Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 4

0006-8993/$ - see frohttp://dx.doi.org/10

nCorresponding autE-mail address: t

Research Report

Fructose-1,6-diphosphate protects againstepileptogenesis by modifying cation-chlorideco-transporters in a model of amygdaloid-kindlingtemporal epilepticus

Yao Ding, Shan Wang, Yan Jiang, Yi Yang, Manman Zhang, Yi Guo,Shuang Wang, Mei-ping Dingn

Department of Neurology, Second Affiliated Hospital, School of Medicine, Zhejiang University, No. 88, Jiefang Road,Hangzhou 310009, China

a r t i c l e i n f o

Article history:

Accepted 25 September 2013

Fructose-1,6-diphosphate (FDP) shifts the metabolism of glucose from glycolysis to the

pentose phosphate pathway and has anticonvulsant activity in several acute seizure

Available online 1 October 2013

Keywords:

Fructose-1,6-diphosphate

Antiepileptogenesis

Amygdaloid kindling

Cation-chloride cotransporters

γ-Aminobutyric acid

nt matter & 2013 Elsevie.1016/j.brainres.2013.09.04

hor. Fax: þ86 571 8778 47inady2002@hotmail.com

a b s t r a c t

animal models. In the present study, we investigated the anti-epileptogenic effects of FDP

in an amygdaloid-kindling seizure model, which is an animal model of the most common

form of human temporal lobe epilepsy. We found that 1.0 g/kg FDP slowed seizure

progression and shortened the corresponding after-discharge duration (ADD). FDP

increased the number of stimulations needed to reach seizure stages 2–5 and prolonged

the cumulative ADD prior to reaching stages 3–5. It also shortened staying days and

cumulative ADD in stages 4–5. However, it demonstrated no significant protective effect

when administered after the animals were fully kindled. In hippocampal neurons, cation-

chloride co-transporters (CCCs) are suggested to play interesting roles in epilepsy by

modulating γ-aminobutyric acid (GABA)ergic activity through controlling GABAA receptor-

mediated reversal potential. We examined the potential link between FDP and the hippo-

campal expression of two main members of the CCCs: the neuron-specific Kþ–Cl�

co-transporter 2 (KCC2) and Naþ–Kþ–Cl�co-transporter 1 (NKCC1). FDP inhibited the

kindling-induced downregulation of KCC2 expression and decreased NKCC1 expression

during the kindling session. Taken together, our data reveal that FDP may have protective

activity against epileptogenesis, from partial to generalized tonic–clonic seizures. Further-

more, our findings suggest that the FDP-induced imbalance between KCC2 and NKCC1

expression may be involved in the neuroprotective effect.

& 2013 Elsevier B.V. All rights reserved.

r B.V. All rights reserved.2

50.(M.-p. Ding).

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 488

1. Introduction

The high-fat and low-carbohydrate ketogenic diet (KD) hasgained acceptance as a treatment for refractory epilepsy(Freeman et al., 2007). Our previous study demonstrated thatKD significantly delayed seizure progression and has aneuroprotective effect against kindling (Jiang et al., 2012).However, even though it has been indicated that KD is welltolerated and does not produce any significant health con-cerns in KD-fed animals, it is still regarded as a difficultregime to follow (Auvin, 2012). This promoted us to investi-gate other potential agents that can mimic the protectiveeffect of KD but with less adverse outcomes. Fructose-1,6-diphosphate (FDP) inhibits glycolysis and diverts glucoseinto the pentose phosphate pathway, and has been shown tobe neuroprotective in several acute chemical seizure models(Lian et al., 2007), and therefore can be considered an antic-onvulsant. However, no investigations of the effects of FDP inseizure progression have been performed. The idea that“seizures beget seizures” was proposed some 130 years agoby Gowers (1881), and it is now widely accepted by contem-porary epileptologists and neuroscientists. The animal modelof amygdaloid kindling recapitulates many of the features ofepilepsy progression (Garriga-Canut et al., 2006; Sharma et al.,2007) and has been used extensively to investigate temporallobe epilepsy, the most common form of epilepsy. Thepresent study was designed to explore the effects of FDP onepileptic progression in a chronic amygdaloid kindling model.The results shed light on the use of FDP as a potentialalternative for KD.

