Signal Transduction and the Ets Family of Transcription Factors

John S Yordy1 and Robin C Muise-Helmericks*,1,2

1Center for Molecular and Structural Biology, Hollings Cancer Center, Medical University of South Carolina, Charleston, South Carolina, SC 29403, USA; 2Department of Cell Biology and Anatomy, Medical University of South Carolina, Charleston, South Carolina, SC 29403, USA

Cellular responses to environmental stimuli are con- expression required for cellular growth, dierentiation trolled by a series of signaling cascades that transduce and survival. One group of downstream eectors of extracellular signals from ligand-activated cell surface these signaling pathways is the Ets family of transcrip- receptors to the nucleus. Although most pathways were tion factors. Ets family members can also be initially thought to be linear, it has become apparent that considered upstream eectors of signal transduction there is a dynamic interplay between signaling pathways pathways controlling the expression of a number of that result in the complex pattern of cell-type speci®c signaling components including both receptor tyrosine responses required for proliferation, dierentiation and kinases and intermediate signaling molecules. survival. One group of nuclear eectors of these The Ets family of transcription factors is de®ned by signaling pathways are the Ets family of transcription a conserved winged helix ± turn ± helix DNA binding factors, directing cytoplasmic signals to the control of domain (Papas et al., 1989; Wasylyk et al., 1993; gene expression. This family is de®ned by a highly Werner et al., 1995). Cumulative data have revealed conserved DNA binding domain that binds the core that this family of transcription factors are down- consensus sequence GGAA/T. Signaling pathways such stream eectors of the Ras-MAPK signaling cascades as the MAP kinases, Erk1 and 2, p38 and JNK, the PI3 (Wasylyk et al., 1998). Phosphorylation of Ets factors kinases and Ca2+-speci®c signals activated by growth by MAPKs controls their subsequent downstream factors or cellular stresses, converge on the Ets family of activity, protein partnerships, target speci®city and factors, controlling their activity, protein partnerships transactivation. Two major groups within this family and speci®cation of downstream target genes. Interest- have been extensively studied, the Ets group, including ingly, Ets family members can act as both upstream and Ets1, Ets2 and Pointed, and the ternary complex downstream eectors of signaling pathways. As down- factors (TCFs) which include Elk1, Sap1a, Sap1b, Fli1 stream eectors their activities are directly controlled by and Net. speci®c phosphorylations, resulting in their ability to Ets1, Ets2 and Pointed (Pt2) each contain a C- activate or repress speci®c target genes. As upstream terminal conserved DNA binding domain and an N- eectors they are responsible for the spacial and terminal domain referred to as the Pointed domain. temporal expression or numerous growth factor recep- This group of Ets family members has a single MAPK tors. This review provides a brief survey of what is phosphorylation site located near the Pointed domain known to date about how this family of transcription (Brunner et al., 1994; Wasylyk et al., 1997). TCFs, on factors is regulated by cellular signaling with a special the other hand, contain a transactivation domain that focus on Ras responsive elements (RREs), the MAP can be phosphorylated on multiple serine and kinases (Erks, p38 and JNK) and Ca2+-speci®c pathways threonine residues (Hipskind et al., 1994a; Treisman, and includes a description of the multiple roles of Ets 1994). Phosphorylation generally enhances their ability family members in the lymphoid system. Finally, we will to activate transcription by binding to speci®c discuss other potential mechanisms and pathways sequences termed Ras-responsive elements (RREs) involved in the regulation of this important family of and serum response elements (SREs) present in the transcription factors. Oncogene (2000) 19, 6503 ± 6513. promoters of many immediate early response genes. Although much work has been done to identify the Keywords: Ets; transcription; MAP kinase; Ras; signal relevant members of the major signaling cascades, there transduction is continual identi®cation of potential new members and modes of regulation in these signal transduction networks. While the eort to map these networks and Introduction identify new members continues, many of the major signaling pathways leading to the regulation of Ets A series of signaling pathways, including the MAP family of transcription factors have been described. A kinase (MAPK) pathways ERK1/2, p38 and JNK, as brief review of a portion of the existing literature will well as the PI3 kinase pathway, among others, are be given and used to re¯ect upon some of the general either activated by growth factors or by cellular stress themes understood at this time and possible future such as UV irradiation (Figure 1). These signaling directions of Ets signaling research. pathways transmit external stimuli to the nucleus and activate numerous transcription factors, resulting in both the temporal and spatial changes in gene Ras-responsive elements

The ®rst RRE was de®ned within the polyoma virus *Correspondence: RC Muise-Helmericks enhancer and is composed of consensus Ets binding Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6504

Figure 1 Schematic representation of the signaling pathways resulting in the control of Elk1 phosphorylation. For simplicity, although SRF is also regulated via these signaling pathways, the focus of the diagram is the eect on Elk1. Activation of the MAPK pathways Erk1/2, p38 and JNK result in the positive regulation of Elk1. Induction of Ca2+ ¯ux induces both a positive signal, resulting in the activation of Erk1/2 and a negative signal via the phosphatase calcineurin

sites (EBS) ¯anking binding sites for AP-1. AP-1 is a transactivation in the absence of MAPK activation heterodimeric transactivating complex whose speci®city (Wasylyk et al., 1990, 1997). results from the dimerization of dierent related Drosophila contains eight Ets-related family members subunits, e.g. c-Fos, c-Jun and is controlled by Ras (Reviewed by Hsu and Schultz, this issue), two of activation (Binetruy et al., 1991; Deng and Karin, which are pointed and yan. Pointed encodes two 1994; Smeal et al., 1991, 1992). Activation of the RRE transcriptional activator proteins, PntP1 and PntP2 within the polyoma enhancer was found to require a (Klambt, 1993; Scholz et al., 1997) while yan,a Ras-dependent activation of both Ets1 and AP1 transcriptional repressor, appears to produce a single activity (Wasylyk et al., 1990). Since this initial protein product (Lai and Rubin, 1992). Of the two description, the transcriptional activation of other proteins encoded by the pointed locus, PntP2 requires genes has also been shown to require RREs. Ras- MAPK phosphorylation to become active (Brunner et stimulated RRE promoter elements are able to activate al., 1994; O'Neill et al., 1994). Multiple roles have been the expression of genes known to be involved in assigned to pointed and yan including the development cellular transformation and migration, including uPA, of the wing, midline glia and retina. An orchestrated stromelysin-1 (MMP-3) and collagenase-1 (MMP-1) balance between PntP2 and Yan is necessary for proper (Aoyama and Klemenz, 1993; Bortner et al., 1993; Drosophila eye development, and both are targets of Chambers and Tuck, 1993; Grant et al., 1995; Wasylyk the Ras-MAPK pathway initiated by Sevenless, a et al., 1991; Wu et al., 1994). Ras-responsiveness can receptor tyrosine kinase (Brunner et al., 1994; Gabay be blocked by dominant-negative members of the AP1 et al., 1996; O'Neill et al., 1994; Rebay and Rubin, and Ets families (Bortner et al., 1993; Galang et al., 1995; Simon et al., 1991). Activation of the Ras- 1994; Gutman and Wasylyk, 1991; Langer et al., 1992; MAPK pathway leads to the phosphorylation of Yan, Lloyd et al., 1991; Wasylyk et al., 1994). Signi®cantly, which is then exported from the nucleus into the dominant-negative Ets proteins inhibit Ras-mediated cytoplasm where it is degraded (Gabay et al., 1996; cellular transformation without inhibiting normal cell Rebay and Rubin, 1995; Rogge et al., 1995). growth in some cell types (Langer et al., 1992; Wasylyk Concurrent with the degradation of Yan, PntP2 is et al., 1994). Ras-MAPK speci®c phosphorylation of phosphorylated by the same signaling cascade, stimu- Ets1 or Ets2 is necessary for the synergistic activation lating its transactivating capacity (O'Neill et al., 1994). of the RRE in conjunction with AP1 and mutations in This allows for the expression of downstream target either threonine 38 in Ets1 or threonine 72 in Ets2 genes necessary for eye development. In parallel with inhibit RRE activation (Yang et al., 1996). However, in the transactivation of RREs by Ets1 and Ets2, PntP2 some cell types Ets1 and Ets2 possess a high level of requires the interaction with the Drosophila jun

