Early diagnosis and treatment of Alzheimer's disease by targeting toxic soluble Aβ oligomers

Abstract

Transient soluble oligomers of amyloid-β (Aβ) are toxic and accumulate early prior to insoluble plaque formation and cognitive impairment in Alzheimer's disease (AD). Synthetic cyclic D,L-α-peptides (e.g., 1) self-assemble into cross β-sheet nanotubes, react with early Aβ species (1-3 mers), and inhibit Aβ aggregation and toxicity in stoichiometric concentrations, in vitro. Employing a semicarbazide as an aza-glycine residue with an extra hydrogen-bond donor to tune nanotube assembly and amyloid engagement, [azaGly⁶]-1 inhibited Aβ aggregation and toxicity at substoichiometric concentrations. High-resolution NMR studies revealed dynamic interactions between [azaGly⁶]-1 and Aβ42 residues F19 and F20, which are pivotal for early dimerization and aggregation. In an AD mouse model, brain positron emission tomography (PET) imaging using stable ⁶⁴Cu-labeled (aza)peptide tracers gave unprecedented early amyloid detection in 44-d presymptomatic animals. No tracer accumulation was detected in the cortex and hippocampus of 44-d-old 5xFAD mice; instead, intense PET signal was observed in the thalamus, from where Aβ oligomers may spread to other brain parts with disease progression. Compared with standard ¹¹C-labeled Pittsburgh compound-B (¹¹C-PIB), which binds specifically fibrillar Aβ plaques, ⁶⁴Cu-labeled (aza)peptide gave superior contrast and uptake in young mouse brain correlating with Aβ oligomer levels. Effectively crossing the blood-brain barrier (BBB), peptide 1 and [azaGly⁶]-1 reduced Aβ oligomer levels, prolonged lifespan of AD transgenic Caenorhabditis elegans, and abated memory and behavioral deficits in nematode and murine AD models. Cyclic (aza)peptides offer novel promise for early AD diagnosis and therapy.

Keywords: Alzheimer’s disease; PET imaging; cyclic D,L-α-(aza)peptide; early diagnosis and therapy; soluble Aβ oligomers.

Fig. 1.

(A) Potential influence of azaGly residue on hydrogen-bonding of cyclic D,L-α-hexapeptide nanotubes which share common cross β-sheet structures with (B) Aβ. For clarity, all side chains have been omitted. (C) Structures of cyclic D,L-α-peptide 1 and azaGly-cyclopeptides (2-7).

Fig. 2.

Inhibitory activity of (aza)peptides 1 and 7 on Aβ aggregation. ThT fluorescence kinetics of Aβ40 in the absence or presence of (A) 1 and (B) 7. (C) Effect of 7 on aggregation of Aβ42. Incubation of Aβ40 (10 µM) and Aβ42 (25 µM), respectively, without or with increasing concentrations of (aza)peptides was done at 37°C, and 100% aggregation was determined based on ThT fluorescence of Aβ alone. Experiments were carried out in triplicate. Data represent one out of the mean ± SD of three experiments. TEM images after 72 h of (D) Aβ40 (10 µM) alone; Aβ40 treated with 7 at (E) 50 μM and (F) 1 μM; 7 alone at (G) 50 μM and (H) 1 μM in PBS. Negatively stained samples are shown. Effects of 7 on Aβ40 oligomer and fibril formation using oligomer- (A11: I Left and J Upper panels) and fibrillar-specific (OC: I Right and J Lower panels) antibodies. Monomeric Aβ40 (30 µM) was aged up to 72 h alone and with (I) 150 µM of 1 and 7 or (J) 3 µM of 7, spotted onto nitrocellulose membranes and probed with A11 and OC antibodies. Capability and kinetics of 7 to disassemble Aβ40 fibrils. (K) Relative fluorescence intensity in ThT assays of 3-d-aged fibrillar Aβ40 (10 µM) incubated for 24 h alone or with a fivefold excess of 1 or 7. (L) Aβ40 fibril disassembly rate with 1 and 7. Results are the average ± SD (n = 3 each) and representative of two experiments. (M) Effects of 7 on Aβ40 secondary structure transition. Time dependent far-UV CD spectra of freshly prepared Aβ40 monomer (10 μM) incubated alone and with substoichiometric concentration (0.1 µM) of 7 in phosphate buffer (50 mM, pH 7.4) after 0 h (full lines) and 24 h (dashed lines). Data are representative of two independent experiments.

Fig. 3.

