HAL Id: hal-03366396
https://hal.archives-ouvertes.fr/hal-03366396
Submitted on 19 Oct 2021
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Ultradian feeding in mice not only aects the peripheral
clock in the liver, but also the master clock in the brain
Satish Sen, Hélène Raingard, Stéphanie Dumont, Andries Kalsbeek, Patrick
Vuillez, Etienne Challet
To cite this version:
Satish Sen, Hélène Raingard, Stéphanie Dumont, Andries Kalsbeek, Patrick Vuillez, et al.. Ul-
tradian feeding in mice not only aects the peripheral clock in the liver, but also the mas-
ter clock in the brain. Chronobiology International, Taylor & Francis, 2017, 34 (1), pp.17-36.
�10.1080/07420528.2016.1231689�. �hal-03366396�
Ultradian feeding in mice not only affects the peripheral clock in the liver, but
also the master clock in the brain
Satish Sen
a,b,c
, Hélène Raingard
a
, Stéphanie Dumont
a
, Andries Kalsbeek
b,c,d
, Patrick Vuillez
a,c
, and
Etienne Challet
a,d
a
5 Regulation of Circadian Clocks team, Institute of Cellular and Integrative Neurosciences, UPR3212, Centre National de la Recherche
Scientifique (CNRS), University of Strasbourg, France
Q1 ;
b
Hypothalamic Integration Mechanisms, Netherlands Institute for Neuroscience (NIN),
Amsterdam, The Netherlands;
c
International Associated Laboratory LIA1061 Understanding the Neural Basis of Diurnality, CNRS, France and
The Netherlands
Q2 ;
d
Department of Endocrinology and Metabolism, Academic Medical Center (AMC), University of Amsterdam, The
Netherlands.
Q3
10 ABSTRACT
Q4 Restricted feeding during the resting period causes pronounced shifts in a number of
peripheral clocks, but not the central clock in the suprachiasmatic nucleus (SCN). By contrast,
daily caloric restriction impacts also the light-entrained SCN clock, as i ndicated by shifted
oscillations of clock (PER1) and clock-con trolled (vasopressin) proteins. To determine if these
15 SCN changes are due to the metabolic or timing cues of the restricted feeding, mice were
challenged with an ultradian 6-meals schedule (1 food access every 4 h) to abolish the daily
periodicity of feeding. Mice fed with ultradian feeding that lost <10% body mass (i.e.
isocaloric) displayed 1.5-h phase-advance of body temperature rhythm, but remained mostly
nocturnal, together with up-regulated vasopressin and down-regulated PER1 and PER2 levels
20 in the SCN. Hepatic expression of clock genes (P er2, Rev-erb α,andClock)andFgf21 was,
respectively, phase-advanced and up-regulated by ultradian feeding. Mice fed with ultradian
feeding that lost >10% body mass (i.e. hypocalori c) became more diurnal, hypothermic in
late night, and displayed larger (3.5 h) advance of body temperature rhythm, m ore reduced
PER1 expression in the SCN, and further modified gene expression in the liver (e.g. larger
25 phase-advance of Per2 and up-regulated levels of Pgc-1α). While glucose rhythmicity was lost
under ultrad ian fe eding, the p hase of daily rhythms in li ver glycogen and plasma c orticos-
terone (albeit increased in amplitude) remained unchanged. In conclusion, the additional
impact of hypocaloric conditions on the SCN are mainly due to the metabolic and n ot the
timing effects o f restricted daytime feeding.
KEYWORDS
30 circadian rhythm; feeding;
6-meal schedule; clock gene;
suprachiasmatic nucleus
Introduction
Biological rhythms are under the control of circadian
os
cillators, including a master circadian clock
located in the suprachiasmatic nucleus (SCN) of
35 the anterior hypothalamus and peripheral oscillators
present in almost every cell of the body (Bray &
Young, 2009). The underlying molecular mechanism
of the clock is based on transcriptional and
translational feedback loops consisting of positive
40 and negative elements (Reppert & Weaver, 2001 ).
