Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

The p75 Neurotrophin Receptor Interacts with Multiple MAGE Proteins

The p75 Neurotrophin Receptor Interacts with Multiple MAGE Proteins THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 277, No. 51, Issue of December 20, pp. 49101–49104, 2002 Accelerated Publication © 2002 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. particular receptor is a difficult bait in conventional yeast The p75 Neurotrophin Receptor two-hybrid. Nonetheless seven candidate interactors have been Interacts with Multiple MAGE reported so far for p75 (3, 5). Most of these candidate interac- tors lack known catalytic domains, suggesting that they must Proteins* recruit additional binding partners to form a functional signal- ing complex. There are no general sequence or functional ho- Received for publication, September 19, 2002, mologies between the described p75 interactors, and the phys- and in revised form, October 21, 2002 iological significance of most of them remains to be established. Published, JBC Papers in Press, October 31, 2002, DOI 10.1074/jbc.C200533200 To identify biologically relevant interactors for p75, we used the Ras rescue system (RRS) for protein interaction trapping Marianna Tcherpakov‡, Francisca C. Bronfman‡, in yeast (6). RRS selects for protein interactions in the mem- Silvestro G. Conticello‡, Anna Vaskovsky‡, Zehava Levy‡, Michio Niinobe§, brane proximal region of the cytoplasm, and there is no re- Kazuaki Yoshikawa§, Ernest Arenas , quirement for nuclear or other translocations of the binding and Mike Fainzilber‡ partners. The system thus provides an appropriate screening From the ‡Molecular Neurobiology Group, Department method for the membrane-proximal binding of a receptor in- of Biological Chemistry, Weizmann Institute of Science, tracellular domain with its primary interactors. Our RRS 76100 Rehovot, Israel, the §Division of Regulation of screens identified two MAGE family proteins as novel p75 Macromolecular Functions, Institute for Protein interactors and highlight type II MAGE proteins as a potential Research, Osaka University, Osaka 565-0871, Japan, ¶ interactor family for type II death domain proteins. and the Department of Medical Biochemistry and Biophysics, Karolinska Institute, S-17177 Stockholm, Sweden MATERIALS AND METHODS RRS Screens, cDNA Cloning, and Sequence Analysis—Plasmids, The p75 neurotrophin receptor has been implicated in yeast strains, and screening protocols for RRS were as described (6). diverse aspects of neurotrophin signaling, but the mech- 274 342 343 416 Linker (K RWN to SLPL ) and death domain (T KRE to ESLC ) anisms by which its effects are mediated are not well regions of p75-ICD were generated by PCR from rat p75. Synthetic understood. Here we identify two MAGE proteins, nec- oligonucleotides were designed to encode the nine-residue (SES- din and MAGE-H1, as interactors for the intracellular TATSPV) tail domain of p75-ICD. A pADH death domain bait was used domain of p75 and show that the interaction is enhanced to screen a mouse embryonic head cDNA library in pMyr. Alignments of MAGE sequences and phylogenetic trees were constructed using by ligand stimulation. PC12 cells transfected with nec- ClustalX. din or MAGE-H1 exhibit accelerated differentiation in Expression Analysis—Rat tissues for Western blot analysis were response to nerve growth factor. Expression of these flash-frozen in liquid nitrogen, macerated to a fine powder, and solubi- two MAGE proteins is predominantly cytoplasmic in lized in 10 mM Tris, pH 8.0, 150 mM NaCl, 10% glycerol, 1 mM PMSF, 1 PC12 cells, and necdin was found to be capable of ho- mM orthovanadate and proteinase inhibitors (Merck) containing 1% modimerization, suggesting that it may act as a cyto- Nonidet P-40. Extracts were separated on SDS-PAGE, blotted, and plasmic adaptor to recruit a signaling complex to p75. probed with anti-necdin antibodies. These findings indicate that diverse MAGE family mem- Cell Culture, Transfections, and Generation of Cell Lines—PC12 cells bers can interact with the p75 receptor and highlight were maintained in Dulbecco’s modified Eagle’s medium containing 6% fetal calf serum and 6% horse serum. For induction of neuronal differ- type II MAGE proteins as a potential family of interac- 5 2 entiation, the cells were seeded at a density of 10 –15  10 cells/cm tors for signaling proteins containing type II death and differentiated for 2– 4 days with NGF (5–50 ng/ml). Transient domains. transfection of COS cells was with DEAE-dextran. Transient and stable transfections of PC12 cells lines were carried out by electroporation with 80 g of DNA per 10 cells in 0.4-cm cuvettes on an ECM-300 Neuronal responses to activation of the p75 neurotrophin (BTX) set to 320 V, 6-ms pulse length. Stably transfected PC12 were receptor range from enhanced outgrowth to increased cell selected under 0.5 mg/ml Geneticin for 4 weeks until resistant colo- nies appeared. death, and p75 null mice exhibit a plethora of defects in both Immunoprecipitation, Western Blots, and Immunofluorescence— neuronal and non-neuronal systems (1, 2). However, the pri- Cells were lysed in 10 mM Tris, pH8.0, 150 mM NaCl, 10% glycerol, 1 mM mary interactors and signaling mechanisms activated by p75 PMSF, 1 mM orthovanadate and proteinase inhibitors (Merck) contain- are not well understood (3). Although p75 belongs to the TNF ing either 0.7% CHAPS or 1% Nonidet P-40. Lysates were spun at receptor superfamily, its intracellular type II death