Alzheimer's Disease

Abstract

Variant forms of Amyloid-beta (Aβ) and kits comprising a variant Aβ are disclosed. The invention also provides uses of these kits and the Aβ variants in Aβ studies, for example in assays and methods for screening novel compounds for use in treating Alzheimer's Disease.

Claims

1-33. (canceled)

34. A variant Amyloid-beta (Aβ) peptide comprising a modified amino acid sequence of a wild-type Aβ peptide, wherein the modified amino acid peptide exhibits reduced propensity to aggregate compared to the wild type peptide.

35. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises one or more modification in amino acids 16-21 or 37-42 of SEQ ID No: 1.

36. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises at least two modifications in amino acids 16-21 or 37-42 of SEQ ID No: 1.

37. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises at least one modification in amino acids 16-21 of SEQ ID No: 1 and at least one modification in amino acids 37-42 of SEQ ID No: 1.

38. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide is formed by modification of amino acid residue F19 or G37 of SEQ ID No: 1.

39. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide is formed by modification of amino acid residue F19 and G37 of SEQ ID No: 1.

40. The variant Aβ peptide according to claim 35, wherein the modification at amino acid residue F19 comprises a substitution with a serine.

41. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises an amino acid sequence substantially as set out in SEQ ID No: 2.

42. The variant Aβ peptide according to claim 35, wherein the modification at amino acid residue G37 comprises a substitution with an aspartic acid.

43. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises an amino acid sequence substantially as set out in SEQ ID No: 3.

44. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises a F19S substitution or a G37D substitution.

45. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises an F19S substitution and a G37D substitution.

46. The variant Aβ peptide according to claim 34, wherein the variant Amyloid-beta (Aβ) peptide comprises an amino acid sequence substantially as set out in SEQ ID No: 4.

47. An isolated nucleic acid molecule encoding the variant Amyloid-beta (Aβ) peptide according to claim 34.

48. The isolated nucleic acid molecule according to claim 47, wherein the isolated nucleic acid molecule comprises a nucleotide sequence substantially as set out in any one of SEQ ID No: 5-7, or functional variant thereof.

49. An Amyloid-beta (Aβ) test kit comprising the variant Amyloid-beta (Aβ) peptide according to claim 34.

50. The kit according to claim 49, wherein the kit comprises wild-type Aβ peptide.

51. The kit according to claim 49, wherein the kit comprises a solvent to disassemble any pre-aggregated peptide, and optionally a buffer and/or a desalting column.

52. The kit according to claim 51, wherein the solvent is hexafluoroisopropanol and/or dimethylsulphoxide.

Description

[0074] For a better understanding of the invention, and to show how embodiments of the same may be carried into effect, reference will now be made, by way of example, to the accompanying Figures, in which:—

[0075] FIG. 1 is a graph produced using WALTZ and shows two peaks that indicate two amyloidogenic regions (residues 16-21 and residues 37-42) in the wildtype “Aβ(1-42)” peptide. The amino acids selected for substitution are highlighted in bold, i.e. F and G;

[0076] FIG. 2 shows the output from WALTZ for a variant peptide, “vAβ42”, in accordance with an embodiment of the invention, showing abolition of the amyloidogenic regions due to substitution of F with S, and of G with D;

[0077] FIG. 3 shows a thioflavine T fluorescence fibril formation assay showing increasing fluorescence of wild type Aβ(1-42) with time, compared to no change in fluorescence of the variant, vAβ42;

[0078] FIG. 4 shows tyrosine fluorescence measured at 300 nm and reveals that both wild type Aβ (1-42) and vAβ42 undergo some conformational changes in the environment of the tyrosine 10 residue;

[0079] FIG. 5 shows CD spectra of wild type Aβ(1-42) with time showing that the wild type peptide forms β-sheet structures rapidly leading to amyloid plaques;

[0080] FIG. 6 shows CD spectra of variant vAβ42 with time showing that vAβ42 remains as a random coiled structure up to the final time point of 48 h, and that no β-sheet structures are formed;

[0081] FIG. 7 shows negative stain transmission electron microscopy. FIG. 7(a) shows fibrillar structures formed by wild type Aβ(1-42) after 48 hours compared to the amorphous structures formed by the variant vAβ42 in FIG. 7(b);

[0082] FIG. 8 shows MIT assays that measure the metabolic activity of SH-SY5Y cells and shows that oligomeric wild type Aβ(1-42) 1 and 10 μM have a significant effect on the cells after 24 hours, whilst variant vAβ42 is the same as buffer only;

[0083] FIG. 9 shows that Aβ(1-42) enters neurons and is rapidly distributed through out processes and cell body whilst vAβ does not appear to enter neurons. AlexaFluor555 tagged Aβ(1-42) and vAβ were incubated with neurons for 24 hours and visualised using confocal microscopy. Figure shows differential interference contrast (DIC) compared to the confocal red channel showing the AlexFluor555 tag; and

[0084] FIG. 10 shows that Aβ(1-42) disrupts long term memory after 24 hour in vivo incubation, whilst vAβ does not. One-way ANOVA p<0.0001 Aβ(1-42) n=55, vehicle n=106, naïve n=65, variant control n=20 Snails were tested for rasp rate to amyl acetate, a measure of the feeding response to the CS. Means ± standard error mean (SEM) values are shown. Asterisks indicate behavioural responses that are significantly lower (***=p<0.0001, *=p<0.05) than those in the vehicle-treated group. One-way ANOVA p<0.0001 Tukey's multiple comparison: Aβ(1-42) v Vehicle: p<0.001; vehicle v naïve: p<0.001; variant control v naïve: p<0.05. All others p>0.05

EXAMPLES

Materials and Methods

[0085] Peptide Design

[0086] Sequence based design was performed using the WALTZ algorithm (2) to explore the effect of amino-acid substitutions on the predicted amyloidogenicity of the wildtype Aβ peptide. The graph produced using WALTZ shows two peaks that indicate the location of two amyloidogenic regions (residues 16-21 and residues 37-42) in the wildtype Aβ(1-42) peptide. Substitutions were introduced within the predicted amyloidogenic regions to examine the effect on the graphical output prediction. A number of variants were shown to reduce the predicted amyloidogenicity and two were selected based on literature studies as well as output from WALTZ.

[0087] Preparation and Systemic Application of A β and Variant Peptides

[0088] Wild-type Aβ (1-42) peptide was purchased from rPeptide (http://www.rpeptide.com). Synthetic variant Aβ peptide, “vAβ42”, was purchased from JPT (jpt.com). Both peptides were prepared in the same way using a preparation previously described which uses HFIP and DMSO to solubilize the peptides followed by complete removal of solvents (3, 4). Peptides were prepared in HEPES buffer (10 mM HEPES, 50 mM NaCl, 1.6 mM KCl, 2 mM MgCl.sub.2, 3.5 mM CaCl.sub.2), designed to mimic the culture media as previously described .sup.8, 9. Briefly, 0.2 mg Ab 1-42 (rPeptide) was solubilized in 200 μL HFIP (Sigma-Aldrich) to disaggregate the peptide. The solution was then vortexed on high for one minute and sonicated in a 50/60 Hz bath sonicator for five minutes. HFIP was dried completely using a low stream of nitrogen gas for five to ten minutes. Once completely dried, 200 μL dry DMSO (Sigma-Aldrich) was added, vortexed for one minute, and sonicated for one minute. Solutions were added to a Zeba buffer-exchange column equilibrated with HEPES buffer with 40 μL HEPES as a stacking buffer. The protein solution was kept on ice and the absorbance at 280 nm measured with a NanoDrop spectrophotometer using a molar absorption coefficient of 1490 M.sup.−1 cm.sup.−1. Solutions were immediately diluted to 50 μM with HEPES buffer and incubated for two hours, by which point oligomers are known to form in Aβ 1-42 preparations, before using in further experiments.

