Mice with the R176Q cardiac ryanodine receptormutation exhibit catecholamine-induced ventriculartachycardia and cardiomyopathyPrince J. Kannankeril*†, Brett M. Mitchell†‡, Sanjeewa A. Goonasekera†§, Mihail G. Chelu‡, Wei Zhang¶, Subeena Sood‡,Debra L. Kearney�, Cristina I. Danila‡, Mariella De Biasi**, Xander H. T. Wehrens‡††, Robia G. Pautler‡, Dan M. Roden¶,George E. Taffet††, Robert T. Dirksen§, Mark E. Anderson‡‡, and Susan L. Hamilton‡§§

Departments of *Pediatrics and ¶Medicine and Pharmacology, Vanderbilt University School of Medicine, Nashville, TN 37232; Departments of ‡MolecularPhysiology and Biophysics, �Pathology, **Neuroscience, and ††Medicine, Baylor College of Medicine, Houston, TX 77030; §Department of Pharmacology andPhysiology, University of Rochester Medical Center, Rochester, NY 14627; and ‡‡Departments of Medicine and Physiology and Biophysics, University of IowaRoy J. and Lucille A. Carver College of Medicine, Iowa City, IA 52242

Edited by Andrew R. Marks, Columbia University College of Physicians and Surgeons, New York, NY, and approved June 12, 2006 (received for reviewJanuary 11, 2006)

Mutations in the cardiac ryanodine receptor 2 (RyR2) have beenassociated with catecholaminergic polymorphic ventricular tachy-cardia and a form of arrhythmogenic right ventricular dysplasia. Tostudy the relationship between RyR2 function and these pheno-types, we developed knockin mice with the human disease-asso-ciated RyR2 mutation R176Q. Histologic analysis of hearts fromRyR2R176Q/� mice revealed no evidence of fibrofatty infiltration orstructural abnormalities characteristic of arrhythmogenic rightventricular dysplasia, but right ventricular end-diastolic volumewas decreased in RyR2R176Q/� mice compared with controls, indi-cating subtle functional impairment due to the presence of a singlemutant allele. Ventricular tachycardia (VT) was observed aftercaffeine and epinephrine injection in RyR2R176Q/�, but not in WT,mice. Intracardiac electrophysiology studies with programmedstimulation also elicited VT in RyR2R176Q/� mice. Isoproterenoladministration during programmed stimulation increased both thenumber and duration of VT episodes in RyR2R176Q/� mice, but notin controls. Isolated cardiomyocytes from RyR2R176Q/� mice exhib-ited a higher incidence of spontaneous Ca2� oscillations in theabsence and presence of isoproterenol compared with controls.Our results suggest that the R176Q mutation in RyR2 predisposesthe heart to catecholamine-induced oscillatory calcium-releaseevents that trigger a calcium-dependent ventricular arrhythmia.

arrhythmogenic right ventricular dysplasia � catecholaminergicpolymorphic ventricular tachycardia � calcium-release channel

The cardiac ryanodine receptor 2 (RyR2) regulates calciumrelease from the sarcoplasmic reticulum in cardiomyocytes (1).

Two inherited arrhythmogenic syndromes have been linked tomutations in RyR2, arrhythmogenic right ventricular dysplasia(ARVD) and catecholaminergic polymorphic ventricular tachycar-dia (CPVT) (2, 3). ARVD and CPVT are both characterized byventricular arrhythmias and a high rate of juvenile sudden death.Patients with CPVT exhibit catecholamine-induced bidirectionalventricular tachycardia (VT) in the setting of a structurally normalheart, whereas patients with ARVD exhibit progressive fibrofattyreplacement of the right ventricular myocardium in addition topolymorphic VT. ARVD arising from RyR2 mutations (ARVD2)is typically associated with exercise-induced ventricular arrhythmiasand relatively mild structural abnormalities compared with otherforms of ARVD and, in some ways, mimics the CPVT phenotype.In fact, the diagnosis of ARVD2 in patients with RyR2 mutationsis controversial because of the differences in degree of cardiacstructural abnormalities between ARVD2 and other forms ofARVD (4).

Disease-causing mutations in RyR2 and the skeletal muscleisoform RyR1 cluster in three highly conserved regions: a cytosolicN-terminal region, a cytosolic central region, and a C-terminal

portion containing the transmembrane and pore regions of thechannel (5, 6). Multiple mutations in RyR2 have been reported inpatients with ARVD2 (7). Families with the RyR2 R176Q mutationalso harbor a second RyR2 mutation, T2504M. The functionalconsequences of these mutations, both individually and in combi-nation, have been studied in vitro and reveal a ‘‘gain of function’’represented by an increased probability of channel opening (8–10).Despite similar in vitro effects, the respective contribution of eachindividual mutation to the overall phenotype is unknown.

