Nr4a1 siRNA Expression Attenuates α-MSH Regulated Gene Expression in 3T3-L1 Adipocytes

Selective attenuation of α-MSH regulated gene expression upon siRNA ablation of Nr4a1 expression demonstrates a novel mechanism of cross-talk between α-MSH and Nr4a signaling in adipose.

Abstract

Several recent investigations have underscored the growing role of melanocortin signaling in the peripheral regulation of lipid, glucose, and energy homeostasis. In addition, the melanocortins play a critical role in the central control of satiety. These observations, and the latest reports highlighting the emerging role of the nuclear hormone receptor (NR) 4A subgroup in metabolism, have prompted us to investigate the cross talk between [Nle⁴, d-Phe⁷] (NDP)-α-MSH and Nr4a signaling in adipose. We have shown that NDP-MSH strikingly and preferentially induces the expression of the NR4A subgroup (but not any other members of the NR superfamily) in differentiated 3T3-L1 adipocytes. Utilization of quantitative PCR on custom-designed metabolic TaqMan low-density arrays identified the concomitant and marked induction of the mRNAs encoding Il-6, Cox2, Pdk4, and Pck-1 after NDP-MSH treatment. Similar experiments demonstrated that the mRNA expression profile induced by cAMP and NDP-MSH treatment displayed unique but also overlapping properties and suggested that melanocortin-mediated induction of gene expression involves cAMP-dependent and -independent signaling. Nr4a1/Nur77 small interfering RNA (siRNA) expression suppressed NDP-MSH-mediated induction of Nr4a1/Nur77 and Nr4a3/Nor-1 (but not Nr4a2/Nurr1). Moreover, expression of the siRNA-attenuated NDP-MSH mediated induction of the mRNAs encoding Il-6, Cox2/Ptgs2, and Pck-1 expression. In addition, Nur77 siRNA expression attenuated NDP-MSH-mediated glucose uptake. In vivo, ip administration of NDP-MSH to C57 BL/6J (male) mice significantly induced the expression of the mRNA encoding Nur77 and increased IL-6, Cox2, Pck1, and Pdk4 mRNA expression in (inguinal) adipose tissue. We conclude that Nur77 expression is necessary for MSH-mediated induction of gene expression in differentiated adipocytes. Furthermore, this study demonstrates cross talk between MSH and Nr4a signaling in adipocytes.

The Nr4a subfamily of orphan nuclear receptors consists of three well-characterized members that include Nur77 (Nr4a1), Nurr1 (Nr4a2), and Nor-1 (Nr4a3) (1). Notably, the Nr4a's are identified as early-response genes and are rapidly expressed in response to a number of physiological and pathophysiological stimuli including stress, fatty acids, neurotransmitters, growth factors, and cytokines (2). Moreover, the Nr4as are implicated in a number of biological processes including DNA repair (3), atherosclerotic vascular disease (4), cell apoptosis and carcinogenesis (5), and metabolism (1, 2, 6–9).

Several groups have demonstrated that β-adrenergic stimulation and/or cAMP signaling markedly and transiently activates the Nr4a subgroup in skeletal muscle, primary hepatocytes and liver, and 3T3-L1 adipocyte cells. Moreover, this was concomitant with the modulation of several genes implicated in lipid, glucose, and energy expenditure (6, 9–13). More recently it was identified that the anorectic neuropeptide α-MSH could also induce the Nr4a family in B16 mouse melanoma cells and primary human melanocytes (14). This is particularly interesting, given that the melanocortins also play a major role in the regulation of feeding behavior and energy expenditure. Furthermore, these melanocortins exert lipogenic effects on peripheral tissue via melanocortin receptor subtypes (15, 16).

To date α-MSH has been mostly studied in skin pigmentation and the regulation of energy homeostasis (17). For example, in the context of metabolism, peripheral administration of α-MSH decreases weight gain in diet-induced obese mice (18) and Pomc-deficient mice (19). In 3T3-L1 adipocyte cells, melanocortin receptors, MC2R and MC5R, are the most abundantly expressed melanocortin receptors (20). ACTH can induce the activity of MC1–5R; however, α-MSH activates all the melanocortin receptors, with the exception of MC2R. Moreover, studies have shown that α-MSH treatment results in lipolysis and the release of free fatty acids in 3T3-L1 adipocyte and isolated mouse adipocytes (20, 21).

Recent reports have underscored the metabolic role of the Nr4as in a number of tissue- and cell-based systems and the induction of this subgroup of nuclear receptors by the anorectic neuropeptide α-MSH. We were interested in examining the functional role of this subgroup in the in vitro adipogenic 3T3-L1 cell line in the context of α-MSH signaling in adipose.

In this study, we have identified that the mRNAs encoding the Nr4a subgroup are strikingly and transiently induced by [Nle⁴, d-Phe⁷] (NDP)-α-MSH, a synthetic α-MSH analog (22, 23) in differentiated 3T3-L1 adipocytes concomitant with the induction of several genes (with important roles in inflammatory signaling and metabolism) including, Il6, Cox2/Ptgs2, Pck-1, and Pdk4. Furthermore, Nur77 small interfering RNA (siRNA) expression attenuated the NDP-MSH-mediated induction of the Nr4a subgroup and Il6, Cox2/Ptgs2, and Pck-1, suggesting a role for cross talk between α-MSH and Nr4a signaling in the modulation of glucose homeostasis in differentiated 3T3-L1 adipocytes.

Results

NDP-MSH strikingly and selectively activates the expression of the Nr4a subgroup

To ascertain whether regulatory cross talk between NDP-MSH and nuclear hormone receptor (NR) signaling occurs in differentiated adipocytes, we used a custom-designed ABI microfluidic TaqMan low-density array (TLDA). This TLDA microfluidic card encoded TaqMan primer sets targeting the entire mouse gene superfamily of NRs that are implicated in the regulation of systems that control endocrine physiology, metabolism, development, and reproduction. These microfluidic cards also incorporate the following five normalization controls: 18S rRNA, Gapdh, Gusb, Hprt1, and 36b4 that span the cycle threshold (Ct) range of the target genes. Three of these controls, 18S rRNA, Gapdh, and 36b4, have been validated as robust normalization controls by the Nuclear Receptor Signaling Atlas (NURSA) consortium for large-scale quantitative real-time PCR (qPCR) mediated NR profiling in several cell culture models, and mouse tissues (24, 25).

Intriguingly, qPCR on the TLDA platform revealed the significant differential expression of several NRs (see Fig. 1A and Table 1). Significance was assigned by the application of the empirical Bayes statistic. Moreover, the assignment of significance was very conservatively refined by the filtering of data using Benjamini-Hochberg analysis to control the P value false discovery rate (FDR), and the application of the Bayes B statistic [which considers and ranks a proportion of differentially expressed genes (P < 0.01)]. FDR, coupled to ranking with the B statistic, resulted in the elimination of the target genes with negative B scores (which suggested < 50:50 chance of differential expression). NDP-MSH treatment (2 h) of 3T3-L1 adipocytes [relative to vehicle (PBS) treatment] resulted in the very significant (adjusted P < 0.001, positive B-statistic scores 1.7–8.2), striking, and selective induction (∼30- to 100-fold) of the three members of the Nr4a subgroup when normalized against the StatMiner-Genorm selected most stable controls, with the least expression variation (18S rRNA, Gapdh, and Gusb). To a markedly lesser extent, only one other NR, Vdr (3.9-fold), was significantly induced. Surprisingly, NDP-MSH did not significantly induce the other members of the NR family (Fig. 1A and Table 1). In summary, the TLDA analysis of the NR superfamily revealed the distinctly selective (and striking) cross talk between MSH and Nr4a signaling in differentiated adipocytes.

Fig. 1.

NDP-MSH selectively activates the expression of Nr4a nuclear receptor subgroup. A, Graph showing relative quantification (log10), i.e. calculated fold differences of nuclear receptor genes after NDP-MSH-treated adipocytes (relative to vehicle, PBS/calibrator). B, Graph showing relative quantification (log10), i.e. fold differences of nuclear receptor genes after 8-Br cAMP treatment (relative to vehicle, PBS/calibrator). Data are derived from Tables 1 and 2, respectively, expressed as fold changes normalized to the mean/median of the most stable Genorm selected controls (18S rRNA, Gapdh, and GusB) with the least expression variation. Differentially expressed genes were identified by linear models (contained in the LIMMA package for Bioconductor R embedded in StatMiner). The relative quantification log 10, i.e. the calculated fold differences [between the target (NDP-MSH and 8-Br cAMP treatment) and the calibrator/reference (vehicle treatment) sample] are displayed in Tables 1 and Table 2, respectively. Data are presented as valid when the Ct values of the gene in the calibrator/reference and target samples (i.e. before and after treatment, respectively) are less than 35 cycles. Moreover, fold differences are flagged as target not detected, calibrator not detected, and no detection when the Ct value of the gene(s) in the target (NDP-MSH or cAMP treated), calibrator (vehicle), and/or both samples are 35 cycles or greater (than the threshold limit). Hence, the reported quantitative “fold change of a gene that is not expressed in some of the biological conditions may not be reliable” (47); however, it does reflect a qualitative difference. The Applied Biosystems SDS software utilized with the ABI7900 instrument assigns Ct values beyond the arbitrarily set threshold (Ct 35) up to a maximum Ct 40, which are utilized to calculate expression (in the calibrator and target not detected genes). Significance is assigned after the application of the empirical Bayes statistic, and Benjamini-Hochberg false detection rate analysis, adjusted P values, and the B scores (>0) are considered for the final assignments (see Materials and Methods). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Table 1.

Analysis of mouse NR gene expression in 3T3-L1 adipocytes treated with NDP-MSH

Detector	B value	t value	P value	Adjusted P value (FDR)	Log10 RQ	Linear RQ	Ct status
Ar-Mm00442688_m1	−7.22399868158938	0.73208	0.49669	0.73078	−0.064420419072092	0.86214	Valid
Esr1-Mm00433149_m1	−7.24969871166557	0.69895	0.51547	0.73078	−0.254771719663615	0.55620	No detection
Esr2-Mm00599819_m1	−7.48900909552998	0.25705	0.80729	0.88068	−0.142688217944727	0.71997	Target not detected
Esrra-Mm00433143_m1	−7.42996	−0.408922825843263	0.69937	0.83127	0.04114	1.09936	Valid
Esrrb-Mm00442411_m1	−7.1667471587223	0.80187	0.45870	0.73078	−0.382809811152694	0.41418	No detection
Esrrg-Mm00516267_m1	−6.80068	−1.17387867497104	0.29284	0.56322	0.97273	9.39136	Calibrator not detected
Hnf4a-Mm00433964_m1	−7.09361679187359	−0.884738688556906	0.41645	0.73078	0.17359	1.49140	Valid
Hnf4g-Mm00443563_m1	−4.94556729942662	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
Dax-Mm00431729_m1	−4.94557	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
Shp-Mm00442278_m1	−7.22980712143738	0.72470	0.50083	0.73078	−0.280660299290717	0.52401	No detection
Rev-erba-Mm00520708_m1	−4.03480	−3.32846032259584	0.02052	0.16418	0.16818	1.47291	Valid
Rev-erb-Mm00441730_m1	−7.38390470987611	−0.497475417556279	0.63977	0.80813	0.04144	1.10012	Valid
Lxrb-Mm00437262_m1	−7.44585	0.37389	0.72368	0.83127	−0.0349194794970221	0.92274	Valid
Lxra-Mm00443454_m1	−7.44176015428787	−0.383196999385467	0.71718	0.83127	0.02669	1.06339	Valid
Fxra-Mm00436419_m1	−7.19653	0.76620	0.47784	0.73078	−0.307451968904812	0.49266	No detection
Fxrb-Mm01308716_m1	−4.94556729942662	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
Pxr-Mm00803092_m1	−6.20661469567789	−1.66706424293002	0.15583	0.41926	0.62303	4.19790	No detection
Car-Mm00437986_m1	−6.15168390378067	1.70975	0.14746	0.41926	−0.566337765175836	0.27143	No detection
Tr2-Mm00449123_m1	−7.52801490090376	0.01864	0.98584	0.99802	−0.00150514997832014	0.99654	Valid
Tlx-Mm00455855_m1	−4.94556729942662	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
Pnr-Mm00443299_m1	−4.94556729942662	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
COUP-TF1-Mm00657937_m1	−2.27214	−4.90398041511805	0.00435	0.04173	0.17520	1.49692	Valid
COUP-TF2-Mm00772789_m1	−7.33692467922958	0.57502	0.58996	0.76536	−0.0309057462215019	0.93131	Valid
Ear2-Mm00438762_m1	−7.51584	0.14413	0.89098	0.95038	−0.00872986987425464	0.98010	Valid
Gr-Mm00433832_m1	−6.71352460251767	−1.25164027630488	0.26558	0.56322	0.04154	1.10038	Valid
Mr-Mm01241595_m1	−6.68485477267374	−1.27665610100014	0.25731	0.56322	0.16296	1.45532	Valid
Nur77-Mm00439358_m1	8.21414	−43.5475587664681	1.06825E-07	0.00001	1.50224	31.78630	Valid
Nurr1-Mm00443056_m1	1.72159	−10.4452421876562	0.00013	0.00192	1.88806	77.27876	Valid
Nor-1-Mm00450074_m1	6.90613	−29.8588159527	7.12918E-07	0.00002	2.00948	102.20581	Valid
Sf-1-Mm00446826_m1	−7.52050115581881	−0.113763265382575	0.91381	0.95354	0.08208	1.20804	Valid
Lrh-1-Mm00446088_m1	−4.94556729942662	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
Gcnf-Mm00599848_m1	−7.18346812239051	−0.78199782533274	0.46930	0.73078	0.09412	1.24200	Valid
Pgr-Mm00435625_m1	−6.3847189015981	−1.52628688212191	0.18693	0.46694	0.77124	5.90526	No detection
Ppara-Mm00440939_m1	−7.29437689329319	0.63803	0.55127	0.73503	−0.122418864903352	0.75436	Valid
Ppard-Mm00803186_g1	−6.75885243212593	−1.2115411614176	0.27934	0.56322	0.04475	1.10854	Valid
Pparg-Mm00440945_m1	−5.18066	−2.44140307841398	0.05809	0.19916	0.07817	1.19720	Valid
Rara-Mm00436264_m1	−6.21541383789805	1.66020	0.15722	0.41926	−0.0782677988726348	0.83509	Valid
Rarb-Mm01319676_m1	−6.80219	1.17251	0.29334	0.56322	−0.518373652533376	0.30313	Valid
Rarg-Mm00441083_m1	−7.27150349322754	−0.669801501139277	0.53240	0.73078	0.09111	1.23342	Valid
Rora-Mm00443103_m1	−5.85406	−1.93695242609036	0.10996	0.35188	0.09141	1.23428	Valid
Rorb-Mm00524993_m1	−7.46572615251607	0.32514	0.75812	0.84628	−0.145999547897031	0.71450	Calibrator not detected
Rorc-Mm00441139_m1	−7.27208	−0.6690192569434	0.53286	0.73078	0.02639	1.06265	Valid
Rxra-Mm00441182_m1	−7.52821835574711	0.00261	0.99802	0.99802	−0.00010034333188812	0.99977	Valid
Rxrb-Mm00441193_m1	−6.42324698260015	−1.49524809558933	0.19456	0.46694	0.05780	1.14235	Valid
Rxrg-Mm00436410_m1	−4.94556729942662	−2.61847886950877	0.04676	0.17264	0.15112	1.41618	No detection
Thra-Mm00617505_m1	−7.24746389971597	0.70188	0.51379	0.73078	−0.0276947596010864	0.93822	Valid
Thrb-Mm00437044_m1	−7.44812283948188	0.36863	0.72736	0.83127	−0.0499709792802206	0.89131	Valid
Vdr-Mm00437297_m1	1.50169	−10.0309216281866	0.00016	0.00192	0.59734	3.95680	Valid

Table displays the P values, and adjusted P values of target genes before and after the application of FDR, respectively. Please note that genes with adjusted P values less than 0.05, after FDR, are differentially expressed in a significant manner. However, some of these genes have a negative B score because Bayes analysis stringently considers a proportion of differentially expressed genes with P < 0.01, and providesinformation about their ranking (Integromics Technical Support, Joaquin Panadero, personal communication). Analysis also includes t, the empirical Bayes moderated t statistic (a variant t test), an empirically moderated estimate of standard error. The relative quantification (RQ) linear and log10 denotes the calculated fold differences [between the target (NDP-MSH treatment) and the calibrator/ reference (vehicle treatment) in a linear and log scale, respectively], normalized against the mean of the Genorm selected controls, 18S rRNA, Gapdh, and GusB. Bold indicates genes differentially expressed in a significant manner.

α-MSH and the synthetic analog, NDP-MSH, induces the activity of all the melanocortin receptors, with the exception of MC2R (20, 23). We observed NDP-MSH mediated induction occurred in the absence of any significant changes in Mc1–5r expression. In addition, and as previously described (26), Mc2-r and Mc5-r were the most abundantly expressed melanocortin receptors in 3T3-L1 adipocytes (data not shown). MSH signaling involves the induction of the cAMP signaling cascade; however, it also involves several other pathways including calcium, p38 MAPK, Ras-Raf-MAPK kinase-ERK, the human homolog of agouti, the agouti signaling protein, c-Jun N-terminal kinase, and AMP-activated protein kinase (AMPK) (20, 27). Therefore, we compared and contrasted the effects of NDP-MSH and cAMP treatment on NR expression and induction in 3T3-L1 adipocytes. There were clear differences in the pattern of differential expression of nuclear receptors (after NDP-MSH and cAMP treatment; see Fig. 1A vs. Fig. 1B, Table 1, and Table 2). However, after application of stringent empirical Bayes statistic and FDR analysis to assign and rank significance, only NR4A1/Nur77 and NR4A3/Nor-1 (not Nr4A2/Nurr1) survived the statistical filtering. Although similar to the effects of MSH, the induction of Nur77 and Nor-1 mRNA after cAMP treatment was less striking. Similarly, Vdr was also significantly induced by cAMP treatment.

Table 2.

Analysis of mouse NR gene expression in 3T3-L1 adipocytes treated with 8-Br cAMP

Detector	B value	t value	P value	Adjusted P value (FDR)	Linear RQ	Log10 RQ	Ct status
Ar-Mm00442688_m1	−5.50815910969828	−2.06127430689319	0.08006	0.23451	1.21734	0.08541	Valid
Esr1-Mm00433149_m1	−7.16414789581768	0.63706	0.54525	0.69985	0.33602	−0.473629813834067	No detection
Esr2-Mm00599819_m1	−6.89807178228891	0.95731	0.37170	0.51395	0.35595	−0.448607232279201	No detection
Esrra-Mm00433143_m1	−3.36143339759365	−3.53756751220189	0.01023	0.05455	1.55960	0.19301	Valid
Esrrb-Mm00442411_m1	−6.90430427917529	−0.95082824889651	0.37475	0.51395	2.55661	0.40766	No detection
Esrrg-Mm00516267_m1	−6.60152603856742	1.23657	0.25787	0.44207	0.01479	−1.83010138686091	No detection
Hnf4a-Mm00433964_m1	−6.68937839287154	−1.15871326825719	0.28624	0.45798	8.12679	0.90992	No detection
Hnf4g-Mm00443563_m1	−7.31435735063327	−0.364596090821052	0.72666	0.85072	4.77107	0.67862	No detection
Dax-Mm00431729_m1	−7.35133976805405	−0.259555666630404	0.80301	0.86184	2.42171	0.38412	No detection
Shp-Mm00442278_m1	−2.4278730568011	−4.2264890643207	0.00432	0.02963	1.51736	0.18109	No detection
Rev-erba-Mm00520708_m1	−3.82501365643782	−3.21134620267899	0.01577	0.06880	2.04760	0.31124	Valid
Rev-erb-Mm00441730_m1	−6.13037922426972	−1.61433639488875	0.15245	0.31296	1.17608	0.07044	Valid
Lxrb-Mm00437262 m1	−4.16445516930638	−2.97721453714941	0.02171	0.08683	1.61585	0.20840	Valid
Lxra-Mm00443454_m1	−4.45056120265712	−2.7820187892694	0.02849	0.10519	1.89310	0.27717	Valid
Fxra-Mm00436419_m1	−7.32856213923481	−0.3281362355285	0.75283	0.85294	4.34285	0.63777	No detection
Fxrb-Mm01308716_m1	−7.38903203164852	0.03464	0.97338	0.97338	0.98431	−0.0068687547461857	No detection
Pxr-Mm00803092_m1	−6.65977958751272	−1.18531466761614	0.27627	0.45727	1.94267	0.28840	No detection
Car-Mm00437986_m1	−6.04637542559818	−1.677049098388	0.13940	0.30706	2.80196	0.44746	No detection
Tr2-Mm00449123_m1	−7.36808364421238	−0.194664866121818	0.85143	0.87759	1.02214	0.00951	valid
Tlx-Mm00455855_m1	−7.20639983844706	−0.572702160560403	0.58556	0.70267	9.75238	0.98911	No detection
Pnr-Mm00443299_m1	−6.02373694581711	1.69378	0.13610	0.30706	0.00196	−2.70690643149241	No detection
COUP-TF1-Mm00657937_m1	−7.35328741921095	−0.252851713322791	0.80797	0.86184	1.03747	0.01597	Valid
COUP-TF2-Mm00772789_m1	−2.74562517816694	3.98642	0.00579	0.03472	0.61922	−0.208154019219964	Valid
Ear2-Mm00438762_m1	−6.82584930950438	−1.03019381520536	0.33870	0.50583	9.25667	0.96645	Valid
Gr-Mm00433832_m1	−7.12140114605729	0.69677	0.50941	0.67922	0.87390	−0.0585394236119403	Valid
Mr-Mm01241595_m1	−3.55486444776139	−3.40040019457488	0.01225	0.05878	1.40592	0.14796	Valid
Nur77-Mm00439358_m1	2.42773	−9.12247662370858	0.00005	0.00082	20.36119	1.30880	Valid
Nurr1-Mm00443056_m1	−1.44142027891247	−5.01700686577354	0.00175	0.01683	13.43164	1.12813	Valid
Nor-1-Mm00450074_m1	5.28930	−13.9160642456297	3.45733E-06	0.00008	48.30491	1.68399	Valid
Sf-1-Mm00446826_m1	−6.24872295010379	1.52411	0.17324	0.33185	0.24698	−0.607346168148251	Target not detected
Lrh-1-Mm00446088_m1	−5.45657647907438	−2.09701999724985	0.07602	0.23451	41.29905	1.61594	No detection
Gcnf-Mm00599848_m1	−6.15468486948772	−1.59599612353876	0.15648	0.31296	1.23803	0.09273	Valid
Pgr-Mm00435625_m1	−6.74926847107076	1.10359	0.30786	0.47669	0.00702	−2.1539605492309	Target not detected
Ppara-Mm00440939_m1	−7.17387235644548	−0.622774614560883	0.55405	0.69985	2.90760	0.46353	Calibrator not detected
Ppard-Mm00803186_g1	−2.09474976017492	−4.48532268747428	0.00318	0.02547	1.29866	0.11350	Valid
Pparg-Mm00440945_m1	−6.84659191386124	−1.00966121064585	0.34776	0.50583	1.10077	0.04169	Valid
Rara-Mm00436264_m1	−6.5853537787917	1.25057	0.25304	0.44207	0.91555	−0.0383181760003797	Valid
Rarb-Mm01319676_m1	−6.04522938763052	−1.67789788368001	0.13923	0.30706	1.55085	0.19057	Valid
Rarg-Mm00441083_m1	−5.2501355189843	2.23897	0.06189	0.21219	0.62312	−0.205428131010783	Valid
Rora-Mm00443103_m1	−0.496484443125155	−5.8521001017043	0.00074	0.00890	1.74168	0.24097	Valid
Rorb-Mm00524993_m1	−7.33417192099774	0.31261	0.76409	0.85294	0.31798	−0.497600401827364	No detection
Rorc-Mm00441139_m1	−7.19257362686531	0.59445	0.57175	0.70267	0.76215	−0.117957620853958	Valid
Rxra-Mm00441182_m1	−6.28245060225431	−1.49793456548665	0.17975	0.33185	1.10146	0.04197	Valid
Rxrb-Mm00441193_m1	−6.05537947610065	−1.67037438314544	0.14074	0.30706	1.20127	0.07964	Valid
Rxrg-Mm00436410_m1	−5.60115826533221	−1.99647435193074	0.08794	0.23451	46.89210	1.67110	No detection
Thra-Mm00617505_m1	−7.37034034396781	−0.184205733984194	0.85931	0.87759	1.01527	0.00658	Valid
Thrb-Mm00437044_m1	−5.5595194952694	2.02555	0.08431	0.23451	0.14235	−0.846640682768079	Target not detected
Vdr-Mm00437297_m1	5.71306	−14.8309173674091	2.28468E-06	0.00008	2.99596	0.47654	Valid

Table displays the P values, and adjusted P values of target genes before and after the application of FDR. Statistical analysis as detailed in Table 1. Bold indicates genes differentially expressed in a significant manner.

The Nr4a genes are rapidly and transiently induced after 30–60 min NDP-MSH treatment

To rigorously assess the temporal response of mRNAs encoding the Nr4a subgroup to NDP-MSH, 3T3-L1 adipocyte cells were differentiated until they acquired the adipocyte phenotype (assessed by oil-red O staining; data not shown) followed by treatment with NDP-MSH over a 15-min to 8-h time period. Expression of the mRNAs was monitored by qPCR. As observed above on NR TLDA analysis, the strongest NDP-MSH-mediated induction was the mRNA encoding Nor-1 (∼200-fold at 2 h after NDP-MSH treatment, normalized against 36b4). The maximal induction of Nur77 and Nurr1 (∼10- to 30-fold, respectively) was also observed at 2 h after NDP-MSH treatment. Induction of mRNA expression was observed within 15 min for Nur77 (Nr4a1) mRNA and within 60 min for Nurr1 and Nor-1 (Nr4a2 and Nr4a3, respectively). Moreover, all the mRNA transcripts encoding the Nr4a1–3 genes rapidly returned to basal levels within 4 and 8 h after NDP-MSH treatment. In summary, NDP-MSH treatment, strikingly, rapidly (15–60 min), and transiently induced the expression of the mRNAs encoding all three Nr4as (Fig. 2, A–C).

Fig. 2.

The Nr4a genes are rapidly and transiently induced after 30–60 min NDP-MSH treatment. Total RNA from 3T3-L1 adipocytes treated with 10 nm NDP-MSH from 15 min to 8 h were analyzed by qPCR for the expression of Nr4a1/Nur77 (A), Nr4a2/Nurr1 (B), and Nr4a3/Nor-1 (C). Results were normalized against 36b4 at each time point and expressed as mean ± sem (n = 3). Statistical significance was calculated using the Student's unpaired t test, where *, P < 0.05; **, P < 0.01; ***, P < 0.001.

The NUR77 protein is induced by NDP-MSH in differentiated 3T3-L1 adipocytes

To assess whether the induction of the mRNAs (encoding the Nr4a genes) translated to changes in protein expression, Western blot analysis was performed. Nuclear protein extracts were obtained from preadipocyte 3T3-L1 cells along with differentiated 3T3-L1 adipocytes after 0, 2, and 4 h treatments with vehicle (PBS/BSA) or NDP-MSH.

We identified an immunoreactive response for NUR77 expression relative to the nuclear loading control, TATA binding protein (TBP) in untreated cells 2 and 4 h after NDP-MSH in differentiated 3T3-L1 cells but not in preadipocytes (Fig. 3). We observed no change in immunoreactivity for NURR1 expression upon NDP-MSH treatment (data not shown). Furthermore, the availability of antibodies for NOR-1 at this time was not suitable for robust and reproducible Western analysis.

Fig. 3.

The NUR77 protein is induced by NDP-MSH in differentiated 3T3-L1 adipocytes. Western blot analysis of NUR77 and TBP in preadipocytes and adipocytes treated with vehicle (PBS) and 10 nm NDP-MSH for 2 and 4 h. Nuclear protein was extracted from the cells as described in Materials and Methods. Thirty micrograms of nuclear protein from cells were resolved.

The induction of the Nr4a subgroup by NDP-MSH is concomitant with the significant differential expression of genes involved in metabolism

In the context of α-MSH in the peripheral regulation of lipid, glucose, and energy homeostasis, we examined the effect of NDP-MSH treatment on metabolic gene expression. Vehicle and NDP-MSH (2 h)-treated 3T3-L1 differentiated adipocytes were analyzed on custom-designed TLDAs targeting 93 critical genes that are involved in (and control) lipid, glucose, and energy metabolism and incorporated several normalization controls, including 18S rRNA, 36b4, Gapdh, GusB, Hprt, and Arbp. FDR, coupled with the ranking with the B statistic, identified five significant changes in gene expression after conservative filtering of data for several genes implicated in metabolism normalized against the Genorm controls (18S rRNA, 36b4, and Gapdh). These included genes implicated in metabolic processes such as inflammation and lipid metabolism; IL-6 (Il-6), glucose metabolism; phosphoenolpyruvate carboxykinase 1 (Pck-1), pyruvate dehydrogenase kinase 4 (Pdk-4), prostaglandin synthase 2/cyclooxygenase 2 (Ptgs2/Cox2), and CCAAT/enhancer-binding protein-β (C/ebp-β) (Fig. 4A and Table 3). Table 3 lists many more genes with adjusted P < 0.05, after FDR, and differentially expressed in a significant manner (involved in lipid and glucose homeostasis). However, these genes had a negative B score because Bayes ranks the differentially expressed genes with P < 0.01 [Integromics (Granada, Spain) technical support, Panadero, J., personal communication].

Fig. 4.

Comparison of effects on the expression of genes involved in metabolic processes between NDP-MSH- and cAMP-treated adipocytes. 3T3-L1 adipocytes were treated with either 10 nm NDP-MSH or 1 mm 8-Br cAMP for 2 h. A, Graph showing relative quantification of metabolic genes in NDP-MSH-treated adipocytes (relative to vehicle, PBS). B, Graph showing relative quantification of metabolic genes of 8-Br cAMP-treated adipocytes (relative to vehicle, PBS). Data are derived from Tables 3 and 4, respectively, expressed as fold changes normalized to the mean of Genorm-selected controls (18S rRNA, Gapdh, and GusB). Statistical analysis was performed as detailed in Fig. 1 and in Materials and Methods.

Table 3.

Analysis of mouse metabolic gene expression in 3T3-L1 adipocytes treated with NDP-MSH

Detector	B value	t value	P value	Adjusted P value (FDR)	Linear RQ	Log10 RQ	Ct status
Abca1-Mm00442646_m1	−7.27570074547357	1.05356	0.33923	0.44865	0.83980	−0.0758261111300267	Valid
Abcg1-Mm00437390_m1	−6.82380564029731	1.45031	0.20532	0.29013	0.86640	−0.0622797613251481	Valid
Acadm-Mm00431611_m1	−5.42425918743045	2.51099	0.05262	0.11043	0.74318	−0.128907733698776	Valid
Acsl4-Mm00490331_m1	−7.86711961853159	−0.116300930841545	0.91184	0.92608	1.02701	0.01157	Valid
Akt2-Mm00545827_m1	−5.37606210054846	2.54696	0.05033	0.11043	0.60982	−0.214801625794898	Valid
Apoe-Mm00437573_m1	−5.42524950676269	2.51025	0.05267	0.11043	0.80969	−0.0916803575683298	Valid
Cav1-Mm00483057_m1	−6.49018146075984	1.71385	0.14583	0.23119	0.77509	−0.110645247295161	Valid
Cav3-Mm01182632_m1	−7.84243423426412	0.23488	0.82341	0.87740	0.32846	−0.483521068590945	Valid
Cd36-Mm00432403_m1	−5.59583051046963	2.38334	0.06170	0.12533	0.82062	−0.0858604443188263	Valid
Cebpa-Mm00514283_s1	−7.7722294963989	−0.418873446254798	0.69230	0.77585	1.09387	0.03897	Valid
Cebpb-Mm00843434_s1	0.65933	−8.94204065039986	0.00026	0.00335	2.09523	0.32123	Valid
Cebpd-Mm00786711_s1	−7.86068754697112	0.15603	0.88198	0.92465	0.94920	−0.0226441452293905	Valid
Cox5a-Mm00432638_m1	−4.39610301250635	3.29953	0.02073	0.07093	0.78247	−0.106531170687753	Valid
Cpt1b-Mm00487200_m1	−7.38764155524402	0.94119	0.38885	0.47689	0.79065	−0.102015720752793	Valid
Cpt2-Mm00487202_m1	−3.36423414156762	4.16470	0.00834	0.03189	0.67695	−0.169446439781525	Valid
Cs-Mm00466043_m1	−6.005817498354	2.07909	0.09084	0.16870	0.71472	−0.145865756787847	Valid
Fabp4-Mm00445880_m1	−7.37245852747716	0.95695	0.38155	0.47689	0.91729	−0.0374949583488132	Valid
Fasn-Mm00662319_m1	−2.20027191613071	5.27582	0.00303	0.01791	0.55214	−0.257949258506736	Valid
Foxo1-Mm00490672_m1	−7.31008243947882	1.01993	0.35349	0.45053	0.83921	−0.076127141125691	Valid
Gys1-Mm00472712_m1	−6.22698294888562	1.91381	0.11245	0.19235	0.67382	−0.171453306419285	Valid
Hif1a-Mm00468875_m1	−7.81212279072606	−0.326746985613668	0.75679	0.83375	1.02677	0.01147	Valid
Hk2-Mm00443385_m1	−7.66526536852216	−0.603429832318206	0.57195	0.66387	1.14481	0.05873	Valid
Il6-Mm00446190_m1	0.75508	−9.095579743778	0.00024	0.00335	5.77883	0.76184	Valid
Insr-Mm00439693_m1	−5.2315109317577	2.65524	0.04407	0.11043	0.79615	−0.0990054207961538	Valid
Irs1-Mm00439720_s1	−4.91149933409588	2.89772	0.03294	0.09398	0.84936	−0.0709092878675154	Valid
Lep-Mm00434759_m1	−7.05689010206662	−1.25404886457099	0.26402	0.36513	1.38522	0.14152	Valid
Lepr-Mm00440181_m1	−7.19973427849448	1.12555	0.31032	0.42022	0.10943	−0.960854298382132	Valid
Lipe-Mm00495359_m1	−0.981165175224435	6.64671	0.00106	0.00860	0.68307	−0.165533049837893	Valid
Lpin1-Mm00550511_m1	−7.29012162119624	−1.03954237168589	0.34512	0.44865	1.08106	0.03385	Valid
Lpl-Mm00434764_m1	−3.27343373949192	4.24564	0.00770	0.03130	0.74887	−0.125596403746472	Valid
Ndufa5-Mm00471676_g1	−5.85515327216234	2.19099	0.07871	0.15503	0.78719	−0.103922244058665	Valid
Ndufb5-Mm00452592_m1	−5.33108511005886	2.58058	0.04829	0.11043	0.75268	−0.123388850444936	Valid
Npy-Mm00445771_m1	−7.47182022954594	−0.850059038036528	0.43322	0.51199	42.23726	1.62570	Valid
Npy2r-Mm01956783_s1	−6.40970549209452	−1.77551759126077	0.13458	0.21870	1.40780	0.14854	Valid
Pck1-Mm00440636_m1	3.42708	−14.5798042878768	0.00002	0.00074	5.28332	0.72291	Valid
Pdk1-Mm00554306_m1	−2.03606996345871	5.44683	0.00263	0.01709	0.59246	−0.227344542280898	Valid
Pdk2-Mm00446681_m1	−3.63381953315276	3.92935	0.01056	0.03815	0.63792	−0.19523467607674	Valid
Pdk3-Mm00455220_m1	−6.20574800517979	1.92976	0.11015	0.19235	0.72923	−0.137135886913592	Valid
Pdk4-Mm00443325_m1	2.01954	−11.3726813856132	0.00008	0.00171	2.59628	0.41435	Valid
Pfkm-Mm00445461_m1	−6.1713180653542	1.95558	0.10653	0.19235	0.75025	−0.124793657091368	Valid
Pnpla2-Mm00503040_m1	−3.12134126930238	4.38322	0.00675	0.02923	0.64772	−0.188612016172131	Valid
Ppargc1a-Mm00447183_m1	−6.27897749572296	−1.87466450180387	0.11831	0.19719	1.31748	0.11974	Valid
Ppargc1b-Mm00504720_m1	−2.88406524116243	−4.6030786876865	0.00549	0.02547	1.37947	0.13971	Valid
Ptgs2-Mm00478374_m1	0.43497	−8.59158033764606	0.00031	0.00338	5.10689	0.70816	Valid
Pygm-Mm00478582_m1	−6.56969069433489	1.65233	0.15799	0.24450	0.75199	−0.123790223772489	Valid
Rab18-Mm00441057_m1	−6.68724133206513	1.56004	0.17812	0.26313	0.80857	−0.0922824175596581	Valid
Retn-Mm00445641_m1	−2.85602538248978	4.62950	0.00535	0.02547	0.68055	−0.167138543148101	Valid
Rock1-Mm00485745_m1	−7.43850817776515	0.88694	0.41481	0.49931	0.87364	−0.0586674013771801	Valid
Rock2-Mm00485761_m1	−7.73161657107397	−0.496300232853125	0.64025	0.73011	1.13376	0.05452	Valid
Scd1-Mm00772290_m1	−4.92220154241346	2.88953	0.03326	0.09398	0.72520	−0.139544126878903	Valid
Scd2-Mm00485951_g1	−0.80611290013729	6.86458	0.00091	0.00845	0.56309	−0.249420075296256	Valid
Slc2a2-Mm00446224_m1	−6.71920501062666	1.53463	0.18409	0.26591	0.00803	−2.09543635203968	Valid
Slc2a4-Mm00436615_m1	−5.34032309258773	2.57367	0.04870	0.11043	0.60574	−0.21771158241965	Valid
Slc2a8-Mm00444634_m1	−5.0247953539966	2.81136	0.03650	0.09885	0.66240	−0.178878712978997	Valid
Socs3-Mm00545913_s1	−7.86633233490374	−0.121855621550162	0.90765	0.92608	1.00960	0.00415	Valid
Srebf1-Mm00550338_m1	−1.6250135810549	5.89266	0.00184	0.01332	0.59948	−0.22222703235461	Valid
Stat3-Mm00456961_m1	−4.85967303649837	2.93743	0.03143	0.09398	0.81138	−0.0907772675813385	Valid
Stat5b-Mm00839889_m1	−2.60569999996316	4.86962	0.00431	0.02333	0.69533	−0.157806613282517	Valid
Ucp1-Mm00494069_m1	9.14953	−45.86660894569	6.77314E-08	4.40254E-06	3025.82138	3.48084	Calibrator not detected
Ucp2-Mm00495907_g1	−4.84322557838101	2.95006	0.03096	0.09398	0.81589	−0.0883690276160257	Valid
Ucp3-Mm00494074_m1	−5.30949498813775	−2.59674255152187	0.04734	0.11043	1.57110	0.19620	Valid

Table displays the P values, and adjusted P values of target genes before and after the application of FDR. Relative quantification (RQ) linear and log10 were normalized against the mean of the Genorm selected controls 18S rRNA, Gapdh, and 36b4. Statistical analysis is as detailed in Table 1. Bold indicates genes differentially expressed in a significant manner.

Our analysis focused on the subset of genes that survived stringent statistical filtering and displayed very high odds of differential expression. We rigorously assessed the temporal response of mRNAs encoding the genes identified above after NDP-MSH treatment. 3T3-L1 adipocyte cells were treated with 10 nm NDP-MSH over a 15-min to 8-h time period, and expression of the mRNAs was monitored by qPCR. Interestingly, we observed maximal induction of Il-6, Ptgs2/Cox2, Pck-1, and Pdk4 after 1, 2, and 4 h (respectively), after NDP-MSH treatment (Fig. 5). In addition, the increase in IL-6 mRNA expression translated to a significant induction of secreted IL-6 measured by ELISA (Fig. 5E).

Fig. 5.

The NDP-MSH induction of the Nr4a subgroup is concomitant with significant differential expression genes involved in metabolism. Total RNA from 3T3-L1 adipocytes treated with 10 nM NDP-MSH from 15 min to 8 h were analyzed by qPCR for the expression of Il6 (A), Cox2/Ptgs2 (B), Pck1 (C), and Pdk4 (D). Results were normalized against 36b4 at each time point and expressed as mean ± sem (n = 3). E, IL-6 secretion to medium (pictograms per milliliter). Medium from 3T3-L1 adipocytes treated for 2 h with 10 nm NDP-MSH and/or vehicle was collected and measured. Results are mean ± sem of three independent experiments performed in quadruplet. Statistical significance was calculated using the Student's unpaired t test, where *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Subsequently we compared and contrasted the effects of NDP-MSH and cAMP treatment on metabolic gene expression and induction in 3T3-L1 adipocytes. There were distinct differences, and some overlap in the pattern of differential expression of metabolic genes (after NDP-MSH and cAMP treatment, Fig. 4B vs. Fig. 4A and Table 3 vs. Table 4). After application of stringent empirical Bayes B statistic and FDR analysis to assign and rank significance, only C/ebpβ, Pck1, and PGC-1α and -β survived the statistical filtering. Similarly, after the application of FDR analysis (without the B score filtering based on P < 0.01), additional genes are identified, and this subset of differentially expressed genes induced by 8-bromoadenosine cAMP (8-Br cAMP) (Fig. 4B and Table 4) are clearly different to the subset induced by NDP-MSH. In summary, this underscores the importance of examining both signaling compounds and that the NDP-MSH-mediated induction of gene expression involves cAMP and cAMP-independent signaling.

Table 4.

Analysis of mouse metabolic gene expression in 3T3-L1 adipocytes treated with 8-Br cAMP

Detector	B value	t value	P value	Adjusted P value (FDR)	Linear RQ	Log10 RQ	Ct status
Abca1-Mm00442646_m1	−6.47203270416448	−0.983136965483622	0.35911	0.58168	1.05340	0.02260	Valid
Abcg1-Mm00437390_m1	−6.94866431433544	−0.255521028063199	0.80585	0.89376	1.04779	0.02028	Valid
Acadm-Mm00431611_m1	−6.95730730202809	−0.223512091415063	0.82968	0.90376	1.02976	0.01274	Valid
Acsl4-Mm00490331_m1	−2.99892919267858	−3.51728759197849	0.01018	0.05644	2.41953	0.38373	Valid
Akt2-Mm00545827_m1	−5.99484250094875	−1.41007012605418	0.20244	0.41631	1.23313	0.09101	Valid
Apoe-Mm00437573_m1	−5.9960202210425	−1.40911893801009	0.20271	0.41631	1.21792	0.08562	Valid
Cav1-Mm00483057_m1	−6.4253521928257	−1.03004427920842	0.33811	0.57291	31.77646	1.50211	Valid
Cav3-Mm01182632_m1	−6.58672998372099	−0.859854864267424	0.41905	0.60862	40.62194	1.60876	No detection
Cd36-Mm00432403_m1	−6.00486578483379	−1.40196357076678	0.20474	0.41631	1.12745	0.05210	Valid
Cebpa-Mm00514283_s1	−5.30471603316879	−1.92631942576746	0.09652	0.26762	1.39442	0.14439	Valid
Cebpb-Mm00843434_s1	0.77510	−6.75103121847619	0.00030	0.00603	2.02460	0.30634	Valid
Cebpd-Mm00786711_s1	−3.65958645529927	−3.05428013704486	0.01906	0.08200	1.56833	0.19544	Valid
Cox5a-Mm00432638_m1	−5.22622058282471	−	0.08903	0.26023	1.28389	0.10853	Valid
Cpt1b-Mm00487200_m1	−6.12619850864466	1.30163	0.23527	0.46294	0.10928	−0.96145793125541	Target not detected
Cpt2-Mm00487202_m1	−6.51347921922502	−0.940037726418302	0.37928	0.58586	1.09284	0.03856	Valid
Cs-Mm00466043_m1	−6.22855102872073	1.21326	0.26536	0.50584	0.91042	−0.0407596072971188	Valid
Fabp4-Mm00445880_m1	−4.4826065167765	−2.49285759987093	0.04228	0.13936	1.35951	0.13338	Valid
Fasn-Mm00662319_m1	−6.9701887124759	−0.164829827780038	0.87385	0.93518	1.01482	0.00639	Valid
Foxo1-Mm00490672_m1	−5.68269700553692	1.65195	0.14364	0.35048	0.83745	−0.0770421500450454	Valid
Gys1-Mm00472712_m1	−5.65091641789539	−1.67564891864883	0.13883	0.35048	1.15577	0.06287	Valid
Hif1a-Mm00468875_m1	−0.529115798487438	−5.48088372472817	0.00101	0.01025	1.59144	0.20179	Valid
Hk2-Mm00443385_m1	−0.132955099353836	−5.84542007115945	0.00070	0.00849	2.17440	0.33734	Valid
Il6-Mm00446190_m1	−6.52307398754988	−0.929844051263675	0.38417	0.58586	4.11340	0.61420	Valid
Insr-Mm00439693_m1	−5.7412347069987	−1.60791890248564	0.15300	0.35896	1.16573	0.06660	Valid
Irs1-Mm00439720_s1	−5.60423389368891	−1.71021227014467	0.13208	0.35030	1.14670	0.05945	Valid
Lep-Mm00434759_m1	−6.47892150309319	0.97607	0.36236	0.58168	0.88855	−0.0513194155804653	Valid
Lepr-Mm00440181_m1	−5.23233463895439	−1.97738797918858	0.08959	0.26023	2.38960	0.37833	No detection
Lipe-Mm00495359_m1	−3.71816685794912	−3.01393427787688	0.02016	0.08200	1.31521	0.11900	Valid
Lpin1-Mm00550511_m1	−1.80312364585439	−4.41121255918593	0.00331	0.02526	1.78729	0.25220	Valid
Lpl-Mm00434764_m1	−4.50941649424105	−2.47464242692057	0.04341	0.13936	1.22113	0.08676	Valid
Ndufa5-Mm00471676_g1	−4.04833801951533	−2.78799300323631	0.02770	0.10243	1.23020	0.08997	Valid
Ndufb5-Mm00452592_m1	−6.55152761517444	−0.899090279456486	0.39922	0.59396	1.11127	0.04582	Valid
Npy-Mm00445771_m1	−6.98517697252056	−0.0273625061673953	0.97895	0.98896	1.11408	0.04691	No detection
Npy2r-Mm01956783_s1	−6.60884715447981	0.83445	0.43227	0.60866	0.03293	−1.48243364236616	Target not detected
Pck1-Mm00440636_m1	1.79158	−7.90330823645043	0.00011	0.00603	3.73391	0.57216	Valid
Pdk1-Mm00554306_m1	−6.71401786122737	−0.70357736013987	0.50499	0.66966	1.05464	0.02311	Valid
Pdk2-Mm00446681_m1	−2.68222632945434	−3.74592162059326	0.00755	0.04996	1.39402	0.14427	Valid
Pdk3-Mm00455220_m1	−6.85057901261473	−0.491669112042786	0.63837	0.81127	1.03864	0.01647	Valid
Pdk4-Mm00443325_m1	−2.76858195305639	−3.6830748059515	0.00819	0.04996	1.45149	0.16182	Valid
Pfkm-Mm00445461_m1	−6.31725965485071	−1.13319036867258	0.29538	0.52994	11.56542	1.06316	Valid
Pnpla2-Mm00503040_m1	−6.9842209070256	−0.0493042992031737	0.96209	0.98896	1.01255	0.00541	Valid
Ppargc1a-Mm00447183_m1	0.81440	−6.79266553187554	0.00029	0.00603	1.71710	0.23480	Valid
Ppargc1b-Mm00504720_m1	0.29347	−6.25776701118953	0.00047	0.00711	2.31327	0.36423	Valid
Ptgs2-Mm00478374_m1	−6.887285384296	−0.418544035436358	0.68841	0.83964	2.25203	0.35257	Valid
Pygm-Mm00478582_m1	−6.78200615886998	−0.606462579597974	0.56383	0.73178	1.12981	0.05300	Valid
Rab18-Mm00441057_m1	−6.29048302233768	−1.15774464650995	0.28588	0.52845	8.89780	0.94928	Valid
Retn-Mm00445641_m1	−3.3695485144151	−3.25555636819725	0.01446	0.06785	1.38660	0.14195	Valid
Rock1-Mm00485745_m1	−6.98548553682099	−0.0143483063672784	0.98896	0.98896	1.00123	0.00053	Valid
Rock2-Mm00485761_m1	−3.26541588310136	−3.32853000857313	0.01310	0.06658	1.32339	0.12169	Valid
Scd1-Mm00772290_m1	−6.90752779559935	−0.372484398662658	0.72083	0.83964	1.03855	0.01643	Valid
Scd2-Mm00485951_g1	−6.87819738040745	−0.437720434408795	0.67511	0.83964	1.04912	0.02082	Valid
Slc2a2-Mm00446224_m1	−6.93605143481426	−0.821666151576023	0.43903	0.60866	21.59963	1.33445	Calibrator not detected
Slc2a4-Mm00436615_m1		0.29619	0.77590	0.87647	0.96984	−0.0133017511643106	Valid
Slc2a8-Mm00444634_m1	−5.93755681705318	−1.45593067223172	0.18983	0.41631	1.11031	0.04544	Valid
Socs3-Mm00545913_s1	−6.98087310901216	0.09124	0.92992	0.97802	0.99436	−0.00245751920288478	Valid
Srebf1-Mm00550338_m1	−6.90082006732795	0.38833	0.70961	0.83964	0.95199	−0.0213659988995519	Valid
Stat3-Mm00456961_m1	−6.91250683871165	−0.360294547303217	0.72952	0.83964	1.09045	0.03761	Valid
Stat5b-Mm00839889_m1	−6.38906309223622	1.06546	0.32291	0.56279	0.89213	−0.0495705205996298	Valid
Ucp1-Mm00494069_m1	−6.71353806416891	0.70422	0.50461	0.66966	0.02429	−1.61448796705966	No detection
Ucp2-Mm00495907_g1	−1.30242634524268	−4.81470379568001	0.00208	0.01809	1.49891	0.17577	Valid
Ucp3-Mm00494074_m1	−4.0794328594065	−2.76681380479372	0.02854	0.10243	1.29796	0.11326	Valid

Table displays the P values, and adjusted P values of target genes, before and after the application of FDR. Relative quantification (RQ) linear and log10 were normalized against the mean of the Genorm selected controls 18S rRNA, Gapdh, and 36b4. Statistical analysis as detailed in Table 1. Bold indicates genes differentially expressed in a significant manner.

Nur77 siRNA expression significantly attenuates Nur77 expression

Nur77 mRNA expression has been shown to be induced during the early phase of 3T3-L1 cell differentiation (28, 29). Therefore, to identify the effect of attenuating Nur77 expression on Nr4a gene expression, we transiently transfected and expressed a siRNA (30 nm) targeting Nur77 (relative to a negative control siRNA) in 3T3-L1 cells and induced to differentiate for 2 h. Expression of the Nur77 siRNA (but not the negative control) significantly and preferentially attenuated Nur77 mRNA expression (∼5-fold; Fig. 6). Although the Nur77 siRNA did not repress Nurr1 mRNA expression, we observed an approximately 2.5 fold attenuation of Nor-1 (Fig. 6).

Fig. 6.

Nur77 siRNA expression significantly attenuates Nur77 expression. 3T3-L1 cells were transfected with 30 nm Nur77 siRNA or negative siRNA (control) for 48 h and differentiated for a further 1 h. qPCR was used to assay the expression of Nur77/Nr4a1, Nurr1/Nr4a2, and Nor-1/Nr4a3. Results were normalized against 36b4 and expressed as mean ± sem (n = 3). Statistical significance was calculated using the Student's unpaired t test, where **, P < 0.01 and ***, P < 0.001.

Stable Nur77 siRNA expression significantly attenuates the NDP-MSH-mediated induction of Nur77: Il6, Cox2, and Pck-1 induction are dependent on activation of Nur77 expression

We next examined NDP-MSH-mediated induction of the Nr4a subgroup and subsequent Il-6, Cox2, Pdk-4, and Pck-1 gene expression in stable 3T3-L1 cells expressing a previously characterized Nur77 siRNA (10, 11) cloned into the pSilencer 3.1 vector. The Nur77 siRNA and negative control-transfected 3T3-L1 cells, were isolated as polyclonal pools (to avoid clonal variation) after 8–10 d of G418 selection. NDP-MSH-mediated induction of Nur77 mRNA was completely ablated in the Nur77 siRNA but not the negative control siRNA transfected/expressing cells (Fig. 7A). In contrast, Nurr1 and Nor-1 remained significantly induced by NDP-MSH in the negative control stable cells; however, the siRNA targeting Nur77 attenuated the NDP-MSH induction of Nurr1 but not Nor-1 (Fig. 7, B and C).

Fig. 7.

Nur77 siRNA stable expression significantly attenuates the NDP-MSH-mediated induction of Nr4a subgroup and metabolic genes. 3T3-L1 Nur77 siRNA (siNur77) and negative siRNA (siNeg) stable cells were differentiated and treated with 10 nm NDP-MSH or vehicle for 2 h. qPCR was used to assay the expression of Nur77/Nr4a1 (A), Nurr1/Nr4a2 (B), Nor-1/Nr4a3 (C), Il6 (D), Cox2/Ptgs2 (E), Pck1 (F), and Pdk4 (G). Results were normalized against 36b4 and expressed as fold changes (n = 3). Statistical significance was calculated using the Student's unpaired t test, where *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Importantly, the stable expression of the Nur77 siRNA in 3T3-L1 cells inhibited the NDP-MSH mediated induction of several metabolic genes including Il-6, Pck-1, and Cox2/Ptgs2 (Fig. 7, D–F). Although NDP-MSH induced the expression of the Pdk-4 transcript in the negative control stable cells, this induction was not significant (Fig. 7G). Moreover, there was no attenuation of the Pdk-4 transcript (in the presence of NDP-MSH) in the cells expressing the Nur77 siRNA (Fig. 7G). In summary, this demonstrated that the NDP-MSH-mediated induction of these important metabolic genes is dependent on the concomitant induction and expression of Nur77.

The MSH-inducible genes identified in this study have all been linked to different aspects of glucose homeostasis and in particular increased Il-6 expression (mRNA and protein) has been demonstrated to enhance glucose transport in 3T3-L1 cells and adipose tissue (30, 31). Moreover, MSH has been demonstrated to induce glucose uptake in hypothalamic cells in the presence of insulin (32). Consequently, we transiently transfected 3T3-L1 cells with siRNA-negative (negative control) and siRNA-Nur77 and grew the cells to confluence. Differentiation medium (which includes insulin) was added with either NDP-MSH or vehicle for 2 h. To examine the effect of siRNA expression, glucose uptake was measured (during the peak of Nur77 expression). We observed that NDP-MSH significantly (P < 0.01) induced glucose uptake (∼1.4 fold). Interestingly, we observed nonsignificant (20%) increase in the basal (vehicle mediated) glucose uptake in the siRNA Nur77-transfected cells [normalized against the siRNA-negative (negative control) transfected cells]; however, transfection of the siRNA-Nur77-attenuated NDP-MSH stimulated glucose uptake (Fig. 8).

Fig. 8.

Effect of Nur77 siRNA on glucose uptake in 3T3-L1 by NDP-MSH. 3T3-L1 transfected with Nur77 siRNA (siNur77) and control (siNeg) for 48 h, followed by 2 h treatment with 10 nm NDP-MSH or vehicle in adipocyte differentiation media. Glucose uptake was measured. Data are expressed as fold changes of 2-deoxyglucose uptake ± se from four independent experiments performed in triplicate. Statistical significance was calculated using the Student's unpaired t test, where *, P < 0.05.

In vivo (ip) administration of NDP-MSH induces expression of the mRNAs encoding Nur77, Il6, Cox2, Pck1, and Pdk4 in (inguinal) adipose tissue

We went on to examine whether NDP-MSH administration induced similar changes in vivo. Seven-week-old C57 BL/6J (male) mice were ip injected with saline (n = 7) and NDP-MSH [1 mg/kg (n = 8)]. Inguinal adipose tissue was collected 4 h after ip injection, RNA was isolated, and qPCR was used to measure the expression of the mRNAs encoding Nur77, IL6, Cox2, Pck1, and Pdk4. We observed that the mRNA encoding Nur77 was very significantly (P < 0.001) induced after 4 h NDP-MSH treatment, relative to saline treatment (Fig. 9). We further examined the expression of the mRNAs encoding Il6, Cox2, Pck1, and Pdk4. We observed 2- to 5-fold increases in the expression of these mRNAs; however, only the approximately 2-fold increase Pck1 mRNA expression attained significance. This suggests that NDP-MSH administration recapitulates similar changes in vivo.

Fig. 9.

NDP-MSH induces Nur77, Il6, Cox2, Pck1, and Pdk4 gene expressions in mouse adipose tissue. Mice were ip injected with vehicle (saline) or 1 mg/kg of NDP-MSH and inguinal adipose tissue collected after 4 h. qPCR was used to assay the mRNA levels, and results were normalized against 36b4 and expressed as fold changes (n = 7–8). Statistical significance was calculated using the Mann-Whitney U test, where *, P < 0.05 and ***, P < 0.001.

Discussion

α-MSHs are important anorectic signaling molecules derived from proopiomelanocortin (POMC), expressed in the hypothalamus, pituitary, skin, and hair follicle. The complexity of melanocortin signaling involves tissue-specific expression of melanocortin receptors and cell specific posttranslational processing of POMC. This complexity drives additional research to elucidate the peripheral actions of α-MSH, which are coordinated, and act in concert with the central nervous system-mediated pathways that control satiety and energy expenditure. Further study of the peripheral action of melanocortins will provide insights into the signaling network controlling feeding, energy expenditure, and body mass (Refs. 15, 26, and 33–35 and references therein).

In this context our study highlights the specific and selective activation of one subgroup of NRs by NDP-MSH in adipocytes. The rapid, striking, and selective nature of this cross talk between Nr4a and MSH signaling is underscored by the observation that other members of the entire NR superfamily of ligand/hormone-dependent DNA binding proteins are refractory to the MSH treatment. Interestingly, this NDP-MSH-mediated activation was associated with the parallel (Nur77 dependent) induction of the mRNAs encoding Il6, Cox2, and Pck1. These data suggest that specific cross talk between Nur77 and MSH play an essential role in the modulation of these metabolic target genes. However, we cannot comment on whether there are primary or secondary targets of Nur77 signaling in 3T3-L1 cells.

The observations from these studies are also consistent with an emerging role and an increasing number of reports that have implicated the Nr4as and MSH action in peripheral tissues in the context of insulin signaling, glucose disposal, oxidative metabolism, adiposity, fatty acid oxidation, leptin expression, etc. For example, the Nur77 orphan nuclear receptor has been demonstrated to regulate Glut4 (mRNA and protein) expression in skeletal muscle (in vitro and in vivo) (10, 11, 36). Moreover, insulin-resistant mouse and rat models, and attenuated expression of genes involved in glucose consumption, are associated with reduced expression of Nur77 and Nor-1 in both skeletal muscle and adipose tissue (7, 10, 36). Furthermore, insulin and thiazolidinedione-peroxisomal proliferator-activated receptor-γ agonists in 3T3-L1 adipocytes markedly induce both Nur77 and Nor-1, and similarly, Nor-1 siRNA expression impairs the insulin-stimulated glucose transport and Akt phosphorylation in adipocytes (6, 7). Finally, Nur77 expression in hepatic tissue activates gluconeogenesis (9).

These roles of the Nr4a subgroup overlap with the emerging role of NDP-MSH in the regulation of lipid metabolism, adiposity, insulin signaling, and glucose metabolism and are in concordance with the associated changes in gene expression observed in these studies (16, 35, 37). For example, the concomitant (Nur77 dependent) induction of the mRNAs encoding IL-6 (Il6), Cox2, and Pck1 are implicated in a number of metabolic functions including insulin signaling, glucose disposal, inducible prostanoid biosynthesis, body fat regulation, adipogenesis, glyceroneogenesis, oxidative metabolism, diabetes, and obesity.

Specifically, Il6 enhances insulin-mediated signaling (in a TNF independent manner) (38), insulin-stimulated glucose disposal, glucose absorption, and lipid oxidation via AMPK-dependent mechanisms (31) and have been implicated in the development of insulin resistance. Moreover, aberrant Il6 expression and function have been shown to increase the risk of obesity. Furthermore, Catania et al. (39) reported iv injections of NDP-MSH in rodent cardiac tissue induced early increases in IL 6 (Il6) mRNA, coupled to increases in Nur77. This is consistent in our study that demonstrated an increase in IL-6 secretion by the 3T3-L1 cells when treated with NDP-MSH.

Another target metabolic gene Cox2, an inducible cyclooxygenase isoform catalyzes an important step in prostanoid synthesis (40). Interestingly, heterozygous (Cox2+/−) lines of mice display increased fat pad sizes, increased body weight, increased serum leptin, and susceptibility to obesity relative to the homozygous Cox2−/− mice. These studies are consistent with a putative role for Cox2 in the regulation of adiposity and body mass (41).

Pck-1 is implicated in glyceroneogenesis, impacting on both the storage and the regulated release of fatty acids via a triglyceride-fatty acid cycle in adipose tissue (42, 43). Moreover, the rate of lipid release is a balance between lipolysis and esterification. Notwithstanding, dysfunctional adipogenic Pck1 expression leads to obesity (due to an increase in triglyceride storage) and insulin resistance (due to a decrease in triglyceride storage) in association with increased and decreased Pck1 activity, respectively (44). Previously Beale et al. (45) discussed a novel hypothesis suggesting that dysfunctional regulation of Pck1 in adipose would be a causal factor in type 2 diabetes and obesity. They argued that Pepck1/Pck1 synthesis is primarily controlled at the transcriptional level, and the promoter functions as a sensor for multiple dietary, nutritional, and signaling pathways. Furthermore, mutations that increase Pepck1/Pck1 activity would increase blood glucose, in association with compromised insulin-mediated control of hepatic gluconeogenesis. In contrast, mutations associated with impaired hepatic Pck1 expression have been discussed to induce hypoglycemia.

In the context of Pdk4, it has been suggested that the pyruvate dehydrogenase complex (attenuated by PDK4) functions as the metabolic switch operating between lipid and carbohydrate utilization, competing for the same substrate (pyruvate) as glyceroneogenesis in adipose tissue (46). In this study, although we clearly identified that NDP-MSH induces the expression of Pdk4, we could not demonstrate that this result was dependent on Nur77 expression. Notwithstanding, studies by Pearen et al. (6) demonstrated that β-adrenergic signaling in skeletal muscle concomitantly induced Nor-1 and Pdk4. These authors also identified that NOR-1 was recruited to the Pdk4 promoter, suggesting that perhaps this member of the Nr4a subgroup regulated Pdk4.

CCAAT/enhancer-binding protein-β came up as differentially expressed in a significant manner, after the 2 h NDP-MSH and cAMP treatment. However, the extent of induction was weak (<3-fold). However, when we examined the temporal response of this member of the CCAAT/enhancer-binding protein-β family of transcription factors to (15 min to 8 h) NDP-MSH treatment, we did not observe consistent and/or robust induction during the time course, in contrast to the other identified target genes (data not shown).

As discussed, the MSH signaling involves the induction of the cAMP signaling cascade; however, melanocortin signaling pathway also involves several other (cAMP independent) pathways including calcium, p38 MAPK, Ras-Raf-MAPK kinase-ERK, the human homolog of agouti, the agouti signaling protein, c-Jun N-terminal kinase, and AMPK (20, 27, 47). Our data indicated that NDP-MSH and cAMP treatment displayed some overlap in the differential expression profile (after treatment) in 3T3-L1 adipocytes. However, we observed distinct and unique effects of NDP-MSH (relative to cAMP) treatment both in the overall pattern of differential expression and the stringently filtered significant hits. For example, NDP-MSH treatment resulted in the very significant induction of Il6, Pck1, Pdk4, and Cox2. In contrast, cAMP treatment did not produce significant differential expression/induction of the mRNAs encoding Il6, Pdk4, and Cox2. This underscores the importance of examining both signaling compounds and that the NDP-MSH-mediated induction of gene expression involves cAMP and cAMP-independent signaling.

Finally, we demonstrated that ip administration of NDP-MSH (in C57 BL/6J male) mice increases the in vivo expression of the mRNAs encoding Nur77, Il6, Cox2, Pck1, and Pdk4 in (inguinal) adipose tissue. As discussed, this is consistent with the observation of Catania et al. (39) that showed iv injections of NDP-MSH in rodent cardiac tissue induced early increases in IL-6 (Il6) mRNA, and Jun et al. (47) demonstrated that tail vein injection of α-MSH induced IL-6 mRNA expression. This suggests that NDP-MSH administration recapitulates similar changes in vivo.

In conclusion, our study clearly demonstrates the cross talk between MSH signaling, Nr4a induction, and the subsequent concomitant regulation of metabolic gene expression. Moreover, we demonstrate that Nur77 mRNA expression is necessary for many aspects of MSH-dependent effects on the expression of metabolic genes (involved in glucose homeostasis). Furthermore, we conclude that α-MSH signaling preferentially targets the Nr4a subfamily because other members of the NR superfamily were refractory to the anorectic hormone. Accordingly, these studies provide novel information in the context of MSH signaling, Nr4a regulation, and subsequent metabolic profiling of target genes in the framework of adipocyte biology.

Materials and Methods

Cell culture

3T3-L1 cells (American Type Culture Collection, Manassas, VA) were cultured and maintained in DMEM medium supplemented with 10% heat-inactivated fetal bovine serum (FBS). Preadipocytes were induced to differentiate at 2 d after confluence in DMEM supplemented with 10% FBS, 0.1 μg/ml dexamethasone (Sigma, St Louis, MO), 4 μg/ml insulin (Sigma), and 115 μg/ml isobutylmethylxanthine (Sigma) for 3 d and then back to DMEM with 10% FBS in which the cells were maintained and refed every 2–3 d. Approximately 80–90% of cells displayed adipocyte phenotype. The 3T3-L1 adipocytes used were further treated with either vehicle (PBS) and a synthetic and more potent analog of α-MSH NDP-MSH or 8-Br cAMP.

Animal procedures

C57BL/6J male mice (7 wk old) were obtained from colony at The University of Queensland and housed in the Queensland Bioscience Precinct vivarium (University of Queensland, St. Lucia, Australia) with a 12 h light, 12-h dark cycle. Mice were injected with vehicle (saline) or 1 mg/kg NDP-MSH ip for 4 h and killed by cervical dislocation. Inguinal fat pads were taken out immediately afterward. Total RNA was extracted from inguinal fat pads using QIAzol reagent (QIAGEN, Clifton Hill, Victoria, Australia) according to the manufacturer's protocol. All aspects of the animal experimentation were approved by The University of Queensland Animal Ethics Committee.

RNA extraction and cDNA synthesis

Total RNA was extracted from 3T3-L1 cells using Tri-Reagent (Sigma) according to the manufacturer's protocol. After treatment with Turbo deoxyribonuclease I (Ambion, Austin, TX) for 30 min at 37 C, RNA was further purified through RNeasy column (QIAGEN). RNA was electrophoresed to determine the integrity of the preparation. Superscript III was used in the synthesis of cDNA from 2 μg of total RNA with random hexamers according to the manufacturer's instruction (Invitrogen, Carlsbad, CA). The cDNA was diluted to 300 μl in nuclease-free water.

Quantitative real-time PCR

qPCR was performed as previously described (1, 12). SYBR primers used are: 36b4 forward, 5′-AGATGCAGCAGATCCGCA-3′, reverse, 5′-GTTCTTGCCCATCAGCACC-3′; Nor-1 forward, 5′-GATCACAGAGCGACATGGGTTA-3′, reverse, 5′-GAGCCTGTCCCTTCCTCTGG-3′; Nur77 forward, 5′-TGATGTTCCCGCCTTTGC-3′, reverse, 5′-CAATGCGATTCTGCAGCTCTT-3′; Nurr1 forward, 5′-GCACTTCGGCGGAGTTG-3′, reverse, 5′-GGAATCCAGCCCGTCAGA-3′; Il6 forward, 5′-AGCCAGAGTCCTTCAGA-3′, reverse, 5′-GGTCCTTAGCCACTCCT-3′; Pdk4 forward, 5′-AAAGGACAGGATGGAAGGAATCA-3′, reverse, 5′-TTTTCCTCTGGGTTTGCACAT-3′; and Pck-1 forward, 5′-GGGAAATCACCGCATAGTCTCT-3′, reverse, 5′-TAACTGCTACAGCTAACGTGAAGAACT-3′. All SYBR primers were designed using Primer Express (Applied Biosystems, Foster City, CA). Commercially available TaqMan gene expression assays (Applied Biosystems) were used to assay expression of Pgts2/Cox2. Statistical analysis was performed on the average of three to four independent assays using unpaired Student's t tests.

TaqMan low-density array

cDNA synthesized from 3T3-L1 cells treated with NDP-MSH, 8-Br cAMP, or vehicle were loaded onto a custom-designed ABI microfluidic TLDAs to analyze the expression of genes involved in metabolism (lipid, carbohydrate, and energy homeostasis). Five internal control genes were used, including the mandatory control (18S rRNA) and four other controls: Gapdh, GusB, Hprt1, and 36b4/Arbp. These controls span the relative abundance/Ct range of the genes on the card, and three (18S rRNA, Gapdh, and 36b4) are validated real-time PCR controls from NURSA-supported nuclear receptor studies (29, 48). The Genorm software imbedded within the ABI/Integromics StatMiner version 4.1 software package was used to compute least expression variation and select the appropriate (and most stable) combination of internal control genes with which to normalize the expression data (against the mean of the most stable controls). However, data can also be normalized against a specific selected control of choice, for example, NURSA used controls. Briefly, significant changes in expression relative were analyzed using the ABI/Integromics StatMiner software package. Differentially expressed genes were identified by linear models (contained in the LIMMA package for Bioconductor R embedded in StatMiner).

Significance is assigned by the application of the empirical Bayes statistic, described as equivalent to shrinkage of the estimated sample variances toward a pooled estimate, resulting in far more stable inference when the number of arrays is small. It returns the empirical Bayes log odds of differential expression (i.e. the probability) that a gene is differentially expressed (a higher score represents a more significant result). For example, the B statistic of zero indicates a 50:50 chance of differential expression, B scores greater than 0 indicate greater than 50:50 chance of differential expression, B less than 0 (negative scores) reflect odds that a gene is more than likely, not differentially expressed. The B statistic considers and ranks a proportion of differentially expressed genes (P < 0.01). Analysis also includes t, the empirical Bayes moderated t statistic (a variant t test), an empirically moderated estimate of standard error (49, 50). Furthermore, we perform conservative data filtering (Benjamini-Hochberg) to control for FDR and correct/adjust the P values. The adjusted P values (after FDR), coupled to the B score, enable stringent refinement the subset of differentially expressed genes.

Relative quantification, i.e. the calculated fold differences [between the target (MSH or cAMP treated) and the calibrator/reference sample/tissue (vehicle treated)] are displayed as valid, when the Ct values of the gene in the target and calibrator/reference samples is less than 35 cycles. Moreover, fold differences are flagged as target not detected, calibrator not detected, and no detection when the Ct value of the gene(s) in the target, calibrator, and/or both tissues/samples are 35 cycles or greater, respectively (than the arbitrarily set threshold limit). Hence, the reported quantitative “fold change of a gene that is not expressed in some of the biological conditions may not be reliable” (50); however, it does reflect a qualitative difference. The data generated by the ABI SDS software from the ABI7900 instrument do not normally contain missing Ct values, and Ct values are assigned beyond the arbitrarily set threshold (Ct 35) up to a maximum Ct 40. In situations in which the sample is undetected and the Ct values are beyond the maximum Ct 40, the StatMiner software imputes a value, set to the maximum Ct.

Protein extraction and Western blotting

Nuclear protein was isolated as previously described (51) from 3T3-L1 adipocytes and concentration determined with BCA protein assay reagent (Pierce, Rockford, IL). Nuclear protein extracted were resolved on 10% SDS-PAGE gels and transferred to polyvinyl difluoride membranes (Milipore, Bedford, MA). The membranes were blocked for 1 h in Tris-buffered saline containing 0.5% Tween 20 and 5% skim milk, followed by an overnight incubation with either Nur77 or TBP antibodies (Santa Cruz Biotechnologies, Santa Cruz, CA). The membranes were further incubated with peroxidase-conjugated secondary antibody for 1 h at room temperature. After being washed with Tris-buffered saline containing 0.5% Tween 20 four times, the signals were detected with ECL Plus Western blotting detection system (GE Healthcare, Little Chalfont, Buckinghamshire, UK), and visualized by autoradiography.

Transfection and generation of siRNA-Nur77 stable cell line

Transient transfection of 3T3-L1 cells were performed in a six-well plate (∼70% confluence) with 30 nm Nur77 siRNA on target plus SMARTpool (Dharmacon, Lafayette, CO) or negative SMARTpool control (Dharmacon). Cells were transfected with Lipofectamine 2000 (Invitrogen) as described in the manufacturer's protocol. Twenty-four hours after transfection, the growth medium was replaced and cells grown for another day until confluent. Differentiation medium (as described above) were added and cells harvested after 1 h. For generation of 3T3-L1, stable cells were transfected with 4 μg of pSilencer 3.1 neomycin Nur77 (11) or 4 μg pSilencer 3.1 neomycin negative in 25-cm² flasks at approximately 40% confluence using Lipofectamine 2000(Invitrogen). Twenty-four hours after transfection, cells were selected for neomycin resistance expression by treatment with 600 mg/ml G418 (Astral Scientific, Caringbah, Australia). G418-resistant cells were maintained in growth medium supplemented with 300 mg/ml G418. The pool of polyclonal cells selected were passaged into three replicates and differentiated into adipocytes and further treated with NDP-MSH or vehicle as described above. RNA was extracted from each replicate independently.

Measurement of secreted IL-6

Medium from differentiated 3T3-L1 adipocytes treated for 2 h with either NDP-MSH or vehicle was collected and analyzed for IL-6 using a mouse IL-6 ELISA kit (BD Biosciences, San Diego, CA) as per manufacturer's instruction.

Glucose uptake

3T3-L1 cells were transiently transfected with Nur77 siRNA or negative control in six-well plates as described above. NDP-MSH or vehicle was added onto the cells together with differentiation medium and assayed after 2 h. Assays were performed as described (52). Briefly, cells were washed twice with Krebs Ringer HEPES buffer (KRH; 121 mm NaCl; 4.9 mm KCl; 1.2 mm MgSO₄; 0.33 mm CaCl₂; 12 mm HEPES, pH 7.4) followed by 13 min incubation at 37 C with either insulin (100 nm) or carrier in 900 μl of KRH. One hundred microliters of working radioactive glucose cocktail, ³H-2-deoxy-d-glucose (final specific activity, 6.7 μCi; Amersham Biosciences, Piscataway, NJ) was added for an additional 15 min. Uptake was terminated with three quick ice-cold washes with KRH, and cells were lysed in 400 μl of 0.1% sodium dodecyl sulfate in KRH. Radioactivity was determined by liquid scintillation counting.