Possible application of melatonin treatment in human diseases of the biliary tract

Abstract

Melatonin was discovered in 1958 by Aaron Lerner. Its name comes from the ability of melatonin to change the shape of amphibian melanophores from stellate to roundish. Starting from the 1980s, the role of melatonin in the regulation of mammalian circadian and seasonal clocks has been elucidated. Presently, several other effects have been identified in different organs. For example, the beneficial effects of melatonin in models of liver damage have been described. This review gives first a general background on experimental and clinical data on the use of melatonin in liver damage. The second part of the review focuses on the findings related to the role of melatonin in biliary functions, suggesting a possible use of melatonin therapy in human diseases of the biliary tree.

MELATONIN SYNTHESIS AND EXCRETION

Melatonin (N-acetyl-5-methoxytryptamine) is a hormone found in animals, plants, and microbes (62). In mammals, melatonin is synthesized from the amino acid tryptophan by the pineal gland as well as peripheral organs, including the intestine and liver (72). The biosynthesis of serotonin and melatonin share some common pathways starting from tryptophan. The specific steps in melatonin synthesis, together with the corresponding enzymes involved, are summarized in Fig. 1. Tryptophan hydroxylase catalyzes the rate-limiting step of serotonin synthesis, the oxidation of tryptophan to 5-hydroxy-tryptophan. Then, aromatic l-amino acid decarboxylase catalyzes the decarboxylation of 5-hydroxy-l-tryptophan to generate serotonin. The enzyme aralkylamine N-acetyltransferase (AANAT) catalyzes the N-acetylation of serotonin to N-acetyl serotonin, the rate-limiting step in the synthesis of melatonin. Finally, the enzyme, acetyl serotonin O-methyltransferase (ASMT) catalyzes the last reaction in melatonin synthesis. AANAT is the enzyme that regulates the circadian rhythm of melatonin synthesis by the pineal gland (19). Indeed, the diurnal fluctuation of AANAT expression is related to the diurnal melatonin synthesis in vertebrates (67). Because the pineal storage of melatonin is undefined, melatonin serum levels are considered indices of the dynamic synthesis of melatonin by the pineal gland (64). Enhanced melatonin secretion has been demonstrated during night hours (peaking between 3 AM and 4 AM), whereas serum melatonin levels are virtually undetectable during light hours; nocturnal melatonin blood levels range between 10 and 80 µg/night. After secretion, melatonin diffuses easily in both aqueous and lipidic phases and circulates in blood bound to albumin for 70% of its total amount. Circulating melatonin easily reaches all body tissues and can cross the blood-brain barrier, as demonstrated by a positron emission tomography study (47).

Fig. 1.

Synthesis of melatonin from tryptophan. The specific steps are reported together with the corresponding enzymes (boxes) involved.

Liver is the main organ involved in melatonin catabolism (90%) through the typical glucuronidation-sulfation pathways. Elimination occurs by urine as either sulfated or unchanged in small amounts (26). Rhythmic melatonin synthesis is sustained by fibers located in the hypothalamus at the level of suprachiasmatic nuclei; the light-dark cycle is the main determinant of melatonin cyclic secretion. Information regarding light- or dark-regulated melatonin synthesis is transmitted to the hypothalamus by the retino-hypothalamic fibers. In support of this view, the exposure to significant artificial light during darkness has been demonstrated to reduce or completely abolish the secretion of melatonin according to the intensity of the illumination. The dark-light cycle of melatonin may stimulate other regions of the central nervous system (CNS), giving information on the 24-h shift. Melatonin synthesis has also been demonstrated in several extrapineal sites, including retina, brain, skin, and the gastrointestinal tract, including the liver (1, 65, 75).

GENERAL MELATONIN PHYSIOLOGICAL EFFECTS

Melatonin regulates several seasonal endocrine and reproductive activities in mammals. However, at least in humans, the main function of this hormone is associated with the regulation of circadian clock with a direct link to the CNS (27). Two subtypes of melatonin receptors have been identified and cloned in mammalian tissues (MT1 and MT2) (21), whereas an MT3 receptor has been identified in amphibian (70). Melatonin receptors have been identified in the nervous (21) and vascular tissue (22), liver (34, 66, 75), and lung (58). Receptor expression seems to be regulated by circulating melatonin levels and circadian cycle (32). In humans, melatonin is beneficial in several sleep disorders, including difficult sleep initiation (69) and shift work sleep impairment (9), as well as blindness (7). Also, reduced circulating levels of melatonin have been demonstrated in patients undergoing pinealectomy (49). Nevertheless, patients undergoing this type of surgery did not show different sleep impairment from individuals undergoing craniotomy. A “pinealoprive syndrome” has been linked to headache, mood, and vision disorders or even seizures in the affected subjects (18).

