Pour isa activites nantes.pdf

Nantes Center Research project
Cardiac diseases and sudden death
The cardiac diseases and sudden death group gathers 5 different themes
THEME A: Clinical and molecular genetics (JJ Schott) THEME B: Functional genomics and transcriptional control of ion channel remodeling (S Demolombe) THEME C: Molecular and cellular physiology (I Baró) THEME D: Experimental arrhythmias and transgenesis (F Charpentier) THEME E: ß-adrenoceptors (C. Gauthier)

Abstract
Sudden cardiac death (SCD) claims almost a million deaths annually in the USA and Europe,
ventricular fibrillation being the most common underlying arrhythmia. For many years, the institut has
been a key player in the field of cardiac arrhythmias. By elucidating the mechanisms responsible for
rare monogenic diseases, our team has produced several breakthroughs, which have considerably
improved our understanding of the molecular basis of normal and abnormal cardiac electrical activity.
The efficiency of our research resulted from synergistic studies conducted in parallel in molecular and
clinical genetics (theme Ia), molecular and cellular physiology (theme Ic), transgenic animals (theme
Id) and genomics of ion channel genes (theme Ib). The institut is one of the very few places worldwide
where all these abilities are gathered in the same laboratory.
A large number of our current projects deal with rare diseases with a special focus on late onset
diseases such as progressive conduction defects and atrial fibrillation. Other projects also aim to
address genetic susceptibility to SCD in the general population. Our hypothesis is that the risk for
SCD is genetically modulated by key pathways and becomes manifest in the face of environmental
triggers such as myocardial ischemia, drugs, or heart failure. Another major project is to elucidate the
mechanisms controlling ion channel transcription based on our previous studies showing that
regulations of ion channel expression play an important role in providing plasticity in response to
genetic, pathological, or pharmacological insults. Studies conducted on transgenic animal or cellular
models also identify important new candidate genes in the context of arrhythmia or conduction
defect. Finally, because ß-adrenoceptors play a key role in modulating cardiac function, we are
evaluating their contribution in diseases that are known to alter ionic channel expression and
predispose to cardiac arrhythmias such as heart failure (theme Ie)
For many years, the institut has been a key player in the field of cardiac arrhythmias. Researchers
and clinicians at the institut have discovered important new concepts by elucidating the mechanisms
responsible for rare monogenic diseases. Indeed, one key development driving our understanding of
the molecular basis of normal and abnormal cardiac electrical behavior has been the identification of
single mutations that greatly heighten the risk for arrhythmias and sudden cardiac death. In this field,
the institut has produced several breakthroughs. This was due not only to the efficiency of its
molecular and clinical genetics group (Prof. Hervé Le Marec, Drs Vincent Probst and Jean-Jacques
Schott) but also because of the synergistic studies conducted in parallel on the mechanisms of cardiac
arrhythmias, based on molecular and cellular physiology (Drs. Isabelle Baró and Gildas Loussouarn),
cell biology (Dr. Jean Mérot), transgenic animals (Dr. Flavien Charpentier), and the genomics of ion
channel genes (Drs. Sophie Demolombe and Françoise Le Bouffant). The institut is one of the very
few places worldwide where all these abilities are gathered in the same laboratory.
This line of research has been central for our success in obtaining the label of 'Centre Thématique de
Recherche et de Soin'. An additional move has been the constitution of a transatlantic network of
excellence sponsored at the level of 6 million dollars by the Leducq Foundation. The network
(Alliance Against Sudden Cardiac Death) coordinated by Prof. D. Escande, includes the department
of Pharmacology at Vanderbilt University (Prof. D. Roden), the department of Cardiology at Miami
University (Prof. R. Myerburg), the department of Cardiology at Johns Hopkins University (Prof. E.
Marbán) and the department of Cardiology at the University of Amsterdam (Prof. A. Wilde). At the
national level, the institut is the only recognized Expert Center for Hereditary Cardiac Arrhythmias.
Sudden cardiac death (SCD) claims almost a million deaths annually in the USA and Western Europe (40,000 in France), and ventricular fibrillation is the most common underlying arrhythmia. Because of the magnitude of the problem, the opportunities for a real public health impact are enormous: even minor improvements (10-20%) in risk detection could reduce the number of victims by 100,000 or more. Device-based approaches have had a remarkable impact on SCD in patients at high risk. However, the fundamental public health problem of SCD is that the majority of victims do not fall into these high-risk groups. Our central hypothesis is that the risk of fatal arrhythmias is modulated by genetically determined variants in key pathways and that this risk becomes manifest in the face of environmental triggers such as myocardial ischemia, drugs, or heart failure. At the institut, our research has focused for many years on rare genetic model diseases (cardiac
channelopathies). Still, a large number of our current projects deals with these rare genetic
syndromes because of the unique opportunity we have of exploring the population of western France,
which is an isolated population almost like Iceland with large pedigrees extending over centuries. More
recently our projects are moving to address genetic susceptibility to SCD in the general
population.
Cardiac channelopathies comprise a group of rare genetic diseases primarily affecting normal
cardiac electrophysiology. This ensemble includes the long QT syndrome (LQTS), the Brugada
syndrome (BS), the inherited Lenègre disease (also called Lev’s disease), the inherited sick sinus
syndrome, catecholaminergic polymorphic ventricular tachycardia and familial atrial fibrillation.
The long QT syndrome results from prolonged repolarization of the ventricular action potential. This
anomaly is associated with a high risk of ventricular arrhythmias known as torsade-de-pointe
arrhythmias. Prolonged cardiac repolarization can result from decreased outward (mainly K+) currents
or increased inward (Na+ or Ca2+) currents. In 1995, the group of Mark Keating1 identified the first gene
responsible for the long QT syndrome. Since then, at least 6 different genes have been identified as
causative for LQTS including the KCNQ1 gene (encoding for the voltage-dependent K+ channel,
KCNQ1 also called KvLQT1) responsible for LQT12, KCNH2 (encoding the human ether-à-gogo
related K+ channel, HERG) responsible for LQT21, SCN5A (encoding the major cardiac Na+ channel)
responsible for LQT33, ANK2, encoding ankyrin B, responsible for LQT44, KCNE1 encoding for the K+
channel subunit KCNE1 (also called MinK or IsK) responsible for LQT55 and the KCNJ2 gene
(encoding the inward rectifier K+ channel Kir2.1) responsible for the Andersen syndrome6, which
associates K+-sensitive periodic paralysis, ventricular arrhythmias, and dysmorphic features.
Our laboratory in conjunction with other groups in the USA has been instrumental in discovering the
role of ankyrin B in LQTS. Ankyrins are a family of adaptor proteins that localize structurally-diverse
ion channels and transporters to specialized membrane domains in various tissues. We have linked
ankyrin B (AnkB) dysfunction with fatal cardiac arrhythmias in humans. The ANK2 A4274G mutation
leading to E1425G causes the “type 4” form of the congenital long QT syndrome. Affected family
members display not only QT interval prolongation, stress- and/or exercise-induced polymorphic
ventricular arrhythmia, syncope, and sudden cardiac death but also sinus node dysfunction
(bradycardia or junctional escape rhythm), and episodes of atrial fibrillation. Subsequent evaluation of
additional probands heterozygous for AnkB E1425G and identification and characterization of the
phenotypes of probands with other AnkB loss-of-function variants demonstrated that the clinical
phenotypes often extend beyond the typical long QT syndrome, leading to the label ‘ankyrin-B-
syndrome’: these phenotypes include bradycardia, sinus arrhythmia, delayed conduction/conduction
block, idiopathic ventricular fibrillation, and catecholaminergic polymorphic ventricular tachycardia.
The Brugada syndrome is a new familial syndrome described in 1992 by the Brugada brothers. The
syndrome associates a typical ECG pattern (ST segment elevation in the right precordial leads
associated with an aspect of atypical right bundle branch block) and a risk of malignant ventricular
arrhythmias such as polymorphic ventricular tachycardia or ventricular fibrillation responsible for
syncope and sudden death. The diagnosis of the disease is often difficult due to potential confounding
factor and the intermittent expression of the ECG abnormalities. In 1998, Chen et al. linked the
syndrome with the α-subunit of the cardiac Na+ channel (Nav1.5) encoded by SCN5A7. Nearly seventy
SCN5A mutations have been identified in patients over the past six years. Most are private mutations
that induce a loss of function of the Na+ channel through various mechanisms. However, in the
Brugada syndrome only 20% of patients are carriers of a SCN5A mutation suggesting other culprit
genes. In 2002, London and co-workers reported in a large family a new locus, BS2, close to SCN5A
on chromosome 3 (3p22-25). In 2006, mutation in the GPD1L gene has been associated with BS2.
