Drinking coffee burns hepatic fat by inducing lipophagy coupled with mitochondrial β-oxidation


  • Wen-Xing Ding Ph.D.

    Corresponding author
    1. Department of Pharmacology, Toxicology and Therapeutics, The University of Kansas Medical Center, Kansas City, Kansas
    • Address reprint requests to: Wen-Xing Ding, Ph.D., Department of Pharmacology, Toxicology and Therapeutics, The University of Kansas Medical Center, MS 1018, 3901 Rainbow Boulevard, Kansas City, KS 66160. E-mail: wxding@kumc.edu; fax: 913-588-7501.

    Search for more papers by this author

  • See Article on Page 1366

  • This work was supported by NIH R01 AA020518 (to W.X.D) and 5P20 RR021940-07 and 8 P20 GM103549-07 (a Pharmacology, Toxicology and Therapeutics Department COBRE grant).

  • Potential conflict of interest : Nothing to report.


acetyl-CoA carboxylase


protein kinase B


AMP-activated protein kinase


adenosine triphosphate


cyclic adenosine monophosphate


free fatty acids


fatty liver disease


lipid droplets


mammalian target of rapamycin complex 1


phosphorylated ACC


peroxisome proliferator-activated receptor gamma coactivator-1alpha


phosphoinositide 3-kinase


peroxisome proliferator-activated receptor alpha


transcriptional factor EB


transforming growth factor β


unc-51-like autophagy-activating kinase 1

Coffee is currently one of the most consumed and popular beverages in the world. According to the National Coffee Association, more than 60% of American adults drink coffee each day, and an average of 3.1 cups of coffee are consumed by each drinker per day. In spite of its popularity, it has been controversial whether drinking coffee is detrimental or beneficial for human health. Though a large body of evidence suggests that drinking coffee may be beneficial for a variety of chronic health conditions including type 2 diabetes, stroke, cancer, and all-cause mortality, some other studies suggest that drinking coffee may be a potential hazard for coronary heart disease and may increase mortality in younger drinkers under 55 years of age.[1, 2] The reasons for the conflicting results from these large, population-based studies could be very complex, but one possibility could be because of the remarkable variety of different types of coffee and the preparation and brewing methods around the world. In contrast to the controversy regarding the health effect of coffee on other organs and tissues, all the experimental and population-based studies support unanimous beneficial effects of drinking coffee on the liver.

The early evidence of the beneficial effects of coffee on the liver came from epidemiologic studies that revealed a strong association of drinking coffee with decreased serum hepatic enzymes, including gamma-glutamyl transferase, aspartate aminotransferase, and alanine aminotransferase, in persons with high risk of liver injury, such as alcoholic, diabetic, and viral infection.[3] Recent epidemiologic studies further support that drinking coffee also reduces the risk for fatty liver, fibrosis, and hepatocellular carcinoma.[4, 5]

Although epidemiological evidence strongly supports the beneficial effects of coffee on liver function, the molecular mechanisms for its actions are less understood. Part of the reasons may be because coffee contains a number of different contents, including caffeine, diterpenoid alcohols cafestol and kahweol, and other antioxidant substances, such as chlorogenic acid and tocopherols. Coffee may increase antioxidant activity to offer hepatoprotective action by directly activating Nrf2 (nuclear factor erythroid 2-related factor) transcription factor or indirectly increasing the expression of UDP glucuronosyltransferase in hepatocytes.[6] Caffeine, the major component of coffee, is metabolized mainly in the liver by cytochrome P450 1A2, which generates three metabolic dimethylxanthines, including paraxanthine (84%), theobromine (12%), and theophyline.[7] It is well known that methylxanthines increase intracellular cyclic adenosine monophosphate (cAMP) levels by inhibiting phosphodiesterase activity. Indeed, caffeine increased intracellular cAMP levels in hepatocytes. As a result, caffeine inhibited liver fibrosis by down-regulation of connective tissue growth factor, an important player for fibrosis mediated by transforming growth factor beta (TGF-β). Mechanistically, it was found that caffeine promoted proteasomal degradation of the TGF-β effector protein, mothers against decapentaplegic homolog 2.[7, 8] Furthermore, coffee might also reduce hepatic lipid accumulation by increasing fatty acid β-oxidation and reducing liver oxidative stress and inflammation, as suggested by a rat model of steatohepatitis.[9]

