MafB-restricted local monocyte proliferation precedes lung interstitial macrophage differentiation

Abstract

Resident tissue macrophages (RTMs) are differentiated immune cells that populate distinct niches and exert important tissue-supportive functions. RTM maintenance is thought to rely either on differentiation from monocytes or on RTM self-renewal. Here, we used a mouse model of inducible lung interstitial macrophage (IM) niche depletion and refilling to investigate the development of IMs in vivo. Using time-course single-cell RNA-sequencing analyses, bone marrow chimeras and gene targeting, we found that engrafted Ly6C⁺ classical monocytes proliferated locally in a Csf1 receptor-dependent manner before differentiating into IMs. The transition from monocyte proliferation toward IM subset specification was controlled by the transcription factor MafB, while c-Maf specifically regulated the identity of the CD206⁺ IM subset. Our data provide evidence that, in the mononuclear phagocyte system, the ability to proliferate is not merely restricted to myeloid progenitor cells and mature RTMs but is also a tightly regulated capability of monocytes developing into RTMs in vivo.

Fig. 1. Lung interstitial macrophage subsets can be defined as Cx3cr1^hiTmem119^hi cells.

a, Heat map showing gene activity in the indicated myeloid cell populations, inferred from microarray data uploaded on the Gene Expression Commons platform. Alv, alveolar; CNS, central nervous system; Int, interstitial; LN, lymph node; Mo, monocyte; Mac, macrophage; PC, peritoneal cavity; SI, small intestine; SLNs, skin-draining lymph nodes; SP, spleen. b, Gene activities of Cx3cr1 and Tmem119 in the indicated myeloid cell populations, as in a. The arrow indicates lung IMs. c, Representative flow cytometry gating strategy showing CD45⁺F4/80⁺CD11c⁺ AMs, AM-excluded CD45⁺SSC^loCD11b⁺F4/80⁺Ly6C⁺CD64⁻ cMo, AM-excluded CD45⁺SSC^loCD11b⁺F4/80⁺Ly6C⁻CD64⁻ pMo, AM-excluded CD45⁺SSC^loCD11b⁺F4/80⁺CD64⁺ bulk IMs further divided into CD206⁺ IMs and CD206⁻ IMs in lungs of wild-type mice at steady state. d,e, Representative flow cytometry histograms (d) and normalized MFI (e) of GFP expression in lung cMo, pMo, AMs and IMs, as in c, and in CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁻XCR1⁺ type 1 conventional DCs (cDC1) and CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁺MAR1⁻ (cDC2) from Cx3cr1^GFP/+ and Cx3cr1^+/+ mice. f,g, Representative histograms (f) and normalized MFI (g) of intracellular Cre protein in lung myeloid cells, as in d and e, from Tmem119^Cre/+ and Tmem119^+/+ mice. Data show the mean ± s.e.m. and individual values (b, e and g: n = 3 replicates, 3 mice and 3 mice, respectively). P values were calculated using a one-way analysis of variance (ANOVA) with Tukey’s post hoc test and compared bulk IMs with cMo, pMo, AMs, cDC1 and cDC2 (e and g). ^**P < 0.01; ^****P < 0.0001. MFI, mean fluorescence intensity. Source data

Fig. 2. Efficiency and specificity of diphtheria toxin-induced interstitial macrophage depletion in IM^DTR mice.

