Portal de Periódicos da CAPES

Notch-1 Controls the Expression of Fatty Acid-activated Transcription Factors and Is Required for Adipogenesis

1997; Elsevier BV; Volume: 272; Issue: 47 Linguagem: Inglês

10.1074/jbc.272.47.29729

1083-351X

Cármen Garcés, Marı́a José Ruiz-Hidalgo, Jaime Font de Mora, Crystal L. Park, Lucio Miele, Julia Goldstein, Ezio Bonvini, Almudena Porrás, Jorge Laborda,

Genetic Syndromes and Imprinting

Notch, a transmembrane receptor member of the homeotic epidermal growth factor-like family of proteins, participates in cell-to-cell signaling to control cell fate during development. Activated Notch-1 constructs lacking the extracellular region prevent differentiation of several mammalian cells in vitro. This effect, however, bypasses the normal mechanisms of cell-to-cell interactions in which Notch-1 participates. We investigated the role of Notch-1 in the hormone-induced adipocyte differentiation of 3T3-L1 fibroblasts, a paradigmatic model of adipogenesis that requires cell-to-cell contact. Unlike other differentiation models, Notch-1 expression and function were necessary conditions for adipogenesis. Impaired Notch-1 expression by antisenseNotch-1 constructs prevented adipocyte differentiation. Strategies aimed at blocking putative Notch/ligand interactions also blocked adipogenesis, implicating Notch as a critical molecule in cell-to-cell signaling necessary for differentiation. Inhibition of Notch-1 expression or function decreased the expression of peroxisomal proliferator-activated receptors δ and γ, transcription factors that control adipocyte differentiation and that are up-regulated at cell confluence. These results implicate Notch in the commitment of 3T3-L1 cells to undergo adipogenesis by controlling the expression of the principal regulators of this process. Notch, a transmembrane receptor member of the homeotic epidermal growth factor-like family of proteins, participates in cell-to-cell signaling to control cell fate during development. Activated Notch-1 constructs lacking the extracellular region prevent differentiation of several mammalian cells in vitro. This effect, however, bypasses the normal mechanisms of cell-to-cell interactions in which Notch-1 participates. We investigated the role of Notch-1 in the hormone-induced adipocyte differentiation of 3T3-L1 fibroblasts, a paradigmatic model of adipogenesis that requires cell-to-cell contact. Unlike other differentiation models, Notch-1 expression and function were necessary conditions for adipogenesis. Impaired Notch-1 expression by antisenseNotch-1 constructs prevented adipocyte differentiation. Strategies aimed at blocking putative Notch/ligand interactions also blocked adipogenesis, implicating Notch as a critical molecule in cell-to-cell signaling necessary for differentiation. Inhibition of Notch-1 expression or function decreased the expression of peroxisomal proliferator-activated receptors δ and γ, transcription factors that control adipocyte differentiation and that are up-regulated at cell confluence. These results implicate Notch in the commitment of 3T3-L1 cells to undergo adipogenesis by controlling the expression of the principal regulators of this process. The Notch gene family has been the object of extensive investigation in both vertebrates and invertebrates (1Artavanis-Tsakonas S. Matsuno K. Fortini M.E. Science. 1995; 268: 225-232Crossref PubMed Scopus (1410) Google Scholar, 2Greenwald I. Curr. Opin. Genet. & Dev. 1994; 4: 556-562Crossref PubMed Scopus (103) Google Scholar, 3Lardelli M. Williams R. Lendahl U. Int. J. Dev. Biol. 1995; 39: 769-780PubMed Google Scholar, 4Muskavitch M.A. Dev. Biol. 1994; 166: 415-430Crossref PubMed Scopus (224) Google Scholar, 5Nye J.S. Kopan R. Curr. Biol. 1995; 5: 966-969Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar).Notch belongs to the family of EGF 1The abbreviations used are: EGF, epidermal growth factor; C/EBP, CAATT enhancer-binding protein; PPAR, peroxisomal proliferator-activated receptor; RT-PCR, reverse transcription-polymerase chain reaction. -like homeotic genes, which encode transmembrane proteins with a variable number of cysteine-rich EGF-like repeats in the extracellular region. In invertebrates, these motifs participate in cell-to-cell like interactions that ultimately regulate differentiation decisions during development (6Rebay I. Fleming R.J. Fehon R.G. Cherbas L. Cherbas P. Artavanis-Tsakonas S. Cell. 1991; 67: 687-699Abstract Full Text PDF PubMed Scopus (603) Google Scholar). In mammals, four Notch genes have been described, Notch-1, Notch-2, Notch-3, and Int-3 (3Lardelli M. Williams R. Lendahl U. Int. J. Dev. Biol. 1995; 39: 769-780PubMed Google Scholar, 7Jhappan C. Gallahan D. Stahle C. Chu E. Smith G.H. Merlino G. Callahan R. Genes Dev. 1992; 6: 345-355Crossref PubMed Scopus (300) Google Scholar, 8Robbins J. Blondel B.J. Gallahan D. Callahan R. J. Virol. 1992; 66: 2594-2599Crossref PubMed Google Scholar), which have been implicated in the differentiation of the nervous system and other structures (9Fortini M.E. Artavanis-Tsakonas S. Cell. 1993; 75: 1245-1247Abstract Full Text PDF PubMed Scopus (112) Google Scholar). The invertebrate EGF-like proteins Delta and Serrate and their mammalian counterparts have been identified as ligands of Notch-1 (10Bettenhausen B. Deangelis M.H. Simon D. Guenet J.L. Gossler A. Development (Camb.). 1995; 121: 2407-2418Crossref PubMed Google Scholar, 11Fleming R.J. Scottgale T.N. Diederich R.J. Artavanis-Tsakonas S. Genes Dev. 1990; 4: 2188-2201Crossref PubMed Scopus (173) Google Scholar, 12Lindsell C.E. Shawber C.J. Boulter J. Weinmaster G. Cell. 1995; 80: 909-917Abstract Full Text PDF PubMed Scopus (540) Google Scholar). In the case of Drosophila and Xenopus, these ligands have been reported to interact with EGF-like repeats 11 and 12 of Notch to trigger its activation (6Rebay I. Fleming R.J. Fehon R.G. Cherbas L. Cherbas P. Artavanis-Tsakonas S. Cell. 1991; 67: 687-699Abstract Full Text PDF PubMed Scopus (603) Google Scholar). It has been proposed that activated Notch functions by maintaining the cells in an undifferentiated state (12Lindsell C.E. Shawber C.J. Boulter J. Weinmaster G. Cell. 1995; 80: 909-917Abstract Full Text PDF PubMed Scopus (540) Google Scholar). The role of Notch in the differentiation of mammalian cells is, however, still incompletely understood. An activated Notchmutant has been shown to suppress neurogenesis and myogenesis, but not gliogenesis in P19 cells (13Nye J.S. Kopan R. Axel R. Development (Camb.). 1994; 120: 2421-2430Crossref PubMed Google Scholar). An inhibitory effect on the function of MyoD1 has been demonstrated in the C2C12 myoblast model (14Kopan R. Nye J.S. Weintraub H. Development (Camb.). 1994; 120: 2385-2396Crossref PubMed Google Scholar, 15Kopan R. Schroeter E.H. Weintraub H. Nye J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1683-1688Crossref PubMed Scopus (425) Google Scholar). Since cell-to-cell contact is required for the differentiation of 3T3-L1 cells to adipocytes (16Green H. Kehinde O. Cell. 1974; 1: 113-116Abstract Full Text PDF Scopus (746) Google Scholar, 17Green H. Kehinde O. Cell. 1976; 7: 105-113Abstract Full Text PDF PubMed Scopus (616) Google Scholar, 18MacDougald O.A. Lane M.D. Annu. Rev. Biochem. 1995; 64: 345-373Crossref PubMed Scopus (943) Google Scholar), we chose this well studied differentiation model to further explore the role of mammalian Notch-1. 3T3-L1 cells are of mesenchymal origin and are committed to undergo adipocyte differentiation when grown to confluence and treated with a combination of hormones, including glucocorticoids and insulin-like growth factor 1/insulin. Following activation of the Ras signaling cascade (19Benito M. Porrás A. Nebreda A.R. Santos E. Science. 1991; 253: 565-568Crossref PubMed Scopus (141) Google Scholar, 20Porrás A. Nebreda A.R. Benito M. Santos E. J. Biol. Chem. 1992; 267: 21124-21131Abstract Full Text PDF PubMed Google Scholar, 21Porrás A. Muszynski K. Rapp U.R. Santos E. J. Biol. Chem. 1994; 269: 12741-12748Abstract Full Text PDF PubMed Google Scholar), the cells undergo gene expression changes that lead to a morphological and biochemical metamorphosis toward the adipocyte phenotype. Cells become round and accumulate lipids in the form of cytoplasmic droplets easily detected by phase-contrast microscopy or when stained with lipophilic agents. The genetic and biochemical changes associated with differentiation of 3T3-L1 cells have been extensively studied (reviewed in Refs. 18MacDougald O.A. Lane M.D. Annu. Rev. Biochem. 1995; 64: 345-373Crossref PubMed Scopus (943) Google Scholar and 22Cornelius P. MacDougald O.A. Lane M.D. Annu. Rev. Nutr. 1994; 14: 99-129Crossref PubMed Scopus (575) Google Scholar), supporting thisin vitro model as representative of in vivoadipogenesis. For instance, in vitro differentiated adipocytes can be injected into animals, where they form adipose tissue indistinguishable from normal (23Green H. Kehinde O. J. Cell. Physiol. 1979; 101: 169-172Crossref PubMed Scopus (234) Google Scholar). The transcription factors that regulate the biochemical and morphological changes leading to adipocyte differentiation belong to the CAATT enhancer-binding protein (C/EBP) family and the thyroid/steroid family. Peroxisomal proliferator-activated receptors (PPARs), members of the latter family, are activated by fatty acid metabolites (24Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. Cell. 1995; 83: 803-812Abstract Full Text PDF PubMed Scopus (2740) Google Scholar, 25Kliewer S.A. Lenhard J.M. Willson T.M. Patel I. Morris D.C. Lehmann J.M. Cell. 1995; 83: 813-819Abstract Full Text PDF PubMed Scopus (1872) Google Scholar) and function by forming heteromeric entities with retinoid X receptor α (26Keller H. Dreyer C. Medin J. Mahfoudi A. Ozato K. Wahli W. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2160-2164Crossref PubMed Scopus (861) Google Scholar, 27Safonova I. Darimont C. Amri E.Z. Grimaldi P. Ailhaud G. Reichert U. Shroot B. Mol. Cell. Endocrinol. 1994; 104: 201-211Crossref PubMed Scopus (98) Google Scholar, 28Safonova I. Reichert U. Shroot B. Ailhaud G. Grimaldi P. Biochem. Biophys. Res. Commun. 1994; 204: 498-504Crossref PubMed Scopus (23) Google Scholar). C/EBPα or PPARγ expression facilitates adipocyte differentiation (29Freytag S.O. Geddes T.J. Science. 1992; 256: 379-382Crossref PubMed Scopus (251) Google Scholar, 30Freytag S.O. Paielli D.L. Gilbert J.D. Genes Dev. 1994; 8: 1654-1663Crossref PubMed Scopus (393) Google Scholar, 31Lin F.T. Lane M.D. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 8757-8761Crossref PubMed Scopus (383) Google Scholar, 32Tontonoz P. Hu E. Spiegelman B.M. Cell. 1994; 79: 1147-1156Abstract Full Text PDF PubMed Scopus (3133) Google Scholar, 33Tontonoz P. Hu E. Graves R.A. Budavari A.I. Spiegelman B.M. Genes Dev. 1994; 8: 1224-1234Crossref PubMed Scopus (2005) Google Scholar), and both transcription factors synergize to promote the adipocyte differentiation program (32Tontonoz P. Hu E. Spiegelman B.M. Cell. 1994; 79: 1147-1156Abstract Full Text PDF PubMed Scopus (3133) Google Scholar,34Hu E.D. Tontonoz P. Spiegelman B.M. Proc. Natl. Acad. Sci. U. S. A. 1994; 92: 9856-9860Crossref Scopus (583) Google Scholar). Adipocyte differentiation obtained by ectopic expression of c/EBPβ and C/EBPδ is mediated by an increase in PPARγ expression (35Wu Z.D. Xie Y.H. Bucher N.L.R. Farmer S.R. Genes Dev. 1994; 9: 2350-2363Crossref Scopus (481) Google Scholar). These data suggest that PPARγ plays a key role as a "master regulator" of the adipocyte differentiation process. We transfected 3T3-L1 cells with a partial Notch-1 construct encompassing an intracellular region of the protein containing the six intracellular ankyrin repeats in either sense or antisense orientation. Differentiation was not affected by ectopic Notchtransfected in the sense orientation. AntisenseNotch-1-transfected cells, however, lost their ability to undergo adipocyte differentiation, an effect that correlated with the inhibition of Notch protein expression in these cells. Moreover, differentiation could also be inhibited by treatment of normal cells with a soluble recombinant protein containing exclusively EGF-like repeats 11 and 12 of Notch-1 (Hurn1EGF1112) or with an antiserum directed against the same EGF-like repeats. Inhibition of Notch-1 expression correlated with morphological changes similar to those induced by culturing 3T3-L1 cells in delipidated medium, a condition that inactivates PPARs (36Chawla A. Lazar M.A. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1786-1790Crossref PubMed Scopus (218) Google Scholar). Analysis of some members of the PPAR family showed that their expression was decreased in the antisenseNotch-1-transfected cells or in cells treated with the recombinant Hurn1EGF1112 protein, whereas the expression of other transcription factors remained unchanged. These results support the conclusion that the expression of Notch-1 is required for adipocyte differentiation of 3T3-L1 cells and that Notch-1 participates in the cell-to-cell interactions that modulate the expression of the transcription factors that regulate this differentiation process. 3T3-L1 cells were obtained from American Type Culture Collection and cultured at 37 °C in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Life Technologies, Inc.) in a humidified atmosphere of 7% CO2 in air. Differentiation assays were performed as described previously (19Benito M. Porrás A. Nebreda A.R. Santos E. Science. 1991; 253: 565-568Crossref PubMed Scopus (141) Google Scholar). Briefly, cells were plated in 10-cm plates and allowed to grow until confluence. After 1–2 days in this state, the medium was replaced with Dulbecco's modified Eagle's medium supplement with 10% fetal bovine serum containing 0.5 mm isobutylmethylxanthine and 500 nm dexamethasone (Sigma). Two days later, the medium was replaced with regular medium to which insulin was added to a final concentration of 1 μm. The cells were maintained in this medium for 5–7 days, with media replacement each 2 days. At the end of this period, differentiation was assessed by red oil O (Sigma) staining and microscopic examination. Hurn1EGF1112 protein or rabbit antisera against the same protein or control antisera were added at the beginning of the differentiation process and maintained at the indicated concentrations for the duration of the differentiation assay. Cell transfections were performed using LipofectAMINE (Life Technologies, Inc.) according to the manufacturer's protocol. Cells were selected in 500 μg/ml hygromycin. Subcloning of hygromycin-resistant cells was performed by limiting dilution and recovery of individual cell colonies. The Notch-1construct used for the experiments described here was obtained by RT-PCR amplification of human spleen total RNA using Pfupolymerase (Stratagene) and a PCR reagent kit (Perkin-Elmer). The sequence of the sense oligonucleotide is 5′-agtgtctgaggccagcaagaagaa-3′, and that of the antisense oligonucleotide is 5′-gaggggcacggacggagact-3′. The region amplified is 1372 nucleotides long and corresponds to the intracellular Notch-1 region containing the ankyrin/Cdc10 repeats from amino acids 1776 to 2232 of the human TAN-1 sequence (GenBank™ accession number M73980). The blunt-end PCR product was cloned into the PvuII site of the plasmid pEBVHisA (Invitrogen) to produce in-frame sense and antisense expression constructs. Production of a human recombinant protein composed of EGF-like repeats 11 and 12 of Notch-1 (Hurn1EGF1112) was accomplished by RT-PCR amplification from human spleen total RNA. The primers used had the following sequences: 5′-acaccatggctgcagaattccatcatcatcatcatcatcaggacgtggatgagtgct-3′ for the sense oligonucleotide and 5′-gataagcttggatcctcatcatgtgttgacctcgcagtgca-3′ for the antisense oligonucleotide. This fragment spanned from positions 1231 to 1471 of the human TAN-1 (Notch-1) sequence. The primers were designed to encode an additional six-histidine tag to enable purification of the recombinant protein by affinity chromatography through nickel columns. The PCR fragment was cloned in PLD101, a bacterial expression plasmid developed for the expression of functional proteins rich in disulfide bonds, as described previously (37Miele L. Cordella-Miele E. Mukherjee A.B. J. Biol. Chem. 1990; 265: 6427-6435Abstract Full Text PDF PubMed Google Scholar), and the protein was expressed in BL21(DE3) bacteria (Novagen). The recombinant protein was purified by Ni2+-nitrilotriacetic acid affinity and size-exclusion chromatography. Purity of the protein was assessed by SDS-polyacrylamide gel electrophoresis, and protein concentration was determined by spectrophotometry. This protein was used to study its effect on differentiation and to produce rabbit antisera. Analysis of the expression of Notch-1, PPARγ, and PPARδ in 3T3-L1 cells was performed by RT-PCR from 2 μg of RNA isolated from these cells. RNA was isolated by using the QIAGEN total RNA purification kit according to the manufacturer's directions. First strand cDNA was prepared by using a reverse transcriptase cDNA kit (CLONTECH) following the manufacturer's recommended procedures. The upper and lower primers used were 5′-gtgtgtggatgagggagataa-3′ and 5′-ggcatagacagcggtagaaa-3′ for Notch-1, 5′-ttgagtgccgagtctgtggggataa-3′ and 5′-cagggaggccagcatcgtgtaga-3′ for PPARγ, 5′-gaggcccgggaagaggagaaagaggaagtg-3′ and 5′-tgaggaagaggctgctgaagttggggatgt-3′ for PPARδ, 5′-cccggccgccttcaacgacgagttc-3′ and 5′-gatgccccgcagcgtgtccagttca-3′ for C/EBPα, 5′-ccgccgcccgccgcctttaga-3′ and 5′-ccgccgtcagctccagcaccttgtg-3′ for C/EBPβ, 5′-ccgacctcttcaacagcaaccacaaag-3′ and 5′-gcatgcgcagtctcttcctcttatctacaaaa-3′ for C/EBPδ, 5′-ccccagctcaccaaatgaccctgttacc-3′ and 5′-accttctccctcaacgcctccacctc-3′ for retinoid X receptor α, and 5′-agacagctcctccccaaatcccctttct-3′ and 5′-tggccaggcacttctgatagcgacagt-3′ for retinoid X receptor β, respectively. As a RNA loading and quality control, 36B4/P0 phosphoriboprotein (38Laborda J. Nucleic Acids Res. 1991; 19: 3998Crossref PubMed Scopus (434) Google Scholar) was also amplified. The primers used for this amplification were 5′-gcactttcgctttctggagggtgtc-3′ for the sense oligonucleotide and 5′-tgacttggttgctttggcgggattag-3′ for the antisense oligonucleotide. PCR negative controls, containing all reagents except cDNA, were set in all cases. To ensure PCR specificity, we used two-step PCR cycles of extension/annealing at 68–70 °C for 2 min and denaturing at 95 °C for 2 s. The specificity of the primers was assessed by agarose gel electrophoresis showing unique bands of the expected sizes. To further confirm specificity, PCR products were sequenced and cloned in PCR II (Invitrogen), and several minipreps were sequenced. Western blots for the analysis of Notch-1 expression in the 3T3-L1 cells and transfectants were performed according to classical procedures (39Harlow E. Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1988: 471-510Google Scholar) using the rabbit antisera raised against the recombinant Hrun1EGF1112 protein. Detection of bound antibody was performed by incubation with 125I-protein A (Amersham Corp.). Pools of hygromycin-resistant 3T3-L1 cells transfected with sense or antisenseNotch-1 constructs were analyzed for their ability to undergo adipocyte differentiation upon treatment with differentiating agents. The cells transfected with sense Notch-1 did not show any changes in their differentiation potential when compared with either vector-transfected or normal cells. However, cells stably transfected with antisense Notch-1 completely lost their ability to differentiate to adipocytes, as determined by the absence of morphological changes and lipid accumulation characteristic of the adipocyte phenotype (Fig.1 A). This behavior could be observed in individual clones. Following a second transfection with the same antisense Notch construct, the antibiotic-resistant cells were cloned by limiting dilution. During this process, we observed two morphologically distinct types of clones (Fig. 1 B). The first type (type N) consisted of cells morphologically indistinguishable from normal 3T3-L1 cells that grew in contact with each other as they expanded from the original cell founder of the colony. The second type of colony (type E) consisted of cells that showed a fusiform or elongated morphology and avoided contact with each other as they multiplied. When isolated and allowed to grow to confluence, cell-to-cell contact was re-established. At confluence, some type E clones formed a monolayer indistinguishable from parental cells (type E1), whereas others formed a more ordered monolayer (type E2) (Fig. 1 B) similar to primary fetal mouse fibroblasts in culture. All the above clones, regardless of their morphology, were unable to differentiate toward the adipocyte phenotype. Type E1 clones, however, when treated at confluence with differentiating agents, underwent morphological changes that made them resemble type E2 clones (data not shown). No morphological changes were observed with the other antisense Notch-1 clone types when treated under the same conditions. Morphological changes or lack of differentiation was not observed in a similar number of clones isolated from mock or senseNotch-1 transfections. Analysis of Notch-1 expression by RT-PCR in normal or transfected cells showed that transfection of 3T3-L1 cells with antisense Notch-1diminished the expression of the Notch-1 gene (Fig.2 A). The extent of decrease in Notch-1 mRNA levels correlated well with the acquisition of the elongated morphology, as confirmed by a semiquantitative RT-PCR comparison of Notch-1 expression in type N clones, showing no morphological changes, and type E clones, showing elongated morphology (Fig. 2 B). We next analyzed whether the decrease in Notch-1 mRNA expression detected by RT-PCR resulted in diminished protein expression levels. A Western blot, performed with normal and antisense transfected cell extracts and antisera raised against EGF-like repeats 11 and 12 of Notch-1, showed that Notch-1 protein expression was exclusively detected in normal cells (Fig. 2 C). 3T3-L1 antisense transfectants and SUP-T1, a T-cell leukemia line that does not express intact Notch-1 (40Ellison L.W. Bird J. West D.C. Soreng A.L. Reynolds T.C. Smith S.D. Sklar J. Cell. 1991; 66: 649-661Abstract Full Text PDF PubMed Scopus (1454) Google Scholar), showed no detectable expression. Absence of membrane expression of Notch-1 protein in the antisense transfected clones was also evidenced by flow cytometry (data not shown). These results demonstrated that transfection with the antisenseNotch-1 construct caused an inhibition of the expression of both Notch-1 mRNA and protein levels. The morphological changes observed in type E1 and E2 clones, which show the largest decrease in Notch-1 expression, resembled the changes observed in 3T3-L1 cells cultured in delipidated medium (Ref. 36Chawla A. Lazar M.A. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1786-1790Crossref PubMed Scopus (218) Google Scholar and data not shown). Under these conditions, activation of the fatty acid metabolite-dependent transcription factors of the PPAR family is inhibited, and adipocyte differentiation does not occur. The striking similarity of the morphological features among some of our antisense transfectant clones and normal cells grown in delipidated medium suggested that Notch-1 inhibition could have an effect on the regulation of the expression of the PPAR family of transcription factors. RT-PCR analysis of the expression levels of two members of this family indicated that while type N clones showed detectable but extremely low levels of PPARγ, no expression of this transcription factor was observed in type E clones (Fig.3, upper panel). The expression of PPARδ was also diminished in both types of clones. This indicates a correlation between the expression of PPARγ and the absence of morphological changes. The inability to differentiate correlated also with the profound decrease in PPARδ and PPARγ, consistent with the key role of these transcription factors in adipogenesis. We also analyzed the expression of other transcription factors potentially involved in adipocyte differentiation. The expression of these transcription factors in nondifferentiated cells was unaffected by transfection with antisense Notch-1 (Fig. 3, lower panels). As happened for PPARγ and PPARδ, antisenseNotch-1 cells induced to differentiate did not up-regulate C/EBPα, although they could down-regulate retinoid X receptor β. The expression of C/EBPβ could not be detected either in nondifferentiated cells or in cells treated to differentiate, including normal 3T3-L1 cells (data not shown). Thus, it appears that Notch-1 specifically regulates the expression of members of the PPAR family. In all cell models studied, in vitro adipogenesis requires induction at confluence (11Fleming R.J. Scottgale T.N. Diederich R.J. Artavanis-Tsakonas S. Genes Dev. 1990; 4: 2188-2201Crossref PubMed Scopus (173) Google Scholar, 40Ellison L.W. Bird J. West D.C. Soreng A.L. Reynolds T.C. Smith S.D. Sklar J. Cell. 1991; 66: 649-661Abstract Full Text PDF PubMed Scopus (1454) Google Scholar), suggesting that cell-to-cell contact is a prerequisite for differentiation. The observations reported here are consistent with the hypothesis that Notch signaling triggered by cell-to-cell contact at confluence is a necessary condition for differentiation. This hypothesis was tested by blocking a potential site for ligand interaction with Notch-1 in 3T3-L1 cells. Two EGF-like repeats, EGF-like repeats 11 and 12, have been reported to constitute the conserved region of ligand interaction for Delta and Serrate inDrosophila. A recombinant Drosophila Notch protein devoid of the other 34 EGF-like repeats of the extracellular domain has been shown to maintain interaction with these ligands. The same EGF-like repeats of Xotch, the Xenopus homolog ofDrosophila Notch, have also been shown to interact withDrosophila Delta (6Rebay I. Fleming R.J. Fehon R.G. Cherbas L. Cherbas P. Artavanis-Tsakonas S. Cell. 1991; 67: 687-699Abstract Full Text PDF PubMed Scopus (603) Google Scholar). Differentiation of 3T3-L1 cells was induced in the presence of various concentrations of a recombinant protein encompassing EGF-like repeats 11 and 12 of Notch-1 (HurnEGF1112). Differentiation of 3T3-L1 cells was also induced in the presence of a rabbit antiserum directed against Hurn1EGF1112 as another method to prevent potential Notch-1/ligand interaction. The results of these experiments showed that differentiation of 3T3-L1 cells was significantly inhibited in the presence of 0.1 μmHurn1EGF1112, and 1 μm Hurn1EGF1112 completely inhibited adipocyte differentiation (Fig.4 A). In addition, a 1:100 dilution of the rabbit antiserum directed against EGF-like repeats 11 and 12 of Notch-1 had identical effects and totally inhibited the differentiation, whereas preimmune serum and other antisera raised against irrelevant antigens had no effect. No morphological changes were observed in the Hurn1EGF1112- or antiserum-treated cells, presumably reflecting residual Notch signal levels. We then studied whether the inhibition of differentiation caused by the recombinant Hurn1EGF1112 protein was associated with decreased levels of expression of PPAR transcription factors. RT-PCR analysis of the expression of PPARγ and PPARδ (Fig. 4 B) showed that 3T3-L1 cells treated with Hurn1EGF1112 had decreased levels of these transcription factors. These results suggest that molecular interference with the normal Notch-1/ligand interaction produces the same effects as the inhibition of Notch-1 expression by genetic means. Thus, Notch-1 function is necessary for the expression of transcription factors of the PPAR family in confluent 3T3-L1 cells. We have observed that abrogation of Notch-1 expression or signaling prevents hormone-induced adipocyte differentiation of 3T3-L1 cells. This effect correlated with down-regulation of PPARγ and PPARδ transcription factors. In addition to its role in the development and differentiation of the nervous system, Notch-1 has been shown to participate in the signaling events controlling the differentiation of Drosophila and mammalian cells of mesodermal origin (12Lindsell C.E. Shawber C.J. Boulter J. Weinmaster G. Cell. 1995; 80: 909-917Abstract Full Text PDF PubMed Scopus (540) Google Scholar, 41Corbin V. Michelson A.M. Abmayr S.M. Neel V. Alcamo E. Maniatis T. Young M.W. Cell. 1991; 67: 311-323Abstract Full Text PDF PubMed Scopus (193) Google Scholar). Our data are consistent with a requirement for Notch-1 in permitting terminal differentiation of mammalian mesodermal cells toward the adipocyte phenotype. We have taken into consideration the potential role of Notch-2, Notch-3, and Notch-4/Int-3 in the control of adipogenesis. 3T3-L1 cells express Notch-2 and Notch-3, but not Int-3 (data not shown). Antisense clones showed no diminished levels of Notch-2 expression. Notch-3, however, was diminished to some extent in the antisense clones (data not shown). Based on these data, a function for Notch-3 in the control of adipogenesis cannot be completely ruled out. Nevertheless, the biological agents used to block Notch-1 interactions are specific for Notch-1. In particular, the anti-Notch-1 antisera able to block differentiation cannot interact with Notch-2 or Notch-3. Taken together, these data indicate that the control of adipogenesis is a function specific of Notch-1, but not Notch-2 or Notch-3. Lindsell et al. (12Lindsell C.E. Shawber C.J. Boulter J. Weinmaster G. Cell. 1995; 80: 909-917Abstract Full Text PDF PubMed Scopus (540) Google Scholar) have proposed that the role of activated Notch-1 is to maintain the cells in an undifferentiated state. This is accomplished by inhibiting the expression or the activity of transcription factors of the helix-loop-helix family. Among these, the Achaete-Scute family of transcription factors inDrosophila and MyoD1 transcription factors have been shown to be targets of Notch action (14Kopan R. Nye J.S. Weintraub H. Development (Camb.). 1994; 120: 2385-2396Crossref PubMed Google Scholar, 15Kopan R. Schroeter E.H. Weintraub H. Nye J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1683-1688Crossref PubMed Scopus (425) Google Scholar, 42Campos-Ortega J.A. Knust E. Eur J. Biochem. 1990; 190: 1-10Crossref PubMed Scopus (37) Google Scholar, 43Heitzler P. Bourouis M. Ruel L. Carteret C. Simpson P. Development (Camb.). 1996; 122: 161-171PubMed Google Scholar). In addition, Notch-1 has been suggested to possess IκB activity (44Guan E. Wang J. Laborda J. Norcross M. Baeuerle P.A. Hoffman T. J. Exp. Med. 1996; 183: 2025-2032Crossref PubMed Scopus (115) Google Scholar). Notch-1, however, is expressed at high levels in certain terminally differentiated cells, such as neurons and T-lymphocytes (40Ellison L.W. Bird J. West D.C. Soreng A.L. Reynolds T.C. Smith S.D. Sklar J. Cell. 1991; 66: 649-661Abstract Full Text PDF PubMed Scopus (1454) Google Scholar, 45Ahmad I. Zaqouras P. Artavanis-Tsakonas S. Mech. Dev. 1995; 53: 73-85Crossref PubMed Scopus (84) Google Scholar). The permissive role of Notch-1 in adipocyte differentiation described here involves maintenance of the expression of transcription factors of the steroid/thyroid family. Interestingly, hormonal down-regulation of c-Myc, a transcription factor of the helix-loop-helix family whose overexpression inhibits adipocyte differentiation of 3T3-L1 cells (29Freytag S.O. Geddes T.J. Science. 