Cigarette smoke extract inhibits fibroblast-mediated ... - CiteSeerX

7 downloads 0 Views 105KB Size Report
Briefly, two cigarettes without filters were combusted with a modified syringe-driven apparatus. The smoke was bubbled through 50 ml of serum-free DMEM.
Cigarette smoke extract inhibits fibroblast-mediated collagen gel contraction S. CARNEVALI,1 Y. NAKAMURA,2 T. MIO,3 X. LIU,4 K. TAKIGAWA,4 D. J. ROMBERGER,4 J. R. SPURZEM,4 AND S. I. RENNARD4 1Dipartimento di Cardiologia, Angiologia e Pneumologia, U. O. di Pneumologia e Fisiopatologia Respiratoria, Universita` degli Studi di Pisa, 56214 Pisa, Italy; 2Third Department of Internal Medicine, School of Medicine, University of Tokushima, Tokushima 770; Pulmonary Medicine, Chest Disease Research Institute, Kyoto University, Kyoto 601, Japan; and Pulmonary and Critical Care Medicine, University of Nebraska Medical Center, Omaha, Nebraska 68198-5300 Carnevali, S., Y. Nakamura, T. Mio, X. Liu, K. Takigawa, D. J. Romberger, J. R. Spurzem, and S. I. Rennard. Cigarette smoke extract inhibits fibroblast-mediated collagen gel contraction. Am. J. Physiol. 274 (Lung Cell. Mol. Physiol. 18): L591–L598, 1998.—Cigarette smoking, the major cause of pulmonary emphysema, is characterized by destruction of alveolar walls. Because tissue destruction represents a balance between injury and repair, we hypothesized that cigarette smoke exposure may contribute to the development of emphysema through the inhibition of tissue contraction during the repair process. To partially evaluate this hypothesis, we investigated the effects of cigarette smoke extract (CSE) on the ability of cultured fibroblasts to mediate collagen gel contraction in vitro: CSE inhibited fibroblastmediated gel contraction in a concentration-dependent manner (P , 0.01). Production of prostaglandin E2, a known inhibitor of fibroblast contraction, was unchanged by CSE as was cell surface integrin expression. In contrast, fibronectin production by fibroblasts was inhibited (P , 0.01), and addition of exogenous fibronectin partially restored the contractile activity, thus suggesting at least one mechanism to explain inhibition of gel contraction by CSE. When CSE was treated to remove volatile components, it showed less inhibitory activity on fibroblast-mediated gel contraction. Therefore, we also examined the effects of acrolein and acetaldehyde, two volatile components of cigarette smoke. Inhibition of contraction was observed at 5 µM acrolein and at 0.5 mM acetaldehyde. In conclusion, cigarette smoke inhibited fibroblast-mediated gel contraction, and this inhibition was due, at least in part, to the volatile components of cigarette smoke and may be mediated, at least in part, by a decrease in fibroblast fibronectin production. By inhibition of repair, these smoke components may contribute to the development of pulmonary emphysema.

blasts, which are recruited to sites of injury throughout the body. These cells can proliferate at sites of injury, produce matrix proteins, and subsequently remodel the newly deposited matrix through a variety of processes including that of contraction (8). Previous studies have suggested that cigarette smoke can impair the wound healing process by inhibiting fibroblast recruitment and proliferation (26). An effect on inhibition of fibroblast-mediated contraction could contribute to the enlarged air spaces that develop in the injuries associated with pulmonary emphysema. The current study, therefore, was designed to determine whether cigarette smoke impairs fibroblast-mediated wound contraction. To accomplish this, a model system using fibroblasts cultured in a three-dimensional (3-D) native collagen gel was utilized (3). In this model, fibroblasts are cultured in a 3-D matrix consisting of native type I collagen fibers. This matrix culture is thought to more closely resemble in vivo conditions than a monolayer culture. Consistent with this, fibroblasts cultured in the 3-D gel have altered phenotype and synthetic capability (3, 21). Among the properties of fibroblasts in a 3-D gel is the ability to cause gel contraction, a phenomenon that resembles the contraction of newly formed scar tissue (3). Our results suggest that cigarette smoke extract (CSE) can impair this process and thus suggest another mechanism by which cigarette smoke could contribute to the development of pulmonary emphysema.

fibronectin; three-dimensional gel

Cell culture. Human fetal lung fibroblasts (HFL1; lung, diploid, human) were obtained from the American Type Culture Collection (Rockville, MD). The cells were cultured in DMEM (GIBCO ) with 10% FCS and refed three times weekly in 100-mm tissue culture dishes (Becton Dickinson Labware, Lincoln Park, NJ). Cells were trypsinized (Trypsin-EDTA, GIBCO; 0.05% trypsin-0.53 mM EDTA-4Na) for use in contraction assays when ‘‘subconfluent,’’ i.e., ,3 3 106 cells/dish. Collagen. Collagen gels were prepared as described by Strom and Michalopoulos (33). Type I collagen was extracted by stirring ethanol-washed adult rat tail tendons for 48 h at 4°C in sterile 4 mM acetic acid. After centrifugation (3,000 rpm for 20 min at 4°C), the supernatant was stored. An aliquot was lyophilized to determine collagen concentration. CSE. CSE was prepared by a modification of the method of Carp and Janoff (7). Briefly, two cigarettes without filters were combusted with a modified syringe-driven apparatus. The smoke was bubbled through 50 ml of serum-free DMEM. The resulting suspension was adjusted to pH 7.4 with concen-