The R176Q mutation in RyR2 corresponds to the R163C muta-tion in RyR1, which causes malignant hyperthermia and central coredisease, suggesting that the R176Q mutation alone may be suffi-cient to alter RyR function and elicit a cardiac phenotype. Wedeveloped mice with the R176Q RyR2 mutation to investigate therole of this mutation in the development of cardiomyopathy andarrhythmic susceptibility to catecholamines. By using homologousrecombination in ES cells, the R176Q point mutation in RyR2 wasengineered in mice by using a knockin strategy (Fig. 1). BecauseARVD and CPVT are autosomal dominant disorders, heterozy-gous (RyR2R176Q/�) mice were studied. RyR2R176Q/� mice exhibitedreduced right ventricular end-diastolic volumes but lacked struc-tural abnormalities of the right ventricle. RyR2R176Q/� mice alsodemonstrated VT after injection of caffeine and epinephrine andwith programmed ventricular stimulation combined with the �-ad-renergic receptor agonist isoproterenol. Additionally, individualcardiomyocytes from RyR2R176Q/� mice showed an increased inci-dence of spontaneous Ca2� oscillations. These findings identify theR176Q RyR2 mutation as sufficient to cause cardiac dysfunctionand catecholamine-dependent arrhythmias and support an overlapbetween ARVD2 and CPVT due to mutations in RyR2.

ResultsFunction, but Not Structure, Is Altered in RyR2R176Q/� Mouse Hearts.Histologic studies revealed no indications of fibrofatty infiltration orfibrosis in hearts from young (16–22 weeks of age) or old (51–57weeks of age) RyR2R176Q/� or WT mice (Fig. 7, which is publishedas supporting information on the PNAS web site; n � 3–7 for eachgroup). Statistically, there were no differences in left or right

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: ARVD, arrhythmogenic right ventricular dysplasia; CPVT, catecholaminergicpolymorphic ventricular tachycardia; RyR 2, ryanodine receptor 2; VT, ventricular tachy-cardia.

†P.J.K., B.M.M., and S.A.G. contributed equally to this work.

§§To whom correspondence should be addressed at: Department of Molecular Physiologyand Biophysics, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030. E-mail:susanh@bcm.tmc.edu.

© 2006 by The National Academy of Sciences of the USA

www.pnas.org�cgi�doi�10.1073�pnas.0600268103 PNAS � August 8, 2006 � vol. 103 � no. 32 � 12179–12184

PHYS

IOLO

GY

ventricular volumes of histological slices from RyR2R176Q/� or WTmice as measured by planimetry (data not shown). There were nodifferences in heart rate under isoflurane during the Dopplerultrasound studies: RyR2R176Q/�, 359 � 113 beats per min vs. WT,383 � 54 beats per min; P � 0.05. However, systolic function wasmodestly reduced in RyR2R176Q/� mice, as demonstrated by de-creased peak aortic velocity: RyR2R176Q/�, 92 � 15 cm�s vs. WT,114 � 15 cm�s; P � 0.05 and mean aortic velocity: RyR2R176Q/�,21 � 5 cm�s vs. WT, 28 � 5 cm�s, P � 0.05 (Fig. 8, which ispublished as supporting information on the PNAS web site).

Using MRI, we found no differences in left or right ventricularend-systolic volumes in 8-week-old RyR2R176Q/� mice comparedwith WT (Fig. 2). However, right ventricular end-diastolic volumewas decreased significantly in RyR2R176Q/� mice: RyR2R176Q/�,25.5 � 1.6 mm3 vs. WT, 31.9 � 1.3 mm3; P � 0.05. There was a trendtoward lower left ventricular end-diastolic volumes in RyR2R176Q/�

mice compared with controls, although this trend did not reachsignificance: RyR2R176Q/�, 41.8 � 2.2 mm3 vs. WT, 47.9 � 1.7 mm3;P � 0.056.

Right ventricular function was assessed in vivo by measurementof right ventricular chamber pressure–volume relationships.RyR2R176Q/� mice (n � 7) exhibited smaller end-diastolic volumes:RyR2R176Q/�, 26.4 � 0.8 mm3 compared with WT (n � 8) litter-mates: WT, 31.4 � 0.9 mm3; P � 0.05, and these values were almostidentical to those obtained by MRI (Fig. 3). Moreover, RyR2R176Q/�

mice had higher right ventricular end-diastolic pressures compared

with WT mice: RyR2R176Q/�, 2.6 � 0.3 mmHg (1 mmHg � 133 Pa)vs. WT, 1.6 � 0.2 mmHg; P � 0.05. RyR2R176Q/� mice also exhibitednonsignificant trends toward decreased right ventricular strokevolume: RyR2R176Q/�, 4.1 � 0.4 �l vs. WT, 4.9 � 0.6 �l; P � 0.25and stroke work: RyR2R176Q/�, 73.2 � 12.9 mm Hg��l�min�1 vs. WT,83.8 � 13.9 mm Hg��l�min�1; P � 0.27 compared with WT mice.The lower end-diastolic volume and higher end-diastolic pressureindicate that RyR2R176Q/� mice exhibit restrictive ventricular filling.

Isoproterenol Induces Spontaneous Ventricular Ectopy only inRyR2R176Q/� Mice. Nine unanesthetized mice (four RyR2R176Q/� andfive WT) were studied with ECG telemetry. Spontaneous arrhyth-mias were not observed in any animal before isoproterenol, andECG intervals in RyR2R176Q/� and WT mice were similar at baseline(Table 1). Multiple premature ventricular beats were observed inRyR2R176Q/� mice after isoproterenol (Fig. 4A), whereas no ven-tricular ectopy was observed in WT mice (Fig. 4B). Heart rateresponses to isoproterenol were somewhat blunted in RyR2R176Q/�

mice compared with WT mice (RyR2R176Q/�, 643 � 16 vs. WT,676 � 19; P � 0.05). No sustained ventricular arrhythmias wereobserved before or after isoproterenol in any mice.

VT Induction in RyR2R176Q/� Mice. Eight animals (four RyR2R176Q/�

and four WT) completed the electrophysiologic study protocol.Programmed electrical stimulation induced multiple short episodes(�0.5 s) of VT in both groups at baseline (Fig. 4C), but low-dose

Fig. 1. Targeted engineering of R716Q mutation in the mouse RyR2 locus. (A) The R176Q mutation along with a new RsrII site was introduced in the RyR2 gene.(pGK-Neo�Tet and pCM-TK cassettes were used as positive and negative selection markers, respectively. Lox P sites are indicated as black triangles.) (B)Allele-specific PCR analysis and restriction digestion. A 765-bp fragment was amplified by PCR, and the product was digested with RsrII. WT band, 765 bp; mutantband, 410 bp and 355 bp. (C) Sequencing of an amplified DNA fragment from a correctly targeted ES cell clone shows the presence of the R176Q mutation andthe new RsrII restriction site on one allele.

12180 � www.pnas.org�cgi�doi�10.1073�pnas.0600268103 Kannankeril et al.

isoproterenol increased the number and duration of VT episodesonly in RyR2R176Q/� mice (Fig. 4 D and E). The average duration ofVT was strikingly increased in RyR2R176Q/� mice after isoproterenolcompared with baseline (baseline, 0.5 � 0.4 s vs. isoproterenol,5.1 � 4.8 s; P � 0.05) and to WT mice in either state (baseline, 0.3 �0.1 s, P � 0.05; isoproterenol, 0.5 � 0.8 s, P � 0.05, Fig. 4F). Therewere no differences in electrophysiologic intervals or refractoryperiods before or after isoproterenol in RyR2R176Q/� mice com-pared with WT (Table 2). Thus, even a low dose of isoproterenol,insufficient to cause a significant increase in heart rate, resulted insignificantly more and longer episodes of VT in RyR2R176Q/� mice.

To further examine the effect of adrenergic agonists on VT inRyR2R176Q/� mice, unanesthetized RyR2R176Q/� and WT mice wereinjected with caffeine (120 mg�kg of body weight) and epinephrine(2 mg�kg of body weight). Two of three RyR2R176Q/� mice experi-enced ventricular ectopy after injection, with one having bidirec-tional VT (Fig. 5). Ventricular ectopy was not evident in WT miceeither before or during acute caffeine�epinephrine treatment.

Increased Incidence of Spontaneous Ca2� Oscillations in Single Car-diomyocytes from RyR2R176Q/� Mice. One obvious possibility is thatmutant RyR2 channels display altered control of Ca2� releaseduring excitation–contraction. To test this possibility, we measured

electrically evoked Ca2� transients in the absence and presence ofisoproterenol (100 nM) in indo-1 AM-loaded ventricular cardio-myocytes isolated from WT and RyR2R176Q/� mice. Under controlconditions, resting Ca2� levels, the magnitude of electrically evoked(0.5 Hz) Ca2� release and maximal response to 10 mM caffeinewere comparable (Table 3, which is published as supporting infor-mation on the PNAS web site). In addition, stimulation with 100 nMisoproterenol resulted in a similar significant increase in the mag-nitude of electrically evoked Ca2� release (Fig. 6 and Table 3).However, a significantly higher incidence of spontaneous, nontrig-gered Ca2� oscillations was observed in RyR2R176Q/� cardiomyo-cytes compared with WT myocytes under both control (14.5% vs.34.7%) and isoproterenol-stimulated conditions (26.3% vs. 57.4%)(Fig. 6; and see Fig. 9C, which is published as supporting informa-tion on the PNAS web site). In the presence of isoproterenol (100nM), RyR2R176Q/� cardiomyocytes exhibited a broad spectrum ofdifferent types of calcium transients, including normal activity (Fig.9A), a slowing in the decay of electrically evoked transients (Fig.9B), nonevoked transients during diastole exhibiting normal kinet-ics (Figs. 6B and 9C), and evoked transients exhibiting prolonged,plateau-like durations (Fig. 9D).

