Mutation in myosin heavy chain 6 causes atrial septal defect.
ABSTRACT Atrial septal defect is one of the most common forms of congenital heart malformation. We identified a new locus linked with atrial septal defect on chromosome 14q12 in a large family with dominantly inherited atrial septal defect. The underlying mutation is a missense substitution, I820N, in alpha-myosin heavy chain (MYH6), a structural protein expressed at high levels in the developing atria, which affects the binding of the heavy chain to its regulatory light chain. The cardiac transcription factor TBX5 strongly regulates expression of MYH6, but mutant forms of TBX5, which cause Holt-Oram syndrome, do not. Morpholino knock-down of expression of the chick MYH6 homolog eliminates the formation of the atrial septum without overtly affecting atrial chamber formation. These data provide evidence for a link between a transcription factor, a structural protein and congenital heart disease.
- SourceAvailable from: PubMed Central[Show abstract] [Hide abstract]
ABSTRACT: CITED2 was identified as a cardiac transcription factor which is essential to the heart development. Cited2-deficient mice showed cardiac malformations, adrenal agenesis and neural crest defects. To explore the potential impact of mutations in CITED2 on congenital heart disease (CHD) in humans, we screened the coding region of CITED2 in a total of 700 Chinese people with congenital heart disease and 250 healthy individuals as controls. We found five potential disease-causing mutations, p.P140S, p.S183L, p.S196G, p.Ser161delAGC and p. Ser192_Gly193delAGCGGC. Two mammalian two-hybrid assays showed that the last four mutations significantly affected the interaction between p300CH1 and CITED2 or HIF1A. Further studies showed that four CITED2 mutations recovered the promoter activity of VEGF by decreasing its competitiveness with HIF1A for binding to p300CH1 and three mutations decreased the consociation of TFAP2C and CITED2 in the transactivation of PITX2C. Both VEGF and PITX2C play very important roles in cardiac development. In conclusion, we demonstrated that CITED2 has a potential causative impact on congenital heart disease.PLoS ONE 05/2014; 9(5):e98157. · 3.53 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Congenital heart disease (CHD) is the most common form of congenital human birth anomalies and a leading cause of perinatal and infant mortality. Some studies including our published genome-wide association study (GWAS) of CHD have indicated that genetic variants may contribute to the risk of CHD. Recently, Cordell et al. published a GWAS of multiple CHD phenotypes in European Caucasians and identified 3 susceptibility loci (rs870142, rs16835979 and rs6824295) for ostium secundum atrial septal defect (ASD) at chromosome 4p16. However, whether these loci at 4p16 confer the predisposition to CHD in Chinese population is unclear. In the current study, we first analyzed the associations between these 3 single nucleotide polymorphisms (SNPs) at 4p16 and CHD risk by using our existing genome-wide scan data and found all of the 3 SNPs showed significant associations with ASD in the same direction as that observed in Cordell's study, but not with other subtypes- ventricular septal defect (VSD) and ASD combined VSD. As these 3 SNPs were in high linkage disequilibrium (LD) in Chinese population, we selected one SNP with the lowest P value in our GWAS scan (rs16835979) to perform a replication study with additional 1,709 CHD cases with multiple phenotypes and 1,962 controls. The significant association was also observed only within the ASD subgroup, which was heterogeneous from other disease groups. In combined GWAS and replication samples, the minor allele of rs16835979 remained significant association with the risk of ASD (OR = 1.22, 95% CI = 1.08-1.38, P = 0.001). Our findings suggest that susceptibility loci of ASD identified from Cordell's European GWAS are generalizable to Chinese population, and such investigation may provide new insights into the roles of genetic variants in the etiology of different CHD phenotypes.PLoS ONE 09/2014; 9(9):e107411. · 3.53 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Patent foramen ovale (PFO) is an atrial septal deformity present in around 25% of the general population. PFO is associated with major causes of morbidity, including stroke and migraine. PFO appears to be heritable but genes involved in the closure of foramen ovale have not been identified. The aim of this study is to determine molecular pathways and genes that are responsible to the postnatal closure of the foramen ovale. Using Sprague-Dawley rat hearts as a model we analysed the dynamic histological changes and gene expressions at the foramen ovale region between embryonic day 20 and postnatal day 7. We observed a gradual loss of the endothelial marker PECAM1, an upregulation of the mesenchymal marker vimentin and α-smooth muscle actin, the elevation of the transcription factor Snail, and an increase of fibroblast activation protein (FAP) in the foramen ovale region as well as the deposition of collagen-rich connective tissues at the closed foramen ovale, suggesting endothelial-to-mesenchymal transition (EndMT) occurring during foramen ovale closure which leads to fibrosis. In addition, Notch1 and Notch3 receptors, Notch ligand Jagged1 and Notch effector HRT1 were highly expressed in the endocardium of the foramen ovale region during EndMT. Activation of Notch3 alone in an endothelial cell culture model was able to drive EndMT and transform endothelial cells to mesenchymal phenotype. Our data demonstrate for the first time that FO closure is a process of EndMT-mediated fibrosis, and Notch signalling is an important player participating in this process. Elucidation of the molecular mechanisms of the closure of foramen ovale informs the pathogenesis of PFO and may provide potential options for screening and prevention of PFO related conditions.PLoS ONE 09/2014; 9(9):e107175. · 3.53 Impact Factor
Mutation in myosin heavy chain 6 causes atrial
Yung-Hao Ching1, Tushar K Ghosh1, Steve J Cross1, Elizabeth A Packham1, Louise Honeyman2,
Siobhan Loughna2, Thelma E Robinson1, Andrew M Dearlove3,10, Gloria Ribas3,10, Andrew J Bonser1,
Neil R Thomas4, Andrew J Scotter4, Leo S D Caves5, Graham P Tyrrell6,10, Ruth A Newbury-Ecob7,
Arnold Munnich8, Damien Bonnet9& J David Brook1
Atrial septal defect is one of the most common forms of
congenital heart malformation. We identified a new locus
linked with atrial septal defect on chromosome 14q12 in a
large family with dominantly inherited atrial septal defect.
The underlying mutation is a missense substitution, I820N, in
a-myosin heavy chain (MYH6), a structural protein expressed
at high levels in the developing atria, which affects the binding
of the heavy chain to its regulatory light chain. The cardiac
transcription factor TBX5 strongly regulates expression of
MYH6, but mutant forms of TBX5, which cause Holt-Oram
syndrome, do not. Morpholino knock-down of expression of
the chick MYH6 homolog eliminates the formation of the atrial
septum without overtly affecting atrial chamber formation.
These data provide evidence for a link between a transcription
factor, a structural protein and congenital heart disease.
We carried out a genome-wide linkage screen in a large family with
dominantly inherited atrial septal defect (ASD) and no other cardiac
abnormalities (ASDF11) using the LMS-MD10 ABI PRISM set of
microsatellite markers (Applied Biosystems). We assumed the most
likely mode of inheritance to be autosomal dominance with incomplete
penetrance, on the basis of inspection of the pedigree. We established
linkage to markers on chromosome 14q12 with a maximum lod score
of 2.5 at recombination fraction (y) of 0.05 for D14S283 and defined a
minimal critical region between D14S261 and D14S275. We carried out
fine mapping of the critical region using seven additional microsatellite
markers and narrowed the disease-associated locus to an interval bet-
ween D14S283 and D14S972, with a peak lod score of 4.22 at a y of 0
for D14S990, assuming a penetrance of 80%. We derived an disease-
associated haplotype from alleles of D14S990, D14S1458 and D14S1457,
which cosegregated with disease or obligate carrier status in all except
one of the younger members of the pedigree (individual A39; Fig. 1).
The best candidate gene located in the critical region was a-myosin
heavy chain (MYH6), which is mainly expressed in atrial tissue1,2. We
screened the 39 exons of MYH6 in all members of family ASD F11
using denaturing high-performance liquid chromatography. We iden-
tified several polymorphisms, but only one (in exon 21; Fig. 2a) had
an allele that cosegregated with disease and obligate carrier status.
Sequencing identified a single nucleotide change (Fig. 2b), a
18429T-A transversion, in all affected family members, all obligate
carriers and individual A39 but not in unaffected family members or
in 200 chromosomes screened from healthy unrelated individuals.
This mutation causes the missense substitution I820N in the MYH6
protein. Amino acid 820 is conserved in type II myosins across all
species examined (data not shown), and never has a hydrophilic
amino acid at the site corresponding to the mutant residue.
The I820N substitution occurs in the MYH6 neck region, at the
start of the IQ domain, a sequence motif characteristic of a binding
site for the calmodulin-like light-chain accessory proteins3. Examina-
tion of the structural context for the homologous scallop myosin4
(Fig. 3) shows that the side chain of I820 projects from the helical
heavy chain into a large hydrophobic pocket on the C-terminal lobe of
the regulatory light chain (RLC). Molecular modeling studies suggest
that the I820N substitution is readily accommodated at this site in
terms of steric interactions (data not shown). But the substitution
places a polar side chain into an apolar environment, which suggests
that the mutant complex is destabilized (DDG of binding) relative to
the wild-type complex.
To understand the functional consequences of this mutation, we
carried out protein-protein interaction studies, first using a glu-
tathione S-transferase (GST) tag pull-down assay and then using
surface plasmon resonance (SPR)5. The mutation in MYH6 signifi-
cantly reduces its ability to pull down RLC (P o 0.001, Student’s
t-test; Fig. 4a,b).
Published online 27 February 2005; doi:10.1038/ng1526
1Institute of Genetics and2Centre for Biochemistry and Cell Biology, University of Nottingham, Queen’s Medical Centre, Nottingham NG7 2UH, UK.3MRC UK HGMP
Resource Centre, Hinxton, Cambridge CB10 1SB, UK.4School of Chemistry, University of Nottingham, Nottingham NG7 2RD, UK.5Department of Biology and
6Department of Chemistry, University of York, York YO10 5YW, UK.7Department of Clinical Genetics, Royal Hospital for Sick Children, Bristol BS2 8JJ, UK.
8Departement de Pediatrie, INSERM U-393, and9Service de Cardiologie Pediatrique, Hopital des Enfants-Malades, 75743, Paris, France.10Present addresses: MRC
geneservice, Babraham Bioincubator, Babraham, Cambridge CB2 4AT, UK (A.M.D.); Centro Nacional de Investigaciones Oncologicas, E-28029 Madrid, Spain (G.R.);
and PITO, London SE1 9QY, UK (G.P.T.). Correspondence should be addressed to J.D.B. (firstname.lastname@example.org).
NATURE GENETICS VOLUME 37 [ NUMBER 4 [ APRIL 2005423
© 2005 Nature Publishing Group http://www.nature.com/naturegenetics
We monitored kinetic interactions between
RLC and mutant and wild-type MYH6 using
SPR5on a BIAcore X instrument. The bind-
ing curves show that, for concentrations of
75–400 nM, GST-tagged mutant MYH6 had
a significantly lower response than did the
wild-type protein (Fig. 4c). We calculated
association and dissociation rates from the
sensorgrams and determined the dissociation
constant (KD). We obtained KDvalues of
1.72 ? 10?9M for wild-type MYH6 and
1.26 ? 10?7M for mutant MYH6, indicating
that the I820N mutation in MYH6 substan-
tially reduces the affinity of MYH6 for
We previously identified fifteen putative
TBX5-binding sites upstream of the tran-
scriptional start of MYH6 (ref. 6). TBX5 is
a transcription factor important during heart
development, and mutations in TBX5 cause
Holt-Oram syndrome, with defects in limb
and heart, including ASD7,8. To determine
whether TBX5 regulates MYH6 expression,
we assayed the activity of the MYH6 promo-
ter in the presence of wild-type TBX5 protein
or mutant TBX5 R237Q and TBX5 R279X
proteins, both of which are associated with
heart defects in Holt-Oram syndrome. Wild-
type TBX5 increased transcriptional activa-
tion by a factor of eight (Fig. 5), whereas
mutant TBX5 R237Q and TBX5 R279X produced significantly lower
levels of activation (TBX5 R237Q, P o 0.05; TBX5 R279X, P o 0.01;
Student’s t-test). This trend was maintained over a range of concen-
trations of protein expression constructs (data not shown).
Family studies and biochemical data suggest that the impaired
function and reduced level of MYH6 are central to the development of
ASD. To determine whether a reduction in MYH6 in vivo compro-
mises the formation of the atrial septum, we investigated the effect of
MYH6 knock-down using morpholinos in the developing chick heart.
We delivered fluorescently tagged morpholinos against chick atrial
myosin heavy chain (MHC) in ovo and observed the consequences for
septal development. MHC is expressed in the chick atrium from
stage 9 (ref. 9), and knock-down was induced at 52–53 h (stage 13/14)
of development, the same time that atrial septation is initiated
(50–53 h)10. We analyzed 14 embryos in which the morpholino
was present throughout the heart at stage 20. Of the seven embryos
in the experimental group, four were positive for the MHC transla-
tional start morpholino and three were positive for the MHC up-
stream morpholino. Of the seven control embryos, three were positive
for the Gene Tools standard control morpholino (mutated human
b-globin). We also analyzed three untreated embryos to determine the
normal range of septum formation. In each of the untreated embryos
and embryos treated with standard control and MHC mismatch
morpholinos, a septum developed from the dorso-cranial wall of the
primitive atrium (Fig. 6a–c). Of the seven ‘knock-down’ embryos
treated with the MHC translational start and MHC upstream mor-
pholinos, four showed no evidence of septum formation (Fig. 6d,f),
D14S1458 3.57 0
D14S1457 3.59 0
Figure 1 Fine mapping of the critical region of chromosome 14q in pedigree F11 with ASD. Family
members diagnosed with ostium secundum type ASD are shown in black and obligate carriers in gray.
Genotypes are shown for markers D14S283, D14S990, D14S1458, D14S1457, D14S972,
D14S264, D14S64 and D14S1032, and the disease-associated haplotype is boxed. Linkage was
calculated between the markers and the disorder assuming autosomal dominant inheritance with
80% penetrance. Individuals A2, A12, A34, A39 and A41 were scored as unaffected. The maximum
lod scores are shown for each marker at the relevant recombination fraction (y).
Retention time (min)
Figure 2 Mutation analysis of MYH6 exon 21. (a) Double peaked pattern observed on denaturing high-performance liquid chromatography analysis with the
extra peak at retention time 3.57 min (arrow). (b) Sequence trace of MYH6 exon 21 in an affected individual (antisense trace). In the sense strand there is
a T-A transversion at nucleotide 18429 of genomic sequence (nucleotide 2459 of cDNA; arrow).
