Electro-Acupuncture Stimulation Acts on the Basal Ganglia Output Pathwayto Ameliorate Motor Impairment in Parkinsonian Model Rats

Jun Jia, Bo Li, Zuo-Li Sun, Fen Yu, Xuan Wang, and Xiao-Min WangCapital Medical University

The role of electro-acupuncture (EA) stimulation on motor symptoms in Parkinsons disease (PD) has notbeen well studied. In a rat hemiparkinsonian model induced by unilateral transection of the medialforebrain bundle (MFB), EA stimulation improved motor impairment in a frequency-dependent manner.Whereas EA stimulation at a low frequency (2 Hz) had no effect, EA stimulation at a high frequency (100Hz) significantly improved motor coordination. However, neither low nor high EA stimulation couldsignificantly enhance dopamine levels in the striatum. EA stimulation at 100 Hz normalized the MFBlesion-induced increase in midbrain GABA content, but it had no effect on GABA content in the globuspallidus. These results suggest that high-frequency EA stimulation improves motor impairment inMFB-lesioned rats by increasing GABAergic inhibition in the output structure of the basal ganglia.

Keywords: electro-acupuncture, DA, GABA, Rota-Rod treadmill, Parkinsons disease

Acupuncture is a popular alternative therapy in patients withParkinsons disease (PD) and might provide some benefit in theclinical setting (Rabinstein & Shulman, 2003). A survey of PDpatients demonstrated that patients who received acupuncture re-ported improvement of their symptoms (Shulman et al., 2002). Inaddition, acupuncture is very safe and well tolerated, and it has noknown interactions with medications (Eng, Lyons, Greene, &Pahwa, 2006).

PD is a neurodegenerative disorder characterized by the pro-gressive loss of dopaminergic neurons in the substantia nigra parscompacta (SNc). The behavioral deficits of PD, including restingtremor, rigidity, and bradykinesia, are expressed when striataldopamine (DA) depletion exceeds 6080% (Bernheimer, Birk-mayer, Hornykiewicz, Jellinger, & Seitelberger, 1973). It hasfollowed from these observations that the best course of therapymight be to increase the DA content of the striatum. However,increasing evidence suggests that a poor relationship exists be-tween the dopamine content of the striatum and the improvementof motor symptoms during treatment (Gash et al., 1996; Tseng,Baetge, Zurn, & Aebischer, 1997). On the other hand, parkinso-nian movement disorders are often associated with abnormalities

in GABA neuron activity in the substantia nigra pars reticulata(SNr) (Zhou, Matta, & Zhou, 2008). Firing intensity and patternsare often altered in the SNr GABAergic output pathways in theparkinsonian brain (Hutchison et al., 2004). Inactivation of SNr ina primate model of PD has been shown to ameliorate parkinsonianmotor symptoms, further demonstrating the importance of SNr inmotor control (Wichmann, Kliem, & DeLong, 2001).

Previous work has demonstrated that high-frequency electro-acupuncture (EA) stimulation significantly reduced abnormal ro-tational behavior and enhanced the survival of degenerating dopa-minergic neurons following medial forebrain bundle (MFB)axotomy (Liang et al., 2003). These effects might be attributable toenhanced synthesis and release of neurotrophic factors (Liang etal., 2002, 2003) and alleviation of inflammatory reactions in theSN (Liu et al., 2004). Furthermore, EA stimulation reversed thelesion-mediated decrease of Substance P content in the ventralmidbrain (Jia et al., 2009). These studies suggested that EA stim-ulation may result in changes in neuronal activity in the basalganglia (Zangen & Hyodo, 2002).

Therefore, the present study was conducted to evaluate theeffects of different frequencies of EA stimulation (0, 2, or 100 Hz)on motor behavior and dopamine content in the striatum in MFB-transected rats. Changes in GABA content in the midbrain and theglobus pallidus (GP) were also measured to assess the modulatoryeffect of EA stimulation on neuronal activity in the basal ganglia.

