Mechanism of action and clinical application of autophagy in multiple sclerosis (Review)

Introduction

Multiple sclerosis (MS)

MS is a complex neurological disorder. Extensive research on its global epidemiology and societal impact has highlighted its significant effects on both individuals and communities. Data from 2015 ranked MS as the 10th most common neurological condition, with an estimated 2.01 million cases at that time (1). According to updated data, an estimated 2.9 million individuals are living with MS globally, with the prevalence having increased in most countries reporting updated data (2). This data represents a significant increase compared to earlier estimates, and the rising prevalence of the disease underscores the importance of ongoing research efforts. However, MS incidence varies considerably by region. Countries with higher socioeconomic status often report a greater frequency of MS, possibly influenced by environmental factors, genetic predispositions and lifestyle choices. For instance, MS is more prevalent in regions like North America and Northern Europe, while it is less common in parts of Asia and Africa. MS typically manifests in young adults, particularly affecting women aged 20 to 40 years, with women being affected at roughly three times the rate of men (3). This gender disparity may be attributed to hormonal, immune and genetic factors.

The management and treatment of MS require significant medical resources, placing a heavy burden on both individuals and society (4). Because MS requires long-term multidisciplinary management, including medication, rehabilitation and psychological support, patients often face substantially higher healthcare costs than individuals without the condition (5). Furthermore, MS often leads to reduced work capacity, contributing to higher unemployment and absenteeism rates. Studies show that individuals with MS experience diminished work abilities throughout their careers, affecting both their financial independence and quality of life (6). Mental health challenges, such as anxiety and depression, are also common among those with MS, further deteriorating their overall health and amplifying the demand for healthcare services (7).

Inadequate social support networks can exacerbate these issues (8). Additionally, societal attitudes and understanding of MS vary widely across cultures and communities (9). Numerous patients face social stigma and misunderstanding, which negatively affects their psychological well-being and social interactions. Therefore, raising public awareness of MS is essential to improving social integration and support for affected individuals (10). Global epidemiological data highlight the rising prevalence of MS as a significant public health concern, with profound implications for both individuals and society (11). Given its rising incidence and impact, advancing research and public health initiatives is crucial to improving the lives of those affected and alleviating the strain on society (12).

Definition and classification of autophagy

Adapter proteins are crucial in selective autophagy, identifying distinct autophagic substrates and attaching them to autophagosomes. Through this method, cells can effectively mark and separate substances requiring degradation (13). This mechanism plays a crucial role in maintaining autophagy selectivity, guaranteeing the breakdown of only certain substances without impacting other cell parts.

Multistep process

Selective autophagy involves initially identifying and isolating the autophagic substrate, then creating the autophagosome, merging it with the lysosome and finally breaking it down. The procedure entails controlling various intracellular signaling routes to guarantee autophagy's efficiency and specificity (14).

Selective autophagy varieties

Depending on the specific substrate and its role, there are multiple subtypes of selective autophagy, including mitochondrial autophagy (mitophagy), which specifically focuses on impaired or dysfunctional mitochondria. Through mitochondrial autophagy, cells are able to eliminate impaired mitochondria, thereby averting oxidative stress and cellular demise. Plasmalemmal autophagy is a type of autophagy that targets certain areas of the cell membrane, eliminating detrimental elements or pathogens to preserve the cell's integrity and functionality (15). Endoplasmic reticulum (ER) autophagy concentrates on eliminating irregular or improperly folded proteins in the ER to preserve its functionality and health (16).

The biological roles of autophagy

Autophagy plays a crucial role in MS pathogenesis by regulating the balance of cellular stress and inflammation (17). In MS, autophagy is particularly involved in controlling oxidative stress and mitigating damage to oligodendrocytes, which are critical for myelin regeneration (18). Three recent studies consistently emphasize that autophagy is not intrinsically beneficial or harmful, but a context-dependent regulatory process whose impact in diseases such as MS depends on cell type, timing and magnitude of activation (19-21). According to the conceptual framework proposed by the three recently studies, autophagy is a highly dynamic and tightly regulated cellular process rather than a uniformly protective pathway. Their critical evaluation of autophagy in MS explicitly challenges the notion of autophagy as an inherently beneficial process, emphasizing that its effects vary across immune and central nervous system (CNS) cell types, disease stages and inflammatory environments. Autophagy is also essential for maintaining cellular quality (22). It selectively eliminates dysfunctional organelles, such as damaged mitochondria, and removes harmful proteins, preventing the accumulation of toxic substances within the cell (23). This process plays a key role in protecting cells from stress and damage. As an adaptive mechanism, autophagy helps cells adjust to environmental changes, particularly during starvation or stress (24). By breaking down internal cellular components, autophagy provides essential nutrients, ensuring ongoing cell survival and functionality. Additionally, autophagy is integral to immune responses (25). It is involved in both anti-infective and immune reactions by eliminating intracellular pathogens and promoting the activation and growth of immune cells through antigen processing (26). As a result, modulating autophagy may offer a novel therapeutic strategy for treating MS (27). In summary, autophagy serves multiple biological roles, from energy distribution and cell quality control to adaptive responses and immune function. Understanding the mechanisms underlying autophagy and its impact on diseases can inspire innovative approaches for studying and treating conditions like MS. Regulating autophagy holds promise for improving disease outcomes, enhancing neuroprotection and slowing disease progression. Seminal work by Mizushima et al (28-30) has established autophagy as a fundamental cellular process essential for maintaining intracellular and tissue homeostasis. Through a series of influential studies and comprehensive reviews, autophagy has been conceptualized not merely as a degradation pathway but as a dynamic system governing cellular quality control, metabolic adaptation and tissue renovation. Extending from the cellular to the tissue level, autophagy has been described as a 'renovation' mechanism that supports long-term tissue maintenance and functional stability, particularly in organs with limited regenerative capacity such as the CNS. This conceptual framework provides a critical foundation for understanding how autophagy may exert both protective and detrimental effects in CNS immune-mediated disorders, including MS, depending on cell type, disease stage and microenvironmental cues.

