Nanoparticles that target nonamyloid and nontau pathways in Alzheimer’s disease Authors: Firoozeh Alavian, Arefeh Hajimohammadi Published: 2026 ABSTRACT Alzheimer’s disease (AD) is a complex neurodegenerative disorder whose progression involves multiple pathways beyond the canonical amyloid and tau cascades. Neuroinflammation, mitochondrial dysfunction, and lysosomal impairment represent key nonamyloid and nontau pathways; preclinical evidence suggests that targeting these pathways may aid in the development of more effective treatments, although clinical validation remains pending. Owing to their ability to cross the blood‒brain barrier and their potential for precise targeting, nanostructured materials represent promising tools for modulating these pathways in preclinical models. Lipid, chitosan, and gold nanoparticles, when employed as carriers of anti-inflammatory and antioxidant compounds such as curcumin and resveratrol, have been shown in animal studies to reduce neuroinflammation and improve mitochondrial function. NPs functionalized with ligands such as triphenylphosphonium specifically target mitochondria, reducing oxidative stress and increasing ATP production by increasing drug bioavailability. Polymeric and carbon nanostructures improve lysosomal function and restore cellular homeostasis. These technologies slow disease progression by reducing neuroinflammation, improving mitochondrial dynamics, and enhancing autophagy processes. This article provides a strictly narrative review of recent advances in the use of nanostructured materials for targeting nonamyloid and nontau pathways in AD and to examine the therapeutic potential of this technology in the development of effective strategies to combat this disease. Introduction AD is the most common form of dementia; it primarily affects elderly individuals and is characterized by progressive neuronal loss. As a result of rising global life expectancy, AD has emerged as one of the most significant health challenges of the twenty-first century [1]. If the age-specific incidence rate of Alzheimer’s disease and related dementia continues to grow at the pace observed over the past decade, it is projected that 115.8 million disability-adjusted life years could be lost annually by 2050. For decades, AD research has been largely shaped by the "amyloid cascade" and "tau hyperphosphorylation" hypotheses, which have guided therapeutic development. However, the repeated failure of clinical trials targeting amyloid-beta (Aβ) and tau proteins underscores the multifactorial nature of the disease and highlights the urgent need to explore alternative pathological mechanisms. The evidence indicates that the NANT pathway plays important roles in the development of AD, including neuroinflammation, mitochondrial dysfunction, lysosomal defects, endoplasmic reticulum (ER) stress, synaptic damage, and neurovascular impairments. These processes often occur before the formation of amyloid plaques and neurofibrillary tangles, indicating their involvement in the early stages of the disease. Chronic neuroinflammation caused by microglial activation and the astrocyte response leads to neuronal damage through the secretion of inflammatory cytokines and reactive oxygen species (ROS) [2]. Similarly, mitochondrial dysfunction leads to energy deficiency and oxidative stress, exacerbating neuronal vulnerability. Defects in lysosomal function and autophagy result in the accumulation of toxic proteins, whereas ER stress disrupts protein balance and calcium homeostasis and activates apoptotic pathways. The loss and dysfunction of synapses, which strongly correlate with cognitive decline, are influenced by glial phagocytosis and metabolic disturbances [3]. Furthermore, recent studies have identified disruptions in the gut‒brain axis (GBA) and epigenetic changes as fundamental factors in Alzheimer’s disease pathology, opening new avenues for therapeutic targets [4]. A critical conceptual advancement in the therapeutic approach to AD lies in distinguishing between pathologies that are fundamentally driven by Aβ aggregation and those capable of autonomous initiation and propagation, even if they subsequently converge with or amplify amyloid-related toxicity. For example, neuroinflammatory processes can be activated systemically through peripheral immune challenges or by genetic risk factors such as TREM2 variants, resulting in microglial dysregulation and synaptic loss before substantial amyloid plaque deposition [5]. Similarly, deficits in mitochondrial bioenergetics represent a core feature of aging and are particularly severe in carriers of the APOE4 allele, regardless of amyloid load, indicating a parallel and contributory pathway to neuronal susceptibility [6]. The nanostrategies discussed in this review are engineered to target these upstream or concomitant dysfunctions. Their therapeutic promise stems from their potential to alter disease progression by correcting failures in essential cellular homeostasis. This provides a complementary or alternative strategy to amyloid reduction, particularly in contexts where therapies centered on amyloid have demonstrated limited clinical efficacy. AD is increasingly recognized as a multifactorial disorder in which amyloid and tau pathology represents only part of the pathogenic landscape. This narrative review synthesizes nanostructured materials engineered to modulate NANT pathways implicated in AD. We map seven convergent axes of pathology—neuroinflammation, mitochondrial dysfunction and bioenergetic failure; lysosomal impairment and autophagy disruption; ER stress and proteostasis imbalance; neurovascular compromise; gut–brain axis signaling; and epigenetic regulation—and assess how engineered nanoparticles influence these processes at the cellular and subcellular scales. Rather than focusing on a single axis or on carrier properties alone, the work integrates mechanistic insights from diverse nanoplatforms and contrasts targeted delivery with nontargeted approaches. We discuss translational considerations for multipathway strategies and outline design principles for multifunctional nanoplatforms, including multiligand surface functionalization and combinatorial payloads, which collectively realize a systems-level nanomedicine framework for AD. The translational potential is underscored by the capacity of nanostructured platforms to cross the BBB and deliver payloads to subcellular targets [7], enabling simultaneous modulation of multiple pathogenic cascades in AD. Literature search and selection methodology This work constitutes a strictly narrative review that synthesizes recent advances in nanotechnology targeting nonamyloid, non‑tau pathways in AD. Although a structured literature search strategy was applied to ensure comprehensive source identification, the subsequent thematic synthesis, critical appraisal, and discussion adhere to a narrative framework rather than a formal systematic review or meta-analysis protocol.. The search strategy targeted three primary electronic databases: PubMed, Scopus, and Web of Science. A defined set of search terms, including “AD,” “nanoparticle,” “nanomedicine,” “neuroinflammation,” “oxidative stress,” “mitochondrial dysfunction,” “synaptic plasticity,” “neurogenesis,” “blood–brain barrier,” and “non‑amyloid,” was applied to titles, abstracts, and keywords. The selection process adhered to explicit inclusion and exclusion criteria. The included studies consisted of articles published from 1997–2026 that investigated nanotherapeutics or nanodiagnostics in experimental models of AD, with a primary focus on mechanisms independent of amyloid-β or tau pathology. Articles whose principal objective involved amyloid‑targeting or tau‑targeting strategies were excluded, alongside editorials, conference abstracts, and non‑English publications. Selected high‑impact reviews and seminal earlier publications were consulted solely for contextual purposes. All retrieved records underwent an initial screening on the basis of title and abstract. The full texts of potentially relevant articles were then obtained and evaluated for final inclusion. This evaluation prioritized studies with clearly described methodologies and results obtained from established AD models. The objective of this structured process was the identification of original research that demonstrated innovative nanomaterial designs and contributed to conceptual frameworks beyond the classical amyloid and tau pathways. Relevant data on nanomaterial properties, experimental conditions, model parameters, and outcomes were extracted and integrated into the subsequent narrative synthesis. Modulation of neuroinflammation via nanostructured materials Neuroinflammation is a pathological process in neurodegenerative diseases such as AD. This condition is characterized by the activation of microglia and astrocytes, which release inflammatory cytokines and ROS, exacerbating damage to neurons [8]. Microglias, which are the resident immune cells of the central nervous system (CNS), play dual roles in neuroinflammation. Under pathological conditions, they polarize into two distinct phenotypes, M1 and M2. The M1 phenotype is inflammatory and involves the secretion of cytokines such as tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-γ), and interleukin 1 beta (IL-1β), leading to neuronal death [8, 9]. In contrast, the M2 phenotype is anti-inflammatory and produces factors such as interleukin 10 (IL-10) and brain-derived neurotrophic factor (BDNF), which have neuroprotective effects and aid in tissue repair. This polarization mechanism is a therapeutic target for modulating neuroinflammation and reducing the progression of inflammatory diseases [8]. Nanostructured materials are suitable tools for targeting neuroinflammation because they have the ability to cross the BBB and can deliver drugs to the desired site in the brain [10]. Among these materials are lipid nanoparticles (LNPs), chitosan nanoparticles, and gold nanoparticles, which have shown high potential [8, 9]. LNPs, which include solid lipid nanoparticles (SLNs) and nanostructured lipid carriers (NLCs), are particularly effective at increasing bioavailability and precise drug targeting. In murine models of AD, SLNs containing resveratrol significantly reduce brain inflammation by lowering TNF-α and IL-1β levels in Alzheimer’s disease models [8, 11]. Additionally, CUR, a natural compound with powerful anti-inflammatory and antioxidant properties, plays a role in inhibiting pathological processes associated with neurodegenerative diseases [12]. Using SLNs to carry CUR, in addition to increasing its bioavailability, helps improve the penetration of this compound across the BBB and enhances its effectiveness in target areas of the brain [13]. SLNs containing methylthioadenosine (MTA) have also increased the bioavailability of the drug in the brain fourfold and and demonstrated neuroprotective effects in murine models of demyelination, underscoring the broader potential of lipid nanocarriers for CNS delivery [8]. Claims of direct symptomatic improvement in human MS patients are not supported by the cited preclinical data. The mechanisms of action of nanostructured materials in modulating neuroinflammation are complex. Controlled drug release is an important feature that provides long-term therapeutic effects and reduces systemic toxicity (8, 9). Furthermore, some nanoparticles reduce oxidative stress through mechanisms such as inhibiting ROS or delivering antioxidants to damaged neurons [14]. Silver nanoparticles coated with CUR (CUR-Ag NPs), for example, have shown favorable performance in combating oxidative stress and inflammatory mediators [15]. Additionally, mitochondrion-targeted nanoparticles, especially those functionalized with TPP+, have also demonstrated efficacy in improving mitochondrial dynamics and biogenesis, which is considered a key factor in neurodegenerative diseases [16]. Overcoming these limitations requires interdisciplinary collaboration to optimize nanoparticle formulations, understand structure‒activity relationships, and establish regulatory frameworks for CNS-targeted therapies. Future research should focus on expanding multidrug nanoformulations by combining phytochemicals with other therapeutic agents, such as anti-inflammatory drugs or gene therapies, to increase their efficacy and reduce risks. Moreover, the integration of nanotechnology with advanced treatments, such as CUR-containing nanoliposomes targeting beta-amyloid deposits, paves the way for advanced therapeutic interventions (Fig. 1)[17]. Selected examples of nanostructured materials that target glial polarization are summarized in Table 1.Fig.1Nanoparticles protect neurons by polarizing microglia from pro-inflammatory M1 to anti-inflammatory M2 phenotype and mitigating oxidative stress in neuroinflammation. M1: warrior (inflammatory); M2: healer (reparative); ROS: reactive oxygen species; CUR (CUR-Ag NPs): curcumin-coated silver nanoparticles; TPP+: triphenylphosphonium cation; IL-10: interleukin 10; BDNF: brain-derived neurotrophic factor; BBB: blood‒brain barrierTable 1Nanostructured materials targeting glial in ADNanostructure typeLigand/coatingTargetingDosage/administrationModel systemOutcomesRefApoE-functionalized SLNsCetyl palmitate, polysorbate 80, ApoEBBB (ApoE-mediated transport)Not reported (in vitro screening only)Human (hCMEC/D3 endothelial cell line; in vitro)BBB permeability: ↑ 1.8-fold; pro-inflammatory cytokines: TNF-α ↓, IL-1β ↓, IL-6 ↓[11]Man-NPsMannose (PLGA-PEG), FTY-loadedGLUT1 (IEB/BBB), mannose receptors on microglia1 mg/kg FTY (oral, once daily × 7 days); 20% glucose gavage 1 h post-doseMouse (strain not specified, adult; in vivo) + Human/Mouse cell lines (Caco-2 intestinal, bEnd.3 endothelial, BV-2 microglial; in vitro)Brain uptake: ↑ 13.2-fold vs. free FTY; intestinal epithelial barrier Papp: ↑ 2.67-fold; microglial phenotype: CD86⁺ ↓, CD206⁺ ↑; astrocyte activation (GFAP⁺): ↓; cytokines: TNF-α ↓, IL-6 ↓, IL-10 ↑, TGF-β ↑; Morris water maze latency: ↓[18]pALRGTAngiopep-2, PEG2000GLP-1R on microglia (BV2, primary) and astrocytes5 mg/kg (i.v., single dose); 1 μM (in vitro exposure)Mouse (strain not specified, adult; in vivo) + Mouse cell lines (BV-2 microglial, primary astrocytes/microglia, bEnd.3 endothelial; in vitro)Microglial activation marker (Iba-1⁺): ↓; astrocyte marker (GFAP⁺): ↓; pro-inflammatory cytokines: IL-1β ↓, IL-6 ↓, TNF-α ↓; anti-inflammatory IL-3: ↑ via GATA3 pathway[19]Ibu&FK@RNPsRAP peptide (CELKVLMEKEL) on ROS-sensitive PCL-PEG (MAH-EDT)RAGE on BBB and parenchymal cells (astrocytes/neurons)FK506: 1 mg/kg + Ibuprofen: 16.5 mg/kg (i.v., every 48 h × 21 days)Mouse (strain not specified, adult; in vivo) + Mouse/Rat cell lines (bEnd.3 endothelial, PC12 pheochromocytoma; in vitro)Astrocyte activation (GFAP⁺): ↓ 12.8-fold; microglial marker (Iba-1⁺): ↓; NF-κB activation: ↓ 1.6-fold; IL-1β: ↓ 4.6-fold; neuronal integrity (Nissl bodies in CA1): ↑ 3.0-fold[20]CS/TPP-NPsBrain-targeted peptide TG (TGNYKALHPHNG), chitosan matrixBBB via TG peptideNot reported (proof-of-concept in vivo study)Mouse (strain not specified, adult; in vivo)Microglial activation: ↓; Aβ aggregation: ↓ via modulation of JNK/AKT/GSK3β signaling pathway[21]RMC/pApoE2MLS (metformin-inspired), RLS (arginine), OA@CeO₂Neurons (CMA, ApoE2 transfection), microglia (crosstalk)20 μg/kg pApoE2 (stereotaxic hippocampal injection, single dose)Mouse (strain not specified, adult; in vivo) + Human/Mouse cell lines (SH-SY5Y neuroblastoma, BV-2 microglial; in vitro)Microglial Aβ phagocytosis: ↑; insulin-degrading enzyme (IDE) activity: ↑; IL-1β: restored to baseline levels; TGF-β: ↑; ROS production: ↓[22]MG-LNPIonizable lipid (cKK-E12), cholesterol, DSPC, PEG-lipidMicroglia (iMGLs, BV2); brain via i.c.st0.5 mg/kg (i.p., single dose) or 0.07 mg/kg (intracisternal, single dose)Mouse (strain not specified, adult; in vivo) + Human iPSC-derived and ES-derived microglia-like cells, mouse BV-2 microglial (in vitro)Astrocyte activation (GFAP⁺ area): ↓; PU.1⁺ microglial nuclei: ↓; IBA1⁺ microglial area: ↓; complement C1Q⁺ area: ↓; GFAP⁺ astrocytic area: ↓[23]PE-SLNPsChitosan, Tween 80Passive (no active ligand)1.47 mg/kg PE-equivalent (oral gavage, once daily)Rat (strain not specified, adult; in vivo)Novel object recognition discrimination index: ↑; oxidative stress markers: ↓; lipid peroxidation: ↓[24]Pd@PEG@BorSH-PEG-COOH, borneol (Bor)Brain1 mg/kg (i.v., every 48 h × 30 days); 20 μg/mL (in vitro, 24 h exposure)Mouse (strain not specified, adult; in vivo) + Human/Mouse cell lines (SH-SY5Y neuroblastoma, bEnd.3 endothelial; in vitro)Neuronal apoptosis rate: ↓; pro-inflammatory IL-6: ↓; microglial activation marker (Iba-1⁺): ↓; astrocyte activation marker (GFAP⁺): ↓[25]SLN, solid lipid nanoparticle; NP, nanoparticle; PLGA, poly(lactic-co-glycolic acid); PEG, polyethylene glycol; FTY, fingolimod; GLUT1, glucose transporter 1; IEB, intestinal epithelial barrier; BBB, blood–brain barrier; Papp, apparent permeability coefficient; CD86/206, cluster of differentiation 86/206; GFAP, glial fibrillary acidic protein; TNF-α, tumor necrosis factor alpha; IL, interleukin; TGF-β, transforming growth factor beta; ApoE, apolipoprotein E; i.v., intravenous; GLP-1R, glucagon-like peptide-1 receptor; pALRGT, PEGylated angiopep-2-liraglutide nanostructure; RAGE, receptor for advanced glycation end products; RAP, receptor-associated protein; PCL, polycaprolactone; MAH-EDT, maleimide-ethylenediamine-thioether; FK506, tacrolimus; PC12, pheochromocytoma cells; CA1, cornu ammonis 1; TPP, tripolyphosphate; TG, TGNYKALHPHNG peptide; JNK, c-Jun N-terminal kinase; AKT, protein kinase B; GSK3β, glycogen synthase kinase 3 beta; RMC, RLS/MLS/OA@CeO₂ assembly; MLS, metformin-inspired lipid; RLS, arginine-containing lipid; OA@CeO₂, oleic acid-modified cerium dioxide; CMA, chaperone-mediated autophagy; IDE, insulin-degrading enzyme; ROS, reactive oxygen species; LNP, lipid nanoparticle; DSPC, 1,2-distearoyl-sn-glycero-3-phosphocholine; i.c.st., intracisternal; i.p., intraperitoneal; iMGL, iPSC-derived microglia-like cell; iPSC, induced pluripotent stem cell; ES, embryonic stem; C1Q, complement component 1q; PE, pomegranate extract; Pd, palladium; Bor, borneol; SH, thiol; h, hour; d, day; min, minute; s, second; vs, versus; ↑, increase; ↓, decrease; → , restored to. Nanoparticles protect neurons by polarizing microglia from pro-inflammatory M1 to anti-inflammatory M2 phenotype and mitigating oxidative stress in neuroinflammation. M1: warrior (inflammatory); M2: healer (reparative); ROS: reactive oxygen species; CUR (CUR-Ag NPs): curcumin-coated silver nanoparticles; TPP+: triphenylphosphonium cation; IL-10: interleukin 10; BDNF: brain-derived neurotrophic factor; BBB: blood‒brain barrier Table 1Nanostructured materials targeting glial in ADNanostructure typeLigand/coatingTargetingDosage/administrationModel systemOutcomesRefApoE-functionalized SLNsCetyl palmitate, polysorbate 80, ApoEBBB (ApoE-mediated transport)Not reported (in vitro screening only)Human (hCMEC/D3 endothelial cell line; in vitro)BBB permeability: ↑ 1.8-fold; pro-inflammatory cytokines: TNF-α ↓, IL-1β ↓, IL-6 ↓[11]Man-NPsMannose (PLGA-PEG), FTY-loadedGLUT1 (IEB/BBB), mannose receptors on microglia1 mg/kg FTY (oral, once daily × 7 days); 20% glucose gavage 1 h post-doseMouse (strain not specified, adult; in vivo) + Human/Mouse cell lines (Caco-2 intestinal, bEnd.3 endothelial, BV-2 microglial; in vitro)Brain uptake: ↑ 13.2-fold vs. free FTY; intestinal epithelial barrier Papp: ↑ 2.67-fold; microglial phenotype: CD86⁺ ↓, CD206⁺ ↑; astrocyte activation (GFAP⁺): ↓; cytokines: TNF-α ↓, IL-6 ↓, IL-10 ↑, TGF-β ↑; Morris water maze latency: ↓[18]pALRGTAngiopep-2, PEG2000GLP-1R on microglia (BV2, primary) and astrocytes5 mg/kg (i.v., single dose); 1 μM (in vitro exposure)Mouse (strain not specified, adult; in vivo) + Mouse cell lines (BV-2 microglial, primary astrocytes/microglia, bEnd.3 endothelial; in vitro)Microglial activation marker (Iba-1⁺): ↓; astrocyte marker (GFAP⁺): ↓; pro-inflammatory cytokines: IL-1β ↓, IL-6 ↓, TNF-α ↓; anti-inflammatory IL-3: ↑ via GATA3 pathway[19]Ibu&FK@RNPsRAP peptide (CELKVLMEKEL) on ROS-sensitive PCL-PEG (MAH-EDT)RAGE on BBB and parenchymal cells (astrocytes/neurons)FK506: 1 mg/kg + Ibuprofen: 16.5 mg/kg (i.v., every 48 h × 21 days)Mouse (strain not specified, adult; in vivo) + Mouse/Rat cell lines (bEnd.3 endothelial, PC12 pheochromocytoma; in vitro)Astrocyte activation (GFAP⁺): ↓ 12.8-fold; microglial marker (Iba-1⁺): ↓; NF-κB activation: ↓ 1.6-fold; IL-1β: ↓ 4.6-fold; neuronal integrity (Nissl bodies in CA1): ↑ 3.0-fold[20]CS/TPP-NPsBrain-targeted peptide TG (TGNYKALHPHNG), chitosan matrixBBB via TG peptideNot reported (proof-of-concept in vivo study)Mouse (strain not specified, adult; in vivo)Microglial activation: ↓; Aβ aggregation: ↓ via modulation of JNK/AKT/GSK3β signaling pathway[21]RMC/pApoE2MLS (metformin-inspired), RLS (arginine), OA@CeO₂Neurons (CMA, ApoE2 transfection), microglia (crosstalk)20 μg/kg pApoE2 (stereotaxic hippocampal injection, single dose)Mouse (strain not specified, adult; in vivo) + Human/Mouse cell lines (SH-SY5Y neuroblastoma, BV-2 microglial; in vitro)Microglial Aβ phagocytosis: ↑; insulin-degrading enzyme (IDE) activity: ↑; IL-1β: restored to baseline levels; TGF-β: ↑; ROS production: ↓[22]MG-LNPIonizable lipid (cKK-E12), cholesterol, DSPC, PEG-lipidMicroglia (iMGLs, BV2); brain via i.c.st0.5 mg/kg (i.p., single dose) or 0.07 mg/kg (intracisternal, single dose)Mouse (strain not specified, adult; in vivo) + Human iPSC-derived and ES-derived microglia-like cells, mouse BV-2 microglial (in vitro)Astrocyte activation (GFAP⁺ area): ↓; PU.1⁺ microglial nuclei: ↓; IBA1⁺ microglial area: ↓; complement C1Q⁺ area: ↓; GFAP⁺ astrocytic area: ↓[23]PE-SLNPsChitosan, Tween 80Passive (no active ligand)1.47 mg/kg PE-equivalent (oral gavage, once daily)Rat (strain not specified, adult; in vivo)Novel object recognition discrimination index: ↑; oxidative stress markers: ↓; lipid peroxidation: ↓[24]Pd@PEG@BorSH-PEG-COOH, borneol (Bor)Brain1 mg/kg (i.v., every 48 h × 30 days); 20 μg/mL (in vitro, 24 h exposure)Mouse (strain not specified, adult; in vivo) + Human/Mouse cell lines (SH-SY5Y neuroblastoma, bEnd.3 endothelial; in vitro)Neuronal apoptosis rate: ↓; pro-inflammatory IL-6: ↓; microglial activation marker (Iba-1⁺): ↓; astrocyte activation marker (GFAP⁺): ↓[25]SLN, solid lipid nanoparticle; NP, nanoparticle; PLGA, poly(lactic-co-glycolic acid); PEG, polyethylene glycol; FTY, fingolimod; GLUT1, glucose transporter 1; IEB, intestinal epithelial barrier; BBB, blood–brain barrier; Papp, apparent permeability coefficient; CD86/206, cluster of differentiation 86/206; GFAP, glial fibrillary acidic protein; TNF-α, tumor necrosis factor alpha; IL, interleukin; TGF-β, transforming growth factor beta; ApoE, apolipoprotein E; i.v., intravenous; GLP-1R, glucagon-like peptide-1 receptor; pALRGT, PEGylated angiopep-2-liraglutide nanostructure; RAGE, receptor for advanced glycation end products; RAP, receptor-associated protein; PCL, polycaprolactone; MAH-EDT, maleimide-ethylenediamine-thioether; FK506, tacrolimus; PC12, pheochromocytoma cells; CA1, cornu ammonis 1; TPP, tripolyphosphate; TG, TGNYKALHPHNG peptide; JNK, c-Jun N-terminal kinase; AKT, protein kinase B; GSK3β, glycogen synthase kinase 3 beta; RMC, RLS/MLS/OA@CeO₂ assembly; MLS, metformin-inspired lipid; RLS, arginine-containing lipid; OA@CeO₂, oleic acid-modified cerium dioxide; CMA, chaperone-mediated autophagy; IDE, insulin-degrading enzyme; ROS, reactive oxygen species; LNP, lipid nanoparticle; DSPC, 1,2-distearoyl-sn-glycero-3-phosphocholine; i.c.st., intracisternal; i.p., intraperitoneal; iMGL, iPSC-derived microglia-like cell; iPSC, induced pluripotent stem cell; ES, embryonic stem; C1Q, complement component 1q; PE, pomegranate extract; Pd, palladium; Bor, borneol; SH, thiol; h, hour; d, day; min, minute; s, second; vs, versus; ↑, increase; ↓, decrease; → , restored to. Nanostructured materials targeting glial in AD SLN, solid lipid nanoparticle; NP, nanoparticle; PLGA, poly(lactic-co-glycolic acid); PEG, polyethylene glycol; FTY, fingolimod; GLUT1, glucose transporter 1; IEB, intestinal epithelial barrier; BBB, blood–brain barrier; Papp, apparent permeability coefficient; CD86/206, cluster of differentiation 86/206; GFAP, glial fibrillary acidic protein; TNF-α, tumor necrosis factor alpha; IL, interleukin; TGF-β, transforming growth factor beta; ApoE, apolipoprotein E; i.v., intravenous; GLP-1R, glucagon-like peptide-1 receptor; pALRGT, PEGylated angiopep-2-liraglutide nanostructure; RAGE, receptor for advanced glycation end products; RAP, receptor-associated protein; PCL, polycaprolactone; MAH-EDT, maleimide-ethylenediamine-thioether; FK506, tacrolimus; PC12, pheochromocytoma cells; CA1, cornu ammonis 1; TPP, tripolyphosphate; TG, TGNYKALHPHNG peptide; JNK, c-Jun N-terminal kinase; AKT, protein kinase B; GSK3β, glycogen synthase kinase 3 beta; RMC, RLS/MLS/OA@CeO₂ assembly; MLS, metformin-inspired lipid; RLS, arginine-containing lipid; OA@CeO₂, oleic acid-modified cerium dioxide; CMA, chaperone-mediated autophagy; IDE, insulin-degrading enzyme; ROS, reactive oxygen species; LNP, lipid nanoparticle; DSPC, 1,2-distearoyl-sn-glycero-3-phosphocholine; i.c.st., intracisternal; i.p., intraperitoneal; iMGL, iPSC-derived microglia-like cell; iPSC, induced pluripotent stem cell; ES, embryonic stem; C1Q, complement component 1q; PE, pomegranate extract; Pd, palladium; Bor, borneol; SH, thiol; h, hour; d, day; min, minute; s, second; vs, versus; ↑, increase; ↓, decrease; → , restored to. The role of nanoparticles in combating mitochondrial dysfunction Oxidative stress is a biochemical process resulting from an imbalance between the production and elimination of ROS [26]. ROS are natural byproducts of cellular processes that have important signaling functions under controlled conditions, but when their levels increase, they can damage DNA, proteins, and lipids. The brain is highly susceptible to oxidative damage due to its high energy demand, elevated oxygen consumption, abundance of peroxidizable unsaturated fatty acids, and high levels of catalytic iron for ROS production [27, 28]. Mitochondrial dysfunction in Alzheimer’s pathology Mitochondrial dysfunction is one of the main pathways of ROS production in the pathogenesis of neurodegenerative diseases such as AD and is independent of amyloid or tau. This dysfunction is characterized by impaired energy metabolism, increased oxidative stress, and an imbalance in calcium homeostasis, all of which play a role in neuronal damage and disease progression [29]. Furthermore, comparisons of mitochondrial DNA (mtDNA) with nuclear DNA (nDNA) have shown that oxidative damage to mtDNA in Alzheimer’s disease patients is 10 to 20 times greater than that to nDNA, indicating the greater sensitivity of mtDNA to ROS-induced damage and the important role of mitochondria in the progression of oxidative stress in AD [30]. Other studies on cell lines whose mtDNA was isolated from Alzheimer’s disease patients have shown that these cells produce higher levels of ROS and free radicals than control cells do. This finding confirms a direct link between mtDNA mutations, excessive ROS production, and the initiation or exacerbation of neurotoxic processes in Alzheimer’s disease (Fig. 2) [31].Fig. 2Mitochondrion-targeted nanoparticles rescue neuronal bioenergetics by scavenging ROS, repairing mtDNA damage, and restoring ATP production in neurodegenerative conditions. GSH-AuNPs: glutathione-conjugated gold nanoparticles; CeO2 NPs: cerium oxide nanoparticles; ROS: reactive oxygen species; mtDNA: mitochondrial DNA Mitochondrion-targeted nanoparticles rescue neuronal bioenergetics by scavenging ROS, repairing mtDNA damage, and restoring ATP production in neurodegenerative conditions. GSH-AuNPs: glutathione-conjugated gold nanoparticles; CeO2 NPs: cerium oxide nanoparticles; ROS: reactive oxygen species; mtDNA: mitochondrial DNA Recent advances in nanotechnology have provided new opportunities to combat mitochondrial dysfunction. These advances have focused mainly on the development of nanoparticles capable of directly improving mitochondrial function, including cerium oxide nanoparticles (CeO2 NPs) and nanoparticles functionalized with various groups that specifically target mitochondria. GSH-AuNPs also improve mitochondrial function and increase ATP production by maintaining the mitochondrial membrane potential, cytochrome c distribution, and respiratory chain activity. The protective effects of these nanoparticles are also stabilized by regulating the expression of genes related to mitochondrial biogenesis (such as PGC-1α, NRF-1, and Tfam) [32]. Mitochondria-targeting ligands The mitochondrial membrane’s high negative electrical potential limits entry of anionic molecules [33]. To overcome this barrier, nanoparticles employ mitochondrion-targeting ligands including TPP⁺, tetramethylrhodamine-5-isothiocyanate, mitochondrial peptides, dequalinium, and natural compounds such as hypericin and glycyrrhetinic acid [34]. Among these ligands, TPP⁺ has been extensively investigated as an effective mitochondriotropic moiety because of its favorable safety profile and capacity to traverse mitochondrial membranes. The conjugation of TPP⁺ to nanoparticles enhances their lipophilicity and exploits the negative mitochondrial membrane potential (approximately − 180 mV), facilitating preferential partitioning into the mitochondrial matrix under experimental conditions. Multiple independent preclinical studies report that TPP⁺-functionalized nanocarriers achieve enhanced subcellular accumulation compared with untargeted formulations. However, these observations remain confined to rodent and cell culture models, and clinical translation requires rigorous validation of in vivo pharmacokinetics, off-target tissue distribution, and dose-dependent toxicity profiles. Seven distinct nanocarrier systems functionalized with TPP⁺ have demonstrated successful mitochondrial delivery in preclinical models of brain disorders, including (i) ceria NPs [35], (ii) PAMAM dendrimers conjugated to N-acetylcysteine [36], (iii) red blood cell membrane-camouflaged human serum albumin NPs loaded with CUR [37], (iv) RBC membrane-coated nanostructured lipid carriers for resveratrol [38], (v) PLGA-b-PEG-TPP+ copolymers for antiretrovirals [39], (vi) tetrahedral DNA framework NPs [40], and (vii) Cu2₋ₓSe-based NPs for CUR delivery [41]. The lipophilic cationic character of TPP⁺ enables these nanoparticles to bypass the impermeability of the inner mitochondrial membrane to anionic therapeutics, resulting in increased intramitochondrial drug concentrations. Consequently, compared with nontargeted formulations (5–50 mg/kg), TPP⁺-functionalized nanoparticles achieve therapeutic efficacy at substantially lower doses (2–5 mg/kg for CUR), thereby minimizing systemic exposure while preserving mitochondrial bioenergetics, the membrane potential, and dynamics [16]. CeO2 NPs possess unique redox properties derived from their ability to switch between Ce2⁺ and Ce3⁺ oxidation states. This feature allows them to scavenge ROS effectively. Studies using transgenic mouse models of AD have shown that CeO2 NPs significantly reduce oxidative stress and thereby slow the progression of dementia [42–44]. The synthesis methods and biological effects of these nanostructured materials play important roles in their efficiency. CeO2 NPs are usually synthesized via precipitation or hydrothermal methods to achieve precise control over the size, shape, and surface chemistry of the particles. Surface modifications with biocompatible polymers or functional groups further enhance the stability and targeting capability of these particles. Preclinical data have shown that CeO2 NPs, when administered intravenously at doses of 0.1 to 10 mg/kg, significantly reduce ROS levels and improve mitochondrial dynamics [42]. These targeted approaches not only increase mitochondrial ATP production but also restore the membrane potential and reduce mitochondrial degradation [45]. Additionally, CUR, an antioxidant, prevents neuron death by reducing mitochondrial oxidative stress, but its targeted delivery to the mitochondria of brain neurons remains challenging. In this context, biomimetic nanoparticles coated with red blood cell membranes (RBC-NPs) have been engineered to efficiently deliver CUR to neuronal mitochondria. These nanoparticles are functionalized with the T807 ligand to facilitate BBB penetration and triphenylphosphine (TPP+) for mitochondrial targeting and are collectively designated T807/TPP+-RBC-NPs. These nanoparticles ensure high stability, controlled drug release, biocompatibility, and prolonged circulation in the blood. CUR loaded in these nanoparticles (CUR-loaded T807/TPP+-RBC-NPs) inhibits neuron death by reducing mitochondrial oxidative stress and improves Alzheimer’s disease symptoms both in vitro and in vivo (Fig. 3) [37].Fig. 3Mitochondrion-targeted ligands enable smart nanoparticle delivery across biological barriers, enhancing therapeutic efficacy at lower doses by precise mitochondrial localization and ROS scavenging. CeO2 NPs: cerium oxide nanoparticles; mtDNA: mitochondrial deoxyribonucleic acid; TPP+: triphenylphosphonium cation; IL-1β: interleukin 1 beta; BBB: blood‒brain barrier; ROS: reactive oxygen species Mitochondrion-targeted ligands enable smart nanoparticle delivery across biological barriers, enhancing therapeutic efficacy at lower doses by precise mitochondrial localization and ROS scavenging. CeO2 NPs: cerium oxide nanoparticles; mtDNA: mitochondrial deoxyribonucleic acid; TPP+: triphenylphosphonium cation; IL-1β: interleukin 1 beta; BBB: blood‒brain barrier; ROS: reactive oxygen species Examples of TPP⁺-functionalized and some mitochondrion-targeted nanoparticles are presented in Table 2.Table 2Mitochondrion-targeted nanostructures for restoring bioenergetics in ADNanostructure typeFunctionalization StatusMitochondrial LigandSubcellular TargetDosage/AdministrationModel systemOutcomesRefTPP⁺-functionalized NPsFunctionalizedTPP⁺Mitochondrial matrix2 mg/kg CUR-equivalent (i.v., once daily × 21 days)Mouse (3 × Tg-AD transgenic; in vivo)ATP production: ↑ 2.4-fold; mtROS: ↓; PGC-1α expression: ↑ 3.1-fold[16]Nanobrake (Mg-CypD-LNC)Dual-functionalizedTPP⁺CypD, mitochondrial Ca2⁺siRNA: 82.5 μg/kg (i.v., weekly × 4 weeks); 100 nM siRNA + 5 μM Aβ₁–₄₂ (in vitro, 48 h exposure)Mouse (5xFAD transgenic; in vivo) + Human/Mouse cell lines (bEnd.3 endothelial, SH-SY5Y neuroblastoma, BV-2 microglial; in vitro)mtROS: ↓ 59%; ΔΨm: ↑ 54.4%; CypD: ↓ 44.9%; ATP: ↑; capillary density: ↑ 44%; cognitive performance: improved[46]TPP-ceria NPsTPP-conjugated, DSPE-PEG, FITCTPP⁺Mitochondrial oxidative stress0.1 mM Ce3⁺ (stereotaxic injection into subiculum, single dose)Mouse (5xFAD transgenic; in vivo) + Human cell lines (SH-SY5Y neuroblastoma, HeLa cervical carcinoma; in vitro)Aβ-induced mtROS: ↓ 60–70%; neuronal density (NeuN⁺): ↑ ~ 40%; glial activation (GFAP⁺/Iba-1⁺): ↓ ~ 50%; lipid peroxidation marker (4-HNE): ↓[35]TPP-MoS₂-QDsTPP-conjugated DSPE-PEG-NH₂, FITCTPP⁺OMM, cristaeNot reported (in vitro characterization study)Human neuronal cell line (AD pathology model; in vitro)Mitochondrial colocalization: ↑; Aβ-induced ROS: ↓; outer mitochondrial membrane/cristae integrity: preserved; neuronal apoptosis: ↓[47]ctGQDsTransferrin, lactoferrin, Abs–AChE inhibitionNot applicable (physicochemical characterization only)In vitro physicochemical characterization (no biological model)AChE inhibitory activity: ↑ cytotoxicity: low; predicted BBB permeability: favorable[48]TDFNsTPP, cholesterol, ASOTPP⁺miRNA-34a, p53/Bax/Bcl-2/caspase-3Antisense oligonucleotide: 2 mg/kg (i.v., weekly × 4 weeks)Mouse (3 × Tg-AD transgenic; in vivo) + Human cell lines (HEK293/TAU; in vitro)miR-34a expression: ↓ 78%; SIRT1: ↑; phosphorylated Tau/p53/Bax: ↓ 58–67%; Bcl-2: ↑; neuronal loss: ↓ 76%; apoptosis-associated neuroinflammation: ↓[40]PN-NPsplain PLGA matrix for sustained NO releaseindirect mitochondrial targeting via nitric oxide releaseMitochondria (↑ biogenesis and function via NO pathway)PLGA: 16 mg/kg + PAPANONOate: 422 μg/kg (i.v., days 1, 4, 7)Mouse (C57BL/6, LPS-induced neuroinflammation model; in vivo)NO release kinetics: sustained (30% at 12 h vs. 87% for free PN); cytochrome c oxidase activity: ↑; ATP production: ↑[49]GSH-AuNPsGSH-conjugated– (indirect redox)Nrf2, antioxidant enzymes, ER stress, mito biogenesis10 ppm GSH-AuNPs (added 48 h post-Aβ₁–₄₂ exposure, 24 h incubation)Human neural stem cells (hNSCs) + 5 μM Aβ₁–₄₂ exposure (in vitro)Caspase-3/9 activity: ↓; Nrf2 pathway activation: ↑; antioxidant enzymes (SOD, GPx, Catalase, HO-1): ↑; ER stress markers (Bip, CHOP, ASK1, p-ASK1): ↓[32]TPP-ceria NPsTPP-DSPE-PEG, FITCTPP⁺Mitochondria0.1 mM Ce3⁺ (3 μL stereotaxic injection into subiculum, single dose)Mouse (5xFAD transgenic; in vivo) + Human cell lines (HeLa, U373 astrocytoma; in vitro)mtROS: ↓; NeuN⁺ neuronal density: ↑; astrocyte/microglial activation (GFAP⁺/Iba-1⁺): ↓; lipid peroxidation (4-HNE): ↓[35]HSA NPs (T807/TPP-RBC-Cur)T807/TPP-RBC-camouflagedT807 (BBB) + TPP⁺Neuronal mitochondriaCurcumin-equivalent: 2–5 mg/kg (i.v., once daily × 10 days); 5 μg/mL (in vitro, 24 h exposure)Mouse (AD transgenic model; in vivo) + Mouse cell line (HT22 hippocampal neurons; in vitro)CA1 hippocampal neuronal survival: ↑; mtROS: ↓; apoptosis rate: ↓ to 17.0% (vs. 41.4% control); glial activation (GFAP⁺/Iba-1⁺): ↓[37]Ab-TP-MDNPsAnti-Aβ Ab, terpolymer-lipid– (indirect: ROS scavenging, O₂ generation)Mitochondria (indirect: ΔΨm, TOMM20, ATP)Manganese: 100 μmol/kg (i.v., twice weekly × 2 weeks for PET imaging; once weekly × 8 weeks for LTP assessment)Mouse (TgCRND8 transgenic; in vivo) + Primary mouse cortical neurons (in vitro)O-GlcNAcylation: ↑ 70%; ATP production: ↑ 32%; oxidative stress markers: ↓; neuroinflammatory cytokines: ↓; cognitive performance: ↑[45]Aβ, amyloid beta; AD, Alzheimer’s disease; PN-NPs, PAPANOate (PN/EBP homologous protein; ASO, antisense oligonucleotide; ATP, adenosine triphosphate; BBB, blood–brain barrier; Bcl-2, B-cell lymphoma 2; Bip, binding immunoglobulin protein; Ce, cerium; CHOP, C/EBP homologous protein; COX, cytochrome c oxidase; CUR, curcumin; CypD, cyclophilin D; d, day(s); DSPE, 1,2-distearoyl-sn-glycero-3-phosphoethanolamine; eq., equivalent; ER, endoplasmic reticulum; FAD, familial Alzheimer’s disease; FITC, fluorescein isothiocyanate; 4-HNE, 4-hydroxynonenal; GFAP, glial fibrillary acidic protein; GSH, glutathione; hNSCs, human neural stem cells; HO-1, heme oxygenase-1; HSA, human serum albumin; i.v., intravenous; Iba-1, ionized calcium-binding adapter molecule 1; LPS, lipopolysaccharide; LTP, long-term potentiation; Mg-CypD-LNC, magnesium-cyclophilin D-loaded lipid nanocapsule; MMP9, matrix metalloproteinase 9; MoS₂ QDs, molybdenum disulfide quantum dots; mtROS, mitochondrial reactive oxygen species; NeuN, neuronal nuclei; NO, nitric oxide; NP, nanoparticle; Nrf2, nuclear factor erythroid 2–related factor 2; OMM, outer mitochondrial membrane; p-, phosphorylated; PEG, polyethylene glycol; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; PLGA, poly(lactic-co-glycolic acid); PN, PAPANONOate; RBC, red blood cell; ROS, reactive oxygen species; SIRT1, sirtuin 1; SOD, superoxide dismutase; TDFNs, tetrahedral DNA framework nanoparticles; Tg, transgenic; TPP⁺, triphenylphosphonium; vs., versus; wk, week; ΔΨm, mitochondrial membrane potential; ↑, increase; ↓, decrease. Table 2Mitochondrion-targeted nanostructures for restoring bioenergetics in ADNanostructure typeFunctionalization StatusMitochondrial LigandSubcellular TargetDosage/AdministrationModel systemOutcomesRefTPP⁺-functionalized NPsFunctionalizedTPP⁺Mitochondrial matrix2 mg/kg CUR-equivalent (i.v., once daily × 21 days)Mouse (3 × Tg-AD transgenic; in vivo)ATP production: ↑ 2.4-fold; mtROS: ↓; PGC-1α expression: ↑ 3.1-fold[16]Nanobrake (Mg-CypD-LNC)Dual-functionalizedTPP⁺CypD, mitochondrial Ca2⁺siRNA: 82.5 μg/kg (i.v., weekly × 4 weeks); 100 nM siRNA + 5 μM Aβ₁–₄₂ (in vitro, 48 h exposure)Mouse (5xFAD transgenic; in vivo) + Human/Mouse cell lines (bEnd.3 endothelial, SH-SY5Y neuroblastoma, BV-2 microglial; in vitro)mtROS: ↓ 59%; ΔΨm: ↑ 54.4%; CypD: ↓ 44.9%; ATP: ↑; capillary density: ↑ 44%; cognitive performance: improved[46]TPP-ceria NPsTPP-conjugated, DSPE-PEG, FITCTPP⁺Mitochondrial oxidative stress0.1 mM Ce3⁺ (stereotaxic injection into subiculum, single dose)Mouse (5xFAD transgenic; in vivo) + Human cell lines (SH-SY5Y neuroblastoma, HeLa cervical carcinoma; in vitro)Aβ-induced mtROS: ↓ 60–70%; neuronal density (NeuN⁺): ↑ ~ 40%; glial activation (GFAP⁺/Iba-1⁺): ↓ ~ 50%; lipid peroxidation marker (4-HNE): ↓[35]TPP-MoS₂-QDsTPP-conjugated DSPE-PEG-NH₂, FITCTPP⁺OMM, cristaeNot reported (in vitro characterization study)Human neuronal cell line (AD pathology model; in vitro)Mitochondrial colocalization: ↑; Aβ-induced ROS: ↓; outer mitochondrial membrane/cristae integrity: preserved; neuronal apoptosis: ↓[47]ctGQDsTransferrin, lactoferrin, Abs–AChE inhibitionNot applicable (physicochemical characterization only)In vitro physicochemical characterization (no biological model)AChE inhibitory activity: ↑ cytotoxicity: low; predicted BBB permeability: favorable[48]TDFNsTPP, cholesterol, ASOTPP⁺miRNA-34a, p53/Bax/Bcl-2/caspase-3Antisense oligonucleotide: 2 mg/kg (i.v., weekly × 4 weeks)Mouse (3 × Tg-AD transgenic; in vivo) + Human cell lines (HEK293/TAU; in vitro)miR-34a expression: ↓ 78%; SIRT1: ↑; phosphorylated Tau/p53/Bax: ↓ 58–67%; Bcl-2: ↑; neuronal loss: ↓ 76%; apoptosis-associated neuroinflammation: ↓[40]PN-NPsplain PLGA matrix for sustained NO releaseindirect mitochondrial targeting via nitric oxide releaseMitochondria (↑ biogenesis and function via NO pathway)PLGA: 16 mg/kg + PAPANONOate: 422 μg/kg (i.v., days 1, 4, 7)Mouse (C57BL/6, LPS-induced neuroinflammation model; in vivo)NO release kinetics: sustained (30% at 12 h vs. 87% for free PN); cytochrome c oxidase activity: ↑; ATP production: ↑[49]GSH-AuNPsGSH-conjugated– (indirect redox)Nrf2, antioxidant enzymes, ER stress, mito biogenesis10 ppm GSH-AuNPs (added 48 h post-Aβ₁–₄₂ exposure, 24 h incubation)Human neural stem cells (hNSCs) + 5 μM Aβ₁–₄₂ exposure (in vitro)Caspase-3/9 activity: ↓; Nrf2 pathway activation: ↑; antioxidant enzymes (SOD, GPx, Catalase, HO-1): ↑; ER stress markers (Bip, CHOP, ASK1, p-ASK1): ↓[32]TPP-ceria NPsTPP-DSPE-PEG, FITCTPP⁺Mitochondria0.1 mM Ce3⁺ (3 μL stereotaxic injection into subiculum, single dose)Mouse (5xFAD transgenic; in vivo) + Human cell lines (HeLa, U373 astrocytoma; in vitro)mtROS: ↓; NeuN⁺ neuronal density: ↑; astrocyte/microglial activation (GFAP⁺/Iba-1⁺): ↓; lipid peroxidation (4-HNE): ↓[35]HSA NPs (T807/TPP-RBC-Cur)T807/TPP-RBC-camouflagedT807 (BBB) + TPP⁺Neuronal mitochondriaCurcumin-equivalent: 2–5 mg/kg (i.v., once daily × 10 days); 5 μg/mL (in vitro, 24 h exposure)Mouse (AD transgenic model; in vivo) + Mouse cell line (HT22 hippocampal neurons; in vitro)CA1 hippocampal neuronal survival: ↑; mtROS: ↓; apoptosis rate: ↓ to 17.0% (vs. 41.4% control); glial activation (GFAP⁺/Iba-1⁺): ↓[37]Ab-TP-MDNPsAnti-Aβ Ab, terpolymer-lipid– (indirect: ROS scavenging, O₂ generation)Mitochondria (indirect: ΔΨm, TOMM20, ATP)Manganese: 100 μmol/kg (i.v., twice weekly × 2 weeks for PET imaging; once weekly × 8 weeks for LTP assessment)Mouse (TgCRND8 transgenic; in vivo) + Primary mouse cortical neurons (in vitro)O-GlcNAcylation: ↑ 70%; ATP production: ↑ 32%; oxidative stress markers: ↓; neuroinflammatory cytokines: ↓; cognitive performance: ↑[45]Aβ, amyloid beta; AD, Alzheimer’s disease; PN-NPs, PAPANOate (PN/EBP homologous protein; ASO, antisense oligonucleotide; ATP, adenosine triphosphate; BBB, blood–brain barrier; Bcl-2, B-cell lymphoma 2; Bip, binding immunoglobulin protein; Ce, cerium; CHOP, C/EBP homologous protein; COX, cytochrome c oxidase; CUR, curcumin; CypD, cyclophilin D; d, day(s); DSPE, 1,2-distearoyl-sn-glycero-3-phosphoethanolamine; eq., equivalent; ER, endoplasmic reticulum; FAD, familial Alzheimer’s disease; FITC, fluorescein isothiocyanate; 4-HNE, 4-hydroxynonenal; GFAP, glial fibrillary acidic protein; GSH, glutathione; hNSCs, human neural stem cells; HO-1, heme oxygenase-1; HSA, human serum albumin; i.v., intravenous; Iba-1, ionized calcium-binding adapter molecule 1; LPS, lipopolysaccharide; LTP, long-term potentiation; Mg-CypD-LNC, magnesium-cyclophilin D-loaded lipid nanocapsule; MMP9, matrix metalloproteinase 9; MoS₂ QDs, molybdenum disulfide quantum dots; mtROS, mitochondrial reactive oxygen species; NeuN, neuronal nuclei; NO, nitric oxide; NP, nanoparticle; Nrf2, nuclear factor erythroid 2–related factor 2; OMM, outer mitochondrial membrane; p-, phosphorylated; PEG, polyethylene glycol; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; PLGA, poly(lactic-co-glycolic acid); PN, PAPANONOate; RBC, red blood cell; ROS, reactive oxygen species; SIRT1, sirtuin 1; SOD, superoxide dismutase; TDFNs, tetrahedral DNA framework nanoparticles; Tg, transgenic; TPP⁺, triphenylphosphonium; vs., versus; wk, week; ΔΨm, mitochondrial membrane potential; ↑, increase; ↓, decrease. Mitochondrion-targeted nanostructures for restoring bioenergetics in AD Aβ, amyloid beta; AD, Alzheimer’s disease; PN-NPs, PAPANOate (PN/EBP homologous protein; ASO, antisense oligonucleotide; ATP, adenosine triphosphate; BBB, blood–brain barrier; Bcl-2, B-cell lymphoma 2; Bip, binding immunoglobulin protein; Ce, cerium; CHOP, C/EBP homologous protein; COX, cytochrome c oxidase; CUR, curcumin; CypD, cyclophilin D; d, day(s); DSPE, 1,2-distearoyl-sn-glycero-3-phosphoethanolamine; eq., equivalent; ER, endoplasmic reticulum; FAD, familial Alzheimer’s disease; FITC, fluorescein isothiocyanate; 4-HNE, 4-hydroxynonenal; GFAP, glial fibrillary acidic protein; GSH, glutathione; hNSCs, human neural stem cells; HO-1, heme oxygenase-1; HSA, human serum albumin; i.v., intravenous; Iba-1, ionized calcium-binding adapter molecule 1; LPS, lipopolysaccharide; LTP, long-term potentiation; Mg-CypD-LNC, magnesium-cyclophilin D-loaded lipid nanocapsule; MMP9, matrix metalloproteinase 9; MoS₂ QDs, molybdenum disulfide quantum dots; mtROS, mitochondrial reactive oxygen species; NeuN, neuronal nuclei; NO, nitric oxide; NP, nanoparticle; Nrf2, nuclear factor erythroid 2–related factor 2; OMM, outer mitochondrial membrane; p-, phosphorylated; PEG, polyethylene glycol; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1-alpha; PLGA, poly(lactic-co-glycolic acid); PN, PAPANONOate; RBC, red blood cell; ROS, reactive oxygen species; SIRT1, sirtuin 1; SOD, superoxide dismutase; TDFNs, tetrahedral DNA framework nanoparticles; Tg, transgenic; TPP⁺, triphenylphosphonium; vs., versus; wk, week; ΔΨm, mitochondrial membrane potential; ↑, increase; ↓, decrease. Comparison of targeted and nontargeted nanoparticles A comparative analysis of mitochondrion-targeted and nontargeted nanoparticles demonstrated context-dependent advantages in experimental models (Tables 2, 5). However, direct performance comparisons are complicated by substantial heterogeneity in study design, including divergent transgenic backgrounds, dosing paradigms, and administration routes across the reviewed literature. Notably, while TPP⁺-functionalized formulations have demonstrated therapeutic efficacy at reduced dosages (2–5 mg/kg CUR equivalent) in head-to-head preclinical evaluations [16, 37], untargeted carriers in parallel neurodegenerative models often require higher systemic exposures (5–50 mg/kg) to achieve comparable biochemical outcomes [50, 51]. Furthermore, the dynamic protein corona that forms in vivo can differentially mask targeting ligands depending on administration route, adding another layer of complexity to cross-study interpretation [52–56]. Within these constraints, mitochondrion-targeted nanoparticles have consistently demonstrated enhanced modulation of bioenergetic regulators (e.g., PGC-1α, SIRT1) in comparative preclinical studies (Table 2), supporting their potential for superior subcellular delivery. Standardized head-to-head evaluations across identical animal models, dosing schedules, and routes remain scarce, underscoring the need for harmonized preclinical benchmarks before clinical extrapolation. To rigorously evaluate these bioenergetic improvements, contemporary studies increasingly rely on direct functional assays rather than indirect behavioral proxies. Quantitative assessments of mitochondrial competence now routinely incorporate oxygen consumption rate (OCR) profiling, ATP-linked respiration measurements, and maximal respiratory capacity assays, which provide objective, pathway-specific readouts of nanoparticle efficacy [16, 57]. The reliability of these core metabolic endpoints is consistently reinforced through orthogonal validation, including high-resolution respirometry, cytochrome c oxidase activity assays, and ultrastructural analysis of cristae morphology. This methodological rigor ensures that observed therapeutic benefits reflect genuine restoration of mitochondrial bioenergetics rather than compensatory or off-target physiological adaptations. While the dose-sparing effect achieved through mitochondrial targeting markedly lowers the risk of off-target toxicity, these findings must be contextualized within established safety thresholds. Although both targeted and untargeted CUR dosing regimens currently exceed the 0–3 mg/kg allowable daily intake for human dietary consumption, the substantial exposure reduction afforded by precise subcellular delivery represents a clinically meaningful step toward mitigating systemic burden and potential hepatorenal strain [37, 58]. These comparative outcomes collectively demonstrate that therapeutic efficacy in mitochondrial dysfunction pathways is governed primarily by targeting precision rather than mere nanoscale dimensions. The convergence of evidence across multiple independent preclinical studies validates core nanostructural design principles for NANT interventions and highlights the critical need for standardized, mechanism-driven evaluation frameworks to accelerate clinical translation. Protein corona formation and its implications for targeted nanodelivery in vivo Upon intravenous administration, nanoparticles undergo rapid adsorption of plasma proteins within seconds to minutes, forming a dynamic biomolecular corona that confers a new biological identity distinct from their engineered physicochemical properties [54]. This corona layer frequently masks surface-conjugated targeting ligands through steric hindrance, substantially diminishing active targeting efficiency in vivo compared with in vitro observations [52, 53]. For example, transferrin-functionalized nanoparticles exhibited diminished cellular targeting following corona formation in human plasma [55]. Recent in vivo corona characterization studies revealed that nanoparticles accumulating in the murine brain parenchyma frequently possess coronas enriched with apolipoprotein E and clusterin, which may facilitate receptor-mediated transcytosis across the experimental BBB [55, 56]. This dynamic shift from engineered ligand targeting to corona-mediated biological identity represents a critical pharmacological consideration. Corona composition is highly sensitive to plasma protein variability, disease stage, and administration route, necessitating standardized in vitro-to-in vivo correlation studies before clinical translation. Nanostructures targeting lysosomal dysfunction in AD Lysosomal dysfunction in AD involves defective acidification, impaired enzyme activity (e.g., cathepsin D), vATPase proton pump defects, altered TFEB-mediated biogenesis, and disrupted endosome-lysosome trafficking. These impairments lead to accumulation of toxic substrates and neuronal death [60, 61]. The application of nanostructures in targeting and regulating lysosomal activity is helpful in the treatment of neurodegenerative diseases such as Alzheimer’s disease [62]. Several nanostructures selectively accumulate in lysosomes, help repair lysosomal damage and restore the optimal pH of lysosomes [63]. Nanostructures such as polymeric nanoparticles, including PLGA, have also been effective in improving lysosomal function. These particles lower the lysosomal pH and increase protein degradation activity [64]. Furthermore, carbon nanostructures such as single-walled carbon nanotubes (SWCNTs) also play a role in modulating lysosomal function. These nanoparticles reduce oxidative stress and improve mitochondrial function [65]. Surface-characterized nanoparticles (nanoparticles whose surface properties—including surface chemistry, surface charge, hydrophilic/hydrophobic properties, presence of ligands, polymer coating, and functional groups—are identified via analytical methods) are also intelligently directed toward lysosomes and enhance autophagy processes [66]. Carbon-based nanoparticles: from structural frameworks to theranostic dots Structural carbon nanoparticles: graphene, nanotubes, and fullerenes Carbon-based nanoparticles encompass a diverse family of structures, including graphene oxide (GO), carbon nanotubes (CNTs), and fullerenes, which have garnered significant attention for AD therapy because of their unique physicochemical properties. Owing to their high surface area, exceptional mechanical strength, and versatile functionalization capabilities, these materials are robust platforms for drug delivery, neuroprotection, and bioimaging. For example, graphene oxide sheets have been shown to mitigate oxidative stress and improve cognitive function in transgenic AD models [67, 68]. Similarly, carbon nanotubes can be engineered to ferry neurotrophic factors across the BBB and promote neuronal survival [69]. The inherent optical and electrical properties of these materials further enable their use as contrast agents for near-infrared imaging, facilitating noninvasive monitoring of disease progression [67]. However, their long-term biocompatibility and potential neurotoxicity require careful surface engineering and rigorous toxicological evaluation [70]. Carbon dots (CDs) A distinct and rapidly advancing subclass within carbon nanoparticles is CDs. These quasispherical nanoparticles (2–10 nm) exhibit exceptional biocompatibility, low toxicity, tunable