Autophagy is an intracellular degradation pathway that involves the formation of a double-membrane autophagosome, which enwraps and delivers cargo to lysosomes, where the contents are degraded. Autophagy is usually activated as a catabolic process when cells lack nutrients and energy. Autophagy was initially thought to be a bulk nonselective degradation pathway for degrading intracellular proteins and excess or damaged organelles. However, a pioneer work from Singh et al. demonstrates that autophagy can selectively degrade intracellular lipid droplets (LDs), a process termed lipophagy.[10] Since then, many follow-up studies including ours have demonstrated that pharmacologically modulating autophagy can attenuate both alcoholic and nonalcoholic steatosis in mouse livers.[11, 12] LDs are organelles enriched with triglycerides and cholesterol esters that are surrounded by a phospholipid monolayer as lipid stores for future use or to detoxify the otherwise toxic free fatty acids (FFAs). When more energy is needed or too much influx of lipids occurs, cells activate the lipolysis process mediated by intracellular lipases to generate FFAs. In addition to the hydrolases, such as proteases, glycases, and nucleases, the lysosome also contains acid lipases (low pH is required for their maximal enzymatic activity).[13] Currently, it is not clear how cytosolic lipases and lysosomes coordinately determine the amount and type of lipids to be degraded. At normal physiological conditions, it is thought that lysosome-mediated lipid degradation is mainly responsible for membranes of organelles or extracellular lipids that reach lysosomes from endocytosis. However, when cells are overloaded with lipids, lipophagy may be able to provide a large amount of FFAs within a short period by degrading cellular LDs.[13] It is well known that FFAs are toxic to cells, and thus too much FFAs generated from lipophagy/lipolysis may harm the cell unless FFAs are quickly cleared. Cells can utilize mitochondrial β-oxidation to burn FFAs to generate adenosine triphosphate (ATP) and, in turn, reduce their toxicities. Therefore, an ideal intervention for treating fatty liver disease (FLD) would be to enhance not only lipophagy, but also the use of FFAs.

In this issue of Hepatology, Sinha et al. report on a study demonstrating that caffeine may be just such an ideal intervention to protect against FLD by enhancing lipophagy and mitochondrial-β oxidation simultaneously.[14] By using a series of autophagic flux assays, they demonstrated that caffeine induces autophagic flux in human hepatoma cells, primary cultured hepatocytes, and mouse livers. They further found that caffeine-induced autophagosomes often contain LDs, suggesting the induction of lipophagy. Metabolomic analysis of hepatic acylcarnitines revealed an increase of hepatic lipolysis by caffeine treatment. More important, caffeine increased mitochondrial β-oxidation activity and inhibited hepatic steatosis in a high-fat-diet-fed mouse model. Interestingly, they found that small interfering RNA knockdown of Atg5, an essential autophagy gene required for biogenesis of autophagosomes, inhibited caffeine-induced mitochondrial β-oxidation and mitochondrial bioenergetics. Selective removal of damaged mitochondria by mitophagy may help to maintain better quality of mitochondria and thus enhance mitochondrial β-oxidation and mitochondrial bioenergetics. Unfortunately, potential changes of mitophagy after caffeine treatment were not addressed in the current study.

How does caffeine trigger autophagy in hepatocytes? Mammalian target of rapamycin complex 1 (mTORC1) and AMP-activated protein kinase (AMPK) are two important regulators for autophagy, which are also two key sensors in response to the changes of cellular nutrients and energy. mTORC1 negatively regulates autophagy by inhibiting unc-51-like autophagy-activating kinase 1 (ULK1) activity by directly phosphorylating ULK1.[15] In contrast, AMPK positively regulates autophagy by at least two mechanisms. AMPK suppresses mTORC1 activity by phosphorylation of tuberous sclerosis 2 and raptor, two essential regulators of mTORC1,[16, 17] and AMPK activates ULK1 complex by directly phosphorylating ULK1 at different sites from mTORC1.[15, 18]

