Alveolar macrophage-expressed Plet1 is a driver of lung epithelial repair after viral pneumonia

Abstract

Influenza A virus (IAV) infection mobilizes bone marrow-derived macrophages (BMDM) that gradually undergo transition to tissue-resident alveolar macrophages (TR-AM) in the inflamed lung. Combining high-dimensional single-cell transcriptomics with complex lung organoid modeling, in vivo adoptive cell transfer, and BMDM-specific gene targeting, we found that transitioning ("regenerative") BMDM and TR-AM highly express Placenta-expressed transcript 1 (Plet1). We reveal that Plet1 is released from alveolar macrophages, and acts as important mediator of macrophage-epithelial cross-talk during lung repair by inducing proliferation of alveolar epithelial cells and re-sealing of the epithelial barrier. Intratracheal administration of recombinant Plet1 early in the disease course attenuated viral lung injury and rescued mice from otherwise fatal disease, highlighting its therapeutic potential.

Fig. 1. scRNA-seq analyses reveal two distinct alveolar BMDM clusters with defined kinetics of appearance over the course of IAV infection.

a Schematic representation of the infection model and sample collection in uninfected mice (D0) and at the indicated time points p.i. The figure was created with BioRender. b Uniform manifold approximation and projection (UMAP) map displaying CD11c⁺ alveolar cell clusters (pooled data from all time points), integrated and embedded using the Harmony algorithm. Colors represent different cell populations after Leiden clustering in 1.4 resolution (58,998 individual cells with signals from 32,286 genes; average number per time point from n = 3–5 mice). c Track plot showing unbiased analysis of the top three to five expressed genes in each particular cluster; pooled data from all time points. d Dot plot representing signature markers for each cell cluster. The size of the dots encodes the percentage of cells within a cluster expressing a particular marker. Color of the dots represents the average expression level throughout all cells within a cluster. e UMAP depiction of alveolar CD11c⁺ cell clusters at the respective time points p.i. f Stacked bar chart showing the percentage of CD11c⁺ cells during the infection course. g Dot plots displaying expression of selected phenotype markers in BMDM1 and BMDM2 populations. Genes shown in the plot were selected among the significantly higher expressed genes (p < 0.05), The Wilcoxon rank sum test was employed as the statistical method. The size of the dots encodes the percentage of cells within a cluster expressing a particular marker. Color of the dots represents the average expression level throughout all cells within a cluster. h Gene ontology (GO)-based GSEA showing normalized enrichment scores (NES) for expression differences of genes comparing BMDM1 compared to BMDM2 on D14 and 21 p.i. An array of statistical tests provided by the fgsea package was employed. Data in (b)–(h) refer to the experiment depicted in (a). Tissue-resident alveolar macrophages (TR-AM), bone marrow-derived macrophages (BMDM), dendritic cell (DC), T lymphocyte (T).

Fig. 2. CD40 and CD206 expression identify BMDM1 and BMDM2 clusters in the lungs of IAV-infected mice.

a Dot plot graph showing expression of CD40 and CD206 in BMDM1 and BMDM2, respectively. The size of the dots encodes the percentage of cells within a cluster expressing the particular marker. Color of the dots represents the average expression level throughout all cells within a cluster. Data refer to the scRNA-seq experiment in Fig. 1. b Representative FACS histograms displaying CD40 expression in BMDM1 and CD206 expression in BMDM2 collected from BALF of IAV-infected mice on days 7 and 21, respectively. c Bar graphs show total BMDM1 and BMDM2 cell counts in BALF samples of mice at the indicated time points. Values are representative of three independent experiments shown as mean ± SEM with individual data points for each group with n = 5, except for D10 samples where n = 3. Probability determined using two-way ANOVA (Holm–Sidak’s multiple comparisons test) ****p < 0,0001, **p = 0.0043; for comparison between BMDM1 and BMDM2. d Heatmap of DNA microarray profiling depicting genes associated with macrophage polarization phenotypes and e with genes associated to epithelial growth factor signaling in CD40^high BMDM1 and CD206^high BMDM2 flow-sorted from BALF of IAV-infected mice at D7 and D21 p.i., respectively; prior gating of BMDM was performed according to Supplementary Fig. 2a. Source data are provided as a Source Data file. BMDM bone marrow-derived macrophage, BALF bronchoalveolar lavage fluid, D day.

Fig. 3. BMDM2 as opposed to BMDM1 exert proliferative and barrier-protective effects on alveolar epithelial cells in vitro and in vivo.