1 Curran ME, Splawski I, Timothy KW, et al. Cell 1995; 80:795-803. 2 Wang Q, Curran ME, Splawski I, et al. Nat Genet 1996;12:17-23. 3 Wang Q, Shen J, Splawski I, et al. Cell 1995;80:805-11. 4 Mohler PJ, Schott JJ, Gramolini AO, et al. Nature 2003;421:634-9. 5 Duggal P, Vesely MR, Wattanasirichaigoon D, et al. Circulation 1998;97:142-6. 6 Plaster NM, Tawil R, Tristani-Firouzi M, et al. Cell 2001;105:511-9. 7 Chen Q, Kirsch GE, Zhang D, et al. Nature 1998;392:293-6. Progressive cardiac conduction defect (PCCD) is characterized by an age-related alteration in the
conduction of the cardiac impulse through the His-Purkinje system. PCCD, also called Lenègre or
Lev disease, is a frequent affection that remains the predominant cause for pacemaker implantation.
It is commonly thought that the disease issues from a slow process developing over decades, affecting
the His bundle and its branches. Lenègre and Lev hypothesized that it was a primary degenerative
disease or an exaggerated aging process of unknown origin with sclerosis affecting selectively the
conducting tissue. As early as the 70’s, familial investigation of patients with chronic bifascicular block
or complete AV block suggested that genetic factors could be involved in the pathophysiology of the
disease. Hereditary cases of bundle branch block have repetitively been reported. We reported a large
French family in which cardiac conduction defect was transmitted as an autosomal dominant trait and
was caused by a splice mutation in the SCN5A gene. SCN5A is the first identified gene responsible for
hereditary Lenègre disease8.
Other cardiac channelopathies include catecholaminergic polymorphic ventricular tachycardia related with mutations in the ryanodine receptor gene (RYR2)9 and in the calsequestrin-2 gene (CASQ2)10. A KCNQ1 mutation has been identified most recently in Chinese kindred with familial atrial fibrillation11. Heart failure is a classical situation where the risk of sudden cardiac death is increased. The failing
human heart is characterized by sustained activation of the sympathetic nervous system. This
increase in adrenergic drive is first beneficial but ultimately detrimental causing important myocardial
damages. The increase in sympathetic stimulation is responsible for down-regulation of myocardial β1-
adrenoceptors in association with alterations in β2-adrenoceptor coupling with its effector system
whereas β3-adrenoceptors are up-regulated resulting in altered myocardial function. Concurrently,
heart failure is associated with vascular β-adrenergic dysfunction, which has been less investigated.
Until recently, the use of β-blockers was disregarded in the treatment of patients with congestive heart
failure. The value of β-blockers is now firmly established even as a first-line treatment of congestive
heart failure. The effects of chronic treatment with β-blockers are variable especially in heart failure of
ischemic etiology and in the elderly. The cause for this variable clinical response is unknown.
The cardiac diseases and sudden death group gathers 5 different themes:
A. Clinical and molecular genetics (Drs. JJ. Schott and V. Probst and Prof. H. Le Marec)
B. Functional genomics and transcriptional control of ion channel remodeling (Drs. S.
Demolombe and F. le Bouffant)
C. Molecular and cellular physiology (Drs. I. Baró, G. Loussouarn and J. Mérot)
D. Experimental arrhythmia and transgenesis (Dr. F. Charpentier)
E. β-adrenoceptors (Dr. C. Gauthier)
THEME A: Clinical and molecular genetics (JJ Schott)
Previous activities
In the past ten years, we have identified new genes responsible for life threatening cardiovascular
pathologies using familial-based genetic approaches. We have mainly focused our research on
cardiac arrhythmias and more recently on late onset diseases such as progressive cardiac
conduction defects and atrial fibrillation but also on valvular diseases such as mitral valve prolapse
and calcific aortic stenosis using an epidemiogenetic approach. Our basic hypothesis is that there is,
for most of these age-related pathologies, a genetic predisposition related to a major gene.
Dominant diseases are - by definition - caused by single gene variants. It is common knowledge among clinicians and human geneticists that these variants show the widest possible spectrum of disease severity ranging in principle from unaffected to severely impaired or dead carriers of the same mutation. Consequently major gene variants are a necessary but not a sufficient condition to develop the respective disease. Other causes for disease manifestation remain to be identified. These may be endogenous (modifier genes) or exogenous (environment, lifetime events, lifestyle choices, diet, etc.). The existence of modulating genes controlling the penetrance and the expressivity in each pathology constitutes another research orientation12. We have identified new pathophysiological mechanisms for three diseases. Elucidation of the
molecular bases for these pathologies resulted from a close cooperation between clinicians,
genealogists, geneticists and physiologists at the institut.
8 Schott JJ, Alshinawi C, Kyndt F, et al. Nat Genet 1999;23:20-1. 9 Priori SG, Napolitano C, Tiso N, et al. Circulation 2000;102:r49-53. 10 Lahat H, Pras E, Olender T, et al. Am J Hum Genet 2001;69:1378-84. 11 Chen YH, Xu SJ, Bendahhou S, et al. Science 2003, 299:251-4. 12 Kaab S, Schulze-Bahr E. Cardiovasc Res 2005, 67(3):397-413. (Review). 1- Identify a mutation in the ANK2 gene leading to LQT413. AnkB E1425G mutation induces not only QT interval prolongation, stress- and/or exercise-induced polymorphic ventricular arrhythmia, syncope, and sudden cardiac death but also sinus node dysfunction, and episodes of atrial fibrillation. Ankyrins are a family of adaptor proteins that localize ion channels and transporters to specialized membrane domains. This work constitutes the first demonstration that a cytoskeleton protein can lead to cardiac arrhythmias. 2- We identified a splice mutation (IVS.22+2 T→C) in the SCN5A gene encoding for the cardiac Na+ channel Nav1.5 in a large family diagnosed with Lenègre disease.14 Functional analysis showed that haploinsufficiency was the disease causing mechanism15. SCN5A is the first identified gene responsible for hereditary Lenègre disease. Screening a second large family identified a second loss of function Nav1.5 G1408R mutation in patients affected by Lenègre disease or Brugada syndrome. Interestingly, the two phenotypes were transmitted separately between different family branches and Brugada phenotype was predominantly diagnosed in males suggesting that gender and/or other modifiers genes affecting the expressivity of Nav1.5 channels added to the complexity of SCN5A channelopathies16. 3- We also identified the first gene responsible for non-syndromic polyvalvular myxomatous dystrophy in a large family showing X-linked mode of inheritance17. Four distinct Filamin A mutations were identified18 inducing similar polyvalvular dystrophies. Filamin A gene encodes a ubiquitously expressed cytoskeleton protein with various signaling properties and structural functions. Although the precise disease causing mechanism still needs to be elucidated, this finding opens new insights in the genetics of valvular dystrophies where 3 loci for monogenic mitral valve prolapse have been identified.19,20,21 4- We demonstrated a familial aggregation for calcific familial aortic stenosis (Monckeberg disease) in the western part of France, suggesting a monogenic origin in this family22. The identification of these new molecular bases of common cardiac pathologies has shed new insight into pathophysiological mechanisms and urged us to reconsider our knowledge in the field of cardiac arrhythmias as well as valvular dystrophies. Current Project
Our objectives are to pursue ongoing projects focused on rare familial cardiac arrhythmias but also to identify susceptibility genes for common cardiac diseases. Strategies developed to achieve these goals are essentially based on whole genome approaches
such as (i) reverse genetics for monogenic diseases; (ii) epidemiogenetic strategies for late onset
disease; (iii) association studies for polygenic pathologies. Alternatively, candidate gene
approaches are used when transgenic animal model show convincing similarities with human cardiac
arrhythmias pathologies. These approaches require high throughput genotyping and sequencing
technologies, which are available at our Génopôle OUEST platform.
