Exploring epigenetic modifications in fatty liver with berberine

Last year, I described a study that found an epigenetic mechanism of peripheral insulin resistance through methylation of the PGC-1alpha promoter.  Recently, Chang et al. explored an epigenetic mechanism in non-alcoholic fatty liver.

Fatty liver is closely linked with insulin resistance and diabetes as well as obesity.  As it is not yet fully understood, the only effective treatment is lifestyle intervention (exercise and weight loss).  However, one compound that has potential, Berberine, is a natural alkaloid that is sold as a dietary supplement in the U.S.  It has been studied in animal and human models in which it ameliorates metabolic problems like dyslipidemia and insulin insensitivity.  The authors also note that it has been found to reduce serum cholesterol and LDL-cholesterol by increasing hepatic LDL-receptor expression.  I have written previously a little about some of these biomarkers- I don’t think (in general) they are good representations for disease, at least yet.  Berberine does reduce fatty liver in some genetically altered mice, but these do not represent a normal milieu.  If berberine can improve the aforementioned dysfunctions and fatty liver in a normal mouse model, it has potential to go on to human trials and possibly be used therapeutically, or its structure used to develop more potent drugs for fatty liver.

So the authors tested this, and elucidated an epigenetic mechanism in which it develops.  A short summary is available at the end. The study was done in Sprague-Dawley rats.  Split into 3 groups, one group consumed a high fat diet to induce fatty liver and received 200 mg berberine per kg/day (BSA HED ~32 mg/kg per day if I calculated correctly), one group consumed a high fat diet and received a placebo, and a “normal” diet group received the placebo.

Genes involved in lipid metabolism were chosen after real-time PCR analysis of liver mRNA expression- genes differentially expressed between the high fat group taking berberine and the high fat group taking placebo were chosen for DNA methylation analysis.

The researchers measured body weight, visceral fat, and lipids for 16 weeks in the groups.  On a high fat diet, rats tended toward obesity as expected, but in the high fat diet with berberine treatment, bodyweight was reduced to levels similar of non high fat diet rats.  It should be noted that this does NOT mean berberine will do this in humans as some supplement companies will claim- many compounds reduce bodyweight in rodents but fail in human trials.  Visceral fat, serum total cholesterol, and LDL-cholesterol, but not serum triglycerides were also lowered in the berberine group, though these were not affected until toward the end of the treatment.

Liver analysis showed that berberine prevents derangement of liver cells, excessive lipid droplets in the cells, and decreased the total liver weight of the liver compared to the placebo group.  Hepatic triglycerides, and serum ALT and AST were also reduced.  These suggest that liver steatosis is improved with berberine.

Fasting glucose and insulin were reduced in the berberine group, as well as AUCs after a glucose challenge.  These suggest that insulin resistance is ameliorated with berberine treatment.

Next came genetic analysis.  mRNA and protein of the following genes were found to be down-regulated in the livers of rats consuming a high fat diet (without berberine) compared to a non high fat diet: CPT-1alpha, MTTP, LDLR.  Berberine prevented the down-regulation of CPT-1alpha and MTTP induced the the high fat diet.  SCD-1 mRNA was lower in the berberine treatment.  PPARgamma and GPAT were not affected by berberine, nor ACC, DGAT1, DGAT2, PPARalpha, and UCP-2, all of which are involved in fat storage and oxidation.  So berberine seems pretty selective in its targets.

So next they chose to examine if these gene expressions are altered by DNA methylation.  They looked at the promoters of MTTP, CPT-1alpha, and LDLR.  They found that the MTTP promoter was methylated differently in each of the test groups, while CPT-1alpha and LDLR was not.

So they focused on the MTTP promoter, finding that on average it was methylated more in the high fat diet rats compared to the non high fat diet rats (38% vs 12%), as well as more at each CpG site that was individually analyzed.  Berberine lowered the methylation level of the promoter with a high fat diet down to about 17% on average, as well as at individual CpG sites. An inverse correlation between MTTP mRNA expression and methylation of 2 of the 3 CpG sites was found.  Together, this suggests that berberine influences gene expressions that influence lipid metabolism (MTTP for sure) by altering the methylation of its (MTTP) promoter, which prevents fatty liver.

