Overexpression of NPTX2 Promotes Malignant Phenotypeof Epithelial Ovarian Carcinoma via IL6-JAK2/STAT3 Signaling Pathway Under Hypoxia


 Background: Epithelial ovarian cancer(EOC) is the main subtype of ovarian cancer and shows an aggressive phenotype and poor prognosis. Neuronal pentraxin II (NPTX2)is a member of the neuronal pentraxin family and plays a contradictory role in different tumors. However, there has been no report about the possible role and effect of NPTX2 in EOC.Methods: Bioinformatics analysis, qPCR, western blotting and immunohistochemistry were used to detect the expression of NPTX2 in EOC. Lentivirus-based transfection for NPTX2 overexpression or knockdown was performed on the EOC cell lines A2780, HEY, SKOV3 and OVCAR-3. The effect of NPTX2 on the malignant phenotype of EOC was examined through methods of MTS assay, Edu assay, transwell assay, western blotting analysis, qPCR analysis, luciferase reporter assay and xenograft experiment.Results: EOC tissues showed higher NPTX2 expression than the normal tissues with poor prognosis. NPTX2 overexpression can promote the proliferation, invasion, migration and tumorigenesis of EOC via IL6-JAK2/STAT3 signaling pathway. Moreover, hypoxia-inducible factor-1(HIF-1) can promote the transcription and expression of NPTX2 under the hypoxic environment.Conclusions: The above results suggest that NPTX2 may play a novel role in ovarian cancer's malignant phenotype and act asa promising treatment target for EOC moleculartherapy.


Introduction
Ovarian cancer is one of the most common malignant tumors in females with the worst prognosis, accounting for 3.4% of all the female malignant tumors and 4.4% of female cancer deaths [1]. Mostly, epithelial ovarian cancer(EOC) is the main subtype of ovarian cancer and shows a more aggressive phenotype than nonepithelial cancers [2]. Although surgery combined with chemotherapy has achieved a certain effect, the patients' long-term prognosis is not ideal. It has been reported that the 5-year survival rate is only 29% in the advanced stage of EOC patients [3]. At present, molecular targeted therapy is regarded as one of the most promising therapies for EOC treatment [4]. Identifying genes associated with tumorigenesis, proliferation, invasion and migration of EOC has signi cant implications for EOC clinical treatment and basic research [5].
Neuronal pentraxin II (NPTX2), also known as neuronal activity-regulated pentraxin, is a member of the neuronal pentraxin family [6]. NPTX2 was initially expressed in the central nervous system (CNS) and played a role in promoting excitatory synapse formation and synaptic remodeling [6,7]. NPTX2 is highly upregulated and participates in developing several CNS diseases, such as Parkinson's disease [8].
Moreover, several studies have also reported that NPTX2 is involved in the occurrence and development of malignant tumors, while it plays a contradictory role in different tumors. For example, NPTX2 promotes colorectal cancer growth and liver metastasis by activating the canonical Wnt/β-catenin pathway via FZD6 [9]. NPTX2 overexpression promotes the cell proliferation, migration and invasion of renal cell carcinoma [10]. However, NPTX2 was downregulated and correlated with the poor prognosis of pancreatic cancer and glioma due to its promoter hypermethylation [11,12]. Although NPTX2 was found to be expressed in female reproductive organs, such as the ovary, cervix and uterine, according to the human protein atlas [13], there has been no report about the possible role and effect of NPTX2 in EOC.
In the present study, we rst found abnormally higher NPTX2 expression in EOC using bioinformatics analyses based on the TCGA database and four GEO datasets (GSE10971, GSE18520, GSE105437 and GSE26712), followed with clinical specimens' validation. Higher NPTX2 expression was correlated with the poor prognosis of EOC. Furtherly, Gene set enrichment analysis (GSEA) and both in vivo and in vitro experiments demonstrated that NPTX2 could promote the malignant phenotype of EOC via IL6-JAK2/STAT3 signaling pathway. Moreover, we found that hypoxia-inducible factor-1(HIF-1) can promote the transcription and expression of NPTX2 under the hypoxic microenvironment. Therefore, NPTX2 is a novel oncogene in the malignant of EOC.

Patients and samples
In the present study, patient samples and adjacent non-tumor tissues were taken from 45 patients diagnosed with EOC via pathology and underwent surgical excision without prior anticancer treatments at Fudan University Shanghai Cancer Center between February 2018 to May 2019. This study was approved by the institutional review board of the Fudan University Shanghai Cancer Center.

