Fibronectin–integrin mediated signaling in human cervical cancer cells (SiHa)
- First Online:
- Cite this article as:
- Maity, G., Fahreen, S., Banerji, A. et al. Mol Cell Biochem (2010) 336: 65. doi:10.1007/s11010-009-0256-5
- 258 Views
Interaction between cell surface integrin receptors and extracellular matrix (ECM) components plays an important role in cell survival, proliferation, and migration, including tumor development and invasion of tumor cells. Matrix metalloproteinases (MMPs) are a family of metalloproteinases capable of digesting ECM components and are important molecules for cell migration. Binding of ECM to integrins initiates cascades of cell signaling events modulating expression and activity of different MMPs. The aim of this study is to investigate fibronectin–integrin-mediated signaling and modulation of MMPs. Our findings indicated that culture of human cervical cancer cell (SiHa) on fibronectin-coated surface perhaps sends signals via fibronectin–integrin-mediated signaling pathways recruiting focal adhesion kinase (FAK) extracellular signal regulated kinase (ERK), phosphatidyl inositol 3 kinase (PI-3K), integrin-linked kinase (ILK), nuclear factor-kappa B (NF-κB), and modulates expression and activation of mainly pro-MMP-9, and moderately pro-MMP-2 in serum-free culture medium.
Integrins are a large family of cell adhesion receptors which mediate cell–cell and cell–matrix adhesion regulating cell survival, proliferation, and motility [1, 2] including invasion and metastasis . Integrins “integrate” signals from extracellular matrix (ECM) proteins (fibronectin, laminin, collagen IV, vitronectin), co-cluster with different kinases, phosphatases and adaptor proteins [focal adhesion kinase (FAK), integrin-linked kinase (ILK), Src (family of protooncogenic tyrosine kinase), Shc] in focal adhesions, signaling molecules [mitogen-activated protein kinase (MAPK), Ras, Raf, Rho family GTPases, phosphatidyl inositol 3 kinase (PI-3K), protein kinase C (PKC), caveolin-1], and cytoskeleton components (vinculin, paxillin, talin, α-actinin) [4–6]. Binding of integrin to ECM proteins stimulates tyrosine phosphorylation of FAK that binds to different signaling and structural proteins such as PI-3K and plays an important role in activation of various MAPK pathways, including extracellular signal-regulated kinase (ERK) and regulate cell adhesion, shape, and motility [5–7]. ERK is phosphorylated and translocated to nucleus in which they activate several transcription factors [8, 9]. ILK connects integrin to actin cytoskeleton and signaling protein acting downstream of PI-3K, activates Akt (protein kinase B) for Activator protein 1 (AP-1) activation, and serves as a link between the ECM and the cytoskeleton [10, 11]. Integrin–ECM interaction thus initiates a signaling cascade resulting in cell migration and release of proteolytic enzymes such as matrix metalloproteinases (MMPs), and reorganization of cytoskeletal proteins . One of the most extensively studied integrins is α5β1, the major fibronectin (FN)-binding receptor in most cell types, which binds to RGD (Arginine-Glycine-Aspartate) and PHSRN (Pro-His-Ser-Arg-Asn) site of FN for maximum activity and triggers intracellular signaling cascade [13, 14]. FN (440 kDa), a potential ligand for most cell types, is present as polymeric fibrillar network in ECM and as a soluble protomer in body fluid. Proteolysis or degradation of ECM surrounding a tumor cell is considered as a key event for invasion and metastasis of malignant cells [15, 16].
Interactions of α5β1 with FN in the regulation of MMP in different tumor cells have been studied. Interaction of FN with α5β1 integrin induced human prostate cancer cell invasion , stimulated tyrosine phosphorylation of FAK in A1-F human foreskin fibroblast , increased MMP-1 in mammary epithelial cell , and increased MMP-9 expression via ERK and PI-3K pathways in lung carcinoma cells [20, 21] and via mitogen-activated protein kinase kinase (MEK1)-MAPK and PI-3K Akt pathways in ovarian cancer cells . FAK-transfected cells showed enhanced MMP-2 and MMP-9 mRNA expression in the presence of FN in human T-lymphocyte [23, 24]. Culture of MCF-7 in the presence of FN showed upregulation of pro-MMP-9 and activation of MMP-2 through α5β1 integrin-mediated signaling, indicating a clear correlation between tumor cell surface integrin and MMPs expression and activity . Moreover, there are also several interesting reports on integrin–MMP relationship [26–30].
Here, we report that culture of human cervical cancer (SiHa) cells on fibronectin-coated dish express and activate mainly pro-MMP-9 and moderately MMP-2 in SFCM via fibronectin–integrin signaling pathways involving FAK, ERK, PI-3K, ILK, and nuclear factor-kappa B (NF-κB).
Materials and methods
Minimal essential medium (MEM) and foetal bovine serum (FBS) were purchased from GIBCO, USA. FN, protein G agarose and protease inhibitor cocktail tablets were purchased from Roche (Germany). Gelatin–Sepharose 4B beads were purchased from GE Healthcare Bio-Sciences (Sweden) and gelatin was from SIGMA, USA. All primary antibodies, secondary antibodies, and NBT-BCIP (nitroblue tetrazolium chloride and 5 bromo 4 chloro 3 indoyl phosphates) were purchased from Santa Cruz (USA). T4 polynucleotide kinase (PNK) was purchased from Promega (USA). Primers and AP-1, Sp1 (Specificity protein 1), and NF-κB probes were synthesized by Operon (Germany). RNAqueous 4 PCR (Total RNA isolation kit) and Retroscript (RT–PCR Kit) were purchased from Ambion (USA). Actin, glyceraldehyde 3 phosphate dehydrogenase (G3PDH) were purchased from Genie (Bangalore).
The human cervical cancer cell line SiHa and human fibrosarcoma cell line HT-1080 were procured from National Centre for Cell Science (NCCS), Pune, India and was grown in MEM with 10% FBS in the presence of 5% CO2 at 37°C.
The SiHa cells were grown on FN (25 μg/ml)-coated and control (without FN) coverslips for 30 min and 1 h. Immunocytochemistry was preformed with anti-phospho-tyrosine or anti-FAK or anti-phospho-FAK or anti-ERK or anti-phospho-ERK or anti-ILK or anti-NF-κB antibody (1:1,000 dilution) according to previous method . Cells were observed under fluorescence microscope (40×).
The SiHa cells (600,000/1.5 ml) were grown on FN (25 μg/ml)-coated and control culture plates for 30 min and 1 h. Cells were extracted in cell extraction buffer (37.5 mM Tris, 75 mM Nacl, 0.5% Triton X-100, protease inhibitor cocktail). FAK, ERK, PI-3K, and ILK were immunoprecipitated from equal amount of cell lysate using respective antibodies and resultant immune complexes run on 7.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE). For ILK blot, equal amount of whole-cell lysate and for NF-κB blot, equal amount of cytoplasmic and nuclear extract (detailed methods below) were run on 7.5% SDS–PAGE. In order to observe co-precipitation of FAK and PI-3K with α5β1 integrin receptor, α5β1 integrin was immunoprecipitated from cell lysate using anti-α5 antibody (for FAK) or with anti-β1 antibody (for PI-3K) (1 μg/ml) and resultant immune-complex was subjected to western blot. Western blot was developed with anti-phospho-tyrosine (phospho-FAK, phospho-ERK, phospho-PI-3K, phospho-ILK) or anti-FAK or anti-ERK or anti-ILK or anti-PI-3K or anti-NF-κB antibody (1:1000) as described earlier . Bands were visualized using NBT/BCIP as substrate.
Enzyme-linked immunosorbent assay (ELISA)
The SiHa cells (600,000/1.5 ml) were grown on FN (25 μg/ml)-coated and control culture dishes (without FN) for 30 min and 1 h. ELISA of MMP-9 and MMP-2 in SFCM (75 μl) was performed as described earlier .
The SiHa cells (600,000/1.5 ml) were grown on FN-coated (25 μg/ml) and control dishes (without FN) for 30 min to 1 h or on type IV collagen or GRGDSP (25 μg/ml)-coated dish for 1 h. For inhibitor study, SiHa cells were pretreated with (E) or without (C) 20 μM NF-κB inhibitor in SFCM for 24 h before FN treatment. Gelatinases were separated from SFCM using Gelatin–Sepharose 4B beads (50 μl) and were subjected to gelatin zymography as described earlier .
Reverse transcription polymerase chain reaction (RT–PCR)
The RNA was extracted from SiHa cells (600,000/1.5 ml) grown on FN-coated (25 μg/ml) or control dish (without FN) for 1 h using RNAqueous 4PCR. RT–PCR was done with equal amounts of total RNA, using specific primers. Condition used for PCR: 40 cycles at 94°C (30 s), 58°C (MMP-9) or 56°C (MMP-2) (30 s), and 72°C (90 s); 25 cycles at 94°C (30 s), 58°C (Tissue inhibitor of metalloproteinase-1, TIMP-1) or 52°C (Tissue inhibitor of metalloproteinase-2, TIMP-2) or 60°C (FAK) (30 s), and 72°C (90 s) in DNA thermal cyclar. The forward and reverse sequence of primers: hMMP-9: 5′-TTGAGTCCGGCAGACAATCC-3′ and 5′-CCTTATCCACGCGAATGACG-3′; hMMP-2: 5′-GTATTTGATGGCATCGCTCA-3′ and 5′-CATTCCTGCAAAGAACACA-3′; TIMP-1: 5′-CACCCACAGACGGCCTTCTGCAAT-3′ and 5′-AGTGTAGGTCTTGGTGAAGCC-3′; TIMP-2: 5′-GTT TTG CAA TGC AGA TGT AG-3′ and 5′-ATG TGG AGA AAC TCC TGC TT-3′; hFAK: 5′-GCGCTGGCTGGAAAAAGAGGAA-3′ and 5′-TCGGTGGGTGCTGGCTGGTAGG-3′. G3PDH primers 5′-CGGAGTCAACGGATTTGGTCGTAT-3′ (forward) and 5′-AGCCTTCTCCATGGTGGTGAAGAC-3′ (reverse) were used to confirm equal loading. A volume of 25 μl of each PCR products was run on 2% agarose gel, and bands were visualized under UV.
Electro mobility shift assay (EMSA)
The SiHa cells (600,000/1.5 ml) were grown on FN (25 μg/ml)-coated or control plates (without FN) for 1 h. Cells were washed with hypotonic buffer (10 mM Hepes pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM PMSF, 0.5 mM DTT) by centrifugation (14,000 rpm × 10 min) and lysed in hypotonic buffer containing 0.5% Nonidet P-40 (NP-40). Cytoplasmic extract was separated by centrifugation; nuclear pellet was lysed in lysis buffer (20 mM Hepes, pH 7.9, 42 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM PMSF, 25% v/v glycerol), and nuclear extract was removed into storage buffer (10 mM Hepes pH 7.9, 50 mM KCl, 0.2 mM EDTA, 20% v/v glycerol, 0.5 mM PMSF, 0.5 mM DTT) after centrifugation. The probes of double-stranded oligonucleotides for NF-κB (5′-TGG AAT TCC CAG), Sp-1 (5′-CCT TCC GCC CCC), and AP-1(5′-CCT GAG TCA GCA) on the basis of human MMP-9 promoter sequence were annealed using annealing buffer (10 mM Tris pH 8, 50 mM NaCl, 1 mM EDTA) by heating at 90°C (2 min), slow cooling to room temperature for 60 min, and end labeled with [γ−32P] ATP using T4 PNK incubating at 37°C (1 h). Nuclear protein (10 μg) was incubated with labeled oligonucleotide probes using 2×-binding buffer [25 mM Hepes (pH 7.6), 1 mM EDTA, 0.5 mM DTT, 5 mM MgCl2, 75 mM KCl, 10% glycerol] at room temperature (30 min), and electrophoresed on 5% polyacrylamide gel using 0.5 × TBE buffer. Gel was dried and subjected to auto radiography.
Wound healing assay
The SiHa cells were kept in SFCM for 1 h with thrice change and were spread over FN (25 μg/ml) coated or uncoated culture dish and grown for 1 h. The monolayer was scratched with sterile tip, washing × 3 with SFCM. The wound was recoated with FN and cell migration was observed under microscope and photographed.
Expression and phosphorylation of protein
Expression, processing, and phosphorylation of FAK
Immunocytochemistry (Fig. 1b) showed prominent localization of FAK at focal adhesion sites and increased phosphorylation of FAK within 30 min (lane E1) upon culture on FN-coated dish compared to control cells (C); however, intensity of phospho-FAK was comparatively lower after 1 h (lane E2), although it still remained higher than control cells. Figure 1c showed appreciable increased expression of FAK and its processing to lower molecular weight forms (110, 74, 62, 50, 40 kDa) and increased expression of phospho-FAK (125, 110 kDa) in cells cultured on FN-coated dishes for 30 min (lane E1) and 1 h (lane E2) compared to control cells (lane C). Actin was used as internal control. RT–PCR (Fig. 1d) clearly indicated upregulation of FAK mRNA (2.5 fold) in SiHa cultured on FN-coated dish for 1 h (lane E) compared to control cells (lane C). G3PDH was used as internal control.
Co-precipitation of FAK and PI-3K with α5β1 integrin receptor
Expression and phosphorylation of ILK
Expression and phosphorylation of ERK and PI-3K
Expression and activity of MMP-9 and MMP-2
m-RNA expression of MMP-9, MMP-2, TIMP-1, and TIMP-2
RT–PCR (Fig. 5d) showed increase in MMP-9 (2.5 fold) and MMP-2 (2 fold) mRNA expression and decrease in TIMP-1(3 fold) and TIMP-2 (2 fold) mRNA expression in SiHa cells cultured on FN-coated culture dishes for 1 h (E) compared with control cells (C). G3PDH was used as internal control.
Translocation of NF-κB from cytosol to nucleus
Assay of binding activity of AP-1, Sp-1, and NF-κB proteins on MMP-9 promoter
Wound healing assay
In this article, we report extracellular matrix protein fibronectin–integrin mediated signaling in SiHa. In a previous article we reported that SiHa cells express α5β1 and bind to FN efficiently. The binding of SiHa cells was appreciably reduced when SiHa cells were treated with anti-α5 monoclonal antibody . Culture of SiHa cells on FN-coated dish induced FAK phosphorylation, processing and localization in focal adhesion sites indicating possible involvement of FAK in FN-integrin-mediated signaling. FAK-mediated signaling and processing were reported during binding of α2 integrin with Collagen IV . Involvement of ERK and PI-3K in FN–integrin mediated signaling has been reported . Our observation showed the possible involvement of ERK and PI-3K pathways in FN-integrin-mediated signaling. Co-precipitation of FAK and PI-3K with α5β1 indicated involvement of FAK and PI-3K in FN-integrin-mediated signaling. Earlier reports have revealed that inhibition or blockade of integrin receptors α5 inhibits activation of PI-3K. In contrast, PI-3K inhibition also reduced α5 integrin expression, indicating a direct link between PI3K activity and α5 integrin expression . ILK expression and activity are known to be increased in many cancer cells  and reported to induce AP-1-mediated MMP-9 expression . Culture of SiHa cells on FN coated surface showed appreciable increase of ILK with distinct punctuates in ILK accumulation sites, and its phosphorylation indicating possible role of ILK in upregulation of MMP-9. FN has been shown to modulate MMP-9 activity and gene expression [20, 22, 24] and MMP-2 activity [25, 31, 34–36] in different cancer cells. Co-expression of MMP-9 and MMP-2 was reported in many human cell lines, including cervical cancer cells . We observed that culture of SiHa cells on FN-coated surface increased mainly pro-MMP-9 activity with a moderate pro-MMP-2 activity in SFCM within 60 min than in control cells. RT–PCR showed increased MMP-9 and MMP-2 mRNA expression which was concomitant with respective increases in MMP-9 and MMP-2 expression and activity in zymogram. MMP-9 may be enzymatically active although proteolytically processed forms cannot be detected . The regulation of MMP activity depends on the levels of TIMPs . Our results showed that TIMP-1 and TIMP-2 mRNA expression was decreased during culture of SiHa cells on FN-coated dish. Down regulation of TIMP-1 and TIMP-2 may also be involved in increased MMP-9 and MMP-2 activity, respectively . Our group reported that culture of SiHa cells in the presence of soluble FN showed higher activity of MMP-2 and low activity of MMP-9 in SFCM . This difference in expression and activity of MMP-9 and MMP-2 in the presence of coated or soluble FN may indicate the difference in mode of FN–integrin-mediated signaling resulting differences in MMP-9 and MMP-2 expression and activity. Culture of SiHa cells with peptide-GRGDSP has no such response indicating role of intact FN molecule with its native conformation eliciting such response as observed by Munshi and Starck [27, 41]. Neither MMP-9 nor MMP-2 activity was observed on culture of SiHa cells on collagen IV (ECM protein)-coated plate. The faster migration of SiHa cells on FN-coated dish in wound healing assay may be due to increased activity of MMP-9 and MMP-2 in SFCM of experimental cells [42, 43].
The MMP-9 promoter contains multiple transcription factor-binding sites, including AP-1, Sp1, and NF-κB, which were responsive to various stimuli [20, 44]. MMP-9 was regulated at transcriptional level in different cell types . The MMP-9 promoter constructs used contain multiple transcription factor-binding sites for NF-kB, Sp1, and AP-1. In order to explore FN regulation of MMP-9 promoter activity, electrophoretic mobility shift assays were performed to identify the transcription factors regulated by FN. As shown in Fig. 7, SiHa cells exposed to FN (25 μg/ml) for 1 h showed significant increases in AP-1 (B), Sp1 (C), and NF-κB (D) DNA binding activities as compared with untreated controls. The increased DNA binding activity of the transcription factors in FN-treated cells as compared to basal level binding in untreated control set indicates that FN might induce MMP-9 transactivation. Collaboration of several transcriptional elements was found necessary for induction of MMP-9 promoter activity [46–49]. We showed NF-κB translocation from cytosol to nucleus during culture of SiHa cells on FN. When SiHa cells were treated with NF-κB inhibitor the pro-MMP activity was appreciably reduced. These results strongly indicated the involvement of NF-κB in α5β1 integrin-mediated signaling and induction of MMP-9 expression and activity.
In conclusion, our study strongly indicates that culture of SiHa cells on fibronectin-coated surface initiates the fibronectin–integrin mediated cell signaling cascades involving FAK, ILK, ERK, PI-3K, and NF-κB inducing appreciable expression and activation of pro-MMP-9 and moderate change of pro-MMP-2 activity. Our findings may help to understand the role of fibronectin–integrin-mediated signaling and its effect in tumor development, which may also have clinical and therapeutic potentials.
We acknowledge the financial support and inspiration received from Dr. Jaydip Biswas, Director, CNCI, and the financial support from Department of Science and Technology, New Delhi (Grant No. SR/SO/HS-59/2004).