Regulation of the Key Epithelial Cancer Suppressor miR-124 Function by Competing Endogenous RNAs

A decrease in the miR-124 expression was observed in various epithelial cancers. Like a classical suppressor, miR-124 can inhibit the translation of multiple oncogenic proteins. Epigenetic mechanisms play a significant role in the regulation of miR-124 expression and involve hypermethylation of the MIR-124-1/-2/-3 genes and the effects of long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) according to the model of competing endogenous RNAs (ceRNAs). More than 40 interactomes (lncRNA/miR-124/mRNA) based on competition between lncRNAs and mRNAs for miR-124 binding have been identified in various epithelial cancers. LncRNAs MALAT1, NEAT1, HOXA11-AS, and XIST are the most represented in these axes. Fourteen axes (e.g., SND1-IT1/miR-124/COL4A1) are involved in EMT and/or metastasis. Moreover, eight axes (e.g., OIP5-AS1/miR-124-5p/IDH2) are involved in key pathways, such as Wnt/b-catenin, E2F1, TGF-β, SMAD, ERK/MAPK, HIF-1α, Notch, PI3K/Akt signaling, and cancer cell stemness. Additionally, 15 axes impaired patient survival and three axes reduced chemo- or radiosensitivity. To date, 14 cases of miR-124 regulation by circRNAs have been identified. Half of them involve circHIPK3, which belongs to the exonic ecircRNAs and stimulates cell proliferation, EMT, autophagy, angiogenesis, and multidrug resistance. Thus, miR-124 and its interacting partners may be considered promising targets for cancer therapy.


Introduction
Recent studies have increasingly shown that less than 1.5% of the human genome genes encode proteins. Non-coding RNAs (ncRNAs) transcribed from non-proteincoding genes are involved in the regulation of all biological processes. Among them, microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) are of the greatest importance. These two large classes differ from each other in length [1]. miRNAs, composed of up to 24 nucleotides, suppress the translation of protein-coding genes of wide functionality. It has been shown that miRNAs are involved in many cellular processes as regulators of homeostasis and their expression is highly conserved in various cells. The key role of miRNAs in carcinogenesis has also been revealed. miRNAs may be oncogenic, suppressor, or may have either function, depending on cancer localization [2,3].
miR-124, first discovered in mice, is a typical suppressor miRNA. This miRNA, found not only in mammals but also in worms, is notable not only for its conservation but also for the unusual decrease in expression associated with the development of the malignant transformation of various etiologies [3,4]. Three genes encode for this miRNA: MIR-124-1 (8p23.1), MIR-124-2 (8q12. 3), and MIR-124-3 (20q13.33). All of them miR-124 is a subject of interest for the study for several reasons, including its high conservativeness in various tissues of many organisms, as well as the fact that its abnormal expression is associated with cancer of various etiologies. miR-124 has been shown to be involved in the regulation of cell growth, differentiation, and development, while the disruption of this miRNA can lead to the development of malignant cell transformation, in particular, to cell cycle arrest, epithelial-mesenchymal transition (EMT), metastasis and resistance to chemotherapy. Therefore, this miRNA is considered a potential biomarker for developing new therapeutic strategies for treating tumors [16].
Several studies have shown the suppressive nature of miR-124 in cancers of various etiologies. Also, for several cancers, the ability to suppress the proliferation, invasion, and migration of tumor cells has been shown. It is noted that, depending on the process and ectopic location, miR-124 uses different targets to suppress a particular type of cancer, although in tumor cells, regardless of etiology, this miRNA is suppressed in various ways, including hypermethylation of genes encoding it. The CBL (Casitas B-lineage lymphoma; encodes ubiquitin ligase E3), PDCD6 (programmed cell death 6), ROCK1 (Rho-associated protein kinase 1), SNAI1/2 (Snail family transcription repressors 1 and 2), TWIST1/2 (Twist family BHLH transcription factors 1 and 2), ASPP (apoptosis-stimulating protein p53), iASPP (inhibitor of apoptosis-stimulating protein p53), SPHK1 (sphingosine kinase 1), NRP1 (neuropilin 1) genes are the previously discovered target genes of this miRNA in different types of cancer. All these genes show oncogenic properties affecting the development and progression of epithelial tumors [3,4].
Note that not only proteins encoding by miR-124 target genes are important in the development of cancer. To assess the picture of pathological transformation, one should consider the cellular signal pathways that trigger when various system processes are disrupted. In particular, the involvement of miR-124 in the process of suppressing tumor activity as part of a cascade of interactions within the Wnt/β-catenin pathway deserves attention. Thus, it was revealed that miR-124 with miR-340 cooperation participates in the specific suppression of the SRGAP1 gene (SLIT-ROBO Rho GTPase Activating Protein 1). As a result, this leads to the suppression of the growth of the gastric cancer cell line. The SRGAP1 gene is an oncogene that, according to several studies, is not only triggered in cancer cells of various etiologies but is also part of the Wnt/β-catenin pathway as the main regulator of the activity of this pathway. Alternatively, miR-124 suppresses the expression of this gene, resulting in the entire cascade "fade" [3,4].
Recent studies provide an opportunity for a more complete assessment of the role of miRNAs in the development of cancers of various localizations. A study by a group of Chinese scientists on the role of miR-124 in the development of malignant transformation of liver cells has revealed a new target. The study consisted of several stages. At the stage conducted on hepatocellular cancer stem cell lines, it was shown that miR-124, whose expression level was high, suppressed the activity of the CAV1 (Caveolin-1) gene found using bioinformatic screening. A mutual regulation within this pair was also noted. Moreover, the results showed tumor progression and revealed an unfavorable prognosis for patients with such gene interaction. One of the possible reasons for this outcome was called resistance to the standard therapeutic agent in this type of cancer when suppressing miR-124 in tumor cells. The authors consider a decrease in resistance to the therapeutic agent and oncogene expression with an artificial increase in miR-124 expression to be an optimistic result. Similar results were obtained for colon cancer [17,18]. Noteworthy, several other target genes have been identified for liver cancer: C/EBPa (CCAAT enhancer binding protein α), which encodes an important transcription factor and is also regulated by miR-124 [19], CRKL (V-crk sarcoma virus CT10 oncogene homolog (avian)-like), which is involved in various cellular processes [20], BIRC3 (Baculoviral IAP Repeat Containing 3) and some others, the suppression of whose expression by miR-124 reduced the proliferative activity of tumor cells, invasion, and migration [21][22][23].
According to some studies, one of the important characteristics of malignant neoplasms is uncontrolled cell growth. The results of a study of a colon cancer-specific marker of the IQGAP1 gene (IQ motif containing GTPase activating protein) have shown that there is a mechanism for suppressing this process. Scientists have found that direct regulation through miR-124 significantly reduces the level of oncogene expression by stopping the growth of tumor cells. It is supposed to be interesting the role of this pair in the launch of the mentioned Wnt/β-catenin pathway [24].
Another important characteristic of cancer is progression. Identifying a marker that allows you to assess the metastatic potential of a tumor is extremely important for understanding the anatomy of any type of cancer. Thus, in assessing the metastatic potential of miR-124 in breast cancer, the ABCC4 (ATP-binding cassette subfamily C member 4) target gene was identified. The miR-124/ABCC4 pair was associated not only with tumor progression but also with drug resistance [25].
When studying the metastatic potential of miR-124, our group obtained some convincing results indicating a high level of metastasis formation in the resection material in ovarian and breast cancers. In a study of 20 miRNA genes in ovarian carcinoma, we showed an association of aberrant methylation of several miRNA genes, including miR-124, with the subsequent development of metastases [26]. It has been shown that because of aberrant methylation, the entire regulatory cascade is disrupted, which leads to the deregulation of the ZEB1 and ZEB2 (zinc finger E-box linking the genes homeobox 1 and 2) genes, involved in EMT, which subsequently leads to the development of metastases.
In our other work, a comprehensive assessment of the interaction between miRNAs, their target genes, and possible regulatory long non-coding RNAs in ovarian carcinoma was carried out. Analysis of the study results revealed a potential interaction between the AURKA (Aurora Kinase A) gene and miR-124 [27].
EMT is considered to be the early and most significant stage of cancer progression. This process is characterized by the transition of adhesive cells into motile cells and is regulated by miRNAs by reducing the expression level of genes responsible for adhesive or epithelial processes, such as E-cadherin, and increasing the expression of genes responsible for mesenchymal processes, such as N-cadherin. Normally, this process is reversible and is extremely important in the formation of various organs during embryonic development, as well as necessary for wound healing [4].
However, as the results of various studies show, violations of the regulation of the expression of several genes associated with this process can lead to the development and progression of cancers of various localizations, as well as affect resistance to different therapeutic markers. In particular, in a study conducted on breast cancer, it was shown that the interaction of miR-124 with the most significant genes regulating EMT, namely ZEB1/2, may be crucial for understanding the formation of cancer. Thus, the study showed that a decrease in the expression level of miR-124 in the tumor increased the expression level of the ZEB2 gene, followed by the launch of EMT and the development of metastases. With a high level of miR-124 expression in the tumor, the reverse process was observed. Thus, using the example of triple-negative breast cancer, it was revealed that this gene is a direct target of miR-124 and is associated with both EMT and metastasis [28].
It should be noted that in the study of clear cell renal cell carcinoma, the synergy of miR-124 with another miR-203 miRNA was found. These regulators jointly suppressed the expression level of the ZEB2 gene, which leads to the suppression of the proliferation and migration of tumor cells [29].
Importantly, in breast cancer, in addition to the EMT genes, another miR-124 target gene was identified. It is a signaling regulator and transcription activator STAT3, which is considered a specific oncogene for breast cancer. The authors note that miR-124 directly suppresses the activity of this gene, as a result reducing the proliferation, invasion, and migration of tumor cells [30].
Along with genomic and epigenomic rearrangements, significant changes occur in the metabolic system of the body with a tumor. A study of pancreatic cancer by a group of Chinese scientists has revealed that miR-124 inhibits the mechanism of metabolic transformation in tumor cells by suppressing the expression of the MCT1 gene (monocarboxylate transporter type 1) and slows down cell growth [31].
Another target was identified in a study of liver cancer formation. The authors found that the suppression of miR-124 expression in the tumor led to increased expression of the AKT1S1 (AKT1 Substrate 1) gene encoding the enzyme PRAS40. The reverse process was detected on cell lines. Based on these results, the authors indicated the diagnostic potential of this marker [32].
Recently, there has been an active study of the effect on the development of malignant transformation of immune control points. One of these points is the PD-L1 gene (Programmed death ligand 1). This gene is a target for miR-124 and is also part of the STAT3 signaling pathway. According to the results of a study on colon cancer, miR-124 plays a huge role in tumors of this localization. It was shown that miR-124 was suppressed in the tumor, while the expression level of PD-L1 was high. The results on cell lines revealed the opposite effect. There was also a concomitant effect of lowering the expression level of several other genes, as IL10 (interleukin 10), IL2 (interleukin 2), TNF-α (tumor necrosis factor α), TGF-β (transforming growth factor beta), IFN-γ (interferon gamma). Additionally, an increase in the expression level of miR-124 decreased proliferation and termination of the cell cycle and triggered both pathways of apoptosis. The authors consider miR-124 as a therapeutic marker [33].
It should be noted that a target gene unrelated to immunity has been identified for this type of cancer. miR-124 enhances the expression of the KiSS1 (Kisspeptin 1) gene reducing the proliferation of tumor cells, their invasion, and migration [34].
In the study of gastric cancer, a potential oncogene was identified, which is a direct target for miR-124. A study on cell lines showed that the SRGAP1 gene (Slit-Robo GTPase-activating protein 1) enhances the proliferation, invasion, and migration of tumor cells. If this gene was suppressed because of a high level of miR-124 expression, the opposite effect was observed [35].
Recently, the studying of the influence of various factors on the properties of malignant transformation of stem tumor cells has become quite popular. This cell type is a precursor to the development of various types of tumors, changing its environment and adjusting the substrate for colonization, providing all further processes inherent. In their remarkable study, a group of Chinese scientists managed to identify interesting patterns. Based on the results of the work, miR-124 directly interacts with the JAMA (junctional adhesion molecule A) gene, which triggers the proliferation of tumor cells, promotes adhesion, and allows for plastic transition. The mutual regulation of this pair is of interest from the perspective of therapy, since a high level of miR-124 expression suppresses the expression of the target gene and thus inhibits the properties of tumor stem cells [36].
There should be also noted several studies in which the authors could trace entire regulatory pathways that trigger the processes of malignant transformation in tumors of various locations. Thus, in the study of squamous cell carcinoma of the esophagus, a pathway regulating the processes of growth and invasion in this type of cancer was identified. In addition to miR-124, the DNMT1 (DNA methyltransferase 1) gene and the BCAT1 (branched-chain amino acid transaminase) long non-coding RNA gene are involved in this pathway [37].
In another study on triple-negative breast cancer, the identified pathway was longer. Scientists have identified several genes and their regulators that are part of the large Wnt signaling pathway. The authors have approved miR-124 as a trigger factor for aberrant processes by targeting Axis inhibition protein 1 (Axin1). There was also a concomitant effect of changing the expression level of several other genes such as BAX (B-cell lymphoma-2 associated X), BCL-2 (B-cell lymphoma-2), and β-catenin that are involved in the Wnt signaling pathway [38]. The same miRNA was also involved in STAT3/VEGF pathway, also detected in breast cancer. The authors claim that the direct target of miR-124 is cyanidin-3-glucoside (C3G) involved in the angiogenesis of breast cancer. This pair has been deemed a factor in the progression of breast cancer [39].
Separately, a recent bioinformatic screening of ovarian cancer databases made it possible to compile a complete network of interacting genes involved in the development and progression of ovarian cancer. According to the results of this study, the most significant targets for miR-124 in ovarian cancer may be genes: CCNB1 (cyclin B1), CEP55 (centrosome protein 55), RACGAP1 (Rac GTPase-activating protein 1), TPX2 (target protein for Xklp2), UBE2C (ubiquitin-conjugating enzyme E2C), ZWINT (ZW10-interacting kinetochore protein), and CENPM (centromere protein M). It is assumed that these genes can also be targets for miR-107, miR-34a-5p, miR-129-2-3p, and some others. The authors suggested that the identified interacting miRNA-mRNA pairs can serve as factors in the progression of ovarian cancer [47].
Bioinformatic database screening for bladder cancer followed by gene co-expression network analysis revealed six new genes (PPARD, CST4, CSNK1E, PTPN14, ETV6, and ADRM1) and several new miRNAs, including miR-124, as drivers in the development and progression of bladder cancer [48]. The predicted regulatory gene cascades involving miR-124 suggest even more multiple functions of this miRNA in epithelial cancers.
The results of all these studies confirmed miR-124 as a perspective biomarker for different cancer types.
It is known that miR-124 itself can be inhibited by hypermethylation of MIR-124-1/-2/-3 genes, encoding miR-124 [5][6][7]26]. The important role of long and circular non-coding RNAs discovered in the last decade in the dysregulation of miR-124 and its targets has been established, which is the subject of the following sections of this review.

Long Non-Coding RNAs in Dysregulation of miR-124 Target Genes in Epithelial Cancers
In the last decade, the mechanism of regulation of protein genes involving not only miRNAs, but also other regulatory ncRNAs along the ncRNA/miRNA/mRNA scheme has received numerous confirmations. The interaction of miRNAs with target mRNAs and with regulatory ncRNAs requires the presence of the miRNA response elements (MRE) in sequences of both messenger RNAs and noncoding RNAs, such as long ncRNAs or circular ones [12,49,50]. It was shown that direct bindings are formed in ncRNA-miRNA and Figure 1. The scheme of differential regulation by miR-124: targets involved in the main biological processes in several common epithelial cancers. Central column-targets, left column-cell processes, right column-cancer types. The data from articles [3,4,16,28,30,[40][41][42][43][44][45][46] were used.
Separately, a recent bioinformatic screening of ovarian cancer databases made it possible to compile a complete network of interacting genes involved in the development and progression of ovarian cancer. According to the results of this study, the most significant targets for miR-124 in ovarian cancer may be genes: CCNB1 (cyclin B1), CEP55 (centrosome protein 55), RACGAP1 (Rac GTPase-activating protein 1), TPX2 (target protein for Xklp2), UBE2C (ubiquitin-conjugating enzyme E2C), ZWINT (ZW10-interacting kinetochore protein), and CENPM (centromere protein M). It is assumed that these genes can also be targets for miR-107, miR-34a-5p, miR-129-2-3p, and some others. The authors suggested that the identified interacting miRNA-mRNA pairs can serve as factors in the progression of ovarian cancer [47].
Bioinformatic database screening for bladder cancer followed by gene co-expression network analysis revealed six new genes (PPARD, CST4, CSNK1E, PTPN14, ETV6, and ADRM1) and several new miRNAs, including miR-124, as drivers in the development and progression of bladder cancer [48]. The predicted regulatory gene cascades involving miR-124 suggest even more multiple functions of this miRNA in epithelial cancers.
The results of all these studies confirmed miR-124 as a perspective biomarker for different cancer types.
It is known that miR-124 itself can be inhibited by hypermethylation of MIR-124-1/-2/-3 genes, encoding miR-124 [5][6][7]26]. The important role of long and circular non-coding RNAs discovered in the last decade in the dysregulation of miR-124 and its targets has been established, which is the subject of the following sections of this review.

Long Non-Coding RNAs in Dysregulation of miR-124 Target Genes in Epithelial Cancers
In the last decade, the mechanism of regulation of protein genes involving not only miRNAs, but also other regulatory ncRNAs along the ncRNA/miRNA/mRNA scheme has received numerous confirmations. The interaction of miRNAs with target mRNAs and with regulatory ncRNAs requires the presence of the miRNA response elements (MRE) in sequences of both messenger RNAs and noncoding RNAs, such as long ncRNAs or circular ones [12,49,50]. It was shown that direct bindings are formed in ncRNA-miRNA and mRNA-miRNA pairs, and competition between ncRNA and mRNA for binding to miRNA is observed. This mechanism is called the competing endogenous RNAs (ceRNA) model [12,49,50]. Vast experimental data have been accumulated, supporting the involvement of lncRNA/miRNA/mRNA interactomes in the development and progression of cancer [51,52]. According to the ceRNA mechanism with the participation of suppressor miRNAs, such as miR-124, activation of target protein genes with oncogenic potential is usually observed. The lncRNAs involved in the inhibition of miR-124 also demonstrated oncogenic properties.
The discovery of new lncRNA/miRNA/mRNA interactomes is performed using a set of methods. Briefly, this is, firstly, a bioinformatics analysis that selects lncRNA/miRNA and miRNA/mRNA pairs from transcriptome databases, such as The Cancer Genome Atlas (TCGA), with a negative correlation of expression levels, but it is also necessary to identify a positive correlation between the levels of lncRNA and mRNA. Then, using the scanMiR tool (https://github.com/ETHZ-INS/scanMiR, accessed on 2 October 2022), the presence of a binding site for a given miRNA in the 3 -UTR mRNA and lncRNA sequences is checked [53]. Other approaches exist (e.g., [54]). Then, the bioinformatically identified transcriptome correlations are confirmed experimentally by analyzing the levels of RNAs of all three types in a representative set of cancer samples and evaluation of the expected expression level relationship (e.g., [55,56]). Further, functional studies are applied on cell lines of this type of cancer, which, in short, include transfection of synthesized RNAs into cells, artificial suppression or overexpression of triplet components, loss and gain of function, analysis of physiological changes in cell culture (as the level of proliferation and apoptosis, migration and cell invasion). Direct bindings between the components of the triplet are established using the luciferase test, RNA-binding protein immunoprecipitation (RIP) assay, and RNA pull-down assay. In some works, the authors confirmed the functional significance of these interactions not only in cell cultures but also with the use of model animals, usually immunodeficient mice (e.g., [55,56]). Some studies also elucidated the effect of lncRNA and interactome, or in other words, the lncRNA/miRNA/mRNA axis, on the survival rate and response to chemotherapy in patients.
In published research (PubMed on 11 September 2022), we have identified more than 40 such interactomes involving miR-124, which are summarized in Table 1. We have presented very briefly and compendiously data on the methods used for axis validation and on the functional role of triplets in the development or progression of the corresponding tumor types, as well as data on their clinical significance for patients (if available in the cited articles).     As shown in Table 1, typical suppressor miR-124 is regulated, more precisely, inhibited in epithelial cancers exclusively by oncogenic lncRNAs, while miR-124 itself also targets mRNA of oncogenic proteins. The role of the most studied lncRNA MALAT1 (Metastasisassociated lung adenocarcinoma transcript 1) in the regulation of miR-124 is shown in nine studies conducted using primary tumors and cell lines or only cell lines of seven types of epithelial cancer: bladder transitional cell carcinoma, breast cancer, cervical cancer, gastric cancer, hepatocellular carcinoma, nasopharyngeal carcinoma, non-small cell lung cancer [40,[73][74][75][76][77][78][79][80]97]. In these studies, mRNAs of the proteins Capn4, CDK4, EZH2, foxq1, GRB2, HBx, SLUG, STAT3, and TGF-β1 were identified as direct targets of miR-124, which is also shown in Figure 2a. Additionally, using a set of methods, the promoting effect of interactomes formed by MALAT1 on the progression of various types of cancer, including increased proliferation, migration, invasion, EMT, and, in contrast, suppression of apoptosis, was established. Experiments on mouse xenografts have shown that interactomes based on MALAT1, miR-124, and oncogenic proteins (as CDK4, EZH2, foxq1, HBx, and SLUG) also stimulate the growth and metastasis of tumors in vivo [74][75][76]78,79,97]. A decrease in the survival rate of patients under the influence of the MALAT1/mir-124 axes, which increase the level of expression of oncogenic proteins CDK4, foxq1, and SLUG [74,76,79,97], has been established. Depending on the target protein, the activation of several signaling pathways was revealed, such as E2F1 signaling (through activation of CDK4 protein expression), PI3K/Akt signaling, and stemness (through HBx activation), as well as TGF-β, SMAD, and ERK/MAPK signaling pathways (through activation of TGF-β1) [74,78,80].  Another highly represented in tumors and actively studied lncRNA NEAT1 (nuclearenriched abundant transcript 1) was regulated in four types of cancer: breast cancer, hepatocellular carcinoma, nasopharyngeal carcinoma, and malignant thyroid nodules [81][82][83][84]. All these studies were conducted both on cell lines and on clinical samples from patients; xenografts of mice were used in two studies [81,82]. Four interactomes were identified: NEAT1/miR-124-3p/ATGL (adipose triglyceride lipase), NEAT1/miR-124-3p/p65 (NF-κB), NEAT1/miR-124/PDCD6 (programmed cell death 6), NEAT1/miR-124-3p/STAT3 (signal transducer and activator of transcription 3) [81][82][83][84], which is also shown in Figure  2b. These interactomes increased cell proliferation, migration, invasion, EMT and inhibit apoptosis in vitro and initiate tumor growth in immune-deficient mice in vivo. NEAT1/miR-124-3p/p65 (NF-κB) axis induces NF-κB signaling pathway in nasopharyngeal carcinoma [82]. NEAT1-mediated abnormal lipolysis promotes the growth of hepatocellular carcinoma cells [81]. Moreover, the activation of ATGL disrupts lipolysis in hepatocellular carcinoma cells, which is accompanied by the activation of DAG (diacylglycerol) + FFA (free fatty acids)/PPARα (peroxisome proliferator-activated receptor alpha) signaling [81].
LncRNA XIST (X-inactive specific transcript) inhibits miR-124 in three types of cancer: bladder cancer, laryngeal squamous cell carcinoma, and tongue squamous cell carcinoma, which activates AR (androgen receptor), EZH2 and JAG1 (jagged 1) proteins [94][95][96]. All three interactomes XIST/miR-124/AR, XIST/miR-124/EZH2, and XIST/miR-124/JAG1 are also shown in Figure 2c. These axes initiated by XIST are involved in promoting cell proliferation, migration, and invasion of three different cancer types as shown in vitro. Using tumor xenografts in nude mice, it was shown that the XIST/miR-124/EZH2 axis increased tumor growth in vivo in laryngeal squamous cell carcinoma [95]. Additionally, the XIST/miR-124/AR axis in bladder cancer has been shown to upregulate the expression of proliferation-associated factors, c-myc and p27, and metastasis-associated factors, MMP9 and MMP13 [94].
LncRNA HOXA11-AS (HOXA11 (Homeobox A11) Antisense RNA) also belongs to the most studied in the regulation of miR-124 and its protein-coding targets. Interactomes with HOXA11-AS have been identified in three types of cancer: hepatocellular carcinoma, gastric cancer, and non-small cell lung cancer, both on clinical samples and on cell lines [61][62][63]. The promoting effect of axes HOXA11-AS/miR-124/EZH2 (enhancer of zeste homolog 2 of polycomb group protein), HOXA11-AS/miR-124-3p/ITGB3 (integrin β3), and HOXA11-AS/miR-124/Sp1 (Sp1 transcriptional factor) on cancer cell proliferation, migration, invasion was shown in vitro. The most recent study [62] showed the effect of the HOXA11-AS/miR-124-3p/ITGB3 axis on growth and metastasis in vivo using mouse xenografts. Additionally, two studies have shown a decrease in patient survival under the action of axes that activate the expression of EZH2 and integrin β3 proteins [61,62].
The work that examined the role of suppressor lncRNA LINC01488 in the regulation of suppressor miR-124 and its target, mRNA of vimentin, in hepatocellular carcinoma [98] is somewhat aloof. Long intergenic non-coding RNA LINC01488 has been revealed as a key negative regulator of this cancer. The LINC01488/cyclin E/miR-124-3p/vimentin interactome does not directly bind the two suppressors, lncRNA LINC01488 and miR-124-3p. LINC01488 activates miR-124-3p through ubiquitination of cyclin E mediator, which can inhibit miR-124-3p. In cell cycle analysis, overexpression of LINC01488 inhibited G1 progression and suppressed S phase entry through effects on cyclin E function. Interestingly, the cyclin E mRNA level was not altered by LINC01488 regulation, indicating specific effects of LINC01488 on cyclin E at the translation level. Results from the RIP assay suggested that LINC01488 binds to cyclin E and decreases the expression of the protein via the ubiquitin-proteasome pathway. This is the only example of an interactome capable of following the function of miR-124-3p itself-to suppress proliferation, migration, invasion, and EMT of hepatocellular carcinoma cells in vitro, tumor growth, and metastasis in vivo. Additionally, this interactome decreases the overall and recurrence-free survival (OS, RFS) of patients. The LINC01488/cyclin E/miR-124-3p/vimentin interactome involves the ubiquitin-proteasome pathway.
Of interest are also works including recently published ones, in which the direct binding of lncRNA to miR-124, that can inhibit the expression of miR-124, and the effect of these interactions on cancer progression is clearly demonstrated, although target genes in these interactomes have not yet been identified. For example, increased expression of the oncogenic lncRNA NEAT1, which can bind and reduce the level of the suppressor miR-124-3p, has also been shown in ovarian cancer [99]. Elevated NEAT1 and decreased miR-124-3p levels are associated with advanced-stage and lymph node metastasis. NEAT1 can be stabilized by the RNA-binding protein HuR. Conversely, an increase in the miR-124-3p suppressor level can reduce the NEAT1 level and can be proposed for antitumor therapy [99].
In the tissues of the tongue squamous cell carcinoma (TSCC), an increase in the expression of the long intergenic non-coding RNA CASC15 (Cancer Susceptibility 15) and, conversely, a decrease in the level of miR-124 was revealed [100]. These changes are associated with poor overall patient survival. Additionally, it has been shown that overexpression of CASC15 can increase, whereas overexpression of miR-124 can reduce the ability of TSCC cells in cultures to migrate and invade [100].
A report on the oncogenic role of lncRNA SNHG16 (small nucleolar RNA host 16) in various cancer types and its involvement in many signaling pathways, such as TGF-β1/SMAD5, mTOR, NF-kB, Wnt, RAS/RAF/MEK/ERK, and PI3K/AKT, was published [92]. Moreover, this lncRNA serves as a sponge for several different miRNAs, among which the involvement of miR-124-3p was also revealed [92].
The effect of SNHG17, another member of the lncRNA family of small nucleolar RNA host genes (SNHGs), on breast cancer through direct binding of SNHG17 to miR-124-3p has been revealed. This study was conducted using rather reliable methods, such as luciferase reporter activity and RIP assays on cell cultures, as well as analysis of the expression in tissues of breast cancer patients and tumor growth in a xenograft model [101].
In non-small cell lung cancer, an inhibitory effect of MALAT1 on the miR-124 level and its association with EMT induction and cancer progression have been shown [102]. Inhibitory direct binding of MALAT1 and miR-124 has also been shown for cervical cancer in experiments on cell cultures and in vivo [103].
The interaction between lncRNA ZNF281 (zinc finger protein 281) and miR-124 predicted via IntaRNA 2.0 (http://rna.informatik.uni-freiburg.de/IntaRNA/Input.jsp, accessed on 2 October 2022) was confirmed using samples from gastric patients and cell cultures with the application of several standard approaches, such as transient cell transfections, qRT-PCR, loss-and gain-of-function, etc. The inhibition of miR-124, mediated by the oncogenic lncRNA ZNF281, promoted the migration and invasiveness of gastric cancer cells [104].
In a 2022 article, the interaction of lncRNA NEAT1 with miR-124-3p in ectopic endometrial cells was found to induce cell proliferation, migration, and invasion, which stimulates the malignant transformation of endometriosis and the development of endometrial cancer [105]. Another 2022 work showed direct inhibitory binding of long non-coding RNA DSCAM-AS1 (DSCAM antisense RNA 1) to miR-124 in hepatocellular carcinoma (HCC), inducing HCC cell proliferation [106].
Of interest, a purely bioinformatics analysis revealed another lncRNA, EMX2OS (EMX2 opposite strand antisense RNA), potentially involved in the regulation of miR-124, and two miR-124 target genes: CALCA (calcitonin-related polypeptide α) and GABRG2 (γ-aminobutyric acid, subunit of γ2 type A receptor) [107]. Recurrent and non-recurrent laryngeal cancer sample datasets were downloaded from the Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus database (GSE27020 and GSE25727). The new interactome EMX2OS/miR-124/CALCA, GABRG2 is presumably associated with molecular mechanisms of regulation of laryngeal cancer recurrent [107].
Thus, further studies must refine the targets in the presented data on incomplete lncRNA/mir-124 axes, and new detailed experimental work must validate the data of a purely bioinformatics analysis [107].

The Circular Long Non-Coding RNAs in Dysregulation of miR-124 Target Genes in Epithelial Cancers
The decisive influence in the formation of the circRNA pool is made by a process called backsplicing. This is the excision by the spliceosome of the looping sections of the transcript. During splicing, introns are removed from the maturing transcript not as a linear fragment, but as a lasso. When cutting off the "tail" of the lasso, usually formed by inverted repeats of neighboring introns, circular RNA will be obtained. These are circular intronic long non-coding RNAs (ciRNAs). Additionally, in the maturing transcript, individual exons, one or more, can also loop out, and such loops can be excised by the spliceosome and covalently closed into a ring, forming exonic circRNAs (ecircRNAs). A region containing several introns and exons can also loop out (sometimes with subsequent continuation of splicing). This process produces exon-intron circRNAs (EIciRNAs). The exonic ecircRNAs accumulate predominantly in the cytoplasm, whereas the other two groups (intronic ciRNAs and exon-intron EIcircRNAs) accumulate predominantly in the nucleus. Relatively few circRNAs are believed to act as miRNA sponges, but there are those for which such interactions have been well demonstrated. First, they are, of course, exonic ecircRNAs. Because of the lack of a free 3 or 5 end, circRNAs have a long half-life and are more stable than linear RNAs. For this reason, circRNAs are considered highly effective biomarkers for various cancer types [108]. Table 2 summarizes the interactomes formed by circRNAs, miR-124, and messenger RNAs of target genes; the methods of their investigation and their functions in the carcinogenesis of epithelial tumors are given.   It can be seen from Table 2 that almost half, six out of 14, articles are devoted to the effect of circHIPK3 on miR-124 expression in various oncological diseases, although this phenomenon has been mostly studied in hepatocellular carcinoma. This ecircRNA is expressed at a high level in the cytoplasm of cells of various tissues (including lungs, heart, stomach, colon, brain). It is formed due to the circularization of the second exon of the HIPK3 gene, flanked by introns containing complementary Alu repeats. circHIPK3 makes a significant contribution to the development of oncological diseases, the study of which can give a lot for their diagnosis and therapy. It can stimulate cell proliferation, autophagy, angiogenesis, EMT transition, inhibit pyroptosis, and cause the development of chemoresistance. Apparently, a significant part of these effects is associated with the influence on the expression of miR-124 and its targets, although the review [123] also mentions its interactions with other miRNAs.
When comparing Tables 1 and 2, it can be noted that the effect of circRNAs on the development of epithelial cancers through the regulation of miR-124 is observed less frequently than the effect of lncRNAs, however, in most of the cited works, it has been convincingly proven by a wide range of methods. In the overwhelming majority of studies, this influence concerns the characteristics important for developing the disease and metastasis, such as the survival of cancer cells, their proliferative activity, and the ability to migrate and invade. It can be noted that in all studies, both circRNA and lncRNA directly interacting with miR-124 exhibit an oncogenic effect, which agrees well with the oncosuppressive effect of miR-124 itself.
In several studies, the influence of many circRNAs, such as circ MTHFD2, circHIPK3, and circPVT1, has been associated with a decrease in sensitivity to chemotherapy, which was studied both by determining resistance to specific drugs [116,117] and by measuring in vitro increase in the expression of the gene that causes multidrug resistance, as MRP4 [112].
For ovarian cancer cells, the expression of cancer stem cell differentiation-related markers in vivo and in vitro was also studied, and it was concluded that circRNA, in particular circ_0026123, plays a significant role in maintaining cancer stemness [121]. In other words, the regulation of miR-124 by circRNA and lncRNA affects the same set of cancer cell characteristics, which indirectly indicates that miR-124 is correctly identified as the most significant miRNA in the interactomes described above.
The work [113] should be noted separately, where the effect of exosomes with miR-124 secreted by breast cancer cells was studied not on the cancer cells themselves, but on their environment (peripheral human endothelial cells). It was shown that miR-124 suppresses angiogenesis, whereas the effect of circHIPK3 on miR-124, in contrast, stimulates angiogenesis.
It should be noted that the role of lncRNA and circRNA in terms of modifying the effect of miR-124 on the microenvironment of cancer cells is not sufficiently disclosed in existing works. Thus, miR-124 significantly affects the immune response, including suppressing the development of various types of malignant tumors [124]. Additionally, exosomal miR-124 has been shown to suppress the transition of normal fibroblasts to cancer-associated fibroblasts in ovarian cancer [125]. The effect of lncRNA and circRNA on miR-124 activity in the tumor microenvironment may be the subject of further research.
We see that the set of those types of cancer, whose pathogenesis is significantly affected by the interaction of miR-124 with ncRNA, is generally very similar for lncRNA and circRNA. This indirectly confirms the thesis about the weakening of miR-124 expression as an important aspect of the pathogenesis of diseases, such as hepatocellular carcinoma, breast cancer, pancreatic cancer, gastric cancer, colorectal cancer, cervical cancer, and prostate cancer [3], and as an effective prognostic factor for these epithelial tumors [126]. Recall that miR-124 and the molecules interacting with it are considered promising targets for therapy in these diseases [3].
More problematic is the question of the targets of the influence of miR-124. Naturally, it should be borne in mind that not only for different types of cancer but also at its different stages, as well as for primary cancer/metastasis, these can be different targets. However, one can expect at least some agreement between the results of different authors. Thus, from the data of Tables 1 and 2, it can be noted that for gastric cancer and hepatocellular carcinoma, on which the largest number of studies have been performed, ncRNA-miR-124 interactions and miR-124-mRNA interactions have been confirmed, and the significance of both of them for carcinogenesis has been proven (Figures 3 and 4).
For ovarian cancer cells, the expression of cancer stem cell differentiation-related markers in vivo and in vitro was also studied, and it was concluded that circRNA, in particular circ_0026123, plays a significant role in maintaining cancer stemness [121]. In other words, the regulation of miR-124 by circRNA and lncRNA affects the same set of cancer cell characteristics, which indirectly indicates that miR-124 is correctly identified as the most significant miRNA in the interactomes described above.
The work [113] should be noted separately, where the effect of exosomes with miR-124 secreted by breast cancer cells was studied not on the cancer cells themselves, but on their environment (peripheral human endothelial cells). It was shown that miR-124 suppresses angiogenesis, whereas the effect of circHIPK3 on miR-124, in contrast, stimulates angiogenesis.
It should be noted that the role of lncRNA and circRNA in terms of modifying the effect of miR-124 on the microenvironment of cancer cells is not sufficiently disclosed in existing works. Thus, miR-124 significantly affects the immune response, including suppressing the development of various types of malignant tumors [124]. Additionally, exosomal miR-124 has been shown to suppress the transition of normal fibroblasts to cancerassociated fibroblasts in ovarian cancer [125]. The effect of lncRNA and circRNA on miR-124 activity in the tumor microenvironment may be the subject of further research.
We see that the set of those types of cancer, whose pathogenesis is significantly affected by the interaction of miR-124 with ncRNA, is generally very similar for lncRNA and circRNA. This indirectly confirms the thesis about the weakening of miR-124 expression as an important aspect of the pathogenesis of diseases, such as hepatocellular carcinoma, breast cancer, pancreatic cancer, gastric cancer, colorectal cancer, cervical cancer, and prostate cancer [3], and as an effective prognostic factor for these epithelial tumors [126]. Recall that miR-124 and the molecules interacting with it are considered promising targets for therapy in these diseases [3].
More problematic is the question of the targets of the influence of miR-124. Naturally, it should be borne in mind that not only for different types of cancer but also at its different stages, as well as for primary cancer/metastasis, these can be different targets. However, one can expect at least some agreement between the results of different authors. Thus, from the data of Tables 1 and 2, it can be noted that for gastric cancer and hepatocellular carcinoma, on which the largest number of studies have been performed, ncRNA-miR-124 interactions and miR-124-mRNA interactions have been confirmed, and the significance of both of them for carcinogenesis has been proven (Figures 3 and 4).  However, in different studies, very different mRNA targets are indicated as significant, there is very little coincidence in them both within the same type of cancer and when comparing different types. The only exception is EZH2, which is characterized by numerous functions. An increase in the expression of EZH2 is associated with increased proliferation, migration, and invasion of cancer cells in various types of cancer: hepatocellular carcinoma, ovarian carcinoma, gastric cancer, laryngeal squamous cell carcinoma, and prostate cancer ( Figure 5). Additionally, the EZH2 gene has been identified in many different axes, both involving regulatory lncRNAs (HOXA11-AS, LINC00511, MALAT1, and XIST) and circRNAs (circ-TRPS1, hsa_circ_0026123), as can be seen by comparing Tables 1 and 2  However, in different studies, very different mRNA targets are indicated as significant, there is very little coincidence in them both within the same type of cancer and when comparing different types. The only exception is EZH2, which is characterized by numerous functions. An increase in the expression of EZH2 is associated with increased proliferation, migration, and invasion of cancer cells in various types of cancer: hepatocellular carcinoma, ovarian carcinoma, gastric cancer, laryngeal squamous cell carcinoma, and prostate cancer ( Figure 5). Additionally, the EZH2 gene has been identified in many different axes, both involving regulatory lncRNAs (HOXA11-AS, LINC00511, MALAT1, and XIST) and circRNAs (circ-TRPS1, hsa_circ_0026123), as can be seen by comparing Tables  1 and 2. At present, ideas about the functions of EZH2 are being revised and supplemented. The canonical role of EZH2 is gene silencing by catalyzing trimethylation of lysine 27 histone H3 (H3K27me3) in a PRC2-dependent manner. However, evidence is accumulating that it can act as a transcriptional coactivator of genes involved in the development of various types of cancer, such as c-Myc, cyclin D1, CXCR4, IL6, TNF, IGF1R, MMPs, and AR. Several studies have shown that biochemical modification of EZH2 induces dissociation from the PRC2 complex and converts EZH2 from a transcriptional repressor to a transcriptional activator. Its participation in interactions at a level other than transcription (binding of various RNAs, suppression of ubiquitination) is also assumed. For this reason, EZH2 is currently considered a novel and promising target for cancer therapy [127].

Conclusions
The typical suppressor miR-124 has a significant anti-oncogenic effect and it can inhibit the translation of at least fifty protein-coding targets. Among the targets regulated  However, in different studies, very different mRNA targets are indicated as significant, there is very little coincidence in them both within the same type of cancer and when comparing different types. The only exception is EZH2, which is characterized by numerous functions. An increase in the expression of EZH2 is associated with increased proliferation, migration, and invasion of cancer cells in various types of cancer: hepatocellular carcinoma, ovarian carcinoma, gastric cancer, laryngeal squamous cell carcinoma, and prostate cancer ( Figure 5). Additionally, the EZH2 gene has been identified in many different axes, both involving regulatory lncRNAs (HOXA11-AS, LINC00511, MALAT1, and XIST) and circRNAs (circ-TRPS1, hsa_circ_0026123), as can be seen by comparing Tables  1 and 2. At present, ideas about the functions of EZH2 are being revised and supplemented. The canonical role of EZH2 is gene silencing by catalyzing trimethylation of lysine 27 histone H3 (H3K27me3) in a PRC2-dependent manner. However, evidence is accumulating that it can act as a transcriptional coactivator of genes involved in the development of various types of cancer, such as c-Myc, cyclin D1, CXCR4, IL6, TNF, IGF1R, MMPs, and AR. Several studies have shown that biochemical modification of EZH2 induces dissociation from the PRC2 complex and converts EZH2 from a transcriptional repressor to a transcriptional activator. Its participation in interactions at a level other than transcription (binding of various RNAs, suppression of ubiquitination) is also assumed. For this reason, EZH2 is currently considered a novel and promising target for cancer therapy [127].