DiscussionA common phenotype in CPVT and ARVD2 patients with muta-tions in RyR2 is the presence of ventricular arrhythmias that canlead to sudden cardiac death (11, 12). Patients with ARVD2 thatharbor the R176Q mutation in RyR2 also possess a second RyR2

Fig. 2. MRI reveals decreased right ventricular end-diastolic volume inRyR2R176Q/� mice. (A) Representative end-diastolic MRI images for two WT andtwo littermate RyR2R176Q/� mice. (B) Right ventricular end-diastolic volumewas decreased in RyR2R176Q/� mice (n � 6) compared with WT mice (n � 6).Error bars indicate 1 SEM. *, P � 0.05 vs. WT. LV, left ventricular; RV, rightventricular; ED, end diastole; ES, end systole.

Fig. 3. Pressure–volume relationships demonstrate right ventricular dia-stolic dysfunction in RyR2R176Q/� mice. (Upper) End-diastolic pressures wereincreased in RyR2R176Q/� mice (n � 7) compared with WT mice (n � 8). Arrow(Upper Right) denotes the increase in right ventricular diastolic pressure inRyR2R176Q/� mice. (Lower) Pressure–volume loops obtained during occlusion ofthe vena cava demonstrate decreased volumes in the right ventricles ofRyR2R176Q/� mice.

Table 1. ECG telemetry data in unanesthetized ambulatory mice

Baseline Isoproterenol

HR, bpm PR, msQRS,ms HR, bpm PR, ms

QRS,ms

RyR2R176Q/� (n � 4) 593 � 44 32 � 1 14 � 2 643 � 16 29 � 1 15 � 1WT (n � 5) 606 � 52 32 � 2 15 � 2 676 � 19 31 � 1 15 � 1P 0.71 0.61 0.53 0.04 0.16 0.64

Values are mean � 1 SD. HR, heart rate; bpm, beats per min; PR, PR interval.

Kannankeril et al. PNAS � August 8, 2006 � vol. 103 � no. 32 � 12181

PHYS

IOLO

GY

mutation on the same allele, T2504M (2). In vitro experimentsrevealed that the R176Q and T2504M mutations, alone or incombination, resulted in altered RyR2 function (8–10). Neverthe-less, it was unknown whether the R176Q RyR2 mutation alone issufficient to cause an increased susceptibility to cardiomyopathyand ventricular arrhythmias in intact animals. Our results demon-strate catecholamine-induced VT in RyR2R176Q/� mice that issimilar to that observed in patients with ARVD2 and CPVT.Furthermore, altered cardiac function was evident in ourRyRR176Q/� mice, mimicking findings in ARVD2. However, fibro-fatty infiltration or fibrosis was not apparent in the right ventriclesof RyR2R176Q/� mice. We conclude that the R176Q RyR2 mutationalone is sufficient to reduce cardiac function and enhance theincidence of adrenergic-dependent and independent spontaneousCa2� oscillations that contribute to catecholamine-induced ventric-ular arrhythmias.

Although there were no structural abnormalities in hearts ofRyR2R176Q/� mice up to �1 year of age, our data indicate that theRyR2 R176Q mutation contributes to a preclinical cardiomyopathy.This study has measured right ventricular pressure–volume curvesin the in vivo mouse heart. The decreased compliance of the right

ventricle and decreased systolic function of the left ventricle inRyR2R176Q/� mice is most likely caused by a RyR2 calcium leak,leading to increased ventricular myocardial tone. Because ARVDis a progressive disease, it is possible that the modest cardiacdysfunction in RyRR176Q/� mice may become severe with age.Meticulous evaluation of ventricular function in humans with RyR2mutations may identify a mild cardiomyopathy in those felt to havestructurally normal hearts.

The modest cardiomyopathy seen in our RyRR176Q/� mice may bemore indicative of CPVT rather than ARVD2. It has been postu-lated that minor cardiac structural abnormalities may be evident inCPVT, but patients that have RyR2 mutations and structuralabnormalities are diagnosed as having ARVD2. Although theT2504M RyR2 mutation through unknown mechanisms may serveto further exacerbate this phenotype, our data establish that theR176Q RyR2 mutation is sufficient to drive catecholamine-sensitivearrhythmias and produce a restrictive ventricular filling in theabsence of gross structural abnormalities. Our findings also rein-force an emerging concept that ‘‘pure’’ ion channel diseases likeCPVT or Brugada syndrome (caused by a loss of sodium channelfunction) may be associated with minor abnormalities of cardiacstructure or contractile function (13, 14).

