MicroRNAs: Diverse Mechanisms of Action and Their Potential Applications as Cancer Epi-Therapeutics

Usually, miRNAs function post-transcriptionally, by base-pairing with the 3′UTR of target mRNAs, repressing protein synthesis in the cytoplasm. Furthermore, other regions including gene promoters, as well as coding and 5′UTR regions of mRNAs are able to interact with miRNAs. In recent years, miRNAs have emerged as important regulators of both translational and transcriptional programs. The expression of miRNA genes, similar to protein-coding genes, can be epigenetically regulated, in turn miRNA molecules (named epi-miRs) are able to regulate epigenetic enzymatic machinery. The most recent line of evidence indicates that miRNAs can influence physiological processes, such as embryonic development, cell proliferation, differentiation, and apoptosis as well as pathological processes (e.g., tumorigenesis) through epigenetic mechanisms. Some tumor types show repression of tumor-suppressor epi-miRs resulting in cancer progression and metastasis, hence these molecules have become novel therapeutic targets in the last few years. This review provides information about miRNAs involvement in the various levels of transcription and translation regulation, as well as discusses therapeutic potential of tumor-suppressor epi-miRs used in in vitro and in vivo anti-cancer therapy.


Introduction
Although research into RNA biology has been ongoing for more than two decades, almost each year brings new discoveries. Until recently, it was thought that microRNAs (miRNAs) act mainly in the cytoplasm at the post-transcriptional level. Interestingly, miRNAs can exert regulatory effect both in the cell (i.e., cytoplasm and nucleus) in which they are produced and in neighboring cells. The latter intracellular transfer of miRNA is mediated by gap junction channels or exosomes [1]. Interestingly, mature miRNAs can regulate one or more mRNA targets, but also a single mRNA transcript can be bound and regulated by many different miRNAs. It is estimated that each miRNA can recognizẽ 100-200 target sites of the transcriptome and the inhibitory effect on expression can be achieved at 1000 copies per cell [1,2]. miRNAs can recognize and bind to 3 UTR, 5 UTR and coding sequence of their targets' mRNA, as well as to promoter regions. Considering miRNAs variety and localization, cell type and cell state, their possibilities to regulate gene expression are limitless.

Inhibition or Activation of Translation
Mature miRNAs (mainly guide strands) form a complex with Argonaut (AGO) proteins called miRNA-induced silencing complex (miRISC) which interact with other proteins including DICER, TRBP, PACT and GW182. The miRNA specific region called 'seed sequence' (which includes nucleotides between 2 and 8, counting from the 5 end of the miRNA) base-pairs with miRNA recognition elements Figure 1. Transport of mature miRNAs and components of RISC (RNA-induced silencing complex) into the nucleus. TNRC6A is shuttled from the cytoplasm into the nucleus either via its own NLS sequence when it interacts with miRNA-AGO complex or independently via its interaction with Importin β (Imp β) and Importin α (Impα). While mature miRNAs loaded into AGO-2 are translocated into the nucleus by Importin 8 (IPO8) miRNA-AGO-TNRC6A complex can be exported back to the cytoplasm by Exportin 1 (XPO1). In the nucleus, miRISC will interact with promoters or enhancers leading to transcriptional gene silencing (TGS) or transcriptional gene activation (TGA). The putative miRNA recognition elements (MREs) could be recognized by miRNAs that mediate chromatin silencing complex assembly or de novo DNA methylation at the promoter region resulting in compact, silent heterochromatin and TGS. Unlike, when miRISC interacts with TATA-box motifs enhancing promoter activities leading to TGA through enrichment of chromatin-remodeling factors and active chromatin marks. Moreover, miRNAs interaction with enhancers result in TGA through chromatin remodeling and the enrichment of active marks at enhancer regions. NLS-nuclear localization signal sequence.
Moreover, it is proposed that miRNA nuclear localization can also be controlled by nuclear localization signal sequences in miRNA molecules or full processing of pre-miRNAs in the nucleus. Several studies show that various motifs, including AGUGUU-motif, 5'-UUGCAUAGU-3' and 5'-AGGUUGKSUG-3' motifs (where K is a uridine or a guanine) as well as the consensus ASUS sequence (where S is a cytosine or a guanidine) are presented in many miRNAs and are engaged in the nuclear translocation [27][28][29]. It is supposed that miRNAs translocation is controlled by RNAbinding proteins (RBPs), however, molecular pathways are now recognized. Regarding processing of pre-miRNA molecules and their loading into nuclear RISC complex, there are many uncertainties that need to be investigated.
Although the functions of nuclear miRNAs have not been fully elucidated, it is suggested that they can regulate both transcriptional rates and post-transcriptional levels of mRNAs. miRNApromoter interaction mediated by AGO proteins may either suppress or activate transcription depending on the location of their target region and epigenetic status of the promoter [22,30]. Genome-wide analysis revealed that human promoters contain miRNA-seed matching sites, suggesting that miRNA-mediated transcription regulation is likely to be a common phenomenon [31]. On the one hand, Benhamed et al. demonstrated that AGO-2 and let-7f are involved in the transcriptional repression of proliferation-promoting genes regulated by the retinoblastoma (Rb)/E2F repressor complex in senescence [32]. The putative MREs for the let-7f have been localized in the promoters of two E2F-target genes CDC2 and CDCA8. Similarly, nuclear miR-522 suppresses transcription of CYP2E1 gene by interacting with its promoter forming a DNA:RNA hybrid which Figure 1. Transport of mature miRNAs and components of RISC (RNA-induced silencing complex) into the nucleus. TNRC6A is shuttled from the cytoplasm into the nucleus either via its own NLS sequence when it interacts with miRNA-AGO complex or independently via its interaction with Importin β (Imp β) and Importin α (Impα). While mature miRNAs loaded into AGO-2 are translocated into the nucleus by Importin 8 (IPO8) miRNA-AGO-TNRC6A complex can be exported back to the cytoplasm by Exportin 1 (XPO1). In the nucleus, miRISC will interact with promoters or enhancers leading to transcriptional gene silencing (TGS) or transcriptional gene activation (TGA). The putative miRNA recognition elements (MREs) could be recognized by miRNAs that mediate chromatin silencing complex assembly or de novo DNA methylation at the promoter region resulting in compact, silent heterochromatin and TGS. Unlike, when miRISC interacts with TATA-box motifs enhancing promoter activities leading to TGA through enrichment of chromatin-remodeling factors and active chromatin marks. Moreover, miRNAs interaction with enhancers result in TGA through chromatin remodeling and the enrichment of active marks at enhancer regions. NLS-nuclear localization signal sequence.
Moreover, it is proposed that miRNA nuclear localization can also be controlled by nuclear localization signal sequences in miRNA molecules or full processing of pre-miRNAs in the nucleus. Several studies show that various motifs, including AGUGUU-motif, 5 -UUGCAUAGU-3 and 5 -AGGUUGKSUG-3 motifs (where K is a uridine or a guanine) as well as the consensus ASUS sequence (where S is a cytosine or a guanidine) are presented in many miRNAs and are engaged in the nuclear translocation [27][28][29]. It is supposed that miRNAs translocation is controlled by RNA-binding proteins (RBPs), however, molecular pathways are now recognized. Regarding processing of pre-miRNA molecules and their loading into nuclear RISC complex, there are many uncertainties that need to be investigated.
Although the functions of nuclear miRNAs have not been fully elucidated, it is suggested that they can regulate both transcriptional rates and post-transcriptional levels of mRNAs. miRNA-promoter interaction mediated by AGO proteins may either suppress or activate transcription depending on the location of their target region and epigenetic status of the promoter [22,30]. Genome-wide analysis revealed that human promoters contain miRNA-seed matching sites, suggesting that miRNA-mediated transcription regulation is likely to be a common phenomenon [31]. On the one hand, Benhamed et al. demonstrated that AGO-2 and let-7f are involved in the transcriptional repression of proliferation-promoting genes regulated by the retinoblastoma (Rb)/E2F repressor complex in senescence [32]. The putative MREs for the let-7f have been localized in the promoters of two E2F-target genes CDC2 and CDCA8. Similarly, nuclear miR-522 suppresses transcription of CYP2E1 gene by interacting with its promoter forming a DNA:RNA hybrid which probably prevents binding of Pol II and transcription factor [33]. On the other hand, Zhang et al. revealed that several miRNAs, such as let-7i, miR-138, miR-92a and miR-181d bind to the TATA-box motifs and enhance the promoter activities of interleukin-2, insulin, calcitonin or c-Myc, respectively [34]. Also, Cyclin B gene has a sequence located in its promoter that interacts with miR-744-5p and miR-466d-3p leading to transcriptional upregulation [35]. A recent study has revealed, that miRNAs (miR-26a-1, miR-339, miR-3179, miR-24-1 and miR-24-2) are able to induce expression of neighboring genes and function as enhancer (cis-acting DNA elements) regulators [36]. Moreover, this study has also shown that miR-24-1 (located in the enhancer region) increases expression of FBP1 and FANCC genes and triggers direct chromatin state alteration of the FBP1 enhancer that activate transcription. Another notable fact is that transcriptional gene silencing (TGS) and transcriptional gene activation (TGA) can be achieved by miRNA-mediated epigenetic regulation. Indeed, miRNA directs the RNA-induced transcriptional silencing complex (RITS), which consists of chromatin remodeling enzymes (e.g., HDAC1, EHMT2 and EZH2) and DNA methyltransferase (DNMT3A), to promoter leading to the transition of active chromatin structure to silent heterochromatin [31]. According to the study carried out by Kim and co-workers miR-320 directs to the promoter region AGO-1 that acts as the effector protein for transcriptional silencing of POLR3D gene [37]. Furthermore, simultaneous enrichment of tri-methyl histone H3 lysine 27 (H3K27me3, a repressive chromatin mark) and EZH2, a histone methyltransferase that mediates H3K27me3, has been observed at the POLR3D promoter [37]. Another study has revealed that miR17-5p and miR20a, encoded within a poly-miRNA cluster miR-17-92, are involved in the acquisition of heterochromatin marks at the promoters through seed-paring manner [38]. miRNA-mediated TGS is involved in cell differentiation processes. For example, during granulopoiesis miR-223-RISC interaction with the promoter of nuclear factor I-A (NFI-A) results in the recruitment of Polycomb group complex and histone-modifying enzymes that repress transcription of NFI-A, an important step for granulocytic differentiation [39]. It is postulated that specific miRNA can initiate TGS through de novo DNA methylation or chromatin modification in human cancer cells. In fact, miR-10a with AGO-1 and AGO-3 reduces HOX4 expression in human breast cells mediating in de novo DNA methylation and accumulation of repressive chromatin marks (H3K27me3 and H3K9me2, di-methyl histone H3 lysine 9) at its promoter [40].
In contrast, AGO-miRNA complex may activate the expression of target loci by either disruption of the recruitment of silencing proteins (e.g., PRC2) to lncRNAs (long non-coding RNAs) or recruitment of protein complex containing transcriptional activators (e.g., transcription factors) [31,41]. In the nucleus, lncRNAs regulate epigenetic silencing of adjacent genes through recruiting chromatin-remodeling factors in close proximity of their promoters [42]. In case of miR-744 and Ccnb1 gene, miRNA-mediated TGA rely on the recruitment of AGO proteins and RNA Pol II enrichment as well as active chromatin marks (such as H3K4me3, tri-methyl histone H3 lysine 4) at the regulated gene promoters [35]. Moreover, miR-373 activates transcription of E-cadherin and CSDC2 genes only via enrichment of RNA Pol II at their promoters [43], while miR-205 induces the expression of IL24 and IL32 tumor suppressor genes by targeting specific sites in their promoters as well enrichment of RNA Pol II and active chromatin modifications [44].
Similar to cytoplasmic miRNAs, nuclear miRNAs can also mediate post-transcriptional gene silencing (PTGS) inducing degradation of target mRNAs. Several studies suggest that miRNAs contribute to the regulation of miRNA precursors and lncRNA transcripts [31]. For instance, mouse nuclear miR-709 is involved in the post-transcriptional regulation of the pri-miR-15a/miR-16-1, binding to a 19-nt recognition element and preventing processing of primary transcripts, thus, nuclear miRNAs can influence the biogenesis of other miRNAs suggesting hierarchical structures among miRNAs [45]. Furthermore, some nuclear-retained lncRNAs are also regulated by AGO-miRNA complexes that interact with miRNA-complementary sequences located in lncRNAs, thus impairing their stability and function [42]. Indeed, the highly abundant lncRNA, metastasis associated lung adenocarcinoma transcript 1 (MALAT1), has two MRE's which are recognized and bound by miR-9 [46]. Subsequently, putative miR-675-5p binding site within H19 RNA transcripts has been identified and the overexpression of miR-675-5p significantly downregulated the level of the H19 transcript [47]. So far, several other non-coding RNAs directly targeted by miRNAs have been identified. Interestingly, a long non-protein coding RNA involved in mammalian X-chromosome inactivation, X (inactive)-specific transcript (XIST), has seed-paring sites for miR-210 which modulates its RNA level [48]. Additionally, miR-671 directs AGO2-mediated cleavage of a circular antisense transcript of the CDR1 gene and negatively regulates this non-coding antisense transcript [49].

Regulation of Alternative Splicing
miRNAs are able to indirectly modulate alternative splicing by regulating translation of various splicing factors. However, mounting evidence suggests that AGO-miRNA complexes can affect the regulation of alternative splicing directly in the nucleus by epigenetic and non-epigenetic mechanisms. A co-immunoprecipitation study has identified multiple AGO-associated splicing factors, moreover, AGO-1, AGO-2 and DICER1 knockdown and overexpression experiments confirmed their involvement in splicing decisions at alternatively spliced exons [50,51]. Advanced molecular analyses were able to identify miRNA binding sites within intronic sequences in mouse and human brain as well as in human myocardial cells [12,52,53]. It is proposed that miRNAs-mediated compaction of chromatin structure at specific exon-intron junctions slows the rate of RNA Pol II elongation, which favors exon inclusion [54]. Surprisingly, exon skipping can be achieved by single-stranded oligonucleotides (ss-siRNA), ss-siRNA is incorporated by AGO-2 in the cytoplasm, then is transported into the nucleus where AGO2-ss-siRNA complex binds to the target mRNA and disrupts association with the splicing machinery [55].
Taken together, the above considerations illustrate the complex regulatory mechanisms of miRNA-mediated gene expression in the cytoplasm and the nucleus. It should be emphasized, that miRNAs are involved in many crucial cellular regulatory processes and may activate or inhibit gene expression at both transcriptional and post-transcriptional level. Thus, deregulation of miRNAs biogenesis and function can disrupt these processes and finally lead to a wide range of human diseases. Hence, miRNAs are valuable as diagnostic and prognostic biomarkers for many diseases, including cancer, diabetes mellitus, cardiovascular pathologies and neurological disorders. Moreover, miRNAs are considered as molecular targets of novel therapies and treatment strategies.

miRNAs As Potential Cancer Epi-Therapeutics
Over the past few decades growing evidence has linked epigenetic mechanisms with the regulation of gene expression. Epigenetic markers such as DNA methylation and post-translational modifications of histone tails can rearrange the structure of chromatin leading either to activation or repression of transcription activity (for details see reviews [56,57]). It is interesting that not only nucleotide sequences determine the level of gene expression but also epigenetic modifications are involved in this process. Epigenetic processes are orchestrated by multiple proteins (e.g., DNA methyltransferases, DNA demethylases and histone modifying enzymes), non-coding RNAs (e.g., miRNAs and lncRNAs) and environmental factors. Typically, loss of DNA methylation (hypomethylation) turns on gene transcription by altering the structure of chromatin. In turn, too much DNA methylation (hypermethylation) induces chromatin compaction and hinders the expression of genes. Therefore, disruption of epigenetic regulation can lead to inappropriate gene expression that impairs crucial biological processes resulting in the development of "epigenetic diseases". The first "epigenetic disease" was cancer and it was established that patients with colorectal cancer had less DNA methylation levels in cancer tissues than from their normal tissue [58]. Growing evidence suggests that epigenetic changes, unlike DNA sequence mutations, are reversible, so it seems that these changes can be an ideal target for epigenetic treatments.
Recently, a subclass of miRNAs, referred to as epi-miRNAs, that influence the expression of genes encoding epigenetic effector and reader proteins, has been identified [59]. Due to the important role of epi-miRs in the modulation of the epigenome, they are currently considered as potential therapeutic targets, especially in cancer. Manipulation of epi-miRs can affect the expression of epigenetically-regulated genes, such as oncogenes and/or tumor suppressor genes, involved in important cellular pathways including DNA replication, cell cycle progression and apoptosis [60,61]. The two types of miRs, oncomiRs and tumor-suppressor miRs, can be distinguished regarding their role in carcinogenesis. Generally, oncomiRs are up-regulated thereby increasing cancer cell proliferation and metastasis, in contrast the expression of tumor-suppressor miRs are down-regulated leading to enhanced tumorigenesis [62]. In this review, we focus on the therapeutic potential of tumor-suppressor epi-miRs that are downregulated in various types of cancer (casi el tinc. Emerging studies found that the decreased levels of epi-miRs promote cell proliferation, colony formation, tumor growth and metastasis [62][63][64]. Moreover, the suppression of some epi-miRs are responsible for the drug resistance of cancer cells [64,65]. Schematic relationship between downregulated tumor-suppressor epi-miRs, chromatin-modifying enzymes and cellular processes is shown in Figure 2.
Biomolecules 2020, 10, x FOR PEER REVIEW 6 of 23 The two types of miRs, oncomiRs and tumor-suppressor miRs, can be distinguished regarding their role in carcinogenesis. Generally, oncomiRs are up-regulated thereby increasing cancer cell proliferation and metastasis, in contrast the expression of tumor-suppressor miRs are downregulated leading to enhanced tumorigenesis [62]. In this review, we focus on the therapeutic potential of tumor-suppressor epi-miRs that are downregulated in various types of cancer (casi el tinc . Emerging studies found that the decreased levels of epi-miRs promote cell proliferation, colony formation, tumor growth and metastasis [62][63][64]. Moreover, the suppression of some epi-miRs are responsible for the drug resistance of cancer cells [64,65]. Schematic relationship between downregulated tumor-suppressor epi-miRs, chromatin-modifying enzymes and cellular processes is shown in Figure 2. To date, several causes have been found that influence the activity of miRNAs, their downregulation is coupled with epigenetic silencing or genomic abnormalities, such as gene amplification, deletions and microdeletions (e.g., at miR-101-1 loci) as well as mutations and chromosomal rearrangements [66,67]. Considering, the drug resistance of cancer chemotherapy (i.e., doxorubicin, cisplatin, paclitaxel), which are related to down-regulation of epi-miRs, their enforced expression appears to be an interesting approach to restore drug sensitivity (Table 1). Fabbri and co-workers revealed that the miR-29 family (29a, -b, and -c) act as tumor suppressor miRs in lung cancer and regulate transcript levels of DNMT3A and DNMT3B [68]. Moreover, it has been established that synthetic epi-miR, miR-29b oligonucleotides, potentiates a hypomethylating effect of DNMT1 inhibitors (decitabine or azacitidine) resulting in better AML response for treatment probably due