NEUROSYSTEMS Age-related changes in metabolic profiles of rat hippocampus and cortices Ve´ronique Paban, 1 Florence Fauvelle 2 and Be ´ atrice Alescio-Lautier 1 1 Universite ´ d’Aix-Marseille I, Laboratoire de Neurobiologie Inte ´ grative et Adaptative, UMR ⁄ CNRS 6149, 3 Place Victor Hugo, 13331 Marseille Cedex 03, France 2 Laboratoire de RMN, CRSSA ⁄ BMC, Unite ´ de Biophysique Cellulaire et Mole ´culaire, 24 Avenue des Maquis du Gre ´ sivaudan, 38 702 Grenoble, France Keywords: frontal cortex, hippocampus, HRMAS NMR, jMRUI, parietal cortex, rhinal cortex Abstract The time course of metabolic changes was investigated in the hippocampus and the parietal, rhinal and frontal cortices of rats from 4 to 30 months old. Samples were analysed by the solid-state high-resolution magic angle spinning nuclear magnetic resonance method. Quantification was performed with the quest procedure of jmrui software. Eighteen metabolites were identified and separated in the spectrum. Six of them were not age sensitive, in particular alanine, glutamine and lactate. In contrast, choline, glycerophosphocholine, myo-inositol, N-acetylaspartate, scyllo-inositol (s-Ins) and taurine (Tau) were notably altered over aging. Interestingly, each age group showed a specific metabolic profile. The concentration of metabolites such as Tau was altered in middle-aged rats only, whereas the s-Ins level decreased in old rats only. Most metabolites showed progressive alteration during the process of aging, which was initiated during the middle-aged period (18 months). Taken together, these results suggest that cell membrane integrity is perturbed with age. Each brain region investigated had distinctive qualitative and ⁄ or quantitative metabolic age-related features. These age-related changes would affect network connectivities and then cognitive functions. Introduction The effect of aging on the brain has been studied at the macroscopic and microscopic levels. Normal aging has been associated with some degree of neuron and volume loss (Peters & Sethares, 2002), downregulation of synaptic protein synthesis involved in structural plasticity of axons and dendrites (Blalock et al., 2003), cumulative mitochondrial damage (Wallace, 2001), inflammatory response (Godbout & Johnson, 2006), alteration of Ca 2+ regulation (Thibault et al., 1998) and reduction in DNA repair ability (Rutten et al., 2003). Some of the effects of aging on the brain are global and affect the CNS as a whole, but in many cases age-related differences are highly circumscribed and confined to specific regions, in particular the frontal and parietal lobes and the hippocampus (Rapp & Heindel, 1994; Small, 2001; Raz et al., 2004). Magnetic resonance spectroscopy (MRS) has become a powerful tool to study the chemical constituents of neural tissues, and provides information on metabolism and neurotransmitters. Altered metabolic profiles have been reported in animal models of brain alteration, for example ischaemia (Macrı ` et al., 2006) or abnormal protein accumu- lation, such as beta-amyloid precursor protein in transgenic mice (Oberg et al., 2008). Most of the 1 H MRS spectrum obtained from a brain in vivo typically consists of resonances attributed to the following key metabolites: N-acetylaspartate (NAA), myo-inositol (m-Ins), choline and its derivatives (Cho), creatine and phosphocre- atine (Cr), lactate (Lac), and glutamate ⁄ glutamine (Glu ⁄ Gln). MRS is an attractive tool, particularly for non-invasive study of metabolism in living systems, but the spectral quality is compromised by magnetic field inhomogeneities and low field strengths. Alternatively, metabolic profiles of tissue samples obtained from biopsy specimens or animals that have been killed can be studied in detail using ex vivo high- resolution nuclear magnetic resonance (NMR), because of its higher sensitivity and the improved resonance dispersion at ultra-high field strengths. Moreover, the solid-state high-resolution magic angle spinning (HRMAS) NMR method uses intact tissue samples and does not require soluble extraction (Cheng et al., 1997; Sitter et al., 2002; Martı ´nez-Bisbal et al., 2004). Therefore, it offers the advanta- ges of retaining tissue morphology and eliminating contamination associated with the extraction procedure. Multivariate statistical analyses, such as partial least squares discriminant analysis (PLS- DA), can then be performed with these data, leading to the ‘metabolic profiling’ used today for classification and prediction. Ex vivo HRMAS NMR has been successfully applied in various tissues and diseases (Cheng et al., 2000; Gonza ´lez et al., 2000; Sitter et al., 2002; Tzika et al., 2002; Wang et al., 2006; Mao et al., 2007). In particular, using 1 H HRMAS NMR, Tsang et al. (2005) have established metabolic profiles in the hippocampus, the frontal cortex, the Correspondence: Dr V. Paban, as above. E-mail: veronique.paban@univ-provence.fr Received 10 September 2009, accepted 4 January 2010 European Journal of Neuroscience, Vol. 31, pp. 1063–1073, 2010 doi:10.1111/j.1460-9568.2010.07126.x ª The Authors (2010). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience