Environmental signals rather than layered ontogeny imprint the function of type 2 conventional dendritic cells in young and adult mice

Abstract

Conventional dendritic cells (cDC) are key activators of naive T cells, and can be targeted in adults to induce adaptive immunity, but in early life are considered under-developed or functionally immature. Here we show that, in early life, when the immune system develops, cDC2 exhibit a dual hematopoietic origin and, like other myeloid and lymphoid cells, develop in waves. Developmentally distinct cDC2 in early life, despite being distinguishable by fate mapping, are transcriptionally and functionally similar. cDC2 in early and adult life, however, are exposed to distinct cytokine environments that shape their transcriptional profile and alter their ability to sense pathogens, secrete cytokines and polarize T cells. We further show that cDC2 in early life, despite being distinct from cDC2 in adult life, are functionally competent and can induce T cell responses. Our results thus highlight the potential of harnessing cDC2 for boosting immunity in early life.

Fig. 1. Developmental heterogeneity of early life cDC2.

a–e Spleens from Clec9a^cre/+Rosa^TOM and Clec9a^cre/creRosa^TOM mice of the indicated ages were analyzed by flow cytometry. a Single live autofluorescence-negative cells were gated and cDCs identified as CD11c⁺MHCII⁺CD64⁻ cells. cDCs were further divided into CD24⁺ and CD11b⁺ cells, and analyzed for XCR-1 and ESAM expression, respectively. b Frequency of CD11c⁺MHCII⁺ cDCs in total splenocytes (n = 13, PND2; n = 25, PND7–8; n = 20, PND14–15; n = 14, PND19–22; n = 12, 4 weeks; n = 5, adult). c Ratio of CD11b⁺ to CD24⁺ cells within CD11c⁺MHCII⁺ cDCs (n = 13, PND2; n = 21, PND7–8; n = 20, PND14–15; n = 14, PND19–22; n = 12, 4 weeks; n = 5, adult). d The percentage of XCR-1⁺ and XCR-1⁻ cells within CD24⁺ cDC1 is shown (n = 12, PND2; n = 21, PND7–8; n = 20, PND14–15; n = 14, PND19–22; n = 12, 4 weeks; n = 5, adult). e The percentage of ESAM^high and ESAM^low cells within CD11b⁺ cDC2 is shown (n = 6, PND2; n = 11, PND7–8; n = 18, PND14–15; n = 14, PND19–22; n = 11, 4 weeks; n = 4, adult). f Representative flow cytometry gating for TOMATO expression in splenic cDC2 from 1-week-old and adult Clec9a^cre/+Rosa^TOM mice. g The frequency of TOMATO⁺ cells within splenic CD11b⁺ cDC2 of Clec9a^cre/+Rosa^TOM mice with age (n = 4, PND2; n = 11, PND7–8; n = 6, PND14–15; n = 10, PND19–22; n = 5, 4 weeks; n = 3, adult). Each dot represents one mouse, horizontal bars represent mean, error bars represent SD, gray rectangles indicate weaning. ns: not significant. Statistical analysis was performed using two-tailed t-test. h, i Bone marrow (h) and spleen (i) from 2-day-old wild-type mice were analyzed by flow cytometry. Lin⁻(CD3, B220, NK1.1, CD4, CD8α, TER119) cells were gated, pre-cDCs were identified as indicated and stained with anti-DNGR-1 (red) or isotype-matched control antibodies (gray). Source data are provided as a Source Data file.

Fig. 2. Fate mapping reveals a lymphoid contribution to cDC2 in early life.

a Spleens from 2-week-old Flt3l^−/−Clec9a^cre/creRosa^YFP (n = 7) and Flt3l^+/+Clec9a^cre/creRosa^YFP (n = 4) mice were analyzed by flow cytometry. Shown are the frequency and number of CD11c⁺MHCII⁺ cells (left) and of cDC2 (middle). b Csf1r^Mer-iCre-Mer dams were mated with male Rosa^YFP mice and injected with 4OH-tamoxifen on E8.5. Spleen and liver from offspring mice were analyzed by flow cytometry on E18.5. The percentage of YFP⁺ cells in the indicated populations is plotted (n = 8). c Spleens from Myb^−/− and Myb^+/− littermate control mice were analyzed on E16.5. The frequency of the indicated populations was calculated and plotted (n = 9). d The percentage of YFP⁺ cells within the respective populations in spleen from Rag1^creRosa^YFP mice at the indicated ages (n = 5, PND2–3; n = 11 PND7–9; n = 8 adult; n = 3 and 4 for pDCs in PND2–3 and adult mice). e The percentage of RFP⁺ cells within the indicated populations in spleen from Il7r^creRosa^RFP mice at the indicated ages (n = 9, PND2–3; n = 5, PND7–9). Each dot represents one mouse, horizontal bars represent mean, error bars represent SD. ***p < 0.001, ****p < 0.0001. Statistical analysis in a, c–e was performed using two-tailed t-test. Source data are provided as a Source Data file.

