TIPE1 Suppresses Growth and Metastasis of Ovarian Cancer

TIPE1, a newly identified member in TIPE (TNFAIP8) family, plays an important role in tumorigenesis and immune regulation, but its role in ovarian cancer, especially in tumor metastasis, remains unknown. In the current study, we aimed to reveal the protein expression spectrum of TIPE1 in normal human tissues and explored its relationship with metastasis in ovarian cancer. The results of IHC staining showed that TIPE1 protein was not only detected in cytoplasm in most human tissues but also expressed in both cytoplasm and nucleus in squamous epithelium and some epithelial-derived cells with secretory functions, such as esophagus, cervix uteri and ovary, and thyroid gland. Moreover, TIPE1 protein was downregulated in ovarian cancer tissues compared with that in the paracancerous. More importantly, TIPE1 suppressed tumorigenesis and metastasis of ovarian cancer in vitro and in vivo, as evidence shows its ability to suppress growth, colony formation, migration, and epithelial-mesenchymal transition (EMT) of ovarian cancer. Taken together, our results demonstrate the suppressor role of TIPE1 in ovarian cancer metastasis, indicating TIPE1 might be a metastasis predictor and a novel therapeutic target for ovarian cancer.


Introduction
Ovarian cancer, the most common gynecological malignancy worldwide, has lower incidence than cervical cancer and uterine body cancer; however the death rate is the highest of all gynecological tumors, which seriously poses a threat to women's lives [1]. Despite making unremitting efforts for improving treatment strategies, involving surgery, chemotherapy, and surgery combined with chemotherapy and target therapy in the past forty years, the survival rate had no obvious improvement [2]. Unfortunately, metastasis and recurrence might occur in more than 80% of patients and the 10-year rate of disease-free survival among patients is below 15% [3]. is is largely due to the abnormal growth and increased cell metastasis potential of ovarian cancer cells. erefore, seeking novel molecular alterations involved in the abnormal cell growth leading to tumor progression would facilitate the realization and development of therapies for ovarian cancer.
TIPE1, a newly identified member of TIPE (tumor necrosis factor-α-induced protein 8) family, was first reported to modulate necroptosis and apoptosis in 2008 [4,5]. Recent studies had suggested that TIPE1 may play a significant role in cancer development and progression. Our previous studies found that TIPE1 was downregulated in hepatocellular carcinoma which could induce caspase-mediated apoptosis and inhibit HCC cell growth both in vitro and in vivo [6]. Similar results were confirmed and presented in lung cancer [7], gastric cancer [8], osteosarcoma [9], breast cancer [10,11], and colorectal cancer [12]. A recent study reported that TIPE1 was decreased in tissue microarrays and inhibited proliferation through inducing apoptosis in ovarian cancer cells [13], but whether TIPE1 related to tumor metastasis was not mentioned. erefore, it is necessary to explore the relationship between TIPE1 and metastasis and to further clarify its anti-tumor effect in ovarian cancer. In addition, TIPE1 seems to act opposite functions in different cancers. Zhao et al. discovered that TIPE1 could promote cervical cancer progression by repression of p53 acetylation [14] and Liu et al. demonstrated that TIPE1 could accelerate NPC cell proliferation by inhibiting autophagy via the AMPK/mTOR signaling pathway [15]. ese findings indicated that the functions of TIPE1 may differ in different tissues. Murine TIPE1 was distributed widely in various mouse tissues [16], but the protein profile of human TIPE1 was still unknown. Hence, it is crucial to clarify the protein expression spectrum of TIPE1 in normal human tissues.
In the current study, we investigated the expression and function of TIPE1 in ovarian cancer. e expression of TIPE in clinical specimens and pathological data indicated that TIPE1 was a suppressor gene in ovarian cancer and it was negatively correlated with cancer metastasis. Furthermore, cell biology assays confirmed that overexpression of TIPE1 could inhibit tumor cell growth and migration. In addition, we presented the expression spectrum of TIPE1 protein in normal human tissues and found that TIPE1 protein was not only detected in cytoplasm of most human tissues, but also detected in nucleus of squamous epithelium and epithelialderived cells, such as esophagus, cervix uteri, ovary, and thyroid gland cells. Our findings revealed the function of TIPE1 in ovarian cancer and suggested that TIPE1 may be a novel therapeutic target for patients.

