Viral Myocarditis
Identification of Five Differentially Expressed Genes in Coxsackievirus
B3–Infected Mouse Heart
Decheng Yang, Jen Yu, Zongshu Luo, Christopher M. Carthy, Janet E. Wilson,
Zhewei Liu, Bruce M. McManus
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
Abstract—Differences in host susceptibility to viral myocarditis caused by a given strain of coxsackievirus B3 (CVB3) are
known to be largely related to host genetic factors. Little is known, however, about the key genes that encode
determinants (mediators) of myocarditis development or the nature of injury. To identify these genes and further
understand the molecular mechanisms of the disease process, we have used a murine model and the differential display
technique to fingerprint mRNAs from CVB3-infected mouse hearts. Total RNA was extracted from hearts of 4- and
10-week-old A/J(H-2a) mice at day 4 after CVB3 infection, and mRNAs were detected by reverse transcriptase–
polymerase chain reaction and subsequently analyzed on polyacrylamide DNA sequencing gels. The differentially
displayed bands were confirmed by Northern hybridization using the bands as cDNA probes. Twenty-eight upregulated
or downregulated bands were selected from the sequencing gels; among these, 2 upregulated and 3 downregulated
cDNA fragments were confirmed by Northern hybridization. DNA sequence analysis and GenBank searching have
determined that 4 of the 5 candidate genes are homologous to genes encoding Mus musculus inducible GTPase, mouse
mitochondrial hydrophobic peptide (a subunit of NADH dehydrogenase), mouse b-globin, and Homo sapiens
cAMP-regulated response element binding protein (CREB) binding protein (CBP), respectively. The remaining
candidate gene matches an unpublished cDNA clone, M musculus Nip21 mRNA (GenBank accession number,
AF035207), which is homologous to human Nip2, a Bcl-2 binding protein. Our data suggest preliminarily that both
structural and nonstructural genes are involved in myocarditis development. For the structural gene, b-globin, we further
confirmed its downregulation at the protein level by measuring the mean cell volume of red blood cells and found it was
marginally reduced in the CVB3-infected group (P,0.06), with no change in hemoglobin concentration. Cardiac
myoglobin concentration was also measured and found to be decreased (P,0.005), with a parallel decrease in total
soluble protein in the CVB3-infected mouse myocardium (P,0.01). We also noted that the ratio of myoglobin to total
protein was not significantly changed; this may be due to the downregulation of additional genes in the host heart, a
number being observed on the differential display gels. The significant downregulation of b-globin major gene
expression in the heart may be relevant to impaired cardiac function in both the early and late postinfection period. The
other identified nonstructural genes are known to be involved in regulation of gene expression, signal transduction
pathways, and apoptotic cell death. The altered expression of structural and nonstructural genes may play important roles
in the mediation of myocarditis development and perhaps other pathological processes in the heart. (Circ Res.
1999;84:704-712.)
Key Words: myocarditis n differential mRNA display n coxsackievirus
C
tion in the heart, pancreas, and other target organs. Viral
genomic RNAs have been clearly shown in cardiac myocytes
by in situ hybridization. Importantly, strong signals from
hybridization of negative-strand viral RNA with sense riboprobes in heart tissue indicates that viruses are replicating
within cardiac myocytes.5– 8 In addition, our previous data
and those of others have shown that virus infection can also
cause extensive direct cell injury in the heart and other
organs.7,9 –11
oxsackievirus B3 (CVB3) is a positive-sense, singlestranded RNA enterovirus. The CVB serogroup was
ranked No. 1 in causation of clinical cardiovascular diseases
by the World Health Organization in global surveillance of
viral diseases from 1975 through 1985.1 Recent molecular
characterization of isolates from community-acquired infections has reaffirmed the pathogenicity of CVB in human
diseases.2– 4 Furthermore, molecular diagnostic approaches
have provided direct and compelling evidence of viral infec-
Received July 1, 1998; accepted January 11, 1999.
From the Department of Pathology and Laboratory Medicine, University of British Columbia-St Paul’s Hospital, Vancouver, British Columbia, Canada.
Correspondence to Dr Decheng Yang, Cardiovascular Research Laboratory, St Paul’s Hospital, 1081 Burrard Street, Vancouver, BC, Canada V6Z 1Y6.
E-mail: dyang@prl.pulmonary.ubc.ca
© 1999 American Heart Association, Inc.
Circulation Research is available at http://www.circresaha.org
704
Yang et al
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
The susceptibility of certain host tissues and cells to CVB3
infection is influenced by many factors, such as sex, age,
nutrition, pregnancy, genetic background, and epidemics;
however, the most important factors are likely host genetic
background and age. In the human population, not everyone
is susceptible to CVB3 infection; despite a high attack rate,
most people will never develop myocarditis in a lifetime.
Similar observations have been made in animals. Different
strains of mice have different levels of susceptibility to active
CVB3 infection, and even those infection-susceptible animals
with different genetic background will have distinctive target
organ susceptibilities. For example, C57BL/6J mice are
resistant to CVB3-induced myocarditis, but they develop
severe hepatitis. A/J(H-2a) and BALB/c mice can develop
prominent myocarditis on viral infection but have a different
likelihood of succumbing to liver injury. In addition to
intrinsic myocyte factors responsible for disease, the immune
system is also important in disease progression. It has been
reported that T-cell populations in mice with different genetic
backgrounds respond differentially to CVB3 infection, influencing the outcome of the disease.10,12
There are also data to invoke age as an important factor for
host susceptibility to CVB3 infection. Infants are more
susceptible to CVB3 infection than young children or adults.
When human neonates acquire coxsackievirus infections at
birth, the mortality is extremely high,13 and myocarditis may
be severe. Neonatal and adolescent mice are susceptible to
CVB3 infection and myocarditis, and the level of susceptibility decreases with increasing age.14 –16 There are also many
studies to show that gene-expression patterns underlying
cardiac development change dramatically with maturation of
the host.17,18 The age effect on myocarditis may be closely
related to genetic background. In other words, a given age
may be an important time point in the maturation of genetic
expression in a host and as such may be a major determinant
of disease.
