Advertisement

Characterization of Apolipoprotein J-Alzheimer's Aβ Interaction (∗)

  • Etsuro Matsubara
    Affiliations
    Department of Pathology, New York University Medical Center, New York, New York 10016
    Search for articles by this author
  • Blas Frangione
    Affiliations
    Department of Pathology, New York University Medical Center, New York, New York 10016
    Search for articles by this author
  • Jorge Ghiso
    Correspondence
    To whom correspondence should be addressed: Dept. of Pathology TH432, New York University Medical Center, 560 First Ave., New York, NY, 10016 . Tel.: 212-263-5775; Fax: 212-263-6751
    Affiliations
    Department of Pathology, New York University Medical Center, New York, New York 10016
    Search for articles by this author
  • Author Footnotes
    ∗ This work was supported by National Institutes of Health Grants AG10953 (Leadership and Excellence in Alzheimer's Disease), AG05891, AG10491, and AG08721. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:March 31, 1995DOI:https://doi.org/10.1074/jbc.270.13.7563
      The main component of Alzheimer's amyloid deposits, Aβ, has been found also as a soluble (sAβ) normal constituent of biological fluids and cell culture supernatants. Whether or not sAβ is the immediate precursor of Aβ, it is clear that peptides with the same amino acid sequence can have both fibrillar and non-fibrillar conformations. The interconversion mechanism from one form to another is presently under intensive investigation. We have previously described that (i) a synthetic peptide Aβ1-40 immobilized on affinity matrices was able to retrieve apolipoprotein J (apoJ) from plasma and cerebrospinal fluid; and (ii) the interaction of sAβ with apoJ occurs in vivo, as demonstrated by the ability of anti-apoJ to co-precipitate sAβ from normal cerebrospinal fluid. We have characterized the binding between Aβ1-40 and apoJ and found that the interaction is saturable, specific, and reversible. The dissociation constant of 2 × 10−9M is indicative of high affinity binding. The stoichiometry of the reaction is 1:1; apoJ has five times more affinity for fresh Aβ1-40 than for the aggregated peptide. Competitive inhibition studies carried out with apolipoprotein E (isoforms E2, E3, and E4), transthyretin, vitronectin, and α1-antichymotrypsin indicate that the complex apoJ•Aβ1-40 cannot be dissociated by any of these competitors at physiologic concentrations. The data strongly suggest that apoJ plays an important role as a carrier protein for sAβ.

      INTRODUCTION

      Amyloid β (Aβ)
      The abbreviations used are: Aβ
      amyloid β
      AD
      Alzheimer's disease
      s
      soluble
      CSF
      cerebrospinal fluid
      ACT
      α1-antichymotrypsin
      Vn
      vitronectin
      TTR
      transthyretin
      PBS
      phosphate-buffered saline
      ELISA
      enzyme-linked immunosorbent assay.
      peptide (39-44 residues) is the main component of the two major neuropathological lesions present in AD, senile plaques, and cerebrovascular amyloid deposits(
      • Wisniewski T.
      • Ghiso J.
      • Frangione B.
      ,
      • Wisniewski T.
      • Lalowski M.
      • Levy E.
      • Marques M.
      • Frangione B.
      ). Although Aβ has high tendency to aggregate and make fibrils, a soluble form has been detected in biological fluids (soluble Aβ, sAβ)(
      • Haass C.
      • Schlossmacher M.
      • Hung A.
      • Vigo-Pelfrey C.
      • Mellon A.
      • Ostaszewski B.
      • Lieberburg I.
      • Koo E.
      • Schenk D.
      • Teplow D.
      • Selkoe D.
      ,
      • Seubert P.
      • Vigo-Pelfrey C.
      • Esch F.
      • Lee M.
      • Dovey H.
      • Davis D.
      • Sinha S.
      • Schlossmacher M.
      • Whaley J.
      • Swindlehurst C.
      • McCormack R.
      • Wolfert R.
      • Selkoe D.
      • Lieberburg I.
      • Schenk D.
      ,
      • Shoji M.
      • Golde T.
      • Ghiso J.
      • Cheung T.
      • Estus S.
      • Shaffer L.
      • Cai X.-D.
      • McKay D.
      • Tintner R.
      • Frangione B.
      • Younkin S.
      ). Whether or not sAβ is the immediate precursor of Aβ, it is clear that the same amino acid sequence can have both fibrillar and non-fibrillar conformations; therefore, the knowledge of the factors that influence its behavior in solution will be a step forward in the understanding of AD pathology. In this regard, the existence of specific components named “desaggrins” was previously suggested based on the fact that Aβ1-40 spontaneous fibril formation in vitro is inhibited in the presence of CSF(
      • Wisniewski T.
      • Castano E.
      • Ghiso J.
      • Frangione B.
      ).
      Extensive immunohistochemical studies indicate that other proteins are co-deposited with Aβ in senile plaques. Amyloid P-component, α1-antichymotrypsin (ACT), apolipoprotein E (apoE), apolipoprotein J (apoJ), complement components, vitronectin (Vn), glycosaminoglycans, and extracellular matrix proteins are among the amyloid-associated proteins described so far(
      • Abraham C.
      • Selkoe D.
      • Potter H.
      ,
      • Coria F.
      • Castano E.
      • Prelli F.
      • Larrondo-Lillo M.
      • van Duinen S.
      • Shelanski M.
      • Frangione B.
      ,
      • Snow A.
      • Willemer J.
      • Kisilevsky R.
      ,
      • Choi-Miura N.-H.
      • Ihara Y.
      • Fukuchi K.
      • Takeda M.
      • Nakano Y.
      • Tobe T.
      • Tomita M.
      ,
      • McGeer P.
      • Kawamata T.
      • Walker D.
      ,
      • Wisniewski T.
      • Frangione B.
      ,
      • Eikelenboom P.
      • Stam F.
      ,
      • Ishii T.
      • Haga S.
      ,
      • Eikelenboom P.
      • Zhan S.
      • Kamphorst W.
      • van der Valk P.
      • Rozemuller J.
      ). It is not clear whether they are innocent bystanders or their presence is related to the mechanism of amyloidogenesis. Several lines of investigation favor the latter notion, at least for some of them (i.e. amyloid P-component and apoE are present in several types of fibrillar deposits but absent in non-fibrillar accumulations representing pre-amyloid lesions)(
      • Gallo G.
      • Wisniewski T.
      • Choi-Miura N.-H.
      • Ghiso J.
      • Frangione B.
      ). In addition, the apoE gene on chromosome 19, particularly the apoE allele ∊4, has been linked to sporadic and late-onset AD(
      • Pericak-Vance M.
      • Bebout J.
      • Gaskell P.
      • Yamaoka L.
      • Hung W.
      • Alberts M.
      • Walker A.
      • Bartlett R.
      • Haynes C.
      • Welsh K.
      • Earl N.
      • Heyman A.
      • Clark C.
      • Roses A.
      ). The inheritance of the apoE4 allele is today considered a risk factor for AD(
      • Wisniewski T.
      • Ghiso J.
      • Frangione B.
      ).
      Biochemical studies performed in vitro have demonstrated a certain degree of binding affinity between Aβ and different proteins, among them apoJ, apoE, transthyretin (TTR), and ACT(
      • Wisniewski T.
      • Frangione B.
      ,
      • Ghiso J.
      • Matsubara E.
      • Koudinov A.
      • Choi-Miura N.-H.
      • Tomita M.
      • Wisniewski T.
      • Frangione B.
      ,
      • Strittmatter W.
      • Saunders A.
      • Schmechel D.
      • Pericak-Vance M.
      • Enghild J.
      • Salvesen G.
      • Roses A.
      ,
      • Wisniewski T.
      • Golabek A.
      • Matsubara E.
      • Ghiso J.
      • Frangione B.
      ,
      • LaDu M.
      • Falduto M.
      • Manelli A.
      • Reardon C.
      • Getz G.
      • Frail D.
      ,
      • Schwarzman A.
      • Gregori L.
      • Vitek M.
      • Lyubski S.
      • Strittmatter W.
      • Enghilde J.
      • Bhasin R.
      • Silverman J.
      • Weisgraber K.
      • Coyle P.
      • Zagorski M.
      • Talafous J.
      • Eisenberg M.
      • Saunders A.
      • Roses A.
      • Goldgaber D.
      ,
      • Ma J.
      • Yee A.
      • Brewer H.
      • Das S.
      • Potter H.
      ). The interactions have been considered in the range of “high avidity binding,” although they were not quantitatively evaluated. Using immobilized synthetic peptides homologous to Aβ (Aβ1-40), we have shown previously its binding association with plasma and CSF apolipoproteins J and E; moreover, the presence of the complex apoJ•sAβ was confirmed in CSF, indicating that the interaction takes place in vivo(
      • Ghiso J.
      • Matsubara E.
      • Koudinov A.
      • Choi-Miura N.-H.
      • Tomita M.
      • Wisniewski T.
      • Frangione B.
      ,
      • Wisniewski T.
      • Golabek A.
      • Matsubara E.
      • Ghiso J.
      • Frangione B.
      ). We are reporting herein the characterization of the complex formation between apoJ and Aβ1-40, in the presence and absence of other amyloid-associated proteins.

