Objective To determine the frequency and suppressive capacity of regulatory T cells (Treg) and their association with clinical parameters in patients with systemic scleroderma (SSc).
Methods Peripheral blood from 25 patients with SSc, 15 patients with localised scleroderma (LS) and 29 healthy controls (HC) was studied. Analysis of CD4+ forkhead box P3 (Foxp3)+ and CD4+CD25++Foxp3+ Treg subpopulations was carried out by flow cytometry and cell proliferation was quantified by 3H-thymidine incorporation. Quantitative analysis of Treg was further performed in skin biopsies from 17 patients with SSc and 21 patients with LS using anti-CD4 and anti-Foxp3 monoclonal antibodies for immunohistochemistry.
Results The frequency of CD4+Foxp3+ and CD4+CD25++Foxp3+ Treg in peripheral blood from patients with SSc was not significantly different from that of patients with LS or HC. The suppressive capacity of CD4+CD25++ Treg in SSc was also found to be similar to that of HC. Phenotypic and functional data revealed no significant difference between the limited or diffuse form of SSc. Moreover, therapy with bosentan showed no significant effect on the frequency of Treg during the course of the disease. However, the frequency of Treg in skin lesions from patients with SSc or LS, determined as the percentage of CD4+ cells expressing Foxp3 in the inflammatory infiltrate, was significantly reduced compared with other inflammatory skin diseases.
Conclusion These results indicate that although the authors found no defect in the frequency or function of peripheral Treg subpopulations, the reduction of CD4+Foxp3+ Treg in the skin of patients with SSc may be important in the pathogenesis of the disease.
Statistics from Altmetric.com
Systemic scleroderma (SSc) is a heterogeneous autoimmune disease in which the extent of cutaneous and systemic organ involvement is highly variable. In recent years, a subclassification of SSc that divides the disease into various subsets based on a number of clinical characteristics has been widely accepted.1 2 Diffuse SSc is the most severe form, with a rapid onset and widespread skin hardening. It generally causes internal organ damage, specifically in the lungs and gastrointestinal tract, but the disease may also affect the skeletal muscles, kidneys and heart. Limited SSc, the most common form (45.5% of all cases of SSc), is milder and has a slower onset and progression. Moreover, skin hardening is usually confined to the hands and the face, and internal organ involvement is less severe. Other subsets of SSc include overlap syndrome and SSc sine scleroderma.
The therapeutic options currently available for the management for patients with SSc are still limited. Calcium antagonists and prostanoids are recommended for SSc-related digital vasculopathy attacks (Raynaud phenomenon, digital ulcers), while immunosuppressive treatment with corticosteroids, methotrexate, azathioprine, cyclosporine or cyclophosphamide may slow progression of the disease.3 4 In addition, there have been some promising developments over the past years with identification of novel candidate targets and innovative strategies, including targeted immunomodulatory therapies, tyrosine kinase inhibitors and agents that may promote vascular repair.5 However, the involvement of specific organs, such as pulmonary hypertension (PAH) due to lung fibrosis, requires symptomatic therapy. Bosentan, which is a competitive receptor antagonist of endothelin-1 (ET-1), is approved for the treatment of PAH.6,–,8 Recently, bosentan has also been shown to be effective in Raynaud's phenomenon and in the prevention of digital ulcers and skin fibrosis in SSc patients.9,–,11 Bosentan binds to both ET receptors (ETA and ETB) with high affinity. Importantly, the expression of these receptors in systemic rheumatic diseases is associated with fibrotic and inflammatory activity and correlates with the reduction of vasodilatative factors.12
Although the cause of SSc remains largely unknown, various factors, such as genetic susceptibility, microvascular injury, immune dysfunction and deposition of extracellular matrix proteins are thought to be involved in its pathogenesis.13 Further evidence suggests that various cytokines can modulate the synthesis of extracellular matrix by fibroblasts. However, it is unclear how autoimmunity and tissue fibrosis are related in SSc. A better understanding of the pathogenesis underlying SSc may help to design more effective therapeutic strategies for SSc in the future. Recently, CD4+CD25+ regulatory T cells (Treg) have been demonstrated to be important in a variety of murine autoimmune disease models.14 Functional analysis of murine Treg has shown that these cells fail to proliferate or secrete cytokines in response to polyclonal or antigen-specific stimulation but inhibit the activation of conventional CD4+CD25- T cells (Tcon).15 Moreover, a lack of Treg or a defect in the suppressive capacity is found in human autoimmune diseases, such as type 1 diabetes,16 17 multiple sclerosis18 and systemic lupus erythematosus (SLE).19 20 In the present study, we investigated the frequency of Treg subpopulations as well as the suppressive capacity of CD4+CD25++ Treg in the peripheral blood of patients with SSc and correlated the results with clinical parameters. Moreover, we evaluated the effect of bosentan therapy on the frequency of peripheral Treg during the course of the disease and defined the frequency of CD4+ forkhead box P3 (Foxp3)+ Treg in skin lesions of patients with different forms of scleroderma.