NMR characterization of the Aβ42 – azapeptide 7 interaction. (A) Annotated aromatic region of the ¹H,¹³C-HSQC spectrum of 20 μM Aβ42 (in 95:5 H₂O:d₆-DMSO) with assignments based on the homonuclear NOESY. (B) The H^aro/H^β region of the Aβ40 homonuclear 2D NOESY spectrum acquired in ²H₂O used for assignment of the Aβ42 aromatic region. Aβ40 allows for higher sample concentrations without aggregation and exhibits identical aromatic signals in the NOESY spectrum, due to the presence of the relevant residues in the N-terminal portion of the sequence. Histidine H^δ is slightly shifted due to a small pH change. (C) Same as (A) after addition of 2 μM of azapeptide 7. (D) Peak intensity ratios (I/I₀) for spectra with and without the azapeptide, calculated as an average over all protons of a given residue. An aggregate value is shown for overlapping H6/H13/H14. Asterisks denote that in the presence of azapeptide 7, F19 and F20 cross-peaks were undetectable above noise level.

Fig. 4.

Effects of peptide 1 and azapeptide 7 on Aβ-mediated toxicity in PC12 cells and in vivo in transgenic C. elegans models. (A) Aβ40 (20 μM) was aged for 48 h alone or with increasing concentrations of 1 and 7, exposed to PC12 cells for 24 h and cell viability was determined by the MTT assay. Results are expressed as a percentage of the control (untreated) cells and are reported as mean ± SD from three assays and analyzed by a one-way ANOVA followed by Tukey’s multiple comparison test (n = 3 each; *P < 0.05, and **P < 0.01). (B) Kaplan–Meier survival plots and (C) median lifespan of transgenic CL2006 and WT CL802 strains treated with 7 (50 µM) or vehicle (5% DMSO) reported as mean ± SD from three experiments and were analyzed by a one-way ANOVA followed by Tukey’s multiple comparison test (n = 25 each; ***P < 0.001, ****P < 0.0001, ns = not significant). (D) Effect of 7 on motility of C. elegans models. Transgenic CL2355 and control CL2122 worms were fed increasing concentrations of 7 at 16°C for 36 h and then at 23°C for another 36 h, before the number of thrashes were counted for 1 min. Data are plotted as a mean ± SD from three experiments and were analyzed by a one-way ANOVA followed by Tukey’s multiple comparison test (n = 20, each; *P < 0.05, and ****P < 0.0001, ns = not significant). (E) Chemotaxis behavior of neuronal Aβ-expressing strain C. elegans CL2355 and transgenic control strain CL2122 in the absence or presence of 7. Worms were fed increasing concentrations of 7 at 16°C for 36 h and then at 23°C for another 36 h. At the end of the incubation time, the animals were placed in the center of an assay plate (100 x 15 mm) containing 1 µL of attractant (0.1% benzaldehyde in ethanol) and 1 µL 1 M sodium azide at two opposite edges of the plate, and 1 µL absolute ethanol with 1 µL 1 M sodium azide on the remaining opposite edges of the plate. The chemotactic index (CI) was then determined after 1-h incubation at room temperature, using the formula: CI = (number of worms at attractant sites - number of worms at control sites)/total number of worms. Results are reported as mean ± SD from three independent experiments and were analyzed by a one-way ANOVA followed by Tukey’s multiple comparison test (n = 20 each; ***P < 0.0005, ****P < 0.0001, ns = not significant). (F) Representative dot-blot analysis of equal amounts of the proteins extracted from C. elegans CL2006 and CL802 strains fed 7 or vehicle, after probing with sequence-specific 6E10 and oligomer-specific A11 antibodies. (G) Representative western blot (WB) analysis of Aβ species in the transgenic C. elegans CL2006 and the control CL802 fed with or without 7. Equal amounts of extracted proteins were loaded onto each gel lane and immunoblotted with an anti-Aβ antibody (6E10) or α-tubulin. (H) Quantification using ImageJ software of Aβ oligomers (the bands at 14 and 27 kDa) in CL2006 and CL802 worms fed either vehicle or 7. Representative images of Aβ deposits (shown as white arrows) in transgenic C. elegans CL2006 fed (I) without or (J) with 7 obtained by thioflavin S staining. (K) Quantitative analysis using ImageJ of Aβ deposits in (I) and (J). The quantity is expressed as mean number ± SD of Aβ deposit area per total examined area of the worm (n = 3 for each analysis).

Fig. 5.

