Long non‐coding RNA LOC100133669 promotes cell proliferation in oesophageal squamous cell carcinoma

Abstract Objectives LOC100133669 is a lncRNA whose function during tumorigenesis remains unclear now. Thus, we aimed to explore its clinical significance and function in oesophageal squamous cell carcinoma (ESCC). Materials and Methods ISH was used to detect LOC100133669 expression in ESCC tissues. The full‐length LOC100133669 was identified by using RACE assay. Subcellular distribution of LOC100133669 was examined by nuclear/cytoplasmic RNA fractionation and qPCR. The role of LOC100133669 in ESCC cell growth was determined by colony formation, MTT and flow cytometry experiments in vitro, as well as xenograft tumour experiment in vivo. RNA pull‐down assay was performed to find LOC100133669‐interacted protein, which was further examined by RIP, IP, Western blot and rescue experiments. Results LOC100133669 was upregulated in ESCC tissues compared with adjacent non‐tumour tissues. High LOC100133669 expression was associated with poor prognosis of patients with ESCC. We defined LOC100133669 to be 831 nt in length and mainly localized in the cytoplasm of ESCC cells. Knockdown of LOC100133669 inhibited ESCC cell proliferation and cell cycle progression, while overexpression of LOC100133669 showed the opposite effects. Furthermore, LOC100133669 could bind to Tim50 and upregulated its protein level through inhibiting ubiquitination. Overexpression of Tim50 in part abolished the LOC100133669 depletion–caused inhibitory effect on ESCC cell proliferation. Conclusions LOC100133669 plays an oncogenic role in ESCC and may serve as a promising diagnostic marker and therapeutic target for ESCC patients.


| INTRODUC TI ON
Oesophageal carcinoma is one of the most lethal gastrointestinal tumours in the world, with the overall five-year survival rate only of 15%-25%. 1 In 2018, 572 000 new cases and 509 000 deaths of oesophageal carcinoma were estimated worldwide. 2 The distribution of oesophageal carcinoma varies with geography. 3,4 China is one of the countries with the highest incidence and mortality of oesophageal carcinoma. 3 According to the histological classification, oesophageal carcinoma is mainly divided into oesophageal squamous cell carcinoma (ESCC) and oesophageal adenocarcinoma (EAC). ESCC counts for most oesophageal carcinoma in China, 5 which may be related to diet and living habits, alcohol consumption and cigarette smoking, environmental and genetic factors. 6,7 Currently, since there is a lack of effective detection markers and treatment targets for patients with ESCC, the exploration of promising biomarkers and molecular mechanisms of ESCC is an urgent goal.
Long non-coding RNAs (lncRNAs) are commonly defined as a class of transcripts that are over 200 nucleotides (nt) in length and cannot encode proteins. 8 According to the location relative to other transcripts, lncRNAs are generally categorized as sense, antisense, bidirectional, intronic and intergenic ones. 9 Recent studies have shown that lncRNAs participate in the regulation of gene expression at epigenetic, 10 transcriptional 11,12 and post-transcriptional 13,14 levels, performing as signals, scaffolds, guides and decoys. 15 Even in the same cellular environment, a lncRNA regulates gene expression in various ways. For example, in the nucleus, CCAT1 acts as a scaffold binding both PRC2 and SUV39H1 to regulate the histone methylation of SPRY4 promoter; in the cytoplasm, CCAT1 serves as a molecular decoy of miR-7 to affect HOXB13 expression. 16 Disorder of lncRNAs is closely related to various diseases, including metabolic diseases, 17,18 neurological diseases, 19,20 autoimmune diseases 21,22 and particularly cancers. 23,24 A great number of lncRNAs are reported to be dysregulated in human cancers and play critical roles during carcinogenesis. 23,24 HOTAIR, a lncRNA localized on the HOXC locus regulating the expression of HOXD genes, 10 is overexpressed and exhibits oncogenic activity in a variety of human cancers. [25][26][27] PCAT-1, a lncRNA originally identified to be overexpressed in prostate cancer, 28 is also upregulated in other human cancers and contributes to cancer progression. 29,30 Although more and more researchers have shifted their focus of cancer research to lncRNAs, the roles and mechanisms of lncRNAs in human cancers, including ESCC, are still unclear.
In this study, we focused our attention on a NCBI RefSeqannotated lncRNA, LOC100133669 (RefSeq accession number: NR_026913.1), which is located on human chromosome 8q24.3.
The function of LOC100133669 during tumorigenesis remains unclear now. Here, we aimed to study the clinical significance and function of LOC100133669 in ESCC and hoped it could be a new molecular target for the diagnosis and treatment of ESCC. The human ESCC cell lines COLO680N, YES2, KYSE30, KYSE70,   KYSE140, KYSE150, KYSE180, KYSE410, KYSE450 and KYSE510 were from Professor Yutaka Shimada of Kyoto University as gifts.