Identifying genes responsible for late onset diseases will become crucial given the aging population.
We plan to extend our strategy to identify late onset monogenic diseases using an
epidemiogenetic approach. Families affected by late onset diseases are by definition difficult to
identify. We plan to take advantage of the regional specificity of the population where families are
generally large with limited geographical mobility creating series of isolates. We propose that the
existence of geographical aggregates for a given disease reflects the existence of founder effects. A
similar strategy applied to the Icelandic population representing a typical geographical isolate has
shown power in human genetics23,24.
In a later perspective, we plan to focus on identifying susceptibility genes for sudden cardiac
death.
A- Monogenic diseases gene identification: new culprit genes for ventricular and atrial
fibrillation
Brugada syndrome remains a true genetic challenge. The molecular data show that the implication of 13 Mohler PJ*, Schott JJ*, Gramolini AO, et al. Nature 2003, 421(6923):634-9. *Equivalent authors 14 Schott JJ, Alshinawi C, Kyndt F, et al. Nat Genet 1999, 23(1):20-1. 15 Probst V, Kyndt F, Potet F, et al. J Am Coll Cardiol 2003, 41(4):643-52 16 Kyndt F, Probst V, Potet F, et al. Circulation 2001, 104(25):3081-6. 17 Trochu JN, Kyndt F, Schott JJ, et al. J Am Coll Cardiol 2000, 35(7):1890-7 18 Kyndt F, Gueffet JP, Legendre A, et al. Circulation 2006, in press 19 Disse S, Abergel E, Berrebi A, et al. Am J Hum Genet 1999, 65(5):1242-51. 20 Freed LA, Acierno JS Jr, Dai D, et al. Am J Hum Genet 2003, 72(6):1551-9. 21 Nesta F, Leyne M, Yosefy C, et al. Circulation 2005, 112(13):2022-30. 22 Probst V, Le Scouarnec S, Legendre A, et al. Circulation 2006, 113(6):856-60. 23 Sveinbjornsdottir S, Hicks AA, Jonsson T, et al. New Engl J Med 2000, 343:1765-1770. 24 Jonsson S, Thorsteinsdottir U, Gudbjartsson DF, et al. JAMA 2004, 292:2977-2983. SCN5A gene is as low as 30% with a penetrance estimated at 11%25. The difficulty to identify large informative families suggests that the penetrance of Brugada syndrome for the other as yet unidentified genes, is probably also very low. We have recruited and characterized new families, large enough for genome wide scan analysis. We have also recruited smaller families, originating from the same geographical area. A large genealogical study showed that these families shared a common ancestor and thus represented a single large pedigree comprising numerous affected patients. This pedigree will also be investigated using a whole-genome SNP approach. Identification of novel genes involved in the Brugada syndrome will be a significant step toward a better understanding of the pathophysiology of this complex disease. As for Brugada syndrome, familial AF remains a challenge for geneticists. Despite the identification of 4 loci26, only KCNQ127 and GJA528 mutants have been identified. Using a classical family based approach, we have identified a new locus for early onset AF on chromosome 20. A common characteristic of the 8 AF patients was the occurrence of low P wave amplitude associated with mild dilation of the atrium. The identification of the disease-causing gene is currently in progress focusing on a 23 mega base interval. In parallel, based on phenotypical evidence showing AF in a Cav1.3 Knock-In mice model, we are currently screening the human CACNA1d gene in a cohort of patients diagnosed with AF or sinus node dysfunction. Preliminary evidences have been obtained for potential disease-causing mutations. Functional analyses are ongoing. B- Epidemiogenetic approaches for late onset pathologies
One of the major finding of this preliminary epidemiogenetic work was the identification of a large heterogeneity (from parish to parish) in the prevalence of several cardiovascular diseases. Since our pioneer work, we have developed new approaches based on heterogeneous geographic distribution of pacemaker implantations for Lenègre disease. Several clusters with high implantation rates were identified. In a first approach, we searched for shared family names in areas with the highest implantation rate. We then screened for other family members affected by Lenègre disease. In one identified family, we mapped a new locus on chromosome 16. Genetic analyses for other families are also ongoing. Similar approaches are used to map genes responsible for calcific aortic stenosis, mitral valve prolapse. In both conditions surgery remains the only efficient medical treatment. C- Sudden Cardiac death in the community: susceptibility genes
Most often, sudden cardiac death (SCD) results from ventricular tachyarrhythmia mainly ventricular
fibrillation complicating an unrecognized coronary heart disease. Most cases occur in asymptomatic
individuals. Currently identified risk factors do not allow stratification of individual risk in the general
population. An important challenge in molecular cardiology is to identify high-risk individual who
could beneficiate from preventive therapy. Genetics could help to improve risk stratification in the
general population and, thus, to establish preventive strategies guided by the individual risk. This
research could have a tremendous impact owing to the numerical importance of SCD.
Our genetic hypothesis is based on three lines of evidence: (1) rare variants in genes encoding ion
channels or regulators are associated with a high risk of sudden death; (2) gene-environment
interactions can lead to a sudden death; and (3) family history of sudden death is an independent risk
factor for sudden death. Rare genetic mutations identified to date only account for a very small part of
SCD in the general population. Most culprit genetic variants remain to be identified.
Our project includes two complementary sections: • to create a registry of out-of hospital SCD in the adult population of Loire-Atlantique, (1 134 000 inhabitants). Incidental cases will be exhaustively reported (estimate: around 600 cases/year). • To establish a DNA bank with blood samples collected by the SAMU at the time of the cardiac arrest (estimate: 250 samples/year). This DNA collection will be used for a case-control genome-wide association study. A control group will be constituted with sudden-death-free subjects matched to cases in age, sex and ethnicity. All subjects will be genotyped using a high-resolution map of SNPs distributed throughout the genome. Positive SNPs will be confirmed in additional population samples obtained through our transatlantic collaborative network. Our ultimate goal is to identify genetic variants associated with an increased risk of SCD and to develop a high-throughput low-cost screening test applicable to the general population for prevention guidance. 25 Viswanathan PC, Balser JR. Trends Cardiovasc Med 2004, 14(1):28-35. Review. 26 Wiesfeld AC, Hemels ME, Van Tintelen JP, et al. Cardiovasc Res 2005, 67(3):414-8. Review. 27 Chen YH, Xu SJ, Bendahhou S, et al. Science 2003, 299: 251-54. 28 Gollob MH, Jones DL, Krahn AD, et al. N Engl J Med 2006, 354(25):2677-88. THEME B: Functional genomics and transcriptional control of ion channel remodeling (S
Demolombe)
Previous activities
Regulations of ion channel expression play an important role in maintaining a stable electrophysiological phenotype in cardiac myocytes as well as in providing plasticity in response to genetic, pathological, or pharmacological insults. Channel expression can be modified at several levels along the biosynthetic pathway, i.e. transcription, translation, trafficking and degradation. We hypothesized that the transcription plays a key role in the electrical plasticity of the heart because of its high specificity and accuracy. We validated this hypothesis by: 1/ evaluating ion channel expression at a genomic scale in response to changing physiological or pathological environment; 2/ by demonstrating that transcriptional modifications induce functional consequences. A- Physiological genomics of ion channel genes
We investigated the regional distribution of ion channel expression in the mouse and non-diseased human hearts.29 Expression of ion channels is heterogeneous within the heart muscle and this heterogeneity is key to specialization of cardiac areas. In a mouse study, we demonstrated that the remarkable signature of the nodal tissues results from increased expression of a large panel of ion channel genes rather than decreased expression. In a human study, we showed that atria, Purkinje fibers and ventricular tissues show specific patterns, whereas inter-tissue differences between epicardium and endocardium, and left and right-sided cavities are subtle. Our data point to significant differences in regionally-determined ion-channel expression, with important potential implications for understanding regional electrophysiology, arrhythmia mechanisms, and responses to ion-channel blocking drugs. We also verified that the expression profile of ion channel genes in the normal mouse heart is dependent on the genetic background,30 and may account for the previously reported differences in ECG characteristics and susceptibility to programmed electrical stimulation-induced ventricular arrhythmias.31 The molecular mechanisms underlying gender-related differences in arrhythmic risks are poorly understood.32 We investigated at a genome-scale the ion-channel expression in non-diseased epicardium and endocardium samples from either sex.33 We demonstrated that transcript expression differences between samples from men and women are subtle but involve key genes implicated in conduction and repolarization. The lower expression in women of several K+-channel α-and β-subunits provides a potential molecular basis for female lower repolarization reserve and poor penetrance of the Brugada syndrome. B- Pharmacological genomics of ion channel genes
