To understand the roles of thyroid hormone receptors (TRs) in adipogenesis, we adopted a loss-of-function approach. We generated 3T3-L1 cells stably expressing either TRα1 mutant (TRα1PV) or TRβ1 mutant (TRβ1PV). TRα1PV and TRβ1PV are dominant negative mutations with a frameshift in the C-terminal amino acids. In control cells, the thyroid hormone, tri-iodothyronine (T3), induced a 2.5-fold increase in adipogenesis in 3T3-L1 cells, as demonstrated by increased lipid droplets. This increase was mediated by T3-induced expression of the peroxisome proliferator-activated receptor γ (PPARγ) and CCAAT/enhancer-binding protein α (C/EBPα), which are master regulators of adipogenesis at both the mRNA and protein levels. In 3T3-L1 cells stably expressing TRα1PV (L1-α1PV cells) or TRβ1PV (L1-β1PV cells), adipogenesis was reduced 94 or 54% respectively, indicative of differential inhibitory activity of mutant TR isoforms. Concordantly, the expression of PPARγ and C/EBPα at the mRNA and protein levels was more repressed in L1-α1PV cells than in L1-β1PV cells. In addition, the expression of PPARγ downstream target genes involved in fatty acid synthesis – the lipoprotein lipase (Lpl) and aP2 involved in adipogenesis – was more inhibited by TRα1PV than by TRβ1PV. Chromatin immunoprecipitation assays showed that TRα1PV was more avidly recruited than TRβ1PV to the promoter to preferentially block the expression of the C/ebpα gene. Taken together, these data indicate that impaired adipogenesis by mutant TR is isoform dependent. The finding that induction of adipogenesis is differentially regulated by TR isoforms suggests that TR isoform-specific ligands could be designed for therapeutic intervention for lipid abnormalities.
Thyroid hormone (tri-iodothyronine, T3) has long been recognized to play a key role in lipid metabolism and thermogenesis. However, the molecular mechanisms by which T3 regulates lipid homeostasis have not been fully elucidated. The genomic actions of T3 are mediated by thyroid hormone nuclear receptors (TRs). There are two TR genes, TRα and TRβ, located on different chromosomes, to encode three major T3 binding receptors, α1, β1, and β2 (Cheng 2000). These TR isoforms, expressed in a tissue-specific and development-regulated manner, regulate growth, differentiation, and development, and maintain metabolic homeostasis. TRs are ligand-dependent transcription factors that bind to specific DNA sequences known as thyroid hormone response elements (TREs) on the promoter of T3 target genes. The regulation of the transcriptional activity of TRs is complex. In addition to T3, their transcriptional activity is further regulated by a host of corepressors and coactivators (Cheng 2000, O'Malley 2003).
Recent progress in the understanding of TR action on lipid homeostasis has been achieved via genetically engineered mice. Mice deficient in the Trα gene exhibit a lower body temperature than the wild-type mice, indicating that Trα1 is important in thermogenesis and energy balance (Wikstrom et al. 1998). Mice lacking the Trβ gene have abnormal cholesterol regulation (Gullberg et al. 2002). Mice devoid of all functional TRs display decreased body temperature and basal metabolic rate, and are cold intolerant due to insufficient heat production (Golozoubova et al. 2004). These studies revealed that TR isoforms act to regulate different aspects of metabolic homeostasis. More insights into TR actions in lipid metabolism came from the use of loss-of-function approach via knock-in mutations of TR genes in mice. Indeed, knock-in mutant mice harboring different mutations of the Trα gene exhibit abnormalities in lipid metabolism. The Trα1P398H knock-in mutant mice have increased body fat accumulation and elevated serum levels of leptin, glucose, and insulin (Liu et al. 2003). In contrast, the Trα1R384C knock-in mutant mice exhibit a lean phenotype with a reduction in fat mass (Sjogren et al. 2007). The Trα1PV mice that harbor a frameshift mutation in the C-terminal 17 amino acids (Fig. 1) display a lean phenotype, partly due to the reduction in the mass of white adipose tissue (WAT; Kaneshige et al. 2001, Ying et al. 2007). The reasons for the different phenotypic manifestations in these three knock-in mutant mice are not clear. The different phenotypes could reflect the differences in the degree of loss of T3 binding and the potency of dominant negative activity of Trα1 mutants. Trα1PV mice completely lose T3 binding and exhibit potent dominant negative activity (Kaneshige et al. 2001). In contrast, Trα1P398H and Trα1R384C mice only partially lose T3 binding activity (Tinnikov et al. 2002, Liu et al. 2003). The weaker dominant negative activity of Trα1R384C is consistent with the restoration of the target gene expression and the rescue of growth retardation in Trα1R384C mice when serum thyroxine concentration was increased tenfold (Tinnikov et al. 2002).
The expression of Trα1PV mutant in WAT impairs adipogenesis via, in part, the transcriptional repression of the master regulator of adipogenesis, peroxisome proliferation-activated receptor γ (Pparγ; Ying et al. 2007). However, despite Trβ expression in WAT (Ying et al. 2007), intriguingly, in the TrβPV mice that harbor the identical PV mutation in the corresponding C-terminal position (Fig. 1; Kaneshige et al. 2000), no apparent changes in WAT are evident (Araki et al. 2009), suggesting mutant TR isoforms mediate distinct impairments in adipogenesis.
To dissect the molecular mechanism underlying the distinct phenotypic manifestation in WAT between Trα1PV and TrβPV mice, we used the 3T3-L1 cell line. This system, developed by Green & Kehinde (1975) in the 1970s, has long been used by investigators as a model cell line to study adipogenesis. To establish the cause–effect relationship, we stably expressed TRα1PV or TRβ1PV in 3T3-L1 cells and evaluated the alterations in adipogenesis as compared with parental cells that contained only the vector backbone (control cells). Clones with similar abundance of TRα1PV (denoted as L1-α1PV cells) or TRβ1PV (L1-β1PV cells) were used to elucidate the molecular basis of the distinct phenotypic expression in adipogenesis in WAT observed in Trα1PV and TrβPV mice (Araki et al. 2009). We found that the T3-induced adipogenesis was more severely impaired in L1-α1PV cells than in L1-β1PV cells. The expression of two master regulators, PPARγ and CCAAT/enhancer-binding protein α (C/EBPα), was found to be more repressed in L1-α1PV cells than in L1-β1PV cells. Consistent with the gene expression patterns, the expression of several key lipogenic enzymes was more severely inhibited by TRα1PV than by TRβ1PV. Further analysis indicated that the preferential repression of the C/ebpα gene by TRα1PV was mediated, at least in part, by more avid recruitment of TRα1PV to the promoter complex of the C/ebpα gene. The finding that TR-mediated adipogenesis is isoform dependent raised the possibility that isoform-specific TR ligands could be considered for therapeutic intervention for lipid abnormalities.
Materials and methods
3T3-L1 cells (ATCC #CL-173) were maintained in DMEM with 10% calf serum as described previously (Ying et al. 2007).
Generation of 3T3-L1 cell lines stably expressing TRα1PV or TRβ1PV by retrovirus transduction
Human TRα1PV and TRβ1PV cDNAs were cloned into pFH-IRESneo plasmid ((Martinez et al. 2001), a generous gift of R G Roeder, Rockefeller University, New York, NY, USA) to obtain FH-α1PV and FH-β1PV respectively as described previously (Ying et al. 2006). The pMSCVhygro-FH-α1PV or pMSCVhygro-FH-β1PV vector was prepared by inserting FH-α1PV or FH-β1PV cDNA from pFH-IRESneo-α1PV or pFH-IRESneo-β1PV respectively into pMSCVhygro (Clontech). All plasmids were verified by DNA sequencing.
EcoPack2-293 ecotropic retrovirus packaging cell lines (Clontech) were cultured in DMEM with 10% fetal bovine serum. Recombinant retroviral transduction of 3T3-L1 cells was performed as described (Ge et al. 2002). Briefly, EcoPack2-293 cells were seeded at a density of 3×106 cells per 10-cm dish. After 24 h, cells were transfected with 22 μg of retrovirus vector by Lipofectamine 2000 reagent (Invitrogen). Forty-eight hours after transfection, the cells were changed into fresh medium. After 12 h, virus-containing supernatant was collected, filtered through 0.45-μm membrane, and supplemented with 8 μg/ml polybrene (Sigma). 3T3-L1 cells were plated at a density of 2×105 cells per 10-cm dish 1 day before infection. Cells were incubated with retrovirus for 48 h and split at a density of 2×105 cells per 10-cm dish. Cells were selected with 150 μg/ml hygromycin (Invitrogen) for 14 days. The expression of TRα1PV or TRβ1PV protein was verified by western blot analysis using anti-FLAG (Sigma–Aldrich), anti-TRβ1 C4 (Bhat et al. 1995), or monoclonal anti-PV antibody (#302; Zhang et al. 2002).
