The LncRNA EPEL Promotes Lung Cancer Cell

Physiol Biochem 2018;45:1270-1283 Cellular Physiology Cell © 2018 The Author(s). Published by S. Karger AG, Basel DOI: 10.1159/000487460 DOI: 10.1159/000487460 © 2018 The Author(s) online:February February 2018 www.karger.com/cpb Published online: 15,15, 2018 Published by S. Karger AG, Basel and Biochemistry Published www.karger.com/cpb Park et al.: EPEL: a Novel lncRNA that Regulates Cell Proliferation Accepted: January 16, 2018

This article is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND) (http://www.karger.com/Services/OpenAccessLicense). Usage and distribution for commercial purposes as well as any distribution of modified material requires written permission.

Original Paper

The LncRNA EPEL Promotes Lung Cancer Cell Proliferation Through E2F Target Activation Seong-Min Parka,b Eun-Young Choib Seon-Young Kima,c Youn-Jae Kimb

Dong-Hyuck Baea,c

Hyun Ahm Sohna

Personalized Genomic Medicine Research Center, KRIBB, Daejeon, bTranslational Research Branch, Research Institute, National Cancer Center, Goyang, Gyeonggi, cDepartment of Bioscience, University of Science and Technology, Daejeon, Republic of Korea a

Key Words Lncrna • E2F • CCNB1 • Lung cancer Abstract Background/Aims: Recent studies have revealed that many long non-coding RNAs (lncRNAs) play oncogenic or tumor-suppressive roles in various cancers. Lung cancer is the leading cause of cancer-related death worldwide, and many lung cancer patients frequently relapse after surgery, even those in the early stages. However, the oncogenic or tumor-suppressive roles and clinical implications of lncRNAs in lung cancer have not been fully elucidated. Methods: The association between an E2F-mediated cell proliferation enhancing lncRNA (EPEL) expression and lung cancer patient survival was accessed using public microarray data with clinical information. Cancer-related phenotypes were analyzed by the siRNA knockdown of EPEL in two lung cancer cell lines. Gene set analysis of gene expression data were performed to identify pathways regulated by EPEL. RNA immunoprecipitation, RT-qPCR, and ChIP assays were performed to explore the functions of selected target genes regulated by EPEL. Results: EPEL, known as LOC90768 and MGC45800, was associated with the relapse and survival of lung cancer patients and promoted lung cancer cell proliferation through the activation of E2F target genes. EPEL knockdown specifically down-regulated the expression of cell cyclerelated E2F target genes, including Cyclin B1 (CCNB1), in lung cancer cells but not that of apoptosis- or metabolism-related E2F target genes. EPEL interacted with E2F1 and regulated the expression of the E2F target genes by changing the binding efficiency of E2F1 to the E2F target promoters. Moreover, the expression levels of EPEL and CCNB1 both alone and in combination were robust prognostic markers for lung cancer. Conclusions: Considering its specific effects on cell cycle-related E2F target genes and its significant association with the prognosis of lung cancer patients, we suggest that the transcriptional regulation of EPEL through E2F target genes is potentially a target for the development of novel therapeutic strategies for lung cancer patients. © 2018 The Author(s) Published by S. Karger AG, Basel

S.-M. Park and E.-Y. Choi contributed equally to this work. Seon-Young Kim and Youn-Jae Kim

Personalized Genomic Medicine Research Center, KRIBB, Daejeon, Translational Research Branch, Research Institute, National Cancer Center, Goyang, Gyeonggi (Korea) E-Mail [email protected], [email protected]

1270

Physiol Biochem 2018;45:1270-1283 Cellular Physiology Cell © 2018 The Author(s). Published by S. Karger AG, Basel DOI: 10.1159/000487460 and Biochemistry Published online: February 15, 2018 www.karger.com/cpb Park et al.: EPEL: a Novel lncRNA that Regulates Cell Proliferation

