As cardiovascular disease remains the global leading cause of death, there is an urgent need to study the pathophysiology of the heart and to effectively evaluate cardioprotective drugs. Due to the difficulty in studying the human heart in vivo and sourcing human heart tissues, induced pluripotent stem cells (iPSCs) have provided a promising alternative for modeling cardiac diseases and evaluating candidate drugs. In recent years, computational methods, including machine learning, have given rise to a new class of tools to evaluate cardiac function more rapidly and comprehensively. In this dissertation, I develop and apply computational tools to probe the function of human iPSC-derived cardiac models (Aim 1), apply machine learning methods in the context of cardiomyocyte disease phenotyping and cardioactive drug profiling (Aim 2), and develop a pipeline for deep learning-driven cardiac fibroblast phenotypic drug discovery (Aim 3).

Human induced pluripotent stem cell (iPSCs)-derived brain organoids have potential to recapitulate the earliest stages of brain development, serving as an effective in vitro model for studying both normal brain development and disorders. In this study, we demonstrate a straightforward approach of generating multiple cerebral organoids from iPSCs on a pillar plate platform, eliminating the need for labor-intensive, multiple transfer and encapsulation steps to ensure the reproducible generation of cerebral organoids. We formed embryoid bodies (EBs) in an ultra-low attachment (ULA) 384-well plate and subsequently transferred them to the pillar plate containing Matrigel, using a straightforward sandwiching and inverting method. Each pillar on the pillar plate contains a single spheroid, and the success rate of spheroid transfer was in a range of 95 - 100%. Using this approach, we robustly generated cerebral organoids on the pillar plate and demonstrated an intra-batch coefficient of variation (CV) below 9 – 19% based on ATP-based cell viability and compound treatment. Notably, our spheroid transfer method in combination with the pillar plate allows miniaturized culture of cerebral organoids, alleviates the issue of organoid variability, and has potential to significantly enhance assay throughput by allowing in situ organoid assessment as compared to conventional organoid culture in 6-/24-well plates, petri dishes, and spinner flasks.

Spinal cord injuries result in permanent loss of motor function, leaving those affected with long term physical and financial burdens. Strategies for spinal cord injury repair must overcome unique challenges due to scar tissue that seals off the injury site, preventing regeneration. Tissue engineering can address these challenges with scaffolds that serve as cell- and drug-delivery tools, replacing damaged tissue while simultaneously addressing the inhibitory environment on a biochemical level. To advance this approach, the choice of cells, biomaterial matrix, and drug delivery system must be investigated and evaluated. This research seeks to evaluate (1) the behaviour of murine induced pluripotent stem cells in previously characterized 3D fibrin matrices; (2) the 3D fibrin matrix as a platform to support the differentiation of human induced pluripotent stem cells; and (3) the ability of an affinity-based drug delivery system to control the release of emerging spinal cord injury therapeutic, heat shock protein 70 from fibrin scaffolds. / Graduate / 0541 / amy.lynn.montgomery@gmail.com

No description available.

