Resident hepatic macrophages are an essential component of the liver’s innate immune response to infectious and toxic-metabolic agents. On the basis of in vitro findings, it has been reported that macrophages assume pro-inflammatory, anti-microbial and anti-tumour roles following stimulation by interferon-γ with lipopolysaccharide or tumour necrosis factor-α (M1-macrophages). M2-macrophages have functional roles that are generally opposed to those of M1-macrophages, and may be induced by interleukin (IL)-4 and/or IL-13 (M2a-macrophages), immune complexes and Toll-like receptor or IL-1 receptor agonists (M2b-macrophages), or IL-10, transforming growth factor-β (TGF-β) or glucocorticoids (M2c-macrophages). The phenotype of hepatic macrophages in the setting of chronic liver disease remains to be clarified.
There are a number of important discrepancies between the reported expression profile of polarised murine macrophages and polarised human monocyte-derived macrophages, which are themselves an expensive model and limited by inter-donor variability. As such, there is a need to develop a convenient human model of polarised macrophages, in order to determine the phenotypic responses of human macrophages to a range of inflammatory and metabolic agents.
Circulating free fatty acids (FFAs) appear to be elevated in the setting of obesity and non-alcoholic fatty liver disease, and may contribute to the pathogenesis of chronic liver disease. Two of the most abundant FFAs in human serum are palmitate and oleate. The influence of these FFAs on aspects of macrophage biology has been studied individually; however, the combined effect of these FFAs (more relevant in vivo) on the phenotype of macrophages is unclear and warrants further investigation.
Finally, although hepatic macrophages may alter metabolic and other functions of hepatocytes, the extent to which such crosstalk may be influenced by the phenotype of resident macrophages remains to be determined.
The aims of this thesis were to develop and characterise an in vitro human model of polarised macrophages, and to determine the effect of elevated concentrations of oleate and palmitate on the phenotype of macrophages. In addition, this thesis aimed to characterise changes to hepatocyte gene expression that occur in response to secreted factors derived from polarised macrophages.
THP-1 monocytic leukaemia cells were differentiated and polarised toward the M1, M2a, M2b and M2c-macrophage phenotypes, and gene expression of a range of macrophage phenotypic markers was characterised by semiquantitative real-time polymerase chain reaction (qPCR). M1-polarised THP-1 macrophages and phorbol-12-myristate-13-acetate (PMA) differentiated (M2) THP-1 macrophages were cultured in the presence of oleate and palmitate in a 2:1 ratio, and their phenotypic expression profile was characterised by qPCR. Additionally, Huh7 cells, a human hepatoma cell line, were cultured in the presence of conditioned media (CM) derived from polarised THP-1 macrophages, and the expression of a number of phenotypic and functional markers was assessed by qPCR.
M1, M2a and M2c-polarised THP-1 macrophages displayed phenotypic profiles that were generally in accord with previous findings from a range of macrophage model systems. The expression profile of M2b-polarised THP-1 macrophages differed from previous reports; however, this phenotype has not previously been characterised in human macrophages.
It was not possible to distinguish between populations of M1, M2a, M2b and M2c-macrophages on the basis of a particular value for the normalised gene expression. However, by calculating ratios of the levels of expressed genes, it was possible to discriminate between populations of M1 and M2-polarised THP-1 macrophages. It was not possible to distinguish between populations of M2a, M2b and M2c-macrophages in this way, which emphasised the degree of overlap in the function and phenotype of the M2-macrophage subtypes.
Culture of M1-macrophages and M2 (PMA)-macrophages with FFAs was not associated with a significant change in expression of the phenotypic markers that were analysed. A significant decrease in a number of M1 phenotype-specific gene expression ratios was demonstrated for M1-polarised THP-1 macrophages exposed to elevated levels of FFAs. Therefore, while the cells did remain M1-polarised, this finding may suggest a shift in their orientation toward the M2 phenotype.
Culture of Huh7 cells with CM derived from M1, M2a, M2b and M2c-macrophages was associated with a significant increase in expression of lipocalin-2, an inflammation associated gene, and TGF-β1, a pro-fibrogenic cytokine. Additionally, culture of Huh7 cells with CM derived from the M2-macrophage subtypes, but not with CM derived from M1-macrophages, was associated with a change in Huh7 cellular morphology toward an elongated, spindle-shaped appearance. There was a significant increase in expression of vimentin, accompanied by a significant decrease in expression of E-cadherin, which is suggestive of a partial epithelial-mesenchymal transition. These results were in agreement with the morphological change that was observed, and suggest hepatocytes may not simply be “bystanders” in injury, but may actively participate in host response. Additionally, this response may be influenced by the phenotype of hepatic macrophages.
The work outlined in this thesis lays a foundation for future experiments that may seek to further explore the nature of macrophage-hepatocyte crosstalk using human material cultured ex vivo.