Net, a new ets transcription factor that is activated...

Net, a new e t s transcription factor that is activated by Ras

Antoine Giovane, Alexander Pintzas, Sauveur-Michel Maira, Peter Sobieszczuk, and Bohdan Wasylyk 1

Centre National de la Recherche Scientifique-Laboratoire de Genetique Moleculaire des Eucaryotes (CNRS--LGME), Institut National de la Sant6 et de la Recherche M6dicale (INSERM)-U. 184, Institut de Chimie Biologique, Facult6 de M4decine, 67085 Strasbourg Cedex, France

Ras signaling appears to be mediated in part by transcription factors that belong to the ets gene family. To identify downstream targets for the Ras signal transduction pathway, we have used Ras-transformed mouse fibroblasts to isolate a new member of the ets gene family, net. Net has sequence similarity in three regions with the ets factors Elkl and SAP1, which have been implicated in the serum response of the fos promoter. Net shares various properties with these proteins, including the ability to bind to ets DNA motifs through the Ets domain of the protein and form ternary complexes with the serum response factor SRF on the fos serum response element, SRE. However, Net differs from Elkl and SAP1 in a number of ways. The pattern of net RNA expression in adult mouse tissues is different. Net has negative effects on transcription in a number of assays, unlike Elkl. Strikingly, Ras, Src, and Mos expression switch Net activity to positive. The study of Net should help in understanding the interplay between Net and other members of the Elk subfamily and their contribution to signal transduction through Ras to the nucleus.

[Key Words: Elkl; oncogene; Src; SRE; SRF]

Received November 18, 1993; revised version accepted May 16, 1994.

The ets gene family codes for transcription factors that are involved in the generation of human cancers, cell transformation, and development (for review, see Seth et al. 1992; Macleod et al. 1992; Janknecht and Nordheim 1993; Wasylyk 1994; Wasylyk et al. 1993; Treisman 1994). The Ets proteins contain a conserved Ets domain that binds specifically to DNA sequences with a core GGA element (ets motifs). They can be grouped into subfamilies on the basis of different criteria such as homology in the Ets domain, position of the Ets domain in the protein, and the presence of other similar sequence elements with comparable functions. The subfamilies appear to have arisen through duplication of an ancestral gene. The Elk subfamily proteins Elkl and SAP1 have three regions with similar sequences (A, B, and C; Rao et al. 1989; Dalton and Treisman 1992; Fig. 1B). They are unusual in the Ets family as the Ets domain (region A) is amino terminal. The B region is required to form a ternary complex with the serum response factor (SRF) and the ets and SRF motifs of the fos serum response element (SRE) (Dalton and Treisman 1992; Janknecht and Nord- heim 1992; Hipskind et al. 1991; Rao and Reddy 1992a; Treisman 1992). The Elkl C region is an activation domain that is stimulated by phosphorylation by MAP kinase (MAPK), and directly imparts regulation on the

1Corresponding author.

SRF (Janknecht et al. 1993; Marais et al. 1993). This model provides an important conceptual framework for understanding the Ras signaling pathway, of which MAPK is a component.

The ras oncogene is frequently mutated in human tu- mors. It is a component of a highly conserved signal transduction pathway that is present from yeast to man (for review, see Blenis 1993; Khosravi and Der 1994; Moodie and Wolfman 1994) and links the exterior of the cell to nuclear effectors of cell growth and differentiation (Chambers and Tuck 1993). Signals flow through recep- tor tyrosine kinases, intermediary linking proteins, Ras, and then a cascade of kinases composed of Raf, MAP kinase kinase kinase (MAPKKK) MAPKK, and MAPK to downstream effectors. Ets factors mediate transformation and regulation of development by Ras. Trans-dom- inant mutants of Ets inhibit Ras transformation (Langer et al. 1992) and revert Ras-transformed cells (A. Giovane et al., unpubl.). Genetic studies in Drosophila show that the Ets-protein Yan/Pok is a negative regulator of pho- toreceptor development that acts antagonistically to the proneural signal mediated by Ras (Lai and Rubin 1992; Tei et al. 1992).

We have identified a new Ets protein, Net, which is another member of the Elk subfamily. It has three regions of similarity with Elkl and SAP1 (A, B, and C) and interacts with SRF to form a ternary complex. Interest- ingly, Net inhibits transcription, in contrast to Elk1. Ras

1502 GENES & DEVELOPMENT 8:1502-1513 �9 1994 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/94 $5.00

Cold Spring Harbor Laboratory Press on April 7, 2019 - Published by genesdev.cshlp.orgDownloaded from

http://genesdev.cshlp.org/

http://www.cshlpress.com

Ras activates Net

A * * ,

TCCACAGACTGCACACCGCCTC'GC~3~T~T~CAGTA~AC~d~G~3AC~CAC~CTCGGCCTGACTGCTCC~CT~PCC~CTCTCACACACA 90 CCAGAGGAGGGGG~AAGAGAAAAA~GAGGAGCGAGAGA~GAAAAAAAAAGAGGGGGGAAAATCAGGATCTCATTAC 18C AAGAGCAACAGACCGTCTGGAGACGCCTGTCAGC~AAAGTCG~TTTCACCC~GTTCCTCC~T~TCTCCCCAAGAAGACTC 270

CCAAACGCTCCCCCACGTCTGGGTATGGAGAGTGCAATCACGCTGTGGCAGTTCCTCTTGCACTTGCTGCTGGACCAGAAACATGAGCAC 32620 MetGIuSerAIalleThrLe~TrDGInPheLeuLeu~isLeuLeuLeuAsDGlnLvs~aGIuHiS

CTCATCTGCTGGA~ATCGAA~GATCGC~A~TTCAA~CTCCTCAAGGCAGAAGAAGTGGC~AAGCTGTC~]GGCCTCCGCAAGAACAAGACC 4~I ~euI~ecv~TrDThrSerA~A~DG~v~]uPheLv~L~uL~uLvsA~aG~uG~uVa~A~aLv~euTrDG]vLeuAraLvgA~nLvgThr

A (Ets domain) AACAT~AACTACGACAAGCTGAGCAGAGCGCTGAGATACTATTACGACAAGAACA~ATCAAGAAAGTGATCC~GCAGAAGTTTGTGTAC 540 AsnMetA~nTvrAsDLvgL@ugerAr~A]aLeuAraTvrTvrTvrA~DLvgA~nI~eI~eLv~Lv~Va~I~eG~vG~nLvs~heVa~T~r 82

AAGTTCGTCTCTTTCCCGGATATCCTGAAAATGGATCCTCACGCGGTAGA•ATCAGCCGGGAGAGCCTCCTGCTGCAGGACGGCGACTGT 630 Lv~Fhe~a~SerPhePr~AZDI~eLeuLysMetAsp~r~isA~aVa~G~uI~eSerArgG~uSerLeuLeuLeuG~nAspG~AspCys IZ2

AAGGTGTCCCCGGAAGGCCGAGAGGTCCACAGGCACGGCTTGTCCTCCCTCAAAAGTGCCAGCCGCAACGAGTACCTCCACTCGGGGCTC 720 LysValSerPr~G~uG~yArgG~uVa~HisArgHisG~yLeuSerSerLeuLysSerA~a~erAraAsnG~uT~rLeuHisSerG~vLe~ ~

TACTCGTCCTTCACCATCAACTCCCTGGAGAACGCTCCAGAGGCCTTCAAGGCCATCAAGACGGAGAAGCTGGAGGAGCCCTGTGATGAC 81C Tvr~er~erPheThrI~eAgn~erLeuG~uAgnA~aPr~G~uA~aPh~ysA~aI~eLysThrG~u~ysLeuG~uG~uPr~CysAspAsp ~72

AGCCCCCCTGTGGAAGAAGTCA~GACTGTGATCAGGTTT~TGACCAATAAAACCGACAAGCACATCACCAGGCCTGTGATGTCCCTGCCT 900 ~erPr~Pr~Va~G~uG~uVa~ArgThrVa~I~eAr~PheVa~ThrAsnLysThrAspLysHi~I~eT~rArgPr~Va~Me~erLeuPr~ 202

TCCACATCCGAGACCGCTGCGGCAGCGGCATCCGCTTTCCTGGCCTCATCTGTCTCAGCCAAGATCTCCTCqTTAATGTTGCCAAATGCT 890 SerThr~erG~uThrA~aA~aAl~AlaA~aSerAlaPheLeuA~a~er~erVa~erA~aLy~I~eSer~erLe~MetLe~Pr~A~nA~m Z32

GCCAGCGTTTCGTCTGTGTCACCCTCTTCATCTCGGTCCCCATCCCTGTCCCCCGACTCTCCCCTCCCTTCTGAACACAGAAGCCTCTTC 10g0 A~Se~ValSerSerV~SerPr~SerSer~erArgSerPr~SerLeu~er~r~AspSerPr~LeuPr~SerG~uHi~Arg~erLeuPhe ~$2

CTC,GAGGCA~CCTG~CATGAGTCGGATTCT~TGGAGCC~CTGAATCTGTCATCG~GCTC~AAAACCAAGTCTCCATCTCTTCCCCCAAAA !170 LeuG•uA•aA•aCysHi•G•uSerAspSerLeuG•u•r•LeuA•nLeu•erSerG•y•erLysThrLysSer•r•SerL•u•r••r•Lys 292

GGCAAAAAACCCAAAGGCTTGGAAATCTCTGCACCCCAACTGTTGCTCTCCGGCACCGACATCGGCTCCATCGCCCTCAACAGCCCAGCC 1280 G~yLy~Lys~r~LysG~yLeuG~uI~e~erA~a~r~GlnLeuLeuLeu$erG~yThrA~pI~eG~y~erI~eA~aLeuAsn5erPrcA~a 322

CTCCCCTCAGGATCCCTCACTCCAGCCTTCTTCACCGCACAGACACCAAGTGGACTGTTTCTGGCCTCGAGTCCGCTGCTGCCCAGCATA 1350 Leu~r~S~rG~y~erLeuThr~r~A~a~h~ph@ThrA~aG~nThrPr~SerG~vLeu~heLeuA~a$erSer~r~LeuLeupr~@r~]e 352

C CACTTCTGGAGCAGCCTTAGTCCGGTCGCCCCACTGAGTCCTGCCAGGCTGCAAGGGCCGAACACACq~CCAGTTCCCCACACT~CTC 1440 ~is~heTrDS~rgerL@u~er~r~Va~A]a~r~Leu~er~r~A]aAraLeuG~nG~vPr~AsnThrLeu~heG~nPh@pr~ThrLeuLeu 382

AACGGTCACATGCCGGTGCCGCTCCCCAGTCTGGACAGAGCTCCATCCCCAGTTCTGCTGTCCCCCAGCTCTCAGAAATCCTGATGATGG 1530 AsnG•yHi•MeE•r•Va••r•LeuPr•Se•LeuA•pArgA•a•r••erPr•Va•LeuLeuSer•r•S•rSerG•nLy•Ser ~09

GCCACCAACTGTCACTGTCATCAACTAAGGACTCAGAACAGATGACGGATGATGCCAGTTTGTCCCCATGGGCTAGq~ACC~GTGTCAT 162~ GAGAAGGACA~'FCTGAAACCTGGTTCATTTGGTq~PGCACTTTTCGTAACATGGATAATCTAGATGTATGTTTAGCATTTTAAAACAAAAG !719 TTTTGGTCTTTTTATATATATATATATATATATTCAGCTCTCTATAAAAGTCTGTTTCGCATTCAGTGAATTT•fAATGTGTGTG7q•I•ITT 18Cr CTTAATCTTGTTAGCTCTGGAGTGTTGAACACTC~CAGGGAGGAGCCqTTCTTAATGTTTTAATGTAACTAATGAG~ATGTG~GCCT~ 1890 TATTGCTTTAACTCT~CATATGCTGAACTCCGC9~GCATACACGATGTAACCAGCTGTGCCTTCCTT?GC~TTTAGTGCTATATGTACAT 1980 GAATTTATGTATTTGTAGTATTG~GAGGAACTGTTTTTAGGACACCTq~GCAATCCAACTGCTCT~TGTAGAGGGT~9Aq`TTTCTCACGTA 2370 GACACGTTCCAAGTGAACCTCCATCGCG~TAAGAACTGCCTTAGGACCT~TTCGGAAC~AAATCCAGTAATGTCCAAGTGATGGGTTTT 2160 TATATAAGAATGAACGAAGCCATATPTAGACAAT.~AACATTGT~CFAAAGAACATGTTC~AAAGTGCACT~I~I~AAAAGAAGAACTq~PGA 22~C

GAGACCAAGCCGCTCAGCCGTTTGTAGAAGCCTT~CTACTTCCTGCAGAAGTCTGAGATGCTAAGGCTCCGTTAAGTTCACACCAGCCAA 2340 ACTCTCAAGGGG~CTTTCCAGGTTTGTATTTAATTTTGAGACTAGCCCTGCTCTCTTCAT~CTACAACCACACCACAACCATCTGGTA 2430 TCATTATCACTACTGGGAGTTAAGGTTTCTTTAGATT~TGG~ATGACAATTAATAGGTAAC~TAATGTIx3TAGGGCTGA~ATATA~TC 2520 TCCATAGTATGTGTAACCTGCATGTGTTGAGACTACGGGTATAACCCCCAGTAATCTGTCTCCCTGTCTCCCCCAAGAAGC~CGTGGTGG 2610 GGAAACAAACTCAGAATGTTTTGGAAGATGCTTACAGATGGTGAACTC 2658

B A (Ets domain) �9 Net MESAITLwQF~LHL~LDQKHEH~Ic~FTSND~GEFKLLKAEEVAKLwG~RKNKTN~NYDKLSRALRYYYDKNIIKKVIGQKFvYKFv~F~DILKMD~HAV . . . . EISR hNet ME~ITLWQFLL~LL~QKHE~LI~T~N0~GEFKL~KAEEVAKLwGL~KNKTNMNY~K~SRAL~YYY~NIIKKVI~QKFVYKFVSFpEILKMD~HAV . . . . EIS~ hSAPla ~DSAITL~FLLQLLQK~QNKH~ICW~SND.~QFKLLQAEEVARLWGIRKNK~NMNY~KLSR~LRYYYVKNIIKK~N~QKFVYKFVSY~EILNN~pMTV.GRIE~C mSAPIO M~SAITL~QFLLQLLQEpQNEH~1~gl̀ ~NN~EFKLLQAEEVARL~TRKNKPNMNYDKLSRA~RYYYVKN~IKKVN~QKFvYKFV~YPETLKM~TV.ARTEGDC hE~kl ~D~SVT~WQFt~Q~REQ&NGH~SWTSRD~GEF~LVDAEEVA~LwGL~NKTN~NY~K~SRALR~YYD~NIIRKV~GQ~VYKF~Su mElkl //VTLwQF~LQ~REQ~NDH~rTS~D~GEFKLVDAEEVARLWGLRKNKTNMNY~KLSRALRYYYDKNIIRKV~GQKFVYKFVSY~EVAGC~TEDCP~Q~Ev~V

B mNet ESLLtQDGD(KV . . . . . . SPEGREVHRHG . . . . . LSSLKSASRNEYLHS~LYSSFTINSL[...NA.PEAFKAIKTEKLEEP.CDDSPPVEEVRTVIRFVTNKTDKH hNet ESLLLQDSDCKV ...... SPEGREAHKHG . . . . . LAVLRSTSRN[YIHSGLYSSFTINSLE...NP.PDAFKAIKREKLEEP.PDDSPPVEEVRTVIRFVTNKTDKH hSAPla ESLNF . . . . SEV . . . . . . ~SSSK~VE~G~K~GAK~RNDYIH~GLYSSFTLN~N.~NVKLF~LIKTE~pAEKLAEKK~PQEPT~SVIKFVTTPSKKP mSAPla EALNS . . . . IET . . . . . . . S~SK~vEYGGKERPpQP~AKTS5RNDYIH~LYSSFTLN~N..~TSNKKLFKSIKIEN~AEKLAEKKA~QEPTP~VIKFVTTPAKK~ hE~kl T~T~NVApA~HAApGDTVSGKP~TpKGA~MA~p~GLARS~RN[Y~&S~LY~TFTIQSLQp~ppBpRpAVV~pN~ApAGA~pp~GSRSTSPSPLEAC~EAEE mE|kl T~AIA.MAPATVHAG~GDTATGKpG~TKGAGMTGQGG~ARS5RNEYMRSG~YSTFTIQS~Q~E~Q~I~RpASVL~NTTPAGVPAPASGSRST~PN~LEAC~EAEE

mNet ITRPV...MSLPSTSETAA~ASAFLASSVSAKISSLMLPNAAS ..... VSSVSPSSSRSP$L.SP..DSPLPSEHRSLFLEAACHs hNet VTRPV...VSLPSTSE..AAAASAFLASSVSAKISSL~LPNAAS . . . . . ISSASPFSSRSPSL.SP.,KSPLPSEHRSLFLEAACHDSDSLE.PLNLSSGSKTKSPS hSAPla P V E ~ V A A T I ~ l G p ~ I ~ E E T I Q A L E T L V S p K ~ p ~ L s

mSAPla pIEpVAAACATSp~L~p~EETIQALET~V~pT~SLETpA~V~I~AT]̀ FN~T~PV~ST~L~LKEP~RTpSP~L~NpDIDTDIE~VASQ~MEL~ENLSLEpKKQDS hElkl AGL~LQVILTp~EApNLKS[s163 . . . . . . . . .

mETkl AGLPLQVILTP,I

C

mNet L . . . . . . . . . ~K~K~K~L[I~ApQLLLS~TDI~S~ALN.~PA~S~S~T~AFFTAQT~FLA~SPLLpSIH~W~S~PvAPLS~ARLQ~PNTLFQFPT~LN hNet L . . . . . . . . . P/~ hSAPlo VL~EKDKVNNSSR~KKp~GL~APT~VIT5~DP~LGIL~Sp~LpTASLTpAFF~SQT~..~]LTp~p[t~SIHFW~Tt~VA~5~AR~Q~ANT~FQF~SVLN mSAPlo A~AEK0KTNN5SQ~KKPK6L[L..TpALVVTGSDPSPSGIL~pS~LpTASLTpALF.SQTP~.ILLTP~PLLSSIHFwSTLSpFApLSPARLQGANMLFQFPSVLN hE1kl . . . . . . . . QpQK•R••RDLEtpLSPSttGG•G•ERT•G•G••S•LQA•G•ALT•SLL•T•TLT••LLTpSSLp•SIHFwST•SpIA•RSpAKLS . . . . . FQFPSSGS mElkl

mNet GHMPVPLPSLDRAPSPVLLSPSSQKS hNet hSAPla SHGPFTLSGLOGPSTPGPFSPOLQKT mSAPlo SHGPLTLSGLEGPSLPGPFSPRLQKT

hE1kl AQVHIPSISVDGLSTPVVLSPGPQKP mEtkl

�9 , , , , ..

