The ICU environment contributes to the endemicity of the “Serratia marcescens complex” in the hospital setting

Serratia marcescens is an opportunistic pathogen historically associated with sudden outbreaks in intensive care units (ICUs) and the spread of carbapenem-resistant genes. However, the ecology of S. marcescens populations in the hospital ecosystem remains largely unknown. We combined epidemiological information of 1,432 Serratia spp. isolates collected from sinks of a large ICU that underwent demographic and operational changes (2019-2021) and 99 non-redundant outbreak/non-outbreak isolates from the same hospital (2003-2019) with 165 genomic data. These genomes were grouped into clades (1-4) and subclades (A and B) associated with distinct species: Serratia nematodiphila (1A), S. marcescens (1B), Serratia bockelmannii (2A), Serratia ureilytica (2B), S. marcescens/Serratia nevei (3), and S. nevei (4A and 4B). They may be classified into an S. marcescens complex (SMC) due to the similarity between/within subclades (average nucleotide identity >95%-98%), with clades 3 and 4 predominating in our study and publicly available databases. Chromosomal AmpC β-lactamase with unusual basal-like expression and prodigiosin-lacking species contrasted classical features of Serratia. We found persistent and coexisting clones in sinks of subclades 4A (ST92 and ST490) and 4B (ST424), clonally related to outbreak isolates carrying bla_VIM-1 or bla_OXA-48 on prevalent IncL/pB77-CPsm plasmids from our hospital since 2017. The distribution of SMC populations in ICU sinks and patients reflects how Serratia species acquire, maintain, and enable plasmid evolution in both "source" (permanent, sinks) and "sink" (transient, patients) hospital patches. The results contribute to understanding how water sinks serve as reservoirs of Enterobacterales clones and plasmids that enable the persistence of carbapenemase genes in healthcare settings, potentially leading to outbreaks and/or hospital-acquired infections.IMPORTANCEThe "hospital environment," including sinks and surfaces, is increasingly recognized as a reservoir for bacterial species, clones, and plasmids of high epidemiological concern. Available studies on Serratia epidemiology have focused mainly on outbreaks of multidrug-resistant species, overlooking local longitudinal analyses necessary for understanding the dynamics of opportunistic pathogens and antibiotic-resistant genes within the hospital setting. This long-term genomic comparative analysis of Serratia isolated from the ICU environment with isolates causing nosocomial infections and/or outbreaks within the same hospital revealed the coexistence and persistence of Serratia populations in water reservoirs. Moreover, predominant sink strains may acquire highly conserved and widely distributed plasmids carrying carbapenemase genes, such as the prevalent IncL-pB77-CPsm (pOXA48), persisting in ICU sinks for years. The work highlights the relevance of ICU environmental reservoirs in the endemicity of certain opportunistic pathogens and resistance mechanisms mainly confined to hospitals.