The dynamic balance of mitochondrial morphology is maintained by the cooperative action of fusion proteins (Mfn1/2 regulating the outer membrane, OPA1 regulating the inner membrane) and fission proteins (Drp1 and its receptor complexes). Imbalance between the two leads to defects in mitochondrial autophagy and the accumulation of dysfunctional mitochondria (Chan, 2020; Chen et al., 2023; Quintana-Cabrera and Scorrano, 2023). Additionally, mitochondrial transport relies on the microtubule network and associated motor protein complexes, which, in conjunction with dynein, regulate the positioning of mitochondria in neuronal axons, with Mfn2 participating in this process through physical binding to the Miro-Milton complex (Chan, 2020; Gao et al., 2017). In the pathological process of PD, mitochondrial dynamics abnormalities manifest as excessive fission and fusion defects, leading to mitochondrial network fragmentation, energy metabolism disorders, and imbalances in reactive oxygen species (ROS) and calcium homeostasis (Subramaniam and Chesselet, 2013; Rani and Mondal, 2020). A study published in the Journal of Neurochemistry in 1990 first confirmed that the activity of mitochondrial complex I in the substantia nigra of PD patients was significantly reduced compared to healthy controls (decreased by 30%, p < 0.05), while the activity of complexes II-III was unaffected (Schapira et al., 1990) Correspondingly, an animal model experiment showed that a single injection of 40 mg/kg MPTP in mice of different ages resulted in age-related declines in complex I function, antioxidant capacity, and increased MAO-B activity, the latter accelerating the conversion of MPTP to toxic metabolite MPP+, suggesting that MPTP-induced oxidative stress and complex I inhibition are potential driving factors of early PD pathology (Ali et al., 1994). To adapt to this functional defect, the density of mitochondria and the expression of complex I/IV proteins in surviving axons increase compensatorily (Reeve et al., 2018). However, the forced enhancement of mitochondrial respiratory chain activity during compensation induces the accumulation of mitochondrial-derived oxidative damage, ultimately triggering or exacerbating neuroinflammatory responses, leading to the collapse of compensatory repair. Pathogenic α-syn mutations (e.g., A30P) significantly reduce its localization in MAM by disrupting its interaction with MAM lipid rafts, leading to decreased physical connections between ER and mitochondria (lower Mander coefficient) and impaired calcium signaling mediated by MAM (Guardia-Laguarta et al., 2014; Jiao et al., 2025); this subsequently leads to reduced mitochondrial calcium uptake and energy metabolism disorders (Calì et al., 2012; Devi et al., 2008). Notably, this fragmentation is not driven by the classical fission pathway mediated by DRP1 but is associated with abnormal cleavage of OPA1 protein, suggesting that α-syn may indirectly affect morphology by regulating mitochondrial fusion-related proteins (Duvezin-Caubet et al., 2006; Guardia-Laguarta et al., 2014). Furthermore, the reduced presence of mutant α-syn in MAM may promote its abnormal accumulation in the mitochondrial membrane, directly interfering with mitochondrial membrane potential and exacerbating oxidative stress by inhibiting complex I activity (Devi et al., 2008), thus forming a vicious cycle of calcium imbalance, energy metabolism disorders, and oxidative damage.

Abnormal accumulation of α-synuclein delays mitochondrial autophagy by upregulating Miro protein levels. Specifically, the clearance rate of damaged mitochondria on the surface of Miro decreases, leading to delayed mitochondrial transport and ultimately resulting in the accumulation of abnormal mitochondria within neurons (Shaltouki et al., 2018). Miro’s abnormal accumulation on the surface of damaged mitochondria causes the motor protein complex (kinesin/Miro/milton) to continuously bind, leading to excessive transport of mitochondria to the distal axon and their retention (Lee and Lu, 2014). Retained mitochondria produce ROS due to electron transport chain dysfunction, and local oxidative stress exacerbates synaptic damage and mtDNA leakage (Verma et al., 2023). Notably, in PD patients, mutations in PINK1 prevent the phosphorylation of Miro at the Ser156 site, leading to the obstruction of Parkin-dependent ubiquitin degradation (Shlevkov et al., 2016; Narendra et al., 2012). The absence of Parkin directly blocks mitochondrial autophagy, resulting in the accumulation of abnormal mitochondria in neuronal cell bodies and axons, ultimately leading to selective death of dopaminergic neurons (Liu et al., 2012; Wang et al., 2011a,b). Accordingly, it can be reasonably inferred that retained damaged mitochondria release DAMPs, such as mtDNA and ROS, activating the NLRP3 inflammasome in microglia and triggering the IL-1β/IL-18 cascade, forming a vicious cycle of “protein abnormal aggregation-mitochondrial damage-neuroinflammation-neuronal death.”

