IL-2 receptor (IL-2R) signaling is essential for optimal stability and function of CD4+CD25hiFOXP3+ regulatory T cells (Treg); a cell type that plays an integral role in maintaining tolerance. Thus, we hypothesized that decreased response to IL-2 may be a common phenotype of subjects who have autoimmune diseases associated with variants in the IL2RA locus, including T1D and MS, particularly in cells expressing the high affinity IL-2R alpha chain (IL-2RA or CD25). To examine this question we used phosphorylation of STAT5 (pSTAT5) as a downstream measure of IL-2R signaling, and found a decreased response to IL-2 in CD4+CD25hi T cells of T1D and MS, but not SLE patients. Since the IL2RArs2104286 haplotype is associated with T1D and MS, we measured pSTAT5 in controls carrying the rs2104286 risk haplotype to test whether this variant contributed to reduced IL-2 responsiveness. Consistent with this, we found decreased pSTAT5 in subjects carrying the rs2104286 risk haplotype. Reduced IL-2R signaling did not result from lower CD25 expression on CD25hi cells; instead we detected increased CD25 expression on naive Treg from controls carrying the rs2104286 risk haplotype, and subjects with T1D and MS. However the rs2104286 risk haplotype correlated with increased soluble IL-2RA levels, suggesting that shedding of the IL-2R may account in part for the reduced IL-2R signaling associated with the rs2104286 risk haplotype. In addition to risk variants in IL2RA, we found that the T1D-associated risk variant of PTPN2rs1893217 independently contributed to diminished IL-2R signaling. However, even when holding genotype constant at IL2RA and PTPN2, we still observed a significant signaling defect in T1D and MS patients. Together, these data suggest that multiple mechanisms converge in disease leading to decreased response to IL-2, a phenotype that may eventually lead to loss of tolerance and autoimmunity.
Citation: Cerosaletti K, Schneider A, Schwedhelm K, Frank I, Tatum M, Wei S, et al. (2013) Multiple Autoimmune-Associated Variants Confer Decreased IL-2R Signaling in CD4+CD25hi T Cells of Type 1 Diabetic and Multiple Sclerosis Patients. PLoS ONE 8(12): e83811. https://doi.org/10.1371/journal.pone.0083811
Editor: Steven Jacobson, National Institutes of Health, United States of America
Received: April 11, 2013; Accepted: November 9, 2013; Published: December 23, 2013
Copyright: © 2013 Cerosaletti et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by Juvenile Diabetes Research Foundation (www.JDRF.org) grants including a Career Development Grant (2-2012-205)to SAL, BRI JDRF Center grant and JDRF (33-2008-398) to JHB. An NIH (www.niaid.org) grant (AI101990-01)to JHB also funded this work. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The cytokine IL-2 is essential for T cell homeostasis. Activated effector T cells (Teff) produce IL-2 and transiently up-regulate the high affinity IL-2RA (CD25) upon activation, enabling them to respond optimally to IL-2 following antigen encounter. Expression of CD25 on Teff is then reduced through negative feed-back loops as the cells come to rest. Enzymes can also cleave CD25 from the surface of T cells upon activation resulting in soluble IL-2RA (sIL-2RA) that can be detected in the serum –. In contrast to Teff, regulatory T cells (Treg) which are essential for suppressing autoimmunity do not produce IL-2 themselves, but paradoxically are highly dependent on IL-2 for their survival and function . Treg constitutively express high levels of CD25 and therefore are highly sensitive to even low doses of IL-2. IL-2R signaling in both Teff and Treg is mediated by a complex of CD25, the IL-2R beta chain (IL-2RB or CD122) and the common gamma chain. Alternatively, the IL-2RB and common gamma chain can serve as a low affinity IL-2R. Upon cytokine binding the IL-2R, a series of sequential phosphorylation events are initiated from the beta and gamma chains including phosphorylation of JAK1, JAK3 and Shc proteins resulting in transcriptional activation of cytokine-targeted genes, including the STAT5-dependent Treg transcription factor FOXP3 and CD25 itself .
Impairment of IL-2/IL-2R signaling has striking consequences on the development of tolerance to self-antigens. This is demonstrated most dramatically in knock-out mice in which a deficiency of Il2, Il2ra, or Il2rb leads to early death due to severe autoimmunity –. In humans, deficiency of IL-2RA can result in autoimmunity , . Consistent with this, we have observed decreased response to IL-2 in CD25+ and memory T cells of T1D subjects . However, it is not clear whether this deficit is common to all autoimmune settings.
The importance of the IL-2R pathway in maintaining tolerance is also revealed by genome wide association studies (GWAS) that have associated multiple variants in the IL-2/IL-2R signaling pathway with susceptibility for type 1 diabetes (T1D), multiple sclerosis (MS), rheumatoid arthritis, Crohn’s disease, Grave’s disease, generalized vitiligo, and alopecia areata. Specifically, autoimmune variants have been identified in the IL2RA gene –, as well as in IL2 itself , , , IL2RB , and the protein tyrosine phosphatase N2 (PTPN2), , a phosphatase in multiple signaling pathways including the IL-2R signaling pathway –. Amongst these genes, IL2RA is unique to the IL-2R pathway, whereas the other genes function in other cytokine pathways in addition to IL-2.
