Fluorescent Peptides Sequester Redox Copper to Mitigate Oxidative Stress, Amyloid Toxicity, and Neuroinflammation

Abstract

Alzheimer’s disease is a progressive neurodegenerative disorder that significantly contributes to dementia. The lack of effective therapeutic interventions presents a significant challenge to global health. We have developed a set of short peptides (PN_Gln) conjugated with a dual-functional fluorophoric amino acid (N_Gln). The lead peptide, P2N_Gln, displays a high affinity for Cu²⁺, maintaining the metal ion in a redox-inactive state. This mitigates the cytotoxicity generated by reactive oxygen species (ROS), which are produced by Cu²⁺ under the reductive conditions of Asc and Aβ₁₆ or Aβ₄₂. Furthermore, P2N_Gln inhibits both Cu-dependent and -independent fibrillation of Aβ₄₂, along with the subsequent toxicity induced by Aβ₄₂. In addition, P2N_Gln exhibits inhibitory effects on the production of lipopolysaccharide (LPS)-induced ROS and reactive nitrogen species (RNS) in microglial cells. In vitro and cellular studies indicate that P2N_Gln could significantly reduce Aβ–Cu²⁺-induced ROS production, amyloid toxicity, and neuroinflammation, offering an innovative strategy against Alzheimer’s disease.

Alzheimer’s disease (AD) is a chronic neurological disorder characterized by the progressive decline of cognitive functions, accompanied by psychiatric symptoms, posing a significant challenge to global health.^1,2 Currently, AD affects an estimated 50 million individuals, with projections suggesting a surge to 152 million by 2050. The pathology of AD is marked by two primary features: the buildup of extracellular amyloid beta (Aβ) plaques and intracellular neurofibrillary tangles (NFTs) composed of abnormal tau protein.²⁻⁴ The complex interplay between copper and the regulation of amyloid precursor protein (APP) in AD has been a subject of investigation.^5,6 Studies indicate that Cu²⁺ may increase APP expression, potentially leading to a rise in Aβ peptide production in the brain.⁷ This overproduction is especially noted in ATP-7A-deficient fibroblast cells from mice and humans, where elevated Cu²⁺ levels stimulate Aβ peptide production.^2,8,9 Aβ peptides undergo several conformational changes, eventually forming toxic oligomers, protofibrils, and fibrillar aggregates with an ordered β-sheet structure.^10,11 The N-terminal region of Aβ (Aβ₁₆: D₁AEFRHDSGYEVHHQK₁₆) contains binding sites for Cu²⁺, and this region does not directly contribute to amyloid aggregation. Structural elucidation studies have identified specific residues such as histidine (H6, H13, and H14) and aspartic acid (D1) to be responsible for the formation of the Aβ–Cu²⁺ complex.¹²⁻¹⁵ This complexation accelerates Aβ aggregation, producing highly toxic species that cause neurodegenerative effects, including membrane and mitochondrial damage.¹ The redox activity of the Aβ–Cu²⁺ complex can generate ROS catalytically in the presence of the physiological reducing agent ascorbate, damaging biomolecules (lipids, proteins, and DNA) and cellular components, leading to neuronal injury and death.^16,17 Understanding and modulation of both Cu-dependent and -independent pathways of Aβ aggregation are vital for combating amyloid toxicity in AD.¹⁸⁻²⁰ Furthermore, the interplay between ROS and neuroinflammation is evident in AD, with activated microglia and astrocytes releasing ROS and RNS, exacerbating the oxidative stress environment in the brain.^21,22 This oxidative stress contributes to increased mitochondrial ROS and mitochondrial damage.²³

Small peptides have been used in the past in order to sequester and redox-silence Cu from Aβ to suppress the ROS formation.²⁴⁻²⁸ In our previous study, we reported natural tripeptide-based peptidomimetic molecules that complex Cu²⁺ and keep it in a redox-dormant state to alleviate Cu-associated toxicity in AD.^22,27,29 Inspired by these findings, we designed and synthesized a set of small peptides functionalized with a naphthalene monoimide (NMI)-based fluorophore (PN_Gln) to monitor and alleviate Cu-associated toxicity (Figure 1a). NMI is known for its anti-amyloidogenic and fluorescence properties, while the peptide fragment is designed for Cu²⁺ sequestration.^27,30,31 Thus, the designed fluorescent peptides are anticipated to concurrently report and capture Cu²⁺, arrest its ROS-producing redox cycle, and inhibit amyloid aggregation. These peptides are promising candidates due to the ease of structural tunability to enhance cell penetration ability and pharmaceutical efficacy. A critical design consideration for these peptide conjugates (PN_Gln) was the preference to complex Cu²⁺ over Zn²⁺, as the latter’s 10-fold higher synaptic concentration and higher disponibility could hinder Cu²⁺ capture.

Figure 1.

