Comparison of Sample Preparation Methods and Evaluation of Intra- and Intersample Reproducibility in Bacteria MALDI-MS Profiling

Abstract

A study is presented on the reproducibility of mass spectral profiles of the whole bacterium E. coli resulting from laser sampling at different regions within and between matrix-assisted laser desorption ionization (MALDI) samples deposited onto the plate. Samples were prepared with different deposition methods and using different MALDI matrices. The three most common matrices used in MALDI-mass spectrometry (MS) bacteria profiling, α-cyano-4-hydroxycinnamic acid (CHCA), sinapinic acid (SA), and ferulic acid (FA), were compared in this study along with two pipet-based sample deposition methods, dried-droplet and premix. Sample variability was determined by analysis of variances (ANOVA), principal component analysis (PCA), and multivariate ANOVA (MANOVA). For the two pipet-based sample deposition methods tested in this study, the intrasample variability (most commonly referred to as "spot-to-spot" reproducibility) was of the same magnitude as the intersample variability for all MALDI matrices tested. By incorporating a spray nebulizer sample deposition method to produce uniform sample/matrix mixtures onto the MALDI plate, we demonstrate that the crystalline morphology of the MALDI sample greatly influences the intrasample reproducibility (i.e., spot-to-spot) of the resulting whole cell MALDI-MS profiles. Overall, for the pipet-based deposition methods, results showed that the smallest variability in bacteria MALDI mass spectral profiles was obtained from samples deposited using the premix method, regardless of the MALDI matrix used, with the best reproducibility obtained with the CHCA matrix. It is concluded that a sample preparation strategy that reduces or eliminates the MALDI matrix morphology heterogeneity can reduce variability (i.e., spot-to-spot) of the bacteria mass spectral profiles by up to 90%.