domain 10,000  g for 5 min and then precleared with 30 l of a 50% suspension does not self-aggregate (4) and does not interact with the major of protein-G-Sepharose beads (Amersham Biosciences) before incuba- tion with primary antibody. Immunoprecipitates were eluted with 80 l binding partners of type I death domains. Identification of of 1 M guanidine HCl in 50 mM Tris, pH 8.0, before running on SDS- specific signaling partners for p75 has been frustrating, as this PAGE. Primary antibodies for precipitation were as follows: anti-p75 MC192 (from Alomone Laboratories), anti-p75 REX (7), monoclonal * This work was supported by grants from the Israel Science Foun- anti-HA (from Roche Molecular Biochemicals). Antibodies and dilutions dation (647/01) and the European Union Fifth Framework Program for Western blots were as follows: anti-necdin polyclonal NC243 (8) at (QLRT-1999-573) (to M. F.) and by a short term European Molecular 1:5000, anti-necdin polyclonal MNF (9) at 1:1000, anti-HA polyclonal Biology fellowship (to M. T.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 The abbreviations used are: RRS, Ras rescue system; PMSF, phen- U.S.C. Section 1734 solely to indicate this fact. ylmethylsulfonyl fluoride; NGF, nerve growth factor; CHAPS, 3-[(3- Incumbent of the Daniel Koshland Sr. Career Development Chair at cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; HA, he- the Weizmann Institute of Science. To whom correspondence should be magglutinin; PBS, phosphate-buffered saline; DRG, dorsal root ganglia; addressed. Tel.: 972-8-934-4266; Fax: 972-8-934-4112; E-mail: mike. BDNF, brain-derived neurotrophic factor; MHD, MAGE homology fainzilber@weizmann.ac.il. domain. This paper is available on line at http://www.jbc.org 49101 This is an Open Access article under the CC BY license. 49102 Multiple MAGE Interactors for p75 FIG.1. Necdin, a MAGE family protein, interacts with p75. A, myristoylated linker and death domain regions of p75 interact with necdin-Ras to rescue transformed yeast grown at 36 °C, whereas no interaction is observed with the tail domain. B, phylogenetic tree show- ing relationships between human type II MAGE genes; * denotes those shown to interact with p75. C, Western blot analysis of necdin expres- sion in rat tissues. 250 g of tissue protein was loaded in each lane, and the blot was probed with the anti-necdin MNF antibody. Equal loading was verified by stripping and reprobing the blot with anti-tubulin antibody (not shown). FIG.2. Association of necdin and MAGE-H1 with p75 in mam- (Santa Cruz) at 1:200, anti-p75 polyclonal 9651 at 1:1000, anti-phospho- malian cells. A, interaction of p75 with MAGE proteins transiently cdc2 polyclonal (R & D Systems) at 5 g/ml, and anti-TrkA RTA (10) at expressed in COS cells. P75 was co-transfected with necdin or MAGE- 1:1000. Cells were fixed for immunofluorescence microscopy for 20 min H1-HA. As a control MAGE proteins were co-transfected with empty with ice-cold 3% paraformaldehyde in PBS at room temperature, vector. Immunoblotting of cell lysates shows equal expression of MAGE quenched in 0.1 M glycine PBS for 15 min, and permeabilized for 10 min proteins with or without p75. P75 was immunoprecipitated with the MC192 antibody and blots were tested for the presence of MAGE in PBS, 5% donkey serum, 1 mg/ml bovine serum albumin, 0.2% sapo- proteins with anti-necdin and anti-HA antibodies. MAGE proteins were nin. Permeabilized cells were incubated with anti-HA (1:4000) for2hat immunoprecipitated only in the presence of p75. B, interaction of p75 room temperature followed by donkey anti-rabbit rhodamine-redX-con- with overexpressed MAGE proteins in transiently transfected PC12 jugated secondary antibody (1:500, Jackson ImmunoResearch) for 1 h cells. PC12 were electroporated with constructs expressing MAGE pro- at room temperature. teins, and co-immunoprecipitations were carried out with the indicated antibodies. An irrelevant antibody was used as control ( in the figure). RESULTS Both necdin and MAGE-H1 were co-precipitated with the p75 antibody, A mouse embryonic head RRS library was screened against and p75 was co-precipitated with anti-MAGE antibodies. C, interaction three baits comprising the death domain, the 69-residue jux- of p75 with necdin is enhanced by NGF or BDNF in a stably transfected PC12 line. A stable PC12 cell line expressing HA-necdin was treated tamembrane “linker,” and the nine-residue C-terminal “tail” of with 100 ng/ml NGF or BDNF for 2 h, followed by immunoprecipitation p75. These screens yielded interacting candidates only for the with the MC192 antibody. Western blotting for necdin reveals increas- death domain bait. One of the clones encoded the C-terminal 81 ing amounts of co-precipitated necdin upon ligand stimulation. D, co- amino acid residues of necdin (11), a MAGE family member immunoprecipitation of necdin with p75 from lysates of freshly dis- sected adult rat DRG. 40 – 45 DRG were used for each lane, and lysates thought to act as a cell cycle regulator. Further yeast co-trans- were precipitated with the anti-p75 monocolonal MC192, followed by fections of necdin with the complete panel of p75-ICD subdo- Western blot with the anti-necdin polyclonal MNF. E, co-immunopre- mains revealed that necdin interacts with both the linker and cipitation of TrkA with p75 from transiently transfected COS cells. death domain segments of p75 (Fig. 1A). The sequence homol- NGF (100 ng/ml) was added to the cells 44 h after transfection, and lysates were processed for co-immunoprecipitation after 4 h incubation. ogies between necdin and NRAGE, which was previously iden- P75 pull-down was with the REX antibody, and TrkA detection was tified by Barker and colleagues as a p75 interactor (12), with the RTA antibody. F, necdin-necdin interaction exemplified by prompted us to examine additional family members. Unfortu- RRS at 36 °C(left panels) and co-immunoprecipitation from transiently nately, a number of MAGE constructs (e.g. NRAGE and MA- transfected COS cells (right panel). GEL2) were found to associate with yeast plasma membrane, thus precluding their use in the RRS system. MAGE-H1/APR-1 Adult rats express robust levels of necdin protein in the hypo- (13) was found to interact robustly with the p75 death domain thalamus and in dorsal root ganglia (DRG) (Fig. 1C), whereas by RRS (data not shown). Thus, at least three type II MAGE necdin levels during development are much lower. proteins (Fig. 1B) are capable of interacting with p75. Tissue Co-immunoprecipitation experiments were then carried out RNA arrays (data not shown) and in silico profiling reveal in both COS and PC12 cells to confirm the p75-necdin and extensive overlaps in the expression profiles of these genes. p75-MAGE-H1 interactions. In accordance with the RRS re- Multiple MAGE Interactors for p75 49103 FIG.3. Necdin and MAGE-H1 accelerate the differentiation of PC12 cells. A, PC12 cells were transiently transfected with necdin, MAGE-H1, or vector control and incubated for 3 days in 50 ng/ml NGF. B, differentiation of two necdin-expressing stable PC12 lines, and two vector-transfected stable lines, in the presence of 50 ng/ml NGF for 2 days. Average  S.E. is shown for a representative experiment. 120 –180 differentiated cells were measured for each point. C, levels of phospho-cdc2 in necdin versus vector-ransfected PC12 cells, after 24 h in 50 ng/ml NGF. Left panel shows a representative experiment, while the right panel shows average  S.E. (n  3). D, immunofluorescence microscopy reveals predominantly cytoplasmic necdin protein in both cycling (upper) and differentiated (lower) PC12-necdin cells. sults, p75 co-precipitated with necdin or MAGE-H1 from trans- known intrinsic catalytic activity and is likely to signal by fected COS cells (Fig. 2A) or PC12 cells (Fig. 2B). In PC12 cell recruiting other molecules to a p75 signaling complex. We lines stably transfected with necdin, application of either NGF therefore conducted additional RRS screens using necdin as or BDNF clearly enhanced the interaction of necdin with p75 bait and identified a number of novel expressed sequence tags (Fig. 2C). Necdin protein levels did not change during the of unknown function as necdin interactors (data not shown). course of this experiment (data not shown). Finally, a p75- Interestingly, one known gene obtained in this screen was necdin interaction at endogenous levels of expression was dem- necdin itself (Fig. 2F, left). Homo-oligomerization of necdin was onstrated by co-immunoprecipitation from lysates of freshly confirmed by co-immunoprecipitation of differentially tagged dissected adult rat DRG (Fig. 2D). necdin proteins from transfected COS cells (Fig. 2F, right). NRAGE was previously found to affect p75-trk association PC12 cells are a well established model for NGF-induced (12), and we therefore examined whether necdin had similar differentiation and do not express endogenous necdin (14). We effects in transiently transfected COS cells stimulated with therefore examined the influence of transient transfection of NGF. TrkA co-precipitated together with p75 in the absence of necdin or MAGE-H1 on NGF effects in PC12 cells. Both necdin- necdin, and this association was lost in necdin-expressing cells and MAGE-H1-transfected cells responded to NGF by a dose- (Fig. 2E). Thus, by interacting with p75, necdin can modulate dependent increase in neurite extension, which was greater association of p75 with other neurotrophin receptors. than observed in vector-transfected cells (Fig. 3A). This accel- Necdin, like other MAGE family proteins, does not have any erated neurite extension was quantified in stable necdin-PC12 49104 Multiple MAGE Interactors for p75 lines (Fig. 3B). Necdin expression levels did not change over the entiation were seen only in ligand-stimulated cells and are time course of these experiments (data not shown). No effect therefore consequent to NGF signaling. Very recently pub- was observed on the number of differentiated cells. Enhanced lished data from antisense experiments on embryonic DRG neurite extension was not observed when BDNF was applied neurons also support a differentiation or survival-promoting instead of NGF or upon application of NGF or BDNF to trans- role for necdin in NGF-responsive neurons (9). This could be fected nnr5 cells (data not shown). To establish whether the due to a change in trk-p75 association (Fig. 2E), thus “freeing” enhanced neurite outgrowth might be a consequence of accel- TrkA from some inhibitory constraint imposed by p75, or could erated differentiation, we quantified the phosphorylation levels be due to an independent signal emanating from p75 and of cdc2, a cell cycle marker (15). Phosphorylation of cdc2 was transduced by necdin. The propensity of necdin for homo-oli- decreased 5-fold more in necdin-PC12 cells as compared with gomerization (Fig. 2F), and its cytoplasmic localization in PC12 vector-PC12 cells 24 h after NGF application (Fig. 3C), indicat- (Fig. 3D), both support the likelihood of necdin acting as a ing that the effect of the