B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Research Report

Ketogenic diet protects dopaminergic neurons against 6-OHDAneurotoxicity via up-regulating glutathione in a rat model ofParkinson's disease

Baohua Chenga,c,⁎,1, Xinxin Yanga,b,1, Liangxiang Anb, Bo Gaoa, Xia Liua, Shuwei Liuc

aJining Medical College, Jining, ChinabRizhao People's Hospital, Rizhao, ChinacMedical College of Shandong University, Jinan, China

A R T I C L E I N F O

⁎ Corresponding author. Fax: +86 537 2217079E-mail address: apple9red@yahoo.com.cn

1 These authors contributed equally to the

0006-8993/$ – see front matter © 2009 Elsevidoi:10.1016/j.brainres.2009.06.060

A B S T R A C T

Article history:Accepted 17 June 2009Available online 25 June 2009

The high-fat ketogenic diet (KD) leads to an increase of blood ketone bodies (KB) level andhas been used to treat refractory childhood seizures for over 80 years. Recent reports showthat KD, KB and their components (D-beta-hydroxybutyrate, acetoacetate and acetone)have neuroprotective for acute and chronic neurological disorders. In our present work, weexamined whether KD protected dopaminergic neurons of substantia nigra (SN) against 6-hydroxydopamine (6-OHDA) neurotoxicity in a rat model of Parkinson's disease (PD) usingNissl staining and tyrosine hydroxylase (TH) immunohistochemistry. At the same time wemeasured dopamine (DA) and its metabolites dihydroxyphenylacetic acid (DOPAC) andhomovanillic acid (HVA) in the striatum. To elucidate the mechanism, we also measuredthe level of glutathione (GSH) of striatum. Our data showed that Nissl and TH-positiveneurons increased in rats fed with KD compared to rats with normal diet (ND) afterintrastriatal 6-OHDA injection, so did DA and its metabolite DOPAC. While HVA had notchanged significantly. The change of GSH was significantly similar to DA. We concludedthat KD had neuroprotective against 6-OHDA neurotoxicity and in this period GSH playedan important role.

© 2009 Elsevier B.V. All rights reserved.

Keywords:Parkinson's diseaseKetogenic dietKetone bodyGlutathione6-hydroxydopamine

1. Introduction

Parkinson's disease (PD) is a neurodegenerative disordercharacterized by motor symptoms including tremor, musclerigidity, paucity of voluntary movements, and posturalinstability (Qu et al., 2007). It is characterized by a progressiveand selective degeneration of dopaminergic neurons of thesubstantia nigra (SN) and the exact cause and underlyingmechanism responsible for the progressive neurodegenera-

.(B. Cheng).

se works.

er B.V. All rights reserved

tion of sporadic PD remains unknown (Chen et al., 2007).Reduced glutathione (GSH) is a cellular reductant, whichprotects against oxidative stress (Leret et al., 2002). Post-mortem research shows 40% decrease of GSH in SN of PDpatients (Sian et al., 1994). Thus, it has been suggested thatlow levels of nigrostriatal GSH contents and consequentoxidative stress might contribute to the degeneration ofdopaminergic neurons in idiopathic PD (Pinnen et al., 2007).Although GSH is not the only antioxidant molecule reported

.

26 B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

to be altered in PD, the magnitude of GSH depletionappears to parallel the severity of the disease and is theearliest known indicator of nigral degeneration (Chinta andAndersen, 2006).

