The ketogenic diet changes metabolite levels in hippocampal extracellular fluid

Abstract

Despite successful use of the ketogenic diet (KD) for the treatment of drug-resistant epilepsy, its mechanism of action is unclear. After KD-feeding, increased plasma d-beta-hydroxybutyrate (BHB) levels appear to be important for protection against seizures. We hypothesized that the KD leads to metabolic changes in the brain, which are reflected in the hippocampal extracellular fluid (hECF). CD1 mice were fed control or KD for 2–3 weeks since weaning. In vivo microdialysis of hECF was used to measure the levels of glucose, lactate, as well as BHB under basal conditions and during 30 min stimulation with 60 mM K⁺, which was retrodialysed. The hECF BHB concentration in KD-fed mice was determined as 43.4 ± 10.1 μM using the zero-flow method and 50.7 ± 5.5 μM based on in vitro recovery. The total BHB concentration in brain homogenate from KD-fed mice was 180 nmol/g. The intracellular BHB concentration is therefore estimated to be about 3-fold higher than the extracellular level, which suggests that BHB in adolescent mouse brains may not be quickly metabolized. The basal hECF glucose concentration was 30% lower in KD-fed mice, indicating that glucose may be less important as an energy source. Lactate levels were similar in control and KD-fed mice. High potassium stimulation elevated lactate by 3–3.5-fold and decreased glucose by 40–50% in both diet groups, consistent with similar anaerobic and aerobic metabolism in both diet groups during high hippocampal activity. Overall, these data (1) defined the BHB concentration in the hippocampal extracellular fluid in KD-fed mice and (2) showed lower glucose metabolism compared to control diet-fed mice. This work will now enable other researchers to mimic the hippocampal extracellular environment in experiments aimed at deciphering the mechanisms of the KD.

Introduction

The ketogenic diet (KD) is a strict high fat, low protein, low carbohydrate diet, and is anticonvulsant in many drug-resistant epileptic children (Freeman et al., 2007, Neal et al., 2009). The KD is also effective in mice as previously reported (Uhlemann and Neims, 1972) and corroborated by us (Samala et al., 2008). The KD leads to the generation of ketone bodies d-β-hydroxybutyrate and acetoacetate) by the liver, reaching plasma levels of BHB in the millimolar range, 1–2 mM in our hands in mice (Samala et al., 2008) or 4 mM in children (Neal et al., 2009). BHB can enter the brain through the monocarboxylate transporters and can then be metabolized in the brain to acetyl-CoA (Nehlig, 2004). Several hypotheses have been proposed about the anticonvulsant mechanism of the KD, including changes in the extracellular milieu (Schwartzkroin, 1999). It was found that ketones directly reduce the spontaneous firing rate of neurons in the substantia nigra pars reticulata and this action was abolished by removal of K(ATP) channels (Ma et al., 2007). Yet, the physiological levels of extracellular ketones in the brain are still unknown. One of the main hypotheses is that the ketogenic diet is anticonvulsant by providing the brain with ketone bodies and thus additional ATP (DeVivo et al., 1978; reviewed by Schwartzkroin, 1999). So far, this hypothesis has been mainly addressed by measuring metabolite and neurotransmitter levels in tissue homogenates or in the cerebrospinal fluid. These measurements have pointed to some metabolic changes in energy and neurotransmitter metabolism (e.g. Bough et al., 2006, DeVivo et al., 1978, Yudkoff et al., 2005). However, total tissue or CSF levels of neurotransmitters and metabolites are not reflective of the neuronal microenvironment, the brain extracellular brain fluid (ECF), which ultimately determines neuronal and astrocytic energy supply and excitability. We showed previously that our 6:1 KD containing the same vitamin, mineral and antioxidant content per caloric density as the control diet, was repeatedly anticonvulsant in the 6 Hz test (Samala et al., 2008). In the present study, we determined the composition of the extracellular fluid in the hippocampus under basal conditions and after stimulation with high potassium levels. We expected to observe a reduced metabolism of glucose in the KD-fed mice, as BHB can be used as an alternative substrate for energy metabolism.