[0089] Thioflavine T Fluorescence

[0090] The sample was prepared with 3.121 μM of ThT in a 10 μM Aβ peptide and added to a 10 mm cuvette. An emission scan between a wavelength of 460 nm-600 nm was performed in a Varian Cary Eclipse Fluorescence Spectrophotometer. The sample compartment was set to 21° C., scan rate of 600 nm/min was used and 3 spectra were averaged for each measurement to improve accuracy.

[0091] Tyrosine Fluorescence

[0092] 130 μl of 50 μM Aβ peptide was added to a 10 mm cuvette and an emission scan between wavelength 290 nm-500 nm was performed in a Varian Cary Eclipse Fluorescence Spectrophotometer. The sample compartment was set to 20° C., scan rate of 300 nm/min was used and 3 spectra were averaged for each measurement to improve accuracy.

[0093] Circular Dichroism

[0094] 500 μl of 50 μM Aβ sample was placed into a 1 mm path length quartz cuvette (Hellma) and was scanned between 180 nm to 275 nm on a JASCO Spectropolarimeter J715.3 spectra were averaged for each measurement to improve accuracy and the samples were equilibrated at 20° C. using a water bath.

[0095] Transmission Electron Microscopy

[0096] TEM grids were prepared using Formvar/carbon film (Agar scientific) coated, 400 mesh copper grids. 4 μl of 50 μM Aβ was placed on the surface of the grid and allowed to be absorbed for 60 s and blotted dry. A 4 μl aliquot of miliQ-filtered water was then added to the grid and blotted dry after 60 s. Immediately after this the grid was negatively stained with 4 μl of 2% (w/v) uranyl acetate for 60 s and blotted dry. The uranyl acetate wash was repeated once more and the grid was left to air dry. All the TEM grids were examined using a Hitachi-7100 TEM at 100 kV and the images were acquired digitally with an axially mounted (2000×2000 pixel) Gatan Ultrascan 1000 CCD camera. Aliquots of Aβ peptide samples were taken at different time points to monitor the fibrillation state and morphology.

[0097] Cell Metabolism Assays

[0098] The Vybrant MTF [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide] cell-proliferation assay (Invitrogen) was used according to the manufacturer's protocol to assess the toxic effect of Aβ42 oligomers on undifferentiated SH-SY5Y cells. SH-SY5Y cells (2×10.sup.5 cells/well) were seeded on uncoated or collagen-I-coated glass coverslips in a 24-well plate 1 day prior to the assay. The cells were incubated 10 or 25 μmol.dm.sup.−3 oligomeric Aβ42 or variant Aβ for 1, 5 or 24 h at 37° C. At specific time points, 12 mmol.dm.sup.−3 MTT solution was added to the cells and further incubated for 2 hours at 37° C. The resulting insoluble dye was dissolved with 50 μL of DMSO and the fluorescence measured at 540 nm with a 620 nm reference filter. Untreated cells served as a reference and the value was set to 100% redox activity and compared to the treated cells, and then converted to percentage survival.

[0099] Immunofluorescence Comparison of Aβ 1-42 Compared to vAβ Internalisation

[0100] Aβ was tagged with AlexaFluor555 or 488 as previously described.sup.8. Briefly, the above protocol was followed up until the addition of DMSO. 10 μL of 113 nM Alexa fluor dye and 20 μL 1M sodium bicarbonate were added to Aâ.sub.42 in DMSO. This was incubated for 15 minutes and the remaining stages of the protocol for Aâ preparation carried out. Alexa Fluor tagged Aβ was added to P0-P1 primary rat hippocampal neurons and incubated the desired length of time, after which the cells were washed once quickly with warmed EBS (external bath solution: 137 mM NaCl, 5 mM KCl, 3 mM CaCl.sub.2, 1 mM CaCl.sub.2, 10 mM D-Glucose, 5 mM HEPES) fixed in 2% paraformaldehyde for 15 minutes, washed three times for 5 minutes each with PBS and mounted in Prolong Gold (Life Technologies).

