SMALL-MOLECULE INHIBITORS OF AGGREGATION INDICATE THAT AMYLOID BETA OLIGOMERIZATION AND FIBRILLIZATION

PATHWAYS ARE INDEPENDENT AND DISTINCT Mihaela Necula§, Rakez Kayed§, Saskia Milton§, and Charles G. Glabe§, *

From the §Department of Molecular Biology and Biochemistry, University of California, Irvine, California, 92697

Running title: Oligomer formation is not an obligatory step for amyloid beta fibrillization Address correspondence to: *Charles G. Glabe, Department of Molecular Biology and Biochemistry, 3438 McGaugh Hall, Irvine, CA, 92697, Tel.: 949-824-6081, Fax: 949-824-8551, E-Mail: cglabe@uci.edu Alzheimer’s disease is characterized by the abnormal aggregation of amyloid beta peptide into extracellular fibrillar deposits known as amyloid plaques. Soluble oligomers have been observed at early time points preceding fibril formation and these oligomers have been implicated as the primary pathological species rather than the mature fibrils. A significant issue that remains to be resolved is whether amyloid oligomers are an obligate intermediate on the pathway to fibril formation or represent an alternate assembly pathway that may or may not lead to fiber formation. To determine whether amyloid beta oligomers are obligate intermediates in the fibrillization pathway, we characterized the mechanism of action of amyloid beta aggregation inhibitors in terms of oligomer and fibril formation. Based on their effects, the small-molecules segregated into three distinct classes: compounds that inhibit oligomerization but not fibrillization, compounds that inhibit fibrillization but not oligomerization, and compounds that inhibit both. Several compounds selectively inhibited oligomerization at substoichiometric concentrations relative to amyloid beta monomer, with some active in the low nanomolar range. These results indicate that oligomers are not an obligate intermediate in the fibril formation pathway. In addition, these data suggest that small-molecule inhibitors are useful for clarifying the mechanisms underlying protein aggregation and may represent potential therapeutic agents that target fundamental disease mechanisms.

Protein aggregation into amyloid fibrils is a

pathological hallmark of many neurodegenerative

diseases, including Alzheimer’s disease (AD)1. AD is characterized, in part, by the aggregation of amyloid beta protein (Aβ) into fibrillar amyloid plaques in select areas of the brain (1). Compelling evidence indicates that Aβ aggregation is critical for neurodegeneration, suggesting that preventing this process may be an effective therapeutic approach for the treatment of AD (2-4).

A number of small-molecules have been reported to inhibit Aβ fibrillogenesis. Since it was initially presumed that toxicity is associated with mature fibers (5-8), the majority of inhibitor screens have been directed toward identifying modulators of Aβ fibrillization. These fibril inhibitor screens have resulted in the discovery of multiple inhibitor molecules (9-24). Some compounds have also been shown to inhibit Aβ-mediated cellular toxicity and this activity was correlated with modulation of fibrillization (13,15,16,21,25).

However, Aβ aggregation is a complicated process and appears to involve more than a simple conversion of soluble monomer to fiber. More recent evidence has pointed to the role of soluble amyloid oligomers or prefibrillar aggregation intermediates as the primary toxic species in degenerative amyloid diseases (2,3). Electron microscopy and atomic force microscopy have identified spherical particles of approximately 3-10 nm that appear at early times of incubation and disappear as mature fibrils appear (26). These spherical oligomers appear to represent intermediates in the pathway of fibril formation because they are transiently observed at intermediate times of incubation during fibril formation. Although oligomers are kinetic intermediates, it is not yet clear whether they are obligate intermediates in the pathway for fibril

1

http://www.jbc.org/cgi/doi/10.1074/jbc.M608207200The latest version is at JBC Papers in Press. Published on February 6, 2007 as Manuscript M608207200

Copyright 2007 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

formation, whether they coalesce directly to form fibrils (26,28-33), or oligomers populate a different aggregation pathway that is distinct from the classic nucleation-dependent fibril assembly pathway (34-36). This same controversy extends to the aggregation of many other amyloidogenic proteins, since many types of amyloids display the same type of kinetically unstable intermediate, the soluble oligomer. Insulin, immunoglobulin light chain, amylin, alpha synuclein, transthyretin, prion protein, β2-microglobulin, β lactoglobulin, phosphoglycerate kinase, and albebetin oligomers have been described as “off pathway” species (37-47). Oligomers of insulin, amylin, huntingtin, and albebetin have been reported to be “on pathway” intermediates or precursors of fibril formation (47-50). Therefore, the published data is equivocal as to whether oligomers are intermediates on the pathway leading to fiber formation or whether they represent “off pathway” aggregates that populate an alternative aggregation pathway (51).

Soluble Aß oligomers have been referred to by a variety of names, including amorphous aggregates, micelles, protofibrils, prefibrillar aggregates and ADDLs (26,52-57). At longer aggregation times, curvilinear fibers form that have a beaded appearance form. These structures have also been called “protofibrils” because they appear to be formed by the coalescence of the spherical subunits (26). These prefibrillar soluble oligomers are specifically recognized by a polyclonal antibody, A11, that recognizes a generic backbone epitope that is common to the oligomeric state independent of the protein sequence (58). A11 does not recognize Aß monomer, Aß dimer, trimer or tetramer or Aß fibrils (58). A11 positive oligomers display the same intermediate kinetics as observed for soluble oligomers and protofibrils by electron microscopy and atomic force microscopy and A11 blocks the toxicity of Aß oligomers, indicating that they represent the primary toxic species. Aß*56 is a soluble oligomeric form of Aß that is closely associated with pathogenesis in the Tg2576 mouse model of AD and Aß*56 is specifically recognized by A11 anti-oligomer antibody (59). Although ADDLs were originally described as low MW trimeric or tetrameric species, more recent investigations indicate that native masses of ADDLs are the same as previously reported for other Aß soluble oligomers (60). A11 and anti-

ADDL antibodies identify the same time course of soluble oligomer accumulation in the 3xTg-AD mouse (61). Therefore, A11 antibody recognizes a significant and important class of oligomers associated with AD that are distinct from amyloid fibrils.

Analysis of the mechanism of action of amyloid aggregation inhibitors holds the promise of clarifying the relationship between oligomers and fibrils, because if oligomers are obligate intermediates on the pathway to fibril formation, then all inhibitors that block oligomer formation would be expected to block fibril formation as well. Alternatively, if fibrils and oligomers represent distinct aggregation pathways, then it would be expected that some inhibitors would block oligomerization, but not fibril formation.

Here we utilize the anti-oligomer antibody, A11, as a primary read out for oligomer formation and we analyze the mechanism of action of small-molecules that have been reported to inhibit the aggregation of different amyloidogenic proteins or their toxicity. The results demonstrate that some inhibitors specifically target oligomers, while others specifically inhibit fibrillization. These data indicate that soluble oligomers are not an obligate intermediate for fibril formation and that oligomers and fibrils represent separate and distinct aggregation pathways. The results further indicate that screening for aggregation inhibitors using fibril-specific assays, like thioflavin T (ThT) fluorescence, will not necessarily identify inhibitors of oligomerization.

Experimental Procedures

Materials. Synthetic Aβ42 was prepared as previously described (57). 200 mesh formvar/carbon-coated nickel grids were obtained from Electron Microscopy Sciences (Ft. Washington, PA) and 4 - 20% Tris-HCl gels were from Bio-Rad (Hercules, CA). 96-well clear, flat bottom microplates were from Nalge Nunc International (Rochester, NY) and 3,3’,5,5’-tetramethylbenzidine was from KPL (Gaitherburg, MD). A11 anti-oligomer antibody is available from InVitrogen (Carlsbad, CA). Horseradish peroxidase-conjugated anti-rabbit IgG (AR) was purchased from Promega (Madison, WI), 6E10 and 4G8 antibodies from Signet (Signet, Dedham, MA), and ECL chemiluminescence kit from

2

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Amersham-Pharmacia (Piscataway, NJ). 0.2 µm nitrocellulose membranes were form Schleicher & Schuell (Germany). Small-molecule compounds and all other reagents were from Sigma (St. Louis, MO) or Calbiochem (San Diego, California). The following small-molecules or analogs of small-molecules reported previously to bind amyloid or to modulate protein aggregation and/or toxicity or screened for such activities were tested here separately for their ability to inhibit Aβ42 oligomer and fiber formation: apigenin (62), azure C (62), basic blue 41 (62), (trans,trans)-1-bromo-2,5-bis-(3-hydroxycarbonyl-4-hydroxy)styrylbenzene (BSB) (63), Chicago sky blue 6B (25), Congo red (10,25,64-66), β-cyclodextrin (67,68), curcumin (11,69), daunomycin hydrochloride (15,20), dimethly yellow (62), direct red 80 (25), 2,2’-dihydroxybenzophenone (62), hexadecyltrimethylammonium bromide (C16) (9), hemin chloride (62,70), hematin (62,70), indomethacin (71), juglone (72), lacmoid (62), meclocycline sulfosalicylate (72), melatonin (18), myricetin (62), 1,2-naphthoquinone (16), nordihydroguaiaretic acid (11), R(–)-norapomorphine hydrobromide (19), orange G (25), o-vanillin (2-hydroxy-3-methoxybenzaldehyde) (24), pherphenazine (62), phthalocyanine (62), rifamycin SV (16,62,73), phenol red (74), rolitetracycline (15,20), quinacrine mustard dihydrochloride (62), thioflavin S (ThS) (25,66), ThT (75), and trimethyl(tetradecyl)ammonium bromide (C17) (9). In addition, diallyltartar, eosin Y, fenofibrate, neocuproine, nystalin, octadecylsulfate, and rhodamine B have also been tested. Inhibition of Aβ42 oligomerization. Aβ42 stock solutions (2 mM) were obtained by dissolving the lyophilized peptide in 100 mM NaOH followed by water bath sonication for 30s. The oligomerization reaction was initiated by diluting the stock solution in phosphate buffered saline (PBS), pH=7.4, 0.02 % sodium azide (45 µM final Aβ42 concentration, final pH=7.4). Because Aβ42 forms oligomers that are free of fibers at early time points, we refer to these conditions as “oligomerization conditions”. The reactions were incubated at 25° C for up to 15 days in the absence or presence of small-molecule compounds (0.01 - 3000 µM) dissolved in DMSO. Incubation at 25° was chosen because the rate of fibril nucleation and elongation are favored by incubation at 37°

(76,77). Control reactions were carried out in the presence of 1% DMSO vehicle. The oligomerization reactions were assayed by dot blot, ELISA, western blot, and transmission electron microscopy (TEM), as described below. In addition, Aβ42 oligomerization was conducted in H2O, pH 2-3, in the absence or presence of select compounds (300 µM) using HFIP-based stock solutions, according to a previously published protocol (58) The effect of select compounds were assessed by the dot blot assay. Inhibition of Aβ42 fibrillization. Aβ42 stock solutions (2 mM) were obtained by dissolving the lyophilized peptide in 100 mM NaOH followed by water bath sonication for 30s. Fibrillization of Aβ42 (45 µM final concentration) was initiated by diluting the stock solution in 10 mM HEPES, 100 mM NaCl, 0.02 % sodium azide, pH=7.4. Because Aβ42 forms fibers that are free of oligomers, we refer to these conditions as “fibrillization conditions”. The reactions were stirred at room temperature for up to 6 days in the presence or absence of small-molecule compounds (30 and 300 µM) dissolved in DMSO. Control reactions were carried out in the presence of 1% DMSO vehicle. The reactions were assayed by light scattering, ThT fluorescence assay, and transmission electron microscopy, as described below. Inhibition of fibrillization was also monitored under oligomerization conditions described above. Dot blot assay. The dot blot assay was performed as previously described (58) to detect Aβ42

oligomer formation. Briefly, 2 µl aliquots of each oligomerization reaction were applied onto nitrocellulose membranes. The membranes were blocked (1 h at room temperature or overnight at 4 °C) with 10% non-fat milk in TRIS-buffered saline containing 0.01% Tween 20 (TBS-T). 0.02% sodium azide was added for overnight blocking. The membranes were then washed and incubated with affinity-purified anti-oligomer antibody (A11, 0.8 ug/ml) (58) diluted in TBS-T containing 5 % milk for 1h at room temperature. The membranes were washed again and incubated with horseradish peroxidase-conjugated anti-rabbit IgG (AR) diluted 1:5,000 in TBS-T containing 5 % milk for 1h at room temperature. Then, the blots were washed and developed with the ECL chemiluminescence kit. Separate membranes served as controls for the presence of Aβ42 and

3

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

were treated as described above, except that the A11 and AR antibodies were replaced with 6E10/4G8 and horseradish peroxidase-conjugated anti-mouse IgG, respectively. These antibodies were diluted 1:10,000 in TBS-T containing 3% BSA. All washes were performed with TBS-T, three times for 5 min, except for the last wash before detection, which was for 10 min. ELISA assay. ELISA assay was performed as previously described (58). Briefly, Aβ42 was subjected to oligomerization in the absence and presence of select small-molecules (0.01 - 3000 µM) identified as oligomerization inhibitors by the dot blot assay, as described above. Control reactions were supplemented with 1% DMSO vehicle. Aliquots of each oligomerization reaction were applied to 96-well clear, flat bottom plates containing 100 µl coating buffer (0.1 M sodium bicarbonate, pH=9.6). The Aβ42 concentration used was within the linear range of the assay. The plates were incubated 2 h at 37 °C, washed, blocked with 10 % BSA/TBS-T for 2h at 37 °C, and washed again. Then, 100 µl of A11 antibody (1:1,500 dilution in 3% BSA/TBS-T) was added to each well and the plates were incubated for 1 h at 37 °C. After washing, 100 µl of AR antibody (1:5000 dilution in 3% BSA/TBS-T) was added to each well and the plates were incubated for 1 h at 37 °C. The plates were then washed and developed with 3,3’,5,5’-tetramethylbenzidine (TMB). The reactions were stopped with 100 µl of 1 M HCl and assayed by absorbance at 450 nm. PBS was used for all washes, which were performed three times. Each reaction was performed in triplicate and the data points were fit to dose-response curves as described before (78), using the Sigmaplot software (Systat Software Inc., Point Richmond, CA). The IC50, defined as the concentration of small-molecule required to attain half-maximal absorbance, was obtained from the fit. Western Blot. 10 µl aliquots of each oligomerization reaction, small-molecule treated or control containing the DMSO vehicle, were mixed in a 1:1 ratio (vol/vol) with 2x SDS sample buffer, boiled for 5 min, and loaded onto 4 - 20% Tris-HCl gels. The gels were transferred to nitrocellulose membranes, which were then subjected to the dot blot assay with the A11 antibody.

ThT fluorescence assay. To investigate the time course of Aβ42 fibril formation under our fibrillization conditions and in the absence of small-molecules, 10 µl aliquots were removed at various time points during the fibrillization reaction and mixed with 120 µl ThT (3 µM, dissolved in the fibrillization buffer). ThT fluorescence was measured at λex = 442 nm and λex = 482 nm until a plateau was reached, in a Gemini XPS plate reader (Molecular Devices, Sunnyvale CA). To investigate the effect of small-molecules on Aβ42 fibril formation and on ThT fluorescence, similar measurements were performed except that all readings were taken after 4 days of incubation and ThT emission was recorded at both 482 nm (in the presence of Aβ42) and 482 and 520 nm (in the absence of Aβ42). Turbidity assay. Aβ42 was incubated under both oligomerization and fibrillization conditions, as described above, in the presence and absence of compounds for 7 and 4 days respectively. Each reaction was then assayed by turbidity at 400 nm wavelength, to estimate the amount of fibrillar material. The reactions were then centrifuged at 14,000 rpm for 30 minutes. Part of each supernatant was removed and assayed by the same method at 400 nm wavelength. The resulting values were then used to correct each of the turbidity readings corresponding to the assembly reactions. Electron microscopy. 1 µl aliquots of the aggregation reactions were adsorbed onto 200 mesh formvar/carbon-coated nickel grids until dry. The grids were then washed with water, stained with 2% uranyl acetate, and washed again. The grids were allowed to dry between all steps and were viewed in a Phillips CM 12 microscope operated at 65 kV.

Results

Kinetics of Aβ42 oligomerization. We have recently described an oligomer specific antibody (A11) that recognizes Aβ oligomers and protofibrils and does not react with monomeric Aβ, fibrillar Aβ, or the amyloid precursor protein (58). We used this antibody to specifically monitor oligomer formation independently from fibril formation, which was measured by light scattering (17,79) and ThT fluorescence (80). Lyophilized Aβ42 was dissolved in 100 mM NaOH

4

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

and the oligomerization reactions were initiated by diluting the resulting stock solution in 1x PBS, pH=7.4, as described in Experimental Procedures section. Aliquots of each reaction were removed at various time points and tested in parallel by immunobloting with A11 antibody and transmission electron microscopy. At time point zero, these aliquots reacted only very weakly with the A11 antibody, had strong reactivity with both 6E10 and 4G8 antibodies (Fig. 1A), and showed little aggregation by TEM (Fig. 1B), consistent with the majority of Aβ42 being in non-aggregated form. Significant A11 immunoreactivity was observed as early as 1 day after the initiation of aggregation (Fig. 1A) and correlated with the appearance of small oligomers by TEM (Fig. 1C). Strong A11 immunoreactivity was observed after 4 days of incubation, indicative of the formation of large amounts of A11-positive oligomers (Fig. 1A). TEM analysis confirmed the presence of oligomers at this time point (Fig. 1D). Longer incubation (i.e., ≥ 6 days) resulted in the appearance of Aβ42 fibrils, which appeared to coexist with A11-positive oligomers for extended periods of time (Fig. 1A, E). The presence of 1% DMSO, which is used as vehicle for the small-molecule compounds, did not significantly alter the kinetics of aggregation (Fig. 1A).

Coincident with the increase in A11 immunoreactivity, diminishing 6E10 and 4G8 immunoreactivity was observed (Fig. 1A). The explanation for the loss of 6E10 and 4G8 immunoreactivity is not entirely clear, but it may be related to decreased exposure of these epitopes as the peptide aggregates. This is consistent with the observation that the immunostaining by these antibodies is greatly enhanced by formic acid treatment or thermal denaturation (81). We have previously reported that oligomers formed by dilution of Aβ stock solutions in HFIP into water and incubation at pH=3 react with both A11 and 6E10 antibodies (58). The absence of 6E10 immunoreactivity in aged oligomers prepared from NaOH stocks diluted in PBS at pH=7.4 indicates that A11 positive oligomers are polymorphic at the 6E10 epitope. Taken together, these data indicate that A11-immunoreactive Aβ42 oligomers formed under the conditions used here are the dominant species early in the aggregation reaction (< 6 days). Both oligomers and fibers appear to populate the later stages of aggregation.

These data suggest that experiments aimed at investigating oligomers formed under these conditions should be conducted at early time points, before the appearance of fibrils. In this study, we conducted all experiments aimed at testing oligomerization inhibitors within the time frame where Aβ42 is mostly oligomeric. Identification of Aβ42 oligomerization inhibitors. The ability of small-molecules to inhibit Aβ42 oligomer formation was examined using oligomer-specific antibody immunoreactivity in a dot blot assay. Control samples were treated with 1% DMSO vehicle and the small-molecule additives were used at 30 and 300 µM, when present. Aggregation was conducted at room temperature, without stirring for up to 5 days under oligomerization conditions. The control reactions showed strong A11 immunoreactivity, indicative of oligomer formation (Fig. 2). Reactions incubated with some of the small-molecules showed decreased A11 immunoreactivity (Fig. 2), suggesting that these compounds may block Aβ42 oligomer formation. The test compounds did not interfere with the binding of Aβ42 to the nitrocellulose membrane, as determined by 6E10 immunoreactivity at time point zero of aggregation (data not shown). The active compounds are numbered and listed in Table I. None of the molecules exhibited A11 reactivity in the absence of Aβ42 with the exception of hemin and hematin (data not shown). Therefore the apparent enhancement of A11 immmunoreactivity of Aβ42 in the presence of hemin (Fig. 2; A3, A4) and hematin (Fig. 2; H7, H8) constitutes an artifact that can be attributed to this phenomenon and this assay cannot determine whether these compounds interfere with oligomer formation. These data indicate that small-molecules are capable of inhibiting Aβ42 oligomerization and some of the molecules that have previously been reported as inhibitors of fibril formation inhibit oligomerization as well. Potency of Aβ42 oligomerization inhibitors. To determine the inhibitor potency, the concentration dependence was determined for each of the compounds with an ELISA assay using the A11 antibody. Aβ42 was subjected to aggregation under oligomerization conditions, in the presence of 1% DMSO (control reaction) or small-molecules inhibitors of oligomerization at concentrations ranging from 0.01-3000 µM. The IC50s for

5

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

inhibition of Aβ42 oligomerization were determined from dose-response curves similar to those presented in supplementary Fig. 1 and ranged from 0.09 to 2787 µΜ (Table I). Eighteen small-molecules inhibited oligomerization at substoichiometric concentrations relative to Aβ42 monomer and four compounds were active in the nanomolar concentration range. These include azure C, basic blue 41, meclocycline sulfosalicylate, and R(–)-norapomorphine hydrobromide (Supplementary Fig. 1). Hemin and hematin were among the substoichiometric inhibitors (Table I) suggesting that the ELISA assay eliminates the interference of these substances with the antibody recognition identified in the dot blot assay. The potency of the remaining molecules varied from low micromolar to low millimolar range (Table I). These data confirm the inhibitory activity of the molecules identified by the dot blot assay and indicate that inhibition of Aβ42 oligomerization can be achieved at concentrations in the low nanomolar range, suggesting that inhibiting oligomer formation may be therapeutically feasible.

HFIP-based Aβ42 solutions were subjected to oligomerization in H2O, as previously described (58) in the absence and presence of 1 % DMSO vehicle. In the absence of DMSO, Aβ42 formed A11 immunoreactive oligomers within 1 day of initiation of aggregation, consistent with previous observations (58) (data not shown). In the presence of DMSO, oligomers were not detected even after extended incubation times (data not shown). Reactions containing 1 % DMSO are the true control for this experiment because all the compounds are delivered in DMSO. Therefore, the ability of the inhibitors of Aβ42 oligomerization (as determined under oligomerization conditions) to inhibit oligomer formation obtained from HFIP-based Aβ42 solutions could not be examined. Effect of inhibitors on Aβ42 oligomers. To further characterize the aggregation state of the products that accumulate in the presence of the inhibitors, aliquots of Aβ42 oligomerization reactions conducted under oligomerization conditions in the presence and absence of inhibitors were assayed by western blot with the anti-oligomer specific antibody A11 (Fig. 3A). Small-molecules were used at 300 µM because this concentration is greater than the IC50 of most of the oligomerization inhibitors. The compound-treated

reactions are presented in the order of the inhibitory potency, starting with the most active as listed in Table I. Control Aβ42 oligomers formed in the absence of inhibitors were reactive with the A11 antibody in the molecular weight range from ~ 18–250 kDa. This is consistent with previous observations that oligomers in this size range react with A11 as determined by size exclusion chromatography (58,59,82). Although most of the immunoreactivity appears as a smear above 75 kDa, discrete bands of ~ 18 and ~ 36 kDa corresponding to Aβ42 tetramers and octamers (Fig 3A, lane 0) were also observed. Samples treated with most of the compounds identified as substoichiometric inhibitors of Aβ42 oligomerization (compounds 1-18) did not exhibit or exhibited only weak A11 immunoreactivity (Fig 3A, lanes 1-18). These data confirm that these molecules are strong inhibitors of Aβ42 oligomerization. Chicago sky blue 6B (Fig. 3A, lane 7), however, constitutes an exception because it did not prevent the formation of A11 positive Aβ42 oligomers. Chicago sky blue 6B may represent a false positive by interfering with A11 binding in the primary screening operation, while compound dissociation from the oligomers in the western blot assay eliminates this artifact. Some of the suprastoichiometric inhibitors, (molecules with 45 µM < IC50s < 244 µM, (Fig 3A, lanes 19-21)), also diminished A11 immunoreactivity on western blots, consistent with bona fide inhibitory activity. Of these, only myricetin and ThT (Fig. 3A, lanes 20 and 21) prevented the formation of the 18 and 36 kDa discrete bands, while juglone (Fig. 3A, lane 19) appears to selectively inhibit the higher MW 75-250 kDa oligomers and not the discrete bands. The stabilization of discrete A11-immunoreactive bands corresponding to tetramers and octamers observed here for juglone may be useful for further structural characterization of these species because of their limited stability and larger size distribution in the absence of inhibitors. Among the remaining suprastoichiometric inhibitors (IC50s > 122 µM, Fig. 3A, lanes 22-29), 2,2’-dihydroxybenzophenone, rhodamine B, phenol red, indomethacin, and eosin y (Fig. 3A, lanes 22, 23, 25, 26, and 29, respectively) showed only weak inhibition of A11 immunoreactivity. This was expected because the concentration of inhibitor used in this assay was close to or up to three times lower than their IC50s which may be to

6

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

low to observe complete inhibition. Pherphenazine (Fig. 3A, lane 28) appeared to rearrange the oligomeric population to stabilize oligomers in the molecular weight range from 37-150 kDa. However, curcumin and quinacrine mustard dihydrochloride (Fig. 3A, lanes 24 and 27, respectively) completely inhibited A11 immunoreactivity, suggesting that their IC50s for inhibition of Aβ42 oligomerization may be lower than determined by ELISA.

We also analyzed the same western blots with 6E10 to visualize the effects of compounds on Aβ42 species that are not detected with A11 (Fig. 3B). In general, we observed the same reciprocal relationship between A11 staining and 6E10 staining for the majority of samples as we had observed previously in the dot blot assay (Fig. 1A). This indicates that the conformation of the 6E10 and A11 epitopes are maintained even in the presence of SDS. The 6E10 staining also revealed differences in the sizes of the products that accumulate in the presence of the inhibitors. Azure C, basic blue 41, meclocycline, and o-vanillin (Fig. 3B, lanes 1-3 and 9) promoted the accumulation of Aβ42 aggregates ranging in size from approximately 50 kDa to material that sticks at the top of the gel. This size range overlaps that of the high molecular weight Aβ42 oligomers stained by A11 in the absence of inhibitors, indicating that conformationally distinct, SDS-resistant aggregates of approximately the same size can be detected by western blots. Since many of the compounds that cause the accumulation of SDS-resistant, A11 negative, 6E10 positive high molecular weight aggregates appear to actually promote fibril formation (see below), the simplest interpretation is that this material may represent fibrils and small fragments of fibrils that are partially sensitive to dissociation by SDS. Indeed, western blots of fibril preparations show a similar size distribution of SDS-resistant 6E10 immunoreactivity that is not recognized by A11 (supplemental Fig. 2). Other compounds (Fig. 3B, lanes 4, 6, 12, 17, 18, 20, 21, and 24) promoted the accumulation of both the high molecular weight aggregates and low molecular weight monomer, dimer and trimer that are A11 negative. A third class of compounds (Fig. 3B, lanes 5, 11, 14, 15, 16, and 19) caused the accumulation of low molecular weight Aβ and/or material that sticks at the top of the gel, but very low amounts of

intermediate-sized aggregates. A few compounds (Fig. 3, lanes 8, 13, and 27) appeared to inhibit the accumulation of both A11 and 6E10 positive bands. The interpretation of this later group is not yet clear. They may promote the formation of products that do not display either A11 or 6E10 epitope or they may favor the precipitation of Aβ42 into a form that is not solubilized by the sample buffer.

These data confirm the inhibitory activity of the molecules identified as Aβ42 oligomerization inhibitors by the dot blot and ELISA assays and suggests that some of the inhibitors may actually promote fibril formation. In addition, the potency qualitatively visualized by western blots correlates well with the potency quantitatively determined by ELISA. Inhibition of fibril formation. Immunoreactivity with the A11 anti-oligomer antibody indicates that only a subset of the small-molecules tested constitute inhibitors of Aβ42 oligomer formation. This suggests that some of the small-molecules tested may be specific inhibitors of fibrillization, since many of them were originally reported as inhibitors of fibril formation. To test this hypothesis directly, Aβ42 was incubated under fibrillization conditions, as described in Experimental Procedures, in the presence 1% DMSO (control reaction) or small-molecules (30 and 300 µM). Fibril formation was assayed by ThT fluorescence, light scattering, and TEM analysis. Under these conditions, Aβ42 supplemented with 1% DMSO readily assembled into fibers through a nucleation-dependent mechanism (Supplementary Fig. 3A). The lag time of assembly was about ~ 13 h and the ThT fluorescence plateau was reached in approximately 4 days. TEM confirmed that aggregation was minimal at time point zero and that Aβ42 assembled into fibrils with classic morphology after four days of incubation (Supplementary Fig. 3B, 3C). Because Aβ42 fibers formed rapidly under these conditions (e.g., the reaction reached plateau levels within 4 days) and because they do not show A11-immunoreactive oligomer contamination by dot blot assay (data not shown), these assembly conditions allow a reasonably rapid assay of the effects of compounds exclusively on fiber formation.

Although ThT fluorescence is useful for assaying aggregation in control reactions and

7

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

establishing working conditions, its utility for quantifying fibrillization in samples containing small organic molecules that adsorb significantly at 442 nm (the excitation wavelength of ThT) is limited by absorption and fluorescence artifacts (Table II). Monitoring ThT fluorescence in the absence of Aβ revealed that most compounds strongly interfered with ThT fluorescence when emission was monitored at both 520 nm (emission maximum of ThT) and 482 nm (emission maximum of ThT in the presence of fibrillar material). Some compounds interfered with ThT absorption causing a decrease of ThT emission, while others increased fluorescence emission due to their intrinsic fluorescent properties (Table II). The decrease of ThT fluorescence observed in reactions containing Aβ42 and small-molecules meclocycline sulfosalicylate, Chicago sky blue 6B, hemin, o-vanillin, C16-30µΜ, hematin, neocuproine, lacmoid, rifamycin SV, rhodamine B, eosin Y, orange G, diallyltartar, direct red, and apigenin (Table II) paralleled the decrease of Aβ42 turbidity (Fig. 4B) and the decrease of fibrillar content by EM (Supplementary Fig. 4 and data not shown), indicating that these compounds inhibit fibrillization. However, the reliability of ThT as an indictor of fibril inhibition in these cases may be questionable, since most of these molecules significantly altered ThT fluorescence in the absence of Aβ (Table II).

Some of the compounds tested, melatonin, β-cyclodextrin, phthalocyanine, fenofibrate, and dimethyl yellow did not significantly interfere with ThT emission at 482 nm. Although these molecules showed strong inhibition of ThT fluorescence in reactions containing Aβ42, none of them reduced Aβ42 turbidity (data not shown). One simple explanation is that these small-molecules may induce the aggregation of Aβ42 in a non-fibrillar form, however abundant fibrillar material was observed by TEM in reactions containing all these molecules (data not shown). This suggests that these small-molecules are not inhibitors of fibril formation, but rather may inhibit the binding of ThT to the fibrils. Low concentrations of hexadecyltrimethylammonium bromide (C16) had no effect on ThT emission at 482 in the absence of Aβ42 and appeared to inhibit ThT-positive fiber formation (Table II), consistent with results obtained by turbidity assay (Fig. 4B). However, higher concentrations of this compound as well as

trimethyl(tetradecyl)ammonium bromide (C17) significantly reduced ThT emission in the absence of Aβ42 (Table II), thus ThT assay cannot reliably determine the effect of these compounds on fiber formation.

Therefore, for the purpose of clarifying the effect of small-molecules on Aβ42 fibrillization, aggregation was also qualitatively assayed by turbidity at 400 nm wavelength, as previously described (17,79). Because small-molecules can undergo significant spectral changes in the presence of proteins (78,83), correcting the turbidity measurements by subtracting buffer blanks containing compounds and lacking the protein is not always adequate. Artifacts related to this phenomenon were avoided by using the supernatant of the fibrillization reactions, obtained after centrifugation, to correct the turbidity readings, as described in Experimental Procedures. Four days after initiation of aggregation and in the absence of compounds, Aβ42 had significant turbidity (Fig. 4A, 4B), and assembled into fibrils (Supplementary Fig. 3C). In the presence of oligomerization inhibitors, the turbidity of most reactions either increased (Fig. 4A) or decreased (Fig. 4B) compared to the fibrillar control. The fact that some compounds increase turbidity compared to controls suggests that they may actually promote fibrillization. The effects on Aβ42 fibrillization were consistent at both compound concentrations examined and are shown here when compounds were tested at the higher concentration, 300 µM (Fig. 4). C16 and C17 however, constitute exceptions and their effect on aggregation is shown at both concentrations. More than half of the small-molecules tested did not inhibit Aβ aggregation into fibrils as shown by constant or increased levels of turbidity in reactions containing these molecules relative to control (Fig. 4A and data not shown).

The compounds that inhibited Aβ42 oligomerization but either promoted fibrillization or did not inhibit it include: azure C, basic blue 41, R(–)-norapomorphine hydrobromide, Congo red, rolitetracycline, daunomycin, C16 (300 µM), 1,2-naphthoquinone, nordihydroguaiaretic acid, C17 (300 µM), juglone, myricetin, ThT, curcumin, indomethacin, quinacrine mustard dihydrochloride, and pherphenazine. All of these oligomerization inhibitors, except for Congo red,

8

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

rolitetracycline, myricetin, and indomethacin, appear to actually promote fibrillization (Fig. 4A). These specific inhibitors of oligomerization are referred to as Class I compounds (Table III). The remaining Aβ42 oligomerization inhibitors, meclocycline sulfosalicylate, hemin, o-vanillin, C16 (30 µM), hematin, C17 (30 µM), neocuproine, lacmoid, rifamycin SV, 2,2’-dihydroxybenzophenone, rhodamine B, phenol red, and eosin Y partially inhibited Aβ42 fibril formation (Fig. 4B). The small-molecules that inhibit both Aβ42 oligomerization and fibrillization are referred to as Class II compounds (Table III). The small-molecules apigenin, Chicago sky blue 2B, diallyltartar, direct red, and orange G did not inhibit oligomerization but partly inhibited Aβ42 fibrillization (Fig. 4B). These compounds are referred to as Class III compounds (Table III). Melatonin, β-cyclodextrin, octadecylsulfate, BSB, ThS, phthalocyanine, fenofibrate, dimethyl yellow, and nystalin did not inhibit either oligomerization or fibrillization (data not shown) when assayed by the methods described in Experimental Procedures.

To qualitatively confirm the effect of the compounds on Aβ42 fibril formation, aliquots of the fibrillization reactions were removed and assayed by TEM. In the control reaction, Aβ42 had minimal aggregation at time zero (Supplementary Fig. 4A) and assembled into fibrils after four days of incubation (Supplementary Fig. 4B). Abundant fibrils with similar morphology formed in the presence of all Class I compounds, confirming that these compounds did not inhibit Aβ42 fibrillization. The lack of inhibition of Aβ42 fibrillization by compounds in this class is illustrated here for basic blue 41, R(-)-norapomorphine hydrobromide, rolitetracycline, and juglone (Supplementary Fig. 4, C-F). Although the lack of well resolved, individual fibrils precludes the quantification of fibrillization by TEM (84), inhibition of fibrillization could be qualitatively detected in the presence of all Class II and III compounds. TEM images of assembly reactions in the presence of C16 (30 µM), C17 (30 µM), neocuproine, lacmoid, rhodamine B, and phenol red are shown in Supplementary Fig. 4 (G-L) as examples.

Since fibrils form after extended incubation times under oligomerization conditions, we also tested the effects of the small-molecule inhibitors under these conditions where both oligomers and

fibrils form under the same conditions. TEM analysis showed that the Class I oligomer specific inhibitory compounds azure C, basic blue 41, R(–)-norapomorphine hydrobromide, daunomycin, C16, 1,2’-naphtoquinone, nordihydroguaiaretic acid, C17, juglone, ThT, curcumin, quinacrine mustard dihydrochloride, and pherphenazine promote Aβ42 fiber formation at after 3 days of incubation when Aβ42 is still present in oligomeric form in the control reaction (Supplementary Fig. 5A-N). These data show that these compounds promote fiber formation and are consistent with the 6E10 western blot data (Fig. 3B) and data obtained under fibrillization conditions (Fig. 4A). The remaining Class I compounds Congo red (Supplementary Fig. 5O), myricetin, and indomethacin did not induce fiber formation at this time point. This is also consistent with the lack of activity of these compounds on Aβ42 fiber formation observed under fibrillization conditions (Fig. 4A).

To test whether the compounds inhibit fibril formation under the same conditions, aliquots of each of these reactions were also removed 7 days after initiation of aggregation and analyzed by turbidity, as explained in the Experimental Procedures. At this time point control Aβ42 solutions in the absence of test compounds exist as a mixture of oligomers and fibers under oligomerization conditions (Fig. 1). Since pure oligomer solutions have a low turbidity, the signal corresponding to fibers is expected to be the main contributor to the turbidity measurements. Consistent with observations obtained at early time points during aggregation (3 days, Supplementary Fig. 4), Class I compounds did not inhibit Aβ42 fibrillization upon prolonged incubation (Supplementary Fig. 6A). In contrast, Class II and Class III compounds inhibited fiber formation (Supplementary Fig. 6B). Data obtained after 7 days of incubation under oligomerization conditions (Supplementary Fig. 6) is also consistent with the effect of compounds on Aβ42 fiber formation assessed under fibrillization conditions (Fig. 4). Taken together, these data indicate that the effects of compounds on fiber formation are consistent regardless of the experimental conditions.

9

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Discussion Aβ42 assembly pathways. The first major set of findings of this study is related to mechanisms underlying Aβ assembly. We characterized the mechanism of action of a set of small-molecules that have been reported as inhibitors of amyloid aggregation or amyloid toxicity in order to determine which steps in the Aβ42 assembly pathways they inhibit. We screened for oligomer inhibitory activity using the anti-oligomer specific antibody A11 (58) and we used the widely employed ThT and light scattering assays for measuring fibril formation (80,85). Since all assays are sensitive to potential artifacts, we also characterized the effects of the inhibitors by western blotting and electron microscopy. We found that the inhibitory compounds segregate into three classes based on their activity. Class I compounds inhibit oligomerization without inhibiting fibrillization, Class II compounds inhibit both, and Class III compounds only inhibit fibrillization. The Class I compounds can be further subdivided on the basis that a subset of compounds actually promote fibril formation, while the remainder have no effect on fibrillization. The finding that Class I compounds block oligomerization without inhibiting fibrillization indicates that Aβ42 oligomer formation is not an obligatory step on the pathway leading to fiber formation. Rather, oligomer formation constitutes an alternate aggregation pathway (Fig. 5). This is consistent with the observations that oligomer formation is considerably more sensitive to urea treatment than fibril formation and that fibril formation can proceed efficiently under concentrations of urea where oligomers are undetectable by A11 (86). Previous results from ultrastructural analysis have been interpreted as indicating that spherical Aβ oligomers coalesce to form “protofibrils”, which then form mature fibrils (36,87), implying that oligomers are intermediates on the fibril assembly pathway. The finding that oligomers are not obligate intermediates does not necessarily imply that oligomers do not ultimately form fibrils, as there may be more than one pathway leading to Aβ fibril formation. Alternatively, oligomers may represent an “off pathway” assembly state that does not directly convert to fibrils, but buffers the concentration of monomer that ultimately

assembles into fibrils. Further work will be necessary to unambiguously clarify this issue.

The existence of oligomeric and fibrillar assembly states and the question of their relationship in the aggregation pathway is not restricted to Aβ, but is a characteristic of the aggregation of many other disease and non-disease-related proteins. Increasing evidence indicates that the propensity to form amyloid fibrils and oligomers is a generic property of misfolded proteins (88,89) and that these aggregation states display common structural properties (90,91) and similar assembly kinetics (92). To the extent that these properties are general, our finding that oligomers are not an obligate intermediate for Aβ fibril formation suggests that oligomer formation may represent an alternative pathway for all amyloids.

An unexpected finding is that Aβ oligomers formed under different conditions display polymorphism in terms of differential epitope accessibility. This type of polymorphism has recently been reported for prion oligomers (93). We found that oligomers prepared by dilution into PBS from NaOH stock solutions are recognized by A11, but display greatly reduced 6E10 and 4G8 immunoreactivity, indicating that these epitopes are either conformationally altered or inaccessible in NaOH oligomers. In contrast, oligomers prepared by dilution of HFIP stock solutions into water at pH=3 are equally immunoreactive with A11, 6E10, and 4G8 (58). Oligomers prepared by the two methods have the same morphology and size distribution by electron microscopy. These results indicate that there are two distinct conformations of A11 positive Aβ oligomers and suggest that sandwich ELISA methods that use 6E10 or 4G8 may not detect all types of Aβ oligomers. Pharmacological implications. Recent advances facilitate the specific screening for small-molecules that inhibit Aβ42 oligomerization and fibrillization. The development of the anti-oligomer antibody A11 (58) allows the specific identification of inhibitors of this important class of oligomers, while careful selection of well-established fibrillization assays allows accurate screening for inhibitors of fibrillization. ThT fluorescence has been widely used to measure amyloid fibrillization (80,85). While this assay is useful for quantifying fibrillar material in many

10

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

circumstances (24), its use for drug screening purposes may be limited by small-molecule interference with fluorophore adsorption (94) and possible competitive binding to amyloid fibers (78).

We identified 29 small-molecules capable of inhibiting oligomer formation. Their potency varies greatly, from compounds that are active at substiochiometric concentrations relative to Aβ42 monomer to compounds that require an approximately 70 fold excess to inhibit oligomer formation by 50%. The substoichiometric inhibitors include azure C, basic blue 41, meclocycline, and R(–)-norapomorphine hydrobromide that are active at nanomolar concentrations. So far, only few compounds have been shown to inhibit Aβ oligomerization. These are curcumin (69), β-cyclodextrin derivatives (68), Congo red (64), Ginkgo biloba extracts (95,96), and benzyl-containing compounds, including o-vanillin (24). While we confirmed this activity for Congo red, o-vanillin, and curcumin, only the former two were active at low micromolar concentrations. These data suggest that oligomers are amenable to drug treatment by a variety of unrelated compounds and the low concentrations required for inhibition predict that such an approach is therapeutically feasible. We were unable to assess the effect of these inhibitors on oligomerization of Aβ42 in distilled H2O at pH 2-3, when the peptide is delivered from HFIP-based stock solutions (58), so it remains unclear if the inhibitors are effective under these conditions.

In addition, we found that 18 compounds partially inhibited Aβ42 fiber formation. Of these, hemin, hematin, lacmoid, rifamycin SV, 2,2,-dihydroxybenzophenone, and apigenin, have been previously shown to inhibit Aβ and tau fibrillization (16,62,70,73). Meclocycline has been reported as an inhibitor of huntingtin aggregation (72), and phenol red was identified as an amylin fibrillization inhibitor (74), suggesting that these compounds may constitute general inhibitors of fiber formation. o-vanillin inhibited Aβ fibrillization (24) and Chicago sky blue and direct red diminished cytotoxicity, an activity correlated with inhibition of Aβ fibrillization (25). Although the concentration required for inhibitory activity differs, the biphasic modulation of Aβ fibrillization observed here with C16 and C17 was previously reported (9). Besides identifying new

inhibitors of Aβ fibrillization, our data confirm such inhibitory activity for the above compounds. However, we could not confirm the previously reported inhibitory activity on fiber formation for azure C, basic blue 41 (62), R(–)-norapomorphine hydrobromide (19), rolitetracycline, daunomycin (15), nordihydroguaiaretic acid (11), myricetin (62), curcumin (11,69), quinacrine mustard dihydrochloride (62), pherphenazine (62), melatonin (18), dimethy yellow, and phtalocyanine (62). One explanation for the discrepancy between the results presented here and previous observations may raise form choice of assays. Because of reasons explained above, we did not rely solely on the ThT assay used previously to test many of these compounds. Small-molecules might have concentration-dependent multiphasic behavior on modulating protein aggregation (herein and (9,94,97-99)), therefore this discrepancy could also be attributed to the choice of compound concentrations used to test activity. Juglone did not inhibit Aβ fibrillization but appears to inhibit huntingtin aggregation (72), suggesting that this molecule is a specific inhibitor of fibrillization (17,100). The compounds were active inhibitors of fibrillization under both of the experimental conditions used.

Taken together, these data suggest that selective inhibition of either Aβ42 oligomerization or fibrillization is possible, which allows the separate targeting of either species. This is consistent with previous observations that calmidazolium chloride promotes the conversion of Aß monomer into clusters of protofibrils, indicating that selective targeting of specific Aβ species is possible (22). The fact that oligomer and fibril formation can be inhibited independently also indicates that screening for fibril inhibitors will only identify a subset of potential oligomer inhibitors.

Mechanism of inhibition. Most of the Class I compounds, in addition to selectively inhibiting oligomerization, appear to promote fiber formation. We speculate that these molecules stabilize Aβ42 conformations that do not support oligomerization but rather favor fiber formation (Fig. 5, I1). The fact that some of them function at a concentration 100-fold lower than that of Aβ suggests that these inhibitors preferentially interact with a rate limiting intermediate, such as a seed or nucleus, that is present at a low concentration. One

11

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

possible explanation is that these compounds inhibit oligomerization by promoting the formation of fibril seeds, which is consistent with the idea that oligomerization and fibrillization are competing alternative pathways. Since Aβ toxicity is dependent on aggregation state and oligomers and fibers appear to be at the opposite ends of the toxicity range, the discovery of molecules that promote fiber formation at the expense of oligomers may be therapeutically useful.

Class I compounds Congo red, rolitetracycline, myricetin, and indomethacin do not significantly increase the amount of fibrillar material at concentrations tested here, but still favor fibrillization over oligomerization and may act by a similar mechanism. Class II compounds inhibit Aβ42 assembly into both oligomers and fibrils. We therefore speculate that these compounds stabilize conformations (Fig. 5, I2) that do not support aggregation. Benzyl-containing compounds were previously reported to belong to this class (24). Here we confirm such action for one of these compounds, o-vanillin. Because, under conditions tested here, curcumin did not inhibit fibril formation, we suggest that this molecule belongs to Class I compounds instead of Class II as previously reported (69). Class III compounds inhibit fibrillization but do not inhibit oligomer formation. This observation predicts that compounds in this class stabilize a conformation (Fig. 5, I3) that supports oligomerization but does not favor fiber formation. Although numerous small-molecules have been previously reported to inhibit fiber formation, few have been tested for their effect on oligomerization. The exceptions include studies that identified naphthalene

sulfonates and catecholamines as members of this class of compounds (101,102).

The inhibitory effects observed here can be attributed solely to the stabilization of compound-bound Aβ42 conformations (Fig. 5, I1, I2, I3) that either do not support assembly or selectively favor aggregation into either oligomers or fibrils. Binding of these molecules to Aβ42 either depletes the assembly competent Aβ species, therefore inhibiting aggregation, or promotes aggregation through alternate pathways via stabilization of conformationally different intermediates. Inhibitors of aggregation reported previously to trap assembly-competent species into incompetent conformations include tau assembly inhibitor N744 (78,100) and alpha-synuclein fibrillization inhibitors baicalein and dopamine (103,104). Inducers or enhancers of aggregation that act by stabilization of early assembly-competent intermediates, as we propose here for compounds pertaining to Classes I and III, include anionic surfactants, planar aromatic dyes, and urea (66,105,106). Alternatively, it is also possible that inhibition/enhancement of aggregation results from direct binding of the small-molecules to oligomeric/fibrillar Aβ42, similar to Congo red (10,107). A more complicated inhibitory action that includes both these mechanisms is also possible.

The data presented here suggests that selective inhibition of specific aggregated Aβ species is feasible and useful for both unraveling mechanisms underlying protein fibrillization and for therapeutic testing in models of neurodegeneration.

References

1. Selkoe, D. J. (1991) Neuron 6, 487-498 2. Hardy, J., and Selkoe, D. J. (2002) Science 297, 353-356 3. Glabe, C. C. (2005) Subcell Biochem 38, 167-177 4. Bloom, G. S., Ren, K., and Glabe, C. G. (2005) Biochim Biophys Acta 1739, 116-124 5. Howlett, D. R., Jennings, K. H., Lee, D. C., Clark, M. S., Brown, F., Wetzel, R., Wood,

S. J., Camilleri, P., and Roberts, G. W. (1995) Neurodegeneration 4, 23-32 6. Pike, C. J., Walencewicz, A. J., Glabe, C. G., and Cotman, C. W. (1991) Brain Res 563,

311-314 7. Pike, C. J., Burdick, D., Walencewicz, A. J., Glabe, C. G., and Cotman, C. W. (1993) J

Neurosci 13, 1676-1687

12

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

8. Seilheimer, B., Bohrmann, B., Bondolfi, L., Muller, F., Stuber, D., and Dobeli, H. (1997) J Struct Biol 119, 59-71

9. Sabate, R., and Estelrich, J. (2005) Langmuir 21, 6944-6949 10. Kuner, P., Bohrmann, B., Tjernberg, L. O., Naslund, J., Huber, G., Celenk, S., Gruninger-

Leitch, F., Richards, J. G., Jakob-Roetne, R., Kemp, J. A., and Nordstedt, C. (2000) J Biol Chem 275, 1673-1678

11. Ono, K., Hasegawa, K., Naiki, H., and Yamada, M. (2004) J Neurosci Res 75, 742-750 12. Cheng, X., and van Breemen, R. B. (2005) Anal Chem 77, 7012-7015 13. De Felice, F. G., Houzel, J. C., Garcia-Abreu, J., Louzada, P. R., Jr., Afonso, R. C.,

Meirelles, M. N., Lent, R., Neto, V. M., and Ferreira, S. T. (2001) Faseb J 15, 1297-1299 14. Bohrmann, B., Adrian, M., Dubochet, J., Kuner, P., Muller, F., Huber, W., Nordstedt, C.,

and Dobeli, H. (2000) J Struct Biol 130, 232-246 15. Howlett, D. R., George, A. R., Owen, D. E., Ward, R. V., and Markwell, R. E. (1999)

Biochem J 343 Pt 2, 419-423 16. Tomiyama, T., Asano, S., Suwa, Y., Morita, T., Kataoka, K., Mori, H., and Endo, N.

(1994) Biochem Biophys Res Commun 204, 76-83 17. Wood, S. J., MacKenzie, L., Maleeff, B., Hurle, M. R., and Wetzel, R. (1996) J Biol

Chem 271, 4086-4092 18. Pappolla, M., Bozner, P., Soto, C., Shao, H., Robakis, N. K., Zagorski, M., Frangione, B.,

and Ghiso, J. (1998) J Biol Chem 273, 7185-7188 19. Lashuel, H. A., Hartley, D. M., Balakhaneh, D., Aggarwal, A., Teichberg, S., and

Callaway, D. J. (2002) J Biol Chem 277, 42881-42890 20. Howlett, D. R., Perry, A. E., Godfrey, F., Swatton, J. E., Jennings, K. H., Spitzfaden, C.,

Wadsworth, H., Wood, S. J., and Markwell, R. E. (1999) Biochem J 340 ( Pt 1), 283-289 21. Lorenzo, A., and Yankner, B. A. (1994) Proc Natl Acad Sci U S A 91, 12243-12247 22. Williams, A. D., Sega, M., Chen, M., Kheterpal, I., Geva, M., Berthelier, V., Kaleta, D.

