adenine nucleotide translocase 1 deficiency results in dilated … · 2016-11-13 · oxidative...

J A C C : C A R D I O V A S C U L A R I M A G I N G V O L . 4 , N O . 1 , 2 0 1 1

© 2 0 1 1 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 1 9 3 6 - 8 7 8 X / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . D O I : 1 0 . 1 0 1 6 / j . j c m g . 2 0 1 0 . 0 6 . 0 1 8

O R I G I N A L R E S E A R C H

Adenine Nucleotide Translocase 1 DeficiencyResults in Dilated Cardiomyopathy With Defectsin Myocardial Mechanics, HistopathologicalAlterations, and Activation of Apoptosis

Nupoor Narula, BS,*† Michael V. Zaragoza, MD, PHD,*‡ Partho P. Sengupta, MBBS,�Peng Li, MD, PHD,§ Nezam Haider, PHD,§ Johan Verjans, MD,§ Katrina Waymire, MA,†Mani Vannan, MD,§ Douglas C. Wallace, PHD*†‡

Irvine, California; and Scottsdale, Arizona

O B J E C T I V E S The aim of this study was to test the hypothesis that chronic mitochondrial energy

deficiency causes dilated cardiomyopathy, we characterized the hearts of age-matched young and old

adenine nucleotide translocator (ANT)1 mutant and control mice.

B A C K G R O U N D ANTs export mitochondrial adenosine triphosphate into the cytosol and have a

role in the regulation of the intrinsic apoptosis pathway. Mitochondrial energy deficiency has been

hypothesized, on the basis of indirect evidence, to be a factor in the pathophysiology of dilated

cardiomyopathies. Ant1 inactivation should limit adenosine triphosphate for contraction and calcium

transport, thereby resulting in early cardiac dysfunction with later dilation and heart failure.

M E T H O D S We conducted a multiyear study of 73 mutant (Ant1�/�) and 57 control (Ant1�/�)

mice, between the ages of 2 and 21 months. Hearts were characterized by cardiac anatomy,

echocardiographic imaging with velocity vector analysis, histopathology, and apoptosis assays.

R E S U L T S The Ant1�/� mice developed a distinctive concentric dilated cardiomyopathy, charac-

terized by substantial myocardial hypertrophy and ventricular dilation, with cardiac function declining

earlier in age as compared to control mice. Left ventricular circumferential, radial, and rotational

mechanics were reduced even in the younger mutants with preserved systolic function. Histopathologic

analysis demonstrated increased myocyte hypertrophy, fibrosis, and calcification in the mutant mice as

compared with control mice. Furthermore, increased cytoplasmic cytochrome c levels and caspase 3

activation were observed in the mutant mice.

C O N C L U S I O N S Our results demonstrate that mitochondrial energy deficiency is sufficient to

cause dilated cardiomyopathy, confirming that energy defects are a factor in this disease. Energy

deficiency initially leads to early mechanical dysfunction before a decline in left ventricular systolic

function. Chronic energy deficiency with age then leads to heart failure. Our results now allow us to use

the Ant1�/� mouse model for testing new therapies for ANT1mutant patients. (J Am Coll Cardiol Img

2011;4:1–10) © 2011 by the American College of Cardiology Foundation

From the *Center for Mitochondrial and Molecular Medicine and Genetics (MAMMAG), University of California, Irvine,
California; †Department of Biological Chemistry, University of California, Irvine, California; ‡Department of Pediatrics,Division of Genetics and Metabolism, University of California, Irvine, California; §Department of Medicine, Division of

Mcd(taau

pwri

mlaoto

Tamfdat

dttrwttec

M

M(CegcsbiCp(Cc6ttrsqlZivLfttIdwG

CCsrIHTMatce

M1

A

A

A

t

L

V

L

d

E

I

L

p

m

d

ROS � reactive oxidative sp

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 4 , N O . 1 , 2 0 1 1

J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.

Ant1 Mouse Cardiomyopathy

2

itochondria generate adenosine triphos-phate by oxidative phosphorylation.The resulting matrix adenosine triphos-phate is exchanged across the mito-

hondrial membrane for cytosolic adenosineiphosphate by the adenine nucleotide translocatorsANTs) (1). Nuclear-encoded ANTs are 30-kDaransmembrane polypeptides proposed to functions homodimers (2). Certain ANT isoforms also playrole in the regulation of apoptosis through mod-lating the mitochondrial permeability transition

See page 11

ore (3–5). Animals harbor up to 4 ANT isoforms,hich exhibit differential tissue-specific and tempo-

al expression patterns. In both human and mouse,soform 1 (ANT1, human; Ant1, mouse) is predom-

inantly expressed in high-energy tissues,including heart, muscle, and brain (6–8).

In humans, ANT1 null, autosomal re-cessive mutations cause mitochondrialmyopathy and cardiomyopathy (9). Anti-bodies against ANT have been commonlyobserved in patients with dilated cardio-myopathy (10). These observations impli-cate ANT1 in the etiology of cardiomy-opathy associated with dilation, reducedfunction, and cardiomyocyte loss by apop-tosis (11,12). Before the discovery ofANT1-deficient patients, our laboratorygenetically inactivated Ant1 in the mouse.The Ant1�/� mice are viable and fertile;however, they develop mitochondrial my-opathy with hyperproliferation of abnor-

al mitochondria, ragged red muscle fibers, andactic acidosis (13). This mouse Ant1 defect is alsossociated with increased mitochondrial reactivexidative species (ROS), antioxidant enzyme induc-ion, and early accumulation of mitochondrial de-xyribonucleic acid (mtDNA) mutations (14).

ardiology, University of California, Irvine, California; and the �Mayolinic Arizona, Scottsdale, Arizona. Ms. Narula presented this manu-

cript at the Annual Meeting of the American Society of Echocardiog-aphy, San Diego, June 2010, for the Arthur E. Weyman, MD Youngnvestigator’s Award. This work was supported by National Institutes ofealth Grants NS21328, NS041850, and DK73691; a Doris Dukeranslational Research Grant; and a California Institute for Regenerativeedicine Comprehensive Grant RC1-00353-1 awarded to Dr. Wallace

nd K08 HL081222 awarded to Dr. Zaragoza. The authors have reportedhat they have no relationships to disclose. Ms. Narula and Dr. Zaragozaontributed equally to this article. A. Jamil Tajik, MD, acted as guestditor for this paper.

r

g

m

ecies

sanuscript received January 26, 2010; revised manuscript received June

5, 2010, accepted June 28, 2010.

hese energetic and ROS aberrations are alsossociated with the coordinate induction of bothtDNA and nuclear deoxyribonucleic acid genes

or mitochondrial biogenesis, induction of nucleareoxyribonucleic acid antioxidant and anti–poptotic genes, and the downregulation of apop-otic genes (15,16).

To test the hypothesis that chronic mitochon-rial energy deficiency causes dilated cardiomyopa-hy, we characterized the nature and progression ofhe cardiomyopathy in Ant1-deficient mice. Theesults revealed that Ant1 deficiency is associatedith a unique cardiomyopathy with significant ven-

ricular hypertrophy and chamber dilation, whereinhe deranged myocardial mechanics occur at anarly age even before systolic dysfunction becomeslinically apparent.

E T H O D S

ice. We used homozygous Ant1B-geo mutantAnt1�/�) and control (Ant1�/�) mice on57BL/6J or 129S4 backgrounds. No clear differ-

nces were seen in results for the different back-rounds. The Ant1 mutant mice and polymerasehain reaction genotyping methods were as de-cribed (13). All animal procedures were approvedy University of California Irvine Institutional An-mal Care and Use Committee (#2002-2399).ardiac imaging. Two-dimensional echocardiogra-hy with M-mode and velocity vector imagingVVI) (Amid, Italy, and Siemens, Mountain View,alifornia), were performed on 73 mutant and 57

ontrol mice (ages 2 to 21 months; 63 female and7 male). The animals were anesthetized and ven-ilated with 98% oxygen and 2% isoflurane. Elec-rocardiography leads were applied to monitor heartate and trigger echo image acquisitions. Ultra-ound images were obtained with a 12-MHz fre-uency linear phased probe (Sequoia, Siemens), ineft ventricular (LV) short-axis view at apex level.oomed images were digitally captured. M-mode

mages at mid LV were used to determine leftentricular dimension at end-diastole (LVIDd) andV dimension at end-systole. The LV ejection

raction (EF) was calculated. Interventricular sep-um (IVS) wall thickness and left ventricular pos-erior wall (LVPW) thickness were obtained, andVS/posterior wall ratio was determined. Echocar-iograms in DICOM 4.0 format were analyzedith Research Arena (TomTec, Unterschleissheim,ermany) based 2-dimensional VVI package (Ver-

B B R E V I A T I O N S

N D A C R O N YM S

NT � adenine nucleotide

ranslocator

V � left ventricle/ventricula

VI � velocity vector imagin

VIDd � left ventricular

imension at end-diastole

F � ejection fraction

VS � interventricular septu

VPW � left ventricular

osterior wall

tDNA � mitochondrial

eoxyribonucleic acid

ion 3.702, Siemens). The endocardial and epicar-

dffpsVfpHmtees6Wmw7r�cat(rSW5pbwmtcanbhSSIPCvmhmgwpp1

cdbr

wtammt(pM

R

AdcBirmcsib

fsattdddoAOawacmitTh�hp


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


3

ial borders were manually identified in a singlerame of a cine-loop, and the borders in otherrames were automatically generated. Segmentaleak systolic velocity, diastolic velocity, radialtrain, and circumferential strain were obtained byVI. Accuracy of VVI for testing myocardial de-

ormation in experimental settings has been donereviously (17).istopathology. Of 130 mice imaged, 72 (ages 2 to 19onths; 20 males and 52 females) underwent his-

opathologic analysis. Cardiac tissue was fixed inither formalin or paraformaldehyde, paraffin-mbedded, and sectioned (7 �m). Sections weretained with hematoxylin and eosin. Sections from9 mice were also stained with Masson’s Trichrome.estern blot analysis. Heart samples (50 mg) from 3utants and 3 control mice, ages 14 to 21 months,ere diced and solubilized in 210 mmol/l mannitol,0 mmol/l sucrose, 1 mmol/l ethylene glycol tet-aacetic acid, 5 mmol/l HEPES pH 7.4, and 110g/�l digitonin for 5 min. The samples were

entrifuged at 10,000 rpm for 10 min at 4°C,nd the supernatant protein concentration was de-ermined by the BioRad Protein Assay reagentBioRad, Hercules, California). Samples wereead spectrophotomically at � � 595 nm withpectronic Genesys 2 (Thermo Fisher Scientific,altham, Massachusetts). Samples were boiled formin, and proteins were separated on 15% SDS-

olyacrylamide gel (8-cm length) with Tris-glycineuffer. After blotting, membranes were processedith first step antibody. We used mouse antibodies:onoclonal anti–cytochrome-c (PharMingen In-

ernational, San Diego, California); polyclonal anti-aspase 3 (Biotechnology, Santa Cruz, California);nd monoclonal anti-GAPDH (Chemicon Inter-ational, Temecula, California). Second-step anti-ody was polyclonal anti-mouse IgG conjugated toorseradish peroxidase (Amersham Biosciences,an Diego, California).tatistical analysis. We used JMP version 7.0 (SASnstitute, Cary, North Carolina) and GRAPHPADRISM software (GraphPad Software, La Jolla,alifornia). Student t test was employed for mean

alues. For 3-group analyses, no corrections wereade for multiple comparisons. To compare the

istopathologic findings between control mice andutants, Pearson chi-square tests of 2 � 2 contin-

ency tables were used; a 2 � 3 contingency tableas used to compare the 3 categories of hypertro-hy (none, mild, and moderate-to-severe). Two-tailvalue was used for histopathologic analysis, and

-tail p value was used for biochemical analysis. We E

onsidered values of p � 0.05 as a significantifference. Due to heterogeneity of variance in theiochemical data, both parametric and nonparamet-ic tests and the Welch t test were employed.

Echocardiographic parameters of LV mechanicsere correlated with LVEF with Pearson correla-

ion coefficient (R) with p value. The diagnosticccuracy of echocardiographic parameters of LVechanics for differentiating mutant from controlice was quantified with receiver-operator charac-

eristic analysis. Area under the curve was calculated18) with binomial confidence interval and com-ared with the method of DeLong et al. (19) withedCalc Software (Mariakerke, Gent, Belgium).

E S U L T S

nt1 cardiomyopathy: concentric hypertrophy withilation. M-Mode echocardiographic images ofontrol and mutant hearts are shown in Figure 1A.y averaging the data (Table 1), we observed an

ncrease in heart weight and heart/body-weightatio and LVIDd, IVS, and LVPW thickness inutant animals, indicative of LV dilation and

oncentric myocardial hypertrophy. Fractionalhortening and EF were reduced in mutant hearts,ndicating presence of LV systolic dysfunction (Ta-le 1).Mutant mice displayed significant regional de-

ects in myocardial contraction. Examples arehown for control and mutant hearts in Figures 1Bnd 1C. Circumferential and radial strains, rota-ion, and rotational velocity for control versus mu-ant hearts are shown in Table 1. Mutant heartsemonstrated 40% reduction in rotation, 25% re-uction in rotational velocity, and more than 50%ecrease in circumferential and radial strain devel-pment, all highly significant.ge-related progression of Ant1 cardiomyopathy.ur results showed that mutant hearts are in

ggregate abnormal in all contractile parameters butith considerable variability. Because we examined

nimals from ages 2 to 21 months, this variabilityould be the result of variable penetrance of theutation or age-related decline in cardiac function

n mutant hearts. To investigate this, we analyzedhe EF of control and mutant hearts by age (Fig. 2).his suggested that the average EF of controlearts increased throughout life (B coefficient �0.39, p � 0.056). By contrast, EF of mutant

earts declined with age (B coefficient� �0.79,� 0.029). There was also significant variability in
F in each age group, greater in mutant than in


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


4

Figure 1. Cardiomyopathy in Ant1�/� Mouse

(A) M-mode echocardiography. Ultrasound images depict left ventricular dimensions in a control (upper panel) and mutant (bottom panel) mouse. Interventricular septum (IVS)thickness is shown as distance between blue and red lines, and left ventricular internal dimension at end-diastole (LVIDd) is shown as distance between red and yellow lines.The LVIDd is increased from 3.3 mm in control to 3.9 mm in mutant. (B) Velocity vector imaging: radial and rotational velocity vectors. Left ventricular short axis in control (a–c)and Ant1 mutant (d–f). Endocardial velocity vectors (a and d) were attenuated at end-ejection in mutant. The 2-dimensional maps of radial velocity and the apical rotationalvelocity are presented in adjacent panels. In contrast to the uniform pattern of radial velocities seen in control mice (b), marked dyssynchrony is seen in the mutant animal (e).Similarly, the velocity of counterclockwise rotation in systole is biphasic in the control mice (c) but is dyssynchronous in the mutant animal (f). (C) Velocity vector imaging: veloc-ity, strain, and strain rates. Images from control (upper panels) and mutant (bottom panels). For each heart, velocity (a and d), strain (b and e), and strain rates (c and f) aredepicted by 3-dimensional color mapping. Uniform motion and shortening are observed in control, whereas dyssynchronous motion and shortening are delineated in the veloc-
ity, strain, and strain rate curves of mutant.

ccfvRp

thE(oETadd

wmpmtLwm(cm(

oEr

fv1rsrCwdmE(s

rain;


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


5

ontrol hearts (Fig. 2). The LVEF in mutants andontrol mice across all ages correlated with circum-erential strain, radial strain, rotation, and rotationalelocity in systole and diastole (R � 0.73, R � 0.65,

� 0.79, R � 0.74, and R � 0.71, respectively,� 0.001 for all).To further evaluate the decline in cardiac func-

ion, we divided the mutant hearts into those thatad EF within the range of control hearts (normalF) and those that fell below the normal range

abnormal EF). We determined the mean and SDf the EF of the control hearts and then selected theF value that was 2 SDs below the control mean.his lower cutoff value was an EF of 56%. With

pplication of this cutoff, 58% of the mutantsemonstrated EF below lower normal limits (Fig. 2,otted line).Next, we divided the mutant hearts into those

ith “normal EF” (n � 31) and those with “abnor-al EF” (n � 42). We then recalculated the major

arameters (Table 2). The mean EF for controlice, normal EF mutants, and abnormal EF mu-

ants was 71%, 65%, and 41%, respectively. TheVIDd of normal EF mutants and control miceere similar, whereas LVIDd of abnormal EFutants was greater than control mice by 12%

Table 2). Normal EF mutants showed 40% in-rease in IVS thickness, whereas abnormal EFutants showed 44% increase in IVS thickness

Table 2).We further characterized contraction mechanics

f LVs. Compared with control mice, both normalF mutants and abnormal EF mutants showed

Table 1. Morphological and Echocardiographic Parameters in C

Parameters Control (n � 57)

TBW (g) 39.39 � 9.65

Heart weight (g)* 0.21 � 0.05

Heart weight/TBW % 0.57 � 0.14

IVS (0.1 mm) 7.69 � 0.92

LVIDd (0.1 mm) 37.53 � 3.87

LVPW (0.1 mm) 7.59 � 0.87

IVS/LVPW 1.02 � 0.12

FS (%) 36.03 � 7.62

EF (%) 70.90 � 7.44

Rotation (°) 1.68 � 0.36

Rotation velocity (°/s) 45.68 � 9.35

CS (%) �12.01 � 2.12

RS (%) 11.96 � 2.60

Values are mean � SD. *Heart weights obtained for 61 of the 130 total mice (CS � circumferential strain; EF � ejection fraction; FS � fractional shortening;at end-diastole; LVPW � left ventricular posterior wall thickness; RS � radial st

eduction in LV rotational velocities, more marked

or abnormal EF mutants (Table 2). Rotationalelocity of normal EF mutants was decreased only1%, whereas that of abnormal EF mutants waseduced 60% (Table 2). Our analyses also revealedtriking differences in circumferential strain andadial strain for the control and mutant hearts.ompared with control mice, circumferential strainas reduced in normal EF mutants (35%), and thisifference was further exaggerated (60%) in abnor-al EF mutants (Table 2). Radial strain of normalF mutants was reduced (43%) and more profound

63%) in abnormal EF mutants, with all differencestatistically significant (Table 2).

10

20

30

40

70

90

80

50

60

EF

(%

)

Sac Age (mo)0 10 20

R2

R2

Figure 2. EF Scatter Plot for Ant1 Mutants and Control Mice

Mutants (X) and control mice (circle). Abnormal ejection fraction (Edefined as �2 SD below the average EF of control animals (dashedzontal line). The best-fit linear regression line is provided with R2 fo

ol and Ant1 Deficient Mice

Mutant (n � 73) p Value

30.89 � 4.47 �0.0001

0.26 � 0.06 �0.001

0.84 � 0.15 �0.0001

10.90 � 1.56 �0.0001

40.15 � 5.81 �0.01

10.95 � 1.73 �0.0001

0.99 � 0.08 0.1466

24.69 � 7.48 �0.0001

51.05 � 15.01 �0.0001

1.02 � 0.56 �0.0001

34.01 � 10.81 �0.0001

�6.05 � 2.35 �0.0001

5.41 � 2.23 �0.0001

ntrol mice and 26 mutants).interventricular septum thickness; LVIDd � left ventricular internal dimensionTBW � total body weight.

= 0.065

= 0.064

F) ishori-r each

ontr

35 coIVS �

group. Sac � sacrificed.

dprapLaep

ptrsd0(am

Chtdcmpfib

csasfmhHA

AUC � area under the cur


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


6

In summary, mutant mice with normal EF,espite having normal LVIDd, show subclinicalerturbations in contractile indexes. Although LVotation has been previously shown to remain un-ltered in hypertrophied hearts (20), LV hypertro-hy in Ant1�/� mutant hearts with preservedVEF was associated with significantly reduced LVpical rotation. This might suggest transmural dis-ase (Table 2). Furthermore, all cardiac dysfunctionarameters declined with age in the mutant animals.Table 3 shows the diagnostic accuracy of various

arameters of LV function for differentiating mu-ant from control mice. Circumferential strain,adial strain, and rotational velocity in diastolehowed greater area under the curve than EF forifferentiating mutants from control mice (p �.02) (Table 3). For hearts with normal EF�56%), all 3 parameters showed good diagnosticccuracy for differentiating mutants from controlice.

Table 2. Echocardiographic Parameters of LV Mechanics in ConObvious Systolic Dysfunction

Parameters Control (n � 57)

EF % 70.90 � 7.44

Age 10.03 � 4.82

LVIDd (0.1 mm) 37.53 � 3.87

IVS (0.1 mm) 7.69 � 0.92

Rotation (degree) 1.68 � 0.36

Rotation velocity (degree/s) 45.68 � 9.35

CS % �12.01 � 2.12

RS % 11.96 � 2.60

Values are mean � SD. *Compared with normal group, p � 0.0001; †compared§compared with mutant normal EF group, p � 0.05; �compared with mutant nAbbreviations as in Table 1.

ccuracy of Echocardiographic Parameters of LV Mechanics

AUC SE 95% CI p V

0.89 0.03 0.82 to 0.94 �0

0.96* 0.01 0.91 to 0.98 �0

0.96* 0.01 0.92 to 0.99 �0

0.81* 0.03 0.73 to 0.87 �0

0.77* 0.04 0.69 to 0.84 �0

ocity 0.94* 0.02 0.89 to 0.97 �0

0.91 0.03 0.84 to �0.96 �0

0.93 0.02 0.86 to �0.97 �0

0.63 0.06 0.52 to �0.73 0

0.52 0.06 0.41 to �0.62 0

ocity 0.88 0.03 0.79 to �0.94 �0

ve; CI � confidence interval; other abbreviations as in Table 1.

ardiac morphology and histopathology. Increasedeart/body-weight ratios, IVS, and LVPW of mu-ants all point to myocardial hypertrophy. Ouretailed hematoxylin and eosin analysis on 25ontrol mice (ages 2 to 19 months) and 47 mutantice (ages 3 to 18.5 months) (Fig. 3) confirmed the

resence of myocyte hypertrophy. In addition, weound evidence of interstitial edema (Figs. 3A to 3D),nterstitial inflammation, myocyte calcification,inucleation, and myofibrillar lysis.Myocyte hypertrophy was significantly more

ommon in mutants than in control mice [chi-quare(2) � 25.6, p � 0.0001] (Fig. 4A). This waslso true of mutant mice �12 months old [chi-quare(2) � 19.5, p � 0.0001] (Fig. 4B). The dif-erence in myocyte hypertrophy was lost in olderice due to the increased presence of myocyte

ypertrophy in older (�12 months) control mice.ypertrophy seems to be an early finding of the

nt1�/� heart.

and Ant1 Deficient Mice With and Without

Mutant

Normal EF (n � 31) Abnormal EF (n � 42)

64.79 � 5.76* 40.90 � 11.09*†

10.18 � 3.79 12.95 � 5.26‡§

37.52 � 3.98 42.10 � 6.21*�

10.77 � 1.50* 11.00 � 1.61*

1.50 � 0.26‡ 0.67 � 0.44*†

44.39 � 4.70 26.36 � 6.90*†

�7.81 � 1.99* �4.75 � 1.66*†

6.80 � 2.09* 4.38 � 1.72*†

mutant normal EF group, p � 0.0001; ‡compared with normal group, p � 0.01;al EF group, p � 0.001.

Cutoff Sensitivity (%) Specificity (%)

64% 79.5 89.5

�8.7% 87.7 96.5

7.7% 87.7 100.0

1° 49.3 100.0

31°/s 46.6 96.5

39°/s 94.5 78.5

�9% 77.4 94.7

7.7% 77.4 100.0

1.6% 77.4 49.1

52°/s 100.0 24.6

39°/s 87.1 78.9

trol

withorm

Table 3. Diagnostic A

Overall Group alue

EF �56% (abnormal) .001

CS* .001

RS .001

Rotation .001

Rotation velocity .001

Reverse rotation vel .001

EF �56% (normal)

CS .001

RS .001

Rotation .02

Rotation velocity .74

Reverse rotation vel .001

*Versus EF, p � 0.02.

iaMcpO(i4smcFotsoACimlftboccinpwwamhsW

D

Ghcascicaih

padttdolpcftcidpmocspshs

mtngm


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


7

To examine fibrosis, we stained myocardial spec-mens from 27 control mice (ages 2 to 19 months)nd 42 mutants (ages 3 to 18.5 months) with

asson’s Trichrome (Figs. 5A to 5F). Fibrosis waslassified as focal or multifocal and interstitial,erivascular, or replacement (Figs. 4C and 4D).verall, fibrosis was increased in mutant hearts