fluorescence, and ease of surface functionalization. Unlike their structural counterparts, CDs offer unique advantages for simultaneous diagnosis and therapy [71]. In addition to their established role in Aβ theranostics [72], the intrinsic physicochemical properties of CDs, such as their potent antioxidant capacity and facile surface engineering, position them as systems with adaptable potential for engaging NANT pathways [71, 73]. The proven efficacy of these compounds in scavenging ROS and their ability to conjugate with various therapeutic ligands theoretically enable them to target key NANT processes such as neuroinflammation and oxidative stress [73]. Furthermore, their small size facilitates efficient BBB penetration and promotes renal clearance, minimizing the risk of long-term accumulation [74]. This combination of properties suggests that strategically engineered CDs may have the potential to evolve into a convergent nanoplatform for multitarget interventions addressing both amyloid and NANT pathologies in AD. Synaptic disorders and synaptic plasticity in AD and nanotechnology-based interventions Loss of synapses is a common and early sign of AD, and the extent of this loss correlates with the severity of cognitive decline [75]. Studies have shown that microglia and astrocytes play important roles in synapse elimination. In the disease state, these cells phagocytose both healthy and damaged synapses and contribute to the spread of damage. The complement system, especially the molecules C1q and C3, is involved in this process and tags synapses for removal by microglia. Studies indicate that activation of the metabotropic glutamate receptor type 1 (mGluR1) in rodent models of AD leads to phagocytosis of glutamatergic synapses by microglia. This process occurs through the upregulation of the complement protein C1q, which is regulated by mGluR1. Activation of mGluR1 causes dephosphorylation of the FMRP protein and increases local translation of C1q mRNA at synapses. The increase in C1q, in turn, stimulates the phagocytosis of synapses by microglia, resulting in synapse loss and cognitive impairments. Inhibition of mGluR1 signaling or a reduction in C1q expression reverses this process and improves synaptic and cognitive function [76]. Additionally, changes in the expression of inflammation-related genes, such as TREM2 and APOE, lead to increased synapse elimination [77, 78]. In recent years, clinical studies have focused on the development of synaptic biomarkers, such as SV2A proteins, neurogranin, and SNAP25, in cerebrospinal fluid to enable early diagnosis of this disease. PET imaging with SV2A ligands has also made it possible to observe synapse loss in the living brain. New treatments targeting interactions between synapses and glial cells, such as CSF1R inhibitors and anti-C1q antibodies, are being tested. These approaches aim to slow disease progression by preserving synaptic integrity [79, 80]. Mitochondria play an important role in synaptic processes, and dysfunction of synaptic mitochondria contributes to synapse degradation in AD, as mitochondria are crucial for maintaining synaptic health. Reduced energy production and oxidative stress caused by mitochondrial dysfunction lead to impaired synaptic function and ultimately neuronal death [81]. Nanostructures such as magnetic or carbon nanoparticles can be used as drug carriers or antioxidant agents to target mitochondria. These materials can facilitate improved mitochondrial function and reduce cellular damage [82]. Additionally, nanostructures can assist in imaging synaptic changes in the early stages of disease [83]. Imaging via magnetic nanoparticles has shown that the gradual loss of synapses often begins decades before the onset of clinical symptoms and is initially evident in areas such as the entorhinal cortex and hippocampus. A reduction in synaptic density and dysfunction, especially in the molecular layer of the dentate gyrus, is associated with episodic memory impairment. NPs, as MRI theranostic agents, can target damaged regions for both early diagnosis through precise imaging and targeted delivery of synapse-repairing drugs. Given that the severity of cognitive impairments is more closely linked to synapse loss than to amyloid or neurofibrillary deposits, the use of surface-functionalized nanoparticles aids in improving targeting and therapeutic performance at the synaptic level. These nanoparticles adapt their biological surface properties to the biological environment and more effectively bind to damaged synapses [84]. Synaptic plasticity, i.e., the ability of synapses to strengthen or weaken over time in response to increased or decreased activity, is a fundamental mechanism in learning, memory, and cognitive functions. This dynamic process is regulated by complex molecular pathways such as calcium signaling, the regulation of neurotrophic factors, and receptor trafficking. However, synaptic plasticity is highly susceptible to damage caused by oxidative stress, mitochondrial dysfunction, and disruption of calcium homeostasis, which are characteristic features of many neurodegenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease [85]. These pathological conditions disrupt the delicate balance necessary for maintaining neuronal communication and lead to decreased cognitive function and gradual degeneration of the nervous system. Consequently, strategies aimed at enhancing synaptic plasticity have emerged as important therapeutic targets for combating neurological disorders. Nanostructures modulating ER stress in AD ER stress plays a significant role in the progression of AD, primarily through the accumulation of misfolded proteins, disruption of homeostasis, increased oxidative stress, and neuroinflammation, which result in neuronal dysfunction and cell death [86, 87]. According to several studies, the Presenilin1 (PS1) protein and amyloid precursor protein (APP) exhibit heightened responses to ER stress. PS1 is involved in regulating the calcium balance within the ER. The enzyme SERCA (Sarco/ER Ca2⁺-ATPase), which is responsible for transferring calcium from the cytosol into the ER, along with ER-associated calcium channels such as the inositol trisphosphate receptor and ryanodine receptor, are influenced by the function of the PS1 protein. Mutations in PS1 linked to familial Alzheimer’s disease (FAD) cause alterations in the activity of these calcium transport channels. These changes in the calcium concentration within the ER are considered stable and effective factors that induce stress in this network [88]. In this context, the use of nanostructures as a novel tool to modulate ER stress in neurodegenerative diseases has been proposed. For example, quercetin-modified nanoparticles (Qc@SNPs) have been designed, which, after crossing the BBB with the help of microbubbles and ultrasound, effectively accumulate in the brain parenchyma. In preclinical models, these nanoparticles have been reported to protect nerve cells by targeting ER stress through reducing neuronal apoptosis, modulating inflammatory responses, regulating calcium homeostasis, and decreasing oxidative stress. The results of behavioral tests, such as the Morris water maze test, revealed that treatment with the Qc@SNPs significantly improved learning and memory in an Alzheimer’s disease mouse model. This smart drug delivery system (Qc@SNP-MBs) is not only considered a safe and effective method for crossing the BBB but also offers a new strategy for treating neurodegenerative diseases such as Alzheimer’s disease by regulating ER stress. The Qc@SNP-MB nanodrug system, composed of quercetin-modified sulfur nanoparticles (Qc@SNPs) and microbubbles, facilitates focused ultrasound-mediated BBB permeabilization in preclinical models. Following intracranial delivery, experimental data indicate reduced ER stress markers, attenuated oxidative damage, and improved behavioral performance in murine models through modulation of calcium homeostasis and suppression of apoptotic pathways (Fig. 4) [87]. These outcomes are confined to acute/short-term animal paradigms and do not yet reflect chronic human disease progression.Fig. 4Nanostructures alleviate ER stress in Alzheimer’s disease by restoring proteostasis, calcium homeostasis, and redox balance thereby preventing neuronal apoptosis and cognitive decline. ER: endoplasmic reticulum; ROS: reactive oxygen species; ONOO–: peroxynitrite; BBB: blood‒brain barrier; SA: syringic acid; GSH-AuNPs: glutathione-conjugated gold nanoparticles; Qc@SNPs: quercetin-modified sulfur nanoparticles; CHOP: C/EBP homologous protein Nanostructures alleviate ER stress in Alzheimer’s disease by restoring proteostasis, calcium homeostasis, and redox balance thereby preventing neuronal apoptosis and cognitive decline. ER: endoplasmic reticulum; ROS: reactive oxygen species; ONOO–: peroxynitrite; BBB: blood‒brain barrier; SA: syringic acid; GSH-AuNPs: glutathione-conjugated gold nanoparticles; Qc@SNPs: quercetin-modified sulfur nanoparticles; CHOP: C/EBP homologous protein Recent studies have also shown that nanoparticles containing plant metabolites such as syringic acid (SA), which regulate ER stress, have protective effects on nerve cells. These compounds prevent neuronal cell death by inhibiting ER stress signaling pathways, including reducing the expression of the genes encoding ATF-6, caspase-8, and caspase-3. SA-containing designed nanoparticles not only cross the BBB but also show high potential for improving neuronal cell function by targeting ER disorders. In vivo studies have shown that these nanoparticles increase the survival of neuroblastoma cells (SH-SY5Y) by reducing oxidative stress and inflammation [83]. Furthermore, glutathione-conjugated gold nanoparticles (GSH-AuNPs) have demonstrated neuroprotective potential by reducing ER stress, which is evident through the regulation of ER stress markers such as Bip, C/EBP homologous protein (CHOP), and ASK1 and the restoration of calcium homeostasis in neural stem cells. These nanoparticles combat oxidative damage by enhancing the Nrf2 antioxidant pathway and increasing the expression of enzymes such as SOD, catalase, and HO-1. By simultaneously targeting ER stress, oxidative damage, and mitochondrial dysfunction, GSH-AuNPs provide a comprehensive environment for protecting nerve cells against neurodegeneration [32]. Research by Huang et al. [89] also revealed that ER stress and the related process called ER-phagy are involved in the pathology of AD. Peroxynitrite (ONOO–), an important biomarker, participates in the initiation and exacerbation of these processes. In one study, a smart fluorescence nanosensor based on AIEgen named DHQM, which has the ability to specifically target the ER network and detect ONOO–, was designed. This sensor, with high sensitivity, was able to track ONOO– fluctuations in an aluminum-induced Alzheimer’s disease cell model through fluorescence imaging. Ferroptosis (a recently discovered type of programmed cell death) is distinct from classical types of cell death, such as apoptosis. This process is iron-dependent and is associated with iron accumulation and severe lipid peroxidation in the cell membrane, leading to cell damage and death) induced by aluminum, leading to excessive production of ONOO–, which in turn activates and exacerbates ER stress. This nanosensor not only provides a valuable tool for understanding the pathological mechanisms of Alzheimer’s disease at the ER network level but also suggests new therapeutic methods for identifying important pathways [89]. Recent advances in nanocarrier engineering have enabled simultaneous modulation of lysosomal function through single multifunctional platforms (Table 3). These systems incorporate stimuli-responsive elements and dual-targeting ligands to address the interconnected nature of subcellular pathologies in AD.Table 3Nanostructures targeting lysosomal and autophagic pathways in ADNanostructure typeTargeting organelleBiological targetDosage/administrationModel systemOutcomesRefMExo-GemLysosomesMannose receptor (microglia)Gemfibrozil-equivalent: 100 μg/mL (i.p., once daily × 7 days)Mouse (strain not specified, adult; in vivo)Morris water maze: escape latency ↓; platform crossings ↑ 2.1-fold; target quadrant time ↑ 1.8-fold; pro-inflammatory cytokines: IL-6 ↓, IL-1β ↓, TNF-α ↓[90]SWCNTs, nanoemulsionsLysosomesLysosomal pH, autophagy, Aβ/tau, neuroinflammationPLGA: 200 μg/mL (in vitro); SWCNTs: ≤ 300 mg/kg (i.v. or intranasal, single or repeated dose as specified in source)Mouse (strain not specified, adult; in vivo) + Mouse/Human cell lines (primary cortical neurons, SH-SY5Y neuroblastoma, BE(2)-M17 neuroblastoma; in vitro)PLGA-NPs: lysosomal pH restored; SWCNTs: autophagic flux ↑ via mTOR pathway normalization and enhanced lysosomal protease activity[91]MSC-PLGA-NPsLysosomesLysosomal pH, autophagy, APP, neuroinflammation20 μM (in vitro exposure); dose not specified (i.p., in vivo)Mouse (C57BL/6, adult; in vivo) + Mouse cell line (N2a neuroblastoma; in vitro)BBB penetration: ↑ 2.3-fold vs. unmodified PLGA; lysosomal pH: restored to physiological range; autophagic markers: LC3-II/I ratio ↓, p62 ↓ 52%; APP clearance: ↑; pro-inflammatory cytokines: IL-6 ↓, TNF-α ↓; hepatorenal toxicity[92]T-LIP@PL-LA/siRC (Nanodredger)Lysosomes, AutophagosomesROCK2, axonal retrograde transport, mito bioenergeticssiRNA: 1.5 OD + Lithospermic acid: 20 mg/kg (i.v., every 72 h × 9 doses); 100 nM siRNA + 18 μM LA (in vitro, 48 h exposure)Mouse (3 × Tg-AD transgenic, female, age not specified; in vivo) + Mouse cell line (HT22 hippocampal neurons + Aβ₂₅–₃₅ exposure; in vitro)Autophagosome retrograde transport velocity: restored to 32.5 nm/s (vs. impaired in AD control); neuronal complexity: ↑; dendritic spine density: ↑; serum neurofilament light chain (NFL): ↓; neuroinflammation: indirectly mitigated[93]PLGA-PEG-RAPLysosomesRAGE on compromised BBBPLGA-PEG-RAP: 60 mg/kg + Simvastatin: 2 mg/kg (i.v., every 48 h × 8 total doses)Mouse (strain not specified, adult; in vivo) + Mouse cell line (bEnd.3 endothelial + RAGE overexpression; in vitro)RAGE expression: ↓ 1.25-fold; ROS production: ↓ 31%; NF-κB p65 nuclear translocation: restored to baseline; neuroinflammation: attenuated (IL-10 ↑, TGF-β ↑)[94]Aβ, amyloid beta; ACh, acetylcholine; AD, Alzheimer’s disease; APP, amyloid precursor protein; BBB, blood–brain barrier; d, day(s); EC, entorhinal cortex; eq., equivalent; Gem, gemfibrozil; h, hour(s); i.n., intranasal; i.p., intraperitoneal; i.v., intravenous; IL, interleukin; LA, lithospermic acid B; MExo, mannose-modified exosome; MSC, mesenchymal stem cell; NFL, neurofilament light chain; NP, nanoparticle; OD, optical density; PLGA, poly(lactic-co-glycolic acid); q3d, once every 3 days; RAGE, receptor for advanced glycation end products; ROCK2, Rho-associated coiled-coil kinase 2; SWCNT, single-walled carbon nanotube; Tg, transgenic; TNF-α, tumor necrosis factor alpha; vs., versus; ↑, increase; ↓, decrease; PLGA-PEG-RAP, PLGA-PEG conjugated with receptor associated protein peptide (ELKVLMEKEL). Table 3Nanostructures targeting lysosomal and autophagic pathways in ADNanostructure typeTargeting organelleBiological targetDosage/administrationModel systemOutcomesRefMExo-GemLysosomesMannose receptor (microglia)Gemfibrozil-equivalent: 100 μg/mL (i.p., once daily × 7 days)Mouse (strain not specified, adult; in vivo)Morris water maze: escape latency ↓; platform crossings ↑ 2.1-fold; target quadrant time ↑ 1.8-fold; pro-inflammatory cytokines: IL-6 ↓, IL-1β ↓, TNF-α ↓[90]SWCNTs, nanoemulsionsLysosomesLysosomal pH, autophagy, Aβ/tau, neuroinflammationPLGA: 200 μg/mL (in vitro); SWCNTs: ≤ 300 mg/kg (i.v. or intranasal, single or repeated dose as specified in source)Mouse (strain not specified, adult; in vivo) + Mouse/Human cell lines (primary cortical neurons, SH-SY5Y neuroblastoma, BE(2)-M17 neuroblastoma; in vitro)PLGA-NPs: lysosomal pH restored; SWCNTs: autophagic flux ↑ via mTOR pathway normalization and enhanced lysosomal protease activity[91]MSC-PLGA-NPsLysosomesLysosomal pH, autophagy, APP, neuroinflammation20 μM (in vitro exposure); dose not specified (i.p., in vivo)Mouse (C57BL/6, adult; in vivo) + Mouse cell line (N2a neuroblastoma; in vitro)BBB penetration: ↑ 2.3-fold vs. unmodified PLGA; lysosomal pH: restored to physiological range; autophagic markers: LC3-II/I ratio ↓, p62 ↓ 52%; APP clearance: ↑; pro-inflammatory cytokines: IL-6 ↓, TNF-α ↓; hepatorenal toxicity[92]T-LIP@PL-LA/siRC (Nanodredger)Lysosomes, AutophagosomesROCK2, axonal retrograde transport, mito bioenergeticssiRNA: 1.5 OD + Lithospermic acid: 20 mg/kg (i.v., every 72 h × 9 doses); 100 nM siRNA + 18 μM LA (in vitro, 48 h exposure)Mouse (3 × Tg-AD transgenic, female, age not specified; in vivo) + Mouse cell line (HT22 hippocampal neurons + Aβ₂₅–₃₅ exposure; in vitro)Autophagosome retrograde transport velocity: restored to 32.5 nm/s (vs. impaired in AD control); neuronal complexity: ↑; dendritic spine density: ↑; serum neurofilament light chain (NFL): ↓; neuroinflammation: indirectly mitigated[93]PLGA-PEG-RAPLysosomesRAGE on compromised BBBPLGA-PEG-RAP: 60 mg/kg + Simvastatin: 2 mg/kg (i.v., every 48 h × 8 total doses)Mouse (strain not specified, adult; in vivo) + Mouse cell line (bEnd.3 endothelial + RAGE overexpression; in vitro)RAGE expression: ↓ 1.25-fold; ROS production: ↓ 31%; NF-κB p65 nuclear translocation: restored to baseline; neuroinflammation: attenuated (IL-10 ↑, TGF-β ↑)[94]Aβ, amyloid beta; ACh, acetylcholine; AD, Alzheimer’s disease; APP, amyloid precursor protein; BBB, blood–brain barrier; d, day(s); EC, entorhinal cortex; eq., equivalent; Gem, gemfibrozil; h, hour(s); i.n., intranasal; i.p., intraperitoneal; i.v., intravenous; IL, interleukin; LA, lithospermic acid B; MExo, mannose-modified exosome; MSC, mesenchymal stem cell; NFL, neurofilament light chain; NP, nanoparticle; OD, optical density; PLGA, poly(lactic-co-glycolic acid); q3d, once every 3 days; RAGE, receptor for advanced glycation end products; ROCK2, Rho-associated coiled-coil kinase 2; SWCNT, single-walled carbon nanotube; Tg, transgenic; TNF-α, tumor