Control systems governing autophagy

Pathway of mammalian target of rapamycin (mTOR)

The mTOR is a central nutrient- and energy-sensing kinase that plays a major, but context-dependent, role in autophagy regulation (31). Rather than acting as a simple inhibitory switch, mTOR integrates multiple environmental cues to dynamically modulate autophagy. In particular, mTOR complex 1 (mTORC1) suppresses autophagy under nutrient-rich conditions by phosphorylating components of the unc-51-like autophagy-activated kinase 1 (ULK1) initiation complex, thereby preventing autophagosome formation (32,33). Growth factor signaling, such as insulin-like growth factor (IGF) binding to IGF1 receptor (IGF1R), activates the PI3K-Akt pathway, leading to inhibition of the tuberous sclerosis 1 (TSC1)/TSC2 complex and subsequent activation of mTORC1, which suppresses autophagy (34-36). Conversely, amino acid deprivation and energy stress suppress mTORC1 activity, enabling autophagy initiation. mTORC1 activity is also regulated spatially through its association with lysosomal membranes, highlighting the importance of subcellular localization in autophagy control (37,38).

Pathway of adenosine monophosphate-activated protein kinase (AMPK)

AMPK functions as a key metabolic sensor that regulates autophagy in response to cellular energy stress (39). Upon glucose deprivation, AMPK activation can promote autophagy by directly phosphorylating ULK1. However, this effect is modulated by mTORC1 activity, as mTORC1-dependent phosphorylation of ULK1 may counteract AMPK-mediated activation (40). In addition, AMPK indirectly influences autophagy through the TSC1/2-mTOR axis, highlighting its role as an upstream regulator within a broader signaling network rather than a unidirectional activator (41).

Pathway of protein kinase A (PKA)

PKA senses glucose availability and has been reported to negatively regulate autophagy under nutrient-rich conditions, primarily through phosphorylation of autophagy-related proteins such as autophagy-related protein 1 (Atg1) and Atg13 (42). Furthermore, PKA may indirectly suppress autophagy by modulating mTORC1 and AMPK signaling, suggesting an indirect and context-dependent regulatory role (43).

Pathway of Jnk-1

The JNK-1 pathway has been implicated in autophagy induction, particularly under stress conditions such as nutrient deprivation and ER stress (44). Activation of JNK-1 can lead to phosphorylation of Bcl-2, thereby disrupting its interaction with Beclin 1 (BECN1) and facilitating autophagy initiation (45). However, JNK-mediated autophagy is highly dependent on stress intensity, duration and cell type, underscoring its role as a stress-responsive modulator rather than a universal autophagy activator.

Pathway of P53

Nuclear p53 can transcriptionally activate autophagy-related genes through the TSC1-mTOR pathway, whereas cytoplasmic p53 has been shown to suppress basal autophagy. Consequently, pharmacological inhibition of p53 may induce autophagy in certain cellular contexts but suppress it in others, highlighting the complexity of p53-mediated autophagy regulation (40).

Free fatty acids

Lipid metabolism is closely connected to autophagy regulation (46). Free fatty acids can activate EIF2S1/eIF2α and MAPK8 through the EIF2AK2/PKR-dependent pathway or inhibit mTORC1, thereby promoting autophagy (47). Lipid droplets, which are dietary fats stored in cells, trigger autophagy, forming autophagosomes that envelop and transport them to lysosomes for degradation. Degradation products, such as free fatty acids, regulate autophagy and prevent lipotoxicity (25).

Pathway of inositol 1,4,5-trisphosphate (IP3)

Inositol monophosphatase (IMPase) reduces the levels of unbound inositol and IP3, which in turn promotes calcium ion release from the ER. This process occurs via the binding of IP3 to its receptors on the reticulum's surface, activating a cascade of calcium-dependent proteases and inhibiting autophagy by cleaving and activating stimulatory alpha subunit of heterotrimeric G proteins (Gsα), resulting in significant cyclic (c)AMP production (48). Additional mechanisms may contribute to this pathway, such as the reduction in IP3 levels, which limits calcium entry from the ER into mitochondria, potentially impairing the mitochondrial respiratory chain and subsequently enhancing autophagy via the AMPK pathway. Furthermore, the IP3 receptor interacts with BECN1, suppressing autophagy (16). Neuropsychotropic drugs such as sodium valproate and carbamazepine are known to stimulate autophagy through this pathway (49).

Molecular impacts of autophagy

The primary mechanism driving autophagy involves the activity of ATG proteins. This process is further regulated by AMPK and mTOR, which modulate autophagy through the phosphorylation of key proteins like ULK1, ULK2 and ATG13. Phosphorylated ULK1 activates BECN1, a homolog of yeast Vps30/Atg6, initiating the formation of autophagosomes through the ULK1-BECN1 complex. As part of the class III phosphatidylinositol 3-kinase (PtdIns3K) complex, BECN1 generates PtdIns3P on the membranes of phagocytic vesicles, serving as a binding site for additional autophagy-related proteins. A pair of ubiquitin-like conjugation systems, comprising ATG12-ATG5-ATG16L1 and the Atg8 family proteins such as microtubule-associated protein 1 light chain 3 (MAP1LC3, commonly referred to as LC3) and γ-aminobutyric acid receptor-associated protein, are essential for the growth and expansion of phagocytic vesicles and the attachment of specific autophagic types (16,48,50) (Fig. 1).

