Down syndrome, autoimmunity and T regulatory cells

Abstract

Autoimmune diseases are more represented in Down syndrome (DS) individuals compared to chromosomally normal people. Natural T regulatory cells (nT_reg) have been considered to be primary in the role of controlling the intensity and targets of the immune response. We have investigated the phenotypical and functional alteration of nT_reg in a group of DS people. The phenotypical characteristic of T_reg cells of 29 DS was analysed and compared with an age-matched healthy control group. The inhibitory potential of CD4⁺CD25^highCD127^low T regulatory cells was evaluated on autologous CD4⁺CD25^- T cell proliferation in response to activation with a mytogenic pan-stimulus (anti-CD2, anti-CD3 and anti-CD28 antibodies). The CD4⁺CD25^high cells in the DS and control groups were 2·692 ± 0·3808%, n = 29 and 1·246 ± 0·119, n = 29%, respectively (P = 0.0007), with a percentage of forkhead box protein 3 (FoxP3)-expressing cells of 79·21 ± 3·376%, n = 29 and 59·75 ± 4·496%, respectively (P = 0.0015). CD4⁺CD25⁺FoxP3⁺ cells were increased in peripheral blood from DS subjects (DS mean 5·231 ± 0·6065% n = 29, control mean 3·076 ± 0·3140% n = 29). The majority of CD4⁺CD25^high were CD127^low and expressed a high percentage of FoxP3 (natural T_reg phenotype). While the proliferative capacity of DS T cells was not altered significantly compared to normal individuals, a reduced inhibitory potential of T_reg compared to healthy controls was clearly observed (mean healthy control inhibition in T_eff : T_reg 1:1 co-culture: 58·9% ± 4·157%, n = 10 versus mean DS inhibition in T_eff : T_reg 1:1 co-culture: 39·8 ± 4·788%, n = 10, P = 0.0075; mean healthy control inhibition in T_eff : T_reg 1:0·5 co-culture: 45·10 ± 5·858%, n = 10 versus DS inhibition in T_eff : T_reg 1:0·5 co-culture: 24·10 ± 5·517%, n = 10, P = 0.0177). DS people present an over-expressed peripheral nT_reg population with a defective inhibitory activity that may partially explain the increased frequency of autoimmune disease.

Introduction

Autoimmune diseases have a high incidence and prevalence among Down syndrome individuals (DS) compared to chromosomally normal people [1–3]. Coeliac disease has a prevalence of 4·5–7%, autoimmune thyroiditis is diagnosed in 5–54% of DS subjects and type 1 diabetes (T1D) is present in 1%. DS is caused by trisomy of human chromosome 21 and occurs in approximately one of 700 newborns. DS shows various complex phenotypes, including developmental abnormalities, immune system deficiency, typical facial features, mental retardation and congenital heart and gastrointestinal malformations [4–6]. Leukaemias and testicular tumours have an abnormally high incidence in DS individuals, while solid tumours are extremely rare [7]. In the past, when DS children were commonly institutionalized, mortality from respiratory infections was particularly elevated [8]. The increased susceptibility to bacterial or viral infections and leukaemias has been attributed to the dysregulation of the immune system that is one pathological feature of the syndrome [9,10]. In 1979, Levin et al. were the first to reveal in a group of 15 infants (aged 1–15 months) that the thymic histological picture was abnormal [11]. They showed that ‘the normal thymic cortico-medullary demarcation was often missing due to marked lymphocyte depletion in the cortex’. In 1992 Murphy et al. demonstrated an interferon (IFN)-γ and tumour necrosis factor (TNF)-α over-expression [12] and in 1995 they proposed a model suggesting that in DS the over-expression of chromosome 21-encoded gene products leads to impaired interaction between immature thymocyte and thymic stromal cells [13]. The thymus has two main functions for sustaining immunological self-tolerance: clonal deletion of self-reactive T cells (negative selection) and the production of natural CD4⁺CD25^high regulatory T cells (T_reg) cells that express the transcription factor forkhead box protein 3 (FoxP3) [14]. The FoxP3-expressing cells differentiate and mature in the human and murine thymus under the influence of thymic stromal lymphopoietin (TSLP) together with IL-7 [15–18] following ligation of high-affinity T cell receptor (TCR) [19]. The TCR of thymocytes is reduced in DS patients [20]. Subsequently the natural T (nT_reg), CD4⁺CD25^high migrate into the periphery, suppress the autoreactive T cells that escape the thymic negative selection and simultaneously regulate the pathogen-induced inflammatory reactions [21–25]. nT_reg play also a role in favouring tumour growth, tolerance towards transplanted organs or suppressing the graft-versus-host reaction in transplanted patients [26,27].

