free radical biology and medicine - connecting repositories · 2016. 12. 7. · proteins. both...

Free Radical Biology and Medicine 96 (2016) 374–384

Contents lists available at ScienceDirect

Free Radical Biology and Medicine

http://d0891-58

n CorrE-m

journal homepage: www.elsevier.com/locate/freeradbiomed

NADPH oxidase 4 deficiency leads to impaired wound repair andreduced dityrosine-crosslinking, but does not affect myofibroblastformation

Dominik Lévigne a,n, Ali Modarressi a, Karl-Heinz Krause b, Brigitte Pittet-Cuénod a

a Division of Plastic, Reconstructive & Aesthetic Surgery, Geneva University Hospitals, Geneva, Switzerlandb Department of Pathology, Faculty of Medicine, University of Geneva, Geneva, Switzerland

a r t i c l e i n f o

Article history:Received 30 November 2015Received in revised form26 April 2016Accepted 28 April 2016Available online 30 April 2016

Keywords:NADPH oxidase 4Dityrosine crosslinkingWound healingWound repairHIF1alphaCD31MyofibroblastMouse modelNeovascularizationROSMPONOX2Collagen crosslinkingExtracellular matrixNOX4VPO1Myeloperoxidase

x.doi.org/10.1016/j.freeradbiomed.2016.04.19449/& 2016 The Authors. Published by Elsevier

esponding author.ail address: [email protected] (D. Lévigne).

a b s t r a c t

NADPH oxidases (NOX) mediate redox signaling by generating superoxide and/or hydrogen peroxide,which are involved in biosynthetic pathways, e.g. thyroid hormone generation, dityrosine crosslinking, aswell as bacterial killing. Data investigating the role of NOX enzymes in cutaneous wound repair is limitedand specifically their function in skin myofibroblast expression is unknown. The isoform NOX4 was re-cently shown to be a pre-requisite for the differentiation of cardiac and pulmonary myofibroblasts.

In this study we investigate the role of NOX4 in wound repair using a wound model in NOX4knockout mice (n¼16) and wildtype mice (n¼16). Wounds were photographed daily until completewound closure. Mice were sacrificed at day 3, 7, 14; wound tissue was harvested.

NOX4-deficient mice healed significantly slower (22 days, SD¼1.9) than wild-type mice (17 days,SD¼1.4, po0.005). However, there was no difference in myofibroblast expression. Strong dityrosineformation was observed, but was significantly weaker in NOX4-/- mice (po0.05). NOX2, HIF1α and CD31expression was significantly weaker in NOX4-/- mice (po0.05).

In this study we show for the first time that NOX4 plays a role in cutaneous wound repair. Our datasuggests that NOX4 mediates HIF1α expression and neoangiogenesis during wound repair. NOX4 dele-tion led to a decreased expression of NOX2, implying a role of NOX4 in phagocytic cell recruitment. NOX4was required for effective wound contraction but not myofibroblast expression. We suggest that myo-fibroblast contraction in NOX4-deficient mice is less effective in contracting the wound because of in-sufficient dityrosine-crosslinking of the ECM, providing the first indication for a physiological function ofdityrosine crosslinking in higher animals.& 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Wound repair is a complex, dynamic and interactive processthat involves a great variety of cells, soluble mediators and ex-tracellular matrix components. In order to guarantee rapid andeffective wound closure, wound contraction is of primordial im-portance. When the wound contracts, the uninjured skin sur-rounding the wound is pulled into the defect, significantly redu-cing the amount of time and tissue needed to reestablish the in-tegrity of the skin barrier after injury. The exact mechanisms un-derlying wound contraction are not yet fully understood. Cellularcontraction of specialized contractile fibroblasts – myofibroblasts –constitutes one of the key components of effective wound

Inc. This is an open access article u

contraction [1,2]. Fibroblast-to-myofibroblast differentiation oc-curs primarily in response to growth factors (mainly transforminggrowth factor beta 1 (TGFβ1)) and mechanical stress. Myofibro-blast are characterized by the expression of α-smooth muscle actin(αSMA) in stress fibers, which is considered to be the molecularbasis for their high contractility [1–3].

In recent years it has become increasingly clear that redoxsignaling is involved in a myriad of physiological cell functionsincluding differentiation, proliferation, apoptosis and migration[4–6]. NOX enzymes are membrane-bound complexes that trans-port electrons across biological membranes to reduce oxygen tosuperoxide-radical and play a key role in mediating redox signal-ing [5]. Seven isoforms of NOX enzymes have been described:NOX1, NOX2, NOX3, NOX4, NOX5, DUOX1, and DUOX2. Despitetheir similar structure, they differ in function and mechanism ofactivation. During wound repair, they could play important roles inkey elements such as coagulation, inflammation, fibroplasia,

nder the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

www.sciencedirect.com/science/journal/08915849www.elsevier.com/locate/freeradbiomedhttp://dx.doi.org/10.1016/j.freeradbiomed.2016.04.194http://dx.doi.org/10.1016/j.freeradbiomed.2016.04.194http://dx.doi.org/10.1016/j.freeradbiomed.2016.04.194http://crossmark.crossref.org/dialog/?doi=10.1016/j.freeradbiomed.2016.04.194&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1016/j.freeradbiomed.2016.04.194&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1016/j.freeradbiomed.2016.04.194&domain=pdfmailto:[email protected]://dx.doi.org/10.1016/j.freeradbiomed.2016.04.194

D. Lévigne et al. / Free Radical Biology and Medicine 96 (2016) 374–384 375

angiogenesis, re-epithelialization and contraction [6,7]. The role ofNOX2 during wound repair is long known in microbial killing as itguarantees the production of large amounts of reactive oxygenspecies (ROS) by immune cells, which is of great importance foreffective host defense [8–11]. In fact, patients with chronic gran-ulomatous disease, a rare congenital abnormality of the NOX2system, show impaired wound repair and are more prone towound infection [12]. Our knowledge concerning the role of NOX-mediated redox signaling in wound repair, especially concerningwound contraction, is still limited and relies mainly on studiesconducted in other fields than skin wound repair.

Recent literature shows that TGFβ1 induces hydrogen peroxideproduction in human fibroblasts [13]. In fact, NOX4 has been re-ported to mediate TGFβ1-induced myofibroblast differentiation ofcardiac [14] and pulmonary fibroblasts [15] as well as the pro-liferation of pulmonary smooth muscle cells [16]. Carnesecchi et alinvestigated the role of NOX4 in fibroblast-to-myofibroblast dif-ferentiation during bleomycin-induced pulmonary fibrosis, theaccumulation of myofibroblasts being a hallmark in advanced andprogressive pulmonary fibrosis. They showed that NOX4 was aprerequisite for myofibroblast differentiation [17]. The accumula-tion of myofibroblasts during lung fibrosis is mainly driven byTGFβ1 signaling [18,19], in line with what has been observed incutaneous wound repair studies [20].

The extracellular matrix (ECM) plays a primordial role inwound repair, not only for myofibroblast differentiation but alsofor wound contraction itself. The crosslinking of ECM proteins iscrucial for the development of a mechanically resistant granula-tion tissue. Dimerization of the phenolic amino acid L-tyrosine hasbeen suggested as a possible mechanism of crosslinking ECMproteins. Both collagen and elastin have been shown to undergodityrosine crosslinking when H2O2 is added exogenously [21,22].In culture, it has been observed that TGFβ1-activated fibroblastsrelease ROS in the extracellular space in a NOX-mediated manner[23]. Larios et al reported that H2O2 generated by TGFβ1-activatedfibroblasts are capable of mediating dimerization of L-tyrosine inthe ECM [24]. A physiological function for NOX/ROS-dependentcross-linking of dityrosine residues in the ECM has so far only bedemonstrated in primitive organisms: stabilization of cutaneoustissue in Caenorhabditis elegans [25]; hardening of the fertilizationenvelope in sea urchin eggs [26]; and stabilization of drosophilawings [27]. We hypothesize that NOX4 plays a role in the devel-opment of the granulation tissue during wound repair by med-iating ROS-induced crosslinking of tyrosine residues in the ECM.

