Maternal Exposure to Dietary Selenium Causes Dopaminergic

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Ecotoxicology and Human Environmental Health

Maternal exposure to dietary selenium causes dopaminergic hyperfunction and cognitive impairment in zebrafish offspring Mohammad Naderi, Maud C. O. Ferrari, Douglas P. Chivers, and Som Niyogi Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b04768 • Publication Date (Web): 18 Oct 2018 Downloaded from http://pubs.acs.org on October 20, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35

Environmental Science & Technology

1

Maternal exposure to dietary selenium causes dopaminergic hyperfunction and cognitive

2

impairment in zebrafish offspring

3 4 5

Mohammad Naderi1*, Maud C. O. Ferrari2, Douglas P. Chivers1, Som Niyogi1,3 1Department

of Biology, University of Saskatchewan, 112 Science Place, Saskatoon, SK S7N 5E2, Canada

6 7

2Department

of Veterinary Biomedical Sciences, University of Saskatchewan, 52 Campus Drive, Saskatoon, SK S7N 5B4, Canada

8 9

3Toxicology

Centre, University of Saskatchewan, 44 Campus Drive, Saskatoon, SK S7N 5B3, Canada

10 11 12 13 14

* Corresponding Author: Mohammad Naderi

15

Email: [email protected]

16

Tel: 1 306 850 7337

17

ACS Paragon Plus Environment


18

Abstract

19

Maternal exposure to environmental contaminants is a predisposing factor for neurodevelopmental

20

disorders with associated cognitive and social deficits in offspring. In this study, we investigated

21

the effects of maternal exposure to selenium (Se), a contaminant of potential environmental

22

concern in aquatic ecosystems, on cognitive performance and the underlying mechanisms in F1-

23

generation adult zebrafish. Adult female zebrafish were exposed to different concentrations of

24

dietary Se (3.5, 11.1, or 27.4 μg Se/g dry weight) for a period of 60 d. Fish were subsequently bred

25

and their offspring were collected and raised for 6 months on a normal diet. Maternal exposure to

26

all concentrations of dietary Se induced learning impairment in F1-zebrafish tested in a latent

27

learning task. The results also showed a hyperfunctioning dopaminergic system in fish exhibiting

28

the learning deficit. The hyperfunction of the dopaminergic system was associated with enhanced

29

oxidative stress and alterations in the mRNA abundance of several immediate early and late

30

response genes in the zebrafish brain. Taken together, these results suggest that maternal exposure

31

to dietary Se via alterations in the dopaminergic system leads to persistent neurobehavioral deficits

32

in F1-generation adult zebrafish.

33 34

Keywords:

35

Zebrafish; Selenium; Dopaminergic system; Neurodevelopmental disorders; Learning impairment

36 37 38 39 40 41

2 Environment ACS Paragon Plus

Page 2 of 35

Page 3 of 35


42

1. Introduction:

43

The early development of the central nervous system (CNS) is a critical period in the ontogeny of

44

vertebrates that requires a choreographed sequence of events. This involves gene transcription,

45

neurogenesis, neuronal migration, differentiation, and synaptic connectivity that eventually lead

46

to the development of a functional brain connectome.1 Neurotransmitters and their receptors

47

appear early during nervous system development and are thought to play important roles in these

48

paramount processes.2 Development, structure and functions of neurotransmitter systems are

49

profoundly affected by maternal nutrition.3 In addition, maternal exposure to environmental

50

contaminants could also result in the transfer of toxic chemicals from the mother to the

51

embryo/fetus, influencing early development of the brain and predisposing offspring to

52

neurodevelopmental disorders.4 Over the last decade, a small but growing number of studies have

53

begun to draw attention to the role of prenatal and postnatal exposure to excess amounts of

54

essential trace elements in the pathogenesis of neurodevelopmental disorders.5-6

55

Selenium (Se) is an essential trace element, but also considered to be a priority aquatic

56

contaminant.7 As a constituent of various selenoproteins, Se is involved in diverse biological

57

functions including neurodevelopment and normal functions of the brain in animals.8-9 Despite its

