Increased Frequency and Compromised Function of T Regulatory Cells in Systemic Sclerosis (SSc) Is Related to a Diminished CD69 and TGFβ Expression

Background Regulatory T cells (Tregs) are essential in the control of tolerance. Evidence implicates Tregs in human autoimmune conditions. Here we investigated their role in systemic sclerosis (SSc). Methods/Principal Findings Patients were subdivided as having limited cutaneous SSc (lcSSc, n = 20) or diffuse cutaneous SSc (dcSSc, n = 48). Further subdivision was made between early dcSSc (n = 24) and late dcSSc (n = 24) based upon the duration of disease. 26 controls were studied for comparison. CD3+ cells were isolated using FACS and subsequently studied for the expression of CD4, CD8, CD25, FoxP3, CD127, CD62L, GITR, CD69 using flow cytometry. T cell suppression assays were performed using sorted CD4CD25highCD127- and CD4CD25lowCD127high and CD3+ cells. Suppressive function was correlated with CD69 surface expression and TGFβ secretion/expression. The frequency of CD4+CD25+ and CD25highFoxP3highCD127neg T cells was highly increased in all SSc subgroups. Although the expression of CD25 and GITR was comparable between groups, expression of CD62L and CD69 was dramatically lower in SSc patients, which correlated with a diminished suppressive function. Co-incubation of Tregs from healthy donors with plasma from SSc patients fully abrogated suppressive activity. Activation of Tregs from healthy donors or SSc patients with PHA significantly up regulated CD69 expression that could be inhibited by SSc plasma. Conclusions/Significance These results indicate that soluble factors in SSc plasma inhibit Treg function specifically that is associated with altered Treg CD69 and TGFβ expression. These data suggest that a defective Treg function may underlie the immune dysfunction in systemic sclerosis.


Introduction
Over the past decade, there have been tremendous advances in our understanding of the basic processes that control immune tolerance. It is now generally accepted that auto-reactive T cells are present in healthy individuals, but that there mere presence does not necessitate the development of autoimmune disease. The identification of CD4+CD25+ regulatory T cells (Tregs) as a crucial component of self-tolerance has opened a major area of investigation and numerous studies have demonstrated the potent influence of Tregs in suppressing autoimmune disease, transplantation and graft-versus-host disease [1,2,3,4,5,6,7]. Studies in rodents have provided the first evidence for the existence of a naturally occurring population of CD4+CD25+ professional regulatory/suppressor T cells, which upon in vitro TCR-mediated stimulation, suppress proliferation of effector T cells [3,8]. In the periphery of young mice not prone to autoimmune disease, Tregs constitute a stable 10% of CD4+ T cells. In contrast, mice genetically prone to autoimmune disease such as diabetes have markedly diminished circulating Tregs [9,10].
Tregs have unique and robust immunosuppressive activity. The cells require specific TCR-mediated activation to develop regulatory capacity, but their effector function appears to be nonspecific, regulating local inflammatory responses through a combination of cell-cell contact and suppressive cytokine production [11,12]. In addition to naturally occurring Tregs, several therapeutic interventions promote Treg development and function [13]. These so-called ''adaptive'' Treg populations share many features attributed to natural occurring Tregs, but can differ in critical cell surface markers [14].
In humans, the important role of Tregs in various autoimmune diseases has been underscored by numerous seminal studies. For instance, Tregs derived from patients with rheumatoid arthritis (RA) are defective in their ability to suppress cytokine production and to convey a suppressive phenotype to CD4+ effector T cells, which was at least partly restored upon treatment of TNFa neutralizing therapies [15]. Moreover, the interaction of Tregs with activated monocytes from patients with RA even led to a diminished suppressive activity possibly underlying their diminished capacity in vivo [16]. Likewise, it was demonstrated by several groups that the number and suppressive capacity of Tregs is altered in patients with systemic lupus erythematosus [17,18,19].
Systemic sclerosis (SSc) is a complex autoimmune disease characterized by an excessive deposition of matrix molecules, leading to fibrosis of multiple organs including the skin, lungs, heart and gastrointestinal tract, and often leading to severe morbidity and premature death. Although the role of immune dysfunction in the pathogenesis of SSc is currently not well understood, alterations in cellular immunity are typified by aberrant T cell biology both in the skin as well as circulation of SSc patients. For example, CD4+ T cells are increased in the circulation of SSc patients [20,21], whereas NKT cells and c/d T cells are decreased [22]. In addition, lesional skin from SSc patients displays various features consistent with T cell activation [21,23,24]. Finally, circulating T cells from SSc patients show altered secretion of various inflammatory mediators compared to T cells from healthy controls [25,26].
