Cell cycle regulation by long non-coding RNAs
- 3.4k Downloads
The mammalian cell cycle is precisely controlled by cyclin-dependent kinases (CDKs) and related pathways such as the RB and p53 pathways. Recent research on long non-coding RNAs (lncRNAs) indicates that many lncRNAs are involved in the regulation of critical cell cycle regulators such as the cyclins, CDKs, CDK inhibitors, pRB, and p53. These lncRNAs act as epigenetic regulators, transcription factor regulators, post-transcription regulators, and protein scaffolds. These cell cycle-regulated lncRNAs mainly control cellular levels of cell cycle regulators via various mechanisms, and may provide diversity and reliability to the general cell cycle. Interestingly, several lncRNAs are induced by DNA damage and participate in cell cycle arrest or induction of apoptosis as DNA damage responses. Therefore, deregulations of these cell cycle regulatory lncRNAs may be involved in tumorigenesis, and they are novel candidate molecular targets for cancer therapy and diagnosis.
KeywordslncRNA DNA damage response Cyclin-CDK CDK inhibitor pRB p53
CDKs and their related pathways control the cell cycle by maintaining exit and entry to the different phases of the cell cycle. In the G1 phase, growth stimuli such as growth factors often activate the MAP kinase pathway, following which genes encoding the cyclin Ds are transcribed. The resulting products bind to and activate CDK4/6 . Cyclin Ds–CDK4/6 complexes phosphorylate retinoblastoma protein (pRB) and its family members, p107 and p130, in the late G1 phase and activate E2F-mediated transcription, which induces the expression of several growth-promoting genes [7, 8]. At the G1/S transition point, cyclin E-CDK2 phosphorylates pRB as well as several proteins involved in DNA replication to promote G1/S progression . Cyclin B-CDK1 has many targets including APC/cyclosome, and promotes maturation of the G2 phase and critically participates in M phase events .
The cellular levels of cell cycle regulators such as cyclins, CDKs, CDK inhibitors, CDC25, RB, and E2F are critical for cell cycle regulation. After the cell cycle regulators complete their functions, they are ubiquitylated by specific E3 ligases and eliminated via the ubiquitin–proteasome pathway [11, 12, 13]. The level of cell cycle regulators is precisely controlled by not only post-translational but also translational mechanisms. For example, several micro-RNAs (miRNAs) participate in cell cycle regulation through translational regulation . MiRNAs are small non-coding RNA molecules containing 22 nucleotides, and negatively regulate translation through binding of the untranslated region of its target mRNAs . The let-7 miRNA family negatively regulates cyclins A and D, and CDK4/6 and CDC25A . The miR-15 family also inhibits the translation of cyclin D, CDK4, and CDC27 [17, 18]. Interestingly, these let-7 and miR-15 family members may be involved in tumorigenesis since they are downregulated in various human cancers [16, 17, 18]. Alternatively, cyclin D1 is a target for not only let-7 and miR-15 miRNAs but also miR-19a, 26a, and 34a . Furthermore, p27 Kip1 is targeted for regulation by the miR-181 family  and the miR-221 family . The roles of other miRNAs in the expression of cell cycle regulators have also been reported . Thus, it has been shown that the cell cycle regulators are critically and precisely controlled by E3 ligases and miRNAs both post-translationally and at the translational level.
LncRNAs involved in the cell cycle control
How it is induced
The effects of the lncRNA on its targets in cell cycle (phase)
Suppression of Cyclin D1 transcription with TLS (G1)
Destabilization of CDK6 mRNA (G1)
High expression In cancer
Promotion of cell-cycle regulators such as cyclin A2 and B1 (G1 and G2/M)
Suppression of Cyclin A, B cdc20. Cdt1 transcription (G1 and G2/M)
Suppression of p15/p16 transcription with PRC1/2 (G1)
High expression In HCC
Suppression of p16. p21, p27 and p57 transcription with PRC2(G0/G1)
Suppression of p18 expression (G1)
Suppression of p57 transcription with PRC2 and G9a (G1?}
Downregulation of RB mRNA via miR675 (G1)
Suppression of p53 mRNA translation (G2/M)
Promotion of p53 target genes transcription (G1?)
Growth inhibition by suppression of miR211 (G1?)
Suppression of transcription of the target genes involved in apoptosis and cell cycle with hnRNA-K (G1?) suppression of β-catenin and Jun B mRNA translation
Suppression of FAS and BIK transcription (G1?)
