MSACL 2025 Abstract

Direct from Specimen Proteomic Profiling of E. coli from Urinary Tract Infections

Linda K Nartey (1), Kenneth Weke (1), Victor Yuen (2), Pamela Kibsey (2), Michael X Chen (2,3), David R Goodlett (1)
(1) University of Victoria, (2) Vancouver Island Health Authority, (3) University of British Columbia

Linda Nartey, MSc (Presenter) University of Victoria

Presenter Bio: Ms. Linda Nartey is a final-year Ph.D. candidate in Microbiology at the University of Victoria, where she works under the supervision of Professor David R. Goodlett and Dr. Michael Chen. Her research focuses on advancing rapid, culture-independent methods for microbial identification by directly analyzing microbial lipidomes and proteomes from clinical and environmental samples. Her work leverages mass spectrometry techniques to profile species-specific lipid markers, with the goal of improving diagnostic speed and accuracy for infectious diseases.

Linda’s scientific journey began with a B.Sc. in Botany from the University of Ghana, where she studied fungal and bacterial contaminants in food. She went on to earn M.Sc. in Biology from Zhejiang Sci-Tech University in China, researching fungal biocontrol agents. Before beginning her graduate studies, she gained hands-on experience at the Noguchi Memorial Institute for Medical Research, contributing to diagnostic research on neglected tropical diseases such as Buruli ulcer and Schistosomiasis.

Her broader research interests include microbial pathogenesis, antimicrobial resistance, and clinical microbiology. Linda aspires to bridge the gap between research and real-world diagnostics by developing faster, more reliable tools to identify pathogens directly from samples, ultimately improving patient care and public health outcomes.

Relevant Financial Disclosures (within past 24 months, reported on Jul 21, 2025) No relevant financial relationship(s) to disclose.

INTRODUCTION
Urinary tract infections (UTIs) remain a significant public health concern affecting approximately 150 million people annually with 20% leading to sepsis, 18% of which are fatal. Traditional diagnostic approaches rely on culture-based methods, which may take 2-3 days for pathogen identification by the Bruker protein biotyper technique. A faster workflow that maintains accuracy of protein-based methods while reducing costs and time to identification is essential. Here, we build on findings from our previous study (PMID: 39283074), where the fast lipid analysis technique (FLAT) was used to improve detection of UTIs in the general population by lipidomics without ex vivo culture. Focusing on this direct from specimen concept and knowing that Escherichia coli are among the most common UTI infections globally we chose to next define the proteomes of E coli taken from patients without ex vivo growth. The proteomic changes that occur when E coli transitions from a host environment to lab remain underexplored. Revealing these protein-level variations is crucial for understanding E. coli’s pathogenicity and enabling the development of effective therapeutic agents.

OBJECTIVES
Understand the differences in proteomes between uropathogenic E. coli strains taken from patients with UTIs without ex vivo growth versus the same patient E coli strains after one generation of growth in the lab.

METHODS
To investigate the proteomic changes of E.coli infected urine after culturing in the lab, we conducted an in-depth, label-free quantitative proteomics investigation of Escherichia coli isolated directly from the urine of five UTI-positive patients, comparing the proteome profiles of each patient’s E. coli to the same E. coli after one generation of laboratory growth. Here, E. coli was isolated directly from UTI-positive urine samples of five patients (termed ‘direct’). Each sample was cultured in laboratory conditions (termed ‘lab-grown’) for one growth generation (Gen1) before harvesting followed by analysis using highly sensitive mass spectrometer (MS).

RESULTS: On average we identified 716-1656 and 1497-1641 proteins from each of the ‘direct’ and Gen1 'lab-grown' samples, respectively. Notably, 23 proteins were consistently absent in all patient-derived E. coli proteomes after laboratory cultivation. These proteins missing from growth in vitro indicate potential adaptive changes of E coli during stressed growth in these patients that may represent biomarkers of infection and/or therapeutic targets.

CONCLUSIONS
Our findings highlight critical proteomic differences between in vivo and in vitro conditions, offering new insights into microbial adaptation and the implications for understanding bacterial behavior within the host environment. These lost proteins are hypothesized to play key roles in E. coli’s adaptation to the host urinary environment. By directly analyzing the proteome from patient specimens, this study provides unique insights into pathogen biology that are often lost in traditional culture-based studies. Our findings underscore the importance of studying pathogens in conditions that closely mimic their natural environment to understand their behavior and identify novel targets for therapeutic intervention.

Collectively, these results highlight the importance of direct from specimen analysis for both improved diagnostics, but also more realistic biomarker discovery.