Search

Comparative modeling and time-course hydrolysis experiments have been applied to investigate two enzyme-inhibitor complexes formed between the wheat subtilisin-chymotrypsin inhibitor (WSCI) and two susceptible proteinases. WSCI represents the first case of a wheat protein inhibitor active against animal chymotrypsins and bacterial subtilisins. The model was created using as template structure that of the CI-2A inhibitor from barley (PDB code: 2CI2), which shares 87% sequence identity with WSCI. Under these conditions of high similarity, the comparative modeling approach can be successfully applied. We predicted the WSCI 3D model and used it to investigate enzyme-inhibitor complex systems. Experimental observations indicated that chymotrypsin, but not subtilisin, in addition to cleavage at the primary reactive site Met48-Glu49, is able to hydrolyze a second peptide bond between Phe58 and Val59. Here, we report on cleavage of the peptide bond at the inhibitor's reactive site (Met48-Glu49) determined using time-course hydrolysis experiments; the same event was investigated for both subtilisin/WSCI and chymotrypsin/WSCI complexes using molecular dynamics simulations. The molecular details of the initial inhibitor-enzyme interactions, as well as of the changes observed during the simulations, allow us to speculate on the different fates of the two WSCI-proteinase complexes.

A site-directed mutagenesis strategy was employed to obtain four mutants of wheat subtilisin/chymotrypsin inhibitor (WSCI), with the aim to produce inactive forms of this protein. The mutants were expressed in Escherichia coli as fusion proteins and, after the tag removal, were purified to homogeneity. Three mutants, containing a single mutation at the sequence positions 49 or 50, were named E49S, E49P and Y50G, respectively. These mutants exhibited anti-subtilisin activities comparable to that of the wild type protein; instead, anti-chymotrypsin activity was detectable only for the mutant E49S. A fourth mutant (M48P-E49G), containing a double amino acid substitution at the inhibitor reactive site (P1-P1′), was inactive against both subtilisin and chymotrypsin. In order to investigate the interactions between the putative susceptible enzymes and the mutated forms of WSCI, we performed time-course hydrolysis experiments by incubating samples of the mutants with subtilisin-agarose and chymotrypsin-agarose, respectively. These experiments yielded information on the E/I complex formation, as well as on the timing of the cleavage pattern of some of these mutants. Molecular modeling studies were carried out with the 3D models of the mutants and of their putative complexes with subtilisin and chymotrypsin. In terms of inter- and intra-chain H-bond networks, the observations made for each theoretical E/I complex were found to be fully coherent with experimental data (kinetic and time-course hydrolysis) and supplied specific modalities of interaction of each mutant with the enzyme counterpart. © 2009 Elsevier Masson SAS. All rights reserved.

Thermal treatment of milk leads to non-enzymatic glycosylation of proteins through Maillard reaction. Free NH2 groups of basic amino acids react with the reducing carbonyl group of lactose forming the so-called Amadori products. Electrospray mass spectrometry analysis shows that β-lactoglobulin (β-LG), the major whey protein, undergoes lactosylation under industrial thermal treatment. In order to investigate the specificity of reactive sites for lactose binding the analysis of trypsin hydrolysates of β-LG isolated from different industrial milks was performed. Results demonstrate that Lys-100 is a preferential lactosylation site of β-LG during industrial milk treatment. These results were confirmed by an analysis of the three-dimensional model of the protein which showed that Lys-100 had the highest solvent accessibility and proximity to another amino group making Lys-100 the best candidate to lactosylation. Lys-47, previously identified by other authors, showed a good proximity to another Lys residue, but an intermediate level of exposition to solvent. Copyright (C) 1998 Elsevier Science B.V.

Homology modelling of the human eIF-5A protein has been performed by using a multiple predictions strategy. As the sequence identity between the target and the template proteins is nearly 30%, which is lower than the commonly used threshold to apply with confidence the homology modelling method, we developed a specific predictive scheme by combining different sequence analyses and predictions, as well as model validation by comparison to structural experimental information. The target sequence has been used to find homologues within sequence databases and a multiple alignment has been created. Secondary structure for each single protein has been predicted and compared on the basis of the multiple sequence alignment, in order to evaluate and adjust carefully any gap. Therefore, comparative modelling has been applied to create the model of the protein on the basis of the optimized sequence alignment. The quality of the model has been checked by computational methods and the structural features have been compared to experimental information, giving us a good validation of the reliability of the model and its correspondence to the protein structure in solution. Last, the model was deposited in the Protein Data Bank to be accessible for studies on the structure-function relationships of the human eIF-5A.