T cell priming by professional antigen presenting cells is tuned by an orchestra of inflammatory mediators, of which TGFb, IL-23, IL-6, IL-22 and IL-1a are considered the most influential. For instance, in the absence of other pro-inflammatory mediators, TGFb production by dendritic cells induces FoxP3, a Treg marker [27,28]. In contrast, TGFb in combination with IL-1a, IL-6 or IL-23 drives the expression of RORcT, a proliferation factor specific for the recently identified Th17 subset [29,30,31,32]. Intriguingly, IL-23, IL-1a and IL-17 have been found increased in the circulation of SSc patients compared to healthy controls [33,34,35,36]. Although TGFb is not increased in SSc plasma, multiple studies have strongly implicated this cytokine as a major stimulus of fibrosis in involved organs. Together, these observations suggest that altered Treg function might play a key role in SSc pathogenesis. To address this issue, we set out to investigate changes in the number and/or function of Tregs in the peripheral blood of patients with SSc, taking into account the different disease phenotypes. In this paper, we show that Tregs are more frequent in SSc patients but are defective in their capacity to suppress proliferation of CD4+ effector T cells. We go on to demonstrate that this diminished suppressive effect of Tregs in SSc is associated with markedly lower expression of the activation marker CD69. Finally, we show that the diminished suppressive capacity and absent upregulation of CD69 upon activation is dependent upon soluble factors present in the plasma of SSc patients. Together these data suggest that diminished T regulatory capacity is present in SSc and that the regulatory deficiency is due to circulating factors rather than an inherent defect of Tregs.

Ethical review board statement
All samples were obtained with written informed consent after approval of the Institutional Review Board at the Boston University School of Medicine, Lund Univeristy medical Hospital and the Radboud University Nijmegen Medical Center.
Study population. Sixty-eight patients presenting to the Arthritis Center, Boston Medical Center were included in the study. This study was approved by the Boston University Medical Center Institutional Review Board. All of the patients met the American College of Rheumatology preliminary criteria for the classification of SSc [37]. Patients were subdivided as having limited cutaneous SSc (lcSSc, n = 20) or diffuse cutaneous SSc (dcSSc, n = 48) on the basis of the extent of their skin involvement [38]. A further subdivision was made between early dcSSc (n = 24) and late dcSSc (n = 24) based upon the duration of disease, defining early dcSSc as patients having a disease duration ,2 years and late dcSSc as patients having a disease duration longer than 3 years. As a comparator group 26 healthy controls were studied. Patients were allowed to use low-dose prednisolone (,10 mg daily) at inclusion of the study. Patients receiving higher doses were excluded.

Isolation
of PBMCs, CD3 + cells and flowcytometry. PBMCs were isolated from heparinized venous blood by using density-gradient centrifugation over Ficoll-Paque (Amersham Bioscience). Next, CD3+ cells were isolated from PBMCs using CD3 microbeads according to manufacturer's protocol (Miltenyi Biotec). To this aim, 10 x 10 4 CD3+ cells were re-suspended in 100 ml buffer (PBS + 1% BSA) on ice. After isolation, cells were directly transferred into RPMI 1640 media supplemented with 2nM L-glutamine, 100 U/mL/ml penicillin/streptomycin (Life technologies), and 10% FBS (BioWhitacker) in 96-well U-bottom plates (Nunc). For flowcytometric analysis, CD3+ were kept on ice and washed extensively with citrated PBS containing 1% FCS. Than, after using the protocol for fixation, intra-cellular staining was achieved using 10 ml of FITC, APC or PE-conjugated antibody that was added and incubated on ice for 20 min. 300 ml FACS buffer was than added and T cells were pelleted, resuspended in 200 ml buffer, and stained for the intracellular marker FoxP3/TGFb were appropriate as conducted by the recommended procedure obtained from the manufacturer (Miltenyi Biotec Inc., CA, USA). Thereafter, cells were washed in buffer, fixed with 2% formaldehyde, washed again in buffer and stored at 4uC. The cells were analyzed using a LSRII FACScan flow cytometer (BD Biosciences) and data were processed using FlowJo software. In all experiments, the purity of CD3+, CD25 high CD127 low and CD25 low CD127 high cells was .97%. plasma from an early dcSSc or plasma alone. After 12 hours of stimulation cells were analyzed on expression of CD69 by flowcytometry as previously described.

