Drug-induced parkinsonism: cinnarizine and flunarizine are potent uncouplers of the vacuolar H⁺-ATPase in catecholamine storage vesicles

Abstract

Cinnarizine (1-diphenylmethyl-4-(3-phenyl-2-propenyl)piperazine) and its di-fluorinated derivative flunarizine inhibit the MgATP-dependent generation of a transmembrane proton electrochemical gradient in chromaffine granule ghosts. The concentrations giving 50% inhibition (IC₅₀) of the MgATP-dependent generation of the pH-gradient were 5.9±0.6 μM (n=6) and 3.0±0.3 μM (n=5) for cinnarizine and flunarizine, respectively. The IC₅₀ values for inhibiting the generation of the membrane potential were even lower, i.e. 0.19±0.06 μM (n=6) and 0.15±0.01 μM (n=4) for cinnarizine and flunarizine, respectively. Cinnarizine (10 μM) also inhibited the energy-dependent vesicular uptake of [¹⁴C]-dopamine (50 μM) by 76%, i.e. from 2.1±0.9 to 0.5 ±0.6 nmol/mg protein/min (n=5, P<0.002). Cinnarizine (10 μM) increased the MgATPase activity of the granule ghosts by 47±26% (n=4) compatible with an uncoupling of the vacuolar H⁺-ATPase activity. The IC₅₀-values observed for the two compounds are in the same range as their reported therapeutic plasma concentrations in vivo, suggesting that cinnarizine and flunarizine may well inhibit proton pumping and catecholamine uptake in storage vesicles also in vivo. This mechanism of action may contribute to the drug-induced parkinsonism seen as a side-effect of the two drugs.

Introduction

Cinnarizine and flunarizine are classified as antihistamines, calcium channel blockers and anti-epileptics and have thus been used for the treatment of a broad variety of diseases and conditions like vertigo (Rascol et al., 1989), migraine and epilepsy (Todd and Benfield, 1989) and nausea (Wilder-Smith et al., 1991). Concerning side effects, both drugs have been linked to a high prevalence of drug-induced parkinsonism in humans (Garcia-Ruiz et al., 1992a, Negrotti et al., 1992). Thus, Baquero et al. (1995) found that 17 of 100 new cases of parkinsonism could be linked to administration of the two drugs. In contrast to true Parkinson’s disease, drug-induced parkinsonism often gives bilateral neurological symptoms (Baquero et al., 1995), indicating a pathogenic mechanism different from the neurodegeneration observed in Parkinsons’s disease. Garzia-Ruiz et al. (1992b) showed that chronic administration of cinnarizine induced a parkinson-like condition in healthy monkeys. They also found a 30–50% reduction in the levels of homovanillic acid (HVA) and 5-hydroxyindolactic acid in the cerebrospinal fluid, reflecting reduced presynaptic brain levels of dopamine and serotonin, respectively. On the other hand, peritoneal injection of flunarizine had no acute effect on rat-brain catecholamines (Gaggi and Gianni, 1990). In vitro, cinnarizine and flunarizine have been shown to reduce the uptake of dopamine into neuroblastoma cells (Mena et al., 1995).

The molecular mechanisms of the pharmacological side effects of cinnarizine and flunarizine are still unclear. In the present study, therefore, the possibility has been considered that the two compounds reduce presynaptic (i.e. vesicular store) dopamine levels via interference with the storage of dopamine in synaptic vesicles. As a model, we have studied the effect of cinnarizine and flunarizine on the bioenergetics of chromaffin granule ghosts linked to the uptake and storage of the amine. Even if the amine carrier differs in chromaffin granules and small synaptic vesicles (Peter et al., 1994), the generation of the proton electrochemical gradient is essentially the same in the two types of vesicles, generated by the fundamental V-type H⁺-ATPase.