Fig. 3. TOM⁻ DC2 phenotypically resemble TOM⁺ cDC2.

a TOM⁺ cDC2 and TOM⁻ DC2 from 1-week-old Clec9a^cre/creRosa^TOM mice were analyzed for expression of the indicated surface markers. Data are representative of at least two independent experiments with four to seven mice. b Spleens from PND8 Clec9a^cre/creRosa^TOM mice were analyzed for expression of CD31 (cyan), CD3 (blue), MHCII (green), TOMATO (red), and CD11b (magenta) by microscopy. Numbered inset squares were magnified on the bottom panels TOM⁺ cDC2 (red arrows) or TOM⁻ DC2 (white arrows) were identified. Data are representative of two independent experiments with three biological replicates.

Fig. 4. TOM⁺ cDC2 and TOM⁻ DC2 in early life are transcriptionally identical.

a–c TOM⁻ DC2 and TOM⁺ cDC2 from 8-day-old and TOM⁺ cDC2 from adult Clec9a^cre/creRosa^TOM mice were sorted and analyzed by mRNA sequencing. a Principal component analysis (PCA) of the top 500 most variable genes across all samples. Dots represent biological replicates. b Pairwise comparison of TOM⁺ cDC2 and TOM⁻ DC2 from 8-day-old mice. c Normalized enrichment score (NES) of gene sets enriched in TOM⁺ cDC2 (red) and TOM⁻ DC2 (blue) from 8-day-old mice. d TOM⁻ DC2 and TOM⁺ cDC2 from 8-day-old Clec9a^cre/creRosa^TOM mice were analyzed by flow cytometry for RORγt, CD90.2, and CD127 expression. Data are representative of three biological replicates. e The percentage of RORγt⁺ cells within TOM⁺ cDC2 and TOM^- DC2 in Clec9a^cre/creRosa^TOM mice at the indicated ages is shown (n = 3, PND8; n = 6, PND13; n = 4, adult). Each dot represents one mouse, horizontal bars represent mean, error bars represent SD. f, g CD19⁻MHCII⁺ cells from spleen of 9-day-old Clec9a^cre/creRosa^TOM mice were sorted after gating out F4/80^high macrophages and subjected to scRNA-Seq. f UMAP display of 1788 cells analyzed by unsupervised graph-based clustering with Seurat algorithm to indicate cluster identity (left) and cell cycle status (right) within clusters. g Heatmap displaying expression of genes enriched in TOM⁻DC2 compared to TOM⁺ cDC2 in bulk mRNA sequencing (Fig. 4b) among scRNA-Seq clusters. h, i The ratio of normalized Tomato reads per cell to normalized reads per cell of the predicted transcript of the unrecombined ROSA locus was calculated (h) and cells with a ratio > 0.5 were identified as Tom⁺, whereas cells with a ratio <−0.5 were identified as Tom⁻. i Tom⁺ and Tom⁻ cells visualized on the UMAP display. Source data are provided as a Source Data file.

Fig. 5. Age causes strongest differences in gene expression and cell function.