Cell treatment
The human EOC cell lines OVCA420, A2780 and OVCAR3 were purchased from the Chinese Academy of Sciences cell bank (Shanghai, China). The wild-type human ovarian epithelial cell line HS832.Tc and human EOC cell lines SKOV3, HEY were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). All cells were maintained in Dulbecco's modi ed Eagle's medium (DMEM, HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Carlsbad, CA, USA) and 1% penicillin/streptomycin (Gibco) at 37 °C with 5% CO 2 . To induce hypoxia, cells were cultured in a hypoxia chamber with 94% N 2 , 5% CO 2 , and 1% O 2 at 37 °C. IL-6-neutralizing antibody (R&D Systems, Minneapolis, MN, USA) was used at a concentration of 0.5 µg/ ml [14]. All cells analyzed were cultured with less than 20 generations.

Western blotting
The total protein of EOC tissues and cell lines was isolated via a cell protein extraction kit (KeyGen Biotechnology, Nanjing, PR China) according to the manufacturer's instructions. The BCA kit (Beyotime Biotechnology, Beijing, PR China) was used to determine the concentration. An equivalent amount of protein from each sample was separated by 4 to 20% SDS-PAGE (Genscript, Nanjing, China), transferred to a nitrocellulose membrane and blocked with 2% bovine serum albumin (KeyGen Biotechnology). The membrane was then incubated withprimary antibodies against NPTX2, IL6, p-JAK2, JAK2, p-STAT3, STAT3 and β-actin (Abcam Technology, Cambridge, UK) at 4 °C overnight, followed with TBST washing and incubated with secondary antibody. The ECL kit (Beyotime) was used to detect the bands on each membrane, and IMAGE J software (National Institutes of Health, Bethesda, MD, USA) was used for quanti cation.

Cell viability assay
According to the manufacturer's instructions, the CellTiter 96® AQueous Non-Radioactive cell proliferation assay kit (Promega, Madison, WI, USA) was used for cell viability detection. Brie y, EOC cells were cultured in 96-well plates at a density of 3 × 10 3 cells / well for 24, 48, 72, 96 and 120 h. Then 20 µl of MTS was added into each well, followed with 1 h incubation at 37℃. An ultraviolet spectrophotometer (Thermo Fisher Scienti c, Waltham, MA, USA) was used to detect the absorbance at 495 nm.

Edu assay
The 5-ethynyl-20-deoxyuridine (Edu) incorporation assay (Beyotime) was performed according to the manufacturer's instructions. EOC cells were seeded into 24-well plates at a density of 1 × 105 cells per well for 20 h, followed with 50 µMEdU treatment for 2 h at 37 ℃. The cells were then xed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100, incubated with 100 µl Click Additive Solution for 30 min, and stained with 100 µl DAPI. The images were taken with a laser scanning confocal microscope (Olympus), and the percentage of Edu-positive cells was calculated.

Transwell assay
For the transwell assay, the 8 µm pore size polycarbonate membrane of the upper transwell chamber (Costar, Corning, NY, USA) was covered with 80 µL of 50 ng/µLMatrigel solution (BD, Franklin Lakes, NJ, USA) for 1 h incubation at 37℃. Then the EOC cells were resuspended in serum-free DMEM at a density of 2 × 10 5 cells/mL, and then 100 µL of cell suspension was seeded into the upper chamber, and 600 µL DMEM with 20% FBS was added to the lower chamber. After incubation at 37 °C for 20 h, the cells were xed with 4% paraformaldehyde (Beyotime) for 10 min and stained with 1% Crystal Violet solution (Beyotime) for 20 min at room temperature. Finally, the cell numbers were counted by calculating the average of ve random elds using an inverted microscope (Olympus). For the migration assay, the upper transwell chamber was not treated with Matrigel solution, and the other steps were the same as the transwell assay.

Immunohistochemistry (IHC)
Immunohistochemistry was performed using an immunohistochemistry kit (Maixin Biotechnology, Fuzhou, Fujian, PR China) according to the manufacturer's instructions. Brie y, tissue samples were para n-embedded, cut into 4-µm sections, and incubated with primary antibodies against NPTX2 and Ki-67 (Abcam) at 4 °C overnight. After DAB staining, the sections were imaged with an optical microscope (Olympus), and staining intensity was assessed according to the German immunohistochemical scoring system [15].