58

biological importance, excess Se can be extremely toxic to a wide range of organisms.7,

59

Anthropogenic activities such as agriculture on seleniferous soils, coal and uranium mining, and

60

fossil-fuel processing operations can release Se into the environment. Selenium eventually enters

61

aquatic ecosystems where it is biotransformed to selenomethionine (SeMet) and transferred

62

through trophic chains to higher trophic level organisms.7 This organic form of Se can be

63

incorporated into egg-yolk protein precursor vitellogenin and maternally transferred to developing

64

embryos. Particularly affected are oviparous vertebrates such as fish, which are at risk of ingesting


10


65

Se-enriched prey.7, 11 Maternal transfer of Se to eggs have been reported to occur in various fish

66

species inhabiting Se-contaminated aquatic ecosystems.12-13 Exposure to Se at this stage

67

commonly manifest as a suite of developmental abnormalities that include spinal curvatures, and

68

craniofacial and fin deformities.14-15 However, to date very little is known about the

69

neurobehavioral consequences of maternal exposure to Se in fish.

70

There is burgeoning evidence that oxidative stress is the main underlying cause of Se (including

71

SeMet) toxicity.16-17 Oxidative stress is, in turn, a common pathogenic mechanism shared by

72

etiological determinants of neurodevelopment disorders.18 Moreover, Se can interfere with and

73

disrupt normal functions of the dopaminergic system.19 Dopamine (DA) is one of the earliest

74

neurotransmitters to emerge in the fish brain,20 and plays an important role in the development and

75

functions of neural circuits involved in the movement, emotion, learning and memory, social and

76

reward-related behaviors.21-23 Importantly, the dopaminergic system is one of the most redox-

77

sensitive transmitter systems in the vertebrate brain. Indeed, a plethora of reactive oxygen species

78

(ROS) is produced by the dopaminergic system itself, which makes it more susceptible to external

79

oxidative insult.24 A wealth of mammalian research points to dysfunction of the dopaminergic

80

system as a common denominator of neurodevelopmental disorders, including attention deficit

81

hyperactivity disorder, autism, and schizophrenia.25 In zebrafish, early transient alterations in the

82

dopaminergic system have also been associated with the persistent behavioral abnormalities in

83

adult zebrafish.26 Moreover, it has been demonstrated that exposure to SeMet during early

84

development (from 2 to 24 hours post fertilization; hpf) leads to a long-lasting spatial learning

85

impairment in adult zebrafish,27 while the underlying mechanism(s) remains to be elucidated. We

86

have previously demonstrated that exposure to dietary SeMet induces oxidative stress and impairs

87

the dopaminergic signaling in the brain, ultimately leading to learning impairment in adult


Page 4 of 35

Page 5 of 35


88

zebrafish.28-29 Therefore, it is logical to hypothesize that embryonic exposure to SeMet via

89

maternal transfer can lead to long-lasting behavioral abnormalities in adult fish, due to the

90

induction of oxidative stress and dysfunction of the DA neurotransmission in the brain.

91

Learning and memory directly or indirectly modulate a broad spectrum of fish behavior in their

92

natural environment. For instance, learning and memory play a decisive role in foraging activities,

93

mate choice strategies, anti-predatory behaviors, agonistic interactions, shoaling behavior, and

94

group joining decision of fish (reviewed in 30). All these behaviors are strongly tied to fish fitness

95

and survival, and thus learning impairment may bring about a wide array of negative consequences

96

for fish. This study aimed to investigate the effects of maternal exposure to dietary SeMet on the

97

learning and memory in zebrafish offspring using a latent learning paradigm. Latent learning is

98

defined as a form of learning that occurs in the absence of any environmental reward and does not

99

immediately manifest itself.30 This form of learning plays a crucial role in the spatial navigation

100

of fish within their environment. Moreover, DA is critically involved in the regulation of this type

101

of learning in zebrafish.31

102

2. Materials and Methods:

103

2.1. Fish maintenance and exposure

104

Adult zebrafish (n = 400) were sourced from the R.J.F. Smith Center for Aquatic Ecology of the

105

University of Saskatchewan. Fish were housed in 30-l glass tanks supplied with aerated de-

106

chlorinated tap water at 28 ± 1 ℃ on a 14/10 h, light/dark cycle. Fish were fed twice daily with

107

flake food (Nutrafin Max flakes, Germany) and frozen brine shrimp (Sally's, San Francisco Bay

108

Brand Inc., USA). Fish were allowed to acclimate to these conditions for at least 3 weeks before

109

the start of the experiment. This acclimation period consisted of a 14-day pre-exposure period to

110

ascertain that unexposed fish were reproductively active.32 Two hundred and fifty-six fish (3.61 ±



111

0.41 cm and 0.67 ± 0.14 g) were then randomly allocated into 16 glass exposure tanks (12 females

112

and 4 males per tank), with four replicate tanks per treatment. Fish were exposed to different

113

nominal concentrations of dietary Se (3, 10, and 30 µg/g dry weight (dw); as SeMet) as described

114

previously, for 60 days.29 These concentrations reflect the range of dietary Se concentration that

115

fish are exposed to in Se contaminated natural waters.33-35 For the first 30 days of exposure, fish

116

were fed everyday with either the control food or the SeMet-spiked food at 5% body weight/day

117

ration. Subsequently, fish received equal portions (2.5%) of control or SeMet-spiked foods and

118

frozen brine shrimp each day for further 30 days. This exposure regime was based on a previous

119

study, which resulted in improved egg production by fish and efficient maternal transfer of Se to

120

eggs.36

121

Since zebrafish have asynchronous ovaries containing follicles at all stages of development,37 they

122

may release eggs with different concentrations of Se among spawning events. Therefore, in this

123

study, we used a modified mixed exposure-breeding regime to minimize the variability of maternal

124

transfer of Se to eggs during the spawning event and possible subsequent effects on developing

125

fish. The rationale behind this exposure regime can be found in Supporting Information (SI), page

126

S2. After 50 days of exposure, adult male fish were removed from each exposure tank while female

127

fish were kept feeding on SeMet-spiked diets. On day 60 of exposure, female fish (n=8) from each

128

exposure tank were netted and placed in breeding tanks. Moreover, 4 sexually mature Se-untreated

129

male fish (originating from the same stock as the fish used in the exposure) were added to each

130

breeding tank to make breeding colonies of 4 Se-untreated males and 8 Se-treated females (4

131

replicates per treatment). Fish were then left to settle overnight. The following morning, spawned

132

eggs were collected from the bottom of the breeding tanks 2 h after the light was turned on.

133

Collected eggs were inspected under a dissecting microscope and scored for viability. Embryos


Page 6 of 35

Page 7 of 35


134

were then transferred to deep Petri dishes (100 × 15mm) containing E3 embryo medium (5 mM

135

NaCl, 0.17 mM KCl, 0.33 mM CaCl2, and 0.33 mM MgSO4) and incubated at 28℃. The percent

136

egg hatchability, larval survival, and larval deformities were determined in F1- generation larval

137

fish as described previously.14 To ascertain chemical transfer of Se from the diets to female fish

138

and subsequent maternal transfer of Se to eggs, Se concentrations in diets (n = 3), whole-body

139

female fish (n = 3), and their eggs (n = 4 replicates of 80-100 pooled eggs) were quantified.

140

Moreover, water samples (n = 3) were collected from each experimental treatment on 30th and 60th

141

days of exposure (1 hr post feeding), and filtered using 0.45 µm disposable nylon filters (Fisher

142

Scientific, Canada) for the measurement of dissolved Se. A sub-group of surviving larval zebrafish

143

with no apparent symptoms of skeletal deformities from each treatment group was reared to 180

144

days post fertilization (dpf) to evaluate the persistent adverse effects of maternal exposure to SeMet

145

on learning performance of adults. Figure S1 illustrates the experimental protocol used in this

146

study.