Going one step further, recent evidence has revealed theimportance of γ-aminobutyric acid (GABA)-mediated excita-tion in epileptogenesis, during which inhibitory GABAergicfunction can be dynamically changed to excitation (Koyamaet al., 2012). Changes in plasmalemmal chloride ion-transportmechanisms, which are instrumental for the chloride ionchannel function, and the regulation of chloride concentra-tion have been suggested to play pivotal roles in GABA-mediated excitation. Among many KD-induced physiologicor biochemical changes in brain cell metabolism, an in vitrostudy showed that a KD may modify cation-chloride co-trans-porters (CCCs) and regulate chloride homeostasis (Rheims et al.,2009). Two major members of CCCs, Kþ–Cl�co-transporter 2(KCC2) and Naþ–Kþ–Cl� co-transporter 1 (NKCC1), control theelectrochemical gradient of chloride across the neuronalplasma membrane, and thy are secondary active transportersthat drive net chloride extrusion or uptake by using the Kþ andNaþ gradients (Loscher et al., 2013). To understand the mechan-ism underlying FDP's action and its possible connection withKD through shared mechanisms, we investigated the effect ofFDP on the regulation of KCC2 and NKCC1 mRNA expressionlevels.

2. Results

The electrodes were found in the right basolateral amygdale in131 rats, and these animals were included in our study; 52 wereinvestigated for seizure semiology and electroencephalogram

(EEG), while 79 rats were killed by decapitation to collect samplesfor molecular analysis. The animals displayed no signs ofabdominal cramps or other abnormal behavior after intraperito-neal (i.p.) injections of FDP.

2.1. Effect of FDP on kindling acquisition

This experiment included three groups of rats pretreatedwith 0.5 g/kg FDP (n¼8), 1.0 g/kg FDP (n¼10), or saline (n¼9).FDP at the dose of 1.0 g/kg slowed seizure stage progression(Po0.05; Fig. 1A) and shortened the corresponding after-discharge duration (ADD; Po0.05, Fig. 1B). However, 0.5 g/kgFDP had no effect on either the progression of seizure stage orthe ADD (Fig. 1A and B).

To further analyze the stepwise progression of kindling,we calculated the numbers of stimulations and cumulativeADD needed to reach each seizure stage or remain at general-ized seizures (GS, stages 4–5). FDP at the dose of 1.0 g/kgincreased the number of stimulations to reach stages 2–5(Po0.05 to reach stages 2 and 4, Po0.01 to reach stages 3 and 5;Fig. 2A) and prolonged the cumulative ADD to reach stages 3–5(Po0.05 to reach stages 4 and 5, Po0.01 to reach stage 3;Fig. 2B). FDP at 1.0 g/kg significantly reduced the number ofdays for which animals stayed in stages 4–5 (Po0.01, Fig. 2C)and shortened the corresponding cumulative ADD (Po0.01,Fig. 2D). However, FDP at the dose of 0.5 g/kg did not differfrom the controls (Fig. 2A–D). In addition, we found that FDP atthe dose of 1.0 g/kg prevented an after-discharge threshold(ADT) decrease at stage 5 compared to controls (Po0.05;Fig. 2E) but showed no such effect on stages 1–4.

2.2. Effect of FDP on fully kindled animals

We next examined the potential anticonvulsant effect of FDPwhen it was administered after the animals were fullykindled by an ADT current. FDP did not decrease the meanseizure stage or the incidence of GS, and it did not shortenthe cumulative ADD or cumulative generalized seizure dura-tion (GSD). There were no differences among the 0.5 g/kg FDP(n¼9), 1.0 g/kg FDP (n¼8), and control groups (n¼8) (Fig. 3).

2.3. Hippocampal KCC2 and NKCC1 mRNA expressionlevels

Relative KCC2 mRNA expression in the hippocampus wasincreased after 4, 8, and 16 days of 1.0 g/kg FDP injection(Po0.01, Fig. 4A), while NKCC1 mRNA expression decreasedafter the 1st day (Po0.05, Fig. 4B) and also after 4, 8, and 16days (Po0.01, Fig. 4B) of daily 1.0 g/kg FDP injections.