Induction of adipocyte differentiation
Adipocyte differentiation was induced as described (Ying et al. 2007). Briefly, after the 3T3-L1 cells reached confluence, they were incubated with DMEM containing 10% resin-stripped calf serum for 48 h. The cells were treated with DMEM containing 10% resin-stripped fetal bovine serum with 1 μM dexamethasone (Sigma) and 0.5 mM isobutylmethylxanthine (IBMX) (Sigma) in the presence or absence of 2 nM T3 for 60 h. Thereafter, cells were subsequently fed every other day with DMEM containing 10% resin-stripped fetal bovine serum with or without 2 nM T3 until being stained with Oil Red O (Sigma) at 7 days after the initiation of stimulation. Quantification of the staining intensities was carried out by drying the stained plates, adding 0.8 ml isopropanol per 3.5-cm dish, and reading at OD510 against the control using a stained dish without cells.
Quantitative real-time RT-PCR
RNAs from control 3T3-L1, L1-α1PV, and L1-β1PV cells were extracted using an RNeasy mini kit according to the manufacturer's instructions (Qiagen). DNA was removed by using TURBO DNA-free kit (Ambion, Austin, TX, USA). The determination of mRNA by real-time reverse transcription-PCR (qPCR) was carried out using total RNA (10–50 ng) and primers for Pparγ, Lpl, and ap2 as described previously (Ying et al. 2007, Araki et al. 2009). The primer sequences for C/ebpα are forward TTACAACAGGCCAGGTTTCC and backward CTCTGGGATGGATCGATTGT.
Western blot analysis
Cells (control 3T3-L1, L1-α1PV, and L1-β1PV cells) were collected before induction or after induction of adipogenesis. For 100 μl of packed cell volume, 500 μl of lysis buffer (10 mM Tris, pH 7.9, 0.1 mM EDTA, 10 mM NaCl, 10 mM dithiothreitol, and protease inhibitors) were used to suspend the cells for 15 min on ice. Subsequently, NP40 was added to make a final concentration of 0.6% to lyse the cells by gentle vortexing. After centrifugation at 10 000 g for 30 s at 4 °C, the supernatant was removed and the pellet was extracted using an extraction buffer (20 mM HEPES/pH 7.9, 0.4 M NaCl, and 1 mM EDTA) by vortexing. The suspension was centrigued at 16 000 g for 10 min at 4 °C. In some experiments, the western blot analysis of lysates was done using cells that were harvested by scraping in buffer (1×PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM sodium phosphate dibasic, and 2 mM potassium phosphate monobasic, pH 7.4) and centrifuged at 2300 g at 4 °C for 4 min, and then the supernatant had been removed. The pelleted cells were incubated in a lysis buffer (50 mM Tris, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 1×Halt phosphatase inhibitor cocktail (Product no.: 78 420, ThermoFisher Scientific, Waltham, MA, USA), and 1× proteinase tablet cocktail inhibitors (Complete Mini EDTA-free; Roche)) for 10 min on ice, vortexed three times, and spun at 16 100 g at 4 °C for 10 min. Supernatant was collected for western blot analysis. The protein concentration of the supernatant was determined by the Bradford method (Pierce Chemical Co., Rockford, IL, USA) with BSA (Pierce Chemical Co.) as the standard. Western blot analysis was carried out as described previously (Ying et al. 2006). For the detection of PPARγ and C/EBPα proteins, nuclear fractions (35 μg) were separated by SDS-PAGE. The primary antibodies used in the western blot analysis were anti-PPARγ antibody (1:100 dilution; sc-7273; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) and anti-C/EBPα (1:100 dilution; sc-61; Santa Cruz Biotechnology, Inc).