Figure 1 Major signaling pathways regulating autophagy. Autophagy is regulated by multiple signaling pathways, including mTOR, AMPK, PKA, JNK-1, p53, free fatty acid-related pathways and IP3 signaling. The mTOR pathway functions as a central negative regulator of autophagy in response to nutrient availability, whereas AMPK activates autophagy under energy stress. PKA generally suppresses autophagy under high-glucose conditions. JNK-1 promotes autophagy by modulating the Beclin 1-Bcl-2 interaction during nutrient deprivation. p53 can either induce or inhibit autophagy depending on its cellular localization and regulatory context. Free fatty acids and lipid droplets modulate autophagy through metabolic stress-related signaling, while IP3 signaling inhibits autophagy via calcium-dependent mechanisms and BECN1binding. IP3, inositol 1,4,5-trisphosphate; mTOR, mammalian target of rapamycin; AMPK, adenosine monophosphate-activated protein kinase; PKA, protein kinase A; ULK1, unc-51-like autophagy-activated kinase 1; Atg12, autophagy-related protein 12; BECN1, Beclin 1.

Collectively, these pathways illustrate that autophagy is governed by an intricate and highly interconnected regulatory network rather than by linear signaling cascades. The functional outcome of autophagy signaling depends on nutrient availability, cellular context, stress intensity and disease state.

This review was based on a comprehensive literature search of peer-reviewed articles related to autophagy and MS. The primary databases searched included PubMed (https://pubmed.ncbi.nlm.nih.gov/), Web of Science (https://www.webofscience.com/) and Scopus (https://www.scopus.com/). Relevant studies were identified using combinations of key words such as 'multiple sclerosis', 'autophagy', 'mitophagy', 'oxidative stress', 'immune regulation' and 'neurodegeneration'. Articles were selected based on their relevance to the role of autophagy in MS pathogenesis, progression and potential therapeutic implications. Priority was given to original research articles, high-quality reviews and recent studies that provided mechanistic insights or experimental evidence. Studies focusing on non-MS neurodegenerative diseases were included when they offered conceptual or mechanistic relevance to MS. Non-English publications and studies lacking sufficient experimental or clinical relevance were excluded.

Connection between autophagy and neurological conditions

Neuronal structures are highly susceptible to disruptions due to the interaction between endosomes/autophagosomes and lysosomes, particularly in the aging brain. Genetic aging and alterations in autophagic processes are linked to the onset of various neurodegenerative diseases, such as Alzheimer's disease (AD), familial Parkinson's disease (PD) and amyotrophic lateral sclerosis (ALS), which arise from impaired lysosomal clearance. Autophagic dysfunction is associated with the accumulation of cytoplasmic aggregates, protein buildup in neurons and mitochondrial dysfunction. These diseases share similarities with MS, characterized by suppressed autophagy. Dysfunction in autophagy leads to cellular dysfunction, further promoting the progression of these diseases. Unlike other cell types, neurons cannot eliminate damaged organelles or misfolded proteins through cell division. The presence of synapses, dendrites and axons, along with their significant distance from the cytosol, further complicates autophagy and intercellular communication (28,37,51,52).

PD

PD, a neurodegenerative disorder, results from the selective degeneration of dopaminergic neurons in the substantia nigra. Neuropathologically, PD is characterized by the deposition of Lewy bodies, which consist of cellular accumulations of ubiquitin and α-synuclein (SNCA). Autophagy plays a critical role in preventing PD onset by degrading SNCA through both macroautophagy and chaperone-mediated autophagy (CMA). Inhibition of these processes leads to the accumulation of both mutant and normal forms of SNCA (53). BECN1 promotes autophagy, thereby enhancing SNCA degradation. The role of BECN1 in promoting autophagy is equally important in MS. The accumulation of SNCA disrupts macroautophagy, as high SNCA levels lead to irregular distribution of ATG9, inhibiting omegasome formation, a precursor to autophagic vesicles, and suppressing CMA, which contributes to progressive neuronal damage (40). The autophagy process in neurons is more complex than in other cell types, further complicating its role in neurodegenerative diseases.

Of note, autophagy may paradoxically contribute to neuronal degeneration under certain conditions. Overactivation of autophagy can exacerbate dopaminergic neuron death, with heightened autophagic activity in PD potentially triggered by oxidative stress that suppresses genes involved in mTOR activation. Supporting this view, autophagy stimulants have been shown to worsen neuronal death in the context of oxidative stress, suggesting that autophagy may act as a double-edged sword in PD (54). Additionally, parkin RBR E3 ubiquitin ligase (PRKN) plays a key role in mitochondrial autophagy by removing damaged mitochondria. Defective PRKN in PD leads to mitochondrial dysfunction and neurodegeneration. Of note, mitochondrial autophagy is dysfunctional in MS, not merely suppressed (55).

A comparative analysis between 20 patients with PD and 20 individuals with idiopathic tremor revealed significantly lower plasma concentrations of the lysosomal hydrolase CTSD (histone D) and BAG chaperone 2, a regulator of the PTEN-induced kinase 1 (PINK1)-PRKN pathway, in those with PD. These findings suggest a reduction in autophagic activity in PD. Similarly, peripheral blood mononuclear cells from patients with PD showed diminished levels of heat shock protein family A member 8/heat shock cognate protein of 71 kDa, a critical component of CMA (56). As a result, autophagy is notably impaired in PD, though it can be either suppressed or upregulated under different circumstances.

