Differential regulation of Effector and Regulatory T cell function by Blimp1

Abstract

The transcriptional regulator Blimp1 plays crucial roles in controlling terminal differentiation in several lineages. In T cells, Blimp1 is expressed in both effector (Teff) and regulatory (Treg) cells, and mice with T cell-specific deletion of Blimp1 (Blimp1CKO mice) spontaneously develop severe intestinal inflammation, indicating a crucial role for Blimp1 in T cell homeostasis regulation. Blimp1 has been shown to function as a direct activator of the Il10 gene and although its requirement for IL10 expression has been demonstrated in both Treg and Teff cells under inflammatory conditions, the intrinsic requirement of Blimp1 for homeostatic maintenance of these T cell subsets had not been investigated. Using mice with Foxp3⁺ Treg-cell specific deletion of Blimp1 and other approaches, here we show that Foxp3⁺ Treg cell-intrinsic expression of Blimp1 is required to control Treg and Teff cells homeostasis but, unexpectedly, it is dispensable to prevent development of severe spontaneous intestinal inflammation. In addition, we show that Blimp1 controls common and unique aspects of Treg and Teff cell function by differentially regulating gene expression in these T cell subsets. These findings document previously unappreciated aspects of Blimp1's role in T cell biology and shed light on the intricate mechanisms regulating Treg and Teff cell function.

Figure 1

Blimp1 expression in CD4⁺ effector and Foxp3⁺ Treg cells under homeostatic conditions. (A) Quantitative Real time-PCR (qRT-PCR) analysis Prdm1 (Blimp1) mRNA (relative to β2-microglobulin) (left) and Western blotting (right) of CD4⁺ Foxp3^GFP+ Treg (nTreg) sorted from Ctrl Foxp3^GFP mice or in vitro differentiated Treg (iTreg,), Th1, Th17 or pathogenic (p) Th17 cells differentiated from naïve cells from the same mice (C57BL/6). (N = 3 mice/group, qPCR and N = 2 mice/sample, Western blotting). (B) FACS plot shows Prdm1 mRNA expression (as reported by YFP, Blimp1^YFP) among peripheral Treg (Foxp3⁺ as determined by intracellular staining of Foxp3 protein), effector (Foxp3⁻ CD44^high) and naïve (Foxp3⁻ CD44^low) TCRβ⁺ CD4⁺ cells in mesenteric lymph nodes (MLN, left) and spleen (SP, right). Bar graph shows the average percentage of Blimp1^YFP+ cells among TCRβ⁺ CD4⁺ Foxp3⁺, TCRβ⁺ CD4⁺ Foxp3⁻ CD44^high or TCRβ⁺ CD4⁺ Foxp3⁻ CD44^low cells. (C) FACS plots and histograms overlay show percent of Blimp1^YFP+ cells in gated TCRβ⁺ CD4⁺ Foxp3⁺ cells from thymus (THY), spleen (SP), mesenteric lymph nodes (MLN) and large (LI) intestines lamina propria (LP) from control (open histogram) and Blimp1^YFP (filled histogram) mice. Gating of Foxp3⁺ cells (as determined by intracellular staining of Foxp3 protein) is shown in FACS plots on the left. Cumulative data from several mice is shown on graph (right). (D) FACS histograms show analysis of Blimp1^YFP expression in gated TCRβ⁺ CD4⁺ Foxp3⁺ Neuropilin-1 (Nrp-1)⁺ (full line, empty histograms) and TCRβ⁺ CD4⁺ Foxp3⁺ Nrp-1⁻ (dashed line, filled histograms) cells in THY, SP, MLN and LI-LP from Blimp1^YFP mice. Lower panel shows percent of Blimp1^YFP+ cells in CD4⁺ Foxp3⁺ Nrp-1⁺ (filled circles) and CD4⁺ Foxp3⁺ Nrp-1⁻ (open circles) cells iin different organs. For all graphs, each symbol represents one mouse. (N = 2–11 mice/group). *P < 0.05 and **P < 0.01, one-way ANOVA (B) and paired t- test (C). Error bars indicate SEM.

Figure 2

Single cell qRT-PCR analysis of Prdm1 (Blimp1) expression in Treg cells. (A) Venn diagram showing frequency of Foxp3, Prdm1, Il10 and Tgfb-expressing cells in all analyzed CD4⁺ CD25^high T cells single cells. (B) C _t values of Foxp3, Prdm1, Tgfb and Il10 in all CD4⁺ CD25^high T cells analyzed. Each symbol represents one cell. (C) Violin plots showing relative expression of Prdm1 (left) and Foxp3 (right) in cells that expressed (positive) or lacked (negative) cytokines (Tgfb and or Il10) expression. Each circle indicates one cell. Yellow triangles indicate median.