Studies have shown that Drp1-dependent fission is a key upstream event in the pathology of PD (Zhang D. et al., 2019; Zhang Q. et al., 2019). MPP + induces energy metabolism disorders by inhibiting mitochondrial complex I, activating Drp1-mediated excessive mitochondrial fission, leading to mitochondrial fragmentation (Santos et al., 2015; Chuang et al., 2016). The ROS produced by fragmented mitochondria and abnormal calcium signaling further positively feedback to enhance Drp1 activity, forming a vicious cycle that ultimately leads to the death of dopaminergic neurons. Inhibiting Drp1 can block fragmentation, almost completely rescuing MPP + -induced ROS generation, loss of mitochondrial membrane potential, and cell death (Santos et al., 2015; Yang et al., 2021; Wang et al., 2011a,b). Based on the above mechanisms, targeting the inhibition of Drp1 activity or regulating its mitochondrial translocation may effectively delay the degeneration of dopaminergic neurons in the prodromal phase of Parkinson’s disease by blocking the vicious cycle of mitochondrial fragmentation-ROS-calcium imbalance, providing a new strategy for early neuroprotection (see Figure 2).

For example, mutations in the Parkin and PINK1 genes lead to early-onset Parkinson’s disease, with mechanisms involving defects in the PINK1/Parkin pathway that cause abnormal mitochondrial quality control. When the E3 ubiquitin ligase activity of Parkin is impaired, the abnormal ubiquitination of Mfn1/2 hinders mitochondrial fission and autophagosome formation, resulting in decreased mitochondrial clearance capacity (Gegg et al., 2010). In PINK1 knockout rat models, dopaminergic neurons in the substantia nigra exhibit impaired mitochondrial respiratory function and elevated oxidative stress levels, confirming that mitochondrial homeostasis imbalance is associated with PD pathology (Wang et al., 2023). Defects in the PINK1/Parkin pathway can induce the leakage of mitochondrial DNA (mtDNA) into the cytoplasm, activating the cGAS-STING pathway as a damage-associated molecular pattern (DAMP), which in turn triggers NLRP3 inflammasome assembly and promotes the release of pro-inflammatory factors such as IL-1β and IL-18 (Wang et al., 2023; Mouton-Liger et al., 2018). The absence of Parkin weakens the negative feedback regulation of NF-κB signaling by downregulating the anti-inflammatory protein A20, further amplifying NLRP3 inflammasome activity (Mouton-Liger et al., 2018). Meanwhile, PINK1 deficiency exacerbates the loss of mitochondrial membrane potential and calcium homeostasis imbalance, activating inflammatory responses in microglia and macrophages, leading to a chronic neuroinflammatory microenvironment (Wang et al., 2023). The continuous release of pro-inflammatory factors such as IL-1β and TNF-α synergizes with mitochondrial dysfunction, resulting in oxidative stress, energy metabolism exhaustion, and abnormal aggregation of α-synuclein in dopaminergic neurons, ultimately triggering neuronal apoptosis and degeneration of the nigrostriatal pathway (Wang et al., 2023). The LRRK2 G2019S mutation causes excessive mitochondrial fission by enhancing the phosphorylation of Drp1 at the Thr595 site, accompanied by increased levels of reactive oxygen species (ROS) and mtDNA damage (Su and Qi, 2013). This damage is manifested as a reduction in mtDNA copy number and its cytoplasmic leakage, directly activating the NLRP3 inflammasome (Pena et al., 2024). He et al. (2023) confirmed that the LRRK2 G2019S mutation enhances NLRP3 activation in astrocytes through the NF-κB pathway, promoting the release of IL-1β and TNF-α, a mechanism validated in mouse models and lymphoblasts derived from patients, with LRRK2 kinase activity inhibition significantly reducing inflammasome activity. Zhu et al. (2013) further revealed that LRRK2 G2019S enhances ULK1-dependent mitophagy by activating the JNK signaling pathway, exacerbating mitochondrial clearance abnormalities and neuronal damage. These studies suggest that defects in the PINK1/Parkin pathway and LRRK2 mutations accelerate the pathological process of PD through mitochondrial stress-related inflammatory signaling, and early intervention during the prodromal phase could target the inhibition of NLRP3 or enhance the expression of the anti-inflammatory protein A20 to delay neuronal loss (Wang et al., 2023; Mouton-Liger et al., 2018). Notably, the mtDNA damage caused by the LRRK2 G2019S mutation due to abnormal kinase activity exhibits significant dynamic reversibility. Treatment with EB-42168 or MLi-2 for 2 h to acutely inhibit G2019S LRRK2 kinase activity can rapidly restore mtDNA damage to normal levels, while the damage phenotype reappears within 2 h after the withdrawal of the inhibitor (Pena et al., 2024). This suggests that mitochondrial dysfunction during the prodromal phase may be in a “metabolic imbalance” state rather than structural damage, and this dynamic balance of damage-repair characteristics provides a critical time window for early intervention based on kinase activity monitoring. Dynamic monitoring of mtDNA damage indices and inflammatory factors may construct a combination of prodromal-specific biomarkers, while the combined application of gene editing technology for targeted mtDNA repair and selective kinase inhibitors may achieve precise regulation from molecular mechanisms to clinical phenotypes. Additionally, using induced pluripotent stem cells to construct patient-specific brain organoid models will provide a new platform for optimizing personalized treatment plans.

Recent studies indicate that DJ-1 is a key downstream mediator of PINK1/parkin-dependent mitophagy (Thomas et al., 2011; McCoy and Cookson, 2011). In fibroblasts and induced dopaminergic neurons from PARK7 mutation patients, the absence of DJ-1 leads to the failure of the autophagy receptor optineurin to be recruited to depolarized mitochondria, directly hindering the clearance of damaged mitochondria and causing the accumulation of mitochondrial fragments (Imberechts and Vandenberghe, 2023). Experiments confirm that the mitochondrial localization of DJ-1 depends on the PINK1/parkin pathway, and artificially targeting DJ-1 to the outer mitochondrial membrane can bypass upstream defects to directly restore optineurin recruitment (Imberechts et al., 2022). These findings indicate that DJ-1 plays a critical role in maintaining mitochondrial quality control, and its deficiency, by blocking downstream steps of autophagy, contributes to the degenerative changes in dopaminergic neurons in conjunction with PINK1/parkin mutations, representing a common pathological mechanism of autosomal recessive PD (see Figure 3).

The NLRP3 inflammasome, as a core regulatory component of innate immunity, is closely related to the neuroinflammatory process and loss of dopaminergic neurons in PD. This complex is composed of the NLRP3 receptor protein, the adaptor protein ASC, and the precursor caspase-1 (Lamkanfi and Dixit, 2014), which triggers polymer assembly and activates caspase-1 by sensing danger signals such as intracellular potassium efflux and mitochondrial reactive oxygen species (ROS) bursts (Fu and Wu, 2023; Zhang et al., 2023), subsequently cleaving pro-IL-1β and pro-IL-18 to generate mature inflammatory factors while inducing pyroptosis (Swanson et al., 2019). Co-culture experiments show that NLRP3-activated microglia significantly increase the death of dopaminergic neurons (SH-SY5Y and MN9D cells) (Lee et al., 2019).