Conversely, inhibiting mTOR may offer therapeutic potential in AD by promoting autophagy (57). In AD, autophagy, a vital process for microglia to clear amyloid-β (Aβ), becomes dysfunctional. Experimental studies have shown that microglia in AD mice exhibit elevated levels of microRNAs (miRNAs), which suppress the expression of autophagy-related proteins. Of note, Aβ accumulation around microglia precedes autophagy impairment in AD (58). Extended exposure to Aβ also triggers miRNA expression in peripheral neurons (59).

Impaired autophagic processes, particularly the dysfunctional formation and transport of autophagosomes, lead to the progressive accumulation of autophagic vesicles around neurons in AD. This accumulation, exacerbated by Aβ, promotes neurodegeneration (60). In addition, a reduction in BECN1 levels, both in AD and aging, results in diminished autophagic activity. This decrease in BECN1 triggers neurodegeneration through enhanced Aβ accumulation and disruptions in autophagy and APP metabolism (17). These findings suggest that excessive autophagy may contribute to the neuropathology of AD, although autophagy malfunctions could also be secondary to disease progression.

Furthermore, the autophagy and BECN1regulatory factor 1 (AMBRA1) protein, which regulates autophagy in the CNS, interacts with BECN1 to strengthen its association with the lipid kinase phosphatidylinositol 3-kinase catalytic subunit type 3 (PIK3C3/VPS34). Together with PIK3C3/VPS34 and BECN1, AMBRA1 is considered a key component of the autophagy core complex, which is critical for regulating this process in AD. The autophagy core complex itself plays an essential role in modulating autophagic activity (61).

Sclerosis of the amyotrophic lateral system

ALS, a progressive neurodegenerative disorder, is characterized by the gradual degeneration of motor neurons. Of the two ALS subtypes, 90% of cases are sporadic, while the familial form accounts for the remaining 10% (62). Autophagy plays a critical role in ALS pathogenesis. Excessive macroautophagy accelerates spinal motor neuron degeneration in ALS mice with the mutant superoxide dismutase 1 (SOD1), which is implicated in ~20% of familial ALS cases (63). Notably, the motor neuron degeneration observed in these mice is closely linked to autophagic dysfunction and autophagosome formation (64). Chen et al (65) highlighted the association between impaired autophagic regulation and ALS-associated proteins, such as mutant SOD1, along with other gene products that contribute to abnormal autophagic activity.

Furthermore, optic nerve phosphatase (OPTN), a highly conserved protein, is essential for regulating various cellular processes, including vesicular transport, inflammation and autophagy (66). Genetic alterations in OPTN are linked to inflammation and glaucoma in patients with ALS, while the loss of OPTN is associated with cerebral atrophy (67).

In summary, the various stages of autophagy influence a range of neurodegenerative disorders in distinct ways. For instance, in AD and ALS, there is an increase in autophagy initiation, while the fusion and maturation stages are disrupted in both conditions. In PD, the nucleation phase is notably diminished. These stage-specific autophagic dysfunctions may be linked to the unique pathophysiology of each neurodegenerative disease.

Investigating autophagy's function in MS

Autophagy plays a dual role in the pathogenesis of MS, similar to its involvement in other diseases. Recent studies have highlighted that autophagy-dependent ferroptosis contributes to the onset of MS (68). Microglia-derived extracellular vesicles (EVs) carry bioactive cargo that modulates neuroinflammation and glia-neuronal interactions in MS, and dysregulation of microglial autophagy has been linked to impaired myelin-debris handling and poorer recovery in demyelinating models. MS-focused studies indicate that altered microglial EV signaling and autophagic dysfunction may contribute to neurodegeneration and disease progression in specific pathological contexts (69).

The link between autophagy and MS was first established in animal models, where altered autophagy markers were observed in hippocampal neurons. One study identified significant increases in the expression of Beclin-1, LC3 and p62, alongside disturbances in the Akt/mTOR signaling pathway, upon examining key autophagy proteins. When compared to control groups, mice with MS exhibited notably reduced Beclin-1/LC3II and p62 levels. Molecular studies further revealed abnormalities in Akt/mTOR pathways, indicating that autophagy activation occurs in hippocampal neurons within this experimental model (70). Additionally, autophagy and mitochondrial autophagy-related markers were found to be upregulated in the bodily fluids of patients with MS (71).

These findings suggest a robust association between autophagy and MS pathology. Of note, ALS-associated proteins, including transactive response DNA-binding protein of 43 kDa and SOD1, have also been implicated in autophagic dysregulation. Several genetic mutations in ALS interfere with autophagy in motor neurons, reinforcing the notion that autophagy dysregulation plays a central role in ALS pathogenesis (72). Furthermore, ATG16L2, an autophagy-related gene, has been proposed as a potential serum biomarker for MS, identified using microbead-based proteomics technology (73).