Bar graphs indicate the percentage of apoptotic (a, Annexin V+) (*p = 0.038, **p = 0.0023, ***p = 0.0005), and proliferating (b, Ki67+) (**p = 0.008, ***p = 0.0018) control (n = 3; 5) or IAV-infected (n = 6; 7) (24 h) primary murine AECs that were co-cultured with BMDM1 (n = 3; 6) or BMDM2 (n = 3; 4) (flow-sorted from BALF of infected mice) for further 24 h. c Schematic representation of BALO-BMDM co-culture assay setup. d Quantification of organoid numbers (*p = 0.041) and e diameter (µm) (**p = 0.0099, ***p = 0.0002, ****p < 0.0001) of BALO after 7D of co-culture, in control experiments (n = 7) (no BMDM) and in co-culture with BMDM1 (n = 9) or BMDM2 (n = 9). f Representative whole-well picture of organoids after 21D in culture. Images were obtained using EVOS FL microscope. Scale bars indicate 50 µm. g Schematic representation of experimental setup of BMDM intrapulmonary transfer into IAV-infected Ccr2^-/- mice referring to data in (h)–(j). The figure was created with BioRender. h Percentage of apoptotic AEC (CD31/45^negEpCam⁺Annexin V⁺⁾ analyzed by FACS. Groups: no cells (n = 6), BMDM1 (n = 6), BMDM2 (n = 10) (*p = 0.049, **p = 0.0012). i Assessment of lung barrier permeability by FITC fluorescence in BALF of Ccr2^-/- recipient mice after intravenous application of FITC-labeled albumin, ratios of BALF to serum fluorescence are given as arbitrary units. Groups: no cells (n = 4), BMDM1 (n = 6), BMDM2 (n = 6) (**p = 0.0079, ***p = 0.0053). j Percentage of proliferating AEC II (CD31/45^negEpCam⁺T1α^negKi67⁺) analyzed by flow cytometry. BMDM1 and BMDM2 were flow-sorted from BALF of IAV-infected WT mice at D7 and D21, respectively for experiments in (a)–(j). Groups: no cells (n = 4), BMDM1 (n = 10), BMDM2 (n = 10) (***p < 0.0001, *p = 0.0236). Bar graphs are representative of three independent experiments showing means ± SEM and individual data points. Statistical significance was calculated using one-Way ANOVA and Tukey’s post-hoc tests except in (d) where Dunnett’s post-test was used, and (e), (i) where Brown Forsythe and Welch ANOVA followed by Games–Howell’s test was used. In (d) and (e), single data points represent means of organoid numbers and diameters per well. BMDM bone marrow-derived macrophage, ACE alveolar epithelial cell, BALO bronchoalveolar lung organoid, D day. Source data are provided as a Source Data file.

Fig. 4. BMDM2 and TR-AM are characterized by increased Plet1 expression and release compared to BMDM1.

a Heatmap of cDNA microarray analysis showing top 10 DEG in BMDM2 versus BMDM1 flow-sorted at D7/D21 p.i., respectively. Genes shown in the heatmap were selected according to p value (p < 0.05). b Violin plots showing expression level (log) of Plet1 in BMDM and c in TR-AM of IAV-infected mice at D7, D14, D21 and D35. Colored areas indicate density distribution in each cluster. Data refer to scRNA-seq experiments in Fig.1b. d Comparative Plet1 mRNA expression analysis by qPCR in BMDM1 of IAV-infected mice obtained at D7; in BMDM2 and TR-AM at D21 p.i. Results are expressed as ΔCt value (Ct reference−Ct target) (n = 6) (**p = 0.0016, ***p = 0.0001). e Mean fluorescence intensity (MFI) of Plet1 (PE) analyzed by FACS in BMDM1 (D7 p.i.) (n = 8), BMDM2 (D21 p.i.) (n = 7) and TR-AM (D21 p.i.) (n = 7) from BALF of IAV-infected mice (**p = 0.0042, ***p = 0.0025). f Quantification of soluble Plet1 in BALF of IAV-infected WT mice at D0/7/21/28/35/60 p.i. n = 3 except on D0 where n = 5 and D7 where n = 6 (*p = 0.0164, ***p = 0.0007). g Quantification of soluble Plet1 in BALF of D7 IAV-infected Ccr2^-/- mice, adoptively transferred at D3 p.i. with BMDM or TR-AM subsets flow-sorted from BALF of IAV-infected WT mice at the indicated time points (n = 3) (**p = 0.0087, ***p = 0.0001, ****p < 0.0001). h Plet1 concentration in supernatant of non-infected or infected AEC (24 h) (n = 5); in supernatant of BALF-isolated TR-AM treated for 12 h with conditioned medium of non-infected (n = 4) or infected AEC (12 h, n = 6, and 24 h, n = 5) or of untreated TR-AM (n = 6) (***p < 0.0001). Data are representative of three independent experiments shown as mean ± SEM statistical differences were calculated using one-way ANOVA and Tukey’s post-hoc tests. BMDM bone marrow-derived macrophage, AEC alveolar epithelial cell, BALF bronchoalveolar lavage fluid, TR-AM tissue-resident alveolar macrophage, D day. Source data are provided as a Source Data file.