Fig. 4C), which became more evident and signif-cant in older hearts (�12 months) [chi-square(1) �.5, p � 0.05] (Fig. 4D). Moreover, multifocal inter-titial fibrosis was significantly more frequent in olderutants, and multifocal replacement fibrosis was ex-

lusively observed in mutants (Figs. 4C and 4D).ibrosis seems to be a later-stage phenomenon andccurs with greater frequency in mutants. Calcifica-ion was also strikingly increased in mutants [chi-quare(1) � 20.8, p � 0.0001] (Figs. 4A and 4B);ccurrence of calcification was independent of age.ctivation of apoptotic pathway in Ant1 mutant mice.ytochrome c levels were found to be 6-fold higher

n the older mutants as compared with older controlice (Fig. 6A). This could be due to the upregu-

ation of oxidative phosphorylation to compensateor the chronic energy deficiency (16) or the syn-hesis and release of mitochondrial cytochrome c oroth. Caspase activation was evaluated on the basisf the relative levels of the full length inactiveaspase 3 and the shorter cleaved and activatedaspase 3. Mutant hearts exhibited a 4-fold increasen amount of full length caspase 3. In contrast to theegligible levels in control myocardium, a largeroportion of cleaved and thus activated caspase 3as observed in the mutant mice. The differenceas statistically significant for cytochrome c release

s well as active caspase 3 by application of para-etric tests; however, due to small sample size and

eterogeneity of variance, the results were nottatistically significant when nonparametric or

elch t tests were applied (Fig. 6B).

I S C U S S I O N

enetic inactivation of the Ant1 gene in the mouseas been found to cause a unique cardiomyopathyharacterized by substantial myocardial hypertrophynd ventricular dilation. Echocardiographic analy-es of circumferential, radial, and rotational me-hanics permitted detection of cardiac dysfunctionn Ant1-deficient mice when global determinants ofontractility were still preserved, even as early as thege of 3 months. Although LV rotation is usuallyncreased or remains unaltered in hypertrophied
earts, LV hypertrophy in Ant1�/� mutants with e
reserved LVEF was associated with reduced LVpical rotation, suggesting presence of transmuralisease. Early evidence of cardiac defects suggestshat cardiac dysfunction might already be present inhe neonatal or prenatal period. This “latent” car-iac dysfunction was confirmed by demonstrationf histological changes in all mutant hearts, regard-ess of age. Mutant hearts had myocyte hypertro-hy, myofibrillar lysis, edema, inflammation, andalcification at all ages. Although none of theseeatures are individually pathognomonic, collec-ively they have been reported in idiopathic dilatedardiomyopathy. This pathology is associated withncreased activated-caspase 3, indicating that Ant1-eficiency is accompanied by activation of apoptoticathways; apoptosis has been reported in cardio-yopathic hearts. This Ant1 deficiency cardiomy-

pathy seems unique, because it produces dilatedardiomyopathy, which is associated with a sub-tantial degree of concentric myocardial hypertro-hy. Furthermore, this cardiomyopathy shows on-et of mechanical dysfunction early on even whenypertrophy is the predominant feature and globalystolic function is normal.

The clinical relevance of our Ant1�/� mutantouse model for human disease has been shown by

he recent report of a patient with a homozygousull missense (A123D) mutation in human ANT1ene. This individual experienced mitochondrialyopathy with ragged red fibers and exercise intol-

Figure 3. Histopathological Alterations in Mutant Animals

(A) Control, 7.5-month-old, with mild hypertrophy; (B) mutant, 9-moshows inflammation (boxed area) and calcification (circled area); (C17-month-old, shows normal myocardial architecture with mild hyp(white arrows); (D) mutant, 18-month-old, shows basophilic degene(white arrow) and myofibrillar lysis (triangles).

nth-old,) control,ertrophyration

rance together with a hypertrophic cardiomyopa-

thtdtmfwwt

a(tApaghPA


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


8

hy. On echocardiography the patient was seen toave a concentric LV hypertrophy with comparablyhickened IVS and LVPW. The LV was notilated, and EF was still normal (9). In contrast tohis human case resulting from a homozygous nullutation (9), other cases of human disease resulting

rom ANT1 gene missense mutations are inheritedith autosomal dominance; these mutations presentith mitochondrial myopathy with progressive ex-

ernal ophthalmoplegia, in absence of cardiomyop-

Figure 4. Quantification of Histopathologic Features

(A and B) Hematoxylin and eosin stained; (C and D) Masson’s trichgate of data from mice �12 months old; (D) aggregate of data fromF � focal; Hyp � hypertrophy; Infl � inflammation; Int � interstitiareplacement.

Figure 5. Cardiac Fibrosis by Masson’s Trichrome Staining

(A) Control, 8.5-month-old, shows multifocal interstitial fibrosis; (B)trol, 14.5-month-old, shows no fibrosis; (D) mutant, 13.5-month-oldfocal perivascular fibrosis and focal interstitial fibrosis (not seen her
multiple foci of interstitial fibrosis.
thy (21,22). Neither the ANT1 (A123D) patient9) nor our mouse (23) exhibited evidence of oph-halmoplegia and ptosis. Hence, the dominantNT1 missense mutations in autosomal dominantrogressive external ophthalmoplegia seem to becting as dominant negative mutations and areenetically and phenotypically different from theomozygous null mutations for cardiomyopathy.erhaps the biochemical influence of the dominantNT1 missense mutations is amplified through

stained; (A and C) aggregate of data from all ages; (B) aggre-ice �12 months old. Binucl � binucleation; Calc � calcification;� multifocal; MFL � myofibrillar lysis; Per � perivascular; Rep �

ant, 5.5-month-old, shows multifocal replacement fibrosis; (C) con-ws multifocal interstitial fibrosis; (E) control, 19-month-old, showsF) mutant, 18.5-month-old, shows focal replacement fibrosis and

romem

l; M

mut, shoe); (

timme

aArThdtrnRampftptt

pmpAo

tatatmAg

AmpmAamwsmcit

RWioPv


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


9

heir inactivation in dimers with mutant subunitsncorporated. Alternatively, dominant missense

utations might change ANT1 function, perhapsaking it leakier for protons and thus reducing

nergy metabolism (24).One of the most striking features of both the

utosomal recessive and autosomal dominantNT1 diseases, in man and mouse, is the age-

elated accumulation of multiple mtDNA deletions.he accumulation of multiple mtDNA deletionsas been documented in muscle for both the humanominant and recessive cases (9,21) and in heart inhe Ant1�/� mouse (14). These mtDNA rear-angements have been proposed to result fromucleotide imbalance (21) or excessive mtDNAOS damage (13). Whatever the mechanism, thege-related accumulation of mtDNA damageight progressively erode mitochondrial energy

roduction, exacerbating the inherited ANT1 de-ect and leading to the age-related progression ofhe cardiomyopathy. If so, one therapeutic ap-roach might be administration of mitochondriallyargeted antioxidants to inhibit mtDNA damage,hus slowing disease progression.

We demonstrated initiation of intrinsic apoptosisathway leading to the activation of caspase 3 inutant hearts. This might indicate that apoptotic

rocesses contribute to the progression of theNT1-deficient cardiomyopathy (25). Expression

Figure 6. Apoptosis Assays by Western Blot

(A) Increased cytochrome c levels in 3 mutants versus 3 control mice,14 kDa, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) � 38error bars are SEM. Inactive caspase 3 � 32 kDa, active caspase 3 �

kDa � kilodaltons; SEM � standard error of the mean.

f a caspase-8 fusion protein in the hearts of w

ransgenic mice resulted in activation of apoptosisnd cardiomyopathy (26), and cardiomyocyte apop-osis has been associated with dilated cardiomyop-thy (27). Therefore, these observations suggesthat therapies that stabilize the mitochondrial per-eability transition pore, such as cyclosporin-derivatives, might play a role in reducing pro-

ression of Ant1�/� cardiomyopathy (28,29).The most direct approach to treating the

NT1�/� cardiomyopathy would be to replace theissing ANT1 gene. We have been successful in

artially restoring ANT1 function in Ant1�/�ouse muscle through transduction of the mousent1 complementary deoxyribonucleic acid withdeno-associated virus. After transduction of theuscle with the Ant1–adeno-associated virus, thereas a striking amelioration of the severe aspects of

keletal muscle pathology (30). By developing theouse Ant1�/� model and performing detailed

haracterization of the cardiomyopathy, we are nown position to evaluate relative benefits of variousherapeutic modalities.

eprint requests and correspondence: Dr. Douglas C.allace, Center of Mitochondrial and Epigenomic Med-

cine, Children’s Hospital of Philadelphia, Departmentf Pathology and Laboratory Medicine, University ofennsylvania, CTRF 6060, 3501 Civic Center Boule-ard, Philadelphia, Pennsylvania 19104-4302. E-mail:

�12 months, p � 0.05, error bars show SEM. Cytochrome c �

. (B) Increased inactive and active caspase 3 levels in mutants,kDa. AU � arbitrary unit; CF � cleaved form; FL � full length;

ageskDa17

[email protected].

mailto:[email protected]


J A N U A R Y 2 0 1 1 : 1 – 1 0

Narula et al.


10

R

1

1

2

2

2

2

2

2

2

2

2

3

Kty

E F E R E N C E S

1. Portman MA. The adenine nucleotidetranslocator: regulation and functionduring myocardial development andhypertrophy. Clin Exp Pharmacol2002;29:334–8.

2. Pebay-Peyroula E, Dahout-GonzalezC, Kahn R, et al. Structure of mito-chondrial ADP/ATP carrier in com-plex with carboxyatractyloside. Nature2003;426:39–44.

3. Kokoszka JE, Waymire KG, Levy SE,et al. The ADP/ATP translocator isnot essential for the mitochondrialpermeability transition pore. Nature2004;427:461–5.

4. Bauer MK, Schubert A, Rocks O,Grimm S. Adenine nucleotidetranslocase-1, a component of the per-meability transition pore, can domi-nantly induce apoptosis. J Cell Biol1999;147:1493–502.

5. Zamora M, Granell M, Mampel T,Vinas O. Adenine nucleotide translo-case 3 (ANT3) overexpression inducesapoptosis in cultured cells. FEBS Lett2004;563:155–60.

6. Li K, Warner CK, Hodge JA, et al. Ahuman muscle adenine nucleotidetranslocator gene has four exons, islocated on chromosome 4, and is dif-ferentially expressed. J Biol Chem1989;264: 13998–14004.

7. Stepien G, Torroni A, Chung AB,Hodge JA, Wallace DC. Differentialexpression of adenine nucleotidetranslocator isoforms in mammaliantissues and during muscle cell differ-entiation. J Biol Chem 1992;267:14592–7.

