STEAP1 facilitates metastasis and epithelial–mesenchymal transition of lung adenocarcinoma via the JAK2/STAT3 signaling pathway

Abstract Six-transmembrane epithelial antigen of prostate-1 (STEAP1) is a relatively newly identified gene target from prostate cancer, breast cancer, and gastric cancer. However, functions of STEAP1 in lung adenocarcinoma (LUAD) are still unknown. In the present study, we explored the molecular and cellular mechanisms of STEAP1 in LUAD. Western blot and Q-PCR were conducted to detect the protein and mRNA expressions respectively. The cell proliferation was tested by CCK8 assay. The effects of STEAP1 on the metastasis and epithelial–mesenchymal transition (EMT) of LUAD were evaluated by EdU assay, wound healing assay, and transwell migratory assay. H1650, H358, HCC827, H1299, H23, A549, H1693 were selected as human LUAD cell lines in the study. Results have shown that STEAP1 expression was up-regulated in LUAD cells compared with normal lung epithelial cells. Knockdowning of STEAP1 suppressed the proliferation, migration, and invasion of LUAD epithelial cells. Importantly, after comparing the proliferation, migration, and invasion of LUAD to the corresponding control groups treated in STAT3 inhibitor ADZ1480, we found that STEAP1 regulates EMT via Janus kinase 2/signal transducer and activator of transcription 3 (JAK2/STAT3) signaling pathway. In conclusion, STEAP1 can serve as a therapeutic target, and it may have important clinical implications for LUAD treatment.


Introduction
Nowadays, lung cancer is one of the major causes of cancer-related deaths [1]. According to the related publications, lung cancer can be generally divided into two subtypes: small-cell lung carcinoma (SCLC) and non-small-cell lung carcinoma (NSCLC), each of them accounts for 15% and 85% of all lung cancer cases, respectively [2]. Lung adenocarcinoma (LUAD) is likely to occur in women, asian people, and non-smokers [3]. LUAD is often metastatic and has a relatively poor prognosis [4]. Chemotherapy, surgery, and radiotherapy are traditional treatments for LUAD [5,6], but the recurrence of early post-operative lung cancer patients are still very high [7]. Meanwhile, chemotherapy and targeted therapies have extensive drug resistance [8,9]. Thus, to explore the molecular mechanism of target genes are important strategy for searching of LUAD treatment.
Six-transmembrane epithelial antigen of prostate-1 (STEAP1) is a novel 339 amino acids cell surface protein. It appears to be an ion channel or transporter, plays a role in cell adhesion and may be related to tumor proliferation and invasion [10]. The relationship between STEAP1 and LUAD is barely reported. The Janus kinase (JAK)2/signal transducer and activator of transcription (STAT)3 signaling pathway are proven often excessive activation in tumors and they are in relevant with various physiological processes, and a wide range genes are affected by them. In this report, we will focus on the regulatory functions of STEAP1 in proliferation, metastasis and invasion of LUAD, and how the JAK2/STAT3 signaling pathway is involved in these processes.

Western blot assay
LUAD cells were washed with cold PBS for three times to prepare whole-cell protein extracts. Then, 8% or 10% sodium dodecylsulfate-poly-acrylamide gel electrophoresis (SDS-PAGE) was used to separate the proteins according to the molecular weight. Proteins were then transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were probed with the primary antibodies after washing three times with PBST, the membranes then probed with appropriate secondary antibodies (Cell Signaling Technology). Enhanced chemiluminescence (ECL) plus detection kit was applied for detecting the immunoreactive bands. The primary antibodies used were as follows: GAPDH (cat:#5174,CST) and STEAP1 (cat:#PA5-20404, Invitrogen).