a, Representative merged UMAP plots of lung single live CD45⁺CD11b⁺ or CD11c⁺ mononuclear cells analyzed by flow cytometry 24 h after 50 ng DT i.p. injection or no treatment in IM^DTR mice (merged data from four mice per group). Cell clusters (left) and heat map plots depicting the expression of Ly6C, CD11c, CD11b, F4/80, CD64 and CD206 (right). b, Representative UMAP plot, as in a, showing cells from either untreated IM^DTR mice or DT-treated IM^DTR mice. c, Representative contour plot of Ly6C and CD64 expression within lung single live AM-excluded CD45⁺SSC^loCD11b⁺F4/80⁺ cells from untreated and DT-treated IM^DTR mice, as in a. d, Absolute numbers of the indicated lung myeloid cell populations quantified by flow cytometry in IM^DTR mice, at 24 h after i.p. injection with DT in doses ranging from 0.1 to 500 ng. Horizontal dotted lines represent the average number of cells in untreated IM^DTR mice. e,f, Absolute numbers of lung CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁻XCR1⁺ cDC1, CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁺MAR1⁻ cDC2, CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁺MAR1⁺ DCs (MAR1⁺ DC) and CD45⁺CD11c⁺MHC-II⁺CD26⁻CD64⁺CD172a⁺ macrophages (CD64⁺ Mac) (e) and lung CD45⁺CD11b⁺Ly6G⁺ neutrophils (Neu) and CD45⁺CD11b⁺SiglecF⁺ eosinophils (Eos) (f) quantified by flow cytometry 24 h after 50 ng DT i.p. injection or no treatment in IM^DTR mice. g,h, Numbers of BM Lin⁻Ly6A/E⁺CD117⁺ LSK, Lin⁻CD16/32⁻CD117⁺CD135⁺CD34⁺CD115⁻ common myeloid progenitors (CMP), Lin⁻CD16/32⁻CD117⁺CD135⁺CD34⁺CD115⁺ monocyte-DC progenitors (MDP), Lin⁻CD16/32⁺CD117⁺CD135⁻CD34⁺CD115⁻Ly6C⁻ granulocyte-monocyte progenitors (GMP), Lin⁻CD16/32⁺CD117⁺CD135⁻CD34⁺CD115⁻Ly6C⁺ granulocyte progenitors (GP), Lin⁻CD16/32⁺CD117⁺CD135⁻CD34⁺CD115⁺Ly6C⁺ monocyte progenitors (cMoP), Lin⁻CD16/32⁺CD117⁻CD115⁺Ly6C⁺ monocytes (Ly6C⁺ BMMo), Lin⁻CD16/32⁻CD117⁻CD135⁺CD115⁺CD34⁻Ly6C⁻ common DC progenitors (CDP) (g), blood CD45⁺CD3⁻CD19⁻Ly6G⁻SiglecF⁻CD115⁺ Ly6C⁺ cMo or Ly6C⁻ pMo, CD45⁺Ly6G⁻SiglecF⁻Ly6C⁻CD115⁺CD11b⁺ F4/80^hi large (LPM) or F4/80^lo small peritoneal macrophages (SPM), liver CD45⁺CD31⁻F4/80⁺CD11b^intCD64⁺ Kupffer cells (KC), spleen Lin⁻F4/80⁺CD11b⁻ red pulp macrophages (RPM), small intestinal (SI) and colon (C) CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ lamina propria macrophages (LPM) and FSC^loCD45^intF4/80⁺CD11b⁺CD64⁺Ly6C⁻ microglia (h), as in e. Data show the mean ± s.e.m. and are pooled from 2–4 independent experiments (d–h: n = 6–15, 10, 8–10, 7, 8–10 mice per group, respectively). P values were calculated using a two-way ANOVA with Tukey’s post hoc test. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001; NS, not significant. Source data

Fig. 3. A vacant interstitial macrophage niche is repopulated by Ccr2-dependent classical monocyte differentiating into interstitial macrophages.

a, Representative plots of Ly6C and CD64 expression within lung AM-excluded CD45⁺F4/80⁺SSC^loCD11b⁺ cells at days 0, 0.5, 1, 2, 3, 7 and 14 after 50 ng DT i.p. in IM^DTR mice. b, Time course of absolute numbers of cMo, pMo, AMs, bulk IMs, CD206⁻ IMs and CD206⁺ IMs quantified by flow cytometry in IM^DTR and littermate controls, as in a. Data show the mean (centerline) ± s.e.m. (colored area) and are pooled from ≥2 independent experiments (n = 8–10 mice per time point). c, Amount of Ccl2 in the lung and serum of IM^DTR and littermate controls at 0, 12, 24 and 48 h after DT i.p. injection. d,e, Representative CD45.1 and CD45.2 contour plots (d) and bar graphs showing the percentage of CD45.1 donor and CD45.2 host chimerism (e) in the indicated cell populations from lethally irradiated thorax-protected CD45.2 IM^DTR mice reconstituted with CD45.1 wild-type BM donor cells, injected or not with 50 ng DT i.p. 4 weeks later and evaluated at day 7 after DT. f,g, Representative CD45.1 and CD45.2 contour plots (f) and bar graphs showing the percentage of Ccr2^+/+ donor, Ccr2^−/− donor and host chimerism (g) in the indicated cell populations from lethally irradiated, thorax-protected CD45.1/CD45.2 IM^DTR mice transplanted with a 1:1 mix of CD45.2 Ccr2^−/− and CD45.1 Ccr2^+/+ BM cells, injected with 50 ng DT i.p. 4 weeks later and evaluated at day 7 after DT. h, Representative contour plot of Ly6C and CD64 expression within lung single live AM-excluded CD45⁺F4/80⁺SSC^loCD11b⁺ cells in CD45.1/CD45.2 IM^DTR mice treated with 50 ng DT i.p., transferred with CD45.1 BM wild-type cMo i.v. 24 h after DT and evaluated at days 2 and 14 after DT. Plots are representative of 5 mice, each of them giving similar results. Data show the mean ± s.e.m. and are pooled from two independent experiments (c, e and g: n = 4–8, 4–8 and 6 mice per group, respectively). P values were calculated by likelihood ratio tests (b), two-way ANOVA with Tukey’s post hoc tests (c,e) or one-way ANOVA with Sidak’s post hoc tests (g). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Source data

Fig. 4. Time-course scRNA-seq analyses of interstitial macrophage niche refilling reveal discrete transitioning cells.

a, Three-dimensional UMAP plot depicting the transcriptional identity of sorted lung CD45⁺SSC^loCD11b⁺F4/80⁺CD64⁻ monocytes and CD45⁺SSC^loCD11b⁺F4/80⁺CD64⁺ IMs merged from IM^DTR mice injected with DT i.p. at 0, 12, 24, 48 and 96 h before the analysis (n = 5 pooled mice per time point). b, UMAP plots from the five separate time points after DT, as in a. Inset indicates the number of cells analyzed (a and b). c, Histogram showing the frequency of each cluster at each time point after DT. d, Heat map depicting the single-cell expression of the ten most upregulated genes within each cluster. e, Dot plots show average expression of the indicated genes and the percentage of cells expressing the genes within each cluster. f, Prevalent pattern of RNA velocities substantiated by arrows and visualized on the same UMAP plot as shown in a. The square on the right shows a higher magnification of the area in the left square.

Fig. 5. Trajectory analyses of interstitial macrophage development identify transient proliferating monocytes.

a, Two-dimensional UMAP plot depicting the transcriptional identity and cell trajectories of lung cMo, Tr-Mo, CD206⁻ IMs and CD206⁺ IMs, as in Fig. 4a, evaluated by Monocle analysis. b, Two-dimensional UMAP plot depicting the pseudotime trajectory values of lung cMo, Tr-Mo, CD206⁻ IMs and CD206⁺ IMs, as in a. c, Heat map plot depicting the DEGs along pseudotime evaluated by tradeSeq in the common trajectory starting from cMo (middle) and ending in CD206⁻ IM and CD206⁺ IM subsets. DEGs are divided into three classes, and examples of genes and the main biological responses enriched in each class are represented on the left and right, respectively. d, Gene expression of the indicated genes along pseudotime evaluated by tradeSeq in both trajectories leading either to CD206⁻ IM or CD206⁺ IM subsets. e, S and G2/M cell cycle scores of single cells within cMo, Tr-Mo, CD206⁻ IMs and CD206⁺ IMs, as depicted by violin plots (height: score; width: abundance of cells). f, cMo and IM signatures, and S and G2/M scores depicted along pseudotime, as in b. P values were calculated by one-way ANOVA with Tukey’s post hoc tests (e). ^***P < 0.001.

Fig. 6. Transitioning monocytes can proliferate via Csf1r-dependent mechanisms.