Luciferase activity analysis
Luciferase reporter assays were performed using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions. Brie y, NPTX2 reporter plasmids were constructed by Gene-Chem (Shanghai, PR China). EOC cells were seeded into 96-well plates at a density of 3 × 10 3 cells per well and transfected with different plasmids. The cells were lysed after 48 h, and luciferase activity was measured using an ultraviolet spectrophotometer (Thermo Fisher Scienti c). Each experiment was independently repeated three times.

Chromatin immunoprecipitation (ChIP) assays
ChIP assays were performed using the EZ-ChIP ™ Immunoprecipitation Kit (Millipore, Billerica, MA, USA) according to the manufacturer's instructions. The chromatin complexes were immuno-precipitated with anti-HIF-1 antibody (Abcam).Then the ChIP Assay Kit (Beyotime) was used to purify DNA samples, and the puri ed DNA was analyzed by qPCR. The primer pairs used to amplify the HIF-1 binding site in the NPTX2promoter was: f: 5′-GTGACACTGCGGGCCCT-3′ and r: 5′-CGTCCGGCGCCTGGTCCT-3′.

Xenograft experiments
Five-week-old female BALB/c nude mice were purchased from Shanghai Laboratory Animal Center and bred in the Laboratory Animal Center of Fudan University Shanghai Cancer Center under speci cpathogen-free conditions. Mice were maintained under a 12 h light/12 h dark cycle with free access to water and standard mouse diet. EOC cell suspensions (5 × 10 5 cells/100 µL of DMEM) were mixed with 100 µL of Matrigel solution (BD) and injected into the back anks of the nude mouse. The tumor size was measured using a Vernier caliper once a week, and the tumor volume was calculated using the following formula: V = (D × d )/2 mm, where D is the longest diameter and d is the shortest diameter of the tumor. The mice were sacri ced 5 weeks after implantation, and the tumors were weighed and photographed.

Bioinformatic analyses
The data on mRNA expression and clinical samples from ovarian carcinoma patients were obtained from The Cancer Genome Atlas (TCGA, http://cancergenome.nih.gov) and Gene Expression Omnibus (GEO, GSE10971, GSE18520, GSE105437 and GSE26712). Gene-set enrichment analysis (GSEA, http://www.broadinstitute.org/gsea/index.jsp) was used to detect any possible signal pathway sets of genes showing statistically signi cant differences between higher and lower NPTX2 expression groups.

Statistical analysis
Statistical analysis was performed using SPSS 23.0 software (IBM, Armonk, N.Y, USA). All experiments were repeated at least three times, and the results are presented as the mean ± SD. The chi-square test,two-tailed Student's t-testand one-way analysis of variance were used to compare the statistical signi cance among different groups. Differences in survival rates were analyzed using the log-rank test and Kaplan-Meier analysis. All statistical tests were two-sided, and statistical signi cance was de ned as a p-value < 0.05.

