Content uploaded by Fabio Arturo Iannotti
Author content
All content in this area was uploaded by Fabio Arturo Iannotti on Mar 05, 2019
Content may be subject to copyright.
Metadata of the article that will be visualized in OnlineFirst
1 Article T itle Exercise training and high-fat diet elicit endocannabinoid
system modifications in the rat hypothalamus and hippocampus
2 Article Sub- T i tl e
3 Article Copyright -
Year
Univ ersity of Nav arra 2017
(This will be the copyright line in the final PDF)
4 Journal Nam e Journal of Physiology and Biochemi stry
5
Correspondi ng
Author
Family Name Gamelin
6 Particl e
7 Given Name Franç ois-Xa v ier
8 Suffix
9 Organizati on Equipe dʼAccuei l 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
10 Division
11 Address Santé. Eurasport, 413 rue Eugène Avi née, Loos
59120
12 Organi zation F-590 00
13 Division Univ Lille Nord de France
14 Address Lill e
15 e-mai l francoi s-xavier.gamel in-2@univ-li lle2.fr
16
Author
Family Name Aucouturier
17 Particle
18 Gi ven Name Julien
19 Suffi x
20 Organi zation F-590 00
21 Division Univ Lille Nord de France
22 Address Lill e
23 Organi zation Equi pe dʼAccueil 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
24 Division
25 Address Santé. Eurasport, 413 rue Eugène Avi née, Loos
59120
26 e-mai l
27
Author
Family Name Iannotti
28 Particle
29 Gi ven Name Fabio Arturo
30 Suffi x
31 Organi zation Institute of Bi omolecul ar Chemistry
32 Division CNR, Endocannabinoid Research Group
33 Address Pozzuol i I-80078
34 e-mai l
35
Author
Family Name Piscitelli
36 Particle
37 Gi ven Name Fabiana
38 Suffi x
39 Organi zation Institute of Bi omolecul ar Chemistry
40 Division CNR, Endocannabinoid Research Group
41 Address Pozzuol i I-80078
42 e-mai l
43
Author
Family Name Mazzarella
44 Particle
45 Gi ven Name Enrico
46 Suffi x
47 Organi zation Institute of Bi omolecul ar Chemistry
48 Division CNR, Endocannabinoid Research Group
49 Address Pozzuol i I-80078
50 e-mai l
51
Author
Family Name Av eta
52 Particle
53 Gi ven Name Teresa
54 Suffi x
55 Organi zation Institute of Bi omolecul ar Chemistry
56 Division CNR, Endocannabinoid Research Group
57 Address Pozzuol i I-80078
58 e-mai l
59
Author
Family Name Leriche
60 Particle
61 Gi ven Name Melissa
62 Suffi x
63 Organi zation F-590 00
64 Division Univ Lille Nord de France
65 Address Lill e
66 Organi zation Equi pe dʼAccueil 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
67 Division
68 Address Santé. Eurasport, 413 rue Eugène Avi née, Loos
59120
69 e-mai l
70
Author
Family Name Dupont
71 Particle
72 Gi ven Name E rw an
73 Suffi x
74 Organi zation F-590 00
75 Division Univ Lille Nord de France
76 Address Lill e
77 Organi zation Equi pe dʼAccueil 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
78 Division
79 Address Santé. Eurasport, 413 rue Eugène Avi née, Loos
59120
80 e-mai l
81
Author
Family Name Cienie w ski -Berna rd
82 Particle
83 Gi ven Name Caroline
84 Suffi x
85 Organi zation F-590 00
86 Division Univ Lille Nord de France
87 Address Lill e
88 Organi zation Equi pe dʼAccueil 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
89 Division
90 Address Santé. Eurasport, 413 rue Eugène Avi née, Loos
59120
91 e-mai l
92
Author
Family Name Leclair
93 Particle
94 Gi ven Name E rw an
95 Suffi x
96 Organi zation F-590 00
97 Division Univ Lille Nord de France
98 Address Lill e
99 Organi zation Equi pe dʼAccueil 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
100 Divi sion
101 Address Santé. Eurasport, 413 rue Eugène Avinée, Loos
59120
102 e-mail
103
Author
Family Name Bastide
104 Parti cle
105 Given Name Bruno
106 Suffix
107 Organizati on F-59 000
108 Divi sion Univ Li l le Nord de France
109 Address Li lle
110 Organizati on Equipe dʼAccuei l 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
111 Divi sion
112 Address Santé. Eurasport, 413 rue Eugène Avinée, Loos
59120
113 e-mail
114
Author
Family Name Di Marzo
115 Parti cle
116 Given Name Vincenzo
117 Suffix
118 Organizati on Institute of Bi om ol ecular Chemistry
119 Divi sion CNR, Endocanna bi noid Research Group
120 Address Pozzuoli I-80078
121 e-mail
122
Author
Family Name Heyman
123 Parti cle
124 Given Name Elsa
125 Suffix
126 Organizati on F-59 000
127 Divi sion Univ Li l le Nord de France
128 Address Li lle
129 Organizati on Equipe dʼAccuei l 7369, URePSSS-Unité de
Recherche Pluridisciplinai re Sport Santé Société-
Equipe Acti vi té Physique, Muscle
130 Divi sion
131 Address Santé. Eurasport, 413 rue Eugène Avinée, Loos
59120
132 e-mail
133
Schedule
Recei ved 6 June 2016
134 Revised
135 Accepted 23 February 20 17
136 Abstract The purpose of the present study was to examine the effect of
chronic exercise on the hypothalamus and hippocam pus levels of
the endocannabinoids (eCBs) anandam ide (AEA) and
2-arachidon oyl glycerol (2-AG) and of two AEA congeners and on
the expression of genes codi ng for CB1, CB2 receptors (Cnr1 and
Cnr2, respecti vely), and the enzym es responsi ble for eCB
bi osynthe sis and degradati on, in rats fed with a standard or high-fat
di et. Mal e Wistar rats (n = 28) were placed on a 12-week hi gh-fat
(HFD) or standard di et period, fol l owed by 12 weeks of exercise
traini ng for half of each group. T issue level s of eCBs and rel ated
lipids were measured by liq ui d chrom atography mass spectrom etry,
and expression of genes coding for CB1 and CB2 receptors and
eCB metabol ic enzyme s was measured by quantitati ve real -time
polym erase chain reaction (qPCR). HFD in duced a si gnificant
increase in 2-AG (p < 0.01) in hypothalamus. High-fat di et paired
with exercise trai ni ng had no effect on AEA, 2-AG, and AEA
congen er l eve ls in the hypothalamus and hi ppocampus. Cnr1
expression l evels were significantly increased in the hippocam pus
in response to HFD, exercise, and the combi nati on of both
(p < 0.05). Our results indi cate that eCB si gnali ng i n the CNS is
sensiti ve to diet and/or exerci se.
137 Keywords
separated by ' - '
2-arachidon oyl glycerol - Anandam ide - CB1 receptor - CB2
receptor - Hippocam pus - Hypothalamus
138 Foot note
information
François-Xavier Gameli n, Vi ncenzo Di Marzo, and Elsa Heyman
share the senior authorship.
UNCORRECTEDPROOF
1
2
3ORIGINAL ARTICLE
4Exercise training and high-fat diet elicit endocannabinoid system
5modifications in the rat hypothalamus and hippocampus
6François-Xavier Gamelin
1,2
&Julien Aucouturier
1,2
&Fabio Arturo Iannotti
3
&
7Fabiana Piscitelli
3
&Enrico Mazzarella
3
&Tere s a Av eta
3
&Melissa Leriche
1,2
&
8Erwan Dupont
1,2
&Caroline Cieniewski-Bernard
1,2
&Erwan Leclair
1,2
&
9Bruno Bastide
1,2
&Vincenzo Di Marzo
3
&Elsa Heyman
1,2
10
11 Received: 6 June 2016 /Accepted: 23 February 2017
12 #University of Navarra 2017
13 Abstract The purpose of the present study was to examine
14 the effect of chronic exercise on the hypothalamus and hip-
15 pocampus levels of the endocannabinoids (eCBs) ananda-
16 mide (AEA) and 2-arachidonoylglycerol (2-AG) and of
17 two AEA congeners and on the expression of genes coding
18 for CB1, CB2 receptors (Cnr1 and Cnr2,respectively),and
19 the enzymes responsible for eCB biosynthesis and degrada-
20 tion, in rats fed with a standard or high-fat diet. Male Wistar
21 rats (n= 28) were placed on a 12-week high-fat (HFD) or
22 standard diet period, followed by 12 weeks of exercise train-
23 ing for half of each group. Tissue levels of eCBs and related
24 lipids were measured by liquid chromatography mass spec-
25 trometry, and expression of genes coding for CB1 and CB2
26 receptors and eCB metabolic enzymes was measured by
27 quantitative real-time polymerase chain reaction (qPCR).
