Advertisement

Hypoxia Impairs Skin Myofibroblast Differentiation and Function

  • Author Footnotes
    4 These authors contributed equally to this work.
    Ali Modarressi
    Footnotes
    4 These authors contributed equally to this work.
    Affiliations
    Department of Plastic, Reconstructive and Aesthetic Surgery, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

    Laboratory of Cell Biophysics, School of Basic Sciences, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
    Search for articles by this author
  • Author Footnotes
    4 These authors contributed equally to this work.
    Giorgio Pietramaggiori
    Footnotes
    4 These authors contributed equally to this work.
    Affiliations
    Department of Plastic, Reconstructive and Aesthetic Surgery, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

    Laboratory of Cell Biophysics, School of Basic Sciences, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
    Search for articles by this author
  • Charles Godbout
    Affiliations
    Matrix Dynamics Group, Laboratory of Tissue Repair and Regeneration, Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada
    Search for articles by this author
  • Enrico Vigato
    Affiliations
    Department of Plastic, Reconstructive and Aesthetic Surgery, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland

    Laboratory of Cell Biophysics, School of Basic Sciences, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
    Search for articles by this author
  • Brigitte Pittet
    Correspondence
    Service de Chirurgie Plastique et Reconstructive, Département de Chirurgie, Faculté de Médecine, Université de Genève, Rue Gabrielle-Perret-Gentil 4, 1211 Genève 14, Switzerland
    Affiliations
    Department of Plastic, Reconstructive and Aesthetic Surgery, University Hospitals of Geneva and University of Geneva, Geneva, Switzerland
    Search for articles by this author
  • Boris Hinz
    Correspondence
    Matrix Dynamics Group, Laboratory of Tissue Repair and Regeneration, Faculty of Dentistry, University of Toronto, Fitzgerald Building, Room 241, 150 College Street, Toronto, Ontario M5S 3E2, Canada
    Affiliations
    Matrix Dynamics Group, Laboratory of Tissue Repair and Regeneration, Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada
    Search for articles by this author
  • Author Footnotes
    4 These authors contributed equally to this work.
      Ischemic wounds are characterized by oxygen levels lower than that of healthy skin (hypoxia) and poor healing. To better understand the pathophysiology of impaired wound healing, we investigated how switching from high (21%) to low (2%) oxygen levels directly affects cultured skin myofibroblasts, essential cells for the normal wound repair process. Myofibroblast differentiation and function were assessed by quantifying α-smooth muscle actin expression and cell contraction in collagen gels and on wrinkling silicone substrates. Culture for 5 days at 2% oxygen is perceived as hypoxia and significantly reduced myofibroblast differentiation and contraction despite high levels of the profibrotic transforming growth factor-β1. Analysis of α-smooth muscle actin expression on wrinkling substrates over time showed that reduced myofibroblast contraction preceded α-smooth muscle actin disassembly from stress fibers after switching from 21 to 2% oxygen. These effects were reversible by restoring high oxygen conditions and by applying mechanical stress. We suggest that mechanical challenge is a clinical relevant strategy to improve ischemic and chronic wound healing by supporting myofibroblast formation.

      Abbreviations

      α-SMA
      α-smooth muscle actin
      HIF-1α
      hypoxia-inducible transcription factor-1α
      PAI-1
      plasminogen activator inhibitor-1
      pO2
      O2 pressure
      SCF
      subcutaneous fibroblast
      TGFβ1
      transforming growth factor-β1
      TGFβ-RII
      TGFβ-receptor type II
      TMLC
      transformed mink lung epithelial cells

      Introduction

      An estimated 2% of the patients in Western countries suffer from chronic wounds (
      • Pelka R.
      [The economic situation of chronic wounds].
      ;
      • Clark R.A.
      • Ghosh K.
      • Tonnesen M.G.
      Tissue engineering for cutaneous wounds.
      ), causing mortality and morbidity in all age groups and representing an important cost factor for the global health-care system (
      • Banwell P.E.
      Topical negative pressure therapy in wound care.
      ). Diabetic patients, the physically disabled, the elderly, and vasculopathic patients are predisposed to the development of impaired healing (
      • Margolis D.J.
      • Bilker W.
      • Knauss J.
      • et al.
      The incidence and prevalence of pressure ulcers among elderly patients in general medical practice.
      ;
      • Falanga V.
      Wound healing and its impairment in the diabetic foot.
      ). Most risk factors for chronic wounds are associated with local ischemia and decreased tissue O2 pressure (pO2), a condition called hypoxia (
      • Tandara A.A.
      • Mustoe T.A.
      Oxygen in wound healing—more than a nutrient.
      ;
      • Sen C.K.
      Wound healing essentials: let there be oxygen.
      ). Although acute ischemia promotes inflammation and angiogenesis at the onset of wound healing, the persistence of hypoxia can dramatically impair physiological healing (
      • Scheid A.
      • Wenger R.H.
      • Christina H.
      • et al.
      Hypoxia-regulated gene expression in fetal wound regeneration and adult wound repair.
      ;
      • Sen C.K.
      • Khanna S.
      • Gordillo G.
      • et al.
      Oxygen, oxidants, and antioxidants in wound healing: an emerging paradigm.
      ;
      • Tandara A.A.
      • Mustoe T.A.
      Oxygen in wound healing—more than a nutrient.
      ).
      During normal skin wound healing, granulation tissue temporarily fills the tissue defect and provides a scaffold for repopulating cells. Among these, myofibroblasts actively remodel the granulation tissue and contract the wound (
      • Hinz B.
      Formation and function of the myofibroblast during tissue repair.
      ). Local dermal fibroblasts and precursor cells from other sources differentiate into myofibroblasts by acquiring de novo α-smooth muscle actin (α-SMA), conferring to these cells high contractile activity (
      • Hinz B.
      • Celetta G.
      • Tomasek J.J.
      • et al.
      Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
      ,
      • Hinz B.
      • Phan S.H.
      • Thannickal V.J.
      • et al.
      The myofibroblast: one function, multiple origins.
      ;
      • Gabbiani G.
      The myofibroblast in wound healing and fibrocontractive diseases.
      ). Soluble growth factors, in particular transforming growth factor-β1 (TGFβ1), and mechanical stress are prerequisites to induce myofibroblast differentiation (
      • Hinz B.
      • Gabbiani G.
      Mechanisms of force generation and transmission by myofibroblasts.
      ;
      • Wang J.
      • Zohar R.
      • McCulloch C.A.
      Multiple roles of alpha-smooth muscle actin in mechanotransduction.
      ). When the injured skin is repaired, myofibroblasts disappear through apoptosis, leaving behind a relatively avascular and collagen-dense scar tissue (
      • Desmouliere A.
      • Redard M.
      • Darby I.
      • et al.
      Apoptosis mediates the decrease in cellularity during the transition between granulation tissue and scar.
      ).
      Using an experimental rat foot wound model, we have recently shown delayed wound closure in chronic ischemic conditions. In this model of ischemic limb by total resection of the external iliac artery, decreased ischemic wound contraction correlated with reduced myofibroblast development (
      • Alizadeh N.
      • Pepper M.S.
      • Modarressi A.
      • et al.
      Persistent ischemia impairs myofibroblast development in wound granulation tissue: a new model of delayed wound healing.
      ). Several mechanisms have been suggested to decrease myofibroblast occurrence in nonhealing wounds, including reduced TGFβ1 expression and/or activation, impaired angiogenesis, and delayed revascularization (
      • Ahn S.T.
      • Mustoe T.A.
      Effects of ischemia on ulcer wound healing: a new model in the rabbit ear.
      ;
      • Wu L.
      • Xia Y.P.
      • Roth S.I.
      • et al.
      Transforming growth factor-beta1 fails to stimulate wound healing and impairs its signal transduction in an aged ischemic ulcer model: importance of oxygen and age.
      ;
      • Tandara A.A.
      • Mustoe T.A.
      Oxygen in wound healing—more than a nutrient.
      ). The objective of our study was to elucidate how hypoxia directly modulates skin fibroblast-to-myofibroblast differentiation with the aim of improving ischemic wound healing.

