Mastermind-like transcriptional co-activator-mediated Notch signaling is indispensable for maintaining conjunctival epithelial identity

REVIEW

Mastermind-like transcriptional co-activators: emerging roles in regulating

cross talk among multiple signaling pathways

AS McElhinny1, J-L Li2 and L Wu3

1Ventana Medical Systems Inc., Tucson, AZ, USA; 2Interdisciplinary Center for Biotechnology Research, University of Florida,Gainesville, FL, USA and 3Department of Molecular Genetics and Microbiology, Shands Cancer Center, University of Florida,Gainesville, FL, USA

A family of Mastermind-like (MAML) genes encodescritical transcriptional co-activators for Notch signaling,an evolutionarily conserved pathway with numerous rolesin both development and human diseases. Notch receptorsare cleaved upon ligand engagement and the intracellulardomain of Notch shuttles to the nucleus. MAMLs form afunctional DNA-binding complex with the cleaved Notchreceptor and the transcription factor CSL, therebyregulating transcriptional events that are specific to theNotch pathway. Here, we review recent studies that haveutilized molecular, cellular and physiological modelsystem strategies to reveal the pivotal roles of the MAMLproteins in Notch signaling. Unexpectedly, however,emerging evidence implicate MAML proteins as excitingkey transcriptional co-activators in other signal transduc-tion pathways including: muscle differentiation andmyopathies (MEF2C), tumor suppressor pathway (p53)and colon carcinoma survival (b-catenin). Thus, theMAML family appears to function in transcriptional co-activation in a multitude of cellular processes. It ishypothesized that MAML proteins mediate cross-talkamong the various signaling pathways and the diverseactivities of the MAML proteins converge to impactnormal biological processes and human diseases, includingcancers.Oncogene (2008) 27, 5138–5147; doi:10.1038/onc.2008.228

Keywords: Mastermind-like co-activator; Notch;MEF2C; p53; Wnt/b-catenin

Introduction

Broadly speaking, cancer results from deregulation ofthe normal cellular processes that mediate proliferation,differentiation and cell death programs. Under physio-logical conditions, cells undergo these processes byresponding to environmental stimuli through specificsignaling transduction pathways. For instance, path-

ways critical for embryonic and postnatal development(including Notch, Hedgehog, Wnt and transforminggrowth factor-b, to name a few) enable cells tocommunicate with the external environment and neigh-boring cells (Pires-daSilva and Sommer, 2003). Theexquisitely specific functional consequences of theresponses generated from the pathways depend uponthe induction and/or repression of downstream targetgene expression. Therefore, a critical yet largelyuncharacterized biological phenomenon is the conver-sion of cellular signaling inputs into the transcriptionalregulation of target genes.

Growing evidence indicates that transcription factorsrequire other molecules (for example, co-activators andrepressors) to regulate their activities (Hall andMcDonnell, 2005; Li et al., 2007; Yu and Reddy,2007). In this review, we focus on the Mastermind-like(MAML) family of transcriptional co-activators that areintegral to the Notch signaling pathway. We also discussunexpected roles for the MAML family in othersignaling pathways that have been revealed recently.These findings highlight pivotal regulatory functions forMAML co-activators in multiple signaling pathways,and also implicate them as players in mediating cross-talks among pathways.

Transcriptional co-activators: molecules that regulatesignaling-activated transcriptional events

In the past, much attention has focused solely on thetranscription factors that regulate gene expression inresponse to specific stimuli. However, a rapidly growingfield offers a more complex view of gene transcriptionand emphasizes the importance of other molecules(including co-activators and co-repressors) in transcrip-tional regulation (Lai, 2002; Hall and McDonnell, 2005;Li et al., 2007; Yu and Reddy, 2007).

In response to specific signals, the first step in geneexpression is to assemble transcriptional complexesupstream of the core promoter (including the TATAbox and transcriptional start site) by binding ofthe pathway-specific transcription factors. The basaltranscriptional machinery is thus positioned at thecore promoter, enabling transcription to proceed.

Correspondence: Dr L Wu, Department of Molecular Genetics andMicrobiology, Shands Cancer Center, University of Florida, 1376Mowry Road, Gainesville, FL 32610-3633, USA.E-mail: [email protected]

Oncogene (2008) 27, 5138–5147& 2008 Macmillan Publishers Limited All rights reserved 0950-9232/08 $30.00

www.nature.com/onc

http://www.nature.com/onc

Transcriptional co-activators are proteins that complexwith transcription factors to regulate specific target geneexpression. They promote transcriptional activation byvarious mechanisms including: assisting with the assem-bly of active transcriptional complexes, altering localchromatin structure for greater accessibility of thecomplex and communicating with the general transcrip-tion apparatus at target gene promoters. The co-activators are often required for maximal transcriptionalactivity, and are now recognized as important targets fordevelopmental and physiological signals in many diversebiological processes.

The MAML proteins, the focus of this review, aretranscriptional co-activators that are essential formediating cellular responses to the Notch signalingpathway. Recent research also implicates their co-activator functions in other signaling pathways, includ-ing those that involve MEF2C, p53 and Wnt/b-catenin.Thus, these recent data suggest broader roles forMAML proteins in regulating various biological pro-cesses.

The MAML family: conserved transcriptional activatorsfor the Notch pathway

Overview of Notch signaling-mediated transcriptionGenetic and cellular evidence clearly demonstrate thatthe MAML proteins are integral to the Notch signalingpathway by regulating the transcriptional activation ofNotch target gene expression. Notch signaling is ahighly conserved pathway that modulates stem/precur-sor cells in their response to developmental cues andinfluences cellular decisions within multiple tissues(Artavanis-Tsakonas et al., 1999; Bray, 2006; Ilaganand Kopan, 2007). This pathway regulates diversebiological processes, including neurogenesis (Laskyand Wu, 2005), myogenesis (Luo et al., 2005), vasculo-genesis (Anderson and Gibbons, 2007) and hemato-poiesis (Aster et al., 2008), to name a few. Notsurprisingly, aberrant Notch signaling (either deficientor abnormally increased) is linked to multiple develop-mental disorders and cancers (Roy et al., 2007). Exactlyhow one signaling pathway can exert such diverse effectson so many different cell types and cellular processesremains elusive. The multiplicity of Notch receptors andligands expressed in different cell types, coupled with thecombinations of tissue-specific transcription factors andtranscriptional co-activators activated in response tospecific stimuli, could at least partially explain thisamazing diversity.

Briefly, the Notch pathway involves the activation ofNotch receptors (Notch1-4 in mammals and Notch inDrosophila) by their ligands that are expressed onneighboring cells (Jagged1, Jagged2, Delta-like1 (Dll1),Dll3 and Dll4 in mammals or Delta and Serrate inDrosophila) (Figure 1a). Upon ligand binding, theNotch receptors undergo proteolytic processing, whichinvolves ADAM metalloproteases and the g-secretasecomplex, resulting in the intracellular domain of Notch

(ICN) being released from the cell membrane andshuttling to the nucleus. This process activates themajor downstream nuclear target for Notch, the CSLfamily of DNA-binding factors (CBF1/RBP-Jk inmammals, Su(H) in Drosophila and Lag-1 in Caenor-habditis elegans). The current prevailing model is thatNotch activation results in the displacement of CSLfamily members’ repressors (including CIR, N-CoR/SMRT, SPEN and KyoT2) and the recruitment oftranscriptional co-activators, including MAMLs, p300and GCN5/PCAF (Lai, 2002; Wu and Griffin, 2004;Kovall, 2007). The transcriptional complex with the corecomponents of ICN, CSL and MAMLs then is able toactivate target gene expression (including the basic helix-loop-helix HES gene family), which in turn regulates theexpression of tissue-specific transcription factors thatcontrol cell fates and other events. Currently, however,only a few Notch target genes have been identified andthese few cannot fully explain the diverse responses ofNotch receptor activation. It is also important to notethat evidence indicates that Notch signaling can bemediated in a CSL-independent manner. However, themechanisms responsible for this particular pathway areless characterized (Brennan and Gardner, 2002).

Identification of Maml proteins and investigations intotheir biochemical roles in Notch-mediated transcriptionThough we now realize the vital function of the MAMLco-activators in the Notch pathway, the exact mechan-isms and the roles of the MAML co-activators insignal-mediated transcription are still unclear. Theidentification of the MAML family members, theirstructural domains and known interactions are dis-cussed below.

The Mastermind homologues have been identified inDrosophila (Yedvobnick et al., 1988; Smoller et al.,1990), C. elegans (Petcherski and Kimble, 2000b),Xenopus (Katada and Kinoshita, 2003), mouse (Petch-erski and Kimble, 2000b; Wu et al., 2004) and human(Wu et al., 2000, 2002; Kitagawa et al., 2001; Lin et al.,2002). The Mastermind (mam) gene first was identifiedin Drosophila as one of the original group of ‘neuro-genic’ loci along with Notch, and loss of the function ofMam resulted in excessive neural cells at the expense ofepithelial cells (Yedvobnick et al., 1988; Xu et al., 1990),indicating that the molecule regulated neuronal cell-fatedecisions. The Mastermind gene has extensive geneticinteractions with Notch components such as Su(H) andDeltex, and was repeatedly identified as a Notchmodifier in genetic screens (Xu and Artavanis-Tsakonas,1990; Xu et al., 1990; Fortini and Artavanis-Tsakonas,1994; Go and Artavanis-Tsakonas, 1998). The gene wasdetermined to encode a novel, glutamine-rich nuclearprotein with novel protein sequence features suggestiveof a role in transcriptional regulation (Smoller et al.,1990). Indeed, immunohistochemical studies revealedMam binding on multiple sites of polytene chromo-somes, likely co-localizing with RNA polymerase andthe groucho co-repressor protein (Bettler et al., 1996).These studies identified Mastermind as an integral

MAML co-activator in Notch and other pathwaysAS McElhinny et al

5139

Oncogene

MAML

CSL

Co-A

ICNSTOP

CSL

ICN DNMAML1

DNMAML1(BD only)

MAML11

1016

Transcription Activation Domains (TADs)

TAD2TAD1

Basic

MAML2

1 1153

1 1133MAML3

1 1596

D. Mam

Acidic Acidic

p300/CBPNotch

Signal-receiving cell

ICN

ICN

Delta JaggedNotch1-4

Signaling cell

BD

Figure 1 Mastermind-like transcriptional co-activator (MAML) co-activates the Notch signaling pathway. (a) The Notch signalingpathway. Ligand binding between neighboring cells induces proteolytic cleavage of Notch receptors, producing the free intracellulardomain of Notch (ICN). ICN translocates to the nucleus and binds to a transcription factor, CSL. An active Notch transcriptionalcomplex, consisting of the ICN/MAML/CSL core components, is formed through displacement of the transcriptional co-repressorcomplex (Co-R), and likely, recruitment of additional, unidentified transcriptional co-activators (Co-A). The transcription of Notchtarget genes is then activated. (b) Structure and interacting domains of MAML co-activators. The human MAML family (MAML1, 2and 3) is structurally conserved with each other and with their Drosophila Mastermind orthologue (D. Mam). They contain an N-terminal basic domain (BD) and two acidic domains. The N-terminal domain is responsible for interactions with the ankyrin domain ofthe Notch receptors. The sequences after the BD have transcriptional activity and are denoted as transcriptional activation domains(TADs). In MAML1, there are two TADs: TAD1, containing the CBP/p300-binding site and TAD2, whose activities are required forNotch signaling in vivo. The diagram is not drawn to scale. (c) Inhibition of Notch-mediated transcription by expression of a dominant-negative MAML1 mutant, dnMAML1. dnMAML1 contains the BD of MAML1 but lacks the entire TADs; therefore, it retains theability to form a complex with Notch and CSL, but fails to activate transcription. It therefore interferes with endogenous Notchsignaling.


5140

Oncogene

component of the Notch signaling pathway in Droso-phila and pointed toward its involvement in transcrip-tional regulation. Moreover, the C. elegans orthologue,LAG-3, was confirmed to have Mastermind-like activityalthough it exhibits limited sequence homology toDrosophila Mastermind and human MAML1 (Petch-erski and Kimble, 2000a, b). LAG-3 formed a ternarycomplex with the LAG-1 DNA-binding protein and theintracellular domain of the receptor (GLP-1) in thenucleus and exhibited potent transcriptional activity;therefore, it was proposed to function as transcriptionalco-activator. The Xenopus Mastermind 1 (XMam1)orthologue has high homology with human MAML1and is required for primary neurogenesis by Notchsignaling (Katada and Kinoshita, 2003). Taken together,these studies laid the foundation for investigating theinvolvement of Mastermind and its orthologues intranscriptional regulation of the Notch signaling pathway.

Our group and others previously identified a family ofthree mammalian MAML co-activators as essentialregulators of Notch-induced transcriptional events(Wu et al., 2000, 2002; Petcherski and Kimble, 2000a;Fryer et al., 2002; Lin et al., 2002; Wu and Griffin,2004). The three human MAML genes encode nuclearproteins and have limited sequence homology with theirothologues, yet are highly conserved in their structure(Figure 1b). The MAML co-activators contain a highlyconserved N-terminal basic domain (BD) and two acidicdomains in their middle region and C terminus. Usingthe functional clues obtained from Drosophila studiesdiscussed above, the hypothesis that human MAMLproteins are involved in the transcriptional regulation ofNotch target gene expression was investigated. Indeed,the human MAML proteins are transcriptional co-activators because although they do not bind to DNAdirectly, they potentiate transcriptional activity whenlocalized to promoters by fusion with the Gal4 DNA-binding domain. Further studies indicate that MAMLproteins contain a BD that is responsible for Notch ICNbinding, and transcriptional domains required for thetranscriptional activation of Notch target genes(Figure 1b).

Biochemically, all MAML proteins appear to formstable DNA-binding complexes with ICNs and thetranscription factor CSL upon Notch activation (Linet al., 2002; Wu et al., 2002) and potentiate transcriptionof Notch target genes. These data led to a model inwhich MAML proteins are recruited to Notch/CSL onthe target gene promoters when Notch receptors areactivated, contributing to transcriptional activation. Insupport of this model, MAML1 mutants that are eitherdeficient in transcriptional activities or incapable ofbinding ICN1 behave as dominant negatives andinterfere with the ability of MAML1 to activate Notchsignaling (Wu et al., 2000). The isolation of a large,stable multiprotein complex containing the endogenousMAML1, ICN and CSL, is also consistent with MAMLproteins being core components of a Notch transcrip-tional complex (Jeffries et al., 2002). Finally, crystalstructure analyses of the Notch core complex confirmedthat the interaction of ICN and CSL creates the

recruitment site for MAML proteins (Nam et al.,2006; Wilson and Kovall, 2006). Mechanistically,MAML1 is required for chromatin-dependent transac-tivation by the reconstituted Notch ICN-CSL enhancercomplex in vitro (Fryer et al., 2002; Wallberg et al.,2002).

Currently, the underlying mechanisms responsible forMAML co-activation functions remain unclear andidentification of interacting molecules is required.Though more detailed mapping studies are needed,two transcriptional activation domains (TAD) of theMAML1 protein have been proposed (Figure 1b). (1)TAD1 is located in the region of aa 75–300 of MAML1,and contains a p300/CBP-binding site. Thus, MAML1recruits the chromatin modification enzyme p300/CBPand leads to nucleosome acetylation at Notch enhancersto activate transcription in vitro (Fryer et al., 2002).However, the recruitment of p300/CBP is not sufficientto activate the transcription of Notch target genes invivo, because an MAML1 mutant, MAML1 (1-302)containing TAD1 and retaining the p300 bindingactivity failed to activate Notch target genes andactually interfered with Notch signaling in vivo (Wuet al., 2000). Moreover, the MAML1 and p300interaction leads to mutual modifications: MAML1causes p300 phosphorylation, whereas p300 mediatesMAML1 acetylation (Fryer et al., 2002; Saint JustRibeiro et al., 2007). The exact functional significance ofthese post-translational modifications in Notch signal-based transcription remains unknown. (2) TAD2extends from the center of MAML1 to its C-terminalregion (aa 303–1016), which contains glutamine-richsequences. TAD2 is required for transcription in vivo.Therefore, unknown proteins that interact with TAD2likely are required for Notch target gene activation. Aprotein that interacts directly with MAML1 is CDK8that can cause ICN1 phosphorylation and degradation(Fryer et al., 2004). The exact binding site was notmapped, but neither MAML1 mutant with deletion aa75–300 nor 1–301 binds to CDK8. This suggests anadditional regulatory role for MAML1 to recruit CDK8to target gene promoters to degrade the Notch enhancercomplex, thereby terminating Notch signaling.

Although more studies have investigated MAML1, ithas been determined that both MAML2 and MAML3interact with Notch receptors and also interact withp300 (Lin et al., 2002; Wu et al., 2002; L Wu et al.,unpublished). Additional interacting proteins have notyet been identified. It is intriguing to speculate that thereare binding proteins specific for each MAML member,contributing to their differential activation and activ-ities. Therefore, much work is required to dissect thestructural domains of the MAML family and to identifytheir specific interacting partners before their functionalroles and mechanisms of actions can be elucidated.

Investigations into MAML co-activators that regulateNotch signaling in various biological processesThe three Maml members exhibit distinct expressionpatterns during embryonic development (Wu et al.,


5141

Oncogene

2002, 2004), supporting the hypothesis that they playspecific roles in different tissues. On the basis of reporterassays in cultured cells, the three human MAMLproteins have overlapping and differential activities interms of their abilities to activate Notch receptor-mediated transcription: MAML1 and MAML2 co-activate all four Notch receptors, whereas MAML3seems only to effectively co-activate Notch4. Thus,MAMLs have the potential to cooperate differentiallywith various Notch receptors in the activation of targetgenes, which could contribute to the wide range ofbiological effects upon activation of the Notch pathway.

Studies have begun to investigate the functional rolesof MAML family members in regulating Notch-basedtranscriptional events in multiple cellular processes. Onevaluable tool is a dominant-negative MAML1 mutant,DNMAML1 (Figure 1c). This mutant consists of a 62amino-acid peptide of the N-terminal BD of MAML1that structurally was shown to interact with the Notch–CSL transcriptional complex (Nam et al., 2006).DNMAML1 is a Pan-Notch inhibitor: it interferes withthe endogenous function of MAML proteins andinhibits transcriptional activation from all four Notchreceptors (Weng et al., 2003). Other strategies that havebeen successful in determining Notch roles in cellularprocesses is by the inhibition of Notch receptorprocessing by inactivating the g-secretase complex withspecific inhibitors (GSI), and development of mousemodels that are deficient of various Notch signalingcomponents.

These studies have shown, for example, that Notchsignaling has well-established roles in lymphoid devel-opment, although the exact functions of the four Notchreceptors are not equal. Loss of the Notch1 receptorperturbs T-cell development, yet the B-cell compartmentappears unaffected (Radtke et al., 1999). Conversely, thetargeted inactivation of the Notch2 receptor or Dll1ligand results in a loss of marginal zone B cells (MZBcells), but T-cell development appears normal (Saitoet al., 2003; Hozumi et al., 2004). Importantly, theseeffects are consistent with phenotypes caused by theinactivation of canonical Notch signaling in thesecompartments, because loss of the transcription factorCSL blocks both T-cell development and MZB cellgeneration (Han et al., 2002; Tanigaki et al., 2002). Notunexpectedly, then, the expression of DNMAML1 inbone marrow cells abrogated canonical Notch signaling,inhibiting both T-cell development and MZB cellgeneration (Maillard et al., 2004). Taken together, thesedata revealed that the MAML genes collectively areessential in mediating physiological Notch functions.

Importantly, DNMAML1 knock-in mouse modelsrecently were generated to investigate the in vivofunctional consequences of inhibiting the endogenousMAML functions in Notch signaling. Defects werereadily determined in several systems, including: T-helper cell development (Tu et al., 2005), skin (Prowelleret al., 2006), vascular smooth muscle (Proweller et al.,2007) and cardiovascular development (High et al.,2007). These studies revealed that MAML proteinsindeed are critical to mediate Notch signaling in

multiple tissues in vivo. It should be noted thatDNMAML1 seems to be specific for Notch signaling.However, the possibility that DNMAML1 might inter-fere with other Notch-independent functions cannot beruled out. More studies are required to characterize theeffects of DNMAML1 in light of newly discoveredNotch-independent MAML functions (discussed be-low).

Current studies have begun to dissect the roles ofindividual MAML co-activators in vivo. A mouse modelwith a targeted disruption of the Maml1 gene wasrecently reported (Shen et al., 2006; Oyama et al., 2007;Wu et al., 2007). The Maml1-knockout mice exhibitedseveral defects (including muscle defects which isdiscussed later). They remained small in size and diedwithin the perinatal period. The effect of Maml1knockout on lymphoid development was characterizedto evaluate its contribution in mediating Notch func-tions. Intriguingly, the Maml1 deficiency had noobvious effect on T-cell development, but did result inthe absence of MZB cells, a phenotype similar to thephenotype of Notch2 or Dll1 ligand deficienciesdescribed above. Moreover, after analyses of the Maml1heterozygote mice, there appeared to be a dosage effectfor Maml1 levels in MZB cell formation. These studiesrevealed an unexpected, in vivo role for Maml1 inmediating the signaling of a specific Notch receptor,Notch2. The specific requirement for Maml1 in the B-cell compartment could not be explained simply byexpression patterns of Mamls, because at least twoMaml members, Maml1 and 2 are expressed in splenic Bcells. Rather, it is hypothesized that Maml1 providesspecific molecular features to the Notch2 transcriptionalactivation complex that are not compensated by otherMaml members. These data favor the model thatindividual Maml co-activators contribute to the mole-cular specificity of Notch receptor functions in vivo.

Currently, the roles for Maml1 in mediating Notchsignaling in other tissues have not been carefullyexamined. However, one additional interesting findingis that there appear to be defects in a subpopulation ofmuscle precursors (satellite cells) that are responsible formuscle regeneration in Maml1-deficient mice (H Shenet al., unpublished). Therefore, we speculate that Maml1loss inhibits the Notch response that is known to beessential for the activation and expansion of satellitecells, the early steps of myogenic process. These ideaswarrant further investigation and are being tested.

In summary, the precise roles of individual MAMLco-activators in vivo, and whether they contribute to thediverse biological effects of Notch signaling, remain tobe addressed. Fortunately, detailed structural informa-tion is now available regarding the trimolecular complexbetween CSL, Notch1 and the N-terminal Notch-binding domain of MAML1 (Nam et al., 2006; Wilsonand Kovall, 2006). Therefore, investigating the struct-ural differences among various combinations oftranscriptional complexes containing distinct Notchand Maml family members will provide valuableinformation about their in vivo specificity. Finally,complete expression profiles of the three MAMLs in


5142

Oncogene

different tissues and at different development stages,as well as characterization of mouse models lackingeach co-activator individually and in combination,will provide a better understanding of the functionsof Maml co-activators in Notch signaling-based re-sponses.

Emerging Notch-independent functions of the MAMLproteins and their involvement with other signalingpathways

MAML1 co-activates MEF2C muscle-basedtranscriptional events and is required for myogenesisThe first indication that MAML functioned indepen-dently of the Notch pathway came from investigationsusing an Maml1 knockout mouse model (Shen et al.,2006). These mice display a severe perturbation ofoverall skeletal muscle structure consistent with amuscular dystrophy-like phenotype, including degenera-tion of myofibers and muscle necrosis. Maml1-nullmuscle embryonic fibroblasts (MEFs) failed toinitiate myogenic differentiation after MyoD transduc-tion, suggesting an intrinsic requirement for Maml1in myogenesis. Indeed, Maml1 is required formuscle differentiation from myoblast to myotubeformation, because Maml1 RNAi-mediated knockdownblocked the differentiation of C2C12 cells (a myoblastcell line). Maml1 also promotes myogenic differentia-tion, as C2C12 cells that overexpress MAML1 exhibiteddramatically enhanced levels of muscle myosin andformed strikingly large myotubes. These data indicatedthat MAML1 expression is essential for muscledifferentiation events, and for mediating muscle geneexpression.

