Compartmentation of Fyn Kinase with Glycosylphosphatidylinositol-anchored Molecules in Oligodendrocytes Facilitates Kinase Activation during Myelination
1999; Elsevier BV; Volume: 274; Issue: 41 Linguagem: Inglês
10.1074/jbc.274.41.29042
ISSN1083-351X
AutoresEva‐Maria Krämer‐Albers, Corinna Klein, Thomas Koch, Monica Boytinck, Jacqueline Trotter,
Tópico(s)Signaling Pathways in Disease
ResumoIn many cell types, glycosylphosphatidylinositol (GPI)-anchored proteins are sequestered in detergent-resistant membrane rafts. These are plasma membrane microdomains enriched in glycosphingolipids and cholesterol and are suggested to be platforms for cell signaling. Concomitant with the synthesis of myelin glycosphingolipids, maturing oligodendrocytes progressively associate GPI-anchored proteins, including the adhesion molecules NCAM 120 and F3, in rafts. Here we show that these microdomains include Fyn and Lyn kinases. Both kinases are maximally active in myelin prepared from young animals, correlating with early stages of myelination. In the rafts, Fyn kinase is tightly associated with NCAM 120 and F3. In contrast, in oligodendrocyte progenitor cells lacking rafts or in raft-free membrane domains of more mature cells, F3 does not associate with Fyn. The addition of anti-F3 antibodies to oligodendrocytes results in stimulation of Fyn kinase specifically in rafts. Compartmentation of oligodendrocyte GPI-anchored proteins in rafts is thus a prerequisite for association with Fyn, permitting kinase activation. Interaction of oligodendrocyte F3 with axonal ligands such as L1 and ensuing kinase activation may play a crucial role in initiating myelination. In many cell types, glycosylphosphatidylinositol (GPI)-anchored proteins are sequestered in detergent-resistant membrane rafts. These are plasma membrane microdomains enriched in glycosphingolipids and cholesterol and are suggested to be platforms for cell signaling. Concomitant with the synthesis of myelin glycosphingolipids, maturing oligodendrocytes progressively associate GPI-anchored proteins, including the adhesion molecules NCAM 120 and F3, in rafts. Here we show that these microdomains include Fyn and Lyn kinases. Both kinases are maximally active in myelin prepared from young animals, correlating with early stages of myelination. In the rafts, Fyn kinase is tightly associated with NCAM 120 and F3. In contrast, in oligodendrocyte progenitor cells lacking rafts or in raft-free membrane domains of more mature cells, F3 does not associate with Fyn. The addition of anti-F3 antibodies to oligodendrocytes results in stimulation of Fyn kinase specifically in rafts. Compartmentation of oligodendrocyte GPI-anchored proteins in rafts is thus a prerequisite for association with Fyn, permitting kinase activation. Interaction of oligodendrocyte F3 with axonal ligands such as L1 and ensuing kinase activation may play a crucial role in initiating myelination. neural cell adhesion molecule detergent-insoluble glycosphingolipid-rich microdomains glycosylphosphatidylinositol myelin-associated glycoprotein postnatal day n dibutyryl cyclic AMP polyacrylamide gel electrophoresis The formation of a myelin sheath is essential for the rapid saltatory propagation of action potentials in the vertebrate nervous system (1Hildebrand C. Remahl S. Persson H. Bjartmar C. Prog. Neurobiol. 1993; 40: 319-384Crossref PubMed Scopus (301) Google Scholar). Myelination in the central nervous system involves sequential stages of interaction between the myelinating glial cell, the oligodendrocyte, and the neuronal process, the axon. Initial recognition and adhesion results in wrapping of the glial process around the axon (ensheathment) followed by the laying down of the multilamellar sheath. Although progenitor cells have the intrinsic potential to differentiate into oligodendrocytes in vitro in the absence of axons (2Raff M.C. Miller R.H. Noble M. Nature. 1983; 303: 390-396Crossref PubMed Scopus (1692) Google Scholar, 3Behar T. McMorris F.A. Novotny E.A. Barker J.L. Dubois-Dalcq M. J. Neurosci. Res. 1988; 21: 168-180Crossref PubMed Scopus (137) Google Scholar, 4Pfeiffer S.E. Warrington A.E. Bansal R. Trends Cell Biol. 1993; 3: 191-197Abstract Full Text PDF PubMed Scopus (737) Google Scholar), myelination in vivodemonstrates a high degree of specificity and requires axonal signals. For example, oligodendrocytes do not normally myelinate dendrites (5Lubetzki C. Demerens C. Anglade P. Villarroya H. Frankfurter A. Lee V.M.Y. Zalc B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6820-6824Crossref PubMed Scopus (112) Google Scholar,6Meyer-Franke A. Barres B. Curr. Biol. 1994; 4: 847-850Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar). Cell adhesion molecules expressed by both axons and glial cells are known to play a crucial role in the establishment of axon-glial contact and subsequent signaling to the oligodendrocyte, driving the process of myelination (7Doyle J.P. Colman D.R. Curr. Opin. Cell Biol. 1993; 5: 779-785Crossref PubMed Scopus (50) Google Scholar, 8Notterpek L.M. Rome L.H. Neuron. 1994; 13: 473-485Abstract Full Text PDF PubMed Scopus (36) Google Scholar). In turn, the myelinating glial cell triggers an axonal reaction, culminating in an increased phosphorylation of neurofilaments and regulation of the axonal diameter (9de Waegh S.M. Lee V.M.Y. Brady S.T. Cell. 1992; 68: 451-463Abstract Full Text PDF PubMed Scopus (679) Google Scholar, 10Hsieh S.T. Kidd G.J. Crawford T.O. Xu Z. Lin W.M. Trapp B.D. Cleveland D.W. Griffin J.W. J. Neurosci. 1994; 14: 6392-6401Crossref PubMed Google Scholar). Members of the cadherin and the Ig superfamily such as L1, NCAM,1 and especially MAG are candidate molecules mediating axon-glial interactions (7Doyle J.P. Colman D.R. Curr. Opin. Cell Biol. 1993; 5: 779-785Crossref PubMed Scopus (50) Google Scholar, 11Payne H.R. Hemperly J.J. Lemmon V. Dev. Brain Res. 1996; 97: 9-15Crossref PubMed Scopus (50) Google Scholar). The formation of morphologically normal myelin sheaths in MAG knockout and even in MAG/NCAM double knockout mice suggests that either these molecules are not involved in myelin formation or additional molecules participate in the early events of myelination (12Li C. Tropak M.B. Gerlai R. Clapoff S. Abramow-Newerly W. Trapp B. Peterson A. Roder J. Nature. 1994; 369: 747-750Crossref PubMed Scopus (327) Google Scholar, 13Montag D. Giese K.P. Bartsch U. Martini R. Lang Y. Blüthmann H. Karthigasan J. Kirschner D.A. Wintergerst E.S. Nave K.A. Zielasek J. Toyka K.V. Lipp H.P. Schachner M. Neuron. 1994; 13: 229-246Abstract Full Text PDF PubMed Scopus (337) Google Scholar, 14Carenini S. Montag D. Cremer H. Schachner M. Martini R. Cell Tissue Res. 1997; 287: 3-9Crossref PubMed Scopus (74) Google Scholar). Our knowledge about the signals exchanged between axons and oligodendrocytes during myelination is also incomplete. A role for oligodendroglial Fyn kinase (a nonreceptor tyrosine kinase of the Src family) has been proposed as Fyn phosphorylates MAG after co-transfection in COS cells and Fyn knockout mice are hypomyelinated (15Umemori H. Sato S. Yagi T. Aizawa S. Yamamoto T. Nature. 