Dr. Mayes has received speaking fees from Actelion and Practice Point Communications (less than $10,000 each), honoraria from Medtelligence for grant reviews (less than $10,000), and royalties from Oxford University Press for a book on scleroderma and from Harry Stewart Talks for a video on scleroderma.
Identification of Cadherin 11 as a Mediator of Dermal Fibrosis and Possible Role in Systemic Sclerosis
Version of Record online: 28 MAR 2014
Copyright © 2014 by the American College of Rheumatology
Arthritis & Rheumatology
Volume 66, Issue 4, pages 1010–1021, April 2014
How to Cite
Wu, M., Pedroza, M., Lafyatis, R., George, A.-T., Mayes, M. D., Assassi, S., Tan, F. K., Brenner, M. B. and Agarwal, S. K. (2014), Identification of Cadherin 11 as a Mediator of Dermal Fibrosis and Possible Role in Systemic Sclerosis. Arthritis & Rheumatology, 66: 1010–1021. doi: 10.1002/art.38275
- Issue online: 28 MAR 2014
- Version of Record online: 28 MAR 2014
- Accepted manuscript online: 18 NOV 2013 11:20AM EST
- Manuscript Accepted: 7 NOV 2013
- Manuscript Received: 16 JUL 2013
- NIH (National Institute of Arthritis and Musculoskeletal and Skin Diseases [NIAMS]). Grant Numbers: 5-T32-AR-052283-03, 1-P50-AR-060780, 2-R01-AR-051089, P50-AR-054144, K23-AR-061436, K08-AR-054404, R01-AR-062056-01A1
- Scleroderma Foundation New Investigator Award
- NIAMS. Grant Number: N01-AR-0-2251
Systemic sclerosis (SSc) is a chronic autoimmune disease clinically manifesting as progressive fibrosis of the skin and internal organs. Recent microarray studies demonstrated that cadherin 11 (Cad-11) expression is increased in the affected skin of patients with SSc. The purpose of this study was to examine our hypothesis that Cad-11 is a mediator of dermal fibrosis.
Biopsy samples of skin from SSc patients and healthy control subjects were used for real-time quantitative polymerase chain reaction analysis to assess Cad-11 expression and for immunohistochemistry to determine the expression pattern of Cad-11. To determine whether Cad-11 is a mediator of dermal fibrosis, Cad-11–deficient mice and anti–Cad-11 monoclonal antibodies (mAb) were used in the bleomycin-induced dermal fibrosis model. In vitro studies with dermal fibroblasts and bone marrow–derived macrophages were used to determine the mechanisms by which Cad-11 contributes to the development of tissue fibrosis.
Levels of messenger RNA for Cad-11 were increased in skin biopsy samples from patients with SSc and correlated with the modified Rodnan skin thickness scores. Cad-11 expression was localized to dermal fibroblasts and macrophages in SSc skin. Cad-11–knockout mice injected with bleomycin had markedly attenuated dermal fibrosis, as quantified by measurements of skin thickness, collagen levels, myofibroblast accumulation, and profibrotic gene expression, in lesional skin as compared to the skin of wild-type mice. In addition, anti–Cad-11 mAb decreased fibrosis at various time points in the bleomycin-induced dermal fibrosis model. In vitro studies demonstrated that Cad-11 regulated the production of transforming growth factor β (TGFβ) by macrophages and the migration of fibroblasts.
These data demonstrate that Cad-11 is a mediator of dermal fibrosis and TGFβ production and suggest that Cad-11 may be a therapeutic target in SSc.
Scleroderma (systemic sclerosis [SSc]) is a multisystem autoimmune disease that is clinically characterized by progressive fibrosis of the skin and internal organs (). The pathogenesis of SSc is complex, involving 3 interrelated processes: inflammation and autoimmunity, vasculopathy, and excessive extracellular matrix (ECM) deposition (). At the cellular level, dendritic cells, T cells, and macrophages contribute to the inflammatory response, ultimately leading to activation of fibroblasts and myofibroblasts (). Transforming growth factor β (TGFβ) is a critical cytokine that stimulates fibroblasts and myofibroblasts to produce large quantities of ECM, leading to tissue fibrosis. At the molecular level, multiple pathways have been implicated in the development of tissue fibrosis, including (but not limited to) TGFβ (), type I interferon ([4-8]), Wnt/β-catenin ([9-12]), and more recently, cadherins ().