MELATONIN EFFECTS DURING LIVER DAMAGE

Several beneficial effects of melatonin have been described during liver injury. For a complete review of the intracellular mechanisms related to melatonin functions, see Ref. 78. In brief, as described in the following subparagraphs, the positive effects of melatonin have been observed in ischemia-reperfusion injury (IRI), nonalcoholic steatohepatitis (NASH), and alcoholic steatohepatitis, as well as in liver cancer models. The favorable properties of melatonin have been related to 1) inhibition of liver cell death by necrosis or apoptosis (44), 2) direct antioxidant actions (20), and 3) attenuation of mitochondrial damage (33).

Melatonin and IRI

With regard to IRI, melatonin administration has been shown to be beneficial in several organs (54). Melatonin exerts favorable effect against mitochondrial dysfunction, which is an important regulator in the onset of IRI. In fact, free radicals generated by IRI are responsible for respiratory chain damage and determine leakage of polar charge and enzymes by mitochondria, thus maintaining an extensive production of free radicals (81). In this setting, melatonin exerts a free radical scavenger activity upregulating the expression of the antioxidative enzymes catalase, glutathione peroxidase (GSH-Px), and superoxide dismutase (SOD) (25). The positive effects of melatonin on mitochondrial functions were also observed in the liver. In this study (60), rats were submitted to vascular clamp of liver vessels for 70 min (ischemia time), followed by 2 h of relapse of clamping (reperfusion time). In the treatment group, melatonin was administered (10 mg/kg body wt ip) 15 min before clamping and at reperfusion. By evaluation of isolated mitochondria, melatonin maintained an adequate ATP synthesis, reducing lipid peroxidation (after IRI) to values similar to those of control values. At the ultrastructural level, mitochondria coming from IRI rats were swollen with loss of cristae, whereas several mitochondria with a normal structure were seen in the melatonin-treated animals. With regard to the specific mechanisms associated with melatonin attenuation of liver IRI, one study in rodents demonstrated the role of c-Jun NH₂-terminal kinase (JNK) and IkB kinase-α (IKKα) (50). These signaling pathways were activated in the course of IRI by stimulation of tumor necrosis factor-α (TNF-α) hepatocyte receptor. In this setting, the increased activity of JNK and IKKα was reduced by melatonin treatment. Further studies (39, 41) linked enhanced Toll-like receptor 4 (TLR4) expression (due to stress conditions such as infection, inflammation, and ischemia) to melatonin effects on a similar model of liver IRI. TLR4 regulates IRI-induced liver damage trough myeloid differentiation factor 88 (MyD88)-dependent downstream activation of nuclear factor kB (NF-κB) and consequent release of proinflammatory cytokines. Melatonin administration (10 mg/kg body wt ip) inhibited TLR4 expression, enhancing the activity of heme oxygenase-1 (HO-1). Supporting this mechanism, the HO1 inhibitor Zinc protoporphyrin reversed the beneficial effect of melatonin, suggesting a HO1-dependent melatonin modulation of TLR4.

A study has evaluated the effects of melatonin in the setting of human liver IRI (15). Specifically, because liver damage is characterized by delayed apoptosis of neutrophils (38), the possible effects of melatonin on neutrophils from resected livers were evaluated. In this study (by evaluation of apoptosis by flow-cytometric assessment of DNA breaks), melatonin restored a normal level of apoptosis in neutrophils, thus suggesting a reduced neutrophilic activity as a possible mechanism for reducing IRI damage in humans. On the basis of these findings, a double-blind clinical study evaluated the possible clinical effects of a single dose of melatonin (50 mg/kg body wt) on outcome after major liver resection in humans (59). The study lacked mechanistic insights, but demonstrated 1) complete safety and good absorption for the administered dose and 2) a trend for improved liver enzymes, reduced intensive care unit, and hospitalization. These findings, however, did not reach the statistical significance, which was likely due to the limited number of patients included in the trial (n = 50).

Melatonin and NASH

NASH is defined in humans as the occurrence of >5% of hepatic steatosis in association with liver inflammation and hepatocyte injury in the absence of other known causes of liver diseases or alcohol abuse (12). This condition, which is frequently related to metabolic alterations, including diabetes and obesity, is gaining importance as a cause of cirrhosis in humans and is becoming one of the major indications for liver transplantation in US (73). Melatonin effects were studied in two rat models of NASH. The first study evaluated the effect of melatonin (2.5, 5, or 10 mg/kg body wt daily ip) in animals on a high-fat (10% fat) diet (61) and demonstrated reduced levels of cholesterol and triglycerides in liver homogenates of melatonin-treated animals compared with controls. In addition, the levels of the antioxidant enzymes SOD and GSH-Px were increased by melatonin treatment. By hematoxylin and eosin staining, melatonin treatment (5 or 10 mg/kg body wt) decreased liver steatosis in comparison with control animals (animals with severe steatosis: 0% with melatonin vs. 60% control, P < 0.01). In the second study, NASH was induced in rats by a methionine- and choline-deficient diet (71). Melatonin that was administered at a higher dose (50 mg·kg body wt⁻¹·day⁻¹ compared with the aforementioned study) (61) 1) significantly reduced serum levels of liver enzymes and inflammatory cytokines [interleukin-1(IL-1)β, interleukin-6 (IL-6), and tumor necrosis factor-α (TNFα)], 2) restored the concentration of glutathione and superoxide dismutase at values similar to those observed in rats under normal diet, and 3) reduced cellular apoptosis evaluated by DNA fragmentation on liver sections. Other studies (11, 30, 31) described the treatment of NASH patients with melatonin and demonstrated that 1) a 5-mg melatonin tablet twice a day was safe, 2) melatonin administration, coupled with a physical and dietary treatment, reduced aspartate transaminase (AST), alanine transaminase (ALT), and γ-glutamyl transferase (GGT) levels compared with diet and physical exercise only, and 3) melatonin treatment decreased the serum levels of the inflammatory cytokines IL1, IL6, and TNFα after 14 mo of treatment. In addition, melatonin improved some metabolic parameters such as insulin, adiponectin, and leptin blood levels (29). However, all of these studies, even if they represent an encouraging advancement, did not show a marked effect of melatonin on the parameters observed. For instance 1) the number of patients displaying normal liver transaminases accounted for only 10–20% of the total number of patients, 2) the decrease of inflammatory cytokines was not corroborated by histological parameters of liver inflammation, and 3) changes in metabolic parameters were not evaluated in long-term treatment. Furthermore, all studies related to the potential therapeutic effects of melatonin on NASH in rodents and humans did not pinpoint the possible molecular mechanisms by which melatonin protects against NASH but rather, focused only on the general antioxidant and cytoprotective properties of melatonin in this setting.

Melatonin and Toxic Hepatitis

The potential therapeutic effects of melatonin on the prevention of toxin-related and sepsis-mediated hepatic damage were summarized in a recent review article (23). In a rat model of liver injury induced by arsenic trioxide (As²O³) melatonin-ameliorated liver inflammation and serum chemistry, the effects were associated with enhanced nuclear factor erythroid-related factor 2 (Nrf2) and OH1 [member of antioxidant response elements (AREs)] expression through activation of the phosphatidylinositol 3-kinases/protein kinase B (PI3K/Akt) pathway (79). Another study has also demonstrated the protective role of melatonin against cadmium (Cd)-induced hepatotoxicity in the tumor cell line Hep-G2 (63). The study focused on the inhibitory role of melatonin on mitochondrial-derived O₂-stimulated autophagic cell death, which was enhanced during Cd-induced liver toxicity through a transduction pathway involving the sirtuin3-SOD2-mROS axis. The limitation of this study is based on the use of Hep-G2 cells rather than normal hepatocytes. A clinical study evaluated the possible role of melatonin in toxic hepatitis in humans, enrolling subjects with statin-induced liver damage (17). In this study, 60 subjects (with increased serum liver enzymes) who underwent therapy with statins were randomized to placebo or melatonin (5 mg twice/day for 6 mo) treatment. After 6 mo of melatonin treatment, liver enzyme levels were decreased (nearly a 40% decrease in comparison with baseline values, 25% normalized their levels vs. no change in controls), suggesting a protective effect of melatonin for toxic-induced human liver damage.

