ADAM17 Mediates MMP9 Expression in Lung Epithelial Cells

The purposes were to study the role of lipopolysaccharide (LPS)-induced tumor necrosis factor (TNF)-α/nuclear factor-κB (NF-κB) signaling in matrix metalloproteinase 9 (MMP9) expression in A549 cells and to investigate the effects of lentivirus-mediated RNAi targeting of the disintegrin and metalloproteinase 17 (ADAM17) gene on LPS-induced MMP9 expression. MMP9 expression induced by LPS in A549 cells was significantly increased in a dose- and time-dependent manner (p<0.05). Pyrrolidine dithiocarbamate (PDTC) and a TNFR1 blocking peptide (TNFR1BP) significantly inhibited LPS-induced MMP9 expression in A549 cells (p<0.05). TNFR1BP significantly inhibited LPS-induced TNF-α production (p<0.05). Both PDTC and TNFR1BP significantly inhibited the phosphorylation of IκBα and expression of phosphorylation p65 protein in response to LPS (p<0.05), and the level of IκBα in the cytoplasm was significantly increased (p<0.05). Lentivirus mediated RNA interference (RNAi) significantly inhibited ADAM17 expression in A549 cells. Lentivirus-mediated RNAi targeting of ADAM17 significantly inhibited TNF-α production in the supernatants (p<0.05), whereas the level of TNF-α in the cells was increased (p<0.05). Lentiviral ADAM17 RNAi inhibited MMP9 expression, IκBα phosphorylation and the expression of phosphorylation p65 protein in response to LPS (p<0.05). PDTC significantly inhibited the expression of MMP9 and the phosphorylation of IκBα, as well as the expression of phosphorylation p65 protein in response to TNF-α (p<0.05). Lentiviral RNAi targeting of ADAM17 down-regulates LPS-induced MMP9 expression in lung epithelial cells via inhibition of TNF-α/NF-κB signaling.


Introduction
Chronic obstructive pulmonary disease (COPD) is a leading cause of chronic morbidity and mortality that imposes a substantial economic and social burden worldwide. The permanent destruction of alveolar walls, which results in airspace enlargement, as well as the loss of elastic recoil and a decrease in surface area for gas exchange and lung hyperexpansion are hallmarks of COPD [1]. Protease-antiprotease imbalance is thought to play a key role in the pathogenesis of COPD. Matrix metalloproteinases (MMPs) are a family of proteolytic enzymes, which belong to the metazincin family and share the conserved zinc-binding motif in their catalytic active site [2]. MMPs not only actively participate in remodeling the extracellular matrix (ECM) by degrading certain constituents, but also are important regulators of extracellular tissue signaling networks, which may regulate cell proliferation, migration, differentiation, apoptosis and angiogenesis [2][3][4]. The ultimate impact of MMPs on ECM degradation may be regulated at several levels, including gene transcription, proenzyme activation and inhibition of active enzymes [5]. MMP9 also known as gelatinase B is involved in the degradation of elastin, aggrecan, and type IV, V and VII collagen [4][5][6][7][8], which may contribute to the development of COPD [9]. An increase in MMP9 has been detected respectively in the sputum and bronchoalveolar lavage fluid (BALF) of patients with COPD [10,11]. Experimental animal models have also shown that over-expression of MMP9 induces pathological changes that are similar to those associated with emphysema, including airspace enlargement and loss of alveolar elastin in mice [6]. Moreover, higher serum concentrations of MMP9 have been linked to airway obstruction and COPD progression [12,13]. However, the complex and highly orchestrated interactions between inflammatory cells, inflammatory cytokines and MMP9 are not well understood. Furthermore, the mechanism that regulates MMP9 expression in lung has not previously been clarified.
Tumor necrosis factor (TNF)-a is a potent pro-inflammatory cytokine that exerts pleiotropic effects on various cell types and plays a critical role in the pathogenesis of chronic inflammatory diseases, such as COPD [14]. In the classical nuclear factor-kB (NF-kB) signaling pathway, stimulation of TNF-a activates the inhibitor of NF-kB kinase (IKK) signalosome, which leads to the phosphorylation of inhibitor of NF-kB (IkB) on 2 conserved Nterminal serine residues. Phosphorylated IkB is then ubiquitinated and subsequently degraded by the S26 proteasome [15]. Finally, NF-kB is activated and translocated into the nucleus where it binds to a decameric consensus motif and facilitates the transcription of target genes. TNF-a is initially synthesized as a transmembrane protein (mTNF-a) with a molecular mass of 26 kDa that is cleaved to yield a 17 kDa soluble product (soluble TNF-a). A disintegrin and metalloproteinase 17 (ADAM17), also called TNF-a converting enzyme (TACE), plays a vital role during ectodomain shedding of TNF-a [16,17]. Several studies have shown that NF-kB signal pathway mediates MMP9 expression that is induced by TNF-a [18,19]. Therefore, we hypothesized that the ADAM17/TNF-a/NF-kB signaling pathway mediates MMP9 expression in lung epithelial cells.
RNA interference (RNAi) is a powerful research tool for studying gene function in vitro and in vivo [20]. Lentiviral vectors can efficiently transduce both dividing and nondividing cells to established sustained transgene expression. These properties make lentiviral vectors attractive vehicles for the delivery of small interfering RNA genes into mammalian cells [21]. Therefore, in the present study, we constructed lentiviral vectors targeting the human ADAM17 gene, investigated whether TNF-a signaling mediated LPS-induced MMP9 expression, and explored the effects of lentiviral RNAi-mediated knockdown of ADAM17 on TNF-a/NF-kB signaling and MMP9 expression in A549 lung epithelial cells treated with LPS.