Representative fused PET-CT images at 1-d postinjection of ⁶⁴Cu-9 into (A) 44-d- and (B) 72-d-old 5xFAD mice (n = 5) monitoring progression of Aβ pathology; (C) 44-d-old WT mice (n = 3) at 1-d postinjection of ⁶⁴Cu-9; (D) 40 min postinjection of ¹¹C-PIB into 95-d-old 5xFAD mice (n = 2). Red and blue represent, respectively, highest and zero uptake in %ID/g. A different intensity scale was used for PET images (D) with ¹¹C-PIB (0.1–1.5 %ID/g) compared with 0.1–0.45 %ID/g for A, B and C. The yellow arrow points to Aβ species accumulation in the thalamus. White arrows point to plaque accumulation in the cortex, hippocampus, and thalamus. In (B and D), coregistration of experimental image data with brain mask to identify mouse atlas space volumes of interest (VOI). 1, striatum; 2, cortex; 3, hippocampus; 4, thalamus; 5, cerebellum; 6, basal forebrain septum; 7, hypothalamus; 8, amygdala; 9, brainstem; 10, central gray; 11, superior colliculi; 12, olfactory bulb; 13, midbrain; 14, inferior colliculi.

Fig. 6.

Immunohistochemical staining of Aβ species in the 2-mo-old 5xFAD mouse brain. (A) Sagittal section of the 5xFAD mouse brain PET imaged with ⁶⁴Cu-9 and immunohistochemically examined with sequence-specific 6E10 antibody (recognizing human full-length APP) and developed with anti-mouse horseradish peroxidase (HRP) secondary antibody and 3, 3′-diaminobenzidine (DAB). Positive 6E10 antibody staining was detected in a) cortex, b) hippocampus, c) thalamus and d) pons regions. (B) Representative immunofluorescence image of a coronal section of 2-mo-old 5xFAD mice developed by oligomer-specific A11 antibody and Alexa Fluor 647-labeled secondary antirabbit antibody. Positive A11 antibody staining was detected in the a) cortex, b) hippocampus, c) thalamus, and e) amygdala. (C) Magnified images (×20) of selected areas in (B). (Scale bars, represent 100 μm.) (D) Quantification of fluorescence intensity of A11-positive Aβ oligomers obtained from a similar area of the cortex (COR), hippocampus (HPC), and c) thalamus (THL), using ImageJ. (E) Representative immunofluorescence image (×60) of punctate intracellular (white arrowhead) and extracellular Aβ oligomers (orange arrowhead) in the thalamus of the 2-mo-old 5xFAD mouse and developed by A11 antibody.

Fig. 7.

Effect of chronic treatment with peptide 1 on behavioral activity of 5xFAD mice. (A) Anxiety is measured by the number of entries to the center of the open-field apparatus (n = 10, each). (B) Basal locomotor activity of the mice in the open-field test as measured by the total distance traveled by mice. (C) Novel object recognition performance of mice as determined by frequency of novel object exploration. Results represent mean ± SEM and were analyzed by a one-way ANOVA with Bonferroni’s multiple comparison test (*P < 0.0003, **P < 0.0095, ns = not significant). (D) Increase in spontaneous alterations behavior of 5xFAD mice treated with peptide 1 in the Y-maze test. Mean ± SEM; a one-way ANOVA followed by Bonferroni’s multiple test (*P < 0.01, ns=not significant). (E) Latency to reach the target. Learning curve in the Barnes maze from 6 d of spatial learning. Results represent mean ± SEM and were analyzed by a two-way ANOVA followed by Tukey’s multiple comparison test (*P < 0.0042). (F) Memory for the target hole on the probe day (day 7) in modified Barnes maze. Transgenic mice show a higher latency time to target, and (G) enter the target hole significantly less often than WT mice (*P < 0.046). Treatment of 5xFAD mice with peptide 1 decreased (F) the latency time (*P < 0.0170) and (G) increased the number of entries to target hole (**P < 0.040). Results in F and G represent mean ± SEM and were analyzed by the Kruskal–Wallis test.

Fig. 8.

Effect of treatment with peptide 1 on Aβ plaque load of 5xFAD mice as determined by IHC using 6E10 antibody. (A, D) Fluorescent images of the brain hemisphere of 5xFAD mice treated with (A) vehicle or (D) peptide 1 for 16 wk. Inserts represent (a) CA1, (b) DG, (c) thalamus, (d) cortex and (e) amygdala. High-magnification images of (B and E) CA1 and (C and F) DG regions. Number of Aβ plaques per equal area of (G) CA1 and (H) DG. *P < 0.05. Data in G and H represent mean ± SD and were analyzed by a two-tailed t test (n = 5).