A Research of STEAP1 on the Biological Behavior of Gastric Cancer

Introduction:Six-Transmembrane Epithelial Antigene of the Prostate 1 (STEAP1) is associated with the occurrence and development of cancer. This study aimed to clarify the role of STEAP1 in gastric cancer tumor growth and metastasis, as well as its molecular mechanism of action. Methods (cid:0) Statistical methods were used for clinical data analysis. Protein expression was detected using immunohistochemistry(IHC). The mRNA and protein expression in the cell cultures were detected using reverse transcription-polymerase chain reaction(RT-PCR) and western blot analysis. Overexpression and silencing models were constructed using plasmid and lentivirus transfection. To detect cell proliferation in vitro, Cell Counting Kit-8(CCK-8), ow cytometry, and colony formation assays were used; transwell and wound healing assays were used to detect cell migration and invasion; For in vivo experiments, nude BALB/c mice were used for detecting subcutaneous tumorigenesis and intraperitoneal implantation. Results (cid:0) We found STEAP1 was overexpressed in gastric cancer tissues and cell lines. Single factor and Cox analyses showed that STEAP1 gene expression level correlated with poor prognosis. Upregulation of STEAP1 increased cell proliferation, migration, and invasion, which decreased after STEAP1 was knocked down. These changes were achieved via the activation of the AKT/FoxO1 pathway and epithelial-mesenchymal transformation (EMT). The in vivo animal experiments showed that STEAP1 knock down, resulted in a decrease in the subcutaneous tumor and peritoneal tumor formation. Conclusions (cid:0) STEAP1 was overexpressed in gastric cancer and closely connected with OS. STEAP1 can regulate the cell cycle via the Akt/FoxO1 pathway to inuence cell proliferation. STEAP1 may affect cell migration and invasion via EMT induction.


Introduction
Gastric cancer is a common kind of malignant tumor that seriously affects people's health. [1] According to the data from the World health Organization (WHO), in contrast to the United States, Australia and New Zealand, China, Japan, and Chile have higher incidence areas of gastric cancer. [2] In China, gastric cancer is the second cause of cancer-related deaths, with 679000 new cases and 498000 deaths. [3] Gastric cancer is predisposed to occur in individuals aged between 50 and 70 years. However, in recent years, it has shown a "younger" trend. [4][5][6][7][8][9] Therefore, it is essential to nd new tumor markers to predict the risk of gastric cancer progression.
Tumor development involves many factors, which are controlled by many genes, including prostate transmembrane epithelial antigen. Six-Transmembrane Epithelial Antigene of the Prostate (STEAP) was found as a prostate-speci c cell surface antigen using suppression subtractive hybridization technique for the rst time [10,11]. STEAP is highly expressed in spontaneous transgenic mouse prostate cancer models and human prostate cancer. In addition, it is also expressed in the pancreas, ovary, gastrointestinal tract, cervix, testis, bladder, Ewing sarcoma, and melanoma cells [10,12]. There are four members in the STEAP protein family, STEAP1-4. The main focus of our study is STEAP1. Gene STEAP1 is located in the 7q21.13 region of the human chromosome; it is 10.4 kb long and contains four introns and ve exons. The transcription of the gene STEAP1 can produce two different kinds of mRNAs: a 1.4 kb and a 4 kb mRNAs. However, only the 1.4 kb mRNA can be processed into a mature protein, which contains 339 amino acids with molecular weight of 36 KD. [10,12], while the 4 kb mRNA contains a 2399 BP large intron, which is not translated into a mature protein [13]. Data has shown that gene STEAP1 is closely related to communication between the adjacent cells, and it seemed to be bene cial for the occurrence and development of tumors. [14] Its structural prediction and the location at the cell-cell contacts indicated that gene STEAP1 product may be a transporter or channel. [10,15] Some previous research on different kinds of cancer have found that STEAP1 was observed in tumor tissue but not in normal tissue. The expression of STEAP1 was closely related to the malignant phenotype of cancer cells. [13,[16][17][18][19][20] However, in Lee's study, they found no correlation between expression of STEAP1 and the clinicopathological factors [21]. These con icting results indicated that the roles of STEAP1 were varied depending on different cancer types. In our study, we will discuss the in uence of STEAP1 on the proliferation, invasion, and in ammatory reactions in gastric cancer.