1994; 367: 572-576Crossref PubMed Scopus (356) Google Scholar). The integrity of the axon-glial unit requires continual bidirectional signaling between the oligodendrocyte and the axon. GPI-anchored proteins expressed by oligodendrocytes and their precursor cells are candidates for recognition molecules dictating the initial interactions between axon and oligodendrocyte and acting as receptors mediating axon-glial signal transduction. Oligodendrocyte precursor cells express a distinct pattern of GPI-anchored molecules, which is retained as the cells mature and is present in preparations of adult myelin. Although both oligodendrocyte precursor cells and mature oligodendrocytes express a similar pattern of GPI-anchored proteins, in oligodendrocytes and myelin, but not in precursor cells, these proteins are associated with the major myelin lipids galactocerebroside, sulfatide, and cholesterol in membrane domains. These domains can be isolated as detergent-insoluble glycosphingolipid-rich microdomains (DIGs) by sucrose density gradient centrifugation (16Krämer E.-M. Koch T. Niehaus A. Trotter J. J. Biol. Chem. 1997; 272: 8937-8945Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar). DIGs can be isolated from several cell types and are thought to represent raft-like microdomains within the plasma membrane, resulting from the lateral assembly of sphingolipids and cholesterol in the exoplasmic leaflet of the lipid bilayer due to weak interactions between their polar headgroups (17Simons K. Ikonen E. Nature. 1997; 387: 569-572Crossref PubMed Scopus (8271) Google Scholar, 18Harder T. Simons K. Curr. Opin. Cell Biol. 1997; 9: 534-542Crossref PubMed Scopus (720) Google Scholar). In polarized epithelial cells, where rafts emerge in the trans-Golgi network, they are thought to be responsible for the apical sorting of glycosphingolipids, GPI-anchored proteins, and other apical marker molecules (17Simons K. Ikonen E. Nature. 1997; 387: 569-572Crossref PubMed Scopus (8271) Google Scholar, 18Harder T. Simons K. Curr. Opin. Cell Biol. 1997; 9: 534-542Crossref PubMed Scopus (720) Google Scholar, 19Brown D.A. Rose J.K. Cell. 1992; 68: 533-544Abstract Full Text PDF PubMed Scopus (2661) Google Scholar). We postulated that in oligodendrocytes, as in epithelial cells, DIGs represent raftlike membrane domains in which oligodendrocyte GPI-anchored proteins together with the major myelin lipids are sorted into the forming myelin sheath (16Krämer E.-M. Koch T. Niehaus A. Trotter J. J. Biol. Chem. 1997; 272: 8937-8945Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar). Mice in which the enzyme UDP-galactose:ceramide galactosyl transferase catalyzing the synthesis of the myelin lipids galactocerebroside and sulfatide has been knocked out exhibit tremors and hind limb paralysis and die prematurely (20Bosio A. Binczek E. Stoffel W. Proc. Natl. Sci. U. S. A. 1996; 93: 13280-13285Crossref PubMed Scopus (289) Google Scholar, 21Coetzee T. Fujita N. Dupree J.L. Shi R. Blight A. Susuki Ki Susuki Ku Popko B. Cell. 1996; 86: 209-219Abstract Full Text Full Text PDF PubMed Scopus (523) Google Scholar). They show deficits in nerve conduction despite the formation of compact myelin. However, the nodal/paranodal structure in these mice is severely perturbed (22Dupree J.L. Coetzee T. Blight A. Susuki K. Popko B. J. Neurosci. 1998; 18: 1642-1649Crossref PubMed Google Scholar, 23Coetzee T. Susuki K. Popko B. Trends Neurosci. 1998; 21: 126-130Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar). Whether these defects result from a loss of functional properties of these lipids such as insulation or from the loss of lipid-associated targeting of specific proteins including signaling molecules to areas of the forming myelin sheath is unelucidated. As was first shown for T-lymphocytes, in many cell types DIG microdomains include nonreceptor tyrosine kinases of the Src family (24Robinson P.J. Immunology Today. 1991; 12: 35-41Abstract Full Text PDF PubMed Scopus (255) Google Scholar, 25Brown D.A. Curr. Opin. Immunol. 1993; 5: 349-354Crossref PubMed Scopus (196) Google Scholar). In this paper, we investigated the association of such kinases with GPI-anchored proteins in oligodendrocytes and myelin, since they could be involved in signal transduction between the wrapping glial cell process and the axon. Our results show that the two major oligodendroglial GPI-anchored proteins, the 120-kDa isoform of NCAM (26Bhat S. Silberberg D.H. J. Neurosci. 1986; 6: 3348-3354Crossref PubMed Google Scholar, 27Trotter J. Bitter-Suermann D. Schachner M. J. Neurosci. Res. 1989; 22: 369-383Crossref PubMed Scopus (144) Google Scholar) and F3 (28Gennarini G. Cibelli G. Rougon G. Mattei M.G. Goridis C. J. Cell Biol. 1989; 109: 775-788Crossref PubMed Scopus (244) Google Scholar, 29Koch T. Brugger T. Bach A. Gennarini G. Trotter J. Glia. 1997; 19: 199-212Crossref PubMed Scopus (79) Google Scholar, 30Einheber S. Zanazzi G. Ching W. Scherer S. Milner T.A. Peles E. Salzer J.L. J. Cell Biol. 1997; 139: 1495-1506Crossref PubMed Scopus (320) Google Scholar), both members of the Ig superfamily of adhesion molecules, colocalize with the Src family kinases Fyn and Lyn in oligodendrocyte and myelin DIGs. The activity of both kinases within the DIGs is developmentally regulated, being most active at the beginning of myelination. Fyn but not Lyn kinase activity is stably associated with both NCAM 120 and F3. Furthermore, antibody-mediated cross-linking of F3 results in stimulation of the Fyn kinase activity localized to oligodendrocyte DIGs. The association of NCAM 120 and F3 with Fyn kinase to a functional signaling complex within raftlike glycosphingolipid-rich microdomains during oligodendrocyte maturation may be critical for signal transduction between axon and glial cell in the early phases of myelination. Radiochemicals ([γ-32P]ATP, l-[35S]Met/Cysin vitro labeling mix) and ECL reagents were from Amersham Pharmacia Biotech (Braunschweig, Germany); human recombinant platelet-derived growth factor (AA) and basic fibroblast growth factor were from TEBU (Frankfurt, Germany); dibutyryl cyclic AMP (dbcAMP), Triton X-100, Nonidet P-40, and sodium deoxycholate were from Sigma (Deisenhofen, Germany); Protein A-Sepharose CL4B was from Amersham Pharmacia Biotech (Freiburg, Germany); Bradford reagent for protein assays was from Bio-Rad (München, Germany); polyvinylidene difluoride membrane was from Millipore (Bedford, MA). The amino-terminal peptide of F3 (KGFGPIFEEQPINT) was synthesized by Dr. R. Frank (ZMBH, University of Heidelberg, Germany). The following rabbit polyclonal antibodies were used: antibodies recognizing NCAM (27Trotter J. Bitter-Suermann D. Schachner M. J. Neurosci. Res. 1989; 22: 369-383Crossref PubMed Scopus (144) Google Scholar), F3 (Ig fraction of a serum raised against the N-terminal F3 peptide; Ref. 29Koch T. Brugger T. Bach A. Gennarini G. Trotter J. Glia. 1997; 19: 199-212Crossref PubMed Scopus (79) Google Scholar), the AN2 antigen (31Niehaus A. Stegmüller J. Diers-Fenger M. Trotter J. J. Neurosci. 1999; 19: 4948-4961Crossref PubMed Google Scholar), Fyn, and Lyn (Santa Cruz Biotechnology, Inc., Heidelberg). The following monoclonal antibodies were used: murine monoclonal antibody 27-11-111 (mAb F11 No. 8) made against chick F11 (32Brümmendorf T. Hubert M. Treubert U. Leuschner R. Tarnok A. Rathjen F.G. Neuron. 1993; 10: 711-727Abstract Full Text PDF PubMed Scopus (176) Google Scholar), which cross-reacts with mouse F3 (56Koch T. Untersuchungen zu Oligodendroglialen Glykosylphosphatidylinositol-Verankerken Molekülen, mit Besonderem Angenmerk auf die Expression von F3. Ph.D. thesis. University of Heidelberg, 1998Google Scholar), kindly provided by Dr. F. Rathjen (Berlin, Germany); rat monoclonal antibody AN2, which recognizes a surface epitope of a 330-kDa glycoprotein expressed by the cell line Oli-neu and primary oligodendrocyte progenitors (31Niehaus A. Stegmüller J. Diers-Fenger M. Trotter J. J. Neurosci. 1999; 19: 4948-4961Crossref PubMed Google Scholar); murine monoclonal antibody 4G10 against phosphotyrosine from Upstate Biotechnology, Inc. (Lake Placid, NY); and mouse monoclonal antibody against Fyn (Pharmingen/Transduction Laboratories). Secondary antibodies were from Dianova (Hamburg, Germany). Primary cultures of oligodendrocytes were prepared from embryonic day 14–16 mice as described (27Trotter J. Bitter-Suermann D. Schachner M. J. Neurosci. Res. 1989; 22: 369-383Crossref PubMed Scopus (144) Google Scholar, 33Sontheimer H. Trotter J. Schachner M. Kettenmann H. Neuron. 1989; 2: 1135-1145Abstract Full Text PDF PubMed Scopus (224) Google Scholar). Oligodendrocytes growing on top of astrocyte monolayers were shaken off and plated in modified Sato medium (27Trotter J. Bitter-Suermann D. Schachner M. J. Neurosci. Res. 1989; 22: 369-383Crossref PubMed Scopus (144) Google Scholar) containing 1% horse serum on poly-l-lysine-coated dishes. To increase the proportion of precursor cells as well as to promote survival, 10 ng/ml human recombinant platelet-derived growth factor (AA), and 5 ng/ml basic fibroblast growth factor were added immediately after the shake and after 24 h in vitro. Oligodendrocytes were kept for 5 days in vitro without further growth factor additions before they were harvested. The resulting population, which was used for all experiments with primary cultures, is enriched for differentiated oligodendrocytes but contains a fraction of progenitor cells (27Trotter J. Bitter-Suermann D. Schachner M. J. Neurosci. Res. 1989; 22: 369-383Crossref PubMed Scopus (144) Google Scholar). The cell line Oli-neu(34Jung M. Krämer E.M. Grzenkowski M. Tang K. Blakemore W. Aguzzi A. Khazaie K. Chlichlia K. von Blankenfeld G. Kettenmann H. Trotter J. Eur. J. Neurosci. 1995; 7: 1245-1265Crossref PubMed Scopus (218) Google Scholar) was cultured in Sato medium containing 1% horse serum. To induce differentiation of the precursor-like Oli-neu cells, cultures were treated with 1 mm dbcAMP for 3–4 days (daily additions to the culture medium). For metabolic labeling primary oligodendrocytes and Oli-neucells were starved for 1 h in SO4/Met/Cys-free DMEM and incubated for 4 h with 100 μCi/mll-[35S]Met/Cys labeling mix. Myelin was isolated from the brains of young postnatal (postnatal days (P)9/10, P12, P16, P20, P30, or P45) and adult NMRI mice of both sexes according to standard procedures (35Smith M.E. J. Neurochem. 1969; 16: 83-92Crossref PubMed Scopus (74) Google Scholar,36Norton W.T. Poduslo S.E. J. Neurochem. 1973; 21: 749-757Crossref PubMed Scopus (1282) Google Scholar). Initially, brains were homogenized in ice-cold 10.5% sucrose using the Ultra-Turrax T25 (IKA, Staufen, Germany). Myelin was removed from the interface between 10.5 and 30% sucrose step gradients and subjected to two rounds of hypoosmotic shock by resuspension in a large volume of ice-cold water and reisolation on the step gradient; this separates myelin membranes from axolemma. Purified myelin was collected from the final sucrose interface, washed twice with cold water, resuspended in a small volume of water, and immediately used or frozen in small aliquots at −80 °C. The protein content of the myelin preparations was determined by using the Bio-Rad protein assay with bovine serum albumin as a protein standard. Detergent extracts were prepared as described (19Brown D.A. Rose J.K. Cell. 1992; 68: 533-544Abstract Full Text PDF PubMed Scopus (2661) Google Scholar,37Sargiacomo M. Sudol M. Tang Z. Lisanti M.P. J. Cell Biol. 1993; 122: 789-807Crossref PubMed Scopus (879) Google Scholar). In brief, primary oligodendrocytes (2–3 × 107) or sonicated myelin (300 μg of total protein) were solubilized at 4 °C in 1 ml of extraction buffer containing 10 mmTris/HCl, pH 7.4, 150 mm NaCl, 5 mm EDTA, 1 mm phenylmethylsulfonyl fluoride, and 2% Triton X-100 (TNE/Triton X-100). The extracts were shaken for 30 min at 4 °C. Detergent extracts were adjusted to 40% sucrose by adding equal volumes of 80% sucrose in TNE without Triton X-100 and placed into an ultracentrifuge tube. A linear gradient from 5 to 30% sucrose (in TNE without Triton X-100) was layered over the lysate. Gradients were centrifuged for 12 h at 35,000 rpm at 4 °C in a Beckmann SW 40 TI rotor (218,000 × g). 1-ml fractions were harvested, and the density was determined by measurement of the refractive index. Proteins in each fraction were analyzed by SDS-PAGE followed by Western blot. Light gradient fractions containing floating GPI-anchored proteins, glycosphingolipids, and cholesterol (DIGs) were collected, diluted with double distilled H2O, and pelleted for 1 h at 218,000 × g and 4 °C. Isolated DIGs were subjected to an in vitro kinase assay and immunoprecipitation. The protein concentration of the DIGs was evaluated with the Bio-Rad protein assay kit using bovine serum albumin as a protein standard. Proteins blotted onto polyvinylidene difluoride membrane were detected by incubation with primary antibodies overnight at 4 °C. The blots were incubated with a second anti-species antibody conjugated with HRP for 30–60 min at room temperature. The blots were developed with ECL reagents according to the manufacturer's instructions. Membranes were stripped with 100 mm glycine, pH 2, for 30 min, blocked, and reprobed with antibodies. Isolated DIGs were resuspended in 0.5 ml of lysis buffer (50 mm Tris/HCl, pH 7.4, 150 mm NaCl, 5 mm EDTA, 1 mm Na3VO4, 1 mm NaF, 1 mm phenylmethylsulfonyl fluoride, and 1% Nonidet P-40), incubated with antibodies and Protein A-Sepharose and washed under stringent conditions in radioimmune precipitation buffer (50 mm Tris/HCl, pH 7.4, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 1 mmdithioerythritol, 100 μm Na3VO4, 10 mm NaF) and once in 50 mm Tris/HCl, pH 7.4, 100 μm Na3VO4, 10 mmNaF. The immunoprecipitate was subjected to a [γ-32P]ATP in vitro kinase assay. To identify associated kinases, some samples were subjected to a second round of immunoprecipitation; immune complexes were denatured and dissociated by two sequential incubations in 50 μl of 50 mm Tris/HCl, pH 7.4, 0.5% SDS, and 1% β-mercaptoethanol for 10 min at 95 °C followed by centrifugation. Supernatants eluted from the Sepharose beads were diluted in 500 μl of lysis buffer and incubated with Fyn antibodies and Protein A-Sepharose for 1 h at room temperature. All samples were analyzed by SDS-PAGE, and radioactive protein bands were detected with a phosphoimager or autoradiography. DIGs that had been subjected to a [γ-32P]ATP kinase assay were immunoprecipitated with antibodies against Fyn and Lyn according to the same procedure, but SDS-PAGE analysis was performed directly after the precipitation. For immunoprecipitation from bottom fractions of the gradients, 250 μl of the fraction was diluted with an equal volume of lysis buffer and processed similarly to the DIG fractions above. Immune complexes were resuspended in 20 μl of kinase buffer (20 mm HEPES, pH 7.4, 5 mm MgCl2, 1 mm MnCl2, 100 μm Na3VO4) and incubated with 5 μCi of [γ-32P]ATP for 20 min at room temperature. The samples were washed and subjected either to SDS-PAGE or a second round of immunoprecipitation to identify associated kinases. DIGs were isolated from sucrose density gradients, resuspended in 50 μl of kinase buffer, and incubated with 10 μCi of [γ-32P]ATP for 20 min at room temperature. Kinase assays were analyzed by SDS-PAGE and autoradiography or subjected to immunoprecipitation. Optical densities from linear exposures of autoradiograms and Western blots developed using ECL were measured using the UltraScan XL Laser Densitometer (Amersham Pharmacia Biotech) and the GelScan XL software. Kinase activities were expressed as a function of tyrosine phosphorylation in relation to the total amount of the kinase. Oli-neu cells (1–2 × 107) were induced to differentiate by incubation with 1 mm dbcAMP for 3–4 days (34Jung M. Krämer E.M. Grzenkowski M. Tang K. Blakemore W. Aguzzi A. Khazaie K. Chlichlia K. von Blankenfeld G. Kettenmann H. Trotter J. Eur. J. Neurosci. 1995; 7: 1245-1265Crossref PubMed Scopus (218) Google Scholar). The cells were washed twice with ice-cold Tris-buffered saline and incubated for 1 h at 4 °C with the monoclonal antibody 27-11-111, which reacts with F3 (32Brümmendorf T. Hubert M. Treubert U. Leuschner R. Tarnok A. Rathjen F.G. Neuron. 1993; 10: 711-727Abstract Full Text PDF PubMed Scopus (176) Google Scholar). The cells were washed three times with ice-cold Tris-buffered saline and incubated with rabbit anti-mouse IgG for 30 min at 4 °C. In control samples, cells were incubated with the monoclonal AN2 antibody (31Niehaus A. Stegmüller J. Diers-Fenger M. Trotter J. J. Neurosci. 1999; 19: 4948-4961Crossref PubMed Google Scholar), followed by rabbit anti-rat IgG, or with primary or secondary antibodies alone. Subsequently, the dishes were transferred to 37 °C for 0, 3, 5, and 10 min (controls were left for 5 min at 37 °C). Following incubation at 37 °C, the cells were immediately lysed in 1.5 ml of extraction buffer, and the extracts were subjected to sucrose density gradient centrifugation as described above. DIGs were isolated from sucrose density gradients, subjected to SDS-PAGE, and analyzed by immunoblotting for tyrosine-phosphorylated Fyn and total Fyn protein with specific antibodies. To examine the localization of GPI-anchored proteins in DIGs during myelination, primary oligodendrocytes or myelin prepared from mice at defined developmental stages between postnatal day 9 (P9) and adult were analyzed by detergent extraction and sucrose density gradient centrifugation. Equal amounts of total myelin protein from each age examined were applied to the gradients. Western blot analysis of the gradient fractions with antibodies recognizing NCAM or F3 showed that a fraction of both GPI-anchored NCAM 120 and F3 float at low densities in the gradient (Fig. 1 A). Transmembrane isoforms of NCAM (NCAM 140 and 180) localize exclusively at the bottom of the gradient in high density fractions. Lipid analysis showed that the low density fractions are enriched in the typical myelin glycosphingolipids galactocerebroside and sulfatide as well as cholesterol (16Krämer E.-M. Koch T. Niehaus A. Trotter J. J. Biol. Chem. 1997; 272: 8937-8945Abstract Full Text Full Text PDF PubMed Scopus (113) Google Scholar). Flotation of GPI-anchored proteins in the gradient shows their specific association with DIGs. DIGs from cultured oligodendrocytes and myelin of P9/10 and P12 mice are distributed in gradient fractions between 18 and 25% sucrose (fractions 4–9). With increasing age of the animals and thus progressing myelination, the myelin DIGs focus at a density of 17–18% sucrose (fractions 8 and 9). The GPI-anchored proteins are enriched in DIGs and in myelin from P45, and they are localized exclusively in DIGs. Western blot analysis of the gradient fractions with antibodies against the Src family tyrosine kinases Fyn (p59) and Lyn (p53/56) showed that both kinases co-localize with NCAM 120 and F3 in DIG fractions (Fig.1 B). In gradients from cultured primary oligodendrocytes and myelin from P9/10 and P12 animals both kinases were found in bottom gradient fractions as well as in DIG fractions, in a distribution similar to that of the GPI-anchored proteins. The expression of Lyn kinase in myelin declined after