NOX4 may also play a role in other processes of wound repair,possibly during angiogenesis as it was previously reported thatgrowth of blood vessels into sponges is abolished in NOX4knockout mice [28]. Hypoxia is arguably the most importanttrigger for angiogenesis during wound repair. NOX4 could play akey role in cell response to hypoxia. Pulmonary artery smoothmuscle cells cultured under hypoxic conditions showed increasedexpression of NOX4 [4,29] and their proliferation was NOX4-de-pended [30]. There is increasing evidence that the stabilization ofhypoxia-inducible factor (HIF) can also be directly induced by ROS[31,32]. In fact, there appears to be a positive feed-forward loopinvolving NOX4 and HIF1α: ROS generated by NOX4 activateHIF1α [33] and HIF1α activates the expression of NOX4 [34]. Thenotion that redox signaling plays an important role in angiogen-esis is supported by evidence suggesting that a large number ofantioxidants limit angiogenesis [35–37].

In this study we aimed to understand 1) whether NOX4 plays arole in cutaneous wound repair, 2) whether NOX4 is necessary foreffective contraction studying myofibroblast expression, collagendeposition and dityrosine crosslinking and 3) whether NOX4 playsa role in HIF1α expression and angiogenesis during wound repair.

2. Material and methods

2.1. Animal wound model

In this study, 16 wild type mice (C57BL/6J) and 16 NOX4knockout mice (B6/129S9-NOX4) were used. We created circularfull thickness wounds of 1.5 cm diameter (1.8 cm2) on the back ofNOX4-deficient knockout mice and of wild type mice, using acircular template. All animal experiments were approved by thelocal veterinary authority (“Direction générale de la santé deGenève”, authorization number G65/3878).

2.2. Wound repair assessment

Wound size and aspect was documented immediately afterwounding and on day 3, 7, 10, 14, 17 and 21. Wounds that wereclose to complete wound closure at any of these time points werephotographed daily from there on in order to determine the day ofcomplete wound closure. Wounds were photographed at a con-stant distance with a ruler next to the wound for scaling. Thewound surface was calculated on photos using a computer-as-sisted image analysis system (Image J).

At complete wound closure (i.e., full epithelialization), thesurface of hairless skin of the scar was measured and considered tocorrespond to the area of the wound healed by epithelialization.The surface of the wound healed by contraction was then esti-mated by subtraction of the epithelialized surface from the woundsurface measured at day 0.

2.3. Histology and immunohistochemistry

Mice were sacrificed at day 3, 7 and 14 (n¼2 per time pointand per group) and the entire wound with the souring uninjuredskin was removed. The tissue was fixed in 4% bufferedformaldehyde.

4 μm thick tissue sections were analyzed by im-munohistochemistry using with the Ventana Discovery automatedstaining system (Ventana Medical Systems, Tucson, AZ, USA).Ventana reagents were used for the entire procedure.

For αSMA (anti-alpha-smooth muscle actin, Dako, #M0851,70mg/ml) and DT (anti dityrosine, JaICA, MDT-020P, 100ug/ml)mouse monoclonal antibodies, no antigen retrieval pre-treatmentwas required. After automatic deparaffinization, slides were in-cubated 30 minutes at 37 °C with primary antibodies diluted at1/300 and 1/400 respectively for αSMA and DT. Then secondaryantibodies were applied at dilution 1/250 (anti-mouse Ig-G1þ IgG2aþ IgG3, abcam, ab133469, 2.03 mg/ml). Detection ofsecondary antibodies was carried out using the rabbit OmniMapkit (Ventana Medical Systems), based on conversion of diamino-benzidine to a dye with multimeric horseradish peroxidase (HRP).

For HIF1α (Novusbio, NB100-479, 1 mg/ml), antigen retrievalwas performed by heating slides 36 mns in standard CC2 EDTAsolution pH 8.4, dilution 1/400, detection Rabbit-OmniM detectionkit.

For CD31 (Abcam, ab28364), antigen retrieval was performedby heating slides 52 min in standard CC1 citrate solution pH 6.0,dilution 1/50, detection Rabbit-OmniM detection kit.

MPO staining was performed using a rabbit polyclonal antibody(Dako,#A0398, 3.2 g/L). Slides were incubated 12 min in CC1 buf-fer for antigen retrieval and primary antibody was incubated30 min at 37 °C at dilution 1/1000. As a negative control, spinalcord tissue of wildtype mice was used. As a positive control, weused spinal cord tissue from an amyotrophic lateral sclerosismouse model which is known to have activated microglia thatexpress MPO [38].

NOX2 staining was performed using a mouse monoclonal

D. Lévigne et al. / Free Radical Biology and Medicine 96 (2016) 374–384376

antibody (Santa Cruz, #sc-130548, 200 μg/ml). Slides were in-cubated 36 min in CC1 buffer for antigen retrieval. First the pri-mary antibody was incubated 30 min at 37 °C at dilution 1/100and then a secondary antibody was applied at dilution 1/250 (anti-mouse IgG1þ IgG2aþ IgG3, abcam, ab133469, 2.03 mg/ml).

Detection was carried out using the rabbit OmniMap kit(Ventana Medical Systems), based on conversion of diamino-benzidine to a dye with multimeric horseradish peroxidase (HRP).

NOX2 antibody control staining was performed using the sameantibody in immunofluorescence on cell cultures of mouse bone-marrow-derived dendritic cells fixed in PFA 4% at a 1:100 dilutionof either wildtype or NOX2-/- mice (data kindly provided by DrTamara Seredenina, Department of Pathology and Immunology,University of Geneva).

Standard Masson’s trichrome staining was used to quantifycollagen expression.

2.4. Image analysis

Sections were scanned with a Mirax Widefield scanner(brightfield detector: Marlin F-146C IRF Medical). Pictures werethen processed using Definiens Tissue Studio software (DefiniensAG, Munich, Germany). Regions of interest (ROIs) were defined inthe granulation tissue and in the epithelium. To evaluate α-SMAexpression, vessels were manually excluded from the image toexclude pericytes. Results were given as % of ROI area stained.

2.5. Statistical methods

All values are expressed as mean7standard deviation (SD).Data was analyzed with Prism6 software (GraphPad Software Inc.,La Jolla, USA). Statistical analysis consisted in a comparison of datafrom wild type versus NOX4-deficient mice, using two-tailed

Fig. 1. A) Wound aspect in wildtype and NOX4-/- mice: Wounds in both groups healewound closure, scale bar¼1 cm. B) Percentage of mice with open wounds over time: wh24 days in the NOX4-/- group, Mantel-cox test comparing the two curves showed a p valsignificantly slower compared to wildtype mice (white par bar), whiskers indicate min

Student t-tests for unpaired comparisons between groups. Differ-ences were considered significant at po0.05. In cases of multiplecomparisons, a post hoc correction with the Bonferroni procedurewas performed. The Mantel-cox test was used for comparing thedifference between the control and NOX4-deficient group in per-centage of wounds closed at different time points.

3. Results

3.1. Macroscopic analysis

We observed a significant delay in wound closure in NOX4-deficient mice compared to the control group. The macroscopicwound aspect showed no obvious difference with regards togranulation tissue formation or fibrin deposition (see Fig. 1a).While it took 18 days until 90% of the wounds were closed in thecontrol group, it took 24 days in the NOX4-/- group (see Fig. 1b).Wounds in the control group took on average 17 days until com-plete wound closure (SD¼1.4) compared to 22 days in the NOX4-/-group (SD¼1.9, po0.005, see Fig. 1c).

When looking at the fraction of the wound area closed by re-epithelialization (hairless skin at the time of complete woundclosure), we observed that wounds healed almost exclusively bycontraction, as expected for this kind of wound repair mousemodel (6.172.3% and 6.772.4% of the surface healed by re-epi-thelialization in wildtype and NOX4-/- mice respectively). Thedifference in the ratio between epithelialization and contractionbetween the two groups was not statistically significant.

3.2. Myofibroblast expression

Staining for αSMA revealed no statistically significant

d predominantly through contraction. NOX4-/- mice showed a significant delay inile it took 18 days until 90% of the wounds were closed in the control group, it tookue of o0.005. C) Time to complete wound closure: NOX4 -/- mice (gray bar) healedand max values, n¼10 per group, po0.005.

