N

CGs

SQ1

Ja

Kb

c

h

••••

a

ARRAA

KGRsC1

PH

1

mcses

h0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

ARTICLE IN PRESSG ModelSL 30546 1–6

Neuroscience Letters xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Neuroscience Letters

jo ur nal ho me page: www.elsev ier .com/ locate /neule t

hronic repetitive transcranial magnetic stimulation enhancesABAergic and cholinergic metabolism in chronic unpredictable mildtress rat model: 1H-NMR spectroscopy study at 11.7 T

ang-Young Kima, Do-Wan Leea, Hyunju -Kimb, Eunjung Bangb,eong-Ho Chaec, Bo-Young Choea,∗

Department of Biomedical Engineering & Research Institute of Biomedical Engineering, College of Medicine, The Catholic University, Seoul, Republic oforeaSeoul Center of Korea Basic Science Institute, Seoul, Republic of KoreaDepartment of Psychiatry, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea

i g h l i g h t s

The CUMS rat model of depression induced the reduction of the sucrose preference.The prefrontal and hippocampal GABA levels were reduced by the CUMS regime.The CUMS-induced GABA deficits were reversed by chronic rTMS treatment.rTMS effects were more pronounced in prefrontal cortex compared to hippocampus.

r t i c l e i n f o

rticle history:eceived 29 December 2013eceived in revised form 22 March 2014ccepted 24 April 2014vailable online xxx

eywords:ABAepetitive transcranial magnetic

a b s t r a c t

Gamma-animobutyric acid (GABA) systems are emerging as targets for development of medicationsfor mood disorders. Deficits in GABA-containing neurons are consistently reported in psychiatric dis-ease, particularly in the prefrontal cortex and hippocampus. Repetitive transcranial magnetic stimulation(rTMS) that use magnetic field to stimulate focal cortical regions with electrical current have a potentialtherapeutic effects with non-invasive and painless method. In this study, we used chronic unpredictablemild stress (CUMS) rat model of depression to investigate the behavioral and neurochemical alter-ations. Furthermore, chronic rTMS treatment effect on neurochemical profile in prefrontal cortex andhippocampus of rats were assessed. The CUMS induced significant reductions in absolute sucrose intake

1

timulation (rTMS)hronic unpredictable mild stress (CUMS)H-NMRrefrontal cortexippocampus

and sucrose preference. In addition, high-resolution H-NMR spectra from brain extracts revealed signifi-cantly reduced prefrontal and hippocampal GABA levels in CUMS rats compared to control. The behavioraland neurochemical changes were reversed by chronic rTMS treatment. Furthermore, chronic rTMS treat-ments results in differential effects on different brain regions. Our results suggest specific and regionallydifferent metabolic response to chronic rTMS treatment in animal model of depression.

42

43

44

45

46

. Introduction

Repetitive transcranial magnetic stimulation (rTMS) that useagnetic field to stimulate focal cortical regions with electrical

urrent has been used for evolving treatment of refractory depres-

Please cite this article in press as: S.-Y. Kim, et al., Chronic repetiticholinergic metabolism in chronic unpredictable mild stress rat modhttp://dx.doi.org/10.1016/j.neulet.2014.04.033

ion [1,2]. Given its non-invasive nature and potential therapeuticffects in neuropsychiatric disorder, the rTMS has been focus of con-iderable and clinical interest in recent years. Furthermore, several

∗ Corresponding author. Tel.: +82 2 2258 7233; fax: +82 2 2258 7760.E-mail address: bychoe@catholic.ac.kr (B.-Y. Choe).

ttp://dx.doi.org/10.1016/j.neulet.2014.04.033304-3940/© 2014 Published by Elsevier Ireland Ltd.

47

48

49

© 2014 Published by Elsevier Ireland Ltd.

meta-analysis revealed that the rTMS showed antidepressant-likeeffects in treatment resistant depression (TRD) [3,4]. Despite of itsbroad use, little information about neither the precise pattern ofbrain activation nor the molecular mechanisms underlying rTMSeffects are known.