Real-time quantitative PCR (Q)-PCR assay
The RNA extraction kit was used to extract total RNA. Then the FastQuant RT Kit (With gDNase) was applied to reverse transcribe RNA into cDNA. The PCR mixtures were prepared according to the instructions of SuperReal PreMix Plus (SYBR Green) Kit. PCR amplification and analysis were conducted on the ABI 7300 real-time PCR machine. All C t values were normalized by the C t value of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Primer lists were as follows: GAPDH F: 5 -TGCACCACCAACTGCTTAGC-3 , R: 5 -GGCATGGACTGTGGTCATGAG-3 ; STEAP1 186-193 F: 5 -AATTCAGATCCTACAGATACAAGCTACTCTA-3 , R: 5 -AGCTTAGAGTAGCTTGTATCTGTAGGATCTG-3 .

Immunohistochemical analysis
Paraffin-embedded tissues were sectioned at 5 μm thickness. Slides were baked at 60 • C for 1 h, then deparaffinized, finally rehydrated for 10 min in sodium citrate buffer. Sections were cooled to room temperature, treated with 3% H 2 O 2 for 10 min and blocked with 5% goat serum for 40 min at room temperature. Primary STEAP1 antibodies were used for incubation for 2-3 h at 37 • C (STEAP1 polyclonal antibody (Cat:#PA5-20404) with a concentration of 25 μg/ml (Thermo Fisher) Then, sections were washed in phosphate-buffered saline (PBS) and incubated with the secondary antibody (biotinylated goat anti-rabbit, diluted 1:200) for 30 min. After washing again with PBS, the sections were dehydrated with ethanol for 2 min, followed by xylene transparency for 5 min then tablets were eventually quickly sealed with neutral gum and ultra-thin cover glass. Sections were observed under ordinary optical microscope.

Wound healing test
Equal numbers of H1299NC, H1299shRNA, A549vector, and A549STEAP1 cells were seeded in 6-cm dishes wells, the next day when the cells were attached, vertical wounds were made on the cell monolayers cells by using 200 μl pipette tips. Then scratched cells were washed off with phosphate buffer saline (PBS). An optical microscope was used to photograph changes of wounds after being scratched for 0 and 48 h.

CCK8 assay
The cell counting kit-8 (CCK-8) was implemented to measure cell proliferation. The cells from each groups were seeded at a density of 2000 cells/well in 100 μl media in the triplicate groups in 96-well plates. Starting from the following day, 10 μl CCK-8 reagent were add into each well and incubated at 37 • C for 2-4 h, and the OD value at 450 nm wavelength was measured using a microplate reader.

Transwell invasion assay and transwell migration assay
Transwell invasion assay was performed using the 24-well invasion chambers (Corning, BioCoat, U.S.A.) with an 8 μm pore size PET membrane with Matriel Matrx being treated. Cells were trypsinized and resuspended, then inoculated into the upper Matrgel chamber in 500 μl of serum-free DMEM medium at a concentration of 1 × 10 5 cells/well. DMEM medium containing 30% FBS in the lower chamber served as the chemoattractant. At the end of incubation, the noninvading cells on the upper membrane surface were erased with cotton swabs. The invaded cells on the lower surface of the membrane were fixed and stained with Giemsa for 5 min. Nine visual fields of each chamber were randomly chosen, and observed under a X71 microscope (Olympus, Tokyo, Japan), and the number of invading cells from pictures was counted by cell counter.

Cell transfection and reagents used
Transfection of shRNA was carried out using lipofectamine 2000 (Invitrogen) following the manufacturer's instructions. Briefly, cells were seeded at a concentration of 2 × 10 5 cells/dish (6 cm) and grown to 70% confluence. Lipofectamine 2000 and shRNA were then mixed and the mixture was incubated in Opti-MEM at room temperature for 15 min. Subsequently, the cells were incubated in medium for 24 h, and then harvested for assays. The recombinant vector was validated by Sanger sequencing. The empty vector pEnter (Vigene) was used as the negative control. The

Statistical analysis
Q-PCR data were analyzed by GraphPad Prism 7 software. T-test was used for the comparison of samples within two groups. P<0.05 was considered statistically significant.