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6505 homologue, D-jun, for downstream function (Treier et shown to mediate a large number of extracellular al., 1995). signals, including tetradecanoyl phorbol acetate (TPA), Gene expression controlled by RREs has multiple lipopolysaccharide (LPS), serum, growth factors and levels of complexity. First, both the AP1 and Ets UV radiation (Janknecht et al., 1995; Treisman, 1995; components are coordinately controlled by Ras activa- Wasylyk et al., 1997). These extracellular signals tion. Second, there are over 25 dierent Ets family activate the SRE by phosphorylating its constituent members. This diversity, combined with the level of members. This phosphorylation either activates the diversity obtained by the number of dierent hetero- resident members and/or leads to an exchange of TCFs dimers that can be formed by individual Fos and Jun present on the SRE (Gille et al., 1992, 1995a; Hipskind family members, demonstrates the complexity of these et al., 1994b; Shaw, 1992; Zinck et al., 1993). The SRF systems and points to probable levels of control. Each binds the CArG box with high anity, and in general association must then be controlled for temporal and is required for the recruitment of the TCF to the low spatial expression, precise downstream target gene anity EBS (Giovane et al., 1994; Janknecht et al., expression and cell-type speci®city. The mechanisms 1993; Price et al., 1995). It has been reported, however, by which cellular signaling networks regulate this that SAP1a can bind and transactivate the c-fos SRE complex array of transcription factors are still not well in the absence of SRF (Masutani et al., 1997). understood. The TCF subgroup of the Ets family of transcription To make matters more complex, other Ras-depen- factors include family members such as Elk1, Sap1a, dent interactions, in addition to interactions with AP1, Sap1b, Fli1 and Net. Although each has unique are becoming apparent. The expression of Cyp24, an features, there are strong similarities in their structural enzyme necessary for the metabolism of vitamin D, is domains. As diagramed in Figure 1, the A (ETS) induced via two vitamin D response elements (VDREs) domain binds DNA, the B (B-box) domain interacts (Dwivedi et al., 2000). An adjacent RAS-dependent with SRF, the C (transactivation) domain and the D- Ets-binding site is required for maximal expression and domain, located between the B- and C-domains (Lopez is independent of AP-1 activity. Instead, a ternary et al., 1994; Price et al., 1995). Additional functional complex between the vitamin D receptor, retinoid X sequences within each TCF provide each transcription receptor and Ets1 results in the synergistic activation of factor with its own unique activity. For example, the Cyp24 promoter and requires RAS-dependent Sap1b lacks the C domain, suggesting that it may phosphorylation of Ets1 for maximal transactivation function as a dominant negative. Fli1 associates and (Dwivedi et al., 2000). forms ternary complexes with SRF via domains located One example of a tissue-speci®c Ras-dependent Ets on either side of the DNA binding domain (Dalgleish partnership is the interaction with Pit1. Pit1 is a and Sharrocks, 2000), indicating that its association pituitary-speci®c homeodomain protein that is required with SRF diers dramatically from that of Elk1 which for the development of dierent pituitary cell lineages binds via a single B-box located C-terminal to the Ets- and is thought to drive the expression of a series of domain (Shore and Sharrocks, 1994). How the pituitary-speci®c genes that are spatially regulated in expression and transactivation activities of these dierent pituitary cell types (Ryan and Rosenfeld, various TCFs are controlled, whether by cell-type 1997). One mechanism by which Pit1 directs the cell- speci®c expression patterns, phosphorylation by spe- type speci®c transcription of the rat prolactin gene is ci®c signaling cascades, formation of unique ternary via an association with Ets1. Expression of prolactin is complexes or speci®c DNA binding anities, remains controlled by thyrotropin-releasing hormone (TRH), to be completely elucidated. Despite these uncertain- which activates a G protein-coupled receptor, in turn ties, some degree of resolution to these questions has activating phospholipase C, and the Ras-MAPK been attained over the past few years. signaling cascade (Ohmichi et al., 1994a,b; Wang and The TCF Elk1 is a target of all three MAPK Maurer, 1999). The prolactin promoter contains a cascades and can activate the c-fos promoter in RRE that is composed of a Pit1 binding site cooperation with SRF (Gille et al., 1995a, b; Janknecht juxtaposed to an Ets-binding site. Ets1 is necessary and Hunter, 1997; Price et al., 1996; Treisman, 1996; for maximal synergistic expression of prolactin and this Whitmarsh et al., 1995). Multiple residues in the C- expression is dependent upon Ras-MAPK phosphor- domain of Elk1 are phosphorylated, leading to ylation of Ets1 (Bradford et al., 1995, 1996, 2000; enhanced DNA binding and transactivation (Price et Kimura et al., 2000). Both factors serve to engender al., 1996; Treisman, 1996). The fact that all three speci®city from a general signal transduction pathway, MAPK cascades can phosphorylate and activate Elk1 thus allowing for unique responses depending on the brings to bear the question of how this comes about: cell type. Do the p38, JNK and ERK pathways phosphorylate Elk1 by the same mechanism or do they employ disparate methods of activation? Indeed, it is becoming Ternary complex factor (TCF) group clear that both the kinase docking sites and the sequence/local context of the phosphoacceptor motifs The ternary complex factor (TCF) subgroup of the Ets is important in determining the substrate speci®city of family is made up of Ets proteins that interact with the the MAPKs (Derijard et al., 1994; Gupta et al., 1996; serum response factor (SRF) on a serum response Kallunki et al., 1996; Livingstone et al., 1995; Sluss et element (SRE) contained within many immediate early al., 1994; Treisman, 1994, 1996; Whitmarsh and Davis, response genes (Figure 1). By far the best characterized 1996). In partial answer to this question, it has been SRE is contained within the c-fos promoter. The SRE shown that the D domain is necessary for the ecient is made up of a CArG box and an adjacent Ets- phosphorylation of the C-domain by ERK2 (Yang et binding site (EBS) (Treisman, 1994) and has been al., 1998c). Further work has shown that the D-domain

