|Home | About | Journals | Submit | Contact Us | Français|
Recent ultrahigh-density tiling array and large-scale transcriptome analysis have revealed that large numbers of long non-coding RNAs (lncRNAs) are transcribed in mammals. Several lncRNAs have been implicated in transcriptional regulation, organization of nuclear structure, and post-transcriptional processing. However, the regulation of expression of lncRNAs is less well understood. Here, we show that the exogenous and endogenous expression of an oncogenic form of small GTPase Ras (called oncogenic Ras) decrease the expression of lncRNA ANRIL (antisense non-coding RNA in the INK4 locus), which is involved in the regulation of cellular senescence. We also show that forced expression of oncogenic Ras increases the expression of lncRNA PANDA (p21 associated ncRNA DNA damage activated), which is involved in the regulation of apoptosis. Microarray analysis demonstrated that expression of multiple lncRNAs fluctuated by forced expression of oncogenic Ras. These findings indicate that oncogenic Ras regulates the expression of a large number of lncRNAs including functional lncRNAs, such as ANRIL and PANDA.
The online version of this article (doi:10.1007/s10616-014-9834-9) contains supplementary material, which is available to authorized users.
Long non-coding RNAs (lncRNAs) range in size from 200 nucleotides to over 10 kb and are spliced and polyadenylated post-transcriptionally. Although mass scale transcriptome analysis has revealed the existence of large numbers of lncRNAs in mammals (Carninci et al. 2005), the function of most lncRNAs remain unclear. Recent studies revealed that several lncRNAs are involved in biological processes including development, cellular senescence, apoptosis, and cancer (Batista and Chang 2013; Fatica and Bozzoni 2014; Kitagawa et al. 2013, 2012). Previously, we and another group have reported the role of the lncRNA ANRIL (antisense non-coding RNA in the INK4 locus) in the regulation of INK4 locus (Kotake et al. 2011; Yap et al. 2010), which encodes two cyclin-dependent kinase inhibitors, p15INK4B and p16INK4A and a positive regulator of p53, ARF. The INK4 locus is frequently mutated or its expression is silenced in human cancers (Ruas and Peters 1998; Sharpless 2005). ANRIL (3.8 kb transcript) is expressed in the opposite direction from the p16INK4A-ARF-p15INK4B gene cluster. ANRIL binds to polycomb repression complex 1/2, that stably silence gene expression and is required for the recruitment and repression of p15INK4B and p16INK4 transcription, resulting in the prevention of cellular senescence. PANDA (p21 associated ncRNA DNA damage activated; 1.5 kb transcript) is a p21 promoter-derived lncRNA induced by DNA damage in a p53-dependent manner (Hung et al. 2011). PANDA directly binds to the transcription factor NF-YA and inhibits its association with promoters of pro-apoptotic genes, such as FAS, NOXA, and PUMA. Depletion of PANDA sensitizes cells to apoptosis by DNA damage, suggesting that PANDA functions by blocking apoptosis to maintain the survival of cells that have sustained DNA damage.
Recent studies have revealed that several lncRNAs, such as MALAT1 and HOTAIR, are also associated with cancer progression (Gupta et al. 2010; Tano et al. 2010). Although the biological importance of lncRNAs has gradually been demonstrated, the regulation of lncRNAs expression is poorly understood. In this study, we analyzed the expression of lncRNAs induced by oncogenic Ras signaling.
WI38 and TIG-3 cells are normal human diploid fetal lung fibroblasts. WI38 cells were obtained from American Type Culture Collection (Frederick, MD, USA) and TIG-3 cells were obtained from Health Science Research Resource Bank (Osaka, Japan). Human colorectal cancer HCT116 cells were obtained from American Type Culture Collection. HKe3 cells were established from HCT116 cells with a disruption in the oncogenic K-Ras (Shirasawa et al. 1993). All cells were cultured in DMEM (Invitrogen, Carlsbad, CA, USA) containing 10 % fetal bovine serum (GIBCO, Grand Island, NY, USA). WI38 and TIG-3 cells were infected with retroviruses expressing H-RasG12V (kindly provided by Dr. CJ. Der, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA). The pMX-puro retrovirus vector was kindly provided by Dr. T Kitamura (The University of Tokyo, Tokyo, Japan). Retrovirus production and transduction were performed as previously described (Kotake et al. 2007).
For western blotting, cells were lysed with RIPA buffer (50 mM Tris–HCl pH 8.0, 150 mM NaCl, 1 % NP-40, 0.5 % DOC (sodium deoxycholate), 0.1 % SDS, 1 mM Na3VO4, 1 mM DTT, 1 mM PMSF) supplemented with protease inhibitors (10 mg/L antipain (PEPTIDE INSTITUTE. INC., Osaka, Japan), 10 mg/L leupepcin (PEPTIDE INSTITUTE. INC.), 10 mg/L pepstatin (PEPTIDE INSTITUTE. INC.), 10 mg/L trypsin inhibitor (Sigma), 10 mg/L E64 (PEPTIDE INSTITUTE. INC.), and 2.5 mg/L chymostatin (PEPTIDE INSTITUTE. INC.)). Western blotting was performed as previously described (Kotake et al. 2007). Antibodies to H-Ras (OP23; Calbiochem-Merck, Darmstadt, Germany) and α-tubulin (Sigma, St. Louis, MO, USA) were purchased commercially. Horseradish peroxidase-conjugated antibodies (Promega, Tokyo, Japan) were used as secondary antibodies and the chemiluminescence system from PerkinElmer (Tokyo, Japan) for detection.
For RT-PCR, total RNA was extracted by RNeasy Plus kit (Qiagen, Tokyo, Japan), and 1 μg of total RNA was used for cDNA synthesis primed with oligo dT primers (Invitrogen). The specific PCR pairs for PANDA, ANRIL, and GAPDH were as follows: PANDA, 5′-AGACCCCAGTGGCACCTGAC-3′ and 5′-GGGCAGAACTTGGCATGATG-3′; ANRIL, 5′-TGCTCTATCCGCCAATCAGG-3′ and 5′-GGGCCTCAGTGGCACATACC-3′; GAPDH, 5′-GCAAATTCCATGGCACCGT-3′ and 5′-TCGCCCCACTTGATTTTGG-3′.
DNA microarray analysis was performed using SurePrint G3 Human GE 8x60K (Agilent Technologies, Santa Clara, CA, USA) according to the manufacturer’s protocol by Hokkaido System Science (Sapporo, Japan). Three chips per group were hybridized. Briefly, total RNA was extracted by RNeasy Mini kit (Qiagen), and 200 μg of total RNA was amplified and labeled with cyanine 3 CTP by the Low Input Quick Amp Labeling kit (Agilent Technologies). Labeled cRNA was purified by RNeasy mini spin column (Qiagen) and 600 ng of cyanine 3 labeled cRNA was hybridized to SurePrint G3 Human GE 8x60K (Agilent Technologies). The chips were analyzed by Agilent Technologies Microarray Scanner.
