LncRNA ADAMTS9-AS2 suppresses the proliferation of gastric cancer cells and the tumorigenicity of cancer stem cells through regulating SPOP.

Abstract Nowadays, research on CSCs is still in an initial stage, and there are few studies reporting the successful isolation and identification of CSCs. In the present study, we attempted to isolate CSCs through cultivating the cell line MKN45 in defined serum‐free medium and study the expression of stem cell markers or related proteins (Oct3/4, Sox2, Nanog and CD44) in CSCs. Moreover, immunofluorescence staining was performed to validate the stem cell markers of spheroid body‐forming cells. Further experiments were used to evaluate the SPOP expression in tumorsphere cells. In addition, ADAMTS9‐AS2 is a lncRNA that contributes to the genesis and development of many cancers, including gastric cancer (GC). We found ADAMTS9‐AS2 functioned as an anti‐oncogene and positively correlated with the expression of SPOP in GC tissues by combining bioinformatics analyses. Furthermore, we reported that ADAMTS9‐AS2 regulated the expression of SPOP in GC cells and tumorsphere cells to inhibit GC progression. Together, our results demonstrated that SPOP and ADAMTS9‐AS2 can be potential targets for GC treatment.

there is a small part of cells in the tumour tissue, and they can possess self-renewal and multi-directional differentiation potential and strongly associated with the malignant degree of the tumour, drug resistance and metastasis, which plays a crucial role in the development and progression of tumour. 3 Also, the American Association for Cancer Research identified these cells as self-renewal, relatively static, multidrug-resistant and multipotent cells in cancer tissue, as well as cells with stronger tumorigenicity and in vivo metastasis, as compared with other cancer cell lines. So, these cells were deemed cancer stem cells (CSCs). 3 Tumour stem cells can proliferate migrate and produce tumours in an appropriate microenvironment. Some reports have been exhibited that GC tissues also exist some cells with stem cell properties, which play a vital role in the pathogenesis and development of GC. [4][5][6] Therefore, to explore the mechanism of gastric CSCs and further studies on their participation is the key process of gastric carcinogenesis.
Speckle-type POZ protein (SPOP), an E3 ubiquitin ligase adaptor, is a TRAF domain and POZ-containing nuclear speckle-related protein, which forms Cul3-based ubiquitin ligases. SPOP binds with Cul3 ubiquitin ligase and interacts with substrate proteins to constitute a ubiquitin ligase complex. 7 It can form heteromeric species, SPOP-like (SPOPL). The molecular rheostat formed by SPOP and SPOPL can fine-tune the E3 ubiquitin ligase activity. Recent studies have shown that SPOP correlated with the mediation of the ubiquitin and subsequent oncogenic hormone receptor SRC-3 proteolysis, which suggests that SPOP plays a specified role in the inhibition of tumour. 8 Recent research analysis found that SPOP protein was frequently mutated in some tumours, such as endometrial cancer, 9 lung cancer 10 and prostate cancer. 11 Our previous studies reported that SPOP also had a similar situation in colon cancer, 12 but it was not very clear in GC.
Long non-coding RNAs (lncRNAs) are a kind of non-coding RNAs with a length of more than 200 nucleotides with little or no protein-coding potential. 13 Increasing evidence suggests that ln-cRNAs play an important role in the development of malignancies, GC included. Considering the important functions of lncRNAs, it is necessary to identify and explore new lncRNAs in GC. By our bioinformatics analysis of the key lncRNAs in GC, we noticed that a ln-cRNA ADAMTS9-AS2 was identified as a newly described tumour suppressor gene.
In the current study, we studied the SPOP expression in gastric CSCs and the different expressions between them and the primary adherent cells. And we further studied the influences of SPOP and ADAMTS9-AS2 on proliferation, apoptosis, and the cycle of GC cells and spheroid formation of tumorsphere cells, in an attempt to explore the potential effects of ADAMTS9-AS2 and SPOP in the initiation and progression of GC.

| Gastric cancer cell lines and Sphere formation
MKN45 cells were cultured in 1640 medium, supplemented with 10% FBS, and plated at the density of 1 × 10 6 cells per 75 cm 2 flask.
After two weeks, the number of suspension spheres was counted under the microscope (Olympus) and spheroid body formation was calculated. When the primary spheroid body reached the size of approximately 200-500 cells per spheroid body, the spheroid bodies were separated at a density of 1000 cells per ml and 100 single-cell suspension (100 μL) was seeded in each well of 96-well ultra-low attachment plate (Corning) in the serum-free medium. After 2 weeks, the spheroid body formation of the second-generation spheroid body was observed.

| Immunofluorescence staining for stem cell markers
Cells plated onto poly-L-lysine-coated glass coverslips were fixed with 4% paraformaldehyde and then washed with PBS. Cells were permeabilized with 0.1% Triton X-100/PBS for 10 minutes.
Furthermore, cells were probed with fluorescein isothiocyanate or Rhodamine-tagged secondary antibodies. Consequently, the fluorescence was photographed under an inverted microscope (Leica). at room temperature for 2 hours. The bands were scanned using an Odyssey infrared imaging system (LI-COR, Lincoln) and quantified.

