Genome sequencing initiated to investigate how chronic liver disease leads on to hepatocellular carcinoma (HCC) has instead uncovered mutations that impact fat metabolism and reduce the sensitivity of hepatocytes to insulin.

Three of these mutations are in genes that are known to play a role in the disordered lipid metabolism seen in fatty liver disease and alcohol-related cirrhosis. These mutations were distributed across the liver and were seen in alcohol-related liver disease (ARLD) and nonalcoholic fatty liver disease (NAFLD), but have rarely been reported in HCC.

Cells that have mutations in these genes do not react to insulin signaling and do not take up fats. That allows them to escape the damage caused by storing excess fat and enables the mutated cells to survive and grow. Although a benefit to individual liver cells, it is suggested this impairs the function of the liver as whole.

It appears that in adapting hepatocytes to survive the effects of intracellular fat accumulation forced on them by ARLD and NAFLD, these mutations are acting as tumor suppressors, said Matthew Hoare, clinician scientist at the Cancer Research UK Cambridge Institute.

"Interestingly, none of the mutations in metabolism genes were linked to the development of liver cancer," said Hoare. That is possibly because hepatocytes carrying the mutations cannot meet the energetic requirements of fast multiplying cancer cells. "This information may prove useful in understanding the changes experienced by liver cancer as it evolves from a background of chronic liver disease," he said.

The patterns of mutations observed by the researchers offer potential new biomarkers of chronic liver disease. That could lead to earlier diagnoses of illnesses which cause few symptoms until the pathology is well advanced. The mutations also could be used to characterize subtypes of disease.

"Liver disease is incredibly interesting because if you pick it up early enough, you can stop it," Hoare said. "We need to think of better ways to identify who has problems and is going to progress. The tests we've got at the moment are incredibly rubbish," he told BioWorld Science.

The research also presents new drug targets. "What the data tell you is when the liver is struggling with liver disease, these mutations are helpful, so they all become plausible drug targets in ARLD and NAFLD," said Hoare.

This is the first time mutations have been identified that have an influence on metabolism and insulin sensitivity in patients with chronic liver diseases. "Mutations in liver cells have not previously been suspected of contributing to the biology of conditions such as obesity and type 2 diabetes," said Peter Campbell, head of cancer, aging and somatic mutations at the Sanger Institute, Cambridge University.

"We started this study hoping to understand how liver cancer emerges from chronic liver disease," Campbell said. Instead, in a paper published in Nature on October 13, the researchers are proposing a new model of chronic liver disease, in which the same mutation is acquired many times independently within the liver, and these mutations collectively occur in a considerable fraction of liver cells.

"The mutations might protect the liver cells from toxicity, but only by allowing these cells to shirk their metabolic duties," Campbell said.

The researchers analyzed somatic mutations from 1,590 genomes across 34 liver samples. These included healthy controls and ARLD and NAFLD patients who had liver transplants.

Seven of 29 patients with liver disease had mutations in FOX-01 (forkhead box protein-01), the major transcription factor in insulin signaling. The mutations were clustered in a single spot within the gene, impairing the insulin-mediated nuclear export of FOX-01.

In 6 of the 7 patients, mutations occurring in the FOX-01 hotspot showed convergent evolution, with variants acquired independently by up to nine distinct hepatocytes per patient. That implies highly specific selection pressures are operating in the liver of a given patient.

The researchers showed in vitro that by preventing nuclear export of FOX-01, the hotspot mutations lead on to insulin resistance, the upregulation of lipid catabolism genes and impaired glucose metabolism.

There also was a significant excess of mutations in the CIDEB gene (cell death-inducing DFFA-like effector b), which regulates lipid droplet metabolism, and GPAM (glycerol-3-phosphate acyltransferase 1, mitochondrial), which processes free fatty acids into storage triacylglycerol.

Knockout mouse models of these two genes are protected against lipotoxicity when fed high-fat diets, providing in vivo evidence for the hypothesis that somatic mutations in CIDEB and GPAM protect hepatocytes from lipotoxicity.

As with FOX-01, there was frequent convergent evolution, with up to 14 independent clones per patient with CIDEB mutations and up to 7 clones per patient with GPAM mutations.

In some patients the mutations collectively impacted up to 15-25% of the entire liver, which the researchers believe could lead to organ-wide changes in liver function.

Further studies are needed in larger cohorts to understand if the patient-to-patient heterogeneity seen in the samples results from different subtypes of disease, if it impacts on disease severity and whether it is predictive of risk of liver cancer, Hoare said. "This is more sequencing [in liver disease] than has been done in the world, but it is still only 34 people."