Биопрепараты: Профилактика, диагностика, лечение (Feb 2018)
Optimization of mathematical processing of calibration curves when determining molecular mass of biologicals by SDS-polyacrylamide gel electrophoresis
Abstract
Identification and purity testing of biologicals as well as determination of their molecular mass are performed on the basis of comparison of test and reference samples by various physico-chemical methods, including Sodium Dodecyl Sulphate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) under reducing or non-reducing conditions. In order to assess the molecular weight of the test sample and reference sample components a calibration curve is drawn using protein markers. The article describes 4 experiments in which gel electrophoresis was used for separation of the most widely used sets of molecular weight markers produced by various manufacturers (Amersham™ LMW Calibration Kit for SDS Electrophoresis (Am), SDS PAGE Molecular Weight Standards, low range (BR), Unstained Protein Molecular Weight Marker (Th), BenchMark™ Protein Ladder (BM), Mark12™ Unstained Standard(M12) (total number of proteins - 45)). The aim of the study was to optimize mathematical processing of calibration curves generated for the above-mentioned sets of markers and to assess their interchangeability. SDS PAGE was performed under reducing conditions using Tris-Glycine gels. The real calculated ( M ) and nominal ( M0) values (indicated in the set of markers instructions for use) of the protein markers molecular mass (MM) were compared using the MM deviation expressed in terms of a molecular mass unit (hereinafter - relative deviation). The estimated average MM relative deviation for sets of markers ranging from 10 to 100 kDa (Am, BR, Th sets), which was calculated using linear regression equations, was equal to about 10 % (8.3 - 12.0 %). In the case of the extended MM range of 10 - 220 kDa (BM, M12 sets) the average relative deviation was almost twice as high (16.5 - 20.3 %). A third-order polynomial regression equation was used to optimize mathematical processing methods used during calculation of the calibration function. This made it possible to reduce the MM relative deviation down to 2.3 - 3.5 % when using sets ranging from 10 to 100 kDa, and to 3.7 - 4.0 % when using the extended range sets, which is important for assessing similarity of different products.
