Changes in COX histochemistry in the brain of mice and rats exposed to chronic subcutaneous rotenone

Highlights

• Mice and rats were chronically injected with rotenone for 1,3 and 7 days.

• Time course of cell death and Cytochrome C oxidase activity in the brain was studied.

• Time course of neuronal loss in SNpc was similar, reaching 44 % and 42 % by the 7th day.

• Cytochrome C oxidase activity changes in mice were evident in cortex and hippocampus.

• In rats, acute changes were observed in SNpc and later in cortex and hippocampus.

Abstract

Exposure of experimental animals to the mitochondrial toxin rotenone is considered to be a model of environmental progression of Parkinson's disease (PD). We investigated the differential vulnerability of various brain regions to generalized inhibition of complex I, induced by subcutaneous rotenone injections for the duration of 1, 3 and 7 days in both rats (2 mg/kg dosage) and mice (4 mg/kg dosage). To examine patterns of metabolic activity changes in the brain, histochemical evaluation of cytochrome C oxidase (COX) activity was performed in post mortem brain sections. Animals displayed a similar time course of neuronal loss in substantia nigra pars compacta (SNpc), reaching 44 % in mice and 42 % in rats by the 7th day. The pattern of COX activity changes, however, was different for the two species. In both experiments, metabolic changes were evident not only in the substantia nigra, but also in non-specific structures (cortex and hippocampus). In mice, a decrease in COX activity was shown mostly for the non-specific areas (V1 cortex and ventral hippocampus) after the single exposure to rotenone. Data from the experiment conducted on rats demonstrated both an acute metabolic decrease in mesencephalic structures (SNpc and nucleus ruber) after a single injection of rotenone and secondary changes in cortical structures (S1 cortex and dorsal hippocampus) after chronic 7 day exposure. These changes reflect the general effect of rotenone on neuronal metabolic rate.

Introduction

Parkinson's disease (PD) is a chronic neurological disorder characterized by the slowly progressing neurodegeneration of the dopaminergic cells in substantia nigra pars compacta (SNpc) (Dauer and Przedborski, 2003). Damage to midbrain dopamine neurons leads to a hypodopaminergic state and changes in the extrapyramidal motor system. The motor symptoms of Parkinson's disease, which include postural instability, rigidity, bradykinesia, and resting tremor, develop only after the death of 50–60 % cells in the SNpc and a 60 % decrease in the mesostriatal pathway dopamine levels (Braak et al., 2004). Thus, the progression of the disease on its presymptomatic stages and the actual cause leading to PD progression remain obscure. The prevalence of PD up to 5–7 % in rural population over 50, compared to 1% in the general population, supports the role of certain environmental factors, such as exposure to certain pesticides, e.g. dieldrin, paraquat, or rotenone (Shastry, 2001). Extensive literature links exposure to rotenone to an increased risk of Parkinson's disease development, although most of the studies have been unable to establish cause-effects relationships. Generally, exposure of experimental animals to rotenone is considered a model of environmental progression of PD (Dauer and Przedborski, 2003). Systemic exposure of rodents, mostly rats, to low doses of rotenone leads to typical features of PD progression: Parkinsonian behaviour (Cannon et al., 2009), gastrointestinal pathology (Drolet et al., 2009), a-synuclein cytoplasmic inclusions in dopaminergic neurons (Sherer et al., 2003b), selective cell death in the SNpc (Betarbet et al., 2000), a decline in the mesostriatal dopamine pathway, and electrophysiological changes in basal ganglia output structures (von Wrangel et al., 2015). However, there are many studies questioning the construct validity of this model. That is, chronic exposure of rats (Zhang et al., 2017) and mice (Inden et al., 2011) to rotenone leads to a high mortality rate from its systemic toxicity, which has nothing to do with PD. The locality of cell death in the rotenone model is also being questioned, and the pattern of neurodegeneration more closely resembles multiple system atrophy (Hoglinger et al., 2003). Time course of neurodegeneration and other pathological processes, such as oxidative damage and mitochondrial dysfunction, is not covered for this model in contrast to the more widely used mitochondrial toxin - 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (Bezard et al., 1998). Mitochondrial dysfunction is often listed as a primary or secondary contributor to neurodegenerative events in Parkinson's disease (Terron et al., 2018). Neurons, with their high energy demand, rely heavily on adenosine triphosphate (ATP) production via oxidative phosphorylation in the mitochondria, which is impaired by rotenone. But, to date, the contribution of this factor to the progression of rotenone-induced pathology in vivo has not been quantified. Cytochrome C oxidase (COX) histochemistry provides useful information about regional metabolic variation in the brain, tends to reflect long-term changes in brain function, and has superior spatial resolution in comparison to other metabolic techniques such as 2-deoxyglucose autoradiography. COX histochemistry was first used to study metabolic patterns in the visual cortex, and the effects of sensory deafferentation on cortical activity (Wong-Riley, 1989). This method is still widely applied in the field of behavioral neurobiology to study networks involved in spatial learning (Begega et al., 2012) and effects of forced exercise or different drugs on brain metabolism (Sampedro-Piquero et al., 2013). Its application is especially interesting in the field of neurodegenerative disease, like Alzheimer's (Strazielle et al., 2009) or Parkinson's (Porter et al., 1994), where mitochondrial dysfunction is thought to be implicated. However, the use of COX measurements in the studies of PD is relatively limited - it was used to show differences in brain metabolism, and presumably brain activity, in the PD model, induced by unilateral 6-hydroxydopamine lesions (Porter et al., 1994; Vlamings et al., 2009). Using COX histochemistry as an indirect measure of relative synaptic activity, these works showed an increase in activity in basal ganglia output structures. However, in both of these studies the neurotoxin was applied locally, and its primary mechanism of action is the generation of reactive oxygen species, rather than complex I inhibition (Kumar et al., 1995). COX metabolic mapping after systemic exposure to the mitochondrial toxin allows to distinguish the brain regions most vulnerable to the mitochondrial damage and changes in metabolic rate.

In this study, we investigated the time course of neuronal damage in the SNpc and region-specific susceptibilities to a neuronal oxidative phosphorylation deficiency, induced by rotenone injections in two different PD models (rats and mice).