Fragmentation features of intermolecular cross‐linked peptides using N‐hydroxy‐ succinimide esters by MALDI‐ and ESI‐MS/MS for use in structural proteomics

The use of mass spectrometry coupled with chemical cross‐linking of proteins has become one of the most useful tools for proteins structure and interactions studies. One of the challenges in these studies is the identification of the cross‐linked peptides. The interpretation of the MS/MS data generated in cross‐linking experiments using N‐hydroxy succinimide esters is not trivial once a new amide bond is formed allowing new fragmentation pathways, unlike linear peptides. Intermolecular cross‐linked peptides occur when two different peptides are connected by the cross‐linker and they yield information on the spatial proximity of different domains (within a protein) or proteins (within a complex). In this article, we report a detailed fragmentation study of intermolecular cross‐linked peptides, generated from a set of synthetic peptides, using both ESI and MALDI to generate the precursor ions. The fragmentation features observed here can be helpful in the interpretation and identification of cross‐linked peptides present in cross‐linking experiments and be further implemented in search engine's algorithms. Copyright © 2011 John Wiley & Sons, Ltd.     

Micelle‐Triggered β‐Hairpin to α‐Helix Transition in a 14‐Residue Peptide from a Choline‐Binding Repeat of the Pneumococcal Autolysin LytA

Choline‐binding modules (CBMs) have a ββ‐solenoid structure composed of choline‐binding repeats (CBR), which consist of a β‐hairpin followed by a short linker. To find minimal peptides that are able to maintain the CBR native structure and to evaluate their remaining choline‐binding ability, we have analysed the third β‐hairpin of the CBM from the pneumococcal LytA autolysin. Circular dichroism and NMR data reveal that this peptide forms a highly stable native‐like β‐hairpin both in aqueous solution and in the presence of trifluoroethanol, but, strikingly, the peptide structure is a stable amphipathic α‐helix in both zwitterionic (dodecylphosphocholine) and anionic (sodium dodecylsulfate) detergent micelles, as well as in small unilamellar vesicles. This β‐hairpin to α‐helix conversion is reversible. Given that the β‐hairpin and α‐helix differ greatly in the distribution of hydrophobic and hydrophilic side chains, we propose that the amphipathicity is a requirement for a peptide structure to interact and to be stable in micelles or lipid vesicles. To our knowledge, this “chameleonic” behaviour is the only described case of a micelle‐induced structural transition between two ordered peptide structures.     

Effective Assignment of α2,3/α2,6‐Sialic Acid Isomers by LC‐MS/MS‐Based Glycoproteomics

Distinct structural changes of the α2,3/α2,6‐sialic acid glycosidic linkages on glycoproteins are of importance in cancer biology, inflammatory diseases, and virus tropism. Current glycoproteomic methodologies are, however, not amenable toward high‐throughput characterization of sialic acid isomers. To enable such assignments, a mass spectrometry method utilizing synthetic model glycopeptides for the analysis of oxonium ion intensity ratios was developed. This method was successfully applied in large‐scale glycoproteomics, thus allowing the site‐specific structural characterization of sialic acid isomers.     

Chemical Proteomic Profiling of Protein 4′‐Phosphopantetheinylation in Mammalian Cells

Protein 4′‐phosphopantetheinylation is an essential post‐translational modification (PTM) in prokaryotes and eukaryotes. So far, only five protein substrates of this specific PTM have been discovered in mammalian cells. These proteins are known to perform important functions, including fatty acid biosynthesis and folate metabolism, as well as β‐alanine activation. To explore existing and new substrates of 4′‐phosphopantetheinylation in mammalian proteomes, we designed and synthesized a series of new pantetheine analogue probes, enabling effective metabolic labelling of 4′‐phosphopantetheinylated proteins in HepG2 cells. In combination with a quantitative chemical proteomic platform, we enriched and identified all the currently known 4′‐phosphopantetheinylated proteins with high confidence, and unambiguously determined their exact sites of modification. More encouragingly, we discovered, using targeted chemical proteomics, a potential 4′‐phosphopantetheinylation site in the protein of mitochondrial dehydrogenase/reductase SDR family member 2 (DHRS2).     

