Search

The utility of high-resolution magic-angle spinning (HRMAS) NMR for studying drug delivery in whole tissues was explored by dosing female Sprague-Dawley rats with topical or injectable benzoic acid (BA). In principle, HR-MAS NMR permits the detection of both intra- and extracellular compounds. This is an advantage over the previous detection of topically applied BA using microdialysis coupled to HPLC/UV as microdialysis samples only the extracellular space. Skin and muscle samples were analyzed by [sup.1]H HR-MAS NMR, and BA levels were determined using an external standard solution added to the sample rotor. One to two percent of the BA topical dose was detected in the muscle, showing that BA penetrated through the dermal and subcutaneous layers. Since BA was not detected in the muscle in the microdialysis studies, the NMR spectra revealed the intracellular localization of BA. The amount of BA detected in muscle after subcutaneous injection correlated with the distance from the dosing site. Overall, the results suggest that HRMAS NMR can distinguish differences in the local concentration of BA varying with tissue type, dosage method, and tissue proximity to the dosing site. The results illustrate the potential of this technique for quantitative analysis of drug delivery and distribution and the challenges to be addressed as the method is refined.

The drug discovery process relies on characterizing structure-activity relationships, since specific ligand-target interactions often result in important biological functions. Measuring diffusion coefficients by nuclear magnetic resonance spectroscopy is a useful way to study binding, because changes can be detected when a small ligand interacts with a macromolecular target. Diffusion coefficients can be miscalculated, however, due to magnetization transfer between the receptor and ligand. This transferred nuclear Overhauser effect (trNOE) disrupts the observed signal decay due to diffusion as a function of the experimental diffusion time. Since longer diffusion times also selectively edit free ligand signal, the measured diffusion coefficients become biased toward the fraction of bound ligand. Despite this discrepancy, under these experimental conditions, the trNOE selectively influences the measured signals of binding ligands and can be used to gain insight into ligand-protein interactions. These phenomena have been studied for caffeine and L-tryptophan, which bind to human serum albumin, and the antimalarial agent trimethoprim, which interacts with dihydrofolate reductase. The results provide insight into the nature of ligand-protein binding and are thus useful for elucidating the molecular features of the ligand that interact with the protein.