Fig. 2. Myofibroblast expression. Immunohistochemistry showing αSMA expression in the granulation tissue at day 3, 7 and 14 after wound infliction in wildtype (leftcolumn) and NOX4-/- mice (middle column). E¼epithelium; G¼granulation tissue, V¼vessel. A strong increase in expression of αSMA was observed at day 7 with a furtherincrease at day 14. Computer-assisted quantification (right column) showed no significant difference between wildtype and NOX4-/- mice in αSMA expression in thegranulation tissue. Values are expressed as % area of the region of interest stained by immunohistochemistry and as mean7SD.


difference between wildtype and NOX4-/- mice. We observed aweak expression of αSMA at day 3 (0.570.47% vs 0.570.2% of ROIstained respectively) and a strongly increased expression at day 7(13.271.6% vs 12.074.3%), which was again increased at day 14(20.071.8% vs 18.374.0%) (see Fig. 2).

3.3. Collagen deposition

Masson’s trichrome staining showed less methyl blue staining(collagen fibers) in the granulation tissue of NOX4-/- mice com-pared to the wildtype group. A tendency toward a decreased

Fig. 3. Masson’s trichrome staining. Pictures of wound edges at day 3, 7 and 14 afterE¼epithelium; G¼granulation tissue. Computer-assisted quantification of methyl blue ingroup compared to control mice. Values are expressed as % area of the region of intere

expression was found at day 3 and 7, a significant difference wasseen at day 14 (65.375.4 % of ROI stained in control animals vs19.9716.5% in NOX4-/- animals; po0.05) (see Fig. 3).

3.4. Dityrosine formation

Dityrosine staining revealed a strong expression in wildtypemice in the granulation tissue at day 3 (23.676.6% of ROI stained)while NOX4-deficient mice showed a significantly weaker ex-pression (4.171.3% of ROI stained, po0.05). At day 7, dityrosineexpression increased in wildtype and NOX4-/- mice (34.873.6% vs

wound infliction in wildtype (left column) and NOX4-/- mice (middle column).the granulation tissue showed significantly lower values at day 14 in the NOX4 -/-

st stained by methyl blue and as mean7SD; *po0.05.

Fig. 4. Dityrosine formation. Immunohistochemistry showing dityrosine formation at day 3, 7 and 14 after wound infliction in wildtype (left column) and NOX4-/- mice(middle column). E¼epithelium; G¼granulation tissue. Dityrosine formation appeared weaker in the NOX4-/- group at all time points in the granulation tissue. No dif-ference was observed in the epithelium at the wound edge. Computer-assisted quantification (right column) showed a significant impair of dityrosine formation in NOX4-/-mice in the granulation tissue at day 3, 7 and 14 as compared to wildtype mice. Values are expressed as % area of the region of interest and as mean7SD; *po0.05.


20.474.6% of ROI stained respectively), but the expression re-mained significantly lower in NOX4-/- mice (po0.05). At day 14,the dityrosine expression slightly decreased in both groups(34.178.6% vs 13.576.7% of ROI stained for wildtype and NOX4-/-respectively), the difference between the two groups being againstatistically significant (po0.05) (see Fig. 4).

In the epithelium, dityrosine formation was generally strongeras compared to the granulation tissue. At day 3, wildtype miceshowed 54.5720.5% and NOX4-/- mice 24.8713.4% of ROIstained. At day 7, 48.4711.9% and 67.971.0% of the ROI wasstained and at day 14, 68.878.4% and 47.470.3% respectively forwildtype and NOX4-/- mice. None of the results in the epitheliumwere statistically different between wildtype and NOX4-/- mice.

3.5. NOX2 expression

NOX2 staining showed no difference between wildtype andNOX4-/- mice at day 3 (2.270.6% and 1.770.2% of ROI stainedrespectively). However, a significantly weaker signal was seen atday 7 in NOX4-/- mice (1.670.3% of ROI stained) compared towildtype mice (5.271.5% of ROI stained). At day 14, there wasagain no significant difference found between wildtype andNOX4-/-, as NOX2 expression decreased in wildtype mice to3.070.4% of ROI stained, being similar to NOX4-/- mice at thistime point (1.470.6% of ROI stained). Specificity of the NOX2antibody was tested by staining cultured bone-marrow-deriveddendritic cells from either wildtype or NOX2-/- mice (see Fig. 5).

3.6. MPO expression

No significant difference was found between NOX4-/- mice andwild type mice when staining granulation tissues for MPO. Thesignal was strongest at day 3 (5.674.6% in wildtype and 4.871.1%of ROI stained in NOX4-/- mice) and gradually decreased. At day 7,4.071.4% of ROI was stained in wildtype mice and 2.171.2% inNOX4-/- mice. At day 14, it was 2.371.5% and 1.770.1%respectively.

0.1% of ROI was stained in our negative control (spinal cordtissue of wildtype mice). 4.1% of ROI was stained in our positivecontrol (spinal cord from an amyotrophic lateral sclerosis mousemodel). Our isotype control showed no staining (see Fig. 6).

3.7. Hypoxia inducible factor

HIF1α staining of the granulation tissue at day 3 showed24.877.4% of ROI stained in the control group compared to5.170.2 in the NOX4-/- group (po0.05). At day 14, expressionincreased in both groups, but remained significantly higher in thecontrol group (33.274.6%) as compared to the NOX4-/- group(13.574.2% of ROI stained, po0.05). At day 14, expressionwas thehighest in both groups being 35.470.4% in the control group and27.771.3% in the NOX4-/- group. The difference at day 14 was notstatistically significant (see Fig. 7).

HIF1α expression was stronger in the epithelium at the woundedge as compared to the granulation tissue. The highest expres-sion was found at day 14 with 52.277.6% in the control group and52.8710.9% in the NOX4-/- group. None of the results in theepithelium were statistically different between the control andNOX4-/- group.

3.8. Neoangiogenesis and CD31 expression

CD31 staining was used to study wound angiogenesis. At day 3,low levels of CD31 expression were found (2.171.3 wildtype vs1.771.0 NOX4-/-). A strong increase was observed in control an-imals at day 7, while this increase was less pronounced in NOX4-/-mice (6.670.7% vs 3.370.4% respectively, po0.05). However,CD31 expression strongly increased between day 7 and 14 inNOX4-/- mice while control animals showed only a slight increasein the same period. In fact, NOX4-/- wounds reached similar levelsof expression by day 14 as compared to the control group (7.770.5% control vs 6.37 0.8% NOX4-/-; ns) (see Fig. 8).

Fig. 5. NOX2 expression. Immunohistochemistry showing NOX2 expression at day 3, 7 and 14 after wound infliction in wildtype (left column) and NOX4-/- mice (middlecolumn). E¼epithelium; G¼granulation tissue. NOX2 expression appeared weaker in the NOX4-/- group at day 7. Computer-assisted quantification (right column) showed asignificant impair of dityrosine expression in NOX4-/- mice in the granulation tissue at day 7 as compared to wildtype mice. Isotype control staining showed no signal (rightcolumn, middle picture). NOX2 antibody control staining on cultured bone-marrow-derived dendritic cells from wildtype and NOX2-/- mice showed specific staining (rightlower corner). Values are expressed as % area of the region of interest and as mean7SD; *po0.05.


4. Discussion

Our study shows for the first time that NOX4-deficiency leadsto significantly impaired cutaneous wound repair. Wounds in bothwildtype and NOX4-deficient mice healed predominately

Fig. 6. Myeloperoxidase. Immunohistochemistry showing myeloperoxidase presence at(middle column). E¼epithelium; G¼granulation tissue, a¼ representative view of cell ac3 and continuously decreased over time. Computer-assisted quantification (right columnthe granulation tissue at day 3, 7 and 14 as compared to wildtype mice. The negative cowas stained in positive control tissues (PCO). Isotype control staining was negative (rig

(9372%) through wound contraction, suggesting that the delay inhealing associated with NOX4-deficiency in our model is mainlydue to impaired wound contraction. However, in our study NOX4was not required for myofibroblast differentiation during woundcontraction and αSMA expression patterns were very similar

day 3, 7 and 14 after wound infliction in wildtype (left column) and NOX4-/- micecumulations observed in both groups. Myeloperoxidase was detected as early as day) showed no significant difference in myeloperoxidase formation in NOX4-/- mice inntrol (NCO) showed a signal in 0.1% of the region of interest (ROI) while 4.1% of ROIht lower corner). Values are expressed as % area of ROI and as mean7SD.