Materials And Methods
Clinical samples 212 samples of gastric cancer tissue and 60 samples of adjacent normal gastric tissue from patients were obtained from the First A liated Hospital of China Medical University from 2003 to 2010. None of the patients received preoperative chemotherapy or radiotherapy, and all of them were proven to have gastric cancer by pathology. The cancer tissue was xed with formalin and preserved in para n. All pathological data were complete, and the postoperative follow-up was su cient. All patients were approved by the ethics committee of China Medical University to participate in the study and provided written informed consent.

Immunohistochemistry of human gastric cancer
To x gastric cancer tissue samples, 10% formalin was used, then the tissue was para n-embedded and cut into 4-µm slices. The xylene and alcohol were used for dewaxing and rehydrating. Endogenous peroxidase activity was blocked by using hydrogen peroxide (30%), the citrate buffer (pH 6.0) was used to boil the the sections fro 3 minutes in a pressure cooker. Next, normal goat serum was used for incubating the sections to reduce the nonspeci c binding. Finally, the tissue sections were incubated (4 °C, 12 h) with anti-STEAP1 antibody (1:200 dilution, B-4, SC-271872, Santa Cruz, USA). Enzyme labeled anti mouse / rabbit IgG polymer was used for secondary antibody. (Mai Xin Biological company,Fuzhou ,China).
Finally, the sections were stained with diaminobezidin (DAB) for 60 sec, stained with hematoxylin for 2 min and sealed with neutral resin. Fluorescence photographic microscope was used for obtaining images(Nikon,Japan).Two pathologists examined all tumor slides randomly. We evaluated STEAP1 staining intensity as follows: scored 0 (negative), 1 (weakly negative), 2 (weak positive), and 3 (strong positive). The percentage scores of positive cells per single eld vision were as follows: scored 1 (0-25%), 2 (26-50%), 3 (51-75%), and 4 (76-100%). We multiplied the two scores above and obtained a nal score ranging from 0 to 12. Tumor samples with a score < 6 were considered as negative expression; on the contrary, the score ≥ 6 was considered as positive expression.

Animals
Twenty-four BALB/c nude female mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were raised in the animal experimental center of of China Medical University. The twenty-four nude mice were randomly divided into four groups. Two groups were used for hypodermic injection, and the other two groups were used for intraperitoneally injected .In the subcutaneous tumorigenesis experiment, 3 × 106 SGC-7901 NC cells per mouse were injected in NC group,and 3*106 SGC-7901 sh-STEAP1 cells per mouse were injected in sh-STEAP1 group .Similarly, in the experiment of intraperitoneal tumorigenesis,3*106 SGC-7901 NC cells per mouse were injected in NC group,and 3 × 106 SGC-7901 sh-STEAP1 cells per mouse were injected in sh-STEAP1 group .

Cell transfection
ShRNA lentivirus was purchased from Shanghai Genechem Co., Ltd. The NC group insertion sequence was TTCTCCGAACGTGTCACGT. There were three shRNA sequences used (shRNA1: CCAACTTCATAATGGAACCAA; shRNA2: CAGCACACACAGGAACTCTTT; and shRNA3: AAGCTAGGAATTGTTTCCCTT). The STEAP1 containing cDNA plasmid and Flag empty plasmid were purchased from Beijing SinoBiological Co., Ltd. Lipofectamine 3000 reagent (Thermo Fisher Scienti c, Inc.) was used for plasmid transfection. The cells were harvested 48 h after transfection. Western blot analysis and RT-PCR were used to check the transfection e ciency.