The higher prevalence of autoimmune disorders present in DS, along with the role of nT_reg cells, led us to investigate if the frequency and the function of circulating nT_reg were normal in a group of 29 children and young adults with DS. For this purpose, T_reg cells, sorted as CD4⁺CD25^highCD127^low, were isolated and cultured with autologous T cells and stimulated with a pan-T stimulus (anti-CD2, anti-CD3 and anti-CD28 monoclonal antibodies-loaded beads). We expected to find a significantly reduced number of nT_reg, but their frequency in the periphery was actually increased. Interestingly, when cultured in vitro, they displayed a reduced suppressive activity, as could be expected according to the increased incidence of autoimmunity in DS individuals.

Materials and methods

Subjects

Eligible patients were selected from a group of subjects with Down syndrome connected to a specific follow-up programme based on AAP (American Academy of Pediatrics) guidelines [3] at our Institution. Flow cytometry data were obtained from 29 DS subjects and 29 healthy age- and sex-matched donors' control group (HD); the proliferation assays were carried out on 10 DS and 10 age- and sex-matched control subjects. DS group included 15 males and 14 females (mean age 11·4 years, range: 1·4–22·8 years). The HD group included 15 males and 14 females (mean age 9·3 years, range: 1·2–23 years); clinical data are summarized in Table 1. In the DS group, five patients (17·2%) had positive antibodies against thyroid peroxidase (TPO) and specific ultrasound scan characteristics for Hashimoto's disease. One patient had positive antibodies against thyrotrophin (TSH) receptor. Positive anti-gliadin antibodies of IgG class were found in 14 patients (45%). Three patients (10·5% overall) presented positive anti-endomysial, anti-tissue transglutaminase immunoglobulin (Ig)A antibodies and a positive duodenal biopsy. One patient was diagnosed with chronic inflammatory demyelinating polyneuropathy (3·5%) and one had vitiligo (3·5%). In our group no patients had type 1 diabetes (T1D). The study protocol was approved by the Ethics Committee of the Insubria University. All parents provided written informed consent prior to participation in the study.

Table 1.

Clinical characteristics of Down syndrome (DS) and healthy donor (HD) subjects

Group	DS	HD
Number (n)	29	29
Gender (male/female)	15/14	15/14
Mean age/range (years)	11·4/1·4–22·8	9·3/1·2–23
Hashimoto thyroiditis	5	–
Grave's disease	1	–
Coeliac disease	3	–
Atopy	2	–
Vitiligo	1	–
Type 1 diabetes mellitus	0	–
Chronic inflammatory demyelinating polyneuropathy	1	–

Flow cytometry analysis of peripheral T cells and assessment of the T_reg phenotype

Human peripheral blood mononuclear cells were freshly separated by Ficoll-Paque (BioWhittaker 17-829E; BioWhittaker, Milan, Italy) from 10–20 ml blood of 29 DS subjects and 29 healthy controls. Immunophenotypic analysis was performed by flow cytometry (BD FACSAria II™ apparatus) and immunofluorescence using the following antibodies: phycoerythrin (PE) mouse anti-human CD25 (clone M-A251; BD Pharmingen, Milan, Italy), PE-cyanin 7 (Cy7) mouse anti-human CD4 (clone: SK3; BD Pharmingen), PE-Cy5 mouse anti-human CD8a (clone: RPA-T8; eBioscience, Milan, Italy) and AlexaFluor® 647 mouse anti-human CD127 (clone HIL-7R-M21; BD Pharmingen). After surface staining, cells were stained intracellularly for FoxP3 according to the manufacturer's recommendations (FoxP3 staining buffer set; eBioscience), and treated with fluorescein isothiocyanate (FITC) rat anti-human FoxP3 antibody (clone: PCH101; eBioscience). Flow cytometry analysis was made with the BD FACSAria II™ apparatus.

In-vitro suppression assay

CD4⁺CD25^highCD127^low T_reg cells [28] and CD4⁺CD25^- responder T cells (T_eff) from patients with DS and HD were isolated from peripheral blood mononuclear cells (PBMCs) by fluorescence-activated cell sorting. The functional characteristic of T_reg cells was tested with the Suppression Inspector human kit (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). It is based on anti-biotin MACSiBead™ particles that are loaded with biotinylated anti-CD2, anti-CD3 and anti-CD28 monoclonal antibodies. One MACSiBead particle per cell (bead-to-cell ratio 1:1) is used for stimulation. T_reg were analysed functionally in vitro by a co-culture assay system with T_eff at different ratios (T_eff : T_reg, 1:0, 1:0·5 and 1:1) in the presence of MACSiBead™ polyclonal stimulus. T_reg alone show a hypoproliferative response (anergy), whereas T_eff alone show a proliferative response. Briefly, 5 × 10⁴ T_eff cells were stained with carboxyfluorescein diacetate succinimidyl ester (CFSE) (5µM CFSE staining solution) and incubated alone or in the presence of 2·5 × 10⁴ or 5 × 10⁴ purified T_reg for 4 days at 37°C in RPMI-1640 culture medium. Because CFSE is incorporated into living cells, its intensity decreases as a function of cell proliferation. The suppressive capacity of T_reg cells towards responder cells in co-culture (T_eff : T_reg ratio 1:0·5 or 1:1) was expressed as the ratio between the percentage of cells proliferating in the presence or absence of T_reg according to the formula [100 × (1 – % CFSE low CD4⁺CD25^-T cells in co-culture/% CFSE low CD4⁺CD25^- T cells alone)].

Statistical analysis

The normality of variable distribution was assessed, and once the hypothesis of normality was accepted (P < 0·05) comparisons were performed by Student's paired or unpaired t-tests, as appropriate. Results were expressed as the mean ± standard error of the mean. Grubbs' test was performed on reference interval data to detect outliers. All statistical analyses were performed using Prism version 5·0 software (GraphPad Software, San Diego, CA, USA; QuickCalcs, http://www.graphpad.com/quickcalcs/index.cfm). P-values less than 0·05 were considered significant.

Results

Circulating T cells

To assess the proportion of distinct subpopulations of T cells in DS patients, a comparative study of CD4⁺ and CD8⁺ T cells was performed. In DS subjects CD4⁺ cells were decreased compared to HD; conversely, CD8⁺ T cells appeared to be slightly increased (see Fig. 1). As a result the CD4/CD8 ratio was decreased significantly (1·4% versus 2·1%).

Fig. 1.

Circulating T cell-subset representation. In Down syndrome (DS) subjects CD4⁺ cells were decreased compared to HD (32·05 ± 1·591% versus 40·40 ± 1·606%, respectively, P = 0.0005, n = 29). No outliers were found in these summaries. Conversely, CD8⁺ T cells appear slightly increased, but with no statistical relevance (23·84 ± 2·029% versus 19·35 ± 1·178%, respectively, P = 0.0610, n = 29); one outlier was found but not removed due to statistical irrelevance. As a result the CD4 : CD8 ratio was decreased significantly (1·4% versus 2·1%).

