Artigo Acesso aberto Revisado por pares

Purification of Apolipoprotein E Attenuates Isoform-specific Binding to β-Amyloid

1995; Elsevier BV; Volume: 270; Issue: 16 Linguagem: Inglês

10.1074/jbc.270.16.9039

ISSN

1083-351X

Autores

Mary Jo LaDu, Terry M. Pederson, Donald E. Frail, Catherine A. Reardon, Godfrey S. Getz, Michael T. Falduto,

Tópico(s)

Nuclear Receptors and Signaling

Resumo

Apolipoprotein E (apoE), particularly the e4 allele, is genetically linked to the incidence of Alzheimer' disease. In vitro, apoE has been shown to bind β-amyloid (Aβ), an amyloidogenic peptide that aggregates to form the primary component of senile plaques. In previous work, we demonstrated that apoE3 from tissue culture medium binds to Aβ with greater avidity than apoE4 (LaDu, M. J., Falduto, M. T., Manelli, A. M., Reardon, C. A., Getz, G. S., and Frail, D. E. (1994) J. Biol. Chem. 269, 23403-23406). This is in contrast to data using purified apoE isoforms as substrate for Aβ (Strittmatter, W. J., Weisgraber, K. H., Huang, D. Y., Dong, L.-M., Salvesen, G. S., Pericak-Vance, M., Schmechel, D., Saunders, A. M., Goldgaber, D., and Roses, A. D. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 8098-8102). Here we resolve this apparent discrepancy by demonstrating that the preferential binding of Aβ to apoE3 is attenuated and even abolished with purification, a process that includes delipidation and denaturation. We compared the Aβ binding capacity of unpurified apoE isoforms from both tissue culture medium and intact human very low density lipoproteins with that of apoE purified from these two sources. The interaction of human Aβ- (1Elshourbagy N.A. Liao W.S. Mahley R.W. Taylor J.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 203-207Google Scholar, 2Boyles J.K. Zoellner C.D. Anderson L.J. Kosik L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Haerter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Google Scholar, 3Snipes G.J. McGuire C.B. Norden J.J. Freeman J.A. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1130-1134Google Scholar, 4Saunders A.M. Strittmatter W.J. Schmechel D. George-Hyslop St., P.H. Pericak-Vance M.A. Joo S.H. Rosi B.L. Gusella J.F. Crapper-MacLachlan D.R. Alberts M.J. Hulette C. Crain B. Goldgaber D. Roses A.D. Neurology. 1993; 43: 1467-1472Google Scholar, 5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar, 6Wisniewski T. Frangione B. Neurosci. Lett. 1992; 135: 235-238Google Scholar, 7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar, 8Ghiso J. Matsubara E. Koudinov A. Choi-Miura N.H. Tomita M. Wisniewski T. Frangione B. Biochem. J. 1993; 293: 27-30Google Scholar, 9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar, 10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar, 11Hixson J.E. Vernier D.T. J. Lipid Res. 1990; 31: 545-548Google Scholar, 12Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 129: 145-166Google Scholar, 13Rall S.C. Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 128: 273-287Google Scholar, 14Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar, 15Kim K.S. Miller D.L. Sapienza V.J. Chen C.-M.J. Bai C. Grundke-Iqbal I. Currie J.R. Wisniewski H.M. Neurosci. Res. Commun. 1988; 2: 121-130Google Scholar, 16Innerarity T.L. Mahley R.W. Biochemistry. 1978; 17: 1440-1447Google Scholar, 17Innerarity T.L. Pitas R.E. Mahley R.W. Methods Enzymol. 1986; 129: 542-565Google Scholar, 18Kowal R.C. Herz J. Weisgraber K.H. Mahley R.W. Brown M.S. Goldstein J.L. J. Biol. Chem. 1990; 265: 10771-10779Google Scholar, 19Nathan B.P. Bellosta S. Sanan D.A. Weisgraber K.H. Mahley R.W. Pitas R.E. Science. 1994; 264: 850-852Google Scholar, 20Ma J. Wee A. Brewer H.B. Das S. Potter H. Nature. 1994; 372: 92-94Google Scholar, 21Sanan D.A. Weisgraber K.H. Russell S.J. Mahley R.W. Huang D. Saunders A. Schmechel D. Wisniewski T. Frangione B. Roses A.D. Strittmatter W.J. J. Clin. Invest. 1994; 94: 860-869Google Scholar, 22Wisniewski T. Castano E.M. Golabek A. Vogel T. Frangione B. Am. J. Pathol. 1994; 145: 1030-1035Google Scholar, 23Whitson J.S. Mims M.P. Strittmatter W.J. Yamaki T. Morrisett J.D. Appel S.H. Biochem. Biophys. Res. Commun. 