S1P1 deletion differentially affects TH17 and Regulatory T cells

Abstract

Sphingosine-1 phosphate receptor 1 (S1P₁) is critical for the egress of T and B cells out of lymphoid organs. Although S1P₁ agonist fingolimod is currently used for the treatment of multiple sclerosis (MS) little is known how S1P₁ signaling regulates Th17 and T_reg cell homeostasis. To study the impact of S1P₁ signaling on Th17 and T_reg cell biology, we specifically deleted S1P₁ in Th17 and T_reg cells using IL-17A ^Cre and Foxp3 ^Cre mice, respectively. Deletion of S1P₁ in Th17 cells conferred resistance to experimental autoimmune encephalomyelitis (EAE). On the other hand, permanent deletion of S1P₁ in T_reg cells resulted in autoimmunity and acute deletion rendered mice more susceptible to EAE. Importantly, our study revealed that S1P₁ not only regulated the egress of T_reg cells out of lymphoid organs and subsequent non-lymphoid tissue distribution but also their phenotypic diversity. Most of the T_reg cells found in S1P₁-deficient mice as well as MS patients on fingolimod therapy had an activated phenotype and were more prone to apoptosis, thus converted to effector T_reg. Our results provide novel insight into the functions of S1P₁ and potential impact of long term fingolimod use on Th17 and T_reg cell biology and general health in MS patients.

Figure 1

S1P₁ is required for peripheral organ distribution and in vivo generation of Th17 cells. (a) EAE scores of S1P ₁ ^Flox IL-17A ^Cre ROSA ^RFPand control mice (S1P ₁ ^WT IL-17A ^Cre ROSA ^RFP) that were immunized with MOG_35–55 and Complete Freund’s Adjuvant. N = 5 mice per group, the experiment was repeated three times. (b) IL-17A and IFN-γ expression of CD4+ T cells from central nervous system (CNS) of S1P ₁ ^Flox IL-17A ^Cre ROSA ^RFP and control mice 15 days after MOG_35–55 immunization (left). The cells were cultured 4 h with PMA/ionomycin and Golgi Plug for intracellular staining. Quantification of flow plots, (right).* indicates p-value <0.05. (c) Th17 and Th1 cell distribution across the peripheral organs in S1P ₁ ^Flox IL-17A ^Cre ROSA ^RFP and control mice, 7 days after MOG_35–55 immunization, and quantification was shown in (d). (e) Representative flow plot of RFP⁺ CD4⁺ cells after rechallenge with media, MOG_35–55 or MOG_35–55 + IL-23 following the first immunization. (f) IL-17A and IFN-γ production by S1P ₁ ^Flox IL-17A ^Cre ROSA ^RFP and control mice in MOG_35–55 + IL-23 rechallenge condition.

Figure 2

T_reg-specific deletion of S1P₁ causes autoimmunity. (a) Representative image of skin lesions that develop around the eyes of 8-week-old S1P ₁ ^Flox Foxp3 ^Cre compared to S1P ₁ ^WT Foxp3 ^Cre mice. (b) Representative splenomegaly and lymphadenopathy in 8 weeks-old S1P ₁ ^Flox Foxp3 ^Cre mice compared to S1P ₁ ^WT Foxp3 ^Cre mice. (c) Representative section of liver and lung stained with hematoxylin & eosin (H&E) depicting higher lymphocytic infiltrates in 6 weeks-old S1P ₁ ^Flox Foxp3 ^Cre mice. Arrows point the lymphocytic infiltrates. PV, portal vein in liver; PV, pulmonary vein in lung; B, bronchiole. (d) Colon inflammation and discoloration in 8-week-old S1P ₁ ^Flox Foxp3 ^Cre mice compared to S1P ₁ ^WT Foxp3 ^Cre mice. (e) Weight chart indicating lack of weight gain over time in S1P ₁ ^Flox Foxp3 ^Cre mice. (*) indicates significance, p < 0.05. (f) Elevated systemic TNF-α and IFN-γ levels in the serum of 8 weeks-old S1P ₁ ^Flox Foxp3 ^Cre mice. (*) indicates significance, p < 0.05. (g) Colorimetric assay measuring hemoglobin in adult S1P ₁ ^Flox Foxp3 ^Cre mice (*) indicates significance, p < 0.05. A total of 7 to 15 mice per group used for experiments (e–g). (h) Activation status of CD4+ or CD8+ T cells in the blood of 8-week-old S1P ₁ ^Flox Foxp3 ^Cre compared to S1P ₁ ^WT Foxp3 ^Cre mice by CD44 and CD62L staining (left) and quantified (right). 3 mice per group used.

