Logo of wtpaEurope PMCEurope PMC Funders GroupSubmit a Manuscript
Nat Genet. Author manuscript; available in PMC Mar 1, 2010.
Published in final edited form as:
Published online Aug 23, 2009. doi:  10.1038/ng.434
PMCID: PMC2749506

Cell-specific protein phenotypes for the autoimmune locus IL2RA using a genotype-selectable human bioresource


Genome-wide association (GWA) studies have identified over 300 regions associated with more than 70 common diseases1. However, identifying causal genes within an associated region remains a major challenge1,2. One approach to resolving causal genes is through the dissection of gene-phenotype correlations. Here we use polychromatic flow cytometry to show that differences in surface expression of interleukin-2 (IL-2) receptor alpha-chain (IL-2RA, or CD25) protein are restricted to particular immune cell types and correlate with several haplotypes in the IL2RA region that have previously been associated to the autoimmune diseases type 1 diabetes (T1D) and multiple sclerosis2-4. We confirm our strongest gene-phenotype correlation at the RNA level by allele-specific expression (ASE). We also define key parameters for the design and implementation of post-GWA gene-phenotype investigations, and demonstrate the usefulness of a large bioresource of genotype-selectable normal donors from whom fresh, primary cells can be analyzed.

We have previously localized the T1D association in the region containing the IL-2RA gene to three independent SNP groups (haplotypes) found in noncoding regions spanning 18 kb directly 5′ of IL2RA to 25 kb into intron 1 of the gene2-4. The most significantly associated T1D SNPs that define the four common disease-associated IL2RA haplotypes are rs12722495(Ref3), rs11594656(Ref2), and rs2104286(Ref4) (Fig. 1​; Supplementary Table 1a-c). Although none of the SNPs are located within known regulatory regions, we hypothesized that the causative variants may alter CD25 protein expression on the surface of immune cells. The existence of multiple IL2RA haplotypes, as well as a distinct pattern haplotype association with multiple sclerosis4, predicts a complex, haplotype-directed CD25 expression pattern. CD25 is expressed constitutively on numerous immune cell types5 such as CD4+ FOXP3+ regulatory T cells (Tregs) and CD4+ memory T cells6 (Fig. 2), and is upregulated upon activation in many cells including certain innate immune cell subsets7. We therefore investigated haplotype-dependent, cell subset-specific variation in surface CD25 levels on peripheral blood CD4+ memory, naïve, and regulatory T cells (as well as stimulated monocyte subsets) by single-cell polychromatic flow cytometry (Table 1). This allows multiple cell types to be studied simultaneously even if they represent only a small peripheral blood subpopulation, as with Tregs6.

Figure 1
The four common IL2RA/CD25 haplotypes and summary of the subject statistics by haplotype group. (a) The three independent T1D-associated SNP groups are located in the non-coding, 5′ and intron 1 regions of the IL-2RA gene on chromosome 10. The ...
Figure 2
Gating strategy for CD4+ T-cell subsets. CD4+ T cells were gated based on their expression of IL-7R (CD127) and CD25 (inset plot shows isotype control staining). CD127hi cells were further subdivided by their CD45RA expression, with CD45RA cells ...
Table 1
Summary of IL2RA genetic effects, repeatability, age effects and sex effects on CD4+ T-cell and monocyte phenotypes.