Gamma-Animobutyric acid (GABA) is a ubiquitous, inhibitoryneurotransmitter found exclusively in the central nervous system.Previous studies suggested that the serotonergic involvement in

ve transcranial magnetic stimulation enhances GABAergic andel: 1H-NMR spectroscopy study at 11.7 T, Neurosci. Lett. (2014),

depression may linked to the action of GABA [5]. Deficits in GABA-containing neurons are consistently reported in psychiatric disease,particularly in the prefrontal cortex and hippocampus [6]. Forexample, there was a report that the density of cortical GABAergic

50

51

52

53

ING ModelN

2 nce Le

idntMicPthwbn

rihra

tGmsar

2

2

tatUrwweta

2

ggwbTbard

2

wttssmd

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

ARTICLESL 30546 1–6

S.-Y. Kim et al. / Neuroscie

nterneurons in the prefrontal cortex is reduced in major depressiveisorder (MDD) compared to control subjects [7]. Proton mag-etic resonance spectroscopy (MRS) methods allow investigatorso directly/noninvasively quantify GABA levels in the brains. Using

RS methods, Sanacora et al. found that the GABA concentrationsn occipital cortex of medication-free MDD patients were signifi-antly reduced compared with healthy controls [8]. More recently,rice et al. reported that occipital cortex and anterior cingulate cor-ex GABA level in subjects with TRD (i.e., three treatment failures)ad the lowest GABA levels compared to unmedicated subjectsith MDD and healthy controls [9]. Taken together, GABA might

e potential candidate for rTMS-induced changes on the centralervous system.

The chronic unpredictable mild stress (CUMS) procedure canesult in a number of behavioral abnormalities that can be seenn patients with depression [10]. The relevance of CUMS model touman depression has been supported by the observation that theeward deficits are reversed by the treatment with major classes ofntidepressant drugs [11,12].

Considering the above, in the present study, we aimed to inves-igate the behavioral and neurochemical alterations (particularly,ABA) induced by the CUMS and the effects of chronic rTMS treat-ent in rat brain. To the best of our knowledge, few studies

imultaneously investigated behavioral and metabolic changes forssessment of therapeutic effects of rTMS treatment in the CUMSat model.

. Materials and methods

.1. Animals

Animal procedures were in accordance with the National Insti-utes of Health Guidelines for Animal Research (Guide for the Carend Use of Laboratory Animals) and were approved by the Insti-utional Animal Care and Use Committee (IACUC, The Catholicniversity of Korea). Experimentally naïve male Sprague–Dawley

ats weighing 160–180 g were housed under a 12 h light–dark cycleith lights on at 0700 h and had continuous access to food andater. Rats were acclimated for one week for adjustment to the

xperimental conditions and to remove the stress induced duringransportation. The animals were singly housed in order to achieve

better acclimatization in these conditions.

.2. CUMS procedures

The animals were divided by two main groups; the experimentalroup exposed to a variety of mild stressors (N = 10), and controlroup that was deprived of food and water for 4 h preceding theeekly sucrose preference test (N = 10). The animals were matched

y the baseline sucrose preference data before the CUMS initiation.he CUMS regime used in this study was similar to that describedy Bekris et al. [12]. A variety of unpredictable mild stressors werepplied for 4 weeks (details in Supplementary Table 1). The controlats were left undisturbed in their cages under the aforementionedescribed maintenance conditions.

.3. Sucrose preference test

After 1 week adaptation of experimental condition, animalsere first trained to consume 1% (w/v) sucrose solution before

he start of CUMS procedures. Training consisted of three 1 hests (Monday, Wednesday and Friday), in which animals could

Please cite this article in press as: S.-Y. Kim, et al., Chronic repetiticholinergic metabolism in chronic unpredictable mild stress rat modhttp://dx.doi.org/10.1016/j.neulet.2014.04.033

elect between two preweighted bottles, one with 1% (w/v) sucroseolution and one with tap water. The sucrose consumption waseasured by comparing bottle weight before and after the 1-h win-

ow. Baseline was measured before the start of CUMS. The sucrose

PRESStters xxx (2014) xxx–xxx

preference (%) was calculated as follows: preference (%) = [sucrosesolution intake (ml)/total fluid intake (ml)] × 100.

