Exposure to peroxynitrite impacts the ability of anastellin to modulate the structure of extracellular matrix

.


Introduction
The ECM plays a key role in determining the properties of tissues by influencing and controlling cell adhesion, proliferation and phenotype, and the binding of growth factors, signalling molecules and enzymes [1].The extracellular matrix (ECM) is a complex and 'plastic' (i.e.adaptable) tissue-dependent 3-dimensional structure that consists of multiple proteins, proteoglycans and glycosaminoglycans.Many of these components interact with fibronectin (FN), a major ECM protein that acts as a scaffold and a "master regulator" of ECM structure [2].FN is secreted by multiple cell types, including smooth muscle cells and fibroblasts, and assembled into insoluble, flexible fibrils that are involved in many aspects of normal cell physiology as well as various pathological processes [3][4][5].In contrast, FN from hepatocytes is mainly released in a soluble form that circulates in plasma, although it also can be incorporated into ECM fibrils under some circumstances [6].FN protomers consist of two monomer chains (~250 kDa) connected via a pair of C-terminal disulfide bonds.Each monomer chain contains multiple repeat modules (12 x type I (FNI), 2 x type II (FNII) and 15-17 x type III (FNIII)), including two alternatively spliced FNIII modules (EDA Abbreviations: ABTS, 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid); AN, anastellin; DAPI, 4′,6-diamidino-2-phenylindole; ECM, extracellular matrix; FN, fibronectin; HBSS, Hanks' balanced salt solution; HCASMC, human coronary artery smooth muscle cells; LDH, lactate dehydrogenase; mAb, monoclonal antibody; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium; ONOO − /ONOOH, the physiological mixture of peroxynitrite (ONOO − ) and peroxynitrous acid (ONOOH) present at pH 7.4; pAb, polyclonal antibody; qPCR, quantitative real-time PCR.* Corresponding author.** Corresponding author.E-mail addresses: davies@sund.ku.dk (M.J. Davies), pmh@sund.ku.dk (P.Hägglund). 1 Present address: College of Food Science and Technology, Zhejiang University of Technology, Hangzhou 310014, China. 2 Joint senior.and EDB) that are usually absent in plasma FN [7].The FN modules are organized into functional domains that provide binding sites for other macromolecules, including collagens, fibrin, heparin and integrins [8].The exact mechanism of FN polymerization is not established, but according to a widely-supported model, cytoskeletal actins exert a pulling force on FN through interactions via membrane-bound integrin receptors (mainly α5β1) and proteoglycans [9][10][11].This results in an unfolding of the compact structure of FN, and particularly the first FNIII module (FNIII 1 ), with this resulting in the exposure of hidden (cryptic) sites.These stabilize interactions between individual protomers and promote polymerization [12].FNIII 1 also contains a cryptic cluster of positively-charged residues that bind to heparin-based polysaccharides and are important for regulation of FN-cell interactions [13].
Anastellin (AN), a recombinant fragment of FNIII 1 , can induce cellfree FN polymerization in vitro into "superfibronectin" (sFN), a fibrillar form, which resembles ECM FN at the microscopic level [14].AN can therefore be used as a model to study the mechanism of FN polymerization in vitro [15,16].AN has also been shown to modulate the structure of the ECM and impact on the post-translational processing of FN in cells co-incubated with AN [17].Furthermore, AN has antiangiogenic and antimetastatic properties, suppresses tumor growth, and influences vascular endothelial growth factor and p38 signaling [18][19][20][21].A cleavage site for matrix metalloproteinase 2 is located near the N-terminal end of AN [22], and it has therefore been proposed that proteolytic fragments related to AN may be generated during ECM turnover in vivo [23].
Previous in vitro studies have demonstrated that AN can be modified by the physiological mixture of the peroxynitrite anion (ONOO − ) and peroxynitrous acid (ONOOH) (henceforth designated as 'peroxynitrite', with the acid form being the more reactive species except in the case of CO 2 ) in a dose-dependent manner [24].Peroxynitrite is released from activated macrophages and is implicated in a variety of inflammatory diseases, including atherosclerosis [25,26], with this oxidant being formed in vivo from reaction of superoxide radical anions (O 2 •-, generated by NADPH oxidases and other processes), and nitric oxide (NO • , produced by nitric oxide synthases [27]).Peroxynitrite reacts rapidly with proteins and a variety of other biological targets, either by direct oxidation or through reactions mediated by reactive radicals (HO • , NO 2 • , and CO 3 •-in the presence of bicarbonate) generated by secondary reactions [28].ECM proteins including FN, laminin, elastin and perlecan are major targets of peroxynitrite in vivo, resulting in structural perturbations, modifications of amino acid side chains (e.g.cysteine, methionine, Tyr, Trp) and the formation of covalent crosslinks (e.g.di-tyrosine and others) [29][30][31][32][33][34].These proteins can also be modified by other oxidants, such as hypochlorous acid (HOCl) generated by myeloperoxidase, which is released from activated neutrophils, monocytes and some tissue macrophages [35].Exposure of ECM proteins and vascular cells to oxidants such as peroxynitrite and HOCl has a major influence on their structure and function, and in the case of cellular exposure results in altered protein expression, cell adhesion and interactions with ECM components [36,37].Upon exposure to peroxynitrite, Tyr and Trp residues in AN are readily converted to 3-nitrotyrosine and 6-nitrotryptophan, respectively, due to the absence of other peroxynitrite-reactive residues such as cysteine and methionine [24].Nearly complete loss of parent Tyr and Trp was observed at a 100-fold molar excess of oxidant.These molecular changes adversely affect the ability of AN to induce polymerization of FN into sFN, although the overall structure and beta-sheet fold of the protein was maintained [24].High doses of peroxynitrite alter the primary, secondary and tertiary structures of AN, and also induce protein aggregation.Corresponding experiments with decomposed peroxynitrite yielded no significant alterations in the structure of AN, suggesting that the reactions are specific for peroxynitrite and not related to other compounds present in the peroxynitrite solution (such as nitrite or nitrate).
In this work, we hypothesized that exposure of AN to peroxynitrite would also impact on the ability of AN to influence the structure of ECM generated by primary human coronary artery smooth muscle cells (HCASMC).These cells play a critical role in the pathogenesis and progression of atherosclerosis [38].Inflammatory oxidants are proposed to act as drivers of a phenotypic switch in smooth muscle cells to a proliferative phenotype that can migrate into atherosclerotic lesions [35,39].Furthermore, modifications to ECM structure and integrity play a key role in determining the stability of the fibrous cap of atherosclerotic lesions [7,40].Rupture of weakened lesions is a key driver of both strokes and myocardial infarctions [41].

