The hippocampal dentate gyrus is essential for generating contextual memories of fear and drug-induced reward V. Hernández-Rabaza a,b , L. Hontecillas-Prieto c , C. Velázquez-Sánchez a , A. Ferragud a , A. Pérez-Villaba a , A. Arcusa a , J.A. Barcia b , J.L. Trejo c , J.J. Canales a, * a Biopsychology & Comparative Neuroscience Unit, Cavanilles Institute (ICBiBE), University of Valencia-General Foundation, Poligono de la Coma s/n, Paterna, 46980 Valencia, Spain b Neuroregeneration & Neural Repair Unit, Centro de Investigación Príncipe Felipe, Valencia, Spain c Cajal Institute, Consejo Superior de Investigaciones Científicas, Madrid, Spain article info Article history: Received 23 April 2008 Revised 17 June 2008 Accepted 18 June 2008 Available online 29 July 2008 Keywords: Dentate gyrus Context Cocaine Place preference Fear conditioning abstract The hippocampus is believed to play a role in processing information relative to the context in which emotionally salient experiences occur but evidence on the specific contribution of the hippocampal den- tate gyrus (DG) to these processes is limited. Here, we have used two classical behavioral paradigms to study the participation of the dorsal DG in context-conditioned reward and context-conditioned fear. Rats received intra-hippocampal vehicle or colchicine injections (4 lg/ll solution; 0.2 ll injections at 10 sites) that damaged the DG but spared other hippocampal subfields. In the first experiment, we used a place conditioning procedure pairing cocaine exposure (20 mg/kg, i.p.) with a specific context and vehi- cle treatment with another. While rats with sham lesions exhibited preference for the cocaine-paired context following conditioning, rats with lesions of the DG showed no evidence of cocaine-induced place preference. In the second experiment, rats with sham or colchicine lesions received a foot shock in a given context and conditioned freezing was measured upon reexposure to the shock-paired context (2, 24, 48 and 96 h after conditioning). Rats with sham lesions exhibited high levels of conditioned freezing when exposed to the conditioning context but rats with lesions of the DG showed impaired conditioning, behaving as controls that had experienced shock in a different context. These observations indicate that the integrity of the DG is essential for establishing a coherent representation of the context to which emo- tional experiences, either hedonic or aversive, are bound. Ó 2008 Elsevier Inc. All rights reserved. 1. Introduction The reexposure to environments in which strong emotional events have taken place can evoke conditioned physiological and psychological responses akin to those originally experienced. Drug abusers can experience craving and renewed desire for consuming drugs if exposed to distinct cues or locations previously associated with drugtaking (See, 2002; Self, 1998; Weiss, 2005). Similarly, environments associated with aversive events can induce fear-re- lated autonomic responses and the recollection of fear-related memories (Grillon, 2002; Hermans, Craske, Mineka, & Lovibond, 2006). Several lines of evidence support a role for the medial tem- poral lobe systems, and particularly for the hippocampal network, in at least some forms of contextual learning (Holland & Bouton, 1999; Ross & Slotnick, 2007; Rudy, Barrientos, & O’Reilly, 2002). First, the hippocampus has been implicated in drug-induced re- ward and context-induced drug-seeking. Lesions or functional inactivation of the dorsal hippocampus impaired in rats the acqui- sition and expression of cocaine-induced conditioned place prefer- ence (CPP) (Meyers, Zavala, & Neisewander, 2003; Meyers, Zavala, Speer, & Neisewander, 2006), a model of context-conditioned re- ward. Moreover, after extinction of cocaine self-administration, inactivation of the dorsal hippocampus impaired context-induced reinstatement of cocaine-seeking (Fuchs et al., 2005). Conversely, stimulation of the hippocampus at theta frequency caused relapse to cocaine-seeking in the drug-taking context, suggesting that the hippocampus might encode an association between context and drug-induced hedonic states (Vorel, Liu, Hayes, Spector, & Gardner, 2001). Second, the participation of the hippocampus in the making of associations between context and aversive stimuli has been am- ply debated (Anagnostaras, Gale, & Fanselow, 2001; Frankland, Cestari, Filipkowski, McDonald, & Silva, 1998; Gewirtz, McNish, & Davis, 2000; Rudy, Huff, & Matus-Amat, 2004). Several observa- tions in the rat suggested that hippocampal lesions do not alter (Gisquet-Verrier, Dutrieux, Richer, & Doyere, 1999) or even poten- tiate context conditioning (Winocur, Rawlins, & Gray, 1987), and conflicting findings were reported in the human as well (Grillon, 1074-7427/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.nlm.2008.06.008 * Corresponding author. Fax: +34 963 543 670. E-mail address: juan.canales@uv.es (J.J. Canales). Neurobiology of Learning and Memory 90 (2008) 553–559 Contents lists available at ScienceDirect Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme