Reduced beta 1 receptor messenger RNA
abundance in the failing human heart.
M R Bristow, … , P E Ray, A M Feldman
J Clin Invest.
1993;
92(6)
:2737-2745.
https://doi.org/10.1172/JCI116891
.
Heart failure in humans is characterized by alterations in myocardial adrenergic signal
transduction, the most prominent of which is down-regulation of beta 1-adrenergic receptors.
We tested the hypothesis that down-regulation of beta 1-adrenergic receptors in the failing
human heart is related to decreased steady-state levels of beta 1 receptor mRNA. Due to
the extremely low abundance of beta 1 receptor mRNA, measurements were possible only
by quantitative polymerase chain reaction (QPCR) or by RNase protection methods.
Because the beta 1 receptor gene is intronless and beta 1 receptor mRNA abundance is
low, QPCR yielded genomic amplification in total RNA, and mRNA measurements had to
be performed in poly (A)(+)-enriched RNA. By QPCR the concentration of beta 1 receptor
mRNA varied from 0.34 to 7.8 x 10(7) molecules/microgram poly(A)(+)-enriched RNA, and
the assay was sensitive to 16.7 zeptomol. Using 100-mg aliquots of left ventricular
myocardium obtained from organ donors (nonfailing ventricles, n = 12) or heart transplant
recipients (failing ventricles, n = 13), the respective beta 1 mRNA levels measured by
QPCR were 4.2 +/- 0.7 x 10(7)/micrograms vs. 2.10 +/- 0.3 x 10(7)/micrograms (P = 0.006).
In these same nonfailing and failing left ventricles the respective beta 1-adrenergic receptor
densities were 67.9 +/- 6.9 fmol/mg vs. 29.6 +/- 3.5 fmol/mg (P = 0.0001). Decreased mRNA
abundance in the failing ventricles […]
Research Article
Find the latest version:
Reduced
flB
Receptor
Messenger RNA Abundance in the Failing Human Heart
MichaelR. Bristow, *WayneA. Minobe,*Mary V. Raynolds, * J. David Port, * RandyRasmussen,*Phillip E.Ray,'and Arthur M. Feldman*
*TempleHoyneBuell Laboratories, Division of Cardiology, University
of
ColoradoHealthSciences Center,Denver, Colorado80262;
tDepartment
of Biology, University of Utah, Salt Lake City, Utah84112;
andgThePeterBelferLaboratoryfor the Molecular Biology ofHeartFailure, The Johns HopkinsUniversitySchool ofMedicine, Baltimore, Maryland 21205Abstract
Heartfailure in humans is characterized by alterations in myo-cardialadrenergic signal transduction, the most prominent of which is down-regulation of
ftl-adrenergic
receptors. We tested thehypothesis
thatdown-regulation of
fl1-adrenergic
receptors in thefailing human heart is related to decreased steady-state levelsof61
receptor mRNA. Due to the extremely low abun-danceoffih
receptormRNA, measurements were possible onlyby
quantitative polymerase chain reaction (QPCR)
or by RNaseprotection
methods. Because thej8
receptor gene is intronless and81
receptor mRNA abundance is low, QPCR yielded genomicamplification
in total RNA, and mRNA mea-surementshadtobeperformed
inpoly(A)'-enriched
RNA.ByQPCR
the concentrationof
jI
receptor mRNA varied from0.34
to7.8
X107 molecules/,g poly(A)'-enriched
RNA, and the assay was sensitive to16.7
zeptomol.Using
100-mg ali-quots of left ventricularmyocardium
obtainedfrom
organ do-nors(nonfailing ventricles, n=12) or heart transplant recipi-ents(failing ventricles,
n=13),
therespective
jI
mRNA levels measured by QPCR were 4.2±0.7 X107/,ug
vs. 2.10±0.3 X107/i.tg
(P
=0.006). In these same nonfailing and failing left ventricles the respectivejBI-adrenergic
receptordensities were67.9±6.9
fmol/mg
vs.29.6±3.5
fmol/mg (P
=0.0001).
De-creased mRNA abundance in thefailing ventricles
wascon-firmed
by RNaseprotection
assays in totalRNA,which
alsodemonstrated
a50% reduction
in#1
message abundance. We conclude thatdown-regulation of
01
receptor mRNA contrib-utestodown-regulation
of#I
adrenergic
receptors in thefailing
human heart.
(J. Clin.
Invest.1993.
92:2737-2745.) Key
words:
,B-adrenergic
* heartfailure
-messenger RNA * polymer-asechain reaction
* receptorIntroduction
The
failing
human heartis characterized by desensitization
toadrenergic
stimulation (
1-5),
which
results inattenuation of
the
ability
of
the hearttoincreasecontractility during
exercise
or
physiologic
stress. Themajority
of
thissubsensitivity
toadrenergic
drive is
the resultof
aselective
down-regulation
of themyocardial
,B,-adrenergic
receptor(2-5).
Inthefailing
hu-Address reprint requeststo Dr. Michael R.Bristow, Temple Hoyne Buell Laboratories, Division ofCardiology, Universityof Colorado HealthSciencesCenter,Denver, CO 80262.
Receivedfor publication12November1992andinrevisedform21
July1993.
man heart ventricular myocardial fl-adrenergic receptors ap-pear tobe down-regulated in all subcellular fractions (4). Since there is nosignificant ,B receptor reserve in the human heart (6, 7),
#I
receptordensity is a major control point in the regulation ofcontractility (5). Less clear are the molecular alterationsresponsible for
myocardialjI
-adrenergic receptor down-regula-tion in thefailing human heart.Inmodel systems the steady-state abundance of
12-adrener-gic receptor mRNA and receptor protein levels correlate closely (8, 9). Further,
studies
in cultured heart cells ( 10, 11 ) suggest that thesamerelationship between mRNA abundance and receptorexpression
may alsoexist for
thef31-adrenergic
receptorsubtype.
Accordingly,
wetested the hypothesis that, in the human heart, the steady-state abundanceof
/3,-adrenergic
receptor mRNAwould correlatewith j3, receptor expression at theprotein level. Because ofthe extremely low abundance of
,B
receptor mRNA, we measured,3,
receptor mRNA levels byquantitative
polymerasechain reaction (QPCR)'
(12-14) orby
RNaseprotection (
15)
assays.Using QPCR
withpoly(A)+-enrichedRNA asthe
starting
materialorRNaseprotection in total RNA, the data support thepremise
thatdown-regulation
of
#,-adrenergic
receptors infailing
human ventricularmyo-cardium is related
todecreased
#3-adrenergic
receptor mRNA abundance.Methods
Tissue procurement
Humanventricularmyocardiumwasobtainedfromsubjects undergo-ing heart or heart/lung transplant, and from organ donors whose heartswereunsuitable for donationowingtoblood typeorsize incom-patibility.15nonfailingleft ventriclesweretakenfrom14organ donors
and onesubject undergoing heart/lungtransplantation, with all 15 heartshaving normal left ventricular function determined by echocardi-ography. Failingleft ventricularmyocardiumwastakenfrom 13 sub-jects with end-stage left ventricular failure dueto idiopathicdilated cardiomyopathywho underwenthearttransplantation.Thesesubjects were notreceiving intravenous,B-adrenergic receptor agonists, phos-phodiesterase inhibitors, orantagonistsbeforetransplantation.Mean
age of thenonfailingcontrolswas28.8±5.1(SEM)yr,comparedwith 41.7±4.7 yr for thesubjects with heart failure (P=NS).Eightof the
nonfailing
controlswerefemale, andseven were male. Of thefailing subjects, 11weremale andtwowerefemale.Allsubjects with end-stageheartfailurehad New York Heart
Asso-ciation classIVsymptoms beforetransplantation.Their
hemodynamic
data includedaleft ventricularejectionfraction of0.16±0.01,cardiac index of 2.01±0.23
liter/min/m2,
meanpulmonarycapillary
wedge pressure of 20.5±2.3 mmHg, mean pulmonary artery pressure of 30.5±2.6mmHg, andmeanrightatrial pressure of 6.6±2.3mmHg.Tissuealiquotswereremovedfrom thestill-beatingheart immedi-ately uponexplantation,andeither immersedinliquid
nitrogen (see
1.Abbreviation usedinthis paper:QPCR, quantitativePCR.
J.Clin.Invest.
© The AmericanSocietyfor Clinical
Investigation,
Inc.0021-9738/93/12/2737/09
$2.00
below)orplaced in ice-cold, oxygenated Tyrode's solutionasdescribed previously (2).
RNA
extraction
1-galiquotsof human left ventricular mid free wall were removedon
explantation, immediately immersed in liquid nitrogen, and stored at -70'C until used.Poly(A)+-enrichedmRNA wasextractedfrom 50-100-mg aliquots of human left ventricular myocardium using oligo(dT) cellulose
(Micro-FastTrack"
mRNA Isolation Kit Ver. 1.2,Invitrogen Corp., SanDiego, CA). Total RNA was extracted from separate 50-100-mg aliquots by the acid guanidinium thiocyanate phenol-chloroform method ( 16) using RNA Stat-60 (TelTest, Inc., Friendswood, TX), aspreviously described ( 13).Quantitative
PCR
1,B
receptormRNA measurement. The basic technique for measuring mRNA abundance by QPCR in small aliquots of human heart has beenpreviously described ( 16). The ethanol precipitated pellet of ex-tractedpoly(A)+-enriched RNA (0.8-3.2Mg)wasdissolved in 16 Ml of diethyl pyrocarbonate-treated H20, and a 4-Ml aliquotwas directly measured by acapillaryspectrophotometer (model DU-64, Beckman Instruments, Inc., Fullerton,CA). ForRNAextracted from both non-failing andfailinghearts 200ng(1-4Ml)ofthe remaining 12 Ml ofRNAwas then added to the reverse transcriptase reaction, along with 0.5 pg ofasynthetic (84mer) cRNA "internal" standard ( 17). Thereverse
transcriptase reactionwascarriedoutaccordingtomanufacturer's in-structions (Promega Corp., Madison, WI), using a20-Mlreaction vol-ume, RNaseH-free Moloney murine leukemia virusreverse transcrip-tase, and randomprimers. The synthetic RNA (or cRNA) was gener-ated by an in vitro transcription reaction from synthetic DNA templates having a T7 RNA polymerase promoter sequence. The cRNA was DNAse (RNase free) treated, phenol/chloroform
ex-tracted, and ethanolprecipitated before resuspension and quantifica-tion by spectrophotometry. Theamountof cRNA addedtothe reverse transcription reaction was determined experimentally toultimately yieldcollinear amplification of the 131 receptor and internal standard
cDNAs.
Extensive characterizationofamplification of reverse transcribed 13, receptor mRNA resulted in selection ofthe
following optimal
condi-tions: 2% DMSO and 2 mMMgCl2 in the Taq polymerasebuffer,the useof RNaseHtodegrade theRNAstrand in theRNA:cDNAduplex afterreversetranscription, 50 pmol of each primer, and4Uof Taq polymerase (PromegaCorp.) per100-Mlreaction. Additionally, the re-action mixturewasdenaturedat94°Cfor 5 min, then cooledtothe annealingtemperature of55°C for 2min, andbroughttothe72°C extension temperature before Taq polymerasewasadded (18). Typi-callya4-Mlaliquot ofthereversetranscriptase/RNAse H-treated reac-tionwasused in the PCRreaction, withafinal volumeof 100M1.Theprimers utilized in the 131 receptor PCR reaction spanned a 202-bp segment ofthe human131 cDNA sequence (+ 150 to +351 [17 ])
andincorporated restriction enzyme recognition sites at the 5' ends for
usein thecloning of this segment for other purposes. The sequences were: 5'.CCGAATTCCGAGCCGCTGTCTCAGCAGTGGACA * 3'
for the 5' primer and 5'- CCAAGCTTGGTGGCCCCGAACGGC-ACCACCAGCA-3'for the 3' primer. For subsequent quantification of PCR product yield after gel electrophoretic separation, an aliquot of
the3'primer was end-labeled with
[y-32P]ATP
using T4 polynucleo-tide kinase (Promega Corp.). All DNA oligonucleotides were pur-chasedfrom Operon Technologies, Alameda, CA. As described previ-ously ( 13),10-,Ml
aliquots were removed from sequential amplification cycles (15through 33) and fractionated with molecular weight stan-dardsby agarose gel electrophoresis. Specific product bands werelo-cated by ethidium bromide staining and ultraviolet irradiation, ex-cised, and quantified by Cerenkov counting.
