N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

Ava i l ab l e on l i ne a t www.sc i enced i r ec t . com

ScienceDirectwww.n r j ou rna l . com

Original Research

Melatonin reduces obesity and restores adipokine

patterns and metabolism in obese (ob/ob) mice

Gaia Faveroa, Alessandra Stacchiotti a, b, Stefania Castrezzati a, Francesca Bonomini a, b,Massimo Albanese c, Rita Rezzani a, b,⁎, Luigi Fabrizio Rodella a, b

a Anatomy and Physiopathology Division, Department of Clinical and Experimental Sciences, University of Brescia, 25123 Brescia, Italyb Interdipartimental University Center of Research “Adaption and Regeneration of Tissues and Organs-(ARTO)”c Department of Oral and Maxillofacial Surgery, University of Verona, Verona, Italy

A R T I C L E I N F O

Abbreviations: SAT, subcutaneous adipose⁎ Corresponding author at: Anatomy and Ph

Brescia, Viale Europa 11, 25123 Brescia, Italy.E-mail address: rita.rezzani@unibs.it (R. R

http://dx.doi.org/10.1016/j.nutres.2015.07.0010271-5317/© 2015 Elsevier Inc. All rights rese

A B S T R A C T

Article history:Received 17 March 2015Revised 28 May 2015Accepted 3 July 2015

The increasing incidence of obesity, leading to metabolic complications, is now recognizedas a major public health problem. The adipocytes are not merely energy-storing cells, butthey play crucial roles in the development of the so-called metabolic syndrome due to theadipocyte-derived bioactive factors such as adipokines, cytokines, and growth factors. Thedysregulated production and secretion of adipokines seen in obesity is linked to thepathogenesis of the metabolic disease processes. In this study, we hypothesized thatdietary melatonin administration would support an anti-inflammatory response and playan important role in energy metabolism in subcutaneous and visceral adipose tissues ofobese mice and so may counteract some of the disruptive effects of obesity. Lean and obesemice (ob/ob) received melatonin or vehicle in drinking water for 8 weeks. Thereafter, theywere evaluated formorphologic alteration, inflammatory cell infiltration, and the adipokinepatterns in visceral and subcutaneous white fat depots. In obese mice treated with vehicle,we observed a significant increase in fat depots, inflammation, and a dysregulation of theadipokine network. In particular, we measured a significant reduction of adiponectin andan increase of tumor necrosis factor α, resistin, and visfatin in adipose tissue deposits.These changes were partially reversed when melatonin was supplemented to obese mice.Melatonin supplementation by regulating inflammatory infiltration ameliorates obesity-induced adipokine alteration, whereas melatonin administration in lean mice wasunaffected. Thus, it is likely that melatonin would be provided in supplement form tocontrol some of the disruptive effects on the basis of obesity pathogenic process.

© 2015 Elsevier Inc. All rights reserved.

Keywords:AdipokineAdiponectinInflammationMelatoninob/ob miceResistinVisfatin

1. Introduction

Obesity is a medical condition potentially affecting all agesand socioeconomic groups [1]. Over the past decade, profound

tissue; TNF-α, tumor necysiopathology Division, DTel.: +39 0303717483; faxezzani).

rved.

changes in nutrition and lifestyle have led to a sharp increasein the prevalence of obesity and its complications. Actually, ithas reached epidemic proportions in many parts of the worldand this trend is expected to continue [1,2].

rosis α; VAT, visceral adipose tissue.epartment of Clinical and Experimental Sciences, University of: +39 0303717486.

892 N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

Obesity is characterized by a chronic inflammatory pro-cess, well evident in the intra-abdominal compartment(visceral adipose tissue, or VAT). It is known that an increaseof VAT is highly correlated with diseases, whereas anaugmentation of subcutaneous adipose tissue (SAT) hasminor metabolic consequences [3]. Although the exactmechanisms underlying the differential role of VAT vs SATin metabolic diseases have yet to be identified, many studiesindicate a higher inflammatory potential of VAT comparedwith SAT in terms of increased production of proinflamma-tory mediators and reduced expression of the protectiveadipokine. Patsouris et al [4] documented that the accumula-tion of proinflammatory macrophages in VAT of obesesubjects is linked to the development of insulin resistance.

Adipose tissue is a secretory and endocrine active organproducing a variety of bioactive proteins, defined asadipokines, that may regulate energy metabolism, foodintake, and insulin sensitivity [5]. Several studies haveshown that the production of some adipokines is altered inobesity, in type 2 diabetes, and in the metabolic syndrome [6],modulating adipocyte size/number and angiogenesis viaparacrine mechanisms and exerting a major role in theregulation of fat mass [7,8]. The increase in adipose tissuemass dysregulates both adipokine and adipocytokine secre-tion patterns [9], which play a pivotal roles in the pathogen-esis of cardiovascular damages through adverse effects onhemostatic balance and vascular function [7,10].

Among the adipokines, herein we report on the perturbedlevel of adiponectin, resistin, and visfatin in obesemice and theeffect of melatonin administration. Adiponectin is a proteinsynthesized in the adipocyte [11,12], and it is released by bothbrown and white adipose tissue [13]. Hypoadiponectinemiacauses endothelial dysfunction by increasing superoxide anionproduction [14], promoting the synthesis of adhesionmoleculesin endothelial cells and the proliferation of vascular smoothmuscle cells [15,16]. Resistin is a cysteine-rich polypeptideadipokine with proinflammatory activity that is synthesized byadipose tissue and may act both in paracrine and in endocrinefashions [15,17]. Also, circulating monocytes and macrophagesseem to be responsible for resistin production in humans [18].Visfatin is a novel adipokine, released from VAT andperivascular adipose tissue, which has an insulin-mimeticeffect [17,19]. Visfatin hasmultiple functions in the vasculature:it stimulates growth of vascular smooth muscle cells andendothelial angiogenesis, and it can also directly affect vascularcontractility. Moreover, it amplifies adipocyte differentia-tion [6,15].

It is evident that maintenance of a “normal” amount ofadipose tissue is essential because an imbalance can causeserious health problems. Melatonin, a pineal indolaminesecreted in a circadian pattern, has well-known antioxidant[20,21] and anti-inflammatory effects [22,23], and it isinvolved in energy expenditure and body fat mass regulation[24,25]. Interestingly, it was shown that the amplitude of thenocturnal pineal [26] and serum [27] melatonin peaksdecreases significantly in obese animals [28].

In this study, we hypothesized that dietary melatoninadministration would support an anti-inflammatory re-sponse in adipose tissue of obese mice that, in turn, inhibitobesity-induced alterations. To test this hypothesis, we

investigated a number of factors such as body weight, fatdepots, white adipose tissue hyperplasia, and inflammationas well as levels of the adipokines, mentioned above, andinflammatory cells in a murine model of obesity. Becauseprior evidence indicates that obesity changes these endpoints, we have been characterized subcutaneous and VATalterations and investigated the beneficial effects of melato-nin on adipose tissue responses in leptin deficient ob/obmice.Based on these results, we proposed melatonin supplemen-tation in drinking water as a suitable dietary addition totemper obesity.

2. Methods and materials

2.1. Animal model

In the current study, 20 lean (B6.-(lean)/OlaHsd) and 20 obesemice (B6.V-Lepob/OlaHsd), defined as ob/ob mice, at 4 weeks ofage, obtained from Harlan Laboratories S.r.l. (Udine, Italy),were housed in standard cages in a temperature-controlledanimal facility (+20°C) with a 12-hour/12-hour light-darkcycle. All animals were fed ad libitum with normal chowand with free access to tap water.

The ob/ob mice, genetically leptin deficient, are an impor-tant model for studying adipose organs and are widely usedfor obesity and diabetes research [29,30].

2.2. Experimental design

Mice were randomly divided into 4 groups of 10 animals each:(a) control lean mice treated with vehicle, (b) lean mice treatedwith melatonin for 8 weeks, (c) control obese mice treated withvehicle, and (d) obese mice treated with melatonin for 8 weeks.

Treatedgroups, instead of tapwaterwithvehicle,were givenmelatonin dissolved in drinking water; in particular, melatonin(kindly provided by Chronolife S.r.l., Roma, Italy) was dissolvedin aminimum volume of ethanol and diluted in drinking waterto yield a dose of 100mg/kg bodyweight per day, as described inour previous studies [31–33]. Fresh melatonin and vehiclesolutions were prepared twice a week, and the melatonin dosewas adjusted to the body weight throughout the study period.Drinking bottles were covered with aluminium foil to protectmelatonin from light [25].

Blood glucose wasmonitored throughout the study in bloodcollected via tail-snip as previously described [31]. Protocolswere approved by the Italian Ministry of Health. All necessarycare was taken to ameliorate any potential animal suffering.

At the conclusion of an 8-week treatment period, when themice were 13 weeks of age, all the animals were killed bycervical dislocation and the dorsal SATand the abdominopelvicVATwere removed and weighed. In Fig. 1, we depict the depotsof white tissue in mice and the areas from which we removedthese deposits.

2.3. Sample processing

Fat specimens were rinsed in physiological solution, and aportion was fixed in buffered formalin, dehydrated, in graded

Fig. 1 –White adipose depots inmice. Representation ofmajorwhite adipose depots (subcutaneous and visceral fat) of mice.Purple identifies the visceral and subcutaneous fat depotsanalyzed in the present study. Modified from Cinti [41].

893N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

ethanol, and then embedded in paraffin wax. Serial paraffinsections (7 μm thick) of each sample were cut with amicrotome and then used for morphometrical evaluationsand immunofluorescences analyses [34]. The other parts ofadipose tissues were fixed in 2.5% glutaraldehyde incacodilate buffer 0.1 M (pH 7.4) for 3 hours at +4°C andpostfixed in 2% osmium tetroxide in cacodilate buffer for 1hour at +4°C for the ultrastructural investigation.

2.4. Transmission electron microscopy

The adipose tissues fixed in glutaraldehyde were treated forultrastructural analysis according to Rezzani et al [34]. Briefly,adipose tissues were dehydrated in increasing ethanolconcentrations and propylene oxide, followed by Araldite-Epon resin embedding. Semithin sections (1 μm thick) wereobtained using an UltraCut E ultramicrotome, then stained bytoluidine blue and observed with a light microscope (Olym-pus, Hamburg, Germany). Subsequently, from representativeblocks, 70- to 80-nm-thick ultrathin sections were obtainedusing a diamond knife, collected on formvar coated grids,double stained by uranyl acetate and lead citrate, andobserved with a transmission electron microscopy (TecnaiG2 Spirit; FEI Company, Eindhoven, the Netherlands) at 80 kV.

2.5. Morphometrical analysis

Alternate paraffin sections were deparaffinized, rehydrated,and stained with hematoxylin-eosin and then were observedwith an optical light microscope (Olympus) at a finalmagnification of ×200. Digital images of SAT and VAT werecaptured and the area of 100 random adipocytes wasmeasured for each animal, as previously described by Nicolaiet al [35]. Individual adipocyte area (μm2) was determinedusing an image analyzer (Image Pro Plus 9.1 Version, Milan,Italy) by 2 observers blinded to the treatment, whoseevaluation was assumed to be correct if the values were not

significantly different. In case of dispute concerning interpre-tation, the case was reconsidered to reach an agreement.

2.6. Immunofluorescence localization of adipokines

Sections were dewaxed in xylene and hydrated using gradedethanols, and the blocking step was performed before theapplication of the primary antibodies with 1% bovine serumalbumin in phosphate-buffered saline for 1 hour in a humidchamber. Subsequently, the sections were incubated overnight at4°C with the following primary antibodies: rabbit polyclonaladiponectin (diluted 1:700; AbCam, Cambridge, UK), goat poly-clonal adiponectin receptor 1 (diluted 1: 100; Santa Cruz Biotech-nology, Santa Cruz, CA, USA), goat polyclonal adiponectinreceptor 2 (diluted 1:100; Santa Cruz Biotechnology), goat poly-clonal resistin (diluted 1:200; Santa Cruz Biotechnology), rabbitmonoclonal visfatin (diluted 1:250; Abcam), and goat polyclonaltumor necrosis factor α (TNF-α; diluted 1:50; Santa Cruz Biotech-nology). Samples were washed in phosphate-buffered saline andlabeled using specific Alexa Fluor (Invitrogen detection technolo-gies, Leiden, The Netherlands) 488 or 543 conjugated secondaryantibodies (diluted 1:200). Finally, the samples were counter-stained with 4′,6-diamidino-2-phenylindole, mounted and ob-served with a confocal microscope (LSM 510 Zeiss, Munich,Germany), as reported previously by Rodella et al [36]. Sectionswithout primaryantibodyand in thepresenceof isotype-matchedIgGs served as negative immunofluorescent control.

Fifteen fields (area of which was 0.04 mm2), randomlyselected from each section, were analyzed and theimmunopositivity for each primary antibody was calculatedusing a software for image acquisition (Image Pro Plus 9.1Version). The evaluation of positive immunostaining wasmade by 2 blinded investigators, whose evaluation wasassumed to be correct if the values were not significantlydifferent. In case of dispute concerning interpretation, thecase was reconsidered to reach an agreement.

2.7. Statistical analyses

The sample size of this research plan was 5 animals per cell, 4cells with a factorial design with 2 factors: animal model-leanor obese (factor A) and treatment-melatonin or vehicle (factorB) at 2 × 2 levels. This design achieves 99.64% power todemonstrate an effect size of 1 with a combination between Aand B factors (fixed-effects analysis of variance poweranalysis). At this aim, we decided to perform in duplicatethe experiment.

The data obtained in the morphometrical and immuno-fluorescence analyses were presented as means ± SD.Statistical analyses were performed using a 1-way analysisof variance test corrected by Bonferroni. Differences amonggroups were considered statistically significant at P < .05.

3. Results

All animals of each experimental group survived and mela-tonin supplementation in drinking water was well tolerated.After 8 weeks of melatonin treatment, ob/obmice resulted in a

894 N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

significative blood glucose reduction with respect to ob/obmice treated with vehicle; however, these mice had higherglucose level than did lean mice treated with melatonin orvehicle (Fig. 2A). As described in our previous studies [31–33],ob/ob mice exhibited a body weight increase of 42.13% withrespect to control mice treated with melatonin or vehicle(Fig. 2B). Moreover, we noted that ob/ob mice treated withvehicle had VAT and SAT masses 91.04% and 88.16%,respectively, greater than those of lean mice treated withmelatonin or vehicle. Melatonin orally supplementationdecreased the body weight gain of ob/ob mice treated withvehicle by 4.43%, and the examination of mice at the grossanatomical level indicated that ob/ob mice treated withmelatonin clearly exhibited reduced white adipose depots,which was confirmed also by the measurement of VAT andSAT depot weights that were lower by 53.03% and 41.11%,respectively (Fig. 2C-H).

The histologic examination of VAT and SAT revealed thatlean mice, treated with melatonin or vehicle, contained

Fig. 2 – Blood glucose, body weight, and adipose tissue depot weidifferent experimental groups (lean mice treated with vehicle orvisceral (C, D) and subcutaneous (E, F) adipose tissue of ob/ob misubcutaneous (H) fat of each experimental group (lean mice treavehicle or melatonin). Scale bar = 1.4 cm. The “*” represents signirepresents significant difference from the lean group treated witthe ob/ob group treated with vehicle. P < .05. Values are means ±

mature adipocytes, which were uniformly endowed with asingle fat droplet and the nucleus in the periphery (Fig. 3A, E).In contrast, VAT and SAT adipocytes in ob/ob mice treatedwith vehicle exhibited a significantly increased size comparedwith the adipocyte cell area of lean mice treated with vehicle(Fig. 3B, F). The, ob/obmice treated with melatonin showed asignificant reduction in adipocyte cell size, an importantindicator of healthy adipose tissue (Fig. 3C, G). These resultswere confirmed also by the morphometrical analyses sum-marized in Fig. 3D and H. Interestingly, VAT depots of ob/obmice treated with vehicle had a significatively elevatedinflammatory infiltrate appeared as crown-like structureswith respect to SAT depots (these VAT infiltrates areidentified in the box of Fig. 3B).

The semithin section analyses confirmed the previousmorphologic evaluation of ob/ob VAT; these large cellscontained a single lipid droplet and a thin nucleus. The cellswere sometimes surrounded by an inflammatory infiltrate(Fig. 3I). At the ultrastructural level, macrophages (Fig. 3L) and

ght. Blood glucose concentrations (A) and body weights (B) inmelatonin and ob/ob mice treated with vehicle or melatonin),ce treated with vehicle, and the weights of visceral (G) andted with vehicle or melatonin and ob/ob mice treated withficant difference from the lean group treated with vehicle, “#”h melatonin, and “+” represents significant difference fromSD; n = 10 mice/group.

Fig. 3 –Morphologic and transmission electronmicroscopy evaluations ofwhite adipose tissue depots. Photomicrographs of VAT(A-C) and SAT (E-G) of lean mice treated with vehicle (A, E), obese mice treated with vehicle (B, F), and obese mice treated withmelatonin (C, G). Hematoxylin-eosin staining. Scale bar: 20 μm. In the photomicrographs, the box identifies the area ofinflammatory infiltration, the “*” shows small adipocytes, and the “§” shows big adipocytes. The graphs represent, respectively,the visceral (D) and subcutaneous (H) adipocyte sizes, expressed inμm2. The “*” represents a significant difference from the leangroup treated with vehicle, “#” represents significant difference from the lean group treated with melatonin, and “+” representssignificant difference from the ob/ob group treatedwith vehicle. P < .05 Values aremeans ± SD, n = 10mice/group. Semithin (I) andtransmission electron microscopy photomicrographs (L-N) of VAT of ob/ob mice treated with vehicle that depict a singlemacrophage near hypertrophic adipocytes (I, L) and a degranulated mast cell in the adipocyte interstitium (M, N). Transmissionelectron microscopy photomicrographs of ob/obmice treated with vehicle visceral fat (O, P) show enlarged lipid droplets withinterstitial cells and connective fibrotic deposits. a, adipocyte; m, macrophage; M, mast cell. The arrow identifies collagen in thespace between 2 adipocytes. Scale bars: 20 μm (A), 2.5 μm (B, E, and F), 10 μm (C), and 5 μm (D).

895N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

mast cells often appeared in a degranulated state (Fig. 3M, N).After melatonin supplementation, the VAT of obese micetreated with vehicle showed regular smaller adipocytes and,at ultrastructural level, regular vessels associated with fibro-blastic cells devoid of inflammatory elements (data not shown).Interesting histopathologic evidence, best characterized bytransmission electron microscopy, showed that VAT wasassociated with fibrosis in obese mice treated with vehicle.Many fibrotic cells were located in abundant connective andfibrotic matrix in the interstitial space (Fig. 3O, P).

Immunofluorescence analyses of TNF-α (identified inred—Fig. 4A-D), important inflammation-related adipokine,confirmed the previous morphologic and ultrastructural

evaluations showing at VAT level of lean mice, treated withmelatonin or vehicle (Fig. 4A, B), a negative expression withrespect to a moderate expression of ob/ob mice treated withvehicle (Fig. 4C). Interestingly, melatonin supplementation inob/ob mice reduced significantly the VAT inflammatorymarker expression (Fig. 4D). Furthermore, immunofluores-cence analyses of resistin (identified in green—Fig. 4E-H) andvisfatin (identified in green—Fig. 4I-N) showed that leanmice,treated with melatonin or vehicle, had no or very weakexpression of both adipokines (Fig. 4E, F, I, L). By comparison,ob/ob mice treated with vehicle showed a moderate/strongexpression of both adipokines at the level of the membrane ofvisceral white adipocytes (Fig. 4G, M). Melatonin treatment for

Fig. 4 – Tumor necrosis factor α, visfatin, and resistin immunofluorescence analyses. VAT expression of TNF-α (A-D), resistin(E-H), and visfatin (I-N), respectively, in lean mice treated with vehicle (A, E, I), lean mice treated with melatonin (B, F, L), ob/obmice treated with vehicle (C, G, M), and ob/ob mice treated with melatonin (D, H, N). Scale bars: 20 μm. In thephotomicrographs, the “*” shows small adipocyte, the “§” shows big adipocytes. The graphs summarize thehistomorphometrical analyses, at the visceral and subcutaneous fat levels, respectively, of TNF-α (O), resistin (P), and visfatin(Q). The “*” represents significant difference from the lean group treated with vehicle, “#” represents significant differencefrom the lean group treated with melatonin, and “+” represents significant difference from the ob/ob group treated withvehicle. P < .05. Values are means ± SD; n = 10 mice/group.

896 N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

ob/ob mice reduced significantly the expressions of resistinand visfatin (Fig. 4H, N). The quantitative data related to VATand SAT immunostainings are summarized in Fig. 4O-Q.

An opposite pattern with respect to the expression of TNF-α, visfatin, and resistin was shown in the immunofluores-cence analyses of adiponectin and its relative receptors. Leanmice, treated with melatonin or vehicle, expressed moderateadiponectin levels (identified in green) at the level of themembrane of visceral white adipocytes (Fig. 5A, B).Adiponectin expression of white adipocytes of ob/ob micetreated with vehicle was significantly decreased, often beingalmost absent with respect to control mice treated withmelatonin or vehicle (Fig. 5C). Interestingly, this effect wasreversed, in part, by oral supplementation of melatonin; thesecells showed a weak adiponectin expression (Fig. 5D). Thepattern of expression of adiponectin receptor 1 (identified ingreen—Fig. 5E-H) and receptor 2 (identified in red—Fig. 5I-N)matched the level of adiponectin; in particular, ob/ob micetreated with vehicle showed a very weak expression

compared with lean mice treated with melatonin or vehicle,and melatonin treatment for ob/ob mice for 8 weeks inducedthe up-regulation of both adiponectin receptors.

These data are displayed for quantitative analyses andsummarized in Fig. 5O-Q, which also shows the SATimmunopositivity levels.

All these obesity-induced alterations and the melatoninprotective effects at adipose tissue level are schematicallysummarized respectively in Fig. 6A and B.

4. Discussion

Dietary interventions that reduce obesity-related alterationsgarner significant research interest. Melatonin supplementa-tion is drawing increased interest in the medical andnutritional research fields [37–39]; actually, much of theresearch focus is on metabolic syndrome obtained mainly inanimal models and suggests that melatonin might increase

Fig. 5 – Immunofluorescence analyses of adiponectin and adiponectin receptors. VAT expression of adiponectin (A-D) andadiponectin receptor 1 (E-H) and adiponectin receptor 2 (I-N) in lean mice treated with vehicle (A, E, I), lean mice treated withmelatonin (B, F, L), ob/obmice treated with vehicle (C, G, M), and ob/obmice treated with melatonin (D, H, N). Scale bars: 20 μm.In the photomicrographs, the “*” shows small adipocyte, whereas the “§” shows big adipocytes. The graphs summarize thehistomorphometrical analyses, at visceral and subcutaneous fat levels, of adiponectin (O), adiponectin receptor 1 (P), andadiponectin receptor 2 (Q). The “*” represents significant difference from the lean group treated with vehicle, “#” representssignificant difference from the lean group treated with melatonin, and “+” represents significant difference from the ob/obgroup treated with vehicle. P < .05. Values are means ± SD; n = 10 mice/group.

897N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

energy expenditure. In the present study, we focused on theanti-inflammatory properties of melatonin and sought todetermine whether a melatonin-supplemented diet for 8weeks was beneficial for attenuating inflammatory infiltra-tion and restoring adipokine pattern in obese mice. Inparticular, we show that in ob/ob mice, both VAT and SATincrease in volume (adipocyte hyperplasia and hypertrophy)and VAT, more than SAT, is strictly associated with localinfiltration of inflammatory cells and alterated adipokineexpression pattern, whereas interestingly as supposed by ourresearch hypothesis, melatonin administration to ob/obmice ameliorated adipose tissue morphologic alterations,reduced adipocyte inflammation, and restored the correctadipokines profile.

With regard to the VAT and SAT increases, our data are inagreement with those obtained from others showing thatunder conditions of positive energy balance, visceral adipo-cytes are smaller than those in the subcutaneous deposits

[40,41]. Moreover, Fain [42] documented that human visceralomental fat cells are smaller than subcutaneous adipocytes,but macrophage accumulation is greatest in visceral omentaldepot, suggesting that something in VAT is essential formaintaining correct cell homeostasis [43]. We extended thesefindings by showing a higher inflammatory cell infiltrationand expression of proinflammatory adipokines in VAT. Undernormal conditions, the VAT adipocytes are involved in lipidsynthesis, storage, and secretion of anti-inflammatory mole-cules, but they can also be induced to secrete a number ofinflammatory factors [5]. Indeed, in an early study of ob/obmice, we observed an increased expression of both TNF-α, asin this study, and CD68, marker of macrophages [31].Furthermore, He et al [44] considered TNF-α as an inflammation-related adipokine, which can modulate insulin signaling andinduce insulin resistance in adipocytes [45], and so it isinvolved in the development of metabolic syndrome [46].Also, the ultrastructural investigation confirmed that VAT

Fig. 6 – Melatonin and obesity dysfunctions. Schematic representations of obesity-induced alterations (A) and of protectiveeffects of melatonin (B) at white adipocyte level.

898 N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

expansion in ob/obmice is associated with a local infiltrationof inflammatory cells, mainly macrophages and mast cells,as previously reported also by Giordano et al [47]. Further-more, Cinti et al [48] suggested that macrophages arelocalized selectively at sites of necrotic-like cell deathwhere they appear as crown-like structures. Whether en-hanced lysis/death of hypertrophic adipocytes is the primarytrigger that accounts for inflammation is unknown. Also,what signal results in greater macrophage accumulation inadipose tissue remains to be investigated; however, VAThypertrophy per se promotes cell death resulting in macro-phage accumulation and aggregation around dead cells.Moreover, it is known that there is a strict correlationbetween adipocyte size and local inflammation [42] that, inturn, promotes and increase obesity damages.

Concerning the alteration of VAT adipokines expression,we observed that VAT expansion in ob/ob mice treated withvehicle requires extensive tissue remodeling which, in turn,probably contributes to the physiological balance of adipokineexpression. Adipokines, which are produced by adipocytes oradipose tissue macrophages, induce a low-grade chronicinflammatory state that plays a central role in obesity-relatedcardiovascular complications and insulin resistance [6,15].Therefore, fat tissue affects not only overall metabolism butalso the functionality of many organs and tissues. In vivo andin vitro studies suggest that VAT, rather than the SAT, is themain source of adiponectin [49]. Here, we observed thatadiponectin and its receptors 1 and 2 are expressed abun-dantly by adipocytes of lean mice treated with melatonin orvehicle. In ob/ob mice treated with vehicle, adiponectin andadiponectin receptor expressions were dramatically reducedand disappear almost completely. Typically, adiponectin isnegatively related to the increase of fat mass likely owing tothe abnormal hormonal milieu mainly caused by the inhib-itory effects exerted by the increased local TNF-α levels (as

also observed in this study), oxidative stress, and theproinflammatory state which prevail in central obesity [10].

With regard to other adipokines, resistin expression isstimulated by the increased VAT expression of TNF-α andproinflammatory factors [50], and visfatin is abundantlyexpressed by VAT and macrophages [51], both of which areincreased in obesity [15]. Herein, we observed that ob/ob micetreated with vehicle showed a higher expression of TNF-α,resistin, and visfatin in VAT comparedwith leanmice, treatedwith melatonin or vehicle.

A final observation in this study is that melatoninadministration for 8 weeks reduced body weight, adiposetissue depots, adipocyte hyperplasia and hypertrophy, bloodglucose, proinflammatory factors, and restored adipokinephysiological profile expression. Our data are in agreementwith previous findings in which melatonin treatment wasreported to counteract obesity alterations and to optimizemetabolism [52–54]. The effects of melatonin in obesity havebeen intensively studied in animal models of diet-inducedobesity [51,55–58]. In the current investigation, we evaluatedthe effect of orally administered melatonin in an animalmodel of obesity, which have a missense mutation of theleptin gene.

The current study has demonstrated that dietary melato-nin supplementation is associated with partial ameliorationof the pathogenesis of obesity, already in the early phases.The biological action of melatonin in this study may haveinvolved (i) the activation of central receptors [59–61],resulting in changes in metabolic rate via sympatheticnervous activity and subsequent effects on lipolysis andadipose tissue plasticity, and/or (ii) a direct effect of melato-nin on adipose tissue [28,62–65]. The activation of adiposetissue receptors may influence energy storage by modulatingadipocyte metabolism or proliferation. The findings indicatethat the adipose tissue is a peripheral target of melatonin for

899N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

the regulation of its overall metabolism [47]. Consistent withthis, Brydon and colleagues [64] reported that MT2 receptorsare expressed in human adipose tissue and participate inadipocyte homeostasis.

A limitation of this study may be that no significativedifference in ob/ob body weight was seen after 8 weeks ofmelatonin treatment; a longer duration of treatment may beneeded to elicit such improvement. Moreover, probablymelatonin receptor expression at VAT and SAT level wouldbe interesting to evaluate together with adipokines expres-sion for better understanding the melatonin mechanism ofaction. These limitations will hopefully be addressed infuture studies.

Our data suggested that melatonin improves the low gradeof adipose tissue inflammation that is associated with obesityand, together with the evidences of melatonin safety andefficacy, make this indoleamine a potential dietary supple-mentation to help people in the prevention of obesity,diabetes, and associated complications. More research, withemphasis on clinical trials, is needed to further strengthenthe link betweenmelatonin and obesity and the effective doserequired to modulate these metabolic responses in humans.

Acknowledgment

The authors sincerely thank Russel J. Reiter for his kindAmerican English and editing revision. The authors alsothank Antonio Lavazza for his support in electron microscopyanalyses and Lorena Giugno for her technical support. Theyalso sincerely thank Chronolife S.r.l. (Roma, Italy) for courte-ously providing melatonin. This study was supported by agrant (ex-60%) from the University of Brescia, Italy.

The authors declare that there are no conflicts of interest.

R E F E R E N C E S

[1] Mokdad AH, Ford ES, Bowman BA, Dietz WH, Vinicor F, BalesVS, et al. Prevalence of obesity, diabetes, and obesity-relatedhealth risk factors, 2001. JAMA 2003;289:76–9.

[2] Yehuda-Shnaidman E, Schwartz B. Mechanisms linkingobesity, inflammation and altered metabolism to coloncarcinogenesis. Obes Rev 2012;13:1083–95.

[3] Rhodes DH, Pini M, Castellanos KJ, Montero-Melendez T,Cooper D, Perretti M, et al. Adipose tissue-specific modula-tion of galectin expression in lean and obese mice: evidencefor regulatory function. Obesity (Silver Spring) 2013;21:310–9.

[4] Patsouris D, Li PP, Thapar D, Chapman J, Olefsky JM, Neels JG.Ablation of CD11c-positive cells normalizes insulin sensitiv-ity in obese insulin resistant animals. Cell Metab 2008;8:301–9.

[5] Gustafson B. Adipose tissue, inflammation and atheroscle-rosis. J Atheroscler Thromb 2010;17:332–41.

[6] Antuna-Puente B, Feve B, Fellahi S, Bastard JP. Adipokines:the missing link between insulin resistance and obesity.Diabetes Metab 2008;34:2–11.

[7] Matsuzawa Y. Therapy insight: adipocytokines in metabolicsyndrome and related cardiovascular disease. Nat Clin PractCardiovasc Med 2006;3:35–42.

[8] Waki H, Tontonoz P. Endocrine functions of adipose tissue.Annu Rev Pathol 2007;2:31–56.

[9] Harman-Boehm I, Blüher M, Redel H, Sion-Vardy N, Ovadia S,Avinoach E, et al. Macrophage infiltration into omentalversus subcutaneous fat across different populations: effectof regional adiposity and the comorbidities of obesity. J ClinEndocrinol Metab 2007;92:2240–7.

[10] Anfossi G, Russo I, Doronzo G, Pomero A, Trovati M.Adipocytokines in atherothrombosis: focus on platelets andvascular smooth muscle cells. Mediators Inflamm 2010;2010:1–26.

[11] Dunajska K, Milewicz A, Jedrzejuk D, Szymczak J,Kuliczkowski W, Salomon P, et al. Plasma adiponectinconcentration in relation to severity of coronary atheroscle-rosis and cardiovascular risk factors in middle-aged men.Endocrine 2004;25:215–21.

[12] Mehta S, Farmer JA. Obesity and inflammation: a new look atan old problem. Curr Atheroscler Rep 2007;9:134–8.

[13] Barseghian A, Gawande D, Bajaj M. Adiponectin and vulner-able atherosclerotic plaques. J Am Coll Cardiol 2011;57:761–70.

[14] Cao Y, Tao L, Yuan Y, Jiao X, Lau WB, Wang Y, et al.Endothelial dysfunction in adiponectin deficiency and itsmechanisms involved. J Mol Cell Cardiol 2009;46:413–9.

[15] Maenhaut N, Van de Voorde J. Regulation of vascular tone byadipocytes. BMC Med 2011;9:25.

[16] Shargorodsky M, Boaz M, Goldberg Y, Matas Z, Gavish D, FuxA, et al. Adiponectin and vascular properties in obesepatients: is it a novel biomarker of early atherosclerosis? Int JObes (Lond) 2009;33:553–8.

[17] Stofkova A. Resistin and visfatin: regulators of insulinsensitivity, inflammation and immunity. Endocr Regul 2010;44:25–36.

[18] Bastard JP, Maachi M, Lagathu C, Kim MJ, Caron M, Vidal H,et al. Recent advances in the relationship between obesity,inflammation, and insulin resistance. Eur Cytokine Netw2006;17:4–12.

[19] Adya R, Tan BK, Punn A, Chen J, Randeva HS. Visfatin induceshuman endothelial VEGF and MMP-2/9 production via MAPKand PI3K/Akt signalling pathways: novel insights intovisfatin-induced angiogenesis. Cardiovasc Res 2008;78:356–65.

[20] Galano A, Tan DX, Reiter RJ. On the free radical scavengingactivities of melatonin's metabolites, AFMK and AMK. JPineal Res 2013;54:245–57.

[21] Zhang HM, Zhang Y. Melatonin: a well-documented antiox-idant with conditional pro-oxidant actions. J Pineal Res 2014;57:131–46.

[22] Hu ZP, Fang XL, Fang N, Wang XB, Qian HY, Cao Z, et al.Melatonin ameliorates vascular endothelial dysfunction,inflammation, and atherosclerosis by suppressing the TLR4/NF-κB system in high-fat–fed rabbits. J Pineal Res 2013;55:388–98.

[23] Jin H, Wang Y, Zhou L, Liu L, Zhang P, Deng W, et al.Melatonin attenuates hypoxic pulmonary hypertension byinhibiting the inflammation and the proliferation of pulmo-nary arterial smoothmuscle cells. J Pineal Res 2014;57:442–50.

[24] Tan DX, Manchester LC, Fuentes-Broto L, Paredes SD, ReiterRJ. Significance and application of melatonin in the regula-tion of brown adipose tissue metabolism: relation to humanobesity. Obes Rev 2011;12:167–88.

[25] Jimenéz-Aranda A, Fernández-Vázquez G, Mohammad A-Serrano M, Reiter RJ, Agil A. Melatonin improves mitochon-drial function in inguinal white adipose tissue of Zückerdiabetic fatty rats. J Pineal Res 2014;57:103–9.

[26] Cano P, Jiménez-Ortega V, Larrad A, Reyes Toso CF, CardinaliDP, Esquifino AI. Effect of a high-fat diet on 24-h pattern ofcirculating levels of prolactin, luteinizing hormone, testos-terone, corticosterone, thyroid-stimulating hormone andglucose, and pineal melatonin content, in rats. Endocrine2008;33:118–25.

900 N U T R I T I O N R E S E A R C H 3 5 ( 2 0 1 5 ) 8 9 1 – 9 0 0

[27] Peschke E, Frese T, Chankiewitz E, Peschke D, Preiss U,Schneyer U, et al. Diabetic Goto Kakizaki rats as well as type 2diabetic patients show a decreased diurnal serum melatoninlevel and an increased pancreatic melatonin-receptor status.J Pineal Res 2006;40:135–43.

[28] Nduhirabandi F, du Toit EF, Lochner A. Melatonin and themetabolic syndrome: a tool for effective therapy in obesity-associated abnormalities? Acta Physiol (Oxf) 2012;205:209–23.

[29] Peng XG, Ju S, Fang F, Wang Y, Fang K, Cui X, et al.Comparison of brown and white adipose tissue fat fractionsin ob, seipin, and Fsp27 gene knockout mice by chemicalshift-selective imaging and (1)H-MR spectroscopy. Am JPhysiol Endocrinol Metab 2013;304:E160–7.

[30] Tang Y, Sharma P, NelsonMD, Simerly R, Moats RA. Automaticabdominal fat assessment in obese mice using a segmentalshape model. J Magn Reson Imaging 2011;34:866–73.

[31] Agabiti-Rosei C, De Ciuceis C, Rossini C, Porteri E, Rodella LF,Withers SB, et al. Anticontractile activity of perivascular fatin obese mice and the effect of long-term treatment withmelatonin. J Hypertens 2014;32:1264–74.

[32] Favero G, Lonati C, Giugno L, Castrezzati S, Rodella LF,Rezzani R. Obesity-related dysfunction of the aorta andprevention by melatonin treatment in ob/ob mice. ActaHistochem 2013;115:783–8.

[33] Stacchiotti A, Favero G, Giugno L, Lavazza A, Reiter RJ, RodellaLF, et al. Mitochondrial and metabolic dysfunction in renalconvoluted tubules of obese mice: protective role of melato-nin. PLoS One 2014;9:e111141.

[34] Rezzani R, Favero G, Stacchiotti A, Rodella LF. Endothelial andvascular smooth muscle cell dysfunction mediated bycyclophylin A and the atheroprotective effects of melatonin.Life Sci 2013;92:875–82.

[35] Nicolai A, Li M, Kim DH, Peterson SJ, Vanella L, Positano V,et al. Heme oxygenase-1 induction remodels adipose tissueand improves insulin sensitivity in obesity-induced diabeticrats. Hypertension 2009;53:508–15.

[36] Rodella LF, Favero G, Rossini C, Foglio E, Bonomini F, Reiter RJ,et al. Aging and vascular dysfunction: beneficial melatonineffects. Age (Dordr) 2013;35:103–15.

[37] Agil A, El-Hammadi M, Jiménez-Aranda A, Tassi M, Abdo W,Fernández-Vázquez G, et al. Melatonin reduces hepaticmitochondrial dysfunction in diabetic obese rats. J Pineal Res2015;59:70–9.

[38] Brum MC, Filho FF, Schnorr CC, Bottega GB, Rodrigues TC.Shift work and its association with metabolic disorders.Diabetol Metab Syndr 2015;7:45.

[39] Magrone T, Jirillo E. Childhood obesity: immune responseand nutritional approaches. Front Immunol 2015;6:76.

[40] Cinti S. The adipose organ. Prostaglandins Leukot EssentFatty Acids 2005;73:9–15.

[41] Cinti S. Transdifferentiation properties of adipocytes in theadipose organ. Am J Physiol Endocrinol Metab 2009;297:E977–86.

[42] Fain JN. Release of inflammatory mediators by humanadipose tissue is enhanced in obesity and primarily by thenonfat cells: a review. Mediators Inflamm 2010;2010:513948.

[43] Murano I, Barbatelli G, Parisani V, Latini C, Muzzonigro G,Castellucci M, et al. Dead adipocytes, detected as crown-likestructures, are prevalent in visceral fat depots of geneticallyobese mice. J Lipid Res 2008;49:1562–8.

[44] He Z, Li M, Zheng D, Chen Q, Liu W, Feng L. Adipose tissuehypoxia and low-grade inflammation: a possiblemechanism forethanol-related glucose intolerance? Br J Nutr 2015;113:1355–64.

[45] Kern PA, Ranganathan S, Li C, Wood L, Ranganathan G.Adipose tissue tumor necrosis factor and interleukin-6expression in human obesity and insulin resistance. Am JPhysiol Endocrinol Metab 2001;280:E745–51.

[46] Hotamisligil GS. Inflammation and metabolic disorders.Nature 2006;444:860–7.

[47] Giordano A, Murano I, Mondini E, Perugini J, Smorlesi A, SeveriI, et al. Obese adipocytes show ultrastructural features ofstressed cells and die of pyroptosis. J Lipid Res 2013;54:2423–36.

[48] Cinti S, Mitchell G, Barbatelli G, Murano I, Ceresi E, Faloia E,et al. Adipocyte death defines macrophage localization andfunction in adipose tissue of obese mice and humans. J LipidRes 2005;46:2347–55.

[49] Ghigliotti G, Barisione C, Garibaldi S, Fabbi P, Brunelli C,Spallarossa P, et al. Adipose tissue immune response: noveltriggers and consequences for chronic inflammatory condi-tions. Inflammation 2014;37:1337–53.

[50] Rabe K, Lehrke M, Parhofer KG, Broedl UC. Adipokines andinsulin resistance. Mol Med 2008;14:741–51.

[51] Moschen AR, Kaser A, Enrich B, Mosheimer B, Theurl M,Niederegger H, et al. Visfatin, an adipocytokine with proin-flammatory and immunomodulating properties. J Immunol2007;178:1748–58.

[52] Ríos-Lugo MJ, Cano P, Jiménez-Ortega V, Fernández-MateosMP, Scacchi PA, Cardinali DP, et al. Melatonin effect on plasmaadiponectin, leptin, insulin, glucose, triglycerides and choles-terol in normal and high fat-fed rats. J Pineal Res 2010;49:342–8.

[53] Roman S, Agil A, Peran M, Alvaro-Galue E, Ruiz-Ojeda FJ,Fernández-Vázquez G, et al. Brown adipose tissue and noveltherapeutic approaches to treat metabolic disorders. TranslRes 2015;165:464–79.

[54] Solís-Muñoz P, Solís-Herruzo JA, Fernández-Moreira D, Gómez-Izquierdo E, García-Consuegra I, Muñoz-Yagüe T, et al. Mela-tonin improves mitochondrial respiratory chain activity andliver morphology in ob/ob mice. J Pineal Res 2011;51:113–23.

[55] Agil A, Navarro-AlarcónM, Ruiz R, Abuhamadah S, El-Mir MY,Vázquez GF. Beneficial effects of melatonin on obesity andlipid profile in young Zucker diabetic fatty rats. J Pineal Res2011;50:207–12.

[56] Cardinali DP, Bernasconi PA, Reynoso R, Toso CF, Scacchi P.Melatonin may curtail the metabolic syndrome: studies oninitial and fully established fructose-induced metabolicsyndrome in rats. Int J Mol Sci 2013;14:2502–14.

[57] Nduhirabandi F, Du Toit EF, Blackhurst D, Marais D, LochnerA. Chronic melatonin consumption prevents obesity-relatedmetabolic abnormalities and protects the heart againstmyocardial ischemia and reperfusion injury in a prediabeticmodel of diet-induced obesity. J Pineal Res 2011;50:171–82.

[58] Sartori C, Dessen P, Mathieu C, Monney A, Bloch J, Nicod P,et al. Melatonin improves glucose homeostasis and endo-thelial vascular function in high-fat diet–fed insulin-resistantmice. Endocrinology 2009;150:5311–7.

[59] Cipolla-Neto J, Amaral FG, Afeche SC, Tan DX, Reiter RJ.Melatonin, energy metabolism, and obesity: a review. J PinealRes 2014;56:371–81.

[60] Reiter RJ, Tan DX, Manchester LC, Pilar Terron M, Flores LJ,Koppisepi S. Medical implications of melatonin: receptor-mediat-ed and receptor-independent actions. Adv Med Sci 2007;52:11–28.

[61] Reiter RJ, Tan DX, Fuentes-Broto L. Melatonin: a multitaskingmolecule. Prog Brain Res 2010;181:127–51.

[62] Slominski RM, Reiter RJ, Schlabritz-Loutsevitch N, Ostrom RS,Slominski AT. Melatonin membrane receptors in peripheraltissues: distribution and functions. Mol Cell Endocrinol 2012;351:152–66.

[63] Bartness TJ, Demas GE, Song CK. Seasonal changes inadiposity: the roles of the photoperiod, melatonin and otherhormones, and sympathetic nervous system. Exp Biol Med(Maywood) 2002;227:363–76.

[64] Brydon L, Petit L, Delagrange P, Strosberg AD, Jockers R.Functional expression of MT2 (Mel1b) melatonin receptors inhuman PAZ6 adipocytes. Endocrinology 2001;142:4264–71.

[65] SongCK, BartnessTJ. CNS sympathetic outflowneurons towhitefat that express MEL receptors may mediate seasonal adiposity.Am J Physiol Regul Integr Comp Physiol 2001;281:R666–72.