Differences in cardiac calcium release channel (ryanodine receptor) expression in myocardium from patients with end-stage heart failure caused by ischemic versus dilated cardiomyopathy

A M Brillantes, P Allen, T Takahashi, S Izumo and A R Markscardiomyopathy.

myocardium from patients with end-stage heart failure caused by ischemic versus dilated Differences in cardiac calcium release channel (ryanodine receptor) expression in

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1992 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.71.1.181992;71:18-26Circ Res. 

http://circres.ahajournals.org/content/71/1/18World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

18

Differences in Cardiac Calcium Release Channel(Ryanodine Receptor) Expression in

Myocardium From Patients With End-StageHeart Failure Caused by Ischemic Versus

Dilated Cardiomyopathy

Anne-Marie Brillantes, Paul Allen, Toshiyuki Takahashi, Seigo Izumo, and Andrew R. Marks

The molecular basis for the systolic and diastolic dysfunction characteristic of end-stage heart failure inhumans remains poorly understood. It has been proposed that both abnormal calcium handling anddefects in the contractile apparatus may contribute to the myocardial dysfunction. Two channels, thecalcium release channel (CRC) or ryanodine receptor of the sarcoplasmic reticulum (SR), and the slowcalcium channel or dihydropyridine receptor (DHPR) of the transverse tubule, play key roles in regulatingintracellular calcium concentration and in excitation-contraction (E-C) coupling in the heart. The DHPRserves as the voltage sensor and plasma membrane calcium channel resulting in activation of the CRCduring E-C coupling in heart muscle. In this study, we investigated the levels ofCRC expression in severalforms of end-stage heart failure in humans. A cardiac CRC cDNA was cloned from rabbit and used as a

probe for Northern blot analyses to determine mRNA levels in the left ventricles of normal (n=4) andcardiomyopathic (n=34) human hearts from patients undergoing cardiac transplantation. Comparedwith normal patients, patients with ischemic cardiomyopathy (n=18) showed a 28% decrease in CRCmRNA levels (p<0.025) and patients with idiopathic dilated cardiomyopathy (n=14) a nonsignificant12% increase. In these same hearts, a-actin levels were unchanged in end-stage heart failure, as has beenpreviously reported. This is the first report indicating that the expression of the CRC mRNA is abnormalin end-stage human heart failure. The decreased expression of the CRC found specifically in patients withischemic myopathy (but not idiopathic dilated cardiomyopathy) may, in part, explain differences incalcium handling and response to therapy observed among patients with different forms of cardiomyop-athy. Decreased CRC expression could be related to abnormal contractile function in cardiomyopathicmuscle resulting from ischemic insult. (Circulation Research 1992;71:18-26)KEY WoRDs * idiopathic dilated cardiomyopathy

channel * mRNA * ryanodine receptor

TNhe calcium release channel (CRC) or ryanodinereceptor of the sarcoplasmic reticulum (SR) isa large (565-kd) protein that forms the foot

structure associating the SR with the transverse tubule

This manuscript from Harvard Medical School was sent toHarold C. Straus, MD, Consulting Editor, for review by expertreferees, editorial decision, and final disposition.From Brookdale Center for Molecular Biology, Molecular Med-

icine Division and Cardiology Division, Department of Medicine(A.-M.B., A.R.M.), Mount Sinai School of Medicine, New York;Department of Anesthesiology (P.A.), Brigham and Women'sHospital, and Molecular Medicine Unit and Cardiovascular Divi-sion (T.T., S.I.), Beth Israel Hospital, Harvard Medical School,Boston.A.-M.B. is supported by a Medical Student Fellowship from the

Howard Hughes Medical Institute, A.R.M. is a Syntex Scholar.Supported by grants from the NIH, AHA, and the Sara ChaitMemorial Foundation.Address for correspondence: Dr. Andrew R. Marks, Brookdale

Center for Molecular Biology and Molecular Medicine Division,Box 1126, Mount Sinai School of Medicine, One Gustave L. LevyPlace, New York, NY 10029.

Received November 13, 1991; accepted April 10, 1992.

* ischemic cardiomyopathy * calcium release

(T tubule) in cardiac and skeletal muscle. The CRC iscoupled by an as yet unknown mechanism to the dihy-dropyridine receptor (DHPR) of the T tubule duringexcitation-contraction (E-C) coupling in striated mus-cle."2 Both calcium channels are involved in the regu-lation of intracellular calcium during diastolic and sys-tolic cycling in the heart.3 The release and reuptake ofCa21 is also controlled by other calcium-handling pro-teins, including the Ca24-ATPases, Ca>2 channels, andCa2 binding proteins found within the cell.4 It has beenproposed that alterations in the expression, regulation,or function of these proteins may result in diastolic andsystolic dysfunction of the myocardium observed inend-stage heart failure.5 Several studies have docu-mented abnormal calcium homeostasis in the failingmyocardium.6-8 Recent clinical trials have documentedadverse effects on patients with heart failure treatedwith organic calcium channel blockers.9-'2 Previousstudies found that the expression of the SR Ca'-ATPase5"3 and the DHPR'3 were decreased in end-stage human heart failure.

