Fig 2: Infiltrating cells are primarily CCR2+ Ly6Chigh. a, b CD45high populations were gated on Ly6C, Cx3CR1, CCR2, and CD45-IV expression. c The number of Ly6Chigh cells increased dramatically at 24 h and remained high before falling at 72 and 96 h (red). d Ly6Clow cells were less abundant at early times, slowly increasing over the first 48 h before abruptly doubling between 72 and 96 h (blue). e CD45-IV+ populations from retinal samples were examined to determine the Ly6C status of cells trapped within the retinal vasculature (yellow). The number of Ly6Chigh cells increased sharply by 24 h and did not return to baseline by 96 h; the number of Ly6Clow cells increased briefly at 24 h before returning to baseline. *p < 0.05, **p < 0.01, ***p < 0.001; error bars are SEM, n = 6–9 retinas per time point

Fig 3: In vivo imaging of monocyte infiltration through retinal vasculature during early neurodegeneration. In vivo retinal images of Arr1−/− CCR2+/rfp mice before and during photoreceptor degeneration. The fluorescence images (SLO) showed only fleeting RFP+ signal prior to degeneration (a), but in the same vessel 24 h later, RFP+ cells studded the interior (b) and could also be seen in other large caliber vessels and within the parenchyma of the central retinal (b′). By 48 h, RFP+ cells had escaped the vasculature and appeared throughout the retina (c, c′). The bright field (reflectance) images taken concurrently with the fluorescence (SLO) show the position of the blood vessel in the same animal over time. SLO images are frames from Additional file 2: Video 0 h; Additional file 3: Video 24 h; Additional file 5: Video 48 h. (d) Color-coded representation of Additional file 4: Video ext 24 h (different Arr1−/− CCR2+/rfp animal, 24 h), where each pixel was colorized based on the time epoch of the video in which RFP+ cells appeared. The largest diameter vessel is largely white because the corresponding pixels were RFP+ throughout the entire session due to persistent RFP+ cell adherence to the vessel walls. Notice transient epochs of cell adherence to finer capillaries as revealed by streaks of blue, green, red, etc. The grayscale projection of maximal pixel intensities over time (right) outlines the vascular network in greater detail. Scale bars are 100 μm in a, b, c, d, and 500 μm in b′, c′

Fig 4: Postnatal development of lung ILC2s contributes to the proper function of gut ILC2s.a, b Analysis of lung and gut ILC2s from mice of indicated postnatal (P) days by flow cytometry and show as mean ± SD. c Expression of CCR2 and CCR4 on ILC2s from CCR2-RFP;CCR4-mNeonGreen mice at postnatal day 12 was analyzed by flow cytometry. d Transfer of lung ILC2s or gut ILC2s isolated from mice at postnatal day 12 to the B-NDG mice. Seven days after transfer, ILC2s from the lung and gut of B-NDG mice were analyzed by flow cytometry. e Treatment of HDM at early age affected the function of lung and gut ILC2s. Mice at postnatal day (P) 10 were i.n. treated with 10 μg HDM for 5 consecutive days, and lung or gut ILC2s were subsequently analyzed by flow cytometry (indicated as P15). Two weeks after first round of HDM treatment, mice were i.p. challenged with 400 ng/mouse/day IL-33 for two days and subjected to flow cytometry analysis (indicated as P30). f Postnatal mice were treated as (e), gut ILC2s from mice with IL-33 treatment at P30 were analyzed by bulk mRNA-sequencing. Blue dots, upregulated genes in HDM-treated (HDM) mice; red dots, upregulated genes in PBS-treated (Ctrl) mice. Representative DEGs are depicted. g Secretion cytokines of ILC2s from HDM-IL-33 treated mice or PBS treated mice (Ctrl) in (e) were analyzed by ELISA and show as mean ± SD. Data are representative of at least three independent experiments. **, P < 0.01 by Two-tailed unpaired Student’s t-test. (P = 0.0015, 0.0030, 0.0029, 0.0048 for IL-4 group; P = 0.0048, 0.0017, 0.0076, 0.0056 for IL-13 group). Source data are provided as a Source Data file.

Fig 5: CCR2 and CCR4 serve as tissue specific markers of ILC2s.a Violin plots showed ILC2 signature genes in the lung and the intestine. Ccr2 was highly expressed in lung ILC2s and Ccr4 is restricted to intestine ILC2s. Box plots indicate median (middle line), 25th, 75th percentile (box) of the data and the maximum and minimum values as endpoints for the whiskers. b Maturation trajectory of lung and intestine ILC2s were analyzed by Monocle pseudotime analysis. c, d Expression of CCR2 and CCR4 on ILC2s. CCR2-mNeonGreen reporter mice (c) and CCR4-mNeonGreen reporter mice (d) were generated as described in “Methods” and analyzed for the expression of CCR2 and CCR4 on ILC2s by flow cytometry. Lin-CD45+CD127+ lymphocytes were gated for analysis of KLRG1 and ST2. (Lin=CD3e,CD8a,CD19,CD11b,CD11c,Gr1,F4/80,Ter119). e, f CCR2-mNeonGreen and CCR4-mNeonGreen mice were intraperitoneally injected with 200 ng/mouse/day IL-25 for three constitutive days. Expression of CCR2-mNeonGreen (e) and CCR4-mNeonGreen (f) on ILC2s from respective mice were analyzed by flow cytometry. g, h CCR2-mNeonGreen mice (g) and CCR4-mNeonGreen mice (h) were analyzed for the expression of CCR2-mNeonGreen and CCR4-mNeonGreen on ILC2s after i.p. injection of 400 ng/mouse/day IL-33 for three constitutive days. (Gate strategies for ILC2s are: Lin−CD127+ST2+KLRG1+ for lung ILC2s, Lin−CD127+ST2−KLRG1+ for iILC2, Lin−CD127+KLRG1+ for gut ILC2s, Lin=CD3e,CD19,CD11b,CD11c,Gr1,F4/80,NK1.1). Data are representative at least three independent experiments. Source data are provided as a Source Data file.