necrosis factor alpha; vs., versus; ↑, increase; ↓, decrease; PLGA-PEG-RAP, PLGA-PEG conjugated with receptor associated protein peptide (ELKVLMEKEL). Nanostructures targeting lysosomal and autophagic pathways in AD Aβ, amyloid beta; ACh, acetylcholine; AD, Alzheimer’s disease; APP, amyloid precursor protein; BBB, blood–brain barrier; d, day(s); EC, entorhinal cortex; eq., equivalent; Gem, gemfibrozil; h, hour(s); i.n., intranasal; i.p., intraperitoneal; i.v., intravenous; IL, interleukin; LA, lithospermic acid B; MExo, mannose-modified exosome; MSC, mesenchymal stem cell; NFL, neurofilament light chain; NP, nanoparticle; OD, optical density; PLGA, poly(lactic-co-glycolic acid); q3d, once every 3 days; RAGE, receptor for advanced glycation end products; ROCK2, Rho-associated coiled-coil kinase 2; SWCNT, single-walled carbon nanotube; Tg, transgenic; TNF-α, tumor necrosis factor alpha; vs., versus; ↑, increase; ↓, decrease; PLGA-PEG-RAP, PLGA-PEG conjugated with receptor associated protein peptide (ELKVLMEKEL). Nanostructures modulating neurovascular dysfunction in AD Cerebrovascular disorders and the role of the BBB in AD have recently attracted significant attention. The investigation of targeted nanoparticles to improve cerebral vascular function and repair the BBB is an important part of NANT treatment [95]. The neurovascular unit (NVU) includes endothelial cells, pericytes, astrocytes, microglia, and neurons, which together maintain homeostasis in the brain. In AD, disruption of the NVU leads to a reduced cerebral blood supply, increased BBB permeability, and neuroinflammation, all of which independently contribute to neuronal damage and cognitive decline, separate from beta-amyloid plaques or tau protein [96, 97]. Owing to their unique physicochemical properties and ability to cross the BBB, nanostructures have provided new therapeutic approaches for restoring NVU function and alleviating vascular deficiencies. One important component of the NVU is endothelial cells, which are damaged in AD, resulting in reduced vasodilation and cerebral blood flow. NPs designed to deliver vascular protective agents or antioxidants can target endothelial cells and reduce oxidative stress and inflammation. Polymeric nanoparticles containing nitric oxide donors or antioxidant enzymes have been shown to significantly improve endothelial function and microcirculation in preclinical models. These nanoparticles overcome the limitations of conventional therapies by increasing drug availability and providing controlled release at the site of vascular injury [18, 98, 99]. Other characteristics of neurovascular disorders in AD include the loss of pericytes, which leads to BBB disruption and the infiltration of toxic substances into brain tissue. Nanostructures designed to preserve pericyte survival or stimulate pericyte regeneration effectively maintain BBB stability. NPs that deliver growth factors such as platelet-derived growth factor (PDGF) to the target site have the potential to increase pericyte proliferation and strengthen vessel walls. Additionally, nanoparticles modified with targeted ligands can selectively bind to pericytes and facilitate therapeutic effects at the desired location [96, 100, 101]. Astrocytes and microglia, which constitute the glial components of the NVU, become activated under neurovascular disorder conditions and release inflammatory cytokines that contribute to vascular and neuronal damage. Nanoparticles capable of modulating glial activity by delivering anti-inflammatory agents or inhibitory RNA molecules have been developed to reduce neuroinflammation [98]. For example, LNPs containing siRNAs targeting inflammatory pathways inhibited microglial activity in animal models and improved vascular function. Similarly, Ralvenius et al. [23] used LNPs to deliver siRNAs specific to a gene called PU.1 to the brain. When these nanoparticles were injected intrathecally into mice, the uptake of mRNA into the brain increased, and within 12 h, they were able to silence approximately 92% of the PU.1 gene. This resulted in reduced production of inflammatory substances in the mouse brain and helped decrease neuroinflammation. Another application of nanostructures is regulating ionic balance and oxidative stress within the NVU. Metal oxide nanoparticles such as CeO2 have inherent antioxidant properties and can protect vascular cells against oxidative damage. These nanoparticles are designed to cross the BBB, accumulate in damaged brain regions, and continuously provide neurovascular protection [102]. Another area where nanostructures have high potential is enhancing the function of the glymphatic system (the brain’s waste and toxin clearance system, which is impaired in AD due to NVU dysfunction) [103]. When magnetic nanoparticles are combined with external magnetic fields, they guide drugs to perivascular spaces, thereby facilitating the removal of metabolic toxins and improving interstitial fluid flow [104]. Furthermore, recent advances in nanotechnology have enabled the creation of biomimetic nanoparticles that mimic cellular vesicles such as exosomes. These nanoparticles can safely and effectively cross the BBB and target specific types of vascular cells. These nanoparticles are capable of delivering miRNAs or other regulatory molecules that stimulate angiogenesis, stabilize the endothelial wall, and support neurovascular communication [105]. Nanotechnology-based modulation of the GBA in AD pathology Modulation of the GBA via nanotechnology is a method employed to combat AD (Table 4). This axis directly addresses the complex bidirectional communication between the gastrointestinal system and the CNS. The GBA includes a network of neural, immune, and metabolic pathways that regulate brain function and behavior. Recent evidence has shown that disruption of this axis plays a role in the development of neurodegenerative diseases such as Alzheimer’s disease. Changes in the composition of the gut microbiota (known as dysbiosis) lead to increased intestinal permeability, systemic inflammation, damage to the BBB, neuroinflammation, neuronal damage, and associated complications, including Alzheimer’s disease [106].Table 4Nanostructured materials targeting the GBA in ADNanostructure typeLigand/coatingPrimary targetDosage/administrationModel systemRole of gut-brain axisOutcomesRefTG-Res-CS/TPP-NPsCS, TPP, TG peptideBrain (BBB)60 mg/kg (oral gavage, once daily × 24 weeks)Mouse (ICR strain, adult; in vivo)↑ microbiota diversity, ↓ Firmicutes/Bacteroidetes, ↓ LPS, modulated inflammation/lipid taxaescape latency ↓, target quadrant time ↑; neuroinflammation: IBA-1⁺ ↓, GFAP⁺ ↓, IL-6 ↓, IL-1β ↓, TNF-α ↓, LPS ↓; signaling: p-JNK/JNK ↓, p-AKT/AKT ↑, p-ERK/ERK ↑, p-GSK3β/GSK3β ↑; antioxidant enzymes: SOD ↑, GSH-px ↑, CAT ↑; MDA: ↓; BBB transport efficiency: 95%[21]PVSEndogenousPassiveNot applicable (observational imaging study)Mouse (APP/PS2 transgenic, age not specified; in vivo)Anatomical conduit for gut–brain immune communicationY-maze spontaneous alternation: ↓; primo vascular system roughness: OS-PVS Rq ↑, M-PVS Rq ↑; primo pore density: ↑[117]Nano-HOHS-15, PEG-400, MCTPassiveHonokiol-equivalent: 20 mg/kg (oral gavage, once daily × 17 weeks)Mouse (TgCRND8 transgenic, age not specified; in vivo)↑ beneficial bacteria, ↓ pro-inflammatory taxa, protected GBA stabilityCognitive performance: escape latency ↓, target quadrant time ↑, platform crossings ↑; pro-inflammatory cytokines: TNF-α ↓, IL-1β ↓, IL-6 ↓; glial activation: IBA-1⁺ ↓, GFAP⁺ ↓; apoptosis markers: caspase-3 ↓, Bcl-2 ↑[118]Res@SeNPs@Res-CS-NPsCS, SeNPsPassive60 mg/kg (oral gavage, once daily × 24 weeks)Mouse (ICR strain, adult; in vivo)↑ microbiota diversity, ↓ pro-inflammatory taxa, inhibited neuroinflammation via GBAGlial activation: IBA-1⁺ ↓, GFAP⁺ ↓; neuroinflammatory cytokines: IL-6 ↓, IL-1β ↓, TNF-α ↓; insulin resistance: ↓; cognitive performance: ↑[114]Probiotic micro/nanocapsulesAlginate, chitosan, gelatin, gum arabic, carrageenan, whey protein, caseinGut microbiota/GBANot reported (review of encapsulation technologies)Mouse (strain not specified; in vivo) + Human (clinical observational; ex vivo)Modulates GBA via SCFAs, ↓ neuroinflammation, ↑ intestinal barrierMolecular docking (GABA–caspase): binding energy –7.3 kcal/mol; pro-inflammatory cytokines: IL-6 ↓, IL-12 ↓, TNF-α ↓; cognitive function: ↑; disease progression: slowed[119]SeNPsChitosan, TGN, Tet-1, LPFFD, chondroitin sulfateGut microbiota, BBBOral administration (dose/frequency not specified in source)Mouse/Rat/Pig (strain not specified, adult; in vivo)↑Akkermansia, ↓ Lachnospiraceae_NK4A136/Rikenella/Helicobacter, ↑ SCFAs, ↓ NLRP3Gut microbiota α-diversity: ↑; beneficial bacterial genera: ↑; amyloid-β burden: ↓; phosphorylated tau: ↓; pro-inflammatory cytokines: IL-6 ↓, IL-1β ↓, TNF-α ↓; cognitive performance: ↑SeNPs (L. casei ATCC 393-SeNPs)Polysaccharides, proteinsMicrobiota–gut–brain axisSelenium: 0.3 mg/kg + L. casei ATCC 393: 0.5 × 10⁹ CFU/g (oral gavage, once daily × 13 weeks)Mouse (C57BL/6, adult, D-galactose + AlCl₃-induced AD model; in vivo)↑ beneficial bacteria, ↓ pro-inflammatory taxa, ↑ SCFAs, ↑ intestinal barrier, ↓ microglia activationMicroglial activation: ↓; pro-inflammatory cytokines: IL-1β ↓, IL-18 ↓; anti-inflammatory cytokines: IL-4 ↑, IL-10 ↑; apoptosis markers: Bax/Bcl-2 ratio ↓, caspase-1 ↓; p53 expression: ↑; neurotransmitters: dopamine ↑, serotonin ↑, GABA ↑[115]AD, Alzheimer’s disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BBB, blood–brain barrier; BUN, blood urea nitrogen; CAT, catalase; CFU, colony-forming unit; CS, chitosan; CREA, creatinine; D-gal, D-galactose; F/B, Firmicutes/Bacteroidetes; GBA, gut–brain axis; GFAP, glial fibrillary acidic protein; GSH-px, glutathione peroxidase; HO, honokiol; IBA-1, ionized calcium-binding adapter molecule 1; IL, interleukin; i.p., intraperitoneal; LPS, lipopolysaccharide; MCT, medium-chain triglyceride; MDA, malondialdehyde; M-PVS, meningeal primo vascular system; MW, molecular weight; Nano-DDS, nanoscale drug delivery system; Nano-HO, honokiol-loaded nanoemulsion; O/W, oil-in-water; OS-PVS, organ surface primo vascular system; PEG, polyethylene glycol; PP, primo pore; Preclin.:Preclinical; PVS, primo-vascular system; Rq, root mean square roughness; SCFAs, short-chain fatty acids; SeNPs, selenium nanoparticles; SMEDDS, self-microemulsifying drug delivery system; SOD, superoxide dismutase; TG, TGNYKALHPHNG peptide; Tg, transgenic; TNF-α, tumor necrosis factor alpha; TPP, sodium tripolyphosphate; vs, versus; wk, week; ↑, increase; ↓, decrease. Table 4Nanostructured materials targeting the GBA in ADNanostructure typeLigand/coatingPrimary targetDosage/administrationModel systemRole of gut-brain axisOutcomesRefTG-Res-CS/TPP-NPsCS, TPP, TG peptideBrain (BBB)60 mg/kg (oral gavage, once daily × 24 weeks)Mouse (ICR strain, adult; in vivo)↑ microbiota diversity, ↓ Firmicutes/Bacteroidetes, ↓ LPS, modulated inflammation/lipid taxaescape latency ↓, target quadrant time ↑; neuroinflammation: IBA-1⁺ ↓, GFAP⁺ ↓, IL-6 ↓, IL-1β ↓, TNF-α ↓, LPS ↓; signaling: p-JNK/JNK ↓, p-AKT/AKT ↑, p-ERK/ERK ↑, p-GSK3β/GSK3β ↑; antioxidant enzymes: SOD ↑, GSH-px ↑, CAT ↑; MDA: ↓; BBB transport efficiency: 95%[21]PVSEndogenousPassiveNot applicable (observational imaging study)Mouse (APP/PS2 transgenic, age not specified; in vivo)Anatomical conduit for gut–brain immune communicationY-maze spontaneous alternation: ↓; primo vascular system roughness: OS-PVS Rq ↑, M-PVS Rq ↑; primo pore density: ↑[117]Nano-HOHS-15, PEG-400, MCTPassiveHonokiol-equivalent: 20 mg/kg (oral gavage, once daily × 17 weeks)Mouse (TgCRND8 transgenic, age not specified; in vivo)↑ beneficial bacteria, ↓ pro-inflammatory taxa, protected GBA stabilityCognitive performance: escape latency ↓, target quadrant time ↑, platform crossings ↑; pro-inflammatory cytokines: TNF-α ↓, IL-1β ↓, IL-6 ↓; glial activation: IBA-1⁺ ↓, GFAP⁺ ↓; apoptosis markers: caspase-3 ↓, Bcl-2 ↑[118]Res@SeNPs@Res-CS-NPsCS, SeNPsPassive60 mg/kg (oral gavage, once daily × 24 weeks)Mouse (ICR strain, adult; in vivo)↑ microbiota diversity, ↓ pro-inflammatory taxa, inhibited neuroinflammation via GBAGlial activation: IBA-1⁺ ↓, GFAP⁺ ↓; neuroinflammatory cytokines: IL-6 ↓, IL-1β ↓, TNF-α ↓; insulin resistance: ↓; cognitive performance: ↑[114]Probiotic micro/nanocapsulesAlginate, chitosan, gelatin, gum arabic, carrageenan, whey protein, caseinGut microbiota/GBANot reported (review of encapsulation technologies)Mouse (strain not specified; in vivo) + Human (clinical observational; ex vivo)Modulates GBA via SCFAs, ↓ neuroinflammation, ↑ intestinal barrierMolecular docking (GABA–caspase): binding energy –7.3 kcal/mol; pro-inflammatory cytokines: IL-6 ↓, IL-12 ↓, TNF-α ↓; cognitive function: ↑; disease progression: slowed[119]SeNPsChitosan, TGN, Tet-1, LPFFD, chondroitin sulfateGut microbiota, BBBOral administration (dose/frequency not specified in source)Mouse/Rat/Pig (strain not specified, adult; in vivo)↑Akkermansia, ↓ Lachnospiraceae_NK4A136/Rikenella/Helicobacter, ↑ SCFAs, ↓ NLRP3Gut microbiota α-diversity: ↑; beneficial bacterial genera: ↑; amyloid-β burden: ↓; phosphorylated tau: ↓; pro-inflammatory cytokines: IL-6 ↓, IL-1β ↓, TNF-α ↓; cognitive performance: ↑SeNPs (L. casei ATCC 393-SeNPs)Polysaccharides, proteinsMicrobiota–gut–brain axisSelenium: 0.3 mg/kg + L. casei ATCC 393: 0.5 × 10⁹ CFU/g (oral gavage, once daily × 13 weeks)Mouse (C57BL/6, adult, D-galactose + AlCl₃-induced AD model; in vivo)↑ beneficial bacteria, ↓ pro-inflammatory taxa, ↑ SCFAs, ↑ intestinal barrier, ↓ microglia activationMicroglial activation: ↓; pro-inflammatory cytokines: IL-1β ↓, IL-18 ↓; anti-inflammatory cytokines: IL-4 ↑, IL-10 ↑; apoptosis markers: Bax/Bcl-2 ratio ↓, caspase-1 ↓; p53 expression: ↑; neurotransmitters: dopamine ↑, serotonin ↑, GABA ↑[115]AD, Alzheimer’s disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BBB, blood–brain barrier; BUN, blood urea nitrogen; CAT, catalase; CFU, colony-forming unit; CS, chitosan; CREA, creatinine; D-gal, D-galactose; F/B, Firmicutes/Bacteroidetes; GBA, gut–brain axis; GFAP, glial fibrillary acidic protein; GSH-px, glutathione peroxidase; HO, honokiol; IBA-1, ionized calcium-binding adapter molecule 1; IL, interleukin; i.p., intraperitoneal; LPS, lipopolysaccharide; MCT, medium-chain triglyceride; MDA, malondialdehyde; M-PVS, meningeal primo vascular system; MW, molecular weight; Nano-DDS, nanoscale drug delivery system; Nano-HO, honokiol-loaded nanoemulsion; O/W, oil-in-water; OS-PVS, organ surface primo vascular system; PEG, polyethylene glycol; PP, primo pore; Preclin.:Preclinical; PVS, primo-vascular system; Rq, root mean square roughness; SCFAs, short-chain fatty acids; SeNPs, selenium nanoparticles; SMEDDS, self-microemulsifying drug delivery system; SOD, superoxide dismutase; TG, TGNYKALHPHNG peptide; Tg, transgenic; TNF-α, tumor necrosis factor alpha; TPP, sodium tripolyphosphate; vs, versus; wk, week; ↑, increase; ↓, decrease. Nanostructured materials targeting the GBA in AD AD, Alzheimer’s disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BBB, blood–brain barrier; BUN, blood urea nitrogen; CAT, catalase; CFU, colony-forming unit; CS, chitosan; CREA, creatinine; D-gal, D-galactose; F/B, Firmicutes/Bacteroidetes; GBA, gut–brain axis; GFAP, glial fibrillary acidic protein; GSH-px, glutathione peroxidase; HO, honokiol; IBA-1, ionized calcium-binding adapter molecule 1; IL, interleukin; i.p., intraperitoneal; LPS, lipopolysaccharide; MCT, medium-chain triglyceride; MDA, malondialdehyde; M-PVS, meningeal primo vascular system; MW, molecular weight; Nano-DDS, nanoscale drug delivery system; Nano-HO, honokiol-loaded nanoemulsion; O/W, oil-in-water; OS-PVS, organ surface primo vascular system; PEG, polyethylene glycol; PP, primo pore; Preclin.:Preclinical; PVS, primo-vascular system; Rq, root mean square roughness; SCFAs, short-chain fatty acids; SeNPs, selenium nanoparticles; SMEDDS, self-microemulsifying drug delivery system; SOD, superoxide dismutase; TG, TGNYKALHPHNG peptide; Tg, transgenic; TNF-α, tumor necrosis factor alpha; TPP, sodium tripolyphosphate; vs, versus; wk, week; ↑, increase; ↓, decrease. Considering the limitations of conventional treatments in targeting these systemic factors affecting AD, nanotechnology offers novel strategies for modulating the GBA through microbiome engineering, targeted drug delivery, and immunomodulation. Nanostructures can restore gut microbial balance by selectively delivering prebiotics, probiotics, or antimicrobial agents[107]. Chiral gold nanoparticles alter the gut microbiota composition in Alzheimer’s disease animal models, improve cognitive function outcomes, and reduce neuroinflammation. These nanoparticles have increased levels of beneficial metabolites, such as indole-3-acetic acid, which can cross the BBB and exhibit neuroprotective effects [108]. Chitosan-based nanoparticles have been used to encapsulate beneficial bacteria or bioactive compounds, enhancing their stability and targeted release in the intestinal lumen. These nanoparticles protect probiotics from gastric degradation and improve their colonization efficiency, thereby facilitating a better microbiome composition associated with reduced systemic inflammation and improved neurodegenerative outcomes such as Alzheimer’s disease [109, 110]. Sodium oligomannate (GV-971), a seaweed-derived compound, has been shown to modulate the gut microbiota composition and reduce neuroinflammation in models of AD. Currently, phase four clinical trials are underway to confirm the efficacy of this drug, highlighting the importance of nanoformulations in enhancing bioavailability and drug properties [111]. In addition to microbiome modulation, preclinical studies suggest that nanotechnology enables targeted delivery of anti-inflammatory and neuroprotective agents to mitigate downstream effects of experimental microbiome dysbiosis. Gut-derived inflammatory mediators may enter systemic circulation and modulate central inflammatory tone via vagal afferents and peripheral immune trafficking [112]. Human translational evidence