Utility of retention prediction model for investigation of peptide separation selectivity in reversed-phase liquid chromatography: impact of concentration of trifluoroacetic acid, column temperature, gradient slope and type of stationary phase.

Peptide separation selectivity in reversed-phase liquid chromatography was investigated using a training set of 165 peptides with a total of 1698 amino acid residues. Gradient separation was performed at selected chromatographic conditions, varying column temperature, gradient slope, ion-pairing reagent concentration, and type of stationary phase. The retention times for each set of experiments were utilized to calculate the amino acid retention coefficients using a published prediction model (Rapid Commun. Mass Spectrom. 2007, 21, 2813-2821). The calculated retention coefficients reflect the contribution of each type of amino acid residue to peptide retention at a given chromatographic condition. For example, the concentration of ion-pairing reagent (trifluoroacetic acid) had the strongest impact on the retention of peptides containing an increasing number of basic (charged) amino acids, such as arginine, lysine, and histidine, while the retention coefficients of other amino acids are minimally affected. Increasing the separation temperature resulted in a moderate decrease of amino acid retention coefficients with the exception of isoleucine, leucine, valine, and proline. This finding suggests that these residues enhance peptide retention at elevated temperature. Amino acid residue retention coefficients were also helpful to understand the impact of sorbent pore size (130 vs 300 A) on peptide retention selectivity. In addition we investigated the selectivity differences of various reversed-phase chromatographic sorbents. The trends suggested by calculated retention coefficients were confirmed using a set of synthetic peptides with specifically designed sequences.