A Novel lncRNA, CTD-2017C7.1, Promotes ESCC Tumorigenesis by Binding to PDIA3 and Activating Oncogenic Gene Expressions

Background lncRNAs are dysregulated in many human cancers, including esophageal squamous cell carcinoma (ESCC), and are associated cancer development and progression. In the current study, we aimed to elucidate the biological roles of lncRNA CTD-2017C7.1 in ESCC. Method The biological functions of CTD-2017C7.1 were determined in vitro and in vivo. RNA pull-down, MS, RIP, RNA-seq, and qRT-PCR assays were employed to investigate the mechanisms of CTD-2017C7.1. Results CTD-2017C7.1 was up-regulated in ESCC tissues and cells, and associated with poor clinical outcome. Overexpression of CTD-2017C7.1 promoted cell proliferation, invasion and migration. CRISPR/Cas9 knockout of CTD-2017C7.1 resulted in reverse effects. Up-regulation of CTD-2017C7.1 also increased ESCC tumor growth in vivo. Mechanistically, CTD-2017C7.1 bound to PDIA3 and activated the expressions of oncogenic genes.

Higher expression of the lncRNA ESCCAL-1 was correlated with worse clinicopathologic features in patients with ESCC [16]. Up-regulated of LOC440173 expression was related to histological grade, tumor invasion depth, lymph node metastasis, and TNM stage [17]. High expression of MALAT1 correlates with shorter disease-free survival and overall survival in ESCC patients who underwent radical surgical resection [2]. Interestingly, lnc-MCEI, an oncogenic lncRNA in ESCC, was found to be able to modify the chemosensitivity of ESCC to cisplatin (DDP) both in vivo and in vitro [18]. These observations indicate that lncRNAs play critical roles in ESCC. However, there are still plenty of ESCC-associated lncRNAs have not been identi ed and functionally characterized.
In our previous study, we have identi ed a novel lncRNA CTD-2017C7.1 that was abundant expressed in exosomes derived from patients with ESCC [19]. This nding suggested that CTD-2017C7.1 might be involved in the pathogenesis of ESCC. However, the biological roles and the molecular mechanisms of CTD-2017C7.1 in ESCC are still entirely unknown. In this study, we con rmed that CTD-2017C7.1 was upregulated in both ESCC cell lines and tumor tissues. We then conducted loss-and gain-of-function assays to determine the in vitro and in vivo impacts of CTD-2017C7.1 on ESCC tumorigenesis. Moreover, investigations were carried out to assess through which mechanisms that CTD-2017C7.1 might play its regulatory roles in ESCC cells. This study provides new insights into the biological functions of CTD-2017C7.1 as well as its regulatory mechanisms in ESCC.

Patient samples
Nine paired ESCC tumor tissues and corresponding adjacent normal tissues were collected from patients undergoing surgical resection at the Department of Thoracic Surgery, Shenzhen People's Hospital, China.
All patients were histologically con rmed by two pathologists. After surgical resection, all tissues were immediately frozen in liquid nitrogen and then stored at -80 o C until use. None of these patients were treated with radiotherapy or chemotherapy prior to surgery.

