MAGNETIC RESONANCE IN CHEMISTRY Magn.Reson.Chem.37,269È273 (1999) NMR spectroscopic diþusion, chemical shift and linewid measurements of low-affinity binding of ibuprofen enantiomers to human serum albumin Yuehong Ma,1,2 Maili Liu,1* Xi-an Mao,1 Jeremy K. Nicholson3 and John C. Lindon3 1 Laboratory of Magnetic Resonance, Atomic and Molecular Physics, Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, China 2 Instrumental Analysis Research Center, Northwest University, Xian 710069, China 3 Centre for Biological Chemistry, Biomedical Sciences Division, Imperial College of Science, Technology and Medicine, University of London, Exhibition Road, London SW7 2AA, UK ABSTRACT: The binding ofracemic, (R)([) and (S)(])-ibuprofen (IBP)to human serum albumin (HSA) was studied using NMR spectroscopy. Having saturated the tight binding sites, the extent of weak binding was then investigated. The molecular di†usion coefficients of IBP in HSA solution at di†erent concentration ratios indicate that there is no signiÐcant di†erence in binding capacity of the two enantiomers of IBP to HSA at high IBP : H ratios.However,there are chemical shift and linewidth changes in the 1NMR spectrum of IBP in the presence of HSA induced by the binding and this suggests that the major binding interaction involves the sec-butyl chain IBP molecule binding to HSA in hydrophobic pockets. Copyright 1999 John Wiley & Sons, Ltd. ( KEYWORDS: NMR ; ibuprofen ; enantiomers ; human serum albumin ; binding ; di†usion coefficient (Received 16 July 1998 ; revised 16 October 1998 ; accepted 27 October 1998) INTRODUCTION Many drugs are bound extensively to human serum albumin (HSA) in bloodplasma.1h5The binding process can be regarded as a reversible Ðrst-order reac- tion to form a molecular complex and thisbinding process can a†ect drug transport to tissues and hence the compoundÏs therapeutic e†ect, toxicity and phar- mocokinetics. Therefore, drug binding to HSA can serve as a meansof drug storage and delivery control to tissue receptors and can prevent the drug from being metabolized rapidly.6,7 Hence there is much interest in the study of the interactions of drugs with HSA. Many clinically used therapeutics are administered as racemates, but increasingly single isomers are preferred for drug development programs. This is because the R and S enantiomers of a drug may have di†erent thera- peutic,pharmacokineticand toxicologicalactivities. Some ofthese e†ects may be modulated by binding of the drug to HSA. Consequently, the binding properties of enantiomers to HSA has become an area of much study.8h13 There are two main kinds of drug binding to a HSA molecule.1 One of these ishigh-affinity binding with three known speciÐc binding sites where each site can only hold one drug molecule. Two of the binding sites have been named Site I (the warfarin binding site) and Site II (the diazepam bindingsite).14In a recent report,11 Site I has been further divided into Site I (the * Correspondence to : Laboratory ofMagnetic Resonance, M. Liu, Atomic and Molecular Physics,Wuhan Instituteof Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, China Contract/grant sponsor : NSFC ; contract/grant number : 29875034. Contract/grant sponsor : CAS ; contract/grant number : KJ951-B1-402. phenylbutazone binding site) and Site III (the digitoxin binding site). The names ofthe sites are simply taken from the model compounds used to probe the sites. The second kind of binding is low-affinity binding and this has high binding capacity since there are often many binding sites leading to the possibility of many mol- ecules being bound. Depending on the concentration of a drug in blood plasma, it is possible that both types of binding could be important as modulators of biological e†ect. Ibuprofen (IBP)is a non-steroidal anti-inÑammatory drug (NSAID) and over 99% of IBP in the blood plasma isbound to HSA after oral administration.1 Therefore, IBP has been used intensively as a model compound to study drugÈprotein binding, particularly the high affinity binding equilibrium. Itoh et al.8 studied IBPÈHSA binding using a chiral Ñuorescent derivatiz- ing reagent and reported that the association constant of (R)([)-IBP was 2.3-fold greater than that of (k a ) (S)(])-IBP. This agrees with the result of a study using high-performance affinity chromatography,11 but the latter method showed that (S)(])-IBP had two major binding regions on HSA withvalues of 1.1 ] 105 and k a 1.2 ] 105 M~1, respectively, and the of (R)([)-IBP k a was 5.3 ] 105 M~1. On the other hand, another result9 indicated that (S)(])-IBP bound to HSA more tightly M~1)than (R)([)-IBP (k a \ 8.0 ] 105 (k a \ 4.8 ] 105 M~1),and that the racemicform bound lesstightly M~1)than either enantiomer based on (k a \ 1.4 ] 105 an analysisusing circulardichroism to monitorthe binding.Although from these results always lies in k a the 105 M~1 range, it clearly shows a method depen- dence. This might be explained by a contribution due to binding at the poorly deÐned diazepam-binding site and the possibility ofa binding-induced HSA conforma- Copyright ( 1999 John Wiley & Sons, Ltd. CCC 0749È1581/99/040269È05 $17.50