Notch Signaling in Mammary Development and Oncogenesis

C 2004) Journal of Mammary Gland Biology and Neoplasia, Vol. 9, No. 2, April 2004 ( Notch Signaling in Mammary Development and Oncogenesis Robert Callahan1 and Sean E. Egan2 With the discovery of an activated Notch oncogene as a causative agent in mouse mammary tumor virus induced breast cancer in mice, the potential role for Notch signaling in normal and pathological mammary development was revealed. Subsequently, Notch receptors have been found to regulate normal development in many organ systems. In addition, inappropriate Notch signaling has been implicated in cancer of several tissues in humans and animal model systems. Here we review important features of the Notch system, and how it may regulate development and cancer in the mammary gland. A large body of literature from studies in Drosophila and C. elegans has not only revealed molecular details of how the Notch proteins signal to control biology, but shown that Notch receptor activation helps to define how other signaling pathways are interpreted. In many ways the Notch system is used to define the context in which other pathways function to control proliferation, differentiation, cell survival, branching morphogenesis, asymmetric cell division, and angiogenesis—all processes which are critical for normal development and function of the mammary gland. KEY WORDS: Notch; mammary development; breast cancer; Int3. subsequently found in flies with mutations in other genes, now collectively known as the Neurogenic genes (5–7). It is this historical context that has led to a strong neurobiological focus of studies following from Poulson’s work. Interestingly, the molecular characterization of Neurogenic gene products has led to the identification of a “Notch signaling transduction pathway” by which many proteins function together to regulate development in the nervous system, and in other ectoderm derived, as well as mesoderm and endoderm derived, tissues in invertebrates and vertebrates (6). INTRODUCTION In 1919 Otto Mohr described a mutation in fruit flies that caused wing Notching in heterozygous females (1,2). Later D. F. Poulson carefully analyzed the phenotypic consequences of the responsible chromosomal deficiency, termed Notch8 (3,4). Interestingly, male Notch mutants, without a normal copy of the Xchromosome to supply the Notch-gene-encoded protein, had a number of dramatic developmental defects affecting many embryonic tissues, including derivatives of all three germ layers. The most dramatic developmental defect observed in Notch null mutants involves an almost complete transformation of surface ectoderm into cells of an expanded nervous system. Indeed, this so-called “Neurogenic” phenotype was NOTCH RECEPTOR PROTEINS In the mid-1980s the fly Notch gene was cloned and found to encode a large transmembrane receptor (8,9). At about the same time, research on the nematode worm C. elegans resulted in cloning of two Notch-related genes, lin-12 and glp-1, which control development of multiple tissues in this organism (10– 12). In each case, the Notch receptor was a single 1 Mammary Biology and Tumorigenesis Laboratory, National Cancer Institute, Bethesda, Maryland 20892; e-mail: rc54d@nih.gov. 2 Program in Developmental Biology, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, M5G 1 × 8, Canada and the Department of Molecular and Medical Genetics, The University of Toronto; e-mail: segan@sickkids.com. 145 C 2004 Plenum Publishing Corporation 1083-3021/04/0400-0145/0  146 pass transmembrane protein with multiple EGF-like repeats and cdc10/ankyrin repeats. Specifically, the Drosophila Notch protein contains 36 extracellular EGF-like repeats, followed by three cysteine-rich Lin12/Notch repeats (LNR), a transmembrane domain, a “RAM23” region, seven cdc10/ankyrin repeats (13), a nuclear localization sequence, and a Cterminal extension with PEST sequences (14). The worm proteins are highly related in structure, although somewhat smaller with 13 and 10 EGF-like repeats for Lin-12 and Glp-1, respectively. The four mammalian Notch proteins (Notch1, 2, 3, and 4) are closely related to the Drosophila protein (Fig. 1) (14). Notch1 and Notch2 contain 36 EGF-like repeats, whereas Notch3 has 34 EGF-like repeats and Notch4 has 29 EGF-like repeats. Notch receptors are proteolytically processed and glycosylated prior to expression on the cell surface in most systems. This first cleavage reaction occurs in the golgi complex, is mediated by Furin or a related preprotein convertase (15,16), and results in the expression of a noncovalently linked heterodimer on the surface of cells (17). NOTCH ACTIVATION SYSTEM: DELTA AND SERRATE/JAGGED LIGANDS Genetic analysis in flies also led to the identification of two related families of Notch ligands (Fig. 1) (18). The first described was a transmembrane protein, Delta. Delta functions to activate Notch receptors on an adjacent cell and to inhibit Notch activation in the same cell (6,19). The Delta extracellular domain contains a single cysteine-rich motif related to the EGF-like repeat, termed DSL on the basis of its identification in the Notch-ligands Delta, Serrate (see below), and Lag-2 (from C. elegans) (20). These domains have only been found in Notch ligands. The DSL domain of Delta is followed by nine EGFlike repeats and a transmembrane domain. Mammals have three Delta genes, Delta-like 1, 3, and 4 (or Dll1, Dll3, and Dll4). Each protein has a single DSL and multiple EGF-like repeats. Dll1 has eight EGF-like repeats, Dll3 has six, and Dll4 has eight. While not formally considered a Delta-family ligand by most authors, there is a fourth mammalian gene that encodes a protein with similarity to Delta ligands (21). This gene is know as Pref-1 or Delta-like (Dlk), and it encodes a protein with an N-terminal EGF/DSL-like domain followed by five EGF-like repeats, a transmembrane domain, and a cytoplasmic domain without obvious domain homology. Like the more widely studied Delta ligands listed above, Pref-1/Dlk can also Callahan and Egan block differentiation and can induce expression of Hes-family proteins (see below) (22). The second Notch ligand discovered in flies is termed Serrate. The mammalian homologues of Serrate are Jagged1 and Jagged2 (18,20). The Serrate/Jagged proteins are also single pass transmembrane proteins. The extracellular domain of each has a DSL domain, followed by 14–16 EGF-like repeats (Serrate has 14 and mammalian Jagged proteins have 16), and a von Willebrand factor type C domain (likely involved in oligomerization). Sequences at the extreme N and C termini of Delta and Serrate ligands are less conserved. N-terminal to the DSL domains in both Delta and Serrate-family ligands is a cysteine-containing sequence which likely controls Notch receptor-binding specificity (18). An artificial Delta–Serrate chimeric molecule, with the N terminus from Delta but DSL, EGF-like repeat region, von Willebrand factor type C domain, transmembrane domain, and C terminus from Serrate, behaves like Delta during fly wing development (see discussion of Fringe modification of ligand specificity below). The intracellular sequences of Delta and Serrate ligands are also somewhat related, although neither contains any recognizable domain structure. ALTERNATIVE FORMS OF NOTCH AND ITS LIGANDS: TRANSCRIPTION, SPLICING, AND GLYCOSYLATION Several Notch receptor and ligand genes in mammals are subject to alternative transcriptional initiation or splicing to generate protein isoforms with distinct domain organization. For example, Imatani and Callahan have described an isoform of Notch4 expressed in breast cancer cell lines, which is transcribed from an alternative promoter (see below) (23). The predicted gene product from this novel mRNA codes for an N-terminally truncated Notch4 product, consisting of most of the intracellular domain. The Dll3 ligand gene can generate multiple protein isoforms through alternative splicing (24). In this case, the alternative isoforms vary at their extreme C termini. The Jagged2 gene is subject to alternative splicing whereby EGF-like repeat 6 can be included, or not (hJAG2. del-E6), depending on the presence of a specific in-frame exon (see Genbank). The extent to which alternative promoter usage and alternative splicing are used to generate multiple Notch receptors and ligands is not clear. Indeed, biological functions for each Notch4, Dll3, and Jagged2 isoform have yet to be determined. Notch Signaling in Mammary Development and Oncogenesis Fig. 1. The Notch activation system. Fringe proteins modify DSL ligands and Notch receptors to control ligand-receptor specificity (top). Note that the positions of carbohydrate structures on Notch and DSL ligands are arbitrary in this figure. Mammalian DSL ligands and Notch receptors (bottom). 147 148 As with most large-cell surface proteins, the Notch receptors and ligands are subject to N-linked glycosylation (17,25). The regulation and function of this form of glycosylation in the Notch system has not been determined, although minimally Nlinked glycosylation is thought to be important for protein folding. In 1998, Haltiwanger and coworkers found that Notch1 is also modified through addition of O-linked Fucose and O-linked Glucose residues (26), each found as a monosaccharide modification of specific Serine or Threonine residues on Notch, or as part of a Serine/Threonine-FucoseGlcNAc-Galactose-Sialic acid tetrasaccharide or a Serine/Threonine-Glucose-Xylose-Xylose trisaccharide, respectively (27,28). These modifications are unusual, having been previously been described on only a limited set of proteins, including the EGF-like repeat containing coagulation factors. During the mid- to late 1990s, Irvine and coworkers characterized a gene in flies, termed Fringe, that functions to control whether Delta or Serrate activate Notch in specific regions of the developing wing disc (25,29). The Fringe protein had homology to bacterial glycosyltransferases, suggesting that it might function by regulating glycosylation of Notch or its ligands. These two observations were connected when several labs collaborated to show that Fringe proteins add a β1,3-linked N-acetyl-Glucosamine (GlcNAc) to the O-linked Fucose residues on Notch (30,31). It has subsequently been shown that Delta and Serrate ligands contain O-linked Fucose and that these can be further modified by Fringes through addition of β1,3-linked GlcNAc (32). In 1997, the mammalian Fringe-family was identified and found to consist of three related genes, termed Lunatic Fringe, Manic Fringe, and Radical Fringe (33,34). The Lunatic Fringe and Manic Fringe gene products, like Drosophila Fringe, are GlcNAc transferases that elongate O-linked Fucose residues on Notch receptors and ligands (30). A biochemical function for Radical Fringe has yet to be described. Genetic studies published to date have highlighted a role for Lunatic Fringe in regulation of Notch activation during somitogenesis (35,36). The biological role of Manic Fringe and Radical Fringe are less