Method

Animal Care and MFB Axotomy

Adult male Wistar rats weighing 200230 g were obtained fromthe laboratory animal center, Capital Medical University, andhoused in a standard 12-hr lightdark cycle with ad libitum accessto food and water. The rats were anesthetized with chloral hydrate(350 mg/kg ip) and then positioned in a stereotaxic apparatus(David Kopf Instruments, Tujunga, CA) with the mouth bar set at

Jun Jia, Bo Li, Zuo-Li Sun, Fen Yu, Xuan Wang, and Xiao-Min Wang,Department of Physiology, Key Laboratory for Neurodegenerative Disor-ders of the Ministry of Education, Capital Medical University, Beijing,Peoples Republic of China.

This study was supported by the National Basic Research Program ofChina (2006CB500700), the National Natural Science Foundation of China(30472245), Key Project of the Beijing Natural Science Foundation and theBeijing Education Committee (kz200510025014), the Talent Training Planof Beijing (20081d 0501800206), and the Capital Medical University fund(107420).

Correspondence concerning this article should be addressed to Xiao-MinWang, Department of Physiology, Capital Medical University, 10 Xi TouTiao, You An Men Wai, Beijing 100069, PR China. E-mail: xmwang@ccmu.edu.cn

Behavioral Neuroscience 2010 American Psychological Association2010, Vol. 124, No. 2, 305310 0735-7044/10/$12.00 DOI: 10.1037/a0018931

305

3.3 mm. MFB lesions were performed using a retractableScouten wire knife as described previously (Tseng et al., 1997).The experimental procedures were approved by the Committee onAnimal Care and Usage, Capital Medical University, and allefforts were made to minimize animal suffering.

EA StimulationRats were randomly divided into five groups: the sham group,

the MFB-lesioned group, and the MFB-lesioned group followed by0-, 2-, and 100-Hz EA stimulation. The EA stimulation wasadministered on Day 2 following MFB lesioning as describedpreviously (Liang et al., 2003). Two stainless steel needles 0.25mm in diameter and 5 mm in length were inserted obliquely at theacupuncture point Dazhui (Du 14, just below the spinous processof the vertebra prominens) and horizontally at the Baihui point (Du21, at the midpoint of the line connecting the two ears). For the0-Hz group, the needles were placed into the acupoints, but noelectrical stimulation was administered. Bidirectional square waveelectrical pulses (0.2-ms duration, 2 Hz and 100 Hz), designated asEA, were given for a total of 30 min each day, 6 days per week.The duration of EA treatment was limited to 4 weeks. The intensityof the stimulation was increased stepwise from 1 to 2 mA and thento 3 mA, with each step lasting 10 min. EA was administered toawake, unrestrained animals while in their cages.

Rota-Rod TestThe motor coordination of rats was evaluated on Rota-Rod

treadmills (Stoelting Company, Wood Dale, IL). The apparatusconsisted of a base platform and a rotating rod with a diameter of6 cm and a nonslippery surface. The accelerating speed of theRota-Rod was set to increase from 6 rpm/min to 40 rpm/minwithin 2 min. The rats were placed on the treadmill, and the timerswere set with acceleration programmed to automatically stop whenthe rat fell off. The maximal cutoff time was set to 600 s. Rats weretested on the Rota-Rod at 7, 14, 21, and 28 days following MFBlesioning.

Tissue Collection and ProcessingFive rats from each group were randomly selected 28 days after

MFB transection. After decapitation, the brain was quickly re-moved. The ventral midbrain (the section of the SN, bregma from4.8 to 6.3 mm) and GP were dissected, rapidly frozen on dryice, and stored at 80C.

High-Performance Liquid Chromatography (HPLC)Concentrations of DA were quantified by a modified method of

HPLC combined with electrochemical detection (HPLC-ECD) asdescribed elsewhere (Kotake, Heffner, Vosmer, & Seiden, 1985).The concentrations of GABA were determined using HPLC withlaser-induced fluorescence detection. On the day of assay, tissuesamples were weighed and then homogenized in 150 l of 0.1 mMacetic acid as described previously (Carlson, Behrstock, Tobin, &Salamone, 2003). Briefly, samples or standards were derivatizedwith the same volume of the o-phthaldialdehyde reagent solution(P0532 of Sigma Chemical, St. Louis, MO). The resulting mixturewas automatically injected by a refrigerated autoinjector into a

symmetry shield-C18 reverse-phase column (150 3.8 mm, 5 mparticle size; Waters Corporation, Milford, MA). The mobile phaseconsisted of 50 mM sodium acetate (pH was adjusted to 6.8 usingacetic acid), methanol, and tetrahydrofuran. The flow rate was 1.0ml/min, maintained with two Shimadzu LC (Agilent, Santa Clara,CA) pumps. Putative GABA peaks were identified by comparingretention times of GABA standard (SigmaAldrich) to retentiontimes of the sample peak. The concentrations were estimated byrationing peak areas of GABA and its external standard (AgilentChemstation analytical software). The running time for each de-termination was 40 min.

