LincHOXA10 Facilitates Colorectal Cancer Development By Regulating HOXA10-Mediated Epithelial-Mesenchymal Transtion

Background: Long non-coding RNAs (lncRNAs) have been reported to play an important role in tumorigenesis and metastasis of human colorectal cancer (CRC). However, the specic role of LincHOXA10 in CRC remains unknown. Methods: The expression of LincHOXA10 and HOXA10 in CRC cells and tissue samples was measured by quantitative reverse transcription PCR (qRT-PCR). The protein expression of HOXA10, E-cadherin, N-cadherin, Vinmentin, p-smad2 and p-smad3 was assessed by Western blotting or immunouorescence staining. Cell proliferation, migration, and invasion were assessed by the MTT and transwell assays. Tumor growth in vivo was carried out by subcutaneous tumor formation in nude mice. Results: In the present study, we found that LincHOXA10 expression was signicantly higher in human CRC tissues than the paired normal tissues. In fact, LincHOXA10 level correlated with the CRC tumor sizes and lymphatic metastasis. In cultured CRC cells, knockdown of LincHOXA10 inhibited cell proliferation, migration and invasion. LincHOXA10 deciency also attenuated CRC tumor growth in vivo. Mechanistically, LincHOXA10 interacted with HOXA10 and regulated its expression. HOXA10 levels were interrelated to the LincHOXA10 level in CRC cells. Functionally, HOXA10 was essential for TGF-β1/SMADs-induced epithelial -mesenchymal transition of CRC cells, and HOXA10 played a critical role in mediating the function of LincHOXA10. Importantly, HOXA10 expression was signicantly up-regulated in human CRC tissues. Conclusions: LincHOXA10 facilitates CRC development and metastasis via regulating HOXA10-mediated epithelial-mesenchymal transition of CRC cells.


Introduction
Colorectal cancer (CRC) is the third most common cancer worldwide and the second leading cause of cancer-associated mortality [1][2][3]. Despite tremendous efforts for improvement in the clinical treatment of CRC, the overall survival of patients with CRC has not improved dramatically owing to cancer recurrence and metastasis. Therefore, it is an acute need to elucidate the underlying mechanism of progression in CRC, which is desirable for the development of new treatment strategies and improvement of patient prognosis.
Homeobox (HOX) genes are classi ed into four subgroups, namely HOX A-D. HOXA10, a member of HOX gene family, encodes a DNA-binding transcription factor that plays critical roles in gene expression, cell differentiation, and morphogenesis [4][5][6]. HOXA10 also takes an important part in embryo implantation [7], endometriosis [8]and hematopoietic lineage commitment [9]. Importantly, HOXA10 is involved in a wide spectrum of biological functions in human cancers, including hepatocellular carcinoma [10,11], lung cancer [12], gastric cancer [13], bladder cancer [14], as well as glioma [15]. However, the expression and role of HOXA10 in CRC has not been previously reported.
Long noncoding RNAs (lncRNAs) are de ned as RNA transcripts longer than 200 nucleotides without or with limited protein-coding potential [16][17][18]. They can regulate gene activity through regulation of DNA and protein modi cation, protein supplement, and RNA interaction [19][20][21]. More lncRNAs are identi ed and play essential roles in biological processes, including cell growth, cell differentiation, and cell cycle control [22,23]. Therefore, lncRNAs may contribute to the development and progression of various human cancers, including CRC. A novel lncRNA, LincHOXA10, is transcribed from the antisense strand of HOXA10 gene locus at chromosome 7p15. 3. The knockdown of LincHOXA10 expression was shown to induce glioma cell apoptosis [24]. Shao et al found that LincHOXA10 drives liver tumor initiating cells (TICs) self-renewal and tumorigenesis [11]. However, the function and underlying mechanism of LincHOXA10 in the development of CRC remain to be elucidated. Here we found LincHOXA10 is upregulated in CRC, and is involved in the proliferation, migration and EMT of CRC cells. Moreover, LincHOXA10 initiates HOXA10 transcription, which mediates TGF-β2/SMADs induced EMT in CRC cells. Our study has identi ed a novel mechanism by which LincHOXA10 regulates CRC tumorigenesis and metastasis.