      MATERIALS AND METHODS

      Synthetic Peptides and Proteins

      Peptide DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVV (Aβ1-40, homologous to residues 672-711 of βPP770) was synthesized at the W. M. Keck Facility at Yale University and characterized as described(
      • Ghiso J.
      • Matsubara E.
      • Koudinov A.
      • Choi-Miura N.-H.
      • Tomita M.
      • Wisniewski T.
      • Frangione B.
      ). A 1 mg/ml stock solution of Aβ1-40 was prepared in 50% acetonitrile containing 0.1% trifluoroacetic acid, aliquoted, and frozen at −80°C until use. For aggregation studies, 100 ng of Aβ1-40 were diluted in 20 μl of PBS (20 mM phosphate, pH 7.4, containing 150 mM NaCl) and incubated at 37°C for different periods of time (0-72 h). Incubation was terminated by the addition of SDS sample buffer and 5 min boiling, and the degree of aggregation was determined by Tris-Tricine 10% SDS-PAGE(
      • Schägger H.
      • von Jagow G.
      ).
      Human plasma apoJ was purchased from Quidel (San Diego, CA); ACT and TTR were obtained from Calbiochem (La Jolla, CA); Vn was purchased from Chemicon (Temecula, CA). Recombinant apoE isoforms 2, 3, and 4 (apoE2, apoE3, and apoE4) were obtained from PanVera (Madison, WI). In all cases, protein purity was corroborated by SDS-PAGE and NH2-terminal sequence.

      Solid-phase Binding Studies

      The interaction apoJ-Aβ1-40 was studied by enzyme-linked immunosorbent assay (ELISA) using Aβ1-40 and purified apoJ. Polystyrene microtiter plates (Immulon 2, Dynatech Lab.; Chantilly, VA) were coated with freshly prepared Aβ1-40 (400 ng/100 μl/well) in 0.1 M NaHCO3, pH 8.6, for 2 h at 37°C. Under these conditions, 10 ng of Aβ1-40 (representing 2.5% of the peptide offered) remained bound to the well, as determined by a modification of Quantigold assay (Diversified Biotech., Boston, MA). After blocking with 1% bovine serum albumin, different concentrations of purified apoJ (0-25 nM; 100 μl/well) in PBS were added to the Aβ1-40-coated wells and incubated for 3 h at 37°C. Bound apoJ was detected with monoclonal IF12 (anti-apoJ α-chain, 1:5000), a generous gift from Dr. N. H. Choi-Miura(
      • Choi N.-H.
      • Tobe T.
      • Hara K.
      • Yoshida H.
      • Tomita M.
      ), followed by alkaline phosphatase-conjugated goat F(ab′)2 anti-mouse IgG (1:3000, BioSource; Camarillo, CA). The reaction was developed for 30 min with p-nitrophenyl phosphate in diethanolamine buffer (Bio-Rad), stopped with 0.4 M NaOH, and quantitated at 405 nm on a 7520 Microplate Reader (Cambridge Technology, Watertown, MA). Nonspecific binding was determined using bovine serum albumin-coated wells and/or omitting apoJ in the assay. Binding data were analyzed with the aid of a curve fitting software (GraphPad Prism Version 1.0, GraphPad Software, San Diego, CA).

