Non-classical tissue monocytes and two functionally distinct populations of interstitial macrophages populate the mouse lung

Abstract

Resident tissue macrophages (RTM) can fulfill various tasks during development, homeostasis, inflammation and repair. In the lung, non-alveolar RTM, called interstitial macrophages (IM), importantly contribute to tissue homeostasis but remain little characterized. Here we show, using single-cell RNA-sequencing (scRNA-seq), two phenotypically distinct subpopulations of long-lived monocyte-derived IM, i.e. CD206⁺ and CD206^-IM, as well as a discrete population of extravasating CD64⁺CD16.2⁺ monocytes. CD206⁺ IM are peribronchial self-maintaining RTM that constitutively produce high levels of chemokines and immunosuppressive cytokines. Conversely, CD206^-IM preferentially populate the alveolar interstitium and exhibit features of antigen-presenting cells. In addition, our data support that CD64⁺CD16.2⁺ monocytes arise from intravascular Ly-6C^lo patrolling monocytes that enter the tissue at steady-state to become putative precursors of CD206^-IM. This study expands our knowledge about the complexity of lung IM and reveals an ontogenic pathway for one IM subset, an important step for elaborating future macrophage-targeted therapies.

Fig. 1

ScRNA-seq analysis of CD64⁺ mononuclear cells in lungs of naive C57BL/6 WT mice. a Gating strategy used for FACS sorting prior to scRNA-seq experiments. b Experimental pipeline of scRNA-seq experiments. c t-SNE plots depicting the CD64-expressing cells analyzed by scRNA-seq. n indicates the number of cells analyzed after quality control and filtering. d Dot plots showing average expression of the indicated genes and percentages of cells expressing the genes within each cluster. Examples of transcripts significantly differentially regulated (P_adj < 10⁻²) between Cluster 4, 2 or 1 vs. the 2 other clusters are depicted. e Representative contour plot of CD16.2 and CD206 expression within CD64⁺ IM, whose quantification is shown in (f). f Percentage of each mononuclear phagocyte subset among CD64⁺ bulk IM, assessed by flow cytometry. Stacked bars represent individual mice, and the % of cells per cluster as identified by scRNA-seq (right bar). g Numbers of each mononuclear phagocyte subset within the steady-state lung. Stacked bars represent individual mice. h Representative contour plot of the indicated markers within CD64⁺ IM. Numbers indicate the percentage of cells within the respective gates. The plots are representative of one of 6 individual mice analyzed, each of them giving similar results. i Representative histograms of surface MHC-II expression within each mononuclear phagocyte subset, whose quantification is shown in (j). j Quantification of MHC-II MFI. f, g Data show individual mice and are pooled from 3 independent experiments (n = 12). j Data show mean ± s.e.m. and are pooled from 2 independent experiments, each symbol representing individual mice (n = 6). P values were calculated using non-parametric f, g Friedman or j Mann–Whitney tests for pairwise comparisons. *P < 0.05; **P < 10⁻²; ***P < 10^−3; ****P < 10⁻⁴. Source data are provided as a Source Data file. AM, alveolar macrophage; Exp, experiment; GEM, gel bead in emulsion; IM, interstitial macrophage (sorted in bulk, as shown in a); MFI, Mean Fluorescence Intensity; Mo, monocyte; RT, reverse transcription

Fig. 2

Morphology, phenotype, tissue association and phagocytic abilities of lung CD64⁺ subpopulations. a Representative photograph of the indicated FACS-sorted populations. b Quantification of the size of cells presented as violin plots (height: cell area; width abundance of cells) and individual dots representing single cells. c Quantification of expression of the indicated markers as compared to control cells. d Experimental outline for panels (e, f). e Representative contour plots showing binding of anti-CD45 in vivo vs. anti-CD45 ex vivo antibodies on the indicated cell populations. Numbers indicate the percentage of double-positive cells. f Percentage of cells positive for anti-CD45 in vivo and ex vivo stainings within the indicated populations. g Experimental outline for panel (h). h Percentage of E. coli bioparticle-positive cells 3 h after i.v. or i.t. administration. Data show (b) individual cells pooled from 3 independent sorting experiments (CD16.2⁺, CD206⁺, CD206⁻, AM: n = 16,77,75,72, respectively), or (c, f, h) mean ± s.e.m., as well as individual mice (c, n = 4-5; f, n = 12; h, n = 5–8/group), and are pooled from 2 independent experiments. P-values were calculated using (c, f) non-parametric Mann–Whitney tests for pairwise comparisons or (h) a linear mixed model on log(y+1)-transformed values with Tukey’s post hoc test. *P < 0.05; **/°°P < 10⁻²; ***P < 10⁻³; ****P < 10⁻⁴; ns, not significant. Empty circles compare % of E. coli⁺ cells after i.t. injection in CD206⁺ vs. CD206⁻ IM subsets. Source data are provided as a Source Data file. i.t., intratracheal; i.v., intravenous; MFI, Mean Fluorescence Intensity. Scale bar = 10 µm