Subcellular fraction location
Nuclear/Cytosol Fractionation Kit (Phygene, PH1466) was used to separate cytoplasmic and nuclear fraction of TE-1 cells in accordance with supplier's suggestions. Trizol reagent (Invitrogen, CA, USA) was applied to isolate total RNA, which was reverse transcribed in a nal volume of 20 µl using random primers under conditions for RT reagent kit (TransGen Biotech, Beijing). QPCR was used to detect CTD-2017C7.1 levels in cytoplasm and nuclear fraction, using the SYBRGreen PCR kit (TransGen Biotech, Beijing). The expression levels of CTD-2017C7.1 in nuclear and cytoplasmic fraction were normalized to GAPDH, respectively.
Cell culture and transfection ESCC cell lines (EC109, KYSE180, KYSE70, and TE-1) were obtained from the cell bank of Shanghai Biological Institute (Shanghai, China). Normal esophageal epithelium cells (NE3) were obtained from Pro.
Fu L [20]. ESCC cells were cultured in RPMI 1640 (Gibco) with 10% FBS and NE3 cells were incubated in a 1:1 mixture of EpiLife and dKSFM (Gibco). All of these cells were maintained at 37°C in a humidi ed atmosphere containing 5% CO 2 . Cell transfections were performed using Lipofectamine 2000 kit (Invitrogen, CA, USA) according to the manufacturer's instructions.
The authentication of cell lines TE-1, EC109, KYSE180, and KYSE70 cells were obtained in January, 2018. All these cell lines were tested and authenticated by STR. Genomic DNAs of TE-1, EC109, KYSE180, and KYSE70 cells were extracted and ampli ed respectively using GenePrint System (Promega). Ampli ed products were processed using the ABI3730xl Genetic Analyzer. Data were analyzed using GeneMapper4.0 software and then compared with the ATCC, DSMZ or JCRB databases for reference matching. The STR results showed that there were no four alleles on the main nine locus, it is certainly no cross contamination of human cells in these cell lines. TE-1, EC109, KYSE180, and KYSE70 cells were tested on Jan. 19, 2018 (Supplementary 1-4).

RNA extraction, quantitative realtime PCR (qRT-PCR)
Total RNA was isolated from tissues or cultured cells using Trizol reagent (Invitrogen, CA, USA). Total RNA (1µg) was reverse transcribed to cDNA using random primers with the SuperScript™ III RT Reagent kit (Invitrogen, CA, USA) following the manufacturer's protocol. QRT-PCR was performed in the CFX96TM Real-Time System (Bio Rad, Hercules, CA, USA). GAPDH was employed as an endogenous control.
Results were normalized to the expression of GAPDH. The relative expression of RNAs was calculated using the 2 −ΔΔCt method. All experiments were performed in triplicate. Sequences of speci c primers are listed in Supplementary Table 1. Subsequently, 10µl CCK-8 reagent was added into each well and incubated with cells for 2h. Finally, cell viability was measured every 24h (0, 24, 48, 72h) using the microplate reader MK3 (Thermo Fisher Scienti c, Shanghai, China) at the absorbance of 450 nm. Five replicates were performed for each treatment group.

Cell invasion and migration assays
For the invasion assay, 5×10 4 cells in serum-free RPMI-1640 medium were placed into the upper chamber (Corning Falcon, USA), and 500 µL of RPMI-1640 with 10 % FBS was added to the lower chamber. After 24 h of incubation, the cells on the upper chamber were removed with cotton swabs, and the cells on the lower membrane surface of the membrane that had migrated or invaded from the upper chamber were xed with 4 % paraformaldehyde for 15 min and then stained with 0.1 % crystal violet for 10 min. The number of invaded cells was counted under a microscope. All elds were counted randomly in each well.
Cell migration was evaluated by wound healing assay. In brief, cells in 6-well plates were allowed to reach 80-90 % con uence, and wounds were scratched in the cell monolayers using a sterile micropipette tip. Wound closure was recorded using a microscope. The distance across the wound was measured at baseline and at 24h, and the wound closure was calculated using the following formula: distance of wound closure = 1 − length of wound (24h)/length of wound (0h).
In vivo animal study BALB/cA nude mice (female, 4-6 weeks old) were purchased from the Experimental Animal Center of Guangdong, China; and were maintained under pathogen-free conditions. The mice were randomly divided into two treatment groups (n = 5 for each group): one group was injected with cells (1×10 6 ) of NC control cells; another group was injected with cells (1×10 6 ) of stably expressing CTD-2017C7.1 (in 0.1ml PBS). Cells were injected subcutaneously into the right anks of mice. Tumor size was measured every 3 days using a Vernier caliper. Tumor volumes were calculated using the following formula: 0.5×L×W2 (L refers to long and W refers to short diameters of the tumor). The tumor weight was measured at the endpoint of the study. At 21 days post-injection, mice were euthanized, and tumor were resected and preserved in formaldehyde for HE or IHC staining. The animal study was approved by the Animal Ethical Committee of Shenzhen University Health Science Center. All experimental procedures were manipulated following the approved protocols.
Immunohistochemical (IHC) analysis IHC assay was conducted to quantify the expression levels of Ki67 protein in the tumor tissues, using Ki67 antibody (Abcam, ab16667) as the primary antibody and HRP-conjugated antibody as the secondary antibody. The detailed procedures for IHC assay have been described previously [21].

RNA pulldown and MS assays
The RNA pull-down assays were performed using the PierceTM Magnetic RNA-Protein Pull-Down Kit (Thermo Fisher Scienti c). Brie y, full-length sense and antisense CTD-2017C7.1 sequences were transcribed into TE-1 cells, using the RNA 3' End Desthiobiotinylation Kit (Thermo Fisher Scienti c) and RNA Production System-T7 (Promega, Madison, WI, USA), respectively. Cells were isolated and lysed using a protein lysis buffer. The biotin-labeled probe was captured using the streptavidin magnetic beads. Then the lysates of TE-1 cells were incubated with biotinylated nucleic acid compounds. The retrieved proteins were puri ed and detected by silver staining and the eluted RNA-binding proteins complex were subjected to mass spectrometry analysis. For LC-MS, proteins were digested by trypsin and separated by liquid chromatography, then detected by Q-Exactive MS (Thermo Fisher Scienti c, Hudson, NH).

RNA immunoprecipitation (RIP)
RNA immunoprecipitation (RIP) analysis was performed using the Imprint® RNA Immunoprecipitation kit (Sigma, RIP-12Rxn) in accordance with the user manual. In brief, TE-1 cells (1×10 7 ) were treated with RIP lysis buffer to obtain the cell extract. The cell lysis was centrifuged at 13,000g at 4℃ for 10 min. Then the collected supernatant was incubated with RIP buffer containing magnetic beads conjugated to speci c antibody or normal anti-rabbit IgG at 4℃ overnight. IgG was considered as the negative control.
Subsequently, beads were incubated with proteinase K Buffer and RIP wash buffer to remove proteins. RNA was puri ed by phenol and chloroform. The puri ed RNAs were furthered subjected to qRT-PCR analysis to assess the expression of precipitated CTD-2017C7.1.

RNA Sequencing (RNA-seq)
To investigate the target genes regulated by CTD-2017c7.1, total RNA was extracted from TE-1 cells stably expressed CTD-2017C7.1 or NC control cells, using Trizol reagent (Invitrogen, CA, USA). RNA quality was assessed using the NanoPhotometer® spectrophotometer (IMPLEN, CA, USA). The NEBNext® UltraTM RNA Library Prep Kit for Illumina® (NEB, USA) was used for generating the sequencing libraries. In brief, magnetic beads attached with poly-T oligo were used to purifying mRNA from total RNA. First strand cDNA synthesis was performed using M-MuLV Reverse Transcriptase (RNase H-), then second strand cDNA synthesis was synthesized using DNA polymerase I and RNase H. The library fragments were puri ed with AMPure XP system (Beckman Coulter, Beverly, USA). After PCR ampli cation, the PCR products were puri ed (AMPure XP system) and library quality was assessed on the Agilent Bioanalyzer 2100 system. TruSeq PE Cluster Kit v3-cBot-HS (Illumina) was used to perform the index-coded samples clustering on a cBot Cluster Generation System based on the manufacturer's instructions. Then the index-coded samples were sequenced on an Illumina Novaseq platform and 150 bp paired-end reads were generated. Differential expression and normalized reads counts (FPKM, fragments per kilobase of gene/transcript model per million mapped fragments) were calculated using the DESeq2 R package (1.16.1).

Ethical conduct of research
The authors state that they have obtained appropriate institutional review board approval or have followed the principles outlined in the Declaration of Helsinki for all human or animal experimental investigations. In addition, for investigations involving human subjects, informed consent has been obtained from the participants involved.

Statistical analysis
Experiments in vitro and in vivo were performed at least 3 times. Results are expressed as the mean ± standard deviation. Differences between two groups were assessed using Student t-test (two-tailed), oneway-ANOVA, or the Mann-Whitney U test, where appropriate. Kaplan-Meier survival analysis was used to compare esophageal cancer patient survival based on dichotomized CTD2017C7.1 expression levels retrieved from the TCGA database. Cox regression was applied to estimate prognosis, adjusting for multiple covariates, including age, gender, and stages. All statistical analyses were performed using the SAS 9.3 program (SAS Corp., NC, USA) and GraphPad Prism 7.0 (GraphPad software, Inc., USA). Statistically signi cant differences are presented as *P < 0.05 and **P < 0.01. In all cases, P < 0.05 was considered statistically signi cant.

