Impaired in vitro regulatory T cell function associated with Wiskott-Aldrich syndrome

Abstract

Wiskott-Aldrich syndrome (WAS) is a primary immunodeficiency characterized by the contradictory coexistence of impaired T-cell function and exaggerated T-cell-mediated pathology, including autoimmunity and eczema. WAS protein (WASp)-deficient mice are also immunodeficient and can develop autoimmune disease. Since defects in regulatory T-cells (Treg) are associated with autoimmunity, we examined the presence and function of these cells in WAS patients and WASp-deficient mice. We found that CD4⁺CD25⁺FOXP3⁺ Treg cells can develop in the absence of WASp expression. However, Treg cells both from WASp-deficient mice and from four out of five WAS patients studied showed impaired in vitro suppressor function. In WASp-deficient mice, this defect could be partially rescued by pre-activation with IL-2, suggesting that inadequate cell activation may play a role in WASp-deficient Treg dysfunction. These findings may provide insights into the complex pathophysiology and paradoxical phenotypes of WAS and suggest new therapeutic modalities for autoimmunity in these patients.

INTRODUCTION

The Wiskott-Aldrich syndrome (WAS) is a rare X-linked primary immunodeficiency characterized by thrombocytopenia, eczema, recurrent infections and a high incidence of malignancy (1-3). A large proportion of WAS patients (40-70%) are also affected by at least one autoimmune disorder (4, 5). The disorder is caused by mutations of WAS, the gene encoding the WAS protein, WASp, a key regulator of actin polymerization that is expressed in non-erythroid hematopoietic cells (2). T cells isolated from WASp-deficient mice and WAS patients show multiple defects including reduced TCR-induced proliferation and IL-2 production associated with impaired actin polymerization, defects which are likely to contribute to the immunodeficiency in WAS (6-9).

A central question in understanding the pathophysiology of WAS is why immunodeficient patients develop symptoms suggestive of hyperactivation of immune compartments, including eczema and autoimmune diseases (5, 10). It has recently been recognized that CD4⁺CD25⁺ Treg cells play an important role in the negative regulation of immune responses and the prevention of autoimmunity (11-13). These cells are non-responsive (anergic) to TCR stimulation and can suppress proliferation of CD4⁺CD25^- T target cells in culture. Importantly, mutations affecting FOXP3, a transcription factor required for the development of this lineage, lead to severe autoimmune disorders in humans and mice (14-19). Abnormal Treg function has also been demonstrated in several human autoimmune disorders, including type 1 diabetes, systemic lupus erythematosus and rheumatoid arthritis (12). We therefore hypothesized that abnormalities in Treg cells may contribute to the pathogenesis of autoimmune complications associated with WAS. To evaluate this question, we examined the presence and function of Treg cells from WASp-deficient mice and humans.

METHODS

Patients

Clinical and molecular characteristics of the WAS patients included in this study are listed in Supplementary Table 1. Mutation analysis of the WAS gene and flow cytometry analysis of WASp expression were performed as described (20). All human studies were approved by the NHGRI Institutional Review Board.

Mice

Was^-/Y and Was^-/- (129S6/SvEvTac-Was^tm1Sbs/J) (8) and control 129S6/SvEvTac mice were obtained from Jackson Laboratory and Taconic, respectively and maintained in sterile housing with sterile food and water. Mice were maintained and experiments performed according to NHGRI Animal Care and Use Committee guidelines. All mice used in these studies were between 6-10 weeks of age.

Cell culture reagents and antibodies

Mouse cells were cultured in RPMI-1640 (Invitrogen), 10% FCS, 100 U/ml penicillin, 100 μg/ml Streptomycin sulfate and 2 mM glutamine. Human cells were cultured in X-VIVO 20 (BioWhitaker, Walkersville, MD) supplemented with 10% heat-inactivated fetal bovine serum (Gemini Bioproducts Inc., Woodland, CA), 100 U/ml penicillin, 100 μg/ml Streptomycin sulfate and 2 mM glutamine (all from Invitrogen, Carlsbad, CA).

