Download:
-
Antibiotic resistance has emerged as a significant threat to global public health. The excessive use of antibiotics is known to hasten the spread of antibiotic resistance. The selective pressure exerted by antibiotics is exceptionally high in acute care hospitals, where an estimated 20%–30% of inpatients in Europe are administered antibiotics during their hospitalization (1). Hospital wastewater typically contains higher concentrations of antibiotics and antimicrobials than municipal wastewater (2–4), making it a critical reservoir for antibiotic resistant bacteria (ARB) and a conduit for the proliferation of these resistant strains (2,5). The occurrence of pathogens resistant to "last-resort" antibiotics in hospital settings, combined with the nosocomial transmission of such resistant mechanisms, represents a profound risk to patient health and the surrounding environment (4).
In this study, 78 metagenomic datasets of hospital wastewater from six Chinese provincial-level administrative divisions (PLADs) were either obtained from the National Center for Biotechnology Information (NCBI) database or generated through direct collection and sequencing. Employing standardized methods, the research set out to analyze data from diverse regions within China to explore regional variations in antibiotic resistance. The investigation revealed that numerous clinically ARGs are carried on plasmids. Notably, the study identified regional differences in the potential bacterial hosts of ARGs at the genus level. Additionally, it was found that the existing hospital wastewater treatment systems are insufficient in effectively removing antibiotic resistance, further emphasizing the substantial ecological and health hazards associated with hospital wastewater discharge.
Metagenomic datasets representing hospital wastewater in China, spanning the years 2018 to 2022, were retrieved from the NCBI database. Additionally, we collected both influent and disinfected effluent samples from a general hospital in Shenzhen City, Guangdong Province, China. The samples underwent metagenomic sequencing on the Illumina NovaSeq 6000 platform, facilitated by Novogene (Tianjin, China), utilizing a paired-end 150 bp sequencing approach. Comprehensive details of the public datasets are presented in
Supplementary Table S1 . To quantify the abundance of ARGs in each sample, we employed the standardized ARGs-OAP pipeline in conjunction with the SARG database (Version 3.2,https://smile.hku.hk/ARGs/Indexing ), conforming to the suggested parameters and thresholds (6-7).Quality assessment of the raw paired-end reads was conducted using Fastp (Version 0.22.0,
https://github.com/OpenGene/fastp ), ensuring that only reads of quality score above 20 were retained for subsequent analysis. These high-quality reads from each sample were individually assembled into contigs with MegaHit (Version 1.2.9,https://github.com/voutcn/megahit ). Predictions of open reading frames (ORFs) were executed using Prodigal (Version 2.6.3,http://prodigal.ornl.gov/ ). We utilized Platon in accuracy mode to differentiate putative plasmid-derived contigs from chromosomal contigs, ensuring meticulous discrimination. Subsequent alignment of ORFs on these assembled sequences against the SARG database was performed (8). For contigs presumed to be of chromosomal origin as determined by Platon, taxonomic classification was executed using Kaiju (Version 1.9.2,https://bioinformatics-centre.github.io/kaiju/ ). The relative abundance of ARG-carrying contigs (ACCs) was calculated in accordance with methodologies previously described in the literature (9).This study identified a total of 1,899 ARG subtypes across 28 different ARG types within the wastewater samples analyzed (Figure 1A). Principal Coordinate Analysis (PCoA) based on the prevalence of ARG subtypes demonstrated distinct clustering of samples from Hebei, Guangdong, and Hunan provinces (Figure 1B). Conversely, in Sichuan Province, samples from the same hospital exhibited more similar ARG profiles. A significant positive correlation was observed between the aggregate abundance of ARGs and the presence of ARG-harboring plasmids (Figure 2B). A diversity of ARGs localized on plasmids was observed, with several subtypes such as aadA, qacEΔ1, cmlA5, sul1, catB3, ereA, OXA-10, and OXA-1 frequently co-occurring. Notably, plasmids routinely carried numerous clinically significant ARGs, including CTX-M-3, KPC-2, NDM-6, and the vanM gene cluster (Figure 2C). At the phylum level, Pseudomonadota was consistently the most prevalent ARG host across all PLADs (Figure 3A). However, at the genus level, variations in ARG genomic landscape were identified across different regions (Figure 3B). Pathogenic genera such as Escherichia, Acinetobacter, Aeromonas, Pseudomonas, Klebsiella, and Enterococcus were associated with a higher diversity of ARGs when compared to non-pathogenic genera.
Figure 1.The occurrence of ARGs in untreated hospital wastewater. (A) The abundance of ARGs in untreated hospital wastewater sampling from different provinces. (B) PCoA analysis based on Bray-Curtis distance of ARG subtypes’ abundance.
Note: Samples including Sichuan_1, Sichuan_2, and Sichuan_3 were taken from a general hospital while Sichuan_4, Sichuan_5, Sichuan_6 were taken from a hospital of traditional Chinese medicines, and Sichuan_7, Sichuan_8, Sichuan_9 were taken from a hospital of stomatology. Samples including Zhejiang_1, Zhejiang_2 and Zhejiang_3 were taken from an eye hospital and Zhejiang_4, Zhejiang_5, Zhejiang_6 were taken from a general hospital.
Abbreviation: MLS=macrolide-lincosamide-streptogramin.
Figure 2.The profile of ARGs located on plasmids. (A) The abundance of ARG-carrying plasmids in different provinces. (B) The Pearson correlation between the total abundance of ARGs and the abundance of ARG-carrying plasmids. (C) The ARG-carrying pattern on the putative plasmid contigs.
Note: The ARGs with more than 10 connections to the plasmid contigs were labeled in the figure.
Abbreviation: MLS=macrolide-lincosamide-streptogramin; ATGs=antibiotic resistance genes.
* P<0.05;
** P<0.01;
*** P<0.001.
-
In this study, the prevalence of antibiotic resistance in hospital wastewater across a national scope was evaluated using standardized methodologies that employed both read-based and assembly-based metagenomic analysis. This examination uncovered distinctions in the profiles of ARGs, with geographical location and hospital type significantly impacting the distribution of these ARGs (
Supplementary Figure S1 ). Although there was variance in the overall abundance of ARGs across different regions, the predominant categories of ARGs, including those conveying resistance to aminoglycosides, beta-lactams, macrolide-lincosamide-streptogramin (MLS), sulfonamides, and tetracyclines, remained consistent across PLADs (10). It is likely that the observed geographical influence can be attributed to regional differences in antibiotic prescription trends. Similarly, the type of hospital appeared to directly affect the variation in antibiotic use (3).Mobile genetic elements (MGEs) are well-documented vehicles for the transference of ARGs, facilitating their propagation (11). This study hones in on plasmids, a critical type of MGE capable of transferring ARGs across diverse phylogenetic boundaries. Our analysis identified a significant positive correlation between the prevalence of ARGs and the presence of plasmids harboring these genes, underscoring plasmids’ role in the mobilization of ARGs. The distribution of ARGs on plasmids was particularly telling, as numerous ARGs with clinical significance were found to be plasmid-borne, highlighting the considerable risk of their transference among bacterial species. Corroborating previous research, our study also detected the vanM gene cluster (4), which imparts resistance to vancomycin, on plasmids. Initially discovered in an Enterococcus faecium isolate from Shanghai Municipality, China, in 2006, vanM has been prevalent in the region since then (12-13). The detection of vanM in hospital wastewater on a national scale indicates widespread dissemination and points to plasmids as a likely contributor to this trend. Our findings demonstrate the pervasive presence of clinically significant ARGs on plasmids, suggesting that plasmids play a pivotal role in the dissemination of ARGs.
Upon further investigation into the ARG genetic context, we discovered that Pseudomonadota consistently emerged as the principal ARG carrier at the phylum level, regardless of the geographic PLADs. At the genus level, however, we observed pronounced regional variations in the ARG genetic context. Notably, the genus thauera exhibited a significantly higher ARG carrier abundance in Hebei Province compared to other PLADs, whereas the genus Acinetobacter was more abundantly identified as an ARG carrier in Gansu Province (P<0.05). Additionally, our findings indicated that pathogenic genera harbored a greater diversity of ARGs compared to their non-pathogenic counterparts (Figure 3C). Complementing these observations, investigation of metagenomic draft genomes from hospital wastewater environments revealed that approximately 60% of the identified potential pathogens (14). Among the E.coli resistant to erythromycin, ampicillin, or trimethoprim/sulfamethoxazole isolated from the same fecal specimen of children, pathogenic isolates were found to be approximately as twice as non-pathogenic isolates (15). This result might be related to the clinical usage of antibiotics against pathogens, as previously confirmed that hospital-specific antibiotic usage was generally associated with increased hospital antibiotic susceptibility (16).
Figure 3.The genetic context of ARGs. (A) The genetic context of ARGs at phylum level. (B) The genetic context of ARGs at genus level. (C) The ARG-carrying patterns of different genera.
Abbreviation: MLS=macrolide-lincosamide-streptogramin; ATGs=antibiotic resistance genes.Wastewater treatment processes are not always effective at eradicating ARGs. Notably, while ARGs were found to be significantly reduced in the wastewater from a general hospital in Zhejiang Province, the total abundance of ARGs in the treated wastewater of eye hospitals actually increased, rather than decreased (
Supplementary Figure S2 ). This suggests that the treatment methods implemented failed to effectively eradicate bacteria harboring multiple ARGs. Prior research has demonstrated variable responses of ARGs to treatment processes, with certain genes such as ermB and tetW showing reductions while others, including blaTEM, sul1, and qnrS, have been found to increase post-treatment (2,17). Zhu et al. conducted an investigation into antibiotic resistance in three hospital wastewater treatment systems and reported that chlorination, typically the final treatment stage before wastewater discharge, was largely ineffective at significantly reducing most ARGs (14). However, compared to relative abundance, absolute quantification of ARGs is more significant for evaluating the environmental implications of ARGs and the effectiveness of treatment technologies. Further investigation based on the absolute abundance of ARGs is desired to explore the treatment efficiency for reducing antibiotic resistance. Furthermore, analysis of the genetic context of ARGs revealed an increased relative abundance of pathogenic genera in treated water (Supplementary Figure S2B ). Pathogenic antibiotic-resistance bacteria (ARB) such as Aeromonas, Pseudomonas, Salmonella, and Klebsiella exhibited higher prevalence in treated wastewaters (Supplementary Figure S2B ). There is mounting evidence suggesting that wastewater treatment plants may actually foster an environment that supports the enrichment and proliferation of ARB and that these facilities are not adequately removing multidrug-resistant pathogens (14,18). Collectively, these findings suggest that treated wastewater could have an elevated presence of pathogenic genera that carry multiple ARGs, posing a potential risk to public health.This investigation provides valuable insights into antibiotic resistance within hospital wastewater across China, illuminating the considerable impact of geographical location and hospital categorization on the distribution of ARGs. In addition, this study reveals a preference for certain clinically significant ARGs to be located on plasmids, as well as regional variations in the genetic context of ARGs at the genus level. The findings also demonstrate that existing wastewater treatment facilities in hospitals are inadequate in effectively eliminating antibiotic resistance. This research highlights the urgent need for enhanced management and treatment of hospital wastewater to reduce the potential spread of antibiotic resistance emanating from healthcare facilities.
HTML
| Citation: |
|