This study was designed to test the hypothesis that induction of the preconditioned state results in a sustained translocation of protein kinase C (PKC) which accounts for the memory associated with preconditioning. Isolated rabbit cardiomyocytes were subjected to established preconditioning protocols using either adenosine or transient ischemia. At timed intervals during induction of preconditioning (PC), post-incubation or final sustained ischemia, cells were harvested, subjected to digitonin lysis and separated into cytosolic and particulate fractions. Samples were evaluated by Western blot analysis with monoclonal antibodies to alpha, epsilon, zeta and gamma PKC isozymes, and bands were quantified by densitometry. Internal controls for each experiment included oxygenated cardiomyocytes and cells with PKC translocation evoked by treatment with phorbol 12-myristate 13-acetate (PMA). For control oxygenated cells, the particulate fraction contained about 30% of PKC epsilon, 5-10% of PKC alpha and 60-70% of PKC zeta. Preconditioning with adenosine (100 μM) or 10 min ischemia had no significant effect on these percentages. Furthermore, the relative amounts of the PKC isozymes associated with the particulate fraction of control and preconditioned cells did not differ after a post-incubation in oxygenated buffer or during a final ischemic incubation. PMA and ingenol completely translocated the epsilon and alpha isoforms, while thymeleatoxin totally translocated PKC alpha but only partially (50%) translocated PKC epsilon. The distribution of PKC zeta between fractions was not affected by any drug, The protein phosphatase inhibitor calyculin A protected cells mimicking preconditioning. This protection was blocked by preincubation with the selective PKC inhibitor calphostin C but was largely retained if calphostin C was added only during the final ischemic period. It is concluded that PKC activity is required for preconditioning, but a sustained translocation of PKC above basal levels is not necessary for protection of rabbit cardiomyocytes in vitro.

Cardiovascular disease (CVD) is the leading cause of death globally, and tissue ischemia induces disorders including myocardial infarction, pulmonary arterial hypertension, and atherosclerosis. Ischemic conditions considerably alter cellular homeostasis as well as metabolism and can result in cardiovascular dysfunction. Therefore, there is a dire need to develop a novel treatment strategy for curing myocardial ischemia via manipulating cellular networking and metabolism. Furthermore, studies have shown that ischemic conditions induce cellular stress such that it modifies the expression of various cellular proteins and may even promote cell death. Cellular proteins must fold into a three-dimensional, native state to become functional. To ensure efficient folding and prevent aggregation, molecular chaperones assist with the folding/refolding process. Additionally, chaperones regulate cellular proteostasis through the ubiquitin-proteosome system (UPS) or autophagy. The heat-shock protein (HSP) is a molecular chaperone family upregulated during stress conditions to assist cells with proper protein folding, stability, and turnover. Specifically, the 70-kDa heat-shock protein (Hsp70) has numerous cytoprotective and immunomodulatory effects through its interaction with components of several cellular pathways. This interaction is dependent on nucleotide exchange factors (NEFs) to hold the open conformation of HSP70's nucleotide-binding domain (NBD). Proteins in the Bcl-2- associated athanogene (BAG) family are a group of co-chaperones that interact with the ATPase domain of HSP70 and help maintain homeostasis. The BAG family proteins (BAG1-6) share an evolutionarily conserved region at their C-termini (the BAG domain), and BAG5 specifically is unique in that it contains five of these domains. BAG5 plays a crucial role in maintaining cellular homeostasis and viability. This project explores the role of BAG5 as a co-chaperone and potential therapeutic tool to improve cardiomyocyte function under ischemic conditions.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23752号 / 医博第4798号 / 新制||医||1055(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 山下 潤, 教授 木村 剛, 教授 井上 治久 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Mitochondria provide the main energy required for cardiac excitation-contraction coupling via aerobic oxidative phosphorylation (OXPHOS) process. Accumulation of reactive oxygen species (ROS), by-products of mitochondrial respiration, within dysfunctional mitochondria results in the activation of cardiac cell death pathways and has been associated with heart failure development. Therefore, maintaining mitochondrial homeostasis as a balance between mitochondrial biogenesis and degradation is of great importance toward cardiac proper functioning. In addition to the importance of mitochondrial energy supply, gap junctions, intercellular channels which connect plasma membrane of adjacent cardiomyocytes, by propagating action potential throughout the myocardium maintain cardiac synchronous beating and rhythm. Gap junctions have a rapid turnover and impair of gap junction quality control impacts cell-to-cell communication; resulting in electrical conduction abnormalities and arrhythmogenesis. Therefore, understanding the underlying mechanism the quality control of mitochondria and gap junctions profoundly contributes toward understating the genesis of cardiomyopathy. Furthermore, cardiovascular problems in HIV (Human immunodeficiency virus) positive patients whose viral load is controlled via antiretroviral therapy remains a problem while the underlying mechanism remains elusive. The current study has used an in vitro model of primary neonatal rat ventricular cardiomyocytes (NRVCs) to discover the molecular mechanisms of mitochondrial as well as gap junction quality control under normal and stress conditions. Furthermore, electrical activities of the primary cardiomyocytes were recorded using microelectrode array (MEA) system and important electrophysiological components such as impulse propagation pattern and conduction velocity were extracted from the complex signal recordings. Overall, we have pursued four main aims; Aim 1. Dysregulation of mitochondrial quality control machinery leads to cardiac death; Aim 2. HIV-1 Tat (transcriptional transactivator) dysregulates cardiac homeostasis via mitochondrial pathway; Aim 3. Impairment of protein quality control impacts the quality of gap junction; Aim 4. Inhibition of gap junction quality dysregulates electrical signal propagation within the culture. / Bioengineering

Cardiomyocyte proliferation plays a crucial role in the developing mammal heart, as it is required for normal morphogenesis and in determining the appropriate heart size. Postnatally, the decrease in cell cycle activity is concomitant with the increase of cell cycle variants, such as endoreduplication and acytokinetic mitosis, which contribute to the hypertrophic growth of the heart. Although the adult mammal heart retains the ability to generate new cardiomyocytes, the extent of cardiomyocyte renewal is insufficient to compensate for the large-scale tissue loss associated with ischemic events. Indeed, ischemic events such as myocardial infarction, lead to a permanent loss of ventricular cardiomyocytes, formation of collagen-containing scar, and consequently cardiac remodeling. The development of therapies able to hamper cardiac remodeling by promoting cardiomyocyte turnover is one of the primary goals in the cardiovascular field. In the present study, we generated a human induced pluripotent stem cell (iPSC) line containing the fluorescence ubiquitination-based cell cycle indicator (FUCCI) under the Troponin T2 (TNNT2) promoter. To gain information about the cell cycle dynamics of human cardiomyocytes, we visualized cell cycle progression in TNNT2-FUCCI human iPSCs-derived cardiomyocytes. Notably, we revealed cardiomyocytes' cell cycle dynamics of cells undergoing proliferation, binucleation, and polyploidization and identified G2 cell cycle arrest in cardiomyocytes undergoing polyploidization. To demonstrate the versatility of the TNNT2-FUCCI human iPSCs line, we developed a live cell screening platform to identify pro-proliferative compounds within an autophagy compound library. We identified Clonidine, an alpha2-adrenergic receptor and imidazoline agonist, as an enhancer of cell cycle activity in TNNT2-FUCCI hiPSC-derived cardiomyocytes. Finally, we investigated the ability of Clonidine to promote cell cycle progression in hiPSC- derived cardiomyocytes and in in vivo and in vitro mouse neonatal cardiomyocytes. We showed that while Clonidine stimulated cardiomyocytes' polyploidization and multinucleation, respectively in in vitro in in vivo mouse cardiomyocytes, the treatment of hiPSC-derived cardiomyocytes with Clonidine enhanced their proliferative capability. In conclusion, we showed that the TNNT2-FUCCI system is a versatile tool for characterizing cardiomyocyte cell cycle dynamics and identifying pro-proliferative molecular candidates with regenerative potential in the mammalian heart.:1. Introduction 1.1 Heart function and composition 1.2 Human cardiac development 1.2.1 Cardiac organogenesis 1.2.2 Metabolic changes in the developing heart 1.3 Cardiomyocytes cell cycle activity 1.3.1 Cardiomyocytes cell cycle regulators and cell cycle arrest 1.3.2 CM multinucleation and polyploidization 1.4. The regenerative capabilities of the mammal heart 1.4.1 Model systems for heart regeneration 1.4.2 Stimulation of cardiomyocyte proliferation as a goal to preserve heart function 1.4.3 Assessment of cardiomyocytes proliferation 1.5 Human-induced pluripotent stem cells to model cardiac development and disease 2. Aim 3. Materials and methods 3.1 TNNT2-FUCCI hiPSC line generation 3.2 hiPSC culture and maintenance 3.3 hiPSC differentiation into CMs 3.4 Imagestream-X Analysis 3.5 TNNT2 expression assessment of hiPSC-derived CMs by immunohistochemistry 3.6 CDK1 immunohistochemistry expression assessment of hiPSC-derived CMs 3.7 Cell area and sarcomere spacing measurement of hiPSC-derived CMs 3.8 Live imaging and timelapse imaging analysis of TNNT2-FUCCI hiPSC 3.9 Murine neonatal CMs cell culture 3.10 Mouse nCM timelapse imaging and analysis 3.11 TNNT2-FUCCI hiPSC-derived CMs culturing and screen conditions 3.11.1 TNNT2-FUCCI screen image acquisition 3.11.2 TNNT2-FUCCI screen automated image analysis 3.11.3 TNNT2-FUCCI screen data analysis 3.12 Primary mouse nCM compound validation and immunohistochemistry 3.13 Mouse nCM immunohistochemistry for AurKB expression assessment 3.14 hiPSC-derived CMs immunohistochemistry for AurKB expression assessment 3.15 Analysis of CM ploidy and binucleation 3.16 in vivo Clonidine treatment of neonatal mice 3.17 Analysis of ploidy and binucleation in in vivo mouse nCMs 3.18 Cell cycle activity assessment in P7 neonatal mouse hearts after Clonidine treatment by immunohistochemistry 4. Results 4.1 TNNT2-FUCCI human iPSC line generation and validation 4.2 TNNT2-FUCCI marks proliferating and non-proliferating CMs 4.3 Live imaging identification of CM cell cycle activity 4.3.1 Cell cycle progression of TNNT2-FUCCI CMs 4.3.2 Cell cycle progression of mouse neonatal cardiomyocytes 4.4 TNNT2-FUCCI live-imaging identifies CM cell cycle activators 4.5 Compound validation in mouse nCMs 4.6 Clonidine elicits cycling activity via alpha1 adrenergic receptor and imidazoline receptor interaction 4.7 Clonidine stimulates hiPSC-derived CM proliferation 4.8 Clonidine stimulates CM polyploidization in mouse nCMs 4.9 Clonidine mediates in vivo CM cell cycle activity in the neonatal mouse 5. Discussion 5.1 TNNT2-FUCCI hiPSC: a new technology to identify cycling CMs 5.2 Study of CM cell cycle activity using TNNT2-FUCCI 5.3 TNNT2-FUCCI hiPSC in combination with a live screening platform revealed enhancer of CMs cell cycle activity. 5.4 Initial validation of pro-proliferative compounds in mouse nCMs reveals Clonidine and Dihydrocapsaicin as enhancers of cell cycle progression 5.5 Clonidine stimulates cell cycle activity in different model systems 5.5.1 Clonidine treatment stimulates proliferation in hiPSC-derived CMs 5.5.2 Clonidine treatment stimulates polyploidization in in vitro- and multinucleation in in vivo mouse nCMs 5.6 Conclusions and future outlooks 6. Appendix 7. Summary 8. Zusammenfassung Acknowledgements References Anlage 1 Anlage 2

