A cationic amphiphilic peptide chaperone rescues Aβ₄₂ aggregation and cytotoxicity

Abstract

Directing Aβ₄₂ to adopt a conformation that is free from aggregation and cell toxicity is an attractive and viable strategy to design therapeutics for Alzheimer's disease. Over the years, extensive efforts have been made to disrupt the aggregation of Aβ₄₂ using various types of inhibitors but with limited success. Herein, we report the inhibition of aggregation of Aβ₄₂ and disintegration of matured fibrils of Aβ₄₂ into smaller assemblies by a 15-mer cationic amphiphilic peptide. The biophysical analysis comprising thioflavin T (ThT) mediated amyloid aggregation kinetic analysis, dynamic light scattering, ELISA, AFM, and TEM suggested that the peptide effectively disrupts Aβ₄₂ aggregation. The circular dichroism (CD) and 2D-NMR HSQC analysis reveal that upon interaction, the peptide induces a conformational change in Aβ₄₂ that is free from aggregation. Further, the cell assay experiments revealed that this peptide is non-toxic to cells and also rescues the cells from the toxicity of Aβ₄₂. Peptides with a shorter length displayed either weak or no inhibitory effect on Aβ₄₂ aggregation and cytotoxicity. These results suggest that the 15-residue cationic amphiphilic peptide reported here may serve as a potential therapeutic candidate for Alzheimer's disease.

A cationic amphiphilic peptide effectively prevents the aggregation of soluble Aβ₄₂ and also disintegrates matured fibrils into soluble precursors. In addition, the peptide also rescues cells from the toxicity of Aβ₄₂.

Introduction

The aggregation of intrinsically disordered polypeptides and proteins into ordered intractable β-amyloid fibrils is a characteristic feature in many neurodegenerative disorders.^1–3 Alzheimer's disease is the most pervasive neurodegenerative disease.^4,5 It accounts for nearly 60–80% of cases of dementia. According to the World Alzheimer's report more than 75 million people will be suffering from this disease by 2030 and 135 million by 2050.⁶ It will be causing a huge social and economic burden on society. Alzheimer's disease is characterized by the aberrant accumulation of Aβ peptide plaques outside neurons in the brain.⁷ β-amyloid peptides are fragments of transmembrane amyloid precursor protein (APP) obtained after proteolysis by β- and γ-secretases inside the transmembrane in neuronal cells.⁸ Among many fragments of APP, peptides Aβ₄₀ and Aβ₄₂ are the most pathologically important species.⁹ Studies suggested that Aβ₄₀ is the most abundant among the fragments and Aβ₄₂ is more prone to aggregation and more toxic.^10,11 Currently, there is no treatment available to cure Alzheimer's disease, however, present medications temporarily improve or slow the progression of the disease. Being a progressive disorder, the effect of these drugs wears off with time. One of the attractive strategies in Alzheimer's treatment is preventing the Aβ-peptides from clumping into plaques or the removal of plaques of Aβ-peptides from the brain.

A panoply of molecules including synthetic peptides derived from the amyloid core region,^12–15 small molecules,^16–19 antibodies,^20–22 nanobodies,²³ affibodies,²⁴ antimicrobial peptides,^25–27 cyclic peptides,^28–30 β-hairpins,^31,32 β-sheet mimetics,^33,34 peptide–polymer conjugates,³⁵ cell permeable peptides,³⁶ foldamers,^37–41 molecular chaperones,⁴² and N-methylated peptides⁴³ have been investigated as aggregation inhibitors of Aβ peptides. Previously, we reported the inhibition of amyloid aggregation by a designed β-hairpin composed of cationic residues.³¹ In this article, we report the prevention of aggregation of Aβ₄₂, disruption of matured Aβ₄₂ peptide plaques into nontoxic species and protection of cells from the toxicity of Aβ₄₂ by a designed 15-residue cationic amphiphilic peptide. At the same time, shorter cationic 8- and 12-residue peptides showed a moderate inhibitory effect on Aβ₄₂ aggregation. Further, no inhibitory effect on Aβ₄₂ amyloid aggregation was observed after replacing Lys residues with Ala residues in the 15-residue peptide.

