Discussion

This study successfully constructed RVG-EXOs and systematically elucidated their therapeutic effects and mechanisms in an AD model. Our results demonstrated that RVG-EXOs exhibited typical exosomal characteristics, and RVG modification significantly enhanced the neuronal targeting capability and brain delivery efficiency. In 3×Tg AD model mice, treatment with RVG-EXOs effectively improved multiple cognitive behavioral deficits, including spatial learning, working memory, and object recognition. At the pathological level, RVG-EXOs demonstrated strong abilities both in vitro and in vivo to promote Aβ clearance, inhibit Tau hyperphosphorylation, and protect synaptic structure and neuronal survival. Mechanistically, our study revealed that RVG-EXOs activated the autophagy pathway via targeted inhibition of RPTOR. On one hand, the enhancement of neuronal autophagy promotes the clearance of pathological proteins such as Aβ and P-Tau, thereby reducing neuronal damage. On the other hand, autophagy-mediated clearance of APP reduces the binding of APP to CD74 receptors on microglia, which in turn inhibits the microglial transition to the DAM phenotype and alleviates neuroinflammation. Through these dual pathways, RVG-EXOs collectively exert neuroprotective effects. Key rescue experiments confirmed that RPTOR overexpression significantly inhibited the therapeutic effects of RVG-EXOs, establishing the core functional axis of “RVG-EXO–RPTOR–autophagy” (Fig. 9A–D).

Fig. 9: Schematic illustration of RVG-EXO fabrication and its therapeutic mechanism in the AD brain. (A) Schematic diagram of the RVG-EXO production. (B) RVG engineering enhances the brain-targeting efficiency of EXOs derived from young donors. (C)RVG-EXOs are capable of targeting neurons and enhancing neuronal autophagy. On one hand, this promotes the clearance of pathological proteins such as Aβ and P-Tau, thereby reducing neuronal damage. On the other hand, autophagy-mediated clearance of APP reduces the binding of APP to CD74 receptors on microglia, which in turn inhibits the microglial transition to the DAM phenotype and alleviates neuroinflammation. Through these dual pathways, RVG-EXOs collectively exert neuroprotective effects. (D) After entering neurons, RVG-EXOs improve autophagy and promote the clearance of pathological proteins by inhibiting Raptor expression.

RVG-29 has been widely used to enhance the brain-targeted delivery efficiency of drug carriers due to its ability to specifically recognize acetylcholine receptors on the neuronal surface [48]. Studies have shown that modifying mesenchymal stem cells with RVG can significantly enhance their migration and enrichment capacity toward brain tissues [49]. Additionally, RVG-modified nanoplatelets can also effectively enhance the delivery of therapeutic agents to the brain in glioma models, further highlighting the value of the RVG peptide in overcoming the BBB and achieving central nervous system (CNS) targeting [50]. Our study successfully constructed RVG-EXOs by conjugating RVG-29 to EXOs isolated from young human plasma using a lipid anchoring method (Fig. 1A–F) and systematically evaluated its delivery efficiency. The results demonstrated that, compared to unmodified natural EXOs, RVG-EXOs exhibited a significantly enhanced cellular uptake capability when co-cultured with SH-SY5Y cells in vitro (Fig. 1G–K). In vivo experiments further confirmed that after tail vein injection, the accumulation signals of RVG-EXOs in mouse brain tissues were significantly higher than those of plain EXOs, demonstrating its enhanced BBB penetration and targeted enrichment capability in the brain (Fig. 1L and M). These results align with the previously observed trend of RVG modification in other delivery systems, validating the applicability and effectiveness of this targeting strategy in EXO engineering.

In recent years, circulating factors (such as plasma and EXOs) from young individuals have attracted significant attention for their roles in reversing age-related functional decline and ameliorating neurodegenerative diseases. Studies have shown that young plasma can reverse cognitive impairment in aged 3×Tg-AD mice [51]. Furthermore, studies have found that EXOs derived from young plasma can reverse age-related cognitive decline in aged mice by improving mitochondrial energy metabolism [9]. EXOs derived from young serum have also been proven to ameliorate age-related cognitive decline in aged mice [10]. In our study, we validated the direct therapeutic effect of RVG-EXOs in the 3×Tg AD model. Behavioral test results demonstrated that treatment of RVG-EXOs significantly improved spatial learning and memory capabilities of AD mice in the MWM, increased the spontaneous alternation rate in the Y-maze, and restored their exploration preference in the NOR test (Fig. 2A–F). These results align with the overall trend observed for “rejuvenating” factors in improving cognitive function and directly demonstrate, within an AD transgenic model, the effectiveness of engineered young plasma EXOs in alleviating multiple cognitive deficits.

The core pathological hallmarks of AD include the deposition of Aβ plaques in the brain and the formation of neurofibrillary tangles composed of P-Tau, which collectively contribute to synaptic dysfunction and neuronal death [2]. Research has shown that young plasma can reduce neuroinflammation, decrease Aβ deposition, lower the level of Tau protein hyperphosphorylation, and protect neurons [51]. Building upon this foundation, our study further constructed and validated the role of RVG-EXOs in intervening in key pathological processes of AD. In the cell model, treatment of RVG-EXOs significantly enhanced the ability of SH-SY5Y cells to clear Aβ, reduced the levels of P-Tau induced by OA, and effectively alleviated neurite damage and cell death caused by Aβ exposure (Fig. 3A–D and S2A, B). In 3×Tg AD model mice, treatment of RVG-EXOs likewise demonstrated clear therapeutic effects: it significantly reduced Aβ plaque burden in the hippocampus, decreased P-Tau protein expression, restored SYAP expression and dendritic spine density, and improved neuronal survival rates (Fig. 3E–J). These results indicate that RVG-EXOs not only retain the neuroprotective activity of young plasma-derived EXOs but also, through enhanced targeted delivery efficiency, effectively modulate the core pathological processes of AD in both in vitro and in vivo systems.

RPTOR serves as the core scaffold protein of the mTORC1 complex. By inhibiting the activity of the ULK1/2 autophagy initiation complex, it negatively regulates the cellular autophagy process [52]. Autophagy is recognized as a key protein degradation pathway responsible for clearing misfolded proteins and plays a key role in various neurodegenerative diseases, including AD [7,14,15]. Studies indicate that autophagic dysfunction is an early and core pathological feature of AD, forming a vicious cycle with the accumulation of Aβ and P-Tau [53]. Therefore, restoring neuronal autophagic function has been regarded as a crucial strategy for intervening in AD, which can significantly mitigate pathological protein deposition, reduce neuronal loss, and improve cognitive behavior [54]. The most critical mechanistic finding of this study lies in elucidating the complete pathway through which young plasma-derived EXOs and RVG-EXOs exert neuroprotective effects by targeting the RPTOR-mTORC1 axis to activate autophagy. We first confirmed that both AD cell and animal models exhibited impaired autophagic flux, as indicated by abnormal LC3B-II alterations and P62 accumulation. Treatment with young plasma-derived EXOs or RVG-EXOs effectively reversed this phenomenon (Fig. 4C–G, 4J–N, and S3A). Further mechanistic investigation traces the origin to RPTOR, a core component of mTORC1. Studies suggest that RPTOR is a susceptibility gene for AD and may represent a potential therapeutic target [55]. Based on sequencing and bioinformatics analyses, a total of six miRNAs, e.g., hsa-miR-23a-3p and hsa-miR-23b-3p, were identified and highly expressed in plasma EXOs and RVG-EXOs from young adults and could target RPTOR gene (Fig. 5A and B). Subsequent in vitro and in vivo experiments confirmed that RVG-EXOs could downregulate RPTOR expression. Conversely, artificial overexpression of RPTOR significantly inhibited the beneficial effects of RVG-EXOs across multiple levels, including improving autophagy, clearing pathological proteins, protecting neurons, and ameliorating behavioral outcomes (Fig. 5C–O, 6A–R, S4K, and S5A–D). This series of evidence collectively and conclusively demonstrates the following causal functioning flow: plasma EXOs and RVG-EXOs from young individuals → inhibition of RPTOR → relief of mTORC1-mediated suppression of autophagy → restoration of autophagic flux → clearance of pathological proteins and neuroprotection. The elucidation of this mechanism elevates the understanding of young plasma EXOs in treating AD from a phenotypic description to the level of a clearly defined molecular pathway.

Our single-cell transcriptomic profiling further extended these mechanistic insights by revealing the cellular and microenvironmental remodeling induced by RVG-EXOs in the 3×Tg mouse brain. Consistent with the neuroprotective effects observed histologically, RVG-EXO treatment significantly increased the proportion of neurons (Fig. 7A–C). More importantly, the downregulation of Rptor expression in neurons and the significant enrichment of autophagy pathways at the single-cell level provided direct in vivo evidence supporting our central mechanistic axis (Fig. 7D–F). This finding corroborates our biochemical and functional data, reinforcing the conclusion that RVG-EXOs promote autophagic flux in neurons through the RPTOR-mTORC1 axis. In addition, studies have found that excitation/inhibition (E/I) imbalance is a hallmark of Alzheimer's disease pathophysiology, and that restoring this balance is associated with improved cognitive function and protection against excitotoxicity [35]. Our single-cell sequencing results show that RVG-EXOs not only increase the absolute numbers of both excitatory and inhibitory neurons, but also elevate the proportion of inhibitory neurons. This alteration contributes to restoring E/I balance and ameliorating cognitive deficits in AD (Fig. 7I and J).

Beyond neuronal changes, the single-cell data unveiled a previously unrecognized effect of RVG-EXOs on microglial dynamics. We observed a significant reduction in the deleterious DAM and proliferating microglia subsets, alongside a concurrent increase in the homeostatic microglial population (Fig. 8G–J). Given that DAM are known to perpetuate neuroinflammation and contribute to synaptic loss in AD [47], this microglial repolarization toward a homeostatic phenotype represents a critical secondary mechanism by which RVG-EXOs exert neuroprotection. This is further supported by the cell-cell communication analysis, which revealed that RVG-EXOs strengthened the neurotrophic Ptn-Sdc3 signaling between neurons while attenuating the App-CD74 signaling axes between neurons and microglia that promotes DAM generation. (Fig. 7M). Collectively, these findings suggest a dual-action model for RVG-EXOs: they directly activate neuronal autophagy to clear pathological proteins and, in parallel, remodel the brain's intercellular communication network to foster an anti-inflammatory, neurosupportive microenvironment. This integrated perspective, derived from unbiased single-cell analysis, not only validates our proposed RPTOR-autophagy mechanism but also uncovers a broader therapeutic landscape wherein RVG-EXOs harmonize both neuronal and glial functions to combat AD pathology.

Several studies have explored RVG-modified EXOs for AD therapy. Notably, Cui et al. demonstrated that RVG-conjugated EXOs derived from mesenchymal stem cells (MSC-RVG-Exos) improved targeting to the brain, reduced Aβ plaque deposition, and modulated inflammatory responses in APP/PS1 mice [49]. This pioneering work established the feasibility of RVG-mediated EXO delivery for AD. However, the mechanistic understanding was primarily centered on anti-inflammatory effects, and the specific molecular cargo responsible for therapeutic benefits was not fully elucidated. More recent studies have employed genetic engineering to generate EXOs displaying RVG peptide and simultaneously enriching therapeutic proteins such as neprilysin for Aβ degradation [56]. While these approaches demonstrate enhanced efficacy, they rely on complex genetic manipulation of parental cells and focus on exogenous cargo loading rather than harnessing the intrinsic therapeutic potential of the EXOs themselves. Compared with these existing approaches, our study offers several distinct advantages and significant advances. Firstly, we employed young human plasma-derived EXOs. This source is particularly advantageous because young plasma EXOs are naturally enriched with neuroprotective and “rejuvenating” factors, including the specific miRNAs (miR-23a-3p, miR-23b-3p, etc.) that were identified as key effectors targeting RPTOR in our study. This leverages the intrinsic therapeutic cargo of young plasma rather than relying on exogenous loading. Secondly, young plasma is more readily accessible and cost-effective compared to MSC culture supernatants or genetically modified cell systems, which require complex and expensive maintenance, transfection, and selection procedures. This practical advantage significantly enhances the translational potential of our strategy for clinical applications. Furthermore, we have elucidated a complete mechanistic pathway. While previous studies on RVG-engineered EXOs primarily focused on phenotypic outcomes, such as Aβ reduction and modulation of inflammatory responses, our study not only examined these phenotypic effects but also clarified the underlying mechanistic pathway.

The core innovations of this study are mainly reflected in the following three aspects. First, it is the first to demonstrate the direct application value of plasma-derived EXOs from young individuals in the treatment of AD. Although young plasma EXOs have been previously investigated in the context of anti-aging research, their role in ameliorating the pathological progression of AD has not been systematically elucidated. Through experiments in vitro and in vivo, this study provides the first validation of their therapeutic potential in AD models. Second, it elucidates the molecular mechanisms by which young plasma-derived EXOs exert their therapeutic effects through the RPTOR-autophagy axis. We identified that miRNAs enriched in plasma EXOs from young individuals, particularly miR-23a-3p and miR-23b-3p, directly target RPTOR. Functional experiments further confirmed that RVG-EXOs promote the clearance of pathological proteins by inhibiting RPTOR expression and activating the autophagy pathway. These findings elevate the therapeutic role of young plasma EXOs from phenomenological descriptions to the molecular pathway level. Third, RVG-engineered modification significantly enhanced the brain-targeting delivery efficiency and therapeutic efficacy of plasma-derived EXOs from young individuals. To address the limitation that natural EXOs are difficult to cross the BBB, this study successfully constructed RVG-EXOs and demonstrated their ability to significantly improve neuron-targeting ability and brain enrichment efficiency, thereby enhancing pathological improvement and cognitive protection in AD models. This provides an effective strategy for the targeted delivery of EXOs in CNS diseases.

Nevertheless, this study has several limitations. First, although we have established the critical role of the RPTOR-autophagy axis in mediating the therapeutic effects of RVG-EXOs, the potential involvement of other molecular pathways cannot be excluded and warrants further investigation. Second, the activation of autophagy is a “double-edged sword” (i.e., its long-term effects and safety in more advanced AD models need to be evaluated over longer observation periods). Finally, translating young plasma EXOs into clinical therapy requires addressing challenges such as large-scale standardized production, quality control, and ethical regulations. Future research should focus on: (1) optimizing engineering strategies to further enhance targeting efficiency; (2) exploring other youth-related active molecules; (3) combining RVG-EXOs with existing drugs to assess synergistic therapeutic effects on the treatment of AD; (4) and advancing relevant preclinical safety evaluations.