miRNAs are single-stranded non-coding RNAs of 19–25 nucleotides with regulatory functions found in eukaryotes (Lu and Rothenberg, 2018). In the nucleus, the biogenesis of miRNA begins with the transcription of miRNA genes into large initial transcripts (pri-miRNA), which then produce a molecule of approximately 70 nucleotides with a stem-loop structure (pre-miRNA) under the shearing action of RNase III endonuclease Drosha and the RNA-binding protein DGCR8 (Brate et al., 2018; Dexheimer and Cochella, 2020). In the cytoplasm, pre-miRNAs are cleaved into mature miRNAs by the RNase III enzyme Dicer and then exported from the nucleus. With the help of helicase, mature miRNAs can shape the RNA-induced silencing complex (RISC), and complementary mRNA sites regulate gene expression via base pairing (Higa et al., 2014). As the smallest nucleic acid, exosome-derived miRNAs in the nervous system constitute a complex network for carrying genetic information and regulating gene expression, indicating that exosomal miRNAs play a key role in the regulation of the occurrence, development, and spread of many human CNS diseases (Lukiw and Pogue, 2020).

Although miRNA transcripts are small, some ncRNAs can surpass 200 nucleotides in length, and they are therefore cataloged as lncRNAs (Derrien et al., 2012; Wu and Kuo, 2020). lncRNAs can exist in both the nucleus and cytoplasm and play different functions according to their subcellular localization (Statello et al., 2021). In the nucleus, lncRNAs are involved in transcriptional regulation of gene expression and mRNA splicing. In the cytoplasm, such as in exosomes, they can affect the stability of mRNA and regulate the biological functions of proteins (Zhu et al., 2013). In addition, lncRNAs share some comparable features with mRNAs: they undergo posttranscriptional modifications, such as 5’-capping, polyadenylation, and alternative splicing (Ponting et al., 2009). The other way around, particularities of numerous lncRNAs are not found in mRNAs, such as cis-regulatory capacity, lacking robustly translated open reading frames, special 3’-terminal processing, templating of nucleic acid polymerization or assembly, and other divergences (Quinn and Chang, 2016). Concurrently, the growing ranks have motivated an increased focus on understanding the roles of exosomal lncRNAs in biology, steadily revealing that exosomal lncRNAs preside over the occurrence of many neurodegenerative diseases and the damage processes of the CNS (Szilagyi et al., 2020).

circRNA is a naturally occurring class of ncRNA molecules that have become a current research focus in the field of RNA (Li et al., 2015). Unlike conventional linear RNA, circRNA has a covalently closed continuous loop and is resistant to RNA exonuclease, so its expression level is more stable, and it is not easily degraded (Qu et al., 2015). In fact, hundreds of circRNAs are enriched in human brain tissues and have distinctive regulatory potency (Rybak-Wolf et al., 2015). Recent research has discovered that exosomal circRNAs can function as miRNA sponges, protein inhibition, and regulators of splicing and transcription; and they also can bind RBP, as well as translate into peptides and proteins, which contribute to neurodegenerative diseases up to a point (Lukiw, 2013; Fanale et al., 2018; Welden et al., 2018). These rising studies indicate that such a circRNA-miRNA competing system is a non-negligible epigenetic regulatory aspect that controls gene expression in some CNS diseases.

Although research on exosomal circRNA is still at a premier period, an emerging number of studies suggest that circRNAs by virtue of exosomes institute their circulation and communicate with the recipient cells. Thus, carrying out exosomal circRNA abundant biological functions (Wang Y. et al., 2019).

The piRNA that is 26–31 nucleotides in length is small non-coding RNA with 2’-O-methylation at their 3’ ends (Pippadpally and Venkatesh, 2020). In contrast to other ncRNAs, which are double-stranded derived RNAs that are dicer-dependent and bind to Argonaute proteins, piRNAs are transcribed from ssRNAs, dicer-independent and bound to PIWI proteins (Aravin et al., 2006). In contrast to miRNAs, piRNAs could silence genes, participate in various aspects, including transposon silencing, transcriptional silencing or activation, posttranscriptional regulation, and other modifications, further elucidating the uniqueness of piRNAs compared with other ncRNAs (Rajasethupathy et al., 2012; Ozata et al., 2019; Wang and Lin, 2021). piRNA biogenesis is unique and complex and is marked by the noncanonical transcription of precursor molecules and self-augmentation mechanisms. Similar to miRNAs and lncRNAs, piRNAs have become a contemporary research hotspot in the field of exosomal ncRNA and could be widely involved in the processes of life.

Interestingly, exosome might be a double-edged sword for ncRNAs (Wang Y. et al., 2019; Cheng et al., 2020). One is that exosome is protected by double-layer membrane wrapping that does help to secure the contents of exosomes from external interference and facilitate diffusion between neurons. Furthermore, in comparison with ncRNAs which exist outside exosomes, exosomal ncRNAs are protected from enzyme degradation and could be more easily delivered to produce targeted biological effects (Liu et al., 2022). On the other hand, exosome could reduce the accumulation of ncRNAs and help to clear ncRNAs which means ncRNAs could be taken up by other specific cells, such as macrophages.

The appearance of complicated diseases is the result of the synergism of multiple interacting genes or RNAs (Zhang et al., 2018). Hence, we should investigate disease mechanisms at the level of cell biology and unearth effective information from a vast network of interacting genes or RNAs.

In 2017, Sarkar observed that the increased miR-34a expression in a specific brain region was related to the severity of AD pathology (Sarkar et al., 2016). They further found that miR-34a targeted dozens of many genes that might result in the dysfunction of synaptic plasticity, energy metabolism, and resting state network activity. Their findings implied that up-regulated miR-34a influenced neuron stability, which might be a potential mechanism of AD biogenesis. Furthermore, a research group demonstrated that miRNA-193b was a regulator of APP derived from blood and CSF (Liu et al., 2014). In samples from APP/PS1 double-transgenic mice, mild cognitive impairment (MCI) and dementia of Alzheimer-type (DAT) patients, this study found that miR-193b overexpression could repress the mRNA and APP protein expression that would influence Aβ generation in brain (Liu et al., 2014). Another interesting study reported that in primary neuronal culture and AD patients’ brain, the miR-29a/b-1 cluster was significantly downregulated, and β-site amyloid precursor protein cleaving the enzyme-1 (BACE1) expression was increased (Hébert et al., 2008). Soon afterward, Seongju Lee found that both miR-29a and miR-29b-1 could bind to the 3’ -UTR of BACE1, and the BACE1 levels were negatively correlated with the expression levels of these miRNAs (Lee et al., 2019). The integration of the information above reveals that miR-125-5p, miR-193b, and miR-29a/b1 could enter cells via exosomes, then influence Aβ production and ultimately lead to the amyloid plaque deposition. Above all, exosomal miRNAs are involved in many pathways that contribute to the biogenesis of AD, such as neuroplasticity, neuron network activity, accumulation of APP and Aβ, and expression, phosphorylation, and aggregation of tau.

In addition to miRNAs, many exosomal lncRNAs are also involved in the pathological process of AD. The expression of lncRNA BACE1-AS (β-site amyloid precursor protein cleaving the enzyme-1-antisense transcript) was significantly increased compared with that in healthy controls (Wang D. et al., 2020). BACE1-AS is the opposite strand to BACE1 and serves as a suppressor of the production of BACE1 (Luo and Chen, 2016). It is suggested that lncRNA BACE1-AS may affect the production of Aβ and participate in the pathology of AD. Moreover, multiple research groups demonstrated that the expression levels of lncRNAs 17A, 51A, and BC200 in plasma exosomes were elevated in certain brain regions in AD patients (Andersen et al., 2005; Cortini et al., 2019; Wang D. et al., 2020). Separately, 51A was found located on the antisense strand of the first intron of the sortilin-related receptor 1 (SORL1), impairing the APP cleavage and promoting neurotoxic Aβ formation (Weiner, 2013). Additionally, lncRNA17A might participate in neuroinflammation, and the binding between BC200 and eukaryotic initiation factor 4A (eIF4A; Ames et al., 2017; Guo et al., 2019). The function of exosomal lncRNA17A results in decoupling ATP hydrolysis from RNA duplex unwinding that preserved long-term neuronal plasticity. All of these studies indicated that BACE1-AS, 51A, 17A, and BC200 might serve as direct or indirect roles in AD pathogenesis and novel therapeutic strategies.

In addition to miRNAs and lncRNAs discussed above, circRNAs may also play a role in the balance of miRNA networks; and the function of exosomal circRNA contributes to Aβ fibrils, oxidative stress, and progressive cognitive deficits, which is regarded as one potential mechanism of AD. For example, circRNA for miRNA-7 (ciRS-7) competes with anti-complementary miRNA complementary content and adsorbs miRNA-7. Hence ciRS-7 quenches the normal function of miRNA-7 and plays an important post-transcriptional regulator of human brain gene expression (Lukiw, 2013). Zhang et al. (2019) further found that circRNA KIAA1586 occurred frequently in AD risk ncRNAs. As a competing endogenous circRNA, its function was to competitively bind three known AD-risk miRNAs (miR-29b, miR-15a, and miR-101) to silence target genes and display the over-expression of Aβ peptide. As mentioned above, exosomal circRNAs serve as the pathological mediators of AD, which is a rapidly growing field of potential biomarkers, though more detailed studies are needed.

Except for the many mechanisms that ncRNAs participate in, one vital pathological characteristic of AD is the stability of genes, which affect the expression of key proteins. Some studies had found that piRNAs take part in gene stability. The function of piRNAs is to silence repetitive genomic regions to mediate genomic stability; and some piRNAs were believed to play an active role in gene expression control, which was related to long-term memory (Landry et al., 2013). In 2019, Jain et al. (2019) first reported that piRNAs are differentially expressed in human CSF exosomes of AD dementia patients. Moreover, exosomal piRNAs might contribute to the pathogenesis of neuropsychiatric illnesses. For AD, piRNAs might influence the expression of APOE in the brain (a key protein-coding gene in AD), suggesting that piRNAs might be involved in the etiological processes of AD (Qiu et al., 2017). Jain demonstrated that piRNAs were significantly correlated with Tau protein levels or the Aβ42/40 ratio, indicating that these piRNAs could reflect changes in Tau protein or amyloid pathology (Jain et al., 2019).

In short, these exosomal ncRNAs have the same distinctive characteristics to participate in AD pathogenesis. Examples include APP and Aβ generation in the nervous system, consecutive hyperphosphorylation of tau, synaptic loss, oxidative stress, and neuronal death (Lee et al., 2019). In conclusion, these relevant studies reveal multiple functions of exosomal ncRNAs in the pathological process of AD and illustrate novel therapeutic targets for AD treatment.

Studies had shown that miRNAs in serum were encapsulated within exosomes, which were secreted by pathological tissues and quite stable (Andersen et al., 2014). More specifically, it had been documented that exosomal miRNAs played a significant role in the regulation of the list of PD-related pathogenic proteins, such as α-SYN, leucine-rich repeat kinase, Parkin, DJ-1/PARK7, and phosphatase (Gui et al., 2015). These biological proteins regulated many key pathways, including protein aggregation, autophagy, inflammation, and hypoxia (Harischandra et al., 2018). In 2019, a research group found that exosomal miR-137 was upregulated and played a vital role in the induction of oxidative stress injury in neurons (Jiang et al., 2019). Jiang further elaborated that miR-137 was found to directly target oxidation resistance 1 (OXR1) and negatively regulate its expression, thus inducing oxidative stress damage in dopaminergic neurons (Billia et al., 2013). The DJ-1 protein is encoded by the PARK7 gene and represents an important protein involved in autosomal recessive primary PD. Chen found that upregulated expression of miR-4639-5p in PD patients, which negatively regulated the post-transcription level of the DJ-1 gene and eventually caused severe oxidative stress and neuronal death (Chen et al., 2017a). The information mentioned above implies that exosomal miR-137-mediated and miR-4639-5p-induced oxidative stress may serve as a potential mechanism in PD pathology. Moreover, miR-325 was shown to be a complementary fragment with a caspase recruit domain to suppress the apoptosis repressor (an anti-autophagic protein), and thus could promote the autophagic cascade (Bo et al., 2014). Coincidentally, a study discovered that Mn exposure significantly increased miR-325 in exosomes, which suggested that exosomal miR-325 participated in dopaminergic neuron autophagy (Harischandra et al., 2018). Although these exosomal miRNAs were found to be able to target genes involved in many vital pathways for PD, some doubts and inexplicable puzzles remain. For instance, Doxakis found that miR-153 directly interacted with the specified regions of the α-SYN gene to inhibit translation from the chimeric transcript, and their effect was cumulative (Doxakis, 2010). Subsequently, a fascinating study reported that the expression of miR-153 was significantly up-regulated in exosomes from the CSF of PD patients compared with healthy controls (Gui et al., 2015). These self-contradictory reports demonstrated the complexity of PD and the immaturity of current techniques. In a word, these exosomal miRNAs can promote our understanding of the mechanisms of PD etiopathology and development. Although, these researches may facilitate the development of new strategies for the diagnosis and therapy of PD, they remind us that more specific studies are needed.

In addition to the classic exosomal miRNAs discussed above, many lncRNAs derived from exosomes are also involved in PD pathogenesis. In 2020, Zou et al. (2020) reported that exosomal linc-POU3F3 activity might shed light on the autophagic-lysosomal system in the pathogenesis of PD. They further applied the lncRNA microarray to detect the expression levels of various exosomal lncRNAs. Eventually, the difference in exosomal lncRNA expression was confirmed, and linc-POU3F3 presented the highest fold change value and the most stable detection density and it was therefore selected as the candidate lncRNA in the pathophysiological process of PD. In the same year, exosomal lnc-MKRN2-42:1 was verified to be related to the pathophysiology of PD by Wang Q. et al. (2020). Through bioinformatics analyses, they concluded that lnc-MKRN2-42:1 could regulate target genes such as EIF4E, BTD, MKNK1, and METTL5 and participated in functions, which correlated with apoptosis, synaptic remodeling, long-term potential, immunity, and glutamate neurotransmitter metabolism. Moreover, Elkouris showed that a mass of lncRNA genes was near to PD-linked protein-coding genes, and four of them were located in exosomes derived from human CSF (Elkouris et al., 2019). They demonstrated that exosomal lncRNAs participated in the direct differentiation of human iPSCs to dopaminergic neurons. Collectively, their data suggested that exosomal lncRNAs were associated with PD pathogenesis.

On the one hand, exosomes may act as vectors to facilitate ncRNA transmission between neurons. Hence, exosomal ncRNA takes part in PD biogenesis and is an emerging therapeutic target. However, further studies will be needed to investigate whether the expression levels of these exosomal ncRNAs would reflect their total brain levels and their more detailed and specific functional associations with PD pathology.

Guo and colleagues found that hypoxia-induced exosomal miR-29a and miR-92a expression leads to the propagation of MDSCs (myeloid-derived suppressor cells), which play vital roles in mediating the formation of an immunosuppressive environment and helping CNS tumors escape the host immune response (Guo et al., 2019). Mechanistically, miR-29a and miR-92a activate the proliferation and function of MDSCs by targeting high-mobility group box transcription factor 1 (Hbp1) and protein kinase cAMP-dependent type I regulatory subunit alpha (Prkar1a), respectively. Altogether, the results of the Guo study provided new insights into the role of glioma exosomal miRNAs in mediating the formation of immunosuppressive microenvironments in tumors and elucidated the underlying exosomal miR-29a/miR-92a-based regulatory mechanism.

Tumorigenesis significantly intensifies the intratumor microvessel density, and angiogenesis is critical for tumor growth, migration and invasion (Ames et al., 2017). Exosomal ncRNAs can have various effects on angiogenesis. First, miR-9 could be secreted from glioma cells via exosomes and was frequently upregulated in glioma specimens (Chen X. et al., 2019). Exosomal miR-9 could significantly enhance the proliferation, migration and invasion of glioma cells, promoting tumorigenesis and an increase in angiogenesis by targeting many malignant genes in glioma cells. Second, the levels of lnc-POU3F3 were upregulated in glioma tissue and significantly correlated with the advanced tumor stage (Lang et al., 2017). That study explored the mechanism by which glioma-derived exosomes affect angiogenesis in more detail. They both used A172-Exos and shA172-Exos to assay the ability of HBMECs human brain microvascular endothelial cells (HBMECs), while HBMECs rapidly internalized A172-Exos and shA172-Exos, and the expression level of linc-POU3F3 in A172-Exos was significantly higher than that in shA172Exos. Furthermore, the gene and protein expression levels of bFGF, bFGFR, VEGFA, and Angio in HBMECs treated with A172-Exos were much higher than those in HBMECs treated with shA172-Exos. These results indicated that gliomas could induce angiogenesis by secreting exosomes enriched in lnc-POU3F3. Exosomal miR-9 and lncRNA-POU3F3 function as promoters of angiogenesis, which is pivotal for glioma pathogenesis and a promising therapeutic target in glioma.

It has been documented that ectopic proliferation is the most basic characteristic of tumors, while invasion and metastasis are the most striking features of tumors (Cheng et al., 2020). Multiple research groups manifested that the expression of exosomal ncRNA is abnormal and that ncRNA has an important impact on tumor proliferation, invasion, and metastasis. In 2015, Zhang found that primary tumor cells with normal expression of PTEN (an important tumor suppressor) but lost PTEN expression after dissemination to the brain (Zhang et al., 2015). Interestingly, brain microenvironment-dependent, reversible PTEN mRNA and protein down-regulation are epigenetically regulated by exosomal miRNAs from astrocytes. Mechanistically, astrocyte-derived exosomes mediated the intercellular transfer of PTEN-targeting miRNAs, leading to increased secretion of cytokine chemokine (C-C motif) ligand 2 (CCL2), which reciprocally enhanced the outgrowth of brain metastatic tumor cells via promoting proliferation and reducing apoptosis. Thus, it can be seen that exosomal miRNA induced PTEN loss, which contributed to primary brain metastasis outgrowth.

In addition to miRNAs, exosomal lncRNAs also unwilling to lag behind that also play a role in the pathogenesis of CNS tumors. In glioma tissues, lncRNA ROR1-AS1 was upregulated and packaged into exosomes derived from tumor cells (Chai et al., 2020). Functional analysis showed that it acted as a sponge of miR-4686 and inhibited its expression to bring about the promotion of glioma development. It is not hard to see exosomal lncRNA ROR1-AS1 derived from tumor cells promoted glioma progression via the miR-4686 axis, and the high expression of ROR1AS1 indicated a poor prognosis in glioma patients. Another type of exosomal ncRNA, circRNA, can also promote glioma progression. Exosomal circRNA 0001445 was taken up and upregulated in glioma cells treated with exosomes (Jitsukawa et al., 2017). In addition, exosomal circRNA 0001445 acted as a sponge for miRNA-127-5p to upregulate the expression of sorting nexin 5 (SNX5), which is a critical regulator in cancers (Han et al., 2021). Taking the information mentioned above, these studies provide an original understanding of the molecular biogenesis of CNS tumor progression, suggesting a participant and therapeutic target role of exosomal ncRNA in CNS tumor patients.

Exosomal ncRNA not only has a beneficial influence on tumor proliferation, invasion, and metastasis but also acts as an inhibitor for drug resistance, tumor progression and invasion (Zhou et al., 2022); from the latter aspect, making exosomal ncRNA is a promising therapeutic tool for CNS tumors. For example, Yao recently selected miR-15a and miR-92a as candidates for further studies and confirmed that they were underexpressed in M2 macrophage exosomes (Yao et al., 2021). The results of target gene validation revealed that miR-15a and miR-92a were bound to CCND1 and RAP1B, respectively. Thus, interference with the expression of CCND1 or RAP1B reduced the phosphorylation levels of AKT and mTOR, indicating that both CCND1 and RAP1B could activate the PI3K/AKT/mTOR signaling pathway. Another study found that mesenchymal stem cell-derived exosomal miRNA-133b suppressed glioma progression via the Wnt/β-catenin signaling pathway by targeting EZH2 (Xu et al., 2019). More detailed particulars are that MSC-derived exosomal miR-133b was found to target and negatively regulate EZH2 expression, and EZH2 silencing resulted in inhibited glioma cell proliferation, invasion, and migration. Therefore, these existing results suggested that exosomal miR-133b could attenuate glioma development by disrupting the Wnt/β-catenin signaling pathway and inhibiting EZH2. Similarly, Yue et al. (2019) showed that the Wnt/β-catenin pathway was also triggered by exosomal miR-301a, which resulted in the suppression of TCEAL7 (tumor suppressor); thus, exosomal miR-301a contributed to glioblastoma resistance to radiotherapy.