The small GTPase Ras controls cell proliferation, survival, and differentiation by regulating downstream effector molecules, such as mitogen-activated protein kinases, phosphoinositide 3-kinase, and Ral guanine nucleotide-dissociation stimulator (Karnoub and Weinberg 2008; Schubbert et al. 2007). Activating Ras mutations including G12V and G13D amino acid substitution lead to oncogenic transformation and occur in a wide range of human cancers. Previously and here, we showed that ANRIL expression is decreased by forced expression of H-RasG12V in WI38 cells (Kotake et al. 2011) (Fig. 1a, b). Furthermore, we examined by using HCT116 and HKe3 cells whether endogenous expression of oncogenic Ras affect the expression of ANRIL. HCT116 cells have a heterozygous K-Ras mutation (G13D). HKe3 cells were generated by disrupting the K-RasG13D of HCT116 cells (Shirasawa et al. 1993). RT-PCR analysis demonstrated that ANRIL is expressed at a low level in HCT116 cells as compared to HKe3 cells (Fig. 1c). We also examined the effect of oncogenic Ras on the expression of PANDA. The stable cell lines expressing H-RasG12V in TIG-3 cells in which PANDA expression was detectable were established, and the expression of PANDA was determined. Western blotting results demonstrated the stable expression of H-RasG12V (Fig. 2a). PANDA expression was substantially increased by H-RasG12V transduction as assessed by RT-PCR analysis (Fig. 2b). These data indicate that oncogenic Ras regulates the expression of two lncRNAs, ANRIL and PANDA, which have a pivotal role in cell fate determination, such as cellular senescence and apoptosis.
Oncogenic Ras might also affect other lncRNAs. Therefore, we next examined the effect of oncogenic Ras on the expression of lncRNAs by global analysis. DNA microarray analysis showed that 3,373 of the protein coding genes (PCG) had at least a twofold change by H-RasG12V transduction (Fig. 3a; Table 1). These data support a previous study demonstrating oncogenic Ras regulates diverse pathways (Karnoub and Weinberg 2008). Additionally, of the 7,419 transcribed regions of lncRNAs, the expression level of 243 (3.3 %) increased more than twofold and the expression level of 168 (2.3 %) decreased by <50 % following H-RasG12V transduction (Fig. 3b; Table 1 and Supplemental Table 1). These data indicate that oncogenic Ras regulates the expression of multiple lncRNAs as well as PCG.
In this paper, we showed that both exogenous and endogenous expression of oncogenic Ras decreased the expression of ANRIL, which represses the transcription of INK4 locus. In mouse and human fibroblasts, the transcription of INK4 locus is activated by oncogenic Ras, causing stable cell cycle arrest to protect cells from hyperproliferative stimulation (Brookes et al. 2002; Serrano et al. 1997). The reduction of ANRIL by oncogenic Ras might be required for the activation of INK4 locus by oncogenic Ras, resulting in the induction of premature senescence. We also found that enforced expression of oncogenic Ras increased the expression of PANDA, which represses apoptosis. Mutational activation of ras can protect some cell strains from apoptosis (Downward 1998). The induction of PANDA by oncogenic Ras might be required for the repression of apoptosis, resulting in the progression of carcinogenesis. Indeed, PANDA is selectively increased in metastatic ductal carcinomas but not in normal breast tissue (Hung et al. 2011). It will be an important issue to reveal the role of PANDA in carcinogenesis. However, the molecular mechanism by which oncogenic Ras regulates ANRIL and PANDA expression is unclear. Recent studies showed that E2F1 transcription factor directly binds to and activates ANRIL transcription (Sato et al. 2010; Wan et al. 2013). It was also revealed that several lncRNAs including PANDA are induced in a p53-dependent manner (Huarte et al. 2010; Hung et al. 2011). Thus, it will be important to identify the downstream effector molecules of oncogenic Ras, which regulate the expression of lncRNAs.
Although approximately 10,000 primate-specific lncRNAs have been identified (Necsulea et al. 2014), the biological functions of most lncRNAs remain unclear. This study demonstrated that the expression of many lncRNAs was altered by oncogenic Ras. Among these lncRNAs regulated by oncogenic Ras, uncharacterized functional lncRNAs associated with malignant transformation may be present.
Below is the link to the electronic supplementary material.
We greatly appreciate Dr. T Kitamura (The University of Tokyo, Tokyo, Japan) and Dr. CJ Der (University of North Carolina at Chapel Hill, Chapel Hill, NC, USA) for providing the plasmids used in this study. This work was supported in part by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to YK, KK, HN, and MK) and Takeda Science Foundation (to YK).