for this cascade remains limited and requires longitudinal clinical cohorts. Lipid nanoparticles functionalized with bacterial extracellular vesicles have been engineered to deliver miRNAs that specifically target the commensal Lactobacillus rhamnosus, thereby increasing bacterial proliferation and the production of indole-3-carboxaldehyde. This approach has been shown to restore intestinal barrier integrity and mitigate inflammation in colitis models, potentially preventing the translocation of bacterial endotoxins such as lipopolysaccharides (LPS), which can activate glial cells and exacerbate neuronal damage [113].Additionally, nanogels containing resveratrol and selenium based on chitosan (Res@SeNPs@Res-CS-NPs) with a high loading capacity (64%) have been designed to reduce inflammation and oxidative stress in Alzheimer’s disease models accompanied by metabolic disorders by the modulation of the gut microbiota. These nanogels improve insulin resistance and lipid accumulation by adjusting the Firmicutes to Bacteroidetes ratio and modulating glucose-related metabolic disorders through the JNK/AKT/GSK3β pathway. Moreover, by reducing LPS production and resulting neuroinflammation, these nanoparticles restore gut microbiota homeostasis. Furthermore, Res@SeNPs@Res-CS-NPs exerts protective effects by regulating bacteria associated with oxidative stress and inflammation, such as Enterococcus and Ruminococcus. This nanosystem has high potential for treating AD accompanied by metabolic disorders by improving cognitive function in animal models [114]. Selenium nanoparticles (SeNPs), which enhance the effects of probiotics, especially Lactobacillus casei ATCC 393 (L. casei ATCC 393-SeNPs), can improve the balance of the gut microbiota. This combination not only strengthens intestinal barrier function by increasing antioxidant capacity and regulating the immune response but also increases the levels of short-chain fatty acids (SCFAs) and neurotransmitters. These changes lead to reduced microglial activation and protect neurons against oxidative stress and neuroinflammation. Additionally, intervention with L. casei ATCC 393-SeNPs modulates the Akt/CREB/BDNF signaling pathway, which is effective in improving cognitive impairment. Therefore, SeNPs enriched with probiotics have significant therapeutic potential in Alzheimer’s disease models through the modulation of the microbiota-GBA [115]. Studies have shown that CS@Se nanoparticles have significant therapeutic effects on Alzheimer’s disease models by modulating the microbiota-GBA. These nanoparticles restore the gut microbiome balance by increasing the diversity and richness of the gut microbiota, increasing the abundance of beneficial bacteria such as Firmicutes and Actinobacteria, and reducing the abundance of harmful bacteria such as Bacteroidetes and Proteobacteria. The resulting changes lead to reduced systemic inflammation and improved intestinal barrier function, which in turn prevents the entry of bacterial toxins such as LPS into the bloodstream. Moreover, CS@Se helps reduce oxidative stress and neuroinflammation by increasing the production of neuroprotective metabolites such as SCFAs. CS@Se nanoparticles play an important role in maintaining neuronal health by enhancing mitochondrial function and improving the synaptic structure in the hippocampus. Moreover, CS@Se prevents overactivation of microglia and the creation of an inflammatory environment in the brain by reducing the abundance of inflammatory bacteria such as Helicobacteraceae. These multiple mechanisms indicate that CS@Se has high potential for improving cognitive and neurological disorders in AD through the regulation of the GBA (Fig. 5) [116].Fig. 5Nanoparticles modulate the GBA by restoring microbial balance, strengthening intestinal barrier, and reducing neuroinflammation ultimately improving cognitive function in AD. GBA: gut-brain axis; SCFAs: short-chain fatty acids; LPS: lipopolysaccharides Nanoparticles modulate the GBA by restoring microbial balance, strengthening intestinal barrier, and reducing neuroinflammation ultimately improving cognitive function in AD. GBA: gut-brain axis; SCFAs: short-chain fatty acids; LPS: lipopolysaccharides Epigenetic regulation via nanostructures: emerging strategies Epigenetic regulation is a new approach in the treatment of AD that, by regulating gene expression without altering the DNA sequence, enables the restoration of normal neuronal function and counteracts neurodegeneration. Epigenetics involves reversible changes such as DNA methylation, histone modification, and the regulation of noncoding RNAs. Through the use of nanostructures, these changes can be precisely and locally targeted in damaged brain regions. Owing to their unique physicochemical properties, nanostructures provide an efficient platform for delivering epigenetic drugs to the brain, crossing the BBB, enhancing treatment efficacy, and reducing systemic side effects [120, 121]. Epigenetic changes play an important role in the pathogenesis of Alzheimer’s disease; these changes affect the expression patterns of genes that regulate synaptic plasticity, neuroinflammation, and neuron survival. Alterations in DNA methylation and histone acetylation lead to the suppression or activation of genes critical for cognitive function and neuronal health. Therapeutic strategies targeting these changes improve synaptic function and slow cognitive decline by restoring epigenetic balance. However, challenges such as limited drug access to the brain, off-target effects, and low bioavailability have hindered the clinical progress of epigenetic drugs [122]. Owing to their special properties, such as smaller particle size, specific morphology, zeta potential, chemical composition, solubility, and functionalization, nanoparticles provide a versatile system for drug delivery to the brain [123]. Researchers, by employing nanostructures, aim to improve the delivery of epigenetic agents, including small interfering RNAs (siRNAs) and other therapeutic compounds, directly to the brain, thereby increasing their bioavailability and reducing off-target effects [124]. One of the known mechanisms in AD is the increased activity of HDAC enzymes, which leads to a reduction in histone acetylation. This change in chromatin structure results in the suppression of genes essential for neurogenesis and neuronal survival. The use of HDAC inhibitors (HDACis) is an effective therapeutic strategy that, by restoring histone acetylation, improves the expression of neuroprotective genes. However, issues such as systemic toxicity and low bioavailability have limited the use of these drugs. NPs, especially carriers such as polymeric nanoparticles, liposomes, and dendrimers, have made it possible for HDACis to be delivered specifically to damaged areas of the brain. These carriers deliver the drug to the appropriate site via passive mechanisms (such as the EPR effect) or active targeting, thereby increasing therapeutic efficacy [125]. Additionally, epigenetic changes, such as dysregulation of the expression of miRNAs, play important roles in the development of AD. MiR-137 is one of these miRNAs that is dysregulated in patients with AD and is used as a noninvasive biomarker in blood for the diagnosis and prediction of this disease. With advances in nanotechnology, gold nanoparticles (AuNPs) have been used as sensitive and precise diagnostic tools in the development of DNA-based nanobiosensors for the detection of miRNAs. In one study, hybridization chain reaction (HCR) amplification and the colorimetric properties of gold nanoparticles were used for visual detection of miR-137 without the need for enzymes, complex chemical reactions, or expensive devices. This system has a detection limit of 0.25 nanomolars and can effectively distinguish between matched, mismatched, and noncomplementary miRNAs. It also demonstrated good performance in analyzing real clinical samples such as blood plasma. This simple and efficient nanobiosensor can potentially be used clinically in the future for the early diagnosis of Alzheimer’s disease by directly measuring miR-137 in blood samples without the need for RNA extraction or amplification [126]. Nanostructures have been investigated as delivery vehicles for advanced epigenetic editing technologies such as CRISPR/dCas9, aimed at modulating gene networks related to synaptic plasticity and neuroinflammatory signaling in experimental models. While in vitro and rodent data indicate promising specificity, the clinical feasibility, immunogenicity of delivery vectors, and long-term epigenetic stability in human neurons remain under active investigation and are not yet established therapeutic modalities.. The use of smart nanoparticles for delivering CRISPR-dCas9 not only reduces brain accessibility issues but also provides longer-lasting and more stable functionality [127]. Additionally, the MOF-CRISPR platform (herein referred to as CMOPKP) is a novel nanozyme-based therapeutic system engineered to regulate redox homeostasis in AD. This multifunctional platform utilizes metal–organic framework (MOF) nanoparticles to deliver a CRISPR-based gene activation system targeting the Nrf2 signaling pathway, thereby upregulating endogenous antioxidant proteins. To enable brain targeting, CMOPKP is functionalized with the peptide KLVFFAED, which facilitates BBB penetration and directs the system to regions under oxidative stress. Furthermore, the incorporation of cerium dioxide (CeO₂) nanoparticles, which exhibit catalase-mimetic activity, enhances the platform’s capacity to scavenge ROS and mitigate oxidative damage. In vivo studies in AD mouse models have demonstrated that CMOPKP significantly ameliorates cognitive deficits. Thus, this integrated approach combining MOF-mediated delivery, CRISPR-based gene activation, and nanozyme activity represents a promising therapeutic strategy for combating oxidative stress in AD [128]. Nanostructures targeting glucose metabolism in AD Glucose is the only energy source in the body capable of crossing the BBB. This molecule, known as blood sugar, provides the necessary energy for normal brain function [129]. Research has shown that in patients with AD and animal models of this disease, the process of glucose uptake and consumption is significantly reduced because of the accumulation of oxidative damage in various tissues of the body, especially in the brain [130, 131]. On the basis of imaging studies using the FDG-PET method (2-[18F] fluoro-2-deoxy-D-glucose), reduced glucose metabolism in specific areas of the brain, such as the parietotemporal cortex and hippocampus, is detected years before the clinical symptoms of Alzheimer’s disease appear. This reduction in metabolism is associated with impaired synaptic and neuronal function and may result from disturbances in glucose transport and phosphorylation. Additionally, oxidative stress caused by disrupted energy metabolism is considered an important mechanism in neuronal damage and disease progression. Studies indicate that these metabolic changes are likely independent of amyloid and tau plaque accumulation and serve as independent markers for predicting disease progression [131]. The development of approaches that can modulate these metabolic disorders has been proposed as one of the treatment methods for Alzheimer’s disease. Among these methods is the use of smart nanostructures that, by targeting metabolic pathways related to glucose, help improve the energy supply and neuronal function. For example, Ab-TP-MDNPs, by crossing the BBB and reaching damaged areas of the brain, enhance glucose metabolism and modify the internal environment of the brain. These MnO2-based nanoparticles reduce oxidative stress and brain hypoxia and, by producing local oxygen, direct glucose metabolism from anaerobic glycolysis toward more efficient oxidative phosphorylation. The resulting metabolic changes lead to increased ATP levels and improved mitochondrial function. Additionally, treatment with this nanodrug increases the level of O-GlcNAcylation, indicating improved glucose availability and better regulation of cellular signaling. Furthermore, the reduction in lactate levels and the expression of MCT2 (lactate transporter) indicate that glucose metabolism has shifted toward greater efficiency. An improved energy supply leads to enhanced synaptic function and increased LTP (long-term potentiation) in the hippocampus, which is essential for memory and learning. These findings emphasize the importance of targeting metabolic disorders in Alzheimer’s disease via the use of multifunctional nanodrugs [45]. On the other hand, metabolic factors such as obesity and insulin resistance also directly affect glucose metabolism in the brain and play a role in the progression of AD. Chitosan nanoparticles carrying resveratrol (TG-Res-CS/TPP+-NPs), by targeting the brain, cross the BBB and help improve cognitive function in an Alzheimer’s disease mouse model. These nanoparticles enhance glucose metabolism in the brain by increasing the expression of glucose transporters and reducing insulin resistance caused by obesity. Additionally, by increasing the activity of antioxidant enzymes and inhibiting microglial activation, they regulate oxidative stress and neuroinflammation. Moreover, TG-Res-CS/TPP+-NPs indirectly affect glucose homeostasis and brain health by modulating the gut microbiota associated with inflammation and lipid metabolism [21]. The link between type 2 diabetes and cognitive disorders has also been well established through mechanisms involving glucose metabolism and oxidative stress. Empagliflozin nanoparticles (EMPA Nps), by selectively inhibiting the SGLT2 transporter, reduce blood glucose levels and prevent metabolic disorders in the brain. Studies have shown that these nanoparticles have protective effects on cognitive function by reducing the levels of inflammatory markers such as IL-6, IL-1β, and TNF-α. Furthermore, EMPA Nps prevent neural damage caused by hyperglycemia by increasing the levels of oxidative stress parameters, including SOD, catalase, and TBARS. Proper nanoparticle formulation increases the efficacy of EMPA compared with the free drug form [132]. Additionally, mannose-fused nanoparticles (FTY@Man NPs) utilize glucose transporter 1 (GLUT1) to cross the BBB through a glucose control strategy. These smart nanoparticles, with a PLGA–PEG structure and surface modification with mannose, first pass through the intestinal epithelial barrier (IEB) and then, by binding to GLUT1, target the brain specifically. After entering the brain, the released fingolimod (FTY) helps improve the metabolic environment of the brain by modulating microglial activity and reducing neuroinflammation. This nanosystem not only optimizes glucose transport but also improves cognitive function in Alzheimer’s disease models by reducing oxidative stress and inflammation. Therefore, FTY@Man NPs, which combine mannose-based targeting and glucose regulation, offer a multifaceted approach for treating Alzheimer’s disease via oral administration (Fig. 6) [18].Fig. 6Glucose-targeting nanoparticles restore brain energy metabolism by enhancing glucose transport, reducing hypoxia/oxidative stress, and improving mitochondrial ATP production rescuing synaptic function in AD. SOD: superoxide dismutase; FTY@Man NP: mannose-fused nanoparticles; GLUT1: glucose transporter 1; ATP: adenosine triphosphate; LTP: long-term potentiation; IL-1β: interleukin 1 beta Glucose-targeting nanoparticles restore brain energy metabolism by enhancing glucose transport, reducing hypoxia/oxidative stress, and improving mitochondrial ATP production rescuing synaptic function in AD. SOD: superoxide dismutase; FTY@Man NP: mannose-fused nanoparticles; GLUT1: glucose transporter 1; ATP: adenosine triphosphate; LTP: long-term potentiation; IL-1β: interleukin 1 beta Bridging the gap: from preclinical promise to clinical reality in NANT-Targeted nanotherapies Expanding landscape of NANT biomarkers in AD The pathophysiological complexity of AD extends beyond Aβ and tau, driving significant research into NANT biomarkers and therapeutic targets. These emerging avenues aim to capture and modulate diverse biological processes central to AD progression. Neurodegeneration-specific markers, such as the neurofilament light chain (NFL), provide a direct measure of axonal injury and neuronal loss. NFL levels in both cerebrospinal fluid (CSF) and blood correlate with disease progression and cognitive decline, independent of Aβ status, highlighting its utility as a dynamic tracker of neurodegeneration [133]. Synaptic dysfunction, an early event in AD, is reflected by CSF biomarkers such as neurogranin and SNAP-25, which are strongly linked to cognitive impairment and synaptic density loss [134, 135]. Neuroinflammatory processes, which are increasingly recognized as critical drivers of pathology, are quantified through proteins such as glial fibrillary acidic protein (GFAP) and chitinase-3-like protein 1 (YKL-40). These markers not only differentiate AD patients from controls but are also are associated with specific pathological subtypes and clinical stages [136, 137]. Furthermore, disruptions in lipid metabolism and protein clearance mechanisms offer additional biomarkers. Altered levels of CSF progranulin and plasma ceramides, for instance, are implicated in AD-related inflammatory responses and vascular copathologies [118, 138]. The systematic exploration of these NANT pathways is critical for addressing gaps in AD drug development, particularly the need for biomarkers that can monitor treatment response and index clinical severity beyond Aβ-targeting therapies. As reviewed by Park and colleagues, the combination of fluid biomarkers spanning neurodegeneration, synaptic integrity such as az neurogranin, neuroinflammation (e.g., YKL-40, progranulin), and other noncore pathologies provides a multidimensional framework for characterizing AD [139]. The integration of such fluid-based NANT biomarkers into multimodal panels enhances the granular characterization of AD