Transactivation of the Moloney Murine Leukemia Virus and T-Cell Receptor Beta-Chain Enhancers by Cbf and Ets Requires Intact Binding Sites for Both Proteins

Dartmouth College Dartmouth Digital Commons

Open Dartmouth: Peer-reviewed articles by Dartmouth faculty Faculty Work

8-1995

Transactivation of the Moloney Murine Leukemia Virus and T-cell Receptor Beta-Chain Enhancers by cbf and ets Requires Intact Binding Sites for Both Proteins.

Wanwen Sun Dartmouth College

Barbara J. Graves University of Utah

Nancy A. Speck Dartmouth College

Follow this and additional works at: https://digitalcommons.dartmouth.edu/facoa

Part of the Medical Biochemistry Commons, Medical Cell Biology Commons, Medical Molecular Biology Commons, Virology Commons, and the Virus Diseases Commons

Dartmouth Digital Commons Citation Sun, Wanwen; Graves, Barbara J.; and Speck, Nancy A., "Transactivation of the Moloney Murine Leukemia Virus and T-cell Receptor Beta-Chain Enhancers by cbf and ets Requires Intact Binding Sites for Both Proteins." (1995). Open Dartmouth: Peer-reviewed articles by Dartmouth faculty. 1139. https://digitalcommons.dartmouth.edu/facoa/1139

This Article is brought to you for free and open access by the Faculty Work at Dartmouth Digital Commons. It has been accepted for inclusion in Open Dartmouth: Peer-reviewed articles by Dartmouth faculty by an authorized administrator of Dartmouth Digital Commons. For more information, please contact [email protected]. JOURNAL OF VIROLOGY, Aug. 1995, p. 4941–4949 Vol. 69, No. 8 0022-538X/95/$04.00ϩ0 Copyright 1995, American Society for Microbiology

Transactivation of the Moloney Murine Leukemia Virus and T-Cell Receptor ␤-Chain Enhancers by cbf and ets Requires Intact Binding Sites for Both Proteins

1 2 1 WANWEN SUN, BARBARA J. GRAVES, AND NANCY A. SPECK * Department of Biochemistry, Dartmouth Medical School, Hanover, New Hampshire 03755,1 and Department of Cellular, Viral and Molecular Biology, University of Utah School of Medicine, Salt Lake City, Utah 841322

Received 25 January 1995/Accepted 28 April 1995

The Moloney murine leukemia virus (Mo-MLV) enhancer contains binding sites (LVb and LVc) for the ets gene family of proteins and a core site that binds the polyomavirus enhancer-binding protein 2/core-binding factor (cbf) family of proteins. The LVb and core sites in the Mo-MLV enhancer contribute to its constitutive activity in T cells. All three binding sites (LVb, LVc, and core) are required for phorbol ester inducibility of the Mo-MLV enhancer. Adjacent binding sites for the ets and cbf proteins likewise constitute a phorbol ester response element within the human T-cell receptor ␤-chain (TCR␤) enhancer and contribute to constitutive transcriptional activity of the TCR␤ enhancer in T cells. Here we show that the CBF␣ subunit encoded by the mouse Cbfa2 gene (the murine homolog of human AML1) and three ets proteins, Ets-1, Ets-2, and GA-binding protein (GABP), transactivate both the Mo-MLV and mouse TCR␤ enhancer in transient-expression assays. Moreover, we show that transactivation by Cbf␣2 requires both intact ets and cbf binding sites. Transactivation by Ets-1, Ets-2, and GABP likewise requires intact binding sites for ets proteins and CBF. Supportive biochemical analyses demonstrate that both proteins can bind simultaneously to a composite enhancer element. These ﬁndings suggest that ets and cbf proteins cooperate in vivo to regulate transcription from the Mo-MLV and TCR␤ enhancers.

The combinatorial use of transcription factors has emerged served in the case of GA-binding protein (GABP), which is as an important theme in the regulation of eukaryotic gene composed of an alpha subunit with an ETS domain and a expression. A typical promoter and enhancer region has a non-DNA-binding beta subunit. Two alpha and two beta sub- diverse array of binding sites for factors. In some cases, the units form a stable tetramer with augmented DNA-binding complexity reflects the many alternative regulatory pathways activity (9, 76). In these two cases, the ets proteins have a that modulate the appropriate tissue-specific and developmen- specific partner. A domain has been mapped in Elk-1 and tal timing of a particular gene. However, in other cases, it has GABP␣ that is important in the protein-protein interaction, become apparent that factors cannot work effectively alone, and only ets proteins bearing that domain can interact with the but rather have cooperative modes of action. The simplest partner. For example, a subset of ets proteins bear the B do- form of this synergism is dimers, which occur as both hetero- main (e.g., Elk-1, SAP-1 [30], and NET [21]) and can interact and homodimeric combinations. This type of cooperation re- with SRF. In the case of GABP, only the GABP␣ subunit, and flects the requirement for a particular DNA-binding motif to no other ets proteins, can interact with GABP␤ (9). dimerize to acquire DNA-binding competence (e.g., the bZIP The partnership requirements of the ets proteins Ets-1 and and bHLH proteins). More intriguing are the cases in which the highly related Ets-2 are less well understood. Full-length two DNA-binding proteins that belong to two totally different Ets-1 contains an inhibitory domain that dampens its DNA- structural families cooperate to mediate transcriptional activa- binding potential (28, 47, 59, 85). Surprisingly, Ets-1 and Ets-2 tion. Although many cases of this type of cooperation have have been reported to cooperate with several different putative recently been reported (see, for example, references 8, 20, and partners in functional studies. Ets-1 works with Sp1 in activat- 31), in no case is the molecular basis for the synergism com- ing transcription from the human T-cell lymphotropic virus pletely understood. type 1 enhancer (18), while on the immunoglobulin M heavy- The ets family of transcription factors provides several ex- chain enhancer, Ets-1 cooperates with PU.1 (58). Ets-1 also amples of synergistic partnerships. The ets proteins contain a synergizes with the Jun-Fos heterodimer on the polyomavirus structural motif (winged helix-turn-helix) that is capable of binding DNA as a monomer (15, 46). However, in several enhancer (84). Myb and Ets-2 cooperation on the mim-1 pro- members of the family, the DNA-binding activity appears to be moter has been reported (16). Cooperative DNA binding that suboptimal in the absence of a partner protein. For example, would relieve the inhibition observed in the full-length Ets-1 the ets protein Elk-1 activates transcription from the c-fos protein has not been reported. Thus, the molecular basis of the promoter in conjunction with the DNA-binding protein serum transcriptional synergism is not known. response factor (SRF) (30). A specific protein-protein interac- In a preliminary study, the polyomavirus enhancer-binding tion stabilizes the binding of Elk-1 and SRF on the c-fos pro- protein 2/core-binding factor (CBF) (37, 81) has been shown to moter (32, 33, 65, 68). A different type of partnership is ob- enhance the binding of Ets-1 to the enhancer of the T-cell receptor (TCR) ␤ subunit gene (88). This putative interaction is particularly interesting, since there is a juxtaposition of cbf * Corresponding author. Phone: (603) 650-1159. Fax: (603) 650- and ets binding sites in a number of TCR gene enhancers, 1128. Electronic mail address: [email protected]. including the T-cell receptor alpha (TCR␣) and TCR␤ chain

4941 4942 SUN ET AL. J. VIROL.