Materials And Methods
Clinical tissue samples A total of 30 CRC tissues and 30 adjacent normal tissue samples were obtained from 30 patients aged . No patients had received chemotherapy or radiotherapy before surgery at the Third A liated Hospital of Xinxiang Medical University Hospital. This study was approved by Ethic Committee of Third A liated Hospital of Xinxiang Medical University, and the tissues were obtained after the consent of patients. All the tissue samples were immediately frozen in liquid nitrogen and stored at − 80 °C until RNA extraction.
Cell Culture And Transfection CRC cell lines SW480, HCT116, LS174T, SW620, RKO and HT29 were obtained from the American Type Culture Collection (ATCC). Normal human fetal colonic mucosa cell line (FHC) was established at our laboratory. All cell lines were cultured in RMPI-1640 medium containing 10% fetal bovine serum (Gibco, USA) and 1% penicillin/streptomycin (Invitrogen, USA) in 5% CO 2 at 37 °C.
Rna Isolation And Real-time Quantitative Pcr (qrt-pcr) Total RNA from tissues or cultured cells was solated using Trizol reagent (Takara, Japan) according to the manufacturer's protocols. RNA (1 µg) was reverse transcribed to cDNA with reverse transcription kit (Takara, Japan). qRT-PCR was performed on ABI 7500 RT-PCR system (Applied Biosystems, USA) with SYBR® Premix Ex Taq™ Kit (TaKaRa, Japan) and Mir-X™ miRNA qRT-PCR SYBR® Kit (Clontech Laboratories, USA) in accordance with the manufacturer's instructions. All experiments were repeated three times. Data were calculated using comparative cycle threshold (CT) (2 −ΔΔCT ) method. The results were normalized to the expression of GAPDH. The primers for LincHOXA10 (F: CCC AGT AAG CCA AAG TCA AGCC, R: CTG AGG TCA ATG GTG CAA AGG ).

Western Blotting
Proteins were lysed in RIPA buffer (KeyGen Biotech, China) containing 100 mmol/L phenyl methane sulfonyl uoride (PMSF), and quanti ed by bicinchoninic acid (BCA) protein quantitative assay (KeyGen Biotech, China). Protein lysates were separated using 10% SDS-PAGE and transferred onto PVDF membranes (Roche, Switzerland). Then, the membranes were incubated with speci c antibodies against HOXA10 (Abcam, England), E-cadherin, N-cadherin, Vimentin, p-smad2 and p-smad3 (Cell Signaling Technology, USA) followed by incubation with the appropriate second antibodies, and α-Tubulin was used as an internal control. Finally, the membranes were detected using an enhanced chemiluminescence (ECL) detection system (FDbio, China), according to the manufacturer's instructions. The results were normalized to the expression of a-tubulin (Proteintech, USA).

Lentiviral Vector Preparation And Plasmid Transfection
The siRNAs targeting LincHOXA10 (si-LincHOXA10) and siRNA negative control (si-NC) were all purchased from GenePharma (China). The full length of HOXA10 was subcloned into pcDNA3.1 to overexpress HOXA10 levels with empty pcDNA3.1 serving as control. The pcDNA3.1 vector was bought from GenePharma (Shanghai). Lipofectamine™ 2000 (Invitrogen) was used for cell transfection according to the manufacturer's instructions.

Immunohistochemistry (ihc)
According to the speci cations of the S-P kit, para n-embedded tissues were cut into 5 µm-thick sections, dehydrated with organic solvent, retrieved with citrate buffer, incubated with primary antibody ( Anti-HOXA10 antibody: Abcam, England.) and then detected with an avidin-biotin complex with 3, 3′diaminobenzidine. The degree of staining was observed and scored independently by two pathologists.

Immuno uorescence Staining
For immuno uorescence staining of cultured cells, cells seeded on confocal dish were transfected with adenoviral vectors. 48 h later, cells were xed with 4% paraformaldehyde for 30 min and permeabilized with 0.5% Triton X-100 for 10 min at room temperature. The cells were then incubated with primary antibodies at 4 °C overnight followed by washes with PBS and incubation with uorescent secondary antibody in dark at room temperature for 1 h. After nal washes with PBS, the confocal dish was mounted using an anti-fade mounting solution containing 4, 6-diamidino-2-phenylindole (DAPI). The staining was examined, and images were captured using an Olympus Confocal laser scanning microscopy FV1200.

