Distinct bacterial community structures and arsenic biotransformation gene profiles in dust

Abstract

Introduction: Microorganisms, which are ubiquitous in the environment, have evolved a diverse array of arsenic biotransformation genes (ABGs). Dust harbors a wide range of microorganisms. However, the distinct characteristics of bacterial community structures and ABG profiles in dust, compared with those in other environments such as soil and water, remain poorly understood.

Methods: In this study, dust samples were simultaneously collected alongside surrounding soil and seawater samples in Xiamen, a coastal city of China, to investigate the distinct profiles and potential sources of bacterial communities and ABGs in dust using 16S rRNA gene amplicon sequencing and metagenomic sequencing.

Results and discussion: Abundant and diverse bacterial communities and ABGs were detected in dust, revealing significant differences in community structures and ABG profiles compared with those in soil and seawater. Soil was identified as the primary source for both bacterial communities and ABGs in dust through fast expectation-maximization microbial source tracking (FEAST). Acetobacteraceae, which showed significantly greater relative abundance (p < 0.001) in dust than in soil and seawater, was also identified as a keystone taxon in the dust bacterial co-occurrence network. Furthermore, metagenome-assembled genomes (MAGs) affiliated with Acetobacteraceae were effectively recovered from dust via metagenomic binning, and these MAGs harbored an array of ABGs, indicating that Acetobacteraceae could be important hosts for ABGs in dust. Overall, our findings offer new insights into bacterial communities and ABGs in dust, thereby improving our understanding of arsenic biogeochemical cycling.

Keywords: arsenic; arsenic biotransformation genes; bacterial communities; dust; metagenomes.

Figure 1

Study area and sampling sites.

Figure 2

Taxonomic compositions and diversity of bacterial communities in dust, soil, and seawater samples. Taxonomic compositions of the bacterial communities (A). Alpha diversity (Shannon index) of the bacterial communities (B). The Wilcoxon rank-sum test was used to compare the differences in Shannon indices of the bacterial communities among dust, soil, and seawater samples (*p < 0.05, **p < 0.01, ***p < 0.001). Beta diversity (principal coordinate analysis (PCoA) based on the Bray–Curtis distance) of bacterial communities (C).

Figure 3

Abundance, diversity, and taxonomic compositions of arsenic biotransformation genes (ABGs). The abundances (genes per million, GPM) of ABGs (A). Alpha diversity (Shannon index) of ABGs (B). The Wilcoxon rank-sum test was used to compare the differences in abundances and Shannon indices of ABGs among dust, soil, and seawater samples (*p < 0.05, **p < 0.01, ***p < 0.001). Beta diversity (principal coordinate analysis (PCoA) based on the Bray–Curtis distance) of ABGs (C). Taxonomic compositions of ABGs at the phylum level (D).

Figure 4

Fast expectation–maximization microbial source tracking (FEAST) analysis for bacteria (A) and ABGs (B). The direction of the arrows indicates the relationship between sources and sinks, and the percentages reflect the contribution of each source.

Figure 5

Key bacterial taxa and arsenic biotransformation genes (ABGs) and the co-occurrence networks of bacterial taxa and ABGs across dust, soil, and seawater. Random forest classification models identifying key bacterial taxa at the family level (A) and ABGs (B). The panels on the left illustrate the relative abundance of these families or the relative proportions of ABGs. The panels on the right show the mean decrease in accuracy for each family or ABG, indicating their importance in the model (*p < 0.05 and **p < 0.01). Network analysis revealing the co-occurrence patterns between ABGs and bacterial communities in dust (C), soil (D), and seawater (E) based on Spearman correlation analysis (|r| > 0.50, p < 0.05).

Figure 6

Phylogenetic tree of the detected bacterial metagenome-assembled genomes (MAGs) with completeness greater than 80% and contamination less than 10% in dust. The line bar indicates a tree scale of 0.2, and the purple dots represent bootstrap values. The heatmap illustrates the presence (blue) or absence (light blue) of arsenic biotransformation genes (ABGs) for each MAG. The bars enclosed by dashed rings depict the relative abundance of each MAG in dust.