IHC Staining.
IHC using the anti-TIPE1 antibody (orb186311, Biorbyt Ltd, Cambridgeshire, UK) was performed according to standard protocols as described pre-viously6. Staining intensity of normal human tissues was scored as follows: no staining � ±; weak staining � +; moderate staining � ++; strong staining � +++. To assess the average degree of staining, multiple regions were analyzed, and at least 100 cells were assessed. e histochemistry score of TIPE1 expression in ovarian cancer tissues and noncancer tissues was assessed by computer system [17]. e formula for the H-score is as follows: Histoscore � Σ (I × Pi), where I � intensity of staining and Pi � percentage of stained cells, producing a cytoplasmic score ranging from 0 to 300, the higher the score, the stronger the positive staining.

MTT Assay.
Cell proliferation analysis was performed by a conventional MTT cell viability assay. Cells were seeded in 96-well plates at density of 2000 per well, and 10 μl of the MTT (5 mg/mL in phosphate-buffered saline) was added per well followed by incubation at 37°C in 5% CO2 for 4 hours before the setting time points. Formazan crystals were solubilized with 100 μl of acidified (0.01 M HCl) 10% sodium dodecyl sulfate (SDS) overnight at 37°C. Absorbance at 570 nm was read on a Bio-Rad 680 microplate reader (Bio-Rad Laboratories, CA, USA).

RNA Extraction and Real-Time RT-PCR.
Total RNA was extracted from ovarian cancer cells using Trizol reagent (Invitrogen, Carlsbad, CA, USA) following manufacturer's instructions. After reverse transcription with PrimeScript ™ RT reagent Kit (Takara RR047 A, Tokyo, Japan), the complementary DNA was used for real-time PCR with TB Green ® Premix Ex Taq ™ II (Takara RR820 A, Tokyo, Japan).

Colony Formation Assay.
Cells were seeded in 6-well plate at a density of 200 per well and cultured at 37°C in 5% CO2 for 10-14 days. e plate was then fixed with 4% paraformaldehyde for 30 minutes and stained with 0.1% crystal violet for 1 hour. e plate was washed three times with PBS. e numbers of colonies with more than 50 cells were counted.

Cell Migration Assay.
e 24-well plate and matched Transwell chambers (Corning, China) were used for cell migration assay. 1 * 105 cells in 100 μl serum-free medium were placed into the upper chamber and 700 μl culture medium with 10% FBS was added to the 24 well plate. After incubating for 48h at 37°C in 5% CO2, the chambers were stained with 0.1% crystal violet and photographed.

Wound
Healing Assay. Cells were seeded in a 12-well plate at a density of 1 * 106 per well. After growing to 90-95% confluence in plate and a wound was made by dragging a plastic pipette tip across the cell surface, the wounds were recorded at 37°C in 5% CO2 for 48h.

Western
Blot. ES2 and SKOV3 cells were washed twice with PBS and then lysed with Lysis buffer (Beyotime, Shanghai, China) supplemented with Protease Inhibitor (Beyotime, Shanghai, China), and centrifuged at 4°C for 15 min. A BCA kit ( ermo Fisher Scientific, Waltham, MA, USA) was used to determine the protein concentration. 40 μg quantity of protein was separated on SDS-PAGE and transferred onto PVDF membranes (Millipore, MA, USA). After blocking with 5% nonfat milk for 2h, the membrane was incubated with primary antibodies overnight and washed three times for 10 min with 1×TBST. Next, the membrane was incubated with the corresponding HRP-conjugated secondary antibodies (anti-mouse or anti-rabbit antibodies, Cell Signaling Technology) for 1 h at room temperature. e reaction was visualized using Clarity Western ECL substrate (Bio-Rad, CA, USA) or detected by exposure to a film. e antibodies used included E-Cadherin, N-Cadherin, Slug, Snail, cl-Caspase9, cl-Caspase3 from Cell Signaling Technology; Flag, Caspase3, GAPDH from Proteintech Group.

In Vivo Study.
Female BALB/c mice (4-6 weeks of age) were purchased from Lingchang Biotech Company (Shanghai, China). Animals were fed according to protocols approved by the Institutional Animal Care and Use Committee of Shandong Cancer Hospital and Institute with full respect to the EU Directive 2010/63/EU for animal experimentation. Approximately 1 * 107 Lenti-TIPE1 overexpression SKOV3 cells in 100 μl PBS were transplanted subcutaneously into the 10 mice. e lentivirus vector group was used as control. After reaching 100 mm 3 in size, size of local tumors and mouse body weight were calculated by measuring two perpendicular diameters (length and width) every three days using a caliper, and the volume was calculated according to the following formula: tumor volume (mm3) � 1/2 × (length × square width). Mice were sacrificed after calculating for 19 days and the tumors were isolated and weighed.
2.11. Statistical Analysis. GraphPad Prism 6.02 (GraphPad Software, San Diego, CA, USA) was used for data analysis. Student's t-test or one-way ANOVA was applied to determine significant differences between groups. e statistical correlation between clinical characteristics of ovarian cancer and TIPE1 expression levels in tissue was analyzed by the unpaired t-test or one-way ANOVA test. p < 0.05 was considered statistically significant.