Thesis (MScMedSc)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Introduction: Primary culture of isolated adult rat cardiomyocytes (ARCMs) is an important model for cardiovascular research, but successful maintenance of these cells in culture for their use in experiments remains challenging (Xu et al, 2009; Louch et al, 2011). Most studies are done on acutely isolated cardiomyocytes immediately after isolation, which is due to low survival of these cells in culture. Obstacles in culture are due to the type of medium and attachment factors (tissue culture adhesives) used to culture and grow these cells. Although we previously identified an optimum medium and adhesive for culture, an adhesive that permits cells to remain attached to the culture surface until after an ischemia/reperfusion insult was elusive. Aims: We therefore aimed to identify the best attachment factor and concentration that will allow adult rat cardiomyocytes to remain attached to the culture surfaces after ischemia/reperfusion experiments. Methods: Cardiomyocytes were isolated from adult Wistar rat hearts and cultured overnight on different concentrations (25 -200 μg/ml) of collagen 1, collagen 4, extracellular matrix (ECM), laminin/entactin (L/E) and laminin. Following overnight cultures, experiments were done in PBS and in PBS versus MMXCB to compare ARCM attachment and viability. Cardiomyocytes cultured on ECM, L/E and L (25−200μg/ml) were subjected to 1 hour of simulated ischemia using MMXCB that contained 3mM SDT and 10mM 2DG, followed by 15 minutes reperfusion. Cell viability was determined by staining cells with JC-1 and images of cells in a field view of 1.17μm/mm2 were captured using fluorescence microscopy. The cells were analysed according to morphology and fluorescence intensity. Results: Total and rod-shaped ARCMs attachment was improved when MMXCB was used as an experimental buffer instead of PBS. Regardless of the buffer used, morphological viability was poor on substrates of Col 1 and Col 4. In contrast to collagens, ARCMs attached efficiently and morphological viability was high on substrates of ECM, L/E and L in MMXCB, but this was greatly reduced in PBS. Mitochondrial viability was high in MMXCB compared to PBS on Col 1 and Col 4 at 75−175μg/ml and on ECM, L/E and L at all concentrations, except at 50 and 150μg/ml ECM, 175μg/ml L/E and 25μg/ml L. When cardiomyocytes cultured on ECM, L/E and L were subjected to simulated ischemia, total ARCMs, rod-shaped and R/G fluorescence (mitochondrial viability) was reduced at all concentrations compared to the control group. Hypercontracted cells were higher in the ischemic treated cells compared to the controls on ECM at 75−150μg/ml and 200μg/ml, L/E at 50,100μg/ml and 175μg/ml and on L at 125μg/ml. Total numbers of ARCMs attached on ECM, L/E and L in the ischemic group consisted of similar numbers of non-viable hypercontracted and viable rod-shaped cells. Conclusion: Cardiomyocytes should be cultured on ECM or L/E or L at concentrations from 25−200μg/ml in MMXCB. PBS is harmful to cultured ARCMs and should thus not be used as an experimental buffer. Ischemia/reperfusion can be simulated on ARCMs cultured on ECM, L/E or L from 25−200μg/ml, provided that a modified culture buffer is used as experimental buffer. / AFRIKAANSE OPSOMMING: Inleiding: Primêre selkulture van geïsoleerde volwasse rot kardiomiosiete (VRKMe) is ‘n belangrike model vir kardiovaskulêre navorsing, maar om hierdie selle suksesvol in kultuur te onderhou is ‘n groot uitdaging (Xu et al, 2009; Louch et al, 2011). Die meeste navorsingstudies maak gebruik van akuut geïsoleerde kardiomiosiete onmiddelik na isolasie omdat oorlewing van hierdie selle in kultuur baie laag is. Die struikelblokke in kultuur is as gevolg van die tipe medium en weefselkultuurgom wat gebruik word. Ons het voorheen 'n optimale medium en weefselkultuurgom geïdentifiseer vir VRKM kultuur oorlewing, maar die weefselkultuurgom was nie effektief genoeg om die selle aan die kultuuroppervlak te laat bly vaskleef, tot na die einde van 'n isgemie/herperfusie eksperiment nie. Doel: Die doel was dus om die beste weefselkultuurgom en konsentrasie te identifiseer, wat sal toelaat dat VRKMe verbonde bly aan die kultuuroppervlaktes tot na die einde van isgemie/herperfusie eksperimente. Metodes: Kardiomiosiete was geïsoleer vanaf volwasse Wistar rotharte en oornag in kultuur op