Genetic association of the IL2RA locus with autoimmunity is complex. Three haplotypes that are tagged by three single nucleotide polymorphisms (SNPs) have been defined in the IL2RA locus: the rs12722495 (previously rs41295061) protective haplotype that is only associated with T1D (OR = 0.62, p = 6.43×10−25), the rs2104286 protective haplotype that is associated with MS (OR = 0.85, p = 6.27×10−7), T1D (OR = 0.80, p = 1.27×10−13) and RA (OR = 0.76, p = 4.9×10−5), and the rs11594656 haplotype that is associated with protection from T1D (OR = 0.87, p = 3.37×10−6) but risk for MS (OR = 1.17, p = 7.67×10−4), , . Other SNPs in the IL2RA locus are associated with Crohn’s disease, Grave’s disease, vitiligo, and alopecia areata , , , , . All of the associated variants are non-coding, falling upstream of the IL2RA gene or in intron 1, and despite being in high linkage disequilibrium (D’ = 1.0); many of these SNPs have low R2 values.
Understanding if altered IL-2R signaling is a phenotype common to multiple autoimmune diseases and how this relates to the genetic heterogeneity at the IL2RA locus is important for elucidating mechanisms common or unique to associated autoimmune diseases. Here, we report reduced IL-2 responsiveness in T1D and MS, and define multiple phenotypes that correlate with the IL2RArs2104286 risk haplotype that is associated with these diseases. We also demonstrate that additional risk alleles in the IL-2R signaling pathway contribute independently to decreased IL-2 responsiveness, collectively impacting tolerance.
Materials and Methods
PBMC were derived from subjects participating in studies under the auspices of the BRI-JDRF Center for Translational Research registry. This study was approved by the Benaroya Research Institute IRB. Written informed consent was obtained from all subjects according to IRB approved protocols at Benaroya Research Institute, Seattle. Subject demographics, genotype and clinical characteristics are described in Table 1. Control participants were selected based on lack of personal or family history of autoimmunity or asthma. Samples for this study were obtained from adult control subjects and subjects diagnosed with T1D, MS and SLE. For some experiments, samples were provided by the Genotype and Phenotype Registry, a service of the Tissue Donation Program at The Feinstein Institute for Medical Research, Manhasset, New York, USA. Samples for all experiments were provided to the investigator by the BRI clinical core in a blinded manner. This enables prospective selection of subjects for individual experiments based on demographic, genetic and clinical characteristics without the experimenter’s prior knowledge of disease state, genotype or subject information. Control and disease subjects or risk and non-risk subjects were included in each batch received from the clinical core to avoid batch effects on experimental data.
IL2RArs12722495, IL2RArs2104286, IL2RArs11594656, and PTPN2rs1893217 were genotyped using fluorescently labeled MGB-Eclipse system (Epoch Biosciences). The genotyping assay was performed using 10ng of genomic DNA, 0.38U JumpStart™ Taq DNA polymerase (Sigma-Aldrich), primers and probes in a 5 µl reaction volume according to the manufacturer’s protocol. PCR was performed and analyzed on an ABI HT7900.The IL2RArs2104286 haplotype allows variation at rs2104286 while holding rs12722495 constant for the A/A genotype and rs11594656 constant for the T/T genotype as described by Dendrou et.al .
Flow cytometric analysis for phosphorylated STAT5
BD Phosphoflow staining was performed as per manufacturer’s instructions. In brief, cells were activated with different concentrations of IL-2 or IL-15 for 10 and/or 20min, fixed with Phosflow Buffer I and permeabilized using BD Phosflow Buffer III prior to staining with Alexa647 anti-pSTAT5(Y694), PerCP CD4 and PE CD25 (clone M-A251). For some experiments, cells were co-stained with FITC anti-CD45RO. Initial experiments were performed using freshly isolated PBMC. As shown previously, similar results were obtained with thawed PBMC. All pSTAT5 data shown in this manuscript are from previously frozen PBMC. Data were acquired using a FACS Calibur and analyzed using FloJo software. Since CD4+CD25hi T cells are known to be sensitive to the freeze/thaw process and fix/perm staining protocols, samples where less than 1% of CD25hi T cells or <500 events were detected were excluded from results. pSTAT5(Y694) mean fluorescence intensity (MFI) data were normalized between experiments by determining the MFI fold increase (geometric MFI of the positive population ÷ geometric MFI of the negative control) as described previously 
Flow cytometric analysis for surface and nuclear markers
Thawed PBMC were surface stained with V500 CD4 and PE CD25 prior to fixation and permeabilization using BioLegend FOXP3 Buffer sets. Permeabilized cells were then stained with Alexa647 anti-FOXP3 and Alexa488 anti-helios antibodies. Invitrogen 8 peak beads were used to normalize flow cytometry settings between experiments by adjusting voltage settings to reach a standard MFI.
Based on preliminary data, we performed all experiments using the number of samples we calculated were required to achieve 80% power to detect a 10% difference between groups. The 10% difference was selected as a biologically relevant difference based on published work ,  and a value that was greater than twice the % CV for repeat measures on the same sample. For observed data reported in this manuscript, all experiments with multiple group comparisons were first analyzed using a Kruskal-Wallis non-parametric ANOVA test to determine whether any group differed from the others. Individual pairings were analyzed using a Mann-Whitney test to determine significance. Results of Mann-Whitney testing including 95% confidence intervals for the difference between the medians are reported in Table S1. False discovery rate was determined using the Benjamini Hochberg procedure. Permutation testing was performed on all pair-wise comparisons. Outliers were identified using both control charts and the Grubb’s test. All outliers are noted in the Figure Legends. For analysis of multiple variables, linear regression was performed as noted in the figure legends. Interaction of IL2RA and PTPN2 genotype on IL-2 responsiveness was assessed using PLINK (http://pngu.mgh.harvard.edu/~purcell/plink/) with an additive SNP genotype model for IL2RA genotype using the “linear” command, including PTPN2 genotype as a covariate with and without the “interaction” command. Association testing was performed using a Chi-square test and interaction of variants was modeled using the R software package. Comparisons required a p-value of <0.05 for the data to be considered significantly different.