NGF-p75-necdin signal in these cells cytoplasmic adaptor for a p75-induced signaling complex. It is indeed to accelerate differentiation. Finally, we examined should, however, be noted that this does not rule out a nuclear the subcellular distribution of necdin expression in both cycling localization or role for necdin in other cell types (16). and differentiated PC12-necdin cells. As shown in Fig. 3D Although these in vitro analyses suggest a role for a p75- necdin protein was primarily cytoplasmic in cycling PC12, and necdin signal in neuronal differentiation, what might be the in this did not change upon differentiation. An appreciable vivo significance of such signaling? The most prominent sites of amount of necdin was found in processes and growth cones of necdin expression in rodents are the DRG and hypothalamus differentiated PC12 (Fig. 3D, lower). Microscopy after harsher (Fig. 1C), both of which are known to also express p75 and trk receptors. A significant (and unexplained) loss of DRG neurons permeabilization or Western blot of nuclear extracts revealed lower amounts of necdin in the nucleus (data not shown). The has been described in p75 null mice (17), and specific hypotha- lamic defects have been described in one line of necdin mutant low levels of nuclear necdin did not change after NGF stimu- lation or during differentiation (data not shown). mice (18). Comparative analyses of these mutant mice may shed light on this interesting question. DISCUSSION To summarize, we have identified necdin and MAGE-H1 as These results establish two members of the type II MAGE novel interactors for the intracellular domain of p75 and sug- gene family, necdin and MAGE-H1, as p75 interactors. Necdin gest that type II MAGE proteins will feature prominently in and MAGE-H1 are related to NRAGE, which was previously p75 signal transduction. identified as a p75 interactor (12). All three interactors contain a conserved sequence called the MAGE homology domain Acknowledgments—We are grateful to Phil Barker, Moses Chao, and Louis Reichardt for generous gifts of antisera; to Ami Aronheim for (MHD), a defining feature of the MAGE gene family (5). Since gracious help with RRS; and to R’ada Massarwa for excellent technical both necdin and MAGE-H1 are essentially comprised of only assistance. the MHD, this structural domain determines the interaction of MAGE proteins with p75. Since most type II MAGE proteins REFERENCES are at least as closely related in sequence to MAGE-H1 or 1. Bentley, C. A., and Lee, K. F. (2000) J. Neurosci. 20, 7706 –7715 2. von Schack, D., Casademunt, E., Schweigreiter, R., Meyer, M., Bibel, M., and necdin or NRAGE as the latter three are to each other (Fig. 1B), Dechant, G. (2001) Nat. Neurosci. 4, 977–978 additional members of this subfamily may also associate with 3. Hempstead, B. L. (2002) Curr. Opin. Neurobiol. 12, 260 –267 4. Liepinsh, E., Ilag, L. L., Otting, G., and Ibanez, C. F. (1997) EMBO J. 16, p75. The death domain of p75 belongs to the structurally dis- 4999 –5005 tinct type II fold (4), and the possibility that type II MAGE 5. Barker, P. A., and Salehi, A. (2002) J. Neurosci. Res. 67, 705–712 proteins form a family of binding partners for type II death 6. Broder, Y. C., Katz, S., and Aronheim, A. (1998) Curr. Biol. 8, 1121–1124 7. Weskamp, G., and Reichardt, L. (1991) Neuron 6, 649 – 663 domain proteins is an intriguing avenue for future study. The 8. Niinobe, M., Koyama, K., and Yoshikawa, K. (2000) Dev. Neurosci. 22, multiplicity of potential interactions between these two fami- 310 –319 9. Takazaki, R., Nishimura, I., and Yoshikawa, K. (2002) Exp. Cell Res. 277, lies may create a compensatory network that is physiologically 220 –232 robust, but at the same time refractory to reductionist analysis. 10. Clary, D. O., Weskamp, G., Austin, L. R., and Reichardt, L. F. (1994) Mol. Biol. For example, the three MAGE interactors for p75 have over- Cell 5, 549 –563 11. Maruyama, K., Usami, M., Aizawa, T., and Yoshikawa, K. (1991) Biochem. lapping expression profiles, which may complicate the analyses Biophys. Res. Commun. 178, 291–296 of null allelles for these genes, at least in the context of p75 12. Salehi, A. H., Roux, P. P., Kubu, C. J., Zeindler, C., Bhakar, A., Tannis, L. L., Verdi, J. M., and Barker, P. A. (2000) Neuron 27, 279 –288 signaling. 13. Zhu, F., Yan, W., Zhao, Z. L., Chai, Y. B., Lu, F., Wang, Q., Peng, W. D., Yang, We examined the effects of NGF-induced p75 signaling via A. G., and Wang, C. J. (2000) BioTechniques 29, 310 –313 the two MAGE interactors in PC12 cells and found that they 14. Aizawa, T., Maruyama, K., Kondo, H., and Yoshikawa, K. (1992) Brain Res. Dev. Brain Res. 68, 265–274 cause an acceleration of neuronal differentiation (Fig. 3). The 15. Dobashi, Y., Shoji, M., Kitagawa, M., Noguchi, T., and Kameya, T. (2000) p75-necdin signal must synergize with a TrkA signaling path- J. Biol. Chem. 275, 12572–12580 16. Taniura, H., and Yoshikawa, K. (2002) J. Cell. Biochem. 84, 545–555 way, since the effects of necdin on differentiation are not seen 17. Fan, G., Jaenisch, R., and Kucera, J. (1999) Neuroscience 90, 259 –268 in nnr5 cells, which lack TrkA, or upon application of BDNF to 18. Muscatelli, F., Abrous, D. N., Massacrier, A., Boccaccio, I., Le Moal, M., Cau, PC12 cells. Importantly, the effects of necdin on PC12 differ- P., and Cremer, H. (2000) Hum. Mol. Genet. 9, 3101–3110 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

The p75 Neurotrophin Receptor Interacts with Multiple MAGE Proteins

Loading next page...
 
/lp/unpaywall/the-p75-neurotrophin-receptor-interacts-with-multiple-mage-proteins-3TvnZNDK14

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.c200533200
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 277, No. 51, Issue of December 20, pp. 49101–49104, 2002 Accelerated Publication © 2002 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. particular receptor is a difficult bait in conventional yeast The p75 Neurotrophin Receptor two-hybrid. Nonetheless seven candidate interactors have been Interacts with Multiple MAGE reported so far for p75 (3, 5). Most of these candidate interac- tors lack known catalytic domains, suggesting that they must Proteins* recruit additional binding partners to form a functional signal- ing complex. There are no general sequence or functional ho- Received for publication, September 19, 2002, mologies between the described p75 interactors, and the phys- and in revised form, October 21, 2002 iological significance of most of them remains to be established. Published, JBC Papers in Press, October 31, 2002, DOI 10.1074/jbc.C200533200 To identify biologically relevant interactors for p75, we used the Ras rescue system (RRS) for protein interaction trapping Marianna Tcherpakov‡, Francisca C. Bronfman‡, in yeast (6). RRS selects for protein interactions in the mem- Silvestro G. Conticello‡, Anna Vaskovsky‡, Zehava Levy‡, Michio Niinobe§, brane proximal region of the cytoplasm, and there is no re- Kazuaki Yoshikawa§, Ernest Arenas , quirement for nuclear or other translocations of the binding and Mike Fainzilber‡ partners. The system thus provides an appropriate screening From the ‡Molecular Neurobiology Group, Department method for the membrane-proximal binding of a receptor in- of Biological Chemistry, Weizmann Institute of Science, tracellular domain with its primary interactors. Our RRS 76100 Rehovot, Israel, the §Division of Regulation of screens identified two MAGE family proteins as novel p75 Macromolecular Functions, Institute for Protein interactors and highlight type II MAGE proteins as a potential Research, Osaka University, Osaka 565-0871, Japan, ¶ interactor family for type II death domain proteins. and the Department of Medical Biochemistry and Biophysics, Karolinska Institute, S-17177 Stockholm, Sweden MATERIALS AND METHODS RRS Screens, cDNA Cloning, and Sequence Analysis—Plasmids, The p75 neurotrophin receptor has been implicated in yeast strains, and screening protocols for RRS were as described (6). diverse aspects of neurotrophin signaling, but the mech- 274 342 343 416 Linker (K RWN to SLPL ) and death domain (T KRE to ESLC ) anisms by which its effects are mediated are not well regions of p75-ICD were generated by PCR from rat p75. Synthetic understood. Here we identify two MAGE proteins, nec- oligonucleotides were designed to encode the nine-residue (SES- din and MAGE-H1, as interactors for the intracellular TATSPV) tail domain of p75-ICD. A pADH death domain bait was used domain of p75 and show that the interaction is enhanced to screen a mouse embryonic head cDNA library in pMyr. Alignments of MAGE sequences and phylogenetic trees were constructed using by ligand stimulation. PC12 cells transfected with nec- ClustalX. din or MAGE-H1 exhibit accelerated differentiation in Expression Analysis—Rat tissues for Western blot analysis were response to nerve growth factor. Expression of these flash-frozen in liquid nitrogen, macerated to a fine powder, and solubi- two MAGE proteins is predominantly cytoplasmic in lized in 10 mM Tris, pH 8.0, 150 mM NaCl, 10% glycerol, 1 mM PMSF, 1 PC12 cells, and necdin was found to be capable of ho- mM orthovanadate and proteinase inhibitors (Merck) containing 1% modimerization, suggesting that it may act as a cyto- Nonidet P-40. Extracts were separated on SDS-PAGE, blotted, and plasmic adaptor to recruit a signaling complex to p75. probed with anti-necdin antibodies. These findings indicate that diverse MAGE family mem- Cell Culture, Transfections, and Generation of Cell Lines—PC12 cells bers can interact with the p75 receptor and highlight were maintained in Dulbecco’s modified Eagle’s medium containing 6% fetal calf serum and 6% horse serum. For induction of neuronal differ- type II MAGE proteins as a potential family of interac- 5 2 entiation, the cells were seeded at a density of 10 –15  10 cells/cm tors for signaling proteins containing type II death and differentiated for 2– 4 days with NGF (5–50 ng/ml). Transient domains. transfection of COS cells was with DEAE-dextran. Transient and stable transfections of PC12 cells lines were carried out by electroporation with 80 g of DNA per 10 cells in 0.4-cm cuvettes on an ECM-300 Neuronal responses to activation of the p75 neurotrophin (BTX) set to 320 V, 6-ms pulse length. Stably transfected PC12 were receptor range from enhanced outgrowth to increased cell selected under 0.5 mg/ml Geneticin for 4 weeks until resistant colo- nies appeared. death, and p75 null mice exhibit a plethora of defects in both Immunoprecipitation, Western Blots, and Immunofluorescence— neuronal and non-neuronal systems (1, 2). However, the pri- Cells were lysed in 10 mM Tris, pH8.0, 150 mM NaCl, 10% glycerol, 1 mM mary interactors and signaling mechanisms activated by p75 PMSF, 1 mM orthovanadate and proteinase inhibitors (Merck) contain- are not well understood (3). Although p75 belongs to the TNF ing either 0.7% CHAPS or 1% Nonidet P-40. Lysates were spun at receptor superfamily, its intracellular type II death domain 10,000  g for 5 min and then precleared with 30 l of a 50% suspension does not self-aggregate (4) and does not interact with the major of protein-G-Sepharose beads (Amersham Biosciences) before incuba- tion with primary antibody. Immunoprecipitates were eluted with 80 l binding partners of type I death domains. Identification of of 1 M guanidine HCl in 50 mM Tris, pH 8.0, before running on SDS- specific signaling partners for p75 has been frustrating, as this PAGE. Primary antibodies for precipitation were as follows: anti-p75 MC192 (from Alomone Laboratories), anti-p75 REX (7), monoclonal * This work was supported by grants from the Israel Science Foun- anti-HA (from Roche Molecular Biochemicals). Antibodies and dilutions dation (647/01) and the European Union Fifth Framework Program for Western blots were as follows: anti-necdin polyclonal NC243 (8) at (QLRT-1999-573) (to M. F.) and by a short term European Molecular 1:5000, anti-necdin polyclonal MNF (9) at 1:1000, anti-HA polyclonal Biology fellowship (to M. T.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 The abbreviations used are: RRS, Ras rescue system; PMSF, phen- U.S.C. Section 1734 solely to indicate this fact. ylmethylsulfonyl fluoride; NGF, nerve growth factor; CHAPS, 3-[(3- Incumbent of the Daniel Koshland Sr. Career Development Chair at cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; HA, he- the Weizmann Institute of Science. To whom correspondence should be magglutinin; PBS, phosphate-buffered saline; DRG, dorsal root ganglia; addressed. Tel.: 972-8-934-4266; Fax: 972-8-934-4112; E-mail: mike. BDNF, brain-derived neurotrophic factor; MHD, MAGE homology fainzilber@weizmann.ac.il. domain. This paper is available on line at http://www.jbc.org 49101 This is an Open Access article under the CC BY license. 49102 Multiple MAGE Interactors for p75 FIG.1. Necdin, a MAGE family protein, interacts with p75. A, myristoylated linker and death domain regions of p75 interact with necdin-Ras to rescue transformed yeast grown at 36 °C, whereas no interaction is observed with the tail domain. B, phylogenetic tree show- ing relationships between human type II MAGE genes; * denotes those shown to interact with p75. C, Western blot analysis of necdin expres- sion in rat tissues. 250 g of tissue protein was loaded in each lane, and the blot was probed with the anti-necdin MNF antibody. Equal loading was verified by stripping and reprobing the blot with anti-tubulin antibody (not shown). FIG.2. Association of necdin and MAGE-H1 with p75 in mam- (Santa Cruz) at 1:200, anti-p75 polyclonal 9651 at 1:1000, anti-phospho- malian cells. A, interaction of p75 with MAGE proteins transiently cdc2 polyclonal (R & D Systems) at 5 g/ml, and anti-TrkA RTA (10) at expressed in COS cells. P75 was co-transfected with necdin or MAGE- 1:1000. Cells were fixed for immunofluorescence microscopy for 20 min H1-HA. As a control MAGE proteins were co-transfected with empty with ice-cold 3% paraformaldehyde in PBS at room temperature, vector. Immunoblotting of cell lysates shows equal expression of MAGE quenched in 0.1 M glycine PBS for 15 min, and permeabilized for 10 min proteins with or without p75. P75 was immunoprecipitated with the MC192 antibody and blots were tested for the presence of MAGE in PBS, 5% donkey serum, 1 mg/ml bovine serum albumin, 0.2% sapo- proteins with anti-necdin and anti-HA antibodies. MAGE proteins were nin. Permeabilized cells were incubated with anti-HA (1:4000) for2hat immunoprecipitated only in the presence of p75. B, interaction of p75 room temperature followed by donkey anti-rabbit rhodamine-redX-con- with overexpressed MAGE proteins in transiently transfected PC12 jugated secondary antibody (1:500, Jackson ImmunoResearch) for 1 h cells. PC12 were electroporated with constructs expressing MAGE pro- at room temperature. teins, and co-immunoprecipitations were carried out with the indicated antibodies. An irrelevant antibody was used as control ( in the figure). RESULTS Both necdin and MAGE-H1 were co-precipitated with the p75 antibody, A mouse embryonic head RRS library was screened against and p75 was co-precipitated with anti-MAGE antibodies. C, interaction three baits comprising the death domain, the 69-residue jux- of p75 with necdin is enhanced by NGF or BDNF in a stably transfected PC12 line. A stable PC12 cell line expressing HA-necdin was treated tamembrane “linker,” and the nine-residue C-terminal “tail” of with 100 ng/ml NGF or BDNF for 2 h, followed by immunoprecipitation p75. These screens yielded interacting candidates only for the with the MC192 antibody. Western blotting for necdin reveals increas- death domain bait. One of the clones encoded the C-terminal 81 ing amounts of co-precipitated necdin upon ligand stimulation. D, co- amino acid residues of necdin (11), a MAGE family member immunoprecipitation of necdin with p75 from lysates of freshly dis- sected adult rat DRG. 40 – 45 DRG were used for each lane, and lysates thought to act as a cell cycle regulator. Further yeast co-trans- were precipitated with the anti-p75 monocolonal MC192, followed by fections of necdin with the complete panel of p75-ICD subdo- Western blot with the anti-necdin polyclonal MNF. E, co-immunopre- mains revealed that necdin interacts with both the linker and cipitation of TrkA with p75 from transiently transfected COS cells. death domain segments of p75 (Fig. 1A). The sequence homol- NGF (100 ng/ml) was added to the cells 44 h after transfection, and lysates were processed for co-immunoprecipitation after 4 h incubation. ogies between necdin and NRAGE, which was previously iden- P75 pull-down was with the REX antibody, and TrkA detection was tified by Barker and colleagues as a p75 interactor (12), with the RTA antibody. F, necdin-necdin interaction exemplified by prompted us to examine additional family members. Unfortu- RRS at 36 °C(left panels) and co-immunoprecipitation from transiently nately, a number of MAGE constructs (e.g. NRAGE and MA- transfected COS cells (right panel). GEL2) were found to associate with yeast plasma membrane, thus precluding their use in the RRS system. MAGE-H1/APR-1 Adult rats express robust levels of necdin protein in the hypo- (13) was found to interact robustly with the p75 death domain thalamus and in dorsal root ganglia (DRG) (Fig. 1C), whereas by RRS (data not shown). Thus, at least three type II MAGE necdin levels during development are much lower. proteins (Fig. 1B) are capable of interacting with p75. Tissue Co-immunoprecipitation experiments were then carried out RNA arrays (data not shown) and in silico profiling reveal in both COS and PC12 cells to confirm the p75-necdin and extensive overlaps in the expression profiles of these genes. p75-MAGE-H1 interactions. In accordance with the RRS re- Multiple MAGE Interactors for p75 49103 FIG.3. Necdin and MAGE-H1 accelerate the differentiation of PC12 cells. A, PC12 cells were transiently transfected with necdin, MAGE-H1, or vector control and incubated for 3 days in 50 ng/ml NGF. B, differentiation of two necdin-expressing stable PC12 lines, and two vector-transfected stable lines, in the presence of 50 ng/ml NGF for 2 days. Average  S.E. is shown for a representative experiment. 120 –180 differentiated cells were measured for each point. C, levels of phospho-cdc2 in necdin versus vector-ransfected PC12 cells, after 24 h in 50 ng/ml NGF. Left panel shows a representative experiment, while the right panel shows average  S.E. (n  3). D, immunofluorescence microscopy reveals predominantly cytoplasmic necdin protein in both cycling (upper) and differentiated (lower) PC12-necdin cells. sults, p75 co-precipitated with necdin or MAGE-H1 from trans- known intrinsic catalytic activity and is likely to signal by fected COS cells (Fig. 2A) or PC12 cells (Fig. 2B). In PC12 cell recruiting other molecules to a p75 signaling complex. We lines stably transfected with necdin, application of either NGF therefore conducted additional RRS screens using necdin as or BDNF clearly enhanced the interaction of necdin with p75 bait and identified a number of novel expressed sequence tags (Fig. 2C). Necdin protein levels did not change during the of unknown function as necdin interactors (data not shown). course of this experiment (data not shown). Finally, a p75- Interestingly, one known gene obtained in this screen was necdin interaction at endogenous levels of expression was dem- necdin itself (Fig. 2F, left). Homo-oligomerization of necdin was onstrated by co-immunoprecipitation from lysates of freshly confirmed by co-immunoprecipitation of differentially tagged dissected adult rat DRG (Fig. 2D). necdin proteins from transfected COS cells (Fig. 2F, right). NRAGE was previously found to affect p75-trk association PC12 cells are a well established model for NGF-induced (12), and we therefore examined whether necdin had similar differentiation and do not express endogenous necdin (14). We effects in transiently transfected COS cells stimulated with therefore examined the influence of transient transfection of NGF. TrkA co-precipitated together with p75 in the absence of necdin or MAGE-H1 on NGF effects in PC12 cells. Both necdin- necdin, and this association was lost in necdin-expressing cells and MAGE-H1-transfected cells responded to NGF by a dose- (Fig. 2E). Thus, by interacting with p75, necdin can modulate dependent increase in neurite extension, which was greater association of p75 with other neurotrophin receptors. than observed in vector-transfected cells (Fig. 3A). This accel- Necdin, like other MAGE family proteins, does not have any erated neurite extension was quantified in stable necdin-PC12 49104 Multiple MAGE Interactors for p75 lines (Fig. 3B). Necdin expression levels did not change over the entiation were seen only in ligand-stimulated cells and are time course of these experiments (data not shown). No effect therefore consequent to NGF signaling. Very recently pub- was observed on the number of differentiated cells. Enhanced lished data from antisense experiments on embryonic DRG neurite extension was not observed when BDNF was applied neurons also support a differentiation or survival-promoting instead of NGF or upon application of NGF or BDNF to trans- role for necdin in NGF-responsive neurons (9). This could be fected nnr5 cells (data not shown). To establish whether the due to a change in trk-p75 association (Fig. 2E), thus “freeing” enhanced neurite outgrowth might be a consequence of accel- TrkA from some inhibitory constraint imposed by p75, or could erated differentiation, we quantified the phosphorylation levels be due to an independent signal emanating from p75 and of cdc2, a cell cycle marker (15). Phosphorylation of cdc2 was transduced by necdin. The propensity of necdin for homo-oli- decreased 5-fold more in necdin-PC12 cells as compared with gomerization (Fig. 2F), and its cytoplasmic localization in PC12 vector-PC12 cells 24 h after NGF application (Fig. 3C), indicat- (Fig. 3D), both support the likelihood of necdin acting as a ing that the effect