FIG. 1. Maps of expression and reporter plasmid constructs. (A) Mo-MLV reporter plasmids used in cotransfection assays. The sequence at the bottom is from the promoter-proximal copy of the enhancer direct repeat. Binding sites for nuclear factors are shown in boxes (70). Point mutations in various binding sites are indicated in lowercase letters above the sequence. Mutations are in both copies of the direct repeat (72). (B) TCR␤ enhancer reporter plasmids. pTCR␤CAT contains the T␤3-T␤4 sequence from the mouse TCR␤ enhancer positioned upstream of the herpesvirus thymidine kinase (tk) promoter and CAT gene. The ets and core recognition sequences within T␤3 and T␤4 are shown below the sequence. pTCR␤(Ets)CAT contains mutations in both ets sites; pTCR␤(core)CAT contains mutations in all three core sites. (C) Cbf␣2 expression construct. A cDNA clone encoding a 451-amino-acid isoform of Cbf␣2 (Cbf␣2-451) was subcloned downstream from the CMV promoter in the vector pcDNA/Amp to yield pcDNA/Cbf␣2-451. (D) Expression constructs for ets proteins. The cDNA encoding each ets protein (Ets-1, Ets-2, Fli-1, Elf-1, PU.1, GABP␣, and GABP␤) was subcloned downstream of the D. melanogaster actin 5c promoter to yield pPac/ets plasmids.

enhancers (26, 34, 63). Interestingly, this arrangement also 61). The CBF␣ proteins share a conserved 128-amino-acid do- forms a highly conserved unit in the enhancers of type C main, called the Runt domain (36) because of its conservation in mammalian leukemia viruses (22). the Drosophila Runt protein (38). This domain mediates DNA Extensive biochemical and genetic studies of cbf and ets binding as well as dimerization to the CBF␤ subunit (36, 54). protein-binding sites have been performed on the enhancer of Cbfa1 and Cbfa2 are expressed in a restricted number of cell Moloney murine leukemia virus (Mo-MLV). Biochemical data types, including thymus (55, 67), whereas Cbfa3 and Cbfb are obtained with proteins expressed in bacteria or purified from expressed ubiquitously (45, 60, 82). It is anticipated that any of the thymus have established that CBF binds the Mo-MLV core site CBF␣ proteins could bind the Mo-MLV enhancer, and this has present in each enhancer repeat (70, 81). CBF proteins are het- been confirmed for the protein encoded by the Cbfa2 gene (12). erodimeric, with a DNA-binding subunit (CBF␣) and a non- Two binding sites for ets proteins, known as the LVb and DNA-binding subunit (CBF␤) (37, 60, 61, 82). Three distinct genes encode a family of CBF␣ subunits: Cbfa1 (AML3/Pebpa2a), LVc sites, flank each of the cbf binding sites on the Mo-MLV Cbfa2 (AML1/Pebpa2b), and Cbfa3 (AML2/Pebpa2c) (1–3, 45, 56, enhancer (Fig. 1). There are more than a dozen different ets 61, 86), while CBF␤ is encoded by an unrelated gene, Cbfb family proteins in vertebrates (39, 50, 83). The DNA-binding (Pebpb2) (1, 48, 60, 82). Any of the CBF␣ subunits encoded by the domain is sufficiently conserved that the majority of ets pro- three Cbfa genes can associate with the CBF␤ subunit (1, 2, 36, teins bind similar sites: a conserved GGA sequence centered in VOL. 69, 1995 ets AND cbf COOPERATION IN VIVO 4943 a 9-bp region with less well conserved flanking sequences (9, site of pBLCAT2 (49). All fragments were subcloned in the same orientation 17, 52, 59, 78, 87, 91). In T lymphocytes, in which the core site (617 to 735) relative to the herpesvirus thymidine kinase promoter, and the nucleotide sequence was confirmed. of Mo-MLV is functional, at least five different ets proteins are The eukaryotic expression vector pcDNA/Amp (Invitrogen) was used to ex- expressed (Ets-1, Ets-2, Fli-1, GABP, and Elf-1 [4, 5, 44, 48, press the CBF␣ subunit encoded by a cDNA derived from the murine Cbfa2 75]). In vitro binding data have shown that both of the GGA gene, which was isolated from a mouse thymus cDNA library (12). This cDNA motifs of the Mo-MLV enhancer can be recognized by Ets-1 encodes a 451-amino-acid polypeptide (Cbf␣2-451) identical to PEBP2␣B1 (1, 3). An EcoRI fragment containing the complete open reading frame was sub- (27, 59). More extensive analysis of the LVb site demonstrates cloned into pcDNA/Amp to give pcDNA/Cbf␣2-451. Expression of Cbf␣2-451 in that Ets-2, Fli-1, GABP, and an ets protein likely to be Elf-1 the pcDNA vector is from the cytomegalovirus (CMV) promoter. recognize this site (27). The ets proteins were expressed from the D. melanogaster actin 5c promoter- In addition to the biochemical data indicating that sites in based pPac expression vector (42). Murine cDNA clones bearing the open reading frames of Ets-1, Fli-1, Elf-1, GABP␣, and GABP␤ (amino acids 2 to 440, the enhancer bind cbf and ets proteins, several lines of genetic 1 to 452, 1 to 619, 1 to 454, and 1 to 380, respectively) were engineered with six evidence suggest that these sites are key components of the consecutive histidine codons (and a BamHI linker) at the N-terminal end of the Mo-MLV enhancer. First, as mentioned above, the LVb site open reading frame. Murine cDNAs of Ets-2 (encoding amino acids 1 to 468) and the core site are highly conserved in all mammalian C-type and PU.1 (1 to 271) spanning the entire open reading frames were ligated into the pPac vector in their native forms. retrovirus enhancers (22). Second, all three sites (LVb, core, Transfections and CAT assays. P19 cells were seeded at 7 ϫ 106 cells per and LVc) are required for transcriptional induction by 12-O- 60-mm dish 1 day before transfection. Transient transfections were performed tetradecanoylphorbol-13-acetate in T cells. Mutations that dis- with LipofectAmine reagent (BRL) following the manufacturer’s instructions. rupt the binding of ets and cbf proteins also attenuate induction Typically, 2 ␮g of the reporter plasmids and 0.1 to 8.0 ␮g of the pcDNA/Cbf␣2- 415 expression vector were mixed with 9 ␮l of LipofectAmine reagent for each (27, 72). Third, the Mo-MLV enhancer is preferentially active transfection. The total amount of expression vector DNA in each transfection in T cells (69), and mutations in the core site attenuate tran- was held constant at 10 ␮g by the addition of either the empty expression vector scription from both the Mo-MLV and the related SL3-MLV pBKCMV (Stratagene) or pcDNA/Amp, to ensure that the amount of CMV enhancer preferentially in T cells (6, 72, 77). Mutations in the promoter was equivalent in each transfection. After5hofincubation in serum- free medium containing the DNA-LipofectAmine mixture, serum was added to LVb site attenuate the constitutive activity of the Mo-MLV each dish to 10% final concentration. The cells were harvested 48 h after trans- enhancer in multiple cell types, including T cells (27, 72). fection. Cell lysates were prepared by freeze-thawing by standard protocols (40). Finally, evidence for the role of the core and LVb sites in Drosophila S3 cells were plated at 3 ϫ 106 per 25-cm2 flask and cotransfected transcription of the Mo-MLV enhancer in T cells is the dra- 24 h later with 5 ␮g of reporter plasmid and 0.5 to 8.0 ␮g of expression plasmid by calcium phosphate precipitation. The total amount of DNA was held constant matic effect of mutations on the disease specificity of Mo- (15 ␮g) by the addition of nonrecombinant pPac expression vector. Thus, the MLV. Mo-MLV causes T-cell leukemias in mice; mutations in concentration of Drosophila actin promoter was the same in each transfection. either the LVb or core site result in a significant shift in disease The DNA was mixed with 375 ␮l of 0.25 M CaCl2 and then added dropwise to specificity from T-cell leukemia to erythroid leukemia (71). 375 ␮lof2ϫHBS (3 mM Na2HPO4, 5 mM KCl, 140 mM NaCl, 25 mM HEPES [pH 7.1]). After 30 min of incubation at room temperature, the slightly cloudy In this report, we extend previous studies by using cDNAs mixture was added to the cells. The flasks were left undisturbed until the time of for cbf and ets proteins to study transactivation of the TCR␤ harvest 48 h later. Cell lysates were prepared by four rounds of freeze-thawing. and Mo-MLV enhancers. The data indicate that several ets and CAT activities were determined as described by Gorman et al. (24). Assay cbf proteins cooperate in vivo. Binding sites for both families of results were normalized to the total protein content of each lysate, and the mixtures were incubated with 0.1 ␮Ci of [14C]chloramphenicol (New England proteins are necessary for in vivo transactivation of the TCR␤ Nuclear). The conversion of [14C]chloramphenicol substrate to its mono- and and Mo-MLV enhancers by Ets-1, Ets-2, and GABP and by the diacetylated forms was monitored by thin-layer chromatography and quantified CBF␣ subunit encoded by the Cbfa2 gene. These findings by PhosphorImager scanning. provide important functional data to support future biochem- DNA-binding assays. Complementary 21-bp oligonucleotides containing the wild-type T␤3 sequence [T␤3(WT): ACAACAGGATGTGGTGGTACA], mu- ical studies. Furthermore, these studies establish the partner- tations in the ets binding site [T␤3(Ets): ACAACAttATGTGGTGGTACA], and ship of ets-cbf families as a model system with which to inves- mutations in the core site [T␤3(core): ACAACAGGATGTttTGGTACA] were tigate synergism between eukaryotic transcription factors. synthesized, annealed, and end labeled with [␥-32P]ATP. Electrophoretic mobility shift assays were performed as described previously (81). Protein expression and purification. A DNA fragment encoding a 17-kDa MATERIALS AND METHODS DNA-binding (Runt) domain from the mouse Cbf␣2-415 protein (amino acids 51 to 202) was amplified by PCR with the following primers: 5Ј (sense) primer, Cell culture. Mouse embryonic carcinoma cell line P19 (kindly provided by CGGAATTCCCATATGGTGGAGGTACTAGCT; 3Ј (antisense) primer, CG GATCCTTACTCCAATTCACTGAGCCG. The amplified fragment was di- Jack Lenz) was cultured at 37ЊCin5%CO2in Dulbecco’s modified Eagle’s medium (Bethesda Research Laboratories [BRL]) supplemented with 10% heat- gested with NdeI and BamHI and subcloned into the NdeI and BamHI sites of a inactivated fetal calf serum (Sigma), 100 U of penicillin per ml, 100 ␮gof modified pET3 vector lacking the codons of T7 gene 10 (59). The plasmid was transformed into Escherichia coli BL21(DE3) containing the pLysS plasmid (73). streptomycin (BRL) per ml, 2 mM L-glutamine (BRL), and 25 mM HEPES (N-2-hydroxyethylpiperazine-NЈ-2-ethanesulfonic acid, pH 7.4). The Drosophila Expression was induced with 0.4 mM IPTG (isopropylthiogalactopyranoside) for melanogaster cell line S3 (kindly provided by Edward Berger) was maintained in 3hat37ЊC to an optical density at 600 nm of 0.3 to 0.4 at the time of induction 1ϫ Drosophila Schneider medium (BRL) supplemented with 10% heat-inacti- and 1 to 1.5 at the time of harvest. The bacterial cell pellet was disrupted by three vated fetal calf serum at 25ЊC. freeze-thaw cycles in lysis buffer (50 mM Tris [pH 8.1] plus protease inhibitors: Plasmids. The Mo-MLV reporter plasmids containing wild-type and mutated 1 mM pefablock, 1 ␮g of leupeptin per ml, 2 ␮g of aprotinin per ml, and 1 ␮gof Mo-MLV enhancers (a 384-bp fragment from the Mo-MLV U3/R region) linked pepstatin per ml; the volume was 1/10 the volume of original bacterial culture) to the bacterial chloramphenicol acetyltransferase (CAT) gene were described and sheared with a 22-gauge needle, and the soluble protein fraction was col- previously (72). Reporter plasmids containing the T␤3 and T␤4 elements from lected following centrifugation (16,000 ϫ g, 15 min, 4ЊC) (11). the murine TCR␤ enhancer (25) were constructed by PCR. Three sets of primers The amino-terminal deletion mutant polypeptide of murine Ets-1 (⌬N331) with BamHI overhangs were used to amplify the wild-type T␤3-T␤4 sequence was produced in bacteria, and pure protein was prepared as described previously covering nucleotides 617 to 735 of the mouse TCR␤ enhancer (43) and to (15). introduce mutations into either ets or cbf binding sites [pTCR␤CAT, pTCR␤(Ets)CAT, and pTCR␤(core)CAT] (Fig. 1B). The sequences of the prim- RESULTS ers are as follows (lowercase letters indicate the mutated nucleotides): TCR␤(wt): sense, 5Ј-CGGGATCCACAACAGGATGTGGTTTG-3Ј, and antisense, 5Ј-CGGGATCCTTGAAGACAGGATGTGGC-3Ј (the Ets core is under- CBF stimulates transcription from the Mo-MLV and TCR␤ lined); TCR␤(Ets): sense, 5Ј-CGGGATCCACAACAttATGTGGTTTGAC-3Ј, enhancers. CBF was first tested for its ability to transactivate and antisense, 5Ј-CGGGATCCTTGAAGACAttATGTGGCAAGTGTGG-3Ј; the Mo-MLV and TCR␤ enhancers in a transient-cotransfec- and TCR␤(core): sense, 5Ј-CGGGATCCACAACAGGATGTttTTTGACAT-3Ј, tion assay in the P19 mouse embryonic carcinoma cell line. A and antisense, 5Ј-CGGGATCCTTGAAGACAGGATGTttCACGTGTttTTCC CAAAATGC-3Ј. cDNA encoding only an ␣ subunit of CBF was used. Neither The PCR products were digested with BamHI and subcloned into the BamHI the Cbfa1 (Pebpa2a) nor Cbfa2 (AML1/Pebpa2b) gene is ex- 4944 SUN ET AL. J. VIROL.

