be perfused with Ca2-free Tyrode solution (Normal tyrode solution in mM: 140 NaCl, 4 KCl, 1 MgCl2, 10 glucose, and 5 HEPES, pH7.4 adjusted with NaOH or HCl) for 5 min. Subsequently, the heart was perfused with a tyrode solution containing Liberase Blendzyme II (0.077 mg ml 1)

(Roche Applied Science, Indianapolis, IN). After 46 min, the heart was removed from the Langendorff apparatus and the ventricles were minced to dissociate the myocyte via trituration. The myocytes were then ltered and centrifuged before being resuspended in a tyrode solution containing 200 mM Ca2 . To obtain broblasts53, the supernatant from the centrifugation step for myocyte isolation was collected and centrifuged at 400 g for 10 min. The pellet obtained was suspended in DMEM media with 10% foetal calf serum and plated on a 6-well tissue culture plate and cultured for 48 h. Media was replaced after 24 h.

High-content microscopy. Myocytes were plated on laminin coated tissue culture plates and xed with 4% paraformaldehyde for 20 min at room temperature and then washed with 1XPBS before scanning. The spot detector algorithm (Cellomics ArrayScanXTI, ThermoFisher) was used to identify rod shaped myocytes in the Hoechst channel by adjusting the ObjectShapeLWR parameter. Over 3000 rod shaped myocytes were analysed for quantication of mCherry uorescence.

Myocyte contraction/relaxation measurements. Myocytes were loaded with Flou-4 Am (10 mM, Molecular Probes, Eugene, OR) and incubated for 30 min. Myocytes were then washed and allotted an additional 30 min for de-esterication52. A Cairn Research Limited (Faversham, UK) epiuorescence system was used for isolated myocyte shortening measurements. Intracellular Ca2 was measured via Fluo-4 epiuorescence with excitation: 48020 nm and emission: 53525 nm. The change in uorescent intensity is expressed as DF/F0, where F is the uorescence intensity and F0 is the uorescence intensity at rest. Shortening data was collected by video edge detection (Crescent Electronics). Myocytes were

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 11

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794

eld stimulated at 1 Hz via platinum electrodes connected to a Grass Telefactor S48 stimulator (West Warwick, RI).

Myocardial infarction and Echocardiography. Permanent MI surgery was performed by anaesthetizing mice with isourane (2%) and mechanical ventilation. After a left thoracotomy, the fourth intercostal space and the lungs were retracted. The left anterior descending coronary artery was identied under a microscope and permanently ligated with a 8-0 silk suture near its origin. Ligation of the artery was conrmed by distal palor of the left ventricular anterior wall. The sham group had similar surgical procedure without tightening of the suture around the artery. The mice were kept on heat pad until recovery of consciousness. All mice were observed for the next 72 h after surgery. Echocardiography was performed using a VEVO 2100 Visual Sonics (Visual Sonics, Toronto) system54. The mice were lightly anaesthetized (1.5% isourane) and the EF, fractional shortening and ventricular chamber dimensions were determined through M mode using the parasternal short axis view.

Pressurevolume loop. Cardiac haemodynamic measurements were assessedvia a closed chest approach using a 1.4 French Millar Pressure catheter(AD Instruments) advanced into the left ventricle through the right carotid artery22. In brief, mice were anaesthetized by ketamine (55 mg kg 1) plus xylazine (15 mg kg 1) in saline solution and placed in supine position on a heat pad.

Following a midline neck incision, the underlying muscles were pulled to expose the carotid artery. Using a 4-0 suture the artery was tied and the pressurevolume catheter was advanced through the artery into the left ventricle of the heart. After 510 min of stabilization, values at baseline and stimulation at varying frequencies (410 Hz) were recorded. Pressurevolume loops were also obtained at varying preloads via inferior vena cava occlusions to get Ees, PRSW and EDPVR. To measure the beta-adrenergic response, 5 mg kg 1 dobutamine was injected intraperitoneal. All the measurement and analysis were performed on LabChart7 (AD Instruments).

ECG. In unanaesthetized, unrestrained mice, the electrocardiogram was measured using an ECGenie system (Mouse Specics, Inc.)43. Mice were placed on a footplate electrode for 30 min and ECG was recorded. For anaesthetized ECG recording55, mice were anaesthetized using 2% isourane and placed in prone position on a heating pad to maintain body temperature. Subcutaneous electrodes were placed in the Lead II position and baseline ECG was measured for 5 min on Powerlab 4/30 (AD Instruments). After 5 min, isoproterenol was infused intra-peritoneally to measure ECG under beta-adrenergic stimulation for 15 min.

