Silver Nanoparticle Targets Fabricated Using Chemical Vapor Deposition Method for Differentiation of Bacteria Based on Lipidomic Profiles in Laser Desorption/Ionization Mass Spectrometry

Abstract

The global threat of numerous infectious diseases creates a great need to develop new diagnostic methods to facilitate the appropriate prescription of antimicrobial therapy. More recently, the possibility of using bacterial lipidome analysis via laser desorption/ionization mass spectrometry (LDI-MS) as useful diagnostic tool for microbial identification and rapid drug susceptibility has received particular attention because lipids are present in large quantities and can be easily extracted similar to ribosomal proteins. Therefore, the main goal of the study was to evaluate the efficacy of two different LDI techniques-matrix-assisted (MALDI) and surface-assisted (SALDI) approaches-in the classification of the closely related Escherichia coli strains under cefotaxime addition. Bacterial lipids profiles obtained by using the MALDI technique with different matrices as well as silver nanoparticle (AgNP) targets fabricated using the chemical vapor deposition method (CVD) of different AgNP sizes were analyzed by the means of different multivariate statistical methods such as principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), sparse partial least squares discriminant analysis (sPLS-DA), and orthogonal projections to latent structures discriminant analysis (OPLS-DA). The analysis showed that the MALDI classification of strains was hampered by interference from matrix-derived ions. In contrast, the lipid profiles generated by the SALDI technique had lower background noise and more signals associated with the sample, allowing E. coli to be successfully classified into cefotaxime-resistant and cefotaxime-sensitive strains, regardless of the size of the AgNPs. AgNP substrates obtained using the CVD method were used for the first time for distinguishing closely related bacterial strains based on their lipidomic profiles and demonstrate high potential as a future diagnostic tool for the detection of antibiotic susceptibility.

Keywords: bacteria; chemical vapor deposition; lipids; mass spectrometry; silver nanoparticles.

Figure 1

SEM images (A) of the bare plates as well as silver coatings obtaining using the CVD process with different amounts of the precursor (5, 10, and 15 mg). Registered XRD patterns for the studied plates (B) confirmed the deposition of the silver nanoparticles on the surface of the steel substrates. Ag 5, 10, 15—silver coating obtained using 5, 10, and 15 mg of the precursor; a.u.—arbitrary units.

Figure 2

Graphical representation of statistical analysis of MS data from MALDI experiment: PCA component 1 vs. 2 (A), PLS-DA component 1 vs. 2 (B), sPLS-DA component 1 vs. 2 (C), and OPLS-DA (D). Dark-grey area represents data for DFI-4 strain while light-grey represents DFI-30 strain.

Figure 3

Dendrograms for MS data from SALDI/NALDI (A), MALDI (B) experiments and with the use of two groups (C) and four groups (D) for all data.

Figure 4

Graphical representation of statistical analysis of MS data from MALDI and SALDI/NALDI experiment with two groups: PCA component 1 vs. 2 (A), PLS-DA component 1 vs. 2 (B), sPLS-DA component 1 vs. 2 (C), and OPLS-DA (D). Dark-grey area represents data for DFI-4 strain while light-grey represents DFI-30 strain.

Figure 5

Graphical representation of statistical analysis of MS data from MALDI and SALDI/NALDI experiment with four groups: PCA component 1 vs. 2 (A), PLS-DA component 1 vs. 2 (B), and sPLS-DA component 1 vs. 2 (C).

Figure 6

Graphical representation of statistical analysis of MS data from SALDI/NALDI experiment: PCA component 1 vs. 2 (A), PLS-DA component 1 vs. 2 (B), sPLS-DA component 1 vs. 2 (C), and OPLS-DA (D). Dark-grey area represents data for DFI-4 strain while light-grey represents DFI-30 strain.