![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Molecular and Cellular Biology, May 2005, p. 3982-3996, Vol. 25, No. 10
0270-7306/05/$08.00+0 doi:10.1128/MCB.25.10.3982-3996.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Division of Molecular Oncology, Institute for Cancer Research and Treatment (IRCC), University of Turin School of Medicine, Str. Prov. 142, Km. 3,95, 10060 Candiolo, Torino, Italy
Received 15 July 2004/ Returned for modification 8 October 2004/ Accepted 11 February 2005
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Spatial and chronological orchestration of the various steps of invasive growth is optimally accomplished by a family of growth factors known as scatter factors, whose prototypic member is hepatocyte growth factor (HGF), together with their tyrosine kinase receptors, in particular, the HGF receptor encoded by the Met protooncogene. HGF induces various morphogenetic responses in epithelial cells deriving from different tissues when these are grown in three-dimensional gels, including formation of branching tubules in kidney, breast, and prostate epithelial cells (for a review, see reference 43), and it is a potent angiogenic factor that promotes remodeling of capillary networks (5). In vivo, HGF stimulates tubulogenesis in the liver and kidneys during organ regeneration and promotes differentiation of the ductal tree in the mammary glands during early pregnancy (24, 29, 36, 52, 65).
Although the role of HGF in the promotion of invasive growth is well documented and the intracellular signals triggered by Met activation and leading to execution of HGF-dependent responses have been extensively investigated (4, 18, 41, 42, 44, 45), fundamental questions remain unanswered. First, which are the nuclear factors that regulate Met transcription? This issue is particularly intriguing in light of the frequently observed neo- or overexpression of Met in human solid tumors (10) and of the recent finding that Met is transcriptionally induced by hypoxic stimuli (39). Second, which are, in turn, the transcriptional targets of Met? In spite of the huge amount of data gathered on the various signaling cascades that are primed upon Met stimulation, very little is known about how these signals affect the nuclear transcription machinery to induce or repress specific gene products. Answers to such questions, together with the functional annotation of the molecules identified, are particularly important to categorize negative regulators of HGF-dependent invasive growth, which can be exploited to hamper cancer progression and metastasis.
In this paper, we took advantage of an in silico analysis of the Met promoter region to identify the transmembrane receptor Notch (and, specifically, its downstream effector HES-1) as a transcriptional repressor of Met, and thus as a physiological inhibitor of Met-dependent invasive growth, both in vitro and in vivo. Intriguingly, in an opposite but complementary fashion, Met activation results in stimulation of the Notch pathway through transcriptional induction of the Notch ligands Delta1 and Delta4, together with HES-1. Taken together, these results unravel an in-built mechanism of negative feedback regulation in which Met activation leads to transcriptional induction of Notch function, which in turn limits HGF activity through repression of Met-triggered signals. Such a regulatory circuit can be exploited to prevent HGF-driven cancer progression and metastasis.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell maintenance and retroviral infection. MDA-MB-435-ß4 cells were grown in Dulbecco's modified Eagle medium (DMEM; Sigma) supplemented with 10% fetal calf serum (FCS; Sigma) and 1 mg/ml G418 (Gibco); MDCK cells were grown in DMEM (Sigma) supplemented with 10% FCS (HyClone). Viral hybrid vectors were produced by transient transfection of 293T cells with an empty pBabe vector or with an N4-TM-pBabe construct, both containing a puromycin resistance cassette, a gag pol packaging construct derived from Moloney murine leukemia virus, and a third plasmid expressing the vesicular stomatitis virus (VSV) envelope. Pools of stably transfected cells were obtained by retroviral infections. After infection, puromycin-resistant cells were isolated by growth in selective medium (DMEM, 10% FCS, 1 mg/ml G418 [Gibco], and 1 mg/ml puromycin [Sigma] for MDA-MB-435-ß4 cells, and DMEM, 10% FCS, and 5 mg/ml puromycin [Sigma] for MDCK cells). Overexpressing HES-1 cells were obtained by transfection of MDA-MB-435-ß4 cells with the construct pcDNA-HES-1 using Lipofectamine (Invitrogen) together with Plus reagent (Invitrogen), according to the instructions of the manufacturer; transfection with an empty pcDNA3.1 vector was used to produce mock cells. The same transfection method was utilized for getting pBabe or N4-TM MDA-MB-435-ß4 cells to express either control or HES-1 small interfering RNAs (siRNAs) (gift of S. Herzig; described in reference 20). All cells were maintained at 37°C in 5% CO2.
Electrophoretic mobility shift assays (EMSA).
