INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The development of alveolar fibrosis after lung injury may lead to significant morbidity and mortality. Survival of patients with lung injury requires the elimination of excess fibroblasts from the air spaces, permitting restoration of anatomic integrity to damaged alveolar units. Failure to eliminate fibrotic air space tissue in a timely fashion is associated with death in up to 40% of nonsurviving patients. Unfortunately, no safe and effective therapy exists for targeting the removal of excess air space fibroblasts. However, recent work indicates that myofibroblast apoptosis occurs within the alveolar air space during the repair phase after lung injury (1). This suggests a role for apoptosis in eliminating excess fibrotic tissue and allowing restoration of normal alveolar anatomy. Therapies designed to promote myofibroblast apoptosis during the repair phase following lung injury, thereby emulating physiologic healing, might facilitate recovery.

The survival of many anchorage-dependent cells requires integrin-mediated adhesion to the extracellular matrix (ECM) (2). In the absence of appropriate ECM contacts, adhesion- dependent cells may undergo apoptosis. For example, the viability of epithelial and endothelial cells requires survival signals generated by the ligation of integrin receptor to the ECM (3).

A number of functionally distinct domains of the ECM protein fibronectin have been identified (Figure 1) (6, 7), including the tripeptide arginine-glycine-aspartic acid (RGD) within the cell-binding domain (8), the CS-1 peptide (LDV sequence) in the alternately spliced type IIICS connecting-segment domain (9), and the FN-C/H-V peptide in the 33/66-kD heparin-binding fragment (10). These three peptides promote cell migration, adhesion, and spreading. By blocking adhesion to the ECM, soluble fibronectin peptides such as RGD trigger apoptosis of epithelial cells (4), endothelial cells (11), and transformed fibroblasts (12).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Location of synthetic peptides within the fibronectin molecule. The three different types of internal homology units are shown. Shaded areas indicate the alternatively spliced segments (ED-B, ED-A, and IIICS). Labels along the top of the figure indicate functional binding domains. Locations of the three fibronectin peptides (RGD, CS-1 and FN-C/H-V) used in the study are shown.

Treatment of fibroproliferative lung diseases has largely been relegated to blocking inflammation. However, the clinical benefit of such treatment is limited despite excellent antiinflammatory efficacy of the agents used, suggesting that therapy targeting fibroblasts is essential. In the study reported here we found that soluble fibronectin peptides trigger apoptosis both of adherent fibroblasts in routine tissue culture and of fibroblasts incorporated into a fibrin gel, through a mechanism that includes disruption of integrin-mediated survival signaling. Our data therefore suggest that the use of small, soluble fibronectin peptides to eliminate fibroblasts may represent a novel approach to the treatment of lung fibrosis.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell Lines

Human lung fibroblasts were prepared and characterized as previously described (13). Briefly, three primary cultures of lung fibroblasts were developed from three patients with acute lung injury who died from respiratory failure at 14, 21, and 30 d after disease onset, respectively. These patients met established criteria for the adult respiratory distress syndrome, and their lung biopies showed diffuse alveolar damage and intraalveolar fibrosis. Immunohistochemical studies indicated that the cells were positive for determinants characteristic of mesenchymal cells, but were negative for epithelial and endothelial determinants as previously described (13). Staining for -smooth-muscle actin revealed a heterogenous population of cells, with approximately 45% to 55% of cells staining positive.

All human fibroblasts were cultured (10% CO₂, 90% air at 37° C) in Dulbecco's modified Eagle's medium (DMEM) containing 10% heat-inactivated fetal calf serum (FCS), 100 U/L penicillin, 100 µg/L streptomycin, and 250 ng/L amphotericin B, and were used before the tenth subcultivation. For routine maintenance, medium was replaced twice weekly and cells were subcultivated at a 1:2 ratio. Two diploid fibroblast strains, normal human lung fibroblasts (ATCC 210; American Type Culture Collection, Rockville, MD) and human fetal skin fibroblasts (ATCC 1475), were also used.