With this technique, separate amplification curves for the unknown
cDNA and the internal standard cDNA were constructed. Since a knownamountofsyntheticRNAinternal standard is carried through the cDNA
synthesis
andamplification,
theoriginal
13
receptor mRNAlevel can be determined by extrapolation from the internal standard cDNA standardcurve. It is essential that thisextrapolation be per-formed during the exponential phase of amplification and that the curves for the unknown and internal standard are collinear, which occurred in allcasesbetweencycles 20 and 30. An RNA samplenot
subjectedto reversetranscription isrun as acontrol for contamination bygenomic DNA.
12 mRNA abundancewasmeasured in total RNA with linear
am-plificationoccurring in all cases between cycle numbers 20 and 30. The assay conditions for measuring 12 receptor mRNA abundance by QPCR have beenpreviouslydescribed (13, 19); 1 ggof total RNA
from bothfailing and nonfailing groups plus 30 pg of cRNAwasused in thereverse transcriptase reaction. Typicallya 1 glaliquot of the reversetranscriptase/RNaseHtreated reactionwasused inthe PCR reaction.
Measurementof
1,I
receptormRNAbyRNaseprotection.The hu-man1,-adrenergic
receptor cDNA(17)wassubcloned intopBluescriptIIKS (Stratagene, Inc., La Jolla, CA) anddigestedwith either SmaIor
EagI. AntisenseRNAtranscribed from the T3 promoter of the SmaI digestedvectorcomprised- 450bpof the terminal end of the 3'
un-translatedregion(SmaI probe, probe I).Antisense RNA transcribed from the T3 promoter of the EagIdigestedvectorcontained - 400bp of the 3' end of thecoding regionandtheentire 3' untranslatedregion (- 700 bp) (EagI probe, probe II).
Antisenseriboprobesweretranscribed from theappropriately di-gested 131 receptor cDNAusingtheMaxiscribe kit(Ambion,Inc.,
Aus-tin, TX): The riboprobes were labeled by incorporation of [
a-32P]UTP800Ci/mM (New England Nuclear, Boston, MA) into the
RNAduringtranscriptionwith T3 DNA-dependentRNApolymerase. Approximately 1-2 X 105 cpm of32P-labeledantisense
1l3
RNAwasused in each nucleaseprotectionassay.
Labeledriboprobeswerehybridizedinsolution with 2or5 yg of totalRNAextractedfromfailingornonfailingexplanted hearts for 18
hat44°C.Unhybridized single-strandedRNAwasdigested with
RNa-seA and RNaseT1. RNA-RNAhybridswereresolvedby electrophore-sis inan8%acrylamide denaturinggel and visualized by autoradiogra-phyat -80°C. The RPAII kit (Ambion, Inc.)wasusedforsolution hybridization and RNase digestion.
ProtectedRNAspecieswerequantified by densitometry.Inall ex-periments in vitro transcribed13,receptor sense strandRNAgenerated by thesamemethodfrom thesame vector astheantisenseriboprobes servedas apositivecontrol.Inaddition,in ordertoverifythespecificity of the 131 probes for
3,1
mRNA,thesyntheticfull-length12mRNA(20) wastranscribed and100pg hybridized with each ofthe two 131 antisense riboprobes.Hybridization intensityfor protected RNA species of the appro-priate sizewasmeasured by densitometry ofautoradiograms developed for2-7d.Two orthreenonfailingcontrolsamples wererunoneach gel and the ODreading averages ofthese samples gave the 100%orcontrol value. Each samplefrom failing or nonfailing hearts was then refer-encedtothis 100% value, with results reportedaspercentageofcontrol.
13-Adrenergic
receptorquantification.#1I
and12
adrenergic receptors werequantified in a crude membrane fraction prepared from 5-g ali-quots of left ventricular free wall,using previously described methods (2).Briefly,the totalpopulation of13I
plus32
receptors wasmeasured by the nonselective radioligand[1225l]iodocyanopindolol.
Maximum binding(Bm..)
and the['1251]iodocyanopindolol
dissociation constant (Kd)weredetermined by nonlinear least-squares computer modeling of thespecificbindingcurve(2).Fractions of1,3
vs.12
receptors in the membranepreparations were determined by computer modeling of competition curves using the highly selective13,
antagonistCGP-2071 2A(4).Multiplication ofthe totalBma,by the fraction of
13,
recep-tors obtained from CGP-20712A/[
'251]iodocyanopindolol
competi-tioncurvesthenyielded3,1
receptordensity.Protein concentrations were determined by the Peterson modifica-tion ofthemethodof Lowry (21 ). RNA concentration and purity was determined bymeasuring absorbance at 260 and 280 nM usingahighly sensitivecapillaryspectrophotometer (BeckmannInstrumentmodel DU-64).
A
B
C D E F G H I J K L M N O P Q R S T U V W
Figure 1. Assessmentof genomicDNAcontaminationlevels betweenpoly(A)'-enrichedRNA(H, I,J,K,P, Q, R, S) and totalRNA(L, M,N, 0, T, U, V, W) fractionssubjectedtoQPCR. 200ngofpoly(A)+-enrichedRNAand IAigof totalRNAextracted fromtwofailingandtwo
nonfailinghearts wereaddedtoduplicatereversetranscriptionreactionsdiffering onlyin whethertheycontainedreversetranscriptaseornot.
Subsequently, 10% of each of the totalRNAreactions and 20% of the
poly(A)'
reactionswereamplified by PCR. All thereversetranscription reactionswithtranscriptase(H I, J., K, L, M, N,0)showstrongamplificationofspecific 202-bp product (comparedwith thepositivesyntheticf,
receptormRNAcontrol,(C)butof the reactions withouttranscriptase (P, Q,R,S, T, U, V, W)onlythepoly(A)+-enriched
RNAreactions (P, Q, R, S) donotamplify the 202-bp product. (A)DNAmolecularweightmarkers. (B) PCRreagentblank.(C)Synthetic, full-length f3, re-ceptormRNA/positiveamplification control. (D, E, F, G) SyntheticfA,
receptormRNA/noreversetranscriptasecontrols.Statistical analysis
Comparison ofnumerical data betweentwo groups wasbyanunpaired
t-test.Correlationanalysiswasbysimple regression.Allstatistical
com-putationwasperformedontheStatView 512 +TM program
(Brain-power, Inc., Calabasas, CA). Statistical significance wastaken as P <0.05 inatwo-taileddistribution.
