Functional Analysis of Hes-1 in Preadipocytes [PDF]

0888-8809/06/$15.00/0 Printed in U.S.A.

Molecular Endocrinology 20(3):698–705 Copyright © 2006 by The Endocrine Society doi: 10.1210/me.2005-0325

Functional Analysis of Hes-1 in Preadipocytes David A. Ross, Sridhar Hannenhalli, John W. Tobias, Neil Cooch, Ramin Shiekhattar, and Tom Kadesch Department of Genetics (D.A.R., S.H., T.K.) and Penn Bioinformatics Core (J.W.T.), University of Pennsylvania School of Medicine; and Wistar Institute (N.C., R.S.), Philadelphia, Pennsylvania 19104 Notch signaling blocks differentiation of 3T3-L1 preadipocytes, and this can be mimicked by constitutive expression of the Notch target gene Hes-1. Although considered initially to function only as a repressor, recent evidence indicates that Hes-1 can also activate transcription. We show here that the domains of Hes-1 needed to block adipogenesis coincide with those necessary for transcriptional repression. HRT1, another basichelix-loop-helix protein and potential Hes-1 partner, was also induced by Notch in 3T3-L1 cells but did not block adipogenesis, suggesting that Hes-1 functions primarily as a homodimer or possibly as a heterodimer with an unknown partner. Purification of Hes-1 identified the Groucho/transducin-

like enhancer of split family of corepressors as the only significant Hes-1 interacting proteins in vivo. An evaluation of global gene expression in preadipocytes identified approximately 200 Hes-1-responsive genes comprising roughly equal numbers of up-regulated and down-regulated genes. However, promoter analyses indicated that the downregulated genes were significantly more likely to contain Hes-1 binding sites, indicating that Hes-1 is more likely to repress transcription of its direct targets. We conclude that Notch most likely blocks adipogenesis through the induction of Hes-1 homodimers, which repress transcription of key target genes. (Molecular Endocrinology 20: 698–705, 2006)

A

The relationship between adipogenesis and Hes-1 is not straightforward, however, because a reduction in Hes-1 expression in preadipocytes also inhibits differentiation (8). Notch signaling is not necessary for adipogenesis (10), arguing that the required levels of Hes-1 are maintained independently of Notch. Hes-1 is one of several related proteins known to be induced directly by Notch, yet its roles in mediating Notch’s effects on development and oncogenesis are not well understood. Hes-1 knockout mice display pancreatic and neuronal phenotypes reminiscent of Notch loss-of-function mutations and opposite to those obtained with constitutively active Notch transgenes (11–13). Hes-1 can also mimic Notch’s abilities to block neurogenesis in vitro as measured by neurite outgrowth (14) and to promote keratinocyte differentiation (15). However, expression of Hes-1 alone does not mimic all of Notch’s diverse effects on cells. Hes-1 does not inhibit myogenesis and does not transform T lymphocytes, suggesting that distinct or additional Notch targets are required for those effects (16, 17). Structurally, Hes-1 is a member of the basic-helixloop-helix (bHLH) family of DNA binding proteins and can recruit Groucho/transducin-like enhancer of split (TLE) corepressors via a WRPW motif positioned at its C terminus (18). Hes-1, like other members of the Hes and HRT/Herp/Hey families, possesses a conserved “Orange” domain (also known as the helix 3-helix 4 domain) that along with the WRPW motif and bHLH domain participates in Hes-1’s ability to block neurite outgrowth (14). The roles of these various protein domains have not been evaluated in other cell types due to the lack of functional assays for Hes-1. Only a few

DIPOCYTE DIFFERENTIATION is a complex process that is orchestrated through extensive transcriptional reprogramming (1). Among the proteins best known to promote differentiation are peroxisome proliferator-activating receptor ␥ and CCAAT/ enhancer binding protein ␣, two transcription factors that are coordinately induced late during the differentiation of preadipocytes in culture and that directly activate expression of many adipocyte-specific genes. Transcription factors that repress adipogenesis and whose expression must therefore be down-regulated include the winged-helix DNA binding proteins Foxo1 and Foxa2, the GATA proteins GATA2 and GATA3, and the zinc finger protein GILZ (2–5). Adipogenesis can also be inhibited by transcription factors that respond to the TGF-␤, Wnt/wingless, and Notch signaling pathways (6–8). Constitutive expression of Hes-1, a DNA binding protein whose expression is induced directly by Notch, also blocks adipogenesis, suggesting that Hes-1 mediates some, if not all of Notch’s effects. Given that Hes-1 expression falls during normal adipogenesis in vitro and in vivo (9), Notch likely acts by maintaining Hes-1 expression inappropriately. First Published Online November 10, 2005 Abbreviations: aP2, Adipocyte-specific fatty acid binding protein; bHLH, basic-helix-loop-helix; CAMK, Ca⫹/calmodulin-dependent protein kinase; LDL, low-density lipoprotein; MASH, mouse homolog of Achaete-Scute homolog-1; TLE, transducin-like enhancer of split; WRPW, tryptophan-arginine-proline-tryptophan. Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