These studies unveiled that exosome is an important carrier for tumor cell communication and that exosomal ncRNAs play suppressor roles in CNS tumor and are potential biomarkers and therapeutic targets in different types of cancer.

Before-mentioned studies summarized the exosomal ncRNAs that have been implicated in the pathogenesis, immune escape, angiogenesis, and treatment of CNS tumors. These intriguing findings suggest that we must broaden our horizons to identify molecular mechanisms and increase our knowledge in the field of CNS tumor pathogenesis.

In 2018, Kimura found that the expression of transfection of exosomal miRNA-let-7i was upregulated and let-7i could inhibit the production of Treg cells, and the effect of MS exosomes was disabled when T cells were incubated with let-7i inhibitor before culture (Kimura et al., 2018). Further analysis revealed that exosomal miRNA-let-7i decreased the expression of insulin-like growth factor 1 receptor (IGF1R) and transformed growth factor β receptor 1 (TGFBR1) on immature CD4+ T cells and then inhibited their differentiation into Treg cells. Compared with healthy controls, in relapsing-remitting MS (RRMS) patients, significantly increased expression of miR-326 in exosomes was observed (Azimi et al., 2019). Junker et al. showed that miR-326 targets CD47 in brain resident cells. In turn, the CD47 molecule inhibits the phagocytic activity of macrophages, consequently reducing its expression, which could increase the degradation of myelin. Similarly, Du and colleagues expounded that miR-326, by targeting Ets-1, a negative regulator of Th17 differentiation, could induce the differentiation of mature T cells into Th17 cells, thereby increasing the severity of MS (Du et al., 2009). Because exosomal miR-326 contributes to the pathogenesis of MS, it is possible that inhibition of miR-326 expression in T-cell-derived exosomes or engineering them to carry selected miRNAs may be considered a promising therapeutic approach for the treatment of MS and may also be a potential clinical target in the course of MS (Azimi et al., 2019). Moreover, gray matter demyelination is increasingly recognized as an important component of MS (Geurts et al., 2007). Subsequently, Pusic found that serum-derived exosomes of young and environmentally enriched patients had high levels of miR-219 (Pusic and Kraig, 2014). They further confirmed that exosomal miR-219 was essential and sufficient for myelinating oligodendrocyte production by depressing the expression of inhibitory regulators of differentiation. As a further confirmation of miR-219 function, they inhibited miR-219 in young serum-derived exosomes and found that the protein transcript levels of these miR-219 target mRNAs decreased, resulting in these altered exosomes no longer increasing myelination in slice culture. In conclusion, exosomal miRNAs have been verified to regulate both immune responses and myelination, making them attractive candidates both for pharmacological intervention and as disease biomarkers (Mycko and Baranzini, 2020).

Emerging evidence has indicated the potential role of ncRNAs in the regulation of gene expression of MS pathogenesis, which provides new opportunities to better understand the course of MS (Yang X. et al., 2018). For instance, lncRNA GAS5 was found to suppress the transcription of TRF4, a key factor controlling M2 polarization, which is a pivotal feature of MS pathogenesis (Sun et al., 2017). Another group discovered that circRNA affected the protein coding transcripts, which elucidated a pathway directly linked with STAT3, a critical factor for the inflammatory demyelination and polarization of the immune response toward Th17 (Zurawska et al., 2019). There have been some reports on the role of ncRNAs in the pathological process of MS, but the association between ncRNAs and exosomes and the effects of their interaction on MS remain to be studied.