      Inhibition Assays

      100 μg of apoJ were biotinylated with Sulfo-NHS biotin (Pierce) according to the manufacturer's specifications; biotin-labeled apoJ was separated from free biotin by chromatography over Sephadex G-10 (Pharmacia) equilibrated in PBS. 0-1136 nM of native unlabeled apoJ was coincubated with 25 nM of biotin-labeled apoJ in Aβ1-40-coated wells (400 ng/well) for 3 h at 37°C. Bound biotinylated apoJ was detected with alkaline phosphatase labeled streptavidin (1:1000, Amersham) and evaluated as described above.
      In separate experiments, variable amounts of fresh Aβ1-40 or 24-h self-aggregated Aβ1-40 (0-1136 nM) were combined with 25 nM apoJ in PBS and incubated for 3 h at 37°C. The mixture was then transferred to Aβ1-40-coated wells and incubated for 3 h at 37°C. Bound apoJ was detected with monoclonal IF12 and alkaline phosphatase-labeled F(ab′)2 goat anti-mouse IgG, as described above.
      Competitive inhibition assays were performed with apoE2, apoE3, apoE4, ACT, Vn, and TTR. 0-2500 nM of the various competitors in PBS were coincubated with 25 nM of apoJ in PBS in Aβ1-40-coated wells (400 ng/well) at 37°C for 3 h. Bound apoJ was determined as described above.

      Complex Characterization

      The apoJ•Aβ1-40 complex was prepared by incubation of 5 μg of apoJ with 5 μg of Aβ1-40 in PBS for 18 h at 37°C. The mixture was separated on native 8% non-SDS-PAGE(
      • Bryan J.
      ), transferred to polyvinylidene difluoride membranes (Immobilon P, Millipore, Milford, MA) using 3-cyclohexylamino-1-propanesulfonic acid pH 11, containing 10% (v/v) methanol, stained with Coomassie Blue, and the protein bands were excised, and sequenced on a 477A protein sequencer (Applied Biosystems, Foster City, CA), as described(
      • Ghiso J.
      • Matsubara E.
      • Koudinov A.
      • Choi-Miura N.-H.
      • Tomita M.
      • Wisniewski T.
      • Frangione B.
      ). For immunoblot detection, apoJ•Aβ1-40 and apoJ•Aβ1-40(agg) complexes were prepared under identical conditions using 1.1 μmol of apoJ and 1.1 μmol of either fresh or 24-h aggregated peptide. After 18 h of incubation at 37°C, the complexes were separated by electrophoresis and electroblotted onto Immobilon P as indicated above. Membranes were blocked for 1 h with 5% low-fat dried milk in PBS and incubated overnight with monoclonal 6E10 (anti-Aβ1-17), kindly provided by Dr. K. S. Kim(
      • Kim K.
      • Wen G.
      • Bancher C.
      • Chen C.
      • Sapienza V.
      • Hong H.
      • Wisniewski H.
      ), at 1:100 dilution, followed by horseradish peroxidase-labeled sheep anti-mouse F(ab′)2 1:2000 (Amersham). Immunoblots were visualized with an enhanced chemiluminescence (ECL) detection kit and exposed to Hyperfilm ECL (Amersham). The resulting bands were scanned on a PDI densitometer and evaluated with the Quantity One (Version 2.4) software (PDI, Huntington Station, NY).

