Human U87 Astrocytoma Cell Invasion Induced by Interaction of βig-h3 with Integrin α5β1 Involves Calpain-2

It is known that βig-h3 is involved in the invasive process of many types of tumors, but its mechanism in glioma cells has not been fully clarified. Using immunofluorescent double-staining and confocal imaging analysis, and co-immunoprecipitation assays, we found that βig-h3 co-localized with integrin α5β1 in U87 cells. We sought to elucidate the function of this interaction by performing cell invasion assays and gelatin zymography experiments. We found that siRNA knockdowns of βig-h3 and calpain-2 impaired cell invasion and MMP secretion. Moreover, βig-h3, integrins and calpain-2 are known to be regulated by Ca2+, and they are also involved in tumor cell invasion. Therefore, we further investigated if calpain-2 was relevant to βig-h3-integrin α5β1 interaction to affect U87 cell invasion. Our data showed that βig-h3 co-localized with integrin α5β1 to enhance the invasion of U87 cells, and that calpain-2, is involved in this process, acting as a downstream molecule.


Introduction
Gliomas have a high incidence rate, and represent the most common form of primary intracranial tumors. They are generally malignant and highly invasive to surrounding structures, and prognosis is largely correlated with tumor stage. Because of these fatal characteristics, it is hard to perform complete resection by surgery. Although much work has been done to find clues as to invasive biomarkers and effective treatment methods, the molecular mechanisms need to be further investigated [1].
Transforming growth factor (TGF)-b-inducible gene-h3 (big-h3) is widely expressed in various types of tumor cells. Though it is not normally expressed in tissues of the central nervous system, it was demonstrated to be expressed in U87 human astrocytoma cells [2,3]. According to its molecular structure and functions, different names have been assigned to the protein, such as TGFBI, RGD-CAP, and MP78/70. Previous studies have demonstrated that by interacting with integrin a3b1 or regulating store-operated Ca 2+ entry, big-h3 promotes the migration and invasive ability of tumor cells [4,5,6]. However, the role of big-h3 in affecting glioma cell invasion in the transduction pathway remains to be investigated.
Integrins are transmembrane heterodimers composed of a and b chains that provide physical and functional links between cellcell and cell-ECM (extracellular matrix) interactions to mediate many cellular activities in tumors [7,8,9]. As we know, the interaction of integrins with ECM is related to cell viability and invasion. Proteins such as EMMPRIN (extracellular matrix metalloproteinase inducer) can interact with integrins to enhance the progression of hepatoma cells [10,11]. In the present study, we found that big-h3 co-localized with integrin a5b1 in U87 cells. However, very little information is available regarding the potential roles of this phenomenon. Given that big-h3 and integrin can be involved in tumor invasion, the interaction of big-h3 with integrin a5b1 may also affect the invasion of U87 cells.
Cell invasion is a characteristic of most malignant tumors, and in glioma cells this process is often mediated by calpain-2, a calcium-dependent thiol proteinase, which consists of a catalytic subunit and a regulatory subunit [12,13]. It can be activated by millimolar levels of Ca 2+ to enhance tumor invasion [14,15,16]. We presume that Ca 2+ is the ''key point'' among big-h3, integrin a5b1 and calpain-2, and therefore attempted to elucidate this relationship.
In the present study, we showed that big-h3 and integrin a5b1 form a complex, and that they activate MMP secretion and enhance invasive potential via its downstream molecule calpain-2 in U87 cells.

Results
siRNA knockdowns can inhibit the expression of big-h3 and calpain-2 in U87 cells Previous studies have shown that big-h3 and calpain-2 are expressed in U87 cells [2,3,12]. To obtain further information about their roles, small interfering RNAs (siRNAs) were transfected into U87 cells for 36 hours to knockdown big-h3 and calpain-2 RNA and protein expression. Silencer negative control siRNAs (Snc-RNAs) were also used as a negative control, according to the manufacturer's protocol. As compared with snc-RNA treated cells, the siRNA knockdowns could effectively decrease the mRNA expression of big-h3 and calpain-2 (47.9%64.1% and 51.1%63.5%, respectively), and the protein expression of big-h3 and calpain-2 was significantly reduced to 43.4%66.5% and 34.6%62.0% (P,0.01, Figure 1A-1D).
Downregulation of big-h3 and calpain-2 decreases MMP secretion and U87 cell invasion big-h3 and calpain-2 promote invasion in tumor cells and they also enhance MMP secretion potential which is implicated in promoting metastasis by degradation of ECM. To further test for the ability to specifically suppress big-h3 and calpain-2, we analyzed the expression level of MMPs and their invasive ability following knockdown of big-h3 and calpain-2 in U87 cells. Snc-RNAs were also used as a negative control, according to the manufacturer's protocol. The results showed that these knockdowns reduced the density of MMPs and the invasive potential of U87 cells (P,0.01, Figure 2A-2D).
big-h3 co-localizes with integrin a5b1 on U87 cell membranes The invasive ability of human hepatoma cells is enhanced by the interaction of big-h3 with integrin a3b1 [5]. According to the study, we hypothesized that big-h3 might interact with integrin a5b1 to affect the invasive ability of U87 cells. So we performed immunofluorescent double-staining and confocal imaging analysis to examine cellular distribution. The results showed that staining overlapped diffusely throughout the surface of U87 cells ( Figure 3A). To further confirm this result, Co-immunoprecipita-tion experiments were performed to detect the immunoreactivity of big-h3 and the integrin a5 and b1 subunits. As Figure 3B-3E indicates, big-h3 and integrin a5b1 interacted in their native conformations in U87 cell lysates.

Interaction of big-h3 with integrin a5b1 mediates MMP secretion and cell invasion potential
To identify the function of the interaction of big-h3 with integrin a5b1, antibodies against the integrin subunits and big-h3 were used. Integrin a5b1, which is composed of an a5 chain and a b1 chain, is a receptor for fibronectin that is a component of ECM. It recognizes the sequence Arg-Gly-Asp (RGD) in its ligands. The ''function blocking'' antibodies, mouse anti-human a5 mAb (P1D6) and mouse anti-human b1 mAb (3S3), neutralize binding of a5b1 to the central cell adhesion domain of fibronectin. Such reaction may result in decrease of invasion potential and MMP secretion [17][18][19][20]. We treated U87 cells with P1D6, 3S3, P1D6+3S3 and si-big-h3 for 36 h, and then examined the invasive ability and the presence of MMPs. As shown in Figure 4A and 4B, no significant cell number and MMP density changes were found between the snc-RNA transfected alone groups and the no antibody treated groups (P.0.01). As compared with the snc-RNA treated alone cells, invasive potential were markedly decreased after treated with P1D6, 3S3 or P1D6+3S3 (P,0.01), but no significant difference were found in the groups, including blank+P1D6, blank+3S3, blank+P1D6+3S3, snc-RNA+P1D6, snc-RNA+3S3 and snc-RNA+P1D6+3S3 groups (P.0.01). Similar results were observed in big-h3 siRNA transfected alone groups compared with snc-RNA treated alone groups and the no antibody treated groups (P,0.01), but there were no significant differences in comparison with blank+P1D6, blank+3S3, blank+P1D6+3S3, snc-RNA+P1D6, snc-RNA+3S3 and snc-RNA+P1D6+3S3 groups (P.0.01). The addition of antibodies big-h3-Integrin-a5b1-Calpain-2 Pathway in U87 Cell against a5 and/or b1 did not further reduce invasion or MMP secretion in big-h3 siRNA transfected cells, compared with the cells treated with big-h3 siRNA alone (P.0.01). These results indicated that big-h3 and integrin a5b1 were both required for and dependent upon the invasion of U87 cells.
Calpain-2 is a downstream signaling molecule of big-h3mediated invasion involving a5b1 We used western blotting analysis, Reverse transcriptase polymerase chain reaction (RT-PCR) assay and invasion assay to test if calpain-2 is upstream or downstream of big-h3 and   big-h3-Integrin-a5b1-Calpain-2 Pathway in U87 Cell integrin a5b1. The results showed that the mRNA and protein expression of big-h3 was not changed following calpain-2 siRNA knockdown (P.0.01, Figure 5A and 5B), but the mRNA and protein expression of calpain-2 was decreased when U87 cells were treated with big-h3 siRNA, (P,0.01, Figure 5C and 5D). And the mRNA and protein expression of calpain-2 was decreased when U87 cells were treated with P1D6, 3S3 and P1D6+3S3 (P,0.01, Figure 5E and 5F). As compared with the snc-RNA treated alone groups, invasive potential was obviously decreased after treated with si-big-h3, si-calpain-2 or si-big-h3+si-calpain-2 (P,0.01), but no significant difference were found in the groups, including snc-RNA+si-big-h3, si-calpain-2 and si-big-h3+si-calpain-2 groups (P.0.01, Figure 5G). All of these results suggest that calpain-2 acts downstream of big-h3 and a5b1 integrin to influence invasion of U87 cells.

Discussion
Studies of the molecular mechanism of invasion could reveal targets for glioma treatment. big-h3, a 68 kDa extracellular matrix protein mainly induced by transforming growth factor-b (TGF-b), was first identified in the human lung adenocarcinoma cell line A549 [21]. It is expressed in many cells and tissues including the heart, liver, stomach, skeletal muscle and kidney [6,21,22]. Although its roles are largely unknown, it has been suggested that it involves in the regulation of many aspects of tumor cell processes, including cell adhesion, spreading, invasion, proliferation and apoptosis [22][23][24][25]. It has also been associated with corneal dystrophy, wound healing, atherosclerosis and many other human diseases [2,26]. It contains four repetitive FAS-1 domains and an integrin recognition site named RGD sequence which can serve as a ligand of integrins. In this study, we focused on the receptors for big-h3, the roles of which still remain largely unknown and tested if they could influence the invasive ability of U87 cells. The results presented here show that big-h3 enhances the invasive potential of U87 cells by interacting with integrin a5b1.
Integrins, a large family of cell matrix adhesion receptors, have been demonstrated to play important roles in many types of tumor cells. Through the interaction with the basement membrane, integrins can mediate adhesion and invasion [8,25,27]. Previous studies have shown that integrin a6b1 can enhance the aggressiveness of U87 glioma cells, and the apoptosis of Ntera2 neuronal cells, which are being evaluated to employ in central nervous system (CNS) transplantation, was delayed by the activation of integrin a5b1 [8,28]. In particular, integrin a5laminins can highly enhance the invasion of all types of glioma cells, and the migration of U251 glioma cells is downregulated by fibronectin, an ECM ligand of integrin a5b1 [29,30]. Therefore, it was not surprising that integrin inhibitors serve as a potential drug to prevent tumor cell invasion. JSM6427 and SJ749, inhibitors of integrin a5b1, attenuate glioma cell proliferation and invasion [22,31]. In the present study, we demonstrated that big-h3 was positively related to the expression of integrin a5b1 and there are no previous reports about the signaling mechanism of big-h3integrin a5b1 interaction. We further observed that antibodies to a5 and/or b1 could effectively reduce big-h3-mediated invasion and MMP secretion, and provided no significant additional inhibitory effect in U87 cells. Hence, we proposed a hypothesis that big-h3 interaction with integrin a5b1 can regulate invasion of U87 cells.
big-h3 plays key roles in tumor cell invasion and previous study has demonstrated that it increases Ca 2+ influx to enhance secretion of MMPs [4]. Ca 2+ is known to be involved in the motility, apoptosis, proliferation of cancer cells, as well as invasion [32,33]. Focal adhesion kinase (FAK), mitogen-activated protein kinase (MAPK) and extracellular regulated protein kinase (ERK) are crucial components in the signaling pathways of some integrins, and all may influence Ca 2+ accumulation. Specific inhibitors of these proteins can not only affect Ca 2+ concentration and its signaling pathway, but significantly block integrin-induced or big-h3-induced invasion and MMP release. Calpains are proteins that belong to the family of calcium-dependent intracellular cysteine proteases, and are ubiquitously expressed in glioma cells. These include mand m-isozymes, and are involved in the degradation of the main components of matrix and glycan, which are correlated with many diseases such as Alzheimer's and stroke [13,34]. Downregulation of calpains after transfection with calpain-1 (mcalpain) or calpain-2 (m-calpain) siRNA could reduce the secretion of MMPs and attenuate the adhesive and invasive potentials of some tumor cells [35,36]. ERK and MAPK are upstream molecules of calpains, and FAK is a common substrate [37,38]. Thus, Ca 2+ might be a ''medium'' to affect the invasive ability promoted by big-h3, integrin a5b1 and calpain-2, and the precise mechanism remains to be elucidated. We speculate that big-h3 enhances the invasion potential of U87 cells via integrin a5b1calpain-2 signaling pathways.
Our data demonstrates that big-h3 activates MMP secretion and promotes invasive potential by interacting with integrin a5b1 via its downstream molecule calpain-2 in U87 cells. These results are significant in that they suggest that mechanisms regulating big-h3-a5b1 interactions, and their role towards calpain-2 signaling may constitute a novel anti-glioma drug target.

Cell culture
Human U87 astrocytoma cell line, which was purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China), were routinely grown in Dulbecco's Modified Eagle's Medium (DMEM) plus 10% fetal bovine serum (FBS) at 37uC in a humidified atmosphere of 5% CO 2 incubators.

RT-PCR
Thirty-six hours after siRNAs transfection, U87 cells were collected to verify the mRNA expression by RT-PCR. Total RNA was extracted using Trizol reagent (Invitrogen, USA) and firststrand complementary DNA (cDNA) was reverse transcribed using the ReverTra Ace kit (Toyobo, China) according to the guidelines. The cDNA was used as the template and was amplified by PCR using a specific primer set for big-h3 and calpain-2, and b-actin was used as the internal control to normalize variances. All primers and probes were Synthesized by Shanghai Sangon Co. The primers used were as follows: big-h3, 59-CATTGAGAA-CAGCTGCATCG-39 (forward) and 59-AGTCTGCTCCGTTCTCTTGG-39 (reverse); calpain-2, 59- Figure 5. The expression of calpain-2 and big-h3 after si-big-h3, si-calpain-2 or integrin a5b1 mAbs treatment. Thirty-six hours after sicalpain-2 treatment, RT-PCR (A) and western blotting (B) were performed to test the mRNA and protein levels of big-h3. Thirty-six hours after si-big-h3 treatment, RT-PCR (C) and western blotting (D) were performed to test the mRNA and protein levels of calpain-2. Thirty-six hours after P1D6, 3S3 and P1D6+3S3 treatment, RT-PCR (E) and western blotting (F) were performed to test mRNA and protein levels of calpain-2. (G) Thirty-six hours after sibig-h3 and si-calpain-2 treatment, alone or in combination, invasion assay was also performed to test invasive potential. Data are representative of three independent experiments. Bars represent the mean of triplicate samples and error bars represent standard deviation. *p,0.01 versus corresponding cells with snc-RNA or no antibody treatment. Bar = 100 um. doi:10.1371/journal.pone.0037297.g005 big-h3-Integrin-a5b1-Calpain-2 Pathway in U87 Cell GCAGCCATTGCCTCCCTCAC-39 (forward) and 59-AC-CTCCACCCACTCGCCGTA-39 (reverse); b-actin, 59-CCCAGCCATGTACGTTGCTA-39 (forward) and 59-TCACCGGAGTCCATCACGAT-39 (reverse). PCR was carried out under the following conditions: 5min denaturation at 94uC, renaturation for 30 cycles at 94uC for 30s, 57uC for 30s, 72uC for 30s and 7 min extension at 72uC. The products were finally separated on 1% agarose gels and were analyzed by ultraviolet light.

Western blotting analysis
Thirty-six hours after siRNAs transfection, the conditioned medium of the U87 cells was collected. After BCA protein assay, equal amounts of protein were separated on 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were separated by electrophoresis and transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore). After being blocked with 5% non-fat milk and washed with TBS-T, the immunoblots were incubated with the designated antibodies. Signals were detected using the Western-Light chemiluminescent detection system (Applied Biosystems). And b-actin was chosen as the internal control.

Co-immunoprecipitation and western blotting analysis
For immunoprecipitation, the interaction of big-h3 with integrin a5b1 was detected by the ProFoundTM Mammalian Co-Immunoprecipitation Kit (Pierce, USA). U87 cells were collected and lysed by M-per reagent. Lysates were incubated with mouse anti-human big-h3 mAb or mouse anti-human a5 mAb (P1D6) (Chemicon) and mouse anti-human b1 mAb (3S3) (Santa Cruz). And then bound proteins were collected onto a coupling gel to elute. After that, by western blotting, the elution was resolved by SDS-PAGE, and then blotted and probed with the designated antibodies to detect integrin a5b1 or big-h3. HRPconjugated rabbit anti-goat IgG (Amersham Pharmacia) was used as the negative control.

Immunofluorescence and confocal microscopy
For the detection of the location of big-h3 and integrin a5b1, after placed onto glass coverslips overnight, U87 cells were fixed in 4% paraformaldehyde in (Phosphate Buffered Saline) PBS and then were pre-incubated with 1% bovine serum albumin (BSA) in PBS for 1h. Coverslips were treated with the primary antibodies (Santa Cruz) at a 1:200 dilution in PBS for 1h in a humid chamber. Primary antibody-treated cells were washed in PBS and then incubated with fluorescein isothiocyanate (FITC)-conjugated donkey anti-goat secondary antibodies (Santa Cruz) at a 1:500 dilution in PBS for 1 h. Cell nuclei were stained with DAPI (Biotium, USA) for 10 min at 37uC. Finally, the cells were enveloped with glycerol and observed by FV1000 laser scanning confocal microscope (Olympus).

Invasion assay
The assay was performed using 24-well Transwell units with a polycarbonate filter (8-mm pore size, Millipore) coated on the upper side with Matrigel at a concentration of 5 mg/ml to form a thin layer. Each lower chamber contained 600 ml of 0.5% FBS as the chemoattractant. After siRNAs transfection, U87 cells were harvested to put into the upper chamber and then incubated for 36 h at 37uC in a humid atmosphere containing 5% CO 2 . After incubation, non-migrating cells on the top of the chambers were completely removed by a cotton swab. Cells that invaded into the lower chambers were fixed in 95% methanol for 5 min and then determined using a colorimetric crystal violet assay.

Gelatin zymography
Thirty-six hours after siRNAs transfection, the sample protein, which was extracted from the conditioned media and was mixed with sample buffer, was separated by 8% acrylamide gels containing 0.1% gelatin. The gels were washed in 2.5% Triton X-100 (Sigma, USA) solution for 15mins twice at room temperature with gentle agitation to remove SDS and were then soaked into reaction buffer (0.05 mol/l Tris-HCl pH 7.5, 0.2 mol/ l NaCl, and 0.01 mol/l CaCl 2 ) at 37uC overnight. The gels were stained with Coomassie brilliant blue for 6 h and were then destained for 0.5 h. The expression and activation of MMP were shown by negative staining.

Statistical analysis
Statistical analysis was performed using SPSS 13.0 statistical software. The results were expressed as mean values 6 SD. And one-way ANOVA was used to evaluate the statistical significance among the groups. The differences were considered significant when P,0.01.