28 HFD induced a significant increase in 2-AG (p<0.01)in
29 hypothalamus. High-fat diet paired with exercise training
30 had no effect on AEA, 2-AG, and AEA congener levels in
31 the hypothalamus and hippocampus. Cnr1 expression levels
32 were significantly increased in the hippocampus in response
33 to HFD, exercise, and the combination of both (p<0.05).
34Our results indicate that eCB signaling in the CNS is sensi-
35tive to diet and/or exercise.
36Keywords 2-arachidonoylglycerol .Anandamide .CB1
37receptor .CB2 receptor .Hippocampus .Hypothalamus
38Introduction
39Strong evidence supports a role of the endocannabinoid sys-
40tem (ECS) in the impaired regulation of food intake and ener-
41gy metabolism that contribute to the development of obesity
42[15,43]. The ECS is a complex endogenous signaling system
43comprising 7-transmembrane domain receptors (cannabinoid
44type 1 (CB1) and type 2 (CB2) receptors), their endogenous
45lipid-derived ligands (the endocannabinoids, eCBs), and en-
46zymes for eCB biosynthesis and degradation. The two most
47studied eCBs are N-arachidonoylethanolamine (AEA), also
48known as anandamide, and 2-arachidonoylglycerol (2-AG).
49eCBs are not stored in cells but are synthesized on demand
50from arachidonic acid containing phospholipid precursors
51through enzyme activation by multiple pathways in the cell
52membrane of most mammalian cells such as neurons, adipo-
53cytes, and skeletal muscle cells, possibly in response to ele-
54vated levels of intracellular calcium [6], membrane depolari-
55zation, and/or receptor stimulation [36].
56In the central nervous system (CNS), the ECS interacts
57with the different systems involved in control of food intake
58and energy expenditure [15]. In rodents, AEA injection in
59hypothalamus elicits increased feeding via CB1 activation
60[27] by modulating probably the expression and the action
61of orexigenic and anorectic mediators, such as the neuropep-
62tide melanin-concentrating hormone or the
63corticotropin-releasing hormone [9,28]. In addition to the
64regulation of food intake, the hypothalamic ECS is also
François-Xavier Gamelin, Vincenzo Di Marzo, and Elsa Heyman share
the senior authorship.
*François-Xavier Gamelin
francois-xavier.gamelin-2@univ-lille2.fr
1
Univ Lille Nord de France, F-59000, Lille, France
2
Equipe d Accueil 7369, URePSSS-Unité de Recherche
Pluridisciplinaire Sport Santé Société-Equipe Activité Physique,
Muscle, Santé. Eurasport, 413 rue Eugène Avinée,
59120 Loos, France
3
CNR, Endocannabinoid Research Group, Institute of Biomolecular
Chemistry, I-80078 Pozzuoli, Italy
J Physiol Biochem
DOI 10.1007/s13105-017-0557-1
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
65 involved in energy expenditure regulation by acting on ther-
66 mogenesis [29].
67 The ECS also plays a role in hedonic aspect of food intake
68 by modulating brain area activities of the reward system such
69 as the nucleus accumbens, ventral tegmental area, or hippo-
70 campus [31][34]. Kirkham et al. [30] observed that the injec-
71 tion of 2-AG into the nucleus accumbens shell produced a
72 short-term stimulatory action on feeding behavior in
73 free-feeding rats. The ECS seems to facilitate the mesolimbic
74 dopamine signaling that stimulates appetite, as Verty et al. [49]
75 observed that a dopamine D1 receptor antagonist attenuates
76 feeding induced by a CB1 agonist.
77 Previous studies in obese rodents suggest the presence
78 of ECS dysregulation in the hypothalamus and hippo-
79 campus [14,34]. Di Marzo et al. [14] found that genet-
80 ically obese rats and mice with disrupted leptin signaling
81 (Zucker fa/fa rats and db/db mice), as well as mice lack-
82 ing leptin (ob/ob mice), have higher hypothalamic
83 endocannabinoid levels compared with wild-type ani-
84 mals. Similar findings were also reported in the hippo-
85 campus of mice with diet-induced obesity [34]. These
86 central ECS alterations, together with peripheral eCB dis-
87 turbances in the adipose tissue and skeletal muscle, may
88 participate in excessive and/or ectopic fat accumulation
89 and related metabolic disorders [12,35].
90 In human obesity, ECS dysregulation is supported by
91 the observation of changes in AEA and/or 2-AG levels
92 intheplasmaandadiposetissue[3,18]. The ECS rep-
93 resents a primary target for the treatment of abdominal
94 obesity and associated metabolic changes, whether its
95 dysregulation is a consequence or a cause of obesity. It
96 is noteworthy that before being withdrawn from the mar-
97 ket due to psychiatric side effects such as anxiety and
98 depression [8], CB1 antagonists were clearly shown to
99 be effective at reducing body weight and waist circum-
100 ference in obese subjects [38].
101 Exercise is a recognized treatment of obesity [37]. One
102 bout of exercise triggers eCB signaling by elevating AEA
103 plasma levels [45][23,40]. However, there is some evidence
104 suggesting that a more long-lasting healthy lifestyle approach
105 may be effective to reverse ECS dysregulation [3,17] and may
106 represent a safe alternative to the pharmacological approach.
107 A 1-year lifestyle modification program including physical
108 activity induced a significant decrease in fasting plasma
109 AEA (−7.1%) and, most importantly, 2-AG (−62.3%) levels
110 in viscerally obese men [17]. In subcutaneous and visceral
111 adipose tissues, chronic exercise limits CB1 gene expression
112 increase induced by high-fat diet in rodents [51]. Thus, chron-
113 ic exercise may counteract ECS dysregulation in these tissues.
114 In the CNS, the ECS mediates exercise-induced hippocam-
115 pus plasticity [25] and reward [20]. However, it remains un-
116 known whether exercise has beneficial effects on the ECS in
117 the CNS of obese animals, and more specifically the
118hypothalamus and hippocampus, involved in the control of
119energy balance [32,48].
120This study aimed therefore at identifying potential changes
121elicited by exercise training in the brain ECS of rats on a
122high-fat diet, compared to rats on a standard diet. With this
123purpose, we determined the hypothalamic and hippocampal
124levels of AEA, 2-AG, and two AEA congeners
125N-oleylethanolamine (OEA) and N-palmitoyl-ethanolamine
126(PEA); the expression of genes encoding for eCB receptors
127(CB1, CB2) and enzymes involved in eCB anabolic
128(α/β-hydrolase 4, glycerophosphodiesterase 1,
129N-acylphosphatidylethanolamine-phospholipase D, protein
130tyrosine phosphatase N22 (respectively ABHD4, GDE-1,
131NAPE-PLD, and PTNP-22) for AEA, OEA, and PEA; diac-
132ylglycerol lipase α(DAGL-α) and diacylglycerol lipase β
133(DAGL-β), for 2-AG) and catabolic (fatty acid amide hydro-
134lase (FAAH), for AEA, OEA, and PEA; α/β-hydrolase 6
135(ABHD6), α/β-hydrolase 12 (ABHD12), and
136monoacylglycerol lipase (MAGL), for 2-AG) pathways [16].
137We also determined the expression of the transient receptor
138potential vanilloid type-1 (TRPV1) channel, which is activat-
139ed by eCBs and AEA congeners and is an ionotropic receptor
140for eCBs [52].
141Materials and methods
142Animals and general procedure
143General procedures were already described previously
144[21]. Briefly, 28 male Wistar rats (3 weeks old) were
145housed in groups of two or three per cage. After 1 week
146of acclimatization, rats were randomly divided into two
147groups and fed ad libitum either with a standard diet
148(energy equivalent: 2.90 kcal g
−1
) or a high-fat diet (en-
149ergy equivalent 5.05 kcal g
−1
)during24weeks.After
15012 weeks, half of the rats were submitted to 12 weeks
151of exercise training (Ctl + training and HFD + training).
152The second half of the rats remained untrained for
15312 weeks (Ctl and HFD groups for rats on standard
154and high-fat diets, respectively). Before and after the
155training period, Ctl + training and HFD + training per-
156formed a maximal aerobic velocity (MAV) test on the
157treadmill. Five days before sacrifice, all rats were sub-
158jected to an oral glucose tolerance test (OGTT). At the
159end of the exercise training period, blood samples were
160collected by cardiac puncture, after which all rats were
161euthanized and the hypothalamus and hippocampus
162removed.