was approximately six- to eightfold on both reporter constructs. Transactivation peaked at 2 to 4 ␮g of input Cbf␣2-451 expression vector and declined at higher doses, suggesting that excess CBF␣2-451 protein squelched some necessary compo- nent of the transcription apparatus (1, 64). Several ets proteins, Ets-1, Ets-2, GABP␣/␤, Fli-1, PU.1, and Elf-1, were also tested for their ability to transactivate pMoCAT and pTCR␤CAT. None of the ets proteins stimulated transcription from either reporter construct in P19 cells, either alone or in conjunction with Cbf␣2-451 (data not shown). Our interpretation of this negative result is that one or more endogenous ets proteins in P19 cells may be present in saturating amounts. Both ets and core motifs are required for transcriptional activation of the Mo-MLV and TCR␤ enhancers by CBF. To identify the cis-acting sequences in the Mo-MLV enhancer required for transactivation by CBF, wild-type and mutated reporter constructs that contained base substitutions in indi- vidual protein-binding sites were used in transactivation assays (Fig. 1). Disruption of the binding site for CBF [pMo(core) CAT] resulted in loss of transactivation by Cbf␣2-451, as ex- pected (Fig. 3). Mutation of one of the adjacent binding sites for the ets proteins, the LVb site [pMo(LVb)CAT], also dis- FIG. 2. Cbf␣2-451 transactivates the Mo-MLV and TCR␤ enhancers in a dose-responsive manner. The CBF expression vector pcDNA/Cbf␣2-451 was rupted transactivation by Cbf␣2-451. In contrast, mutation of cotransfected with 2 ␮g of either the pMoCAT (F) or pTCR␤CAT (E) reporter other protein-binding sites, such as the GRE (binding site for plasmid into P19 cells. Transcription activity of reporter plasmids was deter- the glucocorticoid hormone receptor and the bHLH protein mined by CAT assays as described in Materials and Methods. The fold activation SEF2 [10, 14]), the nuclear factor 1 (NF1) binding site (66), was calculated as the CAT activity induced by transfection of the pcDNA/Cbf␣2- 451 expression vector relative to the CAT activity induced by transfection with an and another ets binding site, LVc, did not affect transactivation empty expression vector. Three independent experiments were conducted. The by Cbf␣2-451. Thus, it appears that transactivation of the Mo- results from one experiment performed with duplicate transfections are plotted MLV enhancer by Cbf␣2-451 requires some protein endoge- (means Ϯ standard deviation). The maximal transactivation varied from 5- to 7-fold on the pMoCAT reporter and 6- to 10-fold on the pTCR␤CAT reporter nous to P19 cells that associates with the LVb site. Transacti- in independent experiments, and the amount of pcDNA/Cbf␣2-451 expression vation by Cbf␣2-451 does not, however, require the other ets vector that induced maximal transactivation varied between 2 and 4 ␮g. binding site in the Mo-MLV enhancer, the LVc site. This is consistent with previous results demonstrating that the LVc site has no role in constitutive expression from the Mo-MLV pressed in P19 cells (1, 61). The ␤ subunit of CBF is expressed enhancer in T cells (27, 72). ubiquitously, and presumably this polypeptide was provided by Transactivation of the TCR␤ enhancer by Cbf␣2-451 also the P19 cells. Consistent with this hypothesis, coexpression of required intact cbf and ets binding sites; mutations in either of both CBF subunits transactivated the Mo-MLV and TCR␤ these sites resulted in decreased levels of transactivation (Fig. enhancers to the same extent as expression of the CBF␣ sub- 4). The similar results of the mutational analysis in both the unit alone (data not shown). The CBF␣ subunit that was used Mo-MLV and TCR␤ enhancers highlight the cbf-ets cassette as for these analyses is one of three isoforms encoded by the a distinct functional unit within these multicomponent enhanc- Cbfa2 gene (1, 3, 12). This isoform, termed Cbf␣2-451, con- ers. tains 451 amino acids, has a predicted molecular mass of 48.6 Both ets and cbf binding sites are required for ets proteins to kDa, and is identical to PEBP2␣B1 (1, 3). transactivate the TCR␤ enhancer in Drosophila S3 cells. Be- Cbf␣2-451 stimulated transcription from both the pMoCAT cause the ets proteins did not transactivate either the Mo-MLV and pTCR␤CAT reporter constructs (Fig. 2; see also Table 1 or TCR␤ enhancer in mouse P19 cells, we chose a second cell for data summary). The maximum stimulation by Cbf␣2-451 line for further analysis, Drosophila S3 cells. Enhancers bearing both wild-type and mutant versions of the ets binding sites were tested in transactivation assays (Fig. 5). Only transactivation TABLE 1. Summary of positive transactivation results that depended upon intact ets sites was interpreted. Ets-1 stim- Reporter activitya ulated transcription from the TCR␤ enhancer at least 15-fold in S3 cells, Ets-2 gave a 5-fold activation, and GABP␣/␤ trans- Transactivator Cell line Mo-MLV TCR␤ activated the TCR␤ enhancer approximately 3-fold. Note that WT Core ets WT Core ets the total level of transactivation of the TCR␤ enhancer by GABP␣/␤ was 16-fold, but that much of this transactivation pcDNA/Cbf␣2-451 P19 ϩb ϪϪϩϪϪ was independent of the mutated GGA sequences [Fig. 5C, pPac/Ets-1 S3 ϩϪϪ compare pTCR␤CAT and pTCR␤(Ets)CAT]. Elf-1-depen- pPac/Ets-2 S3 ϩϪϪ dent transactivation was not at all dependent on the ets sites, pPac/GABP␣/␤ S3 ϩϪϪ and Fli-1 did not transactivate pTCR␤CAT (data not shown). aCAT activity: ϩ, transactivator induced a level higher than in controls with The activity of the ets proteins on the Mo-MLV enhancer empty expression vector; Ϫ, no change relative to controls. The core and ets was tested in S3 cells. The extent of transactivation was much reporters were mutated in the core or ets binding site, as indicated. WT, wild less than that observed with the TCR enhancer and was not type. ␤ b Transactivation of Mo-MLV enhancers containing mutations in LVc, NF1, reproducible (data not shown). We suspect that higher basal or GRE sites was equivalent to that of the wild-type enhancer. transcription from the Mo-MLV enhancer than from the VOL. 69, 1995 ets AND cbf COOPERATION IN VIVO 4945

proposal is consistent with the observations that the Drosophila Runt/Brother heterodimer binds the same consensus core site as the mammalian CBF complex (19, 36). Several additional cell lines (Rat-2, NIH 3T3, HeLa, and SL2) were tested for their ability to support transactivation of the Mo-MLV and TCR␤ enhancers by Cbf␣2-451 and the ets proteins. No cell line was found in which both Cbf␣2-451 and the ets expression vectors, independently or together, would direct transactivation of these enhancers. Therefore, we were unable to perform reconstitution experiments to examine synergistic activation by Cbf␣2-451 and speciﬁc ets proteins. DNA-binding domains of Ets-1 and Cbf␣2 bind simultaneously to the TCR␤ enhancer. The ets and cbf binding sites in the Mo-MLV and TCR␤ enhancers are very closely aligned. They are particularly close in the TCR␤ enhancer, with only a 5-bp center-to-center distance between the two sites. One possible explanation for the observation that mutating the ets site disrupts transactivation by Cbf␣2-451 is that the mutation also disrupts CBF binding. Reciprocally, the inability of Ets-1, Ets-2, or GABP␣/␤ to transactivate the TCR␤ enhancer with mutated cbf binding sites could be caused by their inability to bind. To exclude these trivial explanations, simultaneous binding of Ets-1 and Cbf␣2 to wild-type and mutated TCR␤ enhancers was demonstrated.

FIG. 3. Transactivation of the Mo-MLV enhancer by Cbf␣2-451 requires both intact core and LVb sites. The pcDNA/Cbf␣2-451 expression plasmid was cotransfected with 2 ␮g of the indicated Mo-MLV reporter plasmids in P19 cells, and transactivation relative to the reporter plasmids alone was plotted as described in the legend to Fig. 2. The letters in parentheses indicate the binding site that is mutated [e.g., pMo(LVb)CAT contains a 2-bp mutation in each LVb site in both copies of the enhancer direct repeat; pMo(NF1)CAT contains mutations in both the NF1(a) and NF1(b) sites (see Fig. 1)]. A dose-response curve was generated for each reporter construct; the data shown are from experiments performed with 4 ␮g of pcDNA/Cbf␣2-451. Four independent experiments were performed; data illustrate one representative experiment.

TCR␤ enhancer made it more difﬁcult to transactivate by addition of exogenous ets proteins. The role of the cbf binding sites in the transactivation by several ets proteins in Drosophila S3 cells was also tested. Mutation of the cbf binding sites in the TCR␤ enhancer [pTCR␤(core)CAT] reduced ets-dependent transactivation to the same level as mutation of the ets binding sites (Fig. 5). These ﬁndings suggest that transactivation by Ets-1, Ets-2, and GABP␣/␤ requires interaction with endogenous Drosophila homologs of the CBF␣ and ␤ subunits, called Runt and FIG. 4. Transactivation of the TCR␤ enhancer by Cbf␣2-451 also requires Brother, respectively (13, 19, 38, 61). intact binding sites for ets and cbf proteins. P19 cells were cotransfected with 4 ␮g Cbf␣2-451 did not transactivate expression from either the of pcDNA/Cbf␣2-451 and 2 ␮g of wild-type (pTCR␤CAT) or mutated Mo-MLV or TCR␤ enhancer in S3 cells. As in the case of the [pTCR␤(Ets)CAT and pTCR␤(core)CAT] reporter constructs. Four independent experiments were performed; transfections from one representative exper- P19 cells and endogenous ets proteins, we propose that S3 cells iment are shown graphically on top and numerically in the table below. Analyses contain saturating levels of an endogenous CBF homolog. This were performed as described in the legend to Fig. 2. 4946 SUN ET AL. J. VIROL.

FIG. 5. Both ets and cbf sites are necessary for ets proteins to transactivate the TCR␤ enhancer. Five micrograms of the indicated TCR␤ enhancer reporter plasmids was cotransfected with 2 ␮g of (A) Ets-1, (B) Ets-2, or (C) GABP␣ plus GABP␤ expression constructs. Four independent transfections were performed in the D. melanogaster cell line S3, and the fold activation relative to the reporter construct plus empty pPac expression vector for one representative experiment is shown. Data were collected and presented as described in the legend to Fig. 2.

Truncated versions of Ets-1 and Cbf␣2 which retain the and 2). Adding both Ets-1⌬N331 and CBF␣2RD generates a DNA-binding domains and minimal ﬂanking sequences (Ets- third shifted complex (Fig. 6, lane 3). This complex migrates 1⌬N331 and CBF␣2RD, respectively) were tested for binding more slowly than either the Ets-1⌬N331 or CBF␣2RD protein- to the T␤3 site (25) in the TCR␤ enhancer by the mobility shift DNA complex, suggesting that it represents the simultaneous assay. Both Ets-1⌬N331 and CBF␣2RD form distinct DNA- binding of both proteins. protein complexes on the wild-type T␤3 probe (Fig. 6, lanes 1 The mutations introduced into the ets site abolished the association of Ets-1⌬N331 (Fig. 6, lane 4). Importantly, despite its proximity to the cbf binding site, the ets site mutation did not disrupt binding by CBF␣2RD (Fig. 6, lane 5). Similarly, mutations in the cbf binding site abolished CBF␣2RD but not Ets-1⌬N331 binding (Fig. 6, lanes 6 and 7). Therefore, the inability of Cbf␣2-451 to stimulate transcription from enhancers lacking intact ets sites and of the ets proteins to transactivate an enhancer with core site mutations was not due simply to their inability to bind to the enhancers.

DISCUSSION Transactivation experiments in P19 cells demonstrate that a CBF␣ subunit encoded by the mouse Cbfa2 gene (Cbf␣2-451) stimulates transcription from both the Mo-MLV and TCR␤ chain enhancers in vivo. Transactivation by Cbf␣2-451 is de- FIG. 6. Binding of Ets-1 and Cbf␣2 to wild-type and mutated T␤3 sites in the TCR␤ enhancer. The DNA-binding activity of Ets-1⌬N331 and CBF␣2RD was pendent on a nearby GGA sequence in both enhancers. Mu- tested in the mobility shift assay on T␤3(WT), T␤3(Ets), and T␤3(core) probes, tation of both guanines in the GGA sequence does not disrupt as indicated. Reaction mixtures contained approximately 4 ng of puriﬁed Ets- Cbf␣2-451 binding in vitro. These data strongly suggest that 1⌬N331 or 200 ng of extract prepared from bacteria expressing CBF␣2RD transactivation by Cbf␣2-451 requires an endogenous protein protein and less than 10 fmol of radiolabeled probe. Lanes: 1, 4, and 6, Ets- 1⌬N331; 2, 5, and 7, CBF␣2RD; 3, Ets-1⌬N331 and CBF␣2RD. The positions of in P19 cells whose binding is dependent on one or both of these protein-DNA complexes are indicated by arrowheads. guanines. VOL. 69, 1995 ets AND cbf COOPERATION IN VIVO 4947

Several types of data suggest that ets proteins are the most TABLE 2. Nucleotide sequence preferences of ets proteinsa likely candidates for the Cbf␣2-451 accessory protein. First, Nucleotide(s) at position: the ets proteins are the only known proteins that bind the GGA Protein sequence in both the TCR␤ and the Mo-MLV enhancers. A 123456789 non-ets protein, the mammalian C-type retrovirus enhancer Consensus N N N G G A N N N factor 1 (MCREF-1), also binds the LVb site in the Mo-MLV Ets-1 A, g C C, a G G A A, T G, a C, T enhancer (51, 74). However, a site for MCREF-1 (consensus: GABP␣ G, A C, g C, a G G A A, t G, a T, c CNGGN6CNGG) does not overlap the GGA motif in the Fli-1 A C C G G A A G, a T, c TCR␤ enhancer fragment. The distance between the GGA E74/Elf-1 C, T C C, a G G A A* G, a T motif and the core site is not conserved in the TCR␤ and Mo-MLV enhancers, and this variability is also consistent with T␤3/T␤4 A C A GGAT G T binding by an ets protein. In fact, the center-to-center distance aOptimal sequences for Ets-1 (59, 87), GABP␣ (9), and Fli-1 (52) were from the GGA sequence to the core site is 5 bp further in the determined by selected and amplified binding site analysis. The E74/Elf-1 se- Mo-MLV enhancer than in the TCR␤ enhancer, which would quence was determined for E74, a closely related Drosophila homolog of Elf-1 position binding proteins on different sides of the B-form DNA (79). Two uppercase letters at a position indicate approximately equal selection; a lowercase letter indicates a less-favored nucleotide. The sequence in the TCR␤ helix in the two enhancers. This is consistent with in vitro enhancer is listed at the bottom for comparison. The asterisk at position 7 for binding data that show the insensitivity of ets proteins to sub- E74/Elf-1 indicates that binding is restricted to sites with an A at this position stantial changes in the orientation and spacing between their (80). binding sites and those of their accessory proteins. For example, the spacing between the binding sites for Elk-1 and the SRF has no affect on the ability of these two proteins to form GGA sequences containing eitheraToranAatposition 7 (59, a ternary complex (78). Cooperative binding between Ets-1 87). GABP␣ binds to both GGAA and GGAT sequences but and CBF also has been shown to be indifferent to spacing prefers a GGAA site (9), and Fli-1 has an even greater pref- between the ets and cbf sites (88). Possible candidates for ets erence for GGAA sequences but can bind to GGAT (52). Elf-1 family members in P19 cells include PEA3 and Ets-2, both of is unusual among the mammalian ets proteins insofar as it which are expressed in undifferentiated embryonal carcinoma binds only GGAA sequences (7, 80). These predicted affinities cells (53, 90). correlate with the relative abilities of ets proteins to transacti- In another set of studies, transactivation of the TCR␤ en- vate the TCR␤ enhancer; Ets-1, Ets-2, and GABP transacti- hancer by the ets proteins Ets-1, Ets-2, and GABP␣/␤ was vate with the ranking Ets-1 Ͼ Ets-2 Ͼ GABP, while Fli-1 and demonstrated. Importantly, the activity of these ets proteins in Elf-1 show no transactivation. Ets-1, GABP, and Elf-1 have Drosophila S3 cells is dependent on the integrity of the adja- been reported to be abundant GGA-binding activities in T-cell cent cbf binding site. We speculate that the ets proteins are extracts (27, 89). Taken together, these data predict that Ets-1 cooperating with an endogenous Drosophila cbf homolog to and GABP contribute significantly to transcription from the activate transcription. The obvious candidate for this core- Mo-MLV and TCR␤ enhancers in T cells and do so in collab- binding factor is Runt, the Drosophila protein that is the found- oration with CBF. ing member of the CBF␣ protein family (36, 38), or a partially From the data presented here and by analogy to the well- characterized runt homolog (19). characterized Elk-1 and SRF interaction, we propose the fol- These functional studies indicate that cbf and ets proteins lowing mechanism for CBF cooperation with Ets-1, Ets-2, and cooperate to activate transcription from the Mo-MLV and GABP. Each of these intact ets proteins has attenuated DNA- TCR␤ enhancers in vivo. Although it is certainly possible that binding activity which can be enhanced by deletion of amino- other, as yet unidentified, proteins may be cooperating with terminal sequences (28, 47, 59, 76, 85). Quantitative analyses CBF or the ets proteins on these two enhancers, the most have shown that the low affinity of Ets-1 DNA binding is due straightforward explanation is that one or more members of to instability of the DNA-protein complexes (35). This has also the ets family are cooperating with one or more members of been reported for GABP (76). Thus, we propose that CBF the cbf family of proteins. stabilizes DNA-bound Ets-1, Ets-2, and GABP. Second, we Three ets proteins, Ets-1, Ets-2, and GABP, activated tran- predict that the cbf and ets proteins each contain a domain that scription from the TCR␤ enhancer in Drosophila S3 cells. is responsible for the cooperative interaction. A candidate However, the ets protein Fli-1 did not transactivate the TCR␤ domain for the ets proteins Ets-1, Ets-2, and GABP␣ is a enhancer in Drosophila cells. Elf-1 transactivated the TCR␤ region of high conservation in the amino-terminal half of these enhancer, but transactivation did not require intact ets binding proteins, known as the pointed domain or putative HLH re- sites in the T␤3 and T␤4 elements. Thus, activation by Elf-1 gion (23, 41). There are two possible locations for an interac- was presumably mediated by Elf-1 binding sites elsewhere on tive domain in CBF: either the DNA-binding domain of the the pTCR␤CAT construct. Although we have not tested for CBF␣ subunit, which is highly conserved between Runt and variable expression of the ets proteins that could account for the mammalian CBF␣ proteins, or sequences in the non-DNA- these differences in transactivation activity, we were struck by binding CBF␤ subunit that are conserved between the mam- the following interesting correlation that could also explain malian CBF␤ and Drosophila Brother proteins (19). Cross- these differences: the ability of the ets proteins to transactivate species interactions have been reported for Elk-1, in that Elk-1 the TCR␤ enhancer roughly correlated with their predicted interacts with the conserved DNA-binding domain of both binding affinity for the T␤3 and T␤4 sites. The ‘‘selected’’ mammalian SRF and its yeast homolog MCM1 (57). Finally, consensus sequences for Ets-1, GABP␣, Fli-1, and Elf-1 bind- since cooperative binding between Ets-1 and CBF is relatively ing sites are shown in Table 2. The predicted order of binding indifferent to the spacing between ets and core sites (88), we affinity for the TCR␤ ets sites is Ets-1 ϭ Ets-2 Ͼ GABP␣Ͼ predict that the putative CBF interaction domain in Ets-1 is Fli-1 ϾϾ Elf-1. The most critical nucleotide in this prediction connected to the DNA-binding ETS domain by a flexible hinge is that at position 7, which is a T residue in both the T␤3 and linker. Preliminary data for such a hinge region have recently T␤4 sites. Ets-1 (and presumably Ets-2, which displays 95% been obtained from structural studies of Ets-1 polypeptides amino acid identity in the ETS domain) binds equally well to (62). 4948 SUN ET AL. J. VIROL.