T1D and MS, but not SLE subjects display decreased response to IL-2 in CD4+CD25hi T cells
We have previously observed reduced response to IL-2 in T cells from subjects with T1D . One downstream functional measure of IL-2/IL-2R signaling is phosphorylation of STAT5 (pSTAT5) in response to IL-2 stimulation. To determine if reduced IL-2 responsiveness is a common feature of autoimmunity, we measured pSTAT5 in response to IL-2 in CD4+CD25hi T cells of T1D, MS and SLE patients. Both T1D and MS subjects displayed significantly reduced response to IL-2 in the CD25hi populations, whereas SLE subjects did not differ from controls (Figure 1A). Consistent with previous studies, we found decreased response to IL-2 in the CD25lo population of T1D subjects ( and Figure 1B), although we did not find this same decrease in the CD25lo cells of MS or SLE patients. By ANOVA, age, gender, race, disease status or immune treatment did not significantly contribute to differences in the percentage of pSTAT5 for any cohort. Thus, a common feature of T1D and MS is a reduced response to IL-2 in CD4+CD25hi T cells.
PBMC were thawed, rested and stimulated with media alone or IL-2 (100 IU/ml) prior to fixation, permeabilization and staining with CD4, CD25, CD45RO and pSTAT5. (A) The frequency of pSTAT5+ cells of CD25hi cells was determined by comparing media to 25 IU IL-2 stimulation for 20min. (B) In an independent cohort, the frequency of pSTAT5+ cells of CD25lo cells was determined by comparing media to 100 IU IL-2 stimulation for 10min. In all experiments, patients and controls were prospectively matched with controls for age and ethnicity. Groups in (A) and (B) differed by multiple parameter testing using a Kruskal-Wallis test with p-values of 0.0008 and 0.0198, respectively. Statistical significance for individual pairings using a Mann-Whitney test are shown. Differences shown in (A) were verified with permutation testing where control vs T1D and control vs MS remained significant (<0.001). For (B), after permutation testing the control vs T1D (0.005), T1D vs MS (0.03) and T1D vs SLE (0.01) pairings remained significant. There was one outlier in the T1D cohort in (B). By ANOVA, age, gender, race, disease status or immune treatment did not significantly contribute to differences in % pSTAT5 for any cohort (therapy and disease status analyses are shown in Figure S3). The mean pSTAT5 response is indicated for each group. IL2RArs2104286 genotype is noted by solid black (A/A), solid grey (A/G) and open (G/G) symbols.
The IL2RArs2104286 risk haplotype correlates with decreased IL-2R signaling in CD25hi T cells from control subjects
The IL2RArs2104286 risk haplotype is associated with T1D and MS , , , , but not SLE , , reflecting the response to IL-2 we observed in Figure 1. Thus, we asked whether genetic susceptibility contributes to this disease-specific phenotype. The observed reduced response to IL-2 in the CD25hi cells of the MS and T1D cohorts, was not due to the percentage of subjects carrying the risk allele (control 59%, T1D 50% MS 64% A/A genotype: p = 0.1393, chi-square test Table I). There was also no significant difference in IL-2 response within each disease group when stratified by rs2104286 haplotype (T1D p = 0.2122; MS p = 0.1209).
In order to discern the genetic contribution of rs2104286 in the absence of disease effects, we assessed pSTAT5 in response to IL-2 in healthy control subjects genotyped for the IL2RArs2104286 haplotype. Variation in the rs2104286 haplotype was determined by holding the rs12722495 allele constant at A/A and the rs11594656 allele constant at T/T while looking at variation at rs2104286 as described by Dendrou et.al. . The frequency and MFI of pSTAT5 in CD4+CD25hi T cells (Figure 2A) was significantly lower in control subjects homozygous for the IL2RArs2104286 risk haplotype compared to subjects carrying the A/G genotype at rs2104286. Similar results were found in a smaller but independent cohort (Figure S1).This decrease was also observed in CD45RO+ memory T cells, a subset containing the majority of CD25hi cells, but not in naïve T cells (data not shown). We next determined if this phenotype was specific for expression of CD25, since other cytokine receptors use the gamma chain, including the IL-15R which utilizes both the common gamma chain and IL-2Rβ, but has a unique IL-15RA chain that confers specificity . Although there was a trend, a significant decrease in pSTAT5 was not observed in CD25lo cells stimulated with IL-2 or IL-15 (Figure 2B), cell subsets and stimuli that do not depend on CD25 expression. Thus, responses to IL-2 are decreased in CD25hi T cells of control subjects homozygous for the IL2RArs2104286 risk haplotype and this defect was only observed in cells expressing the IL-2RA chain
PBMC were thawed and stained as in Figure 1. (A) The frequency and MFI fold increase of pSTAT5+ cells of CD25+ gated cells was determined by comparing media to IL-2 stimulated genotyped control samples. (B) The frequency of pSTAT5+ cells of CD25lo was determined following stimulation of genotyped controls with IL-2 or IL-15 (200pg/ml) for 10min. The mean pSTAT5 response is indicated for each group and subjects with IL2RArs2104286 genotypes G/G (▴), A/G (○) and A/A (▪) are differentiated by symbol shapes. Statistical significance was determined using a Mann Whitney test. All controls tested were Caucasian with no significant difference in age or gender between genotypes. *denotes the risk haplotype determined by looking at variation at the rs21042856 allele while holding the rs12722495 allele constant at T/T and the rs11594656 allele constant at T/T.
Subjects homozygous for the IL2RArs2104286 risk haplotype with reduced IL-2 responsiveness do not display decreased CD25 expression on T cells
Decreased response to IL-2 may be caused by lower expression of the IL-2R. Thus, we measured the surface expression of CD25 by flow cytometry on different CD4+ T cell subsets including naïve and memory subsets of Treg and Teff cells as determined by CD45RA and FOXP3 expression (Figure 3A). Interestingly, we did not detect decreased CD25 expression on naïve or memory Teff despite these populations displaying decreased IL-2 responsiveness (Figure 3B). Instead, we found a significant increase in CD25 expression on naïve Treg from subjects homozygous for the rs2104286 risk haplotype (Figure 3C). Moreover, this phenotype was also observed in T1D and MS subjects (Figure 3D). The observed increase in CD25 expression in the naïve Treg population of MS and T1D patients was not due to the percentage of subjects carrying the risk allele and there was no significant difference in CD25 MFI within each disease group when stratified by rs2104286 haplotype. Thus, the decreased response to IL-2 in CD25hi cells observed with the rs2104286 risk haplotype in controls, as well as in T cells from subjects with T1D and MS, is not solely a result of reduced surface expression of IL-2RA.
PBMC were thawed and stained for CD4, CD25, FOXP3, helios and CD45RA as described in materials and methods. (A) Representative gating of CD4+ T cells on FOXP3− Teff and FOXP3+ Treg. CD25 MFI was measured on CD45RA+ naïve and CD45RA− memory Teff (B) and Treg (C) populations of controls and (D) autoimmune subjects. The mean pSTAT5 response is indicated for each group. For (B) and (C), all controls tested were Caucasian with no significant difference in age or gender between IL2RArs2104286 genotypes. G/G (▴), A/G (○) and A/A (▪) subjects are differentiated by symbol shapes. In (D), patients and controls were prospectively matched with controls for age and ethnicity. In (B-C), statistical significance between pairings was determined using a student’s t- test after assessing normality of the data as determined using the Shapiro-Wilk test. P-values were then adjusted using the Benjamini Hochberg false discovery rate procedure. Adjusted p values where significant are shown. Groups in (D) differed by multiple parameter testing using a Kruskal-Wallis test (0.0085). Statistical significance for individual pairings using a Mann-Whitney test is shown. Differences were verified with permutation testing where control vs T1D (0.015) and control vs MS (0.01) remained significant. There were two outliers in the MS cohort. *denotes the risk haplotype determined by looking at variation at the IL2RArs21042856 allele while holding the rs12722495 allele constant at T/T and the rs11594656 allele constant at T/T. By ANOVA, age, gender, race, disease status or immune treatment did not significantly contribute to differences in CD25 MFI for any cohort. IL2RArs2104286 genotype is noted by solid black (A/A), solid grey (A/G) and open (G/G) symbols.
Decreased response to IL-2 in CD25hi T cells inversely correlates with soluble IL-2RA levels in the serum of control subjects homozygous for the IL2RArs2104286 risk haplotype
Previous studies have shown that increased sIL-2RA in the serum is associated with the IL2RArs2104286 risk haplotype in healthy controls , , . To better understand whether there is a relationship between response to IL-2 and the level of sIL-2RA shed from the cell surface, we correlated IL-2R signaling and sIL-2RA levels using serum and PBMC from the same blood draw of genotyped controls. When compared with the response to IL-2, we found an inverse correlation between serum sIL-2RA and response to IL-2 in the CD25hi population (Figure 4). A mechanism which could explain the inverse relationship between sIL-2RA and pSTAT5 in response to IL-2 is the blockade of IL-2 binding to the IL-2R by sIL-2RA shed in the media. However, in our in vitro system this is unlikely since T cells are purified by ficoll gradient centrifugation and are subsequently frozen, but to further test this we washed cells prior to adding IL-2 (thereby eliminating the possible impact of soluble IL-2RA) or incubated overnight (allowing for accumulations of soluble IL-2RA). In both cases we found the response to IL-2 stayed the same as our initial measurement (data not shown). Thus, these data reveal an inverse relationship between response to IL-2 and sIL-2RA, but the mechanisms by which this alters IL-2 signaling are still unclear.
PBMC and serum samples from the same blood draw were assayed for response to IL-2 as in Figure 1 and serum sIL-2RA using Luminex. All controls tested were Caucasian with no significant difference in age or gender between genotypes. Multivariable analysis was performed using linear regression and is shown in the Figure. To confirm these results we performed several additional tests. Random sampling of 80% of the data 1000 times resulted in a Pearson correlation of −0.516 and a p-value of 0.0015.Significance was also tested using a non-parametric correlation (Spearman test = −0.581, p-value of 0.0003) verifying that outliers are not the reason for statistical significance.
PTPN2 and IL2RA independently contribute to reduced response to IL-2 in controls
We have previously shown that the rs1893217 SNP in PTPN2, a gene associated with T1D, RA and CD but not MS, correlates with decreased response to IL-2 in CD25hi and memory CD25lo cells . To determine whether multiple genetic factors could contribute to the decreased response to IL-2 in CD25hi T cells, we stratified control subjects by both the PTPN2rs1893217 genotype and the IL2RArs2104286 haplotype and tested for interaction of these genotypes by linear regression using the PLINK toolset. A significant decrease in pSTAT5 was associated with the IL2RArs2104286 risk haplotype group independent of PTPN2rs1893217 genotype using an additive model with age and gender as covariates (p = 0.028, Figure 5). Likewise, a significant decrease in IL-2 responsiveness was associated with PTPN2rs1893217 risk genotype independent of IL2RArs2104286 haplotype (p = 0.0018), however there was no significant evidence for interaction of IL2RArs2104286 and PTPN2rs1893217 genotypes (p = 0.7688). Thus, both the IL2RArs2104286 risk haplotype and the PTPN2rs1893217 risk allele independently contribute to reduced IL-2R signaling when the risk genotypes are present together.