We also investigated the remodeling of ion channels as a possible mechanism of drugs effects. In this context, we thought that amiodarone might be prototypic. Amiodarone has a remarkable antiarrhythmic efficacy and the pharmacological profile of the drug is complex including direct modulation of ion channel function. This drug is also known to modify thyroid function mostly due to its iodinated nature.34 The question arose as to whether the long-term effects of amiodarone might stem from its molecular interaction with thyroid hormone receptors or others mechanisms including modulation of gene expression in addition to its direct effect on channels.35 The effects of the drug on cardiac ion channel transcripts of treated adult mice are compatible with the pharmacological profile of the drug, which associates decreased conduction and prolonged repolarization, and in nature comparable to that induced by thyroid hormone as we previously analyzed.36,37 Most importantly, remodeling of ion channel expression induced by amiodarone is dose-dependent and correlates with the dose-dependent effects of the drug on the ECG. C- Clinical genomics of ion channel genes
Our group has investigated the remodeling induced by chronic atrial fibrillation (AF) in patients undergoing open-heart surgery.38 Limitations of previous studies of ion channel expression changes in patients with AF include the arbitrary selection of subunits for study and inadequate disease-matched controls. We compared a group of patients with chronic AF and underlying valvular heart disease with 29 Marionneau C, Couette B, Liu J et al. J Physiol 2005;562:223-34. 30 Demolombe S, Marionneau C, Le Bouter S, et al. Cardiovasc Res 2005;67:438-47. 31 Maguire CT, Wakimoto H, Patel VV, et al. Physiol Genomics 2003;15:84-91. 32 James AF, Choisy SC, Hancox JC. Prog Biophys Mol Biol 2005 (in press). 33 Gaborit N, Varro A, Le Bouter S, et al. 2006 (submitted). 34 Kennedy RL, Griffiths H, Gray TA. Clin Chem 1989;35:1882-87. 35 Drvota V, Blange I, Haggblad J, et al. J Cardiovasc Pharmacol 1998;32:654-61. 36 Le Bouter S, El Harchi A, Marionneau C, et al. Circulation 2004;110:3028-35. 37 Le Bouter S, Demolombe S, Chambellan A, et al. Circ Res 2003;92:234-42. 38 Gaborit N, Steenman M, Lamirault G, et al. Circulation 2005;112:471-81. a group of patients with valvular heart disease in sinus rhythm. We conclude that the transcriptional profile of genes involved in ion transport is profoundly affected by valvular heart disease. However, AF is associated with a specific pattern of expression, but whether these are caused by AF or characterize valvular heart disease patients more likely to develop AF is still unknown. A simplistic view of cardiac channelopathies would be that a genetic defect selectively alters the function of a key ion channel leading to abnormal electrogenesis and arrhythmias. However, cardiac cells adapt to the defect by remodeling gene expression. We therefore hypothesized that Brugada syndrome could remodel ion channel gene expression in the myocardium as a result of compensatory molecular mechanisms. Novel methods to accurately investigate the expression of hundreds of transcripts in very small size biological samples have recently been developed in our institut.29 We used these methods to obtain full profiling of ion channel expression in right ventricular septal endomyocardial biopsies from patients with Brugada syndrome.39 The ion channel signature demonstrates a complete discrimination between endomyocardial tissues from patients with Brugada syndrome and those from two groups of control subjects (donor hearts and recently transplanted hearts). We observed that patients with Brugada syndrome share a common molecular signature irrespective of the culprit gene. We are now envisaging a larger biopsy collection from patients receiving an implantable cardiac defibrillator for various indications. Results from this new investigation should conclude whether molecular profiling could reliably diagnose ventricular arrhythmias. Current project
Our projects are dedicated to elucidate the mechanisms controlling ion channel transcription. This project includes two aspects: 1/ to identify the transcription factors (TFs) involved in ion channel transcription; 2/ to identify the feedback pathways linking the electrophysiological phenotype of the cardiomyocytes and the transcription machinery to control phenotype alteration. As a first step, our investigations will focus on the transcriptional regulation of the cardiac voltage-gated Na+ channel (Nav1.5) and the Kv channel-interacting protein 2 (KChIP2) involved in the transient outward potassium current Ito. Both INa and Ito currents play key role in the physiology and pathophysiology of the heart.40,41 Both are also linked to the pathophysiology of the Brugada syndrome.42 The role of transcription factors has been extensively studied in cardiac embryogenesis.43 In comparison, much less is known about their role in the adult heart. Our knowledge of transcriptional regulation of ion channel gene expression is even sparser.44 This is certainly an open and important avenue for new research. Aim #1: to correlate ion channel profiling with transcription factor profiling in different
conditions. To investigate the complex transcriptional regulation of ion channel genes, global
transcription can be measured using high throughput real-time RT-PCR, and computational analysis of
this information can lead to a deeper understanding of the system as a whole. These ‘system
approaches' have been successfully used by Gilchrist et al. in the context of immune regulation.45 We
have already mapped 164 TF RNA species in different regions of the adult mouse and non-diseased
human hearts. We will also investigate the expression dynamics of TFs in relation to progressive
remodeling of cardiac ion channel transcripts as induced by hypertrophy, using a mouse model
developed at the institut by Flavien Charpentier. A collection of human samples from hypertrophic
hearts is also available in our laboratory. However, unlike in mice, dynamic information with
progressing hypertrophy cannot be obtained in human.
Aim #2: to correlate gene profiling with abnormal intracellular Ca2+ handling. Ca2+ ions modulate
many cellular processes including gene expression through transcription factors, such as members of
the CREB, NFAT, and MEF2 families;46,47 it is assumed that Ca2+ channels represent the most
important feedback link between electrical activity and regulation of ion channel gene expression. For
example, it has been recently shown that [Ca2+]i controls the functional expression of Ito in cardiac
myocytes.48 Our goal is to correlate dynamic changes in ion channel expression and transcription
regulators with intracellular Ca2+ anomalies in the mouse model of cardiac hypertrophy.
Aim #3: bioinformatics. We ambition to identify in silico the promoter region for every gene encoding
ion channel subunits and Ca2+ regulators expressed in the heart. Promoter analysis will provide: 1/ a
39 Wichter T, Gaborit N, Varro A, et al. Circulation 2006 (in revision). 40 Balser JR. J Mol Cell Cardiol 2001;33:599-613. 41 Oudit GY, Kassiri Z, Sah R, et al. J Mol Cell Cardiol 2001;33:851-72. 42 Di Diego JM, Cordeiro JM, Goodrow RJ, et al. Circulation 2002;106:2004-11. 43 Bruneau BG. Circ Res 2002;90:509-19. 44 Rosati B, McKinnon D. Circ Res 2004;94:874-83. 45 Gilchrist M, Thorsson V, Li B, et al. Nature 2006;441:173-8. 46 Dolmetsch R. Sci STKE 2003, P E4 47 Wilkins BJ, Molkentin JD. Biochem Biophys Res Commun 2004;322:1178-91. 48 Rossow CF, Dilly KW, Santana LF. Circ Res. 2006, 98:1306-13. number of putative transcription factor binding sites located in predicted promoter regions, 2/ shared organization of co-regulated ion channels. This information will be correlated with our biological data. Aim #4: ChIP on chips approach. Our ChIP on chips approach will be conducted in collaboration
with Dr. Remi Houlgatte.
Aim #5: functional correlate. Depending on the outcome of the preceding aims, different mouse
models invalidated for key ion channels and transcription factor genes will be generated or obtained
through collaborative agreements.
