| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
Two distinct mutations of the RET receptor causing Hirschsprung's disease impair the binding of signalling effectors to a multifunctional docking site
Introduction
Results
Discussion
Materials And Methods
Patients and families
Construction of vectors
Cell culture and antibodies
Production of recombinant retrovirus
Protein analyses
GST-fusion affinity precipitation
Yeast two-hybrid assay
CAT assay
Acknowledgements
References
Two distinct mutations of the RET receptor causing Hirschsprung's disease impair the binding of signalling effectors to a multifunctional docking site
Received March 2, 1999; Revised and Accepted August 5, 1999
The RET gene codes for a transmembrane tyrosine kinase which is a subunit of a multimeric complex that acts as a receptor for four structurally related molecules: the glial cell line-derived neurotrophic factor (GDNF), neurturin, artemin and persephin. Germline mutations of RET cause a dominantly inherited dysgenesis of the enteric nervous system known as Hirschsprung's disease (HSCR; aganglionosis megacolon). The majority of HSCR mutations results either in a reduction of dosage of the RET protein or in the loss of RET function. Two novel distinct mutations of RET that led either to the deletion of codon 1059 (denoted [Delta]1059) or to the substitution of a Pro for Leu1061 have been identified in five HSCR families. In one large pedigree, two children born from asymptomatic consanguineous parents presented a severe form of HSCR and were found to carry the mutation at codon 1061 in the homozygous state. A tyrosine residue at position 1062 is an intracytoplasmic docking site that enables RET to recruit several signalling molecules, including the Shc adaptor protein. We now report that both HSCR mutations impair the fixation of Shc to RET and consequently prevent its phosphorylation. In addition, quantitative analysis in PC12 cells reveals that mutation [Delta]1059 inactivates the ability of RET to transduce a downstream signal whereas mutation L1061P only partially inhibits the signalling of RET. Finally, we provide evidence that these effects are partly mediated via the disruption of the RET/Shc interaction. Collectively, these results demonstrate that HSCR can be ascribed to mutations of RET which interfere with the binding of transduction effectors, such as Shc, and further provide a biochemical explanation for the phenotype of patients carrying a homozygous mutation at codon 1061. Finally, these data indicate that Y1062 is a multifunctional docking site that confers to RET the capacity to engage downstream signalling pathways which exert a crucial role during enteric neurogenesis.
INTRODUCTION
The RET proto-oncogene codes for a transmembrane receptor tyrosine kinase (RTK) which possesses a cadherin-like domain in its extracellular part (1; reviewed in refs 2,3). Two major isoforms which differ by 9 and 51 amino acids, respectively, at the C-terminus (here referred as to RET9 and RET51) are encoded by alternatively spliced transcripts (4,5). Glial cell line-derived neurotrophic factor (GDNF), neurturin (NTN), persephin and artemin, four homologous molecules which exert a trophic effect on various neuronal populations, interact with a multisubunit complex comprising RET and one member of a novel family of related glycosyl-phosphatidylinositol (GPI)-anchored receptors called GFR[alpha] (for references see ref. 6). So far, four distinct GFR[alpha] (1-4) have been characterized. GFR[alpha]-1-4 have been shown to preferentially interact with GDNF, NTN, artemin and persephin, respectively, thus demonstrating that the nature of GFR[alpha] dictates the ligand specificity of the receptor complex (for references see ref. 6). The GDNF signal mediated through RET and GFR[alpha]-1 has a critical role for the development of the enteric nervous system and kidney as attested by the similar phenotype of mice with GDNF, RET or GFR[alpha]-1 null mutations (7-12).
On interaction with its cognate ligands, RET undergoes autophosphorylation; the phosphorylation of intracellular tyrosine residues generates docking sites for an array of signalling effectors. Phospholipase C [gamma] (PLC-[gamma]) interacts with Y1015 while both Shc and Enigma, a protein which displays an N-terminal PDZ domain and three C-terminal LIM domains, bind to Y1062 (13-17). In addition, Grb2 binds directly to a tyrosine specific to the RET51 isoform (Y1096) (18,19). Finally, evidence has been provided that RET interacts with Grb7/Grb10, the Src kinase and Ras-GAP (13,20-22). As exemplified for several RTKs, the generation of a RET-mediated downstream signal depends on the transient assembly and the cellular compartmentalization of multimolecular complexes comprising various adapter proteins and enzymes. The recruitment of one of these complexes requires fixation of Shc, via its Src homology 2 (SH2) or phosphotyrosine-binding (PTB) domain, and subsequent phosphorylation of Shc, presumably on two distinct tyrosines, creates two docking sites for Grb2 (reviewed in ref. 23). The adapter protein Grb2 stably associates with mSOS, a Ras GDP/GTP exchange protein, and the relocalization of Grb2-mSOS at the inner face of the plasma membrane allows the activation of Ras and leads to the stimulation of the mitogen-activated protein kinase (MAPK) pathway (reviewed in ref. 24). It has been recently reported that the PTB domain of Shc interacts with phosphotyrosine 1062 (pY1062) of both RET9 and RET51; furthermore, RET9 but not RET51 is able to associate with Shc SH2, although the Shc PTB displays a higher affinity than does the Shc SH2 (16,17). Consistent with this, the amino acid sequence surrounding Y1062 matches the consensus sequence defined for the binding of the Shc PTB (22-25), i.e. [Psi]-(D/E)-N-(X/P)-X-pY (where [Psi] is a hydro- phobic amino acid) (Fig. 1). The presence of both a hydrophobic residue in position -5 and an asparagine in position -3 has been demonstrated to be required for the ability of the Shc PTB to recognize its target site in a variety of membrane-associated proteins (25-28). In contrast to PTB, SH2 binding specificity is regulated by residues C-terminal to the pY and several studies have suggested that a hydrophobic residue at position +3 is preferred for optimal binding of the Shc SH2 (29). In agreement with this prediction, RET9, which binds to the SHC SH2, has an isoleucine at position +3, while RET51, which poorly binds to the SHC SH2, possesses a serine at the same position. Further work has shown that mutation of Y1062 ablates the transforming ability of oncogenic forms of RET, thus indicating that the interaction with Shc is a crucial step in the RET-mediated activation of neoplastic signalling (15).