Exercise tolerance testing. A metabolic chamber with treadmill and Oxymax analyzer (Columbus Instruments) were used to measure VO2max and distance

run54. The mice were acclimatized at a lower speed (6 m min 1 at 0o incline) for 10 min a day before the actual measurement. After a 5-min initiation period, the treadmill was adjusted to a 6-m min 1 walk for 5 min. The treadmill was then placed at a 20o incline and the speed increased 1 m min 1 every minute until the mice reached exhaustion (mice unwilling to run and neglecting an electrical shock).

VO2max was dened as the absolute maximal value obtained during the procedure.

Statistical and data analysis. Sample size was calculated by power analysis using Minitab 5.1. Data exclusion was determined via the Grubbs test. Investigators collecting and/or analysing the data were blinded to which study group the mouse/ tissue belonged. The survival analysis was performed by log-rank test. All other statistical signicance was determined by Students t-test or analysis of variance followed by a NewmanKeuls post hoc test. All values are expressed as a means.e.m. For all statistical analysis the level of signicance was set at Po0.05.

References

1. Lee, J. A. & Allen, D. G. Modulation of Cardiac Calcium Sensitivity: A New Approach to Increasing the Strength of the Heart347 (Oxford University Press, Inc., 1993).

2. Davis, J. P. & Tikunova, S. B. Ca(2 ) exchange with troponin C and cardiac

muscle dynamics. Cardiovasc. Res. 77, 619626 (2008).3. Hwang, P. M. & Sykes, B. D. Targeting the sarcomere to correct muscle function. Nat. Rev. Drug Discov. 14, 313328 (2015).

4. Kass, D. A. & Solaro, R. J. Mechanisms and use of calcium-sensitizing agents in the failing heart. Circulation 113, 305315 (2006).

5. Liu, B., Lee, R. S., Biesiadecki, B. J., Tikunova, S. B. & Davis, J. P. Engineered troponin C constructs correct disease-related cardiac myolament calcium sensitivity. J. Biol. Chem. 287, 2002720036 (2012).

6. Tikunova, S. B. & Davis, J. P. Designing calcium-sensitizing mutations in the regulatory domain of cardiac troponin C. J. Biol. Chem. 279, 3534135352 (2004).

7. Feest, E. R. et al. Thin lament incorporation of an engineered cardiac troponin C variant (L48Q) enhances contractility in intact cardiomyocytes from healthy and infarcted hearts. J. Mol. Cell. Cardiol. 72, 219227 (2014).

8. Kekenes-Huskey, P. M., Lindert, S. & McCammon, J. A. Molecular basis of calcium-sensitizing and desensitizing mutations of the human cardiac troponin C regulatory domain: a multi-scale simulation study. PLoS Comput. Biol. 8, e1002777 (2012).

9. Norman, C., Rall, J. A., Tikunova, S. B. & Davis, J. P. Modulation of the rate of cardiac muscle contraction by troponin C constructs with various calcium binding afnities. Am. J. Physiol. Heart Circ. Physiol. 293, H2580H2587 (2007).

10. Tikunova, S. B. et al. Effect of calcium-sensitizing mutations on calcium binding and exchange with troponin C in increasingly complex biochemical systems. Biochemistry 49, 19751984 (2010).

11. Wang, D. et al. Structural and functional consequences of the cardiac troponin C L48Q Ca(2 )-sensitizing mutation. Biochemistry 51, 44734487 (2012).

12. Luo, M. & Anderson, M. E. Mechanisms of altered Ca(2)( ) handling in heart

failure. Circ. Res. 113, 690708 (2013).13. Hasenfuss, G. & Teerlink, J. R. Cardiac inotropes: current agents and future directions. Eur. Heart. J. 32, 18381845 (2011).

14. Overgaard, C. B. & Dzavik, V. Inotropes and vasopressors: review of physiology and clinical use in cardiovascular disease. Circulation 118, 10471056 (2008).

15. Kreutziger, K. L. et al. Calcium binding kinetics of troponin C strongly modulate cooperative activation and tension kinetics in cardiac muscle. J. Mol. Cell. Cardiol. 50, 165174 (2011).