When heterodimerized, the positive limb elements
BMAL1 and CLOCK activate transcription of the
negative elements (Period (Per)1, 2, 3 and
Cryptochrome (Cry)1, 2) that in turn inhibit
45 BMAL1/CLOCK transactivation. In parallel, other
clock genes such as Rev-erbα, β and Ror α, β,whose
transcription is also activated by BMAL1/CLOCK,
modulate Bmal1 and Clock transcription (Cho et al.,
2012; Crumbley & Burris, 2011; Preitner et al., 2002).
50Light perceived by the retina is the most potent
synchronizer of the circadian rhythm produced by
the molecular clock mechanism within the SCN. The
molecular clockwork regulates the rhythmic
transcription of clock-controlled genes, such as the
55gene coding for neuropeptide Arginine Vasopressin
(Avp) (Jin et al., 1999 ). The output of the SCN con-
trols the timing of peripheral clocks via nervous,
hormonal and behavioral cues (Froy, 2011). Food
access restricted to the usual resting period can
60phase-shift circadian oscillations in a number of
peripheral organs and brain regions outside the
SCN, while the SCN master clock remains
CONTACT Etienne Challet challet@inci-cnrs.unistra.fr INCI, CNRS UPR3212, 5 rue Blaise Pascal, 67084 Strasbourg, France. Tel: +33 388456693.Q5
CHRONOBIOLOGY INTERNATIONAL
2016, VOL. 00, NO. 00, 1–20
http://dx.doi.org/10.1080/07420528.2016.1231689
© 2016 Taylor & Francis
synchronized to the light-dark cycle (Damiola et al.,
2000;Feilletetal.,2008;Stokkanetal.,2001).
65 However, when daytime restricted feeding is com-
bined with caloric restriction, the master clock is
affected, as assessed by phase-advances in daily
rhythms of body temperature, activity rhythm, and
pineal melatonin, as well as by altered photic reset-
70 ting (Challet, 2010). Moreover, daily caloric restric-
tion leads to phase-shifts in daily oscillations of clock
(PER1) and clock-controlled (AVP) proteins in the
SCN (Mendoza et al., 2007). To avoid the synchro-
nizing effects of daily restricted feeding, a protocol
75 has been developed using a feeding regimen of six
10-min food accesses equally distributed over 24 h
(i.e. one 10-min meal every 4 h) (Kalsbeek &
Strubbe, 1998). In nocturnal rats under light-dark
conditions, this ultradian 6-meals-a-day feeding
80 schedule does not modify the phase of locomotor
activity rhythm, but if food access to the 6-meals is
shortened to cause body mass loss, rats become
partially active during daytime due to a phase-
advance of the rest/activity rhythm (Mendoza et al.,
85 2008).
One recent rat study showed that peripheral
clock gene rhythms are still present during ultra-
dian 6-meals-a-day feeding, despite changes in
amplitude and phase (Su et al., 2016b). Another
90 study suggested that in mice, the peripheral clocks
remain unaffected by meal timing when each meal
is given equally spaced either 2, 3, 4 or 6 times per
day. However, if meal frequency is unevenly dis-
tributed, i.e. with unequal intervals between the
95 meals, then the phase of peripheral clock genes
changes, especially in the kidney. Moreover, that
study also showed that ultradian 6-meal feeding
coupled to caloric restriction was able to produce
phase-advances of peripheral clocks inversely pro-
100 portional to the degree of energy intake (Kuroda
et al., 2012).
In the present study, we aimed at investigating
further whether it is the daily timing of feeding
and fasting or metabolic cues associated with calo-
105 ric restriction that affects the central and periph-
eral clocks. For that purpose, we challenged mice
with a 6-meals-a-day feeding schedule (combined
with isocaloric or hypocaloric conditions) and stu-
died their behavioural and physiological changes,
110 as well as expression of clock and clock-controlled
genes in the master clock and liver.