MDA-MB-435-ß4 nuclear extracts were obtained using a nuclear extraction kit (Panomics), following the instructions of the manufacturer, and then they were dialyzed in a buffer containing 10 mM HEPES, pH 7.9, 25 mM KCl, 1 mM EDTA, 5 mM MgCl2, 1 mM dithiothreitol, 10% glycerol, and 0.05% NP-40 in the presence of protease and phosphatase inhibitors. 5' GCGAGGCAGACAGACACGTGCTGGGGCGGGCAGG 3' together with 5' CTCGCCTGCCCGCCCCAGCACGTGTCTGTCTGCC 3' and 5' GCGAGGCAGACAGACAAAAGCTGGGGCGGGCAGG 3' together with 5' CTCGCCTGCCCGCCCCAGCTTTTGTCTGTCTGCC 3' were annealed to generate, respectively, HES-1-wt and HES-1-mut double-stranded Met-promoter oligonucleotides. Probes were obtained by a Klenow fill-in reaction of 100 ng of each of the two double-stranded oligonucleotides in the presence of 100 µCi of [
-32P]GTP, according to reference 9. The probe for GAA was obtained as described in reference 64. Labeled oligonucleotides (5 x 104 cpm) were incubated with 10 µg of nuclear extracts and 1 mg/ml poly(dIdC) (Amersham-Pharmacia) for 45 min on ice. In some experiments, nuclear extracts were preincubated for 30 min on ice with 2 µg of either control immunoglobulin G (Sigma) or TransCruz Gel Supershift anti-HES-1 antibodies (Santa Cruz) (66). Complexes were separated using a 5% polyacrylamide-0.5% Tris-borate-EDTA gel and visualized using a Storm apparatus (Molecular Dynamics).
Transcriptional analysis. The wild-type Met-promoter construct used for the transcriptional analysis has been previously described (15). The putative HES binding sequence "CACGTG" (+20 from transcriptional start site) was mutated by a recombinant PCR-based approach (34) to "CAAAAG." Accuracy of the mutagenesis procedure was verified by direct sequencing. Analysis of promoter activity was performed in MDA-MB-435-ß4 cells. By using Lipofectamine (Invitrogen), 1.8 x 106 cells/100-mm-diameter plate were transfected with either 20 µg of the appropriate promoter construct, with 3.5 µg of pcDNA-HES-1 or control pcDNA3.1 vector together with 15 µg of the promoter constructs, or with 2.8 µg of the siRNA constructs together with 10 µg of the appropriate promoter constructs. For the rescuing experiments (see Fig. 4E and F), 8 x 106 of MDA-MB-435-ß4 cells/100-mm-diameter plate were transfected with 10 µg of pcDNA-HES-1; after 3 days, these cells were divided and retransfected with increasing amounts of HES-1 siRNAs (8, 16, 24, and 48 µg) or control siRNAs (48 µg). The results showed that 16 and 24 µg of HES-1 siRNAs were sufficient for silencing exogenously induced overexpression of HES-1. After 4 days, these cells were transfected with 25 µg of the appropriate promoter constructs. For the experiments presented in Fig. 5C and D, 8 x 106 N4-TM MDA-MB-435-ß4 cells/100-mm-diameter plate were transfected with 20 µg of HES-1 siRNAs; after 3 days, these cells were divided and retransfected with increasing amounts of pcDNA-HES-1 (4, 8, 12, 24, and 36 µg). The down-regulation of HES-1 visible in HES-1 siRNA-transfected cells is abrogated by 8 and 12 µg of HES-1. After 4 days, these cells were transfected with 25 µg of the appropriate promoter constructs. A TK-Renilla reporter plasmid (Promega) (0.2 µg) was added to each transfection for efficiency standardization. After 48 h, cells were processed using a dual-luciferase reporter assay system (Promega) according to the manufacturer's instructions. Ten microliters of cell lysate was used to determine reporter enzyme activity by using a Lumat LB 9507 luminometer (Berthold). Each experimental point was performed in quadruplicate.
|
|
Biological assays. In all the biological assays we used baculovirus recombinant HGF (37). Conditioned medium of uninfected insect cells, treated as the supernatant derived from HGF-baculovirus-infected cells, provided the negative control. In the migration assays, 5 x 105 cells in 200 µl of DMEM supplemented with 0.2% FCS were seeded in the upper compartment of Transwell chambers (6.5 µm, polycarbonate filter with 8-µm pore size; Costar Corporation). DMEM (500 µl) plus 0.2% FCS, containing either control medium or 40 ng/ml of HGF, was added to the lower chamber. The plates were incubated at 37°C for 12 h, and then the cells on the upper side of the polycarbonate filter were mechanically removed. Cells that had migrated to the lower side of the filter were stained with crystal violet and counted using a Fast-Read 102 New Grid counting chamber (Roche Diagnostics) under a Dialux microscope (Leitz). The invasion assays were performed as the migration assays described above, but here the upper sides of the filters were coated with 15 µg/cm2 of reconstituted Matrigel basement membrane (Collaborative Research). For the tubulogenesis assays, the trypsinized cells were suspended to a final concentration of 1.5 x 105 cells/ml (MDA-MB-435-ß4) or 1 x 105 cells/ml (MDCK) in gelling solution prepared as follows: 800 µl of type I collagen S (3.4 mg/ml; Collaborative Biomed), 1x DMEM, 40 mM HEPES, 37 g/liter NaHCO3, all to a final volume of 1 ml. Eighty microliters of this mixture was seeded into 96 microtiter plate wells and allowed to gel for 10 min at 37°C before overlying with 80 µl DMEM containing 20% FCS. After 2 h at 37°C, the upper aqueous phase was removed and substituted with fresh DMEM medium plus 10% FCS. After further 24 h, either control medium or 80 ng/ml of HGF was added in the upper aqueous phase. Tubules were scored every day for 2 weeks.