Figure 1. Net sequence. (A) The largest ORF codes for a protein (Net) with homology (underlined) to the Ets family (Ets domain, A) and Elk1 and SAPla (A, B, C). Numbers at right correspond to nucleotides and amino acid (bold) sequences. In the 5' region, there are an in-frame stop codon (stars), out-of-frame stop codons (not indicated), and a short ORF (initiation and stop codons underlined). The 3' region has an internal polyadenyla- tion signal (underlined). (B} Net is homologous to SAPla and Elkl in three regions (A, B, and C). Identical amino acids are indicated with a star, similar amino acids (similarity groups

I/M/L/V,T/S, E/D, R/K, F/Y) with a dot. The human SAPla sequence has been corrected from that originally published (R. Treisman, pets. comm.; data base accession number M85165). The deduced amino acid sequences of hNet (human), mElkl (mouse) are partial; they lack the amino-terminal sequence of mElkl and the carboxy-terminal sequences of both mElkl and hNet (indicated by double slashes). Sequences were analyzed with the University of Wisconsin GCG package.

expression leads to transcription activation by Net. The properties of Net are reminiscent of the negative Yan/ Pok factor that participates in Ras signaling.

Results

RNA coding for a n e w Ets protein is present in Ras- transformed fibroblasts

Ets domain-coding sequences were amplified from Ras- transformed NIH-3T3 fibroblast cDNA with degenerate primers (oligonucleotides 1-3, Materials and methods; sequence complexity 30,000-100,000). The PCR product of the expected size was subcloned, and the sequences of individual clones were found to code for the Ets domains of GABPoL (two clones}, Ets2 (three clones), and an un- known protein that we named Net (four clonesl. Net Ets domain-coding sequences were found to be present in cDNA libraries of mouse 10-day embryos, embryonic stem (ES) cells, F9 EC cells, retinoic acid-differentiated F9, and P19 cells and heart, by use of PCR with specific oligonucleotides derived from the cloned sequence (4 and 5; Materials and methods). Seven net cDNA clones ranging in size up to 2.3 kb were isolated from the embryo and ES cell libraries by screening with an ets domain probe and sequenced entirely on both strands (Fig. 1A; see Materials and methods). The compiled cDNA

sequence was estimated to lack no more than 50 bp at the 5' end and up to 450 bp at the 3' end sequences by rapid amplification of cDNA ends {RACE-PCR; Fro- hman 19881. The longest open reading frame (ORF) is preceded by an in-phase termination codon and has a near optimum sequence for translation initiation (purine at - 3 , G at +4; Kozak 1991). The 3' end is AT rich (60.5%), has an uninterrupted stretch of 26 A-T residues, and contains four ATTTA motifs reminiscent of unsta- ble mammalian RNAs that code for proto-oncogenes (Sachs 1993).

The ORF of mouse net codes for an ets domain that is very similar to human Elkl and SAP1 (Fig. 1B, ~ 80% similarity) and is located at the amino terminus, unlike other Ets proteins. Mouse Net also resembles human Elkl and SAPla in regions B and C (-70% and 60%, respectively). To ensure that Net is a third member of the Elk subfamily, we isolated mouse elkl and SAP1 cDNAs from Ras-transformed NIH-3T3 fibroblast RNA as well as several cDNA libraries (ES cell and 10-day embryo), and human net cDNA from a HeLa cell cDNA library (see Materials and methods}. The nucleotide and deduced protein sequences of each member of the subfamily are very similar between species and clearly distinct from each other in each species (Fig. 1B), showing that they are distinct members of the elk subfamily.

GENES & DEVELOPMENT 1503




Giovane et al.

net RNA is abundant in fibroblast cell lines and has a different expression pattern from e lk l and SAP1

Two n e t RNAs migrating at -2 .5 and 4.5 kb were de- tected on RNA blots (Fig. 2). The shorter RNA corre- sponds closely in size to the cloned cDNA (Fig. 1A). Both RNAs hybridize to full-length n e t probes and fragments containing either related (the e t s domain, positions 201- 711, Fig. 1A) or d iss imi lar (B to C region, posi t ions 712 and 1326) sequences. They do no t cross-hybridize to mouse elkl , SAP1, and ets2 probes. The na ture of the longer RNA is under invest igat ion, net RNA levels are variable be tween cell l ines (Table 1) and adul t mouse t issues (Fig. 2). net, e lk l , and SAP1 m R N A s have different sizes and t issue expression pat terns (Fig. 2; e.g., c.f. hear t and brain for net and e/k and lung and l iver for net and SAP1) cons i s ten t w i th thei r being d is t inc t members of the elk subfamily.

Specific D N A binding by Net is inhibi ted by carboxy- terminal sequences

Equimolar amounts of Net and SAPlb (synthesized in rabbit reticulocyte and quantitated by SDS-PAGE of [3SS]methionine-labeled proteins; not shown) were analyzed by mobility shift assay with the PEA3* e t s motif as a probe. Net and SAPlb formed one major and several

Table 1. net mRNA expression pattern in murine and rat cell lines

net mRNA level a Cell lines Description

+ + S--3T3 + + + + NIH-3T3 + + + NIH-3T3 ras

+ + + NIH-3T3 raf + + TM3 + + WEHI3 + + 70Z/3 + + BW5147 + P19 + F9 + F745 + MPC11 BU4 + X63Ag8 + X63Ag8WS 0/+ ES + +* NG108.15 + + * FR 3T3 + + + * FR PyMT + + + * FR ras

fibroblast, Swiss 3T3 fibroblast, NIH-3T3; C11 (FDH) Ki-ras-transformed C11; DT U1

(FDH) v-raf-transformed NIH-3T3; 3611 testis, Leydig cells, BALB/c mouse myelomonocytic leukemia pre-B lymphocyte T-lymphoma embryonal carcinoma embryonal carcinoma erythroleukemia plasmocytoma, lg secreting nonsecreting myeloma secreting myeloma embryonal stem, D3; SV129 mouse neuroblastoma/glioma rat fibroblasts Py mT-transformed FR 3T3; MTT4 ras-transformed FR3T3

Relative expression levels of net RNA are shown in arbitrary units, where + + + + was the highest level (100%), + + + in- termediate (-70%), + + low (-40%), and + very low (-10%), as observed on Northern blots. Semiquantitative comparisons of net-specific mRNA bands were aided by PhosphorImager analysis. Results were adjusted for the amount of loaded mRNA by comparison to ethidium bromide-stained gels and the internal control a-actin. a(,) Rat.

minor complexes (Fig. 3A; C points to the major complex, c.f. lanes 1 and 4 with the lysate control, lane 7). Complex formation was inhibited by excess cold wild- type competitor (lanes 1-9) but not by a mutated competitor altered in the e t s motif (lanes 10-18; Materials and methods). The Net complexes are supershifted with Net antibodies (not shown). Carboxy-terminal deletion mutants of Net were synthesized and quantitated by SDS-PAGE of labeled proteins and PhosphorImager analysis (Fig. 3B). De le t ion up to the Ets doma in progres- s ively increased the aff ini ty for the PEA3* probe - 2 0 - fold (Fig. 3C; the amoun t s were es t imated by Phospho- rImager analysis and corrected for p ro te in levels). These results show tha t the Ets doma in of Ne t is suff icient for D N A binding and tha t carboxy- te rmina l sequences are inhibi tory .

Figure 2. net RNA expression patterns. The blot, prepared with 2 ~g of poly(A) + RNA (MNT-Clontech Labs.), was hybridized with labeled probes for net (2.43-kb EcoRI fragment of clone 26), murine elk (600-bp EcoRI fragment of clone pmElk), murine SAP (700-bp EcoRI fragment of clone pmSap) and [3-actin.

Net forms a ternary complex wi th SRF and D N A containing moti fs for both proteins

The Net-related proteins Elkl and SAP1 form ternary complexes w i th SRF and D N A con ta in ing ets and SRF motifs, such as G wt -wt (Treisman et al. 1992). Ne t bound very weak ly to G wt -wt (Fig. 4A, lanes 1,2, visible on longer exposure; Fig. 4B, lanes 10,11}, whereas SRF

1504 GENES & DEVELOPMENT




Ras activates Net

Figure 3. Specific binding of Net to the ets motif is inhibited by carboxy-terminal sequences. (A) Net binds specifically to the PEA3" motif. Equal molar amounts (quantitated by SDS page of [3sS]-labeled proteins)of SAPlb (lanes 1-3, 10--12)or Net (lanes 4-6, 13-15), or an equal volume of reticulocyte lysate incubated without specific RNA (Lys, lanes 7-9, 16--18) were analyzed by mobility-shift assay with the 32P-labeled PEA3* probe. The indicated molar excess of unlabeled wild-type (WT COMP, lanes 1-9) or mutant (MUT COMP, lanes 10-18) competitors was used. (B,C) Carboxy-terminal sequences inhibit DNA binding by the Ets domain of Net. The amounts of the indicated carboxy-terminal deletion mutants were estimated by SDS-PAGE of [3sS] methionine-labeled proteins (B), and their affinity for the PEA3* probe analyzed by mobility shift (C). Radioactivity was measured with a PhosphorImager. (0) Lysate control; {P) proteins; (C} specific complexes; (F) excess free probe; (M)radioactive protein size markers.

formed a prominent specific complex (C1, lane 3) only when its motif was intact (G wt-mut, lane 11). Net dis- placed the SRF complex to a slower migrating form when present with SRF (C2, lane 4) but only when both the ets and SRF motifs were intact (G mut-wt and G wt-mut, respectively; lanes 8,12). The change in mobility was more dramatic when a shorter form of SRF was used, SRF(122-265), that contains the DNA-binding domain and the Elkl interaction domain (Fig. 4B, lanes 1-3,10,11; the ternary complex is indicated by a small arrowhead). Elkl and SAP1 require the B domain for ternary complex formation. Net mutants retaining the B

domain formed ternary complexes efficiently (lanes 3-7; see Fig. 3 for structures). There was a large increase in binary Net -DNA complex formation when the B domain was deleted (Fig. 4B, see bands labeled with a dot and c.f. especially lanes 7, 8, 15, and 16; see above), which was greater than expected from the increased binding affinity caused by deletion of the B domain. Sim- ilar results were obtained with a natural c-fos SRE containing a nonconsensus ets motif (not shown). We con- clude that the B domain of Net mediates interactions with SRF, similar to Elkl and SAP1.

Net expression represses transcription specifically through ets moti fs

Net expression in NIH-3T3 cells specifically and reproducibly inhibited the low basal activity of a reporter containing ets motifs (Fig. 5A, lanes 1,2,7-9; data not shown), in contrast to cEtsl(p68)(Fig. 5C, lanes 1,2,5,6,11). Net also efficiently inhibited cEtsl(p68) activation (Fig. 5C, 6-8,11-13). These effects were specific to the reporter with the response element llanes 1-3} showing that it was not attributable to nonspecific competi- tion for general transcription factors. Elkl expression stimulated transcription (Fig. 5A, lanes 1,3,7,10,11) and increased activation by cEts (p68)(Fig. 5C, lanes 1,2,4,6,9-11,14,15). Protein blots indicated that similar amounts of Net and Elk1 were expressed (Fig. 5B, lanes 7-11; data not shown). These results suggest that Net could be a negatively acting factor, in contrast to Elkl.

Ha-Ras activates Net

Ha-Ras oncogene expression did not affect the constitutive activity of the ets motif (Fig. 5A, lanes 1,4,7,12), but converted the effects of coexpressed Net from negative to positive (lanes 2,5,7-9,12-14). Ras expression did not affect the activity of cEtsl(p68) or GAL4--VP16 (data not shown; Webster et al. 1988} and had only a small effect on Elkl (at most 1.5-fold; Fig. 5A, lanes 1,3,4,6,7,10- 12,15,16). Protein blots showed that Ras expression did not alter the levels of Net or Elkl (Fig. 5B, lanes 7-16),





Giovane et al.

Figure 4. Net-SRF temary complex formation requires the B domain. (A) Net forms a ternary complex with SRF and specific DNA. Mobility shifts of equimolar amounts of Net and SRF (separately translated) either alone (lanes 2, 6, 10, and 3, 7, 11, respectively) or after mixing (lanes 4,8,12), compared with an equal volume of unprimed incubated lysate (lanes 1,5,9) with probes containing motifs that are wild type for ets and SRF (G wt-wt, lanes 1-4), mutant for ets (G mut-wt, lanes 5-8) or mutant for SRF {G wt-mut, lanes 9-12). (C1)SRF-specific complex; (C2) ternary complex; (F) free probe. (B) The B domain is required for complex formation between Net and SRF (122-265). Equimolar amounts of SRF (amino acids 122-265, lanes 2-9), Net and various deletion mutants (lanes 3-9, 1 I-17; see Fig. 3B for structures) that were synthesized in reticulocyte lysates were analyzed by mobility shift either separately or in combi- nations (as indicated) with the G wt-wt probe. Ternary complexes are indicated by small arrowheads and Net binary complexes with a dot. (SRF) SRF-specific complex; (F) excess free probe.

suggesting that Ras activates Net by post-translational modification.

v-Src and v-Mos s t imulate Net act ivi ty

v-Src specifically stimulated coexpressed Net (Fig. 6, lanes 1,2, 6-8; data not shown) less efficiently than Ras (lanes 9-11). v-Mos also reproducibly activated Net (lanes 1-5). v-Raf expression from several vectors had no detectable effect on Net. However, the effects on control recombinants were relatively small in NIH-3T3 com-

pared with other cell lines (results not shown; Wasylyk et al. 1989), precluding definitive conclusions about activation of Net by Raf. MAPK (p42) expression had no significant effect on Net activity and little effect on Elkl (less than twofold). Activation of MAPK by coexpression with Ras (see Janknecht et al. 1993) led to a small activation of Net (less than twofold) and only somewhat more on Elkl (two- to threefold; data not shown). The effects were even smaller when MAPK was activated with Rafl (not shown; several Raf vectors were used). Elkl is efficiently activated in RK13 cells by coexpress- ing MAPK (p44) and active Rafl compared with either MAPK or Rafl alone (Janknecht et al. 1993), suggesting that MAPK is limiting in these cells. Our results suggest that MAPK is not limiting in NIH-3T3 cells and that inhibition by Net does not result from titration of a limiting quantity of MAPK.

Net expression inhibits the fos SRE

Net expression inhibited the constitutive activity of the SRE but not of the basal promoter (Fig. 7, lanes 1,2,4-6; m in SREm signifies that the FAP motif is mutated). Different reporters and expression vectors gave similar results (DSE-CAT+pBL-CAT2, Robin et al. 1991; pSG5-Net; pJ7fl-Net; p601D-Net; data not shown}. In contrast, Elk1 specifically increased SRE activity [-2.5- fold; lanes 1,3,4,7,8). Protein blots showed that comparable and increasing levels of Net and Elk1 were expressed (data not shown).

Decreasing endogenous Net levels with antisense net RNA specifically increased SRE activity (up to fourfold; Fig. 8A, lanes 1-10; note that the cells were serum starved for longer than above). Serum stimulated the SRE [about sixfold; Fig. 8A, lanes 1,2,6,7; Fig. 8B, lanes 11,12,16,17). Anti-net RNA also specifically stimulated SRE in high serum (up to 10-fold, lanes 11-20). Similar results were obtained in seven different experiments with various amounts of antisense vector and several different plasmid preparations [range between 3- and 10- fold activation). The empty vector p601D, lacking net sequences, had no effect when it was used either in the place of the anti-net vector or to fill up the transfected DNA to the fixed amount applied to cells. Similar results were obtained whether pSG5 or pEMBL were used to fill up the transfected DNA (data not shown). To ensure that Net levels were effectively decreased by antisense expression, we performed several controls. Only - 1 0 % of NIH-3T3 cells are transfected and the decrease in Net expression would go undetected because of the untrans- fected cells. In COS cells we found that antisense net RNA decreased Net expression but had no effect on Elkl (results not shown). Specifically in transfected NIH-3T3 cells, net antisense decreased the negative effects of ex- ogenously expressed Net on either constitutive or cEtsl- induced PALx4 activity. In contrast, it had no effect on cEtsl activity alone, indicating that it did not have a general nonspecific effect (data not shown). These obser- vations indicate that net antisense RNA decreases Net protein levels. The results from both overexpression and





Ras activates Net

Figure 5. Regulation of transcription by Net, Elkl, and Ha-Ras through ets motifs. {A) Ras activates Net and Elkl. The transfections contained pBL-CAT4 (lanes 1-6}, PALx4 (lanes 7-16), pTL2-Net (1 ~g, lanes 8 and 13; 5 ~g, lanes 2, 5, 9, and 14), pTL1-Elkl (1 ~g, lanes 10 and 15; 5 ~g, lanes 3, 6, 11, and 16), pSG5 (5 ~g, lanes 1, 4, 7, and 12), pRCBx2 (5 ~g, lanes 4-6 and 12-16), pARCBx2 (5 ~g, lanes I-3 and 7-11). {B) Protein blot corresponding to CAT assay in (A) Blots prepared with total cell extracts {100 ~g/lane), were probed with Net-specific (top) or Elkl- specific (bottom)antibodies. {C) Net, but not Elkl, inhibits cEtsl(p68). Transfec- tions contained pBL-CAT4 (lanes 1--4), PALx4 (lanes 5-15), pTL2-Net {1 ~g, lanes 7 and 12; 5 ~g, lanes 3, 8, and 13), pTL1-Elkl (1 ~.g, lanes 9 and 14; 5 ~g, lanes 4, 10, and 15) and pSGS-cEtsl(p68) (1 ~g, lanes 6-10; 2 ~g, lanes 2-4 and 11-15), pSG5 (7 ~g, lanes 1 and 5).

down-regulation of endogenous Net suggest that Net could be a negative regulator of transcription.

D i s c u s s i o n

Net is a n e w member of the Ets gene family that belongs to the Elk subfamily

Our objective is to study the role of ets family members in the Ras signal transduction pathway. We searched for new, previously unidentified Ets proteins that might be expressed in Ras-transformed fibroblasts by reverse transcriptase-polymerase chain reaction (RT-PCR) using a panel of degenerate primers against conserved sequences of the ets domain. We identified sequences coding for a new ets factor (Net) as well as GABPot and Ets2. How- ever, this approach was not exhaustive because we also isolated by RT-PCR with specific primers RNAs coding for mouse Elkl and SAP1. We subsequently cloned the mouse cDNAs. The protein sequences of Net, Elkl, and SAP1 are similar, with an amino-terminal ets domain and two further regions of homology, B and C (Fig. 1B).

Net is not the mouse homolog of human Elkl or SAP1. RNAs for Elkl and SAP 1 from Ras-transformed NIH-3T3 cells code for proteins that are almost identical to their human counterparts and clearly distinct from mouse Net (Fig. 1B). Human cells contain RNA coding for Net as well as SAP1 and Elkl. elkl and net genomic sequences are different (data not shown), elkl , net, and SAP1 mRNAs have different sizes and expression patterns (see above; Rao et al. 1989; Treisman 1994; note that an additional SAP2 protein is mentioned in Dalton and Treisman 1992). Etsl and Ets2, which belong to another ets subfamily, have distinct expression patterns and have opposite effects on gene expression (Bhat et al. 1989). Similarly, the Elkl-like factors have different expression patterns and presumably have different functions.

DNA binding by Net is inhibited by sequences outside the Ets domain

Net binds specifically to ets motifs, and its Ets domain (amino acids 1-91) is sufficient for binding, similar to





Giovane et al.

Figure 6. v-Mos and v-Src activate Net. Transfections contained PALx4, pTL2-NET (3 lag, lanes 1-11), pvMCBx2 (1 lag, lane 3; 3 lag, lane 4; 5 tag, lane 5), pvSrc5 (1 ~g, lane 6; 3 lag, lane 7; 5 wg, lane 8), pRCBx2 (1 lag, lane 9; 3 tag, lane 10; 5 vLg, lane 11 ), pSG5 (lane 1 ), pARCBx2 (lane 2). {ONC) Expressed oncoprotein; (VEC) amount of oncogene expression vector.

other Ets proteins (for Elkl and SAP1, see Dalton and Treisman 1992; Janknecht and Nordheim 1992; Rao and Reddy 1992a; Treisman et al. 1992; for others, see re- views by Seth et al. 1992; Macleod et al. 1993, Wasylyk 1994; Wasylyk et al. 1993). Sequences flanking the GGAA/T inner core of the ets motif affect its affinity for different members of the family, so that very divergent members of the family (e.g., Etsl and PU1) will bind exclusively to certain motifs. The Ets domains of Elk1,

Figure 8. Antisense net RNA stimulates the SRE. Transfec- tions contained pBL-CAT4 (lanes 1-5 and 11-15), SREm (lanes 6--10 and 16-20) plasmids, and 0 (lanes 1,2,6,7,11,12,16,17), 1 (lanes 3,8,13,I8), 5 (lanes 4,9,14,19), or 10 (lanes 5,10,15,20) }zg of AntiNet (made up to 20 lag with pSG5). The cells were incubated in medium with 0.05% FCS for 40 hr and then for 8 hr with either 0.05% (A, lanes 1-10) or 20% (B, lanes 11-20) FCS.

Figure 7. Opposite effects of Net and Elkl on the los SRE. Transfections contained pBL-CAT4 (lanes 1-3), SREm (lanes 4-81, pTL2-Net (1 lag, lane 5; 5 tag, lanes 2 and 6), pTL1-Elkl {1 lag, lane 7; 5 lag, lanes 3 and 8), pSG5 (5 tag, lanes 1 and 4).

SAP1, and Net are very similar (Elkl to SAP1, 80%; Elkl to Net, 77%; and SAP1 to Net, 79%). Net, Elkl, and SAP 1 appear to have substantially similar sequence pref- erences (Treisman et al. 1992; see Results; data not shown), although we cannot exclude that there are dif- ferences, as has been observed for SAP1 and Elkl (Treis- man et al. 1992).

DNA binding by Net is inhibited by sequences carboxy-terminal to the Ets domain, similar to Elk l and SAP1 (Janknecht and Nordheim 1992; Rao and Reddy 1992a; Treisman et al. 1992). Inhibition is relieved by deletions near the carboxyl terminus as well as in the middle of the protein (see Results). DNA binding by Etsl is similarly inhibited by sequences throughout the protein, although particular regions have more important effects (Lira et al. 1992; Wasylyk et al. 1992). The inhibitory sequences may mask the DNA-binding domain by forming an intramolecular complex or may otherwise





Ras activates Net

inhibit interactions with DNA (Lim et ah 1992; Wasylyk et al. 19921. Similar intramolecular mechanisms have been described for other factors [e.g., Myc (Kato et al. 1992); El2 (Sun and Baltimore 1991)]. Intramolecular inhibition appears to be a general mechanism to prevent DNA binding by regulatory factors in the absence of appropriate modifications or cofactors.

Net interacts with SRF to form a ternary complex on the fos SRE

A potential cofactor for Net is SRF. Net requires SRF to bind to the fos SRE and form a ternary complex. SRE- related sequences with stronger ets motifs form ternary complexes more efficiently. The B region of Net is required for ternary complex formation, although we cannot exclude that other sequences may also contribute. The related factors Elkl and SAP1 have similar properties (Hipskind et al. 1991; Dalton and Treisman 1992; Janknecht and Nordheim 1992; Rao and Reddy 1992a), consistent with the sequence similarity in the B region (Net to Elkl, 71%; Net to SAP1, 76%; Elkl to SAP1, 67%). However, Elk1 and SAP1 appear to form the ternary complex more efficiently, suggesting that Net may interact less efficiently with SRF, the SRE, or both. Ac- cording to the grappling hook model (Treisman et al. 1992) the closed conformation of Elkl and SAP1 is opened by SRF and DNA. In the ternary complex, essen- tial interactions with SRF are mediated by box B and possibly carboxy-terminal sequences, whereas the Ets domain interacts with DNA and makes no contacts with SRF. The sequence between the Ets domain and region B acts as a flexible tether, allowing the location and orientation of the ets motif to vary with respect to the SRF- binding site. The Elk l tether is dispensable for ternary complex formation (Janknecht and Nordheim 1992; Treisman et al. 1992). The tethers of Net, SAP1, and Elkl have different sequences and lengths, which could affect their functions, such as the ability to form complexes on different SREs.

Inhibition of transcription by Net

We found that Net inhibits transcription under a variety of conditions. Net expression decreases basal TK promoter activity through the inactive ets-responsive element from the stromelysin promoter (Wasylyk et al. 1991). In contrast, cEts 1 (p68) stimulates through this element. Net inhibits cEtsl(p68) activation, even though both proteins bind with similar although diminished affinity to the motif compared with a consensus motif (PEA3*; A. Pintzas, unpubl.). The activity of the SRE motif is also specifically inhibited by Net. Coexpression of SRF did not overcome this inhibition with a variety of reporters, in different serum conditions, and in different cell types (results not shown). This inhibition was not simply a consequence of overexpression of exogenous Net because decreasing endogenous Net levels with antisense net RNA specifically activates the SRE. Some negative transcription factors have negative domains

that are alanine rich. Interestingly, there is a sequence similarity between Net and the inhibitory region of the Drosophila Kr6ppel factor (Licht et al. 1990; AAAASA- FXXXS, Net amino acids 209-219 between domains B and C) that is absent in Elkl and SAP1. It remains to be seen whether Net has a separable negative-acting func- tion. Our results suggest that Net is a negative ets transcription factor but do not prove that Net is a true repressor. The transient expression assay is a deceptively simple system that masks an incredibly complex in-vivo environment in which the relative concentrations and specific activities of the proteins could be different.

Net and Elkl behave differently in a number of assays. Rao and Reddy (1992b) reported that Elkl acts as a transcriptional activator through a trimer of ets motifs (E74) in COS cells. Hill et al. (1993)found that Elkl expression squelched transcription as a result of titration of limiting components or formation of nonfunctional transcription complexes. However, ElM activated transcription when fused to a heterologous DNA-binding domain (Hill et al. 1993; Rao and Reddy 1993a). In agreement with these results, we found that Elkl stimulated transcription in both NIH-3T3 (see above) and COS cells (not shown). Furthermore, Elkl enhances activation by cEtsl{p68). Repression by Net does not appear to involve titration of MAPK because expression of MAPK has little or no effect on Net and Elkl, in contrast to studies in other cell types (Janknecht et al. 1993). These results suggest that Net has a different role from Elk l.

Net activates transcription when coexpressed with Ras or Src

Net activates transcription through the ets-responsive element when it is coexpressed with Ras, Src, and, to some extent, Mos. This suggests that Net participates in the Ras signal transduction pathway, perhaps as a direct target for one of its components. Elkl is phosphorylated in the C domain by serum/growth factor stimulation of cells, and by MAPK (Marais et al. 1993; Rao and Reddy 1993b), a downstream component of the Ras pathway. The carboxyl terminus of Elkl functions as a regulated transcription activation domain, whose activity in vivo is dependent on the integrity of the MAPK sites (Marais et al. 1993; Janknecht et al. 1993). The C domains of Net and Elkl are similar {Net to Elkl, 51%; Net to SAPla, 77%; Elkl to SAPla, 60%}, and the important MAPK sites are conserved between Net, Elkl, and SAPla (positions 359-365, 383-389, and 381-387, respectively; Ma- rais et al. 1993). Ras activation of Net may involve mechanisms similar to serum activation of Elkl. However, serum stimulation did not activate Net, even though we used different reporters and conditions (data not shown). Similarly, net antisense increased SRE activity even when it was stimulated by serum. This raises the in- triguing possibility that serum stimulation of the SRE may have a Ras-independent component. Alternatively, expression of oncogenic Ha-Ras may activate pathways that are unaffected by serum and endogenous c-Ras, and Net, but not Elkl, may be a target for this pathway.





Giovane et al.

There are at least three ets factors tha t in terac t di rect ly w i t h SRF and the SRE, suggest ing tha t t hey m a y media te the effects of different s ignals to th is i m p o r t a n t e lement . In addit ion, these factors can regulate t ranscr ip t ion in the absence of SRF and thus m a y media te signal trans- duc t ion to different p romoters i ndependen t ly of SRF. We have found tha t N e t ac t iv i ty is modu la t ed by Ras expression. The iden t i f i ca t ion of th is new factor should help in unde r s t and ing the m e c h a n i s m s by w h i c h Ras regulates signal t r ansduc t ion pa thways .

