Melatonin and circadian rhythms in liver diseases: Functional roles and potential therapies

Abstract

Circadian rhythms and clock gene expressions are regulated by the suprachiasmatic nucleus in the hypothalamus, and melatonin is produced in the pineal gland. Although the brain detects the light through retinas and regulate rhythms and melatonin secretion throughout the body, the liver has independent circadian rhythms and expressions as well as melatonin production. Previous studies indicate the association between circadian rhythms with various liver diseases, and disruption of rhythms or clock gene expression may promote liver steatosis, inflammation, or cancer development. It is well known that melatonin has strong antioxidant effects. Alcohol drinking or excess fatty acid accumulation produces reactive oxygen species and oxidative stress in the liver leading to liver injuries. Melatonin administration protects these oxidative stress-induced liver damage and improves liver conditions. Recent studies have demonstrated that melatonin administration is not limited to antioxidant effects and it has various other effects contributing to the management of liver conditions. Accumulating evidence suggests that restoring circadian rhythms or expressions as well as melatonin supplementation may be promising therapeutic strategies for liver diseases. This review summarizes recent findings for the functional roles and therapeutic potentials of circadian rhythms and melatonin in liver diseases.

1. Introduction

Circadian rhythms are an internal cycle, which is a biological process with oscillation of approximately 24 hr. Almost all essential physiological events and metabolisms are under control of circadian rhythms¹. For example, in humans, circadian rhythms control sleep-wake cycle, and cognitive activity is upregulated at the wake phase in daytime and the preparation for the sleep phase occurs in nighttime, which is important for memory consolidation². The suprachiasmatic nucleus (SCN) in the hypothalamus is the central pacemaker of the rhythms and synchronizes circadian clocks throughout the body according to light-dark cycle. Synchronization of circadian clocks in different organs controlled by the SCN is mediated by autoregulatory expression of clock genes, such as circadian locomotor output cycles kaput (CLOCK) and brain and muscle ARNT-like 1 (BMAL1) as activators, and period circadian protein homolog 1 (PER1), PER2, cryptochrome circadian regulator 1 (CRY1), and CRY2 as repressors³. Although the brain detects light through retinas and synchronize circadian clocks throughout the body, specific organs, such as the liver, may have their own circadian rhythms or clock gene expression patterns. A previous study has demonstrated that expression levels of clock genes including PER1 and PER2 differ before and after feeding as well as in different tissues, such as the hypothalamus, the skeletal muscle, and the liver, in swine⁴. Koronowski et al. have generated transgenic mice, which have a stop cassette inserted between exon 5 and 6 of BMAL1 resulting in global BMAL1 knockout phenotypes⁵. The stop cassette was flanked by loxP sites and crossing those global BMAL1^−/− mice with mice that express Cre recombinase under the control of α-fetoprotein enhancer (Alfp-Cre mice) allowed the removal of the stop cassette specifically in Alfp-positive hepatic cells⁵. Cre expression is observed selectively in hepatocytes and cholangiocytes in Alfp-Cre mice⁶. As a result, crossed mice expressed BMAL1 only in the liver (hepatocytes and cholangiocytes)⁵. Global BMAL1^−/− mice lost circadian rhythms and expressions, but mice with hepatic BMAL1 expression had rhythmic circadian expressions and metabolisms in the liver without BMAL1 expression and circadian rhythms in the brain and other organs⁵. This liver-specific rhythm was dependent on light-dark cycle, indicating that the liver may have independent circadian rhythms and functions⁵. Since circadian rhythms regulate various physiological events throughout the body, the association of disrupted circadian rhythms with human disorders including liver diseases has been suggested^7,8.

Melatonin (N-acetyl-5-methoxytryptamine) is a hormone associated with circadian rhythms especially in the dark phase. Melatonin synthesis is facilitated at night and inhibited by light at daytime detected by retinas⁹. Melatonin is produced primarily by the pineal gland from the amino acid tryptophan¹⁰. Tryptophan is converted to 5-hydroxytryptophan followed by serotonin (5-hydroxytryptamine). Serotonin is converted to N-acetylserotonin by aralkylamine N-acetyltransferase (AANAT), and finally N-acetylserotonin is converted to melatonin by N-acetylserotonin O-methyltransferase (ASMT) in the pineal gland. Melatonin has a strong antioxidant activity and is an endogenous free radical scavenger. The liver produces melatonin as well as the pineal gland. A study using goldfish identified expressions of AANAT and ASMT in the liver, and expression rhythms of those genes were altered depending on lighting conditions¹¹. Melatonin binds to G protein-coupled receptors MT₁ and MT₂ in humans and rodents, and various hepatic and gastrointestinal cells, which include hepatocytes, gallbladder epithelia, and bile duct epithelia, express these melatonin receptors and can be regulated by melatonin^12–14. Melatonin has protective effects against oxidative stress followed by inflammation caused by reactive oxygen species (ROS) or reactive nitrogen species¹⁵. Therefore, melatonin administration shows therapeutic effects in various disorders including liver diseases^16,17. Melatonin has potentials for novel treatments of liver diseases by decreasing oxidative stress or restoring circadian rhythms and functions. This review summarizes current understandings of the functional roles of circadian rhythms and melatonin signaling in the pathophysiology of liver diseases as well as the potentials of melatonin administration as therapeutic treatments.

