View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 587 (2013) 1644–1649

journal homepage: www.FEBSLetters.org

X-box binding 1 enhances adipogenic differentiation of 3T3-L1 cells through the downregulation of Wnt10b expression ⇑ Yoon Mi Cho, Dae Hun Kim, Su-Nam Kwak, Seong-Whan Jeong, Oh-Joo Kwon

Department of Biochemistry, The Catholic University of Korea, Banpo-dong 505, Socho-gu, Seoul 137-701, Republic of Korea

article info abstract

Article history: Differentiation of preadipocytes into adipocytes is controlled by various transcription factors. Received 29 January 2013 Recently, the pro-adipogenic function of XBP1, a upregulated by endoplasmic Revised 1 April 2013 reticulum stress, has been reported. In this study, we demonstrated that XBP1 suppresses the Accepted 8 April 2013 expression of Wnt10b, an anti-adipogenic Wnt, during the differentiation of 3T3-L1 preadipocytes. Available online 19 April 2013 The expression pattern of XBP1 was reciprocal to that of Wnt10b during the early stage of adipogen- Edited by Lukas Huber esis. The intracellular protein levels of b-catenin were negatively regulated by XBP1. Direct binding of XBP1 to the Wnt10b and the subsequent decrease of the b-catenin signalling pathway represent a novel adipogenic differentiation mechanism. Keywords: XBP1 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Adipogenesis Wnt10b Beta-catenin Transcription

1. Introduction hancer-binding protein a (C/EBPa) in 3T3-L1 cells [8]. In contrast, Han et al. reported that IRE1a did not alter adipocyte differentiation Adipose tissue is a primary metabolic organ that is important in 3T3-L1 cells [9]. for energy homeostasis and insulin sensitivity, and the excess Wnts are secreted glycoproteins that activate the b-catenin- deposition of adipose tissue is an established risk factor for many mediated signalling pathway, which controls a myriad of biological pathologic conditions [1]. Adipogenesis, the molecular events processes [10]. In preadipocytes, Wnt/b-catenin signalling main- underlying the expansion of adipose tissue, is the process by which tains the cells in an undifferentiated state primarily by suppressing fibroblast-like preadipocytes differentiate into adipocytes. adipogenic transcription factors C/EBPa and peroxisome prolifera- X-box binding protein 1 (XBP1) is an important transcription tor-activated c (PPARc). Disruption of Wnt signalling also factor of the unfolded protein response (UPR). During the UPR, causes myoblasts to transdifferentiate into adipocytes in vitro, XBP1 mRNA is converted to its active form, XBP1s, and it then in- indicating the importance of Wnt/b-catenin signalling in adipocyte creases the expression of UPR-associated involved in protein differentiation [11,12]. Of the 19 Wnt ligands found in humans and folding and biogenesis [2]. XBP1 also plays mice, Wnt10b is a major Wnt ligand in preadipocytes and its levels various roles in many UPR-unrelated biological processes. XBP1 in- decline rapidly after the induction of differentiation [11], and the duces the differentiation of many cells, e.g., plasma cells and hepa- overexpression of Wnt10b in 3T3-L1 cells stabilizes b-catenin tocytes [3,4], and it also plays a crucial role in glucose metabolism, and blocks adipogenesis, suggesting it is a molecular switch that inflammation, and insulin function [5]. Recently, a few reports dem- governs adipogenesis [13]. The association of Wnt10b with obesity onstrated the role of XBP1 in the regulation of adipogenesis. For has been also suggested [14]. example, XBP1 is induced during adipogenesis [6,7], and it plays a Although the signalling pathways downstream of Wnt10b that key role in adipocyte differentiation by the induction of CCAAT/en- regulate adipogenesis are being defined, little is known about the mechanisms involved in the control of Wnt10b expression. Re- cently, Chung et al. reported that C/EBPb downregulates Wnt10b Abbreviations: ER, endoplasmic reticulum; UPR, unfolded protein response; expression during adipogenesis [15]. Because XBP1 was strongly XBP1, X-box binding protein 1; Wnt10b, wingless-related MMTV integration site induced by C/EBPb in 3T3-L1 cells [8], here, we examined whether 10b ⇑ XBP1 acts as an upstream regulator of Wnt10b during Corresponding author. Fax: +82 2 596 4435. adipogenesis. E-mail address: [email protected] (O.-J. Kwon).

