Accelerated toxicity screening using NMR and pattern recognition-based methods.

Abstract

(1)H-NMR spectroscopy has proved to be a powerful and efficient means of monitoring the interaction of pharmacological agents with cells and tissues. The application of this technique to biofluid analysis, gives rise to a comprehensive metabolic profile of the low molecular weight components of biofluids, that reflect concentrations and fluxes of endogenous metabolites involved in key intermediary cellular pathways, thereby giving an indication of an organism's physiological or pathophysiological status. Recent developments in spectrometer technology have resulted in increased sensitivity and dispersion. Together with the increased capacity for sample throughput (~ 300 samples/day), arising from the latest advances in flow probe technology and in robotic transfer systems, (1)H-NMR spectroscopic techniques have become viable in terms of toxicological screening. However, the complexity of high-field biofluid spectra in conjunction with the increased capacity for sample handling, leading to a rapid growth in the size of toxicological spectral databases, has placed greater emphasis on the need to develop improved automated procedures for data processing and interpretation. By harnessing chemometric tools to the analysis of complex spectral data, the toxicological consequences of xenobiotic exposure can be evaluated efficiently on-line. Automation of spectral processing procedures and the construction of mathematically-based 'expert systems' for the prediction of drug-induced toxicity founded on 1H-NMR spectral profiles, have now been achieved. In this article, we review the recent developments in NMR and pattern recognition analysis and consider their application in toxicological screening.