Effects of Rotenone and 6-OHDA on Dopaminergic Neurons of the Substantia Nigra Studied In Vitro

Abstract

This study investigated the neurotoxic effects of rotenone and 6-hyroxydopamine (6 OHDA), two compounds which have been implicated in Parkinson’s disease (PD). PD is a neurodegenerative disorder that results in the impairment of movement. During the disease process, a group of dopamine-containing cells in the brain region called the Substantia Nigra pars compacta (SNc), degenerate. Whilst genetic factors contribute to approximately 5% of PD cases, the causes of the remaining 95% are unknown. What does seem clear is the pivotal role of mitochondrial dysfunction as observed in post-mortem human tissue. Mitochondrial dysfunction leads to energy depletion and the generation of harmful reactive oxygen species (ROS). However, despite the fact that the involvement of mitochondria in the disease process has been well established, the cellular events that lead to, and result from, mitochondrial dysfunction remain poorly understood. Rotenone and 6 OHDA have been implicated in PD for two reasons: (1) both toxins can relatively selectively kill SNc neurons in animal models of PD, and (2) there is evidence for both compounds having a potential causative role in the etiology of the disease in humans. When 6 OHDA is injected into the brain, or rotenone applied systemically, both toxins cause degeneration of SNc neurons. This ability makes them excellent tools for studying mechanisms of PD in animal models. In addition, both toxins inhibit mitochondrial function. Despite extensive use in models of PD, the mechanisms by which each toxin cause cell damage remains elusive. The first part of this study investigated the acute responses of dopaminergic SNc neurons to rotenone exposure (5 nM – 1 µM). The experiments were conducted on brain slices obtained from rats. Electrophysiological recordings (whole-cell patch-clamp technique) were used to detect activation of specific membrane channels as well as cell firing and changes to the membrane potential. In addition, imaging of several fluorescent dyes sensitive to specific cellular events was carried out. In voltage-clamp experiments, acute rotenone (200 nM – 1 µM) application evoked a concentration-dependent outward current which was mediated by tolbutamide-sensitive KATP channels. The current was associated with a drop in cell input resistance (Rm) and, in current-clamp, membrane hyperpolarization and inhibition of spontaneous action potentials. The mechanisms by which rotenone activates KATP channels is controversial, with some studies suggesting activation by ATP depletion and others by elevated reactive oxygen species (ROS). To address this issue, experiments were conducted with high levels of ATP in the pipette solution. Since the rotenone-induced outward current was unaffected by high ATP levels, it was concluded that KATP channel activation was due to oxidative stress. Indeed, the antioxidant Trolox significantly attenuated the current response. Confirmation of elevated ROS production was obtained by recording increased mitochondrial superoxide production, using the fluorescent dye MitoSOX. In addition, rotenone evoked depolarization of mitochondrial membrane potential (ΔΨm). Measurements of intracellular Ca2+ and Na+ were performed using the fluorescent dyes Fura 2 and SBFI, respectively. Rotenone evoked increases to both [Ca2+]i and [Na+]i in a concentration-dependent manner. The rotenone-induced [Ca2+]i rise was unaffected by blocking KATP channels with Cs+. The elevation of [Ca2+]i is particularly important in relation to cell death, since [Ca2+]i overload is known to activate pathways leading to necrosis and apoptosis. There has been growing interest in the synergistic action of rotenone with other toxins/conditions which also enhance [Ca2+]i. This concept was explored in the present study by testing the relationship between the baseline [Ca2+]i level and the rotenone-induced [Ca2+]i increase. Two approaches were taken. Firstly, baseline [Ca2+]i was deliberately raised by activation of voltage-gated calcium channels. When rotenone was applied in the presence of this raised baseline calcium level, the rotenone-induced [Ca2+]i rise was significantly greater. The second approach involved post-hoc analysis of the relationship between the normal cellular variation in baseline [Ca2+]i and the rotenone-induced [Ca2+]i elevation. This analysis also revealed a dependency of the rotenone-induced [Ca2+]i elevation on the baseline calcium level. From this finding, as well as the observation that rotenone evoked ROS production, Transient Receptor Protein subtype M2 (TRPM2) channels were proposed as the likely underlying mechanism. The potentiation of the rotenone-induced [Ca2+]i rise by an elevation in baseline calcium level can be attributed to the calcium-dependence of ROS-sensitive TRPM2 channels, known to respond with increased channel opening to increased [Ca2+]i. Recent findings from our laboratory have confirmed TRPM2 involvement in rotenone toxicity, since blockade of these channels with ACA reduced the rotenone-induced [Ca2+]i rise (K. Chung, unpublished). Imaging using the fluorescent dye propidium iodide (PI) to label cells with compromised membrane integrity was also conducted in acute midbrain slices. SNc neurons were retrograde-labelled with FluoroGold and then exposed to various toxic insults. The detergent Triton-X100 caused an increase in PI labelling, whilst rotenone and high concentrations of glutamate were ineffective over the period of time investigated (up to 40 min). The second part of this study, also conducted on acute rat midbrain slices, investigated the acute responses of SNc neurons to 6 OHDA (0.2 – 2 mM) exposure. Extracellular recordings of action potential firing were conducted on SNc neurons. 6 OHDA evoked rapid inhibition of firing in a similar manner to dopamine (100 µM). In the presence of D2 dopamine receptor blocker sulpiride, the inhibition of firing evoked by 6 OHDA was delayed, and an initial increase of firing was observed. Blockade of the dopamine transporter with nomifensine reduced the 6 OHDA-induced inhibition of firing, and prevented the persistent inhibition of firing after 6 OHDA washout. For comparison, the response to 6 OHDA of non-dopaminergic neurons in the subthalamic nucleus was also studied. In the subthalamic nucleus, 6 OHDA evoked an increase of spontaneous action potential firing. Rapid application of 6 OHDA (using the picospritz application technique) in voltage-clamp recorded SNc neurons evoked an outward current, similar to that observed after dopamine application. In the presence of sulpiride, 6 OHDA induced an inward current, consistent with the initial increase of firing activity observed in extracellular recordings. Microfluorometric experiments with Fura 2, showed that 6 OHDA evokes an increase in [Ca2+]i. Loading cells with the fluorescent dye Lucifer Yellow enabled visualization of 6 OHDA-induced swelling of the cell body and damage to proximal dendrites. Imaging of SNc neurons loaded with dextran-rhodamine revealed 6 OHDA-induced damage of distal dendrites. The last part of the study was performed on organotypic cultures obtained from slices of the ventral midbrain. These cultures were prepared from newborn transgenic mice expressing green fluorescent protein (GFP) under the tyrosine hydroxylase-promoter. This fluorescent marker enabled easy identification of dopamine-containing cells (including SNc neurons). Only preliminary experiments were carried out using this preparation. GFP-positive neurons did not show the classic membrane hyperpolarization in response to dopamine. For comparison, recordings from GFP-positive SNc neurons in acute slices obtained from age-matched animals did show a typical hyperpolarizing response to dopamine. GFP-neurons from organotypic cultures also lacked the Ih current – another characteristic feature of SNc neurons in vivo or in acute brain slices. In addition, atypical responses to CNQX (blocker of NMDA receptors) and baclofen (blocker of GABAB receptors) application were identified in GFP-positive neurons. These results demonstrate that the culturing process used in this study alters the functional ‘phenotype’ of dopaminergic neurons, a change which needs to be considered in future studies using this preparation. Chronic exposure of organotypic cultures to low concentration of rotenone (50 nM) evoked a delayed increase of PI labelling indicative of cell death, however technical limitations prevented detection of PI co-localization with GFP was observed. In conclusion, this study identified several key aspects of 6 OHDA and rotenone toxicity in SNc neurons. The most significant novel findings include evidence for ROS activation of KATP channels, presumed involvement of TRPM2 channels in rotenone-induced [Ca2+]i rise, and dopamine-analogous effects of 6 OHDA. The controversial role of KATP channels in neuroprotection was addressed. Findings from this study suggest therapies targeting this channel alone would be of little benefit. The proposed involvement of TRPM2 channels in rotenone-induced [Ca2+]i overload in SNc neurons is particularly interesting as it provides a mechanism for synergism between rotenone and other factors that disrupt [Ca2+]i homeostasis.