Journal of Neuro-Oncology https://doi.org/10.1007/s11060-018-2818-4

LABORATORY INVESTIGATION

AnnexinA5 promote glioma cell invasion and migration via the PI3K/ Akt/NF-κB signaling pathway Chenxing Ji1 · Hua Guo1 · Pei Zhang1 · Wei Kuang1 · Yanghua Fan1 · Lei Wu1  Received: 4 October 2017 / Accepted: 23 February 2018 © Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract As an important member of the Annexins, AnnexinA5 has been attributed important functions in trophoblast membrane repair, anticoagulation and cellular signal transduction. Accumulated studies show that AnnexinA5 is closely associated with various types of carcinomas. However, the potential contribution of AnnexinA5 to glioma cancer progression remains unclear. In this study, we report that AnnexinA5 is significantly upregulated in both high-grade glioma samples and glioma cell lines. Moreover, overexpression of AnnexinA5 promotes cell migration and invasion in vitro and tumorigenicity of glioma cells in nude mice, while knockdown of AnnexinA5 manifests a repressive function during these cellular processes. Importantly, mechanistic studies further reveal that AnnexinA5 is an essential transcriptional target of Snail via activating the PI3K/Akt/NF-κB signaling pathway. Taken together, these findings suggest that AnnexinA5 or the PI3K/Akt/NF-κB pathway may be promising therapeutic molecules to eradicate glioma metastases. Keywords  Glioma · AnnexinA5 · PI3K/Akt/NF-Κb · Progression · Signaling pathway

Introduction According to the World Health Organization (WHO) announced the order of ranking in central nervous system (CNS) tumors, glioma accounts for the most common form of primary brain tumors in the adult [1, 2]. Despite great progress has been made in multimodal treatment of glioma including radiotherapy and chemotherapy following surgical resection, the overall survival of glioma remains very poor [3–6]. Therefore, mounting efforts to understand the molecules and signal pathways involved in the occurrence and development of glioma is considered promising required [7]. Previous studies have showed that deregulation of important genes contributes to malignant outcome in Chenxing Ji, Hua Guo and Pei Zhang contributed equally to this work. Electronic supplementary material  The online version of this article (https​://doi.org/10.1007/s1106​0-018-2818-4) contains supplementary material, which is available to authorized users. * Lei Wu [email protected] 1



Department of Neurosurgery, The Second Affiliated Hospital of Nanchang University, Nanchang 330006, China

adults. Annexins are a superfamily of calcium dependent phospholipid-binding proteins superfamily that have been showed to be involved in various cellular and physiological processes. As an important member of the Annexins, AnnexinA5 is an ubiquitous and widely-expressed intracellular protein in human implicated in numerous membranerelated events along exocytotic and endocytotic pathways with a molecular mass of about 35.7 kDa [8]. AnnexinA5 has been attributed important functions such as trophoblast membrane repair, anticoagulation and cellular signal transduction. Recently, accumulated studies showed that AnnexinA5 is closely associated with various types of carcinomas, including lung cancer, colon cancer and prostate cancer [9–11]. Multivariate analysis by Martin.et al and colleagues reported for the first time that annexin A2 and A5 act as significant ligands for C1q could increase cell apoptosis of systemic lupus erythematosus [12]. In addition, Li et al. revealed that AnnexinA5 was specifically overexpressed in uterine cervical squamous cell carcinomas (UCSCCs), and is significantly associated with the differentiation of UCSCC in patients with an increased serum concentration [13]. There is little knowledge regarding the association between AnnexinA5 and glioma. The data obtained from isobaric peptide tagging chemistry (iTRAQ)-based proteomics profiling revealed that the

13

Vol.:(0123456789)



expression of AnnexinA5 was significantly increased in glioma brain tissues compared with that in normal brain tissues, indicating the potential role of AnnexinA5 as diagnosis biomarker for malignant gliomas [14]. However, the functional effect and mechanism of AnnexinA5 on biology behavior of human glioma cells remains elusive. In the study, AnnexinA5 expression was detected in gliomas specimens of different grades and analyzed its correlation with different clinic-pathological features in glioma tissues. Further, in vitro and in vivo experiments were performed to identify the functional role of AnnexinA5 in migration and invasion of glioma cells, and the underlying mechanisms concerning the functionality of AnnexinA5 in gliomas. Thus, our results will give a new direction utilizing AnnexinA5 as a new promising candidates for inhibiting glioma metastases.

Materials and methods Patient tissues Human glioma specimens were obtained with written consent from glioma patients undergoing underwent surgery at the Department of Neurosurgery in The Second Affiliated Hospital of Nanchang University Samples were collected and immediately preserved in liquid nitrogen after surgical resection. The grade of Glioma specimens were determined pathologically and were further divided into low grade (grade I–II) and high grade (grade III–IV) according to the 2016 World Health Organization (WHO) classification of tumors of the central nervous system. In addition, 57% IDH1 mutations by IDH sequencing are found in some pathological type of adult patients. The clinicopathological features of patients has been summarized in Supplementary Table 1. Written informed consent was taken from every individual enrolled in the current study. This study was performed with the approval of the Institutional Review Board at The Second Affiliated Hospital of Nanchang University. All experiments were carried out in accordance with relevant guidelines and regulations.

