Defective regulatory and effector T cell functions in patients with FOXP3 mutations

Abstract

The autoimmune disease immune dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) is caused by mutations in the forkhead box protein P3 (FOXP3) gene. In the mouse model of FOXP3 deficiency, the lack of CD4+ CD25+ Tregs is responsible for lethal autoimmunity, indicating that FOXP3 is required for the differentiation of this Treg subset. We show that the number and phenotype of CD4+ CD25+ T cells from IPEX patients are comparable to those of normal donors. CD4+ CD25high T cells from IPEX patients who express FOXP3 protein suppressed the in vitro proliferation of effector T cells from normal donors, when activated by "weak" TCR stimuli. In contrast, the suppressive function of CD4+ CD25high T cells from IPEX patients who do not express FOXP3 protein was profoundly impaired. Importantly, CD4+ CD25high T cells from either FOXP3+ or FOXP3- IPEX patients showed altered suppression toward autologous effector T cells. Interestingly, IL-2 and IFN-gamma production by PBMCs from IPEX patients was significantly decreased. These findings indicate that FOXP3 mutations in IPEX patients result in heterogeneous biological abnormalities, leading not necessarily to a lack of differentiation of CD4+ CD25high Tregs but rather to a dysfunction in these cells and in effector T cells.

Figure 1. Schematic representation of the genomic organization of the FOXP3 gene, including the regions encoding each protein domain.

Mutations identified in the 4 IPEX patients analyzed are indicated. Pt1, patient 1; Pt2, patient 2; Pt2b, patient 2bis; Pt3, patient 3.

Figure 2. Phenotype of patients’ and NDs’ PBMCs.

(A) Dot plots of cell-surface expression of CD25. The percentage of CD4⁺CD25⁺ and CD4⁺CD25^– T cells and relative MFI are indicated by the numbers in the plots. Intracytoplasmic CTLA4 and surface GITR expression are shown in histograms; gates for analysis were set on CD4⁺CD25^– T cells (thin lines) and CD4⁺CD25⁺ T cells (thick lines), as shown in the dot plots. MFI values indicated in the histogram plots refer to the CD4⁺CD25⁺-gated cells. (B) Expression of CD45RA, CD62L, HLA-DR, and CD69 relative to CD25 in gated CD4⁺ T cells. Quadrants and regions of positive staining were set based on the isotype control (data not shown), and the numbers indicate the percentages of positive cells in each quadrant.

Figure 3. Phenotype of CD4⁺ CD25^– and CD4⁺ CD25⁺ T cell lines.

Surface expression of CD25 and GITR and intracytoplasmic expression of CTLA4 were determined on T cell lines derived from CD4⁺CD25^– (thin lines) and CD4⁺CD25⁺ (thick lines) T cells, isolated by FACS sorting and expanded in vitro. Staining was performed on resting cells (at least 12 days after activation). Regions of positive staining were set based on the isotype control (data not shown).

Figure 4. Suppressive activity of patients’ CD4⁺ CD25^high T cells.

The ability of freshly isolated CD4⁺CD25^high T cells of patient 2, patient 2bis, patient 3, and NDs to suppress either ND CD4⁺CD25^– effector T cells or autologous CD4⁺CD25^– effector T cells was assessed (in the case of NDs, effector and suppressor cells were derived from the same individuals). Responder cells were activated either in the presence of 50,000 irradiated (60 Gy) APCs plus soluble anti-CD3 mAb (1 μg/ml) (A) or with anti-CD3 plus anti-CD28 mAb–coated beads (B) (see Methods for details). CD4⁺CD25^high T cells (Treg) were added to activated responder cells (R), and [³H]thymidine incorporation was measured after 72 hours. Percentages indicate inhibition of proliferation.

Figure 6. Expression of FOXP3 protein.

(A) Results of Western blot analysis of FOXP3 protein expression in fresh CD4⁺CD25^– T cells and CD4⁺CD25^high Tregs isolated from a normal control subject and in CD4⁺CD25^– or CD4⁺CD25^high T cell lines from patient 1, patient 2, and an ND are shown. The amount of loaded protein was equivalent in each condition as shown by the anti-ERK staining. (B) Results of Western blot analysis of FOXP3 protein expression in T cell lines derived from CD4⁺CD25^– T cells of patient 3 and of an ND, analyzed resting and 24–48 hours after activation, are shown. As a control, results for CD4⁺CD25⁺T cells from an ND tested in parallel are also shown. The amount of loaded protein was equivalent in each condition as shown by the anti-p38 staining. Neg, negative; Pos, positive.

Figure 5. Suppressive activity of patients’ CD4⁺ CD25^high T cell lines.

The ability of CD4⁺CD25^high T cell lines derived from patient 1 (Treg Pt1), patient 2 (Treg Pt2), patient 2bis (Treg Pt2b), or an ND (Treg ND) to suppress CD4⁺CD25^– responders from either allogeneic ND-derived T cell line (left panels) or autologous T cell line (right panels) was assessed. Percentages indicate inhibition of proliferation. Data shown are representative of 3 independent experiments.

Figure 7. Functional activity of WT FOXP3 and mutants on the IL-2 promoter.

Jurkat cells were transfected with a reporter gene construct containing luciferase under the control of the hIL-2 promoter in the presence of an empty vector control (Con), WT FOXP3, or mutant FOXP3 of patients 1 and 2. In all cases, cells were also cotransfected with β-gal under control of the elongation factor 1α promoter. After 16 hours, cells were stimulated with TPA plus ionomycin for an additional 6 hours, then lysed and analyzed for levels of luciferase and β-gal activity. Shown is the percent activity of the IL-2 promoter after normalizing to levels of β-gal, with the empty vector condition set at 100%. Data represent the mean of 4 independent experiments.

Figure 8. Cytokine production by PBMCs isolated from IPEX patients: determination in cell culture supernatants.

PBMCs were stimulated with immobilized anti-CD3 mAb (10 μg/ml) and soluble anti-CD28 mAb (1 μg/ml) (A) or with TPA and ionomycin (B). Supernatants were collected after 12 (IL-2) or 48 hours (all other cytokines), and cytokine concentrations were determined by cytometric bead array (see Methods for details). The values obtained in patient 1 (n = 2 independent determinations) and patient 2 (n = 4 independent determinations) were compared with those obtained in a cohort of age-matched NDs (n = 15) analyzed in parallel for statistical analysis. One single determination was performed in patients 2bis and 3.