Immunity to the microbiota promotes sensory neuron regeneration

Abstract

Tissue immunity and responses to injury depend on the coordinated action and communication among physiological systems. Here, we show that, upon injury, adaptive responses to the microbiota directly promote sensory neuron regeneration. At homeostasis, tissue-resident commensal-specific T cells colocalize with sensory nerve fibers within the dermis, express a transcriptional program associated with neuronal interaction and repair, and promote axon growth and local nerve regeneration following injury. Mechanistically, our data reveal that the cytokine interleukin-17A (IL-17A) released by commensal-specific Th17 cells upon injury directly signals to sensory neurons via IL-17 receptor A, the transcription of which is specifically upregulated in injured neurons. Collectively, our work reveals that in the context of tissue damage, preemptive immunity to the microbiota can rapidly bridge biological systems by directly promoting neuronal repair, while also identifying IL-17A as a major determinant of this fundamental process.

Keywords: S. aureus; Th17; commensal; dorsal root ganglion; interleukin 17; interleukin 17 receptor A; microbiota; neuronal regeneration; repair; sensory neuron.

Figure 1.. Staphylococcus aureus colonization induces Th17 cells that have no impact on host protection

(A) Mice were topically associated (TA) with S. aureus. RORγt expression by skin CD4⁺ T cells at two weeks post-association (left). Absolute numbers and frequencies of Th17 (RORγt⁺CD4⁺Foxp3⁻), Th1 (T-bet⁺CD4⁺Foxp3⁻) and Th2 (Gata3⁺CD4⁺Foxp3) cells in the skin (right). (B) Mice were topically associated (TA) or intradermally injected (ID) with S. aureus. IL-17A and IFN-γ production by skin CD4⁺ T cells (top), two weeks post-association. Frequency quantification (bottom). (C) Mice were transferred with S. aureus-specific TCR-transgenic CD4⁺ T cells (SA1^Tg) and topically associated (TA) with S. aureus. RORγt expression by SA1^Tg cells in the ear skin, two weeks after the first association (left). Absolute numbers and frequencies of SA1^Tg Th17 (RORγt⁺CD4⁺Foxp3⁻), Th1 (T-bet⁺CD4⁺Foxp3⁻) and Th2 (Gata3⁺CD4⁺Foxp3⁻) cells (right). (D) Contour plots (top) and frequency quantification (bottom) of IL-17A and IFN-γ production by SA1^Tg cells in the skin, two weeks post-association. (E) Mice previously associated (TA) with S. aureus were infected with the same strain by intradermal injection (ID). IL-17A production by CD4⁺ T cells in the ear skin, seven days post-infection (left). Absolute numbers of IL-17A⁺CD4⁺ T cells (right). (F) Mice previously transferred with SA1^Tg and topically associated (TA) with S. aureus were infected with the same strain by intradermal injection (ID). Frequency of SA1^Tg cells in the skin, seven days post-infection (left). Absolute numbers of IL-17A production by SA1^Tg cells (right). (G-H) Previously associated (TA) mice with S. aureus were infected with the same strain by intradermal injection (ID). (G) CFU in the skin of WT mice. (H) CFU in the skin-draining lymph nodes of mice treated with anti-IL-17A blocking (αIL-17A) antibody or isotype (Isot.) control (left), Il17a^−/− mice (middle), Th17 deficient mice (Ox40^Cre+Rorc^f/f) (right). Numbers in contour plots, kinetic curve graphs and bar plots indicate means ± SEM. Each dot represents an individual mouse (B, D, F, G and H). Data represent at least two experiments with three to eight mice per group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 and “NS”, not significant as calculated with Student’s t test. See also Figure S1.

Figure 2.. Staphylococcus aureus-induced T cells express a neuronal regeneration transcriptomic signature.

(A) Th17 (CCR6⁺CD4⁺Foxp3⁻) cells from the skin of topically associated (TA) mice and Th1 (CCR6⁻CD4⁺Foxp3⁻) cells from the skin of intradermally infected (ID) mice were sorted for bulk RNA-seq analysis. Scatter plots highlighting differentially expressed genes comparing Th17 (TA) versus Th1 (ID) cells for type 1 and type 17 signature (top right), tissue repair (bottom left) and nerve interaction and regeneration (bottom right). (B) alpha-CGRP-GFP nerve reporter mice were transferred with SA1^Tg-RFP⁺ and topically associated (TA) with S. aureus. Ear pinnae skin was assessed by 2-photon microscopy two weeks after first association. Frame from video reconstruction (top) showing the close interaction between SA1^Tg cells (red) and alpha-CGRP⁺ nerve fibers (cyan). Scale bar (10 μm). Quantification (bottom) of the shortest distances of the SA1^Tg cells from the CGRP⁺ nerve fibers compared with the shortest distances from the blood vessels (gray, stained with anti-CD31). (C) Mice previously associated with S. aureus were infected with the same strain by intradermal injection. This recall group (TA+ID) was compared to mice that received only topical association (TA) or only intradermal injection (ID). Th17 (CCR6⁺CD4⁺Foxp3⁻) and Th1 (CCR6^-CD4⁺Foxp3⁻) cells within the ear skin were simultaneously sorted in each of the three groups for transcriptomic analysis. (D) Top 20 GO terms enriched in Th17 (recall TA+ID) versus Th1 (ID) cells (left). Top 34 genes enriched in Th17 cells related to nerve interaction and regeneration (top right) and tissue repair and wound healing (bottom right). (E) Venn diagram showing the number of up-regulated genes in Th17 cells compared to Th1 cells within each treatment (intradermal infection (ID), topical association (TA), recall (TA+ID)). (F) Heatmap showing the relative expression of genes from the Th17 transcriptomic core set defined in figure 2E. In A and C-F, pools of 40 (TA), 15 (ID) and 10 (TA+ID) mice per group were used. Data in B shows one representative video out of four videos taken in four independent mice and independent experiments. ****p < 0.0001 as calculated with Student’s t test. See also Figure S2, and Video S1.

