Knockdown of STIL suppresses the progression of gastric cancer by down‐regulating the IGF‐1/PI3K/AKT pathway

Abstract SCL/TAL1 interrupting locus (STIL) regulates the mitotic centrosome to promote the centriolar replication and cell cycling, and is associated with malignancies. However, the role and mechanism of STIL in gastric cancer (GC) remain elusive. STIL expression in GC tissue microarray was detected by immunohistochemistry (IHC). GC cells were transduced with control lentivirus or lentivirus for expression STIL‐specific shRNA and the effect of STIL silencing on the malignant behaviors of GC cells was measured in vitro and in vivo. The potential mechanisms underlying the action of STIL were analyzed by transcriptome microarray and bioinformatics. STIL expression was up‐regulated in GC tissues both in our cohort and the data from the cancer genome atlas, and positively associated with T stage and poor overall survival of GC patients. Knockdown of STIL significantly inhibited the proliferation and clonogenicity of human GC cells and attenuated the growth of implanted GC in vivo. Furthermore, STIL silencing induced cell cycle arrest in G2/M phase and apoptosis of GC cells. Transcriptome analysis indicated that STIL silencing modulated many gene expression, particularly for down‐regulating the IGF‐1/PI3K/AKT pathway. In addition, treatment with SC79, an AKT activator, significantly mitigated the effect of STIL‐silencing in GC cells. In conclusion, STIL promotes gastric carcinogenesis and progression by enhancing the IGF‐1/PI3K/AKT signaling, and STIL may be a novel target for intervention of GC.

Chromosome instability, caused by ectopic centriolar amplification, is ubiquitous in human malignancies, and occurs approximately in 50% of GC. 3 SCL/TAL1 interrupting locus (STIL) is a critical regulator of mitotic centrosome to promote the centriolar replication and cell cycling. 4 During the process of cell cycling, STIL expression is induced in an immediate early fashion, reaches the peak levels in mitosis and is degraded after cell cycle exits from mitosis. 5 STIL through its coiled-coil and STAN domains can interact with Polo-like kinase 4, cyclin dependent kinase 1 (CDK1) and SAS-6 to promote centriole duplication and maintain chromosome stability. 6 It is notable that STIL expression is elevated in multiple types of cancers, such as lung cancer and pancreatic cancer and correlated with the expression of several checkpoint genes and mitotic indicators. 7,8 Furthermore, STIL has been reported to be one of the up-regulated 17 genes in primary adenocarcinoma and their elevated expression is associated with metastasis. 9 Hence, STIL acts as an oncogenic factor to promote the progression of several types of cancers. However, the role of STIL in the development and progression of GC has yet been explored. We aimed to examine the expression of STIL in clinical specimens and human GC cells, investigate the impact of STIL silencing on the malignant behaviors of GC cells in vitro and in vivo, and explore the potential mechanisms underlying the action of STIL in gastric carcinogenesis and progression.

| Tissue microarray and immunohistochemistry
The Tissue microarray (TMA), including 100 GC and 80 para-tumor tissues, was purchased from Shanghai Outdo Biotech (Catalog no. HStmA180Su08). The samples were collected from July 2006 to April 2007. The patients were followed-up until July 2015, with the duration from 8.2 to 9 years. Written informed consent was obtained from each patient, and the protocol was approved by the Ethics Committee of the National Engineering Center for Biochip at Shanghai. Additionally, this study was also approved by the Ethics Committee of Inner Mongolia Hospital. The TMA sections (3 µm) were de-paraffinized and rehydrated, followed by antigen retrieval in citrate buffer in a pressure cooker. The TMA sections were treated with 3% hydrogen peroxide for 10 minutes to block the endogenous peroxidase activity and 3% bovine serum albumin. Subsequently, the sections were probed with polyclonal anti-STIL (Abcam, USA, 1:100) at 4°C overnight. The bound antibodies were detected with horseradish peroxidase (HRP)-conjugated second antibodies and visualized using the chromogen of diaminobenzidine, followed by counterstained with hematoxylin.
The intensity of anti-STIL staining was scored by two pathologists using the Image-Pro Plus 6.0 software in a blinded manner. The scores were determined based on the percentages of positive area (a) and staining density (b) and the values of staining index (a × b) were calculated. The samples were stratified by the median value of staining index for high or low STIL expression.

| TCGA dataset analysis
gastric cancer mRNA data were downloaded from the TCGA database. A total of 402 files, containing 375 tumor and 32 non-tumor tissues, were available for analysis. We employed the R package of "Edger" to screen differentially expressed genes (DEGs) and analyzed the levels of STIL expression in GC and matched non-tumor tissues.

| Quantitative reverse PCR
Total RNA was extracted from individual cell lines using HP Total

| Western blotting
The cells were harvested and lyzed in RIPA buffer, followed by cen-

| Proliferation assay
The impact of STIL silencing on the proliferation of SGC-7901 and BGC-823 cells was determined by CCK-8 assay. Briefly, SGC-7901/ shCon, SGC-7901/shSTIL, BGC-823/shCon and BGC-823/shSTIL cells (2 × 10 3 cells/well) were cultured in triplicate in 96-well plates for 5 days. Every day after culture, individual wells of cells were added with 10 μL of CCK-8 solution and cultured for another 4 hours, and measured for the absorbance at 450 nm in a microplate reader.