Results
QPCR
f3 receptor mRNA abundance. Shown in Fig. 1 is a 4% agarose
gel
stained with ethidium bromide
inwhich
were run PCR products generatedfrom four
samples eachof
total andpoly(A)+-enriched
RNA, extractedfrom
twononfailing
and twofailing
humanventricles.
Inthe absenceof
reversetran-scriptase
[RT(-)
condition], genomic material of
the ex-pectedsize for
theAl
receptorPCR productis
clearly present in the total RNAsamplesafter
38 cycles. In contrast,poly(A)+-enriched
RNA shows noamplification
underRT(-)
condi-tions.
Basedonthese resultsweusedpoly
(A)+
-enriched
RNA asthestarting material for QPCR
quantification of
f,
mRNA abundance.Fig.
2 shows anethidium
stained
4% agarosegel inwhich
were separated the contents
of six
PCRreactions
using
poly(A)+-enriched
RNAasthestarting material.
Theampli-fied
cDNAs were made by reversetranscribing 36-44
ngof
poly(A)+-enriched
RNAextractedfrom
six samples of
humanleft ventricle
(designated A, B, C, D, E, and F) alongwith
200-500
fg of
theinternal
standard cRNA. Each PCRreaction
is represented by a set
of nine sequential
I0-yd
aliquots
which
wereremovedfrom thereaction tube at (left to right) cycles 14, 17, 19,
21,
24, 25, 27, 29, and 32. The PCRamplification
product
for
Areceptor mRNA appears as a202-bp band,while
the internal standard appears as an 84-bp band. The gel also includes "genomic controls": that is, poly(A)+-enriched RNA
from
thesix
samples (labeled,respectively, G-L) not subjected to reversetranscription.
There wasprogressive amplification of
the target mRNAand the internal standard cRNA in all six samples (A-F). However, there was no amplification of geno-micDNAin any sample (G-L). Identityof the 202-bp band was
confirmed
bydirect PCRsequencing (22).Fig.
3 showsamplification
curvesfrom four individual
sam-ples (twofailing
and twononfailing)using 1.9-11.4 fg of inter-nal standard cRNA per ngof poly(A)+
RNA. It can be seen that thereis
collinear amplification
between20 and 30 cycles; curves wereaccepted foranalysis
only ifamplification
was lin-earfor
atleast
six
cycles.Inorder to comparethe
efficiency of
reversetranscriptionof
theinternal
standard cRNA to thatof
thef,
receptor mRNA, we measured by QPCR adilution series
of
cDNAs reversetranscribed from
known, equal amounts ( 16.7 zepto-mol to 16.7 azepto-mol of each; or I04 to107
molecules)of
the cRNAandfull-length
f1 receptor mRNA. The latterwas gener-atedby invitro
transcription from
apBluescript
IIKSphage-mid
expression
vector. Anamplification
curve wasconstructedfor
eachconcentration of
theseries,
andquantities of
mRNA vs.cRNA were calculatedfrom equivalent points in theexpo-nential
phaseof
amplification.
Asshownin
Fig.
4, theincorpo-ration
of
labeledprimers into
the PCR productsmadefrom
theinternal
standard cRNAwasessentially identical
tothosefor
the
full-length
f3 receptor mRNA. In ordertodetermine
if thestan-B
MW
-202
-84
C
D
G
H
I
J
K
L
Figure 2. Agarosegelscontainingthefractionatedcontentsof six separate PCR reactionsusing poly(A)+-enrichedRNAsubjectedtoreverse
transcriptionandPCR.Amplified
l,,
receptor mRNA appearsasa202-bpproduct, and the84-bpproductistheinternalstandard. MW, molec-ularweight markers.10-,ul
aliquotswereremovedfrom the PCRreaction incycles 14, 17, 19, 21, 24, 25, 27, 29,and32, inlanes 1-9from lefttoright. G-Larecycles 14,25,and 32for eachof sixsamplesinwhichreverse
transcriptase
wasomitted("genomic control").dard,
the assay waslinear with
respecttoadded mRNAfrom
measured
valuesof 2.5
X107
to18
X107
molecules/jig
poly(A)'
RNA.Table I
gives
theQPCR
quantification of
1l1
mRNAin
poly(A)+-enriched samples from 13
failing
and 12nonfailing
human hearts.
Compared
tononfailing
controls,
failing
heartshad
a50% lower abundance of
#I-adrenergic
receptor mRNA.This
finding
wassimilar
tothe 56%
lowerdensity of
1I
receptor observedin the
failing
vs.nonfailing
heart and
12
receptor mRNAabundance.
132
receptor
mRNA abundance. 132receptor
mRNAabun-dance
wasalso measuredby
QPCR,
in total RNAextracted
from
nonfailing
andfailing
human hearts(Table II). In con-trast to13,
receptormRNA, collinear amplification of
reverse-transcribed
12
messagealways
occurred in <30 cycles.
Geno-mic
amplification
was not observedfor
the132
mRNA assay, presumably becauseof
the lowerPCR
cyclerequirements
(data notshown). As can be seenin
TableI,
there was nodifference
(P= NS)
in
132
receptor mRNA abundance or12
receptorden-sity
betweennonfailing
andfailing
hearts.Relation between 13,
receptormRNAabundance and
13,
or 2receptordensity. Fig.
5gives
thesimple regression
relation
between
QPCR-measured
13 receptor mRNA abundance and1,
or12
receptor
density
inthe 25(13
failing
and 12nonfailing)
left ventricles.
Thereis
asignificant direct relation between
131
mRNAand
3,B
receptordensity (top panel,
r =0.52,
P=0.008)
but no
relation
between 13 mRNA and132
receptordensity
(bottom panel, r =0.07, P=NS).
RNase
protectionNuclease
protection
assays werealso used
to measure 131 mRNAabundance, in total
RNAextracted
from explanted
failing and nonfailing hearts.
Arepresentative autoradiogram
from
an RNaseprotection
assayutilizing probe
Iin total RNAis shown
inFig. 6.
