AD is characterized by the formation of amyloid deposits composed primarily of Aβ [17,18], a natively unfolded protein with little or no ordered structure under physiological conditions [19,20]. Thus, the conditions that decrease the net charge or increase the hydrophobicity could be expected to result in partial folding. Previous studies suggest that the formation of Aβ dimers through the phenolic coupling of tyrosine (termed dityrosine-linkage) could be an early event critical to fibril formation and Aβ deposits in AD [15,16]. Murakami and colleagues reported that Aβ-42 Y10F shows significantly attenuated neurotoxicity compared to that of wild type Aβ-42 [21]. Other reports suggest that the substitution of tyrosine with alanine in the Aβ sequence inhibits the ability of this peptide to catalyze the production of H₂O₂ and to cross-link, as well as inhibiting Aβ neurotoxicity [14]. For these reasons, tyrosine 10 in the chain of Aβ is believed to be critical in the process of AD pathogenesis.

In order to investigate the relationship between amino acid residues of Aβ-40 and conformational changes, our approach was firstly to record circular dichroism (CD) spectra in the far UV region (190–240 nm) to detect the secondary structures of wild type and mutant peptides. The CD spectrum of wild type Aβ-40 was characterized by a strong negative CD band in the 195- to 200-nm region (Figure 1A), indicative of a disordered or random-coil secondary structure [8]. In contrast, the spectral curve of Aβ-40 Y10F displayed a β-sheet spectrum as characterized by a minimum at 220 nm and a positive ellipticity at ~200 nm (Figure 1B).

Aggregation is a characteristic property of partially folded proteins and most aggregating protein systems probably involve a transient partly folded intermediate as the key precursor of fibrillation [22]. To confirm that the Aβ-40 Y10F-induced CD spectrum alteration was directly related to changes in the protein conformation, we performed a kinetic aggregation study to explore the effect of Y10 on the aggregative ability through beta-sheet formation of Aβ-40. The β-aggregation of either wild type Aβ-40 or Aβ-40 Y10F was assessed using the fluorescent dye ThT. Both type of peptides exhibited a gain in fibril, as evident by an increase in ThT fluorescence during the time course. As shown in Figure 2, it was interesting to note that the aggregation of Aβ-40 Y10F showed a more pronounced rise than that of wild type Aβ-40, which could be due to the secondary structural changes of modified Aβ, leading to a greater binding capacity for ThT or an increase in the total amount of aggregation. The aggregation of wild type Aβ-40 reached a steady state at around 60 h, while Aβ-40 Y10F remained in a rapid growth phase until 72 h, suggesting a higher aggregation ability of Aβ-40 Y10F compared to that of wild-type Aβ-40.

Further, peptides were separately incubated at 37 °C without agitation and allowed to fibrillize for 72 h to investigate the aggregation state of Aβ-40 and Aβ-40 Y10F. Both peptides presented fibrillar morphologies though aggregates formed in wild type Aβ-40 apparently consists of shorter fibrils and associate laterally with each, while Aβ-40 Y10F gave rise to longer mature fibrils (long, rigid, and extended fibrils that could be longer than 2 μm in length) and small amounts of spherical structures (Figure 3B). TEM analysis of the fractions of wild type Aβ-40 showed the combined morphologies of fibrils (5–10 nm in diameter and >400 nm in length) and curvilinear aggregates (6–8 nm in diameter and <200 nm in length) (Figure 3A), and the latter are consistent with previously described morphologies for protofibrils that present as dispersed, curvilinear structures of 4–11 nm diameter and <200 nm length [22–24].

The key distinguishing features of protofibrillar aggregates include that they bind to amyloid-specific dyes, such as Congo red and Thioflavin-T (ThT) though much weaker than mature fibrils, and protofibrils are less stable than mature amyloid fibrils [23,25,26]. It is anticipated that the protofibrils or shorter fibrils formed in wild type Aβ-40 are a partially folded form and could be a precursor of fibrils, since the aggregates have contiguous hydrophobic patches on the surface and are prone to foster self-association and hence potentially fibrillation [27], though the structure of wild type Aβ-40 resembles that of other reports under the same concentration of Aβ [25,26]. However, protofilaments as structural subunits of a mature fibril are distinctive from protofibrils [28]. Furthermore, the highly ordered structures such as protofibrils and Aβ-derived diffusible ligands (ADDLs) can be “on pathway” to fibril formation (disappear upon fibril formation) or “off pathway” species (they do not convert into fibrils) [29]. It is important to mention that the protofibrils of wild type Aβ-40 are unlikely to change into fibrils since the images were taken when the process of fibrillization was complete as indicated by the maximum ThT fluorescence (Figure 2). Since synthetic peptides have been shown to have slight biophysical variations in batch-to-batch preparations [30], we repeated EM experiments using three different batches of both peptides to verify the above results. Again, we observed similar conformational patterns using various batches, suggesting that the conformation of Aβ-40 and Aβ-40 Y10F were not due to an unusual synthesis of peptides.

The binding of 1-Anilinonaphthaleine-8-Sulphonic Acid (ANS) to hydrophobic surfaces of protein is represented as enhanced fluorescence and a blue shift in the wavelength of peak emission (λmax), as illustrated by the binding of ANS to wild type Aβ-40 or Aβ-40 Y10F in Figure 4. The results revealed enhanced ANS fluorescence in the presence of both Aβ peptides. After binding to wild type Aβ-40, there was an approximately 1.5-fold increase in quantum yield and a blue shift in maximum emission wavelength for ANS from 506 to 490 nm. By contrast, the change in the fluorescence of ANS upon binding to Aβ-40 Y10F was much more pronounced, with an approximately 2.5-fold increase in quantum yield and a blue shift in maximum emission wavelength to around 476 nm compared to unbound ANS, suggesting structural reorganization of Aβ in the form of more ordered folding when containing the tyrosine to phenylalanine substitution.

The aggregation and toxicity of Aβ highly correlates with its sequence and structure [31]. Previous studies demonstrated that protofibrils or oligomers, rather than Aβ fibrils, mainly contributed to early dendritic and synaptic injury and thereby neuronal dysfunction [11,32]. Using a small molecule methylene blue, Necula et al. successfully inhibited Aβ oligomerization concomitant with the promotion of fibrillization [33], suggesting that oligomers and fibril formation are distinct and competing pathways. According to the above biophysical experiments, Aβ-40 Y10F displayed structural changes and its capability of forming β-sheet rich conformations was apparently increased. To investigate the correlation between the peptides’ aggregation and their physiological toxicity, either Aβ-40 or Aβ-40 Y10F treated cortical neurons were tested to see if residue substitution exerted effects on neurotoxicity.

As determined by MTS assay, exposure to 5 μM wild type Aβ-40 for 96 h showed significantly stronger neurotoxicity than Aβ-40 Y10F, as the cell viability rate was reduced to ~70% (P < 0.01 versus Vehicle, Figure 5). It is possible that the formation of an Aβ tyrosyl radical leads to increased oligomerization via the formation of dityrosine-linkage as an early aggregation step, which is supported by the identification of dityrosine in amyloid plaque [12]. Whilst for neurons treated with 5 μM Aβ-40 Y10F for up to 96 h, the cell viability rate was ~93% (no significant difference versus Vehicle, Figure 5), suggesting a markedly decreased toxicity of Aβ-40 with a tyrosine substitution. This lack of neurotoxicity is in line with Murakami’s reports, which suggest the neurotoxicity of Aβ-42 Y10F was significantly attenuated compared to that of wild type Aβ-42 [21]. However, the reduced toxicity of Aβ-40 with a tyrosine→phenylalanine substitution is contrary to previous study which suggests rodent Aβ-42 (R5G, Y10F, H13R) exhibits similar neurotoxicity to human Aβ-42 [34]. The reason for the discrepancy could probably be attributed to several factors such as differences with the type of peptide employed (various amino acid sequences) or differences in peptide aggregate forms, concentrations, etc.

Aβ has been shown to induce protein oxidation and lipid peroxidation [35,36], while the phenol hydroxyl group of tyrosine (a crucial residue involved in oxidative stress) is easily oxidized through metal ions to give a tyrosyl radical [12]. However, the mechanism of its formation at the molecular level has not fully been clarified. To investigate the contribution of tyrosine to the generation of free radicals, the amount of H₂O₂ produced by wild type Aβ-40 or Aβ-40 Y10F was examined. As shown in Figure 6, Aβ-40 Y10F produced fairly low levels of H₂O₂ as compared to wild type Aβ-40. These results, together with those of the cell viability assay suggest that the Aβ-40 Y10F induced low level of H₂O₂ generation could partly account for its low cytotoxicity provided that H₂O₂ is not the sole factor responsible for Aβ-induced toxicity, confirming that tyrosyl radicals play a critical role in the radical-mediated neurotoxicity of Aβ-40 [15].

All cell culture reagents were obtained from Invitrogen (Carlsbad, USA). 1-anilinonaphthaleine-8-sulphonic acid (ANS) was obtained from Cayman Chemical Company (Michigan, USA). 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), poly-l-lysine, thioflavin T (ThT), and dimethyl sulfoxide (DMSO) were purchased from Sigma-Aldrich (St. Louis, MO, USA). CellTiter 96^® Aqueous One Solution Reagent was procured from Promega (Madison, WI, USA). All other chemical reagents were commercial products of analytical grade and the highest purity available.