Construction and production of lentiviral expression vectors
The human ADAM17 mRNA sequence (GenBank accession number: NM_003183) was used to determine suitable siRNA target sequences and CCTATGTCGATGCTGAACAAA was selected. The siRNAs were converted into short hairpin RNA (shRNA) with a stem-loop-stem conformation and ClaI and MluI (Trono Laboratory, University of Geneva, Switzerland) restriction sites were added at the 59 and 39 ends, respectively. DNA oligonucleotides chemically synthesized by Sangon Biotech Co., Ltd. (Shanghai, China) were annealed and then inserted into pLVTHM vectors following double digestion with ClaI and MluI and ligation with T4 DNA ligase (NEB, Ipswich, MA, USA). The recombinant vectors were transformed into DH5a -competent Escherichia coli cells (Invitrogen, Carlsbad, CA, USA). The orientation of the inserted shRNA cassettes was verified by restriction enzyme analysis and DNA sequencing. A negative control (NC) siRNA sequence (TTCTCCGAACGTGTCACGT) was used as a control for ADAM17 siRNA.

Lentivirus production and transduction
293T cells (Purchased from Shanghai Institute of Biology, Chinese Academy of Sciences) were cultured in DMEM supplemented with 10% FBS. When cell fusion reached 80%, 20 mg recombinant pLVTHM vectors and packaging helper plasmids, including 10 mg pMD2G-VSVG, 10 mg pMDlg-pRRE and 10 mg pRsv-REV (Trono Laboratory, University of Geneva, Switzerland), were co-transfected into 293T cells with calcium phosphate. The medium was replaced with fresh culture medium 12 h after transfection. The cultured supernatants were collected and centrifuged at 8006g for 7 min at 4uC 48 h post-transfection to remove cell debris. The supernatants were then filtered through a 0.45 mm pore filter prior to ultra-centrifugation at 500006g for 90 min at 4uC. Viral particles were precipitated in ice-cold PBS re-suspension solution. Finally, the viral particles, LV-ADAM17-shRNA and LV-NC-shRNA, were stored at 280uC. The viral titer was detected via infection of 293T cells and subsequent flow cytometric analysis. The titer of the recombinant virus was 2.166108 TU/ml.

Enzyme-linked immunosorbent assay (ELISA)
The amount of TNF-a in the supernatants was measured with ELISA kits (R&D Systems, Minneapolis, MN, USA), which were used in accordance with the manufacturer's instructions. Briefly, the supernatants were coated onto the 96-well plates and incubated overnight at 4uC. After blocking for 1 h at room temperature with blocking solution (1% FBS and 0.05% Tween-20 in PBS), monoclonal anti-human TNF-a was added to each well and incubated for 1 h at room temperature. After washing with 0.05% Tween-20 in PBS, peroxidase-conjugated immunopure goat anti-rabbit IgG (H+L) (Catalog# 31460; Pierce, Rockford, IL, USA) was added and incubated for 1 h at room temperature. A color reaction was developed with tetramethyl-benzidine for 10 min at room temperature. Following the addition of a stop solution, absorbance at 450 nm was measured in a microtiter plate reader.

Western blotting
Nuclear and cytoplasmic proteins were extracted from the cells with NE-PER Nuclear and Cytoplasmic Extraction Reagent (Catalog # 78833; Pierce, Rockford, IL, USA) according to the manufacturer's instructions. Protein concentration was measured with a protein assay kit (Micro BCA; Pierce, USA). Equal amounts of cell lysates containing 10 mg protein were incubated for 5 min in boiling water. Proteins were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and then electroblotted onto Hybond-ECL nitrocellulose membrane. The membrane was blocked with TBS solution containing 5% skim milk at room temperature for 1 h. Membranes were then incubated in primary antibodies for 1 h at room temperature on an orbital shaker. The membranes were then washed 5 times, for 5 min each, in Trisbuffered saline Tween (TBS-T) and incubated in diluted peroxidase-conjugated immunopure goat anti-rabbit IgG (H+L)(Catalog#31460; Pierce, USA) for 1 h at room temperature on an orbital shaker. After the membranes were again washed 5 times for 5 min each in TBS-T, the proteins were visualized with an enhanced chemiluminescence solution (Amersham Pharmacia biotech, Buckinghamshire, UK). The images were developed on X-ray film and the band densities were analyzed with a UVP-GDS8000 gel analysis system. The same membranes were stripped and blotted with anti-b-actin antibodies, which provided a loading control. The primary antibodies were obtained from the following sources: MMP9 (G657) Antibody (Catalog#2270), IkBa Antibody which can bind to IkBa that is not phosphorylated at Ser32/Ser36 (Catalog #9242), Phospho-IkBa (Ser32) (14D4) Rabbit mAb (Catalog#2859), TNF-a rabbit polyclonal antibody (Cat-alog#3707S) and phospho-NF-kB p65 (p-p65) (Ser536) Mouse mAb (Catalog# 3036S) from Cell Signaling Technology (Danvers, Massachusetts, USA); and b-actin (N-21) antibody (Cata-log#sc130656) from Santa Cruz Biotechnology (Santa Cruz, California, USA). Protein markers (Catalog #7727) were obtained from Cell Signaling Technology (Danvers, USA).

Statistical analyses
Data were analyzed for statistical significance via one-way analysis of variance (ANOVA), paired-sample t-tests, and multiple comparisons in ANOVA were analyzed via the Student-Newman-Keuls test. Data were expressed as the mean 6 SD. Values of p,0.05 were considered statistically significant.

LPS induced both dose-and time-dependent expression of MMP9 mRNA
When A549 cells were stimulated for 24 h with LPS, at concentrations ranging from 0.1-100 mg/l, MMP9 mRNA expression was induced by LPS in a dose-dependent manner

LPS-induced MMP9 expression is mediated by the TNF-a/ NF-kB signaling pathway in A549 cells
The expression of TNF-a protein in the supernatants was induced by LPS in a dose-and time-dependent manner in A549 cells (p,0.05; Fig. 2A and 2B). When the cells were pretreated for 30 min with TNFR1BP, and then stimulated with LPS for 24 h, the expression of TNF-a protein was significantly decreased (p,0.05; Fig. 2C). The phosphorylation of IkBa and p65 protein expression was induced by stimulation with LPS. But the level of IkBa (unphosphorylated) in the cytoplasm was significantly decreased under the stimulation of LPS (p,0.05; Fig. 2G). When the cells were pretreated for 30 min with TNFR1BP and PDTC, and then stimulated for 24 h by LPS, the level of the IkBa in the cytoplasm was significantly increased (p,0.05; Fig. 2G), and both the phosphorylation of IkBa and expression of p-p65 protein were significantly inhibited (p,0.05; Fig. 2F). LPS-induced MMP9 expression was also markedly inhibited by TNFR1BP and PDTC (p,0.05; Fig. 2D and 2E).
Effects of lentivirus-mediated ADAM17 RNAi on the expression of MMP9 and p-p65 protein, and the phosphorylation of IkBa, in response to LPS A549 cells were infected with LV-ADAM17-shRNA or LV-NC-shRNA for 72 h (Fig. 3), and lentivirus-mediated ADAM17 RNAi significantly inhibited ADAM17 mRNA expression in A549 cells (p,0.05; Fig. 4A). When A549 cells were infected with LV-ADAM17-shRNA or LV-NC-shRNA for 72 h, and then stimulated for 24 h with LPS at 100 mg/l, the production of TNF-a protein in the supernatants was significantly decreased by LV-ADAM17-shRNA (p,0.05; Fig. 4B), on the contrary the levels of TNF-a protein in A549 cells were significantly increased (p,0.05; Fig. 4C). MMP9, phosphorylated IkBa, as well as p-p65 were markedly inhibited by LV-ADAM17-shRNA (p,0.05; Fig. 4D, 4E, 4F). When the cells were infected with LV-ADAM17-shRNA for 72 h, and then stimulated for 24 h with LPS, the level of IkBa (unphosphorylated) in the cytoplasm was significantly increased (p,0.05; Fig. 4G). The results also indicated that LV-NC-shRNA had no effect on the production of TNF-a in response to LPS (p.0.05; Fig. 4B). LPS-induced MMP9 expression, IkBa phosphorylation and p-p65 expression could not be inhibited by LV-NC-shRNA in A549 cells (p.0.05; Fig. 4D, 4E, 4F).

Effects of lentiviral ADAM17 RNAi on MMP9 expression, IkBa phosphorylation, and p-p65 protein expression induced by TNF-a
When A549 cells were pretreated for 30 min with 50 mg/l PDTC, and then stimulated for 24 h with TNF-a at 1 mg/l, PDTC markedly inhibited the phosphorylation of IkBa and p65 in response to TNF-a in A549 cells (p,0.05; Fig. 5C), and the level of IkBa (unphosphorylated) in the cytoplasm was significantly increased (p,0.05; Fig. 5D) Correspondingly, TNF-a-induced MMP9 expression was markedly inhibited by PDTC (p,0.05; Fig. 5A, 5B). When A549 cells were infected with recombinant lentivirus for 72 h, and then stimulated for 24 h with TNF-a at 1 mg/l, LV-ADAM17-shRNA and LV-NC-shRNA had no effect on the expression of MMP9, the level of unphosphorylated IkBa and the phosphorylation of IkBa and p65 in response to TNF-a (p.0.05; Fig. 5E, 5F, 5G, 5H).  Discussion MMP9 is not produced by resident cells in the normal lung; however, in response to various forms of stimulation, MMP9 is expressed, produced, and secreted by many inflammatory cells and some airway structural cells [22,23]. LPS, or endotoxin, is a predominant, integral structural component of the outer membrane of gram-negative bacteria and one of the most potent microbial initiators of inflammation [24]. In the present study, the expression levels of MMP9 mRNA and MMP9 protein were significantly increased by the stimulation of LPS in A549 lung epithelial cells. The transcription factor NF-kB is a dimeric complex composed of members of the Rel protein family, which includes p65 (RelA), p105/p50, p100/p52, RelB, c-Rel, and the viral oncoprotein (v-Rel). In its classic form, NF-kB is a dimer that consists of the transcriptionally inactive p50 subunit and the p65/ RelA (p65) subunit. IkB is responsible for regulating the DNAbinding activity and nucleo/cytoplasmic distribution of NF-kB [25]. The most well characterized IkB protein is IkBa [25,26]. Association of IkBa with NF-kB disrupts DNA binding and masks the nuclear localization signals (NLSs) located in the C-terminal region of the Rel homology domain (RHD). Masking the NLSs is believed to impede the nuclear translocation of NF-kB, which results in the retention of NF-kB in the cytoplasm where it cannot mediate any transcriptional effects. Any of several extracellular stimuli can activate NF-kB by initiating a signal transduction pathway that leads to phosphorylation, ubiquitination and, ultimately, degradation of IkBa [25]. Once IkBa is degraded, the NLSs within the RHD are unmasked, allowing NF-kB to enter the nucleus and initiate transcription of target genes [27]. PDTC, which can protect IkBa from degradation, is a special inhibitor of NF-kB [28]. TNFR1BP can bind to TNF-a and block the biological activities mediated by TNF-a [29]. In the present study, the IkBa antibody (binding to IkBa that is not phosphorylated at Ser32/Ser36) was used to detect the levels of IkBa protein in the cytoplasm. The results showed that the level of IkBa which is not phosphorylated at Ser32/Ser36 in the cytoplasm was significantly decreased under the stimulation of LPS. When the cells were pretreated for 30 min with TNFR1BP and PDTC, and then stimulated for 24 h by LPS, the level of IkBa in the cytoplasm was significantly increased. In the present study, TNFR1BP inhibited the expression of TNF-a and the phosphorylation of IkBa and p65 in response to LPS. Furthermore, LPS-induced the phosphorylation of IkBa and p65 was also inhibited by PDTC. Accordingly, LPS-induced MMP9 expression was markedly decreased when the cells were pretreated with TNFR1BP and PDTC. Therefore, the TNF-a/NF-kB signaling pathway mediated LPS-induced MMP9 expression in A549 lung epithelial cells.
The ADAMs are a family of transmembrane and secreted proteins with a variety of functions, including proteolytic cleavage of cell surface molecules, cell fusion, cell adhesion and intracellular signaling [30]. ADAM17, a member of ADAM family, has been described as ''a signaling scissor''. The most well-known function of catalytically active ADAM17 is to cleave the ectodomains of various transmembrane proteins, such as growth factors, receptors and their ligands, cytokines, and cell adhesion molecules; therefore, ADAM17 is an important regulator of almost every cellular event [31,32]. Though MMP7, MMP14, MMP17 and ADAM10 are also involved in the ectodomain shedding of TNF-a, the specificity constants for cleavage of TNF-a are approximately 100-1000-fold lower than ADAM17 [17]. Therefore, ADAM17 is a key sheddase in the proteolytic cleavage of the extracellular domain of mTNF-a [33]. RNAi has received attention for being one of the most powerful tools for reverse genetics in the postgenome era. Several groups have developed vector-based siRNA expression systems that can induce RNAi in living cells. With lentiviral vectors for shRNA expression, high transduction efficiencies can be achieved, which minimizes the effects of clonal selection of phenotypically diverse cells [34]. Therefore, ADAM17 shRNA expression vectors were constructed in the present study, and ADAM17 expression was significantly inhibited by lentivirusmediated ADAM17 RNAi in A549 cells. Lentiviral ADAM17 RNAi inhibited the expression of TNF-a protein in the supernatants, whereas the level of TNF-a in the cells were increased, which indicated that ADAM17 mediated the shedding of TNF-a. In the present study, lentiviral ADAM17 RNAi inhibited the phosphorylation of IkBa and p65 in response to LPS, and the level of IkBa which is not phosphorylated at Ser32/Ser36 in the cytoplasm was significantly increased. Accordingly, LPS-induced MMP-9 expression was markedly decreased. In contrast, lentiviral ADAM17 RNAi had no effect on the TNF-a-induced phosphorylation of IkBa and p65. Furthermore, ADAM17 RNAi did not inhibit TNFa-induced MMP9 expression in A549 lung epithelial cells. These results indicated that ADAM17 RNAi does not directly inhibit downstream TNF-a signaling, and that ADAM17 mediates LPSinduced MMP9 expression in lung epithelial cells by regulating soluble TNF-a.
In conclusion, TNF-a/NF-kB signaling plays an important role in LPS-induced MMP9 expression in lung epithelial cells. Furthermore, lentiviral RNAi targeting of ADAM17 downregulated LPS-induced MMP9 expression via inhibition of TNFa/NF-kB signaling, which provides a potential target for therapeutic intervention in COPD.