      Results

      Switch to low O2 increases proliferation of rat subcutaneous fibroblasts

      Subcutaneous fibroblasts (SCFs) contribute to the healing of full-thickness wounds in rats (
      • Hinz B.
      • Mastrangelo D.
      • Iselin C.E.
      • et al.
      Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation.
      ). To evaluate the effect of O2 deprivation on cultured SCFs, we switched cells that had been passaged at least two times under environmental O2 (21%) to low O2 for 5 days. To determine the minimal O2 change that is perceived as hypoxia by SCFs, we quantified expression of the heterodimeric hypoxia-inducible transcription factor-1α (HIF-1α;
      • Semenza G.L.
      Hypoxia-inducible factor 1 (HIF-1) pathway.
      ) in 21, 10, 5, and 2% O2. HIF-1α expression by SCFs increased after switching cells from 21 to 5% O2 and to 2% O2 (Figure 1a). This switch did not reduce SCF viability; apoptosis rates were reduced 2-fold (Figure 1b), consistent with 1.8-fold increased cell numbers in 2% O2. We performed all further experiments with 2% O2, referred to as “low O2”.
      Figure thumbnail gr1
      Figure 1Switching to low O2 affects cultured rat subcutaneous fibroblasts (SCFs). SCFs were grown for 5 days in 21% O2 (standard culture), and 10, 5, and 2% O2. (a) Expression of α-smooth muscle actin (α-SMA) and hypoxia-inducible transcription factor-1α (HIF-1α) assessed by western blot, normalized to housekeeping vimentin for densitometric band analysis. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as additional loading control. (b) Cell number was quantified by cell counting; viability was assessed using a colorimetric cell viability assay, and apoptosis with a caspase-2/3 activity kit. Viability and apoptosis were compared between similar cell numbers at the time of the test. Mean values were calculated from at least three independent experiments performed with five samples per condition. Immunofluorescence staining of SCFs in (c) 21% high O2 and (d) 2% low O2 for α-SMA (red), F-actin (phalloidin, green), and nuclei (blue) was used to quantify the percentage of α-SMA-positive myofibroblasts by automated image analysis (e). Scale bar=20μm. Error bars indicate SD of mean. *P<0.01 tested by Student's t-tests against control 21% O2.

      Switch to low O2 inhibits rat SCF-to-myofibroblast differentiation and contraction

      We sought to determine whether the switch from high to low O2 can directly modulate SCF-to-myofibroblast differentiation. In 21% O2 standard culture, ∼25% of SCFs spontaneously express the myofibroblast marker α-SMA (Figure 1c and e). After switching for 5 days to 2% O2, the number of α-SMA-positive cells decreased fivefold (Figure 1d and e), and levels of α-SMA expression were reduced (Figure 1a). We further tested the effect of the low-O2 switch on myofibroblast contraction by quantifying the percentage of cells generating wrinkles in the surface of deformable silicone substrates and by measuring contraction of stress-released collagen gels. In the 21% O2 condition, 21±5% of SCFs generated surface wrinkles in elastomers of a stiffness that restricted visible distortions to highly contractile myofibroblasts (
      • Hinz B.
      • Celetta G.
      • Tomasek J.J.
      • et al.
      Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
      ; Figure 2a, arrowheads, Figure 2c). Five-day culture in 2% O2 almost abolished surface wrinkling (Figure 2b and c). Similar results were observed in 3D culture conditions: contraction of stress-released collagen gels by SCFs decreased 2.5-fold from 35±6% in 21% O2 to 14±1% in low O2 (P<0.01; Figure 2d).
      Figure thumbnail gr2
      Figure 2Effect of low O2 on subcutaneous fibroblast (SCF) contraction. Rat SCFs were grown for 5 days in O2 concentrations of (a) 21% and (b) 2% on wrinkling silicone substrates and observed in phase-contrast microscopy. Substrate stiffness restricts formation of surface wrinkles (arrowheads) to highly contractile myofibroblasts. Scale bar=20μm. (c) The percentages of wrinkling SCFs in 21% O2 and 2% O2 were quantified. (d) Contraction of SCF populations was assessed using attached collagen lattices. A minimum of five lattices were assayed per experimental condition, lattice diameter reduction was normalized to the lattice diameter before release (percent contraction), and mean values were calculated from at least three independent experiments. Error bars indicate SD of mean. *P<0.01 tested by Student's t-tests against 21% O2.
      To test whether SCFs show a tissue-specific response to the low-O2 switch, we assessed α-SMA expression in primary rat lung, liver, and heart fibroblasts. The baseline percentages of α-SMA-positive myofibroblasts (Figure 3a–c) and α-SMA expression levels (Figure 3d) in 21% O2 were higher compared with SCFs: 73±6% for rat lung (Figure 3a), 57±4% for liver (Figure 3b), and 67±8% for cardiac fibroblasts (Figure 3c). In contrast to what has been observed for SCFs, 2% O2 further increased the fraction of differentiated myofibroblasts to 92±9 and 87±8% in rat lung and liver fibroblast populations, respectively (P<0.05; Figure 3a and b). Conversely, rat cardiac fibroblasts responded to 2% O2, similar to SCFs, by decreasing the percentage of α-SMA-expressing cells by ∼5-fold to 14±4% in low O2 (P<0.01; Figure 3c).
      Figure thumbnail gr3
      Figure 3Tissue-specific fibroblast reaction to low O2. Primary explant cultures of rat fibroblasts from (a) lung, (b) liver, and (c) heart were grown for 5 days in 21% O2 (top) and 2% O2 (bottom). Cells were then immunostained for α-smooth muscle actin (α-SMA, red), F-actin (phalloidin, green), and nuclei (blue) (a–c) and processed for (d) western blotting. Scale bar=50μm.
      To evaluate whether rat SCFs are suitable for reproducing the characteristics of human skin fibroblasts, we prepared low-O2 cultures with primary human fibroblasts of various origins that all contribute to skin wound repair, including dermis, fascia, and subcutaneous (hypodermis) tissue. The percentage of α-SMA-positive dermal fibroblasts was reduced from 39±7% in 21% O2 to 8±2% after switching to 2% O2 (P<0.01; Supplementary Figure S1a online). Human SCFs also showed reduced percentages of myofibroblasts in 2% O2, decreasing from 10±2 to 3±1% (P<0.01; Supplementary Figure S1b online). The low percentage of myofibroblasts in 21% O2 fascia culture (8±2%) was reduced in 2% O2 (2±1%; Supplementary Figure S1c online). Hence, the low-O2 switch suppresses myofibroblast differentiation of human fibroblasts from dermis and skin-adjacent layers, similar to rat SCFs.

      Low-O2 conditions have a TGFβ1-desensitizing effect

      Experiments were performed in the absence of exogenous TGFβ1 in the culture medium. To decipher the mechanism through which low-O2 inhibits rat SCF-to-myofibroblast differentiation, we first measured the levels of TGFβ1 produced by rat SCF cultures using a reporter cell bioassay (
      • Abe M.
      • Harpel J.G.
      • Metz C.N.
      • et al.
      An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
      ). Unexpectedly, active and total TGFβ1 were increased by approximately threefold in the medium supernatants of SCFs grown in 2% O2 compared with 21% O2 (Figure 4a). Consistently, addition of exogenous TGFβ1 to the culture medium did not rescue SCF-to-myofibroblast differentiation in 2% O2 (Figure 4c and d). In 21% O2, addition of TGFβ1 increased the percentages of myofibroblasts (Figure 4b) and α-SMA expression (Figure 4e) as expected (
      • Hinz B.
      • Celetta G.
      • Tomasek J.J.
      • et al.
      Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
      ). One possibility of how low O2 desensitizes myofibroblast differentiation in the presence of TGFβ1 is reduced expression of the TGFβ1 receptor subunit TGFβ-RII. TGFβ-RII expression decreased by about 2.5-fold in 2% O2 compared with 21% O2 as shown by western blotting (Figure 4e). Despite downregulation of TGFβ-RII in 2% O2, we measured increased phosphorylation of Smad2 and expression of plasminogen activator inhibitor-1, showing active TGFβ1 signaling (Figure 4e). When assessing noncanonical TGFβ1 signaling, we found that phosphorylation of the p38 mitogen-activated protein kinase was reduced in 2% O2, whereas the levels of phosphorylated ERK1/2 were increased. It is thus possible that low O2 suppresses myofibroblast differentiation by mechanisms independent of Smad and mitogen-activated protein kinase signaling.
      Figure thumbnail gr4
      Figure 4The effect of low O2 on subcutaneous fibroblasts (SCFs) in the presence of active transforming growth factor-β1 (TGFβ1). (a) The levels of active and total TGFβ1 in medium conditioned by SCFs grown for 5 days in 21% and in 2% O2 were determined using transformed mink lung epithelial cells (TMLCs) as reporter cells. Data are presented as the mean of one representative experiment of at least five independent experiments. TGFβ1 (2ngml−1 ) was added to SCFs grown for 5 days in (b) 21% O2 and (c) 2% O2. Cells were then immunostained for α-smooth muscle actin (α-SMA, red), F-actin (green), and nuclei (blue) and (d) the percentage of α-SMA-positive cells was quantified. (e) SCFs extracted from the same conditions were tested by western blotting with loading controls vimentin, α-tubulin, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH); box includes TGFβ1-treated conditions. Contraction of TGFβ1-treated SCFs in (f) 21% and (g) 2% O2 was assessed by growth on silicone substrates that wrinkle only on high myofibroblast contraction. (h) Wrinkling and collagen contraction by TGFβ1-treated SCFs is normalized to 21% O2 without TGFβ1 addition. Scale bar=50μm. Error bars indicate SD of mean. *P<0.01 tested by Student's t-tests against 21% O2.

      Switch from high to low O2 reduces fibroblast and myofibroblast contraction

      In the presence of active TGFβ1, mechanical stress is prerequisite for myofibroblast differentiation (
      • Arora P.D.
      • Narani N.
      • McCulloch C.A.
      The compliance of collagen gels regulates transforming growth factor-beta induction of alpha-smooth muscle actin in fibroblasts.
      ;
      • Hinz B.
      • Mastrangelo D.
      • Iselin C.E.
      • et al.
      Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation.
      ;
      • Goffin J.M.
      • Pittet P.
      • Csucs G.
      • et al.
      Focal adhesion size controls tension-dependent recruitment of alpha-smooth muscle actin to stress fibers.
      ). We hypothesized that the switch to low O2 modulates the high intracellular stress needed for α-SMA expression and organization. Concomitant with the loss of tension in 2% O2 (±TGFβ1; Figures 2, 4g–h), we detected reduced levels of phosphorylated MYPT1, the myosin-binding subunit of myosin phosphatase (Figure 4e). MYPT1 is a downstream target of the small GTPase RhoA and its phosphorylation by the Rho kinase leads to increased actin–myosin contraction. Consistently, another downstream target of the Rho pathway, JNK, showed reduced phosphorylation in low-O2 conditions (Figure 4e).
      To answer whether loss of tension is the cause or effect of the reduced expression of α-SMA, we assessed SCF contraction on soft wrinkling silicone substrates and α-SMA expression in stress fibers over time in 2% O2, without adding TGFβ1 (Figure 5a). In contrast to the previous experiments, in this study we used silicone substrates that were engineered with high compliance, also allowing wrinkling by low-contractile α-SMA-negative fibroblasts. After 5-day culture in 21% O2, 25±2% SCFs formed α-SMA-positive stress fibers, and 75±3% of all SCFs produced sufficient force to wrinkle soft silicone substrates (Figure 5a and b). The percentage of wrinkling SCFs was reduced twofold after just 6hours culture in 2% O2 (45±2%), whereas the percentage of α-SMA-positive cells decreased significantly (11±1%) after only 3 days. After 4 days in 2% O2, the percentages of both α-SMA-positive and wrinkling SCFs decreased to almost zero (Figure 5b). By calculating the percentage of wrinkling SCFs among the α-SMA-positive cells, it became clear that myofibroblasts rapidly lose cell contraction in 2% O2 (Figure 5c). This clearly preceded visible changes in α-SMA organization (Figure 5a) and expression (Figure 5d). The relaxing effect of 2% O2 was not restricted to myofibroblasts, as the percentage of wrinkling cells among the α-SMA-negative SCF population was similarly reduced from approximately 70 to 15% within 6hours (Figure 5a and c). Consistently, low O2 decreased diameter reduction of free-floating collagen gels (Supplementary Figure S2 online), which do not allow formation of the myofibroblast phenotype (
      • Tomasek J.J.
      • Gabbiani G.
      • Hinz B.
      • et al.
      Myofibroblasts and mechano-regulation of connective tissue remodelling.
      ). These results suggest that loss of intracellular stress, presumably via reduced RhoA signaling, may be an important factor for the suppression of myofibroblast differentiation in low O2.
      Figure thumbnail gr5
      Figure 5Correlation of subcutaneous fibroblast (SCFs) contraction and myofibroblast differentiation over culture time in low O2. (a) Rat SCFs were grown in 21% O2 for 5 days minus the indicated times for which the culture was transferred to 2% O2. For the last 6 hours in 2% O2, SCFs were subcultured on soft wrinkling silicone substrates; “0h” cells were always kept in 21% O2. After 6 hours growth on silicone substrates, cells were immunostained for α-smooth muscle actin (α-SMA, red), F-actin (phalloidin, green), and nuclei (blue) (top panel) and overlaid with phase-contrast (wrinkle) images (bottom panel). Note that the soft substrates used here permit even low-contractile fibroblasts to produce wrinkles, in contrast to the stiff substrates used in and . (b) Immunostaining and phase-contrast images of wrinkling substrates were separately analyzed for the percentages of wrinkling cells and α-SMA-positive cells of all cells over time in 2% O2 or (c) to determine the percentages of wrinkling cells of all α-SMA-positive cells (white columns) and of all α-SMA-negative cells (gray columns). (d) SCFs extracted from the same conditions were tested by western blotting. Scale bar=100μm. Error bars indicate SD of mean. *P<0.01 tested by Student's t-tests against 21% O2 in each experimental group.

