FASEB J. Pierce now sold as Thermo Scientific
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fontana, L.
Right arrow Articles by Rifkin, D. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fontana, L.
Right arrow Articles by Rifkin, D. B.
(The FASEB Journal. 2005;19:1798-1808.)
© 2005 FASEB

Fibronectin is required for integrin {alpha}vß6-mediated activation of latent TGF-ß complexes containing LTBP-1

Laura Fontana*, Yan Chen*, Petra Prijatelj*, Takao Sakai{ddagger}, Reinhard Fässler{ddagger}, Lynn Y. Sakai§ and Daniel B. Rifkin*,{dagger},1

Departments of
* Cell Biology and
{dagger} Medicine, New York University School of Medicine, New York, New York, USA; Shriners Hospital for Children and
§ Department of Biochemistry and Molecular Biology, Oregon Health and Science University, Portland, Oregon, USA; and Max Planck Institute of Biochemistry,
{ddagger} Department of Molecular Medicine, Martinsried, Germany

1 Correspondence: Department of Cell Biology, New York University School of Medicine, 550 First Ave., New York, NY 10016, USA. E-mail: rifkid01{at}med.nyu.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Transforming growth factor-ßs (TGF-ß) are secreted as latent complexes consisting of the TGF-ß dimer, the TGF-ß propeptide dimer, and the latent TGF-ß binding protein (LTBP). Although the bonds between TGF-ß and its propeptide are cleaved intracellulary, the propeptide associates with TGF-ß by electrostatic interactions, thereby conferring latency to the complex. We reported that a specific sequence of LTBP-1 is required for latent TGF-ß activation by the integrin {alpha}vß6. Here we describe a 24 amino acid sequence from the hinge domain required for activation. The LTBP-1 polypeptide rL1N, which includes the hinge, associates with fibronectin in binding assays. We present evidence that fibronectin null cells minimally activate latent TGF-ß and poorly incorporate the active hinge sequence into their matrix. In addition, cells missing the fibronectin receptor {alpha}5ß1 exhibit defective activation of latent TGF-ß by {alpha}vß6 and decreased matrix incorporation. The results indicate specificity for integrin-mediated latent TGF-ß activation that include unique sequences in LTBP-1 and an appropriate matrix molecule.—Fontana, L., Chen, Y., Prijatelj, P., Sakai, T., Fässler, R., Sakai, L. Y., Rifkin, D. B. Fibronectin is required for integrin {alpha}vß6-mediated activation of latent TGF-ß complexes containing LTBP-1.


Key Words: LTBP • TGF-ß • fibronectin null cells


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE TRANSFORMING GROWTH FACTOR-ßs (TGF-ß1, 2, and 3) are the prototypical members of a multifunctional cytokine family (1) . The TGF-ßs play pivotal roles in numerous biological processes, such as the immune response (2) , regulation of cell proliferation (1) , angiogenesis (3) , and extracellular matrix (ECM) production (4 , 5) . Anomalous regulation of TGF-ß activity has been related to several human diseases, including tumorigenesis, fibrosis, and autoimmune disease (6) .

TGF-ß is secreted as a biologically inactive large latent complex (LLC) composed of a latent TGF-ß binding protein (LTBP, 125–240 kDa) covalently bound to the latency associated protein [LAP; the amino-terminal propeptide homodimer of the TGF-ß precursor (75 kDa)] and TGF-ß (25 kDa). LAP associates noncovalently with the mature TGF-ß homodimer (7 , 8) ; this complex is called the small latent complex (SLC). The interaction of mature TGF-ß with LAP blocks growth factor binding to its receptors. Thus, LAP, but not LTBP, is responsible for TGF-ß latency. Activation of latent TGF-ß involves disruption of the interaction between LAP and TGF-ß to permit subsequent binding of the cytokine to its receptors (1) .

TGF-ß activation is mediated by several mechanisms. Isolated latent TGF-ß can be activated with heat treatment, acidic or basic pH, and different chaotropic agents or detergents (9) . In tissue culture, TGF-ß can be activated by proteases (10) , thrombospondin-1 (TSP-1) (11) , and {alpha}vß8 (12) or {alpha}vß6 integrins (13) . In the latter case, activation requires the binding between RGD motifs of LAP-ß1 or LAP-ß3 and {alpha}vß6 integrins (13) as well as a specific domain in LTBP-1 (14) . LTBP-3 or LTBP-4 cannot substitute for LTBP-1 (ref 14 and our unpublished data). ß6-Deficient mice show increased inflammation and decreased fibrosis, processes strongly regulated by TGF-ß (15) , suggesting that {alpha}vß6 is an important in vivo activator of latent TGF-ß.

LTBP-1 is required for {alpha}vß6-mediated TGF-ß activation (14) and a sequence of LTBP-1 necessary for this process has been identified (14) . The four LTBP isoforms (LTBP-1, -2, -3, and -4) are composed of three domains containing eight cysteine residues, called cysteine-rich (CR), eight cysteine (8-Cys), or TGF-ß binding (TB) domains, one "hybrid domain" that has sequence similarity to both EGF and CR domains and also contains eight cysteines, and 15–20 epidermal growth factor (EGF)-like repeats (10) . LTBP-1, -3, and -4, but not LTBP-2, can disulfide bind to LAP through the third of the four 8-Cys domains (16) .

LTBP-1 is ubiquitously expressed and is abundant in heart, placenta, lung, spleen, kidney, and stomach (10) . LTBP-1 facilitates the secretion of the SLC (17) and targets the SLC to the ECM (18) through interactions involving the N-terminal region (19 20 21) . LTBP-1 colocalizes with fibronectin (Fn) in cultures of primary osteoblasts (20) or human fibroblasts (22) as visualized by double immunofluorescence. An interaction between LTBP-1 and Fn has also been demonstrated by dot-blot assay (22) .

Fn is an abundant ECM glycoprotein that plays important roles in cell adhesion, migration, growth, and differentiation (23 , 24) . Secreted Fn is a dimer composed of two subunits of 220–250 kDa covalently linked at their C termini by a pair of disulfide bonds. Fn is widely expressed by different cell types, is essential for vertebrate development (25) , and interacts with many molecules including collagen, integrins, fibrin, and heparan sulfate proteoglycans (23) .

Twenty variants in human Fn, generated from alternative splicing of a single pre-mRNA (26) , have been described. Alternative splicing is cell specific; for example, plasma Fn produced from hepatocytes does not contain the ED-A domain included in fibrillar cellular Fn, an abundant component of the ECM (27) . TGF-ß affects the alternative splicing pattern of Fn by increasing expression of the Fn isoform containing the ED-A sequence (28) . TGF-ß also stimulates the expression of ECM proteins, such as tenascin, thrombospondin, Fn, vitronectin, and several proteoglycans (29) , and plays a major role in the regulation of ECM degradation and remodeling (10) . Thus, TGF-ß is a key mediator in matrix metabolism.

In the present study we investigated the role of Fn in {alpha}vß6-mediated TGF-ß activation. The results indicate that Fn participates in latent TGF-ß activation by {alpha}vß6 integrins by binding to a region of the LTBP-1 hinge sequence.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell lines and reagents
Control or Fn–/– mouse fibroblasts were isolated at E13.5 from Fn (flox/flox) mouse embryos (30) and immortalized by transducing an SV-40 large T antigen along with a neomycin gene (31) . After G418 selection, individual cells were cloned using ring cylinders and treated with a Cre-transducing adenovirus to delete the floxed Fn genes. The deletion of Fn alleles was confirmed by PCR and the lack of Fn protein expression by immunoprecipitation using metabolically labeled conditioned media of mutant cells (not shown). CHOB2 and CHOB2/a27 cells were a generous gift from R. L. Juliano, University of North Carolina (32) . ß6-Integrin transfected CHO cells (CHO-ß6) were from D Sheppard (UCSF, San Francisco, CA, USA) (33) CHO-ß6/ECR3E cells stably transfected with the ECR3E region of LTBP-1 were obtained as described (14) . Mink lung epithelial cells (TMLC) stably transfected with a plasmid containing the luciferase cDNA downstream of a TGF-ß-sensitive portion of the plasminogen activator inhibitor 1 promoter (TMLC) were used as described (34) . Luciferase activity is expressed as relative luciferase units (RLU) in which the luciferase activities of experimental samples are divided by the value of the mock control. Cells were grown in DMEM containing 10% heat-inactivated Fn-depleted FCS. Fn was removed from PCS by gelatin-Sepharose chromatography.

Anti-paxillin antibody was from BD Transduction Laboratories and TRITC-labeled phalloidin from Sigma (St. Louis, MO, USA). The secondary anti-mouse IgG peroxidase conjugate antibody was from Amersham Biosciences (Arlington Heights, IL, USA). Mouse anti-{alpha}vß6 Mab 10D5 (35) was a gift from D. Sheppard. Recombinant simian TGF-ß1 LAP was produced as described (36) . Mouse monoclonal antibody HA.11 against the HA epitope was from Covance Research Products BAbCO (Richmond, CA, USA). DAPI was purchased from Sigma.

The purified LTBP-1 polypeptides, rL1N, rL1M, and rL1C are described in ref 37 .

All molecular biology reagents were obtained from Roche Diagnostics Corporation (Indianapolis, IN, USA).

Constructs and vectors
pcDNA3 vector was obtained from Invitrogen (Carlsbad, CA, USA). Simian TGF-ß1 cDNA encoding the complete latent TGF-ß1 was a gift from R. Derynck (UCSF, San Francisco, CA, USA). The ß6-integrin cDNA was a gift from D. Sheppard. pcDNA3-LTBP-1S/HA and the hinge construct pSecTag2cECR3E 403-449 was obtained as described (14) .

LTBP-1 hinge-derived constructs were obtained as follows. The vector pSecTag2cECR3EHA (14) was digested with HindIII and BamHI. This vector contains the coding regions for EGF13, CR3, EGF14, and HA epitope. DNA coding for different hinge regions was cloned upstream the EGF13 coding sequence. These sequences were obtained by the annealing of sense and antisense oligonucleotides containing HindIII and BamHI sticky ends at the 5' and 3' ends, respectively. The sense and antisense oligonucleotides are listed in Table 1. The annealed oligonucleotides were ligated into the HindIII/BamHI digested pSecTag2cECR3EHA vector.


View this table:
[in this window]
[in a new window]
  		Table 1. Oligos used for preparing hinge constructs

TGF-ß bioassay
To measure TGF-ß activation, cells were seeded at 4 x 105 cells/well in 2 mL of medium in 6-well plates and incubated at 37°C for 16 h. Cells were transfected with 1.4 µg of DNA per well using Lipofectamine Plus (Life Technologies, Grand Island, NY, USA). After 16 h cells were collected, counted, and equal numbers of cells were added to TMLC in 96-well plates (34 , 38) . Sixteen to 24 h later, TGF-ß was assessed by measuring luciferase activity in the cell lysates. Total secreted TGF-ß was measured as described (14) . All experiments were done in triplicate and repeated at least three times with similar results. The data presented are the mean and standard deviation of the mean of three measurements of a single experiment.

Measurement of LTBP-1 mRNA levels and secretion in control and Fn null cells
Cells were transiently transfected with LTBP-1-HA expression vector or with a control DNA (mock) as described above. Sixteen to 24 h later, medium was collected, subjected to SDS-PAGE and Western blot with an anti-HA antibody, followed by a secondary anti-mouse IgG antibody peroxidase conjugate. The blot was developed using enhanced chemiluminescence detection (ECL). Total RNA was extracted from cells with Trizol (Invitrogen). cDNA was generated from RNA with reverse transcriptase and used as a template for amplification with Taq polymerase of 290 or 661 bp fragments of LTBP-1 using two different pairs of oligos specific for LTBP-1. PCR amplification with LPBF (5'-TGTCAGCTACAAGGTGTATG-3') and LPBR1 (5'-TCCTTAAAGCAGCTGTGCC-3') generated a 290 bp fragment, whereas amplification with LPF1 (5'-GTCTTTACTCCGAGCATCTG-3') and LPR1 (5'-AGGCCAGGGAGCGCTTTG-3') generated a 661 bp fragment. The result reported refers to the LPBR1/LPBF amplification. A similar result was obtained when the LPF1/LPR1 oligos were used (data not shown). All PCR amplifications were stopped at 30 cycles.

Measurement of LTBP-1 hinge-derived construct mRNA levels in CHO-ß6/ECR3E
CHO-ß6/ECR3E cells were transiently transfected with TGF-ß1 plus the LTBP-1 hinge-derived constructs or a control DNA. Sixteen to 24 h later, total RNA was extracted from cells with Trizol. cDNA was generated from RNA with reverse transcriptase and used as a template for the amplification of a 550 bp fragment using the following pair of oligos, pSec.for (5'-GTTCCAGGTTCCACTGGTGAC-3') and pSec.rev (5'-ATCTCCCCATCCCGCGCCTG-3').

Measurement of ß6 integrin expression by RT-PCR
Total RNA was extracted from cells transiently transfected with a ß6 expression vector. cDNA was generated from RNA with reverse transcriptase and used as a template for the amplification of a 183 bp fragment using the following oligos, ß6.sense (5'-CCGGCTGGCCAAAGAGATGT-3') and ß6.antisense (5'-AGTTAATGGCAAAATGTG CT-3').

Binding studies
Interaction between Fn and LTBP-1 polypeptides was investigated using ELISA or blot overlay assays. For ELISA binding assays, multiwell plates were coated with purified LTBP-1 polypeptides rL1N, rL1M, rL1C (50 nM, 100 µL/well) in 15 mM Na2CO3 and 35 mM NaHCO3 pH 9.2, at 4°C overnight. After the coating solution was discarded, the coated wells were washed and blocked with 5% nonfat dry milk in TBS at room temperature for 1 h. Fn purified from medium conditioned by human skin fibroblasts was added to a final concentration of 25 µg/mL in 2% milk-TBS, containing 5 mM EDTA, and incubated for 3 h at room temperature. After washing the wells with TBS, a monoclonal antibody against Fn used to detect the bound ligand. After washing, an anti-mouse IgG antibody peroxidase conjugate (Amersham) was incubated with the cultures for 1 h at room temperature. The plates were washed and peroxidase activity was detected by incubation with TMB substrate (Sigma). Absorbance was determined at 450 nm using a Dynatech MR7000.

For blot overlay assays, purified recombinant LTBP-1 polypeptides were separated by SDS-PAGE and transferred by electroblotting onto nitrocellulose membranes (Bio-Rad, Hercules, CA, USA). After blocking with 5% nonfat milk in TBS at room temperature for 1 h, the membrane was incubated overnight with Fn (Sigma; 50 µg/mL in 2% nonfat milk-TBS) at 4°C. A monoclonal antibody against Fn, diluted in 2% milk-TBS was used to detect the bound ligand. After washing an anti-mouse IgG antibody peroxidase conjugate was incubated with the membrane for 1 h at room temperature. The blot was developed using enhanced chemiluminescence detection (ECL, Amersham).