verskillende konsentrasies (25 -200 μg/ml) van kollageen 1, kollageen 4, ekstrasellulêre matriks (ESM), laminin/entactin (L/E) en laminin onderhou. Die volgende dag was die VRKMe vir eksperimentasie in PBS en in PBS teenoor MMXCB gebruik, om selbehoud en oorlewing te vergelyk. Kardiomiosiete op ESM, L/E en L (25−200μg/ml) was aan 1 uur van gesimuleerde isgemie blootgestel, in MMXCB wat 3mM SDT en 10mM 2DG bevat het, gevolg deur 15 minute herperfusie. Sel oorlewing was bepaal deur selle te kleur met JC-1 en daarna was fluoressensiebeelde van die selle in ‘n veldgebied van 1.17μm/mm2 geneem. Die selle was volgens selmorfologie en fluoressensie intensiteit ontleed. Resultate: Met die gebruik van MMXCB as eksperimentele buffer in plaas van PBS, het die aantal totale en staafvormige VRKMe verbinding verbeter. Morfologiese onderhoud was sleg op kollageen 1 en 4, ongeag van watter buffer gebruik was. In kontras met die kollagene was die VRKM verbinding en morfologiese onderhoud op ESM, L/E en L in MMXCB effektief verbeter, maar in PBS aansienlik verminder. Mitochondriale lewensvatbaarheid in MMXCB teenoor PBS op kollageen 1 en 4 by 75−175μg/ml, sowel as op ECM, L/E en L by alle konsentrasies, was hoog, behalwe by 50 en 150μg/ml ESM, 175μg/ml L/E en 25μg/ml L. Isgemie blootstelling van kardiomiosiete gekultuur op alle konsentrasies van ESM, L/E en L, het ‘n afname in die totale, staafvormige en R/G fluoressensie (mitochondriale lewensvatbaarheid) teweeggebring. Meer hiperkontrakteerde kardiomiosiete was in die isgemie behandelde groepe as in die kontrole groepe teenwoordig, spesifiek op ESM by 75−150μg/ml en 200μg/ml, op L/E by 50,100μg/ml en 175μg/ml asook op L by 125μg/ml. In die isgemie groepe het die totale aantal VRKMe op ESM, L/E en L meestal uit ‘n gelyke hoeveelheid hiperkontrakteerde en staafvormige selle bestaan. Gevolgtrekking: Kardiomiosiete moet op ESM of L/E of L by konsentrasises van 25−200μg/ml in MMXCB gekultuur word. PBS is nadelig vir VRKMe in kultuur en moet dus nie gebruik word as eksperimentele buffer nie. Isgemie/herperfusie eksperimente kan gesimuleer word op VRKMe wat op 25−200μg/ml ESM, L/E of L gekultuur is, mits ‘n gemodifiseerde kultuur buffer gebruik word as eksperimentele buffer.

L’I/R et la cardioprotection par PreC et PostC impactent la fonction mitochondriale et plus précisément le mPTP. Le mPTP est non seulement modulé par des protéines qui participent à sa formation comme la CypD mais aussi par l’environnement cellulaire. Le but de ma thèse a été d’étudier la régulation du mPTP par la CypD, le complexe I et le Ca2+ durant l’I/R et la cardioprotection. Nos conclusions sont : 1. Le complexe I de la chaîne respiratoire mitochondriale régule l’ouverture du mPTP et cela de façon CypD-dépendante. 2. Le PostC est un inhibiteur du complexe I, l’Iso, est le seul PostC efficace chez le rat in vivo. 3. L’inhibition pharmacologique ou génétique de la CypD cardioprotège en modulant l’ouverture du mPTP et l’homéostasie calcique. 4. La CypD a un nouveau rôle dans la cardioprotection, indépendamment de son action sur le mPTP. En effet, elle module le transfert calcique au niveau des MAM et plus précisément le transfert de Ca2+ entre les RS et la mitochondrie. Son inhibition prévient la surcharge calcique mitochondriale intervenant lors de l’I/R. L’ensemble de ces résultats nous permet de conclure que le mPTP est régulé par de nombreux facteurs interconnectés. Le Ca2+ est l’effecteur principal de l’ouverture du mPTP. La CypD a une action Ca2+ dépendante et module l’homéostasie calcique au niveau des MAM. Le complexe I régule l’ouverture du mPTP de façon CypD dépendante. Enfin les fonctions mitochondriales cardioprotecteurs mPTP-dépendants englobe la CypD, le Ca2+, le complexe I et les fonctions mitochondriales. Cette vision plus large et intégrée de la régulation du mPTP pourra donner des pistes plus efficaces dans le développement de traitements pharmacologiques cardioprotecteurs / Reperfusion of the heart after an ischemic event leads to the opening of a nonspecific pore in the inner mitochondrial membrane, the mitochondrial permeability transition pore (mPTP). Inhibition of mPTP opening is an effective strategy to prevent cardiomyocyte death. For