2. Functional roles and therapeutic potentials of melatonin and circadian rhythms in liver diseases

2.1. Alcoholic liver disease

Excessive alcohol consumption causes alcoholic liver disease (ALD), which is characterized by liver steatosis, hepatitis, and cirrhosis. Alcohol drinking disrupts intestinal permeability leading to elevated serum levels of lipopolysaccharide (LPS), which is a bacterial endotoxin causing inflammatory responses^18,19. A previous study has analyzed the effects of moderate alcohol drinking on circadian rhythms using 11 daytime workers and 11 nighttime workers. In this study, all subjects were told to avoid alcohol drinking and have usual sleep schedule for 1 week²⁰. After an alcohol-free week, subjects were kept in dim light and blood samples were collected and serum melatonin levels were analyzed in every hour for 24 hr (baseline samples)²⁰. In a following week, all subjects were given red wine (0.5 g of ethanol per kg body weight) every day after finishing work and before bedtime for 1 week²⁰. Blood samples were collected with same procedures to baseline samples (post-alcohol samples)²⁰. This study has demonstrated that nighttime workers but not daytime workers have delayed melatonin secretion peak after 1 week alcohol period, and alcohol consumption increases gut permeability only in nighttime workers, suggesting that nighttime workers are more susceptible for alcohol drinking than daytime workers²⁰. This study compared clock gene expression patterns before and after alcohol period in isolated peripheral blood mononuclear cells from daytime and nighttime workers²⁰. Alcohol drinking elevated expression levels of PER1 and decreased for CRY1, and changed circadian oscillation patterns for CRY2 regardless of working time²⁰. Another study has performed acute or chronic alcohol feeding in mice and has demonstrated that both acute and chronic alcohol drinking changes profiles of hepatic circadian transcriptome, especially in sterol regulatory element-binding protein (SREBP) pathways in the liver, indicating the change of circadian rhythms by alcohol drinking²¹. Binge alcohol drinking increases serum levels of alcohol as well as endotoxin (LPS) in healthy individuals¹⁹. Zeitgeber time (ZT) is the standard notation in studies of the circadian cycle. ZT0 indicates the start of the light phase and ZT12 indicates the start of the dark phase. Alcohol drinking may cause different effects depending on the time of drinking, and binge alcohol gavage in different ZT showed different effects in mice²². Mice with binge alcohol drinking at ZT12 had higher serum alcohol levels than ZT0, and serum LPS levels were highest when gavage was performed at ZT4 or ZT8²². Exposure of zebrafish larvae to 5% ethanol for 1 hr can be fatal, and a previous study found that mortality was highest with exposure at ZT2 and lowest at ZT18, indicating the time-dependent effects of alcohol²³. Mice with albumin promoter-driven Cre recombinase (Alb-Cre mice) have Cre expression selectively in hepatocytes and hepatic stellate cells (HSCs)²⁴. Zhang et al. generated Alb-Cre-mediated liver-specific BMAL1 knockout mice and found that these mice had higher serum aminotransferase (ALT) levels and liver steatosis compared to control mice after chronic and binge alcohol feeding²⁵. Hepatic BMAL1 deficiency decreased expression levels of genes associated with de novo lipogenesis and β-oxidation in the liver, indicating the association between clock genes and liver steatosis²⁵. Adenovirus-mediated BMAL1 overexpression in the liver attenuated alcohol-induced liver damage and steatosis in vivo, indicating the protective effects of BMAL1 on ALD²⁵. Another study has demonstrated that Alb-Cre-mediated hepatic BMAL1 knockout mice had higher serum ALT levels after chronic 3% alcohol feeding for 5 weeks by decreasing expression levels of genes associated with glycogen metabolism²⁶.

Alcohol consumption leads to the production of ROS during alcohol metabolism, which induces liver damage²⁷. Therefore, melatonin can protect the liver against ALD with its antioxidant effects²⁸. Melatonin administration (10 mg/kg body weight daily for 10 days via intraperitoneal [i.p.] injection) along with ethanol i.p. injection increased circulating white blood cells in serum as well as antioxidant capacity in the liver, the skeletal muscle, and the kidney in mice²⁹. Another study administered alcohol into pregnant rats by gavage with or without melatonin (10 or 15 mg/kg, i.p. injection during gestational period)³⁰. In this model, melatonin administration decreased levels of alcohol-induced liver damage not only in mothers but also in newborn pups³⁰. Although further studies are required to develop novel treatments for ALD targeting circadian rhythms and clock gene expressions, melatonin administration has shown promising therapeutic effects in these previous studies.

2.2. Non-alcoholic fatty liver disease

Patients with non-alcoholic fatty liver disease (NAFLD) have similar symptoms to ALD, such as liver steatosis, inflammation, and cirrhosis without alcohol consumption. A study of 4,740 male workers has found that night shift workers have higher serum ALT levels than daytime workers with the increase in night shift years, indicating the association between circadian disruption and liver malfunctions³¹. Exposure to light for 2 hr during night time induces glucose intolerance in rats, which is associated with NAFLD³². Shimizu et al. have developed a rat model with 4 hr delayed feeding during ZT16-ZT4 compared to control feeding during ZT12-ZT24³³. In this model, rats with delayed feeding gained increased body weights and perirenal adipose tissues compared to rats with control feeding, and circadian oscillations of hepatic clock and fatty acid synthesis genes were delayed 2–4 hr due to shifted feeding time³³. Another study fed mice with high-fat diet (HFD) during active phase (nighttime) or sleep phase (daytime) and found that mice with sleep phase feeding had increased body weights as well as hepatic concentrations of free fatty acids (FFAs) and triglycerides compared to mice with active phase feeding³⁴. Arrhythmic feeding of mice with one-eighth portion of daily food intake given every 3 hr disrupted rhythmic expression of over 70% of transcriptomes in the liver³⁵. Circadian disruption induces leptin resistance in mice, which may lead to obesity or metabolic syndromes, indicating different impacts of food intake in the liver depending on the timing of intake³⁶. A previous study has demonstrated that Alb-Cre-mediated liver specific BMAL1 knockout mice have elevated mRNA methylation especially in peroxisome proliferator-activated receptor alpha (PPARα) in the liver³⁷. PPARα promotes β-oxidation of FFAs and lipid metabolism³⁸, and methylation of PPARα mRNA causes mRNA degradation and inhibition of PPARα expression leading to accumulation of triglycerides in the liver³⁷. Although detailed mechanisms are undefined, circadian rhythms and rhythmic clock gene expressions may be critical for liver homeostasis and lipid metabolism.