CTD-2017C7.1 is up-regulated in human ESCC tissues and cell lines and correlates with poor prognosis
To investigate the importance of CTD-2017C7.1 in ESCC, we rstly analyzed the expression levels of CTD-2017C7.1 in human ESCC tissues compared with normal tissues. The results showed that CTD-2017C7.1 expression was signi cantly higher in ESCC tissues than that in adjacent normal tissues (Fig. 1A). Furthermore, CTD-2017C7.1 expression levels were determined in ESCC cell lines and NE3 cells. The results revealed that CTD-2017C7.1 expression was increased in ESCC cell lines compared with normal cells (Fig. 1B). To further validate these results, we extracted data from GEPIA (http://gepia.cancerpku.cn/,a TCGA database-based analysis website), and analyzed the CTD-2017C7.1 expression levels in esophageal cancer (including both ESCC and EA subtypes). The results showed that CTD-2017C7.1 expression level in esophageal cancer tissues was signi cantly higher than that in normal tissues (Fig. 1C). We divided TCGA samples into high (above median) and low (below median) CTD-2017C7.1 expression groups based on the median value of CTD-2017C7.1 expression levels. Kaplan-Meier survival analysis was used to examine the association between CTD-2017C7.1 expression and esophageal cancer patient prognosis. Notably, patients with higher CTD-2017C7.1 expression levels had signi cantly shorter overall survival time than those expressing lower levels of CTD-2017C7.1 (log-rank P = 0.039) (Fig. 1D). In addition, multivariate Cox regression analysis veri ed that higher expression of CTD-2017C7.1 was correlated with lower survival probability (HR = 1.70, P = 0.04; Fig. 1D).

enhances ESCC cell growth in vivo
To explore whether CTD-2017C7.1 would also affect tumorigenesis in vivo, TE-1 cells were transfected with CTD2017C7.1-overexpression cells or control cells, and were injected into nude mice, respectively. After 21 days of injection, the tumors formed in the CTD-2017C7.1 over-expression group were signi cantly larger than those in the control group (Fig. 5A, 5F, 5G). Additionally, the mean tumor weight in the CTD-2017C7.1-overexpression group was signi cantly higher than that of control mice (Fig. 5B).
Nevertheless, no signi cant difference of body weight was observed between CTD-2017C7.1 overexpression group and the empty vector group (Fig. 5C). In addition, IHC assay showed that tumors formed from CTD-2017C7.1-overexpression cells exhibited increased positive Ki-67 cells than that from control cells, and HE staining showed similar changes (Fig. 5D, 5E). Collectively, these results indicated that over expression of CTD-2017C7.1 signi cantly promoted tumor growth in vivo.