Wild type Aβ-40 and Aβ-40 Y10F were synthesized by employing solid phase Fmoc chemistry and produced by Biosynthesis Biotech Co., Ltd, Beijing, China. After removal from the resin and de-protection, samples were purified and characterized by reverse phase HPLC using UV/Vis detection, and the purity of all peptides used was at least 95%. The lyophilized peptides were initially dissolved in HFIP to a final concentration of 1 mg/mL, shaken for 2 h at 4 °C to entirely dissolve the peptide, then aliquoted into siliconized tubes and stored at −80 °C [20]. Prior to experiment, the stock solution was spin-vacuumed using the Integrated Speed-Vac System (Thermo Savant, USA) then diluted in 10 mM phosphate buffer (pH 7.4) to the required concentration.

Secondary structural changes in peptides were detected using CD spectroscopy [8,20]. The stock peptide was spin-vacuumed then dissolved in 10 mM phosphate buffer (pH 7.4), and the final concentration of Aβ in each sample was 50 μM. All measurements were performed in quartz cuvette cells with a pathlength of 1 mm and scanned with a J-810 CD spectropolarimeter (Jasco, Japan). CD measurements were carried out between 190 and 240 nm using the following parameters: 2-nm bandwidth, 20 nm/min run speed, 0.5-nm step size, and 2-s response time. Background values for each test were subtracted from the corresponding CDs of each sample, and the spectra were smoothed using the Jasco software FFT filter function and converted to molar ellipticity.

The relative degree of β-aggregation is determined by ThT that specifically binds to fibrillar structures [37,38]. Aβ stock solution was diluted in phosphate buffer to 10 μM, ThT was then added to each sample to a final concentration of 10 μM. Each sample was measured in terms of fluorescence intensity at 37 °C using a Varioskan multimode microplate spectrophotometer (Thermo, USA) under kinetic fluorometric mode. Measurements were performed every 2 h at an excitation wavelength of 450 nm and an emission of 485 nm [8,20]. To account for the background fluorescence, the fluorescence intensity measured from each control solution without Aβ peptide was subtracted from that of each solution containing Aβ.

The solution of 10 μM Aβ peptide diluted in phosphate buffer was incubated at 37 °C for 72 h. Each sample was spotted onto glow discharged carbon-coated 300 mesh copper grids (Ted Pella Inc., Redding, CA, USA) for 1 min, dried, and then negatively stained with 2% uranyl acetate [39]. The excess was wiped off, and the grid was allowed to dry. Specimens were then examined with a JEM-1400 electron microscope (JEOL, Japan) at 80 kV accelerating voltage.

ANS is a fluorescent dye that binds with high affinity to hydrophobic surfaces of proteins. The emission maximum of ANS undergoes a blue shift and increases significantly upon binding to the hydrophobic portions [40,41]. We measured ANS fluorescence to monitor structural changes in Aβ-40 and Aβ-40 Y10F upon 4 d aging. The final concentration of peptide in the solution was 10 μM, and the [ANS]/[Aβ] ratio in the following experiments was kept equals to 5. Emission spectra were recorded from 430 to 650 nm with excitation at 350 nm.

Cortical neuronal cultures were performed as described previously with some modifications [9]. Briefly, embryonic day 17 Sprague Dawley rat cortices were dissociated and suspended in fresh neurobasal medium plus 2% B27 supplements, 1% penicillin-streptomycin, and 0.5 mM l-glutamine, then plated onto poly-l-lysine-coated 96-well culture plates at a density of 5 × 10⁴ cells per well. This method resulted in cultures highly enriched for neurons (>95% purity) with minimal astrocyte and microglial contents. The neuronal cells were allowed to mature for 7 days in vitro before commencing treatment. For treatment, samples were diluted to a concentration of 5 μM in serum-free neurobasal medium, aged for 72 h, and finally added to the cultures for 96 h.

Cell viability was quantitively determined using the MTS assay as described previously [42]. Briefly, cells were cultured in flat bottomed 96-well plates and exposed to various treatments. 20 μL CellTiter 96^® Aqueous One Solution Reagent (MTS) was added to each well according to the manufacturer’s instructions. After 4 h in culture the cell viability was determined by measuring the absorbance at 490 nm using a μQuant MQX200 reader (Bio-Tek). Background readings of MTS incubated in cell-free medium were subtracted from each value before calculations. The data were normalized and calculated as a percentage of untreated control values prior to analysis.

Analyses were carried out using a hydrogen peroxide assay kit (Beyotime Biotech, China) as described previously [9]. Hydrogen peroxide could oxidize Fe²⁺ to Fe³⁺, and then Fe³⁺ reacted with xylenol orange resulting in a colorimetric reaction that could be further detected by a spectrometer. Briefly, test tubes containing 50 μL of supernatants and 100 μL of test solutions were placed at room temperature for 20 min and measured immediately with a spectrometer at a wavelength of 560 nm. The concentration of H₂O₂ released was calculated from a standard concentration curve in triplicate experiments.