To increase the likelihood of identifying gene-phenotype correlations, study participants were specifically selected on the basis of IL2RA haplotype. We used the Cambridge BioResource, a collection of ≈5,000 volunteers living in the Cambridge region who donated blood and saliva samples for research purposes and can be invited to participate in specific research studies based on genotype. The four common IL2RA haplotypes comprise (i) susceptibility alleles at all three SNPs; (ii) a protective rs12722495 allele, which predominantly co-occurs with a protective rs2104286 allele; (iii) a protective rs11594656 allele; and (iv) a protective rs2104286 allele (Fig. 1a; Supplementary Table 1d). We aimed to obtain ~50 individuals homozygous or heterozygous for each of the three protective haplotypes, and 50 individuals homozygous for the susceptible haplotype. Individuals heterozygous for two different protective haplotypes were excluded, and thus all heterozygous individuals studied had one protective and one susceptible haplotype. Of the 2,212 individuals who met our recruitment criteria, we contacted ~100 from each haplotype group (Supplementary Table 2) and 179 individuals were analyzed in total (Fig. 1b). Notably, using this haplotype-specific approach to select samples, we succeeded in analyzing 16 individuals homozygous for the protective rs12722495 allele, even though the frequency of such individuals in the Cambridge BioResource is only 1%. All donors selected were of self-reported non-autoimmune disease status. We assessed healthy donors because genetic effects on a phenotype could be confounded by the presence of disease8, particularly as CD25 is known to be upregulated under inflammatory conditions5.

We found that individuals with one or two protective rs12722495 alleles show 27% higher mean CD25 levels on their CD4+ memory T cells compared to fully susceptible individuals or to donors with protective rs11594656 or rs2104286 alleles (P = 4.30 × 10−9; Fig. 3a). Furthermore, there is a gene dosage effect at rs12722495 (P = 1.16 × 10−10; Fig. 3b; see Fig. 2 for representative profiles), with heterozygous and homozygous protective individuals having 22% and 33% higher mean CD25 levels, respectively, compared to homozygous susceptible individuals. This gene dosage effect is consistent with the T1D allelic risk following a conventional multiplicative model whereby the risk with two doses of an allele is the square of the risk with one allele. No dosage effect was found for rs11594656 or rs2104286 (Supplementary Fig. 1a-c), and this phenotype was not associated with donor age or sex (Table 1). The genetically variable surface CD25 levels on CD4+ memory T cells likely reflect haplotype-dependent transcriptional differences. We found that in total CD4+ T cells and CD4+ memory T cells from donors who were heterozygous for both the protective rs12722495 and fully susceptible haplotypes there were ~30% more pre-mRNA transcripts from the protective haplotype, as determined by cloning-based ASE assays9 designed for SNPs in intron 1 of IL2RA (Supplementary Table 3, Supplementary Fig. 2).

Figure 3
IL2RA genotype and CD4+ T-cell phenotype correlations. (a) Individuals with at least one protective rs12722495 allele have higher CD25 expression on their CD4+ memory T cells (mean CD25 MEF=1,000) than fully susceptible individuals (mean=773) or protected ...

Notably, the surface CD25 levels of CD4+ memory T cells were highly consistent over time (r2 = 0.997; Supplementary Fig. 1d), as demonstrated in 15 donors recalled 3-7 months after their initial donation. This repeatability was compromised, however, when 14 peripheral blood mononuclear cell (PBMC) samples were analyzed 9–10 months after cryopreservation for CD25 levels on the CD4+ memory T cells, as these levels were 28–62% lower than those found on the CD4+ memory T cells analyzed in fresh whole blood on the day the PBMCs were isolated and frozen (Supplementary Fig. 1e). The IL2RA genotype was estimated to account for 18% of CD25 variation in fresh peripheral blood CD4+ memory T cells. Inter-individual variation in donors with homozygous protective rs12722495 alleles was lower than in homozygous susceptible donors (F-test P = 0.003). Given the phenotype repeatability, this indicates that other IL2RA SNPs, variants at other genes or environmental factors alter CD25 expression on CD4+ memory T cells, but that these effects are only observable in individuals having rs12722495 susceptibility alleles. We note that no correlation was observed between mean surface CD25 levels on FOXP3+ T cells and IL2RA haplotype (Supplementary Fig. 3a-d).

In contrast to the rs12722495 genotypic correlation for increased CD25 on CD4+ memory cells, the percentage of CD4+ naïve cells that are CD25+ is lower in individuals with protective alleles at rs2104286 (P = 4.25 × 10−6; Fig. 3c and d; Supplementary Fig. 3e-h), regardless of whether these alleles co-occur on the same haplotype as protective rs12722495 alleles. This phenotype also had good repeatability (r2 = 0.669). We also obtained evidence for a genotype-phenotype correlation for CD25 upregulation on a stimulated monocyte subset (Table 1; Supplementary Fig. 4), which was analogous to that for CD4+ naïve cells.