Antibodies were from BD PharMingen (San Diego, CA), with the exception of anti-human CD25 (Immunotech, Marseille, France), anti-human-CD3 (Ortho Diagnostics, Raritan, NJ), and anti-human and mouse/rat FOXP3 (eBioscience, San Diego, CA).

Treg cell isolation

To obtain mouse Treg cells, CD4^hiCD25^hiCD8^- and CD4^hiCD25^-CD8^- cells were isolated either by cell sorting or AutoMACS (Miltenyi) purification from purified lymph node and splenic CD4⁺ cells pooled from 7-10 mice per genotype, as described (21). Mouse accessory cells (APCs) were obtained from complement-mediated T-cell depleted splenocytes treated with 10 μg/ml mitomycin (Sigma-Aldrich, St. Louis, MO), for 3h at 37°C.

Human Treg cells were purified by sorting CD4⁺CD25^hi cells as previously described (22). Briefly, peripheral blood mononuclear cells (PBMC) were isolated by Ficoll-Hypaque (Amersham Pharmacia, Piscataway, NJ) gradient centrifugation, then stained for 30 minutes at room temperature with anti-CD4-PerCP-Cy5.5, anti-CD25-PE, anti-CD14-APC and anti-CD32-APC in staining buffer (PBS + 2% FBS). CD4⁺ cells were identified on FACSAria (BD Bioscience, Bedford, MA) by gating on lymphocytes using forward and side scatter and excluding APC-positive cells. Based on their degree of CD25 expression, CD4⁺CD25^hi (Treg, top 1% cells), CD4⁺CD25^lo (activated cells), and CD4⁺CD25^- (target cells) were then isolated as previously reported (22). Human accessory cells were obtained by immunomagnetic depletion of CD3⁺ cells using CD3-coated beads (Miltenyi Biotec Inc., Auburn, CA) from PBMC of healthy controls, followed by irradiation (3500 R).

In vitro suppression assays

Mouse CD4⁺CD25⁺ Treg suppression assays were performed in triplicate using 50,000 CD4⁺CD25^- target cells, 50,000 accessory cells and varying numbers of CD4⁺CD25⁺ Tregs in the presence of 0.5 μg/ml soluble anti-CD3 (2C11) as described (21). Proliferative characteristics of Treg cells were assessed by stimulation with anti-CD3 in the presence of accessory cells, but in the absence of CD4⁺CD25^- target cells. Alternatively, CD4⁺CD25⁺ Treg cells were pre-activated with anti-CD3 (1 μg/ml) + 50U IL-2 for three days (23) in the presence of equivalent numbers of accessory cells prior to washing to remove IL-2 and testing in suppression assays.

Suppression assays using human Treg cells were performed essentially as previously described (24). Briefly, CD4⁺CD25^- targets and/or CD4⁺CD25^hi Treg cells were stimulated with a combination of soluble anti-CD3 (OKT3; Ortho Diagnostics) and anti-CD28 (28.2 BD Pharmingen) antibodies (5 μg/ml each) in U-bottom 96-well plates in duplicate at a ratio 1:1 (3×10³ cells/well each) in a final volume of 200 μl with 3×10⁴ allogeneic accessory cells/well. After 5 days of culture, 100 μl of media was removed per well for IL-2 detection using anti-human-ELISA assay kit II (BD Bioscience) and one μCi of [³H]-thymidine (Amersham Biosciences, Piscataway, NJ) was added to each well for 16 h before harvesting.

RESULTS

Treg function is defective in Was knockout mice

To evaluate the requirement for WASp in CD4⁺CD25⁺ Treg cells, we first studied Treg numbers and function in WASp-deficient mice. Although spleens and lymph nodes from WASp-deficient mice had fewer CD4⁺CD25⁺ T cells on average compared to wild type (WT) animals, we found that these cells expressed normal levels of Foxp3 (Fig. 1A). Interestingly, we observed relatively normal numbers of CD4⁺Foxp3⁺ cells in spleens of WASp-deficient animals. Gating on these CD4⁺Foxp3⁺ cells revealed lower levels of CD25 expression than on WT CD4⁺Foxp3⁺ cells, similar to what observed in IL-2 knockout mice (25), and perhaps accounting for the decreased numbers of CD4⁺CD25⁺ cells observed. As expected, isolated CD4⁺CD25⁺ cells from either WT or WASp-deficient mice failed to proliferate in response to anti-CD3 stimulation in the presence of antigen presenting cells (Fig. 1B, “Treg” bars).