Material and methods
Blood samples were obtained from 20 patients with SSc (five male and 15 female, age 52.4±10.4 years), 14 patients with limited SSc (three male and 11 female, age 55.1±10.7 years) and six patients with diffuse SSc (two male and four female, age 46.0±7.2 years) (Table 1). This study also included five patients with SSc (three male and two female, age 58.2±10.6 years) who were treated with the ET-1 receptor antagonist bosentan. Blood was taken form these patients prior to the onset of therapy and again after 3 months of treatment. According to the product information, patients were treated with 62.5 mg bosentan orally twice daily for 4 weeks followed by 125 mg twice daily. For control purposes, blood samples from 15 patients with localised scleroderma (LS) without extracutaneous involvement (two male and 13 female, age 50.5±18.2 years) and 29 healthy controls (HC) (nine male and 20 female, age 42.2±13.9 years) were also obtained.
Immunohistochemistry was performed in skin biopsy specimens from patients other than those included in the flow cytometry analysis. Lesional skin biopsies from 12 patients with SSc (three male and nine female, age 57.3±12.4 years (mean±SD)) and 21 patients with LS (four male and 17 female, age 50.7±18.8 years) were stained for immunohistochemical analysis. In addition, we investigated non-lesional skin biopsies from five patients with SSc (one male and four female, age 46.0±12.3 years) (table 2). For control purposes, results were compared with data from a previously published study21 performed under the same conditions, including lesional skin biopsy specimens from patients with psoriasis (PSO; five male and five female, age 48.4±16.7 years), lichen planus (LP; four male and six female, age 52.6±14.0 years) and atopic dermatitis (AD; four male and six female, age 59.9±18.6 years). Further details from the methodology are included in the supplementary material (supplementary 1).
Frequency of CD4+Foxp3+ and CD4+CD25++Foxp3+ Treg in the peripheral blood of patients with SSc compared with LS patients and HC
The median of percentage of CD4+Foxp3+ Treg was found to be 7.0% in SSc patients (range 3.7–13.7%) and 7.3% in LS patients (range 3.5–13.5%), while samples from HC showed a slightly lower frequency of 6.8% (range 3.7–12.9%) (figure 1A). Statistical analyses revealed no significant differences between the percentage of peripheral CD4+Foxp3+ Treg detected in patients with SSc or LS or HC (Mann–Whitney U test; SSc vs LS p=0.697, SSc vs HC p=0.964, LS vs HC p=0.600). Moreover, a comparison between the percentage of CD4+Foxp3+ Treg detected in the peripheral blood from limited and diffuse SSc patients yielded no significant results (Mann–Whitney U test, p=0.205). There were also no significant differences in the frequency of CD4+Foxp3+ Treg between limited or diffuse SSc compared with LS patients (Mann–Whitney U test; limited SSc versus LS p=0.336, diffuse SSc versus LS p=0.817) and HC (Mann–Whitney U test; limited SSc vs HC p=0.345, diffuse SSc vs HC p=0.329) (data not shown). Further details on determination of peripheral Treg by flow cytometry are included in the supplementary material (supplementary 2).