Results and discussion

Design of and repurposing amphiphilic cationic peptides

We have been interested in the design of antimicrobial cationic amphiphilic peptides. We designed and synthesized peptide P1 to examine its antibacterial activity. The sequence of the peptide is shown in Scheme 1. The analysis revealed that P1 displayed a very weak antimicrobial activity against the bacterial strains E. coli ATCC 25922, S. aureus ATCC 29213, K. pneumoniae BAA 1705, A. baumannii BAA 1605, and P. aeruginosa ATCC 27853. As this peptide displayed a weak antimicrobial activity, we sought to examine whether or not it can be repurposed as a possible Aβ₄₂ aggregation inhibitor. In addition, truncated versions of P1, peptides P2–P4 with complete replacement of Lys residues with Ala, were designed (Scheme 1) and synthesized using solid phase peptide synthesis, purified by RP-HPLC and confirmed by mass spectrometry analysis.

Scheme 1. Sequences of Aβ₄₂ and the peptides.

The monomeric form of the Aβ₄₂ peptide adopts a random coil in aqueous solution, however, upon incubation, it transformed into a β-sheet conformation and eventually led to amyloid fibrils and plaques.^44,45 Peptides P1–P3 possessed net charges of +6, +5 and +3, respectively, at pH 7.0. In comparison with the peptides, Aβ₄₂ had a net charge of −2.7 at neutral pH. The net negative charge of Aβ₄₂ indicates that it can attract and bind to the cationic peptides.

Effect of cationic peptides on Aβ₄₂ amyloid aggregation

We examined the effect of the cationic peptide P1 on Aβ₄₂ aggregation using the thioflavin T (ThT)-fluorescence based assay.^46,47 The free-aggregated soluble form of Aβ₄₂ was prepared using a reported procedure.³¹ The cationic peptide P1 was incubated at different concentrations starting from 0.5 to 10 equivalents with a constant 1 equivalent of soluble Aβ₄₂. The aggregation of Aβ₄₂ in the absence and presence of an inhibitor was quantified using the maximum ThT fluorescence intensity (I_max) against time. The effect of P1 on the aggregation of soluble Aβ₄₂ is shown in Fig. 1a and b. In the absence of an aggregation inhibitor peptide, Aβ₄₂ reaches the maximum fluorescence intensity with a t₅₀ value of 9.2 hours. In comparison with Aβ₄₂ alone, at 0.5 equivalents of P1 with respect to the Aβ₄₂ concentration (P1: Aβ₄₂; 0.5 : 1), the fluorescence intensity was reduced to 20%. Instructively, the ThT fluorescence intensity was diminished to <∼3% at 1 : 1 concentrations of P1 and Aβ₄₂. These results indicate that at 1 : 1 concentrations, P1 effectively prevents the aggregation of Aβ₄₂. The effective inhibition of Aβ₄₂ aggregation was also observed at 1 : 3, 1 : 5 and 1 : 10 concentrations of Aβ₄₂ and P1, respectively.

Fig. 1. Kinetic study of the fibrillation of Aβ₄₂ in the absence and presence of peptides P1–P4. The time dependent aggregation of Aβ₄₂ in the absence and presence of varying concentrations of peptides was monitored using ThT fluorescence. The concentrations of Aβ₄₂ and ThT were 25 μM and 20 μM, respectively, in all experiments. The concentrations of the peptides ranged from 12.5 μM, 25 μM, 75 μM, 125 μM to 250 μM. The ThT mediated fluorescence based time dependent Aβ₄₂ aggregation kinetic profile in the presence of (a) P1, (b) P1 along with error bars, (c) P2, (d) P3, and (e) P4. (f) The comparative aggregation inhibitory profiles of P1 to P3 at a 1 : 1 concentration ratio (25 μM each) with respect to Aβ₄₂.

Based on the exciting results from P1, we further synthesized peptides P2 and P3, the shorter analogues of P1, and examined their effects on the aggregation of Aβ₄₂. Different concentrations of P2 and P3 were incubated with Aβ₄₂ and the ThT fluorescence was examined. Both peptides inhibited Aβ₄₂ aggregation in a concentration dependent manner. The results are shown in Fig. 1c and d, respectively. In comparison with P1, the dodecamer P2 and octamer P3 required higher concentrations to inhibit Aβ₄₂ aggregation. At a 1 : 1 concentration ratio of P2 to Aβ₄₂, we still get a good fluorescence intensity of ThT and it decreases with a further increase in concentration of P2. In addition, the octapeptide P3 with a net charge of +3 also inhibits the aggregation in a concentration dependent manner similar to P2, however, it requires higher concentrations to completely inhibit the Aβ₄₂ aggregation. Further, to understand the importance of cationic residues in the inhibition of Aβ₄₂ aggregation, we synthesized peptide P4 by replacing Lys residues with Ala residues. Peptide P4 with a zero net charge was examined for its effect on Aβ₄₂ aggregation under identical conditions. The peptide did not inhibit Aβ₄₂ aggregation (Fig. 1e), however, it showed a delay in the initial association in a concentration independent manner (see the ESI†). A comparative profile of the aggregation inhibition activities of peptides P1–P3 at a 1 : 1 concentration ratio with respect to Aβ₄₂ is shown in Fig. 1f. The kinetic parameters of the ThT assay are given in the ESI.† The aggregation rate constant, half-time (t_½), and lag phase were extracted from the ThT assay.^48,49 At a 1 : 1 molar ratio of Aβ₄₂ to P1, complete prevention of aggregation of Aβ₄₂ was observed. In the absence of a peptide inhibitor, the aggregation of Aβ₄₂ showed a t_½ value of 9.34 ± 0.25 h and a lag time of 6.35 ± 0.32 h. At a 1 : 0.5 molar ratio of Aβ₄₂ to P1, the t_1/2 and lag time values were reduced to 6.09 ± 0.39 h and 1.8 ± 0.69 h, respectively. The reduction in the t_1/2 value and lag phase suggested the onset of prevention of aggregation of Aβ₄₂. Similar results were also observed in the case of P2 and P3. At a 1 : 1 molar ratio of P2 to Aβ₄₂, we observed t_1/2 to be 9.59 ± 0.37 h and the lag time to be 0.75 ± 0.46 h. In the case of P3, at a 1 : 1 molar ratio with Aβ₄₂, the t_1/2 and lag time values were reduced to 8.22 ± 0.30 h and 2.14 ± 0.32 h, respectively. These results suggested that the peptides prevented aggregation by interfering with the inception of Aβ₄₂ aggregation. The ThT fluorescence based amyloid aggregation kinetic analysis clearly suggested that P1 is a better Aβ₄₂ amyloid aggregation inhibitor compared to P2–P4.

In order to understand whether or not P1 can inhibit the aggregation at lower concentrations, we incubated Aβ₄₂ with 0.1 and 0.3 equivalents of P1. The ThT fluorescence intensity was found to be 40% and 30% for 0.1 and 0.3 concentrations of P1, respectively. Even at these lower concentrations, we observed a drastic diminution in the ThT fluorescence intensity, indicating that P1 inhibits the Aβ-aggregation even at very low concentrations. The results are shown in Fig. 2a. Moreover, the fluorescence intensity decreases with increasing peptide concentration, suggesting that the peptide inhibits the amyloid aggregation in a concentration dependent manner and almost complete inhibition was observed at 1 : 1 concentrations of the peptide and Aβ₄₂.

Fig. 2. Kinetic study of the fibrillation of Aβ₄₂ with P1. (a) The normalized kinetic profile of the self-assembly of 25 μM Aβ₄₂ in the absence and presence of P1 with stoichiometric ratios (2.5 μM, 7.5 μM, 12.5 μM, and 25 μM). (b) Kinetic study of the fibrillation of Aβ₄₂ with peptide P1 with an equimolar concentration (25 μM) at different time points. (i) Incubation of Aβ₄₂ alone (red line), (ii) addition of P1 to the solution of Aβ₄₂ at 0 hours of incubation (green line), (iii) addition of P1 after 16 hours of Aβ₄₂ incubation (black line) and (iv) addition of P1 to the solution of Aβ₄₂ after 24 hours of incubation (blue line). Buffer conditions: 10 mM sodium phosphate, 150 mM NaCl, 1 mM EDTA and pH 7.4. The ThT concentration was 20 μM.

