Dual Targeted Far‐Red Emissive Small Molecules for Mitochondrial Imaging and Multifunctional Modulation in Alzheimer's Disease

ABSTRACT

Alzheimer's disease (AD), the leading cause of age‐related dementia, is a multifactorial neurodegenerative disorder associated with amyloid‐β (Aβ) aggregation, metal ion dyshomeostasis, oxidative stress, and mitochondrial dysfunction. The pathological interplay among these processes necessitates multifunctional small molecules capable of both diagnosis and therapeutic intervention. Herein, we report two triphenylamine (TPA)‐based aggregation‐induced emission (AIE) luminogens, TPA‐RPA (triphenylamine‐rhodaninepropanoic acid) and TPA‐ER (triphenylamine‐ethylrhodanine), designed as mitochondria‐active anti‐amyloid theranostic agents. Among them, TPA‐RPA emerged as the lead candidate, exhibiting far‐red emission and a pronounced ∼5‐fold fluorescence turn‐on response upon selective binding to Aβ40 fibrils, enabling sensitive detection of pathogenic aggregates. TPA‐RPA effectively inhibited Aβ40 fibrillogenesis, disaggregated preformed fibrils, binds with selective metal ions Fe³⁺/Cu²⁺, and suppressed Fe³⁺ mediated amyloid aggregation, highlighting its multifunctional anti‐amyloid activity. In neuronal cells, TPA‐RPA displayed favorable biocompatibility, appreciable mitochondrial localization, and significant protection against Aβ‐induced cytotoxicity, oxidative stress, and mitochondrial dysfunction. Comparative studies identified TPA‐ER as a useful analog with a stronger but more pronounced oxidative response, while TPA‐RPA demonstrated the most balanced overall biological profile, establishing it as a synthetically accessible and mechanistically versatile platform for image‐guided intervention in AD, supported by photophysical, cellular, in silico molecular docking analyses, and blood‐brain barrier/drug‐likeness evaluations.

A mitochondria‐targeted far‐redemissive AIEEgen enables the selective detection of Aβ40 fibrils, inhibition of fibrillogenesis, modulation of metal ions, and suppression of Aβ‐induced mitochondrial apoptosis, providing a unified image‐guided therapeutic platform for Alzheimer's disease.

1. Introduction

Alzheimer's disease (AD), the most prevalent form of age‐associated dementia, continues to pose a formidable biomedical challenge due to its multifactorial etiology and the absence of effective disease‐modifying therapies [1, 2, 3]. Central to the progression of AD is the abnormal aggregation of Aβ peptides into toxic oligomers and insoluble fibrils, which disrupt synaptic transmission, impair mitochondrial function, and ultimately lead to neuronal death [4, 5, 6]. While soluble Aβ oligomers are considered major neurotoxic species, insoluble fibrillar plaques remain central pathological hallmarks of AD and may contribute indirectly through chronic inflammation, seeding, and dynamic release of smaller toxic assemblies [7, 8, 9]. Aβ40 and Aβ42 aggregates are key pathological agents that trigger oxidative stress and neuroinflammation in the aging brain [10, 11]. Despite extensive efforts, therapeutic strategies aimed at targeting Aβ aggregation have yielded limited clinical success, often hindered by poor selectivity, insufficient blood‐brain barrier (BBB) permeability, and a lack of diagnostic integration [12, 13].

The development of small‐molecule theranostic systems that not only interfere with Aβ aggregation but also enable real‐time monitoring of therapeutic efficacy is a promising approach to overcoming these limitations [14, 15, 16]. Theranostic probes, agents combining diagnostic and therapeutic capabilities within a single molecular architecture, represent an emerging class of tools in precision neurotherapeutics [17]. Ideally, such molecules should meet several criteria: high specificity for aggregated Aβ, intrinsic fluorescence for imaging, drug‐like properties for BBB penetration, and additional functional elements that can address mitochondrial dysfunction and metal‐ion dyshomeostasis, both of which are intricately linked to AD pathology [18, 19, 20, 21]. Triphenylamine (TPA) derivatives provide a promising molecular platform for the development of multifunctional theranostic agents [22]. TPA scaffolds possess excellent photophysical properties, including tunable emission, strong donor properties, and aggregation‐induced emission (AIE), making them well‐suited for imaging in complex biological environments [23]. Moreover, the donor–acceptor (D–A) strategy enables fine‐tuning of molecular electronic properties, enhancing both Aβ binding affinity and emission responsiveness. Despite growing interest in TPA‐based fluorophores for bioimaging, the rational engineering of D‐A‐type TPA small molecules with integrated anti‐amyloid, mitochondrial‐targeting, and diagnostic capabilities remains largely unexplored [6]. Mitochondrial dysfunction is an early and critical hallmark of AD, often driven by the abnormal accumulation of amyloid‐β peptides, particularly Aβ40, within mitochondria [4, 24]. This intramitochondrial localization of Aβ40 has been shown to disrupt mitochondrial dynamics, increase reactive oxygen species (ROS) production, impair ATP synthesis, and initiate apoptotic cascades [25, 26, 27]. Targeting Aβ40 within mitochondria, therefore, represents a promising therapeutic strategy to halt or reverse mitochondrial‐associated neurodegeneration in AD [28, 29]. In this context, we have developed two multifunctional small molecules that localize to mitochondria and specifically interact with Aβ, aiming to interfere with its pathological activity at the site of action. By localizing to mitochondria and binding Aβ40, the synthesized molecules demonstrate the potential to attenuate Aβ‐induced mitochondrial toxicity, restore bioenergetic balance, and mitigate oxidative stress, thereby providing a targeted approach to address a key pathological feature of AD.

Rhodanine, a thiazolidine derivative, a five‐membered heterocyclic compound containing sulfur, nitrogen, and carbonyl groups, is a widely utilized electron‐accepting moiety in D–A systems due to its strong electron‐withdrawing capacity, planarity, and synthetic versatility [30, 31]. When integrated into fluorescent molecular scaffolds, rhodanine effectively facilitates intramolecular charge transfer (ICT), enabling the design of probes with red‐shifted absorption and emission, particularly advantageous for NIR imaging applications. Biologically, rhodanine‐based derivatives have demonstrated diverse pharmacological activities, including antimicrobial, anticancer, and anti‐inflammatory properties [24, 32]. The role of rhodanine structures in amyloid pathology and their potential to inhibit β‐sheet formation is still underexplored. Additionally, these structures can modulate pathological protein aggregation processes, such as those involved in AD. Conjugating them into mitochondria‐targeting probes provides a dual‐functional advantage: enhancing optical responsiveness while also offering therapeutic relevance in neurodegenerative conditions.

In this work, the design, synthesis, and biological evaluation of a new class of D‐A‐engineered TPA‐Rhodanine small molecules as multifunctional theranostic probes for AD was reported. By modulating the acceptor strength within the TPA framework, we developed two compounds with selective affinity for Aβ40 fibrils, strong fluorescence turn‐on behavior upon binding, and robust inhibition of fibrillogenesis. Among them, TPA‐RPA emerged as a lead candidate, displaying ∼5‐fold fluorescence enhancement in the presence of Aβ aggregates, potent disaggregation of preformed fibrils, and selective mitochondrial localization, an essential feature for modulating Aβ‐induced mitochondrial toxicity. Furthermore, TPA‐RPA demonstrated selective interaction with pathologically relevant metal ions (Cu²⁺ and Fe³⁺), suggesting its potential to attenuate metal‐mediated aggregation pathways [33, 34, 35]. Molecular docking and theoretical calculations provided mechanistic insights into the binding interactions with Aβ40, revealing stable and specific contacts that support its multifunctional role. The synthesized probes, TPA‐RPA and TPA‐ER, uniquely integrate AIE‐driven NIR fluorescence, selective Aβ recognition, mitochondrial targeting, metal ion modulation, and anti‐amyloid therapeutic activity within a single small‐molecule framework, an uncommon convergence compared to conventional amyloid probes that typically address only imaging or inhibition. Importantly, both molecules are synthesized via a single‐step Knoevenagel condensation, offering a rapid, high‐yielding, and scalable route, with straightforward purification. This synthetic simplicity, combined with multifunctional biological performance, distinguishes these probes from more complex multistep theranostic systems and enhances their translational potential. Together, these results position acceptor‐engineered TPA‐rhodanine probes, particularly TPA‐RPA, as promising chemical tools for the simultaneous imaging and modulation of amyloid pathology in AD (Scheme 1). This study not only expands the scope of small‐molecule theranostics but also highlights the potential of rational molecular engineering to address complex neurodegenerative diseases.

