Cellular senescence prevents deoxyribonucleic acid (DNA) damaged cells undergoing proliferation, protecting against malignancy but at the expense of senescent cell accumulation. Since the proportion of senescent hepatocytes in man correlates closely with liver injury, fibrosis stage and mortality, prevention of malignancy occurs at the expense of impaired organ function. Senescent cells are removed by the innate immune system, including natural killer (NK) cells, during normal embryonic development and healthy wound healing. Since eradication of senescent cells in mice leads to resistance against age-related disorders and functional restoration, one approach to liver injury might be enhanced removal of senescent cells. The non-classical human leukocyte antigen class I molecules E, F and G (HLA-E, -F and -G), are ligands for NK cells and considered tolerogenic. For example, HLA-E binds with the NK inhibitory receptors, killer cell lectin like receptor D1 (CD94) and c-type lectin family receptors (NKG2), preventing NK mediated killing. Less is known of HLA-F, which is present in placental trophoblasts, possibly promoting foetal tolerance. HLA-G is expressed on trophoblast cells and is considered a key component mediating foetal tolerance. Little is known of HLA-E or -F in liver injury, while a few immunohistochemical studies have suggested increased hepatocyte expression of HLA-G in cirrhosis of all aetiologies. My hypothesis was that senescent hepatocytes are removed from healthy liver by the innate immune system, as in healthy wound healing, but that in chronic liver disease there is a failure of this response, leading to an accumulation of senescent cells, fibrosis and cirrhosis. I examined hepatic expression of HLA class I, HLA-E, -F and -G in health and disease. Secondly, I assessed the major histocompatibility complex (MHC) class I ligands, the NK cell receptors. Finally, the various components of the antigen presentation pathway in health and disease were studied. My study shows that HLA class I is present in liver tissue and is upregulated in cirrhosis. HLA-E and -F are expressed in liver tissue and are also upregulated in cirrhosis. The expression of HLA-E and -F was largely intracellular and consistent with retention within the golgi, however, some less marked surface expression was also seen. The data on expression of HLA-G was clear, but inconsistent with the literature. While immunohistochemistry work was indeed congruent with published data showing widespread expression in cirrhosis, all other biochemical techniques, undertaken herein for the first time, failed to detect HLA-G protein in healthy liver tissue or cirrhosis; HLA-G RNA was only identified in hepatitis B infected liver tissue. These data suggest that previous reports linking HLA-G and liver injury may be artefactual. The NK cell receptors, killer cell immunoglobulin like receptors 2DL2 and 2DS2, were significantly upregulated in cirrhosis although the numbers studied were low suggesting that NK cell activation is altered in cirrhosis. Tapasin and tapasin related protein (TAPBPR), both parts of the antigen presentation pathway are expressed in liver tissue and show a slight upregulation in liver disease. This suggests that aberrant expression via this pathway may be part of the liver pathology. In summary there is altered expression of HLA molecules in cirrhosis, specifically HLA-E and -F. Secondly, altered expression of inhibitory molecules is seen on circulating NK cells. These findings are consistent with reduced clearance of damaged hepatocytes in cirrhosis. It proved impossible, despite numerous approaches, to link increased hepatocyte HLA expression with senescence. My study concludes that HLA class I is present in liver tissue and is upregulated in cirrhosis. The expression of HLA-E and -F is identified in liver tissue for the first time and is shown to be upregulated in cirrhosis. This altered expression of HLA-E and -F may have a role in the pathogenesis of cirrhosis and could be a target for future therapies.