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

Brillantes et al Calcium Release Channel Expression in Heart Failure 19

The skeletal and cardiac isoforms of the CRC havebeen cloned and their primary structure determined inrabbit and human skeletal muscle14-16 4nd in rabbitheart.17 The cardiac CRC was found to be homologousto, but distinct from, the skeletal form (-66% homolo-gy). The skeletal and cardiac CRC cDNAs both encodepolypeptides with a mass of 565 kd. Four of these 565 kdpolypeptides form the 2.3 million Dalton junctionalchannel complex that includes the calcium release chan-nel of the SR and the "foot structure" spanning the gapbetween the SR and the T tubule.18-21 To date, little isknown about the regulation of the expression of theCRC in human heart failure. The goal of the presentstudy was to characterize possible alterations in expres-sion of the major intracellular calcium release channelinvolved in E-C coupling in cardiac muscle from pa-tients with end-stage heart failure. CRC mRNA abun-dance was determined in left ventricular (LV) myocar-dial samples from 34 patients with end-stage heartfailure and four normal persons (controls). Patientswith end-stage heart failure were divided into two majorsubgroups, those with ischemic (ICM) and those withidiopathic dilated cardiomyopathies (IDCM). CRCmRNA levels were significantly decreased in the sub-group with ICM (but not in patients with IDCM). Thisdecrease in CRC expression was not caused by ageneralized decrease in myocardial mRNAs, since thea-actin levels were unchanged, in agreement with aprevious report by Boheler et al.22

Materials and MethodsPatientsWe analyzed data from 34 patients with end-stage

heart failure undergoing cardiac transplant surgery atBrigham and Women's Hospital, Boston, and four con-trol patients free of cardiovascular pathology who wereorgan donors but who had been rejected as cardiactransplant donors. The subjects included 18 patientswith coronary artery disease (ischemic cardiomyopathy[ICM]), two with hypertrophic cardiomyopathy, and 14with idiopathic dilated cardiomyopathy (IDCM). Thisstudy was approved by the Committee for the ProtectionofHuman Subjects from Research Risks at the Brighamand Women's Hospital. Clinical characteristics of thepatients are shown in Table 1. There were no significantdifferences in any of the clinical parameters recorded inthis study between the major subgroups of patients withend-stage heart failure, ICM, and IDCM. At the time oftransplantation, hearts were explanted and quicklyweighed, and a piece (1-2 g) of myocardium was excisedfrom the LV free wall and immediately immersed inliquid nitrogen. LV samples were obtained from areasthat were not macroscopically scarred. All samples weretaken from the anterior free wall of the left ventriclenear the apex of the heart. Care was taken to obtainsamples from the same region of each heart. In caseswhere infarcts in the distribution of the left anteriordescending artery had resulted in scar formation at theapex, samples were obtained from regions as close tothe apex as possible outside the left anterior descendingartery distribution. The regions free of scarring in theICM patients were not chosen randomly, but rather

dilated cardiomyopathic samples or from the closestregion at the apex that was free of scar tissue.

cDNA ProbesThe cardiac CRC cDNA probe used for Northern

blot analysis was cloned from rabbit heart total RNAessentially as previously described.23 Sense and anti-sense primers, CR-4860S (GCAGTGTTTG-GATCCTCTGC) and CR-5440A (GCACAAAGAG-GAATTCGGTGG), respectively, were designed basedon the published sequence for the rabbit heart CRC17(numbers designating each oligonucleotide refer to theposition in the published sequence). First-strand cDNAwas synthesized using 1 ,ug of total rabbit heart RNAand 2 pmol of antisense primer with avian myeloblasto-sis virus reverse transcriptase at 42°C for 1 hour. Fortycycles of polymerase chain reaction with Taq DNApolymerase (Perkin-Elmer Cetus) were then performedon 0.1% of the products of first-strand cDNA synthesiswith 20 pmol of each primer through use of the follow-ing parameters: 1 minute denaturation at 94°C, 2 min-utes annealing at 55°C, and 1 minute extension at 72°C.The polymerase chain reaction products were separatedfrom primers on low-melt agarose gels, and a 580-bpcDNA product was excised, purified, digested withEcoRI and BamHI, and subcloned into pBluescript(Stratagene, La Jolla, Calif.). This cDNA clone,HCRC1, was sequenced with the ABI 370 automatedsequenator using the M13 forward and reverse primersand shown to correspond to nucleotides [5027-5647] ofthe cDNA encoding the rabbit cardiac CRC. The actinprobe was pAC 269 from chicken skeletal muscle a-ac-tin.24 This probe recognizes both muscle specific(smooth, skeletal, and cardiac a-actins) and nonmuscle(f3 and y) actin isoforms. The glyceraldehyde-3-phos-phate-dehydrogenase probe (GAPDH) was a 1.3 kb PstI fragment from clone pUC-GAPDH13 containing theentire coding region and part of the 3' untranslatedregion of the rat GAPDH.25 The SR Ca2' ATPase probewas a 0.7 kb Pst I fragment corresponding to thecarboxyl terminus and the 3' untranslated region, thiscDNA hybridizes specifically to the rabbit cardiac/slowtwitch skeletal muscle SR Ca2' ATPase.26 All probeswere uniformly labeled with random primers usingKlenow and a_p32 dCTP to a specific activity of >109cpm/,g.

RNA Preparation and Northern Blot AnalysisTotal cellular RNA was purified from human ventric-

ular tissue samples through use of the lithium chloride/urea method.27 The RNA was quantitated by spectro-photometry at 260 nm, and the ratio of absorbance at 260nm to that at 280 nm was greater than 1.8 for all samples.Total cellular RNA was size fractionated on formalde-hyde/agarose gels run at 30 mA overnight to provideresolution of high molecular weight mRNAs. RNA trans-fer onto nitrocellulose filters was conducted overnightthrough use of lOx SSC transfer buffer. Gels were

photographed before and after transfer, and nitrocellu-lose filters were photographed after transfer to confirmtransfer efficiency of the 28s and 18s ribosomal RNAs.Using these transfer conditions no residual ribosomalRNAs were detected in the gels after transfer. The filterswere then baked at 80°C in a vacuum oven for 2 hours

were all either in the same region as the normal and and prehybridized ovemight in a buffer containing 1 X

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

20 Circulation Research Vol 71, No 1 July 1992

TABLE 1. Clinical Characteristics of Patients With End-Stage Heart Failure

HW CI PCWP LVEDPPatient Age Sex (g) (1/min/m2) (mm Hg) (mm Hg)

ICM

234S6789101112131415161718

Mean

IDCM1234567891011121314

Mean

474455403350546044505457542450413148

46+10

3518544958532451564414235025

40±+16

MMMMMMMMMMMMMFMMMM

MMMMMMMMMMMMMF

405620577510387380460NANA470460595370NA560NA375460

474+87

470625440NA550NANANA500NA300600490NA

497±+102

1.82.0NANA2.03.51.41.51.92.01.42.41.81.42.01.01.81.6

1.8±0.6

2.51.42.01.41.22.91.72.31.71.51.40.91.81.7

1.7±0.5

2423NANA3414343536283438262428203538

29-+-7

212921311914262232381542S14

24±10

29NANANA4117274037NA2735NANANA223016

29+9

26NA2030NA20302330388

32NA15

25±9

HW, heart weight; CI, cardiac index; PCWP, pulmonary capillary wedge pressure; LVEDP, left ventricular end-diastolic pressure; ICM, ischemic cardiomyopathy; IDCM; idiopathic dilated cardiomyopathy.

