Oxidative Modifications of the C-terminal Domain of Tropoelastin Prevent Cell Binding*

Tropoelastin (TE), the soluble monomer of elastin, is synthesized by elastogenic cells, such as chondrocytes, fibroblasts, and smooth muscle cells (SMCs). The C-terminal domain of TE interacts with cell receptors, and these interactions play critical roles in elastic fiber assembly. We recently found that oxidation of TE prevents elastic fiber assembly. Here, we examined the effects of oxidation of TE on cell interactions. We found that SMCs bind to TE through heparan sulfate (HS), whereas fetal lung fibroblasts (WI-38 cells) bind through integrin αvβ3 and HS. In addition, we found that oxidation of TE by peroxynitrite (ONOO−) prevented binding of SMCs and WI-38 cells and other elastogenic cells, human dermal fibroblasts and fetal bovine chondrocytes. Because the C-terminal domain of TE has binding sites for both HS and integrin, we examined the effects of oxidation of a synthetic peptide derived from the C-terminal 25 amino acids of TE (CT-25) on cell binding. The CT-25 peptide contains the only two Cys residues in TE juxtaposed to a cluster of positively charged residues (RKRK) that are important for cell binding. ONOO− treatment of the CT-25 peptide prevented cell binding, whereas reduction of the CT-25 peptide had no effect. Mass spectrometric and circular dichroism spectroscopic analyses showed that ONOO− treatment modified both Cys residues in the CT-25 peptide to sulfonic acid derivatives, without altering the secondary structure. These data suggest that the mechanism by which ONOO− prevents cell binding to TE is by introducing negatively charged sulfonic acid residues near the positively charged cluster.


Tropoelastin (TE), the soluble monomer of elastin, is synthesized by elastogenic cells, such as chondrocytes, fibroblasts, and smooth muscle cells (SMCs). The C-terminal domain of TE interacts with cell receptors, and these interactions play critical roles in elastic fiber assembly. We recently found that oxidation of TE prevents elastic fiber assembly. Here, we examined the effects of oxidation of TE on cell interactions. We found that SMCs bind to TE through heparan sulfate (HS), whereas fetal lung fibroblasts (WI-38 cells) bind through integrin ␣ v ␤ 3 and HS. In addition, we found that oxidation of TE by peroxynitrite (ONOO ؊ ) prevented binding of SMCs and WI-38 cells and other elastogenic cells, human dermal fibroblasts and fetal bovine chondrocytes. Because the C-terminal domain of TE has binding sites for both HS and integrin, we examined the effects of oxidation of a synthetic peptide derived from the C-terminal 25 amino acids of TE (CT-25) on cell binding.
The CT-25 peptide contains the only two Cys residues in TE juxtaposed to a cluster of positively charged residues (RKRK) that are important for cell binding. ONOO ؊ treatment of the CT-25 peptide prevented cell binding, whereas reduction of the CT-25 peptide had no effect. Mass spectrometric and circular dichroism spectroscopic analyses showed that ONOO ؊ treatment modified both Cys residues in the CT-25 peptide to sulfonic acid derivatives, without altering the secondary structure. These data suggest that the mechanism by which ONOO ؊ prevents cell binding to TE is by introducing negatively charged sulfonic acid residues near the positively charged cluster.
Elastic fibers are key extracellular matrix structures that provide the stretch and recoil properties of tissues such as arteries, lungs, and skin. Elastic fibers consist of two major components, elastin and microfibrils. Elastin is the predominant component of elastic fibers, comprising Ͼ90% of the total mass. Tropoelas-tin (TE), 2 the soluble precursor of elastin, is synthesized by elastogenic cells, such as chondrocytes, fibroblasts, endothelial cells, and smooth muscle cells (SMCs). Assembly of monomeric TE into elastic fibers is a multistep process. Upon secretion from cells, TE monomers organize into aggregates on the cell surface. These aggregates are then deposited onto pre-existing microfibrils. Microfibrillar components align the TE aggregates, which undergo cross-linking to form mature elastic fibers. Cell surface molecules, such as the 67-kDa elastin-binding protein (EBP), ␣ v ␤ 3 integrin, and glycosaminoglycans (GAGs), have been proposed to be involved in elastic fiber assembly either by promoting aggregation of TE monomers or by keeping the microfibrils close to the cell membrane through interactions with microfibrillar components (1)(2)(3)(4)(5).