CTD-2017C7.1 directly binds to PDIA3 in ESCC cells
It has been reported that lncRNAs are involved in regulation of cancer cells phenotypes through activation of oncogenes or inactivation of tumor suppressor by binding to speci c RNA-binding proteins [22]. To examine the interactions of CTD-2017C7.1 with potential RNA-binding proteins that regulate malignant phenotypes in ESCC cells, we performed RNA pull-down assay, followed by mass spectrometry analysis, in TE-1 cells. Silver staining of RNA pull down showed that CTD-2017C7.1 indeed bound with many proteins in TE-1 cells (Fig. 4A, Supplementary Table 2). Pathway analysis revealed that proteins identi ed from RNA pull down assay were mainly enriched in cancer-associated pathways such as biosynthesis of amino acids (hsa01230), Glycolysis/Gluconeogenesis (hsa00010), PI3K-Akt signaling pathway (hsa04151), Carbon metabolism (hsa01200), NOD-like receptor signaling pathway (hsa04621), Estrogen signaling pathway (hsa04915) (Fig. 4B). Among the binding proteins identi ed, all of the top 10 binding proteins (Fig. 4D), including PDIA3, TPI1, ANXA2P2, ANXA1, MAPK1, HSP90AB1, HSP90AA1, PKM, PDIA4, and YWHAE, have been previously found to be associated with cancer development and progression, supporting the validity of our ndings. We chose PDIA3 for further validation because PDIA3 was ranked as the top 1 binding protein in MS analysis with the highest score and its expression has been associated with ESCC pathological stage and prognosis [23,24]. RIP experiment con rmed the existence of interaction between CTD-2017C7.1 and PDIA3 in TE-1 cells (Fig. 4C). These data demonstrated that CTD-2017C7.1 regulated ESCC malignancy through interacting with its binding proteins.  Table 3 -4). Principal component analysis showed that gene expression pro les in CTD-2017C7.1 over-expression cells were distinct from that in control cells (Fig. 6C). Gene Ontology (GO) analysis revealed that up-regulation genes were more likely to be involved in cell cycle-related biological processes ( Supplementary Fig. 5A), whereas down-regulated genes were obviously implicated in Wnt signaling processes (Supplementary Fig. 5B). KEGG pathway enrichment analysis suggested that up-regulated genes were enriched in tumor-associated pathways including cell cycle, RNA transport, DNA replication, and splicesome (Fig. 6D). While down-regulated genes were more concentrated in pathways such as Wnt, Hippo, Notch, and lysosome signaling pathways ( Supplementary  Fig. 5C). Among the top genes up-regulated by CTD-2017C7.1 (Fig. 6E), LAMA4, ABCG2, CFH, LUCAT1, and DIO2 have been reported to play important roles in the pathogenesis of cancers. Gene set enrichment analysis (GSEA) demonstrated that differentially expressed genes in CTD-2017C7.1 over-expression cells were signi cantly enriched in cancer-related gene sets RNA degradation, drug metabolism, HIF-1, PI3K-AKT, Platinum drug resistance, cell cycle, Nucleotide excision repair (Fig. 6F-I, Supplementary Table 5-6).