Mitochondrial β-oxidation breaks down FFAs but produces ROS, which causes liver damage. FFAs also induce endoplasmic reticulum (ER) stress, which is caused by accumulation of unfolded and misfolded proteins, and ER stress is another potent source of ROS³⁹. Melatonin has strong antioxidant effects and can be utilized to protect the liver from NAFLD or fatty acid-induced liver damage. Melatonin administration (500 μg/kg daily for 4 weeks, i.p. injection) increased antioxidant activities and decreased lipid peroxidation as well as ER stress in the liver of obese (ob/ob) mice⁴⁰. Another study has demonstrated that melatonin (100 mg/kg daily for 10 weeks, dissolved in drinking water) attenuates ER stress leading to improved liver steatosis and insulin resistance in mice fed with HFD⁴¹. HFD induced elevated hepatic de novo lipogenesis and triglyceride concentrations in hamsters, but melatonin administration (10, 20, or 50 mg/kg daily for 8 weeks via gavage) decreased enzyme activities of acetyl-CoA carboxylase (ACC) and fatty acid synthase (FAS), which are associated with de novo lipogenesis, leading to attenuated liver steatosis⁴². Activation of brown adipose tissues leads to body weight loss, improved glucose tolerance, and decreased lipid accumulation in the liver, and may be a therapeutic target for NAFLD⁴³. Halpern et al. administered melatonin (3 mg) to patients with melatonin deficiency due to radiotherapy or surgical removal of pineal gland for 3 months and analyzed brown adipose tissue mass⁴⁴. Melatonin supplementation increased mass and activity of brown adipose tissues and decreased blood levels of total cholesterol and triglyceride⁴⁴. Pre-treatment with melatonin (0.1–0.3 mM) inhibited accumulation of cholesterol and triglyceride in hepatocellular carcinoma (HCC) line, HepG2 cells, incubated with high concentrations of oleic acid in vitro⁴⁵. Oleic acid elevated expression levels of genes associated with de novo lipogenesis including SREBP-1c, FAS, and stearoyl-CoA desaturase 1 (SCD1) and decreased β-oxidation gene expression such as PPARα and carnitine palmitoyltransferase I (CPT1), but melatonin treatment abolished those effects leading to inhibited lipid accumulation in HepG2 cells⁴⁵.

A previous study has demonstrated that melatonin administration (30 mg/kg daily for 7 weeks, i.p. injection) attenuates HFD-induced liver steatosis, inflammation, and fibrosis in mice⁴⁶. HFD elevated activation of apoptosis signal-regulating kinase 1 (ASK1) and downstream signaling pathways including mitogen-activated protein kinase kinase 3/6 (MKK3/6), MKK4/7, p38, and c-Jun N-terminal kinase (JNK) leading to liver damage, but melatonin ameliorated HFD-induced liver damage by inhibiting activation of these pathways in vivo⁴⁶. Intragastric melatonin administration (10 mg/kg daily for 12 weeks) in another study decreased serum ALT levels, liver steatosis, proinflammatory cytokine expression including interleukin 6 (IL-6) and tumor necrosis factor alpha (TNFα), and activation of mitogen-activated protein kinase (MAPK), JNK, and p38 signaling pathways in HFD mice, indicating that melatonin inhibits signaling pathways associated with cytokine production, inflammation, and apoptosis⁴⁷. Mitochondria play a vital role in β-oxidation of FFAs, and mitochondrial dysfunctions, such as extensive mitochondrial permeability transition pore (mPTP) opening and impaired respiratory functions, lead to excess ROS production causing apoptotic cell deaths⁴⁸. Das et al. have demonstrated that palmitic acid induces mitochondrial fragmentation, calcium overload, and mPTP opening, but melatonin treatment (1 mM for 30 min) inhibits these effects in HepG2 cells in vitro⁴⁹. HFD decreased mitochondrial enzyme activities in mice, such as cytochrome C oxidase and pyruvate dehydrogenase, but melatonin administration (10 or 20 mg/kg daily for 4 weeks, i.p. injection) restored these activities in vivo⁴⁹. Another study also demonstrated mitochondrial dysfunction including calcium overload and mPTP opening caused by HFD in mice, and melatonin administration (20 mg/kg daily for 12 weeks, i.p. injection) reversed these effects by inhibiting expression of nuclear receptor subfamily 4 group A member 1 (NR4A1) and activation of p53, suggesting that melatonin restores mitochondrial functions leading to improved liver conditions in NAFLD⁵⁰. Gut microbiota is associated with the pathophysiology of NAFLD and gut bacteria transplantation has potential therapeutic effects to improve liver conditions⁵¹. Yin et al. have demonstrated that HFD decreases gut microbiota diversity in mice, and melatonin administration (108 mg/kg daily for 2 weeks, supplemented in drinking water) improved this diversity as well as liver steatosis in HFD mice⁵². In this model, HFD altered the mouse gut microbiota balance by elevating abundance of Lactobacillus and decreasing abundance of Bacteroides, which was also observed in cholangiopathies and may be associated with liver damage and fibrosis⁵³, and melatonin reversed these effects⁵². Although further studies are required, melatonin administration may be a promising therapeutic tool for NAFLD by improving mitochondrial activities, inhibiting lipid accumulation in the liver, and maintaining gut microbiota diversity.

2.3. Cholangiopathies

Cholangiopathies or bile duct disorders include primary sclerosing cholangitis (PSC), primary biliary cholangitis (PBC), and biliary atresia. Cholangiopathies are characterized by cholestasis, ductular reaction, biliary inflammation, and liver fibrosis⁵⁴. These disorders are slow but progressive and liver conditions can be severe and cirrhotic especially in the later stages⁵⁵. It has been suggested that circadian rhythm and melatonin synthesis is associated with the pathophysiology of cholangiopathies and liver cirrhosis. A previous study using 74 PBC patients, 60 cirrhosis patients, and 79 healthy individuals has found that sleep timing is significantly delayed in patients with PBC or cirrhosis, and delayed sleep timing is associated with impaired sleep quality in PBC patients⁵⁶. Another study using 13 PBC patients, 7 cirrhosis patients, and 6 healthy individuals found that patients with PBC or cirrhosis had poor sleep quality assessed by the Pittsburgh Sleep Quality Index⁵⁷. This study has demonstrated that 15-day morning bright light treatment for PBC patients using 10,000 lux light box for 45 minutes improves sleep quality and rhythmic secretion of 6-sulphatoxymelatonin in urine, which is a metabolite of melatonin, indicating that cholangiopathies and liver cirrhosis are associated with disrupted circadian rhythms and melatonin production, and restoring rhythms by morning light treatment may be a therapeutic approach for cholangiopathies⁵⁷.

Bile duct ligation (BDL) is a surgical ligation of the common bile duct to mimic cholestasis in rodents, and causes ductular reaction, biliary inflammation, liver damage and fibrosis⁵⁴. BDL elevates expression levels of MT₁ and MT₂ melatonin receptors and clock genes including CLOCK, BMAL1, PER1, and CRY1 in cholangiocytes in vivo^14,58. BDL also elevates expression of AANAT in cholangiocytes, and inhibition of AANAT leads to cholangiocyte proliferation via enhanced expression of vascular endothelial growth factor (VEGF)-A and VEGF-C in BDL rats^59,60. Overexpression of AANAT inhibits VEGF-A and VEGF-C expression as well as cell proliferation in murine cultured cholangiocytes, indicating local melatonin synthesis followed by regulations of cholangiocyte functions during cholestatic liver injury^59,60. A previous study has performed melatonin administration (1 mg/kg per day for 7 days) via intracerebroventricular-implanted cannulas for BDL rats and found that melatonin ameliorated ductular reaction and liver fibrosis by inhibition of gonadotropin releasing hormone (GnRH) secretion in vivo⁶¹. Elevated GnRH secretion activates HSCs leading to robust liver fibrosis⁶². Dark therapy, which maintains animals in complete dark 24 hr for 1 week, facilitates AANAT expression and melatonin secretion form the pineal gland in rats⁶³. Dark therapy for BDL rats attenuated ductular reaction and liver fibrosis by decreasing cAMP levels in cholangiocytes⁶³. A previous study has performed pinealectomy surgery or light treatment, which maintains animals in complete light 24 hr for 1 week, for BDL rats⁶⁴. In this model, both pinealectomy and light treatment decreased serum melatonin levels and AANAT expression in cholangiocytes compared to control BDL rats⁶⁴. Pinealectomy or light treatment exacerbated ductular reaction and liver fibrosis in BDL rats via elevated VEGF-A and proinflammatory cytokine production, such as IL-1β and IL-6, as well as increased ROS levels in the liver⁶⁴. Mdr2^−/− mice are the most common mouse model for human PSC, which have similar conditions such as cholestasis, biliary inflammation, ductular reaction, and liver fibrosis⁵⁴. Dark therapy (complete dark for 1 week) or melatonin administration (2 mg/g body weight per day for 1 week via drinking water) for Mdr2^−/− mice improved liver conditions compared to control Mdr2^−/− mice with 12:12 hr light dark cycle by inhibiting expression of VEGF-A/C in cholangiocytes via decreased levels of microRNA (miRNA) miR-200b⁶⁵. Thioacetamide (TAA) induces biliary damage in rodents and utilized as a model of cholestatic liver injury⁵⁴. A previous study has demonstrated that melatonin administration (5 mg/kg body weight per day for 1 week via i.p. injection) improved liver conditions with lower serum alanine ALT and aspartate aminotransferase (AST) levels, decreased proinflammatory cytokine expression, such as IL-1β and TNFα, as well as profibrogenic factors such as transforming growth factor beta 1 (TGF-β1) and collagen type I in the liver by attenuating oxidative stress during TAA-induced liver injury in rats⁶⁶. Melatonin supplementation may be a promising approach to regulate cholangiocyte functions and maintain liver homeostasis.

2.4. Toxin-induced liver injury

Toxins or carcinogens, such as TAA and carbon tetrachloride (CCl₄), are commonly administered into rodents to cause liver damage and understand the pathophysiology of liver injury⁵⁴. Acute or chronic CCl₄ administration causes oxidative stress-induced liver damage and fibrosis, and melatonin can protect the liver with its antioxidant characteristics. Melatonin administration (2.5–20 mg/kg per day for 6 weeks, i.p. injection) for CCl₄-administered rats decreased serum ALT and AST levels, as well as liver fibrosis by inhibiting expression of TGF-β1 and VEGF-A in the liver^67,68. Although both melatonin treatments before or after CCl₄ administration have protective effects on CCl₄-induced liver injury⁶⁹, a study demonstrated that post-treatment had better therapeutic effects compared to co-treatment in CCl₄-administered rats⁷⁰. Another study demonstrated therapeutic effects of melatonin administration (5 or 10 mg/kg per day, i.p. injection) against CCl₄-induced liver injury in mice with decreased liver fibrosis levels⁷¹. This study found that administration of CCl₄ decreased expression of clock genes, such as BMAL1, CLOCK, PER1/2, CRY1/2, and melatonin treatment restored expression levels of these genes in the liver⁷¹. Stem cell therapy performs transplantation of stem cells to facilitate liver regeneration and improve conditions in various liver diseases including cirrhosis⁷². A previous study treated bone marrow-derived mesenchymal stem cells with or without 5 μM melatonin for 24 hr and infused them into CCl₄-administered rats via tail vein⁷³. Stem cell transplantation decreased serum ALT and AST levels as well as liver fibrosis, and melatonin pre-treatment of stem cells enhanced these therapeutic effects⁷³. Melatonin treatment elevated expression levels of MT₁ and MT₂ melatonin receptors in stem cells, which may be associated with those effects⁷³.

Melatonin shows therapeutic effects on liver damage caused by not only TAA and CCl₄ but also other toxins, indicating its universal effects. α-naphthylisothiocyanate (ANIT) is a chemical compound that causes cholestasis and liver damage. ANIT elevated serum ALT and AST levels, and melatonin supplementation (100 mg/kg in drinking water for 12 hr following the ANIT injection) decreased those levels in rats⁷⁴. Valproic acid (VPA) is a mood-stabilizing agent used for psychiatric disorders, but also causes liver damage. VPA elevated expression of proinflammatory cytokines including IL-1β and TNFα as well as hepatocyte apoptosis, and melatonin administration (10 mg/kg via gavage, 1 hr prior to VPA treatment for 14 days) inhibited those effects⁷⁵. Dexamethasone (DEX) is a type of corticosteroids used to decrease inflammatory responses, but it can cause liver steatosis. A previous study administered DEX to pregnant rats at gestational day 14–21, and rats were sacrificed at postnatal day 120⁷⁶. In this model, prenatal DEX exposure elevated liver steatosis, IL-6, TNFα and TGF-β1 expression, and apoptosis in the liver, and melatonin supplementation (1 mg/kg per day in drinking water, from gestational day 14–21 to postnatal day 120) reversed this DEX-induced liver conditions⁷⁶. Melatonin has strong protective effects on toxin-induced liver damage especially caused by oxidative stress.

2.5. Hepatocellular carcinoma

HCC is the most common malignancy of primary liver cancers. Previous studies suggest the association of disruption of circadian rhythms with cancer development in various organs including the liver⁷⁷. Kettner et al. generated Cry1^−/−Cry2^−/− mice, Per1^−/−Per2^−/− mice, and Alb-Cre-mediated liver-specific BMAL1 knockout mice and maintained them for 90 weeks⁷⁸. These transgenic mice had lower survival rates with various disorders including neurodegeneration, cystic renal dysplasia, liver steatosis, and HCC compared to wild-type mice⁷⁸. This study also maintained wild-type mice with steady 12:12 hr light-dark cycle or weekly transfer to another room with 8 hr difference in light-dark cycle for 4–90 weeks to generate “jet-lagged” mice⁷⁸. Jet-lagged mice had higher lipid deposition in the liver followed by HCC development and poor survival rates compared to mice with steady cycle⁷⁸. Chronic jet lag caused disruption of hepatic cholesterol, bile acid, and xenobiotic metabolism via elevation of constitutive androstane receptor (CAR), which is a regulator of xenobiotic metabolism and associated with hepatocarcinogenesis, indicating the association of circadian rhythms with liver metabolism and HCC development⁷⁸. Another study has found negative correlation of expression levels between BMAL1 and HNF4α, which is important for hepatocyte fate determination and functions⁷⁹. HNF4α had circadian activities with rhythmic expression of downstream targets such as cyclin B1 and D1 in murine hepatocyte line AML12 cells⁷⁹. Human HCC lines, HepG2, Hep3B, and Huh7 cells, expressed high levels of HNF4α and low levels of BMAL1, and overexpression of BMAL1 in HepG2 cells caused elevated apoptosis and decreased tumor growth in xenograft mouse models, indicating the roles of circadian rhythms in hepatocyte carcinogenesis⁷⁹.

HCC tumors often have genetic mutations in various genes or their promoters such as telomerase reverse transcriptase (TERT), tumor protein p53 (TP53), and catenin beta 1 (CTNNB1)⁸⁰. A previous study using 335 HCC patients and 1196 control individuals has identified single nucleotide polymorphisms (SNPs) in MT₁ melatonin receptor, which are associated with the elevated risk of HCC⁸¹. A specific haplotype of four MT₂ SNPs was also associated with HCC development, indicating that functions of receptors and melatonin signaling may be associated with HCC development⁸¹. During 12:12 light-dark cycle, exposure to blue-enriched light (increased transmittance of 462–484 nm and decreased red light greater than 640 nm) at daytime using blue-tinted cages increased plasma levels of melatonin at nighttime in rats⁸². This blue light treatment followed by promotion of melatonin secretion at night inhibited tumor growth in xenograft rat models with Morris 7288CTC hepatoma although detailed mechanisms are undefined⁸³. Diethylnitrosamine (DEN) administration is utilized to induce HCC in rodents as an animal model of HCC⁸⁴. A previous study administered melatonin (5 or 10 mg/kg per day for 10, 20, 30, or 40 weeks via i.p. injection) to DEN mice and demonstrated that melatonin inhibited HCC development and improved liver conditions during DEN-induced liver injury⁸⁵. DEN elevated gene expression of CLOCK and BMAL1, and decreased expression of PER1, PER2, and CRY1, but melatonin injection inhibited these gene expression changes⁸⁵. Another study analyzed the efficacy of the combination of stem cell therapy and melatonin administration in DEN rat models⁸⁶. Although both transplantation of bone marrow-derived MSCs isolated from young rats as well as melatonin administration (20 mg/kg, twice a week for 5 weeks, i.p. injection) attenuated DEN-induced liver damage and HCC development in rat liver, the combination of stem cell transplantation and melatonin administration had the highest therapeutic effects by decreasing expression of anti-apoptosis genes including B-cell lymphoma 2 (Bcl-2) and survivin and by increasing apoptotic gene expression such as caspase3 and Bcl-2-associated X protein (BAX) in this model⁸⁶.

Chemotherapy is commonly performed especially for advanced HCC, and sorafenib has been widely utilized as the standard agent⁸⁷. Sorafenib treatment induced apoptosis by decreasing the Bcl-2/BAX ratio as well as ER stress followed by autophagy via activation of protein kinase R-like endoplasmic reticulum kinase (PERK)/activating transcription factor 4 (ATF4) pathways in HepG2 cells⁸⁸1. Melatonin increased the sensitivity of HepG2 cells to sorafenib leading to enhanced cell death and inhibition of proliferation by inhibiting PERK/ATF4-mediated autophagy in vitro⁸⁸. Melatonin itself suppressed cell proliferation, migration, and invasion of HepG2 and Huh7 cells, and the combination of melatonin and etoposide (VP16) had better anti-cancer effects compared to administrations of melatonin only or VP16 only in xenograft mouse models with Huh7 tumors by inhibiting expression of DNA repair protein RAD51 via long non-coding RNA (lncRNA) RAD51-AS1 in vivo, indicating effective anti-cancer effects of melatonin against HCC⁸⁹.

2.6. Cholangiocarcinoma

Cholangiocarcinoma (CCA) is a type of hepatic malignancy that emerges from the biliary tree and is the second most common primary liver cancer after HCC. Although the association of circadian rhythms with CCA development is still not fully elucidated, a previous study found upregulated and downregulated expression levels of BMAL1 and PER1, respectively in human CCA biopsies or human CCA cell lines, such as Mz-ChA-1 and HuH28 cells⁹⁰. Normal human cholangiocyte line, H69 cells, but not CCA cell lines had circadian expression rhythms for clock genes including BMAL1, CLOCK, PER1, and CRY1⁹⁰. Overexpression of PER1 inhibited cell proliferation in Mz-ChA-1 cells and decreased tumor size in xenograft mouse models, indicating potentials of circadian rhythms and clock genes as therapeutic targets for CCA⁹⁰.

Han et al. has demonstrated that CCA tumor tissues from patients express lower levels of AANAT and ASMT compared to normal liver tissues, and human CCA cell lines including Mz-ChA-1 and HuH28 cells express significantly lower levels of AANAT and ASMT compared to H69 cells⁹¹. Overexpression of AANAT inhibited cell proliferation in Mz-ChA-1 cells, and melatonin administration (4 mg/kg, three times a week for 43 days, i.p. injection) decreased tumor size in the xenograft mouse models with Mz-ChA-1 tumors⁹¹. Melatonin also inhibited cell proliferation in human CCA cell lines KKU-M055 and KKU-M214 cells by elevation of ROS-induced DNA damage and decreased expression of Bcl-2 leading to apoptosis in vitro⁹². Liver fluke infection caused by parasites such as Opisthorchis viverrini is a common risk factor for CCA in Asia⁹³. O. viverrini infection plus carcinogen N-nitrosodimethylamine (NDMA) administration developed CCA in hamsters, and melatonin administration (10 or 50 mg/kg, daily for 120 days, orally) elevated Bcl-2 expression and decreased BAX expression resulting in attenuated CCA development and improved survival rates in this model⁹⁴. CCA tumors are often accompanied with the tumor microenvironment in dense stromal tissues containing fibroblasts and lymphocytes promoting CCA progression⁹⁵. Melatonin inhibited lymphocyte infiltration such as Th17⁺ cells and decreased inflammatory responses including expression of IL-1β and NF-κB in hamsters with fluke infestation and NDMA treatment⁹⁶. Although further evidence is required, melatonin administration may be used for CCA to inhibit tumor growth and regulate functions of CCA microenvironment.

2.7. Liver transplantation and ischemia-reperfusion injury

Liver transplantation is often the sole treatment for end-stage liver diseases, such as liver cirrhosis, late stage PSC and PBC, and liver cancers. However, liver transplantation has a risk of ischemia-reperfusion injury (IRI), which is characterized by oxidative stress, inflammation, and liver damage, and IRI may lead to graft rejection⁹⁷. A previous study monitored blood pressure of 107 liver transplant recipients and found that approximately 90% of patients had arterial hypertension and abnormal circadian blood pressure pattern after operation⁹⁸. Another study analyzed 147 patients who underwent liver transplantation in one institution and found that 45 patients who had transplantation at night suffered higher incidence of intraoperative blood loss and other complications, postoperative abdominal infection, longer operation time, and longer time to restore liver functions after transplantation compared to 102 patients with daytime operation⁹⁹. Although night shift could influence performances of medical workers, circadian rhythms ant the timing of operation may affect conditions of patients after liver transplantation.

Deng et al. clamped first porta hepatis and released it after 35 minutes to recover the blood flow to induce IRI in rats¹⁰⁰. In this model, pretreatment of melatonin (10 mg/kg, 35 and 70 min prior to ischemia, i.p. injection) decreased serum ALT and AST levels showing therapeutic potentials of melatonin against IRI liver damage¹⁰⁰. Living donor liver transplantation is performed using a portion of the liver from a healthy living individual and may compensate the shortage of available liver grafts. However, the donated liver graft from the living person is generally small, and small grafts may cause small-for-size syndrome, which leads to graft dysfunction or complications¹⁰¹. A previous study has performed liver transplantation in mice with small grafts (30% of total liver) and found that melatonin treatments (20 μg/mL in cold saline to perfuse the donor liver and 10 mg/kg 15 min before operation for the recipient, i.p. injection) improved survival rates of recipient mice¹⁰². The combination of IRI and partial hepatectomy (60 min hepatic ischemia and reperfusion combined with 70% or 80% liver excision) caused severe liver damage in mice, and 10 mg/kg melatonin pretreatment decreased serum ALT and AST levels and promoted hepatocyte proliferation and liver regeneration via elevation of IL-6 and TNFα expression¹⁰². Since IRI is caused by oxidative stress during reperfusion, melatonin treatment shows strong antioxidant effects to protect liver damage.

3. Therapeutic potentials of extracellular vesicles

Extracellular vesicles (EVs) are small membrane-bound vesicles secreted from various cells. There are several types EVs including exosomes, microvesicles or microparticles, and apoptotic bodies according to their biogenesis and particle size. EVs contain cargo mediators inside, such as proteins, DNAs, and RNAs, and donor cells can transfer those mediators into recipient cells by secreting EVs. This EV-mediated cell-to-cell communication plays a vital role in the pathophysiology of liver diseases¹⁰³. For example, hepatocytes secrete elevated numbers of EVs during lipid-induced damage, and these EVs contain tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)¹⁰⁴. Lipotoxic hepatocyte-derived EVs induced inflammatory responses such as IL-1β and IL-6 production in macrophages by delivering cargo TRAIL¹⁰⁴. Another study has demonstrated that cholangiocytes secrete EVs that contain elevated levels of lncRNA H19 during cholestatic liver injury, and these H19-enriched EVs activates HSCs leading to liver fibrosis in cholangiopathies¹⁰⁵.