CD4⁺CD25^highFoxP3⁺ T cells are over-represented in DS patients

Flow cytometric analysis in DS patients highlighted a more represented CD4⁺CD25^high population compared to HD (DS mean 2·692 ± 0·3808%, n = 29, HD mean 1·246 ± 0·119%, n = 29), P = 0.0007. FoxP3 was over-expressed in the CD4⁺CD25^high population of DS subjects (mean 79·21 ± 3·376%, n = 29) compared to HD (mean 59·75 ± 4·496%, n = 29), P = 0.0015. Moreover, CD4⁺CD25⁺FoxP3⁺ cells were increased in peripheral blood from DS subjects (DS mean 5·231 ± 0·6065 n = 29, HD mean 3·076 ± 0·3140, n = 29), P = 0.0026 (see Fig. 2). We found a slightly increased CD4⁺CD25⁺FoxP3⁺ population in Down patients with autoimmune disorders (mean population with autoimmunity: 5·7 ± 1·5% n = 9), but with no statistical relevance. We then analysed and compared the CD4⁺CD25⁺ population for expression of the CD127 marker (α-chain of the IL-7 receptor). Low expression of this marker is associated strongly with the T_reg phenotype, with functional suppressive features [28]. We found that the vast majority of CD4⁺CD25^high were also CD127^low (see Fig. 3), with no statistical difference in the proportion between HD and DS patients. Moreover, a high percentage of FoxP3⁺ was seen in the CD4⁺CD25^highCD127^low cell subset (98·3% in DS and 96·4% in HD).

Fig. 2.

CD4⁺CD25⁺forkhead box protein 3 (FoxP3⁺ ) cells in peripheral blood. We can see an increased natural regulatory T cell (nT_reg) population in Down syndrome (DS) subjects compared to healthy donors (HD) (DS mean 5·231 ± 0·6065, n = 29, HD mean 3·076 ± 0·3140, n = 29, P = 0.0026); no outliers were found.

Fig. 3.

Correlation between CD4⁺CD25^high T cells and CD127 expression in a representative subject. The majority of CD4⁺CD25^high T cells display a CD127^low expression phenotype. The CD4⁺CD25^highCD127^low T cell subset in one Down syndrome (DS) expressed 98·3% of forkhead box protein 3 (FoxP3); the CD4⁺CD25^highCD127^low T cell subset in one control expressed 96·4% of FoxP3 (data not shown).

CD4⁺CD25^highCD127^low T cells from patients with DS exhibit impaired suppressive function in vitro

In order to investigate the regulatory potential of T_reg (CD4⁺CD25^highCD127^low) in DS patients, T cell proliferation assays were set up in which DS and HD T_reg were incubated with autologous T_eff (CD4⁺CD25^–) stimulated in vitro by a pan-T stimulus (anti-CD2-, anti-CD3- and anti-CD28-coated beads). No statistical difference between the proliferation rate of DS CD4⁺CD25^- and HD T cells was found after the pan-stimulus in the absence of T_reg (72·8% versus 79·5%, respectively, P = 0.41). Conversely, when T_reg cells from the same subject were added to the culture, a significantly reduced inhibition of T cell proliferation was observed for DS-derived T_reg compared to HD-derived T_reg. Indeed, the percentage of inhibition (see Fig. 4) by 10 healthy control T_reg was 58·9 ± 4·157% (co-culture T_eff : T_reg 1:1) and 45·10 ± 5·858% (co-culture T_eff : T_reg 1:0·5), whereas the percentage of inhibition by 10 DS T_reg was 39·8 ± 4·788% (co-culture T_eff : T_reg 1:1), P = 0.0075 and 24·10 ± 5·517% (co-culture T_eff : T_reg 1:0·5), P = 0.0177.

Fig. 4.