CIGARETTE SMOKING IS THE MAJOR cause of pulmonary emphysema (5). Current concepts suggest that emphysema results from the excessive release of destructive proteases from inflammatory cells that overcome the anti-protease defenses of the lung (20). Excessive tissue damage, however, represents only part of the process in the development of emphysema. In this context, the lung has considerable capacity to mediate repair responses. Net tissue destruction, the defining characteristic of emphysema, must therefore represent tissue damage with inadequate repair responses. Tissue repair after damage is a complex process. Of particular importance appears to be the role of fibro-

MATERIALS AND METHODS

1040-0605/98 $5.00 Copyright r 1998 the American Physiological Society

L591

L592

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

trated NaOH and then filtered through a 0.20-µm pore filter (Lida Manufacturing, Kenosha, WI) to remove bacteria and large particles. CSE was applied to fibroblast cultures within 30 min of preparation. To examine the effect of volatilization, CSE was lyophilized and reconstituted to the initial volume with distilled water. A separate aliquot was bubbled with a stream of N2 gas at room temperature for 30 min. To determine the concentration of acetaldehyde and acrolein in the smoke extracts, four separate extracts were prepared as described above, sealed in screw-cap tubes, frozen on dry ice, and assayed within 48 h. Acetaldehyde and acrolein concentrations were determined by liquid chromatography with fluorescence detection after reaction with 2diphenylacetyl-1,3-indanedione-1-hydrazone (Shawnee Chemical, Springfield, OH) to form azine derivatives (4). Derivatization solution was prepared by adding 0.13 g of the hydrazone reagent to 100 ml of acetonitrile (B&J High-Purity Solvent; Baxter Healthcare, McGaw Park, IL). Catalyst (250 µl of 5 N HCl) and activated 3A molecular sieve (Advanced Specialty Gas Equipment, South Plainfield, NJ) were added to the solution just before use. Immediately after frozen smoke extracts were thawed, 2- and 5-ml aliquots of each smoke extract were added to 100-ml volumes of derivatization solution, and the azine formation reaction was allowed to proceed for 1 h. Each sample was filtered with a 0.45-µm polytetrafluoroethylene membrane Autovial disposable filter assembly (Whatman, Clifton, NJ). Reversed-phase liquid chromatography was based on a Waters NOVA-PAK C18 column and an acetonitrile-water mobile phase. External standard quantitation was performed. Recovery of both compounds was assessed by spiking known concentrations of acetaldehyde and acrolein (Sigma, St. Louis, MO) into samples and handling them in an identical fashion. Contraction assay. Effects of CSE on fibroblast-mediated collagen gel contraction were examined by the method of Bell et al. (3). Briefly, the fibroblasts prepared as described in Cell culture were trypsinized and resuspended in DMEM-1% FCS (Biofluids, Rockville, MD). The cells were gently mixed with a solution containing rat tail tendon collagen, 43 concentrated DMEM, FCS, and distilled water, all prepared at 4°C. Volumes were adjusted so that the final collagen concentration was 0.75 mg/ml, DMEM was 13, FCS was 1%, and cell concentration was 105 cells/ml. Fibroblasts were added after other components had been mixed. Three milliliters of cell suspension containing 3 3 105 cells were dispensed into six-well tissue culture plates, with each well being 35 mm in diameter (Becton Dickinson Labware). The plates were then incubated for 20 min at 37°C to allow the collagen solution to gel. After this, 3 ml of DMEM with 1% FCS containing different concentrations of CSE were added on top of the gel. The gels were then gently detached from the walls and the bottom of the dishes using a sterile metal spatula to prepare a floating gel. The resulting cultures were incubated at 37°C and 5% CO2 and were refed every day with fresh medium containing CSE. Gel contraction was quantified by measuring the area of the gel after 24, 48, and 72 h using an imageanalysis system (Optomax V, Burlington, MA) to quantify an image of the plate prepared on a Xerox copier and expressed as a percentage of the original area. All cultures were performed in triplicate. Cytotoxicity. To assess the cytotoxicity of CSE for human lung fibroblasts, lactate dehydrogenase (LDH) release into gels and into supernatant was measured using a commercially available kit (LDH-20; Sigma). This method is able to detect LDH release from cells incubated with cytotoxic levels of acrolein and acetaldehyde (26).

DNA assay. To estimate cell number in 3-D gels, DNA was assayed fluorometrically with Hoechst dye no. 33258 (Sigma) by a modification of a previously published method (17). Collagen gels were solubilized with collagenase (22), and cell suspensions were collected by centrifugation at 500 g for 10 min and resuspended in 1 ml of distilled water. After freezing and thawing twice, the suspensions were mixed with 2 ml of TNE buffer (3 M NaCl, 10 mM Tris, and 1.5 mM EDTA, pH 7.4) containing 2 µg/ml of Hoechst no. 33258. Fluorescence intensities were measured with a fluorescence spectrometer (LS-5, Perkin-Elmer) with excitation at 356 nm and emission at 458 nm. Prostaglandin production. Prostaglandin E2 (PGE2 ) produced by fibroblasts is known to inhibit collagen gel contraction and could function as an autocrine or paracrine mediator. Therefore, we quantified PGE2 in both supernatant media and solubilized gels (22) by RIA (9, 27) (Advantage Magnetics, Cambridge, MA). Inhibition of cyclooxygenase. To further explore the role of cyclooxygenase and prostaglandin activity in regulating gel contraction, monolayer cell cultures were incubated with 1 µM indomethacin for 45 min before exposure to different concentrations of CSE. After pretreatment, cells were trypsinized and cast into gels that also contained 1 µM indomethacin. To determine whether indomethacin treatment could directly alter fibroblast-mediated gel retraction, control experiments were performed in which indomethacin was added to the cell suspension and mixed with other components. Integrin expression. Cell surface integrins, in particular a2b1-integrin, are thought to be required for fibroblastmediated collagen gel contraction (16, 30). To evaluate cell surface integrin expression, subconfluent fibroblast monolayer cultures were incubated overnight in serum-free DMEM containing different concentrations of CSE. Cell surface expression of integrins was evaluated using epifluorescence flow cytometry. Briefly, cells were suspended in PBS containing 3% BSA and incubated with control mouse ascites or mouse monoclonal anti-integrin antibodies (1:200 dilution), followed by counterstaining with FITC-conjugated antimouse IgG (1:200 dilution, Sigma). The stained cells were fixed in 1% paraformaldehyde, and the relative fluorescence intensity per cell was measured with fluorescence flow cytometry (FACS II, Becton Dickinson, Sunnyvale, CA). Data are expressed as the mean channel fluorescence intensity of 5,000 cells. To compare fluorescence intensities, the mean channel fluorescence intensity for 5,000 cells was converted to linear fluorescence intensity units (25) so that background fluorescence with control antibodies could be subtracted (32). Measurement of fibronectin by ELISA. For quantification of fibronectin production, the media were harvested and the gels were solubilized with collagenase as described in DNA assay. Fibronectin in the gels and the media was assayed by an ELISA that is specific for human fibronectin and does not detect bovine fibronectin (29). Effect of plasma and cellular fibronectin. To evaluate modulation of gel contraction by fibronectin, in separate experiments, 50 µg/ml of human plasma fibronectin (GIBCO BRL, Grand Island, NY) or human cellular fibronectin (Upstate Biotechnology, Lake Placid, NY) were added to the collagen solution before gels were made. Gels were then cast and cultured, and the areas of the gels were measured as described in Contraction assay. Statistical evaluation. The data presented are means 6 SE. Experimental values were compared using a one-way ANOVA for repeated measures and an unpaired two-tailed Student’s t-test for single comparisons. Comparisons were considered statistically significant at P , 0.05.

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

L593

RESULTS

The effect of CSE on fibroblast-mediated gel contraction. Control gels consisting of native type I collagen in which fibroblasts were embedded contracted rapidly during the first 24 h of incubation, reaching 35.8 6 3.4% of their original area, and further decreased their area during the next 72 h (Fig. 1). CSE-exposed gels also contracted but significantly less so. After 24 h of incubation, the area of the gels incubated in media containing 4% CSE was still 57.9 6 1.5% of the original area, and gels incubated in media containing 6% CSE were 90 6 2.0% of the original area (P , 0.01, both comparisons with control by ANOVA). In contrast, the area of the gels incubated with 2% CSE did not differ from control. When media were changed every 24 h to media containing fresh smoke extract, the differences between control and 4 or 6% CSE persisted for up to 72 h (Fig. 1). Because the fibroblasts caused relatively rapid contraction soon after gel release, the effects of adding smoke extract to gels that had been allowed to contract for 24 h were studied. Addition of smoke extract after 24 h retarded the contraction over the subsequent 48 h (Fig. 2; P , 0.05, ANOVA). Conversely, change of the culture medium to smoke-free medium after 24 h partially restored the contractile capacity of the fibroblasts (Fig. 2; P , 0.05, ANOVA). Cytotoxicity of CSE. Cytotoxicity of CSE was assessed by quantification of LDH release into gels and supernatants. No release of LDH was induced at any concentration used (Fig. 3). In contrast, the supernatant of fibroblasts incubated with 1% sodium azide, used as positive control, showed a significant increase in LDH release. As a separate means of evaluating fibroblast viability, fibroblast number was estimated by DNA quantification. No significant differences were observed over 72 h after exposure to CSE (data not shown). PGE2. Fibroblasts are known to produce PGE2 (19), and PGE2 has been reported to inhibit fibroblast-

Fig. 1. Cigarette smoke extract (CSE) inhibition of fibroblastmediated gel contraction. Fibroblasts were cultured in collagen gels and incubated with DMEM-1% FCS containing 0 (s), 2 (r), 4 (k), and 6% (j) CSE, and gel surface areas were measured after 24, 48, and 72 h. Significant inhibition of gel contraction was observed in gels exposed to 4 and 6% at all time points (P , 0.01 compared with control).