THEME C: Molecular and cellular physiology (I Baró)
Our projects are mainly focused on the gene-function relationship of K+ and Na+ channels, which are
key elements of the cardiac action potential. These ion channels are now rather regarded as
macromolecular complexes made of different proteins having various functions: channel per se,
chaperone, regulatory or anchoring protein. However, many molecular mechanisms underlying
channel behavior, and many of the actors of the channel-complexes are still unknown. The genotype-
phenotype relationships in hereditary channelopathies often remain difficult to establish.
Our objectives are 1/ to describe the structural events leading to channel activity and 2/ to identify the channel-complexes actors that may turn out to be new therapeutic targets for cardiac diseases handling. Previous activities
A- Molecular pathophysiology of the KCNQ1-related channelopathies
The pathophysiological sequence that leads to the congenital LQTS sometimes appears as a simple correlation between the mutation and the patient phenotype. In the past, we dedicated ourselves to understand the molecular physiology of the KCNQ1 protein49, and to determine the functional consequences of mutations in its gene in the context of LQTS50,51. Indeed, KCNQ1 gene encodes for a channel protein, which associates with the membrane protein, KCNE1, to constitute a channel-complex responsible for the repolarizing K+ current (IKs) of the myocardium52,53. In LQTS, mutations in the KCNQ1 gene are the most frequent. In previous studies, we and others have shown that mutations leading to a loss of function of the channel are associated with the disease. We have also shown that a KCNQ1 mutation may be associated with another cardiac arrhythmia: the short QT syndrome54. In a patient resuscitated after a sudden death and presenting a QT interval shortening, we detected a mutation inducing a gain of function of the K+ channel. Our most recent discoveries were focused on: 1) Molecular mechanisms for loss-of-function
Mutations may alter the K+ current in different ways, from no expression, impaired trafficking to altered
channel activity in the sarcolemma. Most recently, studying LQT1-associated KCNQ1 mutations, we
unveiled 1/ a new role of the N-terminus of KCNQ1 channel in its trafficking and its implication in
severe forms of LQTS55, 2/ we identified PIP2 and MgATP as regulators of KCNQ1 and showed that
at least three KCNQ1 mutations are linked to a decrease in channel-PIP2 affinity56,57.
2) KCNQ1 channel-complex
Even if, most of the time, the genotype-phenotype relationship may be direct, its remains that the ratio of mutation carriers presenting the disease over the total number of carriers in a family, i.e. its penetrance, may be highly variable. Our hypothesis was that other partners of the complex may contribute to the channel dysfunction and thus explain part of the variable penetrance. Therefore, we initiated a systematic analysis of potential components of KCNQ1 channel-complex. The study of the KCNQ1/KCNE1 complex in a recombinant system identified A kinase anchor proteins
(AKAP) as key component in the adrenergic regulatory transduction pathway for KCNQ1 in
cardiomyocytes58. Very recently, in collaboration with Hugues Abriel (Lausanne, Switzerland) we also
showed that Nedd4/Nedd4-like proteins associate to KCNQ1 and regulate stability and
internalization of KCNQ1 through ubiquitylation processes59.
49 Demolombe S, Baró I, Péréon Y, et al. J Biol Chem 1998, 273:6837-6843 50 Mohammad-Panah R, Demolombe S, Neyroud N, et al. Am J Hum Genet 1999, 64:1015-1023 51 Gouas L, Bellocq C, Berthet M, et al. Cardiovasc Res 2004, 63:60-68 52 Barhanin J, Lesage F, Guillemare E, et al. Nature 1996, 384:78-80 53 Sanguinetti MC, Curran, ME, Spector PS, et al. Nature 1996, 384:80-83 54 Bellocq C, van Ginneken AC, Bezzina CR, et al. Circulation. 2004, 109:2394-2397 55 Dahimène S, Alcoléa S, Escande D, et al. Circ Res in revision 56 Loussouarn G, Park KH, Bellocq C, et al. EMBO J 2003, 22:5412-5421 57 Park KH, Piron J, Dahimène S, et al. Circ Res 2005, 96:730-739 58 Potet F, Scott JD, Mohammad-Panah R, et al. Am J Physiol 2001, 280:H2038-H2045 59 Jespersen T, Membrez M, Nicolas C, et al. submitted B- Molecular pathophysiology of the SCN5A-related channelopathies
The SCN5A gene coding for the cardiac Na+ channel Nav1.5, is also implicated in channelopathies. The functional study of these SCN5A mutations explained the phenotypical impact. Nav1.5 channel generates the Na+ current responsible for the fast depolarization of the action potential (AP). The loss of function of the mutated Na+ channel induces a heterogeneous shortening of the ventricular AP that contributes to ventricular fibrillation associated to BS60,61. In collaboration with the team of Jean-Jacques Schott at the institut we unveiled similar loss of function of Na+ channel for SCN5A mutations in two families presenting with Lenègre disease62,63. However, in Lenègre disease, the loss of function of the Na+ channel induces a conduction defect, a very different phenotype compared to BS. By analogy with KCNQ1 and KCNE1, we suspected that the phenotype of the SCN5A mutation
carriers may be determined by polymorphism of Nav1.5 partners. In a search for new Nav1.5 partners,
we have recently identified 14-3-3 protein as a component of Nav1.5 complex. We showed that 14-3-3
regulates Na+ current availability during the AP64.
The identification of new ion-channel protein partners and characterization of their mechanisms of
interaction at the molecular level may have major implications: 1/ understanding the channel
physiology and the pathophysiology of the congenital cardiac arrhythmias and conduction diseases; 2/
identification of new candidate genes leading to detection of asymptomatic carriers with elevated
risk of sudden death; 3/ ion channel partners may also represent potential targets for new
pharmacological agents.
Current project
In a continuum with our work, our current projects concern the mechanisms implicated in the regulation of the cardiac ion channels KCNQ1 and Nav1.5 in the context of cardiac arrhythmia and conduction defect. A- Partners of channel-complexes
Previously, we used a 'candidate partner' approach. More recently, we have launched a more systematic screening method. 1) KCNQ1 and β-tubulin
Françoise Le Bouffant detected different proteins interacting with Nav1.564 or KCNQ1, using a yeast
two-hybrid screen of a mouse cDNA library. Among them, β-tubulin interacts with the N-terminus part
of KCNQ1. After validation in mammalian expression systems and in cardiomyocytes of the binding
between the full-length channel and the candidate protein by co-immunoprecipitation, we colocalized
both proteins in transfected cells and cardiomyocytes at the level of the intercalated discs and
sarcolemma. We are currently evaluating the regulation of KCNQ1-KCNE1 complex by microtubules in
both models. Our preliminary results show a regulation of KCNQ1-KCNE1 sensitivity to PKA-
dependent stimulation by the microtubules. A second candidate protein interacting with KCNQ1 C-
terminus is also currently under investigation.
2) Proteomics approaches
In order to further detect potential partners of KCNQ1 or Nav1.5, we are also using conventional (pull-down, immuno-precipitation) and new (Blue native/SDS-PAGE adapted for membrane proteins65) proteomics approaches coupled to mass spectrometry analysis (MS). Protein sequencing is performed in the Laboratory of Proteomics of the RIO Platform (INRA) in Nantes who has established the appropriated procedures66. Once identified, channel-partners interactions will be analyzed in expression systems (co-immunoprecipitation and co-localization) and in cardiomyocytes. The following steps will be to investigate the role of the newly identified binding proteins on channel maturation (folding and trafficking) and function (channel activity) in cell lines and in adult cardiomyocytes transfected with cDNA and/or siRNA. B- Molecular physiology of KCNQ1
We are convinced that a better knowledge of the intimate molecular mechanisms of the channel
maturation and activity is required.