Considering our results collectively (Table 1), relative to the susceptible haplotype, the protective rs12722495 haplotype (OR = 0.57; G, protective, minor allele at rs12722495), which is also protective at rs2104286 (G allele), is correlated with three phenotypic changes (in CD4+ memory and naive T cells and in stimulated CD14+ CD16+ monocytes). In comparison, the protective rs2104286 haplotype (OR = 0.88; G, protective, minor allele at rs2104286) is correlated with only two phenotypic changes (in CD4+ naive T cells and stimulated CD14+ CD16+ monocytes). This may, in part, explain their differing strengths of association to T1D. We propose that the CD4+ memory T-cell phenotype is critical for protection from T1D but not from multiple sclerosis, in which rs2104286 (OR = 0.85; G, protective, minor allele at rs2104286) is the most strongly associated SNP4,10, whereas the naïve T-cell and stimulated CD14+ CD16+ monocyte phenotypes are implicated in both diseases.

The IL2RA haplotype-specific CD25 expression patterns in CD4+ naïve and memory T cells are consistent with these cells being distinct cell types, as determined by gene expression changes that occur with T-cell maturation and differentiation11, including those mediated by chromosome remodeling and altered transcription and splicing factors. According to the quantal theory of immunity, T-cell responses depend on cells receiving a critical number of T cell receptor (TCR)- and IL-2R-mediated stimuli12. For CD4+ naïve T cells, the protective rs2104286 genotype reduces the probability of this cell type expressing CD25 thereby reducing the likelihood of activation. Indeed, we found that this genotypic effect persisted upon T cell activation, such that a lower proportion of CD69+ CD4+ naïve T cells upregulated CD25 in donors carrying the protective rs2104286 allele compared to fully susceptible donors (Supplementary Fig. 5). In contrast, because individuals with protective rs12722495 alleles have higher CD25 levels on their CD4+ memory cells, these cells are likely more responsive to IL-2 and to TCR-mediated activation. In mice it has been hypothesized that these nonregulatory CD4+ CD25+ T cells are the primary source in vivo of the IL-2 required for FOXP3+ Treg homeostasis and competitive fitness13-15. This is consistent with the correlation between an ~50% reduction in IL-2 abundance and reduced Treg function in mice carrying two T1D susceptibility alleles at Idd3(Ref16​,​17). By analogy, it is possible that when human nonregulatory CD4+ memory T cells express higher levels of CD25 due to the protective rs12722495 haplotype, homeostatic IL-2 production is increased; this would be consistent with observed defects in IL-2 production shown by individuals with T1D18,19. In support of this hypothesis we have found that upon stimulation, a higher percentage of CD69+ CD4+ memory T cells from protective rs12722495 homozygous donors secrete IL-2 compared with those from homozygous susceptible donors (P = 5.74 × 10−4; Fig. 4). An alternative, or additional, mechanism to reinforce self-tolerance that would be improved with higher CD25 levels on activated, anti-self effector T cells is activation-induced cell death20,21. Modest increases in IL-2 production enable mice to more efficiently establish transplant tolerance, presumably via this mechanism22.

Figure 4
IL-2 secretion by stimulated CD69+ CD4+ T cell subsets. (a) Plot showing that unstimulated CD4+ memory T cells do not upregulate CD69 at their cell surface and do not secrete IL-2. (b) Upon stimulation of isolated PBMCs for 4 hours with 5 μg/ml ...

This study represents the first description gene-phenotype correlations between protein levels on specific subsets of primary human immune cell types and autoimmune disease-associated variants. We provide evidence for three immune cell type-specific quantitative trait effects on protein expression associated with two protective IL2RA haplotypes, providing further support for IL2RA as a T1D-causative gene at the 10p15.1 locus. We also delineate important parameters for the design and practical implementation of gene-phenotype correlation studies. The establishment of a panel of thousands of healthy donors recallable by genotype and resident near the study location was critical to the success of our approach, as was the analysis of fresh rather than frozen primary immune cells. Given the necessity for advances in the field of human immunology23,24, our findings illustrate that although the human population is genetically diverse and individuals are subjected to immune challenges, gene-phenotype correlations can indeed be dissected. Our results will also direct further studies to examine the functional consequences of IL2RA genetic variation in vitro and in animal models of autoimmune disease.



The Cambridge BioResource is funded by the Wellcome Trust, the Juvenile Diabetes Research Foundation (JDRF) International and the National Institute for Health Research Cambridge Biomedical Research Centre. The Cambridgeshire 2 Research Ethics Committee gave ethical approval for the recruitment of 1,000 local volunteers to the Cambridge BioResource, which was formed as a collaboration between the University of Cambridge and the Medical Research Council (MRC). In collaboration with the National Health Service Blood and Transplant (NHSBT) Centre, another 4,000 blood donors were recruited to the Cambridge BioResource, with the ethical approval of the Cambridgeshire 1 Research Ethics Committee. Informed consent was obtained from all volunteers upon recruitment to the Cambridge BioResource for the collection and use of DNA samples for genotyping.

Subjects were selected by genotype from the Cambridge BioResource as part of the ‘Genes and Mechanisms of Type 1 Diabetes in the Cambridge BioResource’ study and were of self-reported white ethnicity. Ethical approval for this study was given by the Peterborough and Fenland Local Research Ethics Committee, and informed consent was obtained for the collection and use of the peripheral blood samples. Subjects were of self-reported non-autoimmune disease status; we specifically excluded individuals with T1D to avoid any confounding due to T1D status or to insulin treatment.

We aimed to obtain samples from ~200 donors enriched for less common IL2RA genotypes and including at least one donor from each of our four haplotype groups on each sample donation day to minimize any day-of-collection effects. The immunophenotyping was conducted in a genotype-blinded fashion. We designed our study to be able to detect an effect size as small as 50% of the genotype-related difference in CTLA-4 expression observed in previous work25. Given these prior estimates a power study shows that a random sampling of the population would provide 10% power to detect a gene-phenotype correlation at P=10−5. With the same total number of donors and the allele distribution resulting from our genotype-specific sampling, we have obtained 87% power at the same significance level.


SNPs were genotyped using Custom TaqMan® SNP Genotyping Assays or Taqman (Applied Biosystems) in accordance with the manufacturers’ protocols. All genotyping data were double scored by a second operator to minimize error.

Antibodies and whole blood immunostaining

The anti-human monoclonal antibodies used for T cell surface immunostaining were allophycocyanin (APC)-conjugated anti-CD25 (clones M-A251 and 2A3; BD Biosciences), Alexa-Fluor®700-conjugated anti-CD4, Alexa-Fluor®488-conjugated anti-CD127, and Pacific BlueTM-conjugated anti-CD45RA (BioLegend). The isotype control antibodies used were APC-conjugated mouse IgG1 (BD Biosciences) and Alexa-Fluor®488-conjugated mouse IgG1 (BioLegend). To minimize potential variation due to antibody batch differences, all antibodies were obtained prior to the start of the experiment and all vials of antibody derived from the same clone and labeled with the same fluorochrome were pooled prior to usage.

To better visualize lower-level surface CD25 expression, we increased CD25 detection sensitivity by simultaneously using two anti-CD25 monoclonal antibodies, labeled with the same fluorochrome (from clones 2A3 and M-A251, BD Biosciences), that do not cross-compete and that recognize distinct epitopes on the CD25 molecule26,27.

Samples were stained within 5 h after venesection. After blocking, samples were stained for 40 min and then lysed with freshly prepared 1X BD FACS™ Lysing Solution (BD Biosciences). After erythrocyte lysis, samples were incubated at 4°C and were washed with BD CellWASH™ (BD Biosciences). The samples were fixed with freshly prepared 1X BD CellFIX™ (BD Biosciences). The samples were stored at 4°C until analysis by flow cytometry or intracellular staining. For intracellular staining, the PE (phycoerythrin)-anti-human FOXP3 Staining Set from eBioscience was used, according to the manufacturer’s instructions.

