The Role of Non-Coding RNAs in Controlling Cell Cycle Related Proteins in Cancer Cells

Cell cycle is regulated by a number of proteins namely cyclin-dependent kinases (CDKs) and their associated cyclins which bind with and activate CDKs in a phase specific manner. Additionally, several transcription factors (TFs) such as E2F and p53 and numerous signaling pathways regulate cell cycle progression. Recent studies have accentuated the role of long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) in the regulation of cell cycle. Both lncRNAs and miRNAs interact with TFs participating in the regulation of cell cycle transition. Dysregulation of cell cycle regulatory miRNAs and lncRNAs results in human disorders particularly cancers. Understanding the role of lncRNAs, miRNAs, and TFs in the regulation of cell cycle would pave the way for design of anticancer therapies which intervene with the cell cycle progression. In the current review, we describe the role of lncRNAs and miRNAs in the regulation of cell cycle and their association with human malignancies.


INTRODUCTION
Cell division has a fundamental role in the development multicellular organisms. This process is accomplished through orderly sequences of happenings that together with each other make the "cell cycle". This cycle comprises precise duplication of the genome throughout the DNA synthesis stage (S phase), and separation of whole sets of chromosomes to one of the daughter cells in the mitosis stage (M phase). Two "Gap" phases also exist in the cell cycle. The first one (G 1 ) links the accomplishment of the M phase to the commencement of S phase in the succeeding cycle. G 2 splits the S and M stages. Cells residing in the G 1 phase might momentarily or enduringly exit the cell cycle and go in an inactive or blocked phase namely G 0 (1). In the mammalian cells, cell cycle is regulated by a number of proteins namely cyclin-dependent kinases (CDKs) and their associated cyclins which bind with and activate CDKs in a phase specific manner (2). Cyclins A and E are activating factors for CDK2. Cyclin B binds with CDK1. Finally, CDK4/6 is activated by cyclin Ds. Binding of cyclins with CDKs leads to phosphorylation of CDKs target proteins which finally permits progression through cell cycle (2). In addition to cyclins, Wee1 kinase and CDC25 phosphatase regulate activity of CDKs by phosphorylation and dephosphorylation reactions, respectively (3). Activity of CDKs and cell cycle progression are inhibited by a number of factors such as p15 ink4b , p16 ink4a , p18 ink4d , p21 Cip1 , p27 Kip1 , and p57 Kip2 . These factors have a specific binding affinity for cyclin-CDK complexes (4)(5)(6). Progression through each phase of cell cycle is regulated by CDKs and their associated proteins/pathways. For instance, MAPK pathway induced by growth factors enhances transcription of cyclin Ds in the G1 phase, leading to activation of CDK4/6 (7). The protein complex formed by cyclin Ds and CDK4/6 phosphorylates retinoblastoma protein (pRB), p107 and p130, in the final stages of G1 phase, thus releasing E2F and enhancing E2F-dependent expression of growth-stimulating genes (7). At the G1/S boundary, the complex constructed by cyclin E and CDK2 phosphorylates pRB and other proteins participating in the regulation of DNA replication to facilitate G1/S transition (8). Cyclin B-CDK1 complex has several functions such as accomplishment of the G2 phase processes (9), negative regulation of cytokinesis (10), and coordination of mitotic-related procedures in the nucleus and the cytoplasm (11). This complex has a number of targets such as the anaphase-promoting complex/cyclosome (APC/C) (10). In addition to CDKs, cyclins, and the APC/C which directly regulate cell cycle progression, other molecules are involved in this process. For instance, p53 functions in numerous stages to warrant that cells do not bring their abnormal DNA through cell division. It halts the cell cycle at the G1 checkpoint through stimulating synthesis of CDK inhibitor (CKI) proteins. These proteins attach CDK-cyclin complexes and inhibit their function extending the time for the activation of DNA repair system. It also induces DNA repair enzymes. If DNA damage cannot be fixed, p53 induces cell apoptosis to prevent transmission of the damaged DNA to the daughter cells (12).