Indeed, caffeine treatment decreased mTORC1 activity, both in cultured hepatocytes and in mouse liver.[14] The mechanism for how caffeine inhibited mTORC1 was not investigated in this study. Intriguingly, caffeine treatment increased intracellular nucleotides, such as nicotinamide adenine dinucleotide, ATP, AMP, and adenosine diphosphate levels, possibly as a result of increased lipolysis and mitochondrial β-oxidation. Increased ATP levels normally result in inhibition of AMPK. Although the direct phosphorylation levels of AMPK were not determined, Sinha et al. found that caffeine treatment increased the ratio of phosphorylated acetyl-CoA carboxylase (p-ACC) versus total ACC. However, the increased ratio of p-ACC/ACC might not necessarily reflect an increased AMPK activity, because caffeine dramatically decreased total ACC levels by unknown mechanisms. Given the increased hepatic ATP levels and decreased total ACC levels, it was less likely that caffeine-induced inhibition of mTORC1 was AMPK dependent. In addition to AMPK, mTORC1 is positively regulated by phosphoinositide 3-kinase/protein kinase B (PI3K-AKT). Although it was not investigated in this study, it has been shown that caffeine inhibits AKT phosphorylation and, in turn, inhibits mTORC1 in nonhepatocytes.[19] Therefore, future works are needed to determine the phosphorylation levels of AKT and AMPK to further dissect the mechanisms by which caffeine inhibited mTORC1 in hepatocytes. In addition to negatively regulating autophagy, mTORC1 positively regulates lipid biosynthesis. It will also be interesting to determine whether caffeine could also decrease gene expression of lipid synthesis genes in addition to induction of lipophagy.

Another intriguing finding in the studies by Sinha et al. was that caffeine increased expression of several autophagy proteins in hepatocytes, suggesting that caffeine may also regulate autophagy and lipophagy at the transcriptional level.[14] Recently, it was found that transcriptional factor EB (TFEB), a basic helix-loop-helix leucine zipper transcription factor of the Myc family, is a master regulator for controlling expression of both autophagy and lysosomal genes. More important, overexpression of TFEB in mouse livers significantly inhibits diet-induced steatosis and obesity. In addition to regulating autophagy, TFEB also activates peroxisome proliferator-activated receptor gamma coactivator-1alpha (PGC-1α) and peroxisome proliferator-activated receptor alpha (PPAR-α), two key transcriptional regulators for mitochondrial biogenesis and lipid catabolism.[20] Interestingly, TFEB is regulated at both transcriptional and post-translational levels. TFEB itself is its own target gene, which perhaps can ensure removal of overloaded cellular lipids more efficiently by making more TFEB. Post-translational modification of TFEB mainly regulates its cellular location, and dephosphorylated TFEB translocates from the cytosol to the nucleus. Three different kinases have been shown to phosphorylate TFEB: extracellular signal-related kinase 2; mTOR; and protein kinase Cβ. Although it was not investigated in this study, caffeine might activate TFEB because of its inhibition of mTORC1. Activated TFEB might be responsible for the caffeine-induced lipophagy and lipolysis, mitochondrial bioenergetics, and β-oxidation observed in this study.

Taken together, this important work expands our understanding of how caffeine benefits liver functions through inducing lipophagy and mitochondrial β-oxidation. However, several important questions still need to be answered. How are LDs selectively recognized and removed by autophagy? Are other selective autophagy receptors, such as ubiquitin and p62/SQSTM1, also involved in selective lipophagy? What is the proper dose of caffeine in terms of drinking coffee to induce lipophagy, because overdose of caffeine may be detrimental to the cardiovascular system? Nevertheless, the fact that caffeine can not only induce lipophagy, but also increase lipid mitochondrial β-oxidation suggests that drinking a couple cups of coffee per day may help to burn the fat out of your liver.

Figure 1.

Proposed molecular signaling events in caffeine-induced lipophagy and mitochondria β-oxidation. Caffeine may inhibit PI3K-AKT and, in turn, inhibit mTOR to trigger autophagy by activating the ULK1 complex, which includes ULK1, Atg13, FIP200, and Atg101. Autophagy selectively removes excess LDs to generate FFAs. Decreased mTOR induces TFEB nuclear translocation by decreasing TFEB phosphorylation. TFEB up-regulates expression of autophagy and lysosomal genes, as well as PGC-1α and PPARα, which burn FFAs by increasing mitochondria β-oxidation. Thus, caffeine protects against fatty liver by coordinately inducing lipophagy and mitochondrial β-oxidation. Question mark (“?”) indicates molecular events that were not studied in this study.

  • Wen-Xing Ding, Ph.D.

  • Department of Pharmacology, Toxicology and Therapeutics

  • The University of Kansas Medical Center

  • Kansas City, Kansas