Modelling and genetic correction of liver genetic diseases

Abstract: The urea cycle is a set of biochemical reactions that converts highly toxic ammonia into urea for excretion. Deficiencies in any of the genes of the cycle can be life-threatening, with liver transplantation currently being the only definitive treatment. However, the scarcity of donor organs dictates the investigation of alternative treatments, which requires appropriate disease models, in vitro and in vivo, that faithfully recapitulate the disease pathology. Recent advancements in the field of genome engineering make interventions in the genetic code less challenging, thereby assisting in the generation of such tools, as well as raising the potential for genetic correction of these conditions. The research conducted in this thesis centres around two broad aims: the investigation of disease models and genetic correction of inherited liver disorders. Induced pluripotent stem cells (iPSC) hold great potential both for disease modelling and as a source of cells for cell therapy. However, their generation through cell reprogramming is sometimes challenging and inefficient. Therefore, in PAPER I we sought to optimize the reprogramming procedure by introducing modifications to the currently existing protocols, and managed to increase the reprogramming efficiency. IPSC could theoretically differentiate into any cell type, including hepatocytes. In order to assess the level of differentiation of the hepatocyte-like cells (HLC) generated from stem cell sources, comparisons with authentic primary liver tissues are necessary. To this end, in PAPER II we created gene expression profiles of fetal and mature (post-natal) liver tissues from a significant number of individuals. The dataset can serve as an accurate and simple assessment tool to evaluate and compare HLC, generated in different laboratories, to authentic human liver tissues. If HLC resemble the functions observed in mature primary hepatocytes, they could be used as in vitro disease models. In addition, programmable nucleases can be applied to either correct or introduce disease-causing of interest in the genome. In PAPER III, we generated iPSC from a patient with a pathogenic variant in the ornithine transcarbamylase (OTC) gene, the most common UCD, corrected the genetic defect and differentiated the cells into HLC. The correction was molecularly, as well as phenotypically confirmed by the restoration of urea cycle function. The thesis also focuses on the investigation of in vivo disease models of UCD. Specifically, in PAPERS IV and V we created liver-humanized mice with hepatocytes from patients with UCD, OTC deficiency (OTCD) or carbamoyl phosphate synthetase 1 deficiency (CPS1D). Highly repopulated animals faithfully recapitulated the clinical manifestations of the disease observed in patients, including hyperammonemia which is considered a hallmark of these UCD. Furthermore, in PAPER V, we investigated the efficacy and safety of ex vivo gene editing of primary OTCD hepatocytes. Ureagenesis was restored in vitro in edited cells, as well as in vivo as mice liver-repopulated with genetically engineered cells partially or completely reversed all markers of the disease investigated. Finally, extensive gene expression and deep sequencing analysis revealed no unspecific mutagenesis effected by the programmable nucleases, pointing out the safety of the application. In conclusion, the research work conducted in this thesis demonstrates the prospects that iPSC and humanized mice possess for the generation of models of liver genetic diseases, in vitro and in vivo. Furthermore, the emergence of genome editing technologies further enhances the aforementioned potentials, as well as raises possibilities for the treatment of liver genetic defects through genome manipulation.