a, b Bulk mRNA sequencing performed as in Fig. 4A–C. a Pairwise comparison of TOM⁺ cDC2 from 8-day-old and adult mice indicating genes with a log₂FC > 1 and padj < 0.05 in red. b NES of gene sets enriched in TOM⁺ cDC2 from 8-day-old (red) or in cDC2 from adult mice (blue). c TOM⁺ cDC2 from 2-week-old and adult Clec9a^cre/+Rosa^TOM mice were analyzed for expression of PD-L1 (n = 5, 2-week-old; n = 4, adult) and CD38 (n = 7, 2-week-old; n = 6, adult) (MFI = mean fluorescence intensity). d Splenic cDC2 from adult wild-type and Ifnar^−/− mice were profiled for expression of PD-L1 and CD38 (n = 5). Expression levels are depicted as MFI. Each dot represents one mouse, horizontal bars represent mean, error bars represent SD. e–g TOM⁻ DC2 and TOM⁺ cDC2 from 2-week-old and TOM⁺ cDC2 from adult Clec9a^cre/+Rosa^TOM mice were sorted, pulsed with OVA_323–339, and co-cultured with CTV-labeled OT-II cells under Th0 or polarizing conditions for 3.5 days. e Left: CTV dilution of OT-II cells co-cultured with the indicated populations or CD11c-enriched splenocytes without OVA_323–339 (gray). Right: division and proliferation indexes of OT-II cells after co-culture with the indicated DC2 populations (n = 5/6, 2-week-old TOM⁺ and TOM⁻ cDC2, n = 4, adult TOM⁺ cDC2). f T cells were analyzed for expression of Foxp3 and IL-17A. For Th17 conditions (right) numbers in each quadrant represent the mean ± SD (n ≥ 7). g The percentage of cytokine or Foxp3-positive cells within proliferated OT-II cells is shown. APC : T-cell ratio 1 : 2 for Th0, Th1, Th17 and 1 : 10 for Treg (n = 6, Th0/Th1; n = 3, Treg; n = 8, Th17; for 2-week-old TOM⁺ cDC2; n = 5, Th0; n = 4, Treg/Th1; n = 8, Th17 for TOM⁻ DC2; n = 4, Th0/Th1/Treg ; n = 7, Th17 for adult TOM⁺ cDC2). h TOM⁺ cDC2 from 2-week-old (n = 5) and adult mice (n = 6, Treg; n = 7, Th17) were sorted as ESAM^high and ESAM^low cells, pulsed with OVA_323–339 and cultured with CTV-labeled OT-II cells under Treg and Th17 polarizing conditions as above. The percentage of cytokine or Foxp3-positive cells within proliferated OT-II cells is shown. i TOM⁻ DC2 and TOM⁺ cDC2 from 2-week-old and TOM⁺ cDC2 from adult Clec9a^cre/+Rosa^TOM mice were co-cultured with OT-II cells in the presence of OVA under Th17 conditions for 4 days. The division index (left) and percentage of IL-17A producing OT-II cells (right, APC : T-cell ratio 1 : 2) are shown (n = 4). Each dot represents one biological replicate from four independent experiments, horizontal bars represent mean, error bars represent SD. ***p < 0.001, ****p < 0.0001. Statistical analysis was performed using two-tailed paired t-test (comparing TOM⁻DC2 and TOM⁺ cDC2 groups), one-way ANOVA (comparing 2-week-old and adult groups) or two-tailed t-test in (c, d, h). Only statistically significant comparisons are indicated. Source data are provided as a Source Data file.

Fig. 6. Early-life cDC2 induce distinct T-cell responses upon targeted antigen delivery compared to adult-life cDC2.

a 2-Week-old Clec9a^cre/+Rosa^TOM mice were injected i.p. with aDCIR2-OVA or isotype-OVA control antibody. After 12 h, TOM⁺ cDC2 and TOM⁻ DC2 were sorted and co-cultured with OT-II cells for 3.5 days. CTV dilution (left) and division index of OT-II cells after co-culture with the indicated DC2 populations (right) are shown (n = 6, 1:1 ratio; n = 5, 1:10 ratio; n = 3, isotype-OVA). b TOM⁺ cDC2 and TOM⁻ DC2 from 2-week-old and TOM⁺ cDC2 from adult Clec9a^cre/+Rosa^TOM mice were sorted and stimulated with CpG-B. Cytokine production was analyzed 18−20 h later. Each dot represents a biological replicate from 4 (IL-6, TNF-α, and IL-10) or two (IL-12p40) independent experiments. c, d 2-Week-old and adult Clec9a^cre/+Rosa^TOM mice were injected i.p with anti-DCIR2-OVA antibody plus CpG-B. After 12 h DC populations were sorted and co-cultured with OT-II T cells as in a. c CTV dilution (left) and division index (right) of OT-II cells after co-culture with the indicated DC populations (n = 10, 2-week-old; n = 8, adult and cDC1 from 2-week-old mice). d Percentage of TNF-α, IFN-γ, and Foxp3-positive cells within proliferated OT-II cells (n = 10, 2-week-old; n = 8, adult). Each dot represents one biological replicate from at least three independent experiments, horizontal bars represent mean, error bars represent SD. ***p < 0.001. Statistical analysis was performed using two-tailed paired t-test (comparing TOM⁻DC2 and TOM⁺ cDC2 groups) or one-way ANOVA (comparing 2-week-old and adult groups). Only statistically significant comparisons are indicated. Source data are provided as a Source Data file.