163All procedures described were approved by the
164Agricultural and Forest Ministry and the National Education
165Ministry (Veterinary service of health and animal protection)
Gamelin et al.
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
166 and were in accordance with the European Union Directive of
167 22 September 2010 (2010/63/UE).
168 High-fat diet
169 Rats were fed with two different types of diet during the
170 24 weeks of the experimentation:
171 –A high-fat diet (Purified Diet 231 HF, Safe, Augy, France)
172 with an energy equivalent of 5.05 kcal g
−1
and containing
173 26.9% of proteins, 39.7% of lipids, and 10.1% of
174 carbohydrates
175 –A standard diet with an energy equivalent of
176 2.90 kcal g
−1
. It contained 16% of proteins, 3% of lipids,
177 60% of carbohydrate, and 21% of other components (fi-
178 ber, mineral, humidity).
179 Fatty acid compositions of the two diets are described in
180 Tab le 1. Food and caloric intake by each rat and their weight
181 gain were estimated two times per week during the
182 experimentation.
183 Maximal aerobic velocity test
184 Animals of in the HFD + training and Ctl + training were
185 familiarized with treadmill running (L810, Bioseb. France)
186 for 10 min for 5 days at a velocity of 20 cm s
−1
and a 0° slope.
187 Electric shocks (intensity <1.2 mA) were used to motivate the
188 rat to run. After the familiarization period, Ctl + training and
189 HFD + training groups performed a graded exercise test to
190 voluntary exhaustion. The test started at 20 cm s
−1
for
191 5 min, followed by speed increment of 3 cm s
−1
every 3 min
192 until the animal could no longer keep up with the treadmill
193 speed. Exhaustion was reached when animal sat longer than
194 10 s on electric shock grid. MAV was defined as the velocity
195 of the last 3-min stage completed. The same protocol was
196 repeated 1 week before rats sacrifice to determine the change
197 in MAV with exercise training (2 days before the OGTT).
198 Exercise training
199 The day after the baseline MAV test, Ctl + training and HFD +
200 training groups started the 12-week exercise training period
201 that consisted of treadmill running for 1 h/day, 5 days/week at
202 an intensity set between 70 and 80% of the MAV. The inten-
203 sity was increased by 1 cm s
−1
every week to take into the
204 adaptations to exercise training. Animal exercised at the same
205 hour of the day at the end of the room dark cycle (7:30 a.m.).
206 Control groups were in the same room during the training
207 session and handled in the same way to induce a similar stress
208 level. Three days before the sacrifice, exercise training was
209 stopped to avoid the confounding fatigue or stress effect of
210 acute exercise.
211Oral glucose tolerance test
212Five days before sacrifice, the animals were fasted over-
213night. Basal blood glucose level, defined as T0, was
214determined using an automatic glucometer (Accu-Chek
215Performa; Roche Diagnostics) before oral administration
216(4 ml kg
−1
of body weight) of a D-glucose solution
217(50%). Tail vein blood glucose was then measured at
21830, 60, 90, and 120 min after the administration. Total
219area under the curve (AUC) was calculated using the
220trapezoidal method [39].
t1:1Table 1 Fatty acid compositions of the standard (Ctl) and the high fat
diet (HFD)
t1:2Ctl HFD
t1:3Total fat (g/kg) 27.50 395.47
t1:4Total saturated fat (g/kg) 6.34 140.36
t1:5C10:0 –0.32
t1:6C12:0 0.03 0.32
t1:7C14:0 0.17 4.19
t1:8C15:0 0.03 0.32
t1:9C16:0 5.30 86.93
t1:10C17:0 0.03 1.29
t1:11C18:0 0.58 45.97
t1:12C20:0 0.10 0.94
t1:13C22:0 0.06 0.07
t1:14C24:0 0.06 –
t1:15Total monounsaturated fat (g/kg) 5.61 171.18
t1:16C16:1 0.19 8.14
t1:17C17:1 0.03 0.65
t1:18C18:1 5.09 158.96
t1:19C19:1 –0.00
t1:20C20:1 0.30 2.80
t1:21C22:1 –0.65
t1:22Total polyunsaturated fat (g/kg) 15.57 86.54
t1:23C18:2 13.70 78.14
t1:24C18:3 1.13 3.24
t1:25C18:4 –2.58
t1:26C20:2 0.03 1.29
t1:27C20:3 –0.65
t1:28C20:4 0.06 –
t1:29C20:5 0.14 –
t1:30C22:1 0.19 –
t1:31C22:4 –0.32
t1:32C22:5 0.06 0.32
t1:33C22:6 0.22 –
t1:34C24:1 0.06 –
t1:35Tot al ω3 fatty acids (g/kg) 1.57 3.91
t1:36Tot al ω6 fatty acids (g/kg) 13.70 75.35
Q1
Endocannabinoid system in brain, diet, and exercise effects
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
221 Sample collection
222 The day before the end of the experiment, rats were fasted in
223 order to obtain the same nutritional state for each. Animals
224 were anesthetized with pentobarbital sodium (60 mg kg
−1
of
225 body weight, i.p.), and blood samples were collected by car-
226 diac puncture. Samples were drawn directly into pre-cooled
227 5-mL EDTA tubes. EDTA blood was immediately centrifuged
228 (less than 5 min after sampling), and plasma was removed and
229 frozen (−80 °C) until analysis. Then, the rats were sacrificed
230 by decapitation. The head was immediately surrounded with
231 ice, and the hypothalamus and hippocampus were quickly
232 removed, weighed, and immediately frozen in liquid nitrogen.
233 Then, they were stored at −80 °C until analyses.
234 Plasma analyses
235 Fasting levels of glucose and insulin were determined in plas-
236 ma. The content of glucose in plasma was measured using a
237 commercially available colorimetric assay kit (Cayman
238 Chemical Company, USA). Plasma insulin was determined
239 using a commercially available rat insulin enzyme immuno-
240 assay kit (SPI-BIO, France).
241 Measurements of endocannabinoids
242 The extraction, purification, and quantification of EC from
243 tissues have been performed as previously described [26].
244 Briefly, the tissues were dounce-homogenized and extracted
245 with chloroform/methanol/Tris–HCl50mmoll
−1
pH 7.5
246 (2:1:1, vol/vol) containing internal standards ([
2
H]
8
AEA;
247 [
2
H]
5
2-AG, [
2
H]
5
PEA, and [
2
H]
4
OEA, 5 pmol each). The
248 lipid-containing organic phase was dried down, weighed, and
249 pre-purified by open-bed chromatography on silica gel.
250 Fractions were obtained by eluting the column with 99:1,
251 90:10, and 50:50 (v/v) chloroform/methanol. The 90:10 frac-
252 tion was used for AEA, 2-AG, PEA, and OEA quantification
253 by liquid chromatography–atmospheric pressure chemical
254 ionization–mass spectrometry by using a Shimadzu
255 high-performance liquid chromatography apparatus
256 (LC-10ADVP) coupled to a Shimadzu (LCMS-2020) quadru-
257 ple mass spectrometry via a Shimadzu atmospheric pressure
258 chemical ionization interface as previously described [26].
259 The amounts of analytes in tissues, quantified by isotope di-
260 lution with the above-mentioned deuterated standards, were
261 expressed as picomole per gram or milligram of wet tissue
262 weight.
263 RNA purification and qPCR
264 Total RNA was isolated from native tissues by use of the
265 TRI-Reagent (Sigma-Aldrich, Milan, Italy), reacted with
266 DNase-I (1 U/ml; Sigma-Aldrich) for 15 min at room
267temperature, and followed by spectrophotometric quantifica-
268tion. Final preparation of RNA was considered DNA- and
269protein-free if the ratio between readings at 260/280 nm was
270≥1.7. Isolated mRNA was reverse transcribed by use of
271SuperScript III Reverse Transcriptase (Life Technologies,
272Monza (MI), Italy). The quantitative real-time PCR was carried
273out in CFX384 real-time PCR detection system (Bio-Rad,
274Segrate (MI), Italy), with specific primers [26], by the use of
275SYBR Green master mix kit (Bio-Rad, Segrate (MI)) (see
276Tab le 2for primer sequence). Samples were amplified simul-
277taneously in quadruplicate in one-assay run with a
278non-template control blank for each primer pair to control for
279contamination or primer-dimer formation, and the cycle thresh-
280old (ct) value for each experimental group was determined. The
281housekeeping gene (the hypoxanthine-guanine
282phosphoribosyltransferase, hprt) was used as an internal control
283to normalize the ct values, using the 2
−ΔCt
formula; differences
284in mRNA content between groups were as expressed as
2852
−ΔΔCt
.
