J. Sep. Sci. 2012, 35, 1771–1778 1771 Eugene Moskovets 1 Anton A. Goloborodko 2 Alexander V. Gorshkov 3 Mikhail V. Gorshkov 2 1 MassTech Inc., Columbia, Maryland, USA 2 Institute for Energy Problems of Chemical Physics, Moscow, Russia 3 N. N. Semenov’s Institute of Chemical Physics, Moscow, Russia Received September 9, 2011 Revised January 13, 2012 Accepted January 13, 2012 Research Article Limitation of predictive 2-D liquid chromatography in reducing the database search space in shotgun proteomics: In silico studies A two-dimensional (2-D) liquid chromatography (LC) separation of complex peptide mix- tures that combines a normal phase utilizing hydrophilic interactions and a reversed phase offers reportedly the highest level of 2-D LC orthogonality by providing an even spread of peptides across multiple LC fractions. Matching experimental peptide retention times to those predicted by empirical models describing chromatographic separation in each LC dimension leads to a significant reduction in a database search space. In this work, we calculated the retention times of tryptic peptides separated in the C18 reversed phase at different separation conditions (pH 2 and pH 10) and in TSK gel Amide-80 normal phase. We show that retention times calculated for different 2-D LC separation schemes utilizing these phases start to correlate once the mass range of peptides under analysis becomes pro- gressively narrow. This effect is explained by high degree of correlation between retention coefficients in the considered phases. Keywords: 2-D LC / Database search / HILIC / Proteomics / Retention time prediction DOI 10.1002/jssc.201100798 1 Introduction Bottom-up proteomics that utilizes liquid chromatography (LC) combined with tandem mass spectrometry (MS) is a powerful method of identification of proteins in biological samples [1]. This approach applied to the analysis of cell cul- tures generates tens of thousands peptides of widely differ- ent abundance [2]. Detection of low-abundance peptides in a complex background of more abundant peptides eluting in the same LC fraction can be difficult due to ion suppression effect in the ionization sources [3, 4]. A fundamental way to reduce the ion suppression effect and to remove limitations on maximum speed of the informative MS/MS analysis in a given tandem instrument is to split liquid sample containing a complex peptide mixture into a large number of gradually eluting LC fractions, where each fraction contains a relatively small number of peptides of different sequence lengths and amino acid compositions. Two-dimensional (2-D) LC sep- Correspondence: Dr. Mikhail V. Gorshkov, Institute for Energy Problems of Chemical Physics, 38 Leninskiy Pr., Building 2, Moscow 119334, Russia E-mail: mike.gorshkov@gmail.com Fax: +7(499)-137-8258 Abbreviations: 2-D LC, two dimensional liquid chromatog- raphy; BioLCCC, Liquid Chromatography of Biomacro- molecules at Critical Conditions; SSRCalc, Sequence Specific Retention Calculator arations where two dimensions are based on “orthogonal” separation principles [5, 6] provide deep fractionation of com- plex peptide mixtures, drastically improving the success rate of MS-based peptide identification. Orthogonality metrics ap- plied to 2-D LC separations in proteomics have been broadly discussed in literature [7, 8]. A combination of hydrophilic interaction liquid chro- matography (HILIC) [9] and reversed-phase (RP) chromatog- raphy shows a high degree of combined peak capacity in the 2-D LC separation space [7, 10]. HILIC and ERLIC [11] (elec- trostatic repulsion hydrophilic interaction chromatography) become increasingly popular for separation of peptides with polar, glycosylated, and phosphorylated groups. In addition to the ability of 2-D LC to reduce the ana- lytical background in MS analysis, the ability to predict re- tention times of peptides separated in complementary chro- matographic systems is another valuable factor. Moreover, because retention time of a peptide depends on its amino acid sequence, the retention time becomes an independent peptide identifier. In MS/MS database peptide-sequencing approach, the first step in peptide identification is selection of those peptide candidates from a database, which molec- ular weights (MW) fit the experimentally measured MW of the peptide under analysis with some error margin deter- mined by mass accuracy of the MS instrument. By compar- ing experimental retention time of the peptide under analysis to the “theoretical” retention times calculated for the mass- selected peptide candidates, one can significantly diminish C  2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com