Although the mechanisms of CVB3-induced myocarditis
are not well understood, it is clear that the occurrence and
progression of myocarditis depend on a complicated series of
events involving interaction among genes of the virus and the
host. Host susceptibility to CVB3 infection may in part reflect
a parallel or sequential expression of certain genes encoding
mediators of viral replication, viral persistence, and cell
viability. These mediators may include proteins involved in
signal transduction pathways, regulation of host gene transcription or translation, and structural integrity of the cell.
During the CVB3 replicative cycle, virus-encoded proteases
can disrupt important host factors, such as eukaryotic translation initiation factor 4G19 and TATA binding protein.20 This
disruption of normal functioning of infected myocytes will
ultimately affect the gene regulation of neighboring myocytes. The cumulative effect is dysfunction and death of
infected myocytes and potential dysfunction and phenotypic
alterations of other cardiac cells.
Much effort has been devoted to the identification of such
regulatory genes; however, previous investigations have usually been confined to selected genes of interest and thus
require information about gene sequences to design oligonucleotide primers or probes to measure such gene expression.
April 2, 1999
705
Therefore, data obtained from studies of preselected genes
reflect rather isolated phenomena in a pathogenetic picture
that is particularly complex in vivo. To understand more
definitively the mechanisms of this disease, we used the
approach of differential mRNA display,21,22 systematically
assessing gene expression at the transcription level in an
A/J(H-2a) mouse enteroviral model. This technique has allowed us to compare thousands of gene expressions from
CVB3-infected and sham-infected mouse hearts side by side
on DNA sequencing gels.
In this article, we report the identification of 5 candidate
genes, the expressions of which were upregulated (2 genes) or
downregulated (3 genes) in the early postinfection period in
CVB3-infected A/J (H-2a) mouse hearts. Of the 5 candidate
genes, 1 is unknown, and 3 are nonstructural genes encoding
enzymes or regulatory factors. The remaining modulated
gene is b-globin. Downregulation of the b-globin major gene
in heart tissue may reflect direct injury of myocardium.
Identification of other nonstructural genes that are differentially expressed in infected heart may also provide new
avenues to understanding molecular mechanisms of myocarditis and other pathological processes in CVB3-infected mice.
The data presented here demonstrate that differential mRNA
display (coupled with other defined end points) may accelerate our capacity to dissect the pathogenesis of viral
myocarditis.
Materials and Methods
Animal and Tissue Processing
All experimental procedures conformed to the regulations of the
Canada Council on Animal Care and were approved by the Institutional Animal Care Committee. Inbred adolescent (4-week-old) and
young adult (10-week-old) A/J (H-2a) mice (12 animals for each age,
6 mice per group) were infected with CVB3 (CG strain, 105
plaque-forming units) or sham-infected with PBS intraperitoneally
and euthanized by CO2 narcosis on day 4 postinfection. The hearts
were harvested, and biventricular transverse slices of tissue were
used for plaque assays, in situ hybridization, and histopathology,
respectively, to confirm the presence and locale of infection, the
occurrence and nature of tissue injury, and the extent of inflammation in hearts with early myocarditis. A remaining portion was
immediately frozen in liquid nitrogen and stored at – 80°C for
subsequent RNA extraction.
Viral Plaque Assay
HeLa cells were cultured in 6-well plates at 3.53105 cells per well.
When the cells were '85% to 90% confluent, serial dilutions of
supernatant from the mouse cell lysate, prepared by homogenization,
were inoculated into each well. One hour after infection, the
inoculum was aspirated and cells in each well were overlaid with 2
mL of 0.7% warm agar containing 13 complete MEM with 10%
FBS. After incubation at 37°C for 2 days, cells were fixed with
Carnoy’s fixative and stained with 1% crystal violet solution for 5
minutes. The plates were washed, and the plaques were counted.
Supernatant of cell lysates from sham-infected tissue was used as a
control.
Histopathology
Transverse sections from the basal two thirds of the ventricular
myocardium from each mouse heart, as well as transverse sections
from each organ, were fixed in fresh 4% paraformaldehyde overnight, embedded in paraffin, sectioned, and stained with hematoxylin
and eosin or with Masson’s trichrome. The severity of virus-induced
disease was evaluated blindly by an experienced cardiovascular
706
CVB3-Induced Differential Gene Expression
pathologist on a scale of 0 to 4 (least to most) for coagulation
necrosis, contraction band necrosis, cytopathic effect, calcification,
cellular infiltrates, stromal collapse, and fibrosis in a fashion similar
to that previously described.7,23 This grading approach allows a clear
distinction between mild, moderate, and severe disease.
In Situ Hybridization
In situ hybridization was performed on both agar-embedded cell
preparations and tissue sections as previously described.24 Paraffinembedded tissue sections were permeabilized with proteinase K,
dehydrated, and hybridized using digoxigenin-labeled probe prepared by in vitro transcription according to the manufacturer’s
instructions. Hybridization was allowed to proceed at 42°C overnight
followed by stringent washing in 50% formamide and 23 SSC. After
anti-digoxigenin antibody was applied to the tissue sections, slides
were enzymatically developed and incubated with color substrate
overnight.
Isolation of Cellular RNAs and Northern Blot
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
Hearts from 4- and 10-week-old A/J (H-2a) mice, showing positive
signals (infection) in the in situ hybridization and myocarditic lesions
in the histopathologic staining, were selected for RNA isolation.
Heart tissue (60 to 100 mg) was homogenized, and total cellular
RNA was extracted with RNAzol B (Tel-Test) according to the
manufacturer’s instructions. Contaminating chromosomal DNA was
eliminated by treating the sample with RNA-free DNase I at 37°C for
15 minutes, followed by phenol/chloroform extraction. Samples of
total RNA (20 mg) were fractionated in 2.2 mol/L formaldehyde/1%
agarose gels and transferred onto nitrocellulose filters. Specific
probes were generated by labeling reamplified or cloned cDNA
fragments with [a-32P]dCTP (random prime DNA labeling kit,
Boehringer Mannheim). A GAPDH probe and ribosomal RNAs were
used as loading control and mRNA size markers, respectively. After
hybridization at 42°C overnight and a high-stringency wash at 60°C
in 0.3 mol/L NaCl, 0.03 mol/L sodium citrate (pH 7.0), and 0.1%
SDS, the filters were exposed to Kodak X-Omat AR film for 3 to 7
days with intensifying screens.
Differential mRNA Display
Differential mRNA display was carried out as described,16,17 except
that RNA samples were isolated from heart tissues rather than from
cell cultures. CVB3-infected and sham-infected A/J (H-2a) mice at
both 4 and 10 weeks of age were compared concurrently. Total RNA
(0.5 mg) was reverse transcribed in a 50-mL reaction with modified
1-base anchored oligo-dT primers (GenHunter Corp). The firststrand cDNAs were then amplified by polymerase chain reaction
(PCR) in the presence of an appropriate 39 1-base anchored oligo-dT
primer, dNTPs, [a-33P]dATP, and an arbitrary 10-mer 59 primer
(RNAimage Kits, GenHunter). Forty PCR cycles were run with the
following parameters: denaturation at 94°C for 30 seconds, annealing at 40°C for 2 minutes, and extension at 72°C for 30 seconds. In
the control, water was substituted for cDNA. Labeled PCR products
were analyzed by electrophoresis in 6% denaturing polyacrylamide
gels. Reproducibility of amplification for selected bands was confirmed by repeating the reactions at least 3 times with different
preparations of cDNA. Differentially upregulated bands were defined as those that were consistently present at higher density in
virus-infected samples than in sham-infected samples. Differentially
downregulated bands were defined as those that were consistently
present at lower density in virus-infected samples than in shaminfected samples.
Identification of the Differentially
Displayed mRNA
PCR bands of interest were excised from the gel and recovered by
rehydration, boiling, and ethanol precipitation. The eluted cDNA
was then reamplified using the same primers as those used in the
differential display reaction but in the absence of isotope. The
reamplified cDNA fragments were cloned directly into plasmid
vector pCRII using the TA cloning kit (Invitrogen). Plasmid DNA of
each clone was prepared following the standard method.25 Inserted
cDNAs were isolated, radiolabeled, and used as probes in a Northern
blot assay as described above. The cDNA fragments that generated
a Northern hybridization signal were sequenced on an Applied
Biosystems automated DNA sequencer using the Sanger et al26
dideoxy chain termination method. Nucleotide sequences obtained
were compared with known sequences by searching the GenBank,
EMBL, and EST (Expressed Sequence Tag) databases with the
BLAST family of programs.27 In some cases, the translated amino
acid sequences were also used to search the databases using the
BLAST programs.
Measurement of Hemoglobin and Myoglobin
To further characterize the b-globin major gene and confirm its
downregulation at the translational level, concentrations of hemoglobin and myoglobin in whole blood and in myocardium of
CVB3-infected mice, respectively, were determined by a method
described previously.28 Briefly, heart tissue was cut into small strips,
frozen on dry ice, and homogenized using an electronic homogenizer. The proteins were extracted with phosphate buffer, 0.4 mol/L at
pH 6.6. After centrifugation at 28 000g for 50 minutes, 3 mL of clear
supernatant was mixed with 1 mL of 0.067 mol/L K2HPO4, and CO
was then bubbled through the solution for 10 minutes. A pinch of
Na2S2O4 was added, and the solution was rebubbled with CO for 5
minutes before optical density measurements were taken at wavelengths of 538 and 568 mm.
Statistical Analysis
All values are presented as mean6SEM. Statistical significance was
evaluated using the Student t test for paired comparison, with
P,0.05 considered statistically significant.
Results
Confirmation of Infection and Disease
To confirm CVB3 infection of heart muscle, in situ hybridization for positive- and negative-strand viral RNA in
paraffin-embedded heart tissue sections was performed using
digoxigenin-labeled sense and antisense RNA probes transcribed from CVB3 cDNA (Figure 1). There was significantly more viral RNA in the myocardium of 4-week-old
mice than in that of 10-week-old mice. Negative-strand
RNAs (Figure 1B and 1E) indicate the replication of CVB3
genomic RNAs in these cells. Pancreatic tissue sections
(Figure 1C and 1F) from each mouse were hybridized with
antisense strand probe to serve as a positive control. The
pancreatic sections from 4- and 10-week-old mice, respectively, did not show as much age-related difference in amount
of infection as seen in the heart, suggesting a cardiac-specific
decrease in susceptibility of 10-week-old mice. To determine
the viral titer in tissues, viral plaque assays were conducted
using cell lysates. Infected heart tissues from 4-week-old and
10-week-old mice both produced plaques on the HeLa cell
monolayer; however, tissue from 4-week-old mice produced
significantly more plaques than that derived from 10-weekold mice, confirming previous reports.29 Histopathologic
evaluation was also carried out to confirm the presence and
severity of myocarditis in the hearts (Figure 2). Tissues from
4-week-old (Figure 2A) mice have more extensive myocyte
injury (coagulation necrosis and cytopathic effect) than those
from 10-week-old (Figure B) mice (statistical data not
shown). These data confirmed previous observations.29 Panels C and D are age-matched sham-infected heart tissues used
as a negative control.
Yang et al
April 2, 1999
707
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
Figure 1. In situ hybridization of CVB3-infected hearts (A, B, D, and E) and pancreas (C and F). Tissue sections of 4-week-old (A through C) and
10-week-old (D through F) A/J(H-2a) mice were hybridized with antisense strand (A, D, C, and F) and sense strand (B and E) RNA probes, respectively. Signals from negative strands of CVB3 in panels B and E suggest the replication of CVB3 in these cells. Two pancreas sections (C and F)
were used as positive controls. Sham-infected mouse tissues of both ages were used as negative controls (data not shown). RNA probes were prepared by in vitro transcription in the presence of digoxigenin-dUTP. All tissues were harvested 4 days after CVB3 inoculation.
Differential mRNA Display
To identify transcriptionally regulated genes of potential
relevance to myocarditis development, we compared differential mRNA display patterns in hearts of 4- and 10-week-old
mice infected with CVB3 with those of sham-infected animals at corresponding ages. PCR amplifications of reversetranscribed first-strand cDNAs were conducted using a 39
anchored oligoT12N primer and a 59 arbitrary 10-mer primer.
Figure 2. Histopathology of CVB3-induced myocarditis in A/J(H-2a) mice. CVB3-infected (A and B) and sham-infected (C and D) mouse
hearts were harvested 4 days after inoculation from 4-week-old (A and C) and 10-week-old (B and D) mice. Tissues were processed
and stained as described in Materials and Methods. Note the absence of pathological changes in hearts of sham-infected mice versus
cardiac damage in hearts of CVB3-infected mice. Also note the decrease in myocyte injury and death in 10-week-old mice (B) as compared with 4-week-old mice (A).
