Hypercholesterolemia increases and exacerbates stress signals leading also to liver damage (LD) and failure. Sirtuin1 (SIRT1) is involved in lifespan extension and it plays an essential role in hepatic lipid metabolism. However, its involvement in liver hypercholesterolemic damage is not yet completely defined. This
Hyperlipidemia, which included hypercholesterolemia (HC), is the most striking risk factor in the development of cardiovascular diseases in occidental population [
Recently, it has been reported that sirtuins, belonging to silent information regulator 2 family, play a key role in the development and rescue of various metabolic diseases, including NAFLD [
It is known that the inhibition of SIRT1 signaling in human fetal hepatocytes induces an increase in intracellular glucose and lipid levels with upregulation of de novo lipogenesis and gluconeogenesis related genes [
Currently available therapies for controlling hyperlipidemia, such as fibrates, bile acid sequestraints, and statins, are almost inefficient in lipid metabolism regulation and cause different side effects in patients [
Melatonin is an indoleamine produced and secreted into blood stream during dark period by the pineal gland. Its physiological effects include regulation of seasonal reproduction, body weight, and energy balance [
Since the mechanisms involved in HC-related LD are not fully understood, the aim of this study was to better investigate liver morphological alterations during HC. Then, we evaluated if melatonin is effective in reducing LD and OS through also the induction of SIRT1 expression. To address this issue, we used ApoE−/− mice, an animal model that spontaneously develops HC, aortic lipid accumulation, and LD in a time-dependent manner. As previously shown, ApoE−/− mice develop HC from the 4th week of life and then showed alterations in hepatic morphology and protein expressions [
In the present study, we provided evidence that melatonin restores liver cytoarchitecture preventing SIRT1 reduction. Furthermore, SIRT1 is able to decrease OS inducing also antioxidants expression. Then, we suggested that melatonin may be a valuable protective alternative strategy for minimizing the LD related to HC, via also a SIRT1-dependent mechanism of action.
Forty C57BL/6 male mice and thirty ApoE−/− male mice (Harlan Laboratories S.r.l., Udine, Italy and Charles River Laboratories S.r.l., Lecco, Italy) were housed in an animal experimental unit with 12 h alternating light–dark cycle and constant temperature. Animal had free access to food and water. All the protocols were approved by the Italian Ministry of Health and complied with “Guiding Principles in the Use of Animals in Toxicology,” which were adopted by the Society of Toxicology in 1989.
Mice were randomly divided into the following seven groups (ten animals per group): (1) control untreated C57BL/6 mice which were 6 weeks old at sacrifice; (2) control untreated C57BL/6 mice which were 15 weeks old at sacrifice; (3) control C57BL/6 mice treated orally with 1% ethanol (melatonin vehicle) dissolved in tap water from the 6th to 15th week of life; (4) control C57BL/6 mice treated orally with melatonin (10 mg/kg/day) from the 6th to 15th week of life; (5) ApoE−/− untreated mice which were 6 weeks old at sacrifice (ApoE−/− of 6 w); (6) ApoE−/− untreated mice which were 15 weeks old at sacrifice (ApoE−/− of 15 w); and (7) ApoE−/− mice treated orally with melatonin (10 mg/kg/day) from the 6th to 15th week of life (ApoE−/− + MEL).
Animals were individually housed in cages with a single water bottle to ensure that all received the correct melatonin dose according to the body weight of the animal. The bottles were wrapped in aluminium foil to protect melatonin from light. Synthetic melatonin was dissolved in 1% ethanol and then diluted in tap water to yield a final dose of 10 mg/kg body weight/day, as previously described by Rezzani et al. [
At the end of the study, all the animals were euthanized and blood and liver samples were collected. Serum cholesterol and triglyceride concentrations were determined by standard laboratory procedures and also serum SOD1 and glutathione (GSH) levels were evaluated, as described later. Furthermore, liver samples were weighted and then fixed in 4% paraformaldehyde and embedded in paraffin wax, according to standard procedures [
Liver paraffin-embedded sections were deparaffinized, rehydrated, and then stained with haematoxylin-eosin, following standard protocols. In detail, haematoxylin-eosin staining was used to evaluate liver morphology and also to measure the hepatocyte nuclear area [
Picrosirius red staining was used to evaluate liver fibrosis [
Liver paraffin-embedded sections were deparaffinized, rehydrated, and incubated for 1 hour at room temperature in specific normal serum. Subsequently, the sections were incubated overnight at 4°C with the following primary antibodies: rabbit polyclonal antibody against SIRT1 (diluted 1 : 100; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or rabbit polyclonal antibody against iNOS (diluted 1 : 200; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or simultaneously with goat antibody against SOD1 (diluted 1 : 100; Santa Cruz Biotechnology, Santa Cruz, CA, US) and rabbit polyclonal antibody against CAT (diluted 1 : 150; Santa Cruz Biotechnology, Santa Cruz, CA, US). Liver sections were then washed in tris-buffered saline (TBS) and, for immunohistochemical analysis of SIRT1 and iNOS, were incubated for 1 hour with specific biotinylated secondary antibody and after 1 hour with the avidin-biotin horseradish peroxidase complex (ABC-peroxidase kit Vector Labs, Burlingame, CA, USA), prepared according to the manufacturer’s instructions. Finally, these liver sections were immersed in a solution of 0.05% 3′-3′-diaminobenzidine tetrahydrochloride (DAB) and 0.03% hydrogen peroxide for 10 minutes, counterstained with haematoxylin, dehydrated, mounted, and observed with a light microscopy (Olympus, Hamburg, Germany) at a final magnification of 400x.
However, for the double immunofluorescence staining with SOD1 and CAT, the liver sections, after the incubation with primary antibodies, were labelled using anti-goat Alexa Fluor 546 and anti-rabbit Alexa Fluor 488 conjugated secondary antibodies (diluted 1 : 200; Invitrogen, UK). Finally, the sections were counterstained with 4′-6-diamidino-2-phenylindole (DAPI), mounted, and observed with a confocal microscope (510 Meta Zeiss, Munich, Germany) at a final magnification of 400x, as previously described by Agabiti-Rosei et al. [
Control reactions for both immunohistochemistry and immunofluorescence analysis were performed in absence of primary antibody and in the presence of isotype-matched IgGs.
Randomly chosen 20 liver fields for each experimental animal were analyzed and the immunostaining for each primary antibody was calculated using an image analyzer (Image Pro Premier 9.1, Media Cybernetics Inc., Rockville, USA). Two blinded investigators performed the histomorphometrical analysis and their evaluation was assumed to be correct if the values were not significantly different. If there was disagreement concerning the interpretation, the case was reconsidered to reach a unanimous agreement. The levels of immunostaining of each primary antibody evaluated were expressed as arbitrary units (AU).
Serum samples were obtained collecting total blood in serum separator tubes and allowed samples to clot for 2 hours at room temperature. Then the samples were centrifuged for 20 minutes at 1000 ×g. The supernatants obtained were subjected to analyses of the levels of SOD1 and GSH through specific ELISA assay kits and following the respective manufacturer’s instructions (LifeSpan BioSciences, Inc., Seattle, USA). In detail, the optical density values for both SOD1 and GSH were determined using a microplate reader set at 450 nm. Furthermore, SOD1 ELISA assay presented intra-assay variations coefficient less than 4.7%, interassay variations coefficient less than 9.5%, and sensitivity less than 1.56 U/mL, whereas the GSH ELISA assay had intra-assay variations coefficient less than 6.2%, interassay variations coefficient less than 9.6%, and sensitivity less than 1 ng/mL.
The data are expressed as mean ± SD. All data were analyzed by one-way analyses of variance for repeated measures (ANOVA), corrected by Bonferroni, to compare the variability of a group with all other experimental groups. Probability values less than 0.05 were considered significant. All experiments were carried out in triplicate and data were collected and analyzed by Origin Pro 9.1 software (OriginLab Corporation, Northampton, MA, USA).
The animal body weight measured before the start and at different time points of the treatment and also the liver weight measured at the end of the study did not show any significant difference among the experimental groups. Table
Liver/body weight ratio and serum parameters.
ApoE−/− |
ApoE−/− |
Control |
Control |
ApoE−/− plus melatonin | |
---|---|---|---|---|---|
Liver/body weights (%) | 5.4 | 5.5 | 5 | 5.2 | 5.1 |
Cholesterol (mg/dL) |
|
|
|
|
|
Triglycerides (mg/dL) |
|
|
|
|
|
As expected, ApoE−/− 6 w and 15 w had increased serum cholesterol and triglyceride levels compared to control mice. In particular, both serum cholesterol and triglyceride concentrations were higher in ApoE−/− 15 w with respect to ApoE−/− 6 w. Interestingly, both serum parameters were significantly reduced after melatonin treatment of ApoE−/− mice (Table
Haematoxylin-eosin staining showed that in untreated control C57BL/6 mice which were 6 weeks or 15 weeks old or in control C57BL/6 mice treated with 1% ethanol (vehicle of melatonin) or with melatonin presented a “normal” hepatic parenchyma without LD. These observations underline that all the above reported experimental groups are similar, so they are considered a single group defined generically as “control” in the following morphological, immunohistochemical/immunofluorescence, and histomorphometrical analysis.
In both untreated ApoE−/− 6 w and 15 w groups, we observed significant LD. In fact, both untreated groups presented hepatocyte ballooning, nuclear polyploidy, and granular cytoplasm with prevalently small lipid droplets deposition, characteristic of microvesicular steatosis, together with inflammatory cells infiltration. The reported hepatic parenchymal alterations were weak/moderate in ApoE−/− 6 w (Figure
Photomicrographs of liver haematoxylin-eosin staining of untreated ApoE−/− of 6 w (a), untreated ApoE−/− of 15 w (b), control (c), and ApoE−/− + MEL (d). Bar: 20
Furthermore, untreated ApoE−/− 15 w showed a significant increased hepatocyte nuclear area with respect to both ApoE−/− 6 w and control groups. Remarkably, liver of ApoE−/− + MEL exhibited a significant hepatocyte nuclear area reduction, reaching value comparable to liver of control mice (Figure
Picrosirius red staining showed a mild parenchymal hepatic perisinusoidal fibrosis in untreated ApoE−/− 6 w (Figure
Photomicrographs of liver Picrosirius red staining of untreated ApoE−/− of 6 w (a), untreated ApoE−/− of 15 w (b), control (c), and ApoE−/− + MEL (d) under polarized light (a–d) or without polarized light (inserts). Bar: 20
Immunohistochemical analysis of SIRT1 showed a very weak/weak expression at parenchymal liver level of both untreated ApoE−/− 6 w (Figure
Photomicrographs of liver SIRT1 immunohistochemical analysis of ApoE−/− of 6 w (a), ApoE−/− of 15 w (b), control (c), and ApoE−/− + MEL (d). Bar: 20
These observations were confirmed also by liver SIRT1 histomorphometrical analysis (Figure
Both untreated ApoE−/− 6 w and 15 w groups showed a significant iNOS expression at parenchymal liver level; in particular, liver of untreated ApoE−/− 6 w had a moderate iNOS expression (Figure
Photomicrographs of liver immunohistochemical analysis of iNOS (a–d) and of liver double immunofluorescence analysis of SOD1 (green staining) and CAT (red staining) (e–h) of ApoE−/− of 6 w (a, e), ApoE−/− of 15 w (b, f), control (c, g), and ApoE−/− + MEL (d, h). Bar: 20
The double immunofluorescence evaluation of endogenous antioxidants SOD1 (identified in green staining) and CAT (identified in red staining) showed that parenchymal liver of untreated ApoE−/− 6 w had a mild hepatic expression of SOD1 and a very weak/weak CAT expression (Figure
All the above reported observations were confirmed also by the histomorphometrical hepatic analysis of iNOS (Figure
Next to histomorphometrical analyses of SOD1 and CAT, we evaluated also the serum levels of the antioxidant enzymes SOD1 and GSH. ApoE−/− 15 w showed a significant reduction of both SOD1 and GSH levels with respect to control mice that showed strong/moderate serum levels. Remarkably, melatonin treatment of ApoE−/− mice showed a significant increase of both serum antioxidants (strong levels). Figure
Serum superoxide dismutase1 (SOD1) (a) and glutathione (GSH) (b) levels.
The LD observed in an ApoE−/− hypercholesterolemic mouse model were significantly reduced by oral supplementation of melatonin.
We observed a severe decrease of SIRT1 expression in liver of ApoE−/− mice, more evident at 15 weeks of life. Growing evidences from both human and experimental studies revealed that impaired SIRT1 signaling is associated with alcoholic liver disease and genetic or pharmacological stimulations of SIRT1 protect against steatosis and/or steatohepatitis [
To highlight the importance of hepatic SIRT1 expression and its involvement in OS, we evaluated the expression of iNOS, a proinflammatory protein commonly related to OS status, and of endogenous antioxidants (SOD1, CAT, and GSH). Our results showed that iNOS is strongly expressed in liver of ApoE−/− mice and this increase plays, in turn, a critical role in the development of liver inflammation and OS [
Remarkably, our results exhibited that the oral supplementation of melatonin restored SIRT1 hepatic expression that, in turn, had an antagonistic effect to LD and OS in ApoE−/− mice. These findings confirmed the literature data, reporting that melatonin had important antioxidant effects and also a strong tendency to attenuate hepatic steatosis [
Other experimental studies showed that melatonin alone or administered in combination with pioglitazone or pentoxifylline reduced the insulin resistance index and total cholesterol and triglycerides and modulated GSH level during NAFLD [
Our results suggest that oral supplementation of melatonin inhibits LD induced by HC and that not only could this effect be due to its known antioxidant properties, but also it may be mediated by a SIRT1-dependent mechanism that, in turn, blocks serum cholesterol and triglycerides increase and hepatic iNOS expression and restores endogenous antioxidants (Figure
Schematic representation of melatonin effects against hypercholesterolemic hepatic alterations. It is important to underline that this indoleamine, together with its antioxidant properties, is able to reduce liver morphological alterations also blocking the decrease of sirtuin1 that in turn inhibits oxidative stress. The red symbols indicate the effect of melatonin in blocking the pathways induced by hypercholesterolemia that are indicated by the blue arrows. CAT: catalase; GSH: glutathione; iNOS: inducible nitric oxide synthase; SIRT1: sirtuin1; SOD1: superoxide dismutase1.
The authors declare that there are no conflicts of interest regarding the publication of this manuscript.
Francesca Bonomini and Gaia Favero contributed equally to the present work.
Thanks are due to Lorena Giugno and Stefania Castrezzati for their technical support. This study was supported by the University of Brescia, Italy, Grant (ex 60%).