ILC2s co-localise with IL-33 in the mouse uterus.

(A) Absolute numbers of uterine ILC2s determined by flow cytometry in virgin (V) mice or dams from ♀WT x ♂WT crosses at mid-gestation (Mid = gestation day 9.5 = gd9.5), and at Term (gestation day 18.5 = gd18.5) (n=5-8). (B) Immunohistochemistry of implantation sites in dams from ♀Il33cit/wt x Il33cit/wt ♂ crosses at mid-gestation. Myometrium is indicated in red, and the ectoplacental cone in white; (C); Absolute numbers of uterine ILC2s determined by flow cytometry in dams from ♀WT x ♂WT crosses (n=7) and dams from ♀Il33citcitt x Il33cit/wt ♂ crosses (n=3 pools of 5 uteri/replicate) at mid-gestation. Data in A and C are displayed as mean ± SEM. Data in A were analysed with Kruskal-Wallis test and data in C were analysed by Student’s t-test.

Fetal growth restriction in foetuses of ILC2 KO dams.

(A) Flow cytometry data showing uILC2 numbers at mid-gestation in dams from ♀WT x ♂WT (n=5), ♀ILC2KO x ♂WT (n=3) and ♀WT x ♂ILC2KO (n=3) crosses. (B) Fetal and placental weights from ♀WT x ♂WT (n=13 litters), ♀ILC2KO x ♂WT (n=8 litters); ♀IL33KO (Il33cit/cit) x ♂WT crosses (n=4 litters); ♀WT x ♂ILC2KO (n=7 litters) crosses. Significance was determined using a mixed model approach that accounts for inter-litter variability but not sex. (C) Measurements of the head diameter and brain segmentation (the measurements are representative of one brain for cross) and femur length per cross and their quantification. (D) 3D Micro-CT imaging of gd18.5 fetuses from ♀ILC2KO x ♂WT crosses (n=9 from 3 dams) and ♀WT x ♂ILC2KO crosses (n=12 from 4 dams). (E) Example of the brain volume demarcation (red, left) and its quantification (right). Data are displayed as mean ± SEM. Data in A and B were analysed by Kruskal-Wallis test, data in D and E were analysed in each tissue by paired Student’s t-test.

Decidual and placental abnormalities in ILC2 KO dams.

(A) Stereological and immunohistochemical quantification of uterine artery thickness and lumen area ratios in the decidua of dams from ♀ILC2KO x ♂WT crosses and dams from ♀WT x ♂WT crosses at midgestation (gd9.5). Pooled data from 4 litters per cross, n=12–20 total implantation sites per cross. (B) Smooth muscle actin (SMA) staining on decidual uterine arteries at midgestation (gd9.5) in dams from ♀WT x ♂WT and ♀ILC2KO x ♂WT crosses; 12.3X magnification, scale bar 100μm. (C) Relative mRNA expression of pro-inflammatory cytokines in the uterus of dams from ♀WT x ♂WT (n=4) and ♀ILC2KO x ♂WT (n=3) crosses at midgestation (gd9.5). (D) Relative mRNA expression of tissue glucose and amino acid transporters from dissected term placentas (gd18.5). of 4 ♀WT x ♂ILC2KO crosses (n=12 placentas) and 5 ♀ILC2KO x ♂WT crosses (n=13 placentas). (E) Representative photomicrographs of sections of placentas at term (gd18.5) stained for lectin, cytokeratin and eosin to highlight fetal vessels (FV), trophoblast (TB) and maternal blood spaces (MBS);. Data in A, C and D are displayed as mean ± SEM. Data in A were analysed by Mann-Whitney test, data in C and D were analysed by paired Student’s t-test.

Uterine ILC2s display gene signatures of enhanced type-2 immunity.

(A) Principal Component Analysis of RNA sequencing data from uILC2s sorted from the uterus of virgin (V) WT mice, dams from ♀WT x ♂WT crosses at mid-gestation (Mid) or term (Term), as well as from ILC2s sorted from lungs (L) and lymph node (LN) of virgin WT mice (n= 3 replicate per group); (B) Mean-Average plot representing differentially expressed genes between ILCs in uterus of WT virgin females or dams from ♀WT x ♂WT crosses compared to ILCs from lung (L) and lymph node of virgin WT mice (LN). Red points indicate genes higher in uterine samples and lower in L and LN samples. Genes with a significant Log2 difference less than −2-fold are the higher in uterine samples (red) or significantly greater than 2-fold are the non-uterine (red). (C-D) Heat maps showing normalised read count of genes identified by GO analysis as associated with (C) type-2 immunity or (D) immune regulation.

Pro-inflammatory environment and defective alternative activation of DCs and macrophages in the uterus of ILC2 KO dams

(A) Fold change (FC) in absolute number of indicated cell types between cells in the uterus of dams from ♀ILC2KO x ♂WT (n=7) and ♀WT x ♂WT crosses (n=10) at mid-gestation (gd9.5); (B) Fold change (FC) in absolute number of indicated cell types between cells in the spleen of dams from ♀ILC2KO x ♂WT (n=3) and ♀WT x ♂WT crosses (n=4) at mid-gestation (gd9.5); (C) Relative mRNA expression of type-2 cytokines in the uterus of dams from ♀WT x ♂WT crosses (n=4) and ♀ILC2KO x ♂WT crosses (n=3) at mid-gestation (gd9.5). (D) Relative mRNA expression of type-2 associated genes (Arg1, Retnla, Chi3le, Clec7a, Mrc1 and Il4) and inflammatory associated genes (Myd88, Il1b and Il6) by purified uterine dentritic cells (DCs) and uterine macrophages of dams from ♀WT x ♂WT (n=3, black) and ♀ILC2KO x ♂WT crosses (n=3, blue) at mid-gestation (gd9.5) (n=3 pooled uteri per experiment; n=3 independent experiments). (E, F) Quantified (left) or representative (right) flow cytometry analysis of ARG-1 expression by purified uterine DCs (E) and macrophages (F) of dams from ♀ILC2KO x ♂WT (n=5) and ♀WT x ♂WT crosses (n=3) at mid gestation (gd9.5). Data are displayed as mean ± SEM and were analysed using Student’s t-test.

Enhanced rate of endotoxin-induced abortion in ILC2 KO dams.

(A) Abortion rate at mid-gestation (gd9.5), 2 days post-treatment in dams from ♀WT x ♂WT crosses (sham-treated with saline, PBS, n=5; or with LPS, n=8) and ♀ILC2KO x ♂WT crosses (PBS, n=4; LPS n=6). (B) Representative images of the uterus of dams from a ♀ILC2KO x ♂WT cross at midgestation (gd9.5), 2 days after PBS- or LPS-treatment; (C) Gene expression analysis of uterine Retnla and Il1b at mid-gestation 1 day after treatment of dams from ♀WT x ♂WT crosses (PBS, n=7; LPS n=10) or ♀ILC2KO x ♂WT crosses (PBS, n=6; LPS n=8). Data in B and C are displayed as mean ± SEM. Data in B were analysed by 2-way ANOVA and data in C by Kruskal– Wallis.

Uterine ILC2s respond to IL-1β by limiting the expansion of LPS-induced IL-1β-producing DCs.

(A) Number of ex-vivo uterine DCs producing IL-1β at midgestation (gd9.5), 1 day after treatment of dams from ♀WT x ♂WT (PBS, n=3; LPS n=4) or ♀ILC2KO x ♂WT crosses (PBS, n=4; LPS n=5). (B) Leucocytes from the uterus of WT mice were stimulated in vitro by LPS for 18h, with or without IL-1R inhibitor. Percentages of uILC2s expressing the indicated type-2 cytokines are shown out of total uILC2s (n=4). (C) Leucocytes from the uterus of dams from ♀WT x ♂WT (n=3) or ♀ILC2KO x ♂WT crosses (n=3) were stimulated in vitro by LPS for 18h, with or without IL-4 or IL-5 and number of uterine DCs producing intracellular IL-1β (left) or TNFα (right) were counted. Data are displayed as mean ± SEM. Data in A were analysed by Kruskal–Wallis, in B with RM_one-way ANOVA and in C by RM one-way ANOVA for paired analysis and one-way ANOVA when comparing across groups.