0014-5793/$36.00 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2013.04.005 Y.M. Cho et al. / FEBS Letters 587 (2013) 1644–1649 1645

Table 1 Primers used for real-time PCR.

Name Primer Sequences (50–30) Size (bp) ID Sense Antisense 36B4 GCTCCAAGCAGATGCAGCA CCGGATGTGAGGCAGCAG 143 NM_007475 mAdiponectin TGTGTATCGCTCAGCGTTCAGTGT AGAGAACGGCCTTGTCCTTCTTGA 224 NM_007678 mC/EBPa AGCTTACAACAGGCCAGGTTTCCT TTCAAGTTCCTTCAGCAACAGCGG 213 NM_007678 mFzd2 TCATCTTTCTGTCCGGCTGCTACA AGCTGGCCATGCTGAAGAAGTAGA 186 NM_020510 mLRP5 TGAAGGAATGGCTGTGGACTGGAT TGTCAAGGTCTCTCCACACAAGCA 125 NM_008513 mWnt1 GGTTTCTACTACGTTGCTACTGG GGAATCCGTCAACAGGTTCGT 121 NM_021279 mWnt10b TGGCTGTAACCACGACATGGACTT CTGACGTTCCATGGCATTTGCACT 170 NM_011718 mWnt4 CTCAAAGGCCTGATCCAGAG TCACAGCCACACTTCTCCAG 281 NM_009523 mWnt5a ACTGGCAGGACTTTCTCAAGGACA GCCTATTTGCATCACCCTGCCAAA 181 NM_009524 mWnt5b CCCCAGGCCAGAGAAAGC CCTCCCCGATGTAGGACAT 60 NM_009525 mWnt6 TGTCAGTTCCAGTTCCGTTTCCGA ACACGAAAGCTGTCTCTCGGATGT 100 NM_009526 mXBP1s CTGAGGTCCGCAGCAGGT TGTCAGAGTCCATGGGAAGA 66 NM_013842 mXBP1t AGCAGCAAGTGGTGGATTTGGAAG AAGAGGCAACAGTGTCAGAGTCCA 299 NM_013842 mb-catenin TGCAGCTTCTGGGTTCCGATGATA AGATGGCAGGCTCAGTGATGTCTT 180 NM_001165902

2. Materials and methods 7300 thermal cycler (Applied Biosystems, Foster City, CA, USA) with SYBR Green Q-PCR Master Mix (Takara). The relative amount 2.1. Cell culture and adipocytes differentiation of mRNAs was calculated using the DDCt method with the 36B4 mRNA as an internal control. The primer sequences for real-time The murine 3T3-L1 preadipocytes (American Type Culture Col- PCR are listed in Table 1. lection, Manassas, VA, USA) were maintained in high-glucose (400 mg/dl) DMEM containing 10% newborn calf serum (Gibco, 2.5. Preparation of cytoplasmic fraction Green Island, NY, USA). Two days after reaching confluence (day 0), the cells were subjected to adipogenic differentiation for 2 days Cells were washed in PBS and scraped in 10 mM Tris–HCl (pH in the medium containing 3-isobutyl-1-methylxanthine (0.5 mM, 7.4) containing 140 mM NaCl, 5 mM EDTA, 2 mM dithiothreitol, Sigma, St. Louis, MO, USA), dexamethasone (1 lM, Sigma), and 0.5 mM PMSF, 2 lg/ml aprotinin, and 1 lg/ml leupeptin. Cells were insulin (167 nM, Sigma). Then the cells were cultured for 2 days lysed by 30 strokes in Potter–Elvehjem homogenizer (Wheaton in the DMEM with 10% FBS and insulin (167 nM), and thereafter Scientific, Millville, NJ, USA) at 4 °C. The lysates were centrifuged the medium was changed to the DMEM/10% FBS until day 8. at 500g for 10 min, and the cleared lysates were subject to ultra- centrifugation at 100000g for 90 min at 4 °C. The supernatants 2.2. Genetic manipulation of XBP1 were collected as the cytoplasmic fraction.

XBP1 knockdown cells were prepared using retroviral vectors 2.6. Immunoblot analysis expressing shRNA targeting mouse XBP1 (SFGneo-iXBP1, from Prof. Glimcher, Cornell Medical College, New York, NY, USA). Viral Cells were lysed in RIPA buffer and spun for 30 min at 16000g. supernatant was prepared as described [16] and was used to trans- Thirty micrograms of protein were separated by 10% SDS–PAGE duce. Briefly, 293T cells were transfected with the viral vectors and transferred onto a nitrocellulose membrane (Schleicher & using Lipofectamin 2000 (Invitrogen, Carlsbad, CA, USA). After Schell, Dassel, Germany). The membranes were blocked with 5% 48 h, virus-containing medium was collected and passed through non-fat milk and incubated overnight at 4 °C with the mouse anti- a 0.22-lm filter. For infection, filter-sterilized Polybrene (hexadi- sera against adiponectin (1:1000; Alexis, Farmingdale, NY, USA), b- methrine bromide; 8 lg/ml, Sigma) was added to the virus-loaded catenin (1:500; BD Bioscience, San Jose, CA, USA) and b-actin medium and applied to proliferating cells for 2 days. Uninfected (1:500; Sigma), and rabbit antisera against XBP1 (1:200; Biovision, cells were removed by culturing in the presence of G418 (1 mg/ Mountain View, CA, USA) and C/EBPa (1:500; Santa Cruz, Santa ml). To make XBP1s-overexpressing cells, recombinant plasmids Cruz, CA, USA). binding was detected using horseradish expressing mouse XBP1s (pcDNA3.1-mXBP1s, from Prof. Glimcher) peroxidase-conjugated anti-rabbit or anti-mouse IgG (1:2,000; Sig- were transfected using Lipofectamin 2000, and selected with G418. ma) and the immunoreactive bands were visualized with ECL method (Amersham Pharmacia Biotech, Little Chalfont, UK). In 2.3. Oil-red O staining some experiments, the density of specific bands was measured and normalized with the control band. Cells were washed twice with PBS, fixed with 10% formalin in distilled water (DW) for 10 min. After being rinsed with DW, cells 2.7. Promoter assay were incubated for 10 min with 0.3% of Oil-red O (Sigma). The monolayer was then extensively washed with DW, and stained 3T3-L1 preadipocytes were seeded into 24-well plates at a den- cells were mounted for microscopic examination (Leica micro- sity of 2 104/well. Reporter plasmids (100 ng) were transfected scope, Wetzlar, Germany). with effector or empty vectors (0.5 or 1 lg). To normalize transfec- tion efficiency, the Renilla luciferase-expressing plasmid (pRL-TK- 2.4. RNA isolation and real-time PCR luc; 10 ng) was cotransfected as internal control. After 48 h, cells were lysed in passive lysis buffer and the luciferase activities of Total RNA was extracted from the cultures using Trizol Reagent both reporters were measured using a dual luciferase assay kit (Invitrogen), and cDNA was synthesized using PrimeScript™ RT kit (Promega, Madison, WI, USA). For Wnt10b promoter assay, we (Takara, Shiga, Japan). The quantitative real-time PCR analysis was used mouse Wnt10b promoter construct [mWnt10b (2675)- performed on at least three independent RNA samples using ABI Luc] which carry a DNA fragment of Wnt10b (2675 to 135 bp) 1646 Y.M. Cho et al. / FEBS Letters 587 (2013) 1644–1649 in pGL2-enhancer vector (a gift from Prof. KS Park, Seoul National Thus, we hypothesize that XBP1s may suppress b-catenin signal- University, Seoul, Korea). For monitoring of Wnt/b-catenin signal- ling through the inhibition of extracellular Wnt ligands. ling pathway, we used Super TOPFlash (Addgene Inc., Cambridge, MA, USA) encoding seven copies of LEF/TCF binding sites linked 3.3. XBP1s inhibits Wnt10b expression to firefly luciferase. Super FOPFlash (Addgene) which has mutated LEF/TCF binding sites was used as negative control. Wnts bind to their cell-surface receptors and allow b-catenin to escape from the destruction complex, and consequently accumu- 2.8. Chromatin immunoprecipitation (ChIP) assay late and translocate to the nucleus. To identify which Wnts were regulated by XBP1s, we examined the mRNA levels of 3 typical ChIP assay was performed according to the supplier’s protocol anti-adipogenic (Wnt1, 10b, and 6) and 3 pro-adipogenic (Wnt4, (Millipore, Billerica, MA, USA). Briefly, cells were differentiated 5a, and 5b) Wnts in XBP1-KD and control cells. Fig. 3A shows that for the indicated times and crosslinked with formaldehyde, and only Wnt10b, the most potent anti-adipogenic Wnt [11], was sig- resuspended and sonicated to generate 200–1000 bp DNA frag- nificantly increased by up to 2-fold following XBP1 knockdown. ments. The fragmented chromatin samples were immunoprecipi- Wnt10b is highly expressed in preadipocytes, and during the tated with rabbit antisera against mouse XBP1 (5 lg, M-186X; early stage of adipogenic differentiation, the so-called critical per- Santa Cruz), and were reverse-cross-linked, purified, and analyzed iod of differentiation, a rapid decrease of Wnt10b levels is required by real-time PCR. Primers used for the amplicon (525 to 316 bp) for adipogenesis [18]. In our study, Wnt10b transcription declined are: CCTAAGGTCTAGATCCGTCCT (forward), and GATCACGATT immediately after hormonal induction and reached its lowest level AGCCCACTCCA (reverse). DNA samples immunoprecipitated with rabbit IgG served as a negative control.

2.9. Statistical analysis

All values were expressed as mean ± S.E.M. from at least three independent experiments. Data was analyzed using one- or two- way ANOVA followed by Tukey’s Multiple Comparison Test. Statis- tical analyses were performed using Graphpad Prism 5 for Win- dows (Graphpad Prism Software Inc., La Jolla, CA, USA). In all cases, statistical significance was set at P<0.05.

3. Results

3.1. XBP1s is a critical factor for 3T3-L1 adipogenesis

During the differentiation of 3T3-L1 preadipocytes, transcrip- tion of both the total and spliced forms of XBP1 was increased in a time-dependent manner (Fig. 1A). To examine the role of XBP1s in adipogenesis, we prepared 3T3-L1 cells in which the expression of XBP1 was knocked down (XBP1-KD), and adipogenic differenti- ation was induced for 8 days. When we compared the degree of adipocyte development using Oil Red O staining between control and XBP1-KD cells at post-induction day 8, it was markedly re- duced in XBP1-KD cells (Fig. 1B). Moreover, mRNA and protein expression of the adipocyte markers C/EBPa and adiponectin was significantly decreased in XBP1-KD cells (Fig. 1C). These results strongly suggest that XBP1s is necessary for the adipogenesis of 3T3-L1 cells.

3.2. XBP1s suppresses intracellular b-catenin levels

To investigate the molecular mechanism for the stimulation of adipogenesis by XBP1s, we examined the Wnt/b-catenin signalling pathway, which is a well-known regulatory mechanism of adipo- genesis. First, we examined the cellular levels of b-catenin, a cen- tral player in the Wnt/b-catenin signalling pathway. Consistent with prior reports [17], the protein levels of b-catenin were signif- icantly and time-dependently decreased immediately after hor- monal induction in wild-type 3T3-L1 cells, whereas it was not Fig. 1. XBP1 enhances adipogenic differentiation of 3T3-L1 cells. Wild-type (CON) or XBP1 knockdown (XBP1-KD) 3T3-L1 preadipocytes were differentiated for altered in XBP1-KD cells (Fig. 2A, left). When we normalized the 8 days, and total RNAs and were prepared at the indicated times. (A) Real- immunoreactive bands using densitometry (Fig. 2A, right), b-cate- time PCR analysis of total (XBP1t) and spliced form (XBP1s) of XBP1 mRNA in wild- nin levels were dramatically decreased from day 1 in wild-type type 3T3-L1 cells (⁄p < 0.05, compared to day 0). (B) Oil-red O staining at 8 days cells; however, they were significantly higher in XBP1-KD cells at (magnification 100). The images shown are representative of 3 experiments. (C) Real-time PCR (upper) and immunoblot (lower) analyses of C/EBPa and adiponec- all times tested. Moreover, the forced expression of XBP1s in tin. Beta-actin was used as loading control. In data from real-time PCR, mRNA level 3T3-L1 cells significantly reduced b-catenin levels (Fig. 2B). How- at day 0 of 3T3-L1 cells was set to 1, and the other values were normalized and ever, b-catenin transcription was not affected by XBP1 (Fig. 2C). represented as the mean ± S.E.M. of 3 independent experiments (⁄P < 0.05, com- pared to the CON). Y.M. Cho et al. / FEBS Letters 587 (2013) 1644–1649 1647