RT-PCR
TRIzol reagent was used to extracted total RNA. The reverse transcription kit PrimeScript RT was purchased from Takara. The primer designs were provided by Huada Gene Co., and the sequences were shown in Table 1.  purchased from Cell Signaling Technology, USA. Then the goat anti-rabbit and anti-mouse IgG secondary antibodies were used to incubate the membrane at room temperature for 60 min. Finally, the ECL was used to visualize and detect the proteins by using BioImaging Systems (UVP Inc., Upland, CA, USA

Colony formation assay
For the colony formation assay, SGC-7901 and MGC-803 cells were transfected with plasmid or shRNA for 36 h and plated into 6-well cell plates (1000 cells/well). The cells were cultured in a 37 °C incubator for 2-3 weeks, xed with alcohol for 30 min, and stained with Trypan Blue for 20 min at room temperature. The colonies with more than 50 cells were counted. Finally, an HD camera was used to obtain the images.The experiment was repeated three times.

Flow cytometry
We used a ow cytometry assay to detect the cell cycle stage. SGC-7901 and MGC-803 cells were transfected with plasmid or shRNA and plated into 6-well plates (1 × 10 5 cells/well). After 24 h, the cells were harvested using 0.25% trypsin in 1.5 ml Eppendorf tubes. Then, the cells were stained with propidium iodide (PI, 500 µ/tube, KeyGEN, Nanjing, China) at 37 °C in the dark for 30 min. Finally, the cells were analyzed using a FACSCalibur ow cytometer (Becton Dickinson, USA). The experiment was repeated three times.

Transwell assay
Cell migration experiments were performed using a 24-well transwell chamber with a pore size of 8 µm (Costar). A total of 5 × 10 4 cells in serum-free DMEM were placed in the upper chamber, and DMEM with 10% FBS was added to the lower chamber. After more than 10 hours, the migration experiment was terminated, and the cells were observed in the medium below. Then, the cells on the membrane in the bottom chamber were xed with 75% alcohol for 30 min and stained with Trypan Blue at room temperature for 20 min. Images were obtained using an inverted microscope. In addition, the transwell chamber was also used for cell invasion experiments. For these experiments, in addition to the above steps, Matrigel (1:9 dilution, BD Bioscience) was added to the upper chamber to observe the change in cell invasion ability.The experiment was repeated three times.
Wound healing assay A wound healing assay was used to observe the migration of cells. In this study, 1 × 10 5 cells were seeded into 6-well plates for every group. After the cells had covered the entire plate, a pipette tip was used to make a scratch in the cell monolayer, and phosphate buffer saline (PBS) was used to wash the oating cells three times. Subsequently, we used serum-free DMEM instead of the former medium. Finally, an inverted microscope (Olympus, Japan) was used to take images at 0 h and 96 h. The difference in scratch distance between the two phases can re ect the difference in the cell migration ability.The experiment was repeated three times.
Statistical analysis GraphPad Prism 7.0 was used for image editing. SPSS 21.0 statistical software was used for data analysis.The data of three repeated experiments were input to analysis and expressed as the means ± SEMs. The chi-square test was used to examine possible correlations between STEAP1 expression and clinicopathological factors. Survival rates were calculated using Kaplan-Meier analysis. The log-rank test was used for single-factor analysis. Cox risk proportion model was used for multi-factor analysis, and a value of P < 0.05 was considered statistically signi cant.

Results
STEAP1 was highly expressed in gastric cancer tissue and closely connected with OS.
The GEPIA database showed that the STEAP1 gene was more highly expressed in gastric cancer than in the normal tissue (Fig. 1a). We detected 212 cases of gastric cancer tissues and 60 cases of paracancerous tissues using IHC and scored them. The results showed that STEAP1 was highly expressed in cancer tissues and was mainly localized to the membrane and cytoplasm in cells (Fig. 1b).
The positive expression rate of STEAP1 was 54.7% (116/212). However, it was expressed at low levels or was negative in paracancerous tissues (Fig. 1c). The 5-year OS in the high expression group was 25.9%, which was signi cantly lower than 60.7% in patients with low expression group (P < 0.001, Fig. d). Figure 1e shows the detailed score in 212 cases of tumor tissues and in 60 cases of paracancerous tissues. The scores between the two groups were signi cantly different (P < 0.001). In addition, 60 matched tissues were scored, and the details are shown in Fig. 1f (P < 0.001). The subsequent study of the data of 212 clinical cases showed that the factors that affected the prognosis in patients included tumor location (P = 0.029), tumor size (P = 0.012), Borrmann type (P = 0.019), STEAP1 expression (P < 0.001), N stage (P < 0.001), T stage (P < 0.001), and distant metastasis (P = 0.005) ( Table 2). Cox multifactor analysis showed that the independent factors in uencing the prognosis of patients included STEAP1 expression (P < 0.001), T stage (P = 0.005), and N stage (P < 0.001) ( Table 3). In the study of the relationship between the expression of STEAP1 and clinicopathological factors, high expression of STEAP1 was closely related to Borrmann type (P = 0.009) and N stage (P < 0.001) ( Table 4).   * High and medium differentiated tubular adenocarcinoma and papillary adenocarcinoma were regarded as differentiated types; mucous adenocarcinoma, signet ring cell carcinoma, low and undifferentiated adenocarcinoma were regarded as undifferentiated types.