Fig. 7. HIF1α expression. Immunohistochemistry showing HIF1α expression at day 3, 7 and 14 after wound infliction in wildtype (left column) and NOX4-/- mice (middlecolumn). E¼epithelium; G¼granulation tissue. HIF1α expression appeared weaker in the NOX4-/- group at all time points in the granulation tissue. No difference was seenin the epithelium at the wound edge. Computer-assisted quantification (right column) showed a significantly lower expression at day 3 and day 7 in the granulation tissue inthe NOX4-/- group compared the control group, no significant difference was seen at day 14. Values are expressed as % area of the region of interest and as mean7SD;*po0.05.


between NOX4-deficient and wildtype mice. This is in contrast toother studies where NOX4 was shown to be a prerequisite formyofibroblast differentiation of cardiac [14] and pulmonary fi-broblasts [15]. Our results suggest that NOX4 is required formyofibroblast differentiation in a situation- and/or tissue-specificmanner.

Fig. 8. CD31 expression. Immunohistochemistry showing CD31 expression at day 3, 7 acolumn). E¼epithelium; G¼granulation tissue; S¼scab. Similar levels of CD31 expresscontrol group which appeared weaker in the NOX4-/- group. Computer-assisted quangranulation tissue in the NOX4-/- group compared the control group. At day 14, woundssimilar levels to the control group. Quantification showed no significant difference at dayand as mean7SD; *po0.05.

The fact that wound contraction was significantly impaireddespite seemingly unchanged myofibroblast expression raises aseries of questions regarding the role of the myofibroblast as themain cell orchestrating wound contraction and regarding the exactmechanisms that lead to the generation of contractile forceswithin the wound bed.

nd 14 after wound infliction in wildtype (left column) and NOX4-/- mice (middleion were seen at day 3 in both groups. A strong increase was seen at day 7 in thetification (right column) showed a significantly lower expression at day 7 in thein NOX4-/- mice showed a strong increase of CD31 expression and had now reached14 between the two groups. Values are expressed as % area of the region of interest


In order to understand why wound contraction is significantlyimpaired in NOX4-deficient mice, we looked for possible ex-planations why myofibroblast contraction could be less effectivedespite preserved αSMA expression. Fully differentiated myofi-broblasts show a strong expression of stress fibers and fibronectin[1,39–41] and are traditionally associated with the expression ofαSMA [42]. The incorporation of αSMA into stress fibers is be-lieved to be the backbone of their contractile activity [43,44].However, recent data suggests that αSMA expression is not a pre-requisite for myofibroblast contraction during wound repair andthat smooth muscle γ-actin and skeletal muscle α-actin cancompletely compensate for a lack of αSMA [45,46]. Possibly, NOX4plays a role in the expression of other muscle actins than αSMA,such as smooth muscle γ-actin or skeletal muscle α-actin inmyofibroblast function during wound repair. If this is the case,NOX4-deficient myofibroblasts could be less contractile despiteexpressing normal amounts of αSMA. Further (in vitro) studiesshould investigate whether NOX4-deficient skin-derived myofi-broblasts show a different pattern of actin expression.

The mechanical properties of the ECM are of primordial im-portance to translate the cellular contraction of myofibroblastsinto a contraction of the wound tissue. While the provisionalmatrix of early granulation tissue is highly compliant, the maturegranulation tissue becomes increasingly stiff over time [47]. Mi-grating fibroblasts remodel the ECM, synthesizing a variety of ECMcomponents, such as collagen and fibronectins [48,49]. It is cur-rently thought that traction forces induced by fibroblast migrationwithin the provisional matrix and their ECM-remodeling activitygradually increase the stiffness of the granulation tissue [1,50–52].NOX4 has been shown to be required for ECM component pro-duction in lung-derived fibroblasts in vitro [15,53]. Targeting NOX4results in attenuation of an established fibrotic response, withreductions in gene transcripts for the extracellular matrix com-ponents collagen 1α1, collagen 3α1, and fibronectin [53]. Accord-ingly, in this study we found that collagen deposition was sig-nificantly reduced in NOX4-deficient mice. A less dense collagenmatrix could explain a lack of translation of the cellular forces intoa contraction of the tissue as a whole. However, a lack of rigidityshould also lead to a reduction in myofibroblast expression, whichis not the case in our study.

In addition to the expression of ECM components, their cross-linking is likely to play an important role in the transduction ofcontractile forces [47,54,55]. This is a little understood processwhere NOX enzymes are particularly likely to play a crucial role. Asoutlined before, we hypothesized that NOX-mediated dityrosinecrosslinking of the ECM is a possible player in wound contraction.In fact, dityrosine formation significantly increased during thewound repair process in our study, suggesting it has a physiolo-gical role. To the best of our knowledge, this study provides thefirst data suggestive of a physiological function of dityrosinecrosslinking in mammals. We could also show that NOX4-defi-ciency leads to significant impairment of dityrosine expression,providing a possible explanation as to how myofibroblast con-traction in NOX4-deficient mice might be less effective in con-tracting the wound. It is conceivable that myofibroblast contrac-tion is only effectively translated into tissue contraction if cells areimbedded in a well cross-linked ECM [56–58].

Tyrosine dimerization is thought to require the presence of ROS(i.e. H2O2) and of a peroxidase [24,59]. In fact, there is strongevidence that peroxidases mediate ECM crosslinking [25,60]. Onepossible source of peroxidase during physiological wound repair ismacrophages and neutrophils as they are known to secrete mye-loperoxidase (MPO), which is mainly known for its role in hostdefense. However, in our study we observe no correlation betweendityrosine formation and MPO expression. Also, there was nodifference in expression between wildtype and NOX4-/- mice,

suggesting that MPO is not the main peroxidase catalyzing dityr-osine formation during wound repair.

A promising candidate for a major role in mediating ECMcrosslinking is vascular peroxidase 1 (VPO1, a.k.a. peroxidasin,PXDN) as it contains, besides its peroxidase domain, modules thatare characteristic of the ECM [61]. VPO1 has also been shown to besecreted by myofibroblasts into the ECM, where it organizes into afibril-like network colocalizing with fibronectin [62]. In fact, Lázáret al. have found that VPO1 mediates the crosslinking of collagenIV in hot spots near the cell surface [63]. It has also been suggestedthat VPO1 catalyzes tyrosyl radical formation and promotes di-tyrosine cross-linking [64]. VPO1 requires H2O2 to function, whichis supplied to the enzyme by a currently unknown cellular source.In a model of hypoxia-induced pulmonary hypertension, Liu et alpropose NOX4 as a provider of hydrogen peroxidase for VPO1during inflammatory reaction [65] but to our knowledge the re-lation between NOX4 and VPO1 is yet to be established. We thinkNOX4 is a promising candidate as a provider of H2O2 for VPO1-mediated collagen crosslinking through tyrosine dimerization,which could be an important mechanism of granulation tissuestiffening during skin wound repair.

Besides tyrosine formation, peroxidases also catalyze otherprotein cross-links, which might also participate along with di-tyrosine to the stiffening of the ECM. For instance, peroxidase-catalyzed cross-link is formed from the deamination of proteinlysyl ϵ-amino groups to form lysyl aldehydes, which then reactwith amino acid residues of adjacent molecules [66,67].

In order to understand whether NOX4 is directly involved inthe dityrosine cross-linking reaction or whether the presence ofdityrosine in the wound could also be explained by a NOX4-de-pendent recruitment and/or activation of phagocytic cells, welooked at NOX2 expression in the granulation tissue. We found asignificantly weaker expression of NOX2 at day 7 in the wounds ofNOX4-/- mice, suggesting that NOX4 plays a role in the recruit-ment of phagocytic cells into the wound. This recruitment couldtake place through NOX4 dependent interleukin-6 (IL-6) expres-sion, similar to what has been shown in human microglia and non-small cell lung cancer cells [56,57]. IL-6 plays an essential role inskin wound repair as evidenced by delayed wound healing in IL-6-deficient mice [58]. These results suggest that NOX4 deficiencyleads to a reduction of dityrosine formation at least partly througha reduction in NOX2-induced ROS in the granulation tissue. NOX2has also been associated with VPO1 expression before and it ispossible that NOX2 provides ROS for VPO1-mediated dityrosineformation [65,68].