Recently, Cerrone et al. (15) examined mice with the R4496Cmutation in RyR2, which underlies CPVT. These mice did notdisplay any structural abnormalities but demonstrated bidirectionalVT in response to adrenergic stress. Cardiac function was notassessed in these mice. We observed bidirectional VT characteristicof CPVT in a RyRR176Q/� mouse after injection of caffeine andepinephrine. However, the VT episodes induced with programmedstimulation after isoproterenol were not bidirectional but, rather,polymorphic, more similar to those seen in patients with ARVD2.The applicability of our findings of ventricular arrhythmias inRyRR176Q/� mice to arrhythmias in humans needs further research.Nonetheless, ARVD2 and CPVT, which both arise from RyR2mutations, may have significant clinical overlap or may representsubtle but distinct variations of a single disorder.

Fig. 4. Isoproterenol provokes ventricular arrhythmias only in RyR2R176Q/�

mice. (A) Isoproterenol administration (100 �g i.p.) revealed ventricular tri-geminy (every third beat is a premature ventricular beat) in a RyR2R176Q/�

mouse. (B) No arrhythmias were induced after isoproterenol administration inWT mice. Surface ECG (lead I) during programmed ventricular stimulationshowing a short episode of inducible VT at baseline (C) and a longer VTepisode after administration of isoproterenol (D) in an RyR2R176Q/� mouse. (E)Isoproterenol administration during programmed stimulation increased thenumber and duration of inducible VT episodes in RyR2R176Q/� mice only. (F)Average duration of VT episodes before and after isoproterenol. Error barsindicate 1 SEM. *, P � 0.05 baseline vs. isoproterenol.

Fig. 5. Caffeine and epinephrine injection induce VT in RyR2R176Q/� miceonly. Tracings in a representative RyR2R176Q/� mouse 5 min after injection ofcaffeine and epinephrine show sinus rhythm (A), ventricular bigeminy (B), andbidirectional VT (C). In B, the upright sinus beats (denoted with asterisks)alternate with premature ventricular beats with an inverted and wider QRScomplex. In C, there are two alternating morphologies of QRS complexes: aninverted QRS similar to the ventricular beats seen in B and a wide QRSmorphology distinct from the previously observed sinus beats.

Table 2. Electrophysiologic data from intracardiac studies in anesthetized mice

Baseline Isoproterenol

RR, ms AH, ms QT, msVERP,

ms RR, ms AH, ms QT, msVERP,

ms

RyR2R176Q/� (n � 4) 119 � 21 27 � 5 51 � 19 36 � 7 123 � 18 29 � 8 49 � 11 31 � 4WT (n � 4) 146 � 46 39 � 15 55 � 10 46 � 19 162 � 54 39 � 13 60 � 12 44 � 13P 0.27 0.13 0.67 0.38 0.23 0.27 0.21 0.11

Values are mean � 1 SD. RR, RR interval; AH, AH interval; QT, QT interval; VERP, ventricular effective refractory period.

12182 � www.pnas.org�cgi�doi�10.1073�pnas.0600268103 Kannankeril et al.

Our results suggest that VT in RyR2R176Q/� mice arises from anincrease in catecholamine-induced spontaneous calcium releasethat likely results in the activation of a Ca2�-dependent membraneconductance (e.g., Na��Ca2� exchange) and subsequent genera-tion of arrhythmogenic early and delayed after depolarizations.This arrhythmogenic mechanism is analogous to that which issuggested to occur upon sarcoplasmic reticulum calcium overloadduring digitalis toxicity (16, 17) and in CPVT that results frommutations in the cardiac isoform of calsequestrin (18). The twoRyR2 mutations (R176Q and T2504M) have been studied alone andin combination in vitro (9). Alone, each mutation augmented peakCa2� release, but demonstrated different Ca2�-release profiles.R176Q had a slower rate of Ca2� release compared with T2504M,suggesting a complex effect of the double mutation (9). Combined,the mutations increased the sensitivity of Ca2� release in responseto caffeine and lumenal Ca2� [store overload-induced Ca2� release(SOICR)] and augmented Ca2� release compared with WTrecombinant RyR2 (8, 10). Our findings in isolated single car-diomyocytes from adult RyR2R176Q/� mice are consistent with theSOICR mechanism seen in cells expressing RyR2 containingboth mutations.

An alternative mechanism of arrhythmogenesis may result fromcatecholamine-induced hyperphosphorylation of RyR2 at Ser-2809, which dissociates FKBP12.6 from the channel, causing adiastolic calcium leak (19, 20). To determine whether this leakunderlies the abnormal calcium release in RyR2R176Q/� cardiomy-

ocytes, we measured RyR2 Ser-2809 phosphorylation in heartsfrom RyR2R176Q/� and WT mice and also after isoproterenoltreatment (100 �g i.p.). Although isoproterenol treatment in-creased RyR2 Ser-2809 phosphorylation, there were no differencesbetween untreated and isoproterenol-treated RyR2R176Q/� and WTmice (Fig. 10, which is published as supporting information on thePNAS web site).