Materials and methods
Animals and housing
Seventy-six5-weekoldmaleC57BL/6Jmice
115(Janvier labs, Le Genest-Saint-Isle, France) were
usedforthisstudy.Theanimalswerehousedin
individual cages equipped with a wheel, at an
ambient temperature of 23 ± 2°C under 12:12 h
light-dark conditions (lights on at 7:00 AM
120(defining Zeitgeber Time (ZT) 0) and off at
19:00 PM (=ZT12)). In a group of 46 animals,
access to food was automatically controlled by
electronic timers for six cages at a time. Thirty
animals served as controls and had ad libitum
125access to food.
All experiments were performed in accordance
with the U.S. National Institutes of Health Guide
for the Care and Use of Laboratory Animals
(1996)
Q6, the French National Law (implementing
130the European Communities Council Directive 86/
609/EEC) and approved in advance by the
Regional Ethical Committee of Strasbourg for
Animal Experimentation (AL/50/57/02/13) and in
compliance with the ethical standards of the jour-
135nal (Portaluppi et al., 2010).
Surgery
Mice were anesthetized with isoflurane (Vetflurane,
Virbac 3% powered by 0.2 l/min O2) to implant a
transponder (Minimitter, Vitalview, Sunriver, OR,
140USA) in the abdominal cavity to record body
temperature and general cage activity. The abdo-
men was shaved and sprayed with antiseptic
(DermaSpray, Bayer) before an incision in the
skin and muscle (8–10 mm) was made. Once the
145transponder was inserted into the abdominal cavity,
the muscle layer was stitched with surgical sutures
(Filapeau, 3.0) and anti-inflammatory medication
was provided in drinking water (Metacam, 0.2
mg/ml, 0.1 ml) for 2 days.
150Experimental procedure
After surgery, the animals were placed in experimen-
tal cages for 2 weeks with drinking water and food ad
libitum. After this, mice were habituated to an ultra-
dian schedule of six meals each day with one food
155access every 4 h (ZT2, ZT6, ZT10, ZT14, ZT18 and
2 S. SEN ET AL.
ZT22). Access to food during restricted feeding was
set automatically by the Food Planning system based
on a food basket allowing and preventing food access
in the low and upper position, respectively
160 (Intellibio, Seichamps, France). Lifting and fall of
the food basket being associated with a brief motor
noise, these auditory cues may have signaled food
availability to the mice. The smaller mesh size of the
trough compared to the size of the food pellets pre-
165 vented any food hoarding in the cage. The duration
of food access was reduced every 4 days gradually
from6×1h,via6×30min,6×20minto6×15
min. This protocol was based on previous studies in
rats (Kalsbeek & Strubbe, 1998 ;Mendozaetal.,
170 2008). The fact th at mic e were fed every day at the
same times has pr obably improved their ability to
adjust to ultradian feeding, as opposed to irregular
meal times (Valle, 1981 ). Food int ake during day-
time and nighttime was measured twice (at the steps
175 of 6 × 1 h and 6 × 15 min) to evaluate the day-night
pattern of food intake. The body mass was measured
weekly. At the end of two weeks of feeding according
to the 6 × 15 min protocol, two groups were
categorized according to individual adaptation to
180 the paradigm, eventually leading to body mass loss.
A cut-off at 10% body mass loss allowed to distin-
guish an isocaloric group including animals with less
than 10% of body mass loss (mean: 5.4 ± 0.5%; n total
=24;n = 4 per ZT) and a hypocaloric group in which
185 animals lost 10% or more (up to 25%) body mass
(mean: 15.5 ± 1.1%; n total = 22; n =3–4perZT).
Animals of the control group were kept with food
and wa ter ad libitum (n =5perZT).