Detection of 414-437 ECR3E in the matrix of control and Fn–/– cells
Control and Fn–/– cells were seeded at 4 x 105 cells/well in 2 mL of medium with Fn-depleted FCS in 6-well plates and incubated at 37°C for 16 h. Where indicated, cellular Fn (Sigma; 10 µg/mL) was added to Fn–/– cells. Cells were transfected with 1.4 µg of DNA per well using Lipofectamine Plus. Sixteen to 24 h later, medium was collected and matrix extracts prepared (39) . Briefly, confluent cultures washed with PBS and lysed in deoxycholate (DOC) buffer (20 mM HEPES pH 7.2, 1% NP-40, 10% glycerol, 1% DOC, and Roche protease inhibitor cocktail) at 4° for 20 min. The cultures were scraped and DOC-insoluble material, representative of the fibrillar ECM, was pelleted by centrifugation at 10,000 RPM for 5 min and solubilized in SDS sample buffer. Matrix extracts were resolved by SDS-PAGE. Western blot was performed using an anti-HA antibody (Covance) and a secondary anti-mouse IgG antibody peroxidase conjugate (Amersham). The blot was developed using ECL. Protein secretion was analyzed by Western blot of the conditioned medium with an anti-HA antibody.

Immunofluorescence
Cells grown on coverslips were fixed for 5 min in 3.7% paraformaldehyde, subsequently permeabilized with 0.5% Triton X-100 in PBS for 20 min, and washed twice with 0.1 M glycine. Immunofluorescence staining for F-actin was done with TRITC-labeled phalloidin and for vinculin with mouse mAb against paxillin. Secondary antibodies were Cy3 goat anti-mouse lgG + lgM (Jackson ImmunoResearch Labs, Inc., West Grove, PA, USA). After washing 3x with PBS, cells were embedded in gelvatol (30) and examined with a Leica confocal microscope.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Identification of a 24 amino acid fragment of the LTBP-1 hinge region required for {alpha}vß6-mediated TGF-ß activation
A region of LTBP-1 containing amino acids 403-449 of the hinge region and the LAP binding domain 8-Cys repeat 3 (CR3) plus flanking EGF-like repeats (EGF 13 and 14) (ECR3E) is sufficient to achieve {alpha}vß6-mediated TGF-ß activation (14) . To determine whether a smaller region of the hinge mediates {alpha}vß6-dependent TGF-ß activation, we designed expression constructs encoding four partially overlapping fragments spanning amino acids 403-449 fused at their C-terminal ends to LTBP-1 ECR3E plus the hemagglutinin epitope (HA) as a tag (Fig. 1 A). Because LAP binds to the CR3 domain, these constructs will all bind to TGF-ß LAP. CHO-ß6/ECR3E cells, which have an impaired ability to activate latent TGF-ß (14) , were transiently transfected with the hinge-derived constructs or a full-length pro-TGF-ß construct and latent TGF-ß activation measured (Fig. 1B ). TGF-ß activation was measured by coculturing the transfected cells with TGF-ß reporter cells (TMLC) (34) . As shown in Fig. 1B , the construct 414-437 ECR3E, containing the hinge sequence 414-437 plus ECR3E, facilitated {alpha}vß6-mediated TGF-ß activation equivalent to either the 403-449 ECR3E construct or full-length LTBP-1 (Fig. 1B ). The other three hinge fragment constructs did not support latent TGF-ß activation (Fig. 1B ). No differences in secretion and complex formation with LAP were found in medium from CHO-ß6 cells transiently transfected with the different hinge constructs (data not shown). Finally, no significant differences in the mRNA levels for each of the transfected constructs were observed by RT-PCR (Fig. 1C ).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 1. Latent TGF-ß activation mediated by hinge-derived constructs. A) Hinge construct schematic. Underlined amino acids represent the sequence in each hinge construct. The constructs contain, from the N- to C-terminal, a portion of the hinge region, EGF13, CR3, EGF14, and the hemagglutinin epitope (HA) as a tag. Therefore, the constructs differ only in the hinge region. B) TGF-ß formation by hinge constructs. CHO-ß6/ECR3E cells were transiently transfected with TGF-ß plus either the hinge-derived constructs or a control DNA (mock). After 16–24 h, cells were cocultured with TMLCs and luciferase activity was measured. Relative luciferase units (RLU) are expressed as the measured luciferase activity divided by the activity of the coculture established with mock-transfected cells. C) Hinge construct expression. CHO-ß6/ECR3E cells were transiently transfected with TGF-ß plus either the hinge-derived constructs or a control DNA (mock). Semiquantative RT-PCR (30 cycles) was performed on RNA extracted from cells 16–24 h later.

Interaction between LTBP-1 fragments and fibronectin
We next examined the interaction between LTBP-1 polypeptides and Fn. We chose to examine the association of the hinge sequence with Fn for several reasons. First, the N-terminal region of LTBP-1 (aa 340-545) associates with the ECM of skin fibroblasts (21) . Second, LTBP-1 colocalizes with Fn in primary osteoblasts (20) and fibroblast cultures (22) . Third, Fn and LTBP-1 interact as demonstrated in dot-blot assays (22) . To test whether the N-terminal region of LTBP-1 binds Fn, we performed ELISA and blot overlay assays with recombinant LTBP-1 polypeptides rL1N (aa 21-629), rL1M (aa 588-1139), and rL1C (aa 1097-1394) (37) . Only the rL1N polypeptide (aa 21-629) bound Fn, indicating the presence of binding site for Fn in the N-terminal region of LTBP-1 (Fig. 2 A). Our attempts to demonstrate binding of the isolated hinge peptide to Fn were unsuccessful. The reasons for this may relate to the relatively unstructured nature of the short peptide. However, competition experiments with the hinge peptide showed an inhibition of latent TGF-ß activation in CHO-ß6/ECR3E cells overexpressing the 414-437 LTBP-1 construct (Fig. 3 ). The interaction between LTBP-1 polypeptides and Fn was also analyzed by ELISA (Fig. 2B, C ). Only the rL1N peptide, containing amino acids 414-437 required for {alpha}vß6-mediated TGF-ß activation, bound to Fn. Control experiments in which no protein or a protein other than Fn were used as ligands gave only a background signal (data not shown). We also analyzed the degree of binding of rL1N vs. increasing amounts of Fn. In this experiment, rL1N polypeptide was coated on the plate and incubated with increasing concentrations of Fn. Bound ligand was detected as described. A dose-response curve was obtained with rL1N polypeptide (Fig. 2C ). A control sample for background binding indicated no Fn binding in the absence of Fn (data not shown). These experiments indicated that Fn interacts with the region of LTBP-1 containing the hinge sequence.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 2. Hinge interactions with Fn. A) Fn blotting to LTBP-1 peptides. Blot overlay assay with LTBP-1 polypeptides and Fn. Purified peptides rL1N, rL1M and rL1C were resolved by SDS-PAGE and transferred to a nitrocellulose filter. Purified Fn was used as ligand and the bound protein detected by anti-Fn followed by a secondary anti-mouse HRP conjugate. B) ELISA of Fn binding. Purified polypeptides rL1N, rL1M, and rL1C were coated on an ELISA plate, Fn added, and the bound ligand detected by anti-Fn followed by a secondary anti-mouse HRP conjugate antibody. A control with only coating buffer (no peptide) was included but showed no binding. Controls with no protein or a protein other than Fn as ligands also gave a background signal (data not shown). Results are reported as A450nm and represent the mean and standard deviation from measurements of 3 experiments. C) Fn binding to rL1N peptide. The rL1N polypeptide was coated on to an ELISA plate and Fn added at increasing concentrations (8–250 nM). Bound Fn was detected by anti-Fn followed by a secondary anti-mouse HRP conjugate antibody. Relative binding is the ratio between the absorbance obtained in the presence of increasing amount of Fn to that obtained in the absence of Fn. Each point represents the mean and the standard deviation from 3 experiments.