PBMC were stained as in Figure 1. The frequency of pSTAT5 of CD25hi T cells in control subjects were stratified by PTPN2rs1893217 and IL2RArs2104286 haplotype. The mean pSTAT5 response is indicated for each group. Statistical significance between rs2104286 haplotypes within groups defined by PTPN2 genotype was determined using a Mann Whitney test. When analyzed as covariates in PLINK, age and gender did not significantly contribute to this phenotype (p = 0.4459 and p = 0.4954, respectively) and there was no evidence for interaction between IL2RA and PTPN2 genotype (p = 0.7688). *denotes the risk haplotype determined by looking at variation at the rs21042856 allele while holding the rs12722495 allele constant at T/T and the rs11594656 allele constant at T/T.
PTPN2 and CD25 risk alleles are not the only factors contributing to decreased response to IL-2 in T1D and MS patients
To determine whether the IL2RArs2104286 risk haplotype and the PTPN2rs1893217 risk alleles together could explain the reduced response to IL-2 in T1D and MS patients relative to control subjects, we held the genotype at IL2RA and PTPN2 constant for the protective alleles and compared responses in disease patients to age and ethnicity matched controls (Figure 6A and B). As in previous analyses, all subjects carried the common risk allele at IL2RArs12722495, a SNP previously shown to display decrease CD25 expression and IL2R signaling , . Even when accounting for IL2RA and PTPN2 genotype, we still find a significant decrease in the response to IL-2 in the CD25hi T cell populations of T1D and MS subjects. In addition, we used the model of y = α + β1xIL2RA +β 2xPTPN2 +β 3xIL2RA*PTPN2 + ε to assess the interaction of each variant in the control, T1D and MS cohorts shown in Figure 1. Although certain combinations of SNP interactions could not be tested based on our sample size, we found no interaction between variants that were present in the disease cohorts (controls p = 0.32; T1D p = 0.89; MS p = 0.19 by ANOVA). Moreover, we found no significant impact on percentage of pSTAT5 when disease status (p = 0.98) and treatment (0.40) were included as variables in the analysis for the MS cohort. Thus, other factors in addition to IL2RA and PTPN2 genotype likely contribute to reduced response to IL-2 in these two autoimmune diseases.
(A) T1D and (B) MS subjects carrying the protective IL2RArs2104286 haplotype and the protective PTPN2rs1893217 SNP were compared to genotype, age and ethnicity matched controls. T1D subjects and matching controls were stimulated with 100 IU/ml IL-2 for 10min while MS samples and matching controls were stimulated with 25 IU/ml IL-2 for 20min. The mean pSTAT5 response is indicated for each group. Statistical significance was determined using a Mann Whitney test. *denotes the risk haplotype determined by looking at variation at the rs21042856 allele while holding the rs12722495 allele constant at T/T and the rs11594656 allele constant at T/T.
Since IL-2 plays a key role in maintaining tolerance, it follows that patients with autoimmune diseases may display defects in the IL-2/IL-2R signaling pathway. Here, we extend our previous finding that response to IL-2 is reduced in CD4+ T cells of T1D subjects  and find a similar defect in CD4+CD25hi T cells of MS patients. Interestingly, this IL-2R signaling defect was not observed in our cohort of SLE patients. However, impaired production of IL-2 has been described in SLE . Together, these results support the hypothesis that defects in the IL-2/IL-2R pathway are present in autoimmunity. We found that the IL2RArs2104286 haplotype that is associated with T1D and MS, but not SLE, was associated with reduced IL-2R signaling in CD4+CD25hi T cells of controls. In addition, we found that the risk alleles at IL2RA and PTPN2, a second IL-2R signaling pathway gene, could act independently to decrease IL-2 responsiveness in control subjects. However, genetic variation at IL2RArs2104286 and PTPN2 does not account for all of the decreased signaling observed in patients. Thus, multiple factors including IL2RA genetic susceptibility contribute to reduced IL-2R signaling in T1D and MS patients.
Despite the similarities in the response to IL-2 in T1D and MS, there remain disease-specific differences. In this study, we found decreased signaling in response to IL-2 in CD25hi T cells of both T1D and MS patients but decreased response in CD4+CD25lo T cells of only T1D patients. Moreover, this decrease did not correlate with therapy or disease status (Figure S3). Thus, IL-2R signaling defects may be more pervasive in T1D. This is consistent with previous studies demonstrating decreased response to IL-2 and IL-15 in T1D patients  and controls carrying the T1D-associated PTPN2rs1893217 risk allele , a gene not associated with MS.
To date, cellular phenotypes linked to the IL2RA locus have been best defined for the T1D-associated IL2RArs12722495 haplotype. These occur in both CD25hi T cells and memory Teff cells and include decreased IL2RA RNA and CD25 surface expression in Treg (confirmed in Figure S2) and memory Teff , , , increased frequency of CD25+ naive T cells , decreased response to IL-2  and increased serum sIL-2RA , . In contrast, the IL2RArs2104286 risk haplotype, associated with both T1D and MS, only shares a subset of these phenotypes, including increased frequency of CD25+ naïve T cells and increased serum sIL-2RA. In this study we did not observe decreased expression of CD25 on memory T cells with the rs2104286 risk haplotype, but we did detect increased CD25 surface expression on naïve Treg, a phenotype that was mirrored in both T1D and MS patients.
