Pleural cavity macrophages promote lung tumor establishment through tissue invasion

Graphical abstract

1

Dear Editor,

Preclinical studies have elegantly demonstrated that distinct populations of cavity-resident macrophages in the peritoneal and pleural spaces are ontogenically, transcriptionally, and functionally related (Buechler et al., 2019). Notably, these cavities are recognized as immunosuppressive environments that commonly facilitate cancer progression (Donnenberg et al., 2019; Morano et al., 2016; Porcel et al., 2015). Cavity-resident macrophages mediate a physiological checkpoint that limited anti-tumor activity at these cancer sites (Chow et al., 2021). Previous genetic lineage tracing has shown that, post-injury, cavity macrophages tend to accumulate on the surfaces of visceral organs, including the lungs (Jin et al., 2021, 2022), rather than deeply infiltrating the parenchyma (Deniset et al., 2019; Wang and Kubes, 2016). However, it remains unclear whether, in the context of tumors, these cavity macrophages penetrate into the lung parenchyma and promote tumor growth. Utilizing dual recombinase-mediated genetic lineage tracing, we observed the infiltration of cavity macrophages into lung tumor metastases. Furthermore, genetic ablation or sequestration of these cavity macrophages significantly reduced tumor growth in the lungs. This observation underscores the crucial role that cavity macrophages play in supporting tumorigenic processes, suggesting that targeting these cells may represent a viable therapeutic strategy for mitigating lung tumor progression.

To simulate the recruitment and invasion of GATA6⁺ cavity macrophages in the B16F10 melanoma lung metastasis mouse model, we employed the Gata6-rox-Stop-rox-CreER (Gata6-iCreER) knock-in mice. In this model, Dre-rox removes a rox-flanked Stop sequence, yielding a Gata6-CreER allele. We crossed these mice with CD45-Dre (Jin et al., 2021) and R26-tdTomato (R26-tdT) (Madisen et al., 2010) mice to enable specific lineage tracing of pleural cavity macrophages (Fig. 1A). CD45 markers can distinguish macrophages (CD45⁺) from CD45⁻ organ cells like hepatocytes, cardiomyocytes, etc. The GATA6 marker can differentiate cavity macrophages (GATA6⁺) from recruited monocytes or tissue-resident macrophages (GATA6⁻). Therefore, theoretically, double-positive cells (CD45⁺GATA6⁺) would be cavity macrophages, distinct from tissue-resident macrophages, monocytes, and other cell lineages such as hepatocytes, cardiomyocytes, or lung epithelial cells (Fig. 1B). For proof of principle, we crossed CD45-Dre; Gata6-iCreER mice with R26-tdT mice and analyzed the labeling of monocytes and macrophages 2 weeks after tamoxifen (Tam) treatment (Fig. 1C). Immunofluorescent (IF) staining revealed that the majority of pleural macrophages expressing GATA6 and F4/80 were tdT⁺ in Tam-treated CD45-Dre; Gata6-iCreER; R26-tdT mice (Fig. 1D). Quantitative analysis revealed that all tdT⁺ cells expressed F4/80, and 79.2% of F4/80⁺ cells were positive for tdT (Fig. 1E). Virtually fluorescence-activated cell sorter (FACS) analysis showed all tdT⁺ pleural cells (>99.5%) were CD11b⁺F4/80⁺ (Fig. 1F and 1G), indicating high specificity of CD45-Dre; Gata6-iCreER for labeling cavity macrophages. Adult CD45-Dre; Gata6-iCreER; R26-tdT mice (8 weeks old) were administered Tam for 1 week to induce gene recombination. Two weeks after the final Tam treatment, mice were intravenously injected with 2 × 10⁵ B16F10 melanoma cells via the tail vein to establish a lung metastasis model (Fig. 1H). In lungs collected from mice treated with phosphate-buffered saline (PBS), we did not detect tdT expression in CD45⁺ cells, excluding ectopic labeling of resident macrophages or other hematopoietic cell lineages in visceral organs (Fig. 1I). Notably, whole-mount fluorescent imaging revealed prominent tdT⁺ signals covering the tumor surface (Fig. 1J). Flow cytometric analysis showed the presence of infiltrated pleural cavity macrophages in the melanoma (Fig. 1K). Quantification data indicated that tdT⁺ macrophages constituted approximately 3.6% of all lung macrophages in the mice inoculated with B16F10 melanoma cells, compared to minimal tdT⁺ macrophages detected in control lungs without tumor inoculation (Fig. 1L). Immunostaining for tdT and F4/80 on lung sections revealed an accumulation of tdT⁺ cavity macrophages within tumors at the lung periphery (Fig. 1M). Importantly, pleural cavity macrophages did not infiltrate the normal (non-tumor) alveolar region or the adventitial space, suggesting their specific involvement in tumors. Quantification of the immunofluorescence-stained images of lung edge tumors showed that tdT⁺ cells constituted about 50% of the macrophages infiltrating the tumors (Fig. 1N), indicating a significant involvement of pleural cavity macrophages in the tumor growth and suggesting a crucial role in shaping the tumor microenvironment. To further investigate whether pleural cavity macrophages are able to infiltrate other types of tumors, we employed two additional tumor cell lines, the mouse hepatoma cell line Hepa1-6 (Hepa1-6) and Lewis lung Carcinoma (LLC) (Fig. S1), both of which are capable of forming lung metastases. Three weeks after tumor cell injection, whole-lung fluorescence imaging of CD45-Dre; Gata6-iCreER; R26-tdT mice revealed a strong accumulation of red fluorescent signals over the tumor regions (Fig. S1A, S1B, S1E, and S1F). This pattern was consistently observed in both Hepa1-6 and LLC lung metastasis models. Further analysis of lung sections by immunofluorescence staining for F4/80 and tdT revealed that pleural cavity macrophages contributed to approximately half of the macrophage population within the tumor regions (Fig. S1C, S1D, S1G, and S1H). These findings are consistent with those observed in the B16F10 melanoma lung metastasis model, suggesting a generalizable role of pleural cavity macrophages across different tumor types.

To specifically elucidate the role of GATA6⁺ cavity macrophages in influencing tumor growth, we aimed to selectively ablate these cells during the progression of tumor development. We engineered a Gata6-RSR-tdT-DTR knock-in mouse line, wherein a genetic construct driven by the Gata6 promoter enables the expression of both the diphtheria toxin receptor (DTR) and tdT following the excision of a rox-Stop-rox sequence in Dre-expressing cells (Fig. 2A). In the CD45-Dre; Gata6-RSR-tdT-DTR mice, Dre-rox recombination specifically removes the RSR cassette in CD45⁺ cell lineages, resulting in the exclusive expression of tdT and DTR in GATA6⁺ pleural or peritoneal cavity macrophages (Fig. 2A and 2B). To assess the efficiency of tdT⁺ cell ablation, adult mice were treated with either diphtheria toxin (DT) or PBS (Fig. 2B). Immunostaining of isolated cavity cells demonstrated the colocalization of tdT and DTR expression within peritoneal and pleural cavity macrophages (Fig. 2C), confirming the successful creation of a cavity macrophage-specific DTR expression system. Additional IF staining and flow cytometric analysis consistently showed that the majority of pleural and peritoneal macrophages expressing GATA6 and F4/80 were tdT positive (Figs. 2D, S2A and S2B). In DT-treated Cd45-Dre; Gata6-RSR-tdT-DTR mice, tdT⁺ cavity macrophages were markedly reduced within 3 days, as confirmed by immunostaining for tdT, compared with PBS controls (Fig. 2D and 2E). Flow cytometric analysis further confirmed a significant reduction in the percentage of tdT⁺ macrophages among CD45⁺ peritoneal or pleural cells (Fig. 2F and 2G). Considering the concern regarding the potential resistance of immune cells to DT-mediated depletion, we sought to evaluate the duration of DT-induced depletion of cavity macrophages. We accordingly evaluated the numbers of pleural and peritoneal cavity macrophages 3 weeks after DT administration, ensuring consistency in the experimental timeline (Fig. S2C). Flow cytometry analysis demonstrated that, 3 weeks after DT treatment, the proportions of pleural and peritoneal cavity macrophages remained significantly lower than those in the PBS control group, indicating a sustained depletion effect (Fig. S2D and S2E). Moreover, an absence of tdT expression in GATA6⁺ cells across all examined visceral organs from CD45-Dre; Gata6-RSR-tdT-DTR mice confirmed that there was no ectopic labeling of ­tissue-resident macrophages or hematopoietic cell lineages in these organs (Fig. S2F and S2G). Overall, the development of the CD45-Dre; Gata6-RSR-tdT-DTR mouse line facilitates specific and efficient ablation of endogenous cavity macrophages.