Fig. 5. Soluble Plet1 drives AEC expansion and promotes alveolar epithelial barrier function in ex vivo models.

a Representative image of BALO co-cultured with BMDM2 for 21D and treated with anti-Plet1 mAb or isotype control. Scale bar represents 50 µm. b Quantification of BALO numbers; single data points represent the mean of organoid numbers per each well (n = 5) (****p < 0.0001). c Mean diameter (µm) of organoids (n = 40–87) per each well, n = 5 well/group, after 7D in culture; single data points represent mean diameter of organoids per each well (****p < 0.0001, ***p = 0.0006). d average numbers of alveoli per BALO after 21D in culture (control) or co-culture with BMDM2 and anti-Plet1 mAb or isotype control (alveoli were counted from n = 3 BALO per group) (*p = 0.0252). e Predicted kinase activity (5 top phosphorylated, red, and de-phosphorylated kinases, green; shown within phylogenetic kinome tree) in AEC treated with 20 ng/ml rPlet-1 for 12 h. f Suggested kinase signaling pathway activated by Plet1. The figure was created with BioRender. g Assessment of Ki67 staining in murine primary AEC culture treated with rPlet1 (20 ng/ml) and/or MEK inhibitor U0126 (10 µM) (n = 3) (*p = 0.0065, **p = 0.004, ***p = 0.0006). h Quantification of tight junction gene expression (Tjp-1, Cldn1, Ocln) by qPCR in murine AEC IAV-infected and treated with rPlet1 (20 ng/ml) (n = 3) (Tjp-1 ***p = 0.0004, ****p < 0.0001; Cldn1 *p = 0.0282, **p = 0.0236; Ocln ***p = 0.0003, ****p < 0.0001). i Representative image of ZO-1 localization in AEC monolayers, IAV-infected and/or treated with rPlet1. Scale bar represents 50 µm. j Time course of transepithelial resistance (TER) in murine (n = 3) and human AEC (n = 4) monolayers on transwells, IAV-infected and/or treated with rPlet1 (*p < 0.0001 for the comparison between AECs+A/PR8 and AECs+A/PR8+rPlet1). k Quantification of tight junction gene expression (Tjp-1, Cldn1, Ocln) by qPCR in murine AEC IAV-infected and treated with rPlet (20 ng/ml) with/without Src activator (10 µM). Results are shown as fold change over IAV-infected AEC without rPlet1 (n = 4) (Tjp-1 ****p < 0.0001; Cldn1 ****p < 0.0001; Ocln **p = 0.0024). Data are representative of three independent experiments showing mean ± SEM calculated using one-way ANOVA and Tukey’s post-hoc tests. j represents a single experiment performed with 3 (mouse) or 4 (human) different biological independent samples showing mean ± SEM and results were analyzed by two-way ANOVA and Tukey’s post-hoc test. BMDM bone marrow-derived macrophage, AEC alveolar epithelial cell, BALO bronchoalveolar lung organoid, TER transepithelial resistance, TR-AM tissue-resident alveolar macrophage, D day, inh inhibitor. Source data are provided as a Source Data file.

Fig. 6. BMDM2-expressed Plet1 drives epithelial progenitor (AEC II) cell expansion and improves lung barrier function in IAV-induced lung injury in vivo.