Melatonin and Alcoholic Hepatitis

Melatonin effects have been assessed in experimental models of alcoholic hepatitis. For example, mice fed a liquid diet with ethanol (5%) developed liver injury characterized by increased liver enzymes as well as hepatic steatosis, necrosis, and inflammation. In this setting, melatonin (from 5 to 20 mg/kg body wt daily by gavage) improved liver damage by decreasing the total hepatic content of triglycerides and levels of inflammatory cytokines such as TNFα, IL-6, and IL1-β (37). In another model, alcoholic hepatitis was studied as a function of matrix metalloproteinase (MMP) activity. MMPs are responsible for the degradation of extracellular matrix protein during damage, as MMP-9 overexpression has been advocated as a possible mechanism of damage in alcoholic hepatitis (43). In rats, the administration of ethanol (a variable dose of 2–8 ml/kg body wt, 50% ethanol twice/day for 3 days), increased 1) MMP-9 activity (8-fold increase in liver tissue and 5-fold increase in serum at the maximum ethanol exposure) and 2) the levels of the pro-inflammatory cytokines TNFα, IL-1β, and IL6 compared with control-treated rats. When rats were treated with melatonin (15 mg/kg body wt ip twice a day for 3 days) before the induction of alcoholic damage, the levels of MMP-9 (in both liver and serum) and metallopeptidase inhibitor 1 (that was decreased in rats exposed to alcohol) returned to values similar to those of control rats (56).

Melatonin and Hepatocellular Carcinoma

The protective effects of melatonin on hepatocellular carcinoma (HCC) are summarized in a recent review (57). HCC progression is in part related to the capacity of tumoral cells to inhibit apoptosis through the synthesis of specific inhibitors of apoptosis proteins (IAPs) (24) that downregulate the activity of caspases, specialized proteases required for cellular apoptosis. Human HCC tissue specimens displayed enhanced immunoreactivity of the IAPs, XIAP, cIAP-1, cIAP-2, and survivin, suggesting that these moieties may play a role in the survival of tumor cells. In neoplastic HepG2 and SMMC-7721 cell lines (treated in vitro with melatonin, 10⁻³–10⁻⁵ mol/L), there was enhanced endoplasmic reticulum stress-induced apoptosis, whereas the protein expression of XIAP and survivin significantly decreased compared with control cells (77).

In a clinical study focusing on the treatment with IL-2 (3 million UI/day sc for 4 wk) plus melatonin (50 mg/day, oral administration, starting 1 wk before IL-2) for advanced gastrointestinal cancers, six subjects underwent the treatment for HCC (52). This type of tumor was the one showing the best complete response rate (17%), having one subject with complete healing from the neoplasm after treatment. In another study (76), 100 patients with unresectable HCC were randomly assigned to receive treatment with trans arterial chemo embolization (TACE) or TACE plus melatonin (20 mg/day on the 7 days before procedure). The results demonstrated improved survival (3-yr survival, TACE vs. TACE + melatonin = 26 vs. 40%, P < 0.05) and reduced liver damage after the procedure, as demonstrated by the decrease of liver enzymes in the melatonin group after TACE. On the basis of these results on HCC and other hepatic diseases, a recent review suggested melatonin diet supplementation as a good strategy to prevent and treat liver diseases (8). Despite the fact that few clinical data are available, the antioxidant and mitochondria-preserving effects of melatonin would be considered a potential therapeutic approach for managing human liver diseases. Large-scale, well-designed, randomized trials are needed to validate this hypothesis. Main results and references regarding studies on melatonin and liver injury are reported in Table 1. Figure 2 summarizes some mechanisms of a melatonin-positive effect in different models of liver injury, as described in the previous paragraphs.

Table 1.

Experimental and clinical use of melatonin in liver injury

Type of liver damage (Ref. No).	Model	Melatonin Treatment (Route)	Results
IRI
60	Rats (vessel clamp)	10 mg/kg body wt ip	ATP synthesis maintained; ↓lipid peroxidation
50	Rats (vessel clamp)	50 mg/kg body wt (gavage)	↓Necrosis/inflammation; ↓JNK and IKKα
40	Rats (vessel clamp)	10 mg/kg body wt ip	↓Liver enzymes; ↓TLR response
41	Rats (vessel clamp)	10 mg/kg body wt ip	↓Liver enzymes; ↑HO-1; ↓TLR4-MyD88
59	Humans with liver resection	50 mg/kg po	↓Liver enzymes; ↓ICU stay; ↓hospital stay
NASH
61	Rats (high-fat diet)	2.5, 5 or 10 mg/kg body wt daily ip	↓Lipid accumulation;↑SOD and GSH-Px
71	Rats (methionine/cholin-deficient diet)	50 mg·kg body wt−1·day−1 ip	↓Liver enzyme; ↓inflammatory cytokines; ↑SOD; ↓apoptosis
30	Humans with NASH	10 mg/day po	↓Liver enzymes
Toxic damage
79	Rats (As2O3)	20 mg/kg body wt ip	↓Histological inflammation; ↓liver enzymes; ↑PI3K/AKT-Nrf2-AREs
63	Hep-G2 cells (Cd)	1 µM	Regulation of autophagy by Sirt3-SOD2 pathway
17	Humans (statins)	5 mg twice/day po	↓Liver enzymes
Alcoholic
37	Mice (5% ethanol)	5−20 mg/kg daily (gavage)	↓Lipid accumulation; ↓inflammatory cytokines; ↓liver inflammation, steatosis, and apoptosis
56	Rats (4–16 ml/kg body wt, 3 days)	60 mg/kg ip (3 days)	↓Inflammatory cytokines; ↑TIMP-1; ↓MMP-9
HCC
24	HepG2 and SMMC-7721 cells	10−3–10−5 M	↓XIAP and survivin; ↑apoptosis
77	HepG2 cells	10−3 M	↓ER stress and COX2; ↑apoptosis
76	Humans on TACE treatment	20 mg/day po (7 days before procedure)	↓Liver damage; ↑survival