T., Cook, K. D., and Wetzel, R. (2005) Proc Natl Acad Sci U S A 102, 7115-7120 23. Ono, K., Hasegawa, K., Naiki, H., and Yamada, M. (2004) Biochim Biophys Acta 1690,

193-202 24. De Felice, F. G., Vieira, M. N., Saraiva, L. M., Figueroa-Villar, J. D., Garcia-Abreu, J.,

Liu, R., Chang, L., Klein, W. L., and Ferreira, S. T. (2004) Faseb J 18, 1366-1372 25. Pollack, S. J., Sadler, II, Hawtin, S. R., Tailor, V. J., and Shearman, M. S. (1995)

Neurosci Lett 197, 211-214 26. Harper, J. D., Wong, S. S., Lieber, C. M., and Lansbury, P. T. (1997) Chem Biol 4, 119-

125 27. Anguiano, M., Nowak, R. J., and Lansbury, P. T., Jr. (2002) Biochemistry 41, 11338-

11343. 28. Harper, J. D., Lieber, C. M., and Lansbury, P. T., Jr. (1997) Chem Biol 4, 951-959 29. Walsh, D. M., Lomakin, A., Benedek, G. B., Condron, M. M., and Teplow, D. B. (1997)

J Biol Chem 272, 22364-22372 30. Blackley, H. K., Sanders, G. H., Davies, M. C., Roberts, C. J., Tendler, S. J., and

Wilkinson, M. J. (2000) J Mol Biol 298, 833-840 31. Nybo, M., Svehag, S. E., and Holm Nielsen, E. (1999) Scand J Immunol 49, 219-223 32. Walsh, D. M., Hartley, D. M., Kusumoto, Y., Fezoui, Y., Condron, M. M., Lomakin, A.,

Benedek, G. B., Selkoe, D. J., and Teplow, D. B. (1999) J Biol Chem 274, 25945-25952 33. Kirkitadze, M. D., Condron, M. M., and Teplow, D. B. (2001) J Mol Biol 312, 1103-1119

13

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

34. Naiki, H., and Nakakuki, K. (1996) Lab Invest 74, 374-383 35. Lomakin, A., Chung, D. S., Benedek, G. B., Kirschner, D. A., and Teplow, D. B. (1996)

Proc Natl Acad Sci U S A 93, 1125-1129 36. Harper, J. D., Wong, S. S., Lieber, C. M., and Lansbury, P. T., Jr. (1999) Biochemistry

38, 8972-8980 37. Gosal, W. S., Clark, A. H., and Ross-Murphy, S. B. (2004) Biomacromolecules 5, 2408-

2419 38. Gosal, W. S., Morten, I. J., Hewitt, E. W., Smith, D. A., Thomson, N. H., and Radford, S.

E. (2005) J Mol Biol 351, 850-864 39. Modler, A. J., Gast, K., Lutsch, G., and Damaschun, G. (2003) J Mol Biol 325, 135-148 40. Hurshman, A. R., White, J. T., Powers, E. T., and Kelly, J. W. (2004) Biochemistry 43,

7365-7381 41. Baskakov, I. V., Legname, G., Baldwin, M. A., Prusiner, S. B., and Cohen, F. E. (2002) J

Biol Chem 277, 21140-21148 42. Kaylor, J., Bodner, N., Edridge, S., Yamin, G., Hong, D. P., and Fink, A. L. (2005) J Mol

Biol 353, 357-372 43. Souillac, P. O., Uversky, V. N., Millett, I. S., Khurana, R., Doniach, S., and Fink, A. L.

(2002) J Biol Chem 277, 12666-12679 44. Souillac, P. O., Uversky, V. N., and Fink, A. L. (2003) Biochemistry 42, 8094-8104 45. Padrick, S. B., and Miranker, A. D. (2002) Biochemistry 41, 4694-4703 46. Grudzielanek, S., Smirnovas, V., and Winter, R. (2006) J Mol Biol. 356, 497-509 47. Morozova-Roche, L. A., Zamotin, V., Malisauskas, M., Ohman, A., Chertkova, R.,

Lavrikova, M. A., Kostanyan, I. A., Dolgikh, D. A., and Kirpichnikov, M. P. (2004) Biochemistry 43, 9610-9619

48. Ahmad, A., Uversky, V. N., Hong, D., and Fink, A. L. (2005) J Biol Chem 280, 42669-42675

49. Green, J. D., Goldsbury, C., Kistler, J., Cooper, G. J., and Aebi, U. (2004) J Biol Chem 279, 12206-12212

50. Poirier, M. A., Li, H., Macosko, J., Cai, S., Amzel, M., and Ross, C. A. (2002) J Biol Chem 277, 41032-41037

51. Gorman, P. M., Yip, C. M., Fraser, P. E., and Chakrabartty, A. (2003) J Mol Biol 325, 743-757

52. Soreghan, B., Kosmoski, J., and Glabe, C. (1994) J Biol Chem 269, 28551-28554 53. Lomakin, A., Teplow, D. B., Kirschner, D. A., and Benedek, G. B. (1997) Proceedings of

the National Academy of Sciences of the United States of America 94, 7942-7947 54. Walsh, D. M., Lomakin, A., Benedek, G. B., Condron, M. M., and Teplow, D. B. (1997)

Journal of Biological Chemistry 272, 22364-22372 55. Lambert, M. P., Barlow, A. K., Chromy, B. A., Edwards, C., Freed, R., Liosatos, M.,

Morgan, T. E., Rozovsky, I., Trommer, B., Viola, K. L., Wals, P., Zhang, C., Finch, C. E., Krafft, G. A., and Klein, W. L. (1998) Proceedings of the National Academy of Sciences of the United States of America 95, 6448-6453

56. Bucciantini, M., Giannoni, E., Chiti, F., Baroni, F., Formigli, L., Zurdo, J., Taddei, N., Ramponi, G., Dobson, C. M., and Stefani, M. (2002) Nature 416, 507-511.

57. Burdick, D., Soreghan, B., Kwon, M., Kosmoski, J., Knauer, M., Henschen, A., Yates, J., Cotman, C., and Glabe, C. (1992) J Biol Chem 267, 546-554

14

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

58. Kayed, R., Head, E., Thompson, J. L., McIntire, T. M., Milton, S. C., Cotman, C. W., and Glabe, C. G. (2003) Science 300, 486-489

59. Lesne, S., Koh, M. T., Kotilinek, L., Kayed, R., Glabe, C. G., Yang, A., Gallagher, M., and Ashe, K. H. (2006) Nature 440, 352-357

60. Hepler, R. W., Grimm, K. M., Nahas, D. D., Breese, R., Dodson, E. C., Acton, P., Keller, P. M., Yeager, M., Wang, H., Shughrue, P., Kinney, G., and Joyce, J. G. (2006) Biochemistry 45, 15157-15167

61. Oddo, S., Caccamo, A., Tran, L., Lambert, M. P., Glabe, C. G., Klein, W. L., and LaFerla, F. M. (2006) J Biol Chem 281, 1599-1604

62. Taniguchi, S., Suzuki, N., Masuda, M., Hisanaga, S., Iwatsubo, T., Goedert, M., and Hasegawa, M. (2005) J Biol Chem 280, 7614-7623

63. Skovronsky, D. M., Zhang, B., Kung, M. P., Kung, H. F., Trojanowski, J. Q., and Lee, V. M. (2000) Proc Natl Acad Sci U S A 97, 7609-7614

64. Podlisny, M. B., Walsh, D. M., Amarante, P., Ostaszewski, B. L., Stimson, E. R., Maggio, J. E., Teplow, D. B., and Selkoe, D. J. (1998) Biochemistry 37, 3602-3611

65. Heiser, V., Scherzinger, E., Boeddrich, A., Nordhoff, E., Lurz, R., Schugardt, N., Lehrach, H., and Wanker, E. E. (2000) Proc Natl Acad Sci U S A 97, 6739-6744

66. Chirita, C. N., Congdon, E. E., Yin, H., and Kuret, J. (2005) Biochemistry 44, 5862-5872 67. Camilleri, P., Haskins, N. J., and Howlett, D. R. (1994) FEBS Lett 341, 256-258 68. Yu, J., Bakhos, L., Chang, L., Holterman, M. J., Klein, W. L., and Venton, D. L. (2002) J

Mol Neurosci 19, 51-55 69. Yang, F., Lim, G. P., Begum, A. N., Ubeda, O. J., Simmons, M. R., Ambegaokar, S. S.,

Chen, P. P., Kayed, R., Glabe, C. G., Frautschy, S. A., and Cole, G. M. (2005) J Biol Chem 280, 5892-5901

70. Howlett, D., Cutler, P., Heales, S., and Camilleri, P. (1997) FEBS Lett 417, 249-251 71. Netland, E. E., Newton, J. L., Majocha, R. E., and Tate, B. A. (1998) Neurobiol Aging

19, 201-204 72. Wang, J., Gines, S., MacDonald, M. E., and Gusella, J. F. (2005) BMC Neurosci 6, 1 73. Tomiyama, T., Shoji, A., Kataoka, K., Suwa, Y., Asano, S., Kaneko, H., and Endo, N.

(1996) J Biol Chem 271, 6839-6844 74. Porat, Y., Mazor, Y., Efrat, S., and Gazit, E. (2004) Biochemistry 43, 14454-14462 75. Klunk, W. E., Wang, Y., Huang, G. F., Debnath, M. L., Holt, D. P., and Mathis, C. A.

(2001) Life Sci 69, 1471-1484 76. Sabate, R., Gallardo, M., and Estelrich, J. (2005) Int J Biol Macromol 35, 9-13 77. Kusumoto, Y., Lomakin, A., Teplow, D. B., and Benedek, G. B. (1998) Proceedings of

the National Academy of Sciences of the United States of America 95, 12277-12282 78. Necula, M., Chirita, C. N., and Kuret, J. (2005) Biochemistry 44, 10227-10237 79. Evans, K. C., Berger, E. P., Cho, C. G., Weisgraber, K. H., and Lansbury, P. T., Jr.