Statistical AnalysisData were expressed as means SEM. Statistical significance

was assessed using one-way analysis of variance followed by theNewmanKeuls post hoc test of difference between groups. Avalue of p .05 was considered statistically significant.

Results

EA Stimulation Improves Motor CoordinationThe Rota-Rod treadmill test was performed following EA stim-

ulation of all experimental rats to investigate the effects of differ-ent frequencies of EA stimulation on motor coordination, TheMFB-lesioned rats demonstrated a decreased treadmill occupancytime compared with the sham group under all conditions (seeFigure 1A). From Week 2, the rats receiving 100-Hz EA stimula-tion demonstrated significantly increased treadmill occupancytime compared with MFB-lesioned rats ( p .05). These improve-ments in motor coordination were retained for the duration of thehigh-frequency EA treatment. These data demonstrate that 100-HzEA stimulation can enhance motor coordination in MFB-transected rats.

EA Stimulation Did Not Increase DA Contentin the Striatum

Wire knife lesions of the MFB dramatically decreased the DAcontent in the ipsilateral striatum in hemiparkinsonian rats comparedwith the sham group ( p .01). In contrast, MFB lesions did not resultin significant changes in DA levels on the contralateral side (seeFigure 1B). The HPLC-ECD analysis data demonstrate that neitherlow- nor high-frequency EA stimulation could enhance the DA levelin the ipsilateral striatum. These results suggest that the improvementin motor coordination is not fully dependent on the restoration of thestriatal DA content, as seen in our EA-treated hemiparkinsonian rats.

Effect of EA on GABA Content in the GPand the Midbrain

To investigate the effects of MFB lesioning and EA treatmenton the output structures of the basal ganglia, we examined theGABA content in the GP and the ventral midbrain area. Our datashow that the GABA content of the ipsilateral GP was higher inMFB-lesioned rats than in sham groups, but no changes wereobserved on the contralateral side (see Figure 2A). This result isconsistent with previous reports that ipsilateral DA lesions induce

306 BRIEF COMMUNICATIONS

a marked increase in the GABA levels in the output nuclei (the GPiand the SNr) of the basal ganglia (Windels, Carcenac, Poupard, &Savasta, 2005). No significant differences in the GABA contentwere detected between rats with the MFB lesion and hemiparkin-sonian rats receiving EA stimulation of any frequency. In contrastto the results from the GP, 100-Hz EA stimulation normalizedmidbrain GABA from the increased levels observed in MFB-lesioned rats. This effect was not seen in hemiparkinsonian ratssubjected to 0- or 2-Hz EA stimulation (see Figure 2B).

DiscussionThe present study demonstrates that high-frequency EA stimu-

lation can improve the motor coordination of MFB-lesioned rats

through a mechanism that involves the potentiation of GABAergicinhibition in the basal ganglia output structure (SNr).

Previous studies have confirmed that a slower rate of degener-ation and progression of motor impairment occurs following MFBaxotomy (Brecknell, Dunnett, & Fawcett, 1995; Knusel et al.,1992). In this model, 100-Hz EA stimulation (but not 2 Hz or 0Hz) significantly improves motor coordination as manifested byperformance on the Rota-Rod test. However, the mechanismsunderlying this improvement are not fully understood.

The striatum is the primary projection target of SNc dopami-nergic neurons. Movement disorders can result when striatal DAdepletion exceeds 70%. The restoration of DA levels in the stria-tum is thought to be critical for treatment of PD. However, manystudies have demonstrated motor improvement without restorationof striatal DA content. For example, intracerebral injection of glial

Figure 2. Effect of electro-acupuncture (EA) stimulation on (A) globuspallidus (GP) and (B) ventral midbrain GABA content in medial forebrainbundle (MFB)-transected rats. The GABA content was measured by high-performance liquid chromatography. The histograms represent the meancontent on the unlesioned side and the lesioned side of MFB-lesioned rats.The values are expressed as percentages SEM of the sham values; n 5; p .05 versus sham rats and # p .05 versus MFB rats.

