Hepatocellular carcinoma (HCC) is the second most common cause of cancer-related mortality worldwide. Despite the extensive multifactorial etiology of HCC, there is growing evidence that disrupted metabolic processes in the liver are the predominant cause of HCC, contributing also to the development of metabolic fatty liver disease (MAFLD), previously termed non-alcoholic fatty liver disease (NAFLD). Maintaining cholesterol balance is one of the most important factors of metabolism, but the role of cholesterol in hepatocarcinogenesis still remains unexplained. To deepen the understanding of the molecular mechanisms that regulate the metabolic-dependent development of HCC, it is still necessary to use animal models since it is impossible, due to ethical and other reasons, to answer all questions in humans. One important aspect is also following the disease progression during the aging that would require following human subjects for several decades. To address the changes in the long term liver pathology related to disballanced cholesterol synthesis we monitored a mouse model (Cyp51 KO) of aging mice (12, 18 and 24 months). The mice represent a targeted knock-out of the lanosterol 14α-demethylase (CYP51) gene from the late part of cholesterol synthesis. Aging Cyp51 KO mice developed liver tumours between 12 and 24 months of age with an incidence of 77.8 % in females and 50 % in males at 24 months. We described that liver tissue damage was more prominent in Cyp51 KO females, where a moderate to severe ductular reaction, accompanied by mild inflammation and severe fibrosis were observed at 12 months of age, when the liver phenotype worsened until the age of 24 months. The influence of chronic metabolic disease caused by deletion of Cyp51 in hepatocytes was reflected in a number of mechanistic adjustments of the liver, which show a significant difference between sexes. All metabolic and transcriptional changes caused by disrupted cholesterol metabolism, which were evaluated in diagnostically confirmed HCC of Cyp51 KO mice and aligned with the latest human literature data, opened horizons for proposing a new sex-dependent mechanistic model for the hepatocarcinogenesis development in Cyp51 KO mice. In both sexes of Cyp51 KO mice, the PI3K/AKT signalling pathway and the ECM-receptor interaction pathway were activated by sex-specific altered genes and significant reduction of lipid-related metabolic processes was present. Elevated plasma cholesterol and cholesterol HDL, inhibited transcription factors FXRα and LXRα:RXRα, and most importantly, the crosstalk between inhibited LXRα and activated TGF-β signalling pathway explain the increased susceptibility to hepatocarcinogenesis in Cyp51 KO females. Additionally, transcription factors (SOX9)2 and PPARα were identified as important targets for female hepatocarcinogenesis, while the elevated expression of Cd36, a target gene of transcription factor RORC, could represent an important sex-dependent regulator of the ECM-receptor interaction signalling pathway in male hepatocarcinogenesis.
Uncovered metabolic reprogramming of cholesterol-dependent pathways represents an important concept of understanding hepatocarcinogenesis in women, where the prevalence of HCC is higher but still lower in comparison to men. We used the comparative functional genomic analysis, as initial results indicated that cholesterol synthesis was reduced also in the evaluated publically available human HCC data sets, included in the analysis. While the inhibition of CYP51 expression has not yet been proven to be an important regulator of metabolically dependent hepatocarcinogenesis in women, our data indicate that the gene is under-expressed in some tumour samples from women with HCC.
To dechipher, which cells of the liver are responsible for the pathological developments upon cholesterol disbalance, we developed another mouse model Cyp51LC. In comparison to the Cyp51 KO model, the Cyp51LC model allows not only the targeted gene deletion in hepatocytes, but also deletion in a time-controlled manner. In our case the doxycycline responsive promotor has been introduced that allows excision of the Cyp51 transgene upon application of the doxycycline at a certain time point, for a certain time period. With such approach we avoid the phenotype changes caused by gene deletion during animal development. Consequently, we can observe only the response of the organ/organism to the change in the adult animal, while still monitoring early phases of the disease.
After application of doxycycline to mice major changes in the ultrastructure of hepatocytes have been observed, indicating extensive damage of cell organelles and, surprisingly, numerous crystals of various shapes whose content is still under investigation. They were more abundant in Kupffer cells compared to hepatocytes and likely include large amount of cholesterol synthesis intermediates. The liver inflammation and ductular reaction were in Cyp51LC livers weaker compared to Cyp51 KO model and fibrosis in liver parenchyma was not observed. The absence of fibrosis in the Cyp51LC mouse model is a large benefit since it allows primary hepatocyte isolation due to les overall liver tissue damage. Primary hepatocytes with depleted cholesterol synthesis showed an increase in endoplasmic reticulum stress and strongly activated lipid droplet degradation. The elevated concentrations of sterol intermedates lanosterol and dehydrolanosterol were found in hepatocytes, this is why we propose that they form crystals. The primary hepatocyte data is important to compare with the obtained on whole liver, since we can now evaluate which changes belong to hepatocyte and which to other cell types of the liver. At the level of sterol intermediates, where other sterol intermediates were also detected, this will be helpful to unravel the composition of crystals found in different liver cell types with depleted cholesterol synthesis. The primary hepatocytes of the Cyp51LC mouse model thus represent an important cellular model for in-depth mechanistic studies of cholesterol-dependent HCC.
|