[0101] For samples that were additionally immunolabelled after incubation with Abeta, cells were first washed and fixed as above, then washed with wash buffer (25% Superblock (Thermo Scientific) diluted into PBS) and permeabilised with 0.3% Triton X-100 for 10 minutes. 50 mM Glycine was added for 5 minutes then the cells were washed and incubated for 30 minutes with Image-iT FX signal enhancer then blocked with Superblock (Thermo Scientific) for 30 minutes. Anti-Abeta oligomer antibody NU1 was added, followed by secondary antibody Anti-Mouse Alexa fluor 555 conjugate, each for 1 hour. Cells were then washed and mounted as described above.

[0102] Confocal Microscopy and Image Overlay

[0103] Samples were imaged on a Leica TCS SP8 using a 63x oil objective. Z stacks were taken with 0.5 um step size and images shown are maximal projections of the stack. Images were prepared using images prepared using FIJI software.sup.10.

[0104] Memory Test in Lymnaea Stagnalis

[0105] Pond snails, Lymnaea stagnalis, were bred at the University of Sussex and maintained in large holding tanks containing 18-22° C. copper-free water, at a 12:12 hour light-dark cycle. The animals were fed Tetra-Phyll (TETRA Werke) twice a week and lettuce three times a week. The peptides were administered to the animals directly after preparation. Using a 1 mL syringe with 30 gauge precision glide needles (Becton Dickinson), 100 μL of the Aβ1-42 or variant control peptide solution was injected into the haemolymph (˜1 mL in volume) of each snail. The estimated final concentration in the animal was 0.1 μM for Aβ1-42 and variant conrol. As there is no blood-brain barrier in Lymnaea (Sattelle and Lane, 1972), the injected peptides have direct access to the animal's central nervous system. For vehicle-injected control animals, 100 μL of normal saline was injected.

[0106] Using well-established methods (I. Kemenes et al., 2006), four-to six-month-old snails were removed from their home tanks and starved in new tanks for two days at the same temperature and light dark cycle as the home tanks. After the starvation period, the animals underwent single-trial food-reward classical conditioning (Alexander et al., 1984) in which the CS (amyl acetate: 0.004% final concentration) and the US (sucrose: 0.6% final concentration) were paired. Initially, each individual snail was left to acclimatise in a 14 cm diameter Petri dish with 90 mL of 18-22° C. copper-free water for ten minutes. After the acclimatisation period, 5 mL of amyl acetate was added to the dish and after thirty seconds, 5 mL of sucrose was added. The snails were then left in their Petri dishes for two minutes, and then removed to their starvation tanks. Both the vehicle-injected and Aβ-injected groups were trained. The naïve groups were not trained, but underwent the same starvation/feeding schedule and handling.

[0107] All animals were tested with the CS. Each individual snail was left to acclimatise in a 14 cm-diameter Petri dish with 90 mL of 18-22° C. copper-free water for ten minutes. After the acclimatisation period, 5 mL of 18-22° C. copper-free water was added to the dish. Rasps, the animals' feeding movements, were manually counted for two minutes to determine a baseline rasping rate (number of rasps per two minutes) for each individual. After two minutes, 5 mL amyl acetate was added to the dish. Rasping was tracked for two minutes. Rasping rates were determined by subtracting the individual animal's baseline rasp from the amyl acetate induced rasp.

[0108] Data that passed the D'Agostino and Pearson omnibus normality test were subjected to parametric tests (one-way analysis of single variance [ANOVA] with Tukey's multiple comparison, or t-tests) to establish significance (criterion, p<0.05). GraphPad Prism software was used for all analyses.

[0109] Results

[0110] Design of a Variant Control Peptide Based on Wild Type Aβ(1-42)

[0111] Referring to FIG. 1, there is shown the wildtype Aβ(1-42) peptide comprising 42 amino acids. All 19 remaining amino acid substitutions were introduced into the two amyloidogenic regions in Aβ42 identified from WALTZ identified as residues 16-21 and residues 37-42. The amino acid substitutions that reduced the peaks for amyloidogenic regions were then shortlisted (see Table 1).