Fig. 2. XBP1 suppresses intracellular b-catenin level during hormonal induction of adipogenesis. (A) Beta-catenin levels in the cytoplasmic fraction of wild-type (CON) and XBP1-KD (KD) cells were measured using immunoblot analysis (left), and the results were expressed in b-catenin/b-actin ratio (right). The ratio at day 0 of CON was set to 1 (⁄P < 0.05, compared to the CON). (B) Immunoblot analysis of b-catenin in the cytoplasmic fraction of 3T3-L1 cells transfected with empty vector (EV) or expression vector carrying XBP1s (XBP1s). The ratio at day 2 of CON was set to 1. Beta-actin was used as loading control (⁄P < 0.05, compared to the EV). (C) Real-time PCR analysis of b-catenin in CON and KD cells. The mRNA level at day 0 of CON was set to 1. Data represent the mean ± S.E.M. of three independent experiments.

Fig. 3. XBP1 inhibits the expression of Wnt10b. (A) The mRNA levels of anti-adipogenic (1, 10b, and 6) or pro-adipogenic (4, 5a, and 5b) isoforms of Wnt were measured in undifferentiated wild-type (CON) or XBP1-KD (KD) 3T3-L1 cells. (B) The mRNA level of Wnt10b was measured during the early adipogenesis in CON and KD cells. (C) Plotting of mRNA levels of Wnt10b (d) and XBP1s (s) for 2 days after differentiation induction. All data represent the mean ± S.E.M. of three independent experiments. (⁄P < 0.05, compared to the CON; #P<0.05, compared to the 0 h of CON). at 6 h in control cells (Fig. 3B). In XBP1-KD cells, Wnt10b mRNA (Fig. 4A). When we cotransfected the XBP1s expression vector levels were significantly higher before induction and at 3 h after and Wnt10b promoter reporter into 3T3-L1 cells, Wnt10b pro- (Fig. 3B). To examine the interrelationship between XBP1s and moter activity was significantly decreased by 50%, indicating the Wnt10b, we plotted their mRNA levels for 2 days after induction, regulation of Wnt10b transcriptional activity by XBP1s (Fig. 4B). and found that their expression patterns were the opposite of each ChIP analysis also revealed that XBP1s directly bound to the other (Fig. 3C). At 6 h, the mRNA levels of XBP1s reached a peak, Wnt10b promoter before and at 6 and 12 h after induction whilst Wnt10b declined to its lowest level. (Fig. 4C). Because the effects of Wnt are mediated through cell sur- face Wnt receptors, we examined whether XBP1s affects the 3.4. Wnt10b is a direct target of XBP1s expression of Frizzled (Fzd) and low-density lipoprotein-related protein 5 (LPR5) during adipogenesis. Corresponding to previous Direct regulation of Wnt10b expression by XBP1s was exam- studies [19], the mRNA levels of Fzd declined rapidly from 3 h, ined using promoter and chromatin immunoprecipitation (ChIP) whilst those of LRP5 were steady and even increased after 24 h analyses. Promoter assay was performed using a 2.5-Kb genomic in wild-type 3T3-L1 cells; however, the expression patterns of fragment of Wnt10b containing putative XBP1 binding sites these Wnt receptors were not affected by XBP1s during the first 1648 Y.M. Cho et al. / FEBS Letters 587 (2013) 1644–1649

