Metabonomics:: Evaluation of nuclear magnetic resonance (NMR) and pattern recognition technology for rapid in vivo screening of liver and kidney toxicants

The Purpose of this study was to evaluate the feasibility of metabonomics technology for developing a rapid-throughput toxicity:screen using 2 known hepatotoxicants: carbon tetrachloride (CCl4) and alpha-naphthylisothiocyanate (ANIT) and 2 known nephrotoxicants: 5-bromoethylamine (BEA) and 4-aminophenol (PAP). In addition, the diuretic furosemide (FURO) was also studied. Single doses of CCl4 (0.1 and 0.5 ml/kg), ANIT (10 and 100 mg/kg), BEA (15 and 150 mg/kg), PAP (15 and 150 mg/kg) and FURO (1 and 5 mg) were administered as single IP or oral doses to groups of 4 male Wistar rats/dose. Twenty-four-h urine samples were collected pretest, daily through Day 4, and on Day 10 (high dose CCl4 and BEA only). Blood samples were taken on Days 1, 2, and 4 or 1, 4, and 10 for clinical chemistry assessment, and the appropriate target organ was examined microscopically. NMR spectra:of urine were acquired and the data processed and subjected to principal component analyses (PCA). The results demonstrated that the metabonomic approach could readily distinguish the onset and reversal of toxicity with good agreement between clinical chemistry and PCA data. In at least 2 instances (ANIT and BEA), PCA analysis suggested effects at low doses, which were not as evident by clinical chemistry or microscopic analysis. Furosemide, which had no effect at the doses employed, did not produce any changes in PCA patterns. These data support the contention that the metabonomic approach represents a promising new technology for the development of a rapid throughput in vivo toxicity screen.