SCHEME 1.

Schematic illustration of the multifunctional TPA‐Rhodanine (TPA‐RPA and TPA‐ER) derivatives as far‐red to NIR donor–acceptor (D–A) probes for Alzheimer's theranostics. Left: Molecular design highlighting AIE‐active transformation and selective metal ion sensing. Middle: High‐affinity binding to Aβ40 monomers and fibrils enables inhibition and disaggregation of amyloid‐β aggregates with turn‐on fluorescence detection. Right: Effective mitochondrial targeting allows for bioimaging and mitigation of Aβ‐induced apoptosis, supporting both diagnostic and therapeutic applications. This schematic was created using ChemDraw and Microsoft PowerPoint.

2. Results and Discussion

2.1. The Rationale of the Design and Synthesis

Rational design forms the basis for integrating known functions from various molecular scaffolds into one unified entity. In this work, we adopted a donor‐π‐acceptor architecture to achieve targeted Aβ binding, metal ion chelation, mitochondrial localization, and fluorescence‐based imaging. The molecular scaffold incorporates a dimethoxy‐TPA unit as a strong electron donor, selected for its excellent electron‐rich character, photostability, and capacity to facilitate long‐wavelength emission via ICT, and the conjugation of a rhodan ine unit (ER and RPA) acts as an acceptor. This donor is conjugated through a π‐system to various acceptor moieties to fine‐tune the probe's photophysical and biological properties. These D‐π‐A fluorophores were synthesised via Knoevenagel condensation between dimethoxy‐diphenyl‐substituted p‐aminobenzaldehyde and the respective acceptors, yielding a set of NIR emissive small molecules with enhanced solvatochromic and AIE behavior. The designed molecules are tailored to (i) produce a “fluorescence turn‐on” response upon binding to Aβ aggregates, (ii) disrupt fibrillogenesis and disaggregate preformed Aβ fibrils, (iii) selectively accumulate in mitochondria, and (iv) interact with specific metal ions (Cu²⁺ and Fe³⁺) that are implicated in Aβ aggregation pathways. The single‐step synthesis, easy purification, and modular tunability of TPA‐RPA and TPA‐ER, combined with their integrated amyloid targeting, mitochondrial localization, and therapeutic functionality, clearly distinguish them from existing probes. This efficient design strategy delivers practical and scalable multifunctional theranostic candidates for translational AD applications.

2.2. Optical Properties of Synthesized Probes

The photophysical properties of these near‐infrared (NIR) probes were assessed in various solvents, including phosphate‐buffered saline (PBS) at 10 mm and pH 7.4. The absorbance spectra of TPA‐RPA and TPA‐ER, both at a concentration of 10 µm, showed absorbance between 460 and 500 nm (Figures S7 and S8) across different polarity solvents: THF, toluene, methanol, chloroform, DMF, DMSO, and water. The fluorescence properties of TPA‐ER and TPA‐RPA were systematically evaluated under various environmental conditions to assess their suitability as environmentally sensitive imaging probes. The emission spectra of the molecules were studied in different solvents. For TPA‐RPA, the emission ranged from 520 to 670 nm upon excitation at 450 nm (Figures S9 and S10). In contrast, TPA‐ER showed emission spanning 590–650 nm upon excitation at 475 nm (Figure S11). Both molecules exhibited characteristic AIE, as seen in DMSO‐water mixtures with increasing water content (Figure 1A,D), with maximum intensity at 80% water fraction for TPA‐RPA (50 µm) and 70% for TPA‐ER (10 µm). The concentration‐dependent study of the molecules was performed in water from 10–100 µm, and a gradual increase in the emission intensity was observed for both molecules, confirming the AIE property (Figures S12 and S13). The fluorescence intensity increased significantly above 60% water fraction, consistent with the restriction of intramolecular motions upon aggregation. The AIE study of TPA‐RPA was further investigated at a 10 µm concentration, and it was observed that the molecule exhibits a 50 nm shift in emission wavelength, indicating an AIEE nature (Figure S14). This AIE response is essential for biological applications, where aggregation in aqueous media or upon target binding can trigger fluorescence turn‐on. Table 1 presents the absorbance and emission maxima in the aggregated state and discusses the electronic properties of the molecules. Additionally, TPA‐RPA exhibited pronounced pH‐dependent behavior (Figure 1B), with strong emission at near‐neutral pH, indicative of potential protonation effects on electron‐withdrawing groups that influence the ICT process. In contrast, TPA‐ER exhibited relatively stable fluorescence across the pH range 2–13 (Figure 1F), suggesting its structural robustness for applications in diverse biological environments. The viscosity‐dependent emission in water‐glycerol mixtures (Figure 1C,E) revealed that both probes are sensitive to microenvironmental rigidity, with increased fluorescence intensity at higher glycerol content. This comprehensive analysis supports their potential for tracking changes in viscosity within cellular compartments, such as mitochondria or during amyloid aggregation. The fluorescence quantum yield (Φ_PL) measurements in water showed values of 2.21% for TPA‐RPA and 4.86% for TPA‐ER, indicating comparatively higher radiative efficiency for TPA‐ER in aqueous media (Figures S15 and S16). Time‐resolved photoluminescence (TRPL) studies further confirmed nanosecond‐scale excited‐state lifetimes for both probes: TPA‐RPA, 1.36 ns, and TPA‐ER, 1.49 ns, consistent with emissive singlet‐state relaxation pathways (Figures S17 and S18). These results, together with the moderate quantum yields, support the environment‐responsive fluorescence behavior of the D–A probes in aqueous conditions.

FIGURE 1.

Fluorescence emission spectra of TPA‐RPA (A–C, H–I) and TPA‐ER (D–G) under various environmental conditions. (A,D) AIE behavior in DMSO‐water mixtures with increasing water fractions (0%–99%). (B,F) pH‐dependent fluorescence response across a range of pH values (2–13). (C,E) Fluorescence intensity variations in water‐glycerol mixtures indicating viscosity sensitivity (0%–100%). (G) Metal ion interaction studies show fluorescence response of TPA‐ER (10 µm) upon addition of various biologically relevant metal ions (100 µm); no significant changes are observed in the presence of metal ions. (H) Fluorescence response of TPA‐RPA (50 µm) upon addition of various biologically relevant metal ions (100 µm), only Cu (II) and Fe (III) show a major intensity change in the emission spectra of TPA‐RPA. (I) Competitive metal ion binding assay with TPA‐RPA in the presence of Fe (III) at various concentrations (0–80 µm), quenching of fluorescent intensity was observed gradually. All measurements were performed at room temperature.

TABLE 1.

Overview of the photophysical and electronic properties of the synthesized AIEgen.

Probes	λabs (nm)	λex (nm)	λemm (nm)	HOMO (eV)	LUMO (eV)	ΔEg (eV)	ΦPL (%) (H2O)	τav (ns) (H2O)
TPA‐RPA	480	450	666	−5.04	−2.32	2.72	2.21	1.36
TPA‐ER	500	480	650	−4.99	−2.25	2.74	4.86	1.49

