- Email: [email protected]
Platelet derived growth factor-CC secreted by M2 macrophages induces alpha-smooth muscle actin expression by dermal and gingival fibroblasts
Immunobiology 218 (2013) 924–929
Contents lists available at SciVerse ScienceDirect
Immunobiology journal homepage: www.elsevier.com/locate/imbio
Platelet derived growth factor-CC secreted by M2 macrophages induces alpha-smooth muscle actin expression by dermal and gingival fibroblasts Judith E. Glim a,b,∗ , Frank B. Niessen b , Vincent Everts c , Marjolein van Egmond a,d , Robert H.J. Beelen a a
Department of Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The Netherlands Department of Plastic and Reconstructive Surgery, VU University Medical Center, Amsterdam, The Netherlands Department of Oral Cell Biology, Academic Centre for Dentistry Amsterdam (ACTA), University of Amsterdam and VU University Amsterdam, Move Research Institute, Amsterdam, The Netherlands d Department of Surgery, VU University Medical Center, Amsterdam, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 25 September 2012 Received in revised form 16 October 2012 Accepted 17 October 2012 Available online 26 October 2012 Keywords: Alpha-SMA Fibrosis M1 M2 macrophages PDGF Wound healing
a b s t r a c t Dermal wounds can heal detrimentally by formation of excess fibrosis or hypertrophic scarring. These phenomena are normally absent in the oral mucosa. Macrophages play an important role in wound repair, have a marked heterogeneity and are thought to contribute to fibrosis. To investigate to what extend macrophages are involved in the occurrence of fibrosis, the effect of differently activated macrophages on dermal and gingival fibroblasts was studied in vitro. Macrophages were differentiated into a classical (M1) or alternative (M2) phenotype, which was assessed by receptor expression (CD40/mannose receptor) and cytokine secretion (interleukin-4 and -12). Fibroblasts were exposed to these macrophages and/or conditioned medium (cm), and differentiation into ␣-SMA-expressing myofibroblasts was quantified. M2, but not M1 macrophages induced ␣-SMA expression in both dermal and gingival fibroblasts. Blocking of transforming growth factor-1 did not decrease the ␣-SMA expression mediated by M2 macrophages. It appeared that this induction was mediated by platelet derived growth factor-CC (PDGF-CC), produced by M2 macrophages. The expression and role of this growth factor was confirmed by ELISA, RT-PCR, and blocking experiments. Our results indicate that M2 macrophages are able to induce myofibroblast differentiation via production of PDGF-CC. Based on our findings we conclude that PDGF-CC may play a hitherto unknown role in the differentiation of myofibroblasts. © 2012 Elsevier GmbH. All rights reserved.
Introduction Wound healing displays a dynamic procedure involving many cells and mediators. The complete process comprises different phases that have been termed coagulation, inflammation, proliferation and remodelling (Reinke and Sorg 2012). One of the cell types that play an essential role in wound healing is the fibroblast. These cells maintain homeostasis of intact tissue by remodelling the extracellular matrix (ECM), and increase the ECM production upon wounding. A proportion of fibroblasts that are located in the vicinity of the wound differentiate into myofibroblasts. These cells are specialized
Abbreviations: ␣-SMA, alpha smooth muscle actin; CM, conditioned medium; ECM, extracellular matrix; IFN-␥, interferon gamma; IL, interleukin; LPS, lipopolysaccharide; MR, mannose receptor; PDGF-CC, platelet derived growth factor-CC; TGF-1, transforming growth factor-beta 1. ∗ Corresponding author at: Department of Molecular Cell Biology and Immunology, VU University Medical Center, PO Box 7057, 1007 MB Amsterdam, The Netherlands. Tel.: +31 20 4448083; fax: +31 20 4448081. E-mail address: [email protected] (J.E. Glim). 0171-2985/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.imbio.2012.10.004
fibroblasts that share characteristics of both fibroblasts and smooth muscle cells. They express ␣-smooth muscle actin (␣-SMA), have high contractile capacities and produce abundant ECM proteins (Hinz 2007). One of the main factors that is considered responsible for modulation of fibroblast into myofibroblasts is transforming growth factor-1 (TGF-1) (Hinz 2010). Myofibroblasts undergo apoptosis when the wound is fully covered by epithelium (Desmouliere et al. 1995), but they persist when a wound is repaired in an aberrant way. Excessive contraction and over-production of ECM by myofibroblasts can cause tissue fibrosis and the scar can become hypertrophic. Hypertrophic scars are elevated above the skin, can be stiff, have a reddish appearance and contractile capacities (van der Veer et al. 2009). Furthermore, increased numbers of fibroblasts and myofibroblasts were found in hypertrophic scars compared to normotrophic scars and intact skin (Nedelec et al. 2001). Next to fibroblasts, macrophages play an important role in wound repair and ablation of these cells greatly affects wound healing. Macrophage depletion in mice resulted in delayed re-epithelialisation, reduced collagen deposition, and impaired angiogenesis (Mirza et al. 2009). Moreover, fibroblasts failed to
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
differentiate into myofibroblasts in macrophage-depleted wounds (Goren et al. 2009). Macrophages form a heterogeneous population of which the M1 and M2 macrophages represent the two most extreme subtypes (Mosser and Edwards 2008). Their differentiation depends on a series of environmental factors. M1 or classically activated macrophages are activated through Toll-like receptors and interferon (IFN)-␥, and are involved in pro-inflammatory responses. M2 or alternatively activated macrophages are activated by anti-inflammatory mediators such as interleukin (IL)-4, IL-10, IL13 or glucocorticoids, and involved in T-helper 2 responses (Varin and Gordon 2009). It is thought that especially M2 macrophages are associated with tissue repair. These macrophages secrete a variety of growth factors and cytokines, thereby modulating fibroblast activity and triggering excessive accumulation of ECM components (Gratchev et al. 2005; Song et al. 2000). Contributing to that, it is thought that M2 macrophages are involved in the occurrence of fibrosis. For example, alveolar macrophages of patients with idiopathic pulmonary fibrosis had an increased M2 phenotype compared to healthy controls (Pechkovsky et al. 2010). In contrast to adult tissue repair, foetal wound healing is associated with ‘scarless’ healing in the first and second trimester of development (Wilgus 2007). Since the foetal immune system is immature in the first developmental phases, it was postulated that the immune system might be involved in the switch between regeneration and fibrotic healing (Wilgus 2007; Redd et al. 2004). Interestingly, intra-oral mucosal healing resembles foetallike healing, characterized by rapid repair and re-epithelialisation with minimal scar formation (Wong et al. 2009). Differences between skin and oral wound repair were attributed to the fibroblasts phenotype, contraction rate, and ␣-SMA expression (Shannon et al. 2006), although the exact cause of scarless healing remains thus far unknown. To further elucidate the role of M1 and M2 macrophages in dermal versus oral wound healing in more detail, we investigated the effect of both types of macrophages on dermal and gingival fibroblasts. The expression of ␣-SMA was assessed as a parameter for fibroblast-to-myofibroblast differentiation.
Methods Isolation of human dermal and gingival fibroblasts Dermal fibroblasts were isolated from split skin biopsies obtained from three healthy donors undergoing abdominal dermolipectomy. Consent according to institutional guidelines was obtained from all donors. Isolation was performed as described by van den Bogaerdt et al. (2002). In short, the dermis was dissected from the epidermis by 0.25% (w/v) dispase II (Boehringer, Mannheim, Germany). For isolation of fibroblasts, the dermis was dissected and incubated for 2 h at 37 ◦ C in 0.25% (w/v) collagenase A/dispase II (Boehringer). Gingival fibroblasts were derived from individuals who underwent extraction of a third molar and signed consent was obtained from all donors. Cells were isolated as described by de Vries et al. (2006). Briefly, free gingiva was cut off the tooth and tissue fragments were washed twice in DMEM (Invitrogen, Breda, The Netherlands) supplemented with 10% FCS and 1% penicillin–streptomycin–amphotericin B (PSA, Sigma–Aldrich, St. Louis, MO, USA). The fragments were cut into pieces and divided in a six-well plate. Both dermal and gingival fibroblasts were cultured in DMEM with 10% FCS and 2% penicillin–streptomycin–glutamine (PSG) at 37 ◦ C, 5% CO2 and passages 4–6 were used for experiments.
925
Human macrophages Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy donor buffy coats (Sanquin Blood Bank, Amsterdam, The Netherlands) using Ficoll (LymphoprepTM, AxisShield, Oslo, Norway) density gradient. Subsequently, monocytes were isolated from the PBMC pool by Percoll (GE Healthcare, Uppsala, Sweden) density centrifugation. Monocytes were seeded in culture plates with a density of 1 × 106 cells/ml in DMEM, supplemented with 5% (v/v) human serum (BioWhittaker Verviers, Belgium) and 2% (v/v) PSG, at 37 ◦ C, 5% CO2 . After 5–7 days of culture, monocytes had differentiated into macrophages, which was confirmed by CD68 (anti-CD68, Clone EBM11, Dako, Glostrup, Denmark) expression. For induction of M1 macrophages, cells were exposed to 1000 U/ml recombinant human IFN-␥ (U-Cytech, Utrecht, The Netherlands) and 10 ng/ml Escherichia coli LPS (0111:B4 strain, LPSEB Ultrapure, Invivogen Toulouse, France). M2 macrophages were generated by incubation with 10 ng/ml recombinant human IL-4 (Immunotools, Friesoythe, Germany). After two days of stimulation supernatant was collected for further experiments. Macrophages were harvested with lidocaine (4 mg/ml; Sigma–Aldrich), added to fibroblasts, or fixed in PBS-formaldehyde 1% at 4 ◦ C. To assess whether macrophages had adapted either an M1 or M2 phenotype, cells were stained for CD40 (M1 macrophages, Serotec, Oxford, UK) and mannose receptor (M2 macrophages, BD Biosciences, San Jose, USA) (Aron-Wisnewsky et al. 2009). Analysis was performed with flow cytometry (FACSCaliburTM , Becton Dickinson, Erembodegem, Belgium). Furthermore, concentrations of IL-12, IL-4, IL-10 (Invitrogen), and platelet derived growth factor (PDGF)-CC (Uscn Life Science Inc., Wuhan, China) in macrophage supernatants were measured by ELISA. Cell culture experiments Fibroblasts were grown to confluence and then seeded on glass chamber slides (NUNC, New York, USA) with a density of 2 × 104 cells per slide. After 24 h cells were exposed to recombinant human TGF-1 (Biovision, Mountain View, USA), recombinant human PDGF-CC (R&D Systems, Abingdon, UK) or macrophage conditioned medium derived from M1 or M2 macrophages. For blocking experiments, 10 g/ml anti-TGF-1 antibody (2Ar2, Abcam, Cambridge, UK), 150 ng/ml anti-PDGF-CC antibody (clone 619346, R&D Systems) or isotype controls (anti-MOG IgG1 antibody, generated in our lab; Mouse IgG2b, Abcam) were used. Cell viability, was assessed by WST-1 reagent (Roche, Almere, The Netherlands) and absorbance was measured at 450 nm (Model 680 Microplate reader, Bio-Rad Laboratories, Veenendaal, The Netherlands). After 72 h of incubation, cells were washed and fixed in acetone or lysed in sample buffer (containing 1 M Tris–HCl pH 6.8, 50% glycerol, 20% SDS, and 0.2% bromophenol blue). For transwell experiments, fibroblasts (1 × 105 cells per well) were seeded on coverslips and 24 h later M1 and M2 macrophages (1 × 105 cells per well) were added in the upper transwell (0.4 m membrane, Corning Costar, Cambridge, MA) compartment. After 72 h cells were fixed. Immunohistochemistry Subsequently to fixation the endogenous peroxidase activity in cells was blocked by 0.25% H2 O2 , for 5 min. Mouse monoclonal antibody to ␣-SMA (1A4, Abcam) was diluted (1:100) in PBS and applied to the cells. After 1 h of incubation at room temperature, cells were rinsed in PBS and incubated with Envision (Dako) for 30 min. Detection was performed using diaminobenzidine (DAB, Dako) chromogen substrate and the nuclei were
926
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
Fig. 1. Phenotypic characteristics of M1 and M2 macrophages. (A) CD40 and (B) mannose receptor (MR) expression on differently activated macrophages was determined by FACS analysis. (C) Quantification of CD40 and MR expression on M1 and M2 macrophages. CD40 is highly expressed by M1 macrophages and MR by M2 macrophages. (D) IL-12 analysis by ELISA showed an abundant release by M1 macrophages, and (E) IL-4 production was the highest in M2 macrophages. *P < 0.05, ***P < 0.001.
visualized by haematoxylin. The cells were dehydrated in ethanol and embedded in entellan. Five randomly taken microscopic images (Nikon Eclipse E800, Nikon Instruments Inc., Melville, USA) were made of every condition and the percentage of ␣SMA positive pixels per area was analysed. This was mediated by computerized scoring by Image J software. First, a threshold correction was set and the amount of positive staining was measured as percentage of the total image area. The average of the five pictures was depicted as the final score for every condition. Western blotting Samples were denatured and subjected to 10% sodium dodecyl sulphate polyacrylamide (SDS-PAGE) gels. The separated protein fractions were blotted to polyvinylidene difluoride (PVDF) membranes (Bio-Rad), blocked with Odyssey® Blocking Buffer (LI-COR Biosciences, Bad Homburg, Germany), and incubated with anti-␣SMA antibody (Abcam) overnight at 4 ◦ C. Blots were then washed with PBS-Tween 0.1% and incubated with fluorescently labelled goat anti-mouse antibody (LI-COR Biosciences). Anti-GAPDH antibody (Cell Signalling Technology, Leiden, The Netherlands) was used as a loading control and detected with the secondary antibody goat anti-rabbit (LI-COR Biosciences). Protein bands were made visible with the Odyssey Infrared Imaging System (LI-COR Biosciences). Real-time polymerase chain reaction (RT-PCR) Total RNA was isolated (Roche) and reverse transcribed into cDNA using oligo(dT) and random primers (Promega, Mannheim, Germany), according to the manufacturer’s protocols. Subsequently, RT-PCR (ViiA7, Applied Biosystems) was performed for PDGF-CC and GAPDH using the following primer pairs (5 –3 ): GAPDH forward: CCATGTTCGTCATGGGTGTG, reverse: GGTGCTAAGCAGTTGGTGGTG, PDGF-CC forward: CAAGGAACAGAACGGAGTAC, reverse: GAGGAAACCTTGGGCTGTG. The expression levels of PDGF-CC were normalized to GAPDH mRNA expression. Statistical analysis Statistical analysis was performed by GraphPad Prism (GraphPad Software Inc., San Diego, USA), using a one-way ANOVA with Bonferroni correction or unpaired t-test. P values < 0.05 were considered as statistically significant.
Results Characterization of M1 and M2 macrophages To determine whether macrophages had adapted an M1 or M2 phenotype, the expression of CD40 and mannose receptor was assessed using flow cytometry (Fig. 1A–C). Both macrophage subtypes expressed CD40 and mannose receptor, though CD40 was highly expressed on M1 macrophages, whereas mannose receptor was increased on M2 macrophages. In addition to membrane marker expression, cytokine secretion was assessed. M1 macrophages released high levels of IL-12, while this cytokine was not secreted by M2 macrophages (Fig. 1D). IL-4 was only produced by M2 macrophages (Fig. 1E) and levels of IL-10, which is considered to be released by M2 macrophages, were too low to detect. From these observations we concluded that IFN-␥/LPS or IL-4 treatment differentiated macrophages into an M1 or M2 phenotype, respectively. M2 macrophages induced ˛-SMA expression in fibroblasts Next, we investigated whether M1 and M2 macrophages had different effects on fibroblasts, with respect to ␣-SMA expression. Therefore, dermal and gingival fibroblasts were exposed to macrophages in a transwell system. In contrast to M1, M2 macrophages induced ␣-SMA protein expression by dermal fibroblasts (Fig. 2A). In gingival fibroblasts a slight increase of ␣-SMA expression was found, although this rise was statistically not significant (Fig. 2B). TGF-1 was included as positive control, and indeed induced ␣-SMA expression in both dermal and gingival fibroblasts (Fig. 2A and B). Interestingly, basal levels of ␣-SMA were higher in gingival compared to dermal fibroblasts (Fig. 2). Subsequently, we investigated whether the same effect was observed with conditioned medium from these macrophages. Again, conditioned medium obtained from M2 macrophages induced ␣-SMA in dermal fibroblasts (Fig. 3A). Now we also found a statistically significant induction with gingival fibroblasts (Fig. 3B). As a control IFN-␥/LPS and IL-4 were included, since the conditioned medium could still contain these agents. Neither IFN-␥/LPS nor IL-4 induced the ␣-SMA expression to the level of M2 conditioned medium. A western blot confirmed the observations found with the stainings (Fig. 3C). The macrophage-conditioned media and the control agents (IFN-␥/LPS and IL-4) did not have any significant effect on cell viability (data not shown).