In a clinical trial, patients with PD remained on a ketogenicdiet (KD) for 28 days in an open trial conducted at theMovement Disorders Clinic of the Beth Israel Medical Centerand Unified Parkinson's Disease Rating Scale (UPDRS) scoreswere determined. The results showed that the changes in theUPDRS scores that occurred in PD patients during the KD. Themean total decrease in UPDRS scores was 43.4%. Amongsymptoms that improved were resting tremor, freezing,balance, gait, mood, and energy level (Vanitallie et al., 2005).D-beta-hydroxybutyrate, the reduced form of the ketones,confers protection against the structural and functionaldeleterious effects of the parkinsonian toxin MPTP; theseinclude degeneration of SN dopaminergic neurons and striataldopaminergic fibers, loss of striatal dopamine, and PD-like

Fig. 1 –Nissl staining of SN (A) ND; (B) KD; (C) ND + 6-OHDA; (D) KDas mean±SEM. (n=5) #p<0.01 compared to control rats (ND withoinjection; +p>>0.05 compared to control rats (ND without 6-OHD

motor deficit (Tieu et al., 2003). In mesencephalic neuronalculture, addition of 4 mM of sodium D-beta-hydroxybutyratesignificantly protected mesencephalic neurons from MPP+toxicity and increased the rate of survival (Koustova et al.,2003). Pretreatment of cells with 8mMD-beta-hydroxybutyrateprovided significant protection to SH-SY5Y cells againsttoxicity induced by rotenone, a PD model in vitro (Imamuraet al., 2006).

In this study, we investigated the effects of KD on the ratmodel of PD induced by 6-hydroxydopamine (6-OHDA).Dopaminergic neurons in the SN and density of fibers in thestriatum were observed using Nissl and tyrosine hydroxylase(TH) immunohistochemistry. HPLC was used for measuringdopamine (DA) and its metabolites dihydroxyphenylaceticacid (DOPAC) and homovanillic acid (HVA). We also detectedthe level of GSH. Our data showed that KD protecteddopaminergic neurons of SN against 6-OHDA neurotoxicityvia up-regulating GSH in a rat model of PD.

+ 6-OHDA; (E) Data of counting cells. Data were representativeut 6-OHDA injection); *p<0.05 compared to ND with 6-OHDAA injection).

27B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

2. Results

2.1. KD protected dopaminergic neurons in the SN andtheir neurites in the striatum against 6-OHDA neurotoxicity

Weexaminedtheeffectsof intrastriatal 6-OHDAinjectiononSNdopaminergic neurons in ND or KD rats for 2 weeks using Nisslstaining and TH immunohistochemistry (Figs. 1 and 2). After2 weeks of 20 μg intrastriatal 6-OHDA injection, Nissl-positiveand TH-positive neurons of SN were significantly decreased inrats fedwithND compared to theNDwithout 6-OHDA injection(control) (p<0.01).WhileNissl-positiveandTH-positiveneuronsof SN in rats fed with KD increased more significantly than inrats fed with ND after 2 weeks of 6-OHDA injection (p<0.05).There were no difference between ND and KD rats withoutintrastriatal 6-OHDA injection (p>0.05). The density of TH-positive fibers had the similar changes in the striata (Fig. 3).

Fig. 2 – TH immunohistochemistry of SN (A) ND; (B) KD; (C) ND + 6representative as mean±SEM. (n=5) #p<0.01 compared to controlwith 6-OHDA injection; +p>0.05 compared to control rats (ND wit

2.2. KD inhibited the decrease of striatal DA and itsmetabolite DOPAC but not HVA induced by 6-OHDA

We used HPLC analysis to measure the levels of striataldopamine and its metabolites DOPAC and HVA. After 2 weeksof 20 μg 6-OHDA intrastrial injection, the levels of striatal DAand its metabolites DOPAC and HVA were significantlydecreased in rats fed with ND compared to the ND ratswithout 6-OHDA injection (control) (p<0.01). While the levelsof striatal DA and its metabolite DOPAC in rats fed with KDincreased more significantly than in rats fed with ND after2 weeks of 6-OHDA injection (p<0.05). The metabolite HVA ofDA in rats fed with KD did not increased more significantlythan in rats fed with ND after 2 weeks of 6-OHDA injection(p>0.05). There were no difference of DA and its metabolitesDOPAC and HVA between ND and KD rats without 6-OHDAinjection (p>0.05) (Fig. 4).