As discussed above, HIF1α has been repeatedly associated withNOX4 [53]. In this study we show that NOX4-deficiency is in factlinked to a significant delay in HIF1α expression in the granulationtissue. However, it appears that other pathways can compensatefor the lack of NOX4 as expression was first delayed in theknockout group but eventually reached levels similar to the con-trol group at day 14. Accordingly, we found that angiogenesis asvisualized by CD31 staining was delayed in NOX4-deficient micewith a significantly lower expression at day 7 that eventuallyreached similar levels to control at day 14. This is in line withearlier studies that showed that NOX4 is required for effectiveangiogenesis [28]. NOX4 has also been extensively studied for itsrole as an oxygen sensor and it has been demonstrated that NOX4is capable of generating hydrogen peroxide as a function of oxygenconcentration throughout a physiological range of pO2 values andto respond rapidly to changes in pO₂ [69].

In an earlier study investigating αSMA expression on wrinklingsubstrates over time we showed that hypoxia reduced myofibro-blast contraction from approximately 70 to 15% [70]. It is con-ceivable that a decrease of HIF1α leads to a reduction in woundvascularization and thus oxygenation, triggering the loss of


contractility of myofibroblasts. However, we had also shown that apersistent state of ischemia significantly reduces αSMA expressionin the wound bed, which was not the case in the current study.In vitro, the loss in contractility clearly preceded the loss in αSMAexpression [70]. Possibly, a light decrease in wound oxygenation –as opposed to severe ischemia – leads to a loss of contractility butnot to a loss of αSMA expression in myofibroblasts. Further studiesshould investigate the effects of hypoxia on the expression of ac-tins other than αSMA in myofibroblasts.

Besides wound contraction, re-epithelialization is an importantprocess for effective wound closure. In this study we did not detectany difference in re-epithelialization but it is possible that NOX4also interferes with keratinocyte activity during re-epithelializa-tion as recent data suggests that keratinocytes are producing ROSvia NOX enzymes, including NOX4 [71,72]. It has been shown thatROS at low concentrations induce keratinocyte migration in vitro[73,74]. Matrix metalloproteinase (MMP) production is essentialfor keratinocyte migration as it contributes to cleaving a paththrough the ECM [75,76]. The collagenase MMP-1 is an importantconstituent of the matrix-degrading apparatus of keratinocytesand is expressed in a NOX4-mediated, ROS-dependent way [77].The isoform NOX1 has been studied in epithelial repair of in-testinal mucosa healing and it has been reported that NOX1mediates epithelial migration through activation and modificationof focal adhesion proteins involved in regulating cell migration[78]. Our analyses of the wound do not indicate any major im-pairment of re-epithelialization in NOX4-deficient mice but thewound model used in this study does not seem appropriate tostudy re-epithelialization as wounds heal overwhelminglythrough contraction. More specific analysis should be done in or-der to better understand the repercussions of NOX4 deficiency onre-epithelialization during wound repair, possibly by using asplintered wound model.

5. Conclusions

In this study we show for the first time that NOX4 plays a rolein cutaneous wound repair. We provide evidence that NOX4 pro-motes dityrosine crosslinking of the ECM as well as the recruit-ment of phagocytic cells, as significantly less NOX2-positive cellswere detected in NOX4-/-mice. NOX4-deficiency also led to re-duced HIF1α and CD31 expression, suggesting a role of NOX4 inneoangiogenesis during wound repair. NOX4 was required for ef-fective wound contraction but – surprisingly – not for myofibro-blast expression, challenging the role of the myofibroblast as theprimordial mediator of wound contraction. We suggest thatmyofibroblast contraction in NOX4-deficient mice is less effectivein contracting the wound because of insufficient dityrosinecrosslinking of the ECM. MPO expression was not significantlyaltered by NOX4 deletion, suggesting that other peroxidases areinvolved in dityrosine formation during wound repair, possiblyVPO1. In fact, this study allows us to propose the first physiologicalrole of dityrosine crosslinking in higher animals, as we observed astrong upregulation in wildtype mice during skin wound repair.

Author declaration

We wish to confirm that there are no known conflicts of in-terest associated with this publication and there has been nosignificant financial support for this work that could have influ-enced its outcome. This research was supported by the SwissNational Science Foundation (grant # 310030_120571) (to BrigittePittet-Cuénod).

We confirm that the manuscript has been read and approved

by all named authors and that there are no other persons whosatisfied the criteria for authorship but are not listed. We furtherconfirm that the order of authors listed in the manuscript has beenapproved by all of us.

We confirm that we have given due consideration to the pro-tection of intellectual property associated with this work and thatthere are no impediments to publication, including the timing ofpublication, with respect to intellectual property. In so doing weconfirm that we have followed the regulations of our institutionsconcerning intellectual property.

We further confirm that any aspect of the work covered in thismanuscript that has involved experimental animals has beenconducted with the ethical approval of all relevant bodies and thatsuch approvals are acknowledged within the manuscript.

We understand that the Corresponding Author is the solecontact for the Editorial process (including Editorial Manager anddirect communications with the office). She is responsible forcommunicating with the other authors about progress, submis-sions of revisions and final approval of proofs. We confirm that wehave provided a current, correct email address which is accessibleby the Corresponding Author and which has been configured toaccept email from [email protected].

References

[1] J.J. Tomasek, G. Gabbiani, B. Hinz, C. Chaponnier, R.A. Brown, Myofibroblastsand mechano-regulation of connective tissue remodelling, Nat. Rev. Mol. CellBiol. 3 (2002) 349–363.

[2] G. Gabbiani, G.B. Ryan, G. Majne, Presence of modified fibroblasts in granu-lation tissue and their possible role in wound contraction, Experientia 27(1971) 549–550.

[3] F. Klingberg, B. Hinz, E.S. White, The myofibroblast matrix: implications fortissue repair and fibrosis, J. Pathol. 229 (2013) 298–309.

[4] R. Rathore, Y.M. Zheng, C.F. Niu, Q.H. Liu, A. Korde, Y.S. Ho, Y.X. Wang, Hypoxiaactivates NADPH oxidase to increase [ROS]i and [Ca2þ]i through the mi-tochondrial ROS-PKCepsilon signaling axis in pulmonary artery smooth mus-cle cells, Free Radic. Biol. Med. 45 (2008) 1223–1231.

[5] K. Bedard, K.H. Krause, The NOX family of ROS-generating NADPH oxidases:physiology and pathophysiology, Physiol. Rev. 87 (2007) 245–313.

[6] S. Roy, S. Khanna, K. Nallu, T.K. Hunt, C.K. Sen, Dermal wound healing issubject to redox control, Mol. Ther.: J. Am. Soc. Gene Ther. 13 (2006) 211–220.

[7] C.K. Sen, The general case for redox control of wound repair, Wound RepairRegen. 11 (2003) 431–438.

[8] M. Schafer, S. Werner, Oxidative stress in normal and impaired wound repair,Pharmacol. Res. 58 (2008) 165–171.

[9] D.B. Allen, J.J. Maguire, M. Mahdavian, C. Wicke, L. Marcocci, H. Scheuenstuhl,M. Chang, A.X. Le, H.W. Hopf, T.K. Hunt, Wound hypoxia and acidosis limitneutrophil bacterial killing mechanisms, Arch. Surg. 132 (1997) 991–996.

[10] A. Gorlach, R.P. Brandes, S. Bassus, N. Kronemann, C.M. Kirchmaier, R. Busse, V.B. Schini-Kerth, Oxidative stress and expression of p22phox are involved inthe up-regulation of tissue factor in vascular smooth muscle cells in responseto activated platelets, FASEB J. 14 (2000) 1518–1528.

[11] T. Inoguchi, T. Sonta, H. Tsubouchi, T. Etoh, M. Kakimoto, N. Sonoda, N. Sato,N. Sekiguchi, K. Kobayashi, H. Sumimoto, H. Utsumi, H. Nawata, Protein kinaseC-dependent increase in reactive oxygen species (ROS) production in vasculartissues of diabetes: role of vascular NAD(P)H oxidase, J. Am. Soc. Nephrol. 14(2003) S227–S232.

[12] J.W. Eckert, S.L. Abramson, J. Starke, M.L. Brandt, The surgical implications ofchronic granulomatous disease, Am. J. Surg. 169 (1995) 320–323.

[13] V.J. Thannickal, B.L. Fanburg, Activation of an H2O2-generating NADH oxidasein human lung fibroblasts by transforming growth factor beta 1, J. Biol. Chem.270 (1995) 30334–30338.

[14] I. Cucoranu, R. Clempus, A. Dikalova, P.J. Phelan, S. Ariyan, S. Dikalov,D. Sorescu, NAD(P)H oxidase 4 mediates transforming growth factor-beta1-induced differentiation of cardiac fibroblasts into myofibroblasts, Circ. Res. 97(2005) 900–907.

[15] L. Hecker, R. Vittal, T. Jones, R. Jagirdar, T.R. Luckhardt, J.C. Horowitz,S. Pennathur, F.J. Martinez, V.J. Thannickal, NADPH oxidase-4 mediates myo-fibroblast activation and fibrogenic responses to lung injury, Nat. Med. 15(2009) 1077–1081.

[16] A. Sturrock, B. Cahill, K. Norman, T.P. Huecksteadt, K. Hill, K. Sanders, S.V. Karwande, J.C. Stringham, D.A. Bull, M. Gleich, T.P. Kennedy, J.R. Hoidal,Transforming growth factor-beta1 induces Nox4 NAD(P)H oxidase and re-active oxygen species-dependent proliferation in human pulmonary arterysmooth muscle cells, Am. J. Physiol. Lung Cell. Mol. Physiol. 290 (2006)L661–L673.

[17] S. Carnesecchi, C. Deffert, Y. Donati, O. Basset, B. Hinz, O. Preynat-Seauve,

http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref1http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref1http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref1http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref1http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref2http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref2http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref2http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref2http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref3http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref3http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref3http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref4http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref5http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref5http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref5http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref6http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref6http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref6http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref7http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref7http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref7http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref8http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref8http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref8http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref9http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref9http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref9http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref9http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref10http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref10http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref10http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref10http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref10http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref11http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref11http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref11http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref11http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref11http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref11http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref12http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref12http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref12http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref13http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref14http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref14http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref14http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref14http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref14http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref15http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref15http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref15http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref15http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref15http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref16http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref17


C. Guichard, J.L. Arbiser, B. Banfi, J.C. Pache, C. Barazzone-Argiroffo, K.H. Krause, A key role for NOX4 in epithelial cell death during development oflung fibrosis, Antioxid. Redox Signal. 15 (2011) 607–619.

[18] J. Gauldie, P. Bonniaud, P. Sime, K. Ask, M. Kolb, TGF-beta, Smad3 and theprocess of progressive fibrosis, Biochem. Soc. Trans. 35 (2007) 661–664.

[19] R.A. Rahimi, E.B. Leof, TGF-beta signaling: a tale of two responses, J. Cell.Biochem. 102 (2007) 593–608.

[20] A. Desmouliere, A. Geinoz, F. Gabbiani, G. Gabbiani, Transforming growthfactor-beta 1 induces alpha-smooth muscle actin expression in granulationtissue myofibroblasts and in quiescent and growing cultured fibroblasts, J. CellBiol. 122 (1993) 103–111.

[21] F. LaBella, F. Keeley, S. Vivian, D. Thornhill, Evidence for dityrosine in elastin,Biochem. Biophys. Res. Commun. 26 (1967) 748–753.

[22] F. LaBella, P. Waykole, G. Queen, Formation of insoluble gels and dityrosine bythe action of peroxidase on soluble collagens, Biochem. Biophys. Res. Com-mun. 30 (1968) 333–338.

[23] G.A. Finlay, V.J. Thannickal, B.L. Fanburg, K.E. Paulson, Transforming growthfactor-beta 1-induced activation of the ERK pathway/activator protein-1 inhuman lung fibroblasts requires the autocrine induction of basic fibroblastgrowth factor, J. Biol. Chem. 275 (2000) 27650–27656.

[24] J.M. Larios, R. Budhiraja, B.L. Fanburg, V.J. Thannickal, Oxidative protein cross-linking reactions involving L-tyrosine in transforming growth factor-beta1-stimulated fibroblasts, J. Biol. Chem. 276 (2001) 17437–17441.

[25] W.A. Edens, L. Sharling, G. Cheng, R. Shapira, J.M. Kinkade, T. Lee, H.A. Edens,X. Tang, C. Sullards, D.B. Flaherty, G.M. Benian, J.D. Lambeth, Tyrosine cross-linking of extracellular matrix is catalyzed by Duox, a multidomain oxidase/peroxidase with homology to the phagocyte oxidase subunit gp91phox, J. Cell.Biol. 154 (2001) 879–891.

[26] J.L. Wong, R. Creton, G.M. Wessel, The oxidative burst at fertilization is de-pendent upon activation of the dual oxidase Udx1, Dev. Cell 7 (2004) 801–814.

[27] N.T. Anh, M. Nishitani, S. Harada, M. Yamaguchi, K. Kamei, Essential role ofDuox in stabilization of Drosophila wing, J. Biol. Chem. 286 (2011)33244–33251.

[28] H.M. Peshavariya, E.C. Chan, G.S. Liu, F. Jiang, G.J. Dusting, Transforminggrowth factor-beta1 requires NADPH oxidase 4 for angiogenesis in vitro andin vivo, J. Cell. Mol. Med. (2014).

[29] M. Mittal, M. Roth, P. Konig, S. Hofmann, E. Dony, P. Goyal, A.C. Selbitz, R.T. Schermuly, H.A. Ghofrani, G. Kwapiszewska, W. Kummer, W. Klepetko, M.A. Hoda, L. Fink, J. Hanze, W. Seeger, F. Grimminger, H.H. Schmidt,N. Weissmann, Hypoxia-dependent regulation of nonphagocytic NADPH oxi-dase subunit NOX4 in the pulmonary vasculature, Circ. Res. 101 (2007)258–267.

[30] S. Ismail, A. Sturrock, P. Wu, B. Cahill, K. Norman, T. Huecksteadt, K. Sanders,T. Kennedy, J. Hoidal, NOX4 mediates hypoxia-induced proliferation of humanpulmonary artery smooth muscle cells: the role of autocrine production oftransforming growth factor-{beta}1 and insulin-like growth factor bindingprotein-3, Am. J. Physiol. Lung Cell. Mol. Physiol. 296 (2009) L489–L499.

[31] N.S. Chandel, D.S. McClintock, C.E. Feliciano, T.M. Wood, J.A. Melendez, A.M. Rodriguez, P.T. Schumacker, Reactive oxygen species generated at mi-tochondrial complex III stabilize hypoxia-inducible factor-1alpha during hy-poxia: a mechanism of O2 sensing, J. Biol. Chem. 275 (2000) 25130–25138.

[32] E. Minet, G. Michel, D. Mottet, M. Raes, C. Michiels, Transduction pathwaysinvolved in Hypoxia-Inducible Factor-1 phosphorylation and activation, FreeRadic. Biol. Med. 31 (2001) 847–855.

[33] S. Bonello, C. Zahringer, R.S. BelAiba, T. Djordjevic, J. Hess, C. Michiels,T. Kietzmann, A. Gorlach, Reactive oxygen species activate the HIF-1alphapromoter via a functional NFkappaB site, Arterioscler., Thromb. Vasc. Biol. 27(2007) 755–761.

[34] I. Diebold, A. Petry, J. Hess, A. Gorlach, The NADPH oxidase subunit NOX4 is anew target gene of the hypoxia-inducible factor-1, Mol. Biol. Cell 21 (2010)2087–2096.

