Effects of Charged Polyelectrolytes on Amyloid Fibril Formation of a Tau Fragment

Abstract

The microtubule associated protein tau is involved in more than 20 different neurological disorders characterized by aberrant intracellular aggregation of tau in the brain. Here, we investigated the aggregation of a novel 20-residue model peptide, tau_298–317, which is derived from the key microtubule binding domain of the full sequence tau. Our results show that tau_298–317 highly mimics the physical and aggregation properties of tau. Under normal physiological conditions, the peptide maintains a disordered random coil without aggregation. The presence of polyanionic heparin (Hep) significantly promotes the aggregation of this peptide to form amyloid fibrils. The Hep-induced aggregation is sensitive to the ionic strength of the solution and the introduction of the negatively charged phosphate group on a serine (Ser) residue in the sequence, suggesting an important role of electrostatic interactions in the mechanism of Hep-mediated aggregation. In addition, two positively charged polysaccharides, chitosan (CHT) and its quaternary derivative N-trimethyl chitosan (TMC), were found to effectively inhibit Hep-induced aggregation of tau_298–317 in a concentration dependent manner. Attractive electrostatic interactions between the positively charged moieties in CHT/TMC and the negatively charged residues of Hep play a critical role in inhibiting Hep-peptide interactions and suppressing peptide aggregation. Our results suggest that positively charged polyelectrolytes with optimized charged groups and charge distribution patterns can serve as effective molecular candidates to block tau-Hep interactions and prevent aggregation of tau induced by Hep and other polyanions.

Graphical Abstract

INTRODUCTION

Microtubule associated protein tau promotes tubulin self-assembly into microtubule polymers and stabilizes the axonal transport system.^1,2 It is also a key component of various enzymatic pathways such as tau-dependent postsynaptic localization of the Fyn kinase.³ In the past decades, tau has received special attention due to its correlation with more than twenty different neurodegenerative diseases, which are collectively known as tauopathies.^4,5 Abnormal tau accumulation in the brain is a characteristic hallmark of tauopathies.⁵ For instance, in Alzheimer’s disease (AD), intracellular tau aggregation generates neurofibrillary tangles (NFTs), whereas extracellular amyloid-β (Aβ) aggregation forms senile plaques.⁶ Tau aggregates have been indicated to be closely related to the pathology of tauopathies.⁶ Understanding the mechanistic details of tau aggregation is critical for unraveling the underlying pathology of tauopathies and developing effective strategies to inhibit tau aggregation for the treatment of these devastating diseases.

Among six different tau isoforms expressed in human, the longest one (2N4R) contains 441 residues.⁷ The residual sequence of 2N4R (Figure 1) can be divided into several regions, where the N-terminus (with two inserts, N1 and N2) is followed by the proline-rich region (P1 and P2), the microtubule binding region composed of four pseudo-repeats (R1–R4), and a shorter C-terminal tail.^6,8,9 Tau is an highly soluble protein with a small number of hydrophobic residues in the sequence, and the molecule has a tendency to fluctuate among different conformations without having specific stable tertiary structure, which makes tau an intrinsically disordered protein (IDP).^6,8,9 In order to aggregate into fibrils, tau usually requires additional external factors or posttranslational modifications that convert it to a pro-aggregant.^10–12 Charged anionic co-factors, such as sulfated glycosaminoglycans, nucleic acids, and fatty acids, have been found to induce aggregation of tau, presumably by assisting in overcoming the nucleation barrier on the aggregation pathway.^10–13 For example, the co-localization of NFTs and heparan sulfate, a highly acidic glycosaminoglycan (GAG) modified by sulfate, in the isolated brain samples of AD patients suggests an important role of this polyanionic co-factor in tau fibrillation in vivo.¹⁴ It is suggested that the polyanionic molecule can interact electrostatically with positively charged residues in the repeat regions of tau, thereby reducing electrostatic repulsions of the repeat regions, alleviating tau-tau interaction, and stabilizing the β-conformation.^6,10,15 Given the importance of electrostatic interactions in the interactions of anionic polymers and tau, as well as polyanion-mediated self-association and aggregation of tau, one would hypothesize that polyelectrolytes containing positive charges may play an opposite role and interfere with tau aggregation, although few such studies have been reported so far.

Figure 1.

Schematic representation of the primary sequence of full-length tau (2N4R) and the sequence of the selected fragment peptide tau_298–317. The structural and functional features of the local regions of the fragment peptide are depicted.

In this work, we investigated the effect of charged polysaccharides on the aggregation of a model tau fragment peptide and evaluated the role of electrostatic interactions in polyelectrolyte-mediated peptide aggregation. Peptide fragments from key regions of amyloidogenic proteins, such as islet amyloid polypeptide (IAPP), Aβ, and prion, have been widely used as useful models for studying the mechanism of full-sequence protein aggregation.^16–18 These approaches greatly reduce the complexity of molecular systems for studying key factors and details of the underlying mechanisms of protein aggregation. Here, a 20-residue peptide fragment derived from tau 2N4R (residues 298–317, referred to as tau_298–317 in this study) was used as a model molecule to mimic the aggregation properties of tau. The peptide sequence is located in the microtubule binding repeating domains of tau and contains the ³⁰⁶VQIVYK³¹¹ local fragment (Figure 1), a key hexapeptide motif capable of triggering fibrillation of tau by forming β-sheet structures.¹⁹ The N-terminus of the selected sequence (Figure 1) starts with a positively charged Lys₂₉₈-His₂₉₉ patch, which were reported to have a strong propensity to bind with negatively charged polyelectrolytes.¹⁵ A positively charged Lys₃₁₁ residue in the middle region of the sequence may also favor binding to polyanions.¹⁵ On the other hand, the ³⁰¹PGGG³⁰⁴ motif in the sequence was found to have a strong tendency to form turns under normal physiological conditions.¹⁵ Moreover, it has also been suggested that the Ser₃₀₅-Asp₃₁₄ local region shows a pronounced propensity to form β-structures.¹⁵ Here, we evaluated the aggregation properties of this novel tau fragment peptide and investigated the effect of polyelectrolytes with different charge properties, i.e., negatively charged heparin (Hep), positively charged chitosan (CHT) and N,N,N-trimethyl chitosan (TMC), on the aggregation of the peptide. Our results show that the peptide closely mimics the aggregation properties of the full sequence tau. The results also suggest that electrostatic interactions play a key role in the polyelectrolyte-mediated amyloid formation of the peptide. Given the importance of the selected fragment in the tau sequence, this study provides valuable information on the effect of polyelectrolytes on aggregation in key regions of tau and may help in the design of novel charged polymer structures to inhibit tau self-association and formation of toxic aggregates.

RESULTS AND DISCUSSION

Characterization of the aggregation properties of tau_298–317.

Tau is an intrinsically disordered protein and is highly soluble in an aqueous solution.⁸ The fragment peptide tau_298–317 prepared in this study contains a number of charged (e.g., Lys, Glu) and polar (e.g., Ser, Gln, Tyr) amino acid residues in the primary sequence (Figure 1) and is also dissolved well in a physiological buffer (50 mM Na-phosphate, pH 7.4), mimicking the molecular properties of the full sequence tau. To study the aggregation properties of the peptide, we examined the fibrillation kinetics of this tau fragment using fluorescence of thioflavin T (ThT). ThT can selectively bind to the fibrillar structure of amyloids and show a significantly enhanced fluorescence intensity in the vicinity of 480 nm. As shown in Figure 2A, there is no notable ThT fluorescence increase over the time course of the experiment for tau_298–317 in a concentration range of 12–200 μM. Increasing the concentration to 400 μM still shows a weak ThT fluorescence change over time. No significant increase of ThT fluorescence was observed either for the studies carried out at different temperature, pH, or with addition of NaCl salt in the buffer (Figure 2B). These results suggest that this tau fragment, like the full sequence tau, does not tend to aggregate under normal in vitro conditions without an effective external aggregation promoter. The atomic force microscopy (AFM) imaging results (Figure 2C) also show that there are no oligomers or fibrils formed in the tau_298–317 sample (12.5 μM) collected at the end of the kinetics study, consistent with the ThT fluorescence results.

Figure 2.

Characterization of the aggregation of tau_298–317. (A) Concentration dependent aggregation kinetics of the peptide at 25 ^oC in pH 7.4 buffer (50 mM Na-phosphate) followed by ThT fluorescence. Data are shown as average of three independent kinetics results. (B) Normalized ThT fluorescence intensity at 40 hrs of the aggregation kinetics of the peptide (12.5 μM) in phosphate buffer (PB), phosphate buffered saline (PBS), acetic buffer, and tris buffer. The concentration of buffers was 50 mM. (C) Tapping mode AFM image of the tau fragment sample (12.5 μM) collected at the end of the aggregation kinetics experiment. (D) CD spectra of the peptide (50 μM) in pH 7.4 buffer (20 mM Na-phosphate). The sample was incubated at room temperature and the spectra were measured at different incubation time points.

The secondary structure of tau_298–317 in aqueous solution (20 mM Na-phosphate, pH 7.4) was studied using circular dichroism (CD) spectroscopy. As shown in Figure 2D, the CD spectra of the tau fragment measured at different incubation time points show a minimum at ~198 nm, indicating a highly random coil conformation.^20,21 The content of specific secondary structures was further estimated by analyzing the CD data using the BeStSel algorithm.²² The results suggest that tau_298–317 contains more than 53% random coil structures (Table S1). Interestingly, The CD analysis results indicate that this tau fragment peptide also contains over 24% antiparallel β-sheets in its native structure. Previous studies on K18 and K19 (truncated forms of full-length tau consisting of only the four and three repeats, respectively) have also reported β-structure in soluble repeat fragments of tau.¹⁵ In addition, tau_298–317 contains ~10–15% turn structures, which could be formed by the ³⁰¹PGGG³⁰⁴ motif commonly seen in β-turns.¹⁵ Taken together, the selected tau fragment peptide, like the full-length tau,²³ contains mainly random coil conformational structures and is not prone to spontaneous aggregation under normal physiological conditions.

Amyloid fibril formation of tau_298–317 induced by heparin (Hep).

Being an intrinsically disordered and highly soluble protein, tau usually requires external co-factors for initiating its aggregation. In vitro, negatively charged co-factor molecules, such as heparin, fatty acids, carboxylated microspheres, or RNA, have been found to drive tau aggregation.^10–13 In particular, since the first report on Hep-mediated tau aggregation,¹⁰ Hep has been recognized as a potent promoter of tau aggregation and has been frequently used in studies of tau fibrillation.^11,24–26 Hep is a highly sulfated linear polysaccharide and the sulfate groups attached along the polymer chain result in a high negative charge density (Figure S1). To assess the impact of Hep on aggregation of tau_298–317, concentration dependent aggregation kinetics of tau_298–317 in the presence of 6.3 μM Hep was monitored using ThT fluorescence. In contrast to that without Hep (Figure 3A vs. 2A), 12.5 μM tau_298–317 with Hep steadily aggregates undergoing a typical sigmoidal fibrillation process consisting of an initial lag phase where monomers form oligomeric nuclei, a growth phase marked by the conversion of the peptides into fibrils, and a plateau phase revealing the formation of mature aggregates.²⁷ The ThT fluorescence intensity at the final plateau phase increases proportionally to the concentration of the peptide (Figure S2). The kinetics curves were fitted using Boltzmen distribution (equation 1, see Experimental Methods) to calculate lag time that is required for nucleus formation in aggregation, and t₅₀ which is the time when the fluorescence intensity reaches half of the maximum. Increasing the peptide concentration leads to a faster aggregation kinetics with shorter lag time and t₅₀ (Figure 3B). The kinetics results clearly show that the presence of Hep significantly promotes the aggregation of tau_298–317.

Figure 3.

Effect of Hep on the aggregation of tau_298–317 peptide. (A) Aggregation kinetics of different concentration of the peptide in the presence of 6.3 μM Hep in pH 7.4 buffer (50 mM Na-phosphate) followed by ThT fluorescence. (B) The t₅₀ and lag time of the aggregation kinetics shown in (A). Error bars are represented as the standard errors of triplicate results. (C) Aggregation kinetics of the tau fragment (12.5 μM) in the presence of different concentration of Hep. (D) AFM image of amyloid fibrils of the tau fragment (12.5 μM) with Hep (1.5 μM) collected at the end of the aggregation kinetics experiment. (D) CD spectra of the peptide (50 μM) in the presence of Hep (6.3 μM). The sample was incubated at room temperature in pH 7.4 buffer (20 mM Na-phosphate) and the spectra were measured at different incubation time points.

The effect of different concentrations of Hep on aggregation of the tau fragment (12.5 μM) was also studied. The result shows that 1.5 μM is the lowest concentration of Hep that can effectively induce peptide aggregation (Figure 3C). Characteristic amyloid fibrils were observed in AFM imaging of the sample collected at the end of the kinetics experiment with 1.5 μM Hep (Figure 3D). Increasing the concentration of Hep above 6.3 μM did not further significantly alter the aggregation kinetics properties of the peptide (Figure S2). Taken together, these results demonstrate that Hep is an effective initiator of tau_298–317 aggregation, similar to its role in aggregation of the full sequence tau.

During aggregation to form NFTs in neurodegenerative diseases, tau undergoes a secondary structural change from disordered random coils to β-sheet-rich structures.^28,29 Here, CD spectroscopy was used to investigate secondary structural changes during Hep-mediated tau_298–317 fibrillation. After 24 hr incubation of the peptide with additional Hep, a negative band at ~224 nm and a positive band at ~195 nm appeared (Figure 3E), suggesting the formation of β-sheet structures upon peptide aggregation. Specifically, the negative band at ~224 nm has been indicated to be a characteristic CD spectrum band of paired helical filaments (PHFs) in AD brain and ADseeded tau fibrils. The results here suggest that aggregated tau_298–317 also adopts secondary structures similar to full sequence tau in PHFs. Further CD data analysis estimates that after aggregation, the peptide contains mainly β-sheet (52%) as its major secondary conformation (Table S2). It is plausible that the ³⁰⁶VQIVYK³¹¹ hexapeptide may be the core region in the formation of β-sheet strcutures.¹⁹ Interestingly, in the presence of Hep, the peptide adopts 28% parallel β-sheet structure, which is absent in the native tau fragment without co-incubating with Hep (Table S1). The fragment peptide also contains ~24% antiparallel β-sheet in the presence of Hep. Remarkably, the ³⁰¹PGGG³⁰⁴ motif present in the fragment was reported to form type II β-turns.¹⁵ The Pro residue in this pattern is capable of introducing a kink in the amide bond, while the subsequent Gly residues may reverse the orientation of the peptide chain by 180^o,^30,31 thereby forming antiparallel β-strands. Antiparallel β-sheets may also be formed from intermolecular peptide chain interactions. On the other hand, parallel β-sheets cannot be formed within the same peptide chain and may only form upon alignment of different peptide chains, which may be the reason why parallel β-sheets are not recognized in the absence of Hep.

Charged moieties in polyelectrolytes have been suggested to play an important role in their regulation of the aggregation of amyloidogenic proteins.³² For instance, the O-sulfo groups of Hep were found to be crucial for accelerating the fibrillation of gelsolin protein.³² A study of Zhao et al. showed that 6-O-sulfo group of Hep plays a significant role in the tau-Hep interaction.³³ Full-length tau contains 44 positively charged Lys residues (10% of the total 441 residues). Unfavorable electrostatic repulsion of these residues may prevent the monomers from self-association.³³ Hep may promote tau aggregation by interacting with tau to form tau-heparin complexes and reducing electrostatic repulsion between monomers.³⁴ To further assess the role of electrostatic interactions in Hep-mediated aggregation of tau_298–317, we tested the effect of ionic strength on the aggregation of this peptide. Increasing the NaCl salt concentration to 300 mM suppresses peptide aggregation in a concentration dependent manner (Figure 4A), in accord with previous studies.³⁵ The effect of NaCl salt concentration on tau_298–317 aggregation indicates the sensitivity of peptide and Hep interactions to ionic strength, suggesting a critical role of electrostatic interactions in the interplay between Hep and tau_298–317 and in Hep-induced peptide self-assembly. In addition, we investigated the aggregation of a tau_298–317 mutant containing a phosphorylated Ser₃₀₅ (pS305) residue. Remarkably, this mutation completely revokes Hep-induced aggregation of the peptide (Figure 4B). It is likely that the addition of a negatively charged phosphate group to the sequence induces unfavorable electrostatic repulsion between the peptide and the anionic Hep, thus interfering with the interaction of Hep and the peptide and preventing the activity of Hep to promote peptide aggregation. Therefore, disrupting the electrostatic interactions between tau and Hep may be an effective strategy for developing inhibitors against Hep-induced tau fibrillation.³⁶

Figure 4.

Effect of electrostatic interactions on Hep-induced tau_298–317 aggregation. (A) Effect of NaCl on ThT fluorescence intensity in the aggregation kinetics of the peptide (12.5 μM) in the presence of Hep (1.5 μM). Fluorescence was obtained at the plateau phase of a 40-hr kinetics measurement. Error bars are represented as the standard errors of triplicate results. Statistical analysis was performed by one-way ANOVA, Bonferroni’s post hoc test. (B) Aggregation kinetics of 12.5 μM tau_298–317 or the pS305 mutant in the absence and presence of 1.5 μM Hep.

