Neonatal Fc receptor expression in dendritic cells mediates protective immunity against colorectal cancer

Abstract

Cancers arising in mucosal tissues account for a disproportionately large fraction of malignancies. Immunoglobulin G (IgG) and the neonatal Fc receptor for IgG (FcRn) have an important function in the mucosal immune system that we have now shown extends to the induction of CD8(+) T cell-mediated antitumor immunity. We demonstrate that FcRn within dendritic cells (DCs) was critical for homeostatic activation of mucosal CD8(+) T cells that drove protection against the development of colorectal cancers and lung metastases. FcRn-mediated tumor protection was driven by DCs activation of endogenous tumor-reactive CD8(+) T cells via the cross-presentation of IgG complexed antigens (IgG IC), as well as the induction of cytotoxicity-promoting cytokine secretion, particularly interleukin-12, both of which were independently triggered by the FcRn-IgG IC interaction in murine and human DCs. FcRn thus has a primary role within mucosal tissues in activating local immune responses that are critical for priming efficient anti-tumor immunosurveillance.

Figure 1. FcRn protects against the development of colorectal cancer through a mechanism independent of intestinal microbiota

(A) Large intestine (LI) tumor incidence at 5 months of age and representative tumor histology in Apc^Min/+ and Apc^Min/+Fcgrt^−/− mice. Scale bar = 100 μm. (B) Tumor incidence in WT and Fcgrt^−/− littermates treated with 8 doses of azoxymethane (AOM). (C) Tumor incidence in AOM/DSS-treated WT and Fcgrt^−/− littermates. (D) Tumor incidence and maximum tumor diameter in WT and Fcgrt^−/− littermates in each of four independent experiments with n ≥ 3 mice per group per experiment. (E) Percent survival of WT and Fcgrt^−/− littermates treated with AOM/DSS. Significance was assessed by Logrank test. (F) Richness indices of microbiota associated with the distal LI of untreated 8-week old WT and Fcgrt^−/− littermates, as revealed by T-RFLP analysis. n = 3-5 mice per group. (G) Abundance of specific microbial species in the distal LI of untreated 7-week old WT andFcgrt^−/− littermates as assessed by qPCR. n = 9 mice per group. Representative results of two (A,B,E) or four (D) independent experiments each with n = 4-10 mice per group. All data represent mean ± s.e.m. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S1.

Figure 2. FcRn drives the activation and retention of tumor-reactive cytotoxic CD8⁺ T cells which confer tumor protection

(A) Frequency of CD8⁺ T cells in the lamina propria lymphocyte (LPL) fraction of tumor and adjacent LI tissue in WT and Fcgrt^−/− littermates (upper panels) following AOM/DSS treatment. Cytotoxic potential of cells within the CD3⁺CD8⁺ gate was assessed by intracellular staining for granzyme B (middle panels) or surface staining of LAMP1 (lower panels). (B) Mean CD8⁺ T cell frequency and cytotoxic potential in WT and Fcgrt^−/− mice, as assessed by flow cytometry, in each of three independent experiments. (C) Cytokine secretion of sorted effector CD8⁺ CD44⁺CD62L⁻ cells from the LP of tumor and adjacent tissue of AOM/DSS treated WT and Fcgrt^−/− mice following 24 h restimulation with anti-CD3 and anti-CD28. (D) Tumor incidence and tumor load (sum of the diameters of all tumors) in recipient mice adoptively transferred with CD8⁺ T cells from WT or Fcgrt^−/− AOM/DSS-treated donors. Significance was assessed by Mann-Whitney test. Representative results of three independent experiments with n ≥ 4 mice per group per experiment. All data represent mean ± s.e.m. NS = not significant. ND = not detected. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S2.

Figure 3. CD8⁻CD11b⁺ DC utilize FcRn to efficiently prime protective anti-tumor CD8⁺ T cell responses

(A) Tumor antigen-specific IgG in the serum or MLN and LI homogenates of AOM/DSS treated WT or Fcgrt^−/− mice. ELISA plates coated with lysates from tumor epithelium were probed with dilutions of serum or tissue homogenates from tumor bearing mice. (B) Transcript profiles of sorted CD8⁻CD11b⁺ and CD8⁺CD11b⁻ DC subsets isolated from the indicated tissue compartment of AOM/DSS-treated WT and Fcgrt^−/− littermates. (C) Tumor incidence and survival in Fcgrt^−/− recipients adoptively transferred with DC from the MLN and LP of AOM/DSS-treated WT or Fcgrt^−/− donors. Endpoint survival was assessed using a Chi-Squared test. (D) CD8⁺ T cell frequency in the LI LP following transfer of WT DC to AOM/DSS-treated Fcgrt^−/− recipients. (E) Tumor incidence and LI LP CD8⁺ T cell frequency in Itgax^creFcgrt^Fl/Fl mice and their littermate Fcgrt^Fl/Fl controls upon treatment with AOM/DSS. (F-G) Tumor incidence (F) and survival (G) of CD8⁺ T cell-depleted Fcgrt^−/− mice adoptively transferred with WT DC. CD8⁺ T cells were depleted by chronic i.p. administration of anti-CD8 antibody (or isotype control). Representative results of three (B-E) or two (A,F) independent experiments with n = 3-6 mice per group per experiment. All data represent mean ± s.e.m. NS = not significant. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S3.