1994; 199: 163-170Google Scholar, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40) -peptide and apoE was analyzed by nonreducing SDS-polyacrylamide gel electrophoresis followed by Western immunoblotting for either Aβ or apoE immunoreactivity. While the level of the apoE3⋅Aβ complex was ~20-fold greater compared with the apoE4⋅Aβ complex in unpurified conditioned medium, apoE3 and apoE4 purified from this medium bound to Aβ with comparable avidity. Moreover, using endogenous apoE on very low density lipoproteins from plasma of apoE3/3 and apoE4/4 homozygotes, apoE3 was again a better substrate for Aβ than apoE4. However, apoE purified from these plasma lipoproteins exhibited little isoform specificity in binding to Aβ. These results suggest that native preparations of apoE may be a more physiologically relevant substrate for Aβ binding than purified apoE and further underscore the importance of subtle differences in apoE conformation to its biological activity. Apolipoprotein E (apoE), particularly the e4 allele, is genetically linked to the incidence of Alzheimer' disease. In vitro, apoE has been shown to bind β-amyloid (Aβ), an amyloidogenic peptide that aggregates to form the primary component of senile plaques. In previous work, we demonstrated that apoE3 from tissue culture medium binds to Aβ with greater avidity than apoE4 (LaDu, M. J., Falduto, M. T., Manelli, A. M., Reardon, C. A., Getz, G. S., and Frail, D. E. (1994) J. Biol. Chem. 269, 23403-23406). This is in contrast to data using purified apoE isoforms as substrate for Aβ (Strittmatter, W. J., Weisgraber, K. H., Huang, D. Y., Dong, L.-M., Salvesen, G. S., Pericak-Vance, M., Schmechel, D., Saunders, A. M., Goldgaber, D., and Roses, A. D. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 8098-8102). Here we resolve this apparent discrepancy by demonstrating that the preferential binding of Aβ to apoE3 is attenuated and even abolished with purification, a process that includes delipidation and denaturation. We compared the Aβ binding capacity of unpurified apoE isoforms from both tissue culture medium and intact human very low density lipoproteins with that of apoE purified from these two sources. The interaction of human Aβ- (1Elshourbagy N.A. Liao W.S. Mahley R.W. Taylor J.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 203-207Google Scholar, 2Boyles J.K. Zoellner C.D. Anderson L.J. Kosik L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Haerter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Google Scholar, 3Snipes G.J. McGuire C.B. Norden J.J. Freeman J.A. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1130-1134Google Scholar, 4Saunders A.M. Strittmatter W.J. Schmechel D. George-Hyslop St., P.H. Pericak-Vance M.A. Joo S.H. Rosi B.L. Gusella J.F. Crapper-MacLachlan D.R. Alberts M.J. Hulette C. Crain B. Goldgaber D. Roses A.D. Neurology. 1993; 43: 1467-1472Google Scholar, 5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar, 6Wisniewski T. Frangione B. Neurosci. Lett. 1992; 135: 235-238Google Scholar, 7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar, 8Ghiso J. Matsubara E. Koudinov A. Choi-Miura N.H. Tomita M. Wisniewski T. Frangione B. Biochem. J. 1993; 293: 27-30Google Scholar, 9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar, 10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar, 11Hixson J.E. Vernier D.T. J. Lipid Res. 1990; 31: 545-548Google Scholar, 12Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 129: 145-166Google Scholar, 13Rall S.C. Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 128: 273-287Google Scholar, 14Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar, 15Kim K.S. Miller D.L. Sapienza V.J. Chen C.-M.J. Bai C. Grundke-Iqbal I. Currie J.R. Wisniewski H.M. Neurosci. Res. Commun. 1988; 2: 121-130Google Scholar, 16Innerarity T.L. Mahley R.W. Biochemistry. 1978; 17: 1440-1447Google Scholar, 17Innerarity T.L. Pitas R.E. Mahley R.W. Methods Enzymol. 1986; 129: 542-565Google Scholar, 18Kowal R.C. Herz J. Weisgraber K.H. Mahley R.W. Brown M.S. Goldstein J.L. J. Biol. Chem. 1990; 265: 10771-10779Google Scholar, 19Nathan B.P. Bellosta S. Sanan D.A. Weisgraber K.H. Mahley R.W. Pitas R.E. Science. 1994; 264: 850-852Google Scholar, 20Ma J. Wee A. Brewer H.B. Das S. Potter H. Nature. 1994; 372: 92-94Google Scholar, 21Sanan D.A. Weisgraber K.H. Russell S.J. Mahley R.W. Huang D. Saunders A. Schmechel D. Wisniewski T. Frangione B. Roses A.D. Strittmatter W.J. J. Clin. Invest. 1994; 94: 860-869Google Scholar, 22Wisniewski T. Castano E.M. Golabek A. Vogel T. Frangione B. Am. J. Pathol. 1994; 145: 1030-1035Google Scholar, 23Whitson J.S. Mims M.P. Strittmatter W.J. Yamaki T. Morrisett J.D. Appel S.H. Biochem. Biophys. Res. Commun. 