The molecular mechanisms underlying the phenotypes associated with IL2RA risk variants appear to be at least two-fold, involving transcriptional regulation of the IL2RA locus and production of sIL-2RA. All associated SNPs in the IL2RA gene to date are located in the promoter region or intron 1, and differential allelic expression of IL2RA transcripts has been reported in HapMap lymphoblastoid cell lines and T1D cases/parents . However, the allele-specific expression was not associated with rs12722495, rs2104286, or rs11594656, but rather other SNPs in intron 1. Subsequent analysis of allele-specific transcription factor binding to SNPs in the IL2RA locus revealed that at least four IL2RA SNPs demonstrate differential binding . This result taken together with the low R2 values for LD between SNPS in IL2RA, suggests that CD25 expression is the combined result of allele-specific transcription factor binding to the haplotype of alleles across the IL2RA promoter and intron 1in any particular individual, and may explain some of the genetic and phenotypic heterogeneity observed with this locus. Added to this composite expression picture are cell extrinsic factors, such as the production of matrix metalloproteases like MMP-9 which can cleave IL-2RA from the cell surface  and glycosylation of cytokine receptors that further impacts stability and function of the receptor . However, production of sIL-2RA also seems to have a genetic component because it is associated with IL2RA SNPs in control subjects, including rs2104286 as we observed here, and is stable over time , . The molecular basis for this genetic association is not clear, since no transcript encoding a sIL2RA isoform has been detected to date.
While increased levels of serum sIL-2RA are present in several autoimmune and inflammatory states and are associated with IL2RA genotype , , , the functional consequences of increased sIL-2RA are not well understood. Here, we find that increased sIL-2RA correlates with decreased response to IL-2. Taken together with others’ findings suggesting that sIL-2RA reduces IL-2 availability both in vitro and in vivo , , , one may conclude that sIL-2RA is functioning as a decoy thereby reducing the amount of IL-2 capable of binding the surface IL-2R. Yet, subjects with high sIL-2RA still displayed reduced pSTAT5 when cells were washed (data not shown), thereby eliminating the likelihood of sIL-2RA functioning as a decoy under these conditions.
To summarize our results and those of others, it appears that multiple mechanisms can cause reduced IL-2 signaling, contributing to autoimmune susceptibility and pathogenesis, as detailed in Table S2. Given the importance of Treg in tolerance and known defects in Treg of T1D and MS subjects , it is likely that decreased stability and function of Treg is a major mechanism by which IL2RA variants contribute risk in associated autoimmune diseases. However, other cell subsets may also be involved, particularly in T1D. Decreased response to IL-2 in Teff populations may cause impaired activation induced cell death leading to increased frequency of autoreactive Teff cells. Additionally, reduced IL-2 responsiveness has the potential to decrease induction of adaptive Treg while increasing Th17 generation . Lastly, increased CD25 expression on naïve Treg (Figure 3) or increased frequency of CD25+ naive T cells may enhance the homeostatic proliferation of these naïve T cell subsets as suggested in a recent study , . Whether this leads to senescence, reduced stability or alterations in the repertoire are currently under investigation, along with other work to more decisively link phenotypes to mechanisms of disease.
Decreased response to IL-2 in CD4+CD25+ T cells correlates with IL2RArs2104286. PBMC from the GAP registry (see materials and methods) were thawed and stained as in Figure 1. The frequency of pSTAT5+ cells of CD25hi and CD25lo gated cells was determined by comparing media to IL-2 stimulated genotyped control samples. Subjects with IL2RArs2104286 genotypes G/G (▴) and A/A (▪) are differentiated by symbol shapes. Statistical significance was determined using a Mann Whitney test. *denotes the risk haplotype determined by looking at variation at the rs21042856 allele while holding the rs12722495 allele constant at T/T and the rs11594656 allele constant at T/T.
Decreased CD25 on memory Treg when stratified by IL2RArs12722495. PBMC were thawed and stained for CD4, CD25, FOXP3, helios and CD45RA as described in materials and methods and gated as shown in Figure 3A.. CD25 MFI was measured on CD45RA+ naïve and CD45RA− memory Treg populations of controls with IL2RArs2104286 genotypes: G/G (▴), A/G (○) and A/A (▪). Statistical significance for individual pairings using a Mann-Whitney test are shown.
Therapy and disease status do not correlate with response to IL-2 in CD25hi cells of the T1D, MS or SLE cohorts. Data from Figure 1A were analyzed against clinical parameters for the (A) T1D, (B-C) MS and (D) SLE cohorts. Statistical significance was determined using linear regression in (A, C) and a Mann Whitney test in (B). The mean pSTAT5 response is indicated for each group in (B, D). IL2RArs2104286 genotype is noted by solid black (A/A), solid grey (A/G) and open (G/G) symbols where measured in the T1D and MS cohorts.
Results of Mann-Whitney association testing and 95% confidence intervals for all comparisons in the figures.
We are very grateful to the investigators and staff of the BRI-JDRF Center for Translational Research at the BRI Diabetes Clinical Research Program for subject recruitment, as well as the BRI Translational Research Clinical Core for sample processing and handling. We are also grateful for the use of samples that were provided by the Genotype and Phenotype Registry, a service of the Tissue Donation Program at The Feinstein Institute for Medical Research, Manhasset, New York, USA. Amber Burt MS provided valuable consultation for genetic interaction analysis under the auspices of the Northwest Institute of Genetic Medicine.
Conceived and designed the experiments: SAL KC JB. Performed the experiments: SAL AS KS IF MT SW. Analyzed the data: SAL KC AS EW. Contributed reagents/materials/analysis tools: CG MK. Wrote the paper: SAL KC JB.