      SCF myofibroblast dedifferentiation under low O2 is reversible

      It is our ultimate goal to improve the healing of ischemic wounds by restoring the myofibroblast phenotype. We first tested whether the effect of the low-O2 switch on myofibroblast differentiation was at all reversible by preculturing SCFs for 5 days in 2% O2 followed by additional 5-day culture in 21% O2 (2% → 21%). Returning to high O2 conditions completely restored the percentages of contractile and α-SMA-positive SCFs (Figure 6a, b, and e). Concomitantly, α-SMA expression was augmented to levels similar to those observed in 21% O2 (Figure 6f), showing reversibility of the low-O2 effects. Because reduced intracellular stress, here observed in low O2, leads to loss of the myofibroblast phenotype (
      • Goffin J.M.
      • Pittet P.
      • Csucs G.
      • et al.
      Focal adhesion size controls tension-dependent recruitment of alpha-smooth muscle actin to stress fibers.
      ), we tested whether application of external stress could compensate for the loss of intracellular tension. SCFs were grown in 2% O2 on silicone membranes that were cyclically stretched by 10% with a frequency of 0.0016Hz (one cycle every 10minutes) over 5 days. Mechanical stimulation rescued myofibroblast differentiation and function in 2% O2 and increased both parameters in 21% O2 compared with static growth on silicone membranes (Figure 6).
      Figure thumbnail gr6
      Figure 6Reversibility of low-O2-induced effects by O2 elevation and mechanical stimulation. Rat subcutaneous fibroblasts (SCFs) were grown for 5 days in 2% O2, followed by another 5 days in (a) 2% O2, (b) 21% O2, (c) 2% O2 under cyclic stretch of 10% with a frequency of 0.0016Hz, and (d) 21% O2 under cyclic stretch. Cells were immunostained for α-smooth muscle actin (α-SMA, red), F-actin (green), and nuclei (blue). (e) SCFs were assessed for percentages of α-SMA-positive cells and of myofibroblasts wrinkling stiff silicone substrates. Collagen gel contraction was assessed after 3 days of growth under the same O2 conditions as used for the “rescue.” Percentage values are calculated using 2 × 5 days 21% O2 culture as reference (dotted line) (5+3 days in the case of collagen contraction). (f) Western blots were performed with cells extracted after each treatment. Scale bar=100μm. Error bars indicate SD of mean. *P<0.01 tested by Student's t-tests against 2 × 5 days (collagen: 5+3 days) 21% O2 culture in each experimental group.

      Discussion

      The epidemics of diabetes and obesity are dramatically increasing the number of chronic wounds, particularly in the lower extremities, areas typically affected by decreased blood flow and lower O2 supply (
      • Sen C.K.
      Wound healing essentials: let there be oxygen.
      ). Recent studies have shown that dysfunction of wound (myo)fibroblasts is a key feature of nonhealing wounds (
      • Clark R.A.
      Oxidative stress and “senescent” fibroblasts in non-healing wounds as potential therapeutic targets.
      ;
      • Schafer M.
      • Werner S.
      Oxidative stress in normal and impaired wound repair.
      ;
      • Wall I.B.
      • Moseley R.
      • Baird D.M.
      • et al.
      Fibroblast dysfunction is a key factor in the non-healing of chronic venous leg ulcers.
      ). We here show that cultured skin myofibroblasts subjected to low O2 downregulate contraction and expression of α-SMA. This effect was reversible; reestablishment of high O2 and application of mechanical stress led to functional and phenotypic recovery of the myofibroblasts.
      Our previous work using a rat hind-limb ischemic wound model suggested that reduced wound contraction in ischemic conditions is related to decreased myofibroblast differentiation (
      • Alizadeh N.
      • Pepper M.S.
      • Modarressi A.
      • et al.
      Persistent ischemia impairs myofibroblast development in wound granulation tissue: a new model of delayed wound healing.
      ). Conversely, hypoxic culture has been shown to stimulate fibrosis and myofibroblast activity in vessels (
      • Das M.
      • Dempsey E.C.
      • Reeves J.T.
      • et al.
      Selective expansion of fibroblast subpopulations from pulmonary artery adventitia in response to hypoxia.
      ;
      • Krick S.
      • Hanze J.
      • Eul B.
      • et al.
      Hypoxia-driven proliferation of human pulmonary artery fibroblasts: cross-talk between HIF-1alpha and an autocrine angiotensin system.
      ;
      • Zhang B.
      • Liang X.
      • Shi W.
      • et al.
      Role of impaired peritubular capillary and hypoxia in progressive interstitial fibrosis after 56 subtotal nephrectomy of rats.
      ), lung (
      • Short M.
      • Nemenoff R.A.
      • Zawada W.M.
      • et al.
      Hypoxia induces differentiation of pulmonary artery adventitial fibroblasts into myofibroblasts.
      ;
      • Leufgen H.
      • Bihl M.P.
      • Rudiger J.J.
      • et al.
      Collagenase expression and activity is modulated by the interaction of collagen types, hypoxia, and nutrition in human lung cells.
      ;
      • Jiang Y.L.
      • Dai A.G.
      • Li Q.F.
      • et al.
      Transforming growth factor-beta1 induces transdifferentiation of fibroblasts into myofibroblasts in hypoxic pulmonary vascular remodeling.
      ), liver, and kidney (
      • Manotham K.
      • Tanaka T.
      • Matsumoto M.
      • et al.
      Transdifferentiation of cultured tubular cells induced by hypoxia.
      ;
      • Shi Y.F.
      • Fong C.C.
      • Zhang Q.
      • et al.
      Hypoxia induces the activation of human hepatic stellate cells LX-2 through TGF-beta signaling pathway.
      ). Comparing for the first time the effect of O2 reduction on myofibroblast differentiation of fibroblasts cultured from different tissues, we show here that the tissue origin determines fibroblast reaction to O2 changes. Whereas subcutaneous, dermal, and heart fibroblasts show a decreased expression of α-SMA, lung and liver fibroblasts exhibit the opposite reaction. This is consistent with the observation that fibroblasts isolated from different tissue origins show different percentages of myofibroblasts in standard culture (
      • Desmouliere A.
      • Rubbia-Brandt L.
      • Abdiu A.
      • et al.
      Alpha-smooth muscle actin is expressed in a subpopulation of cultured and cloned fibroblasts and is modulated by gamma-interferon.
      ;
      • Dugina V.
      • Alexandrova A.
      • Chaponnier C.
      • et al.
      Rat fibroblasts cultured from various organs exhibit differences in alpha-smooth muscle actin expression, cytoskeletal pattern, and adhesive structure organization.
      ) and with the reported fibroblast heterogeneity in normal and wounded tissues (
      • Harper R.A.
      • Grove G.
      Human skin fibroblasts derived from papillary and reticular dermis: differences in growth potential in vitro.
      ;
      • Sorrell J.M.
      • Caplan A.I.
      Fibroblast heterogeneity: more than skin deep.
      ;
      • Wall I.B.
      • Moseley R.
      • Baird D.M.
      • et al.
      Fibroblast dysfunction is a key factor in the non-healing of chronic venous leg ulcers.
      ).
      Although the environmental pO2 of ∼140mmHg (21%) is generally defined as normoxia for cell cultures, the pO2 in normal skin is lower with ∼55–70mmHg (∼5–8%;
      • Sen C.K.
      Wound healing essentials: let there be oxygen.
      ). It is conceivable that fibroblasts perceive the change from 5% tissue O2 to 21% culture O2 levels as a hyperoxic switch, possibly contributing to the spontaneous induction of myofibroblast differentiation. However, studies with cardiac fibroblasts, which experience similar pO2 in the heart (5%), showed the ability of cells to adjust to new pO2 in culture and to “reset” their O2 reference point (
      • Roy S.
      • Khanna S.
      • Bickerstaff A.A.
      • et al.
      Oxygen sensing by primary cardiac fibroblasts: a key role of p21(Waf1/Cip1/Sdi1).
      ). It is appreciated that cells sense changes in O2 rather than absolute levels and that hypoxia and normoxia are relative terms that relate to the respective O2 set point (
      • Sen C.K.
      • Khanna S.
      • Roy S.
      Perceived hyperoxia: oxygen-induced remodeling of the reoxygenated heart.
      ). We can consider that our fibroblasts have reached a “normoxic” set point in 21% O2 after passaging, as a switch to 5 and 2% O2 was always reported as hypoxia by the upregulation of HIF-1α (
      • Semenza G.L.
      Hypoxia-inducible factor 1 (HIF-1) pathway.
      ). But how does hypoxia suppress myofibroblast differentiation?
      Altered TGFβ1 signaling is one important factor responsible for insufficient wound healing in hypoxia (
      • Ahn S.T.
      • Mustoe T.A.
      Effects of ischemia on ulcer wound healing: a new model in the rabbit ear.
      ;
      • Wu L.
      • Xia Y.P.
      • Roth S.I.
      • et al.
      Transforming growth factor-beta1 fails to stimulate wound healing and impairs its signal transduction in an aged ischemic ulcer model: importance of oxygen and age.
      ;
      • Tandara A.A.
      • Mustoe T.A.
      Oxygen in wound healing—more than a nutrient.
      ). Dermal fibroblasts of chronic wounds and such cultured for 2–6 weeks under “chronic” low O2 show decreased levels of TGFβ1 and α1-procollagen expression (
      • Siddiqui A.
      • Galiano R.D.
      • Connors D.
      • et al.
      Differential effects of oxygen on human dermal fibroblasts: acute versus chronic hypoxia.
      ;
      • Kim B.C.
      • Kim H.T.
      • Park S.H.
      • et al.
      Fibroblasts from chronic wounds show altered TGF-beta-signaling and decreased TGF-beta Type II receptor expression.
