Microbiome Related Cytotoxically Active CD8+ TIL Are Inversely Associated With Lung Cancer Development

Abstract

Lung cancer is the most common cancer type around the world. Although major advances in cancer therapy, lung cancer has been the largest proportion of all cancer-related deaths. The respiratory tract contains many types of bacteria and a distinct lung microbiome in lung cancer patients was described in many studies. The specific roles of these lung microorganisms in lung cancer progression remain unclear. In this study, we evaluated the effect of inhalation of bronchoalveolar fluid (BAL) in the lung cancer cell growth. The microbiome-based immune and carcinogenesis was examined in tumor-bearing mouse model. We found that inhalation of BAL collected from non-small cell lung cancer (NSCLC) patients altered the lung microbiota and inhibited tumor cell growth. The inhibitory effect was due to the infiltration of CD3 and CD8⁺ T cells and decrease of M2 macrophages in lungs. The microbial communities of NSCLC BAL inhalation group were dominated by Pasteurella, whereas the microbial communities of non-cancer control and PBS inhalation group were dominated by Delftia. Linear discriminant analysis (LDA) indicated that the genera Pasteurella, Pseudomonas, and Chryseobacterium were increased in NSCLC BAL inhalation group, while genera Delftia, Ezakiella, Blautia, Cloacibacterium, and Microvirga et al. were increased in PBS and Non-cancer group. We demonstrated a significant positive correlation between Pasteurella and cytotoxic CD8⁺ TIL and a negative correlation with M2 macrophages. Coriobacteriaceae was positively correlated with M2 macrophages and negatively correlated with CD8⁺ cells. The abundance of Pasteurella was negatively correlated with tumor cell growth. Our findings provide a promising strategy that can be used as a therapeutic vaccine for lung cancer patients.

Keywords: Pasteurella; TIL; lung cancer; microbiome; therapeutic vaccine.

Figure 1

BAL from lung cancer patients inhibited growth of lung cancer cells. (A) BAL was collected from bronchial and alveolar spaces. (B) Experimental protocol for construction of animal model and treatment scheme. The C57BL/6 mice were pretreated with broad-spectrum antibiotics cefoperazone before aerosolizing BAL. Five days after aerosolization, synergistic luci-LLC cells were intravenously injected into the mice. The in vivo imaging was used to measure the growth of cancer cells in lungs. (C) Bioluminescent images were shown. (D) Activities of luciferase were shown in scatter diagram. (E) Lung metastatic nodules from mice. (F) The number of metastatic tumor modules formed in the lungs was counted and displayed in scatter plot. Data were analyzed with Student’s t test, *p < 0.05, ***p < 0.001. (G) HE staining of lung section.

Figure 2

BAL from NSCLC patients increase CD3⁺ and CD8⁺ T cell infiltration. Formalin fixed, paraffin embedded mouse lung sections were stained with CD8 and CD3 antibodies and nuclei were stained with DAPI. (A) CD3 (green) and CD8 (red) positive T cell subpopulations. The percentage of positive CD3 (B) and CD8 (C) were graphed for NSCLC group, NC group and non-cancer control group. Student’s t test, ***p < 0.001; (D) Positive correlation between number of CD3 and CD8. The area of positive CD3 (E) and CD8 (F) was negatively associated with tumor burden.

Figure 3

Lung microbiome composition after BAL inhalation. (A) Venn diagram representation of the OUT from NSCLC BAL inhalation group, non-cancer control group, and PBS group. The percentage of core lung microbiota genus are shown. (B) Principal Component Analysis of OTUs. (C) Analysis of similarities (ANOSIM) demonstrated the difference in the lung microbiota between NSCLC group and non-cancer control or PBS group. ***p < 0.001; ns, no significance. (D) The composition of major taxonomic groups at genus level in three groups. (E) The composition of each sample based on the taxonomic assignment of 16s rDNA sequences. The X-axis represents the abundance of each taxon. A: non-cancer controls; B: NSCLC patients; C: PBS.

Figure 4

Distinct taxa identified in three groups using LEfSe analysis. (A) LDA scores of difference between groups. (B) Cladogram shows the enrichment of bacteria in three groups. (C) Network of co-occurring microbes was constructed in Cytoscape by correlation data R. Nodes represent microbes, and edges represent the strength of positive correlations between microbes. Correlation between abundance of Pasteurella and tumor burden ^#***p < 0.001; ns, no significance. (D), CD3⁺ T cells (E), and CD8⁺ T cells (F). Correlation between abundance of Coriobacteriaecae and tumor burden (G) and CD8⁺ T cells (H).

Figure 5

Correlation between immune cells and specific genus in lungs. (A) The percentage of CD8+ T cells was determined by flow cytometry. Each dot represents the result for an individual mouse (B) Student's t test, *p < 0.05. The correlation between abundance of Pasteurella and CD8+ T cells (C) and M2 macrophage (D) was shown. (E) The percentage of M2 macrophages was determined by flow cytometry. Each dot represents the result for an individual mouse (F). Student's t test, *p < 0.05. The correlation between abundance of Coriobacteriaceae and CD8+ T cells (G) and M2 macrophage (H) was shown.