Introduction

Recent genome-wide studies have identified thousands of long noncoding RNAs (lncRNAs), RNAs longer than 200 nucleotides without open reading frames, that were determined to be associated with various cellular processes, such as pluripotency, proliferation, development, differentiation and apoptosis [1-9]. Additionally, many differentially expressed lncRNAs have been characterized in various cancer types with oncogenic or tumor-suppressive roles [10-16], with some lncRNAs showing potential as diagnostic or prognostic markers for cancer patients [12, 13, 15-18]. By interacting with chromatin remodelers [12, 14] and transcription factors, such as p53 [19, 20] and E2F [21, 22], many lncRNAs mediate target gene transcription and regulate cancer-related functions, such as cell cycle progression, apoptosis, metastasis and senescence. For cancer progression, lncRNAs often play roles as cell cycle regulators in cyclin-dependent [23-25] or p53-dependent manners [19, 20]. Thus, the characterization of cancer-related lncRNAs is important for the study of cancer biology. The E2F/Rb pathway is a major pathway regulating the mammalian cell cycle [26, 27], and E2Fs are a large family of transcription factors that play various biological roles, including cell cycle control [28, 29]. In a growth control model of E2F1, E2F1 mediates various functions, and even opposing functions, such as cell cycle progression and apoptosis, depending on its binding partners [26, 27, 30]. For cell cycle progression, E2F1 regulates the G1/S [31], S/G2 [32] and G2/M [33] transitions depending on its binding partners. For example, two lncRNAs, TUG1 and MALAT1/NEAT2, regulate the growth control activity of E2F1 by mediating E2F1 SUMOylation [22]. E2F target genes revealed by genome-widestudies [33-38] imply that lncRNAs can regulate E2F target gene expression and various cancer-related functions, including the cell cycle. Lung cancer is the leading cause of cancer-related death worldwide, and lung cancer patients frequently relapse after surgery, even those in the early stages [39-41]. Because understanding lung cancer prognosis is important for developing therapeutic strategies, such as post-operative adjuvant therapy, histological markers have been used for this purpose, but randomized studies for prognosis determination have shown controversial effectiveness [42-45]. Recently, several lncRNAs were reported as potential prognostic markers for lung cancer [46-48]. Thus, studying lncRNAs is a promising way to identify novel molecular markers of lung cancer prognosis and novel therapeutic targets. In this study, we characterized a novel lncRNA, herein named E2F-mediated cell proliferation enhancing lncRNA (EPEL), that promotes lung cancer cell proliferation through E2F target gene activation. Because EPEL over-expression was significantly associated with poor lung cancer prognosis, we suggest its feasibility as a promising prognostic marker for this disease. Materials and Methods

Public data analysis Two public gene expression datasets (GSE31210 and GSE50081) were downloaded from the National Center for Biotechnology Information Gene Expression Omnibus (NCBI GEO) database (http://www.ncbi. nlm.nih.gov/geo/). From the GSE50081 dataset, we used only data for adenocarcinoma patients, and gene expression data were globally normalized using the Robust Multi-array Average (RMA) method [49]. All statistical tests were performed using the R programming language (https://www.r-project.org/), and graphs and heatmaps were prepared using the Excel, R and MeV (http://www.tm4.org/mev.html) programs. Cell culture, siRNAs and transfection A549 and NCI-H1299 cells (ATCC) were maintained in complete RPMI 1640 medium (HyClone) at 37°C in a humidified 5% CO2 incubator. Complete media was supplemented with 10% fetal bovine serum (HyClone), 100 U/ml penicillin/streptomycin (WelGENE), and 2 mM L-glutamine (HyClone). The siRNA target sequences were designed .using the AsiDesigner program [50] and are listed (for all online suppl.

1271


material, see www.karger.com/ doi/10.1159/000487460) in Suppl. Table 1. For siRNA transfection, 1.5 ~ 2.0 x 105 lung cancer cells were plated in 6-well plates and incubated overnight. The siRNAs or non-targeting controls at a concentration of 50 nM were transfected into lung cancer cells using Lipofectamine 2000 in Opti-MEM media. After 4 hours of incubation, the media were changed to complete media, and 48 hours after transfection, gene knockdown was confirmed by RT-qPCR.

Table 1. Multivariate Cox proportional hazard analysis for the prediction of lung adenocarcinoma patient survival Variable

GSE31210 overall EPEL Stage I vs. II MYC copy MYC expression EGFR mutation K-ras mutation ALK fusion Smoking Gender Age GSE31210 relapse-free EPEL Stage I vs. II MYC copy MYC expression EGFR mutation K-ras mutation ALK fusion Smoking Gender Age GSE50081 adeno overall EPEL Stage I vs. II Smoking Sex Age GSE50081 adeno disease-free EPEL Stage I vs. II Smoking Sex Age

Survival Hazard ratio (95% Confidence interval) 2.35572 (0.98232 - 5.649) 3.18932 (1.44192 - 7.054) 2.2642 (0.04776 - 4.084) 2.2172 (0.03805 - 5.345) 1.7231 (0.25698 - 1.311) 1.8281 (0.15222 - 1.966) 1.2315 (0.08912 - 7.398) 1.4187 (0.23627 - 2.103) 1.01752 (0.34867 - 2.969) 1.03656 (0.98417 - 1.092) 2.47593 (1.02082 - 6.005) 3.81864 (1.69569 - 8.599) 1.8125 (0.07397 - 4.115) 1.8750 (0.05600 - 5.079) 1.6537 (0.26507 - 1.380) 1.8982 (0.14622 - 1.898) 1.2232 (0.09164 - 7.293) 1.2313 (0.27124 - 2.432) 1.0144 (0.33562 - 2.896) 1.03654 (0.98366 - 1.092)

p-value

0.05486 0.00419 0.47145 0.52791 0.19048 0.35529 0.85343 0.53059 0.97465 0.17483 0.0449 0.00122 0.56183 0.58461 0.23193 0.32708 0.85681 0.71001 0.97925 0.1792

Reverse transcription qPCR Total RNA was extracted 2.47894 (1.3662 - 4.498) 0.00282 using the RNeasy Mini Kit 2.60230 (1.4412 - 4.699) 0.00151 (QIAGEN) according to the 1.12647 (0.5706 - 2.224) 0.73148 1.39902 (0.7743 - 2.528) 0.26589 manufacturer’s instructions. 1.01346 (0.9835 - 1.044) 0.38329 Reverse transcription was 2.48323 (1.3637 - 4.522) 0.00294 performed with 1 μg of total 2.57987 (1.4205 - 4.686) 0.00185 1.12365 (0.5669 - 2.227) 0.73837 RNA as the template and 1.30645 (0.7211 - 2.367) 0.37804 M-MLV Reverse Transcriptase 1.01037 (0.9805 - 1.041) 0.50017 (Promega). Quantitative realtime PCR (qPCR) reactions were performed in triplicate on a LightCycler 480 machine (Roche) using LightCycler 480 SYBR Green I Master Mix (Roche). cDNA expression was normalized to that of β-actin, and at least three independent biological replicates were included for each reaction. The primers used for qPCR were designed either manually or with the Primer3 program (http://biotools.umassmed.edu/bioapps/primer3_www.cgi). All primer sequences are listed (see suppl. material) in Suppl. Table 2. Proliferation assay Suspensions of 1.0 to 2.0 x 103 cells were seeded into 96-well plates. After 48 to 72 hours of incubation at 37°C, a CyQUANT NF Cell Proliferation dye reagent and deliverer (Invitrogen) mixture was added. After 30 min of incubation, the fluorescence intensity ratio of 530 nm to 485 nm was measured.

Invasion assay Transwell chambers (Corning) were coated with Matrigel Basement Membrane Matrix (BD). Cells were suspended in serum-free media and seeded into the upper chamber at a density of 2.0 to 5.0 x 104 cells per well, and serum-containing media were placed into the lower chamber. After incubation for 24 ~ 72 hours, cells penetrating the pores were stained with Diff-Quik staining solution (Sysmex) and observed under a microscope.

Microarray analysis In total, 750 ng of each cRNA library was prepared from the total RNA samples, hybridized to the HumanHT-12 Gene Expression BeadChip (Illumina), and measured according to the manufacturer’s instructions. The intensity values were analyzed with the GenomeStudio program and globally normalized using the quantile method. Gene set and pathway analyses of differentially expressed genes were performed using the Gene Set Enrichment Analysis (GSEA) program (http://software.broadinstitute.org/gsea/index. jsp). RNA immunoprecipitation assay

1272


RNA immunoprecipitation (RIP) was performed with 5.0 x 106 cells using the Magna RIP kit (Millipore) according to the manufacturer’s instructions. Primary antibodies used for the RIP assay are listed (see suppl. material) in Suppl. Table 3. RIP-qPCR amplifications were performed in triplicate on the LightCycler 480 machine using LightCycler 480 SYBR Green I Master Mix. Data were normalized to the input levels, and at least three independent biological replicates were included for each RIP-qPCR. The primer sequences are listed (see suppl. material) in Suppl. Table 2. Nucleus/cytoplasm fractionation Nuclear and cytosolic fractions were separated using the PARIS kit (Ambion) according to the manufacturer’s instructions

Chromatin immunoprecipitation assay Chromatin immunoprecipitation (ChIP) was performed using Dynabeads Protein A and G (Thermo Fisher Scientific). First, 5.0 x 106 cells were cross-linked with 1% formaldehyde for 10 min at 25°C. Then, the cells were lysed and sonicated using the truChIP Chromatin Shearing Reagent Kit (Covaris) according to the manufacturer’s instructions. The expected fragment size was between approximately 200 and 500 base pairs. Samples were diluted five-fold with low-salt RIPA buffer (0.1% SDS, 1% Triton X-100, 1 mM EDTA, 140 mM NaCl, 4% deoxycholate) and pre-cleared with 50 µl of Dynabeads Protein A and G for 1 hour at 4°C. Primary antibodies were added to pre-cleared supernatants, and the mixtures were incubated overnight at 4°C. The antibodies used for the ChIP assay are listed (see suppl. material) in Suppl. Table 3. Next, 50 µl of Dynabeads Protein A and G were added to the samples, and the mixtures were incubated for 2 hours at 4°C. The beads were subsequently washed with wash buffer (low-salt RIPA, high-salt RIPA, LiCl, and TE). After 15 min of incubation at 65°C, precipitated chromatin was eluted twice in 250 µl of elution buffer (0.1 M NaHCO3 and 1% SDS). Reverse cross-linking was performed for 4 hours at 65°C, and chromatin was then treated with RNase A for 1 hour at 37°C and proteinase K for 1 hour at 45°C. DNA was purified with phenol/ chloroform extraction or a QIAquick PCR Purification Kit (QIAGEN). ChIP-qPCR reactions were performed in triplicate on the LightCycler 480 machine using LightCycler 480 SYBR Green I Master Mix. The data were normalized to the input levels, and at least three independent biological replicates were included for each ChIP-qPCR. The primer sequences are listed (see suppl. material) in Supp. Table 2. Data availability Microarray data have been deposited in the GEO database under accession number GSE102356.

Results

Promotion of lung cancer progression by EPEL The human chromosome 4 (NCBI37/hg19) genomic locus from 183, 059, 813 to 183, 065, 668 is annotated as LOC90768 or MGC45800 and encodes a not yet characterized lncRNA that we named EPEL. According to Encyclopedia of DNA Elements (ENCODE) data, the locus is occupied by active histone markers, and EPEL is transcribed in various cell lines (GM12878, H1-hESC, HeLa-S3, HepG2, HSMM, HUVEC, K562, NHEK, NHLF cell lines) (see suppl. material, Suppl. Fig. 1). EPEL expression was higher in tumor lung tissues (lung-C) than in normal lung tissues (lung-N) according to the Gene Expression across Normal and Tumor tissue (GENT) database (see suppl. material, Suppl. Fig. 2) [51]. We analyzed two public gene expression datasets (GSE31210 and GSE50081) to evaluate the clinical significance of EPEL in lung cancer patients [52, 53]. We first compared EPEL expression between non-relapsed (or non-recurrent) and relapsed (or recurrent) patients and found that EPEL was up-regulated among lung cancer patients with relapse or recurrence (Fig. 1a). Increased EPEL expression was associated with poor survival after surgery (Fig. 1b). By analyzing additional public datasets, we found that increased EPEL expression was also associated with the poor survival of other cancer patients, including breast cancer, Ewing’s sarcoma and melanoma (see suppl. material, Suppl. Fig. 3). Thus, we suggest that the level of EPEL expression is associated with cancer progression. Using multivariate Cox proportional

1273


Park et al. Figure 1

b

GSE31210 Relapse-free survival

0.8

Survival rate

0.2

0.2

0.4

0.6

Survival rate

Expression (log2)

0.8

1.0

GSE31210 Overall survival 1.0

p = 1.0 x 10-6

0.6

GSE31210

0.4

a

0.0

p = 2.4 x 10-3

0.0

p = 4.5 x 10-3 0

1000

Relapse

3000

0

4000

1000

2500

3000

3500

0.8 0.6

Survival rate

0.8 0.6

0.2

0.4 0.2

p = 3.8 x 10-4 0.0

0.0

0

Recurrence

2000

GSE50081 adeno Disease-free survival

p = 9.6 x 10-4 Nonrecurrence

1500

1.0

GSE50081 adeno Overall survival

Survival rate

Expression (log2)

500

Survival day

1.0

GSE50081 adeno p = 1.4 x 10-2

2000

Survival day

0.4

Nonrelapse

2

4

6

Survival year

8

10

0

2

4

6

8

Survival year

Fig. 1. Promotion of lung cancer progression by EPEL. a. Comparison of EPEL expression in relapsed (or recurrent) and non-relapsed (or non-recurrent) lung cancer patients (GSE31210: Okayama et al. 2012, n = 204, p = 1.0 x 10-6, GSE50081 adenocarcinoma: Der et al. 2014, n = 128, p = 1.4 x 10-2, t-test). b. Prognosis of two groups of lung cancer patients classified by EPEL expression (GSE31210: Okayama et al. 2012, n = 204, p = 4.5 x 10-3 for overall survival, p = 2.4 x 10-3 for relapse-free survival, GSE50081 adenocarcinoma: Der et al. 2014, n = 128, p = 9.6 x 10-4 for overall survival, p = 3.8 x 10-4 for disease-free survival, log-rank test). Red: high expression group; Blue: low expression group.

hazard analysis (Table 1), we found that only EPEL expression and histological staging could predict poor survival in both datasets. Thus, we suggest that EPEL expression is a good prognostic marker for lung cancer.