2.3. Theoretical Elucidation of Electronic Structure and Excited‐State Properties

Density functional theory (DFT) and time‐dependent density functional theory (TD‐DFT) calculations were performed to gain molecular‐level insight into the electronic structures and photophysical behavior of the synthesized probes. All calculations were carried out at the B3LYP/6‐311G(d,p) level using the Polarizable Continuum Model (PCM) with water as the solvent and the GD3 dispersion correction to explore the electronic structures of the synthesized probes [36]. Geometry optimization of both probes revealed energetically stable donor–acceptor architectures with twisted conformations, which favorably suppress detrimental π–π stacking while promoting aggregation‐induced emission. Frontier molecular orbital analysis showed that the highest occupied molecular orbital (HOMO) is primarily distributed over the triphenylamine donor framework, whereas the lowest unoccupied molecular orbital (LUMO) is mainly localized on the rhodanine acceptor segment, indicating effective donor‐to‐acceptor electronic communication. This spatial separation confirms an efficient ICT, with an energy gap (ΔE_g) of 2.72 eV for TPA‐RPA (Figure S19). This small band gap facilitates strong NIR emission and supports the probe's sensitivity to microenvironmental changes, critical for amyloid imaging and therapeutic action. Similarly, for TPA‐ER, the HOMO is delocalized over the donor region, whereas the LUMO extends onto the rhodamine acceptor. The calculated ΔE of 2.74 eV is comparable to that of TPA‐RPA, suggesting a similar ICT process (Figure S20). TD‐DFT calculations further revealed that the lowest‐energy singlet excited states (S₁) of TPA‐RPA and TPA‐ER are located at 2.2802 and 2.2883 eV. The lowest triplet states (T₁) were calculated at 1.6270 eV for TPA‐RPA and 1.6372 eV for TPA‐ER, giving singlet‐triplet energy gaps (ΔE_ST) of 0.653 and 0.651 eV, respectively (Tables S2 and S3). The comparable ΔE_ST values suggest similar excited‐state relaxation pathways and accessible intersystem crossing processes in both systems. Natural transition orbital (NTO) analysis provided a more rigorous description of the electronic excitations than frontier orbital inspection alone (Tables S4–S7) [37]. The dominant S₁ transitions in both probes exhibit hybrid local and charge‐transfer (HLCT) character, with the hole density remaining mainly on the electron‐rich triphenylamine segment while the electron density shifts toward the acceptor region (Figures S21 and S22) [38]. Such balanced HLCT behavior is advantageous for maintaining strong radiative transitions while preserving partial charge‐transfer character. Theoretical findings align with the observed far‐red emission, moderate fluorescence lifetimes, and responsive emission profiles of the probes. Strategic D–A engineering can modify the electronic structure of molecules, demonstrate their potential as multifunctional fluorescent probes, and highlight the role of acceptor engineering in optimizing their photophysical properties for the design of amyloid‐targeting agents.

2.4. Crystal Structure and Supramolecular Self‐Assembly Analysis

Single‐crystal x‐ray diffraction analysis shows that both TPA‐ER and TPA‐RPA form distinct supramolecular architectures stabilized by intermolecular interactions. TPA‐ER (Figure S23A–C) creates a highly ordered two‐dimensional lattice with a slipped head‐to‐tail arrangement, indicative of J‐type aggregation [39]. Its π‐extended donor–acceptor backbone promotes face‐to‐face π–π stacking, while C─H···π, O─H···O, and S─H···O hydrogen bonding enhance lateral packing. This results in directional columns along the crystallographic axis (Figure S23D), contributing to efficient exciton coupling and AIE‐enhanced emission. TPA‐RPA (Figure S24A–E) organizes into a three‐dimensional hydrogen‐bonding network with mixed J‐ and X‐type aggregation, which may be involved in the dual emission behavior. The assembly involves C─H···π, π–π, O─H···O, and S···H─C interactions, forming triangular or hexagonal clusters. The presence of both J‐type and cross‐stacked (X‐type) motifs leads to altered excitonic coupling, resulting in a broader emission band and a reduced fluorescence turn‐on response. The side‐view projections (Figure S24B,C) illustrate the hierarchical assembly of discrete molecules into extended supramolecular sheets. These findings confirm that the acceptor unit critically influences intermolecular packing, affecting aggregation mode, exciton delocalization, and emission output. The observed supramolecular architectures highlight the role of acceptor engineering in modulating self‐assembly, photophysics, and biological performance.

2.5. Metal Ion Selectivity of the Probes

The selective interaction of the synthesized TPA‐Rhodanine probes with metal ions like Zn²⁺, Cu²⁺, Ca²⁺, Mn²⁺, Fe²⁺, Al³⁺, and Fe³⁺ offers valuable insight into their potential role in modulating metal‐induced amyloid aggregation. Metal ions such as Cu²⁺, Zn²⁺, and Fe³⁺ are often found associated with β‐sheets of Aβ plaques in AD and are implicated in accelerating peptide aggregation and promoting oxidative stress [20, 34, 40]. The TPA‐ER does not show any interaction with various metal ions (Figure 1G); however, in the case of Al³⁺, a slight blue shift is observed in the emission spectra. TPA‐RPA exhibits differential binding affinities toward these biologically relevant metal ions Fe³⁺ and Cu²⁺, as evidenced by distinct changes in their spectroscopic signatures upon metal coordination (Figure 4H,I). The fluorescence titration of 50 µm TPA‐RPA with varying concentrations of Fe³⁺ revealed a limit of detection (LOD) of 9.2 µm, calculated using the 3σ/slope method (σ = 528.556; slope = 189.23) (Figure S25). This result demonstrates the good sensitivity of TPA‐RPA toward Fe³⁺ recognition, supporting its potential utility for monitoring biologically relevant metal‐ion‐mediated processes. The probes' metal‐ion selectivity suggests they may chelate specific ions involved in Aβ aggregation, thereby reducing metal‐mediated fibrillogenesis. This selective binding enhances both the probes' anti‐amyloidogenic activity and their diagnostic potential for detecting metal‐Aβ complexes, highlighting the multifunctional nature of these TPA‐based molecules in Alzheimer's theranostics.

FIGURE 4.

Molecular docking analysis of TPA‐rhodanine derivatives with Aβ40 fibrils (PDB ID: 2M4J). (A–C) Binding pose and interaction profile of TPA‐RPA within the Aβ40 fibril hydrophobic pocket. (A) Key interactions with bond distance visualized as 3D hydrogen bonds and π‐stacking with hydrophobic residues. (B) Surface hydrophobicity map showing favorable binding pocket occupancy by TPA‐RPA. (C) 2D interaction diagram summarizing hydrogen bonding (green), hydrophobic (pink), and π‐type (orange/yellow) interactions. (D–F) Binding interactions of TPA‐ER with Aβ40 fibrils. (D) 3D docking pose highlighting hydrogen bonding, π–π interactions, and other major interactions with bond distance. (E) Surface view of hydrophobic regions occupied by TPA‐ER. (F) 2D interaction map showing key binding residues and interaction types, including hydrogen bonding, π–sulfur, and alkyl contacts.

2.6. Spectroscopic Interaction and Selectivity Study With Aβ Species

The UV–vis absorption studies reveal significant molecular interactions between the probes and various Aβ species. As shown in Figure 2A,B, the absorbance spectra of TPA‐ER exhibit clear red shifts and intensity changes upon incremental addition of Aβ16‐21 and Aβ40‐fibrils (Aβ‐f), respectively, confirming dose‐dependent binding; at 450 nm, a clear isosbestic point was observed. The absorption of TPA‐RPA notably shifts upon binding to both Aβ16‐21 (KLVFFA, referred to as KA) and Aβ40 fibrils (Aβ‐f) (Figure 2C,D). This suggests a strong affinity for both amyloid components and potential endogenous ligands involved in AD pathology. Furthermore, fluorescence studies (Figure 2E,F) reveal a distinct increase in emission intensity with increasing concentrations of Aβ‐f and Aβ16‐21. Even at low concentrations, specifically with 10 µm TPA‐RPA and 30 µM Aβ‐f, there is an over 5‐fold increase in fluorescence intensity, accompanied by a blue shift, indicating a selective interaction with β‐sheet‐rich structures. In contrast, as the concentration of KA increases, the monomeric peak (at 530 nm) of the probe also increases, suggesting that the monomeric interaction occurs with the shorter peptide. For the full‐length Aβ, both the monomeric and aggregated peaks (530 and 670 nm) increase and are blueshifted. Notably, after 48 h of incubation, the aggregated peak disappears in the case of Aβ‐f, while the monomeric peak shows a slight increase (Figure S26). TPA‐RPA also interacts with Aβ40 at very low concentrations; as the concentration rises from 1.5 to 15 µm, clear changes in fluorescence intensity are observed, accompanied by a blue shift (Figure S27). Also, LOD was found to be 1.8 µm from this fluorescence study (Figure S28). The interaction of TPA‐ER with Aβ16‐21 (KA) and Aβ40‐f was also investigated, revealing that TPA‐ER exhibits minimal fluorescence quenching when interacting with both amyloid species (Figures S29 and S30). The selectivity of the probes was further evaluated in the presence of common biological proteins, including bovine serum albumin (BSA) and lysozyme. Negligible changes in fluorescence intensity were observed under these conditions, indicating minimal nonspecific protein interference (Figures S31–S34). These results highlight the TPA‐RPA's favorable selectivity toward its intended targets in complex biological environments. The TRPL decay of TPA‐ER in combination with Aβ40 at a 1:1 concentration was further studied; however, no significant changes in the lifetime were observed (Figure S35). In contrast, for TPA‐RPA with Aβ40 at the same 1:1 concentration, a significant increase in the lifetime was noted, rising from 1.36 to 1.52 ns (Figure S36). Overall, the interaction studies indicated that TPA‐RPA demonstrates superior binding and turn‐on fluorescence enhancement with amyloid species with specificity, making it particularly significant for imaging and dual‐mode detection in AD.

