Haematopoietic stem cells (HSC) are invaluable, due to their potential to treat malignant and non-malignant diseases. Modern medicine requires a reliable source of human HSCs (hHSCs) for efficient transplantations, which in many cases cannot be obtained from a single donor. Therefore, the ability to amplify donor hHSCs ex vivo would be an ideal alternative. Past attempts to expand hHSCs in vitro, demonstrated that the protocols developed so far have limited success. My research studied the factors which can affect the optimal culture of transplantable HSCs using a 3D culture system that had previously been used to culture HSCs derived from the aorta-gonad-mesonephros (AGM) region of the mouse embryo. This system involved cell culturing at the gas-liquid interface which is particularly sensitive to mechanical disturbances. To overcome this problem, floating Polypropylene support (rings) were designed and tested and I demonstrated that this was able to prolong aggregate culturing for up to 21 days. Further optimisation tests included altering factors such as oxygen levels, and the presence of antioxidants and apoptosis inhibitors in mouse HSCs culture. I have shown that moderate hypoxia (6% O2) did not affect HSCs in culture, while 2% of O2 led to a significant decrease of HSCs activity. Normoxia resulted in higher reactive oxygen species generation, which would likely be detrimental to cells. However, unexpectedly no improvement in repopulation efficiency of cultured HSCs was achieved by the addition of antioxidant. I also found that when the AGM region was dissociated and co-aggregated in the presence of Rho kinase inhibitor a higher level of repopulation was achieved. In addition, troloxpifitrin-a and p38 inhibitor blocked HSC development without affecting progenitor frequency or the total number of live cells. Subclones of mouse stromal cell line (OP9) were used to create a defined haematopoietic niche for hHSC. Functional screening of these lines in co-aggregate culture re- vealed that 3 of the 34 subclones tested were able to maintain hHSC in culture and repopulate immunodeficient mice at a comparable level to uncultured CD34+ cells. The repopulation in engrafted recipients persisted for over 6 months and showed both myeloid and lymphoid potential. These 3 subclones therefore appeared to create a functional niche for hHSCs and were subsequently used to study the impact of a number of factors including SCF, rock inhibitor, TGFb inhibitor, StemRegenin1 (SR), and prolonged culture technique on hHSC expansion. A significant level of fluctuation between experiments was observed and no definitive conclusions could be drawn. I also attempted to establish stromal cell lines from the human AGM region, more specifically from the ventral (AoV) and dorsal (AoD) regions of the dorsal aorta. Despite attempts to immortalise primary stromal cells, all lines went through a growth crisis. Nevertheless, 30 lines were screened for their ability to support haematopoietic cells in co-aggregate culture with results suggesting that lines derived from AoV expanded haematopoietic precursors more efficiently than AoD lines and OP9 control. Many of the tested lines were able to maintain long-term repopulating human HSCs but the level of repopulation was not as high as that achieved from uncultured CD34+ cells. Unfortunately, these human stromal cell lines have an unstable karyotype which may have an impact on their functional characteristics and they may not represent the nature of the primary cells.

Tot i que es coneix l’involucrament de les cèl·lules pancreàtiques acinars en patologies exocrines (pancreatitis i càncer de pàncrees), la manca de models normals basats en cèl·lules ha limitat l’estudi de les alteracions que succeeixen en el programa de diferenciació pancreàtica. Hem demostrat prèviament que les cèl·lules mare embrionàries murines, que són pluripotents, poden adquirir un fenotip acinar in vitro. Això es va aconseguir, en part, amb una combinació de senyals que provenien del cultiu de pàncrees fetals que no era, però, específic del llinatge pancreàtic. L’objectiu d’aquest treball ha estat el de desenvolupar un protocol selectiu pel llinatge acinar basat en l’activació seqüencial de vies de senyalització que recapitulin el desenvolupament pancreàtic in vivo, a través de la formació definitiva de l’endoderm, l’especificació pancreàtica i acinar i l’expansió/diferenciació de progenitors acinars. El tractament de cossos embrionaris amb Activina A va promoure l’expressió de gens d’endoderm com està prèviament descrit. El tractament subsegüent amb àcid Retinoic, FGF10 i Ciclopamina, un inhibidor de la via de Hedgehog, va resutar en la inducció dels marcadors de progenitors pancreàtics Pdx1, Ptf1a i Cpa1 però també d’aquells expressats en el llinatge pancreàtic, que van ser reduïts amb la inhibició de BMPs. Les cèl·lules van ser a continuació cultivades en Matrigel utilitzant un sistema de cultiu en 3D en presència de fol·listatina, dexametasona i KGF comportant una inducció significativa dels nivells de mRNA i proteïna de marcadors acinars i una disminució de l’expressió dels de marcadors acinars. A més, es va veure que Amyl es secretava en el medi. Aquestes dades indiquen que l’activació selectiva del programa de diferenciació acinar en cèl·lules mare embrionàries es pot dur a terme mitjançant una inducció esgraonada de vies de senyalització involucrades en el desenvolupament pancreàtic exocrí proporcionant una eina potencial per estudiar la diferenciació pancreàtica i malalties relacionades amb el pàncrees. / Despite known involvement of pancreatic acinar cells in exocrine pathologies (pancreatitis and pancreatic cancer), the lack of normal cell-based models has limited the study of the alterations that occur in the acinar differentiation program. We have previously shown that mESC (murine embryonic stem cells), which are pluripotent, can acquire an acinar phenotype in vitro. This was achieved, in part, by a combination of signals provided by the culture of foetal pancreases which was, however, no specific for the acinar lineage. The aim of this work was to develop a protocol selective for the acinar lineage based on the sequential activation of signaling pathways that recapitulate pancreatic development in vivo, through the definitive endoderm formation, the pancreatic and acinar specification and the expansion/differentiation of acinar progenitors. Treatment of embryoid bodies with Activin A enhanced the expression of endodermal genes as previously described. Subsequent treatment with Retinoic acid, FGF10 and Cyclopamine, an inhibitor of the Hedgehog pathway, resulted in the enhancement of pancreatic progenitor markers Pdx1, Ptf1a and Cpa1 but also of those expressed in the hepatic lineage, which were reduced by BMPs inhibition. Cells were further cultured in Matrigel using a 3D culture system in the presence of follistatin, dexamethasone, and KGF leading to a significant enhancement of the mRNA and protein levels of acinar markers while decreasing the expression of endocrine ones. Moreover, active Amyl was released into the medium. These data indicate that the selective activation of the acinar differentiation program in ES cells can be achieved by stepwise induction of signaling pathways involved in pancreatic exocrine development providing a potential tool for studying pancreatic differentiation and pancreas-related diseases.