Figure 1. (A) Effects of electro-acupuncture (EA) stimulation on motorcoordination of rats measured by the Rota-Rod treadmill test. The timespent on the treadmill among the five groups was recorded at the end of 1,2, 3, and 4 weeks post-transection. Values are represented as means SEM; n 715; p .05, p .01 versus the medial forebrain bundle(MFB)-lesioned group. (B) Effect of EA stimulation on striatum dopamine(DA) content of MFB-transected rats 28 days after MFB transection. Thestriatum dopamine content was measured by high-performance liquidchromatography with electrochemical detection. The results are given asmeans SEM; n 610; p .01 versus sham lesioned side.

307BRIEF COMMUNICATIONS

cell-derived neurotrophic factor (GDNF) has significantly im-proved parkinsonian behavioral symptoms but does not affect DAlevels in the caudate nucleus or putamen of 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated monkeys (Gash et al.,1996). Moreover, continuous release of low levels of GDNF nearthe SN has also been reported to protect nigral dopaminergicneurons from MFB axotomy-induced degeneration and to signif-icantly improve apomorphine-induced rotational behavior. How-ever, GDNF infusion did not prevent the loss of DA in the striatum(Tseng et al., 1997). Our previous experiments (Jia et al., 2009;Liang et al., 2003) and present study demonstrate that 100-Hz EAstimulation can significantly improve movement disorders inMFB-lesioned rats without affecting DA levels in the striatum(Liang et al., 2003). Analysis of the DA content in the striatumshows that there is no significant difference between the MFB-lesioned group and the groups treated with 2-Hz or 100-Hz EA.However, rats treated with high-frequency EA stimulation showeda slightly higher striatal DA content than those treated with low-frequency EA stimulation. These effects of 100-Hz EA stimulationmight be attributable to enhanced synthesis and release of neuro-trophic factors (Liang et al., 2002, 2003) and alleviation of inflam-matory reactions in the SN (Liu et al., 2004). L-dopa-inducedrotational behavior has suggested that DA levels in the striatum areonly elevated for a short time, whereas DA levels in the SN remainelevated until the behavioral effects of L-dopa have subsided. Thissuggests that elevated DA levels in the SN maintain circlingbehavior when striatal DA levels have begun to decline (Robertson& Robertson, 1989). Morphological evidence also revealed thatEA at 100 Hz protected axotomized dopaminergic neurons fromdegeneration in the SN (Jia et al., 2009). Therefore, a SN-basedmechanism potentially involved in the improvement of motordisorders should be investigated.

SNc dopaminergic neurons release DA not only via axonalterminals that affect neurotransmission within the striatum but alsovia dendrites, some of which densely protrude into the SNr(Nieoullon, Cheramy, & Glowinski, 1977). This anatomical ar-rangement makes it possible for dendritically released DA tointeract with SNr neurons, thus affecting the activity of this keyoutput structure of the basal ganglia (Windels & Kiyatkin, 2006b).In MFB-lesioned rats, we found that high-frequency EA stimula-tion prevented the degeneration of both dopaminergic cell bodiesin the SNc and dopaminergic neurons that arborize in the SNr (Jiaet al., 2009). These data indicate that high-frequency EA stimula-tion may induce an increase in DA release within the SNr, whichmay result in significant alterations in the output pathways of SNrneurons. However, it remains unclear how increased dendritic DArelease might affect SNr GABAergic projection neurons(Gonzalez-Hernandez & Rodriguez, 2000).