286Statistical analyses
287Data are shown as means ± SE. Normal Gaussian distribution
288of the data was verified by the Shapiro-Wilk test. Repeated
289measure analysis of variance (ANOVA) was used to evaluate
290the evolution of weight and accumulated caloric intake during
291the first 12 weeks of the experiment, the evolutions of weight
292and MAV during the exercise-training period and mean caloric
293intake of the next to last week of each period, and the glycemia
294during the OGTT. Multiple comparisons were made with the
295Newman-Keul post hoc test. A two-way ANOVA was used to
296evaluate the effects of diet, exercise training, and the diet–
297exercise interaction on metabolic parameters (fasting glucose,
298insulin, AUC during OGTT) and on eCBs and congeners
299tissue levels. Multiple comparisons were made with the
300Bonferroni post hoc test if significant main effects or interac-
301tion were observed with ANOVA. Concerning gene expres-
302sion data, the Ctl group was compared with other groups by
303using Student’sttest. Statistical significance was set at
304p= 0.05 level for all analysis. All calculations were made with
305Statistica 6.0 (Statsoft, Tulsa, USA).
306Results
307Effect of diet and/or exercise on body mass, caloric intake,
308maximal aerobic velocity, basal glucose and insulin levels,
309and glucose tolerance
310Figure 1shows body weight gain over the diet and exercise
311training period. Over the first 12 weeks, body mass increased
312with time and this is all the more in the case of the high-fat diet
313(Table 3). From the 10th week, rats fed with the high-fat diet
Gamelin et al.
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
314 were significantly heavier than were the rats fed with the con-
315 trol diet (p< 0.01; Fig. 1). During the exercise period, rats fed
316 with the high-fat diet continued to gain more weight with time
317 while exercise training slowed down the time-induced body
318 gain, especially for the HFD + training group (Table 3).
319 Accumulated caloric intake is described in Fig. 2.Overthe
320 first and second periods of the protocol, we observed that rats
321 fed on the high-fat diet accumulated more and more rapidly
322 calorie than did rats fed on the standard diet (Table 3). From
323 the 8th week, rats fed with the high-fat diet accumulated sig-
324 nificantly more calorie than did rats fed with the control diet
325 (p< 0.01; Fig. 2). However, over the second period, the accu-
326 mulated caloric intake was slowed down by exercise training
327 (Table 3).
328MAV was measured only in the exercise-trained groups to
329avoid familiarization in the Ctl and HFD groups that could
330affect the results. Two-way ANOVA for MAV revealed sig-
331nificant effects of diet and time but no significant interaction
332between both factors (Table 4). There were no significant dif-
333ferences for fasting plasma insulin levels between groups.
334Fasting plasma glucose concentration was increased by the
335high-fat diet but this increase was strongly reduced by exercise
336training (Table 4). Between-group comparison indicated that
337in the HFD group, glycemia was significantly higher than in
338the other groups (Table 4). Glucose AUC during the OGTT
339performed 1 week before the animal sacrifice was also signif-
340icantly increased by the high-fat diet (Table 4) but without
341significant protecting effect from exercise training.
Fig. 1 Evolution of body weight during the 24 weeks of the protocol in
control (ctl), high-fat diet (HFD), control with chronic exercise (ctl +
training), and high-fat diet with chronic exercise (HFD + training)
groups. The dashed line corresponds to the introduction of exercise in
ctl + trainingand HFD + training groups. *Significant difference between
rats fed with high-fat and standard diet during the first period of the
protocol: p<0.01
t2:1Tab l e 2 Primers sequence used in qPCR analysis
t2:2Gene Forward sequence (5′-3′) Reverse sequence (5′-3′) Enter accession number Product length (bp)
t2:3abdh
12
CAGGCGTGCGGTCGAAACCA TCAAGCTGCAGTCGGCGTCC NM_001024314.1 189
t2:4abdh
4
TCTGGCGTCAAGCGGAGGGA ACGCCACCCCCAAAGCCATG NM_001108866.1 299
t2:5abdh
6
AGCGTCTGCTCCCATCCCCA TGGCTTGCCAGTGGCGTGAA NM_001007680.1 255
t2:6cnr-1 CTGAGGGTTCCCTCCCGGCA TGCTGGGACCAACGGGGAGT NM_012784.4 285
t2:7cnr-2 GCGGCTAGACGTGAGGTTGGC TCCTTCAGGACCAAGAGTCTCAGCCT NM_020543.4 335
t2:8daglαGGCCGCACCTTCGTCAAGCT ATCCAGCACCGCATTGCGCT NM_001005886.1 380
t2:9daglβAGACCCGGGTGCAATGCTGC GCCCTGGTGTGTGGGTCACG NM_001107120.1 212
t2:10 faah GGCAGAGCCACAGGGGCTATCA TGGGGCTACAGTGCACAGCG NM_024132.3 349
t2:11 gde-1 GCAGCCCCTTCAACGCCTGT GATGGCCGCCAGCGTGTTCT NM_019580.4 172
t2:12 magl CGGAACAAGTCGGAGGTTGA TGTCCTGACTCGGGGATGAT NM_138502.2 220
t2:13 nape-pld AGGCTGGCCTACGAATCACGT ATGGTACACGGGGGACGGCG NM_199381.1 150
t2:14 ptpn-22 TGGTCGTGGGAGAGCCGCTT GGGCCACTTTTTGCGCCTGC NM_001106460.1 263
t2:15 trpv1 AGACATCAGCGCCCGGGACT CCAGCTTCAGCGTGGGGTGG NM_031982.1 151
Q1
Endocannabinoid system in brain, diet, and exercise effects
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
342 Regarding glucose kinetics during the OGTT (Fig. 3), the
343 ANOVA revealed a significant effect of diet (F
(1,
344
26)
=85.23,p< 0.006) and interactions between time and diet
345 (F
(4, 104)
=3.97,p< 0.008) as well time, training, and diet (F
(4,
346
96)
=6.59,p< 0.0001). The HFD group showed higher level
347of blood glucose than did other groups at30 min (p<0.02).At
34890 and 120 min, blood glucose levels for the HFD groups
349were higher than for the Ctl + training group (p< 0.05 for all).
t3:1Tab l e 3 F-statistic and its
associated degrees of freedom
and pvalue of analyses of
variance used to compare the
effects of diet and/or exercise on
body weight; food intake;
maximal aerobic velocity (MAV);
basal glucose and insulin levels;
and oral glucose tolerance test in
control, high-fat diet, control with
chronic exercise, and high-fat diet
with chronic exercise groups
t3:2Weight (g) Main effects by ANOVA Fvalues p
t3:3During diet period Diet
Time
Time × Diet
F
(1, 26)
=53.80
F
(11, 297)
= 1714.96
F
(11, 286)
=30.56
0.00001
0.00001
0.00001
t3:4During exercise period Diet
Ex
Diet × Ex
Time
Time × Diet
Time × Ex
Time × Diet × Ex
F
(1, 26)
=38.91
F
(1, 26)
=0.60
F
(1, 24)
=0.59
F
(12, 288)
= 392.85
F
(12, 312)
=5.34
F
(12, 312)
=16.72
F
(12, 288)
=3.48
0.00001
NS
NS
0.00001
0.00001
0.00001
0.0001
t3:5Accumulated caloric intake (kcal)
t3:6During diet period Diet
Time
Time × Diet
F
(1, 26)
=63.94
F
(11, 297)
= 9136.67
F
(11, 286)
=71.46
0.00001
0.00001
0.00001
t3:7During exercise period Diet
Ex
Diet × Ex
Time
Time × Diet
Time × Ex
Time × Diet × Ex
F
(1, 26)
=61.06
F
(1, 26)
=0.42
F
(1, 24)
=1.28
F
(10, 270)
= 12,597.78
F
(10, 260)
=28.08
F
(10, 260)
=2.99
F
(10, 260)
=0.28
0.00001
NS
NS
0.00001
0.00001
0.01
NS
The
Q2 main effects from ANOVA are as follows: Time time effect, Diet diet effect, Ex exercise training effect, ×
interaction between variables, NS not significant
Fig. 2 Accumulated caloricintake during the 24 weeks of the protocol in
control (ctl), high-fat diet (HFD), control with chronic exercise (ctl +
training), and high-fat diet with chronic exercise (HFD + training)
groups. The dashed line corresponds to the introduction of exercise in
ctl + trainingand HFD + training groups. *Significant difference between
rats fed with high-fat and standard diet during the first period of the
protocol: p<0.01
Gamelin et al.