clear (37). One intriguing possibility is that the mammalian Fringe genes function redundantly in development. While Lunatic and Radical Fringe genes do not overlap in expression or function (37), it will be important to test for redundancy between Lunatic and Manic Fringe, as well as between Manic and Radical Fringe. Using tissue culture assays and a series of mutant CHO cell lines, Pam Stanley’s group Callahan and Egan has shown that the trisaccharide structure FucoseGlcNAc-Galactose must be generated through further elongation of the Fringe-generated disaccharides on Notch, in order for Fringe to effectively block activation of Notch1 by Jagged1 in vitro (38). Interestingly, Fringe proteins do not function in all Notch-dependent developmental systems, and therefore are not always required for Notch activation or inhibition. In contrast, the Fucose transferase enzyme protein O-fucosyltransferase 1 (O-FUT1 or Neurotic) is required for most, if not all, Notch functions in flies (39,40). Deletion of this gene in the mouse gives rise to Notch loss-of-function phenotypes and early embryonic lethality (41). In fact, O-FUT1mediated fucosylation of Notch is required to create high-affinity Delta- and Serrate-binding sites (42). The Fringe-mediated elongation of these O-Fucose residues then affects specificity of Notch for either ligand. For example, addition of both O-Fucose and GlcNAc to Notch EGF-like repeat 12 is required for Fringe to inhibit Serrate binding to Notch in vitro (43). ALTERNATIVE FORMS OF NOTCH AND ITS LIGANDS; PROTEOLYTIC PROCESSING AND β-HYDROXYLATION DURING SYNTHESIS Notch receptors are subject to proteolytic cleavage as they move through the secretory pathway (17). This cleavage occurs at cleavage site 1 (S1), between the LNR repeats and the transmembrane domain, and generates two polypeptides that remain associated through a noncovalent interaction requiring divalent cations (44). A furin-like convertase enzyme is likely responsible for this cleavage, at least in mammalian tissue culture cells (16). Interestingly, full-length uncleaved Notch can also be found at the cell surface, as can Notch proteins that have been cleaved within the EGF-like repeat containing domain and Notch proteins that have been cleaved to remove C-terminal sequences (45–48). Ligand-induced Notch signaling occurs in the absence of furin processing of Notch1 p300 (45). Bush et al. (45) were able to show that interaction of ligands with the uncleaved form of Notch1 can still block mouse C2C12 myoblast differentiation. The nature and function of alternative proteolytic processing of Notch is unclear. It is important to note that proteolytic removal of N-terminal EGF-like repeat containing sequences is sometimes associated with loss of Delta and Serrate/Jagged ligand binding sites (46). In contrast, removal of Notch Signaling in Mammary Development and Oncogenesis C-terminal sequences is associated with removal of disheveled-binding sites and expected to profoundly affect Notch signal transduction (see below) (48). Given the nature of, and variability of, Notch processing, this will remain an active area of research. Many EGF-repeats contain a consensus sequence for β-hydroxylation of aspartic acid and asparagine residues. Dinchuk et al. recently generated aspartyl β-hydroxylase (BAH) mutant mice. These mice had undetectable β-hydroxylation activity in liver preparations. Surprisingly, they displayed phenotypes similar to Jagged2 loss-of-function mutants (49). The authors went on to show that Jagged proteins are subject to β-hydroxylation in tissue culture cells, strongly suggesting that this modification plays a significant role in Notch activation, at least through alterations of Jagged2-Notch signaling. ALTERNATIVE NOTCH ACTIVATION SYSTEMS The extracellular domain of Notch receptors is large and conserved, suggesting that they may interact with a number of proteins beside Delta and Serrate/Jagged ligands and Fringes. In a screen to identify Drosophila gene products with affinity for the Notch extracellular domain, Cedric Wesley identified a number of extracellular and cell surface proteins that can bind to Notch (50). For example, he found that the poorly characterized Pecanex protein seemed to have high affinity for Notch. Perhaps most provocatively, he identified Wingless (Wg), a Drosophila Wnt protein, in this screen. Interestingly, Wg bound to an artificially truncated Notch protein lacking the first 18 EGF-like repeats, indicating that this protein interacts with Notch through a surface, or surfaces, that is distinct from those used to bind DSL ligands. Wg also bound to naturally expressed forms of Notch that were significantly smaller than the predicted fulllength Notch protein (46,50). This binding was dependent on the presence of extracellular calcium. More recent data suggest that the Wingless–Notch interaction may activate alternative forms of signal transduction and play an important role in regulating developmental processes that are dependent on Delta/Serrate ligands and Wnt proteins (see below) (51). A novel Notch ligand, F3/Contactin, was recently shown to activate Notch signaling in oligodendrocytes (52). This cell surface protein has six immunoglobulinlike domains followed by four fibronectin type-three domains and a GPI-link. Interestingly, F3/Contactin 149 activates the disheveled-signaling pathway, a pathway distinct from that activated by Delta and Serratefamily ligands (see below) (52). There are many other complexities in the Notch activation system, some of which have only recently come to light. For example, extracellular calcium concentrations control the affinity of Notch ligand/Notch receptor interactions (53). This phenomenon is exploited to restrict Notch activation during early embryonic development. In addition, the secreted Drosophila protein Scabrous is thought to control endocytic trafficking of Delta and/or Notch in cells that internalized this fibrinogen-related protein, and therefore it is thought to alter the threshold for Notch activation (54). Finally, the recently described Drosophila cell adhesion protein Echinoid is associated with cis-endocytosis of Delta and facilitates efficient activation of Notch by Delta (55). THE DELTA-/SERRATE-INDUCED ACTIVATION EVENT The activation of Notch receptors by Delta or Serrate ligands is a complex cell–cell communication event, which is only partially understood (56). To start with, a Delta or Serrate ligand binds to the extracellular domain of Notch. In some cases this can occur through interaction of ligand on the surface of Delta-induced filopodia with Notch on cells that are not even direct neighbors of the ligand-expressing cell (57). In response to DSL ligand binding to Notch, two related events occur: internalization of a ligand/Notch extracellular domain (ECD) complex into the ligand expressing cell, and cleavage of the Notch extracellular domain at cleavage site 2 (S2). It is not known whether internalization of Delta–Notch ECD or S2 cleavage occurs first, but both are required and presumably occur together. The highly related ADAM metalloproteases TACE and/or Kuzbanian are responsible for S2 cleavage. The S2 cleavage event targets a sequence just outside of the Notch transmembrane domain, and this leaves a membrane-spanning Notch protein fragment with a tiny stub projecting into the extracellular space. Fragments of this sort are targeted for further proteolytic cleavage by a presenilin containing multisubunit protease commonly known as γ -secretase. γ -Secretase cleaves Notch at cleavage site 3 (S3), near the C-terminal end of the transmembrane domain, thus releasing Notch ICD from its membrane tether (58). Notch ICD translocates into the nucleus, 150 where its RAM23 sequences bind to a transcriptional repressor complex through direct contact with the CSL DNA-binding protein (for CBF1, Suppressor of Hairless, or Lag-1). Notch ICD–CSL interaction disrupts the transcriptional repressor complex sitting on DNA, recruits transcriptional activating proteins, and turns on transcription of genes that were repressed prior to Notch activation (see below) (6,56). Interestingly, DSL ligands are also cleaved by Kuzbanian/TACE ADAM proteases (59) and ultimately are cleaved by γ -secretase to liberate ligand ICD fragments that have transcriptional activation properties (60–62). These data suggest that DSL ligand–Notch interaction may activate bidirectional signaling. Much of what has been learned about Notch activation and regulation of transcription has been stimulated by the discovery of activated alleles of Notch receptors in cancer, including the discovery of activated Notch4 in mouse mammary tumor virus- (MMTV) induced breast cancer (63–65). MMTV-/IAP-INDUCED NOTCH4/INT3 TUMORS The potential importance of aberrant Notch signaling in mammary tumor development was first recognized when Notch4 was found to be a common integration site (CIS) for MMTV in 18% of virus-induced mouse mammary tumors (9 out of 45 tumors) of the high-incidence CzechII strain of feral Mus musculus musculus. MMTV is not an acute transforming virus. Instead the virus, as a consequence of integration of its genome into the host genome, activates expression of adjacent cellular genes. In the case of Notch4, all (n = 9) of the viral insertions occur within exons 21 and 22 that span a 174-bp region within the gene. This region encodes a portion of the extracellular domain (ECD) adjacent to the transmembrane domain (TM). The “activated” RNA transcript is initiated within the MMTV long terminal repeat (LTR) and encodes a small portion of ECD, TM, and the complete ICD. The frequency with which Notch4 is activated by MMTV in mouse mammary tumors appears to be influenced by the genetic background of the host mouse strain. In CzechII mice the frequency of integration events in Notch4 was 18% (n = 45), whereas in the high-incidence BR6 inbred and M. m. jyg mouse strains the frequency was 7% (n = 30) and 43% (n = 23), respectively. A similar phenomenon has been noted for the MMTV CISs, Wnt1, and FGF3/FGF4 (reviewed in (66)). These studies strongly imply that alterations fixed in the genetic background Callahan and Egan of the host mouse strain affect the frequency with which particular MMTV integration sites will be selected. Investigators have taken advantage of this consideration to expand studies to MMTV-infected transgenic mouse strains (reviewed in (66)). In these studies the goal was to use the MMTV genome as a molecular tag to identify CIS-associated genes that complement the transgene during the evolution of malignant progression. This approach was taken with erbB2 transgenic mice infected with MMTV (67). In that setting, 2 out of 24 mouse mammary tumors contained MMTV-induced rearrangements of Notch1. In both tumors the viral genome integrated within sequences coding for the last Notch/lin-12 repeats and TM. In one case the viral genome was in the same transcriptional orientation as Notch1, while in the other tumor it was in the reverse orientation. As with MMTV-activated Notch4, the gene product was a protein composed of the Notch1 TM and ICD. MMTV integration within Notch4 is not the only mutagenic event that leads to the expression of “activated” Notch4 ICD. In addition, two studies have reported that intracisternal A-particle (IAP) transposable elements have integrated into Notch4 near MMTV CIS in a spontaneous Balb/c tumor and a CzechII mouse mammary tumor (68,69). In the Balb/c tumor, a fragment of an IAP genome (LTR and gag region) was linked to an extra copy of genomic sequences containing exons 23 and 24 that were inserted in the same transcription orientation as Notch4 in intron 24. Two alternatively spliced transcripts initiated from IAP LTR were detected. The longer of the two coded for the complete intracellular domain of Notch4 from a cryptic translation start signal in the IAP gag region. Translation of the shorter transcript began in exon 24 and therefore coded for a smaller ICD polypeptide that was missing the RAM23 region. In the CzechII mammary tumor, the IAP genome was integrated in the reverse transcriptional orientation from Notch4. In this case transcription of the Notch4 ICD was initiated from a cryptic transcription promoter in the reverse LTR. Translation of a Notch4 TM and ICD containing polypeptide was initiated from an in-frame methionine in the IAP reverse LTR. THE BIOLOGICAL CONSEQUENCES OF “ACTIVATED” NOTCH4 EXPRESSION The biological consequences of “activated” Notch4 expression on mammary gland development and tumorigenesis have been studied in vitro and in Notch Signaling in Mammary Development and Oncogenesis vivo. The HC11 mouse mammary epithelial cell line (70) is a clonal derivative of the COMMA D cell line that was derived from normal midpregnant BALB/c mammary epithelium (71). HC11 cells are not capable of anchorage-independent growth but have retained the capability to differentiate and express milk proteins in response to lactogenic hormones. Taking advantage of these properties, Robbins et al. (72) and Dievart et al. (67) showed that expression of Notch4 and Notch1 ICDs, respectively, in these cells conferred the capability for anchorage-independent growth in soft agar. In addition, expression of Notch4 ICD in these cells blocks their ability to express milk proteins in response to lactogenic hormones (Callahan, unpublished data). Thus, expression of Notch ICD blocks the ability of HC11 cells to differentiate and confers on them growth properties associated with malignant transformation. TAC-2 is another mouse mammary epithelial cell line that has been useful for studying the effect of Notch4 ICD signaling on normal epithelial architecture. These cells have the ability to form well-polarized histotypic structures when grown in collagen gels (73), and when treated with hepatocyte growth factor (HGF) they form branching tubules in culture. Uyttendaele et al. (74) have shown that Notch4 ICD expression inhibits tubule formation by TAC-2 cells. In a subsequent study, Soriano et al. (75) showed that Notch4 ICD signaling blocks the formation of glucocorticoid-induced alveolar-like structures in TAC-2 cells grown in collagen gels and causes loss of contact inhibition in TAC-2 cells grown on collagen-coated dishes. These tissue culture systems offer a means to molecularly dissect the effects of different Notch signaling pathway components on mammary epithelial biology, but do they reflect the consequences of Notch4 ICD expression in vivo? Two transgenic mouse strains have been developed which express the Notch4 ICD from either the MMTV LTR or the whey acidic protein (WAP) promoter (65,76–78). The common phenotype in these strains is that 100% of females are blocked in their ability to lactate, and all develop mammary tumors. In virgin females containing the MMTV LTR-Notch4 ICD transgene, there is minimal ductal development in the mammary gland. After the first pregnancy, ducts do fill the mammary fat pad, but there is little lobular development or cellular differentiation. In contrast, WAP-Notch4 ICD females exhibit normal ductal development, but during pregnancy secretory lobular development is severely impaired. Reciprocal transplantation studies of mammary epithelium from females of each of the transgenic strains 151 with normal FVB/N demonstrated that transgenic mammary epithelium was unable, either to grow in epithelium-divested FVB/N mammary fat pads from virgin mice (from MMTV LTR-Notch4 ICD), or to functionally differentiate in epithelium-divested FVB/N mammary fat pads from parous mice (from WAP-Notch4 ICD). In contrast, FVB/N mammary epithelium was able to grow and differentiate in epithelium-divested transgenic mammary fat pads. Coincident with limited lobular development of the WAP-Notch4 ICD mammary gland during pregnancy, dysplastic lesions appear throughout the gland that do not regress after weaning and progress to frank carcinoma. Immunohistochemical analysis of this tissue indicates that WAP-Notch4 ICD mammary tumors originate from secretory epithelial cell progenitors, whereas the MMTV LTR-Notch4 ICD tumors appear early and arise from ductal progenitor cells or from individual epithelial stem cells. COMPONENTS OF “ACTIVATED” NOTCH4 SIGNALING The Notch ICD is composed of three regions (reviewed in (79)). The RAM23 region, adjacent to TM, contains a nuclear localization signal sequence and is responsible for binding the CSL transcription repressor/factor. Adjacent to RAM23 is the CDC10/Ankryn region, which is composed of seven consecutive 32 amino acid repeat residues (13). The CDC10/Ankryn repeats region binds several proteins that modulate or regulate Notch4 signaling, and also a number of transcriptional coactivators including Mastermind/Maml (see below and reviewed in (79)). Notch1 and Notch2 have a third region that is C-terminal to the CDC10/Ankryn and is defined by its ability to transactivate transcription from GAL4 fusion constructs (80). MMTV integration in Notch4 represents a gain-of-function mutation, since it releases expression and activity of the Notch4 ICD from negative regulatory effects of the Notch4 ECD. All but one of the viral insertions occur 5′ to the sequence encoding a methionine at amino acid residue 1411 (81). The exception occurred 5′ of the sequence encoding a methionine at residue 1381. Although there are three more potential translation start signals 3′ to the Notch4 CIS, none have been used as translation start sites by MMTV integration in viral-induced mammary tumors. This suggests that aberrant expression of the entire ICD may be required for the rapid development of mammary tumors. 152 The complex series of interactions involving Notch ICD and other proteins involved in transmitting the Notch signal, or modifying the nature of this signal during mammary gland development and tumorigenesis, are not yet well characterized. In most developmental contexts studied to date, the major Notch signaling pathway is dependent on interaction of the CSL transcription repressor/factor with the RAM23 region of the Notch ICD. This interaction displaces corepressors SMRT and HDAC1, replacing them with coactivators (82). PRIMARY TARGETS OF CSL-DEPENDENT NOTCH SIGNALING ARE MEMBERS OF THE HES AND HERP GENE FAMILIES Hes (Hairy-Enhancer of Split) and Herp (Hesrelated repressor protein; a.k.a. Hey, Hesr, and Hrt) gene families encode basic helix-loop-helix (bHLH) proteins that are transcriptional repressors of gene expression, functioning downstream of Notch signaling (reviewed by (83,84)). Their expression patterns in different tissues and during development are not completely overlapping. Hes has been shown to act as a transcriptional repressor by three different mechanisms. One of these occurs through formation of a complex with mammalian TLE/Grg (mammalian homologues of Groucho) corepressor proteins. This complex binds to E-box and/or N-box elements in responsive promoters (85). Another is through formation of nonfunctional heterodimers with bHLH transcription activating factors, such as MyoD and H/Mash1. A third, less characterized mechanism described for Hes-1 requires its Orange or helix3–helix4 domain to repress transcription of its own promoter, as well as transcription of the p21waf promoter (83). Herp, in addition to passive repression of transcription by sequestration, can bind a heterologous set of corepressors, N-CoR/mSin3A/HDAC. For the six Hes-family proteins and three Herp proteins, only a limited number of target genes are known. For example, Hes1 represses its own expression, as well as expression of H/Mash, CD4, and acid α-glucosidase, whereas Herp1 is known to repress its own promoter. OTHER GENES DIRECTLY REGULATED BY NOTCH SIGNALING In addition to members of the Hes and Herp gene families, a number of genes directly implicated in can- Callahan and Egan cer are known to be upregulated as a direct consequence of CSL-dependent Notch signaling. One of these genes is erbB2. Chen et al. (86), while studying the transcriptional control of ErbB2, identified a palindrome binding protein that bound the ErbB2 promoter. When its sequence was determined, it was found to be identical to CBF-1/CSL. In 293 tissue culture cells, Notch1 ICD and CBF-1/CSL cooperated to activate transcription from a wild-type, but not mutant, erbB2 promoter. In experiments to study the mechanism by which activated Notch1 transforms HC11 mouse mammary epithelial cells, Dievant et al. were unable to detect upregulation of erbB2 gene expression (67), suggesting that regulation of erbB2 by Notch ICD may depend on cell-type-specific transcription factor expression or be subject to regulation by other signaling pathways. It will be interesting to compare erbB2 expression in tumors with MMTV inserted into Notch1, with erbB2 expression in tumors where Notch1 is not activated. Another gene whose expression is upregulated by CSL-dependent Notch signaling is cyclin D1. Ronchini and Capobianco (87) have previously shown that Notch1 ICD activates transcription of the cyclin D1 gene and CDK2 activity with rapid kinetics in an E1A-immortalized baby rat kidney cell line (RKE). This resulted in stimulation of cell cycle progression from G1 to S-phase. Notch1 ICD mutations that were incapable of activating cyclin D1 transcription failed to transform RKE cells, suggesting that Notch1 ICD transformation of RKE cells requires induction of cyclinD1 expression. Next, this group showed that CSL binds, in a DNA sequence specific manner, to the cyclin D1 promoter, consistent with the idea that Notch1 ICD activates cyclinD1 transcription through the CSL-dependent signaling pathway. Interestingly, Notch1 ICD expression did not significantly increase DNA synthesis in 0.1% serum, indicating that induction of cyclinD1 expression and activation of CDK2 in RKE cells are not sufficient to induce proliferation or cellular transformation, and that other factors are required for oncogenic transformation by Notch1 in this context. There are at least two points to keep in mind in evaluating the role of Notch signaling and cyclinD1 expression. First, at the