AREs, antioxidant response elements; As²O³, arsenic trioxide; Akt, protein-kinase B; ATP, adenosine triphosphate; Cd, cadmium; COX2, cyclooxygenase-2; ER, endoplasmic reticulum; GSH-Px, glutathione peroxidase; HO-1, heme oxygenase 1; ICU, intensive care unit; IKKα, IkB kinase-α; ip, intraperitoneally; IRI, ischemia-reperfusion injury; JNK, c-Jun NH₂-terminal kinase; MMP, matrix metalloproteinase; MyD88, myeloid differentiation factor 88; Nrf2, nuclear factor erythroid 2-related factor 2; PI3K, phosphoinositide 3-kinases; po, per os; Sirt3, sirtuin 3; SOD, superoxide dismutase; TACE, trans arterial chemo embolization; TIMP-1, tissue inhibitor of metalloproteinase 1; TLR-4, Toll-like receptor 4; XIAP, X-linked inhibitor of apoptosis protein; ↑Increase; ↓decrease.

Fig. 2.

Mechanisms of melatonin-positive effect in different models of liver injury. A: melatonin counteracts ischemia-reperfusion injury (IRI) damage both by decreasing the activity of JNK and IkB kinase-α (IKKα) and by direct repression of Toll-like receptor 4 (TLR4) synthesis. B: melatonin reduces arsenic trioxide (As²O³) toxicity stimulation of phosphatidylinositol 3-kinase (PI3K)/protein-chinasi B (Akt) pathway and protects against cadmium (Cd) injury through superoxide dismutase 2 (SOD2) removal of mitochondrial reactive oxygen species (mROS). C: alcohol/matrix metalloproteinase-9 (MMP-9)-mediated cell damage is reversed by melatonin. For details, see text. JNK, c-Jun NH₂-terminal kinase.

MELATONIN EFFECTS ON CHOLANGIOCYTES AND ITS POSSIBLE THERAPEUTIC USE IN HUMAN CHOLANGIOPATHIES

The biliary epithelium is lined by cholangiocytes of different sizes and functions. Studies on rodents identified different subpopulations of cholangiocytes: small (mean diameter ∼8.4 µm) and large (mean diameter ∼14.5 µm) lining small and larger bile ducts, respectively (3). The difference in cholangiocyte size is closely related to functionality diversity, since large 3′,5′-cyclic adenosine monophosphate (cAMP)-dependent cholangiocytes are more differentiated and participate in the secretion of water and bicarbonate (3), whereas small, undifferentiated inositol trisphosphate/Ca²⁺-cholangiocytes are considered a quiescent progenitor subpopulation of cholangiocytes (more resistant to liver injury) that differentiate into large cholangiocytes when these cells are damaged (48). In normal conditions, the most important function of large cholangiocytes is to support the so-called bile acid-independent bile flow, which occurs through the interaction of secretin with a specific secretin receptor (SR; expressed only by cholangiocytes) (5) located on the basolateral membrane of cholangiocytes (2). The interaction of secretin with SR induces an increase in intracellular cAMP levels, activation of cystic fibrosis transmembrane conductance regulator (CFTR) with extracellular extrusion of Cl⁻, and subsequently, activation of the Cl⁻/$HC{O}_3^{–}$ exchanger AE2 stimulating a bicarbonate-rich choleresis while Cl⁻ is recovered back in cholangiocytes (2).

