Antidepressant drugs in convulsive seizures: Pre-clinical evaluation of duloxetine in mice

Accepted Manuscript Antidepressant drugs in convulsive seizures: Pre-clinical evaluation of duloxetine in mice Danielle Santana-Coelho, José Rogerio Souza-Monteiro, Ricardo S.O. Paraense, Guilherme L. Busanello, Gabriela P.F. Arrifano, Jackson R. Mendonça, Mauro E.P. Silveira-Junior, Luiz Fernando F. Royes, Maria Elena Crespo-López PII:

S0197-0186(16)30146-2

DOI:

10.1016/j.neuint.2016.06.001

Reference:

NCI 3879

To appear in:

Neurochemistry International

Received Date: 2 December 2015 Revised Date:

27 May 2016

Accepted Date: 7 June 2016

Please cite this article as: Santana-Coelho, D., Souza-Monteiro, J.R., Paraense, R.S.O., Busanello, G.L., Arrifano, G.P.F., Mendonça, J.R., Silveira-Junior, M.E.P., Royes, L.F.F., Crespo-López, M.E., Antidepressant drugs in convulsive seizures: Pre-clinical evaluation of duloxetine in mice, Neurochemistry International (2016), doi: 10.1016/j.neuint.2016.06.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

1

ANTIDEPRESSANT

DRUGS

IN

CONVULSIVE

2

EVALUATION OF DULOXETINE IN MICE

SEIZURES:

PRE-CLINICAL

3

Danielle Santana-Coelho1, José Rogerio Souza-Monteiro1, Ricardo S. O. Paraense1, Guilherme

5

L. Busanello2, Gabriela P.F. Arrifano1, Jackson R. Mendonça1, Mauro E. P. Silveira-Junior2,

6

Luiz Fernando F. Royes2, Maria Elena Crespo-López1*.

RI PT

4

8

1

9

Federal do Pará (UFPA), Belém, Brazil

SC

7

M AN U

Laboratório de Farmacologia Molecular; Instituto de Ciências Biológicas (ICB), Universidade

10

2

11

Universidade Federal de Santa Maria (UFSM), Santa Maria, Brazil

Laboratório de Bioquímica do Exercício, Centro de Educação Física e Desportos (CEFD),

12

*Corresponding author:

14

Maria Elena Crespo López

15

Laboratório de Farmacologia Molecular

16

Instituto de Ciências Biológicas; Universidade Federal do Pará (UFPA)

17

Rua Augusto Corrêa 01, Campus do Guamá.

18

66075-110 Belém-PA, Brasil

19

Phone: +55 91 32018212; Fax: +55 91 32017930

20

E-mail: [email protected]

21

AC C

EP

TE D

13

ACCEPTED MANUSCRIPT

22

ABSTRACT

23

Convulsive seizures (CS) are deleterious consequences of acute cerebral insults and prejudicial

24

events in epilepsy, affecting more than 50 million people worldwide. Molecular mechanisms of

25

depression

26

neurotransmission provoking oxidative stress (OS). OS is intimately linked to the origin and

27

evolution of CS and is modulated by antidepressant and anticonvulsant drugs. Although newer

28

antidepressants have exhibited a possible protective role in CS, studies analyzing serotonin and

29

norepinephrine reuptake inhibitors merit to be further investigated. Thus, this study challenged

30

the traditional model of pentylenetetrazol-induced CS, with only one administration of

31

duloxetine. Male Swiss mice were treated with duloxetine (dose corresponding to the therapeutic

32

range for human depression or greater, by allometric calculation; 10, 20 or 40 mg/Kg), 30 min

33

before pentylenetetrazol. Behavioral and electroencephalographic alterations were monitored.

34

Lipid peroxidation, nitrites and catalase and superoxidase activities were measured in cortex.

35

Behavioral and electroencephalographic results suggested a possible biphasic effect of

36

duloxetine on CS, with anticonvulsant actions at therapeutic doses and a proconvulsant effect at

37

higher doses. Duloxetine (20 mg/Kg) also prevented lipid peroxidation and decreased catalase

38

and superoxide dismutase activities in the cerebral cortex, with no influence on nitrites levels.

39

These data demonstrated an anticonvulsant effect of duloxetine in CS for the first time. This

40

extra anticonvulsant effect may allow the doses of anticonvulsants to be reduced, causing fewer

41

side effects and possibly decreasing morbidity and mortality due to drug interactions in

42

polytherapy.

include

an

imbalance

between

excitatory

and

M AN U

TE D

EP

AC C

43 44

inhibitory

RI PT

epilepsy

SC

and

Keywords: duloxetine, seizures, SNRI, antidepressant, anticonvulsant, oxidative stress

ACCEPTED MANUSCRIPT

45

Abbreviations: convulsive seizures (CS), pentylenetetrazol (PTZ), selective serotonin reuptake

46

inhibitors

47

electroencephalogram (EEG), malondialdehyde (MDA), catalase (CAT), superoxide dismutase

48

(SOD), tricyclic antidepressants (TCAs), selective norepinephrine reuptake inhibitor (NERI),

49

total time spent in seizure (TTSS).

and

norepinephrine

50 51

Running title: Duloxetine in convulsive seizures

52

AC C

EP

TE D

M AN U

53

reuptake

inhibitors

(SNRIs),

RI PT

serotonin

SC

(SSRIs),

ACCEPTED MANUSCRIPT

54

1. INTRODUCTION Seizures can originate as isolated responses to acute neurological insults or alterations in

56

homeostasis in the brain due to acute disease, such as stroke and head trauma. Also, convulsive

57

seizures are major events in epilepsy, a complex neurological disorder affecting roughly 50

58

million people worldwide (approximately 80% of them from developing countries (WHO, 2012),

59

such as Brazil) and characterized by abnormal hypersynchronous neural activity. This abnormal

60

activity manifests as chronic recurrence of unprovoked and spontaneous seizures, resulting in

61

devastating effects on patients.

SC

RI PT

55

One of the most frequent psychiatric comorbidities in epilepsy is depression, which

63

affects one out of every three patients with epilepsy (Kanner et al., 2012), but the molecular

64

mechanisms are not completely understood. In the last decade, data have supported an

65

association between the high comorbidity of the two disorders and common pathogenic

66

mechanisms, such as decreased GABAergic neurotransmission and increased glutamatergic

67

activity (Kanner, 2012). Interestingly, this imbalance between excitatory and inhibitory

68

neurotransmission is intimately linked to the generation of epileptic seizures (Staley, 2015) and

69

is responsible for oxidative stress in both pathologies and comorbidity (Puttachary et al., 2015;

70

Behr et al., 2012; Shin et al., 2011).

TE D

EP

Oxidative stress is an imbalance between the generation of oxidant species and the

AC C

71

M AN U

62

72

activities of antioxidant defense systems, resulting in an overproduction of free radicals, such as

73

reactive oxygen species and nitric oxide (Puttachary et al., 2015; Cardenas-Rodriguez et al.,

74

2013). Patients with depression have decreased antioxidant enzyme activities and increased

75

oxidative stress biomarkers (Behr et al., 2012). Oxidative stress is also intimately linked to

76

epileptogenic processes, refractoriness to treatment in epileptic patients and worsening of the

ACCEPTED MANUSCRIPT

clinical status of patients with comorbidities (Puttachary et al., 2015; Cardenas-Rodriguez et al.,

78

2013; Martinc et al., 2012; Rowles and Olsen, 2012). The deleterious consequences of oxidative

79

stress on the central nervous system (CNS) have also been shown in several models of

80

experimental epilepsy, such as amygdala kindling, pentilenetetrazol (PTZ) kindling, and acute

81

PTZ-induced seizure models (Aguiar et al., 2013; Souza et al., 2013; Frantseva et al., 2000).

82

Moreover, the therapeutic efficacy of anticonvulsant and antidepressant drugs is associated with

83

the influence of these treatments on oxidative stress (Akpinar et al., 2014; Payandemehr et al.,

84

2012; Martinc et al., 2012).

SC

RI PT

77

The association between antiepileptic drugs and antidepressants in patients is very

86

common in clinical practice. Antidepressants are the third class of drugs most used by patients

87

with epilepsy (Wilner et al., 2014). However, the use of antidepressants in epilepsy is

88

controversial because of evidence supporting an increase in seizure severity, especially with

89

older generations of drugs (e.g., tricyclic antidepressants (TCAs) and bupropion) (Jobe and

90

Browning, 2005; Cardamone et al., 2013). Interestingly, some of the newer antidepressants, such

91

as selective serotonin reuptake inhibitors (SSRIs) and serotonin and norepinephrine reuptake

92

inhibitors (SNRIs), have been highlighted as potential anticonvulsant drugs (Jobe and Browning,

93

2005; Cardamone et al., 2013). A single dose of sertraline, for example, showed more potent

94

anticonvulsant effects than similar doses of anticonvulsant drugs as carbamazepine, lamotrigine

95

or phenytoin in a model with 4-aminopyridine (Sitges et al., 2016). The most studied among

96

these drugs are the SSRIs citalopram and fluoxetine, especially in models with PTZ (Cardamone

97

et al., 2013), which is currently considered a gold standard for screening possible anticonvulsant

98

compounds (Souza et al., 2013; Yuen and Troconiz, 2015). However, results with both drugs

99

have been contradictory, with studies reporting proconvulsant consequences.