      RESULTS

      The interaction between apoJ and Aβ1-40 was characterized by means of solid-phase ELISA experiments. A dose-response relationship that reached saturation was obtained when increasing concentrations of apoJ at pH 7.4 were allowed to interact with a constant amount of immobilized Aβ1-40 (Fig. 1). Non-linear regression analysis of the specific binding data fitted to a rectangular hyperbola and allowed the calculation of the corresponding dissociation constant Kd of 2 nM. The specificity and reversibility of the interaction were assessed at an apoJ concentration of 25 nM (at which 100% saturation of the Aβ1-40-coated plates was achieved) by means of inhibition experiments with either Aβ1-40 or apoJ. When a constant concentration of apoJ in PBS was preincubated with increasing concentrations of freshly prepared Aβ1-40 before the addition to Aβ1-40-coated wells, apoJ-binding to immobilized Aβ1-40 followed a one-site competition curve (Fig. 1, inset). The calculated value for half-maximal inhibition (IC50) was 63 nM. In a separate set of experiments, the binding of biotin-labeled apoJ to immobilized Aβ1-40 was competitively inhibited by increasing concentrations of native apoJ. The remaining bound biotinylated apoJ was quantitated with alkaline phosphataselabeled streptavidin. The data fitted into a one-site competition curve with IC50 = 77 nM (Fig. 1, inset).
      Figure thumbnail gr1
      Figure 1Saturation curve for apoJ-Aβ1-40 interaction. Variable concentrations (0-25 nM) of apoJ were incubated with Aβ1-40-coated wells for 3 h at 37°C. Bound apoJ was detected with monoclonal IF12 and alkaline phosphatase-labeled anti-mouse, as described under “Materials and Methods.” Each point represents the mean (±2 S.D.) of five independent duplicate experiments. Inset, inhibition of apoJ binding to immobilized Aβ1-40. Increasing concentrations (0-1136 nM) of Aβ1-40 were preincubated with apoJ (25 nM) for 3 h at 37°C. The mixture was added to Aβ1-40-coated wells and incubated for another 3 h at the same temperature. Bound apoJ was determined as described under “Materials and Methods.” On separate experiments, native apoJ (0-1136 nM) was coincubated with 25 nM of biotin-labeled apoJ in Aβ1-40-coated wells for 3 h at 37°C. Bound biotinylated apoJ was detected with alkaline phosphatase-labeled streptavidin. Results are expressed as percentage of binding compared with controls incubated with apoJ alone. Data represent the mean of three independent duplicate experiments. S.D. never exceeded ±6%.
      The formation of the apoJ•Aβ1-40 complex was visualized by Coomassie Blue staining after electrophoresis on non-denaturing polyacrylamide gels. As shown in Fig. 2, native apoJ exhibits two molecular forms in this non-SDS system: a monomeric component of ∼80 kDa and a dimeric form of ∼160 kDa (lane 2). When complexed to Aβ1-40, both components shifted their electrophoretic mobility toward higher molecular masses, resulting in ∼85 and 170 kDa bands (an increase of ∼5 and 10 kDa, respectively). Amino-terminal sequence analysis of these 85- and 170-kDa components rendered the sequences DQTVSDNELQEMSNQ, SLMPFSPYEPLNFH, and DAEFRHDSGYEVHHQ corresponding to the first 15 residues of the apoJ α-chain, apoJ β-chain, and Aβ1-40, respectively. Recovery calculations performed for the first 10 steps of the sequence indicated a 1 to 1 stoichiometry (Table 1).
      Figure thumbnail gr2
      Figure 2Identification and characterization of apoJ•Aβ1-40 complexes under non-denaturing conditions. ApoJ (5 μg) and Aβ1-40 (5 μg) were incubated at 37°C for 18 h and the resultant complexes separated on 8% non-SDS-PAGE, transferred to Immobilon P, and stained with Coomassie Blue. The apparent molecular masses were calculated from the Ferguson plots constructed with known molecular mass standards (α-lactalbumin, 14,200 Da; carbonic anhydrase, 29,000 Da; chicken egg albumin, 45,000 Da; bovine serum albumin, 66,000 Da monomer and 132,000 Da dimer; urease, 272,000 Da monomer and 545,000 Da dimer). The complexes (arrowheads) were excised from the membrane and their NH2-terminal sequence determined. Lane 1, Aβ1-40; lane 2, apoJ; lane 3, apoJ•Aβ1-40 complex.
      The influence of the degree of Aβ1-40 aggregation in its ability to form a complex with apoJ was tested using fresh and aggregated peptide; the resulting complex was visualized via immunoblot analysis after SDS-PAGE using anti-Aβ1-17 (monoclonal 6E10), as indicated in Fig. 3. Fresh Aβ1-40 exhibited a major monomeric component and a minor dimeric form while the tetrameric aggregates were almost negligible (lane 1, arrowheads). A similar aliquot of the synthetic peptide that had been incubated for 24 h at 37°C showed an increase in the amount of dimers and tetramers in addition to the typical smear-like electrophoretic appearance which indicates the presence of multiple minor components of higher molecular mass (lane 3). When fresh Aβ1-40 was incubated with apoJ for 18 h at 37°C (lane 2), the presence of the 85-kDa apoJ•Aβ1-40 complex was immunodetected by anti-Aβ1-17 (arrow); the free peptide exhibited the same polymerization pattern as the one shown in lane 1. When 24-h-aggregated Aβ1-40 was incubated with apoJ under identical conditions, the presence of a less intense 85-kDa complex was detected by anti-Aβ1-17 (lane 4, arrow) while the peptide aggregation pattern resembled the one in lane 3. Densitometric evaluation of both complexes (lanes 2 and 4, arrow) indicated that the amount of apoJ-Aβ1-40 formed was 4.6 times lower when aggregated peptide was used. To confirm this apparent different affinity of apoJ for aggregated and non-aggregated Aβ1-40, inhibition assays were carried out on ELISA plates. Both fresh and aggregated peptides were allowed to interact with apoJ in fluid-phase for 3 h at 37°C; the remaining free apoJ was tested for its ability to bind to Aβ1-40-coated wells. As depicted in Fig. 4, aggregated Aβ1-40 exhibited five times less efficiency to form complexes with apoJ (IC50 = 315 nM) than the fresh peptide (IC50 = 63 nM).
      Figure thumbnail gr3
      Figure 3Immunodetection of the apoJ•Aβ1-40 complex formed with fresh or aggregated peptide. ApoJ•Aβ1-40 and apoJ•Aβ1-40(agg) complexes were prepared in PBS using 1.1 μmol of apoJ and 1.1 μmol of either fresh or 24-h heat-aggregated peptide. After 18 h incubation at 37°C, complexes were separated on Tris-Tricine 10% SDS-PAGE and immunoblotted with anti-Aβ1-17 (6E10). Visualization was carried out with peroxidase-labeled anti-mouse followed by ECL. Lane 1, fresh Aβ1-40 (100 ng); lane 2, apoJ-fresh Aβ1-40 complex (1:1 molar ratio); lane 3, 24-h heat-aggregated Aβ1-40 (100 ng); lane 4, apoJ•Aβ1-40(agg) complex (1:1 molar ratio). Solid arrowhead, monomeric Aβ1-40; open arrowheads, dimeric and tetrameric forms of Aβ1-40; arrow, apoJ•Aβ1-40 complex.
      Figure thumbnail gr4
      Figure 4Inhibition of apoJ•Aβ1-40 interaction by either fresh or aggregated Aβ1-40. ApoJ (25 nM) was preincubated with various concentrations of either fresh or 24-h-preaggregated Aβ1-40 prior to the addition to Aβ1-40-coated plates. Bound apoJ was detected as described under “Materials and Methods.” ApoJ binding is expressed as percentage of binding compared to control wells incubated in the absence of inhibitor. Data represent the mean of three independent experiments duplicate experiments. S.D. never exceeded ±6%.
      Competitive inhibition experiments using other plasma/CSF proteins with demonstrated binding affinity for Aβ (apoE2, apoE3, apoE4, ACT, Vn, and TTR) were performed by solid-phase assays. Increasing concentrations (0-2.5 μM) of each competitor were mixed with a constant amount of apoJ and immediately added to Aβ1-40-coated wells. Bound apoJ was detected with monoclonal IF12 after 3 h of incubation. As indicated in Fig. 5, none of the proteins tested exhibited higher affinity for Aβ1-40 than ApoJ. The competition curves distributed themselves into two very well defined groups: the first composed of the three apoE isoforms and the second contained the other proteins. ApoE2 (IC50 = 316 nM) was the strongest competitor of all the apoE isoforms according to the calculated IC50 values, followed by apoE3 (IC50 = 502 nM) and apoE4 (IC50 = 794 nM), indicating that they have 4-10 times lower relative affinity than apoJ for Aβ1-40. None of the other proteins tested reached 50% inhibition of apoJ-Aβ1-40 binding under the conditions tested (TTR, IC50 = 9550 nM; Vn, IC50 = 9820 nM; ACT, IC50 = 11340 nM).
      Figure thumbnail gr5
      Figure 5Competitive inhibition of the apoJ•Aβ1-40 complex formation by amyloid associated proteins. ApoJ (25 nM) was combined with variable concentrations (0-2500 nM) of apoE2 (▲), apoE3 (▼), apoE4 (◆), TTR (▵), ACT (•), and Vn (□) and coincubated with Aβ1-40-coated wells. Bound apoJ was determined as described under “Materials and Methods.” The inhibition of biotin-labeled apoJ•Aβ1-40 complex formation by native apoJ (■) presented on (inset) is shown here for comparison. Results are expressed as percent of the maximum specific binding obtained in the absence of competitors. Data represent the mean of five independent duplicate experiments. S.D. never exceeded ±7%.