Figure 1. Pedigree of the HSCR family with two patients carrying a homozygous mutation at codon 1061 (L1061P) of RET. Black symbols indicate HSCR. Persons are presented with their genotypes at the RET locus: +, wild-type allele; m, mutation at codon 1061.
Hirschsprung's disease (HSCR) is a common congenital disorder (1 in 5000 live births) typified by a lack of enteric neurons along variable lengths of the hindgut leading to intestinal obstruction (3,30). Germline mutations of RET have been identified in ~50% of the familial cases of HSCR and in 15-20% of the sporadic cases (reviewed in refs 3,30). These RET mutations are scattered throughout the coding sequence and include deletion, insertion, frameshift, nonsense and missense mutations (3,30). We and others (31-34) have shown that HSCR mutations disrupt RET function and that the molecular mechanisms involved depend on the position of the mutation. For example, mutations located in the extra- cytoplasmic domain impair the maturation and the translocation of RET at the plasma membrane whereas mutations in the tyrosine kinase domain perturb catalytic activity. Germline mutations of RET are also responsible for a dominantly inherited cancer syndrome called multiple endocrine neoplasia type 2 (MEN 2) (reviewed in refs 2,3).
In the present report, we have analyzed the functional consequences of two distinct RET mutations [deletion of codon 1059 (denoted [Delta]1059) and L1061P] causing HSCR and identified in five pedigrees, including two children carrying a homozygous mutation at codon 1061. We show that these two mutations implicate residues essential for the recognition of RET by the PTB domain of Shc, such that mutation [Delta]1059 completely abolishes the binding of Shc to RET while mutation L1061P partially inhibits the RET-Shc interaction. Consequently, these two mutations disrupt the ability of RET to transduce downstream signals, albeit to different degrees.
RESULTS
During the screening of 95 HSCR patients, two single mutations of RET which led either to the deletion of codon 1059 ([Delta]1059) or to the substitution of a Pro for Leu 1061 (L1061P) were identified in five probands (R.M.W. Hofstra, Y. Wu, R.P. Stulp, J. Osinga, S.M. Maas, L. Breslau-Siderius, A. Brooks, J.J. Van de Ende, V.M.R. Heyendael, N.M.A. Bax, C. Meijers, C.H.C.M. Buys, submitted for publication). The first of these two mutations ([Delta]1059) was identified in two independent families. In the first family, this mutation was found in twin girls who had total colonic aganglionosis and were born from healthy parents. In the second family, the proband presented a short-segment HSCR (aganglionosis restricted to the rectosigmoid colon). The CTC->CCC mutation at codon 1061 (L1061P) was detected in patients of three families. In two of these kindreds, the probands had a short- or a long-segment HSCR, respectively, and there was no family history of HSCR. In the third family, two children affected with a total colonic aganglionosis and born of asymptomatic consanguineous parents were found to carry a homozygous mutation at codon 1061 (Fig. 1); the third case of HSCR diagnosed in this pedigree (VII:3) was a young boy who displayed a long-segment HSCR and was a heterozygous carrier of the mutation at codon 1061 (Fig. 1).
These two HSCR mutations occur adjacent to residue Y1062 of RET and therefore might be expected to impair the ability of RET to bind to transduction effectors, such as Shc. Therefore, our study was aimed at analyzing the functional consequences of these two mutations. The effects of three additional mutations, I1057D, N1059A and I1065D, supposed to specifically prevent RET from binding to either the PTB or the SH2 domain of Shc, were also assessed (Fig. 2). Finally, Y1015, which has been shown to be the PLC-[gamma] binding site on RET, was also included as a control.
Figure 2. Schematic representation of the short RET isoform (RET9). HSCR mutations and additional mutations tested in this study are depicted. SP, signal peptide; TM, transmembrane region; TK, tyrosine kinase domain.
To study the RET-Shc interaction, we used the yeast two-hybrid assay (35). The hybrid bait consisted of the whole intracytoplasmic region of RET9 fused to the dimerization and DNA-binding domain of LexA. Since the LexA moiety promotes RET dimerization and subsequent RET trans-autophosphorylation, this system allowed us to investigate protein-protein interactions which depend on tyrosine phosphorylation of RET. The LexA-RET bait plasmid was co-transformed into yeast strain L40 (36) with plasmids expressing various domains of Shc fused to the GAL4 activation domain (GAD) (37). Transformants were plated on selective medium, replated and analyzed for activation of a LacZ reporter gene by the colony filter assay for [beta]-galactosidase activity, and for activation of the HIS3 reporter gene by growth on selective medium lacking histidine (37). As shown in Table 1, we observed an interaction between RET and either the Shc PTB or the Shc SH2 domain. We further showed that this binding was dependent on the catalytic activity of RET since a mutation in the ATP binding site (K758A) of the RET kinase domain abolished the RET-Shc binding. Consistent with a previous report (19), Grb2 did not bind directly to RET9, thus demonstrating the specificity of the RET-Shc interaction in this assay. Mutation of Y1062 but not of Y1015 disrupted the interaction between Shc and RET. Furthermore, mutations of either I1057 or N1059 blocked the Shc PTB-RET interaction, whereas mutation of I1065 abolished the Shc SH2-RET binding (Table 1). With regard to HSCR mutations, the [Delta]1059 mutation was found to prevent the interaction with the whole Shc molecule, while mutation L1061P eliminated the interaction with the Shc PTB but not the Shc SH2 domain.
Table 1. Two-hybrid analysis of the RET/Shc interaction
| LexA binding domain fusion | Gal4 AD fusion | |||
| Shc | Grb2 | |||
| Full length | PTB | SH2 | ||
| RET WT | + | + | + | - |
| RET K758A | - | ND | ND | - |
| RET Y1015F | + | ND | ND | - |
| RET I1057D | + | - | + | - |
| RET N1059A | + | - | + | - |
| RET [Delta]1059 | - | ND | ND | - |
| RET L1061P | + | - | + | - |
| RET Y1062F | - | ND | ND | - |
| RET I1065D | + | + | - | - |
ND, not done.
Based on the results of the two-hybrid assay, a set of mutations which specifically interfere with the binding of either the Shc PTB or the Shc SH2 were selected and singly introduced into the cDNA coding for a MEN 2A mutant form of RET (RET9-634R; here referred to as RET-634). We and others (38,39) have demonstrated that MEN 2A mutations lead to a ligand-independent constitutive activation of the RET tyrosine kinase. To explore the biological effects of these mutations, NIH3T3 cells were infected with recombinant Babe-Puro retroviruses expressing the RET mutants and a mass population of puromycin resistant cells was selected. RET proteins were immuno- precipitated from lysates of the puromycin resistant cells and samples were subjected to western blot analysis with an anti-RET serum. As shown in Figure 3A, RET proteins were expressed at comparable levels in each cell line and two polypeptides of 150 and 170 kDa were detected, corresponding to an incompletely glycosylated RET precursor and to the mature membrane bound protein, respectively (40) (Fig. 3A). As previously reported, RET-634 displays a marked increase of phosphotyrosine content (Fig. 3B) compared with the RET wild-type (38,39). Furthermore, the different mutations tested in this study did not substantially affect the level of RET phosphorylation as compared with RET-634 (Fig. 3B).
Figure 3. Expression and tyrosine phosphorylation of RET proteins. Comparable amounts of RET proteins were immunoprecipitated from mass-selected NIH3T3 cells stably expressing RET mutants. (A) Immunocomplexes were subjected to western-blot analysis with an anti-RET polyclonal serum. (B) The same immunoblot was stripped and reprobed with an anti-phosphotyrosine monoclonal antibody (4G10).
To examine the phosphotyrosine content of Shc, comparable amounts of the 46, 52 and 66 kDa Shc isoforms (41) were immunoprecipitated from lysates prepared from RET mutant expressing NIH3T3 cells. Western blot analysis with an anti-phosphotyrosine antibody (Fig. 4A) revealed that tyrosine phosphorylation of Shc was significantly increased in cells expressing RET-634 compared with wild-type RET. In accordance with several reports, mutation of Y1062 led to a complete loss of Shc phosphorylation (15-17), while mutation of Y1015, the PLC-[gamma] docking site, did not result in a significant change in Shc phosphorylation (Fig. 4). In cells expressing either RET-634/1059A or RET-634/[Delta]1059, the phosphotyrosine content of Shc was diminished to a level comparable with that observed in cells expressing wild-type RET, while the I1065D mutation did not have a substantial effect (Fig. 4B). In the case of RET-634/1061P, a persistent phosphorylation of Shc was detected, albeit at a significantly reduced level. Quantification by densitometry revealed that the amount of phosphotyrosine in Shc was 4-fold lower in cells expressing RET-634/1061P than in cells expressing RET-634.
A
 |
B
|
Figure 4. Tyrosine phosphorylation of Shc. NIH3T3 cells stably expressing RET mutants were starved overnight (culture medium containing 0.5% FBS) and Shc proteins were immunoprecipitated from whole cell lysates with a rabbit polyclonal anti-Shc antibody. (A and B) Immunoprecipitates were subjected to western-blot analysis using a monoclonal antiphosphotyrosine antibody 4G10 (top); the same immunoblots were reacted with the polyclonal anti-Shc serum (bottom). LacZ, NIH3T3 cells stably expressing [beta]-galactosidase.
We used a `GST pull-down' assay to further explore the interaction between Shc and RET. Recombinant GST-Shc, GST-Shc PTB and GST-Shc SH2 fusion proteins were produced in Escherichia coli, coupled to glutathione-Sepharose beads and used to isolate proteins from NIH3T3 cells expressing different RET mutants. As shown in Figure 4, RET-634 binds to GST-Shc and to GST-PTB. It is worth noting that the interaction between RET-634 and GST-SH2 could not be detected with the anti-RET antibody (Fig. 5) whereas a weak signal corresponding to the molecular weight of RET was detected with an anti-phosphotyrosine antibody (data not shown). These results suggest that RET9 binds to Shc SH2 with a much weaker affinity than to Shc PTB, a finding which is in agreement with the report of Arighi et al. (16). Mutation of Y1062, but not of Y1015, is able to specifically block the interaction of Shc with RET. Consistent with the two-hybrid experiments, the N1059A RET mutant was unable to bind to the Shc PTB. We next assessed the properties of RET-634 carrying the HSCR mutations. Cell lysates containing similar amounts of RET proteins (Fig. 6A) were incubated with GST-Shc fusion proteins and subsequently subjected to western blot analysis with an anti-phosphotyrosine antibody. Figure 6B shows that neither the [Delta]1059 nor the L1061P RET mutants exhibited significant binding to Shc PTB in this assay. Similar results were obtained with Shc SH2, showing that these two mutations hindered the interaction with Shc in the `GST pull-down' assay (data not shown).
