Novel Antiproliferative and Antioxidant Role of BjAnn1, a Mustard Annexin Protein in Human Glioblastoma Cell Lines

Oxidative stress is caused due to the redox imbalance and excessive generation of reactive oxygen species (ROS) including superoxide (O2-), hydrogen peroxide (H2O2) and hydroxyl radical (OH -) as a result of various metabolic enzymatic processes in cells [1-3]. Although basal levels of ROS are required to regulate various signaling pathways in proliferating cells, excessive ROS accumulation induces deleterious modifications on essential macromolecules such as lipids, proteins and DNA leading to cellular damage [3]. To cope with high ROS, cells employ various defense mechanisms using antioxidant enzymes including catalases, peroxidases and superoxide dismutases [3-5]. However, several studies demonstrated the critical role of increased ROS levels in the tumor initiation and progression by increasing cell proliferation, migration and invasion [6]. The characteristic anchorage-independent growth and resistance towards anoikis-mediated cell death was ascribed to the high ROS levels and constant oxidative stress in tumor cells when compared to their normal counterparts [7,8]. High ROS levels were also implicated in the maintenance of stem-like characteristics of the cancer-initiating cells and anchorage-independent growth in the malignantly transformed invading cancer cells [8,9].


Introduction
Oxidative stress is caused due to the redox imbalance and excessive generation of reactive oxygen species (ROS) including superoxide (O 2-), hydrogen peroxide (H 2 O 2 ) and hydroxyl radical (OH -) as a result of various metabolic enzymatic processes in cells [1][2][3]. Although basal levels of ROS are required to regulate various signaling pathways in proliferating cells, excessive ROS accumulation induces deleterious modifications on essential macromolecules such as lipids, proteins and DNA leading to cellular damage [3]. To cope with high ROS, cells employ various defense mechanisms using antioxidant enzymes including catalases, peroxidases and superoxide dismutases [3][4][5]. However, several studies demonstrated the critical role of increased ROS levels in the tumor initiation and progression by increasing cell proliferation, migration and invasion [6]. The characteristic anchorage-independent growth and resistance towards anoikis-mediated cell death was ascribed to the high ROS levels and constant oxidative stress in tumor cells when compared to their normal counterparts [7,8]. High ROS levels were also implicated in the maintenance of stem-like characteristics of the cancer-initiating cells and anchorage-independent growth in the malignantly transformed invading cancer cells [8,9].
Annexins are a superfamily of calcium-binding proteins that bind to negatively charged membrane phospholipids in cells and were reported to be conserved in eukaryotes [10][11][12]. Interestingly emerging evidences have shown that annexins are among the genes, whose expression is differentially altered in various neoplasms [13]. Annexin gene expression is differentially regulated by various stress stimuli and is reported to be involved in exocytosis, actin binding, ion transport, intracellular calcium signaling, growth regulation, apoptosis and inflammation, [14][15][16][17]. Several studies demonstrated the peroxidase activity of plant annexins and this activity was primarily attributed to the presence of sulfur cluster in the first annexin domain and conserved His40 residue at the N-terminal region which is homologous to haembinding domain of horse-radish peroxidase, however, subsequent site-directed mutagenesis studies confirmed that His40 is not essential for maintaining the annexin peroxidase activity [18][19][20]. Conversely, animal annexins lack His40 residue and were not reported to display any peroxidase activity until now. Upon heterologous expression, AnnAt1 from Arabidopsis thaliana displayed peroxidase activity and rescued H 2 O 2 -sensitive E. coli ∆OxyR mutant and mammalian cells from oxidative stress-induced cell death [21][22][23]. Our earlier studies demonstrated that the Brassica juncea annexin 1 protein (BjAnn1), the close homolog of AnnAt1, also exhibits in vitro peroxidase activity [24]. The transgenic overexpression of BjAnn1 in tobacco and cotton conferred tolerance towards multiple stresses including oxidative stress by scavenging stress-induced ROS formation [24][25].
Glioblastoma (GBM) are the most malignant, adult brain tumors and account for ~80% of all primary brain tumors [26,27]. Despite modest improvement in the standard therapies, the median survival is approximately 12-15 months after diagnosis [28]. Therefore, finding new therapeutic options against these aggressive tumors is clinically significant. Elevated ROS has been correlated with the activation of several signaling mechanisms essential for tumor initiation and progression including NF-κB pathway [29]. Constitutive activation of NF-κB pathway is evident in glioblastomas and is associated with enhanced cell growth, cell cycle progression, chemoresistance, migration and invasion [30][31][32]. Previous reports showed the antiproliferative roles of antioxidant enzymes in cancer cell lines by detoxification of ROS [5,33,34]. Among these, manganese-superoxide dismutase (SOD2) was shown to be a tumor suppressor and implicated in oxidative stress response in various cancers [7,35,36]. Overexpression of SOD2 and SOD3 reduced the malignant phenotype and abrogated tumor growth in gliomas [5,35].
Based on the significant sequence homology with human annexins, peroxidase and ROS-detoxifying activities of BjAnn1, the present study was performed to investigate whether BjAnn1 could function in human GBM cell lines across evolutionary boundaries. With the implication of BjAnn1 in oxidative stress, we aimed at identifying its effect on proliferation and migration in GBM cell lines. Our results demonstrated that, BjAnn1 exhibited a novel inhibitory effect on cell proliferation and migration suggesting it as a potential therapeutic target in the future treatment of glioblastoma.