to the inhibition of other DNMT isoforms that are not efficiently suppressed by these agents [69]. Another study showed that synthetic miR-29b mimics inhibit HDAC4 expression in multiple myeloma cell lines, reduce migration potential and increase apoptosis, therefore, this approach could offer a novel targeted therapy [70]. In addition, a recent study has shown that miR-148a combination therapy with either cisplatin or doxorubicin significantly enhanced apoptosis in urothelial cell carcinoma of the bladder cell lines [71]. Importantly, cancer stem cells (CSCs) are characterized by To date, several causes have been found that influence the activity of miRNAs, their down-regulation is coupled with epigenetic silencing or genomic abnormalities, such as gene amplification, deletions and microdeletions (e.g., at miR-101-1 loci) as well as mutations and chromosomal rearrangements [66,67]. Considering, the drug resistance of cancer chemotherapy (i.e., doxorubicin, cisplatin, paclitaxel), which are related to down-regulation of epi-miRs, their enforced expression appears to be an interesting approach to restore drug sensitivity (Table 1). Fabbri and co-workers revealed that the miR-29 family (29a, -b, and -c) act as tumor suppressor miRs in lung cancer and regulate transcript levels of DNMT3A and DNMT3B [68]. Moreover, it has been established that synthetic epi-miR, miR-29b oligonucleotides, potentiates a hypomethylating effect of DNMT1 inhibitors (decitabine or azacitidine) resulting in better AML response for treatment probably due to the inhibition of other DNMT isoforms that are not efficiently suppressed by these agents [69]. Another study showed that synthetic miR-29b mimics inhibit HDAC4 expression in multiple myeloma cell lines, reduce migration potential and increase apoptosis, therefore, this approach could offer a novel targeted therapy [70]. In addition, a recent study has shown that miR-148a combination therapy with either cisplatin or doxorubicin significantly enhanced apoptosis in urothelial cell carcinoma of the bladder cell lines [71]. Importantly, cancer stem cells (CSCs) are characterized by their ability to self-renew and resistance to standard chemotherapy, during remission can regenerate a tumor identical to the original one. An elegant study by Iliopoulos and colleagues uncovered that the combinatorial therapy of doxorubicin with epi-miR (miR-200b) was more effective than doxorubicin alone, blocking tumor growth and preventing relapse [72]. Interestingly, many natural agents, such as resveratrol, curcumin and glabridin used for epigenetic therapy, among others, exert their potent anti-tumor effects by enhancing expression of epi miRs. For instance, resveratrol causes up-regulation of miR-137 in neuroblastoma tumors [73], curcumin increases levels of miR-29a and miR-185 in hepatocellular cancer cells [74], in turn glabridin potentiates expression of miR-148a in breast cancer cells [75].
Epi-miRs-targeted cancer therapy seems to be a promising approach since it is able to influence not only a single gene, but multiple pathways. It is possible to re-establish expression of epi-miRs by delivering synthetic miR mimics (double stranded RNA oligonucleotides directly loaded into RISC) or chemically modified poly(nucleic acids), however, cellular uptake of free synthetic miRs are limited because of the ease in which they a degraded in biofluids [76]. In order to overcome poor in vivo stability and improve efficient and specific-site delivery of miRs to the tumor, innovative delivery systems are required. Currently, both viral and non-viral systems are used to increase stability of miRNA oligonucleotides and enhance their therapeutic effect. Administration of epi-miRs via viral vectors (e.g., adenoviruses, adeno-associated viruses (AAV) or lentiviruses) is very effective, as shown by systemic intravenous injection of epi-miR, miR-26a, packaged into AAV vector, which inhibited progression of hepatocellular carcinoma in a mouse model [77]. However, due to the viral vectors possible toxicity and immunogenicity their use in clinical practice is limited. In this context, non-viral systems seem to be more promising, because of the control of their molecular composition, ease in manufacturing and relatively low immunogenicity. Different delivery systems including, lipid-based delivery system, synthetic polymers (e.g, polyethyleneimine (PEI)) and naturally occurring polymers (e.g., chitosan, protamine and atelocollagen) are applied to protect miRs from degradation (for details see the review [78]). For example, a novel transferrin-conjugated nanoparticle delivery system for synthetic epi-miR, miR-29b, was injected intravenously and significantly prolonged leukemic mice survival [79]. Despite significant advances made in delivery systems of miRs, substantial improvements will be necessary for achieving site-specific delivery.