| Flow cytometry for cell cycle and apoptosis
After

| Xenografted tumor model
The instructions. Briefly, the RNA samples were reversely transcribed into cDNA that were used as the templates, followed by biotin-labelling using the GeneChip 3'IVT Expression Kit (Affymetrix). The microarray hybridization, washing and staining were performed using GeneChip Hybridization Wash and Stain Kit (Affymetrix). The microarrays were scanned using the GeneChip Scanner 3000 7G to produce raw data.
The DEGs were determined with criteria of a P < 0.05 or absolute Fold Change > 1. The potential functional pathways of DEGs were analyzed by Ingenuity Pathways Analysis software.

| Statistical analysis
Statistical analyses were performed using the SPSS 17.0 and R software. Continuous data are expressed as mean ± SD and analyzed by Student's t test for variables normally distributed or U test for nonparametric variables. Survival curves were plotted using the R software.
The difference was considered statistically significant when a P < 0.05.

| Up-regulated STIL expression is associated with the progression of GC
To determine the potential role of STIL in the development of GC, the expression levels of STIL in 100 GC and 80 non-tumor stomach tissues were determined by TMA-based immunohistochemistry. As shown in Figure 1A, the STIL expression was up-regulated in GC compared with that in adjacent non-tumor tissues. The positivity rate of anti-STIL staining in GC tissues (63%) was significantly higher than 43.8% in adjacent non-tumor tissues (P = 0.01). Moreover, TCGA data indicated that the relative levels of STIL mRNA transcripts in 375 GC tissues were significantly higher than that in 32 non-tumor tissues ( Figure 1B).
In addition, varying levels of STIL mRNA transcripts were detected in the different human GC cell lines ( Figure S1). Stratification analysis indicated that the positivity rate of high STIL expression was only significantly associated with T stage of GC in this population (P = 0.006), but not with other clinical parameters (Table 1). Further analysis revealed that high STIL expression was a risk factor of poor overall survival in GC patients in this population ( Figure 1C, Table 1). Hence, up-regulated STIL expression was associated with the progression of GC.

| STIL silencing suppresses the proliferation and clonogenicity of GC cells in vitro
Next, we employed lentivirus-mediated STIL-specific shRNA expres-  Figure 2C). Collectively, such data indicated that STIL silencing inhibited the proliferation and clonogenicity of GC cells in vitro.

| STIL silencing retards the tumor growth in vivo
To further understand the role of STIL in the progression of GC, BALB/c nude mice were implanted with BGC-823/shCon or BGC-823/shSTIL and the dynamic growth of implanted tumors was monitored up to 21 days post inoculation. The dissected tumors were photoimaged in Figure 3A and the dynamic growth of BGC-823/shSTIL tumors was significantly retarded, as compared with that in the BGC-823/shCon group (P < 0.05, Figure 3B). Furthermore, the weights of BGC-823/ shSTIL tumors were significantly less than that of BGC-823/shCon tumors (P < 0.05, Figure 3C). Thus, STIL silencing attenuated the growth of human GC tumors in mice.

| STIL silencing modulates the global gene expression profiling and signaling in GC cells
To gain new insights into the mechanisms by which STIL regu-  Figure 6A).

| Activating the PI3K-AKT pathway enhances the proliferation and clonogencity of GC cells and mitigated the inhibitory effect of STIL silencing
To

| D ISCUSS I ON
SCL/TAL1 interrupting locus has been thought to be an oncogenic factor and its expression is up-regulated in many types of malignancies. However, there is no report on its role in the progression of GC. In this study, we found that up-regulated STIL expression was significantly associated with T stage and poor overall survival of GC Notes: P a value refers to the association between clinicopathological variables and STIL expression; P b value refers to the association between variables and overall survival. The bold values indicates P values < 0.05. and tumorigenicity of GC cells in vitro and in vivo. These findings provide strong evidence that STIL positively regulates gastric carcinogenesis and progression.
Previous studies have shown that STIL mutation causes the leftright asymmetry defects in mouse, 10,11 disorganized mitotic spindles in zebrafish 12 and primary microcephaly in human. 13 However, Pathways with "orange" or "blue" color were predicted to be activated (z-score > 0) and inhibited (z-score < 0), respectively. (B), Eight DEGs were enriched in the IGF-1 pathway. The DEG with "green" color was down-regulated elevated STIL expression can trigger chromosome instability and cancer. 14 Given that STIL is an regulator of cell cycling by promot- ing the centriolar duplication we tested the impact of STIL silencing on the proliferation and clonogenicity of GC cells. We found that STIL silencing significantly inhibited the proliferation and clonogenicity of GC cells in vitro and the growth of implanted GC in vivo.  (Table S1). Based on the interactome landscape from BioGRID, 20 STIL may interact with ubiquitin-specific protease 7 (USP7), 21 which represses the auto-ubiquitination of checkpoint with forkhead and ring finger domains (CHFR)-E3 ubiquitin ligase. 22 Intriguingly, SP3 was predicted as a high-score substrate of CHFR by the UbiBrowser. 23 Taken together, STIL regulates oncogenic IGF-1/PI3K/AKT pathway in GC, which may attribute to its interaction with the USP7/CHFR/SP3 signaling axis.

| CON CLUS ION
Our data indicated that STIL expression was up-regulated in GC and associated with the progression of GC. STIL silencing promoted the proliferation and clonogenicity of GC cells by inducing cell cycle arrest and apoptosis, which was achieved by attenuating the IGF-1/ PI3K/AKT pathway. Our findings suggest that STIL may be a novel therapeutic target for intervention of GC and may provide new insights into the regulation of STIL in promoting GC progression.

ACK N OWLED G EM ENTS
This work was supported by grants from the National Natural

CO N FLI C T O F I NTE R E S T S
The authors declare no conflicts of interest.