Denhardt's solution (0.02% polyvinylpyrrolidone/0.02%Ficoll/0.02% bovine serum albumin), 5 x SSC (lx =0.15M NaCI/0.015 M sodium citrate, pH 7.0), 0.025 Msodium phosphate (pH 7.4), sonicated calf thymus DNA(50 mg/ml), 0.1% SDS, and 50% (vol/vol) formamide.Blots were hybridized with cDNA probes in the samebuffer mixture overnight at 42°C. The cDNA probes foreach individual mRNA species were labeled once, andthe same labeled probe was hybridized to all filterssimultaneously to avoid potential differences in specificactivity of the labeled probes. Filters were washed simul-taneously to a final stringency of 0.2x SSC/0.1% SDS at65°C for 15 minutes. Six individual blots containing allthe samples for this study were hybridized. RNA samplesfor normal, ICM, and IDCM patients were loaded ontoeach gel to avoid potential bias caused by variability inthe efficiency of transfer of a particular gel or region of agel. Each filter was exposed for three different exposuretimes to obtain signals in the linear range for densito-metric analysis of each mRNA species. Exposure was at-80°C on x-ray films (X-OMAT AR, Eastman Kodak,Rochester, N.Y.) with a single intensifying screen. Rel-ative amounts of mRNA were determined by using a

Xerox scanner with MACIMAGE software. Densitometricanalysis was performed using Image 1.36 software on aMacIntosh IIcx that included an algorithm to determinewhether the exposure is in the linear range (a densitysignal <200 indicates exposure in the linear range). Allof the signals used for calculations of mRNA quantitieswere in the linear range. Each densitometric score for theCRC and a-actin mRNAs was then normalized by divid-ing its optical density with the corresponding GAPDHoptical density as an internal control. The mean value forthis ratio for the normal group was set at 1.0. Each CRCdensitometric score was also normalized to a-actin opti-cal density as an additional internal control.

Controls for RNA degradation and the amount ofRNA transferred to each filter included 1) ethidiumbromide visualization of the 28s and 18s ribosomalRNAs on the filters; 2) comparison of Northern blotanalyses with slot blot results (see "Results"); 3) qual-itative analysis of the CRC mRNA signals which weresingle bands without low molecular weight streaking inthe lanes, which would have indicated degradation; 4)control hybridizations to a-actin, and GAPDH probesthat demonstrated that the levels of these messages

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

Calcium Release Channel Expression in Heart Failure 21

were unchanged between normal and cardiomyopathicsamples. To determine that quantitation of mRNAs onNorthern blots was not biased by inefficient transfer ofmRNAs, slot blots were performed on selected RNAsamples (four normal, four ICM, and four IDCM pa-tients). Total RNA was denatured in 50% formamide,7% formaldehyde, and 1x SSC by heating at 65°C for 15minutes followed by cooling on ice. Serial amounts (2.5gig and 5 gg) of each sample examined were loaded induplicate (four slots for each RNA sample) on a slotblot apparatus (Bethesda Research Laboratories) andblotted to nitrocellulose. These filters were hybridizedto CRC and GAPDH cDNA probes as described above.Autoradiograms in which there was a linear decrease insignal corresponding to the amount of RNA loadedwere quantitated by scanning and determining densityof signal as described above. Normalized CRC mRNAlevels for these 12 samples examined by slot blots werecompared with those obtained by Northern blot analy-ses (see "Results").

Statistical AnalysisAll data are expressed as mean+1 SD. Statistical

differences were calculated using the unpaired Stu-dent's t test, and the p<0.05 level was consideredsignificant.

ResultsClinical DataTable 1 summarizes the major clinical characteristics

of the patients with end-stage heart failure. Patientsunderwent clinical evaluations at the Brigham andWomen's Hospital. There was no significant differencebetween the ICM and IDCM groups in terms of meanage (46±10 versus 40±16 years), mean heart weight(474±87 versus 497±102 g), mean cardiac index(1.8±0.6 versus 1.7±0.5 I/minIm2), mean pulmonarycapillary wedge pressures (29+ 7 versus 24+ 10 mm Hg),and mean left ventricular end-diastolic pressure (29±9versus 25±9 mm Hg).

Expression of the Sarcoplasmic Reticulum CalciumRelease Channel and Calcium ATPasemRNAs in Heart Failure

Figure 1 shows representative lanes from Northernblot analyses of human left ventricular RNA hybridizedto a probe specific for the cardiac CRC mRNA. In allRNA samples, the HCRC1 probe hybridized to a singlemRNA species of -16 kb. Nonspecific hybridization(e.g., to ribosomal RNAs) was not observed.