      ). It appears that the exposure time determines the response of fibroblasts to low O2 because culturing human dermal fibroblasts for short periods up to 72hours in 2% hypoxia showed upregulation of TGFβ1, collagen I, and vascular endothelial growth factor and stimulation of cell proliferation and migration (
      • Falanga V.
      • Qian S.W.
      • Danielpour D.
      • et al.
      Hypoxia upregulates the synthesis of TGF-beta 1 by human dermal fibroblasts.
      ,
      • Falanga V.
      • Zhou L.
      • Yufit T.
      Low oxygen tension stimulates collagen synthesis and COL1A1 transcription through the action of TGF-beta1.
      ;
      • Siddiqui A.
      • Galiano R.D.
      • Connors D.
      • et al.
      Differential effects of oxygen on human dermal fibroblasts: acute versus chronic hypoxia.
      ;
      • Steinbrech D.S.
      • Longaker M.T.
      • Mehrara B.J.
      • et al.
      Fibroblast response to hypoxia: the relationship between angiogenesis and matrix regulation.
      ). These data support the notion of acute low O2 as a positive stimulator of wound healing (
      • Frangogiannis N.G.
      • Michael L.H.
      • Entman M.L.
      Myofibroblasts in reperfused myocardial infarcts express the embryonic form of smooth muscle myosin heavy chain (SMemb).
      ). The upregulation of TGFβ1 in hypoxic dermal rat wounds (
      • Alizadeh N.
      • Pepper M.S.
      • Modarressi A.
      • et al.
      Persistent ischemia impairs myofibroblast development in wound granulation tissue: a new model of delayed wound healing.
      ) is consistent with our results showing increased levels of active and total TGFβ1 in response to the low-O2 switch. The absence of effects of TGFβ1 in these conditions may be partly due to the concomitant downregulation of the TGFβ-RII shown here, possibly by a negative TGFβ1 feedback loop. However, TGFβ-RII downstream targets such as phospho-Smad2 and plasminogen activator inhibitor-1 were not affected by low O2 in our experiments, and noncanonical mitogen-activated protein kinase pathways showed no consistent response to low O2.
      The question remains how myofibroblast function is possibly restored in a clinical setting of ischemia if addition of the most potent growth factor TGFβ1 apparently fails, as does that of other growth factors (
      • Chan R.K.
      • Liu P.H.
      • Pietramaggiori G.
      • et al.
      Effect of recombinant platelet-derived growth factor (Regranex) on wound closure in genetically diabetic mice.
      ). The effects of low pO2 are reversible when fibroblasts are returned to 21% O2 after “chronic” (
      • Siddiqui A.
      • Galiano R.D.
      • Connors D.
      • et al.
      Differential effects of oxygen on human dermal fibroblasts: acute versus chronic hypoxia.
      ) and short-term low-O2 culture, as shown by the complete rescue of myofibroblast differentiation in our experiments. Clinical hyperbaric O2 therapy has been applied to stimulate healing by elevating the pO2 in the wound environment with mixed results (
      • Sen C.K.
      • Khanna S.
      • Gordillo G.
      • et al.
      Oxygen, oxidants, and antioxidants in wound healing: an emerging paradigm.
      ;
      • Thackham J.A.
      • McElwain D.L.
      • Long R.J.
      The use of hyperbaric oxygen therapy to treat chronic wounds: a review.
      ). Local O2 therapy provides only limited penetration of O2 into the deep wound environment. In contrast, systemic hyperbaric O2 therapy appears to be effective by increasing the pO2 in the wound bed (
      • Mathieu D.
      Role of hyperbaric oxygen therapy in the management of lower extremity wounds.
      ;
      • Goldman R.J.
      Hyperbaric oxygen therapy for wound healing and limb salvage: a systematic review.
      ;
      • Gordillo G.M.
      • Sen C.K.
      Evidence-based recommendations for the use of topical oxygen therapy in the treatment of lower extremity wounds.
      ). Additional studies with the systemic approach are required—in particular, by assessing its effect on myofibroblasts.
      Mechanical stress is another prerequisite for fibroblast-to-myofibroblast differentiation in vivo and in vitro (
      • Arora P.D.
      • Narani N.
      • McCulloch C.A.
      The compliance of collagen gels regulates transforming growth factor-beta induction of alpha-smooth muscle actin in fibroblasts.
      ;
      • Hinz B.
      • Mastrangelo D.
      • Iselin C.E.
      • et al.
      Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation.
      ;
      • Goffin J.M.
      • Pittet P.
      • Csucs G.
      • et al.
      Focal adhesion size controls tension-dependent recruitment of alpha-smooth muscle actin to stress fibers.
      ;
      • Wipff P.J.
      • Rifkin D.B.
      • Meister J.J.
      • et al.
      Myofibroblast contraction activates latent TGF-beta1 from the extracellular matrix.
      ). Reduction of extracellular and intracellular stress leads to removal of α-SMA from stress fibers within hours; under continued low stress, expression levels of α-SMA are reduced (
      • Hinz B.
      • Gabbiani G.
      • Chaponnier C.
      The NH2-terminal peptide of alpha-smooth muscle actin inhibits force generation by the myofibroblast in vitro and in vivo.
      ;
      • Goffin J.M.
      • Pittet P.
      • Csucs G.
      • et al.
      Focal adhesion size controls tension-dependent recruitment of alpha-smooth muscle actin to stress fibers.
      ). We suggest that the rapid loss of intracellular tension (contraction) under 2% O2 is responsible for the suppression of the myofibroblast phenotype in our study. Importantly, we could compensate for the reduced intracellular tension by applying extracellular cyclic stress to rescue myofibroblast phenotype and function. The mechanism of how low O2 reduces intracellular stress and how extracellular stress can overcome this effect remains elusive. One possibility is the reduced downstream signaling of the RhoA/Rho kinase pathway observed in our study. Implication of RhoA signaling is supported by another study showing that reduced RhoA activity in low-O2 culture correlates with impaired corneal myofibroblast differentiation in response to TGFβ1 (
      • Xing D.
      • Bonanno J.A.
      Hypoxia reduces TGFbeta1-induced corneal keratocyte myofibroblast transformation.
      ). Mechanical stimulation is a promising approach to promote closure of poorly healing wounds. Mechanical devices have been shown to increase vascularization and epithelial cell proliferation in animal models of wound healing (
      • Pietramaggiori G.
      • Liu P.
      • Scherer S.S.
      • et al.
      Tensile forces stimulate vascular remodeling and epidermal cell proliferation in living skin.
      ). This strategy is already clinically applied with negative-pressure wound closure devices. It appears that part of the beneficial effects of this approach is due to the mechanical stimulation of the wound environment (
      • Scherer S.S.
      • Pietramaggiori G.
      • Mathews J.C.
      • et al.
      The mechanism of action of the vacuum-assisted closure device.
      ). It remains to be shown whether the micromechanical forces exerted by the vacuum also affect fibroblast-to-myofibroblast differentiation. In conclusion, our study shows that low O2 is a negative modulator of skin myofibroblast differentiation and function. The possibility of reversing these negative effects by mechanical challenge may be particularly useful for the development of clinical strategies to treat ischemic wounds.

