| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
INSERM Unit 317 (N.V., L.M., S.A., D.Lar., D.Lan.), Institut Louis Bugnard, Université Paul Sabatier, Hôpital Rangueil, 31403 Toulouse Cedex 4, France; and INSERM Unit 449 (H.V.), Faculté de Médecine Laennec, 69372 Lyon Cedex 08, France
Address all correspondence and requests for reprints to: Dr. Nathalie Viguerie, INSERM U317, Institut Louis Bugnard, B
timent L3, Centre Hospitalier Universitaire Rangueil, 31403 Toulouse Cedex 4, France. E-mail: viguenat{at}rangueil inserm.fr.
Abstract
Thyroid hormones are key regulators of metabolism. In adipose tissue, changes in thyroid status result in alterations of lipolytic capacity. The effects of these hormones are mediated by thyroid hormone receptors that modulate gene transcription. Very few target genes have been identified in adipose tissue. To investigate the effect of T3 on gene expression in human adipocytes, primary cultures of human sc adipose tissue explants were treated with T3. 32P-labeled cDNA probes prepared from isolated adipocyte total RNA were hybridized to cDNA arrays representing 1,176 genes. Among the statistically significant variations in mRNA levels with more than 1.3-fold difference, 13 and 6 genes were positively and negatively regulated, respectively (n = 3). The genes encoded proteins that were involved in signal transduction, lipid metabolism, apoptosis, and inflammatory response. Using RT-competitive PCR, we showed a down-regulation of phosphodiesterase 3B, 
2A-adrenergic receptor, and G protein i2 subunit mRNAs, and an up-regulation of ß2-adrenergic receptor mRNA. These regulations may explain the T3-mediated increase in catecholamine-induced lipolysis. The down-regulation of sterol regulatory element binding protein-1c, a transcription factor controlling lipogenic gene expression, may constitute a link between thyrotoxicosis and insulin resistance. Thus, these data suggest that T3 modulates expression of genes with a wide range of function in human adipose tissue.
REGULATION OF FAT mass is controlled by multiple neuroendocrine signals. Thyroid hormones influence adipose tissue development and metabolism. Hyperthyroidism induces a transient hyperplasia concomitant to cell size reduction, whereas an opposite pattern is observed during hypothyroidism (1). T3 modulates both proliferation and differentiation of adipocytes (2, 3). A well known effect of thyroid hormones is the alteration of the lipolytic response to catecholamines. In humans, hyperthyroidism enhances and hypothyroidism decreases lipolysis through different mechanisms. The induction of catecholamine-mediated lipolysis by thyroid hormones results from an increased ß-adrenoceptor (AR) number and a decrease in phosphodiesterase (PDE) activity (4, 5). Both regulations concur to an increase in cAMP level and hormone-sensitive lipase (HSL) activity. Thyroid status has also been shown to affect other pathways in lipid and carbohydrate metabolism in liver, skeletal muscle, and heart (6). Although few studies have been performed in adipose tissue, it is suspected that some regulations are common to different cell types. For example, we recently showed that the up-regulation of uncoupling protein 2 (UCP2) mRNA by T3 is observed in human adipose tissue and skeletal muscle (7).
T3 exerts pleiotropic effects through modulation of gene expression by specific nuclear receptors, thyroid hormone. TR1 and TRß1, the products of two different genes, are expressed in adipose tissue (8). Most of the characterized thyroid- response elements are positive cis-acting elements in which gene transcription is repressed by unliganded TR and activated by T3-occupied TR (9). In the presence of ligand, the TR that is heterodimerized with the retinoid X receptor undergoes a conformational change that results in the replacement of a corepressor complex by a coactivator complex. Although thyroid hormones are key regulators of metabolism, few target genes have been identified. No large-scale studies have been performed in humans. In the present work, we studied the effect of T3 on human adipocyte gene expression using primary culture of human adipose tissue explants and cDNA arrays. Through the simultaneous investigation of thousand of genes, DNA arrays are ideally suited to unravel novel hormonal regulations. Moreover, the molecular basis for T3-mediated regulation of catecholamine-induced lipolysis was determined by the measurement of mRNA levels for key factors of the lipolytic cascade.
Materials and Methods
Chemicals
DMEM, penicillin, streptomycin, gentamycin, and FCS were obtained from Life Technologies (Gaithersburg, MD). T3, collagenase A, and BSA (fraction V) were obtained from Sigma-Aldrich Corp. (St. Louis, MO). Atlas Human 1.2 Array II was purchased from CLONTECH Laboratories, Inc. (Palo Alto, CA). The array includes 1,176 cDNAs immobilized on a nylon membrane. A wide range of biological functions are represented. The complete list of genes represented on the array is available at http://atlas.clontech.com.
Tissue culture
Human adipose tissue was obtained, in agreement with French laws on biomedical research from the sc abdominal fat depots of 14 (age, 44 ± 11 yr; body mass index, 25 ± 4 kg/m2) and 9 (age, 40 ± 8 yr; body mass index, 24 ± 3 kg/m2) Caucasian women for cDNA array and RT- competitive PCR experiments, respectively. Surgical adipose tissue samples were dissected from skin and vessels, rinsed once in warmed PBS, and transferred in a sterile environment. The fat pads were cut into small pieces ranging from 100–400 mg, then placed in DMEM supplemented with 5% FCS, penicillin (200 U/ml), streptomycin (50 µg/ml), and gentamycin (200 µg/ml). The fat pieces were distributed into 75-cm2 polystyrene flasks (Falcon, Becton Dickinson and Co., Meylan, France) up to 25 ml containing from 6–9 g of adipose tissue, then maintained at 37 C in a 7% CO2 chamber. One day later, the medium was changed to FCS-free DMEM, and cultures were treated with T3 (100 nM) or vehicle for 24 h. Medium-free T3 concentration was measured at 1 and 24 h after addition of T3 using RIA kits from Institut Pasteur (Paris, France).
After treatment, the medium was removed, and the fat pieces were immediately digested with 0.5 mg/ml collagenase in DMEM containing 3% (wt/vol) BSA at 37 C under gentle agitation. One hour later, isolated adipocytes were filtered through a nylon mesh and washed twice with PBS. The infranatant was removed, and the resulting packed cells (about 5 ml) were lysed with an equivalent amount of denaturing buffer from RNeasy kit (QIAGEN, Courtaboeuf, France) and then stored at -80 C. Total RNA was extracted using the QIAGEN RNeasy kit and stored at -80 C until analysis.
Differential hybridization of human cDNA expression array
Three cDNA array experiments were performed with adipose tissue preparations from three, seven, and four subjects, respectively. To remove residual genomic DNA in the preparations, total RNA from T3-treated or nontreated adipocytes were first incubated with DNase I at 0.1 U/µg total RNA for 30 min at 37 C. The first strand cDNA synthesis was performed using magnetic bead affinity enrichment of poly A+ RNA on a pool of 40 µg total RNA and direct synthesis of cDNA probe using the gene-specific primer mix from CLONTECH Laboratories, Inc. Probes were prepared using [-32P] dATP label mix and purified with column chromatography as specified by the manufacturer. Equal amounts of purified cDNA probe (2 x 106 cpm/ml) from either control or T3-treated cells were hybridized to two Atlas Human cDNA expression arrays in separate rollers overnight at 68 C in Express Hyb solution (CLONTECH Laboratories, Inc.) The membranes were washed three times in 2x SSC and 1% SDS for 30 min at 68 C, then two times in 0.1x SSC and 0.5% SDS for 30 min at 68 C. The signals were captured on a PhosphorImager (Molecular Dynamics, Inc., Sunnyvale, CA). Images were analyzed with the ImaGene 4.1 software (Biodiscovery Inc., Los Angeles, CA). For each hybridization, the spots with an intensity below 1.4-fold above the background were eliminated. The adjusted intensity (total signal minus background signal) for each good quality spot was then normalized to the average of five housekeeping gene intensities. The housekeeping genes used for signal normalization were ubiquitin, hypoxanthine-guanine phosphoribosyltransferase, HLA class I histocompatibility antigen C-4 , 60S ribosomal protein L13A, and 40S ribosomal protein S9. Statistical significance of the data was assessed using a t test for paired values. Then, we chose a threshold value of 1.3 for up- or down-regulation of significant genes. This threshold value corresponds to variations that we can confidently confirm using RT-competitive PCR as shown for ß2-AR and G protein i2 subunit (Gi2) mRNA variations in Fig. 1
.
|
Quantitation of mRNA levels by RT-competitive PCR was validated and performed as described previously (10) on adipocyte total RNA from nine subjects. The RT step was performed on 100 ng of total RNA with a specific antisense primer and Omniscript reverse transcriptase (QIAGEN). Then, cDNA was amplified by PCR by using sense and antisense primers listed in Table 1 in the presence of known amounts of a specific DNA competitor. DNA competitors were obtained by a deletion in the native sequences using a two-step overlap PCR extension method. Competitive PCR products were separated by capillary electrophoresis and quantified using the ABI PRISM 310 Genetic Analyzer system with the Genescan software (Applied Biosystems, Foster City, CA). To normalize data between RT-competitive PCR experiments, we used, as previously described (7), cyclophilin A (CYP) gene expression. mRNA abundance was therefore expressed as the ratio of target gene to CYP mRNA levels. The t test for paired values was used for comparisons between treated and untreated adipocytes.
|
To determine whether culture conditions induced changes in gene expression, UCP2 mRNA levels were determined using RT-competitive PCR (7, 10) on freshly isolated adipocytes and on adipocytes isolated from explants cultured for 48 h. There was no significant variation in UCP2 gene expression (96 ± 12%; n = 9). Explant cultures were treated with 100 nM T3 for 24 h in serum-free medium. Free T3 concentrations were 40 nM in the medium at 1 and 24 h after addition of T3, suggesting that more than half of T3 was rapidly bound. Free T3 was undetectable in vehicle-treated medium. To verify that T3 treatment induced gene expression in each culture preparation, variations in UCP2 mRNA levels were determined. The fold inductions were 2.3 ± 0.2 (P < 0.01), 3.3 ± 0.9 (P < 0.02), and 2.9 ± 0.3 (P < 0.01) in experiments 1, 2, and 3, respectively. The increase is similar to the induction observed in vivo in human adipose tissue of subjects with a 2-fold increase in plasma free T3 levels for 14 days (7). Then, radioactively labeled probes were synthesized using a mixture of primers specific to the genes spotted onto the membrane. There were no signals on control human genomic DNA spots, indicating that RNA pools used for probe synthesis were free of genomic DNA. Among the 1,176 cDNAs analyzed, 209 (18%) were expressed at sufficient levels in human adipocytes for detection in the 3 hybridization experiments. Among the 19 cDNAs with significant (P < 0.05) and more than 1.3-fold variation (range, 1.3–2.5), 13 were up-regulated (Table 2), and 6 were down-regulated (Table 3). We used information available on the Stanford Online Universal Resource for Clones and Ests (genome-www4. stanford.edu/MicroArray/SMD/), which includes gene ontology annotations (www.geneontology.org) to describe the molecular function, biological process, and cellular component of T3-regulated gene products.
|
|
i2, ß2-AR, 2-AR, HSL, PDE3B, and CYP were determined. As shown in Fig. 1
, SREBP-1c and Gi2 were down-regulated (T3/control, 0.58 ± 0.05, P < 0.02; and 0.76 ± 0.07, P < 0.05, respectively). The data from RT-competitive PCR were therefore in agreement with the data from the cDNA arrays. The 2-AR mRNA level was decreased (0.69 ± 0.06; P < 0.05), whereas the ß2-AR mRNA level was increased (1.39 ± 0.13; P < 0.01). There was no significant change in HSL mRNA levels (1.06 ± 0.27; NS) as observed in cDNA array experiments. T3 down-regulated mRNA expression of PDE3B (0.64 ± 0.05; P < 0.02). Discussion
To study the thyroid hormone regulation of gene expression in human adipocytes, we combined the use of cDNA arrays and primary culture of sc adipose tissue explants. cDNA arrays are well suited to perform high-throughput analysis of differential gene expression. So far, the hormonal regulation of gene expression has not been studied at a large-scale level in human adipocytes. A suitable culture system is necessary for RNA-demanding cDNA array experiments. Culture of freshly isolated mature adipocytes is not appropriate because of decrease in gene expression over time and large differences in the viability of cells from different subjects (11). Primary cultures of human preadipocytes (2) yield cells at different differentiation stages. It is therefore difficult to determine whether the observed hormonal regulation is due to an effect on differentiated adipocytes or on preadipocytes. Moreover, the differentiated cells do not recapitulate the full phenotype of mature adipocytes, i.e. large cells with unilocular lipid droplets and expression of late differentiation markers such as the 2-AR (12). We therefore chose to study hormonal regulation on adipose tissue explants. Explants partly maintain the in vivo structure of human adipose tissue that permits long-term culture with minimal change in gene expression as shown here for UCP2 mRNA level. At the end of the treatment period, mature adipocytes are isolated. Gene expression is therefore measured on one single cell type.
The effect of thyroid hormones on sc adipose tissue lipolysis is well documented (5, 13, 14). Hyperthyroid patients are characterized by an enhanced lipolytic response to catecholamines. Here, we provide part of the molecular basis for thyroid hormone action on lipolysis. T3 increased the mRNA levels of the lipolytic ß2-AR. Accordingly, hyperthyroidism is accompanied by an increase in ß2-AR number with no change in ß1-AR level (14). The activity of low Km phosphodiesterase, an enzyme that hydrolyzes cAMP into 5'AMP, is lower in hyperthyroid patients compared with euthyroid subjects (4). The down-regulation of phosphodiesterase activity contributes to an increase in fat cell cAMP levels and enhanced lipolysis (15). This activity is most likely attributable to PDE3B, which is a key cellular relay in the antilipolytic action of insulin (16). In human adipose tissue culture, T3 treatment led to a 1.6-fold down-regulation of PDE3B mRNA. The 2-AR is highly expressed in human sc adipose tissue in which it mediates through coupling to Gi proteins the antilipolytic effect of catecholamines (12). The concomitant down-regulation of 2-AR and Gi2 mRNA levels may lead to a decreased antilipolytic potency of catecholamines. Together, these regulations favor an increase in intracellular cAMP levels and contribute to the enhanced lipolytic capacity of catecholamines.
Other genes encoding proteins involved in G protein- coupled receptor signal transduction pathways were regulated under T3. A homologue of RDC1, a seven transmembrane spanning receptor that binds calcitonin gene-related peptide and adrenomedullin, was induced during T3 treatment. G protein-coupled receptor kinase 4 belongs to a family of serine/threonine kinases that play a critical role in receptor desensitization. Guanine nucleotide-binding protein ß-subunit-like protein 12 is part of a family of intracellular receptors that anchor activated protein kinase C to the cytoskeleton (17). Through up-regulation of G protein- coupled receptor kinases and down-regulation of receptors for activated protein kinase C, thyroid hormone may exert an inhibitory effect on hormonal regulations of adipose tissue metabolism.
Several proteins involved in lipid metabolism were regulated by T3. SREBP1-c is a member of the basic helix-loop-helix family of transcription factors that is up-regulated during adipocyte differentiation (18). Studies in murine adipocyte cell lines have shown that SREBP1c influences adipogenesis positively and stimulates the expression of genes involved in lipogenesis (19). It has recently been shown that SREBP-1c may be a major mediator of insulin action on hepatic glucose and lipid metabolism gene expression (20). The down-regulation of SREBP-1c by T3 in mature adipocytes suggests that the transcription factor may be involved in the insulin-resistant state associated with hyperthyroidism (21). Apolipoprotein D is a glycoprotein that is primarily associated with high-density lipoproteins in human plasma. Its mRNA is expressed in white adipose tissue (22). An association has been reported between an apolipoprotein D gene polymorphism and obesity (23). In line with an up-regulation in human adipocytes, the human apolipoprotein D promoter contains thyroid hormone responsive elements (24). Through the up-regulation of squalene synthetase, the enzyme that catalyzes the first step in the cholesterol biosynthetic pathway, T3 may modulate cholesterol metabolism in human adipose tissue.
Annexins are a family of Ca2+-dependent phospholipid binding proteins that are preferentially located on the cytosolic face of the plasma membrane (25). Through membrane-membrane and membrane-cytoskeleton interactions, they influence signal transduction pathways and Ca2+ flux. An increase in annexin I, II, and V expression has been described in thyroid hormone-deficient rats; hyperthyroidism gives opposite results (26). Such a regulation is in accordance with the present down-regulation of annexin I in adipose tissue treated with T3. T3 also down-regulated the expression of the gene encoding the S100A3/S100E calcium-binding protein. Several annexins have been shown to bind proteins that belong to the S100 family. The S100 proteins regulate many intracellular events (27). The lower expression of both annexin I and S100A3/S100E protein under T3 could modulate the response to elevation of cytoplasmic Ca2+ levels and modify numerous signal transduction pathways.
In human adipocytes, T3 up-regulated the mRNA for both the lipopolysaccharide (LPS) binding protein and the LPS receptor. On the surface of macrophages, the LPS binding protein/LPS complex interacts with the LPS receptor to initiate an inflammatory reaction (28). The presence of the LPS receptor mRNA has previously been described in human adipocytes and is involved in the release of TNF (29). TNF may play a role in the insulin resistance state associated with obesity (30). Moreover, the cytokine has been shown to stimulate lipolysis and to induce apoptosis in human adipocytes (31, 32). An important event in apoptosis is the opening of the inner mitochondrial membrane permeability transition pore that induces mitochondrion swelling. The peptidyl-prolyl cis-trans-isomerase, which is part of the mitochondrial permeability transition pore (33), was up-regulated by T3. Through these regulations, T3 could influence apoptosis and inflammatory response.
To conclude, cDNA array data show that T3 regulates a large repertoire of genes in human adipocytes. The present results provide a molecular basis to the effect of T3 on catecholamine-induced lipolysis. The down-regulation of SREBP1c may constitute a link between hyperthyroidism and insulin resistance. Moreover, we also show that cellular processes that had not previously been investigated, such as signal transduction, apoptosis, and inflammatory response, may be regulated by thyroid hormone.
Footnotes
This work was supported in part by grants from ALFEDIAM/Novo-Nordisk.
The laboratories involved in this study are members of the Concerted Action FATLINK (FAIR-CT98-4141) supported by the European Commission.
Abbreviations: AR, Adrenoceptor; CYP, cyclophilin; Gi2, G protein i2 subunit; HSL, hormone-sensitive lipase; LPS, lipopolysaccharide; PDE, phosphodiesterase 3B; SREBP-1c, sterol regulatory element binding protein-1c; TR, thyroid hormone receptor; UCP2, uncoupling protein 2.
Received January 26, 2001.
Accepted October 16, 2001.