Fig. 3

Maintenance of lung tissue CD64⁺ mononuclear phagocytes in adult C57BL/6 mice. a Experimental outline for panels (b, c). Briefly, at 4 weeks of age, Cx3cr1^CreERT2.Rosa26-LSL-YFP mice were treated with TAM s.c. 3 times, 48h apart. Mice were analyzed for YFP expression 2, 9, 28, and 52 weeks later. b Representative histograms of YFP expression within the indicated populations. Numbers indicate the percentage of YFP⁺ cells, as quantified in (c). c Percentage of YFP⁺ cells within the indicated populations, assessed by flow cytometry. d Percentages of Ki-67⁺ cells in the indicated populations. e, f Absolute numbers of the indicated cell populations in the lungs of e Ccr2^−/− or f Nr4a1^−/− and control WT mice. c–f Data show mean ± s.e.m., as well as individual mice in (d–f) (c, n = 10; d, n = 4; e, f, n = 9–12) and are pooled from 2 to 3 independent experiments. P-values were calculated using (c) a two-way ANOVA with Tukey’s post hoc test, d non-parametric Mann–Whitney test for pairwise comparisons or e, f a two-tailed unpaired Student’s t-test. °/*P < 0.05; **P < 10⁻²; ***P < 10⁻³; ****P < 10⁻⁴; ns, not significant. In c the empty circle compares CD206⁺ and CD206⁻IM at week 52. Source data are provided as a Source Data file. s.c., subcutaneous; TAM, tamoxifen

Fig. 4

Localization of lung CD64⁺CD16.2⁺ monocytes, CD206⁺ IM and CD206⁻ IM a Confocal microscopy pictures of lung sections from Cx3cr1^GFP/GFP mice (CX3CR1 [green]; CD68 [red]; DAPI [blue]; CD16.2, MHC-II or CD206 [white]). CD64⁺CD16.2⁺ monocytes, CD206⁺ IM and CD206⁻ IM were identified as CX3CR1⁺CD68⁺CD16.2⁺, CX3CR1⁺CD68⁺CD206⁺ and CX3CR1⁺CD68⁺MHC-II⁺cells, respectively. Asterisks indicate CX3CR1⁻CD68⁺ AM; plain or empty arrows indicate CX3CR1⁺CD68⁺ cells expressing or not the marker of interest (i.e., CD16.2, CD206 or MHC-II), respectively. b Preferential distribution of the indicated populations in the peribronchial/perivascular area vs. the alveolar parenchyma. c, d Lung sections of C57BL/6 WT mice were analyzed: CD68 [red]; c CD31 or d Tubb3 [white]; DAPI [blue]; MHC-II, CD206 or CD16.2 [green]). CD206⁺ IM, CD206⁻ IM and CD64⁺CD16.2⁺ monocytes were identified as CD68⁺CD206⁺, CD68⁺MHC-II⁺ and CD68⁺CD16.2⁺ cells, respectively. b Data show mean ± s.e.m. and are pooled from 2 independent batches of mice (n = 4–6). P-values were calculated using a Kruskal–Wallis test, and pairwise comparisons were estimated using Mann–Whitney tests. **P < 10⁻²; ns, not significant. Source data are provided as a Source Data file. Scale bars = (a) 100 µm; (c,d) 50 µm

Fig. 5

Functional properties of IM subpopulations at steady-state and dynamic regulation after airway exposure to microbial products. a Experimental outline for (b). FACS-sorted IM subpopulations were cultured ex vivo overnight with or without LPS, and supernatants were subjected to proteome profiling. b Heatmap depicting the relative abundance of the indicated molecules in the supernatants of non-stimulated (NS) or LPS-stimulated (LPS) lung IM subpopulations. Data represent the mean and are representative of one of 2 independent sorting experiments. c Representative contour plots showing steady-state IL-10 expression as assessed by detection of 450 nm fluorescence (blue fluorescent product of the cleaved CCF4 substrate) in CCF4-loaded cells isolated from IL-10-β-lactamase reporter ITIB or WT control mice. Numbers indicate % of IL-10⁺ cells within the cell populations, as quantified in (d). d Percentages of IL-10⁺ cells in the indicated populations. e Experimental outline for panels (f–i). f Kinetic analysis of numbers of each mononuclear phagocyte subset after i.t. instillation of Pam3CSK4, LPS, and CpG. g Representative contour plot of CD16.2 and CD206 expression within CD64⁺ IM from control (Day 0), and Pam3CSK4-, LPS- and CpG-injected mice 7 days after treatment. h Kinetic analysis of MHC-II expression within the indicated populations after Pam3CSK4, LPS or CpG treatment. i Numbers of IL-10⁺ cells within the indicated subpopulations 7 days after treatment. d, f, h, i Data show mean ± s.e.m., as well as individual mice in (d, i), and are pooled from 2 independent experiments (d, n = 9; f, h, n = 5–6/time point; i, n = 3–5). P-values were calculated using d, i non-parametric Mann–Whitney tests for pairwise comparisons or f, h two-way ANOVA with Tukey’s post hoc tests. *P < 0.05; **P < 10⁻²; ***P < 10⁻³; ****/°°°°P < 10⁻⁴; ns, not significant. In f the empty circles compare numbers of CD206⁺ IM 3 days after LPS vs. day 0. Source data are provided as a Source Data file. i.t., intratracheal

Fig. 6

RNA velocity and trajectory analyses of lung monocyte and IM subpopulations in steady-state C57BL/6 mice. a t-SNE plot depicting the merged scRNA-seq data of lung CD64-expressing cells (see Fig. 1c), Ly-6C^lo patrolling monocytes (see Supplementary Fig. 9) and Ly-6C^hi classical monocytes (see Supplementary Fig. 12). b Prevalent patterns of RNA velocities substantiated by arrows and visualized on the same t-SNE plot as shown in (a). Right panel shows a higher magnification of the area depicted by a black dashed line in the left panel. (see single cell velocities in Supplementary Fig. 13). c Violin plot showing quantification of single cell relative 2D velocities in the indicated cell (sub)populations, as presented in Supplementary Fig. 13. d Visualization of single-step transition probabilities from Ly-6C^lo patrolling monocytes (left), Ly-6C^hi classical monocytes (middle) or CD64⁺CD16.2⁺ monocytes (right) to neighboring cells. Ellipses represent 95% confidence. e, f Slingshot analysis of Ly-6C^lo patrolling monocytes, CD64⁺CD16.2⁺ monocytes, and neighboring CD206⁻ IM. e Suggested pseudo-time trajectory from Ly-6C^lo patrolling monocytes to CD206⁻ IM. Ellipses represent 80% confidence. f Heatmap depicting gene expression profiles of Ly-6C^lo patrolling monocytes, CD64⁺CD16.2⁺ monocytes, and neighboring CD206⁻ IM ordered according to Slingshot pseudo-time trajectory. Left color bars indicate annotation by cell type

Fig. 7

Patrolling monocyte-derived NR4A1-dependent CD16.2⁺ monocytes are local precursors of CD206⁻ IM. a Experimental outline for experiments using mixed BM competitive chimeras shown in (b). b Percentage of BM chimerism and radio-resistance of the indicated populations. c Experimental outline for (d). d Percentages of CD45.1/2⁺ cells within the indicated cell populations. e Survival (% of live animals) were monitored at the indicated mice after CpG treatment. P-values are versus CpG-injected WT mice. f Representative contour plot of Ly-6C and CD64, gated on singlet mononuclear cell-enriched CD45⁺ non-autofluorescent SSC^loF4/80⁺CD11c⁻ cells (see Fig. 1a). Data are representative of 1 of 4 mice analyzed, each giving similar patterns. Data in b, d shown mean ± s.e.m., as well as individual mice in (d), and are pooled from (b) 2 or (d) 3 independent experiments (b, n = 3 [blood] or 7 [lung]; d, n = 6–11/group). e Data were pooled from 2–4 independent experiments (n = 10–26/group). P-values were calculated using b a one-way ANOVA followed by Tukey’s post hoc tests, d non-parametric Mann–Whitney tests for pairwise comparisons or e Mantel–Cox tests. *P < 0.05; **P < 10⁻²; ****P < 10⁻⁴; ns, not significant. i.v., intravenous