TE is mainly synthesized during late fetal and early postnatal stages of development. Synthesis of TE in normal adult tissues is negligible; however, in several cardiovascular and pulmonary diseases, such as atherosclerosis and emphysema/chronic obstructive pulmonary disease (COPD), elastogenesis is re-initiated suggesting elastic fiber repair mechanisms are activated. However, the integrity and organization of the elastic fibers appear abnormal (6 -11) suggesting that there is aberrant assembly of newly synthesized TE into elastic fibers in these conditions.
Oxidative stress has been implicated in the pathogenesis of several cardiovascular and pulmonary diseases. Oxidants can be generated by external factors such as cigarette smoke or internal factors such as inflammatory cells and mitochondrial respiration. These systems produce various reactive oxygen and nitrogen species, such as superoxide anion (O 2 . ), hydrogen peroxide (H 2 O 2 ), hydroxyl radical ( ⅐ OH), nitric oxide (NO), nitrite (NO 2 Ϫ ), and peroxynitrite (ONOO ؊ ) (12)(13)(14). These free radicals can modify proteins, resulting in an alteration of protein structure and function (15,16). We recently reported that oxidative/nitrosative modifications to TE prevent elastic fiber assembly in vitro, at least in part by inhibiting interactions of TE with other components of elastic fibers (17).
In this study, we investigated whether oxidative modification of TE affects its binding to cell surface receptors. We found that SMCs and fetal lung fibroblasts bind to TE via different mechanisms. Similar to previous reports using chondrocytes (2), SMCs bind to TE through cell surface proteoglycans via heparan sulfate (HS) chains, whereas fetal lung fibroblasts bind to TE via both HS moiety and integrin ␣ v ␤ 3 receptors. The binding of TE to the cell surface was mediated through the C-terminal domain of TE. In addition, we found that oxidation of TE or a peptide corresponding to the C-terminal domain of TE (CT-25) by ONOO Ϫ prevented cell binding. Using mass spectrometry, we determined that ONOO Ϫ modified both Cys residues in the C-terminal domain of TE and thereby prevented cell binding.

EXPERIMENTAL PROCEDURES
Cell Culture-Human dermal fibroblasts (HDFs) and human fetal lung fibroblasts (WI-38 cells) were purchased from American Type Tissue Culture Collection (Manassas, VA). Fetal bovine chondrocytes (FBCs) were isolated as described previously (2). Bovine arterial smooth muscle cells (SMCs) were provided by Kurt R. Stenmark (18). Cells were used before passage 6 and were split the night before use in adhesion assays. All cells were maintained in DMEM supplemented with 10% FBS, penicillin (100 units/ml), streptomycin (100 g/ml), and 4 mM L-glutamine.

Recombinant TE and CT-25
Peptides-Full-length bovine TE was expressed as His 6 fusion protein and purified using nickelnitrilotriacetic acid-agarose beads (Qiagen Inc., Valencia, CA) followed by HPLC as described previously (2,17). The peptide corresponding to the C-terminal 25 amino acids of TE (CT-25; FGGALGALGFPGGACLGKSCGRKRK) and a scrambled CT-25 peptide (KACSPARGLCKGFPGRGLGKLPGKG) were synthesized using standard FastMoc chemistry on an ABI-431A synthesizer, purified with reverse phase (C18) HPLC, and the sequence was confirmed by mass spectrometry as described previously (2).
ONOO Ϫ Exposure of TE and CT-25 Peptides-Full-length TE, CT-25, and scrambled CT-25 peptides were treated with ONOO Ϫ (Millipore Corp., Bedford, MA) as described previously (17). Briefly, immediately prior to each assay, the concentration of ONOO ؊ was determined spectrophotometrically at 302 nm (⑀ M ϭ 1670 M Ϫ1 cm Ϫ1 ) and diluted in 0.01 N NaOH. TE and CT-25 peptides were diluted in Tris-buffered saline (50 mM Tris, pH 7.5, and 150 mM NaCl), and ONOO ؊ was added while vortexing. The solutions were incubated at room temperature for 5 min. The pH was monitored to ensure that each reaction was performed at neutral pH.
