NEONATAL ANOXIA IN RATS: HIPPOCAMPAL CELLULAR AND SUBCELLULAR CHANGES RELATED TO CELL DEATH AND SPATIAL MEMORY S. H. TAKADA, a,b * C. A. DOS SANTOS HAEMMERLE, a L. C. MOTTA-TEIXEIRA, c A. V. MACHADO-NILS, c V. Y. LEE, a L. F. TAKASE, d R. J. CRUZ-RIZZOLO, e A. H. KIHARA, b G. F. XAVIER, c I.-S. WATANABE a AND M. I. NOGUEIRA a a Departamento de Anatomia, Instituto de Cie ˆncias Biome ´dicas, Universidade de Sa ˜o Paulo, Sa ˜o Paulo, Brazil b Nu ´ cleo de Cognic ¸a ˜o e Sistemas Complexos, Centro de Matema ´tica, Computac ¸a ˜o e Cognic ¸a ˜o, Universidade Federal do ABC, Santo Andre ´, Brazil c Departamento de Fisiologia, Instituto de Biocie ˆ ncias, Universidade de Sa ˜o Paulo, Sa ˜o Paulo, Brazil d Departamento de Morfologia e Patologia, Centro de Cie ˆ ncias Biolo ´gicas e da Sau ´ de, Universidade Federal de Sa ˜o Carlos, Sa ˜o Carlos, Brazil e Departamento de Cie ˆncias Ba ´sicas, Campus de Arac ¸ atuba, Universidade Estadual Paulista, Arac ¸ atuba, Brazil Abstract—Neonatal anoxia in rodents has been used to understand brain changes and cognitive dysfunction follow- ing asphyxia. This study investigated the time-course of cel- lular and subcellular changes and hippocampal cell death in a non-invasive model of anoxia in neonatal rats, using Terminal deoxynucleotidyl transferase-mediated dUTP Nick End Labeling (TUNEL) to reveal DNA fragmentation, Fluoro-Jade Ò B (FJB) to show degenerating neurons, cleaved caspase-3 immunohistochemistry (IHC) to detect cells undergoing apoptosis, and transmission electron micros- copy (TEM) to reveal fine ultrastructural changes related to cell death. Anoxia was induced by exposing postnatal day 1 (P1) pups to a flow of 100% gaseous nitrogen for 25 min in a chamber maintained at 37 °C. Control rats were similarly exposed to this chamber but with air flow instead of nitro- gen. Brain changes following anoxia were evaluated at post- natal days 2, 14, 21 and 60 (P2, P14, P21 and P60). In addition, spatial reference memory following anoxia and control treatments was evaluated in the Morris water maze, starting at P60. Compared to their respective controls, P2 anoxic rats exhibited (1) higher TUNEL labeling in cornus ammonis (CA) 1 and the dentate gyrus (DG), (2) higher FJB-positive cells in the CA2–3, and (3) somato-dendritic swelling, mitochondrial injury and chromatin condensation in irregular bodies, as well as other subcellular features indi- cating apoptosis, necrosis, autophagy and excitotoxicity in the CA1, CA2–3 and DG, as revealed by TEM. At P14, P21 and P60, both groups showed small numbers of TUNEL- positive and FJB-positive cells. Stereological analysis at P2, P14, P21 and P60 revealed a lack of significant differ- ences in cleaved caspase-3 IHC between anoxic and control subjects. These results suggest that the type of hippocam- pal cell death following neonatal anoxia is likely indepen- dent of caspase-3 activation. Neonatal anoxia induced deficits in acquisition and performance of spatial reference memory in the Morris water maze task. Compared to control subjects, anoxic animals exhibited increased latencies and path lengths to reach the platform, as well as decreased searching specifically for the platform location. In contrast, no significant differences were observed for swimming speeds and frequency within the target quadrant. Together, these behavioral results indicate that the poorer perfor- mance by anoxic subjects is related to spatial memory defi- cits and not to sensory or motor deficits. Therefore, this model of neonatal anoxia in rats induces hippocampal changes that result in cell losses and impaired hippocampal function, and these changes are likely related to spatial memory deficits in adulthood. Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: apoptosis, hippocampal cell death, spatial memory, neonatal hypoxia, asphyxia, transmission electron microscopy. INTRODUCTION Neonatal anoxia is a major cause of brain injury at birth. It affects 0.1–0.3% of full-term infants (Kurinczuk et al., 2010) and approximately 60% of low-weight premature infants, thus constituting a major public health concern (Vannucci et al., 1999; Suguihara and Lessa, 2005). Long-lasting cognitive consequences of neonatal anoxia include deficits in spatial learning and memory (Buwalda et al., 1995; Cannon et al., 2002; Caputa et al., 2005; Rogalska et al., 2006), which may be related to hippocam- pal damage (Leuner et al., 2006; Winocur et al., 2006). The mammalian brain is extremely vulnerable to oxygen deprivation. Such episodes trigger cascades of biochemical events, including disruption of energy metabolism, acid–base imbalance, accumulation of reactive oxygen species and excitatory amino acids in http://dx.doi.org/10.1016/j.neuroscience.2014.08.054 0306-4522/Ó 2014 IBRO. Published by Elsevier Ltd. All rights reserved. * Correspondence to: S. H. Takada, Po´s-doutoranda, Laborato´rio de Neurogene´tica, Centro de Matema´tica, Computac¸a˜o e Cognic¸a˜o, Universidade Federal do ABC, Brazil. E-mail address: silviahonda@usp.br (S. H. Takada). Abbreviations: ANOVA, analysis of Variance; CA, cornus ammonis; DAPI, 4 0 ,6-diamidino-2-phenylindole; DG, dentate gyrus; FJB, Fluoro- Jade Ò B; IHC, immunohistochemistry; ITI, intertrial interval; PBS, phosphate-buffered saline; TEM, transmission electron microscopy; TUNEL, Terminal deoxynucleotidyl transferase-mediated dUTP Nick End Labeling. Neuroscience 284 (2015) 247–259 247