The cadherins are a family of integral transmembrane proteins that mediate homophilic cell-to-cell adhesion via dimerization between ectodomains of identical cadherins on adjacent cells ([14-16]). The cytoplasmic tail of cadherins contains binding sites for multiple intracellular signaling molecules, such as β-catenin. Cadherins play a role in regulating cellular behavior beyond adhesion, including cellular migration and invasion ([17-20]). Cadherins also regulate cell phenotype through epithelial-to-mesenchymal transition, a process that has been implicated in pathologic processes such as cancer and fibrosis ([13, 20-23]).
Cadherin 11 (Cad-11) is a type II classic cadherin initially identified in mouse osteoblast cell lines (). Cad-11 expression has subsequently been reported in other cells, including synovial fibroblasts, where it imparts a mesenchymal phenotype and promotes cellular invasion ([25-28]). Multiple studies have implicated a role of Cad-11 in malignant transformation ([19, 20, 26]). Furthermore, a role of Cad-11 in rheumatoid arthritis has been proposed (). Interestingly, 2 independent microarray studies have noted increased Cad-11 transcripts in the affected skin of SSc patients relative to that in the skin of healthy controls, implicating a role of Cad-11 in another autoimmune disease ([29, 30]).
In the current study, we hypothesized that Cad-11 plays a role in the development of dermal fibrosis. We demonstrated that Cad-11 levels are increased in skin biopsy samples from SSc patients and that Cad-11 modulates the development of dermal fibrosis in the bleomycin-induced mouse model of dermal fibrosis. These studies suggest that Cad-11 may be a therapeutic target for tissue fibrosis in SSc patients.
MATERIALS AND METHODS
Wild-type (WT) mice and Cad-11–knockout (KO) mice (both groups C129/C57BL/6 backcross) were maintained as homozygous inbred lines at the University of Texas Health Science Center (UTHSC) at Houston and Baylor College of Medicine (BCM) ([28, 31]). WT C57BL/6 mice were purchased from Taconic Farms. All studies were conducted with the approval of the Institutional Animal Care and Use Committees of the UTHSC at Houston and BCM.
SSc patients and healthy controls
Skin biopsy samples were obtained from SSc patients and unrelated healthy control subjects at UTHSC at Houston or Boston University Medical Center (BUMC). All SSc patients met the American College of Rheumatology (ACR) classification criteria for SSc (), and all had diffuse SSc. Disease duration was <5 years in patients from UTHSC and 4 years in those from BUMC. The healthy control subjects had no personal or family medical history of autoimmune diseases. Biopsy samples were obtained from the affected skin of the forearms of SSc patients and from a similar location on the forearms of healthy control subjects.
All subjects provided written informed consent, and the study was approved by the institutional review boards at UTHSC and BUMC.
Bleomycin-induced dermal fibrosis model
Mice (6–8 weeks old) were administered bleomycin (0.02 units/day; Teva Parenteral Medicines) dissolved in phosphate buffered saline (PBS) or PBS alone by subcutaneous injection. After 7 or 28 days of injections, mice were killed, and lesional skin for analysis was harvested with an 8-mm skin biopsy punch ().
Five-micrometer–thick sections of formalin-fixed paraffin-embedded skin biopsy samples were stained with hematoxylin and eosin or Masson's trichrome. To evaluate the level of skin fibrosis in mice treated with bleomycin, the thickness of the dermis, defined as the distance between the epidermal–dermal junction and the dermal–adipose layer junction, was measured in each animal at 6 randomly selected sites per microscopic field (). To analyze the accumulated collagen content in the lesional skin, sections were stained with Masson's trichrome ().
Immunohistochemistry (IHC) was performed on formalin-fixed paraffin-embedded skin biopsy sections using antibodies against Cad-11 (Invitrogen), α-smooth muscle actin (α-SMA; Sigma-Aldrich), Mac-3 (BD PharMingen), or β-catenin (BD PharMingen) (). Substitution of the primary antibody with species-specific isotype IgG served as negative controls. Bound antibodies were detected with secondary antibodies from a Histomouse kit (Invitrogen) or an EnVision Plus system–horseradish peroxidase kit (Dako). After counterstaining with hematoxylin, sections were mounted with Permount (Fisher Scientific) and viewed with an Olympus BX60 microscope. Fibroblasts were identified by their spindle-shaped morphology and inflammatory cells by their round morphology ([3, 32]).
