MicroRNAs of the miR-17∼92 family are critical regulators of T(FH) differentiation

Abstract

Follicular helper T cells (T(FH) cells) provide critical help to B cells during humoral immune responses. Here we report that mice with T cell-specific deletion of the miR-17∼92 family of microRNAs (miRNAs) had substantially compromised T(FH) differentiation, germinal-center formation and antibody responses and failed to control chronic viral infection. Conversely, mice with T cell-specific expression of a transgene encoding miR-17∼92 spontaneously accumulated T(FH) cells and developed a fatal immunopathology. Mechanistically, the miR-17∼92 family controlled the migration of CD4(+) T cells into B cell follicles by regulating signaling intensity from the inducible costimulator ICOS and kinase PI(3)K by suppressing expression of the phosphatase PHLPP2. Our findings demonstrate an essential role for the miR-17∼92 family in T(FH) differentiation and establish PHLPP2 as an important mediator of their function in this process.

Figure 1

miR-17~92 family miRNAs regulate T_FH differentiation during protein antigen immunization. (a–c) Expression of miR-17~92 family miRNAs in sorted naive CD4⁺ T and CXCR5^hiPD1^hi T_FH cells (a), naive and activated WT CD4⁺ T cells (b), and CD4tKO CD4⁺ T cells (c, n=3) was determined by Northern blot. Numbers indicate miRNA/U6 ratios normalized to naive CD4⁺ T cells. (d,e) CXCR5^hiPD1^hi T_FH (d) and FAS⁺GL-7⁺ GCB cells (e) were analyzed at days 7 and 14 after intraperitoneal (i.p) immunization with NP-OVA+Alum+LPS (n=6 per group). (f) WT and CD4tKO mice were immunized with 10μg NP-CGG+Alum (i.p.), followed by secondary immunization with 5μg NP-CGG (i.p.) on day 111 after primary immunization. Serum was collected at indicated time points. NP-specific IgG1 antibody concentration was determined by ELISA (n=8 per group). (g) Immunohistochemistry analysis of germinal centers. Spleens of NP-OVA+Alum+LPS immunized mice were examined at day 7 after immunization (i.p). Scale bar, 50 μm. (h) Lethally irradiated B6 Ly5a mice were reconstituted with B6 Ly5a/b and tKO Ly5b bone marrow cells (1:1), immunized with NP-OVA+Alum+LPS (i.p) at eight weeks after reconstitution, and T_FH differentiation in the spleen was analyzed at days 7 or 14 after immunization. Data are representative of two (a) and three (b) independent experiments. All graphs are shown as means± s.e.m. *, p < 0.05; **, p < 0.01.

Figure 2

miR-17~92 family miRNAs regulate T_FH differentiation during chronic viral infection. (a–e) CXCR5^hiPD1^hi (a) and CXCR5^hiBcl-6⁺ (b) T_FH cells, FAS⁺GL-7⁺ GCB cells (c), and IL-21 producing cells (e) were analyzed in spleen of mice at day 30 after LCMV clone-13 infection (a–c, n=8 per group; e, n=4~5 per group). (d) anti-LCMV IgG concentration in serum was determined by ELISA. (f,g) LCMV clone-13 infected mice were bled at the indicated times (f), euthanized at day 140 post-infection (g), and viral titers in serum and tissues were measured by plaque assay (representatives of two independent experiments). All graphs are shown as means± s.e.m. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 3

Spontaneous accumulation of T_FH cells in T cell-specific miR-17~92 transgenic mice. (a) Expression levels of miR-17~92 family miRNAs in TG CD4⁺ T cells (n=3) were determined by Northern blot. Numbers indicate miRNA/U6 ratios normalized to naive WT CD4⁺ T cells. (b) CXCR5^hiPD1^hi T_FH (upper), CXCR5^hi Bcl-6⁺ T_FH (middle) and FAS⁺GL-7⁺ GCB cells (lower) were analyzed in the spleen of 6~8 week old non-immunized mice (n=5~7 per group). (c) Survival curves of TG (n=35) and WT mice (n=24). (d) T_FH (upper) and GCB (lower) cells in splenocytes of 8 month old WT and TG mice. (e) Serum anti-dsDNA antibody concentration in TG and WT mice at 6–8 months of age was determined by ELISA. The serum titers of aged MRL-lpr/lpr mice were arbitrary set at 1x10³. (f) Lethally irradiated B6.Rag1^−/− mice were reconstituted with B6 (Ly5a) and WT or TG (Ly5b) mixed bone marrow cells (1:1), and spontaneous T_FH cell differentiation in the spleen was analyzed at 3 months after reconstitution (n=3 per group). (g) WT naive CD4⁺ T cells were transduced with RV-miR-17~92 or RV-control and activated in vitro in the presence of anti-IL-4, anti-IFNγ, IL-6 and IL-21 for T_FH (upper panel) or anti-IL-4 and IL-12 for T_H1 (lower panel) differentiation. GFP⁺ cells were gated and plotted (representatives of three independent experiments). Bcl-6 expression in in vitro differentiated T_FH cells is presented in Supplementary Fig. 5. All graphs are shown as means± s.e.m. *, p < 0.05; **, p < 0.01.