Figure 5. In vitro interaction between RET mutants and Shc. Cell lysates prepared from NIH3T3 stably expressing RET mutants were subjected to `GST pull-down assay' using beads covered with GST-Shc fusion proteins (GST-Shc, GST-PTB and GST-SH2). Proteins bound to GST fusion proteins were then analysed by western blot using the monoclonal antiphosphotyrosine antibody 4G10 (A) (top). The same immunoblots were stripped and reprobed with an anti-RET polyclonal serum (B) (bottom).
A
 |
B
|
Figure 6. Impaired interaction between Shc and HSCR RET mutants. Cell lysates from NIH3T3 stably expressing comparable amounts of mutant RET proteins (A) (carrying mutation [Delta]1059, L1061P or Y1062F) were analyzed by `GST pull-down assay' using the GST-PTB Shc fusion. (B) Samples were then subjected to western blot analysis using the monoclonal antiphosphotyrosine antibody 4G10.
In order to quantitate the effects of both HSCR mutations on the function of RET, we used a system developed in the rat pheochromocytoma PC12 cell line (42). We demonstrated previously (43,44) that oncogenic forms of RET activate the transcription of immediate early and delayed response genes such as NGFI-A and vgf, respectively. In this study, we assayed the ability of various RET mutants to stimulate the transcription of a chloramphenicol acetyl transferase (CAT) reporter gene driven by the NGFI-A promoter. As reported before (44), cotransfection of the NGFI-A-CAT reporter plasmid with the pBabe-Puro retroviral vector containing RET-634 resulted in marked increase of CAT activity as compared with vector controls (Fig. 7) or to RET wild-type (data not shown). However, expression of either RET-634/1062F or RET-634/[Delta]1059 led to a negligible activation of CAT activity, while expression of RET-634/1061P induced a weak increase in CAT activity (~2.5-fold lower than the level achieved with RET-634) (Fig. 7B). Taken together, this quantitative analysis indicates that mutation [Delta]1059 blocks the engagement of signalling cascades downstream of RET while mutation L1061P results in a partial loss of RET function. The effects of both HSCR mutations could be due either to the disruption of Shc binding or alternatively might involve other signalling pathways which do not involve Shc. To examine this question, we used two truncated mutant forms of Shc comprising either the PTB or the SH2 domain. Since these Shc mutants contain only the protein module implicated in the recognition of phosphotyrosine and lack the two docking sites for Grb2, they should function as Shc dominant-interfering mutants. In accordance with this idea, cotransfection of each Shc mutant construct with both RET-634 and NGF-I-A reporter plasmids resulted in a significant decrease of CAT activity (Fig. 8), thus indicating that the signal relayed by RET and converging on the NGF-I-A promoter requires Shc. Interestingly, regardless of the quantity of plasmid used, the effect of PTB was more pronounced than SH2; i.e. with 1 µg of plasmid the expression of PTB led to a 3-fold decrease of CAT activity versus 1.5-fold with SH2 (Fig. 8). Furthermore, these effects were specific to the dominant-negative forms of Shc proteins since expression of Shc wild-type did not result in a similar inhibition of CAT activity (data not shown). Collectively, these data suggest that both HSCR mutations impair RET signalling, partly through the disruption of Shc binding. In agreement with these findings, we found that RET-634/[Delta]1059 did not exhibit transforming activity, whereas RET-634/1061P showed a persistent but somewhat reduced oncogenic potential (data not shown).
Figure 7. Induction of the NGFI-A promoter by various mutant RET proteins. PC12 cells were transfected with 2 µg of NGFI-A-CAT and 3 µg of RET constructs. At 60 h post-transfection total proteins were isolated and promoter induction was determined by a CAT assay. Representative CAT assays (A) and the bar charts of the relative induction (B) are shown. The results represent the averages of three separate experiments performed in duplicate.
Figure 8. Dominant-negative forms of Shc block RET-634-mediated activation of the NGF-I-A promoter. Two vectors expressing either the PTB or the SH2 domain of Shc were used in this assay. PC12 cells were cotransfected with 2 µg of NGF-I-A reporter plasmid together with 3 µg of RET constructs and either 1 or 3 µg of vectors expressing dominant-negative forms of Shc. At 60 h post-transfection total proteins were isolated and promoter induction was determined by a CAT assay. Representative CAT assays (A) and the bar charts of the relative induction (B) are shown. The results represent the averages of three separate experiments performed in duplicate.
DISCUSSION
Previous reports have offered compelling evidence that HSCR missense mutations disrupt RET function through different molecular mechanisms (31-34). Mutations located in the extracytoplasmic region inhibit the maturation of the RET protein and prevent its translocation at the plasma membrane, whereas mutations within the tyrosine kinase domain either weaken or eliminate the catalytic activity (31-34). Also, mutations in the extracellular or tyrosine kinase domains might cause loss of RET function through a dominant negative mechanism (31,45). The results of the present study support the existence of a novel mechanism of RET inactivation due to HSCR mutations which specifically hamper the binding of transduction molecules and consequently alter RET-mediated signalling. We demonstrate that two distinct HSCR mutations ([Delta]1059, L1061P) interfere with the binding of the Shc adapter to RET. Mutation [Delta]1059, blocks the RET-Shc interaction and abolishes the induction of early genes activated on neuronal differentiation of PC12 cells. Mutation L1061P partially inhibits the binding of Shc and reduces, but does not eliminate, the biological activity of RET.
Systematic analyses of the amino acid sequence that surrounds the Shc phosphotyrosine interaction site have demonstrated that the sequence, [Psi]-(D/E)-N-(X/P)-X-pY (where [Psi] is a hydrophobic amino acid) is the preferential binding site for the PTB domain (25-28). Consistent with these results, our findings support the notion that a hydrophobic residue at position -5 and an asparagine at position -3 are required for the fixation of the Shc PTB to RET and that a hydrophobic residue at position +3 dictates the specificity of recognition of the Shc SH2. Our data indicate that deletion of the codon that specifies the asparagine 1059 in RET or mutation of this asparagine to an alanine abrogate the ability of the Shc PTB domain to link to RET. The two hybrid experiments, however, suggest that the deletion of codon 1059, and not the N1059A mutation, blocks the fixation of Shc not only through the PTB domain but also through the SH2 domain, thus raising the possibility that [Delta]1059 would in fact preclude an efficient recognition and autophosphorylation of Y1062 by RET. Substitution of proline for leucine at position -1 (L1061P) also prevents the ability of Shc to contact RET through its PTB domain. This result was unexpected as previous reports did not point to an important role for the amino acid at the -1 position in the binding of PTB (25-27). However, a study aimed at determining the amino acids required for high affinity binding of Shc to the polyoma virus middle T antigen (mT) did reveal that the threonine at position -1 is critically involved in the Shc PTB-mT interaction which depends on the formation of a [beta]-turn N-terminal to the phosphotyrosine (28). It is conceivable that the bending properties of the proline residue prevent the polypeptide backbone to adopt a conformation optimal for the binding of the PTB domain. Therefore, our results suggest that Leu1061 plays a pivotal role in determining conformation of the PTB docking site in the context of the RET sequence and that replacement by a Pro unfolds this motif and obviates the linking of Shc.
In agreement with the biochemical data, quantitative assay with the NGF-I-A/CAT reporter gene performed in PC12 cells revealed that mutation [Delta]1059 completely inhibits the activating ability of RET-634R. In contrast, the RET mutant L1061P demonstrated a reduced, but persistent, biological activity in this assay. These findings fit with our results showing that Shc is phosphorylated on tyrosine in cells expressing RET-634/1061P, but at a reduced level compared with RET-634R. This can be interpreted as further evidence that mutation L1061P does not lead to a complete loss of RET function. Our results indicate that the presence of a proline at position 1061 is not compatible with efficient binding to the Shc PTB domain, suggesting that Shc might bind to pY1062 via its SH2 domain. Alternatively, it is possible that a residual binding of Shc PTB to RET-634/1061P, not detectable in our biochemical assays, still occurs in vivo. In conclusion, these biological data demonstrate that both HSCR mutations studied in this article disable the ability of RET to transmit a downstream signal.
Recently, it has been reported that Y1062 on RET is a binding site for Enigma, and this interaction is mediated through the second LIM domain of Enigma (46). In contrast to Shc, however, the binding of Enigma to RET does not depend on phosphorylation of Y1062 (46). Further studies have yielded evidence that both Enigma and Shc are required for the mitogenic activity of RET/PTC, a rearranged oncogenic form of RET that has a deleted transmembrane domain. On this basis a model has been proposed where Enigma recruits RET/PTC to the plasma membrane via its PDZ domain (47). Although the results support a role for Enigma in the transforming ability of RET/PTC, the function of this protein in the context of RET remains largely unknown. We cannot exclude that the HSCR mutations analyzed in this study might also impede the binding of Enigma and contribute to some of the observed biological effects, an issue which would deserve to be addressed in future studies. Recently, two of us (G.D.V. and M.S.) have demonstrated that FRS2, a myristylated protein that displays a PTB domain and recruits both Grb2 and the tyrosine phosphatase SHP2 to the FGF receptor, binds to Y1062 of RET (R.M. Melillo, manuscript in preparation). It was next found that both HSCR mutations described in this study prevent the fixation of FRS2 to RET. Since FRS2, like Shc, links RTK to the Ras/MAP kinase signalling pathway it is now important to sort out the respective contribution of these proteins to the RET-mediated GDNF response. However, we have shown that dominant-negative forms of Shc inhibit RET-634 signalling, thereby suggesting that the RET/Shc interaction is a crucial event for the transmission of an intracellular signal. Collectively, these data establish that Y1062 is a multifunctional docking site that plays an essential role in vivo and mutations which interfere with its coupling function compromise the capacity of RET to transduce a neurotrophic signal in enteric neurons.
In a previous study, we predicted that HSCR could be ascribed to mutations which specifically impair the interaction of RET with its transduction effectors (34). Consistent with this idea, we now confirm the existence of a novel class of RET mutations causally involved in the pathogenesis of HSCR. Interestingly, Lorenzo et al. (17) recently reported that an HSCR mutation specific to the RET51 isoform (M1064T) decreases the affinity for the Shc PTB domain, thus confirming both the validity of the proposed classification of HSCR mutations and the essential docking role of Y1062. The results of the present study provide a coherent explanation for the phenotype of individuals carrying the mutation at codon 1061. As demonstrated, mutation L1061P results in a partial loss of RET function which likely accounts for the lack of lethality in patients homozygous for this allele but having a severe aganglionosis. A patient with total colonic aganglionosis and involvement of the small intestine was shown to be homozygous for a missense mutation (R313Q) in the RET cadherin domain (48); this observation suggests that this mutation might also incompletely disrupt RET function. Finally, these data strongly indicate that Y1062 is a multifunctional docking site for several effectors that trigger downstream signals which are critical for the GDNF survival response of enteric neurons.
MATERIALS AND METHODS
Patients and families
A panel of 95 HSCR patients including 78 sporadic cases and 17 familial cases were recruited in The Netherlands. HSCR was diagnosed based on histological and histochemical examination of the aganglionic tract. All the exons of the RET gene were screened either by constant gradient gel electrophoresis (CDGE) or denaturing gradient gel electrophoresis (DGGE) (R.M.W. Hofstra et al., submitted for publication).
Construction of vectors
The construction of the retroviral vectors expressing RET9 and the MEN 2A mutant form of RET9 (RET-634) was previously described (49). Briefly, in these constructs, the expression of the cDNAs encoding RET is driven by the Moloney Murine Leukaemia Virus 5[prime] long terminal repeat (LTR) promoter. The different mutations were introduced using the USE Mutagenesis kit (Pharmacia Biotech, Orsay, France) according to the manufacturer's instructions. The following oligonucleotides were used for mutagenesis:
K758A, 5[prime]-ACCACGGTGGCCGTGGCGATGCTGAAAGAGAAC-3[prime];
Y1015F, 5[prime]-GTTAAGAGGAGAGACTTCTTGGACCTTGCG-3[prime];
I1057D, 5[prime]-CTCCCTTCCACATGGGATGAAAACAAACTCTATG-3[prime];
N1059A, 5[prime]-CCACATGGATTGAAGCCAAACTCTATGGTAG-3[prime];
[Delta]1059, 5[prime]-CCACATGGATTGAAAAACTCTATGGTAGAATTTCC-3[prime];
L1061P, 5[prime]-GATTGAAAACAAACCCTATGGTAGAATTTCC-3[prime];
Y1062F, 5[prime]-GAAAACAGAATTTCCCAT-3[prime];
I1065D, 5[prime]-CAAACTCTATGGTAGAGATTCCCATGCATTTAC-3[prime].
All mutations were verified by DNA sequence analysis.
For the yeast two-hybrid analysis, the cDNA sequence coding for the complete intracytoplasmic domain of RET 9 (amino acids 658-1072) was inserted into the plasmid pLex10 (50) in-frame with the DNA-binding domain of LexA. The plasmid pLex10 contains a Trp+ selection marker. The GAD-Shc hybrid constructs (pACTII vector; kindly provided by S. Elledge, Houston, TX) have been described elsewhere (37). pACTII vector has a Leu+ selection marker.
The Shc dominant-negative mutants were constructed by PCR using the cDNA coding for the human p46 Shc isoform as a template (kindly provided by Emmanuel Fournier, INSERM U119, Marseille, France) and the following two sets of primers:
PTB-f, 5[prime]-CTAGAGATCTCCACCATGGAGCAGAAGCTGATCAGCG- AGGAGGACCTGGGCGGCAATTCGAAGTACTCCGGACCCGGG-3[prime];
PTB-r, 5[prime]-CTAGGAATTCACTCATCCCATGCTGAGCCATC-3[prime];
SH2-f, 5[prime]-CTAGGAATTCCACCATGGAGCAGAAGCTGACAGCGAGG- AGGACCTGGGCGGCCCCCAGTCGGTGTCCATGGCT-3[prime];
SH2r, 5[prime]-GCCACGGGATCCTCACAGTTTCCGCTCCACAG GT-3[prime].
These PCR products were digested with BglII and EcoRI for PTB and EcoRI and BamHI for SH2, cloned in pBluescript SK- (Stratagene, La Jolla, CA), completely sequenced and subcloned in the mammalian expression vector pCI-neo (Promega, Charbonnières, France). The PTB and the SH2 constructs contain the sequence coding for amino acids 2-184 and 219-328 of Shc, respectively. Each protein was epitope-tagged at the N-terminus with a myc epitope which is recognized by the monoclonal antibody 9E10. Similarly, the cDNA coding for the p46 Shc isoform was cloned in pCI-neo and the protein was myc-tagged at the N-terminus. On transient transfections of these expressing vectors in 293T, Shc tagged-proteins which specifically reacted with the 9E10 antibody were detected at the right size, thus ensuring accurate expression of these vectors (data not shown).
Cell culture and antibodies
NIH3T3 cells were grown in Dulbecco's modified Eagle's medium, 10% fetal calf serum and antibiotics. Mass-selected NIH3T3 expressing RET proteins were produced as previously described(49). The two anti-RET rabbit polyclonal sera used in this study were previously described (38,49). Monoclonal anti-phosphotyrosine 4G10 was purchased from UBI (New York, NY) and rabbit polyclonal anti-Shc were purchased from Transduction Laboratory (KI).
Production of recombinant retrovirus
Ecotropic retroviruses were obtained by transient transfection of the retroviral vectors into the BOSC 23 packaging cell line according to the protocol described by Pear et al. (51). Forty-eight hours after transfection, the supernatants were harvested and filtered on 0.45 µm pore size filters. Titre of retroviruses was determined by infecting NIH3T3 cells in the presence of 8 µg/ml of polybrene. Forty-eight hours after infection, the cells were split 1:10 into medium containing 2 µg/ml puromycin. Puromycin-resistant colonies were stained with Giemsa and counted on day 15. Retroviral titres were routinely comprised of between 5 × 105 and 1 × 106 colony-forming units (c.f.u.)/ml.
Protein analyses
RET protein immunoprecipitation and immunoblotting experiments were performed as described in Rossel et al. (49). Mass-selected NIH3T3 cells were lysed in lysis buffer (20 mM Tris-HCl pH 7.8, 150 mM NaCl, 1% Nonidet P-40, 2 mM EDTA), supplemented with a cocktail of proteases and phosphatase inhibitors (1 mM phenylmethylsulphonyl fluoride, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 10 µg/ml pepstatin, 1 mM sodium orthovanadate). Lysates were clarified by centrifugation and protein concentration was estimated by a modified Bradford analysis (Bio-Rad, Ivry sur Seine, France). Comparable amounts of the different RET mutant proteins were immunoprecipitated for 3 h at 4°C with one of the two different anti-RET rabbit polyclonal sera previously described (38,49). Immune complexes were recovered by absorption on protein A coupled to sepharose beads (protein A-Sepharose CL4B; Pharmacia Biotech), washed in lysis buffer, resuspended in Laemmli buffer (20 mM Tris-HCl pH 6.8, 2% SDS, 20% glycerol, 1.4 M [beta]-mercaptoethanol, 20 µg/ml bromophenol blue), then subjected to SDS-7% PAGE and western blotting as previously described (49). Immunoblots were first probed with the monoclonal antiphosphotyrosine antibody 4G10 (UBI), incubated with an anti-mouse conjugated to horse-radish peroxidase (HRP) (Amersham, Les Ulis, France) and revealed with the Amersham enhanced chemiluminescence (ECL) system, then stripped as described (49), reprobed with the anti-RET serum and revealed again using an anti-rabbit-HRP conjugate (Amersham) and the ECL system.
Shc proteins were immunoprecipitated with rabbit polyclonal anti-Shc (Transduction Laboratory) according to the manufacturer's instructions. Immunoprecipitates were resolved by electrophoresis on SDS-polyacrylamide gels and proteins were transferred to polyvinylidenedifluoride (PVDF) membranes (Immobilon P; Millipore, St Quentin Yvelines, France). For analysis of Shc phosphorylation, immunoblots were incubated with anti-phosphotyrosine 4G10 (UBI) then with protein A biotin (Boehringer Mannheim, Meylan, France) and Streptavidin peroxydase (Sigma, L'Isle d'Abeau, France) as detailed by Chappuis-Flament et al. (44). Blots were revealed with the ECL system (Amersham).
GST-fusion affinity precipitation
Bacterially expressed fusion proteins were produced by standard protocols. In vitro interactions were performed using mass-selected NIH3T3 cells expressing RET proteins. Cleared cell lysates (500 µg) were incubated for 2 h at 4°C with 2 µg of GST-fusion protein bound to glutathione-Sepharose beads in a total volume of 500 µl. The beads were washed three times with lysis buffer, resuspended in Laemmli buffer and the bound proteins were fractionated on an SDS-8% polyacrylamide gel. Proteins were transferred to PVDF membranes and probed with either anti-RET or anti-phosphotyrosine (4G10) antibodies.
Yeast two-hybrid assay
The yeast strain L40 (MAT[alpha], tyr1, leu2, his3, LYS2::lexA-His3, URA 3::lexA-lacZ) was previously described (36). Growth conditions and maintenance of the yeast strain L40 were performed essentially as described(37). L40 was transformed simultaneously with the two indicated hybrid plasmids by the improved lithium acetate method (52). Cotransformants were selected on Trp-Leu- plates. Colonies of each transformation were tested for histidine prototrophy after plating on Trp-Leu-His- medium. Synthetic defined dropout yeast media lacking the appropriate amino acids were obtained from BIO 101 (La Jolla, CA).
CAT assay
PC12 cells were grown in RPMI 1640 medium supplemented with 10% horse serum and 5% fetal calf serum (Gibco BRL, Cergy Pontoise, France). For transient transfection assays, cells were plated at 3 × 105 cells in 60 mm diameter tissue culture dishes 24-36 h prior to transfection. Transfection was performed using the lipofectin reagent following the manufacturer's instructions (Gibco BRL). All transfections were carried out with 2 µg of reporter plasmid (pNGFI-A-CAT, pBLCAT2 or RSVCAT) together with 3 µg of the various RET constructs. For the assay of the dominant-negative Shc mutants, the same quantity of reporter and RET plasmids were transfected together with either 1 or 3 µg of Shc expression vectors. The same DNA concentration was reached by adding various amounts of the LTR control vector. Cell extracts were prepared 60 h after transfection and CAT activity was analyzed by thin-layer chromatography (TLC) with 95% chloroform-5% methanol, as previously described (42). Each experimental point was cut from the TLC plate and counted. For each experiment, the percentage of conversion to the acetylated form of chloramphenicol 14C was then calculated. The results of at least three experiments, made in duplicate, were plotted on an arbitrary scale as relative promoter induction. pNGFI-A-CAT induction by RET was specific, since no stimulation was observed when either pBLCAT2 or RSVCAT control plasmid was used as reporter (data not shown).
ACKNOWLEDGEMENTS
We gratefully acknowledge Saskia M. Maas, Liesbeth Breslau-Siderius and Hester Kroes for providing information on the pedigrees reported in this report. We wouldlike to thank Andrea Pasini for his help with retrovirus, John Copeland for critical reading of the manuscript, Laurence Fournier for her excellent technical help, and Mrs Monique Billaud for the graphic work.This work was supported by the Centre National de la Recherche Scientifique, the Association de Recherche sur le Cancer (ARC), the Fondation de France, the Ligue Nationale contre le Cancer, the Region Rhône-Alpes, the Associazione Italiana per la Ricerca sul Cancro and by the EC Projects BMH4-97-2157 and BMH4-CT96-0814. During this work, O.G. and C.B. have been the recipients of a fellowship from the Ligue Nationale contre le Cancer, Comité de l'Indre and Comité de la Drôme, respectively.
REFERENCES
+These authors contributed equally to this work
§Present address: Transcription Laboratory, Imperial Cancer Research Fund, 44 Lincoln's Inn Fields, London WC2A 3PX, UK
¶To whom correspondence should be addressed. Tel: +33 4 78 77 72 13; Fax: +33 4 78 77 72 20; Email: billaud{at}univ-lyon1.fr
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. P. Knowles, J. Murray-Rust, S. Kjaer, R. P. Scott, S. Hanrahan, M. Santoro, C. F. Ibanez, and N. Q. McDonald Structure and Chemical Inhibition of the RET Tyrosine Kinase Domain J. Biol. Chem., November 3, 2006; 281(44): 33577 - 33587. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. C.H. Lui, T. Y.Y. Leon, M.-M. Garcia-Barcelo, R. W. Ganster, B. L.S. Chen, J. M. Hutson, and P. K.H. Tam Novel RET Mutation Produces a Truncated RET Receptor Lacking the Intracellular Signaling Domain in a 3-Generation Family with Hirschsprung Disease Clin. Chem., August 1, 2005; 51(8): 1552 - 1554. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Jain, C. K. Naughton, M. Yang, A. Strickland, K. Vij, M. Encinas, J. Golden, A. Gupta, R. Heuckeroth, E. M. Johnson Jr, et al. Mice expressing a dominant-negative Ret mutation phenocopy human Hirschsprung disease and delineate a direct role of Ret in spermatogenesis Development, November 1, 2004; 131(21): 5503 - 5513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Jijiwa, T. Fukuda, K. Kawai, A. Nakamura, K. Kurokawa, Y. Murakumo, M. Ichihara, and M. Takahashi A Targeting Mutation of Tyrosine 1062 in Ret Causes a Marked Decrease of Enteric Neurons and Renal Hypoplasia Mol. Cell. Biol., September 15, 2004; 24(18): 8026 - 8036. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Zambrano, D. Gianni, P. Bruni, F. Passaro, F. Telese, and T. Russo Fe65 Is Not Involved in the Platelet-derived Growth Factor-induced Processing of Alzheimer's Amyloid Precursor Protein, Which Activates Its Caspase-directed Cleavage J. Biol. Chem., April 16, 2004; 279(16): 16161 - 16169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kjaer and C. F. Ibanez Intrinsic susceptibility to misfolding of a hot-spot for Hirschsprung disease mutations in the ectodomain of RET Hum. Mol. Genet., September 1, 2003; 12(17): 2133 - 2144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Pelicci, F. Troglio, A. Bodini, R. M. Melillo, V. Pettirossi, L. Coda, A. De Giuseppe, M. Santoro, and P. G. Pelicci The Neuron-Specific Rai (ShcC) Adaptor Protein Inhibits Apoptosis by Coupling Ret to the Phosphatidylinositol 3-Kinase/Akt Signaling Pathway Mol. Cell. Biol., October 15, 2002; 22(20): 7351 - 7363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Murakami, Y. Yamamura, Y. Shimono, K. Kawai, K. Kurokawa, and M. Takahashi Role of Dok1 in Cell Signaling Mediated by RET Tyrosine Kinase J. Biol. Chem., August 30, 2002; 277(36): 32781 - 32790. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Mograbi, R. Bocciardi, I. Bourget, T. Juhel, D. Farahi-Far, G. Romeo, I. Ceccherini, and B. Rossi The Sensitivity of Activated Cys Ret Mutants to Glial Cell Line-Derived Neurotrophic Factor Is Mandatory To Rescue Neuroectodermic Cells from Apoptosis Mol. Cell. Biol., October 15, 2001; 21(20): 6719 - 6730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Melillo, M. Santoro, S.-H. Ong, M. Billaud, A. Fusco, Y. R. Hadari, J. Schlessinger, and I. Lax Docking Protein FRS2 Links the Protein Tyrosine Kinase RET and Its Oncogenic Forms with the Mitogen-Activated Protein Kinase Signaling Cascade Mol. Cell. Biol., July 1, 2001; 21(13): 4177 - 4187. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Gabriella De Vita, R. M. Melillo, F. Carlomagno, R. Visconti, M. D. Castellone, A. Bellacosa, M. Billaud, A. Fusco, P. N. Tsichlis, and M. Santoro Tyrosine 1062 of RET-MEN2A Mediates Activation of Akt (Protein Kinase B) and Mitogen-activated Protein Kinase Pathways Leading to PC12 Cell Survival Cancer Res., July 1, 2000; 60(14): 3727 - 3731. [Abstract] [Full Text]
|
|
|
|
|
|
B. Mograbi, R. Bocciardi, I. Bourget, R. Busca, N. Rochet, D. Farahi-Far, T. Juhel, and B. Rossi Glial Cell Line-derived Neurotrophic Factor-stimulated Phosphatidylinositol 3-Kinase and Akt Activities Exert Opposing Effects on the ERK Pathway. IMPORTANCE FOR THE RESCUE OF NEUROECTODERMIC CELLS J. Biol. Chem., November 21, 2001; 276(48): 45307 - 45319. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||