FIGURE 2.

Spectroscopic and binding studies of TPA‐ER and TPA‐RPA with Aβ species. (A,B) UV–vis absorption spectra of TPA‐ER (10 µm) in the presence of varying concentrations of Aβ16‐21 (A) and Aβ‐f (B), showing dose‐dependent spectral changes. (C) UV–vis absorption spectra of TPA‐RPA (10 µm) upon interaction with increasing concentrations of kynurenic acid (KA). (D) UV–vis absorption spectra of TPA‐RPA with incremental concentrations of Aβ‐Fibrils (Aβ‐f), indicating molecular interaction. (E–F) Fluorescence emission spectra of TPA‐RPA (10 µm) with increasing concentrations of Aβ‐f (E) and Aβ16‐21 (F), showing turn‐on fluorescence behavior. (G,H) Isothermal titration calorimetry (ITC) analysis of TPA‐RPA binding to Aβ, demonstrating sequential binding sites and associated thermodynamic parameters. (I) Thioflavin T (ThT) fluorescence assay assessing Aβ40 fibrillation kinetics in the absence and presence of TPA‐RPA and TPA‐ER, indicating inhibitory effects on Aβ aggregation. Data are presented as mean ± SD, n = 3.

2.7. Thermodynamic Binding Analysis of TPA‐RPA and TPA‐ER With Aβ40

To quantify the binding thermodynamics, isothermal titration calorimetry (ITC) experiments were performed, demonstrating that TPA‐RPA engages Aβ with high affinity through sequential binding sites. The derived binding constants and enthalpic contributions indicate a strong, favorable interaction, potentially driven by π–π stacking, hydrogen bonding, and hydrophobic interactions [17]. ITC is an excellent method that was used to investigate the activity of the probe under physiological conditions (pH 7.4, 37°C), revealing that Aβ40‐f (100 µm, pH 7.0, 37°C, 7 days) binds with TPA‐RPA (10 µm) in a sequential binding 2 mode, with a maximum binding constant (K) of 3.67 × 10⁷  m ⁻¹, an enthalpy change (ΔH) of −5.45 × 10⁵ cal/mol, and a Gibbs free energy change (ΔG) of −4.82 × 10² cal/mol. Hence, this reaction is exothermic in nature (Figure 1G). Whereas TPA‐ER (10 µM) binds with Aβ40‐f (100 µm, pH 7.0, 37°C, 7 days) with sequential binding 2 mode, with a maximum binding constant (K) of 1.13 × 10⁵  m ⁻¹, an enthalpy change (ΔH) of −8.44 × 10⁴ cal/mol, and a Gibbs free energy change (ΔG) of −7.5 × 10⁴ cal/mol (Figure 1H). The Gibbs free energy change for the probes indicates that binding between TPA‐RPA and TPA‐ER with Aβ40‐f is highly spontaneous and thermodynamically favorable, likely due to strong interactions such as hydrogen bonding, π–π stacking, and electrostatic interactions.

2.8. Therapeutic Effects: Inhibition and Disaggregation of Aβ40

Aβ40 fibrils are highly hydrophobic, which can be detected using the fluorescent probe Thioflavin‐T (ThT) [41]. The fluorescence intensity of ThT is significantly enhanced when it binds to hydrophobic regions on protein surfaces [42]. Therefore, the influence of TPA‐RPA on the hydrophobicity of Aβ can be reflected by the changes in ThT fluorescence. As shown in Figure 2I, the fluorescence intensity of ThT was used to kinetically study the Aβ aggregation process over a period of 4 days (96 h). The inhibitory effect on fibrillation was monitored through changes in ThT intensity. Notably, both probes completely inhibited the fibrillation process at a 1:1 concentration ratio during incubation. TPA‐ER showed an enhancement in ThT fluorescence; however, a suppression decrease was observed 24 h compared to ThT only. Overall, the ThT aggregation pathway exhibits a distinct kinetic pattern, with the elongation stage commencing after 24 h and the saturation stage commencing at 60 h (Figure 2I). The ThT kinetic assay demonstrated that Fe³⁺ markedly suppresses Aβ aggregation, as evidenced by a decrease in fluorescence intensity compared with Aβ alone. In contrast, the presence of TPA‐RPA completely suppresses the ThT intensity, suggesting effective inhibition of metal‐mediated aggregation kinetics. These results support the dual role of TPA‐RPA in modulating Fe³⁺‐associated amyloid assembly, likely through combined metal coordination and amyloid‐binding interactions (Figure S37). There was a remarkable suppression of Aβ40 fibrillation in both the Aβ and metal‐mediated Aβ conditions, with TPA‐RPA exhibiting a stronger inhibitory effect. We also checked the interaction between ThT fluorescence and both probes and found no interference with ThT fluorescence intensity in their presence (Figures S38 and S39). The fluorescence disaggregation study of both molecules was conducted at a 10 µm probe concentration at different time points. The ThT (10 µm) intensity for Aβ40 fibrils decreased significantly immediately upon addition of TPA‐RPA, and complete disaggregation was observed after 10 min (Figure S40). In contrast, for TPA‐ER, a significant decrease in ThT intensity was noted; however, complete disaggregation was not achieved after 10 min (Figure S41). Further, the ThT fluorescence spectra showed that Fe³⁺‐associated Aβ aggregates retained strong amyloid signals, whereas treatment with TPA‐RPA caused a pronounced decrease in ThT intensity after 5 min, accompanied by the emergence of the probe's far‐red emission band. This dual spectral response indicates effective disassembly or structural remodeling of metal‐stabilized β‐sheet‐rich aggregates into less ordered species (Figure S42). These findings demonstrate that TPA‐RPA can not only inhibit Fe³⁺‐mediated aggregation but also intervene in preformed metal‐associated amyloid assemblies. These results collectively highlight the therapeutic potential of the probes, especially TPA‐RPA, not only for Aβ detection but also for influencing pathogenic aggregation pathways and disaggregating preformed fibrils involved in the progression of AD.

2.9. Morphological Alteration of Aβ40

The therapeutic efficacy of the synthesized probes in modulating Aβ40 aggregation was evaluated by field‐emission transmission electron microscopy (FETEM), which revealed distinct morphological changes upon probe interaction. In the absence of any probe, Aβ40 existed as well‐dispersed monomers (Figure 3A) or matured into long fibrillar structures upon incubation for 5 days, characteristic of pathogenic aggregates (Figure 3B). Upon treatment with TPA‐ER (Figure 3C), amorphous nanostructures were observed, indicating self‐assembly of the probe. Notably, co‐incubation of TPA‐ER with Aβ (Figure 3D) led to the disruption of the typical fibrillar morphology, resulting in short, fragmented aggregates, which highlights its moderate anti‐aggregation ability. In contrast, TPA‐RPA (Figure 3E) formed well‐defined nanoaggregates on its own, and its co‐incubation with Aβ (Figure 3F) markedly suppressed fibril formation, yielding amorphous and disordered aggregates with monomers attaching to the surface of TPA‐RPA, suggestive of strong inhibition of Aβ fibrillogenesis. These findings support that TPA‐RPA demonstrates superior anti‐amyloidogenic activity, effectively interfering with Aβ aggregation and offering potential as a therapeutic candidate for mitigating Aβ‐associated neurotoxicity in AD.

FIGURE 3.

FETEM analysis illustrating the effect of TPA‐Rhodanine probes on Aβ40 aggregation. (A) Monomeric Aβ showing dispersed spherical morphology typical size between 2–3 nm. (B) Mature Aβ40 fibrils show characteristic long, thread‐like structures. (C) Self‐assembled nanostructures of TPA‐ER. (D) Aβ40 treated with TPA‐ER shows partial inhibition and disruption of fibril formation. (E) Self‐assembled structures of TPA‐RPA. (F) TPA‐RPA co‐incubated with Aβ40, revealing amorphous, non‐fibrillar aggregates indicative of strong fibrillogenesis inhibition. Insets highlight higher‐magnification views of the respective morphologies.

