After the discovery that 1-methyl 4-phenyl 1,2,3,6tetrahydropyridine (MPTP) induced a parkinsonian syndrome in individuals who inadvertently injected the by-product of illicit drug manufacturing (Langston et al., 1983; Ramachandiran et al., 2007; Richardson et al., 2005), numerous investigators made the observation that the herbicide paraquat (N,N#dimethyl-4,4#-bipyridinium dichloride, CAS# 1910-42-5) bore a degree of structural similarity to the active metabolite of MPTP, 1-methyl-4-phenylpyridinium (MPPþ; CAS# 3979499-5; see Fig. 1). Since that time, well over 100 papers have been published using paraquat as a cellular or animal model of Parkinson’s disease under the premise that the two compounds are structurally similar. Direct injections of paraquat into the brain demonstrated its ability to damage the dopamine system and helped to identify some potential toxicological manifestations of the compound, but at the same time such studies largely ignored the physical obstacle presented by the blood-brain barrier and the complicated architecture of the brain (Barbeau et al., 1985; Corasaniti et al., 1992; Iannone et al., 1989; Liou et al., 1996). Indeed, it is intuitively obvious that an injection of a strong oxidant directly into nigral neurons or direct application to isolated mitochondria will cause damage, but is this toxicologically relevant? When paraquat is administered systemically, the results have been equivocal (Brooks et al., 1999; Corasaniti et al., 1992; McCormack et al., 2002; Widdowson et al., 1996). Some laboratories observe a consistent loss in nigral neurons (of no more than 25%), but not loss of striatal dopamine, which is considered a hallmark of the disease. Other laboratories have failed to demonstrate toxicity after systemic administration. In contrast, systemically administered MPTP kills a much higher percentage of the neurons (> 90%) and leads to a consistent loss of dopamine, both of which exhibit a clear dose response. Thus, while it does appear that after i.p. injection paraquat enters the brain of certain rodent species and strains, it is not clear if this is at all relevant to the human condition. Many of these widely cited studies, on which the ability of paraquat to damage dopamine neurons is based, are quite artificial from an exposure standpoint and only served to perpetuate the fishy and tenuous connection between the compounds. In retrospect, evidence suggests that the aforementioned research efforts have been somewhat misguided. It has become a veritable dogma that since paraquat looks like MPPþ that it must act like MPPþ (coincidentally, the structure of the related cyperquat is truly identical to that of MPPþ, see Fig. 1). Recent studies cast serious doubt upon this assertion. It has been assumed that paraquat is transported into the dopamine neuron by the dopamine transporter, like MPPþ. Further, it has been assumed (and stated in numerous reviews) that paraquat is a complex I inhibitor. Recent data demonstrate neither to be physiologically plausible. Studies using cell lines expressing the dopamine transporter reveal that paraquat is not transported via the dopamine transporter, nor does it impair dopamine uptake (Ramachandiran et al., 2007; Richardson et al., 2005). Paraquat can inhibit complex I, but only at an IC50 of 7mM, which would be biologically impossible, especially given the fact that in actively respiring mitochondria paraquat does not display further accumulation. Meanwhile, MPPþ inhibits complex I in the 30lM range, which considering the ability of MPPþ to be specifically accumulated in the dopamine neuron by the dopamine transporter and that actively respiring mitochondria do accumulate the toxic compound, is indeed physiologically relevant (Richardson et al., 2005, 2007). For comparison’s sake, the lipophilic piscicide rotenone inhibits complex I in the low nanomolar range. While paraquat can induce redox cycling in the cytosol, which could alter mitochondrial function indirectly, the data suggest that when compared to MPPþ or rotenone, paraquat 1 For correspondence via. E-mail: Gray.miller@emory.edu.
Read full abstract