Several lines of evidence point to the role of non-coding RNAs (ncRNAs) in the regulation of expression or activity of the above-mentioned proteins (13). Cyclins, CDKs and their inhibitors, are targets of regulation by ncRNAs at different levels including transcriptional and post-transcriptional levels (13). Being classified mainly based on their sizes, ncRNAs include micro RNAs (miRNAs) and long non-coding RNAs (lncRNAs). LncRNAs are longer than 200 nucleotides. FIGURE 1 | The lncRNA SNHG3 recruits EZH2 to the promoter of CDKN1A to induce H3K27me3 and decrease expression of this gene. This gene encodes the p21 protein which is an inhibitor of cyclin E/CDK2 (33). The lncRNA FOXD2-AS1 enhances recruitment of EZH2 to the promoter of CDKN1B and decreases its expression via H3K27me3. Therefore, it down-regulates p27 which is an inhibitor of cyclin D (34). These two lncRNAs promote cell progression at G1/S point. Overexpression of ROR1-AS1 has increased cell proliferation, reduced the G0/G1 phase time of cell cycle, and inhibited apoptosis. This lncRNA can bind to EZH2 and suppress expression of DUSP5 (35).  According to their structural features, lncRNAs are categorized into different classes among them are intergenic, intronic and natural antisense lncRNAs (14). In addition to their regulatory roles on gene expression at transcriptional and posttranscriptional levels, lncRNAs act as protein scaffolds to regulate interactions between proteins (15). Expressions of a number of lncRNAs are stimulated by DNA damage. These transcripts contribute in DNA damage responses and carcinogenic processes (13). Meanwhile, there is a reciprocal interaction between miRNAs and a number of cell cycle regulators in a way that miRNAs regulate expression of cell cycle regulators, and expression of miRNA is regulated by cellcycle-dependent transcription factors (16). Such interactive network is implicated in the pathogenesis of a number of disorders particularly cancer. Most of miRNAs are first transcribed from their encoding genes into primary miRNAs. Then, they are changed to precursor miRNAs, and eventually mature miRNAs. These steps are performed in both nucleus and cytoplasm. Mechanistically, miRNAs interact with the 3′ UTR of their target transcripts resulting in degradation their or inhibition of their translation. Yet, miRNAs binding with the 5′ UTR, coding regions, and promoters, has also been demonstrated (17). In the current review, we describe the role of lncRNAs and miRNAs in the regulation of cell cycle and their association with human malignancies.

LNCRNAS AND CELL CYCLE CONTROL
Numerous lncRNAs have been shown to regulate cell cycle progression directly through modulation of expression of CDKs/cyclins or indirectly through regulation of TFs that control expression of CDKs/cyclins. For instance, the known oncogenic lncRNA MALAT1 regulates cell cycle progression at G1 phase since its knock down has resulted in cell cycle arrest at this step and enhanced expression of cell cycle inhibitors p53, p16, p21, and p27, while suppressing expression of cyclin A2 and CDC25A (18). Moreover, this lncRNA has a crucial role in upregulation of expression of B-Myb, an oncogenic TF participating in G2/M progression (18). Thus, MALAT1 regulates cell cycle progression in different phases through modulation of expression of cell cycle regulators. ANRIL is an lncRNA being transcribed from the INK4 locus in an antisense direction to p15 (19). This lncRNA participates in epigenetic suppression of expression of the INK4 locus through recruitment of polycomb repression complex 2 (PRC2). Such function specifically represses expression of p15 (20). Expression of CDK inhibitors is also regulated by other lncRNAs including lncRNA-HEIH. This up-regulated lncRNA in the hepatitis B virus-associated hepatocellular carcinoma decreases expression of p15, p16, p21, and p57, through cooperation with EZH2, thus regulating cell cycle transition at G0/G1 (21).
HOTAIR has been shown to regulate expression of genes which are principally associated with cell cycle progression (22). HOTAIR silencing has led to cell cycle arrest at G0/G1 in addition to modulation of expression of cell cycle-related proteins (22). Experiment in esophageal squamous cell    carcinoma cells has also verified the impact of HOTAIR silencing on suppression of cell proliferation and induction of G1 cell cycle arrest. This lncRNA serves as a molecular sponge to suppress miR-1 expression and subsequently increase expression of cyclin D1 (23). In ovarian cancer cells, HOTAIR increases expression of cyclin D1 and cyclin D2 through negatively regulating miR-206 expression (24). A number of lncRNAs have been shown to regulate expression of p53 thus affecting cell cycle regulation by this TF. These lncRNAs include both oncogenic and tumor suppressor ones. Examples from the former group include PVT1 and ANRIL which enhance MDM2-associated degradation of p53 (25,26). On the other hand, LOC572558 enhances expression of p53 through regulation of its phosphorylation (27). Moreover, MEG3 RNA interacts with the promoter region of p21 to increase p53 accumulation (28). Meanwhile, p53 as a TF can alter expression of several lncRNAs. For example, the lncRNA-p21 has been shown to be transcribed from a genomic region adjacent to p21 Cip1 . This lncRNA is a direct target of p53. LncRNA-p21 silencing has affected expression of a number of p53-target genes with the exception of p21 gene (29). Moreover, the lncRNA PANDA is transcribed from a promoter region of p21 in response to DNA damage through a p53-dependent route (30). Therefore, lncRNAs exert _ (72) functional roles both at upstream and downstream of cell cycleassociated TFs such as p53.
The lncRNA EMS has been identified as a direct target of c-Myc. This lncRNA acts as an oncogenic transcript that facilitates G1/S transition. Functional studies have shown interaction between EMS and the RNA binding protein RALY to increase the stability of E2F1 transcript and enhance its expression (31). The oncogenic lncRNA MIR31HG has been shown to promote cell proliferation, facilitate cell cycle progression, and suppress cell apoptosis. This lncRNA modulates cell cycle transition through regulation of HIF1A and p21 expressions (32). In addition, SHNG3 is another dysregulated lncRNA in diverse cancers. This lncRNA has higher expression in glioma tissues and cell lines compared with normal counterparts. Forced overexpression of SNHG3 has increased proliferation, quickened cell cycle progression, and suppressed cell apoptosis via silencing KLF2 and p21 through recruitment of EZH2 to their promoter (33). FOXD2-AS1 is another oncogenic lncRNA whose knock down results in cell cycle arrest in the G0/G1 stage, inhibition of colony development, cell proliferation, and suppression of tumor FIGURE 2 | A schematic diagram of the regulation of mitochondrial apoptosis, Wnt/b-catenin, JAK-STAT, and PI3K/AKT signaling pathways via different miRNAs in various human cancers. Ectopic expression of some miRNAs including miR-345-5p, miR-561, miR-302b, miR-362-3p, and miR-34a could impede the mitochondrial apoptotic pathway via targeting caspase 3 and 9, Bcl-2, Bax, and Bim which can play an effective role in cell death suppression in variety of tumor cells (94,95). Besides, miR-214, miR-320, miR-188, miR-374a, and miR-574-5p could activate the Wnt/b-catenin pathway in tumor cells through modulating GSK-3b, FOXM1, CCND1, and C-myc, and thereby promoting cell differentiation and proliferation as well as enhancing EMT and cell migration and invasion in different human cancers (96,97). Additionally, miR-340, and miR-574-5p could regulate the JAK-STAT signaling pathway via targeting STAT3, SOCS3, and Survivin which have a significant role in regulating tumor cell growth and metastasis in various tumor cells (98,99). In addition, aberrant expression of miR-214, miR-106b-5p, and miR-561 could negatively modulate PTEN and PIP3 in PI3K/AKT signaling pathway in different human cancers such as ovarian cancer, melanoma, and NSCLC cells (96,100,101).  MiR-486-5p by targeting NEK2 could attenuate proliferation, and confer G0/G1 arrest and also inhibit metastasis via suppressing EMT.  growth in the xenograft model. FOXD2-AS1 and reduced expression of CDKN1B through recruitment of EZH2 to its promoter region (34). ROR1-AS1 in an up-regulated lncRNA in colon cancer tissues, particularly in stage III and IV and more massive tumors. Forced over-expression of ROR1-AS1 has increased cell proliferation, reduced the G0/G1 phase time of cell cycle, and inhibited apoptosis. This lncRNA can bind to EZH2 and suppress expression of DUSP5 (35). Figure 1 shows the molecular mechanism of involvement of a number of lncRNAs in cell cycle regulation. These lncRNAs recruit EZH2 to the promoter regions of their target genes. Table 1 shows the results of studies which assessed the role of lncRNAs in cell cycle control.