424 VOLUME 37 [ NUMBER 4 [ APRIL 2005 NATURE GENETICS
© 2005 Nature Publishing Group http://www.nature.com/naturegenetics
and three had initiated septum formation but the septum was only a
small protuberance (Fig. 6e). Therefore, the knock-down morpholi-
nos disrupted atrial septation, whereas the control morpholinos
allowed septation to occur, producing left and right atrial chambers.
Using the one-tailed Fisher exact test, the knock-down experimental
group (treated with MHC translational start and MHC upstream
morpholinos) was significantly different (P o 0.05) from the control
group (treated with MHC mismatch and standard control morpho-
linos) with respect to the absence of septation. We observed no
obvious abnormalities in other parts of the knock-down embryos,
including other parts of the heart and other organs, compared with
control embryos (data not shown), consistent with the idea that
knock-down of MHC has a specific effect on atrial septation.
Mutations in several myosin genes are responsible for different
human disorders, including hypertrophic cardiomyopathy11,12. A
missense mutation in a conserved region of MYH6 that binds the
myosin essential light chain was identified in a individual with rare
hypertrophic cardiomyopathy of the elderly13. This case is difficult to
reconcile with the findings reported here. There is no evidence of
hypertrophic cardiomyopathy in family ASDF11, and the ventricular
structure is normal in all family members examined (as far as can be
determined). Mutation of MYH7 is not uncommon in individuals
with hypertrophic cardiomyopathy14, and the MYH6 and MYH7
proteins are structurally very similar. But these cardiac-specific pro-
teins have very different distributions in the human heart2,15,16. MYH7
could compensate for a MYH6 deficit in the adult ventricles, as it
comprises 95% of the functional myosin heavy chain protein present.
This may explain why the defect in family ASDF11 affects atrial
structure but not ventricular function. Alternatively, there may be a
fundamental difference in the functional requirement for the two
types of myosin light chain, and disruption of the essential light
chain–heavy chain interaction versus the RLC–heavy chain interaction
could have different phenotypic consequences. Nevertheless, we can-
not exclude the possibility that affected members of family ASDF11
will develop hypertrophic cardiomyopathy in the future, and follow-
up will be required to assess this.
Mutations of the cardiac transcription factors NKX2.5 (ref. 17) and
GATA4 (ref. 18) are associated with congenital heart disease, particu-
larly ASD. ASD is also involved in several syndromes whose under-
lying genes have been identified, including Holt-Oram syndrome and
TBX5. Downstream targets of TBX5, NKX2.5 and GATA4, such as
atrial natriuretic factor18–21, cardiac alpha actin22and connexin 40
(ref. 21) have been reported, although it remains to be determined
whether these genes represent the key targets with respect to ASD.
MYH6 expression is activated by TBX5, and mutations in TBX5 that
cause Holt-Oram syndrome substantially reduce activation of the
MYH6 promoter. Similarly, mutations of GATA4 that result in ASD
Figure 3 A model of myosin. The myosin heavy chain (purple) binds to the
essential light chain (yellow) and RLC (green). The enlarged region shows
the side chain of Ile820 (red) projecting into a hydrophobic pocket formed
by RLC binding to the neck region of the myosin heavy chain.
1 2141 6181101121 141161 181201221241
WT 75 nM
WT 100 nM
WT 200 nM
WT 300 nM
WT 400 nM
MUT 75 nM
MUT 100 nM
MUT 200 nM
MUT 300 nM
MUT 400 nM
Relative light chain binding (%)
Input Luciferase RLC
Figure 4 Interaction studies on wild-type and mutant MYH6 and myosin RLC. (a) GST pull-down assay using
radioactively labeled RLC incubated with immobilized GST, GST-MYH6 or GST-MYH6-I820N. No interaction is
observed with luciferase, but RLC interacts with both wild-type and mutant MYH6. GST protein alone did not
interact with RLC, and neither did an unrelated control protein (luciferase). (b) Binding of RLC by mutant
MYH6-I820N is significantly less than that by wild-type MYH6 (Student’s t-test based on four independent
experiments; ***P o 0.001). Quantification was done using a PhosphorImager (Molecular Dynamics), and
wild-type was set at 100%. (c) Overlayed sensorgrams from SPR studies of the association and dissociation
of wild-type (WT) and mutant (MUT) MYH6 proteins at different concentrations (75–400 nM) with the
immobilized myosin RLC, showing a 100-fold difference in binding affinity. RU, relative units.
NATURE GENETICS VOLUME 37 [ NUMBER 4 [ APRIL 2005425
© 2005 Nature Publishing Group http://www.nature.com/naturegenetics
affect the activation of MYH6 expression18. Therefore, for both
transcription factors, mutations that cause ASD affect the level of
MYH6 promoter activation.