Immunohistochemistry
190 At the end of the experiment, animals were sacrificed
with an overdose of pentobarbital. Mice fed with
ultradian 6-meals schedule were sampled every 4 h
betweenfoodaccesses(i.e.ZT0,ZT4,ZT8,ZT12,
ZT16 and ZT20) to limit direct effects of feeding
195 while avoiding prolonged fasting. Control mice fed
ad libitum were sacrificed at the same times. Blood
was sampled by intracardiac puncture, liver was
sampled in the right lobe, and the heart was perfused
with 50 mL of 0.9% saline followed by 50 mL of 4%
200 paraformaldehyde in phosphate buffer (0.1 M, pH
7.4). Brains were removed, postfixed overnight in
4% paraformaldehyde (4°C) and transferred to a
cryoprotectant buffered sucrose solution (30% at 4°
C) for at least 24 h till brains sank to the bottom due to
205the sucrose density gradient. Brains were then frozen
in isopentane around −50°C and stored at −80°C . Five
series of 30-μm coronal SCN sections were prepared
on a cryostat and collected in Phosphate-Buffered
Saline (0.1 M PBS, 1×) and washed with 1× Tris
210Buffer Saline pH 7.6 (0.1 M TBS 1×). Then, sections
were incubated in 3% H
2
O
2
in TBS (30 min) to
suppress endogenous peroxidase activity, thereby
reducing background staining. Again brain sections
were rinsed in TBS 1×. Brain sections were then
215transferred in a solution containing 10% normal
serum (either goat or horse according to the host
species of the primary antibody) and Triton X-100
(0.1 %) in TBS for 2 h, followed by incubation in the
primary antibody (48 h at 4°C). We used rabbit poly-
220clonal anti arginine-vasopressin (AVP) (1:20000,
Truus, a gift from Dr. Ruud Buijs, Netherlands
Institute for Brain Research, Amsterdam, the
Netherlands), goat polyclonal anti-PER1 (1:750; SC-
7724, Santa Cruz Biotechnologies, Santa Cruz, CA,
225USA) and rabbit polyclonal anti-PER2 (1:3000, #PER-
21A; Alpha Diagnostic International, San Antonio
TX, USA; note that for anti-PER2 immunohisto-
chemistry, PBS indicated below was always replaced
with TBS). The sections were washed in PBS 1×, then
230incubated (2 h at 4°C) with biotinylated goat anti-
rabbit IgG (1:500, PK6101; Vectastain Standard Elite
ABC Kit Vector Laboratories, Inc., Burlingame, CA,
USA) for AVP and PER2 and with biotinylated anti-
goat IgG made in horse (1:500, BA-9500; Vector labs)
235for PER1 immunostaining. After this, sections were
rinsed in PBS 1× and incubated (2 h) in a solution
containing avidin–biotin peroxidase complex
(Vectastain Elite ABC kit; Vector Laboratories Inc.).
Following incubation with ABC reagents, sections
240were rinsed 4 times in PBS, and incubated with
H
2
O
2
(0.015%, Sigma-Aldrich, St Louis, MO, USA)
and 3,3’-diaminobenzidine tetrahydrochloride (0.5
mg/ml, Sigma-Aldrich) diluted in water. Thereafter,
sections were rinsed with PBS, wet mounted on slides
245coated with gelatin, dehydrated through a series of
alcohols, soaked in xylene, and cover slipped.
Photomicrographs were taken on Leica DMRB
microscope (Leica Microsystems) with an Olympus
DP50 digital camera (Olympus France). The number
250of immunopositive cells was counted on one section
in both SCN’s and averaged.