Flow cytometry and data analysis

Immunostained samples were analyzed using a BD™ LSRII Flow Cytometer with BD FACSDiVa Software (BD Biosciences). The flow cytometry data obtained were analyzed using FlowJo (Tree Star, Inc.). Each day donor samples were evaluated, FluoroSpheres were also analyzed (Blank Beads and Calibration Beads, Dakocytomation). The Calibration Beads contain a mixture of bead populations of different fluorescence intensities that have been assigned specific ‘molecules of equivalent fluorochrome’ (MEF) values; thus they allow the transformation of mean or median fluorescence intensity (MFI) into mean or median MEF, which is a quantity proportional to the average number of molecules on a cell. This transformation is linear, MEF=a.MFI, where the slope ‘a’ is a measure of the number of molecules equivalent to one unit of fluorescence.

Statistical analysis of flow cytometry data

All statistical tests were performed using the R statistical software package. Regression analysis and analysis of variance (ANOVA) were used to test for correlations between genotypes and phenotypes, controlling for age and sex. Genotype-phenotype correlations were analysed assuming a linear genotype-phenotype relationship (1-d.f. F-test). A conservative significance threshold was estimated (P=5.56×10−4) using a Bonferroni correction for multiple testing (using a P-value of 0.01 and assuming six potential CD25-related phenotypes and three IL2RA SNPs, giving a total of 18 independent hypotheses). The percentage of IL2RA genotype accounting for the variation in CD4+ memory T-cell CD25 levels was estimated by least-squares regression analysis. One donor was excluded from the CD4+ T cell data and another was excluded from the monocyte data, on the basis that the CD25 levels of these individuals differed from the mean of the rest of the donors by at least 5 standard deviations, and these individuals were thus considered to be outliers.

Peripheral blood mononuclear cell (PBMC) isolation, cryopresesrvation and thawing

PBMC isolation, cryopreservation and thawing were performed as previously described28. PBMC isolation was carried out using Lymphoprep (Fresenius Kabi Norge AS for Axis-Shield PoC AS). PBMCs were cryopreserved in heat-inactivated, filtered human AB serum (Sigma-Aldrich) + 10% DMSO (Hybri-MAX™, Sigma-Aldrich) at a final concentration of 10×106 ml−1 and were stored in liquid nitrogen. Cryovials were thawed in a 37°C water bath. PBMCs were washed by adding 5 ml of room-temperature (20-25°C), heat-inactivated, filtered human AB serum and then 5 ml of RPMI-1640 + GlutaMAX™ I + 25 mM HEPES (GIBCO®, Invitrogen) per 10×106 cells, in a drop-wise fashion.

Cloning-based allele-specific expression assay

PBMCs were isolated from three donors heterozygous for the protective rs12722495 and fully susceptible haplotypes. CD4+ T cells were obtained from PBMCs using the Dynabeads® Untouched™ Human CD4 T Cells kit (Dynal, Invitrogen). CD4+ memory T cell enrichment was carried out using the Dynabeads® Untouched™ Human CD4 T Cells kit with the inclusion of unconjugated anti-CD45RA and anti-HLA-DR antibodies (BD Biosciences), to remove CD4+ naïve T cells and the majority of CD4+ CD127int-low CD25hi regulatory T cells along with non-CD4+ T cells. Cell purity was estimated by flow cytometry.

Genomic DNA was obtained from PBMCs by chloroform extraction. Total RNA was extracted from the negatively selected cells using TRIzol® Reagent (Invitrogen) and further purified using the RNeasy Mini kit (QIAGEN), according to the manufacturers’ instructions. RNA was either left unfragmented or fragmented29 to provide better coverage throughout the gene30. Complementary DNA synthesis was then carried out using Superscript™ III RT kit and oligo dT or random hexamer primers (Invitrogen). After cDNA synthesis using fragmented RNA, 150-350 bp long cDNA molecules were gel-purified using QIAquick® Gel Extraction Kit (QIAGEN). Primers flanking the rs12722495 and rs2104286 SNPs were used to amplify the cDNA and genomic DNA by PCR. PCR products were gel-purified and ligated into pCR®4-TOPO® Vector and transformed chemically into One Shot® Top 10 competent Escherichia coli cells (Invitrogen).