1992; 256: 379-382Crossref PubMed Scopus (251) Google Scholar), is not affected in the antisense Notch-1 transfectants (data not shown). PPARγ is the most prominent player in the transcriptional control of adipogenesis. When activated by its ligands, PPARγ alone is able to drive the complete adipogenic process (24Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. Cell. 1995; 83: 803-812Abstract Full Text PDF PubMed Scopus (2740) Google Scholar, 25Kliewer S.A. Lenhard J.M. Willson T.M. Patel I. Morris D.C. Lehmann J.M. Cell. 1995; 83: 813-819Abstract Full Text PDF PubMed Scopus (1872) Google Scholar, 32Tontonoz P. Hu E. Spiegelman B.M. Cell. 1994; 79: 1147-1156Abstract Full Text PDF PubMed Scopus (3133) Google Scholar). In addition, other transcription factors involved in adipogenesis appear to act by increasing the expression of PPARγ (35Wu Z.D. Xie Y.H. Bucher N.L.R. Farmer S.R. Genes Dev. 1994; 9: 2350-2363Crossref Scopus (481) Google Scholar). PPARγ transcriptional effects are activated by fatty acid metabolites (24Forman B.M. Tontonoz P. Chen J. Brun R.P. Spiegelman B.M. Evans R.M. Cell. 1995; 83: 803-812Abstract Full Text PDF PubMed Scopus (2740) Google Scholar, 25Kliewer S.A. Lenhard J.M. Willson T.M. Patel I. Morris D.C. Lehmann J.M. Cell. 1995; 83: 813-819Abstract Full Text PDF PubMed Scopus (1872) Google Scholar) and the formation of heterodimeric units with retinoid acid receptors (26Keller H. Dreyer C. Medin J. Mahfoudi A. Ozato K. Wahli W. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 2160-2164Crossref PubMed Scopus (861) Google Scholar, 27Safonova I. Darimont C. Amri E.Z. Grimaldi P. Ailhaud G. Reichert U. Shroot B. Mol. Cell. Endocrinol. 1994; 104: 201-211Crossref PubMed Scopus (98) Google Scholar, 28Safonova I. Reichert U. Shroot B. Ailhaud G. Grimaldi P. Biochem. Biophys. Res. Commun. 1994; 204: 498-504Crossref PubMed Scopus (23) Google Scholar). Recently, it has been shown that growth factor inhibition of adipogenesis is mediated by mitogen-activated protein kinase-dependent phosphorylation of PPARγ (46Hu E. Kim J.B. Sarraf P. Spiegelman B.M. Science. 1996; 274: 2100-2103Crossref PubMed Scopus (941) Google Scholar). Thus, PPARγ expression and activation appear to be the key control components for driving the cells to the adipocyte phenotype. PPARγ is expressed preferentially in adipose tissue (47Tontonoz P. Hu E. Spiegelman B.M. Curr. Opin. Genet. & Dev. 1995; 5: 571-576Crossref PubMed Scopus (404) Google Scholar), and its expression increases during the process of adipogenesis (33Tontonoz P. Hu E. Graves R.A. Budavari A.I. Spiegelman B.M. Genes Dev. 1994; 8: 1224-1234Crossref PubMed Scopus (2005) Google Scholar, 59Chawla A. Schwarz E.J. Dimaculangan D.D. Lazar M.A. Endocrinology. 1994; 135: 798-800Crossref PubMed Scopus (617) Google Scholar). The factors that control the preferential expression of PPARγ in preadipocytes and adipocytes are unknown. Our data suggest that Notch-1 is involved in the control of PPARγ gene expression through mechanisms independent of the hormonal induction of the adipocyte differentiation process and therefore controls adipogenesis upstream of PPARγ itself. The signal transduction events in which Notch participates are the object of intense research (1Artavanis-Tsakonas S. Matsuno K. Fortini M.E. Science. 1995; 268: 225-232Crossref PubMed Scopus (1410) Google Scholar). The intracellular regions of the four Notch proteins contain six ankyrin/Cdc10 repeats, which appear to be necessary for signaling. The expression of truncated forms of some Notch family members spanning the ankyrin repeats has been implicated in the development of mammary tumors and T-cell leukemias (7Jhappan C. Gallahan D. Stahle C. Chu E. Smith G.H. Merlino G. Callahan R. Genes Dev. 1992; 6: 345-355Crossref PubMed Scopus (300) Google Scholar, 8Robbins J. Blondel B.J. Gallahan D. Callahan R. J. Virol. 1992; 66: 2594-2599Crossref PubMed Google Scholar, 40Ellison L.W. Bird J. West D.C. Soreng A.L. Reynolds T.C. Smith S.D. Sklar J. Cell. 1991; 66: 649-661Abstract Full Text PDF PubMed Scopus (1454) Google Scholar). Notch signaling may require processing by specific proteases that cleave the Notch intracellular region just distal to the transmembrane domain. The polypeptide released by this cleavage contains the ankyrin repeats (15Kopan R. Schroeter E.H. Weintraub H. Nye J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1683-1688Crossref PubMed Scopus (425) Google Scholar) as well as other regions that appear to be involved in signal cross-talk (48Axelrod J.D. Matsuno K. Artavanis-Tsakonas S. Perrimon N. Science. 1996; 271: 1826-1832Crossref PubMed Scopus (372) Google Scholar). According to this model, upon interaction of Notch with its ligands, the intracellular region is released to the cytoplasm, from where it migrates to the nucleus to inhibit the expression or the function of transcription factors (12Lindsell C.E. Shawber C.J. Boulter J. Weinmaster G. Cell. 1995; 80: 909-917Abstract Full Text PDF PubMed Scopus (540) Google Scholar, 15Kopan R. Schroeter E.H. Weintraub H. Nye J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1683-1688Crossref PubMed Scopus (425) Google Scholar, 44Guan E. Wang J. Laborda J. Norcross M. Baeuerle P.A. Hoffman T. J. Exp. Med. 1996; 183: 2025-2032Crossref PubMed Scopus (115) Google Scholar, 49Jarriault S. Brou C. Logeat F. Schroeter E.H. Kopan R. Israel A. Nature. 