TIPE1 Protein Expression in Normal Human Tissues.
Human TIPE1 protein profile is still unclear till now, so we first detected the protein expression of TIPE1 in normal human tissues, such as digestive system organs ( Figure S1), respiratory, muscular and nervous system organs ( Figure S2), and reproductive system organs ( Figure S3). e IHC results showed that TIPE1 protein was not only detected in the cytoplasm of positively stained cells in most human tissues, but also expressed in both cytoplasm and nucleus of squamous epithelium and some epithelial-derived cells with secretory functions, such as esophagus, cervix uteri, ovary, and thyroid gland.
e IHC data are summarized in Table 1. Considering the strongest staining of TIPE1 protein expressed in both cytoplasm and nucleus of ovary cells, we did the following research.

TIPE1 Was Significantly Decreased in Ovarian Cancer and
Negatively Correlated with Tumor Metastasis. To illuminate the function of TIPE1 in both physiologic and pathologic conditions in ovarian cancer, we firstly detected the mRNA expression of TIPE1 in ovarian cancer tissues and non-cancer tissues from patients who underwent surgery from March 2018 to July 2018 in Shandong Cancer Hospital and Institute. e results showed that TIPE1 mRNA was lower in cancer tissues (Figure 1(a)). To confirm the above results, we collected ovarian tissue sections and purchased tissue microarrays including 281 cases of ovarian cancer tissues and 89 cases of normal or benign ovarian tissues. IHC staining level of TIPE1 in these tissues indicated that TIPE1-positive cells were rarely observed among ovarian cancer but obvious in the ovarian stroma (Figure 1(d)). In addition, mean IHC score in cancer tissues was significantly lower than those in non-cancer tissues (Figure 1(b)). More importantly, mean IHC score displayed that protein levels of TIPE1 decreased significantly in ovarian cancer tissues characterized by tumor metastasis (Figures 1(c) and 1(e)). Statistical analysis of clinical characteristics of these patients also demonstrated that TIPE1 expression was negatively correlated with metastasis in ovarian cancer (Table 2). Besides, we had also analyzed the relationship between TIPE1 expression and patient age, tumor size, histology, differentiation grade, and clinical stage of 281 ovarian cancer patients, but unfortunately, no statistical significance between them was observed.

TIPE1 Inhibited the Migration of Ovarian Cancer Cells.
To further determine the function of TIPE1 on the biological behaviors of ovarian cancer cells, we employed lentivirus-based TIPE1 expressing system to infect ovarian cancer cell lines, and puromycin was added to select the stable expression cells. Metastasis is an important factor leading to death and recurrence of ovarian cancer. Our previous studies suggested that TIPE1 was associated with ovarian cancer metastasis, so we attempted to evaluate the effect of TIPE1 on migration ability in ovarian cancer cells. As shown in Figure 2(a), western blot assay confirmed that the overexpressed TIPE1 lentivirus had reliable and well efficiency. Secondly, overexpressed TIPE1 could increase the protein level of E-Cadherin but decrease the protein level of N-Cadherin, Slug, and Snail, indicating that overexpressed TIPE1 could inhibit the epithelial-mesenchymal transition of ovarian cancer cells (Figure 2(b)). Furthermore, overexpression of TIPE1 in ES2 and SKOV3 cells decreased the closure of wound area compared to the control, and Transwell assays also confirmed that TIPE1 down-regulated migratory abilities of ovarian cancer cells (Figure 2(c)).

TIPE1 Suppressed Ovarian Cancer Cell Proliferation In Vitro.
A recent study reported that TIPE1 could inhibit cancer cells by induing apoptosis, and our studies came to the same conclusion. RT-PCR analysis presented the higher expression of TIPE1 mRNA in TIPE1-infected A2780, ES2, and SKOV3 cells (Figure 3(a)). Both MTT assay and colony formation assays were used to estimate the growth abilities of ovarian cancer cells. Overexpression of TIPE1 dramatically decreased cell growth in A2780, ES2 and SKOV3 cells (Figure 3(b)). Colony formation assay also confirmed the inhibiting function of TIPE1 in these cells (Figure 3(c)). Western blot presented that overexpressed TIPE1 increased the expression of cleaved-caspase3 and cleaved-caspase9 in ES2 and SKOV3 cells (Figure 3(d)). ese results confirmed the inhibition function of TIPE1 in ovarian cancer cell growth and showed a significant anti-tumor activity of TIPE1.  (Figure 4(a)). Tumor growth curve showed that TIPE1 significantly inhibited the growth of SKOV3 xenograft (Figure 4(b)). Consistently, the average tumor weight of lentivirus-TIPE1 group at the time of killing was noticeably less than that of control group (Figure 4(c)). All these data suggested that deficiency of TIPE1 expression was involved in ovarian cancer cells development.

Discussion
TIPE family (tumor necrosis factor-α-induced protein 8) are related to cell death but several members play different roles in tumorigenesis and immune regulation [4,6,[17][18][19]. TIPE [20][21][22] and TIPE3 [18,23,24] had been confirmed to possess a carcinogenic effect in tumor development, while TIPE2 [4,19,25,26] was proved to play opposite roles in cell Ovarian cancer with metastasis Ovarian cancer with no metastasis (e) Figure 1: TIPE1 was significantly decreased in ovarian cancer and negatively correlated with tumor metastasis. (a) Relative mRNA levels of TIPE1 in ovarian cancer tissues (n � 18) and non-cancer tissues (n � 10). (b) Immunohistochemistry score analysis of TIPE1 expression in ovarian cancer tissues (n � 281) and non-cancer tissues (n � 89). (c) Immunohistochemistry score analysis of TIPE1 expression in metastatic ovarian cancer tissues (n � 179) and non-metastatic ovarian cancer tissues (n � 102). (d) Representative photographs of TIPE1 IHC staining in ovarian cancer tissues and ovarian stroma tissues. Scale bar � 50 μm. (e) Representative photographs of TIPE1 IHC staining in metastatic ovarian cancer tissues and non-metastatic ovarian cancer tissues. Scale bar � 50 μm. death and tumorigenesis. Unlike the other members, TIPE1 seemed to play more complicated roles in different tumor progressions. We firstly found that TIPE1 was decreased in hepatocelluar carcinoma, increased HCC cell apoptosis, and inhibited cell growth both [6]. Similar results were shown in lung cancer [7], gastric cancer [8], osteosarcoma [9], breast cancer [10,11], and colorectal cancer [12]. However, some researchers discovered that TIPE1 could promote cell proliferation of cervical cancer and nasopharyngeal carcinoma [15,27], suggesting that TIPE1 may have different effects in different types of tumors. is phenomenon might be related to the TIPE1 protein expression in different human tissues. However, the protein spectrum of human TIPE1 remains unclear.
In this study, we discovered that TIPE1 protein was universally expressed in the cytoplasm of most human tissues. However, the expression of TIPE1 protein was also detected in both cytoplasm and nucleus in squamous epithelium of skin, esophagus, cervix uteri, and epithelial-derived cells such as breast and thyroid gland. Interestingly, the TIPE1 expression was barely detected in lymphoid tissue of the spleen (Figure 3(e)), but had stronger expression in the plasma cells (Figure 1(g), red arrows) of colonic tissues than those cells around the intestinal tissues. Similarly, the dust cells expressed higher TIPE1 staining in cytoplasm than those cells surrounding the lung tissues ( Figure 2(a), right). As we know, plasma cells are derived from B lymphocytes while dust cells originate from macrophages; both of them are activated immune cells. TIPE1 protein is expressed in activated immune cells but not in non-activated immune cells, suggesting that TIPE1 might play a promotional role in immune system. Among all the mentioned human tissues, it should be noted that ovary had the strongest TIPE1 staining, indicating that the function of TIPE1 in ovary might be typical.
Ovarian cancer is a severe gynecological malignancy and it has the highest mortality of all gynecological tumors [1]. Ovarian cancer does not show any obvious symptoms in its early stages; thus most patients are diagnosed at advanced stages accompanied by multiple metastases, limiting the effectiveness of surgery and chemotherapy [14]. ese issues might be explained by the particularity of the tumor microenvironment in ovarian. e tumor microenvironment in ovarian cancer consists of stromal cells and extracellular matrix component, reinforced invasion ability of ovarian cancer cells via the complex cross-talk signaling events [28,29]. erefore, exploring novel genetic sites would be helpful to understand the tumor development and metastasis mechanism of ovarian cancer.
It was reported that TIPE1 was decreased in cancer tissues and inhibited ovarian cancer cell growth [13], but the relationship of TIPE1 and metastasis was not mentioned. In the current study, we proved that TIPE1 was negatively correlated with metastasis and inhibited the migration ability of cells in ovarian cancer. Moreover, overexpressed TIPE1 increased the protein of E-Cadherin but suppressed N-Cadherin, Slug, and Snail expression, which suggested that TIPE1 could inhibit the EMT of ovarian cancer cells.
ese findings would help us to understand the complex metastatic ovarian tumor associated microenvironment and provide new insights for the development of tumor targeted agents [29].