Extensive examination of the molecular mechanisms related to the interplay between melatonin and the biliary epithelium is summarized in a recent review (28). In an early study, the authors compared the effects of S-adenosyl-methionine (SAME; 10 mg·kg body wt⁻¹·day⁻¹) or melatonin (750 µg·kg body wt⁻¹·day⁻¹) administration (both ip for 10 days) on liver functions in the cholestatic model of bile duct ligation (BDL) (53). The antioxidants and hepatoprotective effects of melatonin were superior to those of SAME since melatonin decreased the levels of liver enzymes and malondialdehyde and glutathione at higher rate compared with SAME, levels that were similar to those of sham-operated control rats. We recently extended these findings demonstrating the expression of melatonin receptor subtypes (MT1 and MT2) in bile ducts (by immunohistochemistry in liver sections) and isolated cholangiocytes (by real-time PCR and FACS analysis) with MT1 expression that increased in cholestatic rodents (66). Treatment of BDL rats with melatonin in drinking water (20 mg/L, estimated assumption of 2 mg·g body wt⁻¹·day⁻¹) decreased ductular reaction (DR), serum bilirubin and transaminases levels, the expression of clock genes, cAMP levels, and protein kinase A (PKA) phosphorylation in cholangiocytes by interaction with MT1 (66).

In another study, we examined 1) the expression of AANAT (the limiting-step enzyme for melatonin synthesis) in normal and proliferating cholangiocytes and 2) the effects of AANAT biliary downregulation on changes in DR and ductal secretory activity (65). By immunohistochemistry in liver sections and real-time PCR in isolated cholangiocytes, the study demonstrated the immunoreactivity/expression of AANAT and secretion of melatonin in cholangiocytes, parameters that increased following BDL; minimal expression of AANAT was detected in hepatocytes (65). Downregulation of biliary AANAT (by administration of in vivo AANAT Morpholino) was associated with enhanced DR in liver sections and increased SR, CFTR, and Cl⁻/$HC{O}_3^{–}$ AE2 expression (65), functional indices of biliary hyperplasia (4, 5).

In a recent study, the role of the melatonin brain-liver axis was evaluated by exposing cholestatic BDL rats and Mdr2^−/− mice to total darkness, a condition that increases melatonin secretion from pineal gland (36, 75). When BDL rats were exposed to complete darkness for 1 wk, there was 1) enhanced expression of AANAT in the pineal gland and melatonin serum levels, 2) improved liver morphology, serum levels of liver enzymes, and reduced DR, and 3) decreased deposition of collagen as well as biliary expression of the clock genes, PER1, BMAL1, CLOCK, and CRY1 (36). When Mdr2^−/− mice [that mimic some of the features of human primary sclerosing cholangitis (PSC)] were exposed to total darkness for 1 wk, there were higher serum melatonin levels and reduced DR, collagen deposition, and angiogenesis compared with Mdr2^−/− mice exposed to 12:12-h dark-light cycles (75). The study also demonstrated enhanced expression of miR-200b in both Mdr2^−/− mice and human PSC samples, expression that was reduced in Mdr2^−/− mice subjected to dark exposure or melatonin treatment (75). Also, by in vivo and in vitro downregulation of miR-200b in Mdr2^−/− mice and human biliary cells, respectively, there was reduced DR, collagen deposition, and angiogenesis in liver sections from Mdr2^−/− mice and angiogenesis and fibrosis mRNA expression in biliary lines (75). The role of the melatonin brain-liver axis was further evaluated in BDL cholestatic rats undergoing pinealectomy or prolonged light exposure, maneuvers that reduce melatonin secretion from pineal gland as well as peripheral organs (16). In BDL rats plus pinealectomy or prolonged light exposure, there were increased levels of liver enzymes serum chemistry, ductular reaction, biliary senescence, liver fibrosis, inflammation, angiogenesis, reactive oxygen species (ROS) generation, and expression of miR-200b (which is increased in cholestatic cholangiocytes) compared with BDL rats exposed to 12:12-h light-dark cycles (16). Another study (55) has shown that intracerebroventricular infusion of melatonin to BDL rats reduces ductular reaction and liver fibrosis through inhibition of expression/secretion of hypothalamic gonadotropin-releasing hormone release (GnRH) from cholangiocytes and reduced expression of its receptor (GnRHR); the GnRH-GnRHR axis has been shown to stimulate biliary proliferation and liver fibrosis (45). Enhanced bioavailability of melatonin in the brain may improve the outcome of cholestatic liver diseases. The main results and references of studies regarding melatonin and the biliary tree are summarized in Table 2. A chronological timeline of the major discoveries with regard to melatonin effect on the biliary tract is depicted in Fig. 2.

Table 2.

Results of main experimental studies of melatonin effects on the biliary tree

Ref. No.	Model	Approaches to Study Melatonin Effect	Results
53	BDL rat	Intraperitoneal administration 750 µg·kg−1·day−1	↓Liver enzymes; preserved levels of malondialdehyde and glutathione
66	Normal and BDL rats	Per os administration: 2 mg/g day in water	Identification of MT1 and MT2 receptor on cholangiocytes; melatonin in BDL rat determines: ↓MT1 and MT2 expression, ↓proliferation, ↓secretin stimulated bile flow, and↓intracellular cAMP
65	Normal and BDL rats	Administration: 10−11 M on isolated cells	Identification of AANAT expression in cholangiocytes; melatonin exposure increases AANAT expression in BDL cholangiocytes; identification of melatonin synthesis by BDL cholangiocytes (autocrine loop); AANAT downregulation in BDL induces: ↑proliferation and ↑secretin stimulated bile flow
36	Normal and BDL rats	Prolonged darkness exposure (1 wk)	↑AANAT and melatonin expression in pineal gland; in BDL rats: ↓proliferation, ↓secretin stimulated bile flow
75	Mdr2−/− mice	Prolonged darkness or per os administration: 2 mg·g body wt−1·day−1	↓Fibrosis; ↓expression fibrosis genes; ↓vascular endothelial growth factor A/C; ↓angiopoietin ½; micro RNA-200b
16	Normal and BDL rats	Pinealectomy or prolonged light exposure	↑Fibrosis; ↑proliferation; ↑microRNA-200b

AANAT, aralkylamine N-acetyltransferase; BDL, bile duct ligated; cAMP, cyclic adenosine monophosphate; MT1, melatonin receptor 1A; MT2, melatonin receptor 1B. ↑Increase; ↓decrease.

Cholangiocytes are the target of cholangiopathies, including primary biliary cholangitis (PBC), PSC, and cholangiocarcinoma (CCA), diseases that are characterized by biliary damage/senescence, liver inflammation, and fibrosis (13, 51). Currently, several therapeutic options are evaluated for these diseases, including immunological approaches or use of stem cells (for details we, refer to a recent review in Ref. 14); however, new therapeutic approaches are needed since these cholangiopathies still represent an important cause of liver decompensation and death. A number of experimental studies demonstrated that melatonin might be beneficial for the management of chronic cholestatic liver diseases, as it 1) regulates biliary homeostasis and 2) decreases collagen deposition in the liver. In both PBC and PSC, biliary proliferative activity is important for disease progression. In fact, the increase in the number of bile ducts [the so called ductular reaction (DR)] is a common finding in these diseases (68). DR in PBC and PSC is characterized by functionally ineffective, truncated bile ducts expanding through portal areas in the parenchymal region (6). A recent study in human liver sections comparatively evaluated DR in patients with PBC or PSC (10). In both diseases, there was a statistically significant linear relationship between DR and extent of fibrosis with a correlation coefficient of 0.6 (P < 0.01). The concept that changes in DR regulate the activation of hepatic stellate cells and liver fibrosis is supported by several studies (42, 68, 80). Downregulation of DR (for example, by blockage of the secretin-SR axis) was associated with reduced liver fibrosis mediated by decreased secretion of biliary senescence-associated secretory phenotypes (SASPs; such as TGFβ1), factors that activate hepatic stellate cells (42, 80); upregulation of the secretin-SR-TGFβ1 axis was also seen in liver sections of PSC patients (75). Another study has shown that BDL-induced increases in 1) serum enzyme levels, 2) liver inflammation and ROS levels, 3) DR and liver fibrosis, and 4) liver angiogenesis were exacerbated by both pinealectomy and prolonged exposure to light, maneuvers that decrease melatonin levels (16). The effects of pinealectomy and prolonged exposure to light on these phenotypes were associated with 1) enhanced expression of TGFβ1 and biliary senescence (42, 74) and 2) increased expression of the clock genes CLOCK, ARNTL, CRY1, and PER1 and miR-200b, which were reduced by the administration of melatonin. From these findings, it is apparent that melatonin is able to coordinately modulate key phenotypes of cholangiopathies such as liver inflammation, DR, cellular senescence, and liver fibrosis.

There is growing information with regard to the role of melatonin in the growth of CCA (34, 35). A recent study demonstrated the antiproliferative effects of melatonin on CCA using six different CCA cell lines (Mz-ChA-1, HuH-28, TFK-1, CCLP1, SG231, and HUCC-T1 and the normal human cholangiocyte line H69) and male BALB/c nude mice with CCA established by injection of Mz-ChA-1 cells (34). The study demonstrated an autocrine loop by which the ASMT-AANAT-melatonin-MT1 axis inhibits CCA growth. Specifically, there was decreased immunoreactivity (in sections from liver biopsies) and expression (by FACS analysis and real-time PCR) of ASMT/AANAT but enhanced MT1 expression in CCA tissue/lines compared with normal controls; there were decreased melatonin bile (but not serum) levels in human samples. In the same study, in BALB/c nude mice (with established CCA) treated for 34 days with melatonin (4 mg/kg body wt ip daily injections), there was a significant decrease of tumor volume that was coupled with the enhanced number of apoptotic cholangiocytes. Another study demonstrated that the rhythmic expression of core clock genes (modulated by melatonin) was disrupted in CCA cell lines, since a marked decreased of the clock gene PER1 was observed in liver biopsies and human CCA lines (35). Overexpression of PER1 was coupled with decreased CCA proliferation but enhanced biliary apoptosis both in vitro in CCA cell lines and in vivo in athymic mice. The study also demonstrated that PER1 is a target of miR-34a, since inhibition of miR-34a (overexpressed in CCA) reduced the proliferation and invasiveness of CCA cells compared with normal controls (35). Another study evaluated the effect of melatonin on the apoptosis of the human CCA cell lines KKU-M055 and KKU-M214 that were treated with melatonin (0.5–2 mM for 48 h). In these CCA cell lines, melatonin inhibited cell viability and increased intracellular ROS levels, leading to increased oxidative DNA damage and 8-oxodG formation (46). The main biological determinants of human biliary disease are reported in Figs. 3 and 4 together with the possible melatonin modulatory effects.

Fig. 3.

Time course of experimental findings on melatonin and the biliary tree. Chronological timeline of the major discoveries with regard to the effect of melatonin effect on the biliary tract is reported. BDL, bile duct ligated; MT1, melatonin receptor 1; AANAT, aralkylamine N-acetyltransferase.

Fig. 4.

Biological determinants of main human biliary diseases and the corresponding findings regarding melatonin. The main biological determinants of primary biliary cholangitis (PBC), primary sclerosing cholangitis (PSC), and cholangiocarcinoma (CCA) are reported together with the respective melatonin experimental data, suggesting a possible therapeutic application. For details see text. AANAT, aralkylamine N-acetyltransferase; ASMT, acetyl serotonin O-methyltransferase.

CONCLUSIONS

Several important effects of melatonin have been identified in the last decades. Original studies focusing on beneficial effects on sleep disorders have been implemented by the observation that melatonin is favorable in several pathological conditions involving different organs. Regarding the liver, melatonin determines improvement in several experimental models of damage. Positive effects are observed after toxic, ischemic, and oxidative hepatic injury. Our knowledge on melatonin is now extending on its interaction with specific cells within the liver, such as cholangiocytes. These cells are the main target of cholestatic chronic or neoplastic liver diseases in humans, effects that still are in waiting for a conclusive therapy. Experimental results demonstrate advantages in the use of melatonin in models of damage involving the biliary tree, allowing for speculation on the possible application of this hormone in human therapy. Well-designed clinical studies in the future will address this issue.

GRANTS

This work was supported by the Hickam Endowed Chair, Gastroenterology, Medicine, Indiana University, the Veterans Affairs Merit Awards to G. Alpini (5I01BX000574), H. Francis (1I01BX003031) and F. Meng (1I01BX001724) from the US Department of Veterans Affairs, Biomedical Laboratory Research and Development Service, and National Institutes of Health Grants DK-108959 (H. Francis), AA-026385 (Z. Yang), DK-054811, DK-076898, DK-107310, DK-110035, DK-062975, AA-025997, and AA-025157 (G. Alpini, S. Glaser, and F. Meng) and a grant award from PSC Partners Seeking a Cure (G. Alpini).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

L.B. prepared figures; L.B., T.Z., L.I., P.K., Z.Y., L.C., and H.F. drafted manuscript; L.B., T.Z., S.L., L.I., M.M., F.M., L.K., P.K., Z.Y., L.C., S.G., H.F., and G.A. edited and revised manuscript; S.L., H.F., and G.A. approved final version of manuscript.