LncRNAs regulating cyclins and CDKs
Gadd7 is an lncRNA involved in regulating CDK6 expression  in a posttranscriptional manner. TDP-43 (TAR DNA binding protein) binds to the 3′ untranslated region of CDK6 mRNA to stabilize it. Gadd7 is transcriptionally induced via DNA damage mediated by UV and cisplatin [35, 36], and binds to TDP-43 and dissociates from CDK6 mRNA. The CDK6 mRNA is then degraded, resulting in inhibition of the G1/S transition (Fig. 2b). Therefore, gadd7 negatively controls CDK6 expression, functioning as a translation regulator. Interestingly, gadd7 specifically controls mRNA stability for CDK6, but not CDK4, CDK2, or CCND1, by trapping TDP-43. The physiological relevance of the selective suppression of CDK6 remains to be determined. Gadd7 may be involved in the G1 checkpoint by collaborating with the lncRNA, ncRNACCND1, to downregulate the cyclin D1–CDK6 complex, thereby arresting cell cycle progression in response to DNA damage (Fig. 2a, b). This may represent a novel G1-checkpoint cascade, but further studies are required.
MALAT1, an mRNA splicing mediator , is upregulated in several human cancers and contributes to cancer cell proliferation . MALAT1 depletion results in arrest at G1 and promotes expression of p53 as well as p16, p21, and p27 in human fibroblasts  (Table 1). In contrast, MALAT1 depletion suppresses various genes involved in cell cycle progression such as the genes encoding cyclin A2 and Cdc25A, thereby arresting the cell cycle in G1. Moreover, in G2/M progression, MALAT1 is required for expression of B-Myb, which is involved in the expression of mitotic proteins such as cyclin B1, CDK1, FoxoM1, and PLK by controlling the splicing of B-Myb mRNA . Therefore, MALAT1 may contribute to cell cycle progression in each phase by coordinated control of cell cycle regulators.
Steroid receptor RNA activator (SRA) was identified as an lncRNA that binds to steroid receptors . SRA forms the SRC-1 complex to activate transcription, mediated by steroid receptors such as progesterone receptor and estrogen receptor. It also binds to various other proteins such as myoD, and has multiple cellular functions such as myogenesis. SRA also binds to PPARγ and coactivates gene expression mediated by PPARγ. As such, SRA regulates adipogenesis and insulin sensitivity via PPARγ . Additionally, SRA shows PPARγ-independent activity. Overexpression of SRA in pre-adipocytes downregulates the expression of cell cycle-promoting genes such as those encoding the cyclins [cyclins (A2, B1/2)], CDC20, MCMs (3, 4, 5, 6), and CDT1. Conversely, these genes are upregulated by depletion of SRA. However, it remains to be elucidated whether SRA directly or indirectly suppresses the transcription of these genes, and further investigation into the mechanisms of SRA-regulated gene expression is required.
LncRNAs regulating CDK inhibitors
INK4 family inhibitors
The CDK inhibitory proteins, p16ink4a and p15ink4b (hereafter p16 and p15), bind to and inhibit CDK4 and 6, respectively, via their ankyrin repeats [3, 42]. The p15 and p16 genes (CDKN2B and CDKN2A, respectively) are located at the INK4 locus together with the alternating reading frame gene, ARF . ARF inhibits MDM2-dependent degradation of both p53  and pRB . Therefore, the expression of INK4 locus genes is critical for cell cycle regulation. The INK4 proteins are relatively stable, and their ubiquitin-dependent proteolysis is not particularly important for controlling their cellular levels. Therefore, the INK4 locus genes are mainly regulated by transcription. The participation of several transcription factors, including the ETS family , FOXO , and SP1 , has been reported. Moreover, the locus is regulated epigenetically. It has been suggested that PU.1 cooperates with DNA methyltransferase and is involved in the INK4 locus via methylation of CpG islands . Moreover, PcG is recruited to the INK4 locus, thereby suppressing transcription via histone H3K27 methylation .
