The Yin and Yang of Microglia-Derived Extracellular Vesicles in CNS Injury and Diseases

The mechanisms of exosome biogenesis remain poorly understood. The current consensus in the scientific literature is that EVs originate either from intracellular endocytic trafficking or the plasma membrane [30,31] (Figure 2). The main characteristic of exosomes is their endosomal origin, which distinguishes them from apoptotic bodies and microvesicles that are formed through the inward budding of the endosomal plasma membrane [31,32], leading to the formation of late endosomes and multivesicular bodies (MVBs) inside the cell. These MVBs can either fuse with the lysosome complex for degradation or they are translocated to the plasma membrane where they fuse with the cell membrane and exit the cell to release their intraluminal vesicles as exosomes into the extracellular space [33,34,35]. The Endosomal Sorting Complex Required for Transport (ESCRT) protein plays a key role in orchestrating the cargo loading and vesicle release [36]. This involves binding large multi-subunit protein complexes to ubiquitinated proteins, inducing the inward budding of the endosomal membrane. Each complex within the ESCRT pathway contains multiple ubiquitin-binding domains, which are responsible for sorting ubiquitinated proteins into intraluminal vesicles (ILVs) and determining the cargo of exosomes. On the other hand, there also exists an alternate exosome pathway that is ESCRT-independent and is regulated by phosphorylated RAB31. The alternative pathway employs flotillin proteins in lipid raft microdomains that prevent the fusion of the multivesicular endosomes with lysosomes and instead enables the secretion of the ILVs outside of the cells as exosomes [37].

The release of EVs from microglia is accelerated by various factors that often precede the activation of the parent microglia and are usually associated with environmental changes or cellular stress. Some of the key triggers include cytokines (TNF-α, IL-1β, IFN-γ), chemokines, lipopolysaccharides (LPSs), oxidative stress (ROS), oxygen deprivation or hypoxia, glutamate, serotonin, and adenosine triphosphate (ATP) [17]. A well-studied example that potentiates the release of MGEVs under normal physiological conditions is serotonin, which is released by neurons. Serotonin activates 5-HT2a, 5-HT2b, and 5-HT4 serotonin receptors on the parent microglia cells. The activation of these receptors elevates intracellular calcium levels, promoting the fusion of multivesicular bodies (MVBs) with the plasma membrane with a subsequent release of the EVs by these glial cells [17]. Another potent inducer of exosome secretion from microglial cells is ATP, which is released by damaged or stressed neurons. Elevated concentrations of extracellular ATP present within the microenvironment of microglial cells bind to the purinergic receptors on microglia, leading to their activation and subsequent MGEV release [38]. Both in vitro studies and in vivo experiments conducted by Verderio et al. (2012) in mice with subclinical neuroinflammation demonstrated that MGEVs produced following ATP stimulation triggered a much stronger inflammatory response in activated microglia [39]. Analysis of the proteome of MGEVs released during ATP-induced microglial activation revealed that ATP drives the sorting of proteins involved in cell adhesion, extracellular matrix (ECM) organization, and the autophagy–lysosomal pathway, and enzymes associated with cellular metabolism, which regulate the function of recipient astrocytes and neurons within the local microenvironment [39].

MGEVs contain a diverse array of biologically active cargo molecules. When MGEVs are taken up by recipient cells, these cargoes modulate their function and thus play a crucial role in intercellular communication within the CNS [40]. The cargo composition of MGEVs includes a variety of the molecular constituents of their parent cell of origin, consisting of proteins, lipids, and nucleic acids, including noncoding RNAs, each contributing to diverse biological effects. In general, EVs comprise phospholipid bilayers of phosphatidylserine, phosphatidylcholine, sphingomyelin, ceramides, and cholesterol that combine to maintain the structural integrity and function of these nanovesicles [15]. The EV surface is decorated with various antigens and proteins which interact with corresponding receptors on target cells to trigger specific intracellular signaling pathways [41]. EVs also carry glycoproteins, glycolipids, and signaling molecules including cytokines, growth factors, and an array of metabolites [41].

MGEVs can interact with a variety of cell types within the CNS, including neurons, astrocytes, oligodendrocytes, and other microglia. Through these interactions, they play an important role in modulating the CNS environment [107]. The interaction and uptake of exosomes in recipient cells occurs through various mechanisms [108], including fusion, phagocytosis, pinocytosis, and direct surface contact. MGEV uptake by target cells primarily includes clathrin-mediated endocytosis, micropinocytosis, uptake via lipid raft-associated endocytosis, and direct fusion with the plasma membrane. Clathrin-mediated endocytosis and micropinocytosis are the primary means of EV uptake in microglia. In the former method, exosomes adhere to the plasma membrane and are subsequently internalized into endosomes where they are either degraded in the lysosome or may escape the lysosomes to release their cargoes into the cytoplasm. Macropinocytosis, on the other hand, which involves the plasma membrane forming ruffles in an actin-dependent process to internalize extracellular substances, is also a major means of exosome uptake in microglia, a process that is normally constitutive in these glial cells [109]. Exosome uptake is a more selective process that is driven by the recipient cell type. Recent research indicates that, under healthy physiological conditions, astrocytes are unable to internalize exosomes from oligodendrocytes or microglia [110]. Microglia, on the other hand, have been reported to effectively internalize EVs released from a wide range of neural cell types [107,109,111].

