Tandem MS-Proteotyping: Proteomics- and Genomics-based Characterization and Typing of Infectious Bacteria Edward Moore (Presenter) University of Gothenburg Presenter Bio: Edward Moore earned a PhD in Biology & Biochemistry (Microbiology) from the University of Houston, USA. He worked as Post-doctoral Researcher and Senior Researcher at the National Research Centre (GBF) in Braunschweig, Germany, and, later, as Senior Research Officer at the Macaulay Research Centre and Aberdeen University in Scotland, studying microbial ecology. In 2004, he changed research directions, moving to establish a research group to work on developments in diagnostics of infectious diseases at the Sahlgrenska University Hospital and the Department of Infectious Diseases, University of Gothenburg in Sweden. He is Director and Curator of the Culture Collection University of Gothenburg (CCUG), one of the largest public collections for clinically-relevant microorganisms in the world. In 2005, he was appointed Professor of Bacteriology at the Sahlgrenska Academy, University of Gothenburg.

Authors: Edward R.B. Moore (1,2,3), Lucia Gonzales-Siles (1,2) Francisco Salvà-Serra (1,2,4), Hedvig Engström Jakobsson (1,2), Daniel Jaén-Luchoro (1,3), Beatriz Piñeiro (3), Marga Gomila (4), Roger Karlsson (1,3,5)
(1) Department of Clinical Microbiology, Sahlgrenska University Hospital, Gothenburg, Sweden; (2) Department of Infectious Diseases, Institute for Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; (3) Culture Collection University of Gothenburg (CCUG), Gothenburg, Sweden; (4) Microbiology, Department of Biology, University of the Balearic Islands, Palma de Mallorca, Spain; (5) Nanoxis Consulting AB, Gothenburg, Sweden.

LC-MS/MS proteomics- and genomics-based characterization and typing of microorganisms, i.e., ‘proteotyping’, using peptides digested from expressed proteins and matched to genomic sequence data, can be applied for sensitive and accurate detection and typing of pathogenic bacteria. Proteotyping is capable of resolving and identifying closely-related bacterial taxa with simultaneous detection of virulence and antibiotic resistance features, providing for comprehensive characterizations of infectious bacteria. The methodology may be applied directly to analyses of clinical samples without prior cultivation and isolation, thus providing for rapid, reliable, infectious disease diagnostics.

Introduction

The global expansion of antibiotic resistance (ABR) in bacteria, including human pathogens, presents unconventional challenges for the treatment and prevention of contagious diseases. The World Health Organization (WHO) has predicted the onset of infections for which no effective treatment will be available [1]. With such anticipated escalation of ABR, combined with continuing decline in new antibiotic discovery, the development of innovative approaches for reliable, rapid and cost-efficient characterization of pathogenic microorganisms and diagnostics of infectious disease is increasingly essential to confront rising mortality and costs associated with infections. The routine methodologies used today for diagnosing infections typically depend upon protocols requiring cultivation of the pathogenic bacteria from clinical samples. Faced with patients exhibiting symptoms of infection, physicians resort to prescribing broad-spectrum antibiotics while waiting days or weeks for results from laboratories. With whole-genome sequence data increasingly available, MS-based proteomics are progressively being applied in biological inquiries, providing alternative means for efficient clinical analyses, relying on detection and identification of expressed features, rather than genotypic features that may not be phenotypically relevant. Proteomic analyses of bacterial cells may be considered to be indirect analyses of bacterial genomes; the proteome comprises the entire set of proteins expressed by a cell, an organism or a biological system. ‘Proteotyping’ [2, 3], using LC-MS/MS to detect unique peptide biomarkers from cellular proteins, enables identification of pathogens, differentiating even the most closely related bacterial species, as well as simultaneous detection of ABR and virulence factors, from single MS analytical runs.

Methods

Bacterial reference strains were obtained from the Culture Collection University of Gothenburg (CCUG; https://ccug.se), for optimizing protocols and defining biomarker peptides, and upper respiratory tract samples from the Bacteriology Laboratory of the Department of Clinical Microbiology, Sahlgrenska University Hospital, were used for assessing efficacies of proteotyping analyses of clinical samples. The Molysis Kit (Molzym GmbH & Co. KG) was used for clinical sample clean-up, removing human biomass, while keeping bacteria intact. Bacteria recovered from the clinical samples were treated by bead beating, followed by trypsin digestion, to generate peptides. Triplicate MS analyses were done, using reversed phase nano-LC and Q-Exactive MS (ThermoFisher Scientific, Inc.). MS spectra were acquired and the 'TCUP (Typing and Characterization using Proteomics)' bioinformatics pipeline [4] was used for identifying peptides by matching to genome sequence data in a curated reference database. Genome sequences of relevant reference strains of respective species were determined, to build and reinforce the database.

Results

Hundreds or thousands of peptides could be defined from reference strains as bacterial species-unique biomarkers for identifying pathogenic and non-pathogenic bacteria and for detection of virulence and ABR factors. Clinical respiratory tract samples were analyzed directly (i.e., no cultivation). Respiratory tract pathogens, such as Streptococcus pneumoniae, Haemophilus influenza, Moraxella catarrhalis and Staphylococcus aureus, were able to be detected, using an open, ‘shotgun’, analytical approach. Targeted MS analyses, using identified peptide biomarkers, were able to improve sensitivities of detection of bacteria in clinical samples by as much as three orders of magnitude. Known virulence factors of the different pathogens could be detected and identified. These data were correlated with results from an in-house qPCR panel as validation of the efficacy of the proteomics approach.

Conclusions & Discussion

Proteotyping analyses of generated cellular peptides, enables detection and identification of bacterial pathogens, discriminating closely related bacterial species, as well as subspecies-level differentiation, including detection of ABR and virulence factors, from single MS analytical runs. Accurate and comprehensive genome sequence data is a key for reliable peptide matching and identifying infectious species. The goal of applying the methodology directly to clinical samples without the need for prior cultivation, was demonstrated, by analyses of clinical samples, validating potential protocols, applying protein biomarker detection for reliable and rapid diagnostics. Furthermore, the elucidation of species-unique peptides provides the basis for designing new diagnostic kits, targeting peptide biomarker detection by, e.g., sensitive qPCR and antibody assays. Proteotyping, relying on MS-based detection of expressed proteins, provides an alternative, effective strategy to the traditional, as well as new genotypic- or, even, genomic-based protocols for clinical identification of infectious bacteria.

[1] World Health Organization. 2014. Antimicrobial resistance: global report on surveillance. WHO Press, Geneva, Switzerland.

[2] Karlsson R, Gonzales-Siles L, Boulund F, Svensson-Stadler L, Skovbjerg S, Karlsson A, Davidson M, Hulth S, Kristiansson E, Moore ERB, 2015. Proteotyping: Proteomic characterisation, classification and identification of microorganisms – a prospectus. Systematic & Applied Microbiology 38:246-257.

[3] Karlsson R, Gonzales-Siles L, Boulund F, Lindgren Å, Svensson-Stadler L, Karlsson A, Kristiansson E, Moore ERB. 2017. Proteotyping: tandem mass spectrometry shotgun proteomic characterization and typing of pathogenic microorganisms. In Shah HN, Gharbia SE (eds.). MALDI-TOF and Tandem MS for Clinical Microbiology. John Wiley & Sons, Ltd., New York City, USA.

[4] Boulund F, Karlsson R, Gonzales-Siles L, Johnning A, Karami N, Al-Bayati O, Åhrén C, Moore ERB, Kristiansson E. 2017. Typing and Characterization of Bacteria Using Bottom-up Tandem Mass Spectrometry Proteomics. Molecular & Cellular Proteomics 16:1052-1063.