Cell culture Human glioma cell lines (U87 and U118) were purchased from American Type Culture Collection (Shanghai, China). The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 10% FBS (Gibco) and 1% penicillin/streptomycin in a 5% ­CO2 and humidified atmosphere at 37 °C in incubator.

13

Journal of Neuro-Oncology

RNA extraction and real‑time PCR analyses Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) as described by the manufacturer’s instructions. Reverse transcription reaction was performed using a Reverse Transcription Kit (Takara). Real time PCR was carried out by ABI 7500 real-time PCR system (Applied Biosystems, Foster City, CA, USA) to quantify AnnexinA5 mRNA expression, and GAPDH was used as internal reference.

Western blotting assay Total proteins were extracted from cell or tissues using RIPA buffer (Beyotime Institute of Biotechnology, Jiangsu, China) containing protease/phosphatase inhibitor. Proteins were loaded onto 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) for electrophoresis and then transferred to the PVDF membranes. Prior to probe with the specific primary antibodies (anti-GAPDH antibody diluted at 1:2000; antiAnnexinA5, anti-phosphorylated PI3K, phosphorylated Akt antibodies diluted at 1:500; anti-p65, anti-MMP-2 and antiSnail antibodies diluted at 1:1000, anti-MMP-9 diluted at 1:1000) at 4 °C overnight, the membrane was blocked in 5% non-fat dry milk. The resulting immunocomplexes were visualized by ECL detection solution (Santa Cruz Biotechnology, USA). GAPDH was chosen as an internal control.

Overexpression and knockdown of AnnexinA5 The AnnexinA5 cDNA sequence was synthesized according to the full-length AnnexinA5 sequence and then sub cloned into a pCDNA 3.1 vector (Invitrogen, Shanghai, China). For the transfection, 1 × 106 glioma cells co-transfected with pCDNA-AnnexinA5 and pcDNA3.1-luciferase using Lipofectamine 2000 reagent (Invitrogen, USA) following the product guidelines. Stable transfectants were subsequently selected with a final concentrations 1 mg/ml of G418. For the knockdown of AnnexinA5, lentivirus containing AnnexinA5-specific shRNAs (shRNA1# and shRNA2#), respectively, were obtained from GenePharma (Shanghai, China) and transduced to glioma cells with lentiviruses expressing luciferase. The efficiency of AnnexinA5 overexpression or knockdown was screened through the real-time PCR and western blot analysis.

Cell invasion and migration assay Cell invasion and migration assay was performed as previous study described [15]. stable transfectants (5 × 105) with overexpression and knockdown of AnnexinA5 were

Journal of Neuro-Oncology

cultured in the upper chamber with a matrigel-coated membrane (BD Biosciences) or on the top side of membrane in serum-free 1640 medium (500 µl). Also, the lower chambers were added with culture medium (500 µl) containing 10% FBS. After incubating 48 h, the cells on the lower chambers was fixed with 4% formaldehyde, stained with Giemsa, and photographed. The ability of invasion and migration was determined by the number of cells passing through the membrane. We further performed the wound healing assay to determine the effect of AnnexinA5 expression on migration. The cells of overexpression and knockdown of AnnexinA5 were plated in 6-well plates to form a monolayer, respectively. On the following day, a sterile micropipette tip was used to create a straight scratch in a cell monolayer. The accumulated distance travelled by the cells between both wound edges on average over 24 h can be measured to qualify the cell motility.

Mouse model Forty-eight BALB/c-A nude mice (n = 8 for each treatment group) were purchased from the Animal Center of Chinese Academy of Medical Sciences, Shanghai. Animal research by procedures was approved by the Animal Research Committee of Nanchang University. The methods were performed in accordance with the relevant guidelines and regulations that sterilized water and feed were provided ad libitum. To setup intracranial gliomas models, 1 × 106 U87 cells with overexpression and knockdown of AnnexinA5 were implanted stereotactically in the brains of nude mice. Intracranial tumor growth on days 10, 15 and 20 after implantation was detected by Bioluminescence imaging (Caliper IVIS Spectrum, USA).

Immunohistochemistry analysis We performed an immunohistochemical (IHC) assay in the brain tissues slices resected from the orthotopic Mice staining with hematoxylin and eosin. The expression of components of the extracellular matrix MMPs (MMP-2) was examined with the appropriate primary antibodies, and the brief procedure was performed as previous study described [16]. The image of representative immunohistochemical staining was showed using a microscope.

Dual‑luciferase reporter assay The TOPflash/FOPflash system vector (Promega, Madison, WI, USA) were used to test the effect of AnnexinA5 on Snail transcriptional activity. 1 × 106 stable transfectants with overexpression and knockdown of AnnexinA5 were treated with 1 µg reporter plasmid TOPflash with − 194 to − 78 bp genomic sequence upstream of the Snail promoter

and control plasmid FOPflash with or without 20  pmol PI3K or Akt siRNA using Lipofectamine 2000 reagent (Invitrogen, USA) following the product guidelines. pRLTK Renilla luciferase vector was used for normalization of luciferase activity. After 48 h, cells were lysed and luciferase activity was measured with the Dual-Luciferase Reporter Assay System (Promega).

ChIP assay ChIP assay was performed using the ChIP assay kit (Millipore, Billerica, MA) following the protocol as previously described [17]. Cells (1 × 106) were used to perform the ChIP assay. Chromatins from U87 and U118 cells were incubated with p65 antibody or control IgG overnight. The immunoprecipitated DNA were analyzed by PCR using primers to p65 binding sites within the promoter region of Snail.

Statistical analysis Student’s t test was used to compare the differences between groups. Correlations between expression level of AnnexinA5 and downstream member’s expression level of PI3K/Akt/ NF-κB signaling were analyzed using the Spearman rank correlation test. Data are presented as means ± SD. Statistical analyses were performed using SigmaStat and GraphPad Prism 5.02 software (GraphPad, CA, USA). P < 0.05 was considered significant.

Results AnnexinA5 was highly expressed in high‑grade glioma patients as well as cell lines The relative expression level of AnnexinA5 were examined with qRT-PCR assay in 71 normal brain tissues and 118 gliomas specimens including WHO low grade glioma (I/II) (n = 47), grade III (n = 41) and grade IV (n = 30).The results showed that AnnexinA5 level was dramatically upregulated in glioma samples compared with the normal tissues (P < 0.05) (Fig. 1a, b). We further examined whether AnnexinA5 expression level correlated with clinicopathological characteristics of glioma. As showed in Supplementary Table 1, AnnexinA5 expression is associated with advanced grade (III + IV) of glioma patients. Furthermore, to strengthen the significant quantification of AnnexinA5 in glioma, we enrolled another 62 glioma patient specimens (25, 19 and 18 patients with glioma were classified as WHO I/II, III and IV) and 13 non-tumor specimens to validate the expression of AnnexinA5 using qRT-PCR as suggested in the different malignant gliomas. Consistent

13



Journal of Neuro-Oncology

Fig. 1  AnnexinA5 expression is up-regulated in glioma tissues and glioma cells. a AnnexinA5 expression in Glioma and normal brain tissues. AnnexinA5 expression levels were calculated from the AnnexinA5/GAPDH expression ratio. b The expression levels of AnnexinA5 were significantly higher in glioma samples with grades III or IV than that in the low-grade glioma group and in the controls. c AnnexinA5 level was evaluated in glioma tissues, glioma cell lines as well as the normal human astrocytes cell line (NHA). Data are shown as mean ± SD for three independent experiments (n = 3); *P < 0.05. d The mRNA levels of AnnexinA5 were evaluated by qPCR analysis in glioma cells transfected with scramble (negative control) and pCDNA3.1 vector expressing AnnexinA5 (AnnexinA5-pCDNA3.1). e Images represented the results of protein levels of AnnexinA5 by western blotting analysis in glioma cells transfected with scramble (negative control) and pCDNA3.1 vector

expressing AnnexinA5 (AnnexinA5-pCDNA3.1) (Left). The density of the AnnexinA5 blot was normalized relative to GAPDH which was expressed as fold changes to that of negative control (designated as 1) (Right). f The mRNA levels of AnnexinA5 was evaluated by qPCR in glioma cells, infected with lentivirus expressing two independent shRNAs (AnnexinA5-shRNA#1 and AnnexinA5-shRNA#2). AnnexinA5-shRNA#2 exhibited higher inhibition efficiency, and was used in the further experiments. g Images represented the results of protein levels of AnnexinA5 by western blotting analysis in glioma cells, infected with lentivirus expressing two independent shRNAs (Left). The density of the AnnexinA5 blot was normalized relative to GAPDH which was expressed as fold changes to that of negative control (designated as 1) (Right). Data values indicate means ± SD, *P < 0.05 versus the control

with the above results, expression level of AnnexinA5 in the 62 glioma tissues was correlated with the degree of malignancy. With respect to normal brain tissues and the patients with the low-grades I/II, the patients with grades III or IV exhibited a significantly high AnnexinA5 expression (Supplementary Fig. 1). Meanwhile, the expression of AnnexinA5 in U87 and U118 cells were also confirmed via qRT-PCR assay, and the expression levels of AnnexinA5 were also markedly up-regulated in U87 and U118 cells compared with in the

normal human astrocytes cell line (NHA), which is consistent with the observation in tissues (Fig. 1c).

13

Establishing of stable transfectants with overexpression or knockdown of AnnexinA5 The above results prompted us to construct stable transfectants with AnnexinA5 overexpression or knockdown to elucidate the biological significance of AnnexinA5 in glioma cells. Through mRNA and protein levels analysis, we found

Journal of Neuro-Oncology

that AnnexinA5 expression level was increased in U87 and U118 cells, respectively, compared with the negative control cells (Fig. 1d, e). In addition, we used two shRNAs against AnnexinA5 (shRNA#1and shRNA#2) to construct stable transfectants with AnnexinA5 knockdown. Because of their effectiveness, we successfully established an AnnexinA5 knockdown stable transfectants via AnnexinA5-shRNA#2 (Fig. 1f, g).

