Cdk1-mediated threonine phosphorylation of Sam68 modulates its RNA binding, alternative splicing activity and cellular functions.

Sam68, also known as KHDRBS1, is a member of the STAR family of proteins that directly link signal transduction with post-transcriptional gene regulation. Sam68 controls the alternative splicing of many oncogenic proteins and its role is modulated by post-translational modifications, including serine/threonine phosphorylation, that differ at various stages of the cell cycle. However, the molecular basis and mechanisms of these modulations remain largely unknown. Here, we combined mass spectrometry, nuclear magnetic resonance spectroscopy and cell biology techniques to provide a comprehensive post-translational modification mapping of Sam68 at different stages of the cell cycle in HEK293 and HCT116 cells. We established that Sam68 is specifically phosphorylated at T33 and T317 by Cdk1, and demonstrated that these phosphorylation events reduce the binding of Sam68 to RNA, control its cellular localization and reduce its alternative splicing activity, leading to a reduction in the induction of apoptosis and an increase in the proliferation of HCT116 cells.

Synthesis and Identification of Biologically Active Mono-Labelled FITC-Insulin Conjugate.

Fluorescently labelling proteins such as insulin have wide ranging applications in a pharmaceutical research and drug delivery. Human insulin (Actrapid®) was labelled with fluorescein isothiocyanate (FITC) and the synthesised conjugate identified using reverse phase high performance liquid chromatography (RP-HPLC) on a C18 column and a gradient method with mobile phase A containing 0.1% trifluoroacetic acid (TFA) in Millipore water and mobile phase B containing 90% Acetonitrile, 10% Millipore water and 0.1% TFA. Syntheses were carried out at varying reaction times between 4 and 20 h. Mono-labelled FITC-insulin conjugate was successfully synthesised with labelling at the B1 position on the insulin chain using a molar ratio of 2:1 (FITC:insulin) at a reaction time of 18 h and confirmed by electrospray mass spectroscopy. Reactions were studied across a pH range of 7-9.8 and the quantities switch from mono-labelled to di-labelled FITC-insulin conjugates at a reaction time of 2 h (2:1 molar ratio) at pH > 8. The conjugates isolated from the studies had biological activities in comparison to native insulin of 99.5% monoB1, 78% monoA1, 51% diA1B1 and 0.06% triA1B1B29 in HUVEC cells by examining AKT phosphorylation levels. MonoB1 FITC-insulin conjugate was also compared to native insulin by examining cell surface GLUT4 in C2C12 skeletal muscle cells. No significant difference in the cellular response was observed for monoB1 produced in-house compared to native insulin. Therefore mono-labelled FITC-insulin at the B1 position showed similar biological activity as native insulin and can potentially be used for future biomedical applications.

Biochemical and phenotypic characterisation of the Mycobacterium smegmatis transporter UspABC.

Mycobacterium tuberculosis (Mtb) is an intracellular human pathogen that has evolved to survive in a nutrient limited environment within the host for decades. Accordingly, Mtb has developed strategies to acquire scarce nutrients and the mycobacterial transporter systems provide an important route for the import of key energy sources. However, the physiological role of the Mtb transporters and their substrate preference(s) are poorly characterised. Previous studies have established that the Mtb UspC solute-binding domain recognises amino- and phosphorylated-sugars, indicating that the mycobacterial UspABC transporter plays a key role in the import of peptidoglycan precursors. Herein, we have used a wide array of approaches to investigate the role of UspABC in Mycobacterium smegmatis by analysis of mutant strains that either lack the solute binding domain: ΔuspC or the entire transport complex: ΔuspABC. Analysis of mycobacterial transcripts shows that the uspABC system is functionally expressed in mycobacteria as a contiguous reading frame. Topology mapping confirms an Nin-Cin orientation of the UspAB integral membrane spanning domains. Phenotypic microarray profiling of commercially available sugars suggests, unexpectedly, that the uspC and ΔuspABC mutants had different carbon utilisation profiles and that neither strain utilised glucose-1-phosphate. Furthermore, proteomics analysis showed an alteration in the abundance of proteins involved in sugar and lipid metabolism, crucial for cell envelope synthesis, and we propose that UspABC has an important role in determining the interplay between these pathways.

The RNA-binding protein HuR is a novel target of Pirh2 E3 ubiquitin ligase.