Biochemical methods.
For immunoprecipitation, 1 x 106 cells or 100 embryos were lysed for 15 min at 4°C with 1 ml of lysis buffer (50 mM HEPES, pH 7.4, 5 mM EDTA, 2 mM EGTA, 150 mM NaCl, 10% glycerol, and 1% NP-40), in the presence of protease and phosphatase inhibitors. Extracts were quantified with a bicinchoninic acid protein assay reagent kit (Pierce), and an equal amount of proteins was incubated with MAbs anti-VSV (clone P5D4; Sigma) or anti-hMet-DQ13 (produced in our laboratory). Immune complexes were collected with Sepharose-protein A resin (Amersham-Pharmacia) supplemented with 1.2 µg of rabbit anti-mouse immunoglobulin (RAMIG; Pierce) for 2 h at 4°C and then washed 4x in ice-cold lysis buffer and eluted in boiling Laemmli buffer. Western blot analysis was performed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels. The gels were blotted onto a nylon membrane (polyvinylidene difluoride transfer membrane; Millipore Corporation); the membrane was blocked using 10% bovine serum albumin in Tris-buffered saline-Tween for 2 h at 45°C; the primary antibodies (MAb anti-VSV, Sigma; polyclonal antibody [PAb] anti-Met-C12, Santa Cruz; MAb anti-phosphotyrosine, Upstate Biotechnology; PAb anti-active-MAPK, Promega; PAb anti-MAPK, Cell Signaling; PAb anti-actin, Santa Cruz; PAb anti-active-AKT, Cell Signaling; PAb anti-AKT, Cell Signaling; PAb anti-HES-1, kindly provided by T. Sudo; MAb anti-panRAS, Transduction Laboratories) were incubated in 5% bovine serum albumin in Tris-buffered saline-Tween according to the dilutions suggested by the manufacturers, for 12 h at 4°C. Specific signals were detected with peroxidase-conjugated secondary antibodies (Amersham-Pharmacia) and an enhanced chemiluminescence system (ECL; Amersham-Pharmacia). Pool-down experiments were done according to reference 47; cells were maintained in serum-free medium for 24 h, and then they were either left untreated or stimulated with 50 ng/ml of HGF for the indicated times. The construct Raf-RBD-GST-pGEX (gift of E. Audero) was expressed in bacteria and purified as described in reference 47. In each experimental point, 1 mg of total protein extracts was incubated with 40 µg of Raf-RBD-GST fusion protein. For auto-kinase assays, immunoprecipitated embryonic extracts were washed twice with ice-cold KB buffer (20 mM HEPES, pH 7.1, 5 mM MgCl2, 100 mM NaCl), and the phosphorylation reaction was performed in the presence of [
-32P]ATP and 10 mM unlabeled ATP for 15 min at 37°C. The reactions were stopped by adding boiling Laemmli buffer and the samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels. The signals were visualized with a Storm apparatus (Molecular Dynamics).
Fly stocks. The following fly strains were used in this study: the w1118 strain (gift of S. Roth) was used for P-element-mediated germ line transformation; trachealess enhancer trap line 1-eve-1 (Perrimon, 1991, no. 2014) was a gift of M. Affolter; and the NCo, CyO, and CyO-ftz-LacZ strains were provided from the Bloomington Stock Centre.
Embryo fixation and staining. Embryos were staged according to reference 6. They were collected on apple juice agar plates following standard protocols (63), and then they were processed as previously described (55). For the staining we used the following primary antibodies: MAb 2A12 (developed by C. Goodman and N. Patel; diluted 1:10) and MAb C17.9C6 (anti-Notch, intracellular domain, diluted 1:5; developed by S. Artavanis-Tsakonas), both obtained from the Developmental Studies Hybridoma Bank. Anti-ß-galactosidase (ß- gal) antibodies, purchased from Cappel, were diluted 1:1,000; anti-human Met antibody, clone C12, purchased from SantaCruz, after preabsorption using wild-type embryos, was used 1:100. The primary antibodies were diluted in 1% FCS. The incubations with the primary antibodies were performed at 4°C for 12 h. The secondary antibodies were as follows: biotinylated anti-rabbit and anti-mouse (diluted 1:5,000; Vector Laboratories) and FITC-conjugated anti-rabbit and tetramethyl rhodamine isocyanate-conjugated anti-mouse (diluted 1:500; Molecular Probes). All the secondary antibodies were incubated for 2 h at room temperature. The stainings performed with the biotinylated antibodies were developed using a Vectastain Elite ABC kit (Vector Laboratories) with 3'-3'-diaminobenzidine (Sigma) as chromogen. Subsequently, the embryos were dehydrated, rinsed in acetone, and transferred to a 1:1 mixture of acetone and Durcupan (Fluka) overnight. After acetone evaporation, embryos were observed and photographed by using a Dialux microscope (Leitz). For genotyping, embryos were first stained with biotinylated antibodies to label the tracheas and then with anti-ß-gal and MAb C17.9C6 followed by fluorochrome-conjugated secondary antibodies. NCo torso-Met mutants were identified for the absence of both ß-gal and MAb C17.9C6 staining. They were collected under an Axioplan microscope (Zeiss), mounted in Durcupan (Fluka), and photographed as previously described.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-glucosidase (GAA), a known target gene of the HES-Notch pathway (64), was monitored. As expected, incubation of nuclear extracts with a probe containing the HES binding site on the GAA promoter determined the formation of a complex (Fig. 1C, first lane). The same was true when the nuclear extracts were incubated with a probe containing the wild-type HES putative binding site on the Met promoter, whereas incubation with a probe mutated in this binding site did not lead to complex formation (Fig. 1C, compare the third and second lanes). The specificity of the complex was further examined by competition experiments using either unlabeled HES wild-type or HES mutated Met promoter probes. By adding the mutated competitor, the complex was only slightly reduced in a dose-independent fashion, suggesting an unspecific quenching effect of the mutated competitor over HES-1/Met complex formation (Fig. 1C, lanes six and seven). When a 10-fold molar excess of wild-type competitor was added to the reaction, complex formation was reduced, and it was almost totally abolished in the presence of a 50-fold molar excess (Fig. 1C, lanes eight and nine). The addition of anti-HES-1 antibodies reduced the intensity of the HES-1/Met complex compared to control unspecific immunoglobulins (Fig. 1C, lanes ten and eleven), thus indicating a specific interference of anti-HES-1 antibodies in the complex formation. Altogether, these data confirm the in silico observation by demonstrating that indeed HES-1 binds to Met promoter.
|
|
Intriguingly, Notch has been shown to affect the function of tyrosine kinase receptors through inhibition of Ras signaling (2, 8, 61). Because the need for the integrity of Ras signals in Met-dependent responses has been highlighted in several experimental settings of invasive growth, including tubulogenesis in tridimensional collagen cultures (1, 16, 25) and embryo development (31), we analyzed the activation levels of this Met effector following HGF stimulation in both MDA-MB-435-ß4 and MDCK cells, either mock or N4-TM transduced. As a biochemical readout of Ras activity, we measured the amplitude and duration of HGF-dependent activation of mitogen-activated protein kinase (MAPK). As shown in Fig. 3A, while HGF induced MAPK activation in mock cells, it failed to stimulate MAPK activity in N4-TM transfectants. To further validate this result, we monitored HGF-dependent Ras activation in mock and N4-TM MDCK cells by Ras-GTP pull-downs using the Ras-binding domain of Raf-1 fused to glutathione S-transferase (GST). HGF treatment determined an increase of Ras-GTP in mock cells but not in N4-TM cells, thus confirming a Notch-based repression of HGF-triggered Ras signaling (Fig. 3B). Ras-GTP pull-downs in MDA-MB-435-ß4 cells were hampered by low expression levels of Ras. Differently from Ras, HGF-triggered activation of AKT/PKB, a downstream target of PI3K crucially involved in Met-dependent invasive growth, was unaffected in N4-TM versus mock cells (Fig. 3C). Combined, these data suggest that Notch can, on one hand, decrease Met transcription and, on the other, selectively impinge on specific interruption of the Ras signaling cascade.
|
Met activation induces Notch signaling, which through HES-1 represses Met mRNA synthesis. It is now becoming increasingly clear that subtle regulation of tyrosine kinase activity may rely on negative checkpoints that counter-feedback positive inputs. One of the possible mechanisms whereby this can be achieved is through transcriptional induction of regulatory molecules that, in turn, can affect the activity of transcription factors and coactivators (3, 48). Based on this information, we hypothesized that the inhibitory function of Notch on Met could be induced by Met itself, within an intrinsic circuit of transcriptional cross talk.
Notch activity is first primed by elevation of the cellular levels of its ligands (in vertebrates, two homologues of Serrate [Jagged1 and -2] [26, 60] and four Delta-like genes [Delta 1 to -4] [17, 23, 50]) and further proceeds through nuclear translocation, association with the transcription factor Suppressor of Hairless/CBF-1, and transactivation of several target genes coding for basic helix-loop-helix proteins, namely the genes belonging to the hairy/Enhancer of Split that include HES-1 [E(spl) family] (23).
To validate the prediction that Met stimulates the activity of its own repressor, we performed a time course RT-PCR analysis to track HGF-dependent transcriptional induction of the various Notch ligands and of HES-1. Indeed, HGF stimulation of MDA-MB-435-ß4 cells induced the transient transcription of HES-1 and of two Delta mRNAs, Dl1 and Dl4 (Fig. 4A). Among the other ligands, J1 was expressed constitutively (Fig. 4A), whereas J2 and Dl3 were not expressed nor were they induced by HGF (data not shown).
Given that HGF induces HES-1 mRNA and that activation of Notch, the HES-1 upstream regulator, leads to Met inhibition, we tested if altered expression of HES-1 protein results in modifications of Met mRNA transcription. This was done by monitoring the luciferase activity of Met-promoter reporter constructs (see above) in MDA-MB-435-ß4 cells in which HES-1 was overexpressed by transient transfection. In agreement with the inhibitory function of the Notch pathway on Met synthesis, overexpression of HES-1 led to a 30% decrease in Met-promoter activity (Fig. 4B). As expected, luciferase activity of the Met-promoter construct in which the HES-1 binding site was mutated was insensitive to dose alterations of HES-1 (Fig. 4B). Accordingly, overexpression of HES-1 resulted in down-regulation of the Met protein (Fig. 4C), which leads to reduced biological responses to HGF treatment, measured as growth in tridimensional collagen gel (Fig. 4D). To further support the correlation between HES-1 activity and Met expression, we used the RNA interference technology to progressively down-modulate HES-1 protein levels in MDA-MB-435-ß4 cells exogenously overexpressing HES-1. Addition of a HES-1-specific siRNA resulted in a dose-response increase in Met protein (Fig. 4E, lanes three and four), whereas an unrelated siRNA had no effect (see Fig. 4E, lane five). Accordingly, luciferase activity analysis of the Met-promoter construct in these cells showed that siRNA-mediated inhibition of HES-1 expression abrogated the transcriptional down-regulation of the Met gene (Fig. 4F).
By using an opposite but complementary approach, we monitored luciferase activity of the Met-promoter constructs using MDA-MB-435-ß4 cells in which HES-1 function was knocked down by RNA interference technology. Ablation of HES-1 in cells expressing N4-TM restored a promoter activity similar to that observed in mock cells; as previously shown, mutations in the HES-1 binding domain of the Met promoter region abolished Notch-dependent down-regulation of Met synthesis irrespective of HES-1 expression (Fig. 5A). Consistent with these findings, HES-1 inactivation increased expression of the Met protein in N4-TM-expressing cells (Fig. 5B). Again, gradual increase of HES-1 expression in these HES-1– cells led to recovery of the Notch inhibitory activity, with consequent progressive reduction of Met protein levels (Fig. 5, panels C and D).
If HES-1 is the Notch effector responsible for attenuation of Met synthesis, then overexpression of HES-1 should mimic Notch activity in impairing Met-dependent signals. To test this hypothesis, we compared Met expression, PI3-kinase stimulation (in terms of activated PKB/AKT), and Ras stimulation (in terms of MAPK activity) upon HGF treatment in MDA-MB-435-ß4 cells, expressing either N4-TM or HES-1. As previously shown, both N4-TM- and HES-1-expressing cells displayed reduced Met expression (Fig. 5E, first panel), which correlates with the expression level of HES-1 (Fig. 5E, sixth panel). Moreover, cells expressing either N4-TM or HES-1 exhibited reduced MAPK activation (Fig. 5E, second panel) in response to HGF stimulation, whereas PKB/AKT activity was unaffected (Fig. 5E, fourth panel). The amounts of extracellular signal-regulated kinases (ERKs), AKT/PKB, and actin confirmed equal loading in all the experimental conditions (Fig. 5E, third, fifth, and sixth panels). This indicates that HES-1 overexpression recapitulates the repression of Met signaling observed in N4-TM transfectants.
In summary, these data define a novel epistatic relationship in which Met activation results in induction of Notch signaling that in turn represses Met activity through HES-1.
Cross talk between Met and Notch signaling in Drosophila. To date, four vertebrate Notch genes have been identified (Notch1 to -4) (23) together with six Notch ligands (17, 23, 26, 50, 60). The complexity of Notch and Notch ligand family members, as well as the evidence that loss of either HGF or Met in the mouse results in early embryonic lethality, points to the necessity of exploiting alternative animal models to explore in vivo the interaction between Met and Notch. This model should not require Met function during development and in adult life, and in the meantime it should be easily amenable to genetic manipulation. We thus decided to scrutinize the effects of Met-Notch interactions during development of the tracheal tree in Drosophila melanogaster. The advantages of such an experimental approach are twofold: Met is not expressed in Drosophila (54), so this organism represents a naive recipient for clear-cut dissection of Met-dependent responses without any kind of endogenous developmental interference; moreover, formation of the tracheal tree in Drosophila is a striking example of branching morphogenesis in vivo, in which the role of Notch pathway has been extensively documented (21, 28).
The Drosophila tracheal system is an elaborate network of epithelial tubes that ramifies in some 10,000 branches. The network arises from segmentally repeated clusters of ectodermal cells, the placodes, which invaginate to form sac-like structures called tracheal pits. Each tracheal pit sprouts successively finer branches, some of which ultimately fuse to form an arborescent pattern throughout the embryo. Unlike sprouting of the major tracheal branches, which is simple and stereotyped, ramification of the terminal branches is complex and variable and is affected by several morphogenetic regulators. Among these, the Notch signaling pathway is needed in the tracheal cells for the fusion and terminal branching programs (21, 28). In particular, Notch is first required to single out the cells that will control branch fusion from a group of competent tracheal cells and later to select the correct number of terminal branches arising from some primary branches.
To confirm in this in vivo setting our assumption that Met activation results in induction of Notch function, which in turn represses Met activity, we followed three subsequent experimental lines: first, we analyzed whether expression of a constitutively active Met in Drosophila leads to Notch stimulation (through increased expression of its physiological ligands); then, as a basal read-out of Met function, we examined the tracheal phenotype of Met-expressing flies; finally, we analyzed whether this phenotype is affected in a Notch-null background.
Construction and expression of the chimeric torso-Met receptor in Drosophila. To generate transgenic flies in which Met signaling is constitutively activated, we produced a chimeric fusion protein in which the extracellular and transmembrane domains of human Met were substituted with the corresponding domains of torso4021 (torso-Met). torso4021 is a Drosophila tyrosine kinase mutant receptor bearing a single amino acid substitution (Y 327 C) that leads to spontaneous dimerization and consequent constitutive activation of the receptor (53). The expression of this hybrid receptor was driven by the promoter hsp70, which allows a ubiquitous and heat-inducible expression of the protein.
The torso-Met construct was checked for correct expression by in vitro translation experiments (data not shown) and then introduced into Drosophila embryos by P-element-mediated gene transfer. Six independent P-element insertions were mapped by in situ hybridization onto polytene chromosomes, and different lines were generated using specific marked balancer chromosomes. All the transformant lines were maintained at 18°C to prevent torso-Met activation. To test the expression and the biochemical properties of the torso-Met recombinant protein in vivo, the transgene was induced by heat shock treatment (37°C, 20 min) in the progeny of the transformant flies, and then the embryos were transferred at 18°C for a further 3 h. Immunoprecipitation of lysates from heat-shocked and untreated embryos with specific monoclonal antibodies raised against human Met followed by Western blot analysis with anti-Met and anti-phosphotyrosine antibodies shows that the recombinant protein was expressed exclusively in the embryos exposed to heat shock treatment and had the expected molecular weight of 95,000 (Fig. 6A). Moreover, as for activated human Met receptor, torso-Met was tyrosine phosphorylated (Fig. 6B) and underwent autophosphorylation in autokinase assays (Fig. 6C). This indicates that the recombinant torso-Met protein expressed in Drosophila retains the catalytic properties of human wild-type Met.
|
Phenotypic analysis of the tracheal tree in Met-expressing Drosophila. To analyze the embryonic tracheal tree of torso-Met embryos, transformant flies were crossed with flies bearing the enhancer trap line 1-eve-1, which expresses the bacterial protein ß-gal in all tracheal cells throughout development. The tracheas were detected with antibodies against ß-gal protein or with the antibody 2A12, a marker of terminal tracheal differentiation that recognizes a glycoprotein secreted in the tracheal lumen. The heat shock treatment used for transgene expression did not alter the tracheal trees of embryos resulting from the cross of wild-type and 1-eve-1 flies (data not shown). Since the tracheas of these embryos were indistinguishable from those of uninduced embryos carrying the 1-eve-1 and torso-Met chromosomes, we elected uninduced 1-eve-1; torso-MET embryos as control organisms. About 50% of the heat-shocked embryos resulting from the cross between torso-Met and 1-eve-1 flies that expressed ß-gal protein displayed tracheal alterations.
In the first stages of tracheal development, when the tracheal placodes start to invaginate to form the tracheal pits, expression of torso-Met resulted in scattering of the cells that compose the tracheal pits, which appeared as starred aggregates of loosely associated cells (Fig. 6E), whereas in the immediately subsequent developmental stage, torso-Met led to the formation of supernumerary tracheal branches (Fig. 6F, second row). At later stages of development, the cells that form the primary dorsal branches start a highly coordinated migratory program towards the dorsal side of the embryo that ends up in the formation of the secondary and terminal dorsal tracheal branches. When expressed in this stage, torso-Met impaired this stereotyped migratory behavior (Fig. 6G). The cells that compose the stalk were much more disorganized and, in some cases, one of the uppermost two-paired cells acquired an unusual migratory route, eventually breaking its connection with the neighboring cells. As a consequence, the formation of dorsal secondary and terminal tracheal branches was severely affected (Fig. 6H), resulting in the presence of unfused dorsal branches and an excess of terminal branches.
In summary, these results suggest that exogenous expression of active Met amplifies the overall genetic program of Drosophila tracheal morphogenesis, with enhanced sprouting, primitive hyperbranching, and increased terminal ramification.
Absence of Notch enhances Met-dependent tracheal phenotype in Drosophila. If in invertebrates, as in vertebrates, Notch elicits an inhibitory effect on Met-dependent branching morphogenesis, then abrogation of Notch gene function in Drosophila embryos should enhance torso-Met-controlled tracheal branching.
Accordingly, we expressed torso-Met in flies bearing the mutation encoded by the NCo allele, which lacks the intracellular portion of the receptor and behaves as a dominant negative (62). Moreover, since NCo is an antimorphic allele, the severity of the phenotype may be modulated by the dose of wild-type allele present in the embryo. Males bearing the NCo allele were crossed with torso-Met females carrying a balancer chromosome labeled with ß-gal. In the progeny of this cross, the tracheas were evidenced using 2A12 antibody and the embryos were genotyped using anti-ß-gal antibodies and antibodies raised against the C-terminal portion of Notch receptor; embryos lacking ß-gal staining and expressing barely detectable full-length Notch protein were considered having the genotype NCo/+; torso-Met/+.
The NCo mutants presented the typical alteration of Notch loss-of-function mutants (Fig. 7, second row): the dorsal trunk showed conspicuous interruptions; some dorsal branches were irregularly fused, and some others were missing. Interestingly, when analyzing the tracheal trees following activation of torso-Met in NCo mutants, we found that lack of Notch function results in enhanced tracheal branching. In particular, the terminal portions of the tracheas, instead of being blunt-end tubes, as in torso-Met and NCo mutants (Fig. 7, first and second rows), developed additional ramifications with reiterated segmentations of the distal tips (Fig. 7, third row).