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
927
Fig. 4. ␣-SMA expression in dermal fibroblasts in response to TGF-1 blocking antibodies. Anti-TGF-1 blocking antibodies reduced ␣-SMA expression in cells treated with TGF-1 by approximately 50%. M2 conditioned medium-dependent ␣-SMA induction was not reduced by anti-TGF-1. Isotype (IgG1) control did not have any effect on ␣-SMA expression. Data are representative for three separate experiments and significance was calculated compared to controls. *P < 0.05.
Furthermore, mRNA and protein levels of TGF-1 were not found to be higher in M2 compared to M1 macrophages (data not shown). ˛-SMA upregulation by PDGF-CC Fig. 2. ␣-SMA expression by dermal and gingival fibroblasts cultured with macrophages in transwells. (A) Dermal (DER) and (B) gingival (GIN) fibroblasts were cultured in transwells containing control medium, M1, M2 macrophages, or TGF1 (5 ng/ml). Significant numbers of dermal fibroblast expressed ␣-SMA following incubation with TGF-1 or M2 macrophages. Gingival fibroblasts also increased ␣SMA expression significantly upon incubation with TGF- 1, and a slight but not significant increase was induced by M2 macrophages. Data are representative for three separate experiments and significance was calculated compared to controls. **P < 0.01, ***P < 0.001.
M2-conditioned medium dependent ˛-SMA induction is not caused by TGF-ˇ1 Since M2 macrophages induced ␣-SMA expression in fibroblasts, the question arose by which mechanism this was achieved. One of the candidates responsible for this induction is TGF-1. After all, numerous data in the literature suggest that M2 macrophages produce this growth factor (Mantovani et al. 2002). To examine whether TGF-1 is indeed the factor causing increased ␣-SMA expression, the growth factor was blocked with an activity blocking antibody. As could be expected ␣-SMA expression induced by TGF1 was significantly decreased by addition of anti-TGF-1 antibody (Fig. 4). However, anti-TGF-1 did not reduce ␣-SMA expression in fibroblasts stimulated with M2-conditioned medium (Fig. 4).
The above results showed that TGF-1 is not the factor by which M2 macrophages induce ␣-SMA in fibroblasts. Besides TGF-1, other proteins such as platelet derived growth factor (PDGF)-CC, were proposed to elicit ␣-SMA expression (Ostendorf et al. 2012; Leask 2010). We investigated whether PDGF-CC is able to upregulate ␣-SMA in dermal and gingival fibroblasts. Fig. 5 illustrates that indeed more dermal fibroblasts are positive for ␣-SMA when cultured in the presence of PDGF-CC. Furthermore, the same effect was found in gingival fibroblasts (data not shown). Quantification of the stainings proved that significantly higher numbers of ␣-SMA cells were found upon addition of 10 and 20 ng/ml PDGF-CC (Fig. 5B). M2 macrophages have increased PDGF-CC expression Since it became clear that PDGF-CC has the capacity to elicit ␣SMA expression in fibroblasts, the relation between macrophages and PDGF-CC expression was examined. Macrophages were differentiated to an M1 or M2 phenotype for 48 h and mRNA levels were determined. Significantly higher PDGF-CC expression was found in M2 macrophages, compared to M1 (Fig. 6A). In addition, M2 macrophages also produced more PDGF-CC on protein level (Fig. 6B). These results indicate that the ␣-SMA inducing effect of
Fig. 3. ␣-SMA expression by dermal and gingival fibroblasts upon culturing with macrophage-cm. (A) Dermal (DER) and (B) gingival (GIN) fibroblasts were cultured in the presence of control medium, M1-cm, M2-cm or with TGF-1, IFN-␥+LPS, or IL-4. M2-cm and the positive control TGF-1 induced ␣-SMA in both dermal and gingival fibroblasts. (C) Western blot analysis of ␣-SMA confirming the observations found in (A) and (B). Data are representative for three separate experiments and significance was calculated compared to controls. *P < 0.05, **P < 0.01.
928
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
Fig. 5. Effect of recombinant PDGF-CC on fibroblast ␣-SMA expression. Dermal fibroblasts were treated with TGF-1 (5 ng/ml) or PDGF-CC (5, 10 or 20 ng/ml) for 72 h and stained for ␣-SMA. (A) ␣-SMA staining. (B) Analysis of ␣-SMA positive particles. ␣-SMA expression by dermal fibroblasts is induced PDGF-CC. Data are expressed as mean fold increase related to control values ± SEM of three independent experiments. *P < 0.05, **P < 0.01.
Fig. 6. PDGF-CC expression by macrophages. Macrophages were differentiated towards M1 or M2 for 48 h. (A) mRNA and (B) protein levels of PDGF-CC were measured. M2 macrophages have increased PDGF-CC expression on both mRNA and protein level, compared to M1 macrophages. Data are expressed as mean ± SEM of three independent experiments. **P < 0.01, ***P < 0.001.
M2 macrophages is at least in part mediated by secretion of PGDFCC. Blocking PDGF-CC reduced M2 macrophage mediated ˛-SMA induction Finally, to prove that the ␣-SMA induction in fibroblast by M2 macrophages is caused by PDGF-CC, the protein was blocked. M2conditioned medium was incubated with a neutralizing PDGF-CC antibody and added to dermal fibroblast. 50% reduction of ␣-SMA expression was found when the neutralizing antibody was added (Fig. 7). These results indicate that PDGF-CC is one of the main factors secreted by M2 macrophages that mediate ␣-SMA expression in fibroblasts.
Fig. 7. ␣-SMA expression by dermal fibroblasts in response to PDGF-CC blocking antibodies. Anti-PDGF-CC blocking antibodies (150 ng/ml) reduced ␣-SMA expression in fibroblast cultured with M2-conditioned medium by approximately 50%. Isotype (IgG2b) control did not have any effect on ␣-SMA expression. Data are expressed as mean fold increase related to M2-cm values ± SEM of three independent experiments. **P < 0.01.
Discussion In line with other studies (Lolmede et al. 2009; Mantovani et al. 2004) our investigation showed that IFN-␥ and LPS stimulation on macrophages leads to an upregulation of CD40 and secretion of IL-12, while treatment with IL-4 increased expression of mannose receptor and induced IL-4 production. Since these markers are attributed to M1 (CD40 and IL-12) and M2 (mannose receptor and IL-4) macrophages, we concluded that we had generated M1 and M2 macrophages in vitro. Next, we showed that M2, but not M1 macrophages, and its conditioned medium induce differentiation of dermal fibroblasts into ␣-SMA expressing myofibroblasts. Gingival fibroblasts also increased ␣-SMA expression upon culturing with M2-conditioned medium, but a slight though not significant increase was found when the fibroblasts were cultured with M2 macrophages in transwells. This might be due to the fact that in the transwell experiments the stimuli for M2 differentiation are absent, resulting in a reduced secretion profile by M2 macrophages. The threshold to respond on M2-secretions might be higher for gingival fibroblasts, compared to dermal fibroblasts. All together, these results suggest that there is a difference on macrophage level (M1 versus M2), but the response of these macrophages on ␣-SMA expression is almost equal between oral and dermal fibroblasts. Compared to dermal fibroblasts gingival fibroblasts had constitutively elevated expression of ␣-SMA, thus suggesting a considerable phenotypic difference between these two cell types. In line with our findings, this difference of constitutive expression in vitro was also found by Lygoe et al. (2007). Furthermore, in vivo in red Duroc pigs also enhanced numbers of ␣-SMA expressing myofibroblasts were observed in both wounded and unwounded oral mucosa, compared to the skin (Mak et al. 2009). In addition, gingival fibroblasts had increased contraction capacities, which is in compliance with the elevated ␣-SMA expression (Lygoe et al. 2007).