The discovery of therapeutic epi-miRs potential in cancer therapy makes them attractive candidates for next-generation cancer treatment. It therefore seems likely that profiling of miRs expression and then using appropriate epi-miR-based therapeutics may revolutionize cancer treatment by enabling the reversal of the epigenetic program of tumor cells to a more normal state.
However, we realize that turning epi-miR-based therapy into clinical practice faces challenges. Indeed, some clinical trials with miRNA drugs have not always produced satisfactory results. For example, the FDA (Food and Drug Administration) halted phase I clinical trial of miR-34a mimic (drug: MRX34) used in patients with different types of cancer. Double-stranded miR-34a was encapsulated into a liposome-formulated nanoparticle and administered intravenously [80]. Although, preclinical studies were promising, immune-related serious adverse events (SAEs) appeared during phase I. Due to SAEs this clinical trial was terminated and future phase II trials of MRX34 for melanoma were withdrawn. In contrast human trial of miR-16 mimic (drug: MesomiR-1) exhibited hopeful results in patients with pleural mesothelioma and non-small cell lung cancer. Double-stranded miR-16 was delivered by non-living bacterial minicells with a targeting moiety (i.e., an anti-EGFR bispecific antibody that recognizes EGFR-expressing cancer cells) [81]. This is a new targeted therapy known as TargomiRs. The successful completion of phase I trial confirmed safety and early signs of antitumor activity of TargomiRs so phase II of the trial is expected to begin soon.   ND miR-29b/c overexpression decreases migration and reduces invasive ability; miR-29b/c suppresses the expression of DNMT3A. [95] miR-146a miR-146b UHRF1 GC9811, GC9811-P, MKN28NM and MKN28M pre-miR-146a/b; lentivirus vector miR-146a/b; luciferase reporter assay (in HEK293T and GC9811 cells) Nude mice; metastasis assay: GC9811-P cells infected with miR-146a/b were injected into the tail vein.
pre-miR-148a and pre-miR-152 decreases DNMT-1 protein expression and reduces cell proliferation; miR-148a and miR-152 expression was reduced in tumor cell xenografts in vivo.

Conclusions
Knowledge in the miRNA field is steadily increasing and recent information about the mechanisms of action, especially their involvement in epigenetic regulation has shed new light on cellular regulatory networks.
Interestingly, mature miRNAs are present in both the nucleus and the cytoplasm, therefore they can be involved in the regulation of transcription and translation processes. Nuclear miRNAs can influence gene expression via transcriptional activation or transcriptional gene silencing and shaping alternative splicing, Cytoplasmic miRNAs mainly mediated translation inhibition, however, some miRNAs are capable of activating translation of their target mRNA. A growing body of evidence suggests that miRNAs can act as regulators of the cell epigenome through translation inhibition of proteins engaged in epigenetic control and/or interaction with lncRNA. Considering the pervasive role of miRNAs in numerous biological processes, especially tumorigenesis, better understanding of their role in epigenetic regulation will aid the development of new therapeutic strategies.
Currently, miRNA-based treatment approaches for cancer, including tumor-suppressor epi-miRs, are tested in in vitro and in vivo experiments. Although results seem promising further studies will be needed to clarify the safety and effectiveness of epi-miR therapy in clinical practice. We strongly believe that re-introduction of tumor-suppressor epi-miRs will allow for more effective, personalized therapies in the near future.