| Cell lines and cell culture
The human embryonic kidney cell line HEK293T was obtained from American Type Culture Collection (ATCC, USA). ESCC cells were cultured in RPMI 1640 medium (Gibco), and HEK293T cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco). All culture medium was supplemented with 10% foetal bovine serum (FBS). Cells were maintained in a humidified incubator (Thermo Scientific) at 37°C with 5% CO 2 . following the manufacturer's instructions. The lentiviruses in the medium were collected 72 hours later and then infected the KYSE510 cells in the presence of polybrene (Sigma). KYSE510 control and LOC100133669-stable knockdown cell lines were obtained by puromycin screening. The shRNA sequences are as follows: forward 5′-CCGGACTCAGGAGAGTGAAAGAAACCTCGAGGTTTCTT TCACTCTCCTGAGTTTTTTG-3′, reverse 5′-AATTCAAAAAACT

| Tissue microarrays and RNA in situ hybridization (ISH)
The ESCC tissue microarrays with clinicopathological features and survival times were purchased from Shanghai Outdo Biotech Co., Ltd (SOBC). RNA ISH was performed to detect the expression of LOC100133669. Briefly, after dewaxing and digesting with protease K, ESCC tissue microarrays were hybridized with a double (5′ and 3′) digoxin-labelled LNA probe specific for LOC100133669 (5′-AGTGAGGCTGGAAGGCTGGAT-3′; Exiqon) at 50°C for 1 hour and were subsequently incubated with alkaline phosphatase-conjugated digoxin antibody at 4°C overnight. The comprehensive score = the positive staining intensity × the percentage of positive cells.

| Rapid amplification of cDNA ends (RACE)
The 5′-and 3′-RACE experiments were performed using the SMARTer
Briefly, cell pellet was resuspended with nucleoplasmic fractionation buffer and incubated on ice for 5 minutes. After centrifugation at 5000 × g for another 5 minutes, the supernatant and pellet were collected as the cytoplasmic and nuclear fractions, respectively. RNA was extracted from nuclear/cytoplasmic fractions, and RT-qPCR was then used to evaluate the relative levels of LOC100133669, myc precursor RNA (pre-myc) and GAPDH in each sample.

KYSE510 control and LOC100133669-stable knockdown cells, and
KYSE150/KYSE510 cells transiently transfected with the control siRNA or siRNAs against LOC100133669 for 24 hours were trypsinized into a single-cell suspension and seeded. Ten days later, the colonies were fixed with methanol, stained with crystal violet solution and photographed. Colonies containing more than 50 cells were counted.

| MTT assay
KYSE450 control and LOC100133669-stable overexpression cells, KYSE510 control and LOC100133669-stable knockdown cells, and KYSE150/KYSE510 cells transiently transfected with the control siRNA or siRNAs against LOC100133669 for 24 hours were trypsinized into a single-cell suspension, seeded and cultured for 6 days.
10 μL of MTT (5 mg/mL; Sigma) was added into each well daily. After incubation for 4 hours at 37°C, supernatant was removed and dimethyl sulfoxide (DMSO; Sigma) was added into each well. The viability was evaluated at a wavelength of 492 nm using a microplate reader (Sunrise; TECAN).

| RNA pull-down assay
RNA pull-down assay was performed as described previously. 31 Briefly, template DNA for in vitro transcription of LOC100133669 was obtained by linearizing pcDNA3.1-669 vector with restriction enzyme EcoRI at the 3′ end. Template DNA for in vitro transcription of GAPDH was PCR-amplified using the primers containing T7 promoter sequence as follows: T7-GAPDH, forward, Biotin-labelled RNAs of LOC100133669 and GAPDH were transcribed in vitro using the MEGAscript™ T7 Transcription Kit (Invitrogen) with biotin-16-UTP (Invitrogen). Cell extracts were incubated with RNAs for 30 minutes, followed by adding streptavidin agarose beads (Invitrogen) for further incubation. After washing for 5-6 times, LOC100133669-associated proteins, which were retrieved from beads, were subjected to SDS-PAGE and silver staining. Differential protein bands were excised and identified by mass spectrometry.

