T-Bet Controls Cellularity of Intestinal Group 3 Innate Lymphoid Cells

Abstract

Innate lymphoid cells (ILC) play a significant immunological role at mucosal surfaces such as the intestine. T-bet-expressing group 1 innate lymphoid cells (ILC1) are believed to play a substantial role in inflammatory bowel disease (IBD). However, a role of T-bet-negative ILC3 in driving colitis has also been suggested in mouse models questioning T-bet as a critical factor for IBD. We report here that T-bet deficient mice had a greater cellularity of NKp46-negative ILC3 correlating with enhanced expression of RORγt and IL-7R, but independent of signaling through STAT1 or STAT4. We observed enhanced neutrophilia in the colonic lamina propria (cLP) of these animals, however, we did not detect a greater risk of T-bet-deficient mice to develop spontaneous colitis. Furthermore, by utilizing an in vivo fate-mapping approach, we identified a population of T-bet-positive precursors in NKp46-negative ILC3s. These data suggest that T-bet controls ILC3 cellularity, but does do not drive a pathogenic role of ILC3 in mice with a conventional specific pathogen-free microbiota.

Keywords: ILCs; T-bet; innate lymphoid cells; intestinal inflammation; mucosal homeostasis.

Figure 1

T-bet controls cellularity of intestinal NKp46^- ILC3 cLP. ILC of untreated mice were isolated for flow cytometry analysis. (A) ILC were gated as live CD45⁺ Lin^- CD127⁺ leukocytes. (B) NKp46^-, CCR6⁺ and double-negative ILC3 in C57BL/6 and C57BL/6-background Tbx21 ^-/- mice were analyzed as live CD45⁺ Lin^- CD127⁺ RORγt⁺ leukocytes. (C) Cell number fold change and counts per colon of total NKp46-negative ILC3 and percentage from RORγt⁺ ILC and counts per colon of NKp46^- CCR6^- and CCR6⁺ ILC3 are shown. (D) NKp46^-, CCR6⁺ and double-negative cLP ILC3 from Rag2 ^-/- and TRnUC mice were analyzed as live CD45⁺ Lin^- CD127⁺ RORγt⁺ leukocytes. (E) Cell number fold change and counts per colon of total NKp46-negative ILC3 and percentage of CCR6⁺ and NKp46^- CCR6^- ILC3 from RORγt⁺ ILC are shown. Data shown are representative of a minimum of 7 (B, C) or 4 (D, E) biological replicates.

Figure 2

NKp46^- ILC3 have fate mapper expression of T-bet. Intestinal lamina propria ILC were isolated for flow cytometry analysis. (A) See method section for further details on the design. (B) T-bet^FM and T-bet expression in live CD45⁺ Lin^- CD127⁺ NKp46⁺ NK1.1⁺ cLP ILC and (C) T-bet^FM expression in live CD45⁺ Lin^- CD127⁺ NKp46⁺ and NKp46^- SI LP ILC3 are illustrated. Data shown are representative of one experiment (B) or 2 (C) replicates.

Figure 3

T-bet deficiency promotes intestinal neutrophilia. Leukocytes were isolated from the colonic lamina propria of untreated WT and Tbx21 ^-/- mice for flow cytometry analysis. (A, B) IL-17A and IFNγ expression in live CD45⁺ Lin^- CD127⁺ CD90.2⁺ leukocytes after a 4 h stimulation with PMA and ionomycin was analyzed and statistical analyses are shown. (C, D) CD11b⁺ Gr1⁺ Ly6C⁺ F4/80^- neutrophils were analyzed from a live CD45⁺ cLP leukocyte population and neutrophil counts per colon are shown. (E) CD11b geometric median fluorescence intensity (gMFI) and granularity (SSC-A gMFI) were determined for WT and Tbx21 ^-/- cLP neutrophils. (F–I) Tbx21 ^-/- mice received a fecal transplant with pathogenic microbes derived from TRUC mice (31). (F) Changes in body weights were monitored on a weekly basis in Tbx21 ^-/- mice upon FMT and Tbx21 ^-/- control mice. Colon and spleen mass (G) and cLP neutrophil cellularity (H) were determined 3 weeks upon FMT treatment. (I) CD11b gMFI and SSC-A gMFI in cLP neutrophils 3 weeks upon FMT treatment are illustrated. Data shown are representative of 3 biological replicates.

Figure 4

T-bet deficiency promotes RORγt and CD127 expression by cLP ILC3. ILC were isolated from the colonic lamina propria of C57BL/6 and C57BL/6-background Tbx21 ^-/-, and BALB/c-background Rag2 ^-/- and TRnUC mice for flow cytometry analysis of RORγt and CD127 expression. RORγt gMFI expression in total NKp46^-, CCR6⁺ and double-negative ILC3 from (A) WT and Tbx21 ^-/- and (B) Rag2 ^-/- and TRnUC mice was analyzed within the live CD45⁺ Lin^- CD127⁺ RORγt⁺ leukocyte population. (C) CD127 surface expression in total NKp46-negative ILC3 and ILC2 defined as KLRG1⁺ ICOS⁺ CD127⁺ ILC is illustrated with flow cytometry histograms. Statistical analyses of CD127 surface expression (gMFI) on total NKp46^-, CCR6⁺ and double-negative ILC3 and ILC2 from (D) WT and Tbx21 ^-/- and (E) Rag2 ^-/- mice are presented. Data shown are representative of a minimum of 7 (A–D) or 4 (E) biological replicates.

Figure 5

STAT1 and STAT4 deficiency does not promote ILC3 cellularity. cLP ILC were isolated from mice for flow cytometry analysis. Counts per colons of (A) NKp46^-CCR6^- and CCR6⁺ ILC3 and (B) ILC2 in WT, Stat1^-/- and Stat4 ^-/- mice are shown. Data shown are representative of 4 biological replicates.