-OHDA; (D) KD + 6-OHDA; (E) Data of counting cells. Data wererats (ND without 6-OHDA injection); *p<0.05 compared to NDhout 6-OHDA injection).

Fig. 3 – TH immunohistochemistry of striatum (A) ND; (B) KD; (C) ND + 6-OHDA; (D) KD + 6-OHDA; (E) Data of stiatal TH+ fibersdensity. Data were representative as mean±SEM. (n=5) #p<0.01 compared to control rats (ND without 6-OHDA injection);*p<0.05 compared to ND with 6-OHDA injection; +p>0.05 compared to control rats (ND without 6-OHDA injection).

28 B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

2.3. KD inhibited the decrease of SN and striatal GSHinduced by 6-OHDA

After 2 weeks of 20 μg intrastrial 6-OHDA injection, the levelsof GSH in the SN and striatum were significantly decreased in

Fig. 4 – The levels of striatal DA and its metabolites DOPAC andcompared to control rats (ND without 6-OHDA injection); *p<0.05control rats (ND without 6-OHDA injection); ^p>0.05 compared to

rats fed with ND compared to the ND without 6-OHDAinjection (control) (p<0.01). While the levels of SN and striatalGSH in rats fed with KD increased more significantly than inrats fed with ND after 2 weeks of 6-OHDA injection (p<0.05).There were no difference of GSH levels in the SN and striatum

HVA. Data were representative as mean±SEM. (n=5) #p<0.01compared to ND with 6-OHDA injection; +p>0.05 compared toND with 6-OHDA injection.

Fig. 5 – The levels of SN and striatal GSH. (A) GSH level of SN.(B) GSH level of striatum. Data were representative asmean±SEM. (n=5) #p<0.01 compared to control rats(NDwithout 6-OHDA injection); *p<0.05 compared to NDwith6-OHDA injection; +p>0.05 compared to control rats(ND without 6-OHDA injection).

29B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

between ND and KD rats without intrastriatal 6-OHDAinjection (p>0.05) (Fig. 5).

3. Discussion

In thiswork,we demonstrated that KDprotected dopaminergicneurons of SN against degeneration induced by intrastriatalinjection of 6-OHDA in a rat model of PD. Nissl and TH-positiveneurons increased in rats fed with KD compared to rats fedwith ND after intrastriatal 6-OHDA injection, so did DA, DOPACand GSH. GSH played an important role in the neuroprotectionof KD.

KB, an alternative energy of brain, are produced by the liver(Finn and Dice, 2005) and transported to peripheral tissues(Faria et al., 2007). Most investigators agree that normal serumlevels of ketone bodies can be defined as <0.5 mM (Laffel,1999). There is an increase under conditions of fasting, caloricrestriction, and intake of high-fat and low-carbohydrate dietssuch as KD and the data were not shown in our work. KD is ahigh-fat, low-protein, low-carbohydratediet thathasbeenusedfor decades to treat patientswith refractory epilepsies (Seyfriedet al., 2003). KB are excellent oxidative fuel, free radicalscavengers, and membrane stabilizers (Koustova et al., 2003).Many cases show that KD, or KB (D-beta-hydroxybutyrate,acetoacetate and acetone) have neuroprotections such ascardiopulmonary bypass, status epilepticus and chronic neuro-

logical disorders (Alzheimer's disease and PD) (Tanaka et al.,2002). Our study found KD protected dopaminergic neurons ofSN against degeneration induced by intrastriatal injection of 6-OHDA. And the decrease of DA and its metabolite DOPACinduced by 6-OHDA in the striatum were reversed undercondition of KD.

