Fructose and Follistatin Drive Rapid Liver Disease When Hepatic Insulin Signaling Fails
New mouse study reveals fructose triggers acute fatty liver and fibrosis during complete hepatic insulin resistance, via a follistatin-adipose lipolysis axis.
Summary
Researchers challenged the dogma that insulin-driven lipogenesis is required for fatty liver disease (MASLD). Using mice with complete hepatic insulin resistance (LDKO mice lacking liver IRS1/2), they found that a fructose-enriched diet caused rapid, severe hepatic steatosis and fibrosis—conditions not seen on high-fat diets alone. Isotope tracing showed fructose fueled re-esterification of fatty acids mobilized from adipose tissue, not de novo synthesis. The liver-secreted protein follistatin (Fst) was the key driver: it promoted adipose insulin resistance and uncontrolled lipolysis, flooding the liver with free fatty acids. Blocking Fst prevented steatosis; over-expressing it accelerated disease. In humans, higher serum FST correlated with greater adipose insulin resistance and liver fat accumulation, supporting clinical relevance.
Detailed Summary
For decades, the prevailing model held that insulin-stimulated hepatic de novo lipogenesis (DNL) is necessary for MASLD development. Mouse models of complete hepatic insulin resistance—such as LDKO mice lacking both IRS1 and IRS2 in liver—become diabetic but do not develop fatty liver on high-fat diets, seemingly confirming this view. This study directly challenges that paradigm by demonstrating that dietary fructose unlocks a distinct, insulin-independent pathway to hepatic steatosis and steatohepatitis.
The researchers fed LDKO mice a GAN diet (high-fat/fructose/cholesterol, modeling human MASH) or a high-fructose diet (HFruD 60%). Within ten weeks, LDKO mice developed severe hepatic steatosis, advanced fibrosis (confirmed by Masson trichrome and Sirius Red staining), elevated galectin-3 (inflammatory marker), lipemic serum, and dramatically shortened lifespan—findings absent in control mice or LDKO mice on high-fat diets without fructose. The GAN-fed LDKO mice also had smaller adipose depots and lower body mass, yet markedly enlarged, fat-laden livers.
To dissect the mechanism, the team used [U-13C]fructose gavage and isotope tracing. Excess 13C label appeared in the glycerol backbone of hepatic triacylglycerides but not in fatty acid chains, demonstrating that fructose provided the glycerol-3-phosphate backbone for re-esterification of circulating free fatty acids (FFAs)—not substrate for new fatty acid synthesis. This was confirmed by unchanged or reduced hepatic DNL markers (FASN, ACC, SREBP1c) and by 3H2O labeling experiments.
The source of the excess circulating FFAs was adipose tissue. Complete hepatic IR in LDKO mice drives nuclear FoxO1 activation, which upregulates hepatic follistatin (Fst) secretion. Circulating Fst inhibits activin/myostatin signaling in adipose, impairing adipose insulin sensitivity and promoting uncontrolled lipolysis. Genetic inactivation of hepatic Fst in LDKO mice normalized adipose lipolysis, serum FFAs, and hepatic TAG accumulation, fully preventing GAN-induced MASLD. Conversely, AAV-mediated hepatic overexpression of Fst in wild-type mice accelerated GAN-promoted steatosis and MASH features. Fst-driven lipolysis was further confirmed using activin receptor (ActRIIB-Fc) blockade experiments and adipose-specific insulin receptor knockout crosses.
Translationally, the authors examined participants in the Tübingen Diabetes Family Study. Higher serum FST levels clustered with greater adipose insulin resistance (assessed by oral glucose tolerance adipose IR index) and greater hepatic triacylglyceride content measured by MR spectroscopy, independent of BMI. This human data supports the concept that a liver-Fst→adipose IR→hepatic FFA re-esterification axis, potentiated by dietary fructose, operates in people predisposed to metabolic disease—and suggests FST may be a clinically actionable biomarker or therapeutic target.
Key Findings
- Fructose, not high-fat feeding alone, triggers acute MASLD and advanced fibrosis in mice with complete hepatic insulin resistance.
- Fructose drives hepatic TAG accumulation via glycerol backbone re-esterification of circulating FFAs, not de novo lipogenesis.
- Hepatic follistatin secretion mediates adipose insulin resistance and uncontrolled lipolysis, supplying the FFA flood to the liver.
- Genetic deletion of hepatic Fst prevented fructose-induced MASLD; hepatic Fst overexpression accelerated steatohepatitis in wild-type mice.
- In humans, higher serum FST independently associates with greater adipose insulin resistance and hepatic lipid accumulation.
Methodology
Male LDKO mice (liver-specific Irs1/Irs2 double knockout) and floxed controls were fed GAN or high-fructose diets; hepatic lipid pathways were mapped with [U-13C]fructose isotope tracing and 3H2O DNL assays. Genetic manipulation of follistatin was performed via liver-specific knockout and AAV-mediated overexpression, with human validation in the Tübingen Diabetes Family Study using MR spectroscopy for hepatic fat and oral glucose tolerance-based adipose IR indices.
Study Limitations
The primary mechanistic work used male mice only, limiting sex-generalizability. LDKO mice represent complete hepatic insulin resistance, a more extreme state than typical human MASLD, potentially overstating effect sizes. The human association data are cross-sectional and cannot establish causality between FST levels and liver disease progression.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: