Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Clinics and Research in Hepatology and Gastroenterology (2019) xxx, xxx—xxx

Available online at

ScienceDirect www.sciencedirect.com

ORIGINAL ARTICLE

Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p Kongxi Zhu , Yunxia Wang , Lan Liu , Shuai Li , Weihua Yu ∗ Department of Gastroenterology, The Second Hospital of Shandong University, No. 247, Beiyuan Street, 250033 Jinan, Shandon Province, PR China

KEYWORDS Colon Cancer; Long non-coding RNA MBNL1-AS1; MicroRNA-412-3p; MYL9; Cancer stem cells

∗

Summary Background/Aims: Colon cancer is a common cancer that is a threat to human health. Some long non-coding RNAs (lncRNAs) have been observed to exert roles in colon cancer. Here, the current study is aimed to explore the potential mechanism of lncRNA MBNL1 antisense RNA 1 (MBNL1-AS1) in progression of colon cancer and the associated mechanisms. Methods: Microarray analysis was performed to screen differentially expressed lncRNA and genes associated with colon cancer and its potential mechanism. The functional role of MBNL1AS1 in colon cancer was analyzed, followed identification of the interaction among MBNL1-AS1, microRNA-412-3p (miR-412-3p), and MYL9. Subsequently, CSC viability, migration, invasion, and apoptosis were detected though a series of in vitro experiments. At last, in vivo experiments were performed to assess tumor formation of colon CSCs. Results: MBNL1-AS1 and MYL9 were poorly expressed in colon cancer. MBNL1-AS1 could competitively bind to miR-412-3p so as to promote MYL9 expression. Enhancement of MBNL1-AS1 or inhibition of miR-412-3p was shown to decrease CSC proliferation, migration, and invasion but promote apoptosis. Moreover, MBNL1-AS1 reversed the CSC-like properties as well as xenograft tumor formation in vivo induced by miR-412-3p. Conclusion: Collectively, the present study suggests an inhibitory role of MBNL1-AS1 in colon cancer by upregulating miR-412-3p-targeted MYL9. Thus, this study provides an enhanced understanding of MBNL1-AS1 along with miR-412-3p and MYL9 as therapeutic targets for colon cancer. © 2019 Elsevier Masson SAS. All rights reserved.

Corresponding author. E-mail address: dryu [email protected] (W. Yu).

https://doi.org/10.1016/j.clinre.2019.05.001 2210-7401/© 2019 Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

2

K. Zhu et al.

Introduction Colon cancer ranks third among the most frequently occurring epithelial malignancies, accompanied by high mortality rates worldwide, accounting for more than 9% of all cancer cases [1]. The foremost risk factors for the disease include obesity, physical inactivity, red and processed meat consumption, cigarette smoking and excessive alcohol consumption [2]. Although there are advances in therapy for colon cancer, prognosis of patients with colon cancer is poor, and the recurrence is high [3]. In recent years, biomarkers employed in early diagnosis, accurate staging, response assessment and early prediction of tumor relapse, have been widely explored, with the aim of clinical implementation allowing for a more optimal delivery of treatment [4]. And the theory of cancer stem cells (CSCs), maintained by subpopulations of tumor cells with features of stem cells and progenitors, have become focus of the in-depth investigation in cancer research [5]. It is reported that CSCs are largely involved in tumor progression, recurrence, and drug resistance, including colon cancer [6,7]. Targeting CSCs is known as an effective strategy for colon cancer treatment [8], while the specific mechanisms remain unclear. Thus, it is necessary to identify new treatment strategies against CSCs in colon cancer. Long noncoding RNAs (lncRNAs), comprising over 200 nucleotides in length, serve as critical modulators of cancer biology and have important impacts on cell proliferation, metastasis, and apoptosis [9,10]. For instance, lncRNA ANRIL was upregulated in non-small cell lung cancer (NSCLC), and silenced ANRIL suppressed cell proliferation and promoted cell apoptosis of NSCLC [11]. Also, lncRNAs are considered as nover biomarkers for prognosis and diagnosis of colon cancer, for example, two lncRNAs AFAP1-AS1 and ATB, as tumor promoters, are highly expressed in colon cancer, while CASC2 and CTD903, as tumor suppressors, are poorly expressed in colon cancer [12]. LncRNA muscleblind-like 1antisense RNA 1 (MBNL1-AS1) is located at the site of 3q25.1. It has reported that MBNL1 is poorly expressed in CRC cell HCT-116, while restoration of that reversed epithelial-tomesenchymal transition (EMT) [13]. MBNL1 plays roles in initiation of colorectal cancer (CRC) by reducing expression of microRNA-1307 (miR-1307) [14]. Importantly, it has been revealed that lncRNAs acted as ceRNA of miRNAs in both cancer and disease development, for instance, LINCMD1, the muscle-specific lncRNA could modulate muscle differentiation in human and mouse myoblasts by serving as a ceRNA [15]. miR-412-3p is valuable to predict overall mortality and cancer specific mortality, highlighting critical implications in cancer progression [16], and miR-412-3p was revealed to be highly expressed in extracellular vesicles from patients with oral squamous cell carcinoma (OSCC) [17]. Furthermore, myosin regulatory light chain 9 (MYL9), also known as MLC2, downregulated in prostate cancer (PCa), serves as an indicator for overall and biochemical recurrencefree survivals of patients with PCa [18]. MYL9 is essential for cytoskeletal dynamics, experimental metastasis, and tumor cell migration, and its downregulation represents the decreased median survival rate in patients with colon cancer [19,20]. Based on the abovementioned literature, we obtained colon cancer HT29 cell lines to explore and

developed xenografts in nude mice to verify the possible effects of lncRNA MBNL1-AS1 on cellular processes of colon CSCs via miR-412-3p and MYL9.

Materials and methods Ethics statement The animal experiment procedures were performed in accordance with the protocols approved by the Animal Ethics Committee of The Second Hospital of Shandong University.

Microarray analysis The microarray data related to colon cancer were downloaded from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) with ‘‘colon cancer’’ as the keyword. The Affy package of R language [21] (http://www.bioconductor.org/packages/release/bioc/ html/affy.html) was employed for standard pretreatment of microarray data, and the limma package of R language [22] (http://master.bioconductor.org/packages/release/bioc/ html/limma.html) was used to identify the differentially expressed lncRNAs and differentially expressed genes (DEGs). After correction, the p value was expressed as adj.P.Val. The lncRNAs or genes with |log2FC| > 1.5 and adj.P.Val < 0.05 were regarded as differentially expressed lncRNAs or DEGs. The heat maps of lncRNAs and DEGs were drawn.

Cell culture The colon cancer cell line HT29 from the cell bank of China Center for Type Culture Collection, Chinese Academy of Sciences (Beijing, China) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) in 5% CO2 at 37 ◦ C. The colon CSCs from the HT29 cell line were maintained in DMEM (09211, Shanghai Zhong Qiao Xin Zhou Biotechnology Co., Ltd., Shanghai, China)/F2 medium (09321, Shanghai Zhong Qiao Xin Zhou Biotechnology Co., Ltd., Shanghai, China) supplemented with 2% B27 (xy-E11452, Shanghai Xin Yu Biotech Co., Ltd., Shanghai, China), 20 ng/mL epidermal growth factor (EGF; bs-1810P, Shanghai Xin Yu Biotech Co., Ltd., Shanghai, China), 10 ng/mL bfibroblast growth factor (FGF; bs-1175P, Shanghai Xin Yu Biotech Co., Ltd., Shanghai, China), 5 ng/mL stem cell factor (SCF (WL00959, Shenyang Wan Lei Biotech Co., Ltd., Shenyang, China), 5 ␮g/mL insulin (E-82-96, Suzhou ELSBIO Co., Ltd., Suzhou, China), and 1 ng/mL hydrocortisone (600, Shenzhen Simeiquan Biotech Co., Ltd., Shenzhen, China) to prepare the serum-free medium (SFM) [23]. The HT29 cells were made into single cell suspension. After being stained with trypan-blue (ZYS0051, Shanghai Zeye Biotechnology Co., Ltd., Shanghai, China), the cells (5 × 103 cells/mL) were inoculated into the prepared the SFM medium. Morphological changes of cells were observed and photographed under the inverted microscope (MI13, Guangzhou Micro-shot Technology Co., Ltd., Guangzhou, China) at 24 h, 3rd day, 7th day, 10th day. After

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Role of lncRNA MBNL1-AS1 in colon cancer

3

the microspheres grew regularly, cells were trypsinized. The HT29 cell-derived tumorspheres were cultured in 6-well plates with CSC culture medium (2 mL/well) with 5% CO2 at 37 ◦ C.

Louis, MO, USA) for 10 min. The CSCs were photographed under a confocal laser scanning microscope to record fluorescence images.

Flow cytometry

Dual-luciferase reporter gene essay

According to the previous study [24], 5 ␮g/mL Hoechst 33342 (BB-4135, Shanghai Bestbio Co., Ltd., Shanghai, China) was used to stain the colon cancer cells in the presence or absence of 10 ␮L verapamil ((YJ-P227927, Shanghai Yiji Industrial Co., Ltd., Shanghai, China), followed by staining with 2 ␮L propidium iodide (PI, BD5012, Biogot Biotechnology Co., Ltd., Nanjing, China). The side population (SP) was detected using a flow cytometer (SP6800Z, Sony Biotechnology Inc., Tokyo, Japan). The HT29 cells and HT29 cell-derived tumorspheres were detached with trypsin with concentration adjusted to 1 × 106 cells/mL. After centrifugation, cells were re-suspended using 80 ␮L PBS supplemented with 1% FBS, incubated with 20 ␮L FcR (ab166955, Abcam Inc., Cambridge, MA, USA), 10 ␮L PECD133 antibody (03-01-4876, Cambridge Biologics, Suzhou, China) at 4 ◦ C avoiding exposure to light overnight for 10 min. After being washed with Buffer, cells were centrifuged, and re-suspended by 1 mL Buffer. The flow cytometer was utilized to analyze CD133+ expression.

According to the putative miR-412-3p binding sites in the MBNL1-AS1 and in the MYL9 mRNA 3 -untranslated region (3 -UTR), the wild-type and mutant type sequences were designed and synthesized with Xho I and Not I restriction sites added on the both end of sequences. The synthesized fragments were cloned onto the PUC57 vector (HZ0087, Shanghai Huzhen Industrial Co., Ltd., Shanghai, China). The recombinant plasmids were sub-cloned onto the psiCHECK2 vector (HZ0197, Shanghai Huzhen Industrial Co., Ltd., Shanghai, China), and transformed into Escherichia coli DH5␣ cells. CSCs were inoculated into the 6-well plates (2 × 105 cells/well) cultured for 48 h, and collected. The luciferase activity was measured using Dual-Luciferase Detection Kit from Genecopoeia (D0010, Beijing solarbio science & technology co. ltd., Beijing, China). Luciferase intensity was measured in a GLoma × 20/20 Luminometer from Promega (E5311, Shaanxi Zhongmei Biotechnology, Co., Ltd., Shaanxi, China) using the Dual-Luciferase Reporter Assay System (E1910, Promega, Madison, WI, USA).

Colony formation assay in vitro RNA-pull down assay The colon cancer HT29 CSCs were trypsinized and triturated into single cell suspension, and cells density was adjusted into 1 × 103 cells/mL. Subsequently, 8.6 mL cell suspension was mixed with 0.4 mL 5% agar for preparation of 0.3% upper-layer agar, which was transferred onto the 6-well plates supplemented with 9 mL CSC culture medium and 1 mL 5% agar (100 cells/well) and incubated with 5% CO2 at 37 ◦ C for 21 days, observed and photographed once every three days.

Cell treatment The cells at passage 3 were detached by trypsin, seeded into the 24-well plates, and transfected using lipofectamine 2000 (11668-019, Invitrogen, Carlsbad, CA, USA) with the empty vector plasmids, MBNL1-AS1 overexpression plasmids, miR-412-3p mimic, miR-412-3p inhibitor and their negative controls (NC mimic, and NC inhibitor). All plasmids used above were constructed by Shanghai Sangon Biotechnology Co. Ltd. (Shanghai, China).

Fluorescence in situ hybridization (FISH) The subcellular localization of MBNL1-AS1 was detected using the FISH kit (Roche Diagnostics GmbH, Mannheim, Germany). CSCs were incubated with digoxin-labeled hybridization solution containing MBNL1-AS1 specific probe (Sigma-Aldrich Chemical Company, St Louis, MO, USA), and the antagonistic MBNL1-AS1 probe served as negative control (NC). The nucleus was stained with 4 6-diamidino2-phenylindole (DAPI) (Sigma-Aldrich Chemical Company, St

CSCs were transfected with WT-biotinylated miR-412-3p (50 nM) and MUT-biotinylated miR-412-3p (50 nM). After transfection for 48 h, cells were collected and lysed with specific cell lysis buffer (Ambion, Austin, Texas, USA) for 10 min. A total of 50 mL sample cell lysate was preserved, and the remnant lysate was incubated with M-280 streptavidin-labelled magnetic beads (Sigma, St. Louis, MO, USA) pre-coated by RNase-free and yeast tRNA (Sigma, St. Louis, MO, USA). The antagonistic miR-412-3p probe served as NC. The total RNA was extracted using TrizoL for determination of MBNL1-AS1 level using reverse transcription quantitative polymerase chain reaction (RT-qPCR).

RNA binding protein immunoprecipitation (RIP) CSCs were lysed with RIP Lysis Buffer (N653-100, Shanghai Haoran Biotechnology Co., Ltd., Shanghai, China). The magnetic beads were washed with RIP Wash Buffer (EHJ-BVIS08102, Xiamen Huijia Biotechnology Co., Ltd., Xiamen, China) and incubated with 5 ␮g Ago2 antibody (P10502500, Shenzhen Otwo Biotechnology Co., Ltd., Shenzhen, China). A total of 900 ␮L of RIP Immunoprecipitation Buffer (P10403138, Shenzhen Otwo Biotechnology Co., Ltd., Shenzhen, China) was added into the magnetic beadantibody mixture, which was incubated with CSC lysate overnight. The magnetic bead-protein complex was rinsed with 0.5 mL RIP Wash Buffer and then incubated with 150 ␮L proteinase K buffer at 55 ◦ C for 30 min to purify RNA. The RNA was extracted using the TRIZOL for RT-qPCR.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

4

K. Zhu et al. Table 1

Primer sequences for RT-qPCR.

Genes

Sequences (5’-3’)

MBNL1-AS1

F: CTCCCGCTTCTTCTACCGAC R: TTGGTGCATTTTAAGGCGGC F: CGCCGCTTCACCTGGTCCAC R: CAGCCACAAAAGAGCACAAT F: CGAGGATGTGATTCGCAACG R: TGTTTGAGGATGCGGGTGAA F: ACACTGTGTGCCCATCTACGAGG R: AGGGGCCGGACTCGTCGTCATACT F: ATTGGAACGATACAGAGAAGATT R: GGAACGCTTCACGAATTT

miR-412-3p MYL9 ˇ-actin U6

RT-qPCR: reverse transcription quantitative polymerase chain reaction; MBNL1-AS1: muscleblind-like 1-antisense RNA 1; miR-412-3p: microRNA-412-3p; F: forward; R: reverse.

RNA isolation and quantitation Total RNA of CSCs was extracted using miRNeasy Mini Kit (217004, QIAGEN, Germany). The primer sequences were synthesized by TaKaRa Biotechnology Co. Ltd., (Liaoning, China) (Table 1). Next, total RNA was reversely transcribed into complementary DNA (cDNA) using PrimeScript RT kit (RR036A, TaKaRa Biotechnology Co. Ltd, Liaoning, China). The ABI7500 quantitative PCR instrument (7500, ABI Company, Oyster Bay, NY) was employed to conduct real time ® qPCR based on the instructions of SYBR Premix Ex TaqTM II kit (RR820A, TaKaRa Biotechnology Co. Ltd, Liaoning, China). U6 was regarded as internal reference of MBNL1AS1 and miR-412-3p, and ␤-actin as the internal reference of MYL9. The target gene expression was calculated using the 2-Ct method [25].

Western blot analysis Total protein was extracted using radioimmunoprecipitation assay (RIPA) kit (R0010, Beijing Solarbio Science & Technology Co. Ltd., Beijing, China), separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred onto a nitrocellulose filter membrane. The membrane was blocked with 5% bovine serum albumin (BSA) at 37 ◦ C for 1 h, and incubated overnight at 4 ◦ C with the primary antibodies against rabbit anti-human CD133 (ab16518, 1: 1000), MYL9 (ab64161, 1: 1000), proliferating cell nuclear antigen (PCNA) (ab18197, 1: 1000), high mobility group box 1 (HMGB1) (ab191583, 1: 1000), B-cell lymphoma/leukemia-2 (Bcl-2) (ab59348, 1: 800), and Bcl-2-associated X protein (Bax) (ab53154, 1: 800). The above-mentioned antibodies were from Abcam, Cambridge (MA, USA). The membrane was incubated with the horseradish peroxidase (HRP)-labeled goat ant-rabbit immunoglobulin G (IgG) (1: 5000, Beijing Zhongshan Biotechnology Co. Ltd., Beijing, China) for 1 h. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the internal reference.

3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay After transfection for 48 h, CSCs were collected, counted, seeded into the 96-well plates (3 × 103 ∼ 6 × 103 cells/well). Six replicates were used. Next, cells in each well were added with the prepared 5 mg/mL MTT solution (20 ␮L) and incubated at 37 ◦ C for 2 h. Cells in each well were added with 150 ␮L Dimethyl Sulphoxide (DMSO). The optical density (OD) value was measured using enzyme-linked immune analyzer (NYW-96 M, Beijing Nuoyawei Instruments Co., Ltd., Beijing, China) at the wavelength of 570 nm. Cell viability curve was mapped with time points (24 h, 48 h, and 72 h) as the abscissa and OD value as the ordinate.

Transwell assay CSCs were re-suspended in the Roswell Park Memorial Institute (RPMI) 1640 medium (1 × 105 cells/mL). Then 200 ␮L cell suspension was added into the apical chamber coated with Matrigel, and 600 ␮L RPMI 1640 medium containing 20% FBS was added into the basolateral chamber. After incubation for 24 h, the cells invading into the basolateral chamber through the Matrigel were fixed in 4% paraformaldehyde for 15 min, and stained in 0.5% crystal violet for 15 min. Five fields (200×) were randomly selected and photographed under an inverted microscope (XDS-800D, Shanghai Caikang Optical Instruments Co., Ltd., Shanghai, China). Cells penetrated the membrane were counted. Three replicates were set. The assessment of cell migration was performed as described above except that no Matrigel was used to coat the apical chamber.

Terminal deoxynucleotidyl transferase (TdT)-mediated dNTP nick end-labeling (TUNEL) staining CSCs were incubated in osmotic solution containing 0.2% Triton X-100 and 0.1% sodium citrate for 5 min, followed by incubation with TUNEL reaction mixture (11684817910, Beijing Solarbio Science & Technology, Co., Ltd., Beijing, China), at 37 ◦ C for 30 min. Subsequently, the cells were incubated with the TUNEL-POD transforming agent (AP005, Shanghai 7 Sea Biotechnology Co., Ltd., Shanghai, China) at 37 ◦ C for 30 min. After that, the cells were stained with DAB for 5—10 min, and counter-stained with hematoxylin (PT001, Shanghai Bogoo Biotechnology Co., Ltd., Shanghai, China). Following being differentiated by hydrochloric acid alcohol, the cells were dehydrated by alcohol, cleared by xylene, and mounted. The cells with green fluorescence were regarded as the apoptotic positive cells. Five high-power visual fields were randomly selected under the fluorescence microscope. The percentage of positive cells in total cells was considered to be apoptotic index (AI).

Xenografts in nude mice About 5 × 107 cells/mL cells were subcutaneously injected into armpit of 56 NOD/SCID nude mice (aged 4—5 weeks and weighing 22—24 g, Institute of Medical Laboratory Animals,

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Role of lncRNA MBNL1-AS1 in colon cancer

5

Peking Union Medical College, Beijing, China), which were fed at Animal Experimental Center of The Second Hospital of Shandong University (No. 159). At 28 days after injection, nude mice were euthanized, and the tumor were resected and fixed in the 4% paraformaldehyde for 24 h.

Results

Statistical analysis

The first step in calculating our results in our experiment, the objective of understanding whether MBNL1-AS1 and MYL9 may affect colon cancer, began with screening DEGs and lncRNAs from colon cancer-related microarray datasets GSE41328 and GSE75970, based on the GEO and TCGA database. The MBNL1-AS1 was downregulated in colon cancer (Fig. 1A and B). Moreover, through analysis of GSE75970, MYL9 was also downregulated in colon cancer (Fig. 1C), which suggested that MYL9 expression was related to the prognosis of colon cancer based on the TCGA database (Fig. 1D).

Statistical analysis was performed using SPSS 21.0 (IBM Corp. Armonk, NY, USA), and the measurement data were expressed as mean ± standard deviation. Comparisons of data with normal distribution and homogeneity of variance between two groups were conducted by unpaired t-test, and comparisons among multiple groups were tested by one-way analysis of variance (ANOVA) and repeated measurement Anova with Tukey’s post-hoc test used. P < 0.05 was statistically significant.

MBNL1-AS1 and MYL9 are predicted to be downregulated in colon cancer

Figure 1 LncRNA MBNL1-AS1 may affect colon cancer via MYL9. A. Heat map of differentially expressed lncRNAs from GSE41328. B. MBNL1-AS1 expression in colon cancer tissues and adjacent normal tissues in TCGA database. C. Heat map of differentially expressed genes from GSE75970. D. The overall survival of colon cancer patient with different level of MYL9 expression. MBNL1-AS1, muscleblind-like 1-antisense RNA 1; TCGA, The Cancer Genome Atlas.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

6

ARTICLE IN PRESS K. Zhu et al.

Figure 2 CSCs are enriched in HT29 cells. A. Representative images of cultured HT29 cells (200×). B. SP cell content in HT29 cells. C. Histogram of SP cell content in HT29 cells. D. CD133 positive cells and HT29 microspheres. E. Percentage of CD133 positive cells in HT29 microspheres. F. Colony formation of HT29 microspheres. G. CD133 protein level in HT29 cells and HT29 microspheres measured using Western blot analysis. H. CD133 protein band pattern in HT29 cells and HT29 microspheres detected using Western blot analysis. * P < 0.05 vs. HT29 cells treated with Verapamil or the control group; measurement data were depicted as mean ± standard deviation; comparisons between two groups were analyzed using unpaired t test; n = 8; the experiment was repeated three times. CSCs, cancer stem cells.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Role of lncRNA MBNL1-AS1 in colon cancer

7

Figure 3 Elevated MBNL1-AS1 represses CSC proliferation, invasion, migration, and tumor formation but enhanced apoptosis in colon cancer. A. HT29 CSC cell viability detected using the MTT assay. B—C. HT29 CSC cell migration detected using Transwell assay (200×). D—E. HT29 CSC cell invasion detected using Transwell assay (200×). F—G. HT29 CSC cell apoptosis detected using TUNEL staining (200×). H—J. Tumor growth, tumor volume and tumor weight in vivo using xenografts in nude mice. K-L, protein band patterns and protein levels of MYL9, CD133, PCNA, HMGB1, Bcl-2, and Bax using Western blot analysis. * P < 0.05 vs. cells treated with empty vector; measurement data were depicted as mean ± standard deviation; comparisons among multiple groups were tested by two-way ANOVA; comparisons between two groups were analyzed using unpaired t test; n = 8; the cell experiments were repeated three times. MBNL1-AS1, muscleblind-like 1-antisense RNA 1; CSC, cancer stem cell; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide; TUNEL, terminal deoxynucleotidyl transferase (TdT)-mediated dNTP nick end-labeling; PCNA, proliferating cell nuclear antigen; Bcl-2, B-cell lymphoma 2; Bax, Bcl-2-associated X protein.

Identification of HT29 cells containing abundant CSCs HT29 cells grew adherently under conventional culture conditions. HT29 stem cells showed suspension growth in serum-free medium, and gradually expanded from a single cell to a well-defined tumor stem cell sphere (HT29 microsphere). HT29 microspheres cultured in serum-free medium were replaced by medium containing serum, which showed that HT29 microspheres grew adherently, and a single tumor cell gradually dissociated around the CSC sphere until CSC sphere disappeared and monolayer tumor cells were formed (Fig. 2A). The results of flow cytometry (Fig. 2B—E) showed that compared with HT29 cells treated with verapamil, HT29 cells stained with Hoechst33342 showed increased cells in the SP. Moreover, CD133 positive rate was higher in HT29 cells added with Hoechst33342 than that in HT29 cells added with verapamil (P < 0.05). Due to the similar characteristics between SP cells and CSCs, high ratio of SP cells suggested abundant CSCs in HT29 cells.

The colony formation assay (Fig. 2F) displayed that round and transparent HT29 microspheres appeared in soft agar. At the 3rd day, a few cells formed tiny cell colonies, and at 12th day, a large number of colonies were visible. Furthermore, Western blot analysis (Fig. 2G—H) presented elevated CD133 protein level in HT29 cells stained with Hoechst33342 in comparison to HT29 cells treated with verapamil (P < 0.05). The above findings verified that HT29 cells contained abundant CSCs, and HT29 CSCs were selected for experiments.

MBNL1-AS1 suppresses CSC proliferation, invasion, migration, and tumor formation but promote CSC apoptosis in colon cancer After selection of CSCs, we examined the biological functions of MBNL1-AS1 in CSCs of colon cancer. Considering the low expression of MBNL1-AS1 in colon cancer, we overexpressed MBNL1-AS1 in CSCs. The results of MTT assay (Fig. 3A) showed that the OD value at 48 h and at 72 h significantly decreased by MBNL1-AS1 overexpression (P < 0.05).

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

8

ARTICLE IN PRESS K. Zhu et al.

Figure 4 Overexpressed MBNL1-AS1 suppresses miR-412-3p to promote MYL9. A. Localization of MBNL1-AS1 in HT29 CSCs using LncRNA sub-localization prediction website. B. Localization of MBNL1-AS1 in HT29 CSCs using FISH (400×). C. Binding site between miR-412-3p and MBNL1-AS1. D. Binding site between miR-412-3p and MYL9-3’UTR. E. Luciferase activity of wt-miR-412-3p/MBNL1AS1 and wt-miR-412-3p/MYL9. F. Enrichment of MBNL1-AS1 using RNA pull-down. G. Binding of lncRNA MBNL1-AS1 and miR-412-3p to Ago2 verified using RIP. H. Expression of MBNL1-AS1, miR-412-3p, and MYL9 measured using RT-qPCR. * P < 0.05 vs. HT29 CSCs treated with NC mimic, Bio-probe NC, NC inhibitor, or empty vector; measurement data were depicted as mean ± standard deviation; comparisons between two groups were analyzed using unpaired t test; comparisons among multiple groups were tested by one-way Anova; n = 3; the experiment was repeated three times. LncRNA MBNL1-AS1: long non-coding RNA muscleblind-like 1-antisense RNA 1; miR-412-3p: microRNA-412-3p; FISH: fluorescence in situ hybridization; UTR: untranslated region; wt: wild type; RIP: RNA binding protein immunoprecipitation; RT-qPCR: reverse transcription quantitative polymerase chain reaction; NC: negative control; Anova, analysis of variance.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Role of lncRNA MBNL1-AS1 in colon cancer

9

Figure 5 MBNL1-AS1 overexpression inhibits CSC viability in colon cancer through depletion of miR-412-3p. A. Viability of colon CSCs detected using MTT assay. B. PCNA protein level measured using Western blot analysis. C. PCNA protein band patterns detected using Western blot analysis. * P < 0.05 vs. CSCs treated with NC mimic; #P < 0.05 vs. CSCs treated with NC inhibitor; measurement data were depicted as mean ± standard deviation; comparisons among multiple groups were tested by one-way Anova; n = 3; the experiment was repeated three times. miR-412-3p, microRNA-412-3p; CSC, cancer stem cell; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5diphenyl tetrazolium bromide; PCNA, proliferating cell nuclear antigen; NC, negative control; Anova, analysis of variance.

Figure 6 MBNL1-AS1 upregulation inhibits CSC invasion and migration in colon cancer through depletion of miR-412-3p. A. HT29 CSC cell migration after miR-412-3p elevation or depletion detected using Transwell assay (200×). B. HT29 CSC cell migration rate after miR-412-3p elevation or depletion measured using Transwell assay. C. HT29 CSC cell invasion after miR-412-3p elevation or depletion detected using Transwell assay (200×). D. HT29 CSC cell invasion rate after miR-412-3p elevation or depletion measured using Transwell assay. E. HMGB1 protein level after miR-412-3p elevation or depletion measured using Western blot analysis. F. HMGB1 protein band patterns after miR-412-3p elevation or depletion detected using Western blot analysis. * P < 0.05 vs. HT29 CSCs treated with NC mimic; #P < 0.05 vs. HT29 CSCs treated with NC inhibitor; measurement data were depicted as mean ± standard deviation; comparisons among multiple groups were tested by one-way Anova; n = 3; the experiment was repeated three times. miR-412-3p, microRNA-412-3p; CSC: cancer stem cell; HMGB1, high mobility group box 1; NC: negative control; Anova, analysis of variance.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

10 Moreover, the Transwell assay (Fig. 3B—E), TUNEL staining (Fig. 3F—G), and xenografts in nude mice (Fig. 3H—J) displayed that after overexpression of MBNL1-AS1, CSCs showed reduced migration, invasion, and tumor formation, and increased apoptosis (all P < 0.05). At last, Western blot analysis (Fig. 3K—L) indicated that protein levels of CSC marker CD133, PCNA, HMGB1, and BcL-2 markedly decreased and that of MYL9 and Bax obviously elevated in CSCs overexpressing MBNL1-AS1 (all P < 0.05). These results suggested that restored MBNL1-AS1 inhibited CSC proliferation, invasion, migration, and tumor formation but enhanced CSC apoptosis in colon cancer.

Restored MBNL1-AS1 inhibits miR-412-3p to promote MYL9 Following the results that MBNL1-AS1 elevation contributed to the prevention of CSC proliferation, invasion, migration, and tumor formation in colon cancer, we identified the relationship among MBNL1-AS1, miR-412-3p, and MYL9 to identify the underlying molecular mechanisms. Initially, the lncRNA cellular sub-localization website presented that MBNL1-AS1 mainly concentrated in the cytoplasm (Fig. 4A), which was also verified using FISH (Fig. 4B). After treatment with NC probe, only the nucleus was stained blue by DAPI. Under the conditions of Merge, there was only blue fluorescence in the nucleus, but no red fluorescence. However, MBNL1-AS1-probed red fluorescence was visualized and the nucleus was stained blue by DAPI in CSCs. Under the conditions of Merge, blue fluorescence was observed in the nucleus and MBNL1-AS1-probed red fluorescence was located outside the nucleus, suggesting that MBNL1-AS1 mainly concentrated in the cytoplasm. As shown in Fig. 4C—D, bioinformatics analysis showed that there were specific binding regions between MBNL1-AS1 and miR-412-3p, and between MYL9 and miR-412-3p. miR412-3p could negatively regulates MYL9, and competitively bind to MBNL1-AS1. MBNL1-AS1 and MYL9 were targets of miR-412-3p, which were verified by dual-luciferase reporter gene assay (Fig. 4E). Moreover, the luciferase activity was reduced after co-transfection of miR-412-3p mimic with wtmiR-412-3p/MBNL1-AS1 or wt-miR-412-3p/MYL9 (p < 0.05), while the luciferase activity of mut-3’UTR showed no significant difference (P > 0.05), suggesting that miR-412-3p specifically bound to MBNL1-AS1 and MYL9. Next, the results of RNA-pull down (Fig. 4F) revealed that after treatment with bio-miR-412-3p-wt, CSCs showed evidently increased MBNL1-AS1 expression (P < 0.05), while no obvious difference was found in cells probed with bio-miR-412-3p-mut (P > 0.05), which implied that bio-miR412-3p-wt could promote the enrichment of MBNL1-AS1 around it. It could be concluded that MBNL1-AS1 could competitively bind to miR-412-3p and reduce miR-412-3p expression. Moreover, the findings of RIP (Fig. 4G) showed that in contrast to CSCs transfected with NC inhibitor, the amount of MBNL1-AS1 coprecipitated with Ago2 antibody in cells transfected with miR-412-3p inhibitor markedly elevated, while that of miR-412-3p co-precipitated with Ago2 antibody significantly decreased (both P < 0.05). Furthermore, the amount of MBNL1-AS1 and miR-412-3p precipitated by Ago2 antibody

K. Zhu et al. in cells treated with IgG showed obvious reduction (both P < 0.05). These results revealed that MBNL1-AS1 acted as a ceRNA of miR-412-3p. Finally, RT-qPCR (Fig. 4H) displayed that CSCs transfected with MBNL1-AS1 overexpression plasmid or miR-412-3p inhibitor showed elevated expression of MBNL1-AS1 and MYL9 and decreased miR-412-3p expression, while CSCs transfected with miR-412-3p mimic presented the opposite results (all P < 0.05). However, MBNL1-AS1 overexpression reversed the miR-412-3p-caused reduction of MYL9 (P < 0.05), suggesting that elevated MBNL1-AS1 repressed miR-412-3p so as to upregulate MYL9.

Overexpressed MBNL1-AS1 inhibits CSC viability in colon cancer through inhibiting miR-412-3p With the results in the above section detailing the relationship among MBNL1-AS1 miR-412-3p, and MYL9, the focus of the experiment shifted to effect of miR-412-3p on CSC viability in colon cancer. The results of MTT assay (Fig. 5A) showed that after transfection with miR-412-3p mimic at 48 h and 72 h, CSCs displayed evidently increased OD value (P < 0.05), but exhibited an opposite result after transfection with miR-412-3p inhibitor (P < 0.05). Meanwhile, the Western blot analysis (Fig. 5B—C) presented that after upregulation of miR-412-3p, CSCs showed increased PCNA protein level, but decreased PCNA protein level was found in CSCs after inhibition of miR-412-3p (P < 0.05). Moreover, increases in CSC viability and PCNA protein level triggered by upregulation of miR-412-3p were reversed by MBNL1-AS1 overexpression. The obtained data suggested that depleted miR-412-3p repressed CSC viability in colon cancer, and MBNL1-AS1 impeded CSC viability through inhibiting miR-412-3p.

Overexpressed MBNL1-AS1 inhibits CSC invasion and migration in colon cancer through inhibiting miR-412-3p With the data that verified role of miR-412-3p in CSC viability in colon cancer in hand, the focus of the experiment further studied the mechanisms of miR-412-3p in CSC invasion and migration in colon cancer. Transwell assay (Fig. 6AD) displayed elevated cell migration and invasion in CSCs after upregulation of miR-412-3p, but inhibited cell migration and invasion in CSCs after inhibition of miR-412-3p (both P < 0.05). However, overexpressed MBNL1-AS1 reversed the elevated cell migration and invasion induced by upregulated miR-412-3p expression. Meanwhile, the Western blot analysis (Fig. 6E-F) indicated that after upregulation of miR-412-3p, CSCs showed evidently increased HMGB1 protein level (P < 0.05), which was rescued by overexpressed MBNL1-AS1. On the contrary, after downregulation of miR-412-3p, HMGB1 protein level was reduced in CSCs (P < 0.05). These results suggested that upregulation of MBNL1-AS1 suppressed CSC invasion and migration in colon cancer through downregulation of miR412-3p.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Role of lncRNA MBNL1-AS1 in colon cancer

11

Figure 7 MBNL1-AS1 overexpression enhances apoptosis of CSCs in colon cancer through depletion of miR-412-3p. A. HT29 CSC cell apoptosis after miR-412-3p elevation or depletion detected using TUNEL staining (200×). B. HT29 CSC cell apoptosis rate after miR-412-3p elevation or depletion measured using TUNEL staining. C. Protein levels of BcL-2 and Bax after miR-412-3p elevation or depletion measured using Western blot analysis. D. Protein band patterns of BcL-2 and Bax after miR-412-3p elevation or depletion detected using Western blot analysis. * P < 0.05 vs. HT29 CSCs treated with NC mimic; #P < 0.05 vs. HT29 CSCs treated with NC inhibitor; measurement data were depicted as mean ± standard deviation; comparisons among multiple groups were tested by one-way Anova; n = 3; the experiment was repeated three times. miR-412-3p: microRNA-412-3p; CSCs: cancer stem cells; TUNEL: terminal deoxynucleotidyl transferase (TdT)-mediated dNTP nick end-labeling; Bcl-2: B-cell lymphoma 2; Bax: Bcl-2-associated X protein; NC: negative control; Anova: analysis of variance.

Overexpression of MBNL1-AS1 promotes apoptosis of CSCs in colon cancer via inhibition of miR-412-3p After determination of miR-412-3p expression in invasion and migration of colon CSCs, we adopted TUNEL staining to further examine the molecular mechanisms of miR-412-3p in apoptosis of colon CSCs. The TUNEL staining (Fig. 7A and B) displayed inhibited apoptosis in CSCs overexpressing miR-412-3p, but promoted apoptosis in CSCs after miR-412-3p was inhibited (P < 0.05). However, overexpressed MBNL1-AS1 could reversed the inhibitory role of miR-412-3p in the apoptosis of CSCs. Moreover, the Western blot analysis (Fig. 7C—D) revealed that after transfection with miR-412-3p mimic, CSCs showed markedly increased BCL-2 protein level and reduced Bax (both P < 0.05), while after inhibition of miR-412-3p, opposite results were measured in CSCs (both P < 0.05). Additionally, upregulation of both MBNL1-AS1 and miR412-3p led to lower protein level of Bcl-2 but higher protein level of Bax than miR-412-3p alone, suggesting that upregulated MBNL1-AS1 facilitated apoptosis of CSCs in colon cancer by downregulation of miR-4123p.

MBNL1-AS1 inhibits tumor formation of CSCs in colon cancer via inhibition of miR-412-3p With the findings of MBNL1-AS1 and miR-412-3p regulating molecular mechanisms of colon CSCs, our last objective of observing whether depleted miR-412-3p affected xenograft tumor formation in nude mice was evaluated. The findings of xenografts in nude mice (Fig. 8) revealed that after upregulating miR-412-3p, mice showed suppressed tumor formation, but promoted tumor formation after inhibiting miR-412-3p (P < 0.05), Besides, overexpressed MBNL1-AS1 resulted in suppressed tumor formation which was induced by miR-412-3p, which indicated that upregulated MBNL1AS1 repressed tumor formation of CSCs in colon cancer by downregulating miR-412-3p.

Discussion Colon cancer, one of the most universal and aggressive malignant tumors, is regarded to be the third major cause of cancer-related death around the world [3]. In recent years, it has been attracted great attention that the identification of colon CSC and their properties have emerged as the effective anticancer strategy [6,8]. Accumulating evidences have indicated that lncRNAs are involved in colon

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

12

ARTICLE IN PRESS K. Zhu et al.

Figure 8 MBNL1-AS1 overexpression represses tumor formation of CSCs in colon cancer through depletion of miR-412-3p. A. Tumor growth in vivo using xenografts in nude mice. B. Tumor volume in vivo using xenografts in nude mice. C. Tumor weight in vivo using xenografts in nude mice. * P < 0.05 vs. mice injected with NC mimic-transfected HT29 CSCs; #P < 0.05 vs. mice injected with NC inhibitor-transfected HT29 CSCs; measurement data were depicted as mean ± standard deviation; comparisons among multiple groups were tested by one-way and two-way Anova; n = 8; the experiment was repeated three times. miR-412-3p: microRNA-412-3p; CSCs: cancer stem cells; NC: negative control; Anova: analysis of variance.

Figure 9 MBNL1-AS1 acting as ceRNA of miR-412-3p involved in biological activity of colon CSCs via regulation of MYL9. In colon CSCs, miR-412-3p could negatively regulate MYL9. MBNL1-AS1 acted as a sponge of miR-412-3p and reversed the negative regulation of MYL9 by miR-412-3p, hence resulting in a decline of the proliferation, migration, and invasion yet an increase in apoptosis of colon CSCs. MBNL1-AS1, muscleblind-like 1-antisense RNA 1; ceRNA, competing endogenous RNA; miR-412-3p, microRNA-412-3p; CSCs, cancer stem cells.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

Role of lncRNA MBNL1-AS1 in colon cancer cancer via control of CSCs [26]. The current study was performed with the objective of examining the suppressive role of MBNL1-AS1 in colon cancer in an attempt to elucidate the potential mechanism associated with miR-412-3p and MYL9. The key observations of the current study revealed that CSC cells could be sheltered from proliferation, invasion, and migration in colon cancer by activating MYL9 through the inhibition of miR-412-3p when MBNL1-AS1 expression was restored. Initially, the current study unveiled that MBNL1-AS1 is downregulated in colon cancer. Aberrant expression of several lncRNAs involve in development and progression of multiple human cancers like colon cancer [3,27]. MBNL1 is demonstrated to be poorly expressed in NSCLC [28]. MBNL1 is reported to exert great effects on CRC by reducing expression of miRNAs [11]. Moreover, the current study revealed that MBNL1-AS1 served as a sponge of miR-412-3p. A previous study has indicated that lncRNAs could regulate the mRNA profile in human cancers [29]. miRNAs are aberrantly expressed in the pathogenesis of several diseases, including human cancers [30]. Specific miRNAs, such as miR-195 and miR-20a, are differentially expressed in colon cancer tissues, which was further confirmed in connection with a previously meta-analysis with miRNA microarray datasets [31]. Furthermore, miR-196b is also confirmed to be upregulated in colon cancer [32]. A biological analysis and relevance analysis results confirmed that MYL9 was a target gene of miR-301b-3p, and shared a positive correlation with MBNL1-AS1. The critical role of MYL9 has been verified in some human cancers and it is demonstrated to be downregulated in human colon cancer [19]. These evidences support that MBNL1-AS1 and MYL9 are downregulated in colon cancer, and MBNL1-AS1 increases MYL9 by serving as the ceRNA of miR-412-3p. In addition, the data in the present study confirmed that upregulated MBNL1-AS1 or inhibited miR-412-3p suppresses CSC proliferation, invasion, and tumor formation, and promote apoptosis in colon cancer. Growing evidences have implied that lncRNAs are involved in the proliferation, migration, invasion and apoptosis of colon cancer [33]. Luo et al. have revealed that depleted lncRNA-RP11-317J10.2 enhances cell proliferation, invasion, and tumor growth, which represents the poor prognosis in CRC [34], suggesting that restored lncRNA-RP11-317J10.2 suppresses CRC. Moreover, aberrant expression of miRNAs is implicated in the initiation, progression and angiogenesis of colon cancer, as a consequence of the progressive accumulation of genetic modifications in oncogenes or anti-oncogenes in colonic epithelium [35]. By regulating expression of related signaling pathways, cytoskeleton and membrane proteins, miRNAs give tumor stem cells the macro-biological behavior of recurrence and metastasis [36]. It has been reported that miR-196b facilitates tumorigenesis of colon cancer [32], which indicated that inhibition of miR-196b exerts inhibitory role in colon cancer development. Additionally, MYL9 is also known to be implicated in cell biological processes, such as cell migration [37]. The abovementioned findings suggest that MBNL1-AS1 elevation suppresses proliferation, invasion, and migration and promotes apoptosis of CSCs in colon cancer by inhibiting miR-412-3p via upregulation of MYL9.

13

Conclusion In conclusion, the current study provided evidence that restoration of MBNL1-AS1 could decelerate the occurrence and progression of colon cancer by which MBNL1-AS1 restoration serves as the ceRNA against miR-412-3p to upregulate MYL9 and suppress the proliferation, invasion, and migration and facilitate apoptosis of CSC cells in colon cancer (Fig. 9). Thus, MBNL1-AS1-miR-412-3p-MYL9 network facilitates a novel aspect of the treatment of patients with colon cancer. However, certain limitations exist in the present study. Due to the lack of clinical efficacy and clinical data to support our findings, additional prospective studies are required in order to validate the identified network through multicenter clinical trials.

Authors’ contributions Kongxi Zhu and Yunxia Wang participated in the conception and design of the study. Lan Liu and Shuai Li performed the analysis and interpretation of data. Kongxi Zhu and Weihua Yu contributed to drafting the article. All authors have read and approved the final submitted manuscript.

Disclosure of interest The authors declare that they have no competing interest.

Acknowledgments The authors would like to acknowledge the helpful comments on this paper received from the reviewers.

References [1] Margaritescu C, Pirici D, Cherciu I, Barbalan A, Cartana T, Saftoiu A. CD133/CD166/Ki-67 triple immunofluorescence assessment for putative cancer stem cells in colon carcinoma. J Gastrointestin Liver Dis 2014;23(2): 161—70. [2] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61(2):69—90. [3] Yue B, Qiu S, Zhao S, Liu C, Zhang D, Yu F, et al. LncRNAATB mediated E-cadherin repression promotes the progression of colon cancer and predicts poor prognosis. J Gastroenterol Hepatol 2016;31(3):595—603. [4] Lawrence N, Hinder V, Murray M, Macapagal J, Thompson P, Sharples K, et al. Transient elevation in serum carcinoembryonic antigen while on adjuvant chemotherapy for colon cancer: Is this of prognostic importance? Asia Pac J Clin Oncol 2017;13(2):e124—31. [5] Perona R, Lopez-Ayllon BD, de Castro Carpeno J, BeldaIniesta C. A role for cancer stem cells in drug resistance and metastasis in non-small-cell lung cancer. Clin Transl Oncol 2011;13(5):289—93. [6] Kozovska Z, Gabrisova V, Kucerova L. Colon cancer: cancer stem cells markers, drug resistance and treatment. Biomed Pharmacother 2014;68(8):911—6. [7] Zhao C, Setrerrahmane S, Xu H. Enrichment and characterization of cancer stem cells from a human non-small cell lung cancer cell line. Oncol Rep 2015;34(4):2126—32.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001

