Journal of Chromatography A, 1132 (2006) 289–296 Determination of the binding parameters for antithrombin–heparin fragment systems by affinity and frontal analysis continuous capillary electrophoresis T. Le Saux a,1 , A. Varenne a,∗ , F. Perreau a , L. Siret b,2 , S. Duteil b , L. Duhau b , P. Gareil a a Laboratoire d’Electrochimie et Chimie Analytique, UMR CNRS 7575, Ecole Nationale Sup´ erieure de Chimie de Paris, Universit´ e ParisVI, 11, rue Pierre et Marie Curie, 75231 Paris Cedex 05, France b Sanofi-Aventis, Analytical Sciences, CRVA, 13 Quai Jules Guesde, BP14, 94403 Vitry-sur-Seine Cedex, France Received 5 May 2006; received in revised form 27 June 2006; accepted 28 July 2006 Available online 11 September 2006 Abstract The suitability of affinity capillary electrophoresis (ACE) and frontal analysis continuous capillary electrophoresis (FACCE) for binding constant determination was investigated for complexes between heparin fragments and antithrombin III, one of the main target proteins in the coagulation cascade. In a 100mM ionic strength phosphate buffer (pH 7.4), ACE was suitable to determine weak to medium interactions developed by short oligomeric heparin fragments, but it failed for decasaccharide, which presents a more complex irreversible interaction. However FACCE allowed evaluating the binding constant for these longer oligomeric fragments. Both experimental approaches were complementary for a wide variety of heparinic fragments. © 2006 Elsevier B.V. All rights reserved. Keywords: Antithrombin-heparin binding; Capillary electrophoresis; Affinity; Frontal analysis 1. Introduction Studies on non-covalent molecular interactions are arousing great interest in biology and medicine. For instance, plasma protein binding plays an important role in pharmacokinetics and pharmacodynamics of drugs [1]. Of special recent concern are also the interactions between carbohydrates and proteins, since it has been recognized that they mediate fundamental bio- logical mechanisms, encompassing growth control, apoptosis, fertilization, cell differentiation, proliferation and morphogene- sis as well as physiopathologic disorders like tumoral metastasis, autoimmune diseases, inflammation and host-parasite interac- tions [2,3]. Nevertheless, the difficulties encountered in the analysis of anionic bioactive polysaccharides are important hin- ∗ Corresponding author. Tel.: +33 1 55 42 63 72; fax: +33 1 44 27 67 50. E-mail address: anne-varenne@enscp.fr (A. Varenne). 1 Present address: D´ epartement de Chimie, UMR 8640, Ecole Normale Sup´ erieure, 24 rue Lhomond, 75231 Paris Cedex 05, France. 2 Present address: Laboratoire Franc ¸ais du Fractionnement et des Biotech- nologies, 3 avenue des Tropiques, BP 305, 91958 Courtaboeuf Cedex, France. drances to a better understanding of the strength and the speci- ficity of their interactions with proteins [4]. Among the different methods available for studying biomolecular interactions, capillary electrophoresis (CE) offers powerful attributes, making this separation method coupled to on-line quantitative detection very attractive and well adapted to the study of such non-covalent carbohydrate-protein com- plexes. Compared to other methods like gel filtration chromatog- raphy, dialysis, classical gel electrophoresis or more recently isothermal calorimetry, it requires a small amount of sample, keeping the consumption of carbohydrates and proteins to a minimum. In addition, and contrary to surface plasmon reso- nance, CE allows binding assay in solution where the inter- acting macromolecules remain free at pH and ionic strength values relevant for the biological conditions. Currently, several CE methods are available to measure binding constants, e.g. direct injection (DI), affinity capillary electrophoresis (ACE), Hummel–Dreyer method (HD), vacancy affinity capillary elec- trophoresis (VACE), vacancy peak method (VP) and frontal analysis (FA) [5–8]. In order to estimate a binding constant, the parameter measured experimentally should vary according to 0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.07.092