      Materials and Methods

      Cell culture and reagents

      Primary rat fibroblasts were obtained from explants of subcutaneous tissue, lung, heart, and liver cells and cultured between passages 2 and 5 as described previously (
      • Hinz B.
      • Celetta G.
      • Tomasek J.J.
      • et al.
      Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
      ). Human fibroblasts were grown from explants of dermis, subcutis, and fascia. All cells were cultured for at least two passages in standard (21%) O2 before being used in experiments. Experiments were performed with cells from at least three animals or human samples per cell type. In selected experiments (Figure 4), myofibroblast differentiation was induced by exogenously adding 2ngml−1 TGFβ1 (R&D Systems, Abington, UK) once for 5 days to the culture medium. The study was conducted according to the Declaration of Helsinki Principles. The medical ethical committee of the University Hospitals of Geneva approved all described studies.

      Cell survival, proliferation, and apoptosis

      Proliferation was determined by cell counting after collection. For apoptosis and viability tests, the initially seeded cell numbers were adjusted to obtain similar cell counts at the time of assessment. Cell viability was evaluated by a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide colorimetric assay, measuring absorbance at 570nm in a multiplate spectrophotometer (Centro LB; Berthold Technologies, Regensdorf, Switzerland). Apoptosis was quantified using a Caspase-Glo 3/7 assay kit (Promega, Wallisellen, Switzerland) and multiwell plate luminometer (Berthold Technologies).

      Antibodies and microscopy

      We used primary antibodies against α-SMA (a gift from G Gabbiani, University of Geneva, Switzerland); TGFβ1-RII (Santa Cruz Biotechnologies, Santa Cruz, CA); vimentin (Dako, Baar, Switzerland); phosphorylated and total Smad2, phosphorylated and total p38 mitogen-activated protein kinase, phosphorylated and total ERK1/2, phosphorylated MYPT-1, and phosphorylated JNK (all Cell Signaling Technology, Danvers, MA); plasminogen activator inhibitor-1 (Becton Dickinson, Heidelberg, Germany); α-tubulin, glyceraldehyde-3-phosphate dehydrogenase, and HIF-1α (all from Sigma, Buchs, Switzerland). For immunofluorescence we used secondary Alexa Fluor-conjugated goat anti-mouse and goat anti-rabbit antibodies (Molecular Probes, Invitrogen, Basel, Switzerland), DAPI (Sigma), and phalloidin-Alexa Fluor 488 (Molecular Probes). For western blotting, we used horseradish peroxidase–conjugated goat anti-mouse and goat anti-rabbit secondary antibodies (Jackson ImmunoResearch Laboratories, Suffolk, UK). Microscopy was performed using oil immersion objectives (Plan Neofluar, × 40/1.2 Ph3; Plan Neofluar, × 63/1.4 Ph3; Zeiss, Feldbach, Switzerland) on an inverted microscope (Axiovert 135; Zeiss) and digital CCD camera (Hamamatsu C4742-95-12ERG; Bucher Biotec, Basel, Switzerland). Images were acquired with Openlab 3.1.2 software (Improvision, Basel, Switzerland) and assembled with Adobe Photoshop CS3.

      TGFβ1 bioassay

      To quantify active TGFβ1, we used transformed mink lung epithelial cells (TMLCs), producing luciferase under control of the TGFβ1-responsive plasminogen activator inhibitor-1 promoter (a kind gift from D Rifkin;
      • Abe M.
      • Harpel J.G.
      • Metz C.N.
      • et al.
      An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
      ). TMLCs (50,000cells per cm2) were grown for 4hours before being exposed to fibroblast culture-conditioned medium (active TGFβ1) and to conditioned medium that had been heated for 10minutes at 80°C to activate TGFβ1 (total TGFβ1). TMLCs were incubated in the conditioned medium for another 14hours to produce luciferase as a function of TGFβ1 content and were subsequently lysed. Luciferase activity was assessed by light production from a luciferin substrate (Promega) using a luminometer (Berthold Technologies). All results were normalized to baseline TMLC luciferase production in nonconditioned culture medium.

      Collagen gel contraction, wrinkling substrates, and stretchable membranes

      To quantify the contractility of cell populations, we grew fibroblasts in attached collagen lattices, providing the mechanical conditions for myofibroblast development (
      • Arora P.D.
      • Narani N.
      • McCulloch C.A.
      The compliance of collagen gels regulates transforming growth factor-beta induction of alpha-smooth muscle actin in fibroblasts.
      ;
      • Hinz B.
      • Celetta G.
      • Tomasek J.J.
      • et al.
      Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
      ;
      • Tomasek J.J.
      • Gabbiani G.
      • Hinz B.
      • et al.
      Myofibroblasts and mechano-regulation of connective tissue remodelling.
      ). Populations were initiated with 0.25 to 10.0 × 105 cells per ml collagen type I (1.0mgml−1; Sigma) adapted to obtain similar cell numbers at the time of contraction measurement. For this, lattices were released after 5 days of growth and their diameter was measured after 30minutes.
      Wrinkling silicone substrates (Excellness Biotech, Lausanne, Switzerland) were used to identify individual contractile cells (
      • Hinz B.
      • Celetta G.
      • Tomasek J.J.
      • et al.
      Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
      ). In one experimental series (Figures 2 and 4), substrates were used with a stiffness of 10kPa to restrict wrinkle formation to highly contractile α-SMA-positive cells. In another experimental series (Figure 5), substrates were used with a stiffness of 3kPa, allowing wrinkle formation by almost all fibroblastic cells. Substrates were rendered cell-adhesive with collagen type I (10μgml−1; Sigma). To test the effect of low O2 on fibroblast wrinkling, we pre-grew cells on tissue culture plastic for various durations under low pO2 and then passaged them onto the wrinkling substrates for 6hours for further growth in the respective condition. Percentages of α-SMA-positive and wrinkling fibroblasts were calculated using MetaMorph (Visitron Systems, Munich, Germany) after manual cell selection on a computer screen. Ten random regions of interest were analyzed per experimental condition (∼100 cells per field) and at least five independent experiments were performed.
      For mechanical stimulation, cells were grown on homemade stretchable culture membranes (
      • Wipff P.J.
      • Majd H.
      • Acharya C.
      • et al.
      The covalent attachment of adhesion molecules to silicone membranes for cell stretching applications.
      ) that were isoaxially stretched over 5 days, applying linear stretches of 10% with a cycle frequency of 0.0016Hz using a motorized iris stretch device (Cytomec, Spiez, Switzerland;
      • Majd H.
      • Wipff P.J.
      • Buscemi L.
      • et al.
      A novel method of dynamic culture surface expansion improves mesenchymal stem cell proliferation and phenotype.
      ).