(a) Structures of fluorescent peptides (PN_Gln): P1N_Gln (GGHKN_Gln), P2N_Gln (GHHKN_Gln), P3N_Gln (HKN_Gln), and P4N_Gln (GHKN_Gln). (b) Schematic illustration of tripartite activity of fluorescent peptides PN_Gln: sequestering Cu from Aβ–Cu²⁺ complex, ameliorating amyloid toxicity and oxidative stress, and suppressing neuroinflammation (created with BioRender).

The peptide conjugates (PN_Gln) displayed a marked preference for Cu²⁺ over Fe³⁺, Zn²⁺, and Al³⁺. The ability of P2N_Gln to sequester and form stable, redox-inactive complexes with Cu²⁺ led to a substantial reduction in ROS production triggered by the redox-active Aβ₁₆–Cu²⁺ complex. Our cellular experiments showed that these peptide conjugates are nontoxic and significantly reduce ROS levels. Notably, the peptide P2N_Gln was most effective in sequestering Cu²⁺ from the Aβ₁₆–Cu²⁺ complex, preventing Aβ fibrillation and thus reducing neuronal toxicity. P2N_Gln also penetrated cells and decreased neuroinflammation by reducing lipopolysaccharide (LPS)-induced ROS and RNS in N9 microglial cells, and it helped to restore mitochondrial function and lower mitochondrial ROS production.

In this Letter, we present glycine-histidine-lysine (GHK)-inspired fluorescent peptide conjugates (PN_Gln) as potent Cu²⁺ sequesterers capable of mitigating ROS, oxidative stress, Cu-dependent and -independent amyloid toxicity, and neuroinflammation (Figure 1b). Our findings underscore the potential therapeutic implications of these fluorescent peptide conjugates to ameliorate Cu-associated multifaceted toxicity in AD.

The accumulation of Cu in amyloid plaques suggests its role in exacerbating the pathogenesis of AD in the brain. Therefore, sequestering Cu²⁺ from the Aβ–Cu²⁺ complex emerges as a potential strategy to investigate the role of Cu²⁺ in amplifying Aβ toxicity.^32,33 We designed four molecules by conjugating Cu-binding peptides with a fluorophore component. The fluorophore component consists of NMI-fluorophore-functionalized glutamine (N_Gln). The choice of NMI as the fluorophoric moiety stems from the fact that it can serve as a modulator of Aβ aggregation as well as reporter of metal ion binding in proximity. The tripeptide GHK, a natural peptide found in human plasma, saliva, and urine, has been recognized as a potent Cu-chelating ligand.³⁴ GHK has a strong affinity for Cu²⁺ and readily forms the GHK–Cu complex.^34,35 The formation of the GHK–Cu²⁺ complex is mediated by the α-amino group of Gly (G), the amide nitrogen of the Gly-His (G-H) peptide bond, and the imidazole side chain of His (H).³⁶ The XXH motif is a better redox silencer and stronger ligand compared to XH. XHH, which contains both motifs, is as strong as XXH but binds Cu²⁺ also faster than XXH (more like XH);^29,37,38 to complete the series, H at position 1 was also used. Based on the understandings from our previous studies, we considered HK, GHK, GHHK, and GGHK as Cu-binding peptides.^22,27,36,37 Naphthalene anhydride was condensed with ethylenediamine, which was coupled to glutamine to obtain the fluorescent amino acid conjugate (N_Gln). N_Gln was conjugated to HK, GHK, GHHK, and GGHK. The N-terminal glycine and histidine residues primarily facilitate Cu chelation, while the C-terminal Lys does not participate in the Cu–GHK complexation. N_Gln was conjugated at the C-terminus of peptides to avoid any direct interference in the Cu chelation. Following our rational design strategy, we synthesized four PN_Gln conjugates, namely, P1N_Gln (GGHKN_Gln), P2N_Gln (GHHKN_Gln), P3N_Gln (HKN_Gln), and P4N_Gln (GHKN_Gln) (Figure 1a).

The PN_Gln conjugates were prepared in two stages: synthesis of (i) N_Gln by a solution-phase protocol (Scheme S1) and (ii) copper-binding peptides using a solid-phase protocol (Scheme S2). In the solid-phase synthesis of peptides, Rink amide resin beads were soaked in DCM and DMF, followed by treatment with piperidine at 20% in DMF to deprotect Fmoc protective groups of surface amines. N_Gln was attached to the resin using HBTU, HOBt, and DIPEA in DMF as coupling reagents. The Kaiser test confirmed the complete coupling of N_Gln to the resin, which was then treated with piperidine (20% in DMF) to release the free amine of the N_Gln unit. Subsequently, Lys, His, and Gly were added using the Fmoc-chemistry peptide synthesis protocols as shown in Scheme S2. Finally, the resin was treated with trifluoroacetic acid (TFA) to obtain P1N_Gln, P2N_Gln, P3N_Gln, and P4N_Gln. All the intermediates, N_Gln, and PN_Gln were characterized by NMR spectroscopy and mass spectrometry (LCMS, HRMS and MALDI-TOF) analysis.

