ZEB1, ZEB2, and the miR-200 family form a counterregulatory network to regulate CD8+ T cell fates

Abstract

Long-term immunity depends partly on the establishment of memory CD8⁺ T cells. We identified a counterregulatory network between the homologous transcription factors ZEB1 and ZEB2 and the miR-200 microRNA family, which modulates effector CD8⁺ T cell fates. Unexpectedly, Zeb1 and Zeb2 had reciprocal expression patterns and were functionally uncoupled in CD8⁺ T cells. ZEB2 promoted terminal differentiation, whereas ZEB1 was critical for memory T cell survival and function. Interestingly, the transforming growth factor β (TGF-β) and miR-200 family members, which counterregulate the coordinated expression of Zeb1 and Zeb2 during the epithelial-to-mesenchymal transition, inversely regulated Zeb1 and Zeb2 expression in CD8⁺ T cells. TGF-β induced and sustained Zeb1 expression in maturing memory CD8⁺ T cells. Meanwhile, both TGF-β and miR-200 family members selectively inhibited Zeb2. Additionally, the miR-200 family was necessary for optimal memory CD8⁺ T cell formation. These data outline a previously unknown genetic pathway in CD8⁺ T cells that controls effector and memory cell fate decisions.

Figure 1.

Zeb1 and Zeb2 are reciprocally expressed in CD8⁺ T cells during LCMV infection. (a) Zeb1 and Zeb2 mRNA was measured in purified CD44^lo CD62L^hi naive CD8⁺ T cells (day 0) or D^bGP_33-41⁺ LCMV-specific CD8⁺ T cells from 8 and 80 dpi using quantitative RT-PCR. LCMV-specific CD8⁺ T cells were purified based on D^bGP_33–41-tetramer staining. (b and c) Effector P14⁺ CD8⁺ T cells were isolated 8 dpi based on the expression of KLRG1^hi IL-7R^lo (TE) and KLRG1^lo IL-7R^hi (MP) and Zeb1 and Zeb2 mRNA was measured using quantitative RT-PCR (b) or ZEB1 protein expression was measured using Western blotting (c; β-actin was used as a loading control). (d) Zeb1 mRNA expression was compared between WT and Zeb2^f/f GzmBCre⁺ P14⁺ CD8⁺ T cells at 8 dpi. (e) Zeb2 mRNA expression was compared between WT and Zeb1^f/f GzmBCre⁺ P14⁺ CD8⁺ T cells 30 dpi. (f) ZEB1 ChIP-quantitative PCR was performed on purified CD44^lo CD62L^hi naive CD8⁺ T cells using anti–ZEB1 (black bars) or IgG control (white bars) antibodies and primers to Zeb2 promoter (−550 bp), exon 8 and Il2 promoter. Data shown are representative of two (c) independent or cumulative of three (a, b, d, and e) independent experiments; n = 3–4 mice/group/experiment (c), n = 9–12 mice/group (a, b, d, and e). Data are expressed as mean ± SEM. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.

Figure 2.

ZEB1 promotes the survival of memory CD8⁺ T cells. (a) WT (filled circle) and Zeb1^f/f GzmBCre⁺ (open circle) mice were infected with LCMV-Arm, and splenic D^bGP_33-41 and D^bNP_396-404 tetramer⁺ CD8⁺ T cells were quantified at 8, 15, and 60 dpi. (b) Representative FACS plots of D^bGP_33-41 and D^bNP_396-404 tetramer⁺ CD8⁺ T cells at 8 and 60 dpi in WT and Zeb1^f/f GzmBCre⁺ mice (percentage of tetramer⁺ CD8⁺ T cells ± SEM is indicated). (c–e) Graphs are gated on D^bGP_33-41-tetramer⁺ CD8⁺ T cells from WT or Zeb1^f/f GzmBCre⁺ mice at 30 dpi and show the ratio of BCL-2/BIM expression (c), percentage of Annexin V⁺ and 7-AAD⁺ cells (d), or percentage of Ki67⁺ cells (e) based on flow cytometry. (f) Expression of KLRG1 and IL-7R in D^bGP_33-41 tetramer⁺ CD8⁺ T cells at 8 and 60 dpi in WT and Zeb1^f/f GzmBCre⁺ mice based on flow cytometry. (g–j) WT and Zeb1^f/f GzmBCre⁺ CD8⁺ T cells from 30 dpi were analyzed for production of IFNγ, TNFα, or IL-2 as indicated using intracellular cytokine staining (g) or expression of CD62L (together with absolute number of T_CM and T_EM cells; h), CD122 (i), and EOMES (j). Data shown are representative of five (b and f) or two (d) independent or cumulative of three (a) or five (c, e, and g–j) independent experiments; n = 3–5 mice per group per experiment (b, d, and f), n = 8–14 mice/group (a), amd n = 15–25 mice/group (c, e, and g–j). Data are expressed as mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 3.

ZEB1 is necessary for memory CD8⁺ T cell protective immunity. 100,000 GP_33–41–specific WT or Zeb1^f/f GzmBCre⁺ memory CD8⁺ T cells (Thy1.2/1.2) were transferred into congenically mismatched (Thy1.1/1.1) naive B6 mice that were then infected with recombinant L. monocytogenes expressing the GP_33-41 epitope (LM-33). (a and b) At day 4 after challenge, the frequency and numbers of recalled GP_33–41–specific WT or Zeb1^f/f GzmBCre⁺ CD8⁺ T cells were analyzed in the spleen. (c) LM-33 bacterial titers (CFUs) in the liver and spleen were determined at day 3 after challenge. (d) Histogram shows the expression of Granzyme B in GP_33–41–specific WT or Zeb1^f/f GzmBCre⁺ CD8⁺ T cells day 4 after LM-33 infection. Data shown are representative of two (a and d) or cumulative of two (b and c) independent experiments; n = 3–5 mice/group/experiment (a and d), n = 6–10 mice/group (b and c). Data are expressed as mean ± SEM. *, P < 0.05; **, P < 0.01.

Figure 4.

Zeb1 and Zeb2 are inversely regulated by TGF-β. (a and b) Zeb1 and Zeb2 mRNA expression in naive P14⁺ CD8⁺ T cells or those stimulated (stim) with GP_33–41 peptide for 3 d followed by 2-d culture alone or in the presence of the indicated cytokines (IL-2, IL-15, and TGF-β). (c and d) 50,000 WT or Tgfbr2^f/f LckCre⁺ P14⁺ CD8⁺ T cells were transferred to naive B6 mice followed by LCMV-Arm infection. 45 dpi, the donor cells were purified using FACS and the amount of Zeb1 and Zeb2 mRNA was measured by quantitative RT-PCR. Data shown are cumulative of two (a–d) independent experiments; n = 8 mice (a and b), n = 6–10 mice/group (c and d). Data are expressed as mean ± SEM. *, P < 0.05; **, P < 0.01.

Figure 5.

miR-200 family directly represses Zeb2 but not Zeb1 mRNA in CD8⁺ T cells. (a) Schematic representation of Zeb1 and Zeb2 3′ UTRs (top) showing the miR-200 family seed sequences (dark and light green) and both clusters of miR-200 family genes (bottom) showing the corresponding conserved ZEB-binding sites (orange). (b) Naive (d0) and LCMV-specific effector (8 dpi) and memory (30 dpi) CD8⁺ T cells were sorted and measured for miR-200 family microRNA expression using quantitative RT-PCR. (c) LCMV-specific CD8⁺ T cells were sorted 12 dpi followed by UV cross-linking and immunoprecipitating (IP) with pan-Ago (black), Ago2 (light gray), or IgG control (dark gray) antibodies. Enrichment of specific miR-200a/141 or miR-200b/c/429 binding regions within Zeb2 (left) or Zeb1 (right) 3′ UTR over input was measured by site-specific primers using quantitative RT-PCR. (d) P14⁺ CD8⁺ T cells transduced with RVs designed to overexpress (OE) miR-200a (white bar) or miR-200c (gray bar) were sorted at 8 dpi, and Zeb1 and Zeb2 mRNA was measured by quantitative RT-PCR. (e) LCMV-specific CD8⁺ T cells were purified from WT (black bar) and miR-200^−/− KO (white bar) mice at 30 dpi, and Zeb1 and Zeb2 mRNA was measured by quantitative RT-PCR. Data shown are cumulative of two (b, c, and e) or three (d) independent experiments; n = 6–10 mice/group (b, c, and e), n = 9–15 mice/group (d). Data are expressed as mean ± SEM. **, P < 0.01.