AC C

EP

TE D

M AN U

85

ACCEPTED MANUSCRIPT

On the other hand, the possible anticonvulsant effect of SNRIs has been scarcely studied,

101

although previous studies with venlafaxine, the most studied SNRI, have demonstrated a potent

102

action against convulsive seizures (Borowicz et al., 2011; Ahern et al., 2006). Still, high doses of

103

venlafaxine (75 mg/Kg or more) showed a proconvulsant effect (Santos et al., 2002). Also

104

reboxetine, a selective norepinephrine reuptake inhibitor (NERI), has demonstrated effects in

105

epilepsy (Ahern et al., 2006; Vermoesen et al., 2011; Vermoesen et al., 2012; Popławska et al.,

106

2015; Kumar et al., 2016).

SC

RI PT

100

In line with this idea, the preliminary results in our lab have indicated a possible role for

108

the SNRI duloxetine in convulsive seizures (Coelho et al., 2014). No other work on duloxetine

109

and its possible effects on seizures has been carried out except a recent report showing the

110

efficacy of chronic treatment with duloxetine in a model of genetic absence seizure with

111

depressive-like behavior (Citraro et al., 2015).

M AN U

107

Therefore, the aim of this study was to challenge the traditional model of PTZ-induced

113

convulsive seizures with only one administration of duloxetine using doses in the therapeutic

114

range for human depression or above and to analyze a possible protective role of the drug against

115

convulsive seizures and seizure-related oxidative stress.

118

EP

117

2. MATERIALS AND METHODS

AC C

116

TE D

112

119

2.1. Animals and ethical aspects

120

Male Swiss mice (25-30 g) were housed under standard conditions (21 ± 2°C; 12 h light/dark

121

cycle) with food and water ad libitum. This study was conducted in accordance with the Ethical

122

Principles of Animal Experimentation suggested by the NIH Guide for the Care and Use of

ACCEPTED MANUSCRIPT

Laboratory Animals and ARRIVE guidelines and it was approved by the Committee for Ethics in

124

Experimental Research with Animals of the Federal University of Pará (license number BIO150-

125

13). A total number of 98 animals, randomly grouped, were used in the present study: 58 for

126

convulsive behavior and biochemical tests (n=10 for PTZ-treated groups and n=6 for groups

127

without PTZ) and 40 for electrocorticographic recordings (n=10 for all groups). Animals for

128

electrocorticographic recordings were submitted to surgical procedures before any treatment (see

129

below). No animal died naturally (without euthanasia) from all treatments carried out in the

130

present study. All efforts were made to reduce the number of animals used and to minimize their

131

suffering.

M AN U

SC

RI PT

123

132

2.2. Surgical procedures

134

A set of animals was anesthetized with intraperitoneal Equithesin and placed in a rodent

135

stereotaxic apparatus. Under stereotaxic guidance, two stainless steel screw electrodes were

136

placed over the ipsilateral parietal cortex, along with a ground lead positioned over the nasal

137

sinus. Bipolar nichrome wire teflon-insulated depth electrodes (100 µm) were implanted

138

unilaterally into the ipsilateral hippocampus (coordinates relative to bregma: AP -4 mm, ML 3

139

mm, DV 3 mm). Electroencephalography was performed 3 days after surgery.

EP

AC C

140

TE D

133

141

2.3. Treatments

142

All mice were treated intraperitoneally with duloxetine (10-40 mg/Kg; Lili, USA) or saline.

143

After 30 minutes, all animals were injected with PTZ (60 mg/Kg i.p.; Sigma, Brazil) or saline

144

(Souza-Monteiro et al., 2015) and electroencephalography and behavioral analysis of seizures

145

carried out with continuous monitoring.

ACCEPTED MANUSCRIPT

146

2.4. Electroencephalographic recordings

148

Before treatment, animals were transferred to a Plexiglas cage (25x25x40 cm) and habituated for

149

30 min before electroencephalography. Each animal was then connected to a 100x headstage pre-

150

amplifier (model #8202-DSE3) in a low-torque swivel (Pinnacle Technology Inc, Lawrence, KS,

151

USA) and the electroencephalogram (EEG) recorded using a PowerLab 16/30 data acquisition

152

system (AD Instruments, Castle Hill, Australia). EEG signals were amplified, filtered (0.1 to

153

50.0 Hz, bandpass), digitized (sampling rate 1 kHz), and stored in a PC for off-line analysis.

154

Routinely, a 10-min baseline recording was obtained to establish an adequate control period.

155

After recording the baseline, animals were treated as described above. The animals were

156

observed for the appearance of clonic and generalized tonic–clonic convulsive episodes for 20

157

min based on Ferraro et al. (1999), who described clonic convulsions as episodes characterized

158

by typical partial clonic activity affecting the face, head, vibrissae, and forelimbs. Generalized

159

convulsive episodes were considered to be generalized whole-body clonus involving all four

160

limbs and tail, rearing, and wild running and jumping followed by sudden loss of upright posture

161

and autonomic signs, such as hypersalivation and defecation. The appearance of spontaneous

162

epileptiform events was defined as: isolated sharp waves (≥1.5× baseline); multiple sharp waves

163

(≥2× baseline) in brief spindle episodes (≥1 s, ≥5 s); multiple sharp waves (≥2× baseline) in long

164

spindle episodes (≥5 s); spikes (≥2× baseline) plus slow waves; multispikes (≥2× baseline, ≥3

165

spikes/complex) plus slow waves; or major seizure (repetitive spikes plus slow waves

166

obliterating background rhythm, ≥5 s) (McColl et al., 2003). For quantitative analysis of

167

amplitude on the EEG, we averaged the amplitude over the 20 min of observation. Cortical

168

video-EEG recordings were analyzed off-line using standard functions in LabChart 7.2 software

AC C

EP

TE D

M AN U

SC

RI PT

147

ACCEPTED MANUSCRIPT

169

(AD Instruments). Mice with artifact-prone EEG recordings were excluded from qualitative EEG

170

analysis.

171

2.5. Behavioral analysis of seizures

173

After PTZ injection, the set of animals for behavioral and biochemical analyses were

174

continuously monitored for 20 min as described above. Latencies to first myoclonic jerk and first

175

generalized tonic–clonic convulsion, as well as the total time spent in tonic-clonic convulsions,

176

were recorded (Souza-Monteiro et al., 2015). The animals were then sacrificed by cervical

177

dislocation and the cerebral cortex homogenized in cold 10 mM Tris-HCl buffer (pH 7.4) and

178

stored at 4ºC for biochemical analysis.

179

M AN U

SC

RI PT

172

2.6. Quantitation of lipid peroxidation

181

Lipid peroxidation in the cerebral cortex was spectrophotometrically assessed using

182

malondialdehyde (MDA) as an indicator (Esterbauer and Cheeseman, 1990). Briefly, the samples

183

were centrifuged at 2500g for 10 min and a solution containing methanesulfonic acid and N-

184

methyl-phenyl indole (10.3 mM in acetonitrile) diluted in methanol (1:3) added to the

185

supernatants for 40 min at 45°C. Absorbance was measured at 570 nm and compared to the

186

absorbance of standard concentrations of MDA. The lipid peroxidation values were corrected for

187

the protein concentration of each sample and expressed as nanomole of MDA per milligram of

188

protein.

AC C

EP

TE D

180

189 190

2.7. Nitrite levels

ACCEPTED MANUSCRIPT

Nitrite levels were measured using Griess reagent (1:1 0.1% N-(1-naphthyl)ethylene-diamine-

192

dihydrochloride and 1% sulfanilamide in 5% phosphoric acid) to generate azo compounds

193

(Green et al., 1981). Briefly, aliquots of the homogenates were centrifuged at 21,000 g for 10

194

min at 4ºC and the supernatants incubated at room temperature with Griess reagent for 20 min.

195

Absorbance was measured at 550 nm and compared to the absorbance of standard solutions of

196

sodium nitrite. The nitrite levels were corrected for the protein concentration of each sample and

197

expressed as micromole of nitrites per milligram of protein.

SC

RI PT

191

198

2.8. Assay of catalase activity

200

The activity of the enzyme catalase (CAT) was measured based on the hydrogen peroxide (H2O2)

201

decomposition rate (Aebi, 1984). Briefly, samples of homogenates were centrifuged at 16,000 g

202

for 30 min at 4ºC. The supernatant (100 µl) was then treated with 600 µl of 10 mM potassium

203

phosphate buffer (pH 7.6) and 50 µl of H2O2. The decomposition of H2O2 was recorded in a

204

spectrophotometer at 240 nm for 120 seconds. CAT activity was expressed as units (U) per

205

milligram of protein considering that one unit decomposes 1 µmol of H2O2 per minute at pH 7.0

206

and 37ºC.

TE D

EP

207

M AN U

199

2.9. Determination of superoxide dismutase activity

209

The activity of the enzyme superoxide dismutase (SOD) was measured based on the capacity of

210

the enzyme to inhibit the auto-oxidation of epinephrine (Misra and Fridovich, 1972). Briefly,

211

after centrifugation of samples of homogenates at 2500 rpm for 10 min at 4ºC, 10 to 30 µl of

212

supernatant was incubated with sodium carbonate buffer (50 mM, pH 10.2) and adrenaline (1

213

mM) at 33ºC. Absorbance was continuously monitored at 480 nm and SOD activity expressed as

AC C

208

ACCEPTED MANUSCRIPT

214

units (U) per milligram of protein considering one unit as the amount of enzyme necessary to

215

inhibit 50% of the adrenochrome at pH 10.2 and 33ºC.

216

2.10. Total protein content

218

Total protein content was determined in all samples as described elsewhere (Bradford, 1976).

219

Aliquots of the homogenates were incubated with Bradford reagent (5% ethanol, 8.5%

220

phosphoric acid, 0.25% Coomassie Brilliant Blue G-250) for 2 min at room temperature. The

221

absorbance of each sample was measured at 595 nm and compared to the absorbance of standard

222

solutions of bovine serum albumin.

SC

M AN U

223

RI PT

217

2.11. Statistical analysis

225

Data analysis was performed using the software GraphPad Prism 5.0. P<0.05 was considered

226

significant for all analyses. The normality of the data was tested by the Kolmogorov-Smirnov

227

test. The method of Bartlett was used to detect significant differences between standard

228

deviations (SDs). Parametric and homoscedastic data were showed as mean ± standard error of

229

the mean (SEM) and they were analyzed using two-way ANOVA, followed by the Bonferroni

230

post-hoc test when appropriate. Non-parametric and/or heteroscedastic data were presented as

231

median ± interquartile range and the Kruskal-Wallis test, followed by the Dunn post-hoc test,

232

was applied. Additionally, when a comparison between only two groups was carried out, both

233

Mann-Whitney and Student t test (with Welch correction for different SDs) were used. To avoid

234

misunderstandings, statistical method was included in each figure.

235 236

AC C

EP

TE D

224

ACCEPTED MANUSCRIPT

237

3. RESULTS

238

3.1. Behavioral effects of treatment with duloxetine on convulsive seizures

240

All treatments with duloxetine (10, 20, and 40 mg/Kg) significantly increased the latency to first

241

myoclonic jerk after PTZ treatment (Fig. 1). Interestingly, a dose of 20 mg/Kg duloxetine

242

produced a more potent effect, increasing the latency to the first tonic-clonic seizure (Fig. 1).

243

Data of total time spent in seizures (TTSS) showed Gaussian distribution but different standard

244

deviations between groups, so, non-parametric Kruskal-Wallis test was used revealing a value of

245

P=0.08 (Fig. 2, left). Taking into account that TTSS is a behavior parameter with high variability

246

and the latter P value was really close to significance, an additional analysis, comparing animals

247

treated with PTZ and those treated with 40 mg/Kg of duloxetine and PTZ, was carried (Fig. 2,

248

right). In this analysis, duloxetine significantly increased the TTSS as showed by the same

249

results of both Student t test (with Welch correction for different standard deviations) (Fig. 2,

250

right) and Mann-Whitney test (data not shown). The same analysis with only two groups for the

251

doses of 10 and 20 mg/Kg of duloxetine presented no significant differences between groups

252

(data not shown), suggesting a possible proconvulsant effect of the highest dose of duloxetine.

SC

M AN U

TE D

EP

254

AC C

253

RI PT

239

255

3.2. Influence of duloxetine on electrical alterations caused by seizures

256

EEG revealed that treatment with saline and/or duloxetine did not affect the amplitude of the

257

electroencephalographic waves (Figs. 3 and 4). On the other hand, the administration of PTZ (60

258

mg/Kg) significantly increased the wave amplitude compared to basal levels (Figs. 3 and 4).

259

Although 10 mg/Kg duloxetine did not alter the amplitudes provoked by PTZ, 20 mg/Kg

ACCEPTED MANUSCRIPT

260

duloxetine partially prevented the increase in amplitude. In contrast, 40 mg/Kg duloxetine

261

potentiated the effect of PTZ, resulting in higher amplitudes than all other groups.

262

3.3. Pre-treatment with duloxetine is associated to a decreased seizure-related oxidative

264

stress

265

The administration of PTZ significantly increased the level of lipid peroxidation (one of the main

266

deleterious consequences of oxidative stress) in the cerebral cortex of animals. This increase was

267

not detected when 20 mg/Kg of duloxetine was previously administrated to the animals (Fig. 5).

268

However, the group treated with 40 mg/Kg of duloxetine before PTZ showed similar levels of

269

lipid peroxidation to those of animals that received PTZ alone.

270

One of the major free radicals capable of causing lipid peroxidation is nitric oxide. However,

271

none of the treatments used in this study altered the basal nitrite levels, an indirect marker of

272

nitric oxide production (Fig. 6).

273

Other important molecular mechanisms associated with oxidative stress include the protective

274

role of antioxidant enzymes, such as CAT and SOD. The administration of PTZ significantly

275

decreased the enzymatic activities of both CAT and SOD (Fig. 7). Animals pre-treated with

276

duloxetine (20 and 40 mg/Kg) and then with PTZ showed no significant difference with control

277

group in CAT activity. Also, similar SOD activity was found in both control group and animals

278

treated with 20 mg/Kg duloxetine and PTZ.

280

SC

M AN U

TE D

EP

AC C

279

RI PT

263

4. DISCUSSION

281

Our study demonstrates for the first time that duloxetine has significant anticonvulsant

282

activity against convulsive seizures and seizure-related oxidative stress. Briefly, ten mg/Kg of

ACCEPTED MANUSCRIPT

duloxetine only increased latency to first myoclonic jerk with no effect on other parameters.

284

Twenty mg/Kg of duloxetine increased both latency to first myoclonic jerk and latency to the

285

first tonic-clonic seizure and reduced electrocortical activity of convulsive animals. Forty mg/Kg

286

of duloxetine increased the latency to first myoclonic jerk (an anticonvulsant effect), but it also

287

increased the total time spent in seizures and the electrocortical activity of convulsive animals

288

(both pro-convulsant effects). Although the total time spent in seizures (besides the effects on

289

electrocortical activity) may be a more representative parameter of the effect on seizures than the

290

latency to the first myoclonic jerk, the proconvulsant effect of this drug must be confirmed with

291

additional studies.

M AN U

SC

RI PT

283

292

Duloxetine inhibits the reuptake of both serotonin and noradrenaline and, although it is

293

usually used in clinical practice as an antidepressant drug, our preliminary results previously

294

pointed to an important role for this drug in seizures (Coelho et al., 2014). Usual posology for human depression is 30-120 mg of duloxetine per day, reaching

296

plasma levels of 5-135 ng/ml (Volonteri et al., 2010). During the preparation of this manuscript,

297

new data became available online regarding a protective role of chronic administration of similar

298

doses of duloxetine for 7 weeks in a model of genetic absence epilepsy with depressive-like

299

symptoms (Citraro et al., 2015). Treatment with these doses of duloxetine reduced electric

300

changes (spike wave discharges in the EEG) due to absence seizures with no effect on depressive

301

behavior (forced swimming test). However, the link between the molecular and epileptic

302

phenotype elicited by duloxetine is still poorly explored.

EP

AC C

303

TE D

295

Moreover, typical absence is non-convulsive epilepsy, with a prevalence of only 10% of

304

all epilepsies in children ≤ 15 years of age (the population where this type of epilepsy is more

305

common) (Berg et al., 2014). Consequently, convulsive seizures are more frequent in epileptic

ACCEPTED MANUSCRIPT

306

patients. Additionally, convulsive seizures can be also found as a grave consequence of other

307

pathologies, such as cerebral trauma and stroke, in non-epileptic patients. So, it is extremely

308

important to know the duloxetine behavior in convulsive seizures. Our experimental model included PTZ, a selective blocker of the chloride channel

310

coupled to the GABAA receptor complex. Acute administration of PTZ is an important model of

311

myoclonic and generalized tonic-clonic seizures and currently considered the gold standard for

312

screening potential anticonvulsant drugs (Souza et al., 2013; Yuen and Troconiz, 2015).

313

Moreover, the physiological consequences of convulsive seizures in humans and the response to

314

treatment may be more accurately predicted by this model than others (Yuen and Troconiz,

315

2015). Recent works characterized the behavior of seizures in this model, with the three

316

parameters used in our work (latencies and duration) (Oliveira et al., 2016; Souza-Monteiro et

317

al., 2015; Mehrzadi et al., 2015; Rodrigues et al., 2012).

M AN U

SC

RI PT

309

Our results demonstrate that only one dose of 20 mg/Kg duloxetine is enough to

319

significantly increase the latencies to the first myoclonic jerk and first tonic-clonic seizure and to

320

decrease the amplitudes of discharges caused by seizures. Based on allometric calculation

321

(Reagan-Shaw et al., 2008), this dose of duloxetine corresponds in humans to 113.5 mg

322

approximately, remaining in the therapeutic range for depression (30-120 mg).

EP

EEG and behavioral recordings indicate that the higher dose of duloxetine had adverse

AC C

323

TE D

318

324

effects on the generation and duration of seizures induced by PTZ. These experimental data

325

suggest that changes in the CNS induced by a high dose of duloxetine may affect susceptibility

326

to seizurogenic stimuli or events followed by spontaneous epileptiform activity elicited by PTZ.

327

Although additional studies are necessary, these data may suggest a biphasic effect of duloxetine,

328

i.e., an anticonvulsant action at low doses that becomes proconvulsant at higher doses. Similar

ACCEPTED MANUSCRIPT

329

biphasic modulations have been described for other antidepressants (e.g., mirtazapine or

330

citalopram) (Killic et al., 2011; Payandemehr et al., 2012), but this is the first time that this effect

331

may be shown for an SNRI. Interestingly, a biphasic effect of duloxetine may explain recently published data from

333

patients indicating that large doses of this drug promote the establishment of tonic-clonic

334

seizures (Pellicciari et al., 2012). Supporting this idea, also high doses of the SNRI venlafaxine

335

worsened the severity of seizures and deaths in a PTZ-model with rats (Santos et al., 2002).

336

Although, in the latter work, no anticonvulsant effect was detected for low doses of venlafaxine,

337

an interesting study with patients recently demonstrated that therapeutic doses of the SNRI

338

venlafaxine for 5 months significantly decreased the number of seizures of psychogenic origin

339

(Pintor et al., 2012).

M AN U

SC

RI PT

332

The origin of this biphasic effect is still not well known. For example, serotonergic

341

receptors have demonstrated the ability to modulate convulsions in experimental models.

342

Although no data are available for duloxetine, recent studies with the SSRI citalopram have

343

demonstrated that its anticonvulsant activity is mediated by the 5-HT3 receptor, with no

344

participation of this receptor in the proconvulsant effect at higher doses of the drug (Li et al.,

345

2014; Payandemehr et al., 2012). Activation of the 5-HT3 receptor mitigates the reduction in

346

GABA levels in the cortex and hippocampus of animals treated with PTZ (Li et al., 2014), which

347

could contribute to the anticonvulsant activity.

EP

AC C

348

TE D

340

Moreover, considering that duloxetine has had significant anticonvulsant effects in

349

models as different as genetic absence seizures (Citraro et al., 2015) and PTZ-induced

350

convulsions (Coelho et al., 2014; and this work), the contribution of additional molecular

351

mechanisms to its anticonvulsant effect cannot be discarded. For example, the acute

ACCEPTED MANUSCRIPT

pharmacological inhibition of norepinephrine reuptake by the blockade of norepinephrine

353

transporters has been shown to increase the latency time and decrease the severity of crises in

354

different seizure models (Ahern et al., 2006; Borowicz et al., 2011; Popławska et al., 2015;

355

Kumar et al., 2016). Noradrenergic system plays a role in the control of seizure activity,

356

especially affecting the limbic system. The blockade of the norepinephrine transport with drugs

357

as reboxetine (a NERI) increases the noradrenergic transmission by activation of α2 and β2

358

receptors, inhibiting acute limbic seizures (Vermoesen et al., 2011; Vermoesen et al., 2012).

359

Interestingly, reboxetine also showed proconvulsant effects with the chronic treatment, as

360

revealed by a reduction of the seizure threshold (Ahern et al., 2006). Still, this proconvulsant

361

action would be because the compensatory mechanisms (as the down-regulation of tyrosine

362

hidroxilase, a key enzyme for norepinephrine synthesis) to the maintained blockade more than

363

an effect of the drug per se (Ahern et al., 2006).

M AN U

SC

RI PT

352

Although duloxetine showed lower affinity for molecular targets (receptors and

365

transporters) in dopaminergic innervations than in serotoninergic and noradrenergic systems

366

(Bymaster et al., 2001), the contribution of dopamine action must not be discarded, especially

367

with high doses of duloxetine. Yet, the possible modulation of dopamine by duloxetine is

368

controversial, since similar doses of duloxetine showed contradictory effects (increase of

369

dopamine levels with an unique dose of 10 mg/kg of duloxetine and no significant changes in

370

dopamine levels with four doses of 20 mg/kg) (Muneoka et al., 2009; Kale and Addepalli, 2014).

371

Additional studies are necessary to clarify the role of dopamine in the possible proconvulsant

372

effect of high doses of duloxetine.

AC C

EP

TE D

364

373

Other hypothesis already highlighted to explain the anticonvulsant effect of

374

antidepressants may be the inhibition of ion channels. Sodium and calcium channels, for

ACCEPTED MANUSCRIPT

example, control neurotransmitter release (as glutamate, the main excitatory neurotransmitter of

376

SNC). However, the contribution of this action to the therapeutic effect is less clear. ISSRs (as

377

fluoxetine, paroxetine or citalopram) and TCAs (as desipramine or amitriptyline) already

378

demonstrated to be able of inhibiting sodium channels in cell cultures or brain slices (Lenkey et

379

al., 2006; Yan et al., 2010; Igelström and Heyward, 2012; Thériault et al., 2015). However, other

380

antidepressants as the inhibitor of monoaminooxidase, moclobemide, do not show this property

381

(Thériault et al., 2015). An interesting study with 4-aminopyridine (a blocker of potassium

382

channels that increases the permeability of sodium and calcium channels) showed that sertraline

383

is able to prevent EEG changes induced by this blocker (Sitges et al., 2012). According to the

384

authors, this potent effect may be due to the decreased permeability of brain presynaptic sodium

385

and calcium channels.

M AN U

SC

RI PT

375

Although the studies with duloxetine are extremely scarce, the modulation of ion

387

channels by duloxetine has been reported: duloxetine is capable of inhibiting sodium channels in

388

vitro (Wang et al., 2010), and the same effect was recently described for the calcium channels of

389

neuronal cells (Akpinar et al., 2014). Both sodium and calcium channels have been identified as

390

major targets of traditional antiepileptic drugs, such as phenytoin and ethosuximide, which are

391

used in clinical practice for tonic-clonic seizures and absence epilepsy, respectively.

EP

However, the therapeutic actions of the antidepressant drugs may not completely explain

AC C

392

TE D

386

393

the direct interaction with these pharmacological targets. One of the few molecular events

394

already associated with the possible anticonvulsant effect of antidepressants drugs is the

395

influence on oxidative stress (Akpinar et al., 2014; Payandemehr et al., 2012; Martinc et al.,

396

2012). In recent years, the role of oxidative stress in the development of epilepsy and the

397

refractoriness to treatment has been highlighted (Cardenas-Rodriguez et al., 2013; Aguiar et al.,

ACCEPTED MANUSCRIPT

398

2013; Rowley and Patel, 2013; Martinc et al., 2012; Rowles and Olsen, 2012; Shin et al., 2011),

399

making antioxidant compounds a realistic proposal in the pursuit of new therapeutic tools against

400

seizures (Souza-Monteiro et al., 2015; Branco et al., 2013; Rowles and Olsen, 2012). In our study, PTZ significantly increased the level of lipid peroxidation in the cerebral

402

cortex, confirming previous studies showing similar effects (Souza-Monteiro et al., 2015; Branco

403

et al., 2013; Chowdury et al. 2013). PTZ affects many areas in CNS to induce convulsive

404

seizures since the GABAA receptors (molecular target for PTZ) are widespread. Still, because

405

cortex is one of the main areas involved in convulsive seizures and epileptogenesis, recent works

406

with this model use the whole cortex for the screening of anticonvulsant drugs (Ramos et al.,

407

2012; Nazıroğlu et al., 2013; Coelho et al., 2015; Coitinho et al., 2015; Souza-Monteiro et al.,

408

2015). After the initial screening to prove if one drug has anticonvulsant effects, additional

409

studies could be carried out to clarify the effects on specific areas of the cortex.

M AN U

SC

RI PT

401

In our work, a single dose of 20 mg/Kg duloxetine was able to protect the cerebral cortex

411

of mice, completely avoiding the PTZ-induced increase in lipid peroxidation. However, 40

412

mg/Kg duloxetine did not alter the levels of PTZ-induced lipid peroxidation, a different behavior

413

of that of 20 mg/Kg of duloxetine (P<0.05). Future studies will clarify if higher doses of

414

duloxetine cause a synergistic effect with PTZ.

EP

Although two possibilities may account for the effect of 20 mg/Kg duloxetine (an

AC C

415

TE D

410

416

antioxidant protection exerted by this drug or a consequence of the decrease in total duration and

417

severity of seizures), the latter explanation seems to be less probable if we look to the results

418

with 40 mg/Kg of duloxetine plus PTZ. The latter treatment caused a significant longer time in

419

seizures (Fig. 2) and a significant higher electrical activity (Fig. 4), but the same levels of lipid

ACCEPTED MANUSCRIPT

420

peroxidation than the treatment with PTZ alone. In other words, higher duration and severity of

421

seizures was not associated to higher levels of lipid peroxidation. Despite several in vitro studies with animal models and epidemiological data describing

423

the influence of different antidepressants in oxidative stress (Lee et al., 2013; Behr et al., 2012),

424

duloxetine is almost unknown in this sense. Only recently was the influence of duloxetine on

425

lipid peroxidation described in vitro for the first time (Akpinar et al., 2014); incubation of the

426

PC-12 cell line with 10 µM duloxetine for 24 h reduced basal levels of lipid peroxidation. Our

427

work demonstrates for the first time that a treatment with duloxetine in vivo is able to avoid PTZ-

428

induced lipid peroxidation in the cerebral cortex.