Figure 2 shows Northern blot analyses of representa-tive total RNA samples from human LVs hybridized toprobes for the cardiac CRC, a-actin, and GAPDH.Each probe hybridized to a single mRNA species inhuman myocardial total RNA. The relative decrease inCRC mRNA accumulation in the hearts with ICM isevident. There were no significant differences in thelevels of GAPDH mRNA between normal (1.55±0.25),ICM (1.54±0.34), and IDCM (1.32+0.31) hearts. More-over, ca-actin mRNA levels were not significantly alteredbetween normal (2.12±0.37), ICM (2.0±0.42), andIDCM (1.87±0.38) RNA samples.Table 2 shows the normalized values for CRC and

a-actin mRNA levels in the human LV samples ob-

NL ICM IDCM

28s -

FIGURE 1. Representative Northern blot analysis of CRCmRNAs expressed in human myopathic and normal hearts.Total cellular RNA isolated from normal patients (NL, 20gg) and patients with heart failure due to ischemic cardiomy-opathy (ICM, 30 gg), or idiopathic dilated cardiomyopathy(IDCM, 20 gg) was size fractionated on a formaldehyde-agarose gel and blotted to nitrocellulose. Blots were hybridizedto a-32P-labeled HCRCJ (see "Materials and Methods").This cardiac calcium release channel cDNA probe identifies asingle 16 kb mRNA (28s and 18s ribosomal RNAs are used assize markers). Note that the intensities of the bands in eachlane do not reflect the relative levels of calcium releasechannel mRNA present in the respective heart tissues becausedifferent amounts of RNA were added in each lane tooptimize signals in order to demonstrate that only one mRNAspecies is detected with the probe.

tained by densitometric scanning. The CRC and a-actinmRNA levels were normalized to the GAPDH level foreach sample.

Figure 3 is a summary of the CRC and a-actinexpression data. The CRC mRNA (normalized toGAPDH mRNA levels) was decreased by 28% in thepatients with ICM (n=18, p<00.25) compared withnormal patients and by 46% (p<0.005) compared withthose with IDCM (Table 2 and Figure 2). In the patientswith IDCM (n 14) there was a nonsignificant increasein CRC abundance of 12% compared with normals.Two patients with HCM also showed a decrease in CRCmRNA abundance, but the patient sampling was toosmall to assess the significance of this finding (data notshown). The CRC mRNA (normalized to a-actin) wasdecreased by 28% in the patients with ICM (n =18,p<0.025) compared with normal patients and by 48%(p<0.01) compared with IDCM patients. In the IDCMpatients (n=14) there was a nonsignificant increase inCRC mRNA abundance of 18% compared with normalpatients. We also observed a 44% decrease in normal-ized Ca2'-ATPase mRNA levels in the LVs from ICMpatients (n=18, p<O.00l) compared with normal pa-tients. There was a similar 56% decrease in normalizedCa2'-ATPase mRNA levels in LVs from patients withIDCM (n 14, p<0.001) compared with normal pa-

Brillantes et al

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

22 Circulation Research Vol 71, No 1 July 1992

NL

CRC

1CM

1-:

FIGURE 2. Calcium release channel mRNA lev-els in the left ventricle of normal patients (NL)compared with patients with end-stage heart fail-ure caused by ischemic cardiomyopathy (ICM) oridiopathic dilated cardiomyopathy (IDCM). Rep-resentative Northern analyses from individual pa-tients are shown. In allRNA samples, the HCRC1probe hybridized to a single mRNA species of -16kb, the a-actin probe to a band at -1.6 kb and theGAPDH probe hybridized to an -1.6 kb mRNAspecies. CRC, calcium release channeL

(X-ACTIN

GAPDH

tients. In contrast to the decreases in CRC and Ca2'-ATPase mRNA abundance in the ICM group, a-actinmRNA abundance (normalized to GAPDH) was notsignificantly different among the groups with heartfailure and was unchanged compared with normal pa-tients (Table 2 and Figure 2). To establish that transferof RNA did not bias the results obtained by Northernblot analyses, we performed slot blot analyses of CRCmRNA levels normalized to GAPDH mRNA levels onRNA samples from the normal (n=4), ICM (n=4), andIDCM (n=4) groups. There were no significant differ-ences between the densitometric scores of CRC mRNAlevels (normalized to GAPDH mRNA levels) obtainedby slot blot versus Northern blot analyses (n -12,r=0.995,p<0.01).

DiscussionE-C coupling in the heart involves two calcium chan-

nels, the DHPR located on the T tubule and the CRC ofthe SR (reviewed in Reference 28). In the presentstudy, levels of CRC mRNA in the LVs of patients withend-stage heart failure were shown to be decreased by28% (p<0.025) in those with ICM and increased by12% (p=NS) in those with IDCM compared withnormal patients (Figure 3 and Table 2). Of particularinterest is the observation that CRC mRNA levels weresignificantly decreased in patients with ICM but not inthose with IDCM, indicating a difference in the expres-sion of CRC mRNA between two forms of end-stagecardiomyopathy. The decreased CRC mRNA levels inthe patients with ICM were not caused by a generalizeddecrease in cardiac gene expression or a decrease innumbers of myocytes in the regions sampled, becausethe levels of ca-actin were not altered.

In addition to abnormal expression of the CRC of theSR we have shown that expression of the Ca2'-ATPaseinvolved in SR Ca2' reuptake is also affected duringend-stage heart failure. In agreement with a previousstudy in a similar group of patients5 we found thatexpression of the SR Ca2'-ATPase is decreased. How-ever, in contrast to the CRC, which was decreased onlyin the ICM group, the SR Ca2'-ATPase was decreasedto equivalent levels in both ICM and IDCM groups.Taken together, these data suggest that while reuptakemay be altered in both forms of cardiomyopathy, dimin-ished calcium release may be a feature unique to ICM.

