Programmed Death-1 Culls Peripheral Accumulation of High-Affinity Autoreactive CD4 T Cells to Protect against Autoimmunity

Abstract

Self-reactive CD4 T cells are incompletely deleted during thymic development, and their peripheral seeding highlights the need for additional safeguards to avert autoimmunity. Here, we show an essential role for the coinhibitory molecule programmed death-1 (PD-1) in silencing the activation of high-affinity autoreactive CD4 T cells. Each wave of self-reactive CD4 T cells that escapes thymic deletion autonomously upregulates PD-1 to maintain self-tolerance. By tracking the progeny derived from individual autoreactive CD4 T cell clones, we demonstrate that self-reactive cells with the greatest autoimmune threat and highest self-antigen affinity express the most PD-1. Reciprocally, PD-1 deprivation unleashes high-affinity self-reactive CD4 T cells in target tissues to exacerbate neuronal inflammation and autoimmune diabetes. Reliance on PD-1 to actively maintain self-tolerance may explain why exploiting this pathway by cancerous cells and invasive microbes efficiently subverts protective immunity, and why autoimmune side effects can develop after PD-1-neutralizing checkpoint therapies.

Keywords: T cell activation; autoimmunity; costimulation; peripheral immune tolerance.

Figure 1. PD-1 deprivation unleashes high-affinity self-reactive CD4 T cells to exacerbate autoimmunity

(A) MOG:I-A^b tetramer mean fluorescence intensity (MFI) and PD-1 expression levels for MOG-specific CD4 T cells pooled from spleen and peripheral lymph nodes in 6–8 week old naive B6 mice. Each data point represents the average tetramer and PD-1 staining intensity for cells from an individual mouse prior to exogenous MOG peptide stimulation. (B) Number of MOG-specific CD4 T cells, and their intensity of MOG:I-A^b tetramer staining, recovered from pooled brain and spinal cords of PD-1 deficient compared with wild-type mice 14 days after immunization with MOG_35–55 peptide in complete Freund’s adjuvant (CFA). (C) EAE incidence (p < 0.001) and clinical score after immunization with MOG peptide with CFA for PD-1 deficient compared with wild-type mice. (D) p31:I-A^g7 tetramer MFI and PD-1 expression levels for peripheral chromogranin A -specific CD4 T cells pooled from the spleen and peripheral lymph nodes of 13–16 week old NOD mice. (E) Insulin:I-A^g7 tetramer MFI and PD-1 expression levels for peripheral insulin-specific CD4 T cells pooled from the spleen and peripheral lymph nodes of 15–18 week old NOD mice. (F) Number of insulin-specific CD4 T cell, and their intensity of insulin:I-A^g7 tetramer staining, recovered from the pancreas of 5-week old NOD mice treated with PD-1 neutralizing antibody or controls. Animals lacking insulin-specific CD4 T cells in the pancreas were not included in the analysis of insulin:I-A^g7 tetramer staining intensity. (G) Diabetes onset incidence (p < 0.0001) for PD-1 deficient compared with wild-type mice on the NOD background. Error bar, 1 SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 using non-parametric Mann-Whitney U test. See also Figures S1 and S2.

Figure 2. Activation and expansion of high-affinity CD4 T cells following cognate self-antigen stimulation

(A) Schematic used to identify donor (CD45.2) self-reactive CD4 T cells with 2W1S:I-A^b specificity in CD45.1 recipient mice that ubiquitously express 2W1S as a transmembrane recombinant protein in all cells under the β-actin promoter (Act-2W1S mice). (B) Representative FACS plots and composite analyses of CD4 T cells with 2W1S:I-A^b specificity recovered from the spleen and peripheral lymph nodes of Act-2W1S recipients at the indicated time points after intravenous transfer. (C) Percent CD44⁺, 2W1S:I-A^b tetramer MFI, CD3 MFI and 2W1S:I-A^b TCR affinity (tetramer MFI divided by CD3 MFI) for donor 2W1S-specific CD4 T cells pre- and post-transfer, compared with cells of the same specificity in recipient mice. Error bar, 1 SEM. See also Figures S3 and S4.

Figure 3. Autoreactive CD4 T cells acquire self-tolerance in response to cognate antigen stimulation in peripheral tissues

(A) Schematic for probing the response of donor self-reactive CD4 T cells with 2W1S:I-A^b specificity to Lm-2W1S priming within congenically discordant Act-2W1S recipients. (B) Number of donor (CD45.2) 2W1S-specific CD4 T cells recovered following transfer into wild-type or Act-2W1S CD45.1 recipient mice 8 days after Lm-2W1S challenge compared with naive control mice. (C) Schematic showing ex vivo stimulation of donor self-reactive CD4 T cells with 2W1S:I-A^b specificity recovered from Act-2W1S recipient mice. (D) Representative FACS plots and composite analysis comparing IFN-γ and IL-2 production between CD4 T cells with self-2W1S or irrelevant bulk specificity after PMA-ionomycin stimulation. Error bar, 1 SEM. **p < 0.01, ***p < 0.001, ****p < 0.0001 using non-parametric Mann-Whitney U test.

Figure 4. PD-1 reinforces self-tolerance among self-reactive CD4 T cells

(A) Cell intrinsic expression of each molecule by donor self-reactive CD4 T cells with 2W1S:I-A^b specificity (black line) compared with irrelevant bulk specificity (shaded), 30 days after transfer into Act-2W1S recipients. (B) Schematic illustrating adoptive transfer of donor PD-1 deficient or wild-type CD4 T cells into Act-2W1S recipients, followed by Lm-2W1S challenge 30 days after transfer; and number of donor self-reactive CD4 T cells with 2W1S:I-A^b specificity in spleen and peripheral lymph nodes of mice 8 days after Lm-2W1S challenge compared with no infection naive control mice. Error bar, 1 SEM. ****p < 0.0001 using non-parametric Mann-Whitney U test.

Figure 5. PD-1 expression promotes durable cell-intrinsic self-tolerance among self-reactive CD4 T cells

(A) Schematic showing initial transfer of donor CD45.2 CD4 T cells into Act-2W1S CD45.1 recipients, with secondary transfer into wild-type recipients followed by Lm-2W1S priming. (B) Donor 2W1S-specific CD4 T cells in the spleen and peripheral lymph nodes 8 days after Lm-2W1S challenge compared with no infection naive control mice. (C) PD-1 expression by CD4 T cells with 2W1S (black line) or irrelevant bulk (shaded) for the mice described in panels A and B. (D) Schematic showing initial transfer of donor CD45.2 CD4 T cells into Act-2W1S CD45.1 recipient mice, and secondary transfer into irradiated wild-type recipients followed by Lm-2W1S challenge 15–20 days (HP) or >100 days (post-HP). (E) Donor 2W1S-specific CD4 T cells recovered from the spleen and peripheral lymph nodes enumerated 8 days post Lm-2W1S challenge compared with no infection naive control mice. (F) PD-1 expression by CD4 T cells with 2W1S (black line) or irrelevant bulk (shaded) specificity for the mice described in panels D and E. Error bar, 1 SEM. *p < 0.05 using non-parametric Mann-Whitney U test. See also Figures S5 and S6.

Figure 6. An initial wave of self-reactive T cells that bypass clonal deletion does not impede the expansion and activation of subsequent self-reactive cells with overlapping specificity

(A) Schematic showing two distinct cohorts of congenically discordant donor CD4 T cells co-transferred into the same Act-2W1S recipient mouse 30 days apart. (B) Accumulation of self-reactive CD4 T cells with 2W1S:I-A^b specificity from each donor subset in Act-2W1S recipients. (C) Comparison of 2W1S:I-A^b tetramer staining intensity and PD-1 expression among 2W1S-specific CD4 T cells from each donor subset or recipient mice at the indicated time points post-transfer. Error bar, 1 SEM.

Figure 7. PD-1 dampens peripheral accumulation of individual autoreactive CD4 T cell clones based on TCR self-antigen affinity

(A) Schematic showing transfer of individual donor CD4 T cells with 2W1S:I-A^b specificity into congenically discordant Act-2W1S recipients. (B) Correlation between PD-1 MFI and clonal burst size, and between PD-1 MFI and 2W1S:I-A^b TCR affinity, for each self-reactive CD4 T cell clone derived from a polyclonal repertoire. (C) Schematic showing transfer of individual CD4 T cells with 2W1S:I-A^b specificity from either wild-type or PD-1 deficient donors into congenically discordant Act-2W1S recipients. (D) Clonal burst size and 2W1S:I-A^b TCR affinity for individual self-reactive CD4 T cell clones with or without cell-intrinsic PD-1 expression 8 days after transfer. (E) Two-dimensional analysis of clonal burst size and TCR affinity for individual donor CD4 T cell clones with 2W1S:I-A^b specificity described in panels C and D. Error bar, 1 SEM. *p < 0.05, **p < 0.01 using non-parametric Mann-Whitney U test for panel D and multivariable ANOVA for panel E. See also Figure S7 and Table S1.