The mechanisms by which MGEVs exert their effects is similar to EVs from other cell types and involves the direct transfer of bioactive molecules to recipient cells, the modulation of receptor-mediated signaling pathways, and gene expression changes driven by exosomal miRNAs and other regulatory RNAs [7,112]. MGEVs transfer their cargo constituting miRNAs or proteins that regulate inflammatory signaling cascades in the recipient cells, either by amplifying or dampening neuroinflammatory responses [17,86,90,113,114]. In neuropathological conditions, the role of MGEVs in intercellular communication is important. MGEVs can convey signals that either promote the clearance of insoluble protein aggregates, such as amyloid-beta (Aβ) in Alzheimer’s disease [115], or propagate the spread of pathogenic proteins, such as tau or α-synuclein [25,116,117]. MGEVs, like exosomes from most other cell types, can cross the blood–brain barrier (BBB) [118,119,120]. Using in vitro models, it was demonstrated that MGEVs from reactive microglia dramatically enhanced the permeability of the BBB while reducing the expression of BBB-related proteins like Claudin-1, occludin, and ZO-1. In contrast, MGEVs from resting microglia did not alter the integrity of the BBB [120]. Thus, MGEVs not only mediate local immune responses but, due to their ability to cross the BBB, can affect target cells at a distance and systemically or even be employed as vehicles for delivering therapeutic agents to the brain.

Contrary to the pro-inflammatory role of MGEVs under specific pathological conditions, MGEVs can also exert neuroprotective effects through their delivery of anti-inflammatory cytokines (e.g., IL-4, IL-10) and immunomodulatory molecules that suppress excessive immune responses in the CNS [22,124,125]. This immunomodulatory function of MGEVs helps to mitigate neuroinflammation and attenuate neuronal damage as well as promote neuronal survival and repair [66] in various neurological disorders. The ability of MGEVs to either attenuate or amplify inflammatory responses in the CNS depends on their cargo. In the context of neuroprotection, MGEVs carrying miRNAs, such as miR-146a and miR-124 [55,86,126], have been shown to reduce the expression of pro-inflammatory cytokines, with the MGEVs acting through the negative regulation of the TRAF6/NF-κB signaling axis [127,128,129] and inhibiting the activation of the NLRP3 inflammasome [98]. A study by Ge et al. (2020) [126] demonstrated that MGEVs enriched with miR-124-3p alleviated neurodegeneration and improved cognitive function in TBI [86]. These miRNA species were shown to target TLR4/MyD88/NF-κB p65 signaling, modulate the inflammatory response, and reduce TBI-associated damage.

In addition to protective miRNAs, MGEVs can also deliver neurotrophic factors and antioxidants to neurons. MGEVs from activated microglia can contain brain-derived neurotrophic factor (BDNF), nerve growth factor (NGF), and glial cell-derived neurotrophic factor (GDNF), which are known to promote neuronal survival, enhance synaptic plasticity, and support axonal regeneration [130,131]. A recent study found that following MCAO, TGFβ carried by intravenously administered MGEVs promoted neuronal survival, stimulated angiogenesis, and shifted activated microglia towards a beneficial phenotype in ischemic brain tissue by activating the SMAD-2/3 pathway [132]. MGEVs also carry proteins and enzymes within their exosome cargo that help to reduce oxidative stress in the CNS. Peng et al. [100] conducted studies using a mouse model of SCI, demonstrating that MGEVs helped to mitigate oxidative stress and prevent endothelial dysfunction, while promoting the restoration of endogenous microvascular repair following SCI. It was subsequently demonstrated that these MGEVs promoted the survival, regeneration, and function of endothelial cells by modulating the Keap1/Nrf2/HO-1 signaling axis, which resulted in an increase in downstream antioxidative-related genes including NQO1, Gclc, Cat, and Gsx1 as well as levels of the transcription factor Nrf2 and heme oxygenase 1 (HO-1), which are critical regulators of oxidative stress in cells [133]. Similarly, modified MGEVs derived from IL-4-primed parent microglia reduced pericyte apoptosis, promoted vascular remodeling, and enhanced endothelial cell proliferation, resulting in decreased vascular leakage in rodent models of diabetic retinopathy [134]. These MGEVs not only decreased neuronal impairment and mitochondrial dysfunction in experimental AD models [135] but also provided neuroprotection by reducing A1 astrocyte activation, leading to motor function recovery in rodent models of spinal cord injury (SCI) [101].

In addition to their support of neuronal survival, MGEVs have also been demonstrated to promote neurogenesis and synaptic plasticity through their miRNA cargoes [18,47,77] (Figure 3). Previous studies by Antonucci et al. (2012) [18] demonstrated that microglia-derived microvesicles play a key role in microglia-to-neuron communication by releasing glutamate, which modulates synaptic activity at presynaptic sites. In animal studies, injecting MGEVs into the visual cortex of rats led to an acute increase in excitatory transmission following visual stimuli that was attributed to enhanced sphingolipid metabolism [18].

A few studies have also demonstrated that synaptic plasticity, essential for learning, memory, and cognitive function, is often impaired in neurodegenerative diseases such as AD and PD [136,137]. MGEVs carry not only miRNAs but also synaptic proteins that are vital for maintaining synaptic integrity and function, such as synapsin, a protein associated with synaptic vesicles involved in neural development, and these MGEVs have been found to be released primarily during periods of either increased neuronal activity or cellular stress [138]. The MGEV-induced production of ceramide and sphingosine can elevate neuronal sphingolipid metabolism, driving increased synaptic activity in recipient neurons and increased excitatory transmission [18,138,139]. This capacity to enhance synaptic plasticity and repair makes MGEVs a promising therapeutic approach for restoring cognitive function in neurodegenerative diseases. These findings underline the significance of MGEVs in regulating synaptic activity and their critical role in microglia-to-neuron signaling pathways [18].

MGEVs from differentially activated microglia can also affect inhibitory synaptic transmission. Prada et al. (2018) showed that MGEVs released from inflammatory microglia led to the loss of excitatory synapses. This effect is associated with the transfer of miR-146a-5p, a microglia-specific miRNA that regulates presynaptic synaptotagmin1 (Syt1) and postsynaptic neuroligin1 (Nlg1), both of which are essential for dendritic spine formation and synaptic stability [47]. Other works have suggested that endocannabinoids, known to influence neuronal synaptic communication in the CNS, are transported via MGEVs. These vesicles carry N-arachidonoylethanolamine (AEA) on their surface, which activates type 1 cannabinoid receptors (CB1s), leading to the inhibition of presynaptic transmission in GABAergic neurons [140]. Further research is needed to determine whether the MGEV modulation of neurotransmission is critical for maintaining normal homeostasis or if it contributes to neuropathological conditions by disrupting the balance between excitatory and inhibitory transmission.

MGEVs play a beneficial role in clearing neurotoxic proteins such as amyloid-β in AD and α-synuclein in PD [69]. While more research is needed to confirm the direct involvement of MGEVs in the phagocytic clearance of harmful protein aggregates, evidence suggests that MGEVs contribute to not only the propagation of Aβ aggregation and neuroinflammation, but also aid in the removal of neurotoxic proteins [141]. In AD, MGEVs have been shown to clear amyloid-β peptides by delivering insulin-degrading enzyme (IDE), which prevents Aβ aggregation and plaque formation [142]. The clearance is regulated by proteins such as Glutaminase C (GAC) and TREM2, which promote exosome release and facilitate the microglial phagocytosis of amyloid-β [143,144]. A protective role for TREM2 in reducing exosomal tau pathogenicity has been proposed, with studies showing that the dysfunction or loss of TREM2 expression leads to an increased spread of tau pathology driven by MGEVs [116]. These findings emphasize the beneficial impact of MGEVs in mitigating neurodegenerative diseases by potentiating the clearance of toxic protein aggregates.

The regenerative capacity of MGEVs has been demonstrated across species from leeches to rodents, where they act as a vehicle for transferring pro-regenerative molecules [50,145]. Evaluation of the cargo content of MGEVs from these different sources has identified neurotrophic factors, including TGF-beta and prosaposin [123,146,147,148], that are capable of inducing neurite outgrowth in dorsal root ganglion (DRG) neurons [62], leech neurons [146], and rat embryonic cortical neurons [123], as well as promoting differentiation in PC12 cells [146]. MGEVs also contain cargoes that modulate key metabolic processes involved in axon growth and guidance, dendrite development, and filopodium assembly. In studies with leech microglia, the neurotrophic effects of MGEVs were partially mediated by nGDF (nervous Growth/Differentiation Factor), which is a member of the TGF-β family [123,146,147,148]. Additionally, the cargo evaluation of MGEVs from the N9 microglial cell line revealed the presence of metabolic enzymes that are associated with anaerobic glycolysis and lactate production [39,54], suggesting that MGEVs may serve as reservoirs of energy substrates that are essential for neurite outgrowth.

MGEVs from pro-regenerative microglia actively support myelin repair, and conversely MGEVs generated from inflammatory microglia have been shown to hinder remyelination. Remyelination failure in the CNS is associated with many neurological disorders, such as multiple sclerosis. Lombardi et al. [145] conducted a comparative study to evaluate the effects of MGEVs from either pro-inflammatory or pro-regenerative microglia on oligodendrocyte progenitor cell (OPC) function in a mouse model of lysolecithin-induced demyelination in the corpus callosum. It was found that MGEVs from microglia with regenerative phenotypes significantly enhanced OPC recruitment, migration, differentiation, and myelin repair, whereas MGEVs from pro-inflammatory microglia hindered remyelination. The beneficial effects of these MGEVs were attributed to their surface lipid component, sphingosine 1-phosphate, which played a key role in promoting myelin repair [145]. MGEVs have also been demonstrated to promote the migration and differentiation of OPCs into oligodendrocytes and contribute to myelin regeneration as well as aid in white matter repair to reverse cognitive impairment [92]. The presence of miR-23a-5p in MGEVs has been identified as a key factor in driving OPC migration, differentiation, and myelination [66]. In other studies, miR-23a-5p in MGEVs from anti-inflammatory microglia was found to promote functional recovery in a mouse model of cerebral ischemia [149]. These findings underscore the crucial functional role of MGEVs in OPC biology and the process of myelination.

Peng et al. (2021) showed that MGEVs can improve function after SCI by inhibiting oxidative stress and promoting normal endothelial cell function. MGEVs were found to attenuate the levels of ROS and the protein NADPH oxidase (NOX2), which produces superoxide (O2•−) radicles. The positive effect of MGEVs post-SCI was demonstrated to occur through the activation of Nrf2/HO-1 signaling via the inhibition of keap1 [100]. Similarly, MGEVs derived from IL4-stimulated microglia have been shown to decrease neuronal apoptosis by reducing oxidative stress. This effect was mediated through the transfer of miR-124-3p, which modulates the ROCK/PTEN/AKT/mTOR signaling pathway to inhibit cell death [150]. Table 1 provides an overview of a number of benefecial properties associated with microglial exosomes highlighted in recent studies.

Under pathological conditions, cytokines and chemokines within pro-inflammatory MGEVs such as IL-1β, TNF-α, IL-6, CCL2, and CXCL10 [155,156] activate recipient astrocytes within their microenvironment and drive them towards a harmful phenotype or recruit peripheral immune cells into the CNS to perpetuate neuroinflammation and exacerbate neuronal damage [16,17,46,145,155]. These MGEVs are released from microglia that are activated in response to an injury or disease and as they can also act upon the microglia that secrete them, the MGEVs can therefore create a vicious feedback loop of inflammation [17,66,157,158]. TNF-α, when present in MGEVs, can serve either a physiological function in neuronal circuit formation and synaptic plasticity to maintain homeostasis, or act pathologically to induce neuroinflammation and ensuing tissue degeneration, cell death, and myelin damage [157,159]. In recent work, it was shown that MGEVs derived from LPS-primed microglia were able to activate resting microglia in a similar fashion to direct LPS, triggering a pro-inflammatory response [40]. Similarly, MGEVs derived from TBI mice (or LPS-treated mice) have been reported to promote microglial reactivity when injected into the brains of naïve mice [160]. In vitro experiments have demonstrated that priming microglia with LPS results in the release of MGEVs loaded with elevated levels of IL-1β and miR-155, which are important pro-inflammatory mediators involved in the progression of neuroinflammation in CNS pathologies [160]. These pro-inflammatory MGEVs have been shown to contribute to numerous neurological conditions such as the worsening of stroke outcomes in aging populations [161], release amyloid-β and tau proteins into the extracellular space, and activate resting microglia and astrocytes to accelerate the exacerbation of inflammation in AD [63]. MGEVs have also been implicated in neuropathic pain following spinal nerve ligation where they trigger the IL-1β-mediated P2X7R-p38 pathway to aggravate chronic pain [162]. In HIV-associated neurological disorders, MGEVs activate the NLRP3 inflammasome, causing synaptodendritic injury and functional impairment in recipient neurons to worsen the disease [159].

In AD and PD, MGEVs are key players in the spread of misfolded proteins, amyloid-β and α-synuclein, among neurons [69,163,164], causing protein aggregation and exacerbating neuronal dysfunction to advance disease progression [69]. In neurological disease models, MGEVs have also been shown to deliver reactive radicals and other neurotoxic molecules to nearby cells within the microenvironment, leading to oxidative stress and neuronal injury [165], exacerbating cell death and synaptic dysfunction. MGEVs can influence endothelial cell function, compromising the integrity of the BBB and causing its permeability to peripheral immune cells [155] as well as other deleterious molecules from the systemic circulation. An investigation in a relapsing–remitting EAE mouse model employed the MS drug FTY720, demonstrating that it reduced EVs in the cerebrospinal fluid (CSF) after disease induction [155].