In PD patients, mitochondrial ROS serve as the initial signal promoting NLRP3 activation (Mishra et al., 2021), and Parkin can directly regulate NLRP3 stability. The absence of Parkin leads to the accumulation of NLRP3 protein in microglia, enhancing caspase-1 activation and IL-1β secretion, ultimately resulting in the loss of dopaminergic neurons in the substantia nigra (Yan et al., 2023). Specifically, mitochondrial autophagy impairment hinder s the clearance of damaged mitochondria, and the released ROS and mtDNA activate neuroinflammation through dual mechanisms (Lin and Beal, 2006; Yu et al., 2025): on one hand, excessive ROS production due to mitochondrial dysfunction directly promotes NLRP3 inflammasome assembly (Sarkar et al., 2017; Zhou et al., 2011); on the other hand, the released mtDNA from damaged mitochondria synergizes with ROS, further activating the NLRP3 inflammasome by enhancing mitochondrial membrane potential depolarization and lysosomal rupture, driving caspase-1-dependent maturation and secretion of IL-1β (Shao et al., 2021; Javed et al., 2020; Holley and Schroder, 2020). Additionally, recent studies indicate that LPS-induced metabolic reprogramming enhances glycolysis and succinate accumulation in the TCA cycle, increasing mitochondrial membrane potential (Δψm) by reducing F1F0-ATP synthase-dependent proton backflow, while succinate oxidation maintains the pool of reduced CoQ, collectively driving complex I reverse electron transport (RET)-dependent mtROS bursts that regulate NLRP3 inflammasome activation and IL-1β release (Casey et al., 2025). Furthermore, studies have shown that the oxidized sterol metabolite 27-hydroxycholesterol (27-OHC) triggers a cascade of damage by inducing lysosomal membrane permeabilization (LMP): the destruction of lysosomal membrane integrity leads to abnormal leakage of cathepsin B (CTSB), which activates the NLRP3 inflammasome and drives the cleavage of the pyroptosis execution protein GSDMD in a caspase-1-dependent manner, directly causing neuronal inflammatory death (Chen et al., 2019); on the other hand, the abnormal localization of CTSB further exacerbates lysosomal dysfunction, forming a vicious cycle. It is noteworthy that the functional defects of CTSB have a multiple amplification effect—recent studies reveal that CTSB gene knockout or inhibition leads to a triad effect of decreased lysosomal degradation capacity: reduced GCase activity (synergistic effect with GBA1 mutations), autophagic flow blockage, and compensatory increase in lysosomal generation but decreased degradation capacity. This lysosomal collapse state significantly enhances the pathological aggregation of phosphorylated α-synuclein (pSyn-S129) induced by α-synuclein preformed fibrils (PFF), directly leading to α-synuclein aggregation, exacerbating the pathological progression of PD, and potentially initiating an irreversible neurodegenerative process during the prodromal phase of PD (Jones-Tabah et al., 2024; Moors et al., 2024).

The underlying mechanisms of this pathological network are closely related to mitochondrial-lysosomal interaction disorders. GBA1 mutations lead to decreased β-glucocerebrosidase (GCase) activity, triggering the accumulation of glucosylceramide (GlcCer), which degrades TBC1D15 through the ubiquitin-proteasome pathway, ultimately hindering Rab7-GTP hydrolysis, prolonging mitochondrial-lysosomal (M/L) contact time, and inhibiting Drp1-mediated fission and PINK1-Parkin pathway-dependent autophagy initiation (Kim et al., 2021; Xie et al., 2025), forming a dual defect in mitochondrial quality control. Additionally, Parkin mutations significantly reduce M/L contact frequency by decreasing Rab7 activity (dependent on its E3 ubiquitin ligase function), leading to a deficiency of essential amino acids for mitochondria, such as isoleucine and valine, while causing amino acid accumulation in the lysosomal lumen. This not only impairs mitochondrial oxidative phosphorylation function but also further weakens lysosomal clearance capacity by inhibiting the maturation of lysosomal enzymes such as cathepsin D (Peng et al., 2020). This interaction disorder creates a chain reaction of “quality control failure-metabolic disorder” in a spatiotemporal dimension: GBA1/Parkin mutations lead to abnormal contact time/frequency, while CTSB defects disrupt metabolic terminal clearance capacity, intertwining α-synuclein aggregation and the release of inflammatory factors. The IL-1β released after NLRP3 activation exacerbates α-synuclein aggregation and oxidative stress by activating the P38/STAT1 pathway, while IL-6 secreted by neurons enhances the secretion of pro-inflammatory factors from microglia through the JAK2/STAT3 signaling pathway, forming a cascading amplification effect of “inflammatory factors-protein aggregation” (Lee et al., 2023; Chen et al., 2021).

