Caffeine increases food intake while reducing anxiety-related behaviors

Accepted Manuscript Caffeine increases food intake while reducing anxiety-related behaviors Patrick Sweeney, Russell Levack, Jared Watters, Zhenping Xu, Yunlei Yang PII:

S0195-6663(16)30102-7

DOI:

10.1016/j.appet.2016.03.013

Reference:

APPET 2922

To appear in:

Appetite

Received Date: 12 January 2016 Revised Date:

8 March 2016

Accepted Date: 9 March 2016

Please cite this article as: Sweeney P., Levack R., Watters J., Xu Z. & Yang Y., Caffeine increases food intake while reducing anxiety-related behaviors, Appetite (2016), doi: 10.1016/j.appet.2016.03.013. 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.

ACCEPTED MANUSCRIPT

1 2 3

5 6 7

RI PT

4

Caffeine increases food intake while reducing anxietyrelated behaviors

SC

8 9

Patrick Sweeney1, Russell Levack1, Jared Watters1, Zhenping Xu, and Yunlei Yang*

M AN U

10 11 12 13 14

Department of Neuroscience and Physiology, State University of New York Upstate Medical University, Syracuse, NY 13210, USA

1 These authors contributed equally

EP

* Correspondence: Yunlei Yang Institute For Human Performance, NRB#3609, 505 Irving Ave, New York, NY 13210, USA [email protected] Phone: 315-464-7733 Fax: 315-464-7712

AC C

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

TE D

15

32

1

ACCEPTED MANUSCRIPT

ABSTRACT

34

The objective of this study was to determine the effects of different doses of caffeine on

35

appetite and anxiety-related behavior. Additionally, we sought to determine if withdrawal

36

from chronic caffeine administration promotes anxiety. In this study, we utilized rodent

37

open field testing and feeding behavior assays to determine the effects of caffeine on feeding

38

and anxiety-related behavior (n=8 mice; 4-8 weeks old). We also measured 2 hour and 24

39

hour food intake and body-weight during daily administration of caffeine (n=12 mice; 4-8

40

weeks old). To test for caffeine withdrawal induced anxiety, anxiety-related behavior in

41

rodents was quantified following withdrawal from four consecutive days of caffeine

42

administration (n=12 mice; 4-8 weeks old). We find that acute caffeine administration

43

increases food intake in a dose-dependent manner with lower doses of caffeine more

44

significantly increasing food intake than higher doses. Acute caffeine administration also

45

reduced anxiety-related behaviors in mice without significantly altering locomotor activity.

46

However, we did not observe any differences in 24 hour food intake or body weight

47

following chronic caffeine administration and there were no observable differences in

48

anxiety-related behaviors during caffeine withdrawal. In conclusion, we find that caffeine

49

can both increase appetite and decrease anxiety-related behaviors in a dose dependent

50

fashion. Given the complex relationship between appetite and anxiety, the present study

51

provides additional insights into potential caffeine-based pharmacological mechanisms

52

governing appetite and anxiety disorders, such as bulimia nervosa.

AC C

EP

TE D

M AN U

SC

RI PT

33

53 54 55

2

ACCEPTED MANUSCRIPT

1. Introduction

57

The most broadly consumed behaviorally active substance worldwide is caffeine (Fredholm et

58

al., 1999). Americans consume a daily average of 165 mg of caffeine from beverages alone

59

(Mitchell et al., 2014). Additionally, it has been reported that 85% of Americans will consume a

60

minimum of one caffeinated drink per day (Mitchell et al., 2014). Given that 34.9% of

61

Americans over 20 years of age are considered obese, it is paramount to address the possible

62

effects of caffeine on food intake (Ogden et al., 2014). Interestingly, the effects of caffeine on

63

food intake are controversial. For instance, a study by Retzbach et al. (2014) showed that

64

caffeine increases lever pressing for sucrose, suggesting that caffeine can increase appetitive

65

behavior and feeding. This increase in lever pressing was accompanied by an increase in c-fos

66

expression in the nucleus accumbens, a critical brain region for reward (Berridge &

67

Kringelbach., M.L., 2015) and hedonic control of feeding behavior (O’Conner et al., 2015;

68

Castro & Berridge., 2014). Furthermore, a study by Bonaventura et al. (2012) showed that both

69

high-palatable and low-palatable food intake decreased in rats when A2A adenosine receptors

70

were activated using A2A agonists. Since caffeine exerts its effect by blocking A2A receptors

71

(Ribeiro & Sebastiao, 2010), this finding suggests that caffeine can stimulate food intake.

72

Conversely, Pettenuzzo et al. (2008) found that chronic caffeine administration in rats does not

73

affect the consumption of rat chow, but decreases palatable food intake. Other studies have

74

shown that caffeine has no effect on appetite sensations or energy intake, while at the same time

75

it has been shown that caffeine decreases both consumption of powdered food and body weight

76

in rats (Schubert et al., 2014; Gavrieli et al., 2011; Racotta et al., 1994). A similar effect has been

77

observed in humans where caffeine was shown to intensify the anti-appetitive effects of nicotine

78

(Jessen et al., 2005). In addition to these findings, Gavreli et al. (2013) showed that consumption

AC C

EP

TE D

M AN U

SC

RI PT

56

3

ACCEPTED MANUSCRIPT

79

of caffeinated coffee reduced energy intake in obese individuals. These disparate findings may

80

be explained by the doses of caffeine administered and/or the routes of caffeine administration

81

(injection vs. oral consumption). The neural mechanisms for caffeine's effects on feeding are incompletely understood.

83

However, caffeine has been shown to increase excitatory synaptic input to agouti-related peptide

84

(AgRP) neurons located in the hypothalamic arcuate nucleus (Yang et al., 2011). When

85

activated, AgRP neurons increase feeding behavior by driving consummatory and appetitive

86

behaviors (Aponte et al., 2011; Sternson and Atasoy, 2014). Additionally, caffeine can both

87

increase dopamine levels and activate dopamine neurons in the nucleus accumbens, a brain

88

region critical for motivation and feeding behaviors (Solinas et al., 2002; Retzbach et al., 2014;

89

O’Connor et al, 2015). Taken together, emerging evidence suggests that caffeine increases

90

feeding behavior, at least in part, by increasing the activity of appetite promoting AGRP neurons

91

in the arcuate nucleus and dopamine neurons in the nucleus accumbens. We therefore proposed

92

that acute administration of caffeine would increase food intake, depending on the dose of

93

caffeine administered.

TE D

M AN U

SC

RI PT

82

Beyond caffeine’s effects on feeding, recent reports have implicated caffeine in anxiety-

95

related behaviors. For example, Maximino et al. (2011) showed that caffeine can induce an

96

anxiety-like behavioral response in zebrafish in a dose-dependent manner by blocking A1

97

receptors. However, a recent report showed that caffeine can have anxiolytic effects in rats in the

98

open field test and elevated plus maze (Hughes et al., 2014). Consistently, caffeine

99

administration can exert anxiolytic or anxiogenic effects in rats depending on the anxiety test

100

employed, the rat strain, and the sex of the rat (Hughes and Hancock, 2016). Caffeine has also

101

been reported to have anti-depressive like qualities. For example, a recent study demonstrated

AC C

EP

94

4

ACCEPTED MANUSCRIPT

that chronic pre-treatment with caffeine prior to social defeat stress in mice can reduce the

103

subsequent development of depressive and anxiogenic like phenotypes (Yin et al., 2015). This

104

effect was prevented when caffeine was co-administered with a selective dopamine D1 receptor

105

antagonist, indicating that caffeine reduces stress induced anxiety and depressive phenotypes via

106

dopaminergic signaling. By contrast, caffeine withdrawal is known to cause a large number of

107

behavioral effects including: irritability, sleepiness, nervousness, restlessness, and anxiety (Dews

108

et al., 2002).

SC

RI PT

102

Due to the known interactions between anxiety-related behaviors and feeding, the current

110

study investigated how both acute and chronic caffeine administration affects feeding and

111

anxiety in AgRP-IRES-Cre transgenic mouse, a common transgenic animal for feeding studies

112

(Maniam and Morris, 2012; Hardaway et al., 2015). Based on the known reciprocal interaction

113

between feeding and anxiety behavior (Maniam and Morris, 2012; Hardaway et al., 2015), we

114

hypothesized that low doses of caffeine would exert an anxiolytic effect and increase feeding

115

while high doses would have an anxiogenic effect and decrease feeding. To elucidate the effects

116

of caffeine on feeding behavior and anxiety, we measured free access food intake and open field

117

exploratory behavior following acute caffeine administration. We also tested for caffeine

118

withdrawal induced anxiety by performing open field behavioral testing following withdrawal

119

from chronic caffeine administration.

TE D

EP

AC C

120

M AN U

109

121

2. Methods

122

All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of

123

State University of New York Upstate Medical University, according to US National Institute of

124

Health’s Guide for the Care and Use of Laboratory Animals. AgRP-IRES-Cre transgenic mice

5

ACCEPTED MANUSCRIPT

were obtained from the Jackson Laboratory (Bar Harbor, ME, USA). These mice do not have

126

any known phenotypes when Cre recombinase is not active, as used in this study. AgRP-IRES-

127

Cre mice were used to maintain consistency with potential future studies which will use Cre-loxP

128

recombination strategies to manipulate hypothalamic AgRP neurons to determine their role in

129

caffeine induced feeding behavior. Male and female mice between the ages of 4 and 8 weeks

130

were used for all experiments. Mice were maintained on a 12 hour light/dark cycle (light onset at

131

6:30 am) and housed in ventilated cages with free access to water and standard rodent chow

132

(LabDiet: 5008 Formulab Diet). All behavioral experiments were conducted in the home cage

133

unless otherwise noted.