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
Blocking experiments revealed that TGF-1 was not the factor responsible for the M2-conditioned medium dependent ␣-SMA expression by fibroblasts. M2 macrophages however, had increased mRNA and protein levels of the ␣-SMA inducing protein PDGF-CC. Blocking PDGF-CC in M2 conditioned medium led to a reduced ␣-SMA expression in fibroblasts. From these observations we can conclude that PDGF-CC is one of the mediators by which M2 macrophages increase fibroblasts ␣-SMA expression. Overexpression of this growth factor could contribute to formation of fibrosis. For example, renal fibrosis was associated with overexpression of PDGF-CC (Eitner et al. 2008). Interestingly, it appeared that especially the macrophages in fibrotic kidneys were positive for PDGF-CC, although the authors did not investigate whether the macrophage phenotype was M1 or M2. PDGF-CC antibodies reduced myofibroblast accumulation and a reduction in the severity of fibrosis was found (Eitner et al. 2008). Moreover, impaired wound healing found in diabetic mice, was shown to be enhanced by application of PDGF-CC (Gilbertson et al. 2001). Together, these data indicate that low expression of PDGF-CC may be involved in delayed wound healing, while overexpression contributes to fibrosis. This means that PDGF-CC may presumably be involved in the pathway of fibrosis formation. Production of this growth factor is among others mediated by M2 macrophages, as shown in this study. M2 macrophages are beneficial for proper wound healing, although prolonged presence or overactivation of these macrophages could be detrimental and results in fibrosis. Besides ␣-SMA induction, M2 macrophages were shown to play a role in fibrosis formation at another level. Collagen production of lung fibroblasts was increased when the cells were co-cultured with M2 macrophages (Song et al. 2000). In summary, both dermal and gingival fibroblasts responded almost equally to macrophages regarding to ␣-SMA expression. In contrast to M1, only M2 macrophages induced ␣-SMA expression, which was mediated by PDGF-CC, in both fibroblast populations. We conclude that the M2 macrophage-induced myofibroblast differentiation may contribute to fibrosis and this phenomenon needs further examination. Furthermore, differences between dermal and oral healing may not be attributed to the response of fibroblasts to M2 macrophages, although it would be of importance to investigate the presence of these M2 macrophages in vivo in dermal versus oral wounds. Acknowledgements We thank T.J. de Vries and M.E. Blaauboer from the Department of Oral Cell Biology (ACTA) for providing gingival fibroblasts for the experiments described in this article. In addition, we are grateful to M. Vlig and M.M. Ullrich (VSBN) for the kind gift of the dermal fibroblasts. References Aron-Wisnewsky, J., Tordjman, J., Poitou, C., Darakhshan, F., Hugol, D., Basdevant, A., Aissat, A., Guerre-Millo, M., Clement, K., 2009. Human adipose tissue macrophages: M1 and M2 cell surface markers in subcutaneous and omental depots and after weight loss. J. Clin. Endocrinol. Metab. 94, 4619–4623. de Vries, T.J., Schoenmaker, T., Wattanaroonwong, N., van den, H.M., Nieuwenhuijse, A., Beertsen, W., Everts, V., 2006. Gingival fibroblasts are better at inhibiting osteoclast formation than periodontal ligament fibroblasts. J. Cell Biochem. 98, 370–382. Desmouliere, A., Redard, M., Darby, I., Gabbiani, G., 1995. Apoptosis mediates the decrease in cellularity during the transition between granulation tissue and scar. Am. J. Pathol. 146, 56–66.
929
Eitner, F., Bucher, E., van, R.C., Kunter, U., Rong, S., Seikrit, C., Villa, L., Boor, P., Fredriksson, L., Backstrom, G., Eriksson, U., Ostman, A., Floege, J., Ostendorf, T., 2008. PDGF-C is a proinflammatory cytokine that mediates renal interstitial fibrosis. J. Am. Soc. Nephrol. 19, 281–289. Gilbertson, D.G., Duff, M.E., West, J.W., Kelly, J.D., Sheppard, P.O., Hofstrand, P.D., Gao, Z., Shoemaker, K., Bukowski, T.R., Moore, M., Feldhaus, A.L., Humes, J.M., Palmer, T.E., Hart, C.E., 2001. Platelet-derived growth factor C (PDGF-C), a novel growth factor that binds to PDGF alpha and beta receptor. J. Biol. Chem. 20, 27406–27414, 276. Goren, I., Allmann, N., Yogev, N., Schurmann, C., Linke, A., Holdener, M., Waisman, A., Pfeilschifter, J., Frank, S., 2009. A transgenic mouse model of inducible macrophage depletion: effects of diphtheria toxin-driven lysozyme M-specific cell lineage ablation on wound inflammatory, angiogenic, and contractive processes. Am. J. Pathol. 175, 132–147. Gratchev, A., Kzhyshkowska, J., Utikal, J., Goerdt, S., 2005. Interleukin-4 and dexamethasone counterregulate extracellular matrix remodelling and phagocytosis in type-2 macrophages. Scand. J. Immunol. 61, 10–17. Hinz, B., 2007. Formation and function of the myofibroblast during tissue repair. J. Invest Dermatol. 127, 526–537. Hinz, B., 2010. The myofibroblast: paradigm for a mechanically active cell. J. Biomech. 43, 146–155. Leask, A., 2010. Potential therapeutic targets for cardiac fibrosis: TGFbeta, angiotensin, endothelin, CCN2, and PDGF, partners in fibroblast activation. Circ. Res. 106, 1675–1680. Lolmede, K., Campana, L., Vezzoli, M., Bosurgi, L., Tonlorenzi, R., Clementi, E., Bianchi, M.E., Cossu, G., Manfredi, A.A., Brunelli, S., Rovere-Querini, P., 2009. Inflammatory and alternatively activated human macrophages attract vessel-associated stem cells, relying on separate HMGB1- and MMP-9-dependent pathways. J. Leukoc. Biol. 85, 779–787. Lygoe, K.A., Wall, I., Stephens, P., Lewis, M.P., 2007. Role of vitronectin and fibronectin receptors in oral mucosal and dermal myofibroblast differentiation. Biol. Cell. 99, 601–614. Mak, K., Manji, A., Gallant-Behm, C., Wiebe, C., Hart, D.A., Larjava, H., Hakkinen, L., 2009. Scarless healing of oral mucosa is characterized by faster resolution of inflammation and control of myofibroblast action compared to skin wounds in the red Duroc pig model. J. Dermatol. Sci. 56, 168–180. Mantovani, A., Sica, A., Sozzani, S., Allavena, P., Vecchi, A., Locati, M., 2004. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol. 25, 677–686. Mantovani, A., Sozzani, S., Locati, M., Allavena, P., Sica, A., 2002. Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol. 23, 549–555. Mirza, R., DiPietro, L.A., Koh, T.J., 2009. Selective and specific macrophage ablation is detrimental to wound healing in mice. Am. J. Pathol. 175, 2454–2462. Mosser, D.M., Edwards, J.P., 2008. Exploring the full spectrum of macrophage activation. Nat. Rev. Immunol. 8, 958–969. Nedelec, B., Shankowsky, H., Scott, P.G., Ghahary, A., Tredget, E.E., 2001. Myofibroblasts and apoptosis in human hypertrophic scars: the effect of interferon-alpha2b. Surgery 130, 798–808. Ostendorf, T., Eitner, F., Floege, J., 2012. The PDGF family in renal fibrosis. Pediatr. Nephrol. 27, 1041–1050. Pechkovsky, D.V., Prasse, A., Kollert, F., Engel, K.M., Dentler, J., Luttmann, W., Friedrich, K., Muller-Quernheim, J., Zissel, G., 2010. Alternatively activated alveolar macrophages in pulmonary fibrosis-mediator production and intracellular signal transduction. Clin. Immunol. 137, 89–101. Redd, M.J., Cooper, L., Wood, W., Stramer, B., Martin, P., 2004. Wound healing and inflammation: embryos reveal the way to perfect repair. Philos. Trans. R. Soc. Lond B Biol. Sci. 359, 777–784. Reinke, J.M., Sorg, H., 2012. Wound repair and regeneration. Eur. Surg. Res. 49, 35–43. Shannon, D.B., McKeown, S.T., Lundy, F.T., Irwin, C.R., 2006. Phenotypic differences between oral and skin fibro blasts in wound contraction and growth factor expression. Wound Repair Regen. 14, 172–178. Song, E., Ouyang, N., Horbelt, M., Antus, B., Wang, M., Exton, M.S., 2000. Influence of alternatively and classically activated macrophages on fibrogenic activities of human fibroblasts. Cell Immunol. 204, 19–28. van den Bogaerdt, A.J., van Zuijlen, P.P., van, G.M., Lamme, E.N., Middelkoop, E., 2002. The suitability of cells from different tissues for use in tissue-engineered skin substitutes. Arch. Dermatol. Res. 294, 135–142. van der Veer, W., Bloemen, M.C., Ulrich, M.M., Molema, G., van Zuijlen, P.P., Middelkoop, E., Niessen, F.B., 2009. Potential cellular and molecular causes of hypertrophic scar formation. Burns 35, 15–29. Varin, A., Gordon, S., 2009. Alternative activation of macrophages: immune function and cellular biology. Immunobiology 214, 630–641. Wilgus, T.A., 2007. Regenerative healing in fetal skin: a review of the literature. Ostomy Wound Manage. 53, 16–31. Wong, J.W., Gallant-Behm, C., Wiebe, C., Mak, K., Hart, D.A., Larjava, H., Hakkinen, L., 2009. Wound healing in oral mucosa results in reduced scar formation as compared with skin: evidence from the red Duroc pig model and humans. Wound Repair Regen. 17, 717–729.