In addition, bioinformatics combined with machine learning approaches has been used to explore the role of autophagy-related genes in MS. The 9 autophagy genes most strongly associated with MS were identified, with bioinformatics and experimental data indicating their significant involvement in autophagy regulation and MS development (74). The 9 autophagy genes with the strongest correlations were as follows: Becn1, fibroblast growth factor receptor 2 (Fgfr2), integrin alpha 3 (Itga3), Napsin A aspartic peptidase (Napsa), neurotrophic tyrosine kinase, receptor, type 2 (Ntrk2), ataxia telangiectasia mutated (Atm), solute carrier family 36 member 4 (Slc36a4), member RAS oncogene family (Rab10) and Rous sarcoma oncogene (Src). BECN1 BECN1is one of the core genes involved in autophagy; in MS, reduced expression of BECN1is associated with impaired myelin regeneration. Changes in Fgfr2 expression levels in patients with MS may be associated with neuroinflammation and myelin damage. In MS, Itga3 expression is upregulated, potentially affecting immune cell migration and nerve damage. Napsa may influence immune responses within the nervous system in MS. Reduced expression of Ntrk2 correlates with neurodegeneration and impaired myelin repair. Atm gene defects are associated with enhanced neuroinflammation and immune responses. Slc36a4 may influence neuroinflammation by regulating immune responses. Downregulation of Rab10 may affect autophagosome maturation and the process of neural repair. Upregulation of Src expression may exacerbate immune responses and neuroinflammation in MS. Gene expression profiles of autophagy-related genes in MS also reflected similar patterns, further emphasizing the connection between autophagy and MS (75). These findings collectively highlight the complex relationship between autophagy and MS. The varying stages of autophagy and the progression of MS warrant further investigation to understand how these processes are interlinked and how they can be leveraged for potential therapeutic strategies.

The process of autophagy fosters the progression of MS

Numerous studies have emphasized the protective role of autophagy in preventing the onset and progression of MS (76). Autophagy helps mitigate the development of MS by neutralizing reactive oxygen species (ROS) (77). Compared to healthy individuals, patients with MS show elevated levels of oxidized phospholipids, DNA and oxidative stress markers (78). A reduction in autophagic activity correlates with increased oxidative stress and inflammation due to impaired ROS clearance, leading to higher IL1B/IL-1β production (79). Research has extensively explored the interplay between hypoxia-inducible factor 1, ROS and autophagy in MS (80). The mechanisms through which ER stress and autophagy contribute to MS pathogenesis are well documented.

Similarly, fenofibrate, a peroxisome proliferator-activated receptor-α (PPAR-α) agonist, has been shown to reduce inflammation in MS by modulating autophagy and inhibiting helper T cell 17 (Th17) cell differentiation, along with decreasing pro-inflammatory signals (81). Studies also indicate that miRNA-223 (miR-223) impedes autophagy and exacerbates inflammation in MS by targeting ATG16L1 (82). Furthermore, blocking autophagy has been found to trigger NOD-, LRR-, and pyrin domain-containing protein 3 (NLRP3) inflammasome activation in microglia via phosphodiesterase 10A-cyclic AMP signaling, which leads to increased IL1B/IL-1β levels and enhanced macrophage migration inhibitory factor (MIF) production. Inhibition of NLRP3 inflammasomes with the MCC950 inhibitor reduced MIF expression, alleviated neuroinflammation and reversed neuronal damage in the substantia nigra (83).

In addition to ROS, reactive nitrogen species (RNS), such as nitric oxide and peroxynitrite, are involved in neurodegeneration in MS (84). These molecules can nitrate or nitrosylate the dynamin-related protein 1 (Drp1)/Parkin/PINK1 pathway, promoting excessive mitochondrial autophagy and amplifying neuronal damage. Targeting the RNS-driven excessive autophagy/mitochondrial autophagy pathway holds potential for developing novel anti-MS therapies (85).

Notably, the mTOR pathway, which regulates autophagy, also influences microglial-driven inflammation in MS (86). Targeting specific inhibitors within the PI3K-mTOR signaling axis could offer a promising therapeutic strategy for MS treatment.

Extensive research also supports the inhibitory role of autophagy in MS, potentially through prolonged activation of mTORC1, which may depend on the progression of the lesions. The present findings deepen the understanding of autophagy's role in various stages of MS pathology, suggesting that the mTORC1 pathway could serve as a key regulator, influencing CNS homeostasis and neuroinflammation in MS (87). In this context, rapamycin, an mTOR inhibitor, may stimulate autophagy and promote myelin regeneration by reducing inflammation. Additionally, curcumin has been shown to affect MS metabolism by interacting with AMPK, PPARG coactivator 1 α/PPARγ and the PI3K/Akt/mTOR signaling pathways (88). Several reviews have discussed the mechanisms through which the mTOR complex pathway influences MS (89).

Furthermore, the increase in autophagy during MS could represent an adaptive response to the heightened unfolded protein response (UPR) triggered by ER stress. The UPR serves a cytoprotective function by regulating ER activity, supporting oligodendrocyte function and enhancing myelin production in MS. Thus, stimulating the UPR may reduce MS neuropathology by promoting autophagy.

Echoing previous research, numerous studies emphasize the importance of protein homeostasis in disease development. Impairment of the autophagy-lysosomal pathway, coupled with the detrimental effects of C9orf72 repeat RNA and dipeptide repeat proteins, contributes to disease progression (90). Additionally, optineurin has been shown to initiate autophagy by clearing protein aggregates via a ubiquitin-independent mechanism, thus facilitating the progression of MS (91).

During the active phase of MS, significant increases in autophagy and mitochondrial autophagy markers have been observed in the biological fluids of patients, indicating the activation of these pathways. In line with this, both in vitro and in vivo MS models (induced by pro-inflammatory cytokines, lysyl ovalbumin and copper ketones) exhibit mitochondrial dysfunction, triggering lactate metabolism, enhancing autophagic flux and increasing mitochondrial autophagy. Various autophagy inhibitors, differing in structure and mechanism, have been found to promote myelin synthesis and normalize axonal myelination, while the antipsychotics haloperidol and clozapine significantly reduced dyskinesia induced by copper ketones (92).

