Proceedings of The Physiological Society

Kings College London (2005) J Physiol 565P, PC155


Effect of supplementation with docosahexaenoic acid on P-glycoprotein function in an in vitro model of the brain endothelium

Mallick, AS ; Begley, DJ ; Taylor, JA ; Abbott, NJ ;

1. Centre for Neuroscience Research, King's College London, London, United Kingdom.

The blood-brain barrier (BBB) is formed by cerebrovascular endothelial cells, under the inductive influence of astrocytic glia (Abbott, 2002). It maintains a homeostatic brain environment, by regulating molecular traffic between blood and brain. Closely apposed astrocytes provide fatty acids (FA) to the endothelium (Bernoud et al. 1998); this may modulate cell membranes and the expression of BBB features including tight junctions and transport proteins. P-glycoprotein (Pgp) is a lipid-associated efflux transporter present in the BBB, which can prevent accumulation of toxic lipophilic agents (Romsicki and Sharom, 1999). We hypothesised that supplementation with the FA docosahexaenoic acid (DHA) would increase BBB Pgp activity in the rat brain endothelial cell line (RBE4); previous studies from our lab were contradictory, showing that 18μM and 73μM DHA in medium with 10% fetal calf serum (FCS) inhibited RBE4 Pgp function (transport assay), but increased Pgp expression (Western blot). Control cells were cultured in 96 well plates in standard RBE4 medium with 10% FCS (Begley et al. 1996). For DHA supplementation (over 3 days), the serum level was reduced to 2% FCS, to lower the basal level of DHA, as FCS contains DHA. The effect of DHA on cellular Pgp activity was assessed by measuring uptake (volume of distribution, Vd) of the Pgp substrate [3H]-colchicine over 30 min (Begley et al. 1996). DHA supplementation in the range 0.25-8μM caused a biphasic effect on Pgp activity. All the results were significant when compared to the 2%FCS control (ANOVA plus Tukey-Kramer post-test, P<0.001), with the Pgp activity maximal at 2μM DHA (Vd reduced from 39.03±0.08 to 6.55±0.26 μl/mg protein, mean ± S.E.M., n=6 wells for DHA, 12 for controls). Over a greater DHA range (1-64μM), only 1μM DHA increased Pgp function significantly (P<0.05). In contrast to previous experiments, our DHA source was stored under nitrogen to reduce peroxidation. In an attempt to explain previous results, we stored the DHA under oxidising conditions (without nitrogen). Repeating the previous study (18μM and 73μM DHA) with DHA stored under nitrogen showed no significant increase in Vd. However DHA stored under oxidising conditions did significantly increase Vd at both 18μM and 73μM DHA (P<0.0001), suggesting that previous findings were due to the generation of toxic metabolites by DHA peroxidation.

Where applicable, experiments conform with Society ethical requirements