Nuclear Magnetic Resonance Structure Elucidation of Peptide b2 Ions

Tandem mass spectrometry (MS/MS) is powerful for chemical identification but it is still insufficient for explicit ion structure determination. A strategy is introduced to elucidate MS fragment ion structures using NMR spectroscopy for the first time. In our experiments, precursor ions are dissociated at atmospheric pressure and the resulting fragment ions are identified by mass spectrometry but collected outside the mass spectrometer, making the subsequent NMR measurements possible. This new strategy has been applied to determine the chemical structure of the characteristic b₂ fragment ion, a subject of longstanding debate in MS‐based proteomics.     

Ubiquitin Associates with the N‐Terminal Domain of Nerve Growth Factor: The Role of Copper(II) Ions

Many biochemical pathways involving nerve growth factor (NGF), a neurotrophin with copper(II) binding abilities, are regulated by the ubiquitin (Ub) proteasome system. However, whether NGF binds Ub and the role played by copper(II) ions in modulating their interactions have not yet been investigated. Herein NMR spectroscopy, circular dichroism, ESI‐MS, and titration calorimetry are employed to characterize the interactions of NGF with Ub. NGF_1–14, which is a short model peptide encompassing the first 14 N‐terminal residues of NGF, binds the copper‐binding regions of Ub (K_D=8.6 10^?5 m ). Moreover, the peptide undergoes a random coil–polyproline type II helix structural conversion upon binding to Ub. Notably, copper(II) ions inhibit NGF_1–14/Ub interactions. Further experiments performed with the full‐length NGF confirmed the existence of a copper(II)‐dependent association between Ub and NGF and indicated that the N‐terminal domain of NGF was a valuable paradigm that recapitulated many traits of the full‐length protein.     

Synthesis of Glc1Man9‐Glycoprotein Probes by a Misfolding/Enzymatic Glucosylation/Misfolding Sequence

Glycoproteins in non‐native conformations are often toxic to cells and may cause diseases, thus the quality control (QC) system eliminates these unwanted species. Lectin chaperone calreticulin and glucosidase II, both of which recognize the Glc₁Man₉ oligosaccharide on glycoproteins, are important components of the glycoprotein QC system. Reported herein is the preparation of Glc₁Man₉‐glycoproteins in both native and non‐native conformations by using the following sequence: misfolding of chemically synthesized Man₉‐glycoprotein, enzymatic glucosylation, and another misfolding step. By using synthetic glycoprotein probes, calreticulin was found to bind preferentially to a hydrophobic non‐native glycoprotein whereas glucosidase II activity was not affected by glycoprotein conformation. The results demonstrate the ability of chemical synthesis to deliver homogeneous glycoproteins in several non‐native conformations for probing the glycoprotein QC system.     

Structure of the O‐Glycosylated Conopeptide CcTx from Conus consors Venom

The glycopeptide CcTx, isolated from the venom of the piscivorous cone snail Conus consors, belongs to the κA‐family of conopeptides. These toxins elicit excitotoxic responses in the prey by acting on voltage‐gated sodium channels. The structure of CcTx, a first in the κA‐family, has been determined by high‐resolution NMR spectroscopy together with the analysis of its O‐glycan at Ser7. A new type of glycopeptide O‐glycan core structure, here registered as core type 9, containing two terminal L ‐galactose units {α‐L ‐Galp‐(1→4)‐α‐D ‐GlcpNAc‐(1→6)‐[α‐L ‐Galp‐(1→2)‐β‐D ‐Galp‐(1→3)‐]α‐D ‐GalpNAc‐(1→O)}, is highlighted. A sequence comparison to other putative members of the κA‐family suggests that O‐linked glycosylation might be more common than previously thought. This observation alone underlines the requirement for more careful and in‐depth investigations into this type of post‐translational modification in conotoxins.     

Oxonium Ions Substituting Cesium Ions in the Structure of the New High‐Pressure Borate HP‐Cs1−x(H3O)xB3O5 (x=0.5–0.7)

The new high‐pressure borate HP‐Cs_1?x(H₃O)_xB₃O₅ (x=0.5–0.7) was synthesized under high‐pressure/high‐temperature conditions of 6 GPa/900 °C in a Walker‐type multianvil apparatus. The compound crystallizes in the monoclinic space group C2/c (Z=8) with the parameters a=1000.6(2), b=887.8(2), c=926.3(2) pm, β=103.1(1)°, V=0.8016(3) nm³, R1=0.0452, and wR2=0.0721 (all data). The boron–oxygen network is analogous to those of the compounds HP‐MB₃O₅, (M=K, Rb) and exhibits all three structural motifs of borates—BO₃ groups, corner‐sharing BO₄ tetrahedra, and edge‐sharing BO₄ tetrahedra—at the same time. Channels inside the boron–oxygen framework contain the cesium and oxonium ions, which are disordered on a specific site. Estimating the amount of hydrogen by solid‐state NMR spectroscopy and X‐ray diffraction led to the composition HP‐Cs_1?x(H₃O)_xB₃O₅ (x=0.5–0.7), which implies a nonzero phase width.     

Independent highly sensitive characterization of asparagine deamidation and aspartic acid isomerization by sheathless CZE‐ESI‐MS/MS

Amino acids residues are commonly submitted to various physicochemical modifications occurring at physiological pH and temperature. Post‐translational modifications (PTMs) require comprehensive characterization because of their major influence on protein structure and involvement in numerous in vivo process or signaling. Mass spectrometry (MS) has gradually become an analytical tool of choice to characterize PTMs; however, some modifications are still challenging because of sample faint modification levels or difficulty to separate an intact peptide from modified counterparts before their transfer to the ionization source. Here, we report the implementation of capillary zone electrophoresis coupled to electrospray ionization tandem mass spectrometry (CZE‐ESI‐MS/MS) by the intermediate of a sheathless interfacing for independent and highly sensitive characterization of asparagine deamidation (deaN) and aspartic acid isomerization (isoD). CZE selectivity regarding deaN and isoD was studied extensively using different sets of synthetic peptides based on actual tryptic peptides. Results demonstrated CZE ability to separate the unmodified peptide from modified homologous exhibiting deaN, isoD or both independently with a resolution systematically superior to 1.29. Developed CZE‐ESI‐MS/MS method was applied for the characterization of monoclonal antibodies and complex protein mixture. Conserved CZE selectivity could be demonstrated even for complex samples, and foremost results obtained showed that CZE selectivity is similar regardless of the composition of the peptide. Separation of modified peptides prior to the MS analysis allowed to characterize and estimate modification levels of the sample independently for deaN and isoD even for peptides affected by both modifications and, as a consequence, enables to distinguish the formation of l ‐aspartic acid or d ‐aspartic acid generated from deaN. Separation based on peptide modification allowed, as supported by the ESI efficiency provided by CZE‐ESI‐MS/MS properties, and enabled to characterize and estimate studied PTMs with an unprecedented sensitivity and proved the relevance of implementing an electrophoretic driven separation for MS‐based peptide analysis. Copyright © 2016 John Wiley & Sons, Ltd.     

Photodissociation Spectroscopy of CD3I+ Generated by Mass‐Analyzed Threshold Ionization for Structure Determination

A method is devised better to resolve the subbands of the ground vibronic band in the mass‐analyzed threshold ionization (MATI) spectrum of CD₃I. By selective photodissociation of CD₃I⁺ in these subbands, high‐resolution spectra for the à ²A₁← ²E_3/2 transition are recorded. Spectral analysis confirms our previous suggestion that these subbands are due to cations in different rotational K states; this demonstrates the capability of MATI to generate rovibronically selected ion beams. By using the rotational constants of CH₃I⁺ and CD₃I⁺ obtained by spectral analysis, the zero‐point‐level geometries of the cations in the ²E_3/2 and à ²A₁ states are determined. To the best of our knowledge, this is the first time that the capability of MATI–PD to determine the geometry of a gas‐phase polyatomic cation in an excited electronic state is demonstrated.     

Biopolymer Skeleton Produced by Rhizobium radiobacter: Stoichiometric Alternation of Glycosidic and Amidic Bonds in the Lipopolysaccharide O‐Antigen

The lipopolysaccharide (LPS) O‐antigen structure of the plant pathogen Rhizobium radiobacter strain TT9 and its possible role in a plant‐microbe interaction was investigated. The analyses disclosed the presence of two O‐antigens, named Poly1 and Poly2. The repetitive unit of Poly2 constitutes a 4‐α‐l ‐rhamnose linked to a 3‐α‐d ‐fucose residue. Surprisingly, Poly1 turned out to be a novel type of biopolymer in which the repeating unit is formed by a monosaccharide and an amino‐acid derivative, so that the polymer has alternating glycosidic and amidic bonds joining the two units: 4‐amino‐4‐deoxy‐3‐O‐methyl‐d ‐fucose and (2′R,3′R,4′S)‐N‐methyl‐3′,4′‐dihydroxy‐3′‐methyl‐5′‐oxoproline). Differently from the O‐antigens of LPSs from other pathogenic Gram‐negative bacteria, these two O‐antigens do not activate the oxidative burst, an early innate immune response in the model plant Arabidopsis thaliana, explaining at least in part the ability of this R. radiobacter strain to avoid host defenses during a plant infection process.     

Local Transient Unfolding of Native State PAI‐1 Associated with Serpin Metastability

The metastability of the native fold makes serpin (serine protease inhibitor) proteins prone to pathological conformational change, often by insertion of an extra β‐strand into the central β‐sheet A. How this insertion is made possible is a hitherto unresolved question. By the use of advanced hydrogen/deuterium‐exchange mass spectrometry (HDX‐MS) it is shown that the serpin plasminogen activator inhibitor 1 (PAI‐1) transiently unfolds under native condition, on a second‐to‐minute time scale. The unfolding regions comprise β‐strand 5A as well as the underlying hydrophobic core, including β‐strand 6B and parts of helices A, B, and C. Based thereon, a mechanism is proposed by which PAI‐1 makes transitions through progressively more unfolded states along the reaction coordinate to the inactive, so‐called latent form. Our results highlight the profound utility of HDX‐MS in detecting sparsely populated, transiently unfolded protein states.     

Characterization of different poly(2‐oxazoline) block copolymers by MALDI‐TOF MS/MS and ESI‐Q‐TOF MS/MS

A complete library of poly(2‐oxazoline) block copolymers was synthesized via cationic ring opening polymerization for the characterization by two different soft ionization techniques, namely matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry (MALDI‐TOF MS) and electrospray ionization quadrupole time‐of‐flight mass spectrometry (ESI‐Q‐TOF MS). In addition, a detailed characterization was performed by tandem MS to gain more structural information about the block copolymer composition and its fragmentation behavior. The fragmentation of the poly(2‐oxazoline) block copolymers revealed the desired polymer structure and possible side reactions, which could be explained by different mechanisms, like 1,4‐ethylene or hydrogen elimination and the McLafferty +1 rearrangement. Polymers with aryl side groups showed less fragmentation due to their higher stability compared to polymers with alkyl side groups. These insights represent a further step toward the construction of a library with fragments and their fragmentation pathways for synthetic polymers, following the successful examples in proteomics. © 2010 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem, 2010     

Block length determination of the block copolymer mPEG‐b‐PS using MALDI‐TOF MS/MS

The molar mass determination of block copolymers, in particular amphiphilic block copolymers, has been challenging with chromatographic techniques. Therefore, methoxy poly(ethylene glycol)‐b‐poly(styrene) (mPEG‐b‐PS) was synthesized by atom transfer radical polymerization (ATRP) and characterized in detail not only by conventional chromatographic techniques, such as size exclusion chromatography (SEC), but also by matrix‐assisted laser/desorption ionization tandem mass spectrometry (MALDI‐TOF MS/MS). As expected, different molar mass values were obtained in the SEC measurements depending on the calibration standards (either PEG or PS). In contrast, MALDI‐TOF MS/MS analysis allowed the molar mass determination of each block, by the scission of the weakest point between the PEG and PS block. Thus, fragments of the individual blocks could be obtained. The PEG block showed a depolymerization reaction, while for the PS block fragments were obtained in the monomeric, dimeric, and trimeric regions as a result of multiple chain scissions. The block length of PEG and PS could be calculated from the fragments recorded in the MALDI‐TOF MS/MS spectrum. Furthermore, the assignment of the substructures of the individual blocks acquired by MALDI‐TOF MS/MS was accomplished with the help of the fragments that were obtained from the corresponding homopolymers. © 2010 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem, 2010