[35] C.K. Sen, S. Roy, Redox signals in wound healing, Biochim. Biophys. Acta 1780(2008) 1348–1361.

[36] H. Ashino, M. Shimamura, H. Nakajima, M. Dombou, S. Kawanaka, T. Oikawa,T. Iwaguchi, S. Kawashima, Novel function of ascorbic acid as an angiostaticfactor, Angiogenesis 6 (2003) 259–269.

[37] M.T. Lin, M.L. Yen, C.Y. Lin, M.L. Kuo, Inhibition of vascular endothelial growthfactor-induced angiogenesis by resveratrol through interruption of Src-de-pendent vascular endothelial cadherin tyrosine phosphorylation, Mol. Phar-macol. 64 (2003) 1029–1036.

[38] J.C. Fleming, M.D. Norenberg, D.A. Ramsay, G.A. Dekaban, A.E. Marcillo, A.D. Saenz, M. Pasquale-Styles, W.D. Dietrich, L.C. Weaver, The cellular in-flammatory response in human spinal cords after injury, Brain 129 (2006)3249–3269.

[39] R.A. Ignotz, J. Massague, Transforming growth factor-beta stimulates the ex-pression of fibronectin and collagen and their incorporation into the extra-cellular matrix, J. Biol. Chem. 261 (1986) 4337–4345.

[40] S. Itoh, F. Itoh, M.J. Goumans, P. Ten Dijke, Signaling of transforming growthfactor-beta family members through Smad proteins, Eur. J. Biochem./ FEBS 267(2000) 6954–6967.

[41] I.I. Singer, D.M. Kazazis, D.W. Kawka, Localization of the fibronexus at thesurface of granulation tissue myofibroblasts using double-label immunogoldelectron microscopy on ultrathin frozen sections, Eur. J. Cell Biol. 38 (1985)94–101.

[42] B. Hinz, V. Dugina, C. Ballestrem, B. Wehrle-Haller, C. Chaponnier, Alpha-smooth muscle actin is crucial for focal adhesion maturation in

myofibroblasts, Mol. Biol. Cell 14 (2003) 2508–2519.[43] I. Darby, O. Skalli, G. Gabbiani, Alpha-smooth muscle actin is transiently ex-

pressed by myofibroblasts during experimental wound healing, Lab. Investig.J. Techn. Methods Pathol. 63 (1990) 21–29.

[44] B. Hinz, D. Mastrangelo, C.E. Iselin, C. Chaponnier, G. Gabbiani, Mechanicaltension controls granulation tissue contractile activity and myofibroblast dif-ferentiation, Am. J. Pathol. 159 (2001) 1009–1020.

[45] J.J. Tomasek, C.J. Haaksma, R.J. Schwartz, E.W. Howard, Whole animal knock-out of smooth muscle alpha-actin does not alter excisional wound healing orthe fibroblast-to-myofibroblast transition, Wound Repair Regen. 21 (2013)166–176.

[46] K. Au, H.P. Ehrlich, When the Smad signaling pathway is impaired, fibroblastsadvance open wound contraction, Exp. Mol. Pathol. 89 (2010) 236–240.

[47] B. Hinz, The myofibroblast: paradigm for a mechanically active cell, J. Biomech.43 (2010) 146–155.

[48] P. Martin, Wound healing—aiming for perfect skin regeneration, Science 276(1997) 75–81.

[49] N. Vedrenne, B. Coulomb, A. Danigo, F. Bonte, A. Desmouliere, The complexdialogue between (myo)fibroblasts and the extracellular matrix during skinrepair processes and ageing, Pathol-Biol. 60 (2012) 20–27.

[50] H.P. Ehrlich, T.K. Hunt, Collagen organization critical role in wound contrac-tion, Adv. Wound Care 1 (2012) 3–9.

[51] J.M. Goffin, P. Pittet, G. Csucs, J.W. Lussi, J.J. Meister, B. Hinz, Focal adhesion sizecontrols tension-dependent recruitment of alpha-smooth muscle actin tostress fibers, J. Cell Biol. 172 (2006) 259–268.

[52] H.P. Ehrlich, J.B. Rajaratnam, Cell locomotion forces versus cell contractionforces for collagen lattice contraction: an in vitro model of wound contraction,Tissue Cell 22 (1990) 407–417.

[53] E.R. Jarman, V.S. Khambata, C. Cope, P. Jones, J. Roger, L.Y. Ye, N. Duggan,D. Head, A. Pearce, N.J. Press, B. Bellenie, B. Sohal, G. Jarai, An inhibitor ofNADPH oxidase-4 attenuates established pulmonary fibrosis in a rodent dis-ease model, Am. J. Respir. Cell Mol. Biol. 50 (2014) 158–169.

[54] H.M. Frost, A 2003 update of bone physiology and Wolff's Law for clinicians,Angle Orthod. 74 (2004) 3–15.

[55] G. Garcia-Cardena, J. Comander, K.R. Anderson, B.R. Blackman, M.A. GimbroneJr., Biomechanical activation of vascular endothelium as a determinant of itsfunctional phenotype, Proc. Natl. Acad. Sci. USA 98 (2001) 4478–4485.

[56] B. Li, K. Bedard, S. Sorce, B. Hinz, M. Dubois-Dauphin, K.H. Krause, NOX4 ex-pression in human microglia leads to constitutive generation of reactiveoxygen species and to constitutive IL-6 expression, J. Innate Immun. 1 (2009)570–581.

[57] J. Li, T. Lan, C. Zhang, C. Zeng, J. Hou, Z. Yang, M. Zhang, J. Liu, B. Liu, Reciprocalactivation between IL-6/STAT3 and NOX4/Akt signalings promotes prolifera-tion and survival of non-small cell lung cancer cells, Oncotarget 6 (2015)1031–1048.

[58] Z.Q. Lin, T. Kondo, Y. Ishida, T. Takayasu, N. Mukaida, Essential involvement ofIL-6 in the skin wound-healing process as evidenced by delayed woundhealing in IL-6-deficient mice, J. Leukoc. Biol. 73 (2003) 713–721.

[59] O.M. Lardinois, K.F. Medzihradszky, P.R. Ortiz de Montellano, Spin trappingand protein cross-linking of the lactoperoxidase protein radical, J. Biol. Chem.274 (1999) 35441–35448.

[60] J.W. Heinecke, B.M. Shapiro, The respiratory burst oxidase of fertilization. Aphysiological target for regulation by protein kinase C, J. Biol. Chem. 267(1992) 7959–7962.

[61] R.E. Nelson, L.I. Fessler, Y. Takagi, B. Blumberg, D.R. Keene, P.F. Olson, C.G. Parker, J.H. Fessler, Peroxidasin: a novel enzyme-matrix protein of Droso-phila development, EMBO J. 13 (1994) 3438–3447.

[62] Z. Peterfi, A. Donko, A. Orient, A. Sum, A. Prokai, B. Molnar, Z. Vereb,E. Rajnavolgyi, K.J. Kovacs, V. Muller, A.J. Szabo, M. Geiszt, Peroxidasin is se-creted and incorporated into the extracellular matrix of myofibroblasts andfibrotic kidney, Am. J. Pathol. 175 (2009) 725–735.

[63] E. Lazar, Z. Peterfi, G. Sirokmany, H.A. Kovacs, E. Klement, K.F. Medzihradszky,M. Geiszt, Structure-function analysis of peroxidasin provides insight into themechanism of collagen IV crosslinking, Free Radic. Biol. Med. 83 (2015)273–282.

[64] G. Cheng, H. Li, Z. Cao, X. Qiu, S. McCormick, V.J. Thannickal, W.M. Nauseef,Vascular peroxidase-1 is rapidly secreted, circulates in plasma, and supportsdityrosine cross-linking reactions, Free Radic. Biol. Med. 51 (2011) 1445–1453.

[65] B. Liu, X.J. Luo, Z.B. Yang, J.J. Zhang, T.B. Li, X.J. Zhang, Q.L. Ma, G.G. Zhang, C.P. Hu, J. Peng, Inhibition of NOX/VPO1 pathway and inflammatory reaction bytrimethoxystilbene in prevention of cardiovascular remodeling in hypoxia-induced pulmonary hypertensive rats, J. Cardiovasc. Pharmacol. 63 (2014)567–576.

[66] R.A. Clark, S. Szot, M.A. Williams, H.M. Kagan, Oxidation of lysine side-chainsof elastin by the myeloperoxidase system and by stimulated human neu-trophils, Biochem. Biophys. Res. Commun. 135 (1986) 451–457.

[67] S.L. Hazen, J.P. Gaut, F.F. Hsu, J.R. Crowley, A. d'Avignon, J.W. Heinecke, p-Hy-droxyphenylacetaldehyde, the major product of L-tyrosine oxidation by themyeloperoxidase-H2O2-chloride system of phagocytes, covalently modifiesepsilon-amino groups of protein lysine residues, J. Biol. Chem. 272 (1997)16990–16998.

[68] Y.S. Zhang, L. He, B. Liu, N.S. Li, X.J. Luo, C.P. Hu, Q.L. Ma, G.G. Zhang, Y.J. Li,J. Peng, A novel pathway of NADPH oxidase/vascular peroxidase 1 in med-iating oxidative injury following ischemia-reperfusion, Basic Res. Cardiol. 107(2012) 266.

[69] Y. Nisimoto, B.A. Diebold, D. Cosentino-Gomes, J.D. Lambeth, Nox4: a

http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref17http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref17http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref17http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref17http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref18http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref18http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref18http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref19http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref19http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref19http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref20http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref20http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref20http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref20http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref20http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref21http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref21http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref21http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref22http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref22http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref22http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref22http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref23http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref23http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref23http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref23http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref23http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref24http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref24http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref24http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref24http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref25http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref25http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref25http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref25http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref25http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref25http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref26http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref26http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref26http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref27http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref27http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref27http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref27http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref28http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref28http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref28http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref29http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref30http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref30http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref30http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref30http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref30http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref30http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref31http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref32http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref32http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref32http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref32http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref33http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref33http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref33http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref33http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref33http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref34http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref34http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref34http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref34http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref35http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref35http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref35http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref36http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref36http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref36http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref36http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref37http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref37http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref37http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref37http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref37http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref38http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref38http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref38http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref38http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref38http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref39http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref39http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref39http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref39http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref40http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref40http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref40http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref40http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref41http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref41http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref41http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref41http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref41http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref42http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref42http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref42http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref42http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref43http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref43http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref43http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref43http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref44http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref44http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref44http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref44http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref45http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref45http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref45http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref45http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref45http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref46http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref46http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref46http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref47http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref47http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref47http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref48http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref48http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref48http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref49http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref49http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref49http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref49http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref50http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref50http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref50http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref51http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref51http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref51http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref51http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref52http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref52http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref52http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref52http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref53http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref53http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref53http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref53http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref53http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref54http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref54http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref54http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref55http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref55http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref55http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref55http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref56http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref56http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref56http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref56http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref56http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref57http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref57http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref57http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref57http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref57http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref58http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref58http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref58http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref58http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref59http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref59http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref59http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref59http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref60http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref60http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref60http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref60http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref61http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref61http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref61http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref61http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref62http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref62http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref62http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref62http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref62http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref63http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref63http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref63http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref63http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref63http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref64http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref64http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref64http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref64http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref65http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref65http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref65http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref65http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref65http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref65http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref66http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref66http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref66http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref66http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref67http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref67http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref67http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref67http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref67http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref67http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref68http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref68http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref68http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref68http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref69


hydrogen peroxide-generating oxygen sensor, Biochemistry 53 (2014)5111–5120.

[70] A. Modarressi, G. Pietramaggiori, C. Godbout, E. Vigato, B. Pittet, B. Hinz, Hy-poxia impairs skin myofibroblast differentiation and function, J. Invest. Der-matol. 130 (2010) 2818–2827.

[71] A. Stanley, A. Hynes, C. Brakebusch, F. Quondamatteo, Rho GTPases and Noxdependent ROS production in skin. Is there a connection? Histol. Histopathol.27 (2012) 1395–1406.

[72] H.J. Nam, Y.Y. Park, G. Yoon, H. Cho, J.H. Lee, Co-treatment with hepatocytegrowth factor and TGF-beta1 enhances migration of HaCaT cells throughNADPH oxidase-dependent ROS generation, Exp. Mol. Med. 42 (2010)270–279.

[73] S. Yang, B.J. Misner, R.J. Chiu, F.L. Meyskens Jr., Redox effector factor-1, com-bined with reactive oxygen species, plays an important role in the transfor-mation of JB6 cells, Carcinogenesis 28 (2007) 2382–2390.

[74] Y.M. Go, T.R. Ziegler, J.M. Johnson, L. Gu, J.M. Hansen, D.P. Jones, Selectiveprotection of nuclear thioredoxin-1 and glutathione redox systems against

oxidation during glucose and glutamine deficiency in human colonic epithelialcells, Free Radic. Biol. Med. 42 (2007) 363–370.

[75] P.F. Ridgway, P. Ziprin, D.H. Peck, A.W. Darzi, Hypoxia increases re-epithelialization via an alphavbeta6-dependent pathway, Wound Repair Re-gen. 13 (2005) 158–164.

[76] L.R. Lund, K.A. Green, A.A. Stoop, M. Ploug, K. Almholt, J. Lilla, B.S. Nielsen, I.J. Christensen, C.S. Craik, Z. Werb, K. Dano, J. Romer, Plasminogen activationindependent of uPA and tPA maintains wound healing in gene-deficient mice,EMBO J. 25 (2006) 2686–2697.

[77] L. Grange, M.V. Nguyen, B. Lardy, M. Derouazi, Y. Campion, C. Trocme, M.H. Paclet, P. Gaudin, F. Morel, NAD(P)H oxidase activity of Nox4 in chon-drocytes is both inducible and involved in collagenase expression, Antioxid.Redox Signal. 8 (2006) 1485–1496.

[78] G. Leoni, A. Alam, P.A. Neumann, J.D. Lambeth, G. Cheng, J. McCoy, R.S. Hilgarth, K. Kundu, N. Murthy, D. Kusters, C. Reutelingsperger, M. Perretti, C.A. Parkos, A.S. Neish, A. Nusrat, Annexin A1, formyl peptide receptor, andNOX1 orchestrate epithelial repair, J. Clin. Invest. 123 (2013) 443–454.

http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref69http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref69http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref69http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref70http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref70http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref70http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref70http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref71http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref71http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref71http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref71http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref72http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref72http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref72http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref72http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref72http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref73http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref73http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref73http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref73http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref74http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref74http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref74http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref74http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref74http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref75http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref75http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref75http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref75http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref76http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref76http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref76http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref76http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref76http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref77http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref77http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref77http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref77http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref77http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref78http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref78http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref78http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref78http://refhub.elsevier.com/S0891-5849(16)30212-X/sbref78

NADPH oxidase 4 deficiency leads to impaired wound repair and reduced dityrosine-crosslinking, but does not affect...IntroductionMaterial and methodsAnimal wound modelWound repair assessmentHistology and immunohistochemistryImage analysisStatistical methods

ResultsMacroscopic analysisMyofibroblast expressionCollagen depositionDityrosine formationNOX2 expressionMPO expressionHypoxia inducible factorNeoangiogenesis and CD31 expression

DiscussionConclusionsAuthor declarationReferences

free radical biology and medicine - connecting repositories · 2016. 12. 7. · proteins. both...

Documents