Figure 4. FcRn drives the induction of endogenous tumor-reactive CD8⁺ T cells and can be therapeutically targeted

(A) Incidence of pulmonary metastatic nodules formed by i.v. administered OVA-expressing B16 melanoma cells (OVA-B16) in WT or Fcgrt^−/− mice or Fcgrt^−/− mice pre-immunized with WT or Fcgrt^−/− DC. (B) Frequency of endogenously occurring OVA-specific CD8⁺ T cells in WT and Fcgrt^−/− metastasis-bearing mice. Left panel demonstrates results from individual animals in a single experiment. Right panel shows the results of three independent experiments each with n = 3-6 mice per group. (C) Frequency of pulmonary metastases from mice treated as in (A) and given either a CD8⁺ T cell-depleting antibody or isotype control. (D) Frequency of pulmonary metastatic nodules and OVA-specific CD8⁺ T cells in the lungs of Fcgrt^Fl/Fl and Itgax^creFcgrt^Fl/Fl littermates. (E) Incidence of pulmonary metastatic nodules in WT or Fcgrt^−/− mice or Fcgrt^−/− mice adoptively transferred with OVA-specific CD8⁺ T cells primed ex vivo by DC loaded with OVA-containing IgG IC, FcRn non-binding IHH-IgG IC or soluble OVA. (F) Incidence of pulmonary nodules in OVA-B16-treated WT and Fcgrt^−/− mice pre-immunized with WT DC loaded ex vivo with OVA-containing IC formed with IgG or enhanced FcRn-binding LS-IgG. Representative results of three (A,B,D) or two (C,E,F) independent experiments with n = 3-6 mice per group per experiment. All data represent mean ± s.e.m. NS = not significant. (*) p = 0.09, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S4.

Figure 5. FcRn within DC enables homeostatic CD8⁺ T cell activation and IL-12 production in the LI

(A) IgG isotype content of the serum and LI or MLN homogenates in untreated WT and Fcgrt^−/− littermates. (B) CD8⁺ T cell frequency of the LI LPL fraction of untreated WT and Fcgrt^−/− littermates in a single experiment (left panels) or across three independent experimental repeats (right panel). (C) Frequency of CD8⁺ T cells in the LPL fraction of Fcgrt^Fl/Fl and Itgax^creFcgrt^Fl/Fl littermates. (D) Cytokine secretion by CD8⁺ T cells sorted from LI LP of untreated WT and Fcgrt^−/− mice following 24 h restimulation with anti-CD3 and anti-CD28. (E) Transcript profiles of CD8⁺ T cells sorted from LI LP of untreated littermate control mice. (F) Cytokine secretion from 24 h tissue explant cultures of the MLN and LI of untreated WT and Fcgrt^−/− mice. (G) Transcript profiles of sorted CD8⁻CD11b⁺ DC from the MLN of untreated littermates. Representative results of three independent experiments with n = 3-5 mice per group per experiment. All data represent mean ± s.e.m. NS = not significant. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S5.

Figure 6. IgG IC ligation of FcRn in CD8⁻CD11b⁺ DC induced IL-12 production via activation of a signaling cascade

(A) Induction of IL-12p35 upon ex vivo stimulation of WT CD8⁻CD11b⁺ DC from the spleen or MLN with IgG IC or FcRn non-binding IHH-IgG IC for 6 h. (B) IL-12 secretion after 24 h IgG IC stimulation of CD8⁻CD11b⁺ and CD8⁺CD11b⁻ DC sorted from the MLN of AOM/DSS-treated WT and Fcgrt^−/− mice. (C) Phosphorylation of STAT-1 and nuclear translocation of IRF-1 and NF-κB p65 upon IgG IC stimulation of DC isolated from WT or Fcgrt^−/− mice. (D) IL-12 transcript production by WT or Stat-1^−/− CD8⁻CD11b⁺ DC following stimulation with IgG or IHH-IgG IC for 6 h. (E) Binding of IRF-1 and NF-kB p65 to the promoters of IL-12p35 and IL-12p40 upon stimulation of WT or Fcgrt^−/− DC with IgG IC or IHH-IgG IC for 4 h. (F) Tumor incidence in mice adoptively transferred with WT DC and treated with a neutralizing anti-IL-12 antibody or isotype control. Representative results of three (A-E) or one (F) independent experiments with n = 3-7 mice per group per experiment. All data represent mean ± s.e.m. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S6.

Figure 7. FcRn expressing DC predict survival in human CRC and secrete IL-12 upon FcRn stimulation

(A) Double immunohistochemical staining of FcRn⁺CD11c⁺ DC in the stroma of CRC (upper panels) and CRC-adjacent normal LI (lower panels). FcRn = brown, CD11c = red. Scale bar left panels = 100 μm. Scale bar right panels = 20 μm. (B) Colocalization of FcRn⁺ DC (brown) and CD8⁺ T cells (red) in stroma of CRC (upper panels) and CRC-adjacent normal LI (lower panels). Arrowheads indicate areas of colocalization. (C) Kaplan Meier survival curves of 183 patients with high (≥10 per core) and low (≤ 10 per core) tumor infiltration by CD11c⁺FcRn⁺ cells. (D) Incidence of tumors in chimeric mice treated with AOM/DSS. WT recipients were reconstituted with WT bone marrow. Fcgrt^−/− recipients were reconstituted with Fcgrt^−/−, WT or hFCGRT-hB2M-mFcgrt^−/− bone marrow. Representative result of two independent experiments with n = 4-5 mice per group per experiment. (E) hIL-12p35 and hIL-12p40 transcript expression in hMoDC upon stimulation with FcRn-binding (IgG IC) or FcRn non-binding (IHH-IgG IC) immune complexes. (F) Nuclear translocation of IRF-1 and phosphorylation of STAT-1 in hMoDC upon stimulation with IgG IC or IHH-IgG IC. Data in panels A-B are representative of a total of 50 matched CRC and adjacent normal LI pairs. Data in panels E-F are representative of six donors processed in pairs in each of three independent experiments. All data represent mean ± s.e.m. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.005. See also Figure S7.