Concordance in molecular methods for detection of antimicrobial resistance: A cross sectional study of the influent to a wastewater plant

Kezia Drane; Roger Huerlimann; Rhondda Jones; Anna Whelan; Madoc Sheehan; Ellen Ariel; Robert Kinobe

doi:10.1016/j.mimet.2024.107069

Concordance in molecular methods for detection of antimicrobial resistance: A cross sectional study of the influent to a wastewater plant

J Microbiol Methods. 2024 Nov 17:228:107069. doi: 10.1016/j.mimet.2024.107069. Online ahead of print.

Authors

Kezia Drane¹, Roger Huerlimann², Rhondda Jones³, Anna Whelan⁴, Madoc Sheehan⁵, Ellen Ariel¹, Robert Kinobe⁶

Affiliations

¹ College of Public Health, Medical and Veterinary Sciences, 1 Solander Drive, James Cook University, Townsville, Queensland 4811, Australia.
² Marine Climate Change Unit, Okinawa Institute of Science and Technology, 1919-1 Tacha, Onna-son, Okinawa 904-0495, Japan; College of Science and Engineering, 1 James Cook Drive, James Cook University, Townsville, Queensland 4811, Australia.
³ Graduate Research School, 1 James Cook Drive, James Cook University, Townsville, Queensland 4811, Australia.
⁴ Townsville Water and Waste, Wastewater Operations, Townsville, Queensland 4810, Australia.
⁵ College of Science and Engineering, 1 James Cook Drive, James Cook University, Townsville, Queensland 4811, Australia.
⁶ College of Public Health, Medical and Veterinary Sciences, 1 Solander Drive, James Cook University, Townsville, Queensland 4811, Australia. Electronic address: Robert.kinobe@jcu.edu.au.

PMID: 39557129
DOI: 10.1016/j.mimet.2024.107069

Abstract

Methods that are used to characterise microbiomes and antimicrobial resistance genes (ARGs) in wastewater are not standardised. We used shotgun metagenomic sequencing (SM-Seq), RNA sequencing (RNA-seq) and targeted qPCR to compare microbial and ARG diversity in the influent to a municipal wastewater treatment plant in Australia. ARGs were annotated with CARD-RGI and MEGARes databases, and bacterial diversity was characterised by 16S rRNA gene sequencing and SM-Seq, with species annotation in SILVA/GreenGenes databases or Kraken2 and the NCBI nucleotide database respectively. CARD and MEGARes identified evenly distributed ARG profiles but MEGARes detected a richer array of ARGs (richness = 475 vs 320). Qualitatively, ARGs encoding for aminoglycoside, macrolide-lincosamide-streptogramin and multidrug resistance were the most abundant in all examined databases. RNA-seq detected only 32 % of ARGs identified by SM-Seq, but there was concordance in the qualitative identification of aminoglycoside, macrolide-lincosamide, phenicol, sulfonamide and multidrug resistance by SM-Seq and RNA-seq. qPCR confirmed the detection of some ARGs, including OXA, VEB and EREB, that were identified by SM-Seq and RNA-seq in the influent. For bacteria, SM-Seq or 16S rRNA gene sequencing were equally effective in population profiling at phyla or class level. However, SM-Seq identified a significantly higher species richness (richness = 15,000 vs 3750). These results demonstrate that SM-Seq with gene annotation in CARD and MEGARes are equally sufficient for surveillance of antimicrobial resistance in wastewater. For more precise ARG identification and quantification however, MEGARes presented a better resolution. The functionality of detected ARGs was not confirmed, but general agreement on the putative phenotypic resistance profile by antimicrobial class was observed between RNA-Seq and SM-Seq.

Keywords: Antibiotics; Antimicrobial resistance; Microbiome; Next-generation sequencing; RNA sequencing; Wastewater.