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A second level of negative regulation relies on the ability of Notch to suppress Met-dependent activation of the Ras-MAPK cascade. The interactions between Ras and Notch are complex. Much evidence, both in invertebrates and in vertebrates, suggests that the Ras pathway has a role in Notch activation, but at the same time Ras can be a target of either Notch-mediated activation or suppression (2, 7, 8, 14, 21, 28, 61). Our data show that the final outcome of a constitutive activation of Notch is a dominant inhibition of HGF-dependent Ras activation.
Notch-dependent down-regulation of Met can be dynamically induced by Met itself. We found that activation of Met results in stimulation of Notch signaling by transient transcriptional induction of its Delta ligands. This suggests the existence of a negative loop in which Met, through Delta, activates Notch. Notch then acts by inhibiting Met. This conclusion if further supported by the evidence that Met activation results in transcription of HES-1, a critical nuclear effector of Notch (see above) (1, 23, 31, 44, 59). The identification of Delta and HES-1 as transcriptional targets of Met activity expands the inventory of the Met transcriptome that, so far, includes only a handful of mechanical regulators of cell movement and survival, such as matrix components (33) and matrix proteases (13, 19).
Here we exploited the versatility of the Drosophila system as a feasible model to assess the interactions between Met and Notch in vivo. As a phenotypic readout of genetic analysis, we chose the development of the embryonic tracheal tree, where the function of Notch has been thoroughly analyzed (21, 28).
In Drosophila, expression of an active form of Met induces transcription of the Notch ligand Delta, as in mammalian cells (see above). Moreover, it leads to scattering of the cells that compose the tracheal pits, potentiates the migratory behavior of the primary and secondary tracheal branches and determines the formation of supernumerary tracheal tubes. All these cues are remarkably similar to the morphogenetic processes controlled by the HGF/Met system in vertebrates, namely dissociation of epithelial cells from their original site of accretion and cell reorganization to generate protrusions that bifurcate and expand, eventually resulting in an arborescent pattern of polarized tubules (43). In embryos lacking Notch gene function, Met amplifies tracheal tubulogenesis, in agreement with the inhibitory role of Notch over Met-dependent morphogenesis postulated in this paper.
In conclusion, our findings suggest that Met can self-limit the strength and duration of its positive signals by a twofold mechanism: on the one hand, Met-dependent induction of Notch function leads to a direct repression of Met levels; on the other, Notch can further subsidize this negative tuning by decreasing Ras signaling, with consequent lessening of Met-dependent transduction pathways. This unconventional mechanism of Met negative modulation goes along with the well-established roles of receptor endocytosis and ubiquitin-mediated degradation for attenuation of receptor signaling (12, 40). A delicate balance between positive and negative signals is critical for normal cell homeostasis, and a prevalence of positive signals leading to excessive cell stimulation is commonly found in cancers. Not surprisingly, aberrant execution of HGF-dependent invasive growth in neoplastic cells leads to cancer invasion and metastasis. The finding that Notch can efficiently harness Met-dependent transcription and Met signaling discloses a novel inhibitory pathway that might be exploited to hamper Met-based cancer progression.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by research grants of AIRC, CNR-MIUR, FIRB-MIUR, MIUR-PRIN, and the Foundation Compagnia di San Paolo to P.M.C.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Berset, T., E. F. Hoier, G. Battu, S. Canevascini, and A. Hajnal. 2001. Notch inhibition of RAS signaling through MAP kinase phosphatase LIP-1 during C. elegans vulval development. Science 291:1055-1058.
3. Blume-Jensen, P., and T. Hunter. 2001. Oncogenic kinase signalling. Nature 411:355-365.[CrossRef][Medline]
4. Boccaccio, C., M. Ando, L. Tamagnone, A. Bardelli, P. Michieli, C. Battistini, and P. M. Comoglio. 1998. Induction of epithelial tubules by growth factor HGF depends on the STAT pathway. Nature 391:285-288.[CrossRef][Medline]
5. Bussolino, F., M. F. Di Renzo, M. Ziche, E. Bocchietto, M. Olivero, L. Naldini, G. Gaudino, L. Tamagnone, A. Coffer, and P. M. Comoglio. 1992. Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J. Cell Biol. 119:629-641.
6. Campos-Ortega, A., and V. Hartenstein. 1997. The embryonic development of Drosophila melanogaster, 2nd ed. Springer-Verlag KG, Berlin, Germany.
7. Carmena, A., E. Buff, M. S. Halfon, S. Gisselbrecht, F. Jimenez, M. K. Baylies, and A. M. Michelson. 2002. Reciprocal regulatory interactions between the Notch and Ras signaling pathways in the Drosophila embryonic mesoderm. Dev. Biol. 244:226-242.[CrossRef][Medline]
8. Chihara, T., and S. Hayashi. 2000. Control of tracheal tubulogenesis by Wingless signaling. Development 127:4433-4442.[Abstract]
9. Coligan, S. E., B. M. Dunn, H. L. Ploegh, D. W. Speicher, and P. Wingfield. 1998. Current protocols in protein science, vol. 1. J. Wiley & Sons, Inc., New York, N.Y.
10. Danilkovitch-Miagkova, A., and B. Zbar. 2002. Dysregulation of Met receptor tyrosine kinase activity in invasive tumors. J. Clin. Investig. 109:863-867.[CrossRef][Medline]
11. Dickson, B., F. Sprenger, and E. Hafen. 1992. Prepattern in the developing Drosophila eye revealed by an activated torso-sevenless chimeric receptor. Genes Dev. 6:2327-2339.
12. Dikic, I., and S. Giordano. 2003. Negative receptor signalling. Curr. Opin. Cell Biol. 15:128-135.[CrossRef][Medline]
13. Dunsmore, S., J. Rubin, S. Kovacs, M. Chedid, W. Parks, and H. Welgus. 1996. Mechanisms of hepatocyte growth factor stimulation of keratinocyte metalloproteinase production. J. Biol. Chem. 271:24576-24582.
14. Fitzgerald, K., A. Harrington, and P. Leder. 2000. Ras pathway signals are required for notch-mediated oncogenesis. Oncogene 19:4191-4198.[CrossRef][Medline]
15. Gambarotta, G., S. Pistoi, S. Giordano, P. M. Comoglio, and C. Santoro. 1994. Structure and inducible regulation of the human MET promoter. J. Biol. Chem. 269:12852-12857.
16. Giordano, S., A. Bardelli, Z. Zhen, S. Menard, C. Ponzetto, and P. M. Comoglio. 1997. A point mutation in the MET oncogene abrogates metastasis without affecting transformation. Proc. Natl. Acad. Sci. USA 94:13868-13872.
17. Gray, G., R. Mann, R. Mitsiadis, D. Henrique, M. Carcangiu, A. Banks, J. Leiman, D. Ward, D. Ish-Horowitz, and S. Artavanis-Tsakonas. 1999. Human ligands of the Notch receptor. Am. J. Pathol. 154:785-794.
18. Gual, P., S. Giordano, T. Williams, S. Rocchi, E. Van Obberghen, and P. Comoglio. 2000. Sustained recruitment of phospholipase C-gamma to Gab1 is required for HGF-induced branching tubulogenesis. Oncogene 19:1509-1518.[CrossRef][Medline]
19. Harvey, P., I. Clark, M. Jaurand, R. Warn, and D. Edwards. 2000. Hepatocyte growth factor/scatter factor enhances the invasion of mesothelioma cell lines and the expression of matrix metalloproteinases. Br. J. Cancer 89:1147-1153.
20. Herzig, S., S. Hedrick, I. Morantte, S. H. Koo, F. Galimi, and M. Montminy. 2003. CREB controls hepatic lipid metabolism through nuclear hormone receptor PPAR-gamma. Nature 426:190-193.[CrossRef][Medline]
21. Ikeya, T., and S. Hayashi. 1999. Interplay of Notch and FGF signaling restricts cell fate and MAPK activation in the Drosophila trachea. Development 126:4455-4463.[Abstract]
22. Iso, T., L. Kedes, and Y. Hamamori. 2003. HES and HERP families: multiple effectors of the Notch signaling pathway. J. Cell. Physiol. 194:237-255.[CrossRef][Medline]
23. Joutel, A., and E. Tournier-Lasserve. 1998. Notch signaling pathway and human diseases. Semin. Cell Dev. Biol. 9:619-625.[CrossRef][Medline]
24. Kawaida, K., K. Matsumoto, H. Shimazu, and T. Nakamura. 1994. Hepatocyte growth factor prevents acute renal failure and accelerates renal regeneration in mice. Proc. Natl. Acad. Sci. USA 91:4357-4361.
25. Khwaja, A., K. Lehmann, B. M. Marte, and J. Downward. 1998. Phosphoinositide 3-kinase induces scattering and tubulogenesis in epithelial cells through a novel pathway. J. Biol. Chem. 273:18793-18801.
26. Lindsell, C. E., C. J. Shawber, J. Boulter, and G. Weinmaster. 1995. Jagged: a mammalian ligand that activates Notch1. Cell 80:909-917.[CrossRef][Medline]
27. Liu, Z. J., T. Shirakawa, Y. Li, A. Soma, M. Oka, G. P. Dotto, R. M. Fairman, O. C. Velazquez, and M. Herlyn. 2003. Regulation of Notch1 and Dll4 by vascular endothelial growth factor in arterial endothelial cells: implications for modulating arteriogenesis and angiogenesis. Mol. Cell. Biol. 23:14-25.
28. Llimargas, M. 1999. The Notch pathway helps to pattern the tips of the Drosophila tracheal branches by selecting cell fates. Development 126:2355-2364.[Abstract]
29. Maher, J. J. 1993. Cell-specific expression of hepatocyte growth factor in liver. Upregulation in sinusoidal endothelial cells after carbon tetrachloride. J. Clin. Investig. 91:2244-2252.
30. Mailhos, C., U. Modlich, J. Lewis, A. Harris, R. Bicknell, and D. Ish-Horowicz. 2001. Delta4, an endothelial specific notch ligand expressed at sites of physiological and tumor angiogenesis. Differentiation 69:135-144.[CrossRef][Medline]
