| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2000;102:IV-34.)
© 2000 American Heart Association, Inc.
Special Anniversary Issue |
From the Department of Medicine, Baylor College of Medicine, Houston, Tex.
Correspondence to Robert Roberts, MD, Department of Medicine, Baylor College of Medicine, 6550 Fannin, MS SM-677, Houston, TX 77030
Key Words: genetics cardiomyopathies hypertrophy
As we enter into the new millennium, the aura of molecular genetics conjures up an exciting and challenging future for the cardiologist. The prospects for improvement in the diagnosis, prevention, and treatment of cardiovascular disease are limited only by our present knowledge and imagination. Speculations on the future aside, 2 major scientific feats have already affected cardiology in the past decade and will continue to do so in the next century: the Human Genome Project, which is nearing completion, and the ongoing intense effort to identify genes responsible for cardiovascular function and disease. Cardiac myocytes react to physiological and pathological stimuli with a growth response that leads to an increase in sarcomeres generated in parallel, giving rise to hypertrophy or, in sequence, giving rise to dilatation, or the combination thereof. Insights fundamental to ultimate elucidation of this growth process are hoped to be gleaned from understanding natures errors (inherited disorders) that represent a paradigm of hypertrophy or dilatation. There have been several surprises, namely, the hypertrophy of familial hypertrophic cardiomyopathy (FHCM) being a compensatory response to defects in sarcomeric structural protein, whereas dilated cardiomyopathy (DCM) appears to be an impaired growth response due to defects in cytoskeletal proteins, including the laminae of the nuclear envelope. The genetic revolution provides the engines of ingenuity to achieve even greater progress in the imminent and distant future.
FHCM as a Paradigm for Elucidating the Left Ventricular Hypertrophic Growth Response to Physiological and Pathological Stimuli
Molecular Basis and Pathogenesis of Hypertrophy
in FHCM
Discovery of the first gene responsible for FHCM in
19901 was exciting,
but that it encoded ß-myosin heavy chain (ß-MHC), a sarcomeric
protein, was unexpected. Because the abnormality in HCM is excessive
growth (hypertrophy), a mutation in a growth factor or a
growth-signaling pathway would be more expected. ß-MHC mutations as a
cause for FHCM were detected throughout North
America2 and
Europe.3 A total of 9
genes have now been identified, with multiple mutations responsible for
FHCM, and all of the genes encode for sarcomeric structural proteins.
The 2 most common genes for the disease are MYH7
(ß-MHC) and MYBPC3 (cardiac myosinbinding protein
C).4 5 6 7
ß-MHC mutations were shown to be present in affected family members
but not in unaffected members or the general population. Kochs
postulates were soon satisfied. The human wild-type gene (normal) and
the mutant form of several of these genes were expressed either through
transgenesis8 9 10 11
or homologous
recombination,12 and
the phenotype induced was similar to that observed in humans with FHCM.
The main pathology of human FHCM disease is sarcomere disarray,
increased interstitial fibrosis, and cardiac hypertrophy. Sarcomere
disarray, the hallmark of FHCM, has been consistently observed in these
genetic
models8 9 10 11 12
after expression of ß-MHC, troponin T, myosin binding protein C
mutations, and, most recently, in the rabbit after expression of
ß-MHC.9 Most of the
genetic animal models also exhibit increased interstitial fibrosis and
some alteration in myocardial function; however, very
little,12 if any,
hypertrophy is observed in the mouse
models.9 The
phenotype of HCM was present only in the animals expressing the mutant
gene and not in animals expressing the normal human gene. The heart of
the mouse has
-MHC, whereas in humans 98% of the myosin is ß-MHC.
The cardiac myocyte of the rabbit also has ß-MHC. Expression of the
human mutant ß-MHC gene in the rabbit exhibits a phenotype that is
virtually identical to the phenotype observed in human FHCM, which
includes sarcomere disarray, increased interstitial fibrosis,
hypertrophy, sudden death, and impaired diastolic function, with normal
systolic
function.9
To decipher the molecular events and their temporal sequence
leading to the phenotype required dissection in simpler in vitro
models. ß-MHC is expressed in the right and left ventricles and in
many skeletal
muscles.13 Several
studies show that isolated skeletal muscles expressing the mutant
ß-MHC exhibited impaired
contractility.14 15 16
Contractility and myosin filament formation of the expressed mutant
protein were impaired, as determined by in vitro
models.17 18 19
Analysis of a 3D crystalline structure of skeletal MHC showed that the
ß-MHC mutations involved several domains critical to sarcomeric
contraction, such as impaired actin binding, ATP generation, or calcium
sensitivity.20 21
Thus, there is a specific molecular defect induced in the ß-MHC
molecule to explain the in vitro impaired contractility. Expression of
a mutant ß-MHC gene in the intact feline cardiac myocyte exhibited
sarcomere disarray by 72
hours,22 and similar
results were observed after expression of troponin T in cardiac feline
myocytes.23
Expression of a mutant troponin T in adult cardiac rat
myocytes24 and in
myotubes19 25
was associated with decreased cell shortening and impaired
contractility. In addition, myocytes isolated from the heart of
-MHC
mutant heterozygote mice exhibited impaired contraction and
relaxation.26
Furthermore, detection of cardiac myocyte shortening by a laser system
showed that adult feline cardiac myocytes expressing the mutant ß-MHC
gene exhibited impaired
contractility23
before the development of sarcomere disarray. Most recent studies have
also shown that cardiac contractility is impaired in the transgenic
mouse before the development of sarcomere
disarray.27 The
mutant gene was shown to be incorporated into the feline myocyte
myofibrils23 and,
more recently, into the myofibrils of the heart of transgenic
mice8 and transgenic
rabbits.9 One recent
study involved cardiac myosinbinding protein C
mutations.28
Expression of mutated proteins in fetal rat cardiomyocytes was
associated with altered expression and incorporation in the sarcomeres,
and a novel putative myosin binding site on cardiac myosinbinding
protein C was suggested by hydrophobic cluster
analysis.28 One may
hypothesize that cardiac myosinbinding protein C mutants act as
ß-MHC and troponin T mutants to impair cardiac
contractility.
Thus, the primary genetic defect is impaired contractility,
which stimulates the release of a growth factor(s) that leads to
interstitial fibrosis and hypertrophy. Despite the presence of the
mutant protein in equal abundance in the right and left ventricles,
hypertrophy seldom develops in the right ventricle, suggesting that the
high pressure of the left ventricle is the stimulus. Furthermore,
relief of outflow tract obstruction with septal alcohol injection is
associated with regression of hypertrophy and collagen, indicating that
the stimulus is increased
pressure.29 30
Thus, the defective sarcomeric protein triggers the release of growth
factors that lead to cardiac hypertrophy and collagen formation, as
summarized in Figure 1
, a modification of the figure by
Marian.31
|
Problems to Be Resolved
In the mouse, unlike in human FHCM, there is decreased
systolic function, perhaps because there is only minimal
hypertrophy,8 whereas
in the rabbit, there is extensive hypertrophy with normal systolic
function.9 Techniques
such as subtraction hybridization or DNA microarrays may detect
different growth factors that account for the different hypertrophic
responses in these 2 models.
The culprit responsible for sudden cardiac death remains unknown in FHCM, as it does for sudden cardiac death in myocardial ischemia and cardiac failure, but for all of these, it is postulated to be fibrosis.32 33 The excitement for the future will be elucidating the triggers for the growth response and development of a therapeutic strategy to decrease the incidence of sudden death. We have the genetic animal models, and so the time has come to evaluate known therapies, such as ACE inhibitors and angiotensin II receptor blockers and develop novel drugs, such as specific growth factor inhibitors.
Recognition That the Cytoskeleton Is a Major Determinant of the Cardiac Growth Response
The cytoskeleton is a complex set of protein filaments that is very important to generate shape and movement. The sarcomere is thought to represent an evolutionary specialization of filament proteins that originally served functions that were common to early cells, such as cell motility or chromosome movement. One of these sarcomeric proteins, actin, is also involved in the pathogenesis of DCM.34 In this case, the actin mutations are in the subdomains that interact with actin and not with myosin. The authors thus proposed that the pathogenic mechanism for the development of DCM was a defect in force transmission, whereas hypertrophic cardiomyopathy is caused by chronic reduction of force generation.
Unexpectedly, other classes of cytoskeletal proteins,
dystrophin and the desmin intermediate filaments, were shown to
cause dilated cardiomyopathies. The dystrophin gene, the first
gene to be identified by positional cloning, is part of a multisubunit
complex that confers a structural link between the extracellular matrix
and the actin skeleton. Mutations in the dystrophin gene cause Duchenne
muscular dystrophy and its milder allelic variant, Becker muscular
dystrophy, which develop profound cardiomyopathies. Dystrophin
abnormalities also seem to underlie acquired forms of
DCM.35 Another
important component of the extrasarcomeric cytoskeleton is desmin. A
subset of skeletal and cardiac myopathies have as a hallmark abnormal
deposits of desmin
aggregates.36 These
myopathies are characterized by muscle weakness, restrictive
cardiomyopathy, cardiomyocyte hypertrophy, cardiac dilatation,
conduction blocks, arrhythmias, and heart failure. Mutations were found
not only in the desmin gene but also in the
-B crystallin chaperone
gene.37 38
-B crystallin interacts specifically with desmin and has a role in
desmin intermediate filament assembly. Of particular significance was
the finding that a missense mutation in the desmin gene was responsible
for DCM, with no discernible clinical involvement of the skeletal or
smooth
muscles.39
Thus, a new concept is emerging from these genetic data. The extrasarcomeric cytoskeleton plays a major role in the growth response of the mammalian heart and in the pathogenesis of cardiomyopathies. Morphological studies show that expression of cytoskeletal proteins is altered in acquired forms of heart failure.40 Major hopes now reside in the annotation of the human genome sequence and the availability of high-throughput methods for DNA screening.
Recognition of the Role of the Nuclear Envelope in Cardiac and Skeletal Muscle Disease
Phenotypic Variability of Nuclear
Envelopathies
The story began in 1902 at the "Clinique Nerveuse de
la Salpêtrière," where 2 neurologists described a myopathy with
familial
contractures.41 It
was only in 1979 that it was recognized as a distinct clinical entity
and called Emery-Dreifuss muscular dystrophy
(EDMD).42 By
adulthood, affected individuals invariably develop heart block
requiring pacing or severe dysrrhythmias sometimes requiring an
implantable
defibrillator.43
Left ventricular dilatation and cardiac failure can sometimes occur.
The disease is transmitted as an X-linked trait (X-EDMD) or as an
autosomal dominant trait
(AD-EDMD).44
In 1994, the gene responsible for X-EDMD,
STA, was discovered. Its product was called
emerin.45 It was a
surprise to find that emerin is a nuclear integral membrane protein. In
1999, the lamin A/C gene, LMNA, was identified as the
disease gene of
AD-EDMD.46
LMNA encodes lamins A and C, which derive from
alternative splicing at the 3'-end of the gene. Lamins A and C are
components of the nuclear envelope and are located in the
lamina47
(Figure 2A
). Very unexpectedly, in the majority of
affected members of one of the French families analyzed in that study,
the disease was confined exclusively to the heart and was associated
with arrhythmias, left ventricular dysfunction, DCM, and a high
incidence of sudden
death.48 Recently,
we reported the variability of the phenotype and the spectrum of
LMNA mutations in 53 AD-EDMD
patients.49 Mutation
analysis
(Figure 2A
) identified 18 mutations in LMNA,
distributed between exons 1 and 9, in the region of
LMNA that is common to lamins A and C. Clinical
expression of LMNA mutations ranged from patients
expressing the full clinical picture of EDMD to those characterized
only by cardiac involvement: 12 of the 53 patients showed cardiac
involvement exclusively. There was no clear correlation between the
phenotype and the type or localization of the mutations within the
gene. The obvious immediate clinical application is that in patients
presenting a life-threatening familial or sporadic cardiac-restricted
phenotype similar to that described in the above reports, one should
check for mutations in the lamins A/C gene.