The SNr is usually considered a major output nucleus of thebasal ganglia, and striato-nigral GABAergic projections are pri-mary inputs to SNr neurons. However, D1 DA receptors wereidentified on GABA (Yung et al., 1995) and glutamate terminals(Rosales et al., 1997) synapsing on SNr neurons. Therefore, theeffects of DA on SNr neurons may be indirect, resulting frompresynaptic modulation of GABA or glutamate (GLU) inputs, thetwo opposing forces regulating the activity of these cells. Windelsand Kiyatkin (2006a) have studied the effects of iontophoretic DAand amphetamine on SNr neurons in awake, unrestrained rats.Their results showed that the activity state of GABAergic SNr

neurons may be modulated by dendritic release of DA throughactivation of D1 DA receptors on the terminals of striato-nigralefferents. It has also been suggested that increased somatodendriticDA release can decrease the activity of SNr neurons by increasingGABA input. On the other hand, in vitro work also suggests thatDA may affect SNr activity via modulation of glutamatergic inputs(Wittmann, Marino, & Conn, 2002). Somatodendritically releasedDA may play an additional role in regulating GLU release fromsubthalamic terminals in the SN (Hatzipetros & Yamamoto, 2006).It indicated that somatodendritically released DA in the SN pro-vides negative feedback control on overstimulated or phasic GLUrelease from subthalamic nucleus (STN) terminals within the SNby activation of D2 dopamine heteroreceptors. Moreover, lesion orhigh-frequency stimulation of the STN reduces most of the motorimpairments associated with PD (Parkin et al., 2001; Benazzouz etal., 2000). This effect is presumably mediated by a reduction in theactivity of the subthalamo-nigral pathway.

Although previous studies have indicated that GAD67 mRNAlevels are increased in the GP after dopaminergic neuronal degen-eration (Billings & Marshall, 2004; Soghomonian & Chesselet,1992), this increase in GAD67 mRNA is believed to be secondaryto activation of excitatory subthalamo-pallidal projections. In thisstudy, we found that EA stimulation has no effect on the increasedGABA content in GP in MFB-lesioned rats. This suggests that EAstimulation does not entirely modulate the activity of GP neurons,which is affected by the two processes of degeneration of dopa-minergic neurons: activation of STN inputs to GP neurons and theremoval of nigro-pallidal DA transmission (Billings & Marshall,2004; Soghomonian & Chesselet, 1992).

The GABA content in the midbrain also increased in hemipar-kinsonian rats, which is consistent with other reports (Boulet et al.,2006). Some of this nigral GABA may result from an increase inGP neuronal activity from GABAergic pallido-nigral projectionsto the SNr (Celada, Paladini, & Tepper, 1999). In anesthetizedhemiparkinsonian rats, GP lesions abolish the increase in SNrGABA levels induced by STN high-frequency stimulation (Win-dels et al., 2005). However, SNr inhibition may also result fromactivation of the intranigral axon collateral network of GABAergicSNr cells (Mailly, Charpier, Menetrey, & Deniau, 2003). Accord-ing to the classical model, direct striatonigral GABAergic path-way activation inhibits the tonically active GABAergic neurons ofthe SNr, abolishing target nucleus inhibition in premotor structuresin the thalamus and brainstem (Chevalier & Deniau, 1990). Thus,the inhibition of SNr neurons may be important in mediatingmovement hyperactivity. Here, we show that high-frequency EAstimulation decreases the content of GABA in the midbrain ofMFB-lesioned rats. This implies that EA results in increasedsomatodendritic DA release, resulting in inhibition of the activityof SNr neurons. In keeping with this, high-frequency EA stimu-lation was found to antagonize the MFB lesion-induced increasesin GAD67 mRNA levels in the midbrain (Jia et al., 2009). Previousstudies (Guridi et al., 1996; Salin, Manrique, Forni, &Kerkerian-Le Goff, 2002) have shown that STN high-frequencystimulation or subthalamotomy reverses the DA lesion-mediatedincrease in GAD67 mRNA levels in SNr neurons. The observedreduction of GAD67 mRNA levels in the midbrain indicates thatEA stimulation may reduce inhibitory excitatory efferent activityfrom the output pathways of the basal ganglia. Therefore, EA

308 BRIEF COMMUNICATIONS

treatment may affect GABA transmission to target brain regions,especially to SNr neurons in the basal ganglia output structure.

Rats receiving 0-Hz EA stimulation exhibited no improvementin motor coordination. This observation indicates that electricalstimulation, and not just needle placement, is necessary for thepositive effects of EA therapy. In addition, low-frequency stimu-lation (2 Hz) showed no significant change compared with theMFB-lesioned group. This suggests that different frequencies ofEA stimulation may affect different brain regions or that certainbasal ganglia structures may be less sensitive to the low-frequencystimulation.