Conclusions
The typical suppressor miR-124 has a significant anti-oncogenic effect and it can inhibit the translation of at least fifty protein-coding targets. Among the targets regulated Several studies have shown that biochemical modification of EZH2 induces dissociation from the PRC2 complex and converts EZH2 from a transcriptional repressor to a transcriptional activator. Its participation in interactions at a level other than transcription (binding of various RNAs, suppression of ubiquitination) is also assumed. For this reason, EZH2 is currently considered a novel and promising target for cancer therapy [127].

Conclusions
The typical suppressor miR-124 has a significant anti-oncogenic effect and it can inhibit the translation of at least fifty protein-coding targets. Among the targets regulated by miR-124, we could isolate at least six genes (PDCD6, ROCK1, SLUG, STAT3, TGF-β, ZEB1) common to several epithelial cancers, including lung, stomach, hepatocellular, breast, and ovarian cancers. Through the inhibition of oncoprotein translation, miR-124 induces apoptosis and cell cycle arrest, inhibits proliferation, invasion, EMT, metastasis, cancer cell stemming, chemoresistance, and improves the prognosis of patient survival, which has been shown in various types of cancer. Thus, miR-124 and its targets play a critical role in key biological processes in the development and progression of epithelial cancers.
Usually, the methylation of genes encoding miR-124 is considered first as a factor influencing the decrease in its expression. Without denying the significance of methylation of MIR-124-1/-2/-3 genes, encoding miR-124, and the possibility of influencing its expression through the level of methylation, we consider it important to pay attention to alternative mechanisms.
Indeed, as has been clarified in the last decade, oncogenic lncRNAs and some circular RNAs form interactomes by the mechanism of competing endogenous RNAs (ceRNAs) according to the ncRNA/mir-124/mRNA scheme. These competitive interactions of mRNA and ncRNA (both lncRNA and circRNA) with miR-124 are possible if there are MRE sites for binding this miRNA in mRNA and regulatory ncRNA sequences. Moreover, oncogenic lncRNA and circRNA inhibit the suppressor miR-124, which, as a result, cannot effectively suppress the translation of its oncogenic protein targets.
More than 40 interactomes involving lncRNAs and miR-124 have been identified in epithelial cancers. LncRNAs MALAT1, NEAT1, HOXA11-AS, and XIST are the most represented in these axes. The largest number, nine interactomes, was formed with the participation of the widely studied lncRNA MALAT1 in epithelial cancer of seven types: bladder transitional cell carcinoma, breast cancer, cervical cancer, gastric cancer, hepatocellular carcinoma, nasopharyngeal carcinoma, and non-small cell lung cancer. mRNAs of the proteins Capn4, CDK4, EZH2, foxq1, GRB2, HBx, SLUG, STAT3, and TGF-β1 were identified as direct targets of miR-124 in these axes. LncRNA NEAT is involved in four interactomes, lncRNA HOXA11-AS is involved in three axes, and XIST is also involved in three axes. Interestingly, feedback loops are noted for many interactomes, such as for NEAT1/miR-124/STAT3. Axes involving new, relatively recently discovered lncRNAs, such as PTPRG-AS1/miR-124-3p/CCND1 and PTPRG-AS1/miR-124-3/LHX2, have also been identified. LncRNA PTPRG-AS1 through these axes promotes cell proliferation and cell cycle in vitro and in vivo, and activation of LHX2 induces the Notch pathway and reduces radiosensitivity.
Virtually all lncRNA interactomes inhibiting miR-124 and activating oncogenic proteins stimulate proliferation, inhibit apoptosis, and activate cell motility and cancer invasion, regardless of cancer type. A total of 14 axes, including lncRNAs and miR-124, are involved in EMT and/or metastasis in cancers. Moreover, eight axes are involved in key pathways: Wnt/b-catenin, E2F1, TGF-β, SMAD, ERK/MAPK, HIF-1α, Notch, PI3K/Akt signaling, and cancer cell stemness. Additionally, 15 axes showed a poor outcome or shorter survival time, and three axes decreased the chemo-or radiosensitivity of cancer patients.
To date, 14 interactomes involving circRNAs in the regulation of miR-124 and its targets have been identified. Nearly half of these include circHIPK3, which belongs to the exonic ecircRNAs found predominantly in the cytoplasm. This abundant circHIPK3 can stimulate cell proliferation, autophagy, angiogenesis, EMT, and chemoresistance, which has been shown in various cancers, although most of the work has been done on hepatocellular carcinoma. The influence of circ MTHFD2 and circPVT1 is also associated with a decrease in sensitivity to chemotherapy shown in gastric cancer.
It is important to emphasize that the luciferase reporter assay, RNA pull-down, and RIP assays were used as the most reliable methods for verifying the direct binding of miR-124 to ncRNAs (both, lncRNA and circRNA) and miR-124 to mRNAs in interactomes.
It should also be noted that exosomal miR-124 affects not only tumor cells, but also the microenvironment and significantly affects the immune response of cancer cells. However, there are few data on the analysis of the effect of lncRNAs and circRNAs on miR-124 activity in the tumor microenvironment, which requires further research.
As we have mentioned above, for gastric cancer and hepatocellular carcinoma, on which the largest number of studies have been performed, very different mRNA targets are indicated as significant in different studies. The only exception is EZH2, for which involvement in interactomes with both, lncRNA and circRNA, was shown in five epithelial cancers-hepatocellular carcinoma, ovarian carcinoma, gastric cancer, laryngeal squamous cell carcinoma, and prostate cancer. The EZH2 gene has been identified in many miR-124 axes with four lncRNAs (HOXA11-AS, LINC00511, MALAT1, and XIST), and with two circRNAs (circ-TRPS1, circ_0026123). The multifaceted biological functions of this oncogenic protein and its involvement in all the most important processes of oncogenesis, such as proliferation, migration, invasion, metastasis, and increased stemming of cancer cells, have been noted. For this reason, EZH2 is currently considered a promising target for cancer therapy.
To summarize, the identification of protein target genes of miR-124 and regulatory interactomes involving long and circular ncRNAs reveals the multiple roles of miR-124 in the development and progression of epithelial cancers, which is realized through the ceRNA mechanism. Additionally, miR-124 and molecules interacting with it, demonstrate clinical significance as predictive markers, promising targets, and potential drugs for cancer therapy.