1) Folding, maturation and trafficking of KCNQ1
Our recent works identified impaired trafficking as the cellular mechanism involved in severe forms of Romano Ward LQT1 syndrome55. Our ongoing analysis further highlighted endoplasmic reticulum (ER) retention and protein quality control as key stages in KCNQ1 mis-trafficking. These data point to 60 Dumaine R, Towbin JA, Brugada P, et al. Circ Res 1999, 85:803-809 61 Wan X, Chen S, Sadeghpour A, et al. Am. J. Physiol 2001, 280:H354-H360 62 Schott JJ, Alshinawi C, Kyndt F, et al. Nat Genet 1999, 23:20-21 63 Probst V, Kyndt F, Potet F, et al. J Am Coll Cardiol 2003, 41:643-652 64 Allouis M, Le Bouffant F, Wilders R, et al. Circ Res 2006, 98:1538-1546 65 Brouillard F, Bensalem N, Hinzpeter A, et al. Mol Cell Proteomics 2005, 4:1762-1775 66 Tamaskovic R, Bichsel SJ, Rogniaux H, et al. J Biol Chem 2003, 278:6710-6718 early steps in channel biogenesis as targeted by the mutations. Our objectives are to go one step
further and to analyze at the molecular level 1/ the molecular chaperons that participate in the channel
biogenesis, 2/ the impact of the mutations on the interactions with these chaperons and the fate of the
protein in the ER (retro-translocation - degradation), 3/ the impact of the mutations on the topogenesis
(insertion and orientation of the transmembrane domains) of the channel.
These new aspects of KCNQ1 maturation and trafficking will be analyzed with different experimental approaches already mastered in the laboratory including immunoprecipitation, western blotting, metabolic radiolabeling with 35S-methionin and in vitro translation reaction. 2) Molecular regulations of KCNQ1 activity
KCNQ1 intrinsic regulation. The mechanisms for voltage-dependent channel activation still need
clarification. In this context, we hypothesized that a 6 transmembrane domains (TMDs) Kv channel
may be both structurally and functionally equivalent to a voltage sensor, comprising the four initial
TMDs (S1-S4), merged to a two-TMDs K+ channel comprising TMDs S5-S6 and the pore. One
consequence of such a model would be that the S4-S5 linker behaves like a voltage-dependent
ligand able to bind to the C-terminal part of S6 (CterS6). Preliminary data show that
overexpression of a peptide corresponding to the S4-S5 linker indeed decreases the open state
stability of KCNQ1 whereas a peptide corresponding to the CterS6 increases the open state stability.
These data confirm our initial hypothesis and suggest that complementation of abnormal voltage
dependence by specific peptides could be used in vivo to correct mutant channels.
KCNQ1 osmoregulation and PIP2. During ischemia and reperfusion, cardiac cells have been shown
to undergo significant swelling, which has marked effects on cardiac electrophysiology by modulating
the function of several sarcolemmal ions channels and transporters67. In particular, it has been shown
that KCNQ1 is sensitive to osmolarity, both in presence and absence of KCNE168. We have recently
shown that several KCNQ1 mutants associated with LQTS1 are characterized by a decrease in PIP2
sensitivity57. One of them is almost insensitive to a decrease in PIP2. This mutant is also poorly
sensitive to change in osmolarity, suggesting a role of PIP2 in the osmoregulation of the channel
activity. Using several mutations with various PIP2 affinities, we are currently characterizing the role of
PIP2 in KCNQ1 osmoregulation.
THEME D: Experimental arrhythmias and transgenesis (F Charpentier)
Previous activities
A- Pathophysiology of cardiac conduction defects
In the past years, we characterized the phenotype of a mouse model invalidated, at the heterozygous
state, for Scn5a gene.69,70 We demonstrated that the Scn5a+/- mouse is a good model for Lenègre
disease with similar age-related deterioration in conduction defects. The age effects are due to
progressive occurrence of extensive ventricular fibrosis and remodeling of connexin expression. Our
study provides the first univocal demonstration that a monogenic ion channel defect can lead
with age-related myocardial structural anomalies. It also supports Lenègre's original hypothesis
that progressive conduction defects were due to a slow fibrotic process. Finally, it offers an opportunity
to develop a preventive therapy of fibrosis and progressive conduction defects.
In a subsequent study, we demonstrated that: (i) the expressivity of the conduction anomalies is
variable among Scn5a+/- mice, as also observed in patients with inherited Lenègre disease; (ii) mice
with a strong phenotype exhibit more myocardial rearrangements with aging than mice with a mild
phenotype; (iii) old mice with a strong but not with a mild phenotype have a reduced conduction
reserve and exhibit spontaneous ventricular arrhythmias as also observed in the Brugada syndrome;
(iv) the expressivity of the conduction deficit correlates with the ability of the normal Scn5a allele to
produce larger amount of functional channel proteins. Our data suggest that variable expressivity of
the disease in Scn5a+/- mice results from variable compensatory mechanisms governing the
expression of wild-type Na+ channel proteins. Scn5a+/- mice can be used to investigate for the
mechanisms of variable expressivity of SCN5A loss-of-function mutations in man.
Finally, Scn5a+/- mice present a moderate bradycardia due to abnormal conduction from the sinus
node to surrounding atrial myocytes and sino-atrial blocks.71 Our results provide new insights into
the understanding of the pathophysiology of SCN5A-linked sinus node dysfunction.72
B- Development of a biological pacemaker
67 Vandenberg,J.I., Rees,S.A., Wright,A.R. et al. Cardiovasc Res. 1996, 32:85-97 68 Kubota T, Horie M, Takano M, et al. Jpn J Physiol 2002, 52:31-39. 69 Royer A, van Veen TA, Le Bouter S, et al. Circulation 2005, 111:1738-46. 70 van Veen TA, Stein M, Royer A, et al. Circulation 2005, 112:1927-35. 71 Lei M, Goddard C, Liu J, et al. J Physiol 2005, 567:387-400. 72 Benson DW, Wang DW, Dyment M, et al. J Clin Invest 2003, 112:1019-28. In France, about 180,000 patients are equipped with an electronic cardiac pacemaker and the
implantation rate is continuously increasing with population aging. Although electronic pacemakers are
an efficient medical therapy, they do have limitations including lack of responsiveness to autonomic
neurohumoral regulation, lead- and procedure-related complications and the need for periodic battery
replacement that led to the attempt of developing biological pacemakers by in situ gene transfer.
All studies performed so far used viruses as gene vectors. As an alternative gene delivery system, we
used the tetronic 304, a poloxamine block copolymer developed by Bruno Pitard's team,73 to transfect
locally ventricular myocytes in situ with genes encoding the hyperpolarization-activated pacemaker
channel HCN2 and the ß2-adrenergic receptor. With this method, we generated long-lasting
biological pacemakers that are regulated by ß-adrenergic stimulation and that improve life
expectancy in mice in chronic atrioventricular block.
Current project
A- Pathophysiology of cardiac conduction defects
Characterization of the Scn5a+/- mouse model will continue. Our objectives are now: 1- to identify the mechanisms for variable expressivity of the disease; 2- to identify the gene transcription regulatory pathway(s) responsible for increased fibrosis and 3- to investigate the mechanisms of cardiac arrhythmias; 4- to evaluate the therapeutic potential of drugs that prevent myocardial fibrosis in the context of Variable expressivity of the conduction defects between mice with a mild phenotype and mice with a strong phenotype can be explained either by the presence of modifier genes (polymorphisms), an hypothesis that will be tested with Affymetrix GeneChips® microarrays (with Jean-Jacques Schott, theme IA), or by differential adaptive gene expression remodeling. Differentially-expressed genes between the two groups of mice will be identified at the mRNA level using Applied Biosystems Mouse Genome Survey Microarrays® in collaboration with Marja Steenman (team III). Finally, the mechanisms for higher expression of Nav1.5 protein in mice with a mild phenotype will be investigated with Jean Mérot (theme IC) and Hughes Abriel (Lausanne, Switzerland). The gene transcription regulatory pathway that links moderately impaired ventricular activation in young mice with increased fibrosis with aging will be identified with Sophie Demolombe (theme IB) using high-throughput quantitative RT-PCR (MicroFluidic Cards, Applied Biosystems). In this context, we will determine if the identified pathway is triggered by the conduction slowing per se or by the impaired pattern of activation as previously observed in a canine model of ventricular pacing.74 Other channelopathies, such as the Brugada syndrome, have been linked to loss-of-function mutations in SCN5A.75 The Brugada syndrome is usually attributed to a disequilibrium between a large transient K+ current in the right ventricular subepicardium and a decreased Na+ current.76 But altered right ventricular conduction could also lead to the disease.77 Indeed, the frontier between the Brugada syndrome and isolated cardiac conduction disease remains unclear and in the same family the same mutation can induce both diseases.78 We have shown that Scn5a+/- mice exhibit an earlier and larger conduction delay in the right ventricle and that mice with a severe phenotype have a 50% incidence of ventricular tachycardia. We thus propose to explore further the conduction in the right ventricle by recording epicardial monophasic and/or transmembrane action potentials. Finally, our previous results provide an experimental background to support evaluation of the therapeutic potential of drugs that prevent myocardial fibrosis in the context of Lenègre disease. Many studies have shown that angiotensin converting enzyme (ACE) inhibitors, angiotensin receptors antagonists and aldosterone blockers can prevent or even reverse fibrosis and ventricular remodeling in different cardiac pathologies. We are currently designing a blinded study in the Scn5a+/- mouse model aimed at evaluating the beneficial effects of long lasting administration of ACE inhibitors on conduction parameters, ventricular remodeling and fibrosis. Our project should result in the identification of prognostic markers of the progression of the
disease in patients with Lenègre disease or Brugada syndrome and of new therapeutic targets. If
positive, our study with ACE inhibitors will be followed by a clinical study performed in the Cardiology
Department of the institut du thorax in collaboration with the "Centre d'Investigation Clinique Inserm" of
Nantes. It could provide the first preventive therapy for the Lenègre disease.