Screening of the experimental cell lines and knockdown virus transfection
We detected the STEAP1 mRNA level in GES-1, AGS, SGC-7901, and MGC-803 cells using RT-PCR (Fig. 1g). The results showed that STEAP1 was more highly expressed in SGC-7901 and MGC-803 cells.
The western blot analysis yielded the same conclusion (Fig. 1h). Therefore, we selected SGC-7901 and MGC-803 cells as experimental cell lines. We overexpressed and knocked down the STEAP1 gene by transfecting the STEAP1 plasmid and STEAP1 shRNA. We transfected negative control (NC) virus and three kinds of STEAP1-shRNAs into SGC-7901 cells and detected the knockdown e ciency on mRNA and protein levels using RT-PCR and western blot analysis. The results showed that shRNA2 had the highest knockdown e ciency (Fig. 1i). Subsequently, we transfected NC virus and STEAP1-shRNA2 into SGC-7901 and MGC-803 cells. The results showed successful transfection and knockdown using RT-PCR and western blot analysis (Fig. 1j-k).
STEAP1 gene regulates the cell cycle via the Akt/FoxO1 pathway to in uence cell proliferation.
The CCK-8 assay results showed that the absorbance in the sh-STEAP1 group was lower than that in the NC group at 48 h, 72 h, and 96 h in both the SGC-7901 and MGC-803 cell lines (Fig. 2a). In another group of comparisons, we found that the absorbance of the STEAP1 plasmid vector group was higher than that of the NC group at 24 h, 48 h, 72 h, and 96 h in both the SGC-7901 and MGC-803 cell lines (Fig. 2b). The CCK-8 results indicated that the STEAP1 gene can in uence cell proliferation. In the colony formation assay, we found that when STEAP1 was knocked down, the colony number was lower than that in the NC group (Fig. 2c), while when STEAP1 was overexpressed, the colony number was higher than that in the empty vector group (Fig. 2d). The colony formation assay results also indicated that the STEAP1 gene can in uence cell proliferation. Next, we used ow cytometry to detect the cell cycle. The results showed that when STEAP1 was knocked down, the percentage of the cells in S phase was decreased, and the percentages of cells in G0/G1 and G2/M phase was increased in both SGC-7901 and MGC-803 cells (Fig. 2e). When STEAP1 was overexpressed, the percentage of S phase cells was increased, and the percentages of G0/G1 and G2/M phase cells was decreased in both SGC-7901 and MGC-803 cells (Fig. 2f). The ow cytometry assay indicated that the STEAP1 gene can in uence cell proliferation by in uencing the cell cycle. Finally, we detected cell cycle related proteins and pathway proteins to identify the underlying mechanism. The Western blot analysis showed that when STEAP1 was down regulated, CDK4 and Cyclin D1 were relatively down regulated, and P27 was upregulated. Total AKT (AKT) and total FoxO1 (FoxO1) showed no signi cant change, while phosphorylated AKT (P-AKT) was down regulated, and phosphorylated FoxO1 (P-FoxO1) was upregulated (Fig. 2g). When STEAP1 was upregulated, CDK4 and Cyclin D1 were relatively upregulated, and P27 was down regulated. AKT and FoxO1 also showed no signi cant change, while P-AKT was upregulated and P-FoxO1 was down regulated (Fig. 2h).

STEAP1 regulates cell migration and invasion via EMT.
The transwell assay results showed that the number of migrating cells in the sh-STEAP1 group was lower than that in the NC group in both SGC-7901 and MGC-803 cells (Fig. 3a). The number of migrating cells in the STEAP1 plasmid group was higher than that in the empty vector group (Fig. 3b). The results of the wound healing assay also provided consistent conclusions. In SGC-7901 cells, the migration distance between 0 h and 96 h in the NC group was more obvious than that in the sh-STEAP1 group. The migration distance of the empty vector group was shorter than that of the STEAP1 plasmid group (Fig. 3c). This result was also identi ed in MGC-803 cells. The migration distance of the NC group was longer than that of the sh-STEAP1 group, and the migration distance of the empty vector group was shorter than that of the STEAP1 plasmid group (Fig. 3d). The results of the above two experiments indicated that when STEAP1 was knocked down, the cell migration ability was decreased, whereas it increased after the overexpression of STEAP1. We also used a transwell assay and placed Matrigel into the upper chamber for detecting the effect of the STEAP1 gene on cell invasion. The results showed that the number of invading cells in the sh-STEAP1 group was lower than that in the NC group in both SGC-7901 and MGC-803 cells (Fig. 3e). The number of invading cells in the STEAP1 plasmid group was higher than that in the empty vector group (Fig. 3f). These results indicated that the STEAP1 gene had an effect on cell invasion. When STEAP1 was knocked down, the cell invasion ability was decreased and increased after overexpressing STEAP1. Finally, we detected cell migration-and invasion-related proteins using western blot analysis. The results showed that Vimentin, N-cadherin, MMP-2, and MMP-9 were downregulated and E-cadherin was upregulated after we downregulated the gene STEAP1 (Fig. 3g). In contrast, when STEAP1 was overexpressed, Vimentin, N-cadherin, MMP-2, and MMP-9 were relatively upregulated, and E-cadherin was downregulated (Fig. 3h).

In vivo animal experiments
Twelve BALB/c nude mice were used to study tumor formation and were randomly divided into two groups. We subcutaneously injected 3 × 10 6 of SGC-7901 NC cells per mouse in the rst group, which was called the NC group. The mice in the other group were subcutaneously injected with 3 × 10 6 of SGC-7901 and sh-STEAP1 cells per mouse, which was called sh-STEAP1 group. The tumor sizes were measured every 2 days from 4 to 14 days after the injection and the results are shown in Fig. 4c. Twelve tumor specimens were removed from the mice on the 14th day. The results showed that the tumor size in the NC group was larger than that in the sh-STEAP1 group (Fig. 4a). Then, we carried out para n embedding, sectioning and IHC experiments with the tumor tissue. The results showed that the expression of Ki67 in the NC group was higher than that in the sh-STEAP1 group, while cleaved caspase-3 expression was lower than that in the sh-STEAP1 group (Fig. 4b).
In the intraperitoneal tumorigenesis experiment, similar to the subcutaneous tumorigenesis experiment, twelve mice were divided into the NC and sh-STEAP1 groups, and 3 × 10 6 cells per mouse were injected by intraperitoneal injection. Three weeks later, the mice were sacri ced to observe the number of intraperitoneal tumors,including the mesentery, on the wall of the intestine (Fig. 4d). We used hemostatic forceps to clamp the two sides of the intestine to expand the mesentery,We found a huge number of tumors in the mesentery arranged like beads. However, tumors were rare or absent in the NC group (Fig. 4e).The number of tumors in the abdominal cavity of the two groups were statistically analyzed and the difference was statistically signi cant (Fig. 4f). We enlarged the image of a case in NC group when the abdominal cavity of nude mice was just opened. We can clearly see that the tumors covered the abdominal cavity, and several larger tumors on the mesenteric and intestinal wall were marked at the arrow. (Fig. 4g)

Discussion
As a global health problem, cancer affects the quality of patients' lives worldwide and causes thousands of deaths every year [1]. STEAP1 is overexpressed in many kinds of cancers, such as prostate cancer, colon cancer, bladder cancer, ovarian cancer, pancreatic cancer, testicular cancer, breast cancer, cervical cancer and Ewing sarcoma [10,12]. In our study, STEAP1 was overexpressed in gastric cancer and closely related to the prognosis of patients (Fig. 1b). The 5-year OS of patients with a low expression of STEAP1 was 60.7%, while that of patients with a high expression of STEAP1 was only 25.9% (Fig. 1d). STEAP1 plays a role as an oncogene in gastric cancer, and this result was consistent with the conclusion that STEAP1 is an oncogene in other kinds of cancer. Tumor cell growth, metastasis, proliferation, migration and invasion are basic biological functions [22]. We downregulated or upregulated STEAP1 by using lentivirus knockdown or STEAP1 plasmids, respectively, to detect the changes in the above functions. The results of the CCK-8 and colony formation assays indicated that when we overexpressed STEAP1, the percentage of cells in the S phase increased, that in the G0/G1 and G2/M phases decreased, and the cell proliferation ability was also improved. In contrast, when STEAP1 was knocked down, the percentage of cells in the S phase decreased, that in the G0/G1 and G2/M phases increased, and cell proliferation was reduced ( Fig. 2a-f). Cyclin D1 is known as an oncogene and overexpressed in many kinds of cancers [23].
By binding with CDK4 ( a partner kinases of cyclin D1), cyclin D1 can release transcription factors and advance cell cycle progression from the G1 phase to the S phase. The P27 protein limits cell cycle progression, mainly by inhibiting complex formation, such as CyclinD1-CDK4 and CyclinE-CDK2, to block the cell cycle in the G1 phase. Previous studies have found that the AKT pathway is one of the main signaling pathways in uencing cancer cell proliferation [24][25][26][27][28][29]. Therefore, it was reasonable to consider that STEAP1 can affect cell proliferation via the AKT pathway. Our results also showed that when STEAP1 was downregulated, P-AKT, CDK4 and Cyclin D1 were relatively downregulated, and P27 and P-FoxO1 were upregulated (Fig. 2g). When STEAP1 was upregulated, P-AKT, CDK4 and Cyclin D1 were relatively upregulated, and P-FoxO1 and P27 were downregulated. P-AKT was upregulated and P-FoxO1 was downregulated (Fig. 2h). These results indicated that STEAP1 can regulate the cell cycle via the Akt/FoxO1 pathway to in uence cell proliferation. The results of the transwell and wound healing assays showed that when we overexpressed STEAP1, cell migration and invasion increased. In contrast, when STEAP1 was knocked down, the two abilities above decreased (Fig. 3a-f). Next, EMT-related proteins, MMP2 and MMP9 were detected by western blotting. The EMT-related proteins include N-cadherin ,vimentin and E-cadherin. A previous study showed that when cells tend to migrate and metastasize,the protein expression of vimentin and N-cadherin increases and that of E-cadherin decreases [30,31]. In addition, many studies have identi ed that MMP2 and MMP9 are closely related to tumor migration and invasion and explained the mechanism [32][33][34]. The results of our study showed that when STEAP1 was overexpressed, N-cadherin, Vimentin, MMP-2 and MMP-9 were relatively upregulated, and E-cadherin was downregulated. When STEAP1 was downregulated, N-cadherin, MMP-9 and MMP-2 were downregulated, and E-cadherin was upregulated ( Fig. 3h-g). These results indicated that the STEAP1 gene may regulate cell migration and invasion via EMT. The result of the apoptosis marker cleaved caspase-3 and proliferation-related nuclear antigen Ki-67 by IHC assay also indicated that STEAP1 plays a very important role in cell proliferation in vitro. Until the early 1990s, We always admitted that peritoneal metastasis of gastric cancer is a kind of terminal disease and the effect of systemic chemotherapy is limited for it. In our study, we established a peritoneal metastasis model by intraperitoneal injection of tumor cells into mice. The results showed that the number of tumors on the mesentery in the NC group was higher than that in the sh-STEAP1 group. Through this experiment, we veri ed the effect of STEAP1 on tumor cell invasion and metastasis in vivo.

Conclusions
In