One conundrum that will direct future studies is the appar- or together to specify functionally distinct activities at a single transcriptional ent promiscuity of Ets-1 and CBF for different partners. Ets-1 enhancer. Mol. Cell. Biol. 6:993–1001. 15. Donaldson, L. W., J. M. Petersen, B. J. Graves, and L. P. McIntosh. 1994. has now been reported to interact with five unrelated proteins: Secondary structure of the ETS domain places murine Ets-1 in the super- Jun-Fos, Sp1, PU.1, TFE-3, and CBF (18, 34, 58, 84, 88; this family of winged helix-turn-helix DNA-binding proteins. Biochemistry 33: study). This promiscuity is difficult to reconcile with the hy- 13509–13516. pothesis that these interactions are mediated by a specific 16. Dudek, H., R. V. Tantravahi, V. N. Rao, E. S. Reddy, and E. P. Reddy. 1992. Myb and Ets proteins cooperate in transcriptional activation of the mim-1 domain on the Ets-1 protein. Does Ets-1 contain a separate promoter. Proc. Natl. Acad. Sci. USA 89:1291–1295. interaction domain for each of these proteins or one interac- 17. Fisher, R. J., G. Mavrothalassitis, A. Kondoh, and T. S. Papas. 1991. High- tion domain that binds to all of these proteins? The same affinity DNA-protein interactions of the cellular ETS1 protein: the determi- problem exists for CBF, which has now been reported to re- nation of the ETS binding motif. Oncogene 6:2249–2254. 18. Gegonne, A., R. Bosselut, R. Bailly, and J. Ghysdael. 1993. Synergistic quire two different accessory proteins to activate transcription: activation of the HTLV1 LTR Ets-responsive region by transcription factors an ets protein on the TCR␤ chain enhancer, and c-Myb on the Ets1 and Sp1. EMBO J. 12:1169–1178. TCR␦ enhancer (29). As soon as adequate amounts of pure, 19. Gergen, J. P. 1994. Personal communication. full-length Cbf␣2 subunit are available for biochemical analy- 20. German, M. S., J. Wang, R. B. Chadwick, and W. J. Rutter. 1992. Synergistic activation of the insulin gene by a LIM-homeo domain protein and a basic ses, the experiments necessary to resolve these questions and helix-loop-helix protein: building a functional insulin minienhancer complex. determine the molecular basis of the cbf-ets partnership can be Genes Dev. 6:2165–2176. performed. 21. Giovane, A., A. Pintzas, S. Maira, P. Sobieszczuk, and B. Wasylyk. 1994. Net, a new ets transcription factor that is activated by Ras. Genes Dev. 8:1502– 1513. ACKNOWLEDGMENTS 22. Golemis, E., N. A. Speck, and N. Hopkins. 1990. Alignment of U3 region sequences of mammalian type C viruses: identification of highly conserved We thank Barbara Crute and Amy Lewis for providing the bacterial motifs and implications for enhancer design. J. Virol. 64:534–542. extracts of CBF␣2RD; Jeannine Petersen for bacterial extracts of 23. Golub, T. R., G. F. Barker, M. Lovett, and D. G. Gilliland. 1994. Fusion of Ets-1⌬N331; and Jeff Leiden, Richard Maki, and Steve McKnight for PDGF receptor ␤ to a novel ets-like gene, tel, in chronic myelomonocytic providing the Elf-1, Ets-2, Fli-1, and GABP␣ and ␤ cDNA clones. We leukemia with t(5;15) chromosomal translocation. Cell 77:307–316. also thank Jacqui Wittmeyer and Chenghua Luo for help in cloning 24. Gorman, C. M., L. F. Moffat, and B. H. Howard. 1982. Recombinant ge- several of the ets expression plasmids and Terryl Stacy for advice and nomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2:1044–1051. help with tissue culture and for critically reading the manuscript. 25. Gottschalk, L. R., and J. M. Leiden. 1990. Identification and functional This work was supported by Public Health Service grants CA58343 characterization of the human T-cell receptor ␤ gene enhancer: common (awarded to N.A.S.) and GM38663 (awarded to B.J.G.) and Cancer nuclear proteins interact with the transcriptional regulatory elements of the Center Support grants (CA23018) and (CA42014) to the Norris Cot- T-cell receptor ␣ and ␤ genes. Mol. Cell. Biol. 10:5486–5495. ton Cancer Center and University of Utah Cancer Center, respectively. 26. Grosschedl, R. 1994. Personal communication. March of Dimes Birth Defect Foundation grant 1-10138 to B.J.G. is 27. Gunther, C. V., and B. Graves. 1994. Identification of ETS domain proteins also acknowledged. N.A.S. is a Leukemia Society of America Scholar, in murine T lymphocytes that interact with the Moloney murine leukemia and W.S. was an awardee of a Predoctoral Fellowship from the Norris virus enhancer. Mol. Cell. Biol. 14:7569–7580. 28. Hagman, J., and R. Grosschedl. 1992. An inhibitory carboxyl-terminal do- Cotton Cancer Center at the Mary Hitchcock Hospital. main in Ets-1 and Ets-2 mediates differential binding of ETS family members to promoter sequences of the mb-1 gene. Proc. Natl. Acad. Sci. USA 89: REFERENCES 8889–8893. 1. Bae, S. C., E. Ogawa, M. Maruyama, H. Oka, M. Satake, K. Shigesada, N. A. 29. Hernandez-Munain, C., and M. S. Krangel. 1994. Regulation of the T-cell Jenkins, D. J. Gilbert, N. G. Copeland, and Y. Ito. 1994. PEBP2␣B/mouse receptor ␦ enhancer by functional cooperation between c-Myb and core- AML1 consists of multiple isoforms that possess differential transactivation binding factors. Mol. Cell. Biol. 14:473–483. potentials. Mol. Cell. Biol. 14:3242–3252. 30. Hill, C. S., R. Marais, S. John, J. Wynne, S. Dalton, and R. Treisman. 1993. 2. Bae, S.-C., E. Takahashi, Y. W. Zhang, E. Ogawa, K. Shigesada, Y. Namba, Functional analysis of a growth factor-responsive transcription factor com- M. Satake, and Y. Ito. Cloning, mapping and expression of PEBP2␣C, a third plex. Cell 73:395–406. gene encoding the mammalian Runt domain. Gene, in press. 31. Jackson, D. A., K. E. Rowader, K. Stevens, C. Jiang, P. Milos, and K. Zaret. 3. Bae, S. C., Y. Yamaguchi-Iwai, E. Ogawa, M. Maruyama, M. Inuzuka, H. 1993. Modulation of liver-specific transcription by interactions between hep- Kagoshima, K. Shigesada, M. Satake, and Y. Ito. 1993. Isolation of atocyte nuclear factor 3 and nuclear factor 1 binding DNA in close apposi- PEBP2␣B cDNA representing the mouse homolog of human acute myeloid tion. Mol. Cell. Biol. 13:2401–2410. leukemia gene, AML1. Oncogene 8:809–814. 32. Janknecht, R., and A. Nordheim. 1992. Elk-1 protein domains required for 4. Ben-David, U., E. B. Giddens, K. Letwin, and A. Bernstein. 1991. Erythro- direct and SRF-assisted DNA-binding. Nucleic Acids Res. 20:3317–3324. leukemia induction by Friend murine leukemia virus: insertional activation 33. Janknecht, R., R. Zinck, W. H. Ernst, and A. Nordheim. 1994. Functional of a new member of the ets gene family, Fli-1, closely linked to c-ets-1. Genes dissection of the transcription factor Elk-1. Oncogene 9:1273–1278. Dev. 5:908–918. 34. Jones, K. A. 1994. Personal communication. 5. Bhat, N. K., F. J. Fisher, S. Fujiwara, R. Ascione, and T. S. Papas. 1987. 35. Jonsen, M., and B. J. Graves. 1994. Unpublished data. Temporal and tissue-specific expression of mouse ets genes. Proc. Natl. 36. Kagoshima, H., K. Shigesada, M. Satake, Y. Ito, H. Miyoshi, M. Ohki, M. Acad. Sci. USA 84:3161–3165. Pepling, and J. P. Gergen. 1993. The Runt-domain identifies a new family of 6. Boral, A. L., S. A. Okenquist, and J. Lenz. 1989. Identification of the SL3-3 virus heteromeric DNA-binding transcriptional regulatory proteins. Trends enhancer core as a T-lymphoma cell-specific element. J. Virol. 63:76–84. Genet. 9:338–341. 7. Bosselut, R., J. Levin, E. Adjadj, and J. Ghysdael. 1993. A single amino-acid 37. Kamachi, Y., E. Ogawa, M. Asano, S. Ishida, Y. Murakami, M. Satake, Y. substitution in the Ets domain alters core DNA binding specificity of Ets1 to Ito, and K. Shigesada. 1990. Purification of a mouse nuclear factor that binds that of the related transcription factors Elf1 and E74. Nucleic Acids Res. to both the A and B cores of the polyomavirus enhancer. J. Virol. 64:4808– 21:5184–5191. 4819. 8. Brazas, R. M., and D. J. Stillman. 1993. The Swi5 zinc-finger and Grf10 38. Kania, M. A., A. S. Bonner, J. B. Duffy, and J. P. Gergen. 1990. The homeodomain proteins bind DNA cooperatively at the yeast HO promoter. Drosophila segmentation gene runt encodes a novel nuclear regulatory pro- Proc. Natl. Acad. Sci. USA 90:11237–11241. tein that is also expressed in the developing nervous system. Genes Dev. 9. Brown, T. A., and S. L. McKnight. 1992. Specificities of protein-protein and 4:1701–1713. protein-DNA interaction of GABP␣ and two newly defined ets-related pro- 39. Karim, F. D., L. D. Urness, C. S. Thummel, M. J. Klemsz, S. R. McKercher, teins. Genes Dev. 6:2502–2512. A. Celada, C. V. Van Beveren, and R. A. Maki. 1990. The ETS-domain: a new 10. Corneliussen, B., A. Thornell, B. Hallberg, and T. Grundstrom. 1991. Helix- DNA-binding motif that recognizes a purine-rich core DNA sequence. loop-helix transcriptional activators bind to a sequence in glucocorticoid Genes Dev. 4:1451–1453. response elements of retrovirus enhancers. J. Virol. 65:6084–6093. 40. Kingston, R. E., and J. Sheen. 1989. Reporter system using chloramphenicol 11. Crute, B. E., A. F. Lewis, and N. A. Speck. 1994. Unpublished results. acetyltransferase, p. 9.6.2–9.6.7. In F. M. Ausubel, R. Brent, R. E. Kingston, 12. Crute, B. E., and N. A. Speck. 1993. Unpublished results. D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.), Current 13. Daga, A., J. E. Tighe, and F. Calabi. 1992. Leukaemia/Drosophila homology. protocols in molecular biology. Greene Publishing Associates and Wiley- Nature (London) 356:484. Interscience, New York. 14. DeFranco, D., and K. Yamamoto. 1986. Two different factors act separately 41. Klambt, C. 1993. The Drosophila gene pointed encodes two ETS-like pro- VOL. 69, 1995 ets AND cbf COOPERATION IN VIVO 4949