Previous studies have also highlighted the role of autophagy in macrophages for MS repair (93). Lipophagy in macrophages has been shown to accelerate MS recovery, with microglia containing lipid droplets playing a pivotal role in MS pathology. Notably, miR-223-mediated suppression of cathepsin B in microglia enhances lipophagy following lysolecithin-induced demyelination in mice, indicating that activation of selective autophagy pathways contributes to myelin lipid degradation during demyelinating injury (94).

A recent study introduced a novel theory involving partial neuronal cell death and an autophagy-positive feedback loop, highlighting the role of the neuronal-astrocyte glutamate-glutamine cycle in ALS. Under optimal conditions, disrupting these cycles could slow ALS progression. Brain vascular damage, such as recurrent embolisms and strokes, may initiate neuronal cell death followed by autophagy. Furthermore, ALS impedes the fusion of autophagosomes and lysosomes, exacerbating cell death. Treatment with alginate has been shown to restore the defective fusion phase and significantly delay disease onset, suggesting its potential as a preventative treatment for ALS (95). This study aligns with our long-held belief that autophagy may indirectly influence the development of MS.

Additionally, inhibiting autophagy leads to mitochondrial dysfunction and oxidative stress, which may increase the risk of MS development. There is a notable imbalance in neuronal autophagy in proteopathies, indicating a link between autophagic dysfunction and the onset of neurodegenerative diseases. Patients with MS exhibit a reduction in CD46 receptors, which facilitate autophagy, further supporting the involvement of autophagy in MS pathology. However, the precise role of mitochondrial autophagy in MS remains unclear. Mitochondrial autophagy, a specialized form of autophagy, is crucial for maintaining mitochondrial function and controlling oxidative stress by eliminating damaged mitochondria. This process often involves the activation of PARKIN and PINK1 on the mitochondrial outer membrane, mediated by calcium-binding and coiled-coil domain-containing protein 2 (CALCOCO2)/nuclear dot protein 52 and OPTN receptors. These receptors enhance mitochondrial autophagy by promoting the interaction between phagocytic vesicle membranes and organellar proteins, aiding mitochondrial sequestration within the autophagosome. Alterations in the CALCOCO2 gene are associated with MS progression, contributing to the secretion of pro-inflammatory cytokines like TNF-α. Supporting evidence shows that CALCOCO2 is primarily expressed in B cells of peripheral blood mononuclear cells, where its stimulation through mitochondrial autophagy reduces pro-inflammatory cytokine production. This highlights CALCOCO2's protective function in B cells and its potential involvement in autoimmune diseases like MS (96).

Together, these findings suggest that autophagy plays a protective role in MS by reducing oxidative stress and inflammatory responses, thereby hindering disease onset and progression. Furthermore, the aforementioned molecular mechanisms are summarized in Fig. 2. Although numerous studies have indicated that mitochondrial autophagy in MS is closely associated with neurological damage, these investigations often focus on describing the fundamental processes of autophagy while lacking in-depth analysis of its molecular mechanisms. For instance, while PINK1 and Parkin play crucial roles in clearing damaged mitochondria, the changes in their expression levels among patients with MS remain unclear. Furthermore, research generally lacks experimental data to elucidate how they regulate autophagy activity.

Figure 2 Autophagy-associated pathways involved in MS. Multiple autophagy-related pathways are implicated in the pathogenesis of MS. Dysregulation of autophagy influences inflammatory signaling, mitochondrial quality control, endoplasmic reticulum stress responses and lipid metabolism in neural and glial cells. Key regulators include miR-223-ATG16L1 signaling, NLRP3 inflammasome activation, mitophagy-related pathways and selective autophagy receptors. MS, multiple sclerosis; IL-1β, interleukin-1β; TNF-α, tumor necrosis factor α; Th17, T helper 17 cells; NLRP3, NLR family pyrin domain containing 3; LC3, microtubule-associated protein 1 light chain 3; Atg12, autophagy-related protein 12; PINK1, PTEN-induced kinase 1; PARKIN, Parkinson protein 2, E3 ubiquitin-protein ligase; CALCOCO2, calcium-binding and coiled-coil domain-containing protein 2; CTSB, cathepsin B; ER, endoplasmic reticulum; UPR, unfolded protein response; LD lipid droplet; AMPK, adenosine monophosphate-activated protein kinase; mTOR, mammalian target of rapamycin; PI3K, phosphatidylinositol 3-kinase; PGC-1α, peroxisome proliferator-activated receptor-γ coactivator-1α; cAMP, cyclic adenosine monophosphate; PDE10A, phosphodiesterase 10A; ROS, reactive oxygen species; NO, nitric oxide.

The process of autophagy acts as an inhibitor to the progression of MS

Contrary to the previously discussed findings, several studies-both preclinical and clinical-suggest that autophagy may have a harmful effect on the progression of MS (97). Elevated expression of immune-related GTPase M 1 and ATG5 has been observed in T cells from individuals with relapsing-remitting MS (98). Notably, administration of the mTOR inhibitor rapamycin accelerated the progression of relapsing-remitting MS (99). In parallel, MS lesions displayed increased oxidative stress and inflammation, along with a reduced LC3-II:LC3-I ratio, suggesting altered or impaired autophagic flux in demyelinated regions (100). Autophagy has been implicated in both removal of myelin debris and regulation of remyelination, but defective or dysregulated autophagy may delay efficient clearance of myelin fragments and contribute to sustained neuroinflammation and neuronal injury in MS pathology (18). Additionally, stimulating autophagy by activating STAT1 in microglia led to significant white matter damage in mice. A marked increase in T-cell autophagy correlates with the severity and progression of MS.