It is important to understand how ANRIL expression is regulated. We found that excess RAS signaling promoted by the introduction of activated H-RasG12V into WI38 fibroblasts suppressed ANRIL expression and induced p15 and p16, thereby arresting the cell cycle and inducing senescence-associated beta-galactosidase  (Fig. 3b). Recently, Wan et al.  reported that ANRIL is induced by DNA-damaging agents via the ATM-E2F1 pathway, but p53 is not induced. Moreover, they suggested that depletion of ANRIL decreases homologous recombination after DNA double-strand breaks, although it is unclear whether ANRIL promotes DNA repair via the recruitment of PRC1 and PRC2 to the INK4 locus. Further studies are required on ANRIL function in response to cellular stresses. Moreover, Yang et al. found that lncRNA-HEIH is highly expressed in HBV-related hepatocellular carcinoma. It negatively regulates the expression of CDK inhibitors, such as p15, p16, p21, and p57, via interacting with EZH2, and then plays an important role in G0/G1 arrest  (Table 1).
p18ink4c (hereafter p18) is another INK4 family CDK inhibitor that also inhibits both CDK4 and 6 [3, 54]. Recently, ink4c−/− mice have been shown to develop spontaneous pituitary adenomas , the frequency of which is enhanced by deletion of other CDK inhibitor genes . The combined deletion of the p18 gene (CDKN2C) with the p16 gene is also found in human cancers . Moreover, the expression levels of p16 and p18 are often inversely correlated during the progression of senescence . It has been reported that transcription of the p18 gene is regulated by Menin-RET-signaling and the PI3K-AKT pathway . Du et al.  reported that the lncRNA, HULC, negatively regulates the expression of p18 gene, which is located near the region containing HULC (Table 1). HULC was identified as an lncRNA upregulated in human hepatocellular carcinoma (HCC)  that is transcribed in a CREB-dependent manner . Moreover, the expression of p18 is induced and suppressed by depletion and overexpression of HULC, respectively. The expression of p18 is inversely correlated with the expression of HULC in human HCC tissue specimens. Furthermore, the hepatitis B virus oncogene product, HBx, activates the HULC promoter via CREB to suppress the transcription of the p18 gene by upregulated HULC . Downregulation of the p18 gene by HBx via HULC induction may contribute to the development of HCC, although it is unknown how HULC suppresses the transcription of the p18 gene.
Cip/Kip family inhibitors
The transcription of p57Kip2 gene (CDKN1C), which is located at the KCNQ1 domain, is epigenetically suppressed as an imprinted gene on the paternal chromosome . KCNQ1OT1 is paternally expressed as an antisense RNA of the KCNQ1 domain containing KCNQ1 and the p57Kip2 gene . KCNQ1OT1 functions as a recruiter that associates with the chromatin modifiers, PRC2 and G9a, and recruits them to the KCNQ1 domain to suppress the transcription of p57Kip2 gene (Table 1). As described above, lncRNA-HEIH downregulates the expression of not only the INK4 family inhibitors, p15 and p16, but also the Cip/Kip family inhibitors, p21 and p57 .
LncRNAs regulating the pRB pathway
LncRNAs regulating the p53 pathway
Recently, Melo et al.  reported that enhancer RNAs (eRNAs) are required for coordinated promotion between p53 target genes and p53-bound enhancer regions distant from the target gene, and participate in p53-dependent cell cycle arrest (Table 1). LncRNA loc285194 was suggested to have a tumor suppressor function, but its mechanism was unknown. Liu et al. found that loc285194 is induced by binding of p53 to its binding site in the promoter (Table 1). Moreover, they indicated that loc285194 binds to and inhibits miR-211, thereby downregulating miR-211-mediated cell proliferation . Loc285194 is downregulated in human colon cancer specimens, and thus may contribute to the tumor suppressive function of p53 to inhibit miR-211 .
Huarte et al.  identified lncRNA-p21, which is transcribed near the p21Cip1 gene (CDKN1A) as a p53-target gene. p53 directly binds to its binding element in the lncRNA-p21 promoter. Depletion of lncRNA-p21 alters the expression of some p53-target genes except for p21 gene and inhibits apoptosis (Fig. 5; Table 1). lncRNA-p21 binds to hnRNP-K and recruits it to the target genes, but the mechanism of target gene regulation is unknown. p53 function is partially mediated by gene regulation via lncRNA-p21-hnRNP-K. Moreover, Yoon et al. proposed that lncRNA-p21 functions as a modulator of translation. lncRNA-p21 associates with target mRNAs such as β-catenin and JunB in collaboration with Rck/p54 RNA helicase, and thus the translation of the target mRNAs is repressed . Therefore, lncRNA-p21 regulates both transcription in the nucleus and translation in the cytoplasm.