The interaction between lncRNAs and cell cycle-related proteins can alter response of cancer cells to chemotherapeutic agents. For instance TUG1 has a role in induction of chemoresistance in small cell lung cancer cells through regulation of LIMK2b expression. Knockdown of TUG1 has resulted in the accumulation of cells at G1-phase (67). NNT-AS1 via MAPK/Slug pathway could be involved in cisplatin chemoresistance in non-small cell lung cancer (68). Table 2 summarizes the results of studies which assessed the role of lncRNAs in this regard.
The importance of cell cycle-associated lncRNAs as diagnostic/prognostic markers have been assessed in several studies. Higher expression of NR2F2-AS1, PCAT6, FOXD-AS1, SNHG3, FLVCR-AS1, and some other lncRNAs has been associated with lower OS rate. Table 3 summarizes the results of these studies.

miRNAs and Cell Cycle Control
These small transcripts participate in the regulation of cell cycle control via modulation of checkpoints and DNA repair mechanisms (16,74). Moreover, they regulate expression of cyclins, CDKs, cyclin-dependent kinase inhibitors, and TFassociated proteins such as Rb (16). For instance, the miR-15a-16-1 cluster has been shown to induce cell cycle arrest at the G1 through suppressing expression of CDK1, CDK2, and CDK6 as well as D1, D3, and E1 cyclins (75)(76)(77). miR-188 suppress cell cycle transition at G1/S through inhibition of expression of cyclins D1, D3, E1, and A2 as well as CDK4 and CDK2. This miRNA also reduces Rb phosphorylation and decreases E2F transcriptional activity (78). miR-424 has been shown to regulate cell cycle progression in the G2/M phase through inhibition of expression of CDK1 probably via the Hippo and the extracellular signal-regulated kinase pathways (79). Moreover, regulation of CDK5 by miRNA-26a has been shown to control cell proliferation, apoptosis and tumor growth in an animal model of diffuse large B-cell lymphoma (80). Expression of CDK5 is also regulated by the tumor suppressor miRNA-505-5p in cervical cancer cells (81). Notably, this CDK has a distinct feature from other CDKs which is that it is not activated via interaction with cyclins. Instead, it is activated through binding with p35 and p39, or their cleaved proteins namely p25 and p29 (82).
A number of miRNAs such as miR-504 and miR-1285 regulate expression of p53 (91). Meanwhile, p53 has been shown to alter expression of several miRNAs such as miR-34a/ b/c (92,93). Thus, several miRNAs are implicated in the regulation of cell cycle progression through p53-mediated pathways. Figure 2 illustrates the role of a number of miRNAs in the cell cycle control. Table 4 summarizes the function of miRNAs in cell cycle control.
The interaction between miRNAs and cell cycle controlling proteins has implications in defi ning response to chemotherapeutic agents. For instance miR-192/miR-215 and miR-320 could affect response of cancer cells to 5-Fluorouracil resistance (151,152). In addition, miR-100 could increase cisplatin sensitivity, inhibit cell proliferation, induce conversion from G1 to S phase and promote apoptosis through directly targeting mTOR and PLK1 (140). miR-374a is another miRNA which alters chemoresistance phenotype in nasopharyngeal carcinoma. In this type of cancer, miR-374a decreases proliferation, migratory aptitude, invasiveness, metastatic ability, and resistance to cisplatin. Functionally, miR-374a decreases expression of CCND1 to attenuate activity of the pPI3K/pAKT/c-JUN axis through making a negative-feedback circle. This miRNA also inhibits downstream signals associated with cell cycle transition (144). miR-9600 has been demonstrated to attenuate tumorigenesis and metastatic potential of lung cancer cells via inhibiting expression of STAT3. Besides, miR-9600 improved response of cancer cells to paclitaxel and cisplatin through this axis and enhancement of chemotherapy-associated apoptosis (153). Besides, miR-106b-5p has a crucial impact in modulation of cisplatin resistance in lung cancer through inhibiting expression of PKD2 (154). Table 5 summarizes the data regarding the role of miRNAs in conferring resistance to chemotherapeutic agents.

DISCUSSION
Several lncRNAs and miRNAs have been shown to regulate cell cycle at different stages thus influencing the proliferation rate. Abnormal function of these transcripts might lead to the development of human cancers. Cell cycle is regulated by several lncRNAs through epigenetic modulation of gene expression regulation of transcription factors modulation of translation mRNA stability, and enhancement of proteinprotein interactions (15). Recruitment of EZH2 to the promoter region of target genes is a common mechanism of action of some of these lncRNAs. Dysregulation of cell cyclerelated lncRNAs is regarded as a hallmark of cancer. A number of these lncRNAs also contribute in controlling the proliferation rate of normal differentiated cells or during organogenesis, therefore being important in the context of regenerative medicine and in cell senescence studies. For instance, Malat1 controls differentiation of myogenic cells and muscle regeneration (168). In addition, H19 regulates myoblast differentiation and muscle regeneration (169). However, most of the cell-cycle regulatory roles of lncRNAs have been assessed in the context of cancer. miRNAs also exert functional roles in the regulation of cell cycle. A comprehensive genome-wide screen of cell cycleassociated miRNAs has led to identification of a distinctive group of miRNAs that target almost all cyclins/CDKs. These miRNAs are extremely potent in impeding cancer cell proliferation (170). Systemic administration of a number of these miRNAs using nanoparticle delivery methods has repressed tumor progression in a number of xenograft models indicating the role of these miRNAs in the treatment of cancer (170). The prominent role of cell cycle-regulatory miRNAs in the pathogenesis of cancer has been further highlighted by the observed dysregulation of these transcripts in the cancer stem cells (CSCs) (171). These miRNAs have been shown to target several genes that regulate cell cycle progression among them are PTEN (172), JAK1, SOX4, STAT3, AKT, EZH1, HMGA2 (173), CDK4/6, NOTCH1 (174), and ZEB1/2 (175). Therefore, cell-cycle regulatory miRNAs represent important targets for intervention with the invasive and metastatic properties of cancer cells which are associated with CSC phenotypes. Furthermore, identification of miRNAs with distinctive functions in CSCs and normal stem cells would facilitate design of specific targeted therapies for cancer patients with fewer side effects in normal tissues. This research avenue needs to be explored in future studies.
As lncRNAs have important roles in the regulation of activity of miRNAs through serving as molecular sponges for these small transcripts, identification of this type of interactions between cell cycle-related lncRNAs and miRNAs would pave the way for better recognition of the molecular mechanism of cell cycle progression. Several lncRNAs act as competing endogenous RNAs for miRNAs, thus regulating expression of cell cycleassociated miRNAs. Examples include (but not limited to): loc285194/miR-211 (176), HOTAIR/miR-1 (23), HOTAIR/ miR-206 (24), and ANRIL/miR-384 (177). In addition to this kind of interaction between lncRNAs and miRNAs, some lncRNAs serve as precursors for miRNAs. Such situation exists between H19 and miR-675 (178). While E2F1 enhances expression of H19 lncRNA (179), miR-675 suppresses pRB expression (180). In turn, pRB inhibits E2F-associated transcription of H19 constructing a self-regulated network between H19 and pRB (13). These examples obviously show the complex network between lncRNAs, miRNAs, and TFs.
Expression of cell cycle-associated non-coding RNAs directly influences the survival of patients with diverse cancer types. This speculation is based on the obtained data from both high and low throughput studies. An example of former type of studies is a study which assessed RNA seq data of a large cohort of patients with colorectal cancer. Authors have reported several differentially expressed cell cycle genes and miRNAs which regulate expression of these genes. Subsequently, they verified correlations between expression levels of these genes/miRNAs and patients' survival (181). Moreover, these non-coding RNAs can contribute in the construction of diagnostic panels for diverse types of cancers.
Taken together, cell cycle-associated lncRNAs/miRNAs are potential therapeutic targets for management of cancer and possible biomarkers for prediction of cancer course. However, Ghafouri-Fard et al. Cell Cycle and Non-Coding RNA lack of specificity in the regulatory roles of some of these transcripts limits their application in the clinical settings. Future studies should focus on identification of the network between these two kinds of transcripts and TFs using high throughput techniques. The results of these studies would fulfill the prerequisite step for design of targeted therapies.

AUTHOR CONTRIBUTIONS
SG-F and MT wrote the manuscript and contributed in study design. HS and FTA collected the data and designed the tables. All authors contributed to the article and approved the submitted version.