Production of canine hepatocyte-like cells from stem cell sources

Abstract

The cost of drug development is high with many drugs failing during toxicity testing. This is a particular problem in veterinary medicine where the pharmaceutical market size is so small that it may not be economically viable to develop drugs. The liver and specifically hepatocytes have a crucial role in drug metabolism via oxidation by cytochrome enzymes (CYP), conjugation and excretion into the biliary system. This drug metabolism is unpredictable between species as each has unique CYP profiles. Furthermore there is breed variation of CYP profiles within the canine species. The ability to produce an in vitro source of canine hepatocytes to model drug metabolism in this species and in different breeds would greatly reduce the expense of candidate drug testing. If an unlimited supply could be produced in vitro this would reduce the number of animals required in pre-clinical testing. The aim of this thesis was to produce an in vitro supply of canine hepatocyte-like cells from stem cell sources, namely hepatic progenitor cells (HPC), mesenchymal stem cells (MSC) or induced pluripotent stem cells (iPSC). Cultures of canine primary hepatocytes were produced to use as a gold standard, but also to develop and refine tests of hepatocyte characterisation and function. A panel of primers was developed for use in real time polymerase chain reaction (PCR) as well as optimising tests for low density lipoprotein (LDL) and indocyanine green uptake, albumin production, periodic acid- Schiff staining for glycogen and CYP activity using a luciferase-based system. As primary hepatocytes rapidly lost their defining characteristics and function in vitro, methods of maintaining function using CYP inducers and culture substrates were assessed. Isodensity centrifugation and magnetic-activated cell sorting was employed to isolate HPCs. Selection of cells from the non-parenchymal cell fraction with stem cell marker Prominin 1 demonstrated that these were keratin 7 positive, a HPC marker. Cells morphologically consistent with HPC appeared and expanded in culture after 2 weeks. On passaging, these cells failed to continue expanding, despite plating onto collagen, laminin, SNL feeder cells or using Kubota’s medium (known to allow rapid expansion of rodent and human HPCs). Canine adipose (Ad-MSC) and bone marrow-derived mesenchymal stromal cells (BM-MSC) were isolated post mortem. These were characterised as CD45, 105 and STRO-1 positive, CD11b, 19 and 45 negative cells which could be differentiated into adipocytes, chondrocytes and osteocytes based on staining characteristics and relative gene expression. Protocols published for other species were used to differentiate both Ad-MSC and BM-MSC towards a hepatocyte phenotype. Although a dramatic change in morphology and a reduction in vimentin gene expression were noted, suggesting a loss of mesenchymal phenotype, these protocols did not induce a hepatocyte phenotype. Pre-treatment with 5-Aza-2′-deoxycytidine to cause DNA demethylation and valproic acid to inhibit histone deacetylation also failed to allow transdifferentiation. A polycistronic vector containing Oct-4, c-Myc, Sox2 and Klf4 was successfully transfected into canine epidermal keratinocyte progenitor cells which became alkaline phosphatase positive and assumed a morphology consistent with iPSC. After colony selection and expansion, PCR evidence of plasmid presence was lost, colony morphology changed, and alkaline phosphatase activity reduced, consistent with vector expression factor and pluripotency loss. Canine iPSCs produced by lentiviral method were then differentiated towards hepatocyte phenotype using a published protocol for mouse and human iPSC. These cells were then assessed for hepatocyte characteristics using the developed reagents and primers. These cells demonstrated increased gene expression and morphology consistent with differentiation towards a hepatocyte-like phenotype. This thesis demonstrates successful culture of canine primary hepatocytes and validation of tests of hepatocyte phenotype. This provides a basis for optimising primary hepatocyte function in vitro and assessment of the success of differentiation protocols on stem cell sources. Canine mesenchymal stromal cells do not appear to transdifferentiate towards a hepatocyte-like phenotype using published protocols for other species. Canine iPSC are a promising candidate for an in-vitro source of hepatocyte-like cells.