M AN U

SC

RI PT

422

In addition to lipid peroxidation, several studies have described the role of the nitrergic

430

system in the actions of antidepressants, and new therapeutic targets in this system have been

431

proposed for the development of future antidepressant drugs (Lee et al., 2013). Interestingly, the

432

nitrergic system has already been implicated in the biphasic effect of citalopram against PTZ-

433

induced seizures; the precursor L-arginine exacerbates the proconvulsant effect of high doses of

434

the drug and inhibitors of nitric oxide synthase have a synergistic effect on the anticonvulsant

435

action of low doses of citalopram (Payandemehr et al., 2012). Furthermore, low doses of

436

venlafaxine or duloxetine seem to diminish basal nitrite levels, an indirect marker of nitric oxide

437

production (Zomkowski et al., 2012; Krass et al., 2011). We used a higher dose of duloxetine

438

that had no influence on nitrite levels. These results are in agreement with studies showing that

439

therapeutic doses of venlafaxine do not affect basal nitrite levels (Abdel-Wahab and Salama,

440

2011; Dhir and Kulkarni, 2008) and that the drug only prevents the increase in nitrites in

441

response to deleterious stimuli, such as ischemia, depression, and sleep deprivation, among

AC C

EP

TE D

429

ACCEPTED MANUSCRIPT

442

others (Abdel-Wahab and Salama, 2011; Dhir and Kulkarni, 2008; Gaur and Kumar, 2010;

443

Kumar et al., 2010). In our study, PTZ did not affect nitrite levels compared to basal levels. Comparison with

445

basal levels is relatively uncommon in studies analyzing nitrite levels or other markers of

446

oxidative stress with the PTZ model, reporting only comparisons between the groups treated with

447

PTZ and co-treated with PTZ and other drugs (Aguiar et al., 2013). However, the few studies

448

that included a basal group for comparison used higher doses of PTZ (80-100 mg/Kg) (Lopes et

449

al., 2013; Stojanović et al., 2010; Bikjdaouene et al., 2003), possibly explaining the absence of

450

nitrergic induction in our work. This hypothesis is supported by Guzman et al. (2005), who

451

demonstrated that acute inoculation of 40 mg/Kg PTZ did not alter nitrite levels in the brains of

452

animals.

M AN U

SC

RI PT

444

If the nitrergic system does not play a major role in the oxidative stress found in our

454

model with PTZ and duloxetine, what other factors may be involved? One of the main defense

455

systems against oxidative stress is the antioxidant enzymes responsible for the detoxification and

456

processing of free radicals. SOD is responsible for dismutation of the radical superoxide and

457

CAT catalyzes hydrogen peroxide into water and oxygen, and both are essential to maintaining

458

the balance between pro-oxidant events and antioxidant processes inside the cell. TCAs (e.g.,

459

imipramine) and SSRIs (e.g., fluoxetine) are capable of influencing the enzymatic activities of

460

SOD and CAT in both animal models and patients (Reus et al., 2010; Khanzode et al., 2003).

461

This event could be an alternative to explain the antioxidant protection of duloxetine against

462

seizures.

AC C

EP

TE D

453

463

PTZ significantly reduced the enzymatic activities of SOD and CAT. This action of PTZ

464

on enzymatic activities is well known, and recent studies using the same dose of PTZ (60

ACCEPTED MANUSCRIPT

mg/Kg) reported decreased activities similar to those found here (approximately 50% of the

466

saline group) (Branco et al., 2013; Chowdhury et al., 2013; Rambo et al., 2009). To the best of

467

our knowledge, no data are available on the possible influence of duloxetine on SOD and CAT

468

activities. However, looking at the literature, we hypothesized that duloxetine exerts some

469

influence on these enzymes because the SNRI venlafaxine prevents a decrease of in the enzyme

470

activities in models of ischemia and anxiety-like behavior (Gaur and Kumar, 2010; Kumar et al.,

471

2010).

SC

RI PT

465

Thus, the present study demonstrates for the first time that a dose of duloxetine (20

473

mg/Kg) is associated to the preservation of the enzymatic activities of SOD and CAT in a model

474

of acute seizures induced by PTZ. As described for lipid peroxidation, this effect of duloxetine

475

may be cause and/or consequence of the decrease in seizures and future studies will clarify the

476

exact role of this drug. Still, if the direct influence of duloxetine in SOD and CAT activities is

477

confirmed, the modulation of SOD and CAT would play a major role in the antioxidant

478

protection exerted by the drug as demonstrated by the levels of lipid peroxidation.

TE D

M AN U

472

The latter conclusion is supported by the results obtained with 40 mg/Kg of duloxetine, a

480

dose that is ineffective at preventing the decrease in SOD activity induced by PTZ. Moreover,

481

the total duration and electrical activity in seizures caused by 40 mg/Kg duloxetine plus PTZ

482

(higher than those provoked by PTZ alone) (Fig. 2 and 4) were not correlated with lower

483

activities of the enzymes (CAT was similar to control group and SOD was similar to PTZ

484

group). These facts may reduce the probability of modulation of oxidative stress as a mere

485

consequence of the severity of the seizures. Additional studies with higher doses of duloxetine

486

are being carried out to test a possible biphasic effect of the drug on oxidative stress in the

487

convulsion model.

AC C

EP

479

ACCEPTED MANUSCRIPT

Independent of the possible biphasic effect of duloxetine, recent reports have revealed that,

489

in patients, most cases of death by ingestion of duloxetine is due to the association with other

490

drugs and co-morbidities rather than duloxetine alone (Pilgrim et al., 2014), and duloxetine has

491

fewer cases of reported mortality by overdose than other antidepressants (Taylor et al., 2013).

492

Thus, our data suggest that the relatively safe use of duloxetine for the treatment of depression in

493

epileptic patients would have the additional value of being an anticonvulsant. Although further

494

studies are needed, this extra anticonvulsant effect may allow a reduction of the doses of

495

anticonvulsants, causing fewer side effects and possibly decreasing morbidity and mortality due

496

to drug interactions in polytherapy.

AC C

EP

M AN U

TE D

497

SC

RI PT

488

ACCEPTED MANUSCRIPT

498

ACKNOWLEDGMENTS Our more sincere acknowledgments to the Associate Editor and the Reviewers for really

500

help us to improve this manuscript. We thank Instituto Evandro Chagas (IEC, Brazil) for kindly

501

providing the animals for this study. This work was supported by Conselho Nacional de Ciência

502

e Tecnologia em Pesquisa (CNPq, Brazil; grants numbers 467143/2014-5 and 447568/2014-0)

503

and Pró-Reitoria de Pesquisa da UFPA (PROPESP-UFPA, Brazil). L.F.F. Royes and M.E.

504

Crespo-López thank CNPq for their researcher fellowships. Also, D. Santana-Coelho, J.R.

505

Souza-Monteiro and G.P.F. Arrifano thank Coordenação de Aperfeiçoamento de Pessoal de

506

Nivel Superior (CAPES, Brazil), for their PhD fellowships.

M AN U

SC

RI PT

499

507

AC C

EP

TE D

508

ACCEPTED MANUSCRIPT

REFERENCES

510

Abdel-Wahab, B.A., Salama, R.H., 2011. Venlafaxine protects against stress-induced oxidative

511

DNA damage in hippocampus during antidepressant testing in mice. Pharmacology,

512

Biochemistry, and Behavior 100, 59-65

RI PT

509

Aebi, H., 1984. Catalase in vitro. Methods in enzymology 105, 121-126.

514

Aguiar, C.C., Almeida, A.B., Araujo, P.V., Vasconcelos, G.S., Chaves, E.M., do Vale, O.C.,

515

Macedo, D.S., Leal, L.K., de Barros Viana, G.S., Vasconcelos, S.M., 2013. Effects of

516

agomelatine on oxidative stress in the brain of mice after chemically induced seizures.

517

Cellular and Molecular Neurobiology 33, 825-835.

M AN U

SC

513

518

Ahern, T.H., Javors, M.A., Eagles, D.A., Martillotti, J., Mitchell, H.A., Liles, L.C., Weinshenker,

519

D., 2006. The effects of chronic norepinephrine transporter inactivation on seizure

520

susceptibility in mice. Neuropsychopharmacology 31, 730-738. Akpinar, A., Uguz, A.C., Naziroglu, M., 2014. Agomelatine and duloxetine synergistically

522

modulates apoptotic pathway by inhibiting oxidative stress triggered intracellular calcium

523

entry in neuronal PC12 cells: role of TRPM2 and voltage-gated calcium channels. Journal

524

of Membrane Biology 247, 451-459.

EP

TE D

521

Behr, G.A., Moreira, J.C., Frey, B.N., 2012. Preclinical and clinical evidence of antioxidant

526

effects of antidepressant agents: implications for the pathophysiology of major depressive

527

AC C

525

disorder. Oxidative Medicine and Cellular Longevity 2012, 609421.

528

Berg, A.T., Levy, S.R., Testa, F.M., Blumenfeld, H., 2014. Long-term seizure remission in

529

childhood absence epilepsy: might initial treatment matter? Epilepsia 55, 551-557.

ACCEPTED MANUSCRIPT

530

Bikjdaouene, L., Escames, G., León, J., Ferrer, J.M., Khaldy, H., Vives, F., et al., 2003. Changes

531

in brain amino acids and nitric oxide after melatonin administration in rats with

532

pentylenetetrazole-induced seizures. Journal of Pineal Research 35(1), 54-60. Borowicz, K.K., Golyska, D., Luszczki, J.J., Czuczwar, S.J., 2011. Effect of acutely and

534

chronically administered venlafaxine on the anticonvulsant action of classical

535

antiepileptic drugs in the mouse maximal electroshock model. European Journal of

536

Pharmacology 670, 114-120.

SC

RI PT

533

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities

538

of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248-

539

254.

M AN U

537

Branco, C.S., Scola, G., Rodrigues, A.D., Cesio, V., Laprovitera, M., Heinzen, H., Dos Santos,

541

M.T., Fank, B., de Freitas, S.C., Coitinho, A.S., Salvador, M., 2013. Anticonvulsant,

542

neuroprotective and behavioral effects of organic and conventional yerba mate (Ilex

543

paraguariensis St. Hil.) on pentylenetetrazol-induced seizures in Wistar rats. Brain

544

Research Bulletin 92, 60-68.

TE D

540

Bymaster, F.P., Dreshfield-Ahmad, L.J., Threlkeld, P.G., Shaw, J.L., Thompson, L., Nelson,

546

D.L., Hemrick-Luecke, S.K., Wong, D.T., 2001. Comparative affinity of duloxetine and

547

venlafaxine for serotonin and norepinephrine transporters in vitro and in vivo, human

549

AC C

548

EP

545

serotonin receptor subtypes, and other neuronal receptors. Neuropsychopharmacology 25(6), 871-80.

550

Cardamone, L., Salzberg, M.R., O'Brien, T.J., Jones, N.C., 2013. Antidepressant therapy in

551

epilepsy: can treating the comorbidities affect the underlying disorder? British Journal of

552

Pharmacology 168, 1531-1554.

ACCEPTED MANUSCRIPT

Cardenas-Rodriguez, N., Huerta-Gertrudis, B., Rivera-Espinosa, L., Montesinos-Correa, H.,

554

Bandala, C., Carmona-Aparicio, L., Coballase-Urrutia, E., 2013. Role of oxidative stress

555

in refractory epilepsy: evidence in patients and experimental models. International

556

Journal of Molecular Sciences 14, 1455-1476.

RI PT

553

Chowdhury, B., Bhattamisra, S.K., Das, M.C., 2013. Anti-convulsant action and amelioration of

558

oxidative stress by Glycyrrhiza glabra root extract in pentylenetetrazole- induced seizure

559

in albino rats. Indian Journal of Pharmacology 45, 40-43.

SC

557

Citraro, R., Leo, A., Fazio, P.d., Sarro, G.D., Russo, E., 2015. Antidepressants but not

561

antipsychotics have antiepileptogenic effects with limited effects on comorbid

562

depressive-like behavior in the WAG/Rij rat model of absence epilepsy. British Journal

563

of Pharmacology 172(12), 3177-88.

M AN U

560

Coelho, D.S., Monteiro, J.R.S., Crespo-López, M.E., 2014. Possible anticonvulsant action of

565

duloxetine: Preliminary results in rats. Epilepsy & Behavior 38, 187 (Abstract).

566

http://dx.doi.org/10.1016/j.yebeh.2014.08.047

TE D

564

Coelho, V.R., Vieira, C.G., de Souza, L.P., Moysés, F., Basso, C., Papke, D.K., Pires,

568

T.R., Siqueira, I.R., Picada, J.N., Pereira, P., 2015. Antiepileptogenic, antioxidant and

569

genotoxic evaluation of rosmarinic acid and its metabolite caffeic acid in mice. Life

570

Sciences 122, 65-71.

572 573

AC C

571

EP

567

Coitinho, A.S., da Costa, B.M., Dos Santos, M.T., Fank, B., Hackmann, C.L., Salvador, M., Dani,

C.,

2015.

Vitis

labrusca

leaf

extract

prevents

pentylenetetrazol-

induced oxidative damage but not seizures in rats. Cell Mol Biol 61(2), 39-42.

ACCEPTED MANUSCRIPT

574

Dhir, A., Kulkarni, S.K., 2008. Venlafaxine reverses chronic fatigue-induced behavioral,

575

biochemical and neurochemical alterations in mice. Pharmacology, Biochemistry, and

576

Behavior 89, 563-571.

578

Esterbauer, H., Cheeseman, K.H., 1990. Determination of aldehydic lipid peroxidation products:

RI PT

577

malonaldehyde and 4-hydroxynonenal. Methods in Enzymology 186, 407-421.

Ferraro, T.N., Golden, G.T., Smith, G.G., St Jean, P., Schork, N.J., Mulholland, N., Ballas, C.,

580

Schill, J., Buono, R.J., Berrettini, W.H., 1999. Mapping loci for pentylenetetrazol-

581

induced seizure susceptibility in mice. Journal of Neuroscience 19, 6733-6739.

SC

579

Frantseva, M.V., Perez Velazquez, J.L., Tsoraklidis, G., Mendonca, A.J., Adamchik, Y., Mills,

583

L.R., Carlen, P.L., Burnham, M.W., 2000. Oxidative stress is involved in seizure-induced

584

neurodegeneration in the kindling model of epilepsy. Neuroscience 97, 431-435.

M AN U

582

Gaur, V., Kumar, A., 2010. Protective effect of desipramine, venlafaxine and trazodone against

586

experimental animal model of transient global ischemia: possible involvement of NO-

587

cGMP pathway. Brain Research 1353, 204-212.

TE D

585

Green, L.C., Ruiz de Luzuriaga, K., Wagner, D.A., Rand, W., Istfan, N., Young, V.R.,

589

Tannenbaum, S.R., 1981. Nitrate biosynthesis in man. Proceedings of the National

590

Academy of Sciences of the United States of America 78, 7764-7768.

AC C

EP

588

591

Guzman, D.C., Vazquez, I.E., Mejia, G.B., Garcia, E.H., del Angel, D.S., Olguin, H.J., 2005.

592

Effect of pentylenetetrazole and carbodiimide on oxidation stress markers in rat brain.

593 594 595

Basic & Clinical Pharmacology & Toxicology 96, 512-513.

Igelström, K.M., Heyward, P.M., 2012. The antidepressant drug fluoxetine inhibits persistent sodium currents and seizure-like events. Epilepsy Research 101(1-2), 174-81.

ACCEPTED MANUSCRIPT

597 598 599

Jobe, P.C., Browning, R.A., 2005. The serotonergic and noradrenergic effects of antidepressant drugs are anticonvulsant, not proconvulsant. Epilepsy Behaviour 7(4), 602-19. Kale, P.P., Addepalli, V., 2014. Augmentation of antidepressant effects of duloxetine and bupropion by caffeine in mice. Pharmacol Biochem Behav 124, 238-44.

RI PT

596

Kanner, A.M., Schachter, S.C., Barry, J.J., Hesdorffer, D.C., Mula, M., Trimble, M., Hermann,

601

B., Ettinger, A.E., Dunn, D., Caplan, R., Ryvlin, P., Gilliam, F., LaFrance, W.C., Jr.,

602

2012. Depression and epilepsy: epidemiologic and neurobiologic perspectives that may

603

explain their high comorbid occurrence. Epilepsy & behavior : E&B 24, 156-168.

604

Kanner, A.M., 2012. Can neurobiological pathogenic mechanisms of depression facilitate the

M AN U

605

SC

600

development of seizure disorders? Lancet Neurol. 11(12), 1093-1102. Khanzode, S.D., Dakhale, G.N., Khanzode, S.S., Saoji, A., Palasodkar, R.., 2003. Oxidative

607

damage and major depression: the potential antioxidant action of selective serotonin re-

608

uptake inhibitors. Redox Rep 8(6), 365-70.

609 610

TE D

606

Kilic, F.S.; Dogan, A.E.; Baydemir, C.; Erol, K., 2011. The acute effects of mirtazapine on pain related behavior in healthy animals. Neurosciences 16(3), 217-23. Krass, M., Wegener, G., Vasar, E., Volke, V., 2011. The antidepressant action of imipramine and

612

venlafaxine involves suppression of nitric oxide synthesis. Behavioural Brain Research

AC C

613

EP

611

218, 57-63.

614

Kumar, A., Garg, R., Gaur, V., Kumar, P., 2010. Venlafaxine involves nitric oxide modulatory

615

mechanism in experimental model of chronic behavior despair in mice. Brain Research

616

1311, 73-80.

617

Kumar, U., Medel-Matus, J.S., Redwine, H.M., Shin, D., Hensler, J.G., Sankar, R., Mazarati, A.,

618

2016. Effects of selective serotonin and norepinephrine reuptake inhibitors on depressive-

ACCEPTED MANUSCRIPT

619

and impulsive-like behaviors and on monoamine transmission in experimental temporal

620

lobe epilepsy. Epilepsia 57(3), 506-515. Lee, S.Y., Lee, S.J., Han, C., Patkar, A.A., Masand, P.S., Pae, C.U., 2013. Oxidative/nitrosative

622

stress and antidepressants: targets for novel antidepressants. Progress in Neuro-

623

psychopharmacology & Biological Psychiatry 46, 224-235.

RI PT

621

Lenkey, N., Karoly, R., Kiss, J.P., Szasz, B.K., Vizi, E.S., Mike, A., 2006. The mechanism of

625

activity-dependent sodium channel inhibition by the antidepressants fluoxetine and

626

desipramine. Mol Pharmacol 70(6), 2052-63.