2.10. Molecular Docking Study

The molecular docking analysis elucidates the binding modes and affinities of two designed small‐molecule NIR probes, TPA‐RPA and TPA‐ER, toward the Aβ40 fibril, a key pathological entity implicated in AD. Both compounds successfully interacted with the β‐sheet‐rich hydrophobic cleft of the fibrillar structure, with TPA‐ER demonstrating a higher binding affinity (−7.23 kcal/mol) than TPA‐RPA (−6.13 kcal/mol). This enhanced binding is reflected in the lower inhibition constant (K_i) for TPA‐ER (4.99 µm), suggesting a stronger and more stable interaction than for TPA‐RPA (Ki = 32.21 µm), as shown in Table 2 and Tables S8 and S9.

TABLE 2.

Summary of the Molecular Interactions of TPA‐Rhodanine Probes with Aβ40 Fibrils (PDB ID: 2M4J) via Molecular Docking.

Probes	Highest binding energy (kcal/mol)	Major interactions	Interacting amino acids	Bond distance (Å)	Inhibition constant (Ki)
TPA‐RPA	−6.13	H‐bond	Lys 28, Leu 34	2.17, 1.98	32.21 µm
		π‐anion	Asp 23	3.39
		π‐alkyl	Ala 21	4.20
TPA‐ER	−7.23	H‐bond	Leu 34	2.57	4.99 µm
		π‐sigma	Gly 33	3.62
		Alkyl	Ile 31	4.78

For TPA‐RPA, the binding is primarily stabilized through hydrogen bonding with Lys28 and Leu34, along with π‐anion and π‐alkyl interactions involving Asp23 and Ala21, respectively. The short hydrogen bond distances (2.17 and 1.98 Å) indicate strong directional interactions that contribute to affinity (Figure 4A–C). In contrast, TPA‐ER forms a hydrogen bond with Leu34 (2.47 Å), a π‐sigma interaction with Gly33 (3.62 Å), and a pronounced alkyl interaction with Ile31 (4.78 Å) (Figure 4D–F). The involvement of multiple hydrophobic contacts with residues in the core region of Aβ40 likely contributes to the higher stabilization energy of the TPA‐ER complex. Hydrophobic surface analysis further supports these findings, showing that both ligands appropriately occupy the fibril's hydrophobic grooves and exhibit deeper penetration into the fibrillar pocket. The π‐stacking and non‐covalent interactions observed in both probes play a pivotal role in anchoring them to the fibril surface, thereby potentially disrupting aggregation or stabilizing intermediate peptide forms.

2.11. Biocompatibility and Neuroprotective Efficacy of TPA‐ER and TPA‐RPA Against Aβ‐Induced Toxicity

The evaluation of cytotoxicity and neuroprotective potential of TPA‐based small molecules was performed to identify suitable candidates for further development in AD therapy. Compounds were initially screened for their intrinsic cytotoxicity using an MTT assay in SH‐SY5Y neuronal cells across a broad concentration range (1–100 µm) [43]. TPA‐RPA (Figure 5A) and TPA‐ER (Figure 5B) showed no significant cytotoxicity up to 100 µm, indicating excellent biocompatibility. Also, both molecules show no toxicity in the IMR32 cell line (Figures S43 and S44). These molecules are therefore ideal candidates for further assessment of neuroprotective efficacy. Notably, the consistent cell viability observed with these compounds suggests a favorable safety margin and highlights the importance of structural tuning in the development of TPA‐based molecules. Although mature fibrils are hallmarks of amyloid pathology, increasing evidence suggests that soluble oligomeric intermediates are the more potent neurotoxic species. Accordingly, the biological effects observed here likely reflect the heterogeneous nature of the aggregated Aβ preparations. The neurotoxicity of Aβ40 aggregates (Figure 5C) was validated in a dose‐dependent manner, with increasing fibril concentrations resulting in a sharp reduction in SH‐SY5Y cell viability, confirming the well‐established cytotoxic effects of aggregated Aβ species in neuronal models of AD. Both probes maintained comparatively high cell viability over the 10–100 µm range, with only a modest reduction at the highest tested concentration. Among them, TPA‐RPA exhibited a more favorable cytocompatibility profile and lower oxidative stress induction, supporting its selection as the lead candidate for further biological evaluation.

FIGURE 5.

Cytotoxicity and neuroprotection assessment of TPA‐RPA and TPA‐ER against Aβ40‐fibril‐induced toxicity. (A,B) Cell viability of SH‐SY5Y neuroblastoma cells treated with varying concentrations (10–100 µm) of TPA‐RPA (A) and TPA‐ER (B), showing low inherent cytotoxicity. (C) Dose‐dependent cytotoxicity of Aβ40 fibrils (Aβ40‐f) on SH‐SY5Y cells. (D) Protective effects of TPA‐RPA and TPA‐ER against Aβ40‐f in co‐treatment in 10 and 20 µm, demonstrating attenuation of Aβ‐induced cytotoxicity. (E) Bar graph showing the fold change in intracellular ROS generation for untreated, Aβ‐treated, TPA‐RPA‐treated, TPA‐ER‐treated, Aβ+TPA‐RPA‐treated, and Aβ+TPA‐ER‐treated cells. Statistical analysis was performed using one‐way analysis of variance (ANOVA) followed by Tukey's multiple‐comparison test in GraphPad Prism. The experimental data were expressed as mean ± SEM. Sample size, n = 3. The p‐value < 0.05 was considered statistically significant, where ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001, respectively; ns = not significant.

Co‐incubation of TPA‐RPA and TPA‐ER with Aβ40‐f at 10 and 20 µm significantly restored cell viability compared with cells treated only with Aβ, indicating that these molecules can reduce Aβ40‐induced cytotoxicity (Figure 5D). TPA‐ER showed the highest neuroprotective efficacy, indicating its favorable structure for interacting with Aβ aggregates, modulating aggregation, or enhancing antioxidant defenses. TPA‐RPA also provided strong protection, suggesting that these molecules may function as anti‐amyloid agents by interfering with Aβ fibrillization or by stabilizing cellular stress responses. Although mild intrinsic cellular stress may occur at higher probe concentrations, co‐treatment experiments demonstrated significant recovery of cell viability in Aβ‐treated cells, indicating that the net biological outcome of the probes under amyloid stress conditions is protective rather than detrimental. The differing protective effects of these TPA derivatives likely result from their specific structural features, such as donor–acceptor character and molecular planarity, which affect their interactions with Aβ aggregates and cellular targets. Their potential photophysical properties may also support dual imaging and therapeutic applications for real‐time efficacy monitoring in cellular models.

2.12. Modulation of Aβ‐Mediated Oxidative Stress by TPA‐RPA and TPA‐ER

To evaluate the potential of TPA‐RPA and TPA‐ER to mitigate the ROS‐generating ability of Aβ40 fibrils in diseased neuronal cells, an ROS generation assay was performed. Cellular ROS levels were detected using 2′,7′‐dichlorodihydrofluorescein diacetate (DCFDA)‐based flow cytometry analysis. The results showed that after 6 h of Aβ‐f treatment, cellular ROS levels increased to almost 3.5‐fold in SH‐SY5Y cells compared to untreated cells. On the other hand, TPA‐RPA and TPA‐ER treatment increased cellular ROS levels by 1.2‐ and 2.2‐fold, respectively, compared to untreated cells. Whereas, upon co‐treatment with TPA‐RPA and TPA‐ER along with Aβ‐f, the cellular ROS level decreased to 1.1‐ and 1.4‐fold compared to untreated cells, suggesting a strong potential of the compound to inhibit the ROS‐generating capacity of toxic Aβ aggregates within affected cells (Figure 5E).

