Microorganisms (Apr 2021)

Classification of <i>Salmonella enterica</i> of the (Para-)Typhoid Fever Group by Fourier-Transform Infrared (FTIR) Spectroscopy

  • Miriam Cordovana,
  • Norman Mauder,
  • Markus Kostrzewa,
  • Andreas Wille,
  • Sandra Rojak,
  • Ralf Matthias Hagen,
  • Simone Ambretti,
  • Stefano Pongolini,
  • Laura Soliani,
  • Ulrik S. Justesen,
  • Hanne M. Holt,
  • Olivier Join-Lambert,
  • Simon Le Hello,
  • Michel Auzou,
  • Alida C. Veloo,
  • Jürgen May,
  • Hagen Frickmann,
  • Denise Dekker

DOI
https://doi.org/10.3390/microorganisms9040853
Journal volume & issue
Vol. 9, no. 4
p. 853

Abstract

Read online

Typhoidal and para-typhoidal Salmonella are major causes of bacteraemia in resource-limited countries. Diagnostic alternatives to laborious and resource-demanding serotyping are essential. Fourier transform infrared spectroscopy (FTIRS) is a rapidly developing and simple bacterial typing technology. In this study, we assessed the discriminatory power of the FTIRS-based IR Biotyper (Bruker Daltonik GmbH, Bremen, Germany), for the rapid and reliable identification of biochemically confirmed typhoid and paratyphoid fever-associated Salmonella isolates. In total, 359 isolates, comprising 30 S. Typhi, 23 S. Paratyphi A, 23 S. Paratyphi B, and 7 S. Paratyphi C, respectively and other phylogenetically closely related Salmonella serovars belonging to the serogroups O:2, O:4, O:7 and O:9 were tested. The strains were derived from clinical, environmental and food samples collected at different European sites. Applying artificial neural networks, specific automated classifiers were built to discriminate typhoidal serovars from non-typhoidal serovars within each of the four serogroups. The accuracy of the classifiers was 99.9%, 87.0%, 99.5% and 99.0% for Salmonella Typhi, Salmonella Paratyphi A, B and Salmonella Paratyphi C, respectively. The IR Biotyper is a promising tool for fast and reliable detection of typhoidal Salmonella. Hence, IR biotyping may serve as a suitable alternative to conventional approaches for surveillance and diagnostic purposes.

Keywords