698

Ross et al. • Analysis of Hes-1 in Preadipocytes

gene targets of Hes-1 have been identified, including the Hes1 gene itself (19). One Hes-1 target, human Achaete-Scute homolog-1 (hASH1, and its mouse homolog MASH), is likely to mediate some of Notch’s effects on neurogenesis (20). Another, Calcipressin, links Notch signaling with calcium-mediated activation of the transcription factor NFAT in keratinocytes (15). Most others—CD4, lipocalin-type prostaglandin D synthase, and human acid ␣-glucosidase—have not been shown to be important targets of the Notch signaling cascade (21–23). Recent studies on the MASH promoter in neural stem cells indicate that Ca⫹/ calmodulin-dependent protein kinase (CAMK)II␦ converts Hes-1 from a transcriptional repressor to a transcriptional activator (24). A possible role of Hes-1 as a transcriptional activator of other genes and in other cell types is not yet known. We present here an analysis of Hes-1 in 3T3-L1 preadipocytes. We show that the ability of Hes-1 to block adipogenesis correlates with its ability to recruit corepressors, most likely members of the Groucho/ TLE family of proteins. Interestingly, the related protein HRT1 is also induced by Notch in preadipocytes but did not block adipogenesis, suggesting that the critical targets of Hes-1 in adipocytes are distinct. In an effort to identify those targets, we assessed global gene expression in Hes-1 transduced 3T3-L1 cells and identified approximately 200 Hes-1-responsive genes. Those whose expression was lowered by Hes-1, comprising roughly half the total, were much more likely to have Hes-1 binding sites in their promoter regions, indicating that Hes-1 functions primarily as a transcriptional repressor in preadipocytes. RESULTS Hes-1, along with other members of the Hes family, contains several conserved domains including a bHLH domain, necessary for DNA binding and protein dimerization, an Orange domain (or Helix 3/Helix 4 domain), and an WRPW domain necessary for the binding of Groucho/TLE corepressors. To determine the contribution of these domains toward overall Hes-1 activity in blocking adipogenesis, we obtained a series of mutants including 1) a point mutant that eliminates DNA binding (DB mut); 2) a deletion mutant that lacks the C-terminal region including the WRPW motif (⌬S), and 3) a larger deletion that lacks both the WRPW motif and the Orange domain (⌬R) (Fig. 1A) (14). We generated 3T3-L1 cells transduced with retroviruses expressing each of the Hes-1 mutants and evaluated the effects on adipogenesis. The Hes-1 proteins lacking either an intact DNA binding domain or region containing the WRPW motif were unable to block differentiation as judged by both Oil Red-O staining (Fig. 1B) and adipocyte-specific fatty acid binding protein (aP2) expression (Fig. 1C). The role of the Orange domain could not be assessed in these experiments because the DR mutant also lacks the WRPW motif. Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

Mol Endocrinol, March 2006, 20(3):698–705 699

Fig. 1. The bHLH and WRPW Domains of Hes-1 Are Required to Inhibit the Differentiation of 3T3-L1 Preadipocytes 3T3-L1 cells were transduced with the Hes-1 expressing retroviruses (A), subjected to differentiation for 7 d and then stained for Oil Red-O (B) or evaluated for aP2, Hes-1, and HPRT RNAs (C). WT, Wild type.

We observed these same structural requirements when we evaluated the ability of Hes-1 to repress transcription in transient transfection assays using Luciferase (luc) reporters. For this, we measured the ability of Hes-1 to inhibit the activity of sterol response element binding protein (SREBP) proteins, which are also expressed in adipocytes and can also bind Eboxes (25, 26). Our assays employed ADD11–403, a truncated SREBP that lacks the membrane-tethering domain and is therefore constitutively active. As expected, ADD1–403 activated the fatty acid synthase (FAS) promoter, which contains an E box (27, 28), and this was reduced significantly (⬃3-fold) by Hes-1 (Fig. 2A). Similar results were obtained using the reporter [␮E3]4TATA-luc (29), which contains a simple promoter comprising four E-boxes and a TATA box (Fig. 2B). Hes-1 did not affect the ability of ADD11–403 to activate the low-density lipoprotein (LDL) receptor promoter, LDL-luc (30), which contains a typical sterol response element that binds SREBPs, but not Hes-1 (Fig. 2C). Relative to wild-type Hes-1, the DNA binding mutant and the deletion lacking the WRPW motif each showed a diminished inhibitory effect (Fig. 2D). Hes-1 containing a deletion of both the WRPW motif and the Orange domain did not inhibit ADD11–403. Taken together, our data indicate that the ability of Hes-1 to inhibit adipogenesis correlates with its ability to repress transcription. In addition to certain Hes genes, Notch can also activate transcription of genes encoding additional bHLH repressors, including HRT1, HRT2, and HRT3 (31). In 3T3-L1 cells Notch signaling induced exclu-

700 Mol Endocrinol, March 2006, 20(3):698–705

Ross et al. • Analysis of Hes-1 in Preadipocytes

Fig. 3. HRT1 Inhibits Transcription, but Does Not Block Differentiation of 3T3-L1 Cells A, NIH 3T3 cells were transfected with the [␮E3]4TATA-luc reporter along with expression plasmids for the proteins indicated and luciferase activity was determined 2 d later. B, 3T3-L1 cells were transduced with either the parental virus (pBABE) or viruses expressing either Hes-1 or HRT1 and then analyzed for differentiation by Oil Red-O staining (left) or aP2 RNA expression using RT-PCR (right). Cells were also evaluated for expression of RNA corresponding to the Hes-1 and HRT transgenes, using HPRT as a control (right).