Here we report the identification of a MYH6 mutation in a family
with dominantly inherited ASD and show that knock-down of chick
MHC ablates the formation of the atrial septum. We also show that
wild-type TBX5 activates MYH6 expression, whereas mutant forms of
TBX5 associated with Holt-Oram syndrome result in significantly
lower levels of activation. Our data suggest not only that MYH6 has a
pivotal role in the development of the atrial septum but also that
deficits in its level and function are crucial to the etiology of ASD.
Clinical details. We obtained approval for this study from the Institutional
Review Board of L’Hopital Necker. We obtained consent from adult subjects
and from parents on behalf of their children. All affected individuals had an
ASD of the secundum type. This was defined as an unrestrictive ASD with
increased right ventricular preload and increased pulmonary blood flow of
more than 1.5? systemic flow according to the Doppler continuity equation.
No individual had an ostium primum or sinus venosus defect. We carried out
transthoracic echocardiography (TTE) on all obligate carriers and siblings of
affected individuals. We obtained a subcostal view for all affected individuals of
generation IVand for obligate carriers A34, A12 and A41. We also carried out
TTE on all unaffected siblings in generation III (A29, A24, A17, A4, A8 and
A22) and in the unaffected spouses (A42, A33, A25, A18 and A13). None had
evidence of cardiac disease or indirect evidence for left-to-right shunt. All
offspring in generation IV underwent cardiac examination including echocar-
diography and TTE in the first year of life. In generation III, individual A34 had
a heart murmur noted at the age of 5 years but no TTE was carried out at this
time. Later examination did not detect an ASD, but this does not exclude an
ASD that closed spontaneously in childhood. No ASD detected in childhood
closed spontaneously. Individuals A26, A37, A44 and A40 had surgical closure
of ASD in childhood. Individual A36 awaits closure. Individuals A35 and A14
had ASD diagnosed at birth and repaired percutaneously. No other cardiac
defect was detected in any individual; in particular, there was no evidence of
familial hypertrophic cardiomyopathy. Left ventricular mass was normal in all
individuals who underwent TTE with no evidence of any echocardiography
abnormality related to familial hypertrophic cardiomyopathy using the echo-
cardiography criteria defined previously23. Upper limbs were examined and
were found to be normal. No additional malformations were observed.
Linkage analysis. We isolated DNA samples from blood of all consenting
individuals. We carried out PCR reactions for each marker individually as
described elsewhere24. In addition to the LMS-MD10 marker set, we selected
seven markers from the key interval on 14q for use in genotyping. We analyzed
PCR products in denaturing polyacrylamide gels on an ABI 377 automated
DNA sequencer and scored them using the GeneScan and GenoTyper analysis
software programs (Applied Biosystems). We carried out pairwise linkage
analysis with the MLINK program of the FASTLINK version 4.0P package25,
accessed through the Genetic Linkage User Environment interface (UK Human
Genome Mapping Project Resource Centre). For linkage calculations, we
assumed autosomal dominant inheritance with a gene frequency of 0.001
and penetrance set at 60%, 70%, 80%, 90% and 99%. Individuals A2, A41, A34,
A12 and A39 were scored as unaffected. Microsatellite allele frequencies were
assumed to be equal.
Mutation detection. We carried out denaturing high-performance liquid
chromatography analysis with 34 pairs of primers covering the whole exonic
sequence of MYH6 on the Transgenomic WAVE system using standard
protocols26. We eluted PCR products from the column using an acetonitrile
gradient in a 0.1 M triethylamine acetate buffer (pH 7), at a constant flow rate
of 0.9 ml min–1. We determined the temperature at which heteroduplex
detection occurred from the melting profile of the specific DNA fragments
using WaveMaker software from Transgenomic. We analyzed the fragments
over a 5-min gradient with a total run time of 7.8 min per sample. We adjusted
the linear acetonitrile gradient so that the peaks eluted between 3 and 5 min.
For sequencing, we carried out PCR reactions on 12.5 ng of genomic DNA
using standard protocols with Big Dye terminator Sanger sequencing, resolved
sequences on an ABI 377 and analyzed them with Sequence Navigator.
Mutation modeling. We modeled the effect of the I820N mutation using the
Scallop Myosin Papain digest pdb file (1B7T) in Swiss PDB Viewer V3.7 (b2)
(Glaxo Wellcome Experimental Research, freeware) and rendered it using POV-
Ray for Windows (The Persistance Of Vision Development Team, freeware).
We used the crystal structure of scallop myosin S1 fragment4as the template for
Figure 5 TBX5 activates transcription from the MYH6 promoter. H9c2
cells were cotransfected with a MYH6 reporter plasmid and plasmids
overexpressing TBX5, TBX5 R237Q or TBX5 R279X. Luciferase activity
after transfection of the reporter plasmid and control expression vector was
set at 1. The mutant TBX5 proteins resulted in significantly lower levels of
expression than wild-type TBX5 (Student’s t-test: *P o 0.05, **P o 0.01).
The bars represent an average of four independent experiments in which
each transfection was carried out in triplicate. The values shown are
means 7 s.d.
Figure 6 MHC is required for atrial septation in the chick. Chick hearts
treated with (a) standard control or (b) MHC mismatch morpholinos.
(c) Untreated (wild-type) chick heart. All controls show the developing
septum in the process of dividing the atrial chamber. (d–f) Chick hearts
treated with experimental morpholinos. (d,f) Examples of embryos with no
evidence of atrial septation. (e) One of the three embryos in which septation
had been initiated, resulting in a stump-like protuberance. Antibody staining
shows that MHC is specifically localized to the myocardium of the atrium
(At), including the septum (S) and the sinus venosus, but is absent from
the endocardium and the ventricular chamber.
426VOLUME 37 [ NUMBER 4 [ APRIL 2005 NATURE GENETICS
© 2005 Nature Publishing Group http://www.nature.com/naturegenetics
structural homology modeling with the MODELLER program27. We generated
five models of the MYH6 structure (native and I820N), which differed in their
final objective functionvalue. We chose for subsequent analysis the models with
the lowest objective function, which represented the best balance of the
template-derived constraints and their internal physical interactions.
Application of fluorescently tagged morpholinos to chick embryos in ovo
using Pluronic F-127 gel. We incubated fertile White Leghorn eggs (Henry
Stewart) at 38 1C in a humidified atmosphere. At Hamburger and Hamilton
stage 13/14, we created a window in the shell and dissected away the extra-
embryonic membranes overlying the heart. We removed B3 ml of albumin
and delivered 500 mM morpholino mixed 1:1 with 30% Pluronic
F-127 gel (BASF Corp) in Hank’s balanced salt solution (7 ml) onto the heart
using a pipette at 4 1C. We incubated the embryos for another 29 h before
collecting them and fixing them (in 4% paraformaldehyde) at stage 20.
Pluronic gel is liquid at 0–4 1C but sets when dropped onto the embryo at
physiological temperature. It moulds around the embryo, maintains elasticity
and remains in place for at least 12 h (refs. 28,29).
We obtained morpholinos from Gene Tools. We designed two experimental
morpholinos to target the chick atrial myosin heavy chain mRNA. One of these
(MHC translational start) targets the AUG and 22 bp downstream; the second
(MHC upstream) targets the 25 bp upstream of the AUG. In addition to the
Gene Tools standard control morpholino, we included a second control oligo
(MHC mismatch) designed to the same sequence as MHC translational start
but containing five mismatches to prevent it from binding to the MHC mRNA.
Morpholino sequences are available on request.
Analysis of embryos treated with morpholinos. After fixation, we examined
embryos using a Zeiss SV11 stereomicroscope to determine which had taken up
the fluorescently tagged morpholino into the heart. Only those embryos that
fluoresced throughout the heart were included in the study. We then dehy-
drated the embryos, embedded them in paraffin in a transverse orientation,
sectioned them serially at 8 mm and collected all sections. For immunohisto-
chemistry, we dewaxed the slides in xylene, rehydrated them and washed them.
We carried out antigen unmasking (with 10 mM sodium citrate (pH 6) in the
microwave at 650 W for 10 min) and removed endogenous peroxidase activity
(with 1% hydrogen peroxide for 10 min). We blocked slides (with 5% goat
serum for 60 min) before incubating them with a 1:50 dilution of the chick
MHC antibody30at 4 1C overnight. After washing, we used the avidin-biotin
phosphatase kit (StreptABComplex Duet kit, Dako) to visualize binding in
accordance with the manufacturer’s instructions. We counterstained slides
lightly in haematulum, dehydrated them and mounted them. We carried out
morphological analysis using a Zeiss Axioskop 2.
Protein purification. We generated proteins in Escherichia coli BL21(DE3) cells
after isopropyl-D-thiogalactopyranoside induction. We purified GST or GST
fusion proteins over glutathione sepharose 4B column (Amersham Bioscience).
We purified His-tagged MLC protein on Ni-NTA agarose column (Qiagen). We
estimated total protein using a Bradford assay (Biorad). Because the purified
GST-fusion proteins GST-MYH6 and GST-MYH6-I820N contain different
levels of copurified GST protein, we quantified intact fusion proteins on
SDS-PAGE gels after Coomassie staining.
In vitro pull-down assays. We amplified the coding region of MYH6,
corresponding to amino acid residues 535–976, from an expressed-sequence
tag (EST) clone (primer sequences available on request) and cloned it into
pGEX-2T (Pharmacia) to generate a GST fusion protein. We incorporated the
I820N mutation using a site-directed mutagenesis kit (Stratagene). We ampli-
fied RLC from an EST clone (primer sequences available on request) and
cloned it into pET28a (Novagen) to generate His-tagged RLC. We immobilized
GST, GST-MYH6 and GST-MYH6-I820N proteins on glutathione sepharose
beads and incubated the beads with35S-labeled RLC protein, or a luciferase
control protein, in 100 mM NaCl, 50 mM Tris-HCl (pH 7.4), 10% glycerol,
0.05% Nonidet-P40, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl
fluoride, 0.05% bovine serum albumin buffer at 4 1C for 2 h, before washing
them in buffer without BSA to remove nonspecifically interacting proteins.