| Western blot assay and antibodies
Total proteins extracted from cells were separated by SDS-PAGE and transferred to PVDF membranes. Then, the membranes were blocked with 5% non-fat milk and subsequently incubated with primary antibodies against Tim50 (Proteintech Group, China) or

| In vivo xenograft tumour experiment
In vivo xenograft tumour experiment was performed as described previously. 32

| Statistical analysis
All statistical analyses were performed using SPSS19.0 software or GraphPad Prism 5.0 software. Data are expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used for multigroup comparison, and Student's t test was used for comparison between two groups. Overall survival was analysed by the Kaplan-Meier method, and comparison was performed using log-rank test. Prognosis factors were analysed by Cox proportional hazard regression model. P < .05 was considered statistically significant. was mainly localized in the cytoplasm and was highly expressed in ESCC tissues compared with adjacent non-tumour tissues ( Figure 1A,B). We then divided the 181 ESCC tissues into high-LOC100133669 expression group (n = 49, ISH score ≥ 6) and low-LOC100133669 expression group (n = 132, ISH score < 6). High LOC100133669 expression was not associated with clinicopathological features of ESCC patients (Table S1). However, the patients with high LOC100133669 expression had shorter overall survival time than those with low LOC100133669 expression, as analysed by the Kaplan-Meier method (log-rank test, χ 2 = 6.296, P = .012; Figure 1C), indicating that LOC100133669 overexpression was as-

| Identification and characterization of LOC100133669
The UCSC Genome Browser (http://genome.ucsc.edu) shows that  Figure 2D). CPAT software 33 (http://lilab.research.bcm.edu/ cpat/index.php) also showed that LOC100133669 had limited protein-coding potential ( Figure 2E). In addition, in the 3′-RACE experiment, we found that LOC100133669 had a poly(A) tail. Consistently, the PCR products of LOC100133669, amplified from cDNA reversed with oligo dT primer or random 6 mers, respectively, were basically the same ( Figure 2F), indicating that LOC100133669 is a lncRNA with a poly(A) tail. The localization of lncRNAs in cells is closely  Figure 2G), which is consistent with the results from ISH.

| LOC100133669 promotes ESCC cell proliferation
To study whether LOC100133669 plays roles in ESCC tumorigenesis, we first detected the LOC100133669 expression in  Figure 3C). Consistent results were obtained from MTT assay ( Figure 3D). The tumorigenic role of LOC100133669 was further investigated using xenograft tumour experiment in vivo.

KYSE510 control cells and LOC100133669-stable knockdown cells
were injected subcutaneously into nude mice. As shown in Figure 3E, knockdown of LOC100133669 reduced the ability of KYSE510 cells to form tumours in nude mice. Collectively, these data indicate that LOC100133669 can promote the proliferation of ESCC cells.

| Effect of LOC100133669 on cell cycle progression of ESCC cells
As cell cycle progression plays an important role in determining cell growth, we then aimed to explore whether LOC100133669 contributes to cell cycle progression of ESCC cells. We found that the cell cycle distribution was not significantly changed in KYSE150 and KYSE510 cells upon depletion of LOC100133669, as determined by flow cytometric analysis ( Figure 4A). However, a decrease in G0/ However, myc expression was only remarkably increased after release from serum starvation ( Figure 5).

| LOC100133669 interacts with Tim50
Our next goal was to explore the underling mechanisms by which LOC100133669 plays an oncogenic role in ESCC. lncRNAs have been reported to interact with proteins to perform their functions. 10,12 Thus, we speculated that LOC100133669 may function via the same way. To search for LOC100133669-associated proteins, we performed RNA pull-down assay, followed by SDS-PAGE analysis and silver staining ( Figure 6A). Differential bands between LOC100133669 and negative control (GAPDH) were excised and subjected to mass spectrometry. One of the were served as cytoplasmic and nuclear references, respectively. Data are presented as mean ± SD, n = 3 tein into mitochondria [35][36][37] and has been reported to be involved in tumorigenesis. 38,39 The interaction between LOC100133669

F I G U R E 3
and Tim50 was validated by Western blot analysis ( Figure 6A). To further confirm the association of LOC100133669 with Tim50, RIP assay was conducted and the results demonstrated the specific interaction between LOC100133669 and Tim50 in both KYSE150 and KYSE510 cells ( Figure 6B).

| LOC100133669 regulates Tim50 protein level via ubiquitination
Next, we examined whether LOC100133669 affects the protein level of Tim50. Western blot showed that Tim50 protein level was downregulated in LOC100133669-knockdown cells and was upregulated in LOC100133669-overexpression cells, when compared with corresponding control cells ( Figure 6C). However, knockdown of LOC100133669 had no effect on the mRNA level of Tim50  Indeed, the level of Tim50 ubiquitination in LOC100133669knockdown cells was higher than that in control cells ( Figure 6G).
These data collectively indicate that LOC100133669 binds to Tim50 and upregulates its protein level through inhibiting ubiquitination.

| Tim50 participates in the effect of LOC100133669 on ESCC cell proliferation
To investigate whether Tim50 is involved in the LOC100133669-

| D ISCUSS I ON
In recent years, lncRNAs have become a hotspot in cancer research. A large number of cancer-associated lncRNAs are screened out using the subtractive hybridization method and especially high-throughput chip or sequencing technology. 28,42 For example, MALAT-1, a well-known lncRNA, was identified from differential expression genes between primary non-small-cell lung tumours that metastasized or did not subsequently. 43 MALAT-1 expression was also higher in ESCC tissues than that in adjacent normal tissues and was associated with poor prognosis. 44 ISH results showed that LOC100133669 was mainly localized in the cytoplasm, which is in accordance with the finding from nuclear and cytoplasmic RNA isolation, suggesting that LOC100133669 is likely to interact with molecules in the cytoplasm. Tim50, one of the LOC100133669-associated proteins determined by RNA pull-down assay, attracted our attention. Tim50 is a subunit of the TIM23 complex and is essential for directing translocation of preproteins into mitochondria. [35][36][37] Mitochondria are important energy metabolism organelles, which participate in energy production, metabolism, cell death, cell signalling and oxidative stress. 49 Increasing evidence has shown that mitochondrial dysfunction is closely related to tumorigenesis. 49 Lack of Tim50 has been reported to lead to TIM23 complex and mitochondrial dysfunction. 50 Furthermore, downregulation of Tim50 expression inhibited the growth and chemoresistance of human lung and breast cancer cells harbouring mutant p53. 38 More recently, Tim50 was reported to facilitate NSCLC cell proliferation and invasion via ERK signalling pathway. 39 In our study, knockdown of LOC100133669 downregulated the protein level of Tim50 through promoting its ubiquitination. Cytoplasmic lncRNAs have been reported to participate in regulating protein stability via ubiquitination. 40,41 For example, an oncogenic lncRNA GLCC1 was proven to stabilize transcriptional factor c-Myc from ubiquitination degradation through directly binding to HSP90 chaperon. 51 Moreover, we found that Tim50 overexpression partially abolished the reduction of ESCC cell proliferation caused by LOC100133669 knockdown.
Our study suggests that LOC100133669 might change the expression level of Tim50, thereby affecting the function of mitochondria and then regulating cell growth.
F I G U R E 7 Tim50 overexpression partially compromised the effect of LOC100133669 knockdown on ESCC cell proliferation. A, Western blot analysis of Tim50 protein level in KYSE510 control cells and LOC100133669-stable knockdown cells transiently transfected with pEGFP empty vector or pEGFP-Tim50 vector for 48 h. B, C, Colony formation (B) and MTT (C) assays were performed in cells as shown in A. Data are presented as mean ± SD, n = 3. *P < .05, **P < .01, ***P < .001 In summary, LOC100133669 is highly expressed in ESCC tissues and has an association with patient prognosis. LOC100133669 promotes ESCC cell proliferation and cell cycle progression.
LOC100133669 binds to Tim50 and upregulates its protein level through inhibiting ubiquitination. Tim50 overexpression partially rescues slowed ESCC cell growth due to LOC100133669 knockdown. Our findings identify LOC100133669 as a novel oncogene in ESCC, which may have the potential to be a target for ESCC diagnosis and therapy.

CO N FLI C T O F I NTE R E S T
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.