Promotion of lung cancer cell proliferation and invasion by EPEL As the level of EPEL expression was associated with lung cancer progression, we examined which cancer-related phenotypes were influenced by EPEL expression in lung cancer cells. Using siRNAs for EPEL, we knocked down EPEL in A549 and NCI-H1299 lung cancer cell lines. We first validated that EPEL was successfully knocked down by three siRNAs in both A549 and NCI-H1299 lung cancer cells using the RT-qPCR assay (Fig. 2a) and then performed cell proliferation assays. EPEL knockdown significantly decreased the proliferation of A549 and NCI-H1299 cells (A549: 62% decrease for siEPEL #1, 26% decrease for siEPEL #2, 31% decrease for siEPEL #3; NCI-H1299: 31% decrease for siEPEL #1, 41% decrease for siEPEL #2, 46% decrease for siEPEL #3). We next performed an invasion assay with cells transfected with treatment siRNAs and a control siRNA. EPEL knockdown significantly decreased the invasion of A549 and NCI-H1299 cells (Fig. 2c) (A549: 65% decrease for siEPEL #1, 59% decrease for siEPEL #2, 52% decrease for siEPEL #3; NCI-H1299: 53% decrease for siEPEL #1, 44% decrease for siEPEL #2, 58% decrease for siEPEL #3). Thus, we concluded that EPEL promoted lung cancer proliferation and invasion, which is important for lung cancer

1274


a

1.2

1.0 0.8

**

0.6

**

0.4

*

0.2

Relative expression

Relative expression

1.2

0.0

siEPEL #1

siEPEL #2

**

0.8 0.6

**

***

0.4 0.2

***

siEPEL #2

siEPEL #3

0.2

siControl

siEPEL #1

1

*

0.8

**

0.6

**

0.4 0.2 0

0

siControl

siEPEL #1

siEPEL #2

siEPEL #3

siControl

1.2

1

1

0.8

***

0.4

*

***

0.2

siControl

siEPEL #1

siEPEL #2

siEPEL #3

Relative cell count

1.2

0.6

siEPEL #1

siEPEL #2

siEPEL #3

NCI-H1299 cell Invasion assay

A549 cell Invasion assay

0

***

1.2

1

Relative cell count

Relative cell count

0.4

NCI-H1299 cell Proliferation assay

A549 cell Proliferation assay

Relative cell count

**

0.6

siEPEL #3

1.2

c

1 0.8

0

siControl b

Park et al. Figure 2

NCI-H1299 cell RT-qPCR

A549 cell RT-qPCR

0.8

***

0.6

***

***

0.4 0.2 0

siControl

siEPEL #1

siEPEL #2

siEPEL #3

Fig. 2. Regulation of lung cancer cell invasion and proliferation by EPEL. After EPEL knockdown: a. RT-qPCR (A549: p = 2.7 x 10-3 for siEPEL #1, p = 3.8 x 10-3 for siEPEL #2, p = 2.7 x 10-2 for siEPEL #3; NCI-H1299: p = 5.9 x 10-3 for siEPEL #1, p = 5.7 x 10-4 for siEPEL #2, p = 2.1 x 10-4 for siEPEL #3, t-test). b. Proliferation assay (A549: p = 4.0 x 10-5 for siEPEL #1, p = 7.7 x 10-4 for siEPEL #2, p = 6.4 x 10-4 for siEPEL #3; NCI-H1299: p = 1.4 x 10-2 for siEPEL #1, p = 3.4 x 10-3 for siEPEL #2, p = 1.6 x 10-3 for siEPEL #3, t-test). c. Invasion assay (A549: p = 2.2 x 10-4 for siEPEL #1, p = 1.9 x 10-2 for siEPEL #2, p = 3.8 x 10-4 for siEPEL #3; NCI-H1299: p = 3.4 x 10-4 for siEPEL #1, p = 2.5 x 10-5 for siEPEL #2, p = 3.0 x 10-4 for siEPEL #3, t-test). Data are representative of three independent experiments. The error bars represent the standard error of the mean (s.e.m). *p