a, Representative plots of Ly6C and CD64 expression within lung CD45 i.v.⁺ and CD45 i.v.⁻ AM-excluded CD45⁺F4/80⁺SSC^loCD11b⁺ cells from EdU-pulsed IM^DTR mice treated or not with 50 ng DT i.p. 2 d before. b, Representative histograms of EdU levels in lung cMo and pMo, as in a. c, Representative histograms of EdU levels in lung CD64⁺ cells, as in a. d, Bar graphs showing the percentage of EdU⁺ cells in lung cMo and pMo, and in lung CD64⁺ cells, as in a. e, Representative histograms of DAPI signal in lung CD64⁺ cells, as in a. f, Bar graph showing the percentage of lung extravascular CD64⁺ cells in G1, S and G2/M phases, as in e. g, Expression of the indicated markers in lung cMo, EdU⁺CD64⁺ cells and EdU⁻CD64⁺ cells from EdU-pulsed IM^DTR mice at day 2 after DT and in lung cMo and IMs from untreated IM^DTR mice, as depicted by violin plots (height: MFI; width: abundance of cells). h, Representative plots of MHC-II and CD206 expression within lung CD64⁺ cells from untreated IM^DTR mice and EdU⁻/EdU⁺CD64⁺ cells from DT-treated EdU-pulsed IM^DTR mice, as in g. i, Percentage of MHC-II⁻ CD206⁻ cells and MHC-II⁺ or CD206⁺ cells within lung CD64⁺ cells from untreated IM^DTR mice and EdU⁻/EdU⁺ CD64⁺ cells from DT-treated EdU-pulsed IM^DTR mice, as in h, as depicted by violin plots (height: percentage; width: abundance of cells). j, Representative images of AMs, CD206⁻ IMs, CD206⁺ IMs and CD11b⁺CD206⁻MHC-II^loKi67^hi cells, identified by confocal microscopy on lung sections from untreated and DT-treated IM^DTR mice, at day 2 after DT. k,l, Number of CD206⁻ IMs and CD206⁺ IMs (k) and CD11b⁺CD206⁻MHC-II^loKi67^hi cells (l) per mm², as in j. m, Representative histograms of EdU levels in lung CD64⁺ cells from DT-treated IM^DTR mice, as in a, and treated i.v. with Csf1r antibodies (Ab) or isotype control 6 and 28 h after DT. n, Bar graph showing the percentage of EdU⁺ cells in lung CD64⁺ cells, as in m. Data show the mean ± s.e.m. and are pooled from two independent experiments (d, f, g, i, k, l and n: n = 4–5, 4–5, 4–10, 7–10, 6, 6 and 7–8 mice per group, respectively). P values were calculated using a two-way ANOVA with Sidak’s post hoc tests (b and k), a two-tailed Mann–Whitney test (e, l and n) or one-way ANOVA with Tukey’s post hoc tests (g and i). ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001. Scale bar, 10 µm. Source data

Fig. 7. MafB restricts proliferation and mediates interstitial macrophage development.

a, Heat map depicting predicted transcription factor (TF) activities across lung myeloid cells analyzed by scRNA-seq, as in Fig. 4a, as assessed by SCENIC. b, Expression of Mki67 and Mafb along pseudotime evaluated by tradeSeq in both CD206⁻ IM or CD206⁺ IM trajectories, as in Fig. 5d. c,d, Representative histograms (c) and bar graphs showing normalized MFI (d) of MafB expression in the indicated lung myeloid cell populations from wild-type mice. e, Bar graphs showing expression of MafB in lung cMo and IMs from untreated IM^DTR mice, and in lung cMo, EdU⁻Ki67⁺, EdU⁺Ki67⁺ or EdU⁻Ki67⁻ CD64⁺ cells from EdU-pulsed IM^DTR mice at day 2 after DT. f,g, Representative CD45.1 and CD45.2 plots (f) and bar graphs showing the percentage of wild-type donor, Ms4a3^CreMafb^fl/fl donor and host chimerism (g) in the indicated cell populations from lethally irradiated, thorax-protected CD45.1/CD45.2 IM^DTR mice transplanted with a 1:1 mix of CD45.2 Ms4a3^CreMafb^fl/fl and CD45.1 wild-type BM cells, injected with 50 ng DT i.p. 4 weeks later and evaluated at day 7 after DT. h, Efficiency of Mafb depletion within lung CD64⁺ cells from Lyz2^CreMafb^fl/fl mice evaluated by MafB intracellular staining. Data are representative of five mice, each of them giving similar results. i, Representative UMAP plots of lung CD45⁺CD11b⁺ or CD11c⁺ mononuclear cells analyzed by flow cytometry in Lyz2^CreMafb^fl/fl and Mafb^fl/fl littermate controls (merged data from four mice per group). j–l, Absolute numbers of lung cMo, pMo and CD64⁺ cells (j), bar graphs showing the percentage of EdU⁺ cells within cMo and CD64⁺ cells (k) and bar graph showing the percentage of dead cells within CD64⁺ cells (l) from Lyz2^CreMafb^fl/fl and Mafb^fl/fl mice. Data show the mean ± s.e.m. and are pooled from 2–3 independent experiments (d, e, g, j, k and m: n = 9, 5–6, 4–7, 7, 7–8 and 12 mice per group, respectively). P values were calculated using a one-way ANOVA with Tukey’s post hoc tests (d, e and g), a two-way ANOVA with Sidak’s post hoc tests (j and k) or a two-tailed Mann–Whitney test (m). In d, P values compare bulk IM versus every other population, or CD206⁺ IM versus CD206⁻ IM. In g, P values compare the percentage of donor CD45.1 wild-type chimerism. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001. AUC, area under the curve. Source data