Reduction of the CT-25 Peptide-The CT-25 peptide was incubated with or without ␤-mercaptoethanol overnight at room temperature, lyophilized, and reconstituted in sterile water. The reduction of the CT-25 peptide was confirmed by quantification of sulfhydryl groups using Ellman's reagent as described previously (19).
Cell Binding Assay-Cell binding to TE, CT-25, or scrambled CT-25 peptides was measured as described previously (2). Briefly, non-tissue culture microtiter plates (Costar Corp., Corning, NY) were coated with the designated concentrations of unmodified or ONOO Ϫ -modified TE in 10 mM carbonate buffer, pH 9.6, at 4°C overnight. Equal coating of unmodified and ONOO Ϫ -modified TE to the microtiter plate was confirmed as described previously (17) using the BA4 anti-elastin antibody (Abcam Inc., Cambridge, MA) and an HRP-conjugated anti-mouse secondary antibody (data not shown). The plates were rinsed with PBS and blocked with DMEM containing 1 mg/ml BSA (DMEM/BSA). Cells were detached using 10 mM EDTA in DMEM, resuspended in DMEM/BSA, plated, and incubated at 37°C for 1 h. Nonadherent cells were removed by centrifugation, and the amount of cell binding was quantified using hexosaminidase as a reporter and measured at 410 nm (20).
In some experiments, cells were incubated in the presence of ␣or ␤-lactose (10 mM), HS (100 g/ml), EDTA (10 mM), RGD or RGE peptides (1 mM), or the anti-␣ v ␤ 3 integrin antibody (Millipore Corp.). The concentrations of lactose, EDTA, HS, RGD peptide, and the anti-␣ v ␤ 3 antibody chosen for this study were shown to be inhibitory for cell binding in previous studies (1,2). To confirm the inhibitory effects of EDTA, the RGD peptide, and the anti-␣ v ␤ 3 antibody, cell binding to plates coated with fetal bovine serum was examined (data not shown).
Statistical Analysis-All statistical analysis was performed with the SPSS 13 program. Paired Student's t test was used to analyze the relationship between unmodified and ONOO Ϫmodified conditions. Data are representative of at least three independent experiments performed in triplicate, expressed as means Ϯ S.E. A p value of less than 0.05 was considered significant.
Mass Spectrometry-Mass spectrometric analyses were performed on a Thermo LTQ instrument (Thermo Fisher, San Jose, CA). Mock-treated and ONOO Ϫ -treated CT-25 peptides were desalted using ZipTips (Millipore, Billerica, MA) and analyzed by LC/MS/MS as described previously (21). Briefly, samples were infused into the ion source via a syringe pump at a rate of 2 l/min. The mass spectrometer was operated in positive ion mode with a spray voltage of 4.5 kV. MS spectra were acquired from m/z 150 to 2000 with a maximum ion injection time of 10 ms.
MS/MS spectra were acquired with a Q-TOF micro mass spectrometer (Waters Corp., Milford, MA) operated in datadependent scanning mode. The instrument was operated in positive ion mode with a nano-spray source. The energy setting was 30 eV for collision induced dissociation. Tandem mass spectra were acquired from m/z 50 to 2000 and processed with the MassLynx PepSeq software.
Circular Dichroism-Circular dichroism (CD) spectroscopic measurements of mock-treated, ONOO Ϫ -treated, or ␤-mercaptoethanol-treated CT-25 peptides (100 M in 0.1 M phosphate buffer) were performed using a Jasco J-810 spectropolarimeter. All spectra were collected in the range of 190 -250 nm at room temperature and corrected by subtraction of CD spectra from buffer blanks. The data are expressed as the molar ellipticity () in degrees/cm 2 /dmol Ϫ1 .

Arterial SMCs and WI-38 Lung Fibroblasts Bind to TE in a
Dose-dependent Manner-Elastin has been shown to interact with proteins on the cell surface of elastogenic cells, such as HDFs (1) and FBCs (2). Because SMCs and fibroblasts are the major elastogenic cells in arteries and lungs, respectively, we examined whether SMCs and WI-38 lung fibroblasts bind to TE using a cell binding assay. Non-tissue culture microtiter plates were coated with recombinant TE, and cell binding was determined by measuring the absorbance at 410 nm following addition of hexosaminidase (2). Both SMCs and WI-38 cells bound to TE in a dose-dependent manner (Fig. 1). In contrast, the cells failed to bind to wells coated with BSA. These results suggest that SMCs and lung fibroblasts bind to TE.