Contents lists available at SciVerse ScienceDirect
Immunobiology journal homepage: www.elsevier.com/locate/imbio
Platelet derived growth factor-CC secreted by M2 macrophages induces alpha-smooth muscle actin expression by dermal and gingival fibroblasts Judith E. Glim a,b,∗ , Frank B. Niessen b , Vincent Everts c , Marjolein van Egmond a,d , Robert H.J. Beelen a a
Department of Molecular Cell Biology and Immunology, VU University Medical Center, Amsterdam, The Netherlands Department of Plastic and Reconstructive Surgery, VU University Medical Center, Amsterdam, The Netherlands Department of Oral Cell Biology, Academic Centre for Dentistry Amsterdam (ACTA), University of Amsterdam and VU University Amsterdam, Move Research Institute, Amsterdam, The Netherlands d Department of Surgery, VU University Medical Center, Amsterdam, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 25 September 2012 Received in revised form 16 October 2012 Accepted 17 October 2012 Available online 26 October 2012 Keywords: Alpha-SMA Fibrosis M1 M2 macrophages PDGF Wound healing
a b s t r a c t Dermal wounds can heal detrimentally by formation of excess fibrosis or hypertrophic scarring. These phenomena are normally absent in the oral mucosa. Macrophages play an important role in wound repair, have a marked heterogeneity and are thought to contribute to fibrosis. To investigate to what extend macrophages are involved in the occurrence of fibrosis, the effect of differently activated macrophages on dermal and gingival fibroblasts was studied in vitro. Macrophages were differentiated into a classical (M1) or alternative (M2) phenotype, which was assessed by receptor expression (CD40/mannose receptor) and cytokine secretion (interleukin-4 and -12). Fibroblasts were exposed to these macrophages and/or conditioned medium (cm), and differentiation into ␣-SMA-expressing myofibroblasts was quantified. M2, but not M1 macrophages induced ␣-SMA expression in both dermal and gingival fibroblasts. Blocking of transforming growth factor-1 did not decrease the ␣-SMA expression mediated by M2 macrophages. It appeared that this induction was mediated by platelet derived growth factor-CC (PDGF-CC), produced by M2 macrophages. The expression and role of this growth factor was confirmed by ELISA, RT-PCR, and blocking experiments. Our results indicate that M2 macrophages are able to induce myofibroblast differentiation via production of PDGF-CC. Based on our findings we conclude that PDGF-CC may play a hitherto unknown role in the differentiation of myofibroblasts. © 2012 Elsevier GmbH. All rights reserved.
Introduction Wound healing displays a dynamic procedure involving many cells and mediators. The complete process comprises different phases that have been termed coagulation, inflammation, proliferation and remodelling (Reinke and Sorg 2012). One of the cell types that play an essential role in wound healing is the fibroblast. These cells maintain homeostasis of intact tissue by remodelling the extracellular matrix (ECM), and increase the ECM production upon wounding. A proportion of fibroblasts that are located in the vicinity of the wound differentiate into myofibroblasts. These cells are specialized
Abbreviations: ␣-SMA, alpha smooth muscle actin; CM, conditioned medium; ECM, extracellular matrix; IFN-␥, interferon gamma; IL, interleukin; LPS, lipopolysaccharide; MR, mannose receptor; PDGF-CC, platelet derived growth factor-CC; TGF-1, transforming growth factor-beta 1. ∗ Corresponding author at: Department of Molecular Cell Biology and Immunology, VU University Medical Center, PO Box 7057, 1007 MB Amsterdam, The Netherlands. Tel.: +31 20 4448083; fax: +31 20 4448081. E-mail address: [email protected] (J.E. Glim). 0171-2985/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.imbio.2012.10.004
fibroblasts that share characteristics of both fibroblasts and smooth muscle cells. They express ␣-smooth muscle actin (␣-SMA), have high contractile capacities and produce abundant ECM proteins (Hinz 2007). One of the main factors that is considered responsible for modulation of fibroblast into myofibroblasts is transforming growth factor-1 (TGF-1) (Hinz 2010). Myofibroblasts undergo apoptosis when the wound is fully covered by epithelium (Desmouliere et al. 1995), but they persist when a wound is repaired in an aberrant way. Excessive contraction and over-production of ECM by myofibroblasts can cause tissue fibrosis and the scar can become hypertrophic. Hypertrophic scars are elevated above the skin, can be stiff, have a reddish appearance and contractile capacities (van der Veer et al. 2009). Furthermore, increased numbers of fibroblasts and myofibroblasts were found in hypertrophic scars compared to normotrophic scars and intact skin (Nedelec et al. 2001). Next to fibroblasts, macrophages play an important role in wound repair and ablation of these cells greatly affects wound healing. Macrophage depletion in mice resulted in delayed re-epithelialisation, reduced collagen deposition, and impaired angiogenesis (Mirza et al. 2009). Moreover, fibroblasts failed to
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
differentiate into myofibroblasts in macrophage-depleted wounds (Goren et al. 2009). Macrophages form a heterogeneous population of which the M1 and M2 macrophages represent the two most extreme subtypes (Mosser and Edwards 2008). Their differentiation depends on a series of environmental factors. M1 or classically activated macrophages are activated through Toll-like receptors and interferon (IFN)-␥, and are involved in pro-inflammatory responses. M2 or alternatively activated macrophages are activated by anti-inflammatory mediators such as interleukin (IL)-4, IL-10, IL13 or glucocorticoids, and involved in T-helper 2 responses (Varin and Gordon 2009). It is thought that especially M2 macrophages are associated with tissue repair. These macrophages secrete a variety of growth factors and cytokines, thereby modulating fibroblast activity and triggering excessive accumulation of ECM components (Gratchev et al. 2005; Song et al. 2000). Contributing to that, it is thought that M2 macrophages are involved in the occurrence of fibrosis. For example, alveolar macrophages of patients with idiopathic pulmonary fibrosis had an increased M2 phenotype compared to healthy controls (Pechkovsky et al. 2010). In contrast to adult tissue repair, foetal wound healing is associated with ‘scarless’ healing in the first and second trimester of development (Wilgus 2007). Since the foetal immune system is immature in the first developmental phases, it was postulated that the immune system might be involved in the switch between regeneration and fibrotic healing (Wilgus 2007; Redd et al. 2004). Interestingly, intra-oral mucosal healing resembles foetallike healing, characterized by rapid repair and re-epithelialisation with minimal scar formation (Wong et al. 2009). Differences between skin and oral wound repair were attributed to the fibroblasts phenotype, contraction rate, and ␣-SMA expression (Shannon et al. 2006), although the exact cause of scarless healing remains thus far unknown. To further elucidate the role of M1 and M2 macrophages in dermal versus oral wound healing in more detail, we investigated the effect of both types of macrophages on dermal and gingival fibroblasts. The expression of ␣-SMA was assessed as a parameter for fibroblast-to-myofibroblast differentiation.
Methods Isolation of human dermal and gingival fibroblasts Dermal fibroblasts were isolated from split skin biopsies obtained from three healthy donors undergoing abdominal dermolipectomy. Consent according to institutional guidelines was obtained from all donors. Isolation was performed as described by van den Bogaerdt et al. (2002). In short, the dermis was dissected from the epidermis by 0.25% (w/v) dispase II (Boehringer, Mannheim, Germany). For isolation of fibroblasts, the dermis was dissected and incubated for 2 h at 37 ◦ C in 0.25% (w/v) collagenase A/dispase II (Boehringer). Gingival fibroblasts were derived from individuals who underwent extraction of a third molar and signed consent was obtained from all donors. Cells were isolated as described by de Vries et al. (2006). Briefly, free gingiva was cut off the tooth and tissue fragments were washed twice in DMEM (Invitrogen, Breda, The Netherlands) supplemented with 10% FCS and 1% penicillin–streptomycin–amphotericin B (PSA, Sigma–Aldrich, St. Louis, MO, USA). The fragments were cut into pieces and divided in a six-well plate. Both dermal and gingival fibroblasts were cultured in DMEM with 10% FCS and 2% penicillin–streptomycin–glutamine (PSG) at 37 ◦ C, 5% CO2 and passages 4–6 were used for experiments.
