Ketogenic diet (high-fat, adequate protein, and low carbohydrate diet) and ketone supplementation (exogenous ketone supplement and/or MCT) may evoke nutritional ketosis (Kwiterovich et al., 2003; Poff et al., 2015; Kesl et al., 2016), which enhances conversion of ketone bodies such as beta-hydroxybutyrate and acetoacetate (AcAc) into acetyl-CoA to support ATP production (Yudkoff et al., 2007). Under this condition, the cells utilize ketone bodies as an alternative fuel to enhance brain energy metabolism. The classical KD and variants have proven beneficial in treating epilepsy and other disorders by decreasing seizure activity (Lefevre and Aronson, 2000; Kossoff et al., 2011), likely by modulation of GABAergic, glutamatergic, and adenosinergic processes (Yudkoff et al., 2007; Masino et al., 2011; Rho, 2017). The exact mechanism(s) of the KD/nutritional ketosis on epilepsy are largely unknown and remain an area of intense investigation and interest within the pharmaceutical industry.

Genetically absence epileptic WAG/Rij rats spontaneously generate absence-like seizures, and manifest spike-wave asymmetric discharges (SWDs) on the EEG. A typical SWD contains a train of asymmetric spikes and slow waves starting and ending with sharp spikes (frequency: 7–11 Hz and duration 1–30 s in WAG/Rij rats) (Coenen and Van Luijtelaar, 2003). It was demonstrated that the hyperexcitable neurons of cortical focus in the somatosensory cortex initiate SWDs (cortical focus theory) (Meeren et al., 2002).

It has been demonstrated, that oral administration of βHB and AcAc in their free acid form is difficult to produce sustained ketosis, in contrast to ketone ester and ketone salt supplementation (D’Agostino et al., 2013; Kesl et al., 2016). It is well-known that adenosinergic system may be involved not only in the alleviating effect of KD/ketosis on epileptic seizures (mainly by means of A₁Rs), but also in the modulation of absence epilepsy mechanisms (D’Alimonte et al., 2009; Masino et al., 2011; Kovács et al., 2014; Kim et al., 2016). Thus, we addressed in the present study whether (i) sub-chronically (for 7 days) applied exogenous ketone supplements (KE; KS + MCT: KSMCT) by intragastric gavage can modulate absence epileptic activity and (ii) inhibition of A₁Rs by its specific antagonist can evoke changes in seizure frequency in WAG/Rij rats. At first, to investigate the influence of the ketone supplements on absence epileptic activity, we administered KE (2.5 g/kg/day) and KSMCT (2.5 g/kg/day) alone by gavage. We measured not only glucose, but also βHB levels in the blood collected from tail veins to establish the ketonemia-inducing effect of ketone supplementation as ketonemia is the clinical hallmark associated with metabolic management of seizures (Freeman et al., 1994). As we have no data on the putative effects of intraperitoneally (i.p.) injected specific A₁R antagonist DPCPX on absence epileptic activity in WAG/Rij rats, we also investigated the effect of two doses (i.p. 0.2 and 0.5 mg/kg alone) on SWD number. Subsequently, to decide whether A₁Rs can modify ketone supplements-evoked changes in SWD number, we applied a lower dose (i.p. 0.2 mg/kg) of DPCPX in combination with KSMCT (2.5 g/kg/day, gavage). We hypothesized that (i) exogenous ketone supplementation evokes ketosis, and decreases absence epileptic seizures, and (ii) A₁R blockade modifies the influence of exogenous ketone supplementation on SWD number. Indeed, we demonstrated that exogenous ketone supplements increased blood βHB levels and decreased SWD number when given sub-chronically and inhibition of A₁Rs abolished the exogenous ketone supplements-evoked changes in SWD number in WAG/Rij rats.

The current study is the first to demonstrate that the sub-chronic application of exogenous ketone supplements (both KE and KSMCT) by intragastric gavage decreased absence epileptic activity, and was abolished by pharmacologic inhibition of A₁Rs in WAG/Rij rats.

It has been demonstrated that different types of ketogenic diets represent alternative treatment methods (metabolic-based therapies) for a broad range of seizure disorders, in which decreased epileptic activity may be related to elevations in serum ketone bodies, reduced glucose levels, suppression of insulin signaling as well as different neurotransmitter systems, such as GABAergic and adenosinergic system in different types of epilepsies (Ross et al., 1985; Van Delft et al., 2010; Sharma et al., 2015; Rho, 2017). Although the relationship between KD-evoked ketosis and seizure arrest is still unclear, it is widely accepted that high level ketosis has an important role in antiseizure/anticonvulsant effects of KD: βHB may alter metabolic pathways to increase both GABA and adenosine, which products have anticonvulsant and seizure suppressive effects (Gilbert et al., 2000; Bough et al., 2002; Neal et al., 2009; Van Delft et al., 2010; Sharma et al., 2015). Nevertheless, many patients have difficulty with compliance to the KD because of gastrointestinal symptoms including vomiting, diarrhea and abdominal pain (Ross et al., 1985; Veggiotti and De Giorgis, 2014). Therefore, testing new, non-pharmacological therapeutic strategies, such as application of exogenous ketone supplements instead of strict KD (Lee et al., 2016), which supplements may sustain therapeutic levels of ketosis are needed for some individuals. Exogenous ketone supplementation may generate corresponding beneficial therapeutic effect on epileptic seizures similar to the KD because (i) exogenous ketone supplementation-evoked influences may be similar to effects of KD (such as increased adenosine level), whose actions may mediate the effects of ketone supplementation on different types of epilepsies, (ii) among others, ketone bodies (e.g., βHB) may be responsible for the mediating effects of KD/ketosis as well as ketone supplementation on epileptic seizures due to altering Krebs cycle intermediates and GABAergic activity, and (iii) ketone supplementation (both KE and KSMCT) increase blood βHB levels and evoke nutritional ketosis at levels known to preserve brain homeostasis (Figures 3C,D) (Kwiterovich et al., 2003; Yudkoff et al., 2007; Masino et al., 2011; Poff et al., 2015; Ari et al., 2016; Kesl et al., 2016; Rho, 2017). Moreover, tolerability of oral/intragastric exogenous ketone supplementation in inducing nutritional ketosis is consistent with previous studies (Kesl et al., 2016), and provides a rationale for circumventing the dietary restriction of the KD or to enhance ketogenesis in less restrictive modified versions.

It was previously demonstrated that KD/ketosis may be effective against absence epileptic activity (Ross et al., 1985; Groomes et al., 2011; Thammongkol et al., 2012; Clemens et al., 2013; Kim et al., 2016; Lee et al., 2016). Our results suggest that not only KD, but also sub-chronically applied ketone supplements KE and KSMCT, administered by gavage, can increase βHB levels and decrease SWD number in freely moving WAG/Rij rats (Figures 3A–D). Moreover, SWD number and βHB level returned to the control level after both KE and KSMCT treatments (on the PT days). Thus, our results suggest that after ketone supplementation the absence epileptic activity can decrease in correlation with the increase in βHB level, which strengthens the support for nutritional ketosis as a means to suppress absence epileptic seizures (Ross et al., 1985; Groomes et al., 2011; Thammongkol et al., 2012; Clemens et al., 2013; Kim et al., 2016; Lee et al., 2016).

Moreover, it has been demonstrated that KD and ketone bodies (direct and/or indirect manner) among others may (i) increase adenosine signaling by both decreased expression/activity of adenosine kinase (which enzyme may metabolize adenosine) and enhanced βHB metabolism and, as a consequence, by A₁Rs and (ii) enhance the inhibitory GABAergic effects in the brain by GABA_A receptors (Yudkoff et al., 2007; Masino et al., 2011, 2012; Sharma et al., 2015; Rho, 2017), which systems (GABAergic and adenosinergic) may have a role in the regulation of absence epileptic activity in WAG/Rij rats (Peeters et al., 1989; D’Alimonte et al., 2009; Kovács et al., 2015). It was demonstrated previously that activation of A_2ARs and GABA_A receptors may evoke increases in SWD number in WAG/Rij rats (D’Alimonte et al., 2009; Kovács et al., 2015; Lakatos et al., 2016), which results suggest that the modificatory effects of GABA_A receptors and A_2ARs in ketone supplementation-evoked decrease in absence epileptic activity can be excluded. Density of A₁Rs is uneven in the central nervous system, expression/activity of A₁Rs is decreased in the somatosensory cortex (focus) and thalamus in presymptomatic WAG/Rij rats, A₁Rs has a role in the modulation of different types of epilepsies, and KDs may exert their effect on epileptic activity by means of A₁Rs (D’Alimonte et al., 2009; Kovács et al., 2011b, 2014; Masino et al., 2011). Furthermore, (i) SWDs triggered by excessive hyperexcitability in the cortical focus of absence epilepsy genesis (Meeren et al., 2002; Coenen and Van Luijtelaar, 2003), (ii) brain areas, implicated in absence epilepsy genesis, such as somatosensory cortex, contain A₁Rs (Cremer et al., 2009; Kovács et al., 2011b) and (iii) not only KD, but also ketone supplementation may modulate adenosinergic system by increased adenosine levels (Masino et al., 2011; Sharma et al., 2015; Rho, 2017). It has been demonstrated previously that inhibition of A₁Rs by DPCX induces seizure-like bursting activity in vitro (Thümmler and Dunwiddie, 2000) and A₁R knockout mouse exhibits electrographic seizures in vivo (Masino et al., 2011). Moreover, in accordance with cortical focus theory of absence epilepsy (Meeren et al., 2002; Coenen and Van Luijtelaar, 2003), activation of inhibitory A₁Rs, which may decrease the excessive hyperexcitability in the somatosensory cortex (cortical focus) directly and indirectly by synaptic inhibition with reducing Ca²⁺ influx through voltage-dependent Ca²⁺ channels, and activation of both ATP-sensitive potassium (K_ATP) channels and G protein-gated inwardly rectifying potassium channels hyperpolarizing neuronal membranes (Kovács et al., 2014; Lusardi et al., 2015; Rho, 2017), may decrease absence epileptic activity. In addition, ketosis/ketone bodies may indirectly activate K_ATP channels (Ma et al., 2007). All of these inhibitory effects may be implicated in the modulatory effect of A₁Rs on KE and KSMCT treatment-evoked changes in epileptic seizures (Figure 4C), and/or strengthened the aggravating effect of higher dose of DPCPX on SWD number (Figure 4B) in WAG/Rij rats.

As to the antiepileptic effect of ketone supplementation, additional mechanisms of action similar to KD cannot be excluded (Rho, 2017). For example, depending on the diet formulation, KD/ketosis may result in (i) increased level of two potentially anti-convulsant fatty acids decanoic acid and octanoic acid, (ii) modified synaptic vesicle recycling by means of βHB (Chang et al., 2013; Hrynevich et al., 2016), (iii) appearance of adenosine-associated epigenetic mechanism(s) (Lusardi et al., 2015), (iv) increased threshold for calcium-induced mitochondrial permeability transition (Kim et al., 2015), and (v) decrease in extracellular glutamate release (Juge et al., 2010), which effects may also alleviate epileptic seizures. For example, it has been demonstrated that MCT may evoke anti-seizure effect by decanoic acid through AMPA receptor inhibition in vitro (Chang et al., 2016). Consequently, several mechanisms of action might be involved in KD- and ketone supplementation-evoked antiepileptic/antiseizure effects. However, our data are not sufficient to interpret all of the putative mechanisms, by which (i) exogenous ketone supplementation exerts its antiepileptic effects and (ii) A₁Rs modulate effects of ketone supplements on seizure activity.

The administration of exogenous ketone supplements without dietary restrictions (standard rodent diet) may be an effective alternative way to the KD not only to reach and maintain nutritional ketosis, but also to reduce epileptic activity, and which can be reversed by inhibition of A₁Rs. Our results on WAG/Rij rats and modulation of ketone supplementation-evoked effects on absence epileptic activity provided an opportunity to investigate some new aspects of the pathophysiology of absence epilepsy, several signaling pathways, which are still poorly understood, influence of ketone supplementation on epileptic activity and the correlation between antiepileptic activity of ketone supplementation and their potential to induce ketosis. Our results may contribute to the development of effective antiepileptic strategies such as ketone supplementation not only against absence epilepsy (e.g., childhood absence epilepsy), but also for other therapy-resistant types of epilepsies, which may afford long-lasting seizure protection. Therefore, further studies are needed to reveal the exact molecular and neuropharmacological effects of exogenous ketone supplementation-induced decrease in absence epileptic activity and to disclose the putative link(s) between exogenous ketone supplements-evoked and adenosine-generated effects on seizure activity.

ZK: conception and design of experiments, data collection, interpretation of data, and writing manuscript, DD: interpretation of data, writing manuscript, AD: data analysis, writing manuscript, CA: conception and design of experiments, writing manuscript.

InternationalPatent # PCT/US2014/031237, University of South Florida, D. P. D’Agostino, S. Kesl, P. Arnold, “Compositions and Methods for Producing Elevated and Sustained Ketosis.” The other authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.