Figure 3.. Staphylococcus aureus-elicited Th17 cells promote local nerve regeneration.

(A to E) Mice received (or not) two rounds of topical association (TA) with S. aureus. Subsequently ear pinnae were injured by punch biopsy and analyzed by confocal microscopy, 10 days after punch. (A) Confocal images of the ring of nerve regeneration (β3-tubulin) and CD4⁺ T cell infiltration around the injured site, in unassociated (Ctrl) and associated mice (TA). (B) Quantification of the absolute numbers of CD4⁺ T cells, area, and volume of the pan β3-tubulin nerve fibers, around the injured site. (C) Quantification of area of the nerve regeneration ring in unassociated (Ctrl) and associated mice (TA) treated with anti-IL-17A blocking antibody (αIL-17A) or isotype control (Isot.) and, (D) WT and Il17a^−/− mice. (E) Confocal images (left); and quantification (right) of the area of the nerve regeneration ring, in unassociated (Ctrl) and associated (TA) WT (Ox40^Cre-Rorc^f/f) and Th17-deficient (Ox40^Cre+Rorc^f/f) mice. Scale bars (200 μm). Dot plots show means, and each dot represents an individual mouse. D is a pool of two independent experiment. Data represent at least two experiments with 5–12 mice per group. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 as calculated with Student’s t test. See also Figure S3.

Figure 4.. Il17ra is upregulated by injured neurons

(A) DRG neurons were cultured in presence of IL-17A for 96 hours and analyzed by bulk RNA-seq (left). Heatmap (right) showing relative expression of differentially expressed genes between cultured DRG neurons treated with or without IL-17A. (B) Diagram of punch biopsy and skin nerve innervation. Injured skin sensory neurons whose cell bodies reside in the DRG connect the damaged skin with the spinal cord. (C) RNAscope images of the cervical-DRG2, stained with probes against mRNA transcripts encoding Il17ra (yellow), Atf3 (red), and Tubb3 (magenta). (D) Quantification (mean ± SEM) of Atf3 (left) and Il17ra (right) mRNA expression in the cervical DRG2. (E) Quantification (mean ± SEM) of Il17ra mRNA in Atf3⁺ and Atf3⁻ nerve fibers. (F) Diagram of sciatic nerve transection model. (G) RNAscope images of the lumbar-DRG4, stained with probes against mRNA transcripts encoding Il17ra (yellow), Atf3 (red), and Tubb3 (magenta) (left). Quantification of Atf3 and Il17ra mRNA expression (right). (H) Confocal microscopy images of the lumbar-DRG3, stained with anti-IL-17RA (green), anti-β3-tubulin (magenta) and DAPI for nuclei (blue) (top). Quantification of IL-17RA protein expression (bottom). Graphs in D, E and G show gene expression (number of puncta / μm³) within a segmented neuron. Each dot in G, H represents an individual mouse. Scale bars (50 μm), except for zoom in image (C), where scale bar (5 μm).*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 and “NS”, not significant as calculated with Student’s t test. See also Figure S4.

Figure 5.. Neuronal IL-17RA signaling promotes sensory neuron regeneration and is not associated with aberrant mechanosensation.

(A) RNAscope images of the cervical-DRG2, stained with probes against mRNA transcripts encoding Il17ra (yellow), Trpv1 (cyan), and Tubb3 (magenta). Scale bar (50 μm), except for the zoom in image where the scale bar represents 5 μm. Quantification of Il17ra mRNA expression in Trpv1⁺ and Trpv1⁻ nerve fibers 96 hours post-injury. Bars are gene expression (number of puncta / μm³) within a segmented neuron. (B) Mice received two rounds of topical association (TA) with S. aureus. Subsequently the ear pinnae were injured by punch biopsy and analyzed by confocal microscopy 10 days later. Confocal images of the ring of nerve regeneration, stained with β3-tubulin and alpha-CGRP antibodies, in control (Trpv1^Cre−Il17ra^f/f) mice and mice lacking Il17ra expression in sensory neurons (Trpv1^Cre+Il17ra^f/f). Scale bars (200 μm). (C) Area and volume quantification of the ring of neve regeneration (anti-β3-tubulin). (D) Area and volume quantification of the ring of sensory neuron regeneration (anti-alpha-CGRP). (E-F) Mice received two rounds of association (TA) with S. aureus. Subsequently, footpads were injured by punch biopsy and seven (E) or 28 (F) days later, mechanosensation was analyzed by von Frey test. Data represent at least two experiments with three to four (A), 10 to 12 (C and D), and five to six (E and F) mice per group. Each dot represents an individual mouse (C and D) and the mechanical sensitivity threshold of an individual footpad (E and F). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 and “NS”, not significant as calculated with Student’s t test. See also Figure S5.