2.13. Regulation of Mitochondrial Membrane Potential Under Aβ40‐f

To evaluate the impact of the designed molecules on mitochondrial membrane potential (ΔΨm), a Rhodamine 123‐based fluorescence assay was performed [44]. Rhodamine 123 is a well‐established cationic dye that accumulates in active mitochondria in proportion to ΔΨm [27, 28]. As shown in Figure 6, treatment with Aβ40 alone resulted in a significant reduction in Rhodamine 123 fluorescence, indicating mitochondrial depolarization and compromised ΔΨm, consistent with Aβ‐induced mitochondrial dysfunction. Notably, co‐incubation of Aβ40 with TPA‐ER and TPA‐RPA restored fluorescence intensity to levels close to those of untreated control cells, suggesting effective preservation of mitochondrial membrane integrity. Probe treatment elicited distinct intracellular ROS responses, with TPA‐ER showing a greater increase than TPA‐RPA. However, this was not associated with significant mitochondrial depolarization or loss of viability, indicating that oxidative stress remained within a manageable range under the experimental conditions. These findings highlight the probes' ability to mitigate Aβ‐induced mitochondrial toxicity, possibly by inhibiting Aβ40 aggregation and preventing mitochondrial damage [45]. This protective effect reinforces the potential of the synthesized molecules as therapeutic candidates targeting mitochondrial dysfunction in AD.

FIGURE 6.

Flow cytometry analysis of mitochondrial membrane potential (ΔΨm) using Rhodamine 123 in SH‐SY5Y cells. Untreated cells, TPA‐RPA, and TPA‐ER treatments maintain normal ΔΨm, as evidenced by sharp peaks at high fluorescence intensity. Aβ40 treatment causes a left‐shifted, broadened distribution, indicative of mitochondrial depolarization. Co‐incubation with TPA‐RPA or TPA‐ER markedly restores Rhodamine 123 fluorescence intensity, demonstrating their protective role against Aβ40‐induced mitochondrial dysfunction.

2.14. Mitigation of Aβ‐Induced Apoptotic Cell Death by TPA‐RPA and TPA‐ER

Flow cytometric analysis using Annexin V‐FITC and propidium iodide staining revealed distinct apoptotic profiles in SH‐SY5Y cells under different treatment conditions. Amyloid‐β (Aβ) treatment markedly increased early and late apoptotic populations (Q3 and Q2), with a significant shift from the viable quadrant (Q4) compared to untreated cells, confirming its neurotoxic effects. In contrast, cells treated with TPA‐ER alone showed minimal apoptosis, indicating low intrinsic cytotoxicity. Notably, co‐treatment with TPA‐ER in the presence of Aβ substantially rescued the cells from apoptosis, as evidenced by a pronounced increase in viable populations and reduction in apoptotic cells compared to Aβ alone (Figure 7). This protective effect was prominent with TPA‐ER, suggesting enhanced neuroprotection, possibly due to better interaction with Aβ aggregates or improved subcellular localization. These findings support the potential of TPA‐based molecules as therapeutic candidates for mitigating Aβ‐induced neuronal damage in AD.

FIGURE 7.

Flow cytometric analysis of SH‐SY5Y cell apoptosis after various treatments using Annexin V‐FITC/Propidium Iodide (PI) staining. Cells were treated with Amyloid‐β (Aβ), TPA‐ER, and their combinations, followed by dual staining to differentiate between live (Annexin V⁻/PI⁻, Q4), early apoptotic (Annexin V⁺/PI⁻, Q3), late apoptotic/necrotic (Annexin V⁺/PI⁺, Q2), and necrotic (Annexin V⁻/PI⁺, Q1) populations. Untreated cells served as a control. Aβ treatment significantly increased apoptotic populations, while co‐treatment with TPA‐ER mitigated Aβ‐induced apoptosis, indicating their neuroprotective potential. Percentages in each quadrant represent the proportion of cells in the respective states.

Poly(ADP‐ribose) polymerase (PARP) cleavage is a hallmark of apoptosis, typically activated by cellular stress induced by amyloid toxicity. In this study, Western blot analysis revealed a pronounced band for cleaved PARP in Aβ‐treated SH‐SY5Y cells, indicating strong apoptotic signaling. In contrast, cells treated with TPA‐RPA alone showed no detectable cleaved PARP expression, suggesting that the probe itself is non‐toxic and does not trigger apoptotic pathways. Strikingly, co‐treatment of TPA‐RPA with Aβ substantially suppressed PARP cleavage, restoring expression levels comparable to untreated control cells (Figure 8). These results clearly demonstrate that TPA‐RPA protects neuronal cells from Aβ‐induced apoptosis, likely by interfering with early aggregation processes and alleviating downstream mitochondrial dysfunction and caspase activation. Together with the MMP assays, these findings further validate the therapeutic potential of TPA‐RPA as a dual‐functional probe that mitigates Aβ‐driven cytotoxicity while maintaining neuronal homeostasis.

FIGURE 8.

Western blot analysis of cleaved PARP expression in SH‐SY5Y cells. Aβ treatment induced strong PARP cleavage, whereas TPA‐RPA alone showed no effect. Co‐treatment with TPA‐RPA markedly reduced Aβ‐induced PARP cleavage, indicating protection against apoptotic damage. β‐actin served as a loading control.

2.15. Mitochondrial Colocalization Study of TPA‐RPA and TPA‐ER

To examine intracellular uptake and organelle‐specific localization, confocal fluorescence imaging was performed in SH‐SY5Y neuronal cells using two commercial mitochondrial markers, MitoTracker Green and Rhodamine 123 (Figure 9). Both probes showed efficient cellular internalization and clear intracellular fluorescence, confirming their suitability for live‐cell imaging. In the presence of MitoTracker Green, TPA‐RPA displayed a well‐distributed intracellular fluorescence pattern with substantial signal overlap in the merged images, indicating appreciable mitochondrial association, with a Pearson correlation coefficient (PCC) of 0.85 [46]. TPA‐ER also exhibited distinct punctate fluorescence with significant colocalization, supporting preferential mitochondrial accumulation. Similar trends were observed in the independent co‐staining experiment with Rhodamine 123, where both probes showed clear merged yellow fluorescence, further validating mitochondrial localization.

FIGURE 9.

Cellular co‐localization and confocal fluorescence imaging of TPA‐RPA and TPA‐ER. Representative bright‐field, green channel (Rhodamine 123 and Mitotracker Green), red channel (TPA‐based probes), and merged (bright and dark field) images of cells treated with TPA‐RPA and TPA‐ER (scale bar: 20 µm). Rhodamine 123 and Mitotracker Green, a mitochondria‐specific dye, were used to assess co‐localization and intracellular distribution. Both compounds exhibit distinct red fluorescence signals and significant overlap with Rhodamine 123 and Mitotracker Green in the merged channels, indicating efficient mitochondrial targeting. Confocal microscope image analysis and PCC value calculation using ImageJ.

Notably, for Rhodamine 123 colocalization, TPA‐ER demonstrated stronger, more defined mitochondrial staining than TPA‐RPA, consistent with a PCC of 0.87, both of which are favorable for organelle targeting. In contrast, TPA‐RPA showed moderate but significant mitochondrial affinity, along with a more diffuse intracellular distribution, in both models. Overall, these results demonstrate that both AIEgens possess mitochondrial targeting capability, while subtle structural modification of the terminal functionality effectively modulates localization efficiency. Such tunable organelle selectivity is advantageous for developing multifunctional probes for mitochondrial imaging and AD‐related theranostic studies.

2.16. In Silico Evaluation of BBB Permeability, Toxicity, and Target Prediction

To preliminarily assess the translational potential of the synthesized probes for neurodegenerative applications, comprehensive in silico analyses were performed using multiple predictive platforms, including Log_BBB‐Pred, ADMET, Swiss Target Prediction, and ProTox‐3.0 (Tables S10–S15). Both TPA‐RPA and TPA‐ER exhibited physicochemical parameters within the acceptable range for central nervous system‐active small molecules and were predicted to possess favorable blood‐brain barrier (BBB) permeability, supporting their potential to access brain tissue (Figures S45 and S46) [47, 48, 49, 50]. Toxicity profiling further suggested low acute toxicity tendencies with no major alerts for mutagenicity, hepatotoxicity, or severe organ toxicity, indicating an encouraging preliminary safety profile. In addition, target prediction analyses indicated probable interactions with biologically relevant protein classes, including enzymes, kinases, oxidoreductases, and membrane‐associated receptors, which may be associated with pathways implicated in amyloid pathology, oxidative stress, and neuronal dysfunction. Consistent with the in vitro findings, in silico ADMET and toxicity predictions indicated a favorable safety profile for TPA‐RPA, with no major toxicity alerts and suitable drug‐like characteristics. Collectively, these computational findings support the drug‐like characteristics of the TPA‐RPA and reinforce their promise as multifunctional candidates for AD‐related diagnostic and therapeutic applications.

Taken together, these results highlight the strong impact of subtle structural modification on intracellular localization and biological performance. Both AIEgens exhibited mitochondrial affinity and effective cellular uptake; however, TPA‐RPA emerged as the more balanced lead candidate, combining favorable biocompatibility, appreciable mitochondrial localization, dual‐channel fluorescence response, and significant protection against Aβ‐induced cellular damage. These multifunctional features position TPA‐RPA as a promising platform for mitochondria‐directed image‐guided intervention in AD, where mitochondrial dysfunction and amyloid pathology are closely interconnected (Scheme 2).

SCHEME 2.

Schematic illustration of the multifunctional far‐red‐to‐NIR‐emissive AIEE probe for AD theranostics. The AIEEgen TPA‐RPA binds selectively to Aβ40 aggregates, inhibits fibrillization, and disassembles toxic assemblies while chelating metal ions implicated in the aggregation process. Mitochondrial localization enables suppression of Aβ‐induced ROS generation and membrane depolarization, thereby preventing apoptosis. The system integrates detection, imaging, and therapeutic modulation into a single molecular platform. This schematic illustration was created using Microsoft PowerPoint.

3. Conclusion

In this work, two rhodanine‐based donor–acceptor fluorescent probes, TPA‐RPA and TPA‐ER, were rationally designed to exhibit AIE properties for targeted theranostics in AD. TPA‐RPA displayed stronger binding to Aβ fibrils (exothermic ITC interaction, K_d in the low µM range) and effectively inhibited and disaggregated Aβ40 aggregates, outperforming TPA‐ER. Confocal studies confirmed high mitochondrial accumulation (PCC = 0.87), enabling regulation of the ROS‐MMP pathway and suppression of Aβ‐induced apoptosis, as validated by Annexin V‐FITC/PI and western blot assays. In neuronal cells, both probes rescued Aβ‐mediated cytotoxicity, while flow cytometry validated their ability to suppress mitochondrial‐dependent apoptotic pathways. By integrating selective Aβ recognition, mitochondrial targeting, anti‐aggregation activity, ROS modulation, and fluorescence imaging, TPA‐RPA in particular represents a multifunctional AIEE‐active theranostic candidate for early diagnosis and intervention in AD.

4. Experimental Section

4.1. Synthesis Procedure of the AIEgens

4.1.1. Synthesis of TPA‐ER

(E)‐5‐(4‐(bis(4‐methoxyphenyl)amino)benzylidene)‐3‐ethyl‐2‐thioxothiazolidin‐4‐one (TPA‐ER) was synthesized from 4‐(bis(4‐methoxyphenyl)amino)benzaldehyde (266 mg, 0.8 mmol) and 3‐ethyl‐2‐sulfanylidene‐1,3‐thiazolidin‐4‐one (161 mg, 1 mmol) in ethanol (10 mL). The mixture was heated at 80 °C for 12 h. Piperidine (2 drops) was added to the reaction mixture after both solubilized (Scheme 3). After a few hours, a bright orange precipitate is obtained. The crude product was filtered through Whatman filter paper, washed several times with cold ethanol, and recrystallized from CHCl₃ to obtain the pure product. The bright orange solid of TPA‐ER was obtained with a yield of 80% (Figure S1–S3).

SCHEME 3.

Synthesis procedure of the TPA‐Rhodanine derivatives.

¹H NMR (600 MHz, CDCl₃) δ 7.63 (s, 1H), 7.27 (t, J = 4.2 Hz, 3H), 7.11 (d, J = 8.8 Hz, 4H), 6.87 (dd, J = 14.4, 8.8 Hz, 6H), 4.41 (t, J = 7.6 Hz, 2H), 3.82 (s, 6H), 2.80 (t, J = 7.6 Hz, 2H).

¹³C NMR (151 MHz, CDCl₃) δ 193.09, 167.86, 157.24, 151.15, 138.86, 133.57, 132.50, 127.84, 123.95, 117.94, 117.73, 115.06, 55.52, 39.7, 12.27.

MALSI MS: 476.332

4.1.2. Synthesis of TPA‐RPA

(E)‐3‐(5‐(4‐(bis(4‐methoxyphenyl)amino)benzylidene)‐4‐oxo‐2‐thioxothiazolidin‐3‐yl) propanoic acid (TPA‐RPA) was synthesized from 4‐(bis(4‐methoxyphenyl)amino)benzaldehyde (233 mg, 0.7 mmol) and 3‐(4‐oxo‐2‐sulfanylidene‐1,3‐thiazolidin‐3‐yl)propanoic acid (185 mg, 0.9 mmol) in ethanol (10 mL). A few drops of piperidine were added to the reaction mixture, and it was heated at 80 °C for 12 h. After a few hours, the reaction turns red. The reaction was monitored by TLC. The crude product was dried under vacuum and purified by column chromatography, eluting with 5% methanol in chloroform. The deep red TPA‐RPA solid was obtained with a yield of 72% (Figures S4–S6).

¹H NMR (600 MHz, CDCl₃) δ 7.63 (s, 1H), 7.28 (d, J = 8.5 Hz, 2H), 7.12 (d, J = 8.5 Hz, 4H), 6.88 (t, J = 9.3 Hz, 6H), 4.19 (d, J = 7.1 Hz, 2H), 3.82 (s, 6H), 1.28 (t, J = 7.1 Hz, 3H).

¹³C NMR (151 MHz, CDCl₃) δ 193.35, 168.21, 157.73, 151.76, 139.24, 134.74, 133.12, 128.34, 118.34, 117.49, 115.54, 55.99, 40.15, 31.65.

MALDI MS: 520.339

4.2. Preparation of Aβ40 Monomer and Fibrils

To remove preexisting aggregates, Aβ40 was treated with Hexafluoro‐2‐propanol (HFIP) and Trifluoroacetic acid (TFA) as per the reported protocol. In brief, 1.0 mg Aβ40 was treated with 10 µL TFA and dried in argon. The solution was again treated with 10 µL HFIP and dried in argon. To the solution, 1 mL of 10 mm PBS was added, and the mixture was sonicated for 10 min to obtain a 230 µm Aβ40 monomer stock. To prepare Aβ40 fibrils, the monomeric solution was dissolved in NaOH aqueous solution (20 mm) and then diluted to 20 µm in PBS buffer and incubated at 37°C for 7 days. The fibrils were confirmed by FETEM analysis. Aβ40 samples were prepared under fibrillization conditions; however, partial coexistence of oligomeric or protofibrillar intermediates cannot be excluded and may contribute to the observed biological responses.

4.3. Thioflavin‐T Assay

The ThT fluorescence assay was performed to evaluate the inhibition kinetics of Aβ40 aggregates, specifically focusing on fibrils. A 10 µm solution of Aβ40 monomer was incubated in a 10 mm PBS buffer at 37°C for a total of 96 h. This incubation occurred both in the presence of 0.1 equivalents of the synthesized probes and in their absence as a control. Aliquots were collected at specified time intervals throughout the 96‐h period, beginning at the initial time point. To each sample, 20 µm ThT was added, and the samples were excited at 440 nm. The emitted fluorescence was recorded at 480 nm, and the emission intensity was plotted over time to analyze the inhibition kinetics.

4.4. DFT Study

Vertical and transition energies, including the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) for single‐component crystals, were calculated using density functional theory (DFT). All DFT calculations were performed using the Becke3‐Lee‐Yang‐Parr (B3LYP) hybrid functional with the 6–31G(d,p) basis set, using the Gaussian 16 software package. Images were made using GaussView software. The calculated structures and energy levels were found to be in close agreement with experimental results. The time‐dependent DFT (TDDFT) and Natural Transition Orbital (NTO) calculations in this study were conducted in B3LYP/6‐31G(d,p), using the PCM solvation model and GD3 aberration correction in the Gaussian 16 software package.

4.5. Molecular Docking Study

Molecular docking studies and imaging were conducted using AutoDock Tools 1.5.7 and BIOVIA Discovery Studio 2021. The grid parameters for TPA‐ER and TPA‐RPA specify 40 grid points in each direction (x, y, z) with a grid spacing of 0.508 Å. The number of distinct conformational clusters identified is 10 for TPA‐RPA and 4 for TPA‐ER, based on 20 runs with an RMSD tolerance of 2.0 Å. Additionally, the number of multi‐member conformational clusters found is 4 for TPA‐RPA and 3 for TPA‐ER, also from 20 runs.