Figure 3

Increased Treg numbers in Blimp1 CKO mice is a cell-intrinsic effect associated with increased proliferation and decreased cell death. (A) Expression of Prdm1 (Blimp) mRNA, (qRT-PCR, relative to β2Microglobulin) in sorted Treg (CD4⁺ Foxp3^GFP+) and Teff (CD4⁺ Foxp3⁻ CD44^high) cells from control (Prdm1 ^F/F), Foxp3^CRE CKO (Prdm1 ^F/F Foxp3^cre+), and CD4^CRE CKO (Prdm1 ^F/F CD4^CRE+) mice (B) Left panel; representative FACS plots of TCRβ⁺ CD4⁺ Foxp3⁺ cells (as determined by intracellular staining of Foxp3 protein) in large intestine lamina propria (LI-LP), mesenteric lymph nodes (MLN), spleen (SP) and thymus (THY) from control (Ctrl) or Blimp1 CKO (CD4^creCKO or Foxp3^creCKO) mice. Right graphs; percentage of Foxp3⁺ cells among TCRβ⁺ CD4⁺ cells (top) and absolute numbers of Foxp3⁺ cells (bottom) in LI-LP, MLN, SP and THY. Each symbol represents one mouse (N ≥ 4mice/group). Bars indicate average ± SEM. *P < 0.05, **P < 0.01, one-way ANOVA. (C) Top: graphs showing the average percentage ± SEM of Ki-67⁺ or Zombie (fixable viability dye)⁺ cells among CD4⁺ Foxp3⁺ (as determined by intracellular staining of Foxp3 protein) Foxp3^YFP-cre+ or CD4⁺ Foxp3⁺ Foxp3^YFP-cre- cells from Foxp3^creCKO female mice. Each circle represents one mouse. Bottom: representative FACS plots showing analysis of Ki-67⁺ or Zombie⁺ cells (N = 5 mice). **P < 0.01, paired t -test. In (A) results representative of two different experiments; In (B) cell numbers were calculated based on the total number of cells obtained from each organ and population frequency determined by FACS analysis.

Figure 4

Treg-cell specific deletion of Blimp1 leads to Teff cells accumulation. (A) Representative FACS plots (left) show CD4⁺ (TCRβ⁺ CD4⁺ Foxp3⁻ CD44^high) Teff cells in mesenteric lymphnodes (MLN) and spleen (SP) from control (Ctrl) or Blimp1 CKO (CD4^creCKO or Foxp3^creCKO) mice. Graphs show percentage (middle) and absolute number (right) of TCRβ⁺ CD4⁺ Foxp3⁻ CD44^high cells. (B) Representative FACS plots (left) of T effector CD8⁺ (TCRβ⁺ CD8⁺ Foxp3⁻ CD44^high) cells in -MLN and SP from Ctrl, CD4^creCKO or Foxp3^creCKO mice. Graphs show percentage (middle) and absolute number (right, calculated as described on Fig. 3 legend) of TCRβ⁺ CD8⁺ Foxp3⁻ CD44^high cells. Each symbol represents one mouse (N ≥ 6 mice/group). Bars indicate average ± SEM. *P < 0.05, **P < 0.01, one-way ANOVA.

Figure 5

Treg-cell intrinsic deletion of Blimp1 leads to mild intestinal inflammation. (A) Left: scatter plots showing histology score of cecum (CE), colon (CO) and rectum (RE) among control (Ctrl), CD4^creCKO and Foxp3^creCKO mice. Each symbol represents one mouse (N ≥ 5 mice/group, Bars indicate average ± SEM. *P < 0.05, **P < 0.01, one-way ANOVA. Right: representative pictures (20X magnification) of histological sections of colon of mice in each group. (B) Representative FACS plots (left) and scatter plots (right) showing the frequency of IL17A expression in TCRβ⁺ CD4⁺ Foxp3⁻ cells in spleen (SP) and large intestine lamina propria (LI-LP) from Ctrl, CD4^creCKO or Foxp3^creCKO mice after stimulation. Right scatter plots: Each symbol represents one mouse (N ≥ 4 mice/group). Bars indicate average ± SEM. **P < 0.01, one-way ANOVA. (C) Representative FACS plots (left) and Scatter plots showing the frequency (right, top) or the total numbers (right, bottom) of IL10-producing cells in TCRβ⁺ CD4⁺ Foxp3⁻ or TCRβ⁺ CD4⁺ Foxp3⁺ cells in the SP from Ctrl, CD4^creCKO or Foxp3^creCKO mice. Each symbol represents one mouse (N ≤ 6 mice/group). Bars indicate average ± SEM. *P < 0.05, **P < 0.01, one-way ANOVA.

Figure 6

Blimp1 regulates unique and commonly shared genetic programs in CD4⁺ Treg and Teff cells. (A) Schematic representation of experimental approach used to obtain allotype-marked Blimp1-sufficient (Ctrl, Thy1.1⁺) and deficient (CKO, Thy1.2⁺) Teff (CD4⁺ Foxp3⁻) and Treg (CD4⁺ Foxp³⁺) cells differentiated in the same environment in vivo. (B) Venn diagrams representing the number of deferentially expressed (DE) genes upregulated (top) and downregulated (bottom) in pTreg and Teff cells (CD4^creCKO versus Ctrl) generated from Ctrl and CD4^creCKO CD4⁺ naive T cells co-injected into RAG1^−/− mice and used for mRNA microarray analysis. Three biological replicates were used for each sample. Bar graphs under Venn diagram show fold change for selected genes in each group. (C) IPA analysis of DE genes (1.5-fold at P < 0.05): shown are comparison analysis of diseases and bio functions between common DE genes in both Foxp3^GFP+pTreg (pTreg₎ and Foxp3^GFP− CD44^high (CD4⁺ Teff) cells, DE genes in CD4⁺ Teff cells only and in pTreg cells only (−log₁₀ p value cut-off; 1.3).