Skip to main content

Elucidating Pathways of Steatohepatitis

Periodic Reporting for period 3 - EPoS (Elucidating Pathways of Steatohepatitis)

Reporting period: 2018-05-01 to 2019-10-31

Non-alcoholic fatty liver disease (NAFLD) is strongly associated with the Metabolic Syndrome including obesity, Type 2 Diabetes Mellitus and dyslipidaemia. Its prevalence has increased dramatically in recent years. NAFLD is characterized by substantial inter-patient variability in terms of severity and rate of progression. However, what determines the progression to NASH and beyond is not clear and these groups can only be differentiated by liver biopsy, a costly and potentially risky invasive procedure. Key challenges are to understand the biological and environmental factors that drive inter-patient variability within the NAFLD spectrum and to use this understanding to develop robust methods for diagnosis, risk stratification and therapy so that effective medical care may be targeted to those at greatest risk.

The specific objectives of EPoS are to:
i. generate high quality data defining the pathophysiology of NAFLD using a multi-‘omics’ approach
ii. develop a multi-dimensional pathophysiological profile across the spectrum of NAFLD using a systems medicine platform
iii. validate these findings and identify translatable mechanisms
iv. validate findings against clinical outcomes
v. measure health trends & clinical practice determinants

The ultimate goal is to develop a global understanding of how host and environmental factors interact at the cellular, organ and organism level to promote the development of NAFLD and, importantly, the progression to NASH and end-stage liver disease; and so inform the cost effective diagnosis, prevention and treatment strategies in Europe.
The ‘EPoS European NAFLD Registry’ database and Biobank have been generated with both historic datasets and prospectively recruited patients with longitudinal follow-up to determine natural history and disease outcomes.

Our completed GWAS is the largest worldwide to date on histologically characterised NAFLD patients that addresses the full disease spectrum and is the first reporting the four gene combination of PNPLA3, TM6SF2, GCKR and HSD17B13 as NAFLD risk modifiers. In the case of our transcriptomic analysis by RNA sequencing, a set of 25 genes have been shown to change in expression as NAFLD progresses and our replication analysis has confirmed these changes for 84% of the gene set.

Multi-omics profiling has been performed, including shotgun metagenomics (gut microbiome), genetics, RNA sequencing (liver biopsies), and metabolomics/lipidomics (serum). Integrative analyses included generation of partial correlation network, integrating multi-omics variables together with relevant clinical variables and NAFLD scores. This analysis identified the variables which directly associate with NAFLD scores, and which allowed the identification of multi-omic biomarkers of NAFLD. Additionally, genome-scale metabolic modelling was applied, which identified the key metabolic pathways associated with specific NAFLD groups and also predicted which metabolites were produced or taken up by the liver. These metabolites were verified in patient-matched serum samples.

The results of the gut microbiome analyses highlighted signatures of the gut microbiota tied to liver pathology and fibrosis as well as elucidating the interplay with common co-morbidities.

In subjects with NAFLD, both hepatic insulin resistance and adipose tissue insulin resistance increase with BMI and the degree of liver fibrosis, while muscle insulin sensitivity (OGIS index) is decreased. Moreover, insulin secretion is altered in NAFLD, i.e. increased with both BMI and degree of liver fibrosis, while hepatic insulin clearance is decreased more in subjects with NASH and moderate to severe fibrosis (F2-F4).

The CONSTANS study has been concluded which has provided novel initial insight into healthcare trends and burden within outpatient hepatology clinics throughout western Europe.

Our research on animal models has provided evidence of the relevance of: a) MBOAT7 mediating the remodelling of phosphatidylinositols for liver disease and, more specifically, its effect on inflammatory responses; and b) PEPD, a gene involved in the balance between degradation/remodelling of collagen and its contribution to fibro-inflammation, both important stages in the natural evolution from NAFL to NASH. We have shown that specific dietary interventions leading to fatty liver in murine models, differentially affect the overall phenotype and the severity of liver damage caused by the primary genetic manipulation of MBOAT7 and PEPD genetic models.

Dissemination activities have been excellent, with 66 peer-reviewed journal papers published so far and numerous others in preparation or planned. In addition, 150 conference presentations have been given as well as 38 conference posters.
Data obtained from the GWAS and transcriptomics analysis includes entirely novel observations but also strengthens previously reported findings from both the EPoS team and others. These findings will assist in development of a risk score for NAFLD development. For the transcriptomics, detection of significantly increased expression of certain genes during progression from early fibrosis raises the possibility that this could be the basis of a serum test for advanced disease.

We are currently developing algorithms to identify severe fibrosis in the liver by combining biomarkers of fibrogenesis and of local inflammation with markers of lipolysis and of pancreatic secretion. These algorithms will be validated in different cohorts of patients with NAFLD/NASH.

We have developed a new index, the GSG index (glutamate/(serine+glycine)) that was increased in association with hepatic insulin resistance and liver enzymes (in particular GGT). The GSG-index was able to discriminate subjects with F3-4 vs F0-F2 fibrosis. The adipose tissue insulin resistance and the GSG index have the potential to be used in a large cohort of patients to evaluate metabolic mechanisms associated with NAFLD and its severity, and also to monitor metabolic changes after intervention. The finding of phospholipid metabolites associated with gut microbiota richness and elevated liver enzymes has provided potential metabolite candidate biomarkers.

It has also become clear that it is important to subject NAFLD patients to a glucose tolerance test, not only to evaluate the presence of diabetes/pre-diabetes, but also because the degree of insulin resistance is associated with severity of fibrosis. This will allow early lifestyle intervention in these patients, which has been shown to have positive effects to decrease both NAFLD and glucose intolerance.

We have developed unique research tools (e.g. CRISPR cell lines or bespoke animal models) to study disease biology and evaluate novel therapies that limit lipid accumulation or ROS-induced hepatic injury. We have also found that markers of ECM are related to fibrosis stage in NAFLD patients as well as being correlated to insulin concentration. Both findings are novel.

Our work on murine models will improve the clinical practices of patients carrying SNPs for those aforementioned genes, recently reported in the literature to increase the risk of developing liver disease (MBOAT7), or Metabolic Syndrome (PEPD), by providing new knowledge regarding the pathophysiology associated with the gene dysfunction, and also showing how diet is an important covariate/modifying factor of observed phenotypes.
The Spectrum of NAFLD