STEAP1 expression in lung adenocarcinoma tissues
We investigated the expression of STEAP1 in LUAD tissues and para-carcinoma normal tissues, results have shown that STEAP1 expression level in LUAD tissues was significantly higher than para-carcinoma normal tissues ( Figure  1A-C). The relative STEAP1 expression of LUAD tissues and para-carcinoma normal tissues in the gene expression profiling interactive analysis (GEPIA) dataset were shown in Figure 1D. Data obtained from the GEPIA revealed that overall survival in the low STEAP1 group was higher than that in high STEAP1 group ( Figure 1E). The correlation

STEAP1 promotes proliferation of lung adenocarcinoma cells
To further validate the results listed above, we first nvestigated the protein and mRNA expression levels of STEAP1 in seven common LUAD cell lines (H1650, H358, HCC827, H1299, H23, A549, and H1693). Western blot and Q-PCR showed that among seven LUAD cell lines, H1299 was mostly expressed and A549 was least expression (Figure 2A). Second, to investigate the effects of knocking down and overexpression of STEAP1, we designed shRNA targeting at STEAP1, and its random control shRNA(NS) and transfected them in H1299 cell line. We develop STEAP1 expression plasmid and empty vector control and did overexpression experiment in A549 cells ( Figure 2B). The cell proliferation of each group of cells (H1299NC, H1299shRNA, A549vector, and A549STEAP1) was measured with CCK8 reagent for five consecutive days, and the results showed that the OD450 of A549STEAP1 and H1299shRNA were significantly lower than that of A549vector and H1299NC groups on days 2-5 (both P<0.01, Figure 2C). Ratio of EdU positive cells in H1299NC was significantly higher than that in H1299shRNA group and the ratio of EdU positive cells of A549STEAP1 was significantly higher than that of A549vector ( Figure 2D). We also estimated the expression of proliferation-related proteins, PCNA and cyclinD1, in H1299NC, H1299shRNA, A549vector, and A549STEAP1 groups, Western blot and Q-PCR assays results showed that the expressions of cyclinD1 in H1299NC were significantly higher than that in H1299shRNA, meanwhile, A549STEAP1 was significantly higher than A549vector. Besides, the expression of PCNA in H1299NC was significantly higher than that in H1299shRNA, A549STEAP1 was significantly higher than A549vector ( Figure 2E).

STEAP1 promotes migration and invasion of lung adenocarcinoma cells
Next, wound healing assay was used to examine the migration of cells with each treatment. Results showed that the cell migration of H1299NC group was significantly slower compared with H1299shRNA, the migration of A549STEAP1 group cells was significantly slower compared with A549vector group ( Figure 3A). Moreover, cell numbers of H1299shRNA group were significantly greater than that of H1299NC group, and A549vector group was significantly greater than that of A549STEAP1 group ( Figure 3B). Cell numbers in H1299shRNA group were significantly greater than that in H1299NC group, in addition, A549vector group were significantly greater than that of A549STEAP1 group ( Figure 3C). Furthermore, We analyzed the expression of migration-related proteins, mmp9, mmp2, and snail1, Western blot and Q-PCR results indicated that the expression of snail1 in H1299NC group was significantly higher than H1299shRNA group, the A549STEAP1 group was significantly higher than A549vector group. The expression of mmp2 in H1299NC group was significantly higher than H1299shRNA group, A549STEAP1 group was significantly higher than A549vector group. The expression of mmp9 in H1299NC group was significantly higher than H1299shRNA group, A549STEAP1 group was significantly higher than A549vector group ( Figure 3D).