TABLE-US-00008 TABLE 1  Shows substitution in to the two amyloidogenic regions identified by WALTZ that result in removal of the amyloidogenic propensity peak. Sequence substituted Substitution at F19 that removed Amyloidogenic region 1 KLVFFA F19P (Proline) KLVPFA (SEQ ID NO. 8) F19D (aspartic acid) KLVDFA (SEQ ID NO. 9) F19R (Arginine) KLVRFA (SEQ ID NO. in) F19C (Cysteine) KLVCFA (SEQ ID NO. 11) F19Q (Glutamine) KLVQFA (SEQ ID NO. 12) F19G (Glycine) KLVGFA (SEQ ID NO. 13) F19H (Histidine) KLVHFA (SEQ ID NO. 14) F19K (Lysine) KLVKFA (SEQ ID NO. 15) F19M (Methionine) KLVMFA (SEQ ID NO. 16) F19S (Serine) KLVSFA (SEQ ID NO. 17) Substitutions at G37 that removed Amyloidogenic region 2 GGVVIA G37R (Arginine) RGVVIA (SEQ ID NO. 18) G37D (Aspartic acid) DGVVIA (SEQ ID NO. 19) G37C (Cysteine) CGVVIA (SEQ ID NO. 20) G37E (Glutamic acid) EGVVIA (SEQ ID NO. 21) G37Q (Glutamine) QGVVIA (SEQ ID NO. 22) G37H (Histidine) HGVVIA (SEQ ID NO. 23) G37I (Isoleucine) IGVVIA (SEQ ID NO. 24) G37L (Leucine) LGVVIA (SEQ ID NO. 25) G37K (Lysine) LGVVIA (SEQ ID NO. 26) G37M (Methionine) MGVVIA (SEQ ID NO. 27) G37F (Phenylalanine) FGVVIA (SEQ ID NO. 28) G37P (Proline) PGVVIA (SEQ ID NO. 29) G375 (Serine) SGVVIA (SEQ ID NO. 30) G37T (Threonine) TGVVIA (SEQ ID NO. 31) G37W (Tryptophan) WGVVIA (SEQ ID NO. 32) G37Y (Tyrosine) YGVVIA (SEQ ID NO. 33) G37V (Valine) VGVVIA (SEQ ID NO. 34)

[0112] Out of all the shortlisted substitutions that worked, double substitution F19S, G37D was selected, as shown in FIG. 2. Substituting phenylalanine at position 19 (F19) with Serine (S) and Glycine at position 37 (G37) with Aspartic acid (D) indicates removal of the amyloidogenic regions in the variant Aβ42 peptide (FIG. 2). Therefore, it can be hypothesised that substituting phenylalanine at position 19 that is tightly packed at a hydrophobic region with a smaller and less hydrophobic amino acid-serine and glycine at position 37 which is a small non-polar, neutral amino acid with a negatively charged and acidic aspartic acid has some affect in disrupting the intermolecular contacts and thus prevent variant vAβ42 from aggregating into amyloid plaques.

[0113] Referring to FIG. 3, Thioflavine T fluorescence was used to monitor amyloid assembly with time and showed that whilst wild type Aβ(1-42) showed an increasing fluorescent signal with time, there was no change in fluorescence at 483 nm for the variant vAβ42 suggesting that this peptide does not self-assemble.

[0114] Tyrosine fluorescence has been used previously to monitor the change in fluorescence as the Aβ peptide assembles and changes the environment of the tyrosine residue at position 10 (5). Referring to FIG. 4, there is shown tyrosine fluorescence measured at 300 nm and reveals that both wild type Aβ(1-42) and variant vAβ42 undergo conformational changes in the environment of the tyrosine 10 residue.