1995; 377: 355-358Crossref PubMed Scopus (1218) Google Scholar). These transcription factors, in turn, control the expression of genes involved in terminal differentiation. Binding of the intracellular region of Notch to the transcription factor Supressor of Hairless (Su(H)) in Drosophila or to its mammalian analog (CBF-1) has been reported to be necessary for Notch activity in some experimental models (1Artavanis-Tsakonas S. Matsuno K. Fortini M.E. Science. 1995; 268: 225-232Crossref PubMed Scopus (1410) Google Scholar, 49Jarriault S. Brou C. Logeat F. Schroeter E.H. Kopan R. Israel A. Nature. 1995; 377: 355-358Crossref PubMed Scopus (1218) Google Scholar). Notch signaling interacts with other signaling cascades that can be modulated by or that can modulate Notch signaling itself (48Axelrod J.D. Matsuno K. Artavanis-Tsakonas S. Perrimon N. Science. 1996; 271: 1826-1832Crossref PubMed Scopus (372) Google Scholar, 50Blair S.S. Science. 1996; 271: 1822-1823Crossref PubMed Scopus (17) Google Scholar, 51Hing H.K. Sun X. Artavanis-Tsakonas S. Mech. Dev. 1994; 47: 261-268Crossref PubMed Scopus (71) Google Scholar, 52Kuriyama M. Harada N. Kuroda S. Yamamoto T. Nakafuku M. Iwamatsu A. Yamamoto D. Prasad R. Croce C. Canaani E. Kaibuchi K. J. Biol. Chem. 1996; 271: 607-610Abstract Full Text Full Text PDF PubMed Scopus (187) Google Scholar, 53Miyamoto H. Nihonmatsu I. Kondo S. Ueda R. Togashi S. Hirata K. Ikegami Y. Yamamoto D. Genes Dev. 1995; 9: 612-625Crossref PubMed Scopus (103) Google Scholar, 54Rulifson E.J. Blair S.S. Development (Camb.). 1995; 121: 2813-2824PubMed Google Scholar). Whether the same type of mechanism is involved in the regulation of the transcription factors described in this report remains to be established. However, since we detect a decrease in PPARγ expression in 3T3-L1 cells prior to any differentiation treatment, it is unlikely that changes in Notch-1 expression may affect the hormone-induced signal transduction pathways that lead to adipocyte differentiation. Nonetheless, the carboxyl-terminal amino acid region of Drosophila Notch is involved in the cross-talk with Wingless (48Axelrod J.D. Matsuno K. Artavanis-Tsakonas S. Perrimon N. Science. 1996; 271: 1826-1832Crossref PubMed Scopus (372) Google Scholar), homolog to the mammalian proto-oncogene Wnt-1, which has been implicated in the growth factor-dependent regulation of mitogen-activated protein kinase (55Pan M.G. Wang Y.H. Hirsch D.D. Labudda K. Stork P.J. Oncogene. 1995; 11: 2005-2012PubMed Google Scholar). Since Ras and mitogen-activated protein kinase activation participate in hormone-induced adipogenesis (19Benito M. Porrás A. Nebreda A.R. Santos E. Science. 1991; 253: 565-568Crossref PubMed Scopus (141) Google Scholar, 21Porrás A. Muszynski K. Rapp U.R. Santos E. J. Biol. Chem. 1994; 269: 12741-12748Abstract Full Text PDF PubMed Google Scholar), Notch/Ras cross-talk may further participate in the regulation of differentiation. Notch-1 expression levels may influence the differentiation state of the cell and condition the interpretation of extracellular signals. A stoichiometric relationship between the expression of Notch-1 and its ligands has been shown to play a role in the development ofDrosophila. Thus, mutants expressing half the normal Notch levels can be rescued by another mutation in which Delta expression is also diminished in the same proportion (56Xu T. Rebay I. Fleming R.J. Scottgale T.N. Artavanis-Tsakonas S. Genes Dev. 1990; 4: 464-475Crossref PubMed Scopus (111) Google Scholar). This phenomenon could be at play in 3T3-L1 cells. Pref−/dlk, another member of the transmembrane EGF-like homeotic protein family, has been reported to play a possible role in cell-to-cell communication prior to adipocyte differentiation (57Lee Y.L. Helman L. Hoffman T. Laborda J. Biochim. Biophys. Acta. 1995; 1261: 223-232Crossref PubMed Scopus (55) Google Scholar, 58Smas C.M. Sul H.S. Cell. 1993; 73: 725-734Abstract Full Text PDF PubMed Scopus (564) Google Scholar). Commitment toward the adipocyte phenotype or the different morphologies produced by Notch-1 repression could depend on the levels of expression of Notch-1, its putative ligands, and the stoichiometric relationship between them. Although we failed to detect Notch-1 protein expression by Western blotting in all the 3T3-L1 antisense Notch-1-transfected clones, the different levels of Notch-1 mRNA expressed by these cells could impact on different Notch-1/ligand stoichiometric relationships related to the morphological and differentiation changes observed. Our results suggest that the levels of expression of Notch-1 are involved in the interpretation of extracellular signals. In this regard, the behavior of type E1 clones is representative. When treated to differentiate with the same hormones used to induce adipocyte differentiation, these cells undergo morphological changes toward a different fibroblast-like morphology. This shows that the cells did not become unresponsive to the signals, but rather, the same signals are interpreted in a different way. The requirement of Notch-1 function for the commitment of preadipocytes to differentiation shows that it actively participates in the acquisition of a new phenotype and suggests a novel function for Notch-1 in cell fate determination of mammalian cells. We thank Drs. Eugenio Santos and Steven Bauer for useful discussions and editorial comments.