EVs mediate inflammatory or fibrogenic responses in hepatic cells by delivering cargo mediators, which means that EVs could be used as therapeutic tools or drug carriers to regulate functions of cells by delivering mediators to maintain homeostasis during liver diseases¹⁰⁶. For example, EVs secreted from liver stem cells contained elevated levels of miRNA let-7a, and injection of stem cell-derived EVs ameliorated ductular reaction and liver fibrosis in Mdr2^−/− mice by delivering cargo let-7a¹⁰⁷. Tunicamycin is a drug that causes hepatic ER stress followed by steatosis in mice and used as a drug-induced NAFLD model. Rong et al. isolated EVs secreted from cultured primary adipocytes harvested from mice treated with melatonin (20 mg/kg daily for 2 weeks, i.p. injection) and injected those EVs into tunicamycin-treated mice¹⁰⁸. Tunicamycin caused elevated hepatic triglyceride levels, ER stress gene expression, such as C/EBP homologous protein (CHOP), 78-kDa glucose-regulated protein (GRP78), and inositol-requiring enzyme 1 (IRE1), elevated de novo lipogenesis gene expression including ACC and SREBP-1c, and decreased β-oxidation gene expression such as CPT1¹⁰⁸. EVs isolated from mice with melatonin administration ameliorated tunicamycin-induced gene expression changes and hepatic steatosis in vivo¹⁰⁸. This study has demonstrated that resistin, which is an adipocytokine associated with obesity, is a key cargo mediator and causes hepatic steatosis via EVs delivered from adipose tissues into the liver¹⁰⁸. Although current studies are limited and further experimental evidence is required, EVs may have the great potentials as drug/mediator carriers to deliver melatonin or melatonin-induced elevated levels of proteins or RNAs leading to the regulations of hepatic cells and improvements of liver conditions.

4. Conclusions and future perspectives

Current studies suggest the association between circadian rhythms as well as melatonin secretion with pathophysiological events in various liver diseases. Table 1 summarizes effects of disruption of circadian rhythms leading to liver diseases. Melatonin administration has numbers of therapeutic effects in liver disorders, which are not limited to antioxidant effects. Figure 1 and Table 2 summarize therapeutic effects of melatonin administration shown in recent studies. Disruption of circadian rhythms and altered clock gene expressions have been identified in various liver diseases, and melatonin administration can change expression levels of clock genes. Although detailed mechanisms are still not fully elucidated, expression changes of clock genes and circadian rhythms may contribute to the therapeutic effects of melatonin. Table 3 summarizes clock gene expression in liver diseases and effects of melatonin administration. Melatonin treatments mainly focus on the regulation of functions or cell events in target cells (e.g., restoration of mitochondrial functions in hepatocytes to inhibit lipid deposition during NAFLD⁴⁹). However, hepatic cells communicate with each other and contribute in orchestration to the pathophysiology of liver diseases¹⁰⁹. Melatonin administration in the future may target cell-to-cell communication between hepatic cells to inhibit pathological cellular communication during liver diseases (e.g., improved EV-mediated communication between adipose tissue and the liver during NAFLD¹⁰⁸). Current studies of melatonin therapies have limitations. A number of studies use various doses and administration methods (intraperitoneal injection or via drinking water), and there are no gold-standard methods for melatonin treatments. Although a majority of current studies use 10 mg/kg/day for rats or mice, this dose may not be reasonable for human patients because melatonin tablets that can be purchased over the counter are 3 or 5 mg. Daily melatonin supplementation (3 mg) for 3 months increased brown adipose tissue mass in patients with melatonin deficiency⁴⁴. Although 3 or 5 mg melatonin may be effective in humans, current studies using human subjects are limited and detailed effects and optimal doses of melatonin are still undefined. In addition, melatonin regulates circadian rhythms and melatonin secretion is rhythmic in healthy individuals, but current studies merely supplement melatonin and do not consider the timing of supplementation or rhythms. Melatonin administration via drinking water does not mimic rhythmic melatonin secretion. It is not fully elucidated whether melatonin has therapeutic effects regardless of circadian rhythms or rhythmic melatonin supplementation restores disrupted rhythms and has better effects. Further studies are required to understand the roles of melatonin and circadian rhythms in liver diseases and to develop effective and feasible melatonin treatments for patients with liver diseases.

Table 1.

Selected effects of circadian disruption leading to liver diseases.

Effects	Associated disease	Model
Elevation of gut permeability	ALD	Nighttime workers with moderate alcohol drinking20
Elevated liver damage and steatosis	ALD	Liver-specific BMAL1 knockout mice with alcohol feeding25,26
Increased glucose tolerance	NAFLD	Rats with light exposure at night32
Increased body weight and adipose tissue	NAFLD	Rats with delayed feeding33
Increased hepatic lipid accumulation	NAFLD	Mice with HFD at daytime34
Impaired lipid metabolism	NAFLD	Liver-specific BMAL1 knockout mice37
Elevated liver damage	Cholestasis	BDL rats with complete light64
High HCC development	HCC	Jet-lagged mice78
High incidence of complications	IRI	Patients with liver transplantation at night99

Figure 1. Therapeutic effects of melatonin in liver diseases.

Current studies have represented various therapeutic effects of melatonin administration that are not limited to antioxidant effects. Melatonin decreases de novo lipogenesis in the liver and increases the mass and activities of brown adipose tissues leading to attenuated liver steatosis in NAFLD. Melatonin can increase functions and enzyme activities in mitochondria, which promotes β-oxidation of FFAs and inhibits lipid accumulation in the liver. Melatonin also increases circulating white blood cells and diversity of gut microbiota contributing to improved liver conditions. Proliferation of cholangiocytes can be regulated by melatonin and melatonin administration could be utilized to inhibit ductular reaction and liver fibrosis in cholangiopathies such as PSC. Melatonin inhibits proliferation and tumor growth of HCC and CCA and could be utilized to increase the effects of chemotherapy.

Table 2.

Selected therapeutic effects of melatonin administration in liver diseases.