We did not observe any histological changes in hearts fromRyR2R176Q/� mice up to �1 year old; however the R176Q RyR2mutation was sufficient to alter intracellular calcium dynamics,impair ventricular function, and increase susceptibility to cate-cholamine-induced VT. The phenotype of RyR2R176Q/� mice doesnot precisely correspond to either the CPVT or the ARVD2phenotypes seen in humans, but demonstrates that the R176Q RyR2mutation is critical and, possibly, sufficient to confer key aspects ofthe phenotypes of both diseases and also indicates that these twodiseases may have significant overlap.

Materials and MethodsGeneration of Knockin Mice with the RyR2 R176Q Mutation. Agenomic clone containing exons 7 and 8 of the mouse RyR2 wasisolated from a 129�SvJ �KO-1 library by using the recombinationcloning system in ref. 21. The R176Q mutation, along with a newRsrII restriction site, was introduced into exon 8 of the RyR2 gene.A cassette containing a lox P-flanked NeoR gene expressed from thephosphoglycerate kinase promoter (PGK-NeoR) and a TetR genewas cloned in the unique BlpI site from intron 8 of our genomicclone to obtain the final targeting vector (RyR2R176QNeo). Thetargeting vector was linearized with PmeI and electroporated intoAB2.2 129Sv�J ES cells. DNA was isolated from G418 and gan-cyclovir double-resistant clones and subjected to Southern blotanalysis. A clone found to include the homologous targeted inte-grand was verified by RsrII restriction digestion and direct sequenc-ing and injected into C57BL�6 blastocysts, giving rise to germ-linetransmission resulting in F1 RyR2R176QNeo/� mice. These mice weremated onto Tg(EIIA-Cre) mice (a gift from Graeme Mardon, BaylorCollege of Medicine) to remove the floxed PGK-NeoR�TetR cas-sette. RyR2R176Q/� mice were backcrossed three times on C57�Bl6background. For genotyping, genomic DNA was prepared fromtails and subjected to PCR and digestion with RsrII.

Histology. Hearts from male RyR2R176Q/� mice (n � 7) and WTcontrol littermates (n � 7), aged 16–22 weeks or 51–57 weeks, werefixed with 10% formalin. The ventricular component of each heartwas divided into three cross-sections, submitted in a single paraffinblock, and processed by routine histologic techniques. A complete,full-circumferential section of the middle slice of the ventricles,which consistently included the midportion of the two left ventric-ular papillary muscles, was selected for morphometric analysis. Insome hearts, partial collapse of the right ventricular freewallnecessitated realignment of the collapsed segment before analysis.In these cases, the contour of the affected freewall segment wasredrawn. Measurements of the individual areas of the left and rightventricular chambers from each heart were made by using acomputer-assisted digitizing system with software from Bioquant(Nashville, TN).

Noninvasive Cardiac Function. Cardiac function was analyzed byusing Doppler ultrasound (Fig. 8), MRI, and cardiac catheter-ization. For MRI, heterozygous RyR2R176Q/� (n � 6) and WT(n � 6) mice were initially anesthetized with 4–5% isoflurane(mixed with oxygen) and maintained with 1–2% isofluraneduring imaging. An animal-monitoring system (SA Instruments,Stony Brook, NY) was used to monitor the mouse’s ECG,respiratory rate, and body temperature. Respiratory- and car-diac-gated images were acquired at end-diastole and end-systoleby using a 9.4T, Bruker Avance, 21-cm-bore horizontal scanner.The imaging parameters to acquire cardiac- and respiratory-

Fig. 6. Increased incidence of isoproterenol-induced, nonevoked Ca2� os-cillations in RyR2R176Q/� ventricular myocytes. Representative Ca2� recordingsin the presence of 100 nM isoproterenol (Iso) obtained from indo-1 AM-loaded WT (A) and RyR2R176Q/� (B) single adult ventricular myocytes. A 10 mMcaffeine bolus was applied at the end of the trace to assess sarcoplasmicreticulum calcium content. An expanded time scale of a portion of theelectrically evoked transients is shown below each trace. Arrowheads depictthe delivery of supramaximal extracellular electrical stimuli. Prominent non-evoked Ca2� transients (asterisks) were also observed in the RyR2R176Q/�

myocytes. (C) Average incidence of nonevoked calcium oscillations in WT andRyR2R176Q/� myocytes in the presence and absence of 100 nM isoproterenol(ISO).

Kannankeril et al. PNAS � August 8, 2006 � vol. 103 � no. 32 � 12183

PHYS

IOLO

GY

gated spin echo images were as follows: repetition time (TR), 800ms; echo time (TE), 10 ms; field of view, 3.0 cm; number of slices,10; slice thickness, 1.0 mm; matrix, 256 � 256; and number ofaverages, 1. The multislice scan was performed in the axialorientation to visualize the left and right ventricles, and datawere analyzed by using Amira 3D image processing software(Mercury Computer Systems, Chelmsford, MA). The areasrepresenting the left and right ventricular space in each slicewere summated in the amount of pixels. The volume size of thespace was calculated by using the known volume size of eachpixel (30.0 mm�256 � 30.0 mm�256 � 1.0 mm).