- 1. Brusko TM, Wasserfall CH, Hulme MA, Cabrera R, Schatz D, et al. (2009) Influence of membrane CD25 stability on T lymphocyte activity: implications for immunoregulation. PLoS ONE 4: e7980 10.1371/journal.pone.0007980 [doi].
- 2. Park MJ, Shin JS, Kim YH, Hong SH, Yang SH, et al. (2011) Murine mesenchymal stem cells suppress T lymphocyte activation through IL-2 receptor alpha (CD25) cleavage by producing matrix metalloproteinases. Stem Cell Rev 7: 381–393 10.1007/s12015-010-9203-9 [doi].
- 3. Schulz O, Sewell HF, Shakib F (1998) Proteolytic cleavage of CD25, the alpha subunit of the human T cell interleukin 2 receptor, by Der p 1, a major mite allergen with cysteine protease activity. J Exp Med 187: 271–275.
- 4. Fehervari Z, Yamaguchi T, Sakaguchi S (2006) The dichotomous role of IL-2: tolerance versus immunity. Trends Immunol 27: 109–111 S1471-4906(06)00023-8 [pii];10.1016/j.it.2006.01.005 [doi].
- 5. Malek TR (2008) The biology of interleukin-2. Annu Rev Immunol 26: 453–479.
- 6. Sadlack B, Lohler J, Schorle H, Klebb G, Haber H, et al. (1995) Generalized autoimmune disease in interleukin-2-deficient mice is triggered by an uncontrolled activation and proliferation of CD4+ T cells. Eur J Immunol 25: 3053–3059.
- 7. Suzuki H, Kundig TM, Furlonger C, Wakeham A, Timms E, et al. (1995) Deregulated T cell activation and autoimmunity in mice lacking interleukin-2 receptor beta. Science 268: 1472–1476.
- 8. Willerford DM, Chen J, Ferry JA, Davidson L, Ma A, et al. (1995) Interleukin-2 receptor alpha chain regulates the size and content of the peripheral lymphoid compartment. Immunity 3: 521–530.
- 9. Aoki CA, Roifman CM, Lian ZX, Bowlus CL, Norman GL, et al. (2006) IL-2 receptor alpha deficiency and features of primary biliary cirrhosis. J Autoimmun 27: 50–53.
- 10. Strieder TG, Drexhage HA, Lam-Tse WK, Prummel MF, Tijssen JG, et al. (2006) A reduced IL2R (CD25) expression level in first and second degree female relatives of autoimmune thyroid disease patients. A sign of a poor capability to preserve tolerance? Autoimmunity 39: 93–98.
- 11. Long SA, Cerosaletti K, Bollyky PL, Tatum M, Shilling H, et al. (2010) Defects in IL-2R signaling contribute to diminished maintenance of FOXP3 expression in CD4+CD25+ regulatory T cells of type 1 diabetic subjects. Diabetes 59: 407–415.
- 12. Wellcome Trust Case Control Consortium (2007) Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447: 661–678.
- 13. International Multiple Sclerosis Genetics Consortium (2007) Risk Alleles for Multiple Sclerosis Identified by a Genomewide Study. N Engl J Med 357: 851–862.
- 14. Brand OJ, Lowe CE, Heward JM, Franklyn JA, Cooper JD, et al. (2007) Association of the interleukin-2 receptor alpha (IL-2Ralpha)/CD25 gene region with Graves' disease using a multilocus test and tag SNPs. Clin Endocrinol (Oxf) 66: 508–512.
- 15. Jin Y, Birlea SA, Fain PR, Gowan K, Riccardi SL, et al. (2010) Variant of TYR and autoimmunity susceptibility loci in generalized vitiligo. N Engl J Med 362: 1686–1697 NEJMoa0908547 [pii];10.1056/NEJMoa0908547 [doi].
- 16. Maier LM, Lowe CE, Cooper J, Downes K, Anderson DE, et al. (2009) IL2RA genetic heterogeneity in multiple sclerosis and type 1 diabetes susceptibility and soluble interleukin-2 receptor production. PLoS Genet 5: e1000322.
- 17. Franke A, McGovern DP, Barrett JC, Wang K, Radford-Smith GL, et al. (2010) Genome-wide meta-analysis increases to 71 the number of confirmed Crohn's disease susceptibility loci. Nat Genet 42: 1118–1125 ng.717 [pii];10.1038/ng.717 [doi].
- 18. Matesanz F, Caro-Maldonado A, Fedetz M, Fernandez O, Milne RL, et al. (2007) IL2RA/CD25 polymorphisms contribute to multiple sclerosis susceptibility. J Neurol 254: 682–684.
- 19. Petukhova L, Duvic M, Hordinsky M, Norris D, Price V, et al. (2010) Genome-wide association study in alopecia areata implicates both innate and adaptive immunity. Nature 466: 113–117 nature09114 [pii];10.1038/nature09114 [doi].
- 20. Todd JA, Walker NM, Cooper JD, Smyth DJ, Downes K, et al. (2007) Robust associations of four new chromosome regions from genome-wide analyses of type 1 diabetes. Nat Genet 39: 857–864.
- 21. Hulme MA, Wasserfall CH, Atkinson MA, Brusko TM (2012) Central role for interleukin-2 in type 1 diabetes. Diabetes 61: 14–22 61/1/14 [pii];10.2337/db11-1213 [doi].
- 22. Malek TR, Castro I (2010) Interleukin-2 receptor signaling: at the interface between tolerance and immunity. Immunity 33: 153–165 S1074-7613(10)00287-6 [pii];10.1016/j.immuni.2010.08.004 [doi].