In addition to pro-inflammatory cytokines and pathogenic misfolded proteins, miRNAs in MGEVs have been recognized for their contribution to neurological disease. Pathogenic MGEVs carry miR-146a-5p, which attenuates excitatory synapses by inhibiting presynaptic synaptotagmin-1 and postsynaptic neuroligin-1, which are essential for dendritic spine stability [47]. Similarly, MGEVS that carry the endocannabinoid N-arachidonoylethanolamine on their surface can influence presynaptic transmission by modulating the endocannabinoid system by activating type 1 cannabinoid receptors on GABAergic neurons, which leads to an inhibition of presynaptic transmission [140].

Growing evidence suggests that MGEVs transport misfolded, neurotoxic proteins and contribute to disease progression. In AD, MGEVs has been shown to facilitate the spread of Aβ peptides, leading to plaque formation and neurotoxicity, accelerating protein aggregation and neuronal dysfunction [17,69,166]. Asai et al. (2015) demonstrated that blocking MGEV release reduced tau pathology in a mouse model [28,167]. Additionally, MGEVs from AD brain tissues carry presenilin, APP fragments, and Aβ-producing proteases, promoting the formation of neurotoxic Aβ aggregates [168]. Similarly, in PD, MGEVs carry α-synuclein, the main component of Lewy bodies, spreading it to healthy neurons and promoting its aggregation, driving disease progression [25,163,164]. These studies emphasize MGEVs’ critical role in disseminating misfolded protein aggregates to nearby recipient cells via the exosome–synaptic pathway, propagating synaptic dysfunction in the CNS [25,169,170]. While MGEVs are vital for cellular communication, their ability to propagate toxic proteins poses a significant risk in advancing neurodegenerative diseases. The contribution of MGEVs to the spread of misfolded copper–zinc superoxide dismutase one (mSOD1) between cells, on the other hand, has been questioned in amyotrophic lateral sclerosis (ALS) [171,172], but direct evidence linking MGEVs to the progression of ALS has not been firmly established. A recent study has provided evidence regarding the potential correlation between pro-inflammatory microglia and dysregulated miRNAs in ALS. It has been found that activated microglia from SOD1G93A mice release certain dysregulated miRNA species such as miR-22-3p, miR-125b-5p, miR-146b-5p, miR-155-5p, miR-214-3p, and miR-365-3p that induce microglial activation and inflammatory genes in ALS [172] and can be taken up by neuronal cells in ALS [26]. Collectively, these studies provide evidence for the role of MGEVs in the transmission and propagation of several neurodegenerative conditions.

MGEVs can directly contribute to neuronal death by delivering neurotoxic substances that trigger apoptosis or necrosis in recipient neurons. MGEVs carrying TNF-α and IL-6 have been shown to induce apoptosis in cortical neurons, playing a significant role in neuronal loss in neurodegeneration models [17]. Furthermore, MGEVs carrying elevated levels of reactive species can cause oxidative stress and mitochondrial dysfunction in neurons, ultimately leading to cell death. In glaucoma, where microglial cells are activated by ocular hypertension, they release EVs containing pro-inflammatory cytokines and reactive oxygen species, which induce retinal cell death [173]. In other studies, exposure to α-synuclein has been shown to activate microglial cells, leading to an increased release of MGEVs containing elevated levels of major histocompatibility complex class II and TNF-α, which contribute to neuronal death [174]. MGEVs from microglia exposed to ethanol have been shown to increase the transmission of apoptotic factors, including complement protein C1q and reactive oxygen species (ROS). These MGEVs can alter the cell membranes of β-endorphin neurons during the postnatal developmental period to induce cell death [175]. Recent research by Arvanitaki et al. (2024) shows that microglia from aged murine brains exhibit a DNA repair defect in XPF-ERCC1, leading to the accumulation of dsDNA fragments in the microglial cytosol. These fragments are packaged into MGEVs, which, when released, promote neuronal death [61]. These studies highlight how the cargo of MGEVs can have direct, deleterious effects on neuron survival, particularly when the MGEVs are derived from microglia in a pro-inflammatory or disease-activated state.

Microglia play a key role in the regulation of synaptic pruning and plasticity under both normal physiological conditions like learning and memory as well as during various neurological pathologies associated with cognitive deficits linked to aging, TBI, AD, and HIV-associated neurocognitive disorder [176]. Recent studies suggest that MGEVs influence synaptic transmission, either to cause inhibition or excitation according to their cargo. MGEVs carrying toxic proteins or pro-inflammatory cytokines can disrupt synaptic integrity and signaling, leading to synaptic loss and dysfunction that contributes to cognitive decline and other neurological deficits observed in neurodegenerative diseases. In AD, MGEVs containing Aβ and inflammatory cytokines have been shown to disrupt synaptic plasticity, leading to impaired learning and memory [169]. Prada et al. (2018) suggest that inflammatory microglia release MGEVs enriched with specific miRNAs, such as miR-146a-5p. These miRNA are transferred to neurons to suppresses key synaptic proteins like presynaptic synaptotagmin1 (Syt1) and postsynaptic neuroligin1 (Nlg1), which leads to dendritic spine loss and reduced excitatory synapse density and strength, implicating MGEVs in synaptic gene silencing [47]. Support for the role of miR-146a comes from various studies that demonstrate alterations in miR-146a expression levels during the development and progression of various neurological diseases that are associated with impaired synaptic functions [47,81,177].