teins which are involved in the development of the midline glial cells. De- 67. Satake, M., S. Nomura, Y. Yamaguchi-Iwai, Y. Takahama, Y. Hashimoto, M. velopment 117:163–176. Niki, K. Yukihiko, and Y. Ito. 1995. Expression of the Runt domain encoding 42. Krasnow, M. A., E. E. Saffman, K. Kornfeld, and D. S. Hogness. 1989. PEBP2␣ genes in T cells during thymic development. Mol. Cell. Biol. 15: Transcriptional activation and repression by Ultrabithorax proteins in cul- 1662–1670. tured Drosophila cells. Cell 57:1031–1043. 68. Shore, P., and A. D. Sharrocks. 1994. The transcription factors Elk-1 and 43. Krimpenfort, P., R. de Jong, Y. Uematsu, Z. Dembic, S. Ryser, H. von serum response factor interact by direct protein-protein contacts mediated Boehmer, M. Steinmetz, and A. Berns. 1988. Transcription of T cell receptor by a short region of Elk-1. Mol. Cell. Biol. 14:3283–3291. ␤-chain genes is controlled by a downstream regulatory element. EMBO J. 69. Short, M. K., S. A. Okenquist, and J. Lenz. 1987. Correlation of leukemo- 7:745–750. genic potential of murine retroviruses with transcriptional preference of the 44. LaMarco, K., C. C. Thompson, B. P. Byers, E. M. Walton, and S. L. Mc- viral long terminal repeats. J. Virol. 61:1067–1072. Knight. 1991. Identification of ets- and notch-related subunits in GA binding 70. Speck, N. A., and D. Baltimore. 1987. Six distinct nuclear factors interact protein. Science 253:789–792. with the 75-base-pair repeat of the Moloney murine leukemia virus en- 45. Levanon, D., V. Negreanu, Y. Bernstein, I. Bar-Am, L. Avivi, and Y. Groner. hancer. Mol. Cell. Biol. 7:1101–1110. 1994. AML1, AML2, and AML3, the human members of the runt domain 71. Speck, N. A., B. Renjifo, E. Golemis, T. N. Fredrickson, J. W. Hartley, and gene-family: cDNA structure, expression, and chromosomal localization. N. Hopkins. 1990. Mutation of the core or adjacent LVb elements of the Genomics 23:425–432. Moloney murine leukemia virus enhancer alters disease specificity. Genes 46. Liang, H., X. Mao, E. Olenjniczak, D. G. Nettesheim, L. Yu, R. P. Meadows, Dev. 4:233–242. C. B. Thompson, and S. W. Fesik. 1994. Solution structure of the ets domain 72. Speck, N. A., B. Renjifo, and N. Hopkins. 1990. Point mutations in the of Fli-1 when bound to DNA. Struct. Biol. 1:871–876. Moloney murine leukemia virus enhancer identify a lymphoid-specific viral 47. Lim, F., N. Kraut, J. Framptom, and T. Graf. 1992. DNA binding by c-Ets-1, core motif and 1,3-phorbol myristate acetate-inducible element. J. Virol. but not v-Ets, is repressed by an intramolecular mechanism. EMBO J. 64:543–550. 11:643–652. 73. Studier, F. W., A. H. Rosenberg, J. J. Dunn, and J. W. Dubendorff. 1990. Use 48. Liu, P., S. A. Tarle, A. Hajra, D. F. Claxton, P. Marlton, M. Freedman, M. J. of T7 RNA polymerase to direct expression of cloned genes. Methods En- Siciliano, and F. S. Collins. 1993. Fusion between transcription factor CBF␤/ zymol. 185:60–89. PEBP2␤ and a myosin heavy chain in acute myeloid leukemia. Science 74. Sun, W., M. O’Connell, and N. A. Speck. 1993. Characterization of a protein 261:1041–1044. that binds multiple sequences in mammalian type C retrovirus enhancers. J. 49. Luckow, B., and G. Schutz. 1987. CAT constructions with multiple unique Virol. 67:1976–1986. restriction sites for the functional analysis of eukaryotic promoters and 75. Thompson, C. B., C. Wang, I.-C. Ho, P. R. Bohjanen, B. Petryniak, C. H. regulatory elements. Nucleic Acids Res. 15:5490. June, S. Meisfeldt, L. Zhang, G. J. Nabel, B. Karpinski, and J. M. Leiden. 50. Macleod, K., D. Leprince, and D. Stehelin. 1992. The ets gene family. Trends 1992. cis-Acting sequences required for inducible interleukin-2 enhancer Biochem. Sci. 17:251–256. function bind a novel Ets-related protein, Elf-1. Mol. Cell. Biol. 12:1043–1053. 51. Manley, N. R., M. A. O’Connell, W. Sun, N. A. Speck, and N. Hopkins. 1993. 76. Thompson, C. C., T. A. Brown, and S. L. McKnight. 1991. Convergence of Two factors that bind to highly conserved sequences in mammalian type C Ets- and notch-related structural motifs in a heteromeric DNA binding retroviral enhancers. J. Virol. 67:1967–1975. complex. Science 253:762–768. 52. Mao, X., S. Miesfeldt, H. Yang, J. M. Leiden, and C. B. Thompson. 1994. The 77. Thornell, A., B. Hallberg, and T. Grundstrom. 1988. Differential protein FLI-1 and chimeric EWS-FLI-1 oncoproteins display similar DNA binding binding in lymphocytes to a sequence in the enhancer of the mouse retro- specificities. J. Biol. Chem. 269:18216–18222. virus SL3-3. Mol. Cell. Biol. 8:1625–1637. 53. Martin, M. E., X. Y. Yang, and W. R. Folk. 1992. Expression of a 91- 78. Treisman, R., R. Marais, and J. Wynne. 1992. Spatial flexibility in ternary kilodalton PEA3-binding protein is down-regulated during differentiation of complexes between SRF and its accessory proteins. EMBO J. 11:4631–4640. F9 embryonal carcinoma cells. Mol. Cell. Biol. 12:2213–2221. 79. Urness, L. D., and C. S. Thummel. 1990. Molecular interactions within the 54. Meyers, S., J. R. Downing, and S. W. Hiebert. 1993. Identification of AML-1 ecdysone regulatory hierarchy: DNA-binding properties of the Drosophila and the (8;21) translocation protein (AML-1/ETO) as sequence-specific ecdysone-inducible E74A protein. Cell 63:47–61. DNA-binding proteins: the runt homology domain is required for DNA 80. Wang, C., B. Petryniak, I. Ho, C. B. Thompson, and J. M. Leiden. 1992. 13: binding and protein-protein interactions. Mol. Cell. Biol. 6336–6345. Evolutionarily conserved Ets family members display distinct DNA binding Miyoshi, H., T. Kozu, K. Shimizu, K. Enomoto, N. Maseki, Y. Kaneko, N. 55. specificities. J. Exp. Med. 175:1391–1399. Kamada, and M. Ohki. 1993. The t(8;21) translocation in acute myeloid 81. Wang, S., and N. A. Speck. 1992. Purification of core-binding factor, a leukemia results in production of an AML1-MTG8 fusion transcript. EMBO protein that binds the conserved core site in murine leukemia virus enhanc- J. 12:2715–2721. ers. Mol. Cell. Biol. 12:89–102. 56. Miyoshi, H., K. Shimizu, T. Kozu, N. Maseki, Y. Kaneko, and M. Ohki. 1991. 82. Wang, S., Q. Wang, B. E. Crute, I. N. Melnikova, S. R. Keller, and N. A. t(8;21) breakpoints on chromosome 21 in acute myeloid leukemia are clus- Speck. 1993. Cloning and characterization of subunits of the T-cell receptor tered within a limited region of a single gene, AML1. Proc. Natl. Acad. Sci. and murine leukemia virus enhancer core-binding factor. Mol. Cell. Biol. USA 88:10431–10434. 13:3324–3339. 57. Mueller, C. G. F., and A. Nordheim. 1991. A protein domain conserved 83. Wasylyk, B., S. L. Hahn, and A. Giovane. 1994. The Ets family of transcrip- between yeast MCM1 and human SRF directs ternary complex formation. tion factors. Eur. J. Biochem. 211:7–18. EMBO J. 10:4219–4229. Wasylyk, B. C., P. F. Wasylyk, A. Begue, D. Leprince, and D. Stehelin. 58. Nelson, B., G. Tian, B. Erman, J. Gregoire, R. Make, B. Graves, and R. Sen. 84. 1990. 1993. Regulation of lymphoid-specific immunoglobulin ␮ heavy chain gene The c-ets proto-oncogenes encode transcription factors that cooperate with 346: enhancer by ETS-domain proteins. Science 261:82–86. c-Fos and c-Jun for transcriptional activation. Nature (London) 191– 59. Nye, J. A., J. M. Petersen, C. V. Gunther, M. D. Jonsen, and B. J. Graves. 193. 1992. Interaction of murine Ets-1 with GGA-binding sites establishes the 85. Wasylyk, C., J. P. Kerckaert, and B. Wasylyk. 1992. A novel modulator ETS domain as a new DNA-binding motif. Genes Dev. 6:975–990. domain of Ets transcription factors. Genes Dev. 6:965–974. 60. Ogawa, E., M. Inuzuka, M. Maruyama, M. Satake, M. Naito-Fujimoto, Y. 86. Wijmenga, C., N. A. Speck, N. C. Dracopoli, M. H. Hofker, P. Liu, and F. S. Ito, and K. Shigesada. 1993. Molecular cloning and characterization of Collins. 1995. Identification of a new Runt-domain containing gene Cbfa3, PEBP2␤, the heterodimeric partner of a novel Drosophila runt-related DNA and localization of the human homolog CBFA3, to chromosome 1p34. binding protein PEBP2␣. Virology 194:314–331. Genomics 26:611–614. 61. Ogawa, E., M. Maruyama, H. Kagoshima, M. Inuzuka, J. Lu, M. Satake, K. 87. Woods, D. B., J. Ghysdael, and M. J. Owen. 1992. Identification of nucle- Shigesada, and Y. Ito. 1993. PEPB2/PEA2 represents a new family of tran- otide preferences in DNA sequence recognized specifically by c-Ets-1 pro- scription factor homologous to the products of the Drosophila runt and the tein. Nucleic Acids Res. 20:699–704. human AML1 gene. Proc. Natl. Acad. Sci. USA 90:6859–6863. 88. Wotton, D., J. Ghysdael, S. Wang, N. A. Speck, and M. J. Owen. 1994. 62. Petersen, J. M., and B. Graves. 1995. Unpublished data. Cooperative binding of Ets-1 and core binding factor to DNA. Mol. Cell. 63. Prosser, H. M., D. Wotton, A. Gegonne, J. Ghysdael, S. Wang, N. A. Speck, Biol. 14:840–850. and M. J. Owen. 1992. A novel phorbol ester response element within the 89. Wurster, A. L., G. Siu, J. M. Leiden, and S. M. Hedrick. 1994. Elf-1 binds to human T cell receptor ␤ enhancer. Proc. Natl. Acad. Sci. USA 89:9934–9938. a critical element in a second CD4 enhancer. Mol. Cell. Biol. 14:6452–6463. 64. Ptashne, M., and A. A. F. Gann. 1990. Activators and targets. Nature (Lon- 90. Xin, J., A. Cowie, P. Lachance, and J. A. Hassell. 1992. Molecular cloning don) 346:329–331. and characterization of PEA3, a new member of the Ets oncogene family 65. Rao, V. N., and E. S. Reddy. 1992. elk-1 domains responsible for autonomous that is differentially expressed in mouse embryonic cells. Genes Dev. 6:481– DNA binding, SRE:SRF interaction and negative regulation of DNA bind- 496. ing. Oncogene 7:2335–2340. 91. Zhang, L., V. Lemarchandel, P. H. Romeo, Y. Ben-David, P. Greer, and A. 66. Reisman, D. 1990. Nuclear factor-1 (NF-1) binds to multiple sites within the Bernstein. 1993. The Fli-1 proto-oncogene, involved in erythroleukemia and transcriptional enhancer of Moloney murine leukemia virus. FEBS Lett. Ewing’s sarcoma, encodes a transcriptional activator with DNA-binding 277:209–211. specificities distinct from other Ets family members. Oncogene 8:1621–1630.