SC

624

Li, B., Wang, L., Sun, Z., Zhou, Y., Shao, D., Zhao, J., Song, Y., Lv, J., Dong, X., Liu, C.,

628

Wang, P., Zhang, X., Cui, R., 2014. The anticonvulsant effects of SR 57227 on

629

pentylenetetrazole-induced seizure in mice. PloS One 9, e93158.

M AN U

627

Lopes, K.S., Rios, E.R., Lima, C.N., Linhares, M.I., Torres, A.F., Havt, A., et al., 2013. The

631

effects of the Brazilian ant Dinoponera quadriceps venom on chemically induced seizure

632

models. Neurochemistry International 63(3), 141-145.

TE D

630

Martinc, B., Grabnar, I., Vovk, T., 2012. The role of reactive species in epileptogenesis and

634

influence of antiepileptic drug therapy on oxidative stress. Curr Neuropharmacol 10(4), 328-

635

43.

EP

633

McColl, C.D., Horne, M.K., Finkelstein, D.I., Wong, J.Y., Berkovic, S.F., Drago, J., 2003.

637

Electroencephalographic characterisation of pentylenetetrazole-induced seizures in mice

638

lacking the alpha 4 subunit of the neuronal nicotinic receptor. Neuropharmacology 44(2),

639

234-243.

AC C

636

ACCEPTED MANUSCRIPT

Mehrzadi, S., Shojaii, A., Pur, S.A., Motevalian, M., 2015. Anticonvulsant Activity of

641

Hydroalcoholic Extract of Citrullus colocynthis Fruit: Involvement of Benzodiazepine and

642

Opioid Receptors. J Evid Based Complementary Altern Med. pii: 2156587215615455.

643

Misra, H.P., Fridovich, I., 1972. The generation of superoxide radical during the autoxidation of

644

RI PT

640

hemoglobin. Journal of Biological Chemistry 247, 6960-6962.

Muneoka, K., Shirayama, Y., Takigawa, M., Shioda, S., 2009. Brain region-specific effects of

646

short-term treatment with duloxetine, venlafaxine, milnacipran and sertraline on

647

monoamine metabolism in rats. Neurochemical Research 34(3), 542-55.

SC

645

Nazıroğlu, M., Akay, M.B., Çelik, Ö., Yıldırım, M.İ., Balcı, E., Yürekli, V.A., 2013. Capparis

649

ovata modulates brain oxidative toxicity and epileptic seizures in pentylentetrazol-

650

induced epileptic rats. Neurochemistry Research 38(4), 780-788.

M AN U

648

Oliveira, C.C., Oliveira, C.V., Grigoletto, J., Ribeiro, L.R., Funck, V.R., Grauncke, A.C., Souza,

652

T.L., Souto, N.S., Furian, A.F., Menezes, I.R., Oliveira, M.S., 2016. Anticonvulsant activity

653

of β-caryophyllene against pentylenetetrazol-induced seizures. Epilepsy Behav 56, 26-31.

654

Payandemehr, B., Bahremand, A., Rahimian, R., Ziai, P., Amouzegar, A., Sharifzadeh, M.,

655

Dehpour, A.R., 2012. 5-HT(3) receptor mediates the dose-dependent effects of citalopram

656

on pentylenetetrazole-induced clonic seizure in mice: involvement of nitric oxide. Epilepsy

657

Research 101, 217-227.

659 660

EP

AC C

658

TE D

651

Popławska,

M., Wróblewska,

D., Borowicz,

K.K.,

2015.

Interactions

between

an

antidepressant reboxetine and four classic antiepileptic drugs in the mouse model of myoclonic seizures. Pharmacol Rep. 67(6), 1141-1146.

ACCEPTED MANUSCRIPT

661

Pellicciari, A., Balzarro, B., Scaramelli, A., Porcelli, S., Serretti, A., De Ronchi, D., 2012.

662

Generalized tonic-clonic seizure secondary to duloxetine poisoning: a short report with

663

favorable out come. Neurotoxicology 33, 189-190. Pilgrim, J.L., Gerostamoulos, D., Drummer, O.H., 2014. The prevalence of duloxetine in

665

medico-legal death investigations in Victoria, Australia (2009-2012). Forensic Science

666

International 234, 165-173.

RI PT

664

Pintor, L., Bailles, E., Matrai, S., Carreno, M., Donaire, A., Boget, T., Setoain, X., Rumia, J.,

668

Bargallo, N., 2010. Efficiency of venlafaxine in patients with psychogenic nonepileptic

669

seizures and anxiety and/or depressive disorders. Journal of Neuropsychiatry and Clinical

670

Neurosciences 22, 401-408

672

M AN U

671

SC

667

Puttachary, S., Sharma, S., Stark, S., Thippeswamy, T., 2015. Seizure-induced oxidative stress in temporal lobe epilepsy. BioMed Research International, 745613. Rambo, L.M., Ribeiro, L.R., Oliveira, M.S., Furian, A.F., Lima, F.D., Souza, M.A., Silva, L.F.,

674

Retamoso, L.T., Corte, C.L., Puntel, G.O., de Avila, D.S., Soares, F.A., Fighera, M.R.,

675

Mello,

676

of creatine supplementation and physical exercise against pentylenetetrazol-induced

677

seizures. Neurochemistry International 55(5), 333-40.

Royes,

L.F.,

2009.

Additive

anticonvulsant

effects

EP

C.F.,

TE D

673

Ramos, S.F., Mendonça, B.P., Leffa, D.D., Pacheco, R., Damiani, A.P., Hainzenreder, G.,

679

Petronilho, F., Dal-Pizzol, F., Guerrini, R., Calo, G., Gavioli, E.C., Boeck, C.R., de

680 681 682 683

AC C

678

Andrade, V.M., 2012. Effects of neuropeptide S on seizures and oxidative damage induced by pentylenetetrazole in mice. Pharmacol Biochem Behav 103(2), 197-203.

Reagan-Shaw, S., Nihal, M., Ahmad, N., 2008. Dose translation from animal to human studies revisited. FASEB Journal 22, 659-661.

ACCEPTED MANUSCRIPT

Reus, G.Z., Stringari, R.B., de Souza, B., Petronilho, F., Dal-Pizzol, F., Hallak, J.E., Zuardi,

685

A.W., Crippa, J.A., Quevedo, J., 2010. Harmine and imipramine promote antioxidant

686

activities in prefrontal cortex and hippocampus. Oxidative Medicine and Cellular

687

Longevity 3, 325-331.

RI PT

684

Rodrigues, A.D., Scheffel, T.B., Scola, G., Santos, M.T., Fank, B., de Freitas, S.C., Dani, C.,

689

Vanderlinde, R., Henriques, J.A., Coitinho, A.S., Salvador M., 2012. Neuroprotective and

690

anticonvulsant effects of organic and conventional purple grape juices on seizures in

691

Wistar rats induced by pentylenetetrazole. Neurochem Int 60(8), 799-805.

693 694 695

Rowles, J., Olsen, M., 2012. Perspectives on the development of antioxidant antiepileptogenic

M AN U

692

SC

688

agents. Mini Rev Med Chem 12(10), 1015-27.

Rowley, S., Patel, M., 2013. Mitochondrial involvement and oxidative stress in temporal lobe epilepsy. Free Radic Biol Med 62, 121-131.

Santos, J.G.Jr., do Monte, F.H., Russi, M., Agustine, P.E., Lanziotti, V.M., 2002. Proconvulsant

697

effects of high doses of venlafaxine in pentylenetetrazole-convulsive rats. Braz J Med Biol

698

Res 35(4), 469-72.

TE D

696

Shin, E.J., Jeong, J.H., Chung, Y.H., Kim, W.K., Ko, K.H., Bach, J.H., Hong, J.S., Yoneda, Y.,

700

Kim, H.C., 2011. Role of oxidative stress in epileptic seizures. Neurochemistry

701

International 59, 122-137.

AC C

EP

699

702

Sitges, M., Aldana, B., Reed, R.C., 2016. Effect of the Anti-depressant Sertraline, the Novel

703

Anti-seizure Drug Vinpocetine and Several Conventional Antiepileptic Drugs on the

704 705

Epileptiform EEG Activity Induced by 4-Aminopyridine. Neurochemical Research 41(6), 1365-74.

ACCEPTED MANUSCRIPT

Souza, M.A., Mota, B.C., Gerbatin, R.R., Rodrigues, F.S., Castro, M., Fighera, M.R., Royes,

707

L.F., 2013. Antioxidant activity elicited by low dose of caffeine attenuates

708

pentylenetetrazol-induced seizures and oxidative damage in rats. Neurochemistry

709

International 62, 821-830.

RI PT

706

Souza-Monteiro, J.R., Hamoy, M., Santana-Coelho, D., Arrifano, G.P.F., Paraense, R.S.O.,

711

Costa-Malaquias, A., Mendonça, J.R., da Silva, R.F., Monteiro, W.S.C., Rogez, H.,

712

Oliveira, D.L., do Nascimento, J.L.M., Crespo-López, M.E., 2015. Towards new

713

therapeutic tools for seizures: pre-clinical assay with clarified açai (Euterpe oleracea)

714

juice. Neurochemistry International 90, 20-7.

M AN U

SC

710

Staley, K., 2015. Molecular mechanisms of epilepsy. Nature Neuroscience 18, 367-372.