Preparation of peroxynitrite-modified AN
AN (40 μM) was incubated with different concentrations of peroxynitrite (0-8000 μM; 0-200 fold molar excess) for 30 min at 21 • C in the presence of 100 mM sodium phosphate buffer (pH 7.4).The samples were concentrated and buffer-exchanged to 10 mM sodium phosphate buffer (pH 7.4) using spin filters (Amicon Ultra-0.5 Centrifugal Filter Unit, 3 kD).Samples were centrifuged at 20,000 g for 10 min at 4 • C to remove possible aggregates, and protein concentration was determined using the bicinchoninic acid (BCA) protein assay (Thermo Fisher Scientific), measuring changes in absorbance at 560 nm with a standard curve constructed with bovine serum albumin (BSA).The AN protein concentration was then normalized to 100 μM for subsequent experiments.

Culture of primary human coronary artery smooth muscle cells (HCASMC)
HCASMC (donor 1522, passages 2-5; Cell Applications, San Diego, CA, USA) were cultured at an initial density of 1 × 10 6 cells in 25 mL growth medium (Cell Applications) in T-175 cm 2 flasks in a humidified incubator under an atmosphere of 5% CO 2 at 37 • C for 1 week, with the media changed 3 times/week.The cells were harvested with 10 mL of trypsin/EDTA solution (0.025% trypsin, 0.01% EDTA, in PBS) for 3-5 min and centrifuged at 220 g for 5 min.Cell suspensions with different densities were prepared in growth medium: 1), 1 × 10 5 cells mL − 1 were placed in 6-, 12-or 96-well plates, with the total volume for each well of 2 mL, 1 mL or 50 μL, respectively; 2) 7.5 × 10 4 cells mL − 1 were placed in 8-well chamber slides (VWR International, Dorset, UK) with 300 μL/ well, or in 96-well plates with 200 μL well − 1 overnight.Before further experiments, the cell medium was removed, and cells washed with warm (37 • C) Hanks' Balanced Salt Solution (HBSS; Life Technologies, Hvidovre, Denmark).

Immunofluorescence staining 2.4.1. Detection of FN in HCASMC incubated with peroxynitrite-modified AN
HCASMC (2.25 × 10 4 cells in 300 μL growth medium per well) were seeded into 8-well glass chamber slides overnight, and then incubated with 200 μL peroxynitrite-modified AN (30 μM) in growth medium for 48 or 24 h at 37 • C.After washing with PBS, cells were fixed with 4% (v/ v) formaldehyde at 37 • C for 15 min and permeabilized with 0.5% (v/v) Triton X-100 in PBS, on ice for 5 min.Wells were then blocked with 1% (w/v) BSA in PBS for 1 h at 21 • C, followed by incubation with primary FN antibodies (1:500 dilution of pAb ab2413 and mAb 3E2) overnight at 4 • C. The wells were washed twice with PBS prior to incubation with anti-mouse and anti-rabbit IgG conjugated with either Alexa Fluor 488 or Alexa Fluor 594 antibody (1:500 dilution) for 1 h at 21 • C, washed twice with PBS, and counterstained with 1 μg mL − 1 DAPI for 10 min at 21 • C in the dark.After washing twice with PBS, the slides were imaged using a fluorescence microscope (Olympus, Japan) equipped with cell-Sense Entry v1.5 software.

Detection of FN in decellularized ECM from HCASMC
Cells seeded into 8-well glass chamber slides as described above were incubated in 200 μL growth medium containing 30 μM peroxynitritemodified AN for 48 h at 37 • C. The growth medium was removed and wells were washed twice with PBS before the addition of 1% (w/v) sodium deoxycholate to lyse the cells.The remaining ECM proteins were washed twice with PBS, blocked with 1% (w/v) BSA in PBS for 1 h, followed by incubation with primary FN antibodies (1:500 dilution of pAb 2413 and mAb 3E2) overnight at 4 • C. The wells were washed twice with PBS, incubated with both anti-mouse and anti-rabbit IgG conjugated with Alexa Fluor 488 or Alexa Fluor 594 (1:500 dilution) for 1 h at 21 • C, and washed twice with PBS.A glass cover slip was added, and the slides were imaged using a fluorescence microscope as described above.

Detection of AN binding to ECM
Cells (2.25 × 10 4 cells in 300 μL growth medium per well) were incubated in 8-well chamber slides for 1 week (in the absence of AN), decellularized with 1% (w/v) sodium deoxycholate, and blocked with 1% (w/v) BSA in PBS, before incubation with 200 μL 0.1 μM AN (treated with peroxynitrite as described above) overnight at 4 • C.After rinsing with PBS and blocking with 1% (w/v) BSA, slides were probed with FN pAb (2413; 1:500 dilution) and His-tag mAb (18184; 1:500 dilution) overnight at 4 • C, washed with PBS, and then incubated with either antimouse or anti-rabbit IgG conjugated with Alexa Fluor 488 or Alexa Fluor 594 (1:500 dilution) for 1 h at 21 • C.After washing with PBS, the plates were visualized using a fluorescence microscope as above.The plates were then washed with PBS before incubation with two successive aliquots of 1% (w/v) sodium deoxycholate for 20 min.After washing with PBS, the native ECM remaining on the plates was blocked with 1% (w/v) BSA in PBS for 1 h at 21 was added, and the absorbance was measured at 405 nm using a Spec-traMax i3x microplate reader (Molecular Devices, San Jose, CA).

Detection of AN bound to isolated plasma FN
Peroxynitrite-modified AN (50 μL; 0.25 μM) was added to wells precoated with 50 μL plasma FN (1 μg mL − 1 ) and incubated overnight at 21 • C. Plates were blocked with 1% (w/v) BSA in PBS for 1 h at 21 • C, washed with PBS, and probed with a His-tag antibody (18184) at 4 • C overnight.Wells were washed with PBS, incubated with HRPconjugated anti-mouse IgG secondary antibody (1:1000) for 1 h at 21 • C, and developed with ABTS solution as described above, prior to detection at 405 nm using a SpectraMax i3x microplate reader.

Immunoblot analysis
HCASMC (2 × 10 5 cells) were cultured in 6-well plates (2 mL growth medium per well) overnight before incubation with 1.5 mL growth medium containing peroxynitrite-modified AN (30 μM) for 48 h.The plates were then washed with PBS and incubated with 1% (w/v) sodium deoxycholate (2 × 20 min) with the pooled supernatants transferred and centrifuged (14000 g, 10 min, 4 • C) to remove cell debris.Protein concentrations were determined with BCA (Thermo Fisher Scientific) and normalized to 10 μg total protein before loading on 3-8% Tris-acetate SDS-PAGE gels, under reducing conditions.Using an iBlot system, proteins were transferred onto PVDF membranes that were blocked with 1% (w/v) BSA in TBST (TBS containing 0.1% Tween 20) for 1 h at 21 • C, incubated with primary antibodies (FN pAb (2413) or β-actin mAb (MAB8929); 1:1000 dilutions) overnight at 4 • C, followed by washing with TBST.The membranes were then incubated with HRP-conjugated anti-mouse or anti-rabbit IgG secondary antibodies (1:5000 dilution) for 1 h at 21 • C, washed 3 times with TBST and rinsed once with TBS before development with Western Lightning Plus ECL reagents (Perki-nElmer, Hägersten, Sweden), using a Sapphire Biomolecular Imager (Azure Biosystems, Dublin, CA, USA) to acquire images.

