2. miRNA Involvement in Parkinson’s Disease
2.1. Human and Animal Studies
Cells, either within the same tissue or in different tissues/organs, can communicate across a long distance by sending information from one cell to another to coordinate their behaviours in order to grow, develop and survive
[19]. Recent studies proposed that miRNAs contribute to cell-to-cell communication, by being secreted and transported to other cells via circulation to affect recipient cells
[19]. This hypothesis has been validated by the detection of extracellular/circulating miRNAs in a multitude of biological fluids, such as blood, CSF, saliva, breast milk, urine, and others
[20]. miRNAs can be found circulating in vesicles (exosomes, microvesicles, apoptotic bodies) or associated with proteins (AGO2), and, contrary to cellular RNA, extracellular miRNAs present high stability, providing a desirable characteristic for long-distance cellular communication
[21]. It is now accepted that extracellular/circulating miRNAs can be used as biomarkers and as a therapeutic approach for a wide range of diseases but are also important in cell-to-cell communication.
In the last few years, miRNA dysregulation has been implicated in several neurodegenerative diseases, including PD, where it contributes to neurodegeneration and disease progression
[21].
The importance of miRNAs for CNS integrity has been demonstrated by inducing a selective depletion of Dicer in midbrain dopaminergic neurons in mice, which impairs miRNA biogenesis and results in neurodegeneration and locomotor symptoms mimicking PD
[22]. Furthermore, a multitude of screening studies have reported differentially expressed miRNAs in the brain
[23][24][25][26][27][28], CSF
[29][30][31][32], and blood
[33][34][35][36][37][38][39][40][41][42] of PD individuals. In PD, some miRNAs have been associated with mitochondrial dysfunction, neuroinflammation, and dopaminergic neuron demise, thereby worsening disease pathogenesis. Moreover, several studies have demonstrated that specific miRNAs regulate PD-related genes, such as
SNCA,
PRKN,
DJ-1, and
LRRK2, modulating their functions in different cellular and animal PD models
[43]. It has been observed that overexpression of miR-494 significantly decreased the levels of DJ-1 both in vitro, in 3T3-L1 and Neuro-2a cell lines, rendering cells more susceptible to oxidative stress, and in vivo, in a MPTP mouse model, exacerbating MPTP-induced neurodegeneration
[44]. On the other hand, miR-7 was shown to exert a protective role by repressing expression of α-synuclein and accelerating the clearance of α-synuclein and its aggregates through autophagy
[45]. The reduced levels of miR-7 in the SNpc of a MPTP PD mouse model correlated to nigrostriatal neurodegeneration and α-synuclein upregulation
[46]. A protective role for miR-153 in PD was also hypothesised, since its overexpression decreased MPP
+-induced neurotoxicity in murine cortical neurons
[47]. Kabaria and co-workers observed that miR-34b and miR-34c bind to
SNCA mRNA and reduce α-synuclein expression. In contrast, downregulation of these miRNAs led to increased α-synuclein levels and formation of α-synuclein aggregates. The same group also detected a polymorphism in the binding site of miR-34b, interfering with its binding and consequently leading to α-synuclein overexpression
[48]. Both miR-34b and miR-34c have been found to be downregulated in PD brains
[28]. In vitro studies in differentiated SH-SY5Y cells concluded that depletion of miR-34 b/c resulted in cell death associated with mitochondrial dysfunction and oxidative stress
[28].
LRRK2 gain-of-function mutations have been implicated in both genetic and sporadic PD
[49][50]. Accordingly, LRRK2 levels were found to be increased in PD patients’ brains compared with healthy controls; however, no significant differences in LRRK2 transcripts were detected between both groups, suggesting a post-transcriptional modification of LRRK2 protein expression. In order to understand the reason behind this, Cho and colleagues proceeded to analyse the 3’UTR of
LRRK2 and found a binding site for miR-205, a miRNA found to be downregulated in the brain of PD patients. Further studies confirmed that miR-205 upregulation in cell lines and primary neuron cultures resulted in the downregulation of LRRK2, possibly having a protective effect in the brains of PD patients
[24]. Additionally, mutated LRRK2 seems to be able to negatively regulate miRNA-mediated translational repression in
Drosophila melanogaster brains. Briefly, mutated LRRK2 inhibits the expression of two miRNAs (let-7 and miR-184) known to target E2F transcription factor 1 and dimerization partner transcription factor, leading to defective cell division and neuronal death. These results indicate that mutated LRRK2 might also play a key role in PD pathogenesis by modulating the miRNA pathway
[51]. On the other hand, miRNAs can directly or indirectly modulate the expression of PD-related genes. In fact, a recent study found that miR-421 targets Pink1. Mice treated with MPTP and SH-SY5Y cells treated with MPP
+ were shown to overexpress miR-421. Downregulation of miR-421 attenuated neurodegeneration in MPTP-treated mice and promoted mitophagy in MPP
+-treated SH-SY5Y cells, highlighting the role of miR-421 in regulating mitophagy via the Pink1/Parkin pathway
[52]. Moreover, Zeng and colleagues highlighted the protective role of miR-135b, since it inhibits pyroptosis by targeting FoxO1 in MPP
+-treated SH-SY5Y and PC-12 cells
[53]. A report from 2018 showed that miR-494-3p negatively regulates sirtuin 3 (SIRT3) expression in both MPP
+-treated SH-SY5Y cells and in a MPTP-induced PD mouse model, worsening motor impairment of these mice
[54]. Another study revealed that miR-486-3p targets sirtuin 2 (SIRT2) and reduces its expression levels. However, the authors reported a PD risk-conferring polymorphism in the
SIRT2 gene and showed that a single nucleotide polymorphism (SNP) in this gene alters the binding efficiency of miR-486-3p to
SIRT2, thereby increasing the expression level of SIRT2, which could increase α-synuclein aggregation and toxicity
[55]. Furthermore, miRNAs have been shown to target several genes involved in neuroinflammation, a major hallmark of PD. Specifically, miR-155 was demonstrated to have an important role in the regulation of inflammation. The mentioned miRNA was found to be upregulated in a PD mouse model overexpressing α-synuclein. Deficiency of miR-155 prevented the increase in MHCII and the death of dopaminergic neurons triggered by α-synuclein overexpression. Oppositely, upregulation of miR-155 restored the inflammatory response to α-synuclein fibrils
[56]. miR-7, in addition to having a role in the regulation of α-synuclein as previously described, emerged in the context of neuroinflammation as directly targeting the NLRP3 inflammasome gene. In fact, a study from 2016 determined that transfection of BV2 cells with miR-7 inhibited microglial NLRP3 inflammasome activation, while anti-miR-7 had the opposite effect, aggravating inflammasome activation in vitro. The same group reported that injecting miR-7 mimics directly into the striatum of a MPTP mouse model of PD suppressed NLRP3 inflammasome activation and ameliorated dopaminergic neuronal death
[57]. Studies demonstrated the regulation of TNF-α levels by miR-7116-5p in a microglial cell culture model, where MPP
+ potentiated TNF-α production by downregulating miR-7116-5p. Consistently, overexpressing miR-7116-5p in the microglia of an MPTP mouse model prevented the overproduction of TNF-α and the activation of glia, further reducing the loss of dopaminergic neurons
[58]. Although research on this topic is still in its preliminary stages, and future studies are needed to better understand the role of miRNAs in neuroinflammation, these findings might help identify new therapeutic targets to downregulate microglial activation and potentially diminish dopaminergic neuron death in PD. Importantly, other non-coding RNAs may be involved in PD development progression. As an example, circular RNAs can function as miRNA sponges and act as competitive endogenous RNA to deregulate mRNA by miRNA. For instance, CircSLC8A1, which is increased in the SN of PD patients, regulates miR-128 function and/or activity affecting oxidative stress
[59]. In addition, a recent study showed that transfer RNA fragment profiles, which were hypothesised to function as microRNAs, revealed disease-specific patterns in the CSF and blood of PD patients
[60].
2.2. Gut Microbiota and microRNAs
Given the important involvement of gut dysbiosis and inflammation in PD and the potential of miRNAs to serve as diagnostic biomarkers, Kurz and colleagues
[61] investigated the expression of miRNAs in routine colonic biopsies from PD patients and detected several differentially expressed miRNAs; amongst all, miR-486–5p upregulation showed the highest specificity for PD and correlated with age and disease severity in PD. In a follow-up analysis, 301 target genes of miR-486–5p were identified, as well as the biological processes affected by the mentioned miRNA, with brain development and post-synapse organization processes having the strongest functional association with the miR-486-5p target gene network
[61].
The gut microbiota includes several microorganisms such as bacteria, viruses, protozoans, and fungi. Gut microbiota has a crucial role in the maintenance of gut homeostasis and integrity, since it interacts with the intestinal epithelial barrier (IEB) and intestinal epithelial cells
[62]. Moreover, gut microbiota regulates intestinal epithelium growth, differentiation, and permeability
[62]. Therefore, it comes as no surprise that alterations in gut microbiota composition can disrupt the IEB and lead to gut dysbiosis, with consequent immune and inflammatory response activation. Specific bacterial products (such as short-chain fatty acids (SCFAs), vitamins, or neurotransmitters) can interfere with the regulation of CNS immune and inflammatory processes, including microglial activity, by infiltrating into the bloodstream and traveling to the brain
[63]. Additionally, they have the ability to directly activate circulating immune cells, which then travel to the CNS and regulate brain physiology
[63]. Importantly, the gut–brain axis depends on a tightly regulated interplay between immunity and gut microbiota. As previously stated, it has been shown that gut microbiota is significantly altered in PD patients, compared with healthy controls; it is becoming evident that it might play an important role in the pathogenesis of PD. In addition, recent evidence indicates that miRNAs are involved in PD pathophysiology, suggesting a novel disease-associated mechanism that is now beginning to be explored and positing miRNAs as potential biomarkers for PD screening. Interestingly, miRNAs were found in human faecal samples, mainly derived from intestinal epithelial cells. Furthermore, it was recently demonstrated that secreted miRNAs are able to enter bacteria and regulate bacterial gene transcripts and affect their growth
[64], giving rise to the possibility of the host’s miRNAs’ affecting and shaping their own gut microbiome in PD (
Figure 2). This hypothesis was proposed after
Fusobacterium nucleatum cultured with human miR-515-5p (a miRNA present in human faeces) showed an increased ratio of 16S rRNA/23S rRNA transcripts and altered growth
[64]. Moreover, selective deletion of Dicer in mice promoted gut microbiota imbalance and exacerbation of dextran sulphate sodium (DSS)-induced colitis. These effects were reversed by faecal miRNA transplantation from wild-type littermates
[64]. These studies further corroborate the hypothesis that miRNAs regulate gut microbiota and enhance the role of faecal miRNAs in influencing gut microbiota and preserving intestinal homeostasis. In order to better understand the potential interactions between miRNAs and the gut metagenome, Hewel and co-workers
[65] performed an in silico target screen for binding sites of PD-associated miRNAs on human gut metagenome sequences, from which resulted a massive number of interactions. They found numerous miRNAs that may be key regulators in bacterial pathways relevant to PD, including the bacterial secretion system and lipopolysaccharide (LPS) biosynthesis
[65]. On the other hand, gut microbiota, primarily through gut microbiota metabolites, may modulate human gene expression by affecting the host’s miRNA expression (
Figure 2). In fact, Peck et al. demonstrated that the microbiota modulates miRNA expression in intestinal epithelium cells, which may alter intestinal homeostasis
[66]. Different miRNA expression profiles were observed in the colon and ileum of germ-free mice colonised with gut microbiota from pathogen-free mice, when compared with germ-free littermates
[67]. The analysis of faecal miRNA expression patterns also revealed significant differences between conventional mice and germ-free mice
[68]. Additionally, it has also been demonstrated that depletion of gut microbiota using antibiotics induces alterations in faecal miRNA expression profiles in vivo
[69]. A report from 2014 revealed that adherent-invasive
E. coli, a pathogen highly prevalent in Crohn’s disease, upregulates miRNAs (miR-30c and miR-130a) that target genes involved in the autophagy response (ATG5 and ATG16L1) in mouse enterocytes, which may facilitate adherent-invasive
E. coli replication and aggravation of intestinal inflammation
[70]. Downregulation of miR-144 induced by
Lactobacillus casei was reported to enhance intestinal barrier integrity through upregulation of occludin and zonula occludens 1 levels in intestinal epithelial cells, boosting intestinal barrier function and homeostasis
[71]. Similarly, several probiotics have shown their ability to modulate miRNAs in intestinal epithelial cells and immune cells, altering intestinal barrier function and intestinal immune regulation
[72][73][74]. A link between gut microbiota and brain miRNA expression has been established, since abnormal brain miRNA expression profiles have been described in the amygdala and prefrontal cortex of germ-free mice and mice treated with antibiotics to induce microbiota depletion
[75]. Moreover,
Bacteroides fragilis LPS, a microbial endotoxin, was shown to induce several miRNAs responsible for targeting genes that regulate synaptic architecture, amyloidogenesis, and brain inflammatory signalling
[76]. These studies shed light on bidirectional communication between gut microbiota and miRNAs. Since both gut dysbiosis and differently expressed miRNAs have been reported in PD, the bidirectional interaction of gut microbiota–miRNAs might be involved in the pathophysiology of PD. We can hypothesize that microbial dysbiosis can potentiate the dysregulation of certain miRNAs, which can then target mitochondria and inflammatory pathways in the gut and later on in the brain, triggering PD pathological hallmarks (
Figure 3).
Figure 2. Reciprocal regulation of miRNAs and gut microbiota. Host’s intestinal epithelial cells release miRNAs that have the ability to regulate bacterial gene transcripts, affecting bacteria growth and replication. On the other hand, microbiota regulates host’s miRNA expression. Created with Biorender.
Figure 3. miRNA–microbiota axis in PD. Microbial dysbiosis can potentiate the dysregulation of certain miRNAs, which can then target mitochondria and inflammatory pathways in the gut, leading to a proinflammatory response and to the loss of intestinal barrier integrity. This will allow miRNAs to travel freely or within vesicles through the blood or through the vagus nerve and reach the brain. Within the brain, miRNAs can target mitochondria and activate neuronal innate immunity, ultimately leading to PD hallmarks. Created with Biorender.
3. Conclusions and Future Perspectives
The complexity and heterogeneity of PD have contributed to the challenging nature of PD diagnosis and treatment. The currently used diagnostic methods for PD are based on clinical symptoms that only emerge when the disease has already progressed to a stage of elevated neuronal loss. Therefore, there is a growing need to identify molecular biomarkers that allow an early and precise diagnosis. The possibility of using miRNAs as potential biomarkers for the diagnosis of PD during the prodromal phase gained relevance with the emergence of the hypothesis that miRNAs are involved in the pathogenesis and pathophysiology of PD. Several promising candidate miRNAs have surfaced from the numerous screenings performed so far on CSF, blood, and brain tissue of PD individuals or models. Moreover, considering the involvement of the gut in the aetiology and pathophysiology of PD, studying dysregulated miRNAs in the gut or even in the faecal material of PD patients might lead to the surfacing of an important and useful strategy for the early detection of gut-first PD and for the development of therapies able to slow or halt disease progression before it reaches the brain.
However, although microRNAs show potential to be reliable biomarkers, most of the candidate miRNAs are not organ-specific, and their profile is highly dependent on uncontrollable factors such as individual genetics and ethnicity, as reported in a study from 2020, where differences were observed between different ethnic cohorts of PD patients (American, Asian, and European)
[77].
The potential of using miRNA-based therapies for the treatment of PD has also gained importance over the years. The scientific community has been developing strategies for miRNA modulation for many years, by using miRNA-mimics and antago-miRs to upregulate or downregulate miRNA levels, respectively. Nonetheless, the path to develop miRNA-based therapies has been paved with numerous challenges, including the lack of target specificity of miRNAs and their delivery to specific sites
[78]. Unravelling relevant targets of miRNAs in PD models, followed by human validation of the results, may accelerate finding new biomarkers for an early diagnosis and novel therapeutic strategies for PD. To find a promising gut miRNA that can be a PD biomarker candidate would allow for therapeutic intervention in the prodromal phase of these patients, stopping PD progression and development. Hence, further studies of the microbiome–miRNA axis are a must in PD research.