925
Human macrophages Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy donor buffy coats (Sanquin Blood Bank, Amsterdam, The Netherlands) using Ficoll (LymphoprepTM, AxisShield, Oslo, Norway) density gradient. Subsequently, monocytes were isolated from the PBMC pool by Percoll (GE Healthcare, Uppsala, Sweden) density centrifugation. Monocytes were seeded in culture plates with a density of 1 × 106 cells/ml in DMEM, supplemented with 5% (v/v) human serum (BioWhittaker Verviers, Belgium) and 2% (v/v) PSG, at 37 ◦ C, 5% CO2 . After 5–7 days of culture, monocytes had differentiated into macrophages, which was confirmed by CD68 (anti-CD68, Clone EBM11, Dako, Glostrup, Denmark) expression. For induction of M1 macrophages, cells were exposed to 1000 U/ml recombinant human IFN-␥ (U-Cytech, Utrecht, The Netherlands) and 10 ng/ml Escherichia coli LPS (0111:B4 strain, LPSEB Ultrapure, Invivogen Toulouse, France). M2 macrophages were generated by incubation with 10 ng/ml recombinant human IL-4 (Immunotools, Friesoythe, Germany). After two days of stimulation supernatant was collected for further experiments. Macrophages were harvested with lidocaine (4 mg/ml; Sigma–Aldrich), added to fibroblasts, or fixed in PBS-formaldehyde 1% at 4 ◦ C. To assess whether macrophages had adapted either an M1 or M2 phenotype, cells were stained for CD40 (M1 macrophages, Serotec, Oxford, UK) and mannose receptor (M2 macrophages, BD Biosciences, San Jose, USA) (Aron-Wisnewsky et al. 2009). Analysis was performed with flow cytometry (FACSCaliburTM , Becton Dickinson, Erembodegem, Belgium). Furthermore, concentrations of IL-12, IL-4, IL-10 (Invitrogen), and platelet derived growth factor (PDGF)-CC (Uscn Life Science Inc., Wuhan, China) in macrophage supernatants were measured by ELISA. Cell culture experiments Fibroblasts were grown to confluence and then seeded on glass chamber slides (NUNC, New York, USA) with a density of 2 × 104 cells per slide. After 24 h cells were exposed to recombinant human TGF-1 (Biovision, Mountain View, USA), recombinant human PDGF-CC (R&D Systems, Abingdon, UK) or macrophage conditioned medium derived from M1 or M2 macrophages. For blocking experiments, 10 g/ml anti-TGF-1 antibody (2Ar2, Abcam, Cambridge, UK), 150 ng/ml anti-PDGF-CC antibody (clone 619346, R&D Systems) or isotype controls (anti-MOG IgG1 antibody, generated in our lab; Mouse IgG2b, Abcam) were used. Cell viability, was assessed by WST-1 reagent (Roche, Almere, The Netherlands) and absorbance was measured at 450 nm (Model 680 Microplate reader, Bio-Rad Laboratories, Veenendaal, The Netherlands). After 72 h of incubation, cells were washed and fixed in acetone or lysed in sample buffer (containing 1 M Tris–HCl pH 6.8, 50% glycerol, 20% SDS, and 0.2% bromophenol blue). For transwell experiments, fibroblasts (1 × 105 cells per well) were seeded on coverslips and 24 h later M1 and M2 macrophages (1 × 105 cells per well) were added in the upper transwell (0.4 m membrane, Corning Costar, Cambridge, MA) compartment. After 72 h cells were fixed. Immunohistochemistry Subsequently to fixation the endogenous peroxidase activity in cells was blocked by 0.25% H2 O2 , for 5 min. Mouse monoclonal antibody to ␣-SMA (1A4, Abcam) was diluted (1:100) in PBS and applied to the cells. After 1 h of incubation at room temperature, cells were rinsed in PBS and incubated with Envision (Dako) for 30 min. Detection was performed using diaminobenzidine (DAB, Dako) chromogen substrate and the nuclei were
926
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
Fig. 1. Phenotypic characteristics of M1 and M2 macrophages. (A) CD40 and (B) mannose receptor (MR) expression on differently activated macrophages was determined by FACS analysis. (C) Quantification of CD40 and MR expression on M1 and M2 macrophages. CD40 is highly expressed by M1 macrophages and MR by M2 macrophages. (D) IL-12 analysis by ELISA showed an abundant release by M1 macrophages, and (E) IL-4 production was the highest in M2 macrophages. *P < 0.05, ***P < 0.001.
visualized by haematoxylin. The cells were dehydrated in ethanol and embedded in entellan. Five randomly taken microscopic images (Nikon Eclipse E800, Nikon Instruments Inc., Melville, USA) were made of every condition and the percentage of ␣SMA positive pixels per area was analysed. This was mediated by computerized scoring by Image J software. First, a threshold correction was set and the amount of positive staining was measured as percentage of the total image area. The average of the five pictures was depicted as the final score for every condition. Western blotting Samples were denatured and subjected to 10% sodium dodecyl sulphate polyacrylamide (SDS-PAGE) gels. The separated protein fractions were blotted to polyvinylidene difluoride (PVDF) membranes (Bio-Rad), blocked with Odyssey® Blocking Buffer (LI-COR Biosciences, Bad Homburg, Germany), and incubated with anti-␣SMA antibody (Abcam) overnight at 4 ◦ C. Blots were then washed with PBS-Tween 0.1% and incubated with fluorescently labelled goat anti-mouse antibody (LI-COR Biosciences). Anti-GAPDH antibody (Cell Signalling Technology, Leiden, The Netherlands) was used as a loading control and detected with the secondary antibody goat anti-rabbit (LI-COR Biosciences). Protein bands were made visible with the Odyssey Infrared Imaging System (LI-COR Biosciences). Real-time polymerase chain reaction (RT-PCR) Total RNA was isolated (Roche) and reverse transcribed into cDNA using oligo(dT) and random primers (Promega, Mannheim, Germany), according to the manufacturer’s protocols. Subsequently, RT-PCR (ViiA7, Applied Biosystems) was performed for PDGF-CC and GAPDH using the following primer pairs (5 –3 ): GAPDH forward: CCATGTTCGTCATGGGTGTG, reverse: GGTGCTAAGCAGTTGGTGGTG, PDGF-CC forward: CAAGGAACAGAACGGAGTAC, reverse: GAGGAAACCTTGGGCTGTG. The expression levels of PDGF-CC were normalized to GAPDH mRNA expression. Statistical analysis Statistical analysis was performed by GraphPad Prism (GraphPad Software Inc., San Diego, USA), using a one-way ANOVA with Bonferroni correction or unpaired t-test. P values < 0.05 were considered as statistically significant.
Results Characterization of M1 and M2 macrophages To determine whether macrophages had adapted an M1 or M2 phenotype, the expression of CD40 and mannose receptor was assessed using flow cytometry (Fig. 1A–C). Both macrophage subtypes expressed CD40 and mannose receptor, though CD40 was highly expressed on M1 macrophages, whereas mannose receptor was increased on M2 macrophages. In addition to membrane marker expression, cytokine secretion was assessed. M1 macrophages released high levels of IL-12, while this cytokine was not secreted by M2 macrophages (Fig. 1D). IL-4 was only produced by M2 macrophages (Fig. 1E) and levels of IL-10, which is considered to be released by M2 macrophages, were too low to detect. From these observations we concluded that IFN-␥/LPS or IL-4 treatment differentiated macrophages into an M1 or M2 phenotype, respectively. M2 macrophages induced ˛-SMA expression in fibroblasts Next, we investigated whether M1 and M2 macrophages had different effects on fibroblasts, with respect to ␣-SMA expression. Therefore, dermal and gingival fibroblasts were exposed to macrophages in a transwell system. In contrast to M1, M2 macrophages induced ␣-SMA protein expression by dermal fibroblasts (Fig. 2A). In gingival fibroblasts a slight increase of ␣-SMA expression was found, although this rise was statistically not significant (Fig. 2B). TGF-1 was included as positive control, and indeed induced ␣-SMA expression in both dermal and gingival fibroblasts (Fig. 2A and B). Interestingly, basal levels of ␣-SMA were higher in gingival compared to dermal fibroblasts (Fig. 2). Subsequently, we investigated whether the same effect was observed with conditioned medium from these macrophages. Again, conditioned medium obtained from M2 macrophages induced ␣-SMA in dermal fibroblasts (Fig. 3A). Now we also found a statistically significant induction with gingival fibroblasts (Fig. 3B). As a control IFN-␥/LPS and IL-4 were included, since the conditioned medium could still contain these agents. Neither IFN-␥/LPS nor IL-4 induced the ␣-SMA expression to the level of M2 conditioned medium. A western blot confirmed the observations found with the stainings (Fig. 3C). The macrophage-conditioned media and the control agents (IFN-␥/LPS and IL-4) did not have any significant effect on cell viability (data not shown).
