Mass spectrometric characterization of lipid-modified peptides for the analysis of acylated proteins

The analysis of acylated proteins by mass spectrometry (MS) has largely been overshadowed in proteomics by the analysis of glycosylated and phosphorylated proteins; however, lipid modifications on proteins are proving to be of increasing importance in biomedical research. In order to identify the marker ions and/or neutral loss fragments that are produced upon collision-induced dissociation, providing a means to identify the common lipid modifications on proteins, peptides containing an N-terminally myristoylated glycine, a palmitoylated cysteine and a farnesylated cysteine were chemically synthesized. Matrix-assisted laser desorption/ionization time-of-flight time-of-flight (MALDI-TOF-TOF), electrospray ionization quadrupole time-of-flight (ESI Q-TOF), and electrospray ionization hybrid triple-quadrupole/linear ion trap (ESI QqQ(LIT)) mass spectrometers were used for the analysis. The peptide containing the N-terminally myristoylated glycine, upon CID, produced the characteristic fragments a1 (240.4 Th) and b1 (268.4 Th) ions as well as a low-intensity neutral loss of 210 Da (C14H26O). The peptides containing a farnesylated cysteine residue fragmented to produce a marker ion at a m/z of 205 Th (C15H25) as well as other intense farnesyl fragment ions, and a neutral loss of 204 Da (C15H24). The peptides containing a palmitoylated cysteine moiety generated neutral losses of 238 Da (C16H30O) and 272 Da (C16H32OS); however, no marker ions were produced. The neutral losses were more prominent in the MALDI-TOF-TOF spectra, whereas the marker ions were more abundant in the ESI QqQ(LIT) and Q-TOF mass spectra.     

Neutral loss of isocyanic acid in peptide CID spectra: A novel diagnostic marker for mass spectrometric identification of protein citrullination

Protein citrullination is emerging as an important signaling mechanism that modulates a variety of biological processes. This protein modification constitutes only a 1 Da mass shift, and can be readily confused with other common protein modifications that yield an identical mass shift. In an attempt to develop a robust methodology for detection of protein citrullination sites, we analyzed synthetic citrulline-containing peptides by electrospray ionization tandem mass spectrometry. Collision-induced dissociation (CID) spectra revealed abundant neutral loss of 43 Da from citrullinated peptide precursor ions, which was reconciled by elimination of the HNCO moiety (isocyanic acid) from the citrulline ureido group. The elimination occurs readily in multiple charge states of precursor ions and also in b and y ions. HNCO loss in CID spectra provides a novel diagnostic marker for citrullination, and its utility was demonstrated by the discovery of Arg197 as the specific site of citrullination on nucleophosmin upon peptidylarginine deiminase 4 treatment.     

Molecular characterization of homo- and heterodimeric mercury(II)-<Emphasis Type="Italic">bis</Emphasis>-thiolates of some biologically relevant thiols by electrospray ionization and triple quadrupole tandem mass spectrometry

A series of 24 compounds of general formula R(1)S-Hg-SR(2), R(1) and R(2) being biologically relevant thiol-containing amino acids and peptides (cysteine, homo-cysteine, penicillamine, N-acetyl-cysteine, N-acetyl-penicillamine, cysteinyl-glycine, gamma-glutamyl-cysteine and glutathione) were prepared by direct reaction of mercury(II) ions and thiols in water at millimolar concentration. The obtained products were characterized by electrospray ionization and triple quadrupole tandem mass spectrometry. The source spectra of equimolar mixtures of two different thiols reacting with a stoichiometric amount of mercury(II) show the peak clusters of the three theoretically expected bis-thiolato-mercury(II) complexes with relative intensities close to the theoretically expected 1:2:1 ratio, thus pointing at lack of substantial discrimination between the different thiols, the only observed exception being homo-cysteine, which is less reactive than cysteine and penicillamine. The fragment spectra are structure-specific for the different ligands bound to the metal ion and allow a stand-alone discrimination of some constitutional isomer pairs. Among the peculiar fragmentation processes observed, loss of neutral ammonia from protonated symmetrical and unsymmetrical mercury(II)-bis-thiolates with free, protonizable amino groups leads to the formation of thiirane-carboxylic bound species; this process is suppressed when the protonated amino group is in the gamma-position with respect to the sulfur atom, as in the case of compounds with homo-cysteine. This unusual behavior may hint at unforeseen mechanisms for the interaction of mercury(II) with biological structures, ultimately leading to cellular and organ toxicity. Compounds with N-acetylated amino acids show distinctive fragment ions to which the connectivity of a protonated 2-methyl-oxazoline-5-carboxylic acid may be proposed on the basis of the loss of water and of the elements of formic acid. Finally, the adducts of mercury(II) with glutathione and gamma-glutamyl-cysteine feature a distinctive decomposition channel by loss of a pyroglutamic unit, much the same as protonated glutathione, glutathione disulfide, the S-glutathionyl adducts of biologically occurring electrophiles and other (pseudo)-peptides with gamma-glutamyl bonds.     

Charge transfer through modified peptide nucleic acids

We studied the charge transfer properties of bipyridine-modified peptide nucleic acid (PNA) in the absence and presence of Zn(II). Characterization of the PNA in solution showed that Zn(II) interacts with the bipyridine ligands, but the stability of the duplexes was not affected significantly by the binding of Zn(II). The charge transfer properties of these molecules were examined by electrochemistry for self-assembled monolayers of ferrocene-terminated PNAs and by conductive probe atomic force microscopy for cysteine-terminated PNAs. Both electrochemical and single molecular studies showed that the bipyridine modification and Zn(II) binding do not affect significantly the charge transfer of the PNA duplexes.     

Fragmentation of phosphopeptides by atmospheric pressure MALDI and ESI/ion trap mass spectrometry

An investigation of phosphate loss from phosphopeptide ions was conducted, using both atmospheric pressure matrix-assisted laser desorption/ionization (AP MALDI) and electrospray ionization (ESI) coupled to an ion trap mass spectrometer (ITMS). These experiments were carried out on a number of phosphorylated peptides in order to investigate gas phase dephosphorylation patterns associated with phosphoserine, phosphothreonine, and phosphotyrosine residues. In particular, we explored the fragmentation patterns of phosphotyrosine containing peptides, which experience a loss of 98 Da under collision induced dissociation (CID) conditions in the ITMS. The loss of 98 Da is unexpected for phosphotyrosine, given the structure of its side chain. The fragmentation of phosphoserine and phosphothreonine containing peptides was also investigated. While phosphoserine and phosphothreonine residues undergo a loss of 98 Da under CID conditions regardless of peptide amino acid composition, phosphate loss from phosphotyrosine residues seems to be dependent on the presence of arginine or lysine residues in the peptide sequence.     

Dual nanomolar and picomolar Zn(II) binding properties of metallothionein

Each of the seven Zn(II) ions in the Zn(3)S(9) and Zn(4)S(11) clusters of human metallothionein is in a tetrathiolate coordination environment. Yet analysis of Zn(II) association with thionein, the apoprotein, and analysis of Zn(II) dissociation from metallothionein using the fluorescent chelating agents FluoZin-3 and RhodZin-3 reveal at least three classes of sites with affinities that differ by 4 orders of magnitude. Four Zn(II) ions are bound with an apparent average log K of 11.8, and with the methods employed, their binding is indistinguishable. This binding property makes thionein a strong chelating agent. One Zn(II) ion is relatively weakly bound, with a log K of 7.7, making metallothionein a zinc donor in the absence of thionein. The binding data demonstrate that Zn(II) binds with at least four species: Zn(4)T, Zn(5)T, Zn(6)T, and Zn(7)T. Zn(5)T and Zn(6)T bind Zn(II) with a log K of approximately 10 and are the predominant species at micromolar concentrations of metallothionein in cells. Central to the function of the protein is the reactivity of its cysteine side chains in the absence and presence of Zn(II). Chelating agents, such as physiological ligands with moderate affinities for Zn(II), cause dissociation of Zn(II) ions from metallothionein at pH 7.4 (Zn(7)T <==> Zn(7-n)T + nZn(2+)), thereby affecting the reactivity of its thiols. Thus, the rate of thiol oxidation increases in the presence of Zn(II) acceptors but decreases if more free Zn(II) becomes available. Thionein is such an acceptor. It regulates the reactivity and availability of free Zn(II) from metallothionein. At thionein/metallothionein ratios > 0.75, free Zn(II) ions are below a pZn (-log[Zn(2+)](free)) of 11.8, and at ratios < 0.75, relatively large fluctuations of free Zn(II) ions are possible (pZn between 7 and 11). These chemical characteristics match cellular requirements for Zn(II) and suggest how the molecular structures and redox chemistries of metallothionein and thionein determine Zn(II) availability for biological processes.     

Mass spectrometric evidence for different complexes of peptides and proteins with arsenic(III), arsenic(V), copper(II), and zinc(II) species

Trivalent and pentavalent arsenic were incubated with sulfur-containing amino acid, peptide and protein solutions both as organic compounds (phenylarsine oxide, phenylarsonic acid, dimethylarsinic acid, monomethylarsonic acid) and as inorganic compounds (arsenite, As(III), and arsenate, As(V)). After incubation of phenylarsine oxide solutions with cysteine and glutathione the mass spectra showed a covalent bond between arsenic and sulfur, which was stable at both acidic and neutral pH values. The mass spectra were dominated by monovalent ions at m/z 272 for cysteine samples and at m/z 458 for glutathione samples. Based on these masses the ionic structures could be ascribed to either fragment ions of the covalent arsenic-sulfur complexes or to other arsenic-bonding sites presumably at the amino group. Interestingly, under the same conditions no interactions of inorganic arsenite or arsenate could be measured. In the presence of added Cu(2+) ions all mass signals caused by a reaction of phenylarsine oxide with glutathione disappeared. In these mass spectra only the oxidised form of glutathione (GSSG) was found because of the redox activity of Cu(II). For the model protein lysozyme, no interactions with arsenic could be detected, whereas definite Cu- and Zn-lysozyme complexes with a stoichiometry of 1:1 and 2:1 for Zn(2+) ions and Cu(2+) ions, respectively, were observed. In contrast, for thioredoxin a bonding of As that depended on the concentration of the disulfide-reducing agent tris(2-carboxyethyl) phosphine was demonstrated.For three different phenylarsonic acids and for dimethylarsinic acid that all contain pentavalent arsenic, complexes with glutathione appeared in the mass spectra, which can be attributed to non-covalent interactions or to a covalent bond caused by an additive reaction.The optimisation of the experimental conditions necessary for the mass spectrometric analysis of the interactions of the arsenic species with peptides and proteins is described and the obtained mass spectra that provide information on the kinds of bonds are discussed.     

Interaction of angiotensin peptides and zinc metal ions probed by electrospray ionization mass spectrometry

Electrospray ionization-tandem mass spectrometry experiments were used to provide evidence regarding the sites of interactions between zinc metal ions and angiotensin peptides. The electrospray ionization mass spectra of histidine-containing human angiotensin II (Asp-Arg-Val-Tyr-Ile-His-Pro-Phe) and angiotensin I (Asp-Arg-Val-Tyr-Ile-His-Pro-Phe-His-Leu) in the presence of zinc show abundant multiply charged ions for the zinc-attached peptide [M + aZn²⁺ +(c ? 2a)H⁺]^c+, where a = 1, 2 and c is charge. From collisionally activated dissociation experiments, with both low energy (triple quadrupole mass spectrometry) and high energy collisions (linked scan at constant B/E with a double focusing instrument) of the [M + Zn]²⁺ and [M + Zn + H]³⁺ ions for angiotensin II, a [b ₆ + Zn]²⁺ species is produced as the most abundant product ion, suggesting that the zinc interaction site is in the vicinity of the His⁶ residue. Additionally, tandem mass spectra from the zinc-attached ions for angiotensin I show abundant [b ₆ + Zn]²⁺ and [b ₉ + Zn]²⁺ products, providing evidence that both His⁶ and His⁹ are involved in zinc coordination.     

Sequencing cyclic peptide inhibitors of mammalian ribonucleotide reductase by electrospray ionization mass spectrometry.

Mammalian ribonucleotide reductase (mRR) is a potential target for cancer intervention. A series of lactam-bridged cyclic peptide inhibitors (1-9) of mRR have been synthesized and tested in previous work. These inhibitors consist of cyclic and linear regions, causing their mass spectral characterization to be a challenge. We determined the fragmentation mechanism of cyclic peptides 1-9 using an ion-trap mass spectrometer equipped with an ESI source. Low-energy collision-induced dissociation of sodiated cyclic peptides containing linear branches follows a general pathway. Fragmentation of the linear peptide region produced mainly a and b ions. The ring peptide region was more stable and ring opening required higher collision energy, mainly occurring at the amide bond adjacent to the lactam bridge. The sodium ion, which bound to the carbonyl oxygen of the lactam bridge, acted as a fixed charge site and directed a charge-remote, sequence-specific fragmentation of the ring-opened peptide. Amino acid residues were cleaved sequentially from the C-terminus to the N-terminus. Our findings have established a new way to sequence cyclic peptides containing a lactam bridge based on charge-remote fragmentation. This methodology will permit unambiguous identification of high-affinity ligands within cyclic peptide libraries.     

Differential labeling of free and disulfide-bound thiol functions in proteins

A method for the simultaneous determination of the number of free cysteine groups and disulfide-bound cysteine groups in proteins has been developed based on the sequential labeling of free and bound thiol functionalities with two ferrocene-based maleimide reagents. Liquid chromatography/electrochemistry/mass spectrometry was used to assign the N-(2-ferroceneethyl)maleimide (FEM) labeled free cysteine functionalities in a tryptic digest mixture, whereas a precursor ion scan enables the detection of peptides with ferrocenecarboxylic acid-(2-maleimidoyl)ethylamide (FMEA) labeled disulfide-bound cysteine groups after reduction. Fragment spectra of the labeled peptides yield an excellent coverage of b-type and y-type ions. The ferrocene labeled cysteines were fragmented as 412 Da (FEM) and 455 Da (FMEA). These fragment masses are significantly higher than unlabeled amino acids or dipeptides and are easily detected. The position of free and disulfide-bound cysteine may therefore be assigned in an amino acid sequence.     

Copper-induced oligomerization of peptides: a model study

In this work, copper-binding of the tetraglycine peptide (Gly-Gly-Gly-Gly) was studied by electrospray ionization mass spectrometry. Experiments were performed under alkaline conditions, in the presence of ethanolamine (pH 10.95). We observed that the presence of copper(II) ions induces the aggregation of the peptide and the formation of copper-bound complexes with higher molecular mass is favored, such as the oligomer complexes [3M+2Cu-3H](+) and [4M+3Cu-5H](+). At 1:1 peptide-copper(II) ion ratio, the singly charged [3M+2Cu-3H](+) oligomer complex is the base peak in the mass spectrum. Metal ion-induced oligomer-ization of neurotoxic peptides is well known in the literature; however, there are very few examples in which such oligomerization was directly observed by mass spectrometry. Our results show that application of short peptides can be useful to study the -mechanism of metal ion binding and metal ion-induced oligomerization of peptides.     

Mechanistic studies of Cu(II) binding to amyloid-beta peptides and the fluorescence and redox behaviors of the resulting complexes

Due in large part to the lack of crystal structures of the amyloid-beta (Abeta) peptide and its complexes with Cu(II), Fe(II), and Zn(II), characterization of the metal-Abeta complex has been difficult. In this work, we investigated the complexation of Cu(II) by Abeta through tandem use of fluorescence and electron paramagnetic resonance (EPR) spectroscopies. EPR experiments indicate that Cu(II) bound to Abeta can be reduced to Cu(I) using sodium borohydride and that both Abeta-Cu(II) and Abeta-Cu(I) are chemically stable. Upon reduction of Cu(II) to Cu(I), the Abeta fluorescence, commonly reported to be quenched upon Abeta-Cu(II) complex formation, can be regenerated. The absence of the characteristic tyrosinate peak in the absorption spectra of Abeta-Cu(II) complexes provides evidence that the sole tyrosine residue in Abeta is not one of the four equatorial ligands bound to Cu(II), but remains close to the metal center, and its fluorescence is sensitive to the copper oxidation state and perturbations in the coordination sphere. Further analysis of the quenching and Cu(II) binding behaviors at different Cu(II) concentrations and in the presence of the competing ligand glycine offers evidence supporting the operation of two binding regimes which demonstrate different levels of fluorescence recovery upon addition of the reducing agent. We provide results that suggest the fluorescence quenching is likely caused by charge transfer processes. Thus, by using tyrosine to probe the coordination site, fluorescence spectroscopy provides valuable mechanistic insights into the oxidation state of copper ions bound to Abeta, the binding heterogeneity, and the influence of solution conditions on complex formation.     

