Recent interest in the neurotoxicity of haloperidol is based on its ox
idation in rodents to the pyridinium derivative, HPP+, a structural an
alog of the neurotoxin, 1-methyl-4-phenylpyridinium (MPP(+)). Recently
, we reported that HPP+ and a newly identified reduced pyridinium, RHP
P(+), were present in blood and urine of haloperidol-treated schizophr
enics and that the concentrations of RHPP(+) exceeded those of HPP+. I
n this study, we examined pathways for formation of RHPP(+) in subcell
ular fractions of human liver (n = 5) and brain (basal ganglia, n = 5)
. The major pathway was reduction of HPP+ (20 mu M) to RHPP(+) in cyto
sol (0.17-0.39 and 0.03-0.07 mu M RHPP(+)/g cytosolic protein per h in
liver and brain, respectively). The reactions were inhibited signific
antly by menadione and in brain also by daunorubicin. The inhibition p
rofile, cytosolic location and strict NADPH dependence suggest that th
e enzymes involved are ketone reductases. A second pathway was oxidati
on of reduced haloperidol (50 mu M), a major metabolite of haloperidol
in blood and brain, to RHPP(+). In liver microsomes, 0.17-0.63 mu mol
RHPP(+) was formed /g microsomal protein per h. A potent inhibitor of
the pathway was ketoconazole (IC50, 0.8 mu M), which suggests that P4
50 3A isozymes could be involved. In brain mitochondria but not micros
omes, reduced haloperidol (120 mu M) was oxidised to RHPP(+) at a smal
l but significant rate (0.005-0.020 mu mol RHPP(+)/g mitochondrial pro
tein per h) which was not attenuated by SKF 525A, quinidine, ketoconaz
ole, or monoamine oxidase inhibitors. Further studies are warranted to
establish the biological importance of these metabolites in vivo.