      DISCUSSION

      ApoJ (also known as clusterin or SP-40, 40) is a multifunctional disulfide-linked heterodimeric glycoprotein composed of two ∼40-kDa subunits (named α and β chains)(
      • Murphy B.
      • Kirszbaum L.
      • Walker I.
      • d'Apice A.
      ,
      • de Silva H.
      • Stuart W.
      • Duvic C.
      • Wetterau J.
      • Ray M.
      • Ferguson D.
      • Albers H.
      • Smith W.
      • Harmony J.
      ,
      • Kirszbaum L.
      • Bozas S.
      • Walker I.
      ). The gene for apoJ maps to chromosome 8(
      • Slawin K.
      • Sawczuk I.
      • Olsson C.
      • Buttyan R.
      ,
      • Tobe T.
      • Minoshima S.
      • Yamase S.
      • Choi N.-H.
      • Tomita M.
      • Shimizu N.
      ,
      • Purrello M.
      • Bettuzzi S.
      • Di Pietro C.
      • Mirabile E.
      • Di Blasi M.
      • Rimini R.
      • Grzeschik K.
      • Ingletti C.
      • Corti A.
      • Sichel G.
      ). A single mRNA molecule (
      • Kirszbaum L.
      • Sharpe J.
      • Murphy B.
      • d'Apice A.
      • Classon B.
      • Hudson P.
      • Walker I.
      ) codifies a 449 amino acid chain, and the final apoJ structure is generated by post-translational cleavage at peptide bond Arg205-Ser206. ApoJ message is expressed in almost all mammalian tissues(
      • Collard M.
      • Griswold M.
      ,
      • de Silva H.
      • Harmony J.
      • Stuart W.
      • Gil C.
      • Robbins J.
      ), and the protein has been found in nearly all body fluids(
      • de Silva H.
      • Harmony J.
      • Stuart W.
      • Gil C.
      • Robbins J.
      ,
      • Sylvester S.
      • Morales C.
      • Oko R.
      • Griswold M.
      ). The normal concentration of apoJ in plasma ranges between 35 and 105 μg/ml (0.44-1.3 μM) (
      • Murphy B.
      • Kirszbaum L.
      • Walker I.
      • d'Apice A.
      ), and it is primarily distributed in the high density lipoproteins (
      • de Silva H.
      • Stuart W.
      • Duvic C.
      • Wetterau J.
      • Ray M.
      • Ferguson D.
      • Albers H.
      • Smith W.
      • Harmony J.
      ); it is several times concentrated in seminal fluid while in CSF the values vary between 1.2 and 3.6 μg/ml (15-45 nM) (
      • Choi-Miura N.-H.
      • Ihara Y.
      • Fukuchi K.
      • Takeda M.
      • Nakano Y.
      • Tobe T.
      • Tomita M.
      ,
      • O'Bryan M.
      • Baker H.
      • Saunders J.
      • Kirszbaum L.
      • Walker I.
      • Hudson P.
      • Liu D.
      • d'Apice A.
      • Murphy B.
      ). ApoJ is involved in a variety of physiological processes, including lipid transport(
      • de Silva H.
      • Harmony J.
      • Stuart W.
      • Gil C.
      • Robbins J.
      ), secretion(
      • Hartmann K.
      • Rauch J.
      • Urban J.
      • Parczyk K.
      • Diel P.
      • Pilarsky C.
      • Appel D.
      • Haase W.
      • Mann K.
      • Weller A.
      • Koch-Brandt C.
      ), membrane recycling (
      • Palmer D.
      • Christie D.
      ), spermatogenesis(
      • Kirszbaum L.
      • Sharpe J.
      • Murphy B.
      • d'Apice A.
      • Classon B.
      • Hudson P.
      • Walker I.
      ,
      • Collard M.
      • Griswold M.
      ), and modulation of the complement activity(
      • Choi N.-H.
      • Mazda T.
      • Tomita M.
      ). Within the central nervous system, apoJ is synthesized in neurons (
      • Garden G.
      • Bothwell M.
      • Rubel E.
      ) and astrocytes (
      • Danik B.
      • Chabot J.
      • Mercier C.
      • Benabid A.
      • Chauvin C.
      • Quirion R.
      • Suh M.
      ,
      • Day J.
      • Laping N.
      • McNeill T.
      • Schreiber S.
      • Passinetti G.
      • Finch C.
      ) and is able to cross the blood-brain barrier via a specific receptor/transport mechanism(
      • Zlokovic B.
      • Mackic J.
      • Martell C.
      • Matsubara E.
      • Wisniewski T.
      • McComb G.
      • Frangione B.
      • Ghiso J.
      ). Its production is up-regulated in the degenerative (
      • Dugid J.
      • Bohmont C.
      • Liu N.
      • Tourtellotte W.
      ,
      • May P.
      • Lampert-Etchells M.
      • Johnson S.
      • Poirier J.
      • Masters J.
      • Finch C.
      ) and/or regenerative processes(
      • Buttyan R.
      • Olsson C.
      • Pintar J.
      • Chang C.
      • Bandyk M.
      • Ng P.
      • Sawczuk I.
      ,
      • May P.
      • Johnson S.
      • Poirier J.
      • Lampert-Etchells M.
      • Finch C.
      ). The CSF concentration of apoJ is slightly elevated in AD patients (1.8-4.4 μg/ml), although the differences with the normal population are not statistically significant(
      • Choi-Miura N.-H.
      • Ihara Y.
      • Fukuchi K.
      • Takeda M.
      • Nakano Y.
      • Tobe T.
      • Tomita M.
      ).
      Biochemical data obtained in vitro indicate that apoJ is a major ligand for sAβ in plasma and CSF and that the complex apoJ-sAβ exists in vivo(
      • Ghiso J.
      • Matsubara E.
      • Koudinov A.