716

Stojanović, I., Jelenković, A., Stevanović, I., Pavlović, D., Bjelaković, G., Jevtović-Stoimenov,

717

T., 2010. Spermidine influence on the nitric oxide synthase and arginase activity

718

relationship during experimentally induced seizures. J Basic Clin Physiol Pharmacol.

719

21(2), 169-185.

TE D

715

Taylor, D., Lenox-Smith, A., Bradley, A., 2013. A review of the suitability of duloxetine and

721

venlafaxine for use in patients with depression in primary care with a focus on

722

cardiovascular safety, suicide and mortality due to antidepressant overdose. Therapeutic

723

Advances in Psychopharmacology 3, 151-161.

AC C

EP

720

724

Thériault, O., Poulin, H., Beaulieu, J.M., Chahine, M., 2015. Differential modulation of Nav1.7

725

and Nav1.8 channels by antidepressant drugs. European Journal of Pharmacology 764,

726

395-403

ACCEPTED MANUSCRIPT

727

Vermoesen, K., Massie, A., Smolders, I., Clinckers, R., 2012. The antidepressants citalopram

728

and reboxetine reduce seizure frequency in rats with chronica epilepsy. Epilepsia 53(5),

729

870-8. Vermoesen, K., Serruys, A.S., Loyens, E., Afrikanova, T., Massie, A., Schallier, A., Michotte,

731

Y., Crawford, A.D., Esguerra, C.V., de Witte, P.A., Smolders, I., Clinckers, R., 2011.

732

Assessment of the convulsant liability of antidepressants using zebrafish and

733

mouse seizure models. Epilepsy and Behaviour 22(3), 450-60.

SC

RI PT

730

Volonteri, L.S., Colasanti, A., Cerveri, G., Fiorentini, A., De Gaspari, I.F., Mauri, M.C., Valli,

735

A., Papa, P., Mencacci, C., 2010. Clinical outcome and tolerability of duloxetine in the

736

treatment of major depressive disorder: a 12-week study with plasma levels. Journal of

737

Psychopharmacology 24(8), 1193-9.

Wang, S.Y., Calderon, J., Kuo Wang, G., 2010. Block of neuronal Na+ channels by antidepressant duloxetine in a state-dependent manner. Anesthesiology 113, 655-665.

739 740 741

WHO.

Epilepsy.

TE D

738

M AN U

734

World

Health

Organization.

2012.

Available

at:

www.who.int/mediacentre/factsheets/fs999/en/ Wilner, A.N., Sharma, B.K., Thompson, A., Soucy, A., Krueger, A., 2014. Diagnoses,

743

procedures, drug utilization, comorbidities, and cost of health care for people with

AC C

744

EP

742

epilepsy in 2012. Epilepsy & behavior 41, 83-90.

745

Yan, L., Qiang, W., Qi, F., Qing, Y., Hang, X., Qi, W., 2010. Amitriptyline inhibits currents and

746

decreases the mRNA expression of voltage-gated sodium channels in cultured rat cortical

747

neurons. Brain Research 1336, 1-9.

748

Yuen, E.S., Troconiz, I.F., 2015. Can pentylenetetrazole and maximal electroshock rodent

749

seizure models quantitatively predict antiepileptic efficacy in humans? Seizure 24, 21-27.

ACCEPTED MANUSCRIPT

Zomkowski, A.D., Engel, D., Cunha, M.P., Gabilan, N.H., Rodrigues, A.L., 2012. The role of

751

the NMDA receptors and l-arginine-nitric oxide-cyclic guanosine monophosphate

752

pathway in the antidepressant-like effect of duloxetine in the forced swimming test.

753

Pharmacology, Biochemistry, and Behavior 103, 408-417.

AC C

EP

TE D

M AN U

SC

754

RI PT

750

ACCEPTED MANUSCRIPT

FIGURE LEGENDS

756

Figure 1. Latencies to the first myoclonic jerk (A) and to the first tonic-clonic seizure (B).

757

Animals were intraperitoneally pre-treated with saline or 10, 20, or 40 mg/Kg of duloxetine

758

(DUL10, DUL20, and DUL40, respectively) and then, 30 min later, pentylenetetrazol (PTZ, 60

759

mg/Kg i.p.) was administrated (n=10). Data showed non-gaussian distribution and they are

760

presented as median ± interquartile range, n=10. Kruskal-Wallis followed by Dunn: *P<0.05,

761

**P<0.01 and ***P<0.001 vs. PTZ; ##P<0.01 vs. DUL20+PTZ.

762

Figure 2. Total time spent in seizures (sum of the durations of all seizures during the observation

763

time). Animals were intraperitoneally pre-treated with saline or 10, 20, or 40 mg/Kg of

764

duloxetine (DUL10, DUL20, and DUL40, respectively) and then, 30 min later, pentylenetetrazol

765

(PTZ, 60 mg/Kg i.p.) was administrated (n=10). Data showed Gaussian distribution but

766

significant differences between SDs, so they are showed as median ± interquartile range (left).

767

Kruskal-Wallis test indicated P=0.08. An additional comparison between PTZ and DUL40+PTZ

768

groups was included (right). Because their Gaussian distributions, these two groups at the right

769

are showed as mean ± SEM and Student´s t test with Welch correction for different SDs was

770

applied: *P<0.05 vs. PTZ.

771

Figure 3. Representation of electroencephalographic events, demonstrating the discharge

772

sequence observed in animals treated intraperitoneally with saline (SAL) or 10, 20, or 40 mg/Kg

773

of duloxetine (DUL10, DUL20, and DUL40, respectively) and then, 30 min later, with

774

pentylenetetrazol (PTZ, 60 mg/Kg i.p.) or saline was administrated. Black arrows indicate the

775

first myoclonic seizure and the amplification of the beginning and duration of the generalized

776

seizure are highlighted. The calibration bars denote 50 µV and 30 s.

AC C

EP

TE D

M AN U

SC

RI PT

755

ACCEPTED MANUSCRIPT

Figure 4. Amplitudes of discharges registered in the electroencephalograms. Animals were

778

intraperitoneally treated with saline (SAL) or 10, 20, or 40 mg/Kg of duloxetine (DUL10,

779

DUL20, and DUL40, respectively) and then, 30 min later, pentylenetetrazol (PTZ, 60 mg/Kg

780

i.p.) was administered. Basal records before any treatment (left), levels after saline or duloxetine

781

administration (center), and levels after PTZ injection (right) are shown. Data showed Gaussian

782

distribution with similar SDs and they are presented as mean percentages of basal levels ± SEM

783

(n=10). Two-way ANOVA followed by Bonferroni: ***P<0.001 vs. all groups treated with PTZ;

784

#

785

Figure 5. Lipid peroxidation in the cerebral cortex. Animals were intraperitoneally pre-treated

786

with saline (SAL) or 20 or 40 mg/Kg of duloxetine (DUL20 and DUL40, respectively) and then,

787

30 min later, pentylenetetrazol (PTZ, 60 mg/Kg i.p.) or saline was administered. An outlier data

788

(below the value of mean minus twice the standard deviation) was identified in the PTZ group. It

789

is indicated by a point and it was excluded out of statistical analysis. Data showed Gaussian

790

distribution and no differences between SDs, so they are presented as mean ± SEM, n=6-10.

791

Two-way ANOVA followed by Bonferroni: *P<0.05 vs. SAL, DUL20, DUL40 and

792

DUL20+PTZ groups.

793

Figure 6. Nitrite levels in the cerebral cortex. Animals were treated intraperitoneally with saline

794

(SAL) or 20 or 40 mg/Kg of duloxetine (DUL20 and DUL40, respectively) and then, 30 min

795

later, pentylenetetrazol (PTZ, 60 mg/Kg i.p.) or saline was administrated. Data showed Gaussian

796

distributions with significant differences between SDs. Data are showed as median ±

797

interquartile range, n=6-10. Kruskal-Wallis test revealed no significant differences between

798

groups.

SC

RI PT

777

AC C

EP

TE D

M AN U

P<0.05 vs. all basal records and levels after saline or duloxetine administration.

ACCEPTED MANUSCRIPT

Figure 7. Catalase (A) and superoxide dismutase (B) activities in the cerebral cortex. Animals

800

were treated intraperitoneally with saline (SAL) or 20 or 40 mg/Kg of duloxetine (DUL20 and

801

DUL40, respectively) and then, 30 min later, pentylenetetrazol (PTZ, 60 mg/Kg i.p.) or saline

802

was administered. All data showed Gaussian distributions without significant differences

803

between SDs. Data are presented as mean ± SEM, n=6-10. Two-way ANOVA followed by

804

Bonferroni: *P<0.05 vs. all groups; #P<0.05 vs. SAL, DUL20, DUL40 and DUL20+PTZ.

RI PT

799

AC C

EP

TE D

M AN U

SC

805

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 3

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights Recently, duloxetine (DUL) protected in a model of absence (non-convulsive) epilepsy Yet, the possible effect of DUL in convulsive seizures was unknown

RI PT

Our results suggested a biphasic effect by DUL on PTZ-induced convulsive seizures

DUL (one therapeutic dose) decreased amplitude of discharges caused by PTZ in EEG

AC C

EP

TE D

M AN U

SC

DUL reduced lipid peroxidation by preventing decreased activities of catalase and SOD