Oxidative stress is the loss of the balance between reactiveoxidative species (ROS) and the antioxidants, and it has beenimplicated in the mechanism of PD (Zhou et al., 2008). Post-mortem shows that oxidative stress appear in the SN of thebrain in PD patients and it also is observed in animalmodels ofPD (Greenamyre andHastings, 2004;Maguire-Zeiss et al., 2005).Sullivan et al. show that KD can decrease ROS level and exertneuroprotection (Sullivan et al., 2004). GSH system plays amajor role in controlling cellular redox states and is theprimary defense mechanism for peroxide removal from thebrain (Iida et al., 1999). The intracellular GSH level is controlledby a balance between the rate of production or salvage and therate of consumption or loss (Tanaka et al., 2002). GSH is capableof reacting both in vitro and in vivo with 6-OHDA through anucleophilic attack on the 6-OHDA quinone to form 2-S-(glutathionyl)-6-OHDA, and sulfhydryl antioxidants such asGSH and NAC produce a significant reduction in the autoxida-tion of 6-OHDA (Shimizu et al., 2002). On the other hand, 6-OHDA neurotoxicity has also been shown to involve mito-chondra (Kulich et al., 2007). KD can increaseGSHbiosynthesis,enhances mitochondrial antioxidant status, and protectsmtDNA from oxidant-induced damage (Jarrett et al., 2008). Inour present study, 6-OHDA induced GSH level to decrease inratswithNDbut the turnover appeared in ratswith KD. So GSHplayed an important role in the therapeutics of KD.

SinceKD, D-beta-hydroxybutyrate, acetoacetate and acetonehave the effect of neuroprotection, this becomes a new potenttherapeutics for PD. KTX 0101 is the sodium salt of thephysiological KB (D-beta-hydroxybutyrate) and has been usedtrial and proved neuroprotective, while some side effectscome out and the underlying mechanisms still are not clear(Smith et al., 2005). So further studies need to elucidate themechanisms and eliminate the side effects.

4. Experimental procedures

4.1. Animals

Male Wistar rats (weighing 250–300 g, Medical AnimalCenter of Shandong University, Jinan, Shandong Province,China) were used in this work and housed under control oftemperature (23 °C±2 °C) and light (12h/12 h cycles of dayand night). All experiments were carried out under theguide of the Lab Animal Rules of Shandong University.

4.2. Diet regime

Rats were fed with normal diet (ND) and KD as describedby Hae Sook Noh et al. (2006). The ND is composed of fat(5%), protein (20%), carbohydrate (66%), and inert matter(9%). The KD is composed of fat (90.7%), protein (9%), andcarbohydrate (0.3%). The goal of a KD is to produce andmaintain ketonemia.

30 B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

4.3. 6-OHDA lesion

After 2 weeks fed with ND or KD, rats were anaesthetizedby sodium pentobarbital (Pengyuan Co. China) andinjected unilateral stereotaxically with 20 μg 6-OHDAhydrobromide (Sigma) (dissolved in 4 μl saline with 0.02%ascorbic acid) or the corresponding volume of saline(containing 0.02% ascorbic acid) into the left striatum asdescribed by Carmen Henze et al. (2005). Coordinates(tooth bar: ±0.0 mm; anterior/posterior: +1.0 mm; medial/lateral: +3.0 mm; ventral/dorsal: −4.5 mm) were deter-mined from bregma, according to the atlas of Paxinos andWatson (Paxinos and Watson, 1982). After surgery, ratswere maintained to be fed by KD or ND for 2 weeks.

4.4. Tissue preparation

2 weeks after 6-OHDA injection, some rats of every groupwere anesthetized and perfused with 4% paraformalde-hyde. Then the brains were taken out and kept in 4%paraformaldehyde. After 2 days they were transferred intophosphate-buffered saline (PBS) containing 30% sucrose at4 °C. The tissue was sunk for sectioning in order to Nisslstaining in the SN and TH immunohistochemistry in thestriatum and SN. The left striata of some brains weredissected out and kept at −80 °C for measuring DA, DOPACand HVA using HPLC and the levels of GSH.