InTable
II aregiven the
13,
mRNAabun-dance
measurementsfor probes
Iand
II,
andfor the average of bothprobes
when both were used inthe
samesample.
The abundanceof
13,
mRNA waslessin
failing
than innonfailing
hearts,
confirming
the resultsof
quantitative PCR. Using
the average datafor
bothprobes,
13,
receptor mRNA wasdecreased
by
50.2% and13,
receptordensity
by 65.2%
inthe
samehearts
(Table
II).
As shownin
Fig.
7,13,
mRNA measurement by RNaseprotection
washighly correlated
(r =0.90,
P < .001) with QPCR measurements in the same samples, whichin-TableI.
1,1
and132
Receptor mRNAAbundancesMeasured by QPCR and13,
and 2ReceptorDensities in
Nonfailing
andFailingHumanLeft VentricularMyocardiumGroup
(n) 1,3 mRNAabundance hi, receptor density fl2mRNA abundance 12receptor density
moleculesx
107/,sg
fmol/mg moleculesx107/;Lg fmol/mgpoly(A)+RNA totalRNA
Nonfailing
(12) 4.2±0.7 67.9±6.9 2.2±0.6 22.1±1.9
Failing
(13) 2.1±0.3* 29.6±3.5* 1.6±0.3 17.7±1.3
Valuesaregiven±SEM. *P <
0.01
vs.nonfailing controls.2740 Bristowet al.
MW
E
F
MW
NON-FAILING
NON-FAILING
o BETAI
* CONSTRUCT o 0
-0 0
0 0
1o4Z
: 10'I
0
o*
0
S a
8
15 20 25
CYCLES
FAILING
10'
.
o BETAI * CONSTRUCT
0
. 0 104'
0 0 0 0
0 0 * @00
0
o 0
0
15 20 25
CYCLES
- 3
30 3 5
o
BETAl
* CONSTRUCT
0
.8
S a *
1 0
103
102
101
_
1 0
30 35
.
15 20 25
CYCLES
FAILING
o BETAI
* CONSTRUCT
0 0
0 0 0
0 S
8
15
20 25CYCLES
Figure 3. PCR amplification ofreversetranscription products (cDNA) of
#,
receptormRNAfrom poly(A)+-enrichedRNA (., BETA 1)andinternal standard cRNA from the A,construct(-, CONSTRUCT).
cluded RNAextractedfrom fournonfailingand sixfailing
ven-tricles.
Toverifythespecificityof the antisenseprobes full-length sense human f2-adrenergic receptor mRNA was transcribed
andhybridized to bothantisense A, probes. ThetwoA3
anti-senseprobesusedwerehighly specificfor fl-adrenergic
recep-tormRNA,asneither the SmaI(probe I)northeEagI AI probe (probe II) protectedafragmentofl2mRNA ofthecorrectsize
(datanotshown).
Asthe
#I-adrenergic
receptorgeneisapparently intronless(Fig. 1), protection ofgenomicDNA presentas acontaminant
in the RNApreparationscould beapotential problemasitwas with QPCR. Accordingly, RNA extracted froma nonfailing
heartwastreated with DNAse I beforehybridizationwith the
SmaIriboprobe. Bydensitometry the abundance offlI mRNA inDNAse I treatedtotalRNAwas nodifferent from the
abun-dance of fI mRNA in untreated total RNA from the same
sample (datanotshown).
Discussion
Using QPCRwe havepreviously reportedthat thefailing
hu-manheartexhibits marked abnormalities ofgeneexpression,
including de novo expression of atrial natriuretic peptide
mRNAanddecreasedlevels ofphospholambanmRNA( 13). Furthermore,thesechangesingeneexpressionarepotentially
reversiblewith restoration of normal cardiac function(23). Within the context of the altered receptor-G-protein-adenylyl cyclase signal transduction pathways in the failing heart, the molecular mechanisms responsibleforthe various
described abnormalities( 1-5)remainelusive.ThemRNA
lev-els for thef2adrenergicreceptor( 19)and the Gproteins,Gas
( 19) andGaj3 ( 19), are notdifferent intotal RNA extracted
from failing and nonfailinghuman ventricularmyocardium. This isnottotally unexpected,sincetherespectivegene
prod-uctsare not quantitativelyaltered in thefailinghuman heart
(5). Basedonpreviouswork inthe failinghuman heart(2, 3)
8
0 0 0 0
0 S
t
1o
a
I.
C)
1 0
10
10'4-a
103
102
10
0
30 35
0 0
S
0 S 0
30 3 5
101
-103-,
-2
-* BETA-1
y =3.8360+ 0.56342x RA2=0.965
.
0 INTSTND
y - 3.7934 +0.61796x RA2=0.975
-1 0
LOG amoles, INTERNAL STANDARD OR BETA-1 mRNA
Figure4. Simultaneousmeasurement of identical molar amountsof 3,1 receptor mRNA
(n, BETA-I)
and(31 cRNA internal standard(standard
*,
INTSTND) by quantitative
PCR.andincellculture modelsystems( 10, 11),oneelement ofthe
receptor-G-protein-adenylyl cyclase complexthat isastrong candidate for abnormal gene expression is the
#,(-adrenergic
receptor.
By Northern analysis or QPCR
(3I-adrenergic
receptor mRNA abundance measurements in total RNA extracted from human heart have thus far been unreliable, duetolow abundance of the G + C rich(l3
receptor message (17) and genomic contaminationof PCRproducts. Basedonthe struc-turesofcloselyrelatedgenes(20, 24)and ourPCR data the#,B-adrenergic
receptor
genelikely
contains nointrons,
andtherefore PCRprimers couldnotbechosentodifferentiate the cDNA amplification signal from that ofgenomic DNA. The
needforhigh cyclenumbers(>35)indetectingthis
low-abun-dance messageexacerbatedthegenomicDNAcontamination problem. Consequently, it was necessary to use
poly(A)+-enriched RNAasthe starting material. Inusing these condi-tions, fewer than 32cycles ofPCR wererequiredtoproduce satisfactorylevels ofamplification. Furthermore, under these conditions, amplifications of non-reverse-transcribed
[RT(-)]
RNAproducednodetectablesignal, indicating the elimination
ofgenomic DNA contamination.