a Quantification of BALO numbers after 7D in culture (control) (n = 8) or co-culture with BMDM2 derived from Txf-treated control (Plet1^flx/flx) (n = 4) or Cx3cr1^iCre-Plet1^flx/flx (n = 6) mice; single data points represent mean of organoid numbers per each well (***p = 0.0008, ****p < 0.0001). b BALO diameter (µm) from the same experiment; single data points represent mean diameter of organoids (n = 20–49) per each well. Groups: Control (no cells) (n = 8 well); Plet1^flx/flx (n = 4 well) and Cx3cr1^iCre-Plet1^flx/flx (n = 6 well) (**p = 0.0026, ****p < 0.0001). c Representative image of BALO at 21D of culture, stained with LysoTracker® Green DND. The scale bar in the EVOS FL microscope images (middle) represents 50 µm while in confocal images 25 µm (top) and 50 µm (bottom). d Average numbers of alveoli per BALO from the experiment in (c); quantified from n = 6 BALO/group (*p = 0.0182). e Lung histological sections of IAV-infected, Txf-treated Cx3cr1^iCre-Plet1^flx/flx or Plet1^flx/flx mice obtained after 10/14/21D p.i., stained with H&E. Images underneath are magnifications of the areas within squares. Scale bars represent 100 µm (top) and 50 µm (bottom). f Percentage of Annexin V⁺ AECs in Plet1^flx/flx mice at day 10 (n = 7) and day 14 (n = 6) or Cx3cr1^iCre-Plet1^flx/flx mice at day 10 (n = 8) and day 14 (n = 6) (****p < 0.0001). g Percentage of Ki67⁺ AECs in Plet1^flx/flx mice at day 10 (n = 7) and day 14 (n = 3) or Cx3cr1^iCre-Plet1^flx/flx mice at day 10 (n = 8) and day 14 (n = 3) (****p < 0.0001). h mRNA expression (qPCR) of tight junction component genes in the lung (n = 3) (Tjp-1 D10 **p = 0.01, D14 *p = 0.0224; Cldn1D10 *p = 0.013, D14 *p = 0.0185; Ocln D10*p = 0.0405, D14 *p = 0.0348). i Quantification of barrier dysfunction by FITC-Albumin fluorescence analysis in BALF of Plet1^flx/flx mice at day 10 (n = 7) and day 14 (n = 4) or Cx3cr1^iCre-Plet1^flx/flx mice at day 10 (n = 8) (****p < 0.0001) and day 14 (n = 6) (*p = 0.0368). f–i refer to the experiment described in (e). j Lung sections of IAV-infected Plet1^flx/flx or Cx3cr1^iCre-Plet1^flx/flx mice obtained at D14/21 p.i., stained with H&E. Scale bars represent 50 µm. Images provide a detailed visualization of specific regions from the corresponding images in 6e (bottom) while maintaining the same magnification as in 6e, bottom.  * indicates areas of disrupted, non-repaired alveoli. k Quantification of disrupted alveolar area in lung slides of IAV-infected mice. Single data points represent the mean of three randomly chosen areas from each paraffin lung section (n = 3 biologically independent mouse lung sections for each group) (D14 *p = 0.0439, D21 *p = 0.0367). l Survival analysis of IAV-infected, Txf-treated Cx3cr1^iCre-Plet1^flx/flx (n = 15) or Plet1^flx/flx mice (n = 11) (*p = 0.029). Data are representative of three independent experiments; bar graphs show means ± SEM and single data points. Significance was calculated using Brown Forsythe and Welch ANOVA followed by Games–Howell’s test on (a) and ANOVA followed by Tukey’s post-hoc tests in (b), two-sided Student’s t-test performed in (d), (Welch’s correction) (f) (Welch’s correction) (g, h, i, k) and Log-rank (Mantel–Cox) test in (l). AEC alveolar epithelial cell, BALO bronchoalveolar lung organoid, BALF bronchoalveolar lavage fluid, D day. Source data are provided as a Source Data file.

Fig. 7. PLET1 is present in patients with human IAV-induced ARDS and rPlet1 administration protects lung epithelial barrier function and rescues mice after lethal IAV infection.

a Soluble PLET1 and b total protein concentration in BALF of patients with IAV-induced ARDS (n = 14) or healthy patients (n = 8) (patient characteristics provided in Supplementary Table) (**p = 0.0075, ***p = 0.0005). c PLET1 concentration was negatively correlated with total protein concentration (n = 14). d Schematic representation of the experiments conducted in (e)–(i), rPlet1 (5 µg; squares) or PBS (circles) were applied intratracheally (i.t.) at D3 p.i. The figure was created with BioRender. e Lung histological sections of IAV-infected mice stained with H&E at D7 p.i. Bottom images underneath are magnifications of the top images (red squares); scale bars represent 100 µm (top) and 50 µm (bottom). f Percentage of Annexin V⁺ AECs (n = 6) (**p = 0.0011). g mRNA expression (qPCR) of tight junction component genes in flow-sorted EpCam⁺ cells (n = 3) (*p = 0.0046, ****p < 0.0001). h Quantification of barrier dysfunction by FITC-Albumin fluorescence analysis in BALF (n = 6) (**p = 0.002). i Ki67⁺ percentage of AEC II indicating proliferation (n = 6) (***p = 0.0003). j Survival of IAV-infected mice (5 × 10² pfu) treated with PBS or rPlet1 at D3 p.i. (n = 9/group in two separate experiments) (***p = 0.0001). k Schematic summary: CD206⁺ BMDM2- and BMDM2-derived TR-AM express Plet1 driving AEC II proliferation and upregulation of tight junction genes, ultimately resulting in AEC barrier repair. Application of exogenous, recombinant Plet1 reproduces these effects, highlighting its therapeutic potential. The figure was created with BioRender. Data are representative of three independent experiments showing mean ± SEM. Significance was calculated using two-sided Student’s t-test on (a) (Welch’s correction), (b) (Welch’s correction), (f, h, i) (Welch’s correction), Spearman’s rank correlation test in (c), Log-rank (Mantel–Cox) test in (j), (g) two-way ANOVA (followed by Šídák’s test). AEC alveolar epithelial cell, BALF bronchoalveolar lavage fluid, D day. Panel (k) was generated by BioRender. Source data are provided as a Source Data file.