[0115] CD is used to monitor the conformational change from random coil to β-sheet structure that accompanies amyloid assembly. CD spectra confirm that whilst wild type rapidly forms β-sheet structures, the variant vAβ42 remains random coil conformation for the duration of the experiment, as shown in FIGS. 5 and 6. As such, the vAβ42 is not forming β-sheets which would create amyloid plaques in vivo.

[0116] Electron microscopy was used to examine the morphology of the structures over time. As shown in FIG. 7, after 48 hours, wild type Aβ(1-42) had formed fibrils as expected, whilst the variant vAβ42 forms small spherical structures that appear to be variable and amorphous after 48 hour incubation at 50 μM.

[0117] Wild type Aβ(1-42) has been shown to have a toxic effect on cultured neuroblastoma cells and neurons (6,7). In order to investigate the effect of the variant vAβ42 on cells and to compare to wild type Aβ(1-42), an MTT assay was conducted to assess the effect on metabolic activity of SH-SY5Y cells. As shown in FIG. 8, the results revealed that oligomeric wild type Aβ(1-42) decreases the metabolic activity of the neuroblastoma cells, whilst the variant vAβ has no effect and is comparable to vehicle buffer only.

[0118] Immunofluorescence Comparison of Wild Type AP Compared to Variant vAβ Internalisation

[0119] Tagged wild type Aβ(1-42) and variant vAβ42 were added to neuronal cultures and then visualised using a confocal microscope at time points following addition of 24 hours, as shown in FIG. 9. Clear differences in the pattern of uptake were observed between Aβ(1-42) and vAβ42. In particular, Aβ(1-42) appears to enter the cell body and to associate with the processes of the neurons, whilst vAβ42 is not observed and does not appear to enter the neurons.

[0120] Memory Test in Lymnaea Stagnalis

[0121] Aβ(1-42) and vAβ2 were administered to Lymnaea Stagnalis in a conditioned response memory test as previously described.sup.11. In this test, FIG. 10 shows the reduction in rasp rate following Aβ(1-42) compared to the vehicle (buffer only control). Variant vAβ42 is denoted by “control peptide” and shows a similar rasp rate to the vehicle control showing that vAβ42 does not have the ability to alter the memory in the snails.

[0122] Peptide Preparation Kit

[0123] The inventors have developed a peptide preparation kit which includes the variant vAβ42 peptide as a control, and which can then be used in a variety of assays to explore the effects of the Alzheimer's Aβ. These assays could be wide ranging, including but not limited to: [0124] a) aggregation assays; [0125] b) cell toxicity assays; and [0126] c) animal tests (behavioural tests, molecular, cellular or tissue changes).

[0127] The kit includes:— [0128] (i) a vial containing wild-type Aβ(1-42); [0129] (ii) a vial containing variant vAβ42 (i.e. the peptide of the invention); [0130] (iii) solvent (Hexafluoroisopropanol); [0131] (iv) solvent (Dimethylsulphoxide, dry); [0132] (v) buffer (HEPES, PB etc); and [0133] (vi) desalting column (Invitrogen).

[0134] The wild-type Aβ (1-42) and variant Aβ42 are prepared using the kit in an identical way to ensure consistent starting peptides in disaggregated form, which can then be used in the subsequent assays in a detection kit (described below). The solvents are provided to ensure that the peptides are disaggregated.

[0135] Aggregation/Toxicity Detection Kit

[0136] The two peptides are used in assays, including cell toxicity, cell uptake, membrane permeation, Aβ localisation using live cell imaging and immunofluorescence, immunogold electron microscopy, animal behaviour, molecular studies etc. to compare and contrast the action and behaviour of the wild-type Aβ to the control variant Aβ. This will provide valuable information about the specific effects of wildtype Aβ for understanding its role in Alzheimer's disease. Any of these assays could include the addition of test compounds, but this is not necessary, as many of the assays will focus on finding targets and understanding the biochemical effects rather than drug discovery per se.