Consequently, ATG16L2, a regulator of T-cell autophagy, has potential as a biomarker for predicting relapse in patients with MS (101). Clinical trial data reveal a decrease in ATG16L2 mRNA levels in T cells of individuals with MS, suggesting that targeting autophagy could be a viable therapeutic approach in MS (102).

Notably, there is an upregulation of autophagy-related genes and markers in T lymphocytes and tissues in MS (103). For instance, Atg5 mRNA levels are associated with neurological impairments in patients with MS, with autophagy exacerbating the severity of relapsing-remitting MS (104). Furthermore, the heightened levels of mitochondrial autophagy and general autophagic activity in the biological fluids of patients with active MS point to their potential use as indicators of disease severity (105).

Previous studies have demonstrated a significant upregulation of autophagy in T cells from individuals with MS, as evidenced by elevated ATG5 expression, suggesting a close association between ATG5-driven autophagy and enhanced inflammatory capacity in autoreactive lymphocytes (106,107). In this context, autophagy not only supports T-cell survival under inflammatory stress but also influences the functional balance between pathogenic Th17 cells and immunosuppressive regulatory T cells (Tregs), a key determinant of MS disease activity. Dysregulated autophagy has been proposed to favor Th17 polarization while impairing Treg stability, thereby exacerbating immune imbalance and neuroinflammation (108). Therapeutically, rapamycin has been shown to modulate autophagy through mTOR inhibition and to partially restore immune homeostasis by promoting Treg expansion and dampening neuroinflammation (109). Although early clinical studies indicate that rapamycin is generally well tolerated in MS, its cell type-specific effects on T cells, microglia and oligodendrocytes remain insufficiently characterized. Finally, accumulating evidence links MS pathogenesis to both T- and B-cell dysfunction, with Epstein-Barr virus-mediated B-cell activation further amplifying autoimmune responses (110). How autophagy integrates these peripheral and CNS-specific mechanisms into a unified pathogenic framework warrants further investigation.

Mitochondrial dysfunction in MS is accompanied by activation of mitochondrial autophagy, and experimental inhibition of this process has been shown to enhance myelin synthesis in oligodendrocytes and promote axonal remyelination. Consistently, liraglutide attenuates CNS inflammation and demyelination through modulation of the AMPK and pyroptosis-related NLRP3 pathway, indicating a functional link between mitochondrial stress responses, autophagy and demyelinating pathology (111). Furthermore, antipsychotic medications like clozapine and haloperidol alleviate locomotor impairments in mice caused by copper ketone by inhibiting mitochondrial autophagy. FK506-binding protein 5 (FKBP5) has been identified as a regulator of PPAR-γ, playing a role in mitigating mitochondrial autophagy to create a favorable environment for myelin regeneration. By contrast, a lack of myelin triggers PINK1/Parkin-mediated mitochondrial autophagy, where FKBP5 acts as a critical modulator through PPAR-γ. This discovery underscores FKBP5's role in controlling mitochondrial autophagy during the recovery process in demyelinating conditions, offering a potential target for treating demyelinating disorders (112) (Fig. 3).

Figure 3 Context-dependent roles of autophagy and mitophagy in MS. Autophagy and mitophagy can exert both detrimental and protective effects during MS progression. Dysregulation of autophagy-related genes in T cells, mitochondrial dysfunction and excessive mitophagy contribute to inflammation and impaired remyelination. Pharmacological modulation of autophagy-related pathways influences neuroinflammation, demyelination and myelin regeneration. MS, multiple sclerosis; Atg5, autophagy-related protein 5; LC3, microtubule-associated protein 1 light chain 3; PINK1, PTEN-induced kinase 1; IRGM1, immunity-related GTPase family M member 1; PI3K, phosphatidylinositol 3-kinase; AKT, protein kinase B; mTOR, mammalian target of rapamycin; AMPK, adenosine monophosphate-activated protein kinase; STAT1, signal transducer and activator of transcription 1; NLRP3, NLR family pyrin domain containing 3; PPAR-γ, peroxisome proliferator-activated receptor γ; FKBP5, FK506-binding protein 5.

Overall, these findings indicate that excessive autophagy activation may drive MS neuropathology, suggesting that autophagy inhibitors could be a promising strategy for slowing the progression of MS. In MS research, although autophagy is recognized as playing a crucial role in disease onset and progression, current studies still face significant limitations in understanding its mechanisms.

Collectively, the evidence summarized above suggests that autophagy in MS does not exert a uniform biological effect but instead operates within a context-dependent spectrum ranging from neuroprotection to neurotoxicity. We propose an overarching conceptual framework in which the functional outcome of autophagy in MS is determined by four interrelated factors: Disease stage, cell type, autophagy subtype and inflammatory-metabolic context.

In the early or compensatory phases of MS, moderate activation of autophagy, particularly selective forms such as mitophagy and lipophagy, appears to be predominantly protective. In this context, autophagy promotes mitochondrial quality control, limits oxidative and nitrosative stress, facilitates lipid clearance in microglia and macrophages, and supports oligodendrocyte survival and myelin regeneration. These effects collectively contribute to the maintenance of CNS homeostasis and the attenuation of neuroinflammation.

By contrast, during active or chronic stages of MS, sustained inflammatory signaling, immune cell hyperactivation and metabolic dysregulation may shift autophagy toward a maladaptive state. Excessive or dysregulated autophagy, particularly within T lymphocytes, microglia and neurons, can amplify immune-mediated damage, promote demyelination and exacerbate axonal injury. In this setting, heightened autophagic flux or prolonged mitophagy may lead to energy depletion, impaired cellular repair and reinforcement of pro-inflammatory circuits.

Importantly, this framework reconciles seemingly contradictory findings in the literature by emphasizing that autophagy intensity alone is insufficient to predict biological outcome. Rather, the balance between selective and non-selective autophagy, the cellular compartment involved, and the temporal dynamics of MS pathology jointly determine whether autophagy functions as a protective adaptive response or a driver of neurodegeneration. This integrative model underscores the need for stage-specific and cell type-targeted modulation of autophagy, rather than global activation or inhibition, as a rational therapeutic strategy in MS.

Discussion

MS is a chronic immune-mediated disorder of the CNS characterized by demyelination, axonal injury and progressive neurodegeneration. While inflammatory immune responses have long been regarded as the primary drivers of MS pathology, increasing evidence suggests that cell-intrinsic stress responses, including autophagy, play a critical modulatory role in disease progression rather than acting as mere downstream consequences of inflammation. Accumulating experimental and clinical studies indicate that autophagy is dysregulated in MS across multiple cellular compartments. Altered expression of canonical autophagy markers, such as Beclin-1, has been observed in animal models and patient-derived samples, alongside perturbations in key regulatory pathways, including Akt/mTOR signaling. These findings collectively suggest that autophagic flux is disturbed in MS. However, most available evidence remains correlative, relying heavily on static measurements of autophagy-related markers, which limits the ability to distinguish between increased autophagy induction and impaired autophagic degradation. This methodological limitation represents a major barrier to establishing causality between autophagy dysfunction and MS pathogenesis. Notably, emerging data increasingly implicate mitochondrial dysfunction and mitochondrial autophagy (mitophagy) in MS. Elevated levels of mitophagy-related markers in cerebrospinal fluid and serum, together with associations between mitochondrial injury and neuroaxonal degeneration, support the notion that mitochondrial quality control is particularly vulnerable in MS. Nevertheless, it remains elusive whether mitophagy activation represents a compensatory neuroprotective response or a maladaptive process contributing to cellular energy failure and neurodegeneration. Importantly, the relative contribution of selective mitophagy vs. non-selective autophagy has not been systematically addressed, highlighting a critical knowledge gap in the field.

Comparative insights from other neurodegenerative diseases further emphasize the complexity of autophagy regulation in neurodegeneration. Although MS and ALS differ substantially in etiology, both diseases exhibit convergent features, including mitochondrial impairment, oxidative stress and defective autophagic clearance. These parallels suggest that shared autophagy-related mechanisms may underlie neurodegenerative vulnerability, while disease-specific regulatory contexts likely determine whether autophagy exerts protective or detrimental effects.

Despite substantial progress, several unresolved questions warrant further investigation. First, the cell-type specificity of autophagy dysregulation in MS remains poorly defined. Most studies focus on neurons or oligodendrocytes in isolation, neglecting the dynamic interactions among immune cells, glial populations and neurons within the inflammatory microenvironment. Second, advances in high-resolution imaging, single-cell transcriptomics and artificial intelligence-based data integration offer unprecedented opportunities to dissect autophagy dynamics at spatial and temporal resolutions previously unattainable. The emerging association between autophagy-related pathways and MRI outcomes further underscores the potential clinical relevance of these approaches (113,114). In patients with relapsing-remitting MS, circulating and cerebrospinal fluid markers of autophagy and mitophagy were found to be significantly elevated during active disease phases and showed quantitative correlations with MRI activity (115). Specifically, levels of Parkin and BNIP3 were increased by ~1.5-2.5 fold in patients with gadolinium-enhancing lesions compared with radiologically inactive patients, and these markers correlated positively with the number of contrast-enhancing lesions (r≈0.40-0.55, P<0.01). Furthermore, autophagy- and mitophagy-related markers displayed moderate associations with T2 lesion load (r≈0.35-0.48), suggesting a link between dysregulated autophagy and MRI-defined inflammatory burden in MS. While MRI metrics reflecting long-term neurodegeneration, such as brain atrophy, were not directly assessed in these studies (113-115), their established relationship with axonal injury markers (e.g., neurofilament light chain) indicates that autophagy-related alterations may be indirectly connected to MRI measures of disease progression.

Finally, although autophagy represents an attractive therapeutic target, its dual role in MS poses significant translational challenges. Both insufficient and excessive autophagy may exacerbate pathology, suggesting that context-dependent and stage-specific modulation, rather than global activation or inhibition, will be required. Beyond mitophagy, relatively underexplored selective autophagy pathways, such as ER autophagy and lipophagy, may also contribute to MS pathology and warrant systematic investigation. Bridging the gap between mechanistic insights and clinical application will require integrative strategies combining molecular biology, systems-level analysis and longitudinal patient studies.

In conclusion, autophagy emerges as a central but complex regulator in MS, positioned at the intersection of immune dysregulation, mitochondrial dysfunction and neurodegeneration. Clarifying its precise mechanistic roles and therapeutic potential remains a critical priority for future research.

Availability of data and materials

Not applicable.

Authors' contributions

DW, MW and HF wrote and revised the review. HS, QF, YZ, GL, YB, YY prepared figures. All authors reviewed the manuscript and have read and approved the final manuscript. Data authentication is not applicable.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Acknowledgments

Not applicable.

Funding

This work was supported by the Joint Fund of Henan Provincial Science and Technology R&D Project (grant no. 242103810034), Henan Academy of Innovations in Medical Science 'Three Hundreds' Plan (grant no. HNCMS202437), Henan Provincial Young and Middle-aged Health Science and Technology Innovation Leading Talent Training Program (grant no. LJRC2024019) and Heluo Young Talents Support Program (grant no. 2024HLTJ08).

References