Fig. 4. XBP1 suppresses Wnt10b expression by directly binding to the promoter. (A) Schematic representation of mouse Wnt10b promoter construct used for promoter assays. (B) Promoter assay was performed in 3T3-L1 cells, which were cotransfected with pGL2-Wnt10b reporter vector, and empty vector (EV) or full-length XBP1s- expressing plasmids (XBP1s). Luciferase activity was assessed 48 h after the transfection (⁄P < 0.05, compared to the EV) (C) ChIP analysis. The precipitated chromatin was analyzed by real-time PCR using primers spanning the putative XBP1s binding site within the Wnt10b promoter (–525 to –316 bp) (⁄P < 0.05, compared to the IgG). (D) Real- time PCR analysis of Wnt receptors Frizzled (Fzd) and LRP5. All data represent the mean ± S.E.M. of three independent experiments.

Fig. 5. XBP1 downregulates the transcriptional activity of the b-catenin. (A) 3T3-L1 preadipocytes were cotransfected with Super TOPFlash, and empty vector (EV) or full- length XBP1s-expressing plasmids (XBP1s) for 48 h. Super FOPFlash was used as negative control. Data represent the mean ± S.E.M. of three independent experiments (⁄P < 0.05, compared to the EV). (B) Diagram of XBP1s-mediated Wnt10b regulation during adipogenic differentiation.

2 days (Fig. 4D). These findings strongly suggest that XBP1s may promoter activity by approximately 50% (Fig. 5A, left). Super FOP- induce adipogenesis through the transcriptional repression of Flash activity was not afftected by XBP1s (Fig. 5A, right). Thus, we Wnt10b and the subsequent inhibition of the anti-adipogenic b- conclude that XBP1s may regulate b-catenin-induced transcrip- catenin signalling pathway. tional activity by direct repression of Wnt10b expression.