STEAP1 regulates epithelial-mesenchymal transition via JAK2/STAT3 signaling pathway
Previous studies have showed that proteins in key signaling pathways can regulate tumor growth. We hypothesis that STEAP1 can regulate epithelial-mesenchymal transition (EMT) via JAK2/STAT3 signaling pathway. To verify, Western blot analysis and Q-PCR were employed to detect the protein and mRNA expressions of N-cadherin, E-cadherin, and vimentin in H1299NC, H1299shRNA, A549vector, and A549STEAP1 groups. Our results showed that E-cadherin expression was markedly increased, however, vimentin and N-cadherin expression were decreased after STEAP1 was silenced in A1299 cells ( Figure 4A,B). Besides, the protein and mRNA levels of JAK2, p-JAK2, STAT3, and p-STAT3 in H1299NC, H1299shRNA, A549vector, and A549STEAP1 groups were analyzed through Western blotting and Q-PCR. The results showed JAK2, p-JAK2, STAT3, and p-STAT3 were markedly depleted by H1299shRNA at both protein and mRNA levels. The protein and mRNA levels of JAK2, p-JAK2, STAT3, and p-STAT3 were markedly increased in A549STEAP1 group ( Figure 4C,D). The protein and mRNA expression of JAK2/STAT3 signaling pathway proteins, LCAM1, CCL12, and survivin were markedly decreased in H1299shRNA group. The protein and mRNA levels of LCAM1, CCL12, and survivin were markedly increased in A549STEAP1 group. ( Figure  4E,G). ADZ1480, a specific STAT3 inhibitor, we investigated the influences of ADZ1480 on the EMT in A549 cells, which can be shown in Figure 4F. ADZ1480 increases E-cadherin expression and lower the expression level of vimentin and N-cadherin.

STEAP1 facilitates metastasis and epithelial-mesenchymal transition of lung adenocarcinoma via the JAK2/STAT3 signaling pathway
To determine if STAT3 is involved in STEAP1 regulatary networks, STAT3inhibitor ADZ1480 was using in our experiments. Western blot analysis indicated that the protein expression of p-STAT3 in NC+ADZ1480 was significantly decreased as compared with NC in H1299 cells ( Figure 5A). Cell proliferation in NC+ADZ1480, NC+DMSO, shRNA+DMSO, shRNA+ADZ1480, vector+DMSO, vector+ADZ1480, STEAP1+DMSO, STEAP1+ADZ1480 groups were measured using CCK8 reagent for five consecutive days, results showed that the OD450 of shRNA+ADZ1480 group was significantly lower than NC+DMSO group in H1299 cell, the OD450 of STEAP1+DMSO group was significantly higher than vector+DMSO group in A549 cell ( Figure 5B). The migration assay showed that cell numbers of NC+ADZ1480 group were markedly higher than that of shRNA+ADZ1480 group, cell numbers of shRNA+DMSO group were markedly lower than that of NC+DMSO group in H1299 cell. The invasion assay result was consistent with the migration assay. (Figure 5C,D). Cell numbers of vector+DMSO group were markedly lower than that of STEAP1+DMSO group, cell numbers of vector+ADZ1480 group were markedly lower than that of STEAP1+ADZ1480 group in A549 cell ( Figure 5E,F).

Discussion
In the present study, we concluded that STEAP1 expression was correlated with metastasis and EMT of LUAD. Knocking down of STEAP1 significantly inhibited the proliferation and migration of LUAD cells. In addition, we confirmed that the regulation of STEAP1 in LUAD is associated with by the JAK2/STAT3 signaling pathway. According to relevant literatures, STEAP1 had been studied in several types of cancers, such as breast cancer, prostate cancer, and gastric cancer [11][12][13]. However, our results revealed that STEAP1 was expressed more in H1299 cells and less in A549 cells. The in vitro studies showed that knocking down of STEAP1 inhibited proliferation and migration of the LUAD cells. Thus, STEAP1 plays an crucial role in progression of human LUAD. Furthermore, we found that STEAP1 regulates EMT via JAK2/STAT3 signaling pathway. To our knowledge, this is the first time to report the role of STEAP1 in JAK2/STAT3 signaling pathway in LUAD.
Based on results of the study, STEAP1 can serve as a therapeutic target. Thus, the strategies targeting STEAP1 via JAK2/STAT3 signaling pathway may worth further researches.