Chromatin isolation from control 3T3-L1, L1-α1PV, and L1-β1PV cells was carried out by using a chromatin immunoprecipitation (ChIP) assay kit according to the manufacturer's instructions (Millipore, Billerica, MA, USA). Chromatin solution (1 ml) was immunoprecipitated with the following antibodies: 5 μl of anti-TRβ1 antibody (Rockland Immunochemicals, Gilbertsville, PA, USA code: 600-401-A38), anti-TRα1 antibody (1:300 dilution, Rockland Immunochemicals), monoclonal anti-PV antibody (#302; 4 μg (Zhang et al. 2002)), or IgG (2 μg from kit) for negative control, and anti-AcH3 (2 μg, from ChIP assay's kit) for positive control. The DNA recovered after immunoprecipitation was used as a template for amplification as described previously (Ying et al. 2007). Two percent of the chromatin solution (20 μl) was used for the input DNA as a control. The primer sequences for the analysis of the TR binding sites for C/ebpα were 5′-TAGAGAAGCTGGGCGAAAAGA-3′ (forward) and 5′-CCCACTTCCAGCCAACAGTA-3′ (reverse). The amplified DNA of 339 bp was analyzed on a 1.5% agarose gel with ethidium bromide staining.
All data are expressed as means±s.e.m. Statistical analysis was performed with ANOVA, and a P value <0.05 was considered significant.
Impairment of adipogenesis in 3T3-L1 cells by TR mutants is isoform dependent
The observation that adipogenesis of WAT is impaired in Trα1PV, but not in TrβPV, mice prompted us to use 3T3-L1 cells to further elucidate the molecular mechanisms underlying the differential effects of TR mutants on adipogenesis. 3T3-L1 cells have long been used by investigators as a model cell line to study adipogenesis. We, therefore, generated 3T3-L1 cells stably expressing TRβ1PV (L1-β1PV cells) or TRα1PV (L1-α1PV cells). We chose the clones that stably expressed a similar level of TR mutant isoforms for the study (Fig. 2). Western blot analysis showed that the abundance of TRα1PV protein in L1-α1PV cells (lane 1, Fig. 2A) was similar to that of TRβ1PV protein in L1-β1PV cells (lane 2, Fig. 2A). Lane 3 shows that no TR mutant was expressed in the control cells in which only the vector backbone was transfected (denoted as control cells). Figure 2B shows endogenous TRα1 and TRβ1 protein abundance in L1-β1PV, L1-α1PV, and control cells detected by western blot analysis. Consistent with that found in white adipocytes of wild-type mice (Ying et al. 2007), TRα1 was the major TR isoform detected in these three cell lines. GAPDH was used as loading control in Fig. 2A and B.
To assess whether TRβ1PV and TRα1PV differentially affected the adipogenesis in 3T3-L1 cells, we compared T3-enhanced adipogenesis in control cells with that in L1-β1PV and L1-α1PV cells (Fig. 3). T3 induced an increased adipogenesis in control cells, as indicated by an increased accumulation of lipid droplets stained with Oil Red O (compare panel 1 with panel 4, Fig. 3A). This T3-induced adipogenesis was reduced in L1-β1PV cells (compare panel 5 with panel 4, Fig. 3A) and L1-α1PV (compare panel 6 with panel 4, Fig. 3A). The Oil Red O staining intensities were measured, and the quantitative data in Fig. 3B indicate that T3 induced a 2.5-fold increase in adipogenesis in control cells (bar 2 versus bar 1, Fig. 3B). Importantly, Fig. 3B shows that TRα1PV was more potent (94% reduction) than TRβ1PV (54% reduction) in the inhibition of the T3-induced adipogenesis (compare bar 6 with bar 2 and bar 4 with bar 2; Fig. 3B). The reduction of T3-induced adipogenesis in L1-β1PV and L1-α1PV cells could also be visualized by phase contrast microscopy (Fig. 3C). The finding that TRα1PV was a stronger inhibitor (compare panel f with panel d, Fig. 3C) than TRβ1PV (compare panel e with panel d, Fig. 3C) was also apparent by phase contrast microscopy in that fewer lipid-containing cells (cells with dark intense color) were observed in panel f than in panel e (Fig. 3C). These results are consistent with the findings in vivo that impaired adipogenesis in WAT was detected in Trα1PV mice, but not in TrβPV mice (Ying et al. 2007, Araki et al. 2009).