+Model CLINRE-1276; No. of Pages 14

ARTICLE IN PRESS

14 [8] Todaro M, Francipane MG, Medema JP, Stassi G. Colon cancer stem cells: promise of targeted therapy. Gastroenterology 2010;138(6):2151—62. [9] Yang J, Lin J, Liu T, Chen T, Pan S, Huang W, et al. Analysis of lncRNA expression profiles in non-small cell lung cancers (NSCLC) and their clinical subtypes. Lung Cancer 2014;85(2):110—5. [10] Prensner JR, Chinnaiyan AM. The emergence of lncRNAs in cancer biology. Cancer Discov 2011;1(5):391—407. [11] Nie FQ, Sun M, Yang JS, Xie M, Xu TP, Xia R, et al. Long noncoding RNA ANRIL promotes non-small cell lung cancer cell proliferation and inhibits apoptosis by silencing KLF2 and P21 expression. Mol Cancer Ther 2015;14(1):268—77. [12] Deng H, Wang JM, Li M, Tang R, Tang K, Su Y, et al. Long noncoding RNAs: new biomarkers for prognosis and diagnosis of colon cancer. Tumour Biol 2017;39(6) [1010428317706332]. [13] Tang L, Zhao P, Kong D. Muscleblindlike 1 destabilizes Snail mRNA and suppresses the metastasis of colorectal cancer cells via the Snail/Ecadherin axis. Int J Oncol 2019;54(3):955—65. [14] Tang R, Qi Q, Wu R, Zhou X, Wu D, Zhou H, et al. The polymorphic terminal-loop of pre-miR-1307 binding with MBNL1 contributes to colorectal carcinogenesis via interference with Dicer1 recruitment. Carcinogenesis 2015;36(8):867—75. [15] Tay Y, Rinn J, Pandolfi PP. The multilayered complexity of ceRNA crosstalk and competition. Nature 2014;505(7483):344—52. [16] Lenherr SM, Tsai S, Silva Neto B, Sullivan TB, Cimmino CB, Logvinenko T, et al. MicroRNA expression profile identifies high grade, non-muscle-invasive bladder tumors at elevated risk to progress to an invasive phenotype. Genes (Basel) 2017;8(2), pii: E77. [17] Gai C, Camussi F, Broccoletti R, Gambino A, Cabras M, Molinaro L, et al. Salivary extracellular vesicle-associated miRNAs as potential biomarkers in oral squamous cell carcinoma. BMC Cancer 2018;18(1):439. [18] Huang YQ, Han ZD, Liang YX, Lin ZY, Ling XH, Fu X, et al. Decreased expression of myosin light chain MYL9 in stroma predicts malignant progression and poor biochemical recurrence-free survival in prostate cancer. Med Oncol 2014;31(1):820. [19] Yan Z, Li J, Xiong Y, Xu W, Zheng G. Identification of candidate colon cancer biomarkers by applying a random forest approach on microarray data. Oncol Rep 2012;28(3):1036—42. [20] Wang JH, Zhang L, Huang ST, Xu J, Zhou Y, Yu XJ, et al. Expression and prognostic significance of MYL9 in esophageal squamous cell carcinoma. PLoS One 2017;12(4):e0175280. [21] Gautier L, Cope L, Bolstad BM, Irizarry RA. affy—analysis of Affymetrix GeneChip data at the probe level. Bioinformatics 2004;20(3):307—15. [22] Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 2004;3 [Article 3]. [23] Wang W, Zhang G, Yang J, Gu H, Ding L, Yu H, et al. Digital gene expression profiling analysis of DNA repair pathways in colon

K. Zhu et al.

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

cancer stem population of HT29 cells. Acta Biochim Biophys Sin (Shanghai) 2017;49(1):90—100. Yu CC, Hu FW, Yu CH, Chou MY. Targeting CD133 in the enhancement of chemosensitivity in oral squamous cell carcinoma-derived side population cancer stem cells. Head Neck 2016;38:E231—8. Ayuk SM, Abrahamse H, Houreld NN. The role of photobiomodulation on gene expression of cell adhesion molecules in diabetic wounded fibroblasts in vitro. J Photochem Photobiol B 2016;161:368—74. Dou J, Ni Y, He X, Wu D, Li M, Wu S, et al. Decreasing lncRNA HOTAIR expression inhibits human colorectal cancer stem cells. Am J Transl Res 2016;8(1):98—108. Svoboda M, Slyskova J, Schneiderova M, Makovicky P, Bielik L, Levy M, et al. HOTAIR long non-coding RNA is a negative prognostic factor not only in primary tumors, but also in the blood of colorectal cancer patients. Carcinogenesis 2014;35(7):1510—5. Yu H, Xu Q, Liu F, Ye X, Wang J, Meng X. Identification and validation of long noncoding RNA biomarkers in human nonsmall-cell lung carcinomas. J Thorac Oncol 2015;10(4):645—54. Zhang J, Yuan Y, Wei Z, Ren J, Hou X, Yang D, et al. Crosstalk between prognostic long noncoding RNAs and messenger RNAs as transcriptional hallmarks in gastric cancer. Epigenomics 2018;10(4):433—43. Zhang JG, Wang JJ, Zhao F, Liu Q, Jiang K, Yang GH. MicroRNA21 (miR-21) represses tumor suppressor PTEN and promotes growth and invasion in non-small cell lung cancer (NSCLC). Clin Chim Acta 2010;411(11—12):846—52. Li HG, Zhao LH, Bao XB, Sun PC, Zhai BP. Meta-analysis of the differentially expressed colorectal cancer-related microRNA expression profiles. Eur Rev Med Pharmacol Sci 2014;18(14):2048—57. Fantini S, Salsi V, Reggiani L, Maiorana A, Zappavigna V. The miR-196b miRNA inhibits the GATA6 intestinal transcription factor and is upregulated in colon cancer patients. Oncotarget 2017;8(3):4747—59. Ouyang S, Zheng X, Zhou X, Chen Z, Yang X, Xie M. LncRNA BCAR4 promotes colon cancer progression via activating Wnt/beta-catenin signaling. Oncotarget 2017;8(54):92815—26. Luo J, Xu LN, Zhang SJ, Jiang YG, Zhuo DX, Wu LH, et al. Downregulation of LncRNA-RP11-317J10.2 promotes cell proliferation and invasion and predicts poor prognosis in colorectal cancer. Scand J Gastroenterol 2018;53(1):38—45. Ramalingam S, Subramaniam D, Anant S. Manipulating miRNA Expression: A Novel Approach for Colon Cancer Prevention and Chemotherapy. Curr Pharmacol Rep 2015;1(3):141—53. Fang Y, Xiang J, Chen Z, Gu X, Li Z, Tang F, et al. miRNA expression profile of colon cancer stem cells compared to non-stem cells using the SW1116 cell line. Oncol Rep 2012;28(6):2115—24. Luo XG, Zhang CL, Zhao WW, Liu ZP, Liu L, Mu A, et al. Histone methyltransferase SMYD3 promotes MRTF-A-mediated transactivation of MYL9 and migration of MCF-7 breast cancer cells. Cancer Lett 2014;344(1):129—37.

Please cite this article in press as: Zhu K, et al. Long non-coding RNA MBNL1-AS1 regulates proliferation, migration, and invasion of cancer stem cells in colon cancer by interacting with MYL9 via sponging microRNA-412-3p. Clin Res Hepatol Gastroenterol (2019), https://doi.org/10.1016/j.clinre.2019.05.001