example, inhibition of mPTP opening via ischaemic preconditioning (PreC) and post-conditioning (PostC) decreased the myocardial infarct size after ischemia–reperfusion. Although the molecular composition of the mPTP remains unclear, the matrix protein cyclophilin-D (CypD) is the best defined regulatory component of mPTP. In this thesis, we demonstrated that Complex I of the respiratory mitochondrial chain also regulates mPTP in a CypD-dependent manner. We also proved that inhibition of Complex I by isoflurane prevents lethal reperfusion injury in an in vivo rat model of ischemia-reperfusion. Finally, we proved that cardioprotective inhibition of CypD modulates calcium homeostasis and fluxes between mitochondria and sarcoplasmic reticulum. In summary, our results suggest that mPTP is regulated by several interconnected factors like calcium, CypD, complex I and mitochondrial functions

Les entérovirus du groupe B (EV-B) sont une cause majeure de myocardite aiguë, précurseur de la myocardite chronique et de la cardiomyopathie dilatée (CMD) chez l’homme. Les mécanismes moléculaires viraux impliqués dans la progression de la myocardite aiguë vers la phase chronique et la CMD restent inconnus. En utilisant une approche NGS, nous avons détecté des populations persistantes majoritaires d’EV-B tronquées en extrémité 5’, associées à des formes complètes mineures dans des cas de CMD. Afin évaluer leur impact sur la fonctionnalité des cardiomyocytes, nous avons transfecté dans des cardiomyocytes primaires (HCM) des ARN viraux clonés et identiques à ceux détectés dans les cas de CMD. Les formes EV-B majoritaires tronquées en extrémité 5’, seules ou associées à des populations complètes « auxiliaires » pourraient altérer les fonctions des HCM par des activités de la P2A virale. L'existence de mécanismes de recombinaison génomique entre les populations virales persistantes tronquées et complètes a été étudiée par un test de recombinaison d’ARN EV-B défectifs transfectés dans des HCM. Cette approche in vitro a produit majoritairement des recombinants non-homologues caractérisés par des échanges génétiques dans la région 5’NC (spacer1/2). Nos résultats indiquent l’existence d’événements de recombinaison génomique en région 5’ entre les populations d’EV-B tronquées et complètes qui pourraient contribuer au développement de la CMD. Une meilleure compréhension des mécanismes de persistance virale permettra le développement de nouvelles stratégies thérapeutiques pour lutter contre les infections chroniques par les EV-B. / Group-B Enteroviruses (EV-B) are a common cause of human acute myocarditis, a disease that is a precursor of chronic myocarditis and dilated cardiomyopathy (DCM). However, the viral molecular mechanisms involved in the progression of acute to chronic myocarditis and subsequently to DCM remain unknown. Using NGS approach, we detected persistent major EV-B populations characterized by 5’ terminal genomic deletions ranging from 17 to 50 nucleotides associated with minor complete viral forms in explanted hearts of DCM cases. To assess their impact on cardiomyocyte functions, we transfected viral RNA clones mimicking the viral genomes found in patients’ tissues into primary human cardiomyocytes (HCM). Our findings demonstrated that the major persistent 5’ deleted viral forms alone or associated with full-length populations of helper RNAs could impair cardiomyocyte functions by viral 2Apro activities in EV-DCM cases. To assess the existence of genomic recombination mechanisms between persistent deleted and full-length viral helper populations, we used a recombination assay based on the rescue of non-replicative EV-B RNAs transfected in HCM. This in vitro approach produced major (75%) non-homologous recombinants that nucleotides sequencing characterized modular exchanges into the spacers 1 & 2 of the 5’NC region. Our findings indicate the existence of genomic recombination events through which, 5’ deleted and complete collaborative EV-B populations could significantly contribute to the pathogenesis of unexplained DCM cases. A better understanding of these viral persistence mechanisms will stimulate new therapeutic strategies research for chronic infections caused by EV-B.