1994; 199: 163-170Google Scholar, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40) -peptide and apoE was analyzed by nonreducing SDS-polyacrylamide gel electrophoresis followed by Western immunoblotting for either Aβ or apoE immunoreactivity. While the level of the apoE3⋅Aβ complex was ~20-fold greater compared with the apoE4⋅Aβ complex in unpurified conditioned medium, apoE3 and apoE4 purified from this medium bound to Aβ with comparable avidity. Moreover, using endogenous apoE on very low density lipoproteins from plasma of apoE3/3 and apoE4/4 homozygotes, apoE3 was again a better substrate for Aβ than apoE4. However, apoE purified from these plasma lipoproteins exhibited little isoform specificity in binding to Aβ. These results suggest that native preparations of apoE may be a more physiologically relevant substrate for Aβ binding than purified apoE and further underscore the importance of subtle differences in apoE conformation to its biological activity. INTRODUCTIONApolipoprotein E (apoE), 1The abbreviations used are: apoEapolipoprotein EADAlzheimer' diseaseAββ-amyloidVLDLvery low density lipoproteinTricineN-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycineLDLlow density lipoprotein. a component of several classes of lipoproteins, acts as a ligand for lipoprotein receptors, thus regulating lipid transport and clearance. ApoE is also expressed in the brain (1Elshourbagy N.A. Liao W.S. Mahley R.W. Taylor J.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 203-207Google Scholar) and in response to injury in both the peripheral (2Boyles J.K. Zoellner C.D. Anderson L.J. Kosik L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Haerter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Google Scholar) and central nervous (3Snipes G.J. McGuire C.B. Norden J.J. Freeman J.A. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1130-1134Google Scholar) systems. In humans, apoE has three major isoforms, E2 (Cys112, Cys158), E3 (Cys112, Arg158), and E4 (Arg1121, Arg158), which are products of three alleles at a single gene locus. The presence of cysteine in apoE2 and apoE3 allow these isoforms to form disulfide-linked dimers. Recently, it has been demonstrated that the apoE e4 allele is present with increased frequency in patients with sporadic (4Saunders A.M. Strittmatter W.J. Schmechel D. George-Hyslop St., P.H. Pericak-Vance M.A. Joo S.H. Rosi B.L. Gusella J.F. Crapper-MacLachlan D.R. Alberts M.J. Hulette C. Crain B. Goldgaber D. Roses A.D. Neurology. 1993; 43: 1467-1472Google Scholar) and late-onset familial (5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar) Alzheimer' disease (AD). Due primarily to this genetic linkage, the role of apoE in the pathogenesis of AD is being actively pursued.A major neuropathological feature of AD is the presence of extracellular senile plaques composed predominantly of aggregated β-amyloid peptide (Aβ). The physiological mechanism by which apoE contributes to AD pathology may be by means of an isoform-specific interaction with Aβ. ApoE and Aβ colocalize in senile plaques (6Wisniewski T. Frangione B. Neurosci. Lett. 1992; 135: 235-238Google Scholar), and synthetic Aβ peptides bind in vitro to apoE from tissue culture medium (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar) and cerebrospinal fluid (5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar, 8Ghiso J. Matsubara E. Koudinov A. Choi-Miura N.H. Tomita M. Wisniewski T. Frangione B. Biochem. J. 1993; 293: 27-30Google Scholar, 9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar) as well as to purified apoE (9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar, 10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar). However, studies involving apoE isoform specificity in binding to Aβ have been limited and contradictory. In previous work, we demonstrated that apoE3 from tissue culture medium binds to Aβ with greater avidity than apoE4 (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar). This is in contrast to data from Strittmatter et al. (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar), who used apoE isoforms purified from human plasma as substrate for Aβ. Here we resolve the apparent discrepancy between these studies by demonstrating that the previously observed preferential association of Aβ with apoE3 is attenuated or abolished by purification, a process that includes delipidation and denaturation.MATERIALS AND METHODSExpression of Human ApoE in Cultured CellsHuman apoE3 and apoE4 were expressed in HEK-293 cells stably transfected with human apoE3 or apoE4 (products of the e3 and e4 alleles, respectively) cDNA as described previously (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar). Conditioned medium containing apoE was concentrated (Centriprep, Amicon, Inc.) ~50-fold prior to binding reactions or purification.Purification of ApoEHuman plasma was screened for apoE genotype using a modification of the method of Hixson and Vernier (11Hixson J.E. Vernier D.T. J. Lipid Res. 1990; 31: 545-548Google Scholar). Intermediate and very low density lipoprotein (VLDL) particles (d < 1.02 g/ml) were isolated (12Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 129: 145-166Google Scholar) from the plasma of individuals homozygous for apoE3 and apoE4. Unpurified preparations of VLDL were used within 2 weeks of isolation. Purification of apoE from this lipoprotein fraction and from concentrated conditioned medium was carried out according to standard procedures (13Rall S.C. Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 128: 273-287Google Scholar). Briefly, the conditioned medium and lipoproteins were dialyzed against 0.01% EDTA, lyophilized, and delipidated in CHCl3:MeOH (2:1). Delipidated proteins were pelleted in MeOH and solubilized in 6 M guanidine, 0.1 M Tris, 0.01% EDTA (pH 7.4), and 1% 2-mercaptoethanol. Proteins were fractionated on a Sephacryl S-300 column (Pharmacia Biotech Inc.) equilibrated in 4 M guanidine, 0.1 M Tris, 0.01% EDTA (pH 7.4), and 0.1% 2-mercaptoethanol. Fractions containing apoE were dialyzed in 5 m M NH4HCO3, lyophilized, and resuspended in 0.1 M NH4HCO3. Unpurified and purified apoE proteins were quantified by SDS-polyacrylamide gel electrophoresis, protein staining, and densitometry (Molecular Dynamics, Inc.) of serial dilutions of apoE-containing samples using a purified apoE standard.ApoE⋅Aβ Complex Formation and DetectionFor binding reactions, synthetic human Aβ- (1Elshourbagy N.A. Liao W.S. Mahley R.W. Taylor J.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 203-207Google Scholar, 2Boyles J.K. Zoellner C.D. Anderson L.J. Kosik L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Haerter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Google Scholar, 3Snipes G.J. McGuire C.B. Norden J.J. Freeman J.A. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1130-1134Google Scholar, 4Saunders A.M. Strittmatter W.J. Schmechel D. George-Hyslop St., P.H. Pericak-Vance M.A. Joo S.H. Rosi B.L. Gusella J.F. Crapper-MacLachlan D.R. Alberts M.J. Hulette C. Crain B. Goldgaber D. Roses A.D. Neurology. 1993; 43: 1467-1472Google Scholar, 5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar, 6Wisniewski T. Frangione B. Neurosci. Lett. 1992; 135: 235-238Google Scholar, 7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar, 8Ghiso J. Matsubara E. Koudinov A. Choi-Miura N.H. Tomita M. Wisniewski T. Frangione B. Biochem. J. 1993; 293: 27-30Google Scholar, 9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar, 10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar, 11Hixson J.E. Vernier D.T. J. Lipid Res. 1990; 31: 545-548Google Scholar, 12Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 129: 145-166Google Scholar, 13Rall S.C. Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 128: 273-287Google Scholar, 14Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar, 15Kim K.S. Miller D.L. Sapienza V.J. Chen C.-M.J. Bai C. Grundke-Iqbal I. Currie J.R. Wisniewski H.M. Neurosci. Res. Commun. 1988; 2: 121-130Google Scholar, 16Innerarity T.L. Mahley R.W. Biochemistry. 1978; 17: 1440-1447Google Scholar, 17Innerarity T.L. Pitas R.E. Mahley R.W. Methods Enzymol. 1986; 129: 542-565Google Scholar, 18Kowal R.C. Herz J. Weisgraber K.H. Mahley R.W. Brown M.S. Goldstein J.L. J. Biol. Chem. 1990; 265: 10771-10779Google Scholar, 19Nathan B.P. Bellosta S. Sanan D.A. Weisgraber K.H. Mahley R.W. Pitas R.E. Science. 1994; 264: 850-852Google Scholar, 20Ma J. Wee A. Brewer H.B. Das S. Potter H. Nature. 1994; 372: 92-94Google Scholar, 21Sanan D.A. Weisgraber K.H. Russell S.J. Mahley R.W. Huang D. Saunders A. Schmechel D. Wisniewski T. Frangione B. Roses A.D. Strittmatter W.J. J. Clin. Invest. 1994; 94: 860-869Google Scholar, 22Wisniewski T. Castano E.M. Golabek A. Vogel T. Frangione B. Am. J. Pathol. 1994; 145: 1030-1035Google Scholar, 23Whitson J.S. Mims M.P. Strittmatter W.J. Yamaki T. Morrisett J.D. Appel S.H. Biochem. Biophys. Res. Commun. 1994; 199: 163-170Google Scholar, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40) -peptide, purified by high performance liquid chromatography, was resuspended to 5 m M in 100% Me2SO. ApoE (25 μg/ml, ~700 n M) was incubated for 2 h (except as noted in the legend to Fig. 4) at room temperature with 250 μM Aβ at pH 7.4 as described previously (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar). Control reactions without Aβ contained 5% Me2SO. Reactions were stopped by the addition of 2 × nonreducing Laemmli buffer (14Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar) (4% SDS, no 2-mercaptoethanol) and frozen at −20°C. Samples were boiled for 5 min, electrophoresed on 10-20% SDS-Tricine gels, transferred to Immobilon-P membranes (Millipore Corp.), and probed with antibodies to Aβ or apoE (1:1000 dilution). Monoclonal antibody 4G8 to amino acids 17-24 of Aβ was provided by Drs. H. M. Wisniewski and K. S. Kim (15Kim K.S. Miller D.L. Sapienza V.J. Chen C.-M.J. Bai C. Grundke-Iqbal I. Currie J.R. Wisniewski H.M. Neurosci. Res. Commun. 1988; 2: 121-130Google Scholar). ApoE antiserum was obtained by immunizing rabbits with apoE purified from human serum. Proteins on Western blots were visualized by enhanced chemiluminescence (Amersham Corp.) and quantified by densitometry.RESULTS AND DISCUSSIONGenetic data have defined the presence of the e4 allele of apoE as a major risk factor for the occurrence of sporadic (4Saunders A.M. Strittmatter W.J. Schmechel D. George-Hyslop St., P.H. Pericak-Vance M.A. Joo S.H. Rosi B.L. Gusella J.F. Crapper-MacLachlan D.R. Alberts M.J. Hulette C. Crain B. Goldgaber D. Roses A.D. Neurology. 1993; 43: 1467-1472Google Scholar) and late-onset familial (5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar) Alzheimer' disease. However, the physiological mechanism by which apoE isoform specificity contributes to the pathogenesis of the disease is unknown. Previously, using unpurified apoE from conditioned medium, we showed that the amount of the apoE3⋅Aβ complex was much greater compared with the apoE4⋅Aβ complex (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar). These results are in contrast to data published by Strittmatter et al. (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar). Using apoE purified from human plasma, they reported that apoE4 binds Aβ more rapidly than apoE3. However, the level of Aβ binding to apoE3 and apoE4 after several hours of incubation was comparable. The major differences between these two studies are the source of apoE (plasma versus secreted by cultured cells) and whether the apoE protein was purified prior to use in the binding assays. In this study, we examined whether these differences could account for the discrepancy between these two reports regarding apoE isoform-specific binding to Aβ. We compared the Aβ binding capacity of unpurified apoE from both conditioned medium and intact human VLDL from plasma of apoE3/3 and apoE4/4 homozygotes with the Aβ binding capacity of apoE purified from these two sources. The purification procedure (13Rall S.C. Weisgraber K.H. Mahley R.W. Methods Enzymol. 1986; 128: 273-287Google Scholar) used was the same as that used by Strittmatter et al. (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar).We first examined the effect of purification of apoE from conditioned medium on the formation of apoE⋅Aβ complexes. ApoE3 and apoE4 were purified from the medium of transfected HEK-293 cells, and the binding of this material to Aβ was compared with the binding of unpurified apoE from medium (Fig. 1). Prior to purification, the amount of the apoE3⋅Aβ complex (Fig. 1, A and B, lane 2) was ~20-fold greater compared with the apoE4⋅Aβ complex (lane 4). These results are consistent with our previous data that demonstrated that this preferential binding of Aβ to apoE3 is maintained over time, pH range, and concentration of apoE and Aβ (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar). Purification of the apoE isoforms from this medium resulted in a decrease in Aβ binding to apoE3 (Fig. 1, A and B, lane 6) and an increase in Aβ binding to apoE4 (lane 8). To determine whether an auxiliary component that influences the isoform specificity of apoE binding to Aβ is present in the medium of HEK-293 cells, we reconstituted purified apoE with mock-transfected conditioned medium. The isoform specificity of apoE binding to Aβ was not restored to pre-purification levels, indicating that purification does not remove a component present in the medium that confers the preferential binding of Aβ to apoE3 (compare Figs. 2 and 1 A, lanes 5-8).Figure 1:ApoE/Aβ binding: effect of purifying apoE isoforms from conditioned medium. Shown are Western blots of binding reactions using 25 μg/ml apoE (~700 n M) incubated with (lanes 2, 4, 6, and 8) or without (lanes 1, 3, 5, and 7) 250 μM Aβ-(1-40)-peptide for 2 h at room temperature. Sources of apoE were as follows: conditioned medium of transfected cells (apoE3, lanes 1 and 2; apoE4, lanes 3 and 4) and purified from conditioned medium (apoE3, lanes 5 and 6; apoE4, lanes 7 and 8). Samples were run in nonreducing Laemmli buffer on 10-20% SDS-Tricine gels, transferred to Immobilon-P membrane, and probed with 4G8 antibody (A) or apoE antiserum (B).View Large Image Figure ViewerDownload (PPT)Figure 2:ApoE/Aβ binding: effect of mock medium on apoE isoforms purified from conditioned medium. Shown are Western blots of apoE/Aβ binding reactions assayed under the conditions described in the legend to Fig. 1. Prior to binding reactions, apoE purified from conditioned medium was incubated for 30 min with concentrated conditioned medium from mock-transfected cells to reconstitute the components in unpurified apoE-conditioned medium. Samples were run as described in the legend to Fig. 1 and probed with 4G8 antibody. Lane 1, apoE3 + mock medium; lane 2, apoE3 + mock medium + Aβ;lane 3, apoE4 + mock medium; lane 4, apoE4 + mock medium + Aβ.View Large Image Figure ViewerDownload (PPT)To examine the binding of Aβ to unpurified plasma apoE isoforms, VLDL was isolated from the plasma of apoE3/3 and apoE4/4 homozygotes and used for binding reactions with Aβ. Similar to the results using unpurified apoE from conditioned medium, we observed an ~20-fold difference in the amount of the apoE3⋅Aβ complex (Fig. 3, A and B, lane 2) compared with the apoE4⋅Aβ complex (lane 4) when using intact VLDL particles as the source of apoE. As with purification of apoE from medium, purification of apoE from the VLDL fraction of human plasma attenuated the preferential binding of apoE3 to Aβ (Fig. 3 A and B, lanes 6 and 8). The level of Aβ binding to both apoE4 and apoE3 dimer was increased after purification of apoE from plasma. In apoE3/3 plasma, the apoE immunoreactive species migrating slightly higher than the apoE⋅Aβ complex (~55 kDa) is the apoE3-apolipoprotein A-II heterodimer (Fig. 3 B, lanes 1 and 2). 2M. J. LaDu and M. T. Falduto, unpublished observations. The autoradiographs shown in Figs. 1 and 3 are representative of several experiments with similar results using different apoE-containing preparations. While some variation in the purification-mediated loss of preferential apoE3 binding to Aβ was observed, in general, purification of apoE from both medium and plasma abolished isoform-specific binding to Aβ.Figure 3:ApoE/Aβ binding: effect of purifying apoE isoforms from human plasma. Shown are Western blots of apoE/Aβ binding reactions assayed under the conditions described in the legend to Fig. 1. Sources of apoE were as follows: intact VLDL from apoE3/3 (lanes 1 and 2) and apoE4/4 (lanes 3 and 4) homozygotes and purified from VLDL of apoE3/3 (lanes 5 and 6) and apoE4/4 (lanes 7 and 8) homozygotes. Samples were run as described in the legend to Fig. 1 and probed with 4G8 antibody (A) or apoE antiserum (B). For apoE3 VLDL, total protein was 95 ng/μl, and triglyceride was 600 ng/μl. For apoE4 VLDL, total protein was 110 ng/μl, and triglyceride was 400 ng/μl.View Large Image Figure ViewerDownload (PPT)Strittmatter et al. (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar) used apoE purified from plasma to show that Aβ binds to apoE4 monomer at a faster rate than to apoE3 monomer. However, the contribution of the apoE3-Aβ complex in their time course data is difficult to discern since the blots were probed for apoE and Aβ immunoreactivity simultaneously (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar). Therefore, we repeated the time course study using purified plasma apoE isoforms and probed for Aβ (Fig. 4 A) and apoE (Fig. 4 B) immunoreactivity individually. Similar to the results obtained by Strittmatter et al. (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar), apoE4 monomer did form a complex with Aβ at a faster rate compared with apoE3 monomer. However, when the apoE3 dimer-Aβ complex was taken into account, the total binding of Aβ to apoE3 and apoE4 remained comparable across time.There are several possibilities that could account for the loss of isoform specificity in the binding of purified apoE to Aβ. Purification may result in a change in the oxidative state of apoE that influences its binding to Aβ. Although the apoE isoforms are subject to the same conditions while undergoing purification, oxidation could be specific to the Cys112 in apoE3. This could affect conformation-dependent binding of apoE to Aβ, either directly or via interaction of other amino acids with this residue. However, apoE3 dimer, particularly in its purified form, readily binds Aβ, indicating that at least oxidation of cysteine does not seem to adversely affect apoE3 ⋅ Aβ complex formation. In addition, oxidation provides no obvious explanation for the increase in Aβ binding to purified apoE4 monomer.Another more likely explanation is that apoE acquires a conformation when lipid-associated that confers isoform-specific binding to Aβ. Purification, which includes delipidation, may increase the binding of Aβ to apoE3 dimer and apoE4 monomer by disrupting the endogenous conformation of these apoE species. In each of the unpurified preparations used here, apoE was lipid-associated. ApoE from human plasma was in the VLDL fraction, while the majority of apoE in conditioned medium was in a high density lipid fraction. 3M. J. LaDu, M. T. Falduto, and C. A. Reardon, unpublished observations. Purified apoE has been shown to require lipid to restore its biological activity in other assay systems, presumably by allowing the denatured protein to refold to its functional conformation. These apoE activity assays include such diverse functions as receptor binding and modulation of neuritic growth. While purified apoE is unable to displace 125I-labeled low density lipoprotein (LDL) from LDL receptors (16Innerarity T.L. Mahley R.W. Biochemistry. 1978; 17: 1440-1447Google Scholar), high affinity binding is restored when apoE is added in the presence of a variety of lipid particles, including phospholipid vesicles (17Innerarity T.L. Pitas R.E. Mahley R.W. Methods Enzymol. 1986; 129: 542-565Google Scholar). Similarly, the binding of purified apoE to LDL receptor-related protein is enhanced by the addition of β-migrating VLDL (18Kowal R.C. Herz J. Weisgraber K.H. Mahley R.W. Brown M.S. Goldstein J.L. J. Biol. Chem. 1990; 265: 10771-10779Google Scholar). In primary dorsal root ganglia cultures, purified apoE3 and apoE4 have no effect on neuritic growth. However, when added in the presence of β-migrating VLDL, both apoE isoforms affect neuritic growth, with apoE3 increasing neuritic extension and decreasing branching and apoE4 decreasing both branching and extension (19Nathan B.P. Bellosta S. Sanan D.A. Weisgraber K.H. Mahley R.W. Pitas R.E. Science. 1994; 264: 850-852Google Scholar). Experiments are in progress using the assay system described here to determine the nature of the interaction of Aβ with lipid-associated apoE using both endogenous lipoproteins and reconstituted lipid vesicles.The data presented here are consistent with a function for apoE in the pathogenesis of AD by sequestering Aβ. Avid binding of apoE3 to soluble Aβ in the neuropil could lead to enhanced clearance or altered fibril formation, both of which could prevent the conversion of Aβ into a neurotoxic species. Since lipid-associated apoE3 and apoE4 bind the LDL receptor and LDL receptor-related protein with equal affinity (18Kowal R.C. Herz J. Weisgraber K.H. Mahley R.W. Brown M.S. Goldstein J.L. J. Biol. Chem. 1990; 265: 10771-10779Google Scholar), efficient uptake and clearance of Aβ may depend on its preferential binding to apoE3. Alternatively, purified apoE3 and apoE4 are known to differ with respect to their effect on Aβ fibril formation (20Ma J. Wee A. Brewer H.B. Das S. Potter H. Nature. 1994; 372: 92-94Google Scholar, 21Sanan D.A. Weisgraber K.H. Russell S.J. Mahley R.W. Huang D. Saunders A. Schmechel D. Wisniewski T. Frangione B. Roses A.D. Strittmatter W.J. J. Clin. Invest. 1994; 94: 860-869Google Scholar, 22Wisniewski T. Castano E.M. Golabek A. Vogel T. Frangione B. Am. J. Pathol. 1994; 145: 1030-1035Google Scholar), suggesting that native apoE may play an even greater isoform-specific role in the extracellular aggregation of Aβ due to differences in Aβ binding characteristics. Finally, there is evidence that Aβ-induced toxicity in hippocampal neurons is attenuated by the addition of rabbit apoE (23Whitson J.S. Mims M.P. Strittmatter W.J. Yamaki T. Morrisett J.D. Appel S.H. Biochem. Biophys. Res. Commun. 