Fig. 2. The bHLH and WRPW Domains of Hes-1 Are Required to Inhibit Transcription A–D, NIH3T3 cells were transfected with control vectors (⫺) or expression vectors for ADD1 and wild-type (WT) or mutant Hes-1 proteins as indicated. ADD1 in this and subsequent figures refers to ADD11–403. Reporters carried the FAS promoter (FAS-luc; A), an E-box-based promoter ([␮E3]4TATA-luc; B and D), or the LDL promoter (LDL-luc; C). Activity is shown relative to that obtained with the reporter alone. Mean values and SEM were determined from at least three individual experiments.

sively Hes-1 and HRT1 (data not shown). This raises the possibility that, in preadipocytes, Hes-1 may actually function as a Hes-1/HRT1 heterodimer, which has been reported to be a more effective repressor (32). To assess the potential contribution of HRT1 in mediating Notch’s block to differentiation, we first verified that HRT1 could inhibit transcription from our E-box reporter. NIH 3T3 cells were transfected with the [␮E3]4TATA-luc reporter along with the ADD11–403 expression vector alone or plus Hes-1 and/or HRT1. We found that HRT1 effectively inhibited ADD11–403 and that the combination of HRT1 plus Hes-1 was only slightly more effective (Fig. 3A). We next generated a retrovirus (pBABE-HRT1) that expresses HRT1 and Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

introduced it into 3T3-L1 cells (in parallel with the control parental retrovirus, pBABE, or a retrovirus that expresses Hes-1, pBABE-Hes-1). Cells were then induced to differentiate and either stained with Oil Red-O or assessed for expression of aP2 (Fig. 3B). Whereas Hes-1-transduced cells were unable to differentiate as judged by both assays, cells that harbored the pBABE or pBABE-HRT1 retroviruses differentiated normally. Thus, HRT1 alone cannot mediate the inhibitory effect of Notch on adipogenesis. We conclude that although some Hes-1/HRT1 heterodimers may form as a consequence of Notch signaling, Hes-1 homodimers are likely sufficient to block adipogenesis. Our data cannot rule out the formal possibilities that Hes-1 functions as an obligate heterodimer with an unknown constitutively expressed bHLH protein or that Hes-1/HRT1 heterodimers play a role when adipogenesis is blocked upon Notch signaling. Hes-1 has been shown to bind the Groucho/TLE family of transcriptional repressors (33). Nevertheless, we sought to carry out an unbiased analysis of Hes1-interacting proteins in vivo. We first developed 293T cells stably expressing Flag-tagged Hes-1. Nuclear extracts were then prepared and subjected to affinity purification with an anti-Flag antibody column. Flag peptide eluates revealed the presence of two major bands (Fig. 4), which were individually subjected to trypsin digestion and mass spectrometric sequencing.

Ross et al. • Analysis of Hes-1 in Preadipocytes

Mol Endocrinol, March 2006, 20(3):698–705 701

ried out a global analysis of gene expression using microarrays. We compared RNA extracted from 3T3-L1 cells stably transduced with a control retrovirus with RNA from cells harboring a Hes-1-expressing retrovirus. Because of their availability, we included in our analysis data from cells that had been transduced with an ADD1-expressing virus, with and without the Hes-1 virus. (The number of genes affected by both Hes-1 and ADD1 was relatively low, so this did not significantly influence our results; see Materials and Methods.) Overall, we analyzed six control samples and five Hes-1 samples. Using the overlap of two different statistical approaches [two-way ANOVA and significance analysis of microarrays (SAM; Stanford University, Palo Alto, CA)], we identified 89 genes whose expression was down-regulated in the presence of Hes-1 (including DLK and Hes-1) and 108 genes whose expression was increased. (The full list of genes is published as supplemental data on The Endocrine Society’s Journals Online web site at http:// mend.endojournals.org.) The expression of four novel down-regulated genes (Ltbp2, Clca1, Cri1, and Fgfr2) was confirmed by real-time PCR (Fig. 5). The genes affected by Hes-1 did not fall into any particular class, nor did the list provide any immediate insights into how Hes-1 might inhibit adipogenesis. Given that Hes-1 can both activate and repress transcription, we were interested to know how Hes-1 binding sites were partitioned in the promoters of genes whose expression was either increased or decreased. In the promoters (1 kb) of the 89 downregulated genes, we found 60 occurrences of Hes-1 binding sites, and 36 occurrences in promoters of the 108 up-regulated genes. This corresponds to an increased likelihood of 1.7 in the down-regulated proFig. 4. Members of the Groucho/TLE Family of Corepressors Copurify with Flag-Hes-1 Nuclear extracts of 293T cells (Mock) or 293T cells transduced with a Flag-Hes-1 expression vector were affinity purified and Flag peptide eluates were resolved by SDS-PAGE and silver stained. Sequencing of peptides from the two major bands revealed the presence of the proteins indicated.