Effects of overexpression and knockdown of AnnexinA5 on glioma cells migration and invasion We further examined whether overexpression and knockdown of AnnexinA5 could affect the migration and

Fig. 2  The effects of AnnexinA5 on glioma cells tumorigenicity, migration and invasion. a Overexpression of AnnexinA5 enhanced invasion and migration of glioma cells, while knockdown of AnnexinA5 inhibited the invasion and migration of glioma cells. Data indicates means ± SD (n = 3), *P < 0.05 vs. the control. b Wound healing assay showed that overexpression of AnnexinA5 promoted motility of glioma cells presented. The accumulated distance travelled by the cells between the two boundaries of the acellular area was imaged and measured under a microscope. c Images represented the results of decreased protein levels of two important markers (MMP-2 and

invasion of glioma cells in vitro. As illustrated in Fig. 2a, b, increased expression of AnnexinA5 resulted in markedly increased migration and invasive behavior of glioma cells (P < 0.05), while reduction of AnnexinA5 obviously inhibited migration and invasive activity compared with the negative control cells (P < 0.05). These data indicated that AnnexinA5 plays an important role in glioma cell migration and invasive. Then, we assessed protein level changes of two important genes (MMP-2, MMP-9) corresponding to migration and invasion following AnnexinA5 dyregulation by western blot assay. Protein levels showed significant decrease of MMP-2 and MMP-9 after knocking down AnnexinA5 (Fig. 2c).

MMP-9) corresponding to migration and invasion by western blot analysis in both glioma cell lines with knocked down AnnexinA5. Data are presented as the mean ± SD, *P < 0.05 vs. the control. d The effect of AnnexinA5 on glioma cells tumorigenicity in  vivo (n = 8, each group). The images show representative tumor growth on days 10, 15 and 20 after implantation was detected by Bioluminescence imaging. e The result of IHC showed the expression of MMP-2 in AnnexinA5-knockin and knockdown experiments, compared to negative control

13



A functional link between AnnexinA5 and glioma in vivo In order to investigate the tumor growth effect of AnnexinA5 in vivo, U87 cells with overexpression and knockdown of AnnexinA5 were implanted stereotactically into the brains of mice. Intracranial tumor growth on days 10, 20 and 28 after implantation was measured based on the changes of the bioluminescent signal by Bioluminescence imaging. The results showed that at days 28, the intracranial tumor volume derived from U87 cells with overexpression of AnnexinA5 were significant larger, compared to negative control group; while significantly smaller tumor size were found in AnnexinA5 knockdown group compared to negative control group (Fig. 2d). We further performed an IHC assay in the brain tissues slices resected from the orthotopic mice staining with hematoxylin and eosin. The expression of extracellular matrix MMP-2 was examined with the appropriate primary antibodies. Consistent with the in vitro results, the result of IHC showed the increased expression of MMP-2 in AnnexinA5knockin experiments, compared to negative control (Fig. 2e).

AnnexinA5 increase Snail expression in glioma via modulating PI3K/Akt/NF‑κB signaling pathway AnnexinA5 reportedly is closely associated with tumorigenesis and progression through phosphatidylinositol 3-kinase (PI3K)/Akt/NF-κB pathway [18, 19]. Accordingly, Snail is found a potential candidate target of NF-κB, whose expression is up-regulated directly by NF-κB mechanistically binding to Snail promotor. We hypothesized that AnnexinA5 may regulate the transcriptional activity of Snail promoter in U87 cells. We firstly examined the mRNA levels and protein levels of PI3K or Akt, after glioma cells transfected with siRNAs against PI3K or Akt, respectively. siRNAs against PI3K or Akt could significantly reduce the level of PI3K or Akt. Effective silencing of PI3K or Akt siRNA were presented in Supplementary Fig. 2. The reporter plasmids were transiently co-transfected to stable U87 with overexpression and knockdown of AnnexinA5 with or without PI3K or Akt siRNA, and the relative luciferase activity was examined. As shown in Fig. 3a, the relative luciferase activity of Snail in the glioma cells with stable AnnexinA5 overexpression was significantly higher compared with the negative cells. However, AnnexinA5 had only a slight effect on transcription from the luciferase reporter construct in the absence of PI3K or Akt (Fig. 3b, c). Similar results were observed in U118 cells. We further selected a well-characterized inhibitor of PI3K, LY294002 to further prove the effect of AnnexinA5 on phosphatidylinositol 3-kinase (PI3K)/Akt/NF-κB pathway [20]. Many studies have reported that LY294002 was

13

Journal of Neuro-Oncology

used in malignant tumor to induce the blockage of PI3K/ AKT signaling [21, 22]. Therefore, based on previous study, LY294002 was used in our study. Briefly, the reporter plasmids were transiently co-transfected to stable U87 or U118 cells with overexpression and knockdown of AnnexinA5 followed by a final concentrations of 20 µmol/L LY294002 (Sigma) treatment, and were subsequently analyzed for luciferase activity of Snail. Similar results were observed in the siPI3K or siAkt group that there was a slight effect on transcription in the absence of PI3K or Akt using PI3K/AKT inhibitor LY294002 (Fig. 3b, c). We further examined the expression level of PI3K/Akt activity/signaling in response to AnnexinA5 activation. As illustrated by western blotting analysis (Fig. 3d), the downstream members of PI3K/Akt pathway including phosphorylated PI3K, phosphorylated Akt and NF-κB subunit p65 and Snail were found remarkably activated in U87 cells with overexpression of AnnexinA5. Similar results were observed for downstream members of PI3K/Akt pathway triggered by AnnexinA5 in U118 cells as well. To further confirm that AnnexinA5 regulates Snail expression through PI3K/Akt/NF-κB pathways, we determined whether AnnexinA5 dyregulation affected the binding of NF-κB subunit p65 to Snail promoter in glioma cells using chromatin immunoprecipitation (ChIP). As showed in the Fig. 3e, a remarkable enrichment of p65 was observed in the Snail promoter in the glioma cells with AnnexinA5overexpression. However, upon the silence of AnnexinA5 (Fig. 3f) or in the absence of PI3K (Fig. 3g) or Akt (Fig. 3h), the enrichment of p65 to Snail was either reduced or obviously non-existent. Taken together, these observations revealed that AnnexinA5 might act as an essential regulator of the PI3K/Akt/NF-κB signaling pathway by facilitating the binding of p65 to Snail promoter, contributing to the metastasis and invasion phenotype of glioma cells.

Correlation of AnnexinA5 expression level and downstream member’s expression level of PI3K/ Akt/NF‑κB signaling in clinical samples Based on these results, we further measured the expression level of AnnexinA5 and PI3K/Akt/NF-κB signaling expression level in 50 glioma tissues by western blotting analysis, respectively (Fig. 4a). Intriguingly, among 50 randomly selected tissues from glioma patients, 67.1, 58.6 and 72.5% high-grade glioma showed a significant increase in phosphorylated PI3K (Fig. 4b), phosphorylated Akt (Fig. 4c) and NF-κB subunit p65 level (Fig. 4d), compared with lowergrade glioma. Moreover, glioma patients with a higher AnnexinA5 expression level showed a substantial upregulation of phosphorylated PI3K (Fig. 4e), phosphorylated Akt (Fig. 4f) and NF-κB subunit p65 (Fig. 4g) in glioma tissues. These results further support the hypothesis that AnnexinA5

Journal of Neuro-Oncology

Fig. 3  AnnexinA5 modulate PI3K/Akt/NF-κB signaling pathway. The effect of AnnexinA5 on transcriptional activity of downstream gene Snail of PI3K/Akt/NF-κB signaling pathway in glioma cells with AnnexinA5-overexpression and AnnexinA5-downregulation (a) treated with PI3K or Akt siRNA or PI3K/AKT inhibitor (LY294002) (b, c). d Images represented the results of western blotting illustration of the PI3K/Akt/NF-κB signaling pathway regulated by AnnexinA5 in glioma cells (U87 cells and U118 cells) (Left). Quantification of

the Western blot results in U87 cells and U118 cells (Right). Data are presented as the mean ± SD from triple independent experiments. GAPDH was used as a control. Glioma cells were with AnnexinA5overexpression (e) and AnnexinA5-downregulation (f), or transfected with PI3K or Akt siRNA (g, h), respectively. The enrichment ability of NF-κB subunit p65 to Snail promoter in glioma cells were analyzed using ChIP

could induce tumor metastasis and invasion via activating the PI3K/Akt/NF-κB signaling pathway.

malignant phenotype of glioma may be resulted from the activation of the PI3K/Akt/NF-κB signaling pathway. Annexins are a group of presenting 12 predominantly cytosolic soluble proteins sharing the property of binding to negatively-charged phospholipids in a ­Ca2+-dependent manner [23–25]. Annexins members have been shown to participate in a wide variety of functions, including membrane repair processes, intracellular vesicles fusion membrane organization, membrane-to-cytoskeleton linkage and ion channel activity. As a major intracellular ­Ca2+-binding protein, AnnexinA5 has been described playing an essential

Discussion AnnexinA5 has been reported to strongly associate with tumorigenesis and progression in various tumor types. In the current study, we showed that AnnexinA5 conspicuously promote the migration and invasive behaviors of glioma cells. The potential molecular mechanisms involved in the

13



Journal of Neuro-Oncology

Fig. 4  Linear regression of AnnexinA5′ expression level. a The representative images of the expression profiling of AnnexinA5 and PI3K/Akt/NF-κB signaling in glioma tissues with high-grade (H1H5) and low-grade (L1-L5) by western blotting analysis, respectively. 67.1, 58.6 and 72.5% glioma tumors showed a significant increase in phosphorylated PI3K (b), phosphorylated Akt (c) and NF-κB subunit p65 level (d) by western blotting analysis, respectively, compared with noncancerous glioma tissues patients. Correlations of Annex-

inA5 expression level and downstream members’ expression levels of PI3K/Akt/NF-κB signaling in 50 glioma tissues samples. Paired Student’s t test was used to analyze the significant differences. Data are indicated as means ± SD. *P < 0.05. e AnnexinA5 vs. phosphorylated PI3K ­ (R2 = 0.401, P < 0.01). f AnnexinA5 vs. phosphoryl2 ated Akt ­(R  = 0.479, P < 0.01). g AnnexinA5 vs. NF-κB subunit p65 ­(R2 = 0.632, P < 0.01)

role in cell cycle regulation, cell proliferation and invasion in a number of tumor types [26–28]. However, the effects and potential mechanism of AnnexinA5 on glioma cell migration and invasion are yet to be fully elucidated. Therefore, the aim of the present study was to identify molecular mechanisms of invasion and metastasis in glioma. To prove our hypothesis, we firstly investigate the AnnexinA5 expression levels in glioma tissues with different grades, and found AnnexinA5 were significantly upregulated in high-grade gliomas, which are consistent with data from our previous reports and other groups. Notably, dysexpression of AnnexinA5 reportedly lead to deregulated activation of PI3K/Akt/NF-κB signaling pathway. Meanwhile, in our previous studies, we demonstrated that AnnexinA5 could induce invasion and chemoresistance to temozolomide in glioblastoma multiforme cells via a PI3K-dependent mechanism. The PI3K/Akt signaling pathway is well known to be one of the most significant pathways in the metastasis of malignant tumors [29–31]. Evidence of the important role

of the PI3K/Akt pathway has been well documented, which may contribute tumorigenesis and progression of numerous types of cancer, including proliferation, invasion and migration. In addition, it has been widely accepted that the activation of this pathway could increase NF-κB activation, which upregulate the downstream extracellular matrix components, particularly MMP-2 and -9 and mesenchymal cell-specific protein Snail, causing a series of aggressive phenotypes [32–35]. In the present study, the protein levels of phosphorylated PI3K, phosphorylated Akt, NF-κB subunit p65, MMP-2 and Snail were found remarkably activated in glioma cells with overexpression of AnnexinA5. Dualluciferase reporter assay analysis showed that the relative luciferase activity of Snail in the glioma cells with stable AnnexinA5 downregulation was significantly decreased compared with the negative cells in the absence of PI3K or Akt using PI3K/AKT siRNA or a PI3K inhibitor LY294002. Furthermore, ChIP assay further confirmed the association between AnnexinA5 activity and PI3K/Akt/NF-κB pathway activation. Most importantly, correlation analysis revealed

13

Journal of Neuro-Oncology Fig. 5  A schematic model of AnnexinA5 promote the progression of glioma cancer by upregulating downstream members’ expression of PI3K/ AKT/NF-κB pathway

that patients with a higher AnnexinA5 expression level showed a substantial upregulation of phosphorylated PI3K, phosphorylated Akt and NF-κB subunit p65 in glioma tissues. These results indicate that upregulation of AnnexinA5 could increase the activity of the PI3K/Akt/ NF-κB signaling pathway in glioma cells (Fig. 5). In addition, cell migration and invasion were elucidated by chamber transwell assays. The results revealed that the migratory and invasive behaviors of glioma cells were increased by AnnexinA5 overexpression. Glioma cells with AnnexinA5 overexpression were injected subcutaneously into either side of the posterior flank of the female nude mice. The results showed that tumor volume derived from glioma cells with overexpression of AnnexinA5 were significant larger, compared to negative control group. In conclusion, our present study demonstrates that overexpression of AnnexinA5 significantly promote glioma cells invasion and migration via activating the PI3K/Akt/NF-κB signaling pathway in vitro. Thus, AnnexinA5 or the PI3K/ Akt/NF-κB pathway may be a potential choice target for inhibiting glioma metastases. Funding  This work was supported by the National Scientific Foundation of China grants (No. 81560411).

Compliance with ethical standards  Conflict of interest  The authors declare no conflict of interest. Ethical approval  This study was approved by the Review Board at The Second Affiliated Hospital of Nanchang University, and written informed consent was obtained from each participant. All procedures

performed were in accordance with the ethical standards of the institutional research committees and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

References 1. Chhabda S, Carney O, D’Arco F, Jacques TS, Mankad K (2016) The 2016 World Health Organization Classification of tumours of the Central Nervous System: what the paediatric neuroradiologist needs to know. Quant Imaging Med Surg 6:486–489. https​://doi. org/10.21037​/qims.2016.10.01 2. Komori T, Sasaki H, Yoshida K (2016) Revised WHO classification of tumours of the central nervous system: summary of the revision and perspective. No shinkei geka Neurol Surg 44:625– 635. https​://doi.org/10.11477​/mf.14362​03347​ 3. Jansen M, Yip S, Louis DN (2010) Molecular pathology in adult gliomas: diagnostic, prognostic, and predictive markers. Lancet Neurol 9:717–726. https:​ //doi.org/10.1016/S1474-​ 4422(10)70105​ -8 4. Rock K, McArdle O, Forde P, Dunne M, Fitzpatrick D, O’Neill B, Faul C (2012) A clinical review of treatment outcomes in glioblastoma multiforme–the validation in a non-trial population of the results of a randomised Phase III clinical trial: has a more radical approach improved survival? Br J Radiol 85:e729-733. https:​ //doi. org/10.1259/bjr/83796​755 5. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, Scheithauer BW, Kleihues P (2007) The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 114:97–109. https:​ //doi.org/10.1007/s00401​ -007-0243-4 6. Van Meir EG, Hadjipanayis CG, Norden AD, Shu HK, Wen PY, Olson JJ (2010) Exciting new advances in neuro-oncology: the avenue to a cure for malignant glioma. CA 60:166–193. https​:// doi.org/10.3322/caac.20069​ 7. Maher EA, Furnari FB, Bachoo RM, Rowitch DH, Louis DN, Cavenee WK, DePinho RA (2001) Malignant glioma: genetics

13



8. 9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21.

Journal of Neuro-Oncology and biology of a grave matter. Genes Dev 15:1311–1333. https​:// doi.org/10.1101/gad.89160​1 Gerke V, Moss SE (2002) Annexins: from structure to function. Physiol Rev 82:331–371. https​://doi.org/10.1152/physr​ev.00030​ .2001 Xue G, Hao LQ, Ding FX, Mei Q, Huang JJ, Fu CG, Yan HL, Sun SH (2009) Expression of annexin a5 is associated with higher tumor stage and poor prognosis in colorectal adenocarcinomas. J Clin Gastroenterol 43:831–837. https​://doi.org/10.1097/ MCG.0b013​e3181​9cc73​1 Sun B, Bai Y, Zhang L, Gong L, Qi X, Li H, Wang F, Chi X, Jiang Y, Shao S (2016) Quantitative proteomic profiling the molecular signatures of annexin A5 in lung squamous carcinoma cells. PloS ONE 11:e0163622. https:​ //doi.org/10.1371/journa​ l.pone.016362​ 2 Lee EK, Cho H, Kim CW (2011) Proteomic analysis of cancer stem cells in human prostate cancer cells. Biochem Biophys Res Commun 412:279–285. https​: //doi.org/10.1016/j. bbrc.2011.07.083 Martin M, Leffler J, Blom AM (2012) Annexin A2 and A5 serve as new ligands for C1q on apoptotic cells. J Biol Chem 287:33733–33744. https​://doi.org/10.1074/jbc.M112.34133​9 Li X, Chen L, Liang XJ, Gao YF, Wang XJ, Xu Q, Yan Y, Gao FL (2012) Annexin A5 protein expression is associated with the histological differentiation of uterine cervical squamous cell carcinoma in patients with an increased serum concentration. Mol Med Rep 6:1249–1254. https​://doi.org/10.3892/mmr.2012.1078 Rajcevic U, Petersen K, Knol JC, Loos M, Bougnaud S, Klychnikov O, Li KW, Pham TV, Wang J, Miletic H, Peng Z, Bjerkvig R, Jimenez CR, Niclou SP (2009) iTRAQ-based proteomics profiling reveals increased metabolic activity and cellular cross-talk in angiogenic compared with invasive glioblastoma phenotype. Mol Cell Proteomics 8:2595–2612. https​://doi.org/10.1074/mcp. M9001​24-MCP20​0 Shi X, Zhan L, Xiao C, Lei Z, Yang H, Wang L, Zhao J, Zhang HT (2015) miR-1238 inhibits cell proliferation by targeting LHX2 in non-small cell lung cancer. Oncotarget 6:19043–19054. https:​ // doi.org/10.18632​/oncot​arget​.4232 Duan R, Han L, Wang Q, Wei J, Chen L, Zhang J, Kang C, Wang L (2015) HOXA13 is a potential GBM diagnostic marker and promotes glioma invasion by activating the Wnt and TGF-beta pathways. Oncotarget 6:27778–27793. https​://doi.org/10.18632​ /oncot​arget​.4813 Li W, Zheng J, Deng J, You Y, Wu H, Li N, Lu J, Zhou Y (2014) Increased levels of the long intergenic non-protein coding RNA POU3F3 promote DNA methylation in esophageal squamous cell carcinoma cells. Gastroenterology 146:1714–1726 e1715. https:​ // doi.org/10.1053/j.gastr​o.2014.03.002 Wang J, Zhang Y, Liu X, Ma J, Liu P, Hu C, Zhang G (2014) Annexin A5 inhibits diffuse large B-cell lymphoma cell invasion and chemoresistance through phosphatidylinositol 3-kinase signaling. Oncol Rep 32:2557–2563. https​://doi.org/10.3892/ or.2014.3547 Wu L, Yang L, Xiong Y, Guo H, Shen X, Cheng Z, Zhang Y, Gao Z, Zhu X (2014) Annexin A5 promotes invasion and chemoresistance to temozolomide in glioblastoma multiforme cells. Tumour Biol 35:12327–12337. https​://doi.org/10.1007/s1327​ 7-014-2545-1 Singh AR, Joshi S, George E, Durden DL (2014) Anti-tumor effect of a novel PI3-kinase inhibitor, SF1126, in (12) V-Ha-Ras transgenic mouse glioma model. Cancer Cell Int 14:105. https​://doi. org/10.1186/s1293​5-014-0105-9 Shingu T, Yamada K, Hara N, Moritake K, Osago H, Terashima M, Uemura T, Yamasaki T, Tsuchiya M (2003) Synergistic

13

22.

