Báo cáo khoa học: Collagenous Alzheimer amyloid plaque component assembles amyloid fibrils into protease resistant aggregates pdf

Collagenous Alzheimer amyloid plaque component assembles amyloid fibrils into protease resistant aggregates Linda So ¨ derberg 1, *, Camilla Dahlqvist 1, *, Hiroyoshi Kakuyama 1 , Johan Thyberg 2 , Akira Ito 3 , Bengt Winblad 1 , Jan Na ¨ slund 1 and Lars O. Tjernberg 1 1 Karolinska Institutet and Sumitomo Pharmaceuticals Alzheimer Center (KASPAC), Neurotec, Novum, Huddinge, Sweden 2 Karolinska Institutet, Department of Cell and Molecular Biology, Stockholm, Sweden 3 Sumitomo Pharmaceuticals Research Center, Osaka, Japan Alzheimer’s disease (AD) is characterized by amyloid deposits of the amyloid b-peptide (Ab) in brain [1]. Ab is a 40–42 amino acid peptide that is proteolytically derived from the b-amyloid precursor protein (APP). The longer Ab42 variant aggregates more rapidly than the more abundant Ab40 [2] and is the species deposited initially in the brain in AD and Down’s syndrome [3]. Several lines of evidence suggest that APP processing and Ab levels have a central role in the pathogenesis of AD. The APP gene is located on chromosome 21 (which is present in triplicate in Down’s syndrome), providing an explanation for the elevated levels of APP and Ab, as well as for the AD-like pathology, observed in Down’s syndrome. Moreover, mutations in genes linked to familiar early onset AD generally result in altered APP processing and an increased Ab42 ⁄ Ab40 ratio [4]. It is not known how the amyloid plaques are formed. In vitro studies indicate that unstructured Ab mono- mers spontaneously form soluble oligomers (seeds), which, in turn, form protofibrils and mature fibrils. Other molecules could be of importance for the poly- merization process in vivo, including apolipoproteins E Keywords Alzheimer’s disease; amyloid; CLAC; fibrils; thioflavin T Correspondence C. Dahlqvist, Karolinska Institutet, Neurotec, Novum KASPAC pl. 5, SE-141 57 Huddinge, Sweden Fax: +46 8585 836 10 Tel: +46 8585 836 21 E-mail: Camilla.dahlqvist@neurotec.ki.se *Note These authors contributed equally to this work. (Received 28 January 2005, revised 25 February 2005, accepted 7 March 2005) doi:10.1111/j.1742-4658.2005.04647.x Recently, a novel plaque-associated protein, collagenous Alzheimer amy- loid plaque component (CLAC), was identified in brains from patients with Alzheimer’s disease. CLAC is derived from a type II transmembrane colla- gen precursor protein, termed CLAC-P (collagen XXV). The biological function and the contribution of CLAC to the pathogenesis of Alzheimer’s disease and plaque formation are unknown. In vitro studies indicate that CLAC binds to fibrillar, but not to monomeric, amyloid b-peptide (Ab). Here, we examined the effects of CLAC on Ab fibrils using assays based on turbidity, thioflavin T binding, sedimentation analysis, and electron microscopy. The incubation of CLAC with preformed Ab fibrils led to increased turbidity, indicating that larger aggregates were formed. In sup- port of this contention, more Ab was sedimented in the presence of CLAC, as determined by gel electrophoresis. Moreover, electron microscopy revealed an increased amount of Ab fibril bundles in samples incubated with CLAC. Importantly, the frequently used thioflavin T-binding assay failed to reveal these effects of CLAC. Digestion with proteinase K or tryp- sin showed that Ab fibrils, incubated together with CLAC, were more resistant to proteolytic degradation. Therefore, CLAC assembles Ab fibrils into fibril bundles that have an increased resistance to proteases. We sug- gest that CLAC may act in a similar way in vivo. Abbreviations Ab, amyloid b-peptide; AD, Alzheimer’s disease; APP, b-amyloid precursor protein; CLAC, collagenous Alzheimer amyloid plaque component ⁄ collagen XXV; EM, electron microscopy; NAC, non-amyloid-b-component; ThT, thioflavin T. FEBS Journal 272 (2005) 2231–2236 ª 2005 FEBS 2231 and J, and heparin sulfate proteoglycans [5]. These pro- teins have been suggested to act as ‘pathological chap- erones’ that bind to Ab and favor its deposition. Other proteins may have the opposite effect and attenuate amyloidogenesis. The amorphous Ab aggregates found in so-called diffuse plaques might be precursors of the Ab fibrils found in mature plaques, but the mechanism for this conversion remains to be determined. Recently, a novel plaque-associated protein, colla- genous Alzheimer amyloid plaque component ⁄ collagen XXV (CLAC) [6], was observed in brain from subjects with AD. CLAC is derived from a type II transmem- brane collagen protein, CLAC-P. We have recently shown that the AMY antigen is identical to CLAC [7]. CLAC is a trimer formed from three identical polypep- tides and includes three triple-helical collagen domains that are flanked and separated by nonhelical domains [8]. In vitro studies show that CLAC binds to fibrillized Ab, but not to monomeric Ab [6,8]. The biological function of CLAC, and the contribution of CLAC to plaque formation and the pathogenesis of AD, are still unknown. Previously, CLAC was shown to colocalize with the more mature plaques, but not with Ab42-pos- itive diffuse plaques or with amyloid deposits in cereb- ral blood vessels [9,10]. Recently, it was suggested that CLAC preferentially binds to plaques composed of prefibrillar Ab42 and thus may prevent further matur- ation of amyloid deposits [11]. Therefore, there is cur- rently no clear consensus regarding to which form of amyloid CLAC binds. Here, we examined the effects of CLAC on Ab fibrils in vitro using a v ariety of biochemical techniques. We showed that the addition of CLAC to Ab fibri ls a s sembles the latter i nto p rotease-resistant fibril bundles. W e suggest that CLAC could have a similar effect on amyloid fibrils in vivo an d thus i ncrease the am yloid burden. Results and Discussion Incubation of Ab1–40 fibrils with CLAC results in further aggregation The newly discovered plaque-associated protein, CLAC, binds to Ab fibrils but not to monomeric Ab [6] (H Kakuyama, L So ¨ derberg, K Horigome, C Dahlqvist, B Winblad, J Na ¨ slund & LO Tjernberg, unpublished results). In order to investigate the effect of CLAC on fibrillar Ab, fibrils formed from 45 lm Ab1–40 were incubated in the presence or absence of 100 nm CLAC. The turbidity of the samples was measured at different time-points and found to be increased after only 30 min in the presence of CLAC (Fig. 1A). As we use preformed fibrils that are separated from soluble Ab by centrifugation, the total amount of aggregated Ab cannot increase. Therefore, we conclude that the increase in turbidity arises from the formation of lar- ger aggregates. To verify this result, we centrifuged the samples and subjected them to SDS ⁄ PAGE. In the presence of CLAC, the amount of Ab was increased in the pellet and decreased in the supernatant as com- pared to the control (Fig. 1B). This effect occurred after only 30 min of incubation and remained evident also after a longer incubation (18 h). We suggest that A C B Fig. 1. Incubation of amyloid b-peptide 1–40 (Ab1–40) fibrils with collagenous Alzheimer amyloid plaque component ⁄ collagen XXV (CLAC) results in further aggregation. (A) Turbidity measurements of Ab1–40 fibrils incubated, for 30 min and 18 h, in the presence (100 n M) or absence of CLAC. (B) Coomassie staining of Ab1–40 fibrils incubated in the presence of CLAC (100 n M) or NaCl ⁄ Tris. The pellet samples (P) were sedimented, washed and dissolved in 70% (v ⁄ v) formic acid (FA). The supernatant samples (Sup) were sedimented from the supernatant of the first centrifugation and dis- solved in 70% (v ⁄ v) FA. (C) Electron micrographs of mixtures con- taining preformed Ab1–40 fibrils incubated in the presence of CLAC (100 n M) or NaCl ⁄ P i for 24 h at room temperature. Samples were prepared for electron microscopy as described in the Experimental procedures. Scale bars, 200 nm. Studies on CLAC and Ab interactions in vitro L. So ¨ derberg et al. 2232 FEBS Journal 272 (2005) 2231–2236 ª 2005 FEBS the increased amount of sedimented material is caused by an increase in the size of the aggregates in the pres- ence of CLAC. No CLAC could be detected in the supernatant after incubation, and CLAC was recovered in the carefully washed pellet, as analyzed by western blot and mass spectrometry (data not shown). These data correlate with the turbidity measurements and sup- port a function for CLAC in fibril assembly. The con- trol protein, aldolase, which does not bind to Ab fibrils, had no such effect (data not shown). To investigate whether the effect of CLAC was specific for Ab fibrils, we performed the same set of experiments on fibrils formed from another amyloid forming protein, non- amyloid-b-component (NAC) [12]. CLAC was found to have a similar effect on NAC fibrils, indicating that CLAC could affect other fibrils in a similar way (data not shown). Preformed fibrils incubated for 24 h in the presence or absence of CLAC were subjected to electron microscopy (EM) analysis. A significant increase in the number of large aggregates was observed in the presence of CLAC (Fig. 1C). Thus, EM data also support the notion that CLAC can assemble Ab fibrils into larger aggregates. A similar function has not been described for other proteins that are known to bind Ab fibrils, such as apolipoprotein E or laminin. CLAC protects fibrillar Ab1–40, but not soluble Ab1–40, from proteolysis To investigate the biological relevance of CLAC’s assembly of preformed fibrils, we used proteinase K and trypsin to determine whether the interaction between CLAC and Ab fibrils results in protection of the peptide from proteolysis. Both fibrillar Ab1–40 and soluble Ab1–40 were tested for protease resistance in the pres- ence or absence of CLAC. CLAC and fibrillar Ab were incubated at room temperature for 2 h before the addi- tion of proteases. The extent of Ab proteolysis was eval- uated by SDS ⁄ PAGE and western blot analysis. When the fibrils were incubated without CLAC, only a small amount of Ab remained after 2 h of digestion with pro- teinase K, and the samples were completely degraded after 18 h of digestion (Fig. 2A). In contrast, when the fibrils were preincubated with CLAC, Ab was still pre- sent after 18 h of digestion (Fig. 2A). Digestion with trypsin was less efficient but, in this case, CLAC had a protective effect (Fig. 2B). In contrast, CLAC did not have a protective effect on soluble Ab1–40 (Fig. 2C). Our findings offer an explanation for the increased pro- tease resistance of CLAC positive plaques [13]. There- fore, we speculate that CLAC assembles Ab fibrils into protease-resistant aggregates in vivo and thereby obstructs the clearance of amyloid. Fibril assembly results in decreased thioflavin T (ThT) fluorescence One of the most frequently used methods for studying Ab polymerization is based on the altered fluorescence of ThT upon binding to amyloid aggregates. Therefore, we investigated whether the ThT assay could be used to study the CLAC-induced assembly of Ab fibrils. When Ab fibrils are incubated in the presence of CLAC there is a striking reduction in ThT