As P1 prevents the aggregation of soluble Aβ₄₂ into fibrils, we further sought to investigate whether or not P1 can disrupt the aggregated species of Aβ₄₂ into soluble forms using the ThT-based fluorescence assay. Soluble Aβ₄₂ was incubated along with ThT at 37 °C. Peptide P1 was added to the incubated samples of Aβ₄₂ after 16 hours and 24 hours, the incubation was continued and the ThT fluorescence intensity was monitored. The concentration of P1 was equivalent to the concentration of Aβ₄₂ (1 : 1). The results are shown in Fig. 2b. Instructively, both experiments revealed a gradual diminution of the ThT fluorescence intensity with time. The control experiment without the peptide showed no decrease in the fluorescence intensity of ThT. These results suggest that the peptide disrupted the fibrils of Aβ₄₂ in the solution. Disruption of the amyloid fibrils is one of the most crucial aspects in the design of inhibitors for Alzheimer's disease and similar protein aggregation mediated disorders. However, many inhibitors failed to disrupt the aggregated fibrils of Aβ₄₂. Strikingly, these results suggest that peptide P1 not only prevents the aggregation of soluble Aβ₄₂ into fibrils but also disrupts the matured amyloid fibrils into soluble precursors.

Dynamic light scattering (DLS) study

The disruption of Aβ₄₂ fibrils by P1 in ThT experiments encouraged us to perform dynamic light scattering (DLS) experiments, which measure the size of aggregated particles in the solution. We examined the aggregation of Aβ₄₂ using a DLS experiment after incubation for 24 hours in the absence and presence of P1. The DLS results are shown in Fig. 3a. At 0 hours of incubation, Aβ₄₂ displayed small sized aggregated particles. After incubation for 24 hours, Aβ₄₂ displayed aggregated particles around 800 nm in size. In the presence of P1, at equal concentrations, the particle size was observed to be around 100 nm. Instructively, the peptide alone also displayed an aggregated particle size of around 80 nm. From these experiments, we observed a clear difference in the aggregation profile of Aβ₄₂ in the absence and the presence of P1. The analysis of the DLS experiments further supported that the peptide prevents the aggregation of Aβ₄₂ in solution.

Fig. 3. Effect of P1 on the self-assembly of Aβ₄₂. (a) Dynamic light scattering (DLS) study of 10 μM Aβ₄₂ in the absence and presence of P1 after 24 hours of incubation at 37 °C. The frequency distribution was plotted against the diameter of Aβ₄₂ aggregation. (b) HR-TEM images of Aβ₄₂ (10 μM) after incubation at 37 °C for about 48 hours (i) in the absence of P1 and (ii) in the presence of P1. (c) AFM images of Aβ₄₂ (10 μM) after incubation at 37 °C for about 48 hours (i) in the absence of P1 and (ii) in the presence of 10 μM P1.

In addition, we subjected the Aβ₄₂ samples with ThT to confocal microscopy analysis in the absence and presence of peptide P1. The confocal images are shown in the ESI.† Analysis of the confocal images revealed Aβ₄₂ fibrils with green fluorescence in the absence of the peptide. No fibrils were observed in the presence of P1 (see the ESI†). These observations further support the results obtained in the ThT mediated kinetic analysis and DLS experiments in solution.

Transmission electron microscopy (HR-TEM) and atomic force microscopy (AFM) analysis

Inspired by the disruption of Aβ₄₂ aggregation in solution and also from the confocal experiments, we examined the morphology of Aβ₄₂ in the absence and presence of peptide P1 using HR-TEM and AFM. Before subjecting it to HR-TEM experiments, soluble Aβ₄₂ was incubated at 37 °C in the absence and presence of peptide P1 at a 1 : 1 concentration ratio. The TEM images are shown in Fig. 3b. In the absence of P1, Aβ₄₂ displayed an entangled network of self-assembled fibrils. Fascinatingly, in the presence of P1, no fibrillar network of Aβ₄₂ was observed. These results further support the fact that peptide P1 prevents the aggregation of Aβ₄₂. Similar results were also observed in AFM analysis. The fibrils of Aβ₄₂ observed in the absence of the peptide disappeared in the presence of P1. The AFM images are shown in Fig. 3c.

Enzyme-linked immunosorbent assay (ELISA)

We further carried out an ELISA assay to understand the Aβ₄₂ aggregation kinetics in the absence and presence of the inhibitor P1 using a commercially available ELISA kit. A solution of Aβ₄₂ in PBS buffer was incubated with increasing concentrations of peptide P1 for 24 hours before starting the ELISA experiment. The samples of Aβ₄₂ in the absence and presence of P1 were incubated in a 96 well microplate pre-coated with an Aβ oligomer-specific monoclonal antibody. The bound oligomers of Aβ₄₂ were detected by the biotin conjugated detection antibody followed by the binding of streptavidin conjugated-HRP to biotin. A TMB substrate was used to develop the colour. The absorbance of the samples was measured at 450 nm. The results are shown in Fig. 4a. Strikingly, the wells containing Aβ₄₂ alone showed increased absorbance at 450 nm compared to the sample wells containing Aβ₄₂ along with P1 and the peptide alone. Even at 1 : 0.5 equivalent concentrations of Aβ₄₂ and P1, respectively, a substantial decrease in the absorbance value was observed, suggesting that P1 prevents the aggregation of Aβ₄₂ which cannot be recognized by the antibody. Similar results were also observed for 1, 3, 5 and 10 equivalents of P1 along with Aβ₄₂. The results observed in the ELISA experiments were consistent with the ThT assay, DLS, confocal microscopy and TEM analysis.

Fig. 4. (a) Detection of aggregated Aβ₄₂ fibrils using ELISA in the absence and presence of P1. A substantial decrease in absorbance was observed in the presence of the peptide at concentrations of 0.5 equivalents and above. The peptide also showed weak absorbance under identical conditions. (b) The CD analysis of Aβ₄₂ samples in the absence and presence of peptide P1. Soluble Aβ₄₂ (50 μM) at 0 hours of incubation (blue line), after 24 hours of incubation (red line), and in the presence of an equimolar concentration of P1 (50 μM) after 24 hours of incubation (green line) and peptide P1 alone (black line) are shown.

Circular dichroism analysis

The hindsight of the prevention of aggregation of Aβ₄₂ and the disruption of aggregated Aβ₄₂ fibrils by P1 inspired us to subject them to CD analysis to understand their molecular conformations. The secondary structure contents of the peptide and Aβ₄₂ can be measured using quantitative CD analysis. The results are shown in Fig. 4b.

Aβ₄₂ at 0 hours of incubation without the peptide in PBS showed no regular secondary structures. Similarly, P1 also showed no regular secondary structure signature in the PBS buffer. The CD spectrum of Aβ₄₂ after 24 hours of incubation at room temperature showed the CD minimum at around 220 nm, suggesting the transformation of the random coil into a β-sheet type structure, while the peptide showed no change in its CD pattern even after 24 hours of incubation in the PBS buffer. Instructively, the CD spectrum of a 1 : 1 combination of P1 and Aβ₄₂ after incubating for 24 hours showed a CD spectrum corresponding to a helical secondary structure with CD minima at 205 nm and 221 nm. However, we did not observe a CD maximum at around 200 nm in both β-sheet type and helical type spectra. Based on these results, we speculate that P1 transformed the random coil and/or β-sheet structure of Aβ₄₂ into a helix type structure. This type of transformation of a β-sheet into a helix type structure of Aβ₄₂ was also observed with other aggregation preventing inhibitors.³⁶

NMR analysis

Far-UV CD analysis suggested that there is a conformational rearrangement in Aβ₄₂ upon incubation with P1 at a 1 : 1 ratio. To further probe the structural reorganization within Aβ₄₂ upon interaction with P1, we subjected them to ¹H–¹⁵N HSQC NMR on an 800 MHz spectrometer attached with a cryo probe. This experiment was performed using ¹⁵N labelled Aβ₄₂ at 283 K in 10 mM sodium phosphate buffer, pH 7.4. The amide NH region of the HSQC spectrum of Aβ₄₂ is shown in Fig. 5a. All cross peaks were identified with the help of previously assigned spectra from the literature.^38,39,50 Further, the same HSQC experiment was performed on a sample containing a 1 : 1 mixture of Aβ₄₂ and peptide P1. We observed changes in the positions of the cross-peaks of Aβ₄₂ in the presence of P1. To understand the change in the position of the cross peaks, we overlay the amide NH region of the HSQC spectra of Aβ₄₂ in the absence and presence of peptide P1 which is shown in Fig. 5b.

Fig. 5. The 2D ¹H-¹⁵N HSQC NMR spectrum of a) ¹⁵N-isotopically labelled Aβ₄₂ (40 μM) in the absence of P1. b) Overlay of 2D ¹H–¹⁵N HSQC NMR spectra of ¹⁵N-isotopically labelled Aβ₄₂ (40 μM) in the absence of P1 (blue contours) and in the presence of 80 μM P1 (red contours). Each peak represents the correlation between the nitrogen and the amide proton of individual amino acid residues of Aβ₄₂. c) The changes in the chemical shift of the ¹H resonance of Aβ₄₂ (40 μM) induced by P1 (80 μM) at a stoichiometric ratio of 1 : 2 (Aβ₄₂: P1). The HSQC NMR experiments were conducted using an 800 MHz spectrometer attached with a cryo probe at 283 K in 10 mM sodium phosphate buffer, pH 7.4.

The 2D NMR analysis suggested a clear chemical shift perturbation in many of the residues of Aβ₄₂, however, a dominant effect was observed particularly at the N-terminal residues 3E, 4F, 5R, 7D, 11E and 13H. In addition, the perturbation was also observed for the residues 15Q, 20F, 27 N and 36 V (Fig. 5c). Except for 5R, no strong perturbation was observed for other cationic residues in the sequence. Based on these results, we speculated that the peptide may bind to the N-terminal region and induces conformational reorganization in Aβ₄₂. Among the six negatively charged residues, three residues are strongly involved in the interaction with P1.

In order to understand the structure of P1 in solution, we recorded the ¹H NMR spectrum of an aqueous solution consisting of 5% D₂O. The peptide displayed a poorly dispersed spectrum in the aqueous solution suggesting a lack of a defined secondary structure in aqueous solution. However, in contrast, the peptide displayed a well dispersed ¹H NMR spectrum in CD₃OH. The peptide was subjected to 2D-NMR (TOCSY and ROSEY) in CD₃OH and the structure was calculated using unambiguous NOEs. The peptide adopted a helical conformation in CD₃OH. Further, the CD analysis of the peptide also suggested that the helical conformation in MeOH, however, showed a spectrum similar to that of the random coil type in aqueous solution. The results suggested that the peptide adopted a helical structure in methanol, however, it displayed no defined secondary structure in aqueous solution.

Cell toxicity

The soluble aggregates of Aβ₄₂ are known to be toxic to neuronal cells. As P1 has been shown to prevent and disrupt the aggregation of Aβ₄₂, we examined the toxicity of Aβ₄₂ in the presence of P1 using the MTT (3,(4,5-dimethylthiazol-2-yl)2,5-diphenyltetrazolium bromide) reduction assay on SH-SY5Y human neuroblastoma cell lines. Along with Aβ₄₂, the toxicity of P1 was also examined. The cytotoxicity can be measured based on the ability of live cells to convert MTT into insoluble formazan crystals. Consequently, the quantity of formazan produced in the assay is proportional to the number of live cells.

We used 10 μM Aβ₄₂ as a negative control and cells without Aβ₄₂ as a positive control. Neuronal cells with different concentrations of P1 ranging from 1 μM to 30 μM were incubated at 37 °C along with 10 μM of Aβ₄₂. The results are shown in Fig. 6. Peptide P1 displayed no toxicity to the cells up to a 30 μM concentration (Fig. 6a). Soluble Aβ₄₂ exhibited toxicity to the neuronal cells. Strikingly, with increasing concentrations of P1, Aβ₄₂ showed reduced toxicity to the neuronal cells. No toxicity of Aβ₄₂ was observed at a 1 : 1 ratio of peptide to Aβ₄₂ (Fig. 6c). These fascinating results indicate that P1 rescued the neuronal cells from the toxicity of Aβ₄₂. These results suggested that P1 inhibited the toxicity of 10 μM Aβ₄₂ with an IC₅₀ value of 8.73 μM (Fig. 6d). Peptides P2 and P3 were also non-toxic to the cells at 30 μM concentrations. A comparison of the activities of P3–P1 at a 1 : 1 concentration ratio with respect to Aβ₄₂ is shown in Fig. 6b. In comparison, P1 showed a more notable activity in rescuing the cells from the toxicity of Aβ₄₂ than the other two peptides. In addition to the cytotoxicity experiments, we also examined the haemolytic activity of P1. Red blood cells were treated with different concentrations of P1. The peptide displayed no haemolytic activity when tested up to a 250 μM concentration. These results suggested that P1 is non-toxic to neuronal cells and also non-haemolytic. Overall, these results suggested that P1 prevents the aggregation of Aβ₄₂ by stabilizing its helical conformation and this conformation is non-toxic to neuronal cells. These results strongly justify that P1 is a potent candidate for the therapeutic intervention of Alzheimer's disease.