Fig. 8. Interstitial macrophage identity is severely impaired in myeloid-restricted Mafb-deficient mice.

a, UMAP plots depicting the transcriptional identity of lung CD45⁺SSC^loCD11b⁺F4/80⁺ CD64⁻ monocytes and CD64⁺ cells from Lyz2^CreMafb^fl/fl mice and littermate controls (n = 5 pooled mice per group). b, UMAP feature plots representing single-cell expression of Mrc1 and Lyve1 in lung myeloid cells merged from Lyz2^CreMafb^fl/fl mice and littermate controls, as in a. c, Histogram showing frequency of each cluster in Lyz2^CreMafb^fl/fl mice and littermate controls. d, Heat map depicting the single-cell expression of the ten most upregulated genes within each cluster. e, Volcano plot depicting DEGs between C3 and C4 clusters. Transcripts significantly upregulated in C3 and C4 are colored in green and orange, respectively (log₂ fold change ± 0.5 and adjusted P value < 0.05). f, Expression of the indicated genes within C3 and C4 clusters, as depicted by violin plots (height: expression; width: abundance of cells). g, Surface expression of CD64 and MerTK in lung AMs and CD64⁺ cells, evaluated by flow cytometry in Lyz2^CreMafb^fl/fl and littermate controls. h, GO enrichment analysis performed on the upregulated genes in C4 as compared to C3. i, IM and cMo signature scores within C1, C3 and C4 clusters, as depicted by violin plots (height: scores; width: abundance of cells). Data show the mean ± s.e.m. and are pooled from two independent experiments (g; n = 6–7 mice per group). P values were calculated using a Wilcoxon rank sum test (e and f), a two-way ANOVA with Sidak’s post hoc test (g), a two-tailed Mann–Whitney U test with Benjamini–Hochberg false discovery rate correction (h), or a one-way ANOVA with Tukey’s post hoc test (i). ^***P < 0.001; ^****P < 0.0001. Source data

Extended Data Fig. 1. Flow cytometry gating strategies to delineate lung DCs, BM progenitors, blood immune cells and RTM.

Flow cytometry gating strategy used to delineate lung CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁻XCR1⁺ type 1 conventional DC (cDC1), CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁺MAR1⁻ type 2 conventional DC (cDC2), CD45⁺CD11c⁺MHC-II⁺CD26⁺CD64⁻CD172a⁺MAR1⁺ DCs (MAR1⁺ DC) and CD45⁺CD11c⁺MHC-II⁺CD26⁻CD64⁺CD172a⁺ macrophages (CD64⁺ Mac) (a), BM Lin⁻Ly6A/E⁺CD117⁺ LSK (b), Lin⁻CD16/32⁻CD117⁺CD135⁺CD34⁺CD115⁻ common myeloid progenitors (CMP), Lin⁻CD16/32⁻CD117⁺CD135⁺CD34⁺CD115⁺ monocyte-DC progenitors (MDP), Lin⁻CD16/32⁺CD117⁺CD135⁻CD34⁺CD115⁻Ly6C⁻ granulocyte-monocyte progenitors (GMP), Lin⁻CD16/32⁺CD117⁺CD135⁻CD34⁺CD115⁻Ly6C⁺ granulocyte progenitors (GP), Lin⁻CD16/32⁺CD117⁺CD135⁻CD34⁺CD115⁺Ly6C⁺ monocyte progenitors (cMoP), Lin-CD16/32⁺CD117⁻CD115⁺Ly6C⁺ monocytes (Ly6C⁺ BMMo), Lin⁻CD16/32⁻CD117⁻CD135⁺CD115⁺CD34⁻Ly6C⁻ common DC progenitors (CDP) (c), blood CD45⁺CD3⁺CD19⁻ T cells, CD45⁺CD3⁻CD19⁺ B cells, CD45⁺CD3⁻CD19⁻Ly6G⁻SiglecF⁻CD115⁺ Ly6C⁺ cMo or Ly6C⁻ pMo, CD45⁺CD3⁻CD19⁻Ly6G⁺CD11b⁺ neutrophils (Neu), CD45⁺CD3⁻CD19⁻CD11b⁺SiglecF⁺ eosinophils (Eos) (d), CD45⁺Ly6G⁻SiglecF⁻Ly6C⁻CD115⁺CD11b⁺ F4/80^hi large (LPM) or F4/80^lo small peritoneal macrophages (SPM) (e), liver CD45⁺CD31⁻F4/80⁺CD11b^intCD64⁺ Kupffer cells (KC) (f), spleen Lin⁻F4/80⁺CD11b⁻ red pulp macrophages (RPM) (g), small intestinal CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ lamina propria macrophages (SI LPM) (h) and colon CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ lamina propria macrophages (C LPM) (i). Mac, macrophage.