We then eluted the bound proteins from the beads by adding SDS sample
buffer, resolved them by SDS-PAGE and visualized them by autoradiography.
We scanned band intensity on a PhosphorImager (Molecular Dynamics),
normalized values to those of the equivalent wild-type and mutant MYH6
proteins used in the pull-down and estimated from the same gel after
SPR. We measured association and dissociation kinetics by SPR using a
BIAcore X instrument (Biacore AB). We purified His-tagged RLC protein
and immobilized it on a NTA sensor chip (Biacore) in flowcell 2 (as per NTA
chip instructions). We used flowcell 1 as a reference to allow for background
refractive index correction. We prepared GST fusions of wild-type and mutant
MYH6 proteins in 10 mM HEPES buffer (pH 7.4), 150 mM NaCl, 50 mM
EDTA and 0.005% surfactant P20 buffer (filtered and degassed) at concentra-
tions from 75–400 nM. We exposed immobilized RLC to a 2-min injection of
each MYH6 solution at a flow rate of 20 ml min?1followed by a 2-min delayed
wash (buffer only) to allow the dissociation phase to be recorded. After heavy
chain binding, we carried out complete regeneration of the sensor chip surface
using 350 mM EDTA in running buffer (pH 8.3). We used BIAevaluation
software 3.0 (BIAcore) to analyze binding data using a simple Langmuir 1:1
drifting baseline model. We assessed the best fit of the experimental data for
association and dissociation models by w2analysis. We determined association
(kon, M?1s?1) and dissociation (koff, s?1) rates using a kinetics program
supplied with the system. We calculated the affinity constant KDusing the
relationship KD¼ koff/kon[M]. SPR measurements were done in triplicate at
25 1C and were reproducible to within 5% of stated values.
MYH6 reporter assays. We cloned a 4.4-kb fragment from the promoter region
of MYH6 into the pGL3-Basic reporter plasmid (Promega) and cotransfected
the rat cardiomyocyte cell line H9c2 with 0.75 mg of this plasmid and 0.5 mg
of pcDNA3.1 (Invitrogen) expressing TBX5, TBX5 R237Q or TBX5 R279X
(ref. 6) using Polyfect reagent (QIAGEN) and a total of 1.5 mg of DNA with
2 ng pRLTK (Renilla luciferase) control in six-well plates containing 1.5 ? 105
cells per well. After 24 h, we collected cells and measured levels of luciferase
reporter using the Dual-Luciferase Reporter Assay System (Promega). After
transfection, we carried out western blotting to confirm protein expression and
stability. We prepared cell extracts and detected expression levels using poly-
clonal TBX5 antibody and ECL plus kit (Amersham Bioscience).
URLs. Morpholinos can be obtained from Gene Tools (http://www.gene-tools.
GenBank accession numbers. MYH6 genomic sequence, Z20656; MYH6
coding sequence, D00943; MYH6 EST, AA153146; RLC EST, W17098; chick
MHC mRNA, D63466.
We thank A. Moorman for his gift of chick atrial myosin heavy chain antibody
and C. Nolan for advice on immunolabeling. This work was supported by the
British Heart Foundation, the Wellcome Trust and The Royal Society. The
genome screen, mutation detection and sequencing were done at the Medical
Research Council’s UK Human Genome Mapping Project Resource Centre
COMPETING INTERESTS STATEMENT
The authors declare that they have no competing financial interests.
Received 8 April 2004; accepted 19 January 2005
Published online at http://www.nature.com/naturegenetics/
1. Franco, D., Lamers, W.H. & Moorman, A.F. Patterns of expression in the developing
myocardium: towards a morphologically integrated transcriptional model. Cardiovasc.
Res. 38, 25–53 (1998).
2. Kurabayashi, M., Tsuchimochi, H., Komuro, I., Takaku, F. & Yazaki, Y. Molecular
cloning and characterization of human cardiac alpha- and beta-form myosin heavy
chain complementary DNA clones. Regulation of expression during development and
pressure overload in human atrium. J. Clin. Invest. 82, 524–531 (1988).
3. Rhoads, A.R. & Friedberg, F. Sequence motifs for calmodulin recognition. FASEB J.
11, 331–340 (1997).
4. Xie, X. et al. Structure of the regulatory domain of scallop myosin at 2.8 A resolution.
Nature 368, 306–312 (1994).
5. Fagerstam, L.G., Frostell-Karlsson, A., Karlsson, R., Persson, B. & Ronnberg, I. Bio-
specific interaction analysis using surface plasmon resonance detection applied to
kinetic, binding site and concentration analysis. J. Chromatogr. 597, 397–410 (1992).