| Real-time quantitative PCR
Total RNAs were extracted using TRIzol (Invitrogen). The cDNA Synthesis Kit (Takara) was used for the synthesis of cDNA according to the manufacturer's instructions. qPCR was performed on the Mastercycler Ep realplex (Eppendorf 2S) to detect the mRNA level.

| Flow cytometry analysis
Cell apoptosis analysis was measured using flow cytometry. Cells

| Colony formation assay
Cells were seeded into 6-well culture plates in complete growth medium and cultured for two weeks. Cells in the plates were fixed with 100% methanol for 10 minutes and stained with 0.5% crystal violet for 20 minutes at room temperature, and the number of colonies (>50 cells) was counted under an optical microscope. The experiment was repeated at least three times. SPSS 19.0 software package (SPSS Inc) and Excel 2010 were used for statistical analyses. Student's t test was evaluated to analyse any significant differences. P < .05 was considered to be statistically significant.

| Cancer stem-like properties of tumorsphere cells
To verify whether a cluster of gastric CSCs exists in GC cell lines, The results of Western blot revealed that Oct3/4, Sox2 and CD44 were markedly up-regulated in tumorsphere cells compared with adherent cells ( Figure 1B). In addition, immunofluorescence staining was examined to evaluate the subcellular localization of Oct3/4, Sox2 and CD44 in tumorsphere cells. Double immunofluorescence staining showed that colocalization of Oct3/4 and Sox2 could be found in spheres, which localized to the nucleus of tumorsphere cells. Moreover, the staining of CD44 indicated that CD44 was positively stained in the membrane of tumorsphere cells by using immunofluorescence staining ( Figure 1C).
To confirm whether tumorsphere cells exhibited more tumorigenic than adherent cells in vivo, we examined the tumorigenic capacity between tumorsphere cells and adherent cells. Then, different numbers of tumorsphere cells and adherent cells were injected into the nude mice. We observed that tumorsphere cells formed subcutaneous tumour nodules with larger volume and quicker compared with those from adherent cells in injected mice (Figure 2A).
Representative macroscopic appearances of subcutaneous xenografts in nude mice of tumorsphere cells and adherent cells were shown in Figure S1. These above results indicated that tumorsphere cells were more tumorigenic in vivo.

| The expression of SPOP in gastric cancer stem cells
Interestingly, one previous study showed that SPOP was greatly down-regulated in GC tissues by using Western blot analysis. 14 To examine the role of SPOP in gastric CSCs, we evaluated the expres-

| Identification of ADAMTS9-AS2 as a candidate lncRNA
A prediction model based on protein-coding genes-related lncRNAs was established by calculating the Pearson correlation coefficient as described in Supplemental Methods. The heatmap indicated the SPOP gene-related lncRNAs between GC tissues and adjacent normal tissues, ADAMTS9-AS2 included ( Figure S3). Moreover, statistical software R and edgeR packages were performed to analyse the differentially expressed lncRNAs between tumour and normal tissues. The volcano plot was shown in Figure S4A. The top 100 differentially expressed lncRNAs in GC tissues were screened out and are presented in Figure S4B. As shown in the volcano plot and heatmap, ADAMTS9-AS2 was a lowly expressed lncRNA in GC.
Furthermore, a total of 375 tumour tissues and 32 normal tissues with ADAMTS9-AS2 expression data from the TCGA database across all patient characteristics were analysed. We verified the expression level of ADAMTS9-AS2 using LIMMA software in TCGA data and found that the down-regulation of ADAMTS9-AS2 was in tumour tissues compared with normal tissues ( Figure S4C, P < .001).
To further validate the results, we examined the expression of ADAMTS9-AS2 and the relationship between ADAMTS9-AS2 expression and the expression of SPOP in tumour tissues by using the GEPIA database ( Figure S5).

| ADAMTS9-AS2 partly reversed the effects of SPOP-mediated GC progression
ADAMTS9-AS2 expression was positively correlated with the SPOP expression in GC tissues from Stomach adenocarcinoma (STAD) patients in the TCGA data by using the R package ( Figure 4A). Based on the above-described findings, we detected the effects of SPOP and ADAMTS9-AS2 expression in GC cells by qPCR ( Figure 4B These above results indicated that the expression of SPOP in tumorsphere cells was significantly less than that in adherent cells. *P < .05, **P < .01  investigate CSCs. 15 And sphere formation was regarded as to be an effective approach of isolation or enrichment of stem cell-like cells from multiple tumour types. 16,17 In this paper, in order to isolate gastric CSCs from MKN-45 cells, the spheroid body formation assay was conducted. We found that, as time increased, the cell volume of the suspended spheres increased, and the number increased. Further studies showed that the tumorsphere cells from MKN45 cells possessed higher self-renewal capacity and differentiation potential abilities than adherent cells.

