A variety of data suggest that there are alterations in glucose metabolism before and during seizures. Whether epilepsy changes the routes of glucose metabolism is not known. However, it is clear that alterations in metabolism can affect brain function. For example, the ketogenic diet (KD), which provides energy substrates that bypass glycolysis, can reduce seizure frequency. 2-deoxyglucose (2DG), which inhibits glucose uptake and inhibits phosphoglucose isomerase activity, has acute anticonvulsant activity in animal models (Garriga-Canut et al., 2006).
Fructose-1,6-diphosphate (FDP, also called fructose-1,6-bisphosphate) is an endogenous intermediate of glucose metabolism (Fig. 1) generated by metabolism along the glycolytic pathway. FDP exerts feedback inhibition of the enzyme that generates it—phosphofructokinase. Exogenously administered FDP has been studied for its ability to protect tissue during hypoxia or ischemia and facilitate recovery of that tissue after the injury (Marangos et al., 1998). It also been shown to improve recovery after ischemia and reperfusion injury to the central nervous system (Markov, 1986; Farias et al., 1990). Thus, FDP has neuroprotective activity and may alter neuronal activity.
Recently, FDP has been shown to have dose-dependent anticonvulsant activity in several models of acute seizures in laboratory animals. FDP was more effective than 2DG or the KD against seizures induced by pilocarpine, kainic acid, and pentylenetetrazol (Lian et al., 2007). Administration of FDP before the chemical convulsants slowed the onset of behavioral seizure activity and reduced the overall seizure duration and severity. FDP, given orally in the drinking water at a dose of approximately 600 mg/kg, also blocked the spontaneous seizures that appear weeks after status epilepticus was induced with pilocarpine (Lian et al., 2008). However, FDP was ineffective against spike-wave seizures induced with γ-butyrolactone (Lian et al., 2008). While there remains some debate, the anticonvulsant effect of FDP is most likely due to a direct effect on the central nervous system since intraperitoneal and oral administration of FDP results in a significant, and sustained, increase in the levels of FDP in the brain (Xu & Stringer, 2008a). Interestingly, the levels of FDP in the brain remain elevated long after levels in the blood have fallen near to the baseline levels.
There are a number of possible mechanisms for the anticonvulsant action of FDP. There is some evidence that shifts in glucose metabolism underlie the acute anticonvulsant activity of FDP and 2DG. Because lactate can be oxidized to pyruvate, administration of lactate should provide substrate for energy generation that bypasses the glycolytic pathway (Fig. 1). The effect of pretreatment with lactate on the effectiveness of FDP and 2DG was determined in the pilocarpine model of acute seizures – a model in which both FDP and 2DG are effective (Lian et al., 2007). Pretreatment with lactate completely reversed the efficacy of 2DG, suggesting that the acute anticonvulsant activity of 2DG is due to the suppression of glycolysis. The efficacy of FDP was only partially reversed by pretreatment with lactate. This suggests that FDP is not acting solely by inhibiting glycolysis, and that it has an action that is different from 2DG. Exactly how FDP alters metabolism remains to be determined.
Exogenous administration of FDP could result in the same changes in gene expression that have been described for 2DG (Garriga-Canut et al., 2006). Alterations in the ratio of NAD to NADH change the regulation of a number of genes that contribute to neuronal excitability and that are controlled by the transcription factor neural restrictive silencing factor (NRSF). Whether administration of FDP alters gene regulation in the same pathway remains to be tested. Another possible mechanism for the anticonvulsant activity of FDP is an activation of phospholipcase C and increase in its activity in the nitrogen-activated extracellular signal protein kinase/extracellular signal activated protein kinase (MEK/ERK) signaling pathway (Fahlman et al., 2002).
How might glucose metabolism in the brain be altered by the administration of FDP? Glycolysis is the conversion of glucose to FDP and then the cleavage of the 6-carbon sugar into two 3-carbon molecules, glyceraldehyde 3-phosphate (GAP) and dihydroxyacetone phosphate (DHAP), which is rapidly converted to GAP. The GAP is then converted into pyruvate, which then moves into the citric acid cycle.
The pentose phosphate pathway is an alternate pathway for the metabolism of intracellular glucose. The oxidative portion of this pathway converts glucose-6-phosphate to ribulose-5-phosphate. There is evidence that the pentose phosphate pathway is more active in the brain, particularly in astrocytes, than in other tissues of the body. In glial cells, studies of metabolism have shown that more CO2 is produced through the pentose phosphate pathway than through glycolysis. The functional significance of this is not known. However, metabolism of glucose through the pentose phosphate pathway is known to generate NADPH, which is used to reduce glutathione. Glutathione is found predominantly in astrocytes, so the increased activity of the pentose phosphate pathway matches with the presence of glutathione in these cells.
In a study of metabolism in cultures of glial cells (Kelleher et al., 1995), addition of FDP to the culture medium failed to increase lactate production in normal or hypoxic conditions, suggesting that FDP does not increase glycolysis in astrocytes. The ratio of 14C in CO2 produced by metabolism of glucose was used to determine the relative amount of metabolism through the pentose phosphate pathway compared to glycolysis. In the presence of FDP, most of the metabolized glucose enters the pentose phosphate pathway. Thus, in glial cells, glucose metabolism could be predominantly used to generate NADPH to reduce glutathione.
Glutathione is a tripeptide present in most mammalian tissues in millimolar concentrations. The reduced form of glutathione (GSH) plays an important role in cellular defense against free radicals, peroxides, and electrophilic xenobiotics. As previously noted, in the brain, glutathione is found almost exclusively in astrocytes. Most of the glutathione present in the brain is in the reduced form, suggesting that it is available to neutralize reactive species when they are generated. Interestingly, endogenous GSH has been shown to have antioxidant activity. Using L-buthionine-[S,R]-sulfoximine (BSO), an inhibitor of the synthesis of GSH, to deplete endogenous levels of GSH, Abe et al., (2000) showed that a 40–50% reduction in the basal levels of GSH decreased the latency to convulsions after administration of subcutaneous pentylenetetrazol. In addition, in cortical cultures, hypoxia reduces GSH levels, but the presence of added FDP during the hypoxic period preserves the GSH levels (Vexler et al., 2003). These data suggest that the anticonvulsant action of FDP may be related to the preservation of glutathione levels in astrocytes.
Altogether, the data suggest a hypothesis that exogenously administered FDP gets into the brain and astrocytes where it increases the flux of glucose through the pentose phosphate pathway, generating additional NADPH for the reduction of glutathione, an endogenous anticonvulsant. This hypothesis remains to be directly tested. However, some data are not consistent with this hypothesis. First, administration of FDP to glial cultures significantly increases the production of CO2 from both the glycolytic and the pentose phosphate pathways (Kelleher et al., 1995). However, the total amount of glycolysis in glial cells, even in the presence of FDP, remains quite low. Second, in the pilocarpine model of acute seizures, where FDP has a significant effect on seizure latency, severity, and duration, there is no measurable decrease in glutathione in the time frame in which FDP alters seizure activity (Xu & Stringer, 2008b). If the role of FDP were to maintain endogenous levels of glutathione at times of intense neuronal activity or stress, such as during a seizure, then one would predict a decrease in the levels of GSH early in the seizure. This finding alone does not disprove the stated hypothesis, but does suggest that the answers will not be straightforward.