Measurement of soluble and intracellular
TGFb. Intracellular TGFb expression in CD25 high CD127cells was investigated using a monoclonal antibody for TGFb (BD Bioscience, NJ, USA) and the intracellular staining protocol as used for the FoxP3 staining. After the staining protocol, cells were fixed with 2% formaldehyde, stored at 4uC and analyzed on a flow cytometer the next day. TGFb was assayed using mink lung epithelial cells stably transfected with a plasminogen activator inhibitor-1 promoter/luciferase reporter plasmid (provided by D. Rifkin) as described previously by Abe et al [39]. Statistical analysis. Values are shown throughout the paper as mean6sem. Proportions of lymphocyte subpopulations were compared using the Student's t test for normally or not normally distributed populations where appropriate. Relationships between different values were examined using Pearson's correlation coefficient and Spearman's rank correlation tests. All statistical analyses were performed using Graphpad Prism (GraphPad Prism 4.0 by Graph Pad software Inc.)

Results
CD4 + CD25 + FoxP3 + CD127cells are markedly increased in the circulation of SSc patients irrespective of disease phenotype Human peripheral blood contains a heterogeneous subset of CD4+CD25+ T cells that comprises T regulatory cells (Tregs) and a substantial number of activated effector T cells. To date, the expression of FoxP3 and CD127 remain the best and most specific markers of Tregs [40,41]. Since we postulated that the number and/or phenotype of Tregs in SSc is altered compared to controls, but may also differ among different clinical SSc subtypes, we here studied the number and phenotype of Tregs from patients with limited cutaneous SSc (n = 20), late diffuse cutaneous SSc (n = 24) and early diffuse SSc (n = 24) in comparison with those from healthy controls (n = 26). The clinical characteristics of all patients included in this study are presented in Table 1. Despite similar absolute numbers of CD3+ cells, flowcytometric analysis with the markers CD4, CD25, FoxP3 and CD127, demonstrated that both CD4 + CD25 + (12.461.0 vs. 27.562.8, P,0.0001) and CD25 + FoxP3 + CD127 -(2.960.5 vs. 17.361.9, P,0.0001) cells, (further designated as Tregs) are markedly increased in the circulation of SSc patients compared to controls (Figure 1a, b). Further stratification to SSc disease phenotype revealed a significantly higher number of CD4 + CD25 + (P = 0.01) and CD25 + FoxP3 + CD127 -(P = 0.01) in SSc patients with edSSc compared to ldSSc (Figure 1b), but no other significant differences between SSc phenotypes were detected. Notably, two patients with ldSSc and two with edSSc received cyclophosphamide pulse therapy for their disease. Whereas both ldSSc responded clinically well only one patient with edSSc did. In these three patients the percentage CD25 + FoxP3 + CD127cells was much lower (6.062.1) compared with the other patients that were not treated. The edSSc patient that received cyclophosphamide pulse therapy but did not show a clinical response showed a frequency of 22.8% CD25 + FoxP3 + CD127cells. All the patients had received cyclophosphamide longer than 3 months ago. Further analysis focusing on CD25+ bright (top 10%) and CD25+ very bright (top 2%) cells revealed a similar expression of the markers FoxP3 and CD127 among all individuals, both on the levels of percentage positive cells (Figure 1c, d), as well on the mean fluorescence intensity (MFI, data not shown). Taken together, these data suggest that SSc patients have a markedly increased frequency of T regulatory cells, which is not related to an altered expression of markers characterizing Treg phenotype.