Figure 6.

Overexpression of miR-200 family impairs TE and promotes memory cell formation. As in Figure 5, small numbers of P14⁺ CD8⁺ T cells were transduced with miR-200a (light gray open circle) or miR-200c (dark gray open square) RVs or empty vector control (black filled circle) RVs and transferred into B6 mice that were subsequently infected with LCMV-Arm. (a) Line plots show total numbers of donor RV-transduced P14⁺ CD8⁺ T cells (left) or subsets of KLRG1^hi IL-7R^lo cells (middle) or KLRG1^lo IL-7R^hi cells (right) at 8, 30 and 45 dpi. (b and c) Flow plots show expression of KLRG1 and IL-7R (b) CD62L and CD27 (c) in control, miR-200a and miR-200c OE P14⁺ CD8⁺ T cells at 8 and 30 dpi. (d and e) Bar graphs show amounts of EOMES and T-BET in donor P14⁺ CD8⁺ T cells at 30 dpi (d) or production of IFNγ, TNFα or IL-2 as indicated using intracellular cytokine staining (e; note that IL-2–producing cells were gated on IFNγ⁺ TNFα⁺ CD8⁺ T cells). See Fig. S3 for additional data on effects of miR-200b, miR-429, and miR-141 OE in LCMV-specific CD8⁺ T cells. Data shown are representative of five (b and c) or cumulative of two (a) or three (d and e) independent experiments; n = 3–5 mice/group/experiment (b and c), n = 6–8 mice/group (a), n = 8–10 (d and e). Data are expressed as mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 7.

miR-200 family deficiency results in the loss of memory CD8⁺ T cells. (a) WT (black bar), miR-200^+/− (HET, gray bar), and miR-200^−/− (KO, white bar) mice were infected with LCMV Arm and splenic D^bGP_33-41 and D^bNP_396-404 tetramer⁺ CD8⁺ T cells were quantitated at 45 dpi. (b) Representative contour plots of D^bGP_33-41 and D^bNP_396-404 tetramer⁺ CD8⁺ T cells at 45 dpi in WT, miR-200^+/− HET, and miR-200^−/− KO mice. (c and d) Representative flow cytometry data measuring KLRG1 and IL-7R expression (c) and CD62L and CD27 expression (d) in WT, miR-200^+/− and miR-200^−/− LCMV-specific CD8⁺ T cells 45 dpi. (e) WT (black bar), miR-200^+/− (HET, gray bar), and miR-200^−/− (KO, white bar) CD8⁺ T cells from 45 dpi were analyzed for IFNγ and TNFα (top two bar graphs) or IL-2 (bottom bar graph) expression using intracellular cytokine staining after a 5-h GP_33–41 peptide stimulation. Note that IL-2 producing cells were gated on IFNγ⁺ TNFα⁺ CD8⁺ T cells. Data shown are representative of two (b–d) or cumulative of two (a and e) independent experiments; n = 2–3 mice/group/experiment (b–d), n = 5–6 mice/group (a and e). Data are expressed as mean ± SEM. *, P < 0.05; **, P < 0.01.