It is noteworthy that the characteristic pathological product of PD—fibrillated α-synuclein aggregates—can also act as an endogenous danger signal to activate the NLRP3 inflammasome: studies have confirmed that microglia phagocytose α-synuclein fibrils derived from Lewy bodies, releasing cathepsin B through lysosomal membrane rupture, accompanied by a large generation of ROS, which synergistically activates the NLRP3-caspase-1 axis, driving the maturation and release of IL-1β (Yu et al., 2024; Li G. et al., 2024; Li M. M. et al., 2024; Codolo et al., 2013; Martinon et al., 2002). The released IL-1β further recruits and activates surrounding glial cells, forming a chronic neuroinflammatory microenvironment that exacerbates mitochondrial dysfunction in dopaminergic neurons (Halle et al., 2008). Abnormally aggregated fibrillar α-synuclein drives the inflammatory response through dual mechanisms: on the one hand, by activating Toll-like receptor 2 (TLR2) and TLR3, it synergistically enhances the nuclear factor κB (NF-κB) signaling pathway and upregulates the gene expression of TNF-α and IL-1β (Wang et al., 2019; Daniele et al., 2015), and on the other hand, it relies on lysosomal rupture, reactive oxygen species generation, and cathepsin B release to promote the maturation and secretion of IL-1β by triggering the NLRP3 inflammasome, thereby amplifying neuroinflammation (Fan et al., 2020). This forms a positive feedback loop of “α-synuclein deposition-inflammation amplification-neuronal death” (Codolo et al., 2013). Genetic evidence further supports this mechanism: αSyn-induced neuroinflammation and motor dysfunction are significantly alleviated in NLRP3 gene knockout mice (Gordon et al., 2018). In summary, the activation mechanism of the NLRP3 inflammasome and its role in PD provide important clues for understanding the pathological mechanisms of this disease, while also laying the foundation for developing new therapeutic strategies. Future research needs to further explore the regulatory mechanisms of the NLRP3 inflammasome and its potential applications in PD treatment.

Based on existing research evidence, during the prodromal phase of PD, the interplay between mitochondrial-lysosomal interaction disorders and neuroinflammation creates a unique diagnostic and intervention window. The ROS and mtDNA release caused by mitochondrial autophagy defects can be detected through the following pathways: elevated mtDNA fragments in cerebrospinal fluid, abnormal mitochondrial membrane potential (Δψm), and activation markers of the NLRP3 inflammasome such as caspase-1 cleavage products and IL-1β levels. Lysosomal function assessment should focus on CTSB activity detection, lysosomal membrane stability markers, and GCase activity evaluation (especially for GBA1 mutation carriers). Notably, the oligomer levels of phosphorylated α-synuclein (pSyn-S129) combined with metabolomics analysis may reveal the interaction between “metabolic reprogramming-α-syn pathology, “providing dynamic evidence for risk assessment of the transition from prodromal to clinical stages. Additionally, methods to detect NLRP3 ubiquitination levels in peripheral blood mononuclear cells, such as immunoprecipitation-mass spectrometry combined techniques, can be developed to assess Parkin functional status, providing warning markers for the prodromal phase of PD. It is emphasized that the sensitivity and specificity of this method still need clinical cohort validation, and interference from other E3 ubiquitin ligases must be excluded.

Intervention strategies should be based on the spatiotemporal characteristics of the pathological network: at the metabolic level, drugs that inhibit reverse electron transfer (RET) of complex I block glycolysis-enhanced mtROS bursts (Casey et al., 2025); supplementing essential mitochondrial amino acids such as isoleucine and valine to alleviate metabolic compartmentalization imbalance caused by Parkin mutations. To address lysosomal-mitochondrial interaction disorders, targeting the dynamic balance of Rab7 activity to restore the synergy of mitochondrial fission and autophagy (Jimenez-Orgaz et al., 2018; Liang et al., 2023); CTSB activators or lysosomal membrane stabilizers to restore lysosomal degradation capacity (Prieto Huarcaya et al., 2022), blocking the “seed effect” of α-syn pathological aggregation. At the level of inflammation regulation, combining NLRP3 inhibitors with necroptosis pathway blockers simultaneously inhibits IL-1β secretion and neuronal inflammatory death (Tantra et al., 2024; Zhou et al., 2024); microglia-specific JAK2/STAT3 pathway inhibitors cut off IL-6-mediated inflammation spread (Su and Qi, 2013). For genetically susceptible populations, integrating enzyme replacement therapy with personalized metabolic interventions is necessary to break the “mutation-metabolism-inflammation” vicious cycle. The core of this multidimensional intervention system lies in the early identification of the pathological critical point of “mitochondrial autophagy defects-lysosomal collapse-inflammation cascade, “through spatiotemporally specific drugs, such as nanocarriers targeting mitochondrial-lysosomal contact regions, to reshape cellular homeostasis before irreversible neurodegeneration, providing a new paradigm for disease-modifying treatment in the prodromal phase of PD.

Lactylation modification is a newly discovered epigenetic regulatory mechanism that dynamically regulates protein function through the covalent binding of lactyl groups to lysine residues, becoming a core hub connecting metabolic abnormalities and neurodegenerative diseases (see Figure 4).

Lactylation is involved in physiological processes such as energy metabolism and inflammatory responses. In non-tumor diseases like atherosclerosis and Alzheimer’s disease, lactylation exacerbates pathological damage by regulating inflammation, fibrosis, and cellular senescence (Zhao et al., 2025; Zhang D. et al., 2019; Zhang Q. et al., 2019; Brand et al., 2016). As mentioned earlier, in the pathological process of PD, the abnormal aggregation of α-synuclein has been confirmed to disrupt the integrity of the mitochondria-associated endoplasmic reticulum membrane (MAM), leading to impaired function of mitochondrial complex I and accumulation of reactive oxygen species (ROS) (Guardia-Laguarta et al., 2014). Studies on MPTP-induced Parkinson’s disease models indicate that its toxic metabolite MPP + significantly reduces ATP synthesis efficiency by inhibiting mitochondrial complex I (Ali et al., 1994), prompting cells to activate glycolytic pathways to increase lactate production. This phenomenon shares similar metabolic characteristics with the Warburg effect observed in the tumor microenvironment (Certo et al., 2021). Lactate can inhibit OXPHOS activity through lactylation modification of mitochondrial metabolic enzymes mediated by AARS2, such as PDHA1 K336, forming a negative feedback loop to limit excessive ROS generation (Mao et al., 2024). When this regulatory mechanism is disrupted (e.g., AARS2 deficiency), excessive OXPHOS activity leads to ROS accumulation, potentially triggering mitochondrial oxidative damage and ultimately resulting in the accumulation of dysfunctional mitochondria (Mi et al., 2023). The role of histone lactylation modification in neurological diseases has some experimental support. Research through proteomic analysis has found widespread protein lactylation modifications in mouse brain tissue, and neuronal excitation can significantly increase lactate levels and histone lactylation levels (Hagihara et al., 2021; Kuang et al., 2025; Sun et al., 2025) (Table 1).

In summary, the abnormal accumulation of lactate may exacerbate neuroinflammation through a tripartite attack:

Lactylation is a key link between metabolism and epigenetic regulation. Some studies have found that lactyl-CoA is a donor for lactylation, and its modification is catalyzed by CBP (CREBBP) or p300 (EP300). CBP/p300 acts as the writer enzyme for lactylation modification, catalyzing the lactylation of histone lysines through its acetyltransferase structural domain (Zhang D. et al., 2019; Zhang Q. et al., 2019; Dai et al., 2022). Histone deacetylases HDAC1-3 have the activity to remove H3K18la and act as “eraser enzymes” for lactylation, inhibiting the transcriptional activity of related genes by hydrolyzing lactylation modifications (Dai et al., 2022). Therefore, targeting the regulation of the activity of writer or eraser enzymes may balance the levels of lactylation in the brain, potentially providing a direction for neuroprotection during the prodromal phase of PD, but further experimental validation is needed. The discovery of lactylation modification provides a new perspective for PD treatment: as a triple regulatory node of metabolism, epigenetics, and inflammation, it can explain the co-occurrence of energy metabolism abnormalities and neuroinflammation in PD, and lay a theoretical foundation for developing multidimensional intervention strategies targeting both the NLRP3 inflammasome and mitophagy. Future studies should precisely analyze the spatiotemporal dynamics of lactylation modification and its interaction mechanisms with α-synuclein pathology in PD models using single-cell metabolic imaging combined with CUT&Tag technology.