PANDA (p21-associated ncRNA DNA damage-activated) was identified as a p21 promoter-derived transcript using ultra-high density tiling array of 56 cell-cycle genes. It is induced by DNA damage in a p53-dependent manner  (Table 1). PANDA binds to and inhibits NF-YA transcription factor, which limits the expression of proapoptotic genes such as FAS and BIK and results in the repression of apoptosis. PANDA is selectively induced in metastatic ductal carcinomas but not in normal breast tissue . The results suggest that abnormal overexpression of PANDA may suppress apoptosis induced by DNA damage, which will accumulate and push the genome toward carcinogenesis.
Although the mechanisms of cell cycle regulation via cyclin–CDK, the p53/RB pathway, and the checkpoint pathway have been described in detail, recent studies on lncRNAs strongly suggest that lncRNAs control the expression of cell cycle regulators. Therefore, lncRNAs are critically involved in cell cycle regulation. However, it is unclear why lncRNAs might be deployed to regulate the cell cycle. As described in the “Introduction”, lncRNAs involved in cell cycle regulation are classified into four groups. As shown in Table 1, ANRIL, lncRNA-HEIH, and KCNQ1OT1 are involved in epigenetic regulation of target gene transcription by collaborating with chromatin modifiers, which are classified as epigenetic regulators. ncRNACCND1, SRA, PANDA, and lncRNA-p21 directly interact with the transcriptional machinery on the target genes and collaboratively regulate transcription as transcription factor regulators. Post-transcription regulators including gadd7, MALAT1, lncRNA-RoR, and loc285194 bind to their specific target mRNA to suppress translation and/or to modulate mRNA stability. SRA and MALAT1 also promote protein–protein interactions and are classified as protein scaffolds. Because the general cell cycle is closely associated with various cellular events as well as biological processes, it should be accurately regulated. Post-transcription regulators such as gadd7, MALAT1, H19 lncRNA, and loc285194 can rapidly and transiently suppress translation of their target genes. Transcription factor regulators such as ncRNACCND1, SRA, PANDA, and lncRNA-p21 that directly interact with the transcription machinery on the target genes may also participate in transient regulation. Alternatively, epigenetic regulators such as ANRIL, lncRNA-HEIH, and KCNQ1OT1 may have long-term effects on cellular senescence and imprinting because they mediate epigenetic regulation of cell cycle regulatory genes via chromatin modifiers. From this viewpoint, the cell cycle-regulated lncRNAs mainly control cellular levels of cell cycle regulators via various mechanisms, and may provide diversity and reliability to the general cell cycle.
It is interesting that many lncRNAs are associated with the DNA damage response. As shown in Table 1, 4 of 14 lncRNAs, lncRNACCND1, gadd7, ANRIL and PANDA, are induced by DNA damage. Another 4 lncRNAs, lncRNA-RoR, lncRNAp21, p53-induced eRNA, and loc285194, are induced in a p53-dependent manner, suggesting that they are induced by DNA damage. Therefore, these reported lncRNAs may participate in cell cycle arrest or induction of apoptosis as non-canonical DNA damage responses, whereas the ATM/ATR pathway is involved in a canonical DNA damage response to inactivate CDK activity as a DNA damage checkpoint. LncRNAs-mediated non-canonical pathways may ensure the response to DNA damage is diverse and reliable depending on the cellular context.
Considering the recent progress in lncRNA research, many lncRNAs that have a functional role in cell cycle regulation remain to be identified because the functions of only a small percentage of the total lncRNA population are understood. To clarify the roles of lncRNAs in cell cycle regulation, it should be determined how they regulate the target cell cycle regulators and which signaling pathways induce these lncRNAs. Since abrogation of the cell cycle is closely associated with cancer development and growth, cell cycle regulatory lncRNAs such as ANRIL and PANDA may have oncogenic properties. The importance of lncRNAs in cell cycle regulation will be clarified by further pathological studies. Moreover, these cell cycle regulatory lncRNAs may be novel candidate molecular targets for cancer therapy or diagnosis.
- 1.Nurse P (2002) Cyclin dependent kinases and cell cycle control (nobel lecture). Chem Biochem 3:596–603Google Scholar
- 51.Pasmant E, Laurendeau I, Héron D et al (2007) Characterization of a germ-line deletion, including the entire INK4/ARF locus, in a melanoma-neural system tumor family: identification of ANRIL, an antisense noncoding RNA whose expression coclusters with ARF. Cancer Res 67:3963–3969PubMedCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.