397

Molecular and Cellular Biochemistry 252: 397–400, 2003.© 2003 Kluwer Academic Publishers. Printed in the Netherlands.

Resistin expression in different adipose tissuedepots during rat development

Paula Oliver, Catalina Picó, Francisca Serra and Andreu PalouLaboratori de Biologia Molecular, Nutrició i Biotecnologia, Departament de Biologia Fonamental i Ciències de la Salut,Universitat de les Illes Balears, Palma de Mallorca, Spain

Received 21 October 2002; accepted 17 February 2003

Abstract

Resistin is a hormonal factor synthesised by adipocytes that was first thought to be related with the resistance to insulin inobesity, but whose function is not yet completely established. Here we have studied the ontogenic pattern of resistin mRNAexpression in different white adipose tissue depots (WAT) – epididymal, inguinal, mesenteric and retroperitoneal – and in brownadipose tissue (BAT), as well as the circulating resistin levels, in rats of different ages (from the suckling period to one year ofage). Resistin mRNA was determined by Northern blotting, and serum levels by enzyme immunoassay. In WAT, resistin ex-pression remains almost constant with age, except in early development, where there is a peak of expression in the epididymaland retroperitoneal depots, and a decrease in the inguinal one, while the expression remains constant for the mesenteric depot.Moreover, there is a site-specific difference regarding resistin expression: all the depots express characteristic levels of mRNA,especially at the age of 2 months, the moment when resistin mRNA levels are significantly higher in the epididymal and theretroperitoneal than in the inguinal and mesenteric WAT and than in the BAT. The transient increased resistin expression in theepididymal and the retroperitoneal WAT at a period of time in which there is a change in diet (from milk to chow) suggests acommon nutritional regulation of the resistin gene. Circulating resistin levels increase with age probably reflecting the increasein the body fat content. (Mol Cell Biochem 252: 397–400, 2003)

Key words: resistin, adipose tissue depots, rat development

Short communication

Introduction

Resistin is a 12.5 kDa cysteine-rich polypeptide secreted byadipocytes that was first described as a hormonal signalfrom the adipose tissue involved in the induction of insulinresistance [1]. Serum resistin levels were found to be sig-nificantly elevated in obese and diabetic mice; neutraliza-tion of resistin with antibodies improved insulin sensitivity,and resistin gene expression was down-regulated both invivo (epididymal fat) and in vitro (3T3-L1 adipocytes), af-ter treatment with thiazolidinediones (TZDs) – peroxisomeproliferator-activated receptor γ (PPARγ) agonists with aninsulin-sensitizing action – [1]. These findings suggestedthat resistin serves as a hormonal link between obesity and

insulin resistance, but the precise function of resistin pro-duction by adipocytes remains still unknown. Recently, ithas been shown that resistin is also synthesised in mousebrain and pituitary gland, and it has been hypothesized thatit has multiple roles in non-adipose tissues [2].

Regarding resistin expression in adipose tissue, it appearsthat, in insulin resistant circumstances, its expression shouldbe elevated. However, on the contrary, data from severalgroups working at the mRNA level with different insulinresistant animal models (obese and diabetic animals) foundresistin expression to be decreased [3–5]. Another controver-sial point is the effect of PPARγ agonists upon resistin ex-pression. The first described down-regulation of resistinmRNA expression by TZDs [1] was also demonstrated in

Address for offprints: A. Palou, Departament de Biologia Fonamental i Ciències de la Salut, Universitat de les Illes Balears, Cra. Valldemossa Km 7.5.E-07071-Palma de Mallora, Spain (E-mail: andreu.palou@uib.es)

398

vitro working with 3T3-L1 adipocytes [6], while other datain vivo show that treatment with TZDs increases epididymalresistin mRNA levels in obese mice and diabetic rats [3].

Only sparse data in a few selected depots regarding resistinpattern of expression in the different adipose tissue depotsare available. Resistin levels seem to be much higher in WATthan in BAT [1, 6, 7]. Concerning WAT, some groups showedthat there were no differences in resistin mRNA levels be-tween the different depots [6, 7], while others showed a higherexpression in the gonadal adipose tissue [1]. We have shownthat proteins involved in body weight control and expressedby adipose tissue, such as leptin and the uncoupling proteins,are differently expressed in the fat depots and, moreover, thateach depot has a characteristic ontogenic pattern [8, 9]. Asfar as we know, there is no work describing developmentalchanges in resistin mRNA expression with age.

In this study, the ontogenic pattern of circulating resistinlevels and of resistin gene expression is reported in differentWAT depots (epididymal, inguinal, mesenteric and retroperi-toneal) and BAT in rats, covering from the suckling periodto 1 year of age.

Materials and methods

Animals

Male Wistar rats (CRIFFA, Barcelona, Spain) acclimated to22°C with a 12 h light/12 h dark cycle and free access to astandard chow diet (Panlab, Barcelona, Spain) containing:23.5% proteins, 4.3% fat, 3.7% fibre, 5.8% minerals, 51%carbohydrates and 12% water. Rats (n = 5) were weighed andthen killed by decapitation at different ages: 18 days, 2 months(55 days), 3 months (93 days), 6 months (between 159–212days) and 11 months (between 294–355 days). After killingthe animals the interscapular brown adipose tissue (BAT) andthe epididymal, retroperitoneal, mesenteric and inguinalwhite adipose tissue depots (EWAT, RWAT, MWAT andIWAT, respectively) were rapidly removed in their entirety,weighed, frozen in liquid nitrogen and stored at –70°C untilRNA analysis. Guidelines for the use and care of laboratoryanimals of our University were followed.

Northern blot analysis

Tissue samples were homogenised in Tripure reagent andtotal RNA extracted following the instructions provided byBoehringer Mannheim. Twenty µg of total RNA, denaturedwith formamide/formaldehyde, were fractionated by agarosegel electrophoresis as previously described [10]. The RNAwas then transferred onto a Hybond Nylon membrane in 20× SSC (saline sodium citrate buffer: 1 × SSC in 150 mM

NaCl, 15 mM sodium citrate, pH 7.0) by capillary blottingfor 16 h, and fixed with UV light [10].

The mRNA for resistin was detected by a chemilumines-cence-based procedure, using a 34-mer antisense oligo-nucleotide probe (5′-TCCCACGAGCCACAGGCAGAGCCACAGGAGCAGC-3′) which was synthesised commer-cially (TIB MOLBIOL), labelled at both ends with a singledigoxigenin ligand. CDP-Star™ (Roche) was used to visual-ize the signals, exposing the membranes to Hyperfilm™ ECL(Amersham, Buckinghamshire, UK) for up to 45 min. Bandsin films were analysed by scanner photodensitometry andquantified using the KODAK 1D Image Analysis Software3.5 (Kodak). Finally, blots were stripped and re-probed for18S rRNA as previously described [10], to check the load-ing and transfer of RNA during blotting.

Duplicates of RNA isolation and northern blot analysiswere performed for all samples.

Quantification of resistin levels

Serum resistin concentration was measured with a humanenzyme immunoassay (EIA) kit (Phoenix Pharmaceuticals,Inc., California, USA).

Statistical analysis

All data are expressed as the mean ± S.E.M. The statisticalsignificance of differences in resistin circulating levels for thedifferent ages was assessed by one-way analysis of variance(ANOVA) and the differences in resistin mRNA expressionbetween the adipose tissue depots for the different ages byone-way and two-way ANOVA and least significant differ-ence (LSD) post hoc comparisons. Threshold of significancewas defined at p < 0.05.

Results and discussion

In adult rats (after the 3 months of life), resistin expressionin the different WAT depots studied remains practically un-changed throughout the period studied. However, it is possi-ble to distinguish two levels of resistin expression (p < 0.05,two-way ANOVA): BAT and retroperitoneal depots showinghigher resistin expression than the epididymal, inguinal andmesenteric WAT depots (Fig. 1). On the contrary, before theage of 3 months there is a different profile between the fatdepots. In the inguinal WAT there is a decrease in resistinmRNA levels at the age of 2 months (p < 0.05, one-wayANOVA compared to values at 18 days), while the epidi-dymal and the retroperitoneal WAT depots have a transientpeak of expression at that time, both following the same be-

399

haviour (p < 0.05, two-vay ANOVA) (Figs 1 and 2). Inter-estingly, a similar ontogenical behaviour in these two depotscan also be found when we consider other adipocytokinesproduced by the adipose tissue, such as leptin [8]. BAT, onthe other hand, has a stable resistin expression during devel-opment (p < 0.05, two-way ANOVA), although there is a ten-dency to increase resistin expression at 3 months. Regardingresistin circulating levels, there is an increase during devel-opment (p < 0.05, one-way ANOVA), probably related withthe increase in body fat content [8], until the age of 6 months,remaining constant thereafter (Fig. 3).

Rats are lactating until day 21, when they change to a chowdiet. This change from a fat-rich diet to a carbohydrate-rich dietimplies a wide range of metabolic adaptations. Little is knownfor the moment about regulation of resistin expression in adi-pose tissue, but some results show that there is a tight nutri-tional and hormonal control [7]. Initial studies on the regulationof resistin production have indicated that the expression of thegene is reduced by fasting [1, 7], and there is a rapid increasein expression on refeeding fasted mice with a high carbohy-drate diet, a situation where there is an increase in circulatingleptin levels [7]. In fact, insulin has been reported to stimulate,in vivo, adipose tissue resistin gene expression in diabetic miceand rats [3, 7]. Moreover, several data in vitro indicate that freefatty acids inhibit gene expression of resistin [4], while otherstudies in vivo show, that the lipid content of the diet is not amajor regulator of resistin expression [5].

The first two months of life is a period of changes in theexpression of other genes that are also expressed in the adi-pose tissue such as leptin [8] and UCP2 [9]. In two-month-old rats, leptin mRNA expression increases drastically incomparison with the basal levels present in suckling rats, withlevels remaining high during the rest of the development, and

this increase is only seen in the epididymal and the retro-peritoneal fat depots and not in the mesenteric, and inguinalones [8]. UCP2 mRNA expression is also affected at thisage, showing a transient peak at 2 months in different WATdepots (epididymal, inguinal, mesenteric and retroperitoneal)[9]. The changes in the expression of these genes in the adi-pose tissue have been attributed in both cases to the changein the feeding schedule, reflecting a possible common regu-lation modulated by feeding and food composition.

Resistin expression is site-specific, particularly at the ageof 2 months. At this time, there is a significant higher resistinexpression (p < 0.05, one-way ANOVA) in the epididymaland retroperitoneal WAT compared to the other depots (Fig.2B). This higher expression of resistin in the epididimal andthe retroperitoneal depots at 2 months is not accompanied bya peak in circulating resistin levels (Fig. 3). There are analo-gous differences between depots in leptin mRNA expression,with much higher expression occurring in the internal fatdepots than in the subcutaneous sites [8]. Previous studieshave shown higher WAT resistin mRNA expression comparedto BAT [1, 7]. In our study BAT clearly expressed resistin lev-els which were as high as in the WAT depots during the wholeperiod studied (Fig. 1). In fact, the levels in BAT during de-velopment were even slightly higher when compared to someof the WAT studied depots.

0

50

100

150

200

250

300

350

400

450

500

0 2 4 6 8 10

Age (months)

Res

isti

n m

RN

A (

Au

)

BAT

EWAT

IWAT

MWAT

RWAT

ATAxT

Fig. 1. Resistin mRNA expression levels in interscapular brown adiposetissue (BAT), epididymal (EWAT), inguinal (IWAT), mesenteric (MWAT)and retroperitoneal (RWAT) white adipose tissue depots in rats at differ-ent ages (from 18 days to 12 months) measured by Northern blotting. Re-sults represent mean ± S.E.M. (n = 5) of ratios of specific mRNA levels to18S rRNA (arbitrary units, Au). A – effect of age; T – effect of adipose tissuedepot; AxT – interaction of age and adipose tissue depot (two-way ANOVA,p < 0.05).

Resistin mRNA

18S rRNA

18 days 2 3 6 11

months

Epididymal WAT

Resistin mRNA

18S rRNA

BAT Epididymal Inguinal Mesenteric Retroperitoneal

WAT

2 months

A

B

Fig. 2. Representative Northern blot for resistin mRNA expression. Twentyµg of the total RNA were used for specific determination of the resistinmRNA, using 18S rRNA as a control for quantity of RNA. (A) ResistinmRNA in the epididymal white adipose tissue (WAT) depot of rats at dif-ferent ages. (B) Resistin mRNA in the epididymal, inguinal, mesenteric andretroperitoneal white adipose tissue (WAT) in 2-month-old rats (55 days).

– – – –

400

Our data may help to clarify contradictory resistin expres-sion levels in the different WAT depots found in bibliogra-phy. Kim et al. [7] detected resistin mRNA at similar levelsin all regions of rat white adipose tissues including epidiymaland inguinal fat pads. Results of Haugen et al. [6] in miceagree with these ones: they found that the expression ofresistin mRNA was as high in mice subcutaneous white fatas in the perirenal and epididymal depots. On the other hand,other results in mice demonstrate that resistin mRNA levelsvary as a function of the white adipose depot, with the high-est level of expression in gonadal fat [1]. As we have detected,there is a marked site-specific difference in resistin expres-sion in 2-month-old rats, but it is not so clear at older ages.Two-month old rats showed higher resistin mRNA levels ingonadal and retroperitoneal depots than in the other stud-ied depots. The age of the animals used in previous studies(which is not specified) could explain the discrepancy amongdata.

As we have already mentioned, resistin function remainsstill unknown. Apart from its controversial possible role inconnecting obesity and insulin resistance, the fact that resistinexpression in adipocytes is highly increased during the latestage of adipogenesis [6, 7], together with its expression inadipocytes and its induction by feeding (fasting/refeeding)and by insulin administration, suggests that resistin may beinvolved in sensing the nutritional status of the animals tomodulate adipogenesis [7].

In conclusion, our results show that resistin mRNA is ex-pressed both in BAT and in different WAT depots in consid-erable amounts, and that there is a site-specific expression bythe different depots, particularly at 2 months of age, when

the epididymal and the retroperitoneal depots over-expressresistin. This transient increased resistin expression in theepididymal and the retroperitoneal WAT at a period of timein which there is a change in diet (from milk to chow) sug-gests that resistin gene expression, in these depots, could besensitive to nutritional changes.

Acknowledgements

This work was supported by DGI (BFI2000-0988-C06-01,BFI2000-0988-C06-02 and BFI2000-0988-C06-03) and FIS(FIS01/1379) of the Spanish Government, and by the DGResearch (COST Action 918) of the European Commission.

References

1. Steppan CM, Bailey ST, Bhat S, Brown EJ, Banerjee RR, Wright CM,Patel HR, Ahima R, Lazar MA: The hormone resistin links obesity todiabetes. Nature 409: 307–312, 2001

2. Morash B, Wilkinson D, Ur E, Wilkinson M: Resistin expression andregulation in mouse pituitary. FEBS Lett 526: 26–30, 2002

3. Way JM, Gorgun CZ, Tong Q, Uysal KT, Brown KK, Harrington WW,Oliver WR Jr, Wilson RM, Kliewer SA, Hotamisligil GS: Adipose tis-sue resistin expression is severely suppressed in obesity and stimulatedby peroxisome proliferator-activated receptor gamma agonists. J BiolChem 276: 25651–25653, 2001

4. Juan CC, Au LC, Fang VS, Kang SF, Ko YH, Kuo SF, Hsu YP, KwokCF, Ho LT: Suppressed gene expression of adipocyte resistin in an insu-lin-resistant rat model probably by elevated free fatty acids. BiochemBiophys Res Commun 289: 1328–1333, 2001

5. Le Lay S, Boucher J, Rey A, Castan-Laurell I, Krief S, Ferre P, ValetP, Dugail I: Decreased resistin expression in mice with different sen-sitivities to a high-fat diet. Biochem Biophys Res Commun 289: 564–567, 2001

6. Haugen F, Jorgensen A, Drevon CA, Trayhurn P: Inhibition by insulinof resistin gene expression in 3T3-L1 adipocytes. FEBS Lett 507: 105–108, 2001

7. Kim KH, Lee K, Moon YS, Sul HS: A cysteine-rich adipose tissue-specific secretory factor inhibits adipocyte differentiation. J Biol Chem276: 11252–11256, 2001

8. Oliver P, Picó C, Palou A: Ontogenesis of leptin expression in differ-ent adipose tissue depots in the rat. Pflügers Arch 442: 383–390, 2001

9. Oliver P, Picó, Palou A: Differential expression of genes for uncou-pling proteins 1, 2 and 3 in brown and white adipose tissue depotsduring rat development. Cell Mol Life Sci 58: 470–476, 2001

10. Oliver P, Picó C, Martinez N, Bonet ML, Palou A: In vivo effects ofCGP-12177 on the expression of leptin and uncoupling protein genesin mouse brown and white adipose tissues. Int J Obes Relat MetabDisord 24: 423–428, 2000

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10

Age (months)

Res

isti

n (

ng

/mL

)

Fig. 3. Serum resistin concentration in rats at different ages from 18 daysto 12 months. Resistin levels were measured by EIA. Results representmeans ± S.E.M. (n = 5).