NKp46+ Innate Lymphoid Cells Dampen Vaginal CD8 T Cell Responses following Local Immunization with a Cholera Toxin-Based Vaccine

Abstract

Innate and adaptive immune cells work in concert to generate efficient protection at mucosal surface. Vaginal mucosa is an epithelial tissue that contains innate and adaptive immune effector cells. Our previous studies demonstrated that vaginal administration of Cholera toxin -based vaccines generate antigen-specific CD8 T cells through the stimulation of local dendritic cells (DC). Innate lymphoid cells (ILC) are a group of lymphocytes localized in epithelial tissues that have important immune functions against pathogens and in tissue homeostasis. Their contribution to vaccine-induced mucosal T cell responses is an important issue for the design of protective vaccines. We report here that the vaginal mucosa contains a heterogeneous population of NKp46+ ILC that includes conventional NK cells and ILC1-like cells. We show that vaginal NKp46+ ILC dampen vaccine-induced CD8 T cell responses generated after local immunization. Indeed, in vivo depletion of NKp46+ ILC with anti-NK1.1 antibody or NKG2D blockade increases the magnitude of vaginal OVA-specific CD8 T cells. Furthermore, such treatments also increase the number of DC in the vagina. NKG2D ligands being expressed by vaginal DC but not by CD8 T cells, these results support that NKp46+ ILC limit mucosal CD8 T cell responses indirectly through the NKG2D-dependent elimination of vaginal DC. Our data reveal an unappreciated role of NKp46+ ILC in the regulation of mucosal CD8 T cell responses.

Fig 1. Identification of vaginal NKp46⁺ CD3^- cells.

(A) Frequencies (left panel) and numbers (right panel) of NKp46⁺ CD3^- cells among CD45⁺ leukocytes from spleen, genital lymph nodes (GLN) and vagina of naive C57BL/6 mice. Histogram plots represent results from five independent experiments and are expressed as mean values + SEM, n = 10 mice. (B) Immunofluorescence staining of frozen sections from mouse vagina and spleen stained with anti-CD3 (green) and anti-NKp46 (red) antibodies. Nuclei were visualized with DAPI (blue). White arrows indicate NKp46⁺ CD3^- cells; E: epithelium; LP: lamina propria. WP: white pulp; RP: red pulp. White dotted lines delineate the epithelium from the lamina propria in the vagina and the white pulp from the red pulp in the spleen. Original magnification: x40 (vagina) and x20 (spleen). Data are representative of three independent experiments.

Fig 2. Vaginal NKp46⁺ CD3^- cell population contains conventional Eomes⁺ NK cells and Eomes^- ILC1-like cells.

(A) Histogram plots show stainings with antibodies against specific markers (black line histograms) and with isotype control antibodies (grey line histograms) gated on NKp46⁺ CD3^- cells from spleen, GLN, and vagina of naive C57BL/6 mice. Specific stainings were also performed on NKp46⁺ CD3^- cells from liver and thymus. Numbers in histogram plots represent the MFI (mean fluorescence intensity) and the percentages of positive cells for the marker. Data are representative of at least four independent experiments. (B) Bidimensional dot plots show expression of CD49a and Eomes (upper panel) and CD49a and CD49b on gated NKp46⁺ CD3^- cells from spleen, GLN, vaginas and liver of naive C57BL/6 mice. Numbers in quadrants represent the percentage of cells for each subset.

Fig 3. Functional properties of vaginal Eomes⁺ NK cells and Eomes^- ILC1-like cells.

