ELSEVIER
Neuroscience Letters 220 (1996) 159-162
Enhanced aggregation of -amyloid-containing peptides by extracellular matrix and their degradation by the 68 kDa serine protease prepared from human brain A k i r a M a t s u m o t o a'*, T a i r a E n o m o t o a, Y o s h i s a d a F u j i w a r a a, H i s a m i t s u B a b a b, R e i k o M a t s u m o t o c aDepartment of Radiation Biophysics and Genetics, Kobe University School of Medicine, Kusunoki-cho 7-5-1, Chuo-ku Kobe 650, Japan bThird Division, Department of Internal Medicine, Kobe University School of Medicine, Kobe 650, Japan CFirst Division, Department of Internal Medicine, Osaka Teishin Hospital, Osaka 543, Japan Received 14 October 1996; revised version received 30 October 1996; accepted 11 November 1996
Abstract
To explore whether extracellular matrix components in human brain affect the deposition and aggregation of 13-amyloid containing peptides, human brain samples from patients with sporadic Alzheimer's disease and normal aged were analyzed by Western blot analysis. All major/3-amyloid-containing peptides contained epitope(s) which is recognized by anti heparan sulfate antibody. Incubation of brain /~-amyloid-containing peptides with human collagen type IV in neutral pH efficiently generated a high molecular weight aggregated band, approximately 5-fold that of the control sample. We have previously found a serine protease which is capable of cleaving an oligopeptide at the N-terminus of/3-amyloid. In this study, the protease, which also contains heparan sulfate glycoconjugates, degraded the above brain peptides as natural substrates, although with different efficiency. These findings suggest that extracellular matrix components affect the processing and aggregation of/~-amyloid-containing peptides in human brain. © 1996 Elsevier Science Ireland Ltd. All rights reserved
Keywords: 13-Amyloid;Collagen type IV; Serine protease Alzheimer's disease; Aggregation; Extracellular matrix
Abnormal deposition of fibrous ~-amyloid (A~) as senile plaque and cerebrovascular amyloid is regarded as one of the most essential neuropathological findings in Alzheimer's disease (AD). Mounting evidence suggests that missense mutations in and around AB are responsible for several types of familial AD and related diseases [3,6,20], and transgenic mice homozygous for APP717 missense mutation exhibits AD-mimicking histopathology [5]. More recently, two causative genes for early-onset type familial AD have been identified as S 182 on chromosome 14 [17] and STM-2 on chromosome 1 [10]. The putative gene products, presenilin-1 for S 182 and presenilin-2 for STM-2, are quite homologous membrane proteins on Golgi apparatus and presumably participate in trafficking and processing of newly synthesized proteins [10,17]. These results suggest that impairment in processing and/or trafficking of B-amyloid precursor protein (APP) and A~ is essentially linked to the pathophysiology of AD. On * Corresponding author. Fax: +81 78 67055806.