Cell viability and clonogenic assay
Cell viability was determined by MTT assay in empty vector, pEVand BjAnn1-expressing cells as described earlier [38]. Briefly, U251 and U87 cells were seeded in 96-well plates (2 × 10 3 cells/well) in DMEM medium containing 10% serum. After incubation for 72 h, growth inhibition was determined by using traditional MTT assays. For the clonogenic assays, EV-and BjAnn1-expressing U87 and U251 cells (500 cells/well) were seeded in 6-well plates and allowed to grow for a further period of 8 days. The cells were then fixed in ice cold methanol and stained with 0.5% crystal violet solution to visualize the colonies.

qRT-PCR and cDNA-PCR array
Total RNA was isolated from empty vector (pEV) and pcDNA-BjAnn1 expressing U251 and U87 cell lines using TRIzol ® reagent and semiquantitative RT-PCR analysis were performed as described earlier [39]. The following primers were used for semi-quantitative and quantitative PCR analyses: BjAnn1 sense 5´-ATCTTGCTCTGGACTCTTGAA-3´and antisense 5´-TGGTGTGGTGAGCGACAT-3´and GAPDH sense 5´-GGAGTCAACGGATTTGGTCGTAT-3′ and antisense, 5′-GTCTTCACCACCATGGAGAAGGCT-3′, respectively. The potential changes in the ROS-related gene expression profile by BjAnn1 overexpression were analyzed by performing the cDNA PCR arrays in EV-and BjAnn1-expressing U87 cells using Human Oxidative Stress and Antioxidant Defense (PAHS065) RT 2 Profiler PCR Array kits following the manufacturer's instructions (SA Biosciences, Frederick, MD, USA) as described earlier [40]. The following PCR conditions were used: 1 cycle of 95°C for 10min and 40 cycles of 95°C for 15s, 60°C for 30s, 72°C for 30s, followed by 1 cycle of 72°C for 10min. The data obtained from three independent arrays were analyzed using iCycler IQ version 3.

Western blotting
Whole cell lysates were prepared from GBM cells in modified RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, 50 mM NaF, 5 mM EDTA, 0.5% [wt/vol] sodium deoxycholate and 1% Triton X-100) containing phosphatase and protease inhibitors [37]. The cytosolic and nuclear extracts were prepared as described earlier [38]. Lysates were run on SDS-PAGE followed by Western blotting and immunoreactivity was visualized using enhanced chemiluminescence detection kit.

Wound-healing migration and Transwell-Matrigel invasion assays
Wound-healing cell migration and Transwell-matrigel invasion assays were performed as described earlier [40]. Briefly, the EVand BjAnn1-expressing U251 and U87 cells were cultured in 6-well plates (2×10 5 cells/well) overnight and a straight scratch was made in individual wells using 200-μl pipette tip, which has been considered as point 0 h. After 12 hours, plates were observed for wound closure and average cell migration distances were measured using a microscope calibrated with an ocular micrometer. For matrigel invasion assay, Transwell inserts (ThinCertTM; Greiner Bio-One, Monroe, NC) with 8 mm pores were coated with 200 μL growth factor reduced Matrigel (50 μg/mL in serum-free medium) (Collaborative Research, Inc., Boston, MA) for 2-3 h by placing in 37°C incubator. About 500 μL of cell suspension (1×10 4 cells/well) was added in each well. Invaded cells through matrigel after 24 h were stained with Hema-3 and visualized under light microscope. Percentage of cell invasion and relative migration distance values were represented as mean ± SE obtained from three experimental replicates.

ROS measurement and Immunocytochemistry
Oxidative stress and superoxide content was determined using total ROS/Superoxide Detection Kit following manufacturers' instructions. In brief, the EV-and BjAnn1-expressing U251 and U87 cells were seeded in 96-well plates (1×10 4 /well) and supplemented with 10 μM H 2 O 2 for 2 hours in independent experiments as described above in 6-8 replicates. H 2 O 2 -containing medium was replaced with fresh medium and cells were further incubated for another 12-16 hours. Then cells were washed gently with 1× wash buffer followed by incubation in 100 μl/well of ROS/Superoxide detection mix for 1 h at 37°C. Total ROS content was determined by reading culture plates with a fluorescence microplate reader at 488 nm-530 nm. Green fluorescence of H 2 DCFDA depicting ROS levels was also visualized under a confocal microscope. Immmunocytochemical analysis was performed as described earlier [38]. Briefly, pEV-and BjAnn1-expressing cells were seeded in 4-well chamber slides at a density of 1×10 4 /well and incubated overnight. Cells were fixed with 4% paraformaldehyde and permeabilized in 0.1% Triton-X100. Non-specific binding was blocked by treating the cells with 3% BSA and the cells were incubated with anti-BjAnn1 antibody (1:100 dilutions) overnight at 4°C. Cells were washed and incubated with Alexa Fluor ® -conjugated secondary antibodies for 1 hour at room temperature, washed, mounted and observed under a confocal microscope. The nuclei were counterstained with DAPI (4'-6-diamidino-2-phenylindole).

Reporter gene assays
Reporter gene assays were performed as described earlier [41]. Control (EV) or BjAnn1 expressing U251 and U87 cells were seeded in 6-well plates. After overnight incubation, the cells were transiently transfected with NF-κB-Luc plasmid using lipofectamine TM 2000 for 6 h. After 48 h of transfection, cells were further treated with TNF-α (10 ng/ml) and incubated for an additional period of 12 h. Each transfection was carried out in triplicate and normalized with the β-gal activity and total protein concentration. Luciferase activity was measured by using the luciferase assay system (Promega, Madison, WI). Luciferase activity was expressed as percent of relative light units versus untreated transfected cells.

Statistical analysis
Data obtained from different sets of treatment groups were statistically analyzed by one-way ANOVA with Tukey's test. The mean ± SE values from at least three individual repetitions were presented and the significant difference was denoted at p ≤ 0.01.

Brassica juncea Annexin 1 (BjAnn1) is a close homolog to animal annexins
Annexin proteins are widely conserved in plants and animals.
Sequence analyses revealed that the mustard annexin BjAnn1 shows the most striking amino acid sequence homology up to 73.18% with human annexin A13 ( Figure 1A). BjAnn1 also showed significant sequence homology with human annexins A2 (70.66%), A3 (72.53%) and A8 (70.03%). All annexins are characterized by the presence of a core domain made up of four similar repeats of approximately 70 amino acids long. The four annexin repeats are packed to form a flattened disc-like structure with a slight convex surface on which the type II and type III Ca 2+ binding loops are located and a concave surface where the -NH 2 and -COOH termini come side-by-side [17,45]. To analyze the potential homology between the four conserved annexin repeats, we performed a domain-specific sequence alignment and generated sequence logos using WebLogo 3.3 program ( Figure 1B). Several conserved and semi-conserved residues were identified in all four annexin motifs.

Ectopic expression of BjAnn1 in human GBM cell lines
Our earlier studies demonstrated the peroxidase activity of mustard annexin (BjAnn1) and its beneficial effects in alleviating oxidative  stress upon overexpression in transgenic tobacco and cotton plants from oxidative stress [24,25]. GBM cells have been shown to exhibit high ROS levels when compared to their normal counterparts and ROS is essential in maintaining increased cell proliferation, migration and invasion in tumors. In the present investigation, we aimed at verifying the effect of ROS-scavenging BjAnn1 expression in GBM cells. To accomplish this, we ectopically expressed BjAnn1 in human U251 and U87 GBM cells. The stable expression of BjAnn1 was confirmed by Western blotting (Figure 2A) and RT-PCR ( Figure 2B). To identify the sub-cellular localization of BjAnn1, immunofluorescence analyses was performed. The BjAnn1 protein was primarily localized in the membrane and cytoplasmic regions in the GBM cell lines. These results are in corroboration with the cellular localization of human annexins, which possess phospholipid and F-actin-binding activity and display abundant membrane localization ( Figure 2C).

BjAnn1 expression inhibited proliferation, colony formation, migration and invasion in U251 and U87 GBM cells
To study the role of BjAnn1 on proliferation, an MTT assay was performed. In comparison to EV-treated controls, BjAnn1expressing U251 and U87 cells showed significance decrease in cell proliferation ( Figure 3A). Further, clonogenic assay revealed that colony-forming ability of BjAnn1-expressing cells was prominently inhibited in comparison to EV-controls ( Figure 3B). It is well evident that glioma cells possess high migratory and invasive potential and often metastasized into surrounding normal brain parenchyma. We next determined the effect of BjAnn1 expression on migration and invasion of U87 and U251 cells by performing scratch wound-healing assay and matrigel invasion assay respectively. As shown in Figure 3C, the migratory potential of BjAnn1-expressing cells was significantly decreased when compared to pEV-expressing cells. Correspondingly, the invasive potential through matrigel was significantly inhibited in U87 and U251 upon BjAnn1 expression ( Figure 3D). These results suggest the anti-proliferative and anti-migratory effect of BjAnn1 in GBM cells.

BjAnn1 reduced ROS levels and modulated the expression of oxidative stress and antioxidant defense genes
Characteristic traits of deregulated cell proliferation and invasiveness were in part attributed to the high ROS levels and ROSdependent signaling activation in tumors. Based on the previous reports showing peroxidase activity of BjAnn1, we next determined its effect on ROS levels in U87 and U251 cells. The BjAnn1-expressing cells showed a significant reduction in ROS levels compared to the mock-or pEV-controls ( Figure 4A). Based on this interesting result of significant ROS diminution, we attempted to analyze the possible modulation in expression levels of genes responsive to ROSdependent signaling and metabolism. To achieve this, we performed a human oxidative stress and antioxidant defense PCR array.  Figure 4B). The expression of genes upregulated in response to BjAnn1 expression was summarized in Table 1. With the striking transcriptional elevation of SOD2 (MnSOD) and SOD3 (Cu/ZnSOD) in BjAnn1 expressing GBM cells, we further checked the protein levels. Ectopic expression of BjAnn1 significantly elevated SOD2 and SOD3 expression levels in both cell lines ( Figure 4C). To confirm the specificity of BjAnn1 function in alleviating ROS in GBM cells and to rule out the off-target consequences, we treated U87 cells with H 2 O 2 . Notably, treatment with H 2 O 2 significantly enhanced ROS levels in control and EV-treated cells ( Figure 4D). On the other hand,

BjAnn1 attenuated the NF-κB nuclear translocation and its activation
Deregulated NF-κB signaling is a hallmark in GBM progression. NF-κB is known to be activated in response to ROS and critical in the maintenance of tumor cell proliferation and migration. With the compromised ROS levels upon BjAnn1 expression in GBM cells, we next proceeded to check the levels of NF-κB subunits, p65 and p50. As shown in Figure 5A, the TNF-α induced p50 and p65 nuclear translocation was significantly reduced in BjAnn1 expressing cells compared to pEV-control cells. In contrast, the cytoplasmic p65 levels were high in BjAnn1 expressing cells compared to pEV-control cells.
To further study whether BjAnn1 interfered with the NF-κB pathway activation, we performed gene reporter assays. As shown in Figure 5B, the BjAnn1 overexpression significantly reduced the NF-κB reporter activity compared to pEV-control cells. Further, we also checked the expression of proliferation markers such as cyclin D1 and c-Myc, the direct transcriptional target genes of NF-κB. The expression levels of cyclin D1 and c-Myc were down regulated in BjAnn1 expressing GBM cells compared to control GBM cells ( Figure 5C). These results clearly demonstrated that the expression of BjAnn1 inhibits the GBM cell proliferation by interfering with NF-κB pathway activation.

Discussion
Despite using the advanced multimodal treatments, GBM remain to be the deadliest among the tumors of the central nervous system. Since the prognosis remains poor in GBM, there is an imperative need to develop novel and alternative therapeutic regimens in their control. Recently, increasing evidences suggest the possible functions of plant genes across evolutionary boundaries in the potential therapeutic applications in mammalian systems in vitro and in vivo [37,46]. Earlier studies demonstrated the novel roles of plant proteins in the regulation of cell proliferation when heterologously expressed in a variety of cancer cells [22,23,46]. Recently, we showed that expression of mammalian IκB homologue, NPR1 (non-expressor of pathogenesis related protein 1) from mustard in GBM cells inhibited their proliferation via the inhibition of NF-κB pathway [37]. The Thymidine kinase 1 from Tomato exhibited tumor suppressive role when expressed in human GBM cells and enhanced sensitivity to cytotoxic agents [46].  [23,24]. In both the transgenic plants, ROS scavenging activities were demonstrated. The Oxy5/AtAnn1 protein from Arabidopsis thaliana displayed peroxidase activity and protected oxidative stress-susceptible OxyR-mutant of E.coli from H 2 O 2induced death [21]. In another study, overexpressed Oxy5 in CHO cells and mouse leukemic K36 cells decreased superoxide production, PKC activity and inhibited subcutaneous tumor growth in nude mice [23]. Upon expression in TNF-sensitive HeLa D98 cells, Oxy5 conferred resistance to TNF cytotoxicity and decreased TNF-induced intracellular superoxide formation [22]. It has also been shown that annexins exhibited ROS neutralizing activity [20]. These studies supported the antioxidant role of annexin proteins in heterologous systems. In our present study, we ectopically expressed the BjAnn1 in GBM cells and found that it is localized mostly to the cell membrane and cytoplasm. It has been shown that annexins in eukaryotes bind to F-actin elements, which keep the annexins at cell membrane. The observed membrane staining of BjAnn1 could possibly be due to its binding to F-actin elements.
Plant annexins are expressed throughout the life cycle and their expression is influenced by the environmental cues. It is evident that annexins are capable of linking Ca2 + , redox and lipid signaling to coordinate the development with responses to the biotic and abiotic environment [19,47]. In the present study, phylogenetic analysis of BjAnn1 showed ~73% homology to human annexin A13. Annexin A13 is considered the original founder gene of the vertebrate 12 member annexin A family [44,45] and it is believed to be an earliest branching subfamily in vertebrates and physically non-syntenic to other human annexins [48,49].
In mammalian systems, annexins perform several functions including key roles in carcinogenesis. The expression profiles of annexins are consistently different in various neoplasms with overexpression in specific types of cancers, while others show consistent loss of expression. Several other studies showed that the changes in annexin expression and their subcellular localization contribute to the development and progression of cancer [13]. In our study, we demonstrated that overexpression of BjAnn1 reduced the proliferation, migration and invasion of GBM cells. Further, BjAnn1 overexpression reduced the ROS generation in GBM cells suggesting its evolutionarily conserved ROS scavenging activity. It has been shown that glioma cells express antioxidant enzymes such as superoxide dismutases, catalase and glutathione peroxidase [4]. Oxidative stress is one of the key steps in the progression of carcinogenesis. Antitumor role of antioxidant enzymes was evident in gliomas. In addition, cancer cells have low levels of antioxidant enzymes when compared to their normal cell counter parts [4,33]. The tumor suppressive role of MnSOD (SOD2) and Cu/ZnSOD (SOD3) was earlier reported in several tumors including glioma [5,7,35,36]. Overexpression of MnSOD reduces the malignant phenotype and glioma tumor growth, and generally considered as tumor suppressor in gliomas [35]. In another study when Cu/ZnSOD was overexpressed, tumor growth and metastasis were reduced in mice [5]. Inhibition of Cu/ZnSOD leads to the increased motility of cancer cells. MnSOD catalyzes the dismutation of superoxide radical to hydrogen peroxide and molecular oxygen. In our study, we investigated the changes in the expression of genes related to oxidative stress, which includes peroxidases, superoxide dismutases, oxidative stress responsive genes, antioxidants and genes involved in ROS metabolism. We found a significant deviation of transcriptional levels of oxidative stress responsive genes in BjAnn1 expressing GBM cells. We also demonstrated the increased expression of tumor suppressive antioxidant enzymes, MnSOD and Cu/ZnSOD in BjAnn1 expressing GBM cells. It has also been shown that ROS activates the NF-κB [26] and NF-κB pathway constitutively activated in gliomas. We hypothesize that the reduced proliferative and migration properties of BjAnn1 exert partly through inhibition of NF-κB pathway. In our study we observed that the nuclear translocation of p50 and p65 was also reduced in BjAnn1 overexpressing cells. Further, NF-κB reporter gene activity and its target gene activation such as cyclin D1 and c-Myc were reduced in BjAnn1 expressing GBM cells.
In summary, our study for the first time demonstrated that heterologous expression of BjAnn1 inhibits the proliferation, migration and invasion of GBM cells by enhancing ROS scavenging activity with concomitant activation of antioxidant enzymes and inhibition of NF-κB pathway. This study may initiate the development of new treatment modalities using plant transgenes as therapeutic applications for glioblastoma.