Aberrant expression of phenotypic markers CD62L and CD69 on CD25+FoxP3 bright and CD25+FoxP3+ verybright from SSc patients
Although we observed a markedly increased frequency of CD25 + FoxP3 + CD127cells phenotypically representing Tregs in SSc, these patients continue to have active disease suggesting altered T cell suppressive activity. To address this, we next investigated the expression of markers potentially reflecting T cell activation including GITR, CD62L and CD69. Although the function of glucocorticoid-induced tumor necrosis factor receptor related protein (GITR) remains to be fully elucidated, it is generally accepted that GITR expression is increased upon TCR engagement, reflecting T cell activation [42]. As expected, GITR expression on CD25+Foxp3+, CD25+FoxP3+ bright and CD25+FoxP3+ verybright from healthy donors gradiently increased using flowcytometry (Figure 2a). In addition, the expression of GITR on CD25+FoxP3+, CD25+FoxP3+ bright and CD25+FoxP3+ verybright was comparable between healthy controls and SSc patients and among the investigated SSc phenotypes. In contrast, the expression of CD62L and CD69 was markedly lower in SSc patients compared to healthy controls (Figure 2b,  c). CD62L is a L-selectin that is upregulated upon Treg activation and highly critical for Tregs to enter the lymph node and to carry out their local suppressive function [43,44]. CD69 expression is pivotal for Treg function, potentially via upregulation of TGFb production upon cross-linking [45,46]. CD69 on CD25 high (37.065 vs. 17.665 vs. 5.362) and CD25 veryhigh (35.168 vs. 17.865.7 vs. 2.460.9) T cells significantly decreased in a step-like manner, comparing healthy controls to patients with lSSc, ldSSc and edSSc phenotypes. Intriguingly, and in line with that observed in other autoimmune diseases, the expression of CD69 on CD4 + effector T cells was significantly increased in all SSc patients compared to controls and followed an inverse correlation with the CD69 expression on CD25 high or FoxP3 high cells suggesting, that the regulation of CD69 expression is specifically altered on the Treg population in SSc [47,48,49] (Figure 2d). In the search for potential SSc characteristics that might correlate with CD69 on Tregs in SSc, we found a significant association between the disease duration in lSSc patients whereas no association was present in patients either with ldSSc or edSSc (Figure 2e). Diminished suppressive capacity of CD25 + FoxP3 + CD127regulatory T cells from SSc patients is correlated with CD69 expression and TGFb levels Taken together, our observations imply that although SSc patients have a significantly increased number of CD25 + FoxP3 + CD127cells in the circulation, these cells phenotypically have markers suggesting impaired suppressive activity. To test the regulatory activity of these cells, we studied the capacity of CD25 high CD127 low cells from healthy controls (n = 8), lSSc (n = 6), ldSSc (n = 9) and edSSc (n = 8) patients to suppress the proliferation of CD4 + effector cells. As expected, Tregs from healthy controls efficiently suppressed the proliferation of CD4+ effector cells by 87.3%64.9, whereas nonregulatory T cells (CD25 low CD127 high ) did not (8.5%62.8). In contrast, Tregs obtained from SSc patients all had a markedly diminished suppressive capacity compared to those from healthy donors (Figure 3a) with T regs from lSSc, edSSc and ldSSc suppressing CD4+ effector cell proliferation by, respectively, 28.2%66.0 (P = 0.0001), 56.0%68.5 (P = 0.006) and 18.3%65.2 (P,0.0001). Since CD69 expression by Tregs has been associated with the production of TGFb [46], one of key molecules implicated in suppressor activity, we investigated the possible relationship between CD69 expression and the diminished suppressive effect observed in SSc. Interestingly, the suppressive capacity correlated significantly with CD69 expression in all groups (Figure 3b).
We next investigated the expression levels of TGFb in the Tregs from SSc patients compared to healthy controls and their CD45Ra+ cells. In line with the CD69, which was specifically lower on regulatory T cells in SSc, also TGFb expression was significantly decreased by regulatory T cells obtained from SSc patients compared to those from healthy controls (Figure 3c). TGFb expression on Tregs from ldSSc and edSSc patients was significantly lower compared to that from patients with lSSc (P = 0.008), whereas no difference was observed between ldSSc and edSSc. Measurement of soluble TGFb in the supernatant revealed no measurable TGFb, suggesting that TGFb confers its effect as membrane-bound (data not shown).