the other hand, various factors recently recognized as 'pathological chaperones' have been identified, such as ~l-antichymotrypsin, apolipoprotein E and sulfated glycoconjugates. These factors are considered to enhance A~ aggregation in extracellular space, such as extracellular matrix around senile plaque and interstitial matrix of cerebral vessels, where initial deposition of AB takes place in the brains of early pre-clinical stage AD and Down's syndrome patients. Especially, components of extracellular matrix are important because they regulate APP expression and AB metabolism [1,9], sulfate-containing glycoconjugates directly bind A~ [2] and the interaction with cells plays a crucial role in the regulation of tissue-specific functions [7]. We have found a serine protease in the extracellular fluid of familial AD cells, which cleaves a synthetic oligopeptide (P18, corresponding to -5th -12th amino acids around A/~) at its/3A4-N-terminus [11,12]. Among accumulating A/3-containing peptides in human lymphoid cells, the 16 kDa peptide with its N-terminus 30 amino
0304-3940/96/$12.00 © 1996 Elsevier Science Ireland Ltd. All rights reserved Pll S0304-3940(96) 13235-3
160
A. Matsumoto et aL / Neuroscience Letters 220 (1996) 159-162
acids upstream of the A/3-N-terminus is a natural substrate for this protease [13,14]. And this protease, which is also expressed in the brains of AD patients, contains heparan sulfate glycoconjugates [16]. The present study was carried out to analyze the effect of human extracellular matrix on the aggregation and degradation of natural A/3-containing peptides. Human brain cortex samples were obtained in autopsy within 5 h after death (60 year old male, sporadic AD sample 1; 72 year old female, sporadic AD sample 2; 82 year old female, apoplexy, normal sample 1; 58 year old male, thrombosis after cardiac infarction, normal aged sample 2). Temporal lobe slices were frozen in liquid nitrogen and stored at -95°C. Secreted proteins of lymphoid cells derived from familial AD patients were prepared as previously described [12]. For preparation of brain homogenate protein, frozen brain block (4 g) was thawed in 20 ml of sterile homogenization buffer consisting of 50 mM Tris-HC1, pH 7~6, 1 mM EDTA, 10 mM phenylmethanesulfonyl fluoride, and homogenized in Dounce homogenizer on ice. The homogenate was centrifuged at 39 000 g for 30 min after which the supernatant was precipitated with 40% ammonium sulfate. The precipitate was dialyzed against ice-cold homogenization buffer (2 1) for 3 h and used as brain extract, Some of the protein samples were further purified by immunoprecipitating the above extract using anti A/3 1-14 antibody and applied a limit filtration column (Ultrafree~ C3 Plus, Millipore, USA) to prepare lower molecular weight peptides with molecular mass less than 30 kDa. Anti A/3 8-17 polyclonal antibody, which was raised against 8-17th amino acid residues of A/3, was prepared as described elsewhere [12]. Anti heparan sulfate murine monoclonal antibody which was raised against a human fibroblast heparan sulfate epitope 10E4 (Seikagaku Inc., Japan) and anti A/3 1-14 polyclonal antibody which was raised against 1-14th residues of A/3 (Chemicon Inc., USA), were purchased. The human brain 68 kDa serine protease used in the cleavage analysis was prepared by immunoprecipitating the brain extract from a sporadic AD patient (AD sample 1) using adp°2 antibody as previously described [15]. The yield of this protease from l mg of brain extract was approximately 1 pmol. The brain homogenate substrate for the cleavage experiment was the same as above, except that protease inhibitors were removed from the dialysis buffer. The substrate (10-20/~g) was incubated with the protease (approximately 0.1 pmol) in inhibitor-free homogenization buffer supplemented with 2 mM CaC12 at 37°C for 24 h. Anti A/3 1-14-immunoprecipitate from brain homogenate protein (100 #g) or lymphocyte cytosol (180 #g) was incubated with 10 ~g human collagen type IV (Sigma Chemical Inc., USA), 10 #g human fibronectin (Life Technologies Inc., USA), 10/zg human laminin (Sigma Chemical Inc., USA) or 10 #g human heparan sulfate pro-