The faster migrating band generated peptides corresponding to human Hes-1 as expected (six peptides) and to a lesser extent HRT1 (three peptides). The slower migrating band generated peptides corresponding to human TLE-3 (23 peptides), TLE-4 (15 peptides), and TLE-1 (15 peptides). Although interactions between Hes-1 and Groucho/TLE proteins were not unexpected, our analysis shows that these are the only major Hes-1-interacting proteins in those cells. We did not observe the cadre of TLE1-interacting proteins that have been postulated to form a discreet corepressor complex in these same cells (24), nor did we identify any transcriptional coactivators. To begin to identify components of the Notch pathway in preadipocytes downstream of Hes-1, we carDownloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

Fig. 5. Quantitative PCR of Four Hes-1-Responsive Genes Showing Expression Levels in Control 3T3-L1 Cells and MigR-Hes-1-Transduced 3T3-L1 Cells.

702 Mol Endocrinol, March 2006, 20(3):698–705

moters. Based on 1000 random samplings of 89 promoters from among all annotated mouse promoters, the overrepresentation of HES-1 was significant with P value ⬍ 0.001, whereas the up-regulated promoters had fewer HES-1 sites compared with randomly selected promoters. (The down-regulated promoters were also enriched for binding sites for the DNA binding proteins SRF, Ik-1, and Pax-1, although the significance of this is not clear. The list of down-regulated genes whose promoters contain one or more Hes-1 sites is included in the supplemental data.) We conclude from these analyses that Hes-1 is most likely acting as a repressor in 3T3-L1 preadipocytes and that many of the down-regulated genes, by virtue of their Hes-1 binding sites, are direct targets.

DISCUSSION It is important to define components of the Notch signaling pathway downstream of Hes-1 that lead to the block in adipogenesis. In addition to learning more about Notch signaling itself, this may provide important information concerning the early steps of adipogenesis. We have taken the approach of looking immediately downstream of Notch and Hes-1 because their blocks to adipogenesis occur before the induction of CCAAT/enhancer binding protein ␣ and peroxisome proliferator-activating receptor ␥ (8) where adipogenesis is poorly understood. However, given the recent report that Hes-1 may either activate or repress transcription (24), it was critical for us to determine first how Hes-1 was functioning in preadipocytes. Are the genes immediately downstream from Hes-1 those whose expression increases or those whose expression decreases? Collectively, our data are consistent with Hes-1 functioning primarily as a transcriptional repressor in preadipocytes. First, the domains required for Hes-1 to block adipogenesis are the same as those necessary to repress transcription. Second, we see no evidence of major Hes-1 interacting proteins other than the Groucho/TLE family of transcriptional corepressors. Third, when we compare genes whose expression is either increased or decreased as a consequence of Hes-1, the latter are much more likely to contain Hes-1 binding sites in their promoter regions. The ability of Hes-1 to activate the MASH1 promoter in neural stem cells requires active CAMKII␦, which phosphorylates Hes-1 in the bHLH and Orange domains. Both phosphorylation events are necessary for the recruitment of the coactivator CBP and efficient transcriptional activation (24). Our data show that the bHLH and Orange domains (retained in mutant DS) are not sufficient to mediate the inhibitory effect of Hes-1. (We have not yet confirmed that the bHLH and Orange domains are sufficient to activate transcription from the Mash1 promoter in the presence of active CaMKII␦.) However, we do observe a requirement for Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

Ross et al. • Analysis of Hes-1 in Preadipocytes

amino acids that include the WRPW motif. Although we examined Hes-1-interacting proteins in 293T cells and not 3T3-L1 cells for technical reasons (3T3-L1 cells cannot be transfected efficiently), we found no evidence for the binding of any proteins to Hes-1 other than HRT1 and the Groucho/TLE family of corepressors. We were therefore surprised to detect so many genes whose expression was increased as a consequence of Hes-1; indeed, they slightly outnumbered the down-regulated genes. However, when we looked at the promoter regions of the two classes, there were significantly more putative Hes-1 binding sites in the down-regulated set. Although we acknowledge that looking only at the promoter region (defined here as one kb 5⬘ to the transcription start site) limits our analysis, the result is nevertheless striking and argues that Hes-1 is more likely to bind down-regulated genes than it is up-regulated genes. We propose that the genes whose expression increases are under control of events secondary to the repression events mediated by Hes-1. Our data also underscore the difficulty in elucidating relevant components of the Notch signaling pathway. HRT1 is induced by Notch directly, yet our data show that it does not mimic Notch’s effect on adipogenesis and therefore is not likely to play a role in adipogenesis. By contrast, HRT1 can mimic Notch’s inhibition of myogenesis (Kabak, S., and T. Kadesch, unpublished observations). Similarly, many of the genes directly repressed by Hes-1 in preadipocytes may not be part of the pathway leading to the inhibition of adipogenesis, but may be used in other cell types. Our microarray experiments have provided us with a list of genes affected by Hes-1, each of which will need to be functionally tested for its role in adipocyte development. Furthermore, if we want to establish their positions within the Notch signaling cascade, we will need to determine whether each gene is a direct or indirect target of Hes-1.

Materials and Methods Plasmids and Retroviral Constructs Hes-1 and the Hes-1 mutants, DB mut Hes-1, ⌬S Hes-1, and ⌬R Hes-1 have been described (14) and were kindly provided by Michael Caudy (Burke Medical Research Institute, White Plains, NY). The Hes-1 cDNAs were subcloned into the pMSCV retroviral vector by standard methods. The retroviral vector pBABE-ADD11–403 expresses a constitutively active, nuclear form of SREBP-1c/ADD1 that encompasses amino acids 1–403(34) and was provided by Bruce Spiegelman (Harvard Medical School, Cambridge, MA). MIGR-Hes-1 was the gift of Warren Pear (University of Pennsylvania) and pMSCV-C/EBP␣ was provided by Mitch Lazar (University of Pennsylvania). Production of retroviral supernatant fluid and infection of NIH 3T3 and 3T3-L1 cells were performed as described previously (35). Populations of 3T3-L1 cells transduced with MIGR and MIGR-Hes-1 were generated by infection and subsequent sorting for GFP-positive cells by fluorescence-activated cell sorting. Cell populations harboring pBABE, pBABE-Hes-1, pBABE-HRT1, pBABE-ADD11–403,

Ross et al. • Analysis of Hes-1 in Preadipocytes

pMSCV, pMSCV-Hes-1, pMSCV-DB mut Hes-1, pMSCV⌬SHes-1, pMSCV-⌬RHes-1, and pMSCV-C/EBP␣ were generated by infection followed by selection with 2 ␮g/ml Puromycin. The reporter [␮E3]4TATA-Luc has been described (29). The FAS-luc (36) and LDL-luc (37) reporters, containing the promoters for the FAS and LDL receptor genes, were obtained from Bruce Spiegelman and Timothy Osborne (University of California, Irvine, CA), respectively. Cell Culture and Transfection Assays All cells were maintained in DMEM (Invitrogen Life Technologies, Carlsbad, CA), supplemented with 10% fetal bovine serum, Pen/Strep, glutamine, and the appropriate selective agent if needed. Differentiation of 3T3-L1 cells was performed as described previously (8). Transfections into NIH3T3 cells were carried out with FuGene 6 (Roche Diagnostics Corp., Indianapolis, IN) as per manufacturer’s instructions. Activities of the indicated reporters (50 ng each) were determined by firefly luciferase activity and normalized to Renilla luciferase (2.5 ng of pRL-CMV; Promega, Madison, WI). All transfections were performed at least in triplicate and shown as the average ⫾ SEM. Western, RT-PCR, and Real-Time PCR Analyses Western blotting was performed using standard protocols. Antibodies for C/EBP␣, SREBP-1/ADD-1, and Cdk4 were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Total RNA was prepared using the RNeasy Kit (QIAGEN, Inc., Valencia, CA). First-strand cDNA was prepared by standard protocols. All RT-PCR products were sequenced to verify amplification of the correct cDNA. (Primers sequences are available upon request.) Real-time PCR was performed using the Applied Biosystems Prism 7700 DNA Sequence Detector. Relative levels of mRNA were determined using the comparative threshold (Ct) method. Genes of interest were amplified using FAM-labeled On-demand Assay Kits (Applied Biosystems, Foster City, CA; specific catalog numbers available upon request) and compared with the glyceraldehyde-3phosphate dehydrogenase (GAPDH) VIC probe control RNA amplifications (Taqman Rodent GAPDH Control Kit, Applied Biosystems). Real-time PCR analyses were performed in triplicate and on multiple RNA preparations. Affinity Purification of Flag-Hes-1 Flag-Hes-1 and a selectable marker for puromycin resistance were cotransfected into 293 human embryonic kidney cells. Transfected cells were grown in the presence of 5 ␮g/ml puromycin, and individual colonies were isolated and analyzed for Flag-Hes-1 expression. To purify the complex, nuclear extract from the Flag-Hes-1 cell line was incubated with anti-FLAG M2 affinity gel (Sigma, St. Louis, MO). After extensive washing with buffer A [20 mM Tris-HCl (pH 7.9), 0.5 M KCl, 10% glycerol, 1 mM EDTA, 5 mM dithiothreitol, 0.5% Nonidet P-40], the affinity column was eluted with buffer A containing FLAG peptide (400 ␮g/ml) according to the manufacturer’s instructions (Sigma). Protein identification using liquid chromatography-dual mass spectrometry was performed as detailed previously (38, 39).