708
CVB3-Induced Differential Gene Expression
Figure 4. PCR reamplification of cDNA fragments recovered
from differential display gels. Number below each lane indicates
number of band selected from differential display gels. C indicates PCR negative control. Molecular mass markers were run
in lanes located at both sides and in the middle of the gel.
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
Figure 3. Differential display of mRNAs from hearts of shaminfected control (C) and CVB3-infected (V) mice at ages 4 and
10 weeks old. Hearts were harvested 4 days after CVB3 infection. Total RNAs were isolated and reverse transcribed with T12N
primer (where N is A, G, or C). Resultant cDNAs were amplified
by PCR using the same T12N primer in combination with arbitrary 10-mer primers in the presence of [a-33P]dATP and analyzed by electrophoresis on 6% DNA sequencing gels. Arrowheads indicate upregulated and downregulated bands.
Twenty-four primer combinations were applied to CVB3infected and sham-infected samples. The obtained PCR
products were analyzed on DNA sequencing gels. Figure 3
shows the mRNA display patterns with representative differentially expressed gene fragments, which are reproducible in
repeat experiments. In total, we selected 28 bands that
reflected upregulated or downregulated expression in CVB3infected mouse hearts. Here, we selected 2 gels to show the
altered expression for bands 2, 7, 12, and 15 (data not shown
for other bands). Certain bands (eg, band 2) demonstrated
age-related differential expression in CVB3-infected mice. In
other words, differential gene expression of certain genes is
greater or lesser for given genes in 4-week-old mice then in
10-week-old mice.
blot analysis. The ribosomal RNAs and GAPDH were used as
controls for RNA loading and gene transcriptional regulation.
Relative positions of 18S and 28S ribosomal RNAs are
indicated as size standards. The approximate transcript size
for each gene is listed in the Table. The age-related differential expression of band 2 was also confirmed by Northern
hybridization. Figure 5B shows that upregulation of gene
expression for band 2 is significantly higher in 4-week-old
mice than in 10-week-old mice. Ribosomal RNAs were used
as a loading control.
Identification of Differentially Displayed Genes
Bands confirmed by Northern hybridization were reamplified
by PCR and cloned into a TA cloning vector. The plasmid
DNAs were sequenced by the Sanger et al26 dideoxy chaintermination method. A GenBank search demonstrated that 4
PCR bands shared high sequence homology with respective
known genes in the databases (Table), whereas band 10
Northern Hybridization
To confirm the differential gene expression patterns observed
in gels, 28 altered bands were selected for further study. PCR
fragments in the bands were recovered and reamplified by
PCR (Figure 4) to make probes, which in turn were used to
hybridize RNA blots prepared with 20 mg of total RNA from
CVB3-infected or sham-infected hearts. Bands 8, 10, and 15
generated strong positive signals and also showed downregulation of transcription in CVB3-infected hearts as compared
with sham-infected mice (Figure 5A). On the other hand,
bands 2 and 7 generated strong positive signals and showed
upregulation of transcription in CVB3-infected samples.
Other bands did not produce Northern signals (data not
shown) and were eliminated from further analysis in the
present study. Such transcripts may not have been detected
because their levels were below the sensitivity of the RNA
Figure 5. Northern blot analysis. A, Northern blot confirming
altered gene expression for bands 2, 7, 8, 10, and 15. Total
RNA (20 mg per lane) obtained from CVB3-infected (V) and control (C) mice (4 weeks old) was hybridized with cDNA probes
generated by PCR reamplification of bands recovered from differential display gels. The autoradiograms show that signals
from infected hearts are much stronger (upregulated genes) or
weaker (downregulated genes) than those from control hearts.
Ribosomal RNAs and GAPDH RNA were used as size markers
and loading control, respectively. B, Age-related differential
expression of band 2. The procedure for analysis is the same as
that described above, except that RNA samples from both 4and 10-week-old mice were used. Upregulation of gene expression for band 2 is much higher in 4-week-old mice than in
10-week-old mice. Ribosomal RNAs were used as loading
control.
Yang et al
April 2, 1999
709
Results of GenBank Search of cDNA Fragments Identified by mRNA Differential Display
Band
mRNA
Expression
BLAST Search With DNA Sequence*
Identities, %
Approximate
mRNA size, kb
2
Up
M musculus interferon-g inducibly expressed GTPase
(IGTPase)
137/139 (98)
1.9
7
Up
Mouse hydrophobic peptide (mitochondrially encoded)
77/84 (92)
1.8
8
Down
Mouse b-globin major gene
247/260 (95)
1.2
10
Down
M musculus Nip 21 mRNA
222/248 (89)
2.2
15
Down
Homo sapiens clone cRT16, CBP
58/66 (88)
8.6
*Nucleotide sequences were compared with GenBank, EMBL, and EST databases using the BLAST family of
programs.27
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
matches an unpublished cDNA clone, Mus musculus Nip21
mRNA (AF035207). Nip21 shares high sequence homology
with human Nip2 (U15173).30 The GenBank accession numbers for these 5 clones (clones 2, 7, 8, 10, and 15) are
AF071427, AF071428, AF071431, AF071429, and
AF071430, respectively. The 4 known genes are described as
follows: the cDNA fragment from clone 2, an upregulated
gene, was found to be highly homologous to M musculus
inducibly expressed GTPase (IGTPase). The 139-bp fragment
is 98% identical to the published IGTPase sequence.31,32
Another upregulated cDNA fragment from band 7 shares
92% sequence homology with a mouse mitochondrial gene
encoding a hydrophobic peptide, a subunit of NADH dehydrogenase.33 The remaining 2 candidates are downregulated
genes. The first is from band 8 and has 95% sequence identity
to mouse b-globin major gene. The second downregulated
gene (gene 15) matches very well (88% identity) with human
CREB-binding protein (CBP). CREB is a cAMP response
element binding protein, which plays an important role in
transcriptional initiation.34 –36 CBP is a cotranscriptional factor of CREB.