In the control rats, KCC2 expression decreased after 4, 8,and 16 days of stimulations (Po0.05 after 4 days and Po0.01after 8 and 16 days, Fig. 4C). Compared with controls, ratspretreated with 1.0 g/kg FDP showed increased KCC2 mRNAexpression after 4 (Po0.05), 8, and 16 (Po0.01) days ofstimulations (Fig. 4C).

The expression levels of NKCC1 mRNA did not changesignificantly (Fig. 4D) during kindling in the control group.However, 1.0 g/kg FDP treatment induced a more significant

Fig. 1 – Effects of FDP on amygdaloid-kindling acquisition (n¼9, control group; n¼8, 0.5 g/kg group; n¼10, 1.0 g/kg group).(A) Behavioral stage. (B) After-discharge duration (ADD). Values are expressed as mean7S.E.M. *Po0.05 compared with control.

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 4 89

reduction in NKCC1 mRNA expression than control after the1st day of stimulation, as well as after 4, 8, and 16 days ofstimulations (Po0.05 after 1 day stimulation, Po0.01 after 4,8, and 16 days of stimulation, Fig. 4D).

3. Discussion

A great deal of research has been done on KD, and some studieshave identified molecular players in the anti-epileptic effects ofthe KD (Masino and Rho, 2012; Lutas and Yellen, 2013). Nosingle mechanism has been identified to explain the seizureprotection conferred by the KD (Politi et al., 2011), but itsreplacement for glucose as an energy source for brain hasfascinated researchers. In addition to classic KD, different dietswith carbohydrate levels below normal are also effective in themanagement of intractable epilepsy (Chang et al., 2013), whichinspired studies to determine if approaches that specificallydecrease glucose utilization through the glycolytic pathway canalso protect against seizures (Lian et al., 2007). For instance,administration of the glycolytic-inhibitor 2-deoxy-D-glucose(2-DG) has shown anticonvulsant effects with increased ADT

and delayed progression of epileptogenesis (Garriga-Canutet al., 2006). FDP, an intracellular metabolite of glucose, exertspotent feedback inhibition on the activity of phosphofructoki-nase-1, a rate-limiting enzyme in glycolysis, and is also thoughtto mimic the low-glucose availability associated with reducedcarbohydrate intake in the KD (Stringer and Xu, 2008).

In the present study, we demonstrated that 1.0 g/kg FDPsignificantly delayed seizure progression induced by amyg-daloid kindling, which shortened the corresponding ADD. Incontrast, the effect of FDP at a relatively lower dose (0.5 g/kg) was much weaker. These data confirm that the anti-epileptogenesis function of FDP is dose related, which issimilar to the anticonvulsant activity in rat models of acutegeneralized motor seizures induced by chemical convul-sants, such as pilocarpine, kainic acid, or pentylenetetrazole(Lian et al., 2007). A previous study suggested that oraladministration FDP may protect against generalized tonic–clonic seizures but not absence seizures (Lian et al., 2008).The amygdaloid-kindling seizure model enabled us to ana-lyze epileptogenesis in graded stages. FDP at a dose of1.0 g/kg exerts anti-epileptic action not only in partialseizure stages (stages 2–3) but also delayed epileptogenesisfrom partial to generalized seizures (stages 4–5). At the

Fig. 2 – Effects of FDP on amygdaloid-kindling acquisition (n¼9, control group; n¼8, 0.5 g/kg FDP group; n¼10, 1.0 g/kg FDPgroup). (A) Days of stimulations. (B) Cumulative ADD required reaching each stage (first appearance of each stage). (C) Days ofstimulations. (D) Cumulative ADD in stages 4–5. (E) After-discharge threshold (ADT) in each stage. Values are expressed asmean7S.E.M. *Po0.05 and **Po0.01 compared with control.

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 490

same time, it attenuated the ADT, preventing rats fromreaching stage 5.

Our study on how FDP affects the balance between KCC2and NKCC1 expression is driven by the evolving role ofGABAergic function in epileptogenesis and the recognizedcomplexes of ion-channels/co-transporters involved in chlor-ide transport. GABA is the principal inhibitory neurotransmit-ter in the brain, producing inhibitory postsynaptic potentialsin both feedforward and feedback circuits. A decrease in GABArelease could account for the neuronal hyperexcitabilityobserved in epilepsy (Melo et al., 2010). According to a tradi-tional amino acid hypothesis (Dahlin et al., 2005), KD sup-presses seizures by increasing the level of GABA in the brain.However, this may not be the case because Hartman et al.(2007) reported that GABA levels showed no significant eleva-tion in KD-fed animals. Remarkably, recent studies haveshown the importance of GABA-mediated excitation in epi-leptogenesis. Intracellular chloride concentration, Cli, deter-mines the polarity of GABAA-induced neuronal Cl currents. Inneurons, Cli is relies on a balance between NKCCs, such asNKCC1, which physiologically accumulate Cl in the cell, andKCC2, which extrudes Cl� from the neuronal soma. Altera-tions in the balance between NKCC1 and KCC2 activities may

determine whether the effects of GABA are hyperpolarizing ordepolarizing (Munoz et al., 2007). Accumulating evidence hascorrelated epileptogenesis with altered functional expressionof NKCC1 and KCC2 co-transporters. A study on humanepileptic tissues also revealed that NKCC1 is up-regulated,while KCC2 is down-regulated in subicular cells adjacent to thehippocampus in patients with epilepsy (Huberfeld et al., 2007).FDP has been shown to shift the metabolism of glucose fromthe glycolytic pathway to the pentose phosphate pathway andmodify different signal pathways (e.g., BDNF/TrkB in ourprevious study) (Wheeler and Chien, 2012), and it ultimatelyaffects the GABA neurotransmission system (Lund et al., 2008). Inthe present study, we found that FDP can prevent kindling-induced KCC2 downregulation and decreased NKCC1 expression;this finding suggests that FDP may decrease the intracellularCl� level by modulating the balance between KCC2 and NKCC1expressions, and thus contributes to a reduction of GABA-mediated neuronal excitability.

Conversely, oxidative stress is involved in the pathogen-esis of kindling seizures (Aguiar et al., 2012), and it can resultin a rapid decrease in KCC2 expression in vitro (Wakeet al., 2007). FDP diverts glucose into the pentose phosphatepathway and increases intracellular NADPH production,

Fig. 3 – Effects of FDP on (A) mean seizure stage, (B) cumulative ADD, (C) incidence of generalized seizure (GS), and (D)cumulative generalized seizures duration (GSD) in fully kindled rats (n¼8, control group; n¼9, 0.5 g/kg FDP group; n¼8, 1.0 g/kgFDP group). Values are expressed as mean7S.E.M.

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 4 91

which preserves cellular glutathione levels and decreasesROS production (Park et al., 2004). Therefore, FDP may alsoregulate the change in KCC2 levels through ROS. Thus, theFDP-induced increase in KCC2 expression following kindlingseizures may be also related to the antioxidant propertyof FDP.

In conclusion, we demonstrated the dose-dependent anti-epileptogenic effects of FDP in amygdaloid-kindling acquisition;however FDP had no effect on fully kindled seizures. Addition-ally, FDP inhibited the kindling-induced downregulation of KCC2and decreased hippocampal expression of NKCC1, suggestingthat reduced GABA-mediated excitation induced by alteration ofintracellular chloride level may also play a role in the anti-epileptogenic function of FDP. Because the mechanisms ofcurrent clinical anti-epileptic drugs are unrelated with glucosemetabolism, these findings are of great importance because theyilluminate novel approaches for controlling seizures and maylead to the development of a small molecule replacement for thecomplex KD treatment.

4. Experimental procedures

4.1. Animals and surgery

Male Sprague Dawley rats (260–300 g, Grade II, Certificate No.SCXK2003-0001; Experimental Animal Center, Zhejiang Acad-emy of Medical Science, Hangzhou, China) were housed inindividual cages with a 12-h light–dark cycle (lights on from0800 to 2000). All experiments were carried out in accordancewith the ethical guidelines of the Zhejiang University AnimalExperimentation Committee and were in complete compli-ance with the National Institutes of Health Guide for the Careand Use of Laboratory Animals. Water and food were pro-vided ad libitum. Experiments were carried out between 1000and 1700.

After deep anesthesia with chloral hydrate (400mg/kg, i.p.),rats were mounted on a stereotaxic apparatus (512600; Stoelting,IL, USA). Electrodes were implanted into the right basolateral

Fig. 4 – With daily administration of FDP (1.0 g/kg) without stimulation, the relative levels of KCC2 (A) and NKCC1 (B) mRNAexpression in the right hippocampus were detected on days 1, 4, 8, and 16 (n¼6). Rats daily pretreated with FDP (1.0 g/kg) orsaline were sacrificed after 1, 4, 8, and 16 days of stimulations, and relative levels of KCC2 (C) and NKCC1 (D) mRNA expressionin the right hippocampus are shown in each group (n¼6). *Po0.05 and **Po0.01 versus control. # Po0.05 and ## Po0.01indicated untreated control groups that exhibited statistically significant differences during kindling.

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 492

amygdala (coordinates from the bregma: AP¼�2.4 mm, L¼�4.8mm, and V¼�8.8mm). The electrodes were made of twistedstainless steel wires with a diameter of 0.2mm (A.M. Systems,USA) and Teflon coated except for 0.5mm at the tip. Thedistance between tips was 0.7–0.8mm. The electrodes wereconnected to a miniature receptacle, which was embedded inthe skull with dental cement. Animals were allowed to recoverfrom surgery over a 10-day period before testing.

4.2. Procedure for kindling and threshold measurement

EEGs of the right amygdala were recorded with a digitalamplifier (RM-6240, Chengyi, China). The ADT of each subjectwas determined with a constant current stimulator (YC-2;Chengyi, China). An initial current of 50 μA was used andsubsequently increased stepwise was 20-μA increments. Con-secutive trials were separated by at least 30 min. The ADTwas defined as the lowest current required elicited an after-discharge lasting for at least 5 s on EEG. All animals weresubjected to once-a-day kindling stimulation with the same

current intensity as their own ADT. The daily kindlingstimulus consisted of a 1-s train of monophasic, 1-mssquare-wave pulses at 60 Hz. Animals that exhibited threeconsecutive stage 5 seizures were considered to be fullykindled. Seizure severity was classified according to a mod-ification Racine classification and scored as follows (Racineet al., 1972): (1) facial movement; (2) head nodding; (3)unilateral forelimb clonus; (4) bilateral forelimb clonus andrearing; and (5) bilateral forelimb clonus, rearing, and falling.Stages 1–3 were considered to be focal seizures, while stages 4and 5 were considered GS. In addition to seizure stage, theADD and GSD were recorded.

4.3. Effect of FDP on kindling acquisition and fully kindledanimals

In experiment 1, rats were divided into three groups matchedfor ADTs. Based on pharmacokinetics in a previous study (Xuand Stringer, 2008), FDP (47810; Sigma, USA) was intraperito-neally administered 1 h before kindling stimulation. FDP

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 4 93

was administered once daily at 0.5 and 1.0 g/kg doses; anequivalent volume of saline (vehicle) was administered in thecontrol group. Animals were stimulated daily at ADT inten-sity until the control group animals were fully kindled.

In experiment 2, rats were stimulated until they were fullykindled (without any FDP treatment during kindling) andwere then divided into three groups in the same fashion asexperiment 1. On the next day, three groups received stimuliwith pre-kindling ADT intensity after pretreatment withsaline or 0.5 or 1.0 g/kg FDP. All rats were pretreated andstimulated daily for 5 consecutive days, and the seizure stage,ADD, and GSD were measured by the same procedures usedfor kindling seizures.

4.4. Gene expression study

The rats were killed by decapitation, and the brains wereremoved within 1 h of the last stimulation. The ipsilateralhippocampus was dissected and flash frozen in liquidnitrogen. Total RNA was extracted from the hippocampaltissues with TRIzol reagent (Invitrogen, Carlsbad, CA, USA).Reverse-transcriptase polymerase chain reaction (RT-PCR)was carried out using the SYBRs Green Realtime PCRMaster Mix (QPK-201; Toyobo, Japan) on a LightCyclers

2.0 Instrument (Roche Diagnostics Corporation, Germany)according to manufacturer's instructions. Primers weredesigned and synthesized by Invitrogen BiotechnologyCo. Ltd. (Shanghai, China). The primers sequences wereas follows: GAPDH forward: 5′-GGTGGACCTCATGGCCTA-CAT-3′ and reverse: 5′-GCCTCTCTCTTGCTCTCAGTATCCT-3′; KCC2 forward: 5′-ATCGAGATCCTGCTGGCTTA-3′ andreverse: 5′-ACTTGACGCCCACAAAGACT-3′; NKCC1 forward:5′-GAAGCAAAGGCTCAGATCGT-3′ and reverse: 5′-ATCCGGTC-CAAAGTTCTCAC-3′.

4.5. Histology

Upon completion of the behavioral experiments, electrodeplacements were histologically verified by staining the brainsections with toluidine blue O. Data from animals in whichthe electrodes were placed within the right basolateralamygdala according to the atlas of Paxinos and Watson(2006) were included in the statistical analysis. In the caseof animals used in biochemical experiments, basolateralamygdala electrode placement was visually examined andconfirmed in the process of tissue separation.

4.6. Statistical analysis

Data are presented as mean7standard error of the mean(S.E.M.). Statistical evaluations of group differences in kind-ling acquisition were performed with two-way analysis ofvariance (ANOVA). Other tests were performed with one-wayANOVA when the data were normally distributed and thevariances were homogeneous; otherwise, nonparametricMann–Whitney U tests were used. All analyses were two-sided, and Po0.05 was considered statistically significant.

Acknowledgments

This project was supported by grants from the NationalNatural Science Foundation of China (81100968, 81171227,and 81271435).

r e f e r e n c e s

Aguiar, C.C., Almeida, A.B., Araujo, P.V., de Abreu, R.N., Chaves, E.M., do Vale, O.C., Macedo, D.S., Woods, D.J., Fonteles, M.M.,Vasconcelos, S.M., 2012. Oxidative stress and epilepsy:literature review. Oxidative Medicine and Cellular Longevity2012, 795259.

Auvin, S., 2012. Should we routinely use modified Atkins dietinstead of regular ketogenic diet to treat children withepilepsy?. Seizure 21, 237–240.

Chang, P., Terbach, N., Plant, N., Chen, P.E., Walker, M.C.,Williams, R.S., 2013. Seizure control by ketogenic diet-associated medium chain fatty acids. Neuropharmacology 69,105–114.

Dahlin, M., Elfving, A., Ungerstedt, U., Amark, P., 2005. Theketogenic diet influences the levels of excitatory andinhibitory amino acids in the CSF in children with refractoryepilepsy. Epilepsy Research 64, 115–125.

Freeman, J.M., Kossoff, E.H., Hartman, A.L., 2007. The ketogenicdiet: one decade later. Pediatrics 119, 535–543.

Garriga-Canut, M., Schoenike, B., Qazi, R., Bergendahl, K., Daley,T.J., Pfender, R.M., Morrison, J.F., Ockuly, J., Stafstrom, C.,Sutula, T., Roopra, A., 2006. 2-Deoxy-D-glucose reducesepilepsy progression by NRSF-CtBP-dependent metabolicregulation of chromatin structure. Nature Neuroscience 9,1382–1387.

Gowers, W.R., 1881. Epilepsy and Other Chronic ConvulsiveDiseases: Their Causes, Symptoms and Treatment, vol. 2nd.Dover Publications, London.

Hartman, A.L., Gasior, M., Vining, E.P., Rogawski, M.A., 2007. Theneuropharmacology of the ketogenic diet. Pediatric Neurology36, 281–292.

Huberfeld, G., Wittner, L., Clemenceau, S., Baulac, M., Kaila, K.,Miles, R., Rivera, C., 2007. Perturbed chloride homeostasis andGABAergic signaling in human temporal lobe epilepsy. Journalof Neuroscience 27, 9866–9873.

Jiang, Y., Yang, Y., Wang, S., Ding, Y., Guo, Y., Zhang, M.M.,Wen, S.Q., Ding, M.P., 2012. Ketogenic diet protects againstepileptogenesis as well as neuronal loss in amygdaloid-kindling seizures. Neuroscience Letters 508, 22–26.

Koyama, R., Tao, K., Sasaki, T., Ichikawa, J., Miyamoto, D.,Muramatsu, R., Matsuki, N., Ikegaya, Y., 2012. GABAergicexcitation after febrile seizures induces ectopic granule cellsand adult epilepsy. Nature Medicine.

Lian, X.Y., Khan, F.A., Stringer, J.L., 2007. Fructose-1,6-bisphosphate has anticonvulsant activity in models of acuteseizures in adult rats. Journal of Neuroscience 27,12007–12011.

