Whole genome sequencing revealed microbiome in lung adenocarcinomas presented as ground-glass nodules

Abstract

Background: Emerging evidence has suggested that dysbiosis of the microbiota may play vital roles in tumorigenesis. However, the interplay between the microbiome and lung cancer remains undetermined. In this study, we characterize the microbiome in the early stage of lung cancer, which presented as ground-glass nodules (GGNs).

Methods: We sequenced the whole genomes from 10 GGN lesions and 5 adjacent normal lung tissue samples. After being filtered with human genome sequences, the sequence reads were mapped to prokaryotic genomes refSeq and non-redundant protein database for taxa and gene functions profiling, respectively.

Results: Mycobacterium, Corynebacterium, and Negativicoccus were the core microbiota found in all GGNs and the normal tissue samples. The microbiota composition did not show significant difference between GGNs and normal tissues except the adenocarcinoma (AD) (P=0.047). A significant β diversity in microbiome gene functions was found among different patients. Two individual gene functions, the Secondary Metabolism (1.32 fold with P=0.01) and the Serine Threonine protein kinase (4.23 fold, P<0.001), were significantly increased in GGNs over normal tissue samples.

Conclusions: This study helps shed light on the implication of the microbiome in early stage lung cancer, which encourages the further study and development of innovative strategies for early prevention and treatment of lung cancer.

Keywords: Lung cancer; ground-glass nodule (GGN); microbiome; whole genome sequencing (WGS).

Figure 1

Phylogenetic tree of microbiota taxa in GGNs. The sequence reads were mapped on NCBI prokaryotic genomes and assigned to different levels of taxa by NCBI taxonomy library using Megan 6. The different levels of taxa were displayed by Megan 6: Phylum, Class, Order, Family, Genus, and Species. Each box contains 15 sample bars ordered as AAH1, AAH2, AAH3, AAH4, AAH5, AIS1, AIS2, MIA3, AD4, AD5, NOR1, NOR2, NOR3, NOR4, NOR5. The bar heights represent the normalized read numbers. GGN, ground-glass nodule.

Figure 2

Principal co-ordinate analysis (PCoA) of β diversity of microbiota composition. The phylogenetic distances between samples were calculated by unweighted UniFrac algorithm and displayed by PCoA in QIIME2 package. (A) PCoA of patients. Colored dots represent 5 patient populations. Patients were not obviously classified into groups by any of the three axis; (B) PCoA of in GGNs and normal samples. Colored dots represent 5 GGN types or normal sample populations. The two and 5 normal samples were grouped and separately by the main Axis 1. GGN, ground-glass nodule.

Figure 3

Composition of eggNOG modules in each GGN and normal tissue sample. The sequence reads were assigned to the eggNOG (orthologous groups and functional annotation) and displayed by heatmap in Megan 6. The sequence reads were normalized to 100% scale. A few eggNOGs had a similar proportion among all samples, such as [E], [G] (COG one letter Code description), etc. However, [K], [Z] had different proportion among different samples. GGN, ground-glass nodule.

Figure 4

Kruskal-Wallis test for the composition evenness using QIIME2. The Pielou’s evenness indexes (described in Methods) of 5 patients were calculated and tested for significance between samples by Krustal-Wallis test. Patient groups with 3 samples each were pairwise tested. The Y-axis is the Pielou’s index. Patient 1 (p1) or p2 versus p3 or p4, and p3 versus p5 are significant (*P<0.05).

Figure 5

Gene function annotation. The sequence reads were assigned to the SEED subsystems by Megan 6. (A) Clustering of SEED subsystems in all GGN and normal tissue samples. The sequence reads were normalized by z-scores of GGN samples. Four samples clusters and 4 subsystem clusters were shown by the z-score profiles. However, the sample clusters show neither classifications of GGN types nor different patients. (B) Comparing the gene function annotation between GGNs and normal samples by t-test. Significances were found between the Secondary Metabolism of SEED annotation in GGNs (SM GGN) and in normal sample (SM NOR) and between the COG0515 Serine Threonine protein kinase annotated of eggNOG in GGN (STK GGN) and in normal control tissues (STK NOR). GGN, ground-glass nodule.

Figure S1

Correlation between species richness and abundance. The number of species taxa (species richness) was count in each sample. The percentage of reads mapped on prokaryote genomes for each sample (Table 1). Pearson correlation between the species richness and the percentage of mapped reads was performed. A strong negative (r=–0.84) correlation was found.

Figure S2

Microbiota community diversity in all samples. The filtered sequence reads from whole genome sequencing data were mapped on NCBI prokaryote genomes. The sequence reads assigned on operational taxonomic units (OTUs) by assignment algorithm lowest common ancestor (LCA) using Megan 6 software package. The reads on taxa in 15 samples were displayed by Box plot. Top three taxa were identified.

Figure S3

The top abundant eggNOG function modules in all samples. The filtered sequence reads were mapped on eggNOG database. The reads were assigned to eggNOG annotations using Megan 6. The reads on each eggNOGs of15 samples were displayed by Box plot. The eggNOGs (COG one letter Code description) were ranked by read abundance assigned to.

Figure S4

PCoA of composition of eggNOG annotation. The mapped sequence reads were assigned to eggNOG annotations. The Jaccard distances between patients were calculated using eggNOG composition in each patient and displayed by PCoA in QIIME2 package. Patients such as p5 were dispersed and were not classified into groups by any of the three axis.

Figure S5

PCoA of composition of SEED annotation. The mapped sequence reads were assigned to SEED subsystem annotations. The Jaccard distances between patients were calculated using SEED composition in each patient and displayed by PCoA in QIIME2 package. Patients were not classified into groups by any of the three axis.

Figure S6

PCoA of eggNOG composition. The normal samples (purple) were separated from GGNs. The mapped sequence reads were assigned to eggNOG annotations. The Jaccard distances between GGN types were calculated using eggNOG composition in each GGN type or normal tissue group and displayed by PCoA in QIIME2 package. GGN types were not clustered into groups by any of the three axis but the normal samples (purple) were separated from GGNs. GGN, ground-glass nodule.