The ability of peptides to bind metal ions was assessed by various spectroscopic techniques. The absorption spectra revealed an intense band at 450 nm, signifying the incorporation of the N_Gln moiety in the peptides. The PN_Gln showed a strong emission band in the green spectral region, with emission maximum (λ_max) around 550 nm upon excitation at 450 nm. Numerous studies have established the role of various metal ions (Cu²⁺, Fe³⁺, Zn²⁺, and Al³⁺) to aggravate amyloid toxicity through metal-complexation-induced acceleration of Aβ aggregation.^22,27,39 The selective sequestering of these metal ions from Aβ–metal complexes could provide valuable insights into their role in amyloid toxicity. We treated a homogeneous solution of PN_Gln at 50 μM independently with Cu²⁺, Zn²⁺, Fe³⁺, and Al³⁺ in a 1:1 stoichiometric ratio and recorded fluorescence emission (λ_ex = 450 nm, λ_em = 550 nm). The fluorescence of PN_Gln was selectively reduced in the presence of Cu²⁺, while other metal ions showed minimal effect (Figures 2a and S1). Additionally, P3N_Gln showed a modest affinity toward Zn²⁺. Notably, P1N_Gln, P2N_Gln, and P4N_Gln showed excellent selectivity toward Cu²⁺. The His residue in the PN_Gln contains an imidazole group, which is known for its strong affinity for Cu²⁺. This imidazole group can readily form coordination bonds with Cu²⁺, facilitating its selective binding interaction. A possible explanation for the observed selectivity could be attributed to the ability of Cu²⁺ to deprotonate the amide via coordination at a neutral pH, a property not exhibited by Zn²⁺. Further, Fe³⁺ and Al³⁺, being hard acids, exhibit binding preferences for oxygen over nitrogen.^40,41 The amino group of Gly and the amide group along with the imidazole group of His create a coordination environment that favors Cu binding. These findings collectively validate our design strategy, which aimed at achieving selective complexation of Cu²⁺ by the PN_Gln in the presence of other competing metal ions.

Figure 2.

Cu chelation by the PN_Gln. (a) Selectivity of P2N_Gln among Cu²⁺, Zn²⁺, Al³⁺, and Fe³⁺ from its fluorescence spectra (λ_ex = 450 nm and λ_em = 550 nm). (b) Titration curve monitoring fluorescence of P2N_Gln with increasing concentration of Cu²⁺. (c) Frequency- and amplitude-normalized EPR spectra of Cu(II)–PN_Gln. Conditions: [Cu²⁺] = 450 μM (except for Cu(II)–(P3N_Gln)₂, where [Cu²⁺] = 225 μM), [peptide] = 500 μM, HEPES 50 mM, pH 7.4, glycerol 10% v/v, and T = 100 K.

The titration of Cu²⁺ (0–100 μM) into the PN_Gln solution (50 μM) resulted in a marked decrease in emission intensity, which varied with the concentration of Cu²⁺ (Figures 2b and S1). This reduction in fluorescence intensity upon Cu²⁺ addition confirmed the formation of PN_Gln–Cu²⁺ complexes. Further, the titration experiment showed that PN_Gln and Cu²⁺ formed a complex predominantly in a 1:1 stoichiometric ratio, except for P3N_Gln, which showed 2:1 stoichiometric complexation, in agreement with the literature and electron paramagnetic resonance (EPR) data (Scheme S3 and Figure 2c). The EPR fingerprint of Cu(II)–P1N_Gln suggests a square-planar complex with a four-N coordination environment (consisting of NH₂, N⁻, N⁻ and NIm) (g_∥ ≈ 2.190, A_∥ ≈ 600 MHz), consistent with the expected Cu²⁺ binding to the ATCUN motif, which is a small metal-binding site found in the N-termini of many naturally occurring proteins (Scheme S3).^38,42,43 The conjugation of the NMI fluorophore to peptides in PN_Gln did not affect their Cu²⁺ coordination ability. EPR analysis of Cu(II)–P4N_Gln complex (g_∥ ≈ 2.232, A_∥ ≈ 575 MHz) supported a three-N coordination mode of metal binding involving the N-terminal amine, one amidate, and the imidazole moiety of His2 (NH₂, N^–, N_Im), along with an additional ligand, typically H₂O (Scheme S3).³⁶ In Cu(II)–P2N_Gln, the major component observed is in line with the expected predominance of the four-N coordination of the ATCUN motif at pH 7.4.⁴⁴ The fluorescence quenching observed for Cu–(P3N_Gln)₂ complex (g_∥ ≈ 2.235, A_∥ ≈ 560 MHz) led to EPR spectra acquisition with this stoichiometry, suggesting coordination to three N and one O, similar to the reported crystal structure of Cu(II)–His₂, where the equatorial coordination sphere is provided by the two amino groups (NH₂), one imidazole (N_Im), and one oxygen (Scheme S3).