Figure 1. Impaired suppression function of CD4⁺CD25⁺ Treg cells from WASp-deficient mice.

(A) WASp-deficient mice have similar numbers of CD4⁺Foxp3⁺ regulatory T cells, but these cells express lower levels of CD25. Flow cytometry of splenic cells are shown. (B) Freshly isolated WASp-deficient CD4⁺CD25⁺ Treg cells poorly suppress the proliferation of WT CD4⁺CD25^- target cells. WT CD4⁺CD25^- target cells (50,000/well) were plated with accessory cells and co-cultured in triplicate with the indicated numbers of freshly isolated WT or WASp^-/- CD4⁺CD25⁺ Treg cells in the presence of anti-CD3. Treg sample refers to CD4⁺CD25⁺ T cells cultured with accessory cells in the presence of anti-CD3. Cell proliferation was assessed at the end of a 3-day culture period by [³H]-thymidine incorporation and is presented as the mean ± SEM of triplicate cultures. Data is representative of three independent experiments. (C) [³H]-thymidine incorporation by WASp-deficient and WT Treg cells following 3-day pre-activation (stimulation with anti-CD3 + IL-2). (D) Pre-activated WASp-deficient Treg cells show reduced suppression activity against the proliferation of WT CD4⁺CD25^- target cells. Pre-activated CD4⁺CD25⁺ Treg cells were plated and cultured as in Figure 2B. Cell proliferation was assessed by [³H]-thymidine incorporation after 3-days of culture, as above. Data is the mean of triplicate cultures ± SEM and is representative of two independent experiments. Average proliferation of CD4⁺CD25⁺ was less than 220 cpm.

To determine whether these Treg cells exhibited suppressor function, isolated CD4⁺CD25⁺ cells from wildtype (WT) and WASp-deficient mice were cultured with WT CD4⁺CD25^- cells in an in vitro suppression assay. Since WASp-deficient T cells proliferate poorly in response to TCR stimulation in vitro (8, 9), we used WT CD4⁺CD25^- cells as targets in these assays. CD4⁺CD25⁺ T cells from WT mice suppressed proliferation of WT CD4⁺CD25^- cells in a dose-dependent fashion. In contrast, WASp-deficient CD4⁺CD25⁺ Treg cells failed to effectively suppress target cell proliferation over a wide range of cell ratios, even at suppressor to target ratios of one or greater (Fig. 1B).

Preactivation with IL-2 can partially rescue defects in suppression

Although WASp-deficient T cells respond poorly to TCR stimulation in vitro, their proliferative responses can be partially rescued by addition of IL-2, particularly in the presence of accessory cells (6, 9). To investigate whether the defective suppressor function resulted from a failure of adequate cell activation, we prestimulated CD4⁺CD25⁺ Treg cells in the presence of anti-CD3 plus IL-2 and accessory cells for three days. It has been previously shown that CD4⁺CD25⁺ Treg cells can proliferate and maintain suppressor activity in these culture conditions (26). After prestimulation, we observed similar proliferation of WT and WASp-deficient CD4⁺CD25⁺ cells (p=0.13, Student t-test), despite their lack of proliferation in the absence of IL-2 (Fig. 1C versus 1B, “Treg” bars). Preactivation clearly improved the ability of WASp-deficient Treg cells to suppress WT target cell proliferation (Figure 1D). However, WASp-deficient Treg cells still showed an approximately 4-fold reduction in suppressor activity when compared to WT Treg cells. Thus, although WASp-deficient mice develop CD4⁺CD25⁺Fox3P⁺ Tregs, these cells exhibit impaired suppressor activity in vitro.