The pro-myogenic activities of MAML1 were surpris-ing and unexpected, as previous studies established thatactivation of Notch signaling actually inhibited myo-genic differentiation in C2C12 cells (Kopan et al., 1994;Lindsell et al., 1995). In fact, Notch signaling is inhibitedby the Notch antagonist, Numb, during the later step ofmyogenesis, myogenic differentiation (Luo et al., 2005).Therefore, we hypothesized that Maml1 possessesNotch-independent functions that might account forits pro-myogenic activities. This led us to investigate thefunctional interaction of Maml1 and transcriptionfactors important for myogenic differentiation, resultingin the finding that Maml1 interacts with and co-activatesMEF2C, a critical muscle-specific transcription factor.Furthermore, our preliminary studies indicate thatmyogenin, also a key muscle transcription factor, is adirect target of Maml1, and its expression is regulatedby the Maml1/MEF2C interaction. Consistent withthese data, activation of the Notch pathway (by ligandstimulation or the ectopic expression of the constitu-tively activated form of ICN) inhibited MAML1-enhanced myogenesis in C2C12 cells (Shen et al.,2006). Biochemically, the blockage of MAML1’s pro-myogenic effects by Notch activation was associatedwith the recruitment of MAML1 away from MEF2C tothe Notch transcriptional complex.

These studies revealed novel and unique roles forMAML1 that are independent of the Notch signalingpathway. A model was proposed where the Notchpathway was active during myoblast proliferation, andthen silenced upon activation of myoblasts to differ-entiate into mature muscle. MAML1 then would‘switch’ to function as a potent transcriptional co-activator of MEF2C for muscle differentiation tran-scriptional events to occur. Importantly, these studiessuggested the exciting possibility that MAML familymembers may serve as co-activators for other pathwaysthan Notch, and perhaps mediate cross talks betweenvarious pathways and Notch (Shen et al., 2006).Other studies described below support this intriguinghypothesis.

MAML1 co-activates p53, which is required forp53-mediated germ-cell apoptotic responseThe p53 tumor suppressor mediates cell-cycle arrest andapoptosis in response to cellular stresses, includingDNA damage, hypoxia and oncogene activation, and isimportant in both normal developmental processes andcancer (Vousden and Lane, 2007). Another exciting,recent finding is that MAML1 exhibits Notch-indepen-dent functions in co-activating p53-dependent geneexpression, thereby potentially regulating p53-mediatedresponses (Zhao et al., 2007). Chromatin immunopreci-pitation assays indicate that MAML1 is a component ofthe activator complex that binds to native p53-responseelements within the promoters of p53 target genes. TheMAML1/p53 interaction involves the N-terminal regionof MAML1 and the DNA-binding domain of p53.Overexpression of wild-type MAML1 enhanced thespecific induction of p53-dependent genes in mammaliancells. Conversely, MAML1 knockdown reduced p53-dependent gene expression. Thus, these studies indicatethat MAML1 is essential for p53-dependent geneexpression.

Interestingly, MAML1 also increases the half-life ofthe p53 protein and enhances its phosphorylation/acetylation upon DNA damage in cells. In C. elegans,knockdown of the MAML homologue Lag-3 abrogatedthe p53-mediated germ-cell apoptotic response thattypically occurs in response to DNA damage. Further-more, the knock down of Lag-3 reduced the normalexpression patterns of downstream pro-apoptotic tar-gets of the C. elegans p53 homologue, Cep-1, includingCed-13 and Egl-1. These data clearly demonstrate a rolefor Lag-3 in p53-mediated apoptosis in C. elegans,although more work is required to decipher themechanisms involved. However, these studies haveset the stage for investigating the intriguing regulationof p53-mediated gene expression by MAML familymembers.

MAML1 co-activates b-catenin-based transcriptionalevents and is essential for colon carcinoma cell survivalMisregulation of the Wnt signaling pathway, a pathwaycritical for cellular proliferation and differentiation, isassociated with various disorders including colon


5143

Oncogene

carcinoma (Segditsas and Tomlinson, 2006). Thetight control of nuclear levels of b-catenin is a keyregulatory event for Wnt/b-catenin signaling. In theabsence of Wnt signaling, b-catenin remains sequesteredin the cytosol in a phosphorylated state, which leads toits degradation by a proteosome. The binding of theWnt ligand to Frizzled and LDL receptors on the cellsurface initiates multiple events, including the recruit-ment of Dishevelled and inactivation of glycogensynthase kinase 3b (GSK-3b). Inhibiting the phospho-rylation of b-catenin by GSK-3b leads to the transloca-tion of b-catenin to the nucleus, where it interactswith the T-cell factor (TCF) family of transcriptionfactors and activates target gene expression. AberrantWnt signaling can lead to constitutive transcriptionalactivation of b-catenin/TCF target genes, resulting inenhanced cellular proliferation in the absence ofappropriate cues.

MAML1 recently was determined to be a co-activatorfor b-catenin-mediated transcription (Alves-Guerraet al., 2007). In these studies, MAML1 interactedwith b-catenin in vitro and in vivo and dramaticallyincreased the transcriptional activity of b-catenin onpromoters containing TCF-binding sites. The othertwo MAML family members have similar co-activatingfunctions. Mechanistically, it appears that MAML1 isrecruited by b-catenin on Wnt target gene promoters(for example, cyclin D1 and c-Myc) even in thepresence of the Notch signaling inhibitor, GSI. Thesedata indicate that MAML1 functions in the Wnt/b-catenin pathway independently of Notch signaling.Strikingly, the knockdown of MAML proteins incolonic carcinoma cells, SW480, resulted in cell death,which correlated with decreased b-catenin-inducedexpression of cyclin D1 and c-Myc. Thus, MAMLproteins appear to be essential for SW480 colonic cellsurvival by transcriptional co-activation of b-catenin-mediated transcription.

Other potential Notch-independent functionsfrom Xenopus and Drosophila systemsEvidence that MAML proteins have Notch-independentactivities also has been established from studies usingXenopus and Drosophila systems. In Xenopus embryos,overexpression of XMam1 caused the formation of apigmented cell mass on their surface and inducedexpression of RNA-binding protein nrp-1. However,this effect was not observed when Notch signaling wasactivated ectopically through constitutive activation ofthe Notch receptor (Katada et al., 2006). This indicatesthat Xmam1 affects a neurogenic lineage in a Notch-independent pathway.

Recently, a genetic screen was performed using theExelixis collection of insertional mutations with a goalto identify Mastermind modifiers (Kankel et al., 2007).This screen took advantage of a Drosophila mutant(C96-MamN) that displays a fully penetrant wing-margin phenotype associated with the expression ofthe N-terminal Mastermind protein, MamN (Helmset al., 1999). The screen searched for mutations in the

Exelixis collection that dominantly modified thewing-margin phenotype of C96-MamN. The genesidentified with enhancer or suppressor functions weresubsequently subjected to a second screening to deter-mine Notch pathway-dependent or -independent func-tions. In the end, 175 known and 160 novel geneswere found to interact with Notch components anddesignated as Notch interactors, whereas 79 failed tointeract with the Notch pathway thus were definedas Mastermind-specific interactors (MSI). The MSIgenes were classified into several distinct functionalcategories, including: negative transcriptional regula-tors, factors with RNA polymerase II transcriptionalactivities, factors with small GTPase regulator activitiesand factors that negatively regulate metabolism.It is possible that these MSIs potentially interact withother Notch pathway components that were notincluded in this screen. However, this study verifiedthat the interaction of one MSI gene, the Drosophilab-catenin homologue, armadillo, interacts with Mamindependently of Notch in both the wing and eye. Thesedata confirm the finding that MAML proteinsfunction independently of Notch as a co-activator forb-catenin signaling (as discussed above). Therefore,multiple MSIs from this screen indicate a broad andprominent role for Mam in various developmentalprocesses.

Speculations on functional contributions: cross talkregulation by the MAML co-activators

Figure 2 summarizes the findings that we discussedabove, showing that MAML proteins participate inmultiple signaling pathways: Notch, MEF2C, p53, b-catenin and potentially others. These findings revealpivotal regulatory functions for MAML co-activators indiverse biological processes including cell proliferation,differentiation and survival, and highlight the centralroles for MAML co-activators in signaling cross talk.Many questions remain, however, including whether theMAML co-activators participate in multiple signalingpathways simultaneously or within certain signalingcontexts, and how these activities contribute to thefunctional biological outcomes.

From the biochemical aspects, some clues can beinferred. On the basis of immunoprecipitation studies,the MAML interactions with Notch/CSL appear to bemore stable compared to MAML interactions withMEF2C, p53 and b-catenin (L Wu et al., unpublished).Although preliminary, these data lead to the hypothesisthat MAML proteins mainly function as co-activatorsfor Notch/CSL when Notch receptors are activated, butare co-activators for other signaling pathways whenNotch is inactive. This hypothesis was supported by ourprevious findings that the Notch effects are dominantover MAML1’s myogenic activities with MEF2C(Shen et al., 2006). However, it is anticipated that theremay be more complicated scenarios than just simplefunctional switches. Future investigations into relative


5144

Oncogene

contributions of pathway-specific effects mediated bythe MAML co-activators and the potential interactionsof these pathways in various biological systems will shedlight on their functions in developmental and oncogeniccontexts.

MAML co-activators: links to human cancers

Though it is not a major focus for this review, it isworthwhile to point out that growing evidence impli-cates MAML co-activators in cancers, including muco-epidermoid carcinoma, leukemia and potentially cervi-cal cancer (Wu and Griffin, 2004). This is expectedconsidering the essential regulatory roles for MAMLproteins in Notch signaling as well as other signalingpathways, as discussed in this review.

Importantly, the MAML co-activators may beexcellent candidates as targets to modulate the growingnumber of signaling pathways in which they participate.Currently, we know more about the functions of theMAML proteins in the regulation of Notch signalingthan other signaling pathways. Inhibition of Notchsignaling has been achieved by interference with therecruitment of the MAML proteins to Notch/CSLcomplex (using either DNMAML1 or small molecularcompounds). Thus, more work is needed to supportpreliminary studies that show aberrant Notch signalingcan be controlled to inhibit cell growth and survival incancer cells with overactive Notch signaling (Wenget al., 2003; Liu et al., 2006). In addition, investigationsinto MAML functions in other pathways, includingWnt, p53 and MEF2C-based signaling, may be oftherapeutic importance in the future.

Concluding remarks

The family of MAML genes encodes transcriptionalco-activators required for Notch signaling. Recentdiscoveries of MAML involvement in the regulation ofother signaling pathways have indicated much broaderroles for the MAML co-activators in regulatingbiological functions, and also point toward a centralrole for MAML co-activators in mediating signalingcross talks. However, many questions remain to beaddressed regarding the mechanisms underlying the co-activator functions of MAML proteins in Notch andother relevant pathways. These issues include investiga-tions into the interactions of MAML-regulated signalingpathways and the individual and collective contributionsof the MAML proteins in distinct cell types and diseasedcontexts. Understanding these questions will provideguidance for manipulating signaling pathways criticalfor cancer development and therapeutic treatments.

Acknowledgements

This work was supported in part by NIH (R01 CA097148) andMuscular Dystrophy Association (MDA).

References

Alves-Guerra MC, Ronchini C, Capobianco AJ. (2007). Mastermind-

like 1 is a specific coactivator of beta-catenin transcription

activation and is essential for colon carcinoma cell survival. Cancer

Res 67: 8690–8698.

Anderson LM, Gibbons GH. (2007). Notch: a Mastermind of vascular

morphogenesis. J Clin Invest 117: 299–302.Artavanis-Tsakonas S, Rand MD, Lake RJ. (1999). Notch signaling: cell

fate control and signal integration in development. Science 284: 770–776.

Notch-dependent MAML functions

Notch-independent MAML functions

MAML

CSL

Co-A

ICN

MAML

MEF2C

Co-AMAML

p53

MAML

TCF

Co-A

β-catenin

Figure 2 Summary of the co-activator functions for Mastermind-like transcriptional co-activator (MAML) in both Notch-depen-dent and Notch-independent pathways. (a) MAMLs co-activateNotch-mediated transcription, as explained in Figure 1. (b)MAMLs also co-activate MEF2C-mediated transcription and arerequired for muscle development. MAML/MEF2C interactionmay be direct or indirect. (c) MAMLs also co-activate p53-mediated transcription in a Notch-independent manner. (d)MAMLs also were recently determined to co-activate b-catenin-mediated transcription in a Notch-independent manner. It ispossible that unidentified other co-activators (CoA, yellow) arerequired for these events to occur.


5145

Oncogene

Aster JC, Pear WS, Blacklow SC. (2008). Notch Signaling inLeukemia. Annu Rev Pathol 3: 587–613.

Bettler D, Pearson S, Yedvobnick B. (1996). The nuclear proteinencoded by the Drosophila neurogenic gene Mastermind is widelyexpressed and associates with specific chromosomal regions.Genetics 143: 859–875.

Bray SJ. (2006). Notch signalling: a simple pathway becomes complex.Nat Rev Mol Cell Biol 7: 678–689.

Brennan K, Gardner P. (2002). Notching up another pathway.Bioessays 24: 405–410.

Fortini ME, Artavanis-Tsakonas S. (1994). The suppressor of hairlessprotein participates in Notch receptor signaling. Cell 79: 273–282.

Fryer CJ, Lamar E, Turbachova I, Kintner C, Jones KA. (2002).Mastermind mediates chromatin-specific transcription and turnoverof the Notch enhancer complex. Genes Dev 16: 1397–1411.

Fryer CJ, White JB, Jones KA. (2004). Mastermind recruitsCycC:CDK8 to phosphorylate the Notch ICD and coordinateactivation with turnover. Mol Cell 16: 509–520.

Go MJ, Artavanis-Tsakonas S. (1998). A genetic screen for novelcomponents of the Notch signaling pathway during Drosophilabristle development. Genetics 150: 211–220.

Hall JM, McDonnell DP. (2005). Coregulators in nuclear estrogenreceptor action: from concept to therapeutic targeting. Mol Interv 5:343–357.

Han H, Tanigaki K, Yamamoto N, Kuroda K, Yoshimoto M,Nakahata T et al. (2002). Inducible gene knockout of transcriptionfactor recombination signal binding protein-J reveals its essentialrole in T versus B lineage decision. Int Immunol 14: 637–645.

Helms W, Lee H, Ammerman M, Parks AL, Muskavitch MA,Yedvobnick B. (1999). Engineered truncations in the DrosophilaMastermind protein disrupt Notch pathway function. Dev Biol 215:358–374.

High FA, Zhang M, Proweller A, Tu L, Parmacek MS, Pear WS et al.(2007). An essential role for Notch in neural crest duringcardiovascular development and smooth muscle differentiation.J Clin Invest 117: 353–363.

Hozumi K, Negishi N, Suzuki D, Abe N, Sotomaru Y, Tamaoki Net al. (2004). Delta-like 1 is necessary for the generation of marginalzone B cells but not T cells in vivo. Nat Immunol 5: 638–644.

Ilagan MX, Kopan R. (2007). SnapShot: Notch signaling pathway.Cell 128: 1246.

Jeffries S, Robbins DJ, Capobianco AJ. (2002). Characterization of ahigh-molecular-weight Notch complex in the nucleus of Notch(ic)-transformed RKE cells and in a human T-cell leukemia cell line.Mol Cell Biol 22: 3927–3941.

Kankel MW, Hurlbut GD, Upadhyay G, Yajnik V, Yedvobnick B,Artavanis-Tsakonas S. (2007). Investigating the genetic circuitry ofMastermind in Drosophila, a Notch signal effector. Genetics 177:2493–2505.

Katada T, Kinoshita T. (2003). XMam1, the Xenopus homologue ofMastermind, is essential to primary neurogenesis in Xenopus laevisembryos. Int J Dev Biol 47: 397–404.

Katada T, Ito M, Kojima Y, Miyatani S, Kinoshita T. (2006).XMam1, XenopusMastermind1, induces neural gene expression in aNotch-independent manner. Mech Dev 123: 851–859.

Kitagawa M, Oyama T, Kawashima T, Yedvobnick B, Kumar A,Matsuno K et al. (2001). A human protein with sequence similarityto Drosophila Mastermind coordinates the nuclear form of Notchand a CSL protein to build a transcriptional activator complex ontarget promoters. Mol Cell Biol 21: 4337–4346.

Kopan R, Nye JS, Weintraub H. (1994). The intracellular domain ofmouse Notch: a constitutively activated repressor of myogenesisdirected at the basic helix-loop-helix region of MyoD. Development

120: 2385–2396.Kovall RA. (2007). Structures of CSL, Notch and Mastermind

proteins: piecing together an active transcription complex. Curr

Opin Struct Biol 17: 117–127.Lai EC. (2002). Keeping a good pathway down: transcriptional

repression of Notch pathway target genes by CSL proteins. EMBO

Rep 3: 840–845.

Lasky JL, Wu H. (2005). Notch signaling, brain development, andhuman disease. Pediatr Res 57: 104R–109R.

Li B, Carey M, Workman JL. (2007). The role of chromatin duringtranscription. Cell 128: 707–719.

Lin SE, Oyama T, Nagase T, Harigaya K, Kitagawa M. (2002).Identification of new human Mastermind proteins defines a familythat consists of positive regulators for Notch signaling. J Biol Chem

277: 50612–50620.Lindsell CE, Shawber CJ, Boulter J, Weinmaster G. (1995). Jagged:

a mammalian ligand that activates Notch1. Cell 80: 909–917.Liu ZJ, Xiao M, Balint K, Smalley KS, Brafford P, Qiu R et al. (2006).

Notch1 signaling promotes primary melanoma progression byactivating mitogen-activated protein kinase/phosphatidylinositol3-kinase-Akt pathways and up-regulating N-cadherin expression.Cancer Res 66: 4182–4190.

Luo D, Renault VM, Rando TA. (2005). The regulation of Notchsignaling in muscle stem cell activation and postnatal myogenesis.Semin Cell Dev Biol 16: 612–622.

Maillard I, Weng AP, Carpenter AC, Rodriguez CG, Sai H, Xu Let al. (2004). Mastermind critically regulates Notch-mediatedlymphoid cell fate decisions. Blood 104: 1696–1702.

Nam Y, Sliz P, Song L, Aster JC, Blacklow SC. (2006). Structural basisfor cooperativity in recruitment of MAML coactivators to Notchtranscription complexes. Cell 124: 973–983.

Oyama T, Harigaya K, Muradil A, Hozumi K, Habu S, Oguro H et al.(2007). Mastermind-1 is required for Notch signal-dependent stepsin lymphocyte development in vivo. Proc Natl Acad Sci USA 104:9764–9769.

Petcherski AG, Kimble J. (2000a). LAG-3 is a putative transcriptionalactivator in the C. elegans Notch pathway. Nature 405: 364–368.

Petcherski AG, Kimble J. (2000b). Mastermind is a putative activatorfor Notch. Curr Biol 10: R471–R473.

Pires-daSilva A, Sommer RJ. (2003). The evolution of signallingpathways in animal development. Nat Rev Genet 4: 39–49.

Proweller A, Tu L, Lepore JJ, Cheng L, Lu MM, Seykora J et al.(2006). Impaired Notch signaling promotes de novo squamous cellcarcinoma formation. Cancer Res 66: 7438–7444.

Proweller A, Wright AC, Horng D, Cheng L, Lu MM, Lepore JJ et al.(2007). Notch signaling in vascular smooth muscle cells is requiredto pattern the cerebral vasculature. Proc Natl Acad Sci USA 104:16275–16280.

Radtke F, Wilson A, Stark G, Bauer M, van Meerwijk J,MacDonald HR et al. (1999). Deficient T cell fate specificationin mice with an induced inactivation of Notch1. Immunity 10:547–558.

Roy M, Pear WS, Aster JC. (2007). The multifaceted role of Notch incancer. Curr Opin Genet Dev 17: 52–59.

Saint Just Ribeiro M, Hansson ML, Wallberg AE. (2007). A prolinerepeat domain in the Notch co-activator MAML1 is important forthe p300-mediated acetylation of MAML1. Biochem J 404: 289–298.

Saito T, Chiba S, Ichikawa M, Kunisato A, Asai T, Shimizu K et al.(2003). Notch2 is preferentially expressed in mature B cells andindispensable for marginal zone B lineage development. Immunity

18: 675–685.Segditsas S, Tomlinson I. (2006). Colorectal cancer and genetic

alterations in the Wnt pathway. Oncogene 25: 7531–7537.Shen H, McElhinny AS, Cao Y, Gao P, Liu J, Bronson R et al. (2006).

The Notch coactivator, MAML1, functions as a novel coactivatorfor MEF2C-mediated transcription and is required for normalmyogenesis. Genes Dev 20: 675–688.

Smoller D, Friedel C, Schmid A, Bettler D, Lam L, Yedvobnick B.(1990). The Drosophila neurogenic locus Mastermind encodes anuclear protein unusually rich in amino acid homopolymers. Genes

Dev 4: 1688–1700.Tanigaki K, Han H, Yamamoto N, Tashiro K, Ikegawa M, Kuroda Ket al. (2002). Notch-RBP-J signaling is involved in cell fatedetermination of marginal zone B cells. Nat Immunol 3: 443–450.

Tu L, Fang TC, Artis D, Shestova O, Pross SE, Maillard I et al. (2005).Notch signaling is an important regulator of type 2 immunity. J Exp

Med 202: 1037–1042.


5146

Oncogene

Vousden KH, Lane DP. (2007). p53 in health and disease. Nat Rev

Mol Cell Biol 8: 275–283.Wallberg AE, Pedersen K, Lendahl U, Roeder RG. (2002). p300 and

PCAF act cooperatively to mediate transcriptional activation fromchromatin templates by Notch intracellular domains in vitro. Mol

Cell Biol 22: 7812–7819.Weng AP, Nam Y, Wolfe MS, Pear WS, Griffin JD, Blacklow SC et al.

(2003). Growth suppression of pre-T acute lymphoblastic leukemiacells by inhibition of Notch signaling. Mol Cell Biol 23: 655–664.

Wilson JJ, Kovall RA. (2006). Crystal structure of the CSL–Notch–Mastermind ternary complex bound to DNA. Cell 124: 985–996.

Wu L, Griffin JD. (2004). Modulation of Notch signaling byMastermind-like (MAML) transcriptional co-activators and theirinvolvement in tumorigenesis. Semin Cancer Biol 14: 348–356.

Wu L, Aster JC, Blacklow SC, Lake R, Artavanis-Tsakonas S, GriffinJD. (2000). MAML1, a human homologue of Drosophila Master-mind, is a transcriptional co-activator for NOTCH receptors.Nat Genet 26: 484–489.

Wu L, Kobayashi K, Sun T, Gao P, Liu J, Nakamura M et al. (2004).Cloning and functional characterization of the murine Mastermind-like 1 (Maml1) gene. Gene 328: 153–165.

Wu L, Maillard I, Nakamura M, Pear WS, Griffin JD. (2007). Thetranscriptional coactivator Maml1 is required for Notch2-mediatedmarginal zone B-cell development. Blood 110: 3618–3623.

Wu L, Sun T, Kobayashi K, Gao P, Griffin JD. (2002). Identificationof a family of Mastermind-like transcriptional coactivators formammalian Notch receptors. Mol Cell Biol 22: 7688–7700.

Xu T, Artavanis-Tsakonas S. (1990). Deltex, a locus interacting withthe neurogenic genes, Notch, Delta and Mastermind in Drosophilamelanogaster. Genetics 126: 665–677.

Xu T, Rebay I, Fleming RJ, Scottgale TN, Artavanis-Tsakonas S.(1990). The Notch locus and the genetic circuitry involved in earlyDrosophila neurogenesis. Genes Dev 4: 464–475.

Yedvobnick B, Smoller D, Young P, Mills D. (1988). Molecularanalysis of the neurogenic locus Mastermind of Drosophilamelanogaster. Genetics 118: 483–497.

Yu S, Reddy JK. (2007). Transcription coactivators for peroxisomeproliferator-activated receptors. Biochim Biophys Acta 1771:936–951.

Zhao Y, Katzman RB, Delmolino LM, Bhat I, Zhang Y, GurumurthyCB et al. (2007). The Notch regulator MAML1 interacts with p53and functions as a coactivator. J Biol Chem 282: 11969–11981.


5147

Oncogene

Documents

Mastermind-like transcriptional co-activator-mediated Notch signaling is indispensable for maintaining conjunctival epithelial identity