References
, c-erbAß and Rev-erbA mRNA during the conversion of brown adipose tissue into white adipose tissue. Mol Cell Endocrinol 116:59–65[CrossRef][Medline]
2-adrenoceptors: the regulation of fat cell function and lipolysis. Endocr Rev 16:716–738. Cell 73:625–627[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  M. Alevizaki, K. Saltiki, P. Voidonikola, E. Mantzou, C. Papamichael, and K. Stamatelopoulos Free thyroxine is an independent predictor of subcutaneous fat in euthyroid individuals Eur. J. Endocrinol., September 1, 2009; 161(3): 459 - 465. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hashimoto, E. Ishida, S. Matsumoto, S. Okada, M. Yamada, T. Satoh, T. Monden, and M. Mori Carbohydrate Response Element Binding Protein Gene Expression Is Positively Regulated by Thyroid Hormone Endocrinology, July 1, 2009; 150(7): 3417 - 3424. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wang and M. A. Lazar Bifunctional Role of Rev-erb{alpha} in Adipocyte Differentiation Mol. Cell. Biol., April 1, 2008; 28(7): 2213 - 2220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Polak, C. Moro, E. Klimcakova, M. Kovacikova, M. Bajzova, M. Vitkova, Z. Kovacova, R. Sotornik, M. Berlan, N. Viguerie, et al. The atrial natriuretic peptide- and catecholamine-induced lipolysis and expression of related genes in adipose tissue in hypothyroid and hyperthyroid patients Am J Physiol Endocrinol Metab, July 1, 2007; 293(1): E246 - E251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cosentino, A. M. Fietta, M. Ferrari, E. Rasini, R. Bombelli, E. Carcano, F. Saporiti, F. Meloni, F. Marino, and S. Lecchini Human CD4+CD25+ regulatory T cells selectively express tyrosine hydroxylase and contain endogenous catecholamines subserving an autocrine/paracrine inhibitory functional loop Blood, January 15, 2007; 109(2): 632 - 642. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. T. Rae, O. Gubbay, A. Kostogiannou, D. Price, H. O. D. Critchley, and S. G. Hillier Thyroid Hormone Signaling in Human Ovarian Surface Epithelial Cells J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 322 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P Peeters, W. M van der Deure, and T. J Visser Genetic variation in thyroid hormone pathway genes; polymorphisms in the TSH receptor and the iodothyronine deiodinases. Eur. J. Endocrinol., November 1, 2006; 155(5): 655 - 662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Nichol, M. Christian, J. H. Steel, R. White, and M. G. Parker RIP140 Expression Is Stimulated by Estrogen-related Receptor {alpha} during Adipogenesis J. Biol. Chem., October 27, 2006; 281(43): 32140 - 32147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hashimoto, M. Yamada, S. Matsumoto, T. Monden, T. Satoh, and M. Mori Mouse Sterol Response Element Binding Protein-1c Gene Expression Is Negatively Regulated by Thyroid Hormone Endocrinology, September 1, 2006; 147(9): 4292 - 4302. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Leonardsson, J. H. Steel, M. Christian, V. Pocock, S. Milligan, J. Bell, P.-W. So, G. Medina-Gomez, A. Vidal-Puig, R. White, et al. Nuclear receptor corepressor RIP140 regulates fat accumulation PNAS, June 1, 2004; 101(22): 8437 - 8442. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-Y. Liu, J. J. Schultz, and G. A. Brent A Thyroid Hormone Receptor {alpha} Gene Mutation (P398H) Is Associated with Visceral Adiposity and Impaired Catecholamine-stimulated Lipolysis in Mice J. Biol. Chem., October 3, 2003; 278(40): 38913 - 38920. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. D. Riis, C. H. Gravholt, C. B. Djurhuus, H. Norrelund, J. O. L. Jorgensen, J. Weeke, and N. Moller Elevated Regional Lipolysis in Hyperthyroidism J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4747 - 4753. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