Inhibition of Hep-induced tau_298–317 aggregation by chitosan (CHT) and trimethyl chitosan (TMC).

CHT is the second most abundant cationic polysaccharide in nature and is a widely studied biopolymer in the field of drug delivery systems.^37–39 It carries positive charge upon protonation of the amino groups (Figure S1). CHT has been previously reported to inhibit fibrillogenesis of Aβ,⁴⁰ whereas its effect on tau aggregation is unknown. The kinetics results show that, unlike Hep, the addition of CHT does not facilitate the aggregation of tau_298–317 (Figure S3). Intriguingly, further aggregation kinetics study shows that the presence of CHT suppresses Hep-mediated tau fragment aggregation in a concentration-dependent manner (Figure 5A). In the presence of 0.01 mg/mL CHT, the maximum ThT fluorescence intensity decreases by more than 30%. Moreover, the addition of CHT slightly prolongs the lag time compared to that of the peptide and Hep only (Figure S4), suggesting that CHT may interfere with the formation of oligomeric nuclei during protein aggregation. The addition of 0.1 mg/mL CHT completely eliminated the fluorescence intensity of ThT (Figure 5A), indicating a strong inhibitory activity of CHT on amyloid fibril formation. The CD spectra show that the addition of CHT inhibits the conformational conversion of the peptide to form β-sheets (Figure 5B). Analysis of the CD spectra reveals that in the presence of CHT, tau_298–317 in the mixture of Hep and the peptide still contains largely disordered structures (~51.8% random coil, Table S1) after 24 hr of incubation. In particular, the characteristic parallel β-sheet structures formed in Hep-induced peptide aggregation disappears in the presence of CHT (Table S1).

Figure 5.

Inhibitory effect of CHT on Hep-induced tau_298–317 aggregation. (A) Effect of different concentration of CHT on the aggregation kinetics of the tau fragment (12.5 μM) in the presence of Hep (1.5 μM). (B) CD spectra of the fragment peptide (50 μM) in the absence or presence of Hep (6.3 μM) and/or CHT (0.1 mg /mL) after 24 hr incubation at room temperature. (C-E) AFM image of the tau fragment (12.5 μM) in the presence of Hep (1.5 μM) (C), tau fragment (12.5 μM) in the presence of Hep (1.5 μM) and CHT (0.1 mg/mL) (D), and Hep (1.5 μM) and CHT (0.1 mg/mL) only (E). (F) Effect of CHT on preformed tau_298–317 amyloids. The tau_298–317 amyloids were formed by incubating the peptide (12.5 μM) with Hep (1.5 μM) for 24 hrs before adding CHT (0.1 mg/mL). Data are shown as average of three independent kinetics experiments. (G) Comparison of the ThT fluorescence intensity at 24 hr (before addition of CHT) and 40 hr in the aggregation kinetics of the peptide (red curve in Figure F). Statistical analysis was performed by a two-tailed t-test.

AFM imaging was further measured to verify the inhibitory activity of CHT on Hep-mediated tau_298–317 aggregation. In the presence of 0.1 mg/mL CHT, no fibrillar aggregates were observed, distinct to that without CHT (Figure 5C vs 5D). Only a few of granular particles were found in the AFM image of the peptide-Hep-CHT mixture (Figure 5D). Interestingly, in the absence of the peptide, similar granular particles were also detected in the Hep-CHT mixture (Figure 5E). One speculation is that these granular structures may arise from the coacervation of the oppositely charged polyelectrolytes.⁴¹ A previous study by Liu et al. reported formation of mesoscopic particles between Hep and CHT due to polyelectrolyte complexation.⁴² The size of the particles was further measured using dynamic light scattering (DLS). The diameter of the mesoscopic particles in the Hep-CHT mixture is ~240 nm, whereas the diameter of the particles formed in the peptide-Hep-CHT mixture is ~421 nm (Table S2). It might be inferred that in the presence of tau_298–317, the particles formed by Hep-CHT can trap the peptide and grow to a larger size, while the trapped peptide is not in a conformation that favors critical nucleation and fibrillogenesis.^43,44 Hep-CHT mesoscopic particles have been reported to capture large BSA protein.⁴² Together, a probable mechanism of the inhibitory activity of CHT observed in the present study can be attributed to the interplay between CHT and Hep via favorable electrostatic interaction which in consequence interferes with the Hep-induced peptide aggregation. It has been suggested that the aggregation of tau takes place on the Hep template, and charged groups such as sulfate at N- and 6-O-positions of Hep are critical for the interaction with tau.⁴⁵ CHT may effectively shield these charged moieties of Hep and prevent peptide attachment to Hep. In addition, the disruptive effect of CHT on preformed amyloids was investigated. CHT (final concentration of 0.1 mg/mL) was added at 24 hr of the aggregation kinetic assay of tau_298–317 with Hep, when the aggregation reached the stationary phase. The addition of CHT resulted in a significant decrease in the ThT fluorescence intensity (Figure 5F, 5G), indicating that CHT is able to break the preformed fibrillar structures of tau_298–317.

The effect of TMC, a quaternary derivative of CHT, on Hep-induced tau_298–317 aggregation was also evaluated to determine the role of positively charged polymers in the regulation of peptide fibrillogenesis.^46,47 Unlike the amine groups in CHT, TMC contains a permanent positive charge on the quaternary nitrogen with a trimethyl substitution (N-trimethyl group) in the repeating units of the sequence (Figure S1). Similar to CHT, TMC itself does not promote the aggregation of tau_298–317 (Figure S3). The inhibitory effect of TMC on Hep-mediated peptide aggregation is verified by the reduction of ThT intensity in a TMC concentration-dependent manner (Figure 6A). In the presence of 0.01 mg/mL TMC, the maximum ThT fluorescence intensity decreases by ~52%, slightly greater than the decrease caused by the addition of the same amount of CHT. ThT fluorescence is completely suppressed by 0.1 mg/mL TMC, showing a potent inhibitory activity on the aggregation. No amyloid fibrils were formed in the presence of 0.1 mg/mL TMC (Figure 6D vs 6C), in accord with the ThT fluorescence results. The CD results show that the addition of TMC to the peptide-Hep mixture prevents conformational conversion of the peptide to form β-sheets and keeps the peptide mainly in randomly coiled conformation (Figure 6B, Table S1). TMC also inhibits Hep-mediated tau fragment aggregation at a higher pH of 8.1 (Figure S5). At higher pH, a reduced lag phase observed in aggregation might be related to higher ionization of Hep. Previous studies have shown that the density of negative charges of Hep becomes higher at higher pH, which may lead to more effective complexation between the peptide and Hep.^48,49 Similar to the CHT-Hep mixture, granular structures with a diameter of about 203 nm were also observed in the TMC-Hep solution (Figure 6E, Table S2), suggesting favorable interactions of these oppositely charged polymers to form the mesoscopic particles. Furthermore, TMC (0.1 mg/mL) also appears to disrupt the preformed amyloid fibrils of the peptide (Figure 6F, 6G). Due to its permanent charge in the chain, TMC would be a good alternative polyelectrolyte for the inhibition of tau aggregation under different pH conditions.

Figure 6.

Inhibitory effect of TMC on Hep-induced tau_298–317 aggregation. (A) Effect of different concentration of TMC on the aggregation kinetics of the tau fragment (12.5 μM) in the presence of Hep (1.5 μM). (B) CD spectra of the fragment peptide (50 μM) in the absence or presence of Hep (6.3 μM) and/or TMC (0.1 mg /mL) after 24 hr incubation at room temperature. (C-E) AFM image of the tau fragment (12.5 μM) in the presence of Hep (1.5 μM) (C), tau fragment (12.5 μM) in the presence of Hep (1.5 μM) and TMC (0.1 mg/mL) (D), and Hep (1.5 μM) and TMC (0.1 mg/mL) only (E). (F) Effect of TMC on preformed tau_298–317 amyloids. The tau_298–317 amyloids were formed by incubating the peptide (12.5 μM) with Hep (1.5 μM) for 24 hrs before adding TMC (0.1 mg/mL). Data are shown as average of three independent kinetic experiments. (G) Comparison of the ThT fluorescence intensity at 24 hr (before addition of TMC) and 40 hr in the aggregation kinetics of the peptide (red curve in Figure F). Statistical analysis was performed by a two-tailed t-test.

Taken together, a hypothesized mechanistic effect of the charged polyelectrolytes tested in this study on tau_298–317 aggregation is depicted in Figure 7. This tau fragment peptide mimics the aggregation properties of tau and requires external cofactors, such as Hep, to overcome the nucleation barrier during aggregation. Hep interacts electrostatically with the positively charged residues of the peptide (e.g., 3 Lys, 1 His, and the N-terminal amine group) to assist in the formation of tau_298–317 nuclei for subsequent growth with additional peptides to form amyloid fibrils. Negatively charged moieties such as the O-sulfo group in Hep very likely play a vital role in the peptide-Hep interaction.^33,45 On the other hand, cationic polyelectrolytes CHT and TMC interact competitively with Hep to form polyelectrolyte complexes.^48,50–52 The locally abundant positive charges distributed along the polysaccharide chain favor electrostatic interaction between the charged amine groups of CHT and the sulfonate (–SO₃⁻ ) groups of Hep, contributing to the formation of cationic-anionic polymer complexes.⁴⁸ A recent study by Bueno et al. also reported that the positively charged trimethyl amine [–N⁺(CH₃)₃] groups on TMC and the sulfonate groups on Hep form favorable electrostatic interactions.⁴⁹ The interactions of CHT/TMC and Hep prevent Hep from interacting with tau_298–317, resulting in the inhibition of oligomeric nucleation and amyloid formation of the peptide. The slightly stronger aggregation inhibition activity of TMC compared to CHT implies the importance of the positive charge density along the polyelectrolyte chain in the inhibitory function. In addition, the strong interactions of CHT/TMC with Hep in preformed amyloids may disrupt the interaction between peptide and Hep within the aggregates and disintegrate the preformed fibrillar structure.

Figure 7.

Schematic representation of a possible mechanism of the inhibitory effect of CHT/TMC on Hep-mediated aggregation of the tau fragment peptide.

CONCLUSIONS

In summary, in this study we report the aggregation properties of a novel tau fragment peptide, tau_298–317, which is derived from the key microtubule binding domains of the full sequence tau. Our results show that this short peptide remains disordered and does not aggregate under normal conditions, similar to full sequence tau. The addition of the negatively charged polyelectrolyte Hep significantly promotes the aggregation of this peptide to form amyloid fibrils, most likely due to favorable electrostatic interactions between Hep and the peptide, allowing the monomers of the peptide to localize on Hep and form critical oligomeric nucleus. Furthermore, our results show that the charged polysaccharide CHT and its derivative TMC can effectively inhibit Hep-induced aggregation of tau_298–317. The attractive electrostatic interactions between the positively charged groups in CHT/TMC and the negatively charged residues of Hep play a key role in inhibiting Hep-peptide interactions and suppressing peptide aggregation. Our results suggest that cationic polyelectrolytes with suitable chain length and charge density can effectively block tau aggregation induced by polyanion. For example, chitosan oligosaccharides with controlled degree of polymerization and deacetylation have the potential to be effective tau aggregation inhibitors. A potential challenge of this strategy is that the inhibitory effects of the charged polysaccharides may be sequence nonspecific and the use of these molecules for anti-amyloid therapeutics may cause additional side effects. Designing polyelectrolyte structures with unique positively charged groups and defined charge distribution patterns may lead to the discovery of inhibitors with enhanced specificity for polyanion-induced tau aggregation.

EXPERIMENTAL METHODS

Materials.

All chemical reagents were obtained from commercial suppliers. Hep was obtained (Na salt, from porcine mucosa, Product-No. A3004) from ITW Reagents. Molecular weight of Hep was in the range of 8000–25,000 g/mol, and the average (16,500 g/mol) was used in the concentration calculation. CHT (low molecular weight, 75−85% deacetylated, molecular weight 50–190 kDa, #448869) was purchased from Sigma-Aldrich. TMC synthesis and characterization was carried out following a procedure described in detail previously.⁴⁰ The solution of CHT was prepared by thoroughly mixing appropriate amount of solid CHT in aqueous solution containing 0.25% acetic acid. The solution of TMC was prepared in water.

Synthesis and purification of tau fragment and mutant.

The tau fragment and mutation pS305 (addition of phosphate group at the side chain of Ser₃₀₅) peptides were synthesized on a PS3 solid phase peptide synthesizer (Protein Technologies Inc., Woburn, MA) using Fmoc chemistry. After cleaving the peptide from the resin using trifluoroacetic acid (TFA) cocktail, the crude peptide was purified by high-performance liquid chromatography (HPLC) using a C₁₈ reverse phase column, where mobile phase A (0.1 % TFA, V/V in water) and B (0.1 % TFA, V/V in acetonitrile) set in gradient program. The mutant was purified by using mobile phase A (0.1 % acetic acid, V/V in water) and B (0.1 % acetic acid, V/V in acetonitrile) set in gradient program. The molecular weight of the peptides was verified by matrix-assisted laser desorption ionization (MALDI) mass spectrometry. After purification, the peptides were lyophilized to obtain dry white samples. For further studies, the peptide aliquots were dissolved in MQ water, and the concentration of the peptide solutions were determined by using the Tyr UV absorbance at 280 nm (ε=1,280 M⁻¹cm⁻¹).

Aggregation kinetics of tau fragment using ThT fluorescence.

Aggregation kinetics of the tau fragment in the absence or presence of Hep were measured by using ThT fluorescence. In brief, stock solutions of tau fragment without or with Hep were diluted to a specific final concentration in pH 7.4 buffer (50 mM Na-phosphate). The solution also contained ThT with a final concentration of 20 μM. After mixing with pipetting in an Eppendorf tube, the final solution was transferred into a 96-well microplate (100 μL/well). The plate was sealed with a microplate cover and loaded into a Gemini SpectraMax EM fluorescence plate reader (Molecular Devices, Sunnyvale, CA), where it was incubated at 25°C. The excitation and emission wavelengths for the spectrometer were set at 440 nm and 480 nm, respectively. The fluorescence intensity of ThT was measured every 10 minutes after shaking for 5 seconds. For the assays with CHT or TMC, aliquots of polysaccharides were dissolved in 0.25% aqueous acetic acid (for CHT) or water (for TMC) to obtain the desired concentrations. A specific amount of polysaccharide solution was then added to the peptide solution to a final specific concentration.

The aggregation kinetics curve fitting was done in OriginPro 8.6 to extrapolate the lag time and half time (t₅₀). The data were fitted to the Boltzmann equation (a sigmoidal function) as given by

$$Y=\frac{{\text{A}}_1-{\text{A}}_2}{1+e^{(x-x_0)/dx}}+A_2$$ (1)

where A₁ is the minimum signal, A₂ is the maximum signal, x₀ is the time at which the signal reaches 50% of the maximum, dx is the time constant, the lag time is estimated by x₀ – 2dx.⁵³

Circular dichroism (CD).

Peptide samples were prepared by pipetting tau fragment (50 μM) without or with other polyelectrolytes in pH 7.4 buffer (20 mM Na-phosphate). The peptide solution was incubated at room temperature before measurement. At different time point for measurement, a 200 μL aliquot was transferred into a 0.1 cm quartz cell for far-UV measurements. CD spectra were recorded on a JASCO J-810 spectropolarimeter at a bandwidth of 1.0 nm, a scanning speed of 100 nm/min, a resolution of 0.1 nm, and an averaging time of 2 seconds. Five accumulations were acquired for each measurement. A control buffer was run as blank, which was used for baseline correction. Spectra were smoothed and baselines were corrected using the Spectragryph 1.2 software.

Particle size measurement using dynamic light scattering (DLS).

800 μL aliquots of different sample mixture were prepared in pH 7.4 phosphate buffer (50 mM Na phosphate) and incubated at 25°C for 24 hrs in a vertical rotating mixer. The size measurements were carried out by Zetasizer Nano ZS (Malvern Instruments, Worcestershire, UK) equipped with a 633 nm laser. Measurements were performed at 25 °C using a disposable polycarbonate folded capillary cell (model DTS1070) with a path length of 4 mm. The average diameters (d.nm) for each measurement were calculated from 12~15 runs per measurement using the original software supplied with the device.

Atomic force microscopy (AFM).

AFM images of the peptide samples were acquired using tapping mode measurement. Samples were prepared by pipetting an aliquot of the peptide solution (15 μL) onto the surface of a fresh mica (8 × 8 mm), then incubated in a sealed box to dry overnight at room temperature in the dark. AFM images were recorded using an Asylum Research MFP 3D Bio AFM system with Mikromasch NSC15/AI BS cantilevers. The imaging scan was performed at tip scan rates of 1.0 Hz, using cantilever drive frequencies of 70 kHz. Images were further processed using Gwydion software (2.60 version).