For right ventricular catheterization, RyR2R176Q/� (n � 7) andWT (n � 8) mice were anesthetized with 1.5% isoflurane (mixedwith oxygen) and maintained at 37°C by a heating pad during theexperiment. A 1.4F high-fidelity micromanometer catheter (MillarInstruments) was advanced via the right jugular vein into the rightventricle. Pressure–volume relationships were measured underbaseline conditions and during transient occlusion of the inferiorvena cava. End-diastolic pressure–volume relationships were fittedby using the monoexponential equation. An IOX data acquisitionsystem (Emka Technologies, Falls Church, VA) was used to analyzepressure–volume relationships.

ECG Telemetry. Nine animals (four RyR2R176Q/� and five WT) werestudied with ECG telemetry according to published methods (22,23). Briefly, transmitters (Data Sciences International, St. Paul,MN) were implanted in the abdominal cavity with s.c. electrodes ina lead I configuration. Telemetry was recorded �24–72 h aftersurgery in ambulatory, unanesthetized mice for 30-min intervals atbaseline and after the injection of isoproterenol 100 �g i.p. A subsetof telemetered RyR2R176Q/� and WT mice (n � 3 in each group)were injected with epinephrine (2 mg�kg of body weight i.p.) andcaffeine (120 mg�kg of body weight i.p.) and monitored for 10 min.Data collection was performed by using Dataquest software, andoff-line data analyses were performed by using Physiostat ECGanalysis software, version 3.1 (Data Sciences International).

Intracardiac Electrophysiology Studies. RyR2R176Q/� (n � 4) and WT(n � 4) mice were anesthetized with pentobarbital (0.07 mg�g ofbody weight i.p.). Surface ECG (lead I) signals were obtained froms.c. 27-gauge needles placed in each forelimb. The ECG channelswere amplified (0.1 mV�cm) and filtered between 0.05 and 400 Hz.An octapolar 2F electrode catheter (CIBer cath; NuMED, Hop-kinton, NY) was placed in the right atrium and right ventricle.Bipolar electrogram recordings were obtained from the rightatrium, right ventricle, and His positions. Signals were amplifiedand filtered between 40 and 400 Hz at a 200- to 400-mm�s speed.Bipolar pacing was performed by using a programmable stimulator

(Medtronic 2356) modified by the manufacturer to deliver couplingintervals as short as 10 ms. Pacing threshold (in milliamperes) wasdetermined for each pacing site, and stimulation was performed for1.0- to 2.0-ms pulse width at twice the diastolic-capture threshold.Standard clinical electrophysiologic pacing protocols were used todetermine all basic electrophysiologic parameters. Ventricular ef-fective refractory period was determined at three drive cyclelengths. Single, double, and triple extrastimuli were delivered atthree drive cycle lengths to determine inducibility of VT. Afterbaseline measurements were completed, isoproterenol was admin-istered (100 �g i.p.) and the protocols repeated to assess the effectson conduction and refractoriness.

Intracellular Ca2� Measurements. Single ventricular myocytes wereisolated from the hearts of 8- to 16-week-old WT and RyR2R176Q/�

mice by using standard enzymatic dissociation and plated onlaminin-coated dishes as described by the Alliance for CellularSignaling (AfCS Procedure Protocol ID PP00000 125) (24). Myo-cytes were loaded with 6 �M indo-1 AM (Molecular Probes) for 60min at 37°C and superfused with a Ringer’s solution containing 146mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, and 10 mMHepes (pH 7.4). Indo-1-loaded myocytes were excited at 350 nm byusing a DeltaRam illumination system (Photon Technology Inter-national, Lawrenceville, NJ). Fluorescence emission at 405 and 485nm was collected at 100 Hz by using a photomultiplier detectionsystem and represented as the ratio (R) of F405�F485. For measure-ments of electrically evoked calcium release, myocytes were stim-ulated (8 V, 20 ms, 0.5 Hz for 60 s) by using an extracellularelectrode placed close to the cell of interest in the absence andpresence of 100 nM isoproterenol. A 10 mM caffeine bolus after thepulse train was used to assess sarcoplasmic reticulum Ca2� content.

Statistical Analyses. Electrophysiologic and cardiac function andintracellular Ca2� measurements were compared betweenRyR2R176Q/� and WT animals with Student’s t test. Effects ofisoproterenol on ECG parameters were compared with a paired ttest. A two-tailed P value of �0.05 was considered significant. Forthe data presented in Fig. 6C, statistical significance was deter-mined by using a �2 test. All procedures were approved by therespective institution’s animal care committee.