Effects	Dose	Associated disease	Model
Increase of circulating white blood cells	10 mg/kg/day	ALD	Mouse with ethanol injection29
Decreased liver damage	10 or 15 mg/kg/day	ALD	Pregnant rat with ethanol gavage30
Decreased oxidative and ER stress	500 μg/kg/day	NAFLD	Ob/ob mice40
Improved steatosis and insulin resistance	100 mg/kg/day	NAFLD	HFD mice41
Decreased de novo lipogenesis	10–50 mg/kg	NAFLD	HFD mice42
Increase of brown adipose tissue mass and activity	3 mg/day	NAFLD	Patients with melatonin deficiency44
Restored mitochondria functions	10 or 20 mg/kg/day	NAFLD	HFD mice49,50
Improved gut microbiota diversity	108 mg/kg/day	NAFLD	HFD mice52
Improved liver conditions	2 mg/g/day	PSC	Mdr2−/− mice65
Improved liver conditions	5 mg/kg/day	Bile duct injury	Rats with TAA administration66
Decreased liver damage and steatosis	2.5–20 mg/kg/day	Toxin-induced liver damage	Rats with CCl4 administration67,68
Improved clock gene expression	5 or 10 mg/kg/day	Toxin-induced liver damage	Mice with CCl4 administration71
Inhibition of tumor development	5 or 10 mg/kg/day	HCC	Mice with DEN administration85
Increased sensitivity to chemotherapy	40 mg/kg, 5 days per week	HCC	Xenograft mice with etoposide administration89
Decreased tumor size	4 mg/kg, 3 times per week	CCA	Xenograft mice91
Inhibition of tumor development	10 or 50 mg/kg/day	CCA	Hamsters with O. viverrini infection and NDMA administration94
Reduced liver damage	10 mg/kg	IRI	Rats with porta hepatis clamping and reperfusion100
Promoted liver regeneration	10 mg/kg	Partial liver transplantation	Mice with right lobe clamping and partial hepatectomy102

Table 3.

Expression changes of clock genes in liver diseases and effects of melatonin.

Disease/model	Clock gene expression	Effects of melatonin
Alcohol drinking	PER1↑, CRY1↓ in humans20 CLOCK↑, PER2↓ in mice21	Undefined
Four hour-shifted feeding mice	Shifted expression oscillation for BMAL1, CLOCK, PER1/2, CRY1/233	Undefined
Daytime or nighttime feeding mice	Different expression pattern for PER1 and PER234	Undefined
BDL rats	BMAL1↑, CLOCK↑, PER1↑, CRY1↑14	BMAL1↓, CLOCK↓, PER1↓, CRY1↓14
Human PSC patients	BMAL1↑, CLOCK↑, PER1↑, CRY1↑64	Undefined
CCl4-treated mice	BMAL1↓, CLOCK↓, PER1/2/3↓, CRY1/2↓71	BMAL1↑, CLOCK↑, PER1/2/3↑, CRY1/2↑71
DEN-induced HCC mice	BMAL1↑, CLOCK↑, PER1/2/3↓, CRY1↓85	BMAL1↓, CLOCK↓, PER1/2/3↑, CRY1↑85
CCA	BMAL1↑, PER1↓ in CCA tumors90 Lost rhythmic expression for BMAL1, CLOCK, PER1, CRY1 in CCA cell lines90	Undefined

In conclusion, melatonin has protective effects against various liver diseases, and circadian rhythms and clock gene expressions are promising therapeutic targets for the management of liver conditions.

Abbreviations

AANAT

aralkylamine N-acetyltransferase

ACC

acetyl-CoA carboxylase

Alb

albumin

ALD

alcoholic liver disease

Alfp

α-fetoprotein

ALT

aminotransferase

ANIT

α-naphthylisothiocyanate

ASK1

apoptosis signal-regulating kinase 1

ASMT

N-acetylserotonin O-methyltransferase

AST

aminotransferase

ATF4

activating transcription factor 4

BAX

Bcl-2-associated X protein

Bcl-2

B-cell lymphoma 2

BDL

bile duct ligation

BMAL1

brain and muscle ARNT-like 1

CAR

constitutive androstane receptor

CCA

cholangiocarcinoma

CCl₄

carbon tetrachloride

CHOP

C/EBP homologous protein

CLOCK

circadian locomotor output cycles kaput

CPT1

carnitine palmitoyltransferase I

CRY1

cryptochrome circadian regulator 1

CTNNB1

catenin beta 1

DEN

diethylnitrosamine

DEX

dexamethasone

ER

endoplasmic reticulum

FAS

fatty acid synthase

FFAs

free fatty acids

GnRH

gonadotropin releasing hormone

GRP78

78-kDa glucose-regulated protein

HCC

hepatocellular carcinoma

HFD

high-fat diet

HSCs

hepatic stellate cells

IL

interleukin

IRE1

inositol-requiring enzyme 1

IRI

ischemia-reperfusion injury

JNK

c-Jun N-terminal kinase

lncRNA

long non-coding RNA

LPS

lipopolysaccharide

MAPK

mitogen-activated protein kinase

miRNA

microRNA

MKK

mitogen-activated protein kinase kinase

mPTP

mitochondrial permeability transition pore

NAFLD

non-alcoholic fatty liver disease

NDMA

N-nitrosodimethylamine

NR4A1

nuclear receptor subfamily 4 group A member 1

PBC

primary biliary cholangitis

PER1

period circadian protein homolog 1

PERK

protein kinase R-like endoplasmic reticulum kinase

PPARα

peroxisome proliferator-activated receptor alpha

PSC

primary sclerosing cholangitis

ROS

reactive oxygen species

SCD1

stearoyl-CoA desaturase 1

SCN

suprachiasmatic nucleus

SNPs

single nucleotide polymorphisms

SREBP

sterol regulatory element-binding protein

TAA

thioacetamide

TERT

telomerase reverse transcriptase

TGF-β1

transforming growth factor beta 1

TNFα

tumor necrosis factor alpha

TRAIL

tumor necrosis factor-related apoptosis-inducing ligand

TP53

tumor protein p53

VEGF

vascular endothelial growth factor

VPA

valproic acid

ZT

zeitgeber time