Toxin-Triggered Interleukin-1 Receptor Signaling Enables Early-Life Discrimination of Pathogenic versus Commensal Skin Bacteria

Abstract

The host must develop tolerance to commensal microbes and protective responses to infectious pathogens, yet the mechanisms enabling a privileged relationship with commensals remain largely unknown. Skin colonization by commensal Staphylococcus epidermidis facilitates immune tolerance preferentially in neonates via induction of antigen-specific regulatory T cells (Tregs). Here, we demonstrate that this tolerance is not indiscriminately extended to all bacteria encountered in this early window. Rather, neonatal colonization by Staphylococcus aureus minimally enriches for antigen-specific Tregs and does not prevent skin inflammation upon later-life exposure. S. aureus α-toxin contributes to this response by stimulating myeloid cell production of IL-1β, which limits S. aureus-specific Tregs. Loss of α-toxin or the IL-1 receptor increases Treg enrichment, whereas topical application of IL-1β or α-toxin diminishes tolerogenic responses to S. epidermidis. Thus, the preferential activation of a key alarmin pathway facilitates early discrimination of microbial "foe" from "friend," thereby preventing tolerance to a common skin pathogen.

Keywords: IL-1; commensal; dendritic cells; neonatal; pathogen; regulatory T cells; skin bacteria; skin immunity; staphylococcus.

Figure 1:. Early life colonization by S. epi but not S. aureus facilitates adaptive immune tolerance upon subsequent challenge.

Neonatal mice were colonized with S. epi-2w or S. aureus-2w (Precol) on postnatal day 7, 10 and 13 or left uncolonized (No Precol) before challenge 3-4 weeks later with either bacteria in the context superficial skin abrasion. (A) Representative histology and disease scoring of S. epi or (B) S. aureus treated mice. (C) Representative flow cytometry plots and graphs of % skin neutrophils (gated on a live CD45⁺CD3^neg cells). (D) Representative flow plots and graphs of %Treg in 2w-specific cells in SDLNs and (E) skin of all groups. Plots in D are gated on the live DUMP^negCD45⁺CD3⁺CD4⁺CD44⁺2w⁺ cells in the tetramer-enriched fraction and in E on the same population in total (unenriched) skin. Each point represents an individual mouse. Data are representative of three independent experiments with at least 4 mice per group. See also Figure S1.

Figure 2:. Discrimination between commensal and pathogen occurs during initial neonatal skin colonization and facilitates enrichment of S. epi-specific versus S. aureus-specific Tregs.

(A) Neonatal mice colonized with S. epi-2w or S. aureus-2w on postnatal day 7, 10 and 13 or left uncolonized (Naïve) and then re-challenged 2 weeks later with intradermal 2w peptide in Incomplete Fruend’s Adjuvant (IFA). Representative flow cytometry plots and percentages of 2w-specific Tregs in the SDLNs 8 days later. (B) Neonatal mice were colonized as above and the primary 2w response in the SDLN assessed at weaning age. Representative flow cytometry plots, percentages and absolute numbers of 2w-specific Tregs are shown. All plots are gated on live DUMP^negTCRb⁺CD4⁺CD44⁺2w⁺ cells in the tetramer-enriched fraction. Data are representative of three independent experiments with at least 3 mice per group. See also Figure S2.

Figure 3:. S. epi and S. aureus antigens are concentrated in skin cDC2s following neonatal colonization.

Neonatal mice were colonized with zsg-S. epi or zsg-S. aureus on postnatal day 9 and skin was harvested 18 hours later. (A) Gating strategy to delineate skin dendritic cell subsets. Progressive gating identifies, CD103⁺, CD11b⁺, and Langerhans cells (LHC). (B) Representative flow cytometry plots of zsgreen⁺ CD11b⁺, CD103⁺ and LHCs. (C) Percentage and (D) absolute numbers of zsgreen⁺ DC populations in skin by subtype. Data are representative of two independent experiments with at least 4 mice per group. See also Figure S3.

Figure 4:. Neonatal priming of bacteria-specific CD4⁺ T cells requires CCR7-dependent antigen transport to lymph node.