(A) Bidimensional dot plots (left panel) show expression of CD27 and CD11b on gated Eomes⁺ NKp46⁺ CD3^- cells from spleen, GLN and vaginas of naive C57BL/6 mice. Numbers in quadrants represent the percentage of cells for each subset. Histograms bars (right panel) represent results from five independent experiments expressed as mean values + SEM, n = 10 mice. **p<0.01, ***p<0.001, ****p<0.0001 Mann-Whitney U test. Spl.: spleen; GLN: genital lymph nodes; Vag.: vagina. Frequencies of CD107a⁺ and IFN-γ⁺ producing Eomes⁺ NK cells (B) and Eomes^- ILC1-like cells (C) in cell suspensions from spleen, GLN and vagina of naive C57BL/6 mice. Cells (0.5 million/well) were stimulated in vitro for 4 hours with plate-bound isotype control (IC), anti NK1.1 mAb, PMA/ionomycin, IL-12 + IL-18 cytokines or YAC-1 tumor cells. CD107a and IFN-γ expression on cells were assessed by flow cytometry. Histogram bars represent results from three independent experiments expressed as mean percentages + SEM, n = 5 mice. **p<0.01, ***p<0.001, Mann-Whitney U test. (D) Bidimensional dot plots (left panel) show expression of TNFα and TRAIL on gated Eomes^- ILC1 and Eomes⁺ NK cells from liver and vaginas of naive C57BL/6 mice after stimulation in vitro for 4 hours with PMA/ionomycin. Numbers in quadrants represent the percentage of cells for each subset. Results are representative of two independent experiments with n = 4 mice.

Fig 4. Vaginal NKp46⁺ ILC limit the number of effector CD8 T cells generated by vaginal immunization.

(A) C57BL/6 mice received three intravaginal immunizations at days 0, 14, and 21 with the vaccine (CTBOVA+CT). Groups of mice received injections of either anti-NK1.1 ascite fluid (PK136), anti-NKG2D (CX5) mAb, or isotype-matched control mAb at the indicated days. At day 28, vaginas and GLN were collected and cell suspensions were analyzed by flow cytometry. (B, C and D) Frequencies and numbers of vaginal and GLN NKp46⁺ CD3^- cells and vaginal iNKT cells after NK1.1 depletion or NKG2D/NKG2D ligands blockade from vaccine-immunized C57BL/6 mice. Histogram bars represent results from three independent experiments expressed as mean values + SEM, n = 10 mice. (E) Frequencies and numbers of OVA-specific CD8 T cells in vagina and GLN upon vaccination. Cells were analyzed by flow cytometry at day 28 using OVA-specific MHC I pentamer, anti-CD3 and anti-CD8 mAbs. Histogram bars represent results from three independent experiments expressed as mean values + SEM, n = 10 mice. (F) Frequencies of IFN-γ⁺ OVA-specific CD8 T cells from vaginas and GLN after stimulation in vitro for 4 hours with PMA/ionomycin. IFN-γ expression on OVA-specific CD8 T cells was evaluated by flow cytometry after intracellular staining. Histogram bars represent results from three independent experiments expressed as mean values + SEM, n = 6 mice. *p<0.05, **p<0.01, ***p<0.01, ****p<0.001, n.s: no significatif, Mann-Whitney U test.

Fig 5. NKG2D/NKG2D ligands interaction control vaginal dendritic cell numbers after local immunization.

(A) Naive C57BL/6 mice received three intravaginal immunizations at days 0, 14, and 21 with the vaccine (CTBOVA+CT). At day 28, cells from vaginas and GLN were analyzed by flow cytometry to assess the expression of NKG2D and the NKG2D ligand Rae1 on OVA- specific CD8 T cells. Histogram plots show stainings with antibodies against specific markers (black line) and with isotype control antibodies (grey line). (B) Dendritic cells from vagina (left panel) and GLN (right panel) of vaccine or PBS-treated mice were prepared 7 days after immunization with the vaccine and were analyzed for expression of the NKG2D ligands Rae1 and MULT1 by flow cytometry. Numbers in histograms plots represent the mean fluorescence intensity (MFI) and the percentages of positive cells for the marker. Data are representative of 2 independent experiments. Histogram bars represent results from 2 independent experiments expressed as mean values + SEM, n = 5 mice. (C) After immunization with CTBA488+CT, the number of antigen-bearing dendritic cells (A488⁺ CD11c⁺ MHCII⁺) and other DC (A488^- CD11c⁺ MHCII⁺) were analyzed at day 7 in the vagina of mice treated with anti-NK1.1 ascite fluid, isotype control and anti-NKG2D antibodies. Results from three independent experiments are expressed as mean numbers +SEM, n = 6. *p<0.001, Mann-Whitney U test.