present time it is not known which of the Notch gene(s) normally affect cyclin D1 expression in the mammary gland. Second is the context in which cyclin D1 expression is being evaluated. For instance, it is known that transgenic mice that overexpress cyclin D1 in the mammary epithelium develop mammary hyperplasias and subsequently mammary tumors (88). However, in the HC11 mammary Notch Signaling in Mammary Development and Oncogenesis epithelial cell line, which is frequently used to study the effects of oncogene expression on cell growth and differentiation, a different result is obtained. In these cells, overexpression of cyclin D1 causes inhibition of cell cycle progression and suppression of cell growth (89). In addition, overexpression of cyclin D1 in HC11 cells induces differentiation as defined by increased β-casein expression. A third gene upregulated by Notch1 signaling in some contexts is in fact Notch4, but not Notch2 or Notch3 (90). It is likely that this too is a consequence of CSL-dependent signaling, since the Notch4 promoter contains a CSL-responsive element. In this regard it would be of interest to know if the mammary tumors in which Notch1 is rearranged by MMTV, as well as the HC11 mammary epithelial cells expressing activated Notch ICD (67), also express the complete Notch4 receptor protein. We have found that HC11 cells express three Notch ligands, and that when transfected with a vector expressing the complete Notch4 receptor, these cells acquire the ability to grow in soft agar (Callahan, unpublished data). The role of Notch4 in Notch1 signaling during mammary gland development and tumorigenesis merits further study. Finally, a fourth gene which may function downstream of Notch signaling to control oncogenic behavior in some contexts is NFκ B2. This gene is strongly repressed by CSL in the absence of Notch activation (91). It will be important to test whether NFκ B2 is regulated by Notch in mammary cells. DISTINGUISHING BETWEEN CSL-DEPENDENT AND -INDEPENDENT SIGNALING Initial deletion analysis of activated Notch ICD suggested that signaling detected in the absence of RAM23 sequences reflected CSL-independent Notch signaling (80,92–96). For example, Jeffries and Capobianco (95) showed that RAM23 sequences in ICD were not necessary for transformation of RKE. Moreover, they could not detect a physical interaction between CSL and the RAM23-deleted variant ICD (δRAM23), nor could they detect CSL-dependent reporter gene induction in response to δRAM23 ICD expression. The general conclusion from these studies was that there was a CSL-independent component of Notch signaling that did not require the RAM23 domain. Recently, however, Jeffries et al. (97) have provided evidence that deletion of the RAM23 do- 153 main does not necessarily disrupt CSL-dependent Notch signaling. They showed that a complex of δRAM23 Notch1 ICD, Mastermind-Like-1 (Maml), and CSL can form, and that this complex can stimulate CSL/Notch-dependent transcription. They suggest that Maml acts as a tether between CSL and δRAM23. Further, they speculate that cell lines expressing Maml can support CSL-dependent transcriptional activation with δRAM23 Notch ICD proteins, whereas cell lines that express low levels of Maml cannot. In this regard, Dievart et al. (67) found that only vectors expressing the complete Notch1 ICD were capable of stimulating HC11 mouse mammary epithelial cells to grow in soft agar, suggesting that HC11 cells have low levels of Maml. Imatani and Callahan (23) identified a novel 1.8-kb Notch4/Int3 mRNA species (designated h-Int3sh) that is normally expressed in human testis and is aberrantly expressed in some tumor cell lines. Translation of the product begins at a methionine in the first CDC10/Ankryn repeat of ICD and terminates at the normal translation stop site. It thus represents a naturally occurring δRAM23 Notch4/Int3 ICD. Expression of Int3sh in the MCF10A human mammary epithelial cell line stimulates anchorage-independent growth. It will be interesting to see if MCF10A cells express high levels of Maml, and whether transformation of these cells by h-Int3sh is dependent on Maml expression. Recently we have developed three founder lines of transgenic mice expressing h-Int3sh under control of the WAP promoter (Raafat, Callahan, et al., manuscript in preparation). The phenotype of all three lines with regard to mammary gland development and tumorigenesis is the same. Unlike WAP-Int3 mice, where pregnancy-associated mammary gland development is blocked, lobulo/alveolar development and lactogenic differentiation is normal in WAPh-Int3sh mice. However, like WAP-Int3 females, WAP-Int3sh females develop mammary tumors (but at a lower frequency and with a longer latency). Thus, in our transgenic WAP-h-Int3sh model, mammary tumor development could still be a consequence of CSLdependent Notch signaling. At the present time we are investigating why mammary gland development occurs normally in WAP-h-Int3sh females as compared to WAP-Int3 females. This could reflect the existence of a previously unappreciated regulatory protein that interacts with the RAM23 portion of Notch4 ICD to inhibit mammary gland development, or could indicate that the level/extent of h-Int3sh-CSL interaction mediated by Maml is not sufficient to block mammary gland development and differentiation, but is 154 sufficient to contribute to mammary tumorigenesis. Consistent with the latter possibility, Lin et al. (98) have shown that Notch4/Int3 CSL-dependent signaling is relatively resistant to augmentation by Maml1, -2, and -3. Alternatively, Notch4/Int3-induced mammary tumorigenesis could be a consequence of CSLindependent Notch signaling. In vertebrates, additional evidence for noncanonical Notch signaling exists. For example, CSL-independent signaling is found in the developing avian neural crest, where a Delta-1-activated Notch signal is required to control Slug expression. In this case, regulation of Slug is not altered by expression of a dominant negative CSL protein in the neuroectoderm (99). In addition, Timmerman et al. have shown that Notch signaling can induce an epithelial to mesenchymal transition in a number of cellular contexts, and that this is mediated through upregulation of the Slug relative Snail (100). It will be important to test whether these latter two genes are targets of Notch in mammary epithelium. Interestingly, Slug and Snail expression are associated with invasion and metastasis in human breast cancer (101,102). Perhaps Notch-mediated regulation of Slug or Snail occurs in some of these tumors. Mediator(s) of the many speculated CSLindependent Notch signaling pathways remain largely unknown. However, there are several known transcription factors such as Mef2 (myocyte enhancer factor 2) and LEF1 (lymphocyte enhancer factor 1) that have been shown to interact with Notch ICD (103,104). Mef2 is a member of the MADS-box transcription factors (reviewed in (105)), which interact with the CDC10/Ankryn repeats of Notch1 ICD. This interaction blocks DNA binding by Mef2 and blocks its ability to cooperate with MyoD and myogenin to activate myogenic differentiation (104). LEF1 is a member of the high-mobility-group-box DNA binding protein family (103). This protein binds to the TAD region of Notch1 and Notch2 (but not Notch3), which is located just C-terminal to the CDC10/Ankryn in ICD. The Notch ICD acts as a coactivator of LEF-1 transcription factor activity in this context. Another set of candidates to mediate CSLindependent Notch signaling is the Deltex (mammalian DTX) proteins (106,107). The E47 transcription factor activates B-cell-specific immunoglobulin gene transcription and is required for early B-cell development (108,109). Ordentlich et al. (110) showed that overexpression of Deltex inhibits E47 transcription factor independent of CSL-dependent signaling and that this probably occurs through inhi- Callahan and Egan bition of Ras signaling. In another study, Yamamoto et al. (111) have shown that DTX-1 mediates a Notch signal to block differentiation of neural progenitor cells. They found that a significant fraction of Deltex1 is localized in the nucleus and physically interacts with the transcriptional coactivator p300. Deltex competitively inhibits the binding of the neural-specific helix-loop-helix-type transcription factor Mash1, thereby inhibiting its transcriptional activity and neural differentiation. It seems likely that additional transcription factors and cytoplasmic adaptor proteins will be identified as mediators of CSL-independent oncogenic Notch signaling. In addition, it should be anticipated that these factors will show specificity for individual Notch receptors. THE UBIQUITINATION PATHWAY AND NOTCH SIGNALING Several studies have demonstrated the importance of ubiquitination in the regulation of Notch signaling (reviewed by Lai (112)). The ubiquitination pathway is a critical and conserved pathway for regulation of many cellular processes including protein turnover, trafficking, and transcription. The basic pathway starts with the ubiquitin-activating enzyme (E1) which transfers ubiquitin, a 76 amino acid peptide, to a ubiquitin-conjugating enzyme (E2). An E3 ubiquitin ligase combines with an E2 protein to transfer ubiquitin to a target substrate, or to the end of a polyubiquitin chain. As a consequence of this pathway, soluble target proteins are marked with ubiquitin for degradation in the proteasome. In addition, ubiquitination of transmembrane proteins at the plasma membrane targets them for endocytic internalization, whereas ubiquitination of transmembrane proteins in the sorting endosome targets them for degradation in the lysosome. There are seven classes of E3 ubiquitin ligase enzymes that have been implicated in Notch signal transduction, or in regulation of the strength or duration of Notch signal transduction. These E3 proteins are Su(dx)/Itch, Deltex (DTX), Cbl, Neuralized, Mindbomb, Sel-10/Cdc-4 (a substrate recognition subunit in a multi-subunit E3 ligase), and Lnx. First, Su(dx)/Itch, a Nedd4 class E3 ligase with an Nterminal C2 domain, central WW domains, and a Cterminal Hect domain, binds to the Notch ICD and ubiquitinates it. Interestingly, Su(dx)/Itch can transfer ubiquitin to Notch proteins that are still attached to the plasma membrane, and may therefore control the levels of Notch available for activation at the cell Notch Signaling in Mammary Development and Oncogenesis surface (113). Deltex (Dtx) proteins are ring-finger class E3 ligases, capable of self-ubiquitination (114). In mammals there are four DTX proteins (DTX-1, -2, -2δE, and -3) that influence myogenesis, neurogenesis, and lymphogenesis. The N terminus of DXT-1, DXT-2, and DXT-2δE binds to the Ankryn/CDC10 of Notch ICD, whereas this domain is missing in DXT-3. The role of Deltex (DTX) E3 ligase activity is unclear, since the ring-finger motif is conserved in evolution but is not required for rescue of Deltex lossof-function phenotypes in flies. Cbl is an E3 ligase, most studied in the context of tyrosine kinase signaling. Jehn et al. recently reported that when Notch1 is phosphorylated on tyrosine residues in C2C12 cells, it binds to c-Cbl, and that the tyrosine-phosphorylated form of Notch1 accumulates when cells are treated with the lysosomal inhibitor chloroquine. Thus, Cbl appears to target membrane-bound Notch for destruction in the lysosome, at least in some cells where it is tyrosine-phosphorylated. Neuralized and Mindbomb are unrelated and conserved ring-finger class E3 ligases, which function to ubiquitinate the cytoplasmic domain of Delta-family ligands in order to stimulate their endocytic internalization (115). This internalization of Delta, together with Notch ECD, into signaling cells is required for activation of Notch receptors in neighboring signal-receiving cells (see above). It is not yet clear why both Neuralized and Mindbomb have been conserved to perform the same or similar functions during Notch activation. Sel10/Cdc-4 is an F-box WD40 domain protein in a large multi-subunit E3 ligase known as SCF. Sel-10 binds to the PEST domain of a specific phosphorylated form of Notch ICD found in the nucleus, perhaps in complex with the negative feedback regulator Nrarp (116). Once bound, nuclear phospho-Notch ICD is ubiquitinated and targeted to the proteosome for degradation. The Sel-10 E3 ligase therefore functions to limit Notch-dependent transcriptional regulation after nuclear translocation of Notch ICD (117–119). Interestingly, the Sel-10 gene has recently been deleted from the mouse germline, and this causes a dramatic elevation of Notch4 ICD without increasing expression of Notch1, 2, or 3 ICDs (120). Perhaps these ICDs are ubiquitinated by the related E3 ligase Skp2, or by Sel-10 and Skp2, in vivo (121). In 2001, Steve Reed’s group found that Sel-10 also targets the cyclin E protein for ubiquitination, and therefore for proteolytic degradation (122). In addition, this group found that Sel-10 is a tumor-suppressor gene, mutated in the SUM149PT breast cancer cell line, and in a large number of endometrial carcinomas (122,123). Finally, LNX (ligand of Numb-protein X) is involved in regu- 155 lating Notch signaling during asymmetric cell division (reviewed in (112)). Numb is asymmetrically localized in sibling cells whose fate is decided by Notch signaling (reviewed in (124)). McGill and McGlade (125) have shown that Numb interacts with the WW domain of Su(dx)/Itch, and suggested that Numb acts as an adaptor protein that recruits the Itch E3 ligase and the ubiquitination machinery for the ubiquitination of the Notch ICD complexed with Numb. LNX is an E3 ligase that targets Numb for ubiquitination. Nie et al. (126) suggest that asymmetric distribution of LNX would establish an asymmetric distribution of Numb. It has been proposed that LNX augments Notch signaling by lowering the level of Numb in the daughter cell destined to respond to Notch signaling. It will be important to determine what role these many E3 ligases play in regulating Notch signaling in the mammary gland and breast cancer, and for what biological outcome. Some recent data on mammalian Neuralized speak to this issue (see below). CROSS-TALK BETWEEN THE NOTCH PATHWAY AND OTHER SIGNALING PATHWAYS Many examples of cross-talk between the Notch pathway and other signaling pathways have been documented. For instance, cooperative and antagonistic effects between Receptor Tyrosine Kinase/Ras pathways and the Notch pathway have been reported in Drosophila (127) and C. elegans (128–130). In the Drosophila eye, erbB and Notch signaling are used together to control development of cone cells. In this case, activation of erbB signaling in the neuronal photoreceptor cells is required for expression of Delta ligand in these cells (131). Delta then functions together with erbB signaling to activate cone cell differentiation in neighboring cells (132). An antagonistic interaction between erbB and Notch signaling has been seen during C. elegans vulval development (128). In this case, erbB signaling and Notch signaling are known to interact at multiple levels (133). First, strong activation of erbB signaling in the P6.p vulval precursor cell leads to expression of three Notch ligands (including the secreted Notch ligand, Dsl-1), and to endocytosis and destruction of the worm Notch receptor, Lin-12, in this cell. Once P6.p expresses Notch ligands, these ligands activate Lin-12 in neighboring P5.p and P7.p cells. Lin-12/Notch signaling in P5.p and P7.p activates expression of several redundant inhibitors of erbB signaling in these cells, including 156 Mapk kinase phosphatase, Ack-1 kinase, and a number of proteins implicated in endocytic downregulation of erbB receptors (129,130,133–135). The complex interplay of Notch signaling and erbB signaling in the mammary gland and breast cancer remains to be investigated. However, as discussed above, Notch signaling can upregulate erbB2 expression (86), and activation of Notch1 cooperates with activation of erbB2 to induce breast cancer in mice (67). Notch signaling is also used together with FGF receptor signaling to regulate development in many contexts. In flies, Notch signaling is used to suppress MAP kinase activation downstream of FGF receptor signaling in the developing and branching trachea (136). This interaction between Notch and FGF receptor signaling is found in mammalian limb patterning and tooth development, indeed in many contexts (137,138). In vitro, an antisense oligonucleotide against Jagged1 was found to be a potent enhancer of FGF-induced angiogenesis (139). This result led to the finding that Notch signaling even inhibits FGF-induced transformation of fibroblasts, confirming the antagonistic relationship between Notch and FGF signaling (140). Fitzgerald et al. (78) have attempted to identify which signaling pathways collaborate with Notch4 ICD in mammary tumor development. Using cell lines derived from transgenic MMTV LTR-Notch4 ICD, they showed that inhibition of Erk/MAP kinase and PI-3 kinase pathways, downstream of Ras, blocked anchorage-independent growth of cells in soft agar. Inhibitors of other signaling pathways including Srclike kinases (Lck and Fyn) and protein kinases A and C had no effect in this assay. So, in this setting, there is a synergistic relationship between Notch and Ras signaling pathways in Notch4 ICD initiated mammary tumors. In another experimental system, Weijzen et al. (90) show that oncogenic Ras activates Notch signaling and that wild-type Notch1 is necessary to maintain the neoplastic phenotype in Rastransformed human cells, in vitro and in vivo. In these Ras-transformed cells, expression of Notch1, Delta-1, and presenilin-1 are upregulated, and Ras influences Notch-1 signaling through a p38-mediated pathway. Interactions between Notch and TGFβ signaling pathways have also been reported. For example, Blokzijl et al. (141) have described a synergistic interaction between Notch1 and TGFβ1 to upregulate Hes-1 in myogenic cells. This effect was blocked by expression of a dominant-negative form of CSL. The mechanism underlying the integration of TGFβ1 and Notch signaling is likely through protein–protein interaction between SMAD3, the intracellular trans- Callahan and Egan ducer of TGFβ1 signals, and Notch1 ICD. SMAD4 is required for the functional interaction between the Notch1 and TGFβ1 pathways. In addition, a SMAD3 point mutation that is not able to bind DNA is still capable of potentiating the Notch1 ICD/TGFβ1 signal at the Hes-1 promoter. In reciprocal experiments, using the PAI-1 promoter that is specific for SMAD3 and highly responsive to TGFβ1, Notch1 ICD could further enhance transcription activity of the promoter in a concentration-dependent manner. Whether this functional relationship between Notch1 and SMAD3 can be generalized to other members of the Notch family and to other cellular contexts remains to be determined. As discussed briefly above, the fly Wnt protein Wingless can bind to Notch extracellular domain sequences (50). The function of this interaction was initially very puzzling. More recent studies in flies and mammals have revealed that Wnt and Notch signaling pathways interact at multiple levels, since these two pathways frequently function together to regulate development (51,142). Indeed, Wingless not only binds to Notch, but also activates a cytoplasmic signaling pathway that functions to inhibit DSLligand activated Notch signaling. Wingless activates the cytoplasmic Dishevelled protein, which in turn binds to the Notch cytoplasmic domain to inhibit signaling through Deltex. Wingless signaling also inhibits glycogen-synthase kinase-3 (GSK-3) activity, which is required to stabilize Notch ICD. Uyttendaele et al. (74,143) have presented evidence for negative cross-talk between the Wnt-1 and Notch4 signaling pathways in the TAC-2 mammary epithelial cell line suspended in collagen gels. In this setting Wnt1, like HGF and TGF-β2, induces branching morphogenesis by TAC-2 cells. TAC-2 cells expressing Notch4 ICD did not respond to HGF or to TGF-β2, but underwent branching morphogenesis in the presence of Wnt-1, suggesting that Wnt1 signaling is dominant to Notch4 ICD signaling in TAC-2 cells in the branching morphogenesis assay. Deletion analysis of Notch4 ICD demonstrated that the RAM23 and Ankryn/CDC10 repeats are required for inhibition of ductal morphogenesis. NOTCH IN NORMAL AND PATHOLOGICAL MAMMARY DEVELOPMENT: MANY QUESTIONS What is the role of Notch signaling in development of the mammary gland? This is a difficult Notch Signaling in Mammary Development and Oncogenesis question to address with our current state of knowledge. From what is known about the function of Notch signaling in flies (6) and worms (144), we could imagine a role for Notch signaling in epithelial/mesenchymal transitions, such as those in mammary bud formation (145). Notch signaling might regulate mammary epithelial growth and branching during puberty (146). Notch signaling could be involved in cell-fate specification, perhaps controlling the rationing of cell types within the mammary gland (6). Perhaps Notch signaling is involved in inductive interactions in the mammary gland (6,25,131). For example, Notch signaling could control induction of mammary bud formation or expansion. Maybe Notch signaling controls lactogenic differentiation (6,147–149), cell proliferation (150,151), stem cell self-renewal (152), or even apoptosis/involution (153). The problem here is that there are so many documented biological functions for Notch signaling in both invertebrate and vertebrate systems that we cannot even begin to guess what its most important roles will be in the mammary gland in the absence of some basic information, such as Notch receptor, ligand, and Fringe gene expression during development. In addition, the gain-of-function experiments such as those performed with activated Notch ICD proteins in tissue culture cells and in transgenic mice must be complemented with Notch gene loss-of-function experiments to determine what Notch signaling does in the normal mammary gland. In any case, we have learned some significant lessons already that can hint at Notch functions in this context (see above). For example, in TAC-2 mammary epithelial cells in vitro, activated Notch4 suppressed ductal growth and branching, as well as formation of glucocorticoid-induced alveolar-like structures (74,75). In addition, expression of Notch4 ICD blocked lactogenic differentiation (milk gene expression) in HC11 cell cultures (Callahan, unpublished data). In vivo, expression of Notch4 ICD inhibited ductal elongation and branching, as well as alveolar development and differentiation in MMTVNotch4 ICD transgenic mice. Alveolar development, cellular polarity, and lactogenic differentiation were severely impaired in WAP-Notch4 ICD transgenic mice. Multiple Notch receptors and ligands are expressed in HC11 mammary epithelial cells in vitro (see above), and multiple Notch receptors, ligands, and Fringes are expressed during mouse mammary gland development in vivo (Keli Xu and Egan, unpublished data). Interestingly, the expression pattern of Notch receptors, ligands, and Fringes changes dra- 157 matically with each stage of mammary development, suggesting that Notch signaling may indeed be used for distinct functions during sequential stages. Preliminary loss-of-function experiments on a number of Notch receptors, ligands, and Fringes support this view (Keli Xu and Egan, unpublished data). In 2001, Vollrath et al. reported a mammary phenotype in mice with a targeted mutation in a mammalian homologue of Drosophila Neuralized. Unfortunately, this phenotype has not been characterized in detail, but appears to represent a defect in alveolar development since the mutant gland is still composed primarily of adipocytes during lactation. In addition, lactogenic differentiation of alveolar cells may be somewhat impaired. Since Neuralized is required for Notch activation, this result strongly suggests that Notch activation may play an important role during pregnancy. Interestingly, Notch signaling controls differentiation of adipocytes (154) and development/remodeling of the vascular system (155), two critical stromal elements (156–158) regulating development in the mammary gland. Perhaps Notch regulates epithelial and stromal elements in this context. NOTCH IN HUMAN BREAST CANCER In many ways, it is also very difficult to address the role, or potential role(s), played by Notch signaling in human breast cancer with the current state of knowledge. As a starting point, it will be important to determine which Notch receptors, ligands, and Fringes are expressed in breast cancer, and whether specific expression profiles correlate with specific pathological or clinical features. To this end, such expression analysis has been performed by in situ hybridization on a collection of 25 human breast tumors (Michael Reedijk and Egan, manuscript in preparation). These data have revealed that all four Notch receptors, four Notch ligands, and one of three Fringes are expressed at varying frequencies within this collection of tumors. For example, Notch3 is expressed in approximately half of the tumors and highly expressed in three tumors. Interestingly, Notch3 is very highly expressed in neovessels that have been recruited to many of the tumors, suggesting that this receptor may play an important role in breast tumor angiogenesis (Fig. 2). One of the biggest challenges to assessing the role of Notch signaling in human breast cancer is the heterogeneity of this disease. Clearly a large collection of tumors must be studied to establish any correlation between the Notch system expression profile and tumor pathology 158 Callahan and Egan Fig. 2. Notch-3 expression in human breast cancer samples. In situ hybridization using Notch-3 antisense (A & C) or sense (B & D) RNA probes. Images are shown at 20× magnification using either brightfield or darkfield (insets) microscopy. Tumor 13 (C) demonstrates a fivefold greater level of Notch-3 expression than Tumor 11 (A) as assessed by quantifying silver grains. Solid arrows identify Notch-3-expressing VSMC. or outcomes. Despite these challenges, our preliminary data indicate that Notch receptors, ligands, and Fringes are expressed in human breast cancers and supporting stromal elements, including tumor vessels. Studies on Notch1 and Notch4 ICD in mammary epithelial cell transformation have been extremely informative and have guided our understanding of oncogenic Notch signaling. It will also be important to keep in mind the recent finding that Notch genes can function to suppress tumor growth (159). Perhaps the lack of Notch receptor, ligand, or Fringe expression will be important in some human breast cancers. Clearly, there are more experiments to do. ACKNOWLEDGMENTS We apologize to those whose work has not been directly cited because of space limitations. Sean Egan thanks Drs Michael Reedijk and Keli Xu for permission to cite unpublished results. Robert Callahan thanks Dr Ahmed Raafat and Sharon Bargo for permission to cite unpublished results. S.E.E. is supported by grants from the Terry Fox Foundation, The Canadian Institutes for Health Research, and the U.S. Army/Department of Defense. REFERENCES 1. Mohr OL. Character changes caused by mutation of an entire region of a chromosome in Drosophila. Genetics 1919;4:274282. 2. Mohr OL. A genetic and cytological analysis of a section deficiency involving four units of the X-chromosome in Drosophila melanogaster. Z. f. indukt. Abst. Vererbl. 1923;32:108-232. 3. Poulson DF. Chromosomal deficiencies and embryonic development of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 1937;23:133-137. 4. Poulson DF. The effects of certain X-chromosome deficiencies on the embryonic development of Drosophila melanogaster. The Journal of Experimental Zoology 1940;83:271-325. 5. Jan YN, Jan LY. Neuronal specification. Current Opinion in Genetics and Development 1992;2:608-613. 6. Artavanis-Tsakonas S, Rand MD, Lake RJ. Notch signaling: cell fate control and signal integration in development. Science 1999;284:770-776. Notch Signaling in Mammary Development and Oncogenesis 7. Artavanis-Tsakonas S, Delidakis C, Fehon RG. The Notch locus and the cell biology of neuroblast segregation. Annu. Rev. Cell Biol. 1991;7:427-452. 8. Wharton KA, Johansen KM, Xu T, Artavanis-Tsakonas S. Nucleotide sequence from the neurogenic locus Notch implies a gene product that shares homology with proteins containing EGF-repeats. Cell 1985;43:567-581. 9. Kidd S, Kelley MR, Young MW. Sequence of the Notch locus of Drosophila: relationship of the encoded protein to mammalian clotting and growth factors. Mol. Cell. Biol. 1986;6:3094-3108. 10. Yochem J, Greenwald I. glp-1 and lin-12, Genes implicated in distinct cell-cell interactions in C. elegans, encode similar transmembrane proteins. Cell 1989;58:553-563. 11. Yochem J, Weston K, Greenwald I. The Caenorhabditis elegans lin-12 gene encodes a transmembrane protein with overall similarity to Drosophila Notch. Nature 1988;335:547550. 12. Greenwald I. lin-12, a nematode homeotic gene, is homologous to a set of mammalian proteins that includes epidermal growth factor. Cell 1985;43(3 Pt 2):583-90. 13. Lubman OY, Korolev SV, Kopan R. Anchoring notch genetics and biochemistry; structural analysis of the ankyrin domain sheds light on existing data. Mol Cell 2004;13(5):619-26. 14. Egan SE, St-Pierre B, Leow C-C. Notch Receptors, Partners and Regulators: From Conserved Domains to Powerful Functions. Current Topics in Microbiology and Immunology 1997;228:273-324. 15. Thomas G. Furin at the cutting edge: from protein traffic to embryogenesis and disease. Nat Rev Mol Cell Biol 2002;3(10):753-66. 16. Logeat F, Bessia C, Brou C, Lebail O, Jarriault S, Seidah NG, et al. The Notch1 receptor is cleaved constitutively by a furinlike convertase. PNAS 1998;95:8108-8112. 17. Blaumueller CM, Qi H, Zagouras P, Artavanis-Tsakonas S. Intracellular cleavage of Notch leads to a heterodimeric receptor on the plasma membrane. Cell 1997;90(2):281-91. 18. Fleming RJ. Structural conservation of Notch receptors and ligands. Semin Cell Dev Biol 1998;9:599-607. 19. Jacobsen TL, Brennan K, Arias AM, Muskavitch MA. Cisinteractions between Delta and Notch modulate neurogenic signalling in Drosophila. Development 1998;125:4531-4540. 20. Lissemore JL, Starmer WT. Phylogenetic analysis of vertebrate and invertebrate Delta/Serrate/LAG-2 (DSL) proteins. Mol Phylogenet Evol 1999;11(2):308-19. 21. Smas CM, Sul HS. Pref-1, a protein containing EGF-like repeats, inhibits adipocyte differentiation. Cell 1993;73:725-734. 22. Kaneta M, Osawa M, Sudo K, Nakauchi H, Farr AG, Takahama Y. A role for pref-1 and HES-1 in thymocyte development. J Immunol 2000;164(1):256-64. 23. Imatani A, Callahan R. Identification of a novel NOTCH4/INT-3 RNA species encoding an activated gene product in certain human tumor cell lines. Oncogene 2000;19(2): 223-31. 24. Kusumi K, Sun ES, Kerrebrock AW, Bronson RT, Chi D-C, Bulotsky MS, et al. The mouse pudgy mutation disrupts Delta homologue Dll3 and initiation of early somite boundaries. Nature Genetics 1998;19:274-278. 25. Haines N, Irvine KD. Glycosylation regulates Notch signalling. Nat Rev Mol Cell Biol 2003;4(10):786-97. 26. Moloney DJ, Shair LH, Lu FM, Xia J, Locke R, Matta KL, et al. Mammalian Notch1 Is Modified with Two Un- 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 159 usual Forms of O-Linked Glycosylation Found on Epidermal Growth Factor-like Modules. The Journal of Biological Chemistry 2000;275:9604-9611. Haltiwanger RS. Regulation of signal transduction pathways in development by glycosylation. Curr Opin Struct Biol 2002;12(5):593-8. Haltiwanger RS, Stanley P. Modulation of receptor signaling by glycosylation: fringe is an O-fucose-beta1,3N-acetylglucosaminyltransferase. Biochim Biophys Acta 2002;1573(3):328-35. Irvine KD, Wieschaus E. Fringe, a boundary-specific molecule, mediates interactions between dorsal and ventral cells during Drosophila wing development. Cell 1994;79:595-606. Moloney DJ, Panin VM, Johnston SH, Chen J, Shao L, Wilson R, et al. Fringe is a glycosyltransferase that modifies Notch. Nature 2000;406:369-375. Bruckner K, Perez L, Clausen H, Cohen S. Glycosyltransferase activity of Fringe modulates Notch-Delta interactions. Nature 2000;406:411-415. Panin VM, Shao L, Lei L, Moloney DJ, Irvine KD, Haltiwanger RS. Notch ligands are substrates for protein O-fucosyltransferase-1 and Fringe. J Biol Chem 2002;277(33):29945-52. Cohen B, Bashirullah A, Dagnino L, Campbell C, Fisher B, Leow CC, et al. Fringe boundaries coincide with Notchdependent patterning centers in mammals and alter Notchdependent development in Drosophila. Nature Genetics 1997;16:283-288. Johnston SH, Rauskolb C, Wilson R, Prabhakaran B, Irvine KD, Vogt TF. A Family of Mammalian Fringe Genes Implicated in Boundary Determination and the Notch Pathway. Development 1997;124:2245-2254. Zhang N, Gridley T. Defects in somite formation in Lunatic fringe-deficient mice. Nature 1998;394:374-377. Evrard YA, Lun Y, Aulehla A, Gan L, Johnson RL. Lunatic fringe