Cardiac Myoglobin Concentration
To further assess the b-globin major gene and confirm the
downregulation at the protein level, mean cell volume of red
blood cells was measured and found to be marginally reduced
in the CVB3-infected group (P,0.06) with no change in
hemoglobin concentration (data not shown). Cardiac myoglobin concentration was decreased (P,0.005), with a parallel
decrease in total soluble protein in the CVB3-infected mouse
myocardium (P,0.01) (Figure 6). We also found that the
ratio of myoglobin to total protein was not significantly
changed. Such concordance may be due to the downregulation of additional genes, which have been observed in the
differential display gels.
as induced by a given coxsackievirus. It is postulated that
CVB3 infection of mice induces up- and downregulation of
certain gene expressions encoding determinants (mediators)
for heart disease development. These myocarditic determinants (known and unknown) may be produced by uninfected
or by infected heart cells and/or infiltrating immune cells.
Possible host determinants of disease include regulatory
protein factors (activator or suppressor of gene expression)
and host defense and immune mediators such as proteins
related to interferon and major histocompatability complex,
metabolic proteins, and structural proteins, among many
others, which serve as primary or secondary mediators of
disease development. Modulation of gene expression lies at
the center of regulatory mechanisms that control cellular
responses, and thus intermediate and long-term dysregulation
of certain genes is most likely critical to the etiology and
progression of myocarditis. Therefore, comparisons of differential gene expression between virus-infected and shaminfected tissues have the potential of providing information of
great pertinence to the pathogenesis of viral myocarditis.
Many methods have been developed to discern differential
gene expression in one cell population or another. Subtractive
hybridization37 is one such example; however, this technique
Discussion
CVB3 infection causes remarkable alterations in host cellular
physiology and morphology. This suggests that there is a
complicated interaction between virus and host in disease
development within the main target organs such as the heart.
Until now, most postulates regarding mechanisms of cardiac
myocyte injury and cardiac disease caused by CVB3 have
been based on observations gleaned from histopathology, cell
biology, virology, and immunology. Such studies have indicated that several host factors are pertinent to disease severity
Figure 6. Comparison of cardiac myoglobin concentrations in
myocardium of CVB3-infected and sham-infected mice. Myoglobin was prepared and measured according to the method of
Reynafarje.28 The difference in optical density at 538 and
568 mm was multiplied by the factor 117.3, and the resulting
value expressed the concentration of myoglobin in milligrams
per gram of wet tissue. Data are mean6SEM. *P,0.005;
**P,0.01.
710
CVB3-Induced Differential Gene Expression
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
gives incomplete recovery and selects only for either underexpressed or overexpressed genes. Furthermore, the screening is laborious. Differential mRNA display,21,22 which we
used in this study, can overcome these limitations. This
method has been used to study many diseases.38 – 41 In studies
of cancer, pathogenesis comparisons were made between 2
populations of in vitro cell lines at once. Our approach
involved the comparison of gene expression patterns derived
from RNAs isolated from a whole organ, the heart. This latter
approach preserved the entire pathophysiologic environment
associated with the disease process of myocarditis.
Using 24 primer combinations, we identified 28 differentially expressed cDNA bands that were reproducibly upregulated or downregulated in CVB3-infected mouse hearts for
both 4- and 10-week-old animals. Northern hybridization,
DNA sequencing, and GenBank searching further identified 5
of the 28 bands as genes encoding IGTPase, mitochondrial
hydrophobic peptide (a subunit of NADH dehydrogenase),
b-globin, CBP, and an unpublished cDNA clone homologous
to human Nip2, respectively.
One of the goals of this study was to develop strategies for
identification of host determinants responsible for myocarditis initiation and progression toward altered myocyte phenotype. As mentioned earlier, these mediators may be primary
or secondary factors in the disease process. The link of the 5
identified genes to CVB3-induced myocarditis is especially
interesting, because both structural and nonstructural genes
were found to be involved. Typically, the structural genes are
not considered as important as nonstructural genes in disease
development. This may be not true for all settings. Recently,
2 structural genes encoding dystrophin and muscle LIM
protein have been implicated in dilated cardiomyopathy,42,43 a
common heart disease that has been suggested to result from
long-term infection of CVB3. Modulation of the gene encoding b-globin in viral myocarditis provides a new suggestion
that metabolically active structural genes have importance in
heart disease occurrence. To further determine the relationship between the downregulation of b-globin and the myocardium injury, we measured the concentration of hemoglobin in blood and of myoglobin in the heart tissue from
CVB3-infected mice. Data showed that the concentration of
hemoglobin did not change with myocarditis. However, the
concentration of cardiac myoglobin did show a significant
decrease, with a parallel decrease in total soluble proteins.
This suggests that the downregulation of b-globin major gene
may relate to the decrease of cardiac myoglobin concentration. We also noted that the ratio of myoglobin to total protein
was not changed in the CVB3-infected group; this may be
due to the downregulation of other genes (primary or secondary responses), which have been observed on gels of differential mRNA display. Although many genes showed decreased expression, the downregulation of b-globin major
gene was most visible and reproducible as compared with
other bands. As is known, myoglobin is a cytosolic hemoprotein selectively expressed in cardiac and skeletal myocytes, in which it functions to augment delivery of oxygen for
mitochondrial respiration during heavy contractile work.44
Therefore, downregulation of this gene expression may cause
severe damage of the myocardium. These findings provide
support at the molecular level for the importance of direct
CVB3-induced injury of cardiac myocytes.7,10,11,45
IGTPase is representative of a newly identified group of
interferon-g-induced GTPases, the functions of which are
poorly understood.32 Recent reports indicate that IGTPase is
located predominantly in the endoplasmic reticulum of cells;
however, GTP binding status of IGTPase is independent of its
capacity to localize in the cellular compartment. Thus, the
function of IGTPase may involve protein processing or
trafficking.32 It has been known that interferon-g is a pleiotropic cytokine that regulates a variety of immunological and
inflammatory responses in fighting diseases. Interferon-g,
however, is also thought to be involved in many pathological
responses (eg, in multiple sclerosis), and its production
exacerbates disease symptoms and increases relapse rate.46 In
CVB3-infected hearts, overexpression of interferon-g by
invading immune cells would in turn induce the production of
IGTPase in the heart. The upregulated IGTPase gene expression may trigger signals attempting to stimulate protein
synthesis, process, and trafficking. The consequences of such
signaling may include maintenance of myocyte function in
the face of a destructive insult and, in theory, could contribute
to postinfectious hypertrophy.
