New insights into bioaugmented removal of sulfamethoxazole in sediment microcosms: degradation efficiency, ecological risk and microbial mechanisms

Abstract

Background: Bioaugmentation has the potential to enhance the ability of ecological technology to treat sulfonamide-containing wastewater, but the low viability of the exogenous degraders limits their practical application. Understanding the mechanism is important to enhance and optimize performance of the bioaugmentation, which requires a multifaceted analysis of the microbial communities. Here, DNA-stable isotope probing (DNA-SIP) and metagenomic analysis were conducted to decipher the bioaugmentation mechanisms in stabilization pond sediment microcosms inoculated with sulfamethoxazole (SMX)-degrading bacteria (Pseudomonas sp. M2 or Paenarthrobacter sp. R1).

Results: The bioaugmentation with both strains M2 and R1, especially strain R1, significantly improved the biodegradation rate of SMX, and its biodegradation capacity was sustainable within a certain cycle (subjected to three repeated SMX additions). The removal strategy using exogenous degrading bacteria also significantly abated the accumulation and transmission risk of antibiotic resistance genes (ARGs). Strain M2 inoculation significantly lowered bacterial diversity and altered the sediment bacterial community, while strain R1 inoculation had a slight effect on the bacterial community and was closely associated with indigenous microorganisms. Paenarthrobacter was identified as the primary SMX-assimilating bacteria in both bioaugmentation systems based on DNA-SIP analysis. Combining genomic information with pure culture evidence, strain R1 enhanced SMX removal by directly participating in SMX degradation, while strain M2 did it by both participating in SMX degradation and stimulating SMX-degrading activity of indigenous microorganisms (Paenarthrobacter) in the community.

Conclusions: Our findings demonstrate that bioaugmentation using SMX-degrading bacteria was a feasible strategy for SMX clean-up in terms of the degradation efficiency of SMX, the risk of ARG transmission, as well as the impact on the bacterial community, and the advantage of bioaugmentation with Paenarthrobacter sp. R1 was also highlighted. Video Abstract.

Keywords: Antibiotic resistance genes (ARGs); Bioaugmentation; DNA-stable isotope probing; Metagenomics; Sulfonamide.

Fig. 1

Degradation characteristics of SMX in sediment microcosms. Data are means ± standard deviation, n = 3. Treatments A, C, D, and E represent sterilized control, non-inoculated control, inoculated with Pseudomonas sp. M2, and inoculated with Paenarthrobacter sp. R1, respectively. Phases I, II, and III represent the three different SMX addition periods, namely days 0–8, days 8–10, and days 10–11, respectively

Fig. 2

The boxplots show the bacterial alpha diversity indices in sediments (A). Shannon index during incubation (B). The PCoA plot based on the Weighted_Unifrac distance (C). Relative abundance of the 30 largest bacterial genera (D). Temporal dynamics of Pseudomonas sp. M2 and Paenarthrobacter sp. R1 (E). Comparison of the different genus distribution between SMX-amended treatments (treatments C, D, and E) and non-SMX treatment (treatment B) (F). All taxonomic groups except for the top 30 were merged into the “Others” group. ANOVA with an LSD test (P < 0.05) indicates statistically significant differences denoted by different letters for each assessed parameter (A, C). ALDEx2 was used to identify the microbial groups with significant difference between treatments, and “*” indicates P < 0.05 (centered log-ratio transformed, Wilcoxon test, P values were corrected by Benjamini-Hochberg) (F)

Fig. 3

Co-occurrence network analysis showing the biological interactions in each treatment based on pairwise Pearson’s correlations between ASVs (ρ > 0.6) (A). The color and size of each node represent module class and degree value, respectively. Zi-Pi plot showing the distribution of bacterial ASVs based on their topological roles (B). Zi and Pi are within-module connectivity and among-module connectivity. Network hubs: nodes with Zi > 2.5 and Pi > 0.62; Module hubs: nodes with Zi > 2.5 and Pi ≤ 0.62; Connectors: nodes with Zi ≤ 2.5 and Pi > 0.62; Peripheral nodes: nodes with Zi ≤ 2.5 and Pi ≤ 0.62. CSS: Clostridium sensu stricto. Structural robustness (estimated by natural connectivity) of random removal and targeted removal (the impacts of the loss of the keystone species on robustness of treatments B, C, D, and E were respectively consistent with the loss of 19, 19, 12, and 19 species, reaching 20%, 23%, 10%, and 25%) (C)

Fig. 4

Temporal dynamics (A) and inter-treatment differences (B) of bacterial 16S rRNA gene, sulfonamide degradation monooxygenase encoding gene (sadA gene) and resistance genes (sul1 and sul2 genes) based on qPCR results. The relative abundance of antibiotic resistance genes (ARGs) (C) and mobile genetic elements (MGEs) (D) in sediments based on shot gun metagenomic sequencing. The discrete points outside the box in B with values apparently above or below the data range represent “outliers” for the data

Fig. 5

Relative abundance of bacterial 16S rRNA gene (A) and potential SMX-degraders in the treatments inoculated with Paenarthrobacter sp. R1 (B) and Pseudomonas sp. M2 (C) along buoyant density gradients

Fig. 6

Phylogenetic tree of the putative SA-degraders-related bins and their reference genomes based on 400 marker genes (A) and the proposed SMX assimilation pathway (B). Paenarthrobacter sp. R1 was the strain used in this study. The products were predicted according to our previous studies [22], KEGG (map00627 and map00362) and the composition of functional genes in the SMX-degraders