Discussion
Many lncRNAs have been reported to be associated with cancer development and progression, but only a few lncRNAs have been functionally characterized. Identi cation of ESCC-associated lncRNAs, investigation of their clinical signi cance and biological functions may facilitate the development of lncRNA-directed diagnostic strategy or therapy against this malignancy. In the present study, we identi ed a novel lncRNA CTD-2017C7.1 and demonstrated that it was signi cantly up-regulated in ESCC cell lines as well as in human ESCC tumor tissues. Increased CTD2017C7.1 expression was associated with poor prognosis of esophageal cancer patients. Moreover, higher CTD-2017C7.1 expression promoted ESCC cell proliferation and migration, while silencing of CTD-2017C7.1 suppressed the malignancy phenotypes of ESCC cells. The in vivo studies con rmed that overexpression of CTD-2017C7.1 enhanced tumor growth in ESCC cells. Furthermore, we showed that CTD-2017C7.1 exerted its oncogenic effects by binding to PDIA3 and activating expression of oncogenic genes. Collectively, these ndings indicate that CTD-2017C7.1 might play important roles in ESCC carcinogenesis.
CTD-2017C7.1 is located on chromosome 14 (chr14:102299772-102305868). As a novel lncRNA recently identi ed by RNA-sequencing in plasma exosomes from patients with ESCC [19], the biological functions of CTD-2017C7.1 are unclear. Since prior studies have suggested that some low-expressed cellular lncRNAs were selectively enriched in exosomes [25,26], it raised the question whether measurement of exosomal lncRNA levels can re ect the real abundance or biological functions of exosomal lncRNAs in their parental cells. In the present study, we rst proved that the expression levels of CTD-2017C7.1 were up-regulated in both tumor tissues and in ESCC cells. These observations were in consistent with that of results in exosomes [19]. We then demonstrated that CTD-2017C7.1 promoted ESCC malignancy in vitro and in vivo. Our ndings supported the idea that exosomal lncRNAs such as CTD-2017C7.1 were functional representatives of cellular lncRNAs in ESCC cells or in ESCC tumor tissues.
To explore the mechanism of action of CTD-2017C7.1, we conducted RNA pull-down assays and MS analysis, which led to identi cation of several cancer-associated binding proteins. Among them, PDIA3 was ranked as the top 1 binding protein with the highest score and its bind to CTD-2017C7.1 was con rmed by RIP analysis. PDIA3, also known as ERp57, is a multifunctional member of the PDI family, exhibiting redox and protein disul de isomerase activity [27]. PDIA3 is a molecular chaperone that was found to be associated with transcript factors responsible for expression regulation of genes involved in proliferation, apoptosis and oncogenesis [28]. PDIA3 was found in many different subcellular locations [29,30]. Containing a nuclear location sequence KKKK [31], PDIA3 can translocate to nuclear from cytoplasm when the cell was stimulated by TNF-α or NF-κB [32,33]. Dysregulation of PDIA3 was involved in a variety of disorders including cancer [34,35]. It was found that up-regulation of PDIA3 promoted cancer cell proliferation[36], while downregulation of PDIA3 suppressed cell growth [37]. Some studies revealed that PDIA3 activated AP-1 through participation in nuclear protein complexes in melanoma, and lymphoma cell lines [38]. PDIA3 was also able to interact with NF-κB [33], which promoted cell cycle progression by regulating the expression of genes involved in the cell cycle machinery such as CDK [39] and cyclins [40]. Consistent with previous reports that lncRNAs can regulate gene expression by interacting with RNA binding proteins [41,42], the present study found that CTD-2017C7.1 speci cally bound to PDIA3 in ESCC cells, suggesting that CTD-2017C7.1 might be involved in the pathogenesis of ESCC, at least in part, by interaction with PDIA3. Further studies are required to elucidate the detailed underlying mechanisms of CTD-2017C7.1-PDIA3 interaction in ESCC cells.
Increasing evidence has revealed that lncRNAs contribute to tumorigenesis by silencing tumor suppressor genes or activating oncogenes [15,21]. Since CTD-2017C7.1 mainly localized in the nucleus, RNA sequencing was performed to identify target genes transcriptionally regulated by CTD-2017C7.1. Transcriptome pro ling showed that CTD-2017C7.1 activated differential expression of genes involving in cancer-associated pathways, such as Wnt, Hippo, cell cycle, and DNA replication pathways. Among the top CTD-2017 C7.1-regulated genes, several of them, such as LAMA4, ABCG2, CFH, LUCAT1, and DIO2, have been reported to be involved in cancer development and progression. LAMA4 is a gene functionally associated with PI3K-AKT signaling pathway. Up-regulation of LAMA4 has been correlated with high metastasis potential and poor survival outcome of pancreatic cancer [43]. Abnormally high expression of ABCG2 in esophageal cancer tissues is involved in multidrug resistance of esophageal cancer [44] and gastric cancer [45]. CFH participated in the maintenance of stemness in liver tumor-initiating cells [46]. DIO2 activation was also associated with poor clinical outcomes and treatment resistance [47]. Higher expression level of LUCAT1 can predict a shorter survival time of patients in different cancers [48]. Importantly, LUCAT1 could regulate the stability of DNMT1 and inhibit the expression of tumor suppressors through DNA methylation, leading to the formation and metastasis of ESCC, indicating that LUCAT1 may serve as a potential therapeutic target and as a biomarker for ESCC [49]. Taken together, these ndings suggested that CTD-2017C7.1 might act in concert with LAMA4, ABCG2, CFH, LUCAT1, DIO2, and other oncogenic genes to promote tumorigenesis of ESCC cells. But future studies are needed to clarify CTD-2017C7.1 regulation with other potential target genes and the detailed molecular mechanisms underlying regulatory behaviors in ESCC cells.

Conclusion
The present study provided evidence that the ESCC-associated novel exosomal lncRNA CTD-2017C7.1 is an oncogenic lncRNA that promotes tumorigenesis of ESCC through binding PDIA3 and activating oncogenic gene expressions. These ndings improve our understanding of ESCC pathogenesis, and facilitate the development of lncRNA-directed diagnostics and therapeutics against ESCC.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. (3. Availability of data and materials)

Competing interests
Authors declare no con icts of interest for this article.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.