Blimp-1 represses CD8 T cell expression of PD-1 using a feed-forward transcriptional circuit during acute viral infection

Abstract

Programmed cell death 1 (PD-1) is an inhibitory immune receptor that regulates T cell function, yet the molecular events that control its expression are largely unknown. We show here that B lymphocyte-induced maturation protein 1 (Blimp-1)-deficient CD8 T cells fail to repress PD-1 during the early stages of CD8 T cell differentiation after acute infection with lymphocytic choriomeningitis virus (LCMV) strain Armstrong. Blimp-1 represses PD-1 through a feed-forward repressive circuit by regulating PD-1 directly and by repressing NFATc1 expression, an activator of PD-1 expression. Blimp-1 binding induces a repressive chromatin structure at the PD-1 locus, leading to the eviction of NFATc1 from its site. These data place Blimp-1 at an important phase of the CD8 T cell effector response and provide a molecular mechanism for its repression of PD-1.

Figure 1.

Blimp-1 expression represses PD-1 and is inversely expressed during ex vivo CD8 T cell activation. (A) EL4 cells were transduced with an adenovirus expression vector expressing Flag-tagged Blimp-1 or GFP and were stimulated with PMA + ionomycin for 24 h. PD-1 expression was measured by real-time RT-PCR, and results are plotted relative to the GFP adenovirus-transduced control. The Western blot (right) shows levels of Flag-tagged Blimp-1 as detected using anti-Flag antibodies compared with an actin loading control. (B) Splenic CD8 T cells were stimulated with anti-CD3/CD28 for the indicated times, and expression of PD-1, Blimp-1, and NFATc1 were measured by real-time RT-PCR. The data are normalized to 18S rRNA levels and presented relative to the unstimulated control cells. These data are representative of three independent experiments (mean ± SD). A two-tailed Student’s t test was used to determine significance. **, P < 0.01.

Figure 2.

Blimp-1 represses PD-1 expression through use of site 2 of the PD-1 upstream region. (A) Schematic representation of the PD-1 gene, potential Blimp-1 binding sites, and luciferase reporter constructs used is shown. Regulatory elements CR-C and CR-B are represented by C and B, respectively. Xs refer to the putative Blimp-1 sites mutated in each construct as represented by constructs M1, M2, and M3. (B and C) Control (pcDNA3.1) or Blimp-1 (pcDNA3.1-Flag-Blimp-1) expression vectors were co-transfected with the indicated PD-1 luciferase reporter vector (from A) into EL4 cells, stimulated with PMA/ionomycin, and assayed 24 h later for luciferase activity. Luciferase activity results were plotted relative to a pcDNA3.1 control transfection. (D) EL4 cells were transduced with adenovirus expressing GFP or Flag-Blimp-1 and stimulated with PMA/ionomycin. Expression of NFATc1, Flag-Blimp-1, and actin were analyzed by immunoblotting. Anti-Flag antibody was used to detect Blimp-1 expression. (E) EL4 cells were transduced with adenovirus expressing GFP or Flag-Blimp-1 and stimulated with PMA/ionomycin. Expression of NFATc1 message was analyzed by real-time RT-PCR, and results are presented relative to NFATc1 expression in the GFP control group. (F) EL4 cells were cotransfected with an NFATc1 luciferase reporter gene construct (containing 3.0 kb of the NFATc1 promoter) plus either an empty pcDNA3.1 vector control or the pcDNA3.1-Flag-Blimp-1 construct and stimulated with PMA/ionomycin. Luciferase activity was measured, and the results are presented relative to activity of cells transfected with the empty vector as in B. Data are representative of at least three independent experiments (mean ± SD). A two-tailed Student’s t test was used to determine significance. **, P < 0.01; ***, P < 0.001.

Figure 3.

Blimp-1 binds to site 2 of the PD-1 gene. (A) Schematic diagram of the PCR amplicons for the three potential Blimp-1 binding sites in the PD-1 promoter. (B–D) ChIP analysis of Blimp-1 binding to the PD-1 promoter was performed with chromatin from EL4 cells, splenic CD8 T cells, and LCMV antigen-specific transgenic CD8 T cells. Putative Blimp-1 binding sites shown in A were examined for each experimental setting. (B) EL4 cells were transduced with adenovirus expressing GFP or Flag-Blimp-1 and stimulated with PMA/ionomycin before ChIP analysis. Anti-Flag and control mouse monoclonal (HA,12CA5) antibodies were used for the ChIP assays as indicated. (C) Control, unstimulated (day 0, D0) CD8 T cells and anti-CD3/CD28–stimulated splenic CD8 T cells for 5 d (D5) were used for these ChIP assays. Nonspecific rabbit IgG was used as a control ChIP antibody (right). (D) P14 CD8 T cells were adoptively transferred into WT Thy1.2⁺ C57B/6 mice and analyzed 8 d after infection with LCMV Armstrong. Nonspecific rabbit IgG was used as a control ChIP antibody (right). Results of all ChIP experiments are presented as percentage of input DNA and are representative of three independent experiments (mean ± SD). A two-tailed Student’s t test was used to determine significance. **, P < 0.01; ***, P < 0.001.

Figure 4.

Blimp-1 binding results in the eviction of NFATc1 from CR-C. (A–C) ChIP-PCR was performed for Blimp-1 and NFATc1 binding to the PD-1 promoter in: EL4 cells transduced with adenovirus expressing GFP or Flag-Blimp-1 and stimulated with PMA/ionomycin for 24 h (A); unstimulated (day 0, D0) and ex vivo anti-CD3/CD28–stimulated CD8 T cells for 1 or 5 d (D1, D5; B); and P14 LCMV-specific CD8 T cells that were adoptively transferred into WT C57BL/6 mice (C). In C, mice were infected with LCMV Armstrong and chromatin was prepared on days 6 and 8 after infection. N represents naive cells. Amplicons spanning the NFATc1-binding site within CR-C and the Blimp-1–binding site (site 2) were used to detect NFATc1 and Blimp-1 binding, respectively. The results were presented as percentage of input DNA. For all ChIP assays in this figure, the mouse HA (12CA5 monoclonal antibody) or nonspecific rabbit IgG was used as controls to match the species and type of specific antibodies. (D–F) EL4 cells were transduced with adenovirus vectors expressing an NFATc1-GFP chimeric protein or GFP alone, stimulated with PMA/ionomycin for 24 h and then transduced again with adenovirus vectors expressing either Flag-tagged Blimp-1 or GFP. After a second 24-h period, the cultures were harvested and examined in D by ChIP for NFATc1/NFATc1-GFP (anti-NFATc1 or anti-GPF antibody) or Flag-Blimp-1 (anti-Flag antibody) binding as above (column numbers are for reference to the text); in E for expression of PD-1 mRNA by real-time RT-PCR; and in F by immunoblotting for expression of the transduced genes with anti-GFP and anti-Flag antibodies. Data are representative of at least three independent experiments (mean ± SD). A two-tailed Student’s t test was used to determine significance. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5.

Blimp-1 expression induces the formation of repressive chromatin structure at the PD-1 locus. (A) Real-time PCR-based DNase I hypersensitivity assays using increasing amounts of DNase I were performed on the nuclei isolated from EL4 cells transduced with the indicated adenovirus vectors as above. A control, insensitive region was used to normalize data between experiments as described in the materials and methods. The data were plotted as fold sensitivity relative to samples that were not treated with DNase I. (B and C) ChIP assays were conducted for active and repressive histone modifications on GFP or Flag-Blimp-1 adenovirus-transduced EL4 cells as described above (B) or unstimulated (D0) and ex vivo anti-CD3/CD28 bead-stimulated splenic CD8 T cells for 1 d (D1) or 5 d (D5; C). H3Ac represents histone H3 lysine acetylated at lysines 9 and 14. PCR amplicons for CR-B, CR-C, and a control region (con) from the PD-1 gene that has low levels of modifications are indicated. ChIP results were presented as a percentage of input DNA and were representative of three independent experiments (mean ± SD). A two-tailed Student’s t test was used to determine the significance between control and experimental samples. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 6.

