Drug metabolites are not always safe. Xenobiotic metabolism frequently produces reactive metabolites that could directly or indirectly alter brain function and neuron survival.
The metabolism of neurotransmitter amines by cerebral MAO-A and -B produces equivalent amounts of the corresponding aldehydes, and ammonia. Both products are considered to be neurotoxic and could participate in the progressive neuronal death that occurs during senescence. Moreover, implication of MAOs in xenobiotic metabolism was definitely assessed by the demonstration that MPTP, a meperidine analogue promoting a parkinsonian-like syndrome in humans (66), monkeys (67), and mice (68), was a substrate of MAO-B in most mammalian species assayed (69). The precise mechanism by which MPTP administration induces selective dopaminergic neuronal death in the pars compacta of the substantia nigra is still a subject of debate. Exposure to MPTP results in the formation of MPP + , which is actively taken up by dopaminergic neurons and kills them either by blocking complex 1 of the mitochondrial respiratory chain or by promoting the formation of superoxide and hydroxyl radicals (for a recent review, see Ref. 70). Defects in both mito-chondrial complex 1 and oxidative stress are considered to be the pathological processes that may cause the degeneration of dopaminergic neurons in idio-pathic Parkinson's disease (71). MPP+ is a polar compound: it does not cross the BBB, and accordingly, an intravenous MPP+ injection does not promote neurotoxicity.
MPTP has emerged as an environmental molecule allowing the design of animal models of Parkinson's disease, which is a specific human neurodegeneration of striatal dopaminergic neurons, the etiology of which remains largely unknown (70). Investigations of MPTP also allowed the finding of endogenous molecules that are structurally related to MPP + , such as isoquino-line and P-carboline derivatives. These molecules are oxidized by MAO to reactive isoquinolinium or P-carbolinium species that inhibit mitochondrial respiration and promote oxidative stress. A deregulation of their normal metabolism could participate in the etiology of Parkinson's disease (for a recent review, see Ref. 72). These results suggest that administration of l-deprenyl, a specific MAO-B inhibitor, is of great potential in the treatment of patients with Parkinson's disease.
The very high MAO-B activities observed in isolated rat brain microves-sels have been supposed to protect the brain against MPTP toxicity (24). This seems to be confirmed by a report that an intranigral infusion of low quantities of MPP+ destroyed rat dopaminergic neurons (74).
2. Flavin-Containing Monoxygenases and Cytochromes P450 NADPH-dependent monoxygenases may easily form diverse potentially harmful metabolites (e.g., epoxides, nitrosamines, sulfoxides, imminium spe cies, aldehydes) that may form protein adducts and cytosolic antigens, or promote more or less reversible neurotoxicity. In this chapter, for evident space and topic reasons, we review only some examples of neurotoxicity resulting from cerebral metabolism of drugs or xenobiotics.
A first example concerns the metabolism of the phosphorothionate insecticide parathion to a potent cerebral anticholinesterase metabolite, paraoxon (75). The parent molecule is activated in vivo by CYP-mediated desulfuration in close vicinity of the target enzyme. The environmental use of that molecule may therefore result in severe neurotoxicity.
Nicotine appears to be both a psychostimulant drug and a common environmental pollutant promoting ''passive smoking.'' It should be also considered to be a drug of addiction, since thousands of heavy tobacco smokers are unable to give up smoking, showing therefore a true dependence on nicotine. Smoking and inhalation are routes of administration that allow a very rapid delivery of the drug to the brain, and the first daily puff on a cigarette is considered by tobacco smokers to be ''the best one'' because it efficiently attenuates the morning's withdrawal symptoms. Nicotine is a pharmacologically active tertiary amine, efficiently metabolized to cotinine by both liver and brain CYP (76) through the formation of a reactive nicotine-A[1'(5')j-imminium ion, which is an alkylating species (77). During its metabolism, nicotine undergoes covalent binding to microsomal protein, supporting the concept that reactive metabolic intermediates may play a role in the pharmacology and toxicity of nicotine.
Phencyclidine, originally developed as an anesthetic and analgesic in dentistry, was withdrawn from human use because of significant side effects, including agitation, blurred vision and hallucinations, and paranoid behavior (78). Owing to its hallucinogenic properties, PCP has become a popular drug of abuse, with different behavioral responses among individuals. Liver biotransformation of PCP results in the formation of numerous mono-or dihydroxylated metabolites, suggesting the involvement of several CYP isoforms in this metabolic pathway (79) and explaining in part the different behavioral responses to the drug resulting from genetic polymorphism among individuals (80). PCP is also metabolized in the brain, forming both inactive and pharmacologically active metabolites near their receptors (81, 82). Moreover, PCP metabolism promotes its covalent binding to macro-molecules and inactivation of hepatic CYP2B1 through the NADPH-dependent a-carbon oxidation of PCP, which leads to the formation of the electrophilic PCP imminium ion (83). This irreversible inhibitory effect has also been demonstrated in the rat brain, where PCP inactivates specifically CYP-catalyzed ethoxyresorufin dealkylation (Perrin and Minn, unpublished results).
The selective irreversible binding of reactive intermediates formed by CYP1A metabolism has also been used as a marker of metabolism in some tissues or cells. A high binding in the endothelial cells of the capillary loops of the choroid plexus was described in 1994, suggesting an efficient, BNF-inducible, CYP1A-dependent metabolism in these endothelial cells (84).
No experimental data concerning cerebral mEH-promoted toxicity is available. The anticonvulsants phenytoin and carbamazepine, which possess aryl moieties, are metabolized in the human liver to epoxide intermediates responsible for hepatic necrosis (85). To our knowledge, however, no evidence of such deleterious effects either at the blood-brain barrier or to the brain has been presented. It should be of importance that the human brain displays a 40-fold higher mEH activity than the rat brain (43).
Enzymes catalyzing phase 2 of drug metabolism may also form reactive metabolites. For instance, UGT may promote the formation of protein ad-ducts when it is conjugating carboxylic acids like nonsteroidal antiinflammatory drugs (NSAIDs) (86). But to our knowledge, even if there is evidence that some NSAIDs cross the blood-brain barrier and enter the CNS, the possible formation of reactive metabolites in the brain has never been reported. Formation of reactive metabolites has been also described during hepatic conjugation of arylamines or polycyclic arylmethanol with sulfate and glutathi-one; but once again, no evidence of such activities in the brain has been presented.
The neurotoxicity of a number of molecules has been already well documented, but it seems probable that the long-term intake or consumption of rather common pesticides, herbicides, drug additives, or more or less illicit drugs will be related in the relatively near future to neurological disorders or neurodegenerative diseases.
Was this article helpful?