Endoproteolytic Activation of αv Integrin by Proprotein Convertase PC5 Is Required for Vascular Smooth Muscle Cell Adhesion to Vitronectin and Integrin-Dependent Signaling
Background— Integrins play an important role for vascular smooth muscle cell (VSMC) migration during the development of atherosclerosis and restenosis. Integrin αv-subunit consists of disulphide-bound 125-kDa heavy and 25-kDa light chains, which are generated by endoproteolytic cleavage. This type of activation requires the presence of suitable proprotein convertases (PCs). Based on ex vivo and in vitro data, the PC5 isozyme has been suggested to be the major integrin convertase. We have recently demonstrated that PC5 is upregulated during vascular remodeling in rodents, colocalizing with αv in VSMCs. The aim of this study was to investigate the activation of αv by PCs in VSMCs and its consequences for αv-dependent cell functions.
Methods and Results— Immunoblotting demonstrated that inhibition of PC activity by the specific pharmacological inhibitor dec-CMK inhibits αv cleavage in VSMCs. These results were confirmed using PC5-specific antisense oligonucleotides. PC5-antisense oligonucleotides and dec-CMK inhibited VSMC adhesion to the αvβ3/β5 ligand vitronectin (both P<0.05). Furthermore, PC5-asODNs inhibited VSMC migration on vitronectin-coated wells (P<0.05). Inhibition of PC activity and consequently αv cleavage inhibited the adhesion-dependent focal adhesion kinaseY397-autophosphorylation and subsequent Akt activation, whereas phosphorylation of extracellular signal-regulated kinase 1/2 was not affected. In human endarterectomy lesions, PC5 colocalized with αv integrin in VSMCs in the atherosclerotic plaques.
Conclusions— The present study demonstrates that αv endoproteolytic activation is necessary for integrin-mediated adhesion and migration as well as signaling and requires PC5 in VSMCs. The colocalization of PC5 and αv in human carotid plaques indicates that PC5 might play a key role for αv activation in vivo.
Received June 3, 2003; de novo received August 13, 2003; revision received October 2, 2003; accepted October 6, 2003.
Proliferation and migration of vascular smooth muscle cells (VSMCs) are key elements in atherosclerosis and restenosis.1 Both require integrins, which link the extracellular matrix (ECM) with the cytoskeleton, mediating mechanotransduction2 and bidirectional signaling.3 Activation of nonreceptor protein tyrosine kinases, such as focal adhesion kinase (FAK),4 are involved in transducing integrin signaling, which lack intrinsic tyrosine kinase activity themselves.3 Engagement of integrins involves activation of additional downstream signaling pathways, such as Ras/MEK and PI3-kinase/Akt,3 as well as conformational changes5 and integrin/growth factor receptor clustering at focal adhesion sites.6
The integrins αvβ3 and αvβ5 are upregulated in the neointima after vascular injury7,8 and are expressed in VSMCs in atherosclerotic lesions.9,10 Both integrins are required for VSMC adhesion and migration to vitronectin7,8 and osteopontin11 in vitro. Targeting αv integrins has been demonstrated to prevent neointima formation,7,8,12 emphasizing their critical role in atherosclerosis and restenosis.
Like other arginine-glycine-aspartate (RGD)-recognizing α-integrins,5 the αv subunit consists of disulfide-bound chains of 125 and 25 kDa, which are generated by endoproteolytic cleavage.13 Cleavage of α-subunits can regulate integrin functions14,15 and requires the presence of suitable subtilisin/kexin-like proprotein convertases (PCs). These enzymes are responsible for the activation of a variety of precursor proproteins to become biologically active.16 Based on experiments involving the overexpression of individual PCs (including PC1, PC2, PACE4, furin, PC5, and PC7) and αv, PC5 has been suggested to be the major αv convertase.17 We have recently reported the upregulation of PC5 in PDGF-BB–stimulated proliferating VSMCs, whereas furin was unaffected by this growth factor.18 Furthermore, PC5 and αv are coordinately upregulated and colocalize in neointima VSMCs after balloon injury in rodents,19 supporting the concept that PC5 may play a functional role for αv activation in VSMCs.
In addition to PC-dependent cleavage, an alternative pathway for αv cleavage involving the membrane-bound matrix metalloproteinase MT1-MMP has been reported recently.20 Thus, the present study was done to identify the αv processing enzyme and to investigate the function of αv endoproteolytic cleavage in VSMCs.
Cell culture media and materials were from GIBCO. Rat vitronectin and type I collagen were from Sigma. The broad-spectrum hydroxamate class MMP-inhibitor GM6001 (Ilomastat) was purchased from Chemicon, and the furin-like PC-inhibitor decanoyl-Arg-Val-Lys-Arg-chloromethylketone (dec-CMK) was from Bachem. Phosphorothioate-conjugated antisense oligodesoxynucleotiodes (asODNs) were costume made and purified (>98%) by HPLC (TIB-MOLBIOL). All other chemicals were from Sigma. The following antibodies were used: α-smooth muscle actin (α-SMA; Dako), rabbit antibodies to PC5 and furin (both IRCM), αv N-terminal sequence (VNR139; Calbiochem), and αv C-terminal sequence (AB1930; Chemicon). The MT1-MMP antibody (CL1MMP14) was from Cedarlane, and CD68 was from Dako. Induction of signaling pathways was detected with phospho-specific antibodies, recognizing FAK when phosphorylated at tyrosine 397 (Upstate), ERK1/2 MAP-kinase when phosphorylated at threonine 183, and tyrosine 185 (Promega) and Akt when phosphorylated at serin 473 (Cell Signaling). Membranes were reblotted with antibodies to total FAK, Akt, and ERK1/2, purchased from Santa Cruz (FAK) and Cell Signaling (Akt; ERK1/2). The αvβ5 blocking antibody P1F6 was from GIBCO.
Culture of rat VSMCs was done as described.18 Subconfluent cells of passages 3 through 6 were used and rendered quiescent by serum starvation (0% FCS) overnight. Pharmacological inhibitor experiments were done in the presence of 2.5% FCS. The αvβ5 blocking antibody (P1F6) was used as described by others.10,11 Cell viability was assessed by trypan-blue staining. All experiments were done in triplicate with different preparations of VSMCs.
AsODNs were used to silence PC5 protein levels. The following PC5-asODNs were used: 5′TCCTTACTCCGTCCAC3′ (PC5-asODN-1), 5′GCAACTTGCCAGAGCAT3′ (PC5-asODN-2), and 5′CCAGTCCATGGTCCCGA3′ (PC5-asODN-3). Specificity of PC5-asODN-2 was additionally assessed with sense (5′ATGCTCTGGCAAGTTGC3′) and scrambled (5′CATGACTACGCTCAGAG3′) ODNs. A 19-mer asODNs targeting α1-collagen was used as additional control.21 Transfection was done with the help of OligofectAMINE (GIBCO), used according to the manufacturer. Briefly, subconfluent VSMCs were growth arrested overnight. Cells were then incubated in OptiMEM (GIBCO) supplemented with ODNs for 6 hours. Then 0.2% FCS was added and cells were maintained for an additional 18 hours. After this, the medium was replaced by 2.5% FCS-DMEM supplemented with ODNs. Successful transfection was determined with an FITC-conjugated PC5-asODN-2 by immunofluorescence.
Immunoblotting was performed as described18 using 8% or 10% SDS-PAGE with or without reduction with β-mercaptoethanol. Semiquantitative densitometry was done using the NIH program 1.62 and is expressed in arbitrary units.
Supernatants were electrophoresed in 10% SDS-PAGE containing 0.1% gelatin. Briefly, gels were renaturated by exchanging SDS to Triton X-100 (2.5%). Gels were then incubated for 24 to 48 hours at room temperature in activation buffer (50 mmol/L Tris-base, pH 7.6; 5 mmol/L CaCl2; 0.2 mol/L NaCl; and 0.02% Brij) and stained with Coomassie staining solution (0.5% Coomassie R250; 30% MeOH; 10% acetic acid) overnight, followed by destaining (50% MeOH and 10% acetic acid).
VSMCs were serum-starved for 48 hours and then transfected with asODNs as described above, followed by stimulation with 5% FCS. BrdU was added after 24-hour stimulation for another 8 hours, and supernatants were then analyzed with a commercially available ELISA (Boehringer).
Microtiter plates (96-wells) were coated with rat vitronectin (10 μg/mL) or rat type I collagen (20 μg/mL) for 16 hours at 4°C. Plates were then washed with PBS and incubated with PBS containing 1% BSA for 60 minutes at room temperature to block nonspecific binding. VSMCs (30 000/well) were seeded and allowed to adhere for 2 hours at 37°C. After washing, attached cells were fixed with 4% paraformaldehyde (pH 7.5), stained with 0.5% toluidine blue, and lysed with 1% SDS. Absorbance was measured at 590 nm.
FCS (10%)-directed VSMC migration was examined in transwell cell culture chambers using gelatin-coated (0.2%) or vitronectin-coated (10 μg/mL) polycarbonate membranes with 8-μm pores. Serum-starved VSMCs were transfected with asODNs or incubated with pharmacological inhibitor for 24 hours in serum-deprived DMEM and then used in experiments. The number of VSMCs per high-power field (magnification ×320) that migrated after 4 hours to the lower surface of the filters was determined microscopically. Four randomly chosen high-power fields were counted per filter.
Ten human carotid endarterectomy specimens (Stary stage 4 or above) were investigated. After primary antibody incubation overnight at 4°C, single labeling (PC5, 1:500; α-SMA, 1:50; αv, 1:250; AB1930) was revealed with DAB using the ABC Histostain-Plus kit (Zymed). Double-immunocytochemical stainings of PC5 with α-SMA were done by combining alkaline phosphatase (Vector red) and horseradish peroxidase (DAB) as described.19 Colocalization of polyclonal antibodies to PC5 and αv was done using secondary biotinylated goat anti-rabbit and tertiary horseradish peroxidase Avdin-D antibodies (Vector) with Novared substrate (Vector) for αv stainings first, followed by PC5 staining with secondary goat anti-rabbit alkaline phosphatase antibody using BCIP/NBT (Vector). Specificity controls were done by omission of the first antibody or incubation with nonimmune IgGs.
ANOVA and paired or unpaired t test were performed for statistical analysis as appropriate. Statistical significance was designated at P<0.05. Values are expressed as mean±SD.
Proprotein Convertase Activity Is Required for αv Activation in VSMCs
To investigate the contribution of PCs and MT1-MMP to αv activation, pharmacological inhibitors were used. In controls, the antibody targeting the N-terminal sequence of αv (VNR139) revealed a 150-kDa noncleaved αv on nonreducing gels. However, under reducing conditions, when disulphide-bound heavy chains (125 kDa) and light chains (25 kDa) are separated, a cleaved αv 125-kDa heavy chain is detected (Figure 1A). The PC-inhibitor dec-CMK22 (50 μmol/L; 24 hours) significantly inhibited αv cleavage, demonstrated by the presence of the noncleaved 150-kDa αv on reducing gels (Figure 1A). In contrast, the MMP-inhibitor GM6001 (50 μmol/L; 24 or 48 hours), recently reported to inhibit αv cleavage,20 had no effect (Figure 1A), even though mature MT1-MMP is present (Figure 1B). To assess that the concentration and incubation time of GM6001 was sufficient to inhibit MT1-MMP, conditioned medium was subject to zymography, demonstrating inhibition of pro-MMP-2 activation by GM6001, whereas dec-CMK had no effect (Figure 1C).
Because dec-CMK inhibits furin and PC5, specifically designed asODNs were used to silence PC5, the major integrin convertase.17 Different PC5-asODNs were tested at different concentrations (0.5, 1, and 5 μmol/L), revealing a concentration-dependent inhibition of FCS-induced PC5 levels by PC5-asODNs-1 and -2, whereas PC5-asODNs-3 was inactive (online Figure, panel A, available at http://www.circulationaha.org). PC5-asODNs-2 was the most potent sequence and was therefore used in additional experiments. A control asODNs targeting α1-collagen21 had no effect. Transfection efficiency was monitored by immunofluorescence, demonstrating nuclear uptake of asODNs with the help of transfection medium (online Figure, panel B). At a concentration of 1 μmol/L, PC5-asODNs reduced PC5 protein levels by approximately 60% (#P<0.05 versus controls, online Figure, panel C). Levels of furin were not affected by PC5-asODNs (online Figure, panel C). Because of the incomplete transfection of primary cells (the percentage of VSMCs displaying nuclear uptake of PC5-asODNs-2 was approximately 60%), no complete inhibition of αv cleavage could be achieved. Still, PC5-asODNs-2 significantly inhibited αv activation, evident by the noncleaved 150-kDa αv on reducing gels (Figure 2A). Inhibition of αv activation via inhibition of PC activity was additionally confirmed with an anti-αv integrin antibody (AB1930) recognizing the C-terminal fragment. PC5-asODNs-2 (1 μmol/L; 24 hours) and dec-CMK (50 μmol/L; 24 hours) caused a significant decrease of cleaved αv light chain (25 kDa) on reducing SDS-PAGE (#P<0.05 versus controls), whereas GM6001 (50 μmol/L; 24 hours) had no effect (Figures 2B and 2C).
Proprotein Convertase PC5 Is Required for VSMC Adhesion and Migration on Vitronectin
Silencing PC5 with asODNs-2 (1 μmol/L; 24 hours) or inhibition of PC activity with dec-CMK (50 μmol/L; 24 hours) markedly inhibited VSMC adhesion to the αv-ligand vitronectin (10 μg/mL) (Figure 3A, #P<0.05 versus controls). This inhibition was comparable to the αvβ5 blocking antibody P1F6 (25 μg/mL; #P<0.05 versus nonspecific IgGs). In contrast, neither PC5-asODNs-2 nor dec-CMK significantly inhibited adhesion to type I collagen (20 μg/mL) (Figure 3B). GM6001 (50 μmol/L; 24 hour) did not inhibit adhesion on either of the matrices (Figures 3A and 3B). PC5-asODNs-2 also inhibited VSMC migration on gelatin-coated (0.2%) membranes in a concentration-dependent (0.5 and 1 μmol/L) manner (Figure 4A; #P<0.05 versus controls). Similar results were obtained with vitronectin (10 μg/mL) (#P<0.05 versus controls; Figure 4B), which were comparable to P1F6 (25 μg/mL; #P<0.05 versus nonspecific IgGs). In contrast, FCS-induced VSMC DNA synthesis was not affected by PC5-asODNs-2 (Table).
Activation of αv by Proprotein Convertases Is Required for Integrin-Dependent Signaling
To investigate the requirement of αv cleavage by PCs for integrin-dependent signaling in VSMCs, cells were serum-starved with or without dec-CMK (50 μmol/L) for 24 hours and then detached and allowed to adhere to vitronectin (10 μg/mL) for 45 minutes. On adhesion to vitronectin, phosphorylation of FAK (Figure 5A), Akt (Figure 5B), and ERK1/2 (Figure 5C) was evident. Inhibition of αv-cleavage with dec-CMK resulted in the inhibition of adhesion-induced FAKY397 autophosphorylation and Akt phosphorylation (both #P<0.05 versus vitronectin alone), whereas ERK1/2 phosphorylation was not affected by the status of αv cleavage. To demonstrate that the inhibition of integrin-dependent signaling was not attributable to a defect in the signaling cascades, serum-starved VSMCs, treated with or without dec-CMK, were stimulated with PDGF-BB (5 ng/mL) for 45 minutes. Growth factor stimulation resulted in a vigorous phosphorylation, indicating that signaling through these pathways is intact.
PC5 Expression in Human Carotid Plaques
Carotid atherosclerotic lesions (n=10) obtained by endarterectomy (Figure 6A) contained PC5 in the media underlying fibrous and lipid-rich lesions and in the edges of the lipid core (Figure 6B). Strong PC5 (brown, Figure 6C) and αv (brown, Figure 6E) stainings were noted on serial sections in myofibroblasts (Figure 6D, α-SMA) of the plaque cap (Figure 6F, control). High magnification of PC5 staining in these cells is depicted in Figure 6G. PC5 colocalized with α-SMA (Figure 6H, α-SMA alone; Figure 6J, double-labeling with PC5, brown). In these myofibroblasts, PC5 (blue) colocalizes with αv (brown/red) (Figure 6I).
Adhesive events, essential for cell migration, require dynamic interactions of the cell with the ECM, mediated by integrins. VSMC adhesion and migration requires αvβ3/β5 in vitro,7,8,11 and blocking of αv functions has been demonstrated to prevent neointima formation in vivo.7,8,12 So far, there has been little focus on the characterization of enzymes responsible for the activation of the αv subunit, which will potentially lead to novel strategies to limit VSMC-αv-integrin-matrix interactions in atherosclerosis and restenosis.
The present study demonstrates that PC5 is required for αv activation in VSMCs, thereby controlling VSMC adhesion and migration as well as integrin-dependent signaling. According to our previous studies, αv endoproteolysis occurs within the trans-Golgi-network of VSMCs,19 where furin and PC5 are localized.18 PC5, which is coordinately upregulated with αv in VSMCs after balloon injury in rodents,19 has been shown to be the major αv convertase, being 3 times more potent than furin.17 In our study, the pharmacological PC inhibitor dec-CMK22 significantly inhibited cleavage of the 150-kDa αv into the disulfide-bound 125-kDa heavy and 25-kDa light chains. Because dec-CMK inhibits furin and PC5, specific PC5-asODNs were used, revealing a significant inhibition of αv activation in VSMCs. Furin, which shares several structural, biochemical, and cell biological features with PC5,16 was not affected by PC5-asODNs.
More recently, an alternative pathway of αv cleavage has been demonstrated based on the coexpression of αvβ3 and MT1-MMP in carcinoma cells.20 In contrast to PCs, MT1-MMP generates a disulphide-bounded 115-kDa αv heavy chain with a truncated C-terminus and a 25-kDa light chain, thereby facilitating carcinoma cell adhesion and migration on vitronectin.23 We found that silencing PC5 levels by specific asODNs inhibited VSMC migration on vitronectin. In contrast, we did not find inhibition of αv cleavage or inhibition of adhesion to vitronectin on MT1-MMP inhibition using GM6001 as reported,20 making it unlikely that MT1-MMP significantly contributes to αv activation in VSMCs. MT1-MMP, which activates MMP-2,24 may undergo endoproteolytic activation by furin-like convertases itself.25 Interestingly, dec-CMK (or PC5-asODNs-2; data not shown) did not inhibit the activation of pro-MMP-2 in VSMCs, as noticed by others previously.26 This suggests that furin/PC5 might not contribute to MT1-MMP activation in VSMCs, which can function as a self-convertase.27
In our study, silencing PC5 levels with specific asODNs or inhibiting PC-activity with dec-CMK and thus consequently inhibiting αv activation inhibited VSMC adhesion and migration on vitronectin. It did not affect VSMC adhesion to type I collagen. Our data are in accordance with Berthet el al,15 who demonstrated inhibition of αv cleavage in adenocarcinoma cells transfected with the PC inhibitor α1-PDX (α1-Antitrypsin Portland). Adhesion of α1-PDX transfectants to vitronectin, but not to type I collagen, was significantly inhibited,15 presumably because the α2-subunit, which recognizes type I collagen via a DGEA sequence, is not endoproteolytically cleaved.5 Vitronectin, which is recognized by αv-containing integrins and the platelet αIIbβ3,5 regulates VSMC migration via interaction with αvβ3/β5.7,8 Both αv integrins and vitronectin are upregulated in human atherosclerotic plaques.9,10 Likewise, we found PC5 in human carotid plaques. PC5 colocalized with αv in VSMCs in the atherosclerotic lesions, supporting a role for PC5 in the atherosclerotic process.
There is growing evidence that cleavage of α-subunits may not represent solely precursor maturation but can affect and regulate integrin signaling.14,15,23 Uncleaved α-subunits might have signaling competence, because they are expressed on the cell surface, maintain their RGD-binding properties,14,15 and coimmunoprecipitate with the integrin-signaling adapter molecules Shr/Gbr2.28 A central role in mediating integrin signaling is played by the autophosphorylation of FAK at tyrosine 397 on ECM adhesion, which promotes cell motility.29 It has been shown that adhesion-dependent FAK phosphorylation requires αv cleavage by PCs.15 Whereas autophosphorylated FAKY397 binds PI3-kinase and increases its activity,30 integrin-dependent ERK1/2 activation may occur independently of FAK.31 Activation of PI3-kinase by FAKY397 facilitates cell migration.32 In this study we demonstrate that inhibition of αv cleavage by PCs represses integrin-dependent FAKY397 autophosphorylation and subsequent phosphorylation of Akt, the major PI3-kinase target. In contrast, ERK1/2 phosphorylation was not affected by inhibition of αv cleavage, indicating that the status of αv cleavage could diversely regulate signaling pathways on cell adhesion.
In conclusion, the present study demonstrates that PC5-mediated αv endoproteolytic activation is necessary for αv-dependent VSMC adhesion and migration as well as signaling. Inhibiting PC5 might thus be a novel target to modulate αv-dependent signaling and VSMC functions during the development and progression of atherosclerosis or restenosis.
This work was supported by grants from the Bundesministerium für Bildung und Forschung (BMBF) (CAN02/005) to Drs Stawowy, Fleck, and Graf and Canadian Institutes of Health Research (CIHR) (MGP-44363) to Dr Seidah. Drs Veinot and Chrétien were supported by the Heart and Stroke Foundation of Ontario (NA4337 and T4891) and the Canadian Stroke Network. Dr Chrétien was supported by CIHR (MOP-44362). The authors thank Dr J.S. Munzer for helpful discussions.
An online-only Data Supplement is available at http://www.circulationaha.org.
Ingber DE. Mechanical signaling and the cellular responds to extracellular matrix in angiogenesis and cardiovascular physiology. Circ Res. 2002; 91: 877–887.
Schaller MD, Hildebrandt JD, Shannon JD, et al. Autophosphorylation of the focal adhesion kinase, pp125FAK, directs SH2-dependent binding of pp60src. Mol Cell Biol. 1994; 14: 1680–1688.
Eliceiri BP. Integrin and growth factor receptor crosstalk. Circ Res. 2001; 89: 1104–1110.
Dufourcq P, Couffinhal T, Alzieu P, et al. Vitronectin is up-regulated after vascular injury and vitronectin blockade prevents neointima formation. Cardiovasc Res. 2002; 53: 952–962.
Slepian MJ, Massia SP, Dehdashti B, et al. β3-integrins rather than β1-integrins dominate integrin-matrix interactions involved in postinjury smooth muscle cell migration. Circulation. 1998; 97: 1818–1827.
Hoshiga M, Alpers CE, Smith LL, et al. αvβ3 integrin expression in normal and atherosclerotic artery. Circ Res. 1995; 77: 1129–1135.
Dufourcq P, Louis H, Moreau C, et al. Vitronectin expression and interaction with receptors in smooth muscle cells from human atheromatous plaque. Arterioscler Thromb Vasc Biol. 1998; 18: 168–176.
Coleman KR, Braden GA, Willingham MC, et al. Vitaxin, a humanized monoclonal antibody to the vitronectin receptor (alpha v beta 3), reduced neointima hyperplasia and total vessel area after balloon injury in hypercholesteremic rabbits. Circ Res. 1999; 84: 1268–1276.
Suzuki S, Argraves WS, Pytela R, et al. cDNA and amino acid sequences of the cell adhesion protein receptor recognizing vitronectin reveal a transmembrane domain and homologies with other adhesion protein receptors. Proc Natl Acad Sci U S A. 1986; 83: 8614–8618.
Delwel GO, Hogervorst F, Sonnenberg A. Cleavage of α6A subunit is essential for the activation of the α6Aβ1 integrin by phorbol 12-myristate 13-acetate. J Biol Chem. 1996; 271: 7293–7296.
Berthet V, Rigot V, Champion S, et al. Role of endoproteolytic processing in the adhesive and signaling functions αvβ5 integrin. J Biol Chem. 2000; 275: 33308–33313.
Seidah NG, Mbikay M, Marcinkiewicz M, et al. The mammalian precursor convertases: paralogs of the subtilisin/Kexin family of calcium-dependent serine proteinases. In: Hook VYH, ed. Proteolytic and Cellular Mechanisms in Prohormone and Proprotein Processing. Georgetown, Tex: R.G. Landes Company; 1998: 49–76.
Lissitzky J-C, Luis J, Munzer JS, et al. Endoproteolytic processing of integrin pro-α subunits involves the redundant function of furin and proprotein convertase (PC) 5A, but not paired basic amino acid converting enzyme (PACE) 4, PC5B or PC7. Biochem J. 2000; 346: 133–138.
Stawowy P, Blaschke F, Kilimnik A, et al. Proprotein convertase PC5 regulation by PDGF-BB involves PI3-kinase/p70s6-kinase activation in vascular smooth muscle cells. Hypertension. 2002; 39: 399–404.
Ratnikov BI, Rozanov DV, Postnova TI, et al. An alternative processing of integrin αv subunit in tumor cells by membrane type-1 matrix metalloproteinase. J Biol Chem. 2002; 277: 7377–7385.
Pickering JG, Isner JM, Ford CM, et al. Processing of chimeric antisense oligonucleotides by human vascular smooth muscle cells and human atherosclerotic plaques. Circulation. 1996; 93: 772–780.
Deryugina EI, Ratnikov BI, Postnova TI, et al. Processing of integrin αv subunit by membrane type 1 matrix metalloproteinase stimulates migration of breast carcinoma cells on vitronectin and enhances tyrosine phosphorylation of focal adhesion kinase. J Biol Chem. 2002; 277: 9749–9756.
Toth M, Bernardo M, Gervasi DC, et al. Tissue inhibitor of metalloproteinase (TIMP)-2 acts synergistically with synthetic matrix metalloproteinase (MMP) inhibitors but not with TIMP-4 to enhance the (membrane type-1)-MMP-dependent activation of pro-MMP-2. J Biol Chem. 2000; 275: 41415–41423.
Yana I, Weiss SJ. Regulation of membrane type-1 matrix metalloproteinase activation by proprotein convertases. Mol Biol Cell. 2000; 11: 2387–2401.
Shofuda K, Hasenstab D, Kenagy RD, et al. Membrane-type matrix metalloproteinase-1 and -3 activity in primate smooth muscle cells. FASEB J. 2001; 15: 2010–2012.
Rozanov DV, Strongin AY. Membrane type-1 matrix metalloproteinase functions as a proprotein self-convertase: expression of the latent zymogen in pichia pastoris, autolytic activation, and the peptide sequence of the cleavage forms. J Biol Chem. 2003; 278: 8257–8260.
Cary LA, Chang JF, Guan JL. Stimulation of cell migration by overexpression of focal adhesion kinase and its association with Src and Fyn. J Cell Sci. 1996; 109: 1787–1794.
Chen HC, Guan JL. Association of focal adhesion kinase with its potential substrate phosphatidylinositol 3-kinase. Proc Natl Acad Sci U S A. 1994; 91: 10148–10152.
Lin TH, Aplin AE, Shen Y, et al. Integrin-mediated activation of MAP kinase is independent of FAK: evidence for dual integrin signaling pathways in fibroblasts. J Cell Biol. 1997; 136: 1385–1395.
Reiske HR, Kao SC, Cary LA, et al. Requirement of phosphatidylinositol 3-kinase in focal adhesion kinase-promoted cell migration. J Biol Chem. 1999; 274: 12361–12366.