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
927
Fig. 4. ␣-SMA expression in dermal fibroblasts in response to TGF-1 blocking antibodies. Anti-TGF-1 blocking antibodies reduced ␣-SMA expression in cells treated with TGF-1 by approximately 50%. M2 conditioned medium-dependent ␣-SMA induction was not reduced by anti-TGF-1. Isotype (IgG1) control did not have any effect on ␣-SMA expression. Data are representative for three separate experiments and significance was calculated compared to controls. *P < 0.05.
Furthermore, mRNA and protein levels of TGF-1 were not found to be higher in M2 compared to M1 macrophages (data not shown). ˛-SMA upregulation by PDGF-CC Fig. 2. ␣-SMA expression by dermal and gingival fibroblasts cultured with macrophages in transwells. (A) Dermal (DER) and (B) gingival (GIN) fibroblasts were cultured in transwells containing control medium, M1, M2 macrophages, or TGF1 (5 ng/ml). Significant numbers of dermal fibroblast expressed ␣-SMA following incubation with TGF-1 or M2 macrophages. Gingival fibroblasts also increased ␣SMA expression significantly upon incubation with TGF- 1, and a slight but not significant increase was induced by M2 macrophages. Data are representative for three separate experiments and significance was calculated compared to controls. **P < 0.01, ***P < 0.001.
M2-conditioned medium dependent ˛-SMA induction is not caused by TGF-ˇ1 Since M2 macrophages induced ␣-SMA expression in fibroblasts, the question arose by which mechanism this was achieved. One of the candidates responsible for this induction is TGF-1. After all, numerous data in the literature suggest that M2 macrophages produce this growth factor (Mantovani et al. 2002). To examine whether TGF-1 is indeed the factor causing increased ␣-SMA expression, the growth factor was blocked with an activity blocking antibody. As could be expected ␣-SMA expression induced by TGF1 was significantly decreased by addition of anti-TGF-1 antibody (Fig. 4). However, anti-TGF-1 did not reduce ␣-SMA expression in fibroblasts stimulated with M2-conditioned medium (Fig. 4).
The above results showed that TGF-1 is not the factor by which M2 macrophages induce ␣-SMA in fibroblasts. Besides TGF-1, other proteins such as platelet derived growth factor (PDGF)-CC, were proposed to elicit ␣-SMA expression (Ostendorf et al. 2012; Leask 2010). We investigated whether PDGF-CC is able to upregulate ␣-SMA in dermal and gingival fibroblasts. Fig. 5 illustrates that indeed more dermal fibroblasts are positive for ␣-SMA when cultured in the presence of PDGF-CC. Furthermore, the same effect was found in gingival fibroblasts (data not shown). Quantification of the stainings proved that significantly higher numbers of ␣-SMA cells were found upon addition of 10 and 20 ng/ml PDGF-CC (Fig. 5B). M2 macrophages have increased PDGF-CC expression Since it became clear that PDGF-CC has the capacity to elicit ␣SMA expression in fibroblasts, the relation between macrophages and PDGF-CC expression was examined. Macrophages were differentiated to an M1 or M2 phenotype for 48 h and mRNA levels were determined. Significantly higher PDGF-CC expression was found in M2 macrophages, compared to M1 (Fig. 6A). In addition, M2 macrophages also produced more PDGF-CC on protein level (Fig. 6B). These results indicate that the ␣-SMA inducing effect of
Fig. 3. ␣-SMA expression by dermal and gingival fibroblasts upon culturing with macrophage-cm. (A) Dermal (DER) and (B) gingival (GIN) fibroblasts were cultured in the presence of control medium, M1-cm, M2-cm or with TGF-1, IFN-␥+LPS, or IL-4. M2-cm and the positive control TGF-1 induced ␣-SMA in both dermal and gingival fibroblasts. (C) Western blot analysis of ␣-SMA confirming the observations found in (A) and (B). Data are representative for three separate experiments and significance was calculated compared to controls. *P < 0.05, **P < 0.01.
928
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
Fig. 5. Effect of recombinant PDGF-CC on fibroblast ␣-SMA expression. Dermal fibroblasts were treated with TGF-1 (5 ng/ml) or PDGF-CC (5, 10 or 20 ng/ml) for 72 h and stained for ␣-SMA. (A) ␣-SMA staining. (B) Analysis of ␣-SMA positive particles. ␣-SMA expression by dermal fibroblasts is induced PDGF-CC. Data are expressed as mean fold increase related to control values ± SEM of three independent experiments. *P < 0.05, **P < 0.01.
Fig. 6. PDGF-CC expression by macrophages. Macrophages were differentiated towards M1 or M2 for 48 h. (A) mRNA and (B) protein levels of PDGF-CC were measured. M2 macrophages have increased PDGF-CC expression on both mRNA and protein level, compared to M1 macrophages. Data are expressed as mean ± SEM of three independent experiments. **P < 0.01, ***P < 0.001.
M2 macrophages is at least in part mediated by secretion of PGDFCC. Blocking PDGF-CC reduced M2 macrophage mediated ˛-SMA induction Finally, to prove that the ␣-SMA induction in fibroblast by M2 macrophages is caused by PDGF-CC, the protein was blocked. M2conditioned medium was incubated with a neutralizing PDGF-CC antibody and added to dermal fibroblast. 50% reduction of ␣-SMA expression was found when the neutralizing antibody was added (Fig. 7). These results indicate that PDGF-CC is one of the main factors secreted by M2 macrophages that mediate ␣-SMA expression in fibroblasts.
Fig. 7. ␣-SMA expression by dermal fibroblasts in response to PDGF-CC blocking antibodies. Anti-PDGF-CC blocking antibodies (150 ng/ml) reduced ␣-SMA expression in fibroblast cultured with M2-conditioned medium by approximately 50%. Isotype (IgG2b) control did not have any effect on ␣-SMA expression. Data are expressed as mean fold increase related to M2-cm values ± SEM of three independent experiments. **P < 0.01.
Discussion In line with other studies (Lolmede et al. 2009; Mantovani et al. 2004) our investigation showed that IFN-␥ and LPS stimulation on macrophages leads to an upregulation of CD40 and secretion of IL-12, while treatment with IL-4 increased expression of mannose receptor and induced IL-4 production. Since these markers are attributed to M1 (CD40 and IL-12) and M2 (mannose receptor and IL-4) macrophages, we concluded that we had generated M1 and M2 macrophages in vitro. Next, we showed that M2, but not M1 macrophages, and its conditioned medium induce differentiation of dermal fibroblasts into ␣-SMA expressing myofibroblasts. Gingival fibroblasts also increased ␣-SMA expression upon culturing with M2-conditioned medium, but a slight though not significant increase was found when the fibroblasts were cultured with M2 macrophages in transwells. This might be due to the fact that in the transwell experiments the stimuli for M2 differentiation are absent, resulting in a reduced secretion profile by M2 macrophages. The threshold to respond on M2-secretions might be higher for gingival fibroblasts, compared to dermal fibroblasts. All together, these results suggest that there is a difference on macrophage level (M1 versus M2), but the response of these macrophages on ␣-SMA expression is almost equal between oral and dermal fibroblasts. Compared to dermal fibroblasts gingival fibroblasts had constitutively elevated expression of ␣-SMA, thus suggesting a considerable phenotypic difference between these two cell types. In line with our findings, this difference of constitutive expression in vitro was also found by Lygoe et al. (2007). Furthermore, in vivo in red Duroc pigs also enhanced numbers of ␣-SMA expressing myofibroblasts were observed in both wounded and unwounded oral mucosa, compared to the skin (Mak et al. 2009). In addition, gingival fibroblasts had increased contraction capacities, which is in compliance with the elevated ␣-SMA expression (Lygoe et al. 2007).