A third gene is homologous to a mitochondrial gene
encoding a hydrophobic peptide (ND1), a subunit of NADH
dehydrogenase.33 It has been known that NADH dehydrogenase is an important enzyme in the tricarboxylic acid cycle.
The link between this gene and viral myocarditis is particularly interesting, because mitochondria serve as the center for
energy production via the tricarboxylic acid cycle and thus as
the major power source for contractility of the heart. Altered
expression of NADH dehydrogenase in mitochondria of
muscle cells may directly affect the function of the heart. In
addition, it is increasingly reported that mitochondrion is a
critical organelle involved in the initiation of apoptosis.47,48
Cytochrome c release from mitochondrial membrane can
activate caspases and in turn induce apoptotic cell death.49
NADH dehydrogenase subunit is a hydrophobic peptide
embedded within the mitochondrial membrane. Upregulation
of this gene may alter the membrane permeability and affect
the release of mitochondrially encoded factors. Whether the
altered expression of NADH dehydrogenase is related to the
induction of apoptosis remains to be determined.
A fourth gene encodes a nuclear protein, CREB binding
protein (CBP), that can bind to phosphorylated CREB to
serve as a cofactor in the regulation of transcription of target
genes. CREB is a basic leucine-zipper nuclear transcription
factor and may be involved in the development of dilated
cardiomyopathy.50,51 The activity of CREB is primarily regulated by protein kinase A–mediated phosphorylation of
Ser133.34 Activated CREB can bind to the sequence motif
known as the cAMP response element and participate in
cAMP-regulated gene expression.35,36 Phosphorylation of
CREB at Ser133 specifically enhances its binding ability to
CBP.35 There is also evidence that CBP cooperates with
upstream activators, such as c-jun, that are involved in
mitogen response transcription.35,36 Therefore, as a cofactor,
CBP is recruited to the promoter through interaction with
certain phosphorylated factors. CBP may play a critical role
Yang et al
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
in the transmission of cAMP-induced signals from cell
surface receptors to the transcriptional apparatus.
We predicted that the mediators of host susceptibility to
CVB3 infection would include factors involved in signal
transduction. CBP is indeed such a factor in the cAMP/
protein kinase A pathway. Although the precise mechanism
by which viral infection induces myocarditis via downregulation of CBP is not clear at present, the observation of
downregulation of a vital cofactor, CBP, in cAMP-induced
transcription has provided clues as to why the viral infection
can shut down or inhibit host protein synthesis. It is clear to
some degree that if gene expression is regulated through a
cAMP/protein kinase A pathway (eg, the phosphorylation of
transcription factor CREB must be catalyzed by protein
kinase A), its transcription will be inhibited by CVB3
infection. If such altered gene expression, in turn, regulates
cardiomyocyte metabolism or produces structural proteins in
the heart, such as cardiac myoglobin, viral infection will
cause myocardial injury both physiologically and structurally.
We found that there were more downregulated than upregulated bands on the differential display gels. This can be
attributed in part to the inhibition of key factors and cofactors
(eg, CBP) involved in cAMP-regulated gene transcription and
to the known inactivation of translation initiation factor
eIF4G by enteroviruses.52
The remaining candidate gene, mouse Nip21, may be the
most interesting gene among these 5 candidates in the context
of CVB3-induced cell death. First, this gene product is
homologous (66% similarity in 126 amino acid region) to the
GTPase-activating protein, Rho-GAP,30 which raises the
possibility that Nip21 is involved in signal transduction
pathways. Second, Nip21 shows high sequence homology to
the human Nip2 protein. This human protein is capable of
interacting with the apoptosis regulator Bcl-2 and a homologous protein, the adenovirus E1B 19-kDa protein.53 Mutational analysis indicated that this human Nip2 does not
interact with 19-kDa mutants defective in suppression of cell
death.53 Thus, the human isologue (Nip2) of Nip21 may be a
mediator interacting with Bcl-2 or E1B 19-kDa protein to
promote cell survival. Downregulation of Nip21 by CVB3
infection in the heart may therefore promote myocyte cell
death, an early feature of CVB3-induced disease.7 Furthermore, CVB3 has the potential to activate caspases in infected
cells.54 Such regulation of cell death proteins may be an
important early feature after myocardial infection before
immune infiltration, as well as later.
With the identification of at least 5 candidate determinants
of CVB3-induced myocarditis, we have obtained new clues to
the molecular pathogenesis of CVB3 myocarditis. Meanwhile, we also affirmed the value of differential mRNA
display analysis as a tool to provide insights into molecular
mediators associated with complex processes. In the case of
CVB3 myocarditis, we have used an established animal
model instead of cell lines, providing a real diseaseenvironment assessment. The present study only focused on
gene responses at day 4 after infection. Later time points of
study at days 14 to 21 after infection (and later) may identify
more gene expression related to the immune response and
reparative processes, as well as other secondary gene re-
April 2, 1999
711
sponses. It has been suggested that a subset of dilated
cardiomyopathy is the late phase of CVB3 myocarditis.55
Thus, studies at even later time points may reveal genes
responsible for myocardium hypertrophy and myocardial
remodeling. With progress of our future studies at different
time points, more candidate genes, upregulated and downregulated, known and unknown, are being identified and
characterized, particularly as they relate to cardiac myocyte
injury and impairment. An integrated picture of molecular
pathogenesis of this important form of heart disease will be
required before effective therapies can be developed.
Acknowledgments
These studies have been supported by the British Columbia Health
Research Foundation, the Heart and Stroke Foundation of British
Columbia and Yukon, and the Medical Research Council of Canada.
We are grateful to Dr Mary E. Russell for advice on differential
mRNA display experiments. We also thank George F. Schreiner and
SCIOS Inc for their assistance with the database searching.
References
1. Grist NR, Reid D. Epidemiology of viral infections of the heart. In:
Banatvala JE, ed. Viral Infections of the Heart. London, UK: Edward
Arnold; 1993:23–30.
2. Gupta HL, Khare S, Biswas A, Chattopadhya D, Kumari S. Coxsackie B
virus in the etiology of heart diseases in Delhi. J Commun Dis. 1995;27:
222–228.
3. Gauntt CJ, Pallansch MA. Coxsackievirus B3 clinical isolates and murine
myocarditis. Virus Res. 1996;41:89 –99.
4. McManus BM, Kandolf R. Myocarditis: evolving concepts of cause,
consequence, and control. Curr Opin Cardiol. 1991;6:418 – 427.
5. Hohenadl C, Klingel K, Mertsching J, Hofschneider PH, Kandolf R.
Strand-specific detection of enteroviral RNA in myocardial tissue by in
situ hybridization. Mol Cell Probes. 1991;5:11–20.
6. Kandolf R, Klingel K, Mertsching H, Canu A, Hohenadl C, Zell R,
Reimann B, Mertsching J, McManus BM, Foulis AK, Hofschneider PH.
Molecular studies on enteroviral heart disease: patterns of acute and
persistent infections. Eur Heart J. 1991;12:49 –55.
7. McManus BM, Chow LH, Wilson JE, Anderson DR, Gulizia JM, Guy
KL, Gauntt CT, Klingel KE, Beisel KW, Kandolf R. Direct myocardial
injury by enterovirus: a central role in the evolution of murine myocarditis. Clin Immunol Immunopathol. 1993;68:159 –169.
8. Ukimura A, Deguchi H, Kitaura Y, Fujioka S, Hirasawa M, Kawamura K,
Hirai K. Intracellular viral localization in murine coxsackievirus-B3 myocarditis: ultrastructural study by electron microscopic in situ hybridization. Am J Pathol. 1997;150:2061–2071.
9. Lodge PA, Herzum J, Olszewski J, Huber SA. Coxsackievirus B3 myocarditis: acute and chronic forms of the disease caused by different
immunopathogenic mechanisms. Am J Pathol. 1987;128:455– 453.
10. Andreoletti L, Hober D, Becquart P, Belaich S, Copin MC, Lambert V,
Wattre P. Experimental CVB3-induced chronic myocarditis in two
murine strains: evidence of interrelationships between virus replication
and myocarditis damage in persistent cardiac infection. J Med Virol.
1997;52:206 –214.
11. McManus BM, Chow LH, Wilson JE, Anderson DR, Kandolf R. Direct
damage of myocardium by enterovirus: old and new evidence for a
pre-eminent injurious role in murine myocarditis. In: Idiopathic Dilated
Cardiomyopathy: Structural and Molecular Mechanisms, Clinical Consequences. Berlin, Germany: Springer-Verlag; 1993:284 –293.
12. Blay R, Simperson K, Leslie K, Huber SA. Coxsackievirus-induced
disease, CD41 cells initiate both myocarditis and pancreatitis in DBA/2
mice. Am J Pathol. 1989;135:899 –907.
13. Kaplan MH. Coxsackievirus infection in children under 3 months of age.
In: Bendinelli M, Friedman H, eds. Coxsackievirus: A General Update.
New York, NY: Plenum Press;1988:241–252.
14. Khatib R, Chason JL, Silberberg BK, Lerner AM. Age-dependent pathogenicity of group B coxsackieviruses in Swiss-Webster mice: infectivity
for myocardium and pancreas. J Infect Dis. 1980;141:394 – 403.
15. Grodums EI, Dempster G. The age factor in experimental coxsackie B-3
infection. Can J Microbiol. 1959;5:595– 604.
712
CVB3-Induced Differential Gene Expression
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
16. Lyden D, Olszewski J, Huber S. Variation in susceptibility of Balb/c mice
to coxsackievirus group B type 3-induced myocarditis with age. Cell
Immunol. 1987;105:332–339.
17. Van Houten N, Huber SA. Genetics of coxsackie B3 (CVB3) myocarditis.
Eur Heart J. 1991(suppl D);12:108 –112.
18. Fung YW, Liew CC. Identification of genes associated with myocardial
development. Mol Cell Cardiol. 1996;28:1241–1249.
19. Etchison D, Milburn SC, Edery I, Sonenberg N, Hershey JW. Inhibition
of HeLa cell protein synthesis following poliovirus infection correlates
with the proteolysis of a 220,000-dalton polypeptide associated with
eukaryotic initiation factor 3 and a cap binding protein complex. J Biol
Chem. 1982;257:14806 –14810.
20. Clark ME, Lieberman PM, Berk AJ, Dasgupta A. Direct cleavage of
human TATA-binding protein by poliovirus protease 3C in vivo and in
vitro. Mol Cell Biol. 1993;13:1232–1237.
21. Liang P, Pardee AB. Differential display of eukaryotic messenger RNA
by means of the polymerase chain reaction. Science. 1992;257:967–971.
22. Liang P, Zhu W, Zhang X, Guo Z, O’Connell RP, Averboukh L, Wang
F, Pardee AB. Differential display using one-base anchored oligo-dT
primers. Nucleic Acids Res. 1994;22:5763–5764.
23. Chow LH, Gauntt CJ, McManus BM. Differential effects of myocarditic
variants of coxsackievirus B3 in inbred mice: a pathologic characterization of heart tissue damage. Lab Invest. 1991;64:55– 64.
24. Hohenadl CK, Klingel J, Mertsching PH, Hofschneider PH, Kandolf, R.
Strand-specific detection of enteroviral RNA in myocardial tissue by in
situ hybridization. Mol Cell Probes. 1991;5:11–20.
25. Sambrook T, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory
Manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1989:89 –91.
26. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chaintermination inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463–5467.
27. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W,
Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res. 1997;25:
3389 –3402.
28. Reynafarje B. Simplified method for the determination of myoglobin.
J Lab Clin Med. 1963;61:138 –145.
29. Yu J, Carthy CM, Bohunek L, Ellis J, Wilson JE, McManus BM, Yang
DC. Age-dependent pathogenicity of coxsackievirus B3 in A/J mice:
virus genome, virus titre and cytopathological effects in different organs.
Cardiovasc Pathol. In press.
30. Barfod ET, Zheng Y, Kuang W-J, Hart MJ, Evans T, Cerione RA,
Ashkenazi A. Cloning and expression of a human CDC42 GTPaseactivating protein reveals a functional SH3-binding domain. J Biol Chem.
1993;268:26059 –26062.