Fig. 6. Cell viability of SHSY5Y cells with Aβ₄₂ (10 μM) in the absence and presence of peptides (P1–P3). The cells were incubated at 37 °C up to 24 hours. (a) Treatment of cells at different concentrations of peptides alone. (b) Treatment of cells with Aβ₄₂ in the absence and presence (1 : 1) of peptides P1–P3. The cell viability was compared to the control cells without any peptide or Aβ₄₂. (c) Survival of cells with increasing concentrations of P1 (1 μM, 2 μM, 3 μM, 5 μM, 10 μM and 30 μM) at a constant 10 μM of Aβ₄₂ and (d) dose dependent effect of P1 on Aβ₄₂ induced toxicity in SHSY5Y cells.

There is a large body of evidence emerging from recent studies suggesting that cationic peptides including host defence peptides inhibit the aggregation of β-amyloid peptides.^24–26 However, the toxicity of these peptides hinders their further progress as inhibitors for Alzheimer's disease. The results reported here suggested that peptide P1 displayed no toxicity to cells up to 30 μM and also showed non-haemolytic activity up to a 250 μM concentration (tested). Further, P1 showed no antibacterial activity against E. coli ATCC 25922, S. aureus ATCC 29213, K. pneumoniae BAA 1705, A. baumannii BAA 1605, and P. aeruginosa ATCC 27853 up to 64 μM (tested).

The absence of the elongation phase in the ThT fluorescence kinetic assay, the potent suppression of Aβ₄₂ aggregation by the inhibitor peptide in solution by DLS experiments, the absence of the fibrils of Aβ₄₂ in solution by confocal imaging, nonappearance of Aβ₄₂ fibrils in TEM and AFM, the absence of the aggregation prone β-sheet structure in the CD analysis and structural re-organization Aβ₄₂ in the presence of P1 suggested P1 trapping Aβ₄₂ in a non-toxic monomer form or appropriating oligomers that are nontoxic to cells. The cell toxicity experiments indeed revealed a reduced toxicity of Aβ₄₂ in the presence of an equimolar concentration of P1. Based on these results, we propose that peptide P1 prevents the aggregation of soluble Aβ₄₂ and also disintegrates aggregated Aβ₄₂ fibrils into soluble nontoxic precursors. We speculate that upon interaction P1 induces a conformational change in Aβ₄₂ and this new conformation is not prone to aggregation and also non-toxic to the cells. A cartoon representation of the mechanism of interaction of P1 with Aβ₄₂ is shown in Fig. 7.

Fig. 7. A cartoon representation of the mechanism of interaction of P1 with Aβ₄₂.

Conclusion

In conclusion, we report that a 15-residue cationic amphiphilic peptide not only prevents the aggregation of soluble Aβ₄₂ into fibrils but also disintegrates matured fibrils into soluble precursors. This peptide also rescues neuronal cells by alleviating the toxicity of Aβ₄₂. Inspiringly, the results suggested that this peptide is non-toxic and also non-haemolytic in nature. The shorter versions of the peptide also prevent the aggregation of soluble Aβ₄₂ but at higher concentrations. The experimental evidence suggested that upon interaction, the peptide transforms a random coil of soluble Aβ₄₂ and the beta-sheet structure of aggregated or matured Aβ₄₂ into a helical type conformation. We speculated that the peptide traps this helical type conformation which is not toxic to the neuronal cells. Overall, the peptide transformed aggregated fibrils of Aβ₄₂ into soluble precursors and rescued the cells from the toxicity of Aβ₄₂. These results strongly justify that P1 may serve as one of the potent therapeutic candidates for Alzheimer's disease.

Data availability

Details of the experimental methods, procedures, and additional experimental data are provided in the ESI.†

Author contributions

H. N. G., and D. R. P. designed the project and wrote the manuscript. D. R. P. carried out synthesis, purification and most of the experimental work including the biophysical studies. R. M. R. and S. A. N. assisted D. R. P. in the data analysis, M. S. assisted in the preparation of a few starting materials. D. K. S. and G. G. carried out the NMR experiments. S. R. assisted D. K. S. and G. G. in the NMR analysis part. G. K. and A. D. completed the MIC measurements. S. C. assisted G. K. and A. D. in the analysis of the MIC data.

Conflicts of interest

There are no conflicts to declare.