WT, Ccr7^+/− or Ccr7^−/− pups were colonized with zsgreen-expressing bacteria, zsg-S. epi or zsg-S. aureus, on postnatal day 9 and skin-draining lymph nodes (LN) were harvested 18 hours later. (A) Gating strategy to delineate LN dendritic cells (DC). Progressive gating identifies, CD11c⁺MHC^hi migratory DC (migDC) and CD11c⁺MHC^mid resident DC (rDC). (B) Representative flow cytometry plots (gated as per A) and (C) absolute numbers of zsgreen-containing migDC and rDC following colonization of WT mice with either zsg-S. epi or zsg-S. aureus. (D) Flow plots of total migDC and (E) absolute numbers of zsgreen-containing migDC in LN of Ccr7^+/− or Ccr7^−/− mice colonized with zsg-S. epi or zsg-S. aureus. (F) Ccr7^−/− or Ccr7^−/− pups were colonized with S. epi-2w or S. aureus-2w and LN were stained for tetramer-positive cells at weaning age. Representative flow cytometry plots gated on live DUMP^negCD45⁺CD3⁺CD4⁺ in tetramer-enriched fraction. (G) Absolute numbers of 2W⁺CD4⁺CD44⁺ cells in LN of Ccr7^+/− and CcrT^−/− mice. Data are representative of two independent experiments with at least 4 mice per group. See also Figure S4.

Figure 5:. IL-1β is a key upstream regulator of the transcriptional signature of CD4⁺ T cells in the skin of neonates colonized with S. aureus as compared to S. epi.

WT pups were colonized with S. epi (SE) or S. aureus (SA) every two days from birth and skin Tregs (CD25^hiICOS^hi) and CD4⁺ effector (CD25^negICOS^neg) T cells (Teff) were sorted from the skin at weaning age and processed for RNA sequencing. (A) Principal component analysis plot of the four sorted cell populations. (B) Upstream regulatory analysis was performed in Ingenuity on genes that were differentially expressed between SA vs. SE Teff and (C) SA vs. SE Tregs. Top ten regulators identified for each comparison for which there was a predicted direction of activation are shown in descending order of statistical significance. (D) Volcano plots showing p_adj value and fold change for genes differentially expressed by skin Teff and (E) Tregs from S. aureus vs. S. epi colonized mice. Blue dots represent genes with p_adj < 0.05 and > 2-fold difference in gene expression between groups. Genes in red are those annotated in Ingenuity as being IL-1β regulated. (F) Gene set enrichment analysis for the subset of these genes that are specifically regulated by IL-1β at the transcriptional level in S. aureus vs. S. epi Teff and (G) Tregs. See also Figure S5.

Figure 6:. Myeloid-derived Il-1β during neonatal colonization constrains S. aureus-specific Tregs

Neonatal WT and Il1r1^−/− mice were colonized on postnatal day 7, 10 and 13 and the primary 2w response in the SDLN assessed at weaning age. Representative flow cytometry plots and percentages of 2w-specific Tregs are shown for (A) S. aureus or (B) S. epi colonized mice. Data in A and B are pooled from two of three replicate experiments. All plots are gated on live DUMP^negTCRb⁺CD4⁺CD44⁺2w⁺ cells in the tetramer-enriched fraction. Neonatal WT mice were colonized with S. epi as above with or without (C) topical IL-1β or (D) topical IL-1α then challenged 2 weeks later with intradermal 2w peptide in IFA. Percentages of 2w-specific Tregs in the SDLN 1 week thereafter are shown. 9-day-old mice were colonized overnight with zsg-S. epi or zsg-S. aureus and skin harvested the next day. Intracellular levels of pro-IL-1β in skin (E) macrophages and (F) CD11b⁺ DCs as measured by flow cytometry. (G) Levels of IL-1β by ELISA in media of BMMs exposed for 2 hours to culture supernatants of S. epi or S. aureus. See also Figure S6.

Figure 7:. S. aureus alpha toxin limits accumulation of antigen-specific Tregs

(A) Neonatal mice colonized with S. aureus-2w or S. aureus-2w-ΔagrA, S. aureus-2w-Δhla or S. aureus-2w-Δpsm on postnatal day 7, 10 and 13 and then re-challenged 2 weeks later with intradermal 2w peptide in Incomplete Freund’s Adjuvant (IFA). Representative flow cytometry plots and percentages of 2w-specific Tregs in the SDLNs 8 days later. (B) Neonatal mice were colonized as above with either S. aureus-2w, S. aureus-2w-Δhla or S. aureus-2w-Δpsm and then re-challenged 2 weeks later with intradermal 2w peptide in IFA. (C) Neonatal mice were colonized with S. epi as above with or without recombinant hla and challenged 2 weeks later with intradermal 2w peptide in IFA. Representative flow cytometry and percentages of 2w-specific Tregs 8 days later. All plots are gated on live 2w⁺DUMP^negTCRb⁺CD4⁺CD44⁺ cells in the tetramer-enriched fraction. Data are representative of two independent experiments with at least 3 mice per group. See also Figure S7.