| ADAMTS9-AS2 was involved in SPOPmediated proliferation and spheroid formation
Recently, the expression of CD44, 18 Oct3/4 19 and Sox2 20 have been found in some CSC-like cells, indicating that its expression may be involved in self-renewal and tumorigenesis by activating downstream target genes. And these genes are pluripotent transcription factors and have been identified as stem cell markers. To further identify gastric CSCs, these above different cell-specific surface markers have been assessed. In the present study, our results showed that the expression of Oct3/4, Sox2, CD44 and Nanog was up-regulated in tumorsphere cells. These results indicated that our sorted tumorsphere cells appeared to be enriched with gastric CSCs. In this way, we also verified the existence of gastric CSCs, and we have applied the method 21 to isolate CSCs by serum-free cell culture. More interestingly, the tumorigenicity experiment also exhibited that the tumour formation rate of tumorsphere cells sorted from MKN-45 cell lines was dramatically higher than that of adherent cells. Besides, SPOP is a 374-amino acid protein that comprises a POZ domain downstream of the predicted TD, which has been previously positioned in the nucleus in speckled mode. 22 SPOP is now considered a tumour suppressor because its expression has been shown to inhibit SRC-3-mediated carcinogenic signalling and carcinogenesis. 8 A recently published study showed that SPOP suppressed gastric cancer cell invasion and proliferation by regulating Hh/Gli2 signalling pathway. 14 But so far, there is no research demonstrated that SPOP expression focuses on gastric CSCs. In this study, Western blot and qPCR assays indicated that the expression of SPOP was down-regulated in tumorsphere cells; besides, spheroid formation assay showed that the tumorsphere-SPOP cells formed fewer tumorspheres, indicating that SPOP suppressed the stemness of gastric CSCs. Furthermore, the number of tumorspheres on the tumorsphere-SPOP group was dramatically smaller than that of the control vector group, which indicated that the overexpression of SPOP suppressed the growth of gastric CSCs in vitro.
By some previous studies, we noticed that lncRNA ADAMTS9-AS2 may be strongly associated with some important cancer-related genes. 23 In this study, combining bioinformatics analyses, we discovered the down-regulation of ADAMTS9-AS2 and SPOP in GC cells and revealed that ADAMTS9-AS2 could positively regulate the expression of SPOP. In recent years, emerging research has revealed that lncRNA ADAMTS9-AS2 is identified as a new tumour suppressor gene and is the antisense transcript of ADAMTS9. 24 LncRNA ADAMTS9-AS2 participates in the development of various tumours. Previous researches have reported that down-regulation of ADAMTS9-AS2 was in glioma, 23 colorectal cancer, 25 ovarian cancer 26 and clear cell renal cell carcinoma. 27 In addition, other reports revealed that ADAMTS9-AS2 inhibited the tumour progression by the miR-223-3p/TGFBR3 axis in lung cancer. 28 Shi et al also found that ADAMTS9-AS2 is lowly expressed in breast cancer tissues and drug-resistant breast cancer cells. ADAMTS9-AS2 inhibited PTEN expression and enhanced tamoxifen resistance through targeting microRNA-130a-5p. 29 Moreover, one previous finding has indicated that overexpression of ADAMTS9-AS2 may suppress the proliferation of GC cells, inhibit the migration and invasion of cancer cells and induce apoptosis. The activation of the PI3K/Akt pathway may be an important regulatory mechanism of ADAMTS9-AS2 in GC. 24 However, the role of ADAMTS9-AS2 in gastric CSCs remains unclear. Moreover, in our study, we demonstrated that overexpression of ADAMTS9-AS2 could up-regulate the expression of SPOP, thus suppressing cell growth, proliferation, promote apoptosis and block cell cycle progression of GC and spheroid formation of tumorsphere cells, which suggested that ADAMTS9-AS2 played a crucial role in inhibiting GC progression by targeting SPOP. Our results also suggested that ADAMTS9-AS2 expression may be regulated by DNA methylation of methyltransferases DNMT1. 23 Besides, the expression of SPOP is also controlled by promoter hypermethylation. 30 Therefore, we hold the opinion that the process that lncRNA ADAMTS9-AS2 was involved in SPOP expression regulation in GC might be regulated by DNA methylation. Further researches are needed to study the more precise mechanisms.
Taken together, this study demonstrated that tumorsphere cells from MKN45 cell lines possess gastric CSCs properties. It is apparent that stem cell markers Oct3/4, Sox2 and CD44 maintain the stemness of gastric cancer in tumorsphere cells. SPOP played a vital role in the forming process of CSCs. Moreover, ADAMTS9-AS2 acted as a tumour suppressor that inhibited the progression of GC through regulating the SPOP, and these findings indicated that SPOP and ADAMTS9-AS2 can be potential targets for GC treatment.

ACK N OWLED G EM ENTS
This work was supported by grants of the Nantong Science and Technology Project (MS12019025 and MS12018060) and the Clinical Basic Research Project of Nantong University (2019JY004).

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

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.