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6506 is also required for JNK phosphorylation of the C- continuously growing NIH3T3 cells and, accordingly, domain (Yang et al., 1998b). Interestingly, ERK2 and the c-fos promoter is inactive. Antisense net results in JNK recognize dierent docking sites within the D- an increase in SRE activity in both actively growing domain, while p38 is not targeted to Elk1 via the D- and serum starved cells (Giovane et al., 1997). domain (Yang et al., 1998b). The exact docking site for Additionally, UV induction of the c-fos SRE requires p38 remains to be determined. These studies provide the activation of both the ERK and p38 MAPK evidence for a dual-speci®city docking motif within the pathways, suggesting that a certain threshold of signal D-domain. To add further complexity, Elk1 is is required for activation (Price et al., 1996). Taken responsible for the induction of the pip92 promoter together, this data suggests that there are sucient in both a MAPK-dependent and -independent manor amounts of activated TCFs to transactivate the SRE and appears to be phosphorylated by at least three but that the inhibitory action of Net-b on the c-fos unidenti®ed kinases that reside in a FGF-RAF- SRE must be overcome. This balance of positive and MAPK-independent pathway (Chung et al., 1998). negative regulators provides an elegant mechanism for Furthermore, the rate and length of activation are controlling both the rise and subsequent rapid decrease dierent: Activation of the MAPK-dependent pathway in activity needed for tightly controlling gene expres- results in a rapid and transient induction, while the sion. These ®ndings also support the hypothesis that MAPK-independent cascade induces rapid and pro- TCF exchange on the SRE is responsible, at least in longed transactivation (Chung et al., 1998). The part, for the activation of the SRE. identi®cation of these kinases may lead to the In addition to the negative regulation of TCF description of other docking motifs within Elk1. activity by Net and Net-b, other negative regulators Elk-1 provides a compelling example of how of serum response activities are beginning to be separate cascades can converge on the same eector described. For example, the Id helix ± loop ± helix using dierent recognition motifs. The speci®city of (HLH) proteins bind to the Ets DNA-binding domain each pathway may result in dierent activation kinetics of TCFs and inhibit their transactivation by disrupting of the target. It is likely that separate cascades also the DNA-bound TCF/SRF complexes (Yates et al., activate unique targets, both repressors and activators. 1999). The Id genes are expressed in a fashion that is Depending upon the cell type and circumstances of consistent with delayed-immediate early expression induction, the concert of proteins activated by a signal following mitogenic stimulation (Barone et al., 1994; transduction pathway may be dierent, and have Christy et al., 1991; Deed et al., 1993; Hara et al., dierent kinetics, resulting in a range of outcomes 1994). It is thought that SRE-dependent expression is dependent upon the speci®c set of activated proteins attenuated in a feedback loop by the induction of Id and the dynamics of this activation. The three dierent proteins. The Id proteins could promote dissociation of MAPK pathways, and other MAPK-independent the phosphorylated TCFs from the SRE, allowing the pathways, may activate Elk1, but the proteomic pro®le binding by unphosphorylated TCFs or dominant- of two dierent cell systems, or the same cells in negative TCFs (such as NET-b), resulting in inactiva- dierent environments, will likely be substantially tion of the SRE. dierent. Even within a single cell system, disparity Expression and activity of Elk1 in Xenopus laevis in the outcome is likely to occur depending upon also demonstrates the potential for autoregulatory which, or which combination, of the three dierent loops in the control of gene expression by Ets family MAPK pathways are activated. It will be interesting to members. It has been demonstrated that an FGF- see how these subtleties will be revealed in the coming MAPK signal transduction cascade in Xenopus laevis years. mediates mesoderm induction (Whitman and Melton, Net (ERP/SAP2) is a negative TCF that is switched 1992) and that one of the main eectors of this to a positive regulator via a Ras-dependent signal induction is Xbra (Cornell and Kimelman, 1994; (Ducret et al., 1999). Net contains two inhibitory LaBonne et al., 1995). Peptide growth factors such as domains, the NID domain that forms a helix ± loop ± Activin induce the expression of Xbra that, in turn, helix domain necessary for its strong repressive activity induces the expression of XeFGF. XeFGF activates (Giovane et al., 1994) and the CID domain that is the FGF-MAPK signaling cascade, which upregulates required for interaction with CtBP, a corepressor Xegr1 in an Elk1-dependent manner (Isaacs et al., (Criqui-Filipe et al., 1999). Signi®cantly, the transcrip- 1994; Panitz et al., 1998). Activation of the FGF- tional repression that results from interaction with MAPK pathway by XeFGF, induced by Xbra, is also CtBP requires deacetylase activity (Criqui-Filipe et al., responsible for the maintenance of Xbra expression 1999). Net is mainly a nuclear protein under both (Panitz et al., 1998). Abrogation of the FGF-MAPK serum-deprived and normal conditions. However, signaling cascade by any dominant negative member of activation of the stress activated MAPK, JNK, causes this pathway results in the loss of induction of both a rapid removal of Net from the nucleus (Ducret et al., Xbra and Xegr-1, while dominant negative Elk-1 1999). Therefore, Net activity is regulated on multiple results in only the loss of expression of Xegr-1 when levels by multiple signaling pathways. Ras activation a functional FGF-MAPK cascade is activated (Panitz switches Net from a negative regulator to a positive et al., 1998). In agreement with this observation, an one, JNK activation controls nucleocytoplasmic shuf- activated form of SRF can induce Xegr-1 expression ¯ing and protein association in combination with but not Xbra in the absence of activated MAPK deacetylation modulates the downstream eects of Net. (Panitz et al., 1998). This system demonstrates both a Net-b is a constitutive repressor that results from an signaling cascade leading to the activation of Elk-1 and alternative splice of net that is insensitive to Ras the utilization of a single transduction pathway to signaling (Giovane et al., 1994). Interestingly, Net-b is induce two separate genes via dierent eector the major TCF constituent of the c-fos SRE in transcription factors activated by MAPK.

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6507 Ca2+ immobilization tic cleavage. There is an apparent equilibrium between a folded and an unfolded conformation of Ets1 which An increase in intracellular Ca2+ is a ubiquitous signal involves an inhibitory domain located at the C- that triggers numerous cellular processes such as terminus of the protein. Phosphorylation increases the dierentiation and proliferation. Activation of most stability of the folded state, functionally blocking the growth factor tyrosine kinase receptors, as well as DNA binding domain (Cowley and Graves, 2000). stimulation of G-protein coupled receptors in response Interestingly, this autoinhibition can be relieved by the to hormones or neurotransmitters, results in the release association of another factor, CBFa2, with the of Ca2+ stores from the endoplasmic reticulum inhibitory domain of Ets1 (Goetz et al., 2000). (reviewed in Soderling, 1999). Ca2+ release is controlled In addition to the inhibition of DNA binding as by phosphoinositide-speci®c phospholipase C (PLC) shown for Ets1, Ca2+ signaling also negatively regulates isozymes (Figure 1). The b class of isozymes is activated the activity of Elk1. Ca2+/CaM is responsible for the via association with heterotrimeric G protein subunits activation of a serine/threonine phosphatase, calcineur- and the g class of isozymes is regulated by tyrosine in (Guerini, 1997; Lohse et al., 1995). Calcineurin is kinases. PLC catalyzes the hydrolysis of phosphatidyl- both a positive and negative Ca2+-dependent signaling inositol 4,5-bisphosphate to yield two products, both of component. It is known to dephosphorylate and which act as second messengers. The ®rst, inositol-1,4,5- activate NF-ATc, allowing for its translocation to the trisphosphate mediates the release of Ca2+ stores and nucleus (Luo et al., 1996) and is necessary for the the second, diacylglycerol, activates PKC. PKC can also activation of JNK in T cells (Werlen et al., 1998). As be activated directly by increased Ca2+ levels, and described above, Elk1 is a target of all three MAPK treatment of cells with Ca2+ ionophores stimulates PKC cascades and can activate the c-fos promoter as a TCF activity. Although the speci®city of Ca2+ signaling is still in cooperation with SRF. Although all three MAPK incompletely de®ned, recent studies have shed light on cascades can phosphorylate Elk1, activation of calci- the downstream eectors of Ca2+-mediated signaling. neurin leads to a direct dephosphorylation of Elk1 and Increases in intracellular Ca2+ often induce Ca2+- an inhibition of its activity on the c-fos promoter responsive signaling cascades that exhibit both stimu- regardless of the initiating cascade (Tian and Karin, latory and inhibitory properties. Association of Ca2+ 1999). These results suggest that Ca2+ ¯ux can be both and calmodulin (CaM) form a Ca2+/CaM complex that stimulatory and inhibitory. binds to and activates many signal transducers such as It has been proposed that this dierential signaling is Ca2+/CaM-dependent protein kinases (CaMKs) provided by changes in the range of Ca2+ ¯ux, inducing (Means, 2000; Tokumitsu and Soderling, 1996; Toku- high or low intracellular Ca2+ levels and/or dierential mitsu et al., 1997; Wayman et al., 1997). Binding of oscillatory frequencies. For example, while increases in Ca2+/CaM to the kinase alters its conformation and Ca2+ mediate the activation of NF-AT and NFkB opens the substrate-binding site blocked by an over- under conditions that produce high concentrations of lapping autoinhibitory domain. Initiation of this intracellular Ca2+, at lower levels only NFkBis cascade induces the activation of a number of activated (Dolmetsch et al., 1998). It is interesting that substrates, the best characterized being the transcription Ca2+ exerts both stimulatory and inhibitory in¯uences factors cAMP response element (CRE) binding protein on the c-fos promoter. This could be due to the (CREB), NF-AT, NFkB and SRF (Matthews et al., dierential kinetics of activation of CaMK and 1994; Rao et al., 1997; Stean et al., 1995; Sun et al., calcineurin. This could also be due to the dierential 1996; Villafranca et al., 1996). A great deal of work has Ca2+ sensitivity and anity of CaMK and calcineurin focused on the CRE and the SRE of the c-fos promoter, to CaM (Tian and Karin, 1999). The Kd for CaM both of which can be induced by Ca2+ (Chawla et al., binding to calcineurin is 0.1 nM and to CaMK is 20 ± 1998; Matthews et al., 1994; Miranti et al., 1995). 200 nM (Klee, 1991). Experimental evidence supports In addition to the phosphorylation of the pointed the second explanation where it was shown that 1 nM domain by MAPKs, Ets1 phosphorylation can also be ionomycin, a Ca2+ ionophore, does not induce the c-fos induced by calcium ¯ux. Ca2+-dependent phosphoryla- promoter but was sucient to inhibit activation of the tion targets six serine residues within a region coded for c-fos promoter by inducers that signal through the by exon 7 in the genomic sequence juxtaposed to the MAPK cascade (Tian and Karin, 1999). Therefore, DNA binding domain (Rabault and Ghysdael, 1994) there is a dierential sensitivity of calcineurin, a (Figure 2). This phosphorylation is reportedly con- negative regulator of the SRE and CaM kinase, a trolled by CaMKII in T and B cells (Valentine et al., positive regulator, in their response to Ca2+ ¯ux (Tian 1995) and by myosin light-chain kinase in astrocytes and Karin, 1999). In addition to the regulation of Ets (Fleischman et al., 1995), indicating that dierent family members by the MAPK pathways, signaling kinases may phosphorylate Ets1 depending on the through Ca2+ provides another level of complexity that cellular context. Comparison of wild type and mutant is required to coordinate gene expression based on Ets1 carrying serine to alanine mutations within this dierent extracellular stimuli. Therefore, both the region has indicated that the calcium-speci®c phos- MAPK and the Ca2+-dependent signaling pathways phorylation causes an inhibition of DNA binding are responsible for controlling Elk1 activity by (Rabault and Ghysdael, 1994). In agreement with this phosphorylation (Figure 1). data, the p42 splice variant of Ets1, lacking exon 7, has an increased capacity to bind DNA over the full-length isoform (Fisher et al., 1994). The mechanism by which Ets family and the control of the lymphoid system this phosphorylation inhibits DNA binding has recently been suggested by a series of DNA binding Hematopoiesis is the process by which pluripotential studies coupled with mutational analysis and proteoly- bone marrow stem cells proceed through dierent

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6508

Figure 2 Schematic representation of Ets1, Ets2 and the TCFs Elk1, Sap1a, Net, Netb, Fli1. The common structural domains are indicated. The DNA binding domain is shown as domain A, the B domain is the region that interacts with SRF, the C domain contains the transactivation domain, and the D domain is a region between the B and C domains. Net/Sap2 also contains an inhibitory domain (NID). The pointed domain and single MAPK phosphorylation site are shown for Ets1 and Ets2. The asterisks indicate the sites within Ets1 exon 7 that are phosphorylated in a Ca2+-dependent manner

stages of development, including lineage commitment, Studies using both in vitro and gene targeting and results in mature terminally dierentiated blood strategies have shown the requirement for PU.1 in cells. It is thought that activation of cytokine receptor myeloid and B cell development. For a complete signaling pathways alone does not induce the commit- description of the PU.1 knockout see Bartel et al., this ment of hematopoietic progenitors down a particular issue. One of the most marked eects of PU.1 gene lineage. This process also requires coordinate cell- targeting in this system is the lack of expression of a speci®c gene expression. The function of individual Ets number of cytokine growth factor receptors such as family members has been extensively studied in this GM-CSF, M-CSF (CSF-1), G-CSF and immunoglo- cellular compartment and this system serves as an bulin heavy and light chain genes (Anderson et al., excellent example by which the crosstalk between 1998; Scott et al., 1994). In addition, PU.1, in dierent family members can be controlled through a association with CREB, is necessary for expression of series of signaling pathways. This system also illus- the common signaling b chain (bc gene) of the IL-3, trates how certain Ets family members play important IL-5 and GM-CSF receptors in myeloid cells. Studies roles as upstream regulators of cellular signaling by using PU.1 de®cient myeloid cell lines in `add back' controlling the expression of a number of surface experiments have shown that expression of either the signaling receptors. G-CSF receptor or M-CSF receptor in the PU.1 null One example of an Ets family member that is an cells allows them to survive and proliferate in the important upstream regulator of cellular signaling is presence of the relevant growth factors, whereas PU.1 (Fisher and Scott, 1998). Expression of PU.1 is reintroduction of PU.1 allows for both the response restricted to cells of the hematopoietic lineage, to growth factors and the development of mature including B cells and macrophages, and regulates the macrophages (Anderson et al., 1999; DeKoter et al., expression of growth factor and cytokine receptors. 1998). These studies suggest that although PU.1 is an The tissue speci®c regulation of PU.1 expression is upstream eector of the cytokine growth factor controlled primarily by an octamer binding site and receptors that are required for cellular signaling, Ets-binding site within the PU.1 promoter (Kistler et growth and survival, PU.1 must play an additional al., 1995). In addition, PU.1 has been shown, as with role by either directly or indirectly regulating other other family members, to autoregulate its own genes important for macrophage dierentiation. promoter (Chen et al., 1995). In general the expression Another Ets member important in the dierentiation of PU.1 is usually unaected by most stimuli that of macrophages is Ets2. Ets2 is induced upon regulate gene activation and its expression levels stimulation with M-CSF (Boulukos et al., 1990). Ets2 remain relatively constant. PU.1 is primarily regulated is also expressed in the later stages of macrophage post-transcriptionally by phosphorylation (Lloberas et dierentiation, and its ectopic expression in immature al., 1999). myeloid cell lines can induce dierentiation. Ets2 is