Molecular Docking parameters for TPA‐RPA and TPA‐ER:

• Grid Point Spacing = 0.508 Å

• Even Number of User‐specified Grid Points = 40 x‐points; 40 y‐points; 40 z‐points

• Coordinates of Central Grid Point of Maps = (13.206, ‐5.556, 105.273)

• Minimum coordinates in grid = (3.046, ‐15.716, 95.113)

• Maximum coordinates in grid = (23.366, 4.604, 115.433)

4.6. Toxicity Assay

The MTT assay is a colorimetric technique used to evaluate the cytotoxic effects of medicinal compounds by measuring the activity of NAD(P)H‐dependent oxidoreductase enzymes in viable cells. This assay involves reducing the yellow tetrazolium dye MTT to insoluble purple formazan crystals. To analyze these crystals, a solubilization solution such as dimethyl sulfoxide (DMSO) is added, dissolving the formazan and producing a colored solution. The absorbance of this solution is measured at 570 nm using a UV–vis spectrophotometer, correlating with formazan concentration. In our experiment, we performed the MTT assay using SH‐SY5Y and IMR‐32 human neuroblastoma cell lines cultured in DMEM with 10% fetal bovine serum. The cells were seeded in a 96‐well plate, treated with TPA‐ER and TPA‐RPA at concentrations of 0 to 100 µm, and incubated for 48 h. We then added 10 µL of 5 mg/mL MTT solution and incubated for 2 h. The medium was replaced with DMSO to dissolve the formazan crystals, and the absorbance was measured at 570 nm, with readings at 630 nm for reference. Each sample was tested in triplicate, and results were normalized to the control viability of SH‐SY5Y cells in PBS, set at 100%. MTT graphs, fold change graphs, and statistical analysis conducted using GraphPad Prism.

4.7. Colocalization Study

To investigate cellular uptake, we seeded 1.5 × 10⁵ SH‐SY5Y neuroblastoma cells into 35 mm Petri dishes and incubated them for 24 h to ensure proper attachment. We then treated the cells with TPA‐ER and TPA‐RPA for 6 h, monitoring the uptake of each compound. After treatment, the cells were stained with Rhodamine 123 and Mitotracker Green, a fluorescent dye for mitochondrial visualization, according to the manufacturer's protocol, and incubated for an additional 1 h. Further, Mito‐tracker green dye was used in the same procedure for mitochondrial visualization. Images were captured using a confocal microscope, enabling us to compare compound uptake between treated and untreated groups. Confocal microscope image analysis and PCC value calculation using ImageJ software.

4.8. Measurement of ROS

The generation of ROS is a normal process in healthy cells. ROS production can be initiated by various factors, including photochemical reactions, auto‐oxidation, and several enzymes such as those involved in mitochondrial respiration and NADPH oxidase activity. The cell typically maintains a balance by employing detoxification mechanisms to manage ROS levels. However, oxidative stress occurs when there is a disruption between ROS production and the cell's ability to neutralize them, potentially leading to transient or permanent oxidative damage to cellular components, including membrane lipids, proteins, and nucleic acids. For the detection of ROS generation, 1.5×105 cells were seeded per 35 mm tissue culture petri‐plate and, upon attachment, were treated with Aβ and Aβ+Probe (TPA‐ER/TPA‐RPA) for 6 h. Then, the cells were incubated with 10 µm DCFDA for 30 min. After washing, the trypsinized cells were analyzed by flow cytometry.

4.9. Mitochondrial Membrane Potential (MMP) Assay

Rhodamine‐123 is a cell‐permeant, cationic, green‐fluorescent dye that effectively indicates MMP without causing cytotoxicity. In this study, human neuroblastoma cells (SH‐SY‐5Y) were cultured in DMEM (Dulbecco's Modified Eagle's Medium) at 37°C with 5% CO₂ and seeded in six‐well plates at a density of 1×10⁵ cells per well, with 10% fetal bovine serum (FBS). After 24 h, the media was replaced, and wells were washed with PBS. Aβ and Aβ+Probe (TPA‐ER/TPA‐RPA) were then added at 20 µm for another 48 h. Following this, 10 µm Rhodamine 123 was introduced and incubated for 1 h. Cells were trypsinized, washed with PBS, and were analyzed using flow cytometry to observe the decline of green fluorescence and the diffusion of non‐fluorescent material into the cytoplasm. The data were analyzed using FlowJo software.

4.10. Apoptosis Assay

To assess apoptosis induced by Aβ40 aggregates and the potential protective effects of TPA‐ER, an Annexin V‐FITC/Propidium Iodide (PI) dual staining assay was conducted on SH‐SY5Y neuroblastoma cells. After seeding and allowing the cells to adhere overnight, treatments were applied for 48 h, including: (i) untreated control, (ii) Aβ40 fibril treatment, and (iii) co‐treatment with Aβ40 and TPA‐ER. After incubation, cells were washed, resuspended in binding buffer, and stained using a commercial apoptosis detection kit. Following a 15‐min incubation in the dark, samples were analyzed using a BD FACS Melody flow cytometer, recording 10 000 events per sample. Data analysis with FlowJo (version V10.8.1) facilitated the distinction of viable, early apoptotic, late apoptotic, and necrotic cell populations, allowing for precise quantification of apoptosis under each treatment, and validating the anti‐apoptotic effects of TPA‐based probes against Aβ‐induced neurotoxicity.

4.10.1. Western Blot Analysis of Apoptosis Marker

Proteins were isolated from SH‐SY5Y cells treated with TPA‐RPA or left untreated using 1X RIPA lysis buffer. Protein concentrations were determined using the BCA (Bicinchoninic Acid) reagent, and equivalent amounts were loaded onto an SDS (Sodium Dodecyl Sulfate) gel for electrophoresis. Proteins were transferred to a nitrocellulose membrane, which was then blocked with 5% BSA (Bovine Serum Albumin) in PBST (Phosphate‐Buffered Saline with Tween‐20). Primary antibodies for cleaved PARP (poly(ADP‐ribose) polymerase) and β‐actin were added overnight at 4°C, followed by washes and incubation with secondary antibodies for 2 h. Bands were developed with a chemiluminescent reagent and captured using the Bio‐Rad ChemiDoc system.

4.11. Statistical Analysis

Statistical analysis was performed using one‐way analysis of variance (ANOVA) followed by Tukey's multiple comparison test using GraphPad Prism. The experimental data were expressed as mean ± SEM. The p‐value < 0.05 was considered statistically significant, where ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001, respectively.

4.12. In Silico Prediction of BBB Permeability, ADMET Properties, Toxicity, and Target Identification

The pharmacokinetic and drug‐likeness profiles of the synthesized probes were evaluated using freely accessible computational platforms. The chemical structures of TPA‐RPA and TPA‐ER were first converted into canonical SMILES format and used as input for all predictive analyses. Blood‐brain barrier (BBB) permeability was assessed using the LogBB Pred platform to estimate the likelihood of brain penetration. General absorption, distribution, metabolism, excretion, and toxicity (ADMET) parameters were predicted using the ADMET web server to examine physicochemical suitability, oral drug‐likeness, and pharmacokinetic behavior. Toxicological properties, including acute toxicity class and organ toxicity risks, were estimated using ProTox‐3.0. Potential biological targets were identified using SwissTargetPrediction, where the probability of interaction with human protein targets was calculated based on structural similarity and known ligand‐target relationships. All analyses were performed using default settings provided by each platform. The combined computational outputs were used to assess the suitability of the probes as brain‐accessible multifunctional small molecules for AD‐related applications.

Author Contributions

P.G. and P.K.I. designed schemes and experiments; P.G. synthesized compounds; P.G. and A.S. performed characterization of the molecules, photophysical experiments, and kinetic study; S.M. and S.S.G. performed cell‐culture experiments; P.G. performed biophysical experiments, in silico study including molecular docking; P.G. compiled and analyzed data, made schemes and figures, and wrote the manuscript. All authors reviewed and approved the final version of the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adhm71341‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. CCDC 2455935 and 2500998 contain the supplementary crystallographic data for this paper (Table S1). These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.