Fibronectin Peptides

Fibronectin polypetides were synthesized at the Microchemical facility of the University of Minnesota with a Beckman System 990 peptide synthesizer (Beckman Instrument Inc., Fullerton, CA), and were purified as previously described (14). Peptides used in the study included RGD, CS-1, and one sequence of the 33 kD fragment from the heparin domain of the fibronectin A chain. The scrambled version of the FN-C/H-V peptide and the RGD and CS-1 peptides containing a mutation in their binding sequences were used as controls (Table 1). Sequence analysis was done initially to confirm the composition of the peptides. Thereafter, amino acid analyses were done to confirm the composition of the peptides.

Fibrin Gels

Gels composed of fibrin were prepared under sterile conditions through a modification of the protocol previously described (15). Cells were trypsinized, washed, and resuspended in a 4-(2-hydroxyethyl)-1-piperazine-N'-2-ethanesulfonic acid (Hepes)-saline buffer (0.13 M NaCl, 0.025 M Hepes, and 0.005 M CaCl₂, pH 7.4) containing rabbit fibrinogen (3 mg/ml; Sigma Chemical Co., St. Louis, MO) at a concentration of 5 × 10⁵ cells/ml. Aprotinin (Boehringer Mannheim Corporation, Indianapolis, IN), a plasmin inhibitor, was routinely added to the medium at a concentration of 2 µg/ml. The cell/fibrinogen suspension was placed in a 0.4-µm-pore-size cell culture insert (500 µl/insert; Becton Dickinson, Franklin Lakes, NJ). The inserts were placed in Falcon 24-well dishes, and the cell/fibrinogen solution was clotted with 1 U/ml human thrombin (Sigma) and allowed to polymerize (1 h at 37° C). DMEM containing 0.1% heat-inactivated FCS and 2 µg/ml of aprotinin was added on the top and bottom of the gel. Fibrin gels were incubated (37° C for 12 h) in a humidified incubator. Medium was removed and DMEM containing 0.1% heat-inactivated FCS and 2 µg/ml of aprotinin was added to the top and surrounding gel in the presence of the three fibronectin peptides (RGD, CS-1, and FN-C/ H-V) at concentrations of 500 µg/ml each. Medium was changed at 48 h and the fibrin gels were studied at 96 h.

Analysis of Apoptosis for Fibroblasts Cultured on Tissue Culture Plates and in Fibrin Gels

Cells were examined morphologically for apoptosis through three methods: phase-contrast microscopy, fluorescence microscopy after acridine orange staining (1), and in situ labeling with the terminal deoxynucleotidyl transferase-uridine nucleotide nick-end labeling (TUNEL) assay, using the POD kit for in situ detection of cell death (Boehringer Mannheim) as previously described (16). To study morphologic changes in cells, we fixed fibrin gels in 10% buffered formalin, embedded them in paraffin, sectioned (4 µm) them, and stained the cells with hematoxylin and eosin (H&E). To estimate the percentage of apoptotic cells in fibrin gels, we identified nuclei containing degraded DNA through the TUNEL assay. To assess apoptosis of fibroblasts plated on tissue culture plastic and treated with soluble fibronectin peptides ,we studied DNA integrity with flow cytometry (FACS Caliber; Becton Dickinson, Lincoln Park, NJ) after propidium iodide staining. As an additional corroborative assay for apoptosis, we analyzed cells on tissue culture slides with the TUNEL assay.

Induction of Apoptosis by Fibronectin Peptides

Human lung fibroblasts were seeded into six-well clusters at 1.3 × 10⁴ cells/cm² and cultured in DMEM containing 10% FCS for 2 h. After the removal of medium, cells were washed with phosphate-buffered saline (PBS) and incubated overnight (16 h) with medium containing 0.1% FCS before being treated with fibronectin peptides. For kinetic studies, fibroblasts were washed twice with PBS and incubated for specific intervals with DMEM + 0.1% FCS containing 500 µg/ml of each of the three synthetic fibronectin peptides RGD, CS-1, and FN-C/H-V. Control assays were run with untreated cells and cells treated with the control peptides. The percentage of cells undergoing apoptosis was determined both by quantification of DNA fragmentation through analysis of relative DNA content, using flow cytometry after propidium iodide staining as previously described (1), and by TUNEL analysis.