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6509 regulated both transcriptionally by the classic Ras/Raf/ target genes required for hematopoietic cellular signal- MAPK pathway, via Elk1, and posttranscriptionally ing. The growing list of proteins that are controlled by by phosphorylation (Yang et al., 1996). Although the the Ets family of transcription factors attests to their exact function of Ets2 in the ®nal stages of macro- importance in the regulation of this cellular compart- phage dierentiation has not been completely deli- ment. Although the exact function that each individual neated, it is known that Ets2 inhibits apoptosis under Ets family member plays in this process is in itself very conditions of growth factor deprivation by inducing complex and has yet to be completely elucidated, a the transcription of the anti-apoptotic factor Bcl-XL great deal of information regarding the regulation of (Sevilla et al., 1999). this family can be gleaned from the identi®cation of A widespread role for the Ets family of transcription their downstream target genes. This system demon- factors in the induction or inhibition of apoptosis has strates the interplay between dierent family members been suggested due to the number of family members to coordinate not only hematopoiesis but also the involved in apoptosis: Fli1 and Erg in ®broblasts (Yi et activation of immune cells through cellular signaling al., 1997), PU.1 in erythroblasts (Yamada et al., 1997; mechanisms. Yamada and Oikawa, 1997) and the p42 splice variant of Ets1 in colon carcinoma cells (Li et al., 1999). It is clear that the expression of Ets2 in developing macrophages is Conclusions and perspectives not simply anti-apoptotic but must also control other downstream target genes important for macrophage In this review we have discussed our current under- dierentiation and proliferation. In addition to Ets2, standing of the role that Ets family members play Fli1 activity is also involved, at least in the proliferative within various signal transduction pathways. Most of potential of myeloid cells (Mora-Garcia and Sakamoto, the data to date has focused on the activation of 2000). Activation of the G-CSF receptor results in the various signaling pathways and their role in the upregulation of egr-1, a zinc ®nger transcription factor regulation of transcription. As discussed above, Ets that regulates proliferation and dierentiation (Krish- family members are regulated primarily by phosphor- naraju et al., 1995). Egr-1 is an immediate early gene ylation. One example is the TCF Elk-1, which can be whose expression is controlled by SRF, which is activated via the phosphorylation by kinases controlled constitutively bound to the egr-1 promoter. Stimulation by dierent signaling pathways such as p38, JNK and with G-CSF results in the expression of egr-1 by ERK. Inhibitory proteins such as the Id family and recruiting Fli1 to the SRE (Mora-Garcia and Sakamoto, Net-b are able to inhibit Elk1 binding, thus raising the 2000). The sequential expression and activity of PU.1, threshold for activation and potentially enhancing the Ets2, and Fli1 are pivotal for the regulation of deactivation of Elk1. Cascades resulting from small macrophage development. PU.1 expression and activity second messengers are able to further modulate a is required early during development of this lineage signaling cascade as in the example of the depho- partly because of its involvement in the expression of the sphorylation of Elk1 by the CaMK-dependent phos- appropriate growth factor receptors, the activation of phatase calcineurin. It is likely that signaling to other which results in induction of Ets2 expression and the family members is equally complicated. Members of activity of both Ets2 and Fli1. the Pointed group of Ets family members are Other Ets family members are known to modulate phosphorylated on multiple sites. As discussed, some the transcription of other signaling components of these phosphorylation sites have been mapped and necessary for hematopoiesis and lymphoid function. their eects determined. For example, Ets1 and Ets2 As with macrophages, the coordinate expression of are phosphorylated on a threonine residue near the multiple families is required for the appropriate pointed domain, and Ets1 also contains Ca2+-speci®c expression of downstream target genes involved in phosphorylation sites. However, these proteins are signaling within T cells. Ets1 is highly regulated in phosphorylated on other residues and the role(s) that mature T cells (Bhat et al., 1990), required for these other sites play in the regulation of individual lymphokine production (Romano-Spica et al., 1995) family members remains to be determined. Are there and expression of lck (Leung et al., 1993), a src-related other negative or positive regulatory sites, phosphor- tyrosine kinase that transduces the signal for the T cell ylation sites that induce changes in nucleocytoplasmic antigen receptor (reviewed in Kane et al., 2000). The transport or phosphorylation ± speci®c changes in expression of CD4, a cell surface protein important in protein partners? All these questions are still unan- the development and function of T cells is controlled, swered. Additionally, mechanisms that control the at least in part, by an Ets family member, Elf-1 selection of speci®c downstream target genes remain (Sarafova and Siu, 1999). In adult T cells CD4 is to be established. Each cell type may express multiple critical for antigen recognition, increasing the associa- family members, yet there is speci®city in target gene tion of the T cell receptor and its target and is also selection. Although there is probably some functional required to recruit lck to the T cell receptor complex redundancy, temporal changes in cellular signaling (Kane et al., 2000). Therefore, two family members, somehow converge onto this family of transcription Ets1 and Elf1, are required for the expression of two factors, resulting in appropriate downstream eects. proteins, lck and CD4, respectively, each required in Exactly how these transcription factors are coordi- collaboration to transduce the signal from the T cell nately controlled and balanced within a single cellular receptor complex. system remains to be identi®ed. Establishment of the mammalian immune system Other more speci®c mechanisms of control have yet requires a complex array of both growth factor to be delineated. For example, some reports suggest signaling and coordinated gene expression. Table 1 that the Ets family is involved in the regulation of the shows a list of some of the Ets-speci®c downstream cell cycle. Fli1 is known to control the transcription of

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6510 Table 1 Signaling proteins as Ets-speci®c downstream target genes in the hematopoietic system Target gene ETS-family Cell type Reference

GM-CSF PU.1 Myeloid (Anderson et al., 1998; Scott et al., 1994) M-CSF PU.1 Myeloid (Anderson et al., 1998; Scott et al., 1994) G-CSF PU.1 Myeloid (Anderson et al., 1998; Scott et al., 1994) CD4 ELK-1 T cell (Sarafova and Siu, 1999) LCK ETS-1 T cell (Leung et al., 1993) BTK PU.1 Myeloid, B cell, erythroid (Himmelmann et al., 1996) IL-1b PU.1 Myeloid (Yang et al., 2000) IL-2 receptor ELF-1 T cell (Serdobova et al., 1997) IL-2 ETS-1 T cell (Romano-Spica et al., 1995) IL-5 ETS-1/2 T cell (Blumenthal et al., 1999) IL-12 ETS-2 Myeloid (Gri et al., 1998) TIE-2 ETS-1 Lymphoid/endothelial progenitor (Dube et al., 1999) TCR-ZETA ELF-1 T cell (Rellahan et al., 1998) TNFa ETS-family Myeloid (Kramer et al., 1995) FCg receptor PU.1 Myeloid (Aittomaki et al., 2000) RANTES ETS-family Myeloid (Boehlk et al., 2000) Toll receptor PU.1 Myeloid (Rehli et al., 2000) Ig heavy and light chains PU.1 B cell (Perkel and Atchison, 1998)

the retinoblastoma protein, Rb, a critical cell cycle Growing evidence indicates that other signaling regulator (Tamir et al., 1999). Other important pathways not addressed in this review may also be regulators of the G1 to S transition, such as cyclin regulating the Ets family of transcription factors. Some D1, cdc2 and p21 are known to contain functional Ets studies suggest a crosstalk between the dierent binding sites within their promoter regions (Albanese et signaling pathways that control the Ets family, as al., 1995; Funaoka et al., 1997; Wen et al., 1995). described above, and the JAK/STAT signaling path- Additionally, Ets1 is known to be hyperphosphorylated way. The JAK kinases are historically activated by during mitosis (Fleischman et al., 1993). Given that interferons and other cytokines, although this pathway growth factor induction causes a rapid activation/ can also be activated by interleukins, RTKs, non- expression of many family members, it is likely that receptor tyrosine kinases and G-proteins. Activation of these transcription factors will play central roles on the JAK results in the phosphorylation and activation of G1 to S transition. The exact role of individual family the STAT family of transcription factors (Kotenko and members in cell cycle control remains an open Pestka, 2000). Recent results have shown that IL-2 question. induces the interaction between STAT5 and Ets1, Ets2, Recent evidence has indicated that protein acetyla- and Elf-1 (Lecine et al., 1997; Rameil et al., 2000) and tion regulates a number of cellular processes, especially that this interaction is necessary for the regulation of in the control of transcription. This modi®cation has gene expression in T cells. PU.1 is also known to been suggested to be analogous to phosphorylation and interact with STAT1 (Nguyen and Benveniste, 2000). its control of cellular signaling (Kouzarides, 2000). Ets family members may also play an important role in Although a `cascade' of acetylases has not been TGF-b signaling by controlling TGF-b receptor II described per se, recent evidence suggests that extra- expression (Chang et al., 2000). How individual Ets cellular signals are transduced to acetylases. Acetyla- family members respond to other signaling pathways is tion results in the modulation of the DNA binding also an open question. activity of a number of transcription factors, depending The new tools in biology such as cDNA microarrays on the position of the acetylated lysine residue (Gu and and quantitative proteomics will assist in addressing Roeder, 1997; Martinez-Balbas et al., 2000; Sterner et many of these remaining questions. Ultimately, a al., 1979). Protein : protein association can also be combination of approaches will be needed to more aected by acetylation, such as in the case of T cell fully understand cellular signaling and cellular physiol- factor and armadillo in Drosophila (Waltzer and Bienz, ogy in the context of the organism. The information 1998). To date, no Ets family member has been shown that will be generated in the next years will certainly to be directly under the control of acetylation. increase our understanding of the complex array of However, a number of members do require CBP/p300 signaling pathways responsible for the control of this for transactivation either as a co-adaptor protein, important family of transcription factors. acting as a scaold to bring together transacting factors, or for its HAT activity (Jayaraman et al., 1999; Korzus et al., 1998; Xu et al., 1998; Yang et al., 1998a). Given the growing amount of data suggesting the importance of this process in the control of Acknowledgments transcription, it follows that this group of transcription This manuscript is dedicated to the memory of Takis S factors will not remain untouched by this modi®cation. Papas, friend and advisor. We would like to thank DK Acetylation then, in addition to phosphorylation, may Watson for the critical review of this manuscript. John S provide a means for the selectivity of individual family Yordy is supported in part by the Wachovia Hollings members for their downstream target genes. Cancer Scholarship.

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6511 References

Aittomaki S, Pesu M, Groner B, Janne OA, Palvimo JJ and Dolmetsch RE, Xu K and Lewis RS. (1998). Nature, 392, Silvennoinen O. (2000). J. Immunol., 164, 5689 ± 5697. 933 ± 936. Albanese C, Johnson J, Watanabe G, Eklund N, Vu D, Dube A, Akbarali Y, Sato TN, Libermann TA and Oettgen Arnold A and Pestell RG. (1995). J. Biol. Chem., 270, P. (1999). Circ. Res., 84, 1177 ± 1185. 23589 ± 23597. Ducret C, Maira SM, Dierich A and Wasylyk B. (1999). Mol. Anderson KL, Smith KA, Conners K, McKercher SR, Maki Cell. Biol., 19, 7076 ± 7087. RA and Torbett BE. (1998). Blood, 91, 3702 ± 3710. Dwivedi PP, Omdahl JL, Kola I, Hume DA and May BK. Anderson KL, Smith KA, Perkin H, Hermanson G, (2000). J. Biol. Chem., 275, 47 ± 55. Anderson CG, Jolly DJ, Maki RA and Torbett BE. Fisher RC and Scott EW. (1998). Stem Cells, 16, 25 ± 37. (1999). Blood, 94, 2310 ± 2318. Fisher RJ, Fivash M, Casas-Finet J, Erickson JW, Kondoh Aoyama A and Klemenz R. (1993). Crit.Rev.Oncog.,4, 53 ± A, Bladen SV, Fisher C, Watson DK and Papas T. (1994). 94. Protein Sci., 3, 257 ± 266. Barone MV, Pepperkok R, Peverali FA and Philipson L. Fleischman LF, Holtzclaw L, Russell JT, Mavrothalassitis G (1994). Proc. Natl. Acad. Sci USA, 91, 4985 ± 4988. and Fisher RJ. (1995). Mol. Cell. Biol., 15, 925 ± 931. Bhat NK, Thompson CB, Lindsten T, June CH, Fujiwara S, Fleischman LF, Pilaro AM, Murakami K, Kondoh A, Fisher Koizumi S, Fisher RJ and Papas TS. (1990). Proc. Natl. RJ and Papas TS. (1993). Oncogene, 8, 771 ± 780. Acad. Sci. USA, 87, 3723 ± 3727. Funaoka K, Shindoh M, Yoshida K, Hanzawa M, Hida K, Binetruy B, Smeal T and Karin M. (1991). Nature, 351, 122 ± Nishikata S, Totsuka Y and Fujinaga K. (1997). Biochem. 127. Biophy. Res. Comm., 236, 79 ± 82. Blumenthal SG, Aichele G, Wirth T, Czernilofsky AP, Gabay L, Scholz H, Golembo M, Klaes A, Shilo BZ and Nordheim A and Dittmer J. (1999). J. Biol. Chem, 274, Klambt C. (1996). Development, 122, 3355 ± 3362. 12910 ± 12916. Galang CK, Der CJ and Hauser CA. (1994). Oncogene, 9, Boehlk S, Fessele S, Mojaat A, Miyamoto NG, Werner T, 2913 ± 2921. Nelson EL, Schlondor D and Nelson PJ. (2000). Eur. J. Gille H, Kortenjann M, Thomae O, Moomaw C, Slaughter Immunol., 30, 1102 ± 1112. C, Cobb MH and Shaw PE. (1995a). EMBO J., 14, 951 ± Bortner DM, Langer SJ and Ostrowski MC. (1993). Crit. 962. Rev. Oncog., 4, 137 ± 160. Gille H, Sharrocks AD and Shaw PE. (1992). Nature, 358, Boulukos KE, Pognonec P, Sariban E, Bailly M, Lagrou C 414 ± 417. and Ghysdael J. (1990). Genes Dev., 4, 401 ± 409. Gille H, Strahl T and Shaw PE. (1995b). Curr. Biol., 5, Bradford AP, Brodsky KS, Diamond SE, Kuhn LC, Liu Y 1191 ± 1200. and Gutierrez-Hartmann A. (2000). J. Biol. Chem., 275, Giovane A, Pintzas A, Maira SM, Sobieszczuk P and 3100 ± 3106. Wasylyk B. (1994). Genes Dev., 8, 1502 ± 1513. Bradford AP, Conrad KE, Tran PH, Ostrowski MC and Giovane A, Sobieszczuk P, Ayadi A, Maira SM and Wasylyk Gutierrez-Hartmann A. (1996). J. Biol. Chem., 271, B. (1997). Mol. Cell. Biol., 17, 5667 ± 5678. 24639 ± 24648. Goetz TL, Gu TL, Speck NA and Graves BJ. (2000). Mol. Bradford AP, Conrad KE, Wasylyk C, Wasylyk B and Cell. Biol., 20, 81 ± 90. Gutierrez-Hartmann A. (1995). Mol. Cell. Biol., 15, Grant PA, Thompson CB and Pettersson S. (1995). EMBO 2849 ± 2857. J., 14, 4501 ± 4513. Brunner D, Ducker K, Oellers N, Hafen E, Scholz H and Gri G, Savio D, Trinchieri G and Ma X. (1998). J. Biol. Klambt C. (1994). Nature, 370, 386 ± 389. Chem., 273, 6431 ± 6438. Chambers AF and Tuck AB. (1993). Crit. Rev. Oncog., 4, Gu W and Roeder RG. (1997). Cell, 90, 595 ± 606. 95 ± 114. Guerini D. (1997). Biochem. Biophys. Res. Commun., 235, ChangJ,LeeC,HahmKB,YiY,ChoiSGandKimSJ. 271 ± 276. (2000). Oncogene, 19, 151 ± 154. Gupta S, Barrett T, Whitmarsh AJ, Cavanagh J, Sluss HK, Chawla S, Hardingham GE, Quinn DR and Bading H. Derijard B and Davis RJ. (1996). EMBO J., 15, 2760 ± (1998). Science, 281, 1505 ± 1509. 2770. Chen H, Ray-Gallet D, Zhang P, Hetherington CJ, Gonzalez Gutman A and Wasylyk B. (1991). Trends Genet., 7, 49 ± 54. DA, Zhang DE, Moreau-Gachelin F and Tenen DG. HaraE,YamaguchiT,NojimaH,IdeT,CampisiJ, (1995). Oncogene, 11, 1549 ± 1560. Okayama H and Oda K. (1994). J. Biol. Chem., 269, Christy BA, Sanders LK, Lau LF, Copeland NG, Jenkins 2139 ± 2145. NA and Nathans D. (1991). Proc.Natl.Acad.Sci.USA, Himmelmann A, Thevenin C, Harrison K and Kehrl JH. 88, 1815 ± 1819. (1996). Blood, 87, 1036 ± 1044. Chung KC, Gomes I, Wang D, Lau LF and Rosner MR. Hipskind RA, Buscher D, Nordheim A and Baccarini M. (1998). Mol. Cell. Biol., 18, 2272 ± 2281. (1994a). Genes Dev., 8, 1803 ± 1816. Cornell RA and Kimelman D. (1994). Development, 120, Hipskind RA, Buscher D, Nordheim A and Baccarini M. 453 ± 462. (1994b). Genes Devt., 8, 1803 ± 1816. Cowley DO and Graves BJ. (2000). Genes Dev., 14, 366 ± 376. Isaacs HV, Pownall ME and Slack JM. (1994). EMBO J., 13, Criqui-Filipe P, Ducret C, Maira SM and Wasylyk B. (1999). 4469 ± 4481. EMBO. J., 18, 3392 ± 3403. Janknecht R, Cahill MA and Nordheim A. (1995). Dalgleish P and Sharrocks AD. (2000). Nucleic Acids Res., Carcinogenesis, 16, 443 ± 450. 28, 560 ± 569. Janknecht R, Ernst WH, Pingoud V and Nordheim A. Deed RW, Bianchi SM, Atherton GT, Johnston D, (1993). EMBO J., 12, 5097 ± 5104. Santibanez-Koref M, Murphy JJ and Norton JD. (1993). Janknecht R and Hunter T. (1997). EMBO J., 16, 1620 ± Oncogene, 8, 599 ± 607. 1627. DeKoter RP, Walsh JC and Singh H. (1998). EMBO J., 17, Jayaraman G, Srinivas R, Duggan C, Ferreira E, Swami- 4456 ± 4468. nathan S, Somasundaram K, Williams J, Hauser C, Deng T and Karin M. (1994). Nature, 371, 171 ± 175. Kurkinen M, Dhar R, Weitzman S, Buttice G and Derijard B, Hibi M, Wu IH, Barrett T, Su B, Deng T, Karin Thimmapaya B. (1999). J. Biol. Chem., 274, 17342 ± 17352. M and Davis RJ. (1994). Cell, 76, 1025 ± 1037.

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6512 Kallunki T, Deng T, Hibi M and Karin M. (1996). Cell, 87, Papas TS, Fisher RJ, Bhat N, Fujiwara S, Watson DK, 929 ± 939. Lautenberger J, Seth A, Chen ZQ, Burdett L, Pribyl L. et Kane LP, Lin J and Weiss A. (2000). Curr. Opin. Immunol., al. (1989). Curr. Top. Microbiol. Immunol., 149, 143 ± 147. 12, 242 ± 249. Perkel JM and Atchison ML. (1998). J. Immunol., 160, 241 ± KimuraA,OhmichiM,TasakaK,KandaY,IkegamiH, 252. Hayakawa J, Hisamoto K, Morishige K, Hinuma S, Price MA, Cruzalegui FH and Treisman R. (1996). EMBO Kurachi H and Murata Y. (2000). J. Biol. Chem., 275, J., 15, 6552 ± 6563. 3667 ± 3674. Price MA, Rogers AE and Treisman R. (1995). EMBO J., 14, Kistler B, P®sterer P and Wirth T. (1995). Oncogene, 11, 2589 ± 2601. 1095 ± 1106. Rabault B and Ghysdael J. (1994). J. Biol. Chem., 269, Klambt C. (1993). Development, 117, 163 ± 176. 28143 ± 28151. Klee CB. (1991). Neurochem. Res., 16, 1059 ± 1065. Rameil P, Lecine P, Ghysdael J, Gouilleux F, Kahn-Perles B Korzus E, Torchia J, Rose DW, Xu L, Kurokawa R, and Imbert J. (2000). Oncogene, 19, 2086 ± 2097. McInerney EM, Mullen TM, Glass CK and Rosenfeld Rao A, Luo C and Hogan PG. (1997). Annu.Rev.Immunol., MG. (1998). Science, 279, 703 ± 707. 15, 707 ± 747. Kotenko SV and Pestka S. (2000). Oncogene, 19, 2557 ± 2565. Rebay I and Rubin GM. (1995). Cell, 81, 857 ± 866. Kouzarides T. (2000). EMBO J., 19, 1176 ± 1179. Rehli M, Poltorak A, Schwarz®scher L, Krause SW, Kramer B, Wiegmann K and Kronke M. (1995). J. Biol. Andreesen R and Beutler B. (2000). J. Biol. Chem., 275, Chem., 270, 6577 ± 6583. 9773 ± 9781. Krishnaraju K, Nguyen HQ, Liebermann DA and Homan Rellahan BL, Jensen JP, Howcroft TK, Singer DS, Bonvini E B. (1995). Mol. Cell. Biol., 15, 5499 ± 5507. and Weissman AM. (1998). J. Immunol., 160, 2794 ± 2801. LaBonne C, Burke B and Whitman M. (1995). Development, Rogge R, Green PJ, Urano J, Horn-Saban S, Mlodzik M, 121, 1475 ± 1486. Shilo BZ, Hartenstein V and Banerjee U. (1995). Lai ZC and Rubin GM. (1992). Cell, 70, 609 ± 620. Development, 121, 3947 ± 3958. Langer SJ, Bortner DM, Roussel MF, Sherr CJ and Romano-Spica V, Georgiou P, Suzuki H, Papas TS and Bhat Ostrowski MC. (1992). Mol. Cell. Biol., 12, 5355 ± 5362. NK. (1995). J. Immunol., 154, 2724 ± 2732. Lecine P, Algarte M, Rameil P, Beadling C, Bucher P, Ryan AK and Rosenfeld MG. (1997). Genes Dev., 11, 1207 ± Nabholz M and Imbert J. (1997). Mol. Cell. Biol., 17, 1225. 2351. Sarafova S and Siu G. (1999). J. Biol. Chem., 274, 16126 ± Leung S, McCracken S, Ghysdael J and Miyamoto NG. 16134. (1993). Oncogene, 8, 989 ± 997. Scholz H, Sadlowski E, Klaes A and Klambt C. (1997). Li R, Pei H and Papas T. (1999). Proc. Nat. Acad. Sci. USA, Mech. Dev., 62, 79 ± 91. 96, 3876 ± 3881. Scott EW, Simon MC, Anastasi J and Singh H. (1994). Livingstone C, Patel G and Jones N. (1995). EMBO J., 14, Science, 265, 1573 ± 1577. 1785 ± 1797. Serdobova I, Pla M, Reichenbach P, Sperisen P, Ghysdael J, Lloberas J, Soler C and Celada A. (1999). Immunol. Today, Wilson A, Freeman J and Nabholz M. (1997). J. Exp. 20, 184 ± 189. Med., 185, 1211 ± 1221. Lloyd A, Yancheva N and Wasylyk B. (1991). Nature, 352, Sevilla L, Aperlo C, Dulic V, Chambard JC, Boutonnet C, 635 ± 638. Pasquier O, Pognonec P and Boulukos KE. (1999). Mol. Lohse DL, Denu JM and Dixon JE. (1995). Structure, 3, Cell. Biol., 19, 2624 ± 2634. 987 ± 990. Shaw PE. (1992). EMBO J., 11, 3011 ± 3019. Lopez M, Oettgen P, Akbarali Y, Dendorfer U and Shore P and Sharrocks AD. (1994). Mol. Cell. Biol., 14, Libermann TA. (1994). Mol. Cell. Biol., 14, 3292 ± 3309. 3283 ± 3291. Luo C, Shaw KT, Raghavan A, Aramburu J, Garcia-Cozar Simon MA, Bowtell DD, Dodson GS, Laverty TR and F, Perrino BA, Hogan PG and Rao A. (1996). Proc. Natl. Rubin GM. (1991). Cell, 67, 701 ± 716. Acad. Sci. USA, 93, 8907 ± 8912. Sluss HK, Barrett T, Derijard B and Davis RJ. (1994). Mol. Martinez-Balbas MA, Bauer UM, Nielsen SJ, Brehm A and Cell. Biol., 14, 8376 ± 8384. Kouzarides T. (2000). EMBO J., 19, 662 ± 671. Smeal T, Binetruy B, Mercola D, Grover-Bardwick A, Masutani H, Magnaghi-Jaulin L, Ait-Si-Ali S, Groisman R, Heidecker G, Rapp UR and Karin M. (1992). Mol. Cell. Robin P, Harel-Bellan A and Ait Si Ali S. (1997). Biol., 12, 3507 ± 3513. Oncogene, 15, 1661 ± 1669. Smeal T, Binetruy B, Mercola DA, Birrer M and Karin M. Matthews RP, Guthrie CR, Wailes LM, Zhao X, Means AR (1991). Nature, 354, 494 ± 496. and McKnight GS. (1994). Mol. Cell. Biol., 14, 6107 ± Soderling TR. (1999). Trends Biochem. Sci., 24, 232 ± 236. 6116. Stean NM, Bren GD, Frantz B, Tocci MJ, O'Neill EA and Means AR. (2000). Mol. Endocrinol., 14, 4±13. Paya CV. (1995). J. Immunol., 155, 4685 ± 4691. Miranti CK, Ginty DD, Huang G, Chatila T and Greenberg Sterner R, Vidali G and Allfrey VG. (1979). J. Biol. Chem., ME. (1995). Mol. Cell. Biol., 15, 3672 ± 3684. 254, 11577 ± 11583. Mora-Garcia P and Sakamoto KM. (2000). J. Biol. Chem., Sun P, Lou L and Maurer RA. (1996). J. Biol. Chem., 271, 275, 22418 ± 22426. 3066 ± 3073. Nguyen VT and Benveniste EN. (2000). J. Biol. Chem., 275, Tamir A, Howard J, Higgins RR, Li YJ, Berger L, 23674 ± 23684. Zacksenhaus E, Reis M and Ben-David Y. (1999). Mol. Ohmichi M, Koike K, Nohara A, Kanda Y, Sakamoto T, Cell. Biol., 19, 4452 ± 4464. Zhang ZX, Hirota K and Miyake A. (1994a). Biochem. Tian J and Karin M. (1999). J. Biol. Chem., 274, 15173 ± Biophys. Res. Commun., 201, 642 ± 648. 15180. OhmichiM,SawadaT,KandaY,KoikeK,HirotaK, Tokumitsu H and Soderling TR. (1996). J. Biol. Chem., 271, Miyake A and Saltiel AR. (1994b). J. Biol. Chem., 269, 5617 ± 5622. 3783 ± 3788. Tokumitsu H, Wayman GA, Muramatsu M and Soderling O'Neill EM, Rebay I, Tjian R and Rubin GM. (1994). Cell, TR. (1997). Biochemistry, 36, 12823 ± 12827. 78, 137 ± 147. Treier M, Bohmann D and Mlodzik M. (1995). Cell, 83, Panitz F, Krain B, Hollemann T, Nordheim A and Pieler T. 753 ± 760. (1998). EMBO J., 17, 4414 ± 4425. Treisman R. (1994). Curr. Opin. Genet. Dev., 4, 96 ± 101.