(1995) Proc Natl Acad Sci U S A 92, 763-767 80. LeVine, H., 3rd. (1993) Protein Sci 2, 404-410 81. D'Andrea, M. R., Reiser, P. A., Polkovitch, D. A., Gumula, N. A., Branchide, B.,

Hertzog, B. M., Schmidheiser, D., Belkowski, S., Gastard, M. C., and Andrade-Gordon, P. (2003) Neurosci Lett 342, 114-118

82. Demuro, A., Mina, E., Kayed, R., Milton, S. C., Parker, I., and Glabe, C. G. (2005) J Biol Chem 280, 17294-17300

15

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

83. Nilsson, K. P., Herland, A., Hammarstrom, P., and Inganas, O. (2005) Biochemistry 44, 3718-3724

84. Necula, M., and Kuret, J. (2004) Anal Biochem 329, 238-246 85. Naiki, H., Hasegawa, K., Yamaguchi, I., Nakamura, H., Gejyo, F., and Nakakuki, K.

(1998) Biochemistry 37, 17882-17889 86. Chen, Y. R., and Glabe, C. G. (2006) J Biol Chem. 281, 24414-24422 87. Bitan, G., Kirkitadze, M. D., Lomakin, A., Vollers, S. S., Benedek, G. B., and Teplow, D.

B. (2003) Proc Natl Acad Sci U S A 100, 330-335 88. Dobson, C. M. (1999) Trends Biochem Sci 24, 329-332 89. Dobson, C. M. (2003) Nature 426, 884-890 90. Glabe, C. G. (2006) Neurobiol Aging 27, 570-575 91. Glabe, C. G. (2004) Trends Biochem Sci 29, 542-547 92. Harper, J. D., and Lansbury, P. T., Jr. (1997) Annu Rev Biochem 66, 385-407 93. Chien, P., Weissman, J. S., and DePace, A. H. (2004) Annu Rev Biochem 73, 617-656 94. Kim, Y. S., Randolph, T. W., Manning, M. C., Stevens, F. J., and Carpenter, J. F. (2003)

J Biol Chem 278, 10842-10850 95. Yao, Z., Drieu, K., and Papadopoulos, V. (2001) Brain Res 889, 181-190 96. Chromy, B. A., Nowak, R. J., Lambert, M. P., Viola, K. L., Chang, L., Velasco, P. T.,

Jones, B. W., Fernandez, S. J., Lacor, P. N., Horowitz, P., Finch, C. E., Krafft, G. A., and Klein, W. L. (2003) Biochemistry 42, 12749-12760

97. Esler, W. P., Stimson, E. R., Ghilardi, J. R., Felix, A. M., Lu, Y. A., Vinters, H. V., Mantyh, P. W., and Maggio, J. E. (1997) Nat Biotechnol 15, 258-263

98. Smith, D. L., Portier, R., Woodman, B., Hockly, E., Mahal, A., Klunk, W. E., Li, X. J., Wanker, E., Murray, K. D., and Bates, G. P. (2001) Neurobiol Dis 8, 1017-1026

99. Pertinhez, T. A., Bouchard, M., Smith, R. A., Dobson, C. M., and Smith, L. J. (2002) FEBS Lett 529, 193-197

100. Chirita, C., Necula, M., and Kuret, J. (2004) Biochemistry 43, 2879-2887 101. Ferrao-Gonzales, A. D., Robbs, B. K., Moreau, V. H., Ferreira, A., Juliano, L., Valente,

A. P., Almeida, F. C., Silva, J. L., and Foguel, D. (2005) J Biol Chem 280, 34747-34754 102. Li, J., Zhu, M., Manning-Bog, A. B., Di Monte, D. A., and Fink, A. L. (2004) Faseb J 18,

962-964 103. Conway, K. A., Rochet, J. C., Bieganski, R. M., and Lansbury, P. T., Jr. (2001) Science

294, 1346-1349 104. Zhu, M., Rajamani, S., Kaylor, J., Han, S., Zhou, F., and Fink, A. L. (2004) J Biol Chem

279, 26846-26857 105. Chirita, C. N., and Kuret, J. (2004) Biochemistry 43, 1704-1714 106. Ahmad, A., Millett, I. S., Doniach, S., Uversky, V. N., and Fink, A. L. (2004) J Biol

Chem 279, 14999-15013 107. Klunk, W. E., Pettegrew, J. W., and Abraham, D. J. (1989) J Histochem Cytochem 37,

1273-1281

16

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Footnotes

*This work was supported by NIH NS-38298 and the Larry L. Hillblom foundation.

1The abbreviations used are: AD, Alzheimer’s disease; Aβ, amyloid β protein; Aβ42, amyloid β protein containing 42 amino acids; BSB, (trans,trans)-1-Bromo-2,5-bis-(3-hydroxycarbonyl-4-hydroxy)styrylbenzene; BSA, bovine serum albumin; C17, trimethyl(tetradecyl)ammonium bromide; C16, hexadecyltrimethylammonium bromide; TBS-T, TRIS-buffered saline containing 0.01% Tween 20; TEM, transmission electron microscopy; ThS, thioflavin S; ThT, thioflavin T; TMB, 3,3’,5,5’-tetramethylbenzidine; PBS, phosphate buffered saline;

2Nomenclature: A11, anti-oligomer specific antibody (58). Inhibition of oligomerization refers to inhibition of formation of A11-immunoreactive oligomers. An inhibitor of oligomerization refers to a molecule that inhibits the formation of A11-immunoreactive oligomers. “oligomerization conditions” refer to conditions that favor A11-reactive oligomer visualization. “fibrillization conditions” refer to conditions that favor fiber formation in the absence of contamination with A11-positive oligomers. Fibrillization conditions and oligomerization conditions is a nomenclature used for the sole purpose to describe conditions where Aβ42 forms oligomer-free fibers or fiber free oligomers (at early time points), respectively. These conditions allow separate screening of effects of compounds on fiber or oligomer formation. This nomenclature is not meant to indicate that these are the best or the only conditions to prepare fibers and oligomers.

Figure Legends

Fig. 1. Time course of Aβ42 assembly under conditions that facilitate oligomerization. Aβ42 was incubated for up to 15 days in 1x PBS, pH=7.4 at room temperature without stirring (45 µM, final concentration), in the presence or absence of 1% DMSO, using stock solutions of freshly dissolved peptide. Aliquots were removed at various time points during aggregation and analyzed in parallel by dot blot (A) and TEM (B-E) assays. A, Aliquots were spotted onto nitrocellulose membranes and probed with A11, 6E10, and 4G8 antibodies. Oligomer-specific immunoreactivity formed within one day of aggregation, became strong after 4 days of incubation, and was stable for extended periods of time. 6E10 and 4G8 immunoreactivity diminished as aggregation proceeded. B-E, Aggregation reactions were also assayed by electron microscopy at 50,000-fold magnification. Aβ42 showed minimal aggregation at time point zero (B) but formed oligomers as small as 3 nm within a day of the initiation of aggregation (C). The oligomers appeared to grow in size to yield a heterogeneous population of oligomers and protofibrils by day 4 (D). These oligomers and protofibrils constituted the predominant population for up to 6 days, after which time they coexisted with fibrils with straight morphology (E). Bar = 150 nm. Fig. 2. Select small-molecules inhibit Aβ42 oligomerization. Aβ42 was incubated under oligomerization conditions without stirring (45 µM, final concentration) for up to five days, in the absence of any additive (column 1), in the presence of 1% DMSO vehicle (column 2), or in the presence of small-molecules listed in Experimental Procedures (all other columns). The ability of each compound to inhibit oligomerization was tested at two different concentrations (30 and 300 µM) placed adjacent to each other (i.e., compound 1: row A, column 3 and column 4; compound 2: row A, column 5 and column 6). The active compounds are listed in Table I and their positioning is as follows: 7 (A5, A6), 12 (A7, A8), 10 (A9, A10), 23 (B3, B4), 25 (B5, B6), 27 (B9, B10), 29 (C3, C4), 21 (C5, C6), 11 (D3, D4), 15 (D5, D6), 18 (D7, D8), 13

17

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

(D9, D10), 20 (E9, E10), 4 (F3, F4), 9 (F5, F6), 16 (F7, F8), 5 (F9, F10), 6 (G5, G6), 24 (G7, G8), 2 (H3, H4), 28 (H5, H6), 22 (H9, H10), 17 (I5, I6), 19 (I7, I8), 1 (I9, I10), 3 (J3, J4), 26 (K3, K4). The intensity of each dot represents the amount of oligomer present in each reaction, probed with the oligomer-specific antibody A11. More than half of the small-molecules tested were able to inhibit Aβ42 oligomerization. Buffer treated with 1% DMSO served as negative control (K8, K9) and preformed Aβ42 oligomers were used as positive control (K7, K10). Fig. 3. Effect of inhibitors on Aβ42 oligomeric species. Aβ42 was incubated under oligomerization conditions without stirring (45 µM, final concentration) for 5 days in the presence of 1% DMSO vehicle (O) or 300 µM small-molecules: azure C (1), basic blue 41 (2), meclocycline sulfosalicylate (3), R(–)-norapomorphine hydrobromide (4), Congo red (5), rolitetracycline (6), Chicago sky blue 6B (7), hemin (8), o-vanillin (9), daunomycin hydrochloride (10), C16 (11), 1,2-naphthoquinone (12), nordihydroguaiaretic acid (13), hematin (14), C17 (15), neocuproine (16), lacmoid (17), rifamycin SV (18), juglone (19), myricetin (20), ThT (21), 2,2’-dihydroxybenzophenone (22), rhodamine B (23), curcumin (24), phenol red (25), indomethacin (26), quinacrine mustard dihydrochloride (27), pherphenazine (28), and eosin Y (29). Each reaction was assessed for immunoreactivity with anti-oligomer specific antibody A11 (A) or 6E10 (B) by western blot. The small-molecules inhibited oligomer formation in a manner that is generally consistent with their IC50s. Fig. 4. Effect of compounds on Aβ42 fibrillization. Aβ42 was incubated under fibrillization conditions with stirring (45 µM, final concentration) for four days in the presence of 1% DMSO vehicle or small-molecules (30 and 300 µM). Data is shown here for 300 µM compound concentration except for C16 and C17 for which data is plotted at both concentrations. Aliquots of each reaction were assayed by turbidity at 400 nm wavelength. Small-molecules tested segregated into compounds that do not inhibit (A), partly inhibit (B), or have biphasic behavior (A, B; C16 and C17) on Aβ42 assembly into fibrils. Each reading was obtained in triplicate and is expressed as % of control reaction ± standard deviation. Compounds that neither inhibited oligomerization nor fibrillization are not shown. Fig. 5. Hypothetical model for Aβ42 aggregation in vitro. In vitro, freshly dissolved Aβ42 mostly exists as an unfolded monomer (M). Based on their ability to modulate Aβ42 aggregation into oligomers and fibers, the compounds tested here appear to stabilize at least three types of intermediate Aβ42 conformations. Class I compounds stabilize an Aβ42 intermediate conformation, I1, that is assembly competent and can aggregate directly into fibrils without previous association into oligomers. Class II compounds stabilize an Aβ42 intermediate conformation, I2, that does not favor Aβ42 assembly into either oligomers or fibrils. Class III compounds stabilize an Aβ42 intermediate conformation, I3, that supports Aβ42 oligomer formation but is incompetent for assembly into fibrils. Dotted lines are used to indicate pathways that are not favored but possible. Our data suggest that Aβ42 adopts multiple intermediate conformations that are not prone to aggregation or support selective aggregation into either oligomers or fibers. The presence of oligomers on the classic Aβ42 fibrillization pathway is possible but not obligatory.