Extended Data Fig. 2. Assessment of intracellular Cre protein expression in BM progenitors, blood leukocytes and RTM.

a-c, Representative histograms of Cre intracellular expression in BM progenitors, as in Extended Data Fig. 1b,c (a), in blood leukocytes, as in Extended Data Fig. 1d (b), and in small and large peritoneal macrophages (SPM and LPM, respectively), Kupffer cells (KC), red pulp splenic macrophages (RPM), small intestinal (SI) and colon (C) lamina propria macrophages (LPM), as in Extended Data Fig. 1e-i (c) from Tmem119^+/+ (blue) and Tmem119^Cre/+ (red) mice. d, Flow cytometry gating strategy to show brain FSC^loCD45^intF4/80⁺CD11b⁺CD64⁺Ly6C⁻ microglia. e, Representative histograms of Cre intracellular protein expression in microglia from Tmem119^+/+ (blue) and Tmem119^Cre/+ (red) mice. f-h, Bar graphs showing normalized Cre expression in the indicated cell populations from the BM (f) and the blood (g), and in tissue RTM (h). Data show mean ± SEM (n = 4 mice per group). In h, P values compare microglia vs. every other population and were calculated using a one-way analysis of variance (ANOVA) with Tukey’s post hoc tests. Raw data and P values are provided as a source data file. ^****, P < 0.0001. CDP, common plasmacytoid and dendritic cell progenitor; cMoP, common monocyte progenitor; CMP, common myeloid progenitor; GMP, granulocyte-monocyte progenitor; GP, granulocyte progenitor; LSK, lineage(Lin)⁻Sca-1⁺cKit⁺ multipotent progenitor; MDP, monocyte-dendritic cell progenitor. Source data

Extended Data Fig. 3. Assessment of YFP labeling in Tmem119^CreRosa26^LSL-EYFP mice and of microglia depletion in IM^DTR mice.

a, Flow cytometry gating strategy to show Lineage(Lin)⁻Sca-1⁺c-Kit⁺ (LSK) multipotent progenitors, Lin⁻Sca-1⁻c-Kit^int/hi myeloid lineage-committed progenitors (MyP) and Lin⁻Sca-1^intc-Kit^int common lymphoid progenitors (CLP). b-c, Representative histograms of YFP expression (b) and bar graphs showing % of YFP⁺ cells (c) in BM LSK, MyP and CLP (c) from Tmem119^+/+Rosa26^LSL-EYFP (blue) and Tmem119^Cre/+Rosa26 ^LSL-EYFP (red) mice. d-e, Representative histograms of YFP expression (d) and bar graphs showing % of YFP⁺ cells (e) in blood CD45⁺CD3⁺CD19⁻ T cells, CD45⁺CD3⁻CD19⁺ B cells, CD45⁺CD3⁻CD19⁻Ly6G⁻SiglecF⁻CD115⁺ Ly6C⁺ cMo or Ly6C⁻ pMo, CD45⁺CD3⁻CD19⁻Ly6G⁺CD11b⁺ neutrophils (Neu), CD45⁺CD3⁻CD19⁻CD11b⁺SiglecF⁺ eosinophils (Eos) from Tmem119^+/+Rosa26^LSL-EYFP (blue) and Tmem119^Cre/+Rosa26 ^LSL-EYFP (red) mice. f-g, Representative histograms of YFP expression (f) and bar graphs showing % of YFP⁺ cells (g) in lung cMo, pMo, cDC1, cDC2, AMs, IMs and CD45⁻ structural cells from Tmem119^+/+Rosa26^LSL-EYFP (blue) and Tmem119^Cre/+Rosa26 ^LSL-EYFP (red) mice. h-i, Representative histograms of YFP expression (h) and bar graphs showing % of YFP⁺ cells (i) in small and large peritoneal macrophages (SPM and LPM, respectively), Kupffer cells (KC), red pulp splenic macrophages (RPM), small intestinal (SI) and colon (C) lamina propria macrophages (LPM) and microglia from Tmem119^+/+Rosa26^LSL-EYFP (blue) and Tmem119^Cre/+Rosa26 ^LSL-EYFP (red) mice. j, Numbers of microglia quantified by flow cytometry in IM^DTR and littermate control mice injected i.p. with 50 or 500 ng DT and evaluated 24 or 72 h post-DT injection. Data show mean ± SEM and are pooled from 2 independent experiments (c,e,g,i,j) (c,e,g,i: Tmem119^+/+Rosa26^LSL-EYFP; Tmem119^Cre/+Rosa26 ^LSL-EYFP: n = 4;6 mice per group, respectively; j: n = 3-4 mice per group). P values were calculated using a two-way (c,e,g,i) or a one-way (j) analysis of variance (ANOVA) with Tukey’s post hoc test. In g, P values compare IMs vs. every other population. Raw data and P values are provided as a source data file. ^*, P < 0.05; ^****, P < 0.0001. ns, not significant. Source data

Extended Data Fig. 4. IM niche refilling is independent of Nr4a1 and repopulated IMs are largely similar to native IMs 14 days post-DT in IM^DTR mice.

a, Bar graphs showing % of Nr4a1^+/+ donor, Nr4a1^−/− donor and host chimerism in the indicated cell populations from lethally-irradiated, thorax-protected CD45.1/CD45.2 IM^DTR mice transplanted with a 1:1 mix of CD45.2 Nr4a1^−/− and CD45.1 Nr4a1^+/+ BM cells, injected with 50 ng DT i.p. 4 weeks later and evaluated at day 7 post-DT. b, Principal Component (PC) analysis plot with % indicating the variability explained by each PC component, obtained by bulk RNA-seq analysis of lung cMo, AMs, CD206⁻ IMs and CD206⁺ IMs from untreated IM^DTR mice, and of lung CD206⁻ IMs and CD206⁺ IMs from DT-treated IM^DTR mice at day 14 post-DT (n = 3 pooled mice per replicate, 3 replicates per condition). c, Volcano plots depicting the DEG between native and repopulated CD206⁻ IMs (left) and native and repopulated CD206⁺ IMs (right). Transcripts significantly upregulated in native and repopulated IM subsets are colored in blue and red, respectively (log₂ fold-change ± 1 and adjusted P value < 10⁻²). d, Bar graph showing lung IM numbers assessed by flow cytometry in IM^DTR mice treated or not with DT i.p. at day 0 and 14, and analyzed 24 h after the last DT treatment (day 15). Data show mean ± SEM and are pooled from 2 independent experiments (a,d) (a,d: n = 4,7-8 mice per group, respectively). P values were calculated using a one-way ANOVA with Tukey’s post-hoc tests, and compare donor CD45.1 Nr4a1^+/+ chimerism between cell populations in a. Raw data and P values are provided as a source data file. ^**, P < 0.01; ^****, P < 0.0001. ns, not significant. Source data

Extended Data Fig. 5. Identification of non-classical CD16.2⁺ monocytes in scRNA-seq data of monocyte-to-IM trajectory.

UMAP feature plot depicting the transcriptional identity of sorted lung CD45⁺SSC^loCD11b⁺F4/80⁺CD64⁻ monocytes and CD45⁺SSC^loCD11b⁺F4/80⁺CD64⁺ IMs merged from IM^DTR mice injected with DT i.p. at 0, 12, 24, 48 and 96 hours before the analysis (n = 5 pooled mice per time point), according to their CD16.2⁺ monocyte (Mo) signature score (a) and to the expression of the indicated genes (b-e). f, Two-dimensional UMAP plot, as in a, identifying CD16.2⁺ monocytes.

Extended Data Fig. 6. Tr-Mo are BM-derived cells in transition between cMo and IMs and whose proliferation is inhibited by Csf1r antagonists.

