Les cellules entéroendocrines sécrétrices d’hormones représentent 1% de l’épithélium intestinal mais sont des régulateurs essentiels du métabolisme énergétique et l’altération de leur différenciation provoque de graves pathologies métaboliques. Leur différenciation est régie par une cascade de régulations transcriptionnelles qui est encore peu décrite, particulièrement chez l’homme. L’objectif de ce projet de thèse était d’évaluer l’implication de plusieurs facteurs de transcription préalablement identifiés chez la souris (NGN3, RFX6, ARX, PAX4) dans la différenciation entéroendocrine humaine. Pour ce faire, ces gènes ont été inactivés par la technique CRISPR/Cas9 dans des cellules souches pluripotentes induites humaines (hiPSCs), qui ont ensuite été différenciées in vitro en organoïdes intestinaux (HIOs). Les analyses des HIOs déficients pour NGN3 ne permettent pas de conclure quant à un rôle dans la différenciation entéroendocrine mais indiquent une altération de la régionalisation du tissu formé chez les mutants. RFX6 semble important pour la différenciation et/ou la fonction des cellules entéroendocrines, bien que sa fonction précise n’ait pas pu être déterminée. / Hormone-producing enteroendocrine cells represent 1% of the intestinal epithelium but are key regulators of the energetic metabolism and alteration of their differentiation is associated with severe metabolic disorders. Enteroendocrine differentiation is governed by a transcriptional regulatory cascade that is poorly described, especially in humans. This thesis project aimed to evaluate the implication of several transcription factors, previously identified in mice (NGN3, RFX6, ARX, PAX4), in human enteroendocrine differentiation. To do so, these genes were disrupted with the CRISPR/Cas9 system in human inducible pluripotent stem cells (hiPSCs), which were then differentiated in intestinal organoids (HIOs). Preliminary analysis of NGN3-deficient HIOs did not allow a firm conclusion regarding NGN3 implication in enteroendocrine differentiation but showed a tissue regionalization alteration. RFX6 seems important for the differentiation/function of enteroendocrine cells, although its precise function is still to be determined.

Microbiota colonisation of the intestinal tract makes it difficult for pattern recognition receptors (PRR) to discriminate between beneficial microbes and harmful pathogens. We aim to define the roles of cytosolic Nod-like receptors (NLR) in intestinal immunity and homeostasis. Upon activation, some NLR form inflammasomes that mediate the release of inflammatory cytokines and pyroptosis, an inflammatory form of cell death. NLR activation in the non-hematopoietic compartment was shown to be protective during acute intestinal infection. To identify the cell type responsible for this protection, we generated transgenic mice in which the key inflammasome adaptor molecule Asc is selectively ablated in intestinal epithelial cells (IEC) (Asc^ΔVC) and observed that inflammasomes are important for controlling Citrobacter rodentium clearance in these mice. To further dissect the importance of pathogen clearance by IEC inflammasome, ex vivo cultures of primary IEC organoids were established. Thus far this system has revealed profound differences in inflammasome regulation between IEC organoids and bone marrow-derived macrophages (BMDM). This research will inform our understanding of cell type-specific regulation of inflammasomes.

Human senses are specifically designed to recognize and understand the world that surrounds us. Even though we have five senses, vision alone is responsible for at least 30 % of the sensory input to our brain. The visual process is initiated in a highly specialized cell type, the photoreceptors. These are light-sensitive cells located in the retina, a layered nervous tissue situated at the back of the eye. Retinal degeneration diseases are a highly heterogeneous group of conditions that include mutations affecting the survival, maintenance and proper functioning of photoreceptors or the adjacent retinal pigment epithelium (RPE). Such mutations, alone or in combination with environmental factors, cause the loss of the affected cells, and therefore, impairment of the visual sense. Retinitis Pigmentosa and Age-related Macular Degeneration are typical examples of retinal degenerative diseases eventually leading to blindness. In the first one, rod photoreceptors degenerate and consequently also cone photoreceptors are lost. The second is characterized by malfunction and loss of both, RPE and photoreceptor cells. Many current therapeutic approaches for the treatment of retinal degenerative diseases focus on slowing down the progression of the disease, rather than restoring the visual function. Currently, new therapies with the potential to recover the visual signal are under development. Some of these therapeutic strategies have already reached clinical stages, including gene therapy or retinal prosthesis. However, gene therapy approaches require the presence of remaining photoreceptors and, furthermore, particular targeting of disease-related genes. Retinal prosthesis still require improvement in terms of long-term biocompatibility and relevant visual function recovery. An alternative strategy for vision restoration is cell replacement of the lost photoreceptors, which is potentially suitable for targeting late stages of retinal degeneration diseases, independently of the inherent cause of the disease. Human vision relies primarily on cone photoreceptors, which are the cells responsible for color and high acuity vision under daylight conditions. However, cones represent a minority of the photoreceptors within the retina, and so, due to the low availability of these cells, cone photoreceptor transplantation studies lag behind rod transplantation studies. Consequently, in this study, strategies to increase the numbers of cone photoreceptors within mouse embryonic stem cells (mESC)-derived retinal organoids, which represent a potential cell source for transplantation studies, were explored. In this regard, I manipulated developmental pathways known to be involved in retinal development, such as Notch signaling, through the addition of various compounds in the retinal organoid maturation media. However, early cone markers have not yet been definitively identified, complicating the detection and isolation of cone photoreceptor precursors within the organoids. Therefore, a new early cone-reporter mESC line was generated in the course of this study as a valuable tool with the potential to facilitate the development of novel cone photoreceptor replacement therapies. Equally important in the field of photoreceptor cell replacement is the understanding of how the transplanted donor cells interact with the host retina. Previous studies have shown that visual function improvement is possible after transplanting rod or cone-like photoreceptor precursors into the sub-retinal space of mouse models for retinal degeneration. For many years it has been assumed that the underlying mechanism for the observed vision improvement was the migration and structural integration of donor cells into the host outer nuclear layer, where they mature and establish synaptic connections with the host retinal circuitry. However, experiments performed in this study demonstrate, for the first time, that upon transplantation donor and host photoreceptors exchange cytoplasmic material rather than structurally integrate into the host outer nuclear layer. Furthermore, insights into the transferred cytoplasmic content are given, i.e. that mRNA, but not mitochondria are exchanged by donor and host photoreceptors. This novel way of photoreceptor-photoreceptor communication led to a paradigm change in the field of retinal transplantation, requiring a re-interpretation of former transplantation studies. In addition, the discovery of the material transfer phenomenon might serve as a starting point for the development of novel therapeutic strategies based on cell-cell support for the treatment of retinal degenerative diseases. This study generated new knowledge in two important topics related to the development of cell therapies for retinal degeneration diseases, including the development of tools for cone transplantation studies as well as elucidating the interaction between donor and host cells upon transplantation.