View larger version (10K):
[in this window]
[in a new window]
  		Figure 3. Inhibition of latent TGF-ß activation by competition with a hinge peptide. CHO-ß6/ECR3E cells were transiently transfected with TGF-ß and 414-437 hinge expression constructs. After 16 h the cells were cocultured with TMLC in the presence of the hinge peptide (aa 404-440 of LTBP-1) or an unrelated peptide at different concentrations. 16–24 h later, latent TGF-ß activation was assessed by measuring luciferase activity in the cell lysates. Data are plotted as the % of the control sample, which received either no hinge peptide or no unrelated peptide. The sequence of the hinge peptide was GGMGYTVSGVHRRRPIHHHVGKGPVFVKPKNTQPVAK. This peptide represents amino acids 404-440 of the LTBP-1 hinge domain and was synthesized by the Rockefeller University Protein Chemistry facility. The sequence of the unrelated peptide was TSPDVDLGDISGINASVVNIQKEIDRLNEVAKNLNESLIDL.

Incorporation of 414-437 ECR3E construct into the matrix of control and Fn–/– cells
To test whether 414-437 ECR3E was targeted into the ECM and whether Fn affected this process, we analyzed the incorporation of 414-437 ECR3E into the matrix of control (carrying a floxed Fn allele) and Fn–/– (adenoviral Cre-mediated deletion of the floxed allele) mouse fibroblasts (clone 1-2). Cells were either transiently transfected with the 414-437 ECR3E construct or mock transfected; after 16–24 h the presence of 414-437 ECR3E, the solubilized ECM was analyzed by Western blot with an anti-HA antibody. 414-437 ECR3E was efficiently incorporated into the matrix of control cells but poorly incorporated into the ECM of Fn null cells (Fig. 4 A). Two control constructs 410-429 ECR3E and 426-449 ECR3E, which do not support latent TGF-ß activation (Fig. 1B ), were also tested for matrix incorporation. 426-449 ECR3E did not bind to the matrix produced by control or Fn–/– fibroblasts but 410-429 ECR3E did (Fig. 4A ). Therefore, even though 410-429 ECR3E binds to the ECM produced by either control or Fn–/– cells, it does not support latent TGF-ß activation (Fig. 1B ). The 414-437 ECR3E protein was found in ECM of Fn–/– cells cultured in the presence of cellular Fn, supporting the hypothesis that the hinge sequence interacts with Fn (Fig. 4B ). Western blot of medium from control and Fn–/– cells overexpressing the ECR3E constructs revealed no significant differences in the amounts of the secreted proteins in the conditioned medium (Fig. 3A, B ). This result indicates that differential incorporation into the matrix of control and Fn–/– cells was not due to differential secretion of the 414-437 ECR3E construct. Fn was only observed in the ECM of control cell cultures (Fig. 4C ). Thus, the active hinge region is preferentially incorporated into the ECM of Fn-producing cells.



View larger version (30K):
[in this window]
[in a new window]
  		Figure 4. Matrix incorporation of hinge constructs. A) Matrix incorporation of hinge constructs. Control and Fn–/– fibroblasts were transiently transfected with the indicated hinge ECR3E HA-derived constructs or a control DNA (mock). After 16–24 h, ECM incorporation of the hinge constructs from transfected cells was analyzed by SDS-PAGE and Western blot of DOC-insoluble material with an anti-HA monoclonal antibody followed by an anti-mouse HRP conjugate (upper panel). Secretion of the hinge constructs in the conditioned medium from the transfected cells was analyzed by Western blot (lower panel) as described above. B) Hinge incorporation with Fn. Control and Fn–/– fibroblasts were transiently transfected with the 414-437 ECR3E construct or mock transfected. Alternatively, Fn–/– cells transfected with 414-437 ECR3E were incubated with purified cellular Fn. Incorporation of the 414-437 ECR3E into the ECM (upper panel) and secretion in the conditioned medium (lower panel) were analyzed as described above. C) Fn in the ECM. The presence of Fn in the ECM of control and Fn–/– cells cultured in Fn-depleted FCS was analyzed by Western blot of DOC-insoluble material with an anti-Fn monoclonal antibody followed by an anti-mouse HRP conjugated antibody. The blot was developed as described above.

Hinge incorporation into CHOB2 and CHOB2/a27 cell matrixes
To test whether Fn assembly into a fibrillar matrix is required for the incorporation of 414-437 ECR3E into the ECM, we evaluated the ability of CHOB2 and CHOB2/a27 cells to retain the hinge construct in the DOC insoluble extract of their ECM. CHOB2 is a variant CHO cell line deficient for the Fn receptor {alpha}5ß1 integrin, whereas CHOB2/a27 are CHOB2 cells stably transfected with an {alpha}5 integrin expression vector (32) . CHOB2/a27 cells, but not the parental CHOB2 cell line, assemble a fibrillar Fn matrix as well as CHO control cells (32) . CHOB2 and CHOB2/a27 cells were transfected with the 414-437 ECR3E construct or mock transfected and the solubilized ECM analyzed (Fig. 5 A). The parental CHOB2 cells incorporated significantly less 414-437 ECR3E than did the CHOB2/a27. Incorporation of the hinge sequence into the DOC-insoluble matrix correlated with the amount of fibrillar Fn matrix in the two cell types as revealed by immunofluorescence with an antibody to Fn (Fig. 5B ).



View larger version (42K):
[in this window]
[in a new window]
  		Figure 5. Incorporation of the 414-437 LTBP-1 hinge-derived fragment into the ECM of CHOB2 and CHOB2/a27 transfected cells. A) ECM of CHOB2/CHOB2/a27: hinge incorporation. CHOB2, a CHO variant cell line missing the Fn receptor {alpha}5ß1 and CHOB2/a27, CHOB2 cells stably transfected with an {alpha}5 integrin expression construct, were transiently transfected with the 414-437 ECR3E construct or mock transfected. After 16–24 h, incorporation of 414-437 ECR3E into the ECM was analyzed by SDS-PAGE and Western blot of DOC-insoluble material with an anti-HA monoclonal antibody followed by an anti-mouse HRP conjugate. B) Fn matrix staining. CHOB2 and CHOB2/a27 cells were cultured in the presence of Fn and stained with an anti-Fn antibody followed by a secondary Cy3 conjugated antibody. Bar, 12 µm

TGF-ß activation by control and Fn null cells
To test the potential significance of Fn: hinge interaction, we evaluated {alpha}vß6-mediated latent TGF-ß activation by control and Fn–/– fibroblasts. Cells from two Fn–/– clones (1-2 and 2-8) and two companion control clones were transfected with a ß6 integrin expression vector, the ß6-overexpressing cells cocultured with TMLC, and after 16–24 h the TMLC were assayed for luciferase activity. As shown in Fig. 6 A, cells from the control cell clones activated latent TGF-ß when transfected with the ß6 integrin vector, whereas cells from the two Fn–/– cell clones did not. Latent TGF-ß activation by control/ß6 cells was inhibited by either a monoclonal anti-ß6 antibody or LAP, indicating that the signal in TMLC was both {alpha}vß6 and TGF-ß dependent (Fig. 6B ). TMLC bioassay of heat-activated conditioned medium (38) revealed that the total amount of secreted TGF-ß (endogenous or overexpressed after transfection with a TGF-ß construct) was equivalent in control and Fn–/– (clone 2-8) cells (Fig. 6C ). Therefore, differences in the activation were not due to a differential ability to secrete latent TGF-ß.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 6. {alpha}vß6-mediated TGF-ß activation in control and Fn–/– cells. A) TGF-ß production by Fn–/– cells. Control (clones 1 and 2) and Fn–/– (clones 1-2 and 2-8) cells were transfected with a ß6 expression vector. After 16–24 h, cells were cocultured with TMLC and luciferase activity measured. Relative luciferase activity is the ratio between the activities measured in ß6-expressing cells (white bar) or in mock transfected cells (black bar). B) TGF-ß specificity. Control (clone 1) cells were transiently transfected with a ß6 integrin expression vector. After 16–24 h, cells were cocultured with TMLC in the presence of LAP (200 ng/mL), anti ß6 antibody 10D5 (20 µg/mL), or no additions. Data are expressed as % of luciferase activity obtained in the absence of additions. C) Total TGF-ß production. Control (clone 1: black bar) and Fn–/– (clone 2-8; white bar) cells were transiently transfected with a TGF-ß expression vector (overexpressed) and total secreted TGF-ß was measured after heating the medium (80°C, 10 min). Relative luciferase activity is the ratio between activities measured in the TMLC assay in the presence of serum-free heated conditioned medium and in nonheated serum-free culture medium. Similar results were obtained with clone 1-2 cells. D) LTBP-1 expression. RT-PCR was performed on cDNA amplified from RNA extracted from control and Fn–/– (clone 1-2) cells nontransfected or transiently transfected with an LTBP-1 expression construct. LTBP-1 specific oligos were used for the amplification. A control lane is indicated by –. E) LTBP-1 production. Conditioned medium from control and Fn–/– (clone 1-2) cells mock transfected or transiently transfected with LTBP-1-HA cDNA was subjected to SDS-PAGE and Western blot with an anti-HA monoclonal antibody followed by an anti-mouse HRP conjugate. F) Cytoskeletal organization. Control or Fn–/– were plated on gelatin coverslips (plastic) in DMEM without serum for 4 h, fixed, permeabilized, and immunostained for paxillin (left image), filamentous actin (center image) cells, or merged with additional DAPI staining for nuclei (right image). The immunofluorescent staining revealed no major differences between control and Fn–/– cells, although clone 1-2, Fn–/– cells are slightly smaller. In the merged image paxillin staining is red, actin staining is green, and DAPI is blue.

Using RT-PCR, we evaluated LTBP-1 production by control and Fn–/– cells to insure that any effects on latent TGF-ß activities were not the result of decreased LTBP-1 expression. There was no significant difference in the amount of LTBP-1 mRNA levels in control and Fn–/– cells (clone 1-2) that were either untreated or transfected with a TGF-ß1 expression vector (Fig. 6D ). The same result was obtained with a second pair of LTBP-1-specific oligonucleotides (data not shown). To test for differences in LTBP-1 secretion, control and Fn–/– (clone 1-2) cells were transiently transfected with an LTBP-1-HA expression vector and the conditioned medium from the transfected cells analyzed by Western blot; no differences in the amount of secreted LTBP-1 were observed (Fig. 6E ). Similar results were obtained with clone 2-8 cells as well as cells cotransfected with a TGF-ß expression vector (data not shown). LTBP-1 levels in the ECM were not measured because of the difficulty of quantifying a protein that is cross-linked into the ECM (10 , 19) .

Because of the requirement for an intact cytoskeleton for ß6 integrin-mediated latent TGF-ß activation (13) , we monitored the distribution of paxillin, F-actin, and DAPI in wild-type and Fn–/– cells (Fig. 6F ). Although cells from the Fn–/– clone 1-2 displayed initial slight time-dependent and size differences compared with control cells with respect to spreading and cytoskeletal organization, cells from clone 2-8 appeared identical to control cells when paxillin and filamentous actin were characterized (Fig. 6F ). Moreover, when plated on Fn-coated dishes, cells from the 1-2 clone cells spread as well as control cells (data not shown). Therefore, the compromised ability of the Fn–/– cells to activate latent TGF-ß was not related to obvious differences in their cytoskeleton.

We attempted to complement the inability of the Fn–/– cells to activate latent TGF-ß in several ways. Fn–/– cells transfected with ß6 integrin were either plated on preformed matrix made by control cells or on cellular Fn, incubated with purified cellular Fn, or cocultured with control cells at ratios as high at 5 control to 1 null cell. None of these conditions rescued the ability of null cells to activate latent TGF-ß. Therefore, it appears that Fn must be synthesized by the cell activating latent TGF-ß and cannot be used once it is incorporated into a matrix.

Latent TGF-ß activation by {alpha}5-deficient CHO cells
As a further test of the requirement for organized Fn in {alpha}vß6-mediated latent TGF-ß activation, we checked the ability of CHOB2 and CHOB2/a27 cells to activate latent TGF-ß. CHOB2 and CHOB2/a27 cells were transfected with a ß6 integrin expression vector and after 16–24 h cocultured with the TMLC reporter cell line to measure TGF-ß generation (Fig. 7 A). Coculture of TMLC cells with CHO-ß6 cells was used as a positive control. Although there is some latent TGF-ß activation by ß6 integrins-expressing CHOB2 cells, the activation is significantly less than that observed with the activation obtained with ß6 integrin-overexpressing CHOB2/a27 or CHO-ß6 control cells (Fig. 7A ). The diminution in latent TGF-ß activation in the CHOB2 cells compared with the CHOB2/a27 cells is even more apparent if the mock-transfected values are subtracted. The remaining activation in CHOB2 cells may relate to the ability of CHOB2 cells to assemble Fn using {alpha}vß3 (40) . The level of integrin ß6 expression in the two cell types was equivalent as measured by RT-PCR (Fig. 7B ). These results further support the hypothesis that Fn assembly is required for efficient {alpha}vß6-mediated latent TGF-ß activation.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 7. Decreased latent TGF-ß activation by {alpha}5-deficient CHO cells. A) TGF-ß generation. To measure TGF-ß generation by CHOB2 and CHOB2/a27 cells, the cells were transiently transfected with a ß6 integrin expression vector or with a control DNA (mock). 16–24 h later, cells were cocultured with TMLC and the luciferase activity was measured. CHO-ß6 cells were also included as a positive control. Relative luciferase units (RLU) represent the value of the luciferase activity in the coculture divided by the value of the reporter cells alone. RLU values represent the means and standard deviation from 3 experiments. B) ß6 integrin expression. CHO2 and CHOB2/a27 were transiently transfected with a ß6 integrin expression vector or with a control DNA (mock). ß6 expression levels were measured by RT-PCR (30 cycles) on RNA extracted 16–24 h after transfection.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Increasing evidence suggests that extracellular presentation of growth factors is regulated in part by their binding to the ECM. For example, TGF-ß (18) , insulin-like growth factor (41) , fibroblast growth factor (42) , vascular endothelial growth factor (VEGF) (43) , and hepatocyte growth factor (HGF) (44) all associate with the ECM. Matrix association allows for storage of growth factors that are later readily available without a requirement for de novo biosynthesis. Some growth factors are stored in a latent form and are activated and released when needed (45 46 47) . Localization of the latent TGF-ß complex in the ECM is required as interference with latent TGF-ß association with the ECM blocks activation (48) .

Cells secrete a significant amount of latent TGF-ß, but only a small fraction is activated (49 , 50) . A number of activation mechanisms have been described, and the ECM plays an important role in several of these processes. Proteases associated with ECM degradation, such as plasmin and MMP-2 and -9 (51) , activate latent TGF-ß (52 , 53) , as does the matricellular protein TSP-1 (11) . Two other activation mechanisms involve integrins {alpha}vß6 or {alpha}vß8 (12 , 13) .

In {alpha}vß6-mediated activation of latent TGF-ß, the RGD domain of TGF-ß LAP interacts with integrin. This interaction must induce a conformational change leading to the liberation or exposure of TGF-ß. LTBP-1 participates in this process, and minimal regions of LTBP-1 sufficient for latent TGF-ß activation have been identified (14) . These sequences comprise the third CR repeat, which binds LAP, the flanking EGF-like repeats (EGF 13 and 14) (ECR3E), and the hinge region, encompassing amino acids 403-449. In this paper, we show that a 24 amino acid (aa 414-437) sequence of the hinge plus the ECR3E of LTBP-1 is sufficient to permit {alpha}vß6-mediated TGF-ß activation.

One possible function of the hinge is to anchor the TGF-ß complex through binding to an ECM component. LTBP-1 has been shown by immunoelectron microscopy to localize to microfibrillar structures in the ECM via amino acids 67-467 (20) . LTBP-1 colocalizes with Fn as shown by both double immunofluorescence and sandwich dot-blot assays (20 , 22) . A recent paper by Dallas et al. (54) described the colocalization of Fn and LTBP-1 and suggested that Fn is the major ECM localizer of LTBP-1. TGF-ß induces the synthesis of Fn (4) and promotes the incorporation of Fn into the ECM (4 , 55) . Moreover, TGF-ß up-regulates the expression of the Fn receptor {alpha}5ß1 (56 , 57) . Thus, TGF-ß and Fn may be part of a feed forward loop regulating ECM formation and turnover.

We demonstrated by blot overlay and ELISA assays that the rL1N polypeptide (aa 21-629 of LTBP-1) including the hinge amino acids 414-437 but not polypeptides rL1M and rL1C, which include the ECR3E region, binds Fn (Fig. 2) . Moreover, we demonstrated that the construct 414-437 ECR3E containing the hinge region 414-437 and the LAP binding domain 8-cys repeat 3 (CR3) plus the flanking EGF-like repeats (EGF 13 and 14) (ECR3E) is efficiently incorporated in the ECM of control cells but poorly into the ECM of Fn–/– cells (Fig. 4) . Although we did not demonstrate the interaction of the 414-437 peptide with Fn, the binding of the rL1N polypeptide to Fn, the enhanced incorporation of 414-437 ECR3E into the matrix of cells expressing Fn, and the rescue of incorporation into ECM in the presence of soluble Fn strongly suggest that the hinge sequence binds Fn. Moreover, the association of Fn binding and latent TGF-ß activation imply a mechanistic relationship between Fn binding to the hinge sequence and TGF-ß activation. Perhaps, this interaction is the biological counterpart of the interaction described by Annes et al. (2004) in which a non-{alpha}vß6 integrin activatable soluble form of LLC missing the N- and C-terminal LTBP-1 sequences was activated if the LTBP was bound to the culture disk surface by an antibody.

Dallas et al. (54) refer to unpublished results indicating that Fn and LTBP-1 interactions are indirect. Our binding data indicate direct interactions. At present, we have no explanation for the differences between our results and those of Dallas et al. (54) , but variations in the protocols may explain the different outcomes.

Matrix incorporation by itself is not sufficient to achieve {alpha}vß6-mediated activation of latent TGF-ß. The construct 410-429 ECR3E, which does not promote TGF-ß activation (Fig. 1B ), is incorporated into the matrix of control cells with the same efficiency as the 414-437 ECR3E construct (Fig. 4B ). Unlike 414-437 ECR3E, the 410-429 ECR3E construct is also incorporated in the matrix of Fn–/– cells, indicating that this sequence binds an ECM component other than Fn. This result indicates that incorporation into the ECM is necessary but not sufficient for {alpha}vß6-mediated activation of latent TGF-ß; binding to Fn is required. This raises the question of whether binding of the LLC to additional components of the ECM facilitates activation by other mediators. Such a result would suggest potential regulation of latent TGF-ß by localization of the LLC to different molecules or sites in the matrix.

The importance of Fn in activation by {alpha}vß6 is demonstrated by the failure of Fn–/– fibroblasts to activate latent TGF-ß and the suppression of latent TGF-ß activation by cells with a deficiency in Fn assembly. However, the precise function of Fn in activation is unclear. The ability to generate a Fn fibrillar matrix is not sufficient to permit latent TGF-ß activation as Fn–/– cells expressing ß6 integrin are unable to activate latent TGF-ß when plated on a preformed matrix made by control cells (data not shown) or when cultured in the presence of soluble Fn (data not shown). These facts suggest a mechanism in which the Fn involved in latent TGF-ß activation must be produced and organized by the activating cell at the time of activation and in a local environment, perhaps indicating a requirement for binding to an epitope that exists only transiently as cellular Fn is assembled into a matrix. Experiments to test this are under way.


   ACKNOWLEDGMENTS
 
The authors thank Melinda Vassallo and Rob Ono for expert technical assistance. This work was supported by grants CA34282 and CA78422 (DBR), the Shriners Hospital for Children (L.Y.S.), and DFG (SFB 576) and Fonds der Chemischen Industrie (R.F.).

Received for publication April 18, 2005. Accepted for publication July 18, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Massague, J. (1990) The transforming growth factor-beta family. Annu. Rev. Cell Biol. 6,597-641[CrossRef][Medline]
  2. Wrann, M., Bodmer, S., de Martin, R., Siepl, C., Hofer-Warbinek, R., Frei, K., Hofer, E., Fontana, A. (1987) T cell suppressor factor from human glioblastoma cells is a 12.5-kd protein closely related to transforming growth factor-beta. EMBO J. 6,1633-1636[Medline]
  3. Gohongi, T., Fukumura, D., Boucher, Y., Yun, C. O., Soff, G. A., Compton, C., Todoroki, T., Jain, R. K. (1999) Tumor-host interactions in the gallbladder suppress distal angiogenesis and tumor growth: involvement of transforming growth factor beta1. Nat. Med. 5,1203-1208[CrossRef][Medline]
  4. Ignotz, R. A., Massague, J. (1986) Transforming growth factor-beta stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J. Biol. Chem. 261,4337-4345[Abstract/Free Full Text]
  5. Roberts, A. B., Sporn, M. B., Assoian, R. K., Smith, J. M., Roche, N. S., Wakefield, L. M., Heine, U. I., Liotta, L. A., Falanga, V., Kehrl, J. H., et al (1986) Transforming growth factor type beta: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc. Natl. Acad. Sci. USA 83,4167-4171[Abstract/Free Full Text]
  6. Blobe, G. C., Schiemann, W. P., Lodish, H. F. (2000) Role of transforming growth factor beta in human disease. N. Engl. J. Med. 342,1350-1358[Free Full Text]
  7. Gentry, L. E., Lioubin, M. N., Purchio, A. F., Marquardt, H. (1988) Molecular events in the processing of recombinant type 1 pre-pro-transforming growth factor beta to the mature polypeptide. Mol. Cell. Biol. 8,4162-4168[Abstract/Free Full Text]
  8. Miyazono, K., Hellman, U., Wernstedt, C., Heldin, C. H. (1988) Latent high molecular weight complex of transforming growth factor beta 1. Purification from human platelets and structural characterization. J. Biol. Chem. 263,6407-6415[Abstract/Free Full Text]
  9. Brown, P. D., Wakefield, L. M., Levinson, A. D., Sporn, M. B. (1990) Physicochemical activation of recombinant latent transforming growth factor-beta’s 1, 2, and 3. Growth Factors 3,35-43[Medline]
  10. Koli, K., Saharinen, J., Hyytiainen, M., Penttinen, C., Keski-Oja, J. (2001) Latency, activation, and binding proteins of TGF-beta. Microsc. Res. Tech. 52,354-362[CrossRef][Medline]
  11. Murphy-Ullrich, J. E., Poczatek, M. (2000) Activation of latent TGF-beta by thrombospondin-1: mechanisms and physiology. Cytokine Growth Factor Rev. 11,59-69[CrossRef][Medline]
  12. Mu, D., Cambier, S., Fjellbirkeland, L., Baron, J. L., Munger, J. S., Kawakatsu, H., Sheppard, D., Broaddus, V. C., Nishimura, S. L. (2002) The integrin alpha(v)beta8 mediates epithelial homeostasis through MT1-MMP-dependent activation of TGF-beta1. J. Cell Biol. 157,493-507[Abstract/Free Full Text]
  13. Munger, J. S., Huang, X., Kawakatsu, H., Griffiths, M. J., Dalton, S. L., Wu, J., Pittet, J. F., Kaminski, N., Garat, C., Matthay, M. A., et al (1999) The integrin alpha v beta 6 binds and activates latent TGF beta 1: a mechanism for regulating pulmonary inflammation and fibrosis. Cell 96,319-328[CrossRef][Medline]
  14. Annes, J. P., Chen, Y., Munger, J. S., Rifkin, D. B. (2004) Integrin {alpha}vß6-mediated activation of latent TGF-ß requires the latent TGF-ß binding protein-1. J. Cell Biol. 165,723-734[Abstract/Free Full Text]
  15. Huang, X. Z., Wu, J. F., Cass, D., Erle, D. J., Corry, D., Young, S. G., Farese, R. V., Jr, Sheppard, D. (1996) Inactivation of the integrin beta 6 subunit gene reveals a role of epithelial integrins in regulating inflammation in the lung and skin. J. Cell Biol. 133,921-928[Abstract/Free Full Text]
  16. Saharinen, J., Keski-Oja, J. (2000) Specific sequence motif of 8-Cys repeats of TGF-beta binding proteins, LTBPs, creates a hydrophobic interaction surface for binding of small latent TGF-beta. Mol. Biol. Cell 11,2691-2704[Abstract/Free Full Text]
  17. Miyazono, K., Olofsson, A., Colosetti, P., Heldin, C. H. (1991) A role of the latent TGF-beta 1-binding protein in the assembly and secretion of TGF-beta 1. EMBO J. 10,1091-1101[Medline]
  18. Taipale, J., Miyazono, K., Heldin, C. H., Keski-Oja, J. (1994) Latent transforming growth factor-beta 1 associates to fibroblast extracellular matrix via latent TGF-beta binding protein. J. Cell Biol. 124,171-181[Abstract/Free Full Text]
  19. Nunes, I., Gleizes, P. E., Metz, C. N., Rifkin, D. B. (1997) Latent transforming growth factor-beta binding protein domains involved in activation and transglutaminase-dependent cross-linking of latent transforming growth factor-beta. J. Cell Biol. 136,1151-1163[Abstract/Free Full Text]
  20. Dallas, S. L., Keene, D. R., Bruder, S. P., Saharinen, J., Sakai, L. Y., Mundy, G. R., Bonewald, L. F. (2000) Role of the latent transforming growth factor beta binding protein 1 in fibrillin-containing microfibrils in bone cells in vitro and in vivo. J. Bone Miner. Res. 15,68-81[CrossRef][Medline]
  21. Unsold, C., Hyytiainen, M., Bruckner-Tuderman, L., Keski-Oja, J. (2001) Latent TGF-beta binding protein LTBP-1 contains three potential extracellular matrix interacting domains. J. Cell Sci. 114,187-197[Abstract]
  22. Taipale, J., Saharinen, J., Hedman, K., Keski-Oja, J. (1996) Latent transforming growth factor-beta 1 and its binding protein are components of extracellular matrix microfibrils. J. Histochem. Cytochem. 44,875-889[Abstract]
  23. Pankov, R., Yamada, K. M. (2002) Fibronectin at a glance. J. Cell Sci. 115,3861-3863[Free Full Text]
  24. Hynes, R. O., Yamada, K. M. (1982) Fibronectins: multifunctional modular glycoproteins. J. Cell Biol. 95,369-377[Free Full Text]
  25. George, E. L., Georges-Labouesse, E. N., Patel-King, R. S., Rayburn, H., Hynes, R. O. (1993) Defects in mesoderm, neural tube and vascular development in mouse embryos lacking fibronectin. Development 119,1079-1091[Abstract]
  26. ffrench-Constant, C. (1995) Alternative splicing of fibronectin–many different proteins but few different functions. Exp. Cell Res. 221,261-271[CrossRef][Medline]
  27. Kornblihtt, A. R., Vibe-Pedersen, K., Baralle, F. E. (1984) Human fibronectin: molecular cloning evidence for two mRNA species differing by an internal segment coding for a structural domain. EMBO J. 3,221-226[Medline]
  28. Balza, E., Borsi, L., Allemanni, G., Zardi, L. (1988) Transforming growth factor beta regulates the levels of different fibronectin isoforms in normal human cultured fibroblasts. FEBS Lett. 228,42-44[CrossRef][Medline]
  29. Taipale, J., Saharinen, J., Keski-Oja, J. (1998) Extracellular matrix-associated transforming growth factor-beta: role in cancer cell growth and invasion. Adv. Cancer Res. 75,87-134[Medline]
  30. Sakai, T., Johnson, K. J., Murozono, M., Sakai, K., Magnuson, M. A., Wieloch, T., Cronberg, T., Isshiki, A., Erickson, H. P., Fassler, R. (2001) Plasma fibronectin supports neuronal survival and reduces brain injury following transient focal cerebral ischemia but is not essential for skin-wound healing and hemostasis. Nat. Med. 7,324-330[CrossRef][Medline]
  31. Fassler, R., Pfaff, M., Murphy, J., Noegel, A. A., Johansson, S., Timpl, R., Albrecht, R. (1995) Lack of beta 1 integrin gene in embryonic stem cells affects morphology, adhesion, and migration but not integration into the inner cell mass of blastocysts. J. Cell Biol. 128,979-988[Abstract/Free Full Text]
  32. Wu, C., Bauer, J. S., Juliano, R. L., McDonald, J. A. (1993) The alpha 5 beta 1 integrin fibronectin receptor, but not the alpha 5 cytoplasmic domain, functions in an early and essential step in fibronectin matrix assembly. J. Biol. Chem. 268,21883-21888[Abstract/Free Full Text]
  33. Weinacker, A., Chen, A., Agrez, M., Cone, R. I., Nishimura, S., Wayner, E., Pytela, R., Sheppard, D. (1994) Role of the integrin alpha v beta 6 in cell attachment to fibronectin. Heterologous expression of intact and secreted forms of the receptor. J. Biol. Chem. 269,6940-6948[Abstract/Free Full Text]
  34. Abe, M., Harpel, J. G., Metz, C. N., Nunes, I., Loskutoff, D. J., Rifkin, D. B. (1994) An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal. Biochem. 216,276-284[CrossRef][Medline]
  35. Huang, X., Wu, J., Spong, S., Sheppard, D. (1998) The integrin alphavbeta6 is critical for keratinocyte migration on both its known ligand, fibronectin, and on vitronectin. J. Cell Sci. 111,2189-2195[Abstract]
  36. Munger, J. S., Harpel, J. G., Giancotti, F. G., Rifkin, D. B. (1998) Interactions between growth factors and integrins: latent forms of transforming growth factor-beta are ligands for the integrin alphavbeta1. Mol. Biol. Cell 9,2627-2638[Abstract/Free Full Text]
  37. Isogai, Z., Ono, R. N., Ushiro, S., Keene, D. R., Chen, Y., Mazzieri, R., Charbonneau, N. L., Reinhardt, D. P., Rifkin, D. B., Sakai, L. Y. (2003) Latent transforming growth factor beta-binding protein 1 interacts with fibrillin and is a microfibril-associated protein. J. Biol. Chem. 278,2750-2757[Abstract/Free Full Text]
  38. Mazzieri, R., Munger, J. S., Rifkin, D. B. (2000) Measurement of active TGF-beta generated by cultured cells. Methods Mol. Biol. 142,13-27[Medline]
  39. Verderio, E., Gaudry, C., Gross, S., Smith, C., Downes, S., Griffin, M. (1999) Regulation of cell surface tissue transglutaminase: effects on matrix storage of latent transforming growth factor-beta binding protein-1. J. Histochem. Cytochem. 47,1417-1432[Abstract/Free Full Text]
  40. Wennerberg, K., Lohikangas, L., Gullberg, D., Pfaff, M., Johansson, S., Fassler, R. (1996) Beta 1 integrin-dependent and -independent polymerization of fibronectin. J. Cell Biol. 132,227-238[Abstract/Free Full Text]
  41. Jones, J. I., Gockerman, A., Busby, W. H., Jr, Camacho-Hubner, C., Clemmons, D. R. (1993) Extracellular matrix contains insulin-like growth factor binding protein-5: potentiation of the effects of IGF-I. J. Cell Biol. 121,679-687[Abstract/Free Full Text]
  42. Faham, S., Hileman, R. E., Fromm, J. R., Linhardt, R. J., Rees, D. C. (1996) Heparin structure and interactions with basic fibroblast growth factor. Science 271,1116-1120[Abstract]
  43. Park, J. E., Keller, G. A., Ferrara, N. (1993) The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular matrix-bound VEGF. Mol. Biol. Cell 4,1317-1326[Abstract]
  44. Lyon, M., Deakin, J. A., Mizuno, K., Nakamura, T., Gallagher, J. T. (1994) Interaction of hepatocyte growth factor with heparan sulfate. Elucidation of the major heparan sulfate structural determinants. J. Biol. Chem. 269,11216-11223[Abstract/Free Full Text]
  45. Naldini, L., Tamagnone, L., Vigna, E., Sachs, M., Hartmann, G., Birchmeier, W., Daikuhara, Y., Tsubouchi, H., Blasi, F., Comoglio, P. M. (1992) Extracellular proteolytic cleavage by urokinase is required for activation of hepatocyte growth factor/scatter factor. EMBO J. 11,4825-4833[Medline]
  46. Taipale, J., Koli, K., Keski-Oja, J. (1992) Release of transforming growth factor-beta 1 from the pericellular matrix of cultured fibroblasts and fibrosarcoma cells by plasmin and thrombin. J. Biol. Chem. 267,25378-25384[Abstract/Free Full Text]
  47. Partenheimer, A., Schwarz, K., Wrocklage, C., Kolsch, E., Kresse, H. (1995) Proteoglycan form of colony-stimulating factor-1 (proteoglycan-100). Stimulation of activity by glycosaminoglycan removal and proteolytic processing. J. Immunol. 155,5557-5565[Abstract]
  48. Flaumenhaft, R., Abe, M., Sato, Y., Miyazono, K., Harpel, J., Heldin, C. H., Rifkin, D. B. (1993) Role of the latent TGF-beta binding protein in the activation of latent TGF-beta by cocultures of endothelial and smooth muscle cells. J. Cell Biol. 120,995-1002[Abstract/Free Full Text]
  49. Sato, Y., Rifkin, D. B. (1988) Autocrine activities of basic fibroblast growth factor: regulation of endothelial cell movement, plasminogen activator synthesis, and DNA synthesis. J. Cell Biol. 107,1199-1205[Abstract/Free Full Text]
  50. Annes, J. P., Munger, J. S., Rifkin, D. B. (2003) Making sense of latent TGFbeta activation. J. Cell Sci. 116,217-224[Abstract/Free Full Text]
  51. Werb, Z. (1997) ECM and cell surface proteolysis: regulating cellular ecology. Cell 91,439-442[CrossRef][Medline]
  52. Lyons, R. M., Gentry, L. E., Purchio, A. F., Moses, H. L. (1990) Mechanism of activation of latent recombinant transforming growth factor beta 1 by plasmin. J. Cell Biol. 110,1361-1367[Abstract/Free Full Text]
  53. Yu, Q., Stamenkovic, I. (2000) Cell surface-localized matrix metalloproteinase-9 proteolytically activates TGF-beta and promotes tumor invasion and angiogenesis. Genes Dev. 14,163-176[Abstract/Free Full Text]
  54. Dallas, S. L., Sivakumar, P., Jones, C. J., Chen, Q., Peters, D. M., Mosher, D. F., Humphries, M. J., Kielty, C. M. (2005) Fibronectin regulates latent TGBbeta by controlling matrix assembly of latent TGFbeta binding protein 1. J. Biol. Chem. In press
  55. Muller, G., Behrens, J., Nussbaumer, U., Bohlen, P., Birchmeier, W. (1987) Inhibitory action of transforming growth factor beta on endothelial cells. Proc. Natl. Acad. Sci. USA 84,5600-5604[Abstract/Free Full Text]
  56. Collo, G., Pepper, M. S. (1999) Endothelial cell integrin alpha5beta1 expression is modulated by cytokines and during migration in vitro. J. Cell Sci. 112,569-578[Abstract]
  57. Cai, T., Lei, Q. Y., Wang, L. Y., Zha, X. L. (2000) TGF-beta 1 modulated the expression of alpha 5 beta 1 integrin and integrin-mediated signaling in human hepatocarcinoma cells. Biochem. Biophys. Res. Commun. 274,519-525[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
Q. Chen, P. Sivakumar, C. Barley, D. M. Peters, R. R. Gomes, M. C. Farach-Carson, and S. L. Dallas
Potential Role for Heparan Sulfate Proteoglycans in Regulation of Transforming Growth Factor-beta (TGF-beta) by Modulating Assembly of Latent TGF-beta-binding Protein-1
J. Biol. Chem., September 7, 2007; 282(36): 26418 - 26430.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Biol.Home page
S. S. Chaudhry, S. A. Cain, A. Morgan, S. L. Dallas, C. A. Shuttleworth, and C. M. Kielty
Fibrillin-1 regulates the bioavailability of TGF{beta}1
J. Cell Biol., January 29, 2007; 176(3): 355 - 367.
[Abstract]