M AN U

SC

RI PT

125

134

2.1 Feeding Behavior Assays and Caffeine treatment

136

Prior to feeding behavior experiments, mice were single caged and habituated for one week.

137

Animals were handled daily for the week prior to beginning experiments to habituate them to

138

behavioral assays. All mice were handled by picking them up by their scruff multiple times to

139

mimic handling conditions seen during i.p. injections. Handling was performed to reduce stress

140

responses in response to i.p. injections. Food intake was measured at the indicated time-points by

141

briefly removing the food from the hopper and obtaining its weight. For the acute caffeine intake

142

experiment, 8 mice were used and received the same treatment. Intraperitoneal (i.p.) injections

143

were performed daily at 9:00 am for 8 days. Mice received saline injections (0.9% saline; 200 µl)

144

on days 1-3. Saline injections were performed to obtain baseline food intake measurements prior

145

to caffeine administration. On days 4-8, the mice received an injection of caffeine (Tocris

146

Bioscience, Bristol, UK) every other day with increasing doses (6 mg/kg, 12 mg/kg, and 24

147

mg/kg; dissolved in 200 µl 0.9% saline) and received saline injections on the days where

AC C

EP

TE D

135

6

ACCEPTED MANUSCRIPT

caffeine was not administered. Caffeine injections were administered every other day to prevent

149

habituation to caffeine treatment. Food was replaced daily with approximately 20 grams of fresh

150

standard chow. Food intake was measured at various time points following injections: 30

151

minutes, 1 hour, 2 hours, and 5 hours. Average food intake and body weight were calculated for

152

each mouse.

RI PT

148

For chronic caffeine administration, 12 mice were used and were separated into two equal

154

groups. Animals were assigned to control (saline injections) or experimental groups (caffeine

155

injections). Injections were administered at 1:45 pm and 2:15 pm for all 10 days. On days 1-2,

156

both of the groups received saline injections at each of the two indicated time-points. Saline

157

injections were administered to obtain a baseline measure of food intake prior to treatment. On

158

days 3-6, the experimental group received a saline injection followed by a caffeine injection (20

159

mg/kg in 200 µl 0.9% saline), while the control received two saline injections. Both groups

160

received two saline injections on days 7-10. Saline was administered on days 7-10 to maintain

161

consistency in injection conditions while testing for behavioral symptoms of caffeine

162

withdrawal. Food was replaced daily with approximately 15 grams of fresh standard chow. Body

163

weight was measured immediately following the first injection each day. Food intake was

164

measured at various time points following the second injection: 30 minutes, 1 hour, 2 hours, and

165

24 hours. Average food intake and body weight were calculated for each mouse.

M AN U

TE D

EP

AC C

166

SC

153

167

2.2 Open Field Test

168

To measure the potential effects of caffeine on anxiety-related behaviors, we performed an open

169

field exploration task (Fig. 2). In this task mice were allowed to explore a brightly lit open field

170

for 10 minutes. The bright, exposed center of the field is classically considered to be a less

7

ACCEPTED MANUSCRIPT

anxious environment than the corner of the arena. As such, anxious mice generally enter the

172

center less frequently and spend less time in this region (Hall and Ballenchy, 1932; Calhoon and

173

Tye, 2015). In this task mice were habituated to the behavioral room for 30 minutes prior to

174

beginning the experimental sessions. The open field room contained a brightly lit, 500 mm2

175

arena. The center of the arena was considered to be the middle 250 mm2 of the arena.

176

Exploratory behavior was recorded and analyzed for 10 minutes using ANY-maze software

177

(Stoelting). Between each trial, the arena was cleaned with a 70% ethanol solution. The total

178

distance traveled, average speed, distance traveled in center, time spent in center, number of

179

entries into the center, percent distance traveled in center, distance traveled in periphery, and

180

time spent in periphery were calculated for each mouse. The percent distance traveled in the

181

center area was determined by dividing the distance traveled in the center by the total distance

182

traveled. For the acute caffeine intake experiments, open field testing was run one week after

183

feeding behavior experiments. On the day of open field testing, caffeine (20 mg/kg) or saline was

184

administered via i.p. injection 10 minutes prior to running the open field test. Experiments were

185

performed in a counterbalanced order, where half of the mice received saline on day one and

186

caffeine on day two and the other half received caffeine on day one and saline on day two. Each

187

animal experienced two trials in the open field (caffeine on day 1 & saline on day 2 or saline on

188

day 1 & caffeine on day 2). Average values for saline and caffeine treatments were calculated by

189

averaging day 1 and day 2 values for each treatment condition. For the chronic caffeine intake

190

experiments, open field testing was conducted on three consecutive days following withdrawal

191

from caffeine administration in animals injected with either caffeine or vehicle saline. Relative

192

distance traveled in center was calculated on each day for each mouse as a measure of anxiety

193

from caffeine withdrawal.

AC C

EP

TE D

M AN U

SC

RI PT

171

194 8

ACCEPTED MANUSCRIPT

3. Data Analysis

196

All data were analyzed using GraphPad Prism 6.0 software. For acute feeding and anxiety

197

experiments, average food intake and anxiety measurements were calculated for each mouse for

198

statistical analysis. Feeding behavior experiments were repeated at least twice per mouse for

199

each condition (saline or caffeine) and average food intake was calculated for each mouse for

200

statistical analysis. For acute open field anxiety experiments, each animal experienced two trials

201

in the open field (caffeine on day 1 & saline on day 2 or saline on day 1 & caffeine on day 2).

202

Average values for saline and caffeine treatments were calculating for statistical analysis by

203

averaging day 1 and day 2 values for each treatment condition. For caffeine withdrawal

204

experiments, average food intake or anxiety measurement were obtained for statistical analysis

205

by averaging food intake or anxiety measurements on each individual day for all of the mice in

206

each treatment group (saline vs caffeine withdrawl). 1-way ANOVAs with Tukey’s post hoc test

207

were performed to analyze acute caffeine feeding behavior experiments. Open field data were

208

evaluated with paired t-test. For chronic caffeine feeding experiments, 2-way ANOVAs with

209

Holm-Sidak post hoc testing were performed.

210

EP

TE D

M AN U

SC

RI PT

195

4. Results

212

4.1 Acute caffeine administration markedly increases food intake

213

Both 6 mg/kg and 12 mg/kg doses of caffeine significantly increased food intake after two hours

214

(Fig. 1A; n = 8 mice per group; F (3, 52) = 9.2; p < 0.0001). However, no significant effect of

215

caffeine administration was observed following 24 mg/kg caffeine injections. Five hours

216

following caffeine injections, only 6 mg/kg caffeine significantly increased food intake (Fig. 1B;

217

n = 8 per group; F (3, 52) = 5.8; p = 0.001). Importantly, caffeine administration did not

AC C

211

9

ACCEPTED MANUSCRIPT

significantly affect animals’ locomotor activities (Fig 2A and B; paired Student’s t-test),

219

although a trend towards increasing locomotion was observed with caffeine administration.

220

Taken together, low doses of caffeine treatment increased food intake with less of an effect

221

observed with progressively higher doses.

RI PT

218

222

4.2 Acute caffeine administration decreases anxiety-related behaviors

224

Acute administration of caffeine (i.p.; 20 mg/kg in 200 µl 0.9 % saline) increased entries into the

225

center (Fig. 2C; paired Student’s t-test; t(7) = 4.09; p = 0.004 ), time in the center (Fig. 2D;

226

paired Student’s t-test; t(7) = 4.84; p = 0.0019), and the distance traveled in the center (Fig. 2E;

227

paired Student’s t-test; t(7) = 4.20; p = 0.004), suggesting that caffeine administration reduces

228

anxiety. In contrast, caffeine administration decreased the time mice spent in the less anxiogenic

229

corner of the field (Fig. 2F; paired Student’s t-test; t(7) = 4.85; p = 0.002). No significant

230

difference was detected for the distance traveled in the corner in response to caffeine

231

administration (Fig. 2G).

M AN U

TE D

232

SC

223

4.3 Chronic caffeine administration does not alter daily food intake or body weight

234

Caffeine administration did not significantly affect 24 hour food intake and body weight (Fig. 3A

235

and B). However, consistent with the trends shown in the Fig. 1, two hour food intake was

236

increased following caffeine administration (Fig. 3C; F (6, 70) = 3.09; p = 0.01). As withdrawal

237

from caffeine use has been associated with anxiety, we performed open field tests on three

238

consecutive days following chronic administration of caffeine or saline (Dews et al., 2002). We

239

did not find any significant changes in anxiety during withdrawal days in animals treated with

AC C

EP

233

10

ACCEPTED MANUSCRIPT

240

saline vs. caffeine (Fig. 3D; unpaired Student’s t-test), suggesting that withdrawal from once

241

daily caffeine administration is not strongly anxiogenic.

242

5. Discussion

244

In this study, we find that caffeine increases food intake in a dose-dependent manner. For

245

instance, we find that the lower dose of caffeine markedly increased food intake while the extent

246

of the increase was reduced with higher doses of caffeine (Fig. 1).

SC

RI PT

243

The increase in food intake by caffeine can last at least 5 hours, which may be due to the

248

activation of AgRP neurons localized in the hypothalamic arcuate nucleus (ARC) since previous

249

work has shown that caffeine increases the synaptic strength at AgRP neurons (Yang et al.,

250

2011). It is well demonstrated that activation of AgRP neurons increases food intake (Aponte et

251

al., 2011; Sternson and Atasoy, 2014). Interestingly, the increase in food intake by activation of

252

AgRP neurons can happen over a delayed period via agouti-related peptide action on

253

melanocortin receptors instead of the fast inhibitory GABA transmitter (Krashes et al., 2013),

254

supporting our results that food intake wasn’t significantly increased until two hours after

255

caffeine injections. Additional mechanisms are possible for caffeine induced increases in food

256

intake, such as caffeine acting on nucleus accumbens neurons to facilitate feeding (Retzbach et

257

al., 2014). Therefore, we postulate that caffeine acts on both AgRP neurons and nucleus

258

accumbens neurons through adenosine receptors to stimulate hunger and increase food intake.

259

Future studies will dissect the neural mechanisms underlying caffeine’s effect on feeding.

AC C

EP

TE D

M AN U

247

260

Since acute administration of caffeine increased feeding, we sought to determine if long

261

term caffeine administration can increase daily food intake and body weight. Surprisingly, daily

262

body weight and food intake over the course of 24 hours were not significantly different between

11

ACCEPTED MANUSCRIPT

the caffeine and control vehicle saline-treated groups. A possible explanation for this is the short

264

half-life of caffeine as caffeine has been reported to be completely metabolized in mice after 4-5

265

hours (Hartmann and Czok, 1980). Therefore, after 24 hours, caffeine is likely to be no longer

266

active in the brain. Secondly, as caffeine’s effects are short- lived, counter-regulatory neural

267

circuits are likely to be involved in maintaining a stable level of food intake and body weight, a

268

process that has been shown to be exceedingly complex and highly regulated (Williams and

269

Elmquist, 2012). Another explanation for why our results conflict with other groups’ is that the

270

present findings were conducted during the light cycle. It is possible that caffeine treatment has

271

different effects on feeding and anxiety depending on endogenous adenosine levels, which are

272

naturally low when the animal is awake during the dark period and higher during the light

273

period. As adenosine levels are known to naturally fluctuate during the day, and since caffeine is

274

an adenosine antagonist, future work could be conducted to investigate the potential differences

275

of caffeine treatment during various times in the day (Huang et al., 2011).

TE D

M AN U

SC

RI PT

263

The literature regarding the effects of caffeine on anxiety is controversial. Hughes et al.

277

(2014) showed that caffeine reduces stress in rats. Contrary to this, an anxiogenic effect of

278

caffeine has been observed in rats (Bhattakarya et al., 1997). It has been suggested that

279

increasing concentrations of caffeine can cause different effects by acting through different

280

receptors, as higher doses of caffeine increase anxiety in humans and lower doses decrease

281

anxiety (Cunha et al., 2008).

AC C

282

EP

276

Our results show that caffeine administration reduces anxiety levels in mice during open

283

field exploration (Fig. 2C-F). This is consistent with similar testing performed on rats (Hughes et

284

al., 2014). However, the open field test also showed that caffeine withdrawal had no significant

285

effects on various measures of anxiety (Fig. 3D). Given that anxiety is a symptom of caffeine

12

ACCEPTED MANUSCRIPT

withdrawal, we expected to see an increase in anxiety during caffeine withdrawal (Griffiths and

287

Woodson, 1988). However, since we did not see an increase in anxiety, it is possible that once

288

daily caffeine administration is not sufficient to produce subsequent withdrawal symptoms.

289

Furthermore, previous work showed that consumption of caffeine below doses of 65 mg/kg/day

290

had no effect on the locomotor activity of rats during their withdrawal period (Finn and

291

Holtzman, 1986). Therefore, we postulate that the amount of caffeine injected into our mice may

292

have been too small to produce observable, withdrawal induced anxiety. Furthermore, based on

293

the literature and our current study, caffeine can induce both anxiogenic and anxiolytic effects

294

depending on the dose of caffeine administered.

M AN U

SC

RI PT

286

295

6. Conclusion

297

Overall, our present findings show that caffeine increases appetite in mice. Additionally, our

298

results show that at a dose of 20 mg/kg, caffeine displays an anxiolytic effect. This finding is in

299

accord with the idea that high doses of caffeine exert an anxiogenic effect while low doses exert

300

an anxiolytic effect (Cunha et al., 2008). We also observed that caffeine withdrawal had no effect

301

on the anxiety levels of mice. This is likely due to the dosage of caffeine administered and/or the

302

frequency of caffeine administration. Further testing is needed to determine the precise dose and

303

frequency of administration required to produce caffeine withdrawal symptoms.

EP

AC C

304

TE D

296

Given the high usage of caffeine worldwide, it is important to precisely understand how

305

this compound affects relevant behaviors. Our findings suggest that in addition to caffeine’s well

306

known effects on wakefulness, it may also exert additional effects on appetite and anxiety.

307

Further work is needed to determine the neural correlates and potential targets for caffeine’s

308

wide ranging effects. These targets may reveal additional insights or potential therapeutic

13

ACCEPTED MANUSCRIPT

309

strategies for the many neurological disorders characterized by abnormalities in sleep, feeding or

310

anxiety.

311

Conflict of interest

313

The authors declare that there is no financial interest related to this work.

RI PT

312

314

Author Contributions

316

Y.Y. conceived the idea. P.S., Y.Y. designed the experiments; P.S., R.L., J.W., Z.X. performed

317

the experiments analyzed the data; P.S., R.L., J.W., and Y.Y. wrote the paper with inputs from

318

all other contributors.

M AN U

SC

315

319

Acknowledgements

321

We thank R. Quinn for mouse husbandry, and thank all the members of Yang laboratory for the

322

critical comments on this manuscript. This research was supported by the State University of

323

New York.

EP

324

TE D

320

References

326 327 328 329 330 331 332 333 334 335 336 337 338

Aponte, Y., Atasoy, D., Sternson, S. M. (2011). AGRP neurons are sufficient to orchestrate feeding behavior rapidly and without training. Nature Neuroscience, 14(3), 351-355.

AC C

325

Atasoy, D., Betley, J. N., Su, H. H., Sternson, S. M. (2012). Deconstruction of a neural circuit for hunger. Nature, 488(7410), 172-177. Bhattacharya, S. K., Satyan, K. S., & Chakrabarti, A. (1997). Anxiogenic action of caffeine: an experimental study in rats. Journal of Psychopharmacology (Oxford, England), 11(3), 219. Beer, B., Chasin, M., Clody, D. E., Vogel, J. R., & Horovitz, Z. P. (1972). Cyclic Adenosine Monophosphate Phosphodiesterase in Brain: Effect on Anxiety. Science, (4033), 428430. 14

ACCEPTED MANUSCRIPT

Berridge, K., & Kringelbach, M. (2015). Pleasure Systems in the Brain. Neuron, 86(3), 646-664.

RI PT

Bonaventura, M. D., Cifani, C., Lambertucci, C., Volpini, R., Cristalli, G., & Massi, M. (2012). A2A adenosine receptor agonists reduce both high-palatability and low-palatability food intake in female rats. Behavioural Pharmacology, 23(5-6), 567-574. Calhoon, G & Tye, K.M., (2015). Resolving the neural circuits of anxiety. Nature Neuroscience, 18(10), 1394-1404. Castro, D. C., & Berridge, K. C. (2014). Advances in the neurobiological bases for food ‘liking’ versus ‘wanting’. Physiology & Behavior, 136, 22-30.

M AN U

SC

Cunha, R. A., Ferré, S., Vaugeois, J., & Chen, J. (2008). Potential therapeutic interest of adenosine A2A receptors in psychiatric disorders. Current Pharmaceutical Design, 14(15), 1512-1524. Dews, P. B., O'Brien, C. P., & Bergman, J. (2002). Caffeine: behavioral effects of withdrawal and related issues. Food and Chemical Toxicology, 40, 1257-1261. El Yacoubi, M., Ledent, C., Parmentier, M., Costentin, J., & Vaugeois, J. (2005). Reduced appetite for caffeine in adenosine A2A receptor knockout mice. European Journal of Pharmacology, 519(3), 290-291.

TE D

Finn, I. B., & Holtzman, S. G. (1986). Tolerance to caffeine-induced stimulation of locomotor activity in rats. Journal of Pharmacology and Experimental Therapeutics, 238(2), 542546.

EP

Fredholm, B. B., Bättig, K., Holmén, J., Nehlig, A., & Zvartau, E. E. (1999). Actions of caffeine in the brain with special reference to factors that contribute to its widespread use. Pharmacological Reviews, 51(1), 83-133. Gavrieli, A., Karfopoulou, E., Kardatou, E., Spyreli, E., Fragopoulou, E., Mantzoros, C., & Yannakoulia, M. (2013). Effect of different amounts of coffee on dietary intake and appetite of normal weight and overweight/obese individuals. Obesity, 21(6), 1127-1132.

AC C

339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385

Gavrieli A, Yannakoulia M, Fragopoulou E, Margaritopoulos D, Chamberland JP, Kaisari P, Kavouras SA, Mantzoros CS. (2011) Caffeinated coffee does not acutely affect energy intake, appetite, or inflammation but prevents serum cortisol concentrations from falling in healthy men. The Journal of Nutrition 141: 703–707 Griffiths, R. R., & Woodson, P. P. (1988). Caffeine physical dependence: a review of human and laboratory animal studies. Psychopharmacology, 94(4), 437-451. Hall, C. & Ballachey, E.L. (1932). A study of the rat’s behavior in a field. A contribution to method in comparative psychology. Univ. Calif. Publ. Psychol. 6, 1–12

15

ACCEPTED MANUSCRIPT

Hardaway, J. A., N. A. Crowley, C. M. Bulik and T. L. Kash (2015). Integrated circuits and molecular components for stress and feeding: implications for eating disorders. Genes Brain Behav 14(1): 85-97.

RI PT

Hartmann, M., & Czok, G. (1980). Pharmacokinetics of caffeine in mice and its modification by ethanol. Zeitschrift Für Ernährungswissenschaft, 19(3), 215-227. Huang, Z. L., Urade, Y., & Hayaishi, O. (2011). The role of adenosine in the regulation of sleep. Current Topics in Medicinal Chemistry, 11(8), 1047-1057.

