Infrared MALDI Mass Spectrometry with Laser-Induced Postionization for Imaging of Bacterial Colonies

Ultraviolet matrix-assisted laser desorption ionization mass spectrometry imaging (UV-MALDI-MSI) is a powerful tool to visualize bacterial metabolites in microbial colonies and in biofilms. However, a challenge for the method is the efficient extraction of analytes from deeper within the bacterial colonies and from the cytoplasm of individual cells during the matrix coating step. Here, we used a pulsed infrared (IR) laser of 2.94 μm wavelength to disrupt and ablate bacterial cells without a prior coating with a MALDI matrix. Instead, tissue water or, in some experiments, in addition a small amount of glycerol was exploited for the deposition of the IR laser energy and for supporting the ionization of the analytes. Compared to water, glycerol exhibits a lower vapor pressure, which prolonged the available measurement time window within an MSI experiment. Mass spectra were acquired with a hybrid Synapt G2-S HDMS instrument at a pixel size of 120 μm. A frequency-quadrupled q-switched Nd:YAG laser with 266 nm wavelength served for laser-induced postionization (MALDI-2). In this way, the ion abundances of numerous small molecules such as nucleobases, 2-alkyl-quinolones, a prominent class of Pseudomonas aeruginosa signaling molecules involved in one of the three quorum-sensing pathways, and also the signals of various bacterial phospholipids were boosted, partially by orders of magnitude. We analyzed single and cocultured colonies of Gram-negative P. aeruginosa and of Gram-positive Bacillus subtilis and Staphylococcus aureus as exemplary bacterial systems. To enable a rapid (within 5 s) MSI-compatible steam inactivation in a custom-made autoclave filled with hot water steam, bacterial cultures were grown on porous polyamide membranes. Compared to a UV-MALDI-2-MS measurement of the same systems, mass spectra with a reduced low mass background were generally generated. This resulted in the unequivocal detection of numerous metabolites only with the IR laser. In a fundamental part of our study, and to optimize the IR-MALDI-2 approach for the highest analytical sensitivity, we characterized the expansion dynamics of the particle plume as generated by the IR laser. Here, we recorded the total ion count and the intensities of selected signals registered from P. aeruginosa samples as a function of the interlaser delay and buffer gas pressure in the ion source. The data revealed that the IR-MALDI-2 ion signals are primarily generated from slow particles having mean velocities of ∼10 m/s. Interestingly, two different pressure/delay time regimes of the optimized ionization efficiency for phospholipids and smaller metabolites, respectively, were revealed, a result pointing to yet-unknown convoluted reaction cascades. The described IR-MALDI-2 method could be a helpful new tool for a microbial mass spectrometry imaging of small molecules requiring little sample preparation.