Mitochondrial dysfunction and lipid peroxidation in rat frontal cortex by chronic NMDA administration can be partially prevented by lithium treatment

Abstract

Chronic N-methyl-d-aspartate (NMDA) administration to rats may be a model to investigate excitotoxicity mediated by glutamatergic hyperactivity, and lithium has been reported to be neuroprotective. We hypothesized that glutamatergic hyperactivity in chronic NMDA injected rats would cause mitochondrial dysfunction and lipid peroxidation in the brain, and that chronic lithium treatment would ameliorate some of these NMDA-induced alterations. Rats treated with lithium for 6 weeks were injected i.p. 25 mg/kg NMDA on a daily basis for the last 21 days of lithium treatment. Brain was removed and frontal cortex was analyzed. Chronic NMDA decreased brain levels of mitochondrial complex I and III, and increased levels of the lipid oxidation products, 8-isoprostane and 4-hydroxynonenal, compared with non-NMDA injected rats. Lithium treatment prevented the NMDA-induced increments in 8-isoprostane and 4-hydroxynonenal. Our findings suggest that increased chronic activation of NMDA receptors can induce alterations in electron transport chain complexes I and III and in lipid peroxidation in brain. The NMDA-induced changes may contribute to glutamate-mediated excitotoxicity, which plays a role in brain diseases such as bipolar disorder. Lithium treatment prevented changes in 8-isoprostane and 4-hydroxynonenal, which may contribute to lithium's reported neuroprotective effect and efficacy in bipolar disorder.

Introduction

Excessive glutamate signaling can induce neuronal dysfunction and cell death by producing excitotoxicity, a process that is characterized by a large influx of calcium, resulting in a cascade of events involving disruption of calcium-dependent cellular pathways, mitochondrial dysfunction, production of oxidative stress and activation of apoptosis (Plitman et al., 2014). Glutamate excitotoxicity is hypothesized to be an important contributor in many neurodegenerative and neuropsychiatric disorders, including Parkinson's disease, Alzheimer's disease, Huntington's disease, schizophrenia, and bipolar disorder (BD) (Mehta et al., 2013, Plitman et al., 2014).

Glutamate signaling occurs through ionotropic or metabotropic glutamate receptors. The N-methyl-d-aspartate (NMDA) receptor is an ionotropic glutamate receptor that causes an influx of calcium upon activation, giving it its convulsant properties in high doses due to a dramatic increase in action potentials which can be seen as spike trains in EEG recordings (Corner et al., 2002, Ormandy et al., 1991). Activation of NMDA receptors can also result in the activation of various signaling pathways that are calcium-dependent, including activation of protein kinase C (Fukunaga et al., 1992) and release of arachidonic acid from membrane phospholipid (Dingledine et al., 1999, Lazarewicz et al., 1990, Lazarewicz et al., 1988, Weichel et al., 1999). Interestingly, chronic treatment of rats with mood stabilizers commonly used in BD, including lithium, at therapeutically relevant doses decreased alterations in brain arachidonic acid signaling produced by acute NMDA administration (Rapoport, 2014), suggesting that lithium may be able to ameliorate the effects of excitotoxicity produced by chronic hyperactivation of glutamatergic signaling. In this regard, a model for chronic excitotoxicity has been developed, involving daily i.p. injection of NMDA in rats at daily doses ranging from 25 to 130 mg/kg (Ormandy et al., 1991). The lowest dose produced brief spike trains 5–10 s duration within 10 min, which soon disappeared, suggesting that while the calcium influx produced by the dose is sufficient to produce excitation, it is not sufficient to produce convulsions (Ormandy et al., 1991). Pre-treatment with lithium did not potentiate the severity of EEG changes caused by chronic NMDA. We thought that the low dose NMDA regimen would be of interest in this study.

Excessive calcium influx caused by hyperactivation of NMDA receptors may cause oxidative stress by increasing calcium entry into the mitochondria (Clay et al., 2010, Gao et al., 2007, Giorgi et al., 2011, Hama-Tomioka et al., 2012, Lemasters et al., 2009). In turn, excessive mitochondrial calcium can disturb the electron transfer process, producing reactive oxygen species (ROS) through leakage of electrons from the electron transport chain (ETC), particularly complexes I and III. Furthermore, excess calcium can cause activation of nitric oxide synthase, resulting in increased production of nitric oxide that can form peroxynitrite, a reactive nitrogen species (RNS) (Girouard et al., 2009, Hama-Tomioka et al., 2012). RNS and ROS produced in this process can react with lipids, proteins and DNA to cause various structural and functional modifications (Barzilai and Yamamoto, 2004, Beal, 2002, Clay et al., 2010, Fariss et al., 2005). Lipid peroxidation, which can be produced by ROS, was found to correlate with white matter abnormalities in patients with BD (Versace et al., 2014). Lipid peroxidation may be a particularly relevant target to excitotoxicity produced by hyperactivation of NMDA receptors, since chronic NMDA treatment was found to increase the release of arachidonic acid, which is a favorable substrate for lipid peroxidation (Axelrod, 1990, Bosetti et al., 2002, Chang et al., 1996, Felder et al., 1990, Pellerin and Wolfe, 1991).

Therefore, the aim of this study was to examine whether excessive and chronic activation of NMDA receptors could impair functioning of the mitochondrial ETC and lipid peroxidation in the brain. First, we examined whether alterations in the mitochondrial ETC and products of lipid peroxidation occurred in the frontal cortex of rats that were given chronic (21 days) daily subconvulsive intraperitoneal (i.p.) injection of NMDA. Second, we determined if chronic lithium, at a therapeutically relevant dose that is used to treat BD, could decrease these alterations if they occurred.