CD4⁺CD25⁺FOXP3⁺ T cells are present with the same frequency in healthy controls and WAS patients

To determine whether Treg cell function was similarly affected in WAS, we examined PBMC from 5 WAS patients (Table 1) and 5 healthy control subjects. Staining with anti-CD4 and anti-CD25, and anti-human FOXP3 again demonstrated similar frequencies of CD4⁺CD25⁺, CD4⁺FOXP3⁺ or CD25⁺FOXP3⁺ double positive cells in each group (Fig. 2A and Supplementary Fig. 1).

Table 1.

Patient Characteristics

Subject	Age (years)	Mutation	WASP expression	Lymphocytes/NL (×109/L)	Autoimmunity signs/symptoms	Score*
WAS6	24	C290T/ Arg66Cys	Reduced	1.74/0.46-4.7a	Proteinuria	3
WAS10	23	1305insG/fs stop aa494	Absent	0.88/0.46-4.7a	Relapsing polychondritis	5
WAS12	27	G389A/ Gly119Arg	Reduced	1.89/0.46-4.7a	Positive LA Arthralgias	2
WAS23	2	G291A/ Arg66His	Reduced	3.72/1.7-6.9b	None	2-3
WAS34	4	C995T/ Arg321Stop	Absent	1.74/1.7-6.9b	None	3

NL: normal limits for age (a: adult reference values from NIH Clinical Center Department of Laboratory Medicine; b: pediatric reference values from J. Pediatr. 1997, 130:388)

LA: lupus anticoagulant

^*as described in Zhu et al.6

Figure 2. Phenotype of CD4⁺CD25^hi Treg cells in healthy controls and WAS patients.

(A) Prevalence of CD4⁺FOXP3⁺ or CD25⁺FOXP3⁺ cells in from healthy control (Ctrl) and WAS CD14^-CD32^- lymphocytes. Quadrant markers were set based on staining with the isotype matched control mAbs and percentage of cells in each quadrant are indicated. Data shown are representative of 5 different experiments. (B) FOXP3 expression in healthy control and WAS CD4⁺CD25^hi Tregs were characterized by intracellular staining with anti-human FOXP3-APC antibody. Open histograms represent isotype controls; solid histograms represent FOXP3-specific staining.

To further characterize these cells, WAS CD4⁺CD25^hi “regulatory”, CD4⁺CD25^lo “activated” and CD4⁺CD25^- “target” cell populations were isolated by cell sorting as previously reported (22). Sorted CD4⁺CD25^hi cells represented 0.8-1.5% of the total CD4⁺CD25⁺ T cells. The majority of these cells expressed high levels of FOXP3 protein (Fig. 2B), whereas CD4⁺CD25^lo and CD4⁺CD25^- cells lacked FOXP3 expression (data not shown). CD4⁺CD25^hi cells showed similar scatter characteristics to CD4⁺CD25^- target cells, had minimal proliferation in response to either anti-CD3 plus anti-CD28 or PHA stimulation and lacked IL-2 production after stimulation (Fig. 3 and data not shown). In contrast, CD4⁺CD25^lo T cells were larger and both readily proliferated and produced IL-2 after stimulation (data not shown). Because in humans, FOXP3 can be transiently expressed in some activated CD4⁺ cells (27), we examined CD127 expression on sorted populations, which is inversely correlated with Treg function (28, 29). Importantly, the majority of the cells in the top ∼1% of CD4⁺CD25⁺ cells (our CD4⁺CD25^hi population) were CD127^lo (Supplementary Fig. 2). In contrast, sorted CD4⁺CD25^lo cells expressed high levels of CD127, as expected from activated cells. Together, these findings are consistent with the conclusion that the majority of the CD4⁺CD25^hi cells isolated from both WAS patients and healthy controls represent bona fide Treg cells.

Figure 3. Treg suppression assays after stimulation with anti-CD3 plus anti-CD28 antibodies.