While end-stage cardiomyopathies have generallybeen regarded as a whole in terms of the pathophysiol-ogy of myocardial dysfunction, the current data andother recent studies7,29 provide evidence for potentiallysignificant differences at the molecular level betweentwo forms of cardiomyopathy, both of which result inend-stage heart failure. Previous studies have examinedthe expression of other calcium handling genes in themyocardium in both human end-stage heart failure andin animal models. Expression of the SR Ca2-ATPase isdepressed in animal models of cardiac hypertrophy30 32as well as in human heart failure. To determine theeffects of end-stage heart failure on the expression ofother genes encoding calcium handling proteins in theheart, we had previously examined the abundance of thecardiac DHPR and calsequestrin transcripts in RNAprepared from the patients examined in the currentstudy.13 Compared with normal subjects the DHPRmRNA was decreased by 47%.13 However, in contrast tothe CRC mRNA, no difference was observed betweenthe subgroups with ICM and IDCM in terms of calciumchannel expression.The decreased expression of two calcium channels

involved in E-C coupling in the ICM hearts may berelated to the abnormal calcium handling of myopathichuman cardiac muscle in ischemic heart disease. De-fects in calcium handling with regard to E-C couplingmight be expected to differ between ICM and IDCM,based on the contrasts in regulation of CRC expressionbetween the two forms of cardiomyopathy.

Differences in the function of components of thef3-adrenergic receptor-G protein-adenylate cyclase com-plex have been described in ICM versus IDCM sub-groups.29 For example, 81-adrenergic receptors weredownregulated to a greater extent in dilated versusischemic cardiomyopathies, but P2-adrenergic receptorsshowed more uncoupling from adenylate cyclase in theLVs of ischemic myopathic hearts. This study points outthat while the clinical features of end-stage heart failuremay be similar, irrespective of the etiology of the heartfailure, functional differences at the molecular levelexist.

Regulation of the calcium channels involved in E-Ccoupling could have significant effects on the perfor-mance of the myocardium. Several studies have identi-fied the role of abnormal handling of intracellular

IDCM

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

Brillantes et al Calcium Release Channel Expression in Heart Failure 23

TABLE 2. CRC and Actin mRNA Levels in Normal and Myo-pathic Human Hearts (Normalized Values)

Patient CRC Actin

Normal1 1.02 1.182 1.14 0.953 0.86 0.764 0.98 1.11

Mean

ICM1

23456789101112131415161718

1.00+0.1 1.00±0.2

0.310.841.220.620.630.880.600.030.590.350.410.381.130.820.971.500.521.18

0.72+0.4*Mean

IDCM1

2345

67891011121314

1.160.972.611.310.880.670.740.570.420.550.870.302.191.611.151.341.151.14

1.09±0.6

1.370.391.070.581.080.981.172.040.681.720.810.672.930.23

1.701.391.851.210.921.301.121.540.441.471.121.001.900.33

Mean 1.12±0.7 1.23±0.5

CRC, calcium release channel; ICM, ischemic cardiomyopathy;IDCM, idiopathic dilated cardiomyopathy.

*p<0.025.

calcium as a cause of contractile dysfunction in failinghearts.8,33 Gwathmey et a17 reported that contractionsand Ca2 transients were prolonged in failing hearts.The combination of verapamil and ryanodine, both atmicromolar concentrations, completely abolished theCa2 transients, indicating that calcium flux through theDHPR and the CRC accounted for all of the abnormalCa2 observed in myopathic hearts.7 The observedabnormalities in Ca2 transients and contractions couldbe explained by a combination of abnormal release andreuptake across the SR and abnormal flux across theplasma membrane. Decreased expression of the SRCa2-ATPase could explain defective reuptake. De-creased expression of the CRC and DHPR, on the otherhand, might be expected to reduce SR calcium release

and flux across the plasma membrane. Another possi-bility is that as the expression of the CRC falls out ofproportion to that of the DHPR (in IDCM), an uncou-pling of the two channels develops, leading to defectivehomeostasis of intracellular calcium.Other factors must also be considered, including

alterations in myofilament responsiveness to Ca2`.3435Calcium release channel function might be defectivedue to alterations in the structure of the channels,abnormal interactions with modulators, and defects inother proteins that may interact with the channel.Elucidation of the possible contribution of these mech-anisms requires further study.The present study does not provide data regarding

the mechanism of the decreased expression of the CRCin patients with advanced ICM. Although little is knownabout the regulation of calcium channel gene expres-sion, there is abundant evidence in the case of theDHPR in brain and smooth muscle that different iso-forms can be expressed via alternative splicing of thechannel mRNA.36-39 Alternative splicing has not yetbeen reported for the CRC. Moreover, growth factorsand hormones may play a role in regulating calciumchannel gene expression. For example, Marks et a140have recently shown that expression of the skeletalisoforms of the DHPR and CRC is downregulated bygrowth factors (specifically, fibroblast growth factors) ina myogenic cell line. Fibroblast growth factors may playa role in regulating calcium channel gene expression inthe heart since acidic fibroblast growth factors areexpressed and secreted by cardiac myocytes in culture.41However, regulation of cardiac calcium channel genesby fibroblast growth factors has not been reported. Inanother recent study, regulation of the cardiac CRC bythyroid hormone was observed in experimentally in-duced hyperthyroidism and hypothyroidism in rabbits.42Thus, one effect of thyroid hormone on the heartappears to be regulation of calcium channel gene ex-pression. Whether hormonal regulation of calciumchannel gene expression plays a role in the adaptationsof the heart to end-stage failure remains to bedemonstrated.

Characterization of the expression of genes encodingproteins involved in E-C coupling in the heart in pa-tients with end-stage heart failure does not permit oneto conclude whether the observed differences reflectcausative or responsive phenomena. However, the ob-servation that in one case (ICM) mRNA encoding bothcalcium channels involved in E-C coupling is decreasedmay have implications in terms of the ability of thesehearts to handle calcium and develop contractile force.These observations also may have implications regard-ing the response of patients with end-stage ICM topharmacological interventions, including organic cal-cium channel blockers. For example, patients with de-creased expression of both the DHPR and CRC mightbe expected to respond more profoundly to calciumchannel blockers in terms of reduction in contractileforce of the LV. However, none of the clinical param-eters reported in this study correlated with the observeddifferences in CRC mRNA levels between the twogroups of patients with end-stage heart failure. Thus,clinical identification of patients who might be expectedto be more sensitive to therapy with agents that reduce