4.5. Nissl staining

One section was selected every six sections in the SN areaof each brain and eight sections were selected every brain.All the sections from each brain were mathched as closelyas possible for Nissl staining. The sections were mountedon the slides, dried and kept in xylene overnight. The slideswere then dehydrated in alcohol and washed with dH2O.Following the slides were stained with Cresyl Violet (0.5%cresyl violet acetate). The slides were washed with dH2Oanddehydrated in alcohol again. In the end, the slideswerecoversliped. Using microscope the doperminergic neuronsof SN were counted all the eight sections manually of eachbrain. The data were expressed as percent of control rats(ND without 6-OHDA injection).

4.6. TH immunohistochemistry

The sections of SN were picked up as described as Nisslstaining. The samemethodwasused for picking up sectionsof striatum. The sections were collected and washed withphosphate buffer (PBS). Following treatment with 3%hydrogen peroxide in methanol to eliminate endogenousperoxidase, the sections were treated by 0.04% normal goatserum with 0.4% Triton X-100 in PBS including 0.2 g bovineserumalbumin (BSA) for 1 h at 37 °C, followed by incubationwith TH primary antibody (Dingguo Co. China) overnight at4 °C. Slides were then washed three times with PBS andincubated with biotinylated secondary antibody (BoshideCo. China) 37 °C for 1 h. After washing three times with PBS,slides were incubated with ABC (Boshide Co. China) kit for1 h. Slides were developed in diaminobenzidine (DAB)

(Pengyuan Co. China) for 2–20 min. Using microscope thedoperminergic neurons of SN were counted all the sixsectionsmanually of eachbrain. Thedatawere expressedaspercent of control rats (ND without 6-OHDA injection). ForTH-immunoreactivity fibers in the striatum, the opticaldensitywasmeasured using a computerized image analysissystem (Olympus) and the corpus callosumwere measuredas nonspecific background densities. The optical densityvalues of the striatum in the lesioned sides were expressedas a percentage loss of the same side of control rats (NDwithout 6-OHDA injection).

4.7. HPLC analysis

Tissue levels of DA and its primary metabolites (DOPACandHVA)weremeasured usingHPLC. Datawere expressedas ng/g wet weight of tissue.

4.8. Measurement of GSH

The SN and striata were homogenized and homogenatewas precipitated with 10% trichloroacetic acid. Aftercentrifugation, the supernatant was collected and mixedwith 0.3 mol sodium phosphate buffer (pH 8.0) and 5 mL of0.04% 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB). Then itwas incubated for 5 min at room temperature. Theabsorbance value of the sample was read against theblank at 412 nm. GSH concentration was calculated fromthe standard curve. The data were expressed as percent ofcontrol (ND without 6-OHDA injection).

4.9. Statistics analysis

Data were shown as means±SE. Comparisons betweentwo groups (n=5) were analyzed using t-test. Differenceswere considered significant when p≤0.05.

R E F E R E N C E S

Chen, H.Q., et al., 2007. Biochanin A protects dopaminergicneurons against lipopolysaccharide-induced damage throughinhibition of microglia activation and proinflammatory factorsgeneration. Neurosci. Lett. 417, 112–117.

Chinta, S.J., Andersen, J.K., 2006. Reversible inhibition ofmitochondrial complex I activity following chronicdopaminergic glutathione depletion in vitro: implications forParkinson's disease. Free Radic. Biol. Med. 41, 1442–1448.

Faria, M.H., et al., 2007. Ketone bodiesmetabolism during ischemicand reperfusion brain injuries following bilateral occlusion ofcommon carotid arteries in rats. Acta Cir. Bras. 22, 125–129.

Finn, P.F., Dice, J.F., 2005. Ketone bodies stimulatechaperone-mediated autophagy. J. Biol. Chem. 280,25864–25870.

Greenamyre, J.T., Hastings, T.G., 2004. Biomedicine.Parkinson's—divergent causes, convergent mechanisms.Science. 304, 1120–1122.

Henze, C., et al., 2005. Reactive oxidative and nitrogen species inthe nigrostriatal system following striatal 6-hydroxydopaminelesion in rats. Brain Res. 1052, 97–104.