Gastric cancer is the second leading cause of cancer related deaths and the fifth most common malignancy worldwide. The prognosis of gastric cancer is often poor. Frequently, the lack of clinical signs lead to a delayed diagnosis with three quarters of patients presenting with non-curable advanced disease. The only curative option is surgery, supported in recent years by perioperative chemotherapy. However, known molecular alterations represent possibilities for targeted therapies to improve overall survival. Nevertheless, biomarkers to predict therapy response are missing, resulting in several failed clinical trials for targeted drugs. Organoids are a recently developed three-dimensional culture system derived from different sources, i.e. adult tissue stem cells, embryonic stem cells (ESC) or induced pluripotent stem cells (iPSC). While in ESC or iPSC derived organoids a functional niche is present that maintains stem cells, this niche is missing in adult stem cell derived organoids and needs to be replaced by a definite medium containing the relevant growth factors. Organoids have the ability of proliferation, self-renewal and self-organization. They show a comparable functionality of the organs they are derived from. In sum, organoids are valuable tools to study diseases on a patient level. In this work, we focused on the characterization of gastric cancer by using human and mouse cancer organoids. Firstly, a human gastric cancer organoid biobank was established. The patient derived organoid lines were characterized concerning their molecular profile, treated with classical chemotherapeutics and mutation specific targeting was performed. The generated human cancer organoids showed a high similarity to the tissue they were derived from and allowed a detailed analysis of observed alterations for each individual patient. However, the high number of mutations effected targeted therapies and needed to be interpreted in the whole mutation spectrum of each specific organoid line. In order to establish organoids with defined mutations for in depth analysis of pathway interference, we decided to combine inducible alleles of frequently altered signaling pathways in gastric cancer in mice and derived organoids of the stomach. These organoid lines were further analyzed by their morphology, functionality and drug response. Successful interference with activated pathways demonstrated their potential usefulness as living biomarkers for therapy response testing. In order to analyze gastric cancer in vivo a stomach specific mouse model was established. Intensive literature and database research resulted in the identification of Annexin10 (Anxa10) as potential stomach specific gene which at the same time is expressed in all different cell types of the stomach epithelium. We therefore generated an inducible Cre recombinase mouse line under the Anxa10 promotor. The Anxa10 CreERT2 line showed only stomach specific recombination events and no restriction to a specific cell type. Nevertheless, activation of Cre resulted in a patchy recombination pattern throughout the whole gland and not a uniform recombination in all cells. Due to this patchy expression, the mouse line is an optimal tool for cancer models, where a complete transformation of an organ is not desired. On the other side it is not useful, if a complete knock-out of a certain floxed allele is needed. This new stomach specific mouse line was then used to model gastric cancer subtypes in vivo. Frequently altered pathways and hotspot mutations of each gastric cancer subtype were defined based on the TCGA database. Alterations were mainly found in the following pathways: RTK/RAS, PI3K/AKT, WNT, TGF β, cell adhesion and chromatin remodelling. We generated and analyzed three different mouse models: one for the chromosomal instability (CIN) subtype and two for the genomically stable (GS) subtype. The different models mimicked very closely the histology of known human gastric cancer subtypes. The intestinal CIN model with mutations in Kras, Smad4 and Tp53 developed tumors with glandular and tubular structures showing morphologies to human intestinal type gastric cancer. The first GS model with alterations in Kras, Cdh1 and Smad4 showed cancers with a diffuse tumor cell morphology with the presence of typical signet ring cells. The second GS model with Kras, Cdh1 and Apc alterations showed similarities to the adenomatous tooth like gastric cancer subtype. Taken together, this study demonstrates that gastric cancer organoids might serve as living biomarkers to predict therapy response and resistance in individual patients. Additionally, the generated gastric cancer mouse model is to our knowledge the first model initiating tumor formation exclusively in the stomach with similar characteristics as described for human gastric cancer. This mouse represents a prime tool for further gastric cancer research.

No description available.

No description available.

The 15q13.3 microdeletion is a common genetic disorder associated with multiple neurodevelopmental disorders including autism spectrum disorder, epilepsy, and schizophrenia. Patients have diverse clinical presentations, often prompting genetic assays that identify the CNV in the clinic. This late-stage screening leaves a considerable gap in our understanding of the prenatal and prediagnostic developmental impairments in these individuals, providing a barrier to understanding the disease pathobiology. We provide the first investigation into embryonic brain development of individuals with the 15q13.3 microdeletion by generating multiple 3D neural organoid models from the largest clinical cohort in reported literature. We incorporated unguided and guided forebrain organoid models into our multi-transcriptomic phenotyping pipeline to uncover changes in cell type emergence and disruptions to circuit development, all of which had underlying changes to cell adhesion pathways. Specifically, we identified accelerated growth trajectories in 15q13.3del unguided neural organoids and used single cell RNA sequencing to identify changes in radial glia dynamics that affect neurogenesis. We measured changes in the pseudotemporal trajectory of matured unguided neural organoids, and later identified disruptions in synaptic signaling modules amongst the primary constituents to neural circuitry, excitatory and inhibitory neurons. We leveraged dorsal and ventral forebrain organoid models to better assess circuit dynamics, as they faithfully produce the excitatory and inhibitory neurons in the pallium and subpallium, respectively. We then used the entire 15q13.3del cohort and performed bulk RNA sequencing on each tissue type at two timepoints and discovered convergence on transcriptional dysregulation and disruptions to human-specific zinc finger proteins localized to chromosome 19. We also identified cell type-specific vulnerabilities to DNA damage and cell migration amongst the dorsal and ventral organoids, respectively, which was consistent with the excitatory and inhibitory neural subpopulations amongst the unguided neural organoids scRNA Seq, respectively. We then examined neuron migration in a 3D assembloid model by sparsely labeling dorsal-ventral forebrain organoids from multiple genotype-lineage combinations. Light sheet microscopy identified deficits in inhibitory neuron migration and morphology, but not migration distance, suggesting a complex disruption to cortical circuitry. This novel combination of cell type characterization, pathway identification, and circuitry phenotyping provides a novel perspective of how the 15q13.3 deletions impair prenatal development and can be applied to other NDD models to leverage understanding of early disease pathogenesis. / Dissertation / Doctor of Science (PhD) / The development of the human brain is a highly complex and tightly regulated process that requires the participation of multiple cell types throughout development. Disturbances to the emergence, differentiation, or placement of these cell types can cause disruptions and local miswiring of neural circuits, which is often associated with neurodevelopmental disorders (NDDs). The 15q13.3 microdeletion syndrome is a highly complex condition associated with multiple NDDs and has seldom been studied in a human context. To address this, we used stem cells derived from a 15q13.3 microdeletion syndrome cohort and their typically developing familial controls to generate unguided (“whole brain”) and region-specific organoids to investigate early fetal development across time. We used the largest 15q13.3 microdeletion cohort in reported literature to identify shared disruptions in early developmental milestones such as neurogenesis, neural migration, and neural patterning. We identified expansion of specific cell populations, including progenitors that later give rise to mature neurons. Abnormalities persisted in more mature cell populations, including the inhibitory neurons responsible for establishing critical microcircuitry in the human cortex. By generating guided organoids that enrich for excitatory and inhibitory neural populations, we were able to merge the models to form assembloids, where we captured early migratory and morphological deficits in inhibitory neuron populations, which is supported by the multi-transcriptomics experiments performed in both organoid models. This study provides a framework for examining fetal development in a neurodevelopmental disorder context. By using the 15q13.3 microdeletion background, we found novel disruptions in cell type emergence and circuit formation previously unreported in mouse or 2D neuron models, highlighting the utility of the phenotyping platform for disease modeling.