8. Levy SE, Chen Y, Graham BH, Wal-lace DC. Expression and sequenceanalysis of the mouse adenine nucleo-tide translocase 1 and 2 genes. Gene2000;254:57–66.

9. Palmieri L, Alberio S, Pisano I, et al.Complete loss-of-function of theheart/muscle-specific adenine nucleo-tide translocator is associated with mi-tochondrial myopathy and cardiomy-opathy. Hum Mol Genet 2005;14:3079–88.

0. Schultheiss HP, Bolte HD. Immuno-logical analysis of auto-antibodiesagainst the adenine nucleotide trans-locator in dilated cardiomyopathy. JMol Cell Cardiol 1985;17:603–17.

1. Narula J, Haider N, Virmani R, et al.Apoptosis in myocytes in end-stage

heart failure. N Engl J Med 1996;335:1182–9.

12. Narula J, Haider N, Arbustini E,Chandrashekhar Y. Mechanisms ofdisease: apoptosis in heart failure—seeing hope in death. Nat Clin PractCard 2006;3:681–8.

13. Graham BH, Waymire KG, CottrellB, et al. A mouse model for mito-chondrial myopathy and cardiomyop-athy resulting from a deficiency in theheart/skeletal muscle isoform of theadenine nucleotide translocator. NatGenet 1997;16:226–34.

14. Esposito LA, Melov S, Panov A, Cot-trell B, Wallace DC. Mitochondrialdisease in mouse results in increasedoxidative stress. Proc Natl Acad SciU S A 1999;96:4820–5.

15. Murdock D, Boone BE, Esposito LA,Wallace DC. Up-regulation of nuclearand mitochondrial genes in the skele-tal muscle of mice lacking the heart/muscle isoform of the adenine nucle-otide translocator. J Biol Chem 1999;274:14429–33.

16. Subramaniam V, Golik P, MurdockDG, et al. MITOCHIP assessment ofdifferential gene expression in theskeletal muscle of Ant1 knockoutmice: coordinate regulation of OX-PHOS, antioxidant, and apoptoticgenes. Biochim Biophys Acta 2008;1777:666–75.

17. Pirat B, Khoury DS, Hartley CJ, et al.A novel feature-tracking echocardio-graphic method for the quantitation ofregional myocardial function: valida-tion in an animal model of ischemia-reperfusion. J Am Coll Cardiol 2008;51:651–9.

18. Hanley JA, McNeil BJ. The meaningand use of the area under a receiveroperating characteristic (ROC) curve.Radiology 1982;143:29–36.

19. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areasunder two or more correlated receiveroperating characteristic curves: a non-parametric approach. Biometrics1988;44:837–45.

20. Takeuchi M, Borden WB, Nakai H,et al. Reduced and delayed untwistingof the left ventricle in patients withhypertension and left ventricular hy-pertrophy: a study using two-dimensional speckle tracking imaging.
Eur Heart J 2007;28:2756–62. m
1. Kaukonen J, Juselius JK, Tiranti V, etal. Role of adenine nucleotide translo-cator 1 in mtDNA maintenance. Sci-ence 2000;289:782–5.

2. Napoli L, Bordoni A, Zeviani M, etal. A novel missense adenine nucleo-tide translocator-1 gene mutation in aGreek adPEO family. Neurology2001;57:2295–8.

3. Yin H, Stahl JS, Andrade FH, et al.Eliminating the Ant1 isoform pro-duces a mouse with CPEO pathologybut normal ocular movement. InvestigOphthal Vis Sci 2005;46:4555–62.

4. Brand MD, Pakay JL, Ocloo A, et al.The basal proton conductance of mi-tochondria depends on adenine nucle-otide translocase content. Biochem J2005;392:353–62.

5. Narula J, Pandey P, Arbustini E, et al.Apoptosis in heart failure: release ofcytochrome c from mitochondria andactivation of caspase-3 in human car-diomyopathy. Proc Natl Acad SciU S A 1999;96:8144–9.

6. Wencker D, Chandra M, Nguyen K,et al. A mechanistic role for cardiacmyocyte apoptosis in heart failure.J Clin Invest 2003;111:1497–504.

7. Movassagh M, Foo RS. Simplifiedapoptotic cascades. Heart Fail Rev2008;13:111–9.

8. Angelin A, Tiepolo T, Sabatelli P, etal. Mitochondrial dysfunction in thepathogenesis of Ullrich congenitalmuscular dystrophy and prospectivetherapy with cyclosporins. Proc NatlAcad Sci U S A 2007;104:991–6.

9. Merlini L, Angelin A, Tiepolo T, etal. Cyclosporin A corrects mitochon-drial dysfunction and muscle apoptosisin patients with collagen VI myopa-thies. Proc Natl Acad Sci U S A2008;105:5225–9.

0. Flierl A, Chen Y, Coskun PE, Sam-ulski RJ, Wallace DC Adeno-associated virus-mediated gene trans-fer of the heart/muscle adeninenucleotide translocator (ANT) inmouse. Gene Ther 2005;12:570–8.

ey Words: adenine nucleotideranslocator y aging y apoptosis

cardiomyopathy y
itochondria.

adenine nucleotide translocase 1 deficiency results in dilated … · 2016-11-13 · oxidative...

Documents