16. Bar, C. et al. Telomerase expression confers cardioprotection in the adult mouse heart after acute myocardial infarction. Nat. Commun. 5, 5863 (2014).

17. Du, C. K. et al. Knock-in mouse model of dilated cardiomyopathy caused by troponin mutation. Circ. Res. 101, 185194 (2007).

18. McConnell, B. K. et al. Knock-in mice harboring a Ca(2 ) desensitizing

mutation in cardiac troponin C develop early onset dilated cardiomyopathy. Front. Physiol. 6, 242 (2015).19. Kirk, J. A. et al. Left ventricular and myocardial function in mice expressing constitutively pseudophosphorylated cardiac troponin I. Circ. Res. 105, 12321239 (2009).

20. Murphy, A. M. et al. Transgenic mouse model of stunned myocardium. Science 287, 488491 (2000).

21. Martins, A. S. et al. In vivo analysis of Troponin C knock-in (A8V) mice: evidence that TNNC1 is a hypertrophic cardiomyopathy susceptibility gene. Circ. Cardiovasc. Genet. 8, 653664 (2015).

22. Zhang, B., Davis, J. P. & Ziolo, M. T. Cardiac catheterization in mice to measure the pressure volume relationship: investigating the Bowditch effect. J. Vis. Exp. 100, e52618 (2015).

23. Georgakopoulos, D. et al. In vivo murine left ventricular pressure-volume relations by miniaturized conductance micromanometry. Am. J. Physiol. 274, H1416H1422 (1998).

24. Bers, D. M. & Ziolo, M. T. When is cAMP not cAMP? Effects of compartmentalization. Circ. Res. 89, 373375 (2001).

25. Ehsani, A. A., Biello, D., Seals, D. R., Austin, M. B. & Schultz, J. The effect of left ventricular systolic function on maximal aerobic exercise capacity in asymptomatic patients with coronary artery disease. Circulation 70, 552560 (1984).

26. Kaminsky, L. A. & Tuttle, M. S. Functional assessment of heart failure patients. Heart Fail. Clin. 11, 2936 (2015).

27. Curfman, G. D. Inotropic therapy for heart failurean unfullled promise. N. Engl. J. Med. 325, 15091510 (1991).

28. Toma, M. & Starling, R. C. Inotropic therapy for end-stage heart failure patients. Curr. Treat. Options Cardiovasc. Med. 12, 409419 (2010).

29. Kobayashi, T. & Solaro, R. J. Calcium, thin laments, and the integrative biology of cardiac contractility. Annu. Rev. Physiol. 67, 3967 (2005).

30. Biesiadecki, B. J., Davis, J. P., Ziolo, M. T. & Janssen, P. M. Tri-modal regulation of cardiac muscle relaxation; intracellular calcium decline, thin lament deactivation, and cross-bridge cycling kinetics. Biophys. Rev. 6, 273289 (2014).

31. Loughran, J. H., Chugh, A. R., Ismail, I. & Bolli, R. Stem cell therapy: promising treatment in heart failure? Curr. Heart Fail. Rep. 10, 7380 (2013).

32. Heusch, G. Cardioprotection: chances and challenges of its translation to the clinic. Lancet 381, 166175 (2013).

33. Bussmann, W.-D. Acute and Chronic Heart Failure: Diagnosis and Therapy 286 (Springer-Verlag, 1986).

34. Feldman, D. S., Elton, T. S., Sun, B., Martin, M. M. & Ziolo, M. T. Mechanisms of disease: detrimental adrenergic signaling in acute decompensated heart failure. Nat. Clin. Pract. Cardiovasc. Med. 5, 208218 (2008).

35. Bonne, G., Carrier, L., Richard, P., Hainque, B. & Schwartz, K. Familial hypertrophic cardiomyopathy: from mutations to functional defects. Circ. Res. 83, 580593 (1998).

36. Greenberg, B. et al. Design of a phase 2b trial of intracoronary administration of AAV1/SERCA2a in patients with advanced heart failure: the CUPID 2 trial (calcium up-regulation by percutaneous administration of gene therapy in cardiac disease phase 2b). JACC Heart Fail. 2, 8492 (2014).

12 NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10794 ARTICLE

54. Roof, S. R. et al. Obligatory role of neuronal nitric oxide synthase in the hearts antioxidant adaptation with exercise. J. Mol. Cell. Cardiol. 81, 5461 (2015).

