Figure 1.
Hybrid Th1/2 cells develop in vivo in response to parasite infections.
(A) WT C57BL/6 or Balb/c mice were infected or immunized/challenged with the indicated pathogens. T-bet and GATA-3 expression by CD4+ T cells from indicated organs was analyzed ex vivo without restimulation. Numbers indicate frequencies. Data are representative of three H. polygyrus and three S. mansoni experiments with n = 3–4 mice each. (B) WT C57BL/6 mice were infected with H. polygyrus. T-bet and GATA-3 expression in CD4+ splenocytes was analyzed in a kinetic fashion. Mean frequencies ± SD of T-bet−GATA-3+ (Th2) and T-bet+GATA-3+ (Th1/2) cells are shown (n = 3 mice/time point). Inserted dot plot shows a typical staining on d 6. Data are representative of two independent experiments. (C) Spleen and mLN cells from WT C57BL/6 mice infected with H. polygyrus were restimulated with adult worm lysate at the peak of infection on d 21. Antigen-specific CD4+ T cells as identified by CD40L expression were analyzed for transcription factor and cytokine expression. Numbers indicate frequencies. Data are representative of three independent experiments. (D–F) WT C57BL/6 mice were infected with H. polygyrus. CD4+ splenocytes were analyzed at the peak of infection on d 21. Th2 cells were gated as CD4+T-bet−GATA-3+ and Th1/2 cells as CD4+T-bet+GATA-3+ as shown in (B). (D) Cytokine expression was analyzed upon PMA/ionomycin restimulation (n = 6 mice). Numbers indicate frequencies. Data are representative of two independent experiments. (E) Geometric mean indices ± SD of GATA-3 protein staining are shown (n = 3 mice). Data are representative of three independent experiments. (F) Spleen cells were restimulated as in (D). Geometric mean indices ± SD of cytokine expression calculated as the geometric mean of cytokine+ cells divided by the geometric mean of cytokine− cells are shown (n = 6 mice). Data are pooled from two independent experiments. (G) Expression of CXCR3 and T1/ST2 by CD4+ T cells was analyzed on day 14 of infection. Th2 cells were gated as CD4+T-bet−GATA-3+ and Th1/2 cells as CD4+T-bet+GATA-3+ as shown in (B). As a Th1 reference, the T-bet+GATA-3− population was gated as in the lower right quadrant of the inset in (B). Numbers indicate geometric means. Data are representative of two independent experiments.
Figure 2.
Simultaneous commitment of naive T cells to the Th1 and Th2 differentiation programs.
(A–E) FACS-sorted naive CD4+CD62LhiCD44loCD25−CXCR3−Gr-1− LCMV-TCRtg Th cells were activated with APCs and GP61–80 peptide in the presence of the indicated cytokines and cytokine-blocking antibodies. (A) T-bet and GATA-3 expression is shown in naive cells (top row) or on d 5 of culture (middle row). Geometric mean indices are indicated. Cytokine expression upon PMA/ionomycin restimulation is shown on d 5 of culture (lower row). Numbers indicate frequencies. Data are representative of five independent experiments. (B) Geometric mean indices ± SD of T-bet and GATA-3 stainings versus isotype control stainings on d 5 of culture are depicted as normalization to Th1 cells for T-bet or to Th2 cells for GATA-3. Data are pooled from five independent experiments. (C) Cytokine expression was analyzed after PMA/ionomycin restimulation on d 5 of culture. Mean frequencies ± SD of cytokine+ cells are shown. Data are pooled from five independent experiments. (D) Expression of the indicated cytokine and chemokine receptors (upper row) and transcription factors (lower row) was analyzed on d 5 of culture. Tfh cells were obtained from spleens of WT C57BL/6 mice and identified as CD4+CXCR5+PD-1+ cells. Numbers indicate geometric means. Data are representative of two independent experiments. (E) On d 2 of culture, levels of phosphorylated (upper row) and total (lower row) STAT proteins were determined. Numbers indicate geometric mean indices. Data are representative of two independent experiments. (F) Naive Th cells from WT Balb/c mice were cultured with anti-CD3/anti-CD28 in the presence of polarizing cytokines and cyokine-blocking antibodies to induce Th1, Th2, and hybrid Th1/2 cells. T-bet and GATA-3 expression was analyzed in a kinetic fashion and plotted as geometric mean indices. Data are representative of three independent experiments.
Figure 3.
IFN-γ– and IL-4–producing hybrid Th1/2 cells are distributed over the entire spectrum of T-bet and GATA-3 expression levels.
FACS-sorted naive LCMV-TCRtg CD4+ T cells were cultured for 5 d with APCs and GP64–80 peptide to generate hybrid Th1/2 cells. Transcription factor and cytokine co-expression was analyzed after PMA/ionomycin restimulation. Dot plots show the co-expression of IFN-γ and IL-4 within fractions expressing distinct levels of T-bet (left column) or GATA-3 (right column) that were gated according to the colored histograms on top. As very high fraction, the 10% of cells expressing the highest levels of the respective transcription factor were analyzed. The other fractions each contained approximately 30% of the total population. IFN-γ and IL-4 production in the whole population is depicted in the top center dot plot. Numbers indicate frequencies. Data are representative of two independent experiments.
Figure 4.
Induction of T-bet and IFN-γ in the presence of IL-4 critically depends on IFN-γ signaling.
(A) WT C57BL/6 mice or the indicated gene-deficient mice were infected with H. polygyrus. On d 20 of infection, CD4+ T cells from spleens were analyzed for T-bet and GATA-3 expression (n = 6–7 mice/group). The frequencies of T-bet+GATA-3+ hybrid Th1/2 cells among activated CD4+CD44+ T cells are shown. (B) The frequencies of T-bet+GATA-3+ hybrid Th1/2 cells among activated CD4+CD44+ T cells as shown in (A) in the indicated organs were normalized to the mean of the wild-type group. Data are representative of two independent experiments. (C) FACS-sorted naive CD4+CD62LhiCD44loCD25−CXCR3−Gr-1− Th cells from C57BL/6 or Ifngr−/− mice were activated with anti-CD3/anti-CD28 in the presence of the indicated cytokines and cytokine-blocking antibodies. T-bet and GATA-3 expression was analyzed on d 5 (upper row). Inserted numbers indicate geometric mean indices. Cytokine expression was analyzed upon PMA/ionomycin restimulation on d 5 (lower row). Numbers indicate frequencies. Data are representative of two independent experiments.
Figure 5.
Hybrid Th1/2 memory cells stably maintain their key transcription factor and cytokine profile.
(A–B) LCMV-TCRtg Thy1.1+ Th1, Th2, or hybrid Th1/2 cells were generated in vitro for 2 wk and transferred into WT C57BL/6 mice (n = 3–8 mice/group). Data are representative of two independent experiments. (A) T-bet and GATA-3 expression in CD4+Thy1.1+ cells from peripheral blood was analyzed in a kinetic fashion and plotted as geometric mean indices ± SD. Dot plot depicts a representative staining on d 76 after transfer. (B) Splenocytes were restimulated with GP64–80 peptide on d 42 after transfer and cytokine production of CD4+Thy1.1+ donor cells was analyzed. Numbers indicate frequencies. (C–F) WT C57BL/6 mice were infected with H. polygyrus and cured with Pyrantel at the peak of infection on d 21 (n = 10). Data are representative of two independent experiments. (C) T-bet and GATA-3 expression was determined in CD4+ cells from peripheral blood in a kinetic fashion and plotted as geometric mean indices ± SD. Dot plot (right panel) depicts a representative staining on d 30 after cure. Numbers indicate frequencies. The blue, red, and violet frames in this inset also document the gating strategy for Th1, Th2, and hybrid Th1/2 cells, respectively, which are shown in panels (C), (D), and (F). (D) Cytokine expression upon restimulation with PMA/ionomycin was analyzed in CD4+ cells from spleens of H. polygyrus–infected mice on d 63 after cure. Numbers indicate frequencies. (E) Output of H. polygyrus eggs was determined one day before and 4 d after treatment with Pyrantel. (F) Frequencies ± SD of T-bet−GATA-3+ (Th2) and T-bet+GATA-3+ (hybrid Th1/2) CD4+ cells in the peripheral blood are depicted over time after cure from H. polygyrus infection.
Figure 6.
The hybrid Th1/2 phenotype can be quantitatively modulated but resists full reprogramming into Th1 or Th2 cells.
FACS-sorted naive LCMV-TCRtg CD4+Thy1.1+ T cells were cultured for 5 d with APCs and GP64–80 peptide to generate Th1, Th2, or hybrid Th1/2 cells. (A) Hybrid Th1/2 cells were analyzed on d 5 (left) and further stimulated for 5 d in the presence of either Th1-reprogramming conditions (upper row) or Th2-polarizing conditions (lower row). Geometric mean indices of T-bet and GATA-3 expression are indicated. Cytokine expression was analyzed after PMA/ionomycin restimulation. Numbers indicate frequencies. Data are representative of two independent experiments. (B) 5×106 Th1, Th2, or hybrid Th1/2 cells were transferred into WT mice, followed by subcutaneous LCMV infection of all recipients in the footpad. CD4+Thy1.1+ donor cells were harvested from the feet during the effector phase on d 5–8 after infection. Normalized geometric mean indices of T-bet and GATA-3 expression without restimulation are shown (n = 5–7 mice/group). Data are pooled from two independent experiments. Frequencies of cytokine+ cells were analyzed after PMA/ionomycin restimulation (n = 3–5 mice/group). Data are representative of two independent experiments. (C) 1×106 Th1, Th2, or hybrid Th1/2 cells were transferred into WT C57BL/6 mice, followed by intravenous LCMV infection of all recipients. Splenocytes were analyzed on d 71. Normalized geometric mean indices of T-bet and GATA-3 expression in CD4+Thy1.1+ cells without restimulation are shown for individual mice (left panels). Frequencies of cytokine+ cells among CD4+Thy1.1+ cells after PMA/ionomycin restimulation are depicted (n = 5–6 mice/group). Data are representative of two independent experiments.
Figure 7.
Balance of opposing differentiation programs in hybrid Th1/2 cells translates into attenuated immunopathology.
FACS-sorted naive LCMV-TCRtg CD4+Thy1.1+ T cells were cultured for 5 d with APCs and GP64–80 peptide to generate Th1, Th2, or hybrid Th1/2 cells. (A–E) 5×106 cells were transferred into WT mice, followed by subcutaneous LCMV infection in one footpad and sham infection with medium of the other. (A) Footpad swelling was measured in a kinetic fashion as the difference in thickness between the infected and the sham-treated foot (n = 10–12 mice/group). Pooled data + SD from two independent experiments are shown. (B) CXCR3 expression was determined before transfer. Representative histogram and geometric means ± SD from four independent experiments are depicted. (C) Frequencies of Thy1.1+ donor cells in blood one day before viral infection are shown (n = 10–12 mice/group). (D) Absolute number of CD4+Thy1.1+ cells in challenged feet during the effector phase is depicted (n = 5–7 mice/group). (E) IFN-γ expression by CD4+Thy1.1+ cells isolated from challenged feet was determined upon PMA/ionomycin restimulation during the effector phase on d 5 after infection. Mean frequencies ± SD (n = 3–5 mice/group) are indicated. Data are pooled from (C, D) or representative of (E) two independent experiments. (F–K) 1.5×106 cells were transferred into WT mice. Mice were challenged intranasally with 10 µg GP64–80 in PBS on 4 consecutive days and analyzed 2 days later. (F) Histological analyses of the lungs. Eosinophils are stained with Sirius Red, and examples are indicated by arrows (scale bar, 20 µm; upper row). Mucus-producing goblet cells in the bronchial epithelium are identified by PAS staining, and examples are indicated by arrowheads (scale bar, 100 µm; lower row, n = 4–5 mice/group). (G) Frequency of Siglec-F+CD11clo eosinophils in the lungs of challenged mice was analyzed by FACS (n = 9–10 mice/group). (H) Quantification of PAS+/100 epithelial cells by histology is shown (n = 4–5 mice/group). (I) Absolute numbers of CD4+Thy1.1+ from the lungs were measured by FACS (n = 9–10 mice/group). (J) Frequencies of cytokine+ cells upon restimulation with PMA/ionomycin are depicted (n = 9–10 mice/group). (K) Geometric mean indices of GATA-3 expression by CD4+Thy1.1+ cells without restimulation are shown (n = 4–5 mice/group). Data are representative of (F, K) or pooled from (G–J) two independent experiments.