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

24 Circulation Research

* p<O.OO5

1 T

* 9

ICM

* Cardiac CRCE2 Actin

IIDCM

FIGURE 3. The levels ofcalcium release channel (CRC) mRNA (normalized to GAPDHmRNA levels) were reduced in ischemiccardiomyopathy (ICM) patients by 28% (p<0.025) compared with normal patients (NL) and by 46% (p<0.005) compared withpatients with idiopathic dilated cardiomyopathy (IDCM). The CRC mRNA level in IDCM was increased by a nonsignificant 12%(p=NS) compared with normal patients. Levels of a-actin mRNA in the left ventricle of normal patients were not significantlyaltered compared with patients with end-stage heart failure caused by either ICM or IDCM.

calcium entry (and, hence, calcium-induced calciumrelease) remains problematic.One of the most important questions in the present

study is whether the observed 28% decrease in CRCmRNA levels in the ICM hearts was caused by an actualdecrease in CRCmRNA accumulation or was caused byan artifact of the procedures used to evaluate the RNAsamples. We took several measures in this study toestablish that equivalent amounts of nondegraded RNAwere analyzed for each heart sample. To control for theamounts of RNA bound to nitrocellulose filters, North-ern blots were hybridized to both GAPDH and a-actincDNA probes. CRC mRNA levels in the normal andmyopathic ventricles were normalized to both GAPDHand a-actin mRNA levels. The GAPDH probe was usedas a control because GAPDH is a constitutively ex-

pressed "housekeeping" enzyme that has been usedextensively to control for mRNA levels. In the presentstudy we demonstrated that there were no significantdifferences in the levels of GAPDH mRNA betweennormal, ICM, and IDCM patients. Arai et a142 hadpreviously directly examined the relation of CRCmRNA regulation to that of GAPDH mRNA in therabbit heart. In this earlier study GAPDH mRNA levelswere used as an internal standard for quantifying thelevels of CRC mRNA in hyperthyroid and hypothyroidrabbit hearts. Under conditions in which cardiac CRCmRNA expression was regulated by thyroid hormone,GAPDH mRNA levels were found to be constant.The actin probe was used as a control to normalize

CRC mRNA levels because levels of a-actin mRNA inpatients with end-stage heart failure had recently beenexamined in a study comparing the expression of theskeletal and cardiac actin isoforms.22 No differenceswere seen in the expression of a-actin between patientswith dilated and ischemic cardiomyopathies.22 The lev-els of a-actin mRNA in the myopathic hearts in thepresent study were also unchanged from those in nor-

mal hearts, and there were no significant differences ina-actin mRNA levels between ICM and IDCM (Figure3, Table 2). Thus, both GAPDH and a-actin mRNAlevels were shown in the present study not to beregulated during end-stage heart failure and thus were

suitable for use as internal controls.The present study is the first to examine CRC mRNA

levels in human hearts. For this reason we used North-ern blot analyses to demonstrate that the CRC mRNAwas a single, nondegraded species in the human heartsamples. It is our experience that significant degrada-tion of this large mRNA can be detected (although notquantitated) by observing a smear running down thelane in the Northern blot. We found that the CRCmRNAs were not degraded in the patient samples, andwe did not observe any differences in quality of CRCmRNA signals between the normal patients and eithergroup of cardiomyopathic patients (see Figure 1). Theintegrity of the 28s ribosomal RNA band and the ratioof 28s to 18s ribosomal RNAs are generally used as

indicators of RNA degradation and were used to mon-itor the quality (degree of degradation) of the RNAtransferred to filters for Northern blot analyses. Nosignificant differences were observed between the trans-ferred RNA for the normal, 1CM, and ICDM samples.To control for the possibility that Northern blot transferof the large (16 kb) CRC mRNA was not quantitativewe also used slot blots (in which no RNA transfer isinvolved) and showed that the normalized values forCRC mRNA levels were in agreement with those de-termined from Northern blot analyses (see "Results").These results indicated that RNA transfer did not biasthe quantitation of the CRC mRNA levels. In addition,RNA samples from each type of patient (normal, ICM,and IDCM) were loaded onto each gel to avoid thepossibility that inefficient transfer of a particular gelwould bias the results. All Northern and slot blots werehybridized to a single labeled probe for each cDNA to

2.0-

1.8

X -- 1.5-_>._(D D 1.2-i-JZ W 1.0-

c= DE (xs 0.8-

0.-

0.2-

0.0-

* p<O.025r-

NL

Vol 71, No 1 July 1992

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

Brillantes et al Calcium Release Channel Expression in Heart Failure 25

avoid potential differences in signal caused by variabil-ities in the specific activities of probes labeled at differ-ent times. Films were exposed for three different expo-sure times to ensure that densitometric scanning wasperformed in the linear range. Taken together, thesecontrols indicated that the significant decrease in CRCmRNA levels in the ICM heart samples were not causedby experimentally introduced artifacts, differences inRNA transfer, or the degree of RNA degradationbetween the groups of samples studied.

It could be argued that a systematic degradation ofCRC mRNA in the ischemic tissues that occurredbefore sampling (in vivo) might reduce the CRC mRNAlevels by 25%. The cause of such in vivo degradationmight be a specific decrease in CRC message stability inthe ischemic hearts. Either decreased mRNA stabilityor decreased transcription (or a combination of both)would result in decreased CRC mRNA abundance inthe ischemic hearts. A generalized decrease in mRNAstability (i.e., increased mRNA degradation) in theischemic hearts would have resulted in a decrease in 28sto 18s ribosomal RNA ratios, as well as decreasedGAPDH and a-actin mRNA levels, which we did notobserve. Alternatively, the CRC mRNA could for somereason have been less stable after the tissue was har-vested. This effect would most likely have altered thestability of all large mRNAs, which would have beenreflected in a systematic decrease in the ratio of 28s to18s ribosomal RNAs in the ICM samples, not the CRCmRNA selectively, which we did not observe.

