Organ Toxicity

Good biomarkers for organ toxicity are preferably obtained by non-invasive means; thus, urinalysis by NMR is arguably the most effective tool for cataloguing nephro-toxicity. Conventional clinical biomarkers are enzyme-based and thus relatively inexpensive for screening purposes. These typically include enzymes such as y-GT, intermediates like ALA, and breakdown products in urine (5-hydroxymethyluracil). However, the disadvantage of serum enzymes is that changes tend to be transient and usually only indicate significant pathological damage. Because NMR is nonselective for most low molecular weight intermediates of metabolism, its advantage is in exploring biomarkers that may be detected by other methods only with great difficulty. Compared with bile and plasma, urine is relatively easy to analyze, as reflected by exhaustive reviews of urinalysis for model drug compounds and drugs in development, reported elsewhere [17]. Centrifugation and filtering are the minimal sample preparations required for urine and bile samples. Plasma samples also require a protein-precipitation step. The main problem is the identification of small amounts of metabolite in large dilute urine samples over the background of endogenous urinary components. Sample preparation is clearly an important step in this process if 1H NMR is to provide useful structural data. Bile, however, is a more difficult matrix to analyze, as it contains high bile salt concentrations, micellar components, and detergent properties [17]. Other more exotic biofluids, albeit of less clinical importance, that can be probed for biomarkers of toxicity include amniotic fluid, milk, synovial fluid, aqueous and vitreous humour, saliva, and cyst fluid. These can provide useful windows into toxic mechanisms of specific organ and tissue damage in rodent models, but as yet few studies have been published.

The main organ toxicities studied by NMR have been hepatotoxicity [2, 3, 9, 10] and nephrotoxicity [2, 3, 10]. Early studies identified taurine as a biomarker for liver damage and creatine for testicular damage [35]. Nicholson and coworkers have argued that it is not any single biomarker but rather a combination of altered metabolites which are significantly changed over time that constitutes a more predictive model of toxicity [2, 3, 9, 10]. Building on the above studies, one of the most systematic programs to date using NMR to construct databases and metabolic models of drug toxicity has been the COMET program, as outlined in Section 8.3.3.

0 0

Post a comment