      ACKNOWLEDGMENTS

      We acknowledge J Smith-Clerc for excellent technical assistance and L Buscemi and H Majd for technical help and expert advice. We are grateful to J-J Meister for providing laboratory facilities, equipment, and support. G Gabbiani, C McCulloch, C Chaponnier, R Wells, and D Rifkin are acknowledged for providing antibodies, cells, and reagents. This work was supported by the Swiss National Science Foundation, grants 3200-067254 (to BP) and 3100A0-113733/1 (to BH); the Fondation de Reuter and the Fondation Fernex (to BP); the Gebert Rüf Stiftung; and the Canadian Institutes of Health Research, grant 488342 (to BH).

      SUPPLEMENTARY MATERIAL

      Supplementary material is linked to the online version of the paper at http://www.nature.com/jid

      REFERENCES

        • Abe M.
        • Harpel J.G.
        • Metz C.N.
        • et al.
        An assay for transforming growth factor-beta using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct.
        Anal Biochem. 1994; 216: 276-284
        • Ahn S.T.
        • Mustoe T.A.
        Effects of ischemia on ulcer wound healing: a new model in the rabbit ear.
        Ann Plast Surg. 1990; 24: 17-23
        • Alizadeh N.
        • Pepper M.S.
        • Modarressi A.
        • et al.
        Persistent ischemia impairs myofibroblast development in wound granulation tissue: a new model of delayed wound healing.
        Wound Repair Regen. 2007; 15: 809-816
        • Arora P.D.
        • Narani N.
        • McCulloch C.A.
        The compliance of collagen gels regulates transforming growth factor-beta induction of alpha-smooth muscle actin in fibroblasts.
        Am J Pathol. 1999; 154: 871-882
        • Banwell P.E.
        Topical negative pressure therapy in wound care.
        J Wound Care. 1999; 8: 79-84
        • Chan R.K.
        • Liu P.H.
        • Pietramaggiori G.
        • et al.
        Effect of recombinant platelet-derived growth factor (Regranex) on wound closure in genetically diabetic mice.
        J Burn Care Res. 2006; 27: 202-205
        • Clark R.A.
        Oxidative stress and “senescent” fibroblasts in non-healing wounds as potential therapeutic targets.
        J Invest Dermatol. 2008; 128: 2361-2364
        • Clark R.A.
        • Ghosh K.
        • Tonnesen M.G.
        Tissue engineering for cutaneous wounds.
        J Invest Dermatol. 2007; 127: 1018-1029
        • Das M.
        • Dempsey E.C.
        • Reeves J.T.
        • et al.
        Selective expansion of fibroblast subpopulations from pulmonary artery adventitia in response to hypoxia.
        Am J Physiol Lung Cell Mol Physiol. 2002; 282: L976-L986
        • Desmouliere A.
        • Redard M.
        • Darby I.
        • et al.
        Apoptosis mediates the decrease in cellularity during the transition between granulation tissue and scar.
        Am J Pathol. 1995; 146: 56-66
        • Desmouliere A.
        • Rubbia-Brandt L.
        • Abdiu A.
        • et al.
        Alpha-smooth muscle actin is expressed in a subpopulation of cultured and cloned fibroblasts and is modulated by gamma-interferon.
        Exp Cell Res. 1992; 201: 64-73
        • Dugina V.
        • Alexandrova A.
        • Chaponnier C.
        • et al.
        Rat fibroblasts cultured from various organs exhibit differences in alpha-smooth muscle actin expression, cytoskeletal pattern, and adhesive structure organization.
        Exp Cell Res. 1998; 238: 481-490
        • Falanga V.
        Wound healing and its impairment in the diabetic foot.
        Lancet. 2005; 366: 1736-1743
        • Falanga V.
        • Qian S.W.
        • Danielpour D.
        • et al.
        Hypoxia upregulates the synthesis of TGF-beta 1 by human dermal fibroblasts.
        J Invest Dermatol. 1991; 97: 634-637
        • Falanga V.
        • Zhou L.
        • Yufit T.
        Low oxygen tension stimulates collagen synthesis and COL1A1 transcription through the action of TGF-beta1.
        J Cell Physiol. 2002; 191: 42-50
        • Frangogiannis N.G.
        • Michael L.H.
        • Entman M.L.
        Myofibroblasts in reperfused myocardial infarcts express the embryonic form of smooth muscle myosin heavy chain (SMemb).
        Cardiovasc Res. 2000; 48: 89-100
        • Gabbiani G.
        The myofibroblast in wound healing and fibrocontractive diseases.
        J Pathol. 2003; 200: 500-503
        • Goffin J.M.
        • Pittet P.
        • Csucs G.
        • et al.
        Focal adhesion size controls tension-dependent recruitment of alpha-smooth muscle actin to stress fibers.
        J Cell Biol. 2006; 172: 259-268
        • Goldman R.J.
        Hyperbaric oxygen therapy for wound healing and limb salvage: a systematic review.
        PM R. 2009; 1: 471-489
        • Gordillo G.M.
        • Sen C.K.
        Evidence-based recommendations for the use of topical oxygen therapy in the treatment of lower extremity wounds.
        Int J Low Extrem Wounds. 2009; 8: 105-111
        • Harper R.A.
        • Grove G.
        Human skin fibroblasts derived from papillary and reticular dermis: differences in growth potential in vitro.
        Science. 1979; 204: 526-527
        • Hinz B.
        Formation and function of the myofibroblast during tissue repair.
        J Invest Dermatol. 2007; 127: 526-537
        • Hinz B.
        • Celetta G.
        • Tomasek J.J.
        • et al.
        Alpha-smooth muscle actin expression upregulates fibroblast contractile activity.
        Mol Biol Cell. 2001; 12: 2730-2741
        • Hinz B.
        • Gabbiani G.
        Mechanisms of force generation and transmission by myofibroblasts.
        Curr Opin Biotechnol. 2003; 14: 538-546
        • Hinz B.
        • Gabbiani G.
        • Chaponnier C.
        The NH2-terminal peptide of alpha-smooth muscle actin inhibits force generation by the myofibroblast in vitro and in vivo.
        J Cell Biol. 2002; 157: 657-663
        • Hinz B.
        • Mastrangelo D.
        • Iselin C.E.
        • et al.
        Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation.
        Am J Pathol. 2001; 159: 1009-1020
        • Hinz B.
        • Phan S.H.
        • Thannickal V.J.
        • et al.
        The myofibroblast: one function, multiple origins.
        Am J Pathol. 2007; 170: 1807-1816
        • Jiang Y.L.
        • Dai A.G.
        • Li Q.F.
        • et al.
        Transforming growth factor-beta1 induces transdifferentiation of fibroblasts into myofibroblasts in hypoxic pulmonary vascular remodeling.
        Acta Biochim Biophys Sin (Shanghai). 2006; 38: 29-36
        • Kim B.C.
        • Kim H.T.
        • Park S.H.
        • et al.
        Fibroblasts from chronic wounds show altered TGF-beta-signaling and decreased TGF-beta Type II receptor expression.
        J Cell Physiol. 2003; 195: 331-336
        • Krick S.
        • Hanze J.
        • Eul B.
        • et al.
        Hypoxia-driven proliferation of human pulmonary artery fibroblasts: cross-talk between HIF-1alpha and an autocrine angiotensin system.
        FASEB J. 2005; 19: 857-859
        • Leufgen H.
        • Bihl M.P.
        • Rudiger J.J.
        • et al.
        Collagenase expression and activity is modulated by the interaction of collagen types, hypoxia, and nutrition in human lung cells.
        J Cell Physiol. 2005; 204: 146-154
        • Majd H.
        • Wipff P.J.
        • Buscemi L.
        • et al.
        A novel method of dynamic culture surface expansion improves mesenchymal stem cell proliferation and phenotype.
        Stem Cells. 2008; 27: 200-209
        • Manotham K.
        • Tanaka T.
        • Matsumoto M.
        • et al.
        Transdifferentiation of cultured tubular cells induced by hypoxia.
        Kidney Int. 2004; 65: 871-880
        • Margolis D.J.
        • Bilker W.
        • Knauss J.
        • et al.
        The incidence and prevalence of pressure ulcers among elderly patients in general medical practice.
        Ann Epidemiol. 2002; 12: 321-325
        • Mathieu D.
        Role of hyperbaric oxygen therapy in the management of lower extremity wounds.
        Int J Low Extrem Wounds. 2006; 5: 233-235
        • Pelka R.
        [The economic situation of chronic wounds].
        Krankenpfl J. 1997; 35: 338
        • Pietramaggiori G.
        • Liu P.
        • Scherer S.S.
        • et al.
        Tensile forces stimulate vascular remodeling and epidermal cell proliferation in living skin.
        Ann Surg. 2007; 246: 896-902
        • Roy S.
        • Khanna S.
        • Bickerstaff A.A.
        • et al.
        Oxygen sensing by primary cardiac fibroblasts: a key role of p21(Waf1/Cip1/Sdi1).
        Circ Res. 2003; 92: 264-271
        • Schafer M.
        • Werner S.
        Oxidative stress in normal and impaired wound repair.
        Pharmacol Res. 2008; 58: 165-171
        • Scheid A.
        • Wenger R.H.
        • Christina H.
        • et al.
        Hypoxia-regulated gene expression in fetal wound regeneration and adult wound repair.
        Pediatr Surg Int. 2000; 16: 232-236
        • Scherer S.S.
        • Pietramaggiori G.
        • Mathews J.C.
        • et al.
        The mechanism of action of the vacuum-assisted closure device.
        Plast Reconstr Surg. 2008; 122: 786-797
        • Semenza G.L.
        Hypoxia-inducible factor 1 (HIF-1) pathway.
        Sci STKE. 2007; 2007: cm8
        • Sen C.K.
        Wound healing essentials: let there be oxygen.
        Wound Repair Regen. 2009; 17: 1-18
        • Sen C.K.
        • Khanna S.
        • Gordillo G.
        • et al.
        Oxygen, oxidants, and antioxidants in wound healing: an emerging paradigm.
        Ann NY Acad Sci. 2002; 957: 239-249
        • Sen C.K.
        • Khanna S.
        • Roy S.
        Perceived hyperoxia: oxygen-induced remodeling of the reoxygenated heart.
        Cardiovasc Res. 2006; 71: 280-288
        • Shi Y.F.
        • Fong C.C.
        • Zhang Q.
        • et al.
        Hypoxia induces the activation of human hepatic stellate cells LX-2 through TGF-beta signaling pathway.
        FEBS Lett. 2007; 581: 203-210
        • Short M.
        • Nemenoff R.A.
        • Zawada W.M.
        • et al.
        Hypoxia induces differentiation of pulmonary artery adventitial fibroblasts into myofibroblasts.
        Am J Physiol Cell Physiol. 2004; 286: C416-C425
        • Siddiqui A.
        • Galiano R.D.
        • Connors D.
        • et al.
        Differential effects of oxygen on human dermal fibroblasts: acute versus chronic hypoxia.
        Wound Repair and Regen. 1996; 4: 211-218
        • Sorrell J.M.
        • Caplan A.I.
        Fibroblast heterogeneity: more than skin deep.
        J Cell Sci. 2004; 117: 667-675
        • Steinbrech D.S.
        • Longaker M.T.
        • Mehrara B.J.
        • et al.
        Fibroblast response to hypoxia: the relationship between angiogenesis and matrix regulation.
        J Surg Res. 1999; 84: 127-133
        • Tandara A.A.
        • Mustoe T.A.
        Oxygen in wound healing—more than a nutrient.
        World J Surg. 2004; 28: 294-300
        • Thackham J.A.
        • McElwain D.L.
        • Long R.J.
        The use of hyperbaric oxygen therapy to treat chronic wounds: a review.
        Wound Repair Regen. 2008; 16: 321-330
        • Tomasek J.J.
        • Gabbiani G.
        • Hinz B.
        • et al.
        Myofibroblasts and mechano-regulation of connective tissue remodelling.
        Nat Rev Mol Cell Biol. 2002; 3: 349-363
        • Wall I.B.
        • Moseley R.
        • Baird D.M.
        • et al.
        Fibroblast dysfunction is a key factor in the non-healing of chronic venous leg ulcers.
        J Invest Dermatol. 2008; 128: 2526-2540
        • Wang J.
        • Zohar R.
        • McCulloch C.A.
        Multiple roles of alpha-smooth muscle actin in mechanotransduction.
        Exp Cell Res. 2006; 312: 205-214
        • Wipff P.J.
        • Majd H.
        • Acharya C.
        • et al.
        The covalent attachment of adhesion molecules to silicone membranes for cell stretching applications.
        Biomaterials. 2009; 30: 1781-1789
        • Wipff P.J.
        • Rifkin D.B.
        • Meister J.J.
        • et al.
        Myofibroblast contraction activates latent TGF-beta1 from the extracellular matrix.
        J Cell Biol. 2007; 179: 1311-1323
        • Wu L.
        • Xia Y.P.
        • Roth S.I.
        • et al.
        Transforming growth factor-beta1 fails to stimulate wound healing and impairs its signal transduction in an aged ischemic ulcer model: importance of oxygen and age.
        Am J Pathol. 1999; 154: 301-309
        • Xing D.
        • Bonanno J.A.
        Hypoxia reduces TGFbeta1-induced corneal keratocyte myofibroblast transformation.
        Mol Vis. 2009; 15: 1827-1834
        • Zhang B.
        • Liang X.
        • Shi W.
        • et al.
        Role of impaired peritubular capillary and hypoxia in progressive interstitial fibrosis after 56 subtotal nephrectomy of rats.
        Nephrology (Carlton). 2005; 10: 351-357