After confirming the expected coordination sphere of PN_Gln for Cu²⁺, we sought to assess their ability to sequester Cu²⁺ from the Aβ₁₆–Cu²⁺ complex. The extraction kinetics of Cu²⁺ from the Aβ₁₆–Cu²⁺ complex by PN_Gln was monitored by recording the changes in NMI emission. The NMI fluorescence was found to decrease upon sequestration of Cu²⁺ from Aβ₁₆ by PN_Gln, as shown in Figure 3a–d. PN_Gln showed variable binding kinetics, as evident from the fluorescence quenching within less than 2 min upon addition of Cu²⁺. This result suggests simultaneous binding of all Cu²⁺ in the solution to PN_Gln. P3N_Gln exhibited a rapid but partial quenching (∼65%) upon sequestration of Cu²⁺ from Aβ₁₆–Cu²⁺ complex compared to that in the presence of Cu²⁺ alone (Figure 3c). P4N_Gln showed a quenching effect in the presence of Aβ₁₆–Cu²⁺ complex comparable to that of Cu²⁺ alone after incubation for ∼15 min, indicating slower kinetics of sequestering Cu²⁺ from Aβ₁₆–Cu²⁺ complex (Figure 3d).

Figure 3.

(a–d) Kinetic analysis of Cu²⁺ coordination by PN_Gln through monitoring NMI fluorescence: (a) P1N_Gln; (b) P2N_Gln; (c) P3N_Gln; (d) P4N_Gln. (e, f) MALDI-TOF mass analysis of Aβ₁₆–Cu²⁺ complex in HEPES (10 mM, pH 7.4) buffer (e) and Aβ₁₆–Cu²⁺ complex after the addition of P2N_Gln (f).

P2N_Gln exhibited quenching similar to that of Cu²⁺ alone after ∼10 min (Figure 3b). P1N_Gln displayed significantly slower kinetics and did not show complete quenching even after 20 min (Figure 3a). The overall rate of Cu²⁺ sequestration from Aβ₁₆–Cu²⁺ complex by PN_Gln followed the order P3N_Gln > P2N_Gln > P4N_Gln > P1N_Gln. Circular dichroism (CD) analysis revealed that Aβ₁₆ exhibits a negative Cotton effect at 200 nm, which suggests the presence of a random coil structure. Upon the addition of Cu²⁺, the intensity of this band diminished, thereby confirming the Aβ₁₆–Cu²⁺ complexation. Peptides P1N_Gln, P2N_Gln, and P4N_Gln were able to restore the CD band, with P2N_Gln demonstrating the highest effectiveness (Figure S2a). On the other hand, P3N_Gln was less effective. The fluorescent peptides had a minimal impact on the CD signature of Aβ₁₆ (Figure S2b,c). Overall, P2N_Gln showed the highest efficacy in sequestration of Cu²⁺ from the Aβ₁₆–Cu²⁺ complex.

To substantiate the effectiveness of P2N_Gln in sequestering Cu²⁺, MALDI-TOF mass spectrometry analysis was performed. The Aβ₁₆ complexation with Cu²⁺ resulted in the formation of [Aβ₁₆ + Cu] as a major species with corresponding mass m/z = 2016.80. However, bimetallic and trimetallic complexes [Aβ₁₆ + 2Cu] and [Aβ₁₆ + 3Cu] with corresponding masses m/z = 2079.90 and 2141.93, respectively, were also observed, albeit with very low intensity (Figure 3e). Treatment of [Aβ₁₆–Cu] complex with P2N_Gln revealed distinct peaks at m/z = 933.33 and 997.42, respectively, corresponding to [P2N_Gln+Cu] and bimetallic copper adduct [P2N_Gln+2Cu], while a significant decrease in the intensity of [Aβ₁₆ + Cu] peaks was observed (Figure 3f). These findings confirm the potential of P2N_Gln to selectively sequester Cu²⁺ from the Aβ₁₆–Cu²⁺ complex. The complexation of Aβ and Cu²⁺ has been reported with variable binding affinities of ∼10⁵ to 10⁹ M^–1, depending on the methodologies employed and the competing ligands used in the determination.⁴⁵⁻⁴⁷ Similarly, a number of short peptides including GHK have been reported to complex Cu with variable binding affinities.^{22,25,28,34,35,48,49} Small peptides are anticipated to exhibit higher binding affinities for Cu²⁺ than Aβ, making them potential candidates for sequestration of Cu²⁺ from the Aβ–Cu²⁺ complex.

We employed atomic force microscopy (AFM) to comprehensively study the ability of PN_Gln to sequester Cu²⁺, thereby inhibiting the Aβ aggregation to form toxic species. Treatment of Aβ₄₂ and Aβ₄₂–Cu²⁺ samples with P1N_Gln, P2N_Gln, or P4N_Gln resulted in a significant reduction in the formation of Aβ aggregates. P1N_Gln and P2N_Gln showed good anti-amyloid activity while P3N_Gln exhibited moderate activity, as revealed by the presence of fewer fibrils. This underscores the significant role of P1N_Gln, and P2N_Gln in inhibiting Aβ aggregation (Figure 4a). Notably, we demonstrated the amyloid aggregation inhibition property of the NMI moiety in our previous study.³⁰ In this study, we treated Aβ₄₂ with N_Gln to evaluate its anti-amyloidogenic properties. As expected, N_Gln effectively inhibited Aβ₄₂ aggregation, while relatively lower activity was observed against Aβ₄₂ + Cu²⁺ aggregation, possibly due to its inability to sequester Cu²⁺. The control study showed negligible activity of GHK tripeptide in inhibiting Cu-dependent and -independent amyloid aggregation (Figure S3). The AFM analysis confirmed the ability of P2N_Gln to effectively inhibit Cu²⁺-induced Aβ fibrillation. We employed a molecular docking study to understand the interaction of PN_Gln with Aβ₄₂ fibril (PDB ID 2BEG) (Figures 4b and S4–S7). Among all PN_Gln, P2N_Gln displayed optimal binding affinity (ΔG = −6.6 kcal/mol) by establishing hydrogen bonds with residues A21, D23, and V36 of the Aβ₄₂ fibril. Furthermore, the NMI moiety participated in a π–π stacking interaction with the F19 residue, while the imidazole of P2N_Gln formed a T-shaped π–π bond with residue F20 (Figure 4c).

Figure 4.

(a) AFM images of Aβ₄₂ and Aβ₄₂ + Cu²⁺ fibrils with peptide treatment after incubation for 48 h. (b) Molecular docking of P2N_Gln with Aβ₄₂ fibril (PDB entry 2BEG). (c) Observed interactions between P2N_Gln and Aβ₄₂ fibril. (d) Asc consumption (measured at λ = 265 nm) over time. Asc was added first and triggered by adding Cu²⁺ alone or PN_Gln–Cu²⁺ complexes after 10 min. (e) Asc consumption (measured at λ = 265 nm) over time. Asc was added first and triggered by the addition of Cu²⁺ or Aβ₁₆–Cu²⁺ after 10 min, followed by the addition of PN_Gln.

We explored the potential of PN_Gln in mitigating the Aβ₁₆–Cu²⁺-induced ROS production. Aβ₁₆–Cu²⁺ in the presence of ascorbate (Asc), a physiologically relevant reducing agent, undergoes a redox process and is able to catalyze the production of ROS.⁵⁰ To assess the impact of PN_Gln on the ROS production induced by Cu²⁺ and Aβ₁₆–Cu²⁺, we monitored the consumption of Asc by UV–vis spectroscopy at 265 nm (ε_Asc = 14,500 M^–1 cm^–1) (Figures 4d,e and S8a). PN_Gln was premixed with Cu²⁺ in HEPES buffer (50 mM, pH 7.4) and added after Asc to initiate the redox reaction. Notably, Cu²⁺ was found to be highly redox-active in the presence of rapidly reducing conditions of Asc, culminating in complete reduction within 5 min. In contrast, P1N_Gln–Cu²⁺ and P2N_Gln–Cu²⁺ exhibited negligible oxidation of Asc, aligning with the redox behavior typically associated with the ATCUN–Cu²⁺ complex.^29,38 P4N_Gln–Cu²⁺ displayed relatively slow oxidation of Asc, whereas P3N_Gln–Cu²⁺ showed faster kinetics, reaching completion after ∼10 min (Figure 4d).

To evaluate the impact of PN_Gln on the extraction of Cu²⁺ and inhibition of Aβ₁₆–Cu²⁺-induced ROS production, PN_Gln were introduced during Asc consumption, which was triggered by the addition of either Cu²⁺ or Aβ₁₆–Cu²⁺ complex (Figure 4e). Next, PN_Gln were preincubated with Asc solution followed by the addition of Aβ₁₆–Cu²⁺ complex (Figure S8a). The results show that P3N_Gln, at two different stoichiometric ratios with Cu²⁺ (1:1 and 1:2), was unable to attenuate Asc consumption. Given that P3N_Gln–Cu²⁺ alone exhibited Asc reduction comparable to that of Aβ₁₆–Cu²⁺, it was challenging to ascertain the precise site of Cu binding under these conditions (Figure 4e). P4N_Gln displayed a decrease in Asc consumption compared with P3N_Gln but did not reach a significant saturation plateau. This finding aligns with the capacity of P4N_Gln to exclusively bind Cu²⁺ within a three-N coordination pattern. This interaction gives rise to a marginally redox-active P4N_Gln–Cu²⁺ complex, as observed in only Cu²⁺ (Figure 4e). In contrast, P1N_Gln successfully prevented Asc reduction when added before or during the Cu–Asc redox cycle (Figures 4e and S8a). This finding aligns with the formation of a redox-inert P1N_Gln–Cu²⁺ complex, as illustrated in Figure 3c, where Cu²⁺ is bound to the four-N motif, effectively resisting reduction by Asc. Nevertheless, the initial change in the slope required to reach the saturation line was not instantaneous, indicating the gradual kinetics involved in the abstraction of Cu²⁺ from Aβ₁₆–Cu²⁺ and its subsequent complex formation. This observation correlates with the delayed removal of Cu²⁺ from Aβ, as observed in the fluorescence measurements conducted earlier (Figure 3a–d). In both experiments (Figures 4e and S8a), P2N_Gln demonstrated the highest efficacy in limiting Asc consumption. This implies a rapid uptake of Cu²⁺ from the redox-active Aβ₁₆–Cu²⁺ by the peptide followed by robust Cu²⁺ stabilization, making it nonreducible by Asc.

To evaluate the Cu²⁺ chelation and inhibition of Aβ₁₆–Cu²⁺-induced ROS production by PN_Gln, we monitored the fluorescence kinetics of 7-HO-CCA reacting with HO^•. P3N_Gln exhibited minimal HO^• production due to P3N_Gln–Cu²⁺ complexation, which shows relatively less redox inhibition. In contrast, P1N_Gln and P2N_Gln completely inhibited HO^•, corroborated by the Asc consumption assay, which is attributed to the formation of redox-dormant PN_Gln–Cu²⁺ complexes (Figure S8b).

Aβ aggregates impair neuronal function by increasing the frequency of action potentials and membrane depolarization, resulting in reduced cell viability. This cytotoxic effect is mediated by the ROS production, which in turn is triggered by the interaction of Aβ and Cu²⁺ within the cells.⁵¹ First, the MTT assay was performed by treating the SH-SY5Y cells with PN_Gln (10–100 μM), and cytotoxicity was assessed after 24 h. The PN_Gln were nontoxic to cells, with observed viability of over 90% at 100 μM (Figures 5a and S10a). Next, we assessed rescue of cells from toxicity induced by Cu²⁺ and Asc, which showed the potential of P2N_Gln at 10 μM to restore over 90% cell viability (Figure 5b). Similarly, P2N_Gln at the same concentration exhibited remarkable efficacy in scavenging over 90% of ROS produced by Aβ₁₆–Cu²⁺ and Asc (Figure 5c). 2′,7′-Dichlorofluorescein diacetate (DCFDA) assay was performed to elucidate the mechanism of inhibition of ROS production by PN_Gln. P2N_Gln emerged as the most potent inhibitor of ROS production, confirming its superior antioxidant capacity (Figure S10a). In all the assays, P2N_Gln showed higher efficacy in modulating Aβ₁₆–Cu²⁺-induced ROS production and amyloid toxicity compared to the other peptide conjugates, including GHK control (P4N_Gln) (Figure S10b–e). The in vitro ability of PN_Gln in inhibiting Aβ aggregation led us to perform an in cellulo study to evaluate its effectiveness in mitigating Aβ-induced toxicity under both metal-independent and metal-dependent conditions. SH-SY5Y cells were incubated with Aβ₄₂ fibrils (10 μM) alone and in combination with PN_Gln (10 μM) in the absence or presence of Cu²⁺ (10 μM) for 24 h. Cell viability was assessed using the MTT assay. Aβ₄₂ treatment significantly reduced cell viability to ∼55% and 48% in the absence and presence of Cu²⁺, respectively. Interestingly, PN_Gln treatment significantly improved cell viability from metal-independent cellular toxicity to ∼80% (Figure 5d). The efficacy of all four PN_Gln in mitigating Aβ-fibril-induced toxicity underscores the anti-amyloidogenic potential of the N_Glu moiety. P1N_Gln, P2N_Gln, P3N_Gln, and P4N_Gln exhibited rescue of cells from toxicity induced by Aβ₄₂ fibrils in the presence of Cu²⁺, achieving viabilities of 85%, 92%, 82%, and 86%, respectively. Among them, P2N_Gln showed the highest efficacy in mitigating metal-dependent Aβ toxicity due to its superior ability to sequester Cu²⁺ from the Aβ–Cu²⁺ complex (Figure 5e). Furthermore, P2N_Gln exhibited a concentration-dependent rescue of cells from toxicity induced by Aβ₄₂ oligomers in both the absence and presence of Cu²⁺ (Figure S10f,g). These in cellulo studies highlight the potential of P2N_Gln to effectively protect neuronal cells from both metal-independent and metal-dependent Aβ₄₂ toxicity. A flow cytometry study was carried out to validate the effectiveness of P2N_Gln in suppressing ROS production in SH-SY5Y cells treated with Aβ₁₆, Cu²⁺, and Asc using DCFDA dye (Figure 5f–j). The data showed concentration-dependent inhibition of ROS production by P2N_Gln (Figure 5k,l), which corroborated the results from other experiments. All these results demonstrate the remarkable ability of P2N_Gln to sequester and keep Cu²⁺ in a redox-dormant state to suppress ROS production under cellular conditions.