J.E. Glim et al. / Immunobiology 218 (2013) 924–929
Blocking experiments revealed that TGF-1 was not the factor responsible for the M2-conditioned medium dependent ␣-SMA expression by fibroblasts. M2 macrophages however, had increased mRNA and protein levels of the ␣-SMA inducing protein PDGF-CC. Blocking PDGF-CC in M2 conditioned medium led to a reduced ␣-SMA expression in fibroblasts. From these observations we can conclude that PDGF-CC is one of the mediators by which M2 macrophages increase fibroblasts ␣-SMA expression. Overexpression of this growth factor could contribute to formation of fibrosis. For example, renal fibrosis was associated with overexpression of PDGF-CC (Eitner et al. 2008). Interestingly, it appeared that especially the macrophages in fibrotic kidneys were positive for PDGF-CC, although the authors did not investigate whether the macrophage phenotype was M1 or M2. PDGF-CC antibodies reduced myofibroblast accumulation and a reduction in the severity of fibrosis was found (Eitner et al. 2008). Moreover, impaired wound healing found in diabetic mice, was shown to be enhanced by application of PDGF-CC (Gilbertson et al. 2001). Together, these data indicate that low expression of PDGF-CC may be involved in delayed wound healing, while overexpression contributes to fibrosis. This means that PDGF-CC may presumably be involved in the pathway of fibrosis formation. Production of this growth factor is among others mediated by M2 macrophages, as shown in this study. M2 macrophages are beneficial for proper wound healing, although prolonged presence or overactivation of these macrophages could be detrimental and results in fibrosis. Besides ␣-SMA induction, M2 macrophages were shown to play a role in fibrosis formation at another level. Collagen production of lung fibroblasts was increased when the cells were co-cultured with M2 macrophages (Song et al. 2000). In summary, both dermal and gingival fibroblasts responded almost equally to macrophages regarding to ␣-SMA expression. In contrast to M1, only M2 macrophages induced ␣-SMA expression, which was mediated by PDGF-CC, in both fibroblast populations. We conclude that the M2 macrophage-induced myofibroblast differentiation may contribute to fibrosis and this phenomenon needs further examination. Furthermore, differences between dermal and oral healing may not be attributed to the response of fibroblasts to M2 macrophages, although it would be of importance to investigate the presence of these M2 macrophages in vivo in dermal versus oral wounds. Acknowledgements We thank T.J. de Vries and M.E. Blaauboer from the Department of Oral Cell Biology (ACTA) for providing gingival fibroblasts for the experiments described in this article. In addition, we are grateful to M. Vlig and M.M. Ullrich (VSBN) for the kind gift of the dermal fibroblasts. References Aron-Wisnewsky, J., Tordjman, J., Poitou, C., Darakhshan, F., Hugol, D., Basdevant, A., Aissat, A., Guerre-Millo, M., Clement, K., 2009. Human adipose tissue macrophages: M1 and M2 cell surface markers in subcutaneous and omental depots and after weight loss. J. Clin. Endocrinol. Metab. 94, 4619–4623. de Vries, T.J., Schoenmaker, T., Wattanaroonwong, N., van den, H.M., Nieuwenhuijse, A., Beertsen, W., Everts, V., 2006. Gingival fibroblasts are better at inhibiting osteoclast formation than periodontal ligament fibroblasts. J. Cell Biochem. 98, 370–382. Desmouliere, A., Redard, M., Darby, I., Gabbiani, G., 1995. Apoptosis mediates the decrease in cellularity during the transition between granulation tissue and scar. Am. J. Pathol. 146, 56–66.
929
Eitner, F., Bucher, E., van, R.C., Kunter, U., Rong, S., Seikrit, C., Villa, L., Boor, P., Fredriksson, L., Backstrom, G., Eriksson, U., Ostman, A., Floege, J., Ostendorf, T., 2008. PDGF-C is a proinflammatory cytokine that mediates renal interstitial fibrosis. J. Am. Soc. Nephrol. 19, 281–289. Gilbertson, D.G., Duff, M.E., West, J.W., Kelly, J.D., Sheppard, P.O., Hofstrand, P.D., Gao, Z., Shoemaker, K., Bukowski, T.R., Moore, M., Feldhaus, A.L., Humes, J.M., Palmer, T.E., Hart, C.E., 2001. Platelet-derived growth factor C (PDGF-C), a novel growth factor that binds to PDGF alpha and beta receptor. J. Biol. Chem. 20, 27406–27414, 276. Goren, I., Allmann, N., Yogev, N., Schurmann, C., Linke, A., Holdener, M., Waisman, A., Pfeilschifter, J., Frank, S., 2009. A transgenic mouse model of inducible macrophage depletion: effects of diphtheria toxin-driven lysozyme M-specific cell lineage ablation on wound inflammatory, angiogenic, and contractive processes. Am. J. Pathol. 175, 132–147. Gratchev, A., Kzhyshkowska, J., Utikal, J., Goerdt, S., 2005. Interleukin-4 and dexamethasone counterregulate extracellular matrix remodelling and phagocytosis in type-2 macrophages. Scand. J. Immunol. 61, 10–17. Hinz, B., 2007. Formation and function of the myofibroblast during tissue repair. J. Invest Dermatol. 127, 526–537. Hinz, B., 2010. The myofibroblast: paradigm for a mechanically active cell. J. Biomech. 43, 146–155. Leask, A., 2010. Potential therapeutic targets for cardiac fibrosis: TGFbeta, angiotensin, endothelin, CCN2, and PDGF, partners in fibroblast activation. Circ. Res. 106, 1675–1680. Lolmede, K., Campana, L., Vezzoli, M., Bosurgi, L., Tonlorenzi, R., Clementi, E., Bianchi, M.E., Cossu, G., Manfredi, A.A., Brunelli, S., Rovere-Querini, P., 2009. Inflammatory and alternatively activated human macrophages attract vessel-associated stem cells, relying on separate HMGB1- and MMP-9-dependent pathways. J. Leukoc. Biol. 85, 779–787. Lygoe, K.A., Wall, I., Stephens, P., Lewis, M.P., 2007. Role of vitronectin and fibronectin receptors in oral mucosal and dermal myofibroblast differentiation. Biol. Cell. 99, 601–614. Mak, K., Manji, A., Gallant-Behm, C., Wiebe, C., Hart, D.A., Larjava, H., Hakkinen, L., 2009. Scarless healing of oral mucosa is characterized by faster resolution of inflammation and control of myofibroblast action compared to skin wounds in the red Duroc pig model. J. Dermatol. Sci. 56, 168–180. Mantovani, A., Sica, A., Sozzani, S., Allavena, P., Vecchi, A., Locati, M., 2004. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol. 25, 677–686. Mantovani, A., Sozzani, S., Locati, M., Allavena, P., Sica, A., 2002. Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol. 23, 549–555. Mirza, R., DiPietro, L.A., Koh, T.J., 2009. Selective and specific macrophage ablation is detrimental to wound healing in mice. Am. J. Pathol. 175, 2454–2462. Mosser, D.M., Edwards, J.P., 2008. Exploring the full spectrum of macrophage activation. Nat. Rev. Immunol. 8, 958–969. Nedelec, B., Shankowsky, H., Scott, P.G., Ghahary, A., Tredget, E.E., 2001. Myofibroblasts and apoptosis in human hypertrophic scars: the effect of interferon-alpha2b. Surgery 130, 798–808. Ostendorf, T., Eitner, F., Floege, J., 2012. The PDGF family in renal fibrosis. Pediatr. Nephrol. 27, 1041–1050. Pechkovsky, D.V., Prasse, A., Kollert, F., Engel, K.M., Dentler, J., Luttmann, W., Friedrich, K., Muller-Quernheim, J., Zissel, G., 2010. Alternatively activated alveolar macrophages in pulmonary fibrosis-mediator production and intracellular signal transduction. Clin. Immunol. 137, 89–101. Redd, M.J., Cooper, L., Wood, W., Stramer, B., Martin, P., 2004. Wound healing and inflammation: embryos reveal the way to perfect repair. Philos. Trans. R. Soc. Lond B Biol. Sci. 359, 777–784. Reinke, J.M., Sorg, H., 2012. Wound repair and regeneration. Eur. Surg. Res. 49, 35–43. Shannon, D.B., McKeown, S.T., Lundy, F.T., Irwin, C.R., 2006. Phenotypic differences between oral and skin fibro blasts in wound contraction and growth factor expression. Wound Repair Regen. 14, 172–178. Song, E., Ouyang, N., Horbelt, M., Antus, B., Wang, M., Exton, M.S., 2000. Influence of alternatively and classically activated macrophages on fibrogenic activities of human fibroblasts. Cell Immunol. 204, 19–28. van den Bogaerdt, A.J., van Zuijlen, P.P., van, G.M., Lamme, E.N., Middelkoop, E., 2002. The suitability of cells from different tissues for use in tissue-engineered skin substitutes. Arch. Dermatol. Res. 294, 135–142. van der Veer, W., Bloemen, M.C., Ulrich, M.M., Molema, G., van Zuijlen, P.P., Middelkoop, E., Niessen, F.B., 2009. Potential cellular and molecular causes of hypertrophic scar formation. Burns 35, 15–29. Varin, A., Gordon, S., 2009. Alternative activation of macrophages: immune function and cellular biology. Immunobiology 214, 630–641. Wilgus, T.A., 2007. Regenerative healing in fetal skin: a review of the literature. Ostomy Wound Manage. 53, 16–31. Wong, J.W., Gallant-Behm, C., Wiebe, C., Mak, K., Hart, D.A., Larjava, H., Hakkinen, L., 2009. Wound healing in oral mucosa results in reduced scar formation as compared with skin: evidence from the red Duroc pig model and humans. Wound Repair Regen. 17, 717–729.