Author's personal copy Diffusion of docosahexaenoic and eicosapentaenoic acids through the blood–brain barrier: An in situ cerebral perfusion study Melissa Ouellet a,b , Vincent Emond a,b , Chuck T. Chen c , Carl Julien a,b , Fanchon Bourasset d , Salvatore Oddo e , Frank LaFerla e , Richard P. Bazinet c , Fre ´de ´ ric Calon a,b, * a Faculty of Pharmacy, Laval University, Quebec, QC, Canada b Centre Hospitalier de l’Universite ´ Laval (CHUL) Research Center, Quebec, QC, Canada c Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, ON, Canada d Faculte ´ de Pharmacie, Universite ´ Paris Sud, Cha ˆtenay-Malabry, France e Department of Neurobiology and Behavior, University of California, Irvine, CA, USA 1. Introduction Recent data strongly suggest that the n-3 polyunsaturated fatty acids (n-3 PUFA), docosahexaenoic (DHA; 22:6 n-3) and eicosa- pentaenoic (EPA; 20:5 n-3), can be used as nutraceutical tools for CNS diseases. Owing to its capacity to modulate the brain fatty acid profile and to be incorporated into neuron cell membranes, DHA has been shown to modulate critical aspects of brain function, including membrane fluidity, cell signaling and gene expression (Alessandri et al., 2004; Calon et al., 2004; Jump, 2002). For example, preclinical and clinical studies have underscored beneficial effects of DHA or EPA in mood disorders, peroxisomal disorders, Parkinson’s and Alzheimer’s diseases (Alessandri et al., 2004; Bousquet et al., 2008; Calon and Cole, 2007; Calon et al., 2004; Martinez et al., 2000; Parker et al., 2006; Stahl et al., 2008). Because the brain cannot synthesize DHA or EPA from shorter n-3 PUFA, they must be supplied to the brain by the plasma (Rapoport et al., 2007; Rapoport, 2001). However, it is not clear how exactly blood-borne DHA and EPA are taken up through the blood–brain barrier (BBB) to accumulate into the brain. Both n-3 PUFA in the blood are mainly incorporated into triglycerides, phospholipids, and cholesterol esters found in plasma lipoproteins and/or bound to plasma proteins such as albumin, forming complexes too large to cross the BBB (Hamilton and Brunaldi, 2007; Kampf and Kleinfeld, 2007; Purdon et al., 1997; Rapoport, 2001; Spector, 2001). It is thought instead that DHA and EPA cross the BBB in an unbound fatty acid form (Hamilton and Brunaldi, 2007). Even if the majority of blood DHA is bound to plasma proteins, a rapid dissociation constant is likely to supply unbound DHA and EPA available for brain uptake (Demant et al., 2002; Hamilton and Brunaldi, 2007; Rapoport, 2001). On the other hand, data gathered from in vitro models of the BBB and injection of [ 13 C]- DHA in humans and rodents indicate rather that DHA is quickly taken up by the brain when incorporated into lysophosphatidyl- choline (Bernoud et al., 1999; Lagarde et al., 2001). A model of BBB transport based on the high diffusiblity of fatty acids through cytoplasmic membranes has been recently published. In this Neurochemistry International 55 (2009) 476–482 ARTICLE INFO Article history: Received 4 March 2009 Received in revised form 15 April 2009 Accepted 28 April 2009 Available online 12 May 2009 Keywords: Blood–brain barrier Docosahexaenoic acid Eicosapentaenoic acid In situ brain perfusion ABSTRACT Docosahexaenoic (DHA) and eicosapentaenoic (EPA) acids are n-3 polyunsaturated fatty acids with a therapeutic potential for CNS diseases. Here, using an in situ brain perfusion technique in mice, we show that [ 14 C]-DHA and [ 14 C]-EPA readily cross the mouse blood–brain barrier (BBB) with brain transport coefficients (Clup) of 48  3 mlg 1 s 1 and 52  4 mlg 1 s 1 , respectively. Mechanical capillary depletion of brain homogenates showed that less than 10% of [ 14 C]-DHA or [ 14 C]-EPA remained in endothelial cells of the brain vasculature, demonstrating that both molecules fully crossed the BBB. Addition of bovine serum albumin decreased the Clup of [ 14 C]-DHA to 0.6  0.3 mlg 1 s 1 , indicating that binding to albumin reduced importantly, but not totally, the passage of DHA through the BBB. The Clup of [ 14 C]-DHA or [ 14 C]-EPA was not saturable at concentration up to 100 mM, suggesting that these compounds crossed the BBB by simple diffusion. However, long-term high-DHA dietary consumption reduced the Clup of [ 14 C]-DHA to 33  6 mlg 1 s 1 (20%, p < 0.01). These results confirm that the brain uptake of DHA or EPA perfused with a physiological buffer is comparable to highly diffusible drugs like diazepam, and can be modulated by albumin binding and chronic dietary DHA intake. ß 2009 Elsevier Ltd. All rights reserved. * Corresponding author at: Centre Hospitalier de l’Universite ´ Laval (CHUL) Research Center, Room T205, 2705 Laurier Blvd, Quebec, QC, Canada G1V 4G2. Tel.: +1 418 654 2296/656 4141x48697; fax: +1 418 654 2761. E-mail address: frederic.calon@crchul.ulaval.ca (F. Calon). Contents lists available at ScienceDirect Neurochemistry International journal homepage: www.elsevier.com/locate/neuint 0197-0186/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2009.04.018