Single‐filament imaging mass spectrometry lipidomics in Arthrospira platensis

Abstract

RationaleElucidating intra‐organismal biochemical and lipid organization in photosynthetic biological cell factories of filamentous cyanobacteria, such as Arthrospira platensis (Spirulina), is important for tracking physiological response mechanisms during growth. Little is known about the filaments' biochemical organization and cellular structure and no label‐free imaging techniques exist that provide molecular mapping.MethodsWe applied ultrahigh‐resolution mass spectrometry (MS) with matrix‐assisted laser desorption ionization (MALDI) imaging to immobilized Spirulina filaments to investigate the localization of lipids across distinct physiological regions. We optimized matrix selection and deposition methods with the goal of facilitating high spatial, and intra‐filament, resolution using untargeted multivariate statistical spectral deconvolution across MS pixels.ResultsOur results demonstrate an improved two‐step matrix application with an optimized procedure for intra‐organismal lipid profiling to improve analyte sensitivity and achieve higher spatial resolution. We evaluate several conventional matrices, namely 2,5‐dihydroxybenzoic acid (DHB), superDHB (sDHB), 1,5‐diaminonaphthalene (DAN), and a 50:50 mix of DHB and sDHB, and compare delineation and pixel‐based elucidation of intra‐filament lipidomics. We identified a total of 1626 features that could be putatively assigned a lipid‐like formula based on database query and 46 unique features, with associated lipid assignments that were significantly distinct in their intra‐filament location.ConclusionsMALDI imaging MS with untargeted statistical spectral deconvolution was used to visualize intra‐filament lipidomics organization in Spirulina filaments. Improvements in matrix deposition, including sequential sublimation and pneumatic spraying, increased signal abundance at high spatial resolution and allowed for identification of distinct lipid composition regions. This work outlines a methodology that may be used for micro‐ecological untargeted molecular phenotyping.