      • Choi-Miura N.-H.
      • Tomita M.
      • Wisniewski T.
      • Frangione B.
      ), suggesting that apoJ may act as a carrier protein for sAβ in plasma and CSF. Our results demonstrate that apoJ binds Aβ1-40 with high affinity; the calculated Kd obtained from the saturation curve is 2 × 10−9M. Monomeric and dimeric forms of apoJ (
      • Murphy B.
      • Kirszbaum L.
      • Walker I.
      • d'Apice A.
      ) formed complexes with Aβ1-40; both complexes were constituted by equimolar amounts of apoJ and Aβ1-40, compatible with a 1:1 stoichiometry. The binding was specific and reversible; both, native apoJ and freshly prepared Aβ1-40, inhibited the interaction of apoJ to immobilized Aβ1-40 with a similar IC50. The fact that 100% inhibition can be achieved by incubation of apoJ with Aβ1-40 in solution indicates that the interaction indeed occurs in fluid-phase.
      The self-aggregation rate of Aβ1-40 is pH-, ionic strength-, temperature-, and concentration-dependent(
      • Hilbich C.
      • Kisters-Woike B.
      • Reed J.
      • Masters C.
      • Beyreuther K.
      ,
      • Burdick D.
      • Soreghan B.
      • Kwon M.
      • Kosmoski J.
      • Knauer M.
      • Henschen A.
      • Yates J.
      • Cotman C.
      • Glabe C.
      ,
      • Snyder S.
      • Ladror U.
      • Wade W.
      • Wang G.
      • Barrett L.
      • Matayoshi E.
      • Huffaker H.
      • Krafft G.
      • Holzman T.
      ). The time course of aggregation of Aβ1-40 in PBS at 37°C, evaluated by SDS-PAGE, indicated that our freshly prepared peptide was mainly monomeric while the number of dimers, tetramers, and higher association forms increased as a function of incubation time, reaching a plateau at 24 h that remained without major changes for at least 72 h. ApoJ exhibited five times lower affinity for aggregated Aβ1-40 than for freshly prepared peptide when tested by solid-phase ELISA and by gel electrophoresis followed by scanning evaluation. The presence of dimers, tetramers, and high molecular mass components indicated that apoJ was unable to reverse the aggregation of Aβ1-40. All these data suggest that apoJ has the capability to bind with high affinity to non-aggregated forms of Aβ1-40 but not to Aβ1-40 polymers.
      In order to balance the biological importance of the apoJ-sAβ interaction, competition experiments using other plasma/CSF proteins with demonstrated affinity to Aβ were carried out at physiologic pH. ApoE isoforms were 4-10 times less efficient than apoJ itself in inhibiting the formation of the complex apoJ•Aβ1-40, being apoE4 the least avid competitor under the conditions tested. The rest of the proteins assayed (TTR, ACT, and Vn) exhibited an almost negligible competitive effect; none of them induced more than 10% inhibition at concentrations where apoJ in fluid-phase inhibited 100% the complex formation. It should be noted that all the proteins used in these competition experiments were either purified from plasma or expressed in Sf9 cells. It would be interesting to determine whether other factors such as the association with other proteins, lipoproteins, and/or lipids may affect their interaction with Aβ. These data in the context of the mean physiologic concentrations of apoJ and apoE in plasma as well as in CSF (apoJplasma = 0.87 ± 0.43 μM; apoJCSF = 30 ± 15 nM; apoEplasma = 0.77 ± 0.26 μM; apoECSF = 41 ± 16 nM)(
      • Choi-Miura N.-H.
      • Ihara Y.
      • Fukuchi K.
      • Takeda M.
      • Nakano Y.
      • Tobe T.
      • Tomita M.
      ,
      • Murphy B.
      • Kirszbaum L.
      • Walker I.
      • d'Apice A.
      ,
      • Pitas R.
      • Boyles J.
      • Lee S.
      • Hui D.
      • Weisgraber K.
      ), suggest that normal physiologic conditions favor the formation of apoJ•Aβ complex. In fact, the presence of sAβ in apoJ-containing HDL particles was recently shown(
      • Koudinov A.
      • Matsubara E.
      • Frangione B.
      • Ghiso J.
      ).
      The data indicate the existence of a high affinity binding between apoJ and a peptide with identical primary structure to sAβ. It is conceivable that the interaction is not only related to the transport of the soluble peptide in plasma by apoJ-containing HDL but to the delivery of sAβ through the blood-brain barrier. In this regard, recent in vivo studies performed in guinea pigs have demonstrated the existence of cerebrovascular permeability for Aβ1-40, human apoJ as well as for the apoJ•Aβ1-40 complex(
      • Zlokovic B.
      • Mackic J.
      • Martell C.
      • Matsubara E.
      • Wisniewski T.
      • McComb G.
      • Frangione B.
      • Ghiso J.
      ,
      • Zlokovic B.
      • Ghiso J.
      • Mackic J.
      • McComb J.
      • Weiss M.
      • Frangione B.
      ). Since apoJ has lower affinity for aggregated Aβ, the biological implications for the alteration of the interaction apoJ-sAβ under pathologic conditions should be further investigated as one of the possible mechanisms of peptide aggregation and deposition in AD tissue.