We next examined the role of pleural cavity macrophages in the B16F10 melanoma lung metastasis mouse model by treating CD45-Dre; Gata6-RSR-tdT-DTR mice with DT or PBS (Fig. 2H). DT-treated CD45-Dre; Gata6-RSR-tdT-DTR mice exhibited significant resistance to B16F10 cell metastasis in the lungs compared to PBS-treated controls (Fig. 2I). By day 21, whole-mount bright field imaging showed significantly fewer metastatic melanoma lesions on the lung surface in DT-treated mice (Figs. 2J, S3A and S3B). Additionally, Masson-Fontana staining revealed a marked decrease in the area of lung metastasis in DT-treated CD45-Dre; Gata6-RSR-tdT-DTR mice compared to the control group (Fig. S3A and S3B). These mice also demonstrated an extended survival rate (Fig. 2K). Given that GATA6 is a key transcription factor influencing the survival and proliferation of peritoneal macrophages (Rosas et al., 2014), we proceeded to knock out Gata6 in cavity macrophages. We crossed CD45-Dre; Gata6-iCreER; R26-tdT mice with Gata6-flox mice (Sodhi et al., 2006), and treated the offspring, CD45-Dre; Gata6-iCreER/flox; R26-tdT mice, with Tam between weeks 8 and 9 of age to simultaneously label cavity macrophages and delete Gata6 (Fig. 2L). Two weeks later, these mice were inoculated with B16F10 melanoma cells. Littermate CD45-Dre; Gata6-iCreER; R26-tdT mice served as control. To examine how Gata6 gene knockout affects the number of pleural cavity macrophages, we performed flow cytometric analysis to quantify the absolute numbers of pleural cavity macrophages, peritoneal macrophages, and tumor-infiltrating pleural macrophages. We found that specific deletion of Gata6 led to a marked reduction in the number of cavity macrophages in the peritoneal and pleural spaces, as well as within the tumor (Fig. S3C and S3D). Whole-mount bright field images and Masson-Fontana staining of lungs showed that a significant reduction of tumor numbers in cavity macrophage-specific Gata6 knockout mice compared to control mice (Fig. 2M, 2N, S3E and S3F). Additionally, the survival of CD45-Dre; Gata6-iCreER/flox; R26-tdT mice was notably increased compared to the control mice (Fig. 2O). Taken together, the ablation of cavity macrophages or knockout of Gata6 in these cells resulted in reduced metastasis, decreased tumor burden, and notably extended survival rates.

To assess the role of pleural cavity macrophages in established tumors, we initiated DT treatment 2 weeks after intravenous inoculation of melanoma cells, allowing sufficient time for tumor establishment, and performed analysis in the third week (Fig. S4A). Whole-mount bright field imaging revealed a trend of decreased lung metastases in DT-treated mice compared to PBS controls, but without statistical significance (Fig. S4B–D). Further Masson-Fontana staining revealed a slight decrease in tumor area, and the survival curve suggested a potential improvement in survival following DT treatment (Fig. S4B–F). Based on the above experimental results, DT-mediated depletion of pleural cavity macrophages after tumor establishment showed limited efficacy in suppressing tumor progression. These findings suggest that pleural cavity macrophages play a pivotal role during the early establishment phase of lung tumors, facilitating initial tumor growth and microenvironmental conditioning. Their influence appears to diminish but may still persist to a limited extent once tumors are more established. Therefore, targeting these cells may be most effective at early stages of tumor development. To investigate the role of cavity macrophages in the development of primary tumors derived from inoculated melanoma cells, we directly injected B16F10-Luc cells into the pleural cavity, where they formed primary tumors (Fig. S4G). This model allowed us to explore the specific functions of pleural cavity macrophages in the local tumor microenvironment. Bioluminescence imaging showed that depletion of pleural cavity macrophages markedly inhibited the growth rate of primary tumors (Fig. S4H and S4I). Taken together, these results indicate that pleural cavity macrophages promote the growth of both metastatic and ­primary melanoma tumors.