55. Makara, M. A. et al. Ankyrin-G coordinates intercalated disc signaling platform to regulate cardiac excitability in vivo. Circ. Res. 115, 929938 (2014).

Acknowledgements

We thank Dr. Peter Chen (University of Washington) for providing the Adeno-Asso

ciated Virus expression vector with internal ribosome entry site-mCherry. We thank Dr

Danesh Sopariwala for technical advice regarding collection of VO2max measurements.

We thank Ben Canan and Michael Makara for technical advice on ECG measurements

and Grace Davis and Jonathan O. Davis for collection of ECG data. We thank Drs.

Peter J. Mohler and Muthu Periasamy for critical reading of the manuscript and

providing helpful discussions. This effort was supported by NIH grants R01 HL114904

(B.J.B.), R56 HL091986 and R01 HL091986 (J.P.D.), R01HL074045 and R01HL063043

(S.G.), K02 HL094692 (M.T.Z.), R01113084 (P.M.L.J.) and R01 HL115580 (V.V.F.), and

National Center for Research Resources Award TL1RR025753 (S.C.L.).P.M.L.J. is also

supported by the Fred A. Hitchcock Professorship in Environmental Physiology.

Author contributions

J.P.D. conceptualized idea of the study. V.S., S.C.L., J.P.D. and M.T.Z. designed the

experiments. J.Z. subcloned the constructs and made the rAAV9 viruses. V.S. and S.C.L.

developed and standardized the virus delivery techniques. V.S., B.Z. and S.C.L.acquired

and analysed the in vivo mice data. H.E.S. and B.J.B. performed the western blotting and

analysis. S.R.R. and H.-T.H. conducted isolated myocyte experiments. B.Z. performed the

LAD ligation and PV loop measurements. B.J.H., N.L., N.W. and V.V.F. performed

whole-heart imaging. J.A.R-F, L.B. and S.G. oversaw and performed the histology

experiments, J.K.L and N.W. performed high-content microscopy, F.A performed

broblast culture, P.M.L.J. analysed the ECG measurements. V.S. and M.T.Z. made the

gures. M.T.Z. and J.P.D. performed statistical analysis. V.S., J.P.D. and M.T.Z. wrote the

manuscript.

37. Little, S. C. et al. The rates of Ca2 dissociation and cross-bridge detachment

from ventricular myobrils as reported by a uorescent cardiac troponin C.J. Biol. Chem. 287, 2793027940 (2012).38. Johnson, P. R. et al. Vector-mediated gene transfer engenders long-lived neutralizing activity and protection against SIV infection in monkeys. Nat. Med. 15, 901906 (2009).

39. Glukhov, A. V. et al. Calsequestrin 2 deletion causes sinoatrial node dysfunction and atrial arrhythmias associated with altered sarcoplasmic reticulum calcium cycling and degenerative brosis within the mouse atrial pacemaker complex1. Eur. Heart J. 36, 686697 (2015).

40. Lou, Q. et al. Upregulation of adenosine A1 receptors facilitates sinoatrial node dysfunction in chronic canine heart failure by exacerbating nodal conduction abnormalities revealed by novel dual-sided intramural optical mapping. Circulation 130, 315324 (2014).

41. Hansson, G. K. & Schwartz, S. M. Evidence for cell death in the vascular endothelium in vivo and in vitro. Am. J. Pathol. 112, 278286 (1983).

42. Ohno, M. et al. "Apoptotic" myocytes in infarct area in rabbit hearts may be oncotic myocytes with DNA fragmentation: analysis by immunogold electron microscopy combined with In situ nick end-labeling. Circulation 98, 14221430 (1998).

43. Janssen, P. M. et al. Prednisolone attenuates improvement of cardiac and skeletal contractile function and histopathology by lisinopril and spironolactone in the mdx mouse model of Duchenne muscular dystrophy. PLoS ONE 9, e88360 (2014).

44. Rafael-Fortney, J. A. et al. Early treatment with lisinopril and spironolactone preserves cardiac and skeletal muscle in Duchenne muscular dystrophy mice. Circulation 124, 582588 (2011).

45. Dumont, E. A. et al. Cardiomyocyte death induced by myocardial ischemia and reperfusion: measurement with recombinant human annexin-V in a mouse model. Circulation 102, 15641568 (2000).

46. Kabaeva, Z., Meekhof, K. E. & Michele, D. E. Sarcolemma instability during mechanical activity in Largemyd cardiac myocytes with loss of dystroglycan extracellular matrix receptor function. Hum. Mol. Genet. 20, 33463355 (2011).