Fig. 2. Effects of addition and removal of CSE 24 h after culture of fibroblasts in collagen gels. Fibroblasts were cultured in collagen gels with 1% FCS and incubated with medium containing 0 or 6% CSE for 24 h, after which gel areas were determined. Media were then aspirated, and gels were rinsed once and refed with fresh medium with or without CSE. Contraction was then assessed after 24 and 48 h of further incubation. Squares, gels under control condition for 1st 24 h; circles, CSE-exposed gels for 1st 24 h; open symbols, gels with control medium; solid symbols, gels with CSE.

mediated collagen gel contraction (27). Augmented PGE2 production in response to CSE could be one mechanism to explain reduced gel retraction after exposure to CSE. To evaluate this possibility, the supernatant media from collagen gel cultures were harvested, and the gels were dissolved with collagenase. When the gels were completely dissolved, the resulting solutions were centrifuged to remove cells, and the supernatants were harvested. Release of PGE2 by fibroblasts into the cultures was then quantified in both gels and supernatants. CSE induced a slight but not statistically significant increase in PGE2 release (Fig. 4; P . 0.3, ANOVA). To further explore the role of endogenously produced prostaglandins in modulating

Fig. 3. Cytotoxicity of CSE examined by measurement of lactate dehydrogenase (LDH) release. Three days after exposure to CSE, supernatant was harvested and gels were solubilized (see MATERIALS AND METHODS ). No significant increase in LDH was observed in either supernatants (solid bars) or gels (open bars) for all conditions used. Supernatant from fibroblasts incubated with 1% azide was used as positive control.

L594

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

Fig. 4. Effect of CSE on prostaglandin E2 (PGE2 ) release. Fibroblasts were trypsinized, cast into gels, and incubated with DMEM containing varying concentrations of CSE. After 3 days of exposure, supernatant media were harvested, and gels were solubilized with collagenase. PGE2 was then measured by RIA in both gel (solid bars) and supernatant medium (open bars) for each culture. Amount of PGE2 did not significantly increase in any condition assayed.

gel contraction, fibroblasts were pretreated with or without indomethacin, an inhibitor of prostaglandin synthesis. Indomethacin was able to inhibit PGE2 production by fibroblasts (data not shown) but did not block the ability of CSE to inhibit fibroblast-mediated collagen gel contraction (Fig. 5). a2b1-Integrin expression. Fibroblast contraction of collagen gels is thought to be mediated by cell surface integrins, specifically a2b1-integrin. Inhibition of integrin expression, therefore, could be a possible mechanism to explain CSE-mediated inhibition of fibroblast contraction of collagen gels. All fibroblasts, however, were well stained with antibody to both a2- and b1-

Fig. 5. Effect of indomethacin on CSE inhibition of fibroblastmediated gel retraction. Fibroblast monolayer cultures were pretreated for 45 min with or without indomethacin, trypsinized, and cast into collagen gels (see MATERIALS AND METHODS ). Gels were incubated with medium containing different concentrations of CSE with or without indomethacin. Contraction was significantly reduced by CSE (open bars). Indomethacin treatment had a minimal effect on CSE inhibition of contraction (solid bars).

integrin. CSE exposure did not alter integrin expression assessed as mean fluorescence (data not shown). Fibronectin. Potential mechanisms to explain CSEmediated inhibition of fibroblast contraction of collagen gels also include inhibition of fibronectin production. CSE exposure resulted in a significant decrease in fibronectin production by fibroblasts in gel culture at all conditions used (Fig. 6; P , 0.01). Thus a decrease in fibronectin production may contribute to CSE inhibition of contraction. To further evaluate this possibility, the effect of exogenously added plasma and cellular fibronectin was also investigated. Fibroblasts were cast into gels with or without cellular or plasma fibronectin (50 µg/ml) and then exposed to 0 or 6% CSE. Both cellular fibronectin and plasma fibronectin were able to partially restore the ability of fibroblasts to mediate contraction after exposure to CSE (P , 0.01; Fig. 7). Inhibition of fibroblast-mediated gel contraction and fibronectin production by volatile components of cigarette smoke. To help determine which components of CSE contributed to the inhibition of fibroblast-mediated collagen gel contraction, we assessed the effect of CSE after lyophilization and after ‘‘volatilization’’ to remove volatile components. CSE lost most of the inhibitory activity for fibroblast-mediated collagen gel contraction after removal of volatile components (Fig. 8). We next assessed the effects of two volatile components present in high concentration in cigarette smoke, acrolein and acetaldehyde. Both volatile components of CSE inhibited fibroblast-mediated gel contraction. This inhibition was observed at 1026 M acrolein and at 5 3 1024 M acetaldehyde (P , 0.01; Fig. 9, A and B). Acrolein and acetaldehyde were not cytotoxic at these concentrations, as assessed by LDH release (data not shown), but both significantly reduced the amount of fibronectin recovered from fibroblast-containing gels (Fig. 10). To determine whether these concentrations of volatile components were present in our smoke extracts, we

Fig. 6. Effect of CSE on fibronectin release in gel culture. Three days after exposure to CSE, collagen gels were solubilized and centrifuged. Supernatant media were collected, and fibronectin was quantified by ELISA. Fibronectin release was significantly reduced in gels exposed to CSE compared with nonexposed gels (P , 0.01, ANOVA).

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

Fig. 7. Partial reversal of CSE inhibition of fibroblast-mediated gel retraction by plasma and cellular fibronectin. Fibroblasts were cast into collagen gels without or with either plasma or cellular fibronectin (50 µg/ml) as indicated and exposed to 0 or 6% CSE. After 3 days, size of gels was measured. Plasma and cellular fibronectin were capable of reversing, in part, the inhibitory effect of CSE.

measured the concentrations of both acetaldehyde and acrolein in replica extracts. Recovery of known standard concentrations of acetaldehyde was 70–80%. Despite only partial recovery, acetaldehyde was present in all extracts at concentrations measured to be 4.5 6 1.2 3 1024 M. Recovery of acrolein was much less efficient, being 10% or less in standards, and no acrolein was detectable in smoke extracts. DISCUSSION

The current study demonstrates that cigarette smoke is capable of inhibiting fibroblast-mediated collagen gel contraction. This effect depends on the concentration of smoke extract to which collagen gels are exposed. The effect develops over 24 h and persists for up to 72 h with continual exposure. The effect is, however, at least

Fig. 8. Effect of volatilization and lyophilization on inhibitory activity of CSE for fibroblast-mediated collagen gel contraction. Fibroblasts were cultured (see MATERIALS AND METHODS ) in collagen gels with 1% FCS and incubated with DMEM containing 0 (r) or 6% (s) CSE, volatilized CSE (k), or lyophilized CSE (j). Gel surface areas were measured at various periods of time in culture. CSE lost most of its inhibitory activity for fibroblast-mediated gel contraction with removal of volatile components.

L595

partially reversible. Removal of smoke after 24 h of exposure results in cells resuming their contractile activity. This reversibility suggests that the effect of smoke is not due to a cytotoxic effect, and, consistent with this, cigarette smoke was not associated with LDH release or with any significant change in cell number in the collagen gels as assessed by DNA content. Several mechanisms could explain the inhibitory effect of CSE on fibroblast-mediated collagen gel contraction (1, 10, 24). An increase in endogenous PGE2 production does not appear to account for the majority of the inhibitory effect, however, because cigarette smoke was not associated with significantly augmented PGE2 production nor was indomethacin associated with a blockade of the smoke effect. Similarly, loss of the collagen-binding a2b1-integrin would be expected to be associated with a loss of contractile activity (16, 30), but no changes in fibroblast expression of these integrin chains were observed when monolayer cultures were exposed to CSE. Fibronectin has also been reported to augment fibroblast-mediated gel contraction (2, 11). CSE was observed to inhibit release of fibronectin, thus providing a potential mechanism to account for the effect of CSE on contraction. In support of this mechanism, exogenous fibronectin could partially restore the ability of fibroblasts to support contraction of collagen gels. The inhibitory activity present in cigarette smoke appears to be a volatile component. Moreover, multiple components present in the volatile phase of cigarette smoke are likely to have activity because two active volatile components, acrolein and acetaldehyde, each demonstrated the ability to inhibit both fibroblastmediated collagen gel contraction and fibronectin production. Together, the current study supports the concept that cigarette smoke may impair wound healing processes by altering fibroblast contraction. Contraction of collagen gels by fibroblasts is a complex process, and cigarette smoke could be acting at several levels. This process appears to involve both the collagen-binding a2b1-integrin and fibronectin-mediated processes (2, 11, 30). The current study suggests that no significant changes in a2b1-integrin expression resulted from cigarette smoke exposure. On the other hand, integrin affinity for substrate could be altered, as could the mechanisms by which integrins interact with the cytoskeleton (15). Such an integrin-mediated mechanism is not excluded by the results of the present study. In the present study, increased endogenous PGE2 production does not appear to account for the inhibition of fibroblast-mediated collagen gel contraction caused by cigarette smoke. This is interesting, particularly when compared with the effects of gamma irradiation. Such radiation can also impair fibroblast-mediated collagen gel contraction in a reversible manner (6). The radiation effect, however, appears to be largely mediated through the production of endogenous PGE2, as evidenced by measurement of PGE2 production and reversibility of the effect in the presence of indomethacin. In contrast, cigarette smoke did not increase PGE2 release measured after 72 h. An effect on early PGE2

L596

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

Fig. 9. Inhibition of fibroblast-mediated collagen gel contraction was caused by 2 volatile components of cigarette smoke, acrolein or acetaldehyde, and their cytotoxicity for fibroblasts. To examine collagen gel contraction, fibroblasts were cultured in collagen gels and incubated with medium containing various concentrations of acrolein or acetaldehyde. Gel surface areas were measured after 24 h. A: acetaldehyde. B: acrolein. Inhibition of contraction was observed at 10 µM acrolein and at 1 mM acetaldehyde (P , 0.001 compared with control).

release is not excluded but seems unlikely because indomethacin added both before and together with smoke extract to ensure complete inhibition of PGE synthesis did not reverse the effect of smoke. Thus cigarette smoke appears to be inhibiting fibroblastmediated gel contraction by a mechanism different from that of irradiation. Some investigators have suggested that fibronectin plays a role in fibroblast-mediated gel contraction (2, 11). Our studies demonstrated that fibronectin production was significantly reduced by CSE as well as by acetaldehyde and acrolein. Reduced contraction due to reduction of fibronectin could account for the reduced contraction of cells exposed to CSE. This is supported by the observation that plasma and cellular fibronectin were capable of reversing, in part, the inhibitory effect of CSE. Similar results were observed with acetaldehyde and acrolein (data not shown). These results also

Fig. 10. Effect of acetaldehyde and acrolein on fibronectin release in gel culture. Three days after exposure to 1 mM acetaldehyde or 5 µM acrolein, gels were solubilized and centrifuged. Supernatant media were collected, and fibronectin was quantified by an ELISA. Fibronectin was significantly reduced by both acetaldehyde and acrolein (P , 0.01 compared with control).

suggest that the effect of CSE was not due to a nonspecific toxic effect. Variant forms of fibronectin result from differential splicing of mRNA derived from a single gene (31). Using skin fibroblasts, Asaga et al. (2) suggested that only the ‘‘cellular’’ form was capable of supporting fibroblastmediated gel contraction. In contrast, also working with skin fibroblasts, Gillery et al. (11) suggested that both forms could support this activity. The results from the current study suggest that lung fibroblasts can utilize either form of fibronectin to augment contraction that has been inhibited by CSE. The current study demonstrates that the volatile components of cigarette smoke were particularly important in inhibiting fibroblast-mediated gel contraction. Cigarette smoke contains in excess of 6,000 components, approximately one-half of which are relatively volatile (14). Only a minority of these components have been studied in detail for their toxicities. Among the more toxic species present in high concentration, however, are reactive aldehydes, of which acrolein and acetaldehyde are particularly prominent (13, 14). These moieties are capable of binding to and interacting with various cellular components including contractile proteins (23). Such a mechanism could contribute to the inhibition of contraction by smoke. It was for this reason that these specific species were selected for testing in the current study. It has been estimated that one cigarette can yield 980 µg of acetaldehyde and 85 µg of acrolein. Measured concentrations of acetaldehyde in our smoke extract were ,50% of this theoretical yield. Such a concentration, when diluted according to the protocols used in the current study, would have been just below the concentration-response range tested for acetaldehyde alone. With allowance for the 70–80% recovery we observed for acetaldehyde in standard samples, it is possible that acetaldehyde acting alone could have accounted for some of the observed effect. Recovery of acrolein was much less efficient (10% or less), perhaps because of its

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

greater lability, and the concentration of acrolein was below detectability in all our extracts. However, had acrolein been present in the ratio expected to acetaldehyde in fresh smoke extract, it would have been present at 1.77 µg/ml (2.16 3 1025 M), well within the range at which toxic effects on fibroblast contraction were observed. In addition to acetaldehyde and acrolein, cigarette smoke contains many other components that may also contribute to its toxicity. Because these components of cigarette smoke may interact with multiple targets within the cell, there is the potential that these various toxins could interact in an additive or even synergistic way to impair fibroblast function. Thus, although we believe that we have demonstrated that acetaldehyde and acrolein could contribute to the toxic effects of CSE, we believe that the effects of the extract likely represent a combined effect of these and other toxins. The ability of fibroblasts to mediate contraction of a collagen gel is thought to be a model of part of the wound-repair response (12, 28). This response is also thought to involve fibroblast recruitment and proliferation, both of which are also inhibited by CSE (26). These in vitro findings are consistent with the impaired healing noted in cigarette smokers in a variety of settings. One example of this impaired healing is pulmonary emphysema. In this disease, cigarette smoke results in destruction of alveolar walls and enlargement of air spaces (5, 20). Undoubtedly, the release of potent proteases in excess of anti-protease defenses plays an important role in the damage of the alveolar structures. Emphysema results when this damage is not balanced by an appropriate repair response. The ability of cigarette smoke to impair fibroblast-mediated repair mechanisms could, therefore, be another mechanism by which cigarette smoke contributes to the development of diseases such as pulmonary emphysema. In this regard, it is of interest that emphysema develops variably among smokers (5). Fibrosis-like changes, moreover, may be present at selected sites within emphysematous lungs (18, 34), suggesting that repair processes are active in emphysema and that the final lesion represents a balance among several processes. It is conceivable, therefore, that individuals differ in their susceptibility to cigarette smoke alteration of repair responses, and this could represent an important variable in determining who is at risk for the development of emphysema. In conclusion, we have shown that CSE can inhibit fibroblast-mediated collagen gel contraction. This effect is not likely to be due simply to an effect on cell growth or to cytotoxicity. The effect appears to be dependent on volatile components of cigarette smoke and may be mediated by a decrease in fibroblast production of fibronectin. Impaired wound healing is a feature of cigarette smokers and may be a particularly important feature in the development of some diseases such as pulmonary emphysema. The ability of cigarette smoke to inhibit fibroblast-mediated healing may play a particular role in this process.

L597

We thank Dr. Michael Borgerding for assistance in quantification of acetaldehyde and acrolein and Lillian Richards for administrative assistance. This work was supported in part by the National Research Council of Italy, the State of Nebraska, and the Larson Endowment of the Univ. of Nebraska Medical Center. Address for reprint requests: S. I. Rennard, Univ. of Nebraska Medical Center, 600 South 42nd St., Omaha, NE 68198-5300. Received 31 July 1996; accepted in final form 8 January 1998. REFERENCES 1. Anderson, S. N., Z. Ruben, and G. C. Fuller. Cell-mediated contraction of collagen lattices in serum-free medium: effect of serum and nonserum factors. In Vitro Cell. Dev. Biol. 26: 61–66, 1990. 2. Asaga, H., S. Kikuchi, and K. Yoshizato. Collagen gel contraction by fibroblasts requires cellular fibronectin but not plasma fibronectin. Exp. Cell Res. 193: 167–174, 1991. 3. Bell, E., B. Ivarsson, and C. Merrill. Production of a tissuelike structure by contraction of collagen lattices by human fibroblasts of different proliferative potential in vitro. Proc. Natl. Acad. Sci. USA 76: 1274–1278, 1979. 4. Borgerding, M. F., R. S. Dunn, F. A. Thome, H. L. Chung, D. S. Moore, T. R. Conner, D. L. Heavner, and P. H. Ayres. An improved method for the determination of carbonyl compounds in smoke. Tobacco Chemists’ Research Conference 38th, Atlanta, GA 1984, Paper No. 50. 5. Buist, A. S., and W. M. Vollmer. Smoking and other risk factors. In: Textbook of Respiratory Medicine, edited by J. F. Murray and J. A. Nadel. Philadelphia, PA: W. B. Saunders, 1994, p. 1259–1287. 6. Carnevali, S. Gamma radiation inhibits fibroblast-mediated collagen gel retraction (Abstract). Am. J. Respir. Crit. Care Med. 151: A562, 1995. 7. Carp, H., and A. Janoff. Possible mechanisms of emphysema in smokers. In vitro suppression of serum elastase-inhibitory capacity by fresh cigarette smoke and its prevention by antioxidants. Am. Rev. Respir. Dis. 118: 617–621, 1978. 8. Clark, R. A. F. Overview and general considerations of wound repair. In: The Molecular and Cellular Biology of Wound Repair, edited by R. A. F. Clark and P. M. Henson. New York: Plenum, 1988, p. 3–33. 9. Ehrlich, H. P., and D. J. Wyler. Fibroblast contraction of collagen lattices in vitro: inhibition by chronic inflammatory cell mediators. J. Cell. Physiol. 116: 345–351, 1983. 10. Fukamizu, H., and F. Grinnell. Spatial organization of extracellular matrix and fibroblast activity: effects of serum, transforming growth factor b and fibronectin. Exp. Cell Res. 190: 276–282, 1990. 11. Gillery, P., F. X. Maquart, and J. P. Borel. Fibronectin dependence of the contraction of collagen lattices by human skin fibroblasts. Exp. Cell Res. 167: 29–37, 1986. 12. Grinnell, F. Fibroblasts, myofibroblasts and wound contraction. J. Cell Biol. 124: 401–404, 1994. 13. Houlgate, P. R., K. S. Dhingra, S. J. Nash, and W. H. Evans. Determination of formaldehyde and acetaldehyde in mainstream cigarette smoke by high-performance liquid chromatography. Analyst 114: 355–360, 1989. 14. Huber, G. L., M. W. First, and O. Grubner. Marijuana and tobacco smoke gas-phase cytotoxins. Pharmacol. Biochem. Behav. 40: 629–636, 1991. 15. Hynes, R. O. Integrins: versatility, modulation and signaling in cell adhesion. Cell 69: 11–25, 1992. 16. Klein, C. E., D. Dressel, T. Steinmayer, C. Mauch, B. Eckes, T. Krieg, R. B. Barkert, and T. L. Weber. Integrin a2b1 is upregulated in fibroblasts and highly aggressive melanoma cells in three-dimensional collagen lattices and mediates the reorganization of collagen 1 fibrils. J. Cell Biol. 115: 1427–1436, 1991. 17. Labarca, C., and K. Paigen. A simple, rapid and sensitive DNA assay procedure. Anal. Biochem. 102: 344–352, 1980. 18. Lang, M. R., G. W. Fiaux, M. Gillooly, J. A. Stewart, D. J. S. Hulmes, and D. Lamb. Collagen content of alveolar wall tissue

L598

19.

20.

21. 22.

23.

24.

25. 26.

CIGARETTE SMOKE INHIBITION OF TISSUE REPAIR

in emphysematous and non-emphysematous lungs. Thorax 49: 319–326, 1994. Lin, L. L., A. Y. Lin, and D. L. DeWitt. Interleukin-1a induces the accumulation of cytosolic phospholipase A2 and the release of prostaglandin E2 in human fibroblasts. J. Biol. Chem. 267: 23451–23454, 1992. Lucey, E. G., P. J. Stone, and G. L. Snider. Consequences of proteolytic injury. In: The Lung: Scientific Foundations, edited by R. G. Crystal and J. B. West. New York: Raven, 1991, p. 1789–1802. Mauch, C., A. Hatamochi, K. Scharffetter, and T. Krieg. Regulation of collagen synthesis in fibroblasts within a threedimensional collagen gel. Exp. Cell Res. 178: 493–503, 1988. Mio, T., Y. Adachi, D. J. Romberger, R. F. Ertl, and S. I. Rennard. Regulation of fibroblast proliferation in three dimensional collagen gel matrix. In Vitro Cell. Dev. Biol. 32: 427–433, 1996. Mochitate, K., P. Pawelek, and F. Grinnell. Stress relaxation of contracted collagen gels: disruption of actin filament bundles, release of cell surface fibronectin, and down-regulation of DNA and protein synthesis. Exp. Cell Res. 193: 198–207, 1991. Montesano, R., and L. Orci. Transforming growth factor-b stimulates collagen-matrix contraction by fibroblasts: implication for wound healing. Proc. Natl. Acad. Sci. USA 85: 4894– 4897, 1988. Muirhead, K., T. Schmitt, and A. Muirhead. Determination of linear fluorescence intensities from flow cytometric data accumulated with logarithmic amplifiers. Cytometry 3: 251–256, 1983. Nakamura, Y., D. J. Romberger, L. Tate, R. F. Ertl, M. Kawamoto, Y. Adachi, T. Mio, J. H. Sisson, J. R. Spurzem,

27. 28. 29.

30.

31.

32.

33.

34.

and S. I. Rennard. Cigarette smoke inhibits lung fibroblast proliferation and chemotaxis. Am. J. Respir. Crit. Care Med. 151: 1497–1503, 1995. Pentland, A. P. Collagen lattice effects on fibroblast arachidonic acid metabolism. J. Cell. Physiol. 139: 392–397, 1989. Raghow, R. The role of extracellular matrix in postinflammatory wound healing and fibrosis. FASEB J. 8: 823–831, 1994. Rennard, S. I., R. Berg, G. R. Martin, J. M. Foidart, and P. G. Robey. Enzyme linked immunoassay (ELISA) for connective tissue components. Anal. Biochem. 104: 205–214, 1980. Schiro, J. A., B. M. C. Chan, W. T. Roswit, P. D. Kassner, A. P. Pentland, M. E. Hemler, A. Z. Eisen, and T. S. Kupper. Integrin a2b1 (VLA-2) mediates reorganization and contraction of collagen matrices by human cells. Cell 67: 403–410, 1991. Schwarzbauer, J. E., J. W. Tamkun, I. R. Lemischka, and R. O. Hynes. Three different fibronectin mRNAs arise by alternative splicing within the coding region. Cell 35: 421–431, 1983. Spurzem, J. R., O. Sacco, K. A. Rickard, and S. I. Rennard. Transforming growth factor-beta increases adhesion but not migration of bovine bronchial epithelial cells to matrix proteins. J. Lab. Clin. Med. 122: 92–102, 1993. Strom, S. C., and G. Michalopoulos. Collagen as a substrate for cell growth and differentiation. Methods Enzymol. 82: 544– 555, 1982. Wright, J. L., and A. Churg. Smoke-induced emphysema in guinea pigs is associated with morphometric evidence of collagen breakdown and repair. Am. J. Physiol. 268 (Lung Cell. Mol. Physiol. 12): L17–L20, 1995.