Oncogene Ets and cellular signaling JS Yordy and RC Muise-Helmericks 6513 Treisman R. (1995). EMBO J., 14, 4905 ± 4913. Whitmarsh AJ and Davis RJ. (1996). J. Mol. Med., 74, 589 ± Treisman R. (1996). Curr. Opin. Cell. Biol., 8, 205 ± 215. 607. Valentine MA, Czernik AJ, Rachie N, Hidaka H, Fisher CL, Whitmarsh AJ, Shore P, Sharrocks AD and Davis RJ. Cambier JC and Bomsztyk K. (1995). J. Exp. Med., 182, (1995). Science, 269, 403 ± 407. 1943 ± 1949. Wu H, Moulton K, Horvai A, Parik S and Glass CK. (1994). Villafranca JE, Kissinger CR and Parge HE. (1996). Curr. Mol. Cell. Biol., 14, 2129 ± 2139. Opin. Biotechnol., 7, 397 ± 402. XuL,LavinskyRM,DasenJS,FlynnSE,McInerneyEM, Waltzer L and Bienz M. (1998). Nature, 395, 521 ± 525. Mullen TM, Heinzel T, Szeto D, Korzus E, Kurokawa R, Wang YH and Maurer RA. (1999). Mol. Endocrinol., 13, Aggarwal AK, Rose DW, Glass CK and Rosenfeld MG. 1094 ± 1104. (1998). Nature, 395, 301 ± 306. Wasylyk B, Hagman J and Gutierrez-Hartmann A. (1998). Yamada T, Kondoh N, Matsumoto M, Yoshida M, Trends Biochem. Sci., 23, 213 ± 216. Maekawa A and Oikawa T. (1997). Blood., 89, 1383 ± Wasylyk B, Hahn SL and Giovane A. (1993). Eur. J. 1393. Biochem., 211, 7 ± 18. Yamada T and Oikawa T. (1997). Leukemia, 11 (Suppl 3), Wasylyk B, Wasylyk C, Flores P, Begue A, Leprince D and 480 ± 481. Stehelin D. (1990). Nature, 346, 191 ± 193. Yang BS, Hauser CA, Henkel G, Colman MS, Van Beveren Wasylyk C, Bradford AP, Gutierrez-Hartmann A and C, Stacey KJ, Hume DA, Maki RA and Ostrowski MC. Wasylyk B. (1997). Oncogene, 14, 899 ± 913. (1996). Mol. Cell. Biol., 16, 538 ± 547. Wasylyk C, Gutman A, Nicholson R and Wasylyk B. (1991). Yang C, Shapiro LH, Rivera M, Kumar A and Brindle PK. EMBO J., 10, 1127 ± 1134. (1998a). Mol. Cell. Biol., 18, 2218 ± 2229. Wasylyk C, Maira SM, Sobieszczuk P and Wasylyk B. Yang SH, Whitmarsh AJ, Davis RJ and Sharrocks AD. (1994). Oncogene, 9, 3665 ± 3673. (1998b). EMBO J., 17, 1740 ± 1749. Wayman GA, Tokumitsu H and Soderling TR. (1997). J. Yang SH, Yates PR, Whitmarsh AJ, Davis RJ and Sharrocks Biol. Chem., 272, 16073 ± 16076. AD. (1998c). Mol. Cell. Biol., 18, 710 ± 720. Wen SC, Ku DH, De Luca A, Claudio PP, Giordano A and Yang Z, Wara-Aswapati N, Chen C, Tsukada J and Auron Calabretta B. (1995). Experimental Cell Research, 217, 8± PE. (2000). J. Biol. Chem., 275, 21272 ± 21277. 14. Yates PR, Atherton GT, Deed RW, Norton JD and Werlen G, Jacinto E, Xia Y and Karin M. (1998). EMBO J., Sharrocks AD. (1999). EMBO J., 18, 968 ± 976. 17, 3101 ± 3111. Yi H, Fujimura Y, Ouchida M, Prasad DD, Rao VN and Werner MH, Clore GM, Fisher CL, Fisher RJ, Trinh L, Reddy ES. (1997). Oncogene, 14, 1259 ± 1268. Shiloach J and Gronenborn AM. (1995). Cell, 83, 761 ± Zinck R, Hipskind RA, Pingoud V and Nordheim A. (1993). 771. EMBO J., 12, 2377 ± 2387. WhitmanMandMeltonDA.(1992).Nature, 357, 252 ± 254.

Oncogene