Figure 5.

Cellular studies of PN_Gln. (a) Cell viability assay of PN_Gln (100 μM) in SH-SY5Y cells after 24 h of incubation. (b, c) Effect of PN_Gln on cells treated with (b) Cu²⁺ and Asc or (c) Aβ₁₆–Cu²⁺ and Asc. (d) Cellular rescue from Aβ₄₂-induced toxicity by PN_Gln. (e) Cellular rescue from Aβ₄₂–Cu²⁺-induced toxicity by PN_Gln. (f–j) Flow cytometry analysis of SH-SY5Y cells incubated with Aβ₁₆ and Cu²⁺–Asc followed by treatment with varying concentrations of P2N_Gln and staining with DCFDA. The percentage written in the P1 quadrant represents the percentage of the population present in that quadrant with higher DCFDA-based fluorescence signal than the whole population. (k) Histogram of SH-SY5Y cells in count versus DCFDA intensity. (l) Quantification of the DCFDA fluorescence intensity after the treatment of cells under Aβ₁₆–Cu²⁺ and Asc with various concentrations of P2N_Gln. (The experiment was repeated with n = 3, and statistical analysis was performed using a simple one-way ANOVA; *, p = 0.05).

Anti-inflammatory properties of PN_Gln were investigated by monitoring ROS levels and nitrite secretion in N9 microglial cells. The cellular uptake of P2N_Gln was studied by confocal microscopy by monitoring its intrinsic fluorescence property. N9 cells were incubated with P2N_Gln (5 and 10 μM) in RPMI 1640 medium at 37 °C for 1 h under standard cell culture conditions. The live-cell confocal imaging revealed the localization of P2N_Gln within the cytoplasm of N9 cells, which indicates its successful uptake across the cell membrane (Figure 6a). N9 cells were stimulated with lipopolysaccharide (LPS), which induced the secretion of proinflammatory cytokines and the activation of NADPH oxidase (NOX), leading to ROS production. The concentration-dependent cytotoxicity assessment revealed that PN_Gln were nontoxic in N9 microglia cells up to 50 μM (Figure S10b). DCFDA assay was employed to measure the ROS levels in real time after treating the cells with LPS at a concentration of 1 μg/mL for 24 h. PN_Gln were highly effective in ameliorating the ROS level in the N9 cells (Figure 6b). Treatment with P2N_Gln significantly inhibited the LPS-induced ROS production in a concentration-dependent manner, as shown in Figure 6c. The anti-inflammatory activity of P2N_Gln was further confirmed by the Griess assay, which quantifies the nitric oxide (NO) levels as nitrite (NO₂^–). The inflammation in microglial N9 cells was assessed by measuring the nitrite levels in the medium. The nitrite levels in the LPS-treated cells were considered 100%, and in comparison, the untreated cells showed 50%. The LPS-treated cells incubated with PN_Gln (10 μM) exhibited decrease in nitrite production to 70%, 56%, 85%, and 89% for P1N_Gln, P2N_Gln, P3N_Gln, and PN_Gln, respectively (Figure 6d). Notably, P2N_Gln showed the most potent effect, with an IC₅₀ of 6.45 μM, and inhibited nitrite production in a concentration-dependent manner (Figure 6e). Thus, P2N_Gln effectively modulates inflammatory response by inhibiting LPS-induced microglial activation.

Figure 6.

Effect of P2N_Gln in modulating inflammation in microglia cells. (a) Cellular uptake of P2N_Gln (blue: Hoechst dye was used for nuclear staining; green: P2N_Gln). (b) DCFDA assay in LPS-induced N9 cells treated with PN_Glu (10 μM). (c) Concentration-dependent inhibitory effect of P2N_Gln on ROS production. (d) Effect of PN_Gln in inhibiting nitrite production. (e) Plot of nitrite secretion versus concentration of P2N_Gln and IC₅₀ value of P2N_Gln inhibiting nitrite secretion. (The experiment was repeated with n = 3, and statistical analysis was performed using a simple one-way ANOVA; *, p = 0.05).