(A) Treg suppression activity on autologous target cells. Control (Ctrl) Treg cells and Ctrl target cells were plated alone or co-cultured in a 1:1 ratio and stimulated with anti-CD3 + anti-CD28 in the presence of accessory cells. Similar assays were performed with WAS Treg and WAS target cells. Cell proliferation assessed as incorporation of [³H]-thymidine at the end of 5 days of culture is shown as counts per minute (CPM). Data are the mean ± SD of 5 independent experiments using samples from 5 different healthy controls and the 5 WAS patients described in Supplementary Table 1. (B) Ctrl Treg cells, WAS Treg cells, and Ctrl target cells were plated alone and stimulated with anti-CD3 + anti-CD28 in the presence of accessory cells. Ctrl Treg and WAS Treg were also co-cultured with Ctrl target cells in a 1:1 ratio and stimulated as above. Cell proliferation was assessed as incorporation of [³H]-thymidine at the end of 5 days of culture and is shown as counts per minute (CPM). Data shown are mean values of duplicate cultures.

WAS Treg cells exhibit impaired inhibitory function in vitro

To examine the suppression capability of Treg cells, CD4⁺CD25^hi cells and CD4⁺CD25^- target cells, were stimulated with anti-CD3 and anti-CD28 alone or mixed with each other in a 1:1 ratio in the presence of allogeneic accessory cells. CD4⁺CD25^hi Treg cells isolated from control subjects were anergic and suppressed proliferation when co-cultured with CD4⁺CD25^- target cells (Fig. 3). WAS CD4⁺CD25^hi Treg cells were also anergic to stimulation with either anti-CD3 and anti-CD28 or PHA (Fig. 3 and Supplementary Fig. 3). However, consistent with the proliferative defects of WASp-deficient T cells (30), these experimental conditions did not induce significant proliferation of WAS CD4⁺CD25^- cells (Fig. 3A), thus precluding the assessment of inhibitory effects of WAS CD4⁺CD25^hi Treg cells on autologous target cells.

To bypass these problems, we evaluated the inhibitory effects of WAS Treg cells on CD4⁺CD25^- target cells from allogeneic healthy controls. Treg cells from 4 of the 5 WAS patients studied showed decreased ability to suppress proliferation of control target cells. In two cases (WAS6, WAS23), the addition of WAS Treg cells to allogeneic target cells resulted in a paradoxical enhancement of cell proliferation, as previously observed in other disorders associated with Treg cell dysfunction (31, 32). Thus, in the majority of cases, CD4⁺CD25⁺ Tegs from patients with WAS also show impaired suppressor function.

DISCUSSION

Naturally occurring CD4⁺CD25⁺FOXP3⁺ Treg cells are regarded as major players in mediating peripheral tolerance to autoantigens, and thus, can prevent the onset of autoimmune diseases. In humans, this is substantiated by the high incidence of autoimmunity in patients with genetic loss of FOXP3 and the associated immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) (17-19). In addition, Treg defects have also been reported in patients affected with common autoimmune disorders, including rheumatoid arthritis, multiple sclerosis, psoriasis, and type 1 diabetes (32-35). Our data demonstrates that CD4⁺CD25⁺ Treg cells from WASp-deficient mice have marked defects in suppressor function in vitro. Interestingly, the defects we observe in WASp-deficient cells are somewhat more severe than those described in the very recent report by Humbalt-Baron and colleagues (36). Moreover, our work provides evidence that the same defect is present in patients with WAS. Indeed, CD4⁺CD25⁺ Tregs from four out of five WAS patients studied also had impaired capability to suppress in vitro proliferation of target T cells. Together, these data raise the possibility that abnormal Treg cell function may contribute to the pathogenesis of autoimmune complications associated with WAS.