Suspension Cultures

Suspension cultures were prepared by coating 35-mm tissue culture dishes with 1 ml of 10 mg/ml poly(2-hydroxyethyl methacrylate) (poly[HEMA]; Sigma) dissolved in 95% ethanol and allowing the solution to evaporate at 37° C overnight (4). Suspended human lung fibroblasts were seeded into six-well plates at 1.3 × 10⁴ cells/cm² and were cultured in DMEM containing 0.1% FCS for various periods (8, 16, 24, 32 h) in the absence and presence of RGD, CS-1, and FN-C/ H-V peptides (500 µg/ml).

Western Blot Assay for pp125FAK

Primary lung fibroblasts were cultured in DMEM + 0.1% FCS in the presence of soluble fibronectin peptides or control peptides (72 h, 500 µg/ml). Attached and floating cells were combined and lysed in reducing gel sample buffer (50 mM Tris, pH 7.2; 150 mM NaCl; 1% Triton X-100; 1 mM N-ethylmaleimide; 0.25% Na-deoxycholate; 50 mM NaF; 2 mM Na-orthovanadate; 1 mM phenhylmethyl sulfonyl fluoride, 2 µg/ml aprotinin; 10 µg/ml leupeptin). For each sample, 40 µg of total protein were electrophoresed on 8% sodium dodecyl sulfate- polyacrylamide gels, blotted onto nitrocellulose, and probed with anti-pp125FAK antibody (Upstate Biotechnology, Lake Placid, NY). Immunoreactive bands were visualized by incubation with a goat anti-rabbit IgG (1:5,000 dilution) conjugated with horseradish peroxidase (Sigma), with subsequent development through enhanced chemiluminescence (Amersham, Arlington, IL). Negative controls consisted of nitrocellulose membranes incubated first with normal rabbit serum and then with the antirabbit secondary antibody.

Data Analysis

All data are expressed as mean ± SEM. Experiments were performed three or four times. For the flow-cytometric analyses, all samples were run at least in duplicate. For assessment of the percentage of apoptotic cells in fibrin gels and on slides from cell culture with the TUNEL assay, we used microscopic analysis of at least 200 cells per gel or per slide. Paired evaluations were made for experimental and control conditions within each experiment, and significance was determined with Student's t test. The level of significance was set at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Soluble Fibronectin Peptides Induce Apoptosis of Adherent Nontransformed Fibroblasts

Earlier studies indicate that both epithelial and endothelial cells undergo apoptosis when the cell-matrix interaction is disrupted (2). However, the effect of disrupting the cell-matrix interaction of nontransformed fibroblasts through the use of small fibronectin peptides, which is the primary goal for antifibrotic therapies, has not been determined. In our study, adherent primary human lung fibroblasts incubated for 72 h at 37° C with soluble RGD, CS-1, and FN-C/H-V peptides (500 µg/ ml, each) underwent typical apoptotic morphologic changes, including cytoplasmic condensation and plasma membrane blebbing (Figure 2A, right). Analysis of fibroblast morphology after staining with acridine orange revealed pronounced chromatin condensation and fragmentation, features characteristic of programmed cell death (Figure 2B, right). Control peptides consisting of the scrambled version of the FN-C/H-V peptide and the RGE, and LDE peptides had no effect on cell viability (Figures 2A and 2B, left). To determine whether the ability of fibronectin peptides to induce apoptosis was restricted to primary lung fibroblasts or whether other fibroblasts were similarly susceptible, we used two additional strains of fibroblasts (ATCC 1475 and ATCC 210) in the same culture system used for the primary lung fibroblasts. Fibronectin peptides induced apoptosis in these cells in a similar manner, as in the primary fibroblasts, indicating that the ability of the fibronectin peptides to trigger fibroblast apoptosis was a general phenomenon and not restricted to primary lung fibroblasts. Furthermore, there was no obvious difference in the susceptibility of -smooth-muscle actin-positive and -negative primary human lung fibroblasts to treatment with the fibronectin peptides (data not shown).

View larger version (89K): [in this window] [in a new window]   Figure 2.   Fibronectin peptides induce fibroblast apoptosis. Adherent primary human lung fibroblasts were incubated for 72 h in DMEM + 0.1% FCS containing 500 µg/ml each of RGD, CS-1, and FN-C/H-V peptides or control peptides (scrambled version of FN-C/H-V, RGE, and LDE peptides). (A) Apoptosis was identified by characteristic morphologic changes seen by phase contrast microscopy. Left: Phase contrast microscopic appearance of cells treated with control peptides displaying normal morphologic features. Right: Characteristic morphologic features of apoptosis were present in fibroblasts treated with soluble fibronectin peptides. Note prominent plasma membrane blebbing (arrows) and pyknotic nuclei (original magnification: ×200). (B) Analysis of fibroblast morphology after acridine orange staining. Left: Note normal nuclear morphology in cells treated with control peptides. Right: Acridine orange staining of fibroblasts treated with soluble fibronectin peptides revealed pronounced chromatin condensation and fragmentation, features characteristic of apoptosis (original magnification: ×200; inset: ×400).

The induction of apoptosis in lung fibroblasts was done under low serum-content (0.1% FCS) conditions in nonconfluent cultures of fibroblasts. Serum-starving of the fibroblasts for longer periods (48 h) to attain quiescence before treatment with fibronectin peptides did not alter the apoptotic effect of the peptides, nor did treatment of nonconfluent cultures of fibroblasts with hydroxyurea, to achieve cell-cycle arrest, alter the apoptotic response of these cells to the peptides. However, when the cells were allowed to become densely confluent or were cultured under high serum-content (10% FCS) conditions they became resistant to the effect of the peptides.

Kinetics and Dose Dependency of Fibroblast Apoptosis in Response to Treatment with Soluble Fibronectin Peptides

The kinetics of fibronectin-peptide induction of fibroblast apoptosis were analyzed with primary human lung fibroblasts plated onto tissue culture plates (Figure 3A). Apoptosis was quantified by flow cytometry after propidium iodide staining. After 48 h of incubation, 40 ± 4.5% (mean ± SEM) of fibroblasts exposed to soluble fibronectin peptides (500 µg/ml) were apoptotic. By 96 h, 87 ± 16% of treated fibroblasts had undergone apoptosis. In contrast, at 96 h, only 7 ± 5% of fibroblasts treated with scrambled control peptides or untreated fibroblasts were apoptotic (p < 0.05). This corresponds to an 88% difference in the level of apoptosis achieved with fibronectin peptides and scrambled peptides. Furthermore, in accord with the foregoing results obtained with primary cultures of fibroblasts, the percentages of ATCC 1475 and ATCC 210 fibroblasts undergoing apoptosis at 96 h after peptide treatment were 84 ± 12% and 81 ± 9%, respectively. Fibroblast apoptosis induced by fibronectin peptides occurred in a dose-dependent manner (Figure 3B). An effect was first detected at a 0.5 µg/ml concentration of peptide. The proportion of apoptotic cells increased gradually thereafter, until approximately 60% of the cells underwent apoptosis at a 500 µg/ml concentration at 96 h. The results of flow-cytometric analysis shown in Figure 3 for both the kinetic and dose-dependency studies are representative of four separate experiments.

View larger version (14K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   Figure 3.   Kinetics and dose dependency of fibroblast apoptosis in response to soluble fibronectin peptides. (A) Control (open squares) and soluble fibronectin peptide (sFNP; 500 µg/ml of RGD, CS-1, and FN-C/H-V)-treated (closed squares) human lung fibroblasts were incubated in DMEM + 0.1% FCS for 4 d. Cells were harvested at different time points, fixed with 70% ethanol, and stained with propidium iodide for flow cytometric analysis. Shown are the percentage of apoptotic cells as determined with flow cytometry. (B) Adherent human lung fibroblasts were cultured with various concentrations of fibronectin peptides (0, 0.5, 5, 50, and 500 µg/ml) for 96 h. Shown are the percentages of apoptotic cells as determined with flow cytometry.