teoglycan (Sigma Chemical Inc., USA) at 37°C for 1824 h. The incubation was carried out at neutral pH (6.77.6), resulting from a mixture of the buffer system of substrate protein and those of extracellular matrix proteins determined by suppliers. Electrophoretic separation was carried out by SDS-PAGE of a Tris-tricine system [13]. The gel was then electroblotted to a sheet of Hybond C nitrocellulose membrane (Amersham, UK). Non-specific binding sites in the membrane were blocked by immersing it in 3% bovine serum albumin in Tris-buffered saline (100 mM Tris-C1, pH 7.5 and 140 mM NaCI)/0.1% Tween 20 (TBS-T) for 1 h at 25°C, after which the membrane was washed with three changes of TBS-T for 15 min. The washed membrane was then incubated with the first antibody (A/3 8-17 antibody at a 1:3000 dilution, or anti heparan sulfate antibody and anti A/3 1-14 antibody at 1:1500 dilution) in TBS-T for 1 h at room temperature. After washing as described above, the membrane was incubated with a horseradish peroxidase-labeled anti rabbit second antibody at a 1:2000 dilution for 1 h at room temperature. After washing, signals were detected on an X-ray film using an ECL chemiluminescence detection system (Amersham, UK). Fig. 1 shows Western blot analysis of immunoprecipitated A/3-harboring peptides using anti A¢3 1-14 antibody (Fig. 1A) or anti A/3 8-17 antibody (Fig. 1B). Low molecular weight (smaller than 30 kDa) brain sample from sporadic AD patient sample 1 exhibits 14 and 16 kDa bands (Fig. 1, lane 1), but in samples from normal aged, these bands are almost undetectable (Fig. 1, lane 3). The 16 kDa band co-migrated with the 16 kDa C-terminal peptides (CTPs) of lymphocyte cytosol from sporadic AD and normal aged (Fig. 1B, lanes 5-7). The 14 kDa peptide contains a larger amount of the A/3 1-14 epitope (Fig. IA, lane 1), but the 16 kDa peptide contains a relatively larger amount of the Aj3 8-17 epitope (Fig. IB, B
A kDa
1
2
3
4
1
2
3 4
6
6 7
6.'1-
Fig. 1. A/3-harbouringpeptides in brain homogenate and lymphocyte cytosol. Western blot analysis of a mixture of anti Af3 1-14- and anti A~38-17immunoprecipitatesof brain homogenateand lymphocytecytosol using anti A8 1-14 antibody(A) or anti A/~8-17 antibody(B).Lanes 1-2, Passed-throughlow molecularweight fraction(lane 1) and retained fraction (lane 2) of brain homogenatefrom sporadicAD sample 1; lanes 3-4, passed-through fraction (lane 3) and retained fraction (lane 4) of brain homogenatefrom normal aged sample 1; lanes 5-7, lymphocyte cytosol of sporadicAD sample 2 (lane 5), normal aged 1 (lane 6) and aporadic AD sample 1 (lane 7). Left ordinate indicates migrationpositions of molecularmass markers in kDa, and arrowheadsindicatemigration position of the 16 kDa A/3-harbouringpeptide.
161
A. Matsumoto et al. / Neuroscience Letters 220 (1996) 159-162
A
kDa12345
B12345
3958I 1
4-
e.~Fig. 2. Enhanced expression of aggregated A/3-harbouring peptides by extracellular matrix components. Western blot analysis of low molecular weight AB-harbouring peptides using anti A/3 1-14 antibody (A) or anti AB 8-17 antibody (B) after 24 h incubation with following components. Lane 1, No addition; lane 2, 10/~g human collagen type IV; lane 3, 10/~g human fibronectin; lane 4, 10/zg human laminin; lane 5, 10 #g human heparan sulfate proteoglyan. Left ordinate, migration positions of molecular mass markers in kDa.
lane 2). This finding is in good agreement with the result of lymphoid cells derived from familial AD patients and normal subjects [11]. In cytosol of lymphoid cells, the 16 kDa C-terminal peptide with identified N-terminus (567th amino acid after APP695 numbering) by amino acid sequencing is the most abundant C-terminal peptide in AD cells [13]. Although the N-terminus of the brain 16 kDa peptide is not identified by amino acid sequencing, it is likely that the brain peptide contains the N-terminal residues flanking to the A/3 N-terminus. The amount of large molecular weight bands (larger than 30 kDa), including full-length APP and its fragment, is similar between the samples from AD and normal subjects (Fig. 1A, lanes 2,4). However, the amount of A/3 8-17 epitope in fulllength APP is reduced in the normal aged sample (Fig. 1B, lane 4). This may be due to normal c¢-secretase activity which is able to decrease A/3 8-17 epitope in normal brain, and/or glycosylation(s) which can mask this epitope. Fig. 2 indicates the effect of extracellular matrix components on the aggregation of A~-containing peptides derived from the brains of sporadic AD patients. In the absence of extracellular components, the brain A/3-con-
A kDa
123
B
12
C
12
396.1-
talning peptide forms a barely detectable amount of high molecular weight band (approximately 90 kDa) during 24 h incubation at 37°C in neutral pH (Fig. 2, lane 1). However, in the presence of extracellular components, increases are seen in the amount of the high molecular weight band in the analysis using anti AB 1-14 antibody (Fig. 2A) and anti A/3 8-17 antibody (Fig. 2B); the A~ peptide treated with collagen type IV for 24 h exhibits the largest amount of the high molecular weight band, approximately 5-fold that of the control sample (Fig. 2, lane 2). The sample treated with human fibronectin exhibits a high molecular weight band (approximately 50% more) compared with the control sample (Fig. 2, lane 3). Human laminin and heparan sulfate proteoglycan seem to have no effect (Fig. 2, lanes 4,5). There is no distinctive difference in the amount of anti A~ 1-14 and 8-17 epitopes of the aggregated band. Fig. 3 shows a cleavage experiment of brain AjS-harboring peptides derived from sporadic AD patients by the 68 kDa serine protease with B-secretase-like activity. Western blot analysis using anti A~ 1-14 antibody detects five major A~-harboring peptides; full-length APP (approximately 100 kDa) which is located at the origin of Tristricine gel, 69, 35, 27 and 16 kDa fragment (Fig. 3A). The same blot is further analyzed by anti A/3 8-17 antibody (Fig. 3B) and anti heparan sulfate antibody (Fig. 3C). It is important to note that the 27 and 16 kDa fragments are decreased in the analysis using anti AB 1-14 antibody (Fig. 3A, lane 1). This implies that these two fragments are different from the other three fragments in AB 1-14 epitope, possibly due to a relative abundance of its middle portion, including the c~-secretase site (QIS-K-LI7), of the extracellular domain of AB. In the incubation of the natural brain samples with 0.1 pmol 68 kDa serine protease for 24 h, these fragments are degraded with various efficiencies (Fig. 3A-C, lanes 1,2). Full-length APP, 27 and 16 kDa fragments are excellent substrates for the protease, whereas the 69 and 35 kDa fragments are less efficiently degraded. Like the protease, these five A/3-containing proteins are also detected by anti heparan sulfate antibody which specifically recognizes oligosaccharides containing N-acetylglucosamine residue with O 6-, 03- or O2-sulfation.
68-
14-
Table 1
3,5-
Substrate Fig. 3. Degradation of AB-harbouring peptides by human brain 68 kDa serine protease. Western blot analysis of immunoprecipitates of brain homogenate from sporadic AD sample 1 using anti AB 1-14 antibody (A), anti AB 8-17 antibody (B) or anti heparan sulfate oligosaccharide antibody (C). Lane 1, Immunoprecipitated peptides incubated for 24 h in reaction buffer; lane 2, immunoprecipitated peptides incubated with O.1 pmol protease for 24 b in reaction buffer; lane 3, 0.1 pmol protease incubated for 24 h in reaction buffer. Left ordinate indicates migration positions of molecular mass markers in kDa, and arrowhead indicates migtarion position of the 16 kDa A~-harbouring peptide.
Full-length APP 69 kDa fragment 35 kDa fragment 27 kDa fragment 16 kDa fragment
Efficiency of removal (%)a A/~ 1-14 epitope
A/3 8-17 epitope
Heparan sulfate epitope
26 1 100 33 100
15 14 100 65 100
50 39 100 62 90
apercentage of reduction in intensities of densitometry tracings shown in Fig. 4 during the proteolytic degradation.
