Adenosine Receptors Modulate the Exogenous Ketogenic Supplement-Evoked Alleviating Effect on Lipopolysaccharide-Generated Increase in Absence Epileptic Activity in WAG/Rij Rats

It has been previously demonstrated that KEKS food containing exogenous ketogenic supplement ketone salt (KS) and ketone ester (KE) decreased the lipopolysaccharide (LPS)-generated increase in SWD (spike-wave discharge) number in Wistar Albino Glaxo/Rijswijk (WAG/Rij) rats, likely through ketosis. KEKS-supplemented food-generated ketosis may increase adenosine levels, and may thus modulate both neuroinflammatory processes and epileptic activity through adenosine receptors (such as A1Rs and A2ARs). To determine whether these adenosine receptors are able to modify the KEKS food-generated alleviating effect on LPS-evoked increases in SWD number, an antagonist of A1R DPCPX (1,3-dipropyl-8-cyclopentylxanthine; 0.2 mg/kg) with LPS (50 µg/kg) and an antagonist of A2AR SCH58261 (7-(2-phenylethyl)-5-amino-2-(2-furyl)-pyrazolo-[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine; 0.5 mg/kg) with LPS were co-injected intraperitoneally (i.p.) on the ninth day of KEKS food administration, and their influence not only on the SWD number, but also on blood glucose, R-beta-hydroxybutyrate (R-βHB) levels, and body weight were measured. We showed that inhibition of A1Rs abolished the alleviating effect of KEKS food on LPS-generated increases in the SWD number, whereas blocking A2ARs did not significantly modify the KEKS food-generated beneficial effect. Our results suggest that the neuromodulatory benefits of KEKS-supplemented food on absence epileptic activity are mediated primarily through A1R, not A2AR.


Introduction
Digestion of administered exogenous ketogenic supplements (EKSs), such as ketone salts and ketone esters (KSs and KEs, respectively) can liberate ketone bodies, such as beta-hydroxybutyrate (βHB) and acetoacetate (AcAc) [1]. These molecules can transport to the bloodstream and are able to evoke an increase in ketone body levels (therapeutic ketosis) without dietary restriction, such as a ketogenic diet [2,3]. After that, ketone bodies may be transported to different cells (such as neurons and glia) through monocarboxylate transporters and can be readily used as an energy source in the mitochondria [4,5]. Indeed, adaptation of the brain metabolism to ketone bodies as a main source of fuel was demonstrated [6,7]. It is widely accepted that ketone bodies (e.g., βHB), are not only a potential were implanted epidurally above the frontal cortex and parietal cortex (based on stereotaxic coordinates: AP 2.0 mm, L 2.1 mm and AP −6.5 mm, L 2.1 mm, respectively) [32]. Ground screw electrodes and reference electrodes were implanted above the cerebellar cortex [33]. Electrodes were soldered to a ten-pin socket. Finally, dentacrylate cement (Ivoclar, Liechtenstein) was used to fix the electrodes to the skull. To alleviate post-operative pain, lidocaine ointment (5%; EGIS, Hungary) was used.

EEG Recording and Evaluation
Rats (n = 33) were allowed to recover from surgery for 2 weeks (recovery period) before the recording of EEG. EEG recording was carried out by the Bioamp4 differential amplifier (Supertech Ltd., Pécs, Hungary) and the CED 1401 mkII device (Cambridge Electronic Design Ltd., UK, Cambridge) between 1.30 p.m. and 4.00 p.m. (sampling rate: 500 Hz; bandwidth of the EEG recording: 0.3 Hz to 150 Hz) [34].
As handling may evoke stress, which can induce changes in both the behavior of rats and, thereby, the SWD number for about 20-25 min [20,23,34], we evaluated the SWD number between 30 and 150 min of recording periods similar to previous studies [11,19]. However, it was also observed [23,35] that behavioral changes and its effect on SWD number disappeared within 25-30 min after treatments. One-hour sections of EEG recordings were evaluated separately [23]. SWDs were separated manually from the EEG based on their main features (SWDs consist a train of asymmetric spikes and slow waves; discharge frequency within SWDs: 7-11 Hz; averaged SWD duration: 1-30 s) [20].

Treatments and Animal Groups
After the recovery period, rats were assigned into six groups ( Figure 1). For the adaptation of animals to the experimental procedures, rats were handled and EEG recordings were carried out every day for 5 days. During these 5 days (adaptation period), animals were fed by powdered standard rodent chow, which was mixed with water and 1% saccharine (paste-like standard rodent chow without KEKS) for adaptation of animals not only to EEG recordings, but also to paste-like food. Then, in order to determine the averaged control SWD number, all of the rats were further fed with paste-like standard rodent chow (without KEKS) on three consecutive days. Moreover, all rats were i.p. injected by saline (first injection; 0.3 mL/100 g body weight) and it was followed by the same saline injection (30 min later; second injection; 0.3 mL/100 g body weight) on the three-day control period, and the EEG recording was carried out (three-day control period).  After three-day control periods, animals in groups 1-3 received two i.p. injections (first injections were followed by second injections 30 min later) on the fourth day of the experiments, and EEGs were recorded ( Figure 1). In relation to group 1 (n = 5), 10% DMSO solution (0.3 mL/100 g body weight; first injection) and LPS (50 µg/kg) in 0.3 mL saline/100 g body weight (second injection) were injected. A combined injection of DPCPX (first injection; i.p. 0.2 mg/kg in 0.3 mL 10% DMSO solution/100 g body weight) with LPS (second injection; 50 µg/kg in 0.3 mL saline/100 g body weight n = 5), as well as SCH58261 (first injection; i.p. 0.5 mg/kg in 0.3 mL 10% DMSO solution/100 g body weight) with LPS (second injection; 50 µg/kg in 0.3 mL saline/100 g body weight n = 5) were used in groups 2 and 3, respectively.
After control periods, rats in groups 4, 5, and 6 were fed with KEKS food for 9 consecutive days (KEKS days) ( Figure 1) and were i.p. injected by 0.3 mL saline/100 g body weight (first injection) and by same injections (second injection; 0.3 mL saline/100 g body weight, 30 min later) between the first and eighth days followed by EEG recording. It was previously demonstrated that KEKS food decreased both the number of spontaneously developed SWDs and LPS-generated increase in SWD number between 6-9 days of treatment [19]. However, to strengthen the beneficial effect of KEKS food on LPS-induced enhancement of the SWD number [19], animals in group 4 (n = 6) were injected with 0.3 mL 10% DMSO solution/100 g body weight (first injection) and, 30 min later, with LPS (second injection; 50 µg/kg in 0.3 mL saline/100 g body weight) on the ninth day of KEKS food treatment. Moreover, to investigate the putative effects of combined administration of DPCPX with LPS and SCH58261 with LPS on the SWD number in KEKS-treated animals, in a recent study, rats (group 5 and group 6) received two i.p. injections on the ninth day of KEKS food treatment (these injections were similar to injections used in group 2 and group 3 on the fourth day of the experiments). Namely, in group 5 (n = 6), first i.p. injection (0.3 mL 10% DMSO solution/100 g body weight) contained DPCPX (0.2 mg/kg), whereas the second injection contained LPS (50 µg/kg in 0.3 mL saline/100 g body weight). In group 6 (n = 6), the first i.p. injection was 0.5 mg/kg SCH58261 (in 0.3 mL 10% DMSO solution/100 g body weight), whereas the second i.p. injection contained 50 µg/kg LPS (in 0.3 mL saline/100 g body weight). EEGs were recorded every day. As it has been demonstrated previously that KEKS food alone decreased the SWD number [19], whereas i.p. 0.2 mg/kg DPCPX and i.p. 0.5 mg/kg SCH58261 alone did not change the SWD number in WAG/Rij rats [11,31], these experiments were not carried out again in this study.

Measuring the Level of Blood Glucose and R-βHB, and Body Weight
Blood was taken from the tail vein of rats. We used a glucose and ketone monitoring system (Precision Xtra™, Abbott Laboratories, Irving, TX, USA) to measure the blood glucose (mg/dL) and βHB (R-βHB; mmol/L) levels [2,11]. This equipment only detects blood levels of R-βHB. Consequently, the total blood ketone levels (R-βHB + L-βHB + AcAc + acetone) would be higher than was detected in this study. Blood glucose and R-βHB levels were measured on the third control day (control) and on the first and the ninth days of KEKS food treatment (groups 4-6). We also measured the body weight of rats before KEKS food treatment began (third control day, control) and after the last (ninth) KEKS administration day (groups 4-6).
All results were expressed as means ± SEM (standard error of the mean). The pretreatment control SWD numbers (groups 1-6) and average SWD time (groups 5 and 6) were the grand average counted from the results of control days (three-day control period). In relation to the blood R-βHB and glucose level, as well as body weight, the results were counted from the values determined on the last (third) control days. For data analysis, GraphPad PRISM 9 software was used. Significance was determined by one-or two-way analysis of variance (ANOVA), Tukey's multiple comparisons test, and Šídák's multiple comparisons test [2]. Statistical significance was considered at p < 0.05.

Treatments
After the combined i.p. administration of DPCPX (0.2 mg/kg) with LPS (50 µg/kg) and SCH58261 (0.5 mg/kg) with LPS (50 µg/kg) (groups 5 and 6, respectively) on the ninth day of KEKS food administration, the average SWD time did not change significantly (control/KEKS food + DPCPX + LPS: 7.5 ± 0.21 s/7.6 ± 0.33 s, p > 0.9999; control/KEKS food + SCH58261 + LPS: 7.6 ± 0.42 s/7.5 ± 0.29 s, p > 0.9999). As the average SWD duration did not change after these treatments and the SWD number significantly changed (increased) only after i.p. DPCPX + LPS (group 5; Figure 3B; Table 2), whereas the SWD number did not change significantly after i.p. SCH58261 + LPS (group 6; Figure 3C; Table 2) on the ninth KEKS food administration day, compared to the control, alterations in the total time of SWDs could be different in these two groups. Indeed, the total time of SWDs increased after the combined administration of DPCPX with LPS (control/KEKS food + DPCPX + LPS; 145.3 ± 14.1 s/372.1 ± 60.9 s, p < 0.0001), but the total time of SWDs did not change significantly after co-administration of SCH58261 with LPS (control/KEKS food + SCH58261 + LPS; 198.3 ± 15.9/143.8 ± 39.3 s, p = 0.83) on the ninth day of KEKS food treatment between 30 and 90 min, compared to the control.

KEKS-Generated Changes in Blood R-βHB and Glucose Levels and Body Weight
It was demonstrated that KEKS food treatment effectively increased the blood R-βHB level not only after the first KEKS food treatment, but also on the ninth day of KEKS food administration, independently of the combination of i.p. injections (LPS alone, DPCPX + LPS, SCH58261 + LPS; groups 4, 5, and 6; Figure 4A,C,E, respectively), compared to control levels. Nevertheless, the level of blood glucose was unchanged (groups 4, 5, and 6; Figure 4B,D,F, respectively) ( Table 3).

Discussion
In this study, we further validated our previous result that KEKS-supplemented food (KEKS food) is able to decrease the LPS-induced increase in SWD number [19] and provided new evidence that this alleviating effect of KEKS food may be mediated by the adenosinergic system, likely through A1Rs.

Treatments
The body weight of animals in the three KEKS treated groups (groups 4, 5, and 6) did not change significantly compared to the control (control/treated; group 4: 327.7 ± 3.63 g/323.8 ± 5.26 g, p = 0.5062; group 5: 334.7 ± 5.37 g/331.0 ± 6.68 g, p = 0.5344; group 6: 326.2 ± 6.56 g/323.5 ± 6.71 g, p = 0.8599). Similarly to our previous study [19], food intake was not investigated in the recent study. Nevertheless, the unchanged body weight of animals in groups 4, 5, and 6 suggests that KEKS food treatment did not exert its influence on the SWD number and SWD time through insufficient food intake or calorie restriction.

Discussion
In this study, we further validated our previous result that KEKS-supplemented food (KEKS food) is able to decrease the LPS-induced increase in SWD number [19] and provided new evidence that this alleviating effect of KEKS food may be mediated by the adenosinergic system, likely through A1Rs.
Ketosis may increase the concentration of adenosine [30], and thereby can modulate the activity of adenosine receptors. It was also suggested that EKSs-generated alleviating effects on absence epileptic activity may be mediated through increased βHB level-evoked increase in adenosine level and A1R activity [11,19]. Indeed, EKSs-induced increase in βHB level can increase the level of not only extracellular ATP, but also adenosine [4,43,44]. Thus, adenosine may evoke hyperpolarization of neuronal membranes and decrease neuronal activity through, for example, A1R-mediated opening of ATP-sensitive potassium channels and synaptic inhibition [4,29,44,45] resulting in both moderate hyperexcitability in the cortical focus and a decreased SWD number in WAG/Rij rats. It was also demonstrated that the activation of A2ARs increased the SWD number in WAG/Rij rats [28,46], suggesting that A2ARs are not able to modulate the alleviating influence of EKSs-generated ketosis on the number of spontaneously developed SWDs in WAG/Rij rats [11].
It has been previously demonstrated that administration of a non-selective adenosine receptor antagonist theophylline (i.p. 5 and 10 mg/kg) and i.p. 1 mg/kg SCH58261 decreased the SWD number [46,47] in WAG/Rij rats. In another rat model of human absence epilepsy GAERS (Genetic Absence Epilepsy Rats from Strasbourg), similar results were demonstrated, where another non-selective adenosine receptor antagonist caffeine (i.p. 1, 2.5, 5 and 10 mg/kg) decreased both the SWD number and total SWD time, and the A2AR antagonist DMPX (3,7-dimethyl-1-propargylxanthine; i.p. 0.15 and 0.3 mg/kg) generated a modest decrease in the total SWD time [48]. It was also demonstrated that i.p. 0.5 mg/kg DPCPX increased the SWD number in WAG/Rij rats [11], whereas a different A1R antagonist 8-CPT (8-cyclopentyl-1,3-dimethylxanthine; i.p. 0.625 and 3 mg/kg) decreased the SWD number in GAERS [48]. Based on the above results, we can conclude that A1Rs are able to generate a different modulatory influence on absence epilepsy genesis in these models. Consequently, theoretically, ketosis-generated effects on a number of spontaneously developed SWDs may be different in WAG/Rij rats and GAERS, at least through the adenosinergic system. Indeed, despite that, a ketogenic diet increased the blood βHB level in GAERS, it was not able to alter the SWD number [49]. However, new studies are needed to reveal the reason(s) of the different influences of A1R antagonism on absence epileptic activity in WAG/Rij rats and GAERS (e.g., investigation of A1R expression and distribution in brain areas implicated in absence epilepsy genesis in both rat strains).
It has been demonstrated that adenosine is a key anti-inflammatory mediator [50]. In the central nervous system, adenosine exerts its effects on inflammatory processes mainly through A1Rs and A2ARs [51]. A1Rs and A2ARs are expressed in astrocytes, oligodendrocytes, and microglia modulating inflammatory processes [50,52]. For example, both an A1R agonist (2-chloro-N 6 -cyclopentyl-2'-deoxyadenosine) and an A2AR antagonist (8-chloro-9-ethyl-2-phenethoxyadenine) alleviated the neuroinflammation (which inflammation was evoked by a pro-inflammatory cytokine cocktail, containing TNF-α, IL-1β, and interferon-gamma/IFN-γ). Moreover, this A2AR antagonist also showed antioxidant properties in mixed glial cells and after intracerebroventricular injection of 10 µg LPS [51]. These results suggest that activation of A1Rs and inhibition of A2ARs may evoke alleviating (anti-inflammatory) effects on neuroinflammatory processes. Indeed, activation of A1Rs decreased both the astrocyte proliferation and the excessive activation of microglial cells, thereby attenuating neuroinflammation, whereas an increase in microglial activity and neuroinflammation was demonstrated in A1R knockout mice [53][54][55]. Moreover, A1R activation can attenuate LPS-induced inflammation through inhibition of hypoxia-inducible factor 1 accumulation, thereby downregulation of genes involved in inflammatory processes (e.g., inducible nitric oxide synthase, iNOS) in astrocytes [56]. Activation of A1Rs may also mitigate the harmful influence of ROS (reactive oxygen species) on brain cells [26]. It has also been demonstrated that administration of SCH58261 reduced the level of IL-1β, IL-6, iNOS, and TNFα, whereas an A2AR agonist (CGS21680) increased the level of cytokines in microglial cells [57]. A2AR antagonists (e.g., SCH58261) decreased the LPS-evoked activation of microglia and secretion of IL-1β in microglial cells [58,59]. A2AR activation can increase the activation and proliferation of astrocytes [60,61]. Moreover, activation of A2ARs in microglial cells can increase not only the activity of nitric oxide synthase (NOS) and the level of COX2 expression (one of the enzymes of prostaglandin/PGE synthesis), but also the release of cytokines, PGE 2, and NO [62,63]. Consequently, activation of A1Rs may attenuate, whereas activation of A2ARs may enhance the neuroinflammatory processes and their pathological consequences [56,59,64]. All of the above results suggest that KEKS food-evoked decrease in an LPS-generated increase in SWD number are likely modulated by A1Rs, at least in WAG/Rij rats ( Figure 3A,B). However, it was also suggested that A2ARs can also evoke peripheral anti-inflammatory effects [50,65]; adenosine may decrease LPS-induced cytokine production through A2ARs [27]; activation of A2ARs may inhibit the iNOS expression and NO production [66]; and inhibition of A2ARs did not abolish, but somewhat decreased the KEKS food-evoked alleviating effect on LPS-generated increase in SWD number between 90 and 150 min ( Figure 3C). Thus, the beneficial effect of KEKS food-generated A2AR activation on LPS-induced neuroinflammation thereby on increased SWD number cannot be excluded entirely.

Conclusions
Our results strengthened the potential of ketogenic supplements, such as KEKSsupplemented food, for the treatment of epilepsy through the inhibition of inflammatory pathways. In relation to the mechanism of action, it is likely that KEKS-induced ketosis modulated A1Rs to alleviate the neuroinflammation-induced increase in SWD number. Thus, theoretically, co-administration of EKSs and different modulators of adenosinergic systems (e.g. adenosine receptors) may allow us to develop promising therapeutic tools in the treatment of not only epilepsy, but also inflammation-evoked neurodegenerative diseases. However, further studies are needed to reveal molecular signaling between EKSsevoked alleviating effects on the SWD number, neuroinflammation, and the adenosinergic system in different cells and brain areas implicated in absence epilepsy genesis.