Inusing quantitative QPCR it is mandatory that collinear amplification be obtained with theinternal standard. This,in turn, requires that approximately equal amounts of internal standard and mRNAarepresent in the initialreactionmixture. Additionally, it is important that this collinear amplification
occur at relatively low cycle numbers (i.e., <35), to avoid artifacts causedby PCR. These criteria weresatisfied by the
above describedassayconditions.
Infailing ventricles, steady-state (3, receptormRNAlevels measuredby QPCRwere50%lowerthanin nonfailingcontrol hearts.Inthesesamefailing hearts receptordensitywas56%
lowerthan innonfailing controls. Moreover, when all 25
ven-tricles inwhich (3, receptormRNAabundancewasmeasured
wereconsidered, therewas adirect relation between (3dmRNA abundance and(,3 receptordensity, butnorelation between(,3
mRNAand(32receptordensity. The degree ofdown-regulation
ofmRNAandthedegree ofdown-regulation of (31-adrenergic
TableII.
(3I
ReceptormRNAAbundance MeasuredbyRNase Protection and#,(-Adrenergic
Receptor Densityin Nonfailing and Failing Human Ventricular Myocardium
Probe I Probe II Both MO receptor
Group (Smal digest) (EagI digest) riboprobes density
%ofcontrol fmol/mg
Nonfailing 100.6±13.3 99.5±19.1 99.4±9.5 83.0±10.3 (n = 7) (n =4) (n =7) (n =7) Failing 52.3±9.0* 48.8±9.3* 49.7±6.6* 29.9±4.2*
(n = 6) (n =6) (n =6) (n =6)
Valuesaregiven±SEM.* P <.05vs.nonfailing.
receptors were
therefore
quite similar considering
the vast arrayof factors which could
potentially influence
theregula-tion of both
mRNAand
proteins. These
factors include alteredtranscriptional and translation
frequency,
mRNAstability,
and
receptorprotein
turnover.Also, the fact
that thenonfailing
controls were obtained from organ donors who are invariably
subjected
toveryhigh
levels ofendogenous sympathetic
drive due to braininjury (25)
cannotbeignored.
Sincethe kineticsof
mRNAturnover arelikely
tobefaster than
receptor turn-over(i.e.,
mRNAhalf-life
vs.protein half-life), it
may be that mRNAabundance
wasalready
beginning
todecrease in the
nonfailing controls.
Decreased
(3,
receptor mRNAin
failing left ventricles
wasconfirmed by
RNaseprotection
assay,which showed
arelative
decrease
in
message levelsby 50% in six
failing hearts. The
presence
of
Al,
receptorgenomic material contaminating
the total RNA in these assaysdid
notpose aproblem, presumably
due to the
preference of
theriboprobes for
RNA-RNA vs. RNA-DNAhybridization (26). Of the
tworiboprobes,
the 3' UTRprobe
(SSmaI digest,
probe I)
gavethe best
hybridization
signal.
Both
probes
werespecific
for
(31
receptormRNA,
asin
vitro transcribed
(32
mRNA did not cross-protect against RNasedegradation. The fact that in
two RNAfractions
twototally
independent
methods, QPCR and RNase protection,
yielded decreased
l,,
receptor mRNA abundancemeasure-ments
infailing
human heart provides strong evidence thattranscript
steady-state levelis
oneimportant determinant offl
receptorregulation in
human ventricular myocardium.Despite
the 50% decreasein
flB
mRNA inpoly(A)+-enriched
ortotal RNAextractedfrom
failing human hearts(2
mRNA
abundancein
total RNA was similar in failing and innonfailing ventricles. This is in
agreement with our previously reported data in totalRNA ( 19). Apparently because(32
recep-tormRNAabundance is
greater thanfl,
in the human heart,QPCR
can beperformed in
total RNA with relatively low(<
30) cycle numbers and genomic
contamination can beavoided ( 19).
There aretwoprevious reports of QPCR measurements of
#I
receptor mRNA infailing humanventricular
myocardium, onereporting
anincrease
compared to nonfailing controls (27) andonereporting
adecrease
(28).
Inthe previous reportshow-ing
adecrease (28), 38cycles of PCR were used on total RNA and noRT(-) controlcondition was reported. The report of anincrease
in#I
mRNA infailing ventricles
also used total RNA(27).
Again,
thecombination of
low abundance and an2742 Bristowetal.
C.)
0
-i
y = .043x + 1.125, R-squared: .302
y = -.027x + 3.673, R-squared: .006
8.
a)
Ca) 0
E 74 O
6 03
X
5.
E 1
| z 3 A ; m - - C1 a | Figure5. Simpleregression
13
° O3 °plots of
fB
receptormRNAJ 1 O3
(y-axis)
vs.jl,
receptorden-|3
m10 O |sity (toppanel)and02
recep-0 . . . 2 tor
density
(bottom panel)
7.5 10 12.5 15 17.5 20 22.5 25 27.5 30 32.5 in the 25left ventricles in
BETA2, FMOLUMG which ,S mRNAabundance
wasmeasured byQPCR.
1 2 3 4 5 6 7 8 9 10 II Figure 6. Arepresentative
autoradiogram ofanRNase
protectionassay.Total RNA extracted fromexplanted
hu-manheartswashybridized
withan[a32P]UTP-labeled antisensefl riboprobe(SmaI
450 bp fl,
450-bp
probe). Fragmentsprotected
from RNasedigestionwere
resolved bydenaturinggel
electrophoresisand visual-izedby autoradiogram.Lanes
1and2:SmaIantisensef3l
riboprobewithout (lane 1) and with(lane 2)RNasetreatment;lane 2 contains eight timestheamountof riboprobeaslanes or3-8. The smaller, faster migrating bands
presentinalllanesareprobablyself-protectedprobefragments.Asthe probe isveryG+ Crich(17), higher orderstructuresinthesingle
strandedantisenseprobearelikelytobepresent,evenunderdenaturingconditions. Thesestructureswillprotectportions ofthe probefrom RNasedigestion.Lanes 3, 4, and 5:synthetic f3ImRNA (200, 100,and 50 pg) hybridized withthe labeled,B riboprobe.Theprominentlower
bands in the standard lanesaremostlikely dueto prematureterminationof transcription ofthe syntheticsense
lB
mRNA. Lanes 6, 7, and 8:Total RNAfrom three differentexplantednonfailinghuman heartshybridized withtheantisense
#I
riboprobe. Lanes 9,10,and 11: TotalRNAfrom threedifferentexplantedfailinghumanhearts hybridized with the antisense
fl,
riboprobe. iB mRNA abundance is significantlylowerinfailinghumanheartscomparedtothe abundance in nonfailinghuman heartsasdeterminedby RNase protectionassay.