Bacterial colonies were picked and PCR amplification was performed using an AmpliTaq Gold™ protocol (ABI). PCR products were screened by TaqMan genotyping (ABI). Alleles were scored on SDS v2.1 software (ABI) manually and counted. Statistical analysis was performed using a 2×2 contingency table to calculate one-tailed P-values for T1D-associated SNPs, while a two-tailed P-value was used for rs12244380 as the phase for the alleles of this SNP relative to the T1D-associated SNPs could not be determined. Assuming a 30% allelic imbalance between the protective rs12722495 and fully susceptible haplotypes, we calculated that 350 colonies needed to be picked to provide 80% power to detect a 30% allelic imbalance at the RNA level at P=0.05, while 1,000 colonies would provide 90% power at P=0.001; hence we aimed to pick ~350 bacterial colonies per assay.

Monocyte whole blood activation

Aliquots of whole blood, collected in CPDA VACUETTE® tubes (Greiner Bio-One) were stimulated with 200 U/ml GM-CSF and 200 U/ml IFNγ (PeproTech) for 6 hours at 37°C with shaking at 200 rpm. Unstimulated samples were treated identically except that GM-CSF and IFNγ were not added to the whole blood samples. After activation the blood samples were immunostained. The anti-human monoclonal antibodies used were PE-conjugated anti-CD25 (clones M-A251 and 2A3; BD Biosciences), Pacific Blue™-conjugated anti-CD16 (BioLegend), Pacific Orange™-conjugated anti-CD14 (Invitrogen), and PerCP-conjugated anti-HLA-DR (BD Biosciences).

PBMC activation

Eight pairs of donors were recalled such that in each pair one donor was homozygous for the protective rs12722495 haplotype and one donor was homozygous for the fully susceptible haplotype (and had CD4+ memory cells with CD25 levels <800 MEF). Freshly isolated PBMCs were cultured in 24-well flat-bottom plates at a concentration of 2×106 ml−1 per well in X-VIVO™ 15 (Lonza) supplemented with 1% heat-inactivated, filtered human AB serum (Sigma-Aldrich). Cells were stimulated with 5 μg/ml of SEB (Toxin Technology, Inc.) for 4 hours. Following stimulation, T-cell subsets secreting IL-2 were identified using a PE IL-2 Secretion Assay (Miltenyi Biotec) and immunostaining with fluorescein isothiocyanate (FITC)-conjugated anti-CD25 (clones M-A251 and 2A3; BD Biosciences), APC-conjugated anti-CD69 (BD Biosciences), Alexa-Fluor®700-conjugated anti-CD4, and Pacific Blue™-conjugated anti-CD45RA (BioLegend). Data were analyzed using a paired, one-tailed student’s t-test and a 5% significance level.

T-cell whole blood activation

Pairs of donors were used as described for “PBMC activation”. Aliquots of heparinized whole blood were diluted 1:1 with RPMI-1640 + GlutaMAX™ I + 25 mM HEPES (GIBCO®, Invitrogen). Samples were stimulated with 10 μl of BD FastImmune™ CD2/CD2R (BD Biosciences) per 1 ml of diluted blood at 37°C for 21 hours. Samples were stained with PE-conjugated CD69 (BD Biosciences), APC-conjugated anti-CD25 (clones M-A251 and 2A3; BD Biosciences), Alexa-Fluor®700-conjugated anti-CD4, and Pacific Blue™-conjugated anti-CD45RA (BioLegend), and the FITC anti-human FOXP3 Staining Set from eBioscience was used according to the manufacturer’s instructions. Data were analyzed using a paired, one-tailed student’s t-test and a 5% significance level.

Supplementary Material

Supplementary Information


We gratefully acknowledge the participation of all Cambridge BioResource subjects. We thank R. Bailey, P. Clarke, H. Stevens, S. Duley, D. Harrison, M. Maisuria, T. Mistry, N. Taylor, for DNA sample preparation and D. Smyth for genotype double scoring. We thank J. Casey, W. Herring, K. Hodge, H. Lloyd-Jones, R. Walker, and M. Wiesner for donor coordination and blood sample obtainment and J. Clark for help with immunostaining. We thank T. Attwood for his contribution to sample management. We thank the other members of the Cambridge BioResource Management Committee: D. Dunger, S. O’Rahilly, D. Savage, J. Sambrook, A. Calder, W. Marslen-Wilson, K. Chatterjee, and C. Saunders. We acknowledge use of the DNA from the British 1958 Birth Cohort collection, funded by the Medical Research Council and Wellcome Trust. We thank the Avon Longitudinal Study of Parents and Children laboratory in Bristol and the British 1958 Birth Cohort team, including S. Ring, R. Jones, M. Pembrey, W. McArdle, D. Strachan and P. Burton for preparing and providing the control DNA samples. We also thank the Wellcome Trust, the Juvenile Diabetes Research Foundation (JDRF) International, and the National Institute for Health Research Cambridge Biomedical Research Centre for funding. This work was also supported by a National Institutes of Health (USA) grant P01 AI39671 and a grant from the JDRF UK Centre for Diabetes Genes, Autoimmunity and Prevention (D-GAP; 4-2007-1003). The Cambridge Institute for Medical Research is the recipient of a Wellcome Trust Strategic Award (079895). LSW is a Juvenile Diabetes Research Foundation/Wellcome Trust Principal Research Fellow.


Competing Interests Statement The authors declare no competing financial interests.

URLs. Cambridge BioResource, http://www.cambridgebioresource.org.uk; R, http://www.r-project.org/; T1Dbase, http://www.t1dbase.org/; FlowJo, http://www.treestar.com/flowjo/.


