A comprehensive high-resolution mass spectrometry approach for characterization of metabolites by combination of ambient ionization, chromatography and imaging methods.

RATIONALE: An ideal method for bioanalytical applications would deliver spatially resolved quantitative information in real time and without sample preparation. In reality these requirements can typically not be met by a single analytical technique. Therefore, we combine different mass spectrometry approaches: chromatographic separation, ambient ionization and imaging techniques, in order to obtain comprehensive information about metabolites in complex biological samples. METHODS: Samples were analyzed by laser desorption followed by electrospray ionization (LD-ESI) as an ambient ionization technique, by matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging for spatial distribution analysis and by high-performance liquid chromatography/electrospray ionization mass spectrometry (HPLC/ESI-MS) for quantitation and validation of compound identification. All MS data were acquired with high mass resolution and accurate mass (using orbital trapping and ion cyclotron resonance mass spectrometers). Grape berries were analyzed and evaluated in detail, whereas wheat seeds and mouse brain tissue were analyzed in proof-of-concept experiments. RESULTS: In situ measurements by LD-ESI without any sample preparation allowed for fast screening of plant metabolites on the grape surface. MALDI imaging of grape cross sections at 20 µm pixel size revealed the detailed distribution of metabolites which were in accordance with their biological function. HPLC/ESI-MS was used to quantify 13 anthocyanin species as well as to separate and identify isomeric compounds. A total of 41 metabolites (amino acids, carbohydrates, anthocyanins) were identified with all three approaches. Mass accuracy for all MS measurements was better than 2 ppm (root mean square error). CONCLUSIONS: The combined approach provides fast screening capabilities, spatial distribution information and the possibility to quantify metabolites. Accurate mass measurements proved to be critical in order to reliably combine data from different MS techniques. Initial results on the mycotoxin deoxynivalenol (DON) in wheat seed and phospholipids in mouse brain as a model for mammalian tissue indicate a broad applicability of the presented workflow.

Petroleum crude oil analysis using low-temperature plasma mass spectrometry.

RATIONALE: The analysis of crude oil is a challenging task due to sample complexity. In mass spectrometry, several ionization techniques can be used to perform this task. Herein, we report the use of an atmospheric pressure low-temperature plasma (LTP) probe to desorb and ionize compounds of petroleum crude oil from different sources and residual fuel oil standard reference materials (SRMs). LTP is used to perform rapid screening of low molecular weight and relatively volatile components enabling characterization and differentiation of crude oil samples relying solely on mass spectrometric data. METHODS: Crude oil samples were analyzed without sample preparation or dilution directly from sampling surfaces of different materials such as polytetrafluorethylene, glass and polyethylene. Analyses were performed using Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS) with high mass accuracy and high resolving power of 400,000 at m/z 400 to estimate the elemental composition of the ions produced by LTP. Principal components analysis (PCA) was performed on the LTP data for statistical analysis. RESULTS: LTP was found to generate positive ions of lower mass compounds of low to moderate polarity. Three-dimensional PCA plots efficiently differentiated between SRMs and Azerbaijan crude oil samples. Standards of alkanes, nitrogen heterocycles, sulfur heterocycles, hydrocarbon polycyclic aromatics and saturated acids were investigated for their behavior in LTP ionization. Alkanes were found to form oxidized products to some extent. The LTP probe worked particularly well in the characterization of sulfur compounds. CONCLUSIONS: LTP ionization of crude oils was found to advantageously complement analysis by electrospray ionization. The LTP probe in combination with miniaturized mass spectrometers has the potential to provide direct composition analysis and source identification of crude oil contaminations in the future.

Identification of collagen IV derived danger/alarm signals in insect immunity by nanoLC-FTICR MS.

The immune system can be stimulated by microbial molecules as well as by endogenously derived danger/alarm signals of host origin. Using the lepidopteran model insect Galleria mellonella, we recently discovered that fragments of collagen IV, resulting from hydrolysis by microbial metalloproteinases, represent danger/alarm signals in insects. Here, we characterized immune-stimulatory peptides generated by thermolysin-mediated degradation of collagen IV using nanospray ionization Fourier transform ion cyclotron resonance mass spectrometry (FTICR MS) after separation by nanoscale liquid chromatography (nanoLC). The combination of FTICR MS analysis and de novo peptide sequencing resulted in the identification of 38 specific collagen IV fragments of which several peptides included the integrin-binding motif RGD/E known from numerous mammalian immune-related proteins. Custom-synthesized peptides corresponding either to the presently identified collagen peptide GIRGEHyp or to a well-known integrin-binding RGD peptide (GRGDS) were injected into G. mellonella to determine their immune-stimulatory activities in vivo. Both peptides stimulated immune cells and systemically the expression of lysozyme and a specific inhibitor of microbial metalloproteinases. Further examination using specific MAP kinase inhibitors indicated that MEK/ERK and p38 are involved in RGD/E-mediated immune-signaling pathways, whereas JNK seems to play only a minor role.

High-resolution mass spectrometry driven discovery of peptidic danger signals in insect immunity.

The 'danger model' is an alternative concept for immune response postulating that the immune system reacts to entities that do damage (danger associated molecular patterns, DAMP) and not only to entities that are foreign (pathogen-associated molecular patterns, PAMP) as proposed by classical immunology concepts. In this study we used Galleria mellonella to validate the danger model in insects. Hemolymph of G. mellonella was digested with thermolysin (as a representative for virulence-associated metalloproteinases produced by humanpathogens) followed by chromatographic fractionation. Immune-stimulatory activity was tested by measuring lysozyme activity with the lytic zone assays against Micrococcus luteus cell wall components. Peptides were analyzed by nano-scale liquid chromatography coupled to high-resolution Fourier transform mass spectrometers. Addressing the lack of a genome sequence we complemented the rudimentary NCBI protein database with a recently established transcriptome and de novo sequencing methods for peptide identification. This approach led to identification of 127 peptides, 9 of which were identified in bioactive fractions. Detailed MS/MS experiments in comparison with synthetic analogues confirmed the amino acid sequence of all 9 peptides. To test the potential of these putative danger signals to induce immune responses we injected the synthetic analogues into G. mellonella and monitored the anti-bacterial activity against living Micrococcus luteus. Six out of 9 peptides identified in the bioactive fractions exhibited immune-stimulatory activity when injected. Hence, we provide evidence that small peptides resulting from thermolysin-mediated digestion of hemolymph proteins function as endogenous danger signals which can set the immune system into alarm. Consequently, our study indicates that the danger model also plays a role in insect immunity.