Sequence-specific retention calculator. Algorithm for peptide retention prediction in ion-pair RP-HPLC: application to 300- and 100-A pore size C18 sorbents

Abstract

Continued development of a new sequence-specific algorithm for peptide retention prediction in RP HPLC is reported. Our discovery of the large effect on the apparent hydrophobicity of N-terminal amino acids produced by the ion-pairing retention mechanism has led to the development of sequence-specific retention calculator (SSRCalc) algorithms. These were optimized for a set of approximately 2000 tryptic peptides confidently identified by off-line microHPLC-MALDI MS (MS/MS) (300-A pore size C18 sorbent, linear water/acetonitrile gradient, and trifluoroacetic acid as ion-pairing modifier). The latest version of the algorithm takes into account amino acid composition, position of the amino acid residues (N- and C-terminal), peptide length, overall hydrophobicity, pI, nearest-neighbor effect of charged side chains (K, R, H), and propensity to form helical structures. A correlation with R2 approximately 0.98 was obtained for the 2000-peptide optimization set. A flexible structure for the SSRC programming code allows easy adaptation to different chromatographic conditions. This was demonstrated by adapting the algorithm (approximately 0.98 R2 value) for a set of approximately 2500 peptides separated on a 100-A pore size C18 column. The SSRCalc algorithm has also been extensively tested for a number of real samples, providing solid support for protein identification and characterization; correlations in the range of 0.95-0.97 R2 value have normally been observed.