Co-culture effector T cell–regulatory T cell (T_eff–T_reg) median inhibitory rate. In co-culture T_eff : T_reg 1:1 the median percentage of inhibition by T_reg in control subjects was 58·9 ± 4·157%, whereas in Down syndrome (DS) subjects was 39·8 ± 4·788% (n = 10, P = 0.0075); in co-culture T_eff : T_reg 1:0·5 the median percentage of inhibition by T_reg in control subjects was 45·10 ± 5·858%, whereas in DS subjects was 24·10 ± 5·517% (n = 10, P = 0.0177), calculated as described in Materials and methods. One outlier was identified from the DS 1:0·5 T_eff–T_reg co-culture reference interval data. This data point had a Z-value of 2·40 (P < 0·05), while the critical value of Z for this data set was 2·29. The data were checked and confirmed by a second control; they were not removed because no technical reason was found.

Discussion

DS patients have a characteristically altered immunological asset [9,10]. In our DS population we found a decreased number of circulating CD4⁺ and an increased number of circulating CD8⁺ cells [29]. In this study we show that the proliferative capacity of T cells was not altered significantly compared to T cells from normal individuals, provided that the activation stimulus was exerted through several co-stimulatory molecules such as insoluble anti-CD2, anti-CD3 and anti-CD28 monoclonal antibodies (72·8% versus 79·5%, respectively, P = 0.41). As outlined previously, autoimmune diseases are represented much more in Down syndrome than in healthy controls [1–3]. Although several hypotheses have been put forward in the past to explain the high incidence of autoimmune diseases in DS patients [30,31], the immunological basis of this event is still unclear. An impaired function of nT_reg cells has been shown in many human and murine autoimmune subjects [32]. Mice that lack T_reg (scurfy mice) and individuals with altered expression of FoxP3 gene develop a severe autoimmune-like disease which progresses rapidly to death [immune dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome][33]. Natural T_reg cells are generated in the thymus [18], but the DS thymus presents profound anatomical and architectural abnormalities which may alter the maturation process of these cells. This phenomenon has been documented in patients with Omenn syndrome, a severe congenital immunodeficiency, who carry a severe T_reg defect and a profound anatomical alteration of the thymus gland [34]. DS children also have decreased T cell receptor excision circles (T_REC), which are DNA by-products of a TCR recombination that reflect production of new T cells in the thymus [35]. Similarly, decreased T_RECs as a measure of decreased thymopoiesis are seen in infants with congenital T cell defects [36]. T_reg may be particularly sensitive to this altered maturation process and thus exit the thymus with functional defects. The high incidence of autoimmune diseases in DS may be related at least partially to the functional impairment of T_reg. The mechanisms of suppression by nT_reg include modulation of the cytokine microenvironment, metabolic disruption of the target cell and alteration of dendritic cell activation capacity and cytolysis [37]. In our study we have assessed only their suppressive function, using the dilution of CFSE by proliferating T target cells. Natural T_reg cells which arise in the thymus co-operate with induced T_reg cells which are generated in the periphery following CD4⁺ T cell activation [22].

In our DS group the circulating T_reg cells are increased in number compared to HD, whereas their function is impaired. In IPEX syndrome, nT_reg cells are present but dysfunctional [38].

These characteristics are present in all our DS individuals, irrespective of their actual autoimmune status. We cannot be strongly affirmative on the pathogenic role of the functional defect of these T_reg because different subpopulations of T_regs exist, many of them differentiating in the periphery [32]. We should also take into account that different types of T_reg regulate different types of T helper type 1 (Th1)-, Th2- and Th17-mediated immunity [39–42]: in our in-vitro experiments we used a pan-stimulus for T cells. A functional impairment of these cells is often associated with organ-specific autoimmunity and these disorders can be blocked by the infusion of new T_reg[43]. In DS patients the organ-specific autoimmunity (T1D; Hashimoto thyroiditis and coeliac disease) is remarkably increased in incidence while generalized autoimmunity, such as systemic lupus erythematosus, is present with similar frequency, as in kariotypically normal individuals [44].

Disclosure

None of the Authors has conflicts of interest to declare.

Ethics approval

The study protocol was approved by the Ethics Committee of the Insubria University (Protocol number 0048558). All parents provided a written informed consent prior to participation in the study.