Results
NPTX2 is upregulated inepithelial ovarian carcinoma and correlated with the progression and poor prognosis.
To nd the oncogenes that may participate in the carcinogenesis and development of ovarian carcinoma, we rst searched TCGA and four GEO databases, including GSE10971, GSE18520, GSE105437 and GSE26712, and found that 14 genes were upregulated in these ve databases (Fig. 1a). NPTX2 is the most signi cantly upregulated gene among these genes, while its biological role in ovarian cancer has not been reported (Fig. 1b, c). In addition, we performed Kaplan-Meier survival analysis for the prognostic signi cance of NPTX2 expression in ovarian carcinoma patients, and the average survival time of patients with high NPTX2 expression was signi cantly lower in the TCGA dataset than in patients with low expression (Fig. 1d). We then detected clinical specimens from ovarian carcinoma patients and con rmed the ndings obtained by bioinformatics analysis. All qPCR (Fig. 1e), western blotting (Fig. 1f) and immunohistochemistry (Fig. 1g) showed NPTX2 expressed higher in epithelial ovarian carcinoma than the normal tissues. Moreover, we also detected NPTX2 expression in EOC cell lines OVCA420, A2780, SKOV3, HEY and OVCAR-3 and wild-type human ovarian epithelial cell line HS832.Tc. All EOC cell lines with higher NPTX2 expression than the HS832.Tc (Fig. 2a). The above results NPTX2 is upregulated in ovarian carcinoma and correlated with the progression and poor prognosis.
NPTX2 overexpression promoted the malignant phenotype of epithelial ovarian carcinoma in vitro.
To investigate the role of NPTX2 in EOC, we designed lentiviral-encapsulated NPTX2 overexpression plasmids. According to the expression of NPTX2 in EOC cell lines shown in Fig. 2a, A2780 and HEY with the lowest expression were treated for NPTX2 overexpression. Both western blotting and qPCR con rmed the overexpression of NPTX2 (Fig. 2b, c). Then MTS assays showed the absorbance values were signi cantly higher than controls (Fig. 2d, e), con rming that NPTX2 overexpression promoted EOC cell lines' cell viability. Edu assays showed the Edu positive rates were signi cantly higher than controls (Fig. 2f), con rming that NPTX2 overexpression promoted EOC cell lines' proliferation. Both transwelland migration assays showed that NPTX2 overexpression markedly promoted tumor cell invasion and migration compared with controls (Fig. 2g, h). The above results con rm that NPTX2 overexpression promoted the malignant phenotype of epithelial ovarian carcinoma in vitro.
NPTX2 knockdown inhibited malignant phenotype of epithelial ovarian carcinoma in vitro.
To further demonstrate the role of NPTX2 in the proliferation and metastasis of EOC, we designed two siRNA sequences and transfected OVCAR-3 and SKOV3 to knockdown NPTX2. Western blotting and qPCR were used to con rm the e ciency of NPTX2 knockdown (Fig. 3a, b). First, we used the MTS assay to detect tumor cells' proliferation activity after NPTX2 knockdown, and the absorbance values were signi cantly lower than those of the control group (Fig. 3c, d), con rming that NPTX2 knockdown inhibits the cell viability of EOC cell lines. Edu assays showed the Edu positive rates were signi cantly lower than controls (Fig. 3e), con rming that NPTX2 knockdown inhibits EOC cell lines' proliferation. Subsequently, we examined the effect of NPTX2 knockdown on tumor cell metastasis by both transwell and migration assays. As shown in Fig. 3f, g, NPTX2 knockdown markedly inhibited tumor cell invasion and migration compared with controls. Taken together, these results clearly demonstrated that NPTX2 knockdown inhibited the malignant phenotype of epithelial ovarian carcinoma in vitro.
To further explore the possible signaling pathways in which NPTX2 regulates the proliferation, invasion and migration of EOC, we performed GSEA on the TCGA dataset. The results showed a signi cant IL6-JAK2/STAT3 signaling pathway enrichment in the higher NPTX2 expression group (Fig. 4a). Therefore, we rst examined changes of IL6 after NPTX2 regulation. All the ELISA, qPCR, and western blotting results showed the expression and secretion of IL6 were obviously down-regulated after NPTX2 knockdown in OVCAR-3 and SKOV3 (Fig. 4b, c, f), while the opposite results were got after NPTX2 overexpression in A2780 and HEY (Fig. 4d, e, g). Then the downstream molecules of the IL6-JAK2/STAT3 signaling pathway were detected by western blotting. As shown in Fig. 4f, g, the expression levels of the p-JAK2 and p-STAT3 were signi cantly downregulated after NPTX2 knockdown in OVCAR-3 and SKOV3 (Fig. 4f), while the opposite results were obtained after NPTX2 overexpression in A2780 and HEY (Fig. 4g). The above evidence proves that NPTX2 can upregulate IL6 expression and activate the JAK2/STAT3 signaling pathway inepithelial ovarian carcinoma.
To furtherly verify whether NPTX2 promotes malignant progression in EOC through the IL6-JAK2/STAT3 signaling pathway, we treated EOC cell lines A2780 and HEY after NPTX2 overexpression with an IL-6neutralizing antibody. MTS assays showed higher absorbance values caused by NPTX2 overexpression were reversed after IL-6-neutralizing antibodytreatment (Fig. 5a, b). Edu assays showed a higher Edu positive rate during NPTX2 overexpression were reversed after IL-6-neutralizing antibodytreatment (Fig. 5c). Both transwell and migration assays showed that NPTX2 overexpression markedly promoted tumor cell invasion and migration, while these promoting effects were also reversed after IL-6-neutralizing antibodytreatment (Fig. 5d, e). The above results con rm that NPTX2 overexpression promoted the malignant phenotype of epithelial ovarian carcinoma via IL6-JAK2/STAT3 signaling pathway.