The median percentage of CD4+CD25++Foxp3+ Treg was 1.1% in SSc patients (range 0.7–3.5%), 1.0% in LS patients (range 0.6–4.1%) and 1.3% in HC (range 0.6–2.8%) (figure 1B). When comparing the percentage of CD4+CD25++Foxp3+ Treg in SSc or LS patients or HC, no significant differences were observed (Mann–Whitney U test; SSc vs LS p=0.782, SSc vs HC p=0.953, LS vs HC p=0.762). In addition, no statistically significant differences in the frequency of the CD4+CD25++Foxp3+ Treg population were detected when patients with limited or diffuse SSc were compared with LS patients (Mann–Whitney U test; limited SSc vs LS p=0.385, diffuse SSc vs LS p=0.224) and HC (Mann–Whitney U test; limited SSc vs HC p=0.312, diffuse SSc vs HC p=0.312) (data not shown). Further details on phenotype of peripheral Treg in SSc and LS are included in the supplementary material (supplementary 3).
Correlation of Rodnan skin score with frequency of peripheral Treg in patients with SSc
In order to evaluate whether disease activity in SSc is related to the frequency of CD4+Foxp3+ or CD4+CD25++Foxp3+ Treg, the prevalence of these cells in peripheral blood was correlated with the modified Rodnan skin score of the patients. Linear regression analysis demonstrated that there was no significant correlation between the frequencies of either CD4+Foxp3+ or CD4+CD25++Foxp3+ Treg and the results of the Rodnan skin score in SSc patients (Student t test; p=0.130 and p=0.566, respectively) (figure 2A, B).
Frequency of peripheral Treg in patients with SSc before and after bosentan therapy
In this study, the possible effect of bosentan on the frequency of Treg in peripheral blood was evaluated in patients with SSc. The percentage of CD4+ T cells that were Foxp3+ was similar in each SSc patient with no significant differences detected following 3 months of bosentan therapy (median 7.0 and 8.8, respectively, Mann–Whitney U test, p=0.841 figure 2C).
Functional analysis of CD4+CD25++ Treg from the peripheral blood of patients with SSc compared with HC
We purified CD4+CD25++ Treg from peripheral blood of patients with SSc and from HC to determine their suppressive capacity on CD4+CD25– Tcon. The suppressive capacity of CD4+CD25++ Treg is shown as the percentage of the proliferation of allogeneic Tcon following anti-CD3/CD28 stimulation, as measured by the incorporation of 3H-thymidine. Although a small reduction in the suppressive activity of CD4+CD25++ Treg was observed in individual SSc patients compared with HC, this difference was not statistically significant (Student t test, p=0.772; figure 3).
Frequency of CD4+Foxp3+ T cells in the skin of patients with SSc and LS compared with other inflammatory skin diseases
We analysed the frequency of CD4+ and Foxp3+ cells in a series of skin biopsy specimens obtained from patients with SSc or LS using immunohistochemistry (figure 4A). In addition, we stained for CD68 to exclude that many of the CD4+ cells are macrophages and to prove the specificity of our results. The frequency of CD4+ cells, expressed as the percentage of the total dermal infiltrate, was significantly reduced in lesional skin specimens of patients with SSc (median 47.7%, range 40.3–64.8%) and LS (median 53.9%, range 20.3-60.2%) compared with PSO (median 64.6%, range 46.7–85.5%, p=0.004 and p=0.003, respectively) (figure 4B). In addition, the percentage of CD4+ cells in SSc and LS skin lesions was slightly, but not significantly, reduced when compared with specimens from LP (median 56.8%, range 30.8–70.8%) and AD (median 56.3%, range 37.5–78.2%). The percentage of CD4+ cells detected in lesional skin biopsies of patients with SSc and LS was not statistically different (p=0.89); the comparison of lesional and non-lesional skin of patients with SSc also revealed no significant difference (p=0.10).
In serial sections, the percentage of Foxp3+ cells in the inflammatory infiltrate of lesional skin biopsy specimens of patients with SSc (median 7.3%, range 3.7–14.0%) was significantly less than the percentage observed in the biopsies from the control diseases (PSO p<0.001, LP p=0.004, AD p<0.001) (figure 4C). The percentage of Foxp3+ cells in the dermal infiltrate of lesional skin biopsies of patients with LS (median 9.8%, range 5.4–22.3%) was slightly higher than in patients with SSc, but this difference was not statistically significant. When comparing lesional and non-lesional skin biopsies of patients with SSc, no significant difference was revealed with regard to the percentage of Foxp3+ cells (p=0.69). However, the percentage of Foxp3+ cells in LS biopsies was significantly lower than the percentage observed in PSO, LP and AD (p=0.001, p=0.011 and p=0.005, respectively).
Furthermore, we calculated the percentage of Foxp3+ cells in relation to CD4+ cells in lesional skin biopsy specimens of patients with SSc, LS, and the other inflammatory skin diseases as controls (figure 4D). The percentage of Foxp3+ cells (among CD4+ cells) was 17.2% (range 9.1–21.7%) in SSc and 18.9% (range 9.2–53.7%) in LS. This was, in both cases, significantly lower than the percentage seen in PSO (45.4%, range 14.9–57.7%; compared with SSc p=0.001; compared with LS p=0.005), LP (51.3%, range 16.3–78.7%; compared with SSc p=0.001; compared with LS p=0.005) and AD (33.0%, range 17.0–55.1%; compared with SSc p=0.002; compared with LS p=0.024). There was no significant difference between the percentage of CD4+ cells that were Foxp3+ in lesional skin biopsy specimens of patients with SSc or LS (p=0.033). Moreover, when comparing lesional and non-lesional skin of patients with SSc, no significant difference was found (p=0.23).
It has been proposed that a reduced frequency of CD4+CD25+ Treg and/or a defect in function could potentially lead to human autoimmune diseases.22,–,25 Accordingly, a decrease of peripheral CD4+CD25+ Treg has been suggested in patients suffering from various autoimmune diseases, including type 1 diabetes16 17 and multiple sclerosis.18 In the present study, no significant differences between the percentages of CD4+Foxp3+ and CD4+CD25++Foxp3+ Treg in the peripheral blood from SSc patients and HC were detected. The further phenotyping of peripheral Treg showed also no abnormality (supplementary 4). Moreover, no significant differences were found comparing the percentage of Treg in the peripheral blood of patients with the limited or the diffuse form of SSc or patients with LS. Therefore, the frequency of Treg in patients included in our study suggests that there might be no systemic loss of these cells in the peripheral circulation of such individuals. However, it is important to take into account that most of the patients received systemic treatment, which might have had an influence on the frequency of Treg. In particular the treatment with corticosteroids and/or immunosuppressive agents, such as cyclophosphamide, can significantly restore the CD4+CD25+ Treg population.26 In contrast to our study, Radstake et al.27 recently found a significant increase of peripheral CD4+CD25+ and CD25+Foxp3+CD127– Treg in limited and diffuse SSc. These results are probably due to a different analysis strategy, as the authors set CD4+CD25+ and CD25+Foxp3+CD127– Treg in relation to the population of CD3+ cells. Another group also demonstrated an increased number of Treg in patients with SSc (expressed as patient:control ratio), correlating with activity and severity of the disease.28 However, Antiga et al29 showed a lower number of CD4+CD25++Foxp3+ T cells in patients with SSc and, similarly, there were reduced transforming growth factor (TGF)-β and interleukin (IL)-10 serum levels in patients with SSc and LS.
In order to further evaluate the possible role of Treg in SSc and the association with clinical manifestations, the frequency of CD4+Foxp3+ and CD4+CD25++Foxp3+ Treg was compared with the degree of skin involvement, as measured with the modified Rodnan skin score.30 31 No significant correlation between the frequency of peripheral Treg and the modified Rodnan skin score in patients with SSc was identified. Furthermore, the effect of the ET-1 receptor antagonist bosentan on the percentage of CD4+Foxp3+ Treg in peripheral blood was evaluated in patients with SSc. Interestingly, a recent study demonstrated that bosentan can restore T cell function in patients with SSc.32 The authors speculated that T cells might be sensitive to bosentan due to potential expression of ET-1 receptors. However, no significant difference in the frequency of peripheral Treg before and after 3 months of bosentan therapy was identified in the present study. These results could be explained by the fact that bosentan is highly specific for both ET receptors (ETA and ETB), which are present on many different cells (eg, vascular smooth muscle cells) but have not yet been shown to be expressed by T cells.33 Therefore, the hypothesis needs to be verified with further in vitro studies that address the expression of the ET-1 receptors on T cells and the role of bosentan in modulating the T cell phenotype.