Neuroinflammation instigates the production of ROS, which in turn amplifies mitochondrial ROS and initiates mitochondrial fission.⁵² To assess the effectiveness of P2N_Gln in mitigating the morphological changes in mitochondria caused by neuroinflammation, the control and LPS-treated cells were stained with MitoTracker Orange and visualized through live-cell fluorescence microscopy. The control cells exhibited elongated and interconnected mitochondrial networks, while exposure to LPS (1 μg/mL) led to mitochondrial fragmentation. Remarkably, treatment with P2N_Gln (5 and 10 μM) maintained the tubular structures of the mitochondria (Figure 7a). Subsequently, the mitochondrial ROS production was evaluated using MitoSOX Red. The introduction of LPS caused an increase in mitochondrial ROS, which in turn led to an elevation in MitoSOX intensity. Intriguingly, P2N_Gln was capable of reducing the mitochondrial generation of superoxide anion in a concentration-dependent fashion (Figure 7b–d). These findings underscore the inherent anti-inflammatory attributes of P2N_Gln within the neurological microenvironment.

Figure 7.

Effect of P2N_Gln on mitochondrial damage and mitochondrial ROS in N9 cells after inducing neuroinflammation using LPS for 24 h with or without P2N_Gln. (a) Monitoring mitochondrial fragmentation and its inhibition by P2N_Gln. Mitochondria are stained with MitoTracker Orange. (b) Monitoring mitochondrial ROS using MitoSOX and its inhibition by P2N_Gln. (c) Quantification of mitochondrial ROS. (d) Quantification of relative fluorescence intensity of P2N_Gln showing the amount of P2N_Gln inside the cells after treatment. (Over 200 cells were quantified; the data between two groups were analyzed by the independent-sample t test, and one-way ANOVA was used to compare the means of three groups; *, p = 0.05).

The multifaceted nature of AD etiology presents a formidable challenge, with the alleviation of oxidative stress, amyloid toxicity, and neuroinflammation forming the cornerstone of therapeutic strategies. Peptides, because of their tunability and biocompatibility, are emerging as potential therapeutic agents for AD. We developed fluorophore-conjugated short peptides through rational design strategies to achieve selective sequestration of Cu²⁺ and modulation of oxidative stress, Cu-dependent and -independent amyloid toxicity, and neuroinflammation. These fluorescent peptides exhibited superior chelation capabilities for Cu²⁺ ions and selectively sequestered Cu²⁺ from the Aβ₁₆–Cu²⁺ complex, demonstrating a high specificity for Cu²⁺ over other metal ions. Notably, P2N_Gln emerged as a potent suppressor of ROS production induced by Aβ₁₆–Cu²⁺. The cellular studies validated the biocompatibility of these peptides and their efficacy in preserving cell viability. P2N_Gln was particularly effective in curtailing ROS generation under oxidative stress conditions and mitigating ROS production in microglial cells under inflammatory conditions. This peptide was able to penetrate cells, inhibit nitrite production, rescue mitochondrial fragmentation, and suppress mitochondrial ROS, thereby indicating its anti-inflammatory activity. In conclusion, our findings highlight the therapeutic potential of these fluorescent peptide conjugates in addressing various facets of AD pathology. Further research and optimization could potentially pave the way for the development of innovative therapeutic agents for AD.

Glossary

Abbreviations

AD

Alzheimer’s disease

Aβ

amyloid β

APP

amyloid precursor protein

NMI

naphthalene monoimide

ATCUN

amino terminal copper(II) and nickel(II) binding

CD

circular dichroism

NMR

nuclear magnetic resonance

LCMS

liquid chromatography–mass spectrometry

HRMS

high-resolution mass spectrometry

MALDI-TOF

matrix-assisted laser desorption ionization time-of-flight mass spectrometry

EPR

electron paramagnetic resonance

AFM

atomic force microscopy

ROS

reactive oxygen species

HEPES

4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

Asc

sodium ascorbate

LPS

lipopolysaccharide

DCFDA

2′-7′-dichlorodihydrofluorescein diacetate

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

CCA

coumarin carboxylic acid

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.4c00283.

• Synthetic scheme of fluorescent peptides, characterization data, absorbance and fluorescence spectra, CD spectra, molecular docking figures of the peptides, and AFM images of the Aβ₄₂ fibrils with GHK and fluorescent probe N_Gln (PDF)

Author Contributions

T.G. conceptualized the project. S.M., Y.V.S., S.S., and T.G. designed the experiments. S.S. and S.M. synthesized the fluorescent peptides. S.M. and Y.V.S. performed in vitro studies. B.V. performed EPR studies. S.M. performed Aβ₄₂ expression and purification, AFM microscopy, in silico studies, and cellular studies. All authors contributed to curating and analyzing the data. S.M., Y.V.S., and T.G. wrote the manuscript, and others gave input.

The authors declare no competing financial interest.

Special Issue

Published as part of ACS Medicinal Chemistry Lettersvirtual special issue “Celebrating the 25th Anniversary of the Chemical Research Society of India”.