Electrostimulation at the Dazhui (Du 14) and Baihui (Du 21)acupoints produces a neuroprotective effect on dopaminergic neu-rons in the SN. Park et al. (2003) also demonstrated that acupunc-ture at the Yanglingquan (GB 34) and Taichong (LR 3) acupointsreduces the degeneration of dopaminergic neurons. However, acu-puncture at nonacupoint or nondirectly related acupoints does nothalt the degeneration of dopaminergic neurons (Kim et al., 2005).These results provide evidence for the existence of acupoint spec-ificity. This is consistent with evidence from fMRI studies (Zhanget al., 2004), although additional research is necessary to furthercorroborate this phenomenon.

The GDNF-induced changes in DA levels seen in rhesus mon-keys were limited to the SN, ventral tegmental area, and GP.MPTP-induced depletion of DA in the caudate nucleus and puta-men was not reversed by GDNF administration (Gash et al., 1996).These results suggest that the SN, ventral tegmental area, and GPall play important roles in regulating motor functions. In contrast,it has also been reported that after lesioning with6-hydroxydopamine, DA released per pulse from residual termi-nals in the striatal slices is increased relative to the control, whichmay serve a compensatory function to maintain dopaminergiccontrol over striatal function. However, future studies are neededto determine whether EA stimulation induces changes in DAregulation (release and uptake) or DA receptors in the striatum thatcontribute to the recovery of movement functions.

References

Benazzouz, A., Gao, D. M., Ni, Z. G., Piallat, B., Bouali-Benazzouz, R., &Benabid, A. L. (2000). Effect of high-frequency stimulation of thesubthalamic nucleus on the neuronal activities of the substantia nigrapars reticulata and ventrolateral nucleus of the thalamus in the rat.Neuroscience, 99, 289295.

Bernheimer, H., Birkmayer, W., Hornykiewicz, O., Jellinger, K., & Seitel-berger, F. (1973). Brain dopamine and the syndromes of Parkinson andHuntington. Clinical, morphological and neurochemical correlations.Journal of the Neurological Sciences, 20, 415455.

Billings, L. M., & Marshall, J. F. (2004). Glutamic acid decarboxylase 67mRNA regulation in two globus pallidus neuron populations by dopa-mine and the subthalamic nucleus. The Journal of Neuroscience, 24,30943103.

Boulet, S., Lacombe, E., Carcenac, C., Feuerstein, C., Sgambato-Faure, V.,Poupard, A., & Savasta, M. (2006). Subthalamic stimulation-inducedforelimb dyskinesias are linked to an increase in glutamate levels in thesubstantia nigra pars reticulata. The Journal of Neuroscience, 26,1076810776.

Brecknell, J. E., Dunnett, S. B., & Fawcett, J. W. (1995). A quantitativestudy of cell death in the substantia nigra following a mechanical lesionof the medial forebrain bundle. Neuroscience, 64, 219227.

Carlson, B. B., Behrstock, S., Tobin, A. J., & Salamone, J. D. (2003). Brain

implantations of engineered GABA-releasing cell suppress tremor in ananimal model of parkinsonism. Neuroscience, 119, 927932.

Celada, P., Paladini, C. A., & Tepper, J. M. (1999). GABAergic control ofrat substantia nigra dopaminergic neurons: Role of globus pallidus andsubstantia nigra pars reticulata. Neuroscience, 89, 813825.

Chevalier, G., & Deniau, J. M. (1990). Disinhibition as a basic process inthe expression of striatal functions. Trends in Neurosciences, 13, 277280.

Eng, M. L., Lyons, K. E., Greene, M. S., & Pahwa, R. (2006). Open-labeltrial regarding the use of acupuncture and yin tui na in Parkinsonsdisease outpatients: A pilot study on efficacy, tolerability, and quality oflife. Journal of Alternative and Complementary Medicine, 12, 395399.

Gash, D. M., Zhang, Z., Ovadia, A., Cass, W. A., Yi, A., Simmerman,L., . . . Gerhard, G. A. (1996, March 21). Functional recovery inparkinsonian monkeys treated with GDNF. Nature, 380, 252255.

