HIGH-INTENSITY PHYSICAL EXERCISE DISRUPTS IMPLICIT MEMORY IN MICE: INVOLVEMENT OF THE STRIATAL GLUTATHIONE ANTIOXIDANT SYSTEM AND INTRACELLULAR SIGNALING A. S. AGUIAR JR, a,b G. BOEMER, a,b D. RIAL, a,b F. M. CORDOVA, c G. MANCINI, c R. WALZ, b,d A. F. DE BEM, c A. LATINI, b,c R. B. LEAL, c R. A. PINHO e AND R. D. S. PREDIGER a,b * a Departamento de Farmacologia, Universidade Federal de Santa Ca- tarina, 88049-900 Florianópolis, SC, Brazil b Centro de Neurociências Aplicadas (CeNAp), Hospital Universitário, Universidade Federal de Santa Catarina, 88040-900 Florianópolis, SC, Brazil c Departamento de Bioquímica, Universidade Federal de Santa Cata- rina, 88040-900 Florianópolis, SC, Brazil d Departamento de Clínica Médica, Hospital Universitário, Univer- sidade Federal de Santa Catarina, 88040-900 Florianópolis, SC, Brazil e Laboratório de Fisiologia e Bioquímica do Exercício, Universidade do Extremo Sul Catarinense, 88806-000 Criciúma, SC, Brazil Abstract—Physical exercise is a widely accepted behavioral strategy to enhance overall health, including mental function. However, there is controversial evidence showing brain mi- tochondrial dysfunction, oxidative damage and decreased neurotrophin levels after high-intensity exercise, which pre- sumably worsens cognitive performance. Here we investi- gated learning and memory performance dependent on dif- ferent brain regions, glutathione antioxidant system, and ex- tracellular signal-regulated protein kinase 1/2 (ERK1/2), serine/threonine protein kinase (AKT), cAMP response ele- ment binding (CREB) and dopamine- and cyclic AMP-regu- lated phosphoprotein (DARPP)-32 signaling in adult Swiss mice submitted to 9 weeks of high-intensity exercise. The exercise did not alter the animals’ performance in the refer- ence and working memory versions of the water maze task. On the other hand, we observed a significant impairment in the procedural memory (an implicit memory that depends on basal ganglia) accompanied by a reduced antioxidant capac- ity and ERK1/2 and CREB signaling in this region. In addition, we found increased striatal DARPP-32-Thr-75 phosphoryla- tion in trained mice. These findings indicate an increased vulnerability of the striatum to high-intensity exercise asso- ciated with the disruption of implicit memory in mice and accompanied by alteration of signaling proteins involved in the plasticity of this brain structure. © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: physical exercise, treadmill, memory, hippocam- pus, striatum, oxidative stress. Physical exercise improves mental function and contrib- utes to neuronal plasticity (van Praag et al., 1999; Trejo et al., 2008; Figueiredo et al., 2010), as documented on aged and diseased brain (Gomez-Pinilla et al., 1998; Schmidt- Hieber et al., 2004; Aguiar et al., 2009). However, these exercise-induced outcomes are dependent on the animal’s health, brain region, parameters of physical training, such as ergometer utilized (running wheel and treadmill that permit running movement), and the intensity and duration of physical training (low- to moderate-intensity programs) (Radak et al., 2001a). These exercise parameters modulate the formation of mitochondrial reactive oxygen species (ROS) as a conse- quence of the increased oxygen intake by the tissues (Droge, 2002; Aguiar et al., 2008b) leading to oxidative signaling optimization in several tissues (Radak et al., 2001b; Aguiar Jr and Pinho, 2010), regulating cell signal- ing pathways and gene expression (Droge, 2002; Aguiar Jr and Pinho, 2010). However, depending on concentration, location and context, ROS can be either “friends” or “foes” regarding overall brain functions (Droge, 2002). Therefore, the complex neurobiology of exercise generally demon- strates U-shaped dose–response curves, where low doses are stimulatory and high doses inhibitory (Hu et al., 2009). Concerning high-intensity exercise, direct evidence of increased exercise-induced ROS production is still scarce but it is supported by oxidative stress imbalance in several tissues after exercise (Sen, 1995; Droge, 2002; Aguiar et al., 2008c). In the pro-oxidant brain (Aguiar Jr and Pinho, 2010), even the increase in downstream ROS scavenging enzymes [e.g. GSH peroxidase (GPX) and reductase (GR)] and small antioxidant molecules [e.g. glutathione (GSH)] are not sufficient to prevent mitochondrial dysfunc- tion and oxidative damage after high-intensity exercise (Rosa et al., 2007; Aguiar et al., 2008b). Moreover, there are few studies reporting the possible intracellular signal- ing mechanisms involved in the neurotrophins and mito- chondrial dysfunction (Aguiar et al., 2008a,b; Siamilis et al., 2009), interruption of hippocampal neurogenesis (Lou et al., 2008), and cognitive impairments (Grebot et al., 2003; Rhodes et al., 2003; Rosa et al., 2007; Taverniers et al., 2010) observed in response to high-intensity exercise. In this study, we used a high-intensity exercise para- digm in mice to investigate the associative mnemonic func- tion of the basal ganglia and the cognitive memory system *Correspondence to: R. D. S. Prediger, Laboratório Experimental de Doenças Neurodegenerativas, Departamento de Farmacologia, Uni- versidade Federal de Santa Catarina, Campus Trindade, 88049-900, Florianópolis, SC, Brazil. Tel: +55-48-3721-9764; fax: +55-48-3337- 5479. E-mail address: ruidsp@hotmail.com (R. D. S. Prediger). Abbreviations: BDNF, brain-derived neurotrophic factor; CREB, cAMP response element binding; DARPP, dopamine- and cyclic AMP-regu- lated phosphoprotein; ERK1/2, extracellular signal-regulated protein kinase; GPX, GSH peroxidase; GR, GSH reductase; GSH, glutathi- one; MLSS, maximal lactate steady state; ROS, reactive oxygen spe- cies. Neuroscience 171 (2010) 1216 –1227 0306-4522/10 $ - see front matter © 2010 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2010.09.053 1216