For immunofluorescence, 10-μm frozen sections of lesional skin were prepared and fixed in ice-cold methanol for 15 minutes. Primary antibodies against Cad-11, α-SMA, platelet endothelial cell adhesion molecule 1 (Santa Cruz Biotechnology), S100A4 (Dako), or Mac-3 were incubated for 1 hour at room temperature. Fluorescein isothiocyanate–conjugated anti-goat and phycoerythrin-conjugated anti-rabbit secondary antibodies (Jackson ImmunoResearch) were used, followed by DAPI nucleolus stain. To quantify Cad-11 in lesional dermis, cadherin 11–positive fibroblasts and inflammatory cells were counted, and the ratio of positive cells to total cells was calculated.
Determination of messenger RNA (mRNA) levels by real-time quantitative polymerase chain reaction (qPCR).
Tissue levels of mRNA were determined by real-time qPCR. Total RNA was isolated from skin biopsy samples that had been frozen in RNAlater (Qiagen) using TRIzol reagent (Invitrogen), purified with an RNA Mini kit (Qiagen), and reverse transcribed. Real-time qPCR was performed using validated TaqMan gene expression assays for Cad-11, COL1A1, TGFβ1, interleukin-6 (IL-6), CCL2, connective tissue growth factor (CTGF), and α-SMA (Applied Biosystems) on an Applied Biosystems 7900HT Fast Real-Time PCR system. The 18S RNA or cyclophilin A (PPIA) genes were used as endogenous controls to normalize transcript levels of mRNA in each sample. Data were analyzed with SDS version 2.3 software using the comparative Ct ( ) method.
Quantification of collagen in lesional skin
The collagen content of lesional skin was determined with a Sircol Collagen Assay kit (Biocolor) as described previously (). Total protein assay (Bio-Rad) was used as a control to normalize the collagen content of each sample.
In vitro dermal fibroblast and macrophage cultures
Dermal fibroblasts were explanted from the dorsal skin of 6–8-week-old Cad-11–KO and WT mice, and cells from passage 3–5 were studied in parallel (). Cell migration was determined in 24-well modified Boyden Transwells with 8.0-μm–pore filters (BioCoat; BD Biosciences). Briefly, 2.5 × 104 primary skin fibroblasts from WT and Cad-11–KO mice were placed in Dulbecco's modified Eagle's medium (DMEM) with 0.1% bovine serum albumin (BSA) and added to the upper chamber. Medium with 5% fetal bovine serum (FBS) was added to the lower chamber. After incubation for 6 hours at 37°C, nonmigrating cells were removed from the upper chamber, and the filters were fixed with 100% methanol followed by staining with Diff-Quick stain. Migrating cells were quantitated by counting the number of cells on the underside of the filters.
To determine ERK and Smad2 phosphorylation, dermal fibroblasts were cultured overnight in DMEM/FBS, followed by overnight incubation in DMEM with 0.1% BSA. Cultures were stimulated for 3 hours with 10 ng/ml of TGFβ1 (R&D Systems). Protein lysates were used for Western blotting with antibodies against Smad2, p-Smad2, ERK, and p-ERK (Cell Signaling Technology).
Bone marrow–derived macrophages were obtained from the femurs of WT and Cad-11–KO mice. Cells were isolated, washed, and cultured for 7 days at 37°C in DMEM with 20% FBS supplemented with medium from L929 cells (atmosphere of 5% CO2). For in vitro stimulation, cells were reseeded and cultured for 72 hours with recombinant mouse IL-4 (10 ng/ml; R&D Systems). Supernatants were assessed for TGFβ levels by enzyme-linked immunosorbent assay (R&D Systems).
Results are expressed as the mean ± SD or as the mean ± SEM. The Mann-Whitney U test was used to compare 2 groups of mice in the bleomycin studies. Student's t-test was used for comparison of 2 groups for the in vitro studies. Correlations between Cad-11 and other variables were determined using Spearman's correlation. P values less than 0.05 were considered significant.