heterogeneity, facilitating patient stratification and the monitoring of therapeutic interventions that target these broader pathophysiological mechanisms. Evaluation metrics, validation strategies, and comparative framework across non-amyloid pathways A structured methodological framework is essential for evaluating nanoscale interventions across the diverse non‑amyloid, non‑tau pathways implicated in AD. To ensure analytical consistency, each pathological axis is examined via quantitative, pathway‑specific metrics supported by orthogonal validation strategies. This integrated approach enables direct comparisons between targeted and non‑targeted nanoparticles and provides a coherent basis for interpreting performance differences across studies. Neuroinflammatory modulation is typically quantified through cytokine profiling (TNF‑α, IL‑1β, IL‑6, and IL‑10), microglial phenotype ratios, and the transcriptional activity of inflammatory regulators such as NF‑κB. These measurements were validated through ELISA, immunohistochemistry, and flow cytometry. Comparative analyses consistently show that ligand‑functionalized nanoparticles suppress proinflammatory signaling more effectively and promote stronger restoration of reparative microglial states than nonfunctionalized carriers do [140, 141]. Compared with free compounds, polymeric and nucleic acid–based nanostructures have been shown to restore autophagic flux more effectively in neurodegenerative models, particularly under conditions of early lysosomal impairment [142] ER‑stress regulation in AD is typically characterized through the expression of markers such as CHOP and GRP78/BiP and the activation of the PERK branch of the unfolded protein response. These indicators are widely used across experimental studies to assess ER stress-related vulnerability in neuronal systems [143, 144]. Synaptic integrity was examined through electrophysiological recordings (LTP, LTD), synaptic protein levels, and dendritic spine density. Patch‑clamp recordings, Golgi staining, and high‑resolution microscopy provide orthogonal validation. Compared with nontargeted systems, targeted nanoparticles outperform nontargeted systems in restoring synaptic plasticity in AD models [145, 146]. Neurovascular and BBB‑related dysfunction in AD is commonly assessed through transendothelial electrical resistance, permeability measurements, and tight junction protein expression, such as that of claudin‑5 and occludin. These parameters are well‑established in both in vitro BBB models and in vivo imaging studies of barrier breakdown [147]. Although various ligand‑modified nanoparticles have been explored for enhancing BBB penetration in neurological disorders, direct comparative evidence in Alzheimer‑specific models remains limited and continues to develop. Gut–brain axis modulation was assessed through microbial diversity indices, short‑chain fatty acid levels, and inflammatory metabolite profiling, which were validated via 16S rRNA sequencing and metabolomics. These approaches are well‑established in studies of microbiota–brain interactions [148]. Although nanocarrier‑based delivery systems have been explored for enhancing the stability and bioavailability of polyphenols and anti‑inflammatory agents, direct evidence for microbiome normalization beyond free compounds remains limited in the current literature. Epigenetic regulation is assessed through DNA methylation patterns, histone modification profiles, and the expression of regulatory microRNAs. These metrics were validated through bisulfite sequencing and transcriptomic analyses. Compared with conventional formulations, nanoparticles engineered for nucleic acid delivery demonstrate stronger modulation of epigenetic regulators [149]. Across all domains, validation strategies incorporate multi‑model replication, dose‒response analyses, temporal profiling, and cross‒platform confirmation via biochemical, imaging, and functional assays. Comparative evidence across independent studies has demonstrated that targeted nanoparticles consistently achieve stronger pathway correction, lower required dosages, and improved therapeutic indices than nontargeted formulations do. This integrated evaluation framework supports the performance claims of targeted nanostructures and aligns with methodological expectations for predictive and mechanistic analyses in nonamyloid AD research. Alternative administration routes: intranasal delivery as a translational strategy The intranasal route constitutes a noninvasive administration pathway that circumvents the blood–brain barrier through direct transport along olfactory and trigeminal neural projections, thereby facilitating rapid delivery of therapeutics to the central nervous system [150]. This route eliminates first-pass hepatic metabolism and minimizes systemic exposure, consequently reducing off-target toxicity while enhancing drug bioavailability in cerebral compartments [151]. Nanostructured carriers engineered for intranasal delivery demonstrate enhanced mucoadhesion to the nasal epithelium and sustained release kinetics, which collectively improve the residence time and transmucosal permeation. Comparative preclinical evidence from cerebral ischemia models substantiates the translational feasibility of intranasal nanocarrier systems, demonstrating their capacity to achieve therapeutic brain concentrations while circumventing systemic clearance mechanisms [152]. The intranasal administration of surface-engineered nanoparticles has emerged as a promising modality for delivering anti-inflammatory and mitochondria-protective agents directly to affected brain regions, bypassing the BBB via the olfactory and trigeminal pathways [152–154]. Chitosan-based systems leverage mucoadhesive properties to prolong nasal residence time and enhance paracellular transport across the olfactory epithelium [155]. While mitochondrion-targeted nanoparticles functionalized with TPP⁺ have demonstrated significant efficacy in reducing mitochondrial ROS and restoring the membrane potential in AD models following systemic administration [37, 38], the application of intranasal delivery for mitochondrion-targeted nanotherapeutics remains an underexplored area requiring dedicated preclinical investigation. Overcoming barriers to clinical translation Long-term biocompatibility and neurotoxicity considerations The clinical translation of central nervous system-targeted nanotherapeutics requires a rigorous and proactive evaluation of their long-term biological fate and safety within the uniquely sensitive cerebral microenvironment. A fundamental concern lies in the biopersistence and elimination pathways of nondegradable inorganic nanostructures, such as those composed of gold or silver [114, 156]. Although nanoclusters less than five nanometers in size may be subject to eventual renal clearance, larger particles or those modified with dense polymeric coatings risk sustained sequestration by resident microglia [157, 158]. This long-term retention can precipitate chronic, low-grade inflammation or lysosomal storage disorders over extended periods, potentially spanning years [159]. Additionally, the cumulative impact of repeated therapeutic administration on the functional integrity of the neurovascular unit requires systematic investigation. Certain design components, such as cationic targeting ligands, present specific risks; for example, high local concentrations of molecules such as TPP+ have been shown to disrupt the mitochondrial membrane potential in neural cell lines [160]. These findings underscore the critical need for precise therapeutic dose window characterization. For strategies aiming to durably modulate processes such as neuroinflammation or mitochondrial dysfunctions, which have evolved over decades in AD, material biopersistence has become a critical design parameter rather than a secondary concern. Consequently, the field is witnessing a strategic shift toward intrinsically biodegradable systems [161]. PLGA nanoparticles, with hydrolysis‑driven degradation kinetics tunable from weeks to months, exemplify this approach for sustained drug release without permanent CNS residence [162]. Similarly, silica‑based mesoporous nanoparticles engineered with redox‑ or enzyme‑cleavable disulfide bridges demonstrate stimulus‑responsive disintegration after payload delivery, mitigating long‑term accumulation [163]. This design philosophy extends to inorganic cores; for example, manganese dioxide (MnO₂) nanosystems catalytically decompose into bioeliminable Mn2⁺ ions in the reductive microenvironment of inflamed or hypoxic brain tissue, simultaneously alleviating oxidative stress and ensuring clearance [164, 165]. The integration of such built‑in elimination mechanisms represents a foundational advancement in nanomedicine, aligning the material lifetime with the chronic yet potentially reversible nature of the NANT pathologies they are engineered to correct. Clinical translation status and regulatory pathways Despite considerable preclinical advancements in nanotherapeutic approaches focused on NANT mechanisms, their translation into clinical practice remains at an early stage. To date, no nanoparticle-based formulation has progressed to registered clinical trials explicitly for AD therapy [98, 166]. A notable exception within the broader landscape of NANT therapeutics is sodium oligomannate (GV-971), a seaweed-derived oligosaccharide that received conditional approval from China’s National Medical Products Administration in 2019 for mild-to-moderate AD [168]. Unlike conventional nanomedicines, GV-971 operates through gut microbiota remodeling rather than direct nanoparticle delivery; however, its mechanism aligns with the nanoformulation strategies targeting the GBA described in this review. A phase III randomized controlled trial (NCT02293915) demonstrated statistically significant cognitive improvement as early as week 4, with sustained effects through 36 weeks. Mechanistically, GV-971 has been shown to suppress gut dysbiosis, reduce the peripheral accumulation of phenylalanine and isoleucine, inhibit neuroinflammation driven by Th1 cell infiltration and M1 microglial activation, and ultimately improve cognitive function in both transgenic AD models and human patients [169]. Regulatory frameworks for nanotherapeutics that target subcellular organelles, particularly mitochondria and lysosomes, remain underdeveloped. Although mitochondrion-targeted compounds such as MitoQ have undergone clinical evaluation for Parkinson’s disease, no analogous trials exist for AD [170]. Similarly, the use of lysosome-modulating agents has not progressed beyond preclinical validation despite compelling evidence of lysosomal acidification defects in early Alzheimer’s disease pathogenesis [167, 171]. Critical prerequisites for clinical advancement include standardized characterization of nanoparticle–biological interface interactions, establishment of organ-specific toxicity thresholds, and development of imaging modalities capable of tracking subcellular nanoparticle localization in humans [172]. Integration of these translational elements into future research agendas will bridge the current divide between promising preclinical nanotherapeutics and clinical implementation. Manufacturing scalability and regulatory navigation complexities Reproducing nanoparticles with the precise characteristics required for targeting nonamyloid pathways becomes exponentially more difficult beyond milligram-scale batches [174]. Critical unit operations such as lyophilization require complete reoptimization at larger scales to prevent nanoparticle aggregation and ensure stability [175] while simultaneously preserving the activity of sensitive targeting ligands. This comprehensive process optimization for complex designs can demand extensive development time, with associated production costs increasing significantly for multifunctional systems. Furthermore, regulatory agencies require exhaustive characterization of a nanomaterial’s dynamic biological identity, including protein corona formation and cell-specific interactions, adding another layer of complexity to process validation and control [176, 177]. The combined demands of industrial-scale synthesis and rigorous regulatory evaluation extend the development trajectory for AD nanomedicines, with estimates suggesting a long delay relative to conventional drug development cycles [178]. Safety, toxicology, long-term biodistribution, and regulatory challenges While preclinical nanotherapeutics targeting NANT pathways demonstrate compelling mechanistic efficacy in controlled experimental settings, their clinical translation is constrained by unresolved pharmacological and safety considerations. Upon intravenous administration, rapid adsorption to plasma proteins forms a dynamic corona that masks engineered targeting ligands, alters pharmacokinetics, and may trigger unintended immune recognition or accelerated hepatic clearance [54, 55, 59]. Longitudinal biodistribution data in chronic disease models remain sparse, and the persistent accumulation of non-biodegradable inorganic cores (e.g., Au, Ag, CeO₂) within microglia [156, 157, 179] or lysosomal compartments raises theoretical concerns regarding chronic low-grade inflammation or secondary lysosomal storage-like pathology over multi-year treatment horizons [159]. Toxicological profiling must therefore transition from acute, single-dose assessments to chronic, dose-escalation studies that systematically evaluate neurovascular integrity, microglial activation thresholds, and renal/hepatic elimination kinetics [158, 167, 172]. Furthermore, regulatory frameworks for subcellular-targeted nanomedicines remain fragmented [176]. Health authorities require standardized characterization of the nanomaterial-biological interface, reproducible Good Manufacturing Practice (GMP) scale-up protocols, and validated fluid-based surrogate biomarkers for target engagement in human populations [139, 173, 180]. Until these safety, toxicological, and regulatory gaps are systematically addressed through harmonized preclinical-to-clinical pipelines, human applications will remain incremental [167, 178]. Future perspectives and translational roadmap To translate compelling preclinical data into clinical impact, the field must navigate several important frontiers. First, the rational design of multifunctional nanoparticles must be guided not only by efficacy but also by manufacturability and regulatory predictability. This includes adopting quality-by-design principles early in development to control critical quality attributes such as ligand density and batch-to-batch uniformity. Second, moving beyond single-target approaches, the next generation of smart nanoparticles should incorporate disease-responsive elements to release therapeutics precisely within compromised subcellular compartments. Third, the integration of artificial intelligence and machine learning for in silico prediction of nanoparticle‒protein corona formation and BBB penetrability holds immense promise for accelerating candidate selection. Fourth, establishing validated surrogate biomarkers, such as ligands for synaptic density or mitochondrial complex I activity, in early-phase clinical trials will be crucial for demonstrating target engagement and the biological effect of these sophisticated therapies. A collaborative, interdisciplinary framework uniting material scientists, neurologists, and regulatory experts is paramount to translate this systems-level nanomedicine approach into tangible patient benefit [96, 180, 181]. The impact of such therapies will be further amplified by parallel advances in clinical analytics, such as statistical modeling of longitudinal neurodegenerative trajectories via conditional diffusion frameworks, which enable precise, amyloid-independent disease staging and prognosis [182]. A comparative synthesis of nanoparticles discussed across diverse nonamyloid and nontau pathways is presented in Table 5. This table systematically organizes the performance, advantages, and evidence-based limitations of nanostructured materials in preclinical models and provides a framework for prioritizing multifunctional strategies in translational development.Table 5Summary of some nanoparticles targeting NANT pathways in ADMaterial typeExample formulationMechanism actionTarget pathwayStudy typeAdvantages (specific)Limitations (evidence-based)OutcomesRef (s)LiposomesBaicalein-NLPCytotoxicity↓, uptake↑, oxidative stress↓Oxidative stressHuman/Rat cell lines (SH-SY5Y neuroblastoma, PC12 pheochromocytoma; in vitro)Cytotoxicity↓, neuroprotection-Cell viability: ↔ ~ 100% at 280 μM (vs. untreated control)[183]Silica NPsCatechin-SiO₂ NPsCell survival↑ vs. metal stressCu(II) oxidative stressRat (primary hippocampal neurons; in vitro)Solubility↑, protection vs. metalBare SiO₂ toxicity (needs coating)Cell viability: 98 ± 2% at 5 μM, 97 ± 2% at 25 μM (vs. Cu(II)-stressed control[184]Silica NPsQuercetin-SiO₂ NPsAntioxidant activity↑Cu(II) oxidative stressRat (hippocampal cell line; in vitro)Loading capacity↑, controlled releaseBare SiO₂ causes ROS/inflammationCell viability: 96 ± 2% vs. 55 ± 2% in Cu(II)-stressed control[185]PLGACurcumin-PLGA NPsSolubility↑, ROS↓Nrf2 pathway, ROS↓Human cell line (SK-N-SH neuroblastoma; in vitro)Solubility↑, uptake↑, no cytotoxicityPoor bioavailability of free curcuminROS production: ↓; Nrf2 pathway activation: ↑ at 0.07[186]SLNsQuercetin-SLNsEntrapment efficiency↑Memory/cognitionRat (strain not specified, adult; in vivo) + Zebrafish (Danio rerio, adult; in vivo)High drug loadingLimited long-term toxicity dataLipid peroxidation (MDA): ↓ to 17–25% of control; nitrite levels: ↓ to 16–26% of control[187]NLCsQuercetin-NLCsResidence time↑Stress resistanceHuman cell line (Caco-2 intestinal epithelial; in vitro) + Rat (strain not specified, adult; in vivo)Stress resistance↑, residence time↑No specific limitationsOral bioavailability: ↑ 5.4-fold; brain uptake: ↑ 5.6-fold vs. free drug[188]NPQQuercetin-Zein NPsOral bioavailability↑Cognition↑, astrogliosis↓Mouse (SAMP8 