Complexes of silver(I) with peptides and proteins as produced in electrospray mass spectrometry

Silver(I) forms aqueous phase complexes with both sulfur and nonsulfur containing peptides and proteins. These complexes were introduced into the gas phase via electrospray, and their structures probed by means of tandem mass spectrometry. Experiments with di-, tri-, and oligopeptides show that the abundance of silver(I)-containing ions increases relative to that of proton-containing ions as peptide length increases. This increase is much more dramatic for methionine-containing peptides. Collision-induced dissociation of silver-peptide complexes yields a multitude of product ions that are silver containing. However, even for methioninecontaining peptides, very few of these product ions contain the methionine residue. The solution-phase structure and the gas-phase structure of the silver/peptide complex are not identical. The methionine sulfur acts as the silver anchoring point in solution. Desolvation in the gas phase leads to a rearrangement of the silver/peptide complex such that the silver ion becomes chelated to the nitrogen and oxygen atom on the peptide backbone in addition to the methionine sulfur. This rearrangement decreases the importance of the silver/sulfur bond to the extent that it is frequently broken upon collision activation and leads to the formation of silver/peptide product ions that are nonsulfur bearing.     

Fragmentation of cationized phosphotyrosine containing peptides by atmospheric pressure MALDI/Ion trap mass spectrometry

An investigation of phosphate loss from sodium-cationized phosphotyrosine containing peptide ions was conducted using liquid infrared (2.94 microm) atmospheric pressure matrix-assisted laser desorption/ionization (AP MALDI) coupled to an ion trap mass spectrometer (ITMS). Previous experiments in our laboratory explored the fragmentation patterns of protonated phosphotyrosine containing peptides, which experience a loss of 98 Da under CID conditions in the ITMS. This loss of 98 Da is unexpected for phosphotyrosine, given the structure of its side chain. Phosphate loss from phosphotyrosine residues seems to be dependent on the presence of arginine or lysine residues in the peptide sequence. In the absence of a basic residue, the protonated phosphotyrosine peptides do not undergo losses of HPO(3) (Delta 80 Da) nor HPO(3) + H(2)O (Delta 98 Da) in their CID spectra. However, sodium cationized phosphotyrosine containing peptides that do not contain arginine or lysine residues within their sequences do undergo losses of HPO(3) (Delta 80 Da) and HPO(3) + H(2)O (Delta 98 Da) in their CID spectra.     

Gas-Phase Peptide Sulfinyl Radical Ions: Formation and Unimolecular Dissociation

A variety of peptide sulfinyl radical (RSO?) ions with a well-defined radical site at the cysteine side chain were formed at atmospheric pressure (AP), sampled into a mass spectrometer, and investigated via collision-induced dissociation (CID). The radical ion formation was based on AP reactions between oxidative radicals and peptide ions containing single inter-chain disulfide bond or free thiol group generated from nanoelectrospray ionization (nanoESI). The radical induced reactions allowed large flexibility in forming peptide radical ions independent of ion polarity (protonated or deprotonated) or charge state (singly or multiply charged). More than 20 peptide sulfinyl radical ions in either positive or negative ion mode were subjected to low energy collisional activation on a triple-quadrupole/linear ion trap mass spectrometer. The competition between radical- and charge-directed fragmentation pathways was largely affected by the presence of mobile protons. For peptide sulfinyl radical ions with reduced proton mobility (i.e., singly protonated, containing basic amino acid residues), loss of 62?Da (CH₂SO), a radical-initiated dissociation channel, was dominant. For systems with mobile protons, this channel was suppressed, while charge-directed amide bond cleavages were preferred. The polarity of charge was found to significantly alter the radical-initiated dissociation channels, which might be related to the difference in stability of the product ions in different ion charge polarities.     

On-line electrogeneration of copper-peptide complexes in microspray mass spectrometry

The interaction of copper ions with peptides was investigated by electrospray mass spectrometry. Two electrospray micro-emitters were compared, the first one with a platinum electrode using a copper(II) electrolyte solution containing a peptide sample, and the second one with a sacrificial copper anode in a water/methanol solution containing only a peptide (i.e., angiotensin III, bradykinin, or Leu-enkephalin). The former yielded mainly Cu(2+) complexes either with histidine residues or with the peptide backbone (Cu(+) complexes can be also formed due to gas-phase reactions), whereas the latter can generate a mixture of both Cu(+) and Cu(2+) aqueous complexes that yield different complexation patterns. This study shows that electrospray emitters with soluble copper anodes enable the study of Cu(I)-peptide complexes in solution.     

Nano-electrospray and microbore liquid chromatography-ion trap mass spectrometry studies of copper complexation with MHC restricted peptides

The formation of copper/peptide complex ions by nano-electrospray and microbore HPLC-electrospray mass spectrometry has been investigated for major histocompatibility complex (MHC) class I and class II restricted peptides. Post-column addition of copper(II) acetate following microbore HPLC-MS separation was carried out using a mixing T-piece or via the sheath flow inlet of the electrospray source. Optimal analytical conditions for copper complex ion formation were determined by variation of copper concentration, pH, nebulization gas supply and spray voltage. Tandem mass spectrometry of copper/peptide complex ions provides peptide sequence information and insight into the peptide chelation sites. Copper associated y fragment ions dominate the product ion spectrum for non-histidine containing peptides, but both b and y copper complex ions were observed for the histidine containing MHC class I associated peptide gp70.     

Selective detection of thiosulfate-containing peptides using tandem mass spectrometry

Incubation of proteins or peptides containing disulfide bonds (S-S) with sodium sulfite (Na(2)SO(3)) cleaves S-S bonds producing approximately equimolar amounts of free thiols (-SH) and thiosulfates (-S-SO(3)H), a process known as sulfitolysis. Proteins and peptides containing thiosulfates were separated by reverse-phase high-performance liquid chromatography (RP-HPLC) and characterized by mass spectrometry (MS) and peptide mapping. The mass of the thiosulfate-containing peptide formed from oxidized insulin B chain was 3478.02 Da, 80 Da greater than the reduced peptide and corresponding precisely to addition of sulfur trioxide (SO(3)). Disulfide bond cleavage was also observed using RP-HPLC and MS after incubation of the intramolecular homodimer of mouse S100A8 (mass 20614 Da). The mass of HPLC-separated A8-SH was 10308 Da, and 10388 Da for A8-S-SO(3)H. Loss of SO(3) from multiply charged precursor ions was generally observed at elevated declustering potentials in the source region or within q(2) at relatively low collision energies (approximately 20 V). The characteristic loss of SO(3) at low collision energies preceded peptide backbone fragmentations at higher collision energies. Accurate mass measurement and charge-state discrimination, using a hybrid quadrupole time-of-flight mass spectrometer, allowed specific detection of thiosulfate-containing peptides. An information-dependent acquisition method, where the switch criterion was loss of m/z 79.9568, specifically identified 11 thiosulfate-containing peptides using nano-LC/MS from a tryptic digest of bovine serum albumin (BSA).     

Electrospray ionization detection of inherently nonresponsive epoxides by peptide binding

A small organic molecule that is inherently nonresponsive to electrospray analysis, 1,3-butadiene diepoxide, was analyzed via electrospray ionization (ESI) by binding it to various peptides and observing the product at the characteristic mass shift. The epoxide reacted only with peptides with arginines in their sequence, most likely through a base-catalyzed ring opening to form a covalently bound product. A calibration curve linear over 3 orders of magnitude was generated for the butadiene diepoxide/peptide adduct. Several other epoxides were also reacted with the peptide of choice (angiotensin II), and adducts of these epoxides with the peptide were observed as well, demonstrating the versatility of this method for the analysis of small epoxides. This study demonstrates the possibility of assaying epoxides bound to peptides or proteins in biological samples. Furthermore, it demonstrates an important concept that could be applied to other analytical problems in electrospray: the ability to react an analyte that is nonresponsive to electrospray analysis with an analyte well suited for the technique, and accomplish quantitation based on the adduct formed between the two.     

New example of proline-induced fragmentation in electrospray ionization mass spectrometry of peptides

The positive ion electrospray ionization (ESI+) mass spectra of peptides usually display only protonated molecules provided that soft ionization conditions are applied (low cone voltage to prevent in-source dissociations). Such ions can be multiply charged depending on the molecular weight of the studied compounds. We have experienced an unexpected behavior during the ESI analysis of a modified peptide of relatively high mass (3079 Da). A specific fragmentation occurred even under soft energetic conditions, leading to a mass spectrum containing multiply charged molecular and fragment ions. The selective rupture involved the amide bond between the glutamic acid and proline residues (E-P sequence). The successive replacement of each amino acid by an alanine residue (positional scanning study) was undertaken to assess which part of the sequence induced such selective and abundant fragmentation on multiply charged species. The succession P-P was evidenced as the minimum unit giving rise to the first peptide bond rupture in the sequence X-P-P. Any acidic amino acid at the X position (X = D, E) favored the fragmentation by an intramolecular interaction. Such proline-induced fragmentation occurring readily in the source differed from the literature data on the specific behavior of proline-containing peptides where bond ruptures occur solely in dissociation conditions.