To explore the mechanisms behind the role of cavity macrophages in tumorigenic processes, we included a comprehensive immune profiling analysis to investigate how depletion of cavity macrophages contributes to reduced tumor growth in the B16F10 melanoma lung metastasis model. Flow cytometry was used to analyze changes in the proportions of immune cells in lung tumors and pleural cells following the depletion of cavity macrophages (Fig. S5A). We examined the changes in the proportions of lymphoid and myeloid cells separately. For myeloid cells, monocytes were identified as CD11b⁺LY6G⁻LY6C⁺, neutrophils as CD11b⁺LY6G⁺, and macrophages as F4/80⁺ cells. For lymphoid cells, T cells were identified as CD3⁺, B cells as CD19⁺, and natural killer (NK) cells as CD3⁻CD19⁻NK1.1⁺ cells. Compared with mice not treated with DT, we found no significant changes in the proportions of neutrophils and macrophages among myeloid cells in lung tumors (Fig. S5B). However, the proportion of monocytes was increased, which we speculate may be due to a compensatory response in which monocytes capable of differentiating into macrophages partially compensate for the loss of cavity macrophages. Among lymphoid cells, we observed an increased proportion of T cells, a decreased proportion of B cells, and no significant change in the proportion of NK cells (Fig. S5B). We also analyzed cells from the pleural cavity. In DT-treated mice, macrophages were nearly undetectable, while the proportion of monocytes was markedly increased. In contrast, the proportion of neutrophils remained largely unchanged (Fig. S5C). For lymphoid cells, the changes in cell proportions within the pleural cavity were similar to those observed in the tumor. The proportions of T cells and NK cells were increased, while the proportion of B cells was reduced (Fig. S5D). Given the observed increase in T-cell proportions and the reduced tumor growth following macrophage depletion, we further examined the proportion of cytotoxic CD8⁺ T cells. We found that the depletion of pleural cavity macrophages led to an upregulation of cytotoxic T cells in both the tumor and the pleural cavity (Fig. S5E and S5F). Based on the above analysis of immune cells in the tumor and pleural cavity, we propose that depletion of pleural cavity macrophages promotes the accumulation of cytotoxic T cells, thereby enhancing anti-tumor immune responses and contributing to the suppression of tumor growth.

This study effectively utilized a dual recombinase-mediated genetic system to trace pleural cavity macrophages in a B16F10 melanoma lung metastasis mouse model. Fate mapping demonstrated that these cavity macrophages significantly infiltrate the tumor, constituting a substantial portion of its macrophage population. Crucially, the genetic ablation or knockout of Gata6 in these macrophages led to a decreased tumor burden and improved survival rates. These findings underscore the pivotal role of cavity macrophages in tumor growth and metastasis dynamics, suggesting that targeting these cells may represent a promising therapeutic strategy for tackling tumor progression. The cellular dynamics observed in our work indicate that the molecular mechanisms driving the migration and functional activities of cavity macrophages merit further investigation, potentially revealing new therapeutic targets per se.

References

[1] Buechler MB , Kim KW , Onufer EJ et al. A stromal niche defined by expression of the transcription factor WT1 mediates programming and homeostasis of cavity-resident macrophages. Immunity 2019; 51: 119– 130.e5.

[2] Chow A , Schad S , Green MD et al. Tim-4⁺ cavity-resident macrophages impair anti-tumor CD8⁺ T cell immunity. Cancer Cell 2021; 39: 973– 988.e9.

[3] Deniset JF , Belke D , Lee WY et al. Gata6⁺ pericardial cavity macrophages relocate to the injured heart and prevent cardiac fibrosis. Immunity 2019; 51: 131– 140.e5.

[4] Donnenberg AD , Luketich JD , Dhupar R et al. Treatment of malignant pleural effusions: the case for localized immunotherapy. J Immunother Cancer 2019; 7: 110.

[5] Jin H , Liu K , Huang X et al. Genetic lineage tracing of pericardial cavity macrophages in the injured heart. Circ Res 2022; 130: 1682– 1697.

[6] Jin H , Liu K , Tang J et al. Genetic fate-mapping reveals surface accumulation but not deep organ invasion of pleural and peritoneal cavity macrophages following injury. Nat Commun 2021; 12: 2863.

[7] Madisen L , Zwingman TA , Sunkin SM et al. A robust and high-throughput Cre reporting and characterization ­system for the whole mouse brain. Nat Neurosci 2010; 13: 133– 140.

[8] Morano WF , Aggarwal A , Love P et al. Intraperitoneal immunotherapy: historical perspectives and modern therapy. Cancer Gene Ther 2016; 23: 373– 381.

[9] Porcel JM , Gasol A , Bielsa S et al. Clinical features and survival of lung cancer patients with pleural effusions. Respirology 2015; 20: 654– 659.

[10] Rosas M , Davies LC , Giles PJ et al. The transcription factor Gata6 links tissue macrophage phenotype and proliferative renewal. Science 2014; 344: 645– 648.

[11] Sodhi CP , Li J , Duncan SA. Generation of mice harbouring a conditional loss-of-function allele of Gata6. BMC Dev Biol 2006; 6: 19.

[12] Wang J , Kubes P. A reservoir of mature cavity macrophages that can rapidly invade visceral organs to affect tissue repair. Cell 2016; 165: 668– 678.

RIGHTS & PERMISSIONS

© The Author(s) 2025. Published by Oxford University Press on behalf of Higher Education Press.