senescence-accelerated model, age not specified; in vivo)Bioavailability↑60%, sustained releaseLimited human dataMemory performance: ↑; astrocyte activation (GFAP⁺): ↓ 50%; brain quercetin concentration: 371 ng/g tissue[189]LiposomesLactoferrin-Quercetin LiposomesBBB transcytosis (receptor-mediated)Apoptosis (caspase)Human cell line (SK-N-MC neuroepithelioma; in vitro) + Rat (strain not specified, adult; in vivo)BBB penetration↑, apoptosis↓ at low doseNeeds surface modificationCell viability: ↑ 30–40% vs. control; apoptotic protein expression: ↓[190]AuNPsGSH-AuNPsAntioxidant, Nrf2 activation, ER stress↓Nrf2/ARE, ER stress, mitochondriaHuman neural stem cells (hNSCs; in vitro)Multitarget, stable, low cytotoxicity (10 ppm)In vitro only, no in vivo/BBB dataCell viability: ↑; caspase-3/9 activity: ↓; ROS production: ↓; Nrf2 pathway activation: ↑[32]Albumin NPR-Flurbiprofen-Albumin NPNose-to-brain, mitochondrial uncouplingMitochondrial dysfunctionMouse (strain not specified, adult; in vivo) + Hamster cell line (CHO-APP695; in vitro)BBB bypass, brain deposition↑, mitochondrial function↑Complex production, carrier effects confoundBrain Cₘₐₓ: ↑ 2.5-fold; AUC: ↑ 2.7-fold vs. free drug; mitochondrial respiration: ↑ 60–65%[191]GNPsCitrate-GNPsNADH oxidation (Complex I), NF-κB↓, ROS scavengingMitochondrial OXPHOS, inflammationRat (strain not specified, adult; in vivo)Restores ATP, reduces stress/inflammation, no toxicity (2.5 mg/kg)Incomplete characterization, no long-term safetyMemory performance: ↑; superoxide production: ↓; nitrite levels: ↓; NF-κB activation: ↓[192]SLNsResveratrol-SLNs (R-SLNs)Brain delivery↑, mitochondrial ROS scavengingNrf2/HO-1, mitochondrial stressRat (strain not specified, adult; in vivo)Cognitive deficits↓, oxidative stress↓Animal model only, mechanism unclearBrain resveratrol concentration: ↑ 4.5-fold vs. free drug; ROS production: ↓ 31–38%[193]PdH NPsPdH-H₂ NPsSustained H₂ release, radical scavengingOxidative stress, Nrf2, mitochondriaMouse (3 × Tg-AD transgenic, age not specified; in vivo) + Human AD cell model (in vitro)High H₂ loading, self-catalytic, low toxicity (6.25–25 µg/mL)Invasive injection (intracerebral)Hydroxyl radical (·OH): ↓; cognitive performance: ↑; synaptic density: ↑; caspase-3 activity: ↓[194]AuNPsColloidal AuNPsAnti-inflammatory, antioxidant, mitochondria modulationOxidative stress, cytokines, neurotrophinsRat (strain not specified, adult; in vivo)No toxicity (2.5 mg/kg), memory↑, antioxidant enzymes↑Long-term safety unknown, IP route suboptimalOxidative stress markers: normalized (protein sulfhydryl content restored to baseline,[195]HstNHesperetin (in ethanol)ATP↑, ETC activity↑, ROS↓Mitochondrial bioenergeticsHuman cell line (SH-SY5Y neuroblastoma; in vitro)Direct mitochondrial target, no cytotoxicity (≤ 10 µM)Mechanism unclearATP production: ↑ 9% (p < 0.05); peroxidase activity: ↓[196]Pd NPsPEG/Borneol-Pd NPsROS scavenging, mitochondria protection, BBB penetration↑ROS↓, mitochondria, anti-inflammationHuman cell line (neuroblastoma; in vitro) + Mouse (AD transgenic model, age not specified; in vivo)High BBB permeability, hemolysis < 5% at 200 μg/mL; cell viability > 90%; apoptosis < 4%Long-term toxicity and biodistribution not fully characterizedNeuronal apoptosis rate: ↓ from 27.4% to 5.0%; pro-inflammatory IL-6: ↓; microglial activation (Iba-1⁺): ↓; astrocyte activation (GFAP⁺): ↓[25]PLGANO donor (PN)-PLGA NPsSustained NO release, mitochondrial biogenesis↑Mitochondrial biogenesis/functionMouse (strain not specified, adult; in vivo)No hemolysis, memory↑ (free PN ineffective)Qualitative release/brain accumulation dataNO release kinetics: sustained (threefold slower release vs. free PAPANONOate)[49]HSA NPsT807/TPP-RBC-Curcumin NPsMitochondrion-targeted antioxidant deliveryMitochondrial oxidative stressHuman cell line (neuroblastoma; in vitro) + Mouse/Rat (AD models, age not specified; in vivo)Mitochondrial targeting↑, low toxicity, hemocompatibleLow loading (~ 4.87%), short-term studiesNeuronal apoptosis rate: ↓ from 41.4% to 17.0%[37]SWCNTsSWCNT-ACh complexLysosomotropic oral uptake, MMP collapse → ROSMitochondrial damageMouse (strain not specified, adult; in vivo) + Mouse cell lines (in vitro)Oral bioavailability, crosses BBBDose-dependent toxicity (> 400 mg/kg)Cognitive performance (learning/memory): restored to baseline levels[65]LNPMG-LNP (cKK-E12 based)Anti-PU.1 siRNA delivery to microgliaPU.1 transcription factorHuman iPSC-derived microglia-like cells (in vitro) + Mouse (strain not specified, adult; in vivo)Efficient delivery to inflamed microglia, brain-targetedRequires localized injectionPU.1 gene knockdown: 92% at 500 ng/mL in vitro; neuroinflammatory markers: ↓ in vivo[23]Aβ: amyloid-beta; AD, Alzheimer’s disease; ASO, Antisense Oligonucleotide, ATP, Adenosine Triphosphate, AuNPs, Gold Nanoparticles, BBB, Blood–Brain Barrier, BDNF, Brain-Derived Neurotrophic Factor, CUR, Curcumin, ETC, Electron Transport Chain, GFAP, Glial Fibrillary Acidic Protein, GNPs, Gold Nanoparticles, GSH, Glutathione, HO-1, Heme Oxygenase-1, HSA, Human Serum Albumin, HstN, Hesperetin, IL, Interleukin, iMGLs, induced Microglia-like cells, i.c.st., Intracisternal, i.n., Intranasal, IP, Intraperitoneal, IV, Intravenous, LNP, Lipid Nanoparticle, MDA, Malondialdehyde, miRNA, microRNA, MMP, Mitochondrial Membrane Potential, NF-κB, Nuclear Factor kappa B, NGF, Nerve Growth Factor, NLCs, Nanostructured Lipid Carriers, NO, Nitric Oxide, NP/NPs, Nanoparticle(s), NPQ, Nanoencapsulated Quercetin, Nrf2, Nuclear factor erythroid 2-related factor 2, OXPHOS, Oxidative Phosphorylation, PdH, Palladium Hydride, Pd NPs, Palladium Nanoparticles, PEG, Polyethylene Glycol, PK, Pharmacokinetics, PLGA, Poly(lactic-co-glycolic acid), PN, PAPA NONOate, PU.1, Transcription factor PU.1, RBC, Red Blood Cell, RNS, Reactive Nitrogen Species, ROS, Reactive Oxygen Species, SLNs, Solid Lipid Nanoparticles, SOD, Superoxide Dismutase, SWCNTs, Single-Walled Carbon Nanotubes, TDFNs, Theranostic DNA Frameworks, TNF-α, Tumor Necrosis Factor-alpha, TPP+, Triphenylphosphonium. Table 5Summary of some nanoparticles targeting NANT pathways in ADMaterial typeExample formulationMechanism actionTarget pathwayStudy typeAdvantages (specific)Limitations (evidence-based)OutcomesRef (s)LiposomesBaicalein-NLPCytotoxicity↓, uptake↑, oxidative stress↓Oxidative stressHuman/Rat cell lines (SH-SY5Y neuroblastoma, PC12 pheochromocytoma; in vitro)Cytotoxicity↓, neuroprotection-Cell viability: ↔ ~ 100% at 280 μM (vs. untreated control)[183]Silica NPsCatechin-SiO₂ NPsCell survival↑ vs. metal stressCu(II) oxidative stressRat (primary hippocampal neurons; in vitro)Solubility↑, protection vs. metalBare SiO₂ toxicity (needs coating)Cell viability: 98 ± 2% at 5 μM, 97 ± 2% at 25 μM (vs. Cu(II)-stressed control[184]Silica NPsQuercetin-SiO₂ NPsAntioxidant activity↑Cu(II) oxidative stressRat (hippocampal cell line; in vitro)Loading capacity↑, controlled releaseBare SiO₂ causes ROS/inflammationCell viability: 96 ± 2% vs. 55 ± 2% in Cu(II)-stressed control[185]PLGACurcumin-PLGA NPsSolubility↑, ROS↓Nrf2 pathway, ROS↓Human cell line (SK-N-SH neuroblastoma; in vitro)Solubility↑, uptake↑, no cytotoxicityPoor bioavailability of free curcuminROS production: ↓; Nrf2 pathway activation: ↑ at 0.07[186]SLNsQuercetin-SLNsEntrapment efficiency↑Memory/cognitionRat (strain not specified, adult; in vivo) + Zebrafish (Danio rerio, adult; in vivo)High drug loadingLimited long-term toxicity dataLipid peroxidation (MDA): ↓ to 17–25% of control; nitrite levels: ↓ to 16–26% of control[187]NLCsQuercetin-NLCsResidence time↑Stress resistanceHuman cell line (Caco-2 intestinal epithelial; in vitro) + Rat (strain not specified, adult; in vivo)Stress resistance↑, residence time↑No specific limitationsOral bioavailability: ↑ 5.4-fold; brain uptake: ↑ 5.6-fold vs. free drug[188]NPQQuercetin-Zein NPsOral bioavailability↑Cognition↑, astrogliosis↓Mouse (SAMP8 senescence-accelerated model, age not specified; in vivo)Bioavailability↑60%, sustained releaseLimited human dataMemory performance: ↑; astrocyte activation (GFAP⁺): ↓ 50%; brain quercetin concentration: 371 ng/g tissue[189]LiposomesLactoferrin-Quercetin LiposomesBBB transcytosis (receptor-mediated)Apoptosis (caspase)Human cell line (SK-N-MC neuroepithelioma; in vitro) + Rat (strain not specified, adult; in vivo)BBB penetration↑, apoptosis↓ at low doseNeeds surface modificationCell viability: ↑ 30–40% vs. control; apoptotic protein expression: ↓[190]AuNPsGSH-AuNPsAntioxidant, Nrf2 activation, ER stress↓Nrf2/ARE, ER stress, mitochondriaHuman neural stem cells (hNSCs; in vitro)Multitarget, stable, low cytotoxicity (10 ppm)In vitro only, no in vivo/BBB dataCell viability: ↑; caspase-3/9 activity: ↓; ROS production: ↓; Nrf2 pathway activation: ↑[32]Albumin NPR-Flurbiprofen-Albumin NPNose-to-brain, mitochondrial uncouplingMitochondrial dysfunctionMouse (strain not specified, adult; in vivo) + Hamster cell line (CHO-APP695; in vitro)BBB bypass, brain deposition↑, mitochondrial function↑Complex production, carrier effects confoundBrain Cₘₐₓ: ↑ 2.5-fold; AUC: ↑ 2.7-fold vs. free drug; mitochondrial respiration: ↑ 60–65%[191]GNPsCitrate-GNPsNADH oxidation (Complex I), NF-κB↓, ROS scavengingMitochondrial OXPHOS, inflammationRat (strain not specified, adult; in vivo)Restores ATP, reduces stress/inflammation, no toxicity (2.5 mg/kg)Incomplete characterization, no long-term safetyMemory performance: ↑; superoxide production: ↓; nitrite levels: ↓; NF-κB activation: ↓[192]SLNsResveratrol-SLNs (R-SLNs)Brain delivery↑, mitochondrial ROS scavengingNrf2/HO-1, mitochondrial stressRat (strain not specified, adult; in vivo)Cognitive deficits↓, oxidative stress↓Animal model only, mechanism unclearBrain resveratrol concentration: ↑ 4.5-fold vs. free drug; ROS production: ↓ 31–38%[193]PdH NPsPdH-H₂ NPsSustained H₂ release, radical scavengingOxidative stress, Nrf2, mitochondriaMouse (3 × Tg-AD transgenic, age not specified; in vivo) + Human AD cell model (in vitro)High H₂ loading, self-catalytic, low toxicity (6.25–25 µg/mL)Invasive injection (intracerebral)Hydroxyl radical (·OH): ↓; cognitive performance: ↑; synaptic density: ↑; caspase-3 activity: ↓[194]AuNPsColloidal AuNPsAnti-inflammatory, antioxidant, mitochondria modulationOxidative stress, cytokines, neurotrophinsRat (strain not specified, adult; in vivo)No toxicity (2.5 mg/kg), memory↑, antioxidant enzymes↑Long-term safety unknown, IP route suboptimalOxidative stress markers: normalized (protein sulfhydryl content restored to baseline,[195]HstNHesperetin (in ethanol)ATP↑, ETC activity↑, ROS↓Mitochondrial bioenergeticsHuman cell line (SH-SY5Y neuroblastoma; in vitro)Direct mitochondrial target, no cytotoxicity (≤ 10 µM)Mechanism unclearATP production: ↑ 9% (p < 0.05); peroxidase activity: ↓[196]Pd NPsPEG/Borneol-Pd NPsROS scavenging, mitochondria protection, BBB penetration↑ROS↓, mitochondria, anti-inflammationHuman cell line (neuroblastoma; in vitro) + Mouse (AD transgenic model, age not specified; in vivo)High BBB permeability, hemolysis < 5% at 200 μg/mL; cell viability > 90%; apoptosis < 4%Long-term toxicity and biodistribution not fully characterizedNeuronal apoptosis rate: ↓ from 27.4% to 5.0%; pro-inflammatory IL-6: ↓; microglial activation (Iba-1⁺): ↓; astrocyte activation (GFAP⁺): ↓[25]PLGANO donor (PN)-PLGA NPsSustained NO release, mitochondrial biogenesis↑Mitochondrial biogenesis/functionMouse (strain not specified, adult; in vivo)No hemolysis, memory↑ (free PN ineffective)Qualitative release/brain accumulation dataNO release kinetics: sustained (threefold slower release vs. free PAPANONOate)[49]HSA NPsT807/TPP-RBC-Curcumin NPsMitochondrion-targeted antioxidant deliveryMitochondrial oxidative stressHuman cell line (neuroblastoma; in vitro) + Mouse/Rat (AD models, age not specified; in vivo)Mitochondrial targeting↑, low toxicity, hemocompatibleLow loading (~ 4.87%), short-term studiesNeuronal apoptosis rate: ↓ from 41.4% to 17.0%[37]SWCNTsSWCNT-ACh complexLysosomotropic oral uptake, MMP collapse → ROSMitochondrial damageMouse (strain not specified, adult; in vivo) + Mouse cell lines (in vitro)Oral bioavailability, crosses BBBDose-dependent toxicity (> 400 mg/kg)Cognitive performance (learning/memory): restored to baseline levels[65]LNPMG-LNP (cKK-E12 based)Anti-PU.1 siRNA delivery to microgliaPU.1 transcription factorHuman iPSC-derived microglia-like cells (in vitro) + Mouse (strain not specified, adult; in vivo)Efficient delivery to inflamed microglia, brain-targetedRequires localized injectionPU.1 gene knockdown: 92% at 500 ng/mL in vitro; neuroinflammatory markers: ↓ in vivo[23]Aβ: amyloid-beta; AD, Alzheimer’s disease; ASO, Antisense Oligonucleotide, ATP, Adenosine Triphosphate, AuNPs, Gold Nanoparticles, BBB, Blood–Brain Barrier, BDNF, Brain-Derived Neurotrophic Factor, CUR, Curcumin, ETC, Electron Transport Chain, GFAP, Glial Fibrillary Acidic Protein, GNPs, Gold Nanoparticles, GSH, Glutathione, HO-1, Heme Oxygenase-1, HSA, Human Serum Albumin, HstN, Hesperetin, IL, Interleukin, iMGLs, induced Microglia-like cells, i.c.st., Intracisternal, i.n., Intranasal, IP, Intraperitoneal, IV, Intravenous, LNP, Lipid Nanoparticle, MDA, Malondialdehyde, miRNA, microRNA, MMP, Mitochondrial Membrane Potential, NF-κB, Nuclear Factor kappa B, NGF, Nerve Growth Factor, NLCs, Nanostructured Lipid Carriers, NO, Nitric Oxide, NP/NPs, Nanoparticle(s), NPQ, Nanoencapsulated Quercetin, Nrf2, Nuclear factor erythroid 2-related factor 2, OXPHOS, Oxidative Phosphorylation, PdH, Palladium Hydride, Pd NPs, Palladium Nanoparticles, PEG, Polyethylene Glycol, PK, Pharmacokinetics, PLGA, Poly(lactic-co-glycolic acid), PN, PAPA NONOate, PU.1, Transcription factor PU.1, RBC, Red Blood Cell, RNS, Reactive Nitrogen Species, ROS, Reactive Oxygen Species, SLNs, Solid Lipid Nanoparticles, SOD, Superoxide Dismutase, SWCNTs, Single-Walled Carbon Nanotubes, TDFNs, Theranostic DNA Frameworks, TNF-α, Tumor Necrosis Factor-alpha, TPP+, Triphenylphosphonium. Summary of some nanoparticles targeting NANT pathways in AD Aβ: amyloid-beta; AD, Alzheimer’s disease; ASO, Antisense Oligonucleotide, ATP, Adenosine Triphosphate, AuNPs, Gold Nanoparticles, BBB, Blood–Brain Barrier, BDNF, Brain-Derived Neurotrophic Factor, CUR, Curcumin, ETC, Electron Transport Chain, GFAP, Glial Fibrillary Acidic Protein, GNPs, Gold Nanoparticles, GSH, Glutathione, HO-1, Heme Oxygenase-1, HSA, Human Serum Albumin, HstN, Hesperetin, IL, Interleukin, iMGLs, induced Microglia-like cells, i.c.st., Intracisternal, i.n., Intranasal, IP, Intraperitoneal, IV, Intravenous, LNP, Lipid Nanoparticle, MDA, Malondialdehyde, miRNA, microRNA, MMP, Mitochondrial Membrane Potential, NF-κB, Nuclear Factor kappa B, NGF, Nerve Growth Factor, NLCs, Nanostructured Lipid Carriers, NO, Nitric Oxide, NP/NPs, Nanoparticle(s), NPQ, Nanoencapsulated Quercetin, Nrf2, Nuclear factor erythroid 2-related factor 2, OXPHOS, Oxidative Phosphorylation, PdH, Palladium Hydride, Pd NPs, Palladium Nanoparticles, PEG, Polyethylene Glycol, PK, Pharmacokinetics, PLGA, Poly(lactic-co-glycolic acid), PN, PAPA NONOate, PU.1, Transcription factor PU.1, RBC, Red Blood Cell, RNS, Reactive Nitrogen Species, ROS, Reactive Oxygen Species, SLNs, Solid Lipid Nanoparticles, SOD, Superoxide Dismutase, SWCNTs, Single-Walled Carbon Nanotubes, TDFNs, Theranostic DNA Frameworks, TNF-α, Tumor Necrosis Factor-alpha, TPP+, Triphenylphosphonium. Conclusion The present review attempts to provide an integrative analysis that extends beyond the traditional amyloid/tau framework in AD research by synthesizing preclinical evidence. It systematically examines how sophisticated nanostructures engage with multiple concurrent NANT pathological processes, a conceptual integration not fully addressed in previous studies. Unlike publications that typically address isolated pathways or general nanocarrier characteristics, this work distinctly consolidates and contrasts emerging nanostrategies. The discussion details specific, advanced designs documented within, including mitochondria‑targeted systems functionalized with TPP⁺, chiral gold nanoparticles that remodel the gut microbiota, quercetin‑sulfur nanoparticles that modulate ER stress, and CRISPR‑dCas9‑loaded nanozymes. These examples represent a new class of multifunctional nanotherapeutics informed by disease pathology. Crucially, the development of these sophisticated nanotherapies must proceed in parallel with the validation of robust NANT biomarkers. As detailed in this review, fluid-based markers of neurodegeneration (NFL), synaptic dysfunction (neurogranin), and neuroinflammation (GFAP, YKL-40) provide a multidimensional framework essential for patient stratification and for monitoring the target engagement and clinical efficacy of these emerging interventions. Despite these promising developments, substantial barriers to clinical application remain. The current lack of registered clinical trials for nanoparticle‑based AD treatments points to a critical need to resolve fundamental regulatory questions prior to human studies. Further considerations include persistent biocompatibility issues, possible immunological responses, incomplete long‑term biodistribution and toxicity data, and the practical manufacturing scale‑up of intricate nanoformulations. Subsequent research should concentrate on the logical design of multifunctional yet translationally viable nanoparticles, thorough preclinical validation using pertinent disease models, and increased interdisciplinary cooperation bridging materials science, neurology, and regulatory affairs. Although this review focuses on nonamyloid pathways, contemporary progress in nanomaterial design has also revolutionized approaches for targeting Aβ species, creating opportunities for synergistic, multitarget interventions. Notably, novel dual‑function systems that concurrently mitigate Aβ toxicity and correct neuroinflammation and mitochondrial dysfunction represent a unified strategy against the complex pathology of AD. These developments illustrate the flexibility of nanotechnology in merging amyloid-directed and alternative pathway interventions into a coherent therapeutic paradigm.