Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Novel Regulation of Chondroitin Sulfate Glycosaminoglycan Modification of Amyloid Precursor Protein and Its Homologue, APLP2

1995; Elsevier BV; Volume: 270; Issue: 28 Linguagem: Inglês

10.1074/jbc.270.28.16522

ISSN

1083-351X

Autores

Gopal Thinakaran, Hilda H. Slunt, Sangram S. Sisodia,

Tópico(s)

Connective tissue disorders research

Resumo

Alzheimer's disease is characterized by the presence of parenchymal and cerebrovascular deposits of β-amyloid (Aβ). Aβ is derived from larger amyloid precursor proteins (APP), a member of a family of related polypeptides that includes amyloid precursor-like proteins, APLP1 and APLP2. APP and APLP2 isoforms are encoded by several alternatively spliced APP and APLP2 transcripts, respectively. We previously reported that the APLP2-751 isoform is modified by the addition of chondroitin sulfate glycosminoglycan (CS GAG) at Ser-614. In this report, we demonstrate that the APLP2-763 isoform, which contains an insertion of 12 amino acids immediately N-terminal to Ser-614, is not modified by CS GAG. Finally, we demonstrate that like APLP2-751, APP isoforms that lack sequences encoded by exon 15 (L-APP) are also modified by CS GAG, whereas APP forms containing exon 15 are not. We suggest that CS GAG modification of a subset of APP and APLP2 isoforms represents a means of generating functional diversity for these polypeptides. Alzheimer's disease is characterized by the presence of parenchymal and cerebrovascular deposits of β-amyloid (Aβ). Aβ is derived from larger amyloid precursor proteins (APP), a member of a family of related polypeptides that includes amyloid precursor-like proteins, APLP1 and APLP2. APP and APLP2 isoforms are encoded by several alternatively spliced APP and APLP2 transcripts, respectively. We previously reported that the APLP2-751 isoform is modified by the addition of chondroitin sulfate glycosminoglycan (CS GAG) at Ser-614. In this report, we demonstrate that the APLP2-763 isoform, which contains an insertion of 12 amino acids immediately N-terminal to Ser-614, is not modified by CS GAG. Finally, we demonstrate that like APLP2-751, APP isoforms that lack sequences encoded by exon 15 (L-APP) are also modified by CS GAG, whereas APP forms containing exon 15 are not. We suggest that CS GAG modification of a subset of APP and APLP2 isoforms represents a means of generating functional diversity for these polypeptides. Alzheimer's disease is pathologically characterized by the deposition of β-amyloid (Aβ)1 1The abbreviations used are: Aββ-amyloidAPLPamyloid precursor-like proteinAPPamyloid precursor proteinCSchondroitin sulfateGAGglycosaminoglycanL-APPAPP lacking exon 15PCRpolymerase chain reaction. in senile plaques and cerebrovascular vessels(1Glenner G.G. Wong C.W. Biochem. Biophys. Res. Commun. 1984; 120: 885-890Crossref PubMed Scopus (4379) Google Scholar, 2Masters C.L. Simms G. Weinman N.A. Multhaup G. McDonald B.L. Beyreuther K. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 4245-4249Crossref PubMed Scopus (3791) Google Scholar). Aβ, an ~4-kDa peptide, is derived from larger integral membrane glycoproteins, termed amyloid precursor proteins (APP) (3Kang J. Lemaire H.-G. Unterbeck A. Salbaum J.M. Masters C.L. Grzeschik K.-H. Multhaup G. Beyreuther K. Müller-Hill B. Nature. 1987; 325: 733-736Crossref PubMed Scopus (4067) Google Scholar) encoded by several alternatively spliced transcripts(4Kitaguchi N. Takahashi Y. Tokushima Y. Shiojiri S. Ito H. Nature. 1988; 331: 530-532Crossref PubMed Scopus (935) Google Scholar, 5Ponte P. Gonzalez-DeWhitt P. Schilling J. Miller J. Hsu D. Greenberg B. Davis K. Wallace W. Lieberburg I. Fuller F. Cordell B. Nature. 1988; 331: 525-532Crossref PubMed Scopus (904) Google Scholar, 6Tanzi R.E. McClatchey A.I. Lampert E.D. Villa-Komaroff L. Gusella J.F. Neve R.L. Nature. 1988; 331: 528-530Crossref PubMed Scopus (915) Google Scholar). APP lacking sequences encoded by exon 15, termed L-APP, are expressed in leukocytes, microglial cells, and several nonneuronal tissues(7König G. Mönning U. Czech C. Prior R. Banati R. Schreiter-Gasser U. Bauer J. Masters C.L. Beyreuther K. J. Biol. Chem. 1992; 267: 10804-10809Abstract Full Text PDF PubMed Google Scholar, 8Banati R.B. Gehrmann J. Schubert P. Kreutzberg G.W. Glia. 1993; 7: 111-118Crossref PubMed Scopus (950) Google Scholar, 9Ohgami T. Kitamoto T. Tateishi J. Mol. Brain Res. 1993; 20: 240-244Crossref PubMed Scopus (10) Google Scholar, 10Ohgami T. Kitamoto T. Tateishi J. J. Neurochem. 1993; 61: 1553-1556Crossref PubMed Scopus (7) Google Scholar, 11Sandbrink R. Masters C.L. Beyreuther K. J. Biol. Chem. 1994; 269: 1510-1517Abstract Full Text PDF PubMed Google Scholar). β-amyloid amyloid precursor-like protein amyloid precursor protein chondroitin sulfate glycosaminoglycan APP lacking exon 15 polymerase chain reaction. APP is a member of a larger family of proteins that includes amyloid precursor-like proteins, termed APLP1 (12Wasco W. Bupp K. Magendantz M. Gusella J.F. Tanzi R.E. Solomon F. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 10758-10762Crossref PubMed Scopus (324) Google Scholar) and APLP2(13Sprecher C.A. Grant F.J. Grimm G. O'Hara P.J. Norris F. Norris K. Foster D.C. Biochemistry. 1993; 32: 4481-4486Crossref PubMed Scopus (165) Google Scholar, 14Wasco W. Gurubhagavatula S. Paradis M.D. Romano D.M. Sisodia S.S. Hyman B.T. Neve R.L. Tanzi R.E. Nature Genet. 1993; 5: 95-99Crossref PubMed Scopus (327) Google Scholar, 15Slunt H.H. Thinakaran G. von Koch C. Lo A.C.Y. Tanzi R.E. Sisodia S.S. J. Biol. Chem. 1994; 269: 2637-2644Abstract Full Text PDF PubMed Google Scholar). Like APP pre-mRNA, the primary APLP2 transcript undergoes alternative splicing, and resultant transcripts encode APLP2 isoforms containing a Kunitz protease inhibitory domain (16Petersen L.C. Bjorn S.E. Norris F. Norris K. Sprecher C. Foster D.C. FEBS Lett. 1994; 338: 53-57Crossref PubMed Scopus (22) Google Scholar, 17Van Nostrand W.E. Schmaier A.H. Neiditch B.R. Siegel R.S. Raschke W.C. Sisodia S.S. Wagner S.L. Biochim. Biophys. Acta. 1994; 1209: 165-170Crossref PubMed Scopus (34) Google Scholar) (APLP2-751 and APLP2-763) or a 12-amino acid peptide of unknown function (APLP2-706 and APLP2-763) (13-15, 18). In earlier efforts, we documented that in cultured cells, APLP2-751 matures through a secretory/cleavage pathway similar to that utilized by APP(15Slunt H.H. Thinakaran G. von Koch C. Lo A.C.Y. Tanzi R.E. Sisodia S.S. J. Biol. Chem. 1994; 269: 2637-2644Abstract Full Text PDF PubMed Google Scholar, 19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). Moreover, a chondroitin sulfate glycosaminoglycan (CS GAG) chain is added to Ser-614 of APLP2-751(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). The sequence flanking Ser-614 (ENEGSGMA) is remarkably similar to a previously described recognition signal for CS GAG attachment(20Bourdon M.A. Krusius T. Campbell S. Schwartz N.B. Ruoslahti E. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3194-3198Crossref PubMed Scopus (207) Google Scholar), i.e. S-G-X-G/A flanked by acidic amino acids. In the APLP2-706 and APLP2-763 isoforms, however, a 12-amino acid peptide, presumably encoded by an alternatively spliced exon, is inserted between Glu-612 and Gly-613. We have examined the metabolism of the APLP2-763 isoform, and now report that the 12 amino acid peptide abolishes CS GAG modification of the core APLP2. Thus, our data are consistent with the idea that the level of CS GAG modification of the APLP2 core protein can be modulated by insertion of sequences encoded by an alternatively spliced exon. To assess the generality of the novel mechanism of regulation of CS GAG modification in members of the APP/APLP family, we examined the levels of CS GAG modification of APP isoforms encoded by alternatively spliced APP mRNAs. Indeed, earlier reports indicated that APP is a substrate for modification by CS GAG in C6 glioma cells(21Shioi J. Anderson J.P. Ripellino J.A. Robakis N.K. J. Biol. Chem. 1992; 267: 13819-13822Abstract Full Text PDF PubMed Google Scholar, 22Shioi J. Refolo L.M. Efthimiopoulos S. Robakis N.K. Neurosci. Lett. 1993; 154: 121-124Crossref PubMed Scopus (21) Google Scholar), despite our failure to demonstrate CS GAG modification of transiently expressed APP-695 or −770 isoforms in Chinese hamster ovary and COS-1 cells(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). L-APP isoforms, which lack 18 amino acids encoded by exon 15, contain the sequence ENEGSGL(7König G. Mönning U. Czech C. Prior R. Banati R. Schreiter-Gasser U. Bauer J. Masters C.L. Beyreuther K. J. Biol. Chem. 1992; 267: 10804-10809Abstract Full Text PDF PubMed Google Scholar); this sequence is generated by a fusion of sequences encoded by exons 14 and 16 and is surprisingly similar to sequences flanking the CS GAG attachment site of APLP2-751 (see above). In the present report, we demonstrate that like APLP2-751, L-APP isoforms are also substrates for CS GAG addition. The molecular mechanisms involved in regulating the levels of alternatively spliced transcripts which encode APP/APLP2 isoforms is unknown. Furthermore, the biochemical mechanism(s) by which exon 15 of APP-770/751/695 and the 12-amino acid insert in APLP2-763/706 disrupt CS GAG modification of respective core proteins have not been fully clarified. Nevertheless, we suggest that CS GAG addition represents a cell or developmental-specific mechanism to generate additional functional diversity for each polypeptide. The following oligonucleotide primers were used in this study: SD2, 5'-CCGCTCGAGGAACAGCGAGCG-3' (encodes residues 567-570 of mouse APLP2); AsD2myc, 5'-CCGTCTAGATTAATTCAAGTCCTCTTCAGAAATGAGCTTTT GCTCCATAATCTGCATCTGCTCCAGG-3' (complementary to sequences encoding the C-terminal 6 amino acids of APLP2 and 12 amino acids (MEQKLISEEDLN) of c-Myc); SAPP, 5'-CCGCTCGAGATCAGTTACGGA-3' (encodes amino acids 586-589 of human APP-770); AsAPPmyc, 5'-CC-GTCTAGATTAATTCAAGTCCTCTTCAGAAATG AGCTTTTGCTCCA-TGTTCTGCATCTGCTC-3' (complementary to sequences encoding the C-terminal 5 amino acids of APP and 12 amino acids (MEQKLISEEDLN) of c-Myc). Sequences encoding the C-terminal 196 amino acids of mouse APLP2-763 were amplified by polymerase chain reaction (PCR) from reverse-transcribed E16 embryonic mouse head RNA using the primer pairs SD2 and AsD2myc. The resulting PCR product was digested with XhoII and XbaI and ligated to pSVAPLP2myc(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar), previously digested with XhoII and XbaI, to generate plasmid pSVAPLP2-763myc. The primer pair SAPP and AsAPPmyc was used in PCR to amplify cDNA fragments encoding the C terminus of APP and L-APP polypeptides using the indicated DNA as template: plasmid pEFmo695 (provided by Dr. David Borchelt, The Johns Hopkins University School of Medicine, Baltimore, MD) for the C-terminal 184 amino acids of mouse APP and reverse-transcribed mouse lung RNA for the C-terminal 166 amino acids of mouse L-APP. The PCR products were digested with XhoI and XbaI and ligated to a vector fragment generated by digesting plasmid p770SP (23Sisodia S.S. Koo E.H. Beyreuther K. Unterbeck A. Price D.L. Science. 1990; 248: 492-495Crossref PubMed Scopus (747) Google Scholar) with XhoI and XbaI. The resulting plasmids encode polypeptides in which the N-terminal 380 amino acids of human APP-770 are fused to the C terminus of mouse APP (770ΔmAPP) or mouse L-APP (770ΔmL-APP). COS-1 cells were transfected using a high efficiency calcium phosphate method(24Chen C. Okayama H. Mol. Cell. Biol. 1987; 7: 2745-2752Crossref PubMed Scopus (5153) Google Scholar). Cells were labeled 36 h after transfection with 250 μCi/ml [35S]methionine for 3 h in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 1% dialyzed fetal bovine serum. Detergent lysates were prepared by lysing the cells in immunoprecipitation buffer(23Sisodia S.S. Koo E.H. Beyreuther K. Unterbeck A. Price D.L. Science. 1990; 248: 492-495Crossref PubMed Scopus (747) Google Scholar). Exogenous Myc epitope-tagged APLP2 and APP polypeptides were immunoprecipitated from cell lysates using polyclonal Myc-I antibody(25Lee M.K. Xu Z. Wong P.C. Cleveland D.W. J. Cell Biol. 1993; 122: 1337-1350Crossref PubMed Scopus (321) Google Scholar). Soluble APLP2 derivatives were immunoprecipitated from culture medium using polyclonal antibody D2-I raised against baculovirus-synthesized mouse APLP2-751(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). Soluble chimeric APP-APLP2 derivatives were immunoprecipitated from culture medium with monoclonal antibody P2-1 which recognizes an epitope in the APP ectodomain(26Van Nostrand W.E. Wagner S.L. Suzuki M. Choi B.H. Farrow J.S. Geddes J.W. Cotman C.W. Cunningham D.D. Nature. 1989; 341: 546-549Crossref PubMed Scopus (370) Google Scholar). Immune complexes were collected with protein A-agarose (Pierce). Chondroitinase enzyme digestions of immobilized immunoprecipitates were carried out as described(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). In previous studies, we reported that APLP2-751 is modified by the addition of CS GAG at a single serine residue (Ser-614)(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). Interestingly, the APLP2-763 isoform contains an additional 12-amino acid peptide inserted two amino acids N-terminal to the CS GAG addition site. In order to assess the effect of the 12-amino acid peptide on CS GAG attachment, we transiently transfected cDNA encoding a Myc epitope-tagged mouse APLP2-763 (pSVAPLP2-763myc) into COS-1 cells in parallel with cDNA encoding Myc-tagged mouse APLP2-751 (pSVAPLP2-751). Transfected cells were biosynthetically labeled with [35S]methionine, and radiolabeled mouse APLP2-related molecules were immunoprecipitated from cell lysates with Myc epitope-specific polyclonal antiserum Myc-I. The levels of CS GAG modification was assessed by digestion of the Myc-I immunoprecipitates with chondroitinase ABC prior to electrophoresis on SDS-polyacrylamide gels. APLP2-751 appears as an immature ~120-kDa polypeptide and a heterogeneous population of molecules with a mobility of ~120-200 kDa which result from the addition of variable length CS GAG to the core protein (Ref. 19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar; Fig. 1B, lane 3). Chondroitinase ABC digestion of APLP2-751 resulted in the conversion of the heterogeneous set of polypeptides to a species of ~130 kDa (Fig. 1B, lane 4). In contrast, APLP2-763 appears as an ~118-kDa molecule and a mature, chondroitinase ABC-insensitive polypeptide of ~130 kDa (Fig. 1 B, compare lanes 5 and 6). The latter result was indistinguishable from the behavior of APLP2-751S/A, a molecule that harbors a serine-alanine substitution at Ser-614 and which fails to be modified by CS GAG(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). Notably, the immature form of the APLP2-763 molecule exhibits somewhat accelerated migration relative to the immature form of the shorter APLP2-751 molecule. This deviation from the expected electrophoretic mobility was also observed when newly synthesized APLP2-763 molecules were immunopreciptated from lysates of cells which were pulse-labeled for 5 min (not shown). The reason for this discrepancy is unclear but appears to be dependent of the presence of the APLP2 transmembrane and cytoplasmic segments, since the cognate secreted derivatives migrated as expected (see below, Fig. 1B, compare lanes 8 and 9). As we have demonstrated previously, cells expressing APLP2-751 secreted APLP2 derivatives of ~100-150 kDa, generated following endoproteolytic cleavage of the ~120-200-kDa cell-associated species (Fig. 1B, lane 8). On the other hand, cells expressing APLP2-763 secreted a fairly homogeneous species of ~110 kDa (Fig. 1B, lane 9) which is likely generated following cleavage of the ~130-kDa cell-associated form. The cellular site and sequence requirements for the endoproteolytic cleavage of APLP2-751 and APLP2-763 remain to be determined. In any event, our studies clearly document that the 12-amino acid peptide inhibits CS GAG modification of the APLP2-763 isoform. The mechanism by which the 12-amino acid insert inhibits CS GAG modification of APLP2-763 is not fully clear. In order to assess the requirement for the presence of acidic residues upstream of the CS GAG attachment site on CS GAG addition, we examined the maturation of APLP2-751 in which glutamate residues Glu-610 and Glu-612 were substituted to glutamine. We have demonstrated that the latter molecules fail to be modified by CS GAG (data not shown), a finding that suggests that one, or both, of the acidic residues N-terminal to the Ser-Gly dipeptide is (are) essential for efficient CS GAG modification of APLP2-751. Hence, we propose that one potential mechanism by which the 12-amino acid insert inhibits CS GAG addition is by displacing the acidic residues from the CS GAG attachment site, thus altering the optimal xylosyltransferase recognition sequence(20Bourdon M.A. Krusius T. Campbell S. Schwartz N.B. Ruoslahti E. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3194-3198Crossref PubMed Scopus (207) Google Scholar). However, several caveats to the latter interpretation remain; it is presently uncertain whether presence of the two upstream acidic residues are the sole requirement for efficient CS GAG addition of APLP2-763; confirmation awaits a systematic analysis of a spectrum of APLP2 substrates harboring additional site-directed mutations surrounding the CS GAG attachment site or by analysis of APLP2-763 with dominant mutations that restore CS GAG addition. Moreover, it is not presently clear whether the 12-amino acid peptide inhibits CS GAG addition at the level of CS GAG chain initiation or elongation. Earlier studies demonstrated that APP expressed in C6 glioma cells were modified by CS GAG(21Shioi J. Anderson J.P. Ripellino J.A. Robakis N.K. J. Biol. Chem. 1992; 267: 13819-13822Abstract Full Text PDF PubMed Google Scholar, 22Shioi J. Refolo L.M. Efthimiopoulos S. Robakis N.K. Neurosci. Lett. 1993; 154: 121-124Crossref PubMed Scopus (21) Google Scholar). However, transiently expressed APP-695, −751, or −770 do not appear to be modified by CS GAG in human 293(27Oltersdorf T. Fritz L.C. Schenk D.B. Lieberburg I. Johnson-Wood K.L. Beattie E.C. Ward P.J. Blacher R.W. Dovey H.F. Sinha S. Nature. 1989; 341: 144-147Crossref PubMed Scopus (394) Google Scholar, 28Hung A.Y. Selkoe D.J. EMBO J. 1994; 13: 534-542Crossref PubMed Scopus (114) Google Scholar), COS-1, or Chinese hamster ovary cells(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar, 23Sisodia S.S. Koo E.H. Beyreuther K. Unterbeck A. Price D.L. Science. 1990; 248: 492-495Crossref PubMed Scopus (747) Google Scholar), although heterogeneous [35S]sulfate-labeled APP-related species have been described in PC12 cells(29Weidemann A. König G. Bunke D. Fischer P. Salbaum J.M. Masters C.L. Beyreuther K. Cell. 1989; 57: 115-126Abstract Full Text PDF PubMed Scopus (1084) Google Scholar). However, little information is available about the metabolism of the L-APP isoforms that lack 18 amino acids encoded by exon 15(7König G. Mönning U. Czech C. Prior R. Banati R. Schreiter-Gasser U. Bauer J. Masters C.L. Beyreuther K. J. Biol. Chem. 1992; 267: 10804-10809Abstract Full Text PDF PubMed Google Scholar). Notably, the L-APP isoforms contain a sequence ENEGSGL, generated by fusing exons 14 and 16, which is strikingly similar to the region surrounding the CS GAG attachment site of APLP2-751; remarkably, comparison of a 111-amino acid region immediately N-terminal to the transmembrane domains of APP and APLP2 reveals that the only stretch of identity resides in the ENEGSG sequence (Fig. 2). In order to assess whether L-APP is modified by CS GAG, we transfected COS-1 cells with APP minigenes that encode polypeptides in which the N-terminal 380 amino acids of human APP-770 are fused to the C-terminal 184 amino acids of mouse APP (770ΔmAPP) or 166 amino acids of mouse L-APP (770ΔmL-APP) (Fig. 3A). 770ΔmAPP is synthesized as a polypeptide of ~80 kDa (Fig. 3B, lane 1) and matures to discrete species of ~100 and ~110 kDa (Fig. 3B, lane 3). However, 770ΔmL-APP appears as an immature species of ~85 kDa (Fig. 3B, lane 2) and matures to a heterogeneous set of polypeptides of ~150-200 kDa (Fig. 3B, lane 5). As expected, mature forms of 770ΔmAPP were insensitive to chondroitinase AC digestion (Fig. 3B, lane 4). However, chondroitinase AC converted the heterogeneous ~150-200-kDa 770ΔmL-APP-related species into a homogeneous polypeptide of ~105 kDa, confirming that the chimeric L-APP forms were modified by CS GAG (Fig. 3B, lane 6). Moreover, the vast majority of soluble 770ΔmL-APP derivatives were modified by CS GAG (Fig. 3B, compare lanes 9 and 10; phosphorimaging analysis revealed that >70% of the soluble forms of 770ΔmL-APP were modified), whereas the soluble derivatives of 770ΔmAPP were insensitive to digestion by chondroitinase AC (Fig. 3B, compare lanes 7 and 8). These results demonstrate that mouse L-APP isoforms are substrates for CS GAG attachment and that insertion of sequences encoded by exon 15 into the CS GAG recognition site significantly inhibits CS GAG modification of the precursor.Figure 3:L-APP is modified by the addition of CS GAG. A, structure of chimeric APP polypeptides. Human APP-770 residues 1-380 and mouse APP-770 residues 586-770 are in open box; TMD, transmembrane domain: KPI, Kunitz protease inhibitor domain (hatched box); solid box, C-terminal Myc epitope tag. B, analysis of chimeric APP and L-APP polypeptides in transfected COS-1 cells. Chimeric APP polypeptides were immunoprecipitated from lysates of cells labeled with [35S]methionine for 5 min (lanes 1 and 2) or 3 h (lanes 3-6); immunoprecipitates are from lysates of cells transfected with plasmids encoding 770ΔmAPP (lanes 1, 3, and 4) or 770ΔmLAPP (lanes 2, 5, and 6). Soluble derivatives of chimeric APP polypeptides were immunoprecipitated with monoclonal antibody P2-I from conditioned medium of cells transfected with plasmids encoding 770ΔmAPP (lanes 7 and 8) or 770ΔmLAPP (lanes 9 and 10). The immune complexes were incubated at 37°C without (-, lanes 3, 5, 7, 9) or with (+, lanes 4, 6, 8, 10) chondroitinase AC prior to electrophoresis. The CS GAG-modified L-APP immunoprecipitated from the cell lysate is marked by a bracket.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Aβ, the principal component of senile plaques of Alzheimer's disease, Down's syndrome, and to a lesser extent, normal aging is derived from APP. APP is a member of a larger family of integral membrane glycoproteins that includes APLP1 and APLP2(12Wasco W. Bupp K. Magendantz M. Gusella J.F. Tanzi R.E. Solomon F. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 10758-10762Crossref PubMed Scopus (324) Google Scholar, 13Sprecher C.A. Grant F.J. Grimm G. O'Hara P.J. Norris F. Norris K. Foster D.C. Biochemistry. 1993; 32: 4481-4486Crossref PubMed Scopus (165) Google Scholar, 14Wasco W. Gurubhagavatula S. Paradis M.D. Romano D.M. Sisodia S.S. Hyman B.T. Neve R.L. Tanzi R.E. Nature Genet. 1993; 5: 95-99Crossref PubMed Scopus (327) Google Scholar, 15Slunt H.H. Thinakaran G. von Koch C. Lo A.C.Y. Tanzi R.E. Sisodia S.S. J. Biol. Chem. 1994; 269: 2637-2644Abstract Full Text PDF PubMed Google Scholar). Several alternatively spliced APP and APLP2 mRNA have been described that encode a family of APP and APLP2 isoforms. We recently demonstrated that the APLP2-751 isoform is post-translationally modified by the addition of CS GAG at a single serine residue contained within the sequence ENEGSGMA(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar). To that initial observation, the present report provides two novel insights. First, we document that the APLP2-763 isoform, in which a 12-amino acid peptide is inserted two amino acids N-terminal to the CS GAG attachement site, is not modified by CS GAG. Second, we demonstrate that like APLP2-751, APP isoforms that lack sequences encoded by exon 15, termed L-APP, are also post-translationally modified by the addition of CS GAG. We suggest that L-APP are modified at Ser-638 (of L-APP 752) contained within a junctional sequence, ENEGSGLT, that is remarkably similar to sequences surrounding the CS GAG attachment site of APLP2-751. The latter result resolves the apparent discrepancy between studies which documented that APP is modified by CS GAG in C6 glioma cells (21Shioi J. Anderson J.P. Ripellino J.A. Robakis N.K. J. Biol. Chem. 1992; 267: 13819-13822Abstract Full Text PDF PubMed Google Scholar, 22Shioi J. Refolo L.M. Efthimiopoulos S. Robakis N.K. Neurosci. Lett. 1993; 154: 121-124Crossref PubMed Scopus (21) Google Scholar) and others which failed to demonstrate that human APP-695, −751, and −770 isoforms, when transiently overexpressed in mammalian cells, are substrates for CS GAG addition(19Thinakaran G. Sisodia S.S. J. Biol. Chem. 1994; 269: 22099-22104Abstract Full Text PDF PubMed Google Scholar, 23Sisodia S.S. Koo E.H. Beyreuther K. Unterbeck A. Price D.L. Science. 1990; 248: 492-495Crossref PubMed Scopus (747) Google Scholar, 27Oltersdorf T. Fritz L.C. Schenk D.B. Lieberburg I. Johnson-Wood K.L. Beattie E.C. Ward P.J. Blacher R.W. Dovey H.F. Sinha S. Nature. 1989; 341: 144-147Crossref PubMed Scopus (394) Google Scholar, 28Hung A.Y. Selkoe D.J. EMBO J. 1994; 13: 534-542Crossref PubMed Scopus (114) Google Scholar, 29Weidemann A. König G. Bunke D. Fischer P. Salbaum J.M. Masters C.L. Beyreuther K. Cell. 1989; 57: 115-126Abstract Full Text PDF PubMed Scopus (1084) Google Scholar). Notably, our recent RT-PCR analysis of C6 glioma cell mRNA reveal that approximately 35% of APP mRNA encodes L-APP.2 2G. Thinakaran and S. Sisodia, unpublished results. These latter results are consistent with earlier studies showing that L-APP is highly expressed in microglial cells(7König G. Mönning U. Czech C. Prior R. Banati R. Schreiter-Gasser U. Bauer J. Masters C.L. Beyreuther K. J. Biol. Chem. 1992; 267: 10804-10809Abstract Full Text PDF PubMed Google Scholar, 8Banati R.B. Gehrmann J. Schubert P. Kreutzberg G.W. Glia. 1993; 7: 111-118Crossref PubMed Scopus (950) Google Scholar). In any event, the sequence ENEGSG, contained within the most highly divergent domain between APP and APLP2, appears to be essential for efficient CS GAG modification of both APLP2 and APP. Moreover, CS GAG addition is effectively disrupted by insertion of sequences encoded by an alternatively spliced exon, two amino acids N-terminal to the CS GAG attachment site. To the best of our knowledge, this is the first demonstration that addition of CS GAG to core proteins can be modulated by sequences encoded by alternatively spliced exons. The generality of this novel mechanism of regulation of GAG addition to other proteoglycans awaits a systematic biochemical and molecular analysis of variants of each core protein encoded by alternatively spliced transcripts. Our recent studies of the expression of APLP2 in the mouse nervous system support the view that alternatively spliced APLP2 mRNAs might encode proteins which are functionally distinct.3 3Thinakaran, G., Roskams, A. J. I., Kitt, C. A., Slunt, H. H., Masliah, E., von Koch, C., Ginsberg, S. D., Ronnett, G. V., Reed, R. R., Price, D. L., and Sisodia, S. S. (1995) J. Neurosci., in press. For example, APLP2 isoforms containing the 12-amino acid insert are expressed in mature neuronal populations in cortex, hippocampus, and cerebellum (18Sandbrink R. Masters C.L. Beyreuther K. J. Biol. Chem. 1994; 269: 14227-14234Abstract Full Text PDF PubMed Google Scholar) which have established stable connections with respective targets. On the other hand, APLP2 isoforms lacking the insert are highly and selectively expressed in olfactory sensory neurons, the only neuronal population in the adult mammalian central nervous system which are continuously replaced3; axonal outgrowth from newly born sensory neurons to the olfactory bulb requires elaborate mechanisms to regulate guidance, adhesion and synaptogenesis. Thus, we argue that APLP2 isoforms which are differentially modified by CS GAG are utilized for functionally distinct aspects of neuronal biology. While our manuscript was in review, we learned of a report by Pangelos et al. (Pangelos, M. N., Efthimiopoulos, S., Shioi, J., and Robakis, N. K.(1995) J. Biol. Chem.270, 10388-10391) which demonstrated that APP lacking exon 15 is modified by CS GAGs. We thank Dr. Philip Wong (Neuropathology Laboratory, The Johns Hopkins University School of Medicine) and Dr. Steve Wagner (SIBIA, La Jolla, CA) for providing Myc-I and P2-1 antibodies, respectively; Frances Davenport for assistance with cell culture; Dr. Jeffrey D. Esko (University of Alabama at Birmingham, Birmingham, AL) for helpful suggestions; and Drs. David Borchelt and Donald Price (Neuropathology Laboratory, The Johns Hopkins University School of Medicine) for advice on the manuscript.

Referência(s)