Assessing the Risks of Potential Pathogens and Antibiotic Resistance Genes Among Heterogeneous Habitats in a Temperate Estuary Wetland: a Meta-analysis

Abstract

Temperate estuary wetlands act as natural filters for microbiological contamination and have a profound impact on "One Health." However, knowledge of microbiological ecology security across the different habitats in temperate estuarine wetlands remains limited. This study employed meta-analysis to explore the characteristics of bacterial communities, potential pathogens, and antibiotic resistance genes (ARGs) across three heterogeneous habitats (water, soil, and sediment) within the Liaohe Estuary landscape. The diversity and composition of the three bacterial communities differed with biogeography, temperature, and pH, with the highest α-diversity showing a significantly negative correlation along latitude in soil. Furthermore, aminoglycosides were significantly enriched in water and soil, while dihydrofolate was more likely to be enriched in soil. The potential pathogens, Pseudoalteromonas and Planococcus, were dominant in water and sediment, while Stenotrophomonas was the dominant bacterium in soil. The network topology parameter revealed interspecific interactions within the community. PLS-PM highlights the main direct factors affecting the abundance of potential pathogens and the spread of ARGs, while temperature and pH indirectly influence these potential pathogens. This study advances our understanding of bacterial communities in estuarine wetlands, while highlighting the need for effective monitoring to mitigate the risks associated with potential pathogens and ARGs in these ecosystems.

Keywords: Antibiotic resistance genes; Biogeography; Co-occurrence; Potential pathogens; Risk assessment; Temperate estuary.

Fig. 1

The geo-locations and types of samples collected from seven published studies across three distinct habitats

Fig. 2

Diversity patterns of the bacterial community. a Alpha diversity indices for bacterial communities in water, sediment, and soil (including observed species, Shannon, and PD whole tree indexes). b The relationship between alpha diversity index and latitude across different habitats. Levels of significance: *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 3

Beta diversity patterns across different habitats: non-metric multidimensional scaling (NMDS) showing the community structure of different habitats (a), and the trends with pH (b) and temperature (Temp) (c), based on the Bray–Curtis dissimilarity. Curves illustrating the relationship between community similarity (1—“Bray–Curtis” distance) and geographic distance (d), ∆pH (e), and ∆temperature (f) across different habitats. Residual standard error is used as an indicator of goodness of fit. The Spearman correlation coefficient (ρ) was calculated (P values were all less than 0.01). The curves for water, sediment, and soil are colored red, orange, and green, respectively. The relative standard error (RSE) of the logistic regression was also used to evaluate the goodness of fit of the model

Fig. 4

Species distribution patterns in the three habitats. a Relative abundance of dominant bacterial communities across the three different habitats at the phyla/proteobacterial class level. b Histogram of LDA score distribution, showing species with significant abundance differences among the three habitat groups. The length of the histogram represents the influence of significantly different species, with a threshold value of three. c LEfSe analysis of microbial abundance between sediment, soil, and water samples. Circles radiating from the inside out represent different taxonomic levels (from kingdom to genus). Each small circle represents a classification at that level, with the diameter size of the small circle proportional to the relative abundance. Species with no significant differences are uniformly colored white, while different species are colored according to their respective groups. The species names of biomarkers not shown in the figure are displayed on the right, with the corresponding letter–number codes matching the figure

Fig. 5

Effects of pH and temperature on pathogenic community compositions. a Compositions of the dominant pathogenic bacterial genera in the different habitats. Effects of pH, temperature, and microbial community on pathogenic bacteria in the different habitats based on partial least squares path model (PLS-PM). b Differences in the response of pathogens to pH, temperature, and microbial changes under the total effect of PLS-PM. R² indicates the degree explained by their independent latent variables. Red and blue lines represent direct and indirect effects, respectively. Solid and dashed lines represent significant and insignificant paths, respectively (P < 0.05 and P > 0.05). The thickness of the line represents the absolute value of the path coefficients. c Bar graphs showing the direct, indirect, and total effects of significant pH, temperature, and microbe factors on the pathogenic community, analyzed separately for water, sediment, and soil habitats. The goodness-of-fit index for water, sediment, and soil is 0.656, 0.821, and 0.765, respectively. All data were log-transformed

Fig. 6

Distribution (a) and ternary mapping (b) of ARGs and MGEs in the different habitats. The color of the circle corresponds to the ARGs and MGEs, and the size represents their abundance. The position of each circle reflects the proportion of its abundance in different samples

Fig. 7

Microbial interactions in the different habitats. Network analysis showing co-occurrence patterns among microorganisms, pathogens, ARGs, and MGEs across the different habitats (a all, b sediment, c soil, and d water). Different colored nodes represent different microbial taxa (round), pathogenic bacteria (triangle), ARGs (square), and MEGs (diamond). The size of the circle represents the relative abundance. The size of each node is proportional to the number of connections, and the thickness of each connection between two nodes is proportional to the value of Spearman’s rank correlation coefficient. The pink solid line represents a positive correlation, while the blue solid line represents a negative correlation