Iida, M., et al., 1999. Dopamine D2 receptor-mediated antioxidantand neuroprotective effects of ropinirole, a dopamine agonist.Brain Res. 838, 51–59.

31B R A I N R E S E A R C H 1 2 8 6 ( 2 0 0 9 ) 2 5 – 3 1

Imamura, K., et al., 2006. D-beta-hydroxybutyrate protects dopa-minergic SH-SY5Y cells in a rotenone model ofParkinson's disease. J. Neurosci. Res. 84, 1376–1384.

Jarrett, S.G., et al., 2008. The ketogenic diet increasesmitochondrial glutathione levels. J. Neurochem. 106,1044–1051.

Koustova, E., et al., 2003. Ketone and pyruvate Ringer'ssolutions decrease pulmonary apoptosis in a rat model ofsevere hemorrhagic shock and resuscitation. Surgery 134,267–274.

Kulich, S.M., et al., 2007. 6-Hydroxydopamine inducesmitochondrial ERK activation. Free Radic. Biol. Med. 43,372–383.

Laffel, L., 1999. Ketone bodies: a review of physiology,pathophysiology and application of monitoring to diabetes.Diabetes Metab. Res. Rev. 15, 412–426.

Leret, M.L., et al., 2002. Deprenyl protects from MPTP-inducedParkinson-like syndrome and glutathione oxidation in ratstriatum. Toxicology 170, 165–171.

Maguire-Zeiss, K.A., et al., 2005. Synuclein, dopamine andoxidative stress: co-conspirators in Parkinson's disease? BrainRes. Mol. Brain Res. 134, 18–23.

Noh, H.S., et al., 2006. Ketogenic diet protects the hippocampusfrom kainic acid toxicity by inhibiting the dissociation of badfrom 14-3-3. J. Neurosci. Res. 84, 1829–1836.

Paxinos, G., Watson, C., 1982. The Rat Brain in StereotaxicCoordinates. Academic Press, Sydney.

Pinnen, F., et al., 2007. Synthesis and study of L-dopa-glutathionecodrugs as new anti-Parkinson agents with free radicalscavenging properties. J. Med. Chem. 50, 2506–2515.

Qu, D., et al., 2007. Role of Cdk5-mediated phosphorylation of Prx2in MPTP toxicity and Parkinson's disease. Neuron 55, 37–52.

Seyfried, T.N., et al., 2003. Role of glucose and ketone bodies in themetabolic control of experimental brain cancer. Br. J. Cancer 89,1375–1382.

Shimizu, E., et al., 2002. Roles of endogenous glutathione levels on6-hydroxydopamine-induced apoptotic neuronal cell death inhuman neuroblastoma SK-N-SH cells. Neuropharmacology 43,434–443.

Sian, J., et al., 1994. Alterations in glutathione levels in Parkinson'sdisease and other neurodegenerative disorders affecting basalganglia. Ann. Neurol. 36, 348–355.

Smith, S.L., et al., 2005. KTX 0101: a potential metabolic approachto cytoprotection in major surgery and neurological disorders.CNS Drug Rev. 11, 113–140.

Sullivan, P.G., et al., 2004. The ketogenic diet increasesmitochondrial uncoupling protein levels and activity. Ann.Neurol. 55, 576–580.

Tanaka, K., et al., 2002. GPI1046 prevents dopaminergicdysfunction by activating glutathione system in the mousestriatum. Neurosci. Lett. 321, 45–48.

Tieu, K., et al., 2003. D-beta-hydroxybutyrate rescuesmitochondrial respiration and mitigates features of Parkinsondisease. J. Clin. Invest. 112, 892–901.

Vanitallie, T.B., et al., 2005. Treatment of Parkinson disease withdiet-induced hyperketonemia: a feasibility study. Neurology64, 728–730.

Zhou, C., et al., 2008. Oxidative stress in Parkinson's disease: amechanism of pathogenic and therapeutic significance. Ann.N. Y. Acad. Sci. 1147, 93–104.