Lactate dehydrogenase (LDH) release
HCASMC (1 × 10 5 cells) were cultured in 12-well plates (1 mL growth medium per well) overnight and then incubated in growth medium containing 1 mL peroxynitrite-modified AN (30 μM) for 48 h.Supernatants were collected, and the cells washed with HBSS and lysed with 1 mL H 2 O.The supernatants and lysed cells were centrifuged at 448 g for 5 min at 4 • C to remove cell debris, and 10 μL of the supernatants were added to a 96-well plate and mixed with 200 μL of reaction reagent containing 0.15 mg mL − 1 NADH and 2.5 mM sodium pyruvate in PBS.LDH activity was measured by detecting the decrease in absorbance at 340 nm (from NADH) for 30 min at 5 min intervals with a microplate reader.Cell viability was calculated from the intracellular LDH activity compared to the total intra-and extracellular LDH activity, expressed as a percentage.

Quantitative real-time polymerase chain reaction (qPCR)
HCASMC (1 × 10 5 cells) were cultured in 12-well plates (1 mL growth medium per well) overnight and incubated with 1 mL growth medium containing peroxynitrite-modified AN (30 μM) for 2 h.Total RNA was extracted using a RNeasy kit (Qiagen, Hilden, Germany), and genomic DNA removed using RNase Free DNase (Qiagen).RNA (600 ng) was used for single strand cDNA synthesis using a Quantinova Reverse Transcription Kit (Qiagen) in a total volume of 20 μL.AGATGAGTATGCCTGCCGTG, R: GCGGCATCTTCAAACCTCCA. Triplicate wells were included in each group and data analysis was carried out using the 2 − ΔΔCT method.

Adhesion of HCASMC
Peroxynitrite-modified AN (0.25 μM) either individually or mixed with 25 μL FN (1 μg mL − 1 ) was added to a 96-well black polystyrene microplate (sterile, F-bottom, Corning, NY, USA).Alternatively, the Adhesion of dye-loaded HCAMSC to AN was determined by measuring the fluorescence intensity with λ ex 490 nm and λ em 520 nm in a Spec-traMax i3x microplate reader.

Heparin-affinity chromatography
Peroxynitrite-modified AN (80 μM) in 100 μL 20 mM sodium phosphate buffer pH 7.4 was loaded onto a Hitrap heparin high performance column (Cytiva, UK) on an Äkta Prime FPLC system (Cytiva, UK), and eluted with a 20 min gradient of 0-500 mM NaCl in 20 mM sodium phosphate buffer (pH 7.4) at a flow rate of 1 ml min − 1 at 21 • C. Protein elution was monitored by UV absorbance at 280 nm.

Heparin binding assay
Peroxynitrite-modified AN (1 μM) in 50 μL 10 mM sodium phosphate buffer was adsorbed onto 96-well black polystyrene microplates overnight.Plates were then blocked with 1% (w/v) BSA in PBS for 1 h at 21 • C, incubated with 50 μL of 0.004 mg mL − 1 fluorescein-conjugated heparin overnight at 21 • C in the dark, and fluorescence was analyzed in a microplate reader with λ ex 495 nm and λ em 525 nm (9 nm slit width) and three spectra averaged.

Statistical analyses
Statistical analyses were performed using GraphPad Prism (version 9, GraphPad Software, San Diego, CA, USA) by 2-way ANOVA with Tukey's multiple comparison test.Data are representative of at least three independent experiments and presented as mean ± SD, with p < 0.05 (indicated by * or #) or p < 0.01 (indicated by ** or ##) considered as statistically significant and ns indicating an absence of statistical significance.