It should be noted that as compared with epithelial and endothelial cells, fibroblasts were more difficult to kill (2, 3). Prior studies as well as our own work indicate that both epithelial and endothelial cells rapidly undergo apoptosis in response to soluble fibronectin peptides, with evidence of DNA fragmentation occurring within 8 h and nearly 100% apoptosis of cells by 24 h (data not shown) (2, 3). In contrast, we found that relatively high concentrations of peptides, over a longer time period, were required to induce apoptosis in fibroblasts. Evidence of significant DNA fragmentation by fluorescence-activated cell sorting analysis was apparent at 24 h. However, it required a 2- to 3-d incubation of fibroblasts with peptides to reach the ~ 50% apoptosis level (Figure 3B).

The Combination of All Three Soluble Fibronectin Peptides is Required For Maximal Fibroblast Apoptosis

To determine whether any one of the three synthetic peptides used in the study had a more pronounced effect on induction of fibroblast apoptosis than the others, we tested each of the three fibronectin peptides alone (RGD, CS-1, and FN-C/H-V) and in combination (RGD plus CS-1, RGD plus FN-C/H-V, and CS-1 plus FN-C/H-V) (Figure 4). After 96 h, we quantified apoptosis through flow cytometry after propidium iodide staining. Neither RGD, CS-1, or FN-C/H-V alone nor any two of the peptides in combination were able to reproduce the apoptotic effect of all three of the peptides used together (less than 20% versus 60%) (Figure 4).

View larger version (11K): [in this window] [in a new window]   Figure 4.   Individual effect of RGD, CS-1, and FN-C/H-V peptides on fibroblast viability. Human lung fibroblasts were incubated with RGD, CS-1, or FN-C/H-V peptides individually or with various combinations of the peptides for 96 h. Shown are the percentages of apoptotic cells for each peptide and their combination as determined with flow cytometry.

Soluble Fibronectin Peptides Induce Fibroblast Apoptosis in a Fibrin Gel

To begin to examine whether fibronectin peptides might be important regulators of fibroblast viability in fibrotic air space tissue observed after lung injury, we chose to study their effect in the well-characterized fibrin gel model (15, 16). We found that human lung fibroblasts incorporated into a fibrin matrix resembled loosely formed granulation tissue (Figure 5A, left): fibroblasts extended long, thin cytoplasmic processes and were interspersed throughout the fibrin gel. Fibroblasts incorporated into fibrin gels and cultured (96 h, 37° C) in the presence of soluble fibronectin peptides (500 µg/ml) underwent typical morphologic changes characteristic of apoptosis (Figure 5A, right). Analysis of fibroblast morphology after staining with H&E revealed plasma membrane blebbing and cytoplasm condensation. The percentage of fibroblasts undergoing apoptosis within the fibrin gels was analyzed microscopically after in situ labeling of DNA fragmentation with the TUNEL technique. Positive nuclei stained bright red-brown (Figure 5B, right) and could be easily distinguished from negative nuclei (Figure 5B, left). After treatment with soluble fibronectin peptides, 64 ± 2.8% of fibroblasts in the fibrin gel stained positively, indicating apoptosis in contrast to 17 ± 8.4% in control cultures, representing a 73% difference in the level of apoptosis (p < 0.05). There was no apparent diffusion effect or gradient of fibroblast apoptosis within the gels.

View larger version (93K): [in this window] [in a new window]   Figure 5.   Fibronectin peptides induce apoptosis of fibroblasts incorporated into a fibrin gel. (A) Left: Untreated human lung fibroblasts cultured for 4 d in fibrin gels morphologically resemble granulation tissue (H&E stain; magnification: ×200). Right: Appearance of human lung fibroblasts incubated in presence of RGD, CS-1, and FN-C/H-V peptides (96 h, 500 µg/ml) (H&E stain; original magnification: ×200). Note prominent pyknotic and fragmented nuclei characteristic of apoptosis. (B) TUNEL staining. Left: Untreated human lung fibroblasts in fibrin gel. Right: Fibroblasts cultured in the presence of RGD, CS-1, and FN-C/H-V peptides (original magnification: ×200). Pyknotic and fragmented nuclei are stained dark brown (arrows) and are easily distinguished from normal fibroblasts (shown in left panel ). Shown in the inset is an apoptotic cell containing marked chromatin condensation and fragmentation (original magnification: ×400).