CHRONOBIOLOGY INTERNATIONAL 3
mRNA extraction and quantitative real-time PCR
RNA was extracted from frozen liver samples by
homogenizing liver samples in lysis buffer supplemen-
255 ted with β-mercaptoethanol and using absolutely
RNA miniprep kit (Agilent Technologies, USA. The
sampleswerepurifiedbyprecipitationwithsodium
acetate and isopropyl alcohol. The quality of RNA was
measured on NanoDrop ND-100 spectrophotometer
260 (NanoDrop Technologies, Wilmington, DE, USA;
A260/A280, and A260/A230 values were > 1.8) RNA
integrity was assessed using (Agilent RNA 6000 Nano
Kit) on Aligent 2100 bio-analyzer for all the liver
samples (RIN Value were >7) bio-analyzer. cDNA
265 was synthesized with the High Capacity RNA to
cDNA kit (Applied Biosystem, Foster city CA, USA)
using 1 µg of RNA. Measurement of relative
abundance was performed by real-time PCR analysis
using 1× of TaqMan Gene Expression Master Mix
270 (Life Technologies, Foster city, CA, USA). The
following TaqMan probes (Per2: Mm00478113_m1,
Clock: Mm00455950_m1, Sirt1: Mm00490758_m1,
Fgf21: Mm00840165_g1, Nr1d1 (Rev-erb α):
Mm00520708_m1, Pparα: Mm01208835 m1 and
275 Pgc-1α: Mm00440939_m1) were used for all the
genes with 1µl of cDNA in the reaction mixture of
20 μl. Each reaction PCR was done in duplicate. A
dilution curve was prepared of pooled cDNA samples
using log10 standards to calculate the amplification
280 efficiency for each primer set (values were between
1.85–1.99). Data were normalized to Tbp
(Mm00446971_m1) and analysed the comparative
cycle threshold (Ct) method RQ = 2-ΔΔCt. ΔΔCt =
ΔCt sample – ΔCt reference (Pfaffl, 2001) with effi-
285 ciency corrections. Transcript levels were calculated
relative to the mean of ZT 0 samples.
Plasma metabolic parameters
Plasma samples were obtained after centrifuga-
tion of fresh blood collected with 4% EDTA (10
290 µL for 1 mL of blood) and centrifuged for 10
min (5000 rpm at 4 °C). Plasma gl ucose was
evaluated with GOD-PAP kit (Biola bo, Maizy,
France). The ACS-ACOD method (NEFA-HR2;
Wako, Osaka, Japan) was used for assaying
295 plasma non-esterified fatty acids (NEFA).
Plasma concentrations of corticosterone were
determined by a Rat/Mouse Corticosterone EIA
kit (AC-14F 1, IDS EURL, Paris, FRA NCE). The
limit of s ensitivity of the assay was 0.55 ng/mL.
300Hepatic glycogen assay
Samples of fresh liver were flash-frozen in liquid
nitrogen. Hepatic glycogen was quantified accord-
ing to the method developed by Murat and Serfaty
(Murat & Serfaty, 1974).
305Statistical analysis
Data are presented as mean ± standard error of the
mean (SEM). Statistical analysis was performed by
SigmaPlot (version 12, SPSS Inc, Chicago, IL,
USA). Significance was defined at p < 0.05.
310Two-way analysis of variance (ANOVA) with or
without repeated measures (RM) were performed
to assess the effect of Feeding conditions (food ad
libitum, ultradian iso- or ultradian hypo-caloric
feeding) and the effect of Time (either baseline
315versus experimental condition, or time of day),
and the Interaction between these factors. One-
way analysis of variance (ANOVA) was performed
to assess the effect of Time in the separate feeding
groups. When appropriate, post-hoc analysis was
320performed using Fisher LSD method. For assessing
daily rhythmicity, we used a cosinor analysis to
determine mean level, amplitude and acrophase of
the considered parameter with SigmaPlot software
(Jandel Scientific,Chicago, IL). Data were fitted to
325the following regression: [y = A + B·cos(2π(x − C)/
24)], where A is the mean level, B the amplitude
and C the acrophase of the rhythm.
Results
Food intake
330When mice were fed with 6 meals of 1 h each, they ate
most of their food during their active period (night
time) and less during their resting period (p<0.001).
Post-hoc analysis showed a significant effect of Time
(nightvs.day;FisherLSDtest,p<0.05) (Figure 1A).
335Mice on the 15-min 6-meals-a-day feeding schedule,
both the hypocaloric and isocaloric group, ate equal
amounts of food during day and night. However, the
amount of food ingested by the hypocaloric group
was significantly lower (~20%) than that by the
4 S. SEN ET AL.
