1 23

In Vitro Cellular & DevelopmentalBiology - Animal ISSN 1071-2690 In Vitro Cell.Dev.Biol.-AnimalDOI 10.1007/s11626-014-9733-6

Efficient in vitro adipocyte model of long-term lipolysis: A tool to study the behaviorof lipophilic compounds

Caroline Louis, Carine Van den Daelen,Gilles Tinant, Sophie Bourez, Jean-Pierre Thomé, Isabelle Donnay, YvanLarondelle, et al.

1 23

Your article is protected by copyright and all

rights are held exclusively by The Society

for In Vitro Biology. This e-offprint is for

personal use only and shall not be self-

archived in electronic repositories. If you wish

to self-archive your article, please use the

accepted manuscript version for posting on

your own website. You may further deposit

the accepted manuscript version in any

repository, provided it is only made publicly

available 12 months after official publication

or later and provided acknowledgement is

given to the original source of publication

and a link is inserted to the published article

on Springer's website. The link must be

accompanied by the following text: "The final

publication is available at link.springer.com”.

Efficient in vitro adipocyte model of long-term lipolysis:A tool to study the behavior of lipophilic compounds

Caroline Louis & Carine Van den Daelen & Gilles Tinant &Sophie Bourez & Jean-Pierre Thomé & Isabelle Donnay &

Yvan Larondelle & Cathy Debier

Received: 4 November 2013 /Accepted: 9 January 2014 / Editor: T. Okamoto# The Society for In Vitro Biology 2014

Abstract The triglycerides (TGs) stored in the white adiposetissue are mobilized during periods of negative energy balance.To date, there is no in vitro model of adipocytes imitating a longperiod of negative energy balance in which triglycerides arehighly mobilized. Such model would allow studying themobilization of TGs and lipophilic compounds trapped withinthe adipose tissue (e.g., pollutants and vitamins). The presentstudy aims at developing a performing long-term in vitro lipol-ysis in adipocytes, resulting in a significant decrease of TGstores. Lipolysis was induced on differentiated rat adipocytesby a lipolytic medium with or without isoproterenol for 12 h.The condition with isoproterenol was duplicated, once withmedium renewal every 3 h and once without medium renewal.Adding isoproterenol efficiently triggered lipolysis in a shorttime (3 h). However, a single stimulation by isoproterenol,without medium renewal, was not sufficient to reduce the TGcontent during a longer term (12 h). A reesterification of fattyacids occurred after a few hours of lipolysis, resulting in a novelincrease of cellular lipids. Regular medium renewal com-bined with repeated isoproterenol stimulations led to al-most emptied cells after 12 h. However, medium renewalwithout isoproterenol stimulation for 12 h was as efficientin terms of lipid mobilization. Our study demonstrates that,over a short-term period, isoproterenol is required to exert a

significant lipolytic effect on adipocytes. During a long-termperiod, the presence of isoproterenol is no longer essential.Instead, medium renewal becomes the main factor involved incell emptying. The efficiency of this protocol was demonstratedby visual tracking of the cells and bymonitoring the dynamics ofrelease of a lipophilic compound, PCB-153, from adipocytesduring lipolysis.

Keywords Adipocyte . Lipolysis . Cell culture . Lipophiliccompounds

Introduction

In mammals, white adipose tissue is the major site of energystorage. Adipocytes synthesize and accumulate massiveamounts of triglycerides (TGs), which are stored within intra-cellular lipid droplets (LDs) (Brasaemle et al. 2000; Ahmadianet al. 2010). In vivo, adipocytes are unilocular and LDs fillfrom 90% to 99% of the cell volume (Goldrick 1967). LDs aresurrounded by a phospholipid monolayer that harbors choles-terol and several types of proteins (Brasaemle et al. 2004). Thelipid core is composed of neutral lipids such as TGs for themajor part, but also sterol and vitamin A esters (Wang et al.2008; Chaves et al. 2011). Other lipophilic compounds suchas toxic and persistent lipophilic pollutants are also localizedin this reservoir (de Jong et al. 2004; Le Lay et al. 2004;Bourez et al. 2012b).

During periods of negative energy balance (e.g., lactation,fast, and disease), TGs stored in LDs can be rapidly mobilizedthrough lipolysis. This process leads to the release of free fattyacids (FFAs) and glycerol into the bloodstream so that theycan be used by various tissues (Large et al. 2004; Chaves et al.2011). Lipolysis is regulated by numerous signaling path-ways. The central nervous system, hormones such as insulinand glucagon, as well as autocrine/paracrine factors are able to

Electronic supplementary material The online version of this article(doi:10.1007/s11626-014-9733-6) contains supplementary material,which is available to authorized users.

C. Louis (*) :C. Van den Daelen :G. Tinant : S. Bourez :I. Donnay :Y. Larondelle : C. DebierInstitut des Sciences de la Vie, UCLouvain, Croix du Sud2/L7.05.08, 1348 Louvain-la-Neuve, Belgiume-mail: caroline.louis@uclouvain.be

J.<. ThoméLaboratoire d’Ecologie animale et d’Ecotoxicologie, Université deLiège, Allée du 6 août 15, 4000 Liège, Belgium

In Vitro Cell.Dev.Biol.—AnimalDOI 10.1007/s11626-014-9733-6

Author's personal copy

affect the rate of TG hydrolysis in mature adipocytes(Thompson et al. 2010). Nevertheless, the main pathwaymediating lipolysis is the cAMP-dependent protein kinase A(PKA) pathway. This mechanism involves the binding ofcatecholamines to β-adrenergic receptors coupled to GTP-binding proteins. Adenylate cyclase is then activated by thestimulation of the α-stimulatory subunit of a G-protein. Thisresults in an increase of intracellular cAMP levels, leading tothe activation of PKA. This enzyme is then able to phosphor-ylate and thus activate the hormone sensitive lipase (HSL) andperilipin, which is the most abundant protein associated withLDs (Khoo et al. 1972; Belfrage et al. 1981; Egan et al. 1992;Chaves et al. 2011). Phosphorylation of perilipin is required tofacilitate the action of the phosphorylated HSL. It also pro-motes the freeing of the α/β-hydrolase domain-containingprotein 5 (ABHD5 or CGI-58), which is bound to perilipinin the nonactivated condition. Once released from perilipin,CGI-58 links to adipose triglyceride lipase (ATGL), which isthen able to hydrolyze TGs, leading to diglyceride (DG)production in adipocytes. Phosphorylated and thus activatedHSL together with monoglyceride lipase (MGL) both contrib-ute to the final hydrolysis of DGs and monoglycerides (MGs),respectively (Lass et al. 2006; Brasaemle et al. 2009; Lafontanand Langin 2009; Zimmermann et al. 2009; Chaves et al.2011). Contrary to ATGL and HSL, the activity of MGL isregulated neither by hormones nor by the energy state of thefat cells (Lass et al. 2011).

Although the adipose tissue is often considered as having aprotective role by keeping toxic lipophilic organic pollutantsaway from target organs (Chevrier et al. 2000), it can alsobecome a source of pollutants for the rest of the body duringlipolysis. Indeed, during that metabolic process, lipophiliccontaminants such as persistent organic pollutants (POPs)(e.g., polychlorinated biphenyls [PCBs] and polybrominateddiphenyl ethers [PBDEs]) may be mobilized from the adiposetissue to the blood circulation together with lipids. Severalstudies have reported an increase in circulating concentrationsof POPs during periods of bodyweight loss in humans (Roganet al. 1986; Chevrier et al. 2000; Imbeault et al. 2002; Kimet al. 2011), rodents (Gallenberg and Vodicnik 1987;Gallenberg et al. 1987; Ring et al. 1990; Lee et al. 2002) andmarine mammals (Debier et al. 2003; Debier et al. 2006;Vanden Berghe et al. 2010; Debier et al. 2012; VandenBerghe et al. 2012). These molecules might cause harmfuleffects such as endocrine disruptions, neurobehavioral alter-ations, hepatic and immunotoxicities as well as cancerdevelopments (Golden et al. 1998; Faroon et al. 2000;Kuriyama and Chahoud 2004; Ludewig et al. 2008). Never-theless, the mechanisms involved in the intracellular transportof lipophilic pollutants (Bourez et al. 2012b), as well as thepathways of their mobilization from adipocytes, have not beenelucidated yet. In vitro models are interesting tools for suchmechanistic investigations. However, up to now, no in vitro

model of adipocytes allows studying the behavior of lipophiliccompounds trapped within adipocytes during a long-termlipolytic process.

The aim of this work was to develop a model of efficientand performing in vitro lipolysis in fat cells using primarycultures of rodent adipocytes. Such a model would be aproficient tool enabling not only to study the toxicokineticsof lipophilic pollutants in fat cells, but also to investigate themechanisms by which the lipophilic molecules, in general,behave in adipocytes during periods of negative energybalance. In order to achieve this, we tested (i) the effect ofmedium renewal and (ii) the impact of isoproterenol stimula-tion on the efficiency of the lipolytic process in culturedadipocytes. Indeed, isoproterenol, a well-known syntheticcatecholamine, is usually used in in vitro adipocytes as alipolytic factor (Nagayama et al. 2010; Bourez et al. 2012a).Nevertheless, a unique stimulation of isoproterenol only in-duces a limited breakdown of TGs after a few hours(Nagayama et al. 2010), which does not reflect a prolongedperiod of negative energy balance. We thus compared threedifferent lipolytic treatments during a period of 12 h in order toreach the most efficient lipolytic condition for the fat cells.The first treatment involved the use of isoproterenol combinedto the renewal of the lipolytic medium every 3 h. This treat-ment was compared to a second treatment without isopro-terenol but where the lipolytic medium was also replacedevery 3 h. Eventually, a third treatment involved the use ofisoproterenol but without medium renewal. The efficiency ofthe lipolytic treatment was then illustrated by visual tracking ofthe cells. In addition, the dynamic of release of one lipophiliccompound was followed throughout the lipolytic process. Thetargeted molecule was 2,2′,4,4′,5,5′-hexachlorobiphenyl(PCB-153). It was selected for its ubiquity in human andanimal tissues (Debier et al. 2006; Covaci et al. 2008;Vanden Berghe et al. 2012; Dirtu et al. 2013).

Materials and Methods

Isolation of progenitor cells. The isolation of progenitor cellswas adapted from Rodbell (1964) as described in Bourez et al.(2012a). Adipose tissue was obtained from 2-mo-old maleWistar rats (Centre d’Elevage Janvier, Le Genest Saint Isle,France). The experimental procedure was approved by theAnimal Care and Use Committee of UCLouvain. Rats werekilled by decapitation, and the epididymal fat pads wereresected under sterile conditions. The fat tissue was thenminced and digested into a Dulbecco’sModified Eagle Medium(DMEM, 4.5 g/l glucose, Gibco-Invitrogen, Merelbeke, Bel-gium) containing 1,250 U/ml type II collagenase (Sigma-Aldrich, Bornem, Belgium), 10% (v:v) heat-inactivated fetalbovine serum (FBS, PAA, A&E Scientific, Marcq, Belgium),10 mM HEPES buffer solution (Gibco-Invitrogen), and an

LOUIS ETAL.

Author's personal copy

antibiotic and antifungal mixture (100 U/ml penicillin –100 U/ml streptomycin – 250 ng/ml amphotericin B mixture,Lonza, Verviers, Belgium; 500 ng/ml gentamicin, Gibco-Invitrogen, and 40 U/ml nystatin, Lonza). The digestion wasperformed for 40 min at 37°C under gentle shaking. Thedigested tissue was then filtered through sterile 200 μm nylonmesh, and the filtrate was centrifuged at 115×g for 5 min at4°C in order to isolate the stromal-vascular fraction. Thesupernatant was collected and spun at 145×g for 5 min at4°C. The pellets of the two centrifugations were pooled,filtered through sterile 25 μm nylon mesh, and finally spunat 180×g for 5 min at 4°C. The final pellet of stromal-vascularcells was suspended in a medium composed of DMEM,10% (v:v) heat-inactivated FBS, and the antibiotic andantifungal mixture (same composition as above). Cellswere seeded at a mean density of 18,000 cells per cm2

on 6-well plates (Corning CellBIND Surface, Corning,Elscolab, Kruibeke, Belgium) (day 0) and incubated at37°C in a humidified atmosphere containing 10% CO2 inair for 24 h to allow cell sedimentation and adhesion.

Adipose differentiation of progenitor cells. Twenty-four hoursafter the isolation of progenitor cells (day 1), the medium wasreplaced by a differentiation medium composed of DMEM(4.5 g/l glucose), 10% heat-inactivated FBS, 100 U/ml peni-cillin – 100 U/ml streptomycin – 250 ng/ml amphotericin Bmixture, 10 nM dexamethasone (Sigma-Aldrich), 10 μMciglitizone (Sigma-Aldrich), and 5 μg/ml insulin (Sigma-Aldrich) (Bourez et al. 2012a). This medium was renewedevery 48 h until day 11 in order to obtain differentiated ratadipocytes.

Lipolysis experimental design. At day 11, the differentiationmedium in contact with cells was removed and replaced bya lipolytic medium composed of DMEM (1.0 g/l glucose,Gibco-Invitrogen), 5% (v:v) heat-inactivated FBS, 2%(w:v) bovine albumin (Sigma-Aldrich), and supplementedor not with 1 μM isoproterenol (Sigma-Aldrich). The con-centration was chosen according to Harmancey et al.(2005), Zhou et al. (2011), and Rapold et al. (2013). Thecondition with isoproterenol was duplicated, once withmedium renewal every 3 h and once without mediumchange. For the condition without isoproterenol, mediumrenewal was performed every 3 h (Fig. 1). The lipolyticexperiments were carried out for 12 h. For the conditionwith isoproterenol and without medium renewal, someplates underwent lipolysis for 24 h. Cultures were frequent-ly observed with a phase contrast microscope. FFAs re-leased in the extracellular medium were quantified every3 h in all experiments. Likewise, every 3 h, cells from oneplate (i.e., coming from the same experimental condition)were pooled in order to assess the cellular protein contentand the fatty acids of cellular neutral lipids (NLs).

Extracellular free fatty acid assessment. FFA concentrationswere determined in the extracellular medium using an in vitroenzymatic colorimetric method assay (Wako NEFA HR,Sopachem, Eke, Belgium) following the manufacturer’sinstructions. FFAs initially present in the lipolytic mediumwere quantified and subtracted of results.

Cellular protein assessment. After lipolytic medium removal,wells were washed once with heat up phosphate buffer saline(Sigma-Aldrich). Cells were then collected in 300 μl of lysisbuffer composed of 35 mM sodium dodecyl sulfate (Merck,Darmstadt, Germany), 60 mM Tris buffer (Merck), and10 mM ethylenediaminetetraacetic acid (Sigma-Aldrich).After homogenization, the cellular protein content was deter-mined by the Bicinchoninic Acid Protein Assay kit (Sigma-Aldrich) using bovine serum albumin (BSA, Sigma-Aldrich)as a calibration curve (Bourez et al. 2012b).

Cellular neutral lipid assessment. Cells were collected asdescribed for the determination of protein content. Total cellu-lar lipids were extracted and assessed as detailed inSchneider et al. (2012). Briefly, after homogenization,800 μl of cell lysate were transferred into Pyrex tubes and

collection of cellsrenewal of medium w/ Isorenewal of medium w/o Iso

A

B

C

0 3 6 9 12Lipolysis course [hours]

Figure 1. Schematic representation of the lipolytic experiments carriedout on differentiated rat adipocytes. A and B, Adipocytes were incubatedwith a lipolytic medium supplemented (black triangle) or not (whitetriangle) with 1 μM isoproterenol. Respective lipolytic media wererenewed every 3 h and old media were collected as well as cells (blackcircle). C, Adipocytes were incubated with a lipolytic medium supple-mented with 1 μM isoproterenol. Cells and lipolytic media were collectedevery 3 h (black circle).

EFFICIENT IN VITRO ADIPOCYTE MODEL OF LONG-TERM LIPOLYSIS

Author's personal copy

5 ml of a mixture of chloroform/methanol/water (2:2:1, v:v:v)(Biosolve, Valkenswaard, The Netherlands) was added.Triheptadecanoin (a TG containing three C17:0 fatty acids)(Larodan, Malmö, Sweden) was used as an internal standardin order to quantify fatty acids in NLs corresponding to TGs,DGs, MGs, and cholesterol esters. After spinning, the super-natant was discarded and the chloroform phase was thenevaporated under a moderate nitrogen flux. Samples werethen suspended into chloroform and were loaded on solid-phase extraction columns (Bond Elut NH2, 200 mg, Varian,Middelburg, The Netherlands) previously conditioned withhexane (Biosolve) in order to separate NLs from phospho-lipids and FFAs. After the samples were pulled through, NLswere eluted with chloroform/2-propanol (2:1, v:v) (Biosolve)and collected into new Pyrex tubes in order to evaporate thesolvent under a moderate nitrogen flux. A methylation stepwas performed firstly by adding 0.1 M KOH in methanol at70°C for 1 h and secondly, by adding 1.2 M HCl in methanolat 70°C for 15 min. Fatty acid methyl esters (FAMEs) werethen extracted by adding hexane followed by deionized waterand centrifugation. FAMEs were separated by gas chromatog-raphy as described in Dang Van et al. (2011). Each peak wasidentified by comparison of retention times with puremethyl ester standards purchased from Larodan and fromNu-Check Prep (Elysian, MN). Data processing was oper-ated with the ChromQuest 4.2 software (ThermoFinnigan,Milan, Italy). Thereafter, results were expressed by μmolof fatty acids in cellular NLs per mg of cellular protein.For the sake of simplicity, we refer to μmol NL/mgprotein in the text.

PCB-153 mobilization. At day 10, some differentiated ratadipocytes were incubated with a differentiation mediumsupplemented with PCB-153 congener (Dr. EhrenstorferGmbH, Augsburg, Germany) (37°C – 10% CO2 in air). PCB-153 was added to the culture medium as an ethanolic solution ata concentration of 300 nM, which is within the range ofconcentrations found in in vivo and in vitro studies(Wassermann et al. 1979; Meeker et al. 2011; Bourez et al.2012b). The impact of the ethanol vehicle was tested earlier(Bourez et al. 2012b). Twelve hours later (day 11), some cellswere pooled in order to quantify the intracellular accumulationof PCB-153. Simultaneously, the rest of the cells wereinoculated with a PCB-153-free lipolytic medium supple-mented with 1 μM isoproterenol. The medium wasrenewed every 3 h. The amounts of PCB-153 were quantifiedevery 3 h in the extracellular medium and in the cells. Cellprotein levels were also quantified every 3 h (see “Cellularprotein assessment”).

PCB-153 quantification. Cells and extracellular mediumwerecollected in EPA vials (Alltech, Lokeren, Belgium) with 5 mlof n-hexane (Biosolve) in order to perform a liquid–liquid

extraction through a 10-min shaking. The hexane phase wastransferred into a tube and PCB-112 (Dr. Ehrenstorfer GmbH)was added as internal standard. All samples were then purifiedby acid and Florisil clean-up steps as described in Debier et al.(2003). Purified samples were collected in n-hexane. Fivemicroliters of anhydrous nonane (Sigma-Aldrich) was addedto the samples and the solvent was evaporated under gentlenitrogen flux. The purified extracts were suspended into ahexane solution of Mirex (200 pg/μl) (Dr. EhrenstorferGmbH) used as a standard for the injection. PCB congenerswere separated and quantified with a gas chromatograph (GCTrace, ThermoFinnigan) equipped with an automatic split/splitless type injector (CTCAnalytics, Zwingen, Switzerland),a fused silica capillary column (30 m×0.25 mm internaldiameter; 0.25 μm film) (Rxi-5 ms, Restek, Bellefonte, PA),and a mass spectrometer (Trace DSQ, ThermoFinnigan). Thesystem used helium as the carrier gas at a constant flow rate of1.1 ml per minute. The temperature of injector was 230°C.The oven temperature program was as follows: 2 min at 60°C,gradual heating from 60 to 140°C at the rate of 20°C perminute, 1 min at 140°C, gradual heating from 140 to 290°Cat the rate of 2.5°C per min, and 10 min at 290°C and gradualcooling from 290°C to 60°C at the rate of 10°C per min.Molecules were sent to mass spectrometer by the line transferat 290°C. The ion source of the detector was kept at 230°C.PCBs were identified according to their retention time. Datawere recorded using XCalibur 1.3 software (ThermoFinnigan).Quantification was performed by comparison to an externalstandard composed of 28 congeners (IUPAC numbers: CB-8, -18, -28, -44, -52, -66, -77, -81, -101, -105, -114, -118, -123, -126, -128, -138, -153, -156, -157, -167, -169, -170, -180, -187, -189, -195, -206, and -209) in a certified calibration mixture(AccuStandard, New Haven, CT). Five dilutions (concentra-tion ranging from 25 to 500 pg/μl) were used in order todraw a linear calibration curve for each PCB. Blanks wererun with sample series to control extraction and clean-upsteps. The PCB recovery was calculated on the basis ofthe internal standard, PCB-112. Results were acceptedonly if the recoveries were between 70 and 130%. Allresults were corrected to obtain 100% recovery (Debieret al. 2003). The quality control was assessed through aninterlaboratory comparison.

Cytotoxicity assessment. The potential cytotoxicities of thePCBs and the different lipolytic treatments were assessed bymeasuring the release of lactate dehydrogenase (LDH) in theextracellular medium. The activity of LDH was determinedwith the cytotoxicity detection kit (Roche Diagnostics,Vilvoorde, Belgium) according to the manufacturer’s instruc-tions in (i) the differentiation medium after 12 h of PCBexposure and (ii) the lipolytic media collected every 3 h duringthe lipolytic process. Before PCB exposure and lipolysis,some cells were lysed by exposure to 1% Triton X-100

LOUIS ETAL.

Author's personal copy

(Sigma-Aldrich) and were used as full toxicity control (Bourezet al. 2012a). No treatment appeared toxic (<5% of control) ascompared to the full toxicity control (results not shown).

Lipolysis time lapse. During the isolation of progenitor cells,some of them were seeded at a mean density of 18,000 cellsper cm2 in a tissue culture dish (Nunc, Langenselbold, Ger-many) (day 0). They underwent the same differentiation treat-ment until day 11 as previously described. The differentiationmedium was then removed and replaced by the lipolyticmedium described above and supplemented with 1 μM iso-proterenol. The culture dish was placed inside a cinemato-graphic chamber on the plate of an inverted microscope (20×magnification, Leitz, Sint-Niklaas, Belgium) as described inAlomar et al. (2008). A temperature of 36.5–37.5°C wasmaintained and a warmed and humidified air containing10% CO2 was frequently flushed into the chamber. Thetime-lapse recording equipment consisted of a color videocamera (Leica, Groot-Bijgaarden, Belgium) and a computer.This computer digitalized and recorded the frames every2 min for 12 h using the Leica FW 4000 software (Alomaret al. 2008). The frames from one lipolysis were used in orderto make a movie. Lipolytic medium supplemented with iso-proterenol was renewed every 3 h.

Statistical analysis. Data are presented as mean from 3 inde-pendent experiments ± SEM. The statistical analysis was per-formed by SAS 9.3 (SAS Institute Inc., Cary, NC). Data werelog transformed to achieve normality. Differences betweentreatments were assessed withmixed linear models and Tukey’stests. Differences were deemed significant at p values<0.05.

Results

Development of a lipolytic in vitro model. The aim of thisstudy was to develop a functional and performing model ofin vitro lipolysis by comparing the efficiency of several treat-ments on TGmobilization from differentiated rat adipocytes. Atday 11, the cells were incubated with a basal lipolytic mediumcontaining only 1.0 g/l glucose and 5% heat-inactivated FBS, ascompared to the differentiation medium, which was composedof 4.5 g/l glucose and 10% heat-inactivated FBS. Albumin wasadded to the lipolytic medium in order to bind the FFAsreleased in the culture medium.

Effect of isoproterenol stimulation. At day 11, differentiatedrat adipocytes were incubated with a lipolytic medium supple-mented (Fig. 1A) or not (Fig. 1B) with 1 μM of isoproterenol.Media from both conditions were discarded and renewedevery 3 h for 12 h. At each time point of the experiment, (i)cells from one plate were collected and assessed for total

protein and NL contents, and (ii) media were collected toquantify the amounts of released FFAs. Before lipolytic in-duction, adipocytes contained 4.94± 0.15 μmol NLs/mg pro-tein (Fig. 2A — initial condition).

After 3 h of lipolysis with isoproterenol stimulation, NLswere significantly reduced to 40% of initial level (p=0.003),whereas there was only a slight and nonsignificant reductionto 86% of the initial level for the adipocytes incubated with theisoproterenol-free lipolytic medium (p=0.989) (Fig. 2A). As aconsequence of the lipolytic process, the FFA levels increasedin the medium. Indeed, levels of released FFAs were 1.00±0.12 μmol/ml with isoproterenol supplementation and 0.22±0.13 μmol/ml with the isoproterenol-free lipolytic medium(p<0.001) (Fig. 2B). Sums of FFAs in the medium and thosepresent in the NLs of adipocytes after 3 h of experiment weresimilar to the initial condition (p=0.336 for the stimulatedcondition and p=0.671 for the unstimulated condition).

When examining the results over a longer period (up to12 h), the drop of NLs occurred mainly during the first half ofthe lipolytic process in the isoproterenol condition, whereas it

A

B

Figure 2. Over a short-term period, isoproterenol showed efficient lipo-lytic effect. Adipocytes were incubated with a lipolytic medium (LM)containing (w/Iso) or not (w/o Iso) 1 μM isoproterenol. After 3 h ofexperiment, cellular neutral lipids (A) as well as extracellular free fattyacids (B) were quantified. These results were compared to an initialcondition corresponding to differentiated rat adipocytes before the lipo-lytic induction. Data represent the means of three independent experi-ments ± SEM. Values with different symbols are significantly different(p≤0.01).

EFFICIENT IN VITRO ADIPOCYTE MODEL OF LONG-TERM LIPOLYSIS

Author's personal copy

was more progressive in the condition without isoproterenol(Fig. 3A). Levels of NLs were significantly lower in theisoproterenol-stimulated cells compared to cells without iso-proterenol at each studied time (p<0.001 at 3, 6, and 9 h andp=0.002 at 12 h). However, the difference between the twoconditions decreased as the lipolysis progressed (Fig. 3A).After 12 h, we measured 0.65± 0.41 μmol NLs/mg proteinin the stimulated condition and 0.98 ± 0.59 μmol NLs/mgprotein in the unstimulated condition. It represents a reductionof lipid contents of 87% and 80%, respectively, as comparedto the initial situation (Fig. 3A).

The total FFAs released by the adipocytes throughouta given period were calculated by adding the quantitiesmeasured at each period of 3 h. For example, the amountof total FFAs released after 6 h of lipolysis correspondsto the sum of FFAs released between 0 and 3 h andbetween 3 and 6 h (Fig. 3B). The FFAs released in theextracellular medium increased with time in both condi-tions. This increase was significant during the first 6 h oflipolysis in the stimulated condition (p<0.001 between 0

and 3 h; p=0.007 between 3 and 6 h; p=0.062 between 6and 9 h; p=0.563 between 9 and 12 h) and during thefirst 9 h of lipolysis in the condition without isoprotere-nol (p<0.001 between 0 and 3 h; p=0.010 between 3 and6 h; p=0.028 between 6 and 9 h; p=0.914 between 9 and12 h) (Fig. 3B). The amounts of fatty acids released inthe medium were significantly higher in the stimulatedcondition as compared to the condition without isopro-terenol during the first 9 h of the lipolytic process(p<0.001 at 3 and 6 h, p=0.048 at 9 h). This differencebecame nonsignificant at 12 h (p=0.088).

For a given treatment, the sums of fatty acids present incellular NLs and in media were statistically similar betweentwo consecutive periods for a given treatment (0.194<p<0.999).

Effects of medium renewal. In the present section, we com-pared the lipolytic efficiency of two treatments of 12 h. In thefirst one, the extracellular medium supplemented with 1 μMisoproterenol was replaced every 3 h as described previously(Fig. 1A). In the second one, fresh lipolytic medium supple-mented with 1 μM isoproterenol was added to the cells at thebeginning of the lipolysis treatment but was not renewedduring the 12 h of the lipolytic period (Fig. 1C). Again, every3 h for both treatments, media and cells were collected for theassessment of released FFAs, as well as of cellular protein andNLs.

During the first 3 h of lipolysis (Fig. 4A), both conditionscorresponded to the same treatment: addition of fresh lipolyticmedium supplemented with 1 μM isoproterenol for 3 h.Differences between treatments started from 3 h, with a me-dium renewal every 3 h. Asmentioned in the previous section,there was a drop of NLs in the condition with medium renew-al. By contrast, in the condition without medium renewal,adipocytes appeared to resynthesize TGs. Indeed, levels ofNLs, which are mainly composed of TGs, increased signifi-cantly with time (p=0.002 between 3 and 9 h). After 12 h,these cells contained 4.08± 0.13 μmol NLs/mg protein (cor-responding to 83% of the initial content) and up to 5.96±0.13 μmol NLs/mg protein (corresponding to 121% of theinitial content) after 24 h (results not shown). As a result, levelsof NLs were statistically different at 6, 9, and 12 h (p<0.001)between both conditions.

As mentioned earlier, the total FFAs released by the adipo-cytes throughout a given period were calculated for themedium renewal condition by adding the quantities measuredat each period of 3 h. In the conditionwithout medium renewal,total FFAs correspond to the amounts that were quantified atthe end of a given time period. FFAs were efficiently releasedin the medium during the first 3 h of lipolysis (Fig. 4B). Thequantities released into the medium continued to increase up to6 h in the condition with medium renewal. By contrast, the netrelease stopped in the condition without medium renewal

A

B

Figure 3. During a long-term period, renewal of lipolytic medium isrequired to maintain the lipolysis. Adipocytes were incubated with alipolytic medium (LM) containing (w/Iso) or not (w/o Iso) 1 μM isopro-terenol. The respective media were renewed every 3 h and the assess-ments of cellular neutral lipids (A) and extracellular free fatty acids (B)were carried out. Data represent the means of three independent experi-ments ± SEM. Consecutive values with no common symbol are signifi-cantly different (p≤0.05) within a defined treatment as well as within adefined time.

LOUIS ETAL.

Author's personal copy

(0.070<p<0.995). A slight, but nonsignificant drop was evenvisible (p=0.197). As a result, the amounts of FFAs measuredin the medium were lower in the condition without mediumrenewal (i.e., single initial isoproterenol stimulation) ascompared to the condition with medium renewal (i.e.,repeated isoproterenol stimulation). The difference becameprogressively more pronounced over time (p=0.137 at 6 h,p=0.002 at 9 h, and p<0.001 at 12 h).

Eventually, as shown previously, the sums of fatty acidsmeasured in NLs and inmedia were statistically similar withina given treatment (0.194<p<0.999).

Visual tracking of isoproterenol-stimulated lipolysis. We car-ried out a time course of the cells and made a time-lapse videofrom the frames obtained in order to visualize the lipolyticprocess (Electronic Supplementary Material). This videoshows the unloading of LDs for 12 h. As for both previousexperiments, lipolysis was triggered on differentiated

adipocytes at day 11 (Fig. 5A). The protocol used was theone with isoproterenol stimulation and medium renewal every3 h, as described above. After 3 h of isoproterenol-stimulatedlipolysis, the lipid content was halved (Fig. 5B). The mobili-zation rate was, however, not homogeneous among the cells.Indeed, whereas some adipocytes were already completelyempty (white arrow, Fig. 5B), others still contained significantamounts of lipids and a few of them remained with the samelipid content as the one at 0 h (black arrow, Fig. 5B). After 6(Fig. 5C) and 12 h (Fig. 5D), adipocytes kept mobilizing TGsleading to visibly emptied cells, which reflects the efficiencyof the experimental protocol.

Dynamics of mobilization of PCB-153 during lipolysis. Atday 10, differentiated rat adipocytes were incubated with adifferentiation medium supplemented with 300 nM PCB-153for 12 h. The congener accumulated within adipocytes,reaching a level of 1.9 ± 0.1 nmol/mg protein just before thelipolytic treatment. At day 11, lipolysis was induced withisoproterenol stimulation and medium renewal every 3 h, aspreviously described. The potential presence of PCBs in theinitial lipolytic medium, before being in contact with the cells,has been tested and was negligible (under the detection limit,results not shown). The amounts of PCBs in the lipolyticmedium at the beginning of lipolysis (0 h) were thus set atzero. The contents of PCB-153 were assessed in the cells andlipolytic medium at 3, 6, 9 and 12 h of lipolysis. The PCBlevels in the medium were calculated by adding the quantitiesmeasured at each period of 3 h. For example, the amounts ofPCB-153 released after 6 h of lipolytic treatment correspondto the sum of PCB-153 released between 0 and 3 h andbetween 3 and 6 h. PCB-153 was efficiently released fromadipocytes, leading to a significant reduction of its concentra-tion in the cells and a corresponding rise in the mediumbetween early and late lipolysis (p<0.001 between 0 and12 h) (Fig. 6A and B).

Discussion

Lipolysis in adipose tissue is a physiological process thatoccurs to provide fatty acids as metabolic fuel to an organism,either when the energy demand increases or when a deficiencyof substrate appears (Cherel and Le Maho 1985). In thepresent study, we developed a functional in vitro model ofadipocyte lipolysis. We tested several types of treatments inorder to investigate the effects of isoproterenol (acting as anonselective β-adrenergic agonist) and medium renewal oncell discharge over short-term (3 h) and longer-term (up to12 h) lipolysis.

This study highlights a rapid and significant lipolytic actionof the first isoproterenol dose for the first 3 h. Isoproterenolwas used to mimic the effects of catecholamines in vivo.

A

B

Figure 4. During a long-term period, isoproterenol is not sufficient tomaintain the lipolytic process. Adipocytes were incubated with a lipolyticmedium (LM) supplemented with 1 μM isoproterenol (w/Iso). Werenewed the lipolytic medium with isoproterenol stimulation on half ofthe culture plates every 3 h for 12 h. The other half of the culture platesunderwent the lipolysis with the initial lipolytic medium for 12 h, withoutmedium renewal. Cellular neutral lipids (A) and extracellular free fattyacids (B) were assessed. Data represent the means of three independentexperiments±SEM. Consecutive values with no common symbol aresignificantly different (p≤0.05) within a defined treatment as well aswithin a defined time.

EFFICIENT IN VITRO ADIPOCYTE MODEL OF LONG-TERM LIPOLYSIS

Author's personal copy

Indeed, it activates the classic intracellular pathway of thesehormones. For example, isoproterenol is known to increasethe activity of HSL by the PKA pathway (Anthonsen et al.1998) and to promote the translocation of HSL from thecytosol to the surface of the LDs (Ahmadian et al. 2010).Used at a concentration of 1 μM in the medium, it markedlyinduced the breakdown of TGs in the present study, as ob-served by the significant decrease of intracellular NLs and thecorresponding accumulation of FFAs in the extracellularmedium. The sum of fatty acids in NLs and mediummatched with the fatty acid content of NLs of initial adipo-cytes (i.e., at 0 h), confirming that extracellular FFAs camefrom the mobilization of TGs. These results are in accor-dance with the literature. Indeed, Nagayama et al. (2010)stimulated primary cultures of rat adipocytes with 0.1 μMisoproterenol and reported that phosphorylated HSL wasalready exclusively localized on the LD surface after only2 h of induction. In addition, Clifford et al. (2000) showed asignificant decrease of cytosolic HSL during the first minutesafter 1 μM isoproterenol stimulation in epididymal adipocytes.

During a long-term period (12 h), the effect of isoprotere-nol became less obvious. Instead, the regular renewal oflipolytic medium appeared to be the overriding factor. Indeed,after 12 h of lipolysis withmedium renewal, cellular NLs were

significantly reduced in both conditions (with and withoutisoproterenol). Thus it seems that, even without any exoge-nous catecholamine stimulation, the lipolytic pathway occursin adipocytes, provided that the medium is of lipolytic type(low levels or even absence of agents such as insulin, dexa-methasone, ciglitizone, and glucose) and that it is regularlyrenewed. Vargovic et al. (2011) demonstrated that rat adipo-cytes, isolated from mesenteric adipose tissue, producedcatecholamines de novo, showing that the sympathetic inner-vation is not the only source of catecholamines for the adiposetissue. This production could affect the mobilization of TGsin vitro as well, but further investigations using primarycultures of adipocytes are needed to confirm this hypothesis.In addition, Getty-Kaushik et al. (2005) showed that lipolysiswas enhanced in primary rat adipocytes when increasingamounts of BSA were added to the medium. This growingconcentration of BSA in the medium promoted the outwardmovement of FFAs. In the present study, a regular renewal oflipolytic medium, and thus a regular input of BSA, could alsopromote the lipolytic process within adipocytes. Even if theremaining cellular NLs were close in both conditions (withand without isoproterenol) after 12 h of lipolysis with mediumrenewal, the dynamics of cell emptying were different. Theincubation of adipocytes with repeated doses of isoproterenol

A B

C D

Figure 5. Visual observation ofthe efficient emptying of cellulartriglycerides during a 12-hlipolysis. Cells were incubatedwith a lipolytic mediumcontaining 1 μM isoproterenoland the medium was renewedevery 3 h. A, At 0 h, adipocyteswere fully filled with lipiddroplets. B, After 3 h, more thanhalf of the triglycerides weremobilized. Some adipocytes werealmost emptied of their lipids(white arrow), while some otherswere still filled (black arrow). C,After 9 h, some small lipiddroplets remained observable. D,After 12 h, only a little part of thelipids initially present in the cellswas still trapped into theadipocytes.

LOUIS ETAL.

Author's personal copy

induced a more rapid depletion of lipid stores than the incu-bation without isoproterenol. This may result in part from thefact that the use of lipolytic hormones such as catecholaminesaccelerates the process through the breaking down of lipiddroplets into smaller ones (Brasaemle 2007; Zechner andMadeo 2009), in addition to the effects on HSL and ATGL.A regular use of isoproterenol could thus make TGs moreavailable for the lipolytic enzymes.

Medium renewal allowed a regular input of BSA whileavoiding extended contact between the adipocytes and theproducts of lipolysis, which appeared to promote the lipolyticprocess. Indeed, when the lipolytic medium supplementedwith isoproterenol remained in contact with the adipocytesthroughout the experiment (condition without medium re-newal), we observed that the initially significant NL mobili-zation (and corresponding extracellular FFA accumulation)induced by the isoproterenol stimulation was followed by anincrease of cellular NLs and a slight decrease of FFAs in themedium. These results suggest a reuptake of FFAs from theextracellular medium by adipocytes and a reesterification intoTGs. Reesterification of fatty acids during lipolysis appears toreach significant levels in vivo, particularly in situations of slow

blood circulation (Raclot 2003; Reshef et al. 2003), which isin part reflected in our condition without medium renewal,where the adipocytes remain in contact with the culturemedium. The direct recycling of fatty acids is dependent on theavailability of glycerol-3-phosphate (G3P) and the expression ofreesterification enzymes. In in vivo conditions, glycerol releasedduring lipolysis cannot be recycled to synthesize G3P, seeing thenegligible levels of adipose glycerol kinase (Gyk) in adipocytes(Robinson and Newsholme 1967; Jéquier and Tappy 1999).When glucose is present, the G3P required for fatty acidreesterification andTG storage is rather obtained by the reductionof dihydroxyacetone phosphate, a glycolysis intermediate. Thelipolytic medium used here was composed of glucose (1 g/l) andFBS (5%), which contained among others insulin. Serum heatinactivation at 56°Cmight not have altered insulin, as this proteinappears to be stable at such temperature (Huus et al. 2005). As aresult, without catecholamine stimulation, insulin from heat-inactivated FBS could still promote the translocation of glucosetransporter type 4 from intracellular vesicles to the adipocytemembrane. Glucose could then be able to enter the adipocytesand undergo glycolysis in order to form G3P. In addition, aminoacids and pyruvate, being present in the lipolytic medium(6.21mMand 0.93mM, respectively), could also be at the originof G3P by the glyceroneogenesis pathway. Indeed, during star-vation, adipocytes can synthesize G3P from precursors otherthan glucose or glycerol by the glyceroneogenesis pathway(Duplus and Forest 2002; Nye et al. 2008). This pathway de-pends on the cytosolic phosphoenolpyruvate carboxykinase(PEPCK). Several studies highlighted that in 3T3-F442A adipo-cytes, isoproterenol was a rapid (maximum level obtained at 4 h)and potent (four to five folds) inducer of the PEPCK mRNA(Antras-Ferry et al. 1995; Forest et al. 1997; Tordjman et al.2003). Likewise, Antras-Ferry et al. (1995) have shown that oleicacid can stimulate PEPCK gene transcription. This fatty acid,which represents around 25% of the fatty acid profiles of adipo-cyte NLs (results not shown), might thus have positively influ-enced PEPCK’s gene transcription. A third possibility to obtainG3P is that in vitro adipocytes might express Gyk in particularculture conditions. Indeed, Guan et al. (2002) reported thatthiazolidinediones (TZD), acting as ligand for the nuclear recep-tor peroxisome proliferator-activated receptor gamma (PPARγ),induce Gyk expression. They have noted an important inductionof Gyk mRNA in TZD-stimulated 3T3-L1 adipocytes after 48 hof treatment. Lipolysis was then induced using 10 μM isopro-terenol and a marked incorporation of glycerol into TGs wasnoticed after 5 h of stimulation. A direct utilization of glycerol asa backbone for esterification of FFAs was suggested. Theproducts of lipolysis (glycerol and FFAs) were therebyutilized in the futile cycle allowed by Gyk. In our condi-tions, we cultivated adipocytes with 10 μM ciglitizone, aclass of TZD, for 11 d. Gyk activity might thus have beeninduced and involved in fatty acid reesterification. Thosedifferent pathways to obtain G3P in adipocytes might also

A

B

Figure 6. Mobilization of PCB-153 from adipocytes to extracellularmedium throughout a 12-h lipolytic process. Cells were previously incu-bated with a differentiation medium supplemented with 300 nM PCB-153. Twelve hours later, a lipolysis was performed with a lipolyticmedium containing 1 μM isoproterenol and renewed every 3 h. ThePCB-153 levels in adipocytes and extracellular medium were evaluated.Data represent the means of three independent experiments ± SEM.

EFFICIENT IN VITRO ADIPOCYTE MODEL OF LONG-TERM LIPOLYSIS

Author's personal copy

occur in the conditions with medium renewal. However,in such conditions, FFAs released in the culture mediumwere regularly removed and thus were not available forreuptake by the cells and reesterification into TGs.

Isoproterenol and regular lipolytic medium renewal arethus both efficient strategies to trigger the hydrolysis of cellu-lar triglycerides in in vitro adipocytes. This protocol can beused as a tool to study the behavior of lipophilic moleculesduring lipolysis. Indeed, during periods of negative energybalance, lipophilic pollutants such as PCBs or PBDEs, whichare stored in the adipose tissue (La Merrill et al. 2013; Rooset al. 2013), might be mobilized into the bloodstream. Severalin vivo studies report increases of pollutant concentrations inthe blood circulation during body weight loss episodes(Debier et al. 2006; Kim et al. 2011; Dirtu et al. 2013), raisingpotential toxic impacts on the individual’s health. In thepresent study, we showed that a large amount of PCB-153,which accumulated in differentiated adipocytes, was releasedfrom the cells throughout the lipolytic process and concentrat-ed in the extracellular medium. Further investigations couldbe undertaken in order to study the mechanisms involved inthe mobilization of PCBs as well as other lipophilic pollutantsfrom adipocytes.

Conclusion

In conclusion, isoproterenol showed a fast and efficient lipo-lytic effect during a short-term period (3 h). However, over alonger term (12 h), it was not sufficient to maintain thebreakdown of TGs. Indeed, with a single initial dose ofisoproterenol, in the absence of medium renewal, adipocytesappeared to newly synthesize TGs through the recycling ofFFAs released in the medium. On the contrary, regular mediumrenewal appeared to be the most significant factor involved incell discharge, with or without isoproterenol, leading to animportant hydrolysis of TGs over 12 h of lipolysis. This effi-cient in vitro model allows mimicking an important mobiliza-tion of lipids. It is a useful tool to study the cellular mechanismsinvolved in the release of lipophilic compounds (lipophilicpollutants, vitamins, etc.) from adipocytes during a drasticlipolysis.

Acknowledgments The authors are very grateful to Marie-ThérèseAhn, Daniel Jal, Willy Marteau, and Philippe Bombaerts from “Institutdes Sciences de la Vie” (ISV), UCLouvain, for their technical assistance.We also thank Pauline Beguin, Julie Winand, and Anne-CatherineSchneider from ISV for their help in the fatty acid analyses. Membersof “Support en méthodologie et calcul statistique” (Institutmultidisciplinaire pour la modélisation et l’analyse quantitative,UCLouvain, Belgium) are gratefully acknowledged for the collaborationin the statistical analyses. We also greatly appreciated the help and adviceof Guillaume Bernard for picture processing.

References

Ahmadian M.; Wang Y.; Sul H. S. Lipolysis in adipocytes. Int. J.Biochem. Cell Biol. 42: 555–559; 2010.

Alomar M.; Tasiaux H.; Remacle S.; George F.; Paul D.; Donnay I.Kinetics of fertilization and development, and sex ratio of bovineembryos produced using the semen of different bulls. Anim. Reprod.Sci. 107: 48–61; 2008.

Anthonsen M. W.; Rönnstrand L.; Wernstedt C.; Degerman E.; Holm C.Identification of novel phosphorylation sites in hormone-sensitivelipase that are phosphorylated in response to isoproterenol andgovern activation properties in vitro. J. Biol. Chem. 273: 215–221;1998.

Antras-Ferry J.; Robin P.; Robin D.; Forest C. Fatty acids and fibrates arepotent inducers of transcription of the phosphenol pyruvatecarboxykinase gene in adipocytes. Eur. J. Biochem. 234: 390–396;1995.

Belfrage P.; Fredrikson G.; Nilsson N. O.; Stralfors P. Regulation ofadipose-tissue lipolysis by phosphorylation of hormone-sensitivelipase. Int. J. Obes. 5: 635–641; 1981.

Bourez S.; Joly A.; Covaci A.; Remacle C.; Larondelle Y.; Schneider Y.-J.;Debier C. Accumulation capacity of primary cultures of adipocytesfor PCB-126: influence of cell differentiation stage and triglyceridelevels. Toxicol. Lett. 214: 243–250; 2012a.

Bourez S.; Le Lay S.; Van den Daelen C.; Louis C.; Larondelle Y.; ThoméJ.-P.; Schneider Y.-J.; Dugail I.; Debier C. Accumulation ofpolychlorinated biphenyls in adipocytes: selective targeting to lipiddroplets and role of caveolin-1. PLoS. One. 7: e31834; 2012b.

Brasaemle D.; Subramanian V.; Garcia A.; Marcinkiewicz A.;Rothenberg A. Perilipin A and the control of triacylglycerol metab-olism. Mol. Cell. Biochem. 326: 15–21; 2009.

Brasaemle D. L. Thematic review series: adipocyte biology. The perilipinfamily of structural lipid droplet proteins: stabilization of lipiddroplets and control of lipolysis. J. Lipid Res.48: 2547–2559; 2007.

Brasaemle D. L.; Dolios G.; Shapiro L.; Wang R. Proteomic analysis ofproteins associated with lipid droplets of basal and lipolyticallystimulated 3T3-L1 adipocytes. J. Biol. Chem. 279: 46835–46842;2004.

Brasaemle D. L.; Levin D. M.; Adler-Wailes D. C.; Londos C. Thelipolytic stimulation of 3T3-L1 adipocytes promotes the transloca-tion of hormone-sensitive lipase to the surfaces of lipid storagedroplets. Biochim. Biophys. Acta 1483: 251–262; 2000.

Chaves V. E.; Frasson D.; Kawashita N. H. Several agents and pathwaysregulate lipolysis in adipocytes. Biochimie 93: 1631–1640; 2011.

Cherel Y.; Le Maho Y. Five months of fasting in king penguin chicks:body mass loss and fuel metabolism. Am. J. Physiol. Regul. Integr.Comp. Physiol. 249: 387–392; 1985.

Chevrier J.; Dewailly E.; Ayotte P.; Mauriege P.; Despres J. P.; TremblayA. Body weight loss increases plasma and adipose tissue concen-trations of potentially toxic pollutants in obese individuals. Int. J.Obes. Relat. Metab. Disord. 24: 1272–1278; 2000.

Clifford G. M.; Londos C.; Kraemer F. B.; Vernon R. G.; Yeaman S. J.Translocation of hormone-sensitive lipase and perilipin upon lipolyticstimulation of rat adipocytes. J. Biol. Chem. 275: 5011–5015; 2000.

Covaci A.; Voorspoels S.; Roosens L.; Jacobs W.; Blust R.; Neels H.Polybrominated diphenyl ethers (PBDEs) and polychlorinated bi-phenyls (PCBs) in human liver and adipose tissue samples fromBelgium. Chemosphere 73: 170–175; 2008.

de Jong J. C.; Sørensen L. G.; Tornqvist H.; Jacobsen P. Carbazates aspotent inhibitors of hormone-sensitive lipase. Bioorg. Med. Chem.Lett. 14: 1741–1744; 2004.

Debier C.; Chalon C.; Le Bœuf B. J.; de Tillesse T.; Larondelle Y.; ThoméJ.-P. Mobilization of PCBs from blubber to blood in northern ele-phant seals (Mirounga angustirostris) during the post-weaning fast.Aquat. Toxicol. 80: 149–157; 2006.

LOUIS ETAL.

Author's personal copy

Debier C.; Crocker D. E.; Houser D. S.; Vanden Berghe M.; Fowler M.;Mignolet E.; de Tillesse T.; Rees J.-F.; Thomé J.-P.; Larondelle Y.Differential changes of fat-soluble vitamins and pollutants duringlactation in northern elephant seal mother-pup pairs. Comp.Biochem. Physiol. A 162: 323–330; 2012.

Debier C.; Pomeroy P. P.; Dupont C.; Joiris C.; Comblin V.; Boulengé E.L.; Larondelle Y.; Thomé J.-P. Quantitative dynamics of PCB trans-fer from mother to pup during lactation in UK grey sealsHalichoerus grypus. Mar. Ecol. Prog. Ser. 247: 237–248; 2003.

Dirtu A. C.; Dirinck E.; Malarvannan G.; Neels H.; Van Gaal L.; Jorens P.G.; Covaci A. Dynamics of organohalogenated contaminants inhuman serum from obese individuals during one year of weight losstreatment. Environ. Sci. Technol. 47: 12441–12449; 2013.

Duplus E.; Forest C. Is there a single mechanism for fatty acid regulationof gene transcription? Biochem. Pharmacol. 64: 893–901; 2002.

Egan J. J.; Greenberg A. S.; Chang M. K.; Wek S. A.; Moos M. C.;Londos C. Mechanism of hormone-stimulated lipolysis in adipo-cytes: translocation of hormone-sensitive lipase to the lipid storagedroplet. Proc. Natl. Acad. Sci. U. S. A. 89: 8537–8541; 1992.

Faroon O.; Jones D.; De Rosa C. Effects of polychlorinated biphenyls onthe nervous system. Toxicol. Ind. Health 16: 305–333; 2000.

Forest C.; Franckhauser S.; Glorian M.; Antras-Ferry J.; Robin D.; RobinP. Regulation of gene transcription by fatty acids, fibrates andprostaglandins: the phosphoenolpyruvate carboxykinase geneas a model. Prostaglandins. Leukot. Essent. Fat. Acids. 57:47–56; 1997.

Gallenberg L. A.; Ring B. J.; Vodicnik M. J. Influence of lipolysis on themobilization of 2,4,5,2′4′,5′‐hexachlorobiphenyl from adipocytes invitro. J. Toxicol. Environ. Health 20: 163–171; 1987.

Gallenberg L. A.; Vodicnik M. J. Potential mechanisms for redistributionof polychlorinated biphenyls during pregnancy and lactation.Xenobiotica 17: 299–310; 1987.

Getty-Kaushik L.; Richard A.-M. T.; Corkey B. E. Free fatty acid regu-lation of glucose-dependent intrinsic oscillatory lipolysis inperifused isolated rat adipocytes. Diabetes 54: 629–637; 2005.

Golden R. J.; Noller K. L.; Titus-Ernstoff L.; Kaufman R. H.; MittendorfR.; Stillman R.; Reese E. A. Environmental endocrine modulatorsand human health: an assessment of the biological evidence. Crit.Rev. Toxicol. 28: 109–227; 1998.

Goldrick R.Morphological changes in the adipocyte during fat depositionand mobilization. Am. J. Physiol. 212: 777–782; 1967.

Guan H.-P.; Li Y.; JensenM. V.; Newgard C. B.; Steppan C.M.; Lazar M.A. A futile metabolic cycle activated in adipocytes by antidiabeticagents. Nat. Med. 8: 1122–1128; 2002.

Harmancey R.; Senard J.-M.; Pathak A.; Desmoulin F.; Claparols C.;Rouet P.; Smih F. The vasoactive peptide adrenomedullin is secretedby adipocytes and inhibits lipolysis through NO-mediated β-adrenergic agonist oxidation. FASEB J. 19: 1045–1047; 2005.

Huus K.; Havelund S.; Olsen H. B.; van de Weert M.; Frokjaer S.Thermal dissociation and unfolding of insulin. Biochemistry(Mosc) 44: 11171–11177; 2005.

Imbeault P.; Chevrier J.; Dewailly E.; Ayotte P.; Desprs J. P.; Maurige P.;Tremblay A. Increase in plasma pollutant levels in response toweight loss is associated with the reduction of fasting insulin levelsin men but not in women. Metabolism 51: 482–486; 2002.

Jéquier E.; Tappy L. Regulation of body weight in humans. Physiol. Rev.79: 451–480; 1999.

Khoo J. C.; Fong W. W.; Steinberg D. Activation of hormone-sensitive lipase from human adipose tissue by cyclic AMP-dependent protein kinase. Biochem. Biophys. Res. Commun.49: 407–413; 1972.

Kim M. J.; Marchand P.; Henegar C.; Antignac J. P.; Alili R.; Poitou C.;Bouillot J. L.; Basdevant A.; Le Bizec B.; Barouki R.; Clément K.Fate and complex pathogenic effects of dioxins and polychlorinatedbiphenyls in obese subjects before and after drastic weight loss.Environ. Health Perspect. 119: 377–383; 2011.

Kuriyama S. N.; Chahoud I. In utero exposure to low-dose 2,3′,4,4′,5-pentachlorobiphenyl (PCB 118) impairs male fertility and altersneurobehavior in rat offspring. Toxicology 202: 185–197; 2004.

La Merrill M.; Emond C.; Kim M. J.; Antignac J. P.; Le Bizec B.;Clément K.; Birnbaum L. S.; Barouki R. Toxicological function ofadipose tissue: focus on persistent organic pollutants. Environ.Health Perspect. 121: 162–169; 2013.

Lafontan M.; Langin D. Lipolysis and lipid mobilization in humanadipose tissue. Prog. Lipid Res. 48: 275–297; 2009.

Large V.; Peroni O.; Letexier D.; Ray H.; Beylot M. Metabolism of lipidsin human white adipocyte. Diabetes Metab. 30: 294–309; 2004.

Lass A.; Zimmermann R.; Haemmerle G.; Riederer M.; Schoiswohl G.;Schweiger M.; Kienesberger P.; Strauss J. G.; Gorkiewicz G.;Zechner R. Adipose triglyceride lipase-mediated lipolysis of cellularfat stores is activated by CGI-58 and defective in Chanarin-DorfmanSyndrome. Cell. Metab. 3: 309–319; 2006.

Lass A.; Zimmermann R.; Oberer M.; Zechner R. Lipolysis — a highlyregulated multi-enzyme complex mediates the catabolism of cellularfat stores. Prog. Lipid. Res. 50: 14–27; 2011.

Le Lay S.; Ferre P.; Dugail I. Adipocyte cholesterol balance in obesity.Biochem. Soc. Trans. 32: 103–106; 2004.

Lee S. K.; Ou Y. C.; Yang R. S. H. Comparison of pharmacokineticinteractions and physiologically based pharmacokinetic modeling ofPCB 153 and PCB 126 in nonpregnant mice, lactating mice, andsuckling pups. Toxicol. Sci. 65: 26–34; 2002.

Ludewig G.; Lehmann L.; Esch H.; Robertson L.W.Metabolic activationof PCBs to carcinogens in vivo—a review. Environ. Toxicol.Pharmacol. 25: 241–246; 2008.

Meeker J. D.; Maity A.; Missmer S. A.; Williams P. L.; Mahalingaiah S.;Ehrlich S.; Berry K. F.; Altshul L.; Perry M. J.; Cramer D. W.;Hauser R. Serum concentrations of polychlorinated biphenyls inrelation to in vitro fertilization outcomes. Environ. HealthPerspect. 119: 1010–1016; 2011.

NagayamaM.; Shimizu K.; Taira T.; Uchida T.; Gohara K. Shrinking anddevelopment of lipid droplets in adipocytes during catecholamine-induced lipolysis. FEBS Lett. 584: 86–92; 2010.

Nye C.; Kim J.; Kalhan S. C.; Hanson R. W. Reassessing triglyceridesynthesis in adipose tissue. Trends Endocrin. Met. 19: 356–361;2008.

Raclot T. Selective mobilization of fatty acids from adipose tissue triac-ylglycerols. Prog. Lipid Res. 42: 257–288; 2003.

Rapold R. A.; Wueest S.; Knoepfel A.; Schoenle E. J.; Konrad D. Fasactivates lipolysis in a Ca2+-CaMKII-dependent manner in 3T3-L1adipocytes. J. Lipid Res. 54: 63–70; 2013.

Reshef L.; Olswang Y.; Cassuto H.; Blum B.; Croniger C. M.; Kalhan S.C.; Tilghman S. M.; Hanson R. W. Glyceroneogenesis and thetriglyceride/fatty acid cycle. J. Biol. Chem. 278: 30413–30416;2003.

Ring B. J.; Seitz K. R.; Gallenberg L. A.; VodicnikM. J. The effect of dietand litter size on the elimination of 2,4,5,2′,4′,5′-[14C]hexachlorobiphenyl from lactating mice. Toxicol. Appl. Pharmacol.104: 9–16; 1990.

Robinson J.; Newsholme E. A. Glycerol kinase activities in rat heart andadipose tissue. Biochem. J. 104: 2C–4C; 1967.

Rodbell M. Metabolism of Isolated Fat Cells. J. Biol. Chem. 239: 375–380; 1964.

Rogan W. J.; Gladen B. C.; McKinney J. D.; Carreras N.; Hardy P.;Thullen J.; Tingelstad J.; Tully M. Polychlorinated biphenyls(PCBs) and dichlorodiphenyl dichloroethene (DDE) in human milk:effects of maternal factors and previous lactation. Am. J. PublicHealth 76: 172–177; 1986.

Roos V.; Rönn M.; Salihovic S.; Lind L.; Bavel B.; Kullberg J.;Johansson L.; Ahlström H.; Lind P. M. Circulating levels ofpersistent organic pollutants in relation to visceral and sub-cutaneous adipose tissue by abdominal MRI. Obesity 21:413–418; 2013.

EFFICIENT IN VITRO ADIPOCYTE MODEL OF LONG-TERM LIPOLYSIS

Author's personal copy

Schneider A.-C.; Beguin P.; Bourez S.; Perfield II J. W.; Mignolet E.;Debier C.; Schneider Y.-J.; Larondelle Y. Conversion of t11t13 CLAinto c9t11 CLA in Caco-2 cells and inhibition by sterculic oil. PLoSONE 7: e32824; 2012.

Thompson B. R.; Lobo S.; Bernlohr D. A. Fatty acid flux in adipocytes:the in’s and out’s of fat cell lipid trafficking. Mol. Cell. Endocrinol.318: 24–33; 2010.

Tordjman J.; Khazen W.; Antoine B.; Chauvet G.; Quette J.; Fouque F.;Beale E. G.; Benelli C.; Forest C. Regulation of glyceroneogenesisand phosphoenolpyruvate carboxykinase by fatty acids, retinoicacids and thiazolidinediones: potential relevance to type 2 diabetes.Biochimie 85: 1213–1218; 2003.

Van Dang Q. C.; Focant M.; Mignolet E.; Turu C.; Froidmont E.;Larondelle Y. Influence of the diet structure on ruminalbiohydrogenation and milk fatty acid composition of cows fedextruded linseed. Anim. Feed Sci. Technol. 169: 1–10; 2011.

Vanden BergheM.; Mat A.; Arriola A.; Polain S.; Stekke V.; Thomé J.-P.;Gaspart F.; Pomeroy P.; Larondelle Y.; Debier C. Relationshipsbetween vitamin A and PCBs in grey seal mothers and pups duringlactation. Environ. Pollut. 158: 1570–1575; 2010.

Vanden Berghe M.; Weijs L.; Habran S.; Das K.; Bugli C.; Rees J.-F.;Pomeroy P.; Covaci A.; Debier C. Selective transfer of persistent

organic pollutants and their metabolites in grey seals during lacta-tion. Environ. Int. 46: 6–15; 2012.

Vargovic P.; Ukropec J.; Laukova M.; Cleary S.; Manz B.; Pacak K.;Kvetnansky R. Adipocytes as a new source of catecholamine pro-duction. FEBS Lett. 585: 2279–2284; 2011.

Wang S.; Soni K. G.; Semache M.; Casavant S.; Fortier M.; PanL.; Mitchell G. A. Lipolysis and the integrated physiology oflipid energy metabolism. Mol. Genet. Metab. 95: 117–126;2008.

Wassermann M.; Wassermann D.; Cucos S.; Miller H. J. World PCBsmap: storage and effects in man and his biologic environment in the1970s. Ann. N. Y. Acad. Sci. 320: 69–124; 1979.

Zechner R.; Madeo F. Cell biology: another way to get rid of fat. Nature458: 1118–1119; 2009.

Zhou L.; Wang X.; Yang Y.; Wu L.; Li F.; Zhang R.; Yuan G.; Wang N.;Chen M.; Ning G. Berberine attenuates cAMP-induced lipolysis viareducing the inhibition of phosphodiesterase in 3T3-L1 adipocytes.Biochim. Biophys. Acta (BBA)-Mol. Basis Dis. 1812: 527–535;2011.

Zimmermann R.; Lass A.; Haemmerle G.; Zechner R. Fate of fat: the roleof adipose triglyceride lipase in lipolysis. Biochim. Biophys. Acta1791: 494–500; 2009.

LOUIS ETAL.

Author's personal copy