A liquid chromatography-mass spectrometry platform for the analysis of phyllobilins, the major degradation products of chlorophyll in Arabidopsis thaliana

During senescence, chlorophyll is broken down to a set of structurally similar, but distinct linear tetrapyrrolic compounds termed phyllobilins. Structure identification of phyllobilins from over a dozen plant species revealed that modifications at different peripheral positions may cause complex phyllobilin composition in a given species. For example, in Arabidopsis thaliana wild-type, eight different phyllobilins have structurally been characterized to date. Accurate phyllobilin identification and quantification, which classically have been performed by high performance liquid chromatography (HPLC) and UV/vis detection, are, however, hampered because of their similar physiochemical properties and vastly differing abundances in plant extracts. Here we established a rapid method for phyllobilin identification and quantification that couples ultra-HPLC with high-resolution/high-precision tandem mass spectrometry. Using Arabidopsis wild-type and mutant lines that are deficient in specific phyllobilin-modifying reactions, we identified a total of 16 phyllobilins, among them two that have not been described before in Arabidopsis. The single and collision-induced dissociation tandem mass spectrometry data of all 16 Arabidopsis phyllobilins were collected in a mass spectrometry library, which is available to the scientific community. The library allows rapid detection and quantification of phyllobilins within and across Arabidopsis genotypes and we demonstrate its potential use for high-throughput approaches and genome-wide association studies in chlorophyll breakdown. By extending the library with phyllobilin data from other plant species in the future, we aim providing a tool for chlorophyll metabolite analysis as a measure of senescence for practical applications, such as post-harvest quality control. Significance Statement Phyllobilins have similar physiochemical properties and vastly differing abundance in plant extracts, thus accurate identification and quantification is challenging. Here we describe a LC-MS/MS platform that allows their simultaneous detection and quantification within extracts of senescent Arabidopsis thaliana leaves. This MS/MS spectral data can be referenced when analyzing Arabidopsis senescence mutants and ecotypes and should prove useful for comparative analyses of chlorophyll breakdown in other species.