2.4. Chronic rTMS treatment

The main two groups were subdivided by four groups; controlgroup + sham rTMS treatment (N = 5), control group + rTMS treat-ment (N = 5), CUMS group + sham rTMS treatment (N = 5), CUMSgroup + rTMS treatment (N = 5). The animals were trained to exper-imenter’s hands for 3 days to minimize potential effects on theresults, and gradually habituated to the experimenter’s handing.The rats were treated with daily rTMS for 2 weeks. After the ter-mination of CUMS procedures, rTMS was delivered to rats usinga Neopulse stimulator (Neotonus Inc., Marietta, USA) and a 7 cmdiameter figure-8 coil. The coil was held above the rat’s head atclose proximity, with the site of stimulation at frontal cortex level.Stimulation was delivered at a rate of 10 Hz for 10 min. The intensityof the stimulation was 1.4 T at the surface of the coil. The stimula-tion consisted of 20 trains of 50 pulses with a 25-s pause betweeneach successive train. The sham group was only exposed to theacoustic artifact of rTMS without the stimulation itself by holdingthe coil 20 cm above rats brain.

2.5. Methanol–chloroform–water extraction

After the end of rTMS treatment, the animals were sacrificed andthe brains were rapidly removed, dissected on ice. Subsequentlyprefrontal cortex and hippocampus were isolated. The tissues werekept under liquid nitrogen. The metabolites were extracted bymethanol–chloroform–water extraction method [13]. The upperphase (methanol and water) was separated from the lower phase(organic) using a glass syringe.

2.6. Proton NMR spectroscopy and spectral processing

All dried extract samples were dissolved in 99% D2O contain-ing 0.05% TSP and transferred to 5 mm NMR tube. The protonNMR data were acquired on a Varian 500 MHz NMR spectrome-ter using Carr–Purcell–Meiboom–Gill (CPMG) sequence (32 K datapoints, spectral width: 8000 Hz, relaxation delay: 2 s, spin echodelay: 400 �s, acquisition time: 2 s, number of transients: 128, tem-perature: 283 K, experiment time: 12 min). The proton spectra forboth regions (prefrontal cortex and hippocampus) of rats brainswere analyzed using LCModel software [14]. The metabolites fittedblow than 20% Cramer–Rao lower bounds (CRLB) were included forstatistical analysis.

2.7. Statistical analysis

The PASW software package (Window version 18.0, Chicago, IL,USA) was used for the statistical analysis. The body weight, abso-lute sucrose intake and sucrose preference data were split into twodata (i.e., Baseline-to-Week4 and Week5-to-Week6 data). Baseline-to-Week4 data were analyzed using one-way repeated analysis ofvariance (ANOVA) with time as within-subject factor and stressas between-subject factor. Week4-to-Week6 data were analyzedusing two-way repeated ANOVA with time as within-subject fac-tor, and stress and treatment as between-subject factors. After theANOVA test, separate pair-wise comparisons were performed tofind specific differences between groups.

The each metabolite level was normalized with respect to sum

ve transcranial magnetic stimulation enhances GABAergic andel: 1H-NMR spectroscopy study at 11.7 T, Neurosci. Lett. (2014),

of metabolites signals. Non-parametric Friedman two-way ANOVA(stress × treatment) was used for testing the significant differencebetween four groups. Because the test dose not identify wherethe differences occur or how many differences actually occur, the

166

167

168

169

IN PRESSG ModelN

nce Letters xxx (2014) xxx–xxx 3

Mw

3

3

trepwtbwfonrpb

3

reeppweFtnp[(itrhissd

3

elFwistecfTarg

Fig. 1. The effects of chronic unpredictable mild stress (CUMS) from baseline testto week 4 and chronic rTMS effects on body weight (a), absolute sucrose intake(b), and sucrose preference (c) from week 5 to week 6. Each value representsmean ± standard deviation (*** p < 0.001; baseline measurement vs. week 1 to 4,

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

ARTICLESL 30546 1–6

S.-Y. Kim et al. / Neuroscie

ann–Whitney U test was performed for pair-wise comparisonsith metabolites that showed the significance from Friedman test.