A fraction smaller than 10kD in SSc plasma inhibits the suppressive capacity of regulatory T cells and abrogates the upregulation of CD69 specifically on regulatory T cells
As inflammatory cytokines play an important role in the pathogenesis of SSc and regulatory T cell function, we next investigated whether the diminished suppressive effect of Tregs from SSc could be carried over by soluble factors in the circulation of SSc patients or alternatively could be due to an inherent defect in Tregs. Unexpectedly, the addition of 10% plasma from edSSc patients completely abrogated the suppressive capacity of Tregs on CD4+ effectors cells from healthy controls, an observation that was highly consistent throughout 5 experiments using plasma samples from 5 edSSc and 2 ldSSc patients (Figure 4a). The addition of 25% plasma had a similar effect although somewhat less potent as 10% plasma, a phenomenon that was probably caused by the TGFb present in patients plasma, that partly restored the suppressive capacity of Tregs. In contrast, the addition of plasma obtained from healthy controls did not have a significant effect on the suppressive capacity of Tregs.
Based on our observations that CD69 expression correlates with the diminished suppressive capacity in SSc, we hypothesized that the plasma of SSc patients had a direct effect on the regulation of CD69 expression. To test this, we stimulated freshly isolated CD3+ cells and CD25 high CD127 low T cells from healthy controls with the potent T cell activator PHA. PHA markedly induced CD69 expression both on CD3+ and CD25 high CD127 low T cells (Figure 4b). However, plasma from edSSc patients also significantly increased CD69 expression on CD3+ cells (P = 0.0007) and had an additive effect in combination with PHA (P = 0.02). In contrast with the effect of plasma on CD3+ cells, the addition of edSSc plasma to CD25 high CD127 low T cells did not increase CD69 expression. More intriguingly, the addition   of SSc plasma to CD25 high CD127 low T cells stimulated with PHA completely abrogated the effect of PHA. Since we observed a lower CD69 expression on regulatory T cells freshly isolated from SSc patients, we studied whether these cells still possess the ability to increase CD69 expression upon activation. Co-incubation of regulatory T cells either from patients with lSSc, ldSSc or edSSc led to an clear increase of CD69 expression that could be inhibited by plasma from edSSc patients (Figure 4c). In all experiments presented here, plasma from healthy controls was taken into account but did sort any inhibitory effects as that observed from SSc patients.

Discussion
SSc is an autoimmune disease that reflects several features suggesting dysregulated T cell activation [24,50,51]. The data presented here suggest that dysfunctional Tregs may play an important role in SSc. We show that although the number of Tregs is markedly increased in all clinical SSc phenotypes, these Tregs have a diminished capacity to control CD4 effector T cells. Further we show that their defective function correlates with lower expression of CD69 and TGFb.
Tregs have not been previously characterized in patients with SSc; however, they are critical in maintaining self tolerance and preventing autoimmunity. In several other autoimmune disease Tregs have been implicated in pathogenesis. For example, lupus prone mice, depleted of CD4 + CD25 + cells by thymectomy, have enhanced expansion of autoreactive T cells and accelerated autoantibody production [52]. Conversely, restoration of the CD4 + CD25 + cell population from syngeneic normal mice effectively abrogates the development of autoimmune disease, as has treatment with in vitro expanded Tregs [52,53,54]. Similar evidence originates from experimental arthritis, diabetes and multiple sclerosis models, further highlighting the crucial role of the Tregs in controlling the delicate balance between tolerance and autoimmunity. More recently, several studies performed in patients with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) revealed an aberrant frequency and/or function of Tregs thus indicating their crucial role in human diseases [18,19,55,56,57]. However, none of these studies reported the markedly increased frequency of CD4 + CD25 + and CD25 + / FoxP3 + CD127cells found in our study. In contrast, although some inconsistencies exist, most of these studies found a decreased frequency of circulating Tregs. There appear to be some discrepancies in the literature based on the sole use of CD4 and CD25 as markers for Tregs. However, co-expression of CD4 and CD25 can be induced upon multiple inflammatory events and does not necessarily guarantee suppressive capacity. Therefore, the limited use of these markers could merely reflect activation and thus lead to a false assessment of elevated Treg numbers. More recently, it has been shown that the combination of FoxP3 and CD127 expression is highly specific for discriminating Tregs from activated T cells. FoxP3 expression correlated inversely with CD127 expression, and CD4+CD25 high FoxP3 high CD127 low cells were found to have the most potent suppressive activity [40,58]. In the current study the combination of all these markers was used to characterize and isolate regulatory T cells, confirming our observations of a markedly increased frequency of circulating Tregs in SSc patients.