1994; 199: 163-170Google Scholar), leading to the intriguing possibility that human apoE may contribute to Aβ-induced toxicity in an isoform-specific manner.In summary, using unpurified apoE from tissue culture medium and intact VLDL particles, the apoE3⋅Aβ complex was ~20-fold more abundant than the apoE4⋅Aβ complex. This isoform specificity was attenuated or abolished when apoE purified from these two sources was used in binding reactions with Aβ. ApoE is an apolipoprotein, and as such, its endogenous conformation requires lipid. It is therefore not surprising that the type of apoE preparation used can affect the results. The avidity of Aβ binding to apoE3 compared with apoE4 demonstrated here may be involved in the isoform-specific effect underlying the genetic correlation between the apoE e4 allele and AD. The physiological relevance of this complex to plaque formation or neurodegeneration awaits further investigation. INTRODUCTIONApolipoprotein E (apoE), 1The abbreviations used are: apoEapolipoprotein EADAlzheimer' diseaseAββ-amyloidVLDLvery low density lipoproteinTricineN-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycineLDLlow density lipoprotein. a component of several classes of lipoproteins, acts as a ligand for lipoprotein receptors, thus regulating lipid transport and clearance. ApoE is also expressed in the brain (1Elshourbagy N.A. Liao W.S. Mahley R.W. Taylor J.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 203-207Google Scholar) and in response to injury in both the peripheral (2Boyles J.K. Zoellner C.D. Anderson L.J. Kosik L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Haerter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Google Scholar) and central nervous (3Snipes G.J. McGuire C.B. Norden J.J. Freeman J.A. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1130-1134Google Scholar) systems. In humans, apoE has three major isoforms, E2 (Cys112, Cys158), E3 (Cys112, Arg158), and E4 (Arg1121, Arg158), which are products of three alleles at a single gene locus. The presence of cysteine in apoE2 and apoE3 allow these isoforms to form disulfide-linked dimers. Recently, it has been demonstrated that the apoE e4 allele is present with increased frequency in patients with sporadic (4Saunders A.M. Strittmatter W.J. Schmechel D. George-Hyslop St., P.H. Pericak-Vance M.A. Joo S.H. Rosi B.L. Gusella J.F. Crapper-MacLachlan D.R. Alberts M.J. Hulette C. Crain B. Goldgaber D. Roses A.D. Neurology. 1993; 43: 1467-1472Google Scholar) and late-onset familial (5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar) Alzheimer' disease (AD). Due primarily to this genetic linkage, the role of apoE in the pathogenesis of AD is being actively pursued.A major neuropathological feature of AD is the presence of extracellular senile plaques composed predominantly of aggregated β-amyloid peptide (Aβ). The physiological mechanism by which apoE contributes to AD pathology may be by means of an isoform-specific interaction with Aβ. ApoE and Aβ colocalize in senile plaques (6Wisniewski T. Frangione B. Neurosci. Lett. 1992; 135: 235-238Google Scholar), and synthetic Aβ peptides bind in vitro to apoE from tissue culture medium (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar) and cerebrospinal fluid (5Strittmatter W.J. Saunders A.M. Schmechel D. Pericak-Vance M. Enghild J. Salvesen G.S. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981Google Scholar, 8Ghiso J. Matsubara E. Koudinov A. Choi-Miura N.H. Tomita M. Wisniewski T. Frangione B. Biochem. J. 1993; 293: 27-30Google Scholar, 9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar) as well as to purified apoE (9Wisniewski T. Golabek A. Matsubara E. Ghiso J. Frangione B. Biochem. Biophys. Res. Commun. 1993; 192: 359-365Google Scholar, 10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar). However, studies involving apoE isoform specificity in binding to Aβ have been limited and contradictory. In previous work, we demonstrated that apoE3 from tissue culture medium binds to Aβ with greater avidity than apoE4 (7LaDu M.J. Falduto M.T. Manelli A.M. Reardon C.A. Getz G.S. Frail D.E. J. Biol. Chem. 1994; 269: 23403-23406Google Scholar). This is in contrast to data from Strittmatter et al. (10Strittmatter W.J. Weisgraber K.H. Huang D.Y. Dong L.-M. Salvesen G.S. Pericak-Vance M. Schmechel D. Saunders A.M. Goldgaber D. Roses A.D. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 8098-8102Google Scholar), who used apoE isoforms purified from human plasma as substrate for Aβ. Here we resolve the apparent discrepancy between these studies by demonstrating that the previously observed preferential association of Aβ with apoE3 is attenuated or abolished by purification, a process that includes delipidation and denaturation.

Referência(s)