Biosolids as a Source of Antibiotic Resistance Plasmids for Commensal and Pathogenic Bacteria

Abstract

Antibiotic resistance (AR) is a threat to modern medicine, and plasmids are driving the global spread of AR by horizontal gene transfer across microbiomes and environments. Determining the mobile resistome responsible for this spread of AR among environments is essential in our efforts to attenuate the current crisis. Biosolids are a wastewater treatment plant (WWTP) byproduct used globally as fertilizer in agriculture. Here, we investigated the mobile resistome of biosolids that are used as fertilizer. This was done by capturing resistance plasmids that can transfer to human pathogens and commensal bacteria. We used a higher-throughput version of the exogenous plasmid isolation approach by mixing several ESKAPE pathogens and a commensal Escherichia coli with biosolids and screening for newly acquired resistance to about 10 antibiotics in these strains. Six unique resistance plasmids transferred to Salmonella typhimurium, Klebsiella aerogenes, and E. coli. All the plasmids were self-transferable and carried 3-6 antibiotic resistance genes (ARG) conferring resistance to 2-4 antibiotic classes. These plasmids-borne resistance genes were further embedded in genetic elements promoting intracellular recombination (i.e., transposons or class 1 integrons). The plasmids belonged to the broad-host-range plasmid (BHR) groups IncP-1 or PromA. Several of them were persistent in their new hosts when grown in the absence of antibiotics, suggesting that the newly acquired drug resistance traits would be sustained over time. This study highlights the role of BHRs in the spread of ARG between environmental bacteria and human pathogens and commensals, where they may persist. The work further emphasizes biosolids as potential vehicles of highly mobile plasmid-borne antibiotic resistance genes.

Keywords: antibiotics; antimicrobial resistance; biosolids; pathogens; plasmids.

Figure 1

Spreading of plasmids within a mobile resistome and a modified plasmid capture method to isolate them. (A) Biosolids from wastewater treatment plants (WWTPs) used as agricultural soil fertilizer contain bacteria (green rectangles) with resistance plasmids (orange circles). These biosolids can spread the resistance plasmids further, and through direct or indirect routes transfer to human pathogen and commensal bacteria (red rectangles with black spikes). (B) To determine the mobile resistome of biosolids that can be acquired by human pathogens and commensal bacteria, we used a modified exogenous plasmid isolation approach. Biosolids were mixed individually with different ESKAPE pathogens or E. coli (the recipient hosts). After overnight incubation, the mixtures were spread on a large selective agar plate and stamped with about 10 antibiotic disks. Recipient hosts that acquired a new resistance plasmid from the biosolids formed colonies in the inhibition zone where the recipient alone was sensitive (“control recipient alone”). Assays with biosolids alone were also run in parallel to control for false positives (“controls biosolids alone”).

Figure 2

Genomic maps of the pALTS plasmids. Genes are color-coded by functional/regional categorization. All annotations are referenced in Supplementary Table S1. Replication, Mating Pair Formation, DNA Transfer, Maintenance and Control, Accessory, and Unknown.

Figure 3

Comparative analysis of the pALTS plasmids with close representatives of their respective plasmid incompatibility group. Alignments of plasmids to representative plasmids in their group show high similarities in backbones. Red alignments represent forward alignment whereas blue alignments represent inverted alignment, and the color intensity is proportional to % similarity. Identity cutoff for visual representation was 80%. The plasmids pMLUA1 and pMOL98 are shown as close relatives of pALTS27/pALTS32 and pALTS28, respectively, while R751 and pTL50 (Shintani et al., 2020) are shown as the respective plasmid archetypes of the IncP-1β and IncP-1ɛ groups. Replication, Mating Pair Formation, DNA Transfer, Maintenance and Control, Accessory, and Unknown.

Figure 4

Comparative analysis of the accessory regions of the pALTS plasmids showing key mobile genetic elements in blue, ARGs in red, and other CDSs in yellow. (A) Comparisons of accessory regions of pALTS27, pALST32, and pALTS29. The interrupted class 1 integron integrase gene is shown in green. (B) Comparison of pALTS28 accessory region with its best BLASTn hit (www.ncbi.com), a mobile genetic element of the plasmid pMOS94 (accession: MK671725.1). (C) Comparison of accessory regions of pALTS31 and pALTS33, including details of the duplicated IS6100 in pATS33 showing the inverted conformation of the duplication and the target site duplication (TSD); b’ is the inverted TSD of b. Terminal inverted repeat (TIR) are shown by purple arrows and TSD by a black arrow. Comparative analysis was visualized using Easyfig 2.2.2 (Sullivan et al., 2011) using a predicted protein similarity threshold above 70%. *In1810 carried two putative ARG cassettes dfr40 and ereA4b not identified by AMRfinder. Those two predicted proteins were 100% identical to a trimethoprim-resistant dihydrofolate reductase DfrA and an EreA family erythromycin (E) esterase in the non-redundant protein database from NCBI.

Figure 5

The plasmids were able to persist in their new host over 9 days without selection. Species are coded by shape, and plasmids by color. The proportion of plasmid-carrying cells reported is the median of six replicate cultures. On day 9, the data for E. coli (pALTS31) and S. typhimurium (pALTS33) only represent triplicates. Klebsiella aerogenes (pALTS28), E. coli (pALTS27), E. coli (pALTS31), E. coli (pALTS32), S. typhimurium (pALTS27), S. typhimurium (pALTS29), and S. typhimurium (pALTS33).