. Results

.1. Body weight

For Baseline-to-Week4 data, a repeated one-way ANOVA, withime as within-subject factor and stress as between-subject factor,evealed that there was a statistically significant time and stressffect on body weight [F(4,15) = 300.63, p < 0.001; F(1,18) = 49.53,

< 0.001]. And, there was a significant interaction (time × stress)ith regard to body weight [F(4,15) = 88.91, p < 0.001]. For Week5-

o-Week6 data, Week4 data was used as comparison reference (i.e.,efore rTMS treatment). A repeated two-way ANOVA, with time asithin-subject factor and, stress and treatment as between-subject

actor, revealed that there was a statistically significant time effectn body weight [F(2,15) = 54.17, p < 0.001]. And there were sig-ificant interactions (time × stress, time × stress × treatment) withegard to body weight [F(2,15) = 14.22, p < 0.001; F(2,15) = 4.08,

= 0.039]. The post hoc analysis showed significant differences inody weight between groups (p < 0.05, details in Fig. 1).

.2. Sucrose preference test

For Baseline-to-Week4 data, a repeated one-way ANOVAevealed that there was a statistically significant time and stressffect on both absolute sucrose intake and sucrose prefer-nce [absolute intake: F(4,15) = 9.65, p < 0.001; F(1,18) = 119.37,

< 0.001; preference: F(4,15) = 25.15, p < 0.001; F(1,18) = 332.49, < 0.001]. And, there was a significant interaction (time × stress)ith regard to both absolute sucrose intake and sucrose pref-

rence [absolute intake: F(4,15) = 10.72, p < 0.001; preference:(4,15) = 14.43, p < 0.001]. For Week5-to-Week6 data, a repeatedwo-way ANOVA revealed that there was no statistically sig-ificant time effects on absolute sucrose intake [F(2,15) = 3.28,

= 0.066], while the effects was significant on sucrose preferenceF(2,15) = 5.67, p = 0.015]. And there were significant interactionstime × stress, time × treatment) with regard absolute sucrosentake [F(2,15) = 4.18, p = 0.036; F(2,15) = 4.67, p = 0.027]. Moreover,here were significant interactions (time × stress × treatment) withegard to sucrose preference [F(2,15) = 10.15, p = 0.002]. The postoc analysis showed significant differences in both absolute sucrosentake and sucrose preference between groups. And stressed ratshowed significant decreases in both absolute sucrose intake anducrose preference compared to baseline measurement (p < 0.05,etails in Fig. 1).

.3. 1H-NMR spectroscopy analysis

Fig. 2 shows the representative spectra of rat brain extracts inach group. High-resolution spectra revealed that most of over-apping signals like GABA, Glu, and Gln were clearly separated.ig. 3 shows the normalized metabolites levels that quantifiedith CRLB below than 20%. Total 15 metabolites were included

n further statistical analysis. For the data of hippocampus tissuesamples, non-parametric Friedman two-way ANOVA revealed thathere was a statistically significant difference in metabolites lev-ls (GABA, Cho) between four group [GABA: �2 = 11.14, p = 0.011;holine (Cho): �2 = 14.07, p = 0.003]. In order to find specific dif-erences between groups, we conducted Mann–Whitney U test.

Please cite this article in press as: S.-Y. Kim, et al., Chronic repetiticholinergic metabolism in chronic unpredictable mild stress rat modhttp://dx.doi.org/10.1016/j.neulet.2014.04.033

he Mann–Whitney U test revealed significant increases in GABAnd Cho levels in CUMS + rTMS group compared to CUMS + shamTMS groups (p < 0.05, details in Supplementary Table 2), sug-esting significant rTMS treatment effects on metabolites levels

+ p < 0.05; week 4 vs. week 5 to 6, ## p < 0.005, ### p < 0.001; control group vs.CUMS group or control + sham rTMS vs. CUMS + sham rTMS, p < 0.05, p < 0.005;CUMS + sham rTMS vs. CUMS + rTMS).

in hippocampus. For the data of prefrontal cortex tissues sam-ples, non-parametric Friedman two-way ANOVA revealed thatthere was a statistically significant difference in metabolites levels[creatine (Cr), GABA, glycerophosphocholine (GPC), phosphoryl-choline (PCh), myo-inositol (Ins), and Cho] between four group (Cr:�2 = 12.23, p = 0.007; GABA: �2 = 11.64, p = 0.009; GPC: �2 = 12.29,p = 0.006; PCh: �2 = 10.29, p = 0.016; Ins: �2 = 10.09, p = 0.018; Cho:�2 = 12.64, p = 0.005). The Mann–Whitney U test revealed signif-

ve transcranial magnetic stimulation enhances GABAergic andel: 1H-NMR spectroscopy study at 11.7 T, Neurosci. Lett. (2014),

icant increases in GABA, Cho and Cr + PCr levels in CUMS + rTMSgroup compared to CUMS + sham rTMS groups (p < 0.05, detailsin Supplementary Table 2), suggesting significant rTMS treatmenteffects on the metabolites levels in prefrontal cortex. Averaged

236

237

238

239

ARTICLE IN PRESSG ModelNSL 30546 1–6

4 S.-Y. Kim et al. / Neuroscience Letters xxx (2014) xxx–xxx

F for eacs

GamwtT

4

cwFwitnhe

amsecds[drTm

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

ig. 2. The representative in vitro proton spectra obtained from rat brain extracts

ignals are clearly separated with Glu signals.

ABA spectrum for each group showed significant effects of stressnd rTMS treatment on GABA levels in both brain regions (Supple-entary Fig. 1). The reproducibility of GABA measurements, whichas quantified in terms of coefficient of variations (CVs) was below

han 11% for the spectra of each group (details in Supplementaryable 3).

. Discussion

The present study demonstrated that CUMS procedures in ratsause reductions in absolute sucrose intake and sucrose preference,hich is consistently reported in previous studies [10,12,15,16].

urthermore, the metabolic alterations from extracted rat brainsere observed in CUMS rat model (particularly GABA). The behav-

oral and neurochemical changes were reversed by chronic rTMSreatment, suggesting antidepressant-like effects. The changes ineurochemical levels for both brain regions (prefrontal cortex,ippocampus) may account for the observed different sucrose pref-rence profile.

There was a significant lengthening of inhibitory mechanismsssociated with GABAB receptor mediated inhibitory neurotrans-ission, with higher rTMS frequency [17]. The findings suggest that

ome of the therapeutic effects of rTMS may be coordinated throughnhanced GABA-mediated inhibitory neurotransmission, which isonsistent with the finding that GABAergic neurotransmission isisrupted in MDD [18] and enhanced through either electroconvul-ive therapy (ECT) or treatment with serotonin reuptatke inhibitors8,19]. In the current study, we also found that CUMS-induced

Please cite this article in press as: S.-Y. Kim, et al., Chronic repetiticholinergic metabolism in chronic unpredictable mild stress rat modhttp://dx.doi.org/10.1016/j.neulet.2014.04.033

isrupted GABAergic neurotransmission were reversed by chronicTMS treatment in prefrontal cortex and hippocampus of rats.he antidepressant-like effects of rTMS treatment on other ani-al models of depression like forced swimming test [20], olfactory

h group. Only the spectra of hippocampus tissues are shown. Note that the GABA

bulbectomy [21] were previously reported. We conducted pro-ton NMR spectroscopy measuring metabolites in cytosol fractionin brain extract, which may be slightly different from in vivoexamination measuring intra- and extracelluar metabolites con-tents. However, because extracellular metabolites levels are muchlower than the intracellular one, no significant metabolic differ-ences (especially for GABA) between both measurements wouldbe expected.

Interestingly, we found significant rTMS effects on choliner-gic metabolism (CUMS + sham rTMS vs. CUMS + rTMS) but thechronic stress-induced effects did not reach significance level (con-trol + sham rTMS vs. CUMS + sham rTMS). More specifically, therewere significant increases in prefrontal free Cho and hippocampalPCh and free Cho, while prefrontal GPC levels were significantlydecreased by chronic rTMS treatment compared to CUMS rats. Ren-shaw et al. previously reported an alteration in the metabolism ofcytosolic choline compounds in basal ganglia of depressed subjectswith treatment of fluoxetine [22]. Of choline-containing com-pounds (ChoCC), GPC and PCh contribute to the choline resonancepeak more than 50%. Free choline and acetylcholine are present atmuch lower concentrations and make smaller contributions to thein vivo 1H-MRS choline resonance. Furthermore, the choline res-onance may be partially overlapped with Tau and Ins resonance.Thus, relatively low spectral resolution in vivo proton MR spectralimits the interpretation for altered cholinergic metabolism dueto multiple contributions to the observed ChoCC resonance. Thepresent study overcame the limitation using high-resolution NMRspectroscopy with metabolite extraction method, which allows one

ve transcranial magnetic stimulation enhances GABAergic andel: 1H-NMR spectroscopy study at 11.7 T, Neurosci. Lett. (2014),

to obtain more detailed information (i.e., completely separatedChoCC resonance by GPC, PCh, and free Cho). Although a num-ber of evidence supports the importance of choline metabolismin the expression of mood disorder, it is not clear how cholinergic

298

299

300

301

ARTICLE IN PRESSG ModelNSL 30546 1–6

S.-Y. Kim et al. / Neuroscience Letters xxx (2014) xxx–xxx 5

F respefi n. Nom

naai

aia[msah[bc[emepcddddi

b

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

ig. 3. The hippocampal (a) and prefrontal (c) metabolites levels normalized withts (i.e., CRLB) ((b) hippocampus, (d) prefrontal cortex) from LCModel are also showetabolic differences between groups (* p < 0.05, ** p < 0.005).

eurotransmission mediated by acetylcholine and the levels of GPCnd PCh are related. Our findings for increased PCh and free Cho,nd decreased GPC levels in prefrontal cortex need to be explainedn future study.

Lastly, we observed the significant increase in prefrontal Insnd total creatine (Cr + PCr) levels with chronic rTMS treatmentn CUMS rats. Recently, significantly increased prefrontal Ins levelfter high-frequency rTMS treatment in TRD patients were reported23], which is consistent with our finding. The increased Ins levels

ight be interpreted as restored glial and/or synaptic function bytimulating circuit activity involving regulatory pathway. Total cre-tine peak is composed of Cr and PCr resonances, representing theigh-energy phosphate reserves in the cytosol of neurons and glia24]. The level of creatine in the brain can be significantly alteredy osmotic forces as well as by extracerebral events due to theomplex biosynthetic pathway through liver and kidney enzymes25]. Because of this complexity, it is difficult to attribute differ-nces in creatine simply to local derangements of cellular energyetabolism. Since PCr is linked to ATP through the creatine–kinase

quilibrium, alteration in creatine might be related to changes inrefrontal metabolism. Taken together, our results indicated thathronic rTMS treatment results in different modulatory effect onifferent brain regions. This may be due to brain region locationifference because the magnetic field attenuated along with theistance, and an induced current intensity of different distance is

Please cite this article in press as: S.-Y. Kim, et al., Chronic repetiticholinergic metabolism in chronic unpredictable mild stress rat modhttp://dx.doi.org/10.1016/j.neulet.2014.04.033

istinct. Therefore, the current findings might provide new insightsnto the mechanisms underlying the clinical effects of rTMS.

There are some limitations in this study. First, our results need toe replicated and further clarified in a larger sample size since small

ct to sum of metabolites signals are shown. The corresponding standard errors ofn-parametric Friedman two-way ANOVA (stress × treatment) revealed significant

sub-sample sizes may compromise the power to detect significantdifferences between groups. Second, spectral fitting using LCModelfor high-resolution 1H-NMR spectra from brain extracts are chal-lenging due to the very high demands on the use of correct priorknowledge and line shape. Subtle differences in pH, ionic strength,temperature, protein content, etc., between samples will cause dif-ferences in the NMR-detected peak position and line widths [26].Furthermore, each metabolite is differentially sensitive to theseeffects, thus making a global correction infeasible for a mixture ofmultiple metabolites. To overcome the issues, future study with tar-geted metabolomic approach (i.e., multivariate pattern recognitionanalysis) is needed to elucidate biomarker for treatment response,metabolites patterns or both. And the ultra-high field whole-bodyMRI scanner in the near future would allow more improved quan-tification of metabolites in vivo that would be comparable within vitro data. The benefits are especially expected to be importantfor metabolites having low concentration or overlapping spectralline like GABA. Lastly, future study would be needed to assessrTMS induced-metabolic perturbation in other brain areas relatedto depression.

5. Conclusion

In conclusion, we have revealed that the behavioral andmetabolic alterations were reversed by chronic rTMS treatment,

ve transcranial magnetic stimulation enhances GABAergic andel: 1H-NMR spectroscopy study at 11.7 T, Neurosci. Lett. (2014),

suggesting a role of GABAergic system in the pathophysiology ofdepression. Our results suggest specific and regionally differentmetabolic response to chronic rTMS treatment in animal model ofdepression.