P12, whereas Fyn was still expressed in adult myelin. With progressing myelination, Fyn kinase became progressively associated with DIG fractions, and in myelin from P30 up to adult it localized exclusively to DIGs, in a similar fashion to the shift in localization of the GPI-anchored proteins. Immunoprecipitation of phosphoproteins from the [γ-32P]ATP kinase assays of DIGs with antibodies directed against the Src family kinases Fyn and Lyn showed that both participate in the kinase reactions and are phosphorylated. The phosphorylation of both these kinases in DIGs is down-regulated with ongoing myelination (Fig.2 A). The strongest signal for32P-phosphorylated Fyn was obtained in precipitates of Fyn from oligodendrocyte DIGs. The 32P phosphorylation of Fyn in DIGs from myelin continuously declined between P9/10 and P30 and was virtually absent in DIGs from myelin of P45 and adult mice. High amounts of 32P-phosphorylated Lyn were precipitated from DIGs from oligodendrocytes, whereas only very weak 32P signals were obtained from DIGs of myelin from P9/10 and P12 mice. To investigate whether the reduction in phosphorylation of the kinases is due to a reduction in expression of the respective proteins, the samples were analyzed by Western blotting. Significant levels of Fyn kinase were present in myelin DIGs throughout development until adult (Fig. 2 B). In contrast, expression of Lyn kinase was observed in myelin DIGs up to P12 and was thereafter absent. The phosphorylation of Src family kinases is in most cases due to autophosphorylation and can be taken as a measure of the enzymatic activity (38Thomas S.M. Brugge J.S. Annu. Rev. Cell Dev. Biol. 1997; 13: 513-609Crossref PubMed Scopus (2200) Google Scholar). We measured the optical densities of the autoradiograms showing the 32P phosphorylation as well as the optical densities of the signals from the Western blots showing the total amount of the kinases and expressed the activity of each kinase as relative phosphorylation per unit of protein (Fig. 2 C). The kinase activity of Fyn is highest in DIGs from oligodendrocytes and in DIGs from myelin of P9/10 mice, at a time point where in vivo myelination is commencing. With ongoing myelination, the Fyn activity localized in DIGs is rapidly down-regulated. Lyn kinase activity is maximal in DIGs from cultured oligodendrocytes and weak in DIGs from myelin of P9/10 and P12 mice. We next asked whether the co-localization of the GPI-anchored adhesion receptors and the Src kinases on sucrose density gradients reflects an association between these molecules. We isolated DIGs from oligodendrocyte extracts, used them as a source for immunoprecipitation of NCAM or F3, and performed a [γ-32P]-ATP in vitro kinase assay on the immunoprecipitate. The bottom fractions of the gradient were also isolated and similarly subjected to immunoprecipitation and kinase assay. Phosphorylated proteins associated with the immunoprecipitate were separated by SDS-PAGE and visualized by autoradiography. In contrast to the multiple signals seen when the entire DIG proteins were subjected to a kinase assay (data not shown), a dominant phosphorylated protein of 59 kDa was associated with the immunoprecipitated NCAM 120 and F3 in DIGs (Fig.3 A, lane 5, and Fig. 3 B, lane 5). In both cases, this 59-kDa phosphoprotein was identified as Fyn kinase by reprecipitation with specific antibodies (Fig. 3, A,lane 6, and B, lane 6). No association with Lyn kinase was observed (data not shown). Additional signals of higher molecular weight were observed in the autoradiograph of the kinase assay on the NCAM 120 immunoprecipitate, but these were weak in comparison with the 59-kDa Fyn signal. In the case of the kinase assay on the F3 immunoprecipitate, additional signa
Referência(s)