M a t e r i a l s a n d m e t h o d s

Purification of RNA and cDNA synthesis

Total RNA was extracted with hot phenol {Sambrook et al. 1989) and poly(A) + mRNA with an mRNA purification kit (Pharmacia). For cDNA synthesis, 5 ~g of total RNA (or 2 ~g of poly{A) + mRNA) was heated for 5 rain at 65~ cooled on ice, and immediately incubated with 1 ~g of random hexanucleotide primers [or 0.5 ~g of dT(23)] in 50 lzl containing 50 mM Tris-HC1 (pH 8.3), 75 mM KCI, 10 mM DTT, 3 mM MgC12, 1 mM each dNTP, and 400 units of Moloney murine leukemia virus (M-MLV) reverse transcriptase (GIBCO-BRL) for 1.5 hr at 37~

PCR with degenerate oligonucleotides

Degenerate oligonucleotides 1-3 {below) were chosen on the basis of the most conserved amino acid sequences of the Ets domain. One hundred nanograms of DT cell cDNA and 100 ng of oligonucleotides 1 and 2 or 1 and 3 were incubated in 50 ~1 containing 2.5 units of Taq polymerase (Perkin-Elmer Cetus), 10 mM Tris-HC1 (pH 8.3}, 50 mM KC1, 1.5 mM MgCI2, 0.001% gelatin, and 200 ~M dNTP for 10 min at 94~ and 30 cycles for 1 min at 94~ 2 min at 50~ and 3 min at 72~ Oligonucle- otides with human sequence were used to amplify murine Elk (12-13) and SAP (14-15} cDNAs from DT cells (100 and 110 bp, respectively). The PCR products were digested with EcoRI and BamHI, or EcoRI alone, cloned into the corresponding sites of pSG5 {Green et al. 1988), and sequenced. Sequences for oligonucleotides 1-3 and 12-15 are (1) 5'-gcggaattc- {C/T)TNTGGCA(A/G)TT{T/C){C/T)TNCTN(G/C)A-3'; (2)5'- cgcggatccNC(T/G)NGANA(A/G)(T/C)TTNTCNTC(A/G)TA- (A/G)TTCAT-3'; (3)5'-cgcggatcc(A/G)TC(A/G)TA{A/G)TA(A/ G)TANC(T/G)NA-3'; { 12) 5'-gatgcgaattcGGGAGCAGGAGC- ACCAGTCCA_A-3'; (13) 5'-gatgcgaattcGGGGTCAGGATA- ACCTGCAG-3'; (14) 5'-gatgcgaattcTGCCTTCCCTGGAA- GCCCCAACCT-3'; and {15) 5'-gatgcgaattcGAAGGTGTTC- TGGGAGGTTCCT-3'. In these sequences, N refers to any deoxynucleotide. The sequences in lowercase letters were added to facilitate cloning.

Screening of cDNA libraries

The radioactive probe was synthesized by PCR as described above in 40 ~1 with 500 ng of Net-specific oligonuceotides 4 and 5 {see below), 5 ng of pSG5 +net ets domain sequences, 200 ~M dNTP, 20 ~1 of [a~P]dCTP (3000 Ci/mmole; 3.3 ~M) for 10 min at 94~ then 30 cycles of 1 min at 94~ 2 min at 60~ and 3 min at 72~ (8chowalter and Sommer 1989). The 140-bp product was purified by chromatography on Sephadex G50 (Sambrook et al. 1989). Sequences for oligonucleotides 4 and 5 are (4) 5'-CTG- GACCAGAAACATGACCACCTC-3' and {5) 5'-CGCTCTGC- TCAGCTTGTCGTAGTT-3'.

Two eDNA libraries [10-day mice embryos, random primed;

ES cells, oligo(dT) primed; cloned into the EcoRI site of KZAPII (Stratagene)], which were shown to contain the 140-bp novel Ets domain sequence by PCR with oligonucleotides 4 and 5 (10 s phage/PCR, conditions as above), were screened by classical methods (Sambrook et al. 1989) with the 140-bp probe. Positive eDNA clones were excised in vivo {Short et al. 1989), and the cDNAs [in pBluescript SK (-)] were sequenced by the dideox- ynucleotide chain-termination method either manually (Win- ship 1989) or with an automatic sequencer {Applied Biosys- terns). Five clones from the embryo library {numbers 12, 17, 18, 25, and 26) were shown to have one 0.47-kb insert and two identical 1.0- and 2.43-kb inserts, respectively. Clone 12, which contains the 5' end, was sequenced on both strands. Clones 17 and 18, which contain the ATG (Fig. 1) and stop just before the TGA, were sequenced on one strand. Clones 25 and 26 were sequenced on one and both strands, respectively. Two clones from ES cells {numbers 3 and 5), which had identical 1.21-kb inserts including the ATG but not the TGA, were sequenced on one strand. Sequences 1--473 are from clone 12, 113-1325 are from clones 3 and 5, 268-1270 are from clones 17 and 18, and 220-2658 are from clones 25 and 26.

A set of specific oligonucleotides that was based on the Elk and SAP murine sequences cloned from DT cell was synthesized to screen 10-day mouse embryo and ES cell eDNA libraries by PCR. Two oligonucleotides that were based on K ZapII sequence around the cloning site of the eDNA, and nested primers from murine elk and SAP were used to amplify murine elk and SAP cDNAs. The cDNA fragments were cloned and sequenced. A HeLa cell random-primed cDNA ~ ZAPII library was screened with a murine net eDNA probe under low strin- gency conditions.

Construction of recombinants

- pTL2-Net: net cDNA, with a consensus Kozak sequence GC- CACC replacing sequences 5' to the ATG (nucleotide 295, Fig. 1) in pTL2. An XmaI-BglII fragment generated by PCR {as above) with oligonucleotides 6 and 7 was ligated with a BglII- EcoRI fragment from Net-Bluescript (eDNA from clone 26) between the XmaI and EcoRI sites of pTL2 (pSG5 with an ex- tended polylinker between EcoRI and BglII). The PCR-derived insert was sequenced. Sequences for oligonucleotides 6 and 7 are (6) 5'-aattcccgggGCCACCATGGAGAGTGCAATCACGC- TGTGG-3' and {7) 5'-ACACAGACGAAACGCTGGCAGC- ATTTGCCA-3'. -pSG5-Net: The complete eDNA {clone 26) in the EcoRI site of pSG5. -SREm: Double-stranded sequence of the SRE mutated in the FAP site TCGACAGGATGTCCATATTAGGACATCTGCcT- tAGC was inserted in the SalI site of the pBL-CAT4 polylinker in the sense orientation and sequenced. - p 6 0 1 D - - N e t and p601D-anti-Net: Net eDNA in either orientation in the EcoRI site of p601D. - p 6 0 1 D : The 345-bp rat f3-actin promoter region in the place of the SV40 promoter region of pTL2 {Beddington et al. 1989; Lufkin et al. 1992). -pTL 1-elk: Human HindIII-BamHI elkl cDNA from pT7-elkl (Morais et al. 1993) between the HindIII and BglII sites of pTL1.

RACE

For 5' end amplification, poly (A) § mRNA was reverse tran- scribed, chromatographed on a Sephadex GS0 to remove free nucleotides, and ethanol precipitated with 1 ~g of glycogen. The cDNA was tailed in 25 ~1 containing 0.1 mM sodium cacodyl-





Ras activates Net

ate, 2 mm MnClz, lmM dATP, and 15 units of terminal deoxy- nucleotidyl transferase (Boehringer) for 1 hr at 37~ extracted with phenol and chloroform, ethanol precipitated, and resus- pended in 20 }xl of TE. Tailed eDNA (2 ~xl) was amplified by two rounds of PCR. First round: oligonucleotides 8 and 5, 10 rain at 94~ 20 cycles of 1 min at 94~ 2 rain at 40~ and 3 min at 72~ Second round: oligonucleotide 8 and internal oligonucleotide 9 with 5 ~1 of the first round, 10 min at 94~ 30 cycle of 1 min at 94~ 2 rain at 60~ 3 min at 72~

For 3'-end amplification (the conditions were the same as for 5'-end amplification with 100 ng eDNA and oligonucleotide 8 with 10 for the first round, and with the internal oligonucleotide 11 for the second round. Sequences for oligonucleotides 8-11 are (8) 5'-cgggaattctccgagTTTTTTTTTTTTTTTTTTT- T(N)-3'; (9)5'-ggggagatctGAGGTGCTCATGTTTCTGGTC- CAG-3'; {10) 5'-ACATTGTTTAAAGAACAGTTGTTCC-3'; and ( 11 ) 5'-gggagatctCAAGAAGGGCGTGGTGGGAAAC-3'.

RNA blots

Total mRNA (20 txg) from cell lines, isolated with hot phenol (Sambrook et al. 1989), was electrophoresed on denaturing 1% agarose (running buffer: 0.2 M MOPS at pH 7.0, 3 mM sodium acetate, 1 mM EDTA, and 17.9% formaldehyde), transferred in 20x SSC to Hybond N + (Amersham) and inspected visually (ethidium bromide-stained bands) before and after the transfer. DNA (50--100 ng} was labeled with random hexanucleotide primers (50-100 ng), 16 ~tM dNTPs, 5 ~1 [32p]dCTP {3000 Ci/ mmole) and 6.5 units of Klenow polymerase in 1 x incubation buffer {3x: 120 mM Tris-HC1 at pH 8.0, 12 mM MgC12, 24 mM [3-mercaptoethanol, 480 mM HEPES at pH 6.5, and 0.5 mg/ml of BSA) at 37~ for 2 hr and purified on Prime Quick Columns (Stratagene's protocol). Filters were prehybridized in 0.25 M NaPi (pH 7.0), 7% SDS, and 1 mM EDTA at 65~ for 1 hr, hybridized overnight (Church and Gilbert 1984), washed in 20 mM NaPi, 1% SDS, for several hours, and exposed to X-ray film or PhosphorImager plates for quantification with a PhosphorIm- ager Fuji Bio-Analyzer BAS2000.

In vitro transcription and translation

The templates for transcription by T7 RNA polymerase were either linearized pSG5-derived expression vectors {purified by proteinase K digestion, phenol-chloroform extraction, and ethanol precipitation), or PCR products with a T7 promoter (Kain et al. 1991). RNAs synthesized in the presence of m7GpppG for 1 hr at 37~ were analyzed by agarose gel electrophoresis, purified by phenol-chloroform extraction, and ethanol precipitated. Proteins were synthesized in rabbit reticulocyte lysates (Promega) with [3SS]methionine and optimized amounts of RNA, and analyzed by SDS-PAGE alongside prelabeled markers (Amersham) and fluorography (Amplify, Amersham). They were quantitated according to band intensities with the number of methionines taken into account. Equimolar amounts of proteins were used for mobility shifts.

Templates

Net: pTL2-NET+EcoRI. SAPlb: pT7~link-SAPlb+Spe I IDalton and Treisman 1992). SRF: pTL2-SRF+EcoRI. 122- 264 SRF: PCR, sequence MLPGGYGP . . . . . . . . TKDTLK. Cod- ng strand (CS) primer: 5'-ATGTTACCGGGGGGGCTACGG- GCCGG. Noncoding strand INCSI primer: 5'-AAGCT- TAATTCTCAGGCTTCAGTGTGTCCTTG. 1-344 NET: pSG5-NET + Xho I, sequence MESAI . . . . . GLFLA. 1-327 NET:

PCR, sequence MES . . . . . LPSG. CS primer: 5'-ATGGAGA- GTGCAATCACGCT. NCS primer: 5'-AAGCTTGAATTCT- CCTGAGGGGAGGGCTGGGC. 1-222 NET: pSG5-Net+ BglII, sequence MES . . . . VSA. 1-152 NET: PCR, sequence MES . . . . . SLE. CS primer: 5'-ATGGAGAGTGCAATCACG- CT. NCS primer: 5'-AAGCTTGAATTCCTCCAGGGAGTT- GATGGTGA. 1-131 NET: PCR, sequence MES . . . . . LKS. CS primer: 5'-ATGGAGAGTGCAATCACGCT. NCS primer: 5'- AAGCTTGAATTCACTTTTGAGGGAGGACAAGC. 1-91 NET: PCR, sequence MES . . . . . PDIL. CS primer: 5'- ATGGAGAGTGCAATCACGCT. NCS primer: 5'-AAGCT- TGAATTCCAGGATATCCGGGAAAGAG.

Mobility shift assays

Proteins (1-5 ~1) brought to a constant volume with mock reticulocyte lysate (incubated without added RNA) were incubated for 45 min at 25~ in 20 ~1 of either 20 mM HEPES (pH 7.9), 20% glycerol, 0.1 mM EDTA, 2.5 mM DTT, 1 ~g/20 }xl of poly [d{I-C)], 50 mM KC1 with excess PEA3* probe, or 5 m~a HEPES (pH 7.9), 2.5 mM MgC12, 2.5 mM EDTA, 5 mM NaC1, 2 mM spermidine, 2.5 mM DTT, 1 ~g/20 Ixl of poly[d{I-C/], 2 ~g/lxl of BSA with excess SRE probes. The samples were loaded immediately on prerun [30 rain at 15 mA (75 V)] 4% or 6% polyacrylamide (bis:acrylamide, 1:291 gels in 0.25 x TBE and run for 60 min at 30 mA (150 V) with recirculating buffer at 20~

Oligonucleotides used for mobility shifts: PEA3*, 5'- TC- GAGCCGGAAGTGACGTCGA-3'; PEA3* mut, 5 ' -TCGAG- CATGAAGTGACGTCGA-3'; SRE, 5'- TACACAGGATGTC- CATATTAGGACA-3' spanning the human c-fos promoter from -324 to -300 (Janknecht and Nordheim 1992); G oligo, 5'- GCCCAATGCCGGAAATTGCCCATATAAGGACTCTA- GA-3' selected oligonuceotide with Elkl and SRF motifs (Tre- isman et al. 1992); G mut-wt, 5'-GCCCAATGACTGAAAT- TGCCCATATAAGGACTGTAGA-3' (Treisman et al. 1992); G wt-mut, 5'-GCCCAATGCCGGAAATTGCCCATCGGCTG- ACTCTAGA-3' (Treisman et al. 1992).

Only one strand is shown. The complementary strands form complete double strands with blunt ends. Oligonucleotides were 5'-labeled by T4 polynucleotide kinase and purified on native 10% polyacrylamide gels.

Antibodies and protein (Western) blots

Net and Elkl antisera were raised in rabbits against the ovalbu- min-MBS-coupled peptides [200 ~xg/injection, cysteine (C) was added for coupling): PB263, 385-1C)HMPVPHPSLDRAPSPVLL- SPSSQKS-409 (Net); PCl l , 411-(C)SVDGLSTRVVLSPGPQK- 427 (Elkl).

For Western blots, total ceU extracts from transiently transfected COS (20 ~g) or NIH-3T3 {100 p,g) cells were electrophoresed on 10% SDS-polyacrylamide gels, blotted to nitrocellu- lose filters, and revealed with NET(375)- or Elk(512)-specific antibodies and the ECL detection kit (Amersham's protocol).

Specific DNA-binding affinities of the &NET mutants for the PEA3* probe were estimated from the amount of radioactive probe that formed a complex in mobility shifts with a known amount of protein estimated by SDS-PAGE of proteins labeled with [3SSlmethionine in reticulocyte lysates {with the number of methionines taken into account).

Cell culture and transfections

NIH-3T3 C11 cells (the NIH-3T3 subline was used originally to derive the Ras-transformed line DT) maintained in Dulbecco's modified Eagle medium (DMEM) with 10% fetal calf serum





Giovane et al.

were transfected by the calcium phosphate method with 20 ~g/ 100-ram dish of plasmid DNA containing 2 ~g of pCHl l0 (transfection efficiency control), 1 ~g of reporters, expression vectors (see figure legends), and the appropriate amounts of control vectors {pSGS, pARCBx2). The cells were split, incubated with precipitated DNA for 24 hr, washed twice with DMEM, incubated with DMEM plus 0.05% fetal calf serum for 24 or 48 hr (see Fig. 8), scraped, freeze-thawed three times in solution A (15 mM Tris-HC1 at pH 7.9, 60 mM KC1, 15 mM NaC1, 2 mM EDTA, 0.15 mM spermine, 1 mM DTT, 0.4 mM PMSF), and cen- trifuged./3-Galactosidase activity was measured first to correct for transfection efficiency. For CAT assays, samples were heated to 65~ for 10 rain, acetylated chloramphenicol was sep- arated by thin-layer chromatography with a 95% chloroform/ 5% methanol solvent, and radioactivity was quantitated with a Fuji PhosphorImager. Experiments were repeated at least three times. In the figures, one representative CAT assay is shown, and error bars on the graph indicate one standard deviation.

A c k n o w l e d g m e n t s

We thank Elmar vom Baur and Sissy Fath for their contribution to the isolation of net cDNA during their ESBS final year proj- ect, R. Treisman, T. Lufkin, and C. Marshall for expression vectors, J.M. Gamier and J. Acker for cDNA libraries, the in- valuable help of the staffs of various services (animal house, cell culture, computer, drawing, oligonucleotide synthesis, peptide synthesis, photography, sequencing). A.G., A.P., S.-M.M., and P.S. are recipients of fellowships from the Association pour la Recherche sur le Cancer, the European Community, the Min- ist6re de la Recherche et Technologic, and the Human Frontier Science Program Organisation, respectively. We thank, for fi- nancial assistance, the CNRS, the INSERM, the Centre Hospi- talier Universitaire R6gionale, the Association pour la Recher- che sur le Cancer, the Fondation pour la Recherche M6dicale, the Ligue Nationale Franqaise contre le Cancer, and Bioavenir.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.

N o t e added in proof

Mouse net was cloned independently by T. Liberman and colleagues (ERP; Lopez, M., P. Oettgen, Y. Akbarali, U. Dendorfer, and T.A. Liberman. 1994. ERP, a new member of the ets transcription factor/oncoprotein family: Cloning, characterization, and differential expression during B-cell development. Mol. Cell. Biol. 14: 3292-3309) and human Net by R. Treisman and colleagues (SAP2; Treisman, R., pers. comm.).

The nucleotide sequence data for mouse Elk l, mouse SAP 1 a, and human Net have been submitted to the EMBL, GenBank, and DDBJ data libraries. The sequence data for mouse net have been submitted under accession number 7_32815.

References

Beddington, R.S.P., J. Morgemstem, H. Land, and A. Hogan. 1989. An in situ transgenic enzyme marker for the midges- tation mouse embryo and the visualization of inner cell mass clones during early organogenesis. Development 106: 37-46.

Bhat, N.K., K.L. Komschlies, S. Fujiwara, R.J. Fisher, B.J. Mathieson, T.A. Gregorio, T.A. Young, ].W. Kasik, and K. Ozato. 1989. Expression of mouse genes in mouse thymo- cyte subsets and T cells. J. Immunol . 142: 672-678.

Blenis, J. 1993. Signal transduction via the MAP kinases: Pro- ceed at your own RSK. Proc. Natl. Acad Sci. 90: 5889-5892.

Chambers, A.F. and A.B. Tuck. 1993. Ras-responsive genes and tumor metastasis. Crit. Rev. Oncogen. 4: 95-114.

Church, G. and W. Gilbert. 1984. Genomic sequencing. Proc. Natl. Acad. Sci. 81: 1991-1995.