B- Development of a biological pacemaker
73 Pitard B, Bello-Roufai M, Lambert O, et al. Nucleic Acids Res 2004, 32:e159. 74 van Oosterhout MF, Prinzen FW, Arts T, et al. Circulation 1998, 98:588-95. 75 Chen Q, Kirsch GE, Zhang D, et al. Nature 1998, 392:293-6. 76 Antzelevitch C, Brugada P, Brugada J, et al. J Am. Coll. Cardiol 2003, 41:1665-71. 77 Meregalli PG, Wilde AA, Tan HL. Cardiovasc Res 2005, 67:367-78. 78 Kyndt F, Probst V, Potet F, et al. Circulation 2001, 104:3081-6. After proving the feasibility of generating biological pacemakers with non-viral gene transfer, we now plan i) to improve our pacemakers and ii) to validate in dogs the strategies identified in mice. Further molecular developments to increase the firing rate and the longevity of our engineered pacemakers include the evaluation of different promoters and investigation of the efficacy of mutated HCN2 channels in the mouse model. Longer-term follow-up (> 1 year) of the pacemaker activity is also needed once the most adequate promoter and transgenes will be chosen. Because the mouse cardiac electrical activity differs markedly from the human one, engineered pacemakers need further evaluation in larger mammals such as dogs. Whether or not they will lead to the same firing rate in dogs as in mice will be addressed by these studies. Dogs will also be useful for developing tools and procedures for proper intra-cardiac targeting of gene delivery in man. We will also ensure that the engineered pacemaker can express itself in the atrium, an issue that cannot be solved in the mouse. This part of the project will be performed in collaboration with Marc Vos (Utrecht University Medical Center, NL) who developed a model of dog in chronic atrioventricular block. The expected results of this project are to develop a biological alternative to electronic
pacemakers. We hope that a clinical trial will be ultimately performed in the institut du thorax. Finally,
this project opens new perspectives for the gene therapy of other cardiac diseases.
C- Pathophysiology of atrial fibrillation
In the atrium and sinus node, ICa,L is carried by Cav1.2 (a1C) and Cav1.3 (a1D) subunits. Because previous studies had shown that Cav1.3 knockout mice (Cav1.3-/-) exhibit sinus node dysfunction,79,80 we further characterized them in vivo and demonstrated that they also present spontaneous episodes of atrial fibrillation (AF). The phenotype of Cav1.3-/- mice is thus reminiscent to the brady-tachycardia syndrome ("maladie rythmique de l’oreillette"), which associates sick sinus syndrome, episodes of AF and atrioventricular conduction disturbances. We will use the Cav1.3-/- mouse model to elucidate the pathophysiology of this syndrome. More precisely, we will investigate whether susceptibility to AF is the result of chronic sinus node dysfunction, secondary to the hypertrophy produced by chronic bradycardia or directly related the loss of Cav1.3-related ICa,L component in atrial myocytes. In parallel, we will characterize a transgenic mouse model expressing, in a cardiac specific manner, the S140G-mutated KCNQ1 channel that is responsible for a familial form of idiopathic AF.81 Our preliminary results show that these mice exhibit spontaneous atrial extrasystoles and are more prone to develop AF under programmed electrical stimulation. Our objective is to identify the mechanisms linking the gain of function of the mutated KCNQ1 channel and the occurrence of AF. These studies should provide new insights in understanding the pathophysiology of AF, which not only
is the most common arrhythmia but also represents a challenge for the clinician who lacks
efficient therapies. Our results on the Cav1.3-/- mouse prompted Jean-Jacques Schott (theme IA) to
screen for mutations in the Cav1.3 gene in patients affected by sick sinus syndrome or lone AF. This
ongoing study has already yielded mutants demonstrating the pertinence of the mouse model.
THEME E: ß-adrenoceptors (C. Gauthier)
Previous activities
Our team has characterized the expression and function of ß3-adrenoceptors (ß3-ARs) in the cardiovascular system and their putative involvement in cardiovascular diseases such as heart failure (HF) and hypertension. In human endomyocardial biopsies, we have demonstrated that ß3-AR stimulation, in marked contrast with that of ß1- and ß2-ARs, induces a decrease in contractility through activation of Gi/0 proteins82 and stimulation of the nitric oxide (NO) pathway leading to an intracellular cGMP increase83. Similar results have been obtained in dog. However, ß3-ARs are not endogenously expressed in hearts of small animals as rats or mice. We thought to explore whether rabbits could represent a relevant alternative to dogs. In rabbit cardiomyocytes, ß3-AR stimulation induces negative inotropic and positive lusitropic effects associated with inhibition of L-type Ca2+ current and activation of the delayed rectifier K+ current (IKs) through Gi/o-NOS pathway. These results suggest that ß3-ARs could be relevant target for the treatment of arrhythmias by decreasing Ca2+ overload and preventing prolongation of repolarization. At end stages of human HF, ß3-ARs are upregulated although the negative inotropic effect induced by a ß3-AR agonist is blunted84. At earlier stages, the ß3-AR upregulation may be a compensatory mechanism preventing further cardiomyocyte damage. Indeed, in transgenic mice overexpressing 79 Zhang Z, He Y, Tuteja D, Xu D, et al. Circ Res 2002, 90:981-7. 80 Mangoni ME, Couette B, Bourinet E, et al. Proc Natl Acad Sci USA 2003, 100:5543-8. 81 Chen YH, Xu SJ, Bendahhou S, et al. Science 2003, 299:251-4. 82 Gauthier C, Tavernier G, Charpentier F, et al. J Clin Invest 1996, 98, 556-562 83 Gauthier C., Leblais V., Kobzick L, et al. J Clin Invest 1998, 102:1377-1384. 84 Moniotte S., Kobzik L., Feron O, et al. Circulation 2001, 103(12):1649-1655. human ß3-AR in cardiomyocytes85, no histological evidence of myocyte hypertrophy or fibrogenesis is observed, suggesting that ß3-AR activation is not responsible for cardiac damages during HF. In rat thoracic aorta, the stimulation of endothelial ß3-ARs produces slowly developing relaxation86,87 leading to NO pathway activation and subsequent increase in intracellular cGMP levels. Several K+ channels: BKCa, KATP and Kv, are activated. We obtained similar results in human internal mammary arteries88. Our work highlights the role of β3-AR in vasomotor control of internal mammary arteries and opens new fields of investigation in coronary bypass graft management. In hypertension, there are only few data regarding the potential role of vascular ß3-ARs. Using a canine peri-nephritic hypertension model, we suggested that ß3-AR stimulation might be beneficial by normalizing the blood pressure89. In spontaneously hypertensive rats, we have described a ß3-AR overexpression, although this was not associated with an increased ß3-AR-dependent vasorelaxation90. The relevance of ß-blockers in the treatment of HF is now firmly established even as a first-line treatment of congestive HF. Whether vasodilating properties of ß-blockers may contribute to the clinical effects is debatable. A recent clinical trial demonstrated that nebivolol is an effective and well-tolerated treatment for HF in the elderly91. However, its target is not clearly identified. We have recently shown that in rat aorta, nebivolol-induced relaxation results from activation of ß3-ARs. In addition, we have confirmed the involvement of endothelium and NO pathway in the vascular response92. We have also described a ß3-AR activation by nebivolol in human biopsies. Current projects