Repression of key regulators of adipogenesis by TR mutants is isoform dependent
PPARγ is also a ligand-dependent transcription factor and a member of the nuclear receptor superfamily. Studies have shown that PPARγ is both necessary and sufficient for adipogenesis, and it is therefore known as a ‘master regulator’ of adipogenesis (Rosen & MacDougald 2006). To understand the underlying mechanisms by which the two mutant TR isoforms differentially impair adipogenesis in 3T3-L1 cells, we first examined the effect of T3 on the expression of PPARγ at the protein level before and after induction of adipogenesis (Fig. 4A). Before induction (day 0), very low PPARγ protein was observed under experimental conditions in control, L1-β1PV, or L1-αPV cells whether T3 was present or not (lanes 1, 2, 5, 6, 9, and 10, Fig. 4A). After induction, both PPARγ1 and PPARγ2 were induced in control cells (lanes 3 and 4, Fig. 4A). Importantly, their protein abundance was further increased by T3 (∼2-fold; lane 4). The expression of PPARγ1 and PPARγ2 proteins, however, was strongly decreased in L1-β1PV (lanes 7–8, Fig. 4A) and L1-α1PV cells (lanes 11–12, Fig. 4A). The protein abundance of PPARγ1 and PPARγ2 in L1-α1PV cells was lower than that in L1-β1PV cells (compare lanes 11–12 with lanes 7–8), indicating that TRα1PV inhibited the expression of PPARγ1 and PPARγ2 at the protein level more than TRβ1PV.
C/EBPα is the founding member of a family of basic region/leucine zipper (bZIP) transcription factors. It is required for the differentiation of 3T3-L1 preadipocytes; inhibition of C/EBPα blocks adipocyte differentiation, whereas overexpression of this C/EBP family member triggers adipocyte differentiation (Darlington et al. 1998, Obregon 2008). C/EBPα acts primarily to keep PPARγ levels elevated in the cell. Therefore, we also assessed the effect of T3 on the expression of C/EBPα at the protein level before and after the induction of adipogenesis (Fig. 4B). Before induction (day 0), the C/EBPα protein level was very low in control, L1-β1PV, or L1-α1PV cells (lanes 1, 2, 5, 6, 9, and 10, Fig. 4B). After induction, C/EBPα was induced in control cells (lanes 3 and 4, Fig. 4B). Its protein level was further increased by T3 (∼2-fold, lane 4, Fig. 4B). In L1-β1PV and L1-α1PV cells, markedly decreased signals were observed (lanes 7–8 and 11–12, Fig. 4B), indicating that mutations of TRs interfere with the expression of C/EBPα. However, a lower expression of C/EBPα protein was observed in L1-α1PV cells than in L1-β1PV cells (compare lanes 11–12 with lanes 7–8, Fig. 4B). Panel C shows the corresponding loading control using GAPDH. Taken together, these results demonstrate that T3 regulates the expression of PPARγ and C/EBPα, which are the two master key regulators of adipogenesis.
We further determined the mRNA levels of Pparγ and C/ebpα to ascertain whether the expression of these two key regulators is regulated by T3 and affected by TR mutants at the transcriptional level. Indeed, the expression of Pparγ and C/ebpα mRNAs was lower (∼25–30% reduction) in the absence of T3 than in the presence of T3 (lane 2; Fig. 5A and B respectively). Compared with control, the expression of Pparγ mRNA was 45% lower in L1-β1PV cells and 75% lower in L1-α1PV cells (bars 3 and 5 respectively, Fig. 5A). Consistently, the mRNA level of C/ebpα was 40 and 80% lower in L1-β1PV and L1-α1PV cells respectively compared with the control cells (bars 3 and 5 respectively, Fig. 5B). It is interesting to note that the expression of these two genes was less repressed at the mRNA levels than at the protein level by TR mutants (see Fig. 4). Such discordance could reflect additional post-translational regulation of PPARγ and C/EBPα. In L1-β1PV and L1-α1PV cells, T3 had no significant effect on the extent of PV mutant-mediated repression of Pparγ and C/ebpα mRNAs (Fig. 5A and B respectively). The functional consequences of the decreased expression of Pparγ were evident in that its direct target genes – the lipoprotein lipase (Lpl) and aP2, involved in the lipid synthesis – were concordantly repressed, with a higher extent of repression in L1-α1PV cells than in L1-β1PV cells (compare bars 3 and 5 to bar 1; Fig. 5C and D).
Differential occupancy of mutant TR isoforms on the C/ebpα gene promoter
To further understand how TR mutant isoforms differentially impaired adipogenesis, we focused on the study of the C/ebpα gene, an essential transcription factor in the induction of adipogenesis in 3T3-L1 cells (Darlington et al. 1998, Obregon 2008). The C/ebpα gene is a direct TR target gene with several TREs identified in the promoter of the C/ebpα gene (Menendez-Hurtado et al. 2000). The finding that TR mutant isoforms differentially repressed the expression of C/ebpα (Fig. 5A) prompted us to test the hypothesis that differential recruitment of mutant isoforms to the promoter could underlie differential impairment of adipogenesis by TR mutant isoforms. We, therefore, used ChIP to address this question (Fig. 6). In control cells, when antibodies recognizing specific TR isoforms were used, TRα1 and TRβ1 were found to be recruited to TREs on the promoter of the C/ebpα gene (lanes 5 and 6, Fig. 6). When the monoclonal antibody recognizing specifically the PV mutated sequences (#302) was used, only background noise (lane 7) similar to that observed in negative controls (lane 2, no antibody; lane 3, an irrelevant IgG) was seen. In contrast, in L1-β1PV cells, a strong signal was detected when anti-PV antibody was used, indicating the recruitment of β1PV to the TREs on the promoter of the C/ebpα gene (lane 14). A concomitant decreased recruitment of TRα1 (compare lane 12: 4.5±0.01 units and lane 5: 5.06±0.18 units) and TRβ1 (compare lane 13: 1.64±0.07 units and lane 6: 2.04±0.14 units) was observed, suggesting competition by TRβ1PV with TRα1 and TRβ1 for TREs on the promoter of the C/ebpα gene. The finding that TRα1PV also competed for wild-type TRs for binding to TREs was also demonstrated in L1-α1PV cells. A strong signal was detected when anti-PV antibody was used (lane 21: 4.94±0.07 units) with concomitant decreased intensities when anti-TR antibodies were used for TRα1 (compare lane 19: 1.93±0.05 units and lane 5: 5.06±0.18 units). Furthermore, it is important to point out that the signal intensity detected by anti-PV antibodies in L1-α1PV cells was stronger than that in L1-β1PV cells (compare lane 21: 4.94±0.07 units and lane 14: 4.06±0.17 units), indicating a more avid recruitment of TRα1PV to the promoter of the C/ebpα gene. These results suggest that a higher occupancy by TRα1PV than by TRβ1PV on the promoter of the C/ebpα gene, which, in part, could underlie a more severe impairment of adipogenesis by TRα1 mutations.
It has long been known that T3 plays a critical role in lipid metabolism by regulating genes involved in lipogenesis and lipolysis. The underlying mechanisms, however, have only begun to be unraveled in recent years. Genetically engineered mice are valuable tools to gain insights into the role of TRs in lipid metabolism and energy balance. Mice deficient in Trα1 exhibit lower body temperature (Wikstrom et al. 1998), indicative of the importance of Trα1 in thermogenesis. Mice deficient in Trβ display abnormalities in cholesterol metabolism (Gullberg et al. 2002), demonstrating that Trβ plays a major role in cholesterol homeostasis. Significant advances in the understanding of the roles of Trα1 in adipogenesis came from the mice with knock-in mutations in the Trα gene. Mice that are heterozygous for a Trα1P398H mutation are obese with increased visceral adiposity and hepatic steatosis (Liu et al. 2003). In adipose tissue, the norepinephrine-mediated lipolysis in WAT is blocked, and hormone-sensitive lipase in adipose tissue is decreased. In contrast to the Trα1P398H mutant mice, mice with a Trα1R384C mutation are lean and hypermetabolic with reduction in fat depots (Sjogren et al. 2007). Similar to the lean phenotype exhibited by Trα1R384C mutant mice, Trα1PV mice are dwarfs with reduced WAT mass across a variety of depots (Ying et al. 2007). TRα1PV acts to transcriptionally repress the expression of the Pparγ gene, resulting in impaired adipogenesis. Intriguingly, while Trβ is expressed in WAT (Ying et al. 2007), no apparent reduction in WAT mass was evident in TrβPV mice (Araki et al. 2009), indicating distinct phenotypic expression in the adipogenesis in WAT between these two mutant mice. In the present study, the creation of stable 3T3-L1 cell lines expressing a similar level of TRα1PV or TRβ1PV proteins allowed the dissecting of the differential role of TR isoforms in adipogenesis. Indeed, TRα1PV was a stronger repressor than TRβ1PV in blocking T3-mediated adipogenesis in 3T3-L1 cells by inhibiting the expression of two master regulators, PPARγ and C/EBPα. The stronger repression by TRα1PV was due, in part, to a stronger recruitment of TRα1PV than of TRβ1PV to the promoter of the C/ebpα gene (Fig. 6). As a result, the expression of the C/ebpα gene was repressed, leading to more impaired adipogenesis in L1-α1PV cells than in L1-β1PV cells (Fig. 3). Thus, the present study has provided direct evidence to link the mutations of the TRα gene to the impaired adipogenesis in WAT.
Currently, it is not clear how TRα1PV was preferentially recruited to the promoter complex of the C/ebpα gene. TRα1PV and TRβ1PV differ mainly in the amino-terminal A/B domain (see Fig. 1). Thus, it is reasonable to speculate that the amino-terminal A/B domain of TRα1PV could affect the folding of the tertiary structure such that TRα1PV could bind to TREs stronger than TRβ1PV on the promoter of the C/ebpα gene. Alternatively, the amino-terminal A/B domain could facilitate the formation of a favorable conformation for TRα1PV to interact more favorably with other coregulators (e.g. corepressors) to stabilize the TRα1PV–DNA complex on the C/ebpα gene promoter. These two possibilities are not mutually exclusive and could, in fact, act in concordance to strengthen the interaction of TRα1PV with the promoter to block the expression of the C/ebpα gene, thereby impairing adipogenesis in 3T3-L1 cells.
The finding that TRα1PV had a more deleterious effect than TRβ1PV on the adipogenesis in 3T3-L1 cells provided new insights to understand the observation that WAT mass is reduced in Trα1PV mice, but not in TrβPV mice compared with the wild-type mice (Araki et al. 2009). Trα1 is a major TR isoform in WAT; therefore, mutation of a major TR isoform (i.e. Trα1) results in the loss of function critical for adipogenesis. In TrβPV mice, mutation of a minor TR isoform (i.e. TRβ1) cannot completely block the effect of TRα1, the major TR in WAT. Furthermore, TRβ1PV, as shown in the present study, was not as strong as TRα1 in interacting with TRE on the promoter of the C/ebpα gene promoter, thereby making TRβ1PV less effective in blocking adipogenesis in WAT in TrβPV mice. However, besides the preferential interaction of TRα1PV with the promoter of the C/ebpα gene, other mechanisms, yet to be uncovered, could also contribute to the more severe impairment in adipogenesis by TRα1PV.
As obesity and atherosclerosis have become major health concerns in Western countries, there is an increasing need to develop novel treatment strategies. Thyroid hormone mimetics that have specificity toward TRα1 or TRβ1 are being developed to derive benefits in treating metabolic diseases but without undesirable effects on other target organs (Baxter & Webb 2009). The finding that TRα1PV was more effective than TRβ1PV in blocking adipogenesis suggests that TRα1 could be considered as a potential therapeutic target for decreasing adipose tissue and reducing serum fatty acids. Moreover, 3T3-L1 cells stably expressing TRα1PV or TRβ1PV could be used as model cell lines to further elucidate the role of T3 via TR in adipogenesis.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
The present work was supported by the Intramural Research Program for Cancer Research, National Cancer Institute, and National Institutes of Health.
We thank Dr R G Roeder for the generous gift of pFH-IRESneo plasmid and Dr O Araki for his contributions in the early phase of the present study.
Araki O, Ying H, Zhu XG, Willingham MC & Cheng SY 2009 Distinct dysregulation of lipid metabolism by unliganded thyroid hormone receptor isoforms. Molecular Endocrinology 23 308–315.
Baxter JD & Webb P 2009 Thyroid hormone mimetics: potential applications in atherosclerosis, obesity and type 2 diabetes. Nature Reviews. Drug Discovery 8 308–320.
Bhat MK, McPhie P, Ting YT, Zhu XG & Cheng SY 1995 Structure of the carboxy-terminal region of thyroid hormone nuclear receptors and its possible role in hormone-dependent intermolecular interactions. Biochemistry 34 10591–10599.
Cheng SY 2000 Multiple mechanisms for regulation of the transcriptional activity of thyroid hormone receptors. Reviews in Endocrine and Metabolic Disorders 1 9–18.
Ge K, Guermah M, Yuan CX, Ito M, Wallberg AE, Spiegelman BM & Roeder RG 2002 Transcription coactivator TRAP220 is required for PPAR gamma 2-stimulated adipogenesis. Nature 417 563–567.
Golozoubova V, Gullberg H, Matthias A, Cannon B, Vennstrom B & Nedergaard J 2004 Depressed thermogenesis but competent brown adipose tissue recruitment in mice devoid of all hormone-binding thyroid hormone receptors. Molecular Endocrinology 18 384–401.
Green H & Kehinde O 1975 An established preadipose cell line and its differentiation in culture. II. Factors affecting the adipose conversion. Cell 5 19–27.
Gullberg H, Rudling M, Salto C, Forrest D, Angelin B & Vennstrom B 2002 Requirement for thyroid hormone receptor beta in T3 regulation of cholesterol metabolism in mice. Molecular Endocrinology 16 1767–1777.
Kaneshige M, Kaneshige K, Zhu X, Dace A, Garrett L, Carter TA, Kazlauskaite R, Pankratz DG, Wynshaw-Boris A & Refetoff S 2000 Mice with a targeted mutation in the thyroid hormone beta receptor gene exhibit impaired growth and resistance to thyroid hormone. PNAS 97 13209–13214.
Kaneshige M, Suzuki H, Kaneshige K, Cheng J, Wimbrow H, Barlow C, Willingham MC & Cheng S 2001 A targeted dominant negative mutation of the thyroid hormone alpha 1 receptor causes increased mortality, infertility, and dwarfism in mice. PNAS 98 15095–15100.
Liu YY, Schultz JJ & Brent GA 2003 A thyroid hormone receptor alpha gene mutation (P398H) is associated with visceral adiposity and impaired catecholamine-stimulated lipolysis in mice. Journal of Biological Chemistry 278 38913–38920.
Martinez E, Palhan VB, Tjernberg A, Lymar ES, Gamper AM, Kundu TK, Chait BT & Roeder RG 2001 Human STAGA complex is a chromatin-acetylating transcription coactivator that interacts with pre-mRNA splicing and DNA damage-binding factors in vivo. Molecular and Cellular Biology 21 6782–6795.
Menendez-Hurtado A, Santos A & Perez-Castillo A 2000 Characterization of the promoter region of the rat CCAAT/enhancer-binding protein alpha gene and regulation by thyroid hormone in rat immortalized brown adipocytes. Endocrinology 141 4164–4170.
O'Malley BW 2003 Sequentiality and processivity of nuclear receptor coregulators in regulation of target gene expression. Nuclear Receptor Signaling 1 e010.
Sjogren M, Alkemade A, Mittag J, Nordstrom K, Katz A, Rozell B, Westerblad H, Arner A & Vennstrom B 2007 Hypermetabolism in mice caused by the central action of an unliganded thyroid hormone receptor alpha1. EMBO Journal 26 4535–4545.
Tinnikov A, Nordstrom K, Thoren P, Kindblom JM, Malin S, Rozell B, Adams M, Rajanayagam O, Pettersson S & Ohlsson C 2002 Retardation of post-natal development caused by a negatively acting thyroid hormone receptor alpha1. EMBO Journal 21 5079–5087.
Wikstrom L, Johansson C, Salto C, Barlow C, Campos Barros A, Baas F, Forrest D, Thoren P & Vennstrom B 1998 Abnormal heart rate and body temperature in mice lacking thyroid hormone receptor alpha 1. EMBO Journal 17 455–461.
Ying H, Furuya F, Zhao L, Araki O, West BL, Hanover JA, Willingham MC & Cheng SY 2006 Aberrant accumulation of PTTG1 induced by a mutated thyroid hormone beta receptor inhibits mitotic progression. Journal of Clinical Investigation 116 2972–2984.
Ying H, Araki O, Furuya F, Kato Y & Cheng SY 2007 Impaired adipogenesis caused by a mutated thyroid hormone alpha1 receptor. Molecular and Cellular Biology 27 2359–2371.
Zhang XY, Kaneshige M, Kamiya Y, Kaneshige K, McPhie P & Cheng SY 2002 Differential expression of thyroid hormone receptor isoforms dictates the dominant negative activity of mutant β receptor. Molecular Endocrinology 16 2077–2092.