Gonzalez-Hernandez, T., & Rodriguez, M. (2000). Compartmental orga-nization and chemical profile of dopaminergic and GABAergic neuronsin the substantia nigra of the rat. Journal of Comparative Neurology,421, 107135.

Guridi, J., Herrero, M. T., Luquin, M. R., Guillen, J., Ruberg, M., Laguna,J., . . . Obeso, J. A. (1996). Subthalamotomy in parkinsonian monkeys.Behavioral and biochemical analysis. Brain, 119, 17171727.

Hatzipetros, T., & Yamamoto, B. K. (2006). Dopaminergic and GABAer-gic modulation of glutamate release from rat subthalamic nucleus effer-ents to the substantia nigra. Brain Research, 1076, 6067.

Hutchison, W. D., Dostrovsky, J. O., Walters, J. R., Courtemanche, R.,Boraud, T., Goldberg, J., & Brown, P. (2004). Neuronal oscillations inthe basal ganglia and movement disorders: Evidence from whole animaland human recordings. The Journal of Neuroscience, 24, 92409243.

Jia, J., Sun, Z., Li, B., Pan, Y., Wang, H., Wang, X., . . . Wang, X. (2009).Electro-acupuncture stimulation improves motor disorders in parkinso-nian rats. Behavioural Brain Research, 205, 214218.

Kim, Y. K., Lim, H. H., Song, Y. K., Lee, H. H., Lim, S., Han, S. M., &Kim, C. J. (2005). Effect of acupuncture on 6-hydroxydopamine-induced nigrostratal dopaminergic neuronal cell death in rats. Neuro-science Letters, 384, 133138.

Knusel, B., Beck, K. D., Winslow, J. W., Rosenthal, A., Burton, L. E.,Widmer, H. R., . . . Hefti, F. (1992). Brain-derived neurotrophic factoradministration protects basal forebrain cholinergic but not nigral dopa-minergic neurons from degenerative changes after axotomy in the adultrat brain. The Journal of Neuroscience, 12, 43914402.

Kotake, C., Heffner, T., Vosmer, G., & Seiden, L. (1985). Determinationof dopamine, norepinephrine, serotonin and their major metabolic prod-ucts in rat brain by reverse-phase ion-pair high performance liquidchromatography with electrochemical detection. Pharmacology Bio-chemistry and Behavior, 22, 8589.

Liang, X. B., Liu, X. Y., Li, F. Q., Luo, Y., Lu, J., Zhang, W. M., . . . Han, J. S.(2002). Long-term high-frequency electro-acupuncture stimulation preventsneuronal degeneration and up-regulates BDNF mRNA in the substantianigra and ventral tegmental area following medial forebrain bundle axo-tomy. Brain Research. Molecular Brain Research, 108, 5159.

Liang, X. B., Luo, Y., Liu, X. Y., Lu, J., Li, F. Q., Wang, Q., . . . Han, J. S.(2003). Electro-acupuncture improves behavior and upregulates GDNFmRNA in MFB transected rats. NeuroReport, 14, 11771181.

Liu, X. Y., Zhou, H. F., Pan, Y. L., Liang, X. B., Niu, D. B., Xue, B., . . .Wang, X. M. (2004). Electro-acupuncture stimulation protects dopami-nergic neurons from inflammation-mediated damage in medial forebrainbundle-transected rats. Experimental Neurology, 189, 189196.

Mailly, P., Charpier, S., Menetrey, A., & Deniau, J. M. (2003). Three-dimensional organization of the recurrent axon collateral network of thesubstantia nigra pars reticulata neurons in the rat. The Journal of Neu-roscience, 23, 52475257.

Nieoullon, A., Cheramy, A., & Glowinski, J. (1977, March 24). Release ofdopamine in vivo from cat substantia nigra. Nature, 266, 375377.

309BRIEF COMMUNICATIONS

Park, H. J., Lim, S., Joo, W. S., Yin, C. S., Lee, H. S., Lee, H. J., . . .Chung, J. H. (2003). Acupuncture prevents 6-hydroxydopamine-inducedneuronal death in the nigrostriatal dopaminergic system in the rat Par-kinsons disease model. Experimental Neurology, 180, 9398.