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
350 Effect of diet and/or exercise on AEA, 2-AG, PEA,
351 and OEA levels and the expression of genes coding
352 for eCB receptors and eCB metabolic enzymes in brain
353 tissues
354 Hypothalamus
355 Hypothalamic eCB and AEA congener levels are reported in
356 Table 5. Diet had no significant effect on AEA, and AEA
357 congeners, whereas 2-AG levels were increased by HFD
358 (p< 0.01). Gene expression of the enzymes implicated in
359 eCB synthesis or degradation were affected by the HFD
360 (Fig. 4a). The expression of genes coding for ABDH4 and
361 NAPE-PLD (Abdh4 and Nape-pld,respectively),i.e.,twoen-
362 zymes potentially involved in AEA, OEA, and PEA biosyn-
363 thesis, was significantly decreased in the HFD group com-
364 pared to control rats (p< 0.05), and so was the mRNA coding
365 for FAAH (Faah), a major enzyme for AEA, OEA, and PEA
366 degradation (p< 0.05). Thus, the concurrent downregulation
367 of biosynthetic and inactivating enzyme expression may ex-
368 plain the lack of effect of the HFD on levels of AEA and its
369 congeners. Concerning 2-AG, the expression of the mRNA
370 coding for its biosynthetic enzymes (Magl and Dagl-αor
371 Dagl-β) was unchanged with diet while the mRNA coding
372 for its degradation enzymes was either decreased (Abdh6
373 gene) or increased (Abdh12 gene) in the HFD group
374 (p< 0.05). Finally, the expression of mRNAs coding for
375 CB1 and CB2 (Cnr1 and Cnr2, respectively) was not altered
376 in the HFD group, whereas Trpv1 mRNA was significantly
377 decreased (Fig. 4a;p<0.05).
378 Exercise had no significant effect on AEA, 2-AG, or AEA
379 congener levels (Table 5) nor on the mRNA expression of
380genes coding for eCB receptors (CB1, CB2). Gde-1 and
381Nape-pld mRNA expression was significantly decreased with
382exercise in rats on the standard diet (p< 0.05 for both) but not
383in rats on HFD (Fig. 4a). Expression of mRNA for PTNPN-22
384(Ptnpn-22), another enzyme potentially involved in AEA,
385OEA, and PEA biosynthesis, was instead lowered by exercise
386training only in the rats fed with the HFD. Concerning 2-AG
387biosynthesis or degradation, only Abdh6 mRNA levels were
388significantly decreased with exercise training in lean rats
389(Ctl + training group, p<0.05).
390Hippocampus
391Hippocampal eCB and AEA congener levels are reported in
392Tab le 5and results of qPCR analyses for the hippocampus are
393presented in Fig. 4b. The levels of eCBs and AEA congeners
394were not affected by diet or exercise, whereas Cnr1 mRNA
395expression was significantly increased in HFD, HFD + train-
396ing, and Ctl + training groups compared to the Ctl group
397(p< 0.05 for all). Trpv 1 mRNA levels were also increased in
398the HFD + training group compared to the Ctl group
399(p< 0.05). Concerning gene expression of eCB biosynthetic
400or degrading enzymes, none of them was modified by the
401HFD. The mRNA expression of Gde-1 was increased and that
402of Nape-pld was decreased for Ctl + training and HFD + train-
403ing groups (p< 0.05 for both), and Faah mRNA levels were
404increased in these two groups compared to the respective ctl
405groups (p< 0.05 for both). The mRNA expression of the
4062-AG biosynthetic enzyme DAGL-αwas significantly in-
407creased in the HFD + training group as was that of the 2-AG
408degrading enzymes ABDH6, ABDH12, and MAGL (p<0.05
409for all).
t4:1Tab l e 4 Effect of diet and/or exercise on body weight; food intake; maximal aerobic velocity (MAV); basal glucose and insulin levels; and oral glucose
tolerance test (OGTT) in control (Ctl), high-fat diet (HFD), controlwith chronic exercise (Ctl + training), and high-fat diet with chronic exercise (HFD +
training) groups
t4:2Ctl HFD Ctl + training HFD + training Main effects by ANOVA Fvalues p
t4:3MAV (cm s
−1
)
During diet period
During exercise period
45.9 ± 5.2
61.9 ± 6.0
41.6 ± 3.9
55.8 ± 5.0
Time
Diet
Time × Diet
F
(1, 13)
=95.80
F
(1, 12)
=5.41
F
(1,12)
=0.36
0.0001
0.05
NS
t4:4Plasma insulin (ng ml
−1
) 2.8 ± 0.9 2.9 ± 0.5 2.2 ± 0.8 2.5 ± 0.3 Diet
Ex
Diet × Ex
F
(1, 26)
=0.46
F
(1, 26)
=2.19
F
(1, 24)
=0.08
NS
NS
NS
t4:5Plasma glucose (mg dL
−1
) 83±3 103±12* 86±5 92±5 Diet
Ex
Diet × Ex
F
(1, 26)
=21.09
F
(1, 26)
=2.91
F
(1, 24)
=5.70
0.0001
NS
0.05
t4:6OGTT (AUC) 247.3 ± 12.4 306.7 ± 20.4 247.7 ± 16.5 286.1 ± 9.8 Diet
Ex
Diet × Ex
F
(1, 26)
=71.07
F
(1, 26)
=3.02
F
(1, 24)
=3.26
0.0001
NS
NS
Data are means ± SD
The main effects from ANOVA are as follows: Time time effect, Diet diet effect, Ex exercise training effect, ×interaction between variables
*Significantly different from all groups, p<0.05
Q1
Endocannabinoid system in brain, diet, and exercise effects
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
t5:1Tab l e 5 Hypothalamic and hippocampal concentrations of endocannabinoids and anandamide congeners in control (Ctl), high-fat diet (HFD), control
with chronic exercise (Ctl + training), and high-fat diet with chronic exercise (HFD + training) groups
t5:2Ctl HFD Ctl + training HFD + training Main effects
yANOVA
Fvalues p
t5:3Hypothalamus
t5:4AEA (pmol g
−1
) 27.82 ± 22.13 37.66 ± 16.61 24.54 ± 9.82 36.16 ± 14.82 Diet
Ex
Diet × Ex
F
(1, 26)
=2.56
F
(1, 26)
=0.12
F
(1, 24)
=0.01
NS
NS
NS
t5:52-AG (pmol mg
−1
) 3.71 ± 1.02 4.82 ± 1.06 3.17 ± 1.14 4.62 ± 0.69 Diet
Ex
Diet × Ex
F
(1, 26)
=9.24
F
(1, 26)
=0.78
F
(1, 24)
=0.16
0.01
NS
NS
t5:6PEA (pmol mg
−1
) 1.98 ± 0.27 1.90 ± 0.26 1.91 ± 0.23 1.98 ± 0.38 Diet
Ex
Diet × Ex
F
(1, 26)
=0.01
F
(1, 26)
=0.00
F
(1, 24)
=0.40
NS
NS
NS
t5:7OEA (pmol mg
−1
) 0.56 ± 0.86 0.42 ± 0.09 0.53 ± 0.09 0.54 ± 0.11 Diet
Ex
Diet × Ex
F
(1, 26)
=2.64
F
(1, 26)
=1.84
F
(1, 24)
=3.66
NS
NS
NS
t5:8Hippocampus
t5:9AEA (pmol g
−1
) 61.38 ± 24.61 64.20 ± 17.79 58.20 ± 16.44 59.78 ± 18.27 Diet
Ex
Diet × Ex
F
(1, 26)
=0.07
F
(1, 26)
=0.21
F
(1, 24)
=0.01
NS
NS
NS
t5:10 2-AG (pmol mg
−1
) 1.68 ± 0.31 2.38 ± 1.52 2.03 ± 0.27 1.78 ± 0.40 Diet
Ex
Diet × Ex
F
(1, 26)
=0.39
F
(1, 26)
=0.12
F
(1, 24)
=1.75
NS
NS
NS
t5:11 PEA (pmol mg
−1
) 1.00 ± 0.25 1.07 ± 0.52 0.83 ± 0.15 0.80 ± 0.21 Diet
Ex
Diet × Ex
F
(1, 26)
=0.03
F
(1, 26)
=2.90
F
(1, 24)
=0.15
NS
NS
NS
t5:12 OEA (pmol mg
−1
) 0.33 ± 0.07 0.40 ± 0.15 0.36 ± 0.11 0.31 ± 0.06 Diet
Ex
Diet × Ex
F
(1, 26)
=0.05
F
(1, 26)
=0.37
F
(1, 24)
=1.79
NS
NS
NS
Data are means ± SD
The main effects from ANOVA are as follows: Diet diet effect, Ex exercise training effect, ×interaction between variables, NS non-significant
Fig. 3 Blood glucose levels at 0, 30, 60, 90, and 120 min during the oral
glucose tolerance test in control (ctl), high-fat diet (HFD), control with
chronic exercise (ctl + training), and high-fat diet with chronic exercise
(HFD + training) groups. *Significant difference between HFD group and
all other groups: p<0.01.✝ctl + training significantly different from HFD
and HFD + training groups: p<0.05
Gamelin et al.
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
410 Discussion
411 This study aimed at identifying changes in the hypotha-
412 lamic and hippocampal tissue concentrations of 2-AG;
413 AEA; and two AEA congeners, OEA and PEA, together
414 with corresponding alterations in the expression of genes
415 encoding for eCB receptors (CB1, CB2) and enzymes
416involved in the anabolic (ABHD4, GDE-1, NAPE-PLD,
417and PTNP-22, for AEA, OEA, and PEA; DAGL-αand
418DAGL-β, for 2-AG) and catabolic (FAAH, for AEA,
419OEA and PEA; ABHD6, ABHD12 and MAGL, for
4202-AG) pathways of these mediators, after 12 weeks of
421endurance training in Wistar rats fed with a standard diet
422or a HFD.
Fig. 4 Expression level analysis of the genes related to endocannabinoid
metabolism and functionin control (ctl), high-fat diet (HFD), control with
chronic exercise (ctl + training), and high-fat diet with chronic exercise
(HFD + training) groups. mRNA expression levels of genes encoding for
endocannabinoid receptors (cnr1,cnr2,trpv1)oranabolic(abdh4,gde-1,
nape-pld,ptpn22,dagla,daglb) and catabolic (faah,abdh6,abdh12,
magl) enzymes were measured in the hypothalamus (a)andthe
hippocampus (b). The results obtained by qPCR are reported using the
2
−ΔΔct
formula using hprt as housekeeping gene. Each column represents
the mean ± S.E.M. of at least four independent determinations performed
each in quadruplicate. *Significantly different from ctl group: p<0.05
Q1
Endocannabinoid system in brain, diet, and exercise effects
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
423 Previous studies in diet-induced obesity (DIO) mice or ge-
424 netic models of obesity reported a dysregulation of the eCBs
425 and particularly an increase of 2-AG in the hypothalamus [14]
426 [4]. We confirm these results in all Wistar rats fed with the
427 HFD in the present study. As the ECS has an important role in
428 the regulation of food intake through hypothalamic pathways
429 [30], this 2-AG increase might participate in the higher accu-
430 mulated caloric intake observed in these rats. According to the
431 gene expression results of biosynthetic and degrading en-
432 zymes, this 2-AG increase is difficult to explain for two main
433 reasons. First, the HFD group presented at the same time a
434 decreaseandanincreaseingeneexpressionoftwo
435 2-AG-degrading enzymes (Abdh6 and Abdh12, respectively)
436 that are known to participate only for about 15% in 2-AG
437 hydrolysis [5]. Secondly, these results on Abdh6 and Abdh12
438 expression were not observed in the exercise training group
439 fed the HFD, where the same trends should have been ob-
440 served to account for the similar changes in 2-AG levels.
441 Thus, more than the mRNA expression of biosynthetic and/
442 or degrading enzymes, enzymatic activities and eCB biosyn-
443 thetic precursors might be involved in 2-AG level increase
444 induced by HFD. Di Marzo et al. [14] indeed explain this
445 2-AG increase with defective leptin signaling. Leptin inhibits
446 2-AG biosynthesis by likely decreasing the formation of diac-
447 ylglycerol (DAG) precursors [14]. Nevertheless, these results
448 should be taken with caution as this leptin decrease is not
449 systematically observed in Wistar rats fed with high-fat diet
450 [7], and other feeding-regulated hormones, such as grehlin
451 and glucocorticoids, are also involved in the regulation of
452 hypothalamic eCB levels [43]. Furthermore, in the lateral hy-
453 pothalamus and arcuate nucleus (two areas of the hypothala-
454 mus) of mice fed a HFD, the increased 2-AG levels have been
455 recently shown to be accompanied by increased DAGL-α
456 protein expression [10][11].
457 Our results confirm at the mRNA level that the expression
458 of hypothalamic CB1 is not affected by HFD in rats. Previous
459 rat studies have shown a lack of alteration in CB1 density with
460 HFD [22] models. Considering that CB1 agonist levels might
461 regulate CB1 receptor density in the hypothalamus [44], this is
462 consistent with the fact that only the levels of 2-AG, and not
463 also AEA, were found here to be altered in this brain area. Our
464 results however do not exclude transient changes in CB1 or
465 Cnr1 expression over the course of the study. Indeed, South
466 et al. [44] observed a transient increase in mouse hypothalam-
467 ic CB1 density after 3 weeks of HFD that was normalized at
468 the end of the 20 weeks of HFD, suggesting temporal CB1
469 alterations during obesity induction in rodents.
470 Interestingly, Trpv1 gene expression was decreased with
471 the HFD. To our knowledge only Baboota et al. [2] has re-
472 ported a down-regulation in the expression of this gene in the
473 mouse hypothalamus after a HFD period. The
474 down-regulation of Trpv1, observed now here also in rats,
475 may play a role in the hyperglycemia observed in the HFD
476group. Indeed, Zsombok [52] suggested that TRPV1 activa-
477tion in the paraventricular nucleus of the hypothalamus could
478lower blood glucose.
479Unlike Q3the hypothalamus, the HFD did not alter 2-AG
480levels in the hippocampus in our study neither did the
481levels of AEA and its congeners. However, the ECS in
482this brain area was previously shown to be sensitive to
483this kind of diet [34,41], with increased AEA and 2-AG
484hippocampal concentrations in mice after a 12-week HFD
485[34], and Rivera et al. [41] reporting increased AEA, and
486other acylethanolamide (OEA, PEA) but not 2-AG, levels
487also in rats with 12-week HFD. While these previous
488studies reported an unbalanced ratio between the expres-
489sion of eCB biosynthesis and degradation enzymes, we
490did not observe any significant change in the levels of
491the mRNAs coding for these enzymes, which may explain
492the absence of eCB level alterations. Differences in the
493diet composition between studies may partly explain these
494discrepancies. Different amounts of polyunsaturated fatty
495acid precursors in the diet are for example known to mod-
496ulate the availability of eCB biosynthetic precursors [1].
497Nevertheless, as shown in a previous study [34], where
498the mRNA coding for CB1 (Cnr1) and the CB1 protein
499were overexpressed, the hippocampal ECS may be altered
500following a HFD and contribute to the development of
501obesity or related metabolic alterations. Accordingly, an
502increase in CB1 expression was shown to participate in
503the regulation of neuroplasticity involved in the hedonic
504aspect of eating [34]. Here, we found that rats fed with
505HFD showed increased Cnr1 mRNA expression as com-
506pared to rats fed with standard diet, which correlates with
507a higher accumulated caloric intake in HFD rats.
508The up-regulation of hippocampal Cnr1 expression was
509also found with exercise training combined with HFD.
510Chronic exercise did not appear to counteract ECS
511overactivation and, in fact, seems even to induce this ef-
512fect independently of the diet. Indeed, Cnr1 mRNA and
513the majority of the genes coding for biosynthetic and
514degrading eCB enzymes were increased in Ctl + training
515rats, suggesting an increase in eCB turnover and signaling.
516This result is not surprising as Hill et al. [25]havealready
517observed an increase in AEA levels and in CB1 agonist
518binding density after 8 days of free access to a running
519wheel in the hippocampus of lean Wistar rats. They dem-
520onstrated that this ECS overactivation in this brain area
521was required for chronic exercise-induced neuroplasticity.
522However, exercise does not produce the reversal of CB1
523overactivation induced by HFD in spite of the slowing
524down of calorie intake accumulation observed with exer-
525cise training. Studies of different hippocampal regions will
526be needed to determine whether exercise and HFD selec-
527tively affect eCB signaling in the different subpopulations
528of neurons in this brain area.
Gamelin et al.
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
529 Altogether, these data suggest that exercise training induces
530 hippocampal CB1 signaling, which may be potentially in-
531 volved in enhanced neuroplasticity, thus possibly explaining
532 the improved cognitive performance and mood observed typ-
533 ically with exercise [47].
534 In contrast to the hippocampus, the hypothalamic ECS
535 seems to be less sensitive to exercise, as no change in
536 CB1 and CB2 were observed in HFD and Ctl rats, wheth-
537 er they exercised or not. These results may differentiate
538 chronic exercise from chronic stress, considering that re-
539 peated stress exposure leads to ECS alterations in the hy-
540 pothalamus and other brain area involved in the stress
541 response [42]. Wamsteeker et al. [50] showed a functional
542 downregulation of CB1 receptor in the paraventricular nu-
543 cleus of the hypothalamus after repetitive immobilization
544 stress in adolescent Sprague Dawley rats. Indeed, exercise
545 differs from other stressors. It activates a number of sys-
546 tems related to the stress response but other mechanisms
547 associated with chronic exercise exist to reduce the nega-
548 tive effect of this stressor [46], thus possibly avoiding
549 endocannabinoid-signaling impairment.