Peroxynitrite-modified AN modulates the structure of FN in the ECM in an oxidant dose-dependent manner
The influence of peroxynitrite-modified AN (30 μM protein, pretreated with 0, 2-, 20-or a 200-fold molar excess of peroxynitrite) on the structure of FN in the ECM from HCASMC was examined by immunofluorescence microscopy with an antibody that detects total FN (pAb, FN 2413) and an Ab that specifically recognizes the EDA epitope of cellderived FN (mAb, FN 3E2).With both antibodies, a decreased signal from fibrillar FN in the ECM was detected for cells exposed to native AN, and AN treated with moderate doses of peroxynitrite (≤20-fold molar excess), compared to control cells without added AN (Fig. 1).The FN detected under these conditions was mainly cell-associated.For cells incubated with AN exposed to higher levels of peroxynitrite (200-fold molar excess), an increase in fibrillar FN in the ECM was apparent, relative to cells exposed to native AN.The spatial distribution of the FN detected under these conditions was more heterogeneous than the native ECM in the absence of AN, with aggregates detected as overstained spots in the images of FN fibrils from cells exposed to AN preincubated with 200-fold molar excess of peroxynitrite (Fig. 1).Also, the cells appeared to be clustered together as judged by DAPI staining.Consistent with the results from the cell system, the levels of fibrillar FN in the decellularized ECM samples was increased for cells incubated with AN exposed to a 200-fold molar excess of peroxynitrite, relative to the ECM from cells exposed to native AN or AN modified by moderate doses of peroxynitrite (≤20-fold molar excess; Supplementary Fig. S2).

Exposure of AN to peroxynitrite has a moderate influence on its capacity to bind FN
To study the interactions between peroxynitrite-modified AN and fibrillar FN, decellularized ECM laid down by HCASMC was incubated with AN and probed by immunofluorescence microscopy using a mAb against the His-tag present on the AN.As shown in Fig. 2A, native AN colocalizes with fibrillar FN (detected by pAb, FN 2413) in the ECM, and exposure of AN to a 20-fold molar excess of peroxynitrite does not appear to perturb the detection of AN.However, at a higher oxidant level (200-fold molar excess of peroxynitrite) a marked decrease in the signal from AN was observed.The binding of AN to FN in the ECM was further examined via ELISA.Exposure of AN to a 200-fold molar excess of peroxynitrite resulted in a reduced signal from AN (detected by the His-tag mAb; Fig. 2B), in agreement with the immunofluorescence microscopy data.The influence of peroxynitrite on the ability of AN to bind to isolated plasma FN coated in 96-well plates was also examined.As displayed in Fig. 2C, AN exposed to a 100-fold molar excess of peroxynitrite showed impaired ability to bind plasma FN, with a significantly reduced signal intensity detected relative to native AN.Modification of AN by peroxynitrite did not influence the recognition of AN by the Histag mAb per se (Supplementary Fig. S3).Together these data indicate that exposure of AN to high molar excesses of peroxynitrite impairs the ability of the protein to bind both fibrillar FN in the ECM, and soluble plasma FN.

Peroxynitrite-treated AN modulates cellular FN expression to a modest extent
In order to examine whether exposure of AN to peroxynitrite influences FN synthesis, HCASMC were cultured in growth medium containing 30 μM of peroxynitrite-modified AN for 48 h, followed by immunoblot detection of FN present in the cell lysates.Incubation of HCASMC with native AN resulted in a decrease in the level of intracellular FN (Fig. 3A and B), in agreement with previous data [44].Similar results were obtained in experiments with AN exposed to moderate (20-fold) molar excesses of peroxynitrite.For cells treated with AN exposed to the 200-fold molar excess of peroxynitrite, the FN levels were higher, but not significantly different from the levels detected in cells exposed to native AN (Fig. 3A and B).At the transcript level, exposure of cells to AN resulted in only a small reduction in mRNA expression of the FN1 gene, with no significant difference detected between the samples exposed to native AN, and AN exposed to a 200-fold molar excess of peroxynitrite (Fig. 3C).This indicates that peroxynitrite-treated AN does not have a strong influence on FN synthesis at the transcriptional or translational level.

Peroxynitrite-treated AN induces modest alterations in cellular metabolic activity and viability
The influence of peroxynitrite-modified AN (30 μM) on cell viability was examined by quantifying LDH release from HCASMC after 48 h incubation.No significant differences in LDH activity release were observed between cells exposed to native or peroxynitrite-modified AN, relative to the control without AN (Fig. 4A).A moderate increase in metabolic activity, as indicated by the MTS assay, was observed in cells incubated with native AN, or AN exposed to a 20-fold molar excess of peroxynitrite (Fig. 4B).In contrast, a significant decrease in metabolic activity was observed for cells exposed to AN pre-treated with a 200-fold molar excess of peroxynitrite, relative to native AN.
The mRNA expression of the mitosis-related genes PCNA and CCNB1 (encoding proliferating cell nuclear antigen and cyclin B1, respectively) were examined for cells treated with peroxynitrite -modified, or native AN.For PCNA, a decreased level of mRNA expression was detected in cells exposed to 20-fold molar excess of peroxynitrite, relative to the control without AN, but no significant difference was detected between native and peroxynitrite-modified AN (Fig. 4C).With CCNB1, the transcript levels detected on exposure to AN pre-treated with peroxynitrite were comparable to those of the control samples without added AN (Fig. 4D).

Exposure of AN to peroxynitrite influences AN-mediated alterations in FN cell adhesiveness
The cell adhesiveness of isolated plasma FN co-incubated with either native or peroxynitrite-modified AN, and then adsorbed onto 96-well plates was determined using HCASMC pre-loaded with the fluorescent dye calcein-AM.A significant decrease in the number of adherent dyeloaded cells was observed for FN co-incubated with AN exposed to ≥20-fold excess of peroxynitrite compared to native AN (Fig. 5A).When AN was added to FN pre-adsorbed on the plate, a decrease in cell adhesion was detected in the experiments using AN exposed to a 2-fold molar excess of peroxynitrite, relative to native AN and control without AN (Fig. 5B).The adhesiveness of cells to AN-coated plates in the absence of FN was also tested.A significant increase in cell adherence was detected for AN exposed to a 200-fold excess of peroxynitrite, relative to the native AN.However, the absolute number of adherent cells in the wells coated with AN was much lower than for plates coated with FN (Fig. 5C).

Peroxynitrite-modified AN displays altered heparin binding properties
The influence of peroxynitrite on the ability of AN to bind heparin was determined by affinity chromatography.Two peaks were observed when native AN was eluted from a heparin column (Fig. 6A), in agreement with previous data [44].Exposure of the AN to a 2-fold molar excess of peroxynitrite resulted in a partial merging of the two peaks and a moderate shift in the elution profiles.With exposure to a 200-fold molar excess of peroxynitrite, a large shift in retention time was detected, consistent with impaired binding of the modified protein to heparin.In agreement with these observations, a dose-dependent decrease in binding of fluorescently-labeled heparin to AN samples was observed with peroxynitrite-modified AN (Fig. 6B).

Discussion
The data presented here indicates that the oxidant/protein ratio has a strong influence on the functional properties of AN exposed to peroxynitrite.In agreement with previous findings [44], decreased levels of FN fibrils were detected in the ECM laid down by cells exposed to native (i.e.non-modified) AN.A similar effect was detected for AN exposed to moderate concentrations of peroxynitrite (≤20-fold excess peroxynitrite), but with AN exposed to a 200-fold molar excess of peroxynitrite this decrease was attenuated (Fig. 7).Under these conditions, impaired binding of AN to FN in the ECM of HCASMC was detected by immunofluorescence microscopy and ELISA.These functional changes may be a consequence of the unfolding and aggregation of AN that occurs at high oxidant doses (≥100-fold excess, as determined by far-UV circular dichroism and small-angle X-ray scattering [24]).Steady state concentrations in the low millimolar range corresponding to these high oxidant doses are unlikely under physiological conditions but may reflect the cumulative amount of oxidant that a protein is exposed to over long time periods, especially with extracellular matrix proteins that can have half-lives of many months or years.At lower oxidant doses (1-100-fold), a significant loss of Tyr and Trp, and formation of 3-nitrotyrosine and 6-nitrotryptophan was detected, although the overall structure and beta-sheet fold of AN was maintained [24].It is therefore concluded that the modification of the Tyr and Trp residues in AN that occurs at moderate peroxynitrite levels, does not have a major influence on the ability of AN to modulate the structure of FN in the ECM of HCASMC.In contrast, mutation of one of the four Tyr residues in AN to an alanine (L37AY40A) appear to have a significant influence on the ECM structure, with increased levels of FN fibrils observed in HCASMC exposed to the L37AY40A mutant relative to the wild-type AN [44].
A characteristic feature of AN is that it can increase the cell adhesiveness of FN [14].Exposure of FN to AN pre-incubated with moderate doses of peroxynitrite resulted in decreased cell adhesiveness relative to FN exposed to native AN.It can be hypothesized that peroxynitrite-modified AN binds or blocks access, to cell adhesion sites on FN, thereby decreasing the ability of cells to adhere to FN.In contrast, with AN exposed to peroxynitrite in the absence of FN, a dose-dependent increase in cell adhesiveness was observed, although the number of adherent cells was much lower than what was observed in the presence of FN.This may be due to the major chemical and structural rearrangements generated on AN by moderate-or high molar excesses of peroxynitrite, and/or the formation of cross-linked high-molecular-mass AN aggregates [24].We have demonstrated previously that exposure of AN to moderate doses of peroxynitrite does not influence the ability of AN to polymerize FN [24].It is therefore not possible to conclude that there is a direct relationship between AN-induced modulation of cell adhesiveness and the FN polymerization activity of AN.
The data presented here also indicate that exposure of AN to peroxynitrite influences its chromatographic heparin binding profile, and a significant decrease in heparin binding was observed at moderate oxidant doses (20-fold molar excess).FN contains a heparin-binding RWRPK motif located within the AN-coding region of FNIII 1 [12].This cluster of positively-charged residues is buried inside the protein in its native (relaxed) conformation, but becomes exposed when FN is subjected to the strain induced by cell contractility, with consequent binding to cell-surface proteoglycans such as syndecans [13].These interactions regulate cellular activities such as contraction and cytoskeletal realignment in response to shear stress, as well as FN fibrillogenesis [45,46].In AN, which mimics the strained and partially unfolded FNIII 1 module, this motif is exposed on the protein surface and is accessible to oxidants such as peroxynitrite [47].Peroxynitrite is unlikely to modify the (positively-charged) lysine and arginine residues of this motif, as these amino acids are not major targets of this oxidant [28].However, previous studies have indicated that the Trp19 residue present in this motif is a target of peroxynitrite, with nitrated Trp products observed with equimolar or greater excesses of peroxynitrite [24].Furthermore, a nearby Tyr residue (Tyr15) present on the same side of the beta-sheet structure that contains this motif, has also been detected in a modified form, and it is proposed that changes to these aromatic residues may modulate heparin binding to this motif.The presence of electron-rich nitro groups on Trp19 and Tyr15 would be expected to generate adverse electrostatic interactions with the negative-charged sulfate groups present on heparin.Furthermore, the powerful electron withdrawing effect of the nitro group is known to reduce the pK a of the phenolic group on the Tyr residue (from 10.1 to 7.2-6.8[48,49]) such that at physiological pH, a significant proportion of the 3-nitrotyrosine will be present in its negatively-charged, ionized form.This may contribute to electrostatic repulsive forces with heparin.Diminished heparin binding to AN may also arise from gross changes to the AN structure, or formation of cross-links and aggregates, which result in occlusion of the lysine and arginine residues within the protein structure, thereby making them inaccessible for macromolecular interactions.
In conclusion, this study demonstrates, for the first time, that modifications to AN induced by peroxynitrite influence ECM matrix deposition by cells exposed to AN (Fig. 7).These observations may have pathological implications since alterations in FN processing and deposition have been associated with several pathologies, including atherosclerosis [50,51].The present work demonstrates that peroxynitrite impacts on the ability of AN to interact with the ECM, and modulates its structure.These changes may impact on the capacity to form FN fibrils and thereby drive the assembly of correctly-assembled ECM structures in tissues exposed to oxidants such as peroxynitrite.

Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: MJD declares commercial consultancy contracts with Novo Nordisk A/S.This funder had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish these results.The other authors declare no conflicts of interest with regard to the data presented.

Fig. 1 .
Fig.1.Peroxynitrite-modified AN influences the structure of FN from HCASMC.Primary HCASMC were cultured overnight in an 8-well glass chamber slide at a density of 2.25 × 10 4 cells per well and then incubated in growth medium containing 30 μM peroxynitrite-modified AN for 48 h.The cells were then fixed and permeabilized followed by incubation with primary antibodies.The total matrix FN was visualized by anti-fibronectin (FN 2413, green fluorescence) and cellular FN was detected by an antibody against an EDA epitope (FN 3E2, red fluorescence), nuclei were counterstained using DAPI (blue).Scale bars: 20 μm.

Fig. 2 .
Fig. 2. Exposure to peroxynitrite perturbs binding of AN to both HCASMC-derived and plasma FN.Panel A-B: Primary HCASMC at an initial density of 2.25 × 10 4 cells per well, were cultured for 1 week in growth medium to lay down ECM.The cells were then removed using 1% (w/v) sodium deoxycholate, with the remaining ECM then incubated with either unmodified or peroxynitrite-modified AN (0.1 μM) overnight.Panel A: immunofluorescence staining of FN and AN with fibronectin pAb (FN 2413, green) and a His-tag mAb (18184 mAb, red), respectively.Scale bars: 100 μm.Panel B: ELISA detection of AN bound to ECM from HCASMC (see panel A above).Panel C: ELISA detection of AN bound to isolated plasma fibronectin (20 μg mL − 1 ) in 10 mM phosphate buffer pre-adsorbed in a 96-well plate using a His-tag antibody (18184 mAb) as described in the Materials and methods.The data were analyzed by 2-way ANOVA with Tukey's multiple comparison test and presented as means ± SD from three experiments, expressed as a % of values detected with native AN (not exposed to peroxynitrite).* and ** indicate a significant (p < 0.05 and 0.01) decrease compared to native AN.

Fig. 3 .
Fig. 3. Peroxynitrite-treated AN modulates FN expression to a modest extent.Panel A-B: HCASMC were cultured in 6-well plates at an initial density of 2 × 10 5 cells per well, in growth medium overnight, and then incubated with 30 μM peroxynitrite-modified AN for 48 h in growth medium.Panel A: Immunoblotting of proteins extracted with 1% (w/v) sodium deoxycholate using primary antibodies against fibronectin (FN 2413 pAb) or β-actin.Panel B: densitometric analysis of the fibronectin bands from Panel A normalized to β-actin.Data are expressed as the % compared to the control C (cells without AN added; set to 100%).Panel C: mRNA expression of FN gene in HCASMC grown at an initial density of 10 5 cells per well overnight and then incubated with 30 μM peroxynitrite-modified AN in growth medium in 12-well plates for 2 h.The data are expressed as a fold-change relative to the cells grown without AN after normalization to the housekeeping genes, B2M and 18S rRNA.* and ** indicate statistical significance (p < 0.05 and 0.01) compared to control group by 2-way ANOVA with a Tukey's multiple comparison test, "ns" indicates no statistical significance.

Fig. 4 .
Fig. 4. Peroxynitrite-treated AN induces modest alterations in cellular metabolic activity and viability.(A) HCASMC (10 5 cells per well) were cultured overnight and then incubated in growth medium containing 30 μM peroxynitrite-modified AN for 48 h with viability assessed by measuring release of LDH (LDH activity in the media as a % of the total LDH activity).(B) HCASMC (0.5 × 10 4 cells per well) were incubated in growth medium with 30 μM peroxynitrite-modified AN for 48 h, with metabolic activity then assessed using the MTS assay.(C-D) HCASMC (10 5 cells per well) were incubated in growth medium containing peroxynitrite-modified AN (30 μM) for 2 h with qPCR to determine the mRNA expression of mitosis-related genes encoding for PCNA (C) and CCNB1 (D).The resulting gene expression data are expressed as a fold-change relative to the control cells without AN after normalization to the housekeeping genes, B2M and 18S rRNA.Data were analyzed by 2-way ANOVA with Tukey's multiple comparison test and are presented as means ± SD from three independent experiments.* and ** indicate a significant (p < 0.05 and 0.01) difference to the control with no AN.# and ## indicate a significant (p < 0.05 and 0.01) difference to the native AN (incubated in the absence of peroxynitrite)."ns" indicates no statistical significance.

Fig. 5 .
Fig. 5. Peroxynitrite-modified AN influences FN-mediated cell adhesiveness.A: peroxynitrite-modified AN (0.25 μM) was mixed with FN (1 μg mL − 1 ) overnight at 21 • C, and subsequently adsorbed onto 96-well plates.B: peroxynitrite-modified AN (0.25 μM) was added to plasma FN (1 μg mL − 1 ) preadsorbed on a 96-well plate, with subsequent incubation overnight.C: peroxynitrite-modified AN (0.25 μM) was adsorbed on a 96-well plate overnight (in the absence of FN).HCASMCs (0.5 × 10 4 cells per well) pre-loaded with calcein-AM were added to the coated plates and adherent cells were quantified by fluorescence in a microplate reader with λ ex 490 nm and λ em 520 nm.In all experiments controls with FN (1 μg mL − 1 ) in the absence of AN were also included.Data are presented as means ± SD from three independent experiments and analyzed by 2-way ANOVA with Tukey's multiple comparison test.* and ** indicate significant (p < 0.05 and 0.01) compared to the control with FN (no AN).# indicates a significant (p < 0.05) compared to the native AN (incubated in the absence of peroxynitrite)."ns" indicates no statistical significance.

Fig. 6 .
Fig. 6.Exposure to peroxynitrite modulates heparin affinity of AN.Panel A: Separation of peroxynitrite-modified AN (80 μM) on a HiTrap Heparin HP column.Samples were applied in 20 mM sodium phosphate buffer (pH 7.4) and eluted with a gradient of 0-500 mM NaCl in 20 mM sodium phosphate buffer at a flow rate of 1 mL min − 1 .Panel B: native and peroxynitrite-modified AN (1 μM) were coated onto plates overnight at 21 • C. Subsequently, 4 μg mL − 1 fluorescein-conjugated heparin was added and incubated overnight at 21 • C in the dark, then analyzed using a microplate reader with λ ex 485 nm and λ em 520 nm.Data are presented as means ± SD from three experiments and analyzed by 2-way ANOVA with Tukey's multiple comparison test.** indicate a significant (p < 0.01) decrease compared to native AN (incubated in the absence of peroxynitrite).

Fig. 7 .
Fig. 7. Schematic overview outlining the influence of native and peroxynitrite-modified AN on the structure of FN fibrils in the ECM generated by vascular smooth muscle cells.