Batf3 deficiency reveals a critical role for CD8alpha+ dendritic cells in cytotoxic T cell immunity

Abstract

Although in vitro observations suggest that cross-presentation of antigens is mediated primarily by CD8alpha+ dendritic cells, in vivo analysis has been hampered by the lack of systems that selectively eliminate this cell lineage. We show that deletion of the transcription factor Batf3 ablated development of CD8alpha+ dendritic cells, allowing us to examine their role in immunity in vivo. Dendritic cells from Batf3-/- mice were defective in cross-presentation, and Batf3-/- mice lacked virus-specific CD8+ T cell responses to West Nile virus. Importantly, rejection of highly immunogenic syngeneic tumors was impaired in Batf3-/- mice. These results suggest an important role for CD8alpha+ dendritic cells and cross-presentation in responses to viruses and in tumor rejection.

Figures

Fig. 1. Batf3-/- mice selectively lack the CD8α+ DC subset

(A) Splenocytes from Batf3+/+ (+/+) or Batf3-/- (-/-) mice were stained for CD11c, CD8α and DEC205. Left panels are gated on live cells. Numbers indicate the percentage of splenocytes within the CD11chiCD8α+ gate. Right panels are gated on CD11chi cells. (B) Splenocytes were depleted of B220+ B cells and Thy1.2+ T cells and positively selected for CD11c expression by antibody coated magnetic beads (MACS). Cells were then stained for CD11c, CD11b, and either CD8α and CD4 or CD8α and CD24, and analyzed by FACS. Numbers represent the percentage of cells within the indicated gates. (C) Lymph node cells pooled from cervical, axillary and inguinal lymph nodes and depleted of Thy1.2+ T cells, or light density cells of the thymus were stained for CD11c, CD45RA CD8α, DEC205 or Sirp-α. Plots are gated on the indicated populations.

Fig. 2. Functional loss of CD8α+ cDCs in Batf3-/- mice is cell-intrinsic to the hematopoietic system

(A) Frozen sections from Batf3+/+ (+/+) or Batf3-/- (-/-) mice were stained for B220 (green) and SIGN-R1 (red) expression (left panels) or for B220 (green) and MOMA-1 (red) (right panels). (B) Irradiated F1(B6.SJL/129SvEv) mice (CD45.1+ CD45.2+) were reconstituted with 2 ×107 bone marrow cells from Batf3+/+ (+/+) or Batf3-/- (-/-) CD45.1-CD45.2+ mice. After 10 weeks, donor cells (CD45.1- CD45.2+) were analyzed for CD11c, CD8α, CD4 and CD24 expression. Shown are plots for CD8α and CD4 (left panels) or CD8α and CD24 (right panels) gated on CD11chi donor-derived cells. Numbers represent the percentage of cells within the indicated gates. (C) Batf3+/+ (+/+) or Batf3-/- (-/-) mice were treated i.p. with 10μg FL-Fc. After 10 days, splenocytes were enriched for CD11c+ by MACS and stained for CD11c, CD8α and B220. Plots are gated on live cells (left) or CD11cintCD8α+ cells (right). Numbers represent the percentage of cells within the indicated gates. (D) Batf3+/+ (+/+) or Batf3-/- (-/-) BM cells were cultured in FL (20 ng/ml) for 9 days, and non-adherent cells analyzed for CD11c, CD45RA, CD24 and Sirp-α expression. Plots are gated on live cells (left) or CD11c+ CD45RA- cells (right).

Fig. 3. Lack of cross-presentation and antiviral CTL responses in Batf3-/- mice

(A) Batf3+/+ (+/+) or Batf3-/- (-/-) splenocytes were depleted of B220+ B cells and Thy1.2+ T cells and enriched for CD11c by MACS and cultured with irradiated MHC-class I-/- splenocytes as indicated that were either untreated (-ovalbumin), pulsed with 10 mg/ml soluble ovalbumin (+ovalbumin), or cultured with 1 μM SIINFEKL peptide. CFSE-labeled CD45.1+ OT-I T cells were cultured with these cells and proliferation determined by FACS after 60 hours. Single-color histograms of CD8+CD45.1+ OT-I T cells show the percentage of cells in the indicated gates. (B) Batf3+/+ (+/+) or Batf3-/- (-/-) mice were infected with 100 PFU of WNV. On day 7, isotype-specific anti-WNV E protein titers were measured. (C) Batf3+/+ (+/+) or Batf3-/- (-/-) mice were infected with 100 PFU of WNV, or left uninfected. After 7 days, splenocytes were stimulated in vitro with the WNV-specific NS4B peptide (P33), OVA peptide, or PMA/ionomycin as described. CD8+ T cells were analyzed for expression of intracellular IFN-γ. Data shown are the mean ± SEM (n=9-10). (D) Batf3+/+ (+/+) or Batf3-/- (-/-) CD8+ T cells were transferred i.v. into Rag2-/- recipients. After 24h, mice were infected with 100 PFU of WNV (+ WNV) or left uninfected (-WNV). After 7 days, splenocytes were harvested and analyzed as described in (C). Data shown are the mean ± SEM (n=6). Three independently performed experiments yielded similar results.

Fig. 4. Lack of tumor rejection in Batf3-/- mice

(A) 106 H31m1 fibrosarcoma cells were injected subcutaneously into Batf3+/+ (closed circles), Batf3-/- (open circles), or Rag2-/- (closed triangles) mice and tumor diameter (± SD) (n=10) was measured. (B) Mice were treated as in (A). After 9 days, splenocytes were harvested and cocultured with IFN-γ pre-treated, irradiated H31m1 tumor cells. After 5 days, a CTL killing assay using 51Cr- labeled H31m1 or 1773 tumor cells as target cells was performed. Shown is specific killing activity as described in the Methods. (C) Tumors and spleens from mice treated as in (A) were removed on day 11 and cells analyzed by FACS. Plots are gated on live CD45.2+ cells and show CD3, CD8α and CD4 expression. Numbers represent the percentage of cells within the indicated gate. Results are representative of at least three mice per group.