CD 0 0
C.)O 0E 7.
6
°0a) 6. 0
E 5
3.)
E O k 0 0
-2
OO
0)~~~~~
H 0
m
1.
OO0
0 20 40 60 80 100 120
BETA-1 RECEPTORDENSITY,fmol/mg
R-y = 14.013x + 30.974, R-squared: .817 130. 120 0 110 100 90. 80 * 70 * 60 40B*
30' .* .. . . 0 1 2 3 4 5
BETA-11 mRNA,moleculesx 10'7/mcg
Figure7.Correlation of
0,
mRNAabundancemeasurpoly(A)+-enrichedRNAvs.measurementbyRNasep
total RNA;fournonfailing and sixfailinghumanleft
intronless gene makes measurements ofAl rec
levelsby QPCR challenging,andunreliable in tol
high (> 30) cyclenumbers. Dependingonthe D tio in myocardium comprising the starting
"mRNA" abundance in total RNA could be fc creased ordecreased.
There are two previous reports of RNase p formedassolution hybridization with thehybri
filtration(29, 30). Althoughidenticalriboprobe
these two investigations (29, 30),onestudy in lated cardiomyopathy found an increase in "mRNA" (30) and the other in valvular hear
foundadecrease. Basedonthenumberofself-p
mentsdetected onacrylamide gelsin ourownI tion assays, it is obvious that the approl RNA:RNA hybrid needs to be separatedand i electrophoresis. Unlessit has been documented protected or a verydominantprotected produc
the overwhelming majority ofthe measured ra the solution assay, filtration isolation ofhybric
can lead tospuriousresults.Circumstancessuch
havecontributed tothedivergentresults inthese
studies(29, 30).
Although the starting material (left ventri( dium)usedfor mRNAandreceptormeasureme
cellular,themajorityofcellvolume,protein,an'
tricular myocardium is derived from cardiac
mammalian ventricles adrenergic receptors expressedinmyocardialcells(31 ),andtherefore
thevastmajorityof
A,
receptorproteinand mRin thisinvestigation wasderived from cardiac
havepreviouslyshown thatthisis in factthecasc ergic receptors measured in crude membrane
tracted from humanventricles(2).
Insummary,f1 receptormRNAabundance
failing human heart than in nonfailing, contrc decreasecanfullyaccountfor theobserved dowr
fl1-adrenergicreceptorexpressioninfailing hearts
ies will be required todeterminethe molecular
alteration in geneexpression.
Acknowledgments
Wewish tothankDr.CarlT.Wittwerof theUniversiti
calSchoolforhelpful suggestionsonoptimal conditio 0
M. Benjamin Perryman for critical review of the manuscript, and Frank Stewart for manuscriptpreparation.
Thisworkwassupported byNational Institutes of Health(NIH)
grantsHL-13408-15andHL-48013-0IA2,awardedtoDr.Michael R. Bristow. Dr. Arthur M. Feldman was supported by NIH Program
Grant HL-39719 and The W. W. Smith Charitable Trust.
References
1.Bristow,M.R.,R.Ginsburg,W.Minobe,R.S.Cubicciotti,W. S.Sageman,
K.Lurie,M. E.Billingham, D.C.Harrison,andE. B.Stinson. 1982. Decreased catecholamine sensitivity and 3-adrenergic-receptor density in failinghuman
hearts. N.Engl.J.Med. 307:205-211.
2. Bristow,M. R., R.Ginsburg,M.Fowler, W.Minobe,R. Rasmussen, P. -edby QPCR in Zera, R.Menlove,P. Shah, and E.Stinson. 1986.,B and adrenergicreceptor
)rotectionin subpopulationsin normal andfailing humanventricular myocardium:coupling ventricles. of bothreceptorsubtypesto muscle contraction and selectiveA,receptor
down-regulationinheartfailure. Circ.Res.59:297-309.
3. Brodde, 0. R.,S. Schuler, R. Kretsch, M. Brinkmann, H.G. Borst, R.
,eptor mRNA Hetzer,J. Reidemeister, Warnecke, Regional
distribution of13-adrenoceptorsinthe human heart:coexistenceof function
/B,-talRNA using and,B2-adrenoceptors inbothatriaandventricles inseverecongestive
cardiomy-)NA/RNAra- opathy.J. Cardiovasc.Pharmacol. 8:1235-1242.
material,1 4.Bristow, M. R.,W. Minobe,R.Rasmussen, P.Larrabee, L.Skerl,J. W.
mato
bein-
Klein,
F. L. Anderson, L. Murray, L.Mestroni,
S. V. Karwande,etal. 1992.
)undto be in- -Adrenergic neuroeffector abnormalities in the failing human heartare
pro-duced bylocal, rather thansystemic mechanisms. J.Clin. Invest.89:803-815.
Protection per- 5.Bristow, M. R., R. E. Hershberger, J. D. Port, E. M. Gilbert, A. Sandoval, R. Rasmussen, A. E. Cates,andA.M. Feldman. 1990. O-Adrenergicpathways in Ids isolatedby nonfailingand failinghuman ventricularmyocardium. Circulation. 82(Suppl.
swereusedin I):12-25.
idiopathic di- 6. Port, J. D., and M. R.Bristow. 1988.Lack of spare0-adrenergicreceptorsin
, receptor thehumanheart.FASEB J. 2:A602.
7. Brown,L., N. M.Deighton, S. Bals, W. Sohlmann, H. R.Zerkowski,M.C. I disease (30) Michel, and0. E. Brodde. 1992. Spare receptorsfor f3-adrenoceptor-mediated
Protectingfrag- positiveinotropiceffects of catecholamines inthe human heart. J. Cardiovasc.
RNaseprotec- Pharmacol. 19:222-232.
priately sized
gic8.
Hadcock,J.R.,andC.C.Malbon. 1988.Down-regulationof beta-adrener-gicreceptors: agonist-induced reduction inreceptormRNA levels. Proc.Natl.
solated by gel Acad. Sci. USA. 85:5021-5025.