Transwell Migration And Invasion Assay
Transfected cells (1 × 10 5 ) were seeded in serum-free medium in the top chamber of each transwell well (BD Biosciences, USA), which featured a pore size of 8 µm. Matrigel (BD Biosciences) was used to cover the top side of the membrane for invasion assay and Matrigelfree condition was used for migration assay. The matched lower chamber was lled with complete medium supplemented with 10% FBS. The cells were incubated at 37 °C with 5% CO 2 for 48 hours, the non-traversed cells were removed from the upper lter with a cotton swab, and the traversed cells were xed with formaldehyde and stained with hematoxylin for 30 minutes. Then, the cells that migrated or invaded to the basal portion of the membrane in the lower compartment of the chamber were counted in 5 random visual elds using a microscope (× 200). All experiments were performed in triplicates.
In vivo tumor growth assay CRC cells were harvested by trypsinization, washed twice, and then re-suspended with serum-free medium. To evaluate CRC tumor growth in vivo, 5 × 10 6 of RKO and HCT116 cells stably expressing control or LincHOXA10 shRNA via lentiviral vector were separately injected subcutaneously into the left and right back ank of nude mice (n = 3 per group). Twenty ve days later, tumors were removed and measured.

Statistical analysis
All statistical analyses were performed using SPSS 19.0 (Abbott Laboratories, USA). The quantitative results of all experiments were presented as the mean ± SD. Differences among/between sample groups were analysed by one-way ANOVA or the independent-samples t-test. Relationships between LincHOXA10 expression and clinicopathological characteristics were tested using Pearson's χ2-test. Differences were considered signi cant if P < 0.05*; P < 0.01**; P < 0.001***.

Results
LincHOXA10 expression is upregulated in CRC tissues and correlates with clinicopathologic characteristics of patients with CRC The expression of LincHOXA10 in 30 CRC tissues and paired adjacent non-CRC tissues was detected with qRT-PCR analysis (Fig. 1A). The expression of LincHOXA10 was signi cantly upregulated in CRC tissues as compared with the adjacent non-CRC tissues (Fig. 1B).
The relationship between various clinicopathological characteristics of patients with CRC and LincHOXA10 expression was analyzed (Table 1). Age and gender did not show signi cant correlation with LincHOXA10 expression, while tumor size and lymphatic metastasis showed signi cant correlation with LincHOXA10 expression. Knockdown of LincHOXA10 inhibits cell proliferation, migration, and invasion in CRC cells In cellular level, we detected the expression of LincHOXA10 in six different CRC cells and found that RKO and HCT116 cells exhibited a much higher level of LincHOXA10 expression than other cells (Fig. 6A).
Therefore, we chose RKO and HCT116 cells for the subsequent studies.
To investigate whether LincHOXA10 expression has functional impacts on CRC cells, the expression of LincHOXA10 in two CRC cell lines (RKO and HCT116) was silenced by transfection with LincHOXA10 siRNA (si-LincHOXA10). qRT-PCR results showed that the expression of LincHOXA10 in both RKO and HCT116 cells was downregulated upon transfection with si-LincHOXA10, and the silencing effect of si-LincHOXA10 1# was signi cantly higher than that of si-LincHOXA10 2# and si-LincHOXA10 3# (Fig. 2A).
Thus, si-LincHOXA10 1# was selected for lentiviral vector mediated shRNA knockdown of LincHOXA10 in RKO and HCT116 cells (sh-LincHOXA10), which were used for further functional assays .

Knockdown of LincHOXA10 inhibits tumor growth in vivo
To assess the effect of LincHOXA10 knockdown on tumor growth in vivo, we subcutaneously injected RKO and HCT116 cells that stably express scramble (sh-NC) or LincHOXA10 shRNA into nude mice, and then monitored the growth of the resultant primary tumors. As shown in Fig. 3A and 3B, the xenograft tumors developed at the injection site after 5 days. During a growing period of 25 days, primary tumors derived from LincHOXA10 de cient CRC cells grew signi cantly slower than that derived from control cells ( Fig. 3A and 3B). Moreover, the tumor volumes of the LincHOXA10 de cient groups were signi cantly smaller than those of control groups ( Fig. 3C and 3D).

Knockdown of LincHOXA10 inhibits epithelial to mesenchymal transition (EMT) of CRC cells
EMT is an important mechanism in malignant transformation of tumor cells [25]. We examined whether LincHOXA10 exerted its carcinogenesis through regulating EMT of CRC cells. The key characteristics of EMT are the reduction of E-cadherin along with an increased expression of neuronal cadherin (Ncadherin) and other mesenchymal markers such as vimentin. We found that knockdown of LincHOXA10 in RKO and HCT116 cells signi cantly increased the expression of E-cadherin while decreased N-cadherin and vimentin expression (Fig. 4A-4D), suggesting that down-regulation of LincHOXA10 inhibited the EMT of CRC cells.

Linchoxa10 Interactes With Hoxa10 And Regulates Hohxa10 Protein Expression
It's a common mechanism that lncRNAs can regulate gene expression through participating in the transcription process of nearby genes [26]. Previous studies have shown that LincHOXA10 recruits SNF2L to HOXA10 promoter to regulate its expression [11]. Thus we hypothesize that LincHOXA10 may interact with HOXA10 protein to induce EMT. We detected the expression of HOXA10 in RKO and HCT116 cells when LincHOXA10 was blocked, and found that knockdown of LincHOXA10 signi cantly decreased both the mRNA and protein levels of HOXA10 in RKO (Fig. 5A, 5C and 5E) and HCT116 cells (Fig. 5B, 5C and 5F).
LincHOXA10 is required for TGF-β1/SMADs-induced EMT and HOXA10 rescues LincHOXA10 function It is known that TGF-β/SMADs signaling plays a very important role in EMT of CRC cells [27]. To determine whether LincHOXA10 is required for TGF-β1/SMADs-induced EMT, we knocked down LincHOXA10 expression using shRNA in RKO and HCT116 cells treated with TGF-β1. As shown in Fig. 6A-D, blockade of LincHOXA10 expression restored E-cadherin expression while attenuated Ncadherin and vimentin expression in both RKO and HCT116 cells. Importantly, blockade of LincHOXA10 expression also decreased smad2/3 phosphorylation in RKO and HCT116 cells (Fig. 6A-6D). These results suggested that LincHOXA10 is required for TGF-β1/SMADs -induced EMT of RKO and HCT116 cells.

HOXA10 expression is upregulated in CRC cells and human CRC tissues
To determine whether HOXA10 is expressed in different CRC cells and human CRC tissues, we detected HOXA10 protein expression in control FHC cell and six different CRC cell lines by Western blot and in 30 paired para n-embedded human CRC tissue samples by IHC. The results showed that HOXA10 expression was expressed in all CRC cells detected, and HOXA10 expression in CRC cells was higher than that in FHC cell (Fig. 7A-7D). However, a high level of HOXA10 expression was observed in RKO cells, consistent with LincHOXA10 expression (Fig. 7A). Indeed, HOXA10 expression correlated with LincHOXA10 in most CRC cells (Fig. 7A vs Fig. 7C-7D). Moreover, the protein expression of HOXA10 were signi cantly elevated in 25 out of 30 human CRC tissues (Fig. 7B), consistent with the LincHOXA10 expression (Fig. 1).