162
A. Matsumoto et al. / Neuroscience Letters 220 (1996) 159-162
This result indicates that the above fragments contain the oligosaccharide epitope, which is in good agreement with the recent data showing that g l u c o s a m i n o g l y c a n attachm e n t site ( G A G site) of A P P is the 5 - 9 t h a m i n o acid residues of A/3 [19]. The degradation kinetics of each A P P fragment are presented in Table 1. In addition to the G A G site, it was recently clarified that a n u m b e r of heparan sulfate proteoglycans bind Aft at the putative heparin b i n d i n g motif ( H l 3 - H - Q - K 16) [18]. The 27 kDa fragment has a high ratio of A ~ 8 - 1 7 epitope versus A ~ 1 - 1 4 epitope (Fig. 3 A,B). This fragment, p r e s u m a b l y rich in the heparin-binding motif, is poorly degraded by the protease, which is most obvious in the analysis of the A/3 1 - 1 4 epitope (Fig. 3A). These findings suggest that the moieties of this particular sugar are important in the proper proteolysis of A/3-containing peptides in h u m a n brain. Heparan sulfate glycoconjugates are attached to proteases participating in blood coagulation [4]. It is postulated that they facilitate protease-substrate and protease-inhibitor bindings, restrict proteolysis topologically and provide intracellular storage of proteases in inactive form [8l. In rat dorsal root ganglion neuron, A ~ plays a n o r m a l physiological role by complexing with extracellular matrix c o m p o n e n t s [9]. It is likely that in h u m a n brain also, the topological distribution of collagen type IV or heparan sulfate glycoconjugates affect the physiological processing aggregation and its degradation of A/3-harboring peptides. Although the pathophysiological significance of the serine protease has yet to be clarified, an u n d e r s t a n d i n g of its interaction with extracellular matrix is of crucial importance. This work was supported by a G r a n t - i n - A i d for Priority Areas of Scientific Research, 08278217, from the Ministry of Education, Science, Sports and Culture of Japan. [1] Bronfman, F.C., Soto, C., Tapia, L., Tapia, V. and Inestrosa, N.C., Extracellular matrix regulates the amount of the/3-amyloid precursor protein and its amyloidogenic fragments, J. Cell. Physiol., 166 (1996) 360-369. [2] Buee, L., Ding, W., Anderson, J.P., Narindrasorasak, S., Kisilevsky, R., Boyle, N.J., Robakis, N.K., Delacourte, A., Greenberg, B. and Fillit, H.M., Binding of vascular heparan sulfate proteoglycan to Alzheimer's amyloid precursor protein is mediated in part by the N-terminal region of A4 peptide, Brain Res., 627 (1993) 199-204. [3] Chartier-Harlin, M.-C., Crawford, F., Houlden, H., Warren, A., Hughes, D., Fidani, L., Goate, A., Rosser, M., Roques, P., Hardy, J. and Mullan, M., Early-onset Alzheimer's disease caused by mutations at codon 717 of the/3-amyloid precursor protein gene, Nature, 353 (1991) 844-846. [4] Danielsson, A., Raub, E., Lindahl, U. and Bjrrk, I., Role of temary complexes, in which heparin binds both antithrombin and proteinase, in the asseleration of the reactions between antithrombin and thrombin or factor Xa, J. Biol. Chem., 261 (1986) 15467-15473. [5] Games, D., Adams, D., Alessandrini, R., Barbour, R., Berthelette, P., Blackwell, C., Can', T., Clemens, J., Donaldson, T., Gillespie, F., Guido, T., Hagopian, S., Johnson-Wood, K., Kahn, K., Lee, M.,