HIF-1 can directly induce the expression of NPTX2 under hypoxia
We furtherly investigated the possible upstream mechanism of NPTX2 overexpression in EOC. GSEA analysis based on the TCGA dataset revealed signi cant hypoxia enrichment at a higher NPTX2 expression group (Fig. 6a). Therefore, we investigated whether there is a functional relationship between NPTX2 and hypoxia. The A2780 and HEY cells were treated under hypoxia, and the expression of NPTX2 increased gradually with the extension of the hypoxia treatment time according to both qPCR and western blotting (Fig. 6b, c, d). Since hypoxia-inducible factor-1 (HIF-1) is the most important transcription factor under hypoxic conditions (Fig. 6e), we furtherly discussed whether there is a transcriptional regulation relationship between HIF-1 and NPTX2. We identi ed the possible binding sites of HIF-1 on the promoter region of NPTX2 by JASPAR analysis (Fig. 6f), and the luciferase reporter gene assays were performed. The results showed that NPTX2-wt transfected A2780 and HEY had signi cantly enhanced luciferase activity under hypoxia (Fig. 6g, h). Moreover, ChIP assays revealed NPTX2 enrichment in A2780 and HEY after anti-HIF-1 treatment under hypoxia (Fig. 6i). Besides, qPCR and western blotting further con rmed that NPTX2 expression was remarkably upregulated after HIF-1 overexpression (Fig. 6j, k). In summary, we conclude that HIF-1 directly transcriptionally regulates NPTX2 expression in EOC under hypoxic conditions.

Nptx2 Regulates Epithelial Ovarian Carcinoma Tumorigenesis
Finally, to determine whether NPTX2 can regulate EOC tumorigenesis in vivo, we injected EOC cell lines with NPTX2 overexpression, knockdown or control into the ank regions of nude mice, observed the tumor growths and followed the survival rates. At 5 weeks post-inoculation, the mean volume of the xenograft tumors in the NPTX2 overexpression group was obviously larger than that in the control group (Fig. 7a, b), while the opposite results were obtained after NPTX2 knockdown (Fig. 7d, e). In agreement, the mean weight of the tumors extracted from the NPTX2 overexpression group was obviously larger than the control group (Fig. 7c), while themean weight of the tumors extracted from the NPTX2 knockdown group was smaller than the control group (Fig. 7f). Moreover, IHC was performed to detect the effects of NPTX2 overexpression or knockdown on tumor tissues. In the NPTX2 overexpression group, the staining intensity and expression levels of Ki-67 were allupregulated, while the opposite results were obtained in the NPTX2knockdown group (Fig. 7g). A schematic diagram showing that NPTX2 overexpression promotes malignant phenotype of epithelial ovarian carcinoma via IL6-JAK2/STAT3 signaling pathway under hypoxia (Fig. 7h). Taken together, these results suggested that NPTX2 regulates epithelial ovarian carcinoma tumorigenesis in nude mice.