a, Expression of the indicated markers in lung cMo, EdU⁺CD64⁺ and EdU⁻CD64⁺ cells from EdU-pulsed IM^DTR mice at day 2 post-DT. Values obtained from individual mice are connected with a dashed line. b, Bar graph showing the % of EdU⁺ cells in CD45.1 donor and CD45.2 host cells from lung cMo and CD64⁺ cells from lethally irradiated, thorax-protected CD45.2 IM^DTR mice reconstituted with CD45.1 BM wild-type cells, injected i.p. with DT and evaluated 2 days post-DT and 4 hours after EdU i.p. treatment by flow cytometry. c, Representative pictures of CD11b⁻CD206^hiMHC-II⁻Ki67⁺ AMs, CD11b⁺CD206^loMHC-II^hi IMs (CD206⁻ IM), CD11b⁺CD206^hiMHC-II^lo/int IMs (CD206⁻ IM) and CD11b⁺CD206⁻MHC-II^loKi67^hi cells, identified by confocal microscopy on lung sections from untreated and DT-treated IM^DTR mice, at day 2 post-DT. d, Representative histograms of EdU levels in lung extravascular CD64⁺ cells from DT-treated IM^DTR mice that were treated with DT i.p. 3 days before, with the Csf1r antagonist PLX-3397 or with vehicle i.p. 1 and 2 days before, and with EdU i.p. 8 h before. e, Bar graph showing the % of EdU⁺ cells in lung extravascular CD64⁺ cells, as in d. Data are pooled from 2-3 independent experiments and show individual values (a) (n = 8 mice), or mean ± SEM (b,e) (b,e: n = 10,5-8 mice per group). P values were calculated using a two-way ANOVA with Tukey’s post hoc tests (b) or a two-tailed Student’s t test (e). Raw data and P values are provided as a source data file. ^**, P < 0.01; ^****, P < 0.0001; ns, not significant. Scale bar: 10 µm. Source data

Extended Data Fig. 7. Lung CD64⁺ cells from Lyz2^CreMafb^fl/fl mice exhibit a higher proliferative potential as compared to those from littermate controls.

a, Representative histograms of Ki67 stainings within lung lung extravascular CD45⁺SSC^loCD11b⁺F4/80⁺CD64 ^int/hi cells (CD64⁺ cells) from Lyz2^CreMafb^fl/fl and littermate controls. Insets indicate the % of Ki67⁺ cells within CD64⁺ cells. b, Bar graph showing the % of Ki67⁺ cells within lung CD64⁺ cells. Data show mean ± SEM and are pooled from 2 independent experiments (b) (n = 7 mice per group). P values were calculated using an unpaired two-tailed Mann-Whitney test. Raw data and P values are provided as a source data file. ^***, P < 0.001. Source data

Extended Data Fig. 8. C-Maf specifically controls the identity of the CD206⁺ IM subset.

a, Heatmap plot depicting the DEG along pseudotime evaluated by tradeSeq in the subset-specific trajectories starting from cMo (middle) and ending in CD206⁻ IM and CD206⁺ IM subsets. b, Gene expression of the indicated genes along pseudotime evaluated by tradeSeq in both trajectories leading either to CD206⁻ IM or CD206⁺ IM subsets. c, Representative histograms of c-Maf protein expression in the indicated lung myeloid cell populations from wild-type mice. d, Bar graphs showing normalized MFI of c-Maf expression, as in c. P values compare bulk IMs vs. every other population, or CD206⁺ IMs vs. CD206⁻ IMs. e, Efficiency of Maf depletion within lung IMs from Lyz2^CreMaf^fl/fl mice evaluated by c-Maf protein intracellular staining. Data are representative of 5 mice, each of them giving similar results. f, Absolute numbers of lung cMo, pMo and IMs quantified by flow cytometry in Lyz2^CreMaf^fl/fl and Maf^fl/fl mice. g, UMAP plot depicting the transcriptional identity of lung CD45⁺SSC^loCD11b⁺F4/80⁺ CD64⁻ monocytes and CD64⁺ cells from Lyz2^CreMaf^fl/fl mice and littermate controls (n = 5 pooled mice per group). h, Histogram showing frequency of each cluster in Lyz2^CreMaf^fl/fl and Maf^fl/fl mice. i, Volcano plot depicting the DEG between lung IMs (C3) from Lyz2^CreMaf^fl/fl and Maf^fl/fl mice. Transcripts significantly upregulated in IMs from Maf^fl/fl and Lyz2^CreMaf^fl/fl mice are colored in red and blue, respectively (log₂ fold-change ± 0.5 and adjusted P value < 0.05). j, Surface expression of MerTK and CD206 in lung AMs and IMs, quantified by flow cytometry in Lyz2^CreMaf^fl/fl and Maf^fl/fl mice. (d,f,j) Data show mean ± SEM and are pooled from 3 independent experiments (n = 9 mice per group) (d) or 2 independent experiments (n = 4-5 mice per group) (f,j). P values were calculated using a one-way ANOVA with Tukey’s post-hoc test (for bulk IMs) or a two-tailed Mann-Whitney test (for IM subsets), or a two- way ANOVA with Tukey’s post-hoc test (f,j). Raw data and P values are provided as a source data file. ^*, P < 0.05; ^****, P < 0.0001; ns, not significant. Ab, antibody. Source data