Dalton, S. and R. Treisman. 1992. Characterization of SAP-l, a protein recruited by Serum Response Factor to the c-fos Se- rum Response Element. Cell 68: 597-612.

Frohman, M.A., M.K. Dush, and G.R. Martin. 1988. Rapid pro- duction of full-length cDNAs from rare transcripts: Ampli- fication using a single gene-specific oligonucleotide primer. Proc. Natl. Acad Sci. 85: 8998-9002.

Green, S., I. Issemann, and E. Sheer. 1988. A versatile in-vivo and in-vitro eukaryotic expression vector for protein engi- neering. Nucl. Acids Res. 6: 369-370.

Hill, C.S., R. Marais, S. John, J. Wynne, s. Dalton, and R. Tre- isman. 1993. Functional analysis of a growth factor-responsive transcription factor complex. Cell 73: 395--406.

Hipskind, R.A., V.N. Rao, C.G.F. Mueller, E.S.P. Reddy, and A. Nordheim. 1991. Ets-related protein Elk-1 is homologous to the c-los regulatory factor p62 Tcr. Nature 354:531-534.

Janknecht, R. and A. Nordheim. 1992. Elk protein domains required for direct and SRF assisted DNA-binding. Nucleic Ac- ids Res. 20: 3317-3324.

1993. Gene regulation by Ets proteins. Biochim. Bio- phys. Acta 1155: 346--356.

Janknecht, R., W.H. Ernst, V. Pingoud, and A. Nordheim. 1993. Activation of ternary complex factor Elk-1 by MAP kinases. EMBO J. 12: 5097-5104.

Kain, K., P.A. Orlandi, and D.E. Lanar, 1991. Universal promoter for gene expression without cloning: Expression PCR. BioTechniques 10:366--374

Kato, G.L., W.M.F. Lee, L. Chen, and C.V. Dang. 1992. Max: Functional domains and interaction with c-myc. Genes & Dev. 6: 81-92.

Khosravi-Far, R. and C.J. Der. 1994. The Ras signal transduction pathway. Cancer Metast. Rev. 13: 67-89.

Kozak, M. 1991. Structural features in eukaryotic messenger RNAs that modulate the initiation of translation. J. Biol. Chem. 266: 19867-19870.

Langer, S.J., D.M. Bortner, M.F. Roussel, C.J. Sherr, and M.C.Os- trowski. 1992. Mitogenic signaling by Colony-Stimulating Factor-1 and ras is suppressed by the ets-2 and restored by myc overexpression. Mol. Cell. Biol. 12: 5355-5362.

Lai, Z-C and G.M. Rubin. 1992. Negative control of photore- ceptor development in Drosophila by the product of the yan gene, an ets domain protein. Cell 70: 609-620.

Licht, J.D., M.J. Grossel, J. Figge, and U.M. Hansen. 1990. Drosophila Kriippel is a transcriptional repressor. Nature 346: 76-79.

Lim, F., N, Kraut, J. Frampton, and T. Graf. 1992. DNA binding by c-Ets-1, but not v-Ets, is repressed by an intramolecular mechanism. EMBO I. 11: 643-652.

Lufkin, T., M. Mark, C.P. Hart, P. Doll6, M. LeMeur, and P. Chambon. 1992. Homeotic transformation of the occipital bones of the skull by ectopic expression of a homeobox gene. Nature 359: 835-841.

Macleod, K., D. Leprince, and D. Stehelin. 1992. The ets gene family. Trends Biochem. Sci. 17: 251-256.

Marais, R., J. Wynne, and R. Treisman. 1993. The SRF accessory protein Elkl contains a growth factor-regulated transcriptional activation domain. Cell 73:381-393.

Moodie, S.A. and A. Wolfman. 1994 The 3Rs of life: Ras, Raf and growth regulation. Trends Genet. 10: 44-48.

Rao, V.N. and E.S.P. Reddy. 1992a. elk-1 domains responsible





Ras activates Net

for autonomous DNA binding, SRE: SRF interaction and negative regulation of DNA binding�9 Oncogene 7: 2335-2340.

�9 1992b. A divergent ets-related protein, Elk-1, recognizes similar c-ets-1 proto-oncogene target sequences and acts as a transcription activator�9 Oncogene 7: 65-70�9

�9 1993a. Aelk-1, a variant of Elk-l , fails to interact with the Serum Response Factor and binds to DNA with modu- lated specificity. Cancer Res. 53: 215-220.

�9 1993b. Elk-1 proteins are phosphoproteins and activators of mitogen-activated protein kinase. Cancer Res. 53: 3449-3454�9

Rao, V.N., K. Heubner, M. Isobe, A. AR-Rushdi, C.M. Croce, and E.S. Reddy. 1989. elk, tissue specific ets-related genes on chromosomes X and 14 near translocation breakpoints. Sci- ence 244: 66-70.

Robin, P., D. Trouche, P. Sassone-Corsi, and A. Harel-Bellan. 1991. Universality of c-fos transcriptional regulation: The dyad symmetry element mediates activation by PMA in T lymphocytes. Biochem. Biophys. Res. Com. 1 8 1 : 3 5 5 - 3 6 0

Sachs, A.B. 1993. Messenger RNA degradation in eukaryotes. Cell 74: 413-421.

Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular cloning: A laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Schowalter, D.B. and S.S. Sommer. 1989. The generation of ra- diolabeled DNA and RNA probes with polymerase chain reaction. Anal. Biochem. 177:90-94

Seth, A., R. Ascione, R.J. Fisher, G.J. Mavrothalassitis, N.K. Bhat, and T.S. Papas�9 1992. The ets gene family. Cell Growth Differ. 3: 327-334.

Short, J.M., J.M. Fernandez, J.A. Sorge, and W.D. Huse. 1988. Lambda ZAP: A bacteriophage lambda expression vector with in vivo excision properties�9 Nucleic Acids Res. 16: 7583-7600.

Sun, X-H. and D. Baltimore�9 1991. An inhibitory domain of El2 transcription factor prevents DNA binding in El2 ho- modimers but not in E12 heterodimers. Cell 64: 459-470.

Tel, H., I. Nihonmatsu, T. Yokokura, R. Ueda, Y. Sano, T. Okuda, K. Sato, K. Hirata, S.C. Fujita, and D. Yamamoto. 1992. pokkuri, a Drosophila gene encoding an E26-specific (Ets) domain protein, prevents overproduction of the R7 pho- toreceptor. Proc. Natl. Acad. Sci. 89: 6856-6860.

Treisman, R. 1992. The serum response element. Trends Bio- chem. Sci. 17: 423--426.

�9 1994. Ternary complex factors: Growth factor regulated transcription activators. Curr. Opin. Genet�9 Dev. 4: 96-101.

Treisman, R., R. Marais, and J. Wynne. 1992. Spatial flexibility in ternary complexes betwwen SRF and its accessory proteins. EMBO J. 11: 4631-4640.

Wasylyk, B. 1994. Transformation and transcription: The Ets family�9 Curr. Perspect. Mol. Cell. Oncol. (in press)�9

Wasylyk, B., S.L. Hahn, and A. Giovane. 1993. The ets family of transcription factors. Eur. J. Biochem. 211: 7-18.

Wasylyk, C., B. Wasylyk, G. Heidecker, M. Huleihel, and U.R. Rapp. 1989. Expression of raf oncogenes activates the PEA1 transcription factor motif�9 Mol. Cell. Biol. 9: 2247-2250.

Wasylyk, C., A. Gutman, R. Nicholson, and B. Wasylyk. 1991. The c-Ets oncoprotein activates the stromelysin promoter through the same elements as several non-nuclear oncopro- teins. EMBO J. 1 0 : 1 1 2 7 - 1 1 3 4

Wasylyk, C., J.-P. Kerckaert, and B. Wasylyk. 1992. A novel modulator domain of Ets transcription factors. Genes & Dev. 6: 965-974�9

Webster, N., J.R. Jin, S. Green, M. Hollis, and P. Chambon. 1988. The yeast UASG is a transcription enhancer in human HeLa cells in the presence of the GAL4 transactivator. Cell

52: 169-178�9 Winship, P.R�9 1989. An improved method for directly sequenc-

ing PCR amplified material using dimethyl sulphoxide. Nu- cleic Acids Res. 17: 1266.





10.1101/gad.8.13.1502Access the most recent version at doi: 8:1994, Genes Dev.

A Giovane, A Pintzas, S M Maira, et al. Net, a new ets transcription factor that is activated by Ras.

References

http://genesdev.cshlp.org/content/8/13/1502.full.html#ref-list-1

This article cites 46 articles, 14 of which can be accessed free at:

License

ServiceEmail Alerting

click here.right corner of the article or

Receive free email alerts when new articles cite this article - sign up in the box at the top

Copyright © Cold Spring Harbor Laboratory Press


http://genesdev.cshlp.org/lookup/doi/10.1101/gad.8.13.1502

http://genesdev.cshlp.org/content/8/13/1502.full.html#ref-list-1

http://genesdev.cshlp.org/cgi/alerts/ctalert?alertType=citedby&addAlert=cited_by&saveAlert=no&cited_by_criteria_resid=protocols;10.1101/gad.8.13.1502&return_type=article&return_url=http://genesdev.cshlp.org/content/10.1101/gad.8.13.1502.full.pdf

http://genesdev.cshlp.org/cgi/adclick/?ad=53706&adclick=true&url=http%3A%2F%2Fwww.sbsbio.com%2Fnews%2Fenglishnew%2Fproduct_full.php%3Fy%3D3%26cid%3D131%26pid%3D39



Net, a new ets transcription factor that is activated...

Documents

Transcript of Net, a new ets transcription factor that is activated...