Cadherin 11 expression in SSc skin biopsy samples
Microarray studies have demonstrated an increase in Cad-11 mRNA in SSc skin biopsy tissues ([29, 30]). Relative real-time qPCR was used to determine whether Cad-11 mRNA is increased in skin biopsy samples obtained from SSc patients and healthy controls. Compared to control skin samples (n = 9), SSc skin samples (n = 6) had elevated levels of Cad-11 mRNA (Figure 1A), confirming the published findings of microarray studies ([29, 30]). COL1A1 and CTGF expression was also increased in SSc skin biopsy samples (data not shown).
Using another independent set of skin biopsy samples from patients with diffuse SSc whose disease duration was <4 years, the expression of Cad-11 was determined using microarray expression profiling and was compared to the modified Rodnan skin thickness scores (MRSS). Cad-11 expression levels correlated positively with the extent of skin involvement (Spearman's r = 0.6301, P = 0.0006) (Figure 1B). These data confirm that Cad-11 expression is increased in SSc skin and demonstrate that the expression levels correlate at the cross-sectional level with the extent of dermal fibrosis in SSc patients with disease duration of <4 years. Additional prospective studies are needed to determine if Cad-11 expression changes in the skin over time as the skin worsens or improves.
To determine the cellular expression pattern of Cad-11 in SSc skin, IHC was performed. No immunoreactivity was observed with the isotype control (data not shown). Skin biopsy samples from healthy control subjects had low levels of Cad-11 expression (Figures 1C and D). In contrast, increased Cad-11 reactivity was observed on fibroblasts and inflammatory cells located mostly in the reticular dermis, with occasional cells in the papillary dermis, of SSc biopsy samples (Figures 1E–G). The number of fibroblasts with Cad-11 reactivity was quantified in control samples (n = 4) and SSc samples from patients with early diffuse disease (n = 9). SSc samples had increased numbers of Cad-11–positive fibroblasts (mean ± SD 9.5 ± 1.0/hpf) relative to control samples (5.6 ± 0.6/hpf; P < 0.05) (Figure 1H). Similarly, SSc biopsy samples had increased numbers of inflammatory cells (15.4 ± 1.0/hpf) relative to control biopsy samples (10.3 ± 1.3/hpf; P < 0.05) (Figure 1I). These data demonstrate that Cad-11 expression is increased in skin biopsy samples from SSc patients, where it localizes to fibroblasts and inflammatory cells.
Cad-11 expression in the bleomycin-induced dermal fibrosis model
The bleomycin-induced dermal fibrosis model is well characterized and commonly used to study biologic pathways that are shared between the model and SSc in humans ([35-38]). To determine whether Cad-11 expression is increased in the fibrotic skin of this model, C57BL/6 mice received daily subcutaneous injections of bleomycin, and lesional skin was harvested for analysis. Similar to SSc skin biopsy samples, murine skin injected with bleomycin for 7 days had increased levels of Cad-11 mRNA relative to that in PBS-injected skin, as determined by real-time qPCR (data available upon request from the corresponding author). IHC demonstrated strong expression of Cad-11 on spindle-shaped fibroblast-like cells and on round inflammatory cells in the reticular dermis, with occasional cells in the papillary dermis, of skin biopsy samples from mice injected with bleomycin (day 7 and day 28) (data available upon request from the corresponding author). In contrast, PBS-injected control mice expressed very low levels of Cad-11 in the basal layer of the epidermis, hair follicles, round inflammatory cells, and fibroblasts.
Dual-color immunofluorescence was used to characterize the expression pattern of Cad-11 in the bleomycin-induced dermal fibrosis model (data available upon request from the corresponding author). Dual staining for Cad-11 and α-SMA, a marker of myofibroblasts, confirmed the expression of Cad-11 on myofibroblasts in fibrotic tissue. Furthermore, simultaneous staining for Cad-11 and Mac-3, a macrophage marker, confirmed the expression Cad-11 on macrophages in the fibrotic tissue. Therefore, similar to the findings in the skin of patients with SSc, Cad-11 expression is present on myofibroblasts and macrophages in the fibrotic skin of mice with bleomycin-induced dermal fibrosis.
Cad-11 as a mediator of dermal fibrosis
To investigate the role of Cad-11 in the development of dermal fibrosis, the subcutaneous bleomycin-induced dermal fibrosis model was used in WT mice and Cad-11–KO mice. Lesional skin was analyzed on day 28. Histologic analyses of lesional skin stained with hematoxylin and eosin (Figure 2A) or Masson's trichrome (Figure 2B) demonstrated that subcutaneous injections of bleomycin increased dermal thickness, with dense ECM deposition in the dermis and fibrous replacement of the subcutaneous adipose layer, in WT mice relative to the results following injections of PBS (Figure 2A). In contrast, Cad-11–KO mice showed only a modest increase in dermal thickness, with relative preservation of the subcutaneous adipose layer and less dense ECM deposition. These findings are quantified in Figure 2C and demonstrate that Cad-11–KO mice injected with bleomycin had a statistically significant reduction in dermal thickness relative to WT mice injected with bleomycin.
The soluble collagen content was assessed using a Sircol assay (Figure 2D). As expected, compared to PBS injections, bleomycin injections increased the total collagen content of skin biopsy samples from WT mice. Consistent with the histologic analyses, the total collagen content in lesional skin from Cad-11–KO mice injected with bleomycin was markedly reduced relative to the levels in WT mice. These data strongly support a role of Cad-11 in the development of dermal fibrosis induced by subcutaneously administered bleomycin.
Reduced dermal fibrosis with anti–Cad-11 antibody treatment
The reduction of fibrosis in Cad-11–KO mice suggests that Cad-11 is a potential therapeutic target for dermal fibrosis. To further examine this, anti–Cad-11 monoclonal antibodies were administered intraperitoneally to C57BL/6 WT mice starting on day 14 of the 28-day subcutaneous bleomycin model protocol. Two independent clones of anti–Cad-11 monoclonal antibodies (13C2 and 23C6) were used and were administered every other day according to their ability to decrease inflammatory arthritis in mice (). Histologic analyses of lesional skin stained with hematoxylin and eosin (data available upon request from the corresponding author) and quantitative assessment of dermal thickness (Figure 3A) demonstrated that subcutaneous bleomycin injections increased dermal thickness in WT mice relative to the findings with PBS injections. Treatment with intraperitoneal PBS or isotype control antibody did not alter the amount of dermal fibrosis or the dermal thickness. In contrast, 2 weeks of treatment with either 13C2 or 23C6 resulted in a statistically significant reduction in dermal fibrosis induced by bleomycin treatment relative to treatment with isotype control. These data were further confirmed using a Sircol assay to assess total collagen deposition in lesional skin (Figure 3B). Collagen accumulation in bleomycin-induced lesional skin was significantly reduced in WT mice treated with either of the anti–Cad-11 monoclonal antibodies (13C2 or 23C6) relative to that in skin from mice treated with isotype control.
To determine whether anti–Cad-11 monoclonal antibodies could hasten the resolution of dermal fibrosis in the bleomycin-induced dermal fibrosis model, WT C57BL/6 mice were given subcutaneous injections of PBS or bleomycin daily for 28 days. Mice were subsequently treated with intraperitoneal anti–Cad-11 monoclonal antibody (clone 23C6) or isotype control from day 28 to day 42 (Figure 3C). On day 42, mice administered bleomycin had a slight, but not statistically significant, reduction in dermal thickness relative to that in a group of bleomycin-treated mice euthanized on day 28. Interestingly, after 2 weeks of treatment, anti–Cad-11 monoclonal antibody resulted in a more substantial reduction of dermal thickness than did isotype control antibody treatment in mice euthanized on day 42 and bleomycin treatment in mice euthanized on day 28. Therefore, neutralization of Cad-11 with monoclonal antibodies effectively treated established dermal fibrosis in the bleomycin-induced dermal fibrosis model. These data further confirm a role of Cad-11 in the development of dermal fibrosis and suggest that Cad-11 may be a potential therapeutic target for established dermal fibrosis.
Modulation of fibrotic mediators in Cad-11–deficient mice
Injections of bleomycin induce an early and transient inflammatory response in the dermis, consisting of T cells, mast cells, and macrophages, that leads to increases in levels of inflammatory cytokines and TGFβ and in the numbers of ECM-secreting fibroblasts and myofibroblasts ([32, 37, 38]). To determine whether Cad-11 modulates the early phases of the development of dermal fibrosis, we used real-time qPCR to measure the levels of mRNA for IL-6 and CCL2, two important inflammatory cytokines, in the lesional skin of mice after 7 days of daily bleomycin injections. Bleomycin induced a similar increase in IL-6 and CCL2 mRNA expression in both WT and Cad-11–KO mice (Figures 4A and B). Consistent with this observation, lesional skin from both WT and Cad-11–KO mice demonstrated histologic evidence of an inflammatory response to bleomycin (data available upon request from the corresponding author). After 7 days of bleomycin injections, no differences in the total number of inflammatory cells were observed (data not shown).
To determine whether Cad-11 modulates fibrotic gene expression during the development of dermal fibrosis, real-time qPCR was used to measure mRNA levels of Col1a2, CTGF, and α-SMA in the lesional skin of mice after 7 days of daily bleomycin injections. Interestingly, lesional skin from bleomycin-treated Cad-11–KO mice had significantly less fibrotic gene expression, including Col1a2, CTGF, and α-SMA mRNA, than did lesional skin from bleomycin-treated WT mice (Figures 4C–E). Taken together, these findings demonstrate that bleomycin-induced expression of profibrotic genes, but not IL-6 or CCL2 genes, is reduced in Cad-11–KO mice as compared to WT mice.
Cad-11 and the TGFβ pathway in dermal fibrosis
TGFβ is a potent activator of fibroblasts, inducer of myofibroblast differentiation, and driver of the expression of fibrotic genes, leading to the development of tissue fibrosis. Therefore, we next wanted to determine the extent to which Cad-11 modulates the TGFβ pathway. Cartilage oligomeric matrix protein (COMP) and thrombospondin 1 (TSP-1) are induced by TGFβ and have been shown to correlate with the MRSS in SSc patients (). Using lesional skin biopsy samples from patients with early diffuse SSc, we observed that Cad-11 expression is correlated with COMP expression (Spearman's r = 0.9118, P < 0.0001) (Figure 5A) and with TSP-1 expression (Spearman's r = 0.9186, P < 0.0001) (Figure 5B). These data support a potential interaction of Cad-11 and the TGFβ pathway in SSc patients.
Expression of TGFβ mRNA was determined in lesional skin from WT and Cad-11–KO mice injected with bleomycin. TGFβ mRNA expression was increased in WT mice treated with bleomycin. However, lesional skin from Cad-11–KO mice treated with bleomycin had significantly reduced levels of TGFβ mRNA compared to that in bleomycin-treated WT mice (Figure 5C). Macrophages are one potential source of TGFβ during the development of fibrosis ([37, 40]). Therefore, bone marrow–derived macrophages from WT and Cad-11–KO mice were cultured with IL-4 to determine whether Cad-11 modulates TGFβ production. As seen in Figure 5D, bone marrow–derived macrophages from Cad-11–KO mice produced significantly less TGFβ than did those from WT mice. Therefore, Cad-11 modulates TGFβ production by bone marrow–derived macrophages, and these data suggest that Cad-11 may be a regulator of macrophage TGFβ production in vivo.
Active TGFβ induces both Smad-dependent and Smad-independent intracellular signaling (i.e., ERK) ([41, 42]). To determine whether Cad-11 modulates TGFβ signaling, IHC for Smad2 and ERK were performed on lesional skin from WT and Cad-11–KO mice. Consistent with a reduction in TGFβ levels, the number of fibroblasts and inflammatory cells expressing the phosphorylated form of Smad2 was lower in lesional skin from bleomycin-treated Cad-11–KO mice as compared to that from bleomycin-treated WT mice (Figures 5E and F). Similarly, the numbers of fibroblasts and inflammatory cells expressing the phosphorylated form of ERK were lower in lesional skin from Cad-11–KO mice compared to that from WT mice (Figures 5G and H). These data demonstrate that in the bleomycin-induced dermal fibrosis model, in vivo levels of TGFβ signaling are reduced in Cad-11–KO mice.