J that a single 9. Hadcock, J. R., H. Y. Wang, and C. C. Malbon. 1989.Agonist-induced
for destabilization of
f-adrenergic
receptormRNA:attenuationofglucocorticoid-in-accounts or ducedup-regulationoffl-adrenergicreceptors.J.Biol.Chem.264:19928-19933.
tdioactivity in 10. Bristow, M. R.,C. Durham, J.Klein, R. Rasmussen, J. D. Port, R. R.
lized material Gesteland, W. H. Barry, and A. M. Feldman. 1991.Down-regulationof
f-adren-iasthesemay ergicreceptors andreceptor mRNA in heart cellschronically exposedto norepi-twoprevious nephrine.Clin.Res. 39:256A.(Abstr.)
two previous npI 1.Lazar-Wesley,E.,J. R.Hadcock,C. C. Malbon,G. Kunos, and E. J. Ishac.
1991. Tissue-specific regulationofalB1f1,andf2-adrenergicreceptormRNAsby
cular myocar- thyroid state in the rat.Endocrinology,. 129:1116-1118.
12.Wang, A.M., M. V. Doyle, and D. F. Mark.1989. Quantitationof mRNA ntswas multi- bythepolymerasechainreaction.Proc.Natl.Acad.Sci.USA.86:9717-9721.
JRNA inven- 13. Feldman, A. M., P. E. Ray, C. M.Silan, J.A.Mercer,W.Minobe, and
myocytes. In M.R. Bristow. 1991. Selective geneexpressioninfailinghumanheart.
Circula-are selectively tion.83:1866-1872.
14. Bouaboula, M., P. Legoux, B. Pessegue, B. Delpech, X. Dumont, M.
it islikely that Piechaczyk,P.Casellas, and D.Shire. Standardization ofmRNAtitration usinga
NA measured polymerase chain reactionmethodinvolvingco-amplification witha
multispeci-myocytes. We ficinternalcontrol. 1992.
J. Biol.
Chem. 267:21830-21838.15. Lee, J. J.,and N.A.Costlow. 1987.Amoleculartitrationassaytomeasure
for:3-adren- transcriptprevalancelevels.Methods Enzymol. 152:633-648.
fractions ex- 16.Chomczynski,P., andN.Sacchi. 1987. Single-stepmethodofRNA
isola-tion by acid guanidinionthiocyanate-phenol-chloroform extraction. Anal.
Bio-is lower in the chem. 1622:156-159.
17.Frielle, T., S. Collins, K. W. Daniel, M. G. Caron, R. J. Lefkowitz, and
Al hearts. This B. K. Kobilka. 1987. Cloning ofthe cDNAfor the human beta I-adrenergic
i-regulationof receptor.Proc.Natl. Acad.Sci. USA. 84:7920-7924.
;.Further
18. Kawasaki, E.S.
1990. Amplification of RNA. In PCR Protocols-AGuidetoMethods andApplications.M. A.Innisand D. H.Gelfand,editors.
basis for this AcademicPress, Inc.,San Diego, CA.3-12.
19. Feldman,A. M., P. E. Ray, and M. R. Bristow. 1991. Expression of
a-subunits ofG proteins infailinghuman heart:areappraisalutilizing quantita-tive polymerase chain reaction.J.Mol. Cell. Cardiol.23:1355-1358.
20.Dixon,R.A.F., B. K. Kobilka, D.J.Strader,J. L.Benovic,H.G.
Dohl-man,T. Frielle,M.A.Bolanowski, C.D. Bennett, E. Rands, R.A.Diehl,etal. yofUtahMedi- 1986. Cloning of thegeneand cDNAformammalianfB-adrenergicreceptorand nsforPCR, Dr. homologywith rhodopsin.Nature(Lond.). 321:75-79.
Bristowetal.
OR
zf
z
21. Peterson,G.L. 1977. Asimplification of the protein assay methodof
Lowry etal. which ismoregenerally applicable. Anal. Biochem. 83:346-356. 22. Gyllensten, U. B., andH. A.Erlich. 1988. Generation of single-stranded
DNA by thepolymerase chain reactionandits applicationtodirectsequencing of the HLA-DQAlocus. Proc.Nati.Acad. Sci. USA.85:7652-7656.
23. Ladenson,P.W.,S.I.Sherman,K. L.Baughman,P.E.Ray, andA.M.
Feldman. 1992. Reversible alterations in myocardialgeneexpressionin ayoung
manwith dilated cardiomyopathy and hypothyroidism.Proc.Natl. Acad. Sci. USA.89:5251-5255.
24.Shimomura, H.,and A.Terada.1990. Primary structure of theratbeta-I adrenergic receptor gene. Nucleic AcidsRes. 18:4591.
25.Hamill,R.W.,P. D.Woolf,J. V.McDonald,L. A.Lee,and M.Kelly. 1987.Catecholaminespredictoutcomeintraumatic brain injury.Ann.Neurol. 21:438-443.
26. Sambrook, J.,E. F.Fritsch,and T. Maniatis. 1989. Molecular Cloning, a LaboratoryManual. 2nd edition. Cold Spring Harbor Press, Cold SpringHarbor, NY.
27.IhW-VahI,R., R. Marquetant, andR. H.Strasser. 1992. Subtype-selective regulation ofmRNAfor01-adrenergicreceptor in chronic heart failure.
Circula-tion.86(Suppl.I):I-767.(Abstr.)
28.Ungerer,M., M. Bohm, J. S. Elce, E. Erdmann, and M. J.Lohse. 1993.
Alteredexpression off-adrenergicreceptor kinaseand 3,1-adrenergicreceptorsin
thefailinghuman heart.Circulation.87:454-463.
29.Sylven, C.,M.Bronnegard, E. Jansson,P.Sotonyi, F.Liang-Xiong,F.
Waagstein, and A. Hjalmarsson. 1992. Increased
f3I
receptor expression at mRNAlevel in dilated cardiomyopathy. J.Am.Coll. Cardiol. 19:273A.(Abstr.)30. Sylven, C.,P.Arner, L. Hellstrom, E. Jansson, P. Sotonyi, A. Somogyi, and M. Bronnegard. 1991. Left ventricular
f3,
andl2adrenoceptor mRNAex-pression in normal and volumeoverloaded human heart. Cardiovasc. Res.
25:737-741.