Rapid and Robust Workflows Using Different Ionization, Computation, and Visualization Approaches for Spatial Metabolome Profiling of Microbial Natural Products in Pseudoalteromonas.

Ambient mass spectrometry (MS) technologies have been applied to spatial metabolomic profiling of various samples in an attempt to both increase analysis speed and reduce the length of sample preparation. Recent studies, however, have focused on improving the spatial resolution of ambient approaches. Finer resolution requires greater analysis times and commensurate computing power for more sophisticated data analysis algorithms and larger data sets. Higher resolution provides a more detailed molecular picture of the sample; however, for some applications, this is not required. A liquid microjunction surface sampling probe (LMJ-SSP) based MS platform combined with unsupervised multivariant analysis based hyperspectral visualization is demonstrated for the metabolomic analysis of marine bacteria from the genus Pseudoalteromonas to create a rapid and robust spatial profiling workflow for microbial natural product screening. In our study, metabolomic profiles of different Pseudoalteromonas species are quickly acquired without any sample preparation and distinguished by unsupervised multivariant analysis. Our robust platform is capable of automated direct sampling of microbes cultured on agar without clogging. Hyperspectral visualization-based rapid spatial profiling provides adequate spatial metabolite information on microbial samples through red-green-blue (RGB) color annotation. Both static and temporal metabolome differences can be visualized by straightforward color differences and differentiating m/z values identified afterward. Through this approach, novel analogues and their potential biosynthetic pathways are discovered by applying results from the spatial navigation to chromatography-based metabolome annotation. In this current research, LMJ-SSP is shown to be a robust and rapid spatial profiling method. Unsupervised multivariant analysis based hyperspectral visualization is proven straightforward for facile/rapid data interpretation. The combination of direct analysis and innovative data visualization forms a powerful tool to aid the identification/interpretation of interesting compounds from conventional metabolomics analysis.