La plupart des maladies cardio-vasculaires sont étroitement liées à l’architecture 3D des faisceaux de cardiomyocytes du myocarde humain. Connaitre en détail cette architecture permet de lever un verrou scientifique sur l’organisation spatiale complexe des faisceaux de cardiomyocytes, et offre des pistes pour trouver des solutions pertinentes permettant de guérir ces maladies. A cause de la nature biréfringente des filaments de myosine qui se trouvent dans les cellules cardiomyocyte, l’Imagerie en Lumière Polarisée (ILP) se révèle comme la seule méthode existante permettant d’étudier en détail, l’architecture et l’orientation des faisceaux de cardiomyocytes au sein de la masse ventriculaire. Les filaments de myosine se comportent comme des cristaux uni-axiaux biréfringents, ce qui permet de les modéliser comme les cristaux uni-axiaux biréfringents. L’ILP exploite les propriétés vibratoires de la lumière car l’interaction photonique et atomique entre la lumière et la matière permet de révéler l’organisation structurelle et l’orientation 3D des cardiomyocytes. Le présent travail se base sur la modélisation des différents comportements de la lumière après avoir traversé des faisceaux de cardiomyocytes. Ainsi, un volume 100×100×500 µm3 a été décomposé en plusieurs éléments cubiques qui représentent l'équivalent de l'intersection des cellules de diamètre de 20 µm chacune. Le volume a été étudié dans différentes conditions imitant l’organisation 3D des cardiomyocytes dans différentes régions du myocarde. Les résultats montrent que le comportement du volume change suivant l’arrangement spatial des cardiomyocytes à l’intérieur du volume. Grâce à un modèle analytique développé à l’aide des simulations, il a été possible de connaitre en tout point, l’orientation 3D des cardiomyocytes dans tout le volume. Ce modèle a été implémenté dans un greffon logiciel. Puis, il a été validé avec les piliers des valves auriculo-ventriculaire en comparant les courbes obtenues en simulation numérique à celles obtenues dans la phase expérimentale. De plus, il a été possible de mesurer l’orientation 3D des faisceaux de cardiomyocytes à l’intérieur du pilier. Après cette validation, le modèle a été utilisé sur un cœur humain (sain) en entier. Puis, nous avons extrait les cartographies des orientations 3D (angle azimut, angle d’élévation) des cardiomyocytes, ainsi que la cartographie des niveaux d’homogénéité du myocarde en entier. Pour une confrontation qualitative des mesures de l’orientation 3D obtenues en ILP avec celles en IRM, un cœur humain sain d’un enfant de 14 mois a été prélevé lors de l’autopsie, fixé dans du formol, puis imagé en entier par IRM puis en ILP. Malgré la faible résolution des images en IRM, les résultats obtenus montrent que les mesures de l’orientation 3D des cardiomyocytes issues de ces deux méthodes d’imageries se révèlent quasiment identiques. / Most cardiovascular diseases are closely linked to the 3D cardiomyocytes bundles of the human myocardium. Knowing in detail this architecture allows us to overcome a scientific bottleneck on the complex spatial organization of cardiomyocytes, and offers ways to find appropriate solutions to treat these diseases. The goal of present thesis is then to develop methods and techniques that allow gaining insights into the geometric arrangement of cardiomyocytes or cardiomyocytes bundles in the myocardium. Due to the birefringent nature of myosin filaments that are found in myocardial cells, the Polarized Light Imaging (PLI) appears as the only existing method for studying in detail the architecture and cardiomyocytes bundle orientation in ventricular mass. Myosin filaments react as uniaxial birefringent crystal; thereby it has been modeled as the uniaxial birefringent crystal. The PLI uses the vibration properties of light; the photonic and atomic interaction between light and matter can reveal the structural organization and the 3D cardiomyocytes orientation of the myocardium. The present work is based on modeling the behavior of the light after passing through a cardiomyocytes bundle. Thus, a volume 100 × 100 × 500 μm3 has been decomposed in a number of cubic elements which are equivalent to cardiac cells of diameter of 20 microns. The volume was studied under different conditions to emulate the organization of cardiomyocytes in different regions in human myocardium: isotropic region, heterogeneous region, region with cardiomyocytes bundle crossing. The results showed that the behavior of the volume changes according to the spatial arrangement of cardiomyocytes within the volume. Through an analytical model developed using simulation, it has been possible to know the 3D orientation of cardiomyocytes at any region throughout the volume. This model has been implemented in software as a plugin. Then, it has been validated with the pillars of atrio-ventricular valves by comparing the curves obtained by numerical simulation with those obtained in the experimental phases. Moreover, it has been possible to measure the 3D orientation of cardiomyocytes bundles within the pillars. After validation, the model was applied to an entire human healthy heart. Then, we extracted the mapping of the 3D orientations (azimuth angle, elevation angle) of cardiomyocytes bundles, as well as the mapping of the homogeneity levels of the entire myocardium. For a qualitative comparison of the 3D orientation measurements obtained with the PLI and Magnetic Resonance Imaging (MRI), the healthy human heart of a 14 month old child was extracted at autopsy, then fixed in formalin, and finally imaged by MRI and PLI. Despite the low spatial resolution of MRI images, the results showed that the 3D orientations of cardiomyocytes bundles measured from these two imaging methods appeared almost identical.

No description available.