TGFb is known to potently induce expression of the proliferation factor FoxP3, characterizing Tregs. As TGFb is generally accepted as the key regulator of SSc pathogenesis, the increased frequency of Tregs in SSc was not surprising. TGFb is crucial in the induction of FoxP3 expression and induction of suppressive activity by conversion of CD4+CD25-T cells [59]. Therefore, increased TGFb found in SSc might drive the increased frequency of CD25 high FoxP3 high CD127 -. Indeed, our observation of increased FoxP3, despite comparable levels of CD25 and GITR expression in SSc patients, suggests that Tregs from SSc patients are activated to some extent. The observation that CD62L, a marker that is highly expressed on naturally occurring (thymically-derived) regulatory T cells, is lower in SSc patients suggests that these Tregs originate through conversion of CD4 + CD25 -T cells. These so-called ''adaptive'' Tregs share many features with naturally occurring Tregs, but can differ in critical cell surface biomarkers and functional attributes [14]. For instance, Tregs can mediate their suppressive effects through the production of IL-10 versus TGFb [60,61].
In contrast to CD25 and GITR expression, CD69 expression on Tregs was significantly lower in SSc patients and correlated closely with diminished suppressive activity. Further, upregulation of CD69 by T cell stimulation was completely abrogated by plasma from SSc patients, suggesting the presence of soluble factors in SSc plasma that inhibit CD69 and consequently, the suppressive capacity. Interestingly, the effect of plasma on CD69 expression was highly specific for Tregs, since CD69 regulation on other T cells was not affected. SSc patients show many features suggesting that autoimmune and inflammatory factors may stimulate profibrotic organ damage. For instance, accumulating evidence implicates inflammatory mediators in the Th17 pathway, such as IL-6, IL-1a, IL-23 and IL-17 itself, but also those in the Th2 (IL-10, IL-4), Th1 (IFNc) and other inflammatory pathways, such as IFN type I and TNFa, in this condition (unpublished results [33,34,35,36]). It is therefore tempting to speculate that several mediators could inhibit Treg CD69 expression in SSc patients. In this light, the observation that the three patients who had a clinical response to treatment had a Treg frequency, CD69 expression and suppressive capacity that was almost comparable to that observed in healthy controls is intriguing. Whether these observations are related to lower levels of inflammatory mediators in patients with a therapeutic response will require further investigation.
The potential of Tregs to modulate immune responses has led to considerable interest in their use for clinical intervention in autoimmune diseases. Two broad therapeutic applications have been considered: first, to expand the regulatory T cell compartment ex vivo with the goal of re-infusion and second, to manipulate the immune system in vivo resulting in an increase of Tregs. The latter approach has been shown to be highly applicable by seminal studies by Ehrenstein et al. in which a monoclonal antibody against TNFa led to a re-occurrence of CD4 + CD25 + CD62L -T cells with high suppressive activity [15,62]. Of interest for the current study the suppressive effects of Tregs in these latter studies were found to be contact dependent since the neutralization of TGFb and IL-10 did not block the effect. This is consistent with our observation that intracellular expression of TGFb on Tregs corresponded well with their suppressive capacity, whereas no TGFb was found in the culture supernatants. In our studies we demonstrate that a soluble factors in the plasma of SSc patients is responsible for the dramatic effects observed on suppressive activity, CD69 and TGFb expression. In addition, we did not find evidence for an inherent defect in lower Treg CD69 expression in SSc patients, since activation of these cells led to increased expression.
The factors driving TGFb production are not well resolved. The role of CD69 in the production of TGFb by T cells was shown in several studies. For instance, it was demonstrated that CD69-/mice display greatly prolonged tumor survival that was related to a decreased production of TGFb. CD69 engagement induced TGFb production by NK and T cells [63]. With respect to autoimmunity, CD69-/-mice showed a higher incidence and severity of collagen-induced arthritis, which again were correlated with reduced levels of TGFb [46]. The observation that CD69 surface expression closely mirrors intra-cellular TGFb expression both on CD45Ra as on CD25 high FoxP3 high CD127cells is in line with the notion that CD69 is implicated in TGFb production by T regs.
Altogether, our observations provide a rationale for therapeutic intervention to restore suppressive activity by T regs in SSc. More careful studies designed to identify the nature of factors that moderate the effects in the circulation are warranted.