3.5. XBP1s antagonizes the b-catenin-TCF/LEF-mediated signalling 4. Discussion pathway XBP1s, a key player in UPR activation, is upregulated during Upon Wnt ligand binding, b-catenin binds directly to T cell fac- lipogenesis [21], and UPR activation contributes to the adipogene- tor/lymphoid enhancer factor (TCF/LEF) and stimulates the tran- sis of 3T3-L1 cells [6]. Anti-adipogenic Wnt signalling should also scription of a variety of Wnt-responsive genes [20]. To explore be inhibited for the initiation of adipocyte differentiation [15]. the possibility that XBP1s represses Wnt10b/b-catenin-mediated On the basis of these findings, we examined the involvement of transcriptional activity, we performed a reporter gene assay using XBP1s in Wnt/b-catenin signalling and provided evidence for the a promoter construct containing multiple TCF/LEF binding sites. first time that XBP1s induces adipogenesis by antagonizing the b- When XBP1s was co-expressed with the Super TOPFlash reporter catenin-mediated anti-adipogenic signalling pathway during the construct for 48 h, XBP1s significantly downregulated TCF/LEF early stage of adipogenic differentiation of 3T3-L1 cells. XBP1s Y.M. Cho et al. / FEBS Letters 587 (2013) 1644–1649 1649 directly and specifically inhibited Wnt10b transcription, the most References potent anti-adipogenic Wnt ligand, without affecting the expres- sion of b-catenin or the Wnt receptors Fzd and LPR5. When com- [1] Hotamisligil, G.S. (2006) Inflammation and metabolic disorders. Nature 444, 860–867. bined with a recent report showing the transcriptional repression [2] Sriburi, R., Jackowski, S., Mori, K. and Brewer, J.W. (2004) XBP1: a link between of Wnt10b by C/EBPb [15], our results suggest a new model for the unfolded protein response, lipid biosynthesis, and biogenesis of the the initiation of adipogenesis (Fig. 5B). Both C/EBPb and XBP1s endoplasmic reticulum. J. Cell Biol. 167, 35–41. are upregulated in response to adipogenic induction, and C/EBPb [3] Iwakoshi, N.N., Lee, A.H., Vallabhajosyula, P., Otipoby, K.L., Rajewsky, K. and Glimcher, L.H. (2003) differentiation and the unfolded protein transactivates XBP1 transcription [10]. C/EBPb induces two pro- response intersect at the transcription factor XBP-1. Nat. Immunol. 4, 321– adipogenic transcription factors PPARc and C/EBPa, and XBP1s also 329. increases C/EBPa [8]. Simultaneously with this pro-adipogenic [4] Reimold, A.M. et al. (2000) An essential role in liver development for transcription factor XBP-1. Genes Dev. 14, 152–157. activity, C/EBPb and XBP1s also inhibit anti-adipogenic b-catenin [5] Sha, H., He, Y., Yang, L. and Qi, L. (2011) Stressed out about obesity: IRE1alpha- signaling by the suppression of Wnt10b expression, therefore XBP1 in metabolic disorders. Trends Endocrinol. Metab. 22, 374–381. resulting in the potentiation of adipogenesis. C/EBPb also inhibits [6] Basseri, S., Lhotak, S., Sharma, A.M. and Austin, R.C. (2009) The chemical chaperone 4-phenylbutyrate inhibits adipogenesis by modulating the b-catenin signalling by the direct interaction between b-catenin unfolded protein response. J. Lipid Res. 50, 2486–2501. and PPARc, which was induced by C/EBPb [26]. In addition, [7] Todorcevic, M., Skugor, S., Krasnov, A. and Ruyter, B. (2010) Wnt10b/b-catenin prevents preadipocytes from adipogenic differ- profiles in Atlantic salmon adipose-derived stromo-vascular fraction during differentiation into adipocytes. BMC Genomics 11, 39. entiation by decreasing the expression of PPARc and C/EBPa. Taken [8] Sha, H. et al. (2009) The IRE1alpha-XBP1 pathway of the unfolded protein together, the adipogenic program is initiated and intensified syner- response is required for adipogenesis. Cell Metab. 9, 556–564. gistically by two dual-function transcription factors, i.e., C/EBPb [9] Han, J., Murthy, R., Wood, B., Song, B., Wang, S., Sun, B., Malhi, H. and Kaufman, R.J. (2013) ER stress signalling through eIF2alpha and CHOP, but not and XBP1s. IRE1alpha, attenuates adipogenesis in mice. Diabetologia 56, 911–924. An interesting but unanswered question is how XBP1s re- [10] Clevers, H. and Nusse, R. (2012) Wnt/beta-catenin signaling and disease. Cell presses the transcription of Wnt10b. Since being identified as a 149, 1192–1205. mammalian counterpart of yeast Hac1p, XBP1s is known to act [11] Ross, S.E., Hemati, N., Longo, K.A., Bennett, C.N., Lucas, P.C., Erickson, R.L. and MacDougald, O.A. (2000) Inhibition of adipogenesis by Wnt signaling. Science primarily or exclusively as a transcriptional activator [22].To 289, 950–953. our knowledge, our finding is the first evidence suggesting a neg- [12] Li, H.X., Luo, X., Liu, R.X., Yang, Y.J. and Yang, G.S. (2008) Roles of Wnt/beta- ative regulatory function for XBP1 in mammals. Currently, only catenin signaling in adipogenic differentiation potential of adipose-derived mesenchymal stem cells. Mol. Cell. Endocrinol. 291, 116–124. two reports provide evidence for transcriptional repression by [13] Longo, K.A., Wright, W.S., Kang, S., Gerin, I., Chiang, S.H., Lucas, P.C., Opp, M.R. XBP1. XBP1 inhibits mesodermal and neural tissue formation in and MacDougald, O.A. (2004) Wnt10b inhibits development of white and Xenopus by acting either as transcriptional activator or repressor, brown adipose tissues. J. Biol. Chem. 279, 35503–35509. [14] Christodoulides, C. et al. (2006) WNT10B mutations in human obesity. depending upon the binding of cofactors involved in the forma- Diabetologia 49, 678–684. tion of distinct transcription complexes [23]. Recently, Jafari [15] Chung, S.S., Lee, J.S., Kim, M., Ahn, B.Y., Jung, H.S., Lee, H.M., Kim, J.W. and Park, et al. suggested a different mechanism for XBP1-induced tran- K.S. (2012) Regulation of Wnt/beta-catenin signaling by CCAAT/enhancer binding protein beta during adipogenesis. Obesity (Silver Spring) 20, 482–487. scriptional repression. They suggest that XBP1 differentially regu- [16] Lee, A.H., Iwakoshi, N.N., Anderson, K.C. and Glimcher, L.H. (2003) Proteasome lated the expression of odourant receptors in Drosophila by the inhibitors disrupt the unfolded protein response in myeloma cells. Proc. Natl. location of the binding motif in the regulatory regions [24]. Acad. Sci. USA 100, 9946–9951. [17] Moldes, M., Zuo, Y., Morrison, R.F., Silva, D., Park, B.H., Liu, J. and Farmer, S.R. Although these studies were performed in non-mammalian mod- (2003) Peroxisome-proliferator-activated receptor gamma suppresses Wnt/ el organisms, many reports showing the conservation of UPR [25] beta-catenin signalling during adipogenesis. Biochem. J. 376, 607–613. and IRE1-mediated XBP1s generation mechanisms [26], and of [18] Cawthorn, W.P., Bree, A.J., Yao, Y., Du, B., Hemati, N., Martinez-Santibanez, G. XBP1-binding core sequences [24] during evolution, strongly sug- and MacDougald, O.A. (2012) Wnt6, Wnt10a and Wnt10b inhibit adipogenesis and stimulate osteoblastogenesis through a beta-catenin-dependent gest that XBP1s may also act as a negative regulator in verte- mechanism. Bone 50, 477–489. brates. In this regard, Acosta-Alvear et al. suggested that XBP1 [19] Bennett, C.N., Ross, S.E., Longo, K.A., Bajnok, L., Hemati, N., Johnson, K.W., regulates functionally distinct targets possibly through different Harrison, S.D. and MacDougald, O.A. (2002) Regulation of Wnt signaling during adipogenesis. J. Biol. Chem. 277, 30998–31004. sequence motifs and the differential recruitment of accessory fac- [20] Moon, R.T., Kohn, A.D., De Ferrari, G.V. and Kaykas, A. (2004) WNT and beta- tors [22]. catenin signalling: diseases and therapies. Nat. Rev. Genet. 5, 691–701. In summary, we showed, for the first time, that XBP1s effec- [21] Lee, A.H., Scapa, E.F., Cohen, D.E. and Glimcher, L.H. (2008) Regulation of hepatic lipogenesis by the transcription factor XBP1. Science 320, 1492–1496. tively stimulates preadipocyte differentiation by transcriptional [22] Acosta-Alvear, D. et al. (2007) XBP1 controls diverse cell type- and condition- inhibition of Wnt10b and the subsequent decrease of the anti-adi- specific transcriptional regulatory networks. Mol. Cell 27, 53–66. pogenic b-catenin signalling pathway. With the known regulatory [23] Cao, Y., Knochel, S., Oswald, F., Donow, C., Zhao, H. and Knochel, W. (2006) XBP1 forms a regulatory loop with BMP-4 and suppresses mesodermal and function of C/EBPb on Wnt10b and XBP1, our data provide novel neural differentiation in Xenopus embryos. Mech. Dev. 123, 84–96. information on the early molecular events during the differentia- [24] Jafari, S., Alkhori, L., Schleiffer, A., Brochtrup, A., Hummel, T. and Alenius, M. tion of adipocytes. (2012) Combinatorial activation and repression by seven transcription factors specify Drosophila odorant receptor expression. PLoS Biol. 10, e1001280. [25] Shen, X. et al. (2001) Complementary signaling pathways regulate the unfolded protein response and are required for C. elegans development. Cell Acknowledgment 107, 893–903. [26] Plongthongkum, N., Kullawong, N., Panyim, S. and Tirasophon, W. (2007) Ire1 This research was supported by a Grant (10172KFDA993) from regulated XBP1 mRNA splicing is essential for the unfolded protein response Korea Food and Drug Administration in 2013. (UPR) in Drosophila melanogaster. Biochem. Biophys. Res. Commun. 354, 789– 794.