Changes in in vivo phosphorylated forms of Smad2 and ERK might reflect direct alterations in TGFβ signaling or might be secondary to the lower levels of TGFβ found in lesional skin of Cad-11–KO mice. To determine if Cad-11 directly modulates the responsiveness of the dermal fibroblast to TGFβ, dermal fibroblasts from WT and Cad-11–KO mice were stimulated in vitro with TGFβ. As shown in Figure 5I, lysates of TGFβ-stimulated dermal fibroblasts from WT and Cad-11–KO mice expressed similar levels of phosphorylated Smad2 and ERK. Furthermore, TGFβ-stimulated dermal fibroblasts from WT and Cad-11–KO mice produced similar amounts of CTGF and α-SMA, both of which are TGFβ-stimulated genes (data not shown). Taken together, the in vivo and in vitro data suggest that one mechanism by which Cad-11 regulates the development of tissue fibrosis is through the production of TGFβ and not directly through TGFβ signaling in the dermal fibroblast.
Cad-11 regulation of dermal fibroblasts
Although no differences in TGFβ signaling were observed in dermal fibroblasts from Cad-11–KO mice, given the expression of Cad-11 on dermal fibroblasts, it is reasonable to hypothesize that Cad-11 directly modulates fibroblast behavior through other pathways relevant to fibrosis. Recent data have supported a role of the Wnt/β-catenin pathway in the development of fibrosis ([9, 10, 12, 43]). Given the interaction of β-catenin with cadherins at the cytoplasmic tail (), we hypothesized that Cad-11 may alter the expression of β-catenin in dermal fibroblasts. Cell lysates from dermal fibroblasts isolated from WT and Cad-11–KO mice were immunoblotted for total β-catenin levels. As shown in Figure 6A, the relative expression of β-catenin was lower in dermal fibroblasts from Cad-11–KO mice as compared to WT mice. To further investigate the expression of β-catenin, lesional skin from WT and Cad-11–KO mice that had been injected with bleomycin was assessed by IHC for β-catenin. We found that lesional skin from Cad-11–KO mice had fewer β-catenin–expressing fibroblasts than did skin from WT mice (Figure 6B).
Cadherins regulate cell migration and invasion of malignant cells (). The ability of fibroblasts to migrate and invade the tissue is important in the development of tissue fibrosis. Therefore, the migratory capacity of dermal fibroblasts from WT and Cad-11–KO mice was compared in Transwell migration assays in response to 5% FBS. Dermal fibroblasts from WT mice efficiently migrated through the Transwells in response to 5% FBS (Figure 6C). In contrast, Cad-11–KO dermal fibroblasts showed decreased migratory activity (P < 0.05). These data support a role of Cad-11 in the direct regulation of dermal fibroblast behavior.
In the current study, we found that Cad-11 levels are elevated in SSc patients and correlate with the extent of skin involvement. Studies in Cad-11–deficient mice and studies using anti–Cad-11 monoclonal antibodies clearly demonstrated a role of Cad-11 in the development of dermal fibrosis in the bleomycin-induced dermal fibrosis model. Mechanistic studies demonstrated that Cad-11 modulates TGFβ levels in vivo and the production of TGFβ by macrophages. However, Cad-11 did not alter the in vitro TGFβ signaling in dermal fibroblasts, suggesting that one mechanism by which Cad-11 regulates the development of fibrosis is through TGFβ production. Moreover, Cad-11–deficient mice had lower levels of β-catenin and decreased migratory capacity of dermal fibroblasts relative to WT mice.
The development of tissue fibrosis in patients with SSc involves the complex interaction of multiple molecular pathways in multiple cell types. Early events include endothelial cell dysfunction and activation and recruitment of inflammatory cells, such as macrophages and CD4+ T cells, to the skin (). These events result in an increase in inflammatory cytokines, such as CCL2, IL-6, and type I interferons, as well as profibrotic cytokines, such as IL-13 and TGFβ, leading to tissue damage and activation of wound healing pathways ([2, 4, 44-46]). Under the influence of TGFβ and other profibrotic mediators, fibroblasts and myofibroblasts accumulate in the fibrotic tissue, leading to increased deposition of ECM components and tissue fibrosis.
Given this highly complex network of cells and pathways, targeting an individual molecule that is expressed on a single cell type may prove to be a difficult strategy by which to block this process. Cad-11 expression in fibrotic skin localizes to the dermal macrophage and fibroblast populations, two cells involved in different stages of the fibrotic process. These findings are similar to the pattern of Cad-11 expression in idiopathic pulmonary fibrosis, where Cad-11 also regulates the development of tissue fibrosis (). However, given the higher fold change in Cad-11 mRNA in SSc skin relative to the increased number of fibroblasts and inflammatory cells expressing Cad-11, it remains possible that other populations of cells express Cad-11 at levels beneath the sensitivity of IHC. Since Cad-11 is expressed on both macrophages and fibroblasts, it is likely that Cad-11 regulates multiple steps in the development of fibrosis. Accordingly, in the current study, we provide evidence that Cad-11 regulates the production of TGFβ by macrophages and the behavior of fibroblasts, such as migration.
