| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Correspondence |
We wish to take issue with a paper by Violon [1] recently published in The British Journal of Radiology. This study used a single and perforce limited approach, namely quantum mechanical calculations, to question our model of the stabilization of the hydrophilic sphere of iobitridol [2]. Our model was based on convergent results obtained experimentally (HPLC, 1H-NMR and 13C-NMR spectroscopy) and by molecular modelling simulations. In contrast, Dr Violon's work was not validated by experimental data and the inevitable limitations of computational methods were not taken into account. The term "experiment" was used improperly by Violon to qualify a plain exploration of the conformational behaviour of molecules in the vacuum, using standard quantum mechanical calculations. In addition, no experimental validation was provided, excepting a brief and flawed interpretation of partition coefficients (LogP). The precision and accuracy of partition coefficients decreases markedly close to the lower experimental limit of -3. The differences in LogP values between ioversol (-2.98), iohexol (-2.71), iobitridol (-2.63) and ioxilan (-2.44) are barely meaningful and should be interpreted with the greatest caution.
Our paper [2] had clearly demonstrated the non-equivalence of iobitridol atropisomers in terms of equilibrium populations and dynamic behaviour. The influence of stereochemical factors on the accessibility of molecular surface and on the dynamics of interconversion between isomers was also outlined. These fundamental molecular characteristics of iobitridol were ignored by Violon.
The structural complexity of the compounds under investigation allows marked intramolecular interactions between substituents to occur when the relaxation of the entire molecule, except two torsion angles used to define the rotors, is allowed. As the effect of these interactions was ignored in Violon's study, the observed stabilization of the resulting geometries was in fact owing to inter-substituent interactions rather than to the actual energy of the transition states. In other words, the structural complexity of these compounds and the resulting inter-substituent interactions forbid rotation barriers to be derived from energy differences between low-energy and high-energy geometries, as done by Violon. Furthermore, the perturbing effect of inter-substituent interactions is greatly increased in in vacuo simulations, whose relevance to the conditions existing in concentrated solutions as used clinically is limited at best. These limitations had led us to estimate rotation barriers exclusively from model compounds [2]. The alternative approach of using the Ampac/Mopac algorithms to identify transition states proved impractical when simulating complex molecules. This is owing to the failure of the analysis of Hessian matrix in such cases [3].
Obviously, the stabilization of the hydrophilic sphere of iobitridol we proposed is a hypothesis; one supported by experimental results obtained by independent methods and boundary-aware modelling simulations [2]. This concept has been supported by the crystal structure of iobitridol with elastase [4] and by its kinetics of elimination in rat kidney [5]. Despite extensive clinical use [6], there is no evidence of a larger chemotoxicity of iobitridol that would have invalidated this hypothesis. There is nothing in Violon's paper to support the author's questioning of our conclusion.
Institute of Medicinal Chemistry, Geneva-Lausanne School of Pharmacy, University of Lausanne, CH-1015 Lausanne, Switzerland
Received for publication June 10, 2002. Accepted for publication July 4, 2002.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