Figure 3

S1P₁-deficient T_reg cells exit thymus but show impaired secondary lymphoid tissue egress and reduced non-lymphoid tissue localization. a) Lymphocytes from various organs of 8-week-old control S1P ₁ ^WT Foxp3 ^Cre or S1P ₁ ^Flox Foxp3 ^Cre mice were stained for CD4. T_reg (CD4⁺Foxp3⁺) distribution was analyzed by flow (representative zebra plots in (a) and quantified in (b), n = 4–7 mice), (*) indicates p < 0.05 quantified. (c) Thymus, spleen and LN from 6–8-week-old mice of indicated genotype were stained for CD45, CD3, CD4, CD25, and CD25⁺CD4⁺CD3⁺ T_reg cells gated and among them RFP⁺ or YFP⁺ T_reg cells visualized. A representative plot is provided.

Figure 4

Acute deletion of S1P₁ in T_reg cells renders mice more susceptible to EAE. (a) Treatment schematic (Top), and EAE scores after acute deletion of S1P₁ in S1P ₁ ^Flox Foxp3 ^Cre-ERT2 and control S1P ₁ ^WT Foxp3 ^Cre-ERT2 mice_. (b) Absolute number of CNS-infiltrating cells (left), and intracellular staining of CNS-infiltrating CD4⁺ T cells at the peak of disease in test and control groups as treated in (a). (c) Representative plot for percentage of T_reg cell in pLN and CNS at the peak of disease (left) and quantification of T_reg percentage (right) in CNS and pLN. (d) Th₁₇ cells in the pLN at the peak of the disease, intracellular staining of draining pLN lymphocytes 4 hours after stimulation with PMA/Ionomycin.

Figure 5

S1P₁ deletion intrinsically alters T_reg transcriptional program and promotes central to effector T_reg switch. (a) Heat Map of differentially regulated transcripts in T_reg cells purified from 8 week-old control S1P ₁ ^WT Foxp3 ^Cre and S1P ₁ ^Flox Foxp3 ^Cre mice and (b). Surface expression of indicated T_reg associated markers on YFP⁺ WT (dashed line) or S1P₁ KO (solid filled line) T_reg cells isolated from the LN of healthy female S1P ₁ ^WT Foxp3 ^Cre/WT mice as control, or S1P ₁ ^Flox Foxp3 ^Cre/WT mice as KO source (harboring 50% WT 50% S1P₁KO T_reg due to random X inactivation), and (c) quantification of YFP-Foxp3 protein expression by MFI. d) YFP⁺ WT or YFP⁺S1P₁KO T_reg cells among LN lymphocytes of healthy female S1P ₁ ^WT Foxp3 ^Cre/WT or S1P ₁ ^Flox Foxp3^Cre/WT mice gated. Higher percentages of eT_reg cells (CD44⁺CD62L^low) are observed (bar graphs in (d) bottom panel) and MFI values of CD44 and CD62L are charted for quantification purposes in (e). (f) YFP/Foxp3⁺ T_reg cells were sorted from control S1P ₁ ^WT Foxp3 ^Cre/WT or S1P ₁ ^Flox Foxp3 ^Cre/WT female mice and RNA was extracted. Relative gene expression of indicated eT_reg associated genes was quantified by real time qPCR. Higher levels of indicated genes detected in S1P₁KO T_reg cells. (g) In vitro suppression assay performed with T_reg cells sorted from 8-week-old WT or S1P ₁ ^Flox Foxp3 ^Cre mice. (*) indicates <0.05. n = 3–5 mice per group.

Figure 6

Fingolimod promotes the conversion of cT_reg cells to eT_reg cells in MS patients. (a) Percentage of CD4⁺CD45RO⁺CD25⁺CD127⁻ memory T_reg cells in healthy controls, untreated MS patients (no meds), MS patients treated with fingolimod or dimethyl fumarate. Percentage of CD103+ T_reg cells (CD4⁺CD45RO⁺CD25⁺CD127⁻) (b) and CD103⁺ T helper cells (CD4⁺CD45RO⁺CD25⁻CD127⁺) (c) in the same cohort of MS patients and healthy controls as in (a). Data are mean ± SEM of five or six donors in each group. *p < 0.05; **p < 0.01; ***p < 0.001 (ANOVA). Comparison of the number of naïve (CD45RO⁻) (d) or memory (e) T_reg cells (CD45RO⁺) in MS patients before (gray bars) and after (black bars) fingolimod treatment. (f) Increase in Memory T_reg cells in individual MS patients after fingolimod treatment. CD103 expression on T_reg (g) and T helper (h) cells from MS patients before (grey bars) and after (black bars) fingolimod treatment. Percentage of Foxp3⁺CCR4⁺ (i) and Foxp3⁺CCR4⁻ (j) T cell subsets in MS patients before (filled circle) and after (filled square) fingolimod treatment. Data are mean ± SEM of five to seven donors. *p < 0.05; **p < 0.01; ***p < 0.001 (two-tailed paired t test). All data are presented as frequency among total CD4⁺ T cells.