147

2.2. Latent learning performance

148

In this study, we employed a latent learning paradigm to evaluate learning performance of F1-

149

generation adult zebrafish maternally exposed to dietary Se. The latent learning task was

150

conducted in a maze (Figure S2) comprising a start chamber, a reward chamber, and two tunnels

151

that linked these two chambers to each other. The learning procedure has been described in detail

152

elsewhere.31 Briefly, the learning paradigm consisted of two phases: a training phase followed by

153

a probe phase. During the training phase, only one of the two tunnels leading to the reward chamber

154

was open. Fish in groups of 15 were trained with the maze for 30 min each day while either the

155

right or left tunnel was open. Moreover, no reward was presented in the maze. After 16 consecutive

156

days of training, a probe trial was conducted to test the leaning performance of fish. During this



157

phase, a single fish was located in the start chamber while both right and left tunnels were open

158

and the reward chamber contained a group of fish (n = 6). Since zebrafish are highly social animals,

159

the sight of conspecifics is used as a powerful rewarding stimulus for such studies.38 During the

160

probe trial, zebrafish performance was recorded for 10 min using an over-head HD camera

161

(Logitech c310, USA). The latency to leave the start chamber, the time spent in the correct versus

162

incorrect tunnel, the latency to enter the reward chamber, the time spent in the reward chamber,

163

and locomotion (total distance traveled by the fish) were quantified by MATLAB (Academic

164

version R2015a). At the conclusion of probe trials, fish were euthanized by of Aquacalm (Syndel

165

Laboratories, Canada) and the whole-brain was removed under a dissecting microscope that was

166

equipped with an Axiocam camera (Zeiss, Germany). In addition, the telencephalon was separated

167

from a subset of brains as described previously.28 Brain tissues were stored at −80°C until further

168

analysis.

169

2.3. Measurement of selenium

170

Concentrations of Se in the water, food (SeMet-spiked food and brine shrimp), whole-body tissues,

171

and eggs were measured using a graphite furnace atomic absorption spectrometer (AAnalyst 800,

172

Perkin Elmer, USA) as described previously.28 Briefly, Se concentrations were directly measured

173

in water samples treated with 0.2% (v/v) of concentrated nitric acid. Food, whole-body tissues,

174

and eggs samples were digested with 1N nitric acid (1 to 5 mass (g): volume (ml) ratio) at 60 °C

175

for 48 h. Subsequently, the digested samples were centrifuged at 15,000 g for 4 min and

176

supernatants were collected for Se analysis. Reagent blank and certified reference material (Dolt-

177

4 dogfish liver; National Research Council of Canada) were processed simultaneously to validate

178

the Se measurement procedure used. The recovery rate of Se was found to be 96%.

179

2.4. Biochemical assays


Page 8 of 35

Page 9 of 35


180

Quantification of DA levels in the zebrafish brain (pools of 2-3 brains) was performed using an

181

enzyme-linked immunosorbent assay kit following the manufacturer's instructions (Biovison,

182

USA; detail in ref 29). To evaluate oxidative stress and antioxidant balance in the brain of zebrafish

183

maternally exposed to Se, we measured the ratio of reduced glutathione (GSH) to oxidized

184

glutathione (GSSG) and lipid peroxidation (LPO). For these measurements, 2-3 whole-brains were

185

pooled together. The GSH/GSSG ratio was determined using a fluorometric method in the

186

presence of o-phthalaldehyde as described previously.28-29 LPO was quantified using a

187

commercially available kit following the manufacturer's protocol (Abcam, USA).