Dermal macrophages are one of the major sources of TGFβ in the skin of patients with SSc (). TGFβ is considered to be the master regulator of fibrosis. TGFβ activates intracellular signaling pathways, including Smad and ERK, which leads to the accumulation, differentiation, and activation of myofibroblasts as well as ECM deposition. Based on the current data, when challenged in the subcutaneous bleomycin-induced model of fibrosis, Cad-11–deficient mice had lower levels of TGFβ in lesional skin. Furthermore, bone marrow–derived macrophages from Cad-11–deficient mice produced less TGFβ. These data are consistent with prior observations that Cad-11 also regulates TGFβ production by alveolar macrophages (). These data demonstrate that Cad-11 can regulate the production of TGFβ by multiple macrophage populations. Preliminary studies do not suggest that Cad-11 regulates M2-type macrophage differentiation, and the specific mechanism by which Cad-11 regulates TGFβ production is unknown. However, the current data suggest that one mechanism by which Cad-11 regulates the development of fibrosis is through the regulation of TGFβ production by macrophages.
Fibroblasts, in particular myofibroblasts, are critical downstream effector cells in the fibrotic process. TGFβ stimulates the activation and phosphorylation of the Smad2 and ERK pathways in fibroblasts and the development of fibrosis. Cad-11–deficient lesional skin contained decreased numbers of fibroblasts that were positive for the phosphorylated forms of Smad2 and ERK; however, in vitro, TGFβ induced a similar level of TGFβ signaling. These data suggest that Cad-11 regulates the TGFβ pathway through TGFβ levels and not through the direct ability of the dermal fibroblast to respond to TGFβ. However, Cad-11 may regulate the fibroblasts through other pathways. Recent data have supported a role of the Wnt/β-catenin pathway in the development of fibrosis ([9, 10, 12, 43]). Interestingly, dermal fibroblasts from Cad-11–KO mice have significantly lower levels of β-catenin relative to those from WT mice. It has been shown that activation of Wnt/β-catenin can regulate cell migration in lung fibroblasts (). Consistent with this association, in the current study, we noted that, similar to synovial fibroblasts (), dermal fibroblasts from Cad-11–KO mice showed decreased migratory activity. We therefore propose that Cad-11 is capable of modulating fibroblast behavior indirectly, through the modulation of TGFβ levels, and directly, as noted in the migration assays.
It remains possible that Cad-11 contributes to the development of tissue fibrosis through other pathways. Although we did not observe differences in IL-6 in dermal fibroblasts from Cad-11–KO and WT mice nor in lesional skin from mice with bleomycin-induced skin fibrosis, Cad-11 engagement has been reported to up-regulate IL-6 in human synovial fibroblasts (). This may be due to differences in fibroblast populations. Alternatively, it remains possible that Cad-11 regulates other inflammatory cytokines produced by dermal fibroblasts or dermal macrophages. This is currently under investigation.
In summary, we have identified Cad-11 as a mediator of dermal fibrosis and have shown that antibodies against Cad-11 can prevent the development of fibrosis in the mouse model of bleomycin-induced fibrosis. Furthermore, we showed that Cad-11 antagonism could hasten the resolution of existing fibrosis. These studies provide important preclinical data suggesting that Cad-11 may be a therapeutic target in systemic sclerosis.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Agarwal had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Wu, Pedroza, George, Mayes, Assassi, Tan, Brenner, Agarwal.
Acquisition of data. Wu, Pedroza, Lafyatis, George, Mayes, Assassi, Tan, Agarwal.
Analysis and interpretation of data. Wu, Pedroza, George, Mayes, Assassi, Tan, Agarwal.
- 7A macrophage marker, Siglec-1, is increased on circulating monocytes in patients with systemic sclerosis and induced by type I interferons and Toll-like receptor agonists.Arthritis Rheum2007;56:1010–20., , , , , .