To further verify this, they performed subsequent in vitro experiments with berberine against a demethylating agent, which corroborated the theory.  If you read my previous post about the epigenetic modifications of the PGC-1alpha promoter, you may recall that certain metabolic factors influenced the activity of a DNA methyltransferase isoform which mediated the effect of these variables on the methylation of the promoter.  In this study, that DNA methyltransferases weren’t affected, but serum homocysteine was higher in the high fat diet group compared to the non-high fat diet group, and berberine restored this to normal levels, but other molecules in this pathway (SAM, SAH, SAM/SAH ratio) were not different.  This suggests that berberine prevents methylation by altering the methyl donor pool, or by altering DNA demethylases, or other mechanisms that need to be further explored.

Lastly, since MTTP is needed for VLDL secretion and triglyceride accumulation in VLDL fractions, they found that berberine indeed increased the triglyceride content (as well as higher apoB-100 and -48), compared to a high fat diet which reduced it.

It should be noted that in the previous study on PGC-1alpha that I described, many of the epigenetic changes were on non-CpG sites (which was a novel finding), while this appeared to only examine CpG methylation.  I don’t know why this is.

Their interpretations

The authors describe the functions of MTTP (microsomal triglyceride transfer protein), which include assembling and secreting VLDL and LDL which export lipids from the liver; down-regulation of its expression in other research increases liver dysfunction.  The results of this study match what would be expected if MTTP had a major role in liver lipid homeostasis.  However, other genes come into play that were not affected by methylation- LDLR, which regulates LDL-cholesterol levels, was affected by berberine in a non-methylation fashion.  CPT-1alpha expression, key in mitochondrial beta-oxidation, and SCD-1, which converts saturated fats to monounsaturated fats were also altered through non-methylation mechanisms by berberine.  These alterations could be from other epigenetic mechanisms, e.g. histone modifications or by miRNA.

Quick Summary

Berberine prevents fatty liver in rodents on a high fat diet in one way by reducing the methylation of the MTTP promoter, which restores triglyceride accumulation in VLDL and export from liver cells.  It also affects several other gene expressions through unknown mechanisms, possibly epigenetic, that have not been explored yet.

No practical relevance yet, but interesting to those who like this area.  Elucidating these specific pathways will allow us to find precisely how diet, exercise, and environmental factors interact to influence metabolic functions, and model them for manipulation and medical purposes.


Chang X, Yan H, Fei J, Jiang M, Zhu H, Lu D, & Gao X (2010). Berberine reduces the methylation of the MTTP promoter and alleviates fatty liver induced by a high-fat diet in rats. Journal of lipid research PMID: 20567026

  • Jeffborsato

    Some interesting findings with regards to rats fed a high fat diet, but the PUB MED entry doesn't provide a free copy of the full study, I am concerned that the HFD fed to the rats could have considerable effects on the promotion of fatty liver. I would be curious to find out what type of HFD was fed to the rats and what portion of their total caloric intake was said fat. Rats fed a diet of largely Omega 6 fats may have induced fatty liver at a different rate than one composed of saturated and/or Omega 3 fats. In addition was the feeding controlled across both groups to ensure caloric consumption was the same or were they permitted to unlimited feeding of either diet.High Fat diet seems to be too general of a term to be used even in a study's abstract with what we know about the varying effects of different types of fats on both humans and rodents. I would be interested in knowing if berberine's effects were different or mitigated in any way if the HFD rodents were splint into sub groups (ie: primarily veg oil vs. lard). J

  • http://www.recomp.com Colby

    Hey Jeff,Thanks for the comment and sorry for the delay. The PDF of this paper is actually free online here: http://www.jlr.org/cgi/rapidpdf/jlr.M001958v1I probably should note in posts like these from now on that high fat diets are commonly used to cause metabolic dysfunctions and obesity in rodents, but this does NOT mean it will happen in humans- we do not observe the same effects in human trials. I am only concerned with the mechanistic explorations in this study, so composition of the diet doesn't really matter- the goal was to induce fatty livers so they could observe how berberine was preventing it. They were in fact fed ad libitum to encourage metabolic perturbations, and the high fat diets were mostly of lard which does have a high saturated fat and I believe omega 6 content. The authors state clearly that this provided mechanistic information about the pathogenesis of liver dysfunction with overnutrition- not a high fat diet. This is NOT support against a high fat diet- only a tool to study the effects of high calorie diets on the liver because rodents (generally) have problems tolerating it.I'm sure other variables would come into play when testing the efficacy of berberine- i'm sure human trials will address this down the line if the research continues.

  • Xig9

     Nice information. I thought you may also mention this key study in the molecular mechanism of  berberine on LDLR. Kong Weijia, et al., Berberine is a novel cholesterol-lowering drug working through a unique mechanism distinct from statins, Nature Medicine 2004; 10(12): 1344-1351.