188

2.5. Quantitative real-time polymerase chain reaction

189

The mRNA expression of genes associated with DA receptors (DA receptor D1 [DRD1] and D2

190

[DRD2b, DRD2c, DRD3, DRD4a, DRD4b]) as well as genes involved in DA synthesis (tyrosine

191

hydroxylase1 [TH1]), storage (vesicular monoamine transporter-2 [VMAT2]), re-uptake

192

(dopamine transporter [DAT]), and metabolism (monoamine oxidase [MAO]) was evaluated. We

193

also quantified the mRNA expression of brain-derived neurotrophic factor (BDNF) and early

194

growth response1 (EGR-1), which are involved in the modulation of neuronal growth, maturation,

195

and plasticity. In addition, the transcription level of neuronal differentiation 1 (NEUROD 1) was

196

also examined as an indicator of early neurogenesis in the zebrafish brain.39-41 As described in our

197

previous study, the expression of dopaminergic cell markers and genes involved in synaptic

198

plasticity and neurogenesis were assessed in the telencephalon of the brain.28 This brain region

199

was specifically chosen for its involvement in various forms of learning and memory in fish

200

including latent learning.42 Moreover, the aforementioned genes are mainly localized in the

201

telencephalon.28 Thus, measuring the expression of these genes in the telencephalon specifically

202

provides a higher resolution relative to the whole-brain analysis. Moreover, to further evaluate the



203

induction of oxidative stress in the zebrafish brain, we measured the expression level of genes

204

encoding for antioxidant enzymes, including copper/zinc superoxide dismutase (Cu/Zn-SOD),

205

manganese superoxide dismutase (Mn-SOD), catalase (CAT), and glutathione peroxidase 1a

206

(GPX1a) in the zebrafish whole-brain (pools of 2 brains, n = 5). Due to the fact that ROS generated

207

from the auto-oxidation of DA shows widespread toxicity not only in dopaminergic neurons but

208

also in other brain regions, the level of oxidative stress was quantified in the whole-brain tissues.43

209

Total RNA was extracted from the telencephalon (pools of 3) or the whole-brain tissues (pools of

210

2) using the RNeasy Mini Kit (Qiagen, Germany), followed by a DNase treatment and verification

211

by Nanodrop (NanoDrop, Thermo Scientific, USA). Subsequently, the cDNA was produced using

212

a QuantiTect Reverse Transcription® kit (Qiagen, Germany). The gene encoding β-actin was used

213

as the housekeeping gene. For each sample (n = 5), transcript levels of candidate genes and the

214

reference gene were measured in triplicates in 20 μl reaction volumes on an iCycler Thermal

215

Cycler (Bio-Rad, USA) using SYBR Green PCR Master Mix (SensiFAST, SYBR No-ROX Kit,

216

Bioline, USA) as described previously.28 The relative expression of target genes was calculated by

217

the 2-ΔΔct method.44 The sequences of primers have been reported elsewhere.29

218

2.6. Statistical analysis

219

Data are expressed as mean ± the standard deviation (SD), unless stated otherwise. The data were

220

checked for normality and homogeneity of variance using the Kolmogorov−Smirnov one-sample

221

test and Levene’s test, respectively. Once data met the normal distribution and showed

222

homogeneity of variance, one-way analysis of variance (ANOVA) followed by Tukey's post hoc

223

test was employed to determine significant differences among treatment groups. In the case of

224

heteroscedasticity, the Welch’s test with the Games−Howell post hoc test was performed. If the

225

data did not have equal variance and normal distribution, Kruskal-Wallis test with Dunn's post-


Page 10 of 35

Page 11 of 35


226

hoc test was performed. When data were percentages (percent viability, hatchability, and total

227

deformities) they were normalized using an arcsine square root transformation. A Kaplan–Meier

228

survival analysis with the log-rank test was carried out to determine differences in cumulative

229

survival rates among different treatment groups. The alpha level was set at 0.05. However, the

230

Bonferroni's correction was applied to minimize Type I error rate resulting from multiple

231

comparisons, when appropriate.