PD-1 expression is increased on Blimp-1–deficient CD8 T cells. (A and B) Real-time RT-PCR analysis was performed to determine the expression of PD-1, Blimp-1, NFATc1, and Cre mRNA in anti-CD3/CD28-stimulated primary CD8 T cells from WT and Blimp-1 conditional KO (cKO) mice (A), or FACS-sorted splenic CD8 T cells from day 8 LCMV Armstrong-infected WT and cKO mice (B). The data were plotted relative to the expression of the gene being assayed from unstimulated cells in WT C57Bl/6 mice. (C) ChIP-PCR was performed for the binding of Blimp-1 and NFATc1 to the PD-1 promoter in FACS-sorted CD8 T cells from day 8 LCMV Armstrong-infected WT and Blimp-1 cKO mice as in Fig. 4. (D) ChIP assays were performed for H3K4me3 and H3K27me3 modifications in FACS-sorted CD8 T cells from day 8 LCMV Armstrong-infected WT and Blimp-1 cKO mice. (E) Splenic CD8 T cells from WT or Blimp-1 cKO mice were plated, activated with ex vivo anti-CD3/CD28 beads for 24 h, and left untreated or treated with 1 mg/ml CsA for the remainder of the time course. PD-1 mRNA was measured by real-time RT-PCR, and results were normalized to 18S rRNA and plotted relative to the expression in unstimulated WT CD8 T cells. Data were representative of three independent experiments (mean ± SD). A two-tailed Student’s t test was used to determine significance. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 7.

Transient PD-1 expression on virus-specific CD8 T cells is extended in Blimp-1 cKO mice. (A) WT and Blimp-1 cKO mice were infected with LCMV Armstrong, and the splenic CD8 T cells were analyzed by flow cytometry on day 8 after infection. Representative gating strategy for endogenous naive (CD44lo) and LCMV antigen-specific (GP33, GP276, and NP276) effector CD8 T cell splenocytes. (B) Histogram analysis of PD-1 expression on day 8 GP33, GP276, and NP396-specific effector CD8 T cells versus endogenous naive CD8 T cells from the above WT or Blimp-1 cKO mice. PD-1 mean fluorescence intensity (MFI) on antigen-specific CD8 T cells was listed in each panel. The experiment presented (n = 3 WT and 3 Blimp-1 cKO mice) was representative of two independent experiments. (C–E) 10⁴ Thy1.1⁺ WT naive P14 CD8 T cells were adoptively transferred into WT and Blimp-1 cKO mice 24 h before infection. Chimeric mice were acutely infected with LCMV Armstrong. (C) Histogram analysis of PD-1 expression was performed on transferred P14 and endogenous, antigen-specific CD8 T cells from day 8 WT chimeric mice (blue line) and Blimp-1 cKO chimeric mice (red line). (D) Summary graph of PD-1 MFI on the endogenous WT and Blimp-1 cKO GP276 and NP396-specific effector CD8 T cells compared with endogenous naive cells. These data are representative of two independent experiments with n = 3–4 WT and 3–5 Blimp-1 cKO mice. Data were plotted with the SD of the mean shown. (E) A pilot longitudinal analysis of PD-1 expression (MFI) was performed on endogenous antigen-specific and naive CD8 T cells from the peripheral blood mononuclear cells. This experiment was performed once with an n = 5 WT mice and 8 Blimp-1 cKO mice. Tet+, tetramer positive cells. A two-tailed Student’s t test was used to determine significance. *, P < 0.05; **, P < 0.01.

Figure 8.

Elevated PD-1 expression on Blimp-1–deficient CD8 T cells occurs in mixed bone marrow chimera mice. Bone marrow cells from WT (Ly5.1) and cKO (Ly5.2) mice were depleted of T cells, mixed at a 1:1 or 1:4 ratio, and transferred into lethally irradiated WT (Ly5.1) hosts. Recipient mice were rested for 6 wk to reconstitute the hematopoietic cells and acutely infected with LCMV Armstrong. At day 8 after infection, chimeric mice were sacrificed and PD-1 expression on antigen-specific splenic CD8 T cells was analyzed. (A) Flow cytometry gating on the CD8 T cell population was performed for analysis of LCMV tetramer-positive GP33, GP276, and NP396 CD8 T cells from WT and Blimp-1 cKO mice. (B) Histogram analysis of PD-1 expression from WT and Blimp1 cKO antigen-specific CD8 T cells. (C) Graph of PD-1 MFI on corresponding antigen-specific CD8 T cells. Significance was determined by one-tailed paired Student’s t test. These data are representative of one of three independent experiments performed, each derived from n = 3–5 chimeric mice. *, P < 0.05; **, P < 0.01.