Identification of ß3-ARs in human heart has changed the classically admitted paradigm on the regulation of heart function by the ß-adrenergic system. ß3-AR activation induces negative inotropic effects, associated with acceleration of relaxation and decreased intracellular calcium levels93. Regarding the role of ß3-AR on the heart rate, only a few studies have been performed with no formal conclusion. ß3-ARs stimulation may offer protection against excessive catecholaminergic ß1-AR stimulation. It is well established that ß1-ARs are down-regulated in HF and several studies reported an upregulation of ß3-AR. Current clinical uses of ß-blockers include treatment of primary and secondary prevention of myocardial infarction in patients with coronary artery disease, atrial and ventricular arrhythmias, arterial hypertension, and, most recently, HF. However, ß-blockers are still considered underused in current clinical practice. In addition, elevated heart rate is a well-established independent predictor of coronary artery disease and cardiovascular morbidity and mortality. One possible mechanism of cardioprotection elicited by ß-blockers is bradycardia. Thus, inversely to ß1- or ß2-ARs, ß3-ARs inhibit cardiac contractility. We hypothesized that ß3-ARs may have a buffer or ‘rescue’ function in front of excess catecholamines; such as occur in hyperadrenergic states including heart failure. Until now, only dogs were suitable as a model to analyze cardiac ß3-ARs. Our preliminary results indicate that rabbit could be an appropriate alternative, which reproduces the contractile effects obtained in human heart. A- ß3-adrenoceptors in the heart
1) Function
At present, the precise and specific physiological role of cardiac ß3-ARs is not fully understood, leading to conflicting results in in vitro and in vivo studies. We will evaluate the consequences of ß3-AR stimulation in comparison with ß1- and /or ß2-AR stimulation in rabbit isolated cardiomyocytes (video-imaging), papillary muscles, whole heart (Langendorff) and in vivo (pressure-volume loops). Our preliminary results obtained in isolated rabbit cardiomyocytes show that ß3-AR stimulation produces negative inotropic and positive lusitropic effects associated with a reduction in ICa,L. as a result from Gi/0-NO pathway activation. Surprisingly, these effects are not seen in the whole heart. Additional experiments are needed to explain such discrepancy. It is admitted that ß-AR stimulation increases the heart rate. However, the ß-AR subtype involved in this effect is not yet fully identified. In clinical practice, administration of dobutamine, a ß1-AR agonist, is arrythmogenic. Salbutamol, a ß2-AR agonist, increases heart rate but also prolongs the QT interval. One study suggests that the ß3-AR stimulation produces opposite effects on the heart rate depending 85 Tavernier G, Toumaniantz G, Erfanian M, et al. Cardiovasc Res 2003, 59: 288-296. 86 Trochu J.N., Leblais V., Rautureau Y, et al. Br J Pharmacol 1999, 128: 69-76. 87 Rautureau Y., Toumaniantz G., Serpillon S, et al. Br J Pharmacol 2002, 137(2), 153-161. 88 Rozec B., Serpillon S., Toumaniantz G, et al. J Am Coll Cardiol 2005, 46: 351-359. 89 Donckier J.E., Massart P.E., Van Mechelen H, et al. Eur J Clin Invest 2001, 31(8):681-689. 90 Mallem Y., Toumaniantz G, Serpillon S, et al. Br J Pharmacol 2004, 143: 599-605. 91 Flather, M.D., Shibata, M.C., Coats, A.J, et al. Eur Heart J 2005, 26: 215-225. 92 Rozec B., Tran Quang T., Noireaud J, et al. Br J Pharmacol 2006, 147: 699-706. 93 Rozec B., Gauthier C. Pharmacol Therap 2006,111: 652-673. on the activated signaling pathway94. We will compare the effects of ß1, ß2 and ß3-AR stimulation on rabbit cardiac electrical activity (K+ currents and action potential) and the respective activated signaling pathway will be determined. This project will be carried out in association with F. Charpentier. 2) Localization
To complete the functional studies, we plan to evaluate ß3-AR expression and cellular localization in the different cardiac chambers as well as in the conductive tissue of the rabbit heart. We have successfully tested molecular and biochemical tools used in rats. In addition, in human atria and ventricles, ß3-ARs have been identified, but their stimulation does not lead to the same contractile effect. Indeed, in atria, ß3-AR stimulation does not influence contractility95. Thus, the roles of ß3-ARs in atria remain to be clarified. 3) Signaling pathways
We have demonstrated that ß3-ARs are linked to Gi/0-NO-cGMP pathway in the human ventricle. The cAMP pathway seems to be involved in the regulation of heart rate. Thus, it is necessary to evaluate cAMP and cGMP respective levels and downstream pathways after ß3-AR stimulation on heart rate and contractility. It has been shown that cardiac ß1 and ß2-AR stimulations elicit distinct subsarcolemmal cAMP signals, which are regulated by different phosphodiesterases (PDEs). In collaboration with R. Fischmeister (Inserm U769), we will determine the PDE isoforms involved in ß3-AR effects in rabbit cardiomyocytes and their putative cross-talk with the signaling pathways of ß1 and ß2-ARs. The ß3-AR effects will also be evaluated on the Ca2+ cycle (collaboration with D. Potreau CNRS UMR 6187, Poitiers). B- ß-adrenoceptors subtypes and heart failure
In failing heart, the increased adrenergic tone and the impaired global ß-adrenergic response are well known, but the relative contribution of each ß-AR subtype and their cross-talks have received little or no attention. Different data obtained both in normal and failing hearts support the hypothesis that upregulation and/or stimulation of ß3-AR-mediated pathway may prevent myocardial damage under catecholamine excess, particularly in early stages of the disease. Several animal models could be used to investigate the consequences of cardiac ischemia. In the rabbit, we will evaluate the expression of each ß-AR subtypes after an infarct (collaboration with B. Ghaleh, Inserm U660, Créteil). Another suitable model is the ischemic rat model of HF. This model will permit investigations of either “basal” remodeling of ß-adrenergic system or after treatment with different ß-blockers. We will compare the expression level and localization of each cardiac ß-AR subtypes as well as the electrophysiological and contractile responses at different stages of HF in ischemic rats. As ß3-ARs are overexpressed at end stage of HF and ß3-AR are overexpressed both in endothelial cells and cardiomyocytes, we have developed a transgenic rat model (Tgß3) overexpressing human ß3-AR in endothelial cells. Preliminary results show that endothelial ß3-AR overexpression is associated with ß1-AR decrease in cardiomyocytes whereas ß2-ARs are unchanged. In addition, the isoproterenol-induced positive inotropy is also altered. These results demonstrate for the first time that endothelial ß3-ARs regulate cardiac contractility and ß-AR expression in cardiomyocytes. It remains to determine the cellular mechanisms involved in the cardiac remodeling. Concurrently, and in collaboration with Pr J.L. Balligand (Brussels, Belgium), we will use replication-deficient adenovirus encoding human ß3-AR to evaluate on cardiac expression remodeling and electrical and contractile activities, the effects of ß3-AR overexpression by delivering the adenovirus either into the myocardium bordering the ischemic area, in the whole heart, or in the coronary system. The pro-angiogenic effect of ß3-AR overexpression will be evaluated. Finally, we will determine whether ß3-AR agonist or antagonist properties of different ß-blockers in accurate therapeutic windows could improve outcome, which underlying mechanisms are implied and the putative cross-talks with other pathways involved. We will compare the effects of several ß-blockers on electrophysiological and contractile responses. This project should lead to determine the different ß-AR targets of ß-blockers in order to optimize their use for a given patient. 94 Sterin-Borda L, Bernabeo G, Ganzinelli S, et al. J Mol Cell Cardiol 2006, 40: 580-588. 95 Pott, C., Brixius, K., Bundkirchen, A, et al. Br J Pharmacol 2003, 138 : 521-529.

Source: http://www.allianceagainstscd.org/docs/pdf/Nantes%20Research%20Highlights.pdf