Quantitative reproducibility of mass spectra in matrix-assisted laser desorption ionization and unraveling of the mechanism for gas-phase peptide ion formation

The Influence of MS Imaging Parameters on UV-MALDI Desorption and Ion Yield

Ultraviolet matrix-assisted laser desorption/ionization mass spectrometry imaging (UV-MALDI MSI) is a widely used technique for imaging molecular distributions within biological systems. While much work exists concerning desorption in UV-MALDI MS, the effects of commonly varied parameters for imaging applications (repetition rate, use of continuous raster mode and raster speed), which determine spatial resolution and limits of detection for the technique, remain largely unknown. We use multiple surface characterization modalities to obtain quantitative measurements of material desorption and analyte ion yield in thin film model systems of two matrix compounds, arising from different UV-MALDI MSI sampling conditions. Observed changes in resulting ablation feature point to matrix-dependent spatial resolution and laser-induced matrix modification effects. Analyte ion yields of 10^-9 to 10^-6 are observed. Complex changes in ion yield, between spot and raster sampling and arising from varied laser repetition rate and raster speed, are observed. Graphical Abstract.

Generation and Propagation of MALDI Ion Packets Probed by Sheet-Like Nanosecond UV Laser Light

A sheet-like ultraviolet (UV) probe laser is used to investigate the ejection and propagation of ion packets of matrix CHCA, which are produced by matrix-assisted laser desorption and ionization (MALDI). Laser irradiation of the expanding MALDI plume induced photodissociation of the CHCA-related ions, which existed in a sheet-like volume, leading to their absence in their MALDI signal profiles. The MALDI spectra were measured under varying conditions: the temporal delay of the lasers and the distance of the sheet-like probe laser from the MALDI sample surface. It was found that the center of the (CHCA)H⁺ packets were ejected at 46±11 ns after MALDI laser irradiation, while the (CHCA)₂H⁺ packets were ejected at 64±12 ns, regardless of the magnitude of acceleration static high-voltage in 3.5–5.5 kV. This suggests that (CHCA)₂H⁺ is formed by a proton transfer reaction from (CHCA)H⁺ to (CHCA)₂ in the heated condensed phase and/or near the surface. This study represents the first experimental determination of ion ejection time in the MALDI process, which is also applicable to other species in the MALDI plume.

Matrix Optical Absorption in UV-MALDI MS

In ultraviolet matrix-assisted laser desorption/ionization mass spectrometry (UV-MALDI MS) matrix compound optical absorption governs the uptake of laser energy, which in turn has a strong influence on experimental results. Despite this, quantitative absorption measurements are lacking for most matrix compounds. Furthermore, despite the use of UV-MALDI MS to detect a vast range of compounds, investigations into the effects of laser energy have been primarily restricted to single classes of analytes. We report the absolute solid state absorption spectra of the matrix compounds α-cyano-4-hydroxycinnamic acid (CHCA), para-nitroaniline (PNA), 2-mercaptobenzothiazole (MBT), 2,5-dihydroxybenzoic acid (2,5-DHB), and 2,4,6-trihydroxyacetophenone (THAP). The desorption/ionization characteristics of these matrix compounds with respect to laser fluence was investigated using mixed systems of matrix with either angiotensin II, PC(34:1) lipid standard, or haloperidol, acting as representatives for typical classes of analyte encountered in UV-MALDI MS. The first absolute solid phase spectra for PNA, MBT, and THAP are reported; additionally, inconsistencies between previously published spectra for CHCA are resolved. In light of these findings, suggestions are made for experimental optimization with regards to matrix and laser wavelength selection. The relationship between matrix optical cross-section and wavelength-dependant threshold fluence, fluence of maximum ion yield, and R, a new descriptor for the change in ion intensity with fluence, are described. A matrix cross-section of 1.3 × 10^-17 cm^-2 was identified as a potential minimum for desorption/ionization of analytes. Graphical Abstract ᅟ.

Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry: Mechanistic Studies and Methods for Improving the Structural Identification of Carbohydrates

Although matrix-assisted laser desorption/ionization (MALDI) mass spectrometry is one of the most widely used soft ionization methods for biomolecules, the lack of detailed understanding of ionization mechanisms restricts its application in the analysis of carbohydrates. Structural identification of carbohydrates achieved by MALDI mass spectrometry helps us to gain insights into biological functions and pathogenesis of disease. In this review, we highlight mechanistic details of MALDI, including both ionization and desorption. Strategies to improve the ion yield of carbohydrates are also reviewed. Furthermore, commonly used fragmentation methods to identify the structure are discussed.

Observing Proton Transfer Reactions Inside the MALDI Plume: Experimental and Theoretical Insight into MALDI Gas-Phase Reactions

We evaluated the contribution of gas-phase in-plume proton transfer reactions to the formation of protonated and deprotonated molecules in the MALDI process. A split sample holder was used to separately deposit two different samples, which avoids any mixing during sample preparation. The two samples were brought very close to each other and desorbed/ionized by the same laser pulse. By using a combination of deuterated and non-deuterated matrices, it was possible to observe exclusively in-plume proton transfer processes. The hydrogen/deuterium exchange (HDX) kinetics were evaluated by varying the delayed extraction (DE) time, allowing the desorbed ions and neutrals to interact inside the plume for a variable period of time before being extracted and detected. Quantum mechanical calculations showed that the HDX energy barriers are relatively low for such reactions, corroborating the importance of gas-phase proton transfer in the MALDI plume. The experimental results, supported by theoretical simulations, confirm that the plume is a very reactive environment, where HDX reactions could be observed from 0 ns up to 400 ns after the laser pulse. These results could be used to evaluate the relevance of previously proposed (and partially conflicting) ionization models for MALDI. Graphical Abstract ᅟ.

Bacterial analysis by laser desorption ionization mass spectrometry on amorphous silicon

Lipid profiling in nine bacterial species has been accomplished by laser desorption ionization mass spectrometry (LDI-MS) using amorphous silicon (a-Si) thin film with 100 nm thickness. Lipid ions could be generated by LDI on a-Si regardless of ion acquisition modes because of a thermal property of a-Si to govern laser-induced surface heating. In a comparative study of lipid profiling in Bacillus lichemiformis by LDI-MS and matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS), LDI-MS on a-Si shows a higher efficiency in lipid and lipopeptide detection than MALDI-MS. A total of 53 peaks of lipid ions generated by LDI on a-Si in both acquisition modes for m/z 400-1200 was 1.6 times more than that detected by MALDI-MS using three organic matrices-2,5-dihydroxybenzoic acid, 1,5-diaminonaphthalene, and 2,4,6-trihydroxyacetophenone monohydrate. Also, the authors demonstrate by mass spectrometry imaging (MSI) that LDI-MS provides high detection coverage through whole sample area. MSI results show the detection yield in LDI on a-Si is 94.8% calculated by counting the number of points detected in the analyte ion signal in a whole spot. It means that reproducible detection of lipid ions by LDI-MS is possible even if laser is randomly irradiated at any position within the bacterial sample area applied on a-Si. Lipid profiling by LDI-MS on a-Si was applied to bacterial differentiation of nine bacterial species conducted by performing principal component analysis. Nine bacterial species are successfully distinguishable from each other by LDI-MS lipid profiling.

The Coupled Chemical and Physical Dynamics Model of MALDI

The coupled physical and chemical dynamics model of ultraviolet matrix-assisted laser desorption/ionization (MALDI) has reproduced and explained a wide variety of MALDI phenomena. The rationale behind and elements of the model are reviewed, including the photophysics, kinetics, and thermodynamics of primary and secondary reaction steps. Experimental results are compared with model predictions to illustrate the foundations of the model, coupling of ablation and ionization, differences between and commonalities of matrices, secondary charge transfer reactions, ionization in both polarities, fluence and concentration dependencies, and suppression and enhancement effects.

MALDI ionization mechanisms investigated by comparison of isomers of dihydroxybenzoic acid

Matrix-assisted laser desorption/ionization (MALDI) ion formation mechanisms were investigated by comparison of isomers of dihydroxybenzoic acid (DHB). These exhibit substantially different MALDI performance, the basis for which was not previously understood. Luminescence decay curves are used here to estimate excited electronic state properties relevant for the coupled chemical and physical dynamics (CPCD) model. With these estimates, the CPCD predictions for relative total ion and analyte ion yields are in good agreement with the data for the DHB isomers. Predictions of a thermal equilibrium model were also compared and found to be incompatible with the data. Copyright © 2015 John Wiley & Sons, Ltd.

Ion Yields in the Coupled Chemical and Physical Dynamics Model of Matrix-Assisted Laser Desorption/Ionization

The Coupled Chemical and Physical Dynamics (CPCD) model of matrix assisted laser desorption ionization has been restricted to relative rather than absolute yield comparisons because the rate constant for one step in the model was not accurately known. Recent measurements are used to constrain this constant, leading to good agreement with experimental yield versus fluence data for 2,5-dihydroxybenzoic acid. Parameters for alpha-cyano-4-hydroxycinnamic acid are also estimated, including contributions from a possible triplet state. The results are compared with the polar fluid model, the CPCD is found to give better agreement with the data.

Investigations of Some Liquid Matrixes for Analyte Quantification by MALDI

Sample inhomogeneity is one of the obstacles preventing the generation of reproducible mass spectra by MALDI and to their use for the purpose of analyte quantification. As a potential solution to this problem, we investigated MALDI with some liquid matrixes prepared by nonstoichiometric mixing of acids and bases. Out of 27 combinations of acids and bases, liquid matrixes could be produced from seven. When the overall spectral features were considered, two liquid matrixes using α-cyano-4-hydroxycinnamic acid as the acid and 3-aminoquinoline and N,N-diethylaniline as bases were the best choices. In our previous study of MALDI with solid matrixes, we found that three requirements had to be met for the generation of reproducible spectra and for analyte quantification: (1) controlling the temperature by fixing the total ion count, (2) plotting the analyte-to-matrix ion ratio versus the analyte concentration as the calibration curve, and (3) keeping the matrix suppression below a critical value. We found that the same requirements had to be met in MALDI with liquid matrixes as well. In particular, although the liquid matrixes tested here were homogeneous, they failed to display spot-to-spot spectral reproducibility unless the first requirement above was met. We also found that analyte-derived ions could not be produced efficiently by MALDI with the above liquid matrixes unless the analyte was sufficiently basic. In this sense, MALDI processes with solid and liquid matrixes should be regarded as complementary techniques rather than as competing ones.

Acquisition of the depth profiles and reproducible mass spectra in matrix-assisted laser desorption/ionization of inhomogeneous samples

RATIONALE

In our previous analysis of the matrix-assisted laser desorption/ionization (MALDI) spectra of peptides, we treated their depth profiles in solid samples as homogeneous. Here, we wanted to determine if the reproducible MALDI spectra and linear calibration curves reported previously would be obtained even when the depth profiles were inhomogeneous.

METHODS

We derived a formula relating shot-number-dependent ion abundance data in temperature-controlled MALDI with the analyte depth profile in a solid sample. We prepared samples containing peptides, amino acids, and serotonin in α-cyano-4-hydroxycinnamic acid matrix by vacuum-drying and micro-spotting methods, recorded their MALDI spectra, and analyzed them with the aforementioned formula.

RESULTS

For the samples prepared by vacuum-drying, the analyte depth profiles were inhomogeneous and maximized at the sample surface. Although the MALDI spectra changed as the shot continued, their sum over the entire set of spectra acquired from a spot was reproducible. Similarly, a high-quality calibration curve could be obtained with the spectral data summed over the entire set. Depth profiles were homogeneous for samples prepared by micro-spotting.

CONCLUSIONS

A method has been developed to obtain a reproducible MALDI spectrum and a linear calibration curve for an analyte with an inhomogeneous depth profile in a solid sample.

Quick quantification of proteins by MALDI

Previously, we reported that the matrix-assisted laser desorption ionization spectrum of a peptide became reproducible when an effective temperature was held constant. Using a calibration curve drawn by plotting the peptide-to-matrix ion abundance ratio versus the peptide concentration in a solid sample, a peptide could be quantified without the use of any internal standard. In this work, we quantified proteins by quantifying their tryptic peptides with the aforementioned method. We modified the digestion process; e.g. disulfide bonds were not cleaved, so that hardly any reagent other than trypsin remained after the digestion process. This allowed the preparation of a sample by the direct mixing of a digestion mixture with a matrix solution. We also observed that the efficiency of the matrix-to-peptide proton transfer, as measured by its reaction quotient, was similar for peptides with arginine at the C-terminus. With the reaction quotient averaged over many such peptides, we could rapidly quantify proteins. Most importantly, no peptide standard, not to mention its isotopically labeled analog, was needed in this method.

A Thermal Mechanism of Ion Formation in MALDI

An important recent discovery concerning the fundamentals of matrix-assisted laser desorption/ionization (MALDI) is that the abundance of each ion appearing in a spectrum is fixed, regardless of the experimental condition, when an effective temperature associated with the spectrum is fixed. We describe this phenomenon and the thermal picture for the ion formation in MALDI derived from it. Accepting that matrix-to-analyte proton transfer is in quasi-equilibrium as supported by experimental data, the above thermal determination occurs because the primary (matrix) ion formation processes are thermally governed. We propose that the abundances of the primary ions are limited by the autoprotolysis-recombination process regardless of how they are initially produced. Finally, we note that primary ion formation, secondary (analyte) ion formation, and their dissociations occur sequentially while the effective temperature of the matrix plume falls steadily due to cooling associated with expansion.

Ionization Mechanism of Matrix-Assisted Laser Desorption/Ionization

In past studies, mistakes in determining the ionization mechanism in matrix-assisted laser desorption/ionization (MALDI) were made because an inappropriate ion-to-neutral ratio was used. The ion-to-neutral ratio of the analyte differs substantially from that of the matrix in MALDI. However, these ratios were not carefully distinguished in previous studies. We begin by describing the properties of ion-to-neutral ratios and reviews early experimental measurements. A discussion of the errors committed in previous theoretical studies and a comparison of recent experimental measurements follow. We then describe a thermal proton transfer model and demonstrate how the model appropriately describes ion-to-neutral ratios and the total ion intensity. Arguments raised to challenge thermal ionization are then discussed. We demonstrate how none of the arguments are valid before concluding that thermal proton transfer must play a crucial role in the ionization process of MALDI.

Energetics and kinetics of thermal ionization models of MALDI

Thermal models of ultraviolet MALDI ionization based on the polar fluid concept are re-examined. Key components are very high solvating power of the fluidized matrix and consequent low reaction-free energy, attainment of thermal equilibrium in the fluid, and negligible recombination losses. None of these are found to hold in a MALDI event. The reaction-free energy in the hot matrix must be near the gas phase value, ion formation is too slow to approach equilibrium, and geminate recombination of autoprotolysis pairs greatly increases the initial loss rate. The maximum thermal ion yield is estimated to be many orders of magnitude below experimental values.

Spectral reproducibility and quantification of peptides in MALDI of samples prepared by micro-spotting

Previously, we reported that MALDI spectra of peptides became reproducible when temperature was kept constant. Linear calibration curves derived from such spectral data could be used for quantification. Homogeneity of samples was one of the requirements. Among the three popular matrices used in peptide MALDI [i.e., α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (DHB), and sinapinic acid (SA)], homogeneous samples could be prepared by conventional means only for CHCA. In this work, we showed that sample preparation by micro-spotting improved the homogeneity for all three cases.

Dual track time-of-flight mass spectrometry for peptide quantification with matrix-assisted laser desorption/ionization

RATIONALE

Previously, we reported a method (Anal. Chem. 2012, 84, 10332) for peptide quantification based on matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS). In the method, the peptide-to-matrix ion abundance ratio was utilized. Implementation of the method with a commercial MALDI-TOF can be somewhat inconvenient because matrix-derived ions are routinely deflected away to avoid detector saturation. A solution for this inconvenience is required.

METHODS

We installed a detector to acquire the TOF spectrum of the ions thrown away to avoid detector saturation. By sending the matrix- and peptide-derived ions along two different tracks and detecting them with different detectors, the inconvenience mentioned above could be avoided.

RESULTS

Excellent linearity of the calibration curves obtained by the dual track TOF spectrometry is demonstrated. The method also allows for the acquisition of the tandem mass spectrum of a selected peptide, which can be useful for its identification.

CONCLUSIONS

We devised the dual track MALDI-TOF MS method to avoid detector saturation and demonstrated that the quantification and identification of peptides can be performed simultaneously.

Thermal proton transfer reactions in ultraviolet matrix-assisted laser desorption/ionization

One of the reasons that thermally induced reactions are not considered a crucial mechanism in ultraviolet matrix-assisted laser desorption ionization (UV-MALDI) is the low ion-to-neutral ratios. Large ion-to-neutral ratios (10(-4)) have been used to justify the unimportance of thermally induced reactions in UV-MALDI. Recent experimental measurements have shown that the upper limit of the total ion-to-neutral ratio is approximately 10(-7) at a high laser fluence and less than 10(-7) at a low laser fluence. Therefore, reexamining the possible contributions of thermally induced reactions in MALDI may be worthwhile. In this study, the concept of polar fluid was employed to explain the generation of primary ions in MALDI. A simple model, namely thermal proton transfer, was used to estimate the ion-to-neutral ratios in MALDI. We demonstrated that the theoretical calculations of ion-to-neutral ratios exhibit the same trend and similar orders of magnitude compared with those of experimental measurements. Although thermal proton transfer may not generate all of the ions observed in MALDI, the calculations demonstrated that thermally induced reactions play a crucial role in UV-MALDI.

Chemical aspects of the primary ionization mechanisms in matrix-assisted laser desorption ionization

It has been proposed that the primary ionization mechanism occurring in matrix-assisted laser desorption ionization (MALDI) experiments originates from the presence, in the solid-state matrix-analytes sample, of matrix dimers. These species are formed by the interaction of carboxylic groups present in the matrix molecules with the formation of strong hydrogen bonds. Theoretical calculations proved that the laser irradiation of these structures leads to one or two H-bridge cleavages, giving rise to an "open" dimer structure or to disproportionation with the formation of MH(+) and [M-H](-) species. The ions so formed can be considered highly effective in their reaction with analyte ions, leading to their protonation (or deprotonation). To achieve further evidence for these proposals, in the present study the energetics of the reactions of ions from different aromatic carboxylic acids with two amino acids (glycine and lysine) and three multipeptides (gly-gly, gly-gly-gly and gly-gly-gly-gly) was investigated. The lowest ∆G values were obtained for 2,5- dihydroxybenzoic acid, widely employed as the MALDI matrix. Also, for p-nitrobenzoic acid the reaction is slightly exothermic, while for the other aromatic carboxylic acids derivatives positives values of ∆G are present.

Why do the abundances of ions generated by MALDI look thermally determined?

In a previous study (J. Mass Spectrom. 48, 299-305, 2013), we observed that the abundance of each ion in a matrix-assisted laser desorption ionization (MALDI) spectrum looked thermally determined. To find out the explanation for the phenomenon, we estimated the ionization efficiency and the reaction quotient (QA) for the autoprotolysis of matrix, M + M → [M + H](+) + [M - H](-), from the temperature-controlled laser desorption ionization spectra of α-cyano-4-hydroxycinnamic acid (CHCA) and 2,5-dihydroxybenzoic acid (DHB). We also evaluated the equilibrium constants (KA) for the autoprotolysis at various temperatures by quantum chemical calculation. Primary ion formation via various thermal models followed by autoprotolysis-recombination was compatible with the observations. The upper limit of the effective temperature of the plume where autoprotolysis-recombination occurs was estimated by equating QA with the calculated equilibrium constant. Figure ᅟ

MALDI ionization mechanisms: the coupled photophysical and chemical dynamics model correctly predicts 'temperature'-selected spectra

A number of possible ultraviolet MALDI ionization mechanisms based on different fundamental phenomena have been proposed. Recently, it has been argued, based on 'temperature'-selected spectra, that photoionization models should be rejected in favor of thermal ones. Here, one non-thermal photoionization model, the coupled photophysical and chemical dynamics model, is shown to be fully consistent with the data.

Efficient methods to generate reproducible mass spectra in matrix-assisted laser desorption ionization of peptides

In our previous matrix-assisted laser desorption ionization (MALDI) studies of peptides, we found that their mass spectra were virtually determined by the effective temperature in the early matrix plume, Tearly, when samples were rather homogeneous. This empirical rule allowed acquisition of quantitatively reproducible spectra. A difficulty in utilizing this rule was the complicated spectral treatment needed to get Tearly. In this work, we found another empirical rule that the total number of particles hitting the detector, or TIC, was a good measure of the spectral temperature and, hence, selection of spectra with the same TIC resulted in reproducible spectra. We also succeeded in obtaining reproducible spectra throughout a measurement by controlling TIC near a preset value through feedback adjustment of laser pulse energy. Both TIC selection and TIC control substantially reduced the shot-to-shot spectral variation in a spot, spot-to-spot variation in a sample, and even sample-to-sample variation in MALDI using α-cyano-4-hydroxycinnamic acid or 2,5-dihydroxybenzoic acid as matrix. Based on the utilization of acquired data, TIC control was more efficient than TIC selection by an order of magnitude. Both techniques produced calibration curves with excellent linearity, suggesting their utility in quantification of peptides.