NEW YORK (GenomeWeb News) – Through an exome-wide association study, researchers led by Jonathan Cohen, a professor at the University of Texas Southwestern Medical Center in Dallas, uncovered three variants linked to nonalcoholic fatty liver disease.
As the researchers reported in Nature Genetics yesterday, two of those variants were located in a gene previously linked to disease while one was in the TM6SF2 gene, whose function they traced to the secretion of very-low-density lipoprotein. The variant itself appeared to be associated with higher levels of alanine transaminase, which typically rises in response to liver injury, as well as with lower levels of low-density lipoprotein-cholesterol, triglycerides, and alkaline phosphatase, another marker of liver disease.
"[Our] data indicate that TM6SF2 activity is required for normal VLDL secretion and that impaired TM6SF2 function causally contributes to [nonalcoholic fatty liver disease]," Cohen and his colleagues said.
Nonalcoholic fatty liver disease, or NAFLD, comprises a range of conditions stemming from the accumulation of fat into the liver. The researchers noted that some 30 percent of adults have excess fat in their liver, stored as triglycerides, and that while the condition can be benign, it can also lead to chronic inflammation and cirrhosis. The US National Institute of Diabetes and Digestive and Kidney Diseases reports that NAFLD is becoming more common in the US, possibly due to the increasing number of people with obesity.
Cohen and his colleagues searched for variants associated with hepatic triglyceride content in 2,736 participants from the Dallas Heart Study, sifting through some 138,400 polymorphic sequence variants while adjusting for ancestry, age, body mass index, and gender.
From this, they identified three variants linked to hepatic triglyceride content — two variants in the PNPLA3, which they had previously found to be associated with hepatic triglyceride levels, and one variant in the TM6SF2 gene, whose function was unknown.
The TM6SF2 variant, an adenine-to-guanine substitution leading to lysine replacing glutamate at residue 167, was more common in people of European ancestry than in African Americans or Hispanics, the researchers said. They also noted that the non-substituted form, Glu167, is highly conserved among mammals. People with the variant, though, had elevated hepatic triglyceride content no matter their ancestry.
Additionally, the TM6SF2 variant wasn't linked to other hepatic steatosis risk factors like BMI, insulin resistance, or alcohol consumption. It was also independent from the PNPLA3 variants.
RT-PCR analysis of cDNA from a range of human tissues found that TM6SF2 is highly expressed in the small intestine, liver, and kidneys, though it is found at lower levels in other tissues as well. The TM6SF2 variant is also linked to an increase in serum ALT, a marker of liver injury.
The researchers also confirmed the link between the TM6SF2 variant and NAFLD by drawing two cohorts, one from the Dallas Biobank and one from Copenhagen, including from the Copenhagen City Heart Study and from the Copenhagen General Population Study, with more than 82,000 samples. In both cohorts, the TM6SF2 variant encoding p.Glu167Lys was connected with signifi¬cantly higher activity of ALT in serum.
"These findings further support the hypothesis that the TM6SF2 variant encoding p.Glu167Lys is associ¬ated with NAFLD and are consistent with the notion that the vari¬ant compromises hepatic integrity," Cohen and his colleagues noted.
Further, in all three cohorts, the TM6SF2 variant was linked to lower levels of triglycerides and LDL-C made in the liver.
By expressing the wild-type and variant forms of the protein in a hepatic cell line, the researchers found that wild-type and variant mRNA levels were comparable, but also that much less — 46 percent less — of the variant protein was expressed. This, the researchers said, suggests that the variant protein is misfolded and quickly degraded within the cell.
To determine how that affects hepatic triglyceride content, Cohen and his colleagues developed recombinant adeno-associated viral vectors that expressed short hairpin RNAs targeting TM6SF2 in mouse livers. Two different shRNAs aimed at TM6SF2 led to a more than 90 percent reduction in TM6SF2 mRNA in mouse livers, but not in other tissues, the researchers said.
Inhibiting TM6SF2 in mouse liver led to a three-fold increase of hepatic triglyceride content while also leading to lower plasma cholesterol levels — LDL, HDL, and the triglyceride content of VLDL were reduced, the researchers reported. ALT levels, though, were unchanged.
This, the researchers said, is consistent with a defect in VLDL secretion. To see just how knocking down TM6SF2 affects VLDL secretion, the researchers inhibited the enzyme — intravascular lipoprotein lipase — that breaks down the triglyceride con¬tent of VLDL and measured how quickly plasma triglycerides built up in plasma. In the knockdown mice, the rate of accumulation was much lower than in the control mice.
"These data indicate that TM6SF2 normally acts to promote VLDL secretion and suggest that the increased HTGC associated with the Glu167Lys TM6SF2 variant in humans results from a reduction in TM6SF2 function," Cohen and his colleagues said.
They noted that diets high in sucrose appear to intensify the effect that knocking down TM6SF2 had on hepatic triglyceride content.
Cohen and his colleagues further hypothesized that TM6SF2 might work in the intestine, where it is also expressed, to influence ALP activity. They said that they are currently exploring whether TM6SF2 is involved in lipoprotein synthesis and ALP activity in the intestine as well as the connection between lower ALP levels and increased hepatic triglyceride content linked to the TM6SF2 variant encoding p.Glu167Lys.