Lian, X.Y., Xu, K., Stringer, J.L., 2008. Oral administration offructose-1,6-diphosphate has anticonvulsant activity.Neuroscience Letters 446, 75–77.

Loscher, W., Puskarjov, M., Kaila, K., 2013. Cation-chloridecotransporters NKCC1 and KCC2 as potential targets for novelantiepileptic and antiepileptogenic treatments.Neuropharmacology 69, 62–74.

Lund, I.V., Hu, Y., Raol, Y.H., Benham, R.S., Faris, R., Russek, S.J.,Brooks-Kayal, A.R., 2008. BDNF selectively regulates GABAAreceptor transcription by activation of the JAK/STAT pathway.Science Signaling 1, ra9.

b r a i n r e s e a r c h 1 5 3 9 ( 2 0 1 3 ) 8 7 – 9 494

Lutas, A., Yellen, G., 2013. The ketogenic diet: metabolicinfluences on brain excitability and epilepsy. Trends inNeurosciences 36, 32–40.

Masino, S.A., Rho, J.M., 2012. Mechanisms of ketogenic diet action.In: Jasper's Basic Mechanisms of the Epilepsies, vol., J.L.Noebels, M. Avoli, M.A. Rogawski, R.W. Olsen, A.V. Delgado-Escueta, (Eds.), Bethesda (MD).

Melo, T., Bigini, P., Sonnewald, U., Balosso, S., Cagnotto, A.,Barbera, S., Uboldi, S., Vezzani, A., Mennini, T., 2010. Neuronalhyperexcitability and seizures are associated with changes inglial-neuronal interactions in the hippocampus of a mousemodel of epilepsy with mental retardation. Journal ofNeurochemistry 115, 1445–1454.

Munoz, A., Mendez, P., DeFelipe, J., Alvarez-Leefmans, F.J., 2007.Cation-chloride cotransporters and GABA-ergic innervation inthe human epileptic hippocampus. Epilepsia 48, 663–673.

Park, J.Y., Kim, E.J., Kwon, K.J., Jung, Y.S., Moon, C.H., Lee, S.H.,Baik, E.J., 2004. Neuroprotection by fructose-1,6-bisphosphateinvolves ROS alterations via p38 MAPK/ERK. Brain Research1026, 295–301.

Paxinos, G., Watson, C., 2006. The Rat Brain in StereotaxicCoordinates, 6th. Academic Press, New York.

Politi, K., Shemer-Meiri, L., Shuper, A., Aharoni, S., 2011. Theketogenic diet 2011: how it works. Epilepsy Research andTreatment 2011, 963637.

Racine, R.J., Gartner, J.G., Burnham, W.M., 1972. Epileptiformactivity and neural plasticity in limbic structures. BrainResearch 47, 262–268.

Rheims, S., Holmgren, C.D., Chazal, G., Mulder, J., Harkany, T.,Zilberter, T., Zilberter, Y., 2009. GABA action in immatureneocortical neurons directly depends on the availability ofketone bodies. Journal of Neurochemistry 110, 1330–1338.

Sharma, A.K., Reams, R.Y., Jordan, W.H., Miller, M.A., Thacker,H.L., Snyder, P.W., 2007. Mesial temporal lobe epilepsy:pathogenesis, induced rodent models and lesions. ToxicologicPathology 35, 984–999.

Stringer, J.L., Xu, K., 2008. Possible mechanisms for theanticonvulsant activity of fructose-1,6-diphosphate. Epilepsia49 (Suppl. 8), 101–103.

Wake, H., Watanabe, M., Moorhouse, A.J., Kanematsu, T., Horibe, S.,Matsukawa, N., Asai, K., Ojika, K., Hirata, M., Nabekura, J., 2007.Early changes in KCC2 phosphorylation in response to neuronalstress result in functional downregulation. Journal ofNeuroscience 27, 1642–1650.

Wheeler, T.J., Chien, S., 2012. Protection of rat cardiac myocytesby fructose-1,6-bisphosphate and 2,3-butanedione. PLoS One7, e35023.

Xu, K., Stringer, J.L., 2008. Pharmacokinetics of fructose-1,6-diphosphate after intraperitoneal and oral administration toadult rats. Pharmacological Research 57, 234–238.