Les cellules souches pluripotentes humaines (hPSC) sont des cellules capables à la fois d'autorenouvellement et de se différencier en tous les types cellulaires. Elles peuvent être issues de l'embryon (pour cellules souches embryonnaires humaines, hESC) ou être obtenues par reprogrammation d'une cellule différenciée (pour cellules souches pluripotentes induites humaines, hiPSC). Les hPSC sont au centre d'enjeux scientifiques, médicaux et économiques majeurs, en particulier dans le cadre des maladies génétiques et orphelines. En effet, elles ouvrent la porte à de nouvelles stratégies de modélisation de maladies génétiques humaines in vitro et sont une source potentiellement illimitée de cellules pour une thérapie cellulaire des maladies dégénératives. Cependant, la culture in vitro de hPSC est une étape essentielle avant toute application clinique ou recherche fondamentale. En effet, la culture cellulaire est nécessaire pour l'amplification du nombre des cellules, et est nécessaire pour toute étape de différenciation in vitro. Or c'est une étape délicate pour le succès des applications visées. Mon travail de doctorat s'est focalisé sur deux aspects de la culture des hPSC. Dans un premier temps, j'ai modélisé in vitro une voie de différenciation, le développement trophoblastique humain, en modulant les paramètres de la condition de culture, notamment en jouant sur la concentration du facteur de croissance BMP4. Ce travail m'a permis d'élucider la toute première bifurcation de différenciation cellulaire au cours du développement embryonnaire humain précoce. Dans un second temps, mon travail s'est focalisé sur le changement phénotypique et génomique des hPSC au cours de la culture in vitro. J'ai montré que l'utilisation de certains protocoles de passage cellulaire – en particulier le passage par dissociation cellulaire complète par utilisation de trypsine - se traduit par des acquisitions très précoces d'anomalies génétiques chromosomiques et sub-chromosomiques, et que des anomalies sub-chromosomiques pouvaient précéder l'apparition d'anomalies chromosomiques. Les conséquences de ces observations sont importante pour la recherche de la culture de hPSC : (1) il faut définitivement renoncer à l'utilisation des passages par dissociation cellulaire complète, y compris pour les méthodes de culture en suspension, et (2) il faut, pour valider une technique de culture, compléter systématiquement le caryotype par un examen d'analyse génétique avec une meilleure résolution. / Human pluripotent stem cells (hPSC) are the stem cells capable to self-renew and also to differentiate into all the cell types. These cells can be derived from embryos (for human embryonic stem cells, hESC) but also be obtained by reprogramming the differentiated somatic cells (for human induced pluripotent cells, hiPSC). The hPSC become central stakes of science, medicine and economy, particularly for genetic and rare diseases. In fact, they open up the new perspectives to the novel treatment strategies by remodeling human genetic diseases in vitro and at the same time they are a potentially unlimited cell source for cell therapy for especially degenerative diseases. Meanwhile, the hPSC in vitro culture is one of the most important steps before passing to the clinic applications and in fundamental research, as the proliferation and pluripotency can only be maintained in culture condition as well as many differentiation methods. My PhD work was concentrated on the hPSC in vitro culture. At first, I modeled human trophoblastic development and its differentiation pathway in vitro by modulating the parameters of culture, especially the concentration of BMP4. This work permitted clarifying the first cell lineage bifurcation in early human embryonic development. Secondly, my word was focalized on the phenotypic and genomic changes of hPSC during the in vitro culture. I demonstrated that the use of some passaging protocols in culture, particularly complete cell dissociation by trypsin, was translated by very early acquisitions of chromosomal and sub-chromosomal abnormalities, and that the appearance of sub-chromosomal abnormalities could precede chromosomal abnormalities. The consequences of these observations are important for the hPSC culture research: (1) the use of complete cell-dissociation passaging should be definitively abandoned, including the suspension culture, and (2) the genetic analyses with higher resolution should be added to validate a culture technic.

Les porphyries érythropoïétiques (PE) : Porphyrie Erythropoïétique Congénitale -PEC- et Protoporphyrie Erythropoïétique -PPE- sont caractérisées par le déficit d’une des enzymes de la voie de biosynthèse de l’hème. Le traitement curatif des formes sévères de PE est la transplantation de moelle osseuse allogénique (TMOA). La PPE est parfois compliquée d’une insuffisance hépatique majeure nécessitant une greffe hépatique. Dans un modèle murin de PPE (Fechm1Pas/Fechm1Pas), nous avons démontré l’apparition progressive de lésions hépatiques dès la 2ème semaine de vie. Une TMO précoce (nouveau-né) a permis de prévenir l’apparition de ces lésions hépatiques et de corriger la photosensibilité cutanée démontrant l’efficacité de cette approche thérapeutique pour les formes sévères de PPE. La thérapie génique par greffe de cellules souches hématopoïétiques autologues corrigées représente une alternative à la TMOA en l’absence de donneur HLA-compatible. Nous avons développé des cellules souches pluripotentes induites (iPS) à partir de cellules épidermiques issues de modèles murins de PE et d’un patient PEC. La correction génique a été obtenue par transfert du gène lentiviral (ferrochélatase ou uroporphyrinogène III synthase (UROS). La pluripotence des cellules iPS a été caractérisée in vitro par la formation de corps embryoïdes et in vivo par la formation de tératomes. In vitro, la correction métabolique a été obtenue après différenciation des cellules iPS humaines en progéniteurs hématopoïétiques. Enfin dans une dernière partie, nous nous sommes intéressés à une approche pharmacologique de la PEC. Nous avons montré que les mutations C73R et P248Q entraînaient une instabilité et une dégradation accélérée de l’UROS par la voie du protéasome. Le traitement de souris UrosP248Q par un inhibiteur du protéasome (Velcade®) a permis la correction de la photosensibilité cutanée. Ces travaux ouvrent de nouvelles perspectives pour le traitement des porphyries érythropoïétiques. / Erythropoietic porphyrias (EP) : Congenital Erythropoietic Porphyria -CEP- and Erythropoietic Protoporphyria -EPP-) are characterized by a deficit of one enzyme implicated in heme biosynthetic pathway. The curative therapy for severe cases of EP is an HLA-compatible Bone Marrow Transplantation (BMT). EPP is sometimes complicated by a major hepatic failure requiring hepatic graft. In a murine model of EPP (Fechm1Pas/Fechm1Pas), we have demonstrated that hepatic lesions progressively appear 2 weeks after birth. Early BMT (in neonates) has made it possible to prevent hepatic lesions and correct skin photosensitivity, demonstrating the efficiency of this therapeutic approach in severe cases of EPP. The gene therapy by graft of corrected autologous hematopoietic stem cells represents an alternative to BMT when HLA-compatible donors are lacking. We have developed induced pluripotent stem cells (iPSC) from epidermic cells of murine models of EP and of one PEC patient. The gene correction was obtained by lentiviral gene transfer (ferrochelatase and uroporphyrinogen III synthase -UROS). The pluripotency of iPSC was characterized in vitro by the formation of embryoid bodies and in vivo by the formation of teratomas. In vitro, the metabolic correction was obtained after differentiation of human IPSC into hematopoietic progenitors. In the last part of this thesis, we have focused on a pharmacological approach of CEP. We have shown that C73R and P248Q mutations lead to instability and accelerated degradation of the UROS protein via the proteasome. Treating UrosP248Q mice with a proteasome inhibitor (Velcade®) has allowed the correction of skin photosensitivity. These works offer new prospects for the treatment of erythropoietic porphyrias.

A síndrome de Prader Willi (PWS) é uma doença de neurodesenvolvimento; a principal hipótese de causa de PWS é a ausência da expressão de SNORD116. O SNORD116 fica na região 15q11-q13 que apresenta vários genes com imprinting genômico e é conhecida por ser controlada pela região de controle de imprinting PWS (PWS-IC) que se localiza sobreposta à região promotora e ao exon 1 do gene SNRPN. Em camundongos, uma proteína zinc finger (Zfp57) foi descrita como importante para o estabelecimento e manutenção do imprinting no Snrpn. Através de análise do ENCODE do Genome Browser, verificamos que outra proteína zinc finger (ZNF274) se liga ao SNORD116. ZNF274 é conhecida por formar um complexo com TRIM28 e SETDB1 que inibe a expressão através da trimetilação da lisina 9 na histona 3 (H3K9me3). No atual estudo mostramos que ZNF274 se liga ao SNORD116 preferencialmente ao alelo materno nas células-tronco pluripotente induzidas (iPSCs). Adicionalmente, as proteínas TRIM28 e SETDB1, que formam um complexo com a ZNF274, estão presentes na região do SNORD116, e a modificação H3K9me3 ocorre preferencialmente no alelo materno nas iPSCs. Na análise funcional, mostramos que o knockdown de SETDB1 isoladamente ou combinado com o knockdown de ZNF274 causa aumento na expressão de SNRPN e SNORD116 nas iPSCs. Além disso, ocorre redução do H3K9me3 e aumento da modificação relacionada à ativação da transcrição, H3K4me2 (dimetilação da lisina 4 na histona 3), na PWS-IC. Os knockdowns também afetam a metilação de DNA, ocasionando o aumento de 5-hidroximetliação de citosinas na PWS-IC. Em outros tipos celulares estudados, neurônios derivados de iPSCs e SHEDs, ZNF274 e a modificação H3K9me3 ocorrem em ambos os alelos dentro do SNORD116. É possível que, nas iPSCs, este complexo proteja a região imprintada da desmetilação do DNA de proteína(s) que atue(m) nessa região somente em células pluripotentes. Nossos achados possibilitam melhor compreensão dos mecanismos envolvidos no imprinting da região 15q11-q13, principalmente do SNORD116, e, consequentemente, disponibiliza novas ferramentas para o desenvolvimento de futuras terapias para PWS. / Prader-Willi syndrome (PWS) is a neurodevelopmental disorder. Loss of paternal copies of the cluster of SNORD116 C/D box snoRNAs and their host transcript, 116HG, on human chromosome 15q11-q13 imprinted region is considered to be the major responsible for PWS. PWS-imprinting center (PWS-IC) regulates 15q11-q13 imprinting. PWS-IC is located upstream and in the exon 1 of SNURF-SNRPN gene. In mice, Zfp57 plays an important role in establishment and maintenance of Snrpn imprinting. In human, ENCODE database indicates that ZNF274 binds to SNORD116. Moreover, ZNF274 are C2H2/KRAB zinc finger proteins as Zfp57. We have investigated the mechanism of repression of the maternal SNORD116. Here, we report that the ZNF274, in association with the histone H3 lysine 9 (H3K9) methyltransferase SETDB1, is part of a complex that binds to the silent maternal but not to the active paternal alleles in induced pluripotent stem cells (iPSCs). Knockdown of SETDB1 in PWS-specific iPSCs causes a decrease in the accumulation of H3K9 trimethylation (H3K9me3) at SNORD116. We also show that upon knockdown of SETDB1 in PWS-specific iPSCs, expression of maternally silenced 116HG RNA is partially restored. SETDB1 knockdown in PWS iPSCs also disrupts DNA methylation at the PWS-IC where a decrease in 5-methylcytosine is observed in association with a concomitant increase in 5-hydroxymethylcytosine. In iPSCs-derived neurons and stem cells from human exfoliated teeth (SHEDs) ZNF274/SETDB1 complex binding and H3K9me3 modification occur in both alleles. These observations suggest that the ZNF274/SETDB1 complex bound to the SNORD116 cluster may protect the PWS-IC from DNA demethylation during early development, as indicated by iPSCs. Our findings reveal novel epigenetic mechanisms that function to repress the maternal 15q11-q13 region. The better understanding of epigenetic mechanisms provides new tools for future therapy research.

Os transtornos do espectro autista (TEA) são um grupo de doenças neuropsiquiátricas caracterizadas por um prejuízo na capacidade de comunicação e de interação social e por padrões comportamentais estereotipados. Os TEA são geneticamente heterogêneos o que dificulta a identificação das alterações genéticas que estão contribuindo para estes transtornos. No presente estudo, selecionamos como uma primeira abordagem o estudo de translocações cromossômicas, buscando encontrar genes candidatos para posteriores estudos funcionais. No primeiro caso, uma translocação de novo balanceada envolvendo os cromossomos 2q11 e Xq24, não identificamos nenhum candidato funcional rompido pelos pontos de quebra. Detectamos ainda a presença de uma isodissomia materna do cromossomo 5 nesta paciente. Este resultado sugere que, possivelmente, tanto a translocação cromossômica quanto a isodissomia devem estar contribuindo para a etiologia do TEA nesta paciente, caracterizando este como um caso de efeito poligênico. Já o estudo da translocação de novo balanceada (3,11)(p21,q22) revelou que o gene TRPC6, um canal de cálcio envolvido no desenvolvimento de dendritos e sinapses excitatórias, encontrava-se rompido no cromossomo 11 deste paciente. As análises dos neurônios e células progenitoras neurais deste paciente obtidas através da técnica de reprogramação celular e o estudo global de expressão gênica sugerem fortemente que o rompimento do gene TRPC6 é o fator etiológico do TEA neste caso. Por fim, nós também realizamos um estudo de expressão gênica global de pacientes autistas idiopáticos e verificamos que os genes diferencialmente expressos nestes pacientes estão principalmente envolvidos na regulação da dinâmica do citoesqueleto, indicando que este pode ser o processo biológico comumente afetado nos pacientes autistas. Nosso trabalho mostra que os estudos citogenéticos são importantes para a identificação de genes candidatos para os TEA e reforça a hipótese de que estes transtornos são causados por diferentes variantes genéticas mas que levam ao comprometimento de um processo biológico comum. Acreditamos que o modelo de reprogramação celular contribuirá para o entendimento da implicação de tais processos na etiologia dos TEA. / Autism spectrum disorders (ASD) are a group of neurodevelopmental diseases characterized by impairments in social and communicative skills and repetitive behaviors. The investigation of ASD causes is hampered by the genetic heterogeneity of these neurodevelopmental diseases. In the present study, we mapped the breakpoints associated to chromosomal translocations found in two autistic patients as a first screening approach, trying to identify single candidate genes that could be further investigated by functional analysis. In the first case, a de novo balanced translocation involving the chromosomes 2q11 and Xq24, we did not find any functionally known relevant gene disrupted by the breakpoints but, surprisingly, SNP-array data showed that the patient also presents a maternally inherited isodisomy on chromosome 5. In this case, is possible that ASD is caused by the combination of the molecular results caused by the translocation and the UPD on chromosome 5, which would characterize this case as an example of polygenic effects on ASD etiology. On the other hand, the study of a second case, a boy with a de novo balanced translocation (3;11)(p21;q22), revealed that TRPC6, a calcium channel involved in dendritic spine and excitatory synapse formation, was disrupted by the translocation on chromosome 11. Making use of cellular reprogramming to generate neurons and neuronal progenitor cells from this patient and expression analysis, we demonstrated that TRPC6 disruption can respond for the phenotype seen in this patient. Finally, we also performed a genome-wide expression analysis to investigate idiopathic autistic patients and we verified that ASD DEGs are mainly implicated in cytoskeleton dynamics, suggesting that the regulation of this cellular structure can be one of the common mechanisms of ASD etiology. Our work shows that cytogenetic studies are important for the identification of ASD candidate genes and reinforces the hypothesis that these disorders are caused by different genetic variants that are implicated in a common biological process. We believe that cellular reprogramming will contribute for the understanding of the implication of such biological processes in the etiology of ASD.

Les cellules souches pluripotentes (CSP) humaines sont devenues en quelques années des modèles de choix pour étudier les mécanismes cellulaires et moléculaires qui gouvernent l'apparition de maladies monogéniques, mais également pour le développement de criblages à haut débits afin d'identifier parmi plusieurs milliers de molécules chimiques celles qui ont un potentiel thérapeutique. C'est dans ce contexte de criblage que mes travaux de thèse s'inscrivent, alliant automatisation et miniaturisation de la biologie des CSP dans le cadre de deux maladies monogéniques, l'amyotrophie spinale infantile (SMA) et la dystrophie myotonique de type I (DM1). De manière générale, la mise en place d'une telle stratégie repose sur trois étapes essentielles qui sont l'obtention de CSP porteuses d'une mutation donnée, l'identification d'un modèle d'étude pertinent et la réalisation du criblage à proprement parlé. L'obtention de CSP humaines repose sur deux approches principales. La première consiste en la dérivation de cellules embryonnaires humaine (hES) issues de diagnostiques préimplantatoires et la seconde repose sur la reprogrammation de cellules somatiques par l'induction de pluripotence (iPS). Une partie de mon travail a consisté en la création de cellules iPS modèles de la SMA et leur caractérisation par une approche à haut débit. Par la suite un travail d'optimisation du protocole de génération de motoneurones à partir de CSP humaines a permis d'accélérer et augmenter les rendements de production de ces cellules qui sont principalement affectées dans la SMA. Enfin, l'utilisation de cellules hES porteuses de la mutation causale de la DM1 a permis le criblage de 12000 molécules et a conduit à l'identification d'une famille chimique capable de restaurer plusieurs défauts typiques de cette maladie tels que des défauts d'épissage et de fusion moléculaire. / For only few years, Human pluripotent stem cells (PSC) have become wide spread models in order to study and decipher cellular or molecular mechanims involved in monogenic diseases, but also for the development of large scale screening strategies allowing the identification of new therapeutics among thousands of chemicals. Mythesis research aimed at the development of such strategies, miniaturizing and automating PSC biology within the framework of two monogenic diseases, namely spinal muscular atrophy (SMA) and myotonic dystrophy type 1 (DM1).Basically, PSC based screening programs are generally built around three main steps which are the access to a stem cell model, the identification of a relevant cell type and lastly the screening campaign. There is actually two main ways to generate human PSC. Firstly, human embryonic stem cells (hES) can be derived from the inner cell mass of blastocyte through a pre-implantation diagnosis and secondly, induced pluripotent stem cells (iPS) can be generated after somatic cell reprogramming in vitro. A part of my work has consisted in the generation of hiPS cellular models for SMA by reprogramming fibroplasts that carried SMN1 gene deletion, followed bay the characterization of several dozen of independant clones with high throughput. Then an optimization process of the protocol for the generation of Motoneuron from PSC has been done multiplying experimental conditions. This finally allowed the description of a fast and efficient protocol to generate the most affected cell type in SMA. Finally, DM1 mutated hES were uded for the screening of 12.000 compounds among which a chemical family has been identified to rescue DM1 typical splicing and myogenesis defects.

Heart failure is one of the leading causes of death in the United States. Every year in the United States, more than 800,000 people are diagnosed with heart failure and more than 375,000 people die from heart disease. Current therapies such as heart transplants and bioartificial hearts are helpful, but not optimal. Decellularization of porcine whole hearts followed by recellularization with patient-specific human cells may provide the ultimate solution for patients with heart failure. Great progress has been made in the development of efficient processes for decellularization, and the design of automated bioreactors. In this study, the decellularization of porcine hearts was accomplished in 24 h with only 6 h of sodium dodecyl sulfate (SDS) exposure and 98% DNA removal. Automatically controlling the pressure during decellularization reduced the detergent exposure time while still completely removing immunogenic cell debris. Stimulation of macrophages was greatly reduced when comparing native tissue samples to the processed ECM. Complete cell removal was confirmed by analysis of DNA content. General collagen and elastin preservation was demonstrated by SEM and histology. The compression elastic modulus of the ECM after decellularization was lower than native at low strains but there was no significant difference at high strains. Polyurethane casts of the vasculature of native and decellularized hearts demonstrated that the microvasculature network was preserved after decellularization. A static blood thrombosis assay using bovine blood was also developed. A perfusion bioreactor was designed and right ventricle of the decellularized hearts were recellularized with human endothelial cells and cardiac fibroblasts. An effective, reliable, and relatively inexpensive assay based on human blood hemolysis was developed for determining the remaining cytotoxicity of the cECM and the results were consistent with a standard live/dead assay using MS1 endothelial cells incubated with the cECM. Samples from the left ventricle of the hearts were prepared with 300 µm thickness, mounted on 10 mm round glass coverslips. Human induced pluripotent stem cells were differentiated into cardiomyocytes (CMs) and 4 days after differentiation, cardiac progenitors were seeded onto the decellularized cardiac slices. After 10 days, the tissues started to beat spontaneously. Immunofluorescence images showed confluent coverage of CMs on the decellularized slices and the effect of the scaffold was evident in the arrangement of the CMs in the direction of fibers. This study demonstrated the biocompatibility of decellularized porcine hearts with human CMs and the potential of these scaffolds for cardiac tissue engineering. Further studies can be directed toward 3D perfusion recellularization of the hearts and improving repopulation of the scaffolds with various cell types as well as adding mechanical and electrical stimulations to obtain more mature CMs.