It has recently been shown that, unlike murine cells, activated human CD4⁺CD25^- T cells can transiently express FOXP3. Whether or not these activated FOXP3⁺ T cells have suppressor activity is a current area of debate (27, 37). It has also been recently reported that expression of the IL-7 receptor α chain, CD127, can discriminate between activated CD25⁺CD127⁺ T cells and CD25⁺CD127^- regulatory T cells (28, 29). Although we have not sorted for a CD127-negative population in CD4⁺CD25^hiFOXP3⁺ T cells from WAS patients and controls (due to the limiting numbers of CD4⁺CD25⁺ cells available), we have verified that our stringent selection for CD4⁺CD25^hi cells markedly decreases the fraction of CD127⁺ cells among sorted populations. Moreover, sorted cells from healthy controls and WAS patients contain equal fractions of CD127⁺ cells. In addition, WAS patients and controls in this study showed similar frequencies of CD4⁺DR⁺ cells (Supplementary Fig. 1), indicating that the WAS patients examined did not have increased numbers of activated CD4⁺ lymphocytes. Finally, sorted Treg cells from WAS patients and healthy controls failed to proliferate when stimulated with anti-CD3 and anti-CD28 (Fig. 3) or PHA, which can induce some proliferation of WAS T cells (Supplementary Fig. 2), thus exhibiting anergic characteristics of true Treg cells.

The exception of normal Treg suppression activity by cells from patient WAS34, however, indicates that defective Treg function is not consistent in all WAS patients. It is possible that Treg activity may decline with time, following the reported age-dependent attrition of the T lymphocyte compartment in WAS (38, 39). However, it can also be hypothesized that the defect of Treg function in WAS is not complete and may be modified by other factors. Indeed, we have seen that in vitro suppression can be partially rescued by pre-activation of murine Treg cells with IL-2. However, our observations of marked defects in suppression in Treg cells from WASp-deficient mice, which completely lack protein expression and are more genetically homogeneous, strongly support that WASp is required for full Treg suppressor function.

Although we do not yet understand the mechanism(s) responsible for the defective function of WAS Treg cells, it is likely that multiple factors may contribute. Treg cell suppression function depends on TCR stimulation (40) and requires direct cell-cell interaction (41). WASp deficiency is associated with defective TCR-mediated activation and impaired formation of the immunological synapse (6-9, 42-44), both of which may affect Treg cell suppression activity. Interestingly, the partial rescue of in vitro suppression defects by pre-activation of WASp-deficient Treg cells with IL-2, suggests that optimization of T cell activation may overcome the defect of WAS Treg cells. However, other factors may also contribute to impaired negative regulatory mechanisms in WAS, including the documented defect in T cell receptor downregulation (9), and decreased and delayed expression of CTLA-4 (45), an important T cell inhibitory receptor. We have further found that WAS-deficient murine CD4⁺ T cells show impaired TCR-dependent restimulation-induced cell death (Nikolov et al., PLS, FC, RMS, manuscript submitted). The relative contribution of these potential mechanisms to the breakdown of peripheral tolerance in WAS patients remains to be established. Nevertheless, the demonstration of impaired suppressor function in Tregs from both WASp-deficient mice and WAS patients suggests that defective regulatory T cell function may be an important factor contributing to the immune dysregulation in WAS and suggests avenues for exploration of novel and improved forms of treatment for autoimmune complications of this disease.

Supplementary Material

01

Supplementary Figure 1

Analysis of lymphocytes from WAS patients and healthy control subjects. Isolated PBMC were stained with anti-CD4 and anti-CD25 or anti-DR monoclonal antibody as described in the Materials and Methods section. Percentage of CD4/CD25 and CD4/DR double positive cells were determined by setting quadrant markers on samples stained with isotype controls antibodies. No statistical significant differences were observed between the two groups (Student t-test).

Supplementary Figure 2

Isolated PBMC were stained with anti-CD4 and anti-CD25 and anti-CD127 monoclonal antibodies. Cells were gated on the CD25^hi or CD25^lo cells and CD127 expression levels were compared by histogram. Percentages of CD127^hi cells in the CD4⁺CD25^hi populations are shown.

Supplementary Figure 3

Proliferation of WAS patient cells in response to PHA. While WAS CD4⁺CD25^- cells exhibit proliferation in response to PHA, WAS CD4⁺CD25⁺ Treg cells fail to proliferate, consistent with their anergic phenotype.