SMC and WI-38 Fibroblast Binding to TE Is Independent of the Elastin-binding Protein (EBP)-One mechanism by which elastin binds to cells is via the EBP (22), which is a splice variant of ␤-galactosidase (23). Previous studies have shown that lactose can inhibit the interaction of TE with the EBP (22,24). However, the addition of ␣or ␤-lactose, at concentrations previously shown to be inhibitory of cell binding (22,24), did not disrupt the binding of SMCs or WI-38 cells to TE ( Fig. 2A). These data show that SMC and lung fibroblast binding to TE are not mediated through the EBP.

SMCs Bind to TE via Heparan Sulfate (HS), whereas Binding of WI-38 Cells to TE Is Both HS-and Integrin-mediated-Pre-
vious studies have suggested that both GAGs and integrins can mediate cell binding to TE. For example, FBCs bind to TE via GAGs, with a preference for HS (2). In contrast, HDFs bind to TE via ␣ v ␤ 3 integrin (1). To determine whether SMCs and WI-38 fibroblasts bind to TE via an integrin-or GAG-mediated mechanism, cell binding was performed in the presence of 10 mM EDTA and/or 100 g/ml HS, concentrations previously shown to inhibit cell binding (1,2). We found that the addition of HS significantly inhibited the binding of both SMCs (ϳ80% decrease, p Ͻ 0.005) and WI-38 cells (ϳ45% decrease, p Ͻ 0.05) to TE (Fig. 2B). The addition of EDTA also significantly inhibited the binding of WI-38 fibroblasts (ϳ50% decrease, p Ͻ 0.05) but did not inhibit the binding of SMCs to TE (Fig. 2C). The addition of EDTA also inhibited the binding of WI-38 cells to serum (data not shown). The presence of both EDTA and HS did not have additive effects on the binding of SMCs to TE, as compared with the inhibition of TE binding by HS alone (Fig.  3A). In contrast, the combination of EDTA and HS had an addi-tive inhibitory effect (ϳ90% decrease, p Ͻ 0.005) on WI-38 cell binding to TE (Fig. 3B).
Because the tripeptide RGD inhibits the binding of many integrin receptors (25), we examined the effects of the RGD peptide on the cell binding to TE. The addition of 1 mM RGD or the control RGE peptide had no effect on the adhesion of either SMCs or WI-38 cells to TE (Fig. 2C). However, previous studies have shown that cells can bind to TE via ␣ v ␤ 3 integrin in an RGD-independent manner (1). To determine whether SMCs or WI-38 cells bind to TE via ␣ v ␤ 3 integrin, cell binding was per-  formed in the presence of an anti-␣ v ␤ 3 integrin blocking antibody. We found that the addition of 20 g/ml of the anti-␣ v ␤ 3 integrin antibody significantly inhibited (ϳ53% decrease, p Ͻ 0.05) the binding of WI-38 cells to TE but did not inhibit the binding of SMCs to TE (Fig. 2C). These data show that both SMCs and WI-38 cells bind to TE via HS but WI-38 cells also bind via an integrin-mediated mechanism.
As controls, we examined the effects of HS and EDTA on the binding of FBCs and HDFs to TE. We found that the addition of HS significantly inhibited the binding of both FBCs (ϳ90% decrease, p Ͻ 0.005) (supplemental Fig. 1A) and HDFs (ϳ50% decrease, p Ͻ 0.05) to TE (supplemental Fig. 1B), whereas the presence of EDTA inhibited the binding of HDFs (ϳ60% decreased, p Ͻ 0.005) (supplemental Fig. 1B) but not FBCs (supplemental Fig. 1A). The presence of both EDTA and HS did not have additional effects on the binding of FBCs to TE, as compared with the inhibition of TE binding by HS alone (supplemental Fig. 1A). In contrast, the combination of EDTA and HS had an additive inhibitory effect on HDF cell binding to TE (ϳ90% decrease, p Ͻ 0.005) (supplemental Fig. 1B). These data support our previous findings (2) that FBCs bind to TE via HS but appear to be in conflict with another report (1) that showed that HDF cell binding is mediated via ␣ v ␤ 3 integrin but not through HS.
The addition of the RGD peptide had no effect on the binding of HDFs or FBCs, although the presence of the anti-␣ v ␤ 3 integrin antibody significantly inhibited (ϳ65% decrease, p Ͻ 0.005) the binding of HDFs to TE but did not inhibit the binding of FBCs to TE (supplemental Fig. 2). Together, these data suggest that fibroblasts (HDFs and WI-38 lung fibroblasts) have multiple mechanisms to bind TE, HS-and integrin-mediated, whereas SMCs and chondrocytes mainly use HS for interacting with TE.
SMC and WI-38 Cell Binding Is Mediated through the C-terminal Domain of TE-The C-terminal domain of TE has been implicated in the binding of FBCs and HDFs (1,2). To determine whether the C-terminal domain of TE mediates the binding of WI-38 cells and SMCs, we used a synthetic peptide corresponding to the C-terminal 25 amino acids of TE (CT-25). Both WI-38 cells and SMCs bound to the CT-25 peptide (Fig.  4). To determine whether the mechanisms of binding of the SMCs and WI-38 cells to the C terminus of TE were the same as for full-length TE, binding of the cells to the CT-25 peptide was carried out in the presence of EDTA, HS, or both. The presence of EDTA did not affect the binding of SMCs, whereas HS completely prevented (p Ͻ 0.005) SMC binding to the CT-25 peptide (Fig. 4). In contrast, the presence of either EDTA or HS partially inhibited (ϳ60% decrease, p Ͻ 0.05) the binding of WI-38 cells to the CT-25 peptide. The presence of both EDTA and HS had an additive inhibitory effect and almost completely prevented (p Ͻ 0.005) the binding of WI-38 cells to the CT-25 peptide (Fig. 4).
As controls, we also examined the effects of EDTA and HS on the binding of FBCs and HDFs to the CT-25 peptide. Similar to our findings using full-length TE (supplemental Fig. 1), we found that FBCs bound to CT-25 peptide via HS and that EDTA alone or in combination with HS did not affect the binding of the CT-25 TE peptide (supplemental Fig. 3). In contrast, the presence of EDTA and/or HS significantly reduced the binding of HDFs to the CT-25 peptide (supplemental Fig. 3). Together, these data suggest that SMCs and chondrocytes bind to full-  length TE via HS on the cell surface binding to the C-terminal domain of TE, whereas fibroblasts bind to the C terminus of TE via HS-and integrin-mediated mechanisms.
Oxidative Modifications of TE Prevent Cell Binding-During aging and several cardiovascular and pulmonary diseases, elastin synthesis is re-initiated; however, there is aberrant assembly of newly synthesized TE into elastic fibers (6 -11). These conditions are associated with oxidative stress. Recently, we showed that ONOO Ϫ , an oxidant released by activated inflammatory cells, modifies TE and that these modifications prevent the assembly of TE into elastic fibers in vitro (17). Because the interaction of TE with cells plays a role in elastic fiber assembly, we sought to determine whether oxidative modifications of TE by ONOO Ϫ alter the interactions with cells. We found that SMCs and WI-38 cells were unable to bind to oxidized TE (Fig.  5A). Similar results were obtained using HDFs and FBCs (Fig.  5B). These results indicate that ONOO Ϫ modifies amino acid residues of TE that are critical for cell binding.
Oxidative Modifications of the CT-25 Peptide Prevent Cell Binding-To determine whether oxidation of the C-terminal domain of TE inhibits cell binding, the CT-25 peptide was exposed to ONOO Ϫ and used as a substrate for cell binding. As shown in Fig. 4, the CT-25 peptide promoted binding of SMCs and WI-38 cells (Fig. 6A), but oxidative modification of CT-25 peptide by ONOO Ϫ completely prevented the binding of SMCs and WI-38 cells (Fig. 6A). In addition, oxidative modification of CT-25 also prevented the binding of FBCs and HDFs (Fig. 6B).
The CT-25 peptide has a cluster of positive charges near the C terminus of the peptide (FGGALGALGFPGGACLGK-SCGRKRK) that could contribute to the cell binding. To further delineate the mechanism by which the CT-25 peptide promotes cell binding, we used a scrambled CT-25 peptide that has the same amino acid residues and same overall charge, but the charges are redistributed. We found that the scrambled CT-25 peptide does not promote binding of SMCs or WI-38 cells (Fig.  6A) or the binding of HDFs or FBCs (Fig. 6B). These data suggest that the cluster of positive charges, and not the overall charge, of the CT-25 peptide promotes cell binding.
The CT-25 peptide has two Cys residues (FGGALGALGF-PGGACLGKSCGRKRK) that form a disulfide bond near the cluster of positively charged residues. To determine whether the disulfide-bonded Cys residues in the CT-25 peptide play a role in cell binding, the CT-25 peptide was reduced by treatment with ␤-mercaptoethanol, and we confirmed that the CT-25 peptide remained reduced under the conditions of the experiment using Ellman's reagent (data not shown). We found that the reduced CT-25 peptide retained binding activity for SMCs and WI-38 cells (Fig. 6A) and for HDFs and FBCs (Fig.  6B) with almost equal efficiency as untreated, control CT-25 peptide. These data further suggest that the cell adhesive prop-  erties of the CT-25 peptide are mediated by the cluster of positive charges and that ONOO Ϫ is modifying critical amino acids in or near the cluster, altering the charge in this region, and thus preventing the cell binding.
Cys Residues in the C-terminal Domain of TE Are Modified by ONOO Ϫ -To identify the oxidatively modified amino acids in the C-terminal domain of TE, the CT-25 peptide was exposed to ONOO Ϫ (CT-25-Ox) and analyzed by mass spectrometry. The CT-25 peptide was exposed to degraded ONOO Ϫ to serve as a mock-treated, unoxidized control (CT-25-UnOx). The theoretical molecular mass of the CT-25 peptide is 2408 Da. However, the observed molecular mass (M) of the CT-25-UnOx peptide, as calculated from the triply or quadruply charged ions ([M ϩ 3H] 3ϩ m/z 803 and [M ϩ 4H] ϩ m/z 602, respectively), was 2406 Da (Fig. 7A). This is 2 Da less than the theoretical molecular mass, suggesting that there is a disulfide bond between the two Cys residues within the CT-25 peptide.  (Fig. 7C). This reflects the addition of three oxygen atoms to each Cys residue and the conversion of both Cys residues to sulfonic acid (SO 3 H) derivatives. No other modifications in the CT-25-Ox peptide were found.
Oxidative Modifications of the CT-25 Peptide Do Not Alter Secondary Structure-Previous studies have examined the secondary structure of a peptide corresponding to the C-terminal exon 36 of human TE (GGACLGKACGRKRK) and found some random coil and some ␤-turn structures by circular dichroism analysis, and these structures are not affected upon disruption of the disulfide bond by reduction of peptide (26). Because ONOO Ϫ modified the Cys residues in the CT-25 peptide (FGGALGALGFPGGACLGKSCGRKRK), we sought to determine whether these modifications altered the secondary structure of the CT-25 peptide. Similar to the previous study (26), the CD spectra of the mock-treated, control CT-25 peptide displayed a dominant negative band around 200 nm, but the intensity of this band was markedly decreased in intensity relative to a fully random coil peptide, indicating some ␤-turn structure (Fig. 8). Treatment of the CT-25 peptide with ONOO Ϫ (CT-25-Ox) or ␤-mercaptoethanol (CT-25-Red) did not affect the secondary structure of the CT-25 peptide (Fig. 8). These results suggest that disruption of disulfide bond in the CT-25 peptide, either by oxidative modification or reduction, does not affect the overall secondary structure of peptide.

DISCUSSION
TE has been shown to interact with proteins on the surface of multiple cell types. Cellular interactions with TE influence cell adhesion, migration, proliferation, actin polymerization, mobilization of intracellular Ca 2ϩ , and gene expression (1,2,(27)(28)(29)(30)(31). In addition, cell surface molecules promote elastic fiber assembly by promoting aggregation of TE monomers and/or by keeping microfibrils close to the cell membrane through interac-tions with microfibrillar components (5). TE is available to interact with cells only during elastogenesis, which normally occurs during late fetal and early postnatal stages of development. However, elevated elastin synthesis has been observed during aging and in several pathologic conditions, such as atherosclerosis and emphysema/COPD. As these conditions are associated with oxidative stress, we examined the effects of oxidative modifications of TE on cell binding.
Our studies showed that isolated arterial SMCs and pulmonary fibroblasts (WI-38 cells) bind to recombinant TE using a solid-phase assay. These interactions were specific as cells failed to bind to BSA. In addition, we found that oxidative modifications to TE by ONOO Ϫ prevented the interaction of TE with these cell types, as well as binding of dermal fibroblasts (HDFs) and chondrocytes. These data suggest that oxidative modifications to TE that might occur during pathologic conditions, such as atherosclerosis, COPD, or UV-exposed skin, could affect cellular interactions with TE, and in turn alter cell functions.
TE is synthesized by elastogenic cells, such as SMCs in blood vessels, fibroblasts in skin and lungs, and chondrocytes in elastic cartilage. An early step in elastic fiber assembly is the aggregation of TE on the cell surface. Our data showed that each of these cell types bind to TE through a GAG-mediated mechanism, inhibited by the addition of HS. However, only the two fibroblastic cell types, HDFs and WI-38 lung fibroblasts, bound to TE in an integrin-mediated mechanism, as indicated by the inhibition of cell binding in the presence of EDTA or the anti-␣ v ␤ 3 integrin-blocking antibody. In contrast to our study, Bax et al. (1) found that the interaction of HDFs to TE is solely integrin-mediated and that binding of HDFs to TE is not affected by the addition of HS. The reason for this discrepancy is unknown. Possibilities include different sources of the cell line, culture conditions, or variations in the methodology of the assay, such as how cells were removed from the plate or the use of non-tissue culture plates. Nonetheless, these data show that different cell types use different receptors to bind to TE. This could reflect that different cell types have a different repertoire of cell surface receptors depending on cell type, stage of development, or expression level of TE. Our data also showed that a single cell type can have multiple receptors for TE. It is plausible that these receptors coordinate upon binding of TE. Other GAGs, such as syndecan-4, cooperatively signal with integrins for cell adhesion (32,33), cytoskeletal organization (34,35), and cell migration (36 -38). Thus, for some cells, coordinated recognition of TE may be essential for proper elastic fiber assembly or for cell signaling.
Indirect evidence suggests that the C-terminal domain of TE is important for elastic fiber assembly, as TE molecules that lack this region do not properly assemble into elastic fibers (39). Consistent with previous reports (1, 2), we found that both HS and integrin-binding sites are located in the C-terminal domain of TE. The C terminus of TE contains the only two Cys residues of the protein juxtaposed to a cluster of basic amino acids (CLGKSCGRKRK). Both Cys residues are disulfide-bonded (19,26), which may play a role in the overall structure of the region. The interaction of TE with HS is dependent on the cluster of positive charge. This was supported by the lack of cell binding to the scrambled CT-25 peptide, which has the same overall charge, but the charges are redistributed across the peptide.
Elastogenesis can occur in oxidative environments during several pathologic conditions, including atherosclerosis and emphysema/COPD; however, the assembly of newly synthesized TE is abnormal (6 -11). In addition, recent studies have shown that TE becomes carbonylated in response to cigarette smoke exposure in a mouse model of emphysema (40). Therefore, we recently examined the effects of oxidative modifications of TE on elastic fiber assembly using an in vitro model system (17). We found that TE becomes oxidatively and nitrosatively modified by oxidants, such as ONOO Ϫ and HOCl, and by activated monocytes and macrophages, as indicated by Oxy-Blot and N-Tyr immunoblots, respectively. Furthermore, we showed that these modifications inhibited elastic fiber assembly, at least in part by inhibiting interactions of TE with other components of elastic fibers (17). In this study, we examined the effects of oxidative modifications of TE on cell binding and used mass spectrometry to identify the specific amino acids that were modified by ONOO Ϫ . Because cell interaction sites were located to the C-terminal domain of TE, we used the CT-25 peptide to determine which amino acids were modified by ONOO Ϫ . The CT-25 peptide contains three Lys, two Arg, and two Cys residues with side chains that are potentially susceptible to oxidative modification, any of which could potentially alter the interaction of TE with cells. However, we found that only the Cys residues were modified upon ONOO Ϫ treatment. This is likely because Cys residues are the most susceptible for oxidative modifications by ONOO Ϫ (41,42). Using higher concentrations of ONOO Ϫ may have resulted in the modification of other amino acid residues in the CT-25 peptide; however, the concentrations of ONOO Ϫ used in this study were in a range produced by inflammatory cells (43)(44)(45).
Mass spectrometry showed that treatment of the CT-25 peptide with ONOO Ϫ caused a disruption of disulfide bonds and converted the Cys residues to sulfonic acids. However, CD analysis showed that neither oxidation nor reduction of the disulfide bonds in the CT-25 peptide altered the overall structure of CT-25 peptide. Therefore, we conclude that oxidative modification of the Cys residues in the CT-25 peptide by ONOO Ϫ treatment prevented cell binding by the generation of negatively charged sulfonic acid residues near the positively charged cluster in CT-25. This is further supported by an earlier report where introduction of one negatively charged amino acid residue in the C-terminal domain of TE (G773D), which mimicked a mutation found in a pedigree with severe early onset COPD, in the C-terminal domain of TE resulted in reduced interactions with cells (46). The RKRK sequence of TE has also been shown to bind to ␣ v ␤ 3 integrin (1).
In this study we only examined the effects of ONOO Ϫ on cell binding; however, we speculate that other oxidants could also prevent cell binding to TE. Previously, we have shown that exposure of TE to either ONOO Ϫ or HOCl prevents elastic fiber assembly (17). Because HOCl also has the capability to oxidatively modify the Cys residues in the C-terminal domain of TE, to form sulfenic acid derivatives (Cys-SOH), which would also alter the charge on C-terminal domain, we speculate that oxidation of the CT-25 peptide by HOCl would also prevent cell binding.
The loss of cell interactions with TE due to oxidative modifications could affect biologic processes, which in turn might contribute to the development of various cardiovascular and pulmonary diseases. One possible mechanism is by altering elastic fiber assembly. The data presented here show that the CT-25 peptide derived from the C-terminal domain of TE can become oxidized and loses interactions with cell surface receptors of multiple elastogenic cells, which in turn suggests that oxidation of TE in C-terminal region could impair elastic fiber assembly.
In addition to the potential negative effects of oxidative modifications of TE on elastic fiber assembly, loss of cell interactions with TE could affect cell function. For example, the lack of and/or loss-of-function mutations of elastin cause abnormal proliferation and accumulation of SMCs and excessive deposition of ECM in the intimal region of blood vessels (47)(48)(49)(50)(51). Furthermore, mutations in the region of C-terminal domain of TE have been associated with cutis laxa that has fragmented elastic fibers in skin (52,53).
Oxidation of TE or elastin-derived peptides could affect functions of nonelastogenic cells, such as monocytes, macrophages, neutrophils, and lymphocytes as well. These cells bind to elastin and elastin-derived peptides via the EBP. This receptor recognizes an Xaa-Gly-Xaa-Xaa-Pro-Gly (XGXXPG) motif found repeatedly in elastin. Binding to the EBP triggers chemotaxic migration and the production of proteases and cytokines (54,55). Although our studies focus on the C-terminal domain of TE, which lacks this motif, and its interactions with elastogenic cells, we speculate that TE or elastin-derived peptides that are generated during pathologic conditions might become oxidatively modified, alter the interactions with inflammatory cells, and in turn may contribute in the development of diseases, such as atherosclerosis and emphysema.
In summary, this study investigated the effects of oxidative modifications of TE on its interaction with elastogenic cells. We found that different cell types mediated binding via different mechanisms, GAG-and integrin-mediated, and that a single FIGURE 8. ONOO ؊ treatment of the CT-25 peptide does not alter its secondary structure. The CT-25 peptide was treated with degraded ONOO Ϫ (control), active ONOO Ϫ (CT-25-Ox), or ␤-mercaptoethanol (CT-25-Red), and the secondary structure was analyzed by circular dichroism spectroscopy. All spectra were collected at room temperature and corrected by subtraction of CD spectra from buffer blanks. The data are expressed as the molar ellipticity () in degrees/cm 2 /dmol Ϫ1 . cell can have multiple mechanisms for binding. The cellular interactions with TE were mediated through the C-terminal domain of TE. Exposure of TE or the CT-25 peptide to ONOO Ϫ prevented both HS-and integrin-mediated binding by oxidatively modifying the two Cys residues to form sulfonic acid derivatives. We believe that these studies suggest new mechanisms by which altered cellular interactions with TE could contribute to the pathophysiology of various cardiovascular and pulmonary diseases.