Prediction of Band Profiles of Mixtures of Bradykinin and Kallidin from Data Acquired by Competitive Frontal Analysis Dongmei Zhou, † Xiaoda Liu, † Krzysztof Kaczmarski, ‡ Attila Felinger, † and Georges Guiochon* ,† Department of Chemistry, The University of Tennessee, Knoxville, Tennessee 37996-1600 and Division of Chemical Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee, and Faculty of Chemistry, Rzeszo ´w University of Technology, 35-959 Rzeszo ´w, Poland The competitive adsorption isotherms of two closely related peptides, bradykinin and kallidin, were measured by frontal analysis on a Zorbax SB-C18 microbore column. An aqueous soluton at 20% acetonitrile (0.1% TFA) was used as the mobile phase. The competitive isotherm data were fitted to four different models: Langmuir, Bilangmuir, Langmuir-Freundlich, and Toth. These data fitted best to a Bilangmuir isotherm model. The influence of the pressure on the retention factors of the two peptides was found to be small and was not investigated in detail. The band profiles of large samples of the single components and of their mixtures were recorded. The overloaded profiles calculated using either the equilibrium-dispersive or POR model are in excellent agreement with the experimental profiles in all cases. Our results confirm that the competitive isotherm data derived from mixtures may suffice for a reasonably accurate prediction of the band profiles of all mixtures of the two components, provided their composition is close to 1/1. 1. Introduction Preparative liquid chromatography has become an important method for the separation and purification of peptides and proteins. The chromatographic separation of large biomolecules is usually a complex process, and most such separations are developed empirically. The cost of this approach increases rapidly with increasing sample amount and column size. The commercial success of biotechnology depends much on the development of reliable and efficient large scale purification processes (1, 2). When preparative chromatography is needed, computer-assisted optimization of protein and peptide separations can be a tool of great practical importance. Under nonlinear conditions, the measurement of ad- sorption isotherms is required for the calculation and prediction of experimental profiles. Much work has been reported on the equilibrium thermodynamics of amino acids (3, 4), small peptides (5), and proteins (6-8) in systems used in liquid chromatography. However, few of these works deal with the modeling and the prediction of the chromatographic separation of large biomolecules. Liu et al. (9-11) studied the modeling of preparative reversed-phase HPLC of different insulin variants and measured their isotherms by frontal analysis. Using these data, he could calculate the band profiles of three insulin variants under nonlinear conditions and found good agreement with experimental data. The general goal of this work is to illustrate the extension to peptides and proteins of the concepts and approaches that we have pioneered earlier for conven- tional low molecular weight compounds (12) and that were proven to be most useful to improve the economics of the purification of drug intermediates by preparative liquid chromatography. We selected for this study the nonapeptide bradykinin and the decapeptide kallidin ([Lys 0 ] bradykinin). The sequences of amino acids of these two kinins are as follows: bradykinin, Arg 1 -Pro 2 -Pro 3 - Gly 4 -Phe 5 -Ser 6 -Pro 7 -Phe 8 -Arg 9 (MW ) 1060); kallidin, Lys 1 -Arg 2 -Pro 3 -Pro 4 -Gly 5 -Phe 6 -Ser 7 -Pro 8 -Phe 9 -Arg 10 (MW ) 1188). Bradykinin and kallidin are peptides found in biologi- cal fluids, blood, plasma, and urine and referred to in mammals as “kinins”. Kinin peptide systems are complex and important. A considerable amount of research has been carried out on the biological activity of different bradykinin related compounds (13-17). They cause relaxation of the smooth muscle of veins and hypotension, increase vascular permeability, cause the contraction of the smooth muscles of the guts and airways, leading to increased airway resistance, and stimulate sensory neu- rons. Because of this broad spectrum of activity, kinins have been implicated in many pathophysiologies, includ- ing pain, sepsis, asthma, various symptoms associated with rhinoviral infection, rheumatoid arthritis, and a wide variety of other inflammatory diseases. A bradyki- nin antagonist used clinically to treat any of these disorders would represent a completely novel therapy. Bradykinin-related compounds could be used as new drugs for cancer and inflammation (18). For these reasons, the separation and purification of these two peptides is important. Reversed-phase liquid chromatography (RPLC) is a popular method for the separation of peptides. An important practical goal of this work is to determine the optimum conditions for the separation of bradykinin and kallidin by RPLC. To minimize the cost of the experiments, the determination * To whom correspondence should be addressed. Tel: (865) 974- 0733. Fax: (865) 974-2667. E-mail: guiochon@utk.edu. † The University of Tennessee and Oak Ridge National Labora- tory. ‡ Rzeszo ´w University of Technology. 945 Biotechnol. Prog. 2003, 19, 945-954 10.1021/bp020148h CCC: $25.00 © 2003 American Chemical Society and American Institute of Chemical Engineers Published on Web 04/23/2003