fluorescence (Fig. 3A), which is not accompanied by a reduction in the amount of sedimented Ab, as analyzed by SDS ⁄ PAGE and staining with Coomassie blue (Fig. 3B). Similar results were obtained after incubation with the positive control, laminin, while incubation with the negative control, aldolase, had no effect on ThT fluorescence (data not shown). When freshly solubilized Ab1–40 was incubated for 5 days, a similar reduction in ThT fluorescence was observed in the presence of CLAC (Fig. 4A). No reduc- tion in the amount of fibrils was observed by EM, but the fibrils appeared to aggregate more in the presence of CLAC (Fig. 4B). As CLAC binds to Ab fibrils, but not to freshly dissolved Ab, CLAC does not affect the lag- phase (Fig. 4A), and the reduction in ThT fluorescence at the end-point is probably a result of the effect of A B C Fig. 2. Collagenous Alzheimer amyloid plaque component ⁄ collagen XXV (CLAC) protects fibrillar amyloid b-peptide 1–40 (Ab1–40) from proteolysis. (A) Fibrillar Ab1–40 was preincubated with CLAC (at a final concentration of 200 n M) for 2 h before the addition of protein- ase K (PK) at an enzyme ⁄ Ab ratio of 1 : 10 (w ⁄ w). Samples were taken at the time-points indicated and analyzed on a 10–20% (w ⁄ w) tricine gel, stained with Coomassie blue, and immunoblotted using 4G8 and 6E10 antibodies. (B) Fibrillar Ab1–40 treated with trypsin. (C) Freshly solubilized Ab1–40 (50 l M) was preincubated with CLAC (at a final concentration of 200 n M), digested with PK and analyzed as described in A. L. So ¨ derberg et al. Studies on CLAC and Ab interactions in vitro FEBS Journal 272 (2005) 2231–2236 ª 2005 FEBS 2233 CLAC on the fibrils formed during polymerization. If the reduced ThT fluorescence is caused by the increased fibril assembly in the presence of CLAC, no reduction in fluorescence should be observed if the fibrils are immo- bilized prior to the addition of CLAC. When Ab fibrils were immobilized onto microtiter wells, CLAC had no effect on ThT fluorescence (Fig. 4C). Thus, the assembly of fibrils leads to a reduction in ThT fluorescence, poss- ibly because of a decrease in the number of ThT-binding sites and ⁄ or quenching owing to a high concentration of ThT in the bundles. This is in keeping with a previous report [14]. Based on ThT fluorescence analysis, a num- ber of compounds have been reported to depolymerize fibrillized Ab within a few hours [15,16]. In the light of our present study, we would like to point out that the results from ThT-binding studies must be evaluated carefully and supplemented with data obtained by using alternative techniques. In summary, we have shown that CLAC assembles Ab fibrils into protease-resistant aggregates. We speculate that this process may be of relevance for AD pathogenesis, offering one explanation for the accumulation of amyloid plaques in vivo. Future studies in mice with altered CLAC expression will investigate this hypothesis. Experimental procedures Preparation of fibrillar Ab Ab1–40 and A b 1–42 were purchased from Bachem (Buben- dorf, Switzerland). Ab fibrils were prepared by mixing lyophilized peptide with ultrapure water to obtain a peptide A B Fig. 3. Reduction in thioflavin T (ThT) fluorescence in the presence of collagenous Alzheimer amyloid plaque component ⁄ collagen XXV (CLAC) does not correlate with the amount of sedimented amyloid b-peptide (Ab). (A). Preformed Ab1–40 fibrils incubated in the pres- ence (100 n M) or absence of CLAC were mixed with 200 lLof 10 l M ThT for 15 min. Each data point represents the mean ± SEM of a triplicate result of a representative experiment. (B) The sam- ples (one of the triplicates) were centrifuged, washed, dissolved in 70% (v ⁄ v) formic acid and vacuum dried, then loaded onto a 10–20% (w ⁄ w) tricine gel and stained with Coomassie blue after electrophoresis. The samples from time-points 0, 0.5 and 1 h, for fibrils incubated either with CLAC or with buffer, were analyzed on the same gel to allow comparison of the fibril amount. The samples from 2, 3 and 4 h were analyzed on another gel. A B C Fig. 4. Effect of collagenous Alzheimer amyloid plaque compo- nent ⁄ collagen XXV (CLAC) on the polymerization of amyloid b-pep- tide 1–40 (Ab1–40). (A) A concentration of 25 l M freshly dissolved Ab1–40 in NaCl ⁄ Tris, pH 7.4, was incubated for 5 days at room temperature, with shaking (600 r.p.m.), in the presence or absence of CLAC (final concentration 100 n M). A molar ratio of Ab1– 40 ⁄ CLAC of 250 : 1 was used. Fibril formation was monitored by ThT fluorescence. Each data point represents the mean ± SEM of three separate experiments. (B) Negative stain electron micro- graphs of the 5 day incubation mixtures in (A). Ab1–40 fibrils in the absence of CLAC (left panel) or in its presence (right panel). Scale bar, 100 nm. (C) One nanomol of Ab 1–42 fibrils was allowed to dry onto microtiter wells and incubated with CLAC (a final concentra- tion of 250 n M), followed by incubation with a thioflavin T (ThT) solution of 10 or 100 l M. Fluorescence was measured with an exci- tation of 440 nm and an emission of 490 nm. Studies on CLAC and Ab interactions in vitro L. So ¨ derberg et al. 2234 FEBS Journal 272 (2005) 2231–2236 ª 2005 FEBS concentration of 200 lm. This solution was subsequently stirred gently for 5 min before the addition of 2· NaCl ⁄ P i , pH 7.4, to a final concentration of 100 lm and then vigor- ously stirred for an additional 2–4 days. Before use in assays, fibrils were sedimented by centrifugation at 10 000 g for 10 min and the supernatant was replaced by an equal volume of NaCl ⁄ P i . Ab turbidity and detection Preformed Ab1–40 fibrils were mixed with 100 nm CLAC. The CLAC used in this study was purified from HEK293 cells, as recently described [8]. Ab1–40 fibrils and CLAC or NaCl ⁄ Tris, pH 7.4, were incubated at room temperature with shaking (600 r.p.m.). At the indicated time-points, sam- ples (50 lL) were subjected to turbidity measurements. Turbidity was measured at 355 nm (FLUOstar Galaxy; BMG Labtechnologies GmbH, Offenburg, Germany) and control values of the absorbance of buffer alone were sub- tracted from all measurements. To detect the amount of Ab fibrils after ThT fluorescence measurements, the samples were transferred into Eppendorf tubes and subjected to cen- trifugation for 5 min at 10 000 g. The resulting pellet was washed twice with NaCl ⁄ Tris, while the supernatant was transferred to a new tube and centrifuged for 5 min at 10 000 g. The supernatant was discarded and both pellets were dissolved in 70% (v ⁄ v) formic acid overnight. Samples were vacuum dried to remove the formic acid and then resuspended in 100 mm Tris, pH 10.5, containing 9 m urea, prior to analysis by Coomassie blue staining on a 10–20% (w ⁄ w) tricine gel. Electron microscopy Samples incubated for 24 h were vortexed, and 10 lL aliqu- ots of the sample were then applied to formvar-coated grids. After 5 min, excess fluid was withdrawn and the grids were allowed to dry. Buffer salts were removed by dipping the grids (10 times) into redistilled water. The specimens were negatively stained with 2% (w ⁄ v) uranyl acetate in water, examined in a Philips CM120 electron microscope at 80 kV, and photographed using a MegaView III CCD cam- era (Soft Imaging System, Mu ¨ nster, Germany). Proteolysis of the Ab1–40/CLAC complex Ab1–40 fibrils were prepared as described above. Fibrils were subjected to proteolysis using proteinase K (Sigma Aldrich, St Louis, MO, USA; Signet Laboratories, Ded- ham, MA, USA) and trypsin (Boehringer Mannheim, Milan, Italy) at an enzyme ⁄ protein ratio of 1 : 10 (w ⁄ w) for all experiments. CLAC (at a final concentration of 200 nm) was added to Ab1–40 fibrils (50 lm) or to freshly dissolved Ab1–40 (50 lm) in NaCl ⁄ P i and incubated for 2 h before the addition of proteinase K or trypsin. Aliquots were taken after 0, 2 and 18 h of incubation at 37 °C. The reac- tion was stopped by the addition of formic acid to a final concentration of 70%. Samples were treated, as described above, prior to SDS ⁄ PAGE analysis on 10–20% (w ⁄ w) tricine gels followed by staining with Coomassie blue and immunoblotting using 6E10 and 4G8 antibodies (Signet Laboratories). ThT assay The ThT-binding assay [17] was used to measure the amount of Ab1–40 fibrils in the presence of CLAC. Twenty microlitres of Ab, incubated in the presence or absence of CLAC, laminin or aldolase, was aspirated and further incu- bated for 15 min with 200 lLof10lm ThT solution (10 mm phosphate buffer, 150 mm NaCl, pH 6.0). Fluores- cence spectra of ThT were acquired using a fluorescence spectrometer (FLUOstar Galaxy), with excitation at 440 nm and emission at 490 nm. Control spectra of the ThT solution alone were recorded and subtracted from all measurements. The amount of Ab in the pellet fraction was determined as described above. For the polymerization assay, a stock solution of 1 mgÆmL )1 Ab1–40 in dimethyl- sulfoxide was diluted in NaCl ⁄ Tris to a final concentration of 25 lm.Ab polymerization was carried out in the pres- ence of CLAC (100 nm)orAb1–40 alone at room tempera- ture with continuous shaking at 600 r.p.m. At the indicated time-points, a 50 lL aliquot from each incubation mixture was analyzed for Ab fibril formation by mixing with 200 lL of ThT solution, and the fluorescence was measured as described above. Solid-phase ThT-binding assay A competition assay based on solid-phase binding was used, as previously described, with some minor modifica- tions [6,8]. Briefly, a dimethylsulfoxide stock solution of Ab1–42 was diluted in NaCl ⁄ P i to a concentration of 20 lm. One nanomol of Ab1–42 was allowed to bind to microtiter wells (MaxiSorp; Nunc, Naperville, IL, USA) and dry overnight, at 37 °C. Wells were blocked for 1 h in NaCl ⁄ P i containing 1% (w ⁄ v) BSA (blocking buffer) fol- lowed by washing with NaCl ⁄ P i -T [NaCl ⁄ Pi containing 0.05% (v ⁄ v) Tween-20 (Sigma)] in a microplate washer (ASYS Hitech; Atlantis, GmbH, Eugendolf, Austria). Ab was incubated with 250 nm CLAC in blocking buffer, or in blocking buffer only, for 1 h at room tempera- ture. The wells were washed in NaCl ⁄ Pi-T and incubated with NaCl ⁄ P i containing 10 or 100 lm ThT for 30 min fol- lowed by washing in NaCl ⁄ P i -T. One-hundred microlitres of NaCl ⁄ P i was added to the wells, and binding of ThT was measured in a microplate reader (FLUOstar Galaxy) with excitation at 440 nm and emission at 490 nm. L. So ¨ derberg et al. Studies on CLAC and Ab interactions in vitro FEBS Journal 272 (2005) 2231–2236 ª 2005 FEBS 2235 References 1 Selkoe DJ (2001) Alzheimer’s disease: genes, proteins, and therapy. Physiol Rev 81, 741–766. 2 Jarrett JT & Lansbury PT (1993) Seeding ‘one-dimen- sional crystallization’ of amyloid: a pathogenic mechan- ism in Alzheimer’s disease and scrapie? Cell 73, 1055– 1058. 3 Iwatsubo T, Odaka A, Suzuki N, Mizusawa H, Nukina N & Ihara Y (1994) Visualization of Abeta 42 (43) and Abeta 40 in senile plaques with end-specific Abeta monoclonals: evidence that an initially deposited species is Abeta 42 (43). Neuron 13, 45–53. 4 Hardy J (1997) Amyloid, the presenilins and Alzhei- mer’s disease. Trends Neurosci 20, 154–159. 