Mol Endocrinol, March 2006, 20(3):698–705 703

ADD11–403 together (Hes-1/ADD1). All protocols were conducted as described in the Affymetrix (Santa Clara, CA) GeneChip Expression Analysis Technical Manual using 5 ␮g total RNA. Affymetrix probeset expression values for each array were calculated by the Affymetrix statistical (mas5) algorithm (GCOS version 1.2, Affymetrix Inc.). Expression values were normalized by linear scaling to achieve a trimmed mean (2% trimming) of 150 for each array. The probeset signal values were imported into GeneSpring version 7.2 (Agilent Technologies, Palo Alto, CA), where the arrays were analyzed for intersample consistency using a combination of hierarchical clustering, Principle Components Analysis and pairwise correlations. These analyses revealed one of the Hes-1 samples as an outlier, and it was therefore excluded from subsequent testing for differential gene expression. It was also noted that the differing ADD1 backgrounds had little significance on the overall variation among the samples. Two statistical approaches for the discovery of differentially expressed genes were applied, both with the intent of discovering Hes-1 regulated genes that had consistent changes in the ADD1 and non-ADD1 backgrounds. The mas5 gene signal values from the 11 remaining arrays were evaluated using SAM version 2.0 using two-class unpaired response in the blocked mode, with the ADD1 and the nonADD1 samples forming the experimental blocks. The most significant 369 genes (false discovery rate 0.6%) that were consistently Hes-1 regulated were retained for further analysis. For comparison, a parallel analytical approach was taken using two-way ANOVA as implemented in Partek Pro version 6 (Partek Inc., St. Charles, MO). Gene signal values for the 11 arrays were log2 transformed and the factors Hes-1 (⫹/⫺) and ADD1 (⫹/⫺) were identified. Upon calculation of the two-way ANOVA, genes were ranked by ascending P value for the Hes-1 term. The most significant 363 genes were retained for analysis (1% false discovery rate, by BenjaminiHochberg step-up method). The intersection of the two approaches included 261 Affymetrix probesets, corresponding to 197 unique genes. Binding Site Annotation We extracted the 1-kb regions upstream of the annotated transcripts in the mm5 release of mouse genome from UCSC database (genome.ucsc.edu). We also extracted the HumanMouse alignments for these regions. We searched the 1-kb regions using 531 binding profiles (Positional Weight Matrix or PWM) for vertebrate transcription factors from TRANSFAC version 8.4 (40). The search was done using a tool PWMSCAN (41). The initial hits were based on a P value cutoff of 0.0002, corresponding to an average frequency of 1 hit every 5 kb scanned in the genomic background. We filtered these initial hits further using Human-Mouse alignments. For each hit, we computed the fraction c of binding site bases that were identical between human and mouse. We retained the hits such that either P value ⱕ 0.00002 (1 in 50 kb) or c ⱖ 0.8. This procedure is similar to the one reported previously (41).

Acknowledgments We would like to thank Brian Brunk of the Penn Bioinformatics Core for help with the microarray analyses and members of the Kadesch lab for helpful suggestions.

Microarray Screening and Expression Analyses Microarray screening was performed on triplicate RNA samples isolated from 3T3-L1 cells at d 0 of the adipocyte differentiation protocol (before the addition of differentiation media). The 3T3-L1 cells tested were transduced with 1) the parental retroviruses (MIGR/pBABE), 2) Hes-1 alone (Hes-1/ pBABE), 3) ADD11–403 alone (MIGR/ADD1), or 4) Hes-1 and

Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

Received August 8, 2005. Accepted November 4, 2005. Address all correspondence and requests for reprints to: Tom Kadesch, Department of Genetics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104. E-mail: [email protected]. This work was supported by funds from the National Institutes of Health (RO1 GM58228 to T.K.) and the American

704 Mol Endocrinol, March 2006, 20(3):698–705

Cancer Society (PF-02-120-01-LIB to D.A.R.). D.A.R. was the recipient of the American Cancer Society-IDEC/Genentech/ Ronald Levy postdoctoral fellowship (PF-02-120-01-LIB).

REFERENCES 1. Rosen ED, Walkey CJ, Puigserver P, Spiegelman BM 2000 Transcriptional regulation of adipogenesis. Genes Dev 14:1293–1307 2. Nakae J, Kitamura T, Kitamura Y, Biggs 3rd WH, Arden KC, Accili D 2003 The forkhead transcription factor Foxo1 regulates adipocyte differentiation. Dev Cell 4:119–129 3. Wolfrum C, Shih DQ, Kuwajima S, Norris AW, Kahn CR, Stoffel M 2003 Role of Foxa-2 in adipocyte metabolism and differentiation. J Clin Invest 112:345–356 4. Tong Q, Dalgin G, Xu H, Ting CN, Leiden JM, Hotamisligil GS 2000 Function of GATA transcription factors in preadipocyte-adipocyte transition. Science 290:134–138 5. Shi X, Shi W, Li Q, Song B, Wan M, Bai S, Cao X 2003 A glucocorticoid-induced leucine-zipper protein, GILZ, inhibits adipogenesis of mesenchymal cells. EMBO Rep 4:374–380 6. Choy L, Derynck R 2003 Transforming growth factor-␤ inhibits adipocyte differentiation by Smad3 interacting with CCAAT/enhancer-binding protein (C/EBP) and repressing C/EBP transactivation function. J Biol Chem 278:9609–9619 7. Ross SE, Hemati N, Longo KA, Bennett CN, Lucas PC, Erickson RL, MacDougald OA 2000 Inhibition of adipogenesis by Wnt signaling. Science 289:950–953 8. Ross DA, Rao PK, Kadesch T 2004 Dual roles for the Notch target gene Hes-1 in the differentiation of 3T3-L1 preadipocytes. Mol Cell Biol 24:3505–3513 9. Soukas A, Socci ND, Saatkamp BD, Novelli S, Friedman JM 2001 Distinct transcriptional profiles of adipogenesis in vivo and in vitro. J Biol Chem 276:34167–34174 10. Nichols AM, Pan Y, Herreman A, Hadland BK, De Strooper B, Kopan R, Huppert SS 2004 Notch pathway is dispensable for adipocyte specification. Genesis 40: 40–44 11. Jensen J, Pedersen EE, Galante P, Hald J, Heller RS, Ishibashi M, Kageyama R, Guillemot F, Serup P, Madsen OD 2000 Control of endodermal endocrine development by Hes-1. Nat Genet 24:36–44 12. Ishibashi M, Ang SL, Shiota K, Nakanishi S, Kageyama R, Guillemot F 1995 Targeted disruption of mammalian hairy and Enhancer of split homolog-1 (HES-1) leads to up-regulation of neural helix-loop-helix factors, premature neurogenesis, and severe neural tube defects. Genes Dev 9:3136–3148 13. Ishibashi M, Moriyoshi K, Sasai Y, Shiota K, Nakanishi S, Kageyama R 1994 Persistent expression of helix-loophelix factor HES-1 prevents mammalian neural differentiation in the central nervous system. EMBO J 13: 1799–1805 14. Castella P, Sawai S, Nakao K, Wagner JA, Caudy M 2000 HES-1 repression of differentiation and proliferation in PC12 cells: role for the helix 3-helix 4 domain in transcription repression. Mol Cell Biol 20:6170–6183 15. Mammucari C, Tommasi di Vignano A, Sharov AA, Neilson J, Havrda MC, Roop DR, Botchkarev VA, Crabtree GR, Dotto GP 2005 Integration of Notch 1 and Calcineurin/NFAT signaling pathways in keratinocyte growth and differentiation control. Dev Cell 8:665–676 16. Shawber C, Nofziger D, Hsieh JJ, Lindsell C, Bogler O, Hayward D, Weinmaster G 1996 Notch signaling inhibits muscle cell differentiation through a CBF1-independent pathway. Development 122:3765–3773

Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018

Ross et al. • Analysis of Hes-1 in Preadipocytes

17. Kawamata S, Du C, Li K, Lavau C 2002 Overexpression of the Notch target genes Hes in vivo induces lymphoid and myeloid alterations. Oncogene 21:3855–3863 18. Fisher AL, Ohsako S, Caudy M 1996 The WRPW motif of the Hairy-related bHLH repressor proteins Hairy, Enhancer of Split, and Hes-1 acts as a four amino acid transcription repression and protein-protein interaction domain. Mol Cell Biol 16:2670–2677 19. Hirata H, Yoshiura S, Ohtsuka T, Bessho Y, Harada T, Yoshikawa K, Kageyama R 2002 Oscillatory expression of the bHLH factor Hes1 regulated by a negative feedback loop. Science 298:840–843 20. Sriuranpong V, Borges MW, Strock CL, Nakakura EK, Watkins DN, Blaumueller CM, Nelkin BD, Ball DW 2002 Notch signaling induces rapid degradation of achaetescute homolog 1. Mol Cell Biol 22:3129–3139 21. Kim HK, Siu G 1998 The notch pathway intermediate HES-1 silences CD4 gene expression. Mol Cell Biol 18: 7166–7175 22. Fujimori K, Fujitani Y, Kadoyama K, Kumanogoh H, Ishikawa K, Urade Y 2003 Regulation of lipocalin-type prostaglandin D synthase gene expression by Hes-1 through E-box and interleukin-1 ␤ via two NF-␬B elements in rat leptomeningeal cells. J Biol Chem 278:6018–6026 23. Yan B, Raben N, Plotz PH 2002 Hes-1, a known transcriptional repressor, acts as a transcriptional activator for the human acid ␣-glucosidase gene in human fibroblast cells. Biochem Biophys Res Commun 291:582–587 24. Ju BG, Solum D, Song EJ, Lee KJ, Rose DW, Glass CK, Rosenfeld MG 2004 Activating the PARP-1 sensor component of the groucho/TLE1 corepressor complex mediates a CaMKinase II␦-dependent neurogenic gene activation pathway. Cell 119:815–829 25. Sasai Y, Kageyama R, Tagaawa Y, Shigemoto R, Nakanishi S 1992 Two mammalian helix-loop-helix factors structurally related to Drosophila hairy and Enhancer of split. Genes Dev 6:2620–2634 26. Kim JB, Spotts GD, Halvorsen YD, Shih HM, Ellenberger T, Towle HC, Spiegelman BM 1995 Dual DNA binding specificity of ADD1/SREBP1 controlled by a single amino acid in the basic helix-loop-helix domain. Mol Cell Biol 15:2582–2588 27. Bennett MK, Lopez JM, Sanchez HB, Osborne TF 1995 Sterol regulation of fatty acid synthase promoter. Coordinate feedback regulation of two major lipid pathways. J Biol Chem 270:25578–25583 28. Kim JB, Sarraf P, Wright M, Yao KM, Mueller E, Solanes G, Lowell BB, Spiegelman BM 1998 Nutritional and insulin regulation of fatty acid synthetase and leptin gene expression through ADD1/SREBP1. J Clin Invest 101:1–9 29. Carter R, Ordentlich P, Kadesch T 1997 Selective utilization of bHLHzip proteins at the IgH enhancer. Mol Cell Biol 17:18–23 30. Yieh L, Sanchez HB, Osborne TF 1995 Domains of transcription factor Sp1 required for synergistic activation with sterol regulatory element binding protein 1 of low density lipoprotein receptor promoter. Proc Natl Acad Sci USA 92:6102–6106 31. Iso T, Kedes L, Hamamori Y 2003 HES and HERP families: multiple effectors of the Notch signaling pathway. J Cell Physiol 194:237–255 32. Iso T, Sartorelli V, Poizat C, Iezzi S, Wu HY, Chung G, Kedes L, Hamamori Y 2001 Herp, a novel heterodimer partner of hes/e(spl) in notch signaling. Mol Cell Biol 21:6080–6089 33. Grbavec D, Stifani S 1996 Molecular interaction between TLE1 and the carboxyl-terminal domain of HES-1 containing the WRPW motif. Biochem Biophys Res Commun 223:701–705 34. Kim JB, Spiegelman BM 1996 ADD1/SREBP1 promotes adipocyte differentiation and gene expression linked to fatty acid metabolism. Genes Dev 10:1096–1107

Ross et al. • Analysis of Hes-1 in Preadipocytes

35. Rao PK, Dorsch M, Chickering T, Zheng G, Jiang C, Goodearl A, Kadesch T, McCarthy S 2000 Isolation and characterization of the notch ligand ␦4. Exp Cell Res 260:379–386 36. Athanikar JN, Osborne TF 1998 Specificity in cholesterol regulation of gene expression by coevolution of sterol regulatory DNA element and its binding protein. Proc Natl Acad Sci USA 95:4935–4940 37. Sanchez HB, Yieh L, Osborne TF 1995 Cooperation by sterol regulatory element-binding protein and Sp1 in sterol regulation of low density lipoprotein receptor gene. J Biol Chem 270:1161–1169 38. Bochar DA, Wang L, Beniya H, Kinev A, Xue Y, Lane WS, Wang W, Kashanchi F, Shiekhattar R 2000 BRCA1 is

Mol Endocrinol, March 2006, 20(3):698–705 705

associated with a human SWI/SNF-related complex: linking chromatin remodeling to breast cancer. Cell 102: 257–265 39. Marmorstein LY, Kinev AV, Chan GK, Bochar DA, Beniya H, Epstein JA, Yen TJ, Shiekhattar R 2001 A human BRCA2 complex containing a structural DNA binding component influences cell cycle progression. Cell 104: 247–257 40. Wingender E, Dietze P, Karas H, Knuppel R 1996 TRANSFAC: a database on transcription factors and their DNA binding sites. Nucleic Acids Res 24:238–241 41. Levy S, Hannenhalli S 2002 Identification of transcription factor binding sites in the human genome sequence. Mamm Genome 13:510–514

Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

The Twenty-Fifth Annual University of Kentucky Symposium in Reproductive Sciences and Women’s Health May 18-19, 2006 This annual symposium of the Reproductive Sciences Forum at the University of Kentucky brings faculty, clinicians, and trainees in the field of reproductive sciences and women’s health together to exchange information and ideas with each other and with recognized leaders of the reproductive sciences research community. The program consists of plenary lectures, a poster session, meet-the-professor luncheon and dinner. For registration information visit our web site at http://www2.mc.uky.edu/OBG/Forum/ Symposium/Home.htm. Or contact: Michael Kilgore, PhD Symposium Director Department of Molecular and Biomedical Pharmacology University of Kentucky Lexington, KY 40536 Tel: (859) 323-1821 E-mail: [email protected]

Downloaded from https://academic.oup.com/mend/article-abstract/20/3/698/2738410 by guest on 11 February 2018