As an additional corroborative assay for apoptosis, we analyzed human lung fibroblasts grown on tissue culture slides with the TUNEL assay. In accord with results of the TUNEL analysis of fibroblasts incorporated into fibrin gels and treated with fibronectin peptides, 68 ± 4% of fibroblasts grown on tissue culture slides showed apoptosis. Only 5 ± 3% of fibroblasts treated with scrambled fibronectin peptides underwent apoptosis as determined with the TUNEL assay (p < 0.05). It should be noted that the level of apoptosis as determined with the TUNEL assay was somewhat lower than the approximately 80% of fibroblasts undergoing apoptosis on tissue culture plates as determined through flow cytometry after propidium iodide staining. The TUNEL assay may underestimate the percentage of fibroblasts undergoing apoptosis, since ~ 12% of cells contained no identifiable nucleus, and only the cytoplasm remained. Such cells were not included as positive or apoptotic.

Lung Fibroblast Apoptosis Results from Disruption of Adhesion (Anoikis), with No-Additional Effect of Soluble Fibronectin Peptides on Induction of Apoptosis in Suspended Fibroblasts

The term anoikis has been used to describe a cell's apoptotic response to the absence of cell-matrix interactions. Our data showing the induction of fibroblast apoptosis by blocking of adhesion receptor function with soluble fibronectin peptides but not with scrambled peptides indicate that apoptosis results from disruption of cell-matrix interactions (Figures 2 and 3). Furthermore, phase contrast microscopy indicated that gradual cell rounding and detachment of adherent fibroblasts correlated with the onset of apoptosis (data not shown).

To confirm whether induction of fibroblast apoptosis is due solely to disruption of cell-matrix interactions (8), we studied the effect of the three soluble fibronectin peptides on detached fibroblasts. The percentage of fibroblasts undergoing apoptosis was analyzed when fibroblasts were plated on plastic dishes precoated with a nontoxic film of poly(HEMA) to prevent their attachment. After 32 h of incubation, 56 ± 15% of human lung fibroblasts were apoptotic (Figure 6). This is roughly comparable to the ~ 40% level of apoptosis seen with adherent fibroblasts at 48 h in kinetic experiments (Figure 3A). The presence of soluble fibronectin peptides did not have an additional effect on induction of fibroblast apoptosis (54 ± 13% apoptotic fibroblasts). These results indicate that the mechanism by which soluble fibronectin peptides induce fibroblast apoptosis is through disruption of cell adhesion to the underlying substratum.

View larger version (16K): [in this window] [in a new window]   Figure 6.   Soluble fibronectin peptides have no additional effect on inducing apoptosis on suspended fibroblasts. Untreated (control) (open squares) and soluble fibronectin peptide (sFNP; 500 µg/ml of RGD, CS-1 and FN-C/H-V)-treated (closed squares) human lung fibroblasts were cultured in DMEM + 0.1% FCS for 32 h on poly(HEMA)-coated plates. Cells were harvested at different time points, fixed with 70% ethanol, and stained with propidium iodide for flow cytometric analysis. Shown are the percentage of apoptotic cells as determined with flow cytometry.

pp125FAK is Proteolyzed during Soluble Fibronectin Peptide-Induced Apoptosis of Nontransformed Fibroblasts

pp125FAK plays a pivotal role in integrin and growth-factor signaling pathways, and controls different aspects of cell behavior, especially cell survival (17, 18). In this regard, recent data indicate that focal adhesion kinase (FAK) is an important mediator of integrin-dependent survival signals (19). This prompted us to examine whether fibronectin peptide-induced apoptosis was accompanied by changes in cellular pp125FAK. Data obtained by Crouch and colleagues show that proteolysis of pp125FAK occurs in adherent cells before their commitment to programmed cell death (18), indicating that FAK cleavage is an integral step that leads to cell death rather than a consequence of the apoptotic process. Lung fibroblasts cultured in DMEM + 0.1% FCS with scrambled (control) fibronectin peptides contained intact pp125FAK as determined by Western blot analysis. A 125-kD band was seen, demonstrating the presence of intact pp125FAK (Figure 7, lane 3). In contrast, Western blot analysis of cell lysates derived from lung fibroblasts incubated with fibronectin peptides (72 h, 500 µg/ml each peptide) revealed the presence of an ~ 110-kD degradation product of pp125FAK as well as a band of intact pp125FAK (Figure 7, lane 4). As a reference experiment for demonstrating FAK proteolysis, we used serum-starved, Myc-transformed fibroblasts undergoing apoptosis. Western blot analysis of cell lysates from these serum-starved, Myc transformed fibroblasts showed an ~100-kD proteolytic fragment of pp125FAK in addition to a band of intact pp125FAK (Figure 7, lane 2). Myc-transformed fibroblasts cultured in DMEM + 10% FCS were viable and contained largely intact FAK (Figure 7, lane 1). Different patterns of pp125FAK proteolysis have been described, depending on the antibody and the type of cells used (19). These variations probably explain the presence of the different proteolytic fragments in these two different cell lines. These data indicate that proteolysis of pp125FAK occurs during fibronectin peptide induction of fibroblast apoptosis, and support the hypothesis that fibronectin peptides trigger apoptosis by disrupting integrin-mediated survival signaling.

View larger version (27K): [in this window] [in a new window]   Figure 7.   pp125FAK proteolysis during induction of fibroblast apoptosis by soluble fibronectin peptides. Human lung fibroblasts were treated with RGD, CS-1, and FN-C/H-V peptides at 500 µg/ml concentration for 72 h. Cell lysates of the pooled cell population (adherent and floating cells) were examined for pp125FAK by Western blot analysis. Serum-deprived, Myc-transformed fibroblasts were used as a reference experiment for proteolytic processing of pp125FAK associated with fibroblast apoptosis. A proteolytic cleavage product of pp125FAK appeared after serum withdrawal, whereas intact 125 kD FAK was seen in nonapoptotic Myc-transformed fibroblasts cultured with 10% FCS. Sodium dodecylsulfate-polyacrylamide gel (8%) electrophoresis was used to separate 40 µg/lane of total protein, followed by transfer to nitrocellulose membranes. pp125FAK was detected with an anti-pp125FAK monoclonal antibody and enhanced chemiluminescence. Lane 1 (Myc 10%): Myc-transformed fibroblasts cultured in DMEM + 10% FCS; lane 2 (Myc 0%): Myc-transformed fibroblasts cultured in DMEM without FCS; lane 3 (Control 0.1%): human lung fibroblasts cultured in DMEM + 0.1% FCS with scrambled (control) peptides; lane 4 (FNP 0.1%): human lung fibroblasts cultured in DMEM + 0.1% FCS with soluble fibronectin peptides. Arrowheads show approximate molecular weight of the fragment in kilodaltons.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Therapy of fibroproliferative lung diseases has largely been relegated to blocking inflammation. However, its clinical benefit is limited despite excellent antiinflammatory efficacy of the agents used. This suggests that therapy targeting fibroblasts is essential for such diseases. At least three approaches can be considered. One approach is to block expansion of the fibroblast population by interfering with growth factors, their cognate receptors, or downstream signaling (23, 24). A second approach is to block synthesis of collagen (25). A third approach, which we propose, is the selective elimination of lung fibroblasts by induction of apoptosis because it emulates the physiologic process for eliminating excess fibrotic tissue during integumentary and visceral wound healing (1, 26).

In the present study we found that three soluble fibronectin peptides act as potent inducers of apoptosis of primary lung fibroblasts. Apoptosis was found to be dose and time dependent. Studies in a fibrin gel, a model of early granulation tissue, indicated a potent proapoptotic effect. Fibroblast apoptosis occurred by disruption of adhesion. Treatment of fibroblasts with fibronectin peptides caused proteolysis of pp125FAK, indicating that the mechanism of apoptosis involves disruption of integrin-mediated survival signaling.

RGD, CS-1, and FN-C/H-V, three soluble peptides from the fibronectin molecule, have been shown to block integrin-mediated-functions such as cell adhesion, migration (27), and regulation of the expression of genes encoding matrix metalloproteinases, proinflammatory and antiinflammatory cytokines (28), promoters of cell-cycle progression (29), and modulators of programmed cell death (30). Moreover, through their adhesion-blocking function, the role of these peptides in inducing apoptosis has been demonstrated for endothelial, epithelial, and certain transformed cell lines. In fact, RGDS peptide, by interfering with an integrin recognition site for the fibronectin molecule, decreases cell adherence and so increases apoptosis of bronchial epithelial cells (4). However, the RGD peptide alone was not sufficient to maintain bronchial epithelial cell survival in the absence of growth factors (31). In the same way, transformed rat embryonic fibroblasts incubated with a soluble RGD-containing integrin ligand undergo apoptosis caused by disruption of adhesion (32). Moreover, a YIGSR peptide from laminin induces apoptosis of fibrosarcoma cells (32). Our report is the first to describe an apoptotic effect of the RGD, CS-1, and FN-C/H-V fibronectin peptides on adherent nontransformed fibroblasts, the primary target for antifibrotic therapies.

One difference between induction of apoptosis of fibroblasts versus that of epithelial and endothelial cells with fibronectin peptides is that fibroblasts appear to be more difficult to kill. Induction of apoptosis of epithelial and endothelial cells using fibronectin peptides occurs over a 24-h period (2, 3). In contrast, a 3- to 4-d incubation of adherent fibroblasts with peptides was required in order to achieve a high level of apoptosis. The implications of these data for the potential use of fibronectin peptides as antifibrotic therapy are obvious. Because both epithelial and endothelial cells readily undergo apoptosis (in vitro) in response to fibronectin peptides, targeting delivery of the peptides to air-space fibroblasts would probably be required.

In order to determine whether apoptosis was due to a proapoptotic signal arising from fibronectin peptide binding to receptor, or resulted from the disruption of fibroblast adhesion to the substratum, we investigated the effect of the RGD, CS-1, and FN-C/H-V peptides on nonadherent fibroblasts. Nonadherent human lung fibroblasts (cells in suspension) underwent apoptosis as previously shown (16). However, the soluble fibronectin peptides did not have any additional apoptotic effect on these cells, demonstrating that induction of fibroblast apoptosis by soluble fibronectin peptides was a result of disruption of fibroblast adhesion (anoikis) (33).

The concept of cell-surface-receptor-mediated regulation of cell survival has now been firmly established (2, 4). Integrins are a large family of heterodimeric cell surface receptors involved in cell-cell and cell-matrix interactions (34). RGD, CS-1 and FN-C/H-V are ligands of some integrin receptors whose role has been demonstrated in cell survival signaling (11, 30, 35). To begin to investigate whether induction of fibroblast apoptosis by fibronectin peptides operates by disruption of an integrin-mediated survival pathway, we studied alterations in pp125FAK integrity. pp125 FAK is a tyrosine kinase that belongs to the focal adhesion complex (17). The focal adhesion complex integrates integrin-mediated signal derived from the matrix. As part of the focal adhesion complex, pp125FAK generates signaling biochemical cascades controlling different aspects of cell behavior, such as the integrin-mediated survival signaling pathway (18). Importantly, proteolysis of pp125FAK has been shown to occur before commitment of serum-starved, Myc-transformed fibroblasts to apoptosis, suggesting that FAK proteolysis is a critical step that leads to cell death (19). Furthermore, recent work indicates that caspase-mediated FAK cleavage results in disassembly of focal adhesion complexes, thereby actively disrupting integrin-mediated survival signaling from the extracellular matrix (22). We show that the activity of soluble fibronectin peptides causes FAK proteolysis. This finding supports the hypothesis that soluble fibronectin peptides trigger lung fibroblast apoptosis by disrupting an integrin-mediated survival signaling pathway.

In summary, we have shown that soluble fibronectin peptides can disrupt the adhesion of nontransformed human lung fibroblasts and trigger apoptosis via interference with an integrin-mediated survival pathway. Since conventional therapy for many fibroproliferative lung disorders is ineffective, novel therapeutic strategies need to be identified. One approach is to selectively target the elimination of lung fibroblasts, which for the first time would focus therapy on the principal effector of organ dysfunction. The elimination of air-space fibrotic tissue with small, soluble fibronectin peptides may be an efficacious alternative approach to treating pulmonary fibrosis, provided selective targeting of intraalveolar fibroblasts becomes possible.