is an essential mediator of somite segmentation and patterning. Nature 1998;394:377-381. Zhang N, Norton CR, Gridley T. Segmentation defects of Notch pathway mutants and absence of a synergistic phenotype in lunatic fringe/radical fringe double mutant mice. Genesis 2002;33(1):21-28. Chen J, Moloney DJ, Stanley P. Fringe modulation of Jagged1-induced Notch signaling requires the action of beta 4galactosyltransferase-1. Proc Natl Acad Sci U S A 2001;98(24):13716-21. Okajima T, Irvine KD. Regulation of notch signaling by olinked fucose. Cell 2002;111(6):893-904. Sasamura T, Sasaki N, Miyashita F, Nakao S, Ishikawa HO, Ito M, et al. neurotic, a novel maternal neurogenic gene, encodes an O-fucosyltransferase that is essential for Notch-Delta interactions. Development 2003;130(20):4785-95. Shi S, Stanley P. Protein O-fucosyltransferase 1 is an essential component of Notch signaling pathways. Proc Natl Acad Sci U S A 2003;100(9):5234-9. Okajima T, Xu A, Irvine KD. Modulation of notch-ligand binding by protein O-fucosyltransferase 1 and fringe. J Biol Chem 2003;278(43):42340-5. Lei L, Xu A, Panin VM, Irvine KD. An O-fucose site in the ligand binding domain inhibits Notch activation. Development 2003;130(26):6411-21. Rand MD, Grimm LM, Artavanis-Tsakonas S, Patriub V, Blacklow SC, Sklar J, et al. Calcium Depletion Dissociates 160 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. Callahan and Egan and Activates Heterodimeric Notch Receptors. Molecular and Cellular Biology 2000;20:1825-1835. Bush G, diSibio G, Miyamoto A, Denault JB, Leduc R, Weinmaster G. Ligand-induced signaling in the absence of furin processing of Notch1. Dev Biol 2001;229(2):494-502. Wesley CS, Saez L. Notch responds differently to Delta and Wingless in cultured Drosophila cells. J Biol Chem 2000;275(13):9099-101. Wesley CS, Mok LP. Regulation of Notch signaling by a novel mechanism involving suppressor of hairless stability and carboxyl terminus-truncated notch. Mol Cell Biol 2003;23(16):5581-93. Wesley CS, Saez L. Analysis of Notch Lacking the Carboxyl Terminus Identified in Drosophila Embryos. The Journal of Cell Biology 2000;149:683-696. Dinchuk JE, Focht RJ, Kelley JA, Henderson NL, Zolotarjova NI, Wynn R, et al. Absence of post-translational aspartyl betahydroxylation of epidermal growth factor domains in mice leads to developmental defects and an increased incidence of intestinal neoplasia. J Biol Chem 2002;277(15):12970-7. Wesley CS. Notch and wingless regulate expression of cuticle patterning genes. Mol Cell Biol 1999;19(8):5743-58. Brennan K, Gardner P. Notching up another pathway. Bioessays 2002;24(5):405-10. Hu QD, Ang BT, Karsak M, Hu WP, Cui XY, Duka T, et al. F3/contactin acts as a functional ligand for Notch during oligodendrocyte maturation. Cell 2003;115(2):163-75. Raya A, Kawakami Y, Rodriguez-Esteban C, Ibanes M, Rasskin-Gutman D, Rodriguez-Leon J, et al. Notch activity acts as a sensor for extracellular calcium during vertebrate left-right determination. Nature 2004;427(6970):121-8. Li Y, Fetchko M, Lai ZC, Baker NE. Scabrous and Gp150 are endosomal proteins that regulate Notch activity. Development 2003;130(13):2819-27. Rawlins EL, Lovegrove B, Jarman AP. Echinoid facilitates Notch pathway signalling during Drosophila neurogenesis through functional interaction with Delta. Development 2003;130(26):6475-84. Kopan R. Notch: a membrane-bound transcription factor. J Cell Sci 2002;115(Pt 6):1095-7. De Joussineau C, Soule J, Martin M, Anguille C, Montcourrier P, Alexandre D. Delta-promoted filopodia mediate long-range lateral inhibition in Drosophila. Nature 2003;426(6966):555-9. Struhl G, Adachi A. Requirements for presenilin-dependent cleavage of notch and other transmembrane proteins. Mol Cell 2000;6(3):625-36. Qi H, Rand MD, Wu X, Sestan N, Wang W, Rakic P, et al. Processing of the Notch Ligand Delta by the Metalloprotease Kuzbanian. Science 1999;283:91-94. Ikeuchi T, Sisodia SS. The Notch ligands, Delta1 and Jagged2, are substrates for presenilin-dependent ”gamma-secretase” cleavage. J Biol Chem 2003;278(10):7751-4. Six E, Ndiaye D, Laabi Y, Brou C, Gupta-Rossi N, Israel A, et al. The Notch ligand Delta1 is sequentially cleaved by an ADAM protease and gamma-secretase. Proc Natl Acad Sci U S A 2003;100(13):7638-43. LaVoie MJ, Selkoe DJ. The Notch ligands, Jagged and Delta, are sequentially processed by alpha-secretase and presenilin/gamma-secretase and release signaling fragments. J Biol Chem 2003;278(36):34427-37. Ellisen LW, Bird J, West DC, Soreng AL, Reynolds TC, Smith SD, et al. TAN-1, the human homologue of the Drosophila 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. Notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 1991;66:649-661. Gallahan D, Kozak C, Callahan R. A new common integration region (int-3) for mouse mammary tumor virus on mouse chromosome 17. J Virol 1987;61(1):218-220. Jhappan C, Gallahan D, Stahle C, Chu E, Smith GH, Merlino GT, et al. Expression of an Activated Notch-Related int-3 Transgene Interferes with Cell Differentiation and Induces Neoplastic Transformation in Mammary and Salivary Glands. Genes Dev 1992;6(3):345-355. Callahan R, Smith GH. MMTV-induced mammary tumorigenesis: gene discovery, progression to malignancy and cellular pathways [see comments]. Oncogene 2000;19(8):992-1001. Dievart A, Beaulieu N, Jolicoeur P. Involvement of Notch1 in the development of mouse mammary tumors. Oncogene 1999;18(44):5973-81. Kordon EC, Smith GH, Callahan R, Gallahan D. A novel non-mouse mammary tumor virus activation of the Int3 gene in a spontaneous mouse mammary tumor. J Virol 1995;69(12):8066-9. Lee JS, Haruna T, Ishimoto A, Honjo T, Yanagawa S. Intracisternal type A particle-mediated activation of the Notch4/int3 gene in a mouse mammary tumor: generation of truncated Notch4/int3 mRNAs by retroviral splicing events. J Virol 1999;73(6):5166-71. Ball RK, Friis RR, Schoenenberger CA, Doppler W, Groner B. Prolactin regulation of beta-casein gene expression and of a cytosolic 120-kd protein in a cloned mouse mammary epithelial cell line. Embo J 1988;7(7):2089-95. Danielson KG, Knepper JE, Kittrell FS, Butel JS, Medina D, Durban EM. Clonal populations of the mouse mammary cell line, COMMA-D, which retain capability of morphogenesis in vivo. In Vitro Cell Dev Biol 1989;25(6):535-43. Robbins J, Blondel BJ, Gallahan D, Callahan R. Mouse mammary tumor gene Int-3: a member of the Notch gene family transforms mammary epithelial cells. J. Virol. 1992;66:25942599. Soriano JV, Pepper MS, Nakamura T, Orci L, Montesano R. Hepatocyte growth factor stimulates extensive development of branching duct-like structures by cloned mammary gland epithelial cells. J Cell Sci 1995;108 ( Pt 2):413-30. Uyttendaele H, Soriano JV, Montesano R, Kitajewski J. Notch4 and Wnt-1 proteins function to regulate branching morphogenesis of mammary epithelial cells in an opposing fashion. Dev Biol 1998;196(2):204-17. Soriano JV, Uyttendaele H, Kitajewski J, Montesano R. Expression of an activated Notch4(int-3) oncoprotein disrupts morphogenesis and induces an invasive phenotype in mammary epithelial cells in vitro. Int J Cancer 2000;86(5):652-9. Smith GH, Gallahan D, Diella F, Jhappan C, Merlino G, Callahan R. Constitutive expression of a truncated INT3 gene in mouse mammary epithelium impairs differentiation and functional development. Cell, Growth & Differentiation 1995;6(5):563–577. Gallahan D, Jhappan C, Robinson G, Hennighausen L, Sharp R, Kordon E, et al. Expression of a truncated Int3 gene in developing secretory mammary epithelium specifically retards lobular differentiation resulting in tumorigenesis. Cancer Res 1996;56(8):1775-85. Fitzgerald K, Harrington A, Leder P. Ras pathway signals are required for notch-mediated oncogenesis. Oncogene 2000;19(37):4191-8. Notch Signaling in Mammary Development and Oncogenesis 79. Callahan R, Raafat A. Notch signaling in mammary gland tumorigenesis. J Mammary Gland Biol Neoplasia 2001;6(1):2336. 80. Kurooka H, Kuroda K, Honjo T. Roles of the ankyrin repeats and C-terminal region of the mouse notch1 intracellular region. Nucleic Acids Res 1998;26(23):5448-55. 81. Gallahan D, Callahan R. The mouse mammary tumor associated gene INT3 is a unique member of the NOTCH gene family (NOTCH4). Oncogene 1997;14:1883-1890. 82. Nam Y, Weng AP, Aster JC, Blacklow SC. Structural requirements for assembly of the CSL.intracellular Notch1.Mastermind-like 1 transcriptional activation complex. J Biol Chem 2003;278(23):21232-9. 83. Davis RL, Turner DL. Vertebrate hairy and Enhancer of split related proteins: transcriptional repressors regulating cellular differentiation and embryonic patterning. Oncogene 2001;20(58):8342-57. 84. Iso T, Kedes L, Hamamori Y. HES and HERP families: multiple effectors of the Notch signaling pathway. J Cell Physiol 2003;194(3):237-55. 85. Jennings BH, Tyler DM, Bray SJ. Target specificities of Drosophila enhancer of split basic helix-loop- helix proteins. Mol Cell Biol 1999;19(7):4600-10. 86. Chen Y, Fischer WH, Gill GN. Regulation of the ERBB2 promoter by RBPJkappa and NOTCH. J Biol Chem 1997;272(22):14110-4. 87. Ronchini C, Capobianco AJ. Induction of cyclin D1 transcription and CDK2 activity by Notch(ic): implication for cell cycle disruption in transformation by Notch(ic). Mol Cell Biol 2001;21(17):5925-34. 88. Wang TC, Cardiff RD, Zukerberg L, Lees E, Arnold A, Schmidt EV. Mammary hyperplasia and carcinoma in MMTVcyclin D1 transgenic mice. Nature 1994;369(6482):669-71. 89. Han EK, Begemann M, Sgambato A, Soh JW, Doki Y, Xing WQ, et al. Increased expression of cyclin D1 in a murine mammary epithelial cell line induces p27kip1, inhibits growth, and enhances apoptosis. Cell Growth Differ 1996;7(6):699-710. 90. Weijzen S, Rizzo P, Braid M, Vaishnav R, Jonkheer SM, Zlobin A, et al. Activation of Notch-1 signaling maintains the neoplastic phenotype in human Ras-transformed cells. Nat Med 2002;8(9):979-86. 91. Oswald F, Liptay S, Adler G, Schmid RM. NF-kappaB2 is a putative target gene of activated Notch-1 via RBP- Jkappa. Mol Cell Biol 1998;18(4):2077-88. 92. Nofziger D, Miyamoto A, Lyons KM, Weinmaster G. Notch signaling imposes two distinct blocks in the differentiation of C2C12 myoblasts. Development 1999;126(8):1689-702. 93. Aster JC, Robertson ES, Hasserjian RP, Turner JR, Kieff E, Sklar J. Oncogenic forms of NOTCH1 lacking either the primary binding site for RBP-Jkappa or nuclear localization sequences retain the ability to associate with RBPJkappa and activate transcription. J Biol Chem 1997;272(17): 11336-43. 94. Dumont E, Fuchs KP, Bommer G, Christoph B, Kremmer E, Kempkes B. Neoplastic transformation by Notch is independent of transcriptional activation by RBP-J signalling. Oncogene 2000;19(4):556-61. 95. Jeffries S, Capobianco AJ. Neoplastic transformation by Notch requires nuclear localization. Mol Cell Biol 2000;20(11):3928-41. 