      REFERENCES

        • Wisniewski T.
        • Ghiso J.
        • Frangione B.
        Neurobiol. A. 1994; 15: 143-152
        • Wisniewski T.
        • Lalowski M.
        • Levy E.
        • Marques M.
        • Frangione B.
        Ann. Neurol. 1994; 35: 245-246
        • Haass C.
        • Schlossmacher M.
        • Hung A.
        • Vigo-Pelfrey C.
        • Mellon A.
        • Ostaszewski B.
        • Lieberburg I.
        • Koo E.
        • Schenk D.
        • Teplow D.
        • Selkoe D.
        Nature. 1992; 359: 322-325
        • Seubert P.
        • Vigo-Pelfrey C.
        • Esch F.
        • Lee M.
        • Dovey H.
        • Davis D.
        • Sinha S.
        • Schlossmacher M.
        • Whaley J.
        • Swindlehurst C.
        • McCormack R.
        • Wolfert R.
        • Selkoe D.
        • Lieberburg I.
        • Schenk D.
        Nature. 1992; 359: 325-327
        • Shoji M.
        • Golde T.
        • Ghiso J.
        • Cheung T.
        • Estus S.
        • Shaffer L.
        • Cai X.-D.
        • McKay D.
        • Tintner R.
        • Frangione B.
        • Younkin S.
        Science. 1992; 258: 126-129
        • Wisniewski T.
        • Castano E.
        • Ghiso J.
        • Frangione B.
        Ann. Neurol. 1993; 34: 631-633
        • Abraham C.
        • Selkoe D.
        • Potter H.
        Cell. 1988; 52: 487-501
        • Coria F.
        • Castano E.
        • Prelli F.
        • Larrondo-Lillo M.
        • van Duinen S.
        • Shelanski M.
        • Frangione B.
        Lab. Invest. 1988; 58: 454-458
        • Snow A.
        • Willemer J.
        • Kisilevsky R.
        Lab. Invest. 1987; 57: 120-123
        • Choi-Miura N.-H.
        • Ihara Y.
        • Fukuchi K.
        • Takeda M.
        • Nakano Y.
        • Tobe T.
        • Tomita M.
        Acta Neuropath. 1992; 83: 260-264
        • McGeer P.
        • Kawamata T.
        • Walker D.
        Brain Res. 1992; 579: 337-341
        • Wisniewski T.
        • Frangione B.
        Neurosci. Lett. 1992; 135: 235-238
        • Eikelenboom P.
        • Stam F.
        Acta Neuropath. 1982; 57: 239-242
        • Ishii T.
        • Haga S.
        Acta Neuropath. 1984; 63: 296-300
        • Eikelenboom P.
        • Zhan S.
        • Kamphorst W.
        • van der Valk P.
        • Rozemuller J.
        Virc. Archiv. 1994; 424: 421-427
        • Gallo G.
        • Wisniewski T.
        • Choi-Miura N.-H.
        • Ghiso J.
        • Frangione B.
        Am. J. Pathol. 1994; 145: 526-530
        • Pericak-Vance M.
        • Bebout J.
        • Gaskell P.
        • Yamaoka L.
        • Hung W.
        • Alberts M.
        • Walker A.
        • Bartlett R.
        • Haynes C.
        • Welsh K.
        • Earl N.
        • Heyman A.
        • Clark C.
        • Roses A.
        Am. J. Hum. Genet. 1991; 48: 1034-1050
        • Ghiso J.
        • Matsubara E.
        • Koudinov A.
        • Choi-Miura N.-H.
        • Tomita M.
        • Wisniewski T.
        • Frangione B.
        Biochem. J. 1993; 293: 27-30
        • Strittmatter W.
        • Saunders A.
        • Schmechel D.
        • Pericak-Vance M.
        • Enghild J.
        • Salvesen G.
        • Roses A.
        Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 1977-1981
        • Wisniewski T.
        • Golabek A.
        • Matsubara E.
        • Ghiso J.
        • Frangione B.
        Biochem. Biophys. Res. Commun. 1993; 192: 359-365
        • LaDu M.
        • Falduto M.
        • Manelli A.
        • Reardon C.
        • Getz G.
        • Frail D.
        J. Biol. Chem. 1994; 269: 23403-23406
        • Schwarzman A.
        • Gregori L.
        • Vitek M.
        • Lyubski S.
        • Strittmatter W.
        • Enghilde J.
        • Bhasin R.
        • Silverman J.
        • Weisgraber K.
        • Coyle P.
        • Zagorski M.
        • Talafous J.
        • Eisenberg M.
        • Saunders A.
        • Roses A.
        • Goldgaber D.
        Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 8368-8372
        • Ma J.
        • Yee A.
        • Brewer H.
        • Das S.
        • Potter H.
        Nature. 1994; 372: 92-94
        • Schägger H.
        • von Jagow G.
        Anal. Biochem. 1987; 166: 368-379
        • Choi N.-H.
        • Tobe T.
        • Hara K.
        • Yoshida H.
        • Tomita M.
        J. Immunol. Methods. 1990; 131: 159-163
        • Bryan J.
        Anal. Biochem. 1977; 78: 513-519
        • Kim K.
        • Wen G.
        • Bancher C.
        • Chen C.
        • Sapienza V.
        • Hong H.
        • Wisniewski H.
        Neurosci. Res. Commun. 1990; 7: 113-122
        • Murphy B.
        • Kirszbaum L.
        • Walker I.
        • d'Apice A.
        J. Clin. Invest. 1988; 81: 1858-1864
        • de Silva H.
        • Stuart W.
        • Duvic C.
        • Wetterau J.
        • Ray M.
        • Ferguson D.
        • Albers H.
        • Smith W.
        • Harmony J.
        J. Biol. Chem. 1990; 265: 13240-13247
        • Kirszbaum L.
        • Bozas S.
        • Walker I.
        FEBS Lett. 1992; 297: 70-76
        • Slawin K.
        • Sawczuk I.
        • Olsson C.
        • Buttyan R.
        Biochem. Biophys. Res. Commun. 1990; 172: 160-164
        • Tobe T.
        • Minoshima S.
        • Yamase S.
        • Choi N.-H.
        • Tomita M.
        • Shimizu N.
        Cytogenet. Cell Genet. 1991; 57: 193-195
        • Purrello M.
        • Bettuzzi S.
        • Di Pietro C.
        • Mirabile E.
        • Di Blasi M.
        • Rimini R.
        • Grzeschik K.
        • Ingletti C.
        • Corti A.
        • Sichel G.
        Genomics. 1991; 10: 151-156
        • Kirszbaum L.
        • Sharpe J.
        • Murphy B.
        • d'Apice A.
        • Classon B.
        • Hudson P.
        • Walker I.
        EMBO J. 1989; 8: 711-718
        • Collard M.
        • Griswold M.
        Biochemistry. 1987; 26: 3297-3303
        • de Silva H.
        • Harmony J.
        • Stuart W.
        • Gil C.
        • Robbins J.
        Biochemistry. 1990; 29: 5380-5389
        • Sylvester S.
        • Morales C.
        • Oko R.
        • Griswold M.
        Biol. Reprod. 1991; 45: 195-207
        • O'Bryan M.
        • Baker H.
        • Saunders J.
        • Kirszbaum L.
        • Walker I.
        • Hudson P.
        • Liu D.
        • d'Apice A.
        • Murphy B.
        J. Clin. Invest. 1990; 85: 1477-1486
        • Hartmann K.
        • Rauch J.
        • Urban J.
        • Parczyk K.
        • Diel P.
        • Pilarsky C.
        • Appel D.
        • Haase W.
        • Mann K.
        • Weller A.
        • Koch-Brandt C.
        J. Biol. Chem. 1991; 266: 9924-9931
        • Palmer D.
        • Christie D.
        J. Biol. Chem. 1990; 265: 6617-6623
        • Choi N.-H.
        • Mazda T.
        • Tomita M.
        Mol. Immunol. 1989; 26: 835-840
        • Garden G.
        • Bothwell M.
        • Rubel E.
        J. Neurobiol. 1991; 22: 590-604
        • Danik B.
        • Chabot J.
        • Mercier C.
        • Benabid A.
        • Chauvin C.
        • Quirion R.
        • Suh M.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 8577-8581
        • Day J.
        • Laping N.
        • McNeill T.
        • Schreiber S.
        • Passinetti G.
        • Finch C.
        Mol. Endocrinol. 1990; 4: 1995-2002
        • Zlokovic B.
        • Mackic J.
        • Martell C.
        • Matsubara E.
        • Wisniewski T.
        • McComb G.
        • Frangione B.
        • Ghiso J.
        Biochem. Biophys. Res. Commun. 1994; 205: 1431-1437
        • Dugid J.
        • Bohmont C.
        • Liu N.
        • Tourtellotte W.
        Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 7260-7264
        • May P.
        • Lampert-Etchells M.
        • Johnson S.
        • Poirier J.
        • Masters J.
        • Finch C.
        Neuron. 1990; 5: 831-839
        • Buttyan R.
        • Olsson C.
        • Pintar J.
        • Chang C.
        • Bandyk M.
        • Ng P.
        • Sawczuk I.
        Mol. Cell Biol. 1989; 9: 3473-3481
        • May P.
        • Johnson S.
        • Poirier J.
        • Lampert-Etchells M.
        • Finch C.
        J. Neurol. Sci. 1989; 16: 473-476
        • Hilbich C.
        • Kisters-Woike B.
        • Reed J.
        • Masters C.
        • Beyreuther K.
        J. Mol. Biol. 1991; 218: 149-163
        • Burdick D.
        • Soreghan B.
        • Kwon M.
        • Kosmoski J.
        • Knauer M.
        • Henschen A.
        • Yates J.
        • Cotman C.
        • Glabe C.
        J. Biol. Chem. 1992; 267: 546-554
        • Snyder S.
        • Ladror U.
        • Wade W.
        • Wang G.
        • Barrett L.
        • Matayoshi E.
        • Huffaker H.
        • Krafft G.
        • Holzman T.
        Biophys. J. 1994; 67: 1216-1228
        • Pitas R.
        • Boyles J.
        • Lee S.
        • Hui D.
        • Weisgraber K.
        J. Biol. Chem. 1987; 262: 14352-14360
        • Koudinov A.
        • Matsubara E.
        • Frangione B.
        • Ghiso J.
        Biochem. Biophys. Res. Commun. 1994; 205: 1164-1171
        • Zlokovic B.
        • Ghiso J.
        • Mackic J.
        • McComb J.
        • Weiss M.
        • Frangione B.
        Biochem. Biophys. Res. Commun. 1993; 197: 1034-1040