354

355

356

357

ING ModelN

6 nce Le

A

Q2R0MsC

A

fj

R

[

[

[

[

[

[

[

[ Q3

[

[

[

[

[

[

[

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

ARTICLESL 30546 1–6

S.-Y. Kim et al. / Neuroscie

cknowledgments

This study was supported by the program of Basic Atomic Energyesearch Institute (BAERI) (2009-0078390) and a grant (2012-07883) from the Mid-career Researcher Program funded by theinistry of Education, Science & Technology (MEST) of Korea. This

tudy was conducted using Varian 500 MHz NMR system at Seoulenter of Korea Basic Science Institute.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.neulet.2014.04.033.

eferences

[1] W.J. Triggs, N. Ricciuti, H.E. Ward, J. Cheng, D. Bowers, W.K. Goodman, B.M.Kluger, S.E. Nadeau, Right and left dorsolateral prefrontal rTMS treatment ofrefractory depression: a randomized sham-controlled trial, Psychiat. Res. 178(2010) 467–474.

[2] C. Baeken, M.A. Vanderhasselt, J. Remue, S. Herremans, N. Vanderbruggen, D.Zeeuws, L. Santermans, R. De Raedt, Intensive HF-rTMS treatment in refractorymedication-resistant unipolar depressed patients, J. Affect. Disord. 151 (2013)625–631.

[3] F. Padberg, M. George, Repetitive transcranial magnetic stimulation of the pre-frontal cortex in depression, Exp. Neurol. 219 (2009) 2–13.

[4] B. Dell’Osso, G. Camuri, F. Castellano, V. Vecchi, M. Benedetti, S. Bortolussi,A. Altamura, Meta-review of metanalytic studies with repetitive transcranialmagnetic stimulation (rTMS) for the treatment of major depression, Clin. Pract.Epidemol. Ment. Health 7 (2011) 167–177.

[5] D.A. Slattery, S. Desrayaud, I.F. Cryan, GABAB receptor antagonist-mediatedantidepressant-like behavior is serotonin-dependent, J. Pharmacol. Exp. Ther.312 (2005) 290–296.

[6] G.P. Reynolds, M.K. Harte, The neuronal pathology of schizophrenia: moleculesand mechanisms, Biochem. Soc. Trans. 35 (2007) 433–436.

[7] G. Rajkowska, G. O’Dwyer, Z. Teleki, C.A. Stockmeier, J.J. Miguel-Hidalgo,GABAergic neurons immunoreactive for calcium binding proteins are reducedin the prefrontal cortex in major depression, Neuropsychopharmacology 32(2007) 471–482.

[8] G. Sanacora, G.F. Mason, D.L. Rothman, J.H. Krystal, Increased occipi-tal cortex GABA concentrations in depressed patients after therapy withselective serotonin reuptake inhibitors, Am. J. Psychiatry 159 (2002)

Please cite this article in press as: S.-Y. Kim, et al., Chronic repetiticholinergic metabolism in chronic unpredictable mild stress rat modhttp://dx.doi.org/10.1016/j.neulet.2014.04.033

663–665.[9] R.B. Price, D.C. Shungu, X. Mao, P. Nestadt, C. Kelly, K.A. Collins, J.W. Murrough,

D.S. Charney, S.J. Mathew, Amino acid neurotransmitters assess by proton mag-netic resonance spectroscopy: relationship to treatment resistance in majordepressive disorder, Biol. Psychiatry 65 (2009) 792–800.

[

[

PRESStters xxx (2014) xxx–xxx

10] P. Willner, A. Towell, D. Sampson, S. Sophokleous, R. Muscat, Reduction ofsucrose preference by chronic unpredictable mild stress, and its restorationby a tricyclic antidepressant, Psychopharmacology 93 (1987) 358–364.

11] J.L. Moreau, F. Jenck, J.R. Martin, S. Perrin, W.E. Haefely, Effects of repeated mildstress and two antidepressant treatments on the behavioral response to 5HT1Creceptor activation in rats, Psychopharmacology 110 (1993) 140–144.