Recent studies by Gabrielli et al. (2022) have reported that large extracellular vesicles released by activated microglia play a pivotal role in initiating synaptic dysfunction in AD. This early synaptic impairment is a critical process that gradually extends to larger brain regions as the disease progresses. The study showed that MGEVs can alter dendritic spine density and disrupt synaptic plasticity both in vitro and in vivo, suggesting that MGEVs may drive the spread of early synaptic dysfunction in AD [169]. Work by Falcicchia et al. (2023) showed that MGEVs contribute to cortico-hippocampal network dysfunction in AD. Chronic EEG recordings revealed that a single injection of β-amyloid-laden MGEVs into the mouse entorhinal cortex was able to induce changes in cortical and hippocampal activity similar to those observed in a mouse model of AD and human AD patients. These EEG abnormalities showed a correlation with a progressive decline in cognitive function, further emphasizing the potential role of MGEVs in the propagation of AD [23]. In conclusion, MGEVs play a crucial role in both the maintenance of synaptic plasticity under normal conditions and the disruption of synaptic function in various neurological disorders. MGEVs, depending on their cargo, can either support or impair synaptic integrity. Further investigations into the role of MGEVs in synaptic regulation would provide new insights into therapeutic strategies for preventing and treating neurological diseases.

Pathogenic MGEVs can have detrimental effects, particularly in the context of remyelination repair in demyelinating pathologies of the CNS. They have been shown to impair the function, differentiation, and maturation of oligodendrocyte precursor cells (OPCs), which are crucial for generating new myelin sheaths around demyelinated axons. MGEVs can hinder remyelination and exacerbate demyelination in conditions such as multiple sclerosis and other neurodegenerative conditions as well as contribute to the persistence of a chronic inflammatory state, further impeding the repair of myelin and worsening disease progression. Recent work has shown that MGEVs from reactive microglia have elevated levels of miR-615-5p [178], an inhibitory microRNA, which is known to bind directly to the 3′UTR region of, and inhibit, myelin regulator factor (MYRF), a transcription factor that is critical for OPC maturation. The overexpression of an miR-615-5p sponge with a precise complementary binding site to miR-615-5p in inflammatory microglia in both EAE and cuprizone-induced models of demyelination was able to ameliorate disease progression and promote remyelination [178]. This suggests that MGEVs from pro-inflammatory microglia may prevent myelination via the miR-615-5p/MYRF axis, identifying miR-615-5p as a potential new therapeutic target in demyelinating conditions. Table 2 highlights the detrimental properties of microglial exosomes reported in recent studies.

MGEVs hold the potential to serve as diagnostic and prognostic biomarkers, showing great promise for the treatment of neurological diseases such as AD, PD, and stroke [193,194,195,196,197,198]. Using MGEV specific markers, exosomes identified in the CNS tissue or peripheral body fluids have been identified to be of microglial origin. For example, in PD, exosomes isolated from the CSF of patients, identified by the microglial marker CD11b, have been found to carry higher levels of oligomeric α-synuclein [25]. Similarly, proteomic and biological profiling discovered that in AD patients, brain-derived MGEVs contain elevated levels of signature proteins such as β-amyloid and tau, along with glial-specific molecules like ANXA5, VGF, GPM6A, and ACTZ. In body fluid samples derived from AD patients, the combination of β-amyloid, tau, and these brain-specific proteins could serve as potential biomarkers [24]. Another study demonstrated that CD11b-positive MGEVs could be obtained from the parietal cortex of cryopreserved human brain tissue of late-stage AD and a comparison with age-matched control cases analyzed using integrated multiomics identified an increase in disease-associated markers like FTH1, TREM2, upregulated levels of tau and synaptic proteins, and elevated cholesterol and reduced DHA-containing lipids as well as specific miRNAs (miR-28-5p, miR-381-3p, miR-651-5p, and miR-188-5p) that are linked to immune and cellular senescence [70].

Recent studies have indicated that MGEVs derived from pro-inflammatory microglia show elevated levels of CD11b⁺/TMEM119⁺/CD14⁺ on their surface [66,194]. Combining these MGEV-specific markers with techniques such as immunoprecipitation and fluorescent nanoparticle tracking, these MGEVs can be detected via liquid biopsy, which will be a future tool for the measurement of microglia activity and the extent of neuroinflammation that accompanies the different neurodegenerative disease courses in various CNS pathologies [199]. Analysis of the dynamic exosome cargo will provide accurate insights into disease progression and treatment response, enabling personalized therapeutic strategies. These longitudinal changes in MGEV biomarkers can reflect the effectiveness of treatments or indicate disease progression, providing real-time feedback for clinical decision-making. Recent studies by Santiago et al. demonstrated state-specific proteomic and transcriptomic signatures of unique proteins and mRNA and microRNA signatures in MGEVs influenced by the activation state of the parent cells [200]. As carriers of diverse biomolecules that mirror the physiological and pathological states of their parent cells, MGEVs are therefore emerging as promising biomarkers for CNS diseases. Continued research efforts are needed to elucidate their role as reliable biomarkers and to overcome technical and clinical challenges for their implementation in routine clinical practice.

Another therapeutic approach focuses on modifying the cargo of MGEVs to enhance their beneficial effects and minimize harmful impacts. By altering the miRNA, protein, or lipid content, exosomes can be tailored to improve therapeutic outcomes. For instance, loading MGEVs with miRNAs that target pro-inflammatory pathways or support neuronal survival can promote recovery, as demonstrated in a study where the administration of miR-17-92 cluster-enriched exosomes derived from human bone marrow mesenchymal stem cells (MSCs), which are known to promote neuronal survival and axonal growth, showed tissue protection, enhanced neuroplasticity, and restored function in rat models of stroke [201] and TBI [202]. Similarly, modulating the parent cells to prevent the incorporation of deleterious proteins or miRNAs, such as toxic tau or α-synuclein, in the MGEVs could help mitigate the progression of neurodegenerative diseases like AD and PD.

Several preclinical studies have found that the polarization of the parent microglial cells from an inflammatory to a neuroprotective or pro-reparative phenotype improves histopathological and functional outcomes in several neurological conditions [203,204]. One example in this regard is that the polarization of parent microglial cells with the anti-inflammatory cytokine IL-4 generated MGEVs that exhibited a reduction in neuronal impairment and mitochondrial dysfunction in rodent models of AD [135]. Similarly, the polarization of parent microglial cells resulted in elevated levels of miR-124-3p in MGEVs, which showed neuroprotective effects by reducing neuroinflammation and contributing to neuronal growth and plasticity after TBI [89] as well as alleviated brain damage and improved the outcome following stroke [203].

Another alternate therapeutic strategy to modulate MGEV cargo would be to involve selective manipulations through the preconditioning of the parent microglia to modulate their cargo to enhance beneficial effects and block pathogenicity. One such approach is preconditioning the parent cells to generate anti-inflammatory MGEVs with therapeutic cargoes. Previous studies with MGEVs derived from hypoxia-preconditioned microglia showed that these exosomes promoted angiogenesis, suppressed apoptosis, and were more effective in promoting neurological recovery after ischemic stroke by activating the TGF-β/Smad2/3 signaling pathway [45]. Similarly, preconditioning macrophages to low-dose endotoxins, such as LPS, resulted in the secretion of MGEVs with an enhanced capacity to confer neuroprotection [205,206] and which improve function after ischemic stroke [207].

Despite the therapeutic potential of MGEVs, several key challenges remain to be addressed prior to their use for clinical translation as a therapy for neurological pathologies. These challenges involve refining MGEV isolation and purification techniques, overcoming the inherent variability in exosome populations, improving the efficiency of targeted delivery, and ensuring long-term safety and effectiveness across various neurological disorders, which are critical steps for clinical translation [208]. Although different MGEV isolation techniques have been adapted in research laboratories including ultracentrifugation, ultrafiltration, size exclusion chromatography, immunoaffinity methods, polymer precipitation, and microfluidics-based methods, each method is associated with limitations [209,210]. It is crucial to have a standardization of isolation and characterization methods to overcome batch-to-batch variation and achieve consistent yield and purity. Similarly, the validation of biomarker specificity is essential and dependent on the availability of robust exosome analytical platforms prior to use in clinics. Additionally, it is critical to thoroughly characterize exosomes, particularly in terms of their size, morphology, concentration, the presence of exosome-enriched markers, the cargo they carry, and most importantly to confirm the absence of deleterious contaminants.

EVs have distinct features, including stability, low immunogenicity, and high biocompatibility, making them excellent platforms for drug delivery [211]. This enables them to evade removal by the immune system and effectively deliver drugs to target cells. Additional features of EVs, such as their capacity to cross the BBB for CNS drug delivery and their ability to co-deliver multiple drugs within the same exosome while at the same time preserving their pharmacological properties, further enhance their suitability as drug carriers [212]. In addition, the loaded cargo within the EVs is protected from degradation by enzymes due to their lipid bilayer structure, thereby enhancing the stability of the packaged contents [213]. Currently, engineering techniques such as surface modification and efficient cargo loading methods are actively being researched to improve therapeutic efficacy while maintaining exosome integrity. The incorporation of therapeutic molecules into EVs can be performed either through donor cell engineering or by direct loading [213]. Designing engineered EVs with surface-modified peptides that can target specific cell populations is anticipated to be an effective therapeutic approach for CNS pathologies.