Discussion
Many different mechanisms are implicated in tumorigenesis of different cancers. EMT is one of the mechanisms commonly believed to control the process of cancer cell invasion and metastasis [27]. It's well known that many signaling pathways are involved in EMT regulation, such as TGF-β [28], Wnt [29], Notch [30], TNF [31], and BMPs [32]. Several transcription factors also take part in the regulation of EMT, including the Snail/Slug family, Twist, and SIP1/ZEB2, function as molecular switches for the EMT program [33][34][35]. Recent studies have shown that dysregulated expression of lncRNAs in some cancers may regulate EMT and affect disease progression [36]. Here we identi ed LincHOXA10, A novel lncRNA, drives the EMT of CRC cells though TGF-β1/SMADs pathway. Interestingly, HOXA10, a member of HOX transcription factor family, is regulated by LincHOXA10 and rescues LincHOXA10 function.
LncRNAs execute critical modulators in many physiological and pathological processes, including embryonic pluripotency, developmental transitions, differentiation and epigenetic programs of the transcriptome [22,23]. In addition, lncRNAs may also play important roles in driving tumor suppression or in exerting oncogenic functions in a wide variety of cancer types, by promoting tumor cell proliferation, migration, invasion and metastasis [24,37]. Recently, a lncRNA named LincHOXA10, which resides at HOXA10 promoter and coordinates the activation of multiple 5′HOXA genes, has been identi ed as one of 231 lncRNAs associated with the human HOX loci [24]. Furthermore, expression of LincHOXA10 has been identi ed as a negative prognostic factor in glioma patients [24]. However, the functional role of LincHOXA10 in CRC remains unknown. This present study offers the rst insight into the effect of LincHOXA10 on proliferation, migration, invasion and EMT of CRC cells. Here, we found LincHOXA10, which is upregulated in human CRC tissues and cells compared with non-tumoral tissues and cell lines, enhances proliferation, migration and invasion of CRC cells, as well as EMT. In addition, we show that knockdown of LincHOXA10 inhibites the tumorigenesis of CRC cells both in vitro and in vivo.
The HOX family of homeobox genes encodes transcriptional regulators that play critical roles in many processes. For example, HOXA4, HOXA7, HOXA13 were dysregulated in gastric cancer, and participate in gastric carcinogenesis [38]. Here, we identi ed HOXA10, a member of the HOXA gene cluster, as a predominant HOX TF in CRC tumorigenesis. HOXA10 exerts an oncogenic role in several tumors, including oral squamous cell carcinoma [39], pancreatic carcinoma[40], prostate carcinoma[41], hepatocellular carcinoma [10], and leukemia [42]. Overexpression of HOXA10 leads to tumor propagation and it can serve as a good marker for prognosis of various tumors. Despite of the oncogenic role in various tumors, the role of HOXA10 in CRC tumorigenesis is unclear. Here, we detected the expression and critical role of HOXA10 in CRC cells and human tissues. we found HOXA10 is upregulated in CRC cells and tissues, along with LincHOXA10. Interestingly, HOXA10 overexpression can rescue the function of LincHOXA10 on the EMT and smad2/3 phosphorylation. That is to say, HOXA10 mediates LincHOXA10 function in TGF-β1/SMADs -induced EMT.
It's a common mechanism that lncRNAs can regulate gene expression through participating in the transcription process of nearby genes. LincHOXA10 locates at HOXA10 promoter, we have found LincHOXA10 is upregulated in CRC, along with HOXA10. However, further studies are required to elucidate the detailed mechanism of how LincHOXA10 regulates HOXA10 transcription. The results of the present study demonstrated that LincHOXA10, which binds to HOXA10 promoter to drive HOXA10, recruits SNF2L chromatin remodeling complex to HOXA10 promoter to drive its expression [11]. Therefore, the potential mechanism of LincHOXA10-SNF2L-HOXA10 RNA hybrid may act to stabilize the HOXA10 transcript and required for TGF-β1/SMADs-induced EMT of CRC cells.

Conclusions
Our present study demonstrates that LincHOXA10, which is signi cantly overexpressed in CRC, plays a signi cant role in CRC tumorigenesis and metastasis. LincHOXA10 contributes to the CRC progress by regulating HOXA10 expression and consequently TGF-β1/SMADs-induced EMT of CRC. Since LincHOXA10 level is associated with tumor size and lymphatic metastasis, LincHOXA10 may be used as a biomarker to monitor the progression of CRC cancer in human. All animal experiments were conducted such that the animals received ethical and humane treatment, and all procedures were approved by the Institutional Animal Care and Use Committee of Xinxiang Medical University.

Consent for publication
Not applicable.

Availability of data and material
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.