Evaluation of lipid pretreatment methods for microbial identification using linear negative-ion MALDI-TOF MS

Accurate microbial identification based on lipid profiling offers promising applications in diagnostics and microbial ecology. In this study, five lipid pretreatment methods-Maldixin, chloroform-methanol, tert-butyl methyl ether, isobutyric acid-ammonium hydroxide, and sodium acetate (SA) buffer (pH 4.0)-were systematically evaluated using MALDI-TOF mass spectrometry operated in linear negative-ion mode. Among the tested methods, the SA buffer method provided a favorable balance between operational feasibility, extraction efficiency, and spectral quality. Compared with positive-ion mode, negative-ion mode yielded high-intensity lipid peaks and superior spectral clarity, particularly for Acinetobacter baumannii. Lipid profiles were generated from 203 bacterial strains representing six species, revealing distinct species-specific lipid fingerprints. A lipid fingerprint database was subsequently constructed and validated using an independent set of 144 strains, achieving identification accuracies of 81.9% at the species-level and 97.2% at the genus-level. While overall species-level accuracy remains lower than that of conventional protein-based systems, these results indicate that linear negative-ion mode MALDI-TOF MS provides complementary lipid information that may enhance microbial characterization. Further optimization of sample preparation and database algorithms is warranted to support broader clinical application.