In the present study we have attempted to account forthe possibility that in a sampling from the LV wall,different amounts of myocardial cells might be presentdepending on the extent of fibrosis versus hypertrophyin the hearts. To correct for this possibility we havedetermined mRNA levels for a-actin that should reflectthe myocyte specific mRNA production and correct forany potential loss of myocytes relative to nonmyocytes inthe tissue sampled.

This study does not attempt to determine whether theobserved decreased expression of the CRC gene ispathogenic for LV dysfunction in patients with end-stage heart failure. It is quite possible that this de-creased CRC expression is secondary to other, moreproximal causes of heart failure. Nevertheless, the de-creased levels of CRC mRNA in patients with ischemiccardiomyopathies could lead to altered calcium homeo-stasis and impaired diastolic and systolic function ob-served in myopathic human hearts.8 Decreased expres-sion of the CRC might result in a reduction inexcitation-contraction coupling leading to decreasedcontractility.

AcknowledgmentsWe would like to thank Drs. Richard Gorlin and Milton

Packer for critical reading of the manuscript and for helpfuldiscussions and Dr. Bernardo Nadal-Ginard for encourage-ment. For assistance in obtaining cardiac tissue at the Brighamand Women's Hospital we thank Drs. Rosemarie Maddi andJohn Fox, Department of Anesthesia; Drs. Verdi DiSesa andGregory Cooper, Department of Cardiovascular Surgery; andDrs. Frederick Schoen and Richard Mitchell, Department ofPathology. We also thank Santwana Mukherjee for technicalassistance. This is paper No. 84 submitted from the BrookdaleCenter for Molecular Biology, Mount Sinai School ofMedicine.

References1. Rios E, Brum G: Involvement of dihydropyridine receptors in

excitation-contraction coupling in skeletal muscle. Nature 1987;325:717-720

2. Catterall WA: Excitation-contraction coupling in vertebrate skel-etal muscle: A tale of two calcium channels. Cell 1991;64:871-874

3. Fabiato A: Calcium-induced release of calcium from the cardiacsarcoplasmic reticulum. Am J Physiol 1983;245:C1-C14

4. Fleischer S, Inui M: Biochemistry and biophysics of excitation-contraction coupling. Annu Rev Biophys Biophys Chem 1989;18:333-364

5. Mercardier J, Lompre A, Duc P, Boheler K, Fraysse J, WisnewskyC, Allen P, Komajda M, Schwartz K: Altered sarcoplasmic retic-ulum Ca2+-ATPase gene expression in the human ventricle duringend-stage heart failure. J Clin Invest 1990;85:305-309

6. Smith V, Katz A: Inotropic and lusitropic abnormalities in thegenesis of heart failure. Eur Heart J 1983;4(suppl A):7-17

7. Gwathmey JK, Copelas L, Mackinnon R, Schoen F, Feldman M,Grossman W, Morgan J: Abnormal intracellular calcium handlingin myocardium from patients with end-stage heart failure. Circ Res1987;61:70-76

8. Morgan JP: Abnormal intracellular modulation of calcium as amajor cause of cardiac contractile dysfunction. NEnglJMed 1991;325:625-632

9. Packer M: Pathophysiological mechanism underlying the adverseeffects of calcium channel-blocking drugs in patients with chronicheart failure. (abstract) Circulation 1989;80:IV-59-IV-67

10. Kubo S, Olivari M, Cohn J: Calcium antagonists in heart failure.Ann N YAcad Sci 1991;522:553-564

11. Goldstein R, Boccuzzi S, Cruess D, Nattel S: Diltiazem increaseslate-onset congestive heart failure in post-infarction patients withearly reduction in ejection fraction. Circulation 1991;83:52-60

12. Elkayum V, Amin J, Mehra A, Vasquez J, Weber L, Rahimtoola S:A prospective, randomized double-blind crossover study to com-pare the efficacy and safety of chronic nifedipine therapy with thatof isosorbide dinitrate and their combination in the treatment ofchronic congestive heart failure. Circulation 1990;82:1954-1961

13. Takahashi T, Allen PD, Marks AR, Denniss AR, Schoen FJ,Grossman W, Marsh JD, Izumo S: Altered expression of genesencoding the Ca>' regulatory proteins in the myocardium ofpatients with end-stage heart failure: Correlation with expressionof the Ca2+-ATPase gene. Circ Res 1992;71:1357-1365

14. Marks AR, Tempst P, Hwang KS, Taubman MB, Inui M, Chad-wick C, Fleischer S, Nadal-Ginard B: Molecular cloning and char-acterization of the ryanodine receptor/junctional channel complexcDNA from skeletal muscle sarcoplasmic reticulum. Proc NatiAcad Sci USA 1989;86:8683-8687

15. Takeshima H, Nishimura S, Matsumoto T, Ishida H, Kangawa K,Minamino N, Matsuo H, Ueda M, Hanaoka M, Hirose T, Numa S:Primary structure and expression from complementary DNA ofskeletal muscle ryanodine receptor. Nature 1989;339:439-445

16. Zorzato F, Fujii J, Otso K, Phillips M, Green NM, Lai FA, Meiss-ner G, MacLennan DH: Molecular cloning of cDNA encodinghuman and rabbit forms of the Ca2+ release channel (ryanodinereceptor) of skeletal muscle sarcoplasmic reticulum. J Biol Chem1990;265:2244-2256

17. Otsu K, Willard HF, Khanna VK, Zorzato F, Green NM, MacLen-nan DH: Molecular cloning of cDNA encoding the Ca2+ releasechannel (ryanodine receptor) of rabbit cardiac muscle sarcoplas-mic reticulum. J Biol Chem 1990;265:13472-13483

18. Block BA, Imagawa T, Campbell KP, Franzini-Armstrong C:Structural evidence for direct interaction between the molecularcomponents of the transverse tubule/sarcoplasmic reticulum junc-tion in skeletal muscle. J Cell Biol 1988;107:2587-2600