31. Taylor GA, Jeffers M, Largaespada DA, Jenkins NA, Copeland NG,
Woude GFV. Identification of a novel GTPase, the inducibly expressed
GTPase, that accumulates in response to interferon g. J Biol Chem.
1996;271:20399 –20405.
32. Taylor GA, Stauber R, Rulong S, Hudson E, Pei V, Pavlakis GN, Resan
JH, Vande Woude GF. The inducibly expressed GTPase localizes to the
endoplasmic reticulum, independently of GTP binding. J Biol Chem.
1997;272:10639 –10645.
33. Loveland B, Wang CR, Vonekawa H, Hermel E, Lindahl KF. Maternally
transmitted histocompatibility antigen of mice: a hydrophobic peptide of
a mitochondrially encoded protein. Cell. 1990;60:971–980.
34. Chrivia JC, Kwok RPS, Lamb N, Hagiwara M, Montminy MR, Goodman
RN. Phosphorylated CREB binds specifically to the nuclear protein CBP.
Nature. 1993;365:855– 859.
35. Arias J, Alberts AS, Brindle P, Claret FX, Smeal T, Karin M, Feramisco
J, Montminy M. Activation of cAMP and mitogen responsive genes relies
on a common nuclear factor. Nature. 1994;370:226 –229.
36. Chakravarti D, LaMorte V, Nelson MC, Nakajima T, Schulman IG,
Juguilon H, Montminy M, Evans RM. Role of CBP/P300 in nuclear
receptor signaling. Nature. 1996;383:100 –103.
37. Steeg PS, Bevilacqua G, Kopper L, Thorgeirsson UP, Talmadge JE,
Liatta LA, Sobel ME. Evidence for a novel gene associated with low
tumor metastatic potential. J Natl Cancer Inst. 1988;80:200 –204.
38. Liang P, Averboukh L, Keyomarsi K, Sager R, Pardee AB. Differential
display and cloning of messenger RNAs from human breast cancer versus
mammary epithelial cells. Cancer Res. 1992;52:6966 – 6968.
39. Sager R, Anisowicz A, Naveu M, Liang P, Sotiropoulou G. Identification
by differential display of alpha 6 integrin as a candidate tumor suppressor
gene. FASEB J. 1993;7:964 –970.
40. Utans U, Liang P, Wyner LR, Karnovsky MJ, Russell ME. Chronic
cardiac rejection: identification of five upregulated genes in transplanted
hearts by differential mRNA display. Proc Natl Acad Sci U S A. 1994;
91:6463– 6467.
41. Autiere MV, Feuerstein GZ, Yue TL, Ohstein EH, Douglas SA. Use of
differential display to identify differentially expressed mRNAs induced
by rat carotid artery balloon angioplasty. Lab Invest. 1995;72:656 – 661.
42. Muntoni F, Cau M, Ganau A. Deletion of the dystrophin muscle-promotor
region associated with X-linked dilated cardiomyopathy. N Engl J Med.
1993;329:921–925.
43. Arber S, Hunter JJ, Ross J Jr. MLP-deficient mice exhibit a disruption of
cardiac cytoarchitectural organization, dilated cardiomyopathy, and heart
failure. Cell. 1997;88:393– 403.
44. Wittenberg BA, Wittenberg JB. Myoglobin mediated oxygen delivery to
mitochondria of isolated cardiac myocytes. Proc Natl Acad Sci U S A.
1987;84:7503–7507.
45. Chow LH, Beisel KW, McManus BM. Enteroviral infection of mice with
severe combined immunodeficiency: Evidence for direct viral pathogenesis of myocardial injury. Lab Invest. 1992;66:24 –31.
46. Paty DW, Li DKB. The UBC MS/MRI study group and the IFNb
multiple sclerosis study group. Neurology. 1993;43:662– 667.
47. Liu X, Kim CN, Yang J, Jemmerson R, Wang X. Induction of apoptotic
program in cell-free extract: requirement for dATP and cytochrome c.
Cell. 1996;86:147–157.
48. Kluck RM, Bossy-Wetzet E, Green DR, Newmeyer DD. The release of
cytochrome c from mitochondria: a primary site for Bcl-2 regulation of
apoptosis. Science. 1997;275:1132–1136.
49. Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES,
Wang X. Cytochrome c and dATP-dependent formation of Apaf-1/
Caspase-9 complex initiates an apoptotic protease cascade. Cell. 1997;
91:479 – 489.
50. Leiden JM. The genetics of dilated cardiomyopathy: emerging clues to
the puzzle. N Engl J Med. 1997;337:1080 –1081.
51. Fentzke RC, Korcarz CE, Lang RM, Lin H, Leiden JM. Dilated cardiomyopathy in transgenic mice expressing a dominant-negative CREB
transcription factor in the heart. J Clin Invest. 1998;101:2415–2426.
52. Devaney MA, Vakharia VN, Lloyd RE, Ehrenfeld E, Grubman MJ.
Leader protein of foot-and-mouth disease virus is required for cleavage of
the p220 component of the cap binding protein complex. J Virol. 1988;
62:4407– 4409.
53. Boyd JM, Malstrom S, Subramanian T, Venkatesh LK, Schaeper U,
Elangovan B, D’Sa-Eipper C, Chinnadurai G. Adenovirus E1B 19 kDa
and Bcl-2 proteins interact with a common set of cellular proteins. Cell.
1994;79:341–351.
54. Carthy CM, Granville DJ, Watson KA, Anderson DR, Wilson JE, Yang
DC, Hunt DWC, McManus BM. Caspase activation and specific cleavage
of substrates after coxsackievirus B3-induced cytopathic effect in HeLa
cells. J Virol. 1998;72:7669 –7675.
55. O’Connel JB, Mason JW. Diagnosing and treating active myocarditis.
West J Med. 1989;150:431– 435.
Downloaded from http://circres.ahajournals.org/ by guest on October 1, 2016
Viral Myocarditis: Identification of Five Differentially Expressed Genes in Coxsackievirus
B3−Infected Mouse Heart
Decheng Yang, Jen Yu, Zongshu Luo, Christopher M. Carthy, Janet E. Wilson, Zhewei Liu and
Bruce M. McManus
Circ Res. 1999;84:704-712
doi: 10.1161/01.RES.84.6.704
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1999 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/84/6/704
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/