47. Pusztaszeri, M. P., Seelentag, W. & Bosman, F. T. Immunohistochemical expression of endothelial markers CD31, CD34, von Willebrand factor, and Fli-1 in normal human tissues. J. Histochem. Cytochem. 54, 385395 (2006).

48. Hermiston, M. L., Xu, Z. & Weiss, A. CD45: a critical regulator of signaling thresholds in immune cells. Annu. Rev. Immunol. 21, 107137 (2003).

49. Biesiadecki, B. J. et al. Removal of the cardiac troponin I N-terminal extension improves cardiac function in aged mice. J. Biol. Chem. 285, 1968819698 (2010).

50. Nixon, B. R. et al. Combined troponin I Ser-150 and Ser-23/24 phosphorylation sustains thin lament Ca(2 ) sensitivity and accelerates deactivation in an

acidic environment. J. Mol. Cell. Cardiol. 72, 177185 (2014).51. Rhyner, J. A., Koller, M., Durussel-Gerber, I., Cox, J. A. & Strehler, E. E. Characterization of the human calmodulin-like protein expressed in Escherichia coli. Biochemistry 31, 1282612832 (1992).

52. Kohr, M. J. et al. Biphasic effect of SIN-1 is reliant upon cardiomyocyte contractile state. Free Radic. Biol. Med. 45, 7380 (2008).

53. Accornero, F. et al. Genetic analysis of connective tissue growth factor as an effector of transforming growth factor beta signaling and cardiac remodeling. Mol. Cell. Biol. 35, 21542164 (2015).

Additional information

Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Web End =http://www.nature.com/

http://www.nature.com/naturecommunications

Web End =naturecommunications

Competing nancial interests: The authors declare no competing nancial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions

Web End =http://npg.nature.com/

http://npg.nature.com/reprintsandpermissions

Web End =reprintsandpermissions /

How to cite this article: Shettigar, V. et al. Rationally engineered Troponin C modulates

in vivo cardiac function and performance in health and disease. Nat. Commun. 7:10794

doi: 10.1038/ncomms10794 (2016).

This work is licensed under a Creative Commons Attribution 4.0

International License. The images or other third party material in this

article are included in the articles Creative Commons license, unless indicated otherwise

in the credit line; if the material is not included under the Creative Commons license,

users will need to obtain permission from the license holder to reproduce the material.

To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/

NATURE COMMUNICATIONS | 7:10794 | DOI: 10.1038/ncomms10794 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 13

Word count: 10254

Show less

Abstract

Translate

Treatment for heart disease, the leading cause of death in the world, has progressed little for several decades. Here we develop a protein engineering approach to directly tune in vivo cardiac contractility by tailoring the ability of the heart to respond to the Ca²⁺ signal. Promisingly, our smartly formulated Ca²⁺ -sensitizing TnC (L48Q) enhances heart function without any adverse effects that are commonly observed with positive inotropes. In a myocardial infarction (MI) model of heart failure, expression of TnC L48Q before the MI preserves cardiac function and performance. Moreover, expression of TnC L48Q after the MI therapeutically enhances cardiac function and performance, without compromising survival. We demonstrate engineering TnC can specifically and precisely modulate cardiac contractility that when combined with gene therapy can be employed as a therapeutic strategy for heart disease.

Details

Title

Rationally engineered Troponin C modulates in vivo cardiac function and performance in health and disease

Author

Shettigar, Vikram; Zhang, Bo; Little, Sean C; Salhi, Hussam E; Hansen, Brian J; Li, Ning; Zhang, Jianchao; Roof, Steve R; Ho, Hsiang-ting; Brunello, Lucia; Lerch, Jessica K; Weisleder, Noah; Fedorov, Vadim V; Accornero, Federica; Rafael-fortney, Jill A; Gyorke, Sandor; Janssen, Paul M L; Biesiadecki, Brandon J; Ziolo, Mark T; Davis, Jonathan P

Pages

10794

Publication year

2016

Publication date

Feb 2016

Publisher

Nature Publishing Group

e-ISSN

20411723

Source type

Scholarly Journal

Language of publication

English

DOI

https://doi.org/10.1038/ncomms10794

ProQuest document ID

1767624993

Rationally engineered Troponin C modulates in vivo cardiac function and performance in health and disease

Jump to:

Full text

Abstract

Details

Suggested sources