SC

Hughes, R. N., & Hancock, N. J. (2016). Strain-dependent effects of acute caffeine on anxiety related behavior in PVG/c, Long–Evans and Wistar rats. Pharmacology Biochemistry and Behavior, 140, 51-61.

M AN U

Hughes, R. N., Hancock, N. J., Henwood, G. A., & Rapley, S. A. (2014). Research report: Evidence for anxiolytic effects of acute caffeine on anxiety-related behavior in male and female rats tested with and without bright light. Behavioural Brain Research, 271, 7-15. Jessen, A., Buemann, B., Toubro, S., Skovgaard, I. M., & Astrup, A. (2005). The appetitesuppressant effect of nicotine is enhanced by caffeine. Diabetes, Obesity & Metabolism, 7(4), 327-333.

TE D

Krashes MJ, Shah BP, Koda S, Lowell BB (2013). Rapid versus delayed stimulation of feeding by the endogenously released AgRP neuron mediators, GABA, NPY and AgRP. Cell metabolism. 18(4), 588-595. Maniam, J. and M. J. Morris (2012). The link between stress and feeding behaviour. Neuropharmacology. 63(1), 97-110.

EP

Maximino, C., Lima, M. G., Olivera, K. R. M., Picanço-Diniz, D.,L.W., & Herculano, A. M. (2011). Adenosine A1, but not A2, receptor blockade increases anxiety and arousal in Zebrafish. Basic & Clinical Pharmacology & Toxicology, 109(3), 203-207. Mitchell, D. C., Knight, C. A., Hockenberry, J., Teplansky, R., & Hartman, T. J. (2014). Beverage caffeine intakes in the U.S. Food and Chemical Toxicology, 63, 136-142.

AC C

386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

O’Connor, E. C., Kremer, Y., Lefort, S., Harada, M., Pascoli, V., Rohner, C., & Lüscher, C. (2015). Accumbal D1R Neurons Projecting to Lateral Hypothalamus Authorize Feeding. Neuron, 88, 553-564. Ogden CL, Carroll MD, Kit BK, Flegal KM. (2014). Prevalence of childhood and adult obesity in the united states, 2011-2012. Jama, 311(8), 806-814. Pettenuzzo, L. F., Noschang, C., von Pozzer Toigo, E., Fachin, A., Vendite, D., & Dalmaz, C. (2008). Effects of chronic administration of caffeine and stress on feeding behavior of rats. Physiology & Behavior,95(3), 295-301

16

ACCEPTED MANUSCRIPT

Quinlan, P., Lane, J., & Aspinall, L. (1997). Effects of hot tea, coffee and water ingestion on physiological responses and mood: the role of caffeine, water and beverage type. Psychopharmacology, 134(2), 164-173.

RI PT

Racotta, I. S., Leblanc, J., & Richard, D. (1994). The effect of caffeine on food intake in rats: involvement of corticotropin-releasing factor and the sympatho-adrenal system. Pharmacology, Biochemistry, and Behavior, 48(4), 887-892. Retzbach, E. P., Dholakia, P. H., & Duncan-Vaidya, E. (2014). The effect of daily caffeine exposure on lever-pressing for sucrose and c-fos expression in the nucleus accumbens in the rat. Physiology & Behavior, 135, 1-6.

SC

Ribeiro, J.A., & Sebastiao, A.M. (2010). Caffeine and adenosine. J Alzhermers Dis. 20 Suppl 1, S3-15.

M AN U

Rice, A. M., Fain, J. N., & Rivkees, S. A. (2000). A1 adenosine receptor activation increases adipocyte leptin secretion. Endocrinology, 141(4), 1442-1445. Schubert, M. M., Grant, G., Horner, K., King, N., Leveritt, M., Sabapathy, S., Desbrow, B. (2014). Coffee for morning hunger pangs. An examination of coffee and caffeine on appetite, gastric emptying, and energy intake. Appetite, 83, 317-326.

TE D

Solinas, M., Ferre, S., You, Z. B., Karcz-Kubicha, M., Popoli, P., & Goldberg, S. R. (2002). Caffeine induces dopamine and glutamate release in the shell of the nucleus accumbens. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 22(15), 6321-6324. Sternson, S.M., Atasoy, D. (2014). Agouti-related protein neuron circuits that regulate appetite. Neuroendrocrinology, 2-3(100), 95-102.

EP

Williams, K. W. and J. K. Elmquist. (2012). From neuroanatomy to behavior: central integration of peripheral signals regulating feeding behavior. Nat Neurosci 15(10), 1350-1355. Yang, Y., Atasoy, D., Su, H., & Sternson, S. (2011). Hunger States Switch a Flip-Flop Memory Circuit via a Synaptic AMPK-Dependent Positive Feedback Loop. Cell, 146(6), 9921003.

AC C

433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472

Yin, Y., Zhang, C., Wang, J., Hou, J., Yang, X., & Qin, J. (2015). Chronic caffeine treatment enhances the resilience to social defeat stress in mice. Food & Function, 6(2), 479-491.

473 474

17

ACCEPTED MANUSCRIPT

n.s.

A

**

S a lin e

***

6 m g /k g C a ffe in e 1 2 m g /k g C a ffe in e

RI PT

0.4

SC

0.2

2 4 m g /k g C a ffe in e

M AN U

2 hr food intake (g)

0.6

0.0

n.s.

n.s.

0.8

0.4

EP

***

S a lin e 6 m g /k g C a ffe in e 1 2 m g /k g C a ffe in e 2 4 m g /k g C a ffe in e

AC C

5 hr food intake (g)

1.2

TE D

B

0.0

Sweeney et al., Figure 1

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 1: Acute caffeine administration facilitates feeding. Feeding behavior assays indicating that administration of caffeine increases feeding at both two and five hours following i.p. injections. Injections of caffeine (6 mg/kg-24 mg/kg; dissolved in 0.9% saline) or saline were administered and subsequent levels of food intake were calculated at ascending time-points, as shown. (A) Caffeine shows a dose-dependent effect on appetite with 6 mg/kg exerting the most significant orexigenic effects on feeding after two hours. Higher doses were less effective at increasing feeding although 12 mg/kg also significantly increased food intake. (B) Five hours post injection, only 6 mg/kg administration of caffeine significantly increased food intake. n=8 mice per group for all figures. Data represents mean +/- SEM. ** p<.01; ***p<0.001; ns (not significant).

20

Time in center (s)

**

80 60

40

SC

**

TE D

40

60

Caffeine

M AN U

80

20

20

Saline

Caffeine

**

400

200

0

Saline

EP

0

G

Distance travelled in corner (m)

Entries to center (#)

Saline

Caffeine

D

600 Time in corner (s)

0.05

0

100

F

0.1

0

0

n.s.

Caffeine

Saline

Caffeine

E

Distance travelled in center (m)

40

0.15

RI PT

Average speed (m/s)

60

Saline

C

ACCEPTED MANUSCRIPT

B

n.s.

80

AC C

Total distance travelled (m)

A

80

** 60 40 20 0

Saline

Caffeine

n.s.

80 60 40 20 0

Saline

Caffeine

Sweeney et al., Figure 2

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 2: Acute caffeine administration decreases anxiety related behaviors. (A and B) Open field behavioral analysis indicates that caffeine administration (20 mg/kg) does not significantly affect total distance traveled (A) or average speed (B) in the open field. (C-G) Open field behavioral experiments showing that acute administration of caffeine (20 mg/kg) increases entries to the center (C), time in the center (D), and distance traveled in the center (E) of the open field. (F) Time in the corner of the open field is lower in mice injected with caffeine compared to mice injected with vehicle saline. (G) No significant differences were detected between saline and caffeine conditions in distance traveled in the corner of the open field. Caffeine or saline was administered 10 minutes prior to open field behavioral experiments. All experiments were repeated twice on consecutive days in counterbalanced order. n=8 mice per group; paired Student’s t-test; Data represents mean +/- SEM. **p<0.01; ns (not significant).

ACCEPTED MANUSCRIPT

Saline 20 mg/kg Caffeine Withdrawal

20 Body weight (g)

n.s. 4

4

5

6 7 Experimental Day

8

C 0.4

***

** 0.3

AC C

0.2

EP

*

TE D

3

0.1

0.0 3

4

5

6

Experimental Day

7

10

M AN U

2

2 hr food intake (g)

15

n.s.

SC

3

3

4

5

6 9 8 7 Experimental Day

10

D

Relative distance travelled in center (%)

24 hr food intake (g)

5

Saline 20 mg/kg Caffeine Withdrawal

B

RI PT

A

20

No withdrawal (Saline) Withdrawal (Caffeine)

15 10

5 0 7

9

8 Experimental Day

Sweeney et al., Figure 3

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Figure 3: Chronic caffeine administration does not affect food intake or body weight. (A and B) Average 24 hour food intake and body weight during a long term feeding paradigm. Mice were administered i.p. injections of saline on days 1 and 2, either caffeine (20 mg/kg) or saline on days 3 to 6, and saline on days 7 through 10. No significant changes in daily food intake (A) or body weight (B) were detected during caffeine administration or during caffeine withdrawal (2-way ANOVA). (C) Daily food intake two hours following injections of saline or caffeine. Daily i.p. injections of caffeine increased feeding two hours following caffeine administration. (D) Open field behavioral assays during caffeine withdrawal. Open field experiments were conducted on three consecutive days during withdrawal from caffeine administration in animals previously injected with caffeine or vehicle saline. No significant differences in anxiety behavior were detected between caffeine and saline treated animals in percent distance traveled in the center of the open field on any of the caffeine withdrawal days. Data represents mean +/- SEM. n= 6 mice per group *p<0.05; **p<0.01; ***p<0.001; ns (not significant).