23. 24. 25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

augmentation of antimicrotubule agent-induced cytotoxicity by a phosphoinositide 3-kinase inhibitor in human malignant glioma cells. Cancer Res 63:4044–4047 Gaelzer MM, Coelho BP, de Quadros AH, Hoppe JB, Terra SR, Guerra MC, Usach V, Guma FC, Goncalves CA, Setton-Avruj P, Battastini AM, Salbego CG (2016) Phosphatidylinositol 3-Kinase/AKT pathway inhibition by doxazosin promotes glioblastoma cells death, upregulation of p53 and triggers low neurotoxicity. PloS ONE 11:e0154612. https​://doi.org/10.1371/journ​ al.pone.01546​12 Moss SE, Morgan RO (2004) The annexins. Genome Biol 5:219. https​://doi.org/10.1186/gb-2004-5-4-219 Laohavisit A, Davies JM (2011) Annexins. New Phytol 189:40– 53. https​://doi.org/10.1111/j.1469-8137.2010.03533​.x Mortimer JC, Laohavisit A, Macpherson N, Webb A, Brownlee C, Battey NH, Davies JM (2008) Annexins: multifunctional components of growth and adaptation. J Exp Bot 59:533–544. https​:// doi.org/10.1093/jxb/erm34​4 Shin DW, Kwon YJ, Ye DJ, Baek HS, Lee JE, Chun YJ (2016) Auranofin suppresses plasminogen activator inhibitor-2 expression through annexin A5 induction in human prostate cancer cells. Biomol Ther. https​://doi.org/10.4062/biomo​lther​.2016.223 Hong M, Park N, Chun YJ (2014) Role of annexin a5 on mitochondria-dependent apoptosis induced by tetramethoxystilbene in human breast cancer cells. Biomol Ther 22:519–524. https​://doi. org/10.4062/biomo​lther​.2014.112 Park N, Chun YJ (2014) Auranofin promotes mitochondrial apoptosis by inducing annexin A5 expression and translocation in human prostate cancer cells. J Toxicol Environ Health A 77:1467–1476. https​://doi.org/10.1080/15287​394.2014.95583​4 Tang Y, Lv P, Sun Z, Han L, Luo B, Zhou W (2015) 14-3-3zeta up-regulates hypoxia-inducible factor-1alpha in hepatocellular carcinoma via activation of PI3K/Akt/NF-small ka, CyrillicB signal transduction pathway. Int J Clin Exp Pathol 8:15845–15853 Shan RF, Zhou YF, Peng AF, Jie ZG (2014) Inhibition of AuroraB suppresses HepG2 cell invasion and migration via the PI3K/ Akt/NF-kappaB signaling pathway in  vitro. Exp Ther Med 8:1005–1009. https​://doi.org/10.3892/etm.2014.1844 Rodriguez M, Luo W, Weng J, Zeng L, Yi Z, Siwko S, Liu M (2014) PSGR promotes prostatic intraepithelial neoplasia and prostate cancer xenograft growth through NF-kappaB. Oncogenesis 3:e114. https​://doi.org/10.1038/oncsi​s.2014.29 Felx M, Guyot MC, Isler M, Turcotte RE, Doyon J, Khatib AM, Leclerc S, Moreau A, Moldovan F (2006) Endothelin-1 (ET-1) promotes MMP-2 and MMP-9 induction involving the transcription factor NF-kappaB in human osteosarcoma. Clin Sci 110:645– 654. https​://doi.org/10.1042/CS200​50286​ Ahmad A, Biersack B, Li Y, Kong D, Bao B, Schobert R, Padhye SB, Sarkar FH (2013) Targeted regulation of PI3K/Akt/mTOR/ NF-kappaB signaling by indole compounds and their derivatives: mechanistic details and biological implications for cancer therapy. Anti-cancer Agents Med Chem 13:1002–1013 Zhang LL, Mu GG, Ding QS, Li YX, Shi YB, Dai JF, Yu HG (2015) Phosphatase and tensin homolog (PTEN) represses colon cancer progression through inhibiting paxillin transcription via PI3K/AKT/NF-kappaB pathway. J Biol Chem 290:15018–15029. https​://doi.org/10.1074/jbc.M115.64140​7 Jiao Y, Li H, Liu Y, Guo A, Xu X, Qu X, Wang S, Zhao J, Li Y, Cao Y (2015) Resveratrol inhibits the invasion of glioblastoma-initiating cells via down-regulation of the PI3K/Akt/NFkappaB signaling pathway. Nutrients 7:4383–4402. https​://doi. org/10.3390/nu706​4383