Author's School

School of Medicine

ORCID

https://orcid.org/0000-0003-3243-3874

Author's Department/Program

Biology and Biomedical Sciences

Language

English (en)

Date of Award

12-15-2026

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Guatam Dantas

Committee Members

Megan T. Baldridge, Andrew Kau, Jennie H. Kwon, Fangqiong Ling, Drew J. Schwartz

Abstract

Bacteria are diverse and adaptable and, consequently, present in a wide variety of environments. They are capable of growing under intense pressures, including those brought on by the human immune system and disinfection strategies. To survive these pressures, bacteria develop resistances. Antimicrobial resistance, which occurs when a microbe is no longer responsive to the antibiotic intended to control it, is one such adaptation that presents an especially serious threat, as it limits treatment options for infections caused by antimicrobial resistant microbes. Similarly, the ability of bacteria to survive in environments subjected to heavy disinfection efforts, such as intensive care units (ICUs) in hospitals, can allow these bacteria to colonize or even infect vulnerable hospitalized patients. In order to prevent and monitor healthcare-associated infections (HAIs) and hospital acquired infections, it is essential to understand the microbial reservoirs in ICUs where HAI risk is high. This thesis work genomically describes the bacteria present in two distinct ICU reservoirs: the sink drains of a medical ICU (MICU) in Chapter 2 and the gut microbiomes of neonatal ICU (NICU)-hospitalized infants in Chapter 3.

In Chapter 2, I report on a longitudinal sampling of sink drains from the Barnes Jewish Hospital MICU, investigating the microbial ecology over 47 weeks and evaluating the efficacy of enhanced cleaning on microbial clearance. As enhanced cleaning with peracid-peroxide foam has previously reduced microbial burden in ICU sink drains, I and my collages expected to observe a substantial reduction in bacterial growth from applying this cleaning. We selected four MICU sinks to receive only standard-of-care cleaning and eight MICU sinks to additionally receive peracid + peroxide cleaning. We performed enhanced cleaning on the eight selected sink drains in three periods (one four-week period and two eight-week periods) separated by eight-week follow-up periods with only standard-of-care cleaning. All 12 sink drains were sampled repeatedly immediately before cleaning and during follow-up periods, and these samples were cultured on multiple selective and non-selective agars. Unexpectedly, we found no reduction in bacterial burden during or following enhanced cleaning, which could reflect the different ICU activities in the MICU compared to past studies. We commonly found Pseudomonas aeruginosa and Stenotrophomonas maltophilia, both opportunistic human pathogens known to occupy plumbing niches. Highly related P. aeruginosa and S. maltophilia genomes were collected throughout the study, indicating the likely persistence and resilience of these organisms against cleaning efforts. Persistent sequence types (27 for P. aeruginosa and 5 for S. maltophilia) were associated with greater counts of unique antimicrobial resistance genes and stress-associated genes. The results presented here underscore the context-specific nature of environmental microbes and the necessity of consistent, site-specific monitoring.

In Chapter 3, I examine specific pathobionts (microbes which can cause disease or harmlessly colonize) from the gut microbiomes of NICU-hospitalized preterm infants. Specifically, I focused on taxa known to cause bloodstream infections in NICU patients: Enterococcus faecalis, Enterococcus faecium, Enterobacter spp., Escherichia coli, Klebsiella spp., Staphylococcus aureus, and Staphylococcus epidermidis cultured from stools collected during the first month of life from 110 infants. My co-authors investigated the gut microbiomes of 331 infants longitudinally for the first 60 days of life, and they found that gut microbiome maturation is hampered by antimicrobial exposures, which are common in NICU patients. Among the target pathobionts, we found multiple putative plasmids (circularized contigs shorter than 1 Mb) associated with bacterial chromosomes. We also identified 22 clusters of putative plasmids within 99.99% average nucleotide identity of each other and associated with different bacterial chromosomes. We found six such clusters in distinct infants who were hospitalized in the same NICU room, implicating the NICU built environment in possible transfer of microbes and microbial genetic material between infants; seven putative plasmid clusters were identified in multiple rooms of the same NICU. Interestingly, we detected ten of these putative plasmid clusters in more than one NICU, which we took to suggest that this similarity reflects the few microbes, and therefore plasmids, capable of survival in the intense, highly selective niche of the NICU environment. Functions encoded on clustered putative plasmids were potentially related to bacterial survival, such as antimicrobial resistance and metal homeostasis. Together, these results highlight interconnectedness between patients in the NICU, with implications for infection prevention and future disinfection work.

DOI

https://doi.org/10.48765/skxk-7t69

Download

Share

COinS