[0137] In one embodiment, a test compound is added to the kit following preparation of the two peptides under the protocol contained within the preparation kit. A known amount of a test compound is introduced into the assay (cell toxicity etc), and the amount of aggregation and/or toxicity as detected and quantified, and compared. As discussed above, the variant Amyloid-beta (Aβ) peptide exhibits reduced propensity to aggregate compared to the wild type peptide, and so is used as a negative control against which aggregation of the wild type can be measured. An alteration in aggregation and/or toxicity of the wild-type Aβ peptide in the presence of the test compound compared to that of the variant Aβ peptide indicates that the test compound is a modulator of aggregation or toxicity of wild-type Aβ peptide. The kit can be used to screen a therapeutic agent useful in the prophylaxis or treatment of Alzheimer's disease.

References

[0138] 1. Sipe, J. D. (1992) Amyloidosis. Ann Rev Biochem 61, 947-975

[0139] 2. Maurer-Stroh, S., Debulpaep, M., Kuemmerer, N., de la Paz, M. L., Martins, I. C., Reumers, J., Morris, K. L., Copland, A., Serpell, L., Serrano, L., Schymkowitz, J. W., and Rousseau, F. (2010) Exploring the sequence determinants of amyloid structure using position-specific scoring matrices. Nat Methods 7, 237-242

[0140] 3. Broersen, K., Jonckheere, W., Rozenski, J., Vandersteen, A., Pauwels, K., Pastore, A., Rousseau, F., and Schymkowitz, J. (2011) A standardized and biocompatible preparation of aggregate-free amyloid beta peptide for biophysical and biological studies of Alzheimer's disease. Protein engineering, design & selection: PEDS 24, 743-750

[0141] 4. Williams, T. L., Johnson, B. R., Urbane, B., Jenkins, A. T., Connell, S. D., and Serpell, L. C. (2011) Abeta42 oligomers, but not fibrils, simultaneously bind to and cause damage to ganglioside-containing lipid membranes. The Biochemical journal 439, 67-77

[0142] 5. Al-Hilaly, Y. K., Williams, T. L., Stewart-Parker, M., Ford, L., Skaria, E., Cole, M., Bucher, W. G., Morris, K. L., Sada, A. A., Thorpe, J. R., and Serpell, L. C. (2013) A central role for dityrosine crosslinking of Amyloid-beta in Alzheimer's disease. Acta neuropathologica communications 1, 83

[0143] 6. Soura, V., Stewart-Parker, M., Williams, T. L., Ratnayaka, A., Atherton, J., Gorringe, K., Tuffin, J., Darwent, E., Rambaran, R., Klein, W., Lacor, P., Staras, K., Thorpe, J., and Serpell, L. C. (2012) Visualization of co-localization in Abeta42-administered neuroblastoma cells reveals lysosome damage and autophagosome accumulation related to cell death. The Biochemical journal 441, 579-590

[0144] 7. Irvine, G. B., El-Agnaf, O. M., Shankar, G. M., and Walsh, D. M. (2008) Protein aggregation in the brain: the molecular basis for Alzheimer's and Parkinson's diseases. Mol Med 14, 451-464

[0145] 8. Soura, V. et al. Visualization of co-localization in Abeta42-administered neuroblastoma cells reveals lysosome damage and autophagosome accumulation related to cell death. The Biochemical journal 441, 579-590, doi:10.1042/BJ20110749 (2012).

[0146] 9. Broersen, K. et al. A standardized and biocompatible preparation of aggregate-free amyloid beta peptide for biophysical and biological studies of Alzheimer's disease. Protein engineering, design & selection: PEDS 24, 743-750, doi:10.1093/protein/gzr020 (2011).

[0147] 10. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nature methods 9, 676-682, doi:10.1038/nmeth.2019 (2012).

[0148] 11. Ford, L. et al. Effects of Abeta exposure on long-term associative memory and its neuronal mechanisms in a defined neuronal network. Scientific reports 5, 10614, doi:10.1038/srep10614 (2015).