96. Shawber C, Nofziger D, Hsieh JJ, Lindsell C, Bogler O, Hayward D, et al. Notch signaling inhibits muscle cell differen- 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 161 tiation through a CBF1-independent pathway. Development 1996;122(12):3765-73. Jeffries S, Robbins DJ, Capobianco AJ. Characterization of a high-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 2002;22(11):3927-41. Lin SE, Oyama T, Nagase T, Harigaya K, Kitagawa M. Identification of new human mastermind proteins defines a family that consists of positive regulators for notch signaling. J Biol Chem 2002;277(52):50612-20. Endo Y, Osumi N, Wakamatsu Y. Bimodal functions of Notch-mediated signaling are involved in neural crest formation during avian ectoderm development. Development 2002;129(4):863-73. Timmerman LA, Grego-Bessa J, Raya A, Bertran E, PerezPomares JM, Diez J, et al. Notch promotes epithelialmesenchymal transition during cardiac development and oncogenic transformation. Genes Dev 2004;18(1):99-115. Blanco MJ, Moreno-Bueno G, Sarrio D, Locascio A, Cano A, Palacios J, et al. Correlation of Snail expression with histological grade and lymph node status in breast carcinomas. Oncogene 2002;21(20):3241-6. Hajra KM, Chen DY, Fearon ER. The SLUG zinc-finger protein represses E-cadherin in breast cancer. Cancer Res 2002;62(6):1613-8. Ross DA, Kadesch T. The notch intracellular domain can function as a coactivator for LEF-1. Mol Cell Biol 2001;21(22):7537-44. Wilson-Rawls J, Molkentin JD, Black BL, Olson EN. Activated notch inhibits myogenic activity of the MADS-Box transcription factor myocyte enhancer factor 2C. Mol Cell Biol 1999;19(4):2853-62. Messenguy F, Dubois E. Role of MADS box proteins and their cofactors in combinatorial control of gene expression and cell development. Gene 2003;316(Oct 16):1-21. Diederich RJ, Matsuno K, Hing H, Artavanis-Tsakonas S. Cytosolic interaction between deltex and Notch ankyrin repeats implicates deltex in the Notch signaling pathway. Development 1994;120(3):473-81. Matsuno K, Diederich RJ, Go MJ, Blaumueller CM, Artavanis-Tsakonas S. Deltex acts as a positive regulator of Notch signaling through interactions with the Notch ankyrin repeats. Development 1995;121(8):2633-44. Zhuang Y, Soriano P, Weintraub H. The helix-loop-helix gene E2A is required for B cell formation. Cell 1994;79(5):875-84. Bain G, Maandag EC, Izon DJ, Amsen D, Kruisbeek AM, Weintraub BC, et al. E2A proteins are required for proper B cell development and initiation of immunoglobulin gene rearrangements. Cell 1994;79(5):885-92. Ordentlich P, Lin A, Shen CP, Blaumueller C, Matsuno K, Artavanis-Tsakonas S, et al. Notch inhibition of E47 supports the existence of a novel signaling pathway. Mol Cell Biol 1998;18(4):2230-9. Yamamoto N, Yamamoto S, Inagaki F, Kawaichi M, Fukamizu A, Kishi N, et al. Role of Deltex-1 as a transcriptional regulator downstream of the Notch receptor. J Biol Chem 2001;276(48):45031-40. Lai EC. Protein degradation: four E3s for the notch pathway. Curr Biol 2002;12(2):R74-8. Qiu L, Joazeiro C, Fang N, Wang HY, Elly C, Altman Y, et al. Recognition and ubiquitination of Notch by Itch, a hect-type E3 ubiquitin ligase. J Biol Chem 2000;275(46):35734-7. 162 114. Takeyama K, Aguiar RC, Gu L, He C, Freeman GJ, Kutok JL, et al. The BAL-binding protein BBAP and related Deltex family members exhibit ubiquitin-protein isopeptide ligase activity. J Biol Chem 2003;278(24):21930-7. 115. Le Borgne R, Schweisguth F. Notch signaling: endocytosis makes delta signal better. Curr Biol 2003;13(7):R273-5. 116. Lamar E, Deblandre G, Wettstein D, Gawantka V, Pollet N, Niehrs C, et al. Nrarp is a novel intracellular component of the Notch signaling pathway. Genes Dev 2001;15(15):1885-99. 117. Oberg C, Li J, Pauley A, Wolf E, Gurney M, Lendahl U. The Notch intracellular domain is ubiquitinated and negatively regulated by the mammalian Sel-10 homolog. J Biol Chem 2001;276(38):35847-53. 118. Gupta-Rossi N, Le Bail O, Gonen H, Brou C, Logeat F, Six E, et al. Functional interaction between SEL-10, an F-box protein, and the nuclear form of activated Notch1 receptor. J Biol Chem 2001;276(37):34371-8. 119. Wu G, Lyapina S, Das I, Li J, Gurney M, Pauley A, et al. SEL-10 is an inhibitor of notch signaling that targets notch for ubiquitin-mediated protein degradation. Mol Cell Biol 2001;21(21):7403-15. 120. Tsunematsu R, Nakayama K, Oike Y, Nishiyama M, Ishida N, Hatakeyama S, et al. Mouse Fbw7/Sel-10/Cdc4 is required for notch degradation during vascular development. J Biol Chem 2003. 121. Nakayama K, Nagahama H, Minamishima YA, Matsumoto M, Nakamichi I, Kitagawa K, et al. Targeted disruption of Skp2 results in accumulation of cyclin E and p27(Kip1), polyploidy and centrosome overduplication. Embo J 2000;19(9):2069-81. 122. Strohmaier H, Spruck CH, Kaiser P, Won KA, Sangfelt O, Reed SI. Human F-box protein hCdc4 targets cyclin E for proteolysis and is mutated in a breast cancer cell line. Nature 2001;413(6853):316-22. 123. Spruck CH, Strohmaier H, Sangfelt O, Muller HM, Hubalek M, Muller-Holzner E, et al. hCDC4 gene mutations in endometrial cancer. Cancer Res 2002;62(16):4535-9. 124. Zhong W. Diversifying neural cells through order of birth and asymmetry of division. Neuron 2003;37(1):11-4. 125. McGill MA, McGlade CJ. Mammalian numb proteins promote Notch1 receptor ubiquitination and degradation of the Notch1 intracellular domain. J Biol Chem 2003;278(25):23196203. 126. Nie J, McGill MA, Dermer M, Dho SE, Wolting CD, McGlade CJ. LNX functions as a RING type E3 ubiquitin ligase that targets the cell fate determinant Numb for ubiquitin-dependent degradation. Embo J 2002;21(1-2):93-102. 127. Carmena A, Buff E, Halfon MS, Gisselbrecht S, Jimenez F, Baylies MK, et al. Reciprocal regulatory interactions between the Notch and Ras signaling pathways in the Drosophila embryonic mesoderm. Dev Biol 2002;244(2):226-42. 128. Wang M, Sternberg PW. Pattern formation during C. elegans vulval induction. Curr Top Dev Biol 2001;51:189-220. 129. Yoo AS, Bais C, Greenwald I. Crosstalk between the EGFR and LIN-12/Notch pathways in C. elegans vulval development. Science 2004;303(5658):663-6. 130. Shaye DD, Greenwald I. Endocytosis-mediated downregulation of LIN-12/Notch upon Ras activation in Caenorhabditis elegans. Nature 2002;420(6916):686-90. 131. Tsuda L, Nagaraj R, Zipursky SL, Banerjee U. An EGFR/Ebi/Sno pathway promotes delta expression by inactivating Su(H)/SMRTER repression during inductive notch signaling. Cell 2002;110(5):625-37. Callahan and Egan 132. Flores GV, Duan H, Yan H, Nagaraj R, Fu W, Zou Y, et al. Combinatorial signaling in the specification of unique cell fates. Cell 2000;103(1):75-85. 133. Sternberg PW. Developmental biology. A pattern of precision. Science 2004;303(5658):637-8. 134. Chen N, Greenwald I. The lateral signal for LIN-12/Notch in C. elegans vulval development comprises redundant secreted and transmembrane DSL proteins. Dev Cell 2004;6(2): 183-92. 135. Berset T, Hoier EF, Battu G, Canevascini S, Hajnal A. Notch inhibition of RAS signaling through MAP kinase phosphatase LIP-1 during C. elegans vulval development. Science 2001;291(5506):1055-8. 136. Ikeya T, Hayashi S. Interplay of Notch and FGF signaling restricts cell fate and MAPK activation in the Drosophila trachea. Development 1999;126(20):4455-63. 137. Sidow A, Bulotsky MS, Kerrebrock AW, Bronson RT, Daly MJ, Reeve MP, et al. Serrate2 is disrupted in the mouse limb-development mutant syndactylism. Nature 1997;389(6652):722-5. 138. Harada H, Kettunen P, Jung HS, Mustonen T, Wang YA, Thesleff I. Localization of putative stem cells in dental epithelium and their association with Notch and FGF signaling. J Cell Biol 1999;147(1):105-20. 139. Zimrin AB, Pepper MS, McMahon GA, Nguyen F, Montesano R, Maciag T. An antisense oligonucleotide to the notch ligand jagged enhances fibroblast growth factor-induced angiogenesis in vitro. J Biol Chem 1996;271(51):32499-502. 140. Small D, Kovalenko D, Soldi R, Mandinova A, Kolev V, Trifonova R, et al. Notch activation suppresses fibroblast growth factor-dependent cellular transformation. J Biol Chem 2003;278(18):16405-13. 141. Blokzijl A, Dahlqvist C, Reissmann E, Falk A, Moliner A, Lendahl U, et al. Cross-talk between the Notch and TGFbeta signaling pathways mediated by interaction of the Notch intracellular domain with Smad3. J Cell Biol 2003;163(4): 723-8. 142. Aulehla A, Wehrle C, Brand-Saberi B, Kemler R, Gossler A, Kanzler B, et al. Wnt3a plays a major role in the segmentation clock controlling somitogenesis. Dev Cell 2003;4(3): 395-406. 143. Uyttendaele H, Ho J, Rossant J, Kitajewski J. Vascular patterning defects associated with expression of activated Notch4 in embryonic endothelium. Proc Natl Acad Sci U S A 2001;98(10):5643-8. 144. Greenwald I. LIN-12/Notch signaling: lessons from worms and flies. Genes & Development 1998;12:1751-1762. 145. Hartenstein AY, Rugendorff A, Tepass U, Hartenstein V. The function of the neurogenic genes during epithelial development in the Drosophila embryo. Development 1992;116:12031220. 146. Llimargas M. The Notch pathway helps to pattern the tips of the Drosophila tracheal branches by selecting cell fates. Development 1999;126(11):2355-64. 147. Li Y, Baker NE. Proneural enhancement by Notch overcomes Suppressor-of-Hairless repressor function in the developing Drosophila eye. Curr Biol 2001;11(5):330-8. 148. Anant S, Roy S, VijayRaghavan K. Twist and Notch negatively regulate adult muscle differentiation in Drosophila. Development 1998;125(8):1361-9. 149. Lopez-Schier H, St Johnston D. Delta signaling from the germ line controls the proliferation and differentiation of the Notch Signaling in Mammary Development and Oncogenesis 150. 151. 152. 153. somatic follicle cells during Drosophila oogenesis. Genes Dev 2001;15(11):1393-405. Berry LW, Westlund B, Schedl T. Germ-line tumor formation caused by activation of glp-1, a Caenorhabditis elegans member of the Notch family of receptors. Development 1997;124(4):925-36. Giraldez AJ, Cohen SM. Wingless and Notch signaling provide cell survival cues and control cell proliferation during wing development. Development 2003;130(26):6533-43. Okabe M, Imai T, Kurusu M, Hiromi Y, Okano H. Translational repression determines a neuronal potential in Drosophila asymmetric cell division. Nature 2001;411 (6833):94-8. Miller DT, Cagan RL. Local induction of patterning and programmed cell death in the developing Drosophila retina. Development 1998;125(12):2327-35. 163 154. Garces C, Ruiz-Hidalgo MJ, de Mora JF, Park C, Miele L, Goldstein J, et al. Notch-1 controls the expression of fatty acid-activated transcription factors and is required for adipogenesis. J Biol Chem 1997;272(47):29729-34. 155. Gridley T. Notch signaling during vascular development. Proc Natl Acad Sci U S A 2001;98(10):5377-8. 156. Wiseman BS, Werb Z. Stromal effects on mammary gland development and breast cancer. Science 2002;296(5570):10469. 157. Bahary N, Zon LI. Development. Endothelium–chicken soup for the endoderm. Science 2001;294(5542):530-1. 158. Djonov V, Andres AC, Ziemiecki A. Vascular remodelling during the normal and malignant life cycle of the mammary gland. Microsc Res Tech 2001;52(2):182-9. 159. Radtke F, Raj K. The role of Notch in tumorigenesis: oncogene or tumour suppressor? Nat Rev Cancer 2003;3(10):756-67.