18

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Table I

Potency of compounds active for inhibition of Aβ42 oligomerization. Determined (*) or estimated (**) by ELISA using the anti-oligomer specific antibody A11 (58) from data obtained in triplicate, and expressed as average ± standard deviation. Compound # Name IC50 (µM)* 1 azure C 0.09 ± 0.02 2 basic blue 41 0.12 ± 0.04 3 meclocycline sulfosalicylate 0.25 ± 0.07 4 R(–)-norapomorphine hydrobromide 0.83 ± 0.27 5 Congo red 1.10 ± 0.29 6 rolitetracycline 1.87 ± 0.42 7 Chicago sky blue 6B 2.58 ± 0.57 8 hemin 4.76 ± 1.10 9 o-vanillin 6.15 ± 1.02 10 daunomycin hydrochloride 7.48 ± 2.14 11 hexadecyltrimethylammonium bromide (C16) 7.69 ± 1.88 12 1,2-naphthoquinone 11.48 ± 1.94 13 nordihydroguaiaretic acid 14.47 ± 4.30 14 hematin 15.89 ± 6.68 15 trimethyl(tetradecyl)ammonium bromide (C17) 15.94 ± 5.48 16 neocuproine 17.42 ± 5.20 17 lacmoid 25.51 ± 7.94 18 rifamycin SV 31.64 ± 10.34 19 juglone 67.60 ± 20.67 20 myricetin 106.87 ± 24.30 21 ThT 122.19 ± 32.97 22 2,2’-dihydroxybenzophenone 244.16 ± 78.04 23 rhodamine B 309.61 ± 92.50 24 curcumin 361.11 ± 38.91 25 phenol red 426.25 ± 177.66 26 indomethacin 958 ± nd** 27 quinacrine mustard dihydrochloride 2051 ± nd** 28 pherphenazine 2606 ± nd** 29 eosin Y 2787 ± nd**

19

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Table II

Inhibition of Aβ42 fibrillization by small molecules: correlation between ThT fluorescence and turbidity measurements. *, ** ThT fluorescence (*, λem = 520 nm; **, λem = 482 nm) in the absence of Aβ42 determined from data

obtained in triplicate, and expressed as % control ± standard deviation. λex = 442 nm. *** Inhibition of Aβ42 fibrillization determined by ThT fluorescence (λem = 482 nm) from data obtained

in triplicate, and expressed as % control ± standard deviation. λex = 442 nm. Data was corrected using parallel reactions containing no protein. Negative values indicate higher ThT fluorescence in the absence of protein and suggest compound-specific optical changes in the presence/absence of protein.

**** Inhibition of Aβ42 fibrillization determined by light scattering at 400 nm from data obtained in triplicate (See Fig. 4). I=inhibition of aggregation, NI= no inhibition. The inhibition or lack of inhibition was confirmed by TEM (Supplementary Fig. 3 and data not shown).

º Data not shown. Control reactions contain 1% DMSO vehicle. # Name λem520 nm λem482 nm λem482 nm OD400 nm

* ** *** **** Aβ42 - - + + 1. azure C 541 ± 57 129 ± 14 0.9 ± 1.3 NI 2. basic blue 41 83 ± 25 82 ± 13 -0.4 ± 0.7 NI 3. meclocycline sulfosalicylate 403 ± 83 232 ± 42 -7.6 ± 2.6 I 4. R(–)-norapomorphine hydrobromide 133 ± 16 127 ± 12 4.7 ± 1.8 NI 5. Congo red 53 ± 11 25 ± 5 1.4 ± 0.6 NI 6. rolitetracycline 279 ± 68 235 ± 88 14 ± 6 NI 7. Chicago sky blue 6B 96 ± 15 81 ± 15 1.1 ± 0.9 I 8. hemin 168 ± 37 120 ± 19 -2 ± 1.1 I 9. o-vanillin 203 ± 60 226 ± 139 89 ± 18 I 10. daunomycin hydrochloride 164 ± 56 108 ± 31 -1.1 ± 1.8 NI 11. C16, 300µΜ 87 ± 13 75 ± 13 0.5 ± 0.8 NI C16, 30µΜ 113± 16 106 ± 15 9.8 ± 4.2 I 12. 1,2-naphthoquinone 156 ± 16 75 ± 8 0.3 ± 0.4 NI 13. nordihydroguaiaretic acid 680 ± 81 700 ± 68 -4.5 ± 11 NI 14. hematin 103 ± 12 79 ± 8 1.9 ± 1.9 I 15. C17, 300µΜ 60± 13 55 ± 10 0.7 ± 1.3 NI C17, 30µΜ 84± 12 70 ± 18 112 ± 89 I 16. neocuproine 78 ± 9 77 ± 10 13 ± 12 I 17. lacmoid 1399 ± 169 294 ± 36 2.6 ± 8.1 I 18. rifamycin SV 155 ± 19 206 ± 19 0.7 ± 1 I 19. juglone 71 ± 12 41 ± 10 2.9 ± 0.9 NI 20. myricetin 336 ± 37 253 ± 27 3.9 ± 1.3 NI 21. ThT 148 ± 30 192 ± 30 68 ± 12 NI 22. 2,2’-dihydroxybenzophenone 102 ± 15 125 ± 26 126 ± 22 I 23. rhodamine B 11752 ± 1347 600 ± 71 -0.9 ± 3.1 I 24. curcumin 240 ± 89 127 ± 12 29 ± 7.2 NI 25. phenol red 748 ± 74 112 ± 10 132 ± 26 I 26. indomethacin 147 ± 19 166 ± 22 209 ± 69 NI 27. quinacrine mustard dihydrochloride 21932 ± 2819 25073 ± 2991 -698 ± 193 NI 28. pherphenazine 82 ± 8 72 ± 7 6.4 ± 2.4 NI

20

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

29. eosin Y 49455 ± 4824 10266 ± 2867 -161 ± 172 I 30. orange G 77 ± 8 46 ± 9 10.2 ± 4.2 I 31. diallyltartar 148 ± 50 152 ± 61 -2.5 ± 3.7 I 32. direct red 53 ± 9 32 ± 9 1.8 ± 0.8 I 33. apigenin 293 ± 37 352 ± 1.1 -13 ± 2.8 I 34. dimethyl yellow 91 ± 9 93 ± 10 0.9 ± 0.8 NI º 35. octadecylsulfate 106 ± 12 134 ± 23 26 ± 14 NI º 36. β-cyclodextrin 88 ± 11 100 ± 9 0.1 ± 0.2 NI º 37. BSB 92 ± 24 76 ± 19 2.5 ± 1.2 NI º 38. melatonin 113 ± 21 99 ± 9 0.8 ± 0.7 NI º 39. ThS 145 ± 36 137 ± 31 -0.9 ± 1.9 NI º 40. phthalocyanine 116 ± 42 104 ± 11 1.3 ± 1 NI º 41. fenofibrate 96 ± 22 103 ± 11 0.9 ± 0.6 NI º 42. nystalin 205 ± 28 275 ± 39 -9.2 ± 2.5 NI º 1 % DMSO 100 ± 14 100 ± 12 100 ± 23 100

21

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Table III

Classes of compounds active for inhibition of Aβ42 aggregation. * Assayed by dot blot, ELISA, and western blot using the anti-oligomer specific antibody A11 (58) for

reactions containing 45 µM Aβ42 and 0.01-3000 µM small-molecule inhibitor. ** Assayed by turbidity at 400 nm and TEM for reactions containing 45 µM Aβ42 and 30 and/or 300 µM

small-molecule inhibitor.

Classes of Compounds

Class I (Compounds that inhibit Class II (Compounds that inhibit oligomerization* but do not inhibit both oligomerization* and fibrillization**) fibrillization**)

azure C meclocycline sulfosalicylate basic blue 41 hemin

R(–)-norapomorphine hydrobromide o-vanillin Congo red C16 (30µM)

rolitetracycline hematin daunomycin hydrochloride C17 (30µM) C16 (300µM) neocuproine 1,2-naphthoquinone lacmoid

nordihydroguaiaretic acid rifamycin SV C17 (300µM) 2,2’-dihydroxybenzophenone

juglone rhodamine B myricetin phenol red ThT eosin y

curcumin indomethacin Class III (Compounds that inhibit

quinacrine mustard dihydrochloride fibrillization** and do not inhibit pherphenazine oligomerization*) apigenin Chicago sky blue diallyltartar direct red

orange G

22

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 1

23

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 2

24

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 3

25

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 4

26

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Figure 5

27

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from

Mihaela Necula, Rakez Kayed, Saskia Milton and Charles G. Glabeand fibrillization pathways are independent and distinct

Small-molecule inhibitors of aggregation indicate that amyloid beta oligomerization

published online February 6, 2007J. Biol. Chem. 

  10.1074/jbc.M608207200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2007/02/06/M608207200.DC1

by guest on January 2, 2020http://w

ww

.jbc.org/D

ownloaded from