Dilute liquid crystals used to enhance residual dipolar couplings may alter conformational equilibrium in oligosaccharides Patrick Berthault, a Damien Jeannerat, b Franck Camerel, c Francisco Alvarez Salgado, a,d Yves Boulard, d Jean-Christophe P. Gabriel, c,e Herve ´ Desvaux a, * a Laboratoire Commun de R.M.N., DSM/DRECAM/Service de Chimie Mole ´culaire, URA 331 CNRS, C.E.A./Saclay, F-91191 Gif sur Yvette, France b De ´partement de Chimie Organique, Universite ´ de Gene ` ve, 30 quai Ernest Ansermet, CH-1211 Gene ` ve, Switzerland c Sciences Mole ´culaires aux Interfaces, FRE 2068 CNRS, 2 rue de la Houssinie `re, B.P. 32229, F-44322 Nantes, France d Laboratoire Commun de R.M.N., DSV/DBJC/Service de Biochimie et de Ge ´ne ´tique Mole ´culaire, C.E.A./Saclay, F-91191 Gif sur Yvette, France e Nanomix Inc., 5980 Horton Street, Sixth floor, Emeryville, CA 94608, USA Received 14 October 2002; received in revised form 20 February 2003; accepted 17 May 2003 Abstract The solution structures of a trisaccharide and a pentasaccharide containing the Lewis x motif were determined by two independent approaches using either dipolar cross-relaxation (NOE) or residual dipolar coupling (RDC) data. For the latter, one-bond 13 C  / 1 H RDC enhanced by two different mineral liquid crystals were used alone. Home-written programs were employed firstly for measuring accurately the coupling constants in the direct dimension of non-decoupled HSQC experiments, secondly for transforming each RDC data set into geometrical restraints. In this second program, the complete molecular structure was expressed in a unique frame where the alignment tensor is diagonal. Assuming that the pyranose rings are rigid, their relative orientation is defined by optimizing the glycosidic torsion angles. For the trisaccharide, a good agreement was observed between the results of both approaches (NOE and RDC). In contrast, for the pentasaccharide, strong discrepancies appeared, which seem to result from interactions between the pentasaccharide and the mesogens, affecting conformational equilibrium. This observation is of importance, as it reveals that using simultaneously NOE and RDC can be hazardous as the former represent 99% of the molecules free in solution, whereas the latter correspond to less than 1% of the structure bound to the mesogen. # 2003 Elsevier Ltd. All rights reserved. Keywords: Residual dipolar coupling; Lewis x ; Mineral liquid crystal; Oligosaccharide; Conformer equilibrium 1. Introduction Residual dipolar couplings (RDC) enhanced by dilute liquid crystals are recognized as valuable tools for the structure determination of biomolecules through NMR spectroscopy. 1,2 The small degree of anisotropy intro- duced in the molecular motion leads to non vanishing dipolar interactions and therefore, to supplementary experimental data referenced to the same external frame. The prerequisite is that the 3D structure of the molecule under study is not affected by the presence of the mesogen. Even if only a small fraction of molecules (typically 10 3  /10 4 ) interacts with the mesogen, there is no direct and general proof that the average structure in solution, as determined from dipolar cross-relaxation experiments (NOE), is the same as the one given by RDC. Indeed classical NMR data, such as chemical shifts or relaxation times measured in oriented media and compared to isotropic values cannot answer to this question, since the extracted values cannot figure out the few hundredths of a percent of the molecule bound to the mesogen. The situation may worsen when structures of small flexible compounds are studied, since: (i) most of the atoms are close to the surface and then potentially interact with the mesogen; (ii) the presence of the mesophase can displace the equilibrium between the conformers. This effect has already been directly de- tected for small organic compounds dissolved in chiral * Corresponding author. Tel.:  /33-169-086483; fax:  /33- 169-089806. E-mail address: hdesvaux@cea.fr (H. Desvaux). Carbohydrate Research 338 (2003) 1771  /1785 www.elsevier.com/locate/carres 0008-6215/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0008-6215(03)00243-X