Nonspecific adsorption evaluation and general minimization strategy in peptide analysis based on ultra-performance liquid chromatography-mass spectrometry

Proteomics technology is being increasingly used in the development of novel therapeutic peptides and protein drugs, and also in the intensive search for clinical biomacromolecule diagnostic biomarkers. Ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) is a rapid method to analyze peptides and proteins in low abundance. However, the nonspecific adsorption properties of peptides may lead to the loss or interference of the analytes throughout the analytical process. This unique nonspecific adsorption property is the main reason for the false negative and false positive results obtained through quantification, as well as for the poor precision, accuracy, linear range, and sensitivity, all of which impose significant challenges in the development of analytical methods. Accordingly, a general strategy was established to evaluate and reduce the negative impact of the nonspecific adsorption of peptides on UPLC-MS analysis. In this study, bovine serum albumin (BSA) was used as a model protein to explore the correlation between the physicochemical properties of 50 peptides obtained by the enzymatic digestion of BSA, as well as the degree of nonspecific adsorption. First, these peptides were classified into four categories according to their response and the degree of adsorption in the pretreatment containers and LC system. Next, the factors influencing the adsorption of 12 Class Ⅱ peptides, which were highly responsive and susceptible to adsorption, were systematically studied in terms of several aspects, including: (1) time-dependent adsorption on centrifuge tubes of three kinds (Protein-LoBind polypropylene tube and two types of polypropylene tubes); (2) time-dependent adsorption on sample vials of three kinds (Protein-LoBind polypropylene vial, polypropylene vial, and glass vial); (3) carryovers on chromatographic columns with six different stationary phases (Polar C18, Cortecs C18+, PFP, BEH C18, CSH C18, and BEH C8); (4) carryovers at different chromatographic gradients (2%B-30%B, 2%B-40%B, 2%B-50%B, and 2%B-60%B within 3 min), flow rates (0.2, 0.3, and 0.4 mL/min), and column temperatures (30, 40, 50, and 60 ℃); and (5) carryovers using different washing needle solutions (0.2% formic acid in 10% acetonitrile and 0.2% formic acid in 90% acetonitrile). The results showed that parameters such as the HPLC index and amino acid length of peptides were significantly correlated with their degree of adsorption (p<0.05), However, the above parameters can only explain the adsorption degree of 30% of the peptides. The use of the modified polypropylene material resulted in higher recovery (recovery rate>80% within 24 h) of the peptide solution during storage or pretreatment. During protein/peptide pretreatment and storage, good overall recoveries (recovery rate>80% within 24 h) were obtained using centrifuge tubes and sample vials made of the modified polypropylene material. Analysis and optimization of the LC conditions revealed that the use of the C8 chromatographic column, a high flow rate (0.4 mL/min), slow gradient (2%B-50%B within 3 min), and strong needle solution (0.2% formic acid in 90% acetonitrile) could minimize the carryover. However, the effect of the column temperature on the carryover varied considerably from peptide to peptide, and hence, requires further analysis for specific peptides. The combined optimization of the above experimental conditions resulted in minimal (approximately 1/150) or no adsorption of the Class Ⅱ peptides that were susceptible to adsorption in the analytical process. In this study, a workflow was designed to standardize the procedures for evaluating and reducing peptide adsorption. Detailed data were collected to elucidate the key risk factors and corresponding general mechanism of nonspecific adsorption throughout the analysis. Thus, this study serves as a reference for the development of analytical methods for peptides and proteins with different physicochemical properties. In future work, the risk factors should be assessed during the development and validation of protein-based macromolecular analysis methods. In conclusion, it is important to implement adequate and appropriate measures to ensure risk elimination or minimization.