232

3. Results:

233

3.1. Selenium concentrations

234

Selenium concentrations in water samples (dissolved Se), diets (both in flake food and brine

235

shrimp), whole-body adult zebrafish, and eggs are shown in Table S1. Dietary exposure to SeMet-

236

spiked diets led to a dose-dependent and significant increase in whole-body Se accumulation in

237

female fish (see Table S1). Moreover, our results showed that maternal transfer of Se to eggs were

238

proportional to the Se concentrations in diets fed to the female fish (Table S1). In other words, Se

239

concentrations of 2.3 ± 1.34, 3.95 ± 1.52, 9.21 ± 2.89, and 24.25 ± 6.46 µg/g wet weight were

240

found in eggs of adult zebrafish fed with diets containing 1.27 ± 0.43 (control food), 3.5 ± 0.61,

241

11.1 ± 1.57, and 27.4 ± 0.81 µg Se/g dw, respectively. The Se concentrations in eggs collected

242

from adult female zebrafish fed with SeMet-spiked diets were significantly different relative to the

243

control (F3, 6.25 = 16.03, p < 0.001). A significantly higher Se accumulation was observed in eggs

244

collected from adult female fish treated with 11.1 and 27.4 µg Se/g dw diets (p=0.037 and

245

p=0.017).

246

3.2. Embryo viability, hatchability, larval deformities, and survival rate

247

Maternal exposure to different concentrations of dietary Se did not significantly affect percent egg

248

viability (F3, 12=1.10, p=0.385; Table S1). However, the hatching rate significantly differed among



249

treatment groups (F3, 12 = 7.08, p = 0.005; Table S1). The percent hatch of eggs laid by fish treated

250

with the highest concentration of Se was significantly lower than that of controls (p = 0.045).

251

Maternal transfer of Se to eggs also resulted in larval deformities evaluated at 6 dpf (F3, 12 = 68.68,

252

p < 0.001; Table S1). The occurrence of larval deformities was higher in groups maternally

253

exposed to 11.1 and 27.4 µg Se/g diets compared to the control (both p < 0.001). The survival

254

distributions of embryos/larvae (2-6 dpf) was also significantly different among treatment groups

255

(X23 = 143.55, p < 0.001; Figure S3). A higher mortality rate was observed in embryo/larval fish

256

maternally exposed to 11.1 and 27.4 µg Se/g diet compared to the control (both p < 0.001).

257

3.3. Latent learning performance

258

The latency to leave the start chamber differed significantly among treatment groups (X23 = 36.21,

259

p < 0.001). A prolonged latency to leave the start chamber was observed in all Se-treated groups

260

compared to controls (all p < 0.001; Figure 1A). The amount of time that fish spent in the correct

261

versus incorrect tunnel was also significantly different among groups (F3, 185 = 14.01, p < 0.001;

262

Figure 1B). Fish in groups maternally treated with 3.5, 11.1, and 27.4 μg Se/g diets spent

263

significantly less time in their training tunnel (all p < 0.001). Moreover, the latency to enter the

264

reward chamber was significantly affected in F1-generation adult zebrafish (X23 = 37.57, p < 0.001;

265

Figure 1C). A prolonged latency to reach and enter the reward chamber was found in all groups

266

maternally exposed to dietary Se compared to controls (all p < 0.001). A significant difference in

267

the amount of time fish spent in the reward chamber was found among treatment groups (F3, 185 =

268

7.20, p < 0.001). As represented in Figure 1D, maternal exposure to all SeMet-spiked diets reduced

269

the amount of time that fish spent in the reward chamber, which contained their conspecifics (all

270

p < 0.009). The locomotion (total distance traveled by fish) was another parameter altered

271

significantly among treatment groups (X23 = 17.45, p = 0.001). However, as shown in Figure 1E,


Page 12 of 35

Page 13 of 35


272

the locomotor activity decreased in fish maternally exposed to 3.5 µg Se/g diet compared to fish

273

maternally treated with 27.4 μg Se/g diet (p < 0.001). However, no statistically significant

274

differences were found in Se treated groups when compared with the control group (all p > 0.05).

275

3.4. Dopaminergic cell markers in the brain

276

Maternal exposure to dietary Se induced a significant change in DA levels of the brain (F3, 12 =

277

33.18, p < 0.001; Table 1). An elevated level of DA was found in groups maternally exposed to

278

11.1 and 27.4 μg Se/g diets (p = 0.014 and p

Maternal Exposure to Dietary Selenium Causes Dopaminergic

Recommend Documents