1. Donnelly P. Progress and challenges in genome-wide association studies in humans. Nature. 2008;456:728–731. [PubMed]
2. Lowe CE, et al. Large-scale genetic fine mapping and genotype-phenotype associations implicate polymorphism in the IL2RA region in type 1 diabetes. Nat. Genet. 2007;39:1074–1082. [PubMed]
3. Smyth DJ, et al. Shared, distinct and opposing genetic factors in type 1 diabetes and celiac disease. N. Engl. J. Med. 2008;359:2767–2777. [PMC free article] [PubMed]
4. Maier LM, et al. IL2RA genetic heterogeneity in multiple sclerosis and type 1 diabetes susceptibility and soluble interleukin-2 receptor production. PLoS Genet. 2009;5:e1000322. doi:10.1371/journal.pgen.1000322. [PMC free article] [PubMed]
5. Dendrou CA, Wicker LS. The IL-2/CD25 pathway determines susceptibility to T1D in humans and NOD mice. J. Clin. Immunol. 2008;28:685–696. [PubMed]
6. Baecher-Allan C, Wolf E, Hafler DA. Functional analysis of highly defined, FACS-isolated populations of human regulatory CD4+ CD25+ T cells. Clin. Immunol. 2005;115:10–18. [PubMed]
7. Kniep EM, Strelow I, Lohmannmatthes ML. The monocyte interleukin-2 receptor light chain - production of cell-associated and soluble interleukin-2 receptor by monocytes. Immunology. 1992;75:299–304. [PMC free article] [PubMed]
8. Maier LM, et al. Soluble IL-2RA in multiple sclerosis subjects and the effect of soluble IL-2RA on immune responses. J. Immunol. 2009;182:1541–1547. [PMC free article] [PubMed]
9. Rainbow DB, et al. Commonality in the genetic control of Type 1 diabetes in humans and NOD mice: variants of genes in the IL-2 pathway are associated with autoimmune diabetes in both species. Biochem. Soc. Trans. 2008;36:312–315. [PMC free article] [PubMed]
10. Hafler DA, et al. Risk alleles for multiple sclerosis identified by a genomewide study. N. Engl. J. Med. 2007;357:851–862. [PubMed]
11. McKarns SC, Schwartz RH. Biphasic regulation of Il2 transcription in CD4+ T cells: Roles for TNF-alpha receptor signaling and chromatin structure. J. Immunol. 2008;181:1272–1281. [PMC free article] [PubMed]
12. Smith KA. The quantal theory of how the immune system discriminates between “self and non-self” Med. Immunol. 2004;3:3. [PMC free article] [PubMed]
13. Fontenot JD, Rasmussen JP, Gavin MA, Rudensky AY. A function for interleukin 2 in Foxp3-expressing regulatory T cells. Nat. Immunol. 2005;6:1142–1151. [PubMed]
14. Setoguchi R, Hori S, Takahashi T, Sakaguchi S. Homeostatic maintenance of natural Foxp3(+) CD25(+) CD4(+) regulatory T cells by interleukin (IL)-2 and induction of autoimmune disease by IL-2 neutralization. J. Exp. Med. 2005;201:723–735. [PMC free article] [PubMed]
15. Wu YQ, et al. FOXP3 controls regulatory T cell function through cooperation with NFAT. Cell. 2006;126:375–387. [PubMed]
16. Yamanouchi J, et al. Interleukin-2 gene variation impairs regulatory T cell function and causes autoimmunity. Nat. Genet. 2007;39:329–337. [PMC free article] [PubMed]
17. Sgouroudis E, Albanese A, Piccirillo CA. Impact of protective IL-2 allelic variants on CD4+ Foxp3+ regulatory T cell function in situ and resistance to autoimmune diabetes in NOD mice. J. Immunol. 2008;181:6283–6292. [PubMed]
18. Zier K, Leo M, Spielman R, Baker L. Decreased synthesis of interleukin-2 (IL-2) in insulin-dependent diabetes mellitus. Diabetes. 1984;33:552–555. [PubMed]
19. Kaye WA, et al. Acquired defect in interleukin-2 production in patients with type 1 diabetes mellitus. N. Engl. J. Med. 1986;315:920–924. [PubMed]
20. Lenardo MJ. Fas and the art of lymphocyte maintenance. J. Exp. Med. 1996;183:721–724. [PMC free article] [PubMed]
21. Willerford DM, et al. Interleukin-2 receptor-alpha chain regulates the size and content of the peripheral lymphoid compartment. Immunity. 1995;3:521–530. [PubMed]
22. Pearson T, et al. Islet allograft survival induced by costimulation blockade in NOD mice is controlled by allelic variants of Idd3. Diabetes. 2004;53:1972–1978. [PubMed]
23. Hayday AC, Peakman M. The habitual, diverse and surmountable obstacles to human immunology research. Nat. Immunol. 2008;9:575–580. [PubMed]
24. Davis MM. A prescription for human immunology. Immunity. 2008;29:835–838. [PMC free article] [PubMed]
25. Ueda H, et al. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature. 2003;423:506–511. [PubMed]
26. Tkaczuk J, et al. Effect of anti-IL-2Ra antibody on IL-2-induced Jak/STAT signaling. Am. J. Transplant. 2002;2:31–40. [PubMed]
27. Urdal DL, March CJ, Gillis S, Larsen A, Dower SK. Purification and chemical characterization of the receptor for interleukin 2 from activated human T lymphocytes and from a human T-cell lymphoma cell line. Proc. Natl. Acad. Sci. U. S. A. 1984;81:6481–6485. [PMC free article] [PubMed]
28. Tree TI, Roep BO, Peakman M. Enhancing the sensitivity of assays to detect T cell reactivity: the effect of cell separation and cryopreservation media. Ann. N. Y. Acad. Sci. 2004;1037:26–32. [PubMed]
29. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat. Methods. 2008;5:621–628. [PubMed]
30. Wang Z, Gerstein M, Snyder M. RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet. 2009;10:57–63. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...