NATURE GENETICS VOLUME 37 [ NUMBER 4 [ APRIL 2005427
© 2005 Nature Publishing Group http://www.nature.com/naturegenetics
6. Ghosh, T.K. et al. Characterization of the TBX5 binding site and analysis of mutations
that cause Holt-Oram syndrome. Hum. Mol. Genet. 10, 1983–1994 (2001).
7. Li, Q.Y. et al. Holt-Oram syndrome is caused by mutations in TBX5, a member of the
Brachyury (T) gene family. Nat. Genet. 15, 21–29 (1997).
8. Basson, C.T. et al. Mutations in human TBX5 cause limb and cardiac malformation in
Holt-Oram syndrome. Nat. Genet. 15, 30–35 (1997).
9. Oana, S. et al. The complete sequence and expression patterns of the atrial myosin
heavy chain in the developing chick. Biol. Cell 90, 605–613 (1998).
10. Arrechedera, H., Alvarez, M., Strauss, M. & Ayesta, C. Origin of mesenchymal tissue in
the septum primum: a structural and ultrastructural study. J. Mol. Cell. Cardiol. 19,
11. Geisterfer-Lowrance, A.A. et al. A molecular basis for familial hypertrophic cardiomyo-
pathy: a beta cardiac myosin heavy chain gene missense mutation. Cell 62, 999–1006
12. Poetter, K. et al. Mutations in either the essential or regulatory light chains of myosin
are associated with a rare myopathy in human heart and skeletal muscle. Nat. Genet.
13, 63–69 (1996).
13. Niimura, H. et al. Sarcomere protein gene mutations in hypertrophic cardiomyopathy of
the elderly. Circulation 105, 446–451 (2002).
14. Seidman, J.G. & Seidman, C. The genetic basis for cardiomyopathy: from mutation
identification to mechanistic paradigms. Cell 104, 557–567 (2001).
15. Miyata, S., Minobe, W., Bristow, M.R. & Leinwand, L.A. Myosin heavy chain isoform
expression in the failing and nonfailing human heart. Circ. Res. 86, 386–390 (2000).
16. Reiser, P.J., Portman, M.A., Ning, X.H. & Schomisch Moravec, C. Human cardiac
myosin heavy chain isoforms in fetal and failing adult atria and ventricles. Am. J.
Physiol. Heart Circ. Physiol. 280, H1814–H1820 (2001).
17. Schott, J.J. et al. Congenital heart disease caused by mutations in the transcription
factor NKX2-5. Science 281, 108–111 (1998).
18. Garg, V. et al. GATA4 mutations cause human congenital heart defects and reveal an
interaction with TBX5. Nature 424, 443–447 (2003).
19. Lee, Y. et al. The cardiac tissue-restricted homeobox protein Csx/Nkx2.5 physically
associates with the zinc finger protein GATA4 and cooperatively activates atrial
natriuretic factor gene expression. Mol. Cell. Biol. 18, 3120–3129 (1998).
20. Hiroi, Y. et al. Tbx5 associates with Nkx2-5 and synergistically promotes cardiomyocyte
differentiation. Nat. Genet. 28, 276–280 (2001).
21. Bruneau, B.G. et al. A murine model of Holt-Oram syndrome defines roles of the
T-box transcription factor Tbx5 in cardiogenesis and disease. Cell 106, 709–721
22. Sepulveda, J.L. et al. GATA-4 and Nkx-2.5 coactivate Nkx-2 DNA binding targets:
role for regulating early cardiac gene expression. Mol. Cell. Biol. 18, 3405–3415
23. Charron, P. et al. Diagnostic value of electrocardiography and echocardiography for
familial hypertrophic cardiomyopathy in genotyped children. Eur. Heart J. 19, 1377–
24. Vulliamy, T. et al. The RNA component of telomerase is mutated in autosomal
dominant dyskeratosis congenita. Nature 413, 432–435 (2001).
25. Lathrop, G.M., Lalouel, J.M., Julier, C. & Ott, J. Multilocus linkage analysis in humans:
detection of linkage and estimation of recombination. Am. J. Hum. Genet. 37, 482–
26. Underhill, P.A. et al. Detection of numerous Y chromosome biallelic polymorphisms by
denaturing high-performance liquid chromatography. Genome Res. 7, 996–1005
27. Sali, A. & Blundell, T.L. Comparative protein modelling by satisfaction of spatial
restraints. J. Mol. Biol. 234, 779–815 (1993).
28. Becker, D.L. et al. Roles for alpha 1 connexin in morphogenesis of chick embryos
revealed using a novel antisense approach. Dev. Genet. 24, 33–42 (1999).
29. Becker, D.L. & Mobbs, P. Connexin alpha1 and cell proliferation in the developing
chick retina. Exp. Neurol. 156, 326–332 (1999).
30. de Groot, I.J. et al. Isomyosin expression in developing chicken atria: a marker for
the development of conductive tissue? Anat. Embryol. (Berl) 176, 515–523 (1987).
428VOLUME 37 [ NUMBER 4 [ APRIL 2005 NATURE GENETICS
© 2005 Nature Publishing Group http://www.nature.com/naturegenetics