Fibronectin Fragments Promote Human Retinal Endothelial Cell Adhesion and Proliferation and ERK...

Fibronectin Fragments Promote Human RetinalEndothelial Cell Adhesion and Proliferation and ERKActivation through �5�1 Integrin and PI 3-Kinase

Sylvia H. Wilson,1 Alexander V. Ljubimov,2 Alex O. Morla,3 Sergio Caballero,1

Lynn C. Shaw,1 Polyxenie E. Spoerri,1 Roy W. Tarnuzzer,1 and Maria B. Grant1

PURPOSE. Extracellular matrix degradation is associated withneovascularization in diabetic retinas. Fibronectin fragments(Fn-fs) are generated during vascular remodeling. The effects ofcellular fibronectin (Fn) and selected Fn-fs on adhesion, prolif-eration, and signal transduction in human retinal endothelialcells (HRECs) were characterized.

METHODS. Relative quantitative RT-PCR, flow cytometry, andimmunocytochemistry determined integrin expression onHRECs. Adhesion was evaluated by coating plastic with Fn orFn-fs of 45, 70, 110, or 120 kDa, and MTT conversion was usedto measure proliferation and survival. Peptide inhibitors andblocking antibodies determined adhesive sites and integrinsused for adhesion. Pharmacologic inhibitors and Western anal-yses were used to evaluate intracellular signaling.

RESULTS. HRECs produced significant levels of �2, �3, �5, �v, �1,�3, and �5 integrin subunit mRNA. Flow cytometry of surfaceintegrin expression revealed high levels of �3, �5, and �1 andlower levels of �1, �v, �3, and �5. These results were confirmedby immunocytochemistry. For adhesion to Fn and Fn-fs. the�5�1 integrin was essential. Pharmacologic inhibitors of PI3-kinase blocked adhesion to Fn and Fn-fs, whereas the mito-gen-activated protein (MAP) kinase kinase (MEK) inhibitorPD98059 blocked phosphorylation. The 110- and 120-kDa Fn-fsshowed a concentration-dependent increase in proliferation,whereas 500 ng of the 70 kDa Fn-f–induced proliferation.Addition of III1-C, a matrix assembly domain, increased theproliferative effect of these Fn-fs.

CONCLUSIONS. Fn and its Fn-fs modulate HREC adhesion andproliferation through signal-transduction pathways involvingcoupling of the �5�1 integrin through PI 3-kinase. Mitogenicsignals for endothelial cells from degraded extracellular matrixmay contribute to the development of diabetic retinopathy.(Invest Ophthalmol Vis Sci. 2003;44:1704–1715) DOI:10.1167/iovs.02-0773

Diabetic retinopathy is the leading cause of adult blindnessin the United States.1 Proliferative diabetic retinopathy is

associated with aberrant endothelial cell proliferation, neovas-cularization, vitreous hemorrhage, and traction detachment.Increased amounts of aberrant extracellular matrix (ECM) ob-served in retinal vessels of diabetic patients may contribute tothe endothelial cell dysfunction that is characteristic of thisdisease. Retinal vessels of diabetic patients contain increasedamounts of fibronectin (Fn)2–4 and in vitro exposure of retinalendothelial cells to high glucose levels increases secretion ofFn.5 Retinas of patients with diabetic retinopathy are positivefor the splice variant ED (extra domain)-B Fn, a marker ofangiogenesis,6 suggesting a role for locally synthesized Fn inneovascularization.

Fn is a multifunctional glycoprotein found in plasma andECM that regulates cellular adhesion, migration, oncogenictransformation, wound healing, and hemostasis.7 It exists as a450-kDa dimer with subunits joined by a pair of disulfide bondslocated near the carboxyl termini. The diverse biological activ-ities attributed to Fn have been localized to specific regions ofthe molecule (Fig. 1). These regions were identified by theirbinding affinity for specific molecules such as fibrin, factorXIIIa, gelatin-collagen, and heparin. In addition to bindingdomains, functional domains, such as the matrix assemblydomain located near the amino terminus, and the cell-bindingdomain, which spans type III repeats 8 to 10, have also beenextensively characterized.8,9

Fn is highly susceptible to proteolysis, often generatingfragments (Fn-fs) with greater or different biological activitythan the parent protein.10–12 Degradation of Fn occurs in thevicinity of cells undergoing neoplastic transformation, possiblydue to expression of proteases by cancer cells.13 We havedemonstrated that latent matrix metalloproteinase (MMP)-2,secreted by human retinal endothelial cells (HRECs), exists asa complex with the 30-kDa N-terminal portion of Fn and thatbinding of this fragment to latent MMP-2 inhibits the activationof this latent protease.14 This suggests the active involvementof Fn and Fn-fs in regulating proteolytic enzymes in diabeticneovascularization. We have shown that the 120-kDa N-termi-nal Fn-f is a potent mitogen for HRECs, inducing greater pro-liferation than fibroblast growth factor (FGF)-2.14 Fn-fs inducecell proliferation, cause release of cytokines from vascularcells,15 and modulate adhesion, spreading, and migration ofvascular endothelial cells.14 These studies suggest that the Fnand Fn-fs present in microvascular basement membranes maymodulate neovascularization in the diabetic retina.

Endothelial cells are anchorage dependent and require bothadhesion to the ECM and growth factor stimulation for survival,growth, and differentiation. Many of the adhesive contacts aremediated through integrins, heterodimeric receptor moleculesformed by ion-dependent, noncovalent binding of one � andone � transmembrane glycoprotein subunit.16 The extracellu-lar domain of each subunit binds to several ligands includingthe ECM proteins Fn, vitronectin (Vn), and collagen.

From the 1Department of Pharmacology and Therapeutics, Uni-versity of Florida, Gainesville, Florida; 2Ophthalmology Research, Ce-dars-Sinai Medical Center, Los Angeles, California; and the 3Departmentof Pathology, University of Chicago, Chicago, Illinois.

Supported by National Eye Institute Grants EY12601 andEY07739.

Submitted for publication July 31, 2002; revised October 14 andOctober 31, 2002; accepted November 8, 2002.

Disclosure: S.H. Wilson, None; A.V. Ljubimov, None; A.O.Morla, None; S. Caballero, None; L.C. Shaw, None; P.E. Spoerri,None; R.W. Tarnuzzer, None; M.B. Grant, None

The publication costs of this article were defrayed in part by pagecharge payment. This article must therefore be marked “advertise-ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Corresponding author: Maria B. Grant, Associate Professor, De-partment of Pharmacology and Therapeutics, University of Florida, POBox 100267, Gainesville, FL; [email protected].

Investigative Ophthalmology & Visual Science, April 2003, Vol. 44, No. 41704 Copyright © Association for Research in Vision and Ophthalmology

Similar to growth factor activation, engagement of integrinsinitiates kinase cascades that activate multiple growth-associ-ated kinases including focal adhesion kinase (FAK), phos-phoinositide 3-OH kinase (PI 3-kinase), src, and raf, and themitogen-activated protein (MAP) kinase kinase (MEK).17 Con-versely, lack of attachment to the ECM through integrins in-duces anoikis (cell death by detachment).16

In this study, we determined the expression of integrinsubunits on HRECs by relative quantitative RT-PCR, flow cy-tometry, and immunocytochemistry. We compared the effectof Fn and key Fn-fs on cell adhesion and cellular signalingpathways associated with adhesion, proliferation, and cell sur-vival. Our data indicate that the Fn-fs examined bind to thesame integrin (�5�1) as the parent protein and transduce sig-nals to activate extracellular signal-regulated kinase (ERK)through a PI 3-kinase–dependent pathway. These data suggest,that the interaction of HRECs with Fn and its proteolyticfragments initiate common intracellular signaling events thatcontribute to adhesion and proliferation.

METHODS

Materials and Reagents

Cell culture reagents were obtained from Invitrogen (Carlsbad, CA),except for insulin-transferrin-selenium and endothelial cell growthsupplement, which were purchased from Sigma-Aldrich (St. Louis,MO). Wortmannin, LY294002, U0126, and PD98059 were purchasedfrom Calbiochem (La Jolla, CA); purified cellular Fn and 120-kDa Fn-ffrom Invitrogen; purified 110-kDa Fn-f from Upstate Biotechnology,Inc. (Lake Placid, NY); and purified 70- and 45-kDa Fn-fs from Sigma-Aldrich. Antiserum for detection of phospho-ERK (E10 monoclonal)was from Cell Signaling Technology (Beverly, MA) and total ERKpolyclonal antibody was from Sigma-Aldrich. Horseradish peroxidase(HRP)-conjugated anti-rabbit, HRP-conjugated anti-mouse, and en-hanced chemiluminescence (ECL) reagents were purchased from Am-ersham-Pharmacia Biotech (Piscataway, NJ). Azide-free blocking anti-bodies against integrins �1 (PC4C10), �3 (B3A), and �5 (P1D6) werefrom Invitrogen. Dimeric blocking antibodies �v�3 and �5�1 (LM609and BMA5, respectively) and a second �1 blocking antibody (6S6) werefrom Chemicon International (Temecula, CA). For flow cytometryanalysis, anti-integrin �1 and anti-major histocompatibility complex(MHC) W6.32 were obtained from American Type Culture Collection(ATCC; Manassas, VA), and anti-integrin �2 (PIE6), anti-integrin �3

(PIB5), anti-integrin �5 (P1D6), anti-integrin �v (VNR139), and anti-integrin �4 (3E1) were purchased from Invitrogen. Anti-integrin �6 wasfrom Immunotech (Santa Clara, CA). A second anti-�4 from Beckman-Coulter (Hialeah, FL) was also tested with flow cytometry. Anti-mouse,

anti-rabbit, or anti-goat fluorescein isothiocyanate (FITC)–conjugatedsecondary antibodies were from Sigma-Aldrich. Recombinant humanIII1-C Fn-f was a kind gift from Alex Morla (University of Chicago,Chicago, IL) and additional III1-C was obtained from Sigma-Aldrich.

Isolation and Culture of HRECs

Human eyes were obtained from the National Disease Resource Inter-change (Philadelphia, PA) within 36 hours of death (n � 3 donors).Human retinal endothelial cells were prepared and maintained aspreviously described,18 and cells between passages 3 and 5 were usedfor the present study. The identity of HRECs is typically validated bydemonstrating endothelial cell incorporation of fluorescence-labeledacetylated LDL.18

Relative Quantitative RT-PCR

Relative quantitative RT-PCR was performed as previously described.19

Total RNA was isolated with extraction reagent (TRIzol; GibcoBRL,Grand Island, NY), according to the manufacturer’s instructions. Re-verse transcription was performed with 2 �g of RNA and reversetranscriptase (Superscript MMLU RNase H�; Invitrogen), according tothe manufacturer’s instructions. PCR was performed on the cDNA withpreviously described primers20 or glyceraldehyde-phosphate dehydro-genase (GAPDH) as an internal control. Data were normalized toGAPDH.

Flow Cytometry

A nonenzymatic dissociation was used to remove HRECs from culturedishes, to preserve the integrity of the cell surface molecules. Cellswere washed four times with Ca2�/Mg2�–free PBS and were thenincubated for 15 minutes in 2 mM EDTA in Ca 2�/Mg2�–free PBS at37°C and washed four times with PBS. Cells were incubated for 30minutes on ice with individual integrin antibody diluted in PBS con-taining 1% bovine serum albumin. Cells were washed and then incu-bated with the appropriate fluorescein-labeled secondary antibody for30 minutes on ice. Nonimmune species and isotype-matched antibod-ies were used as negative controls. Appropriate FITC-conjugated sec-ondary antibodies were added and incubated. After a final wash, cellswere fixed in 4% paraformaldehyde in 0.1 M phosphate buffer. Finally,samples were analyzed for integrin expression in flow cytometry per-formed at the University of Florida ICBR Core facility (FACScan; BDBiosciences, Lincoln Park, NJ) and quantified by plotting the relativefluorescence units as histograms over control readings.

Immunocytochemistry

HRECs were grown to 60% confluence (104 cells/cm2) in eight-wellchamber slides (Laboratory-Tek, Naperville, IL) coated with 0.5 �g/mL

FIGURE 1. Fn monomer modular or-ganization and location of fragmentsanalyzed. ED, extra domains (alterna-tively splice regions or sites); CS, cell-specific domains.

IOVS, April 2003, Vol. 44, No. 4 Adhesion and Signaling in Endothelial Cells 1705

Fn (Sigma-Aldrich). The cells were washed with PBS, fixed with 4%paraformaldehyde in 0.1 M phosphate buffer, washed with PBS, andincubated with 10% normal blocking serum in PBS for 30 minutes tosuppress nonspecific binding of IgG. The blocking serum used wasfrom the identical species in which secondary antibody was raised.Cells were then incubated with the primary antibody for 1 hour at37°C and with the appropriate FITC-conjugated secondary antibody,diluted with 1.5% blocking serum in PBS for 30 minutes at 37°C.Nonimmune species and isotype-matched antibodies were used asnegative controls. The cells were washed, mounted in glycerol/PBS,and photographed with a fluorescence microscope (Axiophot; CarlZeiss; Thornwood, NY).

Evaluation of Fn-fs’ Purity and Binding of Fn andFn-fs to Tissue Culture Plastic

Fn and Fn-fs were obtained from three separate suppliers, because thefragments were not available from a single supplier. However, all

FIGURE 2. Surface integrin expression in HRECs. Flow cytometry was used to determine the surface expression of integrin subunits. (A) Arepresentative profile for �1. (B) Quantification of three separate experiments.

TABLE 1. Relative Quantitative RT-PCR Detection of IntegrinmRNA Levels

Subunit mRNA Level

�2 0.459�3 0.457�5 0.142�v 0.839�1 0.392�3 0.933�5 0.484

Data were normalized to glyceraldehyde phosphate dehydroge-nase levels and are expressed as arbitrary units.

1706 Wilson et al. IOVS, April 2003, Vol. 44, No. 4

fragments were purified by the manufacturer, by high-performanceliquid chromatography, and were then reconstituted according tomanufacture’s instructions and stored at �80°C in single-use aliquots.With each purchase of fragments we analyzed each fragment by SDS-PAGE and confirmed that each sample consisted of a single band bysilver staining analysis.21 Tissue culture dishes were coated overnight(4°C) with Fn or the Fn-fs (0.1–10 �g/mL in PBS). Parallel plates werealso coated with 100 �g/mL of each Fn-f and washed twice with PBS.Protein attachment was detected by the bicinchoninic acid (BCA)method (Pierce, Rockford, IL), to ensure that each Fn-f bound to theplastic culture wells with the same avidity. Plates for adhesion andproliferation assays were blocked for 1 hour with 2% BSA in PBS beforeuse.

Adhesion of HRECs

HRECs were incubated for 2 hours in 96-well culture plates coatedovernight with cellular Fn and the 45-, 70-, 110-, or 120-kDa Fn-fsdiluted in PBS, as described earlier. HRECs were serum-starved for 24hours, dissociated in 0.5� trypsin-EDTA, and washed two times in 1%BSA in 1:1 DMEM/F-12. Cells were incubated for 1 hour in 2% BSA inDulbecco’s modified Eagle’s medium (DMEM)/F-12 (1:1) before assayand then plated (5 � 104/well). They were allowed to adhere for 2hours at 37°C, 5% CO2 after preincubation for 10 minutes with RGDS,RGES, or III1-C peptide. Nonadherent cells were removed by washingtwice with PBS. Concentrations of Fn-fs were chosen at which adose-dependent effect was observed, rather than selecting the optimalconcentration, to increase the likelihood of observing the effects of theadded peptides. HREC proliferation was measured by a modified MTTassay, which measures the ability of live cells to use thiozolyl blue andconvert it to dark blue formazan. Complete growth medium containing0.5 mg/mL MTT was added, and MTT conversion was measured usinga microplate spectrophotometer (absorbance, [A]550–A690). For anti-body-blocking experiments, HRECs were processed as described ear-lier and preincubated with antibody to �1, �3, �5, �5�1, or �v�3 dilutedto 10 �g/mL (except for �v�3, which was used at 100 �g/mL) in PBScontaining 2% BSA for 10 minutes before being plated. Control cellswere incubated with nonimmune and isotype matched species anti-bodies. In experiments in which kinase inhibitors were used, cellswere incubated for 30 minutes in the presence of the inhibitors, andcontrol cells were incubated in 0.01% dimethyl sulfoxide (DMSO) forexperiments with wortmannin, LY294002, U0126, and PD98059.

Kinase Activation Assays

Cells were serum-starved for 24 hours (1:1 DMEM/F-12), dissociated asdescribed earlier, allowed to recover for 1 hour, and incubated in thepresence of inhibitors (PD98059, LY294002, wortmannin or 0.01%DMSO as a control) for 30 minutes before stimulation. Cells wereplated on Fn- and Fn-f–coated cell culture dishes (as described foradhesion assays) and allowed to adhere for 2 hours. Nonadherent cellswere collected by centrifugation (800g) at 4°C. The assay was termi-nated with the addition of lysis buffer containing protease and phos-phatase inhibitors (1% Triton X-100, 10 �g/mL aprotinin, 20 �M leu-peptin, 1 �M E-64, 1 mM NaF, 200 �M sodium pervanadate, 1 mMdithiothreitol, 5 mM EDTA, and 25 mM Tris [pH 6.8]) and frozen at�20°C until assayed. Electrophoresis was performed according to themethod described by Laemmli.22 Proteins were fractionated on 10%polyacrylamide gels. Parallel gels were stained with Coomassie blue toverify loading, sample integrity, and protein separation. Proteins weretransferred for 2 hours (50 V) from acrylamide gels to polyvinyl diflu-oride (PVDF) membranes for immunodetection. Membranes wereblocked for 1 hour with 5% nonfat powdered milk in TTBS (25 mMTris-HCl, 150 mM NaCl, 0.05% Tween 20 [pH 7.3]) and probed at roomtemperature (phospho-ERK, 1:1000 in TTBS, 6 hours or total ERK1:40,000 in TTBS for 1 hour). HRP-conjugated anti-rabbit or anti-mousewas used for detection at a dilution of 1:1000. Secondary antibodyincubations were for 1 hour, and membranes were washed three timesin TBS between antibody incubations. Peroxidase activity was detected

using chemiluminescence with 1- to 5-minute exposure times. Densi-tometry was performed on the film (Scion Image, Frederick, MD).Average background density was subtracted, and optical densitieswere plotted on computer (Origin; Microcal, Northampton, MA).

HREC Proliferation Measured by MTT Conversion

The effect of Fn-fs on cell proliferation was measured by MTT conver-sion. Proliferation was measured using 104 cells/well seeded in 96-wellmicrotiter plates coated with proteins as indicated above. HRECs wereincubated for 96 hours in DMEM/F-12 (1:1) supplemented with insulin-transferrin-selenium and 1% endothelial cell growth supplement. Next,10 �L of MTT (5 mg/mL) cells was added to each well (0.5 mg/mL, finalconcentration). Plates were returned to the incubator after addition ofMTT. The assay was terminated by aspiration of the medium withbeveled needle, and MTT was solubilized in 100 �L isopropanol. MTTconversion was measured with a microplate spectrophotometer (A550–A690).

Statistical Analysis

Data were analyzed by ANOVA on computer (Origin Microcal). Eachconcentration was compared with control levels and the differenceconsidered significant at P � 0.05.

RESULTS

A schematic representation of Fn and the regions contained ineach of the five Fn-f examined is shown in Figure 1. The 70-kDaFn-f contains the amino terminal fibrin-heparin and the colla-

FIGURE 3. HRECs stained positive for (A) �5 and (B) �1 subunits byimmunofluorescence. Note the punctate staining. Scale bar, 20 �m.


gen-binding domain. The 45-kDa Fn-f is contained within the70-kDa fragment and includes the collagen-binding domain andthe matrix assembly domain. The 120-kDa Fn-f consists of thefirst 11 type III repeats and contains the cell-binding domain.The 110-kDa Fn-f comprises the first nine type III repeats andcontains a portion of the cell-binding domain. III1-C Fn-f is afragment of the first type III repeat and contains a matrixassembly domain.

Integrin Expression of HRECs

Relative RT-PCR and flow cytometry were used to determinewhich integrins were expressed by HRECs and available tobind Fn. HRECs produced significant levels of �2, �3, �5, �v, �1,�3, and �5 integrin subunit mRNA. Table 1 details the mRNAexpression for integrin subunits in HRECs, normalized toGAPDH. The �v and �3 integrin mRNAs were expressed at thehighest levels, consistent with the role of this integrin dimer inendothelial cell adhesion to various ECM molecules in othervascular beds. Flow cytometry was then used to determinewhether mRNA levels were representative of the surface inte-grin expression. Figure 2 details the cumulative results of flowcytometry analysis of surface integrin expression by HRECs.Integrins containing subunits �3 or �5 were present at highlevels on the cell surface. Integrin subunit �v was also ex-pressed, as was �1, although to a lesser degree than either �3

or �5. The �1 subunit was also highly expressed, and to a lesserdegree �3 and �5. All the integrins detected by RT-PCR werealso detected by flow cytometry, except �1 mRNA that was notdetected by RT-PCR but was detected by flow cytometry andimmunocytochemistry. The immunocytochemical localizationof the �5 and �1 integrin subunits on HRECs is shown inFigures 3A and 3B, respectively. Integrin subunits �3 and �1

were also immunolocalized and reacted with similar intensity,

whereas integrin subunits �1, �v, �3, and �5 reacted with lessintensity (not shown). Negative controls did not stain (notshown).

HREC Adhesion to Fn and Fn-fs

Adhesion was supported on plates coated with cellular (c)-Fnand the 70-, 110-, and 120-kDa Fn-fs (Fig. 4). The 110-kDafragment supported the greatest adhesion up to 10 �g/mL,whereas the 70- and 120-kDa Fn-fs and c-Fn supported similarlevels of adhesion but less than the 110-kDa Fn-f. As expected,the 70-kDa fragment was not as potent at promoting Fn adhe-sion as were the Fn-fs containing the cell-binding domain.Preincubation with the matrix assembly–promoting III1-C pep-tide23 potentiated adhesion to all Fn-fs and to c-Fn (Fig. 4). The5-kDa fragment did not support adhesion, even in the presenceof the III1-C peptide, and it was not capable of blockingadhesion to the other fragments tested or to c-Fn (not shown).In addition, adhesion to all substrates was blocked by preincu-bation with the RGDS peptide, which uncouples RGD-depen-dent integrin binding, but not the control peptide RGES (Fig.4). Adhesion to the 110-kDa Fn-f was only partially blocked byRGDS, indicating either increased affinity of this fragment forthe integrin or the presence of an alternate cell-binding domainor integrin partner. Adhesion to the 70-kDa fragment, whichdoes not contain the cell-binding domain, was also blocked byRGDS. It has been suggested, that this is due to cross-compe-tition between the RGD sequence and a “second” integrin-binding domain on the N-terminal fragment. When this RGD-binding domain of the 70-kDa Fn-f is occupied, the cells areunable to bind through the second integrin-binding site in theN-terminal fragment.24 These data indicate that HRECs adhereto Fn and Fn-fs through integrins and require the presence of

FIGURE 4. Adhesion to the 70- (A),110- (B), and 120-kDa (C) Fn-fs andc-Fn (D). Ninety-six-well cell culturedishes were coated, and cells wereallowed to adhere for 2 hours at37°C, 5% CO2 after preincubation for10 minutes with RGDS, RGES, or theIII1-C peptide (10 �g/mL). Cellswere washed with PBS, and adherentcells were quantified by MTT conver-sion in complete growth medium.For this assay, concentrations of Fn-fswere chosen for which a dose depen-dence had been observed, ratherthan selecting optimal concentra-tions. The cells adhered to all thefragments, and the higher the con-centration of the fragment thegreater the adherence. Note the en-hancement by the III1-C and the ex-pected inhibition by RGDS and ab-sence of inhibition by the controlpeptide RGES. Data represent themean � SD from a single donor (n �4; *P � 0.05; ANOVA, when com-pared with control data). Similar re-sults were obtained in replicate ex-periments with primary culturesobtained from different donors.


the RGD sequence, regardless of whether binding occurs di-rectly at the cell-binding domain.

Inhibition of HREC Adhesion to Fn and Fn-fs byIntegrin Antibodies

Expression data were used to select candidate integrins forantibody blocking experiments. We first tested blocking anti-bodies against the � subunits that were expressed (Fig. 5).Anti-�1 blocked adhesion to the 70- (Fig. 5A), 110- (Fig. 5B),and 120-kDa (Fig. 5C) Fn-fs and to c-Fn (Fig. 5D). Blockingantibodies against �3 and �5 integrins did not block adhesion toany of the proteins tested.

The �1 subunit is capable of forming dimers with several�-subunits. The �5�1 dimer is relatively selective for Fn, al-though Fn can also bind the vitronectin-selective �v�3 integrinunder certain experimental conditions. Anti-�v�1

25 did notblock adhesion to any of the fragments tested (data notshown). Preincubation with anti-�5�1 (10 �g/mL), blockedadhesion to the 70-, 110-, and 120-kDa Fn-fs and Fn-f (Fig. 6).Although some promiscuity exists in integrin-ECM interactions,preincubation with anti-�v�3 (up to100 �g/mL) did not alteradhesion to any of the fragments tested (Fig. 6).

PI 3-Kinase Activity in Adhesion

We next examined the intracellular signaling pathways acti-vated by integrin engagement to promote adhesion. Cells werepretreated with wortmannin or LY294002 (inhibitors of PI3-kinase) and allowed to adhere to each Fn-f or c-Fn for 2 hours.Wortmannin and LY294002 partially blocked adhesion to allfragments, indicating a role for PI 3-kinase in HREC adhesion(Fig.7). Wortmannin (100 nM) was more potent than

LY294002 (50 �M) in inhibiting adhesion, but considerablevariability existed between donors in the degree of observedinhibition with these compounds. When cells were pretreatedwith U0126 and PD98059 inhibitors of MEK/ERK (Fig. 8),adhesion to the 110- and 120-kDa fragments was partiallyblocked by both inhibitors (Figs. 8B, 8C). U0126 (1 �M) wasmore potent than PD98059 (50 �M) in inhibiting adhesion.U0126 blocks both active and inactive MEK/ERK, suggestingthat active (U0126-inhibited but not PD98059-inhibited) MEK/ERK activity may be necessary for adhesion to the 110- and120-kDa Fn-fs. These data suggest a role for ERK in adhesionthat may be mediated solely by the cell-binding domain.

ERK Activation by Adhesion to Fn and Fn-fs

Adhesion to cellular Fn or the 70-, 110-, and 120-kDa Fn-fsincreased activation of ERK, whereas the 45-kDa Fn-f modestlyincreased activation when compared with untreated cells (Fig.9A). Quantification of the p42 ERK isoform is shown in Figure9B. Inhibitors of PI 3-kinase were next tested for their ability toinhibit ERK activation. Both wortmannin and LY294002 re-duced activation of ERK, as did the MEK inhibitor PD98059(Fig. 9B). The 44-kDa ERK isoform was present at lower levelsin HRECs and was below the limit for linear detection by thismethod, which may explain the differential effects on the 44-and 42-kDa isoforms. These data indicate that ERK activationby Fn and Fn-fs occurs through a pathway that is dependent onboth PI 3-kinase and MEK.

HREC Proliferation Measured by MTT Conversion

Cells plated on Fn- and Fn-f–coated dishes (0–10 �g/mL) inserum-free medium showed concentration-dependent prolifer-

FIGURE 5. Integrin antibody �1, butnot �3 or �5, blocked adhesion to the70- (A), 110- (B), and 120-kDa (C)Fn-fs and c-Fn (D). Cells were pre-treated for 10 minutes with blockingantibodies (10 �g/mL) and plated on96-well coated culture dishes. After 2hours at 37°C, 5% CO2, cells werewashed with PBS, and adhesion wasquantified by MTT conversion incomplete growth medium. Data rep-resent the mean � SD, from a singledonor (n � 4; *P � 0.05; ANOVA,when compared with control data).Experiments were repeated withsimilar results with primary culturesobtained from different donors anddifferent sources of antibody.


ation in response to all but the 45-kDa Fn-f. We have demon-strated that the 45-kDa Fn-f inhibits proliferation.14 Exposureof HRECs to the 110- and 120-kDa Fn-f resulted in a concen-tration-dependent increase in the number of cells (Figs. 10B,10D, respectively) The 70-kDa Fn-f, at low concentrations, hadno effect on proliferation, but at concentrations in excess of500 ng, the 70-kDa Fn-f induced proliferation (Fig. 10A). Incontrast, exposure of HRECs to c-Fn had no effect on prolifer-ation (Fig. 10D). Similar to the adhesion process, stimulation ofproliferation on different Fn-fs and c-Fn was significantly po-tentiated by the III1-C peptide containing a matrix assemblysite (Fig. 10). The III1-C fragment alone did not have an effecton MTT conversion (data not shown).

DISCUSSION

Increased serum levels of Fn occur in diabetes, and increasedamounts of Fn are detected in basement membranes of eyes indiabetes. Normalization of blood sugar levels corrects the ele-vated Fn and other serum abnormalities in individuals withdiabetic retinopathy and delays or prevent development of thedisease in individuals with no diabetic retinopathy. However,pancreatic transplantation and subsequent correction of hyper-glycemia does not stabilize or reverse the retinopathy once it isestablished in an individual. Rather, retinopathy continues toprogress in many individuals who have complete correction ofdiabetes after pancreatic transplantation. A possible explana-tion for this observation is that, although the metabolic milieurapidly changes, endothelial cells continue to be exposed tothe altered ECM that contains increased amounts of proteinssuch as Fn and Fn-f. In the current study, HRECs in culture

expressed an increased amount of Fn when exposed to high-glucose conditions (30 mM glucose),26 and HRECs generatedFn-fs in vitro that modulated the activity of MMP-2 and affectedproliferation and migration.

Endothelial cells responded to increased Fn by increasingexpression of an integrin subtype that binds Fn (e.g., �5�1).27

Overexpression of the �5�1 Fn receptor reduces cell migra-tion.28 Thus, increased synthesis and availability of selectedintegrins may mediate the proliferative effects of matrix, fur-ther confirming that integrin binding with ECM componentsactivates intracellular pathways implicated in growth regula-tion.28,29 Depending on the context, integrins can transmitsignals that permit or inhibit growth.30 In the current study,relative quantitative RT-PCR, flow cytometry and immunocyto-chemistry detected multiple integrins expressed by HRECs.However, there were disparities between the mRNA levels andthe cell surface expression detected by flow cytometry. Al-though this may be a result of differential antibody affinity, thedata raise the intriguing possibility that integrin levels in HRECsare subject to posttranscriptional regulation.

In the adhesion experiments, preincubation with the matrixassembly promoting III1-C peptide potentiated adhesion to allfragments, and antibodies to �1 blocked adhesion to Fn and allFn-fs except at the highest concentration of the 70-kDa Fn-ftested. However, although the pattern of adhesion was identi-cal in these experiments, the magnitude of adhesion observedwas different between experiments (Figs. 4, 5). This suggeststhe possibility of donor-to-donor variation in experiments usingprimary cell cultures. However, results were very similar, butnot identical, between donors.

FIGURE 6. Adhesion to the 70- (A),110- (B), and 120-kDa (C) Fn-fs andc-Fn (D) was blocked by antibodiesagainst the �5�1 but not the �v�3

integrin. Cells were pretreated for 10minutes with blocking antibodies (10�g/mL for �5�1, 100 �g/mL for �v�3)and plated on 96-well coated culturedishes. After 2 hours at 37°C, 5%CO2, cells were washed with PBS,and adhesion was quantified by MTTconversion in complete growth me-dium. Data represent the mean � SD(n � 4, *P � 0.05; ANOVA, whencompared with control data). Similarresults were obtained when cellswere incubated in the presence of 10�g/mL �v�3 antibody.


FIGURE 7. Adhesion to the 70- (A),110- (B), and 120-kDa (C) Fn-fs andc-Fn (D) was partially blocked by thePI 3-kinase inhibitors wortmanninand LY294002. Cells were pretreatedfor 30 minutes with 100 nM wort-mannin or 50 �M LY294002 and al-lowed to adhere for 2 hours. Cellswere then washed twice with PBS,and adhesion was quantified by MTTconversion in complete growth me-dium. Data represent the mean � SEof results in four replicate experi-ments using four different donors (D,*P � 0.05; ANOVA, when comparedwith control data).

FIGURE 8. Adhesion to the 110- (B)and 120-kDa (C) fragments was par-tially blocked by the MEK inhibitorsU0126 and PD98059, but adhesion tothe 70-kDa fragment and c-Fn wasnot blocked. Cells were pretreatedfor 30 minutes with each inhibitorand allowed to adhere for 2 hours.Nonadherent cells were washed offwith two changes of PBS, and adhe-sion was quantified by MTT conver-sion in complete growth media. Datarepresent the mean � SE of four rep-licate experiments using four differ-ent donors (*P � 0.05; ANOVA,when compared with control data).


In the current study, the �1 integrin-blocking antibodies didnot completely inhibit adhesion and were unable to block cellbinding to the highest concentration of 70-kDa Fn-f tested. The70-kDa fragment has been characterized as a matrix assemblydomain. We observed this effect with two sources of antibody(Invitrogen and Chemicon) and conclude that the 70-kDa Fnbinds to sites other than those that are blocked by the antibod-ies. In contrast, the antibody raised against the �5�1 dimer, wasa potent inhibitor of adhesion to all Fn-fs tested, including the70-kDa Fn-f. It is possible that this is again a result of differentantibody affinities; however, it may also result from �1-inde-pendent adhesion in the absence of functional �1 subunits.

Fn-fs containing the RGD cell-binding domain induced pro-liferation that was comparable to the 120-kDa or greater thanthe 110-kDa Fn-f. In addition, the 70-kDa Fn-f, which does notcontain the cell-binding RGD sequence, also induced HRECproliferation. Because the 45-kDa fragment is contained withinthe 70-kDa Fn-f and the 45-kDa Fn-f does not support prolifer-ation or adhesion, the adhesive and mitogenic portion of the70-kDa protein must reside within the first 30 kDa of the Nterminus. This 30-kDa N-terminal fragment is generated byHRECs in culture and binds to MMP-2.21

Ligation of integrins activates tyrosine kinases and smallguanosine triphosphatases (GTPases) necessary for the reorga-nization of actin required for cell spreading.31 This is followedby Rho-dependent activation of cell contractility, resulting information of actin stress fibers, clustering of �5�1 integrin andassembly of mature focal adhesions.

A striking correlation exists between the actin cytoskeletonand Fn matrix organization, suggesting that the Fn matrix maybe a potential modulator of actin organization functioning toinfluence cell signaling and growth. The assembly of focaladhesions is associated with the reorganization of Fn intofibrils.32,33 Exogenous full length Fn, as well as its 70-kDa

amino terminal region of Fn, colocalize with Fn fibers. Theinteractions between the amino terminal region of Fn and thecell surface is the initial step in the assembly of exogenous Fninto extracellular matrix and is one of the intermolecular ho-mophilic binding events critical for Fn polymerization.25,34,35

These data suggest that binding of Fn amino terminus to en-dothelial cells has important cytoarchitectural as well as func-tional consequences and that there is an intimate relationshipbetween Fn matrix assembly and cells growth control. This Fnmatrix assembly requires the activity of the integrins, �5�1. Fnmatrix assembly also depends on self-association sites withinFn, in addition to the N-terminal 70-kDa region. The III1-Cfragment is also thought to be particularly important for theproper alignment of Fn molecules during matrix assem-bly.36–40

III1-C was also found to induce spontaneous in vitro disul-fide cross-linking of Fn, to increase binding of cells to Fn and toenhance matrix assembly.41 Herein, we demonstrate that inHRECs, III1-C alone had no effect but when added to Fn andFn-fs of 110, 120, and 70 kDa, it increased proliferation, asmeasured by MTT conversion. Cells can adhere and spread onIII1-C, and both integrins and cell surface proteoglycans medi-ate this adherence. As is the case with most of the Fn-fsexamined, the biological effect of III1-C is dependent on thecell type being examined, the concentration of the fragment,the presence of other Fn-fs, whether the fragment is used forcoating of the culture dish or is added in solution to alreadyadherent cells.

Previous studies have demonstrated that treatment of cellswith III1-C inhibits lysophosphatidic acid–mediated actin orga-nization and tyrosine phosphorylation.42 The ability of III1-C toaffect cytoskeletal function has been attributed to its ability toeither disrupt preexisting Fn matrices or to stimulate increasedFn matrix deposition,43 depending on its concentration. The

FIGURE 9. ERK activation by cellu-lar adhesion to Fn and Fn-fs. HRECswere treated for 30 minutes with0.01% DMSO, PD98059 (50 �M),LY294002 (50 �M), or wortmannin(100 nM) and allowed to adhere tosubstrates coated with c-Fn or Fn-fsfor 2 hours, as described for adhe-sion assays. Cells were harvested inbuffer containing Triton X-100 andprotease and phosphatase inhibitors.Nonadherent cells were collected bycentrifugation and combined withthe adherent fraction. ERK activationwas measured by Western blot withan antibody raised against the hyper-phosphorylated form of the enzyme(A, top). Blots were stripped andprobed with an antibody that recog-nizes total ERK (A, bottom) to verifyequal loading. Cells pretreated withPD98059, LY294002, or wortmanninshowed reduced ERK activationwhen compared with the control(B). Data represent the mean opticaldensities of two replicate Westernblots minus the mean optical densityof nonadherent cells under the samecondition. Similar results were ob-tained using cells derived from a sec-ond donor.


III1-C fragment can partition to caveolin–enriched microdo-mains and allow signals from the ECM to be transmitted to theinterior of the cell to modulate growth and contractility, andthus it is not surprising that III1-C has been shown to modulatesuch complex processes as cell proliferation, angiogenesis, andtumor metastasis.

The ability of each Fn-f to induce proliferation is correlatedwith its ability to support adhesion. Adhesion appears to bemediated by the same integrin (�5�1) for each Fn-f through a PI3-kinase–dependent pathway. Integrin activation by Fn andFn-fs leads to activation of the proliferation-associated kinaseERK through a pathway that is dependent on PI 3-kinase.Activation of both pathways are required for cell survival;however, although each fragment activates both pathways to asimilar degree, there are differences in actin distribution incells adherent to the fragments (Grant MB, unpublished obser-vations, 2001). Differences between Fn-fs may exist in activa-tion of other pathways, such as those coupled to paxillinphosphorylation24 that regulate cytoskeletal dynamics.

ERK activation by Fn can occur through multiple pathways.Association of ECM components with the � integrin subunitactivates a signal transduction cascade, resulting in autophos-phorylation of FAK on Tyr397, which creates a binding site forSrc-family kinases.44,45 Src phosphorylates multiple constitu-ents of the focal adhesion complex, including the dockingprotein p130CAS and FAK itself (e.g., Tyr925).45 Src-dependentphosphorylation of FAK at Tyr925 creates an SH2 docking sitefor the recruitment of Grb2 and Sos, thereby linking integrinsto the ras/raf/ERK cascade.45

FAK-dependent activation of PI 3-kinase may also requireSrc kinase activity.46 The alternate pathway for activation ofERK by integrin engagement involves coupling of the � subunit

to activation and phosphorylation of Shc.30,47 Shc phosphory-lation creates an SH2 binding site for Grb-Sos and links inte-grins to the ras/raf/ERK pathway in a manner that is indepen-dent of FAK.47–51 Shc activation is required for integrin-stimulated proliferation and may cooperate with sustained ERKactivation by FAK to cause cell cycle progression.47,51,52 There-fore, the Fn-fs may modulate ERK signaling through differentpathways that converge upstream of ERK. For example, the70-kDa fragment increases FAK phosphorylation (�70% of thephosphorylation observed with c-Fn) but does not induce pax-illin phosphorylation.24 This indicates either incomplete acti-vation of FAK or differential coupling of the integrin to thetransduction machinery inside the cell. Future experimentswill be designed to determine the divergent pathways acti-vated by Fn-fs that contain the cell-binding domain comparedwith Fn-fs that do not contain the RGD sequence.

Earlier studies have shown that cell adhesion to III1-C re-sults in robust ERK1/2 activation and that this effect is blockedby integrin-blocking antibodies.35,53 Our observations supportthese findings and suggest a possible involvement of Fn assem-bly as a prerequisite for cellular adhesion and proliferation.

In summary, Fn and its proteolytic fragments modulateHREC adhesion and proliferation through similar signal-trans-duction pathways involving coupling of the a5�1 integrinthough PI 3-kinase. Thus, signals from the degraded extracel-lular matrix provide mitogenic signals for HRECs, which maycontribute to the development of diabetic retinopathy.

Acknowledgments

The authors thank Margaret I. Davis for her valuable contribution tothe studies and the manuscript.

FIGURE 10. Proliferation of HRECsin response to the 70- (A), 110- (B),and 120-kDa (C), and Fn-fs and c-Fn(D), alone or in combination withIII1-C. Ninety-six well dishes werecoated with Fn and Fn-fs, and cellswere incubated for 96 hours ingrowth medium without serum ei-ther with or without 50 �M III1-C.Viability was measured using MTTconversion. Data represent themean � SE of results in cells fromtwo donors assayed in quadruplicate(*P � 0.05, when compared with un-coated wells; **P � 0.05 when com-pared with control data at the sameconcentration; ANOVA).


References

1. Frank RN. On the pathogenesis of diabetic retinopathy: a 1990update. Ophthalmology. 1991;98:586–593.

2. Roy S, Cagliero E, Lorenzi M. Fibronectin overexpression in retinalmicrovessels of patients with diabetes. Invest Ophthalmol Vis Sci.1996;37:258–266.

3. Albelda SM, Daise M, Levine EM, Buck CA. Identification andcharacterization of cell-substratum adhesion receptors on cul-tured. J Clin Invest. 1989;83:1992–2002.

4. Allen WE, Jones GE, Pollard JW, Ridley AJ. Rho, Rac and Cdc42regulate actin organization and cell adhesion in macrophages.J Cell Sci. 1997;110:707–720.

5. Munjal ID, McLean NV, Grant MB, Blake DA. Differences in thesynthesis of secreted proteins in human retinal endothelial cells ofdiabetic and nondiabetic origin. Curr Eye Res. 1994;13:303–310.

6. Spirin K, Saghizadeh M, Lewin S, Zardi L, Kenney M, Ljubimov A.Basement membrane and growth factor gene expression in normaland diabetic human retinas. Curr Eye Res. In press.

7. Ruoslahti E. Fibronectin and its receptors. Annu Rev Biochem.1988;57:375–413.

8. Magnusson MK, Mosher DF. Fibronectin: structure, assembly, andcardiovascular implications. Arterioscler Thromb Vasc Biol. 1998;18:1363–1370.

9. Schwarzbauer JE, Sechler JL. Fibronectin fibrillogenesis: a para-digm for extracellular matrix assembly. Curr Opin Cell Biol. 1999;11:622–627.

10. Homandberg GA, Williams JE, Grant D, Schumacher B, EisensteinR. Heparin-binding fragments of fibronectin are potent inhibitorsof endothelial cell growth. Am J Pathol. 1985;120:327–332.

11. Homandberg GA, Meyers R, Xie DL. Fibronectin fragments causechondrolysis of bovine articular cartilage slices in culture. J BiolChem. 1992;267:3597–3604.

12. Homandberg GA, Hui F. Arg-Gly Asp-Ser peptide analogs suppresscartilage chondrolytic activities of integrin-binding and nonbind-ing fibronectin fragments. Arch Biochem Biophys. 1994;310:40–48.

13. Westermarck J, Kahari VM. Regulation of matrix metalloproteinaseexpression in tumor invasion. FASEB J. 1999;13:781–792.

14. Grant MB, Caballero S, Bush DM, Spoerri PE. Fibronectin frag-ments modulate human retinal capillary cell proliferation and mi-gration. Diabetes. 1998;47:1335–1340.

15. Poggi A, Stella M, Donati MB. The importance of blood cell-vesselwall interactions in tumour metastasis. Baillieres Clin Haematol.1993;6:731–752.

16. Frisch SM, Ruoslahti E. Integrins and anoikis. Curr Opin Cell Biol.1997;9:701–706.

17. Chen AF, O’Brien T, Tsutsui M, et al. Expression and function ofrecombinant endothelial nitric oxide synthase gene in canine basi-lar artery. Circ Res. 1997;80:327–35.

18. Grant MG, Guay C. Plasminogen activator production by humanretinal endothelial cells of nondiabetic and diabetic origin. InvestOphthalmol Vis Sci. 1991;32:53–64.

19. Tarnuzzer RW, Macauley SP, Farmerie WG, et al. Competitive RNAtemplates for detection and quantitation of growth factors, cyto-kines, extracellular matrix components and matrix metalloprotein-ases by RT-PCR. Biotechniques. 1996;20:670–674.

20. Dou Q, Williams RS, Chegini N. Expression of integrin messengerribonucleic acid in human endometrium: a quantitative reversetranscription polymerase chain reaction study. Fertil Steril. 1999;71:347–353.

21. Grant MB, Caballero S, Tarnuzzer RW, et al. Matrix metalloprotein-ase expression in human retinal microvascular cells. Diabetes.1998;47:1311–1317.

22. Laemmli UK. Cleavage of structural proteins during the assemblyof the head of bacteriophage T4. Nature. 1970;227:680–685.

23. Chernousov MA, Fogerty FJ, Koteliansky VE, Mosher DF. Role ofthe I-9 and III-1 modules of fibronectin in formation of anextracellular fibronectin matrix. J Biol Chem. 1991;266:10851–10858.

24. Hocking DC, Sottile J, McKeown-Longo PJ. Activation of distinctalpha5beta1-mediated signaling pathways by fibronectin’s cell ad-

hesion and matrix assembly domains. J Cell Biol. 1998;141:241–253.

25. Zhang Z, Morla AO, Vuori K, Bauer JS, Juliano RL, Ruoslahti E. Thealpha v beta 1 integrin functions as a fibronectin receptor but doesnot support fibronectin matrix assembly and cell migration onfibronectin. J Cell Biol. 1993;122:235–242.

26. Munjal ID, McLean NV, Blake DA, Grant MB. Differences in thesynthesis of secreted proteins in human retinal endothelial cellsof diabetic and nondiabetic origin. Curr Eye Res. 1994;13:303–310.

27. Podesta F, Roth T, Ferrara F, Cagliero E, Lorenzi M. Cytoskeletalchanges induced by excess extracellular matrix impair endothelialcell replication. Diabetologia. 1997;40:879–886.

28. Schwartz MA, Lechene C, Ingber DE. Insoluble fibronectin acti-vates the Na/H antiporter by clustering and immobilizing integrinalpha 5 beta 1, independent of cell shape. Proc Natl Acad Sci USA.1991;88:7849–7853.

29. Symington BE, Carter WG. Modulation of epidermal differentiationby epiligrin and integrin alpha 3 beta 1. J Cell Sci. 1995;108:831–838.

30. Wary KK, Mainiero F, Isakoff SJ, Marcantonio EE, Giancotti FG. Theadaptor protein Shc couples a class of integrins to the control ofcell cycle progression. Cell. 1996;87:733–743.

31. Clark EA, King WG, Brugge JS, Symons M, Hynes RO. Integrin-mediated signals regulated by members of the rho family ofGTPases. J Cell Biol. 1998;142:573–586.

32. Mattey DL, Garrod DR. Role of glycosaminoglycans and collagen inthe development of a fibronectin-rich extracellular matrix in cul-tured embryonic corneal epithelial cells. J Cell Sci. 1984;67:189–202.

33. Avnur Z, Geiger B. The removal of extracellular fibronectin fromareas of cell-substrate contact. Cell. 1981;25:121–132.

34. Hocking DC, Sottile J, McKeown-Longo PJ. Fibronectin’s III-1 mod-ule contains a conformation-dependent binding site for the amino-terminal region of fibronectin. J Biol Chem. 1994;269:19183–19187.

35. Morla A, Ruoslahti E. A fibronectin self-assembly site involved infibronectin matrix assembly: reconstruction in a synthetic peptide.J Cell Biol. 1992;118:421–429.

36. Pierschbacher MD, Ruoslahti E. Cell attachment activity of fi-bronectin can be duplicated by small synthetic fragments of themolecule. Nature. 1984;309:30–33.

37. McDonald JA. Receptors for extracellular matrix components.Am J Physiol. 1989;257:L331–L337.

38. McKeown-Longo PJ, Mosher DF. Interaction of the 70,000-mol-wtamino-terminal fragment of fibronectin with the matrix-assemblyreceptor of fibroblasts. J Cell Biol. 1985;100:364–374.

39. Quade BJ, McDonald JA. Fibronectin’s amino-terminal matrix as-sembly site is located within the 29-kDa amino-terminal domaincontaining five type I repeats. J Biol Chem. 1988;263:19602–19609.

40. Schwarzbauer JE. Identification of the fibronectin sequences re-quired for assembly of a fibrillar matrix. J Cell Biol. 1991;113:1463–1473.

41. Mercurius KO, Morla AO. Inhibition of vascular smooth musclecell growth by inhibition of fibronectin matrix assembly. Circ Res.1998;82:548–556.

42. Bourdoulous S, Orend G, MacKenna D, Pasqualini R, Ruoslahti E.Fibronectin matrix regulates activation of RHO and CDC42GTPases and cell cycle progression. J Cell Biol. 1998;143:267–276.

43. Yi M, Ruoslahti E. A fibronectin fragment inhibits tumor growth,angiogenesis, and metastasis. Proc Natl Acad Sci U S A. 2001;98:620–624.

44. Schaller MD, Hildebrand JD, Shannon JD, Fox JW, Vines RR,Parsons JT. Autophosphorylation of the focal adhesion kinase,pp125FAK, directs SH2-dependent. Mol Cell Biol. 1994;14:1680–1688.

45. Schlaepfer DD, Hanks SK, Hunter T, van-der-Geer P. Integrin-mediated signal transduction linked to Ras pathway by GRB2binding to focal. Nature. 1994;372:786–791.

46. Chen HC, Appeddu PA, Isoda H, Guan JL. Phosphorylation oftyrosine 397 in focal adhesion kinase is required for binding


phosphatidylinositol 3-kinase. J Biol Chem. 1996;271:26329–26334.

47. Wary KK, Mariotti A, Zurzolo C, Giancotti FG. A requirement forcaveolin-1 and associated kinase Fyn in integrin signaling andanchorage-dependent cell growth. Cell. 1998;94:625–634.

48. Lin T, Aplin A, Shen Y, et al. Integrin-mediated activation of MAPkinase is independent of FAK: evidence for dual integrin signalingpathways in fibroblasts. J Cell Biol. 1997;136:1385–1395.

49. Schlaepfer DD, Hunter T. Integrin signalling and tyrosinephosphorylation: just the FAKs? Trends Cell Biol. 1998;8:151–157.

50. Schlaepfer DD, Hunter T. Focal adhesion kinase overexpressionenhances ras-dependent integrin signaling to ERK2/mitogen-acti-

vated protein kinase through interactions with and activation ofc-Src. J Biol Chem. 1997;272:13189–13195.

51. Pozzi A, Wary KK, Giancotti FG, Gardner HA. Integrin alpha1beta1mediates a unique collagen-dependent proliferation pathway invivo. J Cell Biol. 1998;142:587–594.

52. Mainiero F, Murgia C, Wary KK, et al. The coupling of alpha6beta4integrin to Ras-MAP kinase pathways mediated by Shc controlskeratinocyte proliferation. EMBO J. 1997;16:2365–2375.

53. Mercurius KO, Morla AO. Cell adhesion and signaling on thefibronectin 1st type III repeat: requisite roles for cell surfaceproteoglycans and integrins (serial on-line). BMC Cell Biol. 2001;2:18.


Fibronectin Fragments Promote Human Retinal Endothelial Cell Adhesion and Proliferation and ERK...

Documents

Transcript of Fibronectin Fragments Promote Human Retinal Endothelial Cell Adhesion and Proliferation and ERK...