of the NGF-p75-necdin signal in these cells cytoplasmic adaptor for a p75-induced signaling complex. It is indeed to accelerate differentiation. Finally, we examined should, however, be noted that this does not rule out a nuclear the subcellular distribution of necdin expression in both cycling localization or role for necdin in other cell types (16). and differentiated PC12-necdin cells. As shown in Fig. 3D Although these in vitro analyses suggest a role for a p75- necdin protein was primarily cytoplasmic in cycling PC12, and necdin signal in neuronal differentiation, what might be the in this did not change upon differentiation. An appreciable vivo significance of such signaling? The most prominent sites of amount of necdin was found in processes and growth cones of necdin expression in rodents are the DRG and hypothalamus differentiated PC12 (Fig. 3D, lower). Microscopy after harsher (Fig. 1C), both of which are known to also express p75 and trk receptors. A significant (and unexplained) loss of DRG neurons permeabilization or Western blot of nuclear extracts revealed lower amounts of necdin in the nucleus (data not shown). The has been described in p75 null mice (17), and specific hypotha- lamic defects have been described in one line of necdin mutant low levels of nuclear necdin did not change after NGF stimu- lation or during differentiation (data not shown). mice (18). Comparative analyses of these mutant mice may shed light on this interesting question. DISCUSSION To summarize, we have identified necdin and MAGE-H1 as These results establish two members of the type II MAGE novel interactors for the intracellular domain of p75 and sug- gene family, necdin and MAGE-H1, as p75 interactors. Necdin gest that type II MAGE proteins will feature prominently in and MAGE-H1 are related to NRAGE, which was previously p75 signal transduction. identified as a p75 interactor (12). All three interactors contain a conserved sequence called the MAGE homology domain Acknowledgments—We are grateful to Phil Barker, Moses Chao, and Louis Reichardt for generous gifts of antisera; to Ami Aronheim for (MHD), a defining feature of the MAGE gene family (5). Since gracious help with RRS; and to R’ada Massarwa for excellent technical both necdin and MAGE-H1 are essentially comprised of only assistance. the MHD, this structural domain determines the interaction of MAGE proteins with p75. Since most type II MAGE proteins REFERENCES are at least as closely related in sequence to MAGE-H1 or 1. Bentley, C. A., and Lee, K. F. (2000) J. Neurosci. 20, 7706 –7715 2. von Schack, D., Casademunt, E., Schweigreiter, R., Meyer, M., Bibel, M., and necdin or NRAGE as the latter three are to each other (Fig. 1B), Dechant, G. (2001) Nat. Neurosci. 4, 977–978 additional members of this subfamily may also associate with 3. Hempstead, B. L. (2002) Curr. Opin. Neurobiol. 12, 260 –267 4. Liepinsh, E., Ilag, L. L., Otting, G., and Ibanez, C. F. (1997) EMBO J. 16, p75. The death domain of p75 belongs to the structurally dis- 4999 –5005 tinct type II fold (4), and the possibility that type II MAGE 5. Barker, P. A., and Salehi, A. (2002) J. Neurosci. Res. 67, 705–712 proteins form a family of binding partners for type II death 6. Broder, Y. C., Katz, S., and Aronheim, A. (1998) Curr. Biol. 8, 1121–1124 7. Weskamp, G., and Reichardt, L. (1991) Neuron 6, 649 – 663 domain proteins is an intriguing avenue for future study. The 8. Niinobe, M., Koyama, K., and Yoshikawa, K. (2000) Dev. Neurosci. 22, multiplicity of potential interactions between these two fami- 310 –319 9. Takazaki, R., Nishimura, I., and Yoshikawa, K. (2002) Exp. Cell Res. 277, lies may create a compensatory network that is physiologically 220 –232 robust, but at the same time refractory to reductionist analysis. 10. Clary, D. O., Weskamp, G., Austin, L. R., and Reichardt, L. F. (1994) Mol. Biol. For example, the three MAGE interactors for p75 have over- Cell 5, 549 –563 11. Maruyama, K., Usami, M., Aizawa, T., and Yoshikawa, K. (1991) Biochem. lapping expression profiles, which may complicate the analyses Biophys. Res. Commun. 178, 291–296 of null allelles for these genes, at least in the context of p75 12. Salehi, A. H., Roux, P. P., Kubu, C. J., Zeindler, C., Bhakar, A., Tannis, L. L., Verdi, J. M., and Barker, P. A. (2000) Neuron 27, 279 –288 signaling. 13. Zhu, F., Yan, W., Zhao, Z. L., Chai, Y. B., Lu, F., Wang, Q., Peng, W. D., Yang, We examined the effects of NGF-induced p75 signaling via A. G., and Wang, C. J. (2000) BioTechniques 29, 310 –313 the two MAGE interactors in PC12 cells and found that they 14. Aizawa, T., Maruyama, K., Kondo, H., and Yoshikawa, K. (1992) Brain Res. Dev. Brain Res. 68, 265–274 cause an acceleration of neuronal differentiation (Fig. 3). The 15. Dobashi, Y., Shoji, M., Kitagawa, M., Noguchi, T., and Kameya, T. (2000) p75-necdin signal must synergize with a TrkA signaling path- J. Biol. Chem. 275, 12572–12580 16. Taniura, H., and Yoshikawa, K. (2002) J. Cell. Biochem. 84, 545–555 way, since the effects of necdin on differentiation are not seen 17. Fan, G., Jaenisch, R., and Kucera, J. (1999) Neuroscience 90, 259 –268 in nnr5 cells, which lack TrkA, or upon application of BDNF to 18. Muscatelli, F., Abrous, D. N., Massacrier, A., Boccaccio, I., Le Moal, M., Cau, PC12 cells. Importantly, the effects of necdin on PC12 differ- P., and Cremer, H. (2000) Hum. Mol. Genet. 9, 3101–3110

Journal

Journal of Biological ChemistryUnpaywall

Published: Dec 1, 2002

There are no references for this article.