12] S. Bekris, K. Antoniou, S. Daskas, Z. Papadopoulou-Daifoti, Behavioural and neu-rochemical effects induced by chronic mild stress applied to two different ratstrains, Behav. Brain Res. 161 (2005) 45–59.

13] J.E. Le Belle, N.G. Harris, S.R. Williams, K.K. Bhakoo, A comparison of cell andtissue extraction techniques using high-resolution 1H-NMR spectroscopy, NMRBiomed. 15 (2002) 37–44.

14] S.W. Provencher, Automatic quantitation of localized in vivo 1H spectra withLCModel, NMR Biomed. 14 (2001) 260–264.

15] J. Gronli, R. Murison, B. Bjorvatn, E. Sorensen, C.M. Potas, R. Crsin, Chronic mildstress affects sucrose intake and sleep in rats, Behav. Brain Res. 150 (2004)139–147.

16] J. Gronli, R. Murison, E. Fiske, B. Bjorvatn, E. Sorensen, C.M. Portas, R. Ursin,Effects of chronic mild stress on sexual behavior, locomotor activity andconsumption of sucrose and saccharine solutions, Physiol. Behav. 84 (2005)571–577.

17] Z.J. Daskalakis, B. Moller, B.K. Christensen, The effects of repetitive transcranial

magnetic stimulation on cortical inhibition in healthy human subjects, Exp.Brain Res. 174 (2006) 403–412.

18] G. Sanacora, G.F. Mason, D.L. Rothman, Reduced cortical gamma-aminobutyricacid levels in depressed patients determined by proton magnetic resonancespectroscopy, Arch. Gen. Psychiatry 56 (1999) 1043–1047.

19] G. Sanacora, G.F. Mason, D.L. Rothman, F. Hyder, J.J. Ciarcia, R.B. Ostroff, R.M.Berman, J.H. Krystal, Increased cortical GABA concentrations in depressedpatients receiving ECT, Am. J. Psychiatry 160 (2003) 577–579.

20] P.S. Sachdev, R. McBride, C. Loo, P.M. Mitchell, G.S. Malhi, V. Croker, Effects ofdifferent frequencies of transcranial magnetic stimulation (TMS) on the forcedswim test model of depression in rats, Biol. Psychiatry 51 (2002) 474–479.

21] P. Vieyra-Reyes, Y.S. Mineur, M.R. Picciotto, I. Túnez, R. Vidaltamayo, R.Drucker-Colín, Antidepressant-like effects of nicotine and transcranial mag-netic stimulation in the olfactory bulbectomy rat model of depression, BrainRes. Bull. 77 (2008) 13–18.

22] P.F. Renshaw, B. Lafer, S.M. Babb, M. Fava, A.L. Stoll, J.D. Christensen, C.M.Moore, D.A. Yurgelun-Todd, C.M. Bonello, S.S. Pillay, A.J. Rothschild, A.A. Nieren-berg, J.F. Rosenbaum, B.M. Cohen, Basal ganglia choline levels in depressionand response to fluoxetine treatment: an in vivo proton magnetic resonancespectroscopy study, Biol. Psychiatry 41 (1997) 837–843.

23] H. Zheng, L. Zhang, L. Li, P. Liu, J. Gao, X. Liu, J. Zou, Y. Zhang, J. Liu, Z.Zhang, Z. Li, W. Men, High-frequency rTMS treatment increases left prefrontalmyo-inositol in young patients with treatment-resistant depression, Prog. Neu-ropsychopharmacol. Biol. Psychiatry 34 (2010) 1189–1195.

24] G.J. Kemp, Non-invasive methods for studying brain energy metabolism: whatthey show and what it means, Dev. Neurosci. 22 (2000) 418–428.

ve transcranial magnetic stimulation enhances GABAergic andel: 1H-NMR spectroscopy study at 11.7 T, Neurosci. Lett. (2014),

25] B. Ross, T. Michaelis, Clinical applications of magnetic resonance spectroscopy,Magn. Reson. Q. 10 (1994) 191–247.

26] A.M. Weljie, J. Newton, P. Mercier, E. Carlson, C.M. Slupsky, Targeted profiling:quantitative analysis of 1H NMR metabolomics data, Anal. Chem. 78 (2006)4430–4442.

448

449

450

451

452