Functional impairment of Treg has also been proposed as a mechanism to explain the loss of tolerance in human autoimmune diseases. Insufficient suppression by Treg has been found in type 1 diabetes,16 multiple sclerosis,18 34 graft versus host disease,35 autoimmune proliferative syndrome type II36 and rheumatoid arthritis,37 but controversial results have been reported in SLE.26 We detected no significant difference in the suppressive capacity of CD4+CD25++ Treg between patients with SSc and HC; however, further studies need to be performed in a greater number of patients. In summary, no alterations in frequency and function of peripheral Treg could be found in SSc patients although the role of a T cell mediated mechanism in the progression of this autoimmune disease has been suggested.38 In contrast to our findings, Radstake et al27 recently found a compromised function of Treg in patients with SSc. These different results might be explained by a different experimental methodology, as the authors correlated a diminished Treg function with CD69 surface expression and co-cultured SSc plasma with Treg from HC.
Importantly, studies on frequency and suppressive capacity of Treg in peripheral blood do not necessarily reflect the situation in the affected organ. Therefore, the analysis of Treg in samples from inflamed tissue, such as skin biopsies, may be important in elucidating the mechanisms underlying autoimmune diseases affecting the skin. In recent immunohistochemical studies, the number of Treg (Foxp3 and glucocorticoid-induced TNFR-related protein (GITR)) has been investigated in the skin. Interestingly, the frequencies of Foxp3+ and GITR+ T cells were similar in all inflammatory skin diseases studied, such as PSO and LP, compared with normal skin.39 40 These results were confirmed by our group in a previous analysis; moreover, a significantly reduced frequency of CD4+Foxp3+ T cells was detected at the side of inflammation in skin lesions of patients with cutaneous lupus erythematosus compared with other inflammatory skin diseases.21 In the present study, the percentage of Foxp3+ cells among skin mononuclear cells as well as the percentage of CD4+ cells expressing Foxp3 in the dermal infiltrate was significantly decreased in lesional skin biopsies of patients with SSc and LS compared with PSO and other control diseases including LP and AD. Consequently, CD4+Foxp3+ T cells were significantly decreased in skin biopsies of patients with SSc (lesional and non-lesional) or LS compared with the control diseases. There were no significant differences in the frequency of Foxp3+ cells (expressed as the percentage of CD4+ cells) in SSc lesions compared with LS lesions. This might be an indication that autoimmunity is also pathogenetically important in LS as hypothesised by some authors.41,–,43 Furthermore, the results may also support the hypothesis that the same factors, such as environmental stimuli and/or genetic agents, might contribute to triggering the abnormal production of collagen.
In conclusion, this study strongly argues against a defect of CD4+Foxp3+ and CD4+CD25++Foxp3+ Treg in the peripheral blood of patients with SSc. However, it can be suggested that a skin-specific Treg abnormality may be linked to the pathogenesis of this autoimmune disease. It is difficult to analyse whether alterations in the Treg population of the skin are causative for the development of cutaneous lesions in either disease or simply an epiphenomenon. More specific knowledge regarding the cellular and molecular mechanisms related to the frequency and function of CD4+CD25+ Treg in different human diseases will help to improve the design of future Treg-based therapeutic strategies.44
Funding This study was supported by a Heisenberg scholarship from the German Research Foundation to A.K. (KU 1559/1-2) and in part sponsored by Actelion Pharmaceuticals Deutschland GmbH, Freiburg, Germany.
Competing interests None.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Ethics Committee, University of Heidelberg, Germany.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.