19. Lai FA, Erickson HP, Rousseau E, Liu QL, Meissner G: Purifica-tion and reconstruction of the calcium release channel from skel-etal muscle. Nature (Lond) 1988;331:315-319

20. Wagenknecht T, Grassucci R, Frank J, Saito A, Inui M, FleischerS: Three-dimensional architecture of the calcium channel/footstructure of sarcoplasmic reticulum. Nature 1989;338:167-170

21. Saito A, Inui M, Wall J, Fleischer S: Mass measurement of the feetstructures/calcium release channel of sarcoplasmic reticulum byscanning transmission electron microscopy (STEM). (abstract)Biophys J 1989;55:206a

22. Boheler K, Carrier L, Bastie D, Allen P, Komajda M, MercardierJ, Schwartz K: Skeletal actin mRNA increases in the human heartduring ontogenic development and is the major isoform of controland failing adult hearts. J Clin Invest 1991;88:323-330

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from

26 Circulation Research Vol 71, No 1 July 1992

23. Marks AR, Tempst P, Chadwick CC, Riviere L, Fleischer S, Nadal-Ginard B: Smooth muscle and brain inositol 1,4,5-trisphosphatereceptors are structurally and functionally similar. J Biol Chem1990;265:20719-20722

24. Fornwald JA, Kuncio G, Peng I, Ordahl CP: The complete nucle-otide sequence of the chick a-actin gene and its evolutionaryrelationship to the actin gene family. Nucleic Acids Res 1982;10:3861

25. Fort P, Marty L, Piechaczyk M, Sabrouty S, Dani C, Jaenteur P,Blanchard J: Various rat adult tissues express only one majormRNA species from the glyceraldehyde-3-phosphate-dehydroge-nase multigenic family. Nucleic Acids Res 1985;13:1431-1442

26. MacLennan DH, Brandl CJ, Korczak B, Green NM: Amino-acidsequence of a Ca2++Mg2+-dependent ATPase from rabbit musclesarcoplasmic reticulum, deduced from its complementary DNAsequence. Nature 1985;316:696-700

27. Auffray C, Rougeon F: Purification of mouse immunoglobulinheavy chain messenger RNAs from total myeloma tumor RNA.EurJ Biochem 1980;107:303-314

28. Bers DM: Excitation-contraction coupling and cardiac contractileforce, in Developments in Cardiovascular Medicine. Boston, KluwerAcademic Publishers, 1991, pp 119-170

29. Bristow M, Anderson F, Port D, Skerl L, Hershberger R, LarrabeeP, O'Connell J, Renlund D, Volkman K, Murray J, Feldman A:Differences in 3-adrenergic neuroeffector mechanisms in ischemicversus idiopathic dilated cardiomyopathy. Circulation 1991;84:1024-1039

30. Komuro I, Kurabayashi M, Shibasaki Y, Takaku F, Yazaki Y:Molecular cloning and characterization of a Ca2+-Mg2+-dependent adenosine triphosphatase from rat cardiac sarcoplasmicreticulum. J Clin Invest 1989;83:1102-1108

31. De la Bastie D, Levitsky D, Rappaport L, Mercardier J, Marotte F,Winewsky C, Brovkovich V, Schwartz K, Lompre A: Function ofthe sarcoplasmic reticulum and expression of its Ca-ATPase genein pressure overload-induced cardiac hypertrophy in the rat. CircRes 1990;66:554-564

32. Nagai R, Zarain-Herzberg A, Brandl C, Fujii J, Tada M, MacLen-nan D, Alpert N, Periasamy M: Regulation of myocardial Ca2+-

ATPase and phospholamban mRNA expression in response topressure overload and thyroid hormone. Proc NatlAcad Sci USA1989;86:2966-2970

33. Katz AM: Cardiomyopathy of overload: A major determinant ofprognosis in congestive heart failure. N Engl J Med 1990;322:100-110

34. Ruegg J, Morano I: Calcium-sensitivity modulation of cardiac myo-fibrillar proteins. J Cardiovasc Phannacol 1989;14(suppl 3):S20-S23

35. Koretsune Y, Marban E: Relative roles of Ca2+-dependent andCa2l-independent mechanisms in hypoxic contractile dysfunction.Circulation 1990;82:528-535

36. Snutch TP, Tomlinson WJ, Leonard JP, Gilbert MM: Distinctcalcium channels are generated by alternative splicing and aredifferentially expressed in the mammalian CNS. Neuron 1991;7:45-57

37. Koch WJ, Hui A, Shull GE, Ellinor P, Schwartz A: Characteriza-tion of cDNA clones encoding two putative isoforms of the a,subunit of the dihydropyridine-sensitive voltage-dependent cal-cium channel isolated from rat brain and rat aorta. FEBS Lett1989;250:386-388

38. Koch WJ, Ellinor PT, Schwartz A: cDNA cloning of a dihydropyr-idine-sensitive calcium channel from rat aorta. J Biol Chem 1990;265:17786-17791

39. Hui A, Ellinor PT, Krizanova 0, Wang J-J, Diebold RJ, SchwartzA: Molecular cloning of multiple subtypes of a novel rat brainisoform of the a, subunit of the voltage-dependent calcium chan-nel. Neuron 1991;7:35-44

40. Marks AR, Taubman M, Saito A, Dai Y, Fleischer S: The Ryan-odine receptor/junctional channel complex is regulated by growthfactors in a myogenic cell line. J Cell Biol 1991;114:303-312

41. Weiner HL, Swain JL: Acidic fibroblast growth factor mRNA isexpressed by cardiac myocytes in culture and the protein is local-ized to the extracellular matrix. Proc NatlAcad Sci USA 1989;86:2683-2687

42. Arai M, Otsu K, MacLennan D, Alpert N, Periasamy M: Effect ofthyroid hormone on the expression of mRNA encoding sarcoplas-mic reticulum proteins. Circ Res 1991;69:266-276

by guest on February 23, 2013http://circres.ahajournals.org/Downloaded from