Stereospecific capillary electrophoresis assays for methionine sulfoxide reductases
Met, either bound in proteins or in its free form is easily oxidized to Met(O) by reactive oxygen species. Met(O)-containing proteins accumulate during ageing and may play a role in degenerative diseases. Because of the chirality of the sulfoxide moiety, Met(O) exists as a pair of diastereomers Met-S-(O) and Met-R-(O). For the reduction of the diastereomers, stereospecific methionine sulfoxide reductase (Msr) enzymes exist. MsrA reduces free and protein-bound Met-S-(O), while MsrB reduces protein-bound Met-R-(O) with little affinity for free Met-R-(O). The latter is reduced by fRMsr. The aim of the present study was the development of stereospecific CE assays for Msr enzymes in order to study the stereospecificity of the enzymes. The substrates included Met(O), Met(O) derivatives as well as Met(O)-containing peptides. Upon establishment of the separation conditions, the methods were validated and subsequently applied for assaying recombinant human and fungal MsrA and MsrB enzymes. An off-line assay was developed using free Met(O) as substrate. The separation of the Met(O) diastereomers, the product Met as well as β-alanine as internal standard was achieved upon derivatization of the analytes with dabsyl chloride by a MEKC method in a SMIL-coated capillary. The capillary coating consisted of a first layer of polybrene and a second layer of dextran sulfate providing a stable strong cathodic EOF and, consequently, highly repeatable analyte migration times. The assay was subsequently applied to screen the stereospecific activity of recombinant human MsrA and fungal fRMsr enzymes as well as to determine the Michaelis-Menten kinetics. An advantage of the present assay over existing assays is the fact that free Met(O) can be used as the natural substrate. In the assay, DTT was used as reducing agent for Msr recycling instead of using the coupled reaction involving Trx and Trx reductase so that no additional enzymes were required. An in-capillary electrophoretically mediated microanalysis (EMMA) assay was established using Fmoc-Met(O) as substrate. The separation of diastereomers of Fmoc-Met(O), the product Fmoc-Met and the internal standard Fmoc-ß-alanine was also achieved in a SMIl-coated capillary by a MEKC method. The partial filling mode was applied because the BGE contained SDS, which would lead to the denaturation of the enzyme. An injection sequence of incubation buffer-enzyme-substrate-enzyme-incubation buffer was selected. The assay was optimized with regard to the mixing time with a mixing voltage and subsequently applied for the analysis of stereospecificity of human MsrA and MsrB2. The Michaelis-Menten constant, Km, and the maximum velocity, Vmax, were determined. Essentially identical data were determined by the EMMA mode compared to off-line incubations. Compared to the off-line assay, the EMMA assay was fully automated and required smaller amounts of enzyme and substrates, thus, reducing the overall cost of the assay. For the development of an assay using peptide substrates, the C-terminally dinitrophenyl-labeled, N-acetylated pentapeptide KIFM(O)K was chosen. The separation of the KIFM(O)K diastereomers and the reduced peptide KIFMK was achieved in a BGE containing sulfated β-CD and 15-crown-5 as buffer additives. The system was optimized using experimental design with regard to the buffer pH, ionic strength, sulfated β-CD and 15-crown-5, as well as capillary temperature and separation voltage. A fractional factorial response IV design was employed for the identification of the significant experimental factors and a five-level circumscribed central composite design for the final method optimization. The assay was successfully applied for the characterization of the stereospecificity of recombinant human MsrA and MsrB2 including the determination of the Michaelis-Menten kinetic data. Using this peptide substrate, lower Km values but significantly higher kcat constants were observed as compared to literature data reported for the substrates dabsyl-Met(O) and Fmoc-Met(O). Thus, the present pentapeptide-based assay may represent Msr activities towards protein-bound Met(O) in a better way as compared to simple amino acid derivative-based assays. In order to better understand previous results that MsrA enzyme had a preference for peptide substrates with positively charged residues flanking Met(O), a group of peptide substrates was employed in combination with wild-type fungal MsrA and three mutants. The present data were in general agreement with the previous non-stereospecific assay except that the DE mutant (Glu99, Glu134) was not found to be more active than the wild-type enzyme for the reduction of KDM(O)DK. Interestingly, a reversed substrate preference was observed between the mutants DN (Glu99, Asn134) and EQ (Gln99, Asp134) reducing KDM(O)NK and KNM(O)DK. The asymmetric negative charges in the active center may explain this behavior. Molecular modelling was performed to rationalize the specificity and indicated that the conserved residue Glu99 in the active site of MsrA was buried in the Met-S-(O) binding groove, which might contribute to the right placing of Met-S-(O) and, consequently, to the catalytic activity of MsrA. Finally, the dual selector system composing of a CD and a crown ether for the separation of the Met(O) peptide diastereomers was studied systematically using a series of N-acetylated Met(O) pentapeptides with a Dnp label at the C-terminus. Depending on the amino acid sequence and the applied CD, the addition of crown ethers, especially of the Krpytofix® diaza-crown ethers, resulted in a significant improvement of the resolution of the diastereomers of peptides containing basic amino acids. Charged CD derivatives such as carboxymethyl-ß-CD and sulfated ß-CD were superior compared to native ß-CD. In contrast, the diastereomer separation of Met(O) peptides containing uncharged amino acids was found superior in a MEKC system in the presence of CDs.