We thank Lingyun Hu and Fredalina Pieri for technical assistance withthe MRI and telemetry data, respectively. This work was supported, inpart, by U.S. Public Health Service Grants AR41802, HL070250,HL046681, AG17899, HL22512, and AR44657. P.J.K. was supported bya Vanderbilt University School of Medicine Clinician Scientist Award,M.E.A. is an Established Investigator of the American Heart Associa-tion, and X.H.T.W. was supported by an American Heart AssociationScientist Development Grant.

1. Taur, Y. & Frishman, W. H. (2005) Cardiol. Rev. 13, 142–146.2. Tiso, N., Stephan, D. A., Nava, A., Bagattin, A., Devaney, J. M., Stanchi, F., Larderet,

G., Brahmbhatt, B., Brown, K., Bauce, B., et al. (2001) Hum. Mol. Genet. 10, 189–194.3. Priori, S. G., Napolitano, C., Tiso, N., Memmi, M., Vignati, G., Bloise, R., Sorrentino,

V. & Danieli, G. A. (2001) Circulation 103, 196–200.4. Priori, S. G., Napolitano, C. (2005) Circ. Res. 97, 1077–1079.5. Tiso, N., Salamon, M., Bagattin, A., Danieli, G. A., Argenton, F. & Bortolussi, M.

(2002) Biochem. Biophys. Res. Commun. 299, 594–598.6. Brini, M. (2004) Biochem. Biophys. Res. Commun. 322, 1245–1255.7. Bauce, B., Rampazzo, A., Basso, C., Bagattin, A., Daliento, L., Tiso, N., Turrini, P.,

Thiene, G., Danieli, G. A. & Nava, A. (2002) J. Am. Coll. Cardiol. 40, 341–349.8. Thomas, N. L., Lai, F. A. & George, C. H. (2005) Biochem. Biophys. Res. Commun.

331, 231–238.9. Thomas, N. L., George, C. H. & Lai, F. A. (2004) Cardiovasc. Res. 64, 52–60.

10. Jiang, D., Wang, R., Xiao, B., Kong, H., Hunt, D. J., Choi, P., Zhang, L. & Chen, S. R.(2005) Circ. Res. 97, 1173–1181.

11. Bauce, B., Nava, A., Rampazzo, A., Daliento, L., Muriago, M., Basso, C., Thiene, G.& Danieli, G. A. (2000) Am. J. Cardiol. 85, 573–579.

12. Francis, J., Sankar, V., Nair, V. K. & Priori, S. G. (2005) Heart Rhythm 2, 550–554.13. Coronel, R., Casini, S., Koopman, T. T., Wilms-Schopman, F. J., Verkek, A. O., de

Groot, J. R., Bhuiyan, Z., Bezzina, C. R., Veldkamp, M. W., Linnenbank, A. C., etal. (2005) Circulation 112, 2769–2777.

14. Royer, A., van Veen, T. A., Le Bouter, S., Marionneau, C., Griol-Charhbili, V., Leoni,A. L., Steenman, M., van Rijen, H. V., Demolombe, S., Goddard, C. A., et al. (2005)Circulation 111, 1738–1746.

15. Cerrone, M., Colombi, B., Santoro, M., di Barletta, M. R., Scelsi, M., Villani, L.,Napolitano, C. & Priori, S. G. (2005) Circ. Res. 96, e77–e82.

16. Priori, S. G. & Corr, P. B. (1990) Am. J. Physiol. 258, H1796–H1805.17. Hauptman, P. J. & Kelly, R. A. (1999) Circulation 99, 1265–1270.18. Viatchenko-Karpinski, S., Terentyev, D., Gyorke, I., Terentyeva, R., Volpe, P., Priori,

S. G., Napolitano, C., Nori, A., Williams, S. C. & Gyorke, S. (2004) Circ. Res. 94,471–477.

19. Wehrens, X. H., Lehnart, S. E., Huang, F., Vest, J. A., Reiken, S. R., Mohler, P. J.,Sun, J., Guatimosim, S., Song, L. S., Rosemblit, N., et al. (2003) Cell 113, 829–840.

20. Lehnart, S. E., Wehrens, X. H., Laitinen, P. J., Reiken, S. R., Deng, S. X., Cheng,Z., Landry, D. W., Kontula, K., Swan, H. & Marks, A. R. (2004) Circulation 109,3208–3214.

21. Zhang, P., Li, M. Z. & Elledge, S. J. (2002) Nat. Genet. 30, 31–39.22. Wu, Y., Temple, J., Zhang, R., Dzhura, I., Zhang, W., Trimble, R., Roden, D. M.,

Passier, R., Olson, E. N., Colbran, R. J., et al. (2002) Circulation 106, 1288–1293.23. Salas, R., Pieri, F., Fung, B., Dani, J. A., De Biasi, M. (2003) J. Neurosci. 23,

6255–6263.24. Sambrano, G. R., Fraser, I., Han, H., Ni, Y., O’Connell, T., Yan, Z. & Stull, J. T.

(2002) Nature 420, 712–714.

12184 � www.pnas.org�cgi�doi�10.1073�pnas.0600268103 Kannankeril et al.