Antibiotic resistance genes and human bacterial pathogens: Co-occurrence, removal, and enrichment in municipal sewage sludge digesters

Understanding which/how antibiotic resistance genes (ARGs) contribute to increased acquisition of resistance by pathogens in aquatic environments are challenges of profound significance. We explored the co-occurrence and removal versus enrichment of ARGs and human bacterial pathogens (HBPs) in municipal sewage sludge digesters. We combined metagenomic detection of a wide spectrum of 323 ARGs and 83 HBPs with a correlation-based statistical approach and charted a network of their co-occurrence relationships. The results indicate that most ARGs and a minor proportion of HBPs (mainly Collinsella aerofaciens, Streptococcus salivarius and Gordonia bronchialis) could not be removed by anaerobic digestion, revealing a biological risk of post-digestion sludge in disseminating antibiotic resistance and pathogenicity. Moreover, preferential co-occurrence patterns were evident within one ARG type (e.g., multidrug, beta-lactam, and aminoglycoside) and between two different ARG types (i.e., aminoglycoside and beta-lactam), possibly implicating co-effects of antibiotic selection pressure and co-resistance on shaping antibiotic resistome in sewage sludge. Unlike beta-lactam resistance genes, ARGs of multidrug and macrolide-lincosamide-streptogramin tended to co-occur more with HBPs. Strikingly, we presented evidence that the most straightforward biological origin of an ARG-species co-occurring event is a hosting relationship. Furthermore, a significant and robust HBP-species co-occurrence correlation provides a proper scenario for nominating HBP indicators (e.g., Bifidobacterium spp. are perfect indicators of C. aerofaciens; r = 0.92-0.99 and P-values < 0.01). Combined, this study demonstrates a creative and effective network-based metagenomic approach for exploring ARG hosts and HBP indicators and assessing ARGs acquisition by HBPs in human-impacted environments where ARGs and HBPs may co-thrive.