|
The year 2000 is adding a very intriguing complexity to the
role of LMNA products in the pathogenesis of human
diseases. First, it was found that mutations in this gene cause another
muscular dystrophy, the limb girdle muscular dystrophy type LGMD1B. The
LGMD1B form is inherited as an autosomal dominant trait. It is slowly
progressive, with age-related atrioventricular cardiac conduction
disturbances and DCM and absence of early
contractures50 (Figure 2B
). We identified LMNA mutations in
3 very-well-characterized LGMD1B families, demonstrating that LGMD1B
and AD-EDMD are allelic
disorders.51 In
addition, a deletion was identified in a family with DCM and skeletal
muscle
abnormalities.52
Second, missense mutations were reported to be implicated in
Dunningan-type familial partial lipodystrophy
(FPLD)53
(Figure 2B).54
Patients with FPLD are born with normal fat distribution but then lose
subcutaneous fat from their extremities, trunk, and gluteal region
after the onset of puberty. The elegant rationale to consider
LMNA as a candidate gene for this disease was that
there is an analogy between the highly specific anatomic site
involvement in AD-EDMD and FPLD.
What Mechanisms Link the Nuclear Envelope
to Cardiac and Skeletal Muscle Diseases and to
Lipodystrophies?
Thus, rare mutations in LMNA, a
ubiquitously expressed gene that encodes nuclear structural proteins,
cause inherited disorders of cardiac and skeletal muscles and of
adipose tissues. The mutations are distributed all along the gene,
without distinct regions being specific to a disease
(Figure 2B). Functional knockout of the mouse gene
that encodes lamin A/C induces a postnatal cardiac and skeletal
muscular
dystrophy.55 The
mice also lack adipose tissue. The lamins have been implicated in
mediating DNA replication, chromatin organization, spatial arrangement
of nuclear pore complexes, nuclear growth, and anchorage of
nuclear-envelope proteins. Because both muscle cells and fat cells
derive from the mesenchymal stem cell, a general model was recently
proposed by
Wilson.56
LMNA mutations would selectively affect the
differentiation, maintenance, repair, or regulation of cells in the
mesenchymal stem cell. Another theory is that forces generated during
skeletal or cardiac muscle contraction or the specific metabolic
activity of the adipose tissue might render these tissues especially
sensitive to nucleus damage produced by the mutation. Finally, the
mutations could alter the interactions between lamins A/C and putative
tissue-specific partners.
ARVD, a Model to Understand the Right Ventricular Response to Injury and Its Associated Right Ventricular Failure and Sudden Death
Increased Awareness of ARVD as a Cause for
Sudden Death
Arrhythmogenic right ventricular dysplasia (ARVD) is
relatively new as a diagnostic entity and was not included in the WHO
classification of cardiomyopathies until 1996. Reports since the days
of Osler in 190557
of patients with partial replacement of the right ventricular
myocardium by fat or fibrous tissue were probably Uhls anomaly rather
than ARVD. In 1978, Frank et
al58 referred to the
new entity as RV dysplasia. In 1982, Fontaine et
al59 added the term
arrhythmogenic because arrhythmias and sudden death seem to be major
clinical manifestations. ARVD is a familial cardiomyopathy of unknown
pathogenesis characterized by a gradual loss of myocytes and
replacement by fatty and fibrous tissue, which, as it progresses, leads
to dilatation of the ventricle and impaired function. The clinical
course is characterized by arrhythmias, sudden death, and heart
failure. The prevalence of the disease in Italy is
1:5000,60 and it
accounts for 22.4% of deaths in athletes. In Olmsted County,
Minnesota, histological features in keeping with ARVD were observed in
9 of 54 sudden death victims 23 to 40 years
old.61 Thus, in this
study, ARVD accounted for 17% of sudden deaths in the young. Despite
the diagnostic difficulties, ARVD is now established as a major cause
of sudden death in the young, and the rate of sudden death is 2.5% per
year, frequently without prior symptoms.
What Are the Implications to Be Gleaned From
Studying ARVD?