Leibowitz, P., Lieberburg, I., Little, S., Masliah, E., McConlogue, L., Montoya-Zavala, M., Mucke, L., Paganini, L., Penninlan, E., Power, M., Schenk, D., Seubert, P., Snyder, B., Sofia.no, F., Tan, H., Vitale, J., Wadsworth, S., Wolozin, B. and Zhao, J., Alzheimertype neuropathology in transgenic mice overexpressing V717F flanwloid precursor protein, Nature, 373 (1995) 593-597. [6] Haass, C., Lemere, C.A., Capell, A., Citron, M., Seubert, P., Schenk, D., Lannfelt, L. and Selkoe, D.J., The Swedish mutation causes early-onset Alzheimer's disease by fl-secretase cleavage within the secretory pathway, Nat. Med., 1 (1995) 1291-1296. [7] Hynes, R.O. and Lander, A.D., Contact and adhesive specificities in the associations, migrations and targeting of cells and axons, Cell, 68 (1992) 302-322. [8] Kjellen, L. and Lindahl, U., Proteoglycans: structures and interactions, Annu. Rev. Biochem., 60 (1991) 443-475. [9l Koo, E.H., Park, L. and Selkoe, D.J., Amyloid/3-protein as a substrate interacts with extracellular matrix to promote neurite outgrowth, Proc. Natl. Acad. Sci. USA, 90 (1993) 4748-4752. [10] Levy-Lehad, E., Wijsman, E.M. and Nemens, E., Anderson, L., Goddard, K.A.B., Weber, J.L., Bird, T.D. and Schellenberg, G.D., A familial Alzheimer's disease locus on chromosome 1, Science, 269 (1995) 970-973. [11] Matsumoto, A. and Fujiwara, Y., Aberrant proteolysis of the /3amyloid precursor protein in familial Alzheimer's disease lymphoblastoid cells, Eur. J. Biochem., 217 (1993) 21-27. [12] Matsamoto, A. and Fujiwara, Y., Ca2+-dependent 68-kilodalton protease in familial Alzheimer's disease cells cleaves the N-terminus of/3-amyloid, Biochemistry, 33 (1994) 3941-3948. [13] Matsumoto, A. and Matsumoto, R., Familial Alzheimer's disease cells abnormally accumulate /3-amyloid-harbouring peptides preferentially in cytosol but not in extracellular fluid, Eur. J. Biochem., 225 (1994) 1055-1062. [14] Matsumoto, A., Matsumoto, R., Baba, H. and Fujiwara, Y., A serine protease in Alzheimer's disease cells cleaves a 16 K-peptide with flanking residues upstream to /3-amyloid-N-terminus as natural substrate, Neurosci. Lett., 195 (1995) 171-174. [15] Matsumoto, A., Matsumoto, R. and Fujiwara, Y., Molecular cloning of human cDNA with a sequence highly similar to that of the dihydrofolate reductase gene in brain libraries derived from Alzheimer's disease patients, Eur. J. Biochem., 230 (1995) 337343. [16] Matsumoto, A., Matsumoto, R., Enomoto, T. and Baba, H., Human brain /3-secretase contains heparan sulfate glycoconjugates, Neurosci. Lett., 211 (1996) 105-108. [17] Sherrington, R., Rogaev, E.I., Liang, Y., Rogaeva, E.A., Levesque, G., Ikeda, M., Chi, H., Lin, C., Li, G., Holman, K., Tsuda, T., Mar, L., Foncin, J.-F., Bruni, A.C., Montesi, M.P., Sorbi, S., Rainero, I., Pinessi, L., Nee, L., Chumakov, I., Pollen, D., Brookes, A., Sanseau, P., Polinsky, R.J., Wasco, W., Da Silva, H.A.R., Haines, J.L., Pericak-Vance, M.A., Tanzi, R.E., Roses, A.D., Fraser, P.E., Rommens, J.M. and St George-Hyslop, P.H., Cloning of a gene bearing missense mutations in early-onset familial Alzheimer's disease, Nature, 375 (1995) 754-760. [18] Snow, A.D., Kinsella, M.G., Parks, E., Sekiguchi, R.T., Miller, J.D., Kimata, K. and Wight, T.N., Differential binding of vascular cell-derived proteoglycans (perlecan, biglycan, decorin and versican) to the beta-amyloid protein of Alzheimer's disease, Arch. Biochem. Biophys., 320 (1995) 84-95. [19] Su, J.H., Cummings, B.J. and Cotman, C.W., Localization of heparan sulfate glycosaminoglycan and proteoglycan core protein in aged brain and Alzheimer's disease, Neuroscience, 51 (1992) 801-813. [20] Van Broeckhoven, C., Hann, J., Bakker, E., Hardy, J.A., Van Hul, W., Wehnert, A., Vecter-Van der Vis, M. and Roos, R.A.C., Amyloid 13protein precursor gene and hereditary cerebral hemorrhage with amyloidosis (Dutch), Science, 248 (1990) 1120-1122.