Figure 4

The effect of miR-17~92 family miRNAs on antigen-specific CD4⁺ T cells during in vivo T_FH differentiation. (a–d) CFSE labeled or unlabeled Ly5b⁺ WT or CD4tKO naive OT-II CD4⁺ T cells were transferred into B6.Ly5a mice, which were subsequently immunized with OVA+Alum subcutaneously. T_FH differentiation (a), cell proliferation (b), expression of activation markers (c), and localization (d) of adoptively transferred Ly5b⁺ OT-II CD4⁺ T cells in inguinal lymph nodes were analyzed by flow cytometry (a–c) or immunohistochemistry (d) at indicated times (3–5 mice per group). IgD⁺ area, IgD⁻ area, and * indicate B cell follicle, T cell zone, and germinal center, respectively. Scale bar, 100 μm (d).

Figure 5

MiR-17~92 family miRNAs specifically regulate T_FH differentiation beyond the initial phase of T cell activation. (a) Expression of miR-17~92 family miRNAs in naive and in vitro activated WT and OX40tKO CD4⁺ T cells were determined by Northern blot (representative of two independent experiments). Numbers indicate miRNA/U6 ratios normalized to naive WT CD4⁺ T cells. (b–c) CXCR5^hiBcl-6⁺ T_FH cells in the spleen were analyzed at day 7 after intraperitoneal (i.p) immunization with OVA+Alum+LPS (n=4~10 per group). Representative flow cytometry plots are shown in (b) and combined bar graphs for T_FH are shown in (c). (d–e) CXCR5^hiBcl-6⁺ T_FH (d) and IFN-γ⁺ T_H1 cells (e) were analyzed at day 8 after LCMV Armstrong infection (n=4~5 per group). Splenocytes were re-stimulated with gp61 peptide loaded APCs to detect antigen specific IFN-γ⁺ T_H1 cells. All graphs are shown as means± s.e.m. *, p < 0.01; n.s., not significant (p >0.05).

Figure 6

Deletion of one copy of the Pten gene in CD4tKO mice partially restores T_FH differentiation. (a) The expression of Pten in WT or CD4tKO naive CD4⁺ T cells was examined by immunoblot (upper panel). The lower panel summarizes Pten/β-actin ratios (n=4 per group). (b) WT or CD4tKO naive CD4⁺ T cells were stimulated wtih anti-CD3+CD28 for indicated amounts of time and the PI3K pathway was analyzed by immunoblot (representative of three independent experiments). (c) The expression of Pten in WT, CD4tKO, or CD4tKO;Pten^fl/+ naive CD4⁺ T cells was examined by Western blot (left panel). The right panel summarizes Pten/β-actin ratios (n=4 per group). (d–e) Flow cytometry analysis of CXCR5^hi Bcl-6⁺ T_FH (d) and FAS⁺GL-7⁺ GCB cells (e) in mice of indicated genotypes at day 7 after i.p. immunization with OVA+Alum+LPS (n=9~12 per group). (f) NP-specific IgG1 and IgM antibody concentration was determined by ELISA on days 7, 14 and 21 after i.p immunization with 10 μg NP-CGG+Alum (n=7~8 per group). All graphs are shown as means± s.e.m. *, p < 0.05; **, p < 0.01.

Figure 7

MiR-17~92 family miRNAs regulate T_FH differentiation through the ICOS-PI3K pathway by suppressing the expression of Phlpp2. (a) Predicted miR-17~92 binding sites in the Phlpp2 3′UTR. (b) Luciferase reporter assay was performed with WT or mutated R1 and R2 regions and normalized with firefly luciferase activity (Rluc/Fluc). (c) naive CD4⁺ T cells were activated with anti-CD3+CD28 and Phlpp2 protein was examined by immunoblot. Rest., CD4⁺ T cells were rested for 2 days in the presence of IL-2 after anti-CD3+CD28 stimulation for 3 days. (d) Ly5b⁺ WT or CD4tKO OT-II CD4⁺ T cells were transferred into B6.Ly5a mice. At day 3.5 after OVA+Alum immunization, OT-II CD4⁺ T cells were examined ex vivo by intracellular staining of pS6 (n=3~6 mice per group). (e) ICOS-induced pAkt and pS6, and Phlpp2 were examined in pre-activated WT or CD4tKO CD4⁺ T cells by immunoblot. (f) Control (RV-CTL) or Phlpp2 shRNA encoding (RV-P2sh) retrovirus transduced WT or CD4tKO OT-II CD4⁺ T cells were adoptively transferred into WT recipient mice, followed by OVA+Alum immunization. CXCR5^hiBcl-6⁺ T_FH cells were analyzed at day 6~6.5 post immunization (n=5~6 per group). (g–h) RV-CTL, RV-P2sh, or RV-miR-17~92 transduced Ly5b⁺ CD4tKO OT-II cells (3x10⁶) (g), or RV-CTL, RV-phlpp2 transduced Ly5b⁺ WT OT-II cells (2x10⁵) (h) were transferred into Ly5a mice. On day 4 after OVA+Alum immunization, draining lymph nodes were analyzed by immunohistochemistry. Images are representatives of two independent experiments (3~4 mice per group for each experiment). Scale bar, 70 μm (g–h). All graphs are shown as means± s.e.m. *, p < 0.05; **, p < 0.01.