Parkin, S., Nandi, D., Giladi, N., Joint, C., Gregory, R., Bain, P., . . . Aziz, T. Z.(2001). Lesioning the subthalamic nucleus in the treatment of Parkinsonsdisease. Stereotactic and Functional Neurosurgery, 77, 6872.

Rabinstein, A. A., & Shulman, L. M. (2003). Acupuncture in clinicalneurology. Neurologist, 9, 137148.

Robertson, G. S., & Robertson, H. A. (1989). Evidence that L-dopa-induced rotational behavior is dependent on both striatal and nigralmechanisms. The Journal of Neuroscience, 9, 33263331.

Rosales, M. G., Martinez-Fong, D., Morales, R., Nunez, A., Flores, G.,Gongora-Alfaro, . . . Aceves, J. (1997). Reciprocal interaction betweenglutamate and dopamine in the pars reticulata of the rat substantia nigra:A microdialysis study. Neuroscience, 80, 803810.

Salin, P., Manrique, C., Forni, C., & Kerkerian-Le Goff, L. (2002). High-frequency stimulation of the subthalamic nucleus selectively reversesdopamine denervation-induced cellular defects in the output structuresof the basal ganglia in the rat. The Journal of Neuroscience, 22, 51375148.

Shulman, L. M., Wen, X., Weiner, W. J., Bateman, D., Minagar, A.,Duncan, R., & Konefal, J. (2002). Acupuncture therapy for the symp-toms of Parkinsons disease. Movement Disorders, 17, 799802.

Soghomonian, J. J., & Chesselet, M. F. (1992). Effects of nigrostriatallesions on the levels of messenger RNAs encoding two isoforms ofglutamate decarboxylase in the globus pallidus and entopeduncularnucleus of the rat. Synapse, 11, 124133.

Tseng, J. L., Baetge, E. E., Zurn, A. D., & Aebischer, P. (1997). GDNFreduces drug-induced rotational behavior after medial forebrain bundletransection by a mechanism not involving striatal dopamine. The Journalof Neuroscience, 17, 325333.

Wichmann, T., Kliem, M. A., & DeLong, M. R. (2001). Antiparkinsonianand behavioral effects of inactivation of the substantia nigra pars reticu-

lata in hemiparkinsonian primates. Experimental Neurology, 167, 410424.

Windels, F., Carcenac, C., Poupard, A., & Savasta, M. (2005). Pallidalorigin of GABA release within the substantia nigra pars reticulata duringhigh-frequency stimulation of the subthalamic nucleus. The Journal ofNeuroscience, 25, 50795086.

Windels, F., & Kiyatkin, E. A. (2006a). Dopamine action in the substantianigra pars reticulata: Iontophoretic studies in awake, unrestrained rats.European Journal of Neuroscience, 24, 13851394.

Windels, F., & Kiyatkin, E. A. (2006b). GABAergic mechanisms inregulating the activity state of substantia nigra pars reticulata neurons.Neuroscience, 140, 12891299.

Wittmann, M., Marino, M. J., & Conn, P. J. (2002). Dopamine modulatesthe function of group II and group III metabotropic glutamate receptorsin the substantia nigra pars reticulata. Journal of Pharmacology andExperimental Therapeutics, 302, 433441.

Yung, K. K., Bolam, J. P., Smith, A. D., Hersch, S. M., Ciliax, B. J., &Levey, A. I. (1995). Immunocytochemical localization of D1 and D2dopamine receptors in the basal ganglia of the rat: Light and electronmicroscopy. Neuroscience, 65, 709730.

Zangen, A., & Hyodo, K. (2002). Transcranial magnetic stimulation in-duces increases in extracellular levels of dopamine and glutamate in thenucleus accumbens. NeuroReport, 13, 24012405.

Zhang, W. T., Jin, Z., Luo, F., Zhang, L., Zeng, Y. W., & Han, J. S. (2004).Evidence from brain imaging with fMRI supporting functional specific-ity of acupoints in humans. Neuroscience Letters, 354, 5053.

Zhou, F. W., Matta, S. G., & Zhou, F. M. (2008). Constitutively activeTRPC3 channels regulate basal ganglia output neurons. The Journal ofNeuroscience, 28, 473482.

Received September 15, 2009Revision received November 30, 2009

Accepted January 15, 2010

310 BRIEF COMMUNICATIONS