550 Our study suffers from several limitations. First, gene
551 expression quantification without data on receptor or en-
552 zyme protein levels or function limits the interpretation of
553 our results. Secondly, the brain areas studied were limited
554 to the hippocampus and hypothalamus. Other brain struc-
555 tures involved in the control of eating and voluntary ex-
556 ercise behaviors, such as the ventral tegmental area and
557 striatum, are highly sensitive to eCB signaling [13,20,
558 44]. Measurements of ECS activities in these areas would
559 have allowed providing a global view of the different
560 structures affected by exercise and food intake. Third,
561 treadmill was chosen instead of running wheel as exercise
562 protocol. Even though both of these protocols are known
563 to induce neural plasticity [33], the forced exercise could
564 induce additional stress compared to the voluntary exer-
565 cise [19] and thus affect the ECS differently. However,
566 our results (i.e., Cnr1 upregulation in the hippocampus)
567 do not seem to be the mere consequence of chronic stress
568 induced by forced exercise. In fact, Hill et al. [24]report-
569 ed hippocampal CB1 down-regulation following 21 days
570 of chronic stress.
571 In summary, we have confirmed here that HFDs are accom-
572 panied by changes in the expression of genes encoding for
573 eCB receptors and enzymes involved in the anabolic and cat-
574 abolic pathways of the ECS in the hippocampus. These chang-
575 es, and particularly the alteration of hippocampal CB1 recep-
576 tor gene expression, may participate in weight gain and glu-
577 cose metabolic perturbation observed with the HFD. While
578 chronic exercise improves weight alterations and avoids hy-
579 perglycemia induced by HFD, the alterations in ECS gene
580 expression occurring during the latter diet are not reversed,
581 and chronic exercise may even result in several hippocampal
582ECS responses similar to those observed with HFD. These
583results highlight the need for additional investigations about
584the role of the ECS in the beneficial brain adaptations induced
585by chronic exercise.
586Acknowledgments The authors thank Gamain J., Barbez P., Max, and
587Mahault G. for their advices and technical assistances.
588
589
Compliance with ethical standards
590Conflict of interest The authors declare that they have no conflicts of
591interest.
592References 593
594
1. Artmann A, Petersen G, Hellgren LI, Boberg J, Skonberg C,
595Nellemann C, Hansen SH, Hansen HS (2008) Influence of dietary
596fatty acids on endocannabinoid and N-acylethanolamine levels in
597rat brain, liver and small intestine. Biochim Biophys Acta 1781:
598200–212
5992. Baboota RK, Murtaza N, Jagtap S, Singh DP, Karmase A, Kaur J,
600Bhutani KK, Boparai RK, Premkumar LS, Kondepudi KK et al
601(2014) Capsaicin-induced transcriptional changes in hypothalamus
602and alterations in gut microbial count in high fat diet fed mice. J
603Nutr Biochem 25:893–902
6043. Bennetzen MF (2011) Investigations of the endocannabinoid sys-
605tem in adipose tissue: effects of obesity/weight loss and treatment
606options. Dan Med Bull 58:B4269
6074. Bisogno T, Mahadevan A, Coccurello R, Chang JW, Allara M,
608Chen Y, Giacovazzo G, Lichtman A, Cravatt B, Moles A et al
609(2013) A novel fluorophosphonate inhibitor of the biosynthesis of
610the endocannabinoid 2-arachidonoylglycerol with potential anti-
611obesity effects. Br J Pharmacol 169:784–793
6125. Blankman JL, Simon GM, Cravatt BF (2007) A comprehensive
613profile of brain enzymes that hydrolyze the endocannabinoid 2-
614arachidonoylglycerol. Chem Biol 14:1347–1356
6156. Cadas H, Gaillet S, Beltramo M, Venance L, Piomelli D (1996)
616Biosynthesis of an endogenous cannabinoid precursor in neurons
617and its control by calcium and cAMP. J Neurosci 16:3934–3942
6187. Chalkley SM, Hettiarachchi M, Chisholm DJ, Kraegen EW (2002)
619Long-term high-fat feeding leads to severe insulin resistance but not
620diabetes in Wistar rats. Am J Physiol Endocrinol Metab 282:
621E1231–E1238
6228. Christensen R, Kristensen PK, Bartels EM, Bliddal H, Astrup A
623(2007) Efficacy and safety of the weight-loss drug rimonabant: a
624meta-analysis of randomised trials. Lancet 370:1706–1713
6259. Cota D, Marsicano G, Tschop M, Grubler Y, Flachskamm C,
626Schubert M, Auer D, Yassouridis A, Thone-Reineke C, Ortmann
627S et al (2003) The endogenous cannabinoid system affects energy
628balance via central orexigenic drive and peripheral lipogenesis. J
629Clin Invest 112:423–431
63010. Cristino L, Busetto G, Imperatore R, Ferrandino I, Palomba L,
631Silvestri C, Petrosino S, Orlando P, Bentivoglio M, Mackie K
632et al (2013) Obesity-driven synaptic remodeling affects
633endocannabinoid control of orexinergic neurons. Proc Natl Acad
634Sci U S A 110:E2229–E2238
63511. Cristino L, Luongo L, Imperatore R, Boccella S, Becker T, Morello
636G, Piscitelli F, Busetto G, Maione S, Di Marzo V (2016) Orexin-a
637and endocannabinoid activation of the descending antinociceptive
638pathway underlies altered pain perception in leptin signaling defi-
639ciency. Neuropsychopharmacology 41:508–520
Q1
Endocannabinoid system in brain, diet, and exercise effects
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
640 12. D'Eon TM, Pierce KA, Roix JJ, Tyler A, Chen H, Teixeira SR
641 (2008) The role of adipocyte insulin resistance in the pathogenesis
642 of obesity-related elevations in endocannabinoids. Diabetes 57:
643 1262–1268
644 13. De Chiara V, Errico F, Musella A, Rossi S, Mataluni G, Sacchetti L,
645 Siracusano A, Castelli M, Cavasinni F, Bernardi G et al Voluntary
646 exercise and sucrose consumption enhance cannabinoid CB1 recep-
647 tor sensitivity in the striatum. Neuropsychopharmacology 35:374–
648 387
649 14. Di Marzo V, Goparaju SK, Wang L, Liu J, Batkai S, Jarai Z, Fezza
650 F, Miura GI, Palmiter RD, Sugiura T et al (2001) Leptin-regulated
651 endocannabinoids are involved in maintaining food intake. Nature
652 410:822–825
653 15. Di Marzo V, Matias I (2005) Endocannabinoid control of food
654 intake and energy balance. Nat Neurosci 8:585–589
655 16. Di Marzo V (2008) Endocannabinoids: synthesis and degradation.
656 Rev Physiol Biochem Pharmacol 160:1–24
657 17. Di Marzo V, Cote M, Matias I, Lemieux I, Arsenault BJ, Cartier A,
658 Piscitelli F, Petrosino S, Almeras N, Despres JP (2009a) Changes in
659 plasma endocannabinoid levels in viscerally obese men following a
660 1 year lifestyle modification programme and waist circumference
661 reduction: associations with changes in metabolic risk factors.
662 Diabetologia 52:213–217
663 18. Di Marzo V, Verrijken A, Hakkarainen A, Petrosino S, Mertens I,
664 Lundbom N, Piscitelli F, Westerbacka J, Soro-Paavonen A, Matias I
665 et al (2009b) Role of insulin as a negative regulator of plasma
666 endocannabinoid levels in obese and nonobese subjects. Eur J
667 Endocrinol 161:715–722
668 19. Dishman RK (1997) Brain monoamines, exercise, and behavioral
669 stress: animal models. Med Sci Sports Exerc 29:63–74
670 20. Dubreucq S, Durand A, Matias I, Benard G, Richard E, Soria-
671 Gomez E, Glangetas C, Groc L, Wadleigh A, Massa F et al
672 (2013) Ventral tegmental area cannabinoid type-1 receptors control
673 voluntary exercise performance. Biol Psychiatry 73:895–903
674 21. Gamelin FX, Aucouturier J, Iannotti FA, Piscitelli F, Mazzarella E,
675 Aveta T, Leriche M, Dupont E, Cieniewski-Bernard C, Montel V,
676 et al. (2016) Effects of chronic exercise on the endocannabinoid
677 system in Wistar rats with high-fat diet-induced obesity. J Physiol
678 Biochem
679 22. Harrold JA, Elliott JC, King PJ, Widdowson PS, Williams G (2002)
680 Down-regulation of cannabinoid-1 (CB-1) receptors in specific
681 extrahypothalamic regions of rats with dietary obesity: a role for
682 endogenous cannabinoids in driving appetite for palatable food?
683 Brain Res 952:232–238
684 23. Heyman E, Gamelin FX, Goekint M, Piscitelli F, Roelands B,
685 Leclair E, Di Marzo V, Meeusen R (2012) Intense exercise in-
686 creases circulating endocannabinoid and BDNF levels in humans
687 —possible implications for reward and depression.
688 Psychoneuroendocrinology 37:844–851
689 24. Hill MN, Patel S, Carrier EJ, Rademacher DJ, Ormerod BK, Hillard
690 CJ, Gorzalka BB (2005) Downregulation of endocannabinoid sig-
691 naling in the hippocampus following chronic unpredictable stress.
692 Neuropsychopharmacology 30:508–515
693 25. Hill MN, Titterness AK, Morrish AC, Carrier EJ, Lee TT, Gil-
694 Mohapel J, Gorzalka BB, Hillard CJ, Christie BR (2010)
695 Endogenous cannabinoid signaling is required for voluntary
696 exercise-induced enhancement of progenitor cell proliferation in
697 the hippocampus. Hippocampus 20:513–523
698 26. Iannotti FA, Piscitelli F, Martella A, Mazzarella E, Allara M,
699 Palmieri V, Parrella C, Capasso R, Di Marzo V (2013) Analysis
700 of the "endocannabinoidome" in peripheral tissues of obese Zucker
701 rats. Prostaglandins Leukot Essent Fatty Acids 89:127–135
702 27. Jamshidi N, Taylor DA (2001) Anandamide administration into the
703 ventromedial hypothalamus stimulates appetite in rats. Br J
704 Pharmacol 134:1151–1154
70528. Jo YH, Chen YJ, Chua SC Jr, Talmage DA, Role LW (2005)
706Integration of endocannabinoid and leptin signaling in an
707appetite-related neural circuit. Neuron 48:1055–1066
70829. Kim KW, Zhao L, Donato J Jr, Kohno D, Xu Y, Elias CF, Lee C,
709Parker KL, Elmquist JK (2011) Steroidogenic factor 1 directs pro-
710grams regulating diet-induced thermogenesis and leptin action in
711the ventral medial hypothalamic nucleus. Proc Natl Acad Sci U S A
712108:10673–10678
71330. Kirkham TC, Williams CM, Fezza F, Di Marzo V (2002)
714Endocannabinoid levels in rat limbic forebrain and hypothalamus
715in relation to fasting, feeding and satiation: stimulation of eating by
7162-arachidonoyl glycerol. Br J Pharmacol 136:550–557
71731. Kirkham TC (2009) Endocannabinoids and the non-homeostatic
718control of appetite. Curr Top Behav Neurosci 1:231–253
71932. Krawczewski Carhuatanta KA, Demuro G, Tschop MH, Pfluger
720PT, Benoit SC, Obici S (2011) Voluntary exercise improves high-
721fat diet-induced leptin resistance independent of adiposity.
722Endocrinology 152:2655–2664
72333. Liu YF, Chen HI, Wu CL, Kuo YM, Yu L, Huang AM, Wu FS,
724Chuang JI, Jen CJ (2009) Differential effects of treadmill running
725and wheel running on spatial or aversive learning and memory:
726roles of amygdalar brain-derived neurotrophic factor and synapto-
727tagmin I. J Physiol 587:3221–3231
72834. Massa F, Mancini G, Schmidt H, Steindel F, Mackie K, Angioni C,
729Oliet SH, Geisslinger G, Lutz B (2010) Alterations in the hippo-
730campal endocannabinoid system in diet-induced obese mice. J
731Neurosci 30:6273–6281
73235. Matias I, Petrosino S, Racioppi A, Capasso R, Izzo AA, Di Marzo
733V (2008) Dysregulation of peripheral endocannabinoid levels in
734hyperglycemia and obesity: effect of high fat diets. Mol Cell
735Endocrinol 286:S66–S78
73636. Pagotto U, Marsicano G, Cota D, Lutz B, Pasquali R (2006) The
737emerging role of the endocannabinoid system in endocrine regula-
738tion and energy balance. Endocr Rev 27:73–100
73937. Pedersen BK, Saltin B (2006) Evidence for prescribing exercise as
740therapy in chronic disease. Scand J Med Sci Sports 16(Suppl 1):3–63
74138. Pi-Sunyer FX, Aronne LJ, Heshmati HM, Devin J, Rosenstock J &
742Group RI-NAS (2006) Effect of rimonabant, a cannabinoid-1 re-
743ceptor blocker, on weight and cardiometabolic risk factors in over-
744weight or obese patients: RIO-North America: a randomized con-
745trolled trial. JAMA 295:761–775
74639. Purves RD (1992) Optimum numerical integration methods for
747estimation of area-under-the-curve (AUC) and area-under-the-
748moment-curve (AUMC). J Pharmacokinet Biopharm 20:211–226
74940. Raichlen DA, Foster AD, Gerdeman GL, Seillier A, Giuffrida A
750(2012) Wired to run: exercise-induced endocannabinoid signaling
751in humans and cursorial mammals with implications for the 'run-
752ner's high. J Exp Biol 215:1331–1336
75341. Rivera P, Luque-Rojas MJ, Pastor A, Blanco E, Pavon FJ,
754Serrano A, Crespillo A, Vida M, Grondona JM, Cifuentes
755M et al (2013) Diet-dependent modulation of hippocampal
756expression of endocannabinoid signaling-related proteins in
757cannabinoid antagonist-treated obese rats. Eur J Neurosci
75837:105–117
75942. Senst L, Bains J (2014) Neuromodulators, stress and plasticity: a
760role for endocannabinoid signalling. J Exp Biol 217:102–108
76143. Silvestri C, Di Marzo V (2013) The endocannabinoid system in
762energy homeostasis and the etiopathology of metabolic disorders.
763Cell Metab 17:475–490
76444. South T, Huang XF (2008) Temporal and site-specific brain alter-
765ations in CB1 receptor binding in high fat diet-induced obesity in
766C57Bl/6 mice. J Neuroendocrinol 20:1288–1294
76745. Sparling PB, Giuffrida A, Piomelli D, Rosskopf L, Dietrich A
768(2003) Exercise activates the endocannabinoid system.
769Neuroreport 14:2209–2211
Gamelin et al.
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTEDPROOF
770 46. Stranahan AM, Lee K, Mattson MP (2008) Central mechanisms of
771 HPA axis regulation by voluntary exercise. NeuroMolecular Med
772 10:118–127
773 47. Tantimonaco M, Ceci R, Sabatini S, Catani MV, Rossi A, Gasperi
774 V, Maccarrone M (2014) Physical activity and the endocannabinoid
775 system: an overview. Cell Mol Life Sci 71:2681–2698
776 48. van Praag H, Shubert T, Zhao C, Gage FH (2005) Exercise en-
777 hances learning and hippocampal neurogenesis in aged mice. J
778 Neurosci 25:8680–8685
779 49. Verty AN, McGregor IS, Mallet PE (2004) The dopamine receptor
780 antagonist SCH 23390 attenuates feeding induced by Delta9-tetra-
781 hydrocannabinol. Brain Res 1020:188–195
78250. Wamsteeker JI, Kuzmiski JB, Bains JS (2010) Repeated
783stress impairs endocannabinoid signaling in the
784paraventricular nucleus of the hypothalamus. J Neurosci
78530:11188–11196
78651. Yan ZC, Liu DY, Zhang LL, Shen CY, Ma QL, Cao TB, Wang LJ,
787Nie H, Zidek W, Tepel M et al (2007) Exercise reduces adipose
788tissue via cannabinoid receptor type 1 which is regulated by perox-
789isome proliferator-activated receptor-delta. Biochem Biophys Res
790Commun 354:427–433
79152. Zsombok A (2013) Vanilloid receptors—do they have a role in
792whole body metabolism? Evidence from TRPV1. J Diabetes
793Complicat 27:287–292
794
Q1
Endocannabinoid system in brain, diet, and exercise effects
JrnlID 13105_ArtID 557_Proof# 1 - 04/03/2017
UNCORRECTED PROOF
AUTHOR QUERIES
AUTHOR PLEASE ANSWER ALL QUERIES.
Q1. Article title has been modified. Please check.
Q2. Please check table footnote if modified correctly.
Q3. The sentence “Unlike the hypothalamus, the HFD did not…” has been modified for clarity. Please
check if the intended meaning was retained and amend as necessary.
