The determination of the amino acid composition and sequence of a peptide, using both conventional and tandem fast atom bombardment mass spectrometry, is presented. First a list of potential empirical formulae is generated using the accurate mass and isotopic peak intensity ratios from the molecular ion cluster. The formulae are mathematically analyzed to ensure that they correspond to linear peptides composed of 19 common amino acids. Then using the ions below m/z 160 Da, which are characteristic of the amino acid content of the peptide, the formulae are decomposed into possible amino acid compositions. From each composition sequences are generated and their predicted fragment ions are compared to ions present in the mass spectrum. A score based on the intensities and the occurrence of consecutive sequence fragments is calculated. Finally sequences with the ten highest scores are retained. Analyses performed on typical peptides with molecular weights below 1500 Da indicate that identification can generally be achieved using this approach.

The peak intensities observed in the molecular ion regions of fast-atom bombardment/liquid secondary ion mass spectra contain contributions from the parent ion species, its one- and two-electron reduction/oxidation products, and chemical background signal due to beaminduced damage. There are several solution and instrumental parameters that can affect the distribution of peak intensities in the molecular ion region. In this study, the analyte concentration and primary beam density and energy were varied systematically to investigate their effects on the measured peak intensities. A computer algorithm, Simbroc (Simulated Background and Reduction/Oxidation Calculations), was designed to deconvolute the observed intensities into their individual components so that the true effects of experimental parameters on redox extent and background levels could be evaluated. The algorithm is based on a comprehensive seven-variable mathematical model for experimental data simulation. The results obtained using the algorithm after its validation indicate that the primary beam energy does not significantly affect redox extent or background levels. Changes in analyte concentration and primary beam density tend to play a more important role in the generation of redox products and beam-induced damage. The background level generally increases as the analyte concentration is lowered for the peptide systems used in this study. An increase in the primary beam density often leads to higher background levels, although the effect is less detectable for samples that have a low (less than 3%) background signal. The apparent two-electron reduction is generally lower at the higher concentrations; however, the "true" reduction occurring for pentaphenylalanine does not show a significant concentration effect.

No abstract

A method for calculating the empirical formulae of peptides from mass spectrometric data is described. Exact mass measurement data and isotopic peak ratios are used to generate a list of potential empirical formulae that fit a given compound within experimental error. The formulae are then analysed by a mathematical algorithm and only those corresponding to chain peptides formed from the basic amino acids are retained. Calculations conducted for typical peptides indicate that the approach may be useful for peptide identification if the experimental values are determined within an acceptable range of errors. Experimental measurements of the exact mass and isotopic peak ratios made using typical peptides demonstrate the feasibility of the approach.

An approach which allows the determination of the specific effects of experimental parameters on the extent of reduction observed in fast-atom bombardmenthiquid-assisted secondary-ion mass spectrometry (FAB/LSIMS) has been developed. The methodology is based on the isolation of the different components-background, reduction extent, and isotopic abundances-which contribute to peak intensities in the molecular-ion cluster. The individual contributions of reduction and background are calculated by an algorithm SRBC (simulated reduction and background calculations) which utilizes a Monte-Carlo method to optimize the values obtained for these variables. This approach has been used to examine the effects of analyte concentration, primary-beam energy and beam density on the extent of reduction in FAB/LSIMS experiments. The results obtained indicate that the beam energy has no significant effect on the apparent reduction, as measured by the experimental (A + 2)IA isotopic peak ratio, while analyte concentration and beam density effects are observed. However, background and reduction data generated by SRBC demonstrate that in some cases there is little effect of these parameters on the true reduction of the analyte and that the increase in apparent reduction with the variation in these parameters results mainly from an artefact created by an increase in the background signal at lower concentrations and higher beam densities.The use of desorption-ionization techniques such as fast-atom bombardment (FAB) and liquid-assisted secondary-ion mass spectrometry (LSIMS) has, over the last decade, expanded the scope of mass spectral analysis. The most unique feature of FAB/LSIMS is the use of a liquid matrix in which the analyte is dissolved.',' Initially, glycerol was employed as the liquid matrix'.' but other materials have since been utilized and these have been the subject of excellent reviews.s5 Several physical and chemical phenomena related, to the use of a li uid matrix have been observed in FAB/LSIMS analysis%7 and although numerous studies have been directed at the investigation of fundamental processes occurring in FAB/LSIMS&" our knowledge of the mechanisms involved is still deficient. One of the phenomena which has been observed under FAB conditions and that has attracted considerable interest is the occurrence of reduction processes in the matrix under bombardment .lLZ9 These reduction processes, which cause the intensities of A + n peaks (A = (M+ H), n 3 1) to be higher than the theoretical values calculated using the natural abundances, have been shown to involve electrons and/or hydrogen 14,21 and have been observed in several chemical systems.The extent of reduction observed in FABILSIMS depends on the nature of the analyt~,'"'6~2~''~3~32 the nature of the r n a t r i~, '~~'~~'~~'~~~~~~~ the presence of matrix additive^'^^'^^'^ and other experimental parameters. Matrices that have electrodradical scavenging properties such as 3-nitrobenzylalcohol (NBA), 2-hydroxyethyl disulfide (HEDS) and 4-hydroxybenzenesulfonic ...