The RING-finger protein Pirh2 is a p53 family-specific E3 ubiquitin ligase. Pirh2 also ubiquitinates several other important cellular factors and is involved in carcinogenesis. However, its functional role in other cellular processes is poorly understood. To address this question, we performed a proteomic search for novel interacting partners of Pirh2. Using the GST-pulldown approach combined with LC-MS/MS, we revealed 225 proteins that interacted with Pirh2. We found that, according to the GO description, a large group of Pirh2-associated proteins belonged to the RNA metabolism group. Importantly, one of the identified proteins from that group was an RNA-binding protein ELAVL1 (HuR), which is involved in the regulation of splicing and protein stability of several oncogenic proteins. We demonstrated that Pirh2 ubiquitinated the HuR protein facilitating its proteasome-mediated degradation in cells. Importantly, the Pirh2-mediated degradation of HuR occurred in response to heat shock, thereby affecting the survival rate of HeLa cells under elevated temperature. Functionally, Pirh2-mediated degradation of HuR augmented the level of c-Myc expression, whose RNA level is otherwise attenuated by HuR. Taken together, our data indicate that HuR is a new target of Pirh2 and this functional interaction contributes to the heat-shock response of cancer cells affecting their survival.

SEMG1/2 augment energy metabolism of tumor cells.

SEMG1 and SEMG2 genes belong to the family of cancer-testis antigens (CTAs), whose expression normally is restricted to male germ cells but is often restored in various malignancies. High levels of SEMG1 and SEMG2 expression are detected in prostate, renal, and lung cancer as well as hemoblastosis. However, the functional importance of both SEMGs proteins in human neoplasms is still largely unknown. In this study, by using a combination of the bioinformatics and various cellular and molecular assays, we have demonstrated that SEMG1 and SEMG2 are frequently expressed in lung cancer clinical samples and cancer cell lines of different origins and are negatively associated with the survival rate of cancer patients. Using the pull-down assay followed by LC-MS/MS mass-spectrometry, we have identified 119 proteins associated with SEMG1 and SEMG2. Among the SEMGs interacting proteins we noticed two critical glycolytic enzymes-pyruvate kinase M2 (PKM2) and lactate dehydrogenase A (LDHA). Importantly, we showed that SEMGs increased the protein level and activity of both PKM2 and LDHA. Further, both SEMGs increased the membrane mitochondrial potential (MMP), glycolysis, respiration, and ROS production in several cancer cell lines. Taken together, these data provide first evidence that SEMGs can up-regulate the energy metabolism of cancer cells, exemplifying their oncogenic features.

Discovery of a heme-binding domain in a neuronal voltage-gated potassium channel.

The EAG (ether-à-go-go) family of voltage-gated K+ channels are important regulators of neuronal and cardiac action potential firing (excitability) and have major roles in human diseases such as epilepsy, schizophrenia, cancer, and sudden cardiac death. A defining feature of EAG (Kv10-12) channels is a highly conserved domain on the N terminus, known as the eag domain, consisting of a Per-ARNT-Sim (PAS) domain capped by a short sequence containing an amphipathic helix (Cap domain). The PAS and Cap domains are both vital for the normal function of EAG channels. Using heme-affinity pulldown assays and proteomics of lysates from primary cortical neurons, we identified that an EAG channel, hERG3 (Kv11.3), binds to heme. In whole-cell electrophysiology experiments, we identified that heme inhibits hERG3 channel activity. In addition, we expressed the Cap and PAS domain of hERG3 in Escherichia coli and, using spectroscopy and kinetics, identified the PAS domain as the location for heme binding. The results identify heme as a regulator of hERG3 channel activity. These observations are discussed in the context of the emerging role for heme as a regulator of ion channel activity in cells.

Plasmodium Condensin Core Subunits SMC2/SMC4 Mediate Atypical Mitosis and Are Essential for Parasite Proliferation and Transmission.

Condensin is a multi-subunit protein complex regulating chromosome condensation and segregation during cell division. In Plasmodium spp., the causative agent of malaria, cell division is atypical and the role of condensin is unclear. Here we examine the role of SMC2 and SMC4, the core subunits of condensin, during endomitosis in schizogony and endoreduplication in male gametogenesis. During early schizogony, SMC2/SMC4 localize to a distinct focus, identified as the centromeres by NDC80 fluorescence and chromatin immunoprecipitation sequencing (ChIP-seq) analyses, but do not form condensin I or II complexes. In mature schizonts and during male gametogenesis, there is a diffuse SMC2/SMC4 distribution on chromosomes and in the nucleus, and both condensin I and condensin II complexes form at these stages. Knockdown of smc2 and smc4 gene expression reveals essential roles in parasite proliferation and transmission. The condensin core subunits (SMC2/SMC4) form different complexes and may have distinct functions at various stages of the parasite life cycle.

Protein kinase B controls Mycobacterium tuberculosis growth via phosphorylation of the transcriptional regulator Lsr2 at threonine 112.

Mycobacterium tuberculosis (Mtb) is able to persist in the body through months of multi-drug therapy. Mycobacteria possess a wide range of regulatory proteins, including the protein kinase B (PknB) which controls peptidoglycan biosynthesis during growth. Here, we observed that depletion of PknB resulted in specific transcriptional changes that are likely caused by reduced phosphorylation of the H-NS-like regulator Lsr2 at threonine 112. The activity of PknB towards this phosphosite was confirmed with purified proteins, and this site was required for adaptation of Mtb to hypoxic conditions, and growth on solid media. Like H-NS, Lsr2 binds DNA in sequence-dependent and non-specific modes. PknB phosphorylation of Lsr2 reduced DNA binding, measured by fluorescence anisotropy and electrophoretic mobility shift assays, and our NMR structure of phosphomimetic T112D Lsr2 suggests that this may be due to increased dynamics of the DNA-binding domain. Conversely, the phosphoablative T112A Lsr2 had increased binding to certain DNA sites in ChIP-sequencing, and Mtb containing this variant showed transcriptional changes that correspond with the change in DNA binding. In summary, PknB controls Mtb growth and adaptations to the changing host environment by phosphorylating the global transcriptional regulator Lsr2.

Molecular insights into an ancient form of Paget's disease of bone.

Paget's disease of bone (PDB) is a chronic skeletal disorder that can affect one or several bones in individuals older than 55 y of age. PDB-like changes have been reported in archaeological remains as old as Roman, although accurate diagnosis and natural history of the disease is lacking. Six skeletons from a collection of 130 excavated at Norton Priory in the North West of England, which dates to medieval times, show atypical and extensive pathological changes resembling contemporary PDB affecting as many as 75% of individual skeletons. Disease prevalence in the remaining collection is high, at least 16% of adults, with age at death estimations as low as 35 y. Despite these atypical features, paleoproteomic analysis identified sequestosome 1 (SQSTM1) or p62, a protein central to the pathological milieu of PDB, as one of the few noncollagenous human sequences preserved in skeletal samples. Targeted proteomic analysis detected >60% of the ancient p62 primary sequence, with Western blotting indicating p62 abnormalities, including in dentition. Direct sequencing of ancient DNA excluded contemporary PDB-associated SQSTM1 mutations. Our observations indicate that the ancient p62 protein is likely modified within its C-terminal ubiquitin-associated domain. Ancient miRNAs were remarkably preserved in an osteosarcoma from a skeleton with extensive disease, with miR-16 expression consistent with that reported in contemporary PDB-associated bone tumors. Our work displays the use of proteomics to inform diagnosis of ancient diseases such as atypical PDB, which has unusual features presumably potentiated by yet-unidentified environmental or genetic factors.

Two Faces of CwlM, an Essential PknB Substrate, in Mycobacterium tuberculosis.

Tuberculosis claims >1 million lives annually, and its causative agent Mycobacterium tuberculosis is a highly successful pathogen. Protein kinase B (PknB) is reported to be critical for mycobacterial growth. Here, we demonstrate that PknB-depleted M. tuberculosis can replicate normally and can synthesize peptidoglycan in an osmoprotective medium. Comparative phosphoproteomics of PknB-producing and PknB-depleted mycobacteria identify CwlM, an essential regulator of peptidoglycan synthesis, as a major PknB substrate. Our complementation studies of a cwlM mutant of M. tuberculosis support CwlM phosphorylation as a likely molecular basis for PknB being essential for mycobacterial growth. We demonstrate that growing mycobacteria produce two forms of CwlM: a non-phosphorylated membrane-associated form and a PknB-phosphorylated cytoplasmic form. Furthermore, we show that the partner proteins for the phosphorylated and non-phosphorylated forms of CwlM are FhaA, a fork head-associated domain protein, and MurJ, a proposed lipid II flippase, respectively. From our results, we propose a model in which CwlM potentially regulates both the biosynthesis of peptidoglycan precursors and their transport across the cytoplasmic membrane.

PknG senses amino acid availability to control metabolism and virulence of Mycobacterium tuberculosis.

Sensing and response to changes in nutrient availability are essential for the lifestyle of environmental and pathogenic bacteria. Serine/threonine protein kinase G (PknG) is required for virulence of the human pathogen Mycobacterium tuberculosis, and its putative substrate GarA regulates the tricarboxylic acid cycle in M. tuberculosis and other Actinobacteria by protein-protein binding. We sought to understand the stimuli that lead to phosphorylation of GarA, and the roles of this regulatory system in pathogenic and non-pathogenic bacteria. We discovered that M. tuberculosis lacking garA was severely attenuated in mice and macrophages and furthermore that GarA lacking phosphorylation sites failed to restore the growth of garA deficient M. tuberculosis in macrophages. Additionally we examined the impact of genetic disruption of pknG or garA upon protein phosphorylation, nutrient utilization and the intracellular metabolome. We found that phosphorylation of GarA requires PknG and depends on nutrient availability, with glutamate and aspartate being the main stimuli. Disruption of pknG or garA caused opposing effects on metabolism: a defect in glutamate catabolism or depletion of intracellular glutamate, respectively. Strikingly, disruption of the phosphorylation sites of GarA was sufficient to recapitulate defects caused by pknG deletion. The results suggest that GarA is a cellular target of PknG and the metabolomics data demonstrate that the function of this signaling system is in metabolic regulation. This function in amino acid homeostasis is conserved amongst the Actinobacteria and provides an example of the close relationship between metabolism and virulence.

A new tool for the chemical genetic investigation of the Plasmodium falciparum Pfnek-2 NIMA-related kinase.

BACKGROUND: Examining essential biochemical pathways in Plasmodium falciparum presents serious challenges, as standard molecular techniques such as siRNA cannot be employed in this organism, and generating gene knock-outs of essential proteins requires specialized conditional approaches. In the study of protein kinases, pharmacological inhibition presents a feasible alternative option. However, as in mammalian systems, inhibitors often lack the desired selectivity. Described here is a chemical genetic approach to selectively inhibit Pfnek-2 in P. falciparum, a member of the NIMA-related kinase family that is essential for completion of the sexual development of the parasite. RESULTS: Introduction of a valine to cysteine mutation at position 24 in the glycine rich loop of Pfnek-2 does not affect kinase activity but confers sensitivity to the protein kinase inhibitor 4-(6-ethynyl-9H-purin-2-ylamino) benzene sulfonamide (NCL-00016066). Using a combination of in vitro kinase assays and mass spectrometry, (including phosphoproteomics) the study shows that this compound acts as an irreversible inhibitor to the mutant Pfnek2 likely through a covalent link with the introduced cysteine residue. In particular, this was shown by analysis of total protein mass using mass spectrometry which showed a shift in molecular weight of the mutant kinase in the presence of the inhibitor to be precisely equivalent to the molecular weight of NCL-00016066. A similar molecular weight shift was not observed in the wild type kinase. Importantly, this inhibitor has little activity towards the wild type Pfnek-2 and, therefore, has all the properties of an effective chemical genetic tool that could be employed to determine the cellular targets for Pfnek-2. CONCLUSIONS: Allelic replacement of wild-type Pfnek-2 with the mutated kinase will allow for targeted inhibition of Pfnek-2 with NCL-00016066 and hence pave the way for comparative studies aimed at understanding the biological role and transmission-blocking potential of Pfnek-2.

The structure of the core NuRD repression complex provides insights into its interaction with chromatin.

The NuRD complex is a multi-protein transcriptional corepressor that couples histone deacetylase and ATP-dependent chromatin remodelling activities. The complex regulates the higher-order structure of chromatin, and has important roles in the regulation of gene expression, DNA damage repair and cell differentiation. HDACs 1 and 2 are recruited by the MTA1 corepressor to form the catalytic core of the complex. The histone chaperone protein RBBP4, has previously been shown to bind to the carboxy-terminal tail of MTA1. We show that MTA1 recruits a second copy of RBBP4. The crystal structure reveals an extensive interface between MTA1 and RBBP4. An EM structure, supported by SAXS and crosslinking, reveals the architecture of the dimeric HDAC1:MTA1:RBBP4 assembly which forms the core of the NuRD complex. We find evidence that in this complex RBBP4 mediates interaction with histone H3 tails, but not histone H4, suggesting a mechanism for recruitment of the NuRD complex to chromatin.

Malaria protein kinase CK2 (PfCK2) shows novel mechanisms of regulation.

Casein kinase 2 (protein kinase CK2) is a conserved eukaryotic serine/theronine kinase with multiple substrates and roles in the regulation of cellular processes such as cellular stress, cell proliferation and apoptosis. Here we report a detailed analysis of the Plasmodium falciparum CK2, PfCK2, demonstrating that this kinase, like the mammalian orthologue, is a dual specificity kinase able to phosphorylate at both serine and tyrosine. However, unlike the human orthologue that is auto-phosphorylated on tyrosine within the activation loop, PfCK2 shows no activation loop auto-phosphorylation but rather is auto-phosphorylated at threonine 63 within subdomain I. Phosphorylation at this site in PfCK2 is shown here to regulate the intrinsic kinase activity of PfCK2. Furthermore, we generate an homology model of PfCK2 in complex with the known selective protein kinase CK2 inhibitor, quinalizarin, and in so doing identify key co-ordinating residues in the ATP binding pocket that could aid in designing selective inhibitors to PfCK2.

DNA damage-induced ubiquitylation of proteasome controls its proteolytic activity.

Stability of proteins is largely controlled by post-translational covalent modifications. Among those, ubiquitylation plays a central role as it marks the proteins for proteasome-dependent degradation. Proteolytic activities of proteasomes are critical for execution of various cellular processes, including DNA damage signaling and repair. However, very little is known about the regulation of proteasomal activity in cells during genotoxic stress. Here we investigated post-translational modifications of the 20S proteasomal subunits upon DNA damage induced by doxorubicin. Using mass-spectrometry, we found novel sites of phosphorylation and ubiquitylation in multiple proteasome subunits upon doxorubicin treatment. Ectopic co-expression of proteasome subunits and tagged ubiquitin confirmed the presence of ubiquitylated forms of PSMA5, PSMA1, PSMA3 and PSMB5 in cells. Moreover, we demonstrated that ubiquitylation in vitro inhibited chymotrypsin-like and caspase-like activities of proteasomes. In vivo, doxorubicin increased the activity of proteasomes, paralleling with attenuation of the overall level of proteasome ubiquitylation. Collectively, our results suggest a novel mechanism whereby the proteolytic activities of proteasomes are dynamically regulated by ubiquitylation upon DNA damage.

Mass spectrometry analysis of human P2X1 receptors; insight into phosphorylation, modelling and conformational changes.

Recombinant FlagHis(6) tagged Human P2X1 receptors expressed in HEK293 cells were purified, digested with trypsin and analysed by mass spectroscopy (96% coverage following de-glycosylation and reduction). The receptor was basally phosphorylated at residues S387, S388 and T389 in the carboxyl terminus, a triple alanine mutant of these residues had a modest ~ 25% increase in current amplitude and recovery from desensitization. Chemical modification showed that intracellular lysine residues close to the transmembrane domains and the membrane stabilization motif are accessible to the aqueous environment. The membrane-impermeant cross-linking reagent 3,3'-Dithiobis (sulfosuccinimidylpropionate) (DTSSP) reduced agonist binding and P2X1 receptor currents by > 90%, and modified lysine residues were identified by mass spectroscopy. Mutation to remove reactive lysine residues around the ATP-binding pocket had no effect on inhibtion of agonist evoked currents following DTSSP. However, agonist evoked currents were ~ 10-fold higher than for wild type following DTSSP treatment for mutants K199R, K221R and K199R-K221R. These mutations remove reactive residues distant from the agonist binding pocket that are close enough to cross-link adjacent subunits. These results suggest that conformational change in the P2X1 receptor is required for co-ordination of ATP action.

A specific binding site for the prorenin propart peptide Arg10-Arg20 does not occur on human endothelial cells.

INTRODUCTION: We looked for novel binding sites for the human prorenin 'decoy peptide' sometimes called 'handle region peptide' on human endothelial cells. METHOD: The biotinylated peptide biotin-Acp-RIFLKRMPSIR (B-PR), an unlabelled peptide PR1 (RIFLKRMPSIR) and a scrambled peptide scPR1 (SRRMIFPIKLR) were synthesized. B-PR was added to human umbilical cord endothelial cells (HUVECs) maintained in serum-free medium, with or without excess unlabelled peptide or 'scrambled' peptide as blocker. Biotin-labelled HUVEC proteins were extracted, the amount of bound tracer was measured, and the identity of the binding proteins was analysed by sodium-dodecyl-sulphate polyacrylamide gel electrophoresis (SDS-PAGE) and by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). RESULTS: Biotinylated peptide bound to the HUVEC proteins with a major labelled band at 68,600 ± 1503 kDa (mean ± SEM, n = 5 runs). Unlabelled peptide and scrambled peptide equally displaced the labelled peptide, indicating that the binding was non-specific for amino acid sequence. LC-MS/MS showed that binding was mainly to cytoskeletal proteins. CONCLUSION: The binding of the human prorenin peptide R¹°IFLKRMPSIR²° to HUVEC proteins is not specific for amino acid sequence and probably involves a general peptide/protein uptake mechanism. We could not detect a specific prorenin propart binding site in these cells.

Conformationally constrained mimetics of laminin peptide YIGSR as precursors for antimetastatic disintegrins.

Conformationally constrained mimetics of the laminin cell-adhesion site, YIGSR, are described. The site is the natural antagonist of the integrin-associated laminin receptor 1 (LAMR1) known to mediate metastatic tumor adhesion. The attachment of selected metastatic cell lines toward the constrained antagonists has been assessed. Observed differential responses prompted by folding preferences of the mimetics revealed stronger attachment activities for turnlike structures. The results permit the conformational design of antimetastatic disintegrins.

Cloning and sequencing of two Ceriporiopsis subvermispora bicupin oxalate oxidase allelic isoforms: implications for the reaction specificity of oxalate oxidases and decarboxylases.

Oxalate oxidase is thought to be involved in the production of hydrogen peroxide for lignin degradation by the dikaryotic white rot fungus Ceriporiopsis subvermispora. This enzyme was purified, and after digestion with trypsin, peptide fragments of the enzyme were sequenced using quadrupole time-of-flight mass spectrometry. Starting with degenerate primers based on the peptide sequences, two genes encoding isoforms of the enzyme were cloned, sequenced, and shown to be allelic. Both genes contained 14 introns. The sequences of the isoforms revealed that they were both bicupins that unexpectedly shared the greatest similarity to microbial bicupin oxalate decarboxylases rather than monocupin plant oxalate oxidases (also known as germins). We have shown that both fungal isoforms, one of which was heterologously expressed in Escherichia coli, are indeed oxalate oxidases that possess < or =0.2% oxalate decarboxylase activity and that the organism is capable of rapidly degrading exogenously supplied oxalate. They are therefore the first bicupin oxalate oxidases to have been described. Heterologous expression of active enzyme was dependent on the addition of manganese salts to the growth medium. Molecular modeling provides new and independent evidence for the identity of the catalytic site and the key amino acid involved in defining the reaction specificities of oxalate oxidases and oxalate decarboxylases.

Stability of the major allergen Brazil nut 2S albumin (Ber e 1) to physiologically relevant in vitro gastrointestinal digestion.

The major 2S albumin allergen from Brazil nuts, Ber e 1, was subjected to gastrointestinal digestion using a physiologically relevant in vitro model system either before or after heating (100 degrees C for 20 min). Whilst the albumin was cleaved into peptides, these were held together in a much larger structure even when digested by using a simulated phase 1 (gastric) followed by a phase 2 (duodenal) digestion system. Neither prior heating of Ber e 1 nor the presence of the physiological surfactant phosphatidylcholine affected the pattern of proteolysis. After 2 h of gastric digestion, approximately 25% of the allergen remained intact, approximately 50% corresponded to a large fragment of M(r) 6400, and the remainder comprised smaller peptides. During duodenal digestion, residual intact 2S albumin disappeared quickly, but a modified form of the 'large fragment' remained, even after 2 h of digestion, with a mass of approximately 5000 Da. The 'large fragment' comprised several smaller peptides that were identified, by using different MS techniques, as deriving from the large subunit. In particular, sequences corresponding to the hypervariable region (Q37-M47) and to another peptide (P42-P69), spanning the main immunoglobulin E epitope region of 2S albumin allergens, were found to be largely intact following phase 1 (gastric) digestion. They also contained previously identified putative T-cell epitopes. These findings indicate that the characteristic conserved skeleton of cysteine residues of 2S albumin family and, particularly, the intrachain disulphide bond pattern of the large subunit, play a critical role in holding the core protein structure together even after extensive proteolysis, and the resulting structures still contain potentially active B- and T-cell epitopes.