5 Dickson DW (1997) The pathogenesis of senile plaques. J Neuropathol Exp Neurol 56 , 321–339. 6 Hashimoto T, Wakabayashi T, Watanabe A, Kowa H, Hosoda R, Nakamura A, Kanazawa I, Arai T, Takio K, Mann DM et al. (2002) CLAC: a novel Alzheimer amyloid plaque component derived from a transmem- brane precursor CLAC-P ⁄ collagen type XXV. EMBO J 21, 1524–1534. 7So ¨ derberg L, Zhukareva V, Bogdanovic N, Hashimoto T, Winblad B, Iwatsubo T, Lee VM, Trojanowski JQ & Na ¨ slund J (2003) Molecular identification of AMY, an Alzheimer disease amyloid-associated protein. J Neuro- pathol Exp Neurol 62, 1108–1117. 8So ¨ derberg L, Kakuyama H, Mo ¨ ller A, Ito A, Winblad B, Tjernberg LO & Na ¨ slund J (2005) Characterization of the Alzheimer’s disease-associated CLAC protein and identification of an amyloid beta-peptide binding site. J Biol Chem 280, 1007–1015. 9 Lemere CA, Grenfell TJ & Selkoe DJ (1999) The AMY antigen co-occurs with abeta and follows its deposition in the amyloid plaques of Alzheimer’s disease and down syndrome. Am J Pathol 155, 29–37. 10 Schmidt ML, Lee VM, Forman M, Chiu TS & Troja- nowski JQ (1997) Monoclonal antibodies to a 100-kd protein reveal abundant Abeta-negative plaques throughout gray matter of Alzheimer’s disease brains. Am J Pathol 151, 69–80. 11 Kowa H, Sakakura T, Matsuura Y, Wakabayashi T, Mann DM, Duff K, Tsuji S, Hashimoto T & Iwatsubo T (2004) Mostly separate distributions of CLAC- versus Abeta40- or thioflavin S-reactivities in senile plaques reveal two distinct subpopulations of beta-amyloid deposits. Am J Pathol 165, 273–281. 12 Ueda K, Fukushima H, Masliah E, Xia Y, Iwai A, Yoshimoto M, Otero DA, Kondo J, Ihara Y & Saitoh T (1993) Molecular cloning of cDNA encoding an unrecognized component of amyloid in Alzheimer disease. Proc Natl Acad Sci USA 90, 11282–11286. 13 Kowa H, Sakakura T, Matsuura Y, Wakabayashi T, Tsuji S, Mann DM, Hashimoto T & Iwatsubo T (2004) CLAC-positive senile plaques in Alzheimer brains: mutually exclusive distribution to Abeta40 and acquisi- tion of protease resistance. Abstracts from the 9th Inter- national Conference on Alzheimer’s Disease and Related Disorders, July 17–22, Abstract: P1–235. Neuro- biol Aging 25, S 163 (Suppl.), 15. 14 Moss MA, Varvel NH, Nichols MR, Reed DK & Rosenberry TL (2004) Nordihydroguaiaretic acid does not disaggregate beta-amyloid (1–40) protofibrils but does inhibit growth arising from direct protofibril association. Mol Pharmacol 66, 592–600. 15 Morgan C, Bugueno MP, Garrido J & Inestrosa NC (2002) Laminin affects polymerization, depolymerization and neurotoxicity of Abeta peptide. Peptides 23, 1229– 1240. 16 Ono K, Hasegawa K, Yoshiike Y, Takashima A, Yamada M & Naiki H (2002) Nordihydroguaiaretic acid potently breaks down pre-formed Alzheimer’s beta-amyloid fibrils in vitro. J Neurochem 81, 434–440. 17 LeVine III H (1993) Thioflavine T interaction with syn- thetic Alzheimer’s disease beta-amyloid peptides: detec- tion of amyloid aggregation in solution. Protein Sci 2, 404–410. Studies on CLAC and Ab interactions in vitro L. So ¨ derberg et al. 2236 FEBS Journal 272 (2005) 2231–2236 ª 2005 FEBS . Collagenous Alzheimer amyloid plaque component assembles amyloid fibrils into protease resistant aggregates Linda So ¨ derberg 1, *, Camilla Dahlqvist 1, *,. vivo. Abbreviations Ab, amyloid b-peptide; AD, Alzheimer s disease; APP, b -amyloid precursor protein; CLAC, collagenous Alzheimer amyloid plaque component ⁄ collagen XXV; EM, electron microscopy; NAC, non -amyloid- b -component; . Incubation of amyloid b-peptide 1–40 (Ab1–40) fibrils with collagenous Alzheimer amyloid plaque component ⁄ collagen XXV (CLAC) results in further aggregation. (A) Turbidity measurements of Ab1–40 fibrils