Indiana University-Purdue University at Indianapolis (IUPUI) / Human pluripotent stem cells (hPSCs) are an attractive tool for the study of different neurodegenerative diseases due to their potential to form any cell type of the body. Due to their versatility and self-renewal capacity, they have different applications such as disease modeling, high throughput drug screening and transplantation. Different animal models have helped answer broader questions related to the physiological functioning of various pathways and the phenotypic effects of a particular neurodegenerative disease. However, due to the lack of success recapitulating some targets identified from animal models into successful clinical trials, there is a need for a direct translational disease model. Since their advent, hPSCs have helped understand various disease effectors and underlying mechanisms using genetic engineering techniques, omics studies and reductionist approaches for the recognition of candidate molecules or pathways required to answer questions related to neurodevelopment, neurodegeneration and neuroregeneration. Due to the simplified approach that iPSC models can provide, some in vitro approaches are being developed using microphysiological systems (MPS) that could answer complex physiological questions. MPS encompass all the different in vitro systems that could help better mimic certain physiological systems that tend to not be mimicked by in vivo models. In this dissertation, efforts have been directed to disease model as well as to understand the intrinsic as well as extrinsic cues using two different MPS. First, we have used hPSCs with Alzheimer’s disease (AD)-related mutations to differentiate into retinal organoids and identify AD related phenotypes for future studies to identify retinal AD biomarkers. Using 5 month old retinal organoids from AD cell lines as well as controls, we could identify retinal AD phenotypes such as an increase in Aβ42:Aβ40 ratio along with increase in pTau:Tau. Nanostring analyses also helped in identification of potential target genes that are modulated in retinal AD that were related to synaptic dysfunction. Thus, using retinal organoids for the identification of retinal AD phenotypes could help delve deeper into the identification of future potential biomarkers in the retina of AD patients, with the potential to serve as a means for early identification and intervention for patients. The next MPS we used to serve to explore non-cell autonomous effects associated with glaucoma to explore the neurovascular unit. Previous studies have demonstrated the degeneration of RGCs in glaucoma due to a point mutation OPTN(E50K) that leads to the degeneration of RGCs both at morphological and functional levels. Thus, using the previous studies as a basis, we wanted to further unravel the impact of this mutation using the different cell types of the neurovascular unit such as endothelial cells, astrocytes and RGCs. Interestingly, we observed the barrier properties being impacted by the mutation present in both RGCs and astrocytes demonstrated through TEER, permeability and transcellular transport changes. We also identified a potential factor TGFβ2 that was observed to be overproduced by the OPTN E50K astrocytes to demonstrate similar effects with the exogenous addition of TGFβ2 on the barrier. Furthermore, the inhibition of TGFβ2 helped rescue some of the barrier dysfunction phenotypes. Thus, TGFβ2 inhibition can be used as a potential candidate that can be used to further study its impact in in vivo models and how that can be used in translational applications. Thus, MPS systems have a lot of applications that can help answer different physiologically relevant questions that are hard to approach using in vivo models and the further development of these systems to accentuate the aspects of neural development and how it goes awry in different neurodegenerative diseases.

Preterminfants frequently suffer frompulmonary complications due to a physiological and structural lung immaturity resulting in significant morbidity and mortality. Novel in vitro and in vivo models are required to study the underlying mechanisms of late lung maturation and to facilitate the development of new therapeutic strategies. Organoids recapitulate essential aspects of structural organization and possibly organ function, and can be used to model developmental and disease processes. We aimed at generating fetal lung organoids (LOs) and to functionally characterize this in vitro model in comparison to primary lung epithelial cells and lung explants ex vivo. LOs were generated with alveolar and endothelial cells from fetal rat lung tissue, using a Matrigel-gradient and air-liquid-interface culture conditions. Immunocytochemical analysis showed that the LOs consisted of polarized epithelial cell adhesion molecule (EpCAM)-positive cells with the apical membrane compartment facing the organoid lumen. Expression of the alveolar type 2 cell marker, RT2-70, and the Club cell marker, CC-10, were observed. Na+ transporter and surfactant protein mRNA expression were detected in the LOs. First time patch clamp analyses demonstrated the presence of several ion channels with specific electrophysiological properties, comparable to vital lung slices. Furthermore, the responsiveness of LOs to glucocorticoids was demonstrated. Finally, maturation of LOs induced by mesenchymal stem cells confirmed the convenience of the model to test and establish novel therapeutic strategies. The results showed that fetal LOs replicate key biological lung functions essential for lung maturation and therefore constitute a suitable in vitro model system to study lung development and related diseases.

Environmental factors such as diet contribute to the pathogenesis of inflammatory bowel disease (IBD). Epidemiological evidence suggests a robust linkage between IBD and the Western diet, which is often characterized by a high intake of food additives. These additives, including synthetic colourants, are widely used, leading to significant human exposure. Allura Red AC (AR) is one of the most popular synthetic colourants, yet little is known about its impact on human health and the role of AR in the pathogenesis of colitis remains elusive. Serotonin (5-hydroxytryptamine; 5-HT), which regulates various gut physiological processes, has been shown to modulate the gut microbiota and enhance susceptibility to colitis. In this thesis, it was discovered that chronic exposure to AR, at a dose found in commonly consumed dietary products, exacerbated dextran sulfate sodium (DSS)-induced colitis and triggered early onset of disease in the CD4+CD45RBhigh T cell-induced colitis model. AR also induced low grade colonic inflammation in naïve C57BL/6 mice. Exposure to AR was associated with increased colonic 5-HT levels and impaired intestinal barrier function via activation of the myosin light chain kinase (MLCK) pathway. However, AR did not promote colitis in mice lacking tryptophan hydroxylase 1 (Tph1), the rate-limiting enzyme responsible for colonic 5-HT synthesis. Further, AR increased colonic 5-HT levels in germ-free (GF) mice and perturbed the gut microbiota composition in specific pathogen-free (SPF) mice. Transfer of this altered microbiota from the dye-exposed SPF mice to GF mice conferred enhanced susceptibility to DSS-induced colitis. Mechanistically, AR induced reactive oxygen species (ROS) generation and promoted 5-HT secretion via the NF-κB pathway in BON cells. Data in this thesis indicate that the widely used synthetic colourant, AR, promotes colitis via colonic 5-HT in microbiota-dependent and -independent pathways. Collectively, these findings provide important information on enhancing public awareness of its detrimental effects on human health. / Thesis / Candidate in Philosophy / Epidemiological and experimental studies suggest a potential link between inflammatory bowel disease (IBD) and diet. The Western diet, often characterized by a high intake of processed foods, is associated with the growing incidence of IBD. Allura Red AC (AR) is a popular artificial food dye found in highly common processed foods, yet little is known about its impact on human health and disease. Serotonin, a key molecule in the gut, has been implicated in large bowel inflammation. Herein, the potential role of AR in the development of colitis was examined. Across multiple models, AR exposure heightened vulnerability to colitis in mice, an effect attenuated by reduced serotonin production in the gut. The effect of AR in enhancing colitis vulnerability occurred via gut microbiota-dependent and -independent pathways. These studies have identified how AR promotes colitis, findings that may advance public health awareness and impact the health of patients with IBD.