Optimizing main spectrum profiles for use in real-time MALDI-TOF mass spectrometry identification of Leptospira serovars

Nickolyn-Martin, Jennifer E.

Permalink

https://hdl.handle.net/2142/108114 Copy

Description

Title

Optimizing main spectrum profiles for use in real-time MALDI-TOF mass spectrometry identification of Leptospira serovars

Author(s)

Nickolyn-Martin, Jennifer E.

Issue Date

2020-04-21

Director of Research (if dissertation) or Advisor (if thesis)

Maddox, Carol W

Committee Member(s)

• Lau, Gee
• Allender, Matthew

Department of Study

Pathobiology

Discipline

VMS - Pathobiology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• MALDI-TOF MS
• leptospirosis
• Leptospira
• canine
• MSP
• serovar
• typing

Abstract

Leptospirosis, a zoonotic disease with a growing number of reported cases worldwide and expansion into new geographic areas, is an increasingly important public and veterinary health concern. Leptospirosis epidemiology is changing for both human and non-human animals. Over 250 pathogenic serovars of Leptospira, the causative agents of leptospirosis, have been identified. Given the proclivity for certain associations between these serovars and reservoir hosts, identification of clinical isolates to the serovar-level is key to understanding the epidemiology of this disease. However, an economical and rapid serovar typing method that can be incorporated into routine diagnostic testing is still lacking. The companion animal most often diagnosed with leptospirosis is the dog. Routine epi-surveillance of canine leptospirosis could benefit both canine and human health. Given the widespread use of Matrix-Assisted Laser Desorption Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS) by diagnostic laboratories for microorganism identification to the species-level, MALDI-TOF MS was explored as a potential method for serovar detection and identification in leptospirosis-positive canine urine samples. Commercial MALDI platforms currently do not include Leptospira serovar mass spectral profiles in their reference databases. Therefore, a custom MALDI Leptospira Main Spectrum Profile (MSP) database, representing seven selected serovars of Leptospira interrogans, was created for this project. An initial database consisted of MSPs constructed according to the manufacturer’s recommended guidelines. MSP specificity was tested using serial dilutions of serovar culture and culture spiked canine urine. Specificity of these original MSPs was found to be insufficient for accurate and consistent serovar identification. Various MALDI sample preparation and deposition methods were tested to determine if any produced higher quality sample spectra that would improve serovar MSP specificity. It was found that the standard ethanol/formic acid protein extraction sample preparation and dried droplet with matrix overlay deposition methods produced the best spectra for MSP creation. Several serovar raw spectra and peak analyses were performed using Bruker’s FlexControl, Compass Explorer, and ClinProTools software to identify potentially-characteristic serovar peak patterns and outlier raw spectra. The data obtained were used in creating several additional MSP types for each serovar, with each type using a different combination of creation parameters. It was found that MSPs designed using creation parameters that differed from those recommended by the manufacturer resulted in the greatest MSP specificity. The best-performing MSP for each serovar was chosen to create a custom Leptospira MSP database. Database specificity was tested in two blind-coded trials using the Bruker MALDI Biotyper Realtime Classification software. In the first trial, which used serovar culture samples for testing, 111 of 112 serovar sample spots returned a correct first match. Specificity ranged from 99 to 100%, while sensitivity ranged from 81 to 100%. In the second trial, which used serovar-spiked canine urine samples, 105 of 112 serovar sample spots returned a correct first match. Specificity for this trial ranged from 97 to 100%, while sensitivity ranged from 75 to 100%. This work demonstrated that, by optimizing custom MSP creation parameters, MALDI-TOF MS can be used to identify Leptospira isolates at the serovar-level within the real-time classification workflow. The second part of this project explored MALDI’s sensitivity for Leptospira serovar detection and identification. The goal was to learn whether MALDI-TOS MS can detect Leptospira at the concentrations typically seen in leptospirosis-positive canine urine specimens submitted to the University of Illinois at Urbana-Champaign’s Veterinary Diagnostic Laboratory (VDL). All qPCR-canine-leptospirosis-positive cases diagnosed by the VDL over a 2.5-year period were analyzed. The average cycle threshold (CT) value for these cases was 35.18. A standard curve was used to estimate a corresponding Leptospira concentration of 1.18 x 103 organisms/mL. Sensitivity trials were performed using serial two-fold dilutions of serovar cultures and serovar-spiked canine urine specimens tested against the custom Leptospira MSP database. The highest dilutions which returned accurate MALDI identifications to the serovar- and genus-levels were noted. Leptospira concentrations and CT values corresponding to these dilutions were determined. The lowest average concentrations that returned accurate MALDI identifications to the serovar- and genus-levels 3.55 x 108 organisms/mL and 3.33 x 108 organisms/mL, respectively. No significant difference was found between the lowest concentrations that returned accurate identifications for serovar culture versus culture-spiked canine specimen dilutions. The difference between the average concentration of leptospirosis-positive specimens received by the VDL and the lowest average concentrations identified by the MALDI revealed that the MALDI’s sensitivity for Leptospira is too low to be used as a routine leptospirosis diagnostic or epi-surveillance tool. Various concentration methods, including centrifugation and filtration, were tested to determine whether they could sufficiently concentrate Leptospira samples for MALDI direct detection. None proved successful. In summary, MALDI’s potential use for routine leptospirosis diagnostics and surveillance remains possible. Different matrices, development of selective enrichment cultures, novel sample preparation and concentration methods, modifications to MALDI software parameters, or a combination thereof, may yet offer a path forward. The work presented here proposes a technique by which custom Leptospira MSPs may be created for use in real-time serovar identification. To the author’s knowledge, this is the first study to report MALDI-TOF MS Leptospira serovar identification within the real-time, rather than offline, classification workflow. Though further testing of additional serovars and clinical isolates is needed, this technique may prove applicable to a variety of serovars other than those used in this study. If so, this technique would offer a method for diagnostic laboratories to create custom Leptospira MSPs for identification and surveillance of the predominant circulating serovars in their respective geographic regions.

Graduation Semester

2020-05

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/108114

Copyright and License Information

Copyright 2020 Jennifer Nickolyn-Martin

Owning Collections