Fig 2: Single-cell RNA-seq identifies a subtype of tumor-supporting alveolar macrophages (AMs) with high expression of INHBA.a–c UMAP plot of single-cell RNA-seq data of AMs from control and tumor-bearing mice. a A total of 13,413 AM cell transcriptomes were analyzed and hierarchically clustered into 14 subgroups using uniform manifold approximation and projection (UMAP) analysis. b, c Ratio of AMs isolated from control mice (blue) to those isolated from tumor-bearing mice (orange). d Schematic image of RNA-velocity analysis results with scVelo. e, f Distribution of Inhba gene expression levels in AMs isolated from tumor-bearing mice (e) and those isolated from control mice (f), shown using binarization by fitting the expression distribution with Gaussian mixture distribution. g, h In vitro culture of mouse primary AMs from bronchoalveolar lavage fluid with or without damage-associated molecular pattern molecules (DAMPs) from LLC cells. g Flow cytometry histogram of AMs with MARCO. Shaded regions indicate staining with only the second antibody (i.e., without the anti-MARCO antibody). h RT-PCR analysis of Inhba expression in AMs after in vitro culture for 1 day with or without DAMPs (n = 3 per group). Means ± s.e.m. for each group are shown. Symbols represent individual wells (h). Statistical significance was determined using one-way ANOVA with Bonferroni’s post hoc test (h).

Fig 3: INHBA expression in alveolar macrophages (AMs) is induced via MyD88-JNK dependent pathway.a Flow cytometry plots of AMs isolated from C57BL/6 wild-type (WT) mice and MyD88 knockout (KO) mice after inoculation with lung cancer cells. b Inhba RT-PCR analysis of AM isolated from tumor-bearing wild-type mice and MyD88 knockout mice (n = 6 mice for wild-type, n = 5 mice for MyD88 KO). c Inhba RT-PCR analysis in the AM cell line AMJ2-C11 after incubation with or without LLC cell line supernatant (Sup.) and JNK inhibitor SP600125 (n = 4 per group). d–f Immunoblotting of JNK phosphorylation in AM cell line with or without stimulation of LLC cell line supernatant after incubation with each inhibitor (d; TAK-242, TLR4 inhibitor, e; ST2825, MyD88 inhibitor, f; takinib, TAK1 inhibitor). These images were representative of three independent experiments with similar results. g, Schematic representation of Inhba signaling in AM. h Viability of LLC cells treated with the ALK4 inhibitor SB505124 and recombinant activin A (rActA) was assessed using WST-1 assay (n = 4 per group except for the one with only rActA administration (n = 3)). i, j Immunoblotting of ERK (i) and Smad2 phosphorylation (j) in LLC cells stimulated with rActA after incubation with the ALK4 inhibitor SB505124. These images were representative of three independent experiments with similar results. Means ± s.e.m. for each group are shown. Symbols represent individual mice (b) and wells (c, h). Statistical significance was determined using unpaired two-tailed Mann–Whitney U-test (b) or one-way ANOVA with Bonferroni’s post hoc test (c, h).

Fig 4: INHBA upregulation in lung alveolar macrophages (AMs) enhances proliferation of lung cancer cells in vivo.a Definitions of R1, R2, and R3 cells in the lungs of the control and tumor-bearing mice. Left images show schematic diagram of control (Ctrl) mice with PBS (upper) or tumor-bearing mice with LLC cells fluorescently labeled with tdTomato (lower). Dot plots on the right indicate the analysis result of CD45+ autofluorescence+ cells (black rectangular regions in the middle dot plots). The red rectangular region and blue pentagonal region indicate AMs and TAMs, respectively. The data are representative of three independent experiments with similar results. b Principal component analysis of R1, R2, and R3 cells by RNA-Seq (n = 3 mice for R1, R2, and R3 populations). c Heatmaps of tumor-associated macrophage marker genes and AM marker genes (left), and the top 15 upregulated genes from R1 to R2 cells (right). d RT-PCR analysis of Inhba expression in AMs and CD45+ (autofluorescence-) cells isolated from control mice, and AMs, CD45+ (autofluorescence-) cells, and tumor cells isolated from tumor-bearing mice (n = 3 mice per group). e Measurement of activin A concentration in AMs sorted from control or tumor-bearing mice using ELISA (n = 6 per group). f WST-1 cell proliferation assay of LLC cells after administration of each dose of recombinant activin A. Viability of LLC cells in each well with or without recombinant activin A was measured on days 0, 1, and 4. Each value indicates the mean ± s.e.m. of the three wells (n = 4 per group for day 0 and day 1, n = 3 per group for day 4). g Comparison of LLC cell number after 2-day culture with or without conditioned media (Sup.) from shRNA-expressing AM cells (AMJ2-C11) (n = 3 per group). h Representative images of the left lung of tumor-bearing mice treated with PBS (left; control) and follistatin (right). Arrowheads indicate the tumor (surrounded by a dashed line). i Tumor volume of tumor-bearing mice treated with PBS (control) or follistatin (n = 3 mice for control, n = 4 mice for follistatin). Means ± s.e.m. for each group are shown. Symbols represent individual mice (d, e, i) or wells (g). Statistical significance was determined using one-way ANOVA with Bonferroni’s post hoc test (d, g) or unpaired two-tailed t-test (e, i).