- 23. Vang T, Miletic AV, Arimura Y, Tautz L, Rickert RC, et al. (2008) Protein tyrosine phosphatases in autoimmunity. Annu Rev Immunol 26: 29–55.
- 24. Lowe CE, Cooper JD, Brusko T, Walker NM, Smyth DJ, et al. (2007) Large-scale genetic fine mapping and genotype-phenotype associations implicate polymorphism in the IL2RA region in type 1 diabetes. Nat Genet 9: 1074–1082.
- 25. Dendrou CA, Plagnol V, Fung E, Yang JH, Downes K, et al. (2009) Cell-specific protein phenotypes for the autoimmune locus IL2RA using a genotype-selectable human bioresource. Nat Genet 41: 1011–1015.
- 26. Chistiakov DA, Chistiakova EI, Voronova NV, Turakulov RI, Savost'Anov KV (2011) A variant of the Il2ra / Cd25 gene predisposing to graves' disease is associated with increased levels of soluble interleukin-2 receptor. Scand J Immunol 74: 496–501 10.1111/j.1365-3083.2011.02608.x [doi].
- 27. Long SA, Cerosaletti K, Wan JY, Ho JC, Tatum M, et al. (2011) An autoimmune-associated variant in PTPN2 reveals an impairment of IL-2R signaling in CD4(+) T cells. Genes Immun 12: 116–125 gene201054 [pii];10.1038/gene.2010.54 [doi].
- 28. Carr EJ, Clatworthy MR, Lowe CE, Todd JA, Wong A, et al. (2009) Contrasting genetic association of IL2RA with SLE and ANCA-associated vasculitis. BMC Med Genet 10: 22 1471-2350-10-22 [pii];10.1186/1471-2350-10-22 [doi].
- 29. Ramos PS, Criswell LA, Moser KL, Comeau ME, Williams AH, et al. (2011) A comprehensive analysis of shared loci between systemic lupus erythematosus (SLE) and sixteen autoimmune diseases reveals limited genetic overlap. PLoS Genet 7: e1002406 10.1371/journal.pgen.1002406 [doi];PGENETICS-D-11-00575 [pii].
- 30. Ring AM, Lin JX, Feng D, Mitra S, Rickert M, et al. (2012) Mechanistic and structural insight into the functional dichotomy between IL-2 and IL-15. Nat Immunol 13: 1187–1195 ni.2449 [pii];10.1038/ni.2449 [doi].
- 31. Maier LM, Anderson DE, Severson CA, Baecher-Allan C, Healy B, et al. (2009) Soluble IL-2RA levels in multiple sclerosis subjects and the effect of soluble IL-2RA on immune responses. J Immunol 182: 1541−1547. 182/3/1541 [pii].
- 32. Garg G, Tyler JR, Yang JH, Cutler AJ, Downes K, et al. (2012) Type 1 Diabetes-Associated IL2RA Variation Lowers IL-2 Signaling and Contributes to Diminished CD4+CD25+ Regulatory T Cell Function. J Immunol 188: 4644–4653 jimmunol.1100272 [pii];10.4049/jimmunol.1100272 [doi].
- 33. Crispin JC, Tsokos GC (2009) Transcriptional regulation of IL-2 in health and autoimmunity. Autoimmun Rev 8: 190–195 S1568-9972(08)00155-9 [pii];10.1016/j.autrev.2008.07.042 [doi].
- 34. Qu HQ, Verlaan DJ, Ge B, Lu Y, Lam KC, et al. (2009) A cis-acting regulatory variant in the IL2RA locus. J Immunol 183: 5158–5162.
- 35. Butter F, Davison L, Viturawong T, Scheibe M, Vermeulen M, et al. (2012) Proteome-wide analysis of disease-associated SNPs that show allele-specific transcription factor binding. PLoS Genet 8: e1002982 10.1371/journal.pgen.1002982 [doi];PGENETICS-D-12-01265 [pii].
- 36. Rubin LA, Galli F, Greene WC, Nelson DL, Jay G (1990) The molecular basis for the generation of the human soluble interleukin 2 receptor. Cytokine 2: 330−336. 1043-4666(90)90062-X [pii].
- 37. Mkhikian H, Grigorian A, Li CF, Chen HL, Newton B, Zhou RW, et al. (2011) Genetics and the environment converge to dysregulate N-glycosylation in multiple sclerosis. Nat Commun 2: 334 ncomms1333 [pii];10.1038/ncomms1333 [doi].
- 38. Russell SE, Moore AC, Fallon PG, Walsh PT (2012) Soluble IL-2Ralpha (sCD25) exacerbates autoimmunity and enhances the development of Th17 responses in mice. PLoS ONE 7: e47748 10.1371/journal.pone.0047748 [doi];PONE-D-12-19869 [pii].
- 39. Long SA, Buckner JH (2011) CD4+FOXP3+ T Regulatory Cells in Human Autoimmunity: More Than a Numbers Game. J Immunol 187: 2061–2066 187/5/2061 [pii];10.4049/jimmunol.1003224 [doi].
- 40. Liao W, Lin JX, Leonard WJ (2013) Interleukin-2 at the crossroads of effector responses, tolerance, and immunotherapy. Immunity 38: 13–25 S1074-7613(13)00011-3 [pii];10.1016/j.immuni.2013.01.004 [doi].
- 41. Pekalski ML, Ferreira RC, Coulson RM, Cutler AJ, Guo H, et al. (2013) Postthymic Expansion in Human CD4 Naive T Cells Defined by Expression of Functional High-Affinity IL-2 Receptors. J Immunol 190: 2554–2566 jimmunol.1202914 [pii];10.4049/jimmunol.1202914 [doi].