Discussion
In the present study, we found for the rst time that NPTX2 was signi cantly upregulated in epithelial ovarian cancer and correlated with the poor prognosis. Since NPTX2 was the top regulated genes among TCGA and four GEO databases (GSE10971, GSE18520, GSE105437 and GSE26712), it is worth discussing whether NPTX2 playa vital role in EOC. We furtherly con rmed the obvious overexpression of NPTX2 in our clinical human EOC tissues and cultured cell lines.
In order to nd the possible role of NPTX2 on EOC, we rst searched the available literature about the roles of NPTX2 in malignant tumors and NPTX2 was reported mainly participated in ve cancers, including glioma, pancreaticobiliary cancer, colorectal cancer, renal cell carcinoma and neuroblastoma [16][17][18][19][20][21]. However, NPXT2 plays a contradictory role in those tumors. Although NPTX2 showed a lower expression due to its promoter hypermethylation and identi ed poor prognosis in glioma and pancreaticobiliary cancer [12,22], only one study demonstrated the direct effect of NPTX2 on the inhibiting proliferation and invasion of pancreatic cancer cells [11]. As for colorectal cancer, renal cell carcinoma and neuroblastoma, all studies demonstrate the direct carcinogenesis effects of NPTX2. NPTX2 promotes colorectal cancer growth and liver metastasis by activating the canonical Wnt/β-catenin pathway via FZD6 [9]. MiR-96 suppresses the proliferation, migration, and invasion ofrenal cell carcinoma via inhibiting NPTX2 [10]. Targeting NPTX2 with the selected peptide can reduce the tumor burden in orthotopic mouse models of human neuroblastoma [20].
Therefore, we prefer to recognize NPTX2 as an possible oncogene in tumors. Our study also con rmed that NPTX2 overexpression could promote the proliferation, invasion and migration of EOC in vitro and in vivo, while NPTX2 knockdown can inhibit those malignant phenotypes of EOC.
To further explore the possible downstream mechanism of NPTX2 on the malignant phenotype of EOC, we performed GSEA on the TCGA dataset, and the results showed a signi cant IL6-JAK2/STAT3 signaling pathway enrichment in the higher NPTX2 expression group. IL6-JAK2/STAT3 signaling pathway is one of the most typical oncogenic signaling pathways in cancers. For example, NFAT1regulated IL6 signaling contributes to the aggressive phenotypes of glioma [23]. Cancer-associated broblasts promote gallbladder cancer growth via activation of the IL6-JAK/STAT3 signal pathway [24]. Ovarian carcinoma-associated mesenchymal stem cells can activate the tumor cell stemness via activating IL6/STAT3 signaling [25]. In our study, we demonstrated that NPTX2 overexpression could promote the expression and secretion of IL6, followed by JAK2/STAT3 signaling pathway activation in epithelial EOC. Moreover, after treatment with IL-6-neutralizing antibody, NPTX2 induced malignant phenotype of EOC was reversed. Therefore, these results demonstrate that NPTX2 promotes EOC's malignant phenotype via upregulation of IL6 and the activation of the JAK2/STAT3 signaling pathway.
Due to the rapid growth of tumor cells, it is frequent to occur in ischemia and hypoxia [26]. Hypoxia can promotethe proliferation, invasion and migration of tumors through upregulation of HIF-1 [27]. HIF-1 is the key transcription factor under hypoxia and is involved in the malignant phenotype of ovary cancers [28]. For example, HIF-1 mediates epidermal growth factor-induced down-regulation of E-cadherin expression and cell invasion in human ovarian cancer cells [29]. In our study, GSEA analysis indicated that NPTX2 might be correlated with a hypoxia environment. Then we inferred whether HIF1 could transcriptionally regulate the expression of NPTX2 under hypoxia. Both the luciferase reporter gene assays and ChIP assays demonstrated that HIF-1 directly transcriptionally regulates NPTX2 expression in EOC under hypoxic conditions.
In conclusion, we found for the rst time that NPTX2 was signi cantly upregulated in EOC tissues and correlated with the poor prognosis. NPTX2 overexpression can promote the proliferation and tumorigenesis of EOC via IL6-JAK2/STAT3 signaling pathway. Moreover, the expression of NPTX2 may be upregulated via HIF-1 due to the hypoxia environment caused by EOC's rapid growth. Therefore, NPTX2 may be a promising treatment target for EOC molecular therapy.

Declarations
Ethics approval and consent to participate All human materials were obtained with informed consent, and the study was approved by the Ethical Review Committee of the World Health Organization of the Collaborating Center for Research in Human Production and authorized by the Human Ethics Committee of FUSCC.

Consent for publication
Not applicable.

Competing interests
The authors declare no con icts of interest. Authors' contributions XH participated in the design of the study. XH, XL and LX performed the experiments and collected the data. YL performed the bioinformatics analysis and analyzed the data. HW and ZF produced the main draft of the text and the gures. XH and XJ contributed to data collection and the interpretation of data. XH wrote the manuscript. XC and XW conceived the project and supervised the research. All authors read and approved the manuscript.  Figure 2 NPTX2 overexpression promoted ovarian carcinoma proliferation in vitro. a: Western blottingshowing theexpressionof NPTX2 in EOC cell lines and wild-type human ovarian epithelial cell line. b, c:Lentiviralbased transfection and the effects on NPTX2 overexpression were validated by western blotting (b) andqPCR (c). d, e: Thecell viability of A2780 (d) and HEY (e) increased after NPTX2 overexpressionas measured by the MTS assay. f: The proliferation of the A2780 and HEY increased after NPTX2 overexpression as measured by the EDU incorporation assay, scale bar=100 μm. g, h: Representative transwell assay(g)and migration assay(h)showing the increase in invasion and migration of A2780 and HEY after NPTX2 overexpression. Scale bar=100 μm. All data are shown as the mean ± SD (three independent experiments). * P < 0.05; * * P < 0.01; * * * P < 0.001.