Although ARVD, like idiopathic or familial DCM, is
associated with a dilated cardiac chamber, it has distinguishing
features with broad biological, physiological, and pathological
implications relating directly to the cardiac response to injury and
the subsequent development of cardiac failure. Whereas familial or
idiopathic DCM is manifested in the left ventricle, with involvement of
the right ventricle being secondary and occurring at a later stage,
ARVD is a mirror image, with the disease initiating exclusively in the
right ventricle, and may only much later involve the left ventricle. In
familial HCM, all of the responsible genes identified so far encode
sarcomeric proteins, which are equally distributed throughout the right
and left ventricles. Thus, restriction of the phenotype to the left
ventricle in FHCM is not because of expression of a chamber-specific
gene but rather from an interaction with the environment, such as the
increased pressure load, which exceeds by several-fold that of the
right ventricle. The 2 genes identified as being responsible for
familial DCM, actin and desmin, are also present throughout the heart,
yet the initial phenotype is in the left ventricle. These observations
would suggest that the restriction of the ARVD phenotype to the right
ventricle is more likely due to a unique stimulus of the right
ventricle rather than a chamber-specific gene. The other intriguing
aspect of ARVD is the possibility that myocardial cells die and are
replaced by fatty-fibrous tissue due to
apoptosis.60 62
Confirmation of the involvement of the apoptosis system would
immediately provide a new framework to direct the development of
specifically targeted therapies.
Although the genetic basis for FHCM has rapidly evolved (9 genes) and that for familial DCM is beginning to emerge (6 loci and 4 genes), a gene responsible for the autosomal dominant form of ARVD is yet to be discovered. Although no gene has yet been identified, 3 loci, 14q23,63 1q42,64 and 2q32,65 have been mapped in Italian families. Recently, 2 loci, 3p2366 and 10p12-p14,67 have been mapped in North American families. Identifying the responsible genes has significantly improved the diagnosis of FHCM and is likely to improve the diagnosis of familial DCM even more. A family with a recessive form of ARVD was identified on the Greek island of Naxos.68 The recessive form of the disease appears to be quite distinct from the dominant form, being associated with palmoplantar keratoderma and woolly hair. The gene responsible for this disease resides on chromosome 17q21 and most recently was shown to encode for plakoglobin.68 Plakoglobin is one of the proteins involved with cell-to-cell adhesions and plays a major role in maintaining myocyte integrity. Whether this will provide a clue to the genes responsible for other loci remains to be determined.
Familial Atrial Fibrillation: A Cornerstone of Our Understanding of Atrial Conduction
The identification of several genes for ventricular tachycardia of the long-QT syndrome and Brugada syndrome69 is in the process of enhancing our understanding of the normal physiology of ventricular conduction and the pathophysiology of arrhythmias. It is likely that genetic defects exist in a host of genes involved with ventricular conduction and arrhythmias. We know very little about atrial conduction and supraventricular arrhythmias, including atrial fibrillation (AF), paroxysmal tachycardia, and Wolff-Parkinson-White syndrome. AF is the most common form of arrhythmia affecting humans; it is associated with extensive morbidity and mortality and accounts for >33% of all strokes in patients >65 years old.
The first chromosomal locus responsible for AF was mapped to chromosome 10q21 in 1997.70 The region containing the locus has since been narrowed from 27 to 0.6 cM, and thus, the gene is expected to be identified in the near future. It is hoped that isolation of the first gene on 10q21 will identify a pathway from which other candidate genes can be derived analogous to that of the sarcomeric genes of FHCM. Once the genes and the mutations have been identified, the collaboration of the cardiologist will again be necessary to further define the precise mechanisms and provide the infrastructure for prevention and treatment.
Acknowledgments
This work was supported in part by
grants from the National Heart, Lung, and Blood Institute, Specialized
Centers of Research (P50-HL-42261-01), INSERM, and the Association
Française contre les Myopathies (6710 and 6491). We thank R.M. Barton
and H. Worman for the diagram shown in Figure 2A
. We greatly appreciate
the administrative assistance of Moira Long and Debbie Graustein in the
preparation of the
manuscript.
References
-cardiac myosin heavy chain gene
mutation impairs contraction and relaxation function of cardiac
myocytes. Am J Physiol. 1999;276:H1780H1787.
/B-crystallin chaperone gene causes a desmin-related myopathy.
Nat Genet. 1998;20:9295.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |