Biochimica et Biophysica Acta 1852 (2015) 1334–1346

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbadis

The flavo-oxidase QSOX1 supports vascular smooth muscle cell migration and proliferation: Evidence for a role in neointima growth Beatriz E. Borges a,1, Márcia H. Appel a,1,2, Axel R. Cofré a,1, Maiara L. Prado a, Chelin A. Steclan a, Frédéric Esnard b, Silvio M. Zanata a, Francisco R.M. Laurindo c, Lia S. Nakao a,⁎ a b c

Departamento de Patologia Básica, Universidade Federal do Paraná, Centro Politécnico, Curitiba 81531-980, Brazil INSERM UMR 1100, Université François Rabelais F-37032 Tours, France Laboratório de Biologia Vascular, Heart Institute, Universidade de São Paulo, São Paulo 05403-900, Brazil

a r t i c l e

i n f o

Article history: Received 11 October 2014 Received in revised form 13 February 2015 Accepted 4 March 2015 Available online 10 March 2015 Keywords: Quiescin sulfhydryl oxidase Cell proliferation Cell migration Vascular smooth muscle cell Neointima

a b s t r a c t Quiescin sulfhydryl oxidase 1 (QSOX1) is a flavoenzyme largely present in the extracellular milieu whose physiological functions and substrates are not known. QSOX1 has been implicated in the regulation of tumor cell survival, proliferation and migration, in addition to extracellular matrix (ECM) remodeling. However, data regarding other pathophysiological conditions are still lacking. Arterial injury by balloon catheter is an established model of post-angioplasty restenosis. This technique induces neointima formation due to migration and proliferation of vascular smooth muscle cells (VSMC), followed by ECM synthesis and remodeling. Here, we show that QSOX1 knockdown inhibited VSMC migration and proliferation in vitro. In contrast, QSOX1 overexpression stimulated these processes. While migration could be induced by the incubation of cells with the active recombinant QSOX1, proliferation was induced by addition of the active and also of an inactive mutant QSOX1 protein. The proliferation induced by both recombinants was independent of intracellular hydrogen peroxide and dependent of the MEK/ERK pathway. To recapitulate in vivo VSMC pathophysiology, balloon-induced arterial injury was performed. The expression of QSOX1 in the neointimal layer of balloon-injured rat carotids was high and peaked at 14 days post-injury. In vivo QSOX1 knockdown led to a significant decrease in PCNA expression at day 14 post-injury and a decreased intima/media area ratio at day 21 post-injury, compared with scrambled siRNA transfection. In summary, our findings demonstrate that QSOX1 induces VSMC migration and proliferation in vitro and contributes to neointima thickening in balloon-injured rat carotids. © 2015 Published by Elsevier B.V.

1. Introduction Quiescin sulfhydryl oxidase 1 (QSOX1) is a flavin-linked sulfhydryl oxidase that localizes to the ER/Golgi apparatus [1,2] or the extracellular milieu [1,3–10]. QSOX1 exists in two isoforms that are produced by alternative splicing [11]. The short QSOX1, the most abundantly expressed isoform [12], is secreted [5,8,13], while the long isoform is a transmembrane protein [14] that was recently demonstrated to be Abbreviations: ALR, augmenter of liver regeneration; bFGF, basic fibroblast growth factor; DAB, 3,3′-diaminobenzidine; ECM, extracellular matrix; ERK1/2, extracellular signal-regulated kinase; HRP, horseradish peroxidase; IgG, immunoglobulin G; MTT, 3(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; PCNA, proliferating cell nuclear antigen; PDGF, platelet-derived growth factor; QSOX1, quiescin sulfhydryl oxidase; siRNA, small interfering RNA; VSMC, vascular smooth muscle cell ⁎ Corresponding author at: Departamento de Patologia Básica, Room 153, Setor de Ciências Biológicas, Centro Politécnico, Universidade Federal do Paraná, Curitiba 81531-980, Brazil. Tel.: +55 41 3361 1760; fax: +55 41 3266 2042. E-mail address: [email protected] (L.S. Nakao). 1 BEB, MHA and ARC contributed equally to this work. 2 Present address: Departamento de Biologia Estrutural, Molecular e Genética, Universidade Estadual de Ponta Grossa, Ponta Grossa, Brazil.

http://dx.doi.org/10.1016/j.bbadis.2015.03.002 0925-4439/© 2015 Published by Elsevier B.V.

proteolytically cleaved to a soluble and secreted variant [15]. Due to the extracellular localization of QSOX1, roles related to extracellular matrix (ECM) remodeling or extracellular protein assembly have been attributed to this protein [12]. Indeed, the requirement for QSOX1 activity to incorporate laminin trimer into the ECM, providing a functional surface, was shown [16]. In addition, extracellular QSOX1 has been associated with tumor progression [17–21] due to the activation of cellular proliferation and matrix metalloproteinases (MMP) [20,22]. In contrast, previous studies have demonstrated that QSOX1 negatively regulates cell cycle progression [4,23–25]. However, despite such findings, data regarding the physiological substrates and roles of QSOX1 remain scarce. Recently, a role for QSOX1 in the cardiovascular system has emerged. It has been found in the supernatants of aortic smooth muscle cells (SMC) and smooth muscle progenitor cells with proteomic approaches [9] and in the necrotic areas of atherosclerotic plaques [26], and it has been described as a possible marker of heart failure [27]. Additionally, protein disulfide isomerase (PDI), which cooperates with QSOX1 during protein folding and isomerization [28], regulates the activity of vascular NADPH oxidase [29], which is a key element of

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

vascular repair after balloon injury [30,31]. Arterial injury by balloon catheter induces neointima formation followed by constrictive remodeling, as primarily determined in experimental models of restenosis [32,33]. It is widely accepted that medial SMC respond to injury, immediately proliferating and migrating to the intima. There, neointimal cells continue proliferating, reaching a maximum rate between 4 and 7 days post-injury [34]. The resulting neointima prominently appears at day 14, after which ECM remodeling takes place [35], while proliferation and migration cease [33,36]. Because neointima thickening is mediated by cell migration, cell proliferation and ECM synthesis/remodeling, we postulated that QSOX1 might have a role in this process. Our findings demonstrate that QSOX1 induces vascular smooth muscle cell (VSMC) migration and proliferation in vitro and contributes to neointima thickening in balloon-injured rat carotids.

1335

megaprimer with a site-directed mutation using the forward primer 5′-CTTCTTTGGCAGTCGTGACTGTGC-3′ (nucleotide substitution underlined) and the reverse primer 5′-ATCGTCATAGACTCTTCTTGA AAGCTTGGG-3′ (HindIII site underlined). The second PCR round was performed with the purified megaprimer (306 bp) as the reverse primer and 5′-GGGGTACCTACTCGTCCTCTGAC-3′as the forward primer (KpnI site underlined). The mutant amplicon was doubled-digested with KpnI and HindIII, and the purified insert was cloned into the pET32a vector (Novagen). Protein concentrations were determined using ε456nm = 12.5 mM− 1 × cm− 1 [43] or Bradford assay. Dithiothreitol oxidase activity was determined as previously described [44]. While the wild type QSOX1 produced an average activity of 123 nmol H2O2/min/mg protein, the mutant protein was completely inactive, as measured before each assay. For some experiments, LPS was removed from the recombinant proteins with Detoxi-Gel® Endotoxin Removing Gel (Thermo Scientific).

2. Material and methods 2.4. Migration assays 2.1. VSMC culture Aortas were obtained from male Wistar rats weighing 250 g. Arteries were dissected and rinsed with PBS containing antibiotics and fungicide. The media layer was then cut into 1-mm2 fragments. These explants were plated with their luminal surfaces facing upward in a 35-mm culture dish and covered with 0.5 mL high glucose DMEM (Life Technologies) containing 20% fetal calf serum (Life Technologies) and 40 mg/L Garamycin® (Schering-Plough). After migration, VSMC were detached with trypsin and subcultured in DMEM containing 10% fetal calf serum and antibiotics. VSMC identity was confirmed by immunofluorescence with anti-smooth muscle actin (Sigma-Aldrich). Cells in passages 3–7 were assayed [37]. This protocol was approved by the Research Ethics Committee (CEUA) of the Biological Sciences Building, protocol number 412/2009. 2.2. QSOX1 knockdown and overexpression in VSMC QSOX1-specific siRNA (targeting both QSOX1 transcripts) and scrambled siRNA duplexes (Stealth RNAi™, Invitrogen) were designed with Block-iT RNAi Designer. Sense sequences corresponding to the QSOX1 and scrambled duplexes were 5′-UGGAGCCUGCCAAGCUGAAG GAUAU-3′ and 5′-UGGCCGUCCAACGGUAAGGAGUAU-3′, respectively. QSOX1 knockdown was achieved by transfecting VSMC with 50 nM QSOX1 siRNA or scrambled siRNA duplexes using Oligofectamine (Invitrogen). Cells (10–25 × 10 3 cells/cm2 ) were transfected in OptiMEM (Life Technologies) for 4 h, after which the transfection medium was replaced by complete medium, unless otherwise stated. For QSOX1 overexpression, VSMC (106) were resuspended in 0.4 mL OptiMEM and electroporated with 10 μg of QSOX1-pCR3.1 [5], corresponding to the short secreted isoform, or empty pCR3.1 (mock) plasmid in a Gene Pulser XCell electroporation device (BioRad). The electroporation settings were: 300 V, 500 μF and a resistance of infinity at exponential decay. Cells were kept at room temperature for 30 min and then plated to recover. Twenty-four hours later, they were seeded at a specific density for each assay [38].

The wound migration assay was performed as previously described [45]. Briefly, 105 cells were plated in 6-well plates. After the transfections, a P-200 pipette tip scraped the dish surface to generate a “wound”. Cells were then cultivated in DMEM with 0.1% fetal calf serum. After 24, 48 and 72 h, cells were gently rinsed with PBS and photographed. Migration was calculated as migration distance. For this purpose, the difference in the distance between the edges of a scratch at the indicated times in relation to the distance measured at time 0 was calculated. In some assays, mQSOX1 or mQSOX1C452S recombinant proteins (50 nM) were added immediately after transfection with siRNA for the indicated times, in DMEM containing 0.1% fetal calf serum. Cell migration was also assessed using a transwell system with 8-μm pores (Corning). VSMC (5 × 105) were seeded in the upper chamber in DMEM containing 0.1% fetal calf serum. DMEM containing 20 ng/mL PDGF-BB (Peprotech) or 10% fetal calf serum was added into the lower chamber at the indicated times to stimulate migration. After 6 h, cells on the upper surface were removed by gentle wiping with a cotton swab, while migrating cells were counted in a hemocytometer [46]. 2.5. Proliferation assay Cells (5 × 104) were plated in 96-well microplates (TPP). Twentyfour, 48 and 72 h after transfection with siRNA or plasmids, cells were collected and analyzed by crystal violet staining as described [47]. In some assays, mQSOX1 or mQSOX1C452S recombinant proteins (5 or 50 nM), in the absence or presence of polyethyleneglycol (PEG)catalase (200 U/mL, Sigma-Aldrich) [48] or MEK1/2 inhibitor U0126 (10 μM, Calbiochem), were added immediately after transfections with siRNA in the culture medium containing 10 or 0.1% serum, for the indicated times. Serum starvation (0.1% fetal calf serum for 24 h) was performed in some experiments, as indicated in the legends. 2.6. Cell viability Cell viability was assessed with an MTT assay, as previously described [47].

2.3. Recombinant mouse QSOX1 production and activity 2.7. Quantitative PCR Recombinant wild type short isoform QSOX1 (herein referred to as mQSOX1) was produced as previously described [39]. Mouse and rat QSOX1 are 91% identical in their amino acid sequences, and a recent study showed that metazoan QSOX share domain organization and enzymatic mechanism [40], enabling the use of mouse proteins with rat cells. The mutant inactive enzyme [41] (herein referred to as mQSOX1C452S) was obtained by mutating the first cysteine residue of the second CXXC motif of the wild type construct using Megaprimer PCR method [42]. A first PCR round was performed to produce a

Total RNA was extracted from 105 VSMC using the Invitrap Spin Cell RNA Mini Kit (Invitek) according to the manufacturer's protocol. Complementary DNA was synthesized from 50 ng total RNA, after treatment of RNA with DNAse. Relative mRNA expression levels were determined using hypoxanthine-guanine phosphoribosyltransferase (HPRT) as a housekeeping gene and the 2−ΔΔCTmethod. Reactions were performed in a Rotor-Gene 6000 thermocycler (Corbett) using a SYBR Green PCR Kit (Qiagen). The QSOX1 primers (which detected both QSOX1

1336

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

isoforms) were: forward 5′-TGCATTCCATAAACGATTGG-3′ and reverse 5′-GAAGTGGAAGAGGACCCACA-3′. The HPRT primers were: forward 5′-CAGGCCAGACTTTGTTGGAT-3′ and reverse 5′-TCCACTTTCGCTGATG ACAC-3′. All primers were validated with the 2−ΔΔCT method [49].

with 100 μg/mL pyronin Y for 10 min [50,51]. Fluorescence (λexc = 350 nm, λem = 461 nm and λexc = 488 nm, λem = 550 nm for Hoechst 33342 and pyronin Y, respectively) was measured in a microplate reader (Tecan Infinity), and the relative RNA quantity was calculated with the pyronin Y/Hoechst 33342 fluorescence ratio.

2.8. SDS-PAGE and immunoblotting 2.10. Brefeldin A assay Cells (5 × 105) were plated in 100-mm dishes (TPP). At 24, 48 and 72 h after transfection, cells were lysed (10 mM Tris–HCl pH 7.4 containing 5 mM EDTA, 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 1% deoxycholate, a protease inhibitor cocktail (Roche) and phosphatase inhibitors for 15 min at 4 °C). After centrifugation (10,000 × g, 10 min), supernatants were collected, and the protein contents were determined with Bradford reagent (BioRad). Proteins were separated on a 10% SDSPAGE gel and electroblotted on to a nitrocellulose membrane. Membranes were blocked with 5% skim milk. The antibodies used were rabbit polyclonal anti-PCNA (1:200, Dako), mouse monoclonal anti β-actin (1:1000, Sigma), mouse anti-phospho ERK1/2 (1:1000, Santa Cruz), mouse monoclonal anti-calponin (1:1000, Sigma), rabbit anti-ERK1/2 (1:1000, Cell Signaling), anti-rabbit IgG coupled to HRP (1:1000, Sigma-Aldrich) and anti-mouse IgG coupled to HRP (1:5000, Sigma-Aldrich). Reactions were developed with a West Pico or Femto chemiluminescence kit (Pierce) and bands were detected on autoradiography films (Thermo). 2.9. Determination of quiescence by fluorescence analyses Cells (104) were harvested, washed twice with PBS and stained with 5 μg/mL Hoechst 33342 solution for 45 min at 37 °C and then

Cells (5 × 105) were plated in 100-mm dishes (TPP), transfected with siRNA duplexes and treated with 2.5 μg/mL brefeldin A (BFA) for 24, 48 and 72 h [52]. Cells were washed and analyzed by crystal violet assay. 2.11. Balloon injury in rat carotid artery Male Wistar rats weighing between 250 and 300 g were obtained from the facility at the Universidade Federal do Paraná (UFPR). Animals were anesthetized with an intraperitoneal injection of 100 mg/kg ketamine hydrochloride (Syntec) and 5 mg/kg xylazine hydrochloride (Syntec). Endothelial denudation and intimal lesion of the left carotid artery were performed with a balloon embolectomy catheter (2F Fogarty, Edwards Life Sciences) as previously described [53]. Briefly, the catheter was inserted through the external carotid artery to the common carotid artery, inflated with 0.2 mL saline and drawn with a rotary motion. After deflation in the common carotid, it was removed. This procedure was performed three times. Some animals were anesthetized and underwent the procedure without the use of the balloon catheter (sham-operated). Following balloon angioplasty, animals were kept warm, and 0.1 mg/kg scopolamine butylbromide plus 5 mg/kg dipyrone

Fig. 1. Effects of QSOX1 knockdown and overexpression on VSMC. QSOX1 mRNA expression in scrambled or QSOX1 siRNA-transfected (A) and in mock or pCR3.1QSOX1-transfected (B) VSMC were evaluated at 24, 48 and 72 h post-transfection. mRNA levels were quantified by quantitative PCR using the 2−ΔΔCt method. Cell viability was assessed by an MTT assay in scrambled and QSOX1 siRNA-transfected (C) or in mock and pCR3.1QSOX1-transfected (D) cells at 24, 48 and 72 h after transfection. In all experiments, cells were cultured in complete medium. The results correspond to the mean ± SD of 3 (A, B) or 4 (C, D) independent experiments. *p b 0.05.

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

(Boehringer Ingelheim) was administered. This protocol was approved by the Research Ethics Committee (CEUA) of the Biological Sciences Building, protocol number 412/2009.

1337

Z.2 slide scanner microscope (Zeiss, Germany) controlled by Metafer 4-VSlide software [55,56]. 2.14. Statistical analysis

2.12. siRNA in vivo transfection A 10 μg aliquot of siRNA duplex (sequences above) was dissolved in 100 μL of a 30% (m/v) pluronic gel (Sigma-Aldrich) solution and perivascularly delivered into the rat carotid artery immediately after injury [54]. The transfection controls were performed with 10 μg of a control siRNA labeled with TyE (IDT) or 10 μg of pEGFP-C1 (Clontech).

Results are shown as the mean ± S.D. The exact n (number of independent experiments with cells or number of rats) is described in each figure legend. Statistical analyses were performed with ANOVA with Tukey post hoc test, using GraphPad Prism 5 or 6 softwares. For all analyses, p b 0.05 was considered statistically significant. 3. Results

2.13. Immunohistochemistry (IHC) and histochemistry Sham-operated, injured and injured/transfected carotids, at 7, 14 and 21 days after balloon catheter injury, as well as carotids from control (uninjured) rats were collected. Next, 5-μm-thick sections of each specimen were subjected to immunohistochemistry at the same time as previously described [39] with the following polyclonal antibodies: anti-QSOX1 [8], anti-GFP (a donation of Dr. Stenio Perdigão, ICC, Curitiba, Brazil) and anti-PCNA (Santa Cruz Biotechnology). Immunoreactions were developed with a peroxidase mouse and rabbit kit (Diagnostic BioSystems) using 3,3′-diaminobenzidine (DAB) as a chromogen (Diagnostic BioSystems). QSOX1 expression was quantified by densitometry in 6 random fields of the same section per animal using Image J software. For histochemistry assays, tissue sections were stained with Weigert Van Gieson. Images were acquired using an Axio Imager

The role of QSOX1 in VSMC proliferation and migration was studied in primary cultures of rat aortic SMC transfected with QSOX1 siRNA or with the pCR3.1QSOX1 plasmid. The differential QSOX1 expression was demonstrated by RT-qPCR. By comparing QSOX1 mRNA expression with corresponding controls, knockdown of approximately 50% was achieved 24 h after siRNA transfection; the knockdown degree slowly decreased through 48 and 72 h (Fig. 1A). An overexpression of 280% was achieved after 24 h, decreasing to 150% after 48 h (Fig. 1B). We then tested whether QSOX1 knockdown or overexpression would interfere with cell viability, as measured by mitochondrial metabolism. With the exception of the time point 72 h after pCR3.1QSOX1 plasmid transfection, compared with the mock control, no significant alteration was observed (Fig. 1C, D). Overall, this result indicates that modulation of QSOX1 expression is not associated with cell death in VSMC.

Fig. 2. Effects of QSOX1 knockdown and overexpression in VSMC migration. Cell migration was assessed by scratch (A, B) and transwell (C, D) assays in scrambled and QSOX1 siRNAtransfected (A, C) and in mock and pCR3.1QSOX1-transfected (B, D) cells at 24, 48 and 72 h after transfection. Migration was calculated as the difference in the distance between the edges of a scratch at the indicated times in relation to the distance measured at time 0. In the transwell assay, migration was calculated as the number of cells attracted by 10% fetal calf serum or by 20 ng/mL PDGF. During the migration periods, cells were cultured in 0.1% fetal calf serum. Data shown correspond to the mean ± SD of 4 (A, B) or 3 (C, D) independent experiments. *p b 0.05.

1338

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

3.1. Extracellular QSOX1 has pro-migratory properties in VSMC Migration was analyzed with the scratch method and with the transwell assay. QSOX1 siRNA-transfected VSMC exhibited a decreased migration rate compared with scrambled siRNA-transfected cells as determined in the scratch assay, for all times analyzed (Fig. 2A). Accordingly, QSOX1 overexpression led to an increased migration rate compared with the mock control (Fig. 2B). These overexpression data reinforce the idea that siRNA effects do not result from off-target effects because they are complementary to the knockdown results. The hypothesis that we were measuring cell proliferation instead of cell migration was excluded, as VSMC did not proliferate when cultivated with 0.1% fetal calf serum for 72 h as measured in the crystal violet assay (data not shown). Migration through transwells confirmed the scratch results. Here, chemotaxis was performed with 10% fetal serum or 20 ng/mL PDGF, a classical chemoattractant agent for VSMC (Fig. 2C, D). These results show that QSOX1 has a pro-migratory effect on VSMC. The most abundant isoform of QSOX1, the short isoform, is reported to be secreted into the extracellular medium [5,8,13]. In addition, the proteolytic cleavage of the long isoform, which produces a secreted active version of QSOX1, was recently reported [15]. Thus, we asked whether the pro-migrative effect of QSOX1 was due to the secreted isoform, implying an extracellular role. To this end, we incubated scrambled or QSOX1 siRNA-transfected cells with mQSOX1 or mQSOX1C452S (Fig. 3). Fifty nanomolar mQSOX1 induced migration of VSMC in all conditions compared with non-stimulated cells, i.e., VSMC cultivated in the absence of mQSOX1. In contrast, the recombinant inactive enzyme mQSOX1C452S did not induce migration, and even inhibited the migration at 48 h (Fig. 3). This response profile to the recombinant proteins was essentially the same in QSOX1 siRNA-transfected or scrambled siRNA-transfected cells. However, QSOX1 knockdown led to significant decreased migration distances, as observed at 48 and 72 h. Efficiency analysis of both recombinants in promoting cell migration showed that cells migrated more when incubated with the wild type than with the

mutant enzyme (Fig. 3). Taken together, these findings show that extracellular QSOX1 induces VSMC migration and that this process is dependent on the sulfhydryl oxidase activity of the enzyme. This extracellular activity is probably derived from a combination of both the recombinant QSOX1 and the endogenous QSOX1. 3.2. Extracellular QSOX1 has pro-proliferative properties in VSMC Proliferation was assessed with crystal violet staining. QSOX1 siRNA treatment induced a significant decrease in cell proliferation compared with the scrambled control, after 48 and 72 h (Fig. 4A). Consistent with this observation, QSOX1 overexpression produced significant increases in the proliferation of pCR3.1QSOX1-transfected cells (Fig. 4B). Additionally, QSOX1 expression paralleled the expression of PCNA, a marker of cell proliferation (Fig. 4C, D). Previous reports indicated that the expression of QSOX1 was associated with cellular quiescence, but recent studies have shown that QSOX1 knockdown inhibits both proliferation and migration of breast [20] and pancreas [22] tumor cells. Because quiescent cells (in G0 state) have less cellular RNA than cycling cells [50], we determined the relative amount of cellular RNA in the cells. The results indicated that the amount of RNA was lower when QSOX1 expression was inhibited and higher when QSOX1 was overexpressed (Fig. 4E, F). These data indicate that QSOX1 expression is not associated with the G0 state in VSMC. Finally, because of the balance between the proliferative/secreting and quiescent/contractile states of VSMC, we analyzed how QSOX1 expression would affect calponin expression, a marker of VSMC contractile phenotype [57]. The results demonstrated that calponin expression is inversely proportional to QSOX1 expression (Fig. 4C, D), confirming the association between high QSOX1 expression and decreased contractile phenotype. To investigate whether the proliferative effect was due to extracellular QSOX1, we incubated QSOX1 siRNA-transfected VSMC with mQSOX1. While 5 nM mQSOX1 had no effect on QSOX1 siRNA-transfected cells at any time point, 50 nM significantly increased proliferation of QSOX1 knockdown cells at 24, 48 and 72 h after the treatment, compared with

Fig. 3. Effect of exogenous extracellular recombinant QSOX1 in VSMC migration. Scrambled or QSOX1 siRNA-transfected cells were scratched and treated with 50 nM wild type mQSOX1 or the inactive recombinant mQSOX1C452S or were left untreated. After 24, 48 or 72 h of treatment, the migrated distance was measured. During the migration period, cells were cultured, in the absence or presence of the recombinant proteins, in 0.1% fetal calf serum. Data shown correspond to the mean ± SD of 4 independent experiments. *p b 0.05.

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

1339

Fig. 4. Effect of QSOX1 knockdown and overexpression in proliferative events in VSMC. Proliferation was assessed by crystal violet assay (A, B), PCNA and calponin expressions by immunoblotting (C, D) and relative RNA content by pyronin Y (PY)/Hoechst 33342 fluorescence ratios (E, F) in scrambled and QSOX1 siRNA-transfected (A, C, E) or in mock and pCR3.1QSOX1transfected (B, D, F) cells cultivated in complete medium for the indicated times after transfection. The results correspond to the mean ± SD of 4 independent experiments, whereas the immunoblotting data are representative blots of 3 independent assays. *p b 0.05.

both the control (non-stimulated) and the 5 nM treatment, indicating a concentration-dependent mitogenic stimulation (Fig. 5). An identical profile was obtained with scrambled siRNA-transfected VSMC. These results indicate that extracellular QSOX1 induces proliferation. When VSMC were treated with brefeldin A, a drug that blocks protein secretion by disrupting ER-Golgi traffic, proliferation was inhibited as measured with the crystal violet assay (data not shown). Although brefeldin A has multiple targets and even induces cell growth arrest, this may be a result of the blockage of QSOX1 secretion. To investigate whether this extracellular trigger of proliferation by QSOX1 was dependent on the sulfhydryl oxidase activity of the extracellular protein, we incubated scrambled siRNA- and QSOX1 siRNA-transfected VSMC with mQSOX1C452S. The results obtained were very similar to those obtained with mQSOX1. The mutant protein stimulated the proliferation of VSMC, with or without QSOX1 knockdown, significantly at 50 nM (Fig. 5). In contrast with the 5 nM mQSOX1, 5 nM mQSOX1C452S was able to promote the proliferation of both scrambled siRNA- and QSOX1 siRNA-transfected cells at 24 h, suggesting a slightly higher efficiency of 5 nM mutant QSOX1. However, in QSOX1 siRNA-transfected cells, 50 nM wild type was more effective than the mutant enzyme at 24 and 48 h (Fig. 5). Therefore, under our experimental conditions, the proliferative effect of QSOX1 in VSMC is

triggered by extracellular QSOX1 and, overall, does not depend on its sulfhydryl oxidase activity. To exclude a possible role of LPS contamination, this assay was performed with LPS-free recombinant proteins. The incubation of non-transfected VSMC with 5 and 50 nM wild type and mutant QSOX1 for 24, 48 and 72 h provided the same results as those shown in Fig. 5 (data not shown), demonstrating that the enhanced proliferative effect was not due to LPS. Also, a 24 h-starvation with 0.1% fetal calf serum, followed by incubation with the recombinant proteins in complete medium provided identical results as those from non-starved cells (data not shown), indicating that cellular synchronization does not affect VSMC proliferation induced by QSOX1. To minimize the effects of growth factors from the bovine serum, we starved cells (24 h-starvation with 0.1% fetal calf serum) and then incubated with recombinant QSOX1 in the same starvation medium. The results showed that even under low serum, incubation of VSMC with QSOX1 and PDGF led to increased cell proliferation, compared with the non-stimulated control cells (Suppl. Fig. 1), and that both the wild type and the mutant QSOX1 induced cell proliferation, compared with the control cells, in a time- and dosedependent way (Suppl. Fig. 1). Moreover, under such low serum condition, the mutant QSOX1 showed an enhanced effect compared with the wild type.

1340

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

Fig. 5. Effect of exogenous extracellular recombinant QSOX1 in VSMC proliferation. Scrambled or QSOX1 siRNA-transfected cells were treated with 5 or 50 nM wild type mQSOX1 or the inactive recombinant QSOX1 mQSOX1C452S or were left untreated, in DMEM containing 10% fetal calf serum. After 24, 48 or 72 h of treatment, proliferation was assessed by crystal violet assay. Data shown correspond to the mean ± SD of 3 independent experiments. *p b 0.05.

Finally, to investigate whether hydrogen peroxide mediates the proliferative effect of exogenous QSOX1, we treated the VSMC with wild type or mutant QSOX1, in medium containing 0.1% fetal serum and PEG-catalase for 48 h, and measured cell proliferation (Fig. 6). Addition of PEG-catalase to the cells inhibited the proliferation induced by PDGF (Fig. 6). However, it did not affect the proliferation induced by mQSOX1 and mQSOX1C452S, indicating that the mitogenic signaling induced by extracellular QSOX1 does not depend on intracellular hydrogen peroxide.

3.3. MEK1/2–ERK1/2 pathway is involved in QSOX1-induced proliferation of VSMC ERK1/2 are usually activated by mitogens. To assess whether this pathway was also activated by recombinant QSOX1, we performed a proliferation assay in the absence or in the presence of U0126, an inhibitor of MEK1/2, the upstream ERK1/2 kinase. Inhibition of ERK1/2 activation by U0126 inhibited cell proliferation promoted by PDGF and by 50 nM mQSOX1 and mQSOX1C452S proteins (Fig. 7A). In agreement,

Fig. 6. Effect of PEG-catalase in VSMC proliferation induced by recombinant QSOX1. VSMC were incubated with 20 ng/mL PDGF-BB, 50 nM wild type mQSOX1 or the inactive recombinant QSOX1 mQSOX1C452S or were left untreated, in DMEM containing 0.1% fetal calf serum, in the presence or absence of 200 U/mL PEG-catalase. After 48 h, proliferation was assessed by crystal violet assay. Data shown correspond to the mean ± SD of at 4 independent experiments. *p b 0.05; ns, non-significant.

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

1341

Fig. 7. Effect of MEK1/2 inhibitor U0126 in VSMC proliferation and in ERK1/2 phosphorylation. (A) VSMC were serum-starved (0.1%) for 24 h and incubated with 20 ng/mL PDGF-BB, 50 nM wild type mQSOX1 or the inactive recombinant QSOX1 mQSOX1C452S or were left untreated, in DMEM containing 0.1% fetal calf serum, in the presence or absence of 10 μM U0126. After 48 h, proliferation was assessed by crystal violet assay. Data shown correspond to the mean ± SD of 3 independent experiments. Statistically significant differences (*p b 0.05) are indicated. (B) VSMC were incubated with 50 nM wild type mQSOX1 or the inactive recombinant QSOX1 mQSOX1C452S or were left untreated, in DMEM containing 0.1% fetal calf serum, for the indicated times, in the absence or presence of 10 μM U0126. Total protein extracts (50 μg) were electrophoresed and immunoblotted with anti-phospho ERK1/2 and anti-ERK1/2 antibodies.

addition of both 50 nM recombinant QSOX1 triggers a rapid ERK1/2 phosphorylation, which is completely abolished by the presence of U0126 (Fig. 7B). Thus, these findings show that both wild type and the inactive QSOX1 induce VSMC proliferation in a MEK1/2– ERK1/2-dependent pathway.

3.4. Contribution of QSOX1 to neointimal growth in balloon-injured rat carotid To investigate the clinical implications of QSOX1 in VSMC, we first analyzed if QSOX1 was expressed by healthy arteries.

1342

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

Immunohistochemistry results showed that basal levels of QSOX1 in rat carotids are relatively low in the media layer (Supp. Fig. 2). As a pathological condition, we employed arterial injury by balloon catheter, a more complex model of vascular disease. Injury by balloon catheter in rat carotids led to intimal hyperplasia, with a prominent neointimal layer observed 14 days after the lesion (Supp. Fig. 3). We then analyzed QSOX1 expression. The results indicated that the medial layer expressed a constant low level of QSOX1 through 21 days post-injury. In the neointimal layer, QSOX1 expression increased until 14 days after lesion, when it is maximal, after which it decreases (Fig. 8). Therefore, in contrast with QSOX1 present in media, QSOX1 in the neointima might play a role in neointimal growth. Thus, we next asked whether decreased QSOX1 expression would interfere with the development of the neointima. We employed pluronic gel to perform an in vivo transfection of siRNA and knockdown QSOX1 expression in the injured carotid. The efficiency of this method was validated by analyzing the transfection of a fluorophore-labeled siRNA and of GFP expression in the rat carotid 48 h after transfection of a plasmid bearing the GFP gene. The results indicated that siRNA-TyE (data not shown) and the perivascularly-delivered plasmid reached the medial cells (Supp. Fig. 4). Transfection of siRNA specific for the QSOX1 gene immediately after the balloon injury induced decreased QSOX1 protein expression by approximately 50% at days 14 and 21 compared with the scrambled siRNA (Fig. 9). Thus, pluronic gel-mediated siRNA transfection provided sustained QSOX1 knockdown in vivo, as previously described [54]. Finally, analysis of the neointima/media area ratio demonstrated that this ratio was significantly decreased by ~30% after QSOX1 knockdown in injured arteries at day 21 compared with the scrambled control (Fig. 10). As neointima growth is a result of VSMC migration, proliferation and matrix secretion, we assessed whether decreased expression of QSOX1 inhibited cell proliferation. Indeed, the neointimal expression of PCNA was highly inhibited (70% compared with the scrambled control)

Fig. 9. QSOX1 expression in neointima after in vivo siRNA transfection in rat carotids. Immediately after balloon catheter injury, 10 μg of scrambled (n = 15 rats/time point) or QSOX1 (n = 15 rats/time point) siRNA were delivered to the injured artery. At days 7, 14 and 21 after the injury, carotid sections were immunostained with anti-QSOX1, and QSOX1 expression was quantified by densitometry in 6 random fields/rat. The values represent the mean ± SD. *p b 0.05.

in the nuclei of QSOX1 knockdown carotids at day 14. Interestingly, at this time, PCNA expression was maximal in the scrambled control arteries, which indicated that scrambled siRNA-transfected neointimal cells were actively dividing on day 14 (Fig. 11). Collectively, these data demonstrate that QSOX1 contributes to neointima thickening after balloon injury in rat carotids.

Fig. 8. QSOX1 expression and quantification in rat carotids after balloon catheter injury. (A) Representative photomicrographs of catheter-injured (15 rats/ time point) carotid sections immunostained with anti-QSOX1 at 7, 14 and 21 days after the injury. The carotid sections were photographed at 400× magnification. QSOX1 expression was determined in neointima (B) and media (C) layers by densitometry in 6 random fields/rat. The values represent the mean ± SD. *p b 0.05.

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

1343

Fig. 10. Effect of QSOX1 knockdown on the neointima/media area ratio. (A) Representative photomicrographs of scrambled (n = 15 rats/time point) or QSOX1 (n = 15 rats/time point) siRNA-transfected injured carotid sections, stained with Weigert Van Gieson solution at 7, 14 and 21 days after lesion. The carotid sections were photographed at 400 x magnification. Line indicates tunica media while dot indicates tunica intima or neointima. (B) Areas were measured with Image J software, and neointima/media area ratios were calculated. The values represent the mean ± SD. *p b 0.05.

4. Discussion To investigate the role of QSOX1 in VSMC, we first modulated its expression by QSOX1 siRNA or QSOX1 plasmid. Our in vitro results clearly demonstrate that QSOX1 knockdown inhibits VSMC migration, while QSOX1 overexpression enhances it. These results are in agreement with recent findings [20,22] demonstrating that QSOX1 knockdown inhibits tumor cell invasion. Migration is one step of the invasion process, and those authors proposed that QSOX1 post-translationally activates MMP-2 and MMP-9. Given that the pro-migratory function depends on sulfhydryl oxidase activity, QSOX1 may introduce disulfide bonds in extracellular substrates such as ECM proteins or exofacial thiol proteins, triggering migration. The importance of the ECM in cell adhesion and migration is well known, and ECM dysfunction underlies vascular pathologies [35,58–60]. Indeed, while our work was in progress, the Fass group demonstrated that QSOX1 knockdown in WI38 fibroblasts led to a decreased number of cells in the monolayer compared with scrambled control fibroblasts [16]. The addition of active recombinant QSOX1 prevented this effect, while the inactive mutant did not. The interpretation of these results was that QSOX1 knockdown impairs the adhesive capacity of the fibroblasts by limiting the incorporation of laminin into the ECM, a process supported by extracellular QSOX1 activity [16]. Exofacial thiol proteins may also be targets of QSOX1, as some of these contain redox-regulated cysteine residues [61–64], which can regulate cell functions [62,63]. Finally, hydrogen peroxide,

the product of QSOX1 catalysis, could also activate intracellular redox signaling, mimicking that triggered by PDGF [65], leading to VSMC migration (for a review, see [66]). Additionally, here we also showed that QSOX1 has a proproliferative role in VSMC. While QSOX1 knockdown leads to a decreased number of cells compared with the scrambled control, QSOX1 overexpression increases the amount of VSMC. One could interpret this finding as indicating that QSOX1 is pro-survival, as has already been described for PC12 cells [67] or similar to the small sulfhydryl oxidase ALR in hepatocytes [68]. However, compared with the scrambled control, QSOX1 knockdown also promoted decreased PCNA expression and decreased relative RNA content in VSMC, indicating that suppression of QSOX1 expression results in less cell cycle activity. In agreement with this, QSOX1 overexpression led to increased PCNA expression and increased RNA content. Collectively, these data support the concept that QSOX1 is mitogenic, stimulating the entrance of VSMC into the cell cycle. QSOX1 upregulation has long been associated with a reversible quiescent state in several cell lines, such as WI38 fibroblasts [4,23], endothelial cells [4,25] and endometrial cells [24]. However, more recent data have shown that QSOX1 expression may also be positively correlated with prostate [17,19], pancreas [18] and breast [20,21] tumor progression, likely because expression of QSOX1 promotes tumor cell proliferation and invasion [20,22], which is in agreement with our findings. The fact that proliferation did not appear to depend on sulfhydryl oxidase activity suggests that QSOX1 could be acting as

1344

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

Fig. 11. Effect of QSOX1 knockdown on PCNA expression in the neointima. (A) Representative photomicrographs of scrambled (n = 12 rats/time point) or QSOX1 (n = 12 rats/time point) siRNA-transfected injured carotid sections immunostained with anti-PCNA at 7, 14 and 21 days after the injury. The carotid sections were photographed at 400× magnification. Line indicates tunica media, while dot indicates tunica intima or neointima. (B) PCNA expression was measured by densitometry in 6 random fields/rat with Image J software. The values represent the mean ± SD. *p b 0.05.

a growth factor, similar to ALR, which binds to a receptor to trigger a mitogenic signaling [69]. In this context, the fact that, under low serum, the mutant enzyme has an enhanced proliferative effect compared with the wild type may indicate that thiol oxidation (in the ECM or at cell surface) may slightly balance the proliferative effect promoted by the structural conformation of QSOX1. Although the detailed mechanism underlying QSOX1-induced mitogenesis remains unknown, our data demonstrate that it does not depend of intracellular hydrogen peroxide, in contrast with the proliferation induced by PDGF [69], and that activation of MEK/ERK pathway is involved. Thus, QSOX1 promotes VSMC proliferation and migration by distinct and independent mechanisms, as reported for in vivo [70]. Regarding the relative roles of each QSOX1 isoform, since our siRNA and qPCR primers recognize both the long and the short versions of the oxidase, we cannot discriminate the contribution of each one. However, because we overexpressed the short QSOX1 and we employed the short QSOX1 in recombinant assays, the pro-proliferative and pro-migratory roles of this protein become clear, although the effects of the long isoform cannot be excluded. Our in vivo results demonstrate that QSOX1 participates in neointimal formation, as QSOX1 knockdown promotes a significant decrease in neointima thickness at day 21. Neointimal growth is a complex process, requiring the activity of countless molecules [33]. Therefore, it is not unexpected that the knockdown of a single protein does not completely block such a process but rather impairs it. Of note, since

QSOX1 knockdown is associated with tumor endothelial cells' proliferation [4,22] and QSOX1 effects are associated with lung fibroblasts' quiescence [24], it is possible that the overall effects of perivascular QSOX1 knockdown also involves, in a more complex way, additional effects on the vasculature, such as endothelialization and adventitial remodeling. Our data also showed that QSOX1 knockdown led to decreased expression of PCNA in neointimal cells at day 14 compared with scrambled siRNA-transfected injured artery, indicating that QSOX1 controls cell proliferation, either directly or indirectly. Because QSOX1 expression within the media layer was not altered by 21 days after lesion formation, the initial migration and/or proliferation of medial VSMC to the intima is most likely induced by factors other than QSOX1, such as PDGF, bFGF, other growth factors or other cytokines. Indeed, it was previously shown that PDGF stimulates SMC proliferation and migration during the first 7 days after endothelial denudation [71]. Our results revealed that cells forming the early neointimal layer express QSOX1 at day 7, and this expression peaks at day 14. Therefore, we suggest that between 7 and 14 days post-injury, neointimal-secreted QSOX1 induces medial cell migration to the intima and proliferation and then activates neointimal cell proliferation in an autocrine manner, as described for ALR in hepatocytes [72]. Thus, the role of QSOX1 in neointimal growth is not immediate; rather, it is associated with its expression by neointimal cells. Suppression of QSOX1 expression by these cells inhibits VSMC migration and proliferation primarily at days 7 and 14, leading to an

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

observable and significantly reduced intima/media area ratio only at day 21. It is important to note that although smooth muscle cells are considered to be the main component of neointima, other cell types are also reported to be present, such as fibroblasts and bone marrowderived cells [73], which might also contribute to QSOX1 expression. In conclusion, our in vitro and in vivo results provide clear evidences that QSOX1 positively regulates proliferation and migration of VSMC, thus contributing to neointima growth induced by balloon injury. Importantly, the relatively modest extent of the effects promoted by QSOX1 knockdown implies that QSOX1 does not play solitary roles in proliferation and migration but rather collaborates with other molecules in such processes. Therefore, our data indicate novel roles for QSOX1 in a disease-related model of vascular repair after injury. Transparency document The Transparency document associated with this article can be found, in the online version. Acknowledgments This work was supported by INCT Redoxoma, Fundação Araucária, CNPq and Ligue contre le cancer (France). The authors thank Joselito Getz for performing the initial in vivo experiments and Dr. Stenio Perdigão and Dr. Lucia de Noronha for providing the anti-GFP and the anti-PCNA antibodies, respectively. Fellowships from CAPES (awarded to BEB, ARC, MLP, CAS) and CNPq (awarded to SMZ, FRML, LSN) are also acknowledged. Appendix A. Supplementary material Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbadis.2015.03.002. References [1] C. Thorpe, K.L. Hoober, S. Raje, N.M. Glynn, J. Burnside, G.K. Turi, D.L. Coppock, Sulfhydryl oxidases: emerging catalysts of protein disulfide bond formation in eukaryotes, Arch. Biochem. Biophys. 405 (2002) 1–12. [2] G. Mairet-Coello, A. Tury, A. Esnard-Feve, D. Fellmann, P.-Y. Risold, B. Griffond, FAD-linked sulfhydryl oxidase QSOX: topographic, cellular, and subcellular immunolocalization in adult rat central nervous system, J. Comp. Neurol. 473 (2004) 334–363. [3] K.L.K.L. Hoober, B. Joneja, H.B.H.B. White, C. Thorpe, A sulfhydryl oxidase from chicken egg white, J. Biol. Chem. 271 (1996) 30510–30516. [4] D. Coppock, C. Kopman, J. Gudas, D.A. Cina-Poppe, Regulation of the quiescenceinduced genes: quiescin Q6, decorin, and ribosomal protein S29, Biochem. Biophys. Res. Commun. 269 (2000) 604–610. [5] B. Benayoun, A. Esnard-Fève, S. Castella, Y. Courty, F. Esnard, Rat seminal vesicle FAD-dependent sulfhydryl oxidase. Biochemical characterization and molecular cloning of a member of the new sulfhydryl oxidase/quiescin Q6 gene family, J. Biol. Chem. 276 (2001) 13830–13837. [6] C. Amiot, J.F. Musard, M. Hadjiyiassemis, M. Jouvenot, D. Fellmann, P.Y. Risold, P. Adami, Expression of the secreted FAD-dependent sulfydryl oxidase (QSOX) in the guinea pig central nervous system, Mol. Brain Res. 125 (2004) 13–21. [7] J. Jaje, H.N. Wolcott, O. Fadugba, D. Cripps, A.J. Yang, I.H. Mather, C. Thorpe, A flavindependent sulfhydryl oxidase in bovine milk, Biochemistry 46 (2007) 13031–13040. [8] S.M. Zanata, A.C. Luvizon, D.F. Batista, C.M. Ikegami, F.O. Pedrosa, E.M. Souza, D.F.S. Chaves, L.F. Caron, J.V. Pelizzari, F.R.M. Laurindo, L.S. Nakao, High levels of active quiescin Q6 sulfhydryl oxidase (QSOX) are selectively present in fetal serum, Redox Rep. 10 (2005) 319–323. [9] D. Simper, U. Mayr, C. Urbich, A. Zampetaki, M. Prokopi, A. Didangelos, A. Saje, M. Mueller, U. Benbow, A.C. Newby, R. Apweiler, S. Rahman, S. Dimmeler, Q. Xu, M. Mayr, Comparative proteomics profiling reveals role of smooth muscle progenitors in extracellular matrix production, Arterioscler. Thromb. Vasc. Biol. 30 (2010) 1325–1332. [10] B.A. Israel, L. Jiang, S.A. Gannon, C. Thorpe, Disulfide bond generation in mammalian blood serum: detection and purification of quiescin-sulfhydryl oxidase, Free Radic. Biol. Med. 69 (2014) 129–135. [11] J. Radom, D. Colin, F. Thiebault, M. Dognin-Bergeret, G. Mairet-Coello, A. EsnardFeve, D. Fellmann, M. Jouvenot, Identification and expression of a new splicing variant of FAD-sulfhydryl oxidase in adult rat brain, Biochim. Biophys. Acta 1759 (2006) 225–233. [12] D.L. Coppock, C. Thorpe, M.A. Liebert, Multidomain flavin-dependent sulfhydryl oxidases, Antioxid. Redox Signal. 8 (2006) 300–311.

1345

[13] S. Matsuba, Y. Suga, K. Ishidoh, Y. Hashimoto, K. Takamori, E. Kominami, B. Wilhelm, J. Seitz, H. Ogawa, Sulfhydryl oxidase (SOx) from mouse epidermis: molecular cloning, nucleotide sequence, and expression of recombinant protein in the cultured cells, J. Dermatol. Sci. 30 (2002) 50–62. [14] S. Chakravarthi, C.E. Jessop, M. Willer, C.J. Stirling, N.J. Bulleid, Intracellular catalysis of disulfide bond formation by the human sulfhydryl oxidase, QSOX1, Biochem. J. 404 (2007) 403–411. [15] J. Rudolf, M.A. Pringle, N.J. Bulleid, Proteolytic processing of QSOX1A ensures efficient secretion of a potent disulfide catalyst, Biochem. J. 454 (2013) 181–190. [16] T. Ilani, A. Alon, I. Grossman, B. Horowitz, E. Kartvelishvily, S.R. Cohen, D. Fass, A secreted disulfide catalyst controls extracellular matrix composition and function, Science 341 (2013) 74–76 (80-.). [17] X. Ouyang, T.L. Deweese, W.G. Nelson, C. Abate-Shen, Loss-of-function of Nkx3. 1 promotes increased oxidative damage in prostate carcinogenesis, Cancer Res. 65 (2005) 6773–6779. [18] K. Antwi, G. Hostetter, M.J. Demeure, B.A. Katchman, G.A. Decker, Y. Ruiz, T.D. Sielaff, L.J. Koep, D.F. Lake, Analysis of the plasma peptidome from pancreas cancer patients connects a peptide in plasma to overexpression of the parent protein in tumors, J. Proteome Res. 8 (2009) 4722–4731. [19] H. Song, B. Zhang, M.A. Watson, P.A. Humphrey, H. Lim, J. Milbrandt, Loss of Nkx3.1 leads to the activation of discrete downstream target genes during prostate tumorigenesis, Oncogene 28 (2009) 3307–3319. [20] B.A. Katchman, I.T. Ocal, H.E. Cunliffe, Y.-H. Chang, G. Hostetter, A. Watanabe, J. Lobello, D.F. Lake, Expression of quiescin sulfhydryl oxidase 1 is associated with a highly invasive phenotype and correlates with a poor prognosis in Luminal B breast cancer, Breast Cancer Res. 15 (2013) R28. [21] M. Soloviev, M.P. Esteves, F. Amiri, M.R. Crompton, C.C. Rider, Elevated transcription of the gene QSOX1 encoding quiescin Q6 sulfhydryl oxidase 1 in breast cancer, PLoS One 8 (2013) e57327. [22] B.A. Katchman, K. Antwi, G. Hostetter, M.J. Demeure, A. Watanabe, G.A. Decker, L.J. Miller, D.D. Von Hoff, D.F. Lake, Quiescin sulfhydryl oxidase 1 promotes invasion of pancreatic tumor cells mediated by matrix metalloproteinases, Mol. Cancer Res. 9 (2011) 1621–1631. [23] D.L.D.L. Coppock, C. Kopman, S. Scandalis, S. Gilleran, Preferential gene expression in quiescent human lung fibroblast, Cell Growth Differ. 4 (1993) 483–493. [24] J.F. Musard, M. Sallot, P. Dulieu, A. Fraîchard, C. Ordener, J.P. Remy-Martin, M. Jouvenot, P. Adami, Identification and expression of a new sulfhydryl oxidase SOx-3 during the cell cycle and the estrus cycle in uterine cells, Biochem. Biophys. Res. Commun. 287 (2001) 83–91. [25] D.M.E.I. Hellebrekers, V. Melotte, E. Viré, E. Langenkamp, G. Molema, F. Fuks, J.G. Herman, W. Van Criekinge, A.W. Griffioen, M. van Engeland, Identification of epigenetically silenced genes in tumor endothelial cells, Cancer Res. 67 (2007) 4138–4148. [26] C.R. de Andrade, B.S. Stolf, V. Debbas, D.S. Rosa, J. Kalil, V. Coelho, F.R.M. Laurindo, Quiescin sulfhydryl oxidase (QSOX) is expressed in the human atheroma core: possible role in apoptosis, In Vitro Cell. Dev. Biol. 47 (2011) 716–727. [27] A. Mebazaa, G. Vanpoucke, G. Thomas, K. Verleysen, A. Cohen-Solal, M. Vanderheyden, J. Bartunek, C. Mueller, J.-M. Launay, N. Van Landuyt, F. D'Hondt, E. Verschuere, C. Vanhaute, R. Tuytten, L. Vanneste, K. De Cremer, J. Wuyts, H. Davies, P. Moerman, et al., Unbiased plasma proteomics for novel diagnostic biomarkers in cardiovascular disease: identification of quiescin Q6 as a candidate biomarker of acutely decompensated heart failure, Eur. Heart J. 33 (2012) 2317–2324. [28] P.C. Rancy, C. Thorpe, Oxidative protein folding in vitro: a study of the cooperation between quiescin-sulfhydryl oxidase and protein disulfide isomerase, Biochemistry 47 (2010) 12047–12056. [29] M. Janiszewski, L.R. Lopes, A.O. Carmo, M. a Pedro, R.P. Brandes, C.X.C. Santos, F.R.M. Laurindo, Regulation of NAD(P)H oxidase by associated protein disulfide isomerase in vascular smooth muscle cells, J. Biol. Chem. 280 (2005) 40813–40819. [30] K. Szocs, Upregulation of nox-based NAD(P)H oxidases in restenosis after carotid injury, Arterioscler. Thromb. Vasc. Biol. 22 (2002) 21–27. [31] M.Y. Lee, A.S.A. Martin, P.K. Mehta, A.E. Dikalova, S.R. Datla, E. Lyons, K. Krause, B. Banfi, J. David, B. Lassègue, K.K. Griendling, A. San Martin, A.M. Garrido, J.D. Lambeth, Mechanisms of vascular smooth muscle NADPH oxidase 1 (Nox1) contribution to injury-induced neointimal formation, Arterioscler. Thromb. Vasc. Biol. 29 (2009) 480–487. [32] C.L. Jackson, Animal models of restenosis, Trends Cardiovasc. Med. 4 (1994) 122–130. [33] G.A. Ferns, T.Y. Avades, The mechanisms of coronary restenosis: insights from experimental models, Int. J. Exp. Pathol. 81 (2000) 63–88. [34] A.W. Clowes, M.M. Clowes, J. Fingerle, M.A. Reidy, Regulation of smooth muscle cell growth in injured artery, J. Cardiovasc. Pharmacol. 14 (Suppl. 6) (1989) S12–S15. [35] W.B. Batchelor, R. Robinson, B.H. Strauss, The extracellular matrix in balloon arterial injury: a novel target for restenosis prevention, Prog. Cardiovasc. Dis. 41 (1998) 35–49. [36] P.F. Leite, M. Liberman, F. Sandoli de Brito, F.R.M. Laurindo, Redox processes underlying the vascular repair reaction, World J. Surg. 28 (2004) 331–336. [37] H.F. McMurray, D.P. Parrott, D.E. Bowyer, A standardised method of culturing aortic explants, suitable for the study of factors affecting the phenotypic modulation, migration and proliferation of aortic smooth muscle cells, Atherosclerosis 86 (1991) 227–237. [38] G. Chu, H. Hayakawa, P. Berg, Electroporation for the efficient transfection of mammalian cells with DNA, Nucleic Acids Res. 15 (1987) 1311–1326. [39] K.F. Portes, C.M. Ikegami, J. Getz, A.P. Martins, L. de Noronha, L.F. Zischler, G. Klassen, A.A. Camargo, S.M. Zanata, E. Bevilacqua, L.S. Nakao, Tissue distribution of quiescin Q6/sulfhydryl oxidase (QSOX) in developing mouse, J. Mol. Histol. 39 (2008) 217–225. [40] K. Limor-Waisberg, S. Ben-Dor, D. Fass, Diversification of quiescin sulfhydryl oxidase in a preserved framework for redox relay, BMC Evol. Biol. 13 (2013) 70.

1346

B.E. Borges et al. / Biochimica et Biophysica Acta 1852 (2015) 1334–1346

[41] E.J. Heckler, A. Alon, D. Fass, C. Thorpe, Human quiescin-sulfhydryl oxidase, QSOX1: probing internal redox steps by mutagenesis, Biochemistry 47 (2008) 4955–4963. [42] J. Kusma, O.M. Chaim, A.C.M. Wille, V.P. Ferrer, Y.B. Sade, Nephrotoxicity caused by brown spider venom phospholipase-D (dermonecrotic toxin) depends on catalytic activity, Biochimie 90 (2008) 1722–1736. [43] E.J. Heckler, P.C. Rancy, V.K. Kodali, C. Thorpe, Generating disulfides with the Quiescin-sulfhydryl oxidases, Biochim. Biophys. Acta 1783 (2008) 567–577. [44] S. Raje, N.M. Glynn, C. Thorpe, A continuous fluorescence assay for sulfhydryl oxidase, Anal. Biochem. 307 (2002) 266–272. [45] K. Naliwaiko, A.C. Luvizon, L. Donatti, R. Chammas, A.F. Mercadante, S.M. Zanata, L.S. Nakao, Guanosine promotes B16F10 melanoma cell differentiation through PKC-ERK 1/2 pathway, Chem. Biol. Interact. 173 (2008) 122–128. [46] S. Zhu, R. Xue, P. Zhao, F.-L. Fan, X. Kong, S. Zheng, Q. Han, Y. Zhu, N. Wang, J. Yang, Y. Guan, Targeted disruption of the prostaglandin E2 E-prostanoid 2 receptor exacerbates vascular neointimal formation in mice, Arterioscler. Thromb. Vasc. Biol. 31 (2011) 1739–1747. [47] B.E. Borges, V.R. Teixeira, M.H. Appel, C.A. Steclan, F. Rigo, F. Filipak Neto, A.M. da Costa Ferreira, R. Chammas, S.M. Zanata, L.S. Nakao, De novo galectin-3 expression influences the response of melanoma cells to isatin-Schiff base copper (II) complex-induced oxidative stimulus, Chem. Biol. Interact. 206 (2013) 37–46. [48] J. Beckman, R. Minor, C. White, J. Repine, G. Rosen, B. Freeman, Superoxide dismutase and catalase conjugated to polyethylene glycol increases endothelial enzyme activity and oxidant resistance, J. Biol. Chem. 263 (1988) 6884–6892. [49] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using realtime quantitative PCR and the 2(-Delta Delta C(T)) Method, Methods 25 (2001) 402–408. [50] H.M. Shapiro, Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and pyronin Y, Cytometry 2 (1981) 143–150. [51] E.S. Cunha, R. Kawahara, M.K. Kadowaki, H.G. Amstalden, G.R. Noleto, S.M.S.C. Cadena, S.M.B. Winnischofer, G.R. Martinez, Melanogenesis stimulation in B16-F10 melanoma cells induces cell cycle alterations, increased ROS levels and a differential expression of proteins as revealed by proteomic analysis, Exp. Cell Res. 318 (2012) 1913–1925. [52] T. Fujiwara, K. Oda, S. Yokota, A. Takatsuki, Y. Ikehara, Brefeldin A causes disassembly of the Golgi complex and accumulation of secretory proteins in the endoplasmic reticulum, J. Biol. Chem. 263 (1988) 18545–18552. [53] D. Accorsi-Mendonça, F.M.A. Corrêa, T.B. Paiva, H.P. de Souza, F.R.M. Laurindo, A.M. de Oliveira, The balloon catheter induces an increase in contralateral carotid artery reactivity to angiotensin II and phenylephrine, Br. J. Pharmacol. 142 (2004) 79–88. [54] L. Wang, J. Zheng, X. Bai, B. Liu, C.-J. Liu, Q. Xu, Y. Zhu, N. Wang, W. Kong, X. Wang, ADAMTS-7 mediates vascular smooth muscle cell migration and neointima formation in balloon-injured rat arteries, Circ. Res. 104 (2009) 688–698. [55] G. Kayser, A. Csanadi, C. Otto, T. Plönes, N. Bittermann, J. Rawluk, B. Passlick, M. Werner, Simultaneous multi-antibody staining in non-small cell lung cancer strengthens diagnostic accuracy especially in small tissue samples, PLoS One 8 (2013) e56333. [56] C.L. Skelly, A. Chandiwal, J.E. Vosicky, R.R. Weichselbaum, B. Roizman, Attenuated herpes simplex virus 1 blocks arterial apoptosis and intimal hyperplasia induced

[57] [58]

[59] [60] [61]

[62]

[63]

[64] [65]

[66] [67]

[68]

[69]

[70] [71]

[72]

[73]

by balloon angioplasty and reduced blood flow, Proc. Natl. Acad. Sci. 104 (2007) 12474–12478. C. Shanahan, P. Weissberg, J. Metcalfe, Isolation of gene markers of differentiated and proliferating vascular smooth muscle cells, Circ. Res. 17 (1993) 193–204. C. Rodríguez, J. Martínez-González, B. Raposo, J.F. Alcudia, A. Guadall, L. Badimon, Regulation of lysyl oxidase in vascular cells: lysyl oxidase as a new player in cardiovascular diseases, Cardiovasc. Res. 79 (2008) 7–13. C.A. Lemarié, P.-L. Tharaux, S. Lehoux, Extracellular matrix alterations in hypertensive vascular remodeling, J. Mol. Cell. Cardiol. 48 (2010) 433–439. S.A. Siefert, R. Sarkar, Matrix metalloproteinases in vascular physiology and disease, Vascular 20 (2012) 210–216. X.M. Jiang, M. Fitzgerald, C.M. Grant, P.J. Hogg, Redox control of exofacial protein thiols/disulfides by protein disulfide isomerase, J. Biol. Chem. 274 (1999) 2416–2423. L.J. Matthias, P.T.W. Yam, X.-M. Jiang, N. Vandegraaff, P. Li, P. Poumbourios, N. Donoghue, P.J. Hogg, Disulfide exchange in domain 2 of CD4 is required for entry of HIV-1, Nat. Immunol. 3 (2002) 727–732. T. Laragione, V. Bonetto, F. Casoni, T. Massignan, G. Bianchi, E. Gianazza, P. Ghezzi, Redox regulation of surface protein thiols: identification of integrin alpha-4 as a molecular target by using redox proteomics, Proc. Natl. Acad. Sci. 100 (2003) 14737–14741. B. Sahaf, K. Heydari, L.A. Herzenberg, L.A. Herzenberg, Lymphocyte surface thiol levels, Proc. Natl. Acad. Sci. 100 (2003) 4001–4005. M. Sundaresan, Z.X. Yu, V.J. Ferrans, K. Irani, T. Finkel, Requirement for generation of H2O2 for platelet-derived growth factor signal transduction, Science 270 (1995) 296–299 (80-.). A. San Martín, K.K. Griendling, Redox control of vascular smooth muscle migration, Antioxid. Redox Signal. 12 (2010) 625–640. C. Morel, P. Adami, J.-F. Musard, D. Duval, J. Radom, M. Jouvenot, Involvement of sulfhydryl oxidase QSOX1 in the protection of cells against oxidative stressinduced apoptosis, Exp. Cell Res. 313 (2007) 3971–3982. C. Thirunavukkarasu, L.F. Wang, S.A.K. Harvey, S.C. Watkins, J.R. Chaillet, J. Prelich, T.E. Starzl, C.R. Gandhi, Augmenter of liver regeneration: an important intracellular survival factor for hepatocytes, J. Hepatol. 48 (2008) 578–588. G. Wang, X. Yang, Y. Zhang, Q. Wang, H. Chen, H. Wei, G. Xing, L. Xie, Z. Hu, C. Zhang, D. Fang, C. Wu, F. He, Identification and characterization of receptor for mammalian hepatopoietin that is homologous to yeast ERV1, J. Biol. Chem. 274 (1999) 11469–11472. A.W. Clowes, S.M. Schwartz, Significance of quiescent smooth muscle migration in the injured rat carotid artery, Circ. Res. 56 (1985) 139–145. A. Jawien, D.F. Bowen-Pope, V. Lindner, S.M. Schwartz, A.W. Clowes, Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty, J. Clin. Invest. 89 (1992) 507–511. Y. Li, G. Xing, Q. Wang, M. Li, H. Wei, G. Fan, J. Chen, X. Yang, C. Wu, H. Chen, F. He, Hepatopoietin acts as an autocrine growth factor in hepatoma cells, DNA Cell Biol. 20 (2001) 791–795. L. Rodriguez-Menocal, M. St-Pierre, Y. Wei, S. Khan, D. Mateu, M. Calfa, A. RahnemaiAzar, G. Striker, S. Pham, R. Vazquez-Padron, The origin of post-injury neointimal cells in the rat balloon injury model, Cardiovasc. Res. 81 (2009) 46–53.