MGEVs originating from activated microglia are recognized as carriers of pro-inflammatory cytokines and chemokines, which can activate astrocytes, sustain neuroinflammation, and trigger cell death [46,155]. In various neuropathological conditions, MGEVs have been observed to transfer reactive oxygen species (ROS) and other neurotoxic molecules to surrounding cells, resulting in oxidative stress and neuronal damage [165]. In neurodegenerative diseases such as AD and PD, MGEVs have been implicated in the transport of pathological proteins like amyloid-β and α-synuclein, facilitating their transfer between neurons [69,163,164]. This process contributes to neuronal dysfunction and disease progression [69]. Thus, developing innovative strategies for bioengineering MGEVs to carry therapeutic cargo such as siRNA, miRNA, neurotrophic growth factor proteins, or selective antibodies targeting pathogenic proteins are urgently needed to mitigate harmful effects, modulate disease progression, and enhance treatment outcomes. An example of the use of bioengineered MGEVs has been carried out in AD, where abnormal lysosomal function in microglia leads to the inefficient clearance of Aβ aggregates. Hao et al. (2020) designed mannose-conjugated macrophage-derived exosomes (MExo) loaded with gemfibrozil (Gem) as a microglia-targeting drug delivery system. This approach selectively delivers Gem to microglia, restoring lysosomal activity to enhance Aβ clearance and improve cognitive function in Alzheimer’s disease [115]. To improve cell-selective uptake within the CNS, the surface functionalization of MGEVs with targeted peptides, antibodies, or proteins by conjugating them with specific membrane proteins like LAMP-2B, a member of the lysosome-associated membrane protein (LAMP) family that is found in abundance on the exosome surface, can be employed for the display of a targeting motif. This approach can boost therapeutic efficacy by enhancing interaction in a cell-selective manner while minimizing off-target effects [214,215,216].

Although MGEV research is still evolving, it is essential to recognize the therapeutic potential that is revealed by several studies which have demonstrated that MGEVs carry bioactive molecules capable of modulating neural environments. This opens avenues for therapy, especially in the context of neurotrauma and neurodegenerative diseases.

Moreover, because MGEVs mirror the phenotypic state of their microglial cell origin, there is the potential to develop therapies that enhance the release of protective EVs while reducing neurotoxic ones. Utilizing MGEVs as a therapeutic approach could also modulate key signaling pathways necessary for maintaining microglial balance and homeostasis. However, precisely controlling EV interactions to achieve targeted modulation is challenging, as the nuanced responses of homeostatic microglia to EV signals can lead to unpredictable outcomes, complicating the design of specific therapies. Additionally, MGEVs are only one part of a plethora of responses elicited by microglia, with other effects produced through the direct secretion of molecules (cytokines, trophic factors) and physical interactions such as the phagocytosis of debris, apoptotic cells, or foreign materials, among other activities. While MGEVs may contain several essential molecules, they do not contain the appropriate machinery like the parent cell to respond to external effectors and ligands, which requires specific signaling as well as the direct physical interaction of the microglia with other cell types. Therefore, MGEVs are only one part, albeit an important part, of the repertoire of microglia activities during disease or repair.

It is also important to take into consideration that the effective use of exosomes also lies in achieving targeted delivery, where exosome uptake occurs in a highly cell-selective manner, ensuring only specific cell types internalize the therapeutic cargo. This specificity is crucial to avoid off-target effects and enhance therapeutic efficacy. Although engineering MGEVs for selective targeting remains complex and requires further development, it nevertheless offers significant promise for advancing CNS repair and resilience therapies.

As of now, the FDA (U.S. Food and Drug Administration) has not approved any exosome-based therapies for clinical use. Exosomes from disparate parent cells are under investigation in various preclinical and clinical studies for their potential in treating a range of conditions, including cancer, neurodegenerative diseases, and other disorders. In recent years, exosomes have been explored in preclinical and clinical studies for various neurodegenerative conditions such as AD, PD, and stroke [217]. Several studies in experimental models of CNS injury and disease have demonstrated the therapeutic benefits of Schwann cell (SC)-, mesenchymal stem cell (MSC)-derived, and neural stem cell (NSC)-derived EVs [218,219,220,221,222], with more recent research highlighting the potential of the repeated administration of allogeneic Schwann cell-derived EVs in an FDA-approved single patient study with ALS that did not